3.1 Clinical characteristics of 556 patients with SCAP.
A total of 556 eligible patients with SCAP were retrospectively included into the retrospective study based on the inclusion and exclusion criteria, and the baseline clinical characteristics of the patients were described in Table 1. 141 in-hospital deaths occurred in the patients, and the in-hospital mortality rate was 25.5%. The patients comprised 383 male patients (69.1%) and 171 female patients (30.9%), with a median age of 71 years old. The common clinical symptoms of SCAP were fever (50.7%), cough (75.3%), expectoration (67.0%), dyspnea (24.2%), and coma (11.4%). 234 (42.2%) patients with SCAP were accompanied by hypertension, and 112 (20.2%) patients with SCAP were accompanied by diabetes mellitus. Among the patients with SCAP and diabetes mellitus, 109 patients (97.3%) were diagnosed with type 2 diabetes mellitus. The vital signs and laboratory examinations were also summarized in Table 1. The median PLR, NLR, MLR, BLR, and ELR were 236.7, 11.91, 0.581, 0.028, and 0.021, respectively.
As shown in Table 1, the age, respiratory rate, PH, base excess (BE), WBC, neutrophil, NLR, BLR, myoglobin, BUN, Scr, Glucose, UA, serum potassium, AST, LDH, PT, D-dimer, procalcitonin, and hs-CRP were significantly elevated in the dead patients compared with the survival patients (all P < 0.05). On the contrary, SBP, DBP, PH, RBC, hemoglobin, and albumin were significantly reduced in the dead patients compared with the survival patients (all P < 0.05). In addition, the dead patients had marginally increased heart rates (P = 0.056) as well as BNP (P = 0.084) and marginally decreased platelets (P = 0.054).
Coma was more frequent occurred in the dead patients than the survival patients (P = 0.001). The comorbidities including cancer, hepatic disease, cardiovascular disease, hypertension, and diabetes mellitus along with the number of comorbidities were associated with in-hospital mortality (all P < 0.05).
Table 1. Comparison of baseline clinical characteristics between survival and dead patients with SCAP.
Clinical characteristics
|
Overall (n = 554)
|
Survival (n = 413)
|
Dead (n = 141)
|
P value
|
Age, years
|
71 (57 – 80)
|
69 (55 – 78)
|
78 (67.5 – 86)
|
1.34e-9
|
Gender, n (%)
|
|
|
|
|
Male
|
383 (69.1%)
|
290 (70.2%)
|
93 (66.0%)
|
|
Female
|
171 (30.9%)
|
123 (29.8%)
|
48 (34.0%)
|
0.344
|
Clinical symptoms, n (%)
|
|
|
|
|
Fever
|
281 (50.7%)
|
216 (52.3%)
|
65 (46.1%)
|
0.204
|
Cough
|
417 (75.3%)
|
308 (74.6%)
|
109 (77.3%)
|
0.517
|
Expectoration
|
371 (67.0%)
|
274 (66.3%)
|
97 (68.8%)
|
0.593
|
Dyspnea
|
134 (24.2%)
|
94 (22.8%)
|
40 (28.4%)
|
0.179
|
Coma
|
63 (11.4%)
|
36 (8.7%)
|
27 (19.1%)
|
0.001
|
Insanity
|
8 (1.4%)
|
7 (1.7%)
|
1 (0.7%)
|
0.661
|
Chest pain
|
25 (4.5%)
|
19 (4.6%)
|
6 (4.3%)
|
0.865
|
Comorbidities, n (%)
|
|
|
|
|
Cancer
|
75 (13.5%)
|
48 (11.6%)
|
27 (19.1%)
|
0.024
|
Hematological disease
|
142 (25.6%)
|
109 (26.4%)
|
33 (23.4%)
|
0.483
|
Hepatic disease
|
62 (11.2%)
|
53 (12.8%)
|
9 (6.4%)
|
0.036
|
Renal disease
|
119 (21.5%)
|
83 (20.1%)
|
36 (25.5%)
|
0.175
|
Cardiovascular disease
|
223 (40.3%)
|
154 (37.3%)
|
69 (48.9%)
|
0.015
|
Hypertension
|
234 (42.2%)
|
162 (39.2%)
|
72 (51.1%)
|
0.014
|
Diabetes mellitus
|
112 (20.2%)
|
70 (16.9%)
|
42 (29.8%)
|
0.001
|
Chronic pulmonary disease
|
236 (42.6%)
|
177 (42.9%)
|
59 (41.8%)
|
0.834
|
Cerebrovascular disease
|
127 (22.9%)
|
92 (22.3%)
