Statistical Analysis:
Data analysis was performed using SPSS 21.0 (Statistical Package for Social Sciences for Windows, Inc, Chicago, Illinois, USA). The compatibility of the data to normal distribution was investigated by Kolmogorov Smirnov test. Data showing the characteristics of continuous variables are expressed as mean ± standard deviation if they are distributed normally, as median (minimum-maximum) if they are not normally distributed, and categorical data were expressed as number and percentage (%). independent groups were compared using Student T test if the data was normally distributed, and Mann-Whitney U test if it was not normally distributed. Categorical variables were compared using the Chi-square test. A p <0.05 value was considered statistically significant.
RESULTS:
During the observation period, a total of 317 patients were hospitalized for COVID-19 pneumonia. After data examined, it was found that a HDS was given to 54 patients whereas 261 patients had LDS therapy. The methylprednisolone doses of 54 patients were 1000 mg, 500 mg, and 250 mg for 9, 5, and 40 patients, respectively. The mean age of included patients was 62.5 (±13.64) years, and 64.7% (n=205) were male. Although males were more in overall, there was no difference for gender between the HDS and LDS groups. HDS group was younger than LDS (58.3±13.8 vs 63.4±13.4, p=0.012).
Clinical characteristics between the high and low-dose corticosteroid groups were spesicified in Table 1. According to demographic features, the body mass index (BMI) was a little higher in the HDS group (28. 96±4.44 vs 30.28±5.08, p=0.05). Both groups had similar rates of comorbidities (Hypertension, Chronic obstructive pulmonary disease, Asthma, Coronary artery disease, Chronic kidney disease, Diabetes mellitus, Congestive heart failure, Interstitial lung diseases, Rheumotology, Malignancy; p=0.28). Admission symptoms [fever (p=0.275), dyspnea (p=0.53), cough (p=0.281)] were found at similar rates in both groups. When the first treatments at hospitalization were compared, no difference was found for Favipiravir and Low-molecular-weight heparin between the groups, whereas the need for adding antibiotics during the hospitalization and added Plaquenil (p=0.009), and vitamin C (p=0.002) more in the HDS group (p<0.001). Bilateral consolidation with ground-glass opacities were detected higher in HDS than LDS group (88.9% vs 59.7%, p<0.001). In addition, the oxygen need of patients at admission and the need for advanced oxygen therapy during hospitalization were found higher in the HDS group (p<0.001). Furthermore, 18.5% of patients in the HDS group had need transfer to the intensive care unit whereas it was 3.8% in LDS (p<0.001). Additionally, the mortality rate was determined higher in the HDS group (25. 9% vs 9.9%, p<0.001).
The median (IQR) length of hospitalization was 10 (7-15), and 6 days (4-9), in HDS and LDS groups, respectively (table 2, p<0.001). Compared with the LDS group, the HDS group had lower saturated O2 on ambient air at admission [IQR, 85% (76-89), p <0.001], and higher ferritin both at admission and before discharge [IQR, 529.5 (270.0-878.5) p = 0.03] and [IQR, 653.5 (453.6-885.0,) p <0.001] (table 2). However, there was no difference between the groups for the other laboratory parameters at admission (C-reactive protein, lymphocytes, neutrophils, neutrophil to lymphocyte ratio, platelets, D-dimer, troponin and creatinin, table 2). Median (IQR) time from symptom onset to hospitalization was 5 (4-7) days in LDS, and 6 (3-10) days in the HDS group (p=0.75). When 317 patients were evaluated, no treatment change was required in the LDS group, while the median (IQR) pulse steroid was applied in the HDS group 3 (2-5) days after hospitalization, and 10 (6-16) days after symptoms appeared. When the HDS group is examined in more detail, it was found that lymphopenia deepened and ferritin increased on the third and fifth days, while pulse steroid was also given on these days (figure 1). Moreover, compared with the LDS group, a decrease in CRP was also lower in the HDS group (table 3, figure 1).
Of overall 317 patients, when the deaths (n=25) and had no regular followed were excluded (n=109), 181 patients were examined for developing fibrosis. All the patients followed up for a mean of 40.2 (±19.38) days after discharge. Of 181 patients, fibrosis was developed in 78 (43.1%) patients whereas was not in 103 (56.9%). Compared with the non-fibrosis group, older age (p=0.001), prolonged hospitalization (p=0.024), the lower saturation at admission (p=0.01), the higher d-dimer (p=0.014), and increased recurrent hospitalizations (p=0.04) were identified in the fibrosis group (table 4). There was no difference in fibrosis development between receive high-dose steroid patients (15.4% vs 26.2%, p=0.11).
