Pediatric HIV infection accounts for a significant mortality and morbidity in childhood(1,2); but the contribution of neonatal HIV infection to neonatal mortality rate has hardly ever been studied. A prospective study in Durban has demonstrated that childhood AIDS patients admitted to pediatric intensive care unit have a worse prognosis as compared to others(3). Since 1996, a group of HIV-infected neonates with opportunistic infections (tuberculosis, syphilis and cytomegalovirus) are shown to have a rapid progressive HIV infection presenting at a mean age of 15 days(4-6) with over 83% of them dying by the age of nine months(6). There is evidence that HIV exposed uninfected infants may have a worse outcome than unexposed infants. The immunological disturbances include lower CD4 counts in HIV exposed but uninfected babies than those who unexposed(7); there is evidence of cell mediated immunity disturbance which may persist over time(8), and, it has been documented that HIV exposed but uninfected babies acquire Streptococcal pneumoniae infection more commonly than HIV unexposed babies (9). A study was carried out to determine if these HIV exposed neonates have a short and medium term prognosis that is worse than that in HIV unexposed babies also requiring intensive care.

A prospective study was carried out at the Neonatal unit of the King Edward VIII Hospital, Durban, South Africa after obtaining clearance of the institutional ethics committee. Babies admitted to the neonatal unit during the 12-month period beginning July 2000 were enrolled in the study after obtaining informed consent. Mothers of all neonates were subjected to HIV testing by HIV-1 ELISA after providing pre-test counseling. Additional tests such as Roche HIV DNA PCR and T cell subset test by flow cytometry were performed in mothers found to be HIV-infected and in their babies after providing post-test counseling. HIV RNA PCR for viral load assessment was performed using Nuclisens Isolation Kit, Organon Teknika, Boxtel, NL. The women who were found to be uninfected and their babies underwent only immunological tests (T cell subset determination by flow cytometry). Maternal details (age, parity, prenatal care received, syphilis serology, mode of delivery, obstetric complications) and clinical details of the neonate (indications for NICU care, outcome, anthropometric para-meters, frequency of infection, neurological complications, haematological and immuno-logical parameters) were noted. Refusal to undertake HIV testing, mothers receiving nevirapine for prevention of perinatal transfer of HIV infection, inability to complete the consent procedure in view of location of the mother in another hospital, maternal death or babies older than seven days postnatal age constituted exclusion criteria. Considering that 30% of mothers would have HIV infection and that the rate of transmission of HIV infection to the fetus is 25%, it was calculated that a sample of 30 HIV-1 exposed babies and 70 HIV-unexposed babies would be required to demonstrate a 35% vs. 70% mortality rate (95% power, OR 4.3). Epi Info version 6 was used for analyzing the data. Non parametric tests were used for statistical analysis. A value less than 0.05 was considered significant.

Ninety nine neonates admitted to the ICU were deemed suitable for inclusion in the study. However, only 70 (71%) mothers consented to participation after pre-test counseling. Thirty mothers (43%) tested HIV positive. Only 14 (47%) of infected mothers consented to their neonates undergoing virological studies (HIV DNA PCR and HIV RNA PCR for viral load assessment). Only eight of these 14 agreed to undergo immunophenotyping in them. Of the 40 HIV uninfected mothers, 17 (42.5%) agreed to undergo immunophenotyping while only 19 (47.5%) consented to subject their neonates to immunophenotyping. As shown in Table I, there were no significant differences between HIV-1 infected and HIV-1 uninfected mothers in terms of age, parity, antenatal care, obstetric complications, mode of delivery and medical difference. However, a significant difference was noted in terms of absolute CD4 counts (539 vs. 943 cells/mm3, P = 0.0006), percentage CD4 cells (30% vs.42%, P = 0.009) and percentage CD8 cells (49% vs. 31%, P = 0.002). As shown in Table II, there were no significant difference between the HIV-exposed and HIV-unexposed neonates in terms of birth weight, gestational age, indications for intensive care admission, complications of ventilation, number of positive cultures, incidence of nosocomial pneumonia, findings noted on cranial ultra-sound examination, duration of intensive care, absolute lymphocyte counts, per cent CD4 counts and incidence of death. In contrast, to the above, there was a significant difference between the two groups of neonates in terms of white cell count and decrease of CD4:CD8 ratio. Twenty one of the 30 HIV-unexposed infants and 14 of the 19 surviving HIV-exposed infants were followed up with the neonatal clinic for 12-18 months. One baby who had demonstrated a positive HIV DNA PCR test became symptomatic at 5 months of age, while the other 13 HIV-exposed infants remained asymptomatic and were noted to have a negative HIV DNA PCR test by 6 months of age.

