Journal of Pakistan Association of Dermatologists . 2015; 25 (4):259-265.

Original Article Cutaneous manifestations associated with

Sajid Ali Mustafvi*, Muhammad Afzal**, Ather Mehmood**, Tahir Ahamd Munir***, Naser Rashid Dar¶, Shadab Ahmad Butt¶¶

* Department of Medicine, Rawal Institute of Health Sciences, Islamabad ** Department of Medicine, Rawal Institute of Health Sciences, Islamabad *** Department of Physiology, Rawal Institute of Health Sciences, Islamabad ¶ Department of Dermatology, CMH, Rawalpindi ¶¶ Department of Anatomy, AMC, Rawalpindi

Abstract Objective To study the frequency of cutaneous manifestations of obesity and their association with severity of disease in Pakistani population.

Methods 100 outdoor patients visiting Rawal Institute of Health Sciences, Islamabad were studied during a period from May 2014 to July 2015. BMI was calculated by measuring weight in kilogram divided by square in height in meter. Cutaneous manifestations in obese patients with class I (CI) 2 2 [BMI >30-34.9Kg/m ] and class II (CII) [BMI 35-39Kg/m ] disease were recorded.

2 Results The mean age was 41.85±8.61 years while the mean BMI was 33.51±2.51 kg/m . A significant difference for diabetes mellitus, striae and was seen between CI and CII groups (p<0.05). BMI showed significant positive correlation with DM ( r=0.280, p=0.005) and acne ( r=0.315, p=0.001) while diabetes mellitus showed with acanthosis nigricans ( r=0.373; p=0.000) and tags ( r=0.218, p=0.029). Acne showed with miliaria (r=0.210, p=0.036) and varicose vein with xanthomas ( r= 0.281, p=0.005).

Conclusion Skin tags, acanthosis nigricans and acne are seen more frequently in obesity.

Keywords Obesity, cutaneous manifestations.

Introduction osteoarthritis, obstructive sleep apnea, and depression, as well as, , endometrial and Obesity is a serious public health issue colonic cancer. It also showed strong relations contributing to the pathogenesis of with nonalcoholic fatty liver disease and cardiovascular, musculoskeletal and metabolic gallstones, diverticulitis, infertility, urinary disorders. Obesity may be present at any age but incontinence, anxiety and impaired social 1,2 most commonly develops in mid-life. Obesity interactions. has shown a well-established relation with conditions as coronary heart disease, type 2 In skin, obesity affects skin barrier functions, diabetes mellitus, , hyperlipidemia, sebaceous glands and sebum productions, sweat glands, lymphatics, structure and its Address for correspondence functions, micro- and macrocirculation and Dr. Tahir Ahmad Munir subcutaneous . Obese patients have larger Department of Physiology, folds and sweat more profusely after Rawal Institute of Health Science, Islamabad becoming overheated because of thick layers of Email: [email protected] sub coetaneous fat, thereby increasing both the

259 Journal of Pakistan Association of Dermatologists . 2015; 25 (4):259-265. frictional and moisture components. 2 Cutaneous hyperlipidemia, and hyperuricemia either manifestations associated with obesity include already diagnosed or under treatment. BMI was lymphedema, acanthosis nigricans, and effects calculated after measuring height in cm and of hyperandrogenesis, acne, miliaria, intertrigo, converting into meter square, weight in Kg, and plantar , skin infections and using the formula - kg/m 2. Height in centimeter varicose veins. 4,5 was measured by asking the patient to stand straight against the wall using weight-height The World Health Organization (WHO) and measuring machine. The patients were divided National Institute of Health (NIH) use body into two groups; class I (CI) [BMI 30-34.9 mass index (BMI) – an index of weight-for- Kg/m 2] and class II (CII) [BMI 35 to 39.9 height commonly used to classify underweight, Kg/m 2]. Patients were examined by a qualified overweight, and obesity. BMI is defined as dermatologist (NRD) for cutaneous weight in kilograms divided by square of height manifestations like skin complexion (dark or 2 2 in meters [Kg/m ]. A BMI of 18.5-24.9 Kg/m is fair), striae, skin tags (acrochordon), miliaria, 2 taken as normal, BMI 25-29.9 Kg/m overweight acanthosis nigricans, varicose veins, xanthoma 2 and BMI >30 Kg/m taken as obese. Obesity can and xanthelasma, plantar hyperkeratosis, and be further characterized as by BMI as class I intertrigo etc. (30-34.9 Kg/m2), class II (35-39.9 Kg/m 2) and class III (>40 Kg/m 2). 6 Results

