Preoperative characteristics and outcomes of patients in the modeling cohort
A total of 746 adult patients underwent cardiac surgery at our center from October 2020 to April 2022. After excluding patients with end-stage renal disease, pregnancy status, renal transplantation, preoperative AKI status, and missing data, 689 patients were enrolled in the modeling cohort (Fig. 1). Patients were divided into AKI group and non-AKI group according to the KDIGO criteria. Table 2 showed the baseline characteristics, general information, and outcomes of the modeling cohort.
Table 3
Comparison of baseline characteristics and outcomes of patients between the AKI and non-AKI groups
Variables | Total (n = 689) | AKI (n = 201) | non-AKI (n = 488) | P values |
Male, n (%) | 417 (60.5) | 109 (54.2) | 308 (63.1) | 0.032* |
Age, years | 57.0 (48.0, 64.0) | 62.0 (54.0, 68.0) | 55.0 (46.0, 62.0) | < 0.001* |
Weight, kg | 60.0 (53.0, 67.0) | 58.0 (52.0, 67.0) | 60.0 (53.0, 67.0) | 0.283 |
Medical History, n (%) | | | | |
Hypertension | 158 (22.9) | 71 (35.3) | 87 (17.8) | < 0.001* |
Diabetes mellitus | 62 (9.0) | 27 (13.4) | 35 (7.2) | 0.012* |
Hyperlipidemia | 5 (0.7) | 1 (0.5) | 4 (0.8) | 1.000 |
CAD | 12 (1.7) | 6 (3.0) | 6 (1.2) | 0.119 |
CKD | 5 (0.7) | 3 (1.5) | 2 (0.4) | 0.152 |
Nephrotoxic drugs, n (%) | 33 (4.8) | 10 (5.0) | 23 (4.8) | 1.000 |
Pre-op LVEF, % | 63.0 (57.0, 67.0) | 63.0 (55.0, 67.0) | 63.0 (57.0, 67.0) | 0.562 |
Pre-op examination | | | | |
Hemoglobin, g/L | 134.0 (120.0, 145.0) | 129.0 (115.0, 139.0) | 136.0 (123.0, 146.0) | < 0.001* |
D-Dimer, ng/mL | 360.0 (240.0, 830.0) | 500.0 (290.0, 1110.0) | 330.0 (220.0, 722.5) | < 0.001* |
Albumin, g/L | 38.5 (36.3, 40.9) | 37.7 (35.1, 40.1) | 38.8 (36.6, 41.3) | < 0.001* |
Baseline sCr, µmol/L | 78.0 (65.3, 93.6) | 77.6 (63.2, 97.3) | 78.5 (66.8, 92.7) | 0.653 |
Outcomes | | | | |
Hospital costs, 10k CNY | 14.8 (11.8, 19.4) | 19.7 (15.7, 26.3) | 13.5 (10.8,16.9) | < 0.001* |
In-hospital mortality, n (%) | 20 (2.9) | 15 (7.5) | 5 (1.0) | < 0.001* |
IABP, n (%) | 43 (6.2) | 27 (13.4) | 16 (3.3) | < 0.001* |
CRRT, n (%) | 19 (2.8) | 15 (7.5) | 4 (0.8) | < 0.001* |
ECMO, n (%) | 11 (1.6) | 6 (3.0) | 5 (1.0) | 0.089 |
MV, h | 20.0 (13.0, 43.0) | 43.0 (19.0, 100.0) | 17.0 (10.8, 24.0) | < 0.001* |
ICU LOS, h | 45.0 (27.0, 112.0) | 110.0 (46.0, 208.0) | 42.0 (22.0, 68.2) | < 0.001* |
AKI staging, n (%) | | | | |
Stage 1 | | 132 (65.7) | | |
Stage 2 | | 48 (23.9) | | |
Stage 3 | | 21 (10.4) | | |
CAD, coronary artery disease; CKD, chronic kidney disease; Nephrotoxic drugs, including immunosuppressants, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, non-steroidal anti-inflammatory drugs, antibiotics (including vancomycin, acyclovir, amphotericin B, aminoglycosides), mannitol, angiography agents, etc.; Pre-op, preoperative; LVEF, left ventricular ejection fraction; sCr, serum creatinine; CNY, Chinese yuan; IABP, intra-aortic balloon pump; CRRT, continuous renal replacement therapy; ECMO, extracorporeal membrane oxygenation; MV, mechanical ventilation; ICU LOS, intensive care unit length of stay; AKI, acute kidney injury; *There is a statistical difference.
