Recognition of possible risk factors for clinically significant drug-drug interactions among Indian people living with HIV receiving highly active antiretroviral therapy and concomitant medications

1 Introduction

Worldwide, Drug-Drug Interactions (DDIs) is the major health concern associated with drug use in elderly patients, longer hospitalization stay and increased number of poly pharmacy (PP) [1]. DDIs are classified into pharmacokinetic and pharmacodynamics interactions. Some of these DDIs are very minor in severity, while some are potentially serious leading to treatment failure and severe in toxicity [2]. Use of Highly Active Antiretroviral Therapy (HAART) in People Living with HIV (PLW-HIV) has dramatically increased the quality of life expectancy. However, in HIV- Seropositive positive patients (HIV-SP), there is an increase in Opportunistic Infections (OIs), comorbid diseases, non-Acquired Immunodeficiency Syndrome (AIDS) related malignancies. Due to these comorbid diseases, a wide range of poly pharmacy are used along with HAART which leads to adverse drug reactions (ADRs) and DDIs [3–5].

In United States of America (USA), [6] 30 to 42% of HIV-SP patients with HAART are associated for clinically significant DDIs, 20–30% in The Netherlands [7] and incidence rate of DDIs was 65.2% in India [8]. Studies have reported association of increased risk for clinically significant drug interactions (CSDIs) in HIV-SP patients were due to increased age and poly pharmacy [9]. Most of the DDIs with HAART were mediated by inhibition or by induction of hepatic drug metabolism. However, several studies [10, 11] on DDIs were focused with HAART and ignoring other concomitant medications, despite the fact that such medications are responsible for possible DDIs with HAART there by results for non-adherence, treatment failure and development of drug resistance. The management of possible DDIs must be taken into consideration during the selection of HAART regimen, along with concomitant medications so as to minimize undesirable DDIs. This study was intended to recognize the possible DDIs, prevalence, and pattern of occurrence, and to identify possible risk factors for DDIs in PLW-HIV receiving HAART and concomitant medications.

2 Methods

A retrospective medical record review was conducted from Kasturba Hospital (South Indian teaching hospital), Manipal, India. The study was approved by the Institutional Ethics Committee. Data were collected from January to December 2014 from In-patients medical records of HIV-SP patients receiving HAART with concomitant medications and were selected, analyzed for DDIs.

HIV-SP patients above 18 years of age were included in the study; and HIV-SP patients who were not on HAART, with chronic renal disease, traditional alternative medicines, pregnant women and out-patients were excluded from the study. In-patients case records was reviewed by a pharm D graduate clinical pharmacist for possible DDIs with HAART with concomitant medications prescribed. The complete drug treatment of these patients was screened from medical records from the day of admission, till day of discharge and follow-ups. All the relevant data were collected from patients’ case records, treatment chart, including the implication of HAART, OIs, comorbid conditions, risk factors, laboratory reports, number of drug prescribed, including demographic details of the patients, viral load and follow-up CD4 T-cell counts. In a suitably designed ‘Individual case record form’ (ICRF) data related to concomitant drugs with dosage and patients’ allergic reactions to food and drugs were noted and documented. University of Liverpool drug interaction database (ULPDIDB) [12] and Micromedex database (MDB) [13] were used for evaluation of DDIs. Additionally, the severity level of the CSDIs were assessed using Stockley’s drug Interactions [14]. As per criteria of ULPDIDB, DDIs were classified based on three indications. 1. If the DDIs is contraindicated [Red Flag Indication (RFI)] or contraindicated drug-drug interactions (XDDIs). 2. If the DDIs needs close monitoring [Orange Flag Indication (OFI)]. 3. If the DDIs has no significant interactions [Green Flag Indication (GFI)] [12]. World Health Organization (WHO) clinical staging of HIV disease in adults, adolescents and children was used to assess the stage of HIV disease [15]. Definition of CSDIs were defined as those drug interactions that needed a dosage adjustment or drug combination that is contraindicated due to its greater potential for adverse clinical effects [9].

