Abstract
Objective:
Adverse drug reactions (ADRs) to antiretroviral therapy (ART) are an important cause of hospitalization, treatment discontinuation, and regimen changes in both developed and developing countries. This study is the first to examine and understand ADRs in HIV-infected patients in Nicaragua.
Methods:
A retrospective descriptive study was conducted from May 2010 to March 2011, in a cohort of HIV-infected patients receiving ART at the largest public hospital in Managua, Nicaragua. Patients were identified based on ADRs reporting on a standardized antiretroviral pharmacotherapy form. Subsequently, chart reviews of these patients were performed in order to document the specific ADRs.
Results:
Six hundred ninety-two patients on ART were included. The incidence of ADRs was 6.4% (95% confidence interval [CI] 4.5-8.2). Females demonstrated a higher incidence, that is, 10.2% (95% CI 5.3-15.1, P = .020). Patients treated with combinations of zidovudine (ZDV)/lamivudine (3TC) and emtricitabine (FTC)/tenofovir (TDF) had fewer ADRs (P < .01) than those using other combinations. Five patients were hospitalized or had a prolonged hospitalization secondary to ADRs, with no mortality attributed to ADR. The most common manifestations of ADRs were central nervous system (20 of 44), gastrointestinal (12 of 44), and dermatologic (8 of 44) reactions. Adverse drug reactions were classified as “likely ADRs” (25 of 44) and “possible ADRs” (19 of 44). No ADRs were preventable.
Conclusion:
Adverse drug reactions most frequently affected the central nervous system. No ADR was life threatening. The frequency of ADRs in this Nicaraguan patient population was less than that reported from other studies in resource-limited settings.
Introduction
The high rates of adverse drug reactions (ADRs) secondary to antiretroviral therapy (ART) in developed and developing countries are a major cause of hospitalization, prolonged hospital stays, and high costs. 1 -7 It has been shown that ADRs are associated with nonadherence to treatment, discontinuation of ART, treatment failure, and changes in ART regimens. 8 -13 Understanding the frequency and distribution of ADRs among HIV-infected patients is imperative to optimally manage their disease, given the significant consequences of ADRs in this patient population.
Despite being the second poorest nation in the Western Hemisphere, Nicaragua has the lowest burden of HIV in Central America. 14 A national HIV/AIDS management strategy is administered under the guidance and supervision of the Nicaragua AIDS Commission. In 2012, an estimated 7000 people were living with HIV and 1700 were receiving ART. The patients were distributed throughout 30 health districts. A single national referral hospital, Hospital Dr Roberto Calderon Gutierrez (HERCG), provides the majority of the ART in the country. 15 -17 As of March 2011, a total of 692 HIV-infected patients were receiving ART through HERCG.
Adverse drug reactions have become an important clinical problem in the Nicaraguan health system and at HERCG and one of the principal contributors to the discontinuation of ART by patients. To date, there have been no publications describing the ADRs in the Nicaraguan population. We aim to describe the main ADRs manifested by HIV-infected patients receiving ART in this setting. This study classifies, analyzes, and evaluates the factors related to the ADRs of ART in Nicaragua. The first-line ART regimen in Nicaragua comprises 2 nucleoside reverse transcriptase inhibitors (NRTIs) with a non-NRTI (NNRTI; note 1). The second-line drug regimen comprises 2 NRTIs with a boosted protease inhibitor (PI; note 2).
We hypothesize that the frequency and distribution of ADRs in HIV-infected patients on ART in Nicaragua are similar to that seen in other resource-limited settings. Although there are many studies about ADRs in Africa, few have been conducted in Latin America and none in Nicaragua. 1,11,18,19
Materials and Methods
This is a retrospective descriptive study conducted in the Department of Infectious Disease at HERCG in Managua, Nicaragua. The study was approved by the HIV/AIDS Health Department of Managua, Nicaragua, and the ethics committee of this country in compliance with the Helsinki declaration. The inclusion criteria were as follows: HIV positive, age 16 years or older, on ART, and documented ADR. An ADR was defined using the World Health Organization (WHO) definition, which states, “a reaction which is noxious and unintended and which occurs at a dose normally used in humans for prophylaxis, diagnosis or therapy of disease or for the modification of physiological function.” 20 The clinical stage of HIV infection for each patient was recorded using the WHO classification system. 21
Data were collected on ADRs by retrospectively reviewing data forms that were kept for each patient as part of the treatment protocol for all patients on ART. Subsequently, each patient’s medical record was reviewed for documentation of any ADRs. Adverse drug reactions were recorded in the medical record by physicians experienced in caring for HIV-infected patients. Data were also collected in a prospective fashion for those patients attending follow-up appointments or who were hospitalized after January 2011. The antiretroviral (ARV) drugs were as follows: efavirenz (EFV), nevirapine (NVP), zidovudine (ZDV), lamivudine (3TC), tenofovir (TDF), emtricitabine (FTC), and lopinavir/ritonavir (LPV/r).
When an ADR was identified, the Hartwig severity scale was used to classify the ADR. 22 The Hartwig severity scale classifies the severity of ADRs into 7 levels. To classify how preventable an ADR may have been, we used a scale developed by Schumock and Thorton, subsequently modified by Otero et al. 23,24 We estimated the probability of developing an ADR using the method put forth by Naranjo et al. 25 We also documented whether the ADR was anticipated or unanticipated for a particular drug, based on the recommendations from the Department of Health and Human Services. 26
Categorical variables were expressed in absolute numbers. For quantitative variables with normal distributions, the mean and standard deviation were reported. For quantitative variables with nonnormal distributions, the median and interquartile range (IQR) were reported. Proportions were compared using chi-square or Fischer exact tests, and the differences were considered significant when the P value was less than .05. Confidence intervals (CIs) of 95% were used. All data were analyzed using the statistical program Intercooled Stata 9.0 (StataCorp LP, College Station, Texas).
