ethiopian journal pediatric
DESCRIPTION
;jo;TRANSCRIPT
1
ETHIOPIAN JOURNAL OF PEDIATRICS AND CHILD HEALTH July 2009, Volume V, Number 5 Original articles The Prevalence of Nosocomial Infections and Associated Risk Factors in Pediatric Patients in Tikur Anbessa Hospital Mikyas Demissie ,M.D. , Sileshi Lulsesed, M.D. Clinical Predictors of Pneumonia Among Under-five Children At Tikur Anbesa Specilaized Hospital Kalid Asrat, MD, Amha Mekasha, MD, MSc Gullian Barre Syndrome in Children At Tikur Anbessa Specialized Hospital Tigist Bacha, MD
Assessment of quality of care of sick under-five children in referral hospitals in Ethiopia.
Sirak Hailu, MD, Solomon Emyu, MD/MPH, Fisseha Mamo, MPH, Tolawaq Kejela MD Management of severe acute malnutrition in children using community based therapeutic care approach: a review of three years data from southern Ethiopia. Efrem Teferi, MD Shiferaw Teklemariam, MD, MPH , Lopiso Erosie, BSC, MPH , Abel Hailu, MD, Tefera Belachew, MD, MSc, DLSHTM, Mohammed A Yassin, MD, MSc, PhD
Review article Child Survival: Progress Towards meeting MDG4 Assaye Kassie, MD
Case report A Case Report : Optic glioma in a child with NF1 Kalid Asrat, MD
Notes for contributors
Ethiopian Pediatric Society E-mail: [email protected] Telephone: 251-011-8602843
Addis Ababa, Ethiopia
2 Ethiopian Journal of Pediatrics and Child Health The official organ of Ethiopian Pediatric Society The Ethiopian Journal of Pediatrics and Child Health aims to contribute towards the improvement of child health in developing countries, particularly in Ethiopia. The journal publishes original articles, reviews, case reports pertaining to health problems of children. Editorial board Amha Mekasha, MD, Msc Editor-in-chief Assaye Kassie, MD Sirak Hialu, MD Tedbab Degife, MD
3
Table of contents Original articles
The Prevalence of Nosocomial Infections and Associated Risk Factors in Pediatric Patients in Tikur
Anbessa Hospital
Mikyas Demissie ,M.D. , Sileshi Lulsesed, M.D.
Clinical Predictors of Pneumonia Among Under-five Children At Tikur Anbesa Specilaized Hospital Kalid Asrat, MD, Amha Mekasha, MD, MSc Gullian Barre Syndrome in Children At Tikur Anbessa Specialized Hospital
Tigist Bacha, MD
Assessment of quality of care of sick under-five children in referral hospitals in Ethiopia.
Sirak Hailu, MD, Solomon Emyu, MD/MPH, Fisseha Mamo, MPH, Tolawaq Kejela MD Management of severe acute malnutrition in children using community based therapeutic care approach: a review of three years data from southern Ethiopia. Efrem Teferi, MD Shiferaw Teklemariam, MD, MPH , Lopiso Erosie, BSC, MPH , Abel Hailu, MD, Tefera Belachew, MD, MSc, DLSHTM, Mohammed A Yassin, MD, MSc, PhD
Review article Child Survival: Progress Towards meeting MDG4 Assaye Kassie, MD
Case report A Case Report : Optic glioma in a child with NF1 Kalid Asrat, MD
Notes for contributors
4
The Prevalence of Nosocomial Infections and Associated Risk Factors in Pediatric Patients in
Tikur Anbessa Hospital
Mikyas Demissie, MD, Sileshi Lulseged, MD, Msc
Abstract
Little is known about nosocomial infections and associated risk factors among children in Ethiopia.
The aim of the study is to generate data on nosocomial infection in children that will serve as a base for
further studies and to develop prevention interventions. A case control study was done on 111 cases and
222 controls from pediatrics wards of Tikur Anbessa Hospital, Aug 2002-Dec 2003. Nosocomial infection
rate was 5 per 100 discharges. A total of 143 nosocomial infections were detected in 111 cases. The
commonest infection was pneumonia 39.8%. Specimen for culture and sensitivity was taken from
63/143(44.1%) infections and organisms were isolated in 43 infections. And 14 different type of bacteria
were found. E coli, klebsiella pneumoniae and pseudomonas species were the most frequently isolated
organisms. The resistance to ampicillin was 91.9% , gentamycin 67.2%, , ceftriaxone 50%, norfloxacin
18.3%, and ciprofloxacin 15.4%. Age less than one year, malnutrition, admission to orthopedics unit,
peripheral intravenous line and prolonged hospitalization were significantly associated with nosocomial
infection. In conclusion surveillance for high risk patients, education of health personnel, proper isolation
technique, hand washing or use of glove and gowns, avoid prolonged hospitalization when possible and
reestablishment of infection control committee are needed for prevention of nosocomial infection. And
antimicrobial therapy should be guided based on drug susceptibility pattern of bacteria isolated from patients
with nosocomial infection in the hospital.
5
Introduction
Nosocomial infections result in considerable
morbidity and mortality, prolongation of
hospitalization and increases patient care
expenditure (1).Published data about nosocomial
infection rate in Ethiopian pediatric patients are not
currently available. But there were two reports on
outbreak of klebsiella at the Etho-Swedish
Children's Hospital. The first report was on
outbreak of gentamycin-resistant klebsiella
bacteremia between February 1988 and February
1990, and infection control committee was
functioning at that time (2).The second report was
on outbreak of klebsiella oxytoca at Ethio-Swedish
hospital in 1992 and 1993(3).
Nosocomial infection rates in pediatric patients in
other countries range from 1.2 to 10.3 infections
per 100 discharges. In most studies the infection
rate varies by age and ward or service. The
highest infection rate was seen in infants younger
than one year of age (1,4-7).
The predominant site of infection differs by the
population studied and the type of surveillance
performed. Respiratory tract and gastrointestinal
tract infections were common in some studies
and surgical wound infection was common in
other studies. Gram negative bacteria, Gram
positive bacteria, and viruses were incriminated
as the predominant causes of nosocomial
infection in various places (1, 5, 7-9).
Determinants of infection include host factors,
prior invasive procedures, use of catheters, use
of antibiotics and exposure to other patients,
visitors or health care providers with contagious
diseases (10). Nosocomial infections can be
transmitted by contact, common source, air or
vectors. Most nosocomial infections in infants
and children result from contact transmission via
the hands of personnel (1,11-13). Malnutrition is
the most common cause of secondary
immunodeficiency (14-16). The cumulative
probability that an individual will experience at
least one nosocomial infection increases with
increasing exposure to the hospital (1,11).
Depressed level of consciousness increases the
likelihood of reflux of gastric contents and
aspiration into lower airway leading to
Pneumonia (11, 16-19). Antimicrobial therapy
reduces the concentration of normal flora and
6
allowing antimicrobial resistant microorganisms
to gain a foothold. Any patient taking antibiotics
for prophylaxis or treatment may become a
source of highly resistant organisms.
Immunocompromised hosts are more likely to
have nosocomial infection(11,12,18,20,21).
Percutaneously inserted peripheral intravenous
catheters are associated with a low rate of local
infection in children and blood stream infections
are rare. Nosocomial infections also are caused
occasionally by contaminated intravenous fluids
or blood products (1,3,11,22,23) .
Surveillance for infection is the first step in
identifying nosocomial infection and suggesting
methods of prevention (10). Surveillance, by
itself, is an effective process to decrease the
frequency of hospital acquired infections (24).
Methods and Materials
A case control study was conducted reviewing
medical records of children who were discharged
from pediatric wards of Tikur Anbessa Hospital
from August 2002 to December 2003. Neonates
were excluded because the risk factors for
nosocomial infections for neonates are different
from others. Additionally the medical records of
the emergency ward were incomplete and those
who were discharged from the emergency ward
were excluded.
The sample size was calculated with assumption
of prevalence of malnutrition in hospitalized
children in Ethiopia to be 50% which makes
p2=0.5 and p1=0.66. Calculated sample size
was: total sample size=333, Cases=111 and
controls=222 with confidence interval (CI) =95%,
odds ratio (OR) =2 , Zα=1.96 and Zβ=0.84.
Cases are patients with nosocomial infection who
were discharged from pediatric wards excluding
neonates and those who were discharged from
the emergency ward. They were identified after
all charts of patients who were discharged from
August 2002 to December 2003 were reviewed.
For each case the next two discharged patients
without nosocomial infection were taken as
controls from registration book excluding
neonates and those discharged from emergency
ward.
7
Data was collected from charts of cases and
controls using structured format that include age,
sex, primary diagnosis, weight, length,
community acquired infection, the ward and the
service the patient was admitted, procedure that
was done, nosocomial infection by site, culture
and sensitivity result, use of antibiotics, use of
chemotherapy, use of steroid, level of
consciousness, HIV status, duration of stay in
hospital, the status of admitted child at discharge.
Operational Definitions:
Nosocomial infections are infections acquired
during hospital care which are not present or
incubating at admission. Infections occurring
more than 48 hours after admission are usually
considered nosocomial. Infection that is acquired
in the hospital becomes evident after hospital
discharge is considered nosocomial. Definition to
identify nosocomial infections have been
developed for specific infection sites. These are
derived from those published by the Center for
Disease Control and Prevention(CDC) in the
United States of America or during international
conferences and are used for surviellance of
nosocomial infections (24,25).
Short duration nasotracheal or orotracheal
intubation is defined as intubation only during
general anaesthesia while Long duration
nasotracheal or orotracheal intubation is
intubation for respiratory failure(19). Orotracheal
or nasotracheal intubation is all episodes of
orotracheal or nasotracheal intubation with in the
period of one week prior to the onset of
nosocomial pneumonia will be considered.
Nasogastric intubation is the presence of a
nasogastric intubation will be accepted when it
was present for at least two days within the
period one week prior to the onset of nosocomial
infection (pneumonia) (19).
Wasting and stunting were based on calculating
weight as a percentage of reference median
weight for height and height as a percentage of
reference median height for age (26). Nutritional
status for under-five was according to the
welcome classification (27).
Data was entered into SPSS 10.4 statistical
software and it was analyzed using EPI info 6.
After analysis result was presented using
descriptive statistics, chi-square determination,
8
and p-value and odds ratio calculation. P-value
was considered as significant if it was less than
0.05. Multiple regression analysis was done
using SPSS to control the effect confounding
factors.
Results
A total of 1701 patients were discharge from
pediatric wards excluding neonatal ward and
emergency ward in 2003. Of these 85 patients
had acquired nosocomial infection at discharge.
Hence, nosocomial infection rate was 5 per 100
discharges in 2003 in these wards. Charts of 111
cases and 222 controls who were discharged
from pediatrics wards from August 2002 to
December 2003, excluding neonatal and
emergency wards were reviewed analyzed.
Among 111 cases 85 were discharged in 2003
and 26 were discharged in 2002.
Of all cases with nosocomial infection 86(77.5%)
had nosocomial infection once, 21(18.9%) twice,
2(1.8%) three times, and 2(1.8) four times. So,
143 nosocomial infections were detected in 111
cases. The majority of nosocomial infections
were detected in the 2nd week and after the 2nd
week . Out of 143 nosocomial infections 10(7%)
of them were detected in the first three days after
admission, 17(11.9%) between the 4th and 7th
day, 50(35%) between 8th and 14th day,
21(14.7%) between 15th and 21st day and
45(31.4%) nosocomial infections were detected
after 21 days of admission.
The commonest site of nosocomial infection was
pneumonia 39.8% followed by gastroenteritis
11.9% and primary blood stream infection
10.9%(Table 1).The predominant site of infection
was pneumonia in pediatrics, general surgery,
neurosurgery, plastic surgery, tumor therapy unit,
and ENT unit. In burn unit the predominant site of
infection was skin and soft tissue infection. And
in orthopedics upper respiratory tract infection
was the commonest nosocomial infection
detected followed by skin and soft tissue
infection. Surgical wound infection was the
second predominant site of nosocomial infection
in patients who were discharged from general
surgery.
Specimen for culture and sensitivity was taken
from 63/143(44.1%) nosocomial infections.
Among 63 nosocomial infections for which
specimen were collected for culture and
sensitivity test 30 (47.6%) were from blood,
11(17.5%) were from urine and 15(23.8%) were
9
from pus and the remaining 7(11.1%) were from
other body fluids.
The result of culture and sensitivity test was
positive in 43/63(68.3%) of nosocomial
infections. A total of 62 different isolates were
found from those with positive culture result. A
single organism was isolated from each of 27
nosocomial infections and two or more bacteria
were isolated from 16 nosocomial infections.
When the isolates were classified by the type of
organism, 14 different types of bacteria were
isolated. Escherichia coli was the most frequently
isolated organism followed by klebsiella
pneumoniae, pseudomonas species, and
coagulase negative staphylococci as shown in
Table 3.
Drug susceptibility test for gentamycin was done
for 61 of isolates and 41/61(67.2%) were
resistant . And 11/60(18.3%)of isolates were
resistant to norfloxacin . In similar way
18/36(50%) were resistant to ceftriaxone. Drug
susceptibility for cloxacillin was done only for
three isolates. (Table 4)
Infants under one year of age (29days-11mo),
children with moderate wasting (Waterlow
classification), children admitted to orthopedics
unit, peripheral intravenous line, and prolonged
hospitalization were found to have significantly
increased nosocomial infection. And factors such
as sex, weight for age according Wellcome
classification, height for age according to
Waterlow classification, ward, infection at
admission, surgical intervention, urgency of
surgery, type of surgical wound, duration of
operative procedure, duration of stay before
surgical intervention, orothracheal intubation,
nasogastric intubation, urinary catheterization,
bronchoscopy, prior treatment with antibiotics,
chemotherapy or steroid, depressed level of
consciousness, and HIV status were not
significantly associated with the development of
nosocomial infection. Majority of children with
nosocomial infection (66.7%) stayed more than
21 days before discharge when compared to only
7.2% of controls stayed more than 21 days.
(Table 5)
Among 222 controls, 175(78.8%) were
discharged after improvement, 18(8.1%) were
discharged in the same or worse condition, and
29(13.1%) died. Of 111 cases, 77(69.4%) were
discharged after improvement, 13(11.7%) were
discharged in the same or worse condition, and
10
21(18.9%) died. Children with nosocomial
infection had no significant increased risk of
death when compared to those without
nosocomial infection ( p value=0.157).
Discussion
Nosocomial infection rate in this study was with
in the range which was observed in other studies
(1, 4, 6, 7). The predominant site of nosocomial
infection was pneumonia followed by
gastroenteritis, and primary blood stream
infection. This was similar to the study done
Welliver and Mc Laughlin. (7).Though 49(44%) of
patients with nosocomial infection were admitted
to surgical service in this study, surgical wound
infection was not common. It accounted for only
5.6% of all nosocomial infections.
Specimen for culture and sensitivity was not
taken in 80(55.9%) of nosocomial infection and
they were treated empirically. But isolated
organisms are usually resistant to commonly
used antibiotics for nosocomial infection like
gentamycin and ceftriaxone. According to this
study, it does not seem rational to treat
nosocomial infection empirically with the above
mentioned antibiotics in Tikur Anbessa hospital.
Antimicrobial therapy should be guided based on
drug susceptibility pattern of organisms isolated.
The proportion of isolated bacteria resistant to
floroquinolones was less when compared to the
proportion of resistance to other antibiotics.
Changing or rotating standard group of antibiotics
used for empiric therapy has been efficacious in
limited studies. The role of floroquinolones in the
treatment of serious infections in children does
not appear to be compromised by safety
concerns since arthralgia and quenolone-induced
cartilage toxicity were low and episodes of
arthralgia were mostly reversible (12,28).
In this study Gram negative bacilli were
commonly isolated organisms which constituted
46/62(74.2%) of all isolated bacteria. In reports
from the 1960s and 1970s from USA Gram
negative bacteria accounted for more than 50%
of the infections which is similar to our study
(1,5,7-9). It is not possible to differentiate
whether the isolated coagulase negative
staphylococci were contaminant or etiologic
agent from this study since only one blood
culture sample was taken and details of blood
culture and clinical response to treatment was
not available. (10,25).
11
Disruption of the physical barrier occurs in burn
patients and in those with degloving injury.
Admission to orthopedics unit was significantly
associated with nosocomial infection in this
study. But admission to burn unit in the hospital
was not significantly associated with nosocomial
infection. (10,12). Increased risk of nosocomial
infection in infants less than one year of age and
in nutritionally compromised patients was seen
in other studies as it was seen in this study (1, 5,
7, 14-17). Children admitted to tumor therapy
unit and those who took anti neoplastic
chemotherapy have no significantly increased
nosocomial infection unlike other studies (18).
The effect of peripheral intravenous line on the
occurrence of nosocomial infection needs further
prospective study since canula related
septicemia is considered when the same
organism is isolated from canula and blood.
Even though increased mortality is expected in
patients with nosocomial infection, mortality of
patients with nosocomial infection was not
increased in this study when compared to the
control group (1, 11, 23).
Effective targeted surveillance for high risk
patients, staff education, use of proper isolation
techniques and effective infection control practice
such as hand washing before and after patient
contact or use of glove and gowns are
recommended for the prevention of nosocomial
infection. Prolonged hospitalization should be
avoided as much as possible since it is
significantly associated with the development of
nosocomial infection. And reestablishment of
infection control committee is also required.
12Table 1. Distribution of nosocomial infection by site , Tikur Anbessa hospital, August 2002-December 2003.
