Nonlinear association between estimated plasma volume status and acute kidney injury in acute pancreatitis patients

Abstract

Background: Acute pancreatitis (AP), a severe pancreatic inflammatory condition, with a mortality rate reaching up to 40%. Recently, AP shows a steadily elevating prevalence, which causes the greater number of hospital admissions, imposing the substantial economic burden. Acute kidney injury (AKI) complicates take up approximately 15% of AP cases, with an associated mortality rate of 74.7%-81%.

Aim: To evaluate the efficacy of estimated plasma volume status (ePVS) in forecasting AKI in patients with AP.

Methods: In this retrospective cohort study, AP cases were recruited from the First College of Clinical Medical Science of China Three Gorges University between January 2019 and October 2023. Electronic medical records were adopted for data extraction, including demographic data and clinical characteristics. The association between ePVS and AKI was analyzed using multivariate logistic regression models, with potential confounders being adjusted. Nonlinear relationship was examined with smooth curve fitting, and infection points were calculated. Further analyses were performed on stratified subgroups and interaction tests were conducted.

Results: Among the 1508 AP patients, 251 (16.6%) developed AKI. ePVS was calculated using Duarte (D-ePVS) and Kaplan-Hakim (KH-ePVS) formulas. After adjusting for covariates, the AKI risk exhibited 46% [odds ratio (OR) = 1.46, 95% confidence interval (CI): 0.96-2.24] and 11% (OR = 1.11, 95%CI: 0.72-1.72) increases in the low tertile (T1) of D-ePVS and KH-ePVS, respectively, and 101% (OR = 2.01, 95%CI: 1.31-3.05) and 51% (OR = 1.51, 95%CI: 1.00-2.29) increases in the high tertile (T3) relative to the reference tertile (T2). Nonlinear curve fitting revealed a U-shaped association of D-ePVS with AKI and a J-shaped association for KH-ePVS, with inflection points at 4.3 dL/g and -2.8%, respectively. Significant interactions were not observed in age, gender, hypertension, diabetes mellitus, sequential organ failure assessment score, or AP severity (all P for interaction > 0.05).

Conclusion: Our results indicated that ePVS demonstrated the nonlinear association with AKI incidence in AP patients. A U-shaped curve was observed with an inflection point at 4.3 dL/g for the Duarte formula, and a J-shaped curve at -2.8% for the Kaplan-Hakim formula.

Keywords: Acute kidney injury; Acute pancreatitis; Cohort study; Duarte formula; Estimated plasma volume status; Kaplan-Hakim formula.

Figure 1

Flowchart of the screening and enrollment of study participants. AKI: Acute kidney injury.

Figure 2

The nonlinear association between estimated plasma volume status and acute kidney injury in patients with acute pancreatitis. A: Duarte formula-based estimated plasma volume status; B: Kaplan-Hakim formula-based estimated plasma volume status. Adjusted for all covariates as model 3. Solid lines indicate the odds ratio of acute kidney injury and dotted lines represent the corresponding 95% confidence interval. Odds ratio = 1 was set as the reference line (takes the upper limit of 99.9%). D-ePVS: Duarte formula-based estimated plasma volume status; KH-ePVS: Kaplan-Hakim formula-based estimated plasma volume status; AKI: Acute kidney injury.

Figure 3

The relationship between estimated plasma volume status and acute kidney injury according to basic features. Except for the stratification component itself, each stratification factor was adjusted for all covariates as model 3. D-ePVS: Duarte formula-based estimated plasma volume status; KH-ePVS: Kaplan-Hakim formula-based estimated plasma volume status; OR: Odds ratio; CI: Confidence interval; AP: Acute pancreatitis; SOFA: Sequential organ failure assessment.