Table I

Maternal Obstetric and Hematological  Data : HIV-1 Infected and HIV-1 Uninfected  Mothers.

Maternal Data

Age (yrs) <18

3

18-35

 23

34

0.33

>35

 4

Parity  0

10

11

0.59

1-4

 20

 25

0.01

>5

0

Antenatal care

 22

33

0.59

Obstetric complications

16

22

0.91

Delivery vaginal

14

10

0.1

Cesarean section

3

 25

0.17

Laboratory Testing

n = 8

n = 17

P value

Hemoglobin

10.3 (±1.2)

11.3 (±1.5)

0.118

White cell count (× 109/L)

8.1 (±3.2)

9.1(±3.4)

0.49

Absolute lymph (× 109/L)

1933 (±796)

2285 (±678)

0.558

Platelet (× 109/L)

395 (±86)

327(±97)

0.074

Abs CD4 (µL)

539 (±239)

 943 (±272)

0.0006

CD4%

30.1 (±10.7)

41.7(±7)

0.009

Abs CD8

935.8 (±555)

721.8 (±378)

0.299

CD8%

49 (±12.9)

30.88 (±8.7)

0.002

CD4:CD8 ratio

0.72 (±0.5)

1.48 (±0.59)

0.004

>2.5 kg

3

< 1.5kg

19

18

>1.5kg

8

17

Small gestational age

14

20

  HMD*

17

22

  Pneumonia

8

7

  Mec aspir†

2

2

  Other

3

9

ICU‡ complications

20

25

Clinical problems

9

13

Positive blood cult

7

5

Postive ETT§ cult

14

15

Antibiotics 2nd line

10

11

Nosocomial pneumonia

10

14

  <7 days

15

27

  8-14 days

10

7

  >15 days

5

6

Deaths

11

10

Laboratory Testing n=14

  n=19

p value

Haemoglobin 12.0(±2.4)

12.5(±2.8)

0.572

White cell count (x109/L) 15.04(±13.1)

10.5(±4.3)

0.002

Absolute lymph (x109/L) 4461(±3778)

3069(±1518)

0.383

Platelet (x109/L) 238 (±143)

212(±165)

0.633

Abs CD4 (µL)  1573 (±1213)

1718 (±953)

0.173

CD4%

48(±14.3)

54 (±13.4)

Abs CD8  1045 (±1187)

1047 (±1394)

0.996

CD8%  22.9 (±8.08)

17.2 (±6.3)

0.04

CD4:CD8 ratio

2.36 (±1.10)

3.73 (±2.4)

Discussion

This study demonstrates that the outcome in HIV-exposed babies is not significantly different from that in HIV-unexposed babies, although some of the hematological parameters may show a significant difference in these two groups of babies. It is also worth recording that most of the HIV-unexposed babies ultimately turn out to be uninfected when tested by virological tests at 6 months of age. A thought is often put forward that with the shortage of resources and paucity of ICU beds, ICU care in resource poor situations should be offered to only those babies in whom it would be worthwhile(10). There is a risk that HIV-exposed babies would be categorized as "not worthwhile". Our study has demonstrated that this is not true and hence withholding ICU care is unjustified not only on ethical grounds but also on hard scientific facts brought out in the study.

