Diabetes mellitus (DM) is a group of metabolic disorders characterised by chronic hyperglycaemia resulting from defects in insulin secretion, insulin action, or both. It has become one of the most important non-communicable diseases of the twenty-first century. The International Diabetes Federation estimated that 463 million adults were living with diabetes worldwide in 2019, and projected that this number would approach 700 million by 2045 [1]. India contributes a very large share of this burden and is often described as the diabetes capital of the world. The ICMR–INDIAB study, which surveyed 15 states of the country, documented an overall diabetes prevalence of 7.3% and showed marked variation between states and between urban and rural populations [2]. Earlier reviews of the Indian scenario had already drawn attention to the rising prevalence of type 2 diabetes, its occurrence at a younger age, and the early onset of complications in Indians [3–5]. Although diabetes is traditionally viewed through the lens of its macrovascular and microvascular complications, the skin is one of the earliest and most accessible organs to show its effects. Skin changes are reported in a substantial proportion of patients, and published prevalence figures vary widely depending on the population studied, the setting and the methodology [6–8]. Cutaneous signs may be the presenting feature that leads to the diagnosis of diabetes, may develop during the course of the disease, or may arise as a consequence of treatment [9,10]. Because the skin can be examined without any special equipment, it offers a simple window through which clinicians can assess metabolic status. The pathogenesis of skin involvement in diabetes is multifactorial. Persistent hyperglycaemia leads to the formation of advanced glycation end-products that alter collagen structure and impair dermal elasticity. Microangiopathy and macroangiopathy reduce cutaneous perfusion, while autonomic and sensory neuropathy cause anhidrosis, dry skin and loss of protective sensation. Hyperglycaemia also impairs neutrophil chemotaxis, phagocytosis and cell-mediated immunity, thereby predisposing to bacterial and fungal infections, and it delays wound healing. In addition, hyperinsulinaemia and insulin resistance stimulate keratinocyte and fibroblast proliferation through insulin-like growth factor receptors, which explains the association of acanthosis nigricans and acrochordons with insulin-resistant states [11–13]. Cutaneous manifestations of diabetes are conventionally grouped into four categories: (i) skin conditions with a strong association with diabetes, such as acanthosis nigricans, diabetic dermopathy, necrobiosis lipoidica, bullosis diabeticorum, scleredema diabeticorum, eruptive xanthomas and granuloma annulare; (ii) cutaneous infections, both bacterial and fungal; (iii) skin conditions that arise as complications of diabetes, such as diabetic foot ulcers, neuropathic changes and xerosis; and (iv) cutaneous reactions to antidiabetic treatment, including injection-site lipohypertrophy, lipoatrophy and drug eruptions. Several of these have practical value as markers. Acanthosis nigricans and multiple skin tags are recognised cutaneous markers of insulin resistance and of pre-diabetic states [14–16], whereas diabetic dermopathy has been regarded as a sign that may point towards underlying microvascular disease [17,18]. The pattern of skin disease in diabetes is not uniform across the world. Climate, humidity, personal hygiene, footwear and barefoot walking, delayed presentation, self-medication with topical corticosteroids, and differences in ethnicity and body habitus all influence the spectrum of lesions that dermatologists encounter. Hospital-based Indian series have documented a high frequency of cutaneous disorders among diabetic patients, with infections and xerosis figuring prominently [19,20], and a study of 500 patients with type 1 diabetes from north India described a wide spectrum of changes in a south Asian population [21]. Studies from other regions of Asia and elsewhere report broadly similar but not identical patterns [22–26]. Local data are therefore valuable in guiding screening and counselling. Despite the high prevalence of skin disease in diabetic patients, systematic dermatological examination is often neglected in routine diabetes care, and many patients are never referred to a dermatologist until complications set in. A better understanding of the cutaneous profile of diabetic patients, and of its relation to disease duration, glycaemic control and systemic complications, may help physicians and dermatologists in early detection and timely intervention. With this background, the present study was undertaken to describe the spectrum of cutaneous manifestations in patients with diabetes mellitus attending a tertiary care teaching hospital and to analyse the factors associated with them. AIM AND OBJECTIVES Aim: To study the cutaneous manifestations in patients with diabetes mellitus. Objectives: • To determine the prevalence and pattern of cutaneous manifestations in patients with diabetes mellitus. • To correlate the cutaneous manifestations with age, sex, type of diabetes and duration of diabetes. • To assess the relationship of cutaneous manifestations with glycaemic control (HbA1c) and with microvascular complications (neuropathy, retinopathy and nephropathy).

