key: cord-0915858-yldn3h0z
authors: Carosi, Giulia; Morelli, Valentina; Del Sindaco, Giulia; Serban, Andreea Liliana; Cremaschi, Arianna; Frigerio, Sofia; Rodari, Giulia; Profka, Eriselda; Indirli, Rita; Mungari, Roberta; Resi, Veronica; Orsi, Emanuela; Ferrante, Emanuele; Dolci, Alessia; Giavoli, Claudia; Arosio, Maura; Mantovani, Prof Giovanna
title: Adrenal insufficiency at the time of COVID-19: a retrospective study in patients referring to a tertiary centre
date: 2020-10-27
journal: J Clin Endocrinol Metab
DOI: 10.1210/clinem/dgaa793
sha: 537a8c0f2efeb810cfbcd5e2c4e0aa1ac1784582
doc_id: 915858
cord_uid: yldn3h0z

CONTEXT: COVID-19 represents a global health emergency and infected patients with chronic diseases often present with a severe impairment. Adrenal insufficiency (AI) is supposed to be associated with an increased infection risk which could trigger an adrenal crisis. OBJECTIVE: Our primary aim was to evaluate the incidence of COVID-19 symptoms and complications in AI patients. DESIGN AND SETTING: We conducted a retrospective case-control study. All patients were on active follow-up and lived in Lombardy, one of the most affected areas. PATIENTS: We enrolled 279 patients with primary and secondary AI and 112 controls (patients with benign pituitary lesions without hormonal alterations). All AI patients had been previously trained to modify their replacement therapy on stress doses. INTERVENTION: By administering a standardized questionnaire by phone, we collected data on COVID-19 suggestive symptoms and consequences. RESULTS: In February-April 2020, the prevalence of symptomatic patients (complaining at least one symptom of viral infection) was similar between the two groups (24% in AI and 22.3% in controls, p 0.79). Highly suggestive COVID-19 symptoms (at least two including fever and/or cough) also occurred equally in AI and controls (12.5% in both groups). No patient required hospitalization and no adrenal crisis was reported. Few nasopharyngeal swabs were performed (n=12) as indicated by sanitary regulations, limiting conclusions on the exact infection rate (two positive results in AI and none in controls, p 0.52). CONCLUSIONS: AI patients who are adequately treated and trained, seem to display the same incidence of COVID-19 suggestive symptoms and disease severity as controls.

COVID-19 is an infectious respiratory syndrome caused by a novel coronavirus called severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (1). It has been firstly identified in Wuhan, China, as responsible for numerous pneumonia cases at the end of 2019. The infection has now spread worldwide, becoming a global health emergency (2) . In Italy, Lombardy was the most hardly hit region by the pandemic and is still one of the most affected regions in the world, with a prevalence of swab-confirmed infection of 0.75% at the end of April (3) . However, in Lombardy pharyngeal swab has been mostly performed in severely symptomatic patients referring to Emergency Rooms, thus the true prevalence of the disease is unknown. The estimated prevalence of infected people based on clinical manifestations, as in early April 2020, is 19.6% (4).

The clinical presentation of symptomatic infection from SARS-CoV-2 ranges from mild to critical. Most infections are mild, however, the 14% and 5% of patients show a severe and critical disease, respectively (5) (6) (7) (8) (9) (10) (11) . In Italy, an overall fatality rate of 13.7 % has been registered, ranging from 0.2% in young patients to 31.7% in the elderly (3) . The main symptoms reported in affected patients were fever, cough, myalgia or fatigue and dyspnea while minor symptoms included headache gastrointestinal symptoms and conjunctivitis (12) .

Risk factors for severe illness are represented by advanced age, smoking, and several comorbidities such as cardiovascular disease, diabetes mellitus, chronic lung disease, cancer, chronic kidney disease, and obesity (12) (13) (14) (15) (16) (17) . Immunocompromising conditions and liver disease were proposed as potential risk factors although specific data are limited (18) . Investigations about COVID-19 presentation in specific fragile groups with other chronic diseases have been started (19) .