.Site of nosocomial infection frequency %
Pneumonia 57 39.9
Gastroenteritis 17 11.9
Primary blood stream infection 15 10.9
Urinary tract infection 13 9.1
Skin and soft tissue infection 13 9.1
Upper respirator tract infection 9 6.3
Surgical wound infection 8 5.6
Central Nervous System infections 2 1.4
Others 7 4.9
Systemic infection 2 1.4
Total 143 100
Table 2. Site of nosocomial infections by the service in Tikur Anbessa hospital, August 2002-December 2003.
Service Site of nosocomial infection
Tot
al
SWI (%)
UTI (%)
URTI (%)
Pneumonia (%)
Skin & soft tissue (%)
primary blood stream infection(%) G
astr
o
Ent
eriti
s (%
)
CN
S(%
)
Oth
er(%
)
Sys
tem
ic
infe
ctio
(%)
(%)in
fect
ion
infe
ctio
m
Pediatrics 0 8 (18.2)
0 27 (61.4)
1 (2.3)
5 (11.4)
4 (9)
0 0 0 44
General surgery
5 (26.3)
2 (10.5)
0 7 (36.8)
1 (5.3)
0 2 (10.5)
1 (5.3)
1 (5.3)
0 19
Neuro surgery
0 0 1 (10)
3 (30)
2 (20)
0 1 (10)
1 (10)
1 (10)
0 10
Plastic surgery
1 (50)
0 0 1 (50)
0 0 0 0 0 0 2
Burn unit 1 (7.1)
1 (7.1)
0 3 (21.4)
4 (28.6)
2 (14.3)
1 (7.1)
0 0 1 (7.1)
14
Tumor therapy
0 1 (3.2)
1 (3.2)
11 (35.5)
1 (3.2)
8 (25.8)
6 (19.3)
0 3 (9.7)
0 31
ENT 0 0 1 (16.7)
3 (50)
0 0 1 (16.7)
0 1(16.7)
0 6
Orthopedics 1 (5.9)
1 (5.9)
6 (35.3)
2 (11.8)
4 (23.5)
0 2 (11.8)
0 0 1 (5.5)
17
Total 8 13 9 57 13 15 17 2 7 2 143
SWI=surgical wound infection UTI=urinary tract infection URTI= upper respiratory tract infection CNS= central nervous system
Table 3 Types of organism isolated from children with nosocomial infection in Tikur Anbessa hospital, August 2002-December 2003.
Organism isolated Frequency %
Escherichia coli 11 17.7
Klebsiella pneumoniae 9 14.5
Pseudomonas species 9 14.5
Coagulase negative staphylococci 6 9.7
Acinetobacter species 5 8.1
Klebsiella oxytoca 4 6.5
Salmonella species 4 6.5
Staphylococcus aureus 3 4.8
Proteus vulgaris 3 4.8
Citrobacter species 3 4.8
Shigella species 2 3.2
Group A streptococci 1 1.6
Morganella morgagne 1 1.6
Enterobacter cloaca 1 1.6
Total 62 100
14
14
Table 4 Drug susceptibility pattern of bacteria isolated from children with nosocomial infections in Tikur Anbessa Hospital, August 2002- December 2003.
Antibiotics No. of isolates tested
sensitive Intermediate sensitive
Resistant
Ampicillin 62 4(6.5%) 1(1.6%) 57(91.9%)
Gentamycin 61 20(32.8%) 0 41(67.2%)
Norfloxacin 60 47(78.3%) 2(3.3%) 11(18.3%)
Trimethoprim-Sulfamethoxazole
58 10(17.2%) 2(3.5%) 46(79.3%)
Tetracycline 58 13(22.4%) 2(3.5%) 43(74.1%)
Chloramphenicol 55 8(14.5%) 0 47(85.5%)
Ceftriaxone 36 13(36.1%) 5(13.9%) 18(50%)
Augmentin 31 8(25.8%) 5(16.1%) 18(58.1%)
Doxycycline 25 5(20%) 1(4%) 19(76%)
Amoxicillin 24 2(8.3%) 1(4.2%) 21(87.5%)
Naldixic Acid 22 12(54.5%) 0 10(45.5%)
Nirofurantoin 19 6(31.6%) 1(5.3%) 12(63.1%)
Amikacin 14 11(78.6%) 1(7.1%) 2(14.3%)
Ciprofloxacin 13 9(69.2%) 2(15.4%) 2(15.4%)
Penicillin G 13 0 2(15.4%) 11(84.6%)
Erythromycin 13 4(30.8%) 3(10%) 6(46.2%)
Methicillin 11 2(18.2%) 3(27.3%) 6(54.5%)
Carbencillin 8 2(25%) 3(37.5%) 3(37.5%)
Kanamycin 4 0 2(50%) 2(50%)
Cephalotin 4 0 2(50%) 2(50%)
Cloxacillin 3 0 2(66.7%) 1(33.3%)
Lincomycin 3 1(33.3%) 2(66.7%) 0
15
15
Table 5 Risk factors associated with nosocomial infection in pediatric patients, Tikur Anbessa Hospital, August 2002-December 2003.
Variables Control Case OR P value Adjusted OR (95% CI)
Number % Number %
Age in month 1-11 35 15.8 31 27.9 2.21(1.03-4.79) 0.042 6.373(2.16-18.83)
12-59 85 38.3 33 29.7 0.97(0.48-1.97) 0.937 0.741(0.296-1.856)
60-119 52 23.4 27 24.3 1.30(0.61-2.76) 0.577 1.155(0.421-3.169)
120-180 50 22.5 20 18 1
Weight for height
>=90% 143 66.2 58 54.7 1
80-89% 44 20.4 21 19.8 1.18(0.62-2.24) 0.709 0.94(0.41-2.18)
70-79% 22 10.2 19 17.9 2.13(1.02-4.46) 0.0447 3.68(1.46-9.30)
<70% 7 3.2 8 7.6 2.82(0.88-9.13) 0.09 2.28(0.52-10.04)
Peripheral intravenous line
No 23 10.4 3 2.7 1
Yes 199 89.6 108 97.3 4.16(1.15-17.82) 0.025 7.566(1.009-56.708)
Use of antineoplastic chemotherapy
No 215 96.8 89 80.2 1
Yes 7 3.2 22 19.8 7.59(2.94-20.35) 0.000001 0.006(0.00-12)
Duration of stay before discharge
<7 days 74 33.3 2 1.8 1
8-14 days 62 27.9 19 17.1 11.34(2.4-73.44) 0.00032 13.43(2.65-68.08)
15-21days 40 18 16 14.4 14.8(3.02-98.38) 0.00005 9.29(1.74-48.64)
>21 days 46 7.2 74 66.7 59.52(13.41-368.48)
0.00000 70.15(13.96-352.46
Service General surgery 72 32.4 16 14.4 1
Pediatrics 90 40.5 38 34.2 1.9(0.94-3.89) 0.079 1.68(0.42-6.71)
Plastic surgery 10 4.5 2 1.8 0.90(0.0-5.09) 0.785 1.21(0.19-7.56)
ENT 14 6.3 4 3.6 1.29(o.31-5.001) 0.945 1.83(0.36-9.21)
Neurosurgery 6 2.7 5 4.5 3.75(o.85-16.41) 0.096 2.89(0.53-15.59)
Orthopedics 21 9.5 15 13.5 3.21(1.26-8.27) 0.012 4.86(1.31-18.01)
Burn unit 4 1.8 7 6.3 7.88(1.77-37.27) 0.003 4.89(0.93-25.83)
Tumor therapy unit
5 2.3 24 21.6 21.6(6.47-77.05) 0.000 1778.85(0.00-34)
16
16
References 1. Jarvis W.R : Epidemiology of nosocomial infections in pediatric patients. Pediatr. Infect. Dis. J. 1987; 6: 344-51 2. Moss W. An outbreak of getamycin-resistant klebsiella bacteremia at a children's hospital. Ethiop Med J. 1992;
30: 197-205 3. Worku B. Klebsiella oxytoca outbreak at Ethio-Swedish children’s hospital (ESCH). Ethiop Med J. 1997; 35:177-
83 4. Cooper R.G., Sumner C. Hospital infection data from a children’s hospital Med J. Aust. 1970; 2:1110-13 5. Gardner P, Carles D.G Infection acquired in a pediatric hospital. J. Pediatr. 1972; 81: 1205-1210 6. Wenzel R.P, Osterman C.A., Hunting K.J. Hospital acquired infection 11. Infection rates by site, service and
common procedures in a university hospital. Am. J. Epidemiol. 1976; 104: 645-51 7. Welliver R.C., Mc Laughlin S. Unique epidemiology of nosocomial infection in a children’s hospital. Am. J. Dis.
Child. 1984; 138: 131-5 8. Gaynes R.P., Edward J.R., Jarvis W.R., Culver D. H, Tolson J. S., Martone W. J. Nosocomial infection among
neonates in high- risk nurseries in the United States. Pediatrics. 1996; 98: 357-61 9. Sohn A. H., Garret D. O., Sinkovitz- Cochran R.L., Grohskopf L. A., Levine G. L., Stover B.H., Seigel J.D.,Jarvis
W.R. Prevalence of nosocomial infections in neonatal intensive care unit patients: Results from the first national point prevalence survey.J.Pediatr .2001;133:821-827.
10. Behrman R, Kleigman R, Jenson H. T lymphocyte, B lymphocyte and Natural Killer cells.-In: Buckley R., ed ; Infection control and prophylaxis.-In: Fisher, ed. Nelson, Text Book of pediatrics, 17th ed. New Delhi, Elsevier, 2004:683, 866, 1184.
11. Feigin, Cherry. Nosocomial infections.-In: Huskin W.C., Goldman D.A., eds., Textbook of pediatric infectious disease, 4th ed. Philadelphia, Elsevier, 1998; 2545-85.
12. Fleming C.A., Balaguera H.U., Craven D.E., Risk factors for nosocomial preumonia.Med chinics of north America 2001;85:1545-1563.
13. Knittle M.A., Eitzman D.V., Baer H, Role of hand contamination of personnel in the epidemiology of gram negative nosocomial infections: J.pediatr 1975; 86:433-437.
14. Chandra R.K: Nutrition, immunity, and infection: Present knowledge and future direction.Lancet.1983; 1:688-91. 15. Puri S, Chandra P.K: Nutritional regulation of host resistance and predictive value of immunlogic test in
assessement of outcome. Pediatric Clinics of North America.1985; 32:499-516. 16. Bhattacharyya N, Kosloske A.M., Macarlhia C.Nosocomial infection in pediatric surgical patients: a study of 608
infants and children: J.pediatr.Surg.1993; 28:338-343. 17. Bhattacharyya N, Koslosk A.M., postoperative wound infectious in pediatric surgical patients: A study of 676
infants and children. J.pediatr surg.1990; 25:125-129. 18. Haley R.W., Houton T.M., Culver D.H., Stanley R.C., Emori T.G., Hardisan C.D., Quade D, Schauchtman
R.H.Schaberg D.R., shah B.V., Schat, G.D., Nosocomial infections in U.S. Hospitals, 1975-1976 estimated frequency by selected characteristics of patients AM J Med 1981; 70: 947-959.
19. Calis R, Torres A, Gatell J.M., Almale M, Rodringuez-Roisin R.,Aqustin vidal A. Noscomial preumonia. A multivariate analysis of risk and prognosis. Chest 1988; 93:318-24.
20. Price D.J., Sleigh J.D., Control of infection due to klebsiella aerogenes in a neurosurgical unit by withdrawal of all antibiotics. Lancet 1970; 2:1213-1215.
21. Young L.S. Nosocomial infection in the immuocompromised adult. Am. J .Med 1981; 70:398-403.
17
17
22. Garlsand I.S.,Nelson D.B., Cheah T.E., Hennes H.H., Johnson T.M. infectious complications during peripheral intravenous therapy with Teflon catheters: a prospective study.pediatr infect dis. J. 1987; 6: 918-921.
23. Tully J.L., Friedland F.H., Baldini L.M., Goldmann D.A., Complications of intravenous therapy with steel needles and Teflon catheters. Am. J Med 1981; 70:702-6.
24. Ducel.G, Fabry.J, Nicolle.L; Prevention of hospital acquired infections A Practical Guide 2nd edition.WHO/CDS/EPH/2002.12
25. Garner J.S., Jarvis W.R., Emori T. G., Horan T.C. , Hughes J .M.. CDC definition for nosocomial infections. Am J Infect Control 1988;16:128-140.
26. Waterlow. J.C. Note on the assessement and classification of protein energy malnutrition in children. Lancet 1973; 2:87-98.
27. Classification of infantile malnutrition. Lancet 1970; 2: 302-303. 28. Grady R., Safety profile of quinolone antibiotics in the pediatric population. Pediatr Infect Dis J 2003; 22: 1128-
32.
Acknowledgements I thank Dr Fithun Lulseged and staff at the statistics office who helped me collect data. I also extend my gratitude to Dr
Adamu Adissie and Dr Firew Mekonnen who helped me in the analysis of the data.
18
18
Guillain Barre Syndrome in Children At Tikur Anbessa Specialized Hospital
Tigist Bacha, MD
Abstract This is a six years retrospective descriptive study conducted in the pediatric and child health department of Tikur Anbessa
Hospital from September, 2001 – September, 2006 G.C to assess the pattern of acute flaccid paralysis (AFP) and the
clinical and epidemiologic features of Guillain Barre syndrome (GBS).
Data was collected from medical records of all patients admitted with diagnosis of AFP, and analyzed using standard
statistical tests with SPSS version 14 software. Out of 70 admitted cases of AFP, forty six (65.7%) were males and 24
(34.3%) were females. Sixty seven cases (95.7%) were diagnosed to have GBS and the rest three were compatible with
poliomyelitis, transverse myelitis and post injection neuritis. Out of the 70 cases 66 (94.3%) have received at least one
dose of polio vaccination and the rest 4 (5.7%) were never vaccinated.
Out of the cases diagnosed to have GBS, 44 (65.7%) met NNCDS diagnostic criteria. History of antecedent event was
obtained in 31/67 (46.3%) patients. Majority of the patients 45 (67.17%) presented with ascending reflexes quadriparesis,
2 (2.98%) patients with descending areflexic quadriparesis, 19(28.35%) only with lower limb involvement and 1 (1.5%)
with typical miller-fisher type. Sensation was affected in 4 patients.
Cytoalbuminological dissociation was found in 27(40.3%). There were 11 deaths (16.4%) of whom five were admitted to
ICU the rest six didn’t. This study showed that the commonest cause of AFP is GBS which is associated with high
mortality. This high mortality rate 11/67 (16.42%) is attributed to absence of pediatrics ICU, late arrival to Hospital after
onset of illness, and poor supportive care.
19
19
INTRODUCTION
Acute flaccid paralysis (AFP) is the clinical condition
diagnosed in any child under 15 years of age with
acute floppiness of one or more limbs or any age in
whom clinician suspect polio (1,2).It is caused by many
conditions including Guillain Barre syndrome (GBS) ,
poliomyelitis, transverse myelitis and metabolic
neuropathy like hypokalemia (3,4).The most common
cause of AFP is GBS followed by transverse myelitis
(2,5,6,7)
In Ethiopia, the polio eradication initiative (PEI) was
started in 1996. Surveillance for acute flaccid paralysis
(AFP) was initiated in May, 1997. AFP surveillance is
the detection of at least one AFP case per 100,000
children under 15 years of age. AFP surveillance
depends on immediate reporting, investigation of AFP
cases, routine monthly reporting of cases including zero
reporting (1, 9).
GBS is an acute inflammatory polyneuropathy. The
cause is unknown but autoimmunity is incriminated.
Unlike polio, GBS is usually symmetrical (asymmetrical
in 9%), and fever at presentation is not present.
Paralysis develops acutely over days, or at most weeks.
After brief plateau the patients’ improvement begins with
gradual resolution that lasts from weeks to months. 50%
has bulbar involvement, 33% of them require ICU
admission 25% required mechanical ventilation 5-10%
mortality rate (2-3% in best ICU) and 80% complete
recovery. Treatment is intensive care support; early
intravenous immunoglobulin (IVIg) therapy and plasma
exchange hasten early recovery. Corticosteroid alone
outcome is controversial (3,8,10).
The objectives of the study are to describe pattern of
AFP, determine the most common cause of AFP, and
determine the clinical and epidemiological feature of
GBS.
Material and method
Retrospective descriptive study was done in pediatric
and child health department of Tikur Anbessa Hospital,
Addis Ababa. Study groups were all children admitted
with the case definition of AFP from September 2001 –
September 2006. The medical report of Statistics Office
registered for AFP surveillance, Pediatric wards,
surgical and medical ICU and neurology clinic
registration books were reviewed to trace all cases. This
hospital is the only pediatric tertiary Hospital serving for
all the country.
In this study:
Antecedent triggering event was defined as the
presence of respiratory, gastrointestinal, febrile illness
or vaccination for the previous 4 weeks.
Polio-compatible case is defined as a case in which
one adequate stool specimen was not collected from a
probable case within 2 weeks of the onset of paralysis,
and there is either an acute paralytic illness with polio-
compatible residual paralysis at 60 days, or death takes
place within 60 days, or the case is lost to follow-up.
The diagnosis of GBS in this study is based on National
Institute of Neurological and Communicative Disorders
and stroke (NNCDS) diagnostic criteria. (11)
1. Progressive weakness of more than one limb
due to neuropathy,
2. Areflexia or hyporeflexia
3. Duration of progress less than 4 weeks,
4. The absence of a sharp sensory level on the
trunk,
5. The absence of other causes of acute
neuropathy
6. Less than 50 mononuclear leukocytes per mm3
of CSF.
