The frequency of fatty liver in patients with alopecia areata: A case–control study

Alopecia areata (AA) is an immune‐mediated nonscarring hair loss disorder with multiple subtypes, including alopecia universalis (AU). Previous studies have shown a link between serum lipid profile and alopecia. We aimed to investigate the frequency of fatty liver in patients with AU and patchy alopecia areata (PAA) compared to a control group.

Methods: This case-control study included patients with AU and PAA referred to a dermatology clinic from September 23, 2019 to September 23, 2020. A group of individuals without hair loss disorders attending the same clinic were selected as controls. Participants' general information, including age, sex, weight, height, and waist circumference (WC), was recorded. Body mass index (BMI) was calculated for all participants. Also, hyperlipidemia and statin use were noted and liver enzymes were evaluated. For AU and PAA patients, disease duration and the Severity of Alopecia Tool (SALT) score were also recorded. Then, all subjects underwent ultrasound to assess fatty liver and its grade.

Results:
Overall, 32 patients were included in each group. All three groups were comparable in age, sex, weight, height, BMI, WC, hyperlipidemia, abnormal liver enzymes, and statin use. Disease duration and SALT score were significantly higher in the AU than in the PAA group (p = 0.009 and p < 0.001, respectively). The frequency of fatty liver was the highest in AU patients (40.6%), followed by 34.4% in PAA patients, and 21.9% in controls (p = 0.263). This was also the case for grade-1 fatty liver, while grade-2 was more common in PAA patients, and grade-3 was only observed in one patient from the AU group (p = 0.496).

Conclusions:
Fatty liver was more frequent in AU and PAA patients than controls, without statistically significant differences. There might be an association between fatty liver and AA, particularly the AU subtype.

K E Y W O R D S
alopecia areata, alopecia universalis, fatty liver 2 | ME THODS

| Participants and study design
This case-control study included patients with AU, and patchy alopecia areata (PAA) referred to a dermatology clinic from September 23, 2019, to September 23, 2020. An expert dermatologist diagnosed AU and PAA. A group of individuals without hair loss disorders attending the same clinic were selected as controls. Patients with known liver diseases were excluded from the study.
The study received ethics approval and complies with the statements of the Declaration of Helsinki. Written informed consent was obtained from all the participants.
First, participants' general information, including age, sex, weight, height, and waist circumference (WC), was recorded. Body mass index (BMI) was calculated for all participants by dividing weight by the square of height. Also, hyperlipidemia and statin use were noted, and liver enzymes were evaluated. For AU and PAA patients, disease duration and the Severity of Alopecia Tool (SALT) score were also recorded. Then, all participants underwent ultrasound to assess fatty liver and its grade.

| Data analysis
We used the Statistical Package for the Social Sciences (SPSS) software (version 26.0; IBM Corp.) for data analysis. Continuous variables were described using means and standard deviations.
Categorical variables were described using frequencies and percentages. The chi-square and Fisher's exact tests were used to compare categorical variables among groups. Based on the central limit theorem, 14 and because the sample size in each group was 32, we used the one-way analysis of variance (ANOVA) test and the independent t-test to compare continuous variables. p values <0.05 were regarded as statistically significant.

| RE SULTS
Overall, 32 patients were included in each group. All three groups were comparable in age, sex, weight, height, BMI, and WC ( Table 1).
The frequency of hyperlipidemia was the highest in AU patients (31.3%), followed by PAA patients (15.6%); however, the difference was not statistically significant among groups (p = 0.131). On the contrary, abnormal liver enzymes were observed in 9.4% of AU patients, 12.5% of PAA patients, and 3.1% of controls (p = 0.524). Statin use was also the highest in AU patients (28.1%), with no statistically significant difference among groups (p = 0.098). Meanwhile, disease duration and SALT score were significantly higher in the AU group than in the PAA group (p = 0.009 and p < 0.001, respectively). Table 2 compares fatty liver and its grade among the three groups. Although the frequency of fatty liver was the highest in AU patients (40.6%), followed by 34.4% in PAA patients and 21.9% in controls, the difference was not statistically significant (p = 0.263).
This was also the case for grade-1 fatty liver, while grade-2 was more common in PAA patients, and grade-3 was only observed in one patient from the AU group (p = 0.496).

| DISCUSS ION
In the current study, we found a higher frequency of fatty liver in patients with AU and PAA compared to controls; however, the difference did not reach statistical significance. The same results were found for grade-1 fatty liver, while grade-2 was more frequent in PAA patients, and grade-3 was only observed in one AU patient.
The higher frequency of fatty liver in AU patients compared with the PAA group can be attributed to the longer disease duration and the higher SALT score in this group. To the best of our knowledge, in previous studies, fatty liver has not been evaluated in patients with AA. Still, recently the replacement of nonalcoholic fatty liver disease (NAFLD) with metabolic dysfunction-associated fatty liver disease (MAFLD) has been proposed, highlighting the association between fatty liver and metabolic syndrome. 15 Additionally, compared to controls, metabolic syndrome has been reported to be higher in patients with AA. 16 Further, in another study by Abdollahimajd et al., although there was no significant association between AA and metabolic syndrome, some of its components were more prevalent in these patients. 17 Insulin resistance is a component of some metabolic syndrome criteria, playing an essential role in its pathophysiology, 18 and AA patients appear to have increased insulin resistance. [19][20][21] There is evidence that immune responses and hair disorders are directly or indirectly linked to metabolic intermediates and lipids in the cholesterol biosynthesis pathway. 22 Also, the relationship between alopecia and dyslipidemia has been demonstrated, 23,24 yet these associations are mainly from studies on patients with androgenic alopecia, which has a different etiology from AA.
Another justification for the potentially higher frequency of fatty liver in AA patients can be an oxidant/antioxidant imbalance, which has been reported in these patients. 25 El-Farargy et al. compared serum paraoxonase 1 (PON1) levels between AA patients and controls and found significantly decreased PON1 levels in AA patients. 25 While the exact function of PON1 is unclear, it seems to prevent the lipid peroxidation of lipoproteins and cells. 26 On top of that, ischemia-modified albumin levels appear to be higher in AA patients, indicating an oxidant/antioxidant imbalance with increased cardiovascular risk. 27 Besides, adiponectin, a serum protein produced primarily by adipose tissue, has been proposed as a novel biomarker of disease severity in AA. 28 In our extensive search through the literature, we found no studies evaluating the frequency of fatty liver in AA patients. This can be regarded as a potential strength for the current study, yet it was not without limitations. First, the sample size was relatively small limiting our findings' generalizability. Thus, the results should be interpreted with caution. Second, the gold standard for the evaluation of fatty liver is liver biopsy, which has been replaced by FibroScan due to its  invasiveness and complications, 29 but we used ultrasound for this