Conference Coverage

Lupus may confer higher risk of death from COVID-19


There is a significantly increased risk for acute respiratory distress syndrome (ARDS)–related death from COVID-19 among people with systemic lupus erythematous (SLE), compared with the general population, according to data collected in Brazil in 2020.

“Special care is therefore necessary for these patients, as well as reinforcement of the importance of preventive measures during a pandemic for this population,” said Eloisa Bonfá, MD, PhD, at the 14th International Congress on Systemic Lupus Erythematosus, which was held together with the 6th International Congress on Controversies in Rheumatology and Autoimmunity.

“We know that lupus patients have an increased susceptibility to infections due to autoimmune dysregulation and use of immunosuppressive therapy,” explained Dr. Bonfá, who is clinical director of the largest tertiary referral center for autoimmune rheumatic diseases in Latin America, the University of São Paulo Faculty of Medicine Hospital Clinics.

“Our study demonstrates for the first time that lupus patients have an increased ARDS severity,” she added.

Prior to the meeting, the study was published in ACR Open Rheumatology.

Collating the evidence

Since the COVID-19 pandemic began, there have been more than 20 million confirmed cases of SARS-CoV-2 infection in Brazil and more than half a million deaths.

Dr. Bonfá presented the results of a cross-sectional study that was part of the country’s national Influenza Epidemiological Reporting Surveillance System. Data from 2020 were used, which included just over 252,000 individuals who had polymerase chain reaction–confirmed SARS-CoV-2 infection. Of these individuals, there were 319 consecutively recruited patients with SLE.

The aim was to look at the effect of being hospitalized for COVID-19–related ARDS on outcomes in people with SLE versus the general population.

ARDS was defined as a positive polymerase chain reaction test and accompanying flu-like symptoms with dyspnea, respiratory discomfort, persistent pressure in the chest, or desaturation less than 95% in room air or having a bluish tinge to the lips or face.

Other telling signs of a serious respiratory infection that were evaluated, but not mandatory for study eligibility, were loss of smell, impaired taste, typical CT findings, or having had contact with a confirmed COVID-19 case in the preceding 2 weeks.

Key findings

The risk for death from COVID-19–related ARDS was “more than double” in patients with SLE, compared with the general population, Dr. Bonfá reported. The relative risk in the fully adjusted, propensity-scored analysis was approximately 2.25.

That analysis did not account for other comorbidities but was fully adjusted for individuals’ age, sex, and region of Brazil where they lived. The latter was important, Dr. Bonfá said, because “we have a high disparity regarding health access and treatment among regions.”

Comorbidities considered as part of the analyses included arterial hypertension, diabetes, malignancies, neurologic disease, and diseases affecting the heart, lung, liver, and kidneys. Researchers also adjusted for smoking, alcohol intake, body weight, pregnancy, and transplantation.

SLE had a greater impact on individuals’ outcomes than all other comorbidities considered.

“We evaluated lupus as one comorbidity compared to all other comorbidities,” Dr. Bonfá explained.

SLE “more than doubled the chances” of dying from ARDS, she said. “This is [a] very impressive finding.”

They found that SLE was associated with an RR for death of 1.73, compared with non-SLE patients, when propensity-score matching without adjustment for comorbidities was used. The RR for death dropped to 1.40 but was still significant when researchers included comorbidities.

Dr. Bonfá and her team also looked at a combined endpoint of death, ICU admission, and need for mechanical ventilation. They found an increased risk in patients with SLE versus the general population in all their analyses, ranging from 1.70 if comorbidities were included in the model to 1.27 if they weren’t to 1.39 if propensity-score matching alone was used.


Next Article: