Baseline characteristics
A total of 1044 patients who underwent CRRT in the ICU were included and divided into quintiles based on PLR. The patients’ mean age was 62.3 years, and 63.1% were male. The mean and median PLR values were 241.3 and 150.6 (83.0–281.6), respectively. And the PLR quintiles 1, 2, 3, 4, and 5 values ranged from < 71.4; 71.4 ≤ PLR < 122.3; 122.3 ≤ PLR < 184.4; 184.4 ≤ PLR < 325.7; and ≥ 325.7, respectively. Detailed baseline characteristics according to PLR quintiles are indicated in Table 1. No significant differences in CCI or other comorbidities were observed among the quintiles. However, age and body weight at ICU admission differed between quintiles (P = 0.011 and P = 0.023, respectively). Notably, the SOFA and APACHE II scores showed significant differences across the quintiles, and both scores were highest in the first quintile (all P < 0.01). The etiology of AKI differed according to PLR quintiles, and septic AKI occurred most frequently in the fifth quintile (56.9%). Laboratory findings, including white blood cell, lymphocyte, and platelet counts; hemoglobin, albumin, BUN, and creatinine levels; and arterial pH, differed significantly among PLR quintiles.
Table 1
| Quintile 1 (n = 208) | Quintile 2 (n = 210) | Quintile 3 (n = 209) | Quintile 4 (n = 208) | Quintile 5 (n = 209) | P |
PLR | 37.4 ± 18.8 | 96.6 ± 14.3 | 151.7 ± 17.7 | 245.3 ± 41.3 | 675.5 ± 392.3 | < 0.001 |
Age, years | 62.5 ± 16.0 | 65.7 ± 15.7 | 64.1 ± 17.3 | 67.1 ± 14.5 | 66.9 ± 15.2 | 0.011 |
Sex, male, n (%) | 138 (66.4) | 135 (64.3) | 129 (61.7) | 128 (61.5) | 129 (61.7) | 0.804 |
ICU admission body weight, kg | 64.5 ± 13.3 | 62.9 ± 12.0 | 61.9 ± 12.3 | 62.5 ± 12.9 | 60.3 ± 11.5 | 0.023 |
ICU admission BMI, kg/m2 | 23.5 ± 4.3 | 23.4 ± 3.9 | 23.9 ± 11.4 | 23.3 ± 4.5 | 22.8 ± 3.5 | 0.575 |
CCI | 3.9 ± 2.4 | 4.0 ± 2.2 | 4.2 ± 2.4 | 4.5 ± 2.5 | 4.4 ± 2.4 | 0.063 |
Causes of AKI, n (%) | | | | | | < 0.001 |
Septic | 77 (37.0) | 65 (31.0) | 65 (31.1) | 68 (32.7) | 119 (56.9) | |
Ischemic | 101 (48.6) | 113 (53.8) | 111 (53.1) | 112 (53.8) | 62 (29.7) | |
Postoperative | 13 (6.3) | 12 (5.7) | 15 (7.2) | 10 (4.8) | 8 (3.8) | |
Nephrotoxic | 9 (4.3) | 10 (4.8) | 10 (4.8) | 8 (3.8) | 12 (5.7) | |
Others | 8 (3.8) | 10 (4.8) | 8 (3.8) | 10 (4.8) | 8 (3.8) | |
Comorbidities, n (%) | | | | | | |
Hypertension | 59 (29.8) | 63 (32.1) | 65 (35.3) | 80 (42.8) | 75 (38.7) | 0.063 |
Diabetes | 58 (27.9) | 59 (28.1) | 56 (26.9) | 66 (31.7) | 53 (25.3) | 0.684 |
Congestive heart failure | 10 (4.8) | 21 (10.0) | 19 (9.1) | 24 (11.5) | 14 (6.7) | 0.101 |
Cerebrovascular accident | 16 (7.7) | 21 (10.0) | 19 (9.1) | 19 (9.1) | 18 (8.6) | 0.947 |
Malignancy | 15 (7.2) | 13 (6.2) | 10 (4.8) | 17 (8.1) | 27 (12.9) | 0.098 |
