Synthetic colloids as priming fluids in cardiopulmonary bypass and postoperative acute kidney injury: a propensity score-matched study

Abstract

Background

The optimal fluid management strategy for patients undergoing on-pump cardiac surgery was controversial regarding the fluid types. This study aimed to compare the impact of hydroxyethyl starch (HES) and succinylated gelatin on postoperative acute kidney injury (AKI).

Methods

This is a single-center, retrospective study. Adult patients undergoing cardiac surgery with cardiopulmonary bypass (CPB) between 2018 and 2022 were included. Patients were grouped by the priming solution used: HES (the HES group) or succinylated gelatin (the gelatin group). The primary outcome was the incidence of postoperative AKI, and secondary outcomes included continuous renal replacement therapy (CRRT), blood transfusions, myocardial infarction, lung infection, cerebrovascular events, chest drainage, duration of mechanical ventilation, intensive care unit (ICU) stay, in-hospital mortality, and associated costs. Propensity score matching (PSM) was conducted to adjust for characteristic differences between the two groups.

Results

A total of 14,443 patients were included. After 1:1 propensity score matching, 1880 pairs were analyzed. Baseline characteristics were comparable between the matched groups. The incidence of AKI was significantly lower in the gelatin group (3.0% vs. 5.1%, P = 0.001). No significant differences were found in transfusion requirements after score-matching. The use of HES was independently associated with increased odds of AKI (adjusted odds ratio [aOR], 1.785; 95% CI, 1.262–2.526; P = 0.001). The effects of HES use on the occurrence of postoperative AKI were greater in elderly and female patients.

Conclusion

The use of HES as a priming solution in on-pump cardiac surgery was associated with an increased risk of postoperative AKI compared to gelatin, although it did not affect transfusion requirements. These findings highlight the importance of considering the priming solution type in fluid management during cardiac surgery.

Introduction

In cardiac surgery, fluid management is a subject of ongoing debate, particularly regarding the types and volumes of fluids used (Ostermann et al. 2024). Effective management is essential to minimize hemodilution caused by fluid infusion and cardiopulmonary bypass (CPB) and to prevent hypovolemia, tissue edema, and cardiovascular complications associated with large blood product transfusions (Casselman et al. 2024). The solutions typically include crystalloids, natural colloids (such as albumin), or synthetic colloids like hydroxyethyl starch (HES) and gelatin. Among these options, albumin has been widely used in recent years. However, research on its benefits in cardiac surgery patients has yielded inconsistent results concerning its impact on renal function or fluid balance (Pesonen et al. 2022; Zhang et al. 2024; Wigmore et al. 2024). Additionally, its high costs constrain its broader clinical adoption, prompting the consideration of synthetic colloids as potential alternatives in specific contexts.

The effects of synthetic colloids on transfusion requirements and kidney function are conflicting. Recent meta-analyses scrutinizing the impact of different priming solutions on postoperative outcomes have found no significant disparities in postoperative mortality, blood loss, or renal function across various priming fluids (Xian-Yu et al. 2023; Ma et al. 2024). However, increasing evidence indicated that HES may be associated with adverse renal outcomes and increased blood loss (Matsunaga et al. 2019; Van der Linden et al. 2005). Consequently, current cardiac surgery guidelines recommend against the use of HES for reducing bleeding and transfusion requirements (Casselman 2024). Despite these recommendations, preferences for synthetic colloids vary across regions, with some areas continuing to use HES while others favor gelatin as an alternative (Miles et al. 2017).

Given the potential differences in prognosis associated with various synthetic colloids and the inconsistent standards of their use, the aim of this study was to compare the effects of HES or gelatin as the priming solution of CPB on the incidence of postoperative acute kidney injury (AKI) in patients undergoing on-pump cardiac surgery.

Materials and methods

Study population

We identified 30,976 patients who underwent cardiac surgery with CPB between January 2018 and June 2022 in our center. The inclusion criteria were as follows: age ≥ 18 years old; undergoing on-pump surgery including coronary artery bypass grafting (CABG), surgical correction of valve disease (mitral, aortic, tricuspid, pulmonary), combined surgeries, and ascending aorta disease treatments; use of HES or gelatin solutions during CPB. Patients who received a combination of the two priming solutions during cardiac surgery, used albumin for priming, had preoperative chronic kidney dysfunction (CKD) or required continuous renal replacement therapy (CRRT), underwent emergency surgery, used minimally invasive extracorporeal circulation systems (MiECC) or modified low-priming CPB system, or had missing data were excluded. A detailed flowchart illustrating the participant selection process is shown in Fig. 1.