|
35 (24.8%)
|
0.535
|
Number of comorbidities
|
|
|
|
|
0 – 1
|
67 (12.1%)
|
59 (14.3%)
|
8 (5.7%)
|
|
2
|
85 (15.3%)
|
70 (16.9%)
|
15 (10.6%)
|
|
≥ 3
|
402 (72.6%)
|
284 (68.8%)
|
118 (83.7%)
|
0.002
|
Vital signs
|
|
|
|
|
Temperature, ℃
|
36.8 (36.5 – 37.7)
|
36.8 (36.5 – 37.6)
|
36.7 (36.45 – 37.8)
|
0.354
|
Respiratory rate, breath/min
|
21 (20 – 25)
|
21 (20 – 25)
|
22 (20 – 28)
|
0.015
|
Heart rate, beat/min
|
99 (83 – 114)
|
97.5 (83 – 112)
|
103 (84 – 121)
|
0.056
|
Systolic blood pressure, mmHg
|
125 (109 – 142)
|
126 (111.5 – 142)
|
120 (101 – 143)
|
0.021
|
Diastolic blood pressure, mmHg
|
72 (62.5 – 84)
|
74 (64 – 85)
|
68 (58 – 79)
|
1.7 × 10-4
|
Blood gas analysis
|
|
|
|
|
PH
|
7.42 (7.33 – 7.47)
|
7.43 (7.36 – 7.48)
|
7.39 (7.28 – 7.46)
|
0.001
|
PaCO2, mmHg
|
37.00 (30.75 – 48.87)
|
37.00 (31.00 – 48.00)
|
36.82 (28.57 – 54.14)
|
0.921
|
PaO2, mmHg
|
68.16 (54.91 – 89.25)
|
71.42 (56.39 – 90.22)
|
64.28 (50.37 – 86.73)
|
0.154
|
BE, mmol/L
|
5.6 (2.3 – 10.1)
|
4.5 (2.1 – 8.8)
|
7.2 (4.3 – 12.1)
|
0.001
|
HCO3--, mmol/L
|
24.8 (20.1 – 31.0)
|
25.0 (21.1 – 30.7)
|
23.6 (17.0 – 32.0)
|
0.145
|
Blood routine
|
|
|
|
|
WBC, ×109/L
|
10.6 (6.7 – 15.9)
|
10.3 (6.5 – 14.8)
|
11.8 (8.0 – 18.6)
|
0.012
|
RBC, ×1012/L
|
3.56 (2.93 – 4.31)
|
3.65 (3.01 – 4.40)
|
3.35 (2.72 – 4.13)
|
0.003
|
Hemoglobin, g/L
|
106 (87 – 128)
|
108 (88 – 130)
|
103 (82 – 123)
|
0.028
|
Platelet, ×109/L
|
167.5 (111.0 – 256.5)
|
170.0 (114.0 – 262.5)
|
155.0 (93.5 – 230.5)
|
0.054
|
Neutrophil, ×109/L
|
8.92 (5.35 – 13.86)
|
8.70 (5.10 – 13.3)
|
9.93 (6.38 – 15.10)
|
0.031
|
Monocyte, ×109/L
|
0.464 (0.261 – 0.735)
|
0.464 (0.276 – 0.720)
|
0.461 (0.213 – 0.817)
|
0.774
|
Lymphocyte, ×109/L
|
0.725 (0.437 – 1.138)
|
0.735 (0.446 – 1.160)
|
0.700 (0.374 – 1.120)
|
0.395
|
Eosinophil, ×109/L
|
0.014 (0.000 – 0.086)
|
0.017 (0.000 – 0.090)
|
0.009 (0.000 – 0.064)
|
0.159
|
Basophil, ×109/L
|
0.020 (0.010 – 0.038)
|
0.020 (0.010 – 0.035)
|
0.220 (0.011 – 0.043)
|
0.156
|
Systemic inflammatory factors
|
|
|
|
|
PLR
|
236.7 (128.8 – 444.0)
|
236.9 (134.7 – 440.5)
|
230.1 (114.9 – 463.9)
|
0.628
|
NLR
|
11.91 (5.70 – 23.27)
|
11.31 (5.57 – 22.59)
|
15.26 (6.65 – 27.57)
|
0.045
|
MLR
|
0.581 (0.337 – 1.000)
|
0.585 (0.349 – 0.966)
|
0.537 (0.302 – 1.134)
|
0.738
|
BLR
|
0.028 (0.013 – 0.054)
|
0.026 (0.013 – 0.049)
|
0.036 (0.013 – 0.066)
|
0.031
|
ELR
|
0.021 (0.000 – 0.098)
|
0.023 (0.000 – 0.106)
|
0.016 (0.000 – 0.089)
|
0.183
|
Other laboratory tests
|
|
|
|
|
Myoglobin, g/L
|
137.2 (60.9 – 403.7)
|
113.4 (53.2 – 305.7)
|
220.3 (106.2 – 745.5)
|
2.89e-7
|
Hs-Tn, μg/L
|
7.20 (0.07 – 41.82)
|
6.25 (0.06 – 40.08)
|
12.10 (0.11 – 58.70)
|
0.134
|
BUN, mmol/L
|
7.89 (5.25 – 13.91)
|
7.39 (4.92 – 11.43)
|
11.13 (6.42 – 20.54)
|
1.65e-7
|
Scr, μmol/L
|
75.0 (54.7 – 127.2)
|
68.9 (52.95 – 112.45)
|
106.6 (61.33 – 187.28)
|
1.0 × 10-5
|
Glucose, mmol/L
|
7.61 (6.01 – 11.18)
|
7.30 (5.90 – 10.51)
|
8.87 (6.44 – 12.68)
|
0.005
|
UA, μmol/L
|
299 (203 – 427)
|
284 (196 – 409)
|
348 (241 – 490)
|
4.0 × 10-4
|
Serum potassium, mmol/L
|
3.97 (4.45 – 3.57)
|
3.92 (3.55 – 4.39)
|
4.08 (3.63 – 4.66)
|
0.033
|
Albumin, g/L
|
30.9 (26.8 – 34.9)
|
31.2 (27.2 – 35.4)
|
29.3 (24.9 – 33.1)
|
0.001
|
Globulin, g/L
|
29.2 (24.3 – 33.9)
|
29.2 (24.6 – 33.6)
|
29.4 (23.7 – 34.2)
|
0.938
|
AST, IU/L
|
38 (27 – 66)
|
37 (26 – 61)
|
43 (28 – 84.8)
|
0.01
|
ALT, IU/L
|
26 (15 – 50)
|
26 (16 – 50)
|
26 (14 – 50)
|