Overall, there were 201 (29.2%) cases of CS-AKI in the cohort, with 132 (65.7%) were in stage 1, 48 (23.9%) in stage 2, and 21 (10.4%) in stage 3. Compared to the non-AKI group, patients in the AKI group were older [62.0 (54.0, 68.0) vs. 55.0 (46.0, 62.0) years old, P < 0.001] and had a higher proportion of hypertension (35.3 vs. 17.8%, P < 0.001) and diabetes mellitus (13.4 vs. 7.2%, P = 0.012). Regarding laboratory results, patients in the AKI group had higher preoperative DDI levels [500.0 (290.0, 1110.0) vs. 330.0 (220.0, 722.5) ng/mL, P < 0.001], but had lower serum albumin concentration [37.7 (35.1, 40.1)) vs. 38.8 (36.6, 41.3) g/L, P < 0.001] and hemoglobin [129.0 (115.0, 139.0) vs. 136.0 (123.0, 146.0) g/L, P < 0.001] than those in the non-AKI group. However, there were no statistical difference in the preoperative use of nephrotoxic drugs, cardiac function, and baseline sCr between the two groups. Notably, patients in the AKI group had a higher proportion of postoperative adverse outcomes, which were associated with higher in-hospital mortality, higher healthcare costs, higher rates of IABP and CRRT, and longer duration of mechanical ventilation and ICU stay.
Intraoperative and postoperative characteristics of patients in the modeling cohort
The surgical procedures, intraoperative, and postoperative characteristics of the modeling cohort are shown in Table 4. Compared to the non-AKI group, the AKI group had a higher proportion of emergency surgery (5.5 vs. 1.8%, P = 0.021) and CABG combined with valve replacement (10.4 vs. 4.7%, P = 0.009). Patients in the AKI group had a longer CPB time [175.0 (130.0, 214.0) vs. 135.0 (106.8, 173.0) min. P < 0.001] and ACC time [103.0 (73.0, 138.0) vs. 84.0 (62.0, 110.0) min, P < 0.001], and had lower minimum intraoperative temperatures [30.9 (29.9, 31.7) vs. 31.1 (30.3, 32.0) ℃, P = 0.005]. There were no significant differences in other surgical procedures (valve replacement, CABG, and other procedures), intraoperative transfusions of red blood cells and hydroxyethyl starch between the two groups.
Table 4
Comparison of intraoperative and postoperative data between the AKI and non-AKI groups
Variables | Total(n = 689) | AKI(n = 201) | non-AKI(n = 488) | P值 |
Surgical procedures, n (%) | | | | |
Valve surgery | 477 (76.8) | 145 (72.1) | 400 (82.0) | 0.072 |
CABG | 51 (7.4) | 21 (10.4) | 30 (6.1) | 0.056 |
CABG + valve surgery | 44 (6.4) | 21 (10.4) | 23 (4.7) | 0.009* |
Other procedures | 49 (7.1) | 14 (7.0) | 35 (7.2) | 1.000 |
Intra-op data | | | | |
CPB duration, min | 145.0 (113.0, 186.0) | 175.0 (130.0, 214.0) | 135.0 (106.8, 173.0) | < 0.001* |
ACC duration, min | 88.0 (64.0, 118.0) | 103.0 (73.0, 138.0) | 84.0 (62.0, 110.0) | < 0.001* |
minimum temperature, °C | 31.0 (30.1, 31.9) | 30.9 (29.9, 31.7) | 31.1 (30.3, 32.0) | 0.005* |
Intra-op transfusion | | | | |
RBC, U | 2.0 (1.5, 2.6) | 2.3 (1.7, 3.1) | 1.8 (1.3, 2.3) | 0.067 |
