Copyright
©The Author(s) 2025.
World J Crit Care Med. Dec 9, 2025; 14(4): 108907
Published online Dec 9, 2025. doi: 10.5492/wjccm.v14.i4.108907
Published online Dec 9, 2025. doi: 10.5492/wjccm.v14.i4.108907
Table 1 Calculation of 4C Mortality Score
| Variable | 4C Mortality Score |
| Age (years) | |
| < 50 | - |
| 50-59 | +2 |
| 60-69 | +4 |
| 70-79 | +6 |
| ≥ 80 | +7 |
| Sex at birth | |
| Female | - |
| Male | +1 |
| 1Number of comorbidities | |
| 0 | 0 |
| 1 | +1 |
| ≥ 2 | +2 |
| Respiratory rate (breaths/minute) | |
| < 20 | - |
| 20-29 | +1 |
| ≥ 30 | +2 |
| Peripheral oxygen saturation on room air (%) | |
| ≥ 92 | - |
| < 92 | +2 |
| Glasgow coma scale score | |
| 15 | - |
| < 15 | +2 |
| Urea (mmol/L) | |
| < 7 | - |
| 7-14 | +1 |
| > 14 | +3 |
| C reactive protein (mg/L) | |
| < 50 | - |
| 50-99 | +1 |
| ≥ 100 | +2 |
Table 2 Baseline characteristics of all patients stratified by hospital outcome, n (%)
| All patients, n = 140 | Alive, n = 121 | Died, n = 19 | P value | |
| Age, years (mean ± SD) | 55.6 ± 16.8 | 53.3 ± 16.6 | 70.1 ± 9.6 | < 0.001 |
| Sex | 0.03 | |||
| Male | 87 (62.1) | 71 (81.6) | 16 (18.4) | |
| Female | 53 (37.9) | 50 (94.3) | 3 (5.7) | |
| No of comorbidities | 0.03 | |||
| 0 | 62 (44.3) | 59 (95.2) | 3 (4.8) | |
| 1 | 41 (29.3) | 33 (80.5) | 8 (19.5) | |
| ≥ 2 | 37 (26.4) | 29 (78.4) | 8 (21.6) | |
| Process measures | ||||
| Prone positioning | 0.57 | |||
| No | 58 (41.4) | 49 (85.5) | 9 (15.5) | |
| Yes | 82 (58.6) | 72 (87.8) | 10 (12.2) | |
| Mechanical ventilation | 0.009 | |||
| No | 89 (63.6) | 82 (92.1) | 7 (7.9) | |
| Yes | 51 (36.4) | 39 (76.5) | 12 (23.5) | |
| NIV + HFNP | 0.15 | |||
| No | 50 (35.7) | 46 (92.0) | 4 (8.0) | |
| Yes | 90 (64.3) | 75 (83.3) | 15 (16.7) | |
| Vasopressors | 0.03 | |||
| No | 97 (69.3) | 88 (90.7) | 9 (9.3) | |
| Yes | 43 (30.7) | 33 (76.7) | 10 (23.3) | |
| CRRT | 0.69 | |||
| No | 138 (98.6) | 120 (87.0) | 18 (13.0) | |
| Yes | 2 (1.4) | 1 (50.0) | 1 (50.0) | |
| COVID specific treatment | ||||
| Antiviral agents | 0.57 | |||
| No | 58 (41.4) | 49 (84.5) | 9 (15.5) | |
| Yes | 82 (58.6) | 72 (87.8) | 10 (12.2) | |
| Monoclonal antibodies | 0.16 | |||
| No | 109 (77.9) | 91 (83.5) | 18 (16.5) | |
| Tocilizumab | 30 (21.4) | 29 (96.7) | 1 (3.3) | |
| Sotrovimab | 1 (0.7) | 1 (100) | 0 (0) | |
| Baricitinib | 0.53 | |||
| No | 97 (69.3) | 85 (87.6) | 12 (12.4) | |
| yes | 43 (30.7) | 36 (83.7) | 7 (16.3) | |
| Length of stay, days | ||||
| ICU, median and [IQR] | 5.9 [3.2-12.8] | 5 [2.8-11.2] | 13 [9.8-18.3] | 0.004 |
| Hospital, median and [IQR] | 13.2 [7.7-22.6] | 11.9 [7.3-21.9] | 18.4 [14.1-22.3] | 0.09 |
Table 3 Physiological and laboratory parameters of all patients stratified by hospital outcome
| All patients, n = 140 | Alive, n = 121 | Died, n = 19 | P value | |
| At hospital admission | ||||
| Respiratory rate (breath/minute) | 28.8 ± 9.8 | 29.4 ± 10.2 | 25.4 ± 6.1 | 0.10 |
| Peripheral oxygen saturation on room air (%) | 92.6 ± 7.3 | 92.2 ± 7.7 | 94.8 ± 2.7 | 0.14 |
| Glasgow coma scale score | 14.7 ± 1.1 | 14.7 ± 1.2 | 14.6 ± 0.8 | 0.55 |
| Urea (mmol/L) | 6.8 ± 5.0 | 6.4 ± 4.5 | 8.9 ± 7.4 | 0.048 |
| C reactive protein (mg/L) [IQR] | 75 [41-121] | 74 [40-125] | 89 [42-107] | 0.81 |
| At ICU admission | ||||
| Respiratory rate (breath/minute) | 28.1 ± 9.5 | 28.2 ± 9.5 | 26.9 ± 9.3 | 0.56 |
| Peripheral oxygen saturation on room air (%) | 93.4 ± 3.9 | 93.5 ± 4.1 | 93.2 ± 2.8 | 0.75 |
| Glasgow coma scale score | 12.9 ± 4.3 | 12.9 ± 4.4 | 13.3 ± 3.8 | 0.72 |
| Urea (mmol/L) | 7.2 ± 4.4 | 6.8 ± 4.1 | 9.6 ± 5.4 | 0.01 |
| C reactive protein (mg/L) [IQR] | 75 [33-125] | 64 [31-124] | 100 [71-134] | 0.09 |
