Cutaneous Adverse Drug Reactions in a Tertiary Teaching Hospital: A Prospective, Observational Study.

OBJECTIVE: To describe clinical patterns, identify associated drugs, and perform causality assessment of cutaneous adverse drug reactions.
MATERIALS AND METHODS: In this prospective, observational study, patients with signs of cutaneous drug reaction from the department of dermatology of a tertiary teaching hospital were included. Patients with viral exanthemas were excluded. The patterns of cutaneous drug reactions, and associated drugs were noted. Naranjo scale was used for causality assessment of adverse drug reactions.
RESULTS: Sixty-five patients (mean age 38.1 years) were included. Skin lesions were seen after 2 days of drug consumption in 25 (38.5%) patients. Pigmentation, fixed drug eruption, and maculo-papular rashes were seen in 13 (20%), 22 (33.8%), and 12 (18.5%) patients, respectively. Thirty-five (53.85%) patients had other associated symptoms. Fever was seen in 22 (33.8%) patients. Thirty-two (49.2%) cases had exposure to antibacterial agents. Most commonly associated drugs with reactions were nonsteroidal anti-inflammatory agents 14 (21.5%) followed by beta-lactam antibiotics 12 (18.5%), and fluroquinolones 7 (10.8%). Fifty-seven (87.7%) adverse drugs reactions had "probable" association with the drug."
CONCLUSION: Fixed drug eruption is the commonest presentation of cutaneous adverse drug reactions and antimicrobial agents are most commonly associated with it. According to the Naranjo scale, the majority of the cutaneous drug reactions have a "probable" association with the offending drugs. Copyright:

Introduction

Drug reactions are an important cause of hospital admissions[12] and in some cases, they can lead to mortality.[3] Adverse reactions contribute significantly to the financial burden on health care resources.[45] In routine clinical practice, almost every clinician encounters at least a few cases of cutaneous adverse drug reactions.[6] Among hospitalized cases, the incidence of cutaneous adverse drug reactions ranges between 1% and 3%.[7] The reported prevalence of acute cutaneous drug reaction is 0.25% with slightly more rates among in-patients than out-patients.[8]

Despite such reactions being common, comprehensive nationwide information about their epidemiology is not available.[6] In many cases, the clinical presentation is not severe. However, early identification of cutaneous adverse reactions and identification of causative drugs is necessary for effective management. Considering clinical picture and relative difficulty or absence of definitive drug-specific laboratory testing, diagnosis is mostly based on clinical judgment.[6] This is important not only for the dermatologists, but all clinicians should be aware of such reactions, should be able to diagnose them at the earliest, and manage them appropriately.[6]

Few investigators from different parts of India have studied the patterns and causative agents associated with cutaneous adverse drug reactions.[91011121314151617] Such studies from Maharashtra[18] are limited.

Objective

The objective of this study was to describe clinical patterns, identify associated drugs, and perform a causality assessment of cutaneous adverse drug reactions.

Material and Methods

In this prospective observational study, we included patients with cutaneous drug reactions presenting to inpatient and outpatient departments in a tertiary teaching hospital from 2015 to 2017. Patients of all age groups and both genders were included in the study. We did not include patients with viral exanthemas. After a detailed history, a clinical examination was conducted to find out the types of skin lesions. Frequency and percentages of each type of skin lesion along with associated symptoms were recorded. Comorbidities in patients presenting with skin reactions, frequency and percentages of drugs consumed, and reasons for consumption of these medicines were also noted. Naranjo scale[19] [≥ 9 definite adverse drug reaction, 5–8 = probable adverse drug reaction, 1–4 = possible adverse drug reaction, 0 = doubtful adverse drug reaction] was used for causality assessment of adverse drug reactions. All cases were included in the study after informed consent and/or assent as necessary. The study was initiated after taking approval from the institutional ethics committee.

Statistical analysis

The collected data were entered into Microsoft Excel for analysis. Categorical data are presented as frequency and percentages, whereas continuous data are presented as mean and standard deviation.

Results

A total of 65 patients (females: 58.5%) were included in this study. The mean age of patients was 38.1 years. Age group-wise distribution of patients is given in Table 1. A history of drug reaction was present in 27 (41.5%) patients.

