COVID-19 is a serious respiratory infectious disease with high case fatality rate [8]. Some studies have shown that 17% of COVID-19 patients with ARDS, and 11% will worsen and lead to death in a short period of time in Wuhan, Hubei Province in the early stage of the epidemic[9]. Therefore, early identification of high-risk patients and active and effective symptomatic support treatment are crucial to the prognosis of the disease. How to grade the severity of COVID-19 patients and evaluate their clinical prognosis is the focus of clinical research and the difficulty of clinical practice.
A total of 409 Omicron COVID-19 pneumonia patients were included in this study, and the in-hospital mortality rate was 13.2%. In an Italian study of 1715 patients, the intensive care unit (ICU) and hospital mortality rates were 48.8% and 53.4%, respectively [10]; A clinical study of 344 COVID-19 intensive care patients showed that the ICU mortality rate was 58.3% [11]; In a series of 138 confirmed cases in Wuhan, China, the mortality rate was 4.3% [12]. Studies have shown that the Omicron variant has high transmission and low pathogenicity, but it can lead to serious diseases in people with comorbidities [13]. The mortality rate of our study is higher than that of previous Omicron mutants, which may be related to the fact that all the patients included in this study are COVID-19 pneumonia patients, and most of them are elderly, and combined with underlying diseases. A systematic review shows that [14] the severity of Omicron variant is lower than that of previous waves, but there are still studies showing that patients infected with Omicron are as severe as other variants. It is emphasized that although the severity of infection in patients hospitalized during the Omicron wave may be lower, there is still a very substantial risk of severe illness.
Novel coronavirus is more likely to infect older male with underlying diseases [15], and may lead to serious or even fatal respiratory diseases, such as ARDS [9]. Studies show that age, comorbidities and ARDS are independent predictors of mortality, the mortality rate of critically patients is 61.5%, oxygenation index is related to the disease severity and prognosis [16–18]. The patients with COVID-19 pneumonia and hospitalized were included in this study, median age was 75 years old, age was an independent risk factor for poor prognosis. More than 80% of the patients were combined with basic diseases, and about 32% of the patients were complicated with 2 or more underlying diseases. The most common diseases were hypertension (49.3%) and diabetes (31.8%). The proportion of the non-survival group combined with heart disease and rheumatic immune disease was significantly higher than that in the survival group. The more serious the condition, the more likely to be combined with high blood pressure, chronic lung disease and rheumatic immune disease, suggesting that cardiovascular disease and immunosuppression will aggravate the disease and have a high risk of mortality. Nevertheless, hypertension is not an independent factor of the mortality in multivariate analysis, which may be due to the inclusion of generally older patients and other laboratory indicators that weaken the predictive value of complications. The proportion of ARDS in mortality and critically group was significantly higher than that in their respective control groups, and ARDS was an independent risk factor for critical illness. Consistent with previous studies [16, 17, 19, 20], the patients of the non-survival and critically group in this study had lower oxygenation index, critically patients often needed more advanced respiratory support, and oxygenation index was an independent risk factor for mortality, it is suggested that hypoxia is related to the prognosis and severity of the disease.
Most of the patients in this study were characterized by cough, expectoration and digestive symptoms, which was similar to the previous studies [17]. The main symptoms in the critically group were expectoration, fatigue, digestive symptoms and dyspnea, and the main symptoms in the mortality group were fatigue and digestive symptoms, expectoration and digestive symptoms are independent risk factors for critical condition, indicating that patients with poor general condition are more likely to lead to more serious illness. The possibility of septic shock caused by concomitant bacterial infection is high [9], in this study, the proportion of mortality and critically group complicated with bacterial infection is higher, and septic shock is an independent risk factor for death, it is suggested that patients with bacterial infection should be identified and treated as soon as possible.
At present, there are few studies on the impact of changes in consciousness on the prognosis of patients, Weixi Xiong et al. [21] found that lower GCS scores were independently associated with mortality. A study showed that disturbance of consciousness was an independent predictor of the development of critical diseases [22]. It is challenging to evaluate the link between consciousness and mortality, because consciousness is a dynamic process and there are many potential confounding factors. In this study, GCS score was obtained within 24 hours of admission, which is an independent risk factor for critically illness, indicating that GCS score can predict the further development of the disease.