Study design and setting This was a hospital-based, cross-sectional, observational study conducted in the Department of Dermatology of a tertiary care teaching hospital, over a period of 12 months. The study was started after obtaining approval from the Institutional Ethics Committee and written informed consent was obtained from every participant in a language they understood. Study population and sample size Consecutive patients with a confirmed diagnosis of diabetes mellitus attending the outpatient departments or admitted to the medical wards were screened for eligibility. Diabetes was diagnosed as per the American Diabetes Association criteria, that is, fasting plasma glucose ≥126 mg/dL, 2-hour plasma glucose ≥200 mg/dL during an oral glucose tolerance test, HbA1c ≥6.5%, or a random plasma glucose ≥200 mg/dL in a patient with classic symptoms, or the use of antidiabetic medication [27]. The sample size was calculated using the formula n = Z²pq/d². Taking an expected prevalence (p) of cutaneous manifestations of 75%, based on published Indian studies [19,20], a 95% confidence level (Z = 1.96) and an absolute precision (d) of 6%, the required sample size was about 200 patients, and this number was enrolled. Inclusion and exclusion criteria Inclusion criteria: patients aged 18 years or older with type 1 or type 2 diabetes mellitus of any duration who gave written informed consent. Exclusion criteria: gestational diabetes; patients receiving long-term systemic corticosteroids or other immunosuppressive drugs; patients with HIV infection, internal malignancy or chronic liver disease; pregnant women; and patients who declined consent. Clinical evaluation A pre-designed proforma was used to record demographic details, type and duration of diabetes, family history, treatment (oral hypoglycaemic agents, insulin or both), history of skin complaints and their temporal relationship to the diagnosis of diabetes, and any topical or systemic treatment taken. Height and weight were measured and body mass index (BMI) was calculated; overweight and obesity were classified using the Asian-Indian cut-offs (23.0–24.9 kg/m² as overweight and ≥25 kg/m² as obese). A complete general physical examination was followed by a detailed dermatological examination of the whole skin, hair, nails, oral cavity, genitalia and flexures under good natural light. The feet were examined for ulcers, callosities, deformities, nail changes and interdigital maceration, and peripheral neuropathy was screened using a 10-g monofilament and a 128-Hz tuning fork. Investigations Fasting and post-prandial plasma glucose and HbA1c were measured in all patients. Glycaemic control was classified as good (HbA1c <7.0%), fair (7.0–8.9%) and poor (≥9.0%). Where clinically indicated, potassium hydroxide (KOH) mounts of skin scrapings, nail clippings and hair, Gram stain and pus culture, Wood’s lamp examination and a Tzanck smear were performed. A punch biopsy was taken after consent in lesions of uncertain diagnosis, such as suspected necrobiosis lipoidica, granuloma annulare or scleredema. Serum creatinine and urine albumin were assessed, and fundus examination was carried out by an ophthalmologist. Operational definitions Neuropathy was defined as the loss of monofilament or vibration sense, or the presence of typical symptoms with an abnormal examination. Retinopathy was defined as any diabetic retinal change on dilated fundus examination. Nephropathy was defined as persistent albuminuria (urine albumin-to-creatinine ratio ≥30 mg/g) or an estimated glomerular filtration rate below 60 mL/min/1.73 m². A patient was considered to have a cutaneous manifestation if one or more dermatoses were diagnosed on examination; patients could have more than one diagnosis, so the percentages for individual conditions add up to more than 100%. Statistical analysis Data were entered in Microsoft Excel and analysed with IBM SPSS Statistics, version 26.0. Categorical variables are shown as numbers and percentages and continuous variables as mean ± standard deviation. Associations between categorical variables were tested with the chi-square test (chi-square test for trend for ordered categories) or Fisher’s exact test when expected cell counts were below five, and continuous variables were compared with the independent t-test. A p value below 0.05 was considered statistically significant.

