Patient Characteristics
The study sample included 57 patients who were admitted to the ICU with a diagnosis of sepsis. Table 1 summarizes patient characteristics at the time of admission and mean serum Gal-3 levels, as well as levels of additional serum biomarkers CysC, NGAL, and PCT. There was no significant difference in mean age between patients that developed S-AKI and patients that did not develop S-AKI following ICU admission (age: 60.6 ± 1.6 years vs. 59 ± 2.0 years, p = 0.53). APACHE II score was significantly higher among patients that developed S-AKI as compared to patients that did not (18.6 ± 1.2 vs 15.2 ± 1.1, p < 0.05).
Table 1
Characteristics of Patients at ICU Admission
Variable | All Patients (n = 57) | Patients Without Subsequent AKI (n = 30) | Patients With Subsequent AKI (n = 27) | P-Value |
Age (years) | 59 (53–66) | 59 (53-66.8) | 61 (55–66) | 0.53 |
Female sex, n (%) | 32 (56) | 18 (60) | 14 (52) | 0.73 |
ICU stay (days) | 13 (9–15) | 11 (8.3–13) | 15 (11.5–19.5) | < 0.001 |
APACHE II score | 16 (12–21) | 14 (11–17) | 20 (13.5–23) | < 0.05 |
Serum Gal-3 (ng/ml) | 5.7 (3.6–8.9) | 4.2 (3.5-7) | 7.9 (4.7–16.6) | 0.002 |
Serum CysC (µg/ml) | 1.5 (1.1–2.2) | 1.2 (0.9–1.5) | 2 (1.7–4.1) | 0.002 |
Serum NGAL (ng/ml) | 302.9 (175.3-514.2) | 243.4 (170.6-421.9) | 430 (215–561) | 0.406 |
Serum PCT (ng/ml) | 11.2 (3.1–52.4) | 7.7 (2.9–27.3) | 32.6 (3.7–91) | 0.007 |
Table 1. Characteristics of patients in AKI and non-AKI groups. Characteristics were compared using chi-square tests or two-tailed t-tests. Data are presented as median (IQR) or n (%). AKI, acute kidney injury; APACHE II, acute physiology and chronic health evaluation II; CysC, cystatin C; Gal-3, galectin-3; ICU, intensive care unit; IQR, interquartile range; NGAL, neutrophil gelatinase-associated lipocalin; PCT, procalcitonin.
Serum Galectin-3, AKI, and Survival
Of the 57 patients admitted to the ICU, 27 (47%) subsequently developed S-AKI. Additionally, 8 patients (14%) died during the ICU stay: 6 were in the S-AKI group and 2 were in the non-AKI group. Mean serum Gal-3 level was significantly higher among patients that developed subsequent S-AKI as compared to patients that did not develop S-AKI (11.2 ± 1.6 ng/ml vs. 5.3 ± 0.5 ng/ml; p = 0.002) (Fig. 2a). Mean serum Gal-3 level was also significantly higher among patients that died as compared to patients that did not die (18.7 ± 3.6 ng/ml vs. 6.4 ± 0.7 ng/ml; p = 0.01) (Fig. 2b). Using multivariate logistic regression, serum Gal-3 was associated with an increased odds of death before and after adjusting for age, AKI occurrence, and APACHE II score (OR = 1.4 [95% CI 1.1–2.2], p = 0.04). Serum Gal-3 was also associated with an increased odds of AKI before and after adjusting for age and APACHE II score (OR = 1.2 [95% CI 1.1–1.4], p = 0.01). The AUC-ROC for serum Gal-3 predicting subsequent AKI was 0.73, and the AUC-ROC for serum Gal-3 predicting ICU mortality was 0.91 (Fig. 2c).
Rat Cecal Ligation and Puncture Model
In total, 54 animals underwent the CLP procedure: 18 rats in the control group, and 18 rats in each P-MCP-treated group.