Hydroxyethyl starch, mL | 354.5 (288.6, 424.3) | 388.7 (338.4, 440.3) | 326.5 (276.7,378.9) | 0.599 |
Emergency surgery, n (%) | 20 (2.9) | 11 (5.5) | 9 (1.8) | 0.021* |
Post-op data | | | | |
APACHE II scores | 9.8 (6.8,12.8) | 11.8 (7.5,16.1) | 8.6 (6.2, 11.0) | < 0.001* |
sCysC, mg/L | 1.0 (0.8, 1.3) | 1.2 (0.9, 1.7) | 0.9 (0.7, 1.1) | < 0.001* |
uNAG, U/g Cre | 8.0 (1.6, 20.8) | 12.2 (4.7, 25.7) | 4.8 (1.3, 18.2) | < 0.001* |
CK, U/L | 530.0 (393.0, 746.0) | 635.0 (451.0, 848.0) | 505.0 (365.0, 690.2) | < 0.001* |
CK-MB, U/L | 61.2 (42.2, 87.8) | 69.9 (47.8, 101.8) | 58.9 (40.3, 83.0) | < 0.001* |
NT-proBNP, pg/mL | 1565.0 (636.6, 3556.0) | 2638.0 (1041.0, 5000.0) | 1336.2 (536.9, 2864.2) | < 0.001* |
cTNI, ng/mL | 654.6 (304.4, 1183.0) | 925.5 (465.8, 2031.0) | 554.5 (265.8, 1027.0) | < 0.001* |
CABG, coronary artery bypass grafting; Other procedures, including correction of congenital heart disease, repair of aortic sinus aneurysm, etc.; Intra-op, intraoperative; CPB, cardiopulmonary bypass; ACC, aortic cross-clamping; RBC, red blood cells; Post-op, postoperative; APACHE, acute physiology and chronic health evaluation; sCysC, serum cystatin C; uNAG, urine N-acetyl-β-D-glucosidase; CK, creatine kinase; CKMB, creatine kinase isoenzyme; NT-proBNP, N-terminal B-type natriuretic peptide precursor; cTNI, cardiac troponin I; *There is a statistical difference.
In terms of postoperative data, patients in the AKI group had higher postoperative APACHE II scores [9.8 (7.5, 12.0) vs. 7.6 (5.2, 10.0), P < 0.01], postoperative sCysC [1.2 (0.9, 1.7) vs. 0.9 (0.7, 1.1) mg/L, P < 0.001], uNAG [12.2 (4.7, 25.7) vs. 4.8 (1.3, 18.2) U/g Cre, P < 0.001], serum CK [635.0 (451.0, 848.0) vs. 505.0 (365.0, 690.2) U/L, P < 0.001], CK-MB [69.9 (47.8, 101.8) vs. 58.9 (40.3 83.0) U/L, P < 0.001], NT-proBNP [2638.0 (1041.0, 5000.0) vs. 1336.2 (536.9, 2864.2) pg/mL, P < 0.001], and cTnI [925.5 (465.8, 2031.0) vs. 554.5 (265.8 1027.0) ng/mL, P < 0.001].
Comparison of clinical characteristics and outcomes of patients with different stages of AKI in the modeling cohort
Clinical data and outcomes of the AKI stage subgroups in the modeling cohort are shown in Table 5. In the AKI group, the levels of postoperative sCysC, uNAG, cTNI, and APACHE II scores increased with increasing AKI stage, whereas preoperative serum albumin, hemoglobin concentration, and minimum intraoperative temperature decreased with increasing AKI stage. In addition, the percentage of emergency surgery and ECMO was significantly higher in patients with stage 3 AKI than in other subgroups. In terms of adverse outcomes, patients with stage 3 AKI had a mortality rate of 19.0%, the incidence of RRT and IABP was 38.1% and 33.3%, respectively, which were significantly higher than those of other subgroups. These patients also had higher hospital costs, prolonged duration of mechanical ventilation and ICU stay.
