Virologic, immunologic and clinical response of infants to antiretroviral therapy in Kampala, Uganda

For this observational cohort study, we recruited and followed HIV-infected infants who were enrolling for care and treatment at the Baylor College of Medicine Bristol Myers Squibb Children’s clinical centre of excellence at Mulago Hospital (Baylor-Uganda) in Kampala, Uganda. Baylor-Uganda is a Non Governmental Organization (NGO) involved in the prevention, care and treatment of HIV-infected children and their families in Uganda. The Foundation is donor funded and runs a clinic that offers free clinical care to all its patients. The clinic is an out-patient treatment center that operates five days a week with an average daily attendance of 170 patients. Over 90% of patients seen at the clinic are children from the age of 6 weeks to 18 years. Parents or adult caretakers constitute the remainder of the patients.

At the clinic, patient care involves HIV diagnosis and treatment, nutritional rehabilitation, diagnosis and treatment of opportunistic infections, counseling and social support. Infants enrolled for care at the clinic are referred from the nearby Mulago National Referral hospital wards and from other centers involved in Prevention of Mother to Child HIV Transmission (PMTCT). Some infants are referred from HIV treatment centers within the country, and others are self referrals.

Study participants were identified and recruited into the cohort from the Baylor-Uganda clinic patient population. Study participants were selected based on a criteria that required them to be clinic patients, confirmed to be HIV positive (by 2 positive blood tests showing the presence of viral nucleic acids in the patient’s blood samples); between 6-weeks and 12 months of age; ART naïve, but willing to start ART as soon as deemed necessary by the clinician; reside within 20 kilometer radius from the clinic; willing to be visited at home; and agree to be regularly followed-up at the clinic. Children without parents or guardians were excluded. Patients less than 6 weeks of age were not included since they were not routinely registered for care at the clinic. Based on WHO guidelines for infant diagnosis [9], and recommendations from the Ugandan Ministry of Health, routine diagnosis of HIV at Baylor-Uganda is carried out from the age of 6 weeks when infants are brought back to hospital for immunization.

From April 2009 to March 2012, all clinic patients that fulfilled the selection criteria were consecutively recruited into the cohort and were subsequently followed up.

Before starting ART, we confirmed HIV infection in the infants by carrying out an HIV DNA Polymerase Chain Reaction (PCR) test followed 2 weeks later by a viral RNA PCR test to quantify the HIV virus in the blood. To be considered HIV infected, both tests had to confirm presence of the virus in blood.

All laboratory- confirmed HIV infected infants were started on ART. Infants enrolled before July 2010 were started on ART based on the 2006 WHO guidelines for ART initiation. Based on these guidelines, all infants with CD4 cell counts below 25% or in advanced WHO clinical stages (stages 3 and 4) were considered eligible for ART and were started on treatment. Following the 2010 revision of WHO treatment guidelines, all HIV-infected infants were started on ART irrespective of their clinical stage or CD4 cell counts. A combination of 2 Nucleoside Reverse transcriptase inhibitor (NRTI) drugs and 1 Non-Nucleoside Reverse Transcriptase (NNRTI) or Protease Inhibitor was the ART combination of choice. Zidovudine (ZDV) or stavudine (D4T) combined with lamivudine (3TC) together with either nevirapine (NVP) or Lopinavir/ritonavir were the main ART drug combinations as recommended by WHO. Stavudine was preferred in patients with anemia (hemoglobin <7.5 g/dl) and Lopinavir/ritonavir was preferred in patients exposed to single dose NVP for PMTCT. Patients on rifampicin for treatment of tuberculosis were put on 3 NRTIs (ZDV or D4T + 3TC + Abacavir).

Baseline measurements and tests were carried out on the date when ART was started and subsequent follow-up visits were scheduled 2, 4, and 8 weeks after starting treatment. Thereafter, patients returned to the clinic monthly, and if an appointment was missed, a home visit was carried out by trained home health workers. In addition, parents were encouraged to bring the children for medical care whenever they fell sick. At baseline and every 6 months thereafter, blood samples were drawn for complete blood count (CBC), CD4 cell count; liver function test [serum alanine amino transferase (ALT)], and HIV viral load.

