key: cord-0989395-8g3j83qi
authors: Baughman, Robert P.; Lower, Elyse E.; Buchanan, Mindy; Rottoli, Paola; Drent, Marjolein; Sellares, Jacobo; Terwiel, Michelle; Elfferich, Marjon; Francesqui, Joel; Barriuso Cabrerizo, María Rita; Sweiss, Nadera; Martone, Filippo; Al-Hakim, Tamara; Judson, Marc A.
title: Risk and outcome of COVID-19 infection in sarcoidosis patients: results of a self-reporting questionnaire
date: 2020-12-16
journal: Sarcoidosis Vasc Diffuse Lung Dis
DOI: 10.36141/svdld.v37i4.10726
sha: 17a8b6151402514481e414b06168795dec582f00
doc_id: 989395
cord_uid: 8g3j83qi

BACKGROUND: It has been suggested that sarcoidosis patients, especially those on immunosuppressive medications, are at increased risk for COVID-19 infection and more severe disease. METHODS: A questionnaire was developed in four languages (English, Dutch, Italian, and Spanish). The questionnaire queried whether patients had been infected with COVID-19 and outcome of the infection. Risk factors for COVID-19 infection were collected. RESULTS: A total of 5200 sarcoidosis patients completed the questionnaire with 116 (2.23%) reporting infection and 18 (15.8%) required hospitalization. Increased hazard ratio (HR) for COVID-19 infection were seen for those with a COVID-19 infected roommate (HR=27.44, p<0.0001), health care provider (HR=2.4, p=0.0001), pulmonary sarcoidosis (HR=2.48, p=0.001), neurosarcoidosis (HR=2.02, p<0.01), or rituximab treatment (HR=5.40, p<0.0001). A higher rate of hospitalization was found for those with underlying heart disease (HR=3.19 (1.297-7.855), p<0.02). No other feature including race, other immunosuppressive agent, age, or underlying condition was associated with a significant increased risk for infection or more severe disease. CONCLUSION: The overall rate of COVID-19 was 2.23%, suggesting an increased rate of COVID-19 infection. However, when an analysis of the questionnaires of sarcoidosis and non-sarcoidosis patients was performed in one localized area over this time period, the rate of COVID-19 infection was similar in both groups. Sarcoidosis patients who cohabitated with COVID-19 infected individuals, worked in health care, had pulmonary or neurologic sarcoidosis, or were treated with rituximab had an increased risk for COVID-19 infection. No significant increased risk for hospitalization could be identified based on age, race, gender or any specific immunosuppressive treatment. (Sarcoidosis Vasc Diffuse Lung Dis 2020; 37 (4): e2020009)

The COVID-19 pandemic has dramatically changed and challenged the practice of medicine. Both sarcoidosis patients and their health care providers are concerned that sarcoidosis may increase the risk of contracting COVID-19 and may be associated with poor outcomes from COVID-19 infection (1) . Sarcoidosis patients may have several risk factors associated with an increased rate and a poor outcome from COVID-19 infection including underlying lung disease and the use of glucocorticoids and other immunosuppressive agents (2;3) . However, it is not clear that use of immunosuppressive therapy alone is a risk factor for increased rate of 5) . A worse outcome from COVID-19 infection may be result of the common presence of other co-morbidities including heart disease, diabetes, and hypertension (2;6-8) . Because of the potential risk of COVID-19 infection in those receiving immunosuppressive medications, sarcoidosis experts have proposed modifying the treatment of sarcoidosis (9). However, these recommendations were based almost exclusively on expert opinion and extrapolation from other medical conditions because of the absence of sarcoidosis-specific outcome data concerning COVID-19 infection.

Because of the lack of specific information concerning risk of acquiring COVID-19 infection and its outcomes in sarcoidosis patients, we initiated an IRB approved questionnaire survey of sarcoidosis patients between April and July 2020 to investigate the prevalence of COVID-19 infection, clinical outcomes and possible risk factors for contracting in several sarcoidosis cohorts across several countries. The questionnaire was distributed through several platforms and was available in four languages: English, Dutch, Italian, and Spanish. Five versions of the questionnaire that were minimally different (vide infra) were distributed to various cohorts. One version of the questionnaire was distributed to sarcoidosis patients who participated in a previously described registry (10) . In order to compare these findings with another high-risk group, we also surveyed a cohort of breast cancer patients during this time period. The results of these five questionnaires were pooled and analyzed to provide data concerning the frequency, severity, potential risk factors and outcomes for COVID-19 infection in sarcoidosis patients.

