Human immunodeficiency virus (HIV) infected patients with highly active antiretroviral therapy(HAART) have led to substantial reductions in morbidity and mortality. Pattern of use of HAART regimens result in developing adverse drug reactions (ADRs).¹ Worldwide, 33 million people are infected with HIV infection and around 3 million people have access to HAART.² India as a developing country stands at second position in having highest burden of HIV/Acquired immunodeficiency syndrome(AIDS) and related opportunistic infections.³ In India, HIV infected patients receive a fixed dose of antiretroviral therapy (ART) regimen consisting of either zidovudine (ZDV) or stavudine (D4T) with lamivudine (3TC) in combination with either efavirenz (EFV) or nevirapine (NVP).Most often under reporting of ADRs was observed among Indian clinicians due to inadequate training on use of drugs and patient safety monitoring, with increase in use of antiretroviral increases the risk for ADRs, resulting in humanistic and economic burden to the HIV infected patients as well as to the society.⁴The aim of this study was to assess patterns of use, occurrence, incidence, severity and causality of ADRs to HAART in Indian HIV positive patients.

A prospective study was conducted in HIV-infected hospitalized in-patients and out-patient visit during follow-up from August 2009 to May 2012 at National AIDS Control organization (NACO) approved; Antiretroviral therapy (ART) Centre, district government hospital, Udupi, India. The study protocol was approved by the University ethics committee. The World Health Organization (WHO) definition⁵ of ADRs was adopted to detect ADRs with respect to red blood cell disorders, skin and appendages disorders, gastro-intestinal system disorders, central and peripheral nervous system disorders, liver and biliary system disorders, urinary system disorders, body as a whole, general disorders, psychiatric disorders, metabolic and nutritional disorders and resistance mechanism disorders by using system organ classes and codes using WHO adverse reaction terminologies (WHO-ART)⁶.By active follow-up after treatment and the adverse event was detected by asking patients directly or by screening patient’s case records. HIV-infected patients of either sex who were receiving fixed dose of first-line ART combination regimens as per national treatment guidelines were included and HIV-infected patients treated with “traditional medicines” practiced in India like ayurveda, yoga, naturopathy, unani, siddha and homeopathy were excluded. Study procedure was explained and informed consent was obtained. The enrolled patients with average number of ten days of hospital stay as in-patient were followed for short term ADRs during ward rounds in collaboration with the clinicians and long term intensive monitoring of ADRs with their pattern of use of HAART regimens was documented by the clinical pharmacist at every six months of regular follow-up at out-patient visit. Suspected ADRs to HAART was documented by the department of pharmacy practice where the national pharmacovigilance programs for reporting of ADRs exist. The enrolled patients were followed for investigation for anemia i.e. (baseline hemoglobin level) and baseline symptoms of peripheral neuropathy i.e. (numbness, pins- and needles, burning and aching sensation, sensation of temperature in the legs and feet and cramping) before initiation of HAART and thereafter at every six months of regular follow-up was documented. Clinical pharmacist reviewed patient’s treatment chart and case notes was made on demographic details, past medical history, diagnosis of HIV status, time of initiation of ART, duration of HAART, type of opportunistic infections and HAART regimen implicated. The laboratory investigations such as liver function tests, lipid profile, blood sugar level and renal function test along with the pattern of occurrence of ADRs to HAART was documented using ADR documentation forms with details of suspected antiretroviral drugs involved, treatment of management of ADRs and their outcome. Necessary intervention to the treatment for prevention of ADRs was made by the clinical pharmacist in cooperation with the treating physician. Naranjo’s algorithm⁷ was used for causality assessment of ADRs. The ADR was documented as “not predictable” if the drug had previously been well tolerated by the patient at the same dose and route of administration. The ADR was recorded as “predictable” If there was a previous history of allergic reactions to the drugs or history of exposure. Reported ADR with a literature incidence of ≥1/100 was documented as “predictable.” The severity of suspected ADRs was assessed using the modified Hart wig and Siegel scale.⁸ Modified Shumock and Thornton criteria⁹ were used to assess the preventability of ADRs.