The data were transferred to a structured form which
was then entered into computer data base. SPSS
Version 14, 2005 was used to process the statistical
data.
Results
In this six year period 70 AFP cases (46 male and 24
female) were admitted and majority 67 (95.7%) were
GBS. Three cases were suspected to have a
compatible poliomyelitis, transverse myelitis and post
injection neuritis each accounting 1.4%. The mean age
of AFP is 5.82 year with a range of 1-12 year. Cases
were reported from different regions of Addis Ababa 44,
Oromia 12, Amhara 11 and South Nations Nationalities
and Peoples region (SNPPR) 3 (Fig 1). The distribution
over the years is shown in Fig 2 from 2001-06 (1993-
1998 Eth C). 1More cases were seen in 1996 Eth C.
Stool sample was taken for polio in 65 cases and was
not taken in 5 of them. Out of the 70 cases 66 (94.3%)
of the cases got at least one dose of polio vaccination
and the rest 4 (5.7%) never vaccinated (table 1). Prior
injection history is found in 7 (1%) of the patients.
Addis AbabaSNNPROromiyaAmhara
Region
50
40
30
20
10
0
NO
of
AF
P C
ases
21
21
Fig. 1 AFP case Regional distributions Fig 2. AFP cases distribution per year
Table 1. Cranial nerve involvement
Cranial nerve involved Frequency Percentage
Facial nerve 3 27.2
Opthalmolopelgia 1 9.1
Glossopharyngeal nerve 5 45.4
Vagus nerve 1 9.1
Multiple cranial nerve 1 9.1
Total 11 100
199819971996199519941993
Admition Year
25.0%
20.0%
15.0%
10.0%
5.0%
0.0%
AF
P c
ases
in P
erce
nt
22
22
Table 2: CSF analysis Result
Frequency Percent
CSF cell count >10/cumm and protein >45mg
3 4.5
CSF cell count >10/cumm and protein <45mg
7 10.4
CSF cell count <10/cumm and protein >45mg
27 40.3
CSF cell count <10/cumm and protein <45mg
21 31.3
Not Documented 9 13.4
Total 67 100.0
From those diagnosed as having GBS 44(62.8%) met
NNCDS diagnostic criteria. The mean age was 5.9 with
ranges of 15 months -12 years. Most of them were male.
History of antecedent event was obtained in 31(46.3%)
patients. Respiratory symptoms accounted 19(27.1%),
gastrointestinal symptoms 9 (11.4%), both gastrointestinal
and respiratory symptoms 1 (1.4%), 1 (1.4%) post
vaccine (Rabies), and 1 (1.4%) had prior malarial attack.
The pattern of progression of paralysis was ascending
areflexic quadriparesis in 45 (67.2%) patients, descending
areflexic quadriparesis in 2(3%), patients with only lower
limb involvement were 19 (28.3%) and one (1.4%) was
Miller - Fisher variant. The mean interval between onset
of symptoms to hospital admission was 6 days with a
range of 5 hrs-14days. Rapidity of the progression ranged
from 1 day to 14.0 days the mean being 3.44 days. The
mean duration of hospital stay is15.5days. Out of the 17
patients whose blood pressure was measured 12 patients
had normal measurement and 5 had transient
hypertension. Bladder and bowel dysfunction was
reported in 44 out of 64 documented cases. Cranial nerve
palsy was reported in 11 patients. As shown in table xxx
the commonest were cranial IX (45.4%) followed by facial
nerve (27.2%). Multiple cranial nerve involvement was
found in 1(9.1%) patient. Sensation was affected in 4
(6%) cases. Ataxia was documented in 8(11.9%) patients.
Cytoalbuminological dissociation was found in 27(40.3%)
patients. see table 5. EMG was done in 5 patients with
GBS out of which 2 were demyelinating, 2 axonal and 1
mixed axonal and demyelinating.
Fifteen15/67 (22.39%) patients required ICU care out of
which 6 didn’t get the service. There were 11(16.4%)
deaths. Five died in ICU and 6 patients died in the pediatric
ward. Four of them came within 4 days of onset of illness
and 7 of them after 4 days. Respiratory failure was
considered as a cause of death in 6, Respiratory failure and
infection were cause of death in 5 of them (3 had
pneumonia, 1 urosepsis). Specific treatment such as
plasmapheresis and Immunoglobulin was given to none of
the patients. Prednisolon was given only for 4 patients
(1.0%).HIV screening was done only for one patient and
the result was non reactive.
Discussion
This is the first study done on pattern of AFP and clinical and
epidemiology of GBS in the pediatric age groups. The
predominance of male is similar to other studies (6-10). The
common AFP identified was GBS which also correlates with
other studies in Australia, Bangladesh, and Honduras (2,6-8).
Regarding GBS the antecedent events are lower in this study
than other studies done in Kenya, Tanzania, Nigeria and
including the study done in this hospital in adults (12-15).
Similar to other studies respiratory symptoms were the
commonest antecedent events (12). The cranial nerve
involvement are common findings similar to that of Kenya
and adult Ethiopian patient’s (12, 13).
In this study one patient gave history of antecedent malarial
attack the species not documented. From other studies the
development of Guillain–Barré syndrome was reported in 10
patients who had had acute P. falciparum malaria during its
seasonal exacerbation is reported (16).
The high mortality rate is higher to the report from other
studies done in Kenya (13) and Tanzania (9). Also a
higher mortality was observed in adult study done in the
same hospital (12). The main attributable cause is lack of
good pediatric ICU in this Hospital
which could be compounded by the late appearance of
patients to the hospital. In addition specific therapies such
as immunoglobulin and plasmapheresis are not available in
the setup.
This study showed a high need of having pediatric ICU. We
can see the mortality rate from GBS is high some of which
could have been prevented if there was a pediatric ICU.
The care of critical patients and universal precautions for
infection prevention should be encouraged. In addition the
use of specific treatments such as plasmaphereis and
immunoglobulin should be introduced.
For five cases stool sample for polio was not taken out of
whom four died. Taking specimen as soon as possible
especially on critical patients, including on weekend may
increase the surveillance. Improving the AFP surveillance
increased by case investigation reporting cases as early as
possible especially on critical patients should be
encouraged.
In addition care seeking behaviour should be improved so
that patients are brought to hospital as early as possible.
Finally a further large scale study with more investigation
modalities should be done in the near future.
24
24
References
1. Acute Flacc
2. id paral
3. ysis surveillance: Ministry of Health and WHO Ethiopia June 2006: 5
4. Koul R, Chacko A, Javed H et al Acute flaccid paralysis in Australian children. . J Paediatr Child Health.
2003; 39 (1):22-6.
5. Hahn AF. Guillain-Barre syndrome. Lancet 1998; 352: 635-41.
6. Lovecchio F, Jacobson S. Approach to generalized weakness and peripheral neuromuscular disease.
Emerg Med clin N am 1997; 15(3):605-23
7. Rehman A, Idris M, Elahi M, Arif A. Guillain-Barre syndrome. The leading cause of Acute Flaccid Paralysis
in Hazara Division. J ayub Med Coll Abbottabad 2007;19:26-28.
8. Rasul CH, Das PL, Alam S, Ahmed S, Ahmed M. Clinical profile of acute flaccid paralysis. Med J Malaysia.
2002 ; 57 (1):61-5.
9. Molinero MR, Varon D, Holden KR, Sladky JT, Molina IB, Cleaves F. Epidemiology of childhood Guillain-
Barre syndrome as a cause of acute flaccid paralysis in Honduras: 1989-1999.
J Child Neurol. 2003; 18(11):741-7..
10. Kuwabara S. Guillain-Barre syndrome: epidemiology, pathophysiology and management. Drugs.
2004;64(6):597-610. .
11. Berhane Beyene, Ayele Gebremariam, Tilahun Teka et al. Laboratory and Epidemiology communications,
Regional Distribution of Acute Flaccid Paralysis Cases in Ethiopia in 2000 – 2002. Jpn.J.Infect. Dis., 2004;
52:. 72.
12. Harvey B.Sarnat. Neuromuscular Disorders. In: Richard E. Behrman, Robert M. Kliegman and Hal B.
Jenson, Editors. Nelson Text Book of Pediatrics, 17th Edition .Philadelphia Pennsylvania, 2004:2080-81.
13. Asbury Ak. Assessment of current diagnostic criteria for Guillian-Barre syndrome. Ann Neurol
1990:27(suppl):s21-4
14. Zenebe Melaku, Guta Zenebe, Abera Bekele .Guillaian barre syndrome in Ethiopian patients. Ethiop Med J
2005; 43( 1) :21
15. Bahemuka M. Guillain-Barre syndrome in Kenya: a clinical review of 54 patients.
J Neurol. 1988; 235(7):418-21.
16. Howlett WP, Vedeler CA, Nyland H, Aarli JA. Gullian-Barrre syndrome in Northern Tanzania: a comparison
of epidemiological and clinical findings with western Norway. Acta Neurol Scand 1987:75:95-100.
17. Osuntookun BO, Agebebi K. Prognosis of Guillain–Barre syndrome. Medicine in the African Region: The
Nigerian experience. Neurol Neurosurg Psychatry 1973:36:478-83.
25
25
Acknowledgments:
I extend my thanks to Dr. Ahmed Bedru for his invaluable advice in the conduct of the research. Ato Tilahun
Zimita for his assistance in data entry and analysis.
26
26
ASSESSMENT OF QUALITY OF CARE OF SICK UNDER-FIVE CHILDREN IN REFERRAL HOSPITALS IN ETHIOPIA
Sirak Hailu, MD1, Solomon Emyu, MD/MPH2, Fisseha Mamo, MPH3, Tolawaq Kejela MD4
ABSTRACT
Background: About 10-20% of sick children presenting to a primary care facility require referral to hospital for inpatient care. Improvement of the quality of pediatric referral care has a major contribution to the child survival efforts by ensuring the continuum of care and averting mortality. Objective: To assess the quality of care for children in selected referral hospitals based on the minimum standards derived from the “WHO Pocket book of Hospital Care for Children, 2005” and thereby to initiate pediatric referral care quality improvement process in the country. Methods: A qualitative assessment of pediatric referral care was conducted in 8 hospitals selected by convenient sampling, January – July 2008. A team composed of experienced pediatricians and health officer used an adapted WHO hospital assessment tool to assess the quality of triage, emergency care and case management practices & hospital infrastructure and support services. Results: None of these hospitals were practicing the standard triaging process by assessing children immediately on arrival for emergency and priority signs. All of them were not appropriately organized and fully equipped to handle pediatric emergencies effectively. Overall, the case management of common neonatal and childhood illnesses was not optimal. Generally, there was shortage of some essential drugs and lack of materials such as nasal prongs, infant and child size bag & masks, nebulizers, heaters and oxygen concentrators. Hygienic facilities were below the expected standard. Staff were not trained in ETAT (Emergency Triage Assessment and Treatment) and there were no protocols for pediatric referral care. There was no clearly designated high dependency area where very sick children receive highest attention and no special rooms for providing appropriate neonatal care in majority of the hospitals. The overall case fatality rate was 11% (10-16%) but first 24 hours mortality could not be determined due to problems with the recording system.
Conclusions: The quality of pediatric referral care needs serious attention and coordinated efforts utilizing the opportunity of the national hospital management initiative and the BPR (Business Process Re-engineering) to institutionalize ETAT and standards of hospital care for children. This has to be complemented with availing of appropriate job aids, essential supplies and equipments, and improvement of health worker skills through training, clinical mentoring and regular supportive supervision.
1 WHO/Ethiopia, 2 WHO/Ethiopia, 3 FMOH/Addis Ababa, 4 Medical Faculty/AAU,
27
27
INTRODUCTION
The Ethiopian health service delivery system is organized as a four-tier system, characterized by a Primary Health Care Unit (PHCU) – 1st tier, then the district hospital – 2nd tier, zonal hospital – 3rd tier and specialized hospital- 4th tier. Services given at each level of these tiers have a crucial role in averting the morbidity and mortality burden of children and contributing a lot to the achievement of healthy society. In relation to this, Ethiopia has developed a comprehensive national child survival strategy, which is part of the national HSDP-3, and is implementing the IMNCI approach at a wide scale at health facility and community levels. Currently the national IMNCI coverage of the country at Health Center levels is about 60%. These interventions at a community and frontline health facilities (PHCU) levels are very important to address the majority of the health problems of children and also to make the services close and easily available to the society. The IMCI/IMNCI strategy seeks to strengthen prevention and care for children through appropriate community and household care, primary care, referral
practices, and care at the first-level hospital. On the basis of current guidelines, it has been estimated that 10% to 20% of sick children who present for primary care (i.e., the most severely ill) require referral to a first referral or district hospital. The quality of care provided in these hospitals is likely therefore to have a major impact on the health and lives of millions of children each year. It is estimated that 40-60% of deaths at the referral hospitals occurs in the first 24 hours of admission. Many of these deaths could be prevented if very sick children are identified soon after their arrival in the health facility, and treatment is started immediately. The Emergency Triage Assessment and Treatment (ETAT) tool is designed to enable health workers triage all sick children when they arrive at a health facility, into those with emergency signs, with priority signs, or non-urgent cases. It also enable them to provide emergency treatment for life-threatening conditions. The standards of care of the ETAT guidelines correspond to the minimum standards that should be maintained even in small hospitals and where resources are limited.
Early assessment and prioritization for management of sick children attending a health service are critical to achieving good health outcomes. Experience of Malawi in implementing ETAT showed a reduction of total inpatient mortality from 10-18% (median 12.4%) to 6-8% (median 5.7%) over a 2 year period (2001-
2003) and deaths within 24 hours of admission from 36% to 12.5%. This survey tries to assess the current situation of the quality of pediatric care in selected government hospitals.
OBJECTIVES & METHODOLOGY The objective of this study was to assess the quality of care for children in selected referral hospitals based on the minimum standards derived from the “WHO Pocket book of Hospital Care for Children, 2005” and to initiate pediatric referral care quality improvement process by identifying key areas that need immediate and long term action. Eight hospitals (Adama, Ambo, Bishoftu, Debre Birhan, Dessie, Yekatit 12, Yirgalem and Zewditu hospitals); one district, six zonal and one regional referral hospital, were included in the assessment using convenient sampling. The main criterion used for including a hospital was having functional pediatrics outpatient and inpatient services. WHO’s generic assessment tool, “Assessment of the quality of care for children in hospitals”, was
adapted and used for data collection. Adaptation of the tool was done by practicing pediatricians, general practitioners and nurses and it was pretested during a five days national orientation workshop on pediatric referral care. The assessment tool had 12 major sections: General hospital information, Hospital layout and structure, Hospital support systems (drugs, equipment and laboratory), Emergency care, Pediatric wards (layout, facilities, staff, supplies & equipments), Case management in the ward (Cough or difficult breathing, Diarrhoea, Fever conditions, Malnutrition, HIV/AIDS), Supportive care & nutrition, Monitoring of patients, Neonatal Care (layout and staff, Routine neonatal care and Sick newborn care), Paediatric surgery and rehabilitation, Hospital administration and Access to hospital.
28
28
The adapted assessment tool was used to collect information from all hospitals, January to July 2008. Data was collected by two teams each composed of a pediatrician and a health officer who were trained and also involved in the adaptation of the data collection tool. Four different methods of data collection were used namely; hospital visit, case management observations, records review and interviews of caretakers and providers which took two full days for assessing each hospital. Hospital visit: Direct observation of the hospital layout and structure, OPD attendance, admission rates, availability of essential drugs, availability of diagnostic and therapeutic equipments were made. Areas of doubt were clarified
by interviews. Case observations: The care for admitted children to the hospital was observed without interference from the assessors. This is complemented by discussion of the cases with staff, review of the case records and monitoring charts, and interviewing the caretakers. Records review: Assessors obtain information on the quality of care for admitted and recently discharged patients by checking records. If there are not sufficient patients for direct case observations, assessors reviewed at least 3-5 records for each clinical problem. Interviews: Assessors conducted interviews with hospital staff and caretakers to gain some idea of their perception of care for children in the hospitals.
Each major assessment section consisted of a number of standards and each standard was qualified by several detailed criteria. Thus, each of the 12 sections of the assessment tool was scored based on the standards and the criteria to meet these standards. The standards are the minimum requirements for good quality of care for children as defined by the WHO “Pocket book of Hospital Care for Children, 2005”. The detailed criteria of each standard were rated/scored based on the need for improvement as “good” or “need to be improved”, where “Good” means the criteria is similar to the
standard, and “To be improved” means the component is below the expectation of the WHO standard and needs improvement. A summary score from 5 to 1 was awarded at the end of each section whereby 5 indicates “Good’’ practice complying with standards of care while scores from 4 to 1 indicate practices “To be improved” (4 = Little improvement, 3 = Some improvement, 2 = Considerable/Significant improvement is needed to reach standards of care, and 1 = Services not provided, totally inadequate care or potentially life-threatening practices).
Finally, a total summary evaluation score for complete assessment of all sections was marked in the summary evaluation sheet. The total summary score can assist in monitoring hospital improvements over time and to make also inter-hospital comparisons to some extent.
At the end of each hospital visit, assessors met with hospital administration and staff for a debriefing and to identify priority actions for improvement of the hospital services.
