If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author at: Department of Community Medicine, Faculty of Health Sciences, UiT—The Arctic University of Norway, Hansine Hansens veg 18, 9019 Tromsø, Norway.
Department of Community Medicine, Faculty of Health Sciences, UiT—The Arctic University of Norway, Tromsø, NorwayDepartment of Public Health and Nursing, Faculty of Medicine and Health Sciences, NTNU The Norwegian University of Science and Technology, Trondheim, Norway
Limited data are available regarding the incidence of transient viraemia in children in resource-limited settings.
•
All patterns of detectable viraemia after initial viral suppression are more common in children.
•
Children are more prone to remain non-suppressed after 12 months of antiretroviral therapy than adults.
•
There is a need for closer viral load monitoring of children growing up with HIV.
Abstract
Objective
To investigate the incidence and predictors of viraemia among individuals on antiretroviral therapy (ART) in Harare, Zimbabwe.
Methods
Children (0–19 years) and adults (>19 years) starting ART between 2013 and 2015 were followed for a median of 2.8 and 2.7 years, respectively. The incidence rates of virological failure (VF), low-level viraemia (LLV), and viral blips were assessed and the predictors of viraemia were determined using logistic and parametric survival regression analyses.
Results
A total of 630 individuals initiated ART, and 19.7% of children and 5.6% of adults did not achieve viral suppression by 12 months. Younger age and CD4 count ≤200 cells/mm3 at baseline were associated with not being virally suppressed at 12 months in adults. Among those who achieved viral suppression during the follow-up period, the incidence of VF was higher in children (4.0/100 person-years vs. 0.4/100 person-years in adults; p< 0.001), as was the incidence of LLV (1.9/100 person-years vs. 0.3/100 person-years in adults; p = 0.03). The incidence rate of blips was 10.9 per 100 person-years in children and 4.0 per 100 person-years in adults.
Conclusions
Children are less likely to reach viral suppression and are at higher risk of viraemia while on ART than adults. The significance of LLV and blips needs further study.
). According to World Health Organization (WHO) guidelines, an HIV viral load (VL) of ≥1000 copies/ml should prompt enhanced counselling and a repeat VL test within 6 weeks, and two sequential VL measurements of ≥1000 copies is considered virological failure (VF) and should lead to regimen change (
A proportion of individuals on ART experience low-level viraemia (LLV) and/or transient viraemia; for example, viral blips have been reported in up to 40% of HIV-positive individuals on ART (
Factors associated with HIV viral load “blips” and the relationship between self-reported adherence and efavirenz blood levels on blip occurrence: a case-control study.
Immune activation despite suppressive highly active antiretroviral therapy is associated with higher risk of viral blips in HIV-1-infected individuals.
Data on detectable viraemia after ART initiation are scarce and particularly the incidence and significance of transient and/or LLV in low-income settings, where – unlike high-income settings – regular VL monitoring is not the norm (
The significance of HIV ‘blips’ in resource-limited settings: is it the same? analysis of the treat Asia HIV Observational Database (TAHOD) and the Australian HIV Observational Database (AHOD).
). However, children are potentially at a higher risk of developing viraemia due to weight-based dosing, which may lead to variable drug levels, poor tolerability of drugs, and suboptimal adherence (
). Moreover, psychosocial factors including dependence on a caretaker and delayed disclosure of HIV status to the child may put children at higher risk of remaining viraemic after ART initiation (
Review: an urgent need for research on factors impacting adherence to and retention in care among HIV-positive youth and adolescents from key populations.
In this study, the incidence of and risk factors for detectable viraemia including VF, LLV, and blips were investigated in a cohort of HIV-infected children and adults initiated on ART in Harare, Zimbabwe.