0.654
|
TBIL, µmol/L
|
13.55 (9.20 – 19.95)
|
13.6 (9.3 – 20.3)
|
13.5 (8.8 – 19.8)
|
0.927
|
LDH, U/L
|
337 (245 – 512)
|
320 (239 – 484)
|
372 (272 – 631)
|
0.001
|
PT, s
|
12.8 (11.7 – 14.4)
|
12.6 (11.6 – 14.0)
|
13.3 (12.1 – 16.5)
|
3.2 × 10-5
|
D-dimer, μg/mL
|
2.98 (1.28 – 6.56)
|
2.46 (1.09 – 5.53)
|
4.44 (2.27 – 9.01)
|
8.75e-7
|
Procalcitonin, μg/L
|
0.84 (0.20 – 4.14)
|
0.66 (0.18 – 3.30)
|
1.51 (0.45 – 7.08)
|
2.4 × 10-4
|
BNP, ng/mL
|
203.7 (69.7 – 645.6)
|
184.1 (60.2 – 644.4)
|
227.2 (101.0 – 654.1)
|
0.084
|
hs-CRP, mg/L
|
78.06 (22.65 – 138.28)
|
71.77 (19.46 – 132.43)
|
94.97 (39.45 – 160.07)
|
0.008
|
Abbreviations: SCAP, severe community-acquired pneumonia; PaCO2, arterial carbon dioxide tension; PaO2, arterial oxygen tension; BE, base excess; WBC, white blood cell; RBC, red blood cell; PLR, platelet-to-lymphocyte ratio; NLR, neutrophil-to-lymphocyte; MLR, monocyte-to-lymphocyte ratio; BLR, basophil-to-lymphocyte ratio; ELR, eosinophil-to-lymphocyte ratio; Hs-Tn, high-sensitivity troponin; BUN, blood urea nitrogen; Scr, serum creatinine; UA, uric acid; AST, aspartate aminotransferase; ALT, alanine aminotransferase; TBIL, total bilirubin; LDH, lactic dehydrogenase; PT, prothrombin time; BNP, brain natriuretic peptide; hs-CRP, high-sensitivity C-reactive protein.
3.2 Correlations of clinical characteristics with in-hospital mortality in patients with SCAP.
The age, respiratory rate, PH, BE, WBC, neutrophil, NLR, BLR, myoglobin, BUN, Scr, Glucose, UA, serum potassium, AST, LDH, PT, D-dimer, procalcitonin, hs-CRP, SBP, DBP, PH, RBC, hemoglobin, albumin, coma, cancer, hepatic disease, cardiovascular disease, hypertension, diabetes mellitus, number of comorbidities, heart rate, BNP, and platelet were included into binary logistic regression to evaluate the prognostic value of these clinical characteristics for in-hospital mortality in the patients with SCAP. As shown in Table 2, the age, coma, respiratory rate, heart rate, cancer, cardiovascular disease, hypertension, diabetes mellitus, number of comorbidities, BE, WBC, NLR, BLR, myoglobin, BUN, UA, AST, LDH, PT, D-dimer, and hs-CRP were significant risk factors for in-hospital mortality (all OR values > 0, P < 0.05). Increased SBP, DBP, PH, RBC, hemoglobin, and albumin were significantly correlated with reduced in-hospital mortality (all P < 0.05), and hepatic disease was significantly associated with lowered in-hospital mortality (P = 0.040). Raised platelet levels were marginally related to reduced in-hospital mortality (P = 0.064), while both increased glucose (P = 0.062) and procalcitonin (P = 0.076) were marginally related to elevated in-hospital mortality.
Thus, the age, coma, respiratory rate, heart rate, cancer, hepatic disease, cardiovascular disease, hypertension, diabetes mellitus, number of comorbidities, PH, BE, WBC, NLR, BLR, myoglobin, BUN, UA, AST, LDH, PT, D-dimer, hs-CRP, SBP, DBP, RBC, hemoglobin, an albumin as significant prognostic factors were included into multivariate binary logistic regression analysis to screen independent prognostic factors for in-hospital mortality in the patients with SCAP. As shown in Table 3, the age, heart rate, DBP, diabetes mellitus, and RBC were significant independent prognostic factors for in-hospital mortality (all P < 0.05).
Table 2. Univariate binary logistic regression analysis of clinical characteristics associated with in-hospital mortality in SCAP.
