Atopy is predictive of a decreased need for hospitalization for coronavirus disease 2019

The coronavirus disease 2019 (COVID-19) pandemic has caused high utilization of health care resources, including hospitalization and intensive care unit treatment. There has been considerable interest in determining which clinical factors stratify patients into high or low risk for severe COVID-19 illness to aid with clinical decision making. Advanced age, cardiovascular disease, and diabetes have been associated with increased COVID-19 severity.¹ Asthma seems to be underrepresented as a COVID-19 comorbidity compared with the global prevalence of the disease.^{1 , 2} To date, the effect of atopic conditions on the disease course of COVID-19 has yet to be fully elucidated. This study is a large, 2-site cohort of patients positive for COVID-19 designed to understand the association between atopic conditions and COVID-19 disease severity.

The study was approved by the institutional review boards of both the participating universities. An electronic medical record database search was performed to identify patients tested for COVID-19. All encounters of these patients were carefully reviewed and only cases who were active patients at the health systems, with a detailed medical history available before the diagnosis of COVID-19, were included in the study. Through extensive chart review, demographic and clinical factors related to allergy evaluations, including active or historical diagnoses of asthma, allergic rhinitis (AR), eczema, and food allergy, and variables related to COVID-19 infection severity were retrieved from the electronic medical records. To confirm these allergy-related variables, charts were initially reviewed by trained researchers for codes of the International Classification of Diseases, Tenth Revision, and the diagnoses were subsequently verified by clinical history and evidence of allergic sensitization by means of skin prick or serum immunoglobulin E testing by board-certified allergists.

Logistic regressions were used to compare COVID-19–related outcomes in association with preexisting AR, asthma, eczema, and food allergy while adjusting for age, sex, body mass index, and race. Patients with a history of 1 or more of the following: AR, eczema, or food allergy were labeled as having atopy. Similar regression analysis was performed to compare individuals with and without atopy. Patients with asthma were first analyzed as a single group compared with those without asthma. Next, patients with asthma were divided into allergic asthma, defined as patients who received a diagnosis of both AR and asthma, and nonallergic asthma for subgroup analysis.

The combined series included 2013 patients with positive nasopharyngeal polymerase chain reaction test results for severe acute respiratory syndrome coronavirus 2. Complete data on demographic variables, confirmed allergy diagnoses, and COVID-19 management variables were available for 1043 patients who were used for our analysis. A total of 970 patients were excluded because they were not active patients in the medical center's electronic medical records before their COVID-19 diagnosis. The mean (SD) age was 50.16 (16.77) years, 43.3% were men, and 58.1% were African Americans. There was no difference in the demographics between the groups with and without atopy. Among the 1043 cases, 257 (24.6%) had atopy. Atopy was associated with significantly lower odds of hospitalization for COVID-19 (Table 1 ); 27.6% of those with atopy and 37.8% of those without atopy but diagnosed as having COVID-19 were hospitalized (adjusted P = .004). Furthermore, atopy was associated with a decreased duration of hospitalization for COVID-19. When analyzing AR, eczema, and food allergy separately, both AR and eczema were associated with lower odds of COVID-19–related hospitalization (Table 1). AR was also associated with a decreased duration of hospitalization and intubation for COVID-19. Asthma was associated with increased intubation time; mean intubation time was 13.4 vs 8.1 days in those with asthma vs those without asthma (adjusted P = .016). When the patients with asthma were grouped into those with allergy and those without allergy, only the patients without allergic asthma were associated with prolonged intubation time (Table 1). There was no difference in mortality between the groups with and without atopy. Furthermore, the treatment regimens for COVID-19 were not notably different between these 2 groups, with minimal use of systemic corticosteroids in this cohort.

Table 1.

Characteristics and Outcomes of COVID-19 Patients in Association With Atopic Conditions in 1043 COVID-19 Positive Patients in a 2-Center Study From the United States