Methods
A retrospective cohort study was conducted using data collected from patients who attended Newlands Clinic, Harare, Zimbabwe, a not-for-profit HIV clinic that provides care for children and adults. ART is provided free of charge according to national guidelines. Routine 6-monthly VL monitoring is performed (during the study period, this was done using the Roche COBAS AmpliPrep/COBAS TaqMan48 version 2.0 test; Roche Molecular Systems, Pleasanton, CA, USA). Adherence and psychosocial counselling is provided at each clinic visit (at least every 3 months). Adherence is assessed by pill count for each antiretroviral drug at every clinic visit. In the case of suspected VF (a VL ≥1000 copies/ml), the patient receives intensified adherence counselling for 6 weeks and repeated VL testing. Those with a VL ≥ 1000 copies/ml in two consecutive VL measurements are considered to have VF and are switched to second-line ART (protease inhibitor (PI)-based regimen and change of at least one nucleoside reverse transcriptase inhibitor (NRTI)). Those on second-line ART who have a VL ≥1000 copies/ml despite counselling undergo HIV drug resistance testing and are considered for third-line ART.
Individuals who were ART-naïve and initiated ART between August 2013 and August 2015 and who had at least two VL tests after ART initiation were included in the study. The following data were extracted: age, sex, date of ART initiation, ART regimen, adherence, height, weight, clinical history (WHO HIV disease stage, history of tuberculosis, opportunistic infections, chronic comorbidities), and laboratory parameters (VL, CD4 count, haemoglobin) at the time of ART initiation and during the follow-up period until September 28, 2017.
Data analysis
All patient data are stored in a secure electronic database. The data were anonymized prior to analysis. Statistical analyses were performed in Stata 14 (StataCorp LLC, College Station, TX, USA). The outcomes were the incidence of VF, LLV, and viral blips. Viral blip was defined as a VL measurement ≥50 copies/ml preceded and followed by a VL below the limit of detection (<50 copies/ml) (
The significance of HIV ‘blips’ in resource-limited settings: is it the same? analysis of the treat Asia HIV Observational Database (TAHOD) and the Australian HIV Observational Database (AHOD).
). LLV was defined as a VL ≥50 to <1000 copies/ml in at least two consecutive VL tests. The WHO definition for VF was used (VL ≥1000 copies/ml in two consecutive VL measurements) (
The proportion of participants who did not achieve viral suppression by 12 months of ART (the 12-month cut-off was chosen to allow for a VL test to confirm suppression at month 12) was estimated and the factors associated with virological non-suppression were studied using logistic regression.
The incidence rates of viral blips, LLV, and VF among those who had achieved viral suppression during the follow-up period and had at least two VL tests after initial VL suppression were estimated. Nelson–Aalen cumulative hazard curves were plotted to evaluate the incidence of VF, LLV, and viral blips after VL suppression by age at ART initiation (children aged 0–19 years and adults aged >19 years).
The factors associated with the occurrence of viral blips were investigated using survival analysis. For the model, participants were included in the analysis at time 0 (time of first suppressed VL test after ART initiation) and followed until a viral blip occurred, or for those patients who remained suppressed until the last VL test available, by the end of the follow-up. Participants who reported treatment interruption in ART for more than 2 weeks were excluded from this analysis. Since the estimated cumulative hazard of blips increased exponentially with time, we fitted the parametric survival regression with Weibull distribution stratified by age group. A value of P (the shape parameter) >1 confirmed that the hazard of failure (viral blip) increased with time.
Age, sex, body mass index (BMI), stunting (in children only), pre-treatment VL, CD4 count, anaemia, WHO clinical stage at ART initiation, history of tuberculosis before ART initiation, chronic comorbidities (adults only), ART regimen, and average adherence were investigated as predictors of detectable viraemia. Stunting was defined as a height-for-age Z-score of <−2 (children only). Height-for-age Z-scores and BMI-for-age Z-scores in children were calculated using WHO reference standards. Anaemia was defined according to WHO criteria (haemoglobin <11 g/dl for children <5 years; haemoglobin <11.5 g/dl for children 5–11.99 years; haemoglobin <12 g/dl for children 12–14.99 years; haemoglobin <12 g/dl for females aged ≥15 years; haemoglobin <13 g/dl for males aged ≥15 years). Adherence was calculated as a percentage of the number of tablets dispensed at the last visit minus the number of tablets returned at the current visit divided by number of tablets that should have been consumed between visits. The average adherence over the study period was calculated for each participant. Age, sex, and CD4 count were adjusted for a priori. All statistical tests were two-tailed and p-values of <0.05 were considered statistically significant.