3 Results

During the study, a total of 240 HIV-SP patients with HAART [(males 152 (63.3%), females 88 (36.7%)] were screened from medical case records. Out of which 131 (54.6%) were in the age group of 41 to 59 years and 23 (9.6%) of patients were elderly ≥ 60 years of age. The average age of the HIV-SP patients was found to be 44.72±10.5 years. The majority 139 (57.9%) of patients were presented with BMI of 18.5 to 24.9. Social habits of smoking and alcoholic were 57 (23.8%) and 56 (23.3%) respectively. The CD4 T-cell count in the majority of patients 183 (76.2) was ≤ 200 cell/μl. Stage III HIV diseases was found to be 184 (76.7%). Number of poly pharmacy per prescription excluding HAART regimen ranges from nine to eleven drugs 57 (23.75%). Most of the patients 105 (43.75%) has one comorbid disease. Patient characteristics are shown in Table 1.

The comorbid diseases in the majority of HIV-SP patients was respiratory tract infections; 41 (17.1%), followed by anemia; 34 (14.2%), hypertension; 28 (11.7%) and gastritis; 27 (11.2%). Majority of OIs was tuberculosis; 76 (31.7%) followed by candidiasis; 41 (17.1) and 98 (40.8%) of patients experienced at least one OIs during their previous stay in hospital. Pattern of comorbid diseases and occurrence of opportunistic infections are shown in Table 2.

During the study, 267 DDIs to HAART were reported in 107 patients. Prevalence of DDIs was higher in male population 71 (66.4%) compared to female 36 (33.6%). The majority 60 (56.1) of DDIs were observed in 41 to 59 years of age followed by 18 to 40 years; 36 (33.6%) and 60 years of age and above patients were also included. The DDIs in smoker 34 (31.8%) and alcoholic were 33 (30.8%). The baseline CD4 T-cell count in the majority of patients with DDIs 84 (78.5) was ≤ 200 cell/μl. DDIs were higher in stage III 85 (79.4%), followed by stage II 18 (16.8%) and stage I, 4 (3.8) of HIV diseases. Most 28 (26.2%) of patients who developed DDIs were presented with nine to eleven drugs in their prescription excluding standard HAART regimen and reported with at least one comorbid disease 35 (32.7). DDIs associated with comorbid diseases was respiratory tract infections; 26 (24.3%), followed by hypertension; 18 (16.8%), diabetes mellitus; 15 (14%); and anemia 13 (12.1%). Patients with DDIs were much associated with OIs of tuberculosis; 35 (32.7%) followed by candidiasis; 20 (18.7). In our study, 46 (42.9%) of patients developed one DDIs and 25 (23.3%) developed two DDIs. In a single patient maximum of 16 DDIs were reported. The majority 97 (90.6%) patients had developed ≤5 DDIs, 8 (7.5) developed six to eleven DDIs and 2 (1.86%) patients developed ≥12 DDIs. Characteristics of patients with DDIs are shown in Table 3.

During the study, 267 DDIs to HAART and concomitant medications were observed in 107 patients using Liverpool data and majority of DDIs were of OFI 253 (94.8%) and RFI 14 (5.2%). The prevalence of pharmacokinetic DDIs was 238 (89.1%) and pharmacodynamics DDIs was 29 (10.9%). The prevalence of DDIs was higher in male 178 (66.7%) compared to female 89 (33.3%). The majority of DDIs were reported between 41 to 59 years of age. Most (49.4%) of number of DDIs reported were receiving greater than eight other medications excluding standard HAART regimen. Prevalence of DDIs was higher with HAART regimen of lamivudine+zidovudine+nevirapine (29.2%) and lowest with tenofovir+emtricitabine+nevirapine (0.4%). Assessment of drug-drug interactions as per Liver pool are summarized in Table 4.

Of 267 DDIs to HAART 136 DDIs were able to be analyzed according to MDB. Of the remaining 131 DDIs were excluded because of lack of information’s for DDIs in MDB. Most of the time of onset of DDIs was delayed in nature 87 (64%), followed by not specified 40 (29.4) and very few are rapid in action 9 (6.6%). Severity of DDIs were minor (44.9%), major 39 (28.6%) and moderate in 36 (26.5%). In most of the DDIs, documentation was ‘probable’ (78.7%) followed ‘established’ by 28 (20.6%). Significance rating of possible DDIs was Level-3, 61 (44.9%), followed by Level-1, 42 (30.9%) and Level-2, 32 (23.5%). Assessment of drug-drug interactions (as per MDB) are shown in Table 5.

