Baseline characteristics of participants
A total of 2355 participants stratified by quartiles of serum chloride levels were included in this study, their mean age was 67.31 years respectively and 61.53% of them were male (Table 1). Patients in the highest quartile of serum chloride had higher age, sodium, and SBP, whereas these patients had lower weight, WBC, platelet, RDW, potassium, calcium, glucose, PT, PTT, INR, anion gap, urea nitrogen, creatinine, disease severity score (SOFA, SIRS, APS III, and SAPS II), and SP02. In addition, they had a lower risk of heart failure, COPD, cirrhosis prevalence, and lower MV treatment, accompanied by lower hospitalization days and mortality, compared to patients in the lower quartiles.
Table 1
Baseline characteristics of critically ill patients with GIB grouped by chloride quartiles
| Total (n = 2355) | Q1 (n = 589) | Q2 (n = 586) | Q3 (n = 512) | Q4 (n = 668) | P |
Age, years | 67.31 (15.99) | 64.91 (15.81) | 67.72 (16.18) | 67.89 (16.22) | 68.63 (15.60) | < 0.001 |
Gender, n(%) | | | | | | 0.277 |
female | 906 (38.47) | 224 (38.03) | 209 (35.67) | 199 (38.87) | 274 (41.02) | |
male | 1449 (61.53) | 365 (61.97) | 377 (64.33) | 313 (61.13) | 394 (58.98) | |
Weight, kg | 80.80 (21.90) | 83.15 (25.76) | 80.65 (20.88) | 80.94 (20.02) | 78.77 (20.25) | 0.006 |
Laboratory tests | | | | | | |
WBC, k/µL | 10.33 (7.24, 14.30) | 11.22 (7.76,15.70) | 10.34 (7.23,14.40) | 9.96 (7.45,13.31) | 9.89 (6.94,13.71) | < 0.001 |
RBC, m/µL | 3.19 (0.62) | 3.14 (0.68) | 3.24 (0.67) | 3.18 (0.57) | 3.19 (0.57) | 0.082 |
Platelet, k/µL | 162.50 (108.00, 228.00) | 162.00 (93.25,236.00) | 174.50 (114.62,237.46) | 161.62 (117.46,227.58) | 155.45 (110.08,208.25) | 0.016 |
RDW, % | 16.35 (2.51) | 16.85 (2.80) | 16.37 (2.53) | 16.05 (2.44) | 16.13 (2.20) | < 0.001 |
Sodium, mmol/L | 138.99 (4.83) | 134.65 (4.80) | 138.35 (3.09) | 139.84 (2.89) | 142.72 (3.97) | < 0.001 |
Potassium, mmol/L | 4.20 (0.60) | 4.32 (0.72) | 4.20 (0.56) | 4.16 (0.53) | 4.14 (0.55) | < 0.001 |
Calcium, mg/dL | 8.14 (0.71) | 8.32 (0.78) | 8.23 (0.66) | 8.08 (0.62) | 7.95 (0.72) | < 0.001 |
Glucose, mg/dL | 138.37 (51.56) | 144.64 (59.35) | 141.19 (52.72) | 133.55 (45.78) | 134.08 (46.43) | < 0.001 |
PT, sec | 16.78 (6.52) | 18.57 (7.67) | 16.77 (6.81) | 15.82 (4.85) | 15.96 (5.92) | < 0.001 |
PTT, sec | 37.10 (16.34) | 41.79 (18.93) | 37.43 (16.91) | 34.46 (12.60) | 34.68 (14.97) | < 0.001 |
INR | 1.54 (0.62) | 1.70 (0.72) | 1.54 (0.65) | 1.44 (0.48) | 1.45 (0.57) | < 0.001 |
Anion gap, mmol/l | 14.00 (12.00, 16.50) | 16.33 (14.00,19.63) | 14.00 (12.00,16.00) | 13.00 (11.50,15.50) | 12.67 (11.00,14.75) | < 0.001 |
Urea nitrogen, mg/dl | 27.60 (17.29, 46.71) | 34.00 (19.00,59.00) | 25.00 (16.50,43.22) | 24.00 (15.00,40.50) | 28.00 (18.00,44.87) | < 0.001 |
Creatinine, mg/dl | 1.10 (0.77, 1.75) | 1.53 (0.87,2.94) | 1.05 (0.75,1.58) | 0.94 (0.70,1.44) | 1.00 (0.75,1.48) | < 0.001 |
Severity of illness | | | | | | |
SOFA | 5.00 (2.00, 8.00) | 6.00 (4.00,11.00) | 4.00 (2.00,8.00) | 4.00 (2.00,7.00) | 4.00 (2.00,7.00) | < 0.001 |