Adrenal insufficiency (AI) is a chronic condition of inadequate cortisol production from different origins (20) . As reported by a recent statement released by the European Society of Endocrinology, patients with A c c e p t e d M a n u s c r i p t 6 primary (PAI), mainly Addison's disease (AD), as well as secondary adrenal insufficiency (SAI), should be considered at risk of being affected by COVID-19 (21, 22) . Firstly, because of the overall increased risk of infections reported in AI patients (23) , probably as a consequence of their inefficient innate immune response (24) , and secondly, because of the increased mortality rate observed facing respiratory infections, mainly linked to adrenal crisis exacerbations (25) . Some clinical guidelines on how to manage AI patients at the time of COVID-19 pandemic have been issued (22, 26, 27) . However, so far, data on COVID-19 presentation and outcome in adrenal insufficient subjects are lacking.

We conducted an observational case-control study involving adult patients with chronic adrenal insufficiency, both primary (PAI) and secondary (SAI). As control group, we recruited patients with nonfunctioning pituitary microadenomas (NFPA) and empty sella, without hormonal deficiencies. We made this choice because we expected, in this group, the same incidence of infective diseases as normal subjects and, moreover, most anamnestic data were already present in medical records, avoiding the need for a protracted recruitment phase. Hypoadrenal patients with a previous diagnosis of Cushing syndrome have been excluded from the enrolment considering hypercortisolism a condition highly associated with an increased infection risk which may persist after surgical treatment (28) . Patients with 21-hydroxylase deficiency were excluded too, on the basis of the potential role of hyperandrogenism on the severity of COVID-19 presentation (29) (30) (31) .

Diagnosis of PAI and SAI was previously performed following International Guidelines (32, 33) . Other pituitary deficiencies in SAI patients, if present, were adequately replaced with specific hormonal therapies.

A c c e p t e d M a n u s c r i p t 7 Inclusion criteria for both groups were represented by an active follow up at our Endocrinology Unit, Ospedale Maggiore Policlinico, in Milan (last visit on site performed in the last two years) and by a stable residence in Lombardy region.

All patients' medical records sorted by name in our Centre were systematically analysed, and we extracted all patients presenting the cited inclusion criteria. All patients answered the phone call.

From 1 st to 15 th May 2020, we retrospectively collected clinical data using phone calls, asking patients to report any symptoms of viral infection that occurred from January 1 st to April 30 th 2020 and any specific exam performed. According to Italian epidemiological data on COVID-19 outbreak, (34) we considered as suggestive for SARS-CoV-2 infection only symptoms occurred from February 14 th .

We registered the presence of fever (higher than 37.5 °C), cough, myalgia, fatigue, dyspnea, gastrointestinal symptoms, conjunctivitis, anosmia, ageusia, upper respiratory tracts symptoms, or others (thoracic pain, headaches, otalgia), along with the need to increase oral replacement therapy dosage or parenteral administrations of steroids. We collected data on nasopharyngeal swabs and chest imaging results. In the absence of swab and imaging data, we considered highly suggestive (HS) of COVID-19 patients reporting at least two symptoms including fever or cough which are the most frequent. We also analysed data defining HS-symptoms with stricter criteria, including the presence of at least two out of three of the cardinal symptoms: fever, cough and loss of sense of smell and/or taste. Hospitalization and emergency unit access for suspicious/confirmed COVID-19 infection were also registered.

Data on contacts with infected people, flu shot injection in the winter season, and smoking habits were collected. Working habits acquired from February were also collected.

A c c e p t e d M a n u s c r i p t 8 Personal data (age and sex), cause of adrenal insufficiency, presence of pituitary deficiencies other than SAI, and the type of replacement therapy were extracted from patients' medical records.

All AI patients have been contacted at the start of pandemic receiving a brief reminder about keeping adherence to their replacement therapy and adopting sick day rules when indicated. We also invited patients to strictly observe national sanitary rules at work.

