A randomized controlled trial of topical benzoyl peroxide 2.5% gel with a low glycemic load diet versus topical benzoyl peroxide 2.5% gel with a normal diet in acne (grades 1-3)

Introduction

Acne is a very common skin disorder affecting individuals of all age groups.^[1],[2] Populations that consume low glycemic load diets, such as the Kitavan islanders of Guinea, Ache hunters, Inuits and rural Brazilians, tend to be acne-free.^[3] An increase in the incidence of acne was observed in Inuits, Okinawa islanders and Chinese after they started following a western diet.^[3] These observations suggest that lifestyle factors like diet may have a role in acne pathogenesis. A high glycemic load diet causes hyperinsulinemia, high androgen bioavailability, increased insulin-like growth factor-1 and acne.^[4],[5] Increased insulin-like growth factor-1 further increases dihydrotestosterone and dehydroepiandrosterone sulfate, which increase sebum secretion and sebocyte proliferation.^[4],[5] Increased insulin levels in blood elevate plasma levels of epidermal growth factor, transforming growth factor-β and free nonesterified fatty acids, thereby causing sebaceous gland inflammation and acne.^[6] We conducted this study to determine the efficacy of a low glycemic load diet in the management of acne vulgaris.

Methods

The study was designed as a randomized controlled investigator-blinded trial. After obtaining institutional ethical committee clearance, a total of 84 patients (male and female, age 14–29 years) with acne of grades 1, 2 and 3 were included randomly from among dermatology outpatients in a tertiary care centre between September 2014 and May 2016. Patients already on topical medication, those on oral retinoid in the last 6 months or oral antibiotics in the last 15 days and pregnant women were excluded.

Patients were randomly assigned to two groups of 42 participants each, using the chit in the box method. Allotment was done by a registrar not involved in the rest of the study. Detailed history was taken from all participants including details of diet, duration of acne, treatment history, family history, personal history, cosmetic use and any specific aggravating factors.

Alow glycemic load diet (comprising 45% of energy from low glycemic load carbohydrates, 25% from proteins and 30% from fats) was advised for patients in Group 1.

(Glycemic load = Glycemic index of the food item × its carbohydrate content (g)/100. Glycemic index values were referred from a reference table,^[7] from the Sydney University's Glycemic index website ^[8] and from the National Institute of Nutrition, ICMR, Hyderabad, India.^[9])

Constituents of the low glycemic load diet were: Total calories: 1900 (±300 kcal) Carbohydrates: 47% (±5%). Proteins: 23% (±3%) Fats: 30% (±3%) [Table - 1].

Table 1: Diet Chart

Every participant in group 1 received an individualized diet plan from the dietitian on the first visit which matched their baseline diet. Group 1 patients maintained a diet chart with a daily menu which was examined during every visit to determine adherence.

Topical 2.5% benzoyl peroxide gel and a mild noncomedogenic cleanser were advised for all patients in both the groups.

Acne lesions (face) were scored and graded at all visits (baseline, 4, 8 and 12 weeks). Skin evaluation was done by modified Cunliffe–Leeds lesion count technique.^[10] Body mass index and homeostasis model assessment of insulin resistance (fasting glucose mg/dl × fasting insulin μU/ml/405) were calculated at baseline and 12 weeks.

Statistical Package for the Social Sciences (SPSS) version 17.00 (SPSS Inc. 233, South Walker Drive, Chicago, IL) was used for statistical analysis. Student's unpaired t-test was used to analyze mean values of all quantitative variables. P < 0.05 was considered significant.

Sample size was determined by the formula: η = 2(Z_α+ Z_β)^[2]σ^[2] / d ^[2]. With the power of study 90% and the confidence level of 95%, substituting in the above formula, the number of participants required for the study was 84. A total of 111 were enrolled in the study; though some patients were lost to follow-up and some more discontinued the treatment, we retained the required number [CONSORT Flowchart].

Results

The study included 42 individuals in group 1 who followed a low glycemic load diet and 42 patients in group 2 with no dietary intervention. Fifty-four (64.2%) of patients were in the age group of 20–24 years and a majority (57,67.8%) were women [Table - 2]. No significant differences were seen between the two groups in age distribution (P = 0.443) or sex distribution. Both groups had higher female populations (P = 0.815). A majority (68, 80.9%) in both the groups had grade 3 acne, with no significant difference in the grade of acne between the 2 groups (P = 0.581).

Table 2: Demography of analyzed patients and allocated patients

There was also no significant difference in the duration of acne between two groups. Duration of acne in most patients in our study was between 1 month and 5 years (P = 0.430). Thirty-eight (90.4%) patients in group 2 had a family history of acne, compared to18 (42.8%) patients in group 1 (P = 0.0001). Thirty one (73.8%) patients in group 2 had a history of usage of cosmetics compared to 28.5% of patients in group 1 (P = 0.0001).