4.1 Demographic and clinical profile A total of 200 patients with diabetes mellitus were included. The age of the patients ranged from 24 to 84 years with a mean of 54.6 ± 12.3 years, and the largest group (64 patients, 32.0%) belonged to the age group of 51–60 years. There were 112 men (56.0%) and 88 women (44.0%), giving a male-to-female ratio of 1.27:1. A little more than half of the patients (108; 54.0%) lived in urban areas and 96 (48.0%) gave a family history of diabetes (Table 1). Table 1: Socio-demographic profile of the study population (n = 200) Variable Category Number Percentage (%) Age group (years) 20–30 6 3.0 31–40 22 11.0 41–50 46 23.0 51–60 64 32.0 61–70 44 22.0 >70 18 9.0 Sex Male 112 56.0 Female 88 44.0 Residence Urban 108 54.0 Rural 92 46.0 Family history of diabetes Present 96 48.0 Absent 104 52.0 Mean age 54.6 ± 12.3 years (range 24–84 years); male-to-female ratio 1.27:1. Type 2 diabetes was present in 186 patients (93.0%) and type 1 diabetes in 14 patients (7.0%). The duration of diabetes was 5–10 years in the largest group (74 patients, 37.0%), and 27 patients (13.5%) had diabetes for more than 15 years. The mean HbA1c was 8.2 ± 1.9%; good glycaemic control was seen in only 52 patients (26.0%), while 64 (32.0%) had poor control. Nearly half of the patients (96; 48.0%) were obese by Asian-Indian criteria. Oral hypoglycaemic agents alone were being taken by 128 patients (64.0%), insulin alone by 38 (19.0%), and both by 34 (17.0%) (Table 2). Neuropathy was found in 78 patients (39.0%), retinopathy in 62 (31.0%) and nephropathy in 44 (22.0%). Table 2: Diabetes-related profile of the study population (n = 200) Variable Category Number Percentage (%) Type of diabetes Type 2 186 93.0 Type 1 14 7.0 Duration (years) <5 58 29.0 5–10 74 37.0 11–15 41 20.5 >15 27 13.5 HbA1c (%) <7.0 (good) 52 26.0 7.0–8.9 (fair) 84 42.0 ≥9.0 (poor) 64 32.0 BMI (kg/m²) <18.5 8 4.0 18.5–22.9 52 26.0 23.0–24.9 44 22.0 ≥25.0 96 48.0 Treatment Oral agents only 128 64.0 Insulin only 38 19.0 Insulin + oral agents 34 17.0 Mean HbA1c 8.2 ± 1.9%. 4.2 Overall prevalence and pattern of cutaneous manifestations Cutaneous manifestations were present in 164 of the 200 patients, an overall prevalence of 82.0%, whereas 36 patients (18.0%) had no skin disease. In 58 of the 164 affected patients (35.4%) the skin complaint had been noticed before or at the time diabetes was diagnosed. Xerosis and pruritus together were the most frequent group (96 patients, 48.0%), followed by diabetes-associated dermatoses (92 patients, 46.0%) and cutaneous infections (84 patients, 42.0%) (Table 3). Patients could have more than one condition; the 164 affected patients had a total of 477 diagnoses, or 2.9 per patient, and 74 patients (37.0%) had three or more (Table 5). Table 3: Distribution of cutaneous manifestations by category (n = 200) Category Patients (n) Percentage (%) Xerosis and/or pruritus 96 48.0 Diabetes-associated dermatoses 92 46.0 Cutaneous infections 84 42.0 Foot-related changes (ulcers, callosities) 30 15.0 Other dermatoses (lichen planus, psoriasis, vitiligo, non-fungal nail changes) 30 15.0 Treatment-related changes 20 10.0 Patients with any cutaneous manifestation 164 82.0 Patients without cutaneous manifestation 36 18.0 Patients may appear in more than one category. Among the infections, fungal infections were the most common and were seen in 62 patients (31.0%). These comprised dermatophytosis in 34 (17.0%), cutaneous candidiasis in 22 (11.0%), onychomycosis in 18 (9.0%) and pityriasis versicolor in 6 (3.0%). Bacterial infections were found in 24 patients (12.0%), mainly furuncles and carbuncles (6.0%), cellulitis (3.0%) and erythrasma (2.5%), and viral infections in 9 patients (4.5%), the commonest being herpes zoster (2.5%). Among the diabetes-associated dermatoses, acanthosis nigricans (23.0%), skin tags (19.0%) and diabetic dermopathy (15.0%) were the commonest; necrobiosis lipoidica (3.0%), bullosis diabeticorum (1.5%), scleredema diabeticorum (2.0%), eruptive xanthomas (2.0%) and granuloma annulare (2.5%) were less frequent (Table 4, Figure 1). Table 4: Frequency of individual cutaneous manifestations (n = 200) Cutaneous manifestation Number Percentage (%) A. Cutaneous infections Dermatophytosis 34 17.0 Cutaneous candidiasis 22 11.0 Onychomycosis 18 9.0 Pityriasis versicolor 6 3.0 Furuncle / carbuncle 12 6.0 Cellulitis 6 3.0 Erythrasma 5 2.5 Abscess 3 1.5 Impetigo 2 1.0 Herpes zoster 5 2.5 Herpes simplex 3 1.5 Viral warts 2 1.0 B. Diabetes-associated dermatoses Acanthosis nigricans 46 23.0 Skin tags (acrochordons) 38 19.0 Diabetic dermopathy 30 15.0 Diabetic thick skin / digital sclerosis 10 5.0 Necrobiosis lipoidica 6 3.0 Granuloma annulare 5 2.5 Scleredema diabeticorum 4 2.0 Eruptive xanthomas 4 2.0 Bullosis diabeticorum 3 1.5 C. Xerosis, pruritus and foot-related