ARTICLES

Subtotal perforations have a greater likelihood of failure,while small perforations had a higher than expected successrate. The overall success rate of 78% compares favourablywith the literature, and the relatively short follow-up periodshould not compromise these results because, as stated byPacker et al. ,'4 the success of both take rate and hearingresults was no different at 6 months than in the longer termover 1 - 2 years.

The hearing results indicate that myringoplasty is abeneficial procedure, with a large majority of patientsachieving socially acceptable hearing in the operated ear,viz. 79% (122 of 154 patients).

The incidence of postoperative sensorineural cochleardamage is in keeping with the literature, and can beascribed to excessive manipulation of the ossicles duringthe procedure. Minimal bone drilling was performed onthese patients. We conclude that myringoplasty is abeneficial procedure, with regard to both the successfulclosure of the tympanic membrane perforation andimprovement of the hearing.

REFERENCES

1. Wullstein H. The theory and practice of tympanoplasty. Laryngoscope 1956: 66:1076-1093.

2. Zoellner F. The principles of plastic surgery of the sound-conducting apparatus. JLaryngol Otol 1956; 69: 637-652.

3. Austin D, Shea J. A new system of tympanoplasty using vein graft. Laryngoscope1961; 71: 596-611.

4. Hough J. Tympanoplasty with the interior fascial graft technique and ossicularreconstruction. Laryngoscope 1970; 80: 1385-1413.

5. Liele D, Sheehy J. Standard classification for surgery of chronic ear infection.Arch Otolaryngo/ 1965; 81: 204-205.

6. Smyth G. Sensorineural hearing loss in chronic ear surgery. Ann Otal 1977; 86:3-8.

7. Ojala Kt Sorri M. Late post-operative hearing results correlated with the severityof tissue changes in ears with chronic otitis media. J Laryngol Oto/1 983; 97:131-139.

8. Wehrs R. Hearing results in tympanoplasty. Laryngoscope 1985; 95: 1301-1306.9. Harder H, Jervall L, Kylen P, Ekvall L. Calculation of hearing results after

tympanoplasty. Clin Otolaryngo/1982; 7: 221-229.10. Bluestone C, Cantekin E, Douglas G. Eustachian tube function related to the

results of tympanoplasty in children. Laryngoscope 1979; 89: 450-458.11. Tos M, Lau T, Plate S. Sensorineural hearing loss following chronic ear surgery.

Ann Otol Rhinal Laryngol 1984; 93: 403-409.12. Bellucci A. Cochlear hearing loss in tympanoplasty. Otolaryngol Head Neck Surg

1985; 93: 482-485.13. PaJva T, Karja J, Palva A. High-tone sensorineural losses following chronic ear

surgery. Arch Otolaryngo/1973; 98: 176-178.14. Packer P, Mackendrick A, Solar M. What's best in myringoplasty: underlay or

overlay, dura or fascia? J LaryngoJ Otal 1982; 96: 25-41.15. Blanshard J, Robsen A, Smith I, Maw A. A long term view of myringoplasty in

children. J Laryngol Oto/1990; 104: 758-762.16. Puhakka H, Virolainen E, Rahka T. Long-term results of myringoplasty with

temporalis fascia. J Laryngol Otol 1979; 93: 1081-1086.17. Adkins W, White B. Type 1 tympanoplasty: influencing factors. Laryngoscope

1984; 94: 916-918.18. Palva A. Karja J. Eustachian tube patency in chronic ears - pre-operative

evaluation correlated to postoperative results. Acta Otolaryngol 1970; 263: 25-28.19. MacKinnon D. Relationship of pre-operative Eustachian tube function to

myringoplasty. Acta Otolaryngo/ 1970; 69: 100-106.20. Tos M. Tympanoplasty and age. Arch btolaryngol 1972; 96: 493-498.21. Tos M. Lau T. Stability of tympanoplasty in children. Otolaryngol eJin North Am

1989; 22: 15-28.22. Liden G. Nilssan G. Anderson H. Narrow band masking with white noise. Acta

Otolaryngol 1959; 50: 116-124.

Accepted 28 Sep 1993.

Macrosomia - maternaland fetal risk factorsJ. K. Essel, E. T. Opai-Tetteh

Risk factors associated with fetal macrosomia were

studied in 348 pregnancies resulting in the delivery of an

infant weighing 4 000 g or more in a black population.

