Effect of baseline immunological condition, virological response and

duration of HAART on growth in HIV-infected adolescents.

Lazarus E1; Otwombe K1; Mohapi L1; Cescon A2,3; Violari A1; Laher F1

Institutions: 1Perinatal HIV Research Unit, Soweto, South Africa, 2Simon Fraser University, Faculty of Health Sciences, Burnaby, Canada, 3British Columbia Centre for Excellence in

HIV/AIDS, Vancouver, Canada, 1

Effect of HIV on child & adolescent growth from the literature

• HIV is associated with poor growth in children1-3 • Stunting (short-for-age) is the most common

abnormal growth pattern4

• Height-for-age is more closely linked to disease progression than weight-for age5

• Survival related to height growth velocity6

1Gertner JM , J Pediatr 1994; 2Moye J Jr, J Pediatr 1996; 3Carey VJ, AIDS 1998; 4Arpadi, JAIDS 2000; 5Benjamin DK, AIDS 2003; 6Chantry CJ, Pediatr Infect Dis J, 2003

Background Methods Results Conclusions

2

Effect of HAART on child & adolescent growthfrom the literature

• HAART improves weight and height1-8

• Improved growth may be associated with –Virological response1,5 – Immunological response3,6

– Immunological stage at initiation of ART8

–Age at start of HAART3,7,8

1Guillén S, Pediatr Infect Dis J 2007; 2Kabue MM, AIDS Patient CARE STDS 2008; 3 Nachman SA, Pediatr Infect Dis J 2005; 4Song R, Pediatrics 2007; 5Verweel G, Pediatrics 2002; 6Ghaffari G, Pediatrics 2004; 7Bakeera-Kitaka S, J Pediatr Infect Dis, 2008; 8Kekitiinwa A, JAIDS,2008

Background Methods Results Conclusions

3

Aim

• To determine the correlation with adolescent growth of: – baseline CD4 – age at initiation of HAART – virological response– treatment duration

Background Methods Results Conclusions

4

Setting

• Retrospective, cross-sectional analysis • Perinatal HIV Research Unit, Soweto, S Africa• HAART-recipients aged 11-19 years

Background Methods Results Conclusions

5

How is growth measured?

• Wt & ht plotted on centile chart• Z-score calculated using SAS

software provided by the CDC• Z-score: measure of standard

deviations from mean • 50th centile z-score = 0• Example

– 13 year old female– Wt = 35kg WAZ = -

2.48– Ht = 140 cm HAZ = -1.57

Background Methods Results Conclusions

6

What is growth?

• Difference between gender adjusted growth measurements between most recent clinic visit and baseline:– Weight for age z-score (WAZ)– Height for age z –score (HAZ)

Background Methods Results Conclusions

7

Statistical analysis

• Chi-square contingency table analysis: association between growth and – duration of treatment– baseline CD4– age at baseline – most recent viral load.

• Where an association existed, a univariate analysis was performed to determine the predictor of growth.

• 5% significance level

Background Methods Results Conclusions

8

Background characteristicsBackground Methods Results Conclusions

N=107

Sex: Number (%) Female 57 (53)Age at HAART initiation: median years (IQR) 8.4 (6.1 – 11.4)Duration HAART use: median years (IQR) 6.1 (2.4 – 9.1)

Baseline CD4 count: median cells/ml (IQR)<200: n (%)200 – 350: n (%)>350: n (%)

246482435

(107 – 475)(45)(22)(33)

Most recent viral load: n (%)<400 cp/ml≥400 cp/ml

8423

(79)(21)

9

GrowthBackground Methods Results Conclusions

-0.15-0.1

-0.050

0.050.1

0.150.2

0.25

HAZ WAZ

Baseline CD4

CD4 <200CD4 200-350CD4 >350

10

GrowthBackground Methods Results Conclusions

-0.4

-0.2

0

0.2

0.4

0.6

HAZ WAZ

Duration of HAART

<6.1 yrs>6.1 yrs

11

GrowthBackground Methods Results Conclusions

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

HAZ WAZ

Most recent viral load

VL >400VL <400

12

ResultsBackground Methods Results Conclusions

Height-for-age z-score (HAZ) Weight-for-age z-score (WAZ)

Variable OR (CI) p-value OR (CI) p-value

Odds Ratio (OR) for gain in HAZ and WAZ

Baseline CD4 ≥200 <200

2.7 (1.18, 6.20)1

0.019 1.2 (0.56, 2.67)1

0.61

Baseline CD4 ≥350<350

2.9 (1.25, 6.66)1

0.013 1.7 (0.74, 3.79)1

0.22

Most recent VL<400 ≥ 400

1.5 (0.55, 3.98)1

0.44 1.1 (0.41, 2.72)1

0.91

Duration Rx>6.1 yrs<6.1 yrs

1.9 (0.84, 4.1)1

0.13 0.7 (0.30, 1.43)1

0.29

Age at BL 0.9 (0.83, 1.05) 0.28 0.9 (0.81, 1.01) 0.1

13

Conclusion

• Higher baseline CD4 count predicts better growth in height

• To optimise growth outcomes into adolescence, we recommend HAART initiation at baseline CD4 count >200 cells/ml for children >5 years old

Background Methods Results Conclusions

14

START early

GROWbetter

15

Limitations

• Retrospective• Non-randomised

Background Methods Results Conclusions

16

Acknowledgments

• Clinic staff of adult & paediatric clinics– Doctors Haseena Cassim, Xolile Dlamini, Mfundo

Kuhlase, Afaaf Liberty, Mpolokeng Melamu, Geraldinah Mjali, Ashaan Naidoo, Faeezah Patel

– Sisters Noreen Boikanyo, Jackie Brown, Sizakele Ncongwane, Rebecca Phofa

– Nurses Gloria Tshabalala, Mpho Kokoele, Cathy Lephoto, Mitha Madiba, Patricia Masiya, Busi Saba

• USAID for providing funding to care for these adolescents.

• Our adolescent patients and their caregivers

Background Methods Results Conclusions

18