Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

44 Abstracts 12163-12167

12th World AIDS Conference

12163 Morbidity and mortality among HIV-1 infected and
uninfected Kenyan children
Joanne Embreel, S. Njenga2, P. Datta2, P. Nyange2, J.O. Ndinya-Achola2,
F.A. Plummer1. 'Dept. of Medical Microbiology, University of Manitoba, Winnipeg
MB, Canada; 2University of Nairobi, Nairobi, Kenya
Background and Objective: To determine the excess morbidity and mortality
experienced by HIV-1 perinatally and postnatally infected African children not
receiving antiretroviral therapy.
Design: Prospective cohort study.
Methods: The mortality and morbidity in HIV-1 perinatally, postnatally, and
uninfected cohorts of children enrolled at birth to HIV-1 seropositive mothers
and a concurrently enrolled cohort of infants born to HIV-1 seronegative mothers
followed in The Nairobi Mother to Child HIV-1 Transmission and Pediatric AIDS
Study were compared.
Results: There were 102 perinatally, 41 postnatally, 465 uninfected and 570
control children followed for 226, 144, 1217 and 1401 person years respectively.
Cumulative mortality rates over 10 years were 40.2%, 31.2%, 4.2% and 4.6%
respectively (p < 0.001). The following were significantly (p < 0.001) increased
among perinatally infected children compared with controls: hospitalizations (3x)
and the incidence of sepsis/meningitis (2.5x), tuberculosis (10x), measles (3x),
pneumonia (2x), acute otitis media (4.5x), chronic otitis media (6.5x), thrush
(4.5x), other mouth ulcers (3.5x), skin rashes (2x), and failure to thrive (1.5x).
Fewer illnesses were more frequent among postnatally infected children compared with controls and included: hospitalizations (3x), sepsis/meningitis (2.5x),
tuberculosis (5x), acute otits media (3.5x), chronic otitis media (3.5x), thrush
(2.5x), skin rashes (2.5x), and failure to thrive (2x). Only tuberculosis was
increased among uninfected children compared with controls (p < 0.05).
Conclusion: Both perinatal and postnatal HIV-1 infection significantly affected
childhood mortality and morbidity. The pattern of illnesses was slightly different
for perinatal versus postnatal infection. There were no "AIDS" defining illnesses
which would be useful for clinical screening in lieu of serological testing or which
would be predictive of early infant mortality due to HIV-1 infection.
12164 HIV viral load in children with lymphocytic interstitial
pneumonitis
F. Bell', M. Richardson2, H. Lyall3, D. Gibb4, V. Novelli5, S. Walters3,
G. Tudor-Williams3, M. Sharland2. 'Dept of Pediatrics Sheffield Childresn
hospital S102th; 2St. Georges Hospital London; 3St. Mary's Hospital London;
4Institute of Child Health London; 5Great Osmond Street Hospital London, UK
Background: Up to one third of children with vertically acquired HIV develop
lymphocytic interstitial pneumonitis (LIP). The natural history of LIP is poorly
understood but affected children tend to have a better prognosis than children
without LIP. However, the lymphoproliferation seen in LIP suggests a high viral
load, a factor that is usually associated with progressive disease.
Methods: Retrospective case note review at three centres in London. Viral
loads, CD4 counts and CD8 counts in children with LIP were compared with
those from an equal number of age-matched HIV-infected controls.
Results: The notes of 127 patients were reviewed. 34 children were identified
with recent radiological evidence of LIP. Mean age in LIP patients and controls
was 4.7 years. 19 patients and 18 controls were receiving antiretroviral therapy.
Clinical status (CDC classification) was similar for the two groups. Median viral
load was 127 700 copies/ml in children with LIP compared with 35 000 in controls
(p < 0.05, Wilcoxon paired test). Median CD4 counts were 610 x 106/1 (LIP) vs.
750 (controls) (p < 0.05), CD8 counts: 1620 (LIP) vs. 1260 (controls) (p = 0.06).
Conclusions: LIP is usually considered a good prognostic factor in paediatric
HIV infection. In this study children with LIP had high HIV viral loads and low
CD4 counts, findings normally associated with a poor outcome. Further work is
required to explain this apparent paradox.
S12165 P Fever of unknown origin in HIV infected children
Jesus Ruiz-Contreras, M.I. De Jose, L. Ciria, M.J. Mellado, J.T. Ramos,
J. Clemente, V. Rodriguez-Cerrato. Ctra de Andalucia KM5, 4 Madrid 28041,
Hospital 12 de Octubre Dpto Pediatria, Spain
Background: There are few reports of fever of unknown origin (FUO) in HIVinfected children. The aims of this study are: 1) to describe the number and
characteristics of episodes of FUO in HIV-infected children less than 13 years
of age; 2) to assess diagnostic methods that were carried out; 3) to know the
survival time since the episode of FUO was diagnosed.
Methods: Retrospective review of the charts of HIV-infected children in four
tertiary university hospitals in Spain. FUO was defined as fever for 7 or more
days in children whose initial history, physical examination and laboratory data
(including bacterial cultures) failed to reveal the cause.
Results: Overall 316 charts of HIV-infected children were recovered and 67
episodes of FUO were identified in 53 (17%) children. Most of these episodes
(83%) occurred in patients with HIV advanced disease (C3, 43 cases; B3, 8 case.s)

