Potential drug interactions in patients given antiretroviral therapy

ABSTRACT Objective: to investigate potential drug-drug interactions (PDDI) in patients with HIV infection on antiretroviral therapy. Methods: a cross-sectional study was conducted on 161 adults with HIV infection. Clinical, socio demographic, and antiretroviral treatment data were collected. To analyze the potential drug interactions, we used the software Micromedex(r). Statistical analysis was performed by binary logistic regression, with a p-value of ≤0.05 considered statistically significant. Results: of the participants, 52.2% were exposed to potential drug-drug interactions. In total, there were 218 potential drug-drug interactions, of which 79.8% occurred between drugs used for antiretroviral therapy. There was an association between the use of five or more medications and potential drug-drug interactions (p = 0.000) and between the time period of antiretroviral therapy being over six years and potential drug-drug interactions (p < 0.00). The clinical impact was prevalent sedation and cardiotoxicity. Conclusions: the PDDI identified in this study of moderate and higher severity are events that not only affect the therapeutic response leading to toxicity in the central nervous and cardiovascular systems, but also can interfere in tests used for detection of HIV resistance to antiretroviral drugs.


Introduction
HIV infection affects 36.9 million people worldwide, representing about 0.6% of the world's population.
There are an estimated 1.6 million deaths yearly due to acquired immunodeficiency syndrome (AIDS) (1) . This disease causes a negative impact of multidimensional order into the lives of people. However, a great transformation occurred in the epidemiological profile with the emergence of highly active antiretroviral therapy (HARRT) (2) .
The use of HARRT, which can reduce the viral load to undetectable levels and raise the count of CD4 + T lymphocytes, resulted in the mortality reduction and increased survival rate of the infected individuals (2) .
However, the success of HARRT is associated with the maintenance of a high rate of patient compliance and the prevention and management of drug-drug interactions (DDI) (3) .

DDI is defined as a clinical or pharmacological effect
that results from the co-administration of medications, which alter a patient's response to treatment. DDI occurs when the action of one drug (object, substrate) is altered by the presence of another drug (precipitant, interacting drug) (4) . DDI represents one of the most frequent adverse drug events that result in hospitalization, increase of cardiovascular risk, and abandonment of treatment. These induce adverse events or reduce the therapeutic efficacy, particularly in individuals subjected to polypharmacy (5) .
Polypharmacy, combined with factors such as age, alcohol consumption, illicit drug use, and potentially interactive features of some antiretroviral drugs such as protease inhibitors and reverse transcriptase inhibitors but not nucleoside analogues increase the complexity of therapeutic management and the risks pertaining to DDI (4- 7) . Antiretroviral therapy (ART) agents represent one of the main therapeutic groups with the greatest potential for DDI. Both protease inhibitors and nucleoside analogues are substrates and modulators of the cytochrome P450 enzyme system (6)(7)(8)(9) . International consensus and national guidelines for the management of patients undergoing HARRT, must be put in place to avoid hardships arising from DDI. Despite this, studies conducted in different countries indicate that the prevalence of DDI in the users of ART in outpatient context varies from 21.5% to 67.1%, depending on the age of individuals, the therapeutic classes involved, and the database used to analyze DDI (8)(9) . Participants exposed to DDI showed reduced treatment adherence (9) .
Post-marketing surveillance reported the use of HARRT focused on the identification of potential drug-drug interactions (PDDI), particularly in Brazil, where there are over 405,000 individuals involved in the treatment and this surveillance can contribute to a better understanding and management of clinically relevant PDDI (10) . The term PDDI refers to the possibility of a particular medication altering the intensity of the pharmacological effects of another medication therefore increasing or decreasing the therapeutic effect and/or adverse reactions or the responses other than those originally stemming from the medications (10) .
In this context, it is fundamental that health professionals have knowledge regarding PDDI in people subjected to antiretroviral treatment, as the prescription must consider the characteristics of the drugs and especially the possibilities of these interactions. The scientific literature shows that few studies in the area are carried out by nurses, even though the routine of medications should occupy a strategic position leading to interactions in order to enable nurses to examine their daily work and interfere with medication routine geared to to prevent the occurrence of adverse reactions due to drug interactions. The objective of this study was to determine the prevalence of potential drug-drug interactions in patients with HIV infection undergoing HARRT and to identify the major PDDI in this group and associated factors.

