key: cord-0739129-ses5i6mh authors: Ibe, Yuta; Ishigo, Tomoyuki; Fujii, Satoshi; Fujiya, Yoshihiro; Kuronuma, Koji; Tsugawa, Takeshi; Takahashi, Satoshi; Fukudo, Masahide title: Delayed dexamethasone treatment at initiation of oxygen supplementation for coronavirus disease 2019 is associated with the exacerbation of clinical condition date: 2022-03-18 journal: J Infect Chemother DOI: 10.1016/j.jiac.2022.03.007 sha: 4ac006aca56db0a0715715944a29b125f98ea74e doc_id: 739129 cord_uid: ses5i6mh INTRODUCTION: Coronavirus disease 2019 (COVID-19) frequently causes inflammatory lung injury as its symptoms progress. While dexamethasone reportedly reduces inflammation and prevents progression to respiratory failure, the appropriate time to administer dexamethasone in patients with COVID-19 remains unclear. METHODS: This was a single-center, retrospective cohort study, where we consecutively enrolled patients hospitalized with COVID-19 who received oxygen and oral dexamethasone (n = 85). We assessed the association between the number of days to the initiation of dexamethasone and the cumulative rate of exacerbation defined as death or initiation of mechanical ventilation within 28 days of symptom onset. RESULTS: The optimal cut-off value from the initiation of oxygen supplementation to that of dexamethasone administration was two days (sensitivity, 85%; specificity, 59%), whereas that from oxygen saturation (SpO(2)) < 95% to the initiation of dexamethasone administration was five days (sensitivity, 78%; specificity, 59%). adjusting for age, sex, body mass index, Charlson comorbidity index score, time of oxygen supplementation (two or more days), and SpO(2) < 95% (five or more days), Cox regression analysis results showed that delayed dexamethasone administration since the initiation of oxygen supplementation was significantly associated with a higher risk of death or greater need for mechanical ventilation (hazard ratio: 5.51, 95% confidence interval, 1.79–16.91). CONCLUSIONS: In patients with COVID-19 and hypoxemia, early administration of dexamethasone, preferably less than two days from initiation of oxygen supplementation, may be required to improve clinical outcomes. The new coronavirus infection, first confirmed in China toward the end of 2019, is still prevalent worldwide [1] . The characteristic symptom of coronavirus disease 2019 (COVID-19) is respiratory dysfunction, which can lead to fatal comorbidities, including acute respiratory distress syndrome (ARDS) [2] . In the Wuhan province of China, 67-85% of patients admitted to the intensive care unit (ICU) due to critical COVID-19 were diagnosed with ARDS, with a mortality rate of 61.5% [3, 4] . Similarly, high mortality rates have been reported in a recent study [5] . Therefore, in addition to the treatment of COVID-19, measures are needed to prevent the development of ARDS. One such treatment is the synthetic steroid dexamethasone [6] , which prevents progression to ARDS associated with cytokine storm by reducing inflammatory lung injury and preventing systemic organ damage [7] . Furthermore, dexamethasone was shown to reduce 28-day mortality in patients with COVID-19 receiving either oxygen supplementation or mechanical ventilation [8] . Nonetheless, the timing of dexamethasone administration in patients with COVID-19 remains controversial. On the one hand, since ARDS can cause self-perpetuating inflammation with subsequent loss of organ function, and is associated with high mortality levels, dexamethasone should be started as early as possible following the onset of ARDS [9] . On the other, studies have shown that early corticosteroid administration within seven days of symptom onset was associated with a subsequent increase in plasma viral load [10, 11] and exacerbated COVID-19 symptoms [12] . There are no clinical or laboratory indications for the initiation of dexamethasone administration, and the optimal timing of treatment remains unclear. The aim of this study was therefore to determine the optimal timing for initiating J o u r n a l P r e -p r o o f dexamethasone to prevent the exacerbation in patients with COVID-19. This was a single-center, retrospective cohort study. We included adult patients with positive reverse transcription-polymerase chain reaction or antigen test for severe acute respiratory syndrome coronavirus (SARS-CoV-2), who received oral dexamethasone and oxygen at Sapporo Medical University Hospital between October 1, 2020, and May 31, 2021. The following patients were excluded: those who did not require oxygen supplementation at dexamethasone treatment initiation, those who received high-dose methylprednisolone prior to dexamethasone initiation, and those under 18 years. Information concerning age, sex, body mass index (BMI), vital signs, oxygen saturation (SpO2), chest computed tomography (CT), blood biochemistry, urinalysis, coagulation function, C-reactive protein, comorbidities, and medications was obtained from patient files. The Charlson comorbidity index (CCI) was used to evaluate the severity of comorbidities [13, 14] . This study was approved by the Clinical Investigation Ethics Committee of Sapporo Medical University Hospital (Number 322-65). Oxygen supplementation was initiated if the SpO2 breathing room air dropped to ≤ 93%, according to the US Centers for Disease Control and Prevention definition of severe disease, or if the SpO2 was estimated to be ≤ 93% based on the chest