SIRS | 2.62 (0.93) | 2.69 (0.89) | 2.57 (0.96) | 2.53 (0.90) | 2.66 (0.97) | 0.011 |
APS III | 49.96 (21.32) | 57.50 (24.34) | 47.92 (20.15) | 44.82 (18.72) | 49.06 (19.55) | < 0.001 |
SAPS II | 38.85 (14.37) | 42.03 (15.57) | 37.88 (13.85) | 36.64 (14.21) | 38.57 (13.36) | < 0.001 |
Vital signs | | | | | | |
SBP, mmHg | 114.84 (105.46, 126.74) | 111.78 (102.87,123.31) | 115.87 (106.39,126.17) | 114.84 (106.50,126.84) | 116.61 (107.15,128.80) | < 0.001 |
DBP, mmHg | 62.09 (55.87, 69.25) | 61.91 (54.84,70.46) | 63.11 (56.16,69.31) | 61.74 (56.15,67.76) | 61.75 (56.19,69.22) | 0.589 |
MBP, mmHg | 76.19 (10.80) | 76.05 (12.25) | 76.65 (10.60) | 75.60 (9.63) | 76.37 (10.45) | 0.410 |
SPO2, % | 97.29 (1.93) | 96.81 (2.29) | 97.17 (1.69) | 97.48 (1.75) | 97.67 (1.80) | < 0.001 |
Comorbidities | | | | | | |
Hypertension, n(%) | | | | | | < 0.001 |
no | 1439 (61.10) | 412 (69.95) | 351 (59.90) | 278 (54.30) | 398 (59.58) | |
yes | 916 (38.90) | 177 (30.05) | 235 (40.10) | 234 (45.70) | 270 (40.42) | |
Diabetes, n(%) | | | | | | 0.232 |
no | 1657 (70.36) | 402 (68.25) | 405 (69.11) | 361 (70.51) | 489 (73.20) | |
yes | 698 (29.64) | 187 (31.75) | 181 (30.89) | 151 (29.49) | 179 (26.80) | |
Heart failure, n(%) | | | | | | < 0.001 |
no | 1725 (73.25) | 384 (65.20) | 413 (70.48) | 389 (75.98) | 539 (80.69) | |
yes | 630 (26.75) | 205 (34.80) | 173 (29.52) | 123 (24.02) | 129 (19.31) | |
CKD, n(%) | | | | | | 0.106 |
no | 1836 (77.96) | 442 (75.04) | 467 (79.69) | 412 (80.47) | 515 (77.10) | |
yes | 519 (22.04) | 147 (24.96) | 119 (20.31) | 100 (19.53) | 153 (22.90) | |
Stroke, n(%) | | | | | | 0.039 |
no | 2186 (92.82) | 549 (93.21) | 541 (92.32) | 488 (95.31) | 608 (91.02) | |
yes | 169 (7.18) | 40 (6.79) | 45 (7.68) | 24 (4.69) | 60 (8.98) | |
COPD, n(%) | | | | | | < 0.001 |
no | 2191 (93.04) | 529 (89.81) | 540 (92.15) | 484 (94.53) | 638 (95.51) | |
yes | 164 (6.96) | 60 (10.19) | 46 (7.85) | 28 (5.47) | 30 (4.49) | |
Cirrhosis, n(%) | | | | | | < 0.001 |
no | 1779 (75.54) | 398 (67.57) | 447 (76.28) | 406 (79.30) | 528 (79.04) | |
yes | 576 (24.46) | 191 (32.43) | 139 (23.72) | 106 (20.70) | 140 (20.96) | |
Treatments | | | | | | |
MV, n(%) | | | | | | < 0.001 |
no | 588 (24.97) | 101 (17.15) | 146 (24.91) | 141 (27.54) | 200 (29.94) | |
yes | 1767 (75.03) | 488 (82.85) | 440 (75.09) | 371 (72.46) | 468 (70.06) | |
Enteral nutrition, n(%) | | | | | | 0.273 |
no | 2291 (97.28) | 574 (97.45) | 573 (97.78) | 501 (97.85) | 643 (96.26) | |
yes | 64 (2.72) | 15 (2.55) | 13 (2.22) | 11 (2.15) | 25 (3.74) | |
Outcomes | | | | | | |
Length of ICU stay, days | 8.08 (4.48, 16.06) | 10.98 (5.77,21.79) | 8.63 (4.46,17.27) | 7.11 (4.11,14.02) | 6.91 (4.12,12.70) | < 0.001 |
Length of hospital stay, days | 2.41 (1.59, 4.59) | 2.86 (1.75,6.15) | 2.42 (1.58,4.83) | 2.14 (1.49,3.66) | 2.31 (1.57,4.20) | < 0.001 |
Hospital-mortality, n(%) | | | | | | < 0.001 |
no | 2004 (85.10) | 445 (75.55) | 499 (85.15) | 466 (91.02) | 594 (88.92) | |
yes | 351 (14.90) | 144 (24.45) | 87 (14.85) | 46 (8.98) | 74 (11.08) | |
ICU-mortality, n(%) | | | | | | < 0.001 |