Ethical approval for the study was obtained from the Newlands Clinic Research Committee, the Medical Research Council of Zimbabwe, and Regional Committee for Medical and Health Research Ethics (Norway).
Results
Of the 725 participants who initiated ART during the study period, 17 were excluded as they were suppressed at baseline (and thus may have received ART previously) and 78 were excluded for having <2 VL measurements available following ART initiation (Figure 1). The baseline demographic, clinical, and laboratory characteristics of the remaining 630 participants (127 children and 503 adults) are shown in Table 1. The median follow-up time was 2.8 (interquartile range (IQR) 2.3–3.2) years for children and 2.7 (IQR 1.8–2.8) years for adults. The sex distribution was equal in the paediatric group (49.6% female vs. 50.4% male), while there were more adult female than male participants (62.2% female vs. 37.8% male), consistent with studies in Sub-Saharan Africa, which have shown higher HIV treatment coverage among women (
). Twenty-one children, all aged below 3 years at ART initiation, were commenced on PI-based regimens, as per the national guidelines, which recommend PIs as part of the first-line regimen in children below 3 years. Ten adults initiated treatment with a PI-based regimen based on clinician judgement, with reasons including severe anaemia, Kaposi sarcoma, WHO stage 4, third trimester of pregnancy, and breastfeeding.
Table 1Characteristics of study participants. Results are presented as the number (percentage), or as the median (interquartile range), unless indicated otherwise.
Characteristics
Children (n = 127)
Adults (n = 503)
Demographic
Male
64 (50.4%)
190 (37.8%)
Age at ART initiation, years
10 (3–15)
37 (31–44)
Stunted at ART initiation (height-for-age Z-score <−2)
24 (18.9%)
Underweight at ART initiation (BMI Z-score <−2 for children or BMI <18.5 kg/m2 for adults)
After ART initiation, had at least one episode of one of the following: oral candidiasis, necrotizing gingivitis, herpes zoster, histoplasmosis, cryptococcal meningitis, molluscum contagiosum, genital warts, tonsillitis.
ART, antiretroviral therapy; BMI, body mass index; WHO, World Health Organization; NRTI, nucleoside reverse transcriptase inhibitor; NNRTI, non-nucleoside reverse transcriptase inhibitor; PI, protease inhibitor.
a Chronic comorbidities included hypertension, diabetes, or renal failure diagnosed before inclusion in the study.
b After ART initiation, had at least one episode of one of the following: oral candidiasis, necrotizing gingivitis, herpes zoster, histoplasmosis, cryptococcal meningitis, molluscum contagiosum, genital warts, tonsillitis.
Of the 630 participants, significantly more children than adults did not achieve viral suppression by 12 months post ART initiation (19.7% children vs. 5.6% adults; p< 0.001). Younger age and CD4 count ≤200 cells/mm3 at baseline were associated with not achieving viral suppression in adults (Table 2). Over the follow-up period, 106 (83.5%) children and 482 (95.8%) adults reached VL suppression, with the median time to VL suppression being 0.5 (IQR 0.4–1.3) years for children and 0.5 (IQR 0.2–0.8) years for adults.
Table 2Logistic regression analysis of the risk factors for not achieving viral suppression by 12 months of ART.
Of the 588 participants who achieved viral suppression after ART initiation, 516 had sufficient VL data to study the incidence of detectable viraemia. Over the follow-up period, 57 (11%) participants experienced a viral blip. Blips occurred more commonly in children than in adults (22.9% vs. 8.3%; p< 0.001), with the incidence rate being 10.9 (95% confidence interval (CI) 7.2–16.6) per 100 person-years in children and 4.0 (95% CI 2.8–5.5) per 100 person-years in adults (Figure 2A). Fifty percent of blips in children and 71.4% of blips in adults were of low magnitude (50–199 copies/ml). The median time from viral suppression to a blip was 1.9 (IQR 1.4–2.3) years in children and 1.8 (IQR 1.0–2.1) years in adults.
Figure 2Nelson–Aalen cumulative hazard curves for the development of (A) viral blips, (B) low-level viraemia (LLV), and (C) virological failure (VF) after viral load suppression. Curves for children are presented in red and curves for adults in blue.