At every visit, physicians reviewed the infant, documented and managed any adverse events or opportunistic infections. In addition, nutritional assessment including measurement of weight and height was conducted. Anthropometric computations and comparisons were conducted using the WHO anthro software [10] which is based on WHO child growth standards. Underweight was defined as a weight for age z-score (WAZ-score) less than -2 standard deviations (SD) from the WHO reference median values. Stunting was defined as height for age z-score (HAZ-score) less than -2SD from the reference values and weight for height z-score (WHZ-score) less than -2SD from the WHO reference median defined wasting. Adherence to medication was encouraged by both the physician and a trained counselor.

Antiretroviral therapy was considered the exposure variable and the primary outcome variable was viral suppression 6 months after ART initiation. Other outcomes of interest included change in CD4 cell count; occurrence of ART-related adverse events; death during follow-up; and change in nutritional status as measured by weight-for-age, height-for-age and weight for height z-scores. Virologic suppression was defined as the attainment of HIV viral loads less than 400 copies/ml after 6 months of ART.

Patients’ age, sex, WHO clinical stage, ART combination, presence of opportunistic infections and concurrent use of other medication, were considered potential confounders and were therefore documented and considered in the analysis.

All patient information was recorded in structured forms and was subsequently entered into the clinics electronic medical database which is designed to check for consistency of the data.

Written approval to conduct the study was granted by Makerere University College of Health Sciences’ Research and Ethics committee, Uganda National Council of Science and Technology, and Baylor College of Medicine Institutional Review Board (USA). The parents or caretakers gave written informed consent to participate in the study.

Of the 78 infants that had ART for at least 6 months, successful viral suppression was seen in 56(71.8%) infants. Twenty two (28.2%) infants continued to have detectable viremia at month-6 of treatment. In univariate and multivariable regression analysis; none of the baseline factors such as age, gender, CD4 cell percent, WHO clinical stage, choice of antiretroviral agent, weight for age or height for age z-scores could predict treatment success at month-6 of therapy (Table 2).

At month-12 of ART, 8 of the 22 patients that had detectable viremia at month-6, attained viral suppression; 1 patient got lost to follow-up, 1 was switched to second line ART, and 12 patients continued to have detectable viremia. On subsequent clinic visits, 6 of the 12 patients were switched to second-line ART, 4 were lost to follow-up and 2 died (Table 3). Among patients that attained viral suppression at month-6, no viral rebound was seen at month-12 of ART.

Of 47 patients that completed 24 months of ART; 44(93.6%) were still on first line treatment. Of these, 37(84.1%) continued to have viral suppression; and 7 patients had detectable viremia, all of whom had earlier achieved viral suppression.

Altogether 7 patients were switched to second-line treatment (Abacavir + 3TC + Lopinavir/ritonavir) as a result of persistent viremia;5 out of the 7 attained viral suppression 6 months after starting second-line ART, and 2 patients were yet to complete 6 months of second-line ART.

The cumulative probabilities of attaining viral suppression at month-6 and month-12 of ART were 71.8% (95% CI: 61.6-81.3) and 83.1% (95% CI: 73.8-90.5) respectively.

The choice of third agent (Nevirapine versus Lopinavir/r) was not associated with risk of viremia at month-6, even after adjusting for the choice of NRTI backbone and baseline WHO clinical stage, Risk Ratio(RR)=1.1 (95% CI: 0.8-1.5), P=0.152. There was no association between WHO guideline used for patient treatment (2006 versus 2010) and risk of virologic failure at month-6 of ART; RR=1.24 (95% CI: 0.9-1.6).

There was a significant increase in CD4 cells from a baseline mean of 23% (SD=10.1) to 30% (SD=9.6) at month-6, P<0.001; and to 33% (SD=8.9) at month-12 of ART; (P=0.009, for the change between month-6 and month-12). By the end of the second year of ART, the mean CD4 cell percentage rose to 36% (SD=8.05) (Figure 1). The proportion of infants with very low CD4 cell counts (<15%) dropped from 23.8% at baseline to 8.5% at 6 months of treatment, and by the end of the second year, only one patient continued to have CD4 cell percentage below 15%. This patient was an infant, with no prior exposure to nevirapine for PMTCT, who started ART (D4T+3TC+NVP) at the age of 4 months with a baseline CD4 of 16%. His CD4 percentage dropped to 15% at month-6 and to 14% at month-24 of ART. He never attained virologic suppression and was considered a case of treatment failure due to possible primary HIV resistance. He was subsequently started on second-line ART.