Frequencies with their percentage were used to represent demographic details of the patient. Statistical analysis was carried out using Statistical Package for Social Science (SPSS) version 17.0 (Systat Software, Inc., Chicago, Illinois).

A total of 1982 HIV positive patients were enrolled with inclusion criterion 1181(59.6%) males, 801(40.4%) females fromNACO approved ART Center. Out of which 426 ADRs to HAART were reported and documented during the study period. Of the 426 ADRs, majority of female patients 236 (55.4%) developed ADRs compared to males 190 (44.6%). The overall incidence of ADRs to HAART in our study was 21.4% (426/1982). Higher incidence rate of ADRs was noted with 3TC + NVP + D4T (22.9%) and 3TC + NVP + ZDV combinations (22.5%) while the incidence rate of ADRs was lowest with 3TC + D4T + EFV combination (16.8%). Incidence of ADRs to HAART was significantly higher in the age groups 41-60 years (p <0.001), (25.7%) than in other age groups. Incidence of ADRs to HAART was significantly higher with CD4⁺T-cell counts of 350-500 cells/µl (p <0.001), (47.2%). The incidence of ADRs in our study was higher in female population (p <0.001), 11.9 % (236/1982 compared to males {9.5% (190/1982)}. Patients’ demographic details are shown in Table 1.

HAART regimen commonly implicated in ADRs was noted with 3TC+NVP+ZDV 226 (53%). The suspected antiretroviral drug was withdrawn for the management of the ADRs in majority 119 (27.9%) of the reports. Symptomatic treatment was instituted in 91.8% of the ADR cases. Most (86.4%) of the patients were recovered from ADRs on the last day of hospital discharge. Three fatal ADRs of NVP induced cutaneous drug eruptions of Steven Johnson Syndrome (SJS) and Toxic Epidermal Necrolysis (1.1%) were observed during the study period. A total of 7.3% of patients with ADRs were discharged from the hospital against medical advice, which led to loss to follow-up and resulted in an unknown outcome of ADRs.

We found that ‘no-de-challenge’ and “no- re-challenge” accounted for 9.6% and 80.5% of the reported ADRs, respectively. The causality assessment was performed using Naranjo's scale, out of 426 ADRs reported, 294 (69%) were possible, 92 (21.6%) were probable. As per modified Schumock and Thornton scale, 31(7.3%) were ‘definitely preventable ADRs. zidovudine induced anemia were considered under “preventable ADRs” and 4(0.9%) ‘Probably preventable’, 391(91.8%) were ‘non-preventable’. The majority of ADRs 387 (90.8%) were ‘non-predictable’ and 39 (9.2%) were ‘predictable and preventable ADRs’ which includes zidovudine induced vomiting. The characteristic details of the reported ADRs to HAART are shown in Table 2.

The commonly observed ADRs to HAART were anemia 135 (31.7%), vomiting 66 (15.5%) followed by skin rash 55 (12.9%) and peripheral neuropathy 45(10.7%). Pattern of adverse drug reactions reported with highly active antiretroviral therapy are shown in Figure 1. Severity of ADRs was assessed using modified Hart wig and Siegel scale and reported ‘mild ADRs’ i.e. level 1 and level 2 accounted for 0.5% and 50.2% respectively. ‘Moderate ADRs’ i.e. level 3, level 4a and level 5 accounted for 10.3%, 31% and 7.3% respectively. Only three (0.7%) of level 4b ADRs related to hospital admissions that was lead to level 7 ‘severe ADRs’ or fatal ADRs. Results are shown in Figure 2.

Figure 1. Pattern of adverse drug reactions reported with highly active antiretroviral therapy.