CT findings and subjective symptoms of respiratory distress. The decision to initiate dexamethasone treatment was based J o u r n a l P r e -p r o o f as applicable. Differences in continuous variables were tested using the unpaired Student's ttest and Mann-Whitney U test for parametric and non-parametric data, respectively. The χ 2 test or Fisher's exact test was used to compare categorical variables, as appropriate. Receiver operating characteristic (ROC) analysis was used to determine the area under the curve (AUC). To predict the outcomes of death or need for mechanical ventilation within 28 days, the optimal cut-off value was calculated for the number of days from the onset of (a) symptoms, (b) oxygen supplementation, and (c) SpO2 < 95% until the initiation of dexamethasone administration, to predict the outcomes of death or need for mechanical ventilation within 28 days. The optimal cut-off value was determined based on the Youden index. Kaplan-Meier curves and log-rank tests were performed to assess the cumulative rates of death or mechanical ventilation from the initiation of dexamethasone treatment. Univariate and multivariate Cox regression analyses were performed to identify significant predictors across the dependent outcomes of interest. Explanatory variables were forced to incorporate the late treatment with dexamethasone in addition to age, sex, BMI, and CCI score. Statistical significance was set at P < 0.05. All analyses were performed using JMP ® 15 (SAS Institute Inc., Cary, NC, USA). One hundred and sixty-one patients received dexamethasone, and the data of 85 patients were finally used for analysis of effectiveness. Patients who did not require oxygen during dexamethasone treatment (n = 45) and those treated with high-dose methylprednisolone (n = 31) were excluded (Fig. 1 ). The patients included in the study are shown in Table 1 and (Table 1) . Twenty-five patients (30%) received favipiravir prior to dexamethasone or in combination with dexamethasone. The median time from onset of symptoms to the initiation of dexamethasone was nine days (IQR 7-11 days). Seventeen of the 85 patients experienced exacerbations. Of them, five died, and 12 underwent mechanical ventilation within 28 days of onset. Of the 12 patients who were managed with ventilators, four patients improved, six did not improve and were transferred to other hospitals, and two died. During the study period, elevated blood glucose levels due to dexamethasone occurred in 36 (42%) patients, of whom 20 had diabetes mellitus as a comorbidity. Insulin was continuously infused to 31 patients. Most cases of inadequate glycemic control were resolved J o u r n a l P r e -p r o o f by the time of discharge. Other common adverse events of dexamethasone, such as thrombosis, gastrointestinal disorders, and bacterial infections, were not observed. The number of days from each event to the initiation of dexamethasone administration was compared between the exacerbation and non-exacerbation groups (Fig. 2) . There was no difference between the groups in the number of days from symptom onset to initiation of dexamethasone administration (9 [IQR 7-11] vs. 9 [IQR 7-11] days, p = 0.658) (Fig. 2a) . The duration from oxygen treatment to initiation of dexamethasone administration was significantly longer in the exacerbation group compared with the non-exacerbation group (1 [IQR 1-2] vs. 2 [IQR 1-3] days, p < 0.001) (Fig. 2b ). In addition, the time for SpO2 to fall below 95% to initiation of dexamethasone administration was longer in the exacerbation group compared with the non-exacerbation group (2 [IQR 1-4] vs. 5 [IQR 2-9] days, p = 0.029) (Fig. 2c ). We performed a subgroup analysis (Table 3 ). Regarding the time from onset of symptoms to the initiation of dexamethasone, there was a significant difference in sex and SpO2 at the time of hospitalization between the early and late dexamethasone treatment groups (Table 3) . Regarding the time from oxygen supplementation, there was a significant difference in CCI score and favipiravir treatment (Table 3 ). Regarding the time from SpO2 falling below 95%, there was a significant difference in dyslipidemia, CCI score, and favipiravir and tocilizumab treatment (Table 3) . Optimal cut-off value for the number of days until initiation of dexamethasone for clinical outcome ROC curves were drawn to calculate the optimal cut-off values for the number of days until initiation of dexamethasone for various clinical outcomes occurring within 28 days of symptom onset. The cut-off value from symptom onset to the initiation of dexamethasone was nine days (AUC, 0.54; sensitivity, 49%; specificity, 67%), and the cut-off value from the initiation of oxygen supplementation to the initiation of dexamethasone was two days (AUC, 0.72; sensitivity, 85%; specificity, 59%) ( Table 3 ). The cut-off value from SpO2 < 95% to the initiation of dexamethasone was five days (AUC, 0.69; sensitivity, 78%; specificity, 59%) ( Table 2) . The patients were divided into subgroups (early and late treatment groups) based on the optimal cut-off value. First, the subgroups stratified by the optimal cut-off value for the number of days from symptom onset to initiation of dexamethasone administration had no difference in the cumulative rate of exacerbation (Fig. 3a) . Regarding that from oxygen treatment, the cumulative rate of exacerbation was significantly lower in the early treatment group (