Influential risk factors for development of DDIs was seen with Ischemic heart disease (p = 0.025), Diabetes mellitus (p = 0.036), cardiovascular attacks (p = 0.003), hypertension (p = 0.026), respiratory tract infections (p = 0.008) and psychiatric disorder (p = 0.001) and pneumocystis carini pneumonia (p = 0.002). A couple of factors such as gender (p = 0.723), age (p = 0.805), thyroid disorder (p = 0.078) were not significantly associated with DDIs. Possible risk factors for DDIs are shown in Table 6.

The highest number of DDIs were reported with efavirenz based HAART combinations 49 (18.4%) followed by zidovudine based HAART 44 (16.5%) (Table 7). CSDIs with PIs and concomitant medications were reported in (n = 44) number of DDIs. HAART regimen much associated for CSDIs with PIs were tenofovir+emtricitabibe+atazanavir+ritonavir. CSDIs with PIs based interactions includes atazanavir +fluconazole 4 (8.85%), ritonavir+ fluconazole 4 (8.85%), ritonavir + quetiapine, lopinavir+ quetiapine, atazanavir + lithium. The drug interactions with PIs are shown in Table 8. Out of 267 DDIs, 14 XDDIs were reported with RFI and 253 DDIs were reported with OFI which needs close monitoring. Medication classes significantly associated for XDDIs were atazanavir with fluconazole 4 (29%), ritonavir with fluconazole 4 (29%), nevirapine with rifampicin 2 (14.3%), atazanavir with pantoprazole 1 (7.1%) and ritonavir with quetiapine 1 (7.1%) (Table 9). General management of DDIs are shown in Table 10. Among 267 DDIs, antiretroviral drug class associated were Nucleoside reverse transcriptase inhibitors (NRTIs), n = 135, (50.6%), Non-Nucleoside reverse transcriptase inhibitors (NNRTIs) n = 88, (32.9%), and Protease Inhibitors (PIs) n = 44, (16.5%) (Fig. 1).

OPEN IN VIEWER

Fig.1

Antiretroviral associated with drug-drug interactions. NRTIs: Nucleoside Reverse Transcriptase Inhibitors; NNRTIs: Non-Nucleoside Reverse Transcriptase Inhibitors; PIs: Protease Inhibitors.

CSDIs with HAART and concomitant medications are common. In India, no study has previously reported risk factors for all possible DDIs with HAART and concomitant medications and also no study reports for prescriber awareness on CSDIs of HAART and concomitant medications in HIV patients. There is lack of data for all possible DDIs associated with HAART and concomitant medications in Indian setup on safety of practice of medicine. Hence, our study evaluating medical record review for recognizing allpossible DDIs to HAART, prevalence, nature of occurrence, and associated risk factors of DDIs to HAART and concomitant medications in PLW-HIV in India is important. The study observed significant pharmacokinetic (89.1%) and pharmacodynamics DDIs associated with use of poly pharmacy (concomitant use of greater than eight drugs), with use of NRTIs, NNRTIs and PIs based HAART regimen, which was similar to that observed in other studies [16, 17]. The prevalence of DDIs in our study was (44.9%) and there exist difference in prevalence of DDIs from various other studies conducted elsewhere [7, 9, 18]. However, Recognition of all possible DDIs from this study will focus to help the treating clinician in India to promote awareness for all possible DDIs associated with HAART and concomitant medications, which will also promote prescriber for preventing CSDIs while prescribing concomitant medications along with HAART regimen.

In our study, risk factors associated for major DDIs was observed with comorbid conditions of cardiovascular disease. Greater risk for DDIs with HAART were observed among patients when treated for hypertension, ischemic heart disease followed by myocardial infraction (p < 0.001). Most of the DDIs were observed among patients who were on HAART with antithrombotic therapy, antiplatelet drugs like clopidogrel, aspirin that are used to control cardiovascular disease risk. Greater risk for DDIs were reported in cases of efavirenz with warfarin, tenofovir with aspirin, nevirapine with amlodipine, nivirapine with clopidogrel and efavirenz with acenocoumarol. These findings of our results will help clinical cardiologists to familiarize with all possible DDIs between HAART and concomitant cardiovascular drugs, so as to prevent DDIs. However, our findings supports the review study of Raposeiras-Roubin et al. in which ischemic heart disease is an associated risk factor in HIV patients treated with HAART and cardiovascular risk [19].

