Clinical study of D-lactate and intestinal fatty acid binding protein in evaluating disease severity and prognosis of ICU patients: secondary analysis of a prospective, multicenter clinical study

Abstract

To analyze the relationship between D-lactate, intestinal fatty acid binding protein (I-FABP) and the severity as well as the prognosis of patients in intensive care unit (ICU). Using a retrospective approach, the data derived from a prospective, randomized, single-blind, multicenter clinical study published earlier by our group were further analyzed to evaluate the effect of fat-modified enteral nutrition (EN) suspension on the intestinal barrier in ICU patients. In this study, a total of 141 patients were recruited from 7 ICUs in South China, and 15 healthy volunteers were included as healthy control group. Clinical data of patients were collected, including gender, age, disease severity related indicators such as acute physiology and chronic health evaluation II (APACHE II), sequential organ failure assessment (SOFA), C-reactive protein (CRP) and initial ICU diagnosis, as well as prognostic indicators such as length of stay in ICU, length of stay in hospital and prognosis of patients at 28 days. To compare the difference of serum D-lactate and I-FABP between ICU patients and healthy control group on day 0 (the day before EN reached 500 mL) and day 5 (EN ≥ 2 092 kJ/d for 5 days). According to the hemodynamic and/or mechanical ventilation status on day 5 (compared with day 0), 141 patients were divided into the improvement group (101 cases) and the non-improvement group (40 cases), and the changes of D-lactate and I-FABP in the two groups were observed. Spearman correlation analysis was used to analyze the correlation between serum D-lactate, I-FABP and the severity of the disease, as well as the predictive value of dynamic changes of D-lactate and I-FABP on the prognosis of 28 days. Compared with the healthy volunteers, the serum D-lactate and I-FABP levels of ICU patients on day 0 were significantly increased [D-lactate (mg/L): 10.82 (3.31, 25.48) vs. 6.63 (1.54, 17.70), I-FABP (ng/L): 519.60 (159.06, 1 362.14) vs. 84.40 (30.78, 108.57), both P < 0.01], and D-lactate and I-FABP on day 5 were both decreased compared with the levels on day 0, but still higher than the healthy volunteers. I-FABP in the improvement group was significantly lower than that in the non-improvement group on day 0, and there was no significant difference in D-lactate levels between the two groups, D-lactate and I-FABP in both groups were significantly lower on day 5 than those on day 0, and D-lactate and I-FABP in the improvement group on day 5 were significantly lower than those in the non-improvement group [D-lactate (mg/L): 7.61 (1.71, 27.22) vs. 9.38 (2.09, 20.56), I-FABP (ng/L): 378.65 (152.56, 864.62) vs. 521.21 (205.93, 1 413.11), all P < 0.05]. Correlation analysis showed that D-lactate was significantly positively correlated with APACHE II score and SOFA score on day 0 and day 5 (R12 = 0.367, P < 0.001; R22 = 0.240, P = 0.012); I-FABP was significantly positively correlated with APACHE II score on day 0 (R2 = 0.264, P = 0.004); D-lactate on day 5 and I-FABP on day 0 and day 5 were significantly negatively correlated with prognosis on day 28 (R12 = -0.203, P = 0.022; R22 = -0.208, P = 0.023; R32 = -0.211, P = 0.021). The area under curve (AUC) analysis showed that D-lactate on day 5 and I-FABP on day 0 and day 5 had independent predictive value for 28-day prognosis, with AUC of 0.634, 0.638 and 0.652, P values of 0.023, 0.024 and 0.017, 95% confidence intervals (95%CI) of 0.533-0.734, 0.523-0.754 and 0.525-0.778, respectively. When the cut-off values were 7.71 mg/L, 593.55 ng/L and 468.10 ng/L, the sensitivity were 51.5%, 68.5% and 75.3%, and the specificity were 91.0%, 60.0% and 60.0%, respectively. Serum D-lactate and I-FABP were increased significantly and decreased with the improvement of the condition, and these two molecular biomarkers have certain value in predicting the prognosis of ICU patients.