Galectin-3
Differences in serum Gal-3 concentrations between control and P-MCP-treated rats are shown in Fig. 3. Baseline serum values were not significantly different among the three rat groups (F = 3.2, p = 0.15). In both P-MCP-treated groups, mean serum Gal-3 levels were significantly lower than control at 2 h post procedure (400 mg P-MCP: 0.83 ± 0.05 ng/ml vs 1.63 ± 0.22 ng/ml, p = 0.003; 1200 mg P-MCP: 0.82 ± 0.08 ng/ml vs 1.63 ± 0.22 ng/ml, p = 0.001). Serum Gal-3 levels in all groups peaked at 2 h post-CLP and were significantly higher than their respective baseline values (control: p = 0.0001; 400 mg P-MCP: p = 0.001; 1200 mg P-MCP: p = 0.004). Gal-3 levels in all three groups subsided rapidly and were no longer significantly elevated at 8 h post CLP compared to their respective baseline values. The AUC-ROC for serum Gal-3 predicting subsequent AKI was 0.75, and the AUC-ROC for serum Gal-3 predicting ICU mortality was 0.88 (Fig. 4a).
Interleukin-6
Differences in serum IL-6 concentrations between control and P-MCP-treated rats are shown in Fig. 5. Baseline values were not significantly different among the three groups (F = 3.2, p = 0.56). Concentrations of serum IL-6 in both P-MCP-treated groups were significantly lower than control values at all post-CLP time points, with the maximum difference at 24 h post-CLP (400 mg P-MCP: 468.0 ± 89.5 pg/ml vs 1344.5 ± 1103.4 pg/ml, p = 0.015; 1200 mg P-MCP: 458.8 ± 251.9 pg/ml vs 1344.5 ± 1103.4, p = 0.02) (Fig. 5). Serum IL-6 in the control group increased over time and peaked at 24 h post-CLP (3,344.5 ± 1,103.4 pg/ml). In both Gal-3 inhibitor groups, IL-6 levels decreased between 8 h and 24 h post-procedure. The reduction in IL-6 levels between 8 h and 24 h post-CLP was statistically significant in the 1200 mg P-MCP group only (631 [IQR 494.7-828.2] pg/ml vs 187.2 [IQR 156.5-272.6] pg/ml, p = 0.006). The AUC-ROC for serum IL-6 predicting subsequent AKI was 0.74, and the AUC-ROC for serum IL-6 predicting ICU mortality was 0.81 (Fig. 4b).
Creatinine
Differences in serum Cr concentrations between control and P-MCP-treated rats are shown in Fig. 6. Baseline values were not significantly different between the three groups (F = 3.2, p = 0.54). Serum Cr concentration in the 1200 mg P-MCP-treated group was significantly lower than control at all time points post-CLP, reaching a peak difference at 24 h post-procedure (51.8 ± 6.7 µmol/l vs 104.6 ± 18.8 µmol/l, p = 0.016). Serum Cr concentration in the 400 mg P-MCP-treated group was significantly lower than control group at 2 h and 24 h post-CLP (2 h: 26.3 ± 0.9 µmol/l vs 31.3 ± 1.6 µmol/l, p = 0.009; 24 h: 53.5 ± 4.2 µmol/l vs 104.6 ± 18.8 µmol/l, p = 0.005).
AKI
The percent of animals that developed AKI at 24 h post-CLP in the control and P-MCP-treated groups are shown in Fig. 7. As determined by RIFLE criteria, 89% (16) of rats in the control group developed AKI, while 44% (8) of rats developed AKI in the 400 mg P-MCP group and 44% (8) developed AKI in the 1200 mg P-MCP group. There was a statistically significant difference in AKI rate between the control group and both the 400 mg P-MCP group (p = 0.007) and 1200 mg P-MCP group (p = 0.007). In the control group, 8 (44%) rats were classified as RIFLE-R, 2 (11%) rats were RIFLE-I, and 6 (33%) rats were RIFLE-F. In the 400 mg P-MCP group, 1 (6%) rat was classified as RIFLE-R, 4 (22%) rats were RIFLE-I, and 3 (17%) rats were RIFLE-F. In the 1200 mg P-MCP group, 1 (6%) rat was classified as RIFLE-R, 4 (22%) rats were RIFLE-I, and 3 (17%) rats were RIFLE-F.
Survival
Survival data are shown in Fig. 8. Each group began with 18 rats prior to CLP. At 7 days post-CLP, 11 (61%) rats had died in the control group compared to 5 (28%) rats in the 400 mg P-MCP group and 4 (22%) rats in the 1200 mg P-MCP group. Mortality was significantly lower in both P-MCP-treated groups as compared to controls (400 mg P-MCP: p = 0.03; 1200 mg P-MCP: p = 0.001).