1. Cecchini, Michael
  2. Tsoulis, David J.
  3. Velykoredko, Yuliya

Article Content


Egeberg, A., Hansen, P. R., Gislason, G. H., & Thyssen, J. P. (2016). Clustering of autoimmune diseases in patients with rosacea. Journal of the American Academy of Dermatology, 74, 667.e1-672.e1.


Rosacea is a common skin disease predominantly affecting the face characterized by redness, flushing, and dilated blood vessels. There are four overlapping subtypes including erythematotelangiectatic, inflammatory papulopustular, phymatous, and ocular. It is viewed as a disease caused by environmental triggers in a genetically predisposed host. Triggers may include ultraviolet radiation, spicy foods, a mite (Demodex folliculorum), and alcohol. There is evidence that about 50% of the risk of rosacea is related to various genetic factors. Recent analyses have identified that genetic loci related to rosacea are linked to Type 1 diabetes mellitus (T1DM), and other studies have shown genetic relationships between T1DM, rheumatoid arthritis (RA), celiac disease, and multiple sclerosis.


Egeberg and colleagues designed a case-controlled study to examine possible relationships between rosacea and the autoimmune disorders discussed above. They used two Danish registries, examining data from January 1997 to December 2011, to collect information on disease status and other variables such as age, gender, time of encounter, socioeconomic status, smoking history, and so forth. For every case of rosacea, they identified five healthy controls. The investigators also examined their data set for potential sources of bias. To ensure that there was lower misclassification bias, they examined their data with and without patients who also had a diagnosis of systemic or cutaneous discoid lupus erythematosus. To ensure that there was no detection bias given the large number of patients, they examined an unrelated condition, colon cancer, for association.


The study identified 6,759 cases (2,489 men, 4,270 women) and 33,795 controls. All of the autoimmune conditions examined were found to be associated with rosacea. Specific adjusted odds ratios were 2.59 for T1DM (95% CI [1.41, 4.73]), 1.65 for multiple sclerosis (95% CI [1.20, 2.28]), 2.03 for celiac disease (95% CI [1.35, 3.07]), and 2.14 for RA (95% CI [1.82, 2.52]). The findings were consistently statistically significant after removal of the population with systemic or cutaneous discoid lupus erythematosus, and there was no observed association between rosacea and colon cancer. Interestingly, only the association between rosacea and RA was significant in men when the analysis was performed by gender (2.05, 95% CI [1.48, 2.84]).


REMARKS: Overall, this was the first large-scale study examining the possible association between rosacea and these autoimmune diseases. An association is plausible considering the previously observed genetic overlap, and the findings confirm this. It is important to note that this observation does not prove a causative relationship between the disorders but, rather, there is likely a complex interplay between shared genetic and environmental risk factors that results the observed association. Egeberg et al. specifically mention sun exposure and/or Vitamin D as factors possibly involved in all five diseases. Some of the limitations of this study include lack of patient stratification based on the four rosacea subtypes along with the lack of generalizability outside Danish or predominantly White populations. Regardless, the study contributes to the accumulating evidence of overlap between rosacea and other autoimmune conditions. This could raise the index of suspicion for diagnosis among various healthcare providers and help identify patients with autoimmune conditions earlier in their disease course.



Hessam, S., Sand, M., Gambichler, T., & Bechara, F. G. (2015). Correlation of inflammatory serum markers with disease severity in patients with hidradenitis suppurativa (HS). Journal of the American Academy of Dermatology, 73(6), 998-1005.


Hidradenitis suppurativa (HS) is a chronic inflammatory condition affecting apocrine gland-bearing skin. The disease favors the groin, inframammary folds, and axillae; however, the perianal, gluteal, and abdominal areas may also be involved. Multiple pathogenic mechanisms have been proposed, resulting in the formation of painful nodules, abscesses, purulent-draining sinus tracts, and scarring. Although clinical tools for assessing disease severity in HS exist, they do not incorporate the use of objective inflammatory markers. To date, there are no known studies that have evaluated the use of routine inflammatory markers in association with a scoring system of HS disease severity.


Hessam et al. sought to evaluate the association between disease severity (by Hurley classification and modified HS score [mHSS]) and markers of inflammation (c-reactive protein [CRP], white blood cell count [WBC], and neutrophil count). The study was a retrospective cross-sectional study of 275 patients from one academic center in Germany in 2013. Charts were included if they had a confirmed diagnosis based on clinical criteria as well as available mHSS and laboratory data. Patients excluded from the study were younger than 18 years old, were receiving active antibiotic or immunosuppressive therapy, received antibiotics or immunosuppression or had surgery within 4 weeks before their initial visit, or had a suspected infectious disease. It was noted that increased WBC was driven by neutrophilia; therefore, they continued their analysis only examining neutrophil counts.