RESULTS
General hospital information All surveyed hospitals had isolated pediatrics OPD and ward except Bishoftu hospital whose pediatric inpatient room was part of the adult medical ward. Separate pediatric waiting area and archive rooms were found in only 3 and 2 hospitals respectively. Six of the 8 hospitals had separate room for admitting paediatric infectious cases while only 3 had separate
room for admitting newborns. In 6 hospitals, children with surgical conditions were admitted in the general pediatric ward. None of the hospitals had a clearly designated high dependency area or room where very sick children are cared for and receive closest attention.
Electricity and running water were available in all hospitals even though most of the hand washing and toilet facilities were non-functional in half of the
hospitals. Five hospitals had backup power supply. All hospitals had appropriate sharp disposal boxes.
29
29
The pediatric service in these hospitals caters for children up to the age of 14 years. Based on available data for the period July 2006 to June 2007, there were on average 29 (range 12-66) children seen in the Paediatric OPD with about 5 admissions per day (range 1-9). The two commonest causes of OPD attendance and admission were severe pneumonia and diarrhoea with severe dehydration or dysentery. The mean bed capacity of the assessed hospitals was 39 (range 11-68) and the average bed occupancy rate was 58% (range 31% in Zewditu & 90% in Ambo). Children under 5 account for 64% of admissions to the pediatric wards. Based on available data from 5 hospitals, the average daily emergency patient load was 7 and the average all cause mortality in under fives was 11% (10-16%).
Hospital support systems (drugs, equipments and laboratory)
As shown in Tables 1 & 2, the availability of emergency drugs and equipments in the emergency area and the wards was inadequate. Drugs like parenteral Phenobarbitone (long acting anticonvulsant) were non-existent and those drugs that were available were often not immediately accessible. Only 3 hospitals had ambu bags (big mask) in the emergency area and the wards while infant size masks, nasal prongs and nebulizers were totally absent. Oxygen concentrator and heat sources were found in only one hospital. However, all hospitals had good laboratory facilities to perform the five basic laboratory tests for managing emergencies (RBS, blood film, HGB, CSF microscopy and blood group and cross match) even though results were delivered timely in half of the hospitals. Bilirubin and other chemistries were being done in 5 while culture facilities were available in only 2 hospitals.
Table 1: Availability of Essential Drugs
No Emergency area Ward Pharmacy/store
Drugs
1 Glucose 40% i.v. 5 (63%) 6 (75%) 7 (88%)
2 Glucose 5% i.v. 6 (75%) 7 (88%) 8 (100%)
3 Normal saline i.v. 6 (75%) 6 (75%) 7 (88%)
4 Ringer’s lactate i.v. 8 (100%) 7 (88%) 8 (100%)
5 Epinephrine (Adrenaline) s.c. 8 (100%) 7 (88%) 7 (88%)
6 Corticosteroids i.v. or p.o. 6 (75%) 5 (63%) 7 (88%)
7 Furosemide i.v. 5 (63%) 5 (63%) 5 (63%)
8 Diazepam i.m., i.v. 5 (63%) 5 (63%) 5 (63%)
9 Phenobarbital i.m., i.v. 0 (0%) 0 (0%) 0 (0%)
10 Paracetamol 6 (75%) 6 (75%) 8 (100%)
11 Ampicillin inj. 6 (75%) 4 (50%) 7 (88%)
12 Benzyl penicillin 5 (63%) 5 (63%) 7 (88%)
13 Cloxacillin 5 (63%) 5 (63%) 6 (75%)
14 3rd generation Cephalosporins 3 (38%) 4 (50%) 5 (63%)
15 Chloramphenicol 3 (38%) 3 (38%) 4 (50%)
16 Gentamicin 5 (63%) 5 (63%) 6 (75%)
17 *All anti-malaria drugs 4 (50%) 4 (50%) 8 (100%)
18 Digoxin 3 (50%) 4 (50%) 6 (75%)
19 F75 milk 2 (25%) 6 (75%) 6 (75%)
20 F100 milk 3 (50%) 7 (88%) 7 (88%)
21 Ready to Use Therapeutic Food 2 (25%) 5 (63%) 6 (75%)
22 Vitamin K i.m. injection 0 (0%) 0 (0%) 3 (38%)
23 ORS 6 (75%) 7 (88%) 8 (100%)
30
30
Table 2: Availability of essential Equipments & supplies
No Equipments Emergency area Ward Pharmacy/ store
1 Resuscitation table/area 5 (63%) 2 (25%) 3 (38%)
2 Torch 4 (50%) 4 (50%) 4 (50%)
3 Otoscope 5 (63%) 6 (75%) 6 (75%)
4 Scales for children 7 (88%) 8 (100%) 7 (88%)
5 Stethoscopes 7 (88%) 7 (88%) 8 (100%)
6 Thermometers 8 (100%) 8 (100%) 7 (88%)
7 Heat source 1 (13%) 2 (25%) 0 (0%)
8 Lumbar puncture set 5 (63%) 6 (75%) 3 (38%)
9 Oxygen source: 7 (88%) 7 (88%) 4 (50%)
oxygen cylinder 7 (88%) 7 (88%) 4 (50%)
oxygen concentrator 1 (13%) 1 (13%) 0 (0%)
central supply 0 (0%) 0 (0%) 0 (0%)
10 Flow-meters for oxygen? 7 (88%) 7 (88%) 5 (63%)
11 Equipment for the administration of oxygen? 7 (88%) 7 (88%) 6 (75%)
nasal prongs 0 (0%) 0 (0%) 0 (0%)
catheters 7 (88%) 7 (88%) 6 (75%)
Masks 0 (0%) 1 (13%) 0 (0%)
12 Self inflating bags for resuscitation 3 (38%) 3 (38%) 2 (25%)
13 Masks
infant size 0 (0%) 0 (0%) 0 (0%)
child size 3 (38%) 4 (50%) 2 (25%)
adult size 1 (13%) 1 (13%) 0 (0%)
14 Butterflies and/or cannulas of paediatric size 7 (88%) 7 (88%) 7 (88%)
15 NG-tubes, paediatric size 4 (50%) 7 (88%) 7 (88%)
16 Suction equipment 5 (63%) 7 (88%) 6 (75%)
17 Nebulisers for administration of Salbutamol 0 (0%) 0 (0%) 0 (0%)
Emergency care Most of the surveyed hospitals (5/8) were not appropriately organized, fully staffed and equipped to handle pediatric emergencies effectively. Most of these hospitals were not practicing the standard triage process by assessing children immediately on arrival for emergency or priority signs before administrative procedures. However, three hospitals, two of them implementing the new hospital improvement initiative, had assigned qualified nurses
at the reception to facilitate smooth patient flow and prioritization of the management of emergency cases. Even in these three hospitals the triaging and management of emergency cases was not up to the expected standards of ETAT. None of the staff working in the OPD and emergency areas had been trained in ETAT and there was lack of job aids and standard protocol for pediatric referral care (Table 3).
31
31 Table 3: Summary of grading of the emergency setup in the hospitals
No Standards and criteria Good To be
improved
1 Children are assessed for severity/ priority signs (triaged) immediately on arrival as per the ETAT standard
0 (0%) 8 (100%)
2 Patients do not have to wait for their turn, registration, payment etc. before a first assessment is done and action taken.
0 (0%) 8 (100%)
3 A wall chart or job aid for identifying children by severity of condition is located in the emergency admissions area.
0 (0%) 8 (100%)
Drugs, equipment and supplies
4 Essential drugs for emergency conditions (anticonvulsants, glucose, iv fluids) are always available and free of charge to the family
0 (0%) 8 (100%)
5 Essential lab tests (glucose, Hb or PCV) are available and results are obtained timely 4 (50%) 4 (50%)
6 Essential equipment (needles and syringes, naso-gastric tubes, oxygen equipment, self-inflating resuscitation bags with masks of different sizes, nebulisers or spacers) is available
0 (0%) 8 (100%)
Staffing
7 A qualified staff member is designated to carry out triage. 3 (38%) 5 (63%)
8 A health professional is available without delay to manage children determined to have an emergency condition.
3 (38%) 5 (63%)
Case management
9 Staff doing triage is trained in the ETAT guidelines and can implement them appropriately when the emergency room gets busy during peak hours
0 (0%) 8 (100%)
10 Staff is skilled in the management of common emergency conditions and starts treatment without delay: Management of convulsions, lethargy, severe respiratory distress, shock and severe dehydration.
0 (0%) 8 (100%)
Pediatrics ward
Table 4 summarizes the status of the 8 hospitals against the WHO standards and criteria for pediatric wards.
32
32
33
33 Table 4: Standard's for children ward
No Standards and criteria Good To be improved
1 Children are seen in OPD only by the designated health professional in the designated room/area.
6/7 (86%) 1/7 (14%)
Closest attention for the most seriously ill
2 The most seriously ill children are cared for in a section where they receive closest attention.
0 (0%) 8 (100%)
Separate ward for children.
4 Children are kept in a separate ward or separate area of a ward. 7 (88%) 1 (13%)
5 Severely ill children are kept apart from adults in wards such as for infectious diseases or intensive care.
4 (50%) 4 (50%)
6 Children with surgical conditions are at least kept in a separate room, with staff aware of the special needs for children such as feeding and warmth.
3 (38%) 5 (63%)
7 Arrangements are made to meet these needs. 1 (13%) 7 (88%)
8 In cold climates, the ward has an efficient and safe heat source. 0 (0%) 8 (100%)
Separate room for sick neonates with mothers
9 Sick new-borns are kept separate from healthy babies. 3 (38%) 5 (63%)
10 Mothers of sick new-borns are rooming in with their babies, and have adequate facilities. 0 (0%) 8 (100%)
Hygiene and accident prevention
11 Staff has access to hand washing facilities The ward is kept clean and dangerous items are inaccessible for children
3 (38%) 5 (63%)
12 Sharps are disposed of in a special container preventing accidents 8 (100%) 0 (0%)
Hygienic and sufficient services facilitate the stay of mother and child
13 There are sufficient and adequate toilets which are easily accessible 0 (0%) 8 (100%)
14 Mothers have access to running water and to an appropriate space, near the ward, to wash themselves and their child.
0 (0%) 8 (100%)
15 Mothers have access to a washing facility, in order to wash her and her child’s clothes. 0 (0%) 8 (100%)
16 Patients are kept in a bed/cot with a clean mattress. 4 (50%) 4 (50%)
17 Patients receive bed sheets 6 (75%) 2 (25%)
34
34
Case management practices Overall, the case management of common childhood problems including pneumonia, diarrhea, fever conditions and malnutrition was not optimal in almost all hospitals. Pneumonia was diagnosed and its severity correctly classified more or less based on diagnostic signs in 2 hospitals. Similarly, appropriate use of antibiotics and oxygen was observed in a third of the hospitals (3/8). Patient monitoring and supportive care were inadequate in all hospitals. None of the hospitals had nebulizers for inhalation therapy. Correct assessment of dehydration was documented in half of the hospitals and the monitoring and management of dehydration was inadequate in almost all hospitals. Inappropriate use of antibiotics for diarrhea was observed and supportive care especially feeding was inadequate.
In the majority of cases, differential diagnosis of fever was not considered exhaustively and investigations were incomplete. In some instances planned LP tests were not done and patients were treated empirically. Inappropriate choice and administration of antibiotics was observed and documentation of patient progress and treatment given was not up to the standard in almost all hospitals. Overall, patient records and monitoring charts were poorly recorded. Most of the hospitals had the dietary supplies needed for the management of severe malnutrition. Even though routine antibiotics were given for malnourished children as per the national guideline, feeding was not given according to the recommended schedule especially at night. Prevention and management of other complications like dehydration and hypothermia was inadequate. None of the surveyed hospitals had rooms with heaters for malnourished children. Monitoring of patients and recording of progress was poor in the majority of cases.
All the necessary guidelines and tools were in place for counseling, diagnosing and staging of paediatric HIV and for the treatment and monitoring of antiretroviral therapy. There were well trained staff who were regularly mentored and most of the HIV
services were conducted in line with the national standard especially at the outpatient level. However, inpatient management of some opportunistic infections and the supportive care of these patients need improvement.
Supportive care and nutrition Although breastfeeding was encouraged, none of the hospitals provide appropriate routine pediatric diet for children. Misuse of intravenous fluids was common including in children with severe acute malnutrition.
Multiple antibiotics were often prescribed some times for longer durations. Screened blood was used in all hospitals but the indications for its use and the administration procedure needs improvement.
Monitoring of patients All admitted children were assessed by a doctor at admission and majority of them were re-evaluated about once a day during working days and upon consultation during weekends. A qualified nurse was available 24 hours per day in the children's wards in all hospitals and a medical doctor in 6 of them. Nutritional status was assessed at admission using weight for age in most hospitals. All patients had individual charts, vital sign and medication sheets, but the case histories, progress notes and the other
monitoring charts were not properly and regularly recorded in the majority of cases both by physicians and nurses. There was no designated area where very ill children receive highest attention. Except for the Severe Acute Malnutrition Chart, there were no standardized monitoring charts consisting of all relevant parameters which could simplify patient monitoring.
35
35
Neonatal care
Early and exclusive breastfeeding, skin to skin contact and proper thermal protection was not practiced adequately in almost all hospitals. Eye and Vitamin K prophylaxis and immunizations were given routinely in only 2 hospitals. Neonatal resuscitation flow chart was available in only one hospital. Overall, only three hospitals had a separate room for sick newborn babies with bed capacity ranging from 3 to 17. In the other 5 hospitals neonates were kept in the general pediatric ward or the maternity room. Only
one hospital was well staffed and equipped (heaters, oxygen sources, phototherapy and suction machines, resuscitation materials and other supplies) to closely monitor seriously ill newborns with 24 hour availability of skilled nursing staff. However, rooming-in and hygienic facilities for mothers were inadequate in all hospitals. As shown on Table 5, the case management of common neonatal problems is inadequate and needs strong effort to improve the situation.
Table 5: Sick newborn case management summary table
No Good To be improved
1 Neonatal sepsis is appropriately diagnosed. 2/6 (33%) 4/6 (67%)
2 Neonatal sepsis is appropriately treated. 1/6 (17%) 4/6 (67%)
3 Specific feeding needs of sick young infants and those with low birth weight, are met.
1/6 (17%) 5/6 (83%)
4 Jaundice is adequately recognized and managed. 1/6 (17%) 5/6 (83%)
Pediatric surgery & rehabilitation Of the 8 hospitals surveyed, one (Ambo Hospital) was not doing any major surgery for children. In 6 of the hospitals, children with surgical conditions were admitted in the general pediatric ward. Pre- and post-operative starving was kept to a minimum in the majority of the hospitals which perform major surgery. Overall, frequent post-operative monitoring of vital signs and the readiness for resuscitation was inadequate in nearly half of the hospitals. Only one hospital had basic rehabilitation facilities.
Hospital administration & access to hospital Nationally, there were no pediatric treatment guidelines for referral hospitals and none of the hospitals practice pediatric death audits. Availability of essential drugs and equipment that are basic for provision of quality pediatric emergency and inpatient services are inadequate in almost all of the hospitals surveyed. However, the handling of available drugs including the stock management was good and old drugs were used first.
Economic and transportation barriers were serious concerns for all the care takers interviewed and most of them had tried traditional medicine before seeking care from providers. Caretakers were not satisfied with the communication from providers regarding the problems of their sick children, explanations about ward procedures and the treatments and follow-up details. Almost all care takers indicated that the
condition of the toilet and washing facilities was very poor.
Summary The following table and figures 1 & 2 show the summarized score by major sections of the assessment and by facility. Table 6 is a summary sheet showing the details of the summary scores of the 8 hospitals in the different service areas.
36
36
Table 6: Details of summary score of the 8 hospitals in the different service areas
No Summary score of different service areas Good To be improved
5 4 3 2 1
1 Essential drugs, supplies and equipment ** 5 3
2 Laboratory support 6 2
3 Emergency area and management 6 2
4 Children's ward and facilities 8
5 Cough or difficult breathing child management 2 1 5
6 Diarrhoea patient management 2 6
7 Febrile child management 2 5 1
8 Malnourished child management 1 7
9 Management of child with HIV/AIDS 3 5
10 Summary score of supportive care 6 2
11 Summary score in monitoring 4 4
12 Routine neonatal care service 1 6 1
13 Neonatal Care Unit facilities 1 1 6
14 Case management and sick newborn care 1 7
15 Paediatric surgery and rehabilitation 2 4 1 1
16 Summary score hospital administration 4 4
17 Summary score access to hospital 5 3
Total score = 405 30 84 222 68 1
Hospital summary score = total score / 8*17 2.98
NB: - ** = Five of 8 hospitals had a summary score of 3 out of 5, while 3 hospitals has a summary score of 2 out of 5 for emergency drugs, supplies and equipment section.
Similarly, Figure 1 shows average scoring of the quality of the different pediatric care services in the 8 hospitals out of a maximum score of 5. As it can be seen clearly from the figure, there are marked variations in quality of the different pediatric
care/services rendered. The quality of care given to children with HIV/AIDS is relatively in far better condition compared to services given to neonates, scores of 4.4 and 2.3 out of 5 respectively.
37
37
Figure 1: Average summary score of the different pediatric care services out of a maximum score of 5
As shown in Figure 2 below, the hospitals’ quality of
care total summary score ranged from 2.6 to 3.44 out
of 5 with an average of 2.98 (~3.0) which indicates
that all hospitals need substantial actions for
improvement to reach defined standards. This figure
also shows the difference in the performance of the
eight hospitals.
