# The value of transpulmonary thermodilution parameters in predicting hemodynamic instability in intensive care patients undergoing continuous renal replacement therapy

## TL;DR

This study shows that low cardiac index and high mean arterial pressure can predict hemodynamic instability in ICU patients undergoing kidney therapy.

## Contribution

The study identifies novel independent predictors of hemodynamic instability during CRRT using transpulmonary thermodilution parameters.

## Key findings

- Low cardiac index and high mean arterial pressure are independent predictors of HIRRT.
- HIRRT occurred in 55.5% of patients, often within the first hour of CRRT.
- Higher systemic vascular resistance index was associated with HIRRT but not independently predictive.

## Abstract

Hemodynamic instability related to renal replacement therapy (HIRRT) is a serious complication of continuous renal replacement therapy (CRRT) in critically ill patients with acute kidney injury (AKI), significantly increasing mortality risk. The pathophysiology involves complex interactions between cardiac output and systemic vascular resistance. Therefore, identifying early and reliable predictive parameters for HIRRT is clinically crucial.

This prospective, observational cohort study was conducted in an Internal Medicine Intensive Care Unit between July 2023 and October 2024. Thirty-six patients undergoing CRRT with invasive monitoring via the PiCCO® device were enrolled. Hemodynamic parameters were recorded. HIRRT was defined exclusively as a decrease in systolic blood pressure ≥ 20 mmHg or a decrease in mean arterial pressure ≥ 10 mmHg. Changes in vasopressor or inotropic therapy were not included in the diagnostic criteria for HIRRT. To avoid classifying transient or clinically insignificant fluctuations as HIRRT, the decrease in blood pressure had to be present in at least two consecutive measurements (approximately 10–15 min). This definition is consistent with prior CRRT studies that use objective blood pressure thresholds. Statistical analyses included the Mann-Whitney U test, chi-square test, logistic regression, and receiver operating characteristic (ROC) curve analysis.

The mean age of the patients was 68.1 ± 18.3 years, and 69.4% were male. The most common indication for CRRT was uremic complications (52.8%), and the most frequent comorbidity was hypertension (55.6%). A total of 83.3% of patients were on vasopressor support, and 38.9% were on mechanical ventilation. HIRRT occurred in 55.5% of patients during CRRT, and in 60% of these patients, it developed within the first hour. The HIRRT group had a significantly lower cardiac index (CI) (p = 0.002) and a higher systemic vascular resistance index (SVRI) (p = 0.003). Additionally, the HIRRT group had higher baseline mean arterial pressure (p = 0.015) and baseline diastolic blood pressure (p = 0.003), and a significantly greater total ultrafiltration volume (p = 0.018). Multivariate analysis identified a low CI (p = 0.018) and a high mean arterial pressure (p = 0.031) as independent predictors. ROC analysis revealed that the optimal cut-off value for the mean arterial pressure was 78 mmHg (AUC: 0.759, 95% CI: 0.597–0.922) and that for the CI was 2.61 L/min/m² (AUC: 0.794, 95% CI: 0.642–0.946).

Low baseline cardiac index and elevated mean arterial pressure emerged as independent predictors of HIRRT, whereas higher SVRI was significant only in univariate analysis, indicating an association but not independent predictive value. Collectively, these findings suggest that a reduced capacity to increase cardiac output in response to hemodynamic stress, together with alterations in vascular tone regulation, may play a central role in the development of HIRRT. While these parameters may facilitate early identification of high-risk patients and support individualized management strategies, further validation in larger, multicenter cohorts is required.

Not applicable.