changes Xerosis 72 36.0 Pruritus (generalised or localised) 52 26.0 Callosities / plantar keratoderma 20 10.0 Diabetic foot ulcer 16 8.0 D. Treatment-related changes Insulin injection-site lipohypertrophy 14 7.0 Lipoatrophy 3 1.5 Cutaneous drug eruption 4 2.0 E. Other dermatoses Non-fungal nail changes 14 7.0 Vitiligo 8 4.0 Psoriasis 6 3.0 Lichen planus 4 2.0 Percentages are calculated on the total of 200 patients. Lipohypertrophy occurred in 14 of 72 insulin users (19.4%). Table 5: Number of cutaneous diagnoses per patient (n = 200) Number of diagnoses Patients (n) Percentage (%) None 36 18.0 One 38 19.0 Two 52 26.0 Three 36 18.0 Four or more 38 19.0 4.3 Cutaneous manifestations and duration of diabetes The proportion of patients with cutaneous manifestations rose steadily with the duration of diabetes, from 65.5% among those with diabetes for less than five years to 96.3% among those with a duration of more than 15 years (chi-square test for trend, p < 0.001) (Table 6). Acanthosis nigricans and skin tags were seen at all durations, whereas necrobiosis lipoidica, scleredema, bullosis diabeticorum and diabetic foot ulcers were found only in patients with diabetes of more than 10 years’ duration. Table 6: Cutaneous manifestations according to duration of diabetes Duration (years) Patients (n) With manifestations n (%) Without manifestations n (%) <5 58 38 (65.5) 20 (34.5) 5–10 74 62 (83.8) 12 (16.2) 11–15 41 38 (92.7) 3 (7.3) >15 27 26 (96.3) 1 (3.7) Total 200 164 (82.0) 36 (18.0) Chi-square test for trend, p < 0.001. 4.4 Cutaneous manifestations and glycaemic control Glycaemic control showed a strong relationship with the skin. Any cutaneous manifestation was present in 57.7% of patients with good control (HbA1c <7%), 85.7% of those with fair control and 96.9% of those with poor control (p < 0.001). The gradient was even steeper for infections, which affected 19.2%, 38.1% and 65.6% of the three groups respectively (p < 0.001) (Table 7, Figure 2). Patients with cutaneous manifestations had a higher mean HbA1c than those without (8.5 ± 1.8% versus 6.9 ± 1.2%; p < 0.001). Table 7: Cutaneous manifestations and infections according to glycaemic control HbA1c (%) Patients (n) Any manifestation n (%) Cutaneous infection n (%) <7.0 52 30 (57.7) 10 (19.2) 7.0–8.9 84 72 (85.7) 32 (38.1) ≥9.0 64 62 (96.9) 42 (65.6) Total 200 164 (82.0) 84 (42.0) Chi-square test for trend, p < 0.001 for both outcomes. .5 Cutaneous manifestations and microvascular complications Cutaneous manifestations were significantly more common in patients with microvascular complications. They were present in 97.4% of patients with neuropathy compared with 72.1% of those without; in 96.8% of patients with retinopathy compared with 75.4% of those without; and in 97.7% of patients with nephropathy compared with 77.6% of those without (p < 0.001 for each) (Table 8). Of the 30 patients with diabetic dermopathy, 21 (70.0%) had at least one microvascular complication. All 16 patients with diabetic foot ulcers had peripheral neuropathy. Table 8: Cutaneous manifestations in relation to microvascular complications Complication Status Patients (n) With manifestations n (%) p value Neuropathy Present 78 76 (97.4) <0.001 Absent 122 88 (72.1) Retinopathy Present 62 60 (96.8) <0.001 Absent 138 104 (75.4) Nephropathy Present 44 43 (97.7) <0.001 Absent 156 121 (77.6) Chi-square test. 4.6 Sex and type of diabetes The overall prevalence of cutaneous manifestations did not differ between men (80.4%) and women (84.1%). However, skin tags (27.3% versus 12.5%; p = 0.008) and cutaneous candidiasis (15.9% versus 7.1%; p = 0.049) were more frequent in women, whereas diabetic dermopathy was more frequent in men (19.6% versus 9.1%; p = 0.038). Dermatophytosis and diabetic foot ulcers were more common in men, and acanthosis nigricans and xerosis in women, but these differences were not statistically significant (Table 9). The prevalence of cutaneous manifestations was similar in type 1 and type 2 diabetes (78.6% versus 82.3%; p = 0.72), but vitiligo was significantly more frequent in type 1 diabetes (21.4% versus 2.7%; p = 0.01) (Table 10). Table 9: Distribution of selected manifestations by sex Manifestation Male (n = 112) n (%) Female (n = 88) n (%) p value Any manifestation 90 (80.4) 74 (84.1) 0.49 Dermatophytosis 24 (21.4) 10 (11.4) 0.06 Cutaneous candidiasis 8 (7.1) 14 (15.9) 0.049 Acanthosis nigricans 20 (17.9) 26 (29.5) 0.051 Skin tags 14 (12.5) 24 (27.3) 0.008 Diabetic dermopathy 22 (19.6) 8 (9.1) 0.038 Xerosis 36 (32.1) 36 (40.9) 0.20 Diabetic foot ulcer 12 (10.7) 4 (4.5) 0.11 Chi-square test. Table 10: Cutaneous manifestations by type of diabetes Manifestation Type 1 (n = 14) n (%) Type 2 (n = 186) n (%) p value Any manifestation 11 (78.6) 153 (82.3) 0.72 Vitiligo 3 (21.4) 5 (2.7) 0.01 Fisher’s exact test.