no | 2122 (90.11) | 491 (83.36) | 530 (90.44) | 483 (94.34) | 618 (92.51) | |
yes | 233 (9.89) | 98 (16.64) | 56 (9.56) | 29 (5.66) | 50 (7.49) | |
365-day mortality, n(%) | | | | | | < 0.001 |
no | 1470 (62.42) | 276 (46.86) | 360 (61.43) | 366 (71.48) | 468 (70.06) | |
yes | 885 (37.58) | 313 (53.14) | 226 (38.57) | 146 (28.52) | 200 (29.94) | |
Abbreviations: GIB, gastrointestinal bleeding; WBC, white blood cell; RBC, red blood cell; RDW, red blood distribution width; PT, prothrombin time; PTT, partial thromboplastin time; INR, international normalized ratio; SOFA, Sequential Organ Failure Assessment; SIRS, Systemic Inflammatory Response Syndrome; APS III, Acute Physiology Score III; SAPS II, Simplified Acute Physiology Score II; SBP, systolic blood pressure; DBP, diastolic blood pressure; MBP, mean blood pressure; SPO2, percutaneous arterial oxygen saturation; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; MV, mechanical ventilation. |
After grouping patients according to their 365-day survival status in Table 2, we found that survivors were younger, had lower WBC, RDW, phosphorus, calcium, glucose, anion gap, urea nitrogen, and creatinine, and had higher RBC, platelets, sodium, chloride, systolic and diastolic blood pressure, and SP02. Non-survivors had higher disease severity indices (SOFA, SIRS, APS III, and SAPS II) and poorer coagulation (PT, PTT, and INR). The prevalence of heart failure, CKD, COPD, and cirrhosis was also higher among non-survivors. In addition, a higher proportion of non-survivors required mechanical ventilation and enteral nutrition.
Table 2
Characteristics of GIB patients by 365-day survival status
| Total (n = 2355) | survival (n = 1470) | non-survival (n = 885) | P |
Age, years | 67.31 (15.99) | 65.49 (16.22) | 70.33 (15.14) | < 0.001 |
Gender, n(%) | | | | 0.825 |
female | 906 (38.47) | 563 (38.30) | 343 (38.76) | |
male | 1449 (61.53) | 907 (61.70) | 542 (61.24) | |
Weight, kg | 80.80 (21.90) | 81.44 (20.85) | 79.74 (23.52) | 0.068 |
Laboratory tests | | | | |
WBC, k/µL | 10.33 (7.24, 14.30) | 9.70 (7.03, 13.07) | 11.63 (7.92, 16.53) | < 0.001 |
RBC, m/µL | 3.19 (0.62 | 3.22 (0.60 | 3.13 (0.65 | < 0.001 |
Platelet, k/µL | 162.50 (108.00, 228.00) | 170.50 (118.00, 228.00) | 147.43 (91.75, 226.60) | < 0.001 |
RDW, % | 16.35 (2.51) | 15.88 (2.21) | 17.14 (2.77) | < 0.001 |
Sodium, mmol/L | 138.99 (4.83) | 139.32 (4.20) | 138.43 (5.69) | < 0.001 |
Potassium, mmol/L | 4.20 (0.60) | 4.15 (0.55) | 4.30 (0.66) | < 0.001 |
Calcium, mg/dL | 8.14 (0.71) | 8.11 (0.67) | 8.19 (0.78) | 0.016 |
Chloride, mmol/L | 105.22 (6.23) | 106.06 (5.54) | 103.82 (7.01) | < 0.001 |
Glucose, mg/dL | 138.37 (51.56) | 135.74 (49.34) | 142.76 (54.80) | 0.002 |
PT, sec | 16.78 (6.52) | 15.75 (5.96) | 18.50 (7.02) | < 0.001 |
PTT, sec | 37.10 (16.34) | 34.51 (14.53) | 41.39 (18.20) | < 0.001 |
INR | 1.54 (0.62) | 1.44 (0.57) | 1.70 (0.67) | < 0.001 |
Anion gap, mmol/L | 14.00 (12.00, 16.50) | 13.00 (11.50, 15.50) | 15.33 (13.00, 18.25) | < 0.001 |
Urea nitrogen, mg/dL | 27.60 (17.29, 46.71) | 23.55 (15.50, 39.96) | 35.00 (21.67, 57.50) | < 0.001 |
