Large cohort study shows increased risk of developing atopic dermatitis after COVID-19 disease 2023, Schmitt et al

[Sly Saint]

To the Editor,
Atopic dermatitis (AD) is a frequent, chronic inflammatory disease constituting significant burden to patients, their families and healthcare systems.1 The pathophysiology is multifactorial involving genetic predisposition, epidermal dysfunction, and cutaneous inflammation.2 Systemic infections trigger AD flares and are related to the manifestation of new-onset AD.1 Following the acute phase of a SARS-CoV-2 infection, some people develop long-lasting symptoms, known as post- or long-COVID.3 Different incident diseases are associated with prior COVID-19 disease, including cardiovascular and respiratory diseases, mental health problems, fatigue, and autoimmune diseases.4, 5 Due to the role of viral infections in the pathophysiology of AD, we hypothesized that the risk of new-onset AD is increased in individuals with previous SARS-CoV-2 infection.
To test this, we undertook a large matched cohort study based on German routine healthcare data covering inpatient and outpatient care, diagnoses, prescriptions and demographic data. Patients with polymerase chain reaction (PCR)-confirmed COVID-19 infection in the year 2020 were matched 1:3 by age, sex, previous occurrence of an autoimmune disease and comorbidity propensity score to control subjects without COVID-19-infection and followed up to 15 months through June 2021. Patients with prevalent AD in the four quarters (one inpatient diagnosis or two outpatient diagnoses in two different quarters with ICD-10: L20 or L30 for adults) before the initial SARS-CoV-2 infection or their assigned index date were excluded. Following the NICE guidelines on long-COVID,3 we defined the post-COVID-19-phase starting 3 months after the assigned index date. Primary outcome was new-onset AD. Patients were considered as having new-onset AD, if they received at least two physician documented diagnoses of AD (ICD-10: L20 or L30 for adults), no more than two quarters apart or an inpatient diagnosis in the post-COVID period. Additionally, we requested at least one prescription of topical or systemic treatment approved for AD (Table S1). We calculated incidence rates (IR) per 1000 person-years for the entire study population and predefined subgroups using Poisson models to estimate the IR-ratios (IRR) for the development of AD as a function of a prior diagnosis of COVID-19.5 Because of the non-interventional nature of routine healthcare data, no consent to participate was collected. This waiver for informed consent was confirmed by ethics committee of the Faculty of Medicine Carl Gustav Carus at the TU Dresden (BO-EK [COVID]-482,102,021).
In total, 641,704 COVID-19-patients, and 1,907,992 matched control cases without COVID-19 were included (Figure S1). 23,740 patients in the COVID-19-group and 111,818 cases in the control cohort were excluded because of prior AD. The IR of AD 3 to 15 months after the assigned index date was 7.35 (95%-CI 7.11–7.59) per 1000 person-years in the COVID-19 group and 5.53 (95% CI: 5.32–5.74) in the control group. The largest risk difference was observed among those under 18 years of age. Thus, patients with prior COVID-19 infection had a 33% increased risk of developing AD compared to controls (IRR 1.33; 95%-CI 1.26–1.40). The risk for new-onset AD was significantly increased in both sex groups, medication groups and all age groups. However, the confidence intervals of the relative risks overlapped between these groups. (Table 1, Figure 1).
TABLE 1. Incidence rate (IR) for an atopic dermatitis 3 to 15 months after the index date among persons with versus without SARS-CoV2 infection.

All	3477	7.35	2616	5.53	1.82	<0.01
Female	2220	8.19	1637	6.03	2.16	<0.01
Male	1257	6.22	979	4.84	1.38	<0.01
Age <18	622	14.59	456	10.70	3.90	<0.01
Age 18–64	2346	6.61	1803	5.08	1.53	<0.01
Age 65–79	282	6.31	200	4.46	1.86	<0.01
Age 80+	227	7.30	158	5.03	2.27	<0.01
All with medication	1284	2.66	1013	2.10	0.56	<0.01
Female with medication	771	2.78	644	2.32	0.46	<0.01
Male with medication	513	2.50	368	1.79	0.71	<0.01
Age < 18 with medication	210	4.72	150	3.38	1.34	<0.01
Age 18–64 with medication	829	2.29	698	1.93	0.36	<0.01
Age 65–79 with medication	130	2.87	91	1.99	0.88	<0.01
Age 80+ with medication	115	3.65	74	2.33	1.31	<0.01

[th]

Outcomes​

[/th][th]

Events COVID-19​

[/th][th]

IR of AD COVID-19​

[/th][th]

Events weighted controlsa​

[/th][th]

IR of AD weighted controls​

[/th][th]

Difference in IR of AD​

[/th][th]

p-value binomial probability test​

[/th]​

• Abbreviations: AD atopic dermatitis; IR- Incidence rate.
• a Weighted events have been rounded, while the estimates used the exact values. Medication includes all topical or systemic treatments, which are currently in use in Germany (TableS1).

​

FIGURE 1
Open in figure viewerPowerPoint
Forest plot comparing incident risk ratios for any first-onset of atopic dermatitis 3 to 15 months after a SARS-CoV-2 infection by subgroup.​

In summary, our study shows consistent and significantly increased new-onset of AD in patients with previous SARS-CoV-2 infection. Limitations of the presented study include its observational nature so that causal conclusions can only be drawn with caution. A major strength is the large sample size and the robustness of the findings in several analyzed subgroups. This new evidence strengthens previous studies that suggested a relevant pathophysiological role of viral infections in AD.6 Future research is necessary to further investigate the role of the COVID-19 pandemic on the global burden of AD.