Seven (1.4%) participants (four children and three adults) experienced LLV and 12 (2.3%) participants (eight children and four adults) developed VF after initial suppression during follow-up. The incidence of LLV was 1.9 (95% CI 0.7–5.1) per 100 person-years in children and 0.3 (95% CI 0.1–1.0) per 100 person-years in adults (Figure 2B). VF was more common in children compared to adults (8.3% vs. 0.9%; p< 0.001). The incidence of VF was 4.0 (95% CI 2.0–7.9) per 100 person-years in children and 0.4 (95% CI 0.2–1.2) per 100 person-years in adults (Figure 2C). Notably, 2.2% of children <10 years and 13.7% of those ≥10 years developed VF (p = 0.05). All four adults with VF had re-suppression (three following a switch from a non-nucleoside reverse transcriptase inhibitor (NNRTI)-based to a PI-based regimen, one remaining on the same NNRTI-based regimen), while only three children had re-suppression (two following a switch from an NNRTI-based to a PI-based regimen, one remaining on the same NNRTI-based regimen) during the follow-up period.
Four out of 22 children (18.2%) experienced both viral blips and VF during the follow-up. Blips were followed by VF in two participants, while blips occurred after VF and re-suppression in the other two cases. In those participants with blips followed by VF, the magnitude of blips was low, while in those who had blips after VF, blips were of medium (200–499 copies/ml) and high (>500 copies/ml) magnitude. Among two children with blips after VF, one had re-suppression on the same ART regimen with adherence counselling, while another had re-suppression after a switch to a PI-based regimen. Two children with blips followed by VF remained unsuppressed by the end of follow-up.
Due to low rates of LLV in the study participants after initial VL suppression, risk factors for only viral blips were investigated in the survival analysis. The survival analysis included 507 participants (94 in the paediatric group and 413 in the adult group). No baseline characteristics were found to be associated with an increased risk of viral blips in children or adults (Table 3).
Table 3Risk factors for the development of viral blips by age.
Among individuals starting ART, a significantly higher proportion of children compared to adults did not achieve viral suppression by 12 months. Likely reasons include poor tolerability to antiretroviral drugs, inadequate ART dosing, or suboptimal adherence (
Factors associated with virological failure and suppression after enhanced adherence counselling, in children, adolescents and adults on antiretroviral therapy for HIV in Swaziland.
). Some individuals may require longer periods to achieve suppression, as shown in the present study, where over the follow-up period, 83.5% of children and 95.8% of adults did achieve viral suppression. One possibility is that perinatally HIV-infected young children tend to have sustained high VL in the first years of life with slow reduction of peak VL with increasing age (
). This potentially could contribute to incomplete initial VL suppression and a higher risk of viraemia. Yet the present study results do not support this, as the exclusion of children aged ≤3 years did not change the rates of viral non-suppression in the paediatric study group. Lower CD4 count at baseline was associated with not being suppressed at 12 months in adults, a finding consistent with other studies (
The effects of age on associations between markers of HIV progression and markers of metabolic function including albumin, haemoglobin and lipid concentrations.
This study found a significantly higher incidence of viral blips among children compared to adults. The reported incidence of blips varies in different populations (
Factors associated with HIV viral load “blips” and the relationship between self-reported adherence and efavirenz blood levels on blip occurrence: a case-control study.
The significance of HIV ‘blips’ in resource-limited settings: is it the same? analysis of the treat Asia HIV Observational Database (TAHOD) and the Australian HIV Observational Database (AHOD).
), largely due to different definitions and variability of the assays for VL. We used a common definition of viral blip, as a single VL measurement ≥50 copies/ml preceded and followed by VL <50 copies/ml (
). Compared to studies that have used a similar definition, it was found that a lower proportion of adults in the present study experienced blips: 8.3% during 2.1 years of follow-up compared to 18.6–40% followed up from 1.08 to 1.58 years (
reported a high incidence of blips (40%) in a cohort of patients who received an unboosted PI-based treatment regimen in a high-income country. In addition, the median number of VL measurements was 17 per participant, compared to an average of six measurements in the present study in which only 2.1% of adults were receiving a ritonavir-boosted PI-based regimen.
