This prospective observational study was conducted in the outpatient department of skin and venereal disease in a tertiary care hospital from June 2013 to June 2014. The study was approved by ethics committee of the institute to which this hospital is attached.

All patients irrespective of age and gender presenting to dermatology outpatient department with suspected CADR after systemic drug consumption and who were willing to give informed consent were enrolled in the study. ADR was defined as, “A response to a drug which is noxious and unintended, and which occurs at doses normally used in man for the prophylaxis, diagnosis, or therapy of disease, or for the modifications of physiological function” [12].

The patients in whom proper drug history could not be elicited due to problems such as inability to recall names of medicines consumed, language barrier etc. were not involved in the study. Patients in whom cutaneous manifestations were suspected to be due to use of topical medications or indigenous (homeopathic, ayurvedic etc.) medications and patients who declined to participate were also excluded from the study.

The patients satisfying inclusion and exclusion criteria were enrolled in the study. Detailed clinical history, drug history and relevant information like onset of the reaction, it’s duration and temporal association with drug intake if any, enlistment of all drugs taken preceding the onset of reaction, past history of drug rashes, reports of relevant laboratory investigations undertaken to arrive at a clinical diagnosis etc. was recorded in a pre-designed case record form.

For knowing pattern, CADRs were categorized into various morphological types such as maculopapular rash, urticaria, fixed dose eruption etc. The diagnosis of CADR was made by a dermatologist. Various CADRs were defined according to standard reference [13]. Maculopapular (MP) rash was graded from grade 1-3 as per Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 [14]. According to this criteria, Grade1 MP rash is defined as macules/papules covering <10% Body Surface Area (BSA) with or without symptoms (e.g., pruritus, burning, tightness), Grade2 MP rash as macules/papules covering 10-30% BSA with or without symptoms (e.g., pruritus, burning, tightness etc.); limiting instrumental Activities of Daily Living (ADL) and Grade3 MP rash as macules/papules covering >30% BSA with or without associated symptoms; limiting self-care ADL.

Causality assessment of CADRs was done on the basis of World Health Organization-Uppsala Monitoring Centre causality assessment system [15]. CADRs with causality assessment possible and above were included in the final analysis. Severity of CADRs was assessed by modified Hartwig and Siegel scale [16].

The data was entered in the excel sheet. Demographic data was expressed as percentage or mean ±SD. Descriptive statistics was used to express the results regarding pattern, severity and causality assessment of the ADRs.

Demographic characteristics of study population are shown in [Table/Fig-1]. Seventy percent patients experiencing CADR belonged to female gender. About 68% of patients fell in the age group of 21-40 years. The clinical pattern of CADRs is shown in [Table/Fig-2]. MP rash was predominant pattern of CADRs and occurred in 76.67% of cases. Most of these patients (66.67%) exhibited Grade 3 MP rash. Urticaria was next common finding that occurred in 8.89% of patients. Steven-Johnson Syndrome (SJS) occurred in 4.44% of cases. Pruritis accompanying CADR was seen in 46.67 %(42/90) cases. All patients of SJS were hospitalized and treated successfully. Amongst the drug classes suspected to have caused CADRs [Table/Fig-3] antiretroviral agents were implicated in 75.56% of CADRs while antibacterial, antiepileptic and non-steroidal anti-inflammatory drugs (NSAIDs) were implicated in about 20% of CADRs. The individual drugs implicated in various CADRs with their frequency are shown in [Table/Fig-4]. As can be seen from this table, Nevirapine (NVP) was the suspected drug in 39 cases of MP rash, 5 cases of urticaria, 4 cases of SJS, 2 cases of pustular rash and 2 cases of angiodema and accounted for 57.78% (52/90) of CADRs. Causality assessment is shown in [Table/Fig-5]. Causality was probable in 57.78% and possible in remaining 42.22%. Severity assessment is shown in [Table/Fig-6]. On modified Hartwig and Siegel scale, 78.89% CADRs belonged to mild category, 16.67% to moderate category while 4.44% belonged to severe category.

The skin is most commonly involved organ in adverse drug reactions. Although few CADRs are potentially life threatening and cause significant morbidity and mortality, most CADRs have favourable course and generally resolve after discontinuation of the offending agent. Consequently, most of the patients with CADRs are likely to present and get treated in an outpatient setting, making prospective surveillance of CADRs in an outpatient setting essential.

The age range in our study was 3-67 years with majority of patients falling in the age group of 21-40 years. This is similar to other studies and shows that no age is exempted from CADRs [17,18].

In present study, females were predominantly affected by CADRs and male to female ratio was 1:2.33. Earlier studies have shown either female preponderance [6], male preponderance [17] or no gender difference [18]. Predominance of females in present study could be explained by the fact that most of CADRs in the study were attributed to anti-retroviral medications (nevirapine in particular) and previous studies indicate potential gender differences in the frequency and severity of CADRs to anti-retroviral drugs [19]. It is well known that pharmacokinetics and/or pharmacodynamics of a drug may differ in males and females but how these differences lead to increased frequency of ADRs for certain drugs in females is not precisely known [20].

The causative agents and pattern of CADRs differs from centre to centre depending upon the drugs used and services offered. Our hospital has Anti-Retroviral Therapy (ART) centre established in the year 2004 and regularly provides ART to HIV positive patients. HIV infected patient’s present complex immunological alterations which pose these patients at a higher risk of developing CADRs due to drug hypersensitivity [21].