A questionnaire regarding COVID-19 infection was developed by three of the authors (RPB, EEL, and MAJ). The questionnaire was approved by the University of Cincinnati Institutional Review Board and is shown in Supplement S-1. IRB approval for the FSR questionnaire was obtained from Advarra (Columbus, MD), where the registry number for the FSR Registry is Pro00008556 and modification number for this sub-study is: MOD00087736.

The questionnaire queried whether patients had been infected with COVID-19. If infected, they were asked to provide details regarding level of care (treated at home, hospital, or intensive care unit). They were also queried concerning risk factors for COVID-19 infection including household contacts and whether they were health care providers. The patients were asked if they had underlying medical conditions associated with increased risk for COVID-19 infection. They were also asked to provide information regarding their age, sex, and selfdeclared race. They provided their current residence, including their state for those living in the United States. Patients with sarcoidosis were queried about the duration of disease, specific organ involvement, and current and past immunosuppressive therapy. The questionnaire responses were collected in an anonymous manner with no patient identifiers captured. At time of completing the questionnaire, patients were asked to provide consent for use of their questionnaire responses. The distribution was meant to reach as wide an audience as possible. Patients were enrolled without incentives, since the survey was anonymous. Respondents were asked to complete the questionnaire even if they did not have any symptoms related to COVID-19 infection. There was no restriction for patients to complete more than one questionnaire.

A total of five questionnaires were distributed (Supplement S-1 to S-4). Table 1 summarizes the features of the five questionnaires. All were based on the University of Cincinnati/Albany Medical Center (UC/AMC) questionnaire. The same questionnaire was distributed to those patients who were Foundation for Sarcoidosis Research (FSR) registry. The questionnaire was translated into Dutch and distributes by the Dutch Sarcoidosis Society (Sarcoidose. nl) and the ild care foundation. The Italian Asso-ciation for Sarcoidosis patients (ACSI) prepared the Italian version of the COVID-19 questionnaire. The Spanish association of patients with sarcoidosis (ANES-Asociación Española de) prepared a Spanish version of the COVID-19 questionnaire.

The hazard ratio (HR) with 95% confidence interval (CI) was calculated for various factors for each questionnaire using a statistical software package (MedCalc Software limited, Ostend, Belgium). For those questions which were identical (except for language), the results were summed. HR were calculated for the individual questionnaires and for summary data when available. A p value of less than 0.05 was considered significant.

Supplement Figures S1-S5 show the numbers of patients who completed the questionnaire for each site, including those who did not give consent for final analysis or were excluded for other reasons. Table 2 summarizes the values for 5200 sarcoidosis patients who were analyzed. A total of 116 (2.23%) reported COVID-19 infection. The overall rate of COVID-19 infection is shown for each site and ranged from 0.8 to 4.76%. As shown in the table, there was no significant difference in the rate of infection based on sex, race, or age. In the Dutch questionnaire an additional question revealed that 96 out of 973 COVID-19 negative patients (10%) reported COVID-19 symptoms but were not tested.

The results of the individual questionnaires are provided in Supplement S5-S9. Table 3 summarizes For sarcoidosis immunosuppressive therapy, rituximab treatment was associated with an increased risk of COVID-19 infection (HR=5.40 (2.621-11.123), p<0.0001). Only the UC/AMC and FSR included more than one patient treated with rituximab. However, both of these identified an increased risk with rituximab use. There was no significant increase in risk for COVID-19 infection for those on any other immunosuppressive therapy. For those receiving prednisone, there was no increased risk for patients prescribed 10 mg or more a day versus a lower dose. A total of 425 patients were prescribed hydroxychloroquine. There was no difference in the risk for COVID-19 among the other questionnaires or for the summary data of all sarcoidosis patients compared to all others. In the Dutch questionnaire, patients who reported to receive any medication for sarcoidosis had a decreased COVID-19 risk (HR=0.40, 0.165 to 0.958, P<0.05). There was no significant difference for the larger UC/AMC questionnaire (HR=1.49, 0.916 to 2.437, p>0.10) or for the combined data.