SOFA | 14.1 ± 5.0 | 12.7 ± 4.7 | 12.2 ± 4.6 | 11.8 ± 4.4 | 11.7 ± 4.6 | < 0.001 |
APACHE II | 27.6 ± 8.5 | 25.9 ± 7.9 | 24.7 ± 8.1 | 24.4 ± 7.3 | 25.6 ± 8.1 | 0.001 |
Laboratory findings | | | | | | |
WBC count, ×103/µL | 20.2 ± 11.3 | 12.7 ± 7.9 | 11.9 ± 7.3 | 12.4 ± 7.9 | 11.5 ± 8.9 | < 0.001 |
Neutrophil, ×103/µL | 10.9 ± 9.0 | 10.7 ± 7.4 | 10.2 ± 6.8 | 11.0 ± 7.4 | 10.5 ± 8.4 | 0.877 |
Lymphocyte, ×103/µL | 4.2 ± 1.3 | 1.3 ± 0.9 | 1.0 ± 0.6 | 0.7 ± 0.4 | 0.4 ± 0.3 | < 0.001 |
Platelet count, ×103/µL | 81.6 ± 84.4 | 128.0 ± 82.9 | 146.9 ± 82.0 | 169.1 ± 93.3 | 203.4 ± 145.2 | < 0.001 |
Hemoglobin, g/dL | 10.1 ± 3.3 | 10.5 ± 2.9 | 10.5 ± 2.8 | 10.7 ± 2.4 | 10.0 ± 2.5 | 0.039 |
Sodium, mEq/L | 138.1 ± 8.7 | 139.2 ± 7.7 | 138.3 ± 7.8 | 137.7 ± 7.6 | 137.3 ± 8.1 | 0.142 |
Potassium, mEq/L | 4.9 ± 1.2 | 4.7 ± 1.1 | 4.9 ± 1.2 | 4.8 ± 1.0 | 4.8 ± 1.1 | 0.270 |
Albumin, g/dL | 2.8 ± 0.7 | 3.0 ± 0.7 | 3.0 ± 0.7 | 3.1 ± 0.6 | 2.8 ± 0.7 | < 0.001 |
BUN, mg/dL | 53.1 ± 36.1 | 54.4 ± 31.0 | 63.7 ± 40.0 | 58.6 ± 33.2 | 65.0 ± 37.0 | < 0.001 |
Creatinine, mg/dL | 3.4 ± 3.0 | 3.3 ± 2.0 | 4.1 ± 3.3 | 3.9 ± 3.2 | 4.0 ± 3.1 | 0.010 |
Arterial pH | 7.23 ± 0.15 | 7.26 ± 0.16 | 7.27 ± 0.23 | 7.28 ± 0.12 | 7.28 ± 0.12 | 0.005 |
Data are presented as mean ± standard deviation or number (%) |
Abbreviations: PLR platelet-to-lymphocyte ratio; ICU intensive care unit; BMI body mass index; CCI Charlson Comorbidity Index; AKI acute kidney injury; SOFA Sequential Organ Failure Assessment; APACHE II Acute Physiology and Chronic Health Evaluation II; WBC white blood cell; BUN blood urea nitrogen.
In-hospital mortality outcomes
Among patients treated using CRRT, 651 (62.4%) died during the follow-up period. The mean length of hospital stay was 26.1 days (range, 5–32 days). Patients were admitted to the ICU for an average of 12.5 days and received CRRT treatment for approximately 4.3 days. Overall, PLR (the outcomes of this study) as a continuous variable, showed a U-shaped relationship in the cubic spline regression model (Fig. 1). In-hospital mortality and other in-hospital information according to PLR quintiles are presented in Table 2. Briefly, we observed significantly higher in-hospital mortality rates in the first [157 (75.5%)] and fifth [148 (70.8%)] quintile groups than in the third quintile group [93 (44.5%)], which had the lowest mortality rate. Furthermore, the 30-day and 90-day mortality rates showed a significant increase at both ends of the quintiles (Table 2). The ratio of patients who received mechanical ventilation and used vasopressors differed significantly by quintile; with the highest and lowest rates in the first and fifth quintiles, respectively (both P < 0.05).