Creatinine, mg/dL | 1.10 (0.77, 1.75) | 0.95 (0.70, 1.45) | 1.40 (0.87, 2.29) | < 0.001 |
Severity of illness | | | | |
SOFA | 5.00 (2.00, 8.00) | 4.00 (2.00, 6.00) | 7.00 (4.00, 11.00) | < 0.001 |
SIRS | 2.62 (0.93) | 2.53 (0.93) | 2.76 (0.92) | < 0.001 |
APS III | 49.96 (21.32) | 43.99 (17.06) | 59.89 (23.86) | < 0.001 |
SAPS II | 38.85 (14.37) | 34.31 (12.34) | 46.38 (14.34) | < 0.001 |
Vital signs | | | | |
SBP, mmHg | 114.84 (105.46, 126.74) | 117.27 (108.18, 128.79) | 110.42 (102.03, 122.18) | < 0.001 |
DBP, mmHg | 62.09 (55.87, 69.25) | 63.15 (57.55, 70.52) | 59.79 (53.77, 67.41) | < 0.001 |
MBP, mmHg | 76.19 (10.80) | 77.75 (10.78) | 73.60 (10.32) | < 0.001 |
SPO2, % | 97.29 (1.93) | 97.45 (1.80) | 97.02 (2.09) | < 0.001 |
Comorbidities | | | | |
Hypertension, n(%) | | | | < 0.001 |
no | 1439 (61.10) | 852 (57.96) | 587 (66.33) | |
yes | 916 (38.90) | 618 (42.04) | 298 (33.67) | |
Diabetes, n(%) | | | | 0.144 |
no | 1657 (70.36) | 1050 (71.43) | 607 (68.59) | |
yes | 698 (29.64) | 420 (28.57) | 278 (31.41) | |
Heart failure, n(%) | | | | < 0.001 |
no | 1725 (73.25) | 1145 (77.89) | 580 (65.54) | |
yes | 630 (26.75) | 325 (22.11) | 305 (34.46) | |
CKD, n(%) | | | | < 0.001 |
no | 1836 (77.96) | 1182 (80.41) | 654 (73.90) | |
yes | 519 (22.04) | 288 (19.59) | 231 (26.10) | |
Stroke, n(%) | | | | 0.806 |
no | 2186 (92.82) | 1366 (92.93) | 820 (92.66) | |
yes | 169 (7.18) | 104 (7.07) | 65 (7.34) | |
COPD, n(%) | | | | 0.002 |
no | 2191 (93.04) | 1386 (94.29) | 805 (90.96) | |
yes | 164 (6.96) | 84 (5.71) | 80 (9.04) | |
Cirrhosis, n(%) | | | | < 0.001 |
no | 1779 (75.54) | 1163 (79.12) | 616 (69.60) | |
yes | 576 (24.46) | 307 (20.88) | 269 (30.40) | |
Treatments | | | | |
MV, n(%) | | | | < 0.001 |
no | 588 (24.97) | 446 (30.34) | 142 (16.05) | |
yes | 1767 (75.03) | 1024 (69.66) | 743 (83.95) | |
Enteral nutrition, n(%) | | | | 0.002 |
no | 2291 (97.28) | 1442 (98.10) | 849 (95.93) | |
yes | 64 (2.72) | 28 (1.90) | 36 (4.07) | |
Abbreviations: GIB, gastrointestinal bleeding; WBC, white blood cell; RBC, red blood cell; RDW, red blood distribution width; PT, prothrombin time; PTT, partial thromboplastin time; INR, international normalized ratio; SOFA, Sequential Organ Failure Assessment; SIRS, Systemic Inflammatory Response Syndrome; APS III, Acute Physiology Score III; SAPS II, Simplified Acute Physiology Score II; SBP, systolic blood pressure; DBP, diastolic blood pressure; MBP, mean blood pressure; SPO2, percutaneous arterial oxygen saturation; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; MV, mechanical ventilation. |
Relationship between serum chloride and 365-day all-cause mortality
Kaplan-Meier survival analyses revealed a significant association between serum chloride and the 365-day all-cause mortality in patients with GIB (Fig. 2). The curves demonstrate a clear stratification of survival probabilities, with the lowest quartile (Q1) exhibiting the lowest survival rate. The log-rank test confirmed the difference in survival distributions across the four groups was statistically significant (P < 0.05).