Fig. 1

The flow chart of different strategies of fluid management in patients undergoing on-pump cardiac surgery. Abbreviations: CKD, chronic kidney injury, CPB, cardiopulmonary bypass, CRRT, continuous renal replacement therapy, MiECC, minimal invasive extracorporeal circulation, HES, hydroxyethyl starch, PSM, propensity score-matching. *Modified low-priming system is a novel low-priming modified CPB system which has been used in our center since 2010. The low-priming CPB system consists of a short-tubing circuit, an oxygenator with an integrated arterial filter, a vacuum-assisted venous drainage device, and microcardioplegia. †Other types of surgery included heart transplantation, left ventricular assist device implantation, and thoracoscopic surgery

Study design and data collection

This is a retrospective cohort study. We categorized patients into the following two groups according to the priming solution used during cardiac surgery: the HES group or the Gelation group. The Ethics Committee of our institution approved this study (Approval Number: 2024–2379), and the requirement for informed consent was waived due to the retrospective nature of the study.

Patient data were obtained from the cardiac surgery patient database of our center and medical record system using anonymized records. The data included patient demographics (age, sex, body mass index [BMI], and body surface area [BSA]); EuroSCORE (European System for Cardiac Operative Risk Evaluation); personal medical history; echocardiography data; laboratory tests; vital signs; perioperative medication; intraoperative information; and postoperative data.

Surgical management

In the operating room, all patients were subjected to general anesthesia and mechanical ventilation, accompanied by received standard monitoring. During CPB, the bypass flow was maintained between 60 and 70 ml kg⁻¹ min⁻¹, mean arterial pressure (MAP) was kept within 50–70 mmHg, and the core temperature was regulated between 32 and 34 ℃. The CPB circuit consisted of a roller pump, an open venous reservoir, a polypropylene membrane oxygenator, and uncoated polyvinyl chloride tubing, with a static priming volume of 1200–1600 ml. The priming strategies included two options: a mixture of sodium acetate Ringer’s solution with succinylated gelatin or a mixture of sodium acetate Ringer’s solution with 6% HES 130/0.4 (mean molecular weight, 130 kDa; degree of substitution), both in a 1:2 ratio. The decisions regarding the types of priming solutions were made by the perfusionists, with no standardized protocols established at our center.

Study endpoints

The primary endpoint of the study was the prevalence of postoperative AKI, defined according to the Kidney Disease Improving Global Outcomes (KIDGO) Guideline (Ronco et al. 2019): creatinine ≥ 2.0 times the baseline, or increased to more than 4.0 mg/dL, or requirement for CRRT.

Secondary endpoints included the incidence of major complications during the postoperative period, including receiving CRRT, rates of blood products transfusion, myocardial infarction, lung infection, cerebrovascular events, chest drainage, duration of mechanical ventilation, intensive care unit (ICU) stay and length of stay (LOS), in-hospital mortality, ICU costs, and in-hospital costs.

Statistical analysis

Categorical variables were presented as numbers (percentages) and analyzed for significance using the chi-square test or Fisher’s exact tests. Continuous variables were expressed as mean and standard deviation (SD) and tested for normality using the Shapiro–Wilk test. Skewed variables were shown as median (25–75% interquartile range) and assessed using the Mann–Whitney U test. To compensate for baseline differences and reduce selection bias between the two groups, a propensity score-matching (PSM) was conducted using the nearest neighbor algorithm with 1:1 matching without replacement and a caliper distance of < 0.2 of the logit propensity score. The baseline disequilibrium and the adequacy of covariates balance between groups before and after matching were assessed using P-values, with a P-value > 0.05 considered as an unbalanced distribution. Furthermore, the variables were chosen for their clinical relevance and distinctiveness as determined by clinical expertise (Table 1). In the matched cohort, multivariable logistic regression was applied to assess the relationship between two types of solution and postoperative AKI. The independent variables included age, BSA, types of priming fluid, EuroSCORE, history of cerebrovascular disease, history of cardiac surgery, preoperative diuretic use, preoperative blood urea nitrogen, preoperative hematocrit, types of surgery, intraoperative fluid balance, types of cardioplegia. Additionally, sensitivity analyses were performed excluding patients with a total priming volume greater than 1000 ml to further assess the association between volumes of solution and postoperative AKI, using a logistic model approach. Subgroup analyses were conducted to identify other factors affecting patient outcomes. The association between risk factors and outcomes was evaluated using adjusted odds ratio (aOR) and 95% confidence intervals (95% CI). All analyses were performed using R software version 4.4.2 (R Foundation for Statistical Computing, Vienna, Austria).