Table 2. In-hospital information of PLR quintile groups
|
Quintile 1
|
Quintile 2
|
Quintile 3
|
Quintile 4
|
Quintile 5
|
P
|
30-day mortality, n (%)
|
150 (72.1)
|
120 (57.1)
|
90 (43.1)
|
112 (53.9)
|
136 (65.1)
|
<0.001
|
90-day mortality, n (%)
|
155 (74.5)
|
129 (61.4)
|
92 (44.0)
|
114 (54.8)
|
137 (65.6)
|
<0.001
|
In-hospital mortality, n (%)
|
157 (75.5)
|
132 (62.9)
|
93 (44.5)
|
121 (58.2)
|
148 (70.8)
|
<0.001
|
CRRT duration, days
|
4.0 ± 4.1
|
4.8 ± 4.8
|
4.3 ± 5.0
|
4.3 ± 3.7
|
4.2 ± 3.6
|
0.341
|
ICU duration, days
|
10.5 ± 17.0
|
12.8 ± 14.7
|
13.4 ± 19.2
|
13.4 ± 22.2
|
12.5 ± 28.9
|
0.646
|
ICU admission to CRRT initiation, days
|
2.7 ± 4.6
|
2.9 ± 5.3
|
2.7 ± 4.1
|
3.1 ± 11.5
|
2.8 ± 4.3
|
0.957
|
Length of hospital stay, days
|
24.1 ± 37.6
|
24.5 ± 33.1
|
27.5 ± 32.8
|
27.3 ± 47.5
|
27.1 ± 37.1
|
0.812
|
Need for mechanical ventilation, n (%)
|
134 (64.4)
|
134 (63.8)
|
118 (56.5)
|
121 (58.7)
|
98 (47.1)
|
0.002
|
Vasopressor requirements, n (%)
|
175 (84.1)
|
157 (74.8)
|
163 (78.0)
|
151 (73.3)
|
149 (71.6)
|
0.024
|
Prescribed target clearance, mL/kg/hour
|
35.8 ± 9.3
|
35.5 ± 9.9
|
35.7 ± 10.2
|
36.0 ± 8.7
|
35.5 ± 9.4
|
0.989
|
Data are presented as mean ± standard deviation or number (%)
Abbreviation: CRRT continuous renal replacement therapy; ICU intensive care unit
Correlation between PLR ratio and mortality
Kaplan-Meier curves showed a significant difference in cumulative survival according to PLR quintiles (Fig. 2). The first and fifth quintiles had the highest mortality rates, whereas the third quintile had the lowest mortality rate. Similar differences were observed in the 30-day and 90-day mortality rates (all P < 0.05; Fig. 2). In model 4, after adjusting for various factors, including age, sex, comorbidities, and severity indicators, compared with the third quintile of PLR (reference group), the lowest (adjusted HR [aHR] 1.94, 95% CI 1.44–2.62, P < 0.001) and highest (aHR 1.60, 95% CI 1.18–2.18; P = 0.002) quintiles had significantly greater in-hospital mortality (Table 3). Figure 3 shows the model 4’s aHRs of the 30-day (quintile 1: aHR 1.93, 95% CI 1.43–2.61, P < 0.001; quintile 5: aHR 1.63, 95% CI 1.20–2.22, P = 0.002), 90-day (quintile 1: aHR 1.97, 95% CI 1.47–2.65, P < 0.001; quintile 5: aHR 1.70, 95% CI 1.26–2.30, P < 0.001), and in-hospital mortality, showing U-shaped aHRs across the PLR quintiles. Supplementary Table 1 presents the detailed results of the Cox regression analysis for 30-day and 90-day mortality across the PLR quintiles (see Additional file 1).
Table 3. Cox regression analyses for in-hospital mortality in PLR quintile groups
|
Model 1
|
|
Model 2
|
|
Model 3
|
|
Model 4
|
|
|
HR (95% CI)
|
P
|
aHR (95% CI)
|
P
|
aHR (95% CI)
|
P
|
aHR (95% CI)
|
P
|
Quintile 1
|
2.19 (1.69–2.83)
|
<0.001
|
2.34 (1.78–3.09)
|
<0.001
|
2.26 (1.70–3.00)
|
<0.001
|
1.94 (1.44–2.62)
|
<0.001
|
Quintile 2
|
1.55 (1.19–2.03)
|
0.001
|
1.50 (1.13–1.99)
|
0.006
|
1.46 (1.09–1.95)
|
0.011
|
1.42 (1.05–1.94)
|
0.024
|
Quintile 3
|
Reference
|
|
Reference
|
|
Reference
|
|
Reference
|
|
Quintile 4
|
1.29 (0.98–1.70)
|
0.068
|
1.28 (0.95–1.72)
|
0.102
|
1.26 (0.93–1.70)
|
0.135
|
1.28 (0.93–1.75)
|
0.128
|
Quintile 5
|
1.57 (1.20–2.04)
|
<0.001
|
1.62 (1.22–2.14)
|
<0.001
|
1.57 (1.17–2.10)
|
0.003
|
1.60 (1.18–2.18)
|
0.002
|
Model 1: unadjusted
Model 2: adjusted for age, sex, and body weight
Model 3: adjusted for age, sex, body weight, CCI, hypertension, and malignancy.
Model 4: adjusted for age, sex, body weight, CCI, hypertension, malignancy, SOFA score, APACHE II score, mechanical ventilator use, and vasopressor use.
