Effects of Ulinastatin on Postoperative Renal Function in Patients Undergoing Cardiac Surgery with Cardiopulmonary Bypass: A Prospective Cohort Study with 10-Year Follow-Up

Cardiac surgery-associated acute kidney injury (CSA-AKI) is a frequent and severe complication in adult patients undergoing cardiac surgery with cardiopulmonary bypass (CPB). It is the second most common cause of AKI in the intensive care unit (ICU) [1]. Notably, AKI is associated with the increased need for renal replacement therapy (RRT), length of hospital stay, hospitalization cost, and mortality [1]. The KDIGO criteria define AKI severity according to serum creatinine (SCr) and urinary output. However, the SCr increases relatively slower than the onset of kidney injury, leading to the delayed detection of AKI. Neutrophil gelatinase-associated lipocalin (NGAL), an AKI biomarker, is promising for identifying acute tubular damage, especially in the subclinical AKI stage. It can represent acute tubular damage before the presence of clinical renal dysfunction [2].

Prior investigations indicated that CSA-AKI was associated with systemic inflammatory response syndrome (SIRS), renal ischemia-reperfusion injury, renal hypoperfusion, blood transfusion, and nephrotoxic drugs [3]. Moreover, kidney dysfunction was linked with CPB-induced inflammatory responses [3, 4]. However, efficacious drugs that could reduce AKI incidence and the requirement for RRT remain unknown [5]. Ulinastatin is a glycoprotein purified from healthy human urine as a broad-spectrum protease inhibitor. It was demonstrated that CPB could reduce complications in patients undergoing cardiac surgery through its anti-inflammatory and antioxidant activities [6‒8]. In addition, it has been widely used in clinical practice for sepsis and septic shock patients. With the increasing research on the ulinastatin effect on inflammation, oxidative stress, apoptosis, and renal fibrosis, tremendous clinical practices have fully affirmed its efficacy in treating acute or chronic pancreatitis, severe infection, and acute respiratory distress syndrome [9, 10]. However, whether ulinastatin prevents the kidney from CSA-AKI and can reduce morbidity and mortality associated with CSA-AKI remains controversial. The present study aimed to determine the potential effects of ulinastatin on renal function and long-term survival rate in patients receiving cardiac surgery with CPB.

This single-center prospective cohort study was approved by the Ethics Committee of Fuwai Hospital, and written informed consent was not required. The period of enrollment was from August 2008 to January 2011.

The inclusion criteria were (1) patients aged 18–70 years, (2) patients with normal renal function with GFR ≥60 mL/min/1.73 m², and (3) patients who underwent cardiac surgery with CPB. Patients with preoperative or perioperative renal dysfunction who underwent kidney transplantation within the past 365 days were excluded. The patients with acute inflammatory reactions were excluded. The ulinastatin group patients received intravenous ulinastatin (1 × 10⁶ U) after induction of anesthesia, and the patients without ulinastatin served as the control group. According to the routine practice in our institution, blood samples were assessed before anesthesia induction (T₀), on arrival in the ICU (T₁), and at 6 (T₂), 12 (T₃), and 24 h (T₄) after surgery.

The primary outcome of this study was the rate of new-onset AKI. According to the KDIGO criteria, AKI was identified when urine output was <0.5 mL/kg/h × 6 h, scores increased by >1.5-fold over 7 days compared with baseline, or scores increased by ≥ 0.3 mg/dL over 48 h [11]. The secondary outcome was the need for RRT and the changes in AKI biomarkers, including plasma NGAL (pNGAL) and interleukin-6 (IL-6), during the postoperative course. The CSA-NGAL Score is a tubular damage score based on the level of NGAL and a clinical index for early detection of subclinical AKI [2]. According to the CSA-NGAL Score, when a patient reaches absolute cutoff values of 150 ng/mL for urinary NGAL (uNGAL) and 200 ng/mL for pNGAL, or uNGAL is > 100 ng/mL with a second value >125 ng/mL and pNGAL is > 100 ng/mL with a second value >150 ng/mL, acute tubular damage related to cardiac surgery can be reasonably identified [2]. SCr, pNGAL, and serum IL-6 concentrations were detected by ELISA for each sample, in which pNGAL and IL-6 were detected by emulsion enhanced immune turbidimetry assay and double antibody sandwich ELISA separately. In addition, in-hospital mortality and morbidity (stroke, postoperative myocardial infarction, respiratory failure) and adverse outcomes (sudden cardiac arrest, readmission to ICU, reoperation, intra-aortic balloon pump, and extracorporeal membrane oxygenation) were observed. The present study defines respiratory failure as prolonged ventilator support for more than 48 h.

