A retrospective study based on the analysis of patient records in the HPGRB (Provincial General Reference Hospital in Bukavu) to sort out all cases of adverse effects of PLIHV from 2018 to 2024. This was a retrospective cross-sectional study based on a review of medical records of people with HIV receiving antiretroviral therapy at the Provincial General Reference Hospital of Bukavu. Data were extracted at a single point in time from existing records, which did not allow longitudinal follow-up.

Figure 1. The Provincial General Reference Hospital of Bukavu (HPGRB) [11].

Figure 1 shows the picture of the site. HPGRB is a tertiary medical structure of the Congolese State ceded to the Archdiocese of Bukavu in 1995 for its management, located at n˚02 on Avenue Michombero, Nkafu district, Kadutu commune. At the secondary level, it is a teaching hospital for Universities, Higher Institutes of Medical Techniques and Institutes of Health Sciences. With a capacity of 450 beds, 52 private rooms and 16 standard rooms for the hospitalization of patients [11]. It is structured in departments including internal Medicine; Surgery; Gynecology and Obstetrics; Pediatrics; Specialties; Medical Biology and Anatomopathology; Medical Imaging; Intensive Care Unit; Emergency Care Service, and the Mortuary. The medical staff includes 90 doctors, 207 nurses, pharmacists, laboratory technicians, midwives and anesthesiologists.

This is a cross-sectional study conducted from January 2018 to January 2024 at the Outpatient Department of Internal Medicine of the Hospital in the care of people with HIV. After authorization from the HPGRB, patients carrying an HIV retrovirus, followed at the HPGRB for at least three months, aged 18 years and over, were included, whose medical records were consulted. Patients were excluded if they were under 18 years of age, pregnant at the time of ART initiation, or had a medical record missing key baseline variables, such as age, sex, antiretroviral regimen, or follow-up data. The exact number of records screened before these exclusions were applied was not systematically logged. This limits the reproducibility of the selection process, and is acknowledged as a limitation below. Sociodemographic data (age, sex, marital status, level of education, profession), clinical-biological data, data on antiretroviral treatment and side effects of treatment were collected on a standardized data collection form. The degree of severity of adverse effects was assessed using the scale for rating the severity of adverse events in adults validated by the Guide to Integrated HIV Management in the Democratic Republic of the Congo.

These categories from 1 to 4 allow us to classify the most frequent adverse effects of antiretrovirals according to their severity, as follows: (a) Grade 1: mild intolerance, not influencing the patient’s daily lifestyle and not requiring immediate drug treatment [12]; (b) Grade 2: moderate intolerance, partially influencing the patient’s daily lifestyle and requiring drug treatment; (c) Grade 3: severe intolerance, completely influencing the patient’s lifestyle, requiring drug treatment or even emergency hospitalization; (d) Grade 4: intolerance endangering the patient’s life [13]. The medical record allowed us to collect information concerning the adverse effects of ARVs when significant organ dysfunction occurs. Treatment interruption was defined as any temporary or permanent discontinuation of antiretroviral therapy, as reported by the patient or documented in the medical record. Information obtained from patient interviews was cross-checked with pharmacy attendance records to minimize recall bias.

This study was conducted in accordance with the fundamental principles of the Declaration of Helsinki. The protocol was approved by the ethics committee of the Provincial General Reference Hospital of Bukavu (HPGRB). Adverse effects were defined according to the national HIV management guidelines of the Democratic Republic of the Congo. They included any clinical or biological abnormality occurring after initiation of antiretroviral therapy. Severity was graded using the national adverse event severity scale, ranging from Grade 1 (mild) to Grade 4 (severe). Adverse effects were documented by clinicians during routine follow-up visits and recorded in patients’ medical files. Both clinician-documented and patient-reported adverse effects were included. Data were retrospectively extracted from medical records. Incomplete or missing records were excluded from severity analyses. Although the study covered a six-year period, analyses were based on the most recent documented clinical information for each patient. Repeated measurements over time were not analyzed, which helped to minimize time-related bias.

Data were entered and analyzed using descriptive statistical methods. Categorical variables were summarized as frequencies and percentages. Continuous variables were summarized using medians and interquartile ranges. Associations between adverse effects and treatment interruption were assessed using bivariate analyses. The exact statistical tests underlying each comparison were not systematically documented in the source data, so only the resulting p-values are reported here. A p-value below 0.05 was considered statistically significant. Multivariable analysis was not performed, because several key covariates, including exact treatment start date and complete viral load data, were missing for a large share of patients. This point is discussed further in the Limitations section (Table 1).

Generally, HIV infection predominantly affects the young sexually active population, as shown by the median age of our study population, which was 44 [35 - 52] years. Studies conducted in the DRC and Burkina Faso reported comparable median ages [20 - 21]. Within the study population, women were predominantly represented, at 55.8%. These results are comparable to those found in the DRC in 2020 and in Mali in 2015, which respectively showed a female predominance of 77.2% and 61.41% [14][15]. UNAIDS in 2020 reports that the female sex accounts for approximately 48% of all new HIV infections in 2019 worldwide and 59% in sub-Saharan Africa [16]. Moreover, women often have more access to screening services than men due to prenatal consultations [17]. To this is added a combination of biological factors and inequalities related to gender and circumcision, which protects men from sexually transmitted infections by 51% to 60% [18][19].