One hundred four patients met criteria for inclusion in the study. With regard to possible confounders, smokers had a statistically significantly higher neutrophil count than nonsmokers, but there was no difference between these groups with regard to their mHSS. There was also a significant positive correlation between body mass index and elevated CRP levels (r = .366, p = .0001).


CRP and neutrophil counts increased with disease severity based on both Hurley staging and mHSS. With regard to predicting disease severity, severe disease defined as Hurley stage III was independently predicted by CRP level (odds ratio = 1.077, 95% CI [1.013, 1.145]). Severe disease, defined by mHSS70, was independently predicted by both CRP and body mass index.


REMARKS: The results of this study suggest that CRP and neutrophil count could play a quantitative role in stratifying disease severity in patients with HS. In addition, CRP may also be useful for monitoring responses to anti-inflammatory therapy and for assessing subclinical disease activity. Although other studies assessing the use of CRP and neutrophil count are known in the literature, it was not possible to compare those findings with the current study because of the differences in patient populations, variable laboratory instrumentation, and normal reference ranges. Future research should be directed at prospective studies to examine if inflammatory markers can predict disease course and response to treatment. Given that these are markers of general inflammation, it will be important to consider their variability in the general population and the influence of various demographic and disease factors on their levels. This would increase their utility in HS and other diseases.



Mikailov, A., Cohen, J., Joyce, C., & Mostaghimi, A. (2016). Cost-effectiveness of confirmatory testing before treatment of onychomycosis. JAMA Dermatology, 152(3), 276-281.


Onychomycosis is a fungal infection of the nail. It is the most common nail disease in North American adults with a prevalence of 7%-14%. In patients seeking care for nail dystrophy, the prevalence can be as high as 65%-95%. Two antifungal treatment options for onychomycosis include oral terbinafine and the recently approved topical agent eficonazole 10% solution. Current treatment guidelines are outdated and urge confirmatory diagnostic testing before initiating treatment. These confirmatory tests include periodic-acid-Schiff (PAS) stain, potassium hydroxide (KOH) testing, or fungal culture. Because these recommendations were established, the costs of treatment and confirmatory testing have changed dramatically. The goal of Mikailov and colleague's study was to evaluate the cost-effectiveness of confirmatory testing before treating onychomycosis with oral terbinafine or topical eficonazole as well as the potential harm associated with the use of these therapies empirically.


Partners Healthcare in Boston, MA, conducted the study between April 2014 and September 2015. The investigators performed a statistical model analysis that compared three approaches with the management of onychomycosis: (a) empiric treatment for all patients clinically suspected to have onychomycosis, (b) in-clinic KOH screen followed by treatment if positive and PAS test for KOH-negative patients before treatment, and (c) PAS testing before treatment. The cost analyses were modeled based on disease prevalence of 30%, 60%, and 90%. The terbinafine regimen consisted of 250 mg by mouth per day for a 12-week course, whereas eficonazole regimen was 1 drop per day to each affected nail for 48 weeks. All the cost analyses were performed in U.S. dollars and specific for American pharmacies with appropriate reimbursement codes. Terbinafine-induced liver injury, which is the most concerning treatment-associated side effect, was factored into the analysis.


The calculated cost of treatment and liver enzyme monitoring for a single 12-week course of terbinafine was $53. Treatment with eficonazole for one nail was $2,307. Confirmatory testing with KOH and PAS cost $6 and $148, respectively. Treating empirically with terbinafine was more cost-effective than confirmatory testing with all modeled disease prevalences. However, the cost savings were highly significant when a diagnosis was confirmed before treating with eficonazole.


Experience with terbinafine has shown that it is relatively safe and associated with few toxic liver injuries. Specifically, the Clinical and Research Information on Drug-Induced Liver Injury Database affiliated with the National Library of Medicine estimates clinically apparent liver injury from terbinafine as 1 case per 50,000-120,000 patients. Because this side effect is so rare, the study noted that the cost burden to prevent one case of liver injury would range between $9.62 and $233.89 million.


REMARKS: Drug costs and side effects are major factors when choosing between empiric treatment or pursuing confirmatory testing for onychomycosis. This study does not support the current guidelines, which urge confirmatory testing before treating nail fungal infections. These results should be interpreted with caution because they are based on a statistical modeling and not real-world clinical experience. The costs are also specific for the United States and may not be generalizable to other countries. There are great discrepancies in drug and health delivery costs between countries, which are generally higher in the United States. Furthermore, the office practices of physicians seeing patients with onychomycosis can be variable as many do not perform in-office KOH testing. Nonetheless, this study highlighted the importance of health economics in guiding clinical decision making. Perhaps, when the clinical diagnosis of onychomycosis is very compelling, some providers may consider treating empirically with terbinafine especially if the patient has no underlying liver disease, but further studies are warranted.