Clinical characteristics
|
β
|
SE
|
Wald χ2
|
P value
|
OR
|
95% CI
|
Lower limit
|
Upper limit
|
Age
|
0.039
|
0.007
|
29.363
|
6.00e-8
|
1.040
|
1.025
|
1.054
|
Coma
|
0.908
|
0.276
|
10.823
|
0.001
|
2.480
|
1.444
|
4.261
|
Respiratory rate
|
0.041
|
0.015
|
7.749
|
0.005
|
1.042
|
1.012
|
1.072
|
Heart rate
|
0.010
|
0.004
|
5.362
|
0.021
|
1.010
|
1.002
|
1.019
|
SBP
|
-0.010
|
0.004
|
6.446
|
0.011
|
0.990
|
0.983
|
0.998
|
DBP
|
-0.024
|
0.006
|
14.556
|
1.36 × 10-4
|
0.977
|
0.965
|
0.989
|
Cancer
|
0.588
|
0.263
|
4.989
|
0.026
|
1.801
|
1.075
|
3.018
|
Hepatic disease
|
-0.770
|
0.375
|
4.222
|
0.040
|
0.463
|
0.222
|
0.965
|
Cardiovascular disease
|
0.477
|
0.197
|
5.882
|
0.015
|
1.612
|
1.096
|
2.370
|
Hypertension
|
0.480
|
0.196
|
5.989
|
0.014
|
1.617
|
1.100
|
2.375
|
Diabetes mellitus
|
0.732
|
0.226
|
10.477
|
0.001
|
2.079
|
1.335
|
3.238
|
Number of comorbidities
|
|
|
|
|
|
|
|
0 – 1
|
|
|
|
|
1.000 (Reference)
|
|
|
2
|
0.458
|
0.472
|
0.940
|
0.332
|
1.580
|
0.626
|
3.987
|
≥ 3
|
1.120
|
0.392
|
8.146
|
0.004
|
3.064
|
1.420
|
6.612
|
PH
|
-2.872
|
0.902
|
10.138
|
0.001
|
0.057
|
0.010
|
0.332
|
BE
|
0.076
|
0.024
|
10.013
|
0.002
|
1.079
|
1.029
|
1.130
|
WBC
|
0.018
|
0.009
|
4.282
|
0.039
|
1.018
|
1.001
|
1.036
|
Neutrophil
|
0.017
|
0.011
|
2.311
|
0.128
|
1.017
|
0.995
|
1.039
|
RBC
|
-0.332
|
0.104
|
10.181
|
0.001
|
0.718
|
0.586
|
0.880
|
Hemoglobin
|
-0.009
|
0.004
|
6.025
|
0.014
|
0.991
|
0.984
|
0.998
|
Platelet
|
-0.002
|
0.001
|
3.442
|
0.064
|
0.998
|
0.997
|
1.000
|
NLR < 11.91
|
|
|
|
|
1.000 (Reference)
|
|
|
NLR ≥ 11.91
|
0.450
|
0.197
|
5.219
|
0.022
|
1.569
|
1.066
|
2.309
|
BLR < 0.028
|
|
|
|
|
1.000 (Reference)
|
|
|
BLR ≥ 0.028
|
0.499
|
0.198
|
6.383
|
0.012
|
1.647
|
1.118
|
2.426
|
Myoglobina
|
1.26 × 10-4
|
4.7 × 10-5
|
7.240
|
0.007
|
1.000
|
1.000
|
1.000
|
BUN
|
0.041
|
0.010
|
18.543
|
1.7 × 10-5
|
1.042
|
1.023
|
1.062
|
Scr
|
4.43 × 10-4
|
4.29 × 10-4
|
1.067
|
0.302
|
1.000
|
1.000
|
1.001
|
Glucose
|
0.033
|
0.018
|
3.493
|
0.062
|
1.034
|
0.998
|
1.070
|
UA
|
0.002
|
0.001
|
15.965
|
6.5 × 10-5
|
1.002
|
1.001
|
1.003
|
Serum potassium
|
-0.008
|
0.025
|
0.100
|
0.752
|
0.992
|
0.945
|
1.042
|
Albumin
|
-0.052
|
0.016
|
10.163
|
0.001
|
0.949
|
0.919
|
0.980
|
AST
|
0.001
|
0.000
|
3.987
|
0.046
|
1.001
|
1.000
|
1.002
|
LDHa
|
3.79 × 10-4
|
1.90 × 10-4
|
3.978
|
0.046
|
1.000
|
1.000
|
1.001
|
PT
|
0.089
|
0.023
|
15.016
|
1.07 × 10-4
|
1.094
|
1.045
|
1.144
|
D-dimer
|
0.017
|
0.008
|
4.790
|
0.029
|
1.017
|
1.002
|
1.033
|
Procalcitonin
|
0.006
|
0.003
|
3.158
|
0.076
|
1.006
|
0.999
|
1.013
|
BNP
|
7.4 × 10-5
|
4.9 × 10-5
|
2.311
|
0.128
|
1.000
|
1.000
|
1.000
|
Hs-CRP
|
0.003
|
0.001
|
5.410
|
0.020
|
1.003
|
1.000
|
1.005
|
aThe precise OR (95% CI) value for myoglobin was 1.000126 (1.000034 – 1.000219); the precise OR (95% CI) value for LDH was 1.000380 (1.000007 – 1.000753).
Abbreviations: SCAP, severe community-acquired pneumonia; SBP, systolic blood pressure; DBP, diastolic blood pressure; BE, base excess; WBC, white blood cell; RBC, red blood cell; NLR, neutrophil-to-lymphocyte; BLR, basophil-to-lymphocyte ratio; BUN, blood urea nitrogen; Scr, serum creatinine; UA, uric acid; AST, aspartate aminotransferase; LDH, lactic dehydrogenase; PT, prothrombin time; BNP, brain natriuretic peptide; hs-CRP, high-sensitivity C-reactive protein.
Table 3. Multivariate binary logistic regression analysis of prognostic factors for in-hospital mortality in patients with SCAP.