To explore the role of laboratory examination in predicting the disease severity and death risk is the focus of research in recent years, based on which the establishment of a scoring system or the comparison of the existing scoring system has important clinical significance. Ruan Q et al. [18] showed that there were significant differences in WBC, lymphocyte, platelet, serum albumin, total bilirubin, BUN, creatinine, myoglobin, Troponin I, CRP and Interleukin-6 between the survival and non-survival group. Mo P, Xing Y et al. [23] showed that the refractory patients of COVID-19 had higher levels of neutrophil, AST, LDH and CRP, lower levels of platelet and serum albumin, and patients with refractory COVID-19 were more likely to receive mechanical ventilation. A meta analyze [24] showed that the changes of WBC, neutrophil and lymphocyte count were related to the critical condition and risk of mortality of COVID-19 patients. A retrospective study of Bernal-Monterde et al. [25] showed that increased levels of gamma-glutamyl transpeptidase and alkaline phosphatase and decreased albumin were associated with an increased risk of COVID-19 's mortality. A review shows that the main risk factors of severe clinical process and outcome in COVID-19 patients are coagulation disorder, leukocytosis, lymphopenia, eosinophil decrease, and the higher AST, BUN, CK, Troponin I, CRP, PCT, Interleukin-6 [26].
In this study, the levels of lymphocyte and serum albumin in the critically and mortality group were lower, and NLR, CRP, PCT, Interleukin-6, BUN, creatinine, AST, LDH, glucose, CK-MB, troponin I, NT-ProBNP, fibrinogen, D-dimer, lactate acid were higher. Some studies have shown that the significant decrease in the total number of lymphocyte count indicates that coronavirus consumes many immune cells and inhibits the cellular immune function of the body [27]. T- lymphocyte damage may be an important factor leading to the deterioration of the patient's condition [11], the degree of increase in neutrophil count may suggest that the intensity of inflammatory response in patients with COVID-19 is related to the storm of cytokines induced by virus invasion [17], and coagulation activation may be related to persistent inflammatory response. In addition, the degree of decrease in lymphocyte count also indicates the degree of damage to the immune system caused by viral infection [28]. Therefore, NLR can be used as a useful factor to reflect the imbalance of inflammation and immune response in patients with COVID-19 [27]. The levels of creatinine and BUN in COVID-19 patients of the critically and mortality group were higher, indicating that novel coronavirus has a significant effect on human kidney, and acute renal injury may be related to the direct effects of virus, hypoxia and shock [17]. A study of 701 patients showed that the increase of creatinine level on admission was correlated with disease severity [29], which was consistent with this study. In one study of refractory patients, LDH and CRP levels were significantly increased, and LDH is a predictor of inflammation in many lung diseases, such as obstructive disease, interstitial lung disease, severe pneumonia [23]. CRP is a widely used biochemical indicator of inflammation, which reflects the acute and severe systemic inflammatory response caused by viral infection. In a recent study, COVID-19 patients treated in the ICU had higher levels of LDH and CRP than those who did not receive treatment in the ICU [17], it is suggested that CRP and LDH are mostly related to poor prognosis. In this study, LDH is an independent risk factor for critically and mortality, while CRP is not an independent risk factor, some studies have shown that CRP cannot directly reflect the degree of cytotoxicity like LDH, so it may not be as effective as LDH in evaluating the degree and prognosis of COVID-19 pneumonia [30]. Some studies have shown that high levels of glucose are associated with poor prognosis [31], which consistent with the results of this paper, the increase of glucose levels may be related to stress caused by inflammation and islet cell damage.
