Clinical characteristics of patients
After screening by the inclusion and exclusion criteria, a total of 231 patients were included into the study. For these cases, the median age was 70 years old (range from 18 to 96 years old) and males accounted for 61.9%. A total of 73 patients passed away within 3 month during the study period, resulting in the mortality of 31.6%. The survival patients generally had lower level of t-PAIC, TAT, PT, INR, APTT, TT, FDP, D-dimer, Creatinine, Lac, Heartrate, SOFA, APACHEII, and higher levels of PLT, HB, and PaO2 in the two cohorts. Besides, the survival patients exhibited higher level of serum PH value in validation cohort. The other detailed clinical characteristics and results of univariate analysis are shown in Table 1.
Table1 Patients’ baseline characteristics at ICU admission
Clinical characteristics
|
Training cohort
|
Validation cohort
|
Survival patients
(n =112)
|
Non-survival patients
(n = 49)
|
p value
|
Survival patients
(n =46)
|
Non-survival patients (n =24 )
|
p value
|
Male, n (%)
|
69(61.6)
|
30(61.2)
|
0.963
|
29(63.0)
|
15(62.5)
|
0.964
|
Age ≥ 57(y), n (%)
|
47(42.0)
|
32(65.3)
|
0.006
|
30(65.2)
|
22(91.7)
|
0.016
|
Comorbidity
|
Diabetes, n (%)
|
17(15.2)
|
10(20.4)
|
0.414
|
17(37.0)
|
9(37.5)
|
0.964
|
Hypertension, n (%)
|
42(37.5)
|
26(53.1)
|
0.066
|
21(45.7)
|
12(50.0)
|
0.729
|
COPD, n (%)
|
6(5.4)
|
6(12.2)
|
0.126
|
8(17.4)
|
6(25.0)
|
0.450
|
CKD, n (%)
|
10(8.9)
|
6(12.2)
|
0.518
|
2(4.3)
|
3(12.5)
|
0.209
|
Source of infection
|
Pulmonary, n (%)
|
71(63.4)
|
33(67.3)
|
0.629
|
27(58.7)
|
19(71.2)
|
0.087
|
Urinary tract, n (%)
|
8(7.1)
|
1(2.0)
|
0.195
|
5(10.9)
|
0(0)
|
0.094
|
Abdominal, n (%)
|
27(24.1)
|
14(28.6)
|
0.550
|
13(28.3)
|
3(12.5)
|
0.136
|
Skin (%), n (%)
|
6(5.4)
|
2(4.1)
|
0.732
|
2(4.3)
|
1(4.2)
|
0.972
|
TM ≥ 13.1(TU/mL), n (%)
|
17(15.2)
|
22(44.9)
|
0.000
|
17(37.0)
|
17(70.8)
|
0.007
|
TAT(ng/ml)
|
8.2(4.6-18.0)
|
17.2(5.7-46.8)
|
0.002
|
8.7(5.6-17.0)
|
13.4(6.2-30.9)
|
0.162
|
PIC(ug/mL)
|
1.16(0.62-2.16)
|
1.04(0.57-2.28)
|
0.742
|
1.10(0.75-1.48)
|
1.43(0.69-2.83)
|
0.421
|
t-PAIC(ng/ml)
|
12.2(7.6-24.1)
|
21.7(11.3-41.7)
|
0.003
|
14.2(9.4-23.9)
|
21.3(13.8-47.1)
|
0.020
|
PT (s)
|
14.2(12.7-16.2)
|
16.4(14.0-21.4)
|
0.000
|
13.7(13-15.3)
|
15(13.6-19.6)
|
0.008
|
INR
|
1.2 (1.1-1.3)
|
1.4(1.2-1.8)
|
0.000
|
1.14(1.08-1.27)
|
1.25(1.13-1.60)
|
0.008
|
APTT (s)
|
31.6(26.6-38.4)
|
37.4(32.0-47.7)
|
0.000
|
31.4(26.7-40.5)
|
33.9(29.2-48.7)
|
0.087
|
FIB (s)
|
2.9±1.09
|
2.6±1.2
|
0.143
|
2.9±0.9
|
2.7±1.3
|
0.308
|
TT (s)
|
15.8(14.5-17.3)
|
17.2(14.8-18.7)
|
0.003
|
15.5(14.0-17.4)
|
16.8(14.9-19.4)
|
0.070
|
FDP(μg/L)
|
8.69(3.67-18.92)
|
14.45(4.53-38.00)
|
0.030
|
7.56(4.51-13.12)
|
11.37(6.99-27.95)
|
0.033
|
D-dimer(μg/L)
|
2.59(1.03-5.97)
|
4.91(1.65-11.00)
|
0.016
|
2.19(0.87-4.53)
|
3.19(2.54-7.76)
|
0.015
|
Platelet(× 109/L)
|
179±90
|
138±94
|
0.010
|
182±108
|
209±128
|
0.358
|
Hemoglobin(g/L)
|
111±29
|
100±31
|
0.038
|
109±31
|
104±31
|
0.525
|
ALT(U/L)
|
31.9(12.9-73.5)
|
21.5(13.3-116.8)
|
0.597
|
27.3(13.3-58.7)
|
29.6(11.1-64.2)
|
0.921
|
AST(U/L)
|
43.0(23.3-84.3)
|
42.1(26.4-131.2)
|
0.483
|
33.2(19.8-72.5)
|
30.3(19.1-76.7)
|
0.843
|
TBil(umol/L)
|
13.5(7.9-22.5)
|
13.9(7.4-32.5)
|
0.514
|
14.5(6.8-23.4)