Table 1

Baseline characteristics of patients

Parameter	Result
Age group n (%)
≤ 20 years	11 (16.9%)
21-40 years	33 (50.8%)
41-60 years	11 (16.9%)
61-80 years	10 (15.4%)
Mean (± SD) age in years	38.1±19.1
Age range in years	7-80
Gender
Female n (%)	38 (58.5%)
Male n (%)	27 (41.5%)
History of drug reaction	27 (41.5%)

The onset of skin reaction was seen in 1 day of drug consumption in 13 (20%) study participants and 2 days of drug consumption in 25 (38.5%). A total of 17 (26.2%) and 10 (15.4%) patients had onset after 3 days and 5 days, respectively [Figure 1]. Urticaria, fixed drug eruption, macular papular rash, erythroderma, Stevens–Johnson syndrome (SJS), and toxic epidermal necrolysis (TEN) appeared within 24 h.

Figure 1

Time for the onset of skin rash

On examination, pigmentation was seen in 13 (20%) patients. Mucosal involvement in the form of erythema and/or ulcer was seen in eight (12.3%) patients out of which genital, lips, and oral mucosa were involved in 2 (3.1%), 5 (7.7%), and 1 (1.5%) patients, respectively.

Fixed drug eruption was observed in 22 (33.8%) patients, whereas maculopapular rash was present in 12 (18.5%) patients. Other patterns of skin lesion in the study population are enlisted in Table 2.

Table 2

Skin lesion pattern in the study population

Skin lesion	n (%)
Fixed drug eruption	22 (33.8%)
Maculopapular eruption	12 (18.5%)
Stevens-Johnson syndrome (SJS)	06 (9.8%)
Toxic epidermal necrolysis (TEN)	03 (4.9%)
Urticaria	07 (10.5%)
Angioedema	03 (4.6%)
Erythroderma	03 (4.6%)
Erythema multiforme	02 (3.1%)
Lupus erythematosus	02 (3.1%)
Leucocytoclastic vasculitis	02 (3.1%)
Dapsone hypersensitivity syndrome	01 (1.5%)
Lichenoid reaction	01 (1.5%)
Pitriasis rosea	01 (1.5%)

A total of 35 (53.85%) patients had other associated symptoms. Fever and joint pain were seen in 22 (33.8%) and 7 (10.8%) patients, respectively. Two (3.1%) patients had both fever and joint pain. One (1.5%) patient each had burning, erosion, itch, and pain. Diabetes mellitus, tuberculosis, and hypertension were present in 7 (10.8%), 6 (9.2%), and 2 (3.1%) patients, respectively. Only in one patient, the viral disease was associated with SJS/TEN.

A total of 32 (49.2%) cases had exposure to antibacterial agents, whereas intake of non-steroidal anti-inflammatory drugs (NSAIDs), antifungal agents, antiepileptic agents, opioids, anticoagulants, and other drugs were reported by 14 (21.5%), 5 (7.7%), 5 (7.7%), 3 (4.6%), 2 (3.1%), and 4 (6.2%) patients, respectively [Figure 2].

Figure 2

Classes of drugs associated with cutaneous adverse reactions

The most commonly aasociated drugs with cutaneous reactions were antibacterials and non-steroidal antiinflammatory agents [Table 3]. In one patient, fixed drug eruption was seen after amikacin administration [Figure 3].