Results
The subjects included 161 participants undergoing HARRT, of which 52.2% (n = 84) were exposed to PDDI.
The average viral load was 5658.89 ± 30020.70 copies/ ml, and the average CD4 + T-cell counts was 476.17 ± 269.69 cells/µl. About 44% of the population had some opportunistic disease in the last year, however there were not present at the time of data collection.
The groups of individuals exposed to potential PDDI and unexposed to PDDI, showed no differences pertaining to gender, age, alcohol consumption, drug use, adherence to therapy, or adverse reaction reporting.
The polypharmacy (p = 0.000) and time of treatment (p = 0.00) showed a significant association with the presence of DDI.
The average age in the group of individuals exposed to potential PDDI and unexposed to PDDI was 44.1 ± 10.5 years and 41.0 ± 10.3 years, respectively (range: 22-67 years). Among individuals exposed to PDDI, 7.1% were elderly, whereas among individuals unexposed to PDDI, 5.2% were elderly. There was statistically significant difference (p = 0.001) among the mean of drugs used in the PDDI group: the mean was 5.08 ± 0.92 and in the non-PDDI group was 4.01 ± 0.14. The average of ART was statistically significant (p To evaluate the adherence to HARRT, we used the "questionnaire for the evaluation of adherence to ART in people with HIV/AIDS (CEAT-VIH)" translated and validated in Portuguese (11). We used the CAGE questionnaire to evaluate alcohol consumption (12).
In the group with PDDI, the average of interactions between ART per patient was 2.07 ± 0.75 and among all therapeutic classes was 2.63 ± 1.42.  Among patients with PDDI, 60.7% showed two PDDIs and 16.7% showed three PDDIs. About two out of ten individuals were exposed to four to nine PDDIs.

Discussion
Despite the evidence of international and Brazilian guidelines for HARRT and the issue of interactions and potential adverse events associated with it (8)(9)15) , the prevalence in this study (52.2%) was higher in comparison to other research conducted with adults for outpatient treatments in India (21.5%) (8) , the United Kingdom (27%) (9) , and England (35%) (9) . Possible explanation for this difference is the fact that this study looked at the PDDI between medicines of groups with high-potential interactions (antiretroviral, central nervous system, and ethanol). In addition, the overall average of drugs consumed and age of the participants in this investigation was higher compared to other studies.  (16) .
The association between polypharmacy and PDDI was confirmed, a fact in line with other investigations that analyzed ART (9) . Polypharmacy is a risk factor in patients undergoing HARRT and relates particularly to those individuals who have a treatment regimen with at least a medication not belonging to the ART group, which can be aggravated by age (17) . Elderly patients have 51% probability of DDI and youngsters a 35% chance in the case of use of 6 and 7 medications, respectively (17) .
Each medicine added in the therapy increases the risk of adverse events by 10%, including DDI (17) .
Despite the risk of DDI associated with polypharmacy, this strategy is critical in HIV-infected patients. The first line of initial treatment typically includes three ART, two Nucleoside analogues, and non-Nucleoside analogs. The first-line treatment consists of tenofovir, lamivudine, and efavirenz and the second-line treatment consists of protease inhibitor and nucleoside analogues (15) .
In addition, in cases where there is a presence of opportunistic infections or co-morbidities, polypharmacy is mandatory (15) .  (18) .
Accordingly, longer exposure to HARRT causes increased frequency of adverse reactions to medication.
Evidence suggests that the incidence of adverse reactions to medication is about 50% in adults who receive HARRT in the outpatient setting (19) . The presence of adverse reactions to medication is 1.6 times higher in people with a CD4 + T-cell count below 200 cells/µl (20) . This may be more associated with the greatest number of drugs used by individuals than with a lower number of these cells. Therefore, therapeutic resources are essential to combat HIV infection and opportunistic infections.
In this study, the values of CD4 + T cell counts and viral load showed no association with the presence of PDDI. However, other authors observed an association of the event with a count less than 200 cells/µl (9,20) , and stated that DDI may affect patients' health. responses, including DDI (21) . In evaluating a specific DDI, it is important to note the relative inhibitory potential of the drug to the particular enzyme (22) .

Co-administration of ART and other medications can
result in important changes in the serum levels, many of which are related to preventable adverse events.
There is evidence that the HARRT, alone or combined with other medications, including action in the CNS, alters the CYP450 metabolism, which was observed in the present study, particularly in the PDDIs of moderate and higher severity (6)(7) .
Regardless of the treatment time, 68.3% of PDDI were classified as moderate. Among these,

12.8% included ART and agents that act in the CNS
(benzodiazepines, antidepressants, and neuroleptics) with a clinical impact of excessive sedation and confusion (6)(7) . Whereas the users of these drugs are in the community, including developing industrial communities, this reaction can interfere with the quality of life and lead to negative outcomes. In these cases, a modification of therapy should be considered. For example, in the PDDI of diazepam and ritonavir, the use of a benzodiazepine such as lorazepam could prevent an increase in sedation effects (23) .
Although there has not been any PDDI reported between ART and other medicines and alcohol that inhibitor association with ritonavir provides higher serum levels, stable and long-lasting protease inhibitor, increasing its power of viral inhibition and reducing the occurrence of resistance mutations (25) .

Conclusions
Risk factors were found for the occurrence of