Clinical characteristics
|
β
|
SE
|
Wald χ2
|
P value
|
OR
|
95% CI
|
Lower limit
|
Upper limit
|
Age
|
0.029
|
0.014
|
3.961
|
0.047
|
1.029
|
1.000
|
1.058
|
Coma
|
0.876
|
0.587
|
2.229
|
0.135
|
2.401
|
0.760
|
7.582
|
Respiratory rate
|
0.036
|
0.037
|
0.940
|
0.332
|
1.036
|
0.964
|
1.114
|
Heart rate
|
0.031
|
0.011
|
7.794
|
0.005
|
1.032
|
1.009
|
1.054
|
SBP
|
-0.005
|
0.011
|
0.254
|
0.614
|
0.995
|
0.974
|
1.016
|
DBP
|
-0.038
|
0.019
|
4.144
|
0.042
|
0.963
|
0.928
|
0.999
|
Cancer
|
0.590
|
0.558
|
1.119
|
0.290
|
1.804
|
0.604
|
5.385
|
Hepatic disease
|
-1.447
|
0.799
|
3.277
|
0.070
|
0.235
|
0.049
|
1.127
|
Cardiovascular disease
|
0.678
|
0.424
|
2.555
|
0.110
|
1.970
|
0.858
|
4.525
|
Hypertension
|
0.112
|
0.468
|
0.058
|
0.810
|
1.119
|
0.447
|
2.801
|
Diabetes mellitus
|
0.993
|
0.465
|
4.553
|
0.033
|
2.698
|
1.084
|
6.715
|
Number of comorbidities
|
|
|
|
|
|
|
|
0 – 1
|
|
|
|
|
1.000 (Reference)
|
|
|
2
|
0.216
|
1.151
|
0.035
|
0.851
|
1.241
|
0.130
|
11.849
|
≥ 3
|
1.597
|
1.020
|
2.449
|
0.118
|
4.938
|
0.668
|
36.484
|
PH
|
-3.297
|
2.046
|
2.598
|
0.107
|
0.037
|
0.001
|
2.038
|
BE
|
0.044
|
0.037
|
1.407
|
0.235
|
1.045
|
0.972
|
1.123
|
WBC
|
-0.024
|
0.025
|
0.887
|
0.346
|
0.976
|
0.929
|
1.026
|
RBC
|
-0.784
|
0.367
|
4.559
|
0.033
|
0.457
|
0.222
|
0.938
|
Hemoglobin
|
0.012
|
0.013
|
0.749
|
0.387
|
1.012
|
0.985
|
1.039
|
NLR < 11.91
|
|
|
|
|
1.000 (Reference)
|
|
|
NLR ≥ 11.91
|
-0.406
|
0.440
|
0.850
|
0.356
|
0.667
|
0.281
|
1.579
|
BLR < 0.028
|
|
|
|
|
|
|
|
BLR ≥ 0.028
|
0.608
|
0.443
|
1.887
|
0.170
|
1.837
|
0.771
|
4.374
|
Myoglobina
|
0.000
|
0.000
|
2.125
|
0.145
|
1.000
|
1.000
|
1.000
|
BUN
|
0.013
|
0.029
|
0.201
|
0.654
|
1.013
|
0.957
|
1.072
|
UA
|
-0.003
|
0.002
|
2.893
|
0.089
|
0.997
|
0.994
|
1.000
|
Albumin
|
0.023
|
0.041
|
0.330
|
0.566
|
1.024
|
0.945
|
1.109
|
AST
|
0.001
|
0.002
|
0.125
|
0.724
|
1.001
|
0.997
|
1.004
|
LDH
|
0.000
|
0.001
|
0.124
|
0.725
|
1.000
|
0.999
|
1.001
|
PT
|
0.053
|
0.047
|
1.269
|
0.260
|
1.055
|
0.961
|
1.158
|
D-dimer
|
0.006
|
0.008
|
0.552
|
0.458
|
1.006
|
0.991
|
1.021
|
Hs-CRP
|
0.000
|
0.003
|
0.006
|
0.939
|
1.000
|
0.994
|
1.005
|
aThe precise OR (95% CI) value for myoglobin was 1.000119 (0.999964 – 1.000275).
Abbreviations: SCAP, severe community-acquired pneumonia; SBP, systolic blood pressure; DBP, diastolic blood pressure; BE, base excess; WBC, white blood cell; RBC, red blood cell; NLR, neutrophil-to-lymphocyte; BLR, basophil-to-lymphocyte ratio; BUN, blood urea nitrogen; Scr, serum creatinine; UA, uric acid; AST, aspartate aminotransferase; LDH, lactic dehydrogenase; PT, prothrombin time; BNP, brain natriuretic peptide; hs-CRP, high-sensitivity C-reactive protein.
3.3 Establishment and validation of a predictive model.
The age, heart rate, DBP, diabetes mellitus, and RBC were used to establish a nomogram to predict in-hospital mortality in the patients with SCAP (Figure 1). Each prognostic factor designated a score presented on the top line of the nomogram. A summation of the scores for all prognostic factors was obtained, the total score reflected a score presented on the bottom line of the nomogram, and it denoted the probability of in-hospital mortality in a patient with SCAP. The nomogram had good prognostic value for in-hospital mortality, with an index of concordance (C-index) of 0.745 (95% CI: 0.700 – 0.790). The C-index and prognostic value of the nomogram verified by a ROC curve (Figure 2). A calibration curve and decision curve analysis (DCA) were used to evaluate the clinical applicability of the nomogram [16]. The calibration curve was plotted through bootstrap sampling 2000 times, and it showed that the predicted values by the nomogram were in line with the actually observed values (Figure 3A). The DAC curve indicated that the nomogram had good net benefits in a range of 0.2 – 0.6 of threshold probabilities, demonstrating the good clinical applicability of the nomogram (Figure 3B).