Table 3

Drugs associated with adverse cutaneous drug reactions

Drug	n (%)
Antibacterials (n=32)
Ofloxacin	6 (9.2%)
Amoxycillin plus clavulanic acid	5 (7.7%)
Amoxycilin	3 (4.6%)
Metronidazole	2 (3.1%)
Ornidazole	2 (3.1%)
Isoniazid	3 (4.6%)
Amikacin	1 (1.5%)
Cefadroxil	1 (1.5%)
Cefixime	1 (1.5%)
Ceftriaxone	1 (1.5%)
Dapsone	1 (1.5%)
Minocycline	1 (1.5%)
Norfloxacin	1 (1.5%)
Isoniazid	1 (1.5%)
Rifampicin	1 (1.5%)
Streptomycin	1 (1.5%)
Cefotaxime	1 (1.5%)
Non-steroidal anti-inflammatory agents (NSAIDs) and paracetamol (n=14)
Diclofenac	5 (7.7%)
Paracetamol	5 (7.7%)
Nimesulide	2 (3.1%)
Ibuprofen	1 (1.5%)
Other NSAID	1 (1.5%)
Antifungals (n=5)
Terbinafine	2 (3.1%)
Fluconazole	1 (1.5%)
Itraconazole	2 (3.1%)
Antiepileptics (n=5)
Carbamazepine	3 (4.6%)
Phenytoin	2 (3.1%)
Opioids (n=3)
Tramadol	3 (4.6%)
Anticoagulants (n=2)
Warfarin	1 (1.5%)
Rivaroxaban	1 (1.5%)
Others (n=4)
Hyoscine	1 (1.5%)
Furosemide	1 (1.5%)
Ranitidine	1 (1.5%)
Telmisartan	1 (1.5%)

Figure 3

Fixed drug eruption after amikacin administration

A total of 7 patients out of 65 were below 18 years of age. Out of them, three had fixed drug eruption and two had SJS and TEN. One patient developed lichenoid drug eruption due to carbamazepine.

The patients were prescribed suspected drugs for primary illness of pain 17 (26.2%), diarrhea 9 (13.8%), infection 8 (12.3%), tuberculosis 6 (9.2%), fever 6 (9.2%), fungal infection 5 (7.7%), and seizures 5 (7.7%). Acidity, acne, cataract, congestive cardiac failure, chronic kidney disease, ear infection, dyslipidemia, edema, and vomiting were the initial complaints of one patient each.

Classification of cutaneous adverse drug reactions according to Naranjo Scale showed 57 (87.7%) adverse drugs reactions as “probable,” whereas 8 (12.3%) were “possible” adverse drug reactions.

Upon withdrawal of the offending drug, 44 (67.7%) patients showed a reduction in symptoms, whereas in 21 (32.4%) patients, there was no response.

Discussion

The risk of adverse drug reactions is a common concern for both clinicians and patients as it may affect the compliance of therapy and may lead to hospitalization and result in other complications sometimes even death. A large study from two large general hospitals in England reported a 6.5% prevalence of admission due to drug-related adverse events.[4] It is therefore important for physicians to understand the patterns of such reactions so as to identify them early and treat them appropriately. In this study, we examined clinical presentation, identified drugs associated with cutaneous adverse drug reactions, and causality of these reactions in our setting.

Cutaneous adverse drug reactions can occur in any age group. A study by Patel and Marfatia,[10] reported minimum age of 1 year and a maximum of 80 years. This suggests that no age is immune to the occurrence of cutaneous adverse drug reactions. In our study also, the maximum age was 80 years, and the minimum age was 7 years. Several studies[10111220] have reported its common occurrence in adult patients. In our study also, adults predominated the study population. The mean age of patients in our study was almost similar to another study from India.[11] In a study by Patel and Marfatia,[10] age group of 41–50 years was the most affected. In a study from a South Indian hospital, 21–40 years was the most commonly affected age group.[12] In other studies, people of the age group 26–37 years of age[20] and 20–39 years were most commonly affected.[11] In our study, patients between 21 and 40 years were maximum in number. This could be because of more awareness and reporting of drug reactions by the adult age group. It is expected that elderly patients may have a higher risk of drug interactions and drug reactions because of polypharmacy, however, the occurrence of cutaneous drug reactions in this age group was not high compared to adults at least in these studies.

Studies have reported more occurrences of cutaneous reactions in females as compared to males. A study[11] reported male to female ratio of 0.87:1. Another retrospective study reported a higher rate in females compared to males.[1] In our study also, females outnumbered the male population. Is there a hormonal correlation?

Many patients in our study group, had a history of drug reactions. It is important to check a history of drug reactions before prescribing medication. A careful history of drug-related allergy can help in reducing the incidence of adverse reactions.

Studies have reported the development of lesions after 1 to 45 days of drug intake.[1016] In our study, all cutaneous reactions were observed within 5 days of drug intake.