The Local Ethical Committee (Fondazione IRCCS Ca' Granda, Milan) approved the protocol study and patients gave their verbal consent to participate.

All categorical data, including diagnosis of COVID-19 and suggestive symptoms, were expressed as percentage (%) and proportion (/) and analysed using Chi-squared test or Fisher exact test when indicated.

Continuous parameters with normal distribution were described as mean ± standard deviation (SD) and compared using T test while non-Gaussian data were expressed using the median and the interquartile range (IQR) and compared using a non-parametric test (Wilcoxon-Mann-Whitney). Assuming that patients with hypoadrenalism showed a higher incidence of infections than the general population (we considered a RR of 2.5) (23), we estimated that at least 102 patients per group needed to be enrolled to detect a significant difference in COVID-19 incidence (with an alpha error of 5% and power of 90%). P-values <0.05 were considered as statistically significant. All statistical analyses were performed using SPSS, version 26 (IBM, Cary, NC, USA).

A c c e p t e d M a n u s c r i p t 9

On the basis of inclusion criteria, we enrolled 279 AI patients: 219 SAI and 60 PAI with a mean age 57.5±15.6 and 56.8±15.7, respectively (p 0.754). AI females were 139/279 (49.8%) with a higher prevalence in PAI than SAI (68.3% vs 44.7%, respectively, p 0.001).

The In the control group, 112 patients were recruited (89 patients with NFPA and 23 with empty sella without hormonal alterations as inclusion criteria). Mean age was 57.5±14.3 and females were 52.7%.

AI and controls' characteristics (age, sex distribution, smoking, and working habits) were comparable, see Table 1 . Only the prevalence of flu shots was higher in AI patents, as expected as a fragile group.

A c c e p t e d M a n u s c r i p t 10

Symptoms occurrence over time is displayed in Figure 1 . In both AI and control groups, most patients reported the outset in March 2020 (38% and 36%, respectively). while another peak of symptoms was reported in controls in former February which overlaps with flu peak period (35) (21% vs 10% of symptomatic controls and AI respectively, p 0.096). Splitting the hypoadrenal group in symptomatic PAI and SAI, we confirmed that most patients in both groups complained their symptoms during COVID-19 outbreak (13% vs 21.5% in the second half of February, respectively, p=0.55 and 52% vs 32% in March, respectively, p=0.18).

As stated before, only symptoms that occurred from February 14 th were considered as suggestive for COVID-19 and thus included in the following analysis of symptomatic patients.

The main clinical features and collected data of both AI patients and controls are shown in Table 1 . The prevalence of symptomatic patients (complaining at least one symptom of viral infection, mentioned above) was similar between the two groups (24% in AI and 22.3% in controls, p 0.65). Specific symptoms distribution in the whole AI group, in SAI/PAI and in controls, is represented in Figure 2 . Except for the higher prevalence of gastrointestinal symptoms in controls vs AI patients (42.3% vs 19.4%, respectively, p 0.03), no other differences between these groups were found. Comparing SAI and PAI, only fever prevalence was significantly higher in SAI vs PAI (60.5% vs 26.3% respectively, p 0.02).

The prevalence of HS-symptoms was equal in AI and controls (12.5%). Even considering stricter criteria, as defined in methods, the prevalence of HS-symptoms resulted once more similar (10.8 % in AI and 8.9 % in controls, p >0.99). A higher prevalence of HS-symptoms was observed in SAI than in PAI group, although not statistically significant (14.2% and 6.7%, respectively, p 0.19). Contacts with verified COVID-A c c e p t e d M a n u s c r i p t 11 19 cases was also similar in AI and controls. HS-symptoms duration was unexpectedly longer in controls than AI (median 15 vs 7 respectively, p 0.04).