4.4
3.8
3.6
3.3
3.3
3.1
3.0
3.0
2.9
2.8
2.8
2.6
2.6
2.5
2.5
2.4
2.3
0.0 1.0 2.0 3.0 4.0 5.0
HIV/AIDS mgt
Laboratory support
Cough mgt
Diarrhoea mgt
Malnutrition
Febrile child mgt
Ward and facilities
Routine neonatal care
Surgery and rehabilitation
Emergency area & mgt
Supportive care
Ess drugs & equipment
Access to hospital
Monitoring
Hospital administration
Neonatal Care Unit facilities
Sick newborn care
38
38
Figure 2: Total summary score of the 8 hospitals, out of a total score of 5
CONCLUSIONS AND RECOMMENDATIONS
This assessment has the following limitations. First,
sampling of the hospitals was by convenience;
majority of them being zonal hospitals the results may
not fully reflect the situation in the more peripheral
district hospitals. Second, the study was largely
observational and assessments were in part based on
the judgments of the observers. Besides, the quality
of available hospital statistics was inadequate to
disaggregate patient load at OPD, emergency and
admission levels by appropriate age categories (0-28
days, 1-12 months, 1-5 years and > 5 years). The
statistics problem definitely under-estimates the
reported hospital mortality rate. Similarly, it was
impossible to determine the mortality rate within the
first 24 hours of admission which is a very sensitive
indicator of the quality of emergency care in any
hospital setting.
Despite the potential limitations, this study provides
some basic information about the status of pediatric
hospital care that could guide the national efforts in
improving the quality of referral care for children.
The overall case fatality rate of 11% (10-16%) in this
survey was comparable to that reported from Zambia
(12-15.8%) and Kenya (4-15%) but the possibility of
under-reporting is there.
All of the surveyed hospitals were not appropriately
organized and fully equipped to handle pediatric
emergencies effectively. None of these hospitals were
practicing the standard triaging process by assessing
children immediately on arrival for emergency and
priority signs & the overall management of emergency
cases was not up to the expected standards of ETAT.
In the majority of the surveyed hospitals (5/8), there
were no separate pediatric waiting areas which are
3.12.9 2.9
2.6
3.0
3.4
3.1 3.0
0.0
1.0
2.0
3.0
4.0
5.0
Adama Ambo Bishoftu Debre Birhan Dessie Yekatit 12 Yirgalem Zewditu
Hospitals
sco
re
39
39
essential for the proper implementation of ETAT.
Even though the current quality improvement process
through the hospital BPR initiative is a very good
opportunity for the overall improvement of hospital
care for children, the centralized triaging mechanism
that keeps patients of all age groups together in one
waiting area is not conducive for the effective
implementation of the ETAT standards.
Generally, there was lack of some essential drugs
and materials such as nasal prongs, infant and child
size bag & masks, nebulizers, heaters and oxygen
concentrators. The total absence of long acting
parenteral anti-convulsants poses serious
shortcoming in the management of pediatric
neurologic emergencies.
In almost all hospitals, there was no clearly
designated and properly arranged/equipped high
dependency area where very sick children receive
highest attention. Majority of the hospitals (5/8) do not
have any special arrangement and facilities for
providing appropriate neonatal care.
Hygienic facilities were below the expected standard
in majority of the hospitals and none of the hospitals
provide appropriate routine pediatric diet for children.
Overall, the case management of common neonatal
and childhood illnesses was not optimal. None of the
hospital staff had been trained in ETAT and there
were no job aids or protocols for pediatric referral
care which could partly contribute to the problems
observed in the case management.
In summary, the quality of pediatric referral care needs serious attention and coordinated systematic improvement effort using the opportunity of the national hospital management initiative and the BPR process to institutionalize ETAT and standards of hospital care for children. This has to be complemented with availing of appropriate job aids, essential supplies and equipment, and improvement of health worker skills through training, clinical mentoring and regular supportive supervision.
40
40
RECOMMENDATIONS
1. The ETAT standards should be incorporated into the Ethiopian Hospital Management Initiative blue print and the
BPR documents by the FMOH to ensure its systematic implementation.
2. National plan of action should be developed for systematic and phased implementation of ETAT in referral
hospitals.
3. Staff should be trained in ETAT through in-service courses & ETAT should also be introduced into the pre-service
teaching to ensure its sustainability.
4. National pediatric referral care protocol & job aids should be availed in all hospitals
5. Coordinated efforts and more resources needed for availing essential drugs, supplies and equipments in all
hospitals.
6. All hospitals should have arrangements/rooms for the care of sick neonates, high-dependency areas for critically
sick children and isolation rooms for infectious cases.
7. Efforts should be made to improve the hygienic facilities & the quality of routine hospital diet for children.
8. The recording and reporting system in the hospitals need to be improved and pediatric data should be
disaggregated by appropriate age categories.
9. Mechanisms for regular supportive supervision and mentoring needed to achieve sustained improvement in the
quality of pediatric referral care.
10. The Ethiopian Pediatric Society, the Medical teaching institutions, WHO, UNICEF and other partners should
support the quality improvement process.
41
41
ACKNOWLEDGEMENTS
We would like to thank the FMOH especially the former Family Health and Health Services Departments, the Directors and staff of the surveyed hospitals for facilitating the whole process. We are very grateful to the World Health Organization for the financial and technical support provided for the assessment. Sincere gratitude is expressed to Dr Neghist Tesfaye (FMOH), Dr Teshome Desta (WHO/ICST-ESA), Dr Wilson Were (WHO/HQ) and Prof. Elizabeth Molyneux (Queen Elizabeth Central Hospital, College of Medicine, Blantyre/Malawi) for their valuable technical support.
42
42Management of severe acute malnutrition in children using community based therapeutic care approach:
a review of three years data from southern Ethiopia.
Efrem Teferi, MD1, Shiferaw Teklemariam, MD, MPH1 , Lopiso Erosie, BSC, MPH1 , Abel Hailu, MD1
Tefera Belachew, MD, MSc, DLSHTM2, Mohammed A Yassin, MD, MSc, PhD1,3
Abstract
The objective of the study is to assess the outcome of community-based therapeutic care (CTC) for children with severe
malnutrition in Southern Ethiopia. Diagnosis of severe malnutrition was made based on anthropometric measurement and
all children received therapeutic food according to the protocol and were discharged from the feeding program when they
their weight for height was more than 80% of the reference for 2 consecutive weeks. Data on the number of admissions,
discharges, weight gain and length of stay in the program were recorded using standard formants and reports were sent to
the Regional Health Bureau monthly. The data was entered using EPI-Info proportions and means were compared using
chi-square test. This is a retrospective review of reports retained in the Bureau. A total of 12,316 patients with severe acute
malnutrition, 56.2% marasmic and 43.8% kwashiorkor cases were treated in CTC program from 2003 to 2005. The average
cure and death rates were 91% (9871) and 2.5% (217) and the average weight gain was 5.3 and 5.8 grams /kg/day and the
average length of stay was 49 and 42 days for cases of Marasmus and Kwashiorkor, respectively. Except for weight gain
and length of stay, our findings exceeded the minimum sphere standards for treatment outcome measures. In conclusion
the CTC approach has a comparable outcome to Therapeutic feeding centers and could be expanded quickly during
emergency situation. As majority of patients are treated at home, the workload for the health worker would be reduced, so it
is an alternative approach for management of severe malnutrition where human resource and space in health facilities are
limited.
43
43
Introduction
Ethiopia has a long history of food insecurity and nutritional problems affecting a large proportion of the population. (1) Even during a relatively non-drought years, malnutrition in children in Ethiopia is extremely high exposing the survival of this group of the population at a great threat. An estimated 47% of Ethiopia’s under-five children are moderately or severely stunted contributing to an under five-mortality rate of 123/1000 live births. (2) Ethiopia stands 6th among countries with the highest number of under-five deaths in the world, with more than 472,000 under-five dying each year, (3) Malnutrition, even in its milder form, accounts directly or indirectly for 53% of all under-five deaths in Ethiopia (4). The Southern Nations, Nationalities and Peoples Regional State (SNNPR) is located in southwestern part of the country and has a population of about 14 million. The region is divided into 14 administrative Zones and 8
special Woredas (districts) with a total of 104 Woredas. There are 56 nationalities in the region , which gives it the unique feature in encompassing rich and diverse culture. According to the 2005 Regional Health Bureau annual report, there were 16 hospitals, 160 health centers, 1336 health posts with potential health service coverage of 50%. The regional ; DPT3 coverage was 90%, family planning 47%, antenatal care 60% and deliveries attended by health professionals of 18%. The major causes of mortality among children under five years of age are pneumonia, diarrhea and malaria. Malnutrition has become a common feature occurring in the region in recent years with large-scale famines reported in 2003 and cases continuing to be reported. Dependency on rainfall, poor resource management, uncontrolled population growth, poor farming planning and limited food reserves of households complemented with failure of rains in 2002 exacerbated the living condition of vulnerable communities leading to exhaustion of coping mechanisms.
In response to the recent famine, many international humanitarian agencies and the government opened Therapeutic Feeding Centers (TFC) to mitigate the crisis. A Decentralized therapeutic program was initiated with Concern, Save the Children US in collaboration with other international organizations. In 2004, the Ministry of health in cooperation with UNICEF initiated enhanced outreach strategy (EOS) for child survival. This involves screening for malnutrition, Vitamin A supplementation, strengthening immunization and health education for child survival and was started in 54 woredas of the region (17) and expanded to the entire region in 2006. Improvement in child survival is strongly associated with decrease in malnutrition in countries characterized by high rates of general malnutrition such as in Ethiopia (5-7). To reduce mortality due to severe acute malnutrition, TFCs were established in SNNPR in 2003, which provided a high quality individual inpatient care and
patients receive formula 75 (F75) and formula 100 (F100) milk with routine drugs (18). A high cure rate (87.2%) and a relatively low death rate (3.6%) was reported from 25 TFCs opened in the region in 2003-05 with an average length of stay of patients of 21-25 days (8). This high intensive care phase of treatment is very important for patients with complicated malnutrition associated with anorexia, septicemia, hypothermia, hypoglycemia and severe dehydration. However, management of children with malnutrition in TFC requires skilled staff, room for inpatient care, materials including bedding, cooking utensils and caretakers’ food (8-11). The centers were opened and ran by non-governmental organizations (NGO) and coordinated by the regional health bureau. When the activities were integrated into the routine health services and handed over to health facilities, it was difficult to continue with the limited work force, space and resource in the facilities.
-4- The traditional model of inpatient treatment of severe acute malnutrition (SAM) does not consider the social aspects of management of malnutrition and hence has high opportunistic costs to mothers/care givers (9). Mothers/care givers had to stay in centers for several weeks leaving their other children and family members at home and hindering them from their daily activities. Community therapeutic care (CTC) is a nutritional
intervention designed with the capacity to address SAM in both emergency and development contexts (9). Its underlying aims are to maximize coverage and access. In practice, this means giving priority to provision of care for acutely malnourished over inpatient care for a few extreme cases.
44
44CTC was introduced in SNNPR in 2003 and expanded in 2004 and 2005. It integrates supplementary and therapeutic feeding with an emphasis on outreach and community based support. In out patient therapeutic program (OTP), the therapeutic product used is ready to use (plumpy nut and BP 100 biscuit) with outpatient drug treatment protocol. There were few referrals for inpatient treatment, which includes those with complication and who were admitted in stabilization centers (SC) and
treated for few days and referred to OTP like in phase one TFC (13). During emergency, CTC approach can quickly provide good coverage and treatment for a large number of severely malnourished people. This paper reviews reports of CTC to assess the outcome of malnourished children treated in community therapeutic care established in the southern Ethiopia and compared with the results of children treated in TFC.
Materials and methods
Children with severe acute malnutrition were admitted to therapeutic care established in response to the famine encountered in the region. Diagnosis of severe malnutrition was made based on anthropometric measurement and brief examination for bilateral pitting pedal edema. At admission, patients were assessed for hydration, anemia and signs of infection. Patients were given oral doses of Vitamin A, Folic acid, Amoxicillin (5-day course), Mebendazol, treated for dehydration with Resomal and given ready to use therapeutic food (RUTF) according to the protocol (18). Patients were discharged from the outpatient therapeutic feeding program when their weight for height was more than 80% of the reference for 2 consecutive weeks and if they did not have signs of infection. They were followed in supplementary feeding program until they reach 85% of the reference for 2 consecutive weeks. At each follow up visit, weight, extent of pitting edema, presence of infection and treatment were recorded. In CTC, plumpy
nut was used instead of F100. The main difference between F100 and plumpy nut is that part of dried skimmed milk in the F100 was replaced with peanut butter (with a 25% total weight). Plumpy nut has an energy density 5 times more than that of F100. Plumpy nut was used in all the centers except for few weeks when BP100 was used due to shortage of plumpy nut.Data on the number of monthly admissions, discharges, average weight gain and length of stay in the program were recorded in each therapeutic centre using standard formants and reports were compiled by the health facilities and NGOs supporting the programs and were sent to the Regional Health Bureau (RHB) monthly. The data was entered using EPI- Info programme(CDC Atlanta) and analyzed using descriptive statistics, proportions and means were compared using chi-squared test and p values value <0.05 was considered as significant.
This is a retrospective review of reports retained in the RHB. All reports from CTC sent to RHB were included in this paper, except reports of 243 cases from Kembata Tembaro zone and 44 admissions from Sidama Zone
because of incompleteness. Patients who were defaulters, referrals and non-respondents were excluded as their outcome was not known. Ethical approval was obtained from the SNNPR Health Bureau.
Results
A total of 12,316 patients with severe acute malnutrition, 56.2% marasmic and 43.8% kwashiorkor cases were treated in CTC program from 2003 to 2005. Of these, 1540 (12.5%) were treated in 2003, 1955 (16%) in 2004 and 8791(71%) in 2005. The majority (90%) of the cases were age between 6 months to 5 years old. The average cure and death rates for the region were 91% (9871) and 2.5% (217) respectively. The highest (99%) cure rate
was recorded in Boricha and Damot Gale districts (Woredas), while the lowest cure rate (80.6%) was observed in Malgano in Sidama zone. Death rate above minimum standard was not reported from any of the CTC, the highest death rate (8.4%) was reported from Bedessa and the lowest (0.4%) in Arbegona as shown in Table 1.
The number of admissions was not the same as the number of cases discharged, as some of the cases were on follow up when the report was compiled. There were also cases that defaulted from the program, referred to other places and non respondents making the total
number of cases discharged higher than the sum of deaths and cured cases. The average weight gain for the region was 5.3 and 5.8 grams /kg/day for cases with Marasmus and Kwashiorkor, respectively. The average length of stay was 49 and 42 days for marasmic and Kwashiorkor cases, respectively (p > 0.05 for both) as shown in Table
45
452. The outcome of patients treated in CTC was comparable to those treated in TFC in the region during
the same period with cure rate of 91% vs. 87% (p>0.05) and death rates of 2.5% vs. 3.6% (p>0.05), respectively.
However, there was statistically significant difference in the mean length of stay which were 21 and 25 vs. 42 and 49 days for marasmus and kwashiorkor patients respectively and the average weight gain of 13.4 and14
vs. 5.3 and 5.8 g/kg/day for patients treated in TFC than in CTC (p<0.01 for both) (Table 3). A similar trend was observed when the outcome from the CTC program was compared to the sphere standard as shown in Table 4.
Discussion
Given the spatial arrangement of health service units and their limited capacity to handle a large number of cases of severe acute malnutrition during emergency and crisis in Ethiopia, seeking an alternative solution for management of such cases cannot be overlooked. Community base therapeutic care provides a promising alternative option to the TFCs and facility-based stabilization centers (9-12). Though CTC cannot totally replace an inpatient therapeutic care as some cases with complications, such as infections, may still need an inpatient care, it is complementary to therapeutic and supplementary feeding programs (9).
The experience in implementing therapeutic feeding programs in the region in the last few years enabled us to understand and expand CTC very quickly. The total number of cases treated in the region (12,316) was sizable enough to assess the outcome of the therapeutic feeding programs. Overall, the outcome of patients treated in the CTC approach were promising considering the minimum sphere standards for the outcome indictors of therapeutic feeding programs (9-11,15,). In addition, CTC handles the majority of cases by creating access and capacity which is very difficult to meet during emergency situation (9-11). The outcome of children treated by the CTC approach in the SNNPR exceeded the minimum standard for both cure and death rates although the average weight gain was low and the average length of stay was longer. The latter may be due to sharing of RUTF with siblings within the family as patients in CTC are treated at home. Our results are similar to the findings in study done in Badewacho in Hadiya in SNNPR, which reported a cure rate of 85% and death rate of 4%, mean weight gain of 4.8 g/kg/day with a mean length of stay of 42 days among 170 children treated in CTC (16). The results of our study were better compared to findings reported from other parts of
Ethiopia (Amhara, Oromia) and other African countries (Sudan, Malawi, and Niger) (13,17). The average length of stay was also shorter than results in other parts of Ethiopia, which reported 36-91 days with a similar average weight gain from studies in other parts of Ethiopia (3-6.5 g/kg/d) but less than those reported from other parts of Africa (4.2-10g/kg/day). (13, 17)
The cure and death rates of patients treated in CTC was comparable with those treated in TFC in SNNPR during a similar period. However, the duration of stay was shorter and the mean weight gain was much better for patients treated in TFC than those treated in CTC program. This could be due to the unavoidable sharing of RUTF with siblings and even adults where the bulk of the treatment period in CTC approach is based at home. The fact that most of the data from TFC were collected during emergency and data from the CTC were collected during non-emergency situation makes comparison difficult. However, according to reports (data not shown) from Boricha district from where data from TFC and CTC programs were collected during emergency situations in 2003 and 2005, the treatment outcome was similar confirming that CTC program was effective even during emergency situation. It was observed that inpatient treatment schemes had no additional advantage over CTC except the higher average weight gain obtained compared to the workload it incurs to health workers and the difficulty to admit all patients with the limited space and health professionals available in the health facilities.