There was no increased risk with any of the associated with age, race, sex, duration of disease, or the comorbidities investigated. In the Dutch questionnaire, there was no difference in HR for those with a BMI above 25.

There was an increased risk for sarcoidosis patients with lung involvement (HR=2.48 (1.446-4.249 ), p=0.001). However, the FSR was the only individual questionnaire which identified this as a significant risk. The summary identified neurosarcoidosis as an increased risk factor for COVID-19 infection (HR=2.02 (1.234-3.307 ), p<0.01). For neurosarcoidosis, there was a significant HR seen for the Spanish and FSR questionnaire and borderline for UC/AMC (p=0.0619).

Most patients responding to the UC/AMC questionnaire provided their country of residence or state of residence if they lived in the United States, and the percentage of patients with COVID-19 infection by residence is shown in Table S-10. There was no significant difference in rates of infections between the United States and non-United States residents. For the United States, there was a wide range but not a significantly different rate of infection between states.

For the UC/AMC questionnaire, we analyzed the risk of COVID-19 infection for two groups of patients seen by either EEL or RPB at the University of Cincinnati from April1 to June 30, 2020. During that time, 547 cancer patients were seen at University of Cincinnati (UC cancer). Two (0.37%) reported COVID-19 infection. Only 3 of 541 (0.55%) sarcoidosis patients seen at the UC sarcoidosis clinic during this time reported COVID-19 infection. The hazard ratio for COVID-19 infection in sarcoidosis patients at the University of Cincinnati was not significantly higher than those with cancer (HR=1.52, 0.254 to 9.041, P>0.10). Table 4 shows the reported rate of hospitalization for those with COVID-19 infection for each questionnaire and for the total group. A mean of 15.8% (range 13 to 27%) of COVID-19 infected patients were hospitalized with about one-third cared for in the intensive care unit at some time during hospitalization.

For 105 of the COVID-19 infected patients we had information regarding current immunosuppressive therapy. Nineteen (18.1%) of these were hospitalized. Table 5 summarizes these outcomes and calculates the HR for hospitalization for various immunosuppressive therapies. There were no significant HR with any specific therapy. For the 96 COVID-19 infected patients from three questionnaires (UC/AMC, FSR, and Dutch), we were able to analyze the clinical outcome versus presence of diabetes, hypertension, COPD, and heart disease. Table 6 shows the results of this analysis. Only those with underlying heart disease had an increased rate of hospitalization (HR=3.19, 1.297-7.855, p<0.02).

In this analysis of five surveys of sarcoidosis patients from the USA and Europe concerning COV-ID-19 infection, we found evidence that the rate of COVID-19 infection in sarcoidosis patients was higher than in the general population. We found that sarcoidosis patients who were healthcare workers or were living with a person infected with COVID were at higher risk of COVID infection; these data have been reported in the general population (12;13) and support the well-established fact that COVID-19 is highly infectious and is easily transmitted to individuals who are in close proximity to an actively infected person (14) . In regard to immunosuppressive therapy, only rituximab was associated with in-creased risk for COVID-19 infection in sarcoidosis patients. The lack of association of prednisone use with the development of COVID-19 infection held even when comparing > 10 versus < 10 mg/day. In sarcoidosis patients with diabetes, hypertension, heart disease, and co-existing chronic obstructive pulmonary disease no increased risk of acquiring COVID-19 infection was identified, even though these comorbidities have also been identified as risk factors for COVID-19 infection (6;8;15) .