Table 1 Variables used for propensity score matching

Results

Patient characteristics

During the study period, a total of 14,443 patients met the inclusion and exclusion criteria, and there were 1921 (13.3%) patients who received succinylated gelatin as priming solution (the gelatin group) and 12,522 (86.7%) patients who received HES (the HES group). As a result, 1880 pairs were created after PSM for the final analysis (Fig. 1). The baseline characteristics and operative data of the patients before and after PSM are summarized in Table 2. In the unmatched population, patients in the gelatin group had a higher level of EuroSCORE (EuroSCORE ≥ 6: 3.6% vs. 4.6%, P < 0.001), a higher percentage of ejection fraction (EF) (62 ± 7% vs. 61 ± 7%, P = 0.002), and higher levels of preoperative platelet counts [197 (173, 230) × 10⁹/L vs. 197 (168, 227) × 10⁹/L, P = 0.027] compared to those in the HES group. Patients in the gelatin group had a significantly longer duration of CPB time and aortic cross-clamp (ACC) time compared to those in the HES group [114 (92, 144) min vs. 109 (87, 139) min, P < 0.001; 83 (65, 108) min vs. 79 (61, 104) min, P < 0.001, respectively]. The median fluid balance was significantly higher in the gelatin group compared to the HES group [570 (10, 1200) mL vs. 500 (− 50, 1100) mL, P < 0.001]. Blood cardioplegia was more frequently used in the HES group than in the gelatin group (72.7% vs. 65.5%, P < 0.001). During the study period, there was a downward trend in the use of HES and an upward trend in the use of gelatin. After PSM was adjusted, all clinical covariates were well balanced.

Table 2 Demographic and clinical variables before and after propensity score matching

Patient outcomes

The primary outcome of postoperative AKI occurred in 57 (3.0%) patients in the gelatin group compared with 567 (4.5%) in the HES group (P = 0.002) (Table 3). The gelatin group had a significantly higher incidence of perioperative red blood cell (RBC) and fresh frozen plasma (FFP) transfusions and higher postoperative chest drainage volume on the first postoperative day, ICU costs, and hospital costs (all P < 0.005). After adjustment using the PSM approach, the incidence of AKI remained significantly different between the HES group and the gelatin group (3.0% vs. 5.1%, P = 0.001). The median chest drainage volume on the first postoperative day was higher in the gelatin group compared to the HES group (360 (270, 500) mL vs. 350 (250, 480) mL, P = 0.008). The use of HES as the priming solution was associated with significantly increased odds of AKI (adjusted odds ratio [aOR], 1785; 95% CI [confidential interval], 1.262–2.526; P = 0.001) (Table 4). The sensitivity analyses, excluding patients with a total priming volume greater than 1000 mL, were consistent with the primary analyses, showing that the HES as a priming solution was associated with the increased odds of AKI (aOR, 2.535; 95% CI, 1.594–4.031; P < 0.001).

Table 3 Patient outcomes before and after propensity score matching

Table 4 Multivariate logistic regression analysis of predictors of the occurrence of AKI

Subgroup analyses

The subgroup analyses of the primary outcome identified a quantitative interaction between fluid type and age, sex, and EuroSCORE, and there was no significant interaction with type of surgery (Fig. 2). The use of HES as a priming solution was associated with increased risks of AKI in patients aged ≥ 60 years. The use of HES as a priming solution was associated with increased odds of AKI among female patients, and this effect was less notable among male patients. Patients with a EuroSCORE of 3 to 5 had an increased risk of developing AKI when HES was used as the priming solution.

Fig. 2

Subgroup analysis of postoperative acute kidney injuryAbbreviations, CABG, coronary artery bypass grafting, CI, confidence interval, HES, hydroxyethyl starch

Discussion

In this retrospective cohort study involving 14,443 patients undergoing on-pump cardiac surgery, the use of HES as a priming solution was associated with an increased risk of postoperative AKI but did not increase the risk of transfusion. These findings suggest that the type of solution, rather than the total volume of fluid administered, may have a more direct role in the development of AKI. Subgroup analysis indicated that the association between HES exposure and AKI was consistent among female patients, further supporting the potential impact of solution type on renal outcomes.

Current guidelines and consensus recommend against the routine use of synthetic colloids for both priming CPB and intraoperative fluid administration due to concerns about increased mortality and impaired postoperative kidney function (Ostermann 2024; Casselman 2024). In this study, we found that the use of HES for priming was associated with a higher incidence of AKI compared to gelatin. The physicochemical properties of HES, including the degree of molar substitution and hydroxymethylation, may contribute to its slower degradation within the body (Westphal et al. 2009), resulting in the accumulation of HES in various tissues and organs, including the kidneys. Variations in the molecular structure, metabolism, and tissue distribution of colloids may significantly impact renal outcomes (Vuiblet et al. 2015; Glassford et al. 2018). Specifically, the accumulation of HES in renal tubular cells and extrarenal tissues is thought to impair renal function, potentially leading to kidney damage (Pisano et al. 2016).