In 31.3% symptomatic AI patients, the replacement therapy dose was increased, especially in PAI group 

Adrenal insufficiency has been shown to be associated with increased infection risk and immunological alterations (23, 24, 36) . These data mainly come from studies involving patients with PAI while no consistent data are available in SAI. It is well known that infections also represent the most frequent trigger for adrenal crisis which primarily contributes to the increased mortality rate reported in this group (25) . Recently, some experts have remarked about the possible risk of COVID-19 contagion and complications in hypoadrenal patients, but the impact of the virus in this group has not been proved yet.

Our specific investigation in this group pointed out some useful information.

According to our data, hypoadrenal patients appeared not to be exposed to a significantly higher risk of manifest COVID-19 typical symptoms than controls even if the exposure risk, based on contacts with infected people and working habits, seemed equal. Also, they did not display severe disease complications, without requiring hospital admissions. Along with that, no adrenal crisis was reported. No differences, even in mortality, have been found. Only one SAI patient died, but a relationship with SARS-CoV-2 infection was unlikely.

The study group included exclusively patients who were on active follow up in a dedicated tertiary endocrinological centre and thus, they have been highly instructed on the possible risks related to infections' complications and on how to manage their own replacement therapy, if necessary. These characteristics may have contributed to the similar clinical outcome we found in AI and controls.

A c c e p t e d M a n u s c r i p t 13 Epidemiological studies on COVID-19 in Italy reported a peak of notified cases in March which overlaps with the peak of suggestive symptoms we observed in both AI group and controls in that month (34) . The other peak observed in former February, encountered in controls only, might be related to the seasonal flu virus peak (35) . The higher number of flu shots in AI patients could explain this difference.

Unexpectedly we observed a longer duration of HS-symptoms in controls vs AI patients along with a higher prevalence of gastrointestinal symptoms. As previously stated, it is possible that AI patients, who are highly instructed on infectious diseases complications, adopted a safer behaviour, limiting their social contacts and applying better medical managing of virus-related symptoms. Moreover, we could speculate that the high prevalence of gastrointestinal symptoms observed in controls, which is in accordance to the 50% reported in recent studies, may be related to a release of the cytokine in the GI tract, a mechanism that could be mitigated by the steroids and by the inefficient innate immune response in IA patients (37) .

The main study limitation is represented by the lack of certain diagnostic data on SARS-CoV-2 detection.

According to sanitary regulations in Lombardy, nasopharyngeal swabs were performed in specific situations only, namely hospital admissions. The method of data collection, fully based on what patients reported on phone calls, is an additional important study limit. On the other hand, the major strengths are represented by the large number of AI patients recruited, the presence of a control population, and the absence of subjects who did not respond to the call, thus excluding the possibility to have missed hospitalized COVID-positive or dead patients. Future investigations, like serum SARS-CoV-2 antibodies evaluation, will be necessary to clarify the real infection rate in AI patients. A second limitation in our assessment is the incomplete knowledge base at the time of the interview that determined the questions we asked. However, except for cutaneous manifestations (39, 40) , all other signs and symptoms resulting today associated with COVID-19 were included (41) . Lastly, we highlight that we decided to consider AI patients as a whole group in our main case-control analyses because, according to our statistical power analyses, the number of patients with PAI was limited.

A c c e p t e d M a n u s c r i p t 14 From our data, we can conclude that at the time of COVID-19 pandemic, patients with adrenal insufficiency who are adequately informed on infections related risks and who are adequately trained on replacement therapy self-adjustment, do not appear to be exposed to an increased risk of developing a more severe clinical impairment than controls.

Some datasets generated and analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

A c c e p t e d M a n u s c r i p t M a n u s c r i p t 19 Legends Table 1 Age is expressed as mean and range while symptoms duration as median and IQR (interquartile range).