Identifiable maternal risk factors included a mother in her

3rd decade of life, multiparity, maternal weight of 70 kg or

more at the end of pregnancy, prolonged or post-term

pregnancy, abnormal glucose tolerance and previous

history of a macrosomic infant. Male infants had a higher

risk of being macrosomic. Macrosomic infants accountedfor 3,4% of all singleton deliveries, with their caesarean

section rate of 33,9% being almost three times that of

control infants. The importance of antenatal prediction of

fetal weight is emphasised and suggestions for reduction

of the high perinatal mortality and morbidity rates, as well

as maternal morbidity, are discussed.

S Afr Med J 1995; 85: 43-46.

Little attention has been paid to fetal macrosomia in blackAfrican populations, despite the fact that as a high-riskfactor in pregnancy and delivery macrosomia probablydeserves as much attention as is currently accorded lowbirth weight, albeit for different reasons. Although perinatalmortality has decreased considerably in recent years,' therisks to mother and fetus are serious enough to warrantearly detection of macrosomia during labour, and preferablyduring pregnancy. In fact, Parks and Ziel2 suggested that therisks are sufficiently high to justify consideration of electivecaesarean section for delivery of such infants. The onlystudy emanating from Africa known to the authors is that byOkpere et al. 3 from Nigeria.

Our study attempted to identify maternal and fetal riskfactors associated with macrosomia in a black Africanpopulation from another geographical region of Africa. Wedefined a macrosomic infant as one weighing 4 000 g ormore at birth, the currently accepted definition in thedeveloped world. 4

Patients and methodsUmtata General Hospital, Eastern Cape, serves a populationof predominantly low and medium socio-economic status.Between 1 February 1989 and 31 August 1990, 10 507singleton infants were delivered at the hospital, of whom 360were macrosomic. Complete documentation was available

Department of Obstetrics and Gynaecology, University of Transkeiand Umtata General Hospital, Umtata, Eastern Cape

J. K. Essel, M.B. CH.B., FWA.C.S.

E. T. Opai-Tetteh, M.B. CH.8., D.T.M.&H., D.f'H.

SAMJ Volume 85 No. 1 January 1995

Results

for 348 of the 360. An equal number of babies, eachdelivered within minutes of a macrosomic infant butweighing between 3 000 and 3 200 g, were selected ascontrols. Mothers and infants in the macrosomic group werecompared with those in the control group in respect of age,height, parity, weight at delivery, previous history ofmacrosomia, glucose tolerance, pregnancy complications,mode of delivery, intrapartum and postpartum complicationsand neonatal outcome. All pregnancies included in the studyhad to be at least term by dates (i.e. ;. 37 completedweeks' gestation), irrespective of booking status or place ofprenatal care.

Results were analysed by the simple X2- test. Values of

P < 0,05 were statistically significant.

The 360 macrosomic infants delivered in the study periodconstituted 3,43% of all singleton deliveries; 327 of theirmothers were booked patients, compared with 329 of thecontrol mothers.

Table I sets out the identified characteristics of themothers in the two groups. There was a statisticallysignificant increased incidence of macrosomia in the agegroup 30 - 39 years. Height did not appear to be asignificant factor, since about 50% of the macrosomicinfants' mothers were less than 160 cm tall. The incidence ofmacrosomia generally increased with parity and becamemore obvious in the grand multiparous group. Pre­pregnancy maternal weight and weight gain duringpregnancy could not be ascertained because most mothersbooked in the second trimester. However, a weight of 70 kgor more at the end of pregnancy was found to be asignificant risk factor, as was a previous history ofmacrosomia (28,4% of mothers in the macrosomic group,compared with 1,1 % of controls).

Table I. Maternal characteristics

Macrosomic infants Controls

------ --- --------

Of the 184 mothers in the macrosomic group (52,9%) whohad a glucose tolerance test (GTT) performed, 34 (18,5%)had an abnormal result and 19 (10,3%) were classified asgestational diabetics. The prevalence of gestational diabetesamong 51 mothers whose infants weighed 4 500 g or morewas 9,8%.

Complications of pregnancy in the two groups are set outin Table 11. Post-term delivery occurred in 16,4% of themacrosomic group, compared with 2,9% of controls.

Table 11. Complications of pregnancy

Macrosomic infants Controls

No. % No. %

Post-term pregnancy 57 16,4 10 2,9"

Hypertension/pre-eclampsia 18 5,2 13 3,7

Abruptio placentae 2 0,6 2 0,6

Premature rupture of 4 1,1 5 1,4membranes• P< 0,05.