Results
A total of 692 patients were identified and included in this analysis. The overall incidence of ADRs was 6.4% (95% CI: 4.5-8.2). In women, the incidence of ADRs was 10.2% (95% CI: 5.3-15.1) and in men it was 5.1% (95% CI: 3.2-7.1). The difference in ADR incidence between women and men was statistically significant (P = .020). The time on treatment before an ADR developed ranged from 2 days to 7 years and 3 months (2675 days), with a median of 71.5 days (IQR: 21-167 days; Table 1).
Adverse Drug Reaction Incidence in Patients on Antiretroviral Therapy.
Abbreviations: ADRs, adverse drug reactions; ART, antiretroviral therapy.
a Women with higher ADR incidence than that of men (P = .020).
b Median and interquartile range.
Of those patients with ADRs (44 cases), 61% (27 of 44) were male and the average age was 35 years (standard deviation ±9.6). More than half of patients with ADRs belonged to HIV/AIDS clinical stage III or IV (11 and 13 patients, respectively). Fifteen patients were of stage II and 5 patients were of stage I. According to the Naranjo probability scale, 57% (25 of 44) of cases were “likely” ADRs and 43% (19 of 44) were “possible” ADRs. None of the ADRs were classified as “certain” (Table 2). Utilizing the Hartwig severity scale, we did not identify any severe ADRs, which would have warranted management in intensive care units. No fatalities were reported from ADRs. Using the Schumock and Thorton scale, none of the ADRs identified were “preventable.” The most common ADRs were in the central nervous system (dizziness, hallucination, aggressiveness, somnolence), in the gastrointestinal system (nausea, vomiting, and diarrhea), and the dermatologic system (skin rash) (20 cases, 12 cases and 8 cases respectively). Less frequently reported manifestations included the hematopoietic system (anemia; 3 cases), the renal system (azotemia; 1 case), and the endocrine system (1 case; gynecomastia) (Table 3).
Efavirenz was implicated in the most ADRs (26 of 44 patients), followed by lopinavir/ritonavir (8 of 44 patients) and zidovudine (5 of 44 patients). There was no significant difference (p > 0.05) between the use of efavirenz, nevirapine, tenofovir, or lopinavir/ritonavir in predisposing to higher frequency of ADRs. When a single agent was identified as the causative agent of ADRs, zidovudine and lamivudine presented a lower proportion of ADRs compared to the other drugs. Finally, the combination tenofovir/emtricitabine demonstrated a lower rate of ADRs in comparison to lopinavir/ritonavir (p < 0.01) (Table 4).
Characteristics of Adverse Drug Reactions.
Abbreviations: ADRs, adverse drug reactions; WHO, World Health Organization; SD, standard deviation.
a Mean age 35 years (SD ±9.6 years).
b No ADRs between 5 and 7 Hartwig severity scale.
c No ADRs preventable in Schumock and Thorton scale modified by Otero et al.
d No ADRs classified as certain by Naranjo questionnaire.
Adverse Drug Reaction Incidence by Organ System.
Antiretroviral Drugs Suspected to Cause Adverse Drug Reaction.
Abbreviations: ARV, antiretroviral; ADRs, adverse drug reactions.
a Fisher exact P value.
Discussion
The incidence of documented ADRs in Nicaragua was 6.45%, which is lower than that of most other countries, including both developing and developed countries. 27,28 Other studies have reported ADR incidence as 34.5% in Brazil; 66% in Lima, Peru; 66.35% in Piura, Peru; 86% in Chhattisgarh, India; and 92.2% in Belo Horizonte, Brazil. 9,29,30,31 The striking difference between the rate we report in our population and the rates reported in these studies may be a difference in how ADRs were defined and documented. Many studies do not describe how they defined ADR, and in 1 study with a very high rate (92.2% in Belo Horizonte, Brazil) the ADRs were based on patient reports. Our method, using physician documentation for the ADR, may be more stringent. Another reason that we may have had such low rates is because of our limited use of stavudine (d4T) and didanosine (ddI). 32 Both drugs have been removed from the recommendations in the Guidelines for Antiretroviral Management in Nicaragua. 17
Our population demonstrated a notable lack of lipodystrophy, which lies in sharp contrast to numerous other studies, with some reporting rates as high as 46.5%. 16,31,33,34 The lack of lipodystrophy in our population is likely secondary to the short follow-up interval as this ADR is seen with prolonged use of ART. Similarly, the newer generation NRTIs used in Nicaragua are known to have a lower rate of lipodystrophy than that of some earlier generation NRTIs. The final reason for lower than expected rates of lipodystrophy may have been that anthropometric measurements were inconsistently recorded in the medical records.
Notably, this study gives a much more detailed classification of the ADRs than other studies in similar settings. Our study, using detailed classification systems, should be able to be reproduced in other resource-limited settings as very few expenses incurred in the design and execution of the study.
In conclusion, we have come to understand that a minority of patients receiving ART in Nicaragua experience an ADR. It has been well documented that the availability of simplified ARV regimens by using combination tablets reduces the occurrence of ADRs. 12,13,14,17,35 We hope to be able to acquire these formulations in the coming years in order to maintain or decrease our rate of ADR even further.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.