Diabetes mellitus affects almost every organ, and the skin is among the earliest to bear witness. In this study of 200 patients, 82.0% had at least one cutaneous manifestation, and patients with skin disease had a greater number of conditions, longer-standing diabetes, poorer glycaemic control and more microvascular complications than those without. These findings reaffirm that the skin is a sensitive indicator of the metabolic and vascular status of the diabetic patient. The mean age of our patients was in the sixth decade, and the largest group was aged 51–60 years, which is in keeping with the age profile of type 2 diabetes reported in Indian hospital-based series. The slight male preponderance may reflect gender differences in health-seeking behaviour and access to a tertiary hospital rather than a true difference in prevalence, since community surveys have found broadly similar rates in men and women. Type 2 diabetes constituted more than nine-tenths of our sample, as expected for an adult population in which type 1 diabetes is relatively rare, and nearly half the patients were obese by Asian-Indian criteria, in line with the central role of adiposity and insulin resistance in the Indian diabetic phenotype [28]. The overall prevalence of 82.0% in our study is high but falls within the wide range published in the studies. Indian studies from tertiary centres have reported that a large majority of diabetic patients have at least one skin disorder and studies from Pakistan, Kuwait, Italy, Turkey and elsewhere have reported prevalences that vary from roughly one-third to more than four-fifths of patients [29]. This variation is explained largely by differences in the definition of a manifestation, the inclusion of common conditions such as xerosis and pruritus, the proportion of type 1 and type 2 patients, the duration and control of diabetes, and by whether the examination was done by a dermatologist. Because we included a careful, full-body dermatological examination by a dermatologist and counted common conditions such as xerosis, our figure is expected to lie at the upper end of the range. Reviews of the subject also note that studies in which dermatologists examine all patients report higher figures than those relying on physician-recorded complaints. It is also noteworthy that in more than one-third of the affected patients the skin condition had been noticed before or along with the diagnosis of diabetes. This supports the view that cutaneous signs, particularly acanthosis nigricans, skin tags, recurrent candidiasis, furunculosis and generalised pruritus, should prompt screening for diabetes. Infections formed the commonest group of dermatoses in our patients (42.0%), with fungal infections predominating (31.0%). This pattern has been reported in most Indian series, in which candidiasis, dermatophytosis and bacterial infections constitute a considerable proportion of the skin disease encountered in diabetics. The hot and humid climate of the country, occlusive clothing, poor foot hygiene, the widespread use of inadequate or inappropriate antifungal treatment, and the common practice of using topical corticosteroid–antifungal combinations contribute to the high rates of dermatophytosis seen in India. Lugo-Somolinos and Sanchez demonstrated that dermatophyte infections, especially of the feet and nails, are significantly more frequent in diabetic than in non-diabetic subjects [30], and the high frequency of onychomycosis (9.0%) and tinea pedis in our study is consistent with this. Cutaneous candidiasis was significantly more common in women (15.9% versus 7.1%). Candidal intertrigo, vulvovaginitis, balanitis and paronychia are favoured by glycosuria, obesity, moisture and impaired local immunity, and women, because of the vulvovaginal and submammary flexural involvement as well as occupational wet work, are particularly susceptible. Bacterial infections, namely furuncles, carbuncles, cellulitis and erythrasma, were found in 12.0% of our patients. Impaired neutrophil function, reduced chemotaxis and phagocytosis, and microvascular disease are held responsible for this susceptibility [31]. Herpes zoster was seen in 2.5% of patients, in keeping with the observation that diabetes is a risk factor for reactivation of the varicella-zoster virus. An important finding of this study was the steep increase in infections with worsening glycaemic control: 19.2%, 38.1% and 65.6% in patients with good, fair and poor control respectively. This dose–response relationship strongly suggests that hyperglycaemia is itself a major driver of infection in the skin, and that careful metabolic control is the single most effective