Table III shows that the caesarean section rate was almostthree times higher in the macrosomic group than among thecontrols. Cephalopelvic disproportion accounted for 90% ofabdominal deliveries of macrosomic infants. No forcepsdeliveries were performed in either group. Thirteen of 22mothers of macrosomic infants who had a prolonged firststage of labour were given oxytocin and subsequently had aspontaneous uncomplicated vaginal delivery.

Table Ill. Mode of delivery

Macrosomic infants Controls

No. % No. %

Spontaneous vaginal 219 62,9 297 85,3deliveryVacuum extraction 11 3,2 7 2,0Caesarean section 118 33,9 44 12,6"• P< 0,005.

No. % No. %

Age (yrs) The incidence of uterine rupture was higher in the

< 20 20 5,7 93 26,7 macrosomic group (8,6/1 000) than among the controls

20- 29 172 49,4 185 53,2 (2,9/1 000). Of the patients who had vaginal deliveries only

30 - 39 139 39,9 62 17,8* 1 mother of a macrosomic infant developed primary;. 40 17 4,9 8 2,3 postpartum haemorrhage, compared with 4 in the control

Height (cm) group; 3 cases of second-degree perineal laceration

,,; 160 cm 170 48,9 253 72,7 occurred in the macrosomic group, with 2 in the control

> 160 cm 178 51,2 95 27,3 group. There was no maternal death in either group.

Parity Table IV shows that 224 of the macrosomic infants were

0 39 11,2 130 37,4 male (male/female ratio 1,8:1,0); the ratio in the control1 - 4 226 64,9 174 50,Ot group was 1,07:1. The incidence of birth asphyxia (defined;.5 83 23,9 44 12,6" as a 1-minute Apgar score of less than 7) among

Weight at delivery (kg) macrosomic infants delivered vaginally was 4,6%, compared< 70 74 21,3 248 71,3 with 5,2% in the control group. There was no difference in70 - 90 208 59,8 95 27,3* the incidence of meconium aspiration. Shoulder dystocia;. 90 66 19,0 5 1,4* complicated the delivery of 13 (3,7%) of the macrosomic

Previous macrosomia 99 28,4 4 1,1* infants: 1 of these developed Erb's palsy in the neonatalAbnormal GTT 34 18,5 0 0 period, while another had clavicular fracture. When fetuses

·P<0,01. who were dead when the mother arrived at the hospital weretP< 0,05. excluded, the neonatal mortality rate for the macrosomic

infants was 23/1 000, compared with 11,5/1 000 for the

Volume 85 No. I January 1995 SAMJ

ARTICLES

control group. Six of 9 perinatal deaths in the macrosomicgroup were due to avoidable causes, viz. 3 cases ofruptured uterus, 2 cases of obstructed labour, and 1 case ofeclampsia. All of these mothers had been referred fromperipheral district hospitals.

Comparison of the incidence of macrosomia among differentracial groups is fraught with difficulties because of thedifferent birth weights and denominators used by variousauthors.3.5-? In recent years a birth weight of 4000 g andabove has been used by most workers. Using this definition,Modanlou et al.· observed an increase in the incidence ofmacrosomia from 7% in 1960 to 10,7% in 1980. Ourincidence of 3,42% is similar to the 3% reported byMarinho? from Nigeria but lower than the 5,5% reported byGross et al.· among 7 123 infants in the USA. While thelower incidence in Africa is difficult to explain, it may berelated to the generally lower pre-pregnant weights of blackAfrican women. The preponderance of male infants in ourstudy is in accord with findings from other studies.'·3

In order to make the diagnosis of fetal macrosomiaantenatally, it is vital to be aware of the predisposing factors.In our environment these were found to be as follows: amother in her 3rd decade of life who is multiparous or grandmultiparous, weighs over 70 kg at term, and has a previoushistory of a macrosomic infant, a current post-termpregnancy, and gestational diabetes. In fact our 16,4%incidence of post-term or prolonged pregnancy is similar tothe 16% reported by Boyd et al. ' The above risk factors areall similar to those identified by other workers.3.• LikeModanlou et al.: our study also confirmed the significanceof a previous history of macrosomia.

