Our observational study compared real-world outcomes with baricitinib versus control for treatment in patients hospitalized with severe COVID-19 on IMV in a propensity score-matched and retrospective cohort. The study found high mortality among patients hospitalized with severe COVID-19 on IMV. The result suggested that baricitinib didn’t reduce 28-days mortality in severe COVID-19 patients on IMV.
One of the major strengths of our study was the PSM design. There are a few known risk factors for severe COVID-19, which could have impact on prognosis in a retrospective study. So far, age and comorbidities such as diabetes, COPD, chronic kidney disease, have been reported to be independent predictors of mortality for COVID-19 patients(3, 18–23). Vaccine has been shown to bring survival benefit in COVID-19, which can also influence the outcomes(24, 25). With the expanding coverage of COVID-19 vaccination, its influence on reducing mortality could not been overlooked. So the current study excluded patients with unknown vaccination status. Given these potentially confounding variables and the retrospective, observational nature of current study, PSM was applied to minimize bias. Therefore, the outcomes were assessed between two comparable groups, which increased validity of our study. Moreover in our study, the infection biomarkers such CRP, IL-6 and procalcitonin, were balanced between both groups, which suggested that the severity of infection was similar. So the balanced baseline demographics, comorbidity profile and severity of infection lend credibility to our findings.
Our study did not show a significant difference in the 28-day all-cause mortality between severe COVID-19 patients on IMV treated with baricitinib versus control. In less severe hospitalized COVID-19 patients, the benefit of baricitinib on reducing mortality was reported by numerous studies. It was roughly estimated that baricitinib or another JAK inhibitor was associated with a 20% proportional reduction in mortality(15). However, in severe COVID-19, there were discrepant results about efficacy of baricitinib among studies. On one hand, some studies found that baricitinib reduced mortality in severely ill patients. In an exploratory trial conducted in severe patients on IMV or ECMO, treatment with baricitinib compared with placebo reduced 28-day all-cause mortality by 46% and 60-day all-cause mortality by 44%(16). But this trial had a relatively small sample size. On the other hand, some studies showed that baricitinib brought no benefit on the mortality in severely ill patients. In the Bari-SolidAct study, no statistically significant difference was observed on 60-day mortality in hospitalized patients with severe/critical COVID-19 receiving either baricitinib or placebo(17). Of note, the trial was stopped before reaching planned sample size, so the results should be treated with caution. In subgroup analysis of RECOVERY study, baricitinib didn’t reduce mortality among patients on IMV(15). Our findings agreed with those reports, and didn’t support the use of baricitinib in severe COVID-19 patients on IMV. So the efficacy of baricitinib in severe COVID-19 was still controversial, more research was needed to draw a definite conclusion.
Despite the use of standard care, our study reported high mortality of 73.9% for baricitinib group and 84.1% for control group, which were much higher than reported in previous studies. In an exploratory trial conducted in severe patients, the overall 28-day all-cause mortality for those on IMV or ECMO at baseline was 39% in participants who received baricitinib and 58% in those who received placebo(16). The open-label RECOVERY study reported that 28-day mortality of 49% in participants who received tocilizumab while on IMV versus 51% in those who received standard of care(15). In one large meta-analysis, the case fatality rate of 45% was reported for patients with severe COVID-19 requiring IMV(16). This discrepancy in mortality may be explained by the following reasons. First, our study included older patients than previous studies. In our study, the mean age was 78.80 ± 9.04 years for baricitinib group and 82.57 ± 9.27 years for control group. The mean participant age was 58·6 ± 13·8 years in the above-mentioned exploratory trial, and 58·1 ± 15·5 years in the RECOVERY study. Studies have shown that older age is a risk factor for fatality related to COVID-19(3, 26). Second, our study had a large proportion of patients with comorbidities, especially heart failure (41.3% − 52.3%) and renal insufficiency (41.3% − 45.5%). In the RECOVERY study, only about 18%-19% of patients had heart disease and about 2% had severe kidney impairment(15). The discrepancy indicated that the patients in our study were in poorer health state, which may contribute to a worse prognosis and high mortality. Third, delayed use of antiviral drugs may also contribute to high morbidity. Most of our patients received antiviral drugs after intubation instead of in early phase of COVID-19, so maybe it was too late. It was well-known that early use of antiviral drugs lead to better prognosis(5). The high mortality in our study may partially explained why baricitinib failed to bring survival benefit.