Waist circumference, measured at the end of The mean age of patients of group CI (n=50) normal expiration at the level of iliac crests, is was 41.2±7.55 years while that of group CII another reasonable way of assessing obesity. (n=50) was 42.25±9.33 years ( p=0.397). The Circumference values that indicate a significant mean height showed a non-significant difference increased relative risk are more than 88cm for (p=0.123) between group CI (169.70±7.80 cm) women and more than 102cm for men. 7 and group CII (172.15±8.97 cm). A similar trend was seen when pulse rate ( p=0.147) and blood This study was undertaken to assess the pressure ( p=0.183) of the two groups were frequency of cutaneous manifestations compared, however, difference was significant associated with obesity and their correlation in terms of weight (89.67±8.40 Kg, with severity of disease. 108.64±23.23Kg; p=0.000) and BMI (31.18±0.91, 35.72±0.74; p=0.000) between the Methods groups.

A total of 100 obese patients visiting the Rawal Table 1 represents cutaneous manifestations of Institute of Health Sciences, Islamabad for their obesity. A non-significant greater percentage skin problems were selected in the study for a (OR=0.783; p=0.545) of females was noted in period from June 2014 to July 2015. After both (CI and CII) groups. A non-significant patient consent, demographic parameters like higher percentage for hypertension (OR=0.777; age, sex, pulse, blood pressure (using mercury p=0.347), hyperuricemia (OR=0.583, p=0.372), sphygmomanometer, taking average of the three hyperlipidemia (OR=0.706, p=0.405), family readings after completely relaxing the patient) history of obesity (OR=0.911; p=0.829), miliaria were recorded. The patients were also inquired (OR=0.599; p=0.260) and plantar hyperkeratosis regarding history of diabetes mellitus, (OR=0.545; p=0.147) was noticed in patients of

260 Journal of Pakistan Association of Dermatologists . 2015; 25 (4):259-265. group CI compared to those of group CII; while stretch marks ( r=0.174, p=0.083), skin tags dark skin complexion (OR=0.592; p= 0.211), (r=0.218, p=0.029), varicose veins ( r=0.114, varicose veins (OR=2.550; p=0.182) and p=0.261), xanthomas ( r=0.154, p=0.126), xanthomas/ xanthelasma (OR=1.278; p=0.539) miliaria ( r=059, p=0.559) and intertrigo were of non-significant greater percentage in (r=0.101, p=0.317), while a significant negative group CII compared to patients of group CI. correlation was seen with plantar hyperkeratosis (r= -0.016, p=0.873) and acne ( r= -0.032, A significant higher percentage for diabetes p=0.752). mellitus (OR=2.455; p=0.028), striae distensae (OR=2.279; p=0.044), and acanthosis nigricans Acne showed significant negative correlation (OR=3.857; p=0.001) was seen when patients of only with skin tags ( r= -0.217, p=0.030), and group CII and CI were compared; while acne positive correlation with BMI ( r=0.315, was found to be more statistically significant in p=0.001) and miliaria (r=0.210, p=0.036) while patients of group CI compared to group CII the varicose veins had significant positive (OR=0.259; p=0.001). The patients of group CI correlation with xanthomas ( r= 0.281, p=0.005). and CII showed same percentage for intertrigo Miliaria had significant negative correlation with (OR=1.000; p=1.000) acanthosis nigricans (r= -0.245, p=0.014).