Abbreviation: PLR platelet-to-lymphocyte ratio; HR hazard ratio; aHR adjusted hazard ratio; CI confidence interval; CCI Charlson Comorbidity Index; SOFA Sequential Organ Failure Assessment; APACHE II Acute Physiology and Chronic Health Evaluation II.
Subgroup analysis
We divided the patients into subgroups according to age, sex, body mass index, comorbidities (hypertension and diabetes), SOFA and APACHE II scores. The first and fifth quintiles showed significant associations with increased in-hospital mortality in most subgroups including those with older age, female sex, hypertension, diabetes, and higher SOFA score (Table 4).
Table 4. Subgroup analyses for in-hospital mortality in PLR quintile groups
|
Quintile 1
|
Quintile 2
|
Quintile 3
|
Quintile 4
|
Quintile 5
|
|
aHR (95% CI)
|
aHR (95% CI)
|
Reference
|
aHR (95% CI)
|
aHR (95% CI)
|
Age, years
|
|
|
|
|
|
≤68.0
|
1.67 (1.10–2.54)a
|
1.39 (0.89–2.16)
|
1.00
|
1.28 (0.80–2.03)
|
1.50 (0.96–2.34)
|
>68.0
|
2.43 (1.59–3.73)c
|
1.54 (1.01–2.35)a
|
1.00
|
1.29 (0.84–2.00)
|
1.85 (1.22–2.80)b
|
Sex
|
|
|
|
|
|
Male
|
1.54 (0.95–2.50)
|
1.23 (0.76–2.00)
|
1.00
|
1.17 (0.71–1.94)
|
1.69 (1.05–2.72)a
|
Female
|
2.08 (1.41–3.07) c
|
1.65 (1.11–2.46)a
|
1.00
|
1.32 (0.88–1.99)
|
1.69 (1.14–2.50)b
|
BMI, kg/m2
|
|
|
|
|
|
22.9
|
2.04 (1.33–3.12)b
|
1.74 (1.14–2.64)b
|
1.00
|
1.38 (0.89–2.16)
|
1.64 (1.09–2.48)a
|
>22.9
|
2.02 (1.32–3.08)b
|
1.30 (0.83–2.03)
|
1.00
|
1.25 (0.80–1.96)
|
1.86 (1.19–2.90)b
|
Hypertension
|
|
|
|
|
|
Yes
|
2.59 (1.52–4.42)c
|
1.57 (0.91–2.72)
|
1.00
|
1.60 (0.93–2.73)
|
2.39 (1.40–4.07)b
|
No
|
1.73 (1.21–2.48)b
|
1.37 (0.95–1.98)
|
1.00
|
1.12 (0.76–1.66)
|
1.39 (0.97–2.01)
|
DM
|
|
|
|
|
|
Yes
|
2.45 (1.35–4.48)b
|
1.98 (1.08–3.62)a
|
1.00
|
1.91 (1.01–3.61)a
|
2.53 (1.37–4.68)b
|
No
|
1.70 (1.20–2.41)b
|
1.32 (0.92–1.88)
|
1.00
|
1.12 (0.78–1.61)
|
1.42 (1.00–2.01)a
|
SOFA
|
|
|
|
|
|
13.0
|
1.59 (0.97–2.61)
|
1.60 (0.99–2.58)
|
1.00
|
1.00 (0.62–1.64)
|
1.73 (1.11–2.71)a
|
>13.0
|
2.16 (1.46–3.17)c
|
1.49 (1.00–2.22)
|
1.00
|
1.57 (1.03–2.39)a
|
1.62 (1.07–2.46)a
|
APACHE II
|
|
|
|
|
|
26.0
|
1.74 (1.06–2.86)a
|
1.95 (1.23–3.08)b
|
1.00
|
1.48 (0.93–2.36)
|
1.69 (1.08–2.66)a
|
>26.0
|
1.99 (1.36–2.93)c
|
1.21 (0.80–1.83)
|
1.00
|
1.15 (0.74–1.77)
|
1.62 (1.08–2.44)a
|
aP <0.05; bP <0.01; cP <0.001.
The statistical values are based on the multivariable Cox regression model 4 except for the variables corresponding to each subgroup.
Abbreviations: PLR platelet-to-lymphocyte ratio; aHR adjusted hazard ratio; CI confidence interval; BMI body mass index; DM diabetes mellitus; SOFA Sequential Organ Failure Assessment; APACHE II Acute Physiology and Chronic Health Evaluation II.