The role of maternal height remains controversial.Although our study failed to establish any associationbetween maternal height and macrosomia, Boyd et al.' andOkpere et al. 3 found height above 169 cm and 160 cm,respectively, to be significant risk factors. Although otherstudies,,3 have reported a weight gain in pregnancyexceeding 13 kg at term as an important risk factor, wecould not be certain that it applied in our study population,because most of our patients booked late for antenatal care.The incidence of gestational diabetes of 9,8% amongmothers of infants weighing more than 4 500 g is muchhigher than the 0,7% reported by Oats et al. 9 in a similar

Macrosomic infants

No. %

Controls

group of mothers. We believe that this finding warrantsroutine screening for diabetes in mothers with risk factorsfor macrosomia. Parks and Zie!' reported that a symphysis­fundus height of 40 cm or more may be an importantfinding. We have yet to determine its significance in ourenvironment.

In the final analysis, ability to estimate fetal weightappears to be of the utmost importance in identification ofthe macrosomic infant. Clinical estimation can serve as auseful guide in experienced hands, but predictions are oftenfalsely positive or negative. lO We agree with others"'" thatreal-time ultrasonography gives the best estimate of fetalweight if available and should be used routinely. Although aslow active phase of labour may forewarn of a large baby,14

our study demonstrates that this per se does not excludejudicious use of oxytocin to augment labour.

Despite the fact that macrosomic infants constitute asmall percentage of all singleton deliveries, macrosomia isassociated with considerable maternal morbidity and highneonatal mortality and morbidity. The high caesarean sectionrate recorded in this study is similar to rates reported byothers.,,3.•,lO Although our incidence of shoulder dystocia(3,7%) is similar to others,'·15 we experienced fairly lowincidences of brachial plexus injury, facial palsy, clavicularfractures, birth asphyxia and meconium aspiration. Webelieve that this may be because we do not performmidforceps deliveries in our institution. The recent use ofultrasonography for antenatal prediction of shoulderdystocia5

.'. should help minimise the above risks. The highperinatal mortality rate recorded in our study is grounds forconcern, in that most of the causes of death were avoidable.

While awareness of antenatal risk factors is certainlyimportant in the prediction and subsequent management ofmacrosomia, fetal and maternal outcome depends largely onhow well labour is managed. This assertion is supported bythe fact that the majority of perinatal deaths in our serieswere due to obstructed labour and ruptured uterus. It ispertinent to emphasise that these mothers were referred toour institution from peripheral hospitals and that in no casewas uterine rupture associated with the presence of a scarfrom a previous caesarean section or injudicious use ofoxytocin in a multigravida. The major underlying cause ofthese perinatal deaths was late diagnosis of obstructedlabour due to macrosomia.

We believe that the unacceptably high perinatal mortalityrate and maternal morbidity can be avoided if midwives andlabour room doctors are properly trained in the concept ofactive management of labour, and hence early diagnosis offailure to progress. Clinical suspicion of a large babycoupled with a slow active phase of labour, especially arrestof cervical dilatation over a 2-hour period in the presence ofadequate uterine contractions, constitutes an early sign offailure to progress which should not be ignored. In thesecond stage of labour, intervention is required if there is nodescent of the presenting part after 30 minutes of bearingdown, or if the patient is undelivered after 45 minutes ofpushing.

We express our gratitude to all consultants in the Departmentof Obstetrics and Gynaecology, Umtata General Hospital, forpermission to include their patients in this study. Our thanksalso go to the professional nurses in the labour ward for theirco-operation.

1,7

51,7

48,3

5,2

0,9

No. %

6

180

168

18

3

64,4

35,6

4,6

1,1

3,7

0,3

0,3

0,3

2,6

224

124

16

4

13

1

1

1

9

Table IV. Outcome

Male infants

Female infants

Birth asphyxia

Meconium aspiration

Shoulder dystocia

Brachial plexus palsy

Clavicular fracture

Facial palsy

Perinatal deaths

Discussion

SAMJ Volume 85 No. 1 lanuary 1995

REFERENCES

1. Boyd ME, Usher R, McLean FH. Fetal macrosomia: prediction, risks and proposedmanagement. Obstet Gyneco/1983; 61: 715-722.

2. Parks DG, Ziel HK. Macrosomia: a proposed indication for primary caesareansection. Obstet Gyneco/1978; 52: 407-409.

3. Okpere EE, Ezimohai M, Agbopuonwu I. Maternal and fetal risk factors associatedwith macrosomic babies in Benin City, Nigeria. Nigerian J Obstet Gyneco/1984;4{2): 51-55.