Adherence to the low glycemic load diet in group 1 as determined from their daily diet chart was up to 75%.

Even though there was a significant reduction in acne counts at the end of the study in both the groups, [Figure - 1] and [Figure - 2] there was no significant difference [Table - 3] between the two groups (P = 0.931). The difference between group 1 and group 2 in respect to reduction of body mass index [Table - 4] during the study period was statistically significant (P = 0.0001). The difference in homeostasis model assessment of insulin resistance [Table - 5] during the study period between the two groups was also statistically significant (P = 0.0001).

Figure 1: Patient in Group 1 at 0 and 12 weeks

Figure 2: Patient in Group 2 at 0 and 12 weeks

Table 3: Acne counts of the analyzed patients and allocated patients

Table 4: Body mass index of analyzed patients and allocated patients

Table 5: Homeostasis model assessment of insulin resistance of analyzed patients and allocated patients

Discussion

Nearly all individuals of 15-17 years age suffer from acne ^{[11],[12],[13]} and in 15–20% of young individuals, acne is moderate to severe.^{[11],[14],[15]} Smith et al., in a randomized controlled trial on 43 male participants aged up to 25 years with mild to moderate facial acne, found that acne was a common problem faced by that age group.^[16] The severity of acne vulgaris increases with maturity, and prepubertal girls with severe acne have higher dehydroepiandrosterone sulfate levels.^[17],[18] In a German study, the authors found that 64% of people in the age group of 20-29 years and 43% of those in the age group of 30-39 years had acne.^[19],[20]

A large Chinese study conducted among undergraduate students found that 78% of first-degree relatives had acne, which was heritable in nature.^[21] Many retrospective studies have identified the genetic basis and clustering of familial cases in acne.^{[22],[23],[24],[25],[26]}

In a cross-sectional study including Kitavan and Ache hunters, Cordain et al. observed that there was no acne in these populations consuming low glycemic load diets.^[2] They proposed that a high glycemic load diet is a major factor contributing to acne in Western populations.^[2] Kim et al. observed that consumption of fermented milk with lactoferrin decreases the total non-inflammatory and inflammatory lesion counts.^[27] Kwon et al. also found that patients consuming low glycemic load diets showed decrease in the number of acne lesions after 5 weeks.^[28] Di Landro et al. found that there was a positive association between acne and frequent consumption of milk and skimmed milk.^[29] In our study, patients who consumed a low glycemic load diet showed decrease in body weight after a period of 12 weeks while those on a regular diet did not. Consuming high glycemic index and carbohydrate-rich foods repeatedly causes acute hyperinsulinemia in adolescents. Hyperinsulinemia has been implicated in acne pathogenesis due to its association with high a bioavailability of androgens and high concentrations of free insulin-like growth factor I.^[4],[5] Rapid increase in insulin levels results in a rapid lowering of blood sugar. In turn, androgen released by the adrenal glands signals the liver to secrete glycogen from its storage to increase blood sugar towards normal. Decrease in blood sugar also stimulate food cravings. Finally, these cycles continue in a chronic manner. Frequent hyperinsulinemia increases appetite and then body weight and body mass index.^[30]

Nagpal et al. conducted a cross-sectional study to determine the prevalence of insulin resistance and metabolic syndrome in men affected by acne and found that post-adolescent males with acne and higher body mass index more commonly have high insulin resistance.^[31]

These studies indicate correlations between insulin resistance, body mass index and acne. In our study too, only group 1 patients had significant improvements in homeostasis model assessment of insulin resistance values, indicating that a low glycemic load diet helps decrease insulin resistance. However insulin resistance and body mass index values did not correlate with the number of acne lesions in our study .

Limitation of our study is application of mild topical cleanser in patients of both groups. Topical cleanser improves the normal skin barrier along with topical application of 2.5% of benzoyl peroxide gel in both groups. Topical benzoyl peroxide itself is a potent topical agent against acne. These might be the main factors responsible for significant reduction in acne counts in group 1 and group 2, so no significant difference in acne counts in between the two groups was observed. So a study on the effect of low glycemic load diet alone on acne may help us to overcome this limitation. Even the adherence to diet chart was only 75% in group 1 and this might be a limitation of the study. The other limitation would be that no specific diet chart was given to the group 2, so it is possible that some/many of them paid more attention to their diets once they were a part of the study.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the forms the patients have given their consent for their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Acknowledgement

We are grateful to Jestina Rachel Kurien for helping us to modify the manuscript.

Financial support and sponsorship

Partly sponsored by Manipal Academy of Higher Education.

Conflicts of interest

There are no conflicts of interest.