The current data surveyed sarcoidosis patients mostly in United States and Europe from April through July 2020. This time frame co-indices with the onset of the pandemic in these two parts of the world. In our analysis of a questionnaire administered to 5200 sarcoidosis patients, 2.23% or 22,308 cases per million had become infected with COVID-19 during this time. During the time period in which this questionnaire was administered, the number of confirmed cases of COVID-19 in the United States was estimated as 1,060 per million (https://en.wikipedia. org/wiki/Template:COVID-19_pandemic_data/ United_States_medical_cases). The same site reported that the cumulative rate of COVID-19 infection in Spain was 5197 per 1 million, Italy 3853 per 1 million, and Netherlands 3141 per 1 million. How- These data suggest that the rate of COVID infection is higher in sarcoidosis patients than the general population. However, the rate of COVID-19 infection varies by the time period as well as from country to country and even within the United States. It is therefore possible, that the rate of COVID-19 in sarcoidosis patients was no different from the general population. A significant number of the sarcoidosis patients completing the UC/AMC questionnaire were seen at the University of Cincinnati Sarcoidosis Clinic, which is in southwestern Ohio. As of July 1, 2020, the reported cumulative rate of COVID-19 infection for this area was 5103 per 1 million (https:// coronavirus.ohio.gov/wps/portal/gov/covid-19/dashboards/overview). This is similar to the rate of 5545 per 1 million for our sarcoidosis patients and 3656 per 1 million for the cancer patients. The lower rate of COVID-19 infections for sarcoidosis patient at University of Cincinnati clinic versus other patients completing the questionnaires may be due to lower overall rate of COVID-19 infection in the area and/or the more rigorous criteria for diagnosis, since at our clinic we required verification by culture. This criterion may underestimate the number of cases of COVID-19 infection (16). The hospitalization rates for COVID-19 may be more accurate, since such cases are usually confirmed by cultures.

For sarcoidosis patients, we identified five features associated with increased for COVID-19 infection. Sarcoidosis patients with a COVID-19 infected roommate had a greater than 20-fold increased risk for COVID-19 infection. In one meta-analysis, the risk of in home transmission of disease has been estimated as ten-fold (17). There was also a nearly two-fold increased risk for sarcoidosis patients who were health care workers. This increased risk has been noted for some time (12). The higher risk in Spain may be a reflection that some areas were hit sooner than other parts of the world. The widespread use of N-95 and other respiratory policies later in the pandemic and in other parts of the world may have blunted this risk factor (13;18). Comorbidities in sarcoidosis include diabetes, hypertension, heart disease, and co-existing chronic obstructive pulmonary disease (10;19) . These have also been identified as risk factors for COVID-19 infection and more severe disease (2;6;8;15) . However, in the current study, none of these was associated with an increased risk for infection.

In regards to immunosuppression therapy, only rituximab was associated with increased risk for COVID-19 infection for sarcoidosis patients. This is not surprising, since rituximab has been noted to have increased the risk for acquiring viral infections (20). In addition, viral infections are more severe when patients are receiving rituximab. We studied only seven patients with COVID-19 infection treated with rituximab.

Sarcoidosis is a multi-organ disease and the effect of sarcoidosis on different organs may affect the patient's ability to avoid COVID-19 infection. In this study, we found that patients with lung or neurologic involvement were at increased risk for COVID-19 infection. Chronic lung disease has been identified as a risk factor for infection and more severe disease (21).

In terms of the outcomes of COVID-19 infected sarcoidosis patients, we found that less than twenty percent required hospitalization. In a recent, retrospective study of 37 sarcoidosis patients with COVID-19 infection, the rate of hospitalization for infection was 60% and no different from the nonsarcoidosis patients seen at that center (3) . However, the rate of adverse outcome as defined by requiring intubation and/or mortality was significantly higher than the non-sarcoidosis patients (3) . The current study would have missed the very severe cases, at least the ones who died, as it was a study analyzing a self-reported questionnaire. In our study, the use of immunosuppressive therapy was not associated with a significant increased risk for hospitalization. The overall outcome of these patients appears more favorable than that reported in rheumatoid arthritis patients treated with immunosuppression (22-24). In one study (23), the use of > 10 mg per day of prednisone or its equivalent was associated with increased risk. That study analyzed 600 COVID-19 infected patients and our study may have been underpowered to detect that difference. Interestingly, that study found that anti-TNF therapy was associated with a significantly lower risk for COVID-19 infection (23).

Hydroxychloroquine has been proposed as a potential therapy for patients with COVID-19 infection and ongoing studies are evaluating this drug (25). Over 400 of our sarcoidosis patients were receiving hydroxychloroquine at time of survey. There was no change in the rate of infection or rate of hospitalization for the seven patients who developed COVID-19 infection while on hydroxychloroquine. This has also been noted in a study of patients treated with hydroxychloroquine for various rheumatologic conditions (23).

Several comorbidities have been associated with a worse clinical outcome from COVID-19 infection (7;8;21;23). We were able to analyze the outcome of 96 sarcoidosis patients with COVID-19 infection and reported comorbidities. Underlying heart disease was associated with an increased risk for hospitalization. Other comorbidities examined included diabetes, COPD, and hypertension were not felt to be significant risk factors.