The effects of synthetic colloids on the development of postoperative AKI after cardiac surgery are conflicting (Matsunaga 2019; Heringlake et al. 2020; Vives et al. 2016). Some studies suggest that gelatin may be more strongly associated with an increased risk of AKI compared to HES (Heringlake 2020). A large cohort study involving 6478 patients analyzed the effects of changing the primary fluid replacement solution over three consecutive 3-year periods (Bayer et al. 2013), showing that treatment with either HES or gelatin was significantly associated with an increased risk of renal replacement therapy, even after adjusting for potential confounders through propensity score matching. In cardiac surgery, due to the complexity of the procedure and the kidney’s sensitivity to fluid load, CPB may exacerbate several physiological processes detrimental to renal function, such as inadequate perfusion, inflammation, and oxidative stress (Yang et al. 2024). Additionally, factors such as fluid volume and the type of colloid used (e.g., HES or gelatin) may influence renal blood flow and function in distinct ways. In this study, while PSM effectively adjusted for known confounders, unmeasured variables such as perioperative management strategies and individual patient characteristics may still contribute to the observed differences in AKI risk. At our center, while the use of HES has been gradually reduced, we remain cautious with the use of gelatin due to its potential impact on renal function and acknowledge the need for further investigation to clarify its role in kidney injury.

In our preliminary analysis, to minimize the potential impact of different preloading volumes, we excluded patients who underwent surgery with the MiECC or modified low-priming system, which both use lower priming volumes. However, the potential influence of large volumes of colloids could not be entirely disregarded. To address this, we performed a sensitivity analysis excluding patients who received more than 1000 mL of colloid, and the results were consistent with our primary analysis (Table 4). In the present study, the volume of colloids administered did not appear to influence the development of AKI, as confirmed by these sensitivity analyses. This finding aligns with previous studies, which also observed no significant dose–effect relationship between colloid volume and AKI severity (Matsunaga 2019; Lagny et al. 2016). Our analysis suggests that the type of colloid, rather than the volume administered, may be more closely associated with renal injury.

The subgroup analyses of the primary outcome revealed that the use of HES as a priming strategy in on-pump cardiac surgery was associated with an increased risk of AKI among elderly and female patients. In elderly patients, the increased risk of AKI may be attributed to age-related declines in renal function, such as reduced renal blood flow and glomerular filtration rate, making them more vulnerable to hemodynamic changes (O'Sullivan et al. 2017). Furthermore, elderly individuals often have comorbidities like diabetes and hypertension, which further elevate the risk of AKI. Females have been suggested as an established risk factor for the development of cardiac surgery-associated AKI (Neugarten et al. 2016). In female patients, the increased incidence of AKI could be related to their smaller BSA and renal physiological characteristics. Women are more sensitive to fluid load and hemodynamic changes due to their smaller BSA, which may result in a relatively greater dilution effect and volume overload when the same volume of priming fluid is administered, potentially exacerbating renal injury. Additionally, gender differences may be associated with variations in gene expression profiles, hormonal influences, and differences in renal blood flow and filtration function (Curtis 2024), which could make women more susceptible to renal injury when HES is used. However, the literature on gender differences in the use of different colloid types is limited, and further research is needed to clarify the specific impact of gender on the relationship between colloid types and renal injury.

Previous randomized controlled trials comparing gelatin and HES in cardiac surgery found no significant differences in blood loss and transfusion requirements (Kimenai et al. 2013; Skhirtladze et al. 2014). Our results are consistent with a meta-analysis in cardiac surgery, which also found no evidence of increased bleeding complications in the subgroup analysis of different colloid types (Chen et al. 2024). In this study, no significant difference in transfusion needs was observed after PSM between the HES and gelatin groups, despite initial disparities prior to matching. These findings suggest that the choice of colloid may not be a critical factor in minimizing transfusion requirements. Instead, patient baseline characteristics, surgical complexity, and perioperative management strategies may have a greater impact on transfusion needs.

This study has several limitations. First, its retrospective nature limits the ability to establish causality. Second, while PSM helped control for confounding, unmeasured variables such as biomarkers for early AKI detection were not included. Third, this study was limited to synthetic colloids, primarily HES and gelatin, and thus, the results may not be generalizable to other types of colloids or crystalloid solutions. Finally, this was a single-center study, and the choice of colloids may reflect institutional preferences that could limit broader applicability. Future research may involve larger, multicenter cohort studies aimed at exploring optimal priming strategies during cardiac surgery, with the goal of enhancing early detection and improving patient outcomes.

Conclusions

In patients undergoing on-pump cardiac surgery, the use of HES as a priming solution, compared to gelatin use, was associated with an increased risk of postoperative AKI. However, HES did not result in a higher requirement for blood transfusions. These findings highlight the importance of carefully selecting the type of solution used in cardiac surgery, as the choice of colloid may have a more significant impact on renal outcomes than the volume of fluid administered.