The present study found that antiretroviral drug nevirapine (NVP) was most commonly associated with CADRs in an outpatient setting. NVP was incriminated in 39 cases of MP rash out of 69 (39/69, 56.12%) and 5 cases of urticaria out of 8 cases (5/8, 62.5%). All four cases of SJS, two cases of pustular rash and two cases of angioedema were attributed to NVP [Table/Fig-4].

NVP is most commonly used non-nucleoside reverse transcriptase inhibitor (NNRTI) as a part of first-line ART. According to the revised National AIDS Control Organization (NACO) ART initiation guidelines November 2011, ART should be started if CD4 count is < 350 cells/mm³ in stages 1 & 2 and irrespective of CD4 count in stages 3 & 4. It is also mentioned by NACO that patients who initially were on NVP-based ART and shifted to efavirenz due to anti-tubercular treatment (ATT) should again be shifted to NVP without any lead in dose after completion of rifampicin-based ATT [22].

A recent study by Lokhande et al., has shown that there is striking increase in the incidence of NVP-induced cutaneous rashes of all forms (4.64% patients treated before November 2011 vs 9.03% patients treated after November 2011) and considerable increase in frequency of severe kind of reactions with the revised guidelines [23]. Number of previous studies has shown NVP to be the most common anti-retroviral agent associated with CADRs including severe CADRs like Stevens-Johnson syndrome, Toxic epidermal necrolysis and Stevens-Johnson Syndrome-Toxic epidermal necrolysis overlap [24–27].

Evidence suggests that there is strong genetic predisposition to cutaneous reactions with NVP and certain Human Leukocyte Antigen (HLA) genotypes are strongly associated with CADRs [28,29]. Generation of electrophilic quinoid species during NVP metabolism and their covalent reaction with bionucleophiles has also been suggested to be possible underlying mechanism for NVP toxicity [30].

Besides NVP, other anti-retroviral drugs namely efavirenz, zidovudine, lamivudine and atazanavir were also suspected in CADRs especially MP rash in present study [Table/Fig-4]. Earlier studies showing anti-retroviral medications associated with diverse and frequent skin manifestations support our finding [31,32].

In present study, anti-retroviral medications were suspected to be responsible in more than 75% of cases. In remaining 25% of cases, anti-bacterials, anti-epileptics and NSAIDs were suspected to be culprits. Antimicrobials, NSAIDs, and anti-epileptics have been consistently reported to be causative agents in CADRs by several studies. In a systematic review by Patel et al., based on 18 prospective studies of CADRs during January 1995 to April 2013, major drugs associated with CADRs were antimicrobials (45.46%), NSAIDs (20.87%) and anti-epileptics (14.5%) [5]. A prospective, observational study in an outpatient setting by Saha et al., found that antimicrobials (sulfonamides, fluoroquinolones, and β lactams), analgesics and anti-epileptics accounted for most of CADRs [18]. Despite established ART centre, this study did not report any CADR to anti-retroviral medications probably due to practice of managing ART-induced ADRs by the ART centre itself. Other studies conducted in an out-patient setting also indicate that antimicrobials, anticonvulsants and NSAIDs are commonly associated with CADRs in an out-patient setting [33,34]. Here again, anti-retroviral drugs were not reported to cause CADRs indicating variation in drug-use or practices-followed.

The most common morphological pattern of CADR in our study was MP rash followed by urticaria, SJS and fixed dose eruption. Our findings are supported by several previous studies [18,27,33,34] and a systematic review by Patel et al., [5].

The causality assessment was probable in 57.78% and possible in 42.42% of cases in our study which is similar to previous studies [34]. Majority of CADRs (96%) belonged to category of mild to moderate severity while about 4% of reactions were severe. Earlier studies support this finding [5,34].

There are some limitations to present study. As most of the patients attending OPD of this hospital come from lower socio-economic strata and primarily rely on drugs supplied free of cost by the hospital, the suspect-drug data and pattern of CADRs generated from this study may not be reflective of the pattern of other tertiary care centers catering to patients of higher socio-economic strata or offering different services than ours. Causality assessment is not always straight-forward especially in polypharmacy cases and none of the case in our study could be classified as definite as rechallenge was not attempted deliberately in out-patient setting owing to potential risk associated with it. In this regard, it is interesting to know that in a study by Gangar et al., [35], an attempt to rechallenge with NVP or delaviridine resulted in recurrence of rash in >75% patients and in 70% who were crossed over to the alternative agent.

Some patients with minor, self-limiting drug eruptions may not have reported to skin OPD in present study and might have been treated by other departments. This is suggested by our observation that >66% of patients of MP rash belonged to Grade 3.

Anti-retroviral drugs especially NVP was suspected in most of the CADRs in an outpatient setting of a tertiary care hospital providing ART. Older drugs i.e. antimicrobials, NSAIDs and anti-epileptics still caused substantial CADRs. MP rash was most common morphological pattern while severe life-threatening CADRs occured in 4% of cases. Female gender was twice as susceptible to CADRs as male gender.

With increasing recognition of variety of clinical manifestations of CADRs with anti-retroviral and other drugs, elucidation of underlying cellular and molecular mechanisms is needed. Deeper understanding of underlying mechanisms coupled with identification of risk factors for serious CADRs and appropriate genetic screening of groups at higher risk may improve outcomes of skin reactions in future.