PAI=primary adrenal insufficiency; SAI=secondary adrenal insufficiency; AI=adrenal insufficiency; p 1 =pvalue PAI vs SAI; p 2 =p-value AI vs controls; CO=cortone acetate; HC=hydrocortisone; m-HC=modified release hydrocortisone; Symptomatic patients=patients presenting at least one suggestive symptom of viral infection; HS-symptoms=highly suggestive symptoms of COVID-19 (at least two including fever or cough); Contacts=direct contacts with infected people; * = p<0.05.

This Figure shows the distribution of symptomatic patients over time, from January to April 2020. We expressed, as a percentage, the ratio between patients who complained symptoms in a specific period and the total of symptomatic patients, in each group. Most patients, both in AI and controls, complained their suggestive symptoms during COVID-19 outbreak, especially on March 2020. AI, patients with adrenal insufficiency; PAI, primary adrenal insufficiency; SAI, secondary adrenal insufficiency. 

Coronaviridae Study Group of the International Committee on Taxonomy of Viruses. The species Severe acute respiratory syndrome-related coronavirus: classifying 2019-nCoV and naming it SARS-CoV-2

WHO Director-General's opening remarks at the media briefing on COVID-19. 2020. Available at

and their implications on cancer prevention, diagnosis and management

Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China

A familial cluster of pneumonia associated with the 2019 novel coronavirus indicating person-to-person transmission: a study of a family cluster

-nCoV Persons Under Investigation Team, 2019-CoV Persons Under Investigation Team. Persons Evaluated for 2019 Novel Coronavirus -United States

Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study

Clinical characteristics of novel coronavirus cases in tertiary hospitals in Hubei Province

Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention

Cancer patients in SARS-CoV-2 infection: a nationwide analysis in China

Obesity in patients younger than 60 years is a risk factor for Covid-19 hospital admission

Cardiovascular Disease, Drug Therapy, and Mortality in Covid-19

Preliminary Estimates of the Prevalence of Selected Underlying Health Conditions Among Patients with Coronavirus Disease 2019 -United States

People who are at higher risk for severe illness

Italian Group for the Study of Inflammatory Bowel Disease (IG-IBD). Outcomes of COVID-19 in 79 patients with IBD in Italy: an IG-IBD study

Adrenal insufficiency

COVID-19 and endocrine diseases. A statement from the

Endocrinology in the time of COVID-19

Increased Infection Risk in Addison's Disease and Congenital Adrenal Hyperplasia

Effect of once-daily, modified-release hydrocortisone versus standard glucocorticoid therapy on metabolism and innate immunity in patients with adrenal insufficiency (DREAM): a single-blind, randomised controlled trial

on behalf of the EU-AIR Investigators. Mortality data from the European Adrenal Insufficiency Registry-Patient characterization and associations

COVID-19 infection and glucocorticoids: update from the Italian Society of Endocrinology Expert Opinion on steroid replacement in adrenal insufficiency

Use of glucocorticoids in patients with adrenal insufficiency and COVID-19 infection

Multisystem Morbidity and Mortality in Cushing's Syndrome: A Cohort Study

Androgen sensitivity gateway to COVID -19 disease severity

Androgen hazards with COVID-19

SARS-CoV-2, Testosterone and frailty in males (PROTEGGIMI): A multidimensional research project

Diagnosis and Treatment of Primary Adrenal Insufficiency: An Endocrine Society Clinical Practice Guideline

Corticotropin Tests for Hypothalamic-Pituitary-Adrenal Insufficiency: A Metaanalysis

COVID-19 in Italy: impact of containment measures and prevalence estimates of infection in the general population

World Health Organization -Regional Office for Europe -Flu news -Situation update 24

Primary adrenal insufficiency is associated with impaired natural killer cell function: a potential link to increased mortality

Clinical Insights into the Gastrointestinal Manifestations of COVID-19

for the COVID-19 Lombardy ICU Network

Patients Infected With SARS-CoV-2 Admitted to ICUs of the Lombardy Region, Italy

A dermatologic manifestation of COVID-19: Transient livedo reticularis

Cutaneous manifestations in COVID-19: a first perspective

WHO health topics -coronavirus. 2020. Available at