46
46The limitation of this study was the fact that it was based on a retrospective record review and there
was no information on the follow up of individual cases and suffers incompleteness and missing data.
In conclusion CTC as a new strategy for management of severe malnutrition was successfully implemented in SNNPR. It has a high coverage, low cost and could be expanded very quickly in emergency situations. The majority of patients can be treated in the outpatient program without disrupting caregivers from their daily activities and only very few patients who had infections and very severe malnutrition required inpatient care for few days. The product used in CTC was ready to use therapeutic food (plumpy nut, BP 100 biscuit) and could be implemented as an outreach and by Health Extension Workers.
CTC is an alternative approach for management of severe malnutrition where shortage of health professionals and limited space for admission are the major hurdles. However, problems including shortage of health professionals, drugs, therapeutic products, transportation, lack of space to admit complicated cases, misunderstanding among some who consider CTC as an NGO business, lack of training and high turnover of health workers should be anticipated during integration of CTC into the routine health system. As CTC is a community based approach, deployment of health extension workers in the region in the recent years would be a good opportunity for outreach activities including follow up and referral of cases in the villages.
For the CTC to be taken, scaled up and to be part of the routine service, continuous in-service training of health professionals, inclusion of CTC to the pre-service training, facilitation of transportation of materials required for CTC, local production of therapeutic food and integration of CTC into the routine health service is recommended. The program should be supported with proper nutrition behavioral change communication on infant and young children feeding so that to prevent malnutrition and its recurrence.
47
47
-11-
Table 1: Outcome of cases of severe acute malnutrition treated in the CTC program from 2003-
2005, South Ethiopia
Zone Woredas
Number of cases Admitted
Treatment Outcomes
Cured Death Total Discharged N Percent N Percent
Silte Dalocha 762 631 92.79 19 2.8 680 Lanfuro 907 795 88.24 21 2.3 901
Sidama Malgano 1159 815 80.61 14 1.4 1011 Boricha 1255 1140 98.96 16 1.4 1152 Arbegona 248 215 93.48 1 0.4 230 Bensa 213 61 89.71 2 2.9 68 Bedessa 399 206 86.92 20 8.4 237 Shebedino 575 495 95.56 7 1.4 518 Awssa Zu. 748 479 89.53 11 2.1 535
Wolaita Boloso Sore 2593 2408 93.12 43 1.7 2586 Offa 801 641 86.50 20 2.7 741 Damot Gale 578 514 98.66 4 0.8 521 Humbo 159 142 95.30 5 3.4 149 Sodo Zurria 278 185 89.37 9 4.3 207
Kembata T. Modulla 313 141 86.50 6 3.7 163 Konso Karat 1328 1003 88.60 19 1.7 1132
Total 12316 9871 90.9 217 2.5 10831
CTC = Community based Therapeutic care, N = Number
48
48
-12-
Table 2: Mean length of stay and average weight gain for cases of severe acute malnutrition
admitted to the Community based Therapeutic care program from 2003 to 2005, South Ethiopia.
Centers
Average length of stay in
days
Average weight gain in gram/kg/day
Marsmus Kwashiorkor Marasmus Kwashiorkor
Dalocha 63 47 5.2 4.0
Lanfuro 61 40 5.6 5.1
Malgano 37 32 7.8 10.0
Shebedino 40 32 4.6 5.0
Awassa Zu. 59 47 3.0 2.8
Arbegona 58 50 5.2 4.4
Boricha 45 32 6.0 5.5
Bedessa 70 66 3.0 18.0
Offa 62 52 3.9 3.1
Sodo Zur. 40 42 7.5 4.8
Boloso So. 58 52 5.0 4.2
Damot Ga. 36 33 5.4 4.4
Humbo 41 30 5.5 3.8
Modulla 32 30 5.3 6.3
Karat 39 37 6.0 5.6
Regional average 49.4 41.5 5.3 5.8
49
49
-13-
Table 3: Comparison of outcome of patients treated in TFC and CTC
in the southern region treated during similar period (2003-2005)
CTC = Community based Therapeutic care, TFC = Therapeutic Feeding Centre
-14-
Table 4: Outcome of children treated by the Community based Therapeutic care e program from 2003- 2005 compared to the minimum sphere standard, South Ethiopia.
Indicator Sphere standard* Our Study findings
Cure rate >75% 90.9%
Death rate <10% 2.5%
Mean weight gain >8grams/kg/day 5.34-5.8 gram/ kg/ day
Mean duration of stay 30-40 days 42-49 days
Program Cure rate
Death rate
Average length of stay (day)
Average weight gain (g/kg/d)
Marasmus Kwashiorkor Marasmus Kwashiorkor
TFC 87.2 3.6 25 21 14 13.4
CTC 91 2.5 49 42 5.3 5.8
50
50
*Guideline for management of severe acute malnutrition,FMOH,2007, Referrence.22
References
1. Birhane G. Running a national early warning system: the Ethiopian experience, Addis Ababa Relief
and Rehabilitation commission, 1991.
2. CSA & ORC Macro (Central Statistical Authority (Ethiopia) and OCR Macro), 2006, Ethiopia
Demographic and Health Survey 2005. Addis Ababa, Ethiopia and Calverton, Maryland, U.S.A:
Central Statistical Authority and OCR Macro.
3. Robert E Black, Saul S Morris, Jennifer Bryce. Where and why are 10 million children dying every
year? The Lancet 2003;28: 361,
4. BASICS II. Basic support for institutionalizing child survival. The Second Child Survival Revolution,
Summary of the Lancet Child Survival Series: BASICS II, 2003.
5. Pelletier, D., & E. Frongillo. 2002. Changes in Child Survival are Strongly Associated with
Changes in Malnutrition in Developing Countries. FANTA, Academy for Educational Development:
Washington DC, USA.
6. Federal Ministry of Health, Family Health Department. National strategy for child survival in
Ethiopia, Addis Ababa, July 2005.
7. WHO. Reducing severe and moderate malnutrition in Children. Bull WHO 1995, 73(4): 443-48.
8. Teferi E et al. Treatment Outcome of children admitted to Therapeutic Feeding centers in Southern
Region. (in press).
9. De Waal, A. Taffesse, L. Carruth . Child survival during the 2002–2003 droughts in Ethiopia,
Special Issue: Humanitarian Crises: The Emergency Rooms of Global Health, Taylor &
Francis,June 2006, 1(2).
10. Collins S, Dent N, Binns P etal. Management of Sever acute Malnutrition in children: Review,
online www.thelancet.com September 25, 2006.
51
51
11. Collins, S. ‘Changing the Way we Address Severe Malnutrition during Famine’. The Lancet 2001,
358:498-501.
12. Collins, S. and Sadler, K. ‘The Outpatient care for severely malnourished children in emergency
relief programs: a retrospective cohort study’. The Lancet 2002, 360:1824-30.
13. Community based therapeutic care (CTC) in Ethiopia. Proceeding of workshop in Addis Ababa,
Ethiopia, 22- 23 June 2004.
14. Federal MoH/UNICEF/MOST. Guideline for enhanced outreach strategy (EOS) for child survival
interventions, revised version, Addis Ababa. July 2005.
15. EL HadjiIssakhaDiop, Nicolle Idohou, Marieb am Adeline Ndour, Andre Brined and Salimata Wade.
Comparison of
the efficacy of ready to use food and liquid milk Based diet for rehabilitation of severely
malnourished children. Am Clin Nutr 2003; 78: ; 302-7.
16. Steve Collins, Kate Sadler, Outpatient care for severely malnourished children in emergency relief
programmes; a retrospective Cohort study. The Lancet 2002;360;1824-30.
17. Community based approach to managing severe malnutrition, Proceedings of an interagency
workshop, Dublin October 2003.
18. Federal Ministry of Health . National guideline for the management of severe acute malnutrition for
Ethiopia. MoH, Addis Ababa, May 2004.
52
52
Acknowledgments
We would like to thank members of Laboratory and Research Department for their valuable comments on
the draft. We also thank save the children USA, International Medical Corpus, Action Contra La Faim for
helping to implement the program and health workers in zones and woreda for working hard to save the
lives of patients. I t would have been difficult to implement the program with out the help of UNICEF (United
Nations Children’s Fund), who provided therapeutic products drugs, and conducted trainings in cooperation
with Regional Health Bureau.
53
53
CHILD SURVIVAL: PROGRES TOWARDS MEETING MDG4
Assaye Kassie1
Abstract Few causes are responsible for the majority of under-five deaths in Ethiopia: pneumonia (28%), neonatal causes (25%), malaria (20%), diarrhea (20%), measles (4%) and AIDS (1%). To prevent these deaths, and to achieve Millennium Development Goal 4 (MDG4) (“to reduce by two thirds, between 1990 and 2015, the under 5 mortality rate”), it is necessary to ensure the implementation of cost-effective interventions that are listed in the National Child Survival Strategy. There are examples of remarkable achievements in coverage increase within short time periods, including training of 30,000 Health Extension Workers (HEWS) in the last 4 years, rapid increase, from 1% in 2005 to 42% in 2007, in percentage of children under the age of five years who slept under a Long Lasting Insecticide-treated Nets (LLINs), and increase in coverage of Vitamin A supplementation from 45 % in 2005 to 91% in 2007. These successes can serve as benchmarks to scaling up of other interventions. Among the major killers, the ones that are poorly addressed are childhood pneumonia and perinatal problems which are the leading causes of under-five mortality in Ethiopia. Realizing the continuum of care approach at delivery level and sustaining it over time, and searching for an alternative way of improving access to treatment of childhood pneumonia and essential newborn care, are crucial challenges for child survival in Ethiopia. Furthermore, there is growing consensus that a primary bottleneck to achieving MDGs in low-income countries is health systems that are too fragile and fragmented to deliver the volume and quality of services to those in need. Major shortfalls are identified in the health workforce, lack of donor coordination, and week information systems. It is for this reason that the Health Sector Development Programme (HSDP) in Ethiopia is focusing on cost-effective health interventions and on health systems strengthening in order to achieve the dual goals of improving population health and reducing health inequalities.
54
54
1) Introduction: an historical perspective Following the success of the 1979 International Year of the Child, in the early 1980’s signs of hope were emerging for childrens’ causes. The evolution of Basic Service and Primary Health Care (PHC) approaches gave the practitioners of child health and human development a new sense of purpose and in 1982 an initiative known as the “Child Survival Revolution” (later including child development) was launched (19). The “Child Survival Revolution” was initiated to promote Growth monitoring, Oral rehydration therapy, Breast feeding, Immunization, the provision of Food and Family planning. These interventions have collectively come to be known as “GOBIFF” (19). Ethiopia was one of the first countries to implement these high-impact child survival interventions. The Ethiopian Expanded Programme on Immunization (EPI) was launched in 1980 but until recently coverage of all of the above key interventions, including EPI, remained very low. Regardless of the fact that so many lives could have been saved with the implementation of such simple, high-impact interventions, around the mid-nineties, it was noticed that child mortality was not yet receiving
1 UNICEF, Addis Ababa enough attention. This was mainly because the world’s attention had been, understandably, focused on the growing HIV pandemic and HIV associated opportunistic infections such as tuberculosis and the like. While progress in reducing under-five mortality has in many low income countries slowed, in others it has totally stopped declining and in some cases even regressed significantly (17;16). Following the Lancet Child Survival publications in 2003 (4), the second global “Child Survival Revolution” was launched. In 2004, the National Child Survival Conference was organised in Addis Ababa. The Federal Ministry of Health (FMOH) and its partners all participated and in 2005 the National Child Survival Strategy was developed (6). The overall objective of this strategy was to reduce under five mortality by two thirds between 1990 and 2015 to achieve the Millennium Development Goal 4 (MDG4). Primarily, the National Child Survival Strategy focuses on the health sector, but important distant determinants of child survival, like reducing poverty, improving household food security, raising levels of maternal education and providing safe water and sanitation, are also recognized in the document.
The National Child Survival Strategy has been instrumental in the scaling-up of child survival interventions through the active participation of partners, relevant sectors and the community at large (6). This has wide implications also in terms of poverty reduction. In fact, the focus on child (and maternal) care provides more of a poverty orientation than reliance on other services, since the disease burden at an early age or at childbirth is particularly important among the poor. In fact, not only are death rates higher among the poor, compared to the rich, but the highest poor-rich mortality ratio is observed
for complications of pregnancy and childhood infectious diseases (13).
2) Causes of mortality: what are Ethiopian Children dying from? The Child Survival Strategy identifies the direct causes that are responsible for under-five mortality (Figure 1): pneumonia (28%), neonatal causes (25%), malaria (20%), diarrhea (20%), measles (4%) and AIDS (1%). Underlying conditions are also identified, manly malnutrition and HIV/AIDS (6).
55
55
Neonatal, 25%
Malaria, 20%
Pneumonia, 28%
Diarrhea, 20%
AIDS, 1%
Measles, 4%
Other, 2%
Figure 1. Causes of under-5 deaths in Ethiopia (1). In Ethiopia, neonatal mortality contributes to about one fourth of the overall under-five mortality. Newborns die mainly due to infections,
followed by perinatal asphyxia and prematurity/low birth weight (Figure 2).
Figure 2. Causes of neonatal deaths in Ethiopia (1).
Fortunately, for each target condition there are both preventive and curative interventions that can prevent around 72% of deaths in under five children. To prevent these deaths and to achieve
MDG4, it is necessary to ensure effective implementation of a limited number of interventions that are listed in the National Child Survival Strategy. In 2003, the Lancet child
Other7%
Diarrhoea3%
Congenital4%
Tetanus7%
Infection37%
Preterm birth17%
Asphyxia25%
Causes of neonatal
deaths
Infections and tetanus account for 47% of neonatal deaths in Ethiopia
Malnutrition
57%
HIV/AIDS
56
56
survival series estimated that, with 99% coverage of the high impact interventions for which there is sufficient evidence for effect in prevention or treatment, it would be possible to prevent 65% of deaths due to pneumonia, 55% of deaths due to neonatal complications, 91% of deaths due to malaria, 88% of deaths from diarrhea, 100% of deaths from measles, and 48% of those due to AIDS (12).
3) Millennium Development Goal 4: where do we stand?
Achieving the MDG4 for child survival in Ethiopia demands focused and coordinated action to strengthen the health systems, improve nutrition and reduce inequities in access to effective interventions against all the diseases which kill under-five children (5; 20).
According to the Ethiopia Demographic and Health Survey carried out in 2005, there was an improvement in under 5 mortality rates, with a decrease from 165 to 123 per 1,000 live births between 1990 and 2005 (3). The plan is to decrease U5MR to 54 per 1,000 in the year 2015 to meet MDG4 (Figure 3).
165153
140
123109
95
165153
140
123
89
54
0
40
80
120
160
1990 1995 2000 2005 2010 2015
Years
Un
der
5 M
ort
alit
y R
ate
Current Trend MDG Trend
HSDP I II
Figure 3. Trend in under 5 Mortality Rate in the period 1990-2005 and projections until 2015 in Ethiopia.
57
57
4) Opportunities to achieve MDG4: implementation of child survival interventions An integrated approach is in place in order to achieve MDG 4, with: Focus on the community: Ethiopia is investing a lot in the Health Extension Program (HEP) to reach the poorest of the poor with basic and essential life saving care by focusing mainly on the mothers, newborns and children of the rural population. HEP institutionalizes and standardizes the village health-care delivery system to empower care-takers, families and communities to take care of their own health. It
increases access and utilization of most of the high impact preventive and curative interventions which are listed in the National Child Survival Strategy. Accelerated Expansion of Primary health Care: within the context of HSDP III, the FMOH has initiated an accelerated expansion of PHC services. This new initiative aims to accelerate physical infrastructure expansion, a base for improving access to basic health care services in rural Ethiopia. Besides physical infrastructure expansion, the initiative also entails an increase in the number of health professionals mainly at primary health care unit level.
Child Survival Strategy: HSDP III is incorporated as the de facto health component of the Plan for Accelerated and Sustained Development to End Poverty (PASDEP) (15). Maternal and Child Health (MCH) are major focuses of HSDP III, with Child Survival Strategy being a major component of HSDP III. Availability of Experienced Programs: there are experienced programs relevant to child health in Ethiopia, including EPI, Integrated
Management of Neonatal and Childhood Illnesses (IMNCI), Nutrition, Safe Motherhood and Malaria control programs. Partnership for Maternal, Newborn and Child Health (PMNCH): there is a growing partnership between the Government, UN organizations, bilateral partners, Private institutions and Non-Governmental Organizations for Child Survival.
5) Benchmarks: scaling-up is possible Through proper and rational utilization of available opportunities in the country, it is possible to scale up the implementation of high impact child survival interventions. These are some examples of successes and remarkable achievements in coverage increase within short time periods that can serve as benchmarks to
scaling up of other interventions. These include training of 24,600 HEWs within a period of 3 years, and rapid increase in percentage of children under the age of five years who slept under Long Lasting Insecticide-treated Nets (LLINs) from 1% in 2005 (3) to 42% in 2007 (10). Furthermore, through the Enhanced Outreach Strategy, it has been possible to scale up bi-annual supplementation of Vitamin A from the 2005 baseline level of 45 % to 91% in 2007 (Figure 4).