Table 2 shows analysis of variance between and Discussion within the groups. Analysis of variance showed a significant difference for BMI ( p=0.000), acne Our results are in consistence with a number of (p=0.009), diabetes mellitus ( p=0.028), studies which showed association of obesity acanthosis nigricans ( p=0.002), skin tags with cutaneous manifestations.3,8,9,10 (p=0.016), stretch marks ( p=0.044), and blood pressure ( p=0.026), within and between the Among the skin manifestations skin tags (52%) groups. However, the skin complexion followed by striae distensae (48%) were found (p=0.215), miliaria (p=0.265), xanthoma/ more frequent (52%) than other cutaneous xanthelasma ( p=0.163), varicose veins manifestations. These results are in consistence (p=0.186) plantar hyperkeratosis ( p=0.150) and with Tamega et al.11 and Boza et al.12 who intertrigo ( p=0.334) were found to be non- showed a significant correlation ( p<0.05) with significant within and between the groups diabetes mellitus and dyslipidemia. Our result also showed a significant difference ( p=0.04) Table 3 showed correlations with variables. when skin tags were compared within and BMI showed a positive correlation with DM between the two study groups. Our results (r=0.280, p=0.005), acne ( r=0.315, p=0.001), showed a significant correlation ( p<0.05) with miliaria (r=0.119, p=0.238), and plantar diabetes mellitus. hyperkeratosis ( r=0.167, p=0.097), while negative correlation with skin tags (r= -0.332, Acanthosis nigricans was seen in 44% of the p=0.001), varicose vein (r= -0.132; p=0.190), selected patients. Other comorbid conditions like intertrigo ( r= -0.084, p=0.406), and xanthomas diabetes mellitus showed significant association (r= -0.120, p=0.236). with acanthosis ( p=0.000), and miliaria (p=0.014), however, it also showed a positive Diabetes mellitus showed positive correlation correlation with skin tags and striae distensae. A with acanthosis nigricans (r=0.373; p=0.000) significant difference (p=0.002) was seen for

261 Journal of Pakistan Association of Dermatologists . 2015; 25 (4):259-265.

Table 1 Different systemic and cutaneous manifestations of obesity in study population (n=100). Variable C1 CII OR P value CI 95% n= 50 (50%) n= 50 (50%) Sex Male 23 (46.0%) 20 (40%) 0.783 0.545 0.354 – 1.730 Female 27 (54.0%) 30 (60%) Skin complexion Dark 29 (58.0%) 35 (70%) 0.592 0.211 0.259 – 1.351 Fair 21 (42.0%) 15 (30%) Family H/O Obesity Yes 35 (70.0%) 34 (68%) 0.911 0.829 0.390 – 2.126 No 15 (30.0%) 16 (32%) Hypertension Yes 32 (64.0%) 29 (58%) 0.777 0.347 0.347 – 1.738 No 18 (36.0%) 21 (42%) Diabetes mellitus Yes 21 (42%) 32 (64%) 2.455 0.028 1.097 – 5.494 No 29 (58%) 18 (36%) Hyperuricemia Yes 08 (16%) 06 (12%) 0.583 0.372 0.177 – 1.925 No 42 (84%) 44 (88%) Hyperlipidemia Yes 20 (40%) 16 (32%) 0.706 0.405 0.311 – 1.603 No 30 (60%) 34 (68%) Acne Yes 30 (60%) 14 (28%) 0.259 0.001 0.112 – 0.599 No 20 (40%) 36 (72%) Acanthosis nigricans Yes 14 (28%) 30 (60%) 3.857 0.001 1.670 – 8.911 No 36 (72%) 20 (40%) Striae distensae Yes 23 (46%) 33 (66%) 2.279 0.044 1.017 – 5.108 No 27 (54%) 17 (34%) Acrochordon (skin tag) Yes 20 (40%) 32 (64%) 2.667 0.016 1.188 – 5.985 No 30 (60%) 18 (36%) Miliaria Yes 25 (50%) 20 (40%) 0.667 0.315 0.302– 1.472 No 25 (50%) 30 (60%) Varicose veins Yes 03 (06%) 07 (14%) 2.550 0.182 0.620 – 10.492 No 47 (94%) 43 (86%) Xanthoma/xanthelasma Yes 18 (36%) 21 (42%) 1.287 0.539 0.575 – 2.881 No 32 (64%) 29 (58%) Plantar hyperkeratosis Yes 19 (38%) 18 (36%) 0.918 0.836 0.407 – 2.067 No 31 (62%) 32 (64%) Intertrigo Yes 10 (20%) 10 (20%) 1.000 1.000 0.375 – 2.664 No 40 (80%) 40 (80%) acanthosis within and between the groups. These association ( p<0.05) of acanthosis nigricans with results are in association with Phiske 10 and Al obesity. A positive correlation was also noticed Mutari 13 who showed statistically significant with skin tag, and stretch marks.