It is likely that more frequent VL testing will increase the probability of detecting blips, which may explain the discrepancies between our findings and those of studies conducted in high-income settings. However, in a study comparing the incidence of blips in high-, middle-, and low-income countries, 21% of the individuals in the first category experienced blips, while only 11% of participants did so in the middle/low income group, despite a similar ratio of blips to the number of VL measurements across the settings (
The significance of HIV ‘blips’ in resource-limited settings: is it the same? analysis of the treat Asia HIV Observational Database (TAHOD) and the Australian HIV Observational Database (AHOD).
). HIV plasma RNA measurements are subject to pre-analytic and laboratory errors, variations in cut-off point, and variations in lower limit of detection among the different VL assays, which may have an impact on the incidence of blips (
Factors associated with virological failure and suppression after enhanced adherence counselling, in children, adolescents and adults on antiretroviral therapy for HIV in Swaziland.
Virological response and resistance among HIV-infected children receiving long-term antiretroviral therapy without virological monitoring in Uganda and Zimbabwe: observational analyses within the randomised ARROW trial.
Virological response and resistance among HIV-infected children receiving long-term antiretroviral therapy without virological monitoring in Uganda and Zimbabwe: observational analyses within the randomised ARROW trial.
). This is considerably higher than the results of the present study, but may be explained by the use of WHO 2006 criteria for ART initiation, which relied on clinical and/or immunological assessment and not just the diagnosis of HIV infection. The majority of studies in low-income countries have investigated the incidence of any detectable viraemia in one or two VL measurements. For example, a cross-sectional study among adolescents with HIV conducted in an urban setting in Cameroon utilized two consecutive VL measurements to detect sustained VL suppression (VL < 50 copies/ml in two VL tests) (
). In that study, 18.6% of participants had a VL measurement above 50 copies/ml in a single test. Another study conducted in Ethiopia found detectable viraemia (defined as an HIV-1 RNA of 41–1000 copies/ml) in a single VL measurement in 13% of HIV-positive children who had received first-line ART for a median of 24 months (
). In a study by Jobanputra et al., 38% of children aged <10 years and 34% of children aged 10–19 years had single detectable VL (>100 copies/ml) followed by VL re-suppression (
Factors associated with virological failure and suppression after enhanced adherence counselling, in children, adolescents and adults on antiretroviral therapy for HIV in Swaziland.
). Although the results from these studies are not directly comparable to the present study findings, they show that detectable viraemia is common in paediatric patients. A multicentre cohort study of data on children with HIV in the UK and Ireland showed that 22% of participants with sustained viral suppression experienced transient viraemia (defined as a single VL >50 copies/ml between two VL tests below 50 copies/ml), with an incidence of 12 per 100 person-years, similar to the findings in our study (
The incidence of LLV and VF after reaching VL suppression was also significantly higher in children than in adults. Furthermore, in the present study, adolescents (those aged ≥10 years) were more likely to experience VF compared to younger children, a finding that has also been reported by other studies conducted in Africa (
Rates and predictors of non-adherence to antiretroviral therapy among HIV-positive individuals in Kenya: results from the second Kenya AIDS indicator survey, 2012.
High rates of virological failure and drug resistance in perinatally HIV-1-infected children and adolescents receiving lifelong antiretroviral therapy in routine clinics in Togo.
). A possible reason for the higher VF rates reported in other studies is the cross-sectional design of studies and lack of information regarding initial VL suppression prior to inclusion in the study (
High rates of virological failure and drug resistance in perinatally HIV-1-infected children and adolescents receiving lifelong antiretroviral therapy in routine clinics in Togo.
). Due to regular VL monitoring and the longitudinal design of our study, it was possible to distinguish between those who developed VF and those who never reached suppression. Another explanation for low VF in our cohort is the high standard of care provided at Newlands Clinic, with routine VL testing, continuous adherence monitoring support, and a rapid response with intensified adherence counselling and/or ART switch among those who have a non-suppressed VL (
). Routine VL testing, which is not available as standard of care in many low-income settings, can detect VF earlier than targeted VL monitoring, i.e., VL testing prompted by clinical or immunological deterioration.