In critically ill patients with GIB, the Cox proportional hazards model elucidated the relationship between serum chloride and 365-day all-cause mortality (Table 3). Elevated serum chloride levels were associated with a lower risk of death; in unadjusted crude models, the risk of death decreased by 5% for each unit increase in serum chloride (HR = 0.95, 95% CI: 0.94–0.96). This association persisted after adjustment for multiple potential confounders, including comorbidities and severity of illness in Model 2 (HR = 0.97, 95% CI: 0.95–0.99, P = 0.005). Dividing the serum chloride into quartiles, with lower mortality risks for each quartile increase compared to the reference first quartile (Q1). In the fully adjusted Model 2, the fourth quartile (Q4) showed a markedly increased risk (HR = 0.75, 95% CI: 0.57–0.99, P = 0.039).
Table 3
Association between chloride and 365-day all-cause mortality of critically ill patients with GIB
| Model 1 | | Model 2 | | Model 3 |
HR (95%CI) | P | HR (95%CI) | P | HR (95%CI) | P |
Chloride | 0.95 (0.94 ~ 0.96) | < 0.001 | | 0.95 (0.94 ~ 0.96) | < 0.001 | | 0.97 (0.95 ~ 0.99) | 0.005 |
Chloride quartiles | | | | | | | | |
Q1 | 1.00 (Reference) | | | 1.00 (Reference) | | | 1.00 (Reference) | |
Q2 | 0.63 (0.53 ~ 0.74) | < 0.001 | | 0.64 (0.54 ~ 0.76) | < 0.001 | | 0.90 (0.74 ~ 1.10) | 0.299 |
Q3 | 0.43 (0.35 ~ 0.52) | < 0.001 | | 0.45 (0.37 ~ 0.56) | < 0.001 | | 0.72 (0.57 ~ 0.92) | 0.009 |
Q4 | 0.46 (0.39 ~ 0.55) | < 0.001 | | 0.49 (0.41 ~ 0.59) | < 0.001 | | 0.75 (0.57 ~ 0.99) | 0.039 |
HR: Hazard Ratio, CI: Confidence Interval |
Model 1: no adjusted; |
Model 2: adjusted for age, gender, weight, hypertension, diabetes, and heart failure; |
Model3: adjusted for age, gender, weight, WBC, RBC, platelet, RDW, Sodium, potassium, calcium, glucose, anion gap, PT, PTT, urea nitrogen, creatinine, hypertension, diabetes, heart failure, CKD, cirrhosis, hyperlipidemia, COPD, MV, SOFA, APS III, SIRS, SAPS II, SPO2, and enteral nutrition. |
Dose-response relationship
To investigate the potential non-linear association between serum chloride and 365-day all-cause mortality in critically ill patients with GIB, we used an RCS to visualize the association between the two above (Fig. 3). After adjusting for confounders in this study, we further found a significant association between serum chloride and all-cause mortality, while a non-linearity test revealed that the association was nonlinear (P for nonlinear = 0.030). In addition, we divided the population into two groups according to whether serum chloride was higher than 106 mmol/L and found that higher GIB critically ill patients had lower all-cause mortality (Supplementary Fig. 1).