DISCUSSION:
Herein, we investigated the efficiency of short-term high-dose steroid therapy in the course of severe COVID-19 pneumonia outside the intensive care unit. We did not find a lower mortality rate among high-dose GS patients statistically. However, it should be kept in mind that this study was retrospectively designed, and the groups were not homogenous, and the high-dose steroid group consisted of patients with more severe pneumonia than the low-dose steroid group.
Today most deaths due to COVID-19 are caused by respiratory failure (stage 3) from cytokine storms and there is no effective and specific treatment is not yet for all the stages. A limited effect on the early stages (stage 1-2) with antiviral agents such as remdesivir, favipiravir, darunavir, lopinavir/ritonavir has been shown besides with the randomized controlled studies still continue on (10). Stage 3 is related to cytokine release syndrome which demonstrates high levels of interleukins (IL) (IL-1 β, IL-1RA, IL-6, IL-8, IL-9, IL-10, IL-17), macrophage inflammatory protein, vascular endothelial growth factor, tumor necrosis factor-alpha (TNF-α), and other pro-inflammatory chemokines, cytokines, and signaling proteins (11). Corticosteroids are known to functions anti-inflammatory, anti-allergic, and hypothermic due to reducing levels of IL-8, MCP-1, and IP-10 and inhibiting the gene expression of IL-6, IFN-γ, and IL-4 (3,12). Based on the effects on these cytokines it is suggested that using corticosteroid as adjunctive therapy in COVID-19 treatment (4,5, 13-20). Due to its immunosuppression property, potential risks of corticosteroid therapy including secondary infections, long-term complications, and delayed virus clearance, the benefits and harms should be carefully weighed before using corticosteroids, and time to administration should be decided carefully (21). So there is not a consensus on GC s use and the optimal dose. Considering that the conflicting results regarding the use of steroids in the treatment of patients with severe COVID-19 pneumonia so far, we aimed to demonstrate the benefits of high-dose steroid usefulness in these patients outside the intensive care unit.
Although there was also not found a lower mortality rate in the HDS group in our study such as Chen’s and Bartoletti’s studies (15, 22), more clinical improvement was observed in COVID-19 patients given HDS than LDS. However, when use a propensity score matching, significantly lower mortality was found in the Cox regression survival analysis (15). But there couldn’t apply the propensity score matching for our study due to technical problems. On the other hand, the time to give steroid is extremely important to these patients that the early high-dose GCs were shown to decrease the progression of the disease and improve the resolution of alterations in the pulmonary structures (20). Concurrent administration of high doses of steroids to these patients together with the increase in ferritin levels on the 1st, 3rd, and 5th days of hospitalization, a slower decrease in CRP value, and the deepening of lymphopenia, and the prominence of clinical worsening, supports this information.
In this recent study, similar to the literature, higher BMI, the oxygen need of patients at admission, and the need for advanced oxygen therapy, more bilateral subpleural multifocal ground-glass opacities with consolidation and hyperferritinemia were found significantly different in the HDS group than LDS group (23). However, unlike the literature, the patients were younger in the HDS group with higher mortality. Moreover patients in the HDS group spent 4 days more in the hospital, although that may be partially explained by a more severe course of the disease. Physicians should be kept in mind that it also prolongs the duration of virus clearance (24).
The secondary objective of this study was to investigate whether fibrosis develops in patients with high-dose steroid therapy on follow-up in three months. However, as expected, the rate of fibrosis was not lower among patients receiving high doses of steroids. Nevertheless, older age, prolonged hospitalization, lower saturation at admission, higher d-dimer, and increased recurrent hospitalizations were identified as the indicator for fibrosis in our study. There is not yet a study for predicting fibrosis in the literature. Although the number of patients is small, this study can be a guide for future studies.
The limitations of the study were the retrospective design that prevents randomization, the small number of patients with HDS, and the difference between groups in baseline characteristics.
In conclusion, the use of HDS in hospitalized COVID-19 patients remains unclear. Along with this, our study demonstrated the use of high-dose corticosteroids might not be associated with a lower mortality rate among hospitalized severe COVID-19 patients. Prospective multicenter studies with a large number of controlled randomized patients will be guiding in the future.