A number of studies have shown that episodes of LLV and viral blips are associated with an increased risk of subsequent VF in adults (
A single quantifiable viral load is predictive of virological failure in human immunodeficiency virus (HIV)-infected patients on combination antiretroviral therapy: the Austrian HIV cohort study.
). While some studies have suggested that episodes of transient viraemia may result in the selection of drug-resistant HIV strains, leading to an increased risk of VF (
Performance of HIV-1 drug resistance testing at low-level viremia and its ability to predict future virologic outcomes and viral evolution in treatment-naive individuals.
The significance of HIV ‘blips’ in resource-limited settings: is it the same? analysis of the treat Asia HIV Observational Database (TAHOD) and the Australian HIV Observational Database (AHOD).
The significance of HIV ‘blips’ in resource-limited settings: is it the same? analysis of the treat Asia HIV Observational Database (TAHOD) and the Australian HIV Observational Database (AHOD).
). However, these studies have been conducted largely in adults. In the present study, not only did children have a higher risk of VF, but they also had a higher incidence of viral blips as well as LLV, and the significance of these in predicting VF needs further study. Furthermore, careful virological monitoring is warranted in children.
The strengths of this study are the use of a standardized definition for viral blips, regular VL monitoring, availability of detailed clinical data including adherence data, and good follow-up rates. The limitations of the study include the relatively small sample size and low number of blips in each age group, and there may therefore be inadequate power to detect an association between the covariates and the incidence of viral blips. Furthermore, the follow-up period was not long enough to investigate whether blips or LLV increase the risk of subsequent VF. Adherence may have been over-estimated, as the pill count is not the optimal measure of adherence.
In conclusion, this study demonstrated that detectable viraemia is common among children, although its role with regard to long-term outcomes necessitates further study in low-income settings, especially where HIV prevalence is high and the availability of VL monitoring remains limited.
Acknowledgements
We are grateful to the staff and patients of Newlands Clinic for their assistance.
Funding
This study was supported by grant from HelseNord (HNF 1387-17). RAF is funded through a Wellcome Trust Senior Fellowship in Clinical Science (206316/Z/17/Z). The funding sources were not involved in the analysis, interpretation of the data, writing of the report, or in the decision to submit the article for publication.
Ethical approval
Ethical approval was given by Newlands Clinic Research Committee, the Medical Research Council of Zimbabwe (MRCZ/E/188), and the Regional Committee for Medical and Health Research Ethics, Norway (REK 2015/1650).
Factors associated with HIV viral load “blips” and the relationship between self-reported adherence and efavirenz blood levels on blip occurrence: a case-control study.
Performance of HIV-1 drug resistance testing at low-level viremia and its ability to predict future virologic outcomes and viral evolution in treatment-naive individuals.
Factors associated with virological failure and suppression after enhanced adherence counselling, in children, adolescents and adults on antiretroviral therapy for HIV in Swaziland.
The significance of HIV ‘blips’ in resource-limited settings: is it the same? analysis of the treat Asia HIV Observational Database (TAHOD) and the Australian HIV Observational Database (AHOD).
Review: an urgent need for research on factors impacting adherence to and retention in care among HIV-positive youth and adolescents from key populations.
A single quantifiable viral load is predictive of virological failure in human immunodeficiency virus (HIV)-infected patients on combination antiretroviral therapy: the Austrian HIV cohort study.
Rates and predictors of non-adherence to antiretroviral therapy among HIV-positive individuals in Kenya: results from the second Kenya AIDS indicator survey, 2012.
High rates of virological failure and drug resistance in perinatally HIV-1-infected children and adolescents receiving lifelong antiretroviral therapy in routine clinics in Togo.
The effects of age on associations between markers of HIV progression and markers of metabolic function including albumin, haemoglobin and lipid concentrations.
Virological response and resistance among HIV-infected children receiving long-term antiretroviral therapy without virological monitoring in Uganda and Zimbabwe: observational analyses within the randomised ARROW trial.
Immune activation despite suppressive highly active antiretroviral therapy is associated with higher risk of viral blips in HIV-1-infected individuals.
34567-3&id=gr1.jpg){kind=link}
34567-3&id=gr2.jpg){kind=link}