Incidence and predictors of initial antiretroviral therapy regimen change among children in public health facilities of Bahir Dar City, Northwest Ethiopia, 2021: multicenter retrospective follow-up study

Although various plans and policies have been implemented in the past 40 years to combat the impact of HIV/AIDS, the current reduction in HIV/AIDS-related morbidity and mortality is also rested on early diagnosis, early initiation and continuous use of antiretroviral therapy. The inconsistent ART use increases the likelihoods of adverse outcomes in HIV care and treatment programs, including treatment failure [1, 2].

Antiretroviral therapy regimen change (ARTRC) refers to the substitution of one or more of the antiretroviral drugs in the initial antiretroviral therapy (ART) regimen. The drugs may belong to the same or different drug classes [3, 4]. If the antiretroviral treatment regimen is changed for the first time, it is called an initial regimen change (IRC). Often, changes in the antiretroviral treatment regimen will occur as long as the child receives HIV treatment and care for a variety of reasons within a few months and beyond. The reasons may be market availability, toxicity, treatment failure, drug stock-out, or simplification [5]. A one or more ART drugs can also be changed from the initial regimen to another regimen, depending on the patient’s condition and recommendations from WHO or specific country guidelines [6]. Before making changes to the patient’s ART regimen, clinicians will consider the patient’s past regimen, past ART failure episodes, past drug resistance test results, drug costs, and the patient’s ability to tolerate the new regimen [7].

The magnitudes of RC varied across the previously conducted studies in children [3, 8, 9]. In America and Malaysia, the prevalence of initial regimen change among children was 72% [8] and 71.8% [3], respectively. Moreover, a randomized trial of PIs and NNRTIs in the United States of America showed that 72% of participants experienced at least one treatment disruption during 8 years of follow up [8]. In Spain, a follow-up study among 212 HIV-infected children also reported that only 24% of children remained on their first combination therapy [10]. In addition, a prospective study in Ghana reported that 3.3% of the participants had experienced initial RC [11]. Another study in South Africa also revealed that overall incidence of initial ARTRC was 85.9 per 1000 patient-years, only 72 and 26% of children were remained on initial regimen at three and 5 years of ART during 10 years of study follow-up, respectively [12]. Another two different studies in South Africa showed that 11.3 and 26% of the study participants experienced IRC [12, 13]. Similarly, 2 years regimen durability study in South Africa revealed that there were about 15.3% single-drug substitutions, 7.2% treatment interruptions and 3.3% switched to second-line therapy [9].

The initial or subsequent RC could be affected by a variety of factors which might vary greatly with baseline laboratory values CD4 count [14], co morbidity [8, 15], initial regimen type [8, 14], having poor adherence [8, 12], the occurrence of adverse drug reaction/toxicity (ADR) [14], treatment failure [8, 12, 14] and availability of new drug [15]. These factors often cause significant morbidity and antiretroviral treatment failure. The predisposing factors were also varying across the findings of the previous studies [3, 12, 14, 16].

A serious confront of ART drug substitution, switching, interruption and discontinuation are the emergences of drug resistance, making future therapeutic interventions ineffective and narrowing the subsequent possible alternatives [17]. Additionally, RC also affects the success of HIV treatment to achieve Joint United Nation Program on HIV/AIDS (UNAIDS) 90–90-90 and 95–95-95 goals [18, 19].

In order to improve success of the initial ART regimen in HIV care and treatment program; optimizing the limited available ART regimens is vital [20]. Several studies done on predictors of first line HIV treatment failure were also reported that ART regimen change is a predictor of the subsequent ART failure [21, 22]. Even though the availability of new drugs has improved the patient acceptance, satisfaction to treatment, and safety profile of new regimens, the definition of the best available starting ART regimen is hindered by the lack of data on the exact impact of long-term durability of initial ART regimens [23]. As a result, the knowledge of how often a combination regimen is changed can be a good substitute marker of the regimen success. Even though several studies have tried to determine incidence and predictors of initial ARTRC, they were not consistent. So far, there were a limited number of studies on predictors of initial ARTRC among children in Ethiopia.

Therefore, this study aimed to determine incidence and predictors of initial ARTRC among children in public health facilities of Bahir Dar city, Northwest Ethiopia, 2021.

This retrospective follow-up study was aimed to determine the incidence and predictors of initial ART regimen change in Bahir Dar city, Ethiopia. Based on this objective, the overall incidence of initial ART regimen change was 1.85per 100 PMO which were predicted by adherence level, regimen type, tuberculosis and sex of child.

The incidence rate was higher than study done in sub Saharan Africa (0.41per 100 PMO) [16] and Mali (0.28 per 100 PMO [24]. The discrepancy could be due to studies in sub Saharan Africa and Mali were about only switching to second-line ART but this study included both drug substitution and switching to second-line regimen. Furthermore, treatment failure had been diagnosed through the level of CD4 count and WHO clinical stage in the earlier years of HIV care and treatment especially in developing countries but currently, WHO recommends routine viral load test which is important for early diagnosis of treatment failure to meet the 95–95-95 UNAIDS goals [25, 26]. As the finding of this study revealed, the clinicians need to careful for their practice of individual regimen management whether they have been adhered to standards and evidence-based practice along with the national guideline while changing the regimens or not.