PSI and CURB-65 score have passed a retrospective analysis of large samples in CAP, and have been proved to be effective, which can accurately stratify CAP patients and help clinicians make medical decisions [32, 33]. MuLBSTA score is an early warning model established by Chinese scholars to predict the mortality of viral pneumonia in recent years [5], which is used to assess the 90-day death risk of viral pneumonia, it is found that the sensitivity of the model is 77.6% and the specificity is 77.8%, which is better than that of CURB-65 score (AUROC = 0.773 vs. 0.717, p < 0.001). At the same time, it is pointed out that MuLBSTA score can accurately classify hospitalized patients with viral pneumonia into related risk categories, and can provide guidance for further clinical decision-making.
Satici et al. [34] reported that PSI score is superior to CURB-65 score in a series of cases, which is similar to our results. PSI score was originally designed to help clinicians determine which low-risk CAP patients can receive treatment in outpatient settings safely [35], although it has since been used to assess the mortality of inpatients after discharge. In this study, PSI score is better than CURB-65 and MuLBSTA score in predicting the severity of the disease. When PSI score is combined with MuLBSTA score, the effectiveness of predicting the disease is the highest, with an AUROC of 0.777, a sensitivity of 82.6% and a specificity of 58.7%. When PSI score is used to predict the risk of mortality, its efficiency is higher than that of MuLBSTA and CURB-65 score, the AUROC of the three is 0.85,0.787 and 0.736 respectively, the sensitivity of predicting mortality is 96.3%, 83.3% and 70.4%, and the specificity is 60.8%, 59.4% and 65.4%, the risk of mortality is higher when PSI score ≥ 102.5, CURB- 65 score ≥ 1.5 and MuLBSTA score ≥ 12.5.
In this study, we tried to use the clinical classification of moderate, severe and critical to predict the mortality of patients. The results showed that when the disease was severe and critical, the risk of mortality was higher, the AUC was 0.809, and the sensitivity and specificity of predicting mortality were 83.3% and 73.8%, respectively. We tried to use the combination of PSI score and clinical classification to evaluate the risk of mortality, and found that the efficacy of the combination of the two models was higher than that of the single, with an AUROC of 0.904, sensitivity specificity of 94.4% and 79.2%, respectively. Although there is no specific index of viral infection in PSI score parameters, it is the most effective in predicting the severity and prognosis of COVID-19 pneumonia in this study. However, some studies have found that the risk of death of viral and non-viral CAP increases with the increase of PSI grades, suggesting that age, vital signs and complications may have a more significant impact on prognosis, which is similar to the subjects included in this study [36].
Studies have shown that MuLBSTA score performs well in predicting mortality and risk stratification [37], in one study, PSI score has a good value in predicting the risk of mortality [38], which is consistent with our study, but MuLBSTA score is better than PSI score in predicting ICU admission needs. PSI score puts too much emphasis on the influence of age and comorbidities on prediction, while laboratory and imaging examinations have less weight, which can lead to underestimation of the risk of some patients, especially those who are young and without comorbidities. Through the monitoring of vital signs, CURB-65 score pays attention to distinguish the patients with high risk of mortality caused by severe CAP, which is simple and operable, but ignores the impact of comorbidities on the disease. MuLBSTA score includes smoking history, but for retrospective studies, there is information bias, which needs to be demonstrated by prospective studies; and for bacterial infection, it is difficult to have a strict definition. In this study, researchers mainly use infection indicators and imaging findings to determine whether bacterial infection is complicated, so there is a selection bias.
This study is a single-center retrospective study, the general condition of the patients is poor, the age is generally older and there are more underlying diseases, the state of disease is severe, and there is a certain selection bias, so the mortality rate is higher than the previous statistics; secondly, for the retrospective study, there is inevitable information bias, which needs to be further verified by prospective studies. We missed a large number of asymptomatic and mild patients, as well as patients treated at home and outpatient clinics, so our cohort may represent a more serious outcome for Omicron COVID-19 pneumonia. Omicron COVID-19 pneumonia is a new respiratory virus infectious disease, the strain is constantly changing, and the cognition for its basic and clinical is still being explored and updated, so the assessment of the risk of severity and mortality of Omicron COVID-19 pneumonia still needs large-scale clinical studies to confirm.