Ten (11.9%) of the 84 patients that started ART were diagnosed with tuberculosis at baseline or shortly after ART initiation. Anti-tuberculosis treatment was started but one patient died after 5 months of ART and tuberculosis medication. One 7-month old infant was diagnosed with disseminated Bacillus Calmette-Guerin (BCG) infection 3 months after starting ART. He was treated as a case of immune reconstitution inflammatory syndrome (IRIS) and improved.

At ART initiation, 54 (64.3%) of 84 patients had a WAZ-score below -2SD of the WHO reference median. This prevalence of underweight varied according to age and clinical stage of the disease. Thirty six (80%) of 45 older infants (7-12 months of age) compared to 18(46.2%) of 39 younger infants (1-6 months of age) were underweight at baseline; prevalence ratio= 1.7 (95% CI: 1.19-2.51). Of 47 infants with baseline advanced HIV disease (WHO clinical stage 3&4), 36 (76.6%) were underweight; and among 37 infants with milder clinical disease (WHO clinical stage 1&2), 18 (48.6%) were found to be underweight at baseline; prevalence ratio= 1.6 (95% CI: 1.09-2.27). There was no association between baseline underweight and gender or baseline CD4 cell counts.

Among the 78 infants that received ART for at least 6 months, the proportion that were underweight reduced from 65.4% at baseline to 30.8% at 6 month of ART, p<0.001

Eighty one patients were included in the analysis of height for age z-scores (HAZ-score). Of these, 32 (39.5%) patients had baseline HAZ-score less than -2. Twenty (46.5%) out of 43 infants aged 6-12 months compared to 12 (31.5%) out of 38 infants younger than 6 months of age were stunted; prevalence ratio=1.5 (95% CI: 0.84-2.60). Twenty three (51.1%) of 45 patients in WHO stage 3or 4 compared to 9 (25%) of 36 patients in WHO stage 1 or 2 were stunted at baseline; prevalence ratio=2 (95% CI: 1.08-3.85).

There was no significant change in stunting in the first 6 month of ART. Of 75 patients that had both baseline and follow-up HAZ-score measurements, the prevalence of stunting increased from 40% at baseline to 48% after 6 months of ART, p=0.134.

Of 65 patients that were included in the analysis of weight for height z-scores (WHZ-score), 22 (33.8%) had baseline WHZ-scores less than -2. However, after 6 months of ART, only 9 (13.8%) patients continued to have WHZ-scores less than -2; p=0.007.

A total of 11 (12.1%) patients died during the follow-up period. Four of the 11 died before ART could be initiated. All 4 patients were in WHO clinical stage 4; between 6-12 months of age at the time of death; and had been in care for 6-11.7 weeks. The illness preceding death was tuberculous pericarditis for one female infant, dysentery and lymphoma for other 2 female infants and Pneumocystis Jirovecii Pneumonia for the only male infant. Of the 7 infants that died while on ART, 4 died within the first 6 months, 1 died in the 8^th month and 2 died in the second year of ART. The overall mortality rate among children on ART was 4.09 (95% CI: 1.95-8.58) per 100 child-years. Post mortem examinations were not carried out; however, the majority of patients that died had underlying malnutrition (Table 4).

Within the first 6 months of ART, adverse events were reported in 7 (8.3%) of the 84 patients that started treatment. Of the seven, 3 patients developed diarrhea, 1 had nausea and vomiting, 2 had anemia and 1 developed a skin rash. One of the 3 infants with diarrhea and the one infant with nausea and vomiting were on a PI based regimen. All patients with anemia were on ZDV and the one patient with skin rash was on nevirapine. All patients that developed anemia or skin rash had ART regimen change. Patients with diarrhea, nausea and vomiting had mild and transient illness not requiring ART regimen change.