262 Journal of Pakistan Association of Dermatologists . 2015; 25 (4):259-265.

Table 2 Correlation among different cutaneous manifestations (n=100). Family Plant. h/o Skin Acan. Varicos Xantho- Hyperkera Intertrig BMI obesity DM Acne Striae tag Miliaria nigrican e veins mas -tosis o BMI Pearson correlation 1 .002 .280 .315 -.157 -.332 .119 -.332 -.132 -.120 .167 -.084 Sig. (2-tailed) .987 .005 .001 .120 .001 .238 .001 .190 .236 .097 .406 Family Pearson correlation .002 1 .105 -.029 .059 .178 .177 -.073 .223 .073 -.021 .137 h/o Sig. (2-tailed) .987 .297 .773 .558 .076 .078 .471 .025 .471 .835 .174 obesity Diabetes Pearson correlation -.280 .105 1 -.032 .174 .218 .059 .373 .114 .154 -.016 .108 mellitus Sig. (2-tailed) .005 .297 .752 .083 .029 .559 .000 .261 .126 .873 .287 Acne Pearson correlation .315 -.029 -.032 1 -.125 -.217 .210 -.183 .114 .032 .038 .099 Sig. (2-tailed) .001 .773 .752 .214 .030 .036 .068 .257 .751 .707 .329 Striae Pearson correlation -.157 .059 .174 -.125 1 .156 .096 .119 .161 .168 .113 -.098 distensae Sig. (2-tailed) .120 .558 .083 .214 .120 .341 .239 .109 .095 .261 .331 Skin tag Pearson correlation -.332 .178 .218 -.217 .156 1 .002 .107 .053 .165 -.114 .098 Sig. (2-tailed) .001 .076 .029 .030 .120 .986 .291 .598 .101 .257 .334 Sweat Pearson correlation .119 .177 .059 .210 .096 .002 1 -.245 .128 .184 .000 -.055 rash Sig. (2-tailed) .238 .078 .559 .036 .341 .986 .014 .206 .068 .996 .587 Acan. Pearson correlation -.332 -.073 .373 -.183 .119 .107 -.245 1 .047 -.110 -.171 .099 nigrican Sig. (2-tailed) .001 .471 .000 .068 .239 .291 .014 .641 .277 .089 .329 Varicose Pearson correlation -.132 .223 .114 .114 .161 .053 .128 .047 1 .281 -.090 .000 veins Sig. (2-tailed) .190 .025 .261 .257 .109 .598 .206 .641 .005 .375 1.000 Xantho- Pearson correlation -.120 .073 .154 .032 .168 .165 .184 -.110 .281 1 .091 .034 mas Sig. (2-tailed) .236 .471 .126 .751 .095 .101 .068 .277 .005 .367 .735 Plantar Pearson correlation .167 -.021 -.016 .038 .113 -.114 .000 -.171 -.090 .091 1 -.091 hyper- Sig. (2-tailed) .097 .835 .873 .707 .261 .257 .996 .089 .375 .367 .368 Intertrigo Pearson correlation -.084 .137 .108 .099 -.098 .098 -.055 .099 .000 .034 -.091 1 Sig. (2-tailed) .406 .174 .287 .329 .331 .334 .587 .329 1.000 .735 .368 Acan. nigrican = Acanthosis nigricans, DM=diabetes mellitus,