3.4 Association of systemic inflammatory factors with hospital length of stay (LOS) and overall survival (OS).
The median (25th–75th centile) hospital LOS was 12 (6 – 19) days in terms of its abnormal distribution. The correlations of systemic inflammatory factors with hospital LOS was determined using Spearman correlation coefficients. As shown in Figure 4A, NLR was negatively correlated with hospital LOS (r = -0.109, P = 0.010) while ELR was positively correlated with hospital LOS (r = 0.154, P = 2.78 × 10-4).
The association of systemic inflammatory factors with OS was determined using the Kaplan–Meier method, and the values of systemic inflammatory factors were cut-off by the corresponding median values. As shown in Figure 4B and Table 4, both increased NLR and BLR were associated with worse OS (both P < 0.05), while increased ELR was associated with better OS (P = 0.013).
Table 4. Association of systemic inflammatory factors with overall survival in patients with SCAP.
Systemic inflammatory factors
|
N
|
Mean ± SE (day)
|
Median (25th – 75th percentile, day)
|
HR (95% CI)
|
P value
|
NLR
|
|
|
|
|
|
< 11.91
|
278
|
85.3 ± 18.3
|
54.0 (198.0 – 23.0)
|
1.000 (reference)
|
|
≥ 11.91
|
276
|
63.9 ± 5.8
|
30.0 (114.0 – 11.0)
|
1.582 (1.129 – 2.216)
|
0.007
|
BLR
|
|
|
|
|
|
< 0.028
|
279
|
105.8 ± 16.9
|
47.0 (198.0 – 22.0)
|
1.000 (reference)
|
|
≥ 0.028
|
275
|
55.5 ± 7.3
|
36.0 (114.0 – 14.0)
|
1.468 (1.047 – 2.058)
|
0.026
|
ELR
|
|
|
|
|
|
< 0.021
|
279
|
60.0 ± 7.1
|
47.0 (114.0 – 11.0)
|
1.000 (reference)
|
|
≥ 0.021
|
275
|
78.3 ± 15.4
|
54.0 (123.0 – 20.0)
|
0.657 (0.469 – 0.921)
|
0.013
|
Abbreviations: SCAP, severe community-acquired pneumonia; SE, standard error; HR, hazard ratio; CI, confidence interval; NLR, neutrophil-to-lymphocyte; BLR, basophil-to-lymphocyte ratio; ELR, eosinophil-to-lymphocyte ratio.
3.5 Relationships between systemic inflammatory factors and ICU admission, length of stay (LOS), and mortality.
Among the included patients with SCAP, 386 patients (69.7%) were admitted into ICU based on critical conditions. The values of systemic inflammatory factors were cut-off by the corresponding median values, and the relationships between systemic inflammatory factors and ICU admission were identified using the chi-square test. As shown in Table 5, increased PLR, NLR, MLR, and BLR were all correlated with elevated ICU admission rates (all P < 0.05), while increased ELR was correlated with reduced ICU admission rates (P = 0.016).
Table 5. Relationships between systemic inflammatory factors and ICU admission in patients with SCAP.
Systemic inflammatory factors
|
N
|
ICU admission
|
P value
|
No
|
Yes
|
PLR
|
|
|
|
|
< 236.7
|
277
|
96 (34.7%)
|
181 (65.3%)
|
|
≥ 236.7
|
277
|
73 (26.4%)
|
204 (73.6%)
|
0.034
|
NLR
|
|
|
|
|
< 11.91
|
278
|
113 (40.6%)
|
165 (59.4%)
|
|
≥ 11.91
|
276
|
56 (20.3%)
|
220 (79.7%)
|
1.96e-7
|
MLR
|
|
|
|
|
< 0.581
|
277
|
99 (35.7%)
|
178 (64.3%)
|
|
≥ 0.581
|
277
|
70 (25.3%)
|
207 (74.7%)
|
0.007
|
BLR
|
|
|
|
|
< 0.028
|
279
|
101 (36.2%)
|
178 (63.8%)
|
|
≥ 0.028
|
275
|
68 (24.7%)
|
207 (75.3%)
|
0.003
|
ELR
|
|
|
|
|
< 0.021
|
279
|
72 (25.8%)
|
207 (74.2%)
|
|
≥ 0.021
|
275
|
97 (35.3%)
|
178 (64.7%)
|
0.016
|
Abbreviations: SCAP, severe community-acquired pneumonia; ICU, intensive care unit; PLR, platelet-to-lymphocyte ratio; NLR, neutrophil-to-lymphocyte; MLR, monocyte-to-lymphocyte ratio; BLR, basophil-to-lymphocyte ratio; ELR, eosinophil-to-lymphocyte ratio.