Table 5
Comparison of clinical information and outcomes in various stages of CS-AKI
Variables | Total | non-AKI | AKI stage 1 | AKI stage 2 | AKI stage 3 | P value |
Number of cases | 689 | 488 | 132 | 48 | 21 | |
Male, n (%) | 417 (60.5) | 308 (63.1) | 75 (56.8) | 24 (50.0) | 10 (47.6) | 0.117 |
Age, years | 57.0 (48.0, 64.0) | 55.0 (46.0, 62.0) | 62.5 (54.0, 68.0) | 62.0 (54.8, 67.2) | 65.0 (53.0, 68.0) | < 0.001* |
Weight, kg | 60.0 (53.0, 67.0) | 60.0 (53.0, 67.0) | 58.0 (52.0, 67.0) | 57.2 (52.5, 67.2) | 56.0 (52.5, 70.0) | 0.745 |
Medical History, n (%) | | | | | | |
Hypertension | 158 (22.9) | 87 (17.8) | 47 (35.6) | 14 (29.2) | 10 (47.6) | < 0.001* |
Diabetes mellitus | 62 (9.0) | 35 (7.2) | 17 (12.9) | 7 (14.6) | 3 (14.3) | 0.047 |
Hyperlipidemia | 5 (0.7) | 4 (0.8) | 0 (0.0) | 0 (0.0) | 1 (4.8) | 0.187 |
CAD | 12 (1.7) | 6 (1.2) | 6 (4.5) | 0 (0.0) | 0 (0.0) | 0.104 |
CKD | 5 (0.7) | 2 (0.4) | 2 (1.5) | 0 (0.0) | 1 (4.8) | 0.096 |
Nephrotoxic drugs, n (%) | 33 (4.8) | 23 (4.8) | 5 (3.8) | 4 (8.3) | 1 (4.8) | 0.067 |
Pre-op LVEF, % | 63.0 (57.0, 67.0) | 63.0 (57.0, 67.0) | 63.0 (55.0, 67.0) | 62.0 (54.8, 66.0) | 64.0 (57.0, 66.0) | 0.859 |
Pre-op examination | | | | | | |
Hemoglobin, g/L | 134.0 (120.0, 145.0) | 136.0 (123.0, 146.0) | 129.0 (115.8, 138.0) | 128.5 (118.0, 141.5) | 125.0 (96.0, 142.0) | 0.001* |
D-Dimer, ng/mL | 360.0 (240.0, 830.0) | 330.0 (220.0, 722.5) | 510.0 (295.0, 1132.5) | 390.0 (275.0, 915.0) | 670.0 (320.0, 1190.0) | < 0.001* |
Albumin, g/L | 38.5 (36.3, 40.9) | 38.8 (36.6, 41.3) | 37.7 (35.1, 39.6) | 38.1 (36.0, 40.6) | 37.0 (32.5, 40.1) | 0.001* |
Baseline sCr, µmol/L | 78.0 (65.3, 93.6) | 78.5 (66.8, 92.7) | 76.8 (62.7, 96.4) | 79.5 (63.3, 91.0) | 82.1 (71.1, 150.5) | 0.249 |
Surgical procedures, n (%) | | | | | | |
Valve surgery | 477 (76.8) | 400 (82.0) | 48 (36.4) | 15 (31.2) | 14 (66.6) | 0.184 |
CABG | 51 (7.4) | 30 (6.1) | 15 (11.4) | 6 (12.5) | 0 (0.0) | 0.054 |
CABG + valve surgery | 44 (6.4) | 23 (4.7) | 12 (9.1) | 7 (14.6) | 2 (9.5) | 0.016* |
Other procedures | 49 (7.1) | 35 (7.2) | 11 (8.3) | 3 (6.2) | 0 (0.0) | 0.707 |
Intra-op data | | | | | | |
CPB duration, min | 145.0 (113.0, 186.0) | 135.0 (106.8, 173.0) | 172.0 (125.0, 200.0) | 189.5 (139.0, 251.8) | 175.0 (143.0, 211.0) | < 0.001* |
ACC duration, min | 88.0 (64.0, 118.0) | 84.0 (62.0, 110.0) | 104.0 (70.0, 129.2) | 104.5 (74.8, 147.2) | 102.0 (86.0, 142.0) | < 0.001* |
Table 5. Comparison of clinical information and outcomes in various stages of CS-AKI (Continued) |
minimum temperature, °C | 31.0 (30.1, 31.9) | 31.1 (30.3, 32.0) | 30.9 (30.0, 31.8) | 30.8 (29.6, 31.6) | 30.6 (29.7, 31.4) | 0.025* |