4. Treharne I. Obesity in pregnancy_ In: Studd JWW, ed. Progress in Obstetrics andGynaecology. Va1. 4. Edinburgh: Churchill Livingstone, 1984: 127-138.

5. Modanlou HO, Komatsu G, Dorchester WG, Freeman RK, Sasu SK. Large-for­gestational age neonates: anthropometric reasons for shoulder dystocia. ObstetGyneco/1982; 80: 417-422.

6. Gross TL, Sokol SJ. Williams J, Thompson T. Shoulder dystocia: a fetal-physicianrisk. Am J Obstet Gyneco/1987; 156: 1408-1418.

7. Marinho AO. Sirthweight and low birthweight among live singleton Nigerian infants.Trop J Obstet Gyneco/ 1980; 55: 420-424.

8. Modanlou HO, Oorchester WL, Thorosian A, Freeman RK, Macrosomia - maternal,fetal and neonatal implications. Obstet Gynecol 1980; 55: 420-424.

9. Oats IN, Abell OA, Seischer NA, Broomhart GR. Maternal glucose tolerance duringpregnancy with excessive size infants. Obstet Gynecol 1980; 55: 184-186.

10. Svigos JM. The macrosomic infant a high risk complication. Med J Aust 1981; 1:245-246.

11. Campbell S, Wilkin O. Ultrasonic measurements of fetal abdominal circumference inestimation of fetal weight. Br J Obstet Gynecol 1975; 82: 689-697.

12. Ogata ES, Sabbagha R, Metzger BE, Phelps RJ, Depp R, Freinkel N. Serialultrasonography to assess evolving fetal macrosomia. JAMA 1980; 243: 2405~2408.

13. Sampson MS, Thomason JL, Kelly SL, Work BA. Prediction of intrauterine fetalweight using real time ultrasound. Am J Obstet Gynecol 1982; 42: 554-556.

14. Khatree MHO, Gamsu HR, Rudd P, Studd JWW. Features predictive of brachialplexus injury during labour. S Afr Med J 1982; 61: 232-233.

15. Golditch IM, Kirkman K. The large fetus: management and outcome. Obstet Gynecol1978; 52: 26-30.

16. Hopwood HG. Shoulder dystocia: fifteen years' experience in a community hospital.Am J Obstet Gyneco/1982; 144: 162-166.

Accepted 18 Apr 1994.

A wolf in wolf's clothingthe abdominalcompartment syndromeR. Burrows, J. Edington, J. V. Robbs

Four patients are described in whom massive abdominal

distension after laparotomy led to increased airway and

central venous pressure and severely reduced urine

output. All cases were associated with massive fluid

resuscitation and operative findings were a grosslyoedematous bowel with free fluid under pressure in the

abdomen. These findings are consistent with the diagnosis

of intra-abdominal compartment syndrome. In 1 case

trauma was remote from the abdomen indicating that

abdominal surgery or trauma may not be a prerequisite for

the development of the condition. Recognition of the

features of the condition is essential as it can only be

treated by decompression of the abdominal contents.

S Afr Med J 1995; 85: 46-48.

Oliguria following raised intra-abdominal pressure has beendescribed previousIY,',2 but the term intra-abdominalcompartment syndrome appears to have been coined byFietsam et al. 3 in 1989 when they described 4 cases ofabdominal distension in patients undergoing repair of aruptured aortic aneurysm. The intra-abdominal compartmentsyndrome usually occurs after emergency aortic surgery anddirect abdominal trauma, but we describe 1 case where thecondition followed elective aortic surgery and 1 case wherethe condition occurred follOWing trauma remote from theabdomen.

Case reports

Case 1A 70-year-old man underwent elective infrarenal aorticaneurysmectomy. The only complicating factor of note inthe pre-operative phase was evidence of an old inferiormyocardial infarct on electrocardiography. During theoperation, which lasted 3112 hours, he required 5 litres ofcrystalloid fluid and 6 units of blood as replacement forblood lost. Urine output for the procedure was 2 litres.Otherwise the operation was uneventful and he wasadmitted to the intensive care unit (ICU). On admission,

Intensive Care Unit, Addington Hospital, Durban

R. Burrows. M.B. B.CH., FFA (SA)

St Augustine's Medical Centre, Berea, Durban

J. Edington, M.B. B.CH., F.F.A. (SA)

Department of Surgery, University of Natal, Durban

J. V. Robbs, CH.M., F.A.C.5. (EDIN.)

Volume 85 No. 1 January 1995 SAMJ

•