Effect of therapeutic plasma exchange on acquired hypocholesterolemia in patients with septic shock: a post hoc analysis of the two exchange trials

To the editor

Cholesterol, a sterol lipid with pleiotropic functions, has recently gained increasing interest in sepsis pathobiology [1]. Cholesterol bound to its associated lipoproteins such as high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C) plays diverse roles in the human body, including immunomodulatory functions and bacterial toxin scavenging [1]. Moreover, cholesterol concentrates in lipid rafts - lipid-rich domains within cell membranes - where it is crucial for maintenance of cellular membrane integrity, membrane fluidity, transepithelial resistance and downstream signaling cascades including the beta-adrenergic pathways, which is of relevance in sepsis [1]. It has been demonstrated that serum total cholesterol (TC) levels, HDL-C and LDL-C, but not triglycerides, fall early in sepsis and that the magnitudes of the declines and subsequent recovery are directly linked to patient prognosis and outcomes [1].

A recently emerging approach towards explorative sepsis treatment is therapeutic plasma exchange (TPE) [2]. Previous studies have suggested faster hemodynamic stabilization and clearance of lactate in septic shock patients undergoing TPE [3, 4]. TPE removes circulating injurious substances from the patients´ circulation while replenishing depleted, protective molecules [3]. In analogy, TPE could influence circulating lipid levels, e.g. through elimination of cholesterol-lowering substances such as bacterial toxins, or by directly replenishing cholesterol. However, so far, no data on TPE and its impact on cholesterol levels in septic patients exist.

In this post-hoc secondary analysis from both a prospective single-center non-randomized study (EXCHANGE-Pilot) [4] and a prospective, open-label RCT (EXCHANGE-1) [3], trial data and blood samples were obtained from a total of 49 patients (n = 14 receiving standard of care (SOC) and n = 35 receiving one additional TPE) with early and severe septic shock (onset < 24 h) requiring an norepinephrine (NE) dose of ≥ 0.4 µg/kg/min (Suppl. Figure 1). No significant differences between SOC and TPE groups concerning demographics, disease severity, comorbidities, side of infection and type of causative pathogen at study inclusion were seen (Suppl. Table 1).

Concentrations of TC, HDL-C, LDL-C and triglycerides in serum of septic patients undergoing early TPE compared to SOC were assessed at study inclusion/randomization and after 6 h. Hypocholesterolemia was observed in both SOC and TPE group to comparable extent at baseline (p = 0.8) (Fig. 1, Panel A). TC did not further change within 6 h after randomization in the SOC group, but was significantly increased by TPE (p < 0.001) (Fig. 1, Panel A(a)). LDL- and HDL-C were similar reduced at baseline in both groups, respectively, and increased in the TPE group 6 h after randomization (p < 0.001) (Fig. 1, Panel A(b-c)). Triglycerides, in contrast, were not reduced at baseline, yet reduced within the normal range in the TPE group (p = 0.004) (Fig. 1, Panel A(d)). Median changes for different cholesterol fractions between baseline and 6 h after randomization are shown in Suppl. Figure 2.

Fig. 1.

Impact of TPE on acquired hypocholesterolemia in septic shock patients and effect of a cholesterol concentrate on endothelial permeability in vitro. Panel A shows serum concentrations of (a) total cholesterol (TC), (b) LDL-cholesterol, (c) HDL-cholesterol and (d) triglycerides as violin plots at study inclusion and 6 h after randomization in patients with septic shock who received either standard of care (SOC) alone or SOC in combination with therapeutic plasma exchange (TPE). Blood samples were drawn at study inclusion/randomization and after 6 h. For blood sampling serum was obtained by centrifugation, divided into aliquots and stored at − 80° until assayed. Circulating TC, HDL-, LDL-C, and triglyceride blood concentrations were quantified using enzymatic colorimetric tests on a cobas 8000 automated platform (Roche Diagnostics, Mannheim, Germany). Panel B demonstrates the effect of cholesterol concentrate on endothelial permeability in vitro. Human umbilical vein endothelial cells (HUVECs) were grown in either low serum media (control medium, 0.625%) or medium containing cholesterol concentrate (cholesterol medium, 0.5 nM) (a). Additionally, cells were either stimulated with 50 ng/ml TNFa (b) or 5 nM Thrombin (c) and transendothelial electrical resistance (TER) was measured at different time points by an electric cell-substrate impedance sensing (ECIS) system. Values were plotted over time. Resistance data were normalized to the resistance values at time of media change.Impedance measuring was performed with the ECIS^® Z-Theta instrument (Applied Biophysics Inc, NY, USA) to monitor the effects of total cholesterol on the barrier integrity of HUVECs over time. Paired-t-test or Wilcoxon signed-rank test was used as appropriate for within-group effects between the chosen two time points (randomization, 6 h after randomization). Comparisons between groups were analyzed by means of Mann-Whitney U test. Wilcoxon matched-paired signed rank tests were performed for comparing cholesterol and control longitudinal TER results in the ECIS ex-vivo experiments. For all statistical analyses a two-tailed p-value < 0.05 was considered statistically significant. GraphPad Prism 7 (Graph Pad, La Jolla, CA, USA), SPSS Statistics Version 25 (SPSS Inc., Chicago, IL, USA) and the R environment for statistical computing version 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria) were used for data analysis and graph generation