way to prevent infection. Similar associations between poor control and infection have been described in both type 1 and type 2 diabetes. Acanthosis nigricans was seen in 23.0% of our patients and skin tags in 19.0%. Both are cutaneous markers of hyperinsulinaemia and insulin resistance, and they are closely linked to obesity. Hud and colleagues documented the strong relationship of acanthosis nigricans with obesity and with hyperinsulinaemia and Thappa found a significant association of skin tags with diabetes in an Indian population, advocating that patients with multiple acrochordons be screened for diabetes. A systematic review has since confirmed that acanthosis nigricans, skin tags and some other signs are valid skin markers of diabetes and pre-diabetes. In our study, nearly half of the patients were obese, which explains the frequency of these markers. Skin tags were significantly more common in women, which may be related to the higher proportion of obese women, to hormonal factors and to friction in the neck and axillary folds. Diabetic dermopathy (shin spots) was seen in 15.0% of patients and was significantly more frequent in men, a finding that has been described by other authors and is attributed to greater exposure to trauma of the shins. In our series, seven of every ten patients with dermopathy had at least one microvascular complication. This accords with the view that diabetic dermopathy is not a trivial finding but a marker of microangiopathy, and that its presence should lead to a search for retinopathy, nephropathy and neuropathy. The classical but uncommon diabetic dermatoses, namely necrobiosis lipoidica (3.0%), granuloma annulare (2.5%), scleredema diabeticorum (2.0%), eruptive xanthomas (2.0%) and bullosis diabeticorum (1.5%), were seen in a small minority of patients, in agreement with other reports [32,33]. Necrobiosis lipoidica, which predominantly affects the shins of women, was found in six patients in our series, all with a long duration of diabetes. Scleredema diabeticorum typically occurs in long-standing, obese, type 2 diabetics with poor control, and this matched our four patients. Eruptive xanthomas, which signal severe hypertriglyceridaemia in uncontrolled diabetes, were seen only in patients with markedly raised HbA1c, and bullosis diabeticorum was restricted to patients with long-standing neuropathy. Xerosis was the single commonest finding (36.0%), followed by pruritus (26.0%). Dry skin in diabetes is multifactorial: autonomic neuropathy reduces sweating, hyperglycaemia-induced osmotic diuresis causes dehydration and impaired stratum corneum hydration, and microangiopathy affects dermal nutrition [34]. Seité and colleagues emphasised that treatment of xerosis is important because dry, cracked skin, particularly over the feet, provides an easy portal of entry for infection and can progress to ulceration. Pruritus may be due to xerosis, candidiasis, neuropathy or associated renal disease, and it was much more frequent among patients with poor control in our series. Diabetic foot ulcers were present in 8.0% of patients, all of whom had peripheral neuropathy, and 10.0% had callosities. A foot ulcer is the culmination of neuropathy, deformity, repeated unrecognised trauma and ischaemia, and the lifetime risk of developing a foot ulcer in a person with diabetes has been estimated at about 19% to 34% [35]. Barefoot walking, poorly fitting footwear and delay in seeking care are particular problems in India and increase the risk of infection, which, when deep, requires early and aggressive management according to published guidelines [36]. A detailed examination of the feet at every visit is therefore a simple, cost-effective measure. Insulin-injection-site lipohypertrophy was seen in 14 of the 72 patients (19.4%) who were on insulin, and lipoatrophy in three. Lipohypertrophy results from repeated injections at the same site and can lead to erratic insulin absorption and unexplained glycaemic variability. Its frequency in our series stresses the need to teach and review injection technique and site rotation at every visit. Cutaneous drug eruptions, mostly due to oral hypoglycaemic agents, were seen in 2.0%, and the offending drug was changed in each case. The prevalence of cutaneous manifestations rose from 65.5% in patients with diabetes of less than five years’ duration to 96.3% in those with a duration of more than 15 years, a significant trend. Longer exposure to hyperglycaemia causes cumulative glycation of dermal proteins, vascular damage and neuropathy, and this explains why certain manifestations such as necrobiosis lipoidica, scleredema, bullosis diabeticorum and foot