In our study, male gender (p < 0.001) were identified as risk factors for the development of DDIs. This may be due to the fact that most of the male patients were first hospitalized and treated for OIs of tuberculosis, candidiasis, and pneumocystis carnie pneumonia who experienced DDIs at much higher rate whereas, female patients were admitted to hospital at the late or end stage of HIV disease due to social stigma. Other risk factors for DDIs in our study were due to respiratory tract infections followed by psychiatric disorder (p < 0.001). In our study use of antibiotics, drugs for treating OIs, drugs for the treatment of psychiatric disorders such as lithium, Olanzapine, escitalopram, zolpidem, valproic acid, Quitiapine, alprazolam, escitalopram, clonazepam are highly implicated DDIs with HAART regimen. Whereas, Piscitelli et al. study observed a highly significant association of DDIs between the use of drugs for only treating OIs [2].

In our study, majority of DDIs were pharmacokinetic (89.1%) compared to pharmacodynamics (10.9%). This is mainly because most of HAART regimen are extensively metabolized by the cytochrome (CYP) isoenzyme, CYP3A415, CYP450 and CYP3A4. However, other studies have reported higher incidence of pharmacodynamics DDIs in clinical practice [20, 21]. The overall risk of XDDIs was significantly associated with use of PIs compared to NNRTI with lowest risk. XDDIs was observed in patients who were on fluconazole + atazanavir (28.6%), fluconazole + ritonavir (28.6%), nevirapine + rifampicin (14.4%), nevirapine + dexamethasone, ritonavir + quetiapine, lopinavir+ quetiapine and atazanavir + pantoprozole. The risk of XDDIs was highest with HAART regimen with PIs which is similar to other studies [22, 23]. However, differs with various studies [10, 24, 25] in which they found a higher risk for XDDIs was much associated with NNRTI based HAART regimen. In addition, risk of XDDIs (i.e., RFI) was highest in combination of tenofovir+emtricitabine+atazanavir+ritonavir regimen. This may be due to interactions of atazanavir with tenofovir resulting in reduced plasma concentration of the PIs. However this is in contrast to previous reports [26, 27] that when ritonavir is used as a booster, interactions of atazanavir with tenofovir is not clinically relevant.

HIV-SP patients being treated with HAART experienced DDIs with warfarin at a much higher rate when used with PIs or NNRIs [28]. Depending on the type of antiretroviral agent, either inhibition or induction of warfarin metabolism do occur [29]. We observed occurrence of DDIs with warfarin and efavirenz in two cases. DDIs with warfarin and efavirenz is due to alteration in CYP2C9 metabolism. However, we could not directly correlate with anticoagulation effects with increase in international normalization ratio (INR). DDIs with cotrimoxazole was observed in patients who were on lamivudine (14.6%), emtricitabine (11.4%), zidovudine (8.7%) and stavudine (1.9%) containing regimen. In ninety three cases, cotrimoxazole DDIs with HAART regimen suggesting a likely association of risk for DDIs. Our study findings are similar to a study [30] wherein use of cotrimoxazole manifested for higher DDIs in HIV patients.

DDIs with statins are attributed to CYP3A4 inhibitors [31]. In this study, we found five cases of CSDIs with hypolipidemic agents such as atorvastatin and HAART regimens containing nevirapine and efavirenz. This may be due to the fact that in our hospital setup, atorvastatin were the most commonly used class of hypolipidemic drugs prescribed in clinical practice. These findings of CSDIs of atorvastatin with nevirapine or efavirenz containing HAART regimen will help practitioners to recognize and prevent risk for CSDIs associated with the concomitance use of statins along with HAART regimen containing nevirapine and efavirenz. In our hospital setup, extensive use of proton pump inhibitors (PPIs) are common in clinical practice. In our study, one case of XDDIs with the use of PPIs (i.e. Pantoprazole) with atazanavir was reported. These finding are consistent with the study carried out by Lewis et al. [32]. On the other hand, study that was carried by Saberi et al. [33] reported to minimize or to avoid use of PPIs while taking PIs based HAART regimen. However concomitant use of PPIs with PIs are acceptable for less than 30 days of indications [33].

5 Conclusion

HIV patients who are on HAART and treated for opportunistic infections, respiratory tract infections, psychiatric disorder, cardiovascular disease and who are on antiplatelet drugs, antithrombotic therapy need intensive monitoring for DDIs. Recognition of all possible risk factors for clinically significant drug interaction will help clinicians to prevent DDIs. Furthermore, clinicians strictly need to pay attention to avoid “Red Flag Indication combinations” while prescribing concomitant medications along with HAART regimen.

Conflict of interest

The authors declare no conflicts of interest.