|
17.6(10.9-28.1)
|
0.192
|
Cr (μmol/ml)
|
92.6(62.3-163.8)
|
136(76.5-241.4)
|
0.017
|
70.3(54.5-132.5)
|
113.4(78.3-150.0)
|
0.056
|
Temperature(°C)
|
36.7(36.5-37.5)
|
36.6(36.3-37.3)
|
0.350
|
36.7(36.2-37.3)
|
36.4(36.0-36.8)
|
0.176
|
Heart rate(min-1)
|
96±20
|
106±25
|
0.013
|
98±26
|
107±26
|
0.179
|
MAP (mmHg)
|
90±17
|
88±22
|
0.547
|
91±17
|
87±18
|
0.352
|
SOFA
|
7(5-10)
|
9(7-15)
|
0.000
|
7(5-10)
|
9(6-13)
|
0.065
|
APACHEII
|
21±6
|
24±6
|
0.008
|
22±7
|
27±7
|
0.004
|
PH
|
7.41 (7.35-7.45)
|
7.38(7.29-7.50)
|
0.133
|
7.42(7.34-7.49)
|
7.29(7.19-7.43)
|
0.006
|
PaCO2 (mmHg)
|
36(31-42)
|
34(29-40)
|
0.149
|
34(28-41)
|
39(32-46)
|
0.040
|
PaO2 (mmHg)
|
110(81-157)
|
93(64-140)
|
0.017
|
112(80-167)
|
97.1(64-152)
|
0.366
|
Lac (mmol/L)
|
1.7(1-3.2)
|
3.2(1.5-6.6)
|
0.000
|
2.0(1.2-3.3)
|
3.8(1.8-9.5)
|
0.010
|
COPD chronic obstructive pulmonary disease, CKD chronic kindey disease, TM thrombomodulin, TAT thrombin-antithrombin complex, PIC α2-plasmininhibitor-plasmin complex, tPAIC tissue plasminogen activator-inhibitor complex,PT prothrombin time, APTT activated partial thrombin time, FIB fibrinogen, INR international normalized ratio, TT thrombin time, FDP fibrinogen degradation product, ALT alanine transaminase, AST aspartate transaminase, TBil total bilirubin, MAP mean arterial pressure, SOFA Sequential Organ Failure Assessment, APACHEII Acute Physiology and Chronic Health Evaluation II, PH potential of hydrogen, PaO2 arterial partial oxygen pressure, PaCO2 arterial partial pressure of carbon dioxide, Lac lactate.
Development of an Individualized Prediction Model
Multiple logistic regression analysis identified the age, INR, Lac, and TM as independent predictors (Table 2). This model that contained the above independent predictors was developed and presented as the nomogram (Fig 1).
Table 2 Factors independently associated with 90-day mortality of patients with sepsis in the multivariate logistic analysis
Variables
|
OR
|
95% CI
|
P value
|
Age (≥57 y vs.<57 y)
|
1.20
|
0.36-2.04
|
0.005
|
TM(≥13.1 TU/mL vs. <13.1 TU/mL)
|
1.30
|
0.39-2.21
|
0.005
|
INR
|
1.52
|
0.23-2.80
|
0.021
|
Lac(mmol/L)
|
0.17
|
0.04-0.29
|
0.008
|
INR international normalized ratio, TM thrombomodulin, Lac lactate,
Validation of the prediction nomogram
The calibration curve of this nomogram for the probability of 90-day mortality demonstrated an excellent conformity between prediction and observation in the training and validation cohorts although the logistic calibration curve and nonparametric curve slightly deviated from ideal line (Fig 3). In addition, the Hosmer-Lemeshow test showed a nonsignificant statistic (P >0.05), which indicated that there was no violation of perfect fit.
Clinical application of the nomogram
The decision curve analysis (DCA) was performed in the training and validation cohorts,in order to assess the clinical usefulness of the predictive nomogram. It is presented in Fig. 4A and B, respectively. The decision curve showed that the nomogram could provide a good clinical utility. The result showed that if the threshold probability of a patient or doctor is approximate > 15%, using the nomogram to predict 90-day death can reap more benefit than either the treat-all-patients scheme or the treat-none scheme.