It is necessary to develop a quick, accurate scale to identify patients who are most at risk of dying. In the past, scoring systems have been applied to the healthcare system to stratify risk, predict outcomes, and efficiently manage patients (7). Numerous risk stratification systems in COVID-19 are used to forecast death. In the current study, we used the 4C mortality score among several COVID-19 patient categories. The range of the score is 0 to 15. Considering this, the anticipated mortality rates is as follows: Low-risk groups ranged from 1.2 to 1.7%, intermediate-risk groups from 9.1 to 9.9%, high-risk groups from 31.4 to 34.9%, and extremely high-risk groups from 61.5 to 66.2% (3).
Our study revealed that the average 4C mortality score in non-severe patients was 4.8. This value corresponds to an intermediate risk group for mortality with an in-hospital mortality rate of 9.9% in Knight et al., 2020 study (3), but none of the patients in this group in the present study passed away. The severe group got a 4C mortality score of 9.2 and this value had a hospital mortality rate of 34.9% in the original study (3), while just 3.1% of the patients in this group died in the current research. When we calculated this score for the critical group, we found that its mean value was 12.5 and that it corresponded to a high-risk category with 34.9% of in-hospital mortality. However, we found a significantly higher mortality in this group, where 86 out of 121 cases passed away, representing a 71.1% in-hospital mortality rate.
At a cut-off value of 9.5 for the 4C mortality score across all studied patients, this value had an AUC of 0.84 (95% CI = 0.84–0.88, p = 0.001), a sensitivity of 89.9%, a specificity of 62.8%, and positive and negative predicted values for mortality of 51% and 93.5%, respectively. According to Knight and colleagues (3), AUC was 0.79, 95% confidence interval: 0.78 to 0.79) and sensitivity 92.5% at a cut-off value of ≥ 9 to rule in mortality.
Using the same score in a Saudi Arabian study (8), at a cutoff value of 4C mortality score > 9 it had a sensitivity of 70.5%, specificity of 73.97%, positive predictive value of 62.4%, and negative predictive value of 80.2% at AUC of 0.81.
A CT scan can be a useful tool for determining the severity of each patient's condition (5). Quantitative approaches or software that uses deep learning algorithms to quantify the percentage of affected lung volumes can also be utilized to evaluate the severity (9). In the present study, we used visual method for assessing CT severity and the mean value of CT severity score among all studied patients was 12.6 (moderate severity) with a sensitivity and specificity of 84.3% and 64.7% respectively for predicting mortality. The mean CT-SS of COVID-19 patients in another Egyptian study was 13.6 (moderate severity) (10).
We found that there was a moderate to strong significant positive correlation between 4C mortality score and CT severity score in COVID-19 patients. Our study is the first to use the 4C mortality score explicitly along with the CT severity score as a prognostication tool. Therefore, a moderate severity CT severity score can be utilized as a complementary for a mortality risk indicator.
The current study revealed that the overall mortality rate was 30.1% and patients of critical group represented 75% of this number. A previous Egyptian multicenter, retrospective study (11) was conducted on all PCR confirmed COVID-19 cases admitted to 6 quarantine hospitals with a total of 3712 hospitalized patients were included; of them, 900 deaths were recorded (24.2%). The higher mortality in our study than an aforementioned one could be attributed to that, patients of critical group represented 41% of the studied patients. No classification of patients according to severity was done in the multicenter study.
On comparison the baseline characteristics of survivors and non-survivors, we found that about 70% of deceased patients, their ages more than 60 years. This finding is consistent with the results of previous study (12) that showed a higher mortality rate in older patients with COVID-19, particularly in those aged ≥ 60 years. We found also that a higher existing comorbidities and mortality, this coincides to the results of a prospective study that was conducted in Spain, in which comorbidities especially hypertension was the most reported previous comorbidity in the non-survivors’ group (61.10%, p < 0.0001) (13).In addition, both of 4C mortality score and CT severity score were significantly higher in deceased patients than alive ones. Regarding the use of ventilatory support, it was revealed that the use of NIV and IMV were more in non-survivors. Another study in China showing that non-survivors were more likely to have received noninvasive mechanical ventilation [57%, p < 0.001], invasive mechanical ventilation [35%, p < 0.001] than survivors (14).
On analysis the associated factors related to mortality, we found that age more than 60 had a significant role in mortality with an odd ratio of 2.7 and this coincides with results of Abdel Ghaffar et al study (11). Data from China and Italy reported also that the case fatality rate of COVID-19 significantly increases with age (15). This may be attributed to immunosenescence, and risk for immunopathology in elderly with decreased B and T lymphocyte activities are major contributors to older persons' susceptibility to severe infection and death (16).
It was found that male gender had a significant risk factor of mortality (OR 1.7) in comparison with females Owing to the higher expression of ACE II in males, which is the main receptor for the binding of SARS-CoV-2 to host cells (17), can contribute to this finding. We found also that either of hypertension and diabetes mellitus had a high odd ratio for mortality on univariate analysis (OR 3.13, p = 0.0001, OR 3.14, p = 0.0001 respectively). A meta-analysis included a total of 24 observational studies with 99,918 COVID-19 patients found that hypertensive patients have a 2.17-fold higher risk for COVID-19 mortality (OR: 2.17; 95% CI: 1.67 − 2.82; P < 0.001) (18).
Another meta – analysis included thirteen studies with a total number of 3027 patients with SARS-CoV-2 infection; they reported that diabetes is associated with an almost fourfold greater risk for severe disease and death in patients with COVID-19 (odds ratio (OR) = 3.68, 95% confidence interval (CI) [2.68–5.03]; P < 0.001) (19).
The current research found that both of 4C mortality score more than 9.5 and CT-SS > 12 had a significant odd ratio of mortality in both of univariate and multivariate analyses. No previous studies deal with this issue.
In conclusion, on using 4C mortality score as a simple risk stratification score of mortality, it overestimates mortality in non-severe and severe COVID-19 patients, yet, underestimates it in critical patients. Semiquantative radiological assessment by CT severity score can be used as a surrogate for prediction of mortality.
4C mortality score > 9.5 and CT severity score > 12 represent the highest value for prediction of mortality from COVID in comparison to other patient characteristics.