Intra-op transfusion | | | | | | |
RBC, U | 2.0 (1.5, 2.6) | 1.8 (1.3, 2.3) | 1.9 (1.4, 2.5) | 1.8 (1.3, 2.4) | 2.0 (1.5, 2.5) | 0.087 |
Hydroxyethyl starch, mL | 354.5 (288.6, 424.3) | 326.5 (276.7,378.9) | 365.5 (316.410.7) | 398.3 (338.4, 454.5) | 386.5 (326.7,448.9) | 0.268 |
Emergency surgery, n (%) | 20 (2.9) | 9 (1.8) | 6 (4.5) | 3 (6.2) | 2 (9.5) | 0.020* |
Post-op data | | | | | | |
APACHE II scores | 9.8 (6.8,12.8) | 8.6 (6.2, 11.0) | 10.3 (8.1,12.6) | 12.2 (9.7,14.6) | 13.8 (11.7,15.9) | < 0.001* |
sCysC, mg/L | 1.0 (0.8, 1.3) | 0.9 (0.7, 1.1) | 1.1 (0.9, 1.4) | 1.4 (1.1, 1.9) | 1.9 (1.2, 2.6) | < 0.001* |
uNAG, U/g Cre | 8.0 (1.6, 20.8) | 4.8 (1.3, 18.2) | 10.9 (3.2, 18.6) | 19.3 (10.6, 32.4) | 24.2 (20.8, 30.5) | < 0.001* |
CK, U/L | 530.0 (393.0, 746.0) | 505.0 (365.0, 690.2) | 547.5 (422.0, 778.0) | 754.0 (533.8, 952.5) | 643.0 (448.0, 843.0) | < 0.001* |
CK-MB, U/L | 61.2 (42.2, 87.8) | 58.9 (40.3, 83.0) | 69.8 (47.7, 93.5) | 76.0 (48.9, 117.6) | 69.1 (50.0, 119.0) | 0.001* |
NT-proBNP, pg/mL | 1565.0 (636.6, 3556.0) | 1336.2 (536.9, 2864.2) | 2451.0 (901.0, 4523.2) | 2549.5 (1047.8, 5253.5) | 3678.0 (2115.0, 13684.0) | < 0.001* |
cTNI, ng/mL | 654.6 (304.4, 1183.0) | 554.5 (265.8, 1027.0) | 843.6 (396.6, 1507.0) | 985.5 (669.6, 2555.0) | 1076.0 (864.7, 2657.0) | < 0.001* |
Outcomes | | | | | | |
ICU LOS, h | 45.0 (27.0, 112.0) | 42.0 (22.0, 68.2) | 85.0 (40.8, 161.2) | 134.0 (79.2, 212.5) | 222.0 (154.0, 383.0) | < 0.001* |
CRRT, n (%) | 19 (2.8) | 4 (0.8) | 3 (2.3) | 4 (8.3) | 8 (38.1) | < 0.001* |
IABP, n (%) | 43 (6.2) | 16 (3.3) | 9 (6.8) | 11 (22.9) | 7 (33.3) | < 0.001* |
ECMO, n (%) | 11 (1.6) | 5 (1.0) | 1 (0.8) | 3 (6.2) | 2 (9.5) | 0.005* |
MV, h | 20.0 (13.0, 43.0) | 17.0 (10.8, 24.0) | 25.0 (18.0, 72.2) | 65.0 (37.5, 142.2) | 132.0 (47.0, 218.0) | < 0.001* |
Hospital costs, 10k yuan | 14.8 (11.8, 19.4) | 13.5 (10.8,16.9) | 17.8 (14.1, 23.3) | 21.6 (18.0,27.4) | 30.0 (20.8, 36.8) | < 0.001* |
In-hospital mortality, n (%) | 20 (2.9) | 5 (1.0) | 3 (2.3) | 8 (16.7) | 4 (19.0) | < 0.001* |
CAD, coronary artery disease; CKD, chronic kidney disease; Nephrotoxic drugs, including immunosuppressants, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, non-steroidal anti-inflammatory drugs, antibiotics (including vancomycin, acyclovir, amphotericin B, aminoglycosides), mannitol, angiography agents, etc.; Pre-op, preoperative; LVEF, left ventricular ejection fraction; CABG, coronary artery bypass grafting; Other procedures, including correction of congenital heart disease, repair of aortic sinus