Steven Johnson Syndrome (SJS)/ Toxic Epidermal Necrolysis (TEN), Drug

hypersensitivity syndrome (DHS), Immune reconstitution syndrome (IRIS)

Figure 2. Severity of adverse drug reactions reported with highly active antiretroviral therapy.

In our study, female HIV positive patients were found to be more susceptible to ADRs to HAART with higher incidence of 11.9% compared to male HIV positive patients. This result confirms the findings of various studies^{10, 11}that female gender have increased incidence for the development of ADRs. Most of ADRs to HAART were observed with 3TC + NVP + ZDV and 3TC + NVP + D4T. This may be due to large number of naïve HIV positive patients were diagnosed and initiated with first-line HAART similar to another study.¹² In this study, we found that highest percentage of ADRs reported to HAART were among age group of 21- 40 years (61.5%) followed by 41-60 years (34.2%), a finding consistent with the study carried out by Obiako et al.¹³However, other study¹⁰that were carried out reported higher percentage of ADRs to HAART in children’s and elderly patients. This may be due to the fact that in the present study, most patients 1266 (63.9%) who were admitted to the study site were in the age group 21- 40 years.

The occurrence of anemia had the highest percentage (31.7%) with ZDV containing HAART regimen during first four weeks of initiation of HAART. After replacement to ZDV with D4Tcontaining HAART regimen, an improvement in hemoglobin levels (8.0-9.4 g/dl) was observed with our patients. This finding is in accordance with previous published studies from our department where incidence of ZDV induced anemia was higher in comparison with D4Tcontaining HAART regimen in Indian HIV patients.² With the use of ZDV containing HAART regimen, gastrointestinal adverse effects such as vomiting was observed to be major cause for discontinuation of HAART regimen among Indian patients within 4 to 8 weeks of initiation of ZDV. This finding consistent with various studies.^{11, 14, 15}

In our study dermatological manifestations of skin rashes was reported in patients with NVP therapy along with cutaneous drug eruptions of Steven Johnson Syndrome (SJS) and Toxic Epidermal Necrolysis (TEN) was observed (1.1%). In few patients, erythematous macules that progressed to flaccid blisters along with symptoms of fever, stinging eyes and painful swallowing was observed with NVP therapy. In our study patients, with the occurrence of SJS and TEN was managed by discontinuation of NVP and with supportive measures of intravenous fluid administration, antimicrobial therapy, electrolyte maintenance, and with intensive skin care. Three patients were admitted to our study site, due to NVP induced SJS with level of 4b severity of Hart wig and Siegel scale. These three patients was not completely recovered from NVP induced SJS, even after discontinuation of NVP and resulted to level 7 severity of fatal adverse reactions. This is in accordance with published studies where NVP induced SJS and TEN has resulted in life threatening fatal ADRs in HAART treatment.^{16, 17}

Drug Hypersensitivity Syndrome (DHS) with efavirenz (EFV) containing HAART regimen (12.9%) showed definite improvement after discontinuation of EFV. In our study, three cases of EFV induced maculopapular skin eruptions that prolonged into exfoliative erythroderma were presented with moderate in severity. Complete recovery from these ADRs was seen in our patients after four weeks of discontinuation of EFV. The onset of maculopapular skin eruptions due to EFV therapy ranged from 12 to 14 days after initiation of treatment. Our study findings are similar to observations in a published studies.^{18, 19}

During the present study, the occurrence of drowsiness (0.9%) and severe depression (1.4) was observed in patients with EFV therapy. This patients developed aggressive reactions, agitation, emotional liability, neurosis, paranoid and mania. Such psychiatric symptom was greatly reduced by our clinical pharmacist intervention during ward rounds with physician for prescribing EFV therapy at bed time and all patience with EFV treatment at bed time was followed for adverse psychiatric symptoms at every six months of regular follow-up at out-patient visit. These observations are in agreement with the previously published study where EFV therapy resulted with higher incidence of central nervous system disorders.²⁰

In our study, Forty five of our patients experienced D4T induced peripheral neuropathy. This patients developed pain in the hands and at feet with numbness, tingling sensation. The symptom associated with D4T induced peripheral neuropathy was resolved after two to eight weeks after discontinuation of D4T therapy. In our study the incidence of D4T induced peripheral neuropathy was found to be less i.e. 10.7% compared to Browne et al.²¹ where they reported 55% incidence of D4T induced peripheral neuropathy.

