3.1 Baseline characteristics of patients
The study included 850 eligible AP patients, comprising 595 in the training cohort and 255 in the validation cohort. The training cohort consisted of 353 (59.3%) males and 242 (40.7%) females with AP with a median age of 60 years (IQR=48–71 years). The validation cohort comprised 130 (51.0%) males and 125 (49.0%) females with a median age of 59 years (IQR=45–71 years). Most of the patients in both cohorts were white (>65%), male (>50%), and married (>60%), and had Medicare (46.1%) or private (36.0%) insurance. Most patients did not have an infection or MODS.
Only the sex distribution and glucose differed significantly differences between the training and validation cohorts. The baseline clinicopathological data were similar in the training and validation cohorts, as indicated in Table 1. The median length of stay in the ICU was 6 days (IQR=0–102 days). Long-term outcome data were available for all 850 patients: the 28-day, 60-day, and 90-day mortality rates were 12.9% (n=110), 18.7% (n=159), and 53.2% (n=452), respectively.
3.2 Prognostic factors for 28-day, 60-day, and 90-day mortality
Univariate analyses revealed that the significant variables were weight, sex, insurance status, explicit sepsis, SAPSII score, Elixhauser score, bilirubin, anion gap, creatinine, hematocrit, hemoglobin, RDW, SpO2, heart rate, and respiratory rate, which were included in multivariate Cox regression analyses. The multivariate analyses showed that the positive factors for survival were weight (HR=0.989, P=0.001), being female (HR=0.728 vs male, P<0.05), having private insurance (HR=0.496 vs Medicare, P<0.01), having Medicaid (HR=0.674 vs Medicare, P=0.096), having government insurance (HR=0.269 vs Medicare, P<0.05), creatinine (HR=0.820, P<0.001), hemoglobin (HR=0.678, P<0.001), SpO2 (HR=0.938, P=0.001), and heart rate (HR=0.986 P<0.001). In addition, the risk factors affecting survival were explicit sepsis (HR=2.052 vs without explicit sepsis, P<0.001), SAPSII score (HR=1.037, P<0.001), Elixhauser score (HR=1.024, P=0.002), bilirubin (HR=1.043, P=0.010), anion gap (HR=1.038, P=0.001), hematocrit (HR=1.116, P=0.001), RDW (HR=1.148, P<0.001), and respiratory rate (HR=1.045, P=0.009).
3.3 Prognostic nomogram for 28-day, 60-day and 90-day mortality
The results of the multivariate regression model presented in Table 2 were used to establish a nomogram (Figure 2). The nomogram contained of all important independent factors predicting 28-day, 60-day, and 90-day mortality in the training cohort. The nomogram indicates that hemoglobin is the most important factor affecting prognosis, and it also includes sex, insurance status, explicit sepsis, Elixhauser score, weight, bilirubin, anion gap, creatinine, hematocrit, hemoglobin, RDW, heart rate, respiratory rate, and SpO2.
3.4 Performance and clinical usefulness of the nomogram
The C-index analysis of the training cohort indicated that the nomogram provided high 28-day, 60-day, and 90-day survival C-indexes, of 0.705, 0.713, and 0.720, respectively. The C-indexes of the nomogram were similarly high in the internal validation cohort, at 0.722, 0.737, and 0.751, indicating the good discriminative ability of the model (Figure 3). All of the C-indexes exceed 0.700, and the calibration curve has good consistency with the 45-degree ideal line (Figure 4). The DCA curves in Figure 5 display the large net benefits of the new model in predicting survival at 28 days, 60 days, and 90 days.
3.5 Predictive accuracy of the nomogram
The NRI values at the 28-day, 60-day, and 90-day follow-ups were 0.501, 0.704, and 0.732, respectively, in the training cohort, and 0.170, 0.299, and 0.314 in the validation cohort. The results show that the new model has better prediction performance than the SAPSII model. Moreover, the IDI values at the 28-day, 60-day, and 90-day follow-ups were 0.084, 0.107, and 0.118, respectively, in the training cohort, and 0.041, 0.077, and 0.085 in the validation cohort.