and 79% had a previous history of opportunistic infections (mostly oral and/or
esophageal candidiasis). In 75% and 73% of episodes of FUO, the patients were
receiving cotrimoxazol prophylaxis and antirretroviral treatment respectively. The
diagnoses were: FUO, 24 cases (36%); disseminated MAC infection, 11 cases
(16%); tuberculosis, 11 patients (extrapulmonary disease in 3 of them); B lymphoma, 4 cases (6%); disseminated CMV infection, 3 cases (4%). Other less
frequent diagnoses were: drug fever, endocarditis, HIV primary infection, pancre
atic abscess, Pseudomonas aeruginosa bacteremia and visceral leishmaniasis.
Duration of fever before the admission was 10.5 ~ 12 days (median 7) and total
duration was 25 ~ 17.6 days (median 21). Invasive procedures were carried out
in 23 cases (34%). Forty (75%) patients have died since the episode of FUO
(survival 12 ~ 17 months). In 17 of them the death was directly related to the
episode of FUO.
Conclusions: Fever of unknown origin is a common entity in HIV infected
children especially in those with advanced disease and consequently it presents
a poor prognosis, as 75% of the children die shortly after diagnosis. In about half
of the cases no cause is found; disseminated MAC infection, tuberculosis and
tumours are the most frequently identified causes of FUO in Spain.
172*/12166 1 The impact of HIV infection on in-hospital
paediatric mortality at Chris Hani Baragwanath
Hospital (CHBH), South Africa
Karen Zwi', Tammy Meyers2, J.M. Pettifor2, R.N. Soderlund3, J.S. Galpin4. 1PO
Bertsham 2013 Chris Hani Baragwanath Hospital & Univ Witwaters Rand;
2Department Paediatrics C.H. Baragwanath Hosp; 3Centre for Health Policy;
4Dep't Statistics & Actuarial Science CHBH, Johannesburg, South Africa
The prevalence of HIV admissions to CHBH has increased from 2.6% in 1992
to nearly 30% in 1996. This paper documents the impact that this has had on
in-hospital mortality.
Methods: Hospital records from January 1992 to the end of April 1997 were
used to obtain retrospective data on general paediatric admissions. HIV tests
were performed where clinically indicated although from June to December 1996
prospective data was obtained on a random sample of admissions where most
children were screened. Pre- and post- test counselling was conducted and
parental consent obtained.
Results: Records were available for 22 633 admissions retrospectively and
549 in the group studied prospectively. In-hospital mortality for all children increased by 42% from 4.3% in 1992 to 6.1% in 1997. Mortality was 13.2% in the
HIV infected children compared to 5.1% in the uninfected and 3.1% in untested
children (p < 0.001). The proportion of HIV related deaths increased from 6.7%
in 1992 to 54.3% in 1997 (p < 0.001). Mortality rates declined in the uninfected
children from 5.3% to 4.6% in 1997 (x trend 3.3; p = 0.06). This did not occur
in the HIV infected children. Young infants were more likely to die. Prospective
data showed that 76.5% of HIV infected children were <6 months of age at
death compared to 53.3% of uninfected children. Case fatality rates were higher
for both pneumonia and gastroenteritis in the HIV infected children, 19.9% and
13.7% respectively, compared to 3.7% and 2.9% in the negative and untested
children (p < 0.001). Acute respiratory infections were the most common cause
of death in the HIV infected children (64.7%). The rest of the HIV deaths were
attributable to infectious diseasesan malnutrition. Deaths in the HIV uninfected
children were predominantly associated with infections but included a wide spread
of different pathologies such as congenital abnormalities and poisonings. Aside
from H. Influenza type B, not part of routine vaccination schedule in South Africa.,
there was a decline in vaccine-preventable deaths in all children.
Conclusion: The mortality rate of children has increased at CHBH. The majority
of deaths are now HIV related. Children who are not HIV infected have declining
mortality. HIV infection is threatening the advances that have been made on child
survival in South Africa over the last few decades.
S12167 Relationship between dental eruption delay and the
immunological deficit in HIV carrier infants
Maria Regina Baptista'.2, T.M.I. Nishimoto2, A.S.P. Gibbons2, C.S. Nucci3
'Rbassin Nagib Trabulsi No 145 Apto 062 Ponta Da Praia Santos SP CEP
11030-540; 2NIC of AIDS Municipal Program Santos SP; 3Coordinator of
Municipal Oral Health Santos SP, Brasil
Objective: To determine the relationship between dental eruption delay and
immunological deficit among infants carrying HIV.
Patients and Methods: Cohort of 100 infants vertically infected, class B3
and C3, for the CDCP Classification for Pediatric Infection, who have been sequentially followed since birth up to the present. 38 out of 100 infants present
dental eruption delay, 18 presenting deciduous dentition delay and 20 presenting
effective dentition delay. The infants have been duly submitted to the CD4 cell
count and clinically examined by skilled dentist surgeon and a pediatrician. The
evaluation criterion adopted meets the international chart comparison.
18 months 36 months 6 years 8 years 10 years 12 years
16 deciduous All of the 1st effective All effective Former All effective
teeth deciduous molar molar pre molars 28 teeth
Results:
Age 18 months 36 months 6 years 8 years 10 years 12 years
Infants amount 7 11 10 10 11 3

Dental eruption less than less than 20 Absence of Absence of Absence of Less than
delay 16 teeth deciduous teeth 1st molar the incisors 1st pre molar 28 teeth
CD4 count 50-490 02-409 02-183 02-192 02-153 30-190
Conclusion: The dental eruption delay is related to: severe immunological
supression; normal or moderate immunological supression infants were not verified; mastication, phonetic articulation and swallowing alterations are noticed;