aneurysm, etc.; Intra-op, intraoperative; CPB, cardiopulmonary bypass, ACC, aortic cross clamping; RBC, red blood cells; Post-op, postoperative; APACHE, acute physiologic and chronic health evaluation; sCysC, serum cystatin C; uNAG, urine N-acetyl-β-D-glucosidase; NT-proBNP, N-terminal B-type natriuretic peptide precursor; cTNI, cardiac troponin I; ICU LOS, intensive care unit length of stay; CRRT, continuous renal replacement therapy; IABP, intra-aortic balloon pump; ECMO, extracorporeal membrane oxygenation; MV, mechanical ventilation. *There is a statistical difference.
Clinical early prediction model based on LASSO regression
We included all 31 preselected variables from the modeling cohort of 689 postoperative cardiac patients and used LASSO regression to develop a clinical prediction model. In LASSO regression, the regularization parameter λ was obtained by 10-fold cross-validation with the minimum criterion. The partial likelihood difference curves were plotted by Log(λ), and vertical dashed lines are plotted at the minimum criterion and the optimal value of the 1-SE criterion (Fig. 2A). In this study, log(λ) was determined to be 0.03064213 (1-SE criterion) after using 10-fold cross-validation. Figure 2B shows the LASSO coefficient profiles generated from the log(λ) series for all variables selected. After plotting vertical lines at the determined values using 10-fold cross-validation, the 11 best nonzero coefficient predictors were screened out, which were age, hypertension, baseline sCr, preoperative albumin level, preoperative hemoglobin concentration, postoperative NT-proBNP, postoperative cTNI, CABG combined with valve surgery, duration of CPB, sCysC, and uNAG (Fig. 2C). The parameters of each factor in the model are shown in Table 6.
Table 6
Parameters of the CS-AKI prediction model
Variables | OR | 95% CI | P值 |
Age | 1.039 | 1.020–1.059 | < 0.001 |
Hypertension | 2.111 | 1.339–3.327 | 0.001 |
Baseline sCr | 0.998 | 0.998–0.999 | 0.035 |
Pre-op albumin | 0.957 | 0.905–1.013 | 0.130 |
Pre-op hemoglobin | 0.991 | 0.980–1.002 | 0.101 |
Post-op NT-proBNP | 1.000 | 1.000–1.000 | 0.005 |
Post-op cTNI | 1.000 | 1.000-1.001 | < 0.001 |
CABG + valve surgery | 0.596 | 0.329–1.081 | 0.088 |
CPB duration | 1.007 | 1.004–1.011 | < 0.001 |
sCysC | 2.694 | 1.742–4.165 | < 0.001 |
uNAG | 1.017 | 1.006–1.028 | 0.003 |
sCr, serum creatinine; Pre-op, preoperative; Post-op, postoperative; NT-proBNP, N-terminal B-type natriuretic peptide precursor; cTNI, cardiac troponin I; CABG, coronary artery bypass grafting; CPB, cardiopulmonary bypass; sCysC, serum cystatin C; uNAG, urine N-acetyl-β-D-glucosidase.