Table 4 Predictive scores of all patients stratified by hospital outcome
| All patients, n = 140 | Alive, n = 121 | Died, n = 19 | P value | |
| Mortality prediction scores | ||||
| 4C Mortality Score at hospital admission | 7.6 ± 4.2 | 7.2 ± 4.2 | 10.6 ± 2.7 | < 0.001 |
| 4C Mortality Score at ICU admission | 8.4 ± 4.2 | 7.8 ± 4.1 | 12.1 ± 2.6 | < 0.0001 |
| APACHE II | 12.9 ± 5.4 | 12.3 ± 5.3 | 16.9 ± 4.5 | 0.0004 |
| APACHE III | 50.7 ± 18.0 | 48.1 ± 16.7 | 67.2 ± 17.5 | < 0.0001 |
Table 5 Predictive accuracy of 4C mortality and Acute Physiology and Chronic Health Evaluation scores
| AUROC | 95%CI | |
| 4C Mortality Score at hospital admission | 0.75 | 0.65-0.84 |
| 4C Mortality Score at ICU admission | 0.80 | 0.71-0.88 |
| APACHE II | 0.75 | 0.65-0.85 |
| APACHE III | 0.79 | 0.69-0.90 |
Table 6 Optimal cut-off value for the four predictive scores
| Cut-off value | Sensitivity | Specificity | AUROC | 95%CI for cut-off value | |
| Hospital 4C mortality | 9 | 0.74 | 0.72 | 0.73 | 7.2-10.8 |
| ICU 4C mortality | 10 | 0.84 | 0.72 | 0.78 | 8.9-11.1 |
| APACHE II | 14 | 0.68 | 0.67 | 0.68 | 11.8-16.2 |
| APACHE III | 56 | 0.79 | 0.73 | 0.76 | 52.0-60.0 |
Table 7 Comparison of mortality rates across intensive care unit 4C score risk groups, n (%)
| Risk groups | Total, n = 140 | Alive, n = 121 | Died, n = 19 |
| Low (0-3) | 15 | 15 | 0 |
| Intermediate (4-8) | 59 | 58 | 1 (1.7) |
| High (9-15) | 57 | 41 | 16 (28.1) |
| Very high (16-21) | 9 | 7 | 2 (22.2) |
Table 8 Comparison with previous studies
| Ref. | Study design | Sample size | Mortality% | 4C Mortality Score calculated at | Comparison | AUROC1 | 95%CI |
| Knight et al[2], United Kingdom | Multicenter prospective observational | 22361 | 30.9 | Hospital admission | Derivation and validation | 0.77 | 0.76-0.77 |
| Covino et al[12], Italy | Single centre retrospective | 210 | 20 | Hospital admission | COVID-GRAM, qCSI, NEWS | 0.79 | 0.73-0.85 |
| Wellbelove et al[10], United Kingdom | Exploratory analysis | 53 | 11 | Hospital admission | CURB-65, CRB-65, qSOFA, NEWS | 0.83 | 0.71-0.95 |
| Kuroda et al[11], Japan | Multicenter retrospective | 693 | 34.4 | Hospital admission | RISE-UP, A-DROP, REMS | 0.84 | 0.80-0.88 |
| Jones et al[13], Canada | Multicenter retrospective | 959 | 23 | Hospital admission | 4C score in wave1, 2 and 3 | 0.77 | 0.74-0.80 |
| Lombardi et al[3], France | Multicenter retrospective | 14343 | 35.3 | Hospital admission | 32 scores | 0.78 | 0.77-0.79 |
| Vicka et al[6], Lithuania | Single centre retrospective | 249 | 41.4 | Hospital admission | APACHE II, SOFA, SAPS | 0.75 | 0.69-0.81 |
| Mohammad et al[8], Saudi Arabia | Multicenter retrospective | 506 | 6.1 | Hospital admission | External validation | 0.90 | 0.85-0.95 |
| Durie et al[5], Australia | Multicenter prospective observational | 1492 | 13.2 | ICU admission | APACHE II | 0.79 | 0.68-0.90 |
| Altreby et al[4], Saudi Arabia | Single centre retrospective | 1493 | 37.9 | ICU admission | External validation | 0.81 | 0.79-0.83 |
| Ocho et al[9], Japan | Multicenter retrospective | 206 | 10.1 | Hospital admission | A-DROP, CURB-65, SOFA | 0.84 | 0.76-0.92 |
| Our study | Single centre retrospective | 140 | 13.6 | Hospital admission, | APACHE II, | 0.75, | 0.65-0.84, |
- Citation: Deshpande K, Tripathi D. Predictive accuracy of 4C Mortality Score and Acute Physiology and Chronic Health Evaluation scores for mortality in COVID-19 patients admitted to intensive care unit. World J Crit Care Med 2025; 14(4): 108907
- URL: https://www.wjgnet.com/2220-3141/full/v14/i4/108907.htm
- DOI: https://dx.doi.org/10.5492/wjccm.v14.i4.108907