Next, the impact of cholesterol administration on transendothelial resistance (TER), by exposing endothelial cells to inflammatory conditions mimicking a septic environment, was investigated in-vitro using an Electric cell-substrate impedance sensing (ECIS^®) assay. First, the effect of cholesterol on baseline endothelial permeability showed a dose-dependent effect of cholesterol on TER (Fig. 1, Panel B(a)). Next, it was tested if treatment of endothelial cells with cholesterol can rescue mediator-induced vascular permeability. Indeed, cholesterol concentrate was able to abolish both long-acting permeability by TNFa (Fig. 1, Panel B(b)) and rapid induction of endothelial permeability by thrombin (Fig. 1, Panel B(c)). In previous in vitro experiments, we could show that exposing endothelial cells to plasma from septic shock patients leads to a drop in TER a surrogate of permeability. If the experiments were repeated after TPE this phenotype could be reversed, indicating that plasma exchange could have barrier protective effects in septic shock [4]. The current data from this study add to the current knowledge a potentially fundamental role of cholesterol supplementation in barrier protection using TPE. These findings are also consistent with the recent data, involving rat models, where cholesterol appears to stabilize endothelial integrity via modulation of lipid raft domains, crucial for maintaining cellular membrane stability and signaling pathways involved in vascular permeability [5].

This study has important limitations. Firstly, its relatively small sample size and bicentric setting affect the power and the generalizability of the results. A fixed dose was used for the intervention, which precludes us from providing data on effects at different dosages or time frames. Since TPE was performed within 6 h of randomization, TPE could have been performed very early or late in this timeframe potentially influencing biomarker levels. Repeated measurements at preselected timepoints including the days following treatment should be implemented in follow-up trials to capture the variability due to biomarker assessment time.

In summary, the current study demonstrates that adjunctive TPE therapy in septic shock is associated with a significant increase of circulating cholesterol, which may contribute to reduced capillary leakage and ultimately, the early hemodynamic improvement observed in septic shock patients following TPE.

Abbreviations

AKI

Acute kidney injury

BMI

Body mass index

CRP

C-reactive protein

FFP

Fresh frozen plasma

HDL

High-Density lipoprotein

IQR

Interquartile range

LDL

Low-Density lipoprotein

MAP

Mean arterial pressure

NE

Norepinephrine

PCT

Procalcitonin

RCT

Randomized controlled trial

RRT

Renal replacement therapy

SOC

Standard of care

SOFA

Sequential Organ Failure Assessment

TC

Total cholesterol

TNFa

Tumor necrosis factor alpha

TPE

Therapeutic plasma exchange

WBC

White blood cell

Author contributions

TP and KS collected clinical data from the PDMS. TP, KS and SD generated the figures for publication. DMH, RL and KB performed the laboratory experiments. TP, DAH, HS, CP, BS, CB, KS and SD recruited patients. TP, DAH, DMH, CB, KS and SD interpreted data. TP, DAH, CB, KS and SD wrote the manuscript. KS, CB and SD had the original idea for both trials and wrote the proposals. All authors read an approved the final manuscript.

Funding

No funding was received.

Data availability

The datasets used and analyzed are during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The ethical committee of Hannover Medical School (No. 2786 − 2015 and No. 8852_MPG_23b_2020) and University Medicine Bonn (No. 024/20) approved protocols of both studies, and written informed consent was obtained from participants or authorized representatives. The study was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments.

Consent of publication

Not applicable.

Competing interests

The authors declare no competing interests.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13054-025-05790-0.