In the 386 patients admitted into ICU, 114 patients (29.6%) died during ICU therapy, and the corresponding median (25th – 75th percentile) PLR, NLR, MLR, BLR, and ELR were 243.2 (135.6 – 479.9), 14.72 (7.28 – 25.88), 0.639 (0.357 – 1.023), 0.032 (0.015 – 0.057), and 0.017 (0.000 – 0.091), respectively. The median (25th–75th centile) ICU LOS was 9 (4 – 15) days in terms of its abnormal distribution. The correlations of systemic inflammatory factors with ICU LOS was determined by Spearman correlation coefficients. As shown in Figure 5, ELR was positively correlated with ICU LOS (r = 0.153, P = 0.003), whereas PLR, NLR, MLR, or BLR had no significant correlation with ICU LOS (all P > 0.05). Both higher NLR and BLR were associated with elevated ICU mortality (both P < 0.05, Table 6).
Table 6. Association of systemic inflammatory factors with ICU mortality in patients with SCAP.
Systemic inflammatory factors
|
N
|
Death in ICU
|
OR (95% CI)
|
P value
|
No
|
Yes
|
PLR
|
|
|
|
|
|
< 243.2
|
193
|
137 (71.4%)
|
55 (28.6%)
|
1.000 (reference)
|
|
≥ 243.2
|
192
|
134 (69.4%)
|
59 (30.6%)
|
1.097 (0.708 – 1.699)
|
0.679
|
NLR
|
|
|
|
|
|
< 11.91
|
193
|
146 (75.6%)
|
47 (24.4%)
|
1.000 (reference)
|
|
≥ 11.91
|
192
|
125 (65.1%)
|
67 (34.9%)
|
1.665 (1.069 – 2.593)
|
0.023
|
MLR
|
|
|
|
|
|
< 0.581
|
192
|
135 (70.3%)
|
57 (29.7%)
|
1.000 (reference)
|
|
≥ 0.581
|
193
|
136 (70.5%)
|
57 (29.5%)
|
0.993 (0.641 – 1.538)
|
0.973
|
BLR
|
|
|
|
|
|
< 0.028
|
193
|
145 (75.1%)
|
48 (24.9%)
|
1.000 (reference)
|
|
≥ 0.028
|
192
|
126 (65.6%)
|
66 (34.4%)
|
1.582 (1.017 – 2.461)
|
0.041
|
ELR
|
|
|
|
|
|
< 0.021
|
192
|
129 (67.2%)
|
63 (32.8%)
|
1.000 (reference)
|
|
≥ 0.021
|
193
|
142 (73.6%)
|
51 (26.4%)
|
0.735 (0.474 – 1.141)
|
0.170
|
Abbreviations: SCAP, severe community-acquired pneumonia; ICU, intensive care unit; PLR, platelet-to-lymphocyte ratio; NLR, neutrophil-to-lymphocyte; MLR, monocyte-to-lymphocyte ratio; BLR, basophil-to-lymphocyte ratio; ELR, eosinophil-to-lymphocyte ratio.
3.6 The levels of systemic inflammatory factors in patients with SCAP and healthy volunteers.
Age, gender, blood cell counts, hemoglobin, systemic inflammatory factors were compared between the patients with SCAP and healthy volunteers. The comparison of gender ratios was performed using the chi-square test, and the other comparisons were performed using the Mann–Whitney U test based on the abnormal distribution of the clinical data. As shown in Table 7, the patients with SCAP had significantly higher age and male-female ratios than the healthy volunteers (both P < 0.05). PLR, NLR, MLR, and BLR were substantially decreased in the patients with SCAP compared with the healthy volunteers, meanwhile ELR was markedly elevated (all P < 0.0001).
413 of 554 (74.4%) patients with SCAP met the condition of hospital discharge after therapy and were discharged from the hospital. In these patients, blood gas analyses were performed in 214 patients (51.8%), and blood routine examinations were performed in 253 patients (61.3%). The Wilcoxon signed rank test was used to perform the paired comparison of the clinical characteristics at hospital admission and discharge in the patients with SCAP on the basis of the abnormal distribution of the difference values. As shown in Table 8, both PaCO2 and PaO2 were significantly improved at hospital discharge compared to hospital admission (both P < 0.05). PLR, NLR, and MLR were lowered whereas ELR was improved at hospital discharge compared to hospital admission (all P < 0.05).
Taken together, PLR, NLR, and MLR were improved in patients with SCAP compared to healthy volunteers, and were reduced when the patients were in recovery after treatment. Meanwhile, ELR exhibited a complete opposite trend. In addition, BLR was also elevated in patients with SCAP compared with healthy volunteers.
Table 7. Comparisons of clinical characteristics between patients with SCAP and healthy volunteers.