There are several limitations to our study. The questionnaire did not try to quantitate severity of disease, especially significant pulmonary fibrosis. Therefore, we could not comment on impact of severe lung disease on risk or outcome of COVID-19 infection. Because of the low number of incident cases, we may have been underpowered to detect smaller, but significant risk factors including comorbidities and the impact of immunosuppression therapy. The questionnaires were completed by the patients, usually on-line. With the exception of those seen at the University of Cincinnati, there was no attempt to verify COV-ID-19 infection. Patients who had severe disease or even died would be unlikely to be able to complete the questionnaire, so this group was underrepresented. Also, the rate of COVID-19 infection in sarcoidosis patients varies based on local conditions and time into the pandemic. The rate of COVID-19 infection in the general population during the study period may be better understood over time, especially as serologic testing becomes more widely used. However, the rate of infection for COVID-19 infection in the general population may be higher. We compared our results to a standard reporting site which provided cumulative rates for various parts of the world. In the sub-study at University of Cincinnati, the rate of infection was not significantly different from cancer patients seen at the same time period. Future reports may provide a better understanding of the rate of COVID-19 infection in the community studied. Future rates may be affected as vaccines become available. Vaccines for other conditions have proved to be effective in preventing infections in sarcoidosis patients (26).

In summary, our data suggests an increased rate of COVID-19 infection in sarcoidosis patients. However, when compared to non-sarcoidosis patients in the same area and time of the study, the rate of COVID-19 infection was not significantly different. The most obvious risk factor for COVID-19 infection was having a roommate with COVID-19. This means that hygiene measures and distancing are extremely important at home as well as in public. To facilitate research on prevalence and risk factors of COVID-19 infection in chronic diseases, sarcoidosis specifically, it would be helpful to report in population registries not only numbers of patients with COVID-19, but also their characteristics, such as comorbidities and medication use. (3) 

Patients with interstitial lung disease and pulmonary sarcoidosis are at high risk for severe illness related to COVID-19

Epidemiology of COVID-19: A systematic review and meta-analysis of clinical characteristics, risk factors, and outcomes

Moderate or Severe Impairment in Pulmonary Function is Associated with Mortality in Sarcoidosis Patients Infected with SARS-CoV-2

Door deze vragenlijst in te vullen, stemt u ermee in dat de gegevens anoniem verwerkt worden. Een samenvatting van het onderzoek zal, op basis van anonieme gegevens

• Weet het niet zeker, ben niet getest, maar heb wel symptomen

Zo ja, wat was er nodig voor de behandeling? (vink aan wat van toepassing is) : • Kon thuis worden behandeld • Ben opgenomen geweest in een ziekenhuis, maar ben niet op de intensive care verpleegd • Ben opgenomen geweest op een intensive care afdeling • Ben beademd met een ventilator

Zijn er mensen die in hetzelfde huis/ appartement wonen als u, die besmet zijn geweest met COVID-19

Hoe lang heeft u al sarcoïdose? (jaren, maanden)

Ik heb sarcoïdose in de volgende organen/delen van mijn lichaam: (vink aan wat van toepassing is

Nadat ik COVID-19 heb gehad, zijn mijn klachten: • verergerd • minder geworden dan voorheen • hetzelfde gebleven • niet van toepassing

Wat is uw gewicht?

Bent u in de afgelopen 3 maanden behandeld voor sarcoïdose met: (vink aan wat van toepassing is, meerdere antwoorden mogelijk)

• Prednison/prednisolon (graag de huidige dagelijkse dosis onderaan deze vraag invullen)

Remicade/Renflexis/Inflectra)

Plaquenil)/Chloroquine • Rituximab (Rituxan)

We would like to thank Sarcoidose.nl, the Dutch Sarcoidosis Patient Society and the ild care foundation for preparing and supplying the Dutch version of the questionnaire. For the English version, we thank sarcoidosis patients who participated through the Foundation for Sarcoidosis Research. For the Italian questionnaire, we thank the Amici Contro la Sarcoidosi Italia ONLUS. For the Spanish questionnaire, we thank the Asociación de Enfermos de Sarcoidosis.