Factors that influenced early termination of the study included the reluctance of mothers to consent to HIV testing in order to participate in the study. Fear of a positive result is well known, however, little is written around the topic. Two recent reports one from the Zimbabwe and the other from the Latino countries stress the fear, and stigma associated with the epidemic(11,12). The introduction of the antiretroviral, nevaripine, as standard of care to reduce perinatal transmission of the HI virus in this setting clearly influenced the recruitment of cases into to the study(13)

We wish to thank the Chief Medical Manager of King Edward VIIIth Hospital for permission to undertake the study. Our thanks go to the Ms J. Ramjee for data collection and punching, paediatric registrars and Ms Fikile Sibonyoni, the HIV counsellor who assisted with the study. MA was in receipt of a grant from the University of Natal Research Fund for this study.

Contributors: MA, PJ, TP conceptualized the protocol; collected clinical and laboratory data; AM helped in data collection; PK and SC were involved in virological working.

Funding: University of Natal Research Funds, South Africa.

Competing interests: None.

1. Pillay K, Colvin M, Williams R, Coovadia HM. Impact of HIV-1 infection in South Africa. Arch Dis Child 2001; 85: 50-51.

2. Zwi K, Pettifor J, Soderlund N, Meyers T. HIV infection and in-hospital mortality at an academic hospital in South Africa. Arch Dis Child 2000; 83: 227-230.

3. Jeena PM, Coovadia HM, Bhagwanjee S. Prospective, controlled study of the outcome of human immunodeficiency virus -1 antibody positive children admitted to an intensive care unit. Crit Care Med 1996; 24: 963-967.

4. Adhikari M, Pillay T, Pillay D. Perinatal tuberculosis : an emerging problem. Pediatr Infec Dis J 1997; 16: 1108-1112.

5. Pillay T, Adhikari M. Congenital tuberculosis in a neonatal intensive care unit. Clin Infect Dis 1999; 29: 467-468.

6. Pillay T, Adhikari M, Mokhili J. Severe rapidly progressive HIV-1 infection in newborns. Pediatr Infec Dis 2001; 20: 404-410.

7. Gesner M, Papaevangelu V, Kim M, Chen S, Moore T, Krasinski K, Borkowsky W. Alteration in the proportion of CD4 T lymphocytes in a subgroup of human immunodeficiency virus-exposed-uninfected children. Pediatrics 1994; 93: 624-630.

8. Clerici M, Saresella M, Colombo F, Fossati S, Sala N, Bricalli D, et al. T-lymphocyte maturation abnormalities in uninfected newborns and children with vertical exposure to HIV. Blood 2000;96:3866-3671.

9. Farley JJ, King JC, Nair P, Hines SE, Tressler RL, Vink PE. Invasive pneumococcal disease among infected and uninfected children of mothers with human immunodeficiency virus infection type 1. J Pediatr 1994,124: 853-858.

10. Mathiva LR, Luyt D, Hon H, Pence M, Litmanovitch M, et al. Outcome of mechanical ventilation in children infected with the human immunodeficiency virus. S Afr Med J 1998,88;1447-1451.

11. Daily HIV/AIDS Report. Global Challenges/ Fear of HIV, Associated stigma hamper Zimbabwe’s Efforts to Lower Infant Mortality Rate. www.kaisernetwork.org/daily_reports/rep_index.cfm

12. HIV testing in the Latino Community www.PositiveWords.com(c)2003 Dallabridge associates.

13. Moodley D. The Saint Trial : Nevirapine (NVP) versus zidovudine (ZDV) + lamivudine (3TC) in prevention of peripartum HIV transmission, Abstract LbOr2.. Thirteenth International AIDS Conference Durban South Africa 2000; Jul 9-14, Durbam, South Africa.