58
58
Figure 4. Coverage in Vitamin A supplementation among children aged 6-59 months in Ethiopia during the period 2004-07. To realize the continuum of care that starts from the household and continues up to the facility level, the Family Health Department (FHD) of the Federal Ministry of Health (FMOH) launched the case management IMCI training in 1996 and, following that, community IMCI was adopted by a national workshop in 2001. In 2004, 36 % of the Health Centers had at least one health worker trained in IMCI (5). Assessment, classification and management of early neonatal problems were incorporated into the formal IMCI training guideline in 2006 and since then IMCI was renamed as Integrated Management of Newborn and Childhood illnesses (IMNCI). According to the March 2008
IMNCI annual review meeting report of the FHD, 60% of the Health Centers have at least one person trained in IMNCI (11). On the other side, Community-integrated IMNCI (C-IMNCI) coverage has shown a significant increase from 2 woredas in 2004 to 180 woredas in 2008 (11). In the past five years, immunization coverage is also showing an increasing trend. In 2003, DPT3 coverage was 50 % and in 2007 the Pentavalent coverage reached 73% (8). Most importantly, the introduction of Haemophilus influenzae type b (Hib) vaccine in 2007 had brought a paramount benefit to prevent childhood pneumonia and meningitis in under five children (9).
6) Lessons learned and way forward: more of the same is not enough
Recent evidence suggests that, based on current trends, many low-income countries are unlikely to achieve the MDG health target by 2015 (21). This is despite the fact that there are a growing number of cost-effective interventions, as well as increasing international assistance for specific disease control programmes. There is growing consensus that a primary bottleneck to achieving MDGs in low-income countries is health systems that are too fragile and fragmented to deliver the
volume and quality of services to those in need (18). Major shortfalls are identified in the health workforce, lack of donor coordination, and week information systems as critical challenges to achieving MDGs. It is for this reason that HSDP in Ethiopia is focusing on cost-effective health interventions and on health systems strengthening in order to achieve the dual goals of improving population health and reducing health inequalities.
Child Survival: Children (6-59 months) supplemented with Vitamin AAchievements per round as of November 2007
11,423,171 11,926,235
1,389,4382,576,620
4,946,811
11,567,721
8,269,75310,154,375
0
2,000,000
4,000,000
6,000,000
8,000,000
10,000,000
12,000,000
2004 - 1stROUND
2004 - 2ndROUND
2005 - 1stROUND
2005 - 2ndROUND
2006 - 1stROUND
2006 - 2ndROUND
2007 - 1stROUND
2007 - 2ndROUND
Achievements: # of children covered with vitamin A
National target children in Child Survival
Child Survival: Children (6-59 months) supplemented with Vitamin AAchievements per round as of November 2007
11,423,171 11,926,235
1,389,4382,576,620
4,946,811
11,567,721
8,269,75310,154,375
0
2,000,000
4,000,000
6,000,000
8,000,000
10,000,000
12,000,000
2004 - 1stROUND
2004 - 2ndROUND
2005 - 1stROUND
2005 - 2ndROUND
2006 - 1stROUND
2006 - 2ndROUND
2007 - 1stROUND
2007 - 2ndROUND
Achievements: # of children covered with vitamin A
National target children in Child Survival
Child Survival: Children (6-59 months) supplemented with Vitamin AAchievements per round as of November 2007
11,926,23511,423,171
1,389,4382,576,620
4,946,811
11,567,721
8,269,753
10,154,375
0
2,000,000
4,000,000
6,000,000
8,000,000
10,000,000
12,000,000
2004 - 1stROUND
2004 - 2ndROUND
2005 - 1stROUND
2005 - 2ndROUND
2006 - 1stROUND
2006 - 2ndROUND
2007 - 1stROUND
2007 - 2ndROUND
Achievements: # of children covered with vitamin A
National target children in Child Survival
59
59
Training a sufficient number of health professionals and construction of an adequate number of health facilities in a very short time are some of the grand achievements of the FMOH. However, the anatomy alone may not take us to the destination until and unless it is complimented with the physiology. The health
workers that we have trained until now must have the necessary knowledge and skills to fulfill their jobs, and, equally, health facilities should be equipped and supplied with essential items. Moreover, there should be regular and continuous supervision to achieve the ultimate goal of child survival.
Among the major killers, the ones that are poorly addressed are childhood pneumonia and perinatal problems which are the leading killers of under-five children in Ethiopia. According to the 2005 EDHS, only 4.9% of pneumonia cases have
had access to antibiotic therapy and, from the same source, skill delivery coverage was also alarmingly low, with only 6% of total deliveries being conducted by a skilled attendant (3).
The Child Survival Strategy identifies Health Extension Program as an important vehicle that carries most high impact child survival interventions to the community. At this juncture, the potential of HEWs to implementing most of the child survival interventions is not fully exploited. There are multiple reasons for this, including problems related to competency, shortage of supplies, lack of regular supportive supervision, and lack of community ownership. Therefore, the conclusion is that achieving MDG4 in Ethiopia is feasible, but demands addressing the following serious concerns:
• Mobilizing adequate amounts of resources to fully implement the Health Extension Program;
• Realizing the continuum of care approach at delivery level and sustaining it over time;
• Searching for an alternative way of improving access to treatment of childhood pneumonia and essential newborn care, which are the two major killers of the under five children: this may include provision of community-based pneumonia and neonatal infection treatment;
• Strengthening Monitoring and Evaluation, which is the life blood of the child survival strategy to track progress towards the goal and to ensure quality of service rendered to the community.
60
60
References
1. Child health in Ethiopia. Background document for the National Child Survival Conference – April 22-22, 2004, Addis Ababa, Ethiopia. World Health Organization, Addis Ababa.
2. [http://www.afro.who.int/cah/documents/situational_analysis/et_final_cs_situation_analysis.pdf.
Accessed June 2008].
3. CSA, 2006. Ethiopia Demographic and Health Survey 2005. Central Statistic Agency, Addis Ababa and ORC Macro, Calverton.
4. Editorial, 2003. The world’s forgotten children. The Lancet, 361: 1.
5. FMOH, 2004. National review and planning meeting. Report of the Family Health Department of
the FMOH. Federal Ministry of Health, Addis Ababa.
6. FMOH, 2005a. National strategy for child survival. Family Health Department. Federal Ministry of Health, Addis Ababa.
7. FMOH, 2005b. HSDP III. Health Sector Strategic Plan 2005/06-2009/10. Federal Ministry of
Health, Addis Ababa.
8. FMOH, 2007a. Health and health related indicators2006/07. Federal Ministry of Health, Addis Ababa.
9. FMOH, 2007b. Annual performance report of HSDP III. EFY 1999 (2006/2007). Federal Ministry of
Health, Addis Ababa.
10. FMOH, 2008a. National Malaria Indicator Survey. Federal Ministry of Health, Addis Ababa.
11. FMOH, 2008b. IMNCI and C-IMNCI national review meeting. Report of the Family Health Department of the FMOH. Federal Ministry of Health, Addis Ababa.
12. Gareth, J., Steketee, R., W., Black, R., E., Bhutta, Z. A., Morris S. S., and the Bellagio Child
Survival Study Group, 2003. How many children deaths can we prevent this year? The Lancet, 362: 65-71.
13. Gwatkin, D.R., 2000. Health inequalities and the health of the poor: What do we know? What can
we do? Bull. World Health Organ. 78, 3-17.
14. Lawn J., Kerber, K., 2006. Opportunities for Africa’s newborns. Practical data, policy and programmatic support for newborn care in Africa. The partnership for maternal newborn and child health, Cape Town.
15. MOFED, 2007. Ethiopia: Building on progress. A Plan for Accelerated and Sustained Development
to End Poverty (PASDEP). Annual Progress Report 2006/07. Ministry of Finance and Economic Development, Addis Ababa.
16. Save The Children, 2006. New Report shows improvements in child survival in Africa , but more
than a million African babies die in the first month of life. Save The Children, US.
61
61
17. Schuftan C., 1990. The child survival revolution: a critique. Family practice, 7(4):329-332.
18. Travis, P., Bennet, S., Haynes, A., Pang, T., Bhutta, Z., Hyder, A.A., Pielemeier, N.R., Mills, A.,
Evans, T., 2004. Overcoming health-systems constraints to achieve Millennium Development Goals. Lancet, 364: 900-906.
19. UNICEF, 1996. The state of the World Children 1996. The 1980s: Campaign for child survival.
page 1-5. United Nations Children’s Fund, New York.
20. UNICEF, 2006. Preliminary Report. Trends in Child Mortality in Eastern and Southern Africa 1990- 2005. United Nations Children’s Fund, New York.
21. World Bank, 2003. The Millennium Development Goals for Health: Rising to the Challenges (draft).
World Bank, Washington, DC.
62
62
A CASE REPORT: Optic glioma in a child with NF1 Kalid Astrat, 1 MD
Abstract A 10 year old female patient presented with progressive right eye proptosis ( Fig 1) and skin rash of three years duration was seen at Tikur Anbessa Specialized Hospital, department of pediatrics hematology /oncology unit. Physical examination showed mildly decreased visual acuity and cafe au lait spot ( Fig 2) AND axillary freckling and Orbital CT ( Fig 3) showed right intraorbital mass with an assessment of right optic nerve glioma she is to be started on weekly vinblastine at a dose of 6mg/m2.
63
63
INTRODUCTION Optic nerve glioma (also known as optic pathway glioma) is the most common primary neoplasm of the optic nerve. Along with reducing visual acuity in the affected eye, the tumor sometimes produces additional symptoms as it grows. A low-grade form of this neoplasm, benign optic glioma, occurs most often in pediatric patients. Another form, aggressive glioma, is most common in adults; it is frequently fatal, even with treatment.1
Optic-pathway glioma accounts for 1-5% of all brain tumors in children [1]. About half of these cases occur in children with neurofibromatosis type 1 [2]. The diagnosis is usually rendered before age 6 years, although there are some reports of older ages [3,4]. The vast majority of optic-pathway gliomas in children are pilocytic astrocytomas [2,5]. The tumor may arise anywhere along the optic pathway, from just behind the globe to the lateral geniculate body [5,6]. In patients with neurofibromatosis type 1, the tumor is usually smaller than in sporadic (non-neurofibromatosis type 1-associated) cases [5,7].
The clinical presentation is variable. In patients with neurofibromatosis type 1, 40-80% of optic-pathway gliomas are asymptomatic at diagnosis, whereas in sporadic cases, they are symptomatic [3,5,8-10]. The most common signs are vision-related: mainly visual loss, decreased visual acuity, and strabismus. Other findings include endocrine disturbances, signs of increased intracranial pressure, and hydrocephalus [2,5,8,9,10-13]. Ophthalmologic examination may reveal decreased visual acuity, pathologic visual fields, proptosis and more [9,10,13,14]. The generally young age of the children and high prevalence of neurofibromatosis type 1-associated attention deficit hyperactivity disorder render the ophthalmologic examination difficult, and decrease its sensitivity and specificity [2,15]. Early diagnosis is important so that the tumor can be carefully monitored and treatment can be administered early, before visual deterioration. The diagnosis can be made functionally by visual-evoked potentials, but as is
the case for eye examinations, their efficacy is limited, and specificity is low [2,16]. Modern neuroimaging modalities provide excellent characterization of optic-pathway gliomas, obviating the need for biopsy [17]. Magnetic resonance imaging was found to be superior to computed tomography for the detection and evaluation of extensive tumor involvement. It has a higher specificity, and can be used to assess disease progression [18,19]. The biological behavior of optic-pathway glioma varies. The tumor may progress rapidly, remain stable for years, or even shrink spontaneously or after biopsy, mostly with clinical improvement [2,5,7,20]. Regrowth after a stable period or after biopsy was also reported [2]. Tumor progression is apparently affected by the presence of neurofibromatosis type 1, patient age, and tumor location [21]. Less progression was evident in patients with neurofibromatosis type1 than in sporadic cases, and in children who were older at diagnosis [2,5,7,9,10,19,22]. Tumors situated at the optic nerve or chiasma tend to grow more slowly and less aggressively than chiasmatic/hypothalamic gliomas, with lower mortality [23,24]. Posterior involvement may also lead to significant morbidity and mortality [6,14,25].The natural 1 Department of Pediatrics and Child Health, Medical faculty, Addis
Ababa University. history of optic-pathway gliomas in neurofibromatosis type 1 is considered unpredictable [2,26]. Hence deciding whether, or when, to initiate treatment becomes difficult. The presence of an optic-pathway glioma in a patient without neurofibromatosis type 1 is considered an indication for treatment [5]. Although neurofibromatosis type 1 is thought to be relatively benign, given the risk of visual impairment, blindness, neurologic deficits, or death [27,28], patients affected by the tumor should be monitored routinely for its size and visual function, and an adverse change in either should be considered an indication for treatment [2]. If a glioma tends to remain stable, the
64
64
intervals between magnetic resonance examinations can be gradually increased [10]. Nevertheless, the subjective timing of imaging scans and the lack of objective references to identify deviations from normality place patients at risk of either unnecessary or insufficient neuroimaging. An optic-pathway glioma tends to grow along the optic pathways by increasing its width, rather than as one ‘‘concentric’’ mass that grows in all directions [5]. As such, stereotypical patterns of growth as seen on imaging scans of children with optic-pathway glioma are often highly comparable, because the tumor tends to involve the same brain structure, and the manner of growth is very similar.
CASE REPORT
A 10 year old female patient presented with progressive right eye proptosis and skin rash of three years duration was seen at Tikur Anbessa Specialized Hospital, department of pediatrics hematology /oncology unit. Since the last 6 months the proptosis was more progressive to attain the current size. Family history is positive for paternal unilateral loss of vision unrelated to trauma. Physical examination showed mildly decreased visual acquity (OD =6/9, OS =6/6) and cafe au lait spot( Fig 2) and axillary freklings other wise no other findings on the musculoskletal and CNS. Orbital CT( Fig 3) showed right intraorbital mass with an assessment of right optic nerve glioma she was started on weekly vinblastine at a dose of 6mg/m2.
Fig 1
65
65
Fig 2
66
66
Fig 3 Discussion This is one of the rarely reported case of a 10 year old Ethiopian child with type 1 neurofibromatosis and right optic nerve glioma. In most young patients with optic glioma the presenting symptom is painless proptosis. Optic atrophy is common, as is reduced visual acuity, although the latter may be a late symptom. A large lesion may compress the optic chiasm, causing nystagmus or other symptoms. Hypothalamic symptoms, such as changes in appetite or sleep, also may occur. Massive lesions may compress the third ventricle, resulting in obstructive hydrocephalus accompanied by headache, nausea, and vomiting also may occur but these findings were not found in this patient. Historically, surgery and radiotherapy have played a primary role in management, however, in the last 15 years, chemotherapy has evolved into the first-line treatment of choice. The case presented was started on weekly Vinblastine at a dose of 6 mg/m2 for one year with regular
ophthalmologic and Orbital CT/MRI if feasible.
67
67
References 1. Alshail E, Rutka JE, Becker LE, Hoffman HJ. Optic chiasmatic-hypothalamic glioma. Brain Pathol
1997;7:799-806. 2. Shuper A, Horev G, Kornreich L, et al. Visual pathway glioma: An erratic tumor with therapeutic
dilemmas. Arch Dis Child 1997;76: 259-63. 3. Thiagalingam S, Flaherty M, Billson F, North K. Neurofibromatosis type 1 and optic pathway
gliomas: Follow-up of 54 patients. Ophthalmology 2004;111:568-77. 4. Listernick R, Ferner RE, Piersall L, Sharif S, Gutmann DH, Charrow J. Late-onset optic pathway
tumors in children with neurofibromatosis1. Neurology 2004;63:1944-6. 5. Kornreich L, Blaser S, Schwarz M, et al. Optic pathway glioma: Correlation of imaging findings with
the presence of neurofibromatosis. AJNR 2001;22:1963-9. 6. Liu GT, Brodsky MC, Phillips PC, et al. Optic radiation involvement in optic pathway gliomas in
neurofibromatosis. Am J Ophthalmol 2004;137:407-14. 7. Astrup J. Natural history and clinical management of optic pathway glioma. Br J Neurosurg 2003;17:327-35. 8. Guillamo JS, Creange A, Kalifa C, et al. Prognostic factors of CNS tumors in neurofibromatosis 1
(NF1): A retrospective study of 104 patients. Brain 2003;126:152-60. 9. Czyzyk E, Jozwiak S, Roszkowski M, Schwartz RA. Optic pathway gliomas in children with and
without neurofibromatosis 1. J Child Neurol 2003;18:471-8. 10. Listernick R, Charrow J, Greenwald M, Mets M. Natural history of optic pathway tumors in children
with neurofibromatosis type 1: A longitudinal study. J Pediatr 1994;125:63-6. 11. Cnossen MH, Stam EN, Cooiman LC, et al. Endocrinologic disorders and optic pathway gliomas in
children with neurofibromatosis type1. Pediatrics 1997;100:667-70. 12. Shuper A, Kornreich L, Michowitz S, Schwartz M, Yaniv I,Cohen IJ. Visual pathway tumors and
hydrocephalus. Pediatr Hematol Oncol 2000;17:463-8. 13. Khafaga Y, Hassounah M, Kandil A, et al. Optic gliomas: A retrospective analysis of 50 cases. Int J
Radiat Oncol Biol Phys 2003;56:807-12. 14. Sigorini M, Zuccoli G, Ferrozzi F, et al. Magnetic resonance findings and ophthalmologic
abnormalities are correlated in patients with neurofibromatosis type 1(NF1). Am J Med Genet 2000;93:269-72.