263 Journal of Pakistan Association of Dermatologists . 2015; 25 (4):259-265.

Table 3 Analysis of variance for variables between and within groups. Sum of squares Df Mean square F Sig. Between groups 1.000 1 1.000 5.115 .026 Blood pressure Within groups 19.160 98 .196 Between groups .360 1 .360 1.556 .215 Skin complexion Within groups 22.680 98 .231 Between groups 2.250 1 2.250 9.906 .002 Acne Within groups 22.260 98 .227 Between groups 1.000 1 1.000 4.146 .044 Striae distensae Within groups 23.640 98 .241 Between groups 1.440 1 1.440 6.000 .016 Skin tags Within groups 23.520 98 .240 Sweat Between groups .250 1 .250 1.259 .265 Rash Within groups 19.460 98 .199 Between groups 2.250 1 2.250 9.906 .002 Acanthosis nigricans Within groups 22.260 98 .227 Between groups .160 1 .160 1.774 .186 Varicose veins Within groups 8.840 98 .090 Between groups .360 1 .360 1.973 .163 Xanthomas Within groups 17.880 98 .182 Between groups .490 1 .490 2.104 .150 Plantar hyperkeratosis Within groups 22.820 98 .233 Between groups 4.000 1 4.000 .942 .334 Intertrigo Within groups 416.000 98 4.245 Between groups 542.890 1 542.890 648.029 .000 BMI Within groups 82.100 98 .838 Between groups .010 1 .010 .046 .831 Family h/o obesity Within groups 21.380 98 .218 Between groups 1.210 1 1.210 5.003 .028 Diabetes mellitus Within groups 23.700 98 .242

The percentage of acne and that of acanthosis varicose veins (10%) was much less than that of nigricans was the same, and showed significant xanthomas (44%) but the difference within and association with BMI ( p=0.001), miliaria between the groups was found to non- (p=0.036), xanthomas, varicose vein and plantar significant. The intertrigo and plantar hyperkeratosis. It also showed a significant hyperkeratosis did not establish correlation with difference ( p=0.001) when compared within and obesity and BMI. between the groups. These results are in accordance with Phiske,10 Tamega et al.11 who Conclusion showed association of acne with obesity and diabetes mellitus. Obesity, a major public health problem, affecting nearly all body systems and all Our results showed significant association disciplines of medicine may encounter obesity- (p=0.005) of xanthomas with that of varicose related complications. Skin tags, acanthosis veins, however, the patient percentage for nigricans and acne are more frequently

264 Journal of Pakistan Association of Dermatologists . 2015; 25 (4):259-265. associated with obesity. The cases of varicose board of the National Foundation. vein with venous ulcers, lymphedema and J Am Acad Dermatol. 2010; 63 :1058-69. 7. WHO study group. Defining the problem intertrigo may pose difficulty in treatment. of over-weight and obesity in obesity: preventing and managing the global References epidemic report of a WHO consultation on obesity. Geneva: WHO. 1997. pp7-16 8. Iwao S, Jwao N, Muller DC et al . Does 1. National Task force on the preventions and waist circumference and to the predictive treatment of obesity: overweight, obesity power of the body mass inside for coronary and health risk. Arch Intern Med. risk? Obes Res. 2001; 9:685-95. 2010; 160 :808-904. 9. Garcia-Hidalgo L. Dermatological 2. Yosipritch G, Devore A, Dawn A. Obesity complications of obesity. Am J Clin and the skins: skin physiology and skin Dermatol . 2002; 3:497-506. manifestations of obesity. J Am Acad 10. Phiske MM. An approach to acanthosis Dermatol . 2007: 56 :901-16. nigricans. Indian Dermatol Online J. 3. Guida B, Nino M, Perrino NR. The impact 2014; 5:239-49. of obesity on skin disease and epidermal 11. Tamega AA, Aranha A, Guitoka M. permeability barrier status. J Eur Acad Association between skin tag and insulin Dermatol Venereol . 2009; 24:191–5. resistance. An Bras Dermatol . 2010; 85 :25- 4. Verma SB, Wollina U. Acne keloidalis 31. nuchae: another cutaneous symptom of 12. Boza J, Trindade E, Peruzzo J. Skin metabolic syndrome, truncal obesity, and manifestation of obesity; a comparative impending/overt diabetes mellitus? Am J study. J Eur Acad Dermatol Venereol . Clin Dermatol . 2010; 11 :433-6. 2012; 26 :1220-3. 5. van Gent WB, Wilschut ED, Wittens C. 13. Al Mutiari N. Associated cutaneous Management of venous disease. BMJ . diseases in obese adult patients; a 2010; 341 :45-65 prospective study from a skin referral care 6. Bremmer S, van Voorhees AS, Hsu S. centre. Med Princ Pract . 2011; 20 :248-52. Obesity and psoriasis: from the medical

265