Anemia, diet and therapeutic iron among children living with HIV: a prospective cohort study
Baseline characteristics
Between February 2011 and August 2012, 286 children were screened at all participating
sites. Parents’ HIV status was unknown in 7 children, there were 34 children who did
not fulfill age criteria and were either below 2Â years or older than 12Â years, duration
of ART was less than 6Â months in 2 children during the period of recruiting, and caregivers
of 3 children refused to give consent. The final number of children recruited for
the study was 240 and were equally distributed among the three sites. Among the 240
children enrolled in the study, mean age was 7.7Â years (SD 2.6), and there were 131
males (54.6Â %). Distribution of WHO Clinical staging among the children was as follows:
stage 1 and 2: 80.8Â %; stage 3 and 4: 19.2Â %. Median CD4 percentage was 25Â % (IQR
18, 33), median absolute CD4 count was 773 cells/mm
3
(IQR?=?507, 1251) and proportion of children with baseline CD4??350 cells/mm
3
was 27/240 (11.3Â %). There was a high prevalence of malnutrition at baseline; proportion
of children with stunting (height for age Z score HAZ???2) was 40.0Â %; those underweight
(weight for age Z score WAZ???2) was 45.4Â %, and those with wasting (weight for height
Z score WHZ???2) was 23.3Â %, and those children with low BMI (body mass index Z score
BMIZ???2) was 29.2Â %. The proportion of children on ART at baseline was 104/240 (43.3Â %).
ART regimens included zidovudine or stavudine, with lamivudine and nevirapine, efavirenz
or lopinavir/ritonavir. The prevalence of intestinal helminthic infestation was 28/240
(11.7Â %) in this population, and included Ascaris lumbricoides (11/28), Giardia lamblia (7/28), Enterobius vermicularis (7/28), Trichomonas hominis (2/28) and Entamoeba histolytica (1/28).
Anemia prevalence and associations
Anemia was prevalent in 113/240 children (47.1Â %), while severe anemia was seen in
16/240 (6.7Â %). Overall iron deficiency was prevalent in 154/240 (64.4Â %). Vitamin
A deficiency prevalence was 43/240 (17.9Â %), while folate and vitamin B
12
deficiencies were 1/240 (0.4Â %) and 15/240 (6.3Â %) respectively. Risk factors for
anemia in a bivariate model included stunting, CD4??25Â %, detectable viral load ?400
copies/ml and absence of ART and vitamin A deficiency (Table 1). When all the potential risk factors were added to a multivariate model, we found
that significant independent risk factors for anemia were stunted status (OR 1.9,
95%CI 1.1-3.4), low CD4 count (OR 3.2, 95%CI 1.8-5.7), detectable viral load (OR 2.4,
95%CI 1.1-5.4) and vitamin A deficiency (OR 2.5, 95%CI 1.1-5.6) (Table 1). Drugs such as co-trimoxazole did not have an impact on anemia prevalence.
Table 1. Bivariate and multivariate analysis of risk factors of anemia in HIV
Etiology of anemia
Among those children with anemia, iron deficiency was the commonest micronutrient
deficiency; 74/113 (65.5Â %) had iron deficiency anemia, Vitamin A deficiency was seen
in 30/113 (26.6Â %), while 1/112 (0.9Â %) and 9/112 (8.0Â %) had folate and vitamin B
12
deficient respectively. Anemia of inflammation was seen in 66/113 (58.4Â %) of anemic
children. There were several overlapping micronutrient deficiencies as well as evidence
of inflammation that was associated with anemia (Fig. 1). At the initial testing stage, there were no children whose anemia could be attributed
to zidovudine. The presence of ART at baseline did not impact the etiology pattern
of anemia.
Fig. 1. Etiology of anemia in HIV. Proportions of micronutrient deficiency and inflammation
contributing towards the etiology of anemia among HIV-infected children
Dietary intake
Median intake of nutrients expressed as percentage of RDA was 36Â % for iron, and 72Â %
for energy, indicating that on an average, most of the children were obtaining less
than half of the recommended dietary allowance for iron. The proportion of children
getting less than minimum RDA (at least 75Â % RDA) for iron was as high as 79.9Â % and
for energy was 43.3 % (Table 2). These proportions did not change significantly over the 1-year follow-up period.
Table 2. Dietary intake among children with HIV
Follow-up and iron therapy
Follow-up data were available for 194/240 (80.8Â %) children who returned for their
6-month and 12Â month visit. There were 18 children who were transferred out to a different
ART center and could not return for follow-up visits, 8 children died, 1 child whose
caregiver withdrew consent, and 19 were lost-to-follow-up. During the follow-up period,
ART was initiated among 42 children, and 32 received zidovudine-based treatment. Among
these, 3 children developed zidovudine-related bone marrow suppression with severe
anemia and were switched to stavudine or abacavir-based ART.
Among 113 children who were initially anemic, 77 children received therapeutic iron
after the baseline visit. Of the remaining 35 children who did not receive iron, 25
children had haemoglobin 11 g/dl and were not considered “anemic†and the clinical
decision to give iron was not taken. Two children died within 1Â month of baseline
visit, and the remaining 8 children were amongst those who were lost-to-follow-up
as described earlier. Assessment of adherence to iron, assessed by telephone or personal
contact during the clinic visit, indicated that 70Â % of the children reported 100Â %
adherence, 20 % missed 1–2 weeks of therapy, and 10 % missed over 2 weeks of iron
therapy. Mild adverse effects to iron were reported by 17/77 (22Â %), and included
dark colored stools, nausea, diarrhea or constipation and mild abdominal discomfort.
All these were minor side effects that diminished after 2–3 weeks of first reporting.
The number of hospitalizations for intercurrent infections (pneumonia, tuberculosis
and other infections) was similar in both groups (3 among iron supplementation group
and 2 in the non-iron group). No malaria was reported among these children.
Effect of iron therapy
Among children who received iron for 3Â months, median hemoglobin increased from 10.4Â g/dl
to 10.9 mg/dl (Table 3). Hemoglobin change was maximum after 1 year, and increased to 11.3 mg/dl among children
who received iron for up to 6Â months. Children who were on ART plus iron had a greater
hemoglobin increase compared to children who were on ART alone, without iron (Hb change
1.3 versus 0.4 gm/dl respectively, p?=?0.009). The prevalence of iron deficiency also decreased from 68.1 to 49.2Â % (p?=?0.04) among those who received iron. In addition, this group also showed a decreased
trend in clinical severity; severe WHO clinical stage (Stage 3, 4) decreased from
25.7Â % at baseline to 10.9Â % at 1Â year of follow-up. A smaller decrease in clinical
severity stage (16.7–12.5 %) was seen among those who did not receive iron. There
was no significant change in the presence of chronic inflammation among those who
received iron supplements. Iron did not independently affect growth or CD4 parameters;
overall improvement of WAZ and HAZ were seen over one year irrespective of iron supplements.
Table 3. Change in parameters following iron therapy