The finding of this study showed that the hazard of initial regimen change among children who initiate NVP based regimen was higher than those children who initiated EFV based ART regimen. The finding was supported by large sample study in France [27] and multicenter study in Europe and Thailand [14]. This risk difference might be due to NVP toxicity. NVP toxicity is associated with immune competency results in NVP based regimen to have higher regimen change rate [28]. This could be after WHO recommendations of ART initiation regardless of CD4 count in children [29]. In addition, the finding was deferred from the study done in Malaysia which stated that NNRTI based categories of regimen had no difference for regimen change [3]. This could be due to sample size and analysis model used by this study. The risk of regimen change difference between NVP and EFV based regimens could be due to the adverse effect of EFV more frequently occur within months and self-resolved without regimen change within months. Additionally, EFV has high half life than NVP which result in low frequency of daily dose and good adherence [30]. Besides, the hazard difference between these drugs could be due to the CD4 dependent serious side effect of NVP and its self inducing characteristics of its metabolism [30]. The other hazard difference between the two groups might be due to the high barrier resistance of EFV and its viral suppressing effect over NVP [31]. On the other hand, those children who started LPVr based regimen had lower hazard of initial regimen change than EFV based regimens which is consistent with study done in Thailand [14]. This could partly mirror the reluctance of clinicians to switch children failing LPVr based regimen due to the difficulty in deciding what to switch to, and recommendations to first address adherence issues due to the high resistance barrier and guideline restriction for children aged less than 3 years [26, 32]. This implies that LPVr based regimen is preferable than EFV and NVP based regimens to advance the potential durability of the initial regimen which supports the current WHO recommendation of ARV drug preference lists.

In this study, the hazard of initial regimen change among children who had poor adherence was higher than those children who had good adherence. The finding was similar with the study done in Malaysia [3]. The hazard difference between the two groups could be due to immature disclosure especially in old children. In addition, palatability and preparation of the drugs might be lead to poor adherence in this age group. Similarly, frequency of dosing, drug side effects, child’s age and developmental stage, psychosocial, behavioral, and socio-demographic characteristics of children and caregivers might be associated with poor adherence [33]. Furthermore, infants and children are dependent on others for medication administration. As a result adult caregivers may face barriers that contribute to poor adherence in children including forgetting doses, changes in routine, being too busy, and child refusal [34]. This finding showed that despite the current strategies to boost ART adherence practice of the child, adherence is still challenging in clinical practice of HIV care and treatment program.

This study finding also showed that hazard of initial ART regimen change was higher for being male than being female. The finding was comparable with an international cohort collaboration regimen change study reported that male had higher rate of switching to second line regimen over female [35] and the Pediatric International Epidemiology Databases to Evaluate AIDS (IeDEA) study member countries (Asia-pacific and African countries) [36]. The gender discrepancy might be due to psychosocial transition difference. In addition, sex discrepancies in treatment outcomes are thought to result from differences in both biologic and behavioral factors including antiretroviral metabolism, antiretroviral adherence and retention in care. A related study on sex difference in pharmacologic effect of antiretroviral therapy among adult reported that sex-related variations in the expression and activities of drug transporter genes, proteins, and enzymes involved in phases of the drugs biotransformation [37]. The difference in maintaining initial ART regimen between male and female child implies the requirement of further close supervision for male child by clinicians and other responsible bodies.

Moreover, the finding of this study revealed that the hazard of initial regimen change among children who has history of tuberculosis was higher than those children who hadn’t history of tuberculosis. The finding was consistent with studies conducted in Ghana [11] and Uganda [38]. The hazard difference between the two groups could be due to the occurrences of adverse effect of ART drugs together with the extra dose frequencies of both drugs (ART and anti-TB) which lead to poor adherence [39, 40]. In addition, TB-HIV co infection may impair or delay immune recovery after ART initiations. Furthermore, TB is the most frequent life-threatening opportunistic infection and the leading cause of death among HIV positive children [5]. TB speeds up the viral replication in HIV infected individuals which in turn leads to treatment failure due to ineffective antiretroviral drugs [26]. In addition, the initial ART regimen could be changed due to drug-drug interaction in children who developed tuberculosis while they were on ART [41].

Since the data were collected from secondary source, some important predictors such as family’s wealth index, viral load at baseline and care giver educational level were missed. Additionally, it also missed some important variables like laboratory diagnostic tests (blood chemistry, organ function test) and some socioeconomic factors. Furthermore, it might be prone for selection bias. Another limitation of this study could be over estimation of incidence of initial ART regimen change due to inclusion of all ART regimen changes following national guideline changes.

In a conclusion, the incidence of initial regimen change was high as compared to the previous studies. The possibility of initial regimen change is predicted for HIV positive children who had initiated ART with history of tuberculosis, NVP based regimen, being male by sex and poor adherence. LPVr based regimen is preferable than EFV and NVP based regimens to advance the potential durability of the initial regimen which supports the current WHO recommendation of ARV drug preference lists. Therefore, strengthening the health care providers’ adherence counseling capability, strengthening tuberculosis screening and prevention strategies and care of initial regimen type choice needs attention in the HIV/AIDS care and treatment programs. Moreover, further prospective study is recommended to be conducted by including family’s wealth index, viral load at baseline, care giver educational level, hepatic function test and renal function test.