Predictive effects of relevant variables on CS-AKI
Based on the screening results of the LASSO regression, we further used ROC to evaluate the predictive effectiveness of the measures independently associated with CS-AKI. Figure 3 shows the ROC and box plots of age (Fig. 3A-B), CPB duration (Fig. 3C-D), sCysC (Fig. 3E-F), uNAG (Fig. 3G-H), cTNI (Fig. 3I-J), and NT-proBNP (Fig. 3K-L) for the prediction of CS-AKI independently, with ROC-AUC of 0.650 (0.610–0.700), 0.670 (0.630–0.710), 0.730 (0.680–0.770), 0.650 (0.610–0.690), 0.650 (0.610–0.700), and 0.640 (0.590–0.690), respectively. As can be seen, these variables alone are not powerful enough to predict CS-AKI.
Model effectiveness evaluation
We developed a CS-AKI clinical prediction model including 11 variables screened by LASSO regression including age, hypertension, baseline sCr, preoperative albumin, preoperative hemoglobin, postoperative NT-proBNP, cTNI, sCysC, uNAG, CABG combined with valve surgery, and CPB duration. Then we used ROC, calibration curve, DCA to evaluate their differentiation, calibration and clinical benefit, respectively. Figure 4A shows that the ROC-AUC of the model was 0.830 (0.800–0.860), suggesting that its differentiation and prediction efficacy were satisfactory. Figure 4B shows the calibration curve of the model, suggesting that its prediction accuracy was good. Figure 4C shows the DCA of the model, with the vertical coordinate representing the net benefit, and the horizontal coordinate representing the threshold probability (i.e., the expected benefit of no treatment is equal to that of full treatment), the red line representing the early prediction model, the green dashed line representing full treatment, and the blue dashed line representing no treatment. The results showed that this model increased net clinical benefit than the "all treatment" or "no treatment" scenarios at a threshold probability in the range of 0.09–0.80.
Establishment of validation cohort and external validation of the model
This study continued to enroll 313 adult cardiac surgery patients from May 2022 through December 2022 according to the aforementioned inclusion and exclusion criteria, which were used to establish an external validation cohort (Additional Fig. 1). Additional Table 1 shows the clinical data and outcomes of the validation cohort compared to the modeling cohort. A total of 124 patients in the validation cohort had AKI within one week after surgery, with an incidence rate of 39.6%, which was higher than that in the modeling cohort. Of these, 77 cases had stage 1 AKI (24.6%), 25 cases had stage 2 AKI (8.0%), and 22 cases had stage 3 AKI (7.0%). In terms of the patient distribution, the validation cohort had a higher incidence of stage 1 AKI (24.6% vs. 19.2%, P < 0.01) and stage 3 AKI (7.0% vs. 3.0%, P < 0.05) than the modeling cohort. Based on the variables screened by LASSO regression, the CS-AKI early prediction model for the validation cohort was established with the variables of age, hypertension, baseline sCr, preoperative albumin, preoperative hemoglobin, CABG combined with valve surgery, CPB duration, and postoperative NT-proBNP, cTNI, sCysC, and uNAG. Figure 5A shows that the ROC-AUC was 0.840 (0.790–0.880), which was not significantly different from the ROC-AUC of the modeling cohort [0.830 (0.800–0.860), P > 0.05]. The calibration curve indicated a good fit of the model (Fig. 5B).
Plotting the nomogram of the CS-AKI clinical prediction model
To facilitate the clinical use of the early prediction model, we drew a nomogram for CS-AKI (Fig. 6). This nomogram allows the clinician to obtain the corresponding score on the scale based on the values of each variable. The sum of each score gives the calculated "RISK" on the total score axis, which is the risk for CS-AKI. Figure 7 shows the Nomo-ROC and Nomo-DCA of the nomogram. The ROC-AUC of the nomogram is significantly larger than the ROC-AUC of each variable in the model (Fig. 7A). And the area under the DCA curve is also larger than the area under the DCA curve of each variable in the model (Fig. 7B). These results suggested the rationality and the diagnostic effectiveness of the nomogram were satisfactory.