68
68
15. Wolsey DH, Larson SA, Creel D, Hoffman R. Can screening for optic nerve gliomas in patients with neurofibromatosis type I be performed with visual-evoked potential testing? J AAPOS 2006;10:307-11.
16. North K, Cochineas C, Tang E, Fagan E. Optic gliomas in neurofibromatosis type 1: Role of visual
evoked potentials. Pediatr Neurol 1994;10:117-23. 17. Pepin SM, Lessell S. Anterior visual pathway gliomas: The last 30 years. Semin Ophthal. 2006;21:117-24. 18. Van Es S, North KN, McHugh K, De Silva M. MRI findings in children with neurofibromatosis type
1: A prospective study. Pediatr Radiol 1996;26:478-87. 19. Chateil JF, Soussotte C, Pedespan JM, Brun M, Le Manh C, Diard F. MRI and clinical differences
between optic pathway tumours in children with and without neurofibromatosis. Br J Radiol 2001;74: 24-31.
20. Parsa CF, Hoyt CS, Lesser RL, et al. Spontaneous regression of optic gliomas: Thirteen cases
documented by serial neuroimaging. Arch Ophthalmol 2001;119:516-29. 21. Chan MY, Foong AP, Heisey DM, Harkness W, Hayward R, Michalski A. Potential prognostic
factors of relapse-free survival in childhood optic pathway glioma: A multivariate analysis. Pediatr Neurosurg 1998;29:23-8.
22. Grill J, Laithier V, Rodriguez D, Raquin MA, Pierre-Kahn A, Kalifa C. When do children with optic
pathway tumors need treatment? An oncological perspective in 106 patients treated in a single center. Eur J Pediatr 2000;159:692-6.
23. Schroder S, Baumann-Schroder U, Hazim W, Haase W, Mautner VF. Long-term outcome of
gliomas of the visual pathway in type 1 neurofibromatosis. Klin Monatsbl Augenheilkd 1999;215:349-54.
24. Tow SL, Chandela S, Miller NR, Avellino AM. Long-term outcome in children with gliomas of the
anterior visual pathway. Pediatr Neurol 2003;28:262-70. 25. Balcer LJ, Liu GT, Heller G, et al. Visual loss in children with neurofibromatosis type 1 and optic
pathway gliomas: Relation to tumor location by magnetic resonance imaging. Am J Ophthalmol 2001;131:442-5.
26. Serova NK, Lazareva LA, Gorelychev SK, Ozerova VI, Pronin IN. A follow-up of patients with
anterior optic tract glioma concurrent with type 1 neurofibromatosis. Vestn Oftalmol 2006;122:39-42.
69
69
27. Luh GY, Bird CR. Imaging of brain tumors in the pediatric population. Neuroimag Clin North Am 1999;9:691-716.
28. Kosa E, Csakvary V. Neurofibromatosis type 1 in children—With special consideration of
ophthalmologic symptoms. Orv Hetil 2004;145:473-8. Acknowledgments I would like to thank Dr David N Korones pediatric oncologist at Rocester medical center , New York and Dr Ibrhaim Qaddumi , oncologist at St Jude Children’s Research hospital for their constructive ideas and providing me with important references.
70
70
Clinical Predictors of Pneumonia Among Under-five Children At Tikur Anbesa Specilaized Hospital
Kalid Asrat1, MD, Amha Mekasha1, MD, MSc
Abstract
The aim of the study is to identify simple clinical signs and symptoms in under five children which are predictors of pneumonia at Tikur Anbesa Specialized Hospital (TASH), Department of Pediatrics and Child Health, emergency and regular out-patient units. The design of the study is a prospective cross-sectional study carried out during Aug 2004 – Sep 2005. All children between the age of 2-59 months who attended the regular and emergency pediatrics units at TASH during the study period who had either cough or difficulty of breathing or chest x-ray evidence of pneumonia were included in the study. A calculated sample of 164 was taken. Data analysis was done using SPSS and EPINFO version 1.1.2. Chi-square test was used to calculate the differences in distribution of clinical signs and symptoms between groups with and without chest x-ray (CXR) evidence of pneumonia. A total of 179 patients were studied of whom 102 were males and 77 females (M:F 1:0.75). Clinical symptoms and signs were related to CXR pneumonia. Tachypnea (94.5% with CXR pneumonia Vs 59.3% without CXR pneumonia) and retraction (86.3% CXR Vs 40.6% without CXR pneumonia) were the best clinical predictors of pneumonia. Tachypnea, flaring of alae nasi and retraction were also independently associated with CXR pneumonia. Vomiting, refusal to feed, history of rapid breathing , grunting and chest findings did not predict pneumonia. In conclusion pneumonia is a significant cause of morbidity and mortality in developing countries like
Ethiopia . Using simple clinical indicators pneumonia can be diagnosed by primary health workers and mortality can be
reduced significantly.
71
71
INTRODUCTION Around 10.6 million children die every year before reaching their 5th birth day. Almost all of these deaths occur in low income and middle income countries mainly in Africa and south east Asia. Most deaths among under fives are still attributable to just a handful of conditions, acute respiratory infection(ARI) mostly pneumonia accounts for 19% of all deaths(1,2,3,). Age Specific mortality from lower respiratory infection( LRI) in young Gambian children has been estimated at 10 per 1000 each year(4 ) . In Ethiopia infant mortality rate is 77 per 1000, under five mortality is 120 per 1000 (5), ARI being one of the major cause of morbidity and mortality. Taking these into consideration WHO in 1981 initiated a programme for the control of ARI on a case management of pneumonia. One of the strategies is to improve case detection and patient management by primary health care workers (7) and so reduce mortality. The aim of this study is to evaluate the usefulness of simple clinical symptoms and signs in the diagnosis of LRI, mainly pneumonia, which can easily be used by primary health care workers. Despite LRI (mainly pneumonia) being a condition commonly encountered by clinicians, uncertainty remains over the
diagnosis, investigation and treatment of the condition. Infants and children may present with a number of different clinical symptoms and signs such as fever, cough and tachypnea. Minority of children may present with fever of unknown origin ( FUO) and may have no respiratory symptoms or signs.
WHO has developed algorithm (8) to aid medical and non-medical health care workers in diagnosis of LRI with out radiological confirmation. The WHO algorithm stresses the importance of tachypnea which has a 74% sensitivity and 67% specificity for radiologically defined pneumonia (8). However, in children who had diseases for less than three days (9), tachypnea had a lower sensitivity and specificity of illness. Clinicians must be aware that the absence of tachypnea does not necessarily mean the absence of pneumonia (10 ). Grunting and nasal flaring increase the chance of pneumonia, but their absence cannot be relied upon to rule out the chance of pneumonia (9). Other signs that relate to severity of pneumonia are chest indrawing, nasal flaring and cyanosis. High fever in young children (age up to 3 years) was found to be a sign of pneumonia (11,12).The British thoracic
1 Department of Pediatrics and Child Health, Medical faculty, Addis
Ababa University. society ( BTS) guidelines suggested that in children less than 3 years combination of fever > 38.5ºc,chest indrawing and RR >50/min indicate pneumonia, breathing difficulty is more reliable sign in older children (18). Chest x-ray (CXR) is still considered to be the gold standard for diagnosing pneumonia in the developed world. However, there is poor concordance between radiological changes which constitute pneumonia( inter and intra observer variation), consolidation on CXR was most commonly identified by the radiologist was generally agreed to represent pneumonic changes(12) . WHO has recognized the difficulties with CXR interpretation and developed a tool to standardize the reporting of CXR use in epidemiological studies of pneumonia (8) this system classifies CXR as: normal appearance, infiltrates and end stage consolidation defined
as significant amount of alveolar type of consolidation. So does a normal x-ray rule out pneumonia? There is anecdotal evidence for having pneumonia with a normal CXR. Fever and tachypnea may present before CXR changes are seen. Subjects and methods Study setting: The study was conducted at Tikur Anbessa Specialized Hospital (TASH), department of pediatrics and child health emergency and regular OPD which is one of the very few centers in the country which gives pediatric out patient and in patient services. Study design: a cross sectional survey was under-taken among children under five children in the period of (AUG 2004- SEP 2005) who have either cough or difficulty of breathing or CXR evidence of pneumonia
Study population: All children between the age of 2-59 months who attended the regular as well as the emergency pediatrics OPD at TASH during the study period(AUG 2004- SEP 2005) who have either cough or difficulty of breathing or CXR evidence of pneumonia were included. A calculated sample of 164 was taken. Data analysis: Data was entered and analysis was done using SPSS and EPINFO version 1.1.2 software. Chi-
square test was used to calculate the differences in distribution of clinical signs and symptoms between groups with and without CXR pneumonia. Signs and symptoms which were significant on univariate analysis were taken for multiple logistic regression analysis. Ethical consideration: Permission to undertake the study was obtained from the department of pediatrics and child health of TASH.
Results
72
72
A total of 179 children among whom 102 (57%) males and 77 (43%) females who fulfilled the inclusion criteria’s were
included in this study. The age distribution is shown in table 1 where half of the patients were infants.
21 patients have severe PEM,13 patients have rickets,3 have sero-proven pediatrics HIV infection and one patient each have gastroenteritis, malaria, downs syndrome and meningitis (table 3). Table 2 shows clinical signs that best correlate with CXR
pneumonia, these are tachypnea, flaring and retraction ,the other clinical signs and symptoms like vomiting, refusal to feed, cough, fever and grunting and chest findings did not
correlate with CXR pneumonia. Table 4 shows the
sensitivity, specificity, 95% confidence intervals and prevalence of selected symptoms and signs and table 5
shows the frequency of clinical signs and symptoms. In children between the age of 2-59 months, tachypnea (94.5% with CXR pneumonia Vs 59.3% without) and retraction (86.3% Vs 40.6%) were the best predictors of CXR pneumonia. Tachypnea, flaring and retraction were independently associated with CXR pneumonia.
73
73
74
74
75
75
Discussion In rural areas of developing countries the case fatality rate from LRI in children is most likely to be reduced if primary health care workers can identify the most serious forms of LRI and deal with them appropriately, accurate guidelines to detect LRI(pneumonia) who can be safely treated with antibiotics as outpatients from those who require immediate referral must be based on symptoms and signs that can be readily assessed. This study has attempted to look into clinical signs and symptoms predictors of pneumonia in less than 5 year children such as vomiting rapid breathing, tachypnea and chest retraction, age and sex distribution ,frequency of each clinical signs and symptoms at emergency and regular pediatric out-patient department of TASH over one year period. On a study done by Campbell, et al showed that in infants a fever of >38.5ºc, refusal to feed (breast feeding) or the presence of vomiting best correlated with CXR pneumonia. In children aged 1-4 year, a fever of > 38.5ºc, RR >60/min are best correlated with CXR pneumonia(11).Another study done by Cherian et al showed that RR >50/min in infants and RR >40/min in
children 12-35 months of age, as well as history of rapid breathing and the presence of chest retraction in both age groups were found to be sensitive and specific indicators of LRI. Increased RR and history of rapid breathing were also sensitive in diagnosing less severe LRI that did not necessitate admission to the wards, whereas chest retraction was not. All these clinical signs had a lower sensitivity in diagnosing LRI in children aged 36 months and over (19).Study done by Shan et al showed that chest indrawing was a reliable sign in children between the age of 0-4 years with cough for assessment of severe LRI (pneumonia) and among other children with cough a RR of >50/min is a reliable basis for diagnosing LRI (14). A multicentric study done by the WHO young infant study group indicated that the best threshold for predicting pneumonia in infants aged less than 2 months was RR > 60/min( 15).The best combination of sensitivity( 78-82%) and specificity (73-89%) was achieved by wing thresholds of 50 and 40 breaths/min for children aged 2-11 months and 1-4 years, respectively (16).
The Bangladesh study suggested that chest indrawing was more specific as a sign of severe pneumonia (17).Tachypnea has been exhaustively demonstrated to be an excellent predictor of radiologically defined pneumonia in a study done by Levantine J et al (10,14). In the present study tachypnea and retraction are very specific signs of CXR pneumonia, this is also consistent with other studies done on the subject(10,14,17,18) and their usefulness is further increased by their high sensitivity and are also easier to teach to non medical staff like primary health care workers. Vomiting, rapid breathing, fever, cough and grunting were not satisfactory predictors of CXR pneumonia.
Since our study is a hospital based study it should be interpreted with caution when community health workers (CHW) training and national guidelines are setup. Such programmes must adopt policies that take into consideration the health resource s available to implement them. In rural areas of many developing countries including Ethiopia the referral systems are poorly developed and much of the primary care must be developed to CHWs, this should come from community based studies. In conclusion pneumonia is still the significant cause of morbidity and mortality in developing countries like Ethiopia . If pneumonia is diagnosed earlier using simple clinical parameters which can also be used by non medical staff like community health workers and other health professional at each level mortality could be reduced significantly. Tachypnea (94.5%) and retraction(86.3%) were the best predictors of pneumonia as evidenced by
suggestive CXR (CXR pneumonia). Further studies should be conducted on the subject especially at the community level.
76
76
References
1. World health report 2005 2. Demographic estimates of the annual number of infants and child defense in the
world.The pediatric health 1984. 3. Handbook of health information of India. Central Bureaus of health Intelligence, New
Delhi,1984. 4. Deaths in infancy and early childhood in well vaccinated rural west African population.
Ped 1994 ,91-99. 5. MOH. Health and health related indicators.1998. 6. Williams B et al estimate of world wide distribution of deaths from ARI. Lancet inf dis J
2002,2:25-32. 7. WHO. A programme for controlling ARI in children memorandum,WHO meeting. Bull
WHO 1994 ,62:45-58. 8. WHO ,management of ARI in children ,practical guidelines for outpatient care.
Geneva WHO 1995. 9. Plafox M etal diagnostic value of tachypnea in pneumonia defined radiologically. Arch
Dis child 2000,82:41-5 10. Leventhal J. clinical predictors of pneumonia as a guide to ordering chest X-ray.
Clinical pediatr 1982;21; 730-40 11. Campbell H et al .Assessment of clinical criteria for identification severe LRI in
children ,the lancet 1989,297-9 12. Campbell SM, Shann F, et al. The effect of panel membership and feedback in ratings
in a two round Delphi survey ,results of randomized controlled trial ,medicinal care 1999,37,964-8
13. Davies H et al. Reliability of chest X-ray of LRI in young children. pediatr inf dis 1996,16,600-4
14. Shann F et al. LRI in children, possible criteria for selection of patients for antibiotic therapy and hospital admission, bull of the WHO 1984,62,749-53
15. The WHO young infant study group, clinical prediction of serious bacterial infection in young infants in developing countries. Pediatr inf dis J 1999,18 suppl 8: 523-31
16. Mulholland EK et al. Standardized diagnosis of pneumonia in developing countries. Pediatr inf dis J 1992,11,77-81
17. Kalteer HD et al. Identifying sick children requiring referral to hospital in Bangladesh .Bull WHO 1997,75 supp1,65-75
18. BTS/SIGN ,British guideline on the management of Asthma Thorax 2003,58: supp 1 19. Cherian T et al evaluation of simple clinical signs for the diagnosis of LRI Lancet 1988 2:125-8
77
77
Instructions to contributors
1. All material submitted to the considered for publication must be submitted
exclusively to the Journal. All manuscripts should be typed in double spacing
and should be sent in triplicates accompanied by an electronic copy to the
Editor- in- chief of Journal.
2. Abstracts should be prepared in a structured format of about 250 words must
be included. The abs tract should include: objectives(S) ,design, setting,
subjects of participants invitations, interventions, main outcome measures,
result and conclusion.
3. Format and style: Original articles should include: abstract, Introduction
materials and methods, Results and Discussion, Acknowledgements and
References.
4. Abbreviations may be used after the long form has been written, e.g. the
world health Organization (WHO) united nation (UN). Drugs should be
referred to be approved and proprietary names. Scientific measurements
should be given in SI units. Tables and Figure should be typed on separated
Sheet. It is advisable that table s should not exceed five in number.
5. The title page should contain (i) the title of the article, (ii) Author (s) name (s)
and signature (s), degrees, designation, name of institution where the work
was carried out, present mailing address, telephone and E- mail and (iii)
name and address of the author to whom all correspondences should be
addressed.
6. References made to Journal should use the name of the journal abbreviated
according to the style used in Index Medic us, IT should follow the Vancouver
style: Year, volume, first and last page number E.g. Samson T. Lakech G
.Malnutrition and enteric parasitizes among under five children in Aynalem
Village, Tigray, Ethiop J HLth Dev 2000; 14:67-75
7. References made to books should include the authors or editors last name.
Initils, titles edition number, place of publication, publisher and year E.g.
Hailu B,Ali H. Chronic cough in children. In the tropics. Addis Ababa Medical
books pub 2006: 110-120
78
78
8. All manuscripts submitted to the Journal per peed reviewed. Manuscripts
are sent to two external referees selected from pool of experts.
9. All authors must give signed consent to publication. Authors are responsible
for the views, opinions and authenticity of the material published in the
journal.
10. Rejected articles will not be returned to the authors
Acknowledgements The printing of the Journal was made possible through the generous contribution of UNICEF Ethiopia