ulcers were confined to patients with long-standing disease. Comparable associations have been reported in Indian and international studies. The relationship with glycaemic control was equally strong. The prevalence of manifestations rose from 57.7% to 96.9% as HbA1c worsened, and the mean HbA1c of patients with skin disease was significantly higher than that of those without. Although this is a cross-sectional association and causality cannot be inferred, our findings support the idea that effective glycaemic control not only prevents classical micro- and macrovascular complications but also reduces the burden of skin disease. Patients with neuropathy, retinopathy and nephropathy had significantly more cutaneous disease, with over 96% having at least one skin manifestation in each group. Neuropathy is directly responsible for xerosis, callosities, neuropathic ulcers and bullosis, while microangiopathy contributes to dermopathy, necrobiosis lipoidica and impaired wound healing. Demirseren and colleagues, in their analysis of 750 patients, likewise found that skin diseases were significantly related to extracutaneous complications of diabetes. This shared pathogenesis is the reason why the skin may act as a mirror of internal disease, and why every patient with prominent skin findings should be screened for systemic complications. The overall prevalence of skin disease was similar in type 1 and type 2 diabetes in our small group of type 1 patients, and our numbers are insufficient for firm conclusions. Of interest, vitiligo was significantly more frequent in patients with type 1 diabetes (21.4% versus 2.7%), which is consistent with the autoimmune basis shared by the two conditions. A study of 500 south Asian patients with type 1 diabetes found vitiligo and other autoimmune and xerotic changes to be frequent, and an Indian study had earlier suggested that vitiligo is part of a systemic autoimmune process [37]. Pavlović and co-workers similarly reported a spectrum of changes in young patients with type 1 diabetes. Strengths and limitations The strengths of this study include a full-body dermatological examination of every patient by a dermatologist, objective measurement of glycaemic control using HbA1c, and the systematic assessment of microvascular complications. Its limitations include its single-centre, hospital-based design, which may overestimate the prevalence because patients attending a tertiary centre are likely to be sicker; the cross-sectional design, which cannot establish causality or the temporal sequence between skin lesions and diabetes; the lack of a non-diabetic control group; the small number of patients with type 1 diabetes; and the absence of lipid profile and insulin-resistance indices, which would have refined the interpretation of metabolic skin markers. Multicentre community-based studies with matched controls are needed to confirm these findings.

Cutaneous manifestations are extremely common in patients with diabetes mellitus, affecting more than four of every five patients in this study. Infections, xerosis, acanthosis nigricans, skin tags and diabetic dermopathy were the commonest conditions. The burden of skin disease increased significantly with longer duration of diabetes, poorer glycaemic control and the presence of neuropathy, retinopathy and nephropathy. Skin disease, particularly recurrent infections, may be the first indication of undiagnosed or poorly controlled diabetes, and markers such as acanthosis nigricans and skin tags can point to insulin resistance in an otherwise unsuspected patient.

1. Saeedi P, Petersohn I, Salpea P, Malanda B, Karuranga S, Unwin N, et al. Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: Results from the International Diabetes Federation Diabetes Atlas, 9th edition. Diabetes Res Clin Pract. 2019;157:107843. 2. Anjana RM, Deepa M, Pradeepa R, Mahanta J, Narain K, Das HK, et al. Prevalence of diabetes and prediabetes in 15 states of India: results from the ICMR-INDIAB population-based cross-sectional study. Lancet Diabetes Endocrinol. 2017;5(8):585-96. 3. Mohan V, Sandeep S, Deepa R, Shah B, Varghese C. Epidemiology of type 2 diabetes: Indian scenario. Indian J Med Res. 2007;125(3):217-30. 4. Unnikrishnan R, Anjana RM, Mohan V. Diabetes mellitus and its complications in India. Nat Rev Endocrinol. 2016;12(6):357-70. 5. Pradeepa R, Mohan V. Epidemiology of type 2 diabetes in India. Indian J Ophthalmol. 2021;69(11):2932-8. 6. Duff M, Demidova O, Blackburn S, Shubrook J. Cutaneous manifestations of diabetes mellitus. Clin Diabetes. 2015;33(1):40-8. 7. Lima AL, Illing T, Schliemann S, Elsner P. Cutaneous manifestations of diabetes mellitus: a review. Am J Clin Dermatol. 