Except for the standard of care therapies, the additional treatments such as antiviral drugs, tocilizumab, or ECMO were left at the discretion of the physicians in the study. During the surge there was a national drug shortage, which made the use of additional treatments variable among patients. Our study found that patients in the baricitinib group were more likely to receive nirmatrelvir/ritonavir and IVIG. Nirmatrelvir/ritonavir could reduce risk of progression to severe COVID-19 and mortality for non-hospitalized adults, and its efficacy on severe COVID-19 was unknown(27, 28). In the current study, most of patients received nirmatrelvir/ritonavir after intubated and admitted to ICU instead of within five days of the onset of symptoms, which was not supported by evidence. The efficacy of delayed use of nirmatrelvir/ritonavir complicated matters even further. Regarding IVIG, available data do not support its use in severe COVID-19. In a multicenter, double-blind, placebo-controlled, phase 3 trial of patients with moderate-to-severe COVID-19-associated acute respiratory distress syndrome, IVIG did not improve clinical outcomes at day 28 and tended to be associated with an increased frequency of serious adverse events(29). Two retrospective studies reported similar results and found IVIG was not associated with significant changes in mortality in severe COVID-19 patients(30, 31). In conclusion, although published studies showed that neither nirmatrelvir/ritonavir nor IVIG reduce mortality in severe COVID-19, their effect on outcome were difficult to account for in a retrospective study design.
Due to PSM, most comorbidities were balanced between groups except hypertension. Our study reported that baricitinib group had a higher proportion of patients with hypertension. Hypertension has been identified as the most prevalent cardiovascular comorbidity in patients infected with COVID-19(20). So far, the evidence had been mixed about the association between hypertension and COVID-19 prognosis(32–34). In a study conducted in Italy, hypertension was not an independent predictor of mortality in the multivariate analysis(23). Another European registry study, which included more than 9000 patients, reported that hypertension was not independently related with in-hospital mortality(35). On the other hand, a large-scale analysis in China showed that hypertension was a significant risk factor for poor outcomes including admission to an intensive care unit, invasive ventilation, or death(20). A meta-analysis found that hypertension acted as an independent risk factor for deterioration of COVID-19(36). So inconsistent results about association between hypertension and COVID-19 prognosis make the interpretation of our findings troublesome. The retrospective design of our study further made it difficult to clarify the impact of hypertension on the outcome.
The current study also found there were disagreement about CK and CK-MB in patients. Compared to baricitinib group, the control group had higher level of CK-MB and similar levels of CK. Previous results showed that the cardiovascular system of the COVID-19 patients had been notably damaged, and the degree of damage could be evaluated by cardiac biomarkers such as CK and CK-MB. Most studies showed that the elevated levels of CK and CK- MB were significantly associated with an increased risk of the mortality in COVID-19 infected patients(37). For example, in a retrospective analysis including 2954 COVID-19 patients, those with higher CK-MB had a significantly higher mortality rate compared to patients with normal levels(38). Two meta-analysis revealed that higher CK and CK-MB were associated with the mortality and severe disease in COVID-19 patients(39, 40). But there were some studies reporting conflicting results as well. A study found CK-MB and CK had no significant difference in the prediction effect of the mortality in COVID-19(41). The disagreement of CK and CK-MB found in our study made it difficult to clarify whether patients in the control group had more severe cardiac damage than patients in the baricitinib group. As a result, the impact of cardiac damage on the outcome could be evaluated neither.
Some limitations of the study merit consideration. First, the present study was a retrospective study, so there were inherent problems related to this design. Second, we did not have data about the specific variant or sub-variant of each patient in our cohort. Third, some disparities in baseline characteristics and concomitant treatment were present which may have influenced the outcome parameters.