COVID-19 outcomes	Conditions
	Atopya		Allergic rhinitis		Eczema		Food allergy		Allergic asthmab			Nonallergic asthma
	n = 257		n = 171		n = 58		n = 80		n = 86			n = 179
	Odds ratio (95% CI) or beta (95% CI)	P valuec	Odds ratio (95% CI) or beta (95% CI)	P valuec	Odds ratio (95% CI) or beta (95% CI)	P valuec	Odds ratio (95% CI) or beta (95% CI)	P valuec	Odds ratio (95% CI) or beta (95% CI)	P valuec	Odds ratio (95% CI) or beta (95% CI)	P valuec
Hospitalized	0.62 (0.44-0.86)	.004	0.64 (0.43-0.92)	.02	0.51 (0.25-0.90)	.045	0.97 (0.57-1.62)	.90	0.77 (0.57-1.03)	.08	1.03 (0.75-1.41)	.85
Duration of hospitalization, mean (SD)	−1.67 (−2.9 to −0.4)	.008	−2.33 (−3.6 to 0.9)	.001	−2.2 (−4.5 to 0.2)	.07	1.79 (−0.57 to 4.17)	.13	−0.68 (1.81-0.45)	.24	1.00 (0.26-2.27)	.12
Admitted to ICU	0.75 (0.46-1.17)	.21	0.71 (0.40-1.19)	.21	0.65 (0.22-1.55)	.37	0.97 (0.41-2.01)	.93	1.02 (0.69-1.51)	.90	1.31 (0.85-1.99)	.22
Intubated	0.61 (0.33-1.07)	.10	0.75 (0.38-1.37)	.37	0.18 (0.01-0.87)	.09	0.71 (0.21-1.83)	.53	1.06 (0.66-1.68)	.81	1.45 (0.87-2.36)	.14
Duration of intubation, mean (SD)	−1.1 (−2.3 to +0.2)	.10	−1.4 (2.80 to −0.07)	.039	−0.93 (−3.4 to 1.5)	.46	1.08 (−2.08-4.25)	.50	0.21 (−1.00-1.44)	.72	1.18 (0.41-3.21)	.011

Abbreviations: BMI, body mass index; CI, confidence interval; COVID-19, coronavirus disease 2019; ICU, intensive care unit.

NOTE: Bold font indicates statistical significance.

^aAtopy is defined as presence of allergic rhinitis and/or eczema and/or food allergy.

^bAllergic asthma is defined by asthma diagnosed based on the Global Initiative for Asthma guidelines and evidence of inhalant allergen sensitization, and nonallergic asthma are individuals with asthma without inhalant allergen sensitization. In this article, those with comorbid asthma and allergic rhinitis are grouped under allergic asthma and those with asthma without allergic rhinitis are grouped under nonallergic asthma.

^cP values for categorical variables are calculated by logistic regression comparing categorical variables adjusted for age, race, sex, and BMI, or analysis of covariance comparing the adjusted means of continuous variables adjusting for age, race, sex, and BMI. Odds ratio (95% CI) are reported for categorical variables and beta is reported for continuous variables.

In our series, AR and eczema were associated with a lower rate of hospitalizations, and AR was associated with a lower duration of hospitalization and intubation for COVID-19 infections. This adds greater credence to an earlier report from China, in which respiratory allergies were not identified as risk factors for severe COVID-19 illness.¹ In a recent study, Jackson et al³ found that expression of angiotensin-converting enzyme 2, the cell receptor utilized by severe acute respiratory syndrome coronavirus 2 for cell entry, was decreased in individuals with atopy. Interestingly, reductions in angiotensin-converting enzyme 2 expression levels correlate with allergic sensitization, higher levels of total immunoglobulin E, and type 2 inflammatory cytokines.3, 4, 5 Interleukin 13, a major type 2 inflammatory cytokine, is found to significantly reduce angiotensin-converting enzyme 2 expression in airway epithelial cells.⁵ Our findings of the association of AR and eczema with decreased need of hospitalization for COVID-19 provide robust clinical data to support these mechanistic findings.

The role of asthma and its association with COVID-19 severity is more complicated.^{6 , 7} Asthma was not reported in previously published cohorts of COVID-19 from China^{1 , 2}; although data from the Centers for Disease Control indicate that asthma is present in as high as 27% of hospitalized COVID-19 patients in the United States in the 20 to 49 year age range.⁶ This could be explained by the lower rates of asthma in China (2%-4%) than those in the United States (8%-11%).^{8 , 9} In the current report, allergic asthma was not associated with any COVID-19 outcome variable despite AR being protective against hospitalization. Furthermore, nonallergic asthma was associated with a prolonged intubation time which confirms an earlier study.¹⁰ It is possible that asthma, as a chronic pulmonary disease susceptible to viral-induced exacerbations, places those with more severe COVID-19 illness at risk for more prolonged lung involvement. However, a coexisting atopic background may mitigate the severe inflammatory response syndrome of COVID-19 in those with allergic asthma, leading to the absence of the prolonged intubation time reported in individuals with nonallergic asthma.

The knowledge that atopy is associated with less severe COVID-19 outcomes can be instructive in clinical risk stratification. Further studies are needed to understand the underlying mechanism of these apparent protective physiological factors that may prove advantageous in future prevention and treatment strategies.