2017;18(4):541-53. 8. Mendes AL, Miot HA, Haddad Junior V. Diabetes mellitus and the skin. An Bras Dermatol. 2017;92(1):8-20. 9. Huntley AC. The cutaneous manifestations of diabetes mellitus. J Am Acad Dermatol. 1982;7(4):427-55. 10. Jelinek JE. Cutaneous manifestations of diabetes mellitus. Int J Dermatol. 1994;33(10):605-17. 11. Perez MI, Kohn SR. Cutaneous manifestations of diabetes mellitus. J Am Acad Dermatol. 1994;30(4):519-31. 12. Ferringer T, Miller F 3rd. Cutaneous manifestations of diabetes mellitus. Dermatol Clin. 2002;20(3):483-92. 13. Rosen J, Yosipovitch G. Skin manifestations of diabetes mellitus. In: Feingold KR, Anawalt B, Boyce A, et al., editors. Endotext [Internet]. South Dartmouth (MA): MDText.com, Inc.; 2018. 14. Hud JA Jr, Cohen JB, Wagner JM, Cruz PD Jr. Prevalence and significance of acanthosis nigricans in an adult obese population. Arch Dermatol. 1992;128(7):941-4. 15. Thappa DM. Skin tags as markers of diabetes mellitus: an epidemiological study in India. J Dermatol. 1995;22(10):729-31. 16. Bustan RS, Wasim D, Yderstraede KB, Bygum A. Specific skin signs as a cutaneous marker of diabetes mellitus and the prediabetic state – a systematic review. Dan Med J. 2017;64(1):A5316. 17. Morgan AJ, Schwartz RA. Diabetic dermopathy: a subtle sign with grave implications. J Am Acad Dermatol. 2008;58(3):447-51. 18. Demirseren DD, Emre S, Akoglu G, Arpacı D, Arman A, Metin A, et al. Relationship between skin diseases and extracutaneous complications of diabetes mellitus: clinical analysis of 750 patients. Am J Clin Dermatol. 2014;15(1):65-70. 19. Ragunatha S, Anitha B, Inamadar AC, Palit A, Eshwarappa M. Cutaneous disorders in 500 diabetic patients attending diabetic clinic. Indian J Dermatol. 2011;56(2):160-4. 20. Mahajan S, Koranne RV, Sharma SK. Cutaneous manifestation of diabetes mellitus. Indian J Dermatol Venereol Leprol. 2003;69(2):105-8. 21. Sawatkar GU, Kanwar AJ, Dogra S, Bhadada SK, Dayal D. Spectrum of cutaneous manifestations of type 1 diabetes mellitus in 500 south Asian patients. Br J Dermatol. 2014;171(6):1402-6. 22. Romano G, Moretti G, Di Benedetto A, Giofrè C, Di Cesare E, Russo G, et al. Skin lesions in diabetes mellitus: prevalence and clinical correlations. Diabetes Res Clin Pract. 1998;39(2):101-6. 23. Yosipovitch G, Hodak E, Vardi P, Shraga I, Karp M, Sprecher E, et al. The prevalence of cutaneous manifestations in IDDM patients and their association with diabetes risk factors and microvascular complications. Diabetes Care. 1998;21(4):506-9. 24. Al-Mutairi N, Zaki A, Sharma AK, Al-Sheltawy M, Nour-Eldin O. Cutaneous manifestations of diabetes mellitus. Med Princ Pract. 2006;15(6):427-30. 25. Pavlović MD, Milenković T, Dinić M, Misović M, Dakić D, Todorović S, et al. The prevalence of cutaneous manifestations in young patients with type 1 diabetes. Diabetes Care. 2007;30(8):1964-7. 26. Ahmed K, Muhammad Z, Qayum I. Prevalence of cutaneous manifestations of diabetes mellitus. J Ayub Med Coll Abbottabad. 2009;21(2):76-9. 27. American Diabetes Association. 2. Classification and diagnosis of diabetes: Standards of Medical Care in Diabetes—2021. Diabetes Care. 2021;44(Suppl 1):S15-33. 28. Misra A, Gopalan H, Jayawardena R, Hills AP, Soares M, Reza-Albarrán AA, et al. Diabetes in developing countries. J Diabetes. 2019;11(7):522-39. 29. Wahid Z, Kanjee A. Cutaneous manifestations of diabetes mellitus. J Pak Med Assoc. 1998;48(10):304-5. 30. Lugo-Somolinos A, Sanchez JL. Prevalence of dermatophytosis in patients with diabetes. J Am Acad Dermatol. 1992;26(3 Pt 1):408-10. 31. Sibbald RG, Landolt SJ, Toth D. Skin and diabetes. Endocrinol Metab Clin North Am. 1996;25(2):463-72. 32. Muller SA, Winkelmann RK. Necrobiosis lipoidica diabeticorum: a clinical and pathological investigation of 171 cases. Arch Dermatol. 1966;93(3):272-81. 33. Cole GW, Headley J, Skowsky R. Scleredema diabeticorum: a common and distinct cutaneous manifestation of diabetes mellitus. Diabetes Care. 1983;6(2):189-92. 34. Seité S, Khemis A, Rougier A, Ortonne JP. Importance of treatment of skin xerosis in diabetes. J Eur Acad Dermatol Venereol. 2011;25(5):607-9. 35. Armstrong DG, Boulton AJM, Bus SA. Diabetic foot ulcers and their recurrence. N Engl J Med. 2017;376(24):2367-75. 36. Lipsky BA, Berendt AR, Cornia PB, Pile JC, Peters EJ, Armstrong DG, et al. 2012 Infectious Diseases Society of America clinical practice guideline for the diagnosis and treatment of diabetic foot infections. Clin Infect Dis. 2012;54(12):e132-73. 37. Gopal KV, Rama Rao GR, Kumar YH, Appa Rao MV, Vasudev P, Srikant. Vitiligo: a part of a systemic autoimmune process. Indian J Dermatol Venereol Leprol. 2007;73(3):162-5.