In the present study, Lipodystrophy occurred in patients receiving a D4T containing regimen with the incidence of 3.2% which is similar to that of other studies^{22, 23}. In most of our patients, central fat accumulation was observed within the abdomen as well as in the dorso-cervical spine region and upper trunk. Metabolic abnormalities such as elevated cholesterol, elevated triglycerides level, hyperlactataemia, low levels of high-density lipoprotein (HDL) cholesterol and type-2 diabetes mellitus was presented with lipodystrophy. Our study findings are similar to other studies^{21, 23}where in most of the patients developed a greater increase in triglycerides and cholesterol level compared with patients who do not develop lipodystrophy.

In our study, the occurrence of hepatotoxicity was observed after 18 weeks with use of NVP therapy with elevated transaminase, prolonged partial thromboplastin time, with and without hyperbilirubinemia, liver tenderness, and hepatomegaly. Out of seven cases of NVP induced hepatotoxicity, six cases showed greater increase in serum AST and ALT levels more than five times the upper limit of normal values. Also we found that woman, including pregnant women was at a higher risk of developing NVP induced hepatotoxicity than men. Our study findings is agreement with various studies^{25, 26, 27, 28, 29}where they observed similar incidence of severe hepatotoxicity with NVP use.

In the present study, Immune reconstitution syndrome was reported in two patients, with 3TC + NVP + ZDV during initial eight months period of HAART, developed pneumocystis pneumonia and Guillain-Barre syndrome. The finding of our study supported with various studies ^{30, 31, 32, 33, 34} Nephrolithiasis was observed in one patient, who was receiving 3TC + ZDV combination along with additional protease inhibitor. The occurrence of nephrolithiasis in our study suggesting that renal effects was more likely associated with use of additional protease inhibitors rather than the reverse transcriptase inhibitors. This finding is agreement with the studies.^{35, 36}

Strategies to either avoid or minimize the ADRs to HAART greatly influence quality of life in HIV infected patients. zidovudine induced anemia can be prevented either by careful monitoring of complete blood count, blood transfusion or use of epoetin alfa, use of myelosuppressive treatment with sulfamethoxazole-trimethoprim or gancyclovir and ultimately discontinuation of zidovudine containing regimen.³⁷ zidovudine induced vomiting can be prevented by taking medication with meals, nutritional counseling, knowing triggers for nausea/vomiting with specific food products, use of antiemetic drugs, acid reducing agents, tea with ginger or switching medications.³⁸ Skin rash can be managed by using antihistamines or corticosteroids, and discontinuing offended drug or switching therapy.³⁹ The main stays of management of peripheral neuropathy with use of amitriptyline, gabapentine or pre-gabalin, Non-Steriodal Anti-Inflammatory drugs (NSAIDs), opiate analgesics, acupuncture treatment and keeping the extremities warm.⁴⁰ Treatment for lipodystropy involves polylactic acid injections, exercise for fat accumulation, weight reduction, breast reduction surgery and liposuction for dorso-cervical fat accumulation.⁴¹

In India, the pattern of ADRs to HAART in HIV infected patients treated with lamivudine +nevirapine +zidovudine combination predicts higher incidence of anemia with zidovudine use.

HIV infected female patients, patients with age distribution of 41-60 years; CD4⁺ T-cell counts 350-500 (cells/µl) need to be advised to monitor toxicity of HAART. Physician must focus for monitoring all lab investigations for early detection and prevention of adverse effects associated with HAART.

The authors wish to thank the staff of ART Centre, District Hospital, Udupi, for their support and encouragement.