Clinical characteristics
|
Patients with SCAP (n = 554)
|
Healthy volunteers (n= 300)
|
P value
|
Age, years
|
71 (57 – 80)
|
51 (39 – 60)
|
6.89e-47
|
Gender, n (%)
|
|
|
|
Male
|
383 (69.1%)
|
137 (45.7%)
|
|
Female
|
171 (30.9%)
|
163 (54.3%)
|
1.97e-11
|
WBC, ×109/L
|
10.6 (6.7 – 15.9)
|
6.6 (5.7 – 7.9)
|
6.47e-34
|
RBC, ×1012/L
|
3.56 (2.93 – 4.31)
|
4.92 (4.54 – 5.31)
|
6.85e-74
|
Hemoglobin, g/L
|
106 (87 – 128)
|
144 (133 – 155)
|
7.11e-69
|
Platelet, ×109/L
|
167.5 (111.0 – 256.5)
|
264.5 (230.0 – 300.8)
|
1.43e-35
|
Neutrophil, ×109/L
|
8.92 (5.35 – 13.86)
|
3.55 (2.82 – 4.45)
|
1.79e-63
|
Monocyte, ×109/L
|
0.464 (0.261 – 0.735)
|
0.480 (0.370 – 0.590)
|
0.575
|
Lymphocyte, ×109/L
|
0.725 (0.437 – 1.138)
|
2.350 (1.970 – 2.825)
|
4.08e-101
|
Eosinophil, ×109/L
|
0.014 (0.000 – 0.086)
|
0.130 (0.083 – 0.210)
|
3.97e-56
|
Basophil, ×109/L
|
0.020 (0.010 – 0.038)
|
0.030 (0.020 – 0.040)
|
1.96e-9
|
Systemic inflammatory factors
|
|
|
|
PLR
|
236.7 (128.8 – 444.0)
|
112.0 (90.5 – 136.2)
|
3.32e-43
|
NLR
|
11.91 (5.70 – 23.27)
|
1.51 (1.21 – 1.87)
|
1.14e-102
|
MLR
|
0.581 (0.337 – 1.000)
|
0.204 (0.159 – 0.237)
|
2.60e-72
|
BLR
|
0.028 (0.013 – 0.054)
|
0.012 (0.008 – 0.017)
|
1.90e-28
|
ELR
|
0.021 (0.000 – 0.098)
|
0.056 (0.035 – 0.092)
|
9.02e-16
|
Abbreviations: SCAP, severe community-acquired pneumonia; WBC, white blood cell; RBC, red blood cell; PLR, platelet-to-lymphocyte ratio; NLR, neutrophil-to-lymphocyte; MLR, monocyte-to-lymphocyte ratio; BLR, basophil-to-lymphocyte ratio; ELR, eosinophil-to-lymphocyte ratio.
Table 8. Comparison of clinical characteristics at hospital admission and discharge in patients with SCAP.
Clinical characteristics
|
N
|
At hospital admission
|
At hospital discharge
|
P value
|
Blood gas analysis
|
|
|
|
|
PH
|
214
|
7.40 (7.30 – 7.47)
|
7.40 (7.28 – 7.45)
|
0.289
|
PaCO2, mmHg
|
214
|
40.25 (32.93 – 52.25)
|
45.88 (36.79 – 59.39)
|
0.007
|
PaO2, mmHg
|
214
|
67.83 (53.00 – 90.69)
|
74.71 (56.38 – 100.50)
|
0.029
|
BE, mmol/L
|
134
|
5.8 (2.5 – 10.8)
|
6.2 (3.2 – 11.9)
|
0.260
|
HCO3--, mmol/L
|
151
|
27.3 (21.1 – 33.5)
|
28.5 (22.2 – 35.2)
|
0.243
|
WBC, ×109/L
|
253
|
9.9 (6.3 – 13.6)
|
6.6 (5.0 – 8.7)
|
4.50e-17
|
RBC, ×1012/L
|
253
|
3.61 (3.00 – 4.40)
|
3.44 (2.85 – 4.04)
|
1.85e-7
|
Hemoglobin, g/L
|
253
|
107 (88 – 130)
|
102 (85 – 119)
|
3.13e-8
|
Platelet, ×109/L
|
253
|
164 (110 – 256)
|
205 (140 – 289)
|
4.8 × 10-5
|
Neutrophil, ×109/L
|
253
|
8.30 (4.78 – 11.71)
|
4.62 (3.19 – 6.77)
|
2.24e-20
|
Monocyte, ×109/L
|
253
|
0.490 (0.294 – 0.702)
|
0.445 (0.327 – 0.612)
|
0.153
|
Lymphocyte, ×109/L
|
253
|
0.735 (0.446 – 1.144)
|
1.018 (0.676 – 1.457)
|
4.95e-10
|
Eosinophil, ×109/L
|
253
|
0.020 (0.000 – 0.099)
|
0.111 (0.040 – 0.213)
|
2.06e-15
|
Basophil, ×109/L
|
253
|
0.019 (0.009 – 0.033)
|
0.027 (0.013 – 0.043)
|
1.1 × 10-5
|
Systemic inflammatory factors
|
|
|
|
|
PLR
|
253
|
231.9 (128.5 – 414.2)
|
187.2 (126.8 – 298.7)
|
0.003
|
NLR
|
253
|
10.22 (5.12 – 20.35)
|
4.43 (2.46 – 7.91)
|
1.04e-19
|
MLR
|
253
|
0.609 (0.350 – 0.962)
|
0.425 (0.277 – 0.661)
|
2.83e-9
|
BLR
|
253
|
0.023 (0.010 – 0.045)
|
0.026 (0.012 – 0.043)
|
0.404
|
ELR
|
253
|
0.025 (0.000 – 0.111)
|
0.101 (0.043 – 0.208)
|
4.90e-11
|
Abbreviations: SCAP, severe community-acquired pneumonia; PaCO2, arterial carbon dioxide tension; PaO2, arterial oxygen tension; BE, base excess; WBC, white blood cell; RBC, red blood cell; PLR, platelet-to-lymphocyte ratio; NLR, neutrophil-to-lymphocyte; MLR, monocyte-to-lymphocyte ratio; BLR, basophil-to-lymphocyte ratio; ELR, eosinophil-to-lymphocyte ratio.