As reported in other studies, we too observed fixed drug eruption, maculopapular eruption, and urticaria as common presentations in patients with cutaneous adverse drug reactions. Sharma and colleagues reported maculopapular rash (34.6%), fixed drug eruption (30%), and urticaria (14%) as the most common presentations.[9] Mahapatra and Keshri reported maculopapular drug reaction, fixed drug eruption, and urticaria in 23%, 14%, and 13% patients, respectively.[13] A study from South India reported maculopapular rash (42.7%) as the most common type of reaction.[12] In another retrospective study, macular and maculopapular rashes were the most common adverse cutaneous reactions (42.0%).[1] In our study, the percentage of patients with fixed drug eruption was much higher than those with maculopapular eruption. Our observations are similar to a study by Pudukadan and Thappa[11] who also reported fixed drug eruption as the most common (31.1%) cutaneous drug reaction. In their study, maculopapular rash was present in 12.2% of the patients.[11] In another study from India (n = 200), fixed drug eruption was observed in 61 (30.5%) patients.[10] A study from Nagpur, Maharashtra, reported maculopapular rash, fixed drug eruption, and urticarial in 37.73%, 17.2%, and 14.56% patients, respectively.[18] Maculopapular rashes were often associated with pruritus and bilateral and symmetrical in nature beginning over the trunk. Unlike drug-induced rash, they were associated with prodromal symptoms and high-grade fever.

The severity of cutaneous adverse reactions may range from mild to life-threatening.[7] SJS and TEN are two serious types of cutaneous adverse drug reactions. The rates of these reactions among patients with cutaneous drug reactions range from 4% to 25%. In a study from eastern India, these two reactions were seen in 24.50% of the cases.[16] In another study, SJS and TEN were seen in 25% of the cases of cutaneous drug reactions.[13] In a study, from a hospital in south India, SJS was present in 19.5% of the cases.[12] In another study involving 200 patients with cutaneous drug reactions, 6 (3%) cases had SJS, and 2 (1%) patients had TEN.[10] In our study, SJS and TEN were present in 13.8% of the cases.

The rate of cutaneous reactions differs from drug to drug and for specific drugs, it may be more than 10%. Similar, to a study by Sharma and colleagues,[9] we also observed antimicrobials to be the most commonly implicated agents in the development of cutaneous drug reactions. Other studies also support this observation.[81218] Beta-lactam antibiotics are among the most commonly associated medicines with cutaneous adverse reactions.[8] In our study also, it was the most common class of agents responsible for cutaneous drug reactions. Among them, aminopenicillins, particularly amoxicillin-containing agents are the most common offending drugs.[1] Our observations are in line with this.

Other common agents associated with cutaneous drug reactions include NSAIDs and antiepileptic agents.[9121318] In our study, NSAIDs were the second most common category after antimicrobial agents to be associated with cutaneous drug reactions.

Sulfa drugs are another important category of medications implicated in such reactions. In one study, cotrimoxazole (22.2%) and dapsone (17.7%) were the most common drugs associated with cutaneous adverse drug reactions.[11] Cotrimoxazole was associated with cutaneous reaction in 26 (13%) in another study.[10] In our study, there was only one case of cutaneous drug reaction with dapsone. Anticonvulsants have been reported to contribute a large percentage to the life-threatening TEN and SJS.[9]

Some studies have used the World Health Organization-Uppsala Monitoring Centre (WHO-UMC) criteria for causality assessment.[1617] In a study, out of 231 cases, 3 (1.3%) were classified as “Certain,” 68 (29.4%) as “Probable,” and 160 (69.3%) cases as “Possible.”[17] In another study, certain and probable cases were 18.9% and 41.5%, respectively.[16] In our study, according to the Naranjo scale for causality assessment, maximum reactions were “probable” association and some had “possible” relation with the drug.

Single-center study and a small number of patients are the limitations of this study. Larger, multi-centric studies are needed to confirm our observations.

Conclusion

Fixed drug eruption was the commonest presentation of cutaneous adverse drug reactions. Fever was associated with skin reactions in almost one-third of the patients. Antimicrobial agents are most commonly associated with it. According to the Naranjo scale, the majority of the cutaneous drug reactions have a “probable” association with the offending drugs.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.