The patients were predominantly at clinical stages 2 (65.2%; n = 150/230) and 3 (13.0%; n = 30/230) of the WHO with a median CD4 lymphocyte count of 269 cells/mm³ at the beginning of care. This result is comparable to data from other studies conducted in Burkina Faso and Senegal in 2018, which respectively show a predominant clinical stage 2 and 3 of the WHO (59%) among patients with an average CD4 T lymphocyte count of 245.8 cells/mm³ and an advanced stage of infection with a CD4 T lymphocyte count below 200 cells/mm³ in 45.6% of cases [20].

This situation could be explained by the fact that the patients would have mostly discovered their serological status at an advanced stage of HIV infection, characterized by a predominance of opportunistic infections (64.8%; n = 149/230) as the circumstance of HIV diagnosis, with tuberculosis being the most frequent among these opportunistic infections (58.6%). Indeed, it is the most frequently associated opportunistic infection with AIDS in tropical areas [21]. Data on viral load was available for less than a third (23.80%) of the patients. This data is comparable to that found in a study conducted in Burkina Faso in 2018, during which the viral load was very rarely measured. These data could be justified by a lack of financial resources because despite the health insurance present in the country, the non-negligible costs of blood tests are borne by the patient.

Overall, antiretroviral treatment was well-tolerated. Nearly half of the patients (41.7%; n = 88/210) experienced at least one adverse effect. Most side effects were reported in patients taking the TDF-3TC-EFV combination. Among the 88 patients with at least one adverse effect, digestive disorders were the most frequent. Diarrhea affected 63% of them (n = 55/88) and vomiting affected 60% (n = 53/88). Nausea followed at 48% (n = 42/88) and abdominal pain at 30% (n = 26/88). Neurological disorders were also frequent. Headaches affected 55% of these patients (n = 48/88) and dizziness affected 40% (n = 35/88). Insomnia and behavioral disorders followed at 18% (n = 16/88). Cutaneous manifestations were less frequent. Skin rash affected 40% of patients with adverse effects (n = 35/88), pruritus affected 28% (n = 25/88), and prurigo affected 32% (n = 28/88). Hematological disorders were dominated by anemia, at 40% (n = 35/88). Metabolic disorders affected 24% of patients (n = 21/88), mostly through blood sugar disorders, at 22% (n = 19/88). Each patient could report more than one adverse effect. This is why the percentages within this subgroup add up to more than 100%.

These results are comparable to those found in a study in Mali in 2017 in which digestive disorders were predominantly observed, followed by neurological, cutaneous, hematological, and metabolic disorders [22]. This high frequency of digestive disorders was also found during the work of Denget al. in China in 2022 among patients on TDF-3TC-EFV. Unlike a study conducted in Mali in 2019, which reported a predominance of neurological disorders in 46.2% of cases, marked by headaches and dizziness, followed by digestive disorders dominated by anorexia (37.2%) and diarrhea (28.1), cutaneous and mucosal manifestations with a predominance of pruritus (33.4%), hematological disorders marked by anemia (19.4%), and metabolic disorders dominated by elevated blood sugar levels (47.2%) [23].

Although efavirenz-containing therapeutic regimens are known to be more often associated with central nervous system side effects [24], these differences could be explained by the short duration of our study, whereas the Mali study took into account side effects in patients over 15 years old in which all patients underwent intensive screening for long-term adverse effects, whereas in our cross-sectional study, adverse effects were mainly self-reported. However, although usually of mild or moderate intensity, these digestive disorders posed real problems with adherence and therapeutic management, sometimes leading to complete or partial interruption of treatment. The problem was more acute because this combination was the most prescribed. The availability since the end of 2020 of the TDF-3TC-DTG association in Gabon has provided a salutary alternative, which should enhance the overall effectiveness of HIV management [25].

The predominance of efavirenz-based regimens observed in this study reflects national treatment practices during the earlier years of the study period. During this time, dolutegravir-based regimens were not yet widely implemented in routine HIV care. The progressive transition toward dolutegravir-based regimens in more recent years should be considered when interpreting these findings, as efavirenz is known to be more frequently associated with neuropsychiatric and digestive adverse effects.

Adherence was assessed by combining two methods: interview (subjective method) and regularity at ARV dispensing appointments (objective method). Adherence to antiretroviral treatment in the study population was reported in less than one-third of the patients, regardless of the therapeutic regimen used, with non-adherence rates of 71% under TDF + 3TC/FTC + EFV and 68% in patients under TDF + 3TC/FTC + DTG. These results corroborate with a study conducted in Gabon in 2019, in which patients were non-adherent at 66% [25]. On the other hand, these results are higher than those presented in Congo in 2020, which showed a non-adherence rate of 25.5% [14] respectively. Our low reported adherence rate could be due to different population characteristics, the quality of health services, study parameters, and the adherence evaluation system used. The focus should therefore be on adherence to antiretroviral treatment, particularly during counselling sessions, as this is necessary for an optimal clinical response and complete viral suppression.

Several factors influence the quality of adherence to ART: treatment-related factors, patient-related factors, environmental factors, and healthcare provider-related factors [24]. The main determinants of non-adherence in this study were primarily: adverse effects, denial, stigmatization, religion, medication shortages, forgetfulness, travel. It had also been noted that the patient/healthcare provider relationship, the patients’ level of knowledge about HIV infection, and the support of a close relative could be considered as factors influencing adherence. Colinnet et al. in Gabon in 2019 had found as reasons for non-compliance: misunderstanding of the disease, denial, and the accessibility of adverse effects.

This study provides useful information for developing future interventions aimed at improving adherence to antiretroviral therapy. Due to a wide range of obstacles and facilitators to adherence, it has been difficult to develop effective interventions to improve long-term adherence to medications. Future interventions aimed at increasing self-efficacy in adherence and more effectively managing side effects through education and guidance from healthcare providers are necessary.