The analysis of data in our study revealed that 18 (60%) of our patients were between 50-70 years of age (Table 1, 2). It highlights the fact that majority patients affected were in relatively elderly age group. These possibly could also be due to the fact that younger patients (<12 years) were excluded from the study & those (>70 years) are more likely to remains at home due to various other disabilities and lack of family support and hence are less likely to get exposed to SARS-CoV-2 infection. In Gao et al.’s study [7], which included older (≥65 years) COVID-19 patients, the deceased group had more morbidities including cardiovascular diseases (49% vs. 20%, P<0.001), respiratory diseases (51% vs. 11%, P<0.001), chronic kidney disease (29% vs. 5%, P<0.001) and cerebrovascular disease (20% vs. 3%) than the discharged group. In a study by Vrillon et al [8] included the very elderly (≥85years) COVID-19 patients, the non-survivor group had more cardio-neurovascular diseases (68.2% vs. 37.0%, P=0.013), more complications like acute respiratory disease syndrome (95.5% vs. 1.9%, P<0.001), and needed more frequent treatment with oxygen (95.5% vs. 46.3%, P<0.001) compared with the survivor group. Yong Sub Na et al [9] also reported in-hospital mortality rate of elderly patients with severe SARS-CoV-2 was 25.5%. Non-survivor group were older, had more underlying comorbidities, frailer, and more severe by severity scores than the survivor group.

Infection rates among males & females were almost equals (Table 2). Chen J et al. [10] also in their study did not find any significant differences w.r.t age ((>65  all patients 106(63.5), survival 101(62.7), and dead 5(83.3), age <65 all patients 61(36.5), survival 60(37.3), and dead 1(16.7) and sex, (male all patients 84(50.3), survival 78(48.4), and dead 6(100.0), female all patient 83(49.7), survival 83(51.6) and dead 0(0.0).

We did not includes patients with mild infection (spo2 >94%)(6) as per guidelines of govt. as these patients were categorised as having mild illness  and so were not candidate for hospital admission. O2 saturation difference was highly significant between those who died and survived respectively (p<0.001) (Table 3) & these patients were so severely ill that even on O2 therapy the difference of oxygenation was highly significant (p<0.001) (Table 3). Hypoxia is extremely detrimental to all body organs and is main factor in pathogenesis of multi organ failure. CURB-65 score in our study was significantly different between survivors & those who did not survive (Median [IQR] 1.00(0-2) & 2.00(2-3)) respectively (Table 4) highlighting presence of severe infection in our patients who died (p=0.001) (Table 4, Figure 2). Further large no of our patients required mechanical ventilators due to inclusion criteria. Invasive ventilators has been reported to be having highly significant correlation with mortality(p<0.001)(11) due to ventilators associated pneumonia, barotrauma and lower cardiac output and  various other factors.

Investigation done on our patients who died revealed significantly higher total leucocyte count (p=0.004) (Table 3, Figure 1). Atieh et al [12] in a meta-analysis of 19 articles reported leucocytosis (the lymphocyte counts lower than 0.8 × 10⁹ /L is associated with COVID-19 severity, number of neutrophil higher than 3.5 × 10⁹ /L is associated with a poor clinical outcome). Further Haung G et al [13]. Have reported that leucocytosis at time of admission was associated with worse prognosis in this study patients categorized as having severe illness tended to have lower lymphocyte count ( pooled MD -0.36, 95% CI -0.50 to 0.27; p<0.00001) and higher leukocyte count ( pooled MD 1.32, 95% CI 0.62 to 2.02; p,0.00001). SARS Cov-2 induces severe inflammation resulting even cytokine storm which is directly associated with high mortality by causing severe lung injury. Chen J et al [14] reported severe positive correlation with mortality (p<0.001). Further neutrophil release reactive oxygen molecules & metalloproteinase which is also injurious to various tissues.

Results of our study show that APACHE-II score ranged from [8-10 (Median IQR) 9.00] and [18-23 (Median IQR) 19.00] respectively among those who survived & those who expired the difference was statistically highly significant (p=0.001) (Table 4, Figure 3). The ROC curve showed Area undercover 1.0 with Ideal cut-off value 12.50 with 100% specificity & sensitivity signifies that this score is a robust score for predicting adverse outcome. Our results are similar to that of Beigmohammadi M Taghi et.al [11] who reported mortality rate for ICU based on APACHE-II score value mean APACHE-II and mean SOFA scores were significantly higher in the non-survivor than in the survivor group (14.4 ±5.7 vs. 9.5 ± 5.1, p≤0.001, 7.3 ± 3.1 vs.3.1 ± 1.1, p≤0.001, respectively). Their study showed an increase in mortality rate at higher score (p≤0.001) at cut off score of 13. Zou et.al^.(15)reported APACHE-II score(23.23 ± 6.05 vs. 10.87 ± 4.40; p<0.001) and SOFA score at  (4.56 ± 2.81 vs. 1.63 ± 1.25; p<0.001) were higher in those who died . They further reported that APACHE-II was a better predictor of hospital mortality than SOFA score.

In our study, ROC curve regarding SOFA score indicates that AUC was 1.00, at ideal cut-off value 5.50 had sensitivity & specificity 100% (Table 5, Figure 5). Indicating that it is good score to predict adverse outcome at time of admission of patients. Nays(9)reported that SOFA score showed the best performance in predicting the prognosis of elderly patients as compared to APACHE-II & CURB-65 score (AOC =0.766 & p<0.001). They concluded that SOFA score is an efficient tool for assessing in hospital mortality in elderly patients with severe SARS-CoV-2 in their study APACHE-II score was (11.0 (8.0-14.0) & 14.0 (11.0-19.0) in survivor and non-survivor respectively (p<0.001). Similarly, SOFA score was1.0 (0.0-3.0) & 4.0(2.0-8.5) respectively in survivors than on survivor with p-value<0.001) and CURB-65 score was (1.0 (1.0-2.0) & 2.0(1.5-3.0) in survivor and non-survivor respectively (<0.001.) Beigmohammadi M Tyagi [11] also reported that mean SOFA score was significantly higher in non survivors than in survivors group (7.3 ± 3.1 vs 3.1 ± 1.1, p≤0.001). AUC was 89.5% for SOFA and 73% for APACHE-II score & both scores showed increase in mortality at higher score value (p≤0.001) with cut off value of 13 for APACHE-II & 5 for SOFA score mean daily SOFA score had a predictive performance (p≤0.001).

Analysis of data in our study revealed that CURB-65 score in non survivors and survivors was 2.0 and 1.0 respectively & this difference was highly significant (p≤0.001) (Table 4, Figure 4). ROC curve to assess diagnostic validities of scale to predict clinical outcome showed that AUC was 83 within ideal cut off value 1.50 has sensitivity of 80% & specificity 66.7% (Table 5, Figure 5). It indicates that CURB-65 scoring system is also good scoring system to predict mortality at time of admission of SARS-CoV-2 infection. However, score does not seembetter than APACHE-II & SOFA score as sensitivity & specificity are less than both these scores (100% vs 80%).

So, the early identification of these parameters and judicious use of APACHE-II, SOFA & CURB-65 score can identify and risk stratify patients at risk of death & so to mitigate the adverse outcome appropriated therapeutic strategies should be instituted. APACHE-11 and SOFA scores are equally good, CURB-65 is not as good as these two scores but still a valuable predicative tool to categorize the patients with SARS-CoV-2 infection. So, we recommend further larger prospective studies to arrive at definitive conclusion.