The follow-up period was until January 2021 by telephone call, outpatient service, a face-to-face visit, recent chat tools (QQ), and WeChat. For those without a telephone, follow-up calls were provided by family members or neighbors who were willing to answer on their behalf. When a participant reported renal dysfunction, a detailed record was documented, including the renal functional status and the illness onset date as diagnosed by local hospitals. The data ascertained by the follow-up team were as follows: onset of death, time for loss follow-up, renal dysfunction, presence or absence of heart failure, hematosepsis, or readmission for renal dysfunction, RRT, intermittent hemodialysis, or potentially nephrotoxic drugs, levels of blood glucose, and results for renal function. All of these items were recorded. Information on the date and the primary and secondary causes of death were obtained for death cases. Loss to follow-up occurred when the telephone number offered by participants did not work or family members could not be contacted.

Continuous variables were presented as mean ± standard deviation or median (Q1, Q3), the Shapiro-Wilk test was used to estimate the normal distribution assumption, and the Levene test was applied to assess the equal variance assumption. Differences in these characteristics between the two groups were evaluated by an independent two-sample t-test or Mann-Whitney Wilcoxon test. Categorical variables were summarized by frequency and percentage and compared with the χ² or Fisher’s exact test. Inter-group differences in serum IL-6, pNGAL, and SCr were assessed by a two-way repeated-measures analysis of variance (ANOVA) with a Bonferroni post hoc test. Survival analysis was performed by the Kaplan-Meier method and a log-rank test. Moreover, those variables with a p value <0.1 were considered to enter multivariable Cox regression models to find the potential risk factors for long-term survival. All the analyses were performed using SPSS software (Version 18.0, SPSS Inc.). p value <0.05 was considered statistically significant.

This study enrolled 413 patients, of whom 135 patients in the ulinastatin group and 278 without ulinastatin served as the control group (Fig. 1). In addition, the two groups’ demographic characteristics were not statistically significant, except for repairing the atrial septal defect. In addition, the detailed demographics are presented in Table 1.

The incidence of AKI in the ulinastatin group was significantly lower than in the control group (20.00 vs. 32.37%, p = 0.009). There was no statistical significance in RRT between the ulinastatin group and the control group (0/135 vs. 6/278, p = 0.09). Furthermore, a significant difference was observed between the two groups of the subclinical AKI calculated by CSA-NGAL Scores (p = 0.002) (Table 2).

The interaction effects of group and time in SCr, pNGAL, and IL-6 were evaluated by two-way repeated-measures ANOVA with the Bonferroni post hoc test. For SCr, there was no significant difference between the two groups (p = 0.23; Fig. 2a) and no interaction between time and group (p = 0.235). However, a considerable time effect (p < 0.001) was observed in Scr. For pNGAL, statistical significance was observed in groups (p = 0.007, Fig. 2b), time effect (p < 0.001), and interaction between time and group (p = 0.006). Moreover, a statistical significance of IL-6 was observed between the two groups (p = 0.001, Fig. 2c), time effect (p < 0.001), and interaction between time and group (p < 0.001).

The corresponding levels of SCr, pNGAL, and IL-6 in the ulinastatin and control groups at different sampling times were as follows: (1) SCr, T₁: 1.07 versus 1.08 mg/dL (p = 0.94), T₂: 1.46 versus 1.5 mg/dL (p = 0.17), T₃: 1.46 versus 1.57 mg/dL (p = 0.14), T₄: 1.28 versus 1.32 mg/dL (p = 0.39). (2) pNGAL, T₁: 89.78 versus 129.41 ng/mL (p = 0.005), T₂: 135.2 versus 247.92 ng/mL (p = 0.001), T₃: 131.06 versus 212.16 ng/ml (p = 0.003), T₄: 110.32 versus 159.78 ng/mL (p = 0.01). (3) IL-6, T₁: 0.37 versus 0.40 pg/mL (p = 0.30), T₂: 0.78 versus 0.93 ng/mL (p = 0.012), T₃: 0.83 versus 1.05 g/mL (p < 0.001), T₄: 0.58 versus 0.89 ng/mL (p < 0.001, T₄).

There was no significant difference in the in-hospital mortality rate (0.74 vs. 0.36%; p = 0.60) between the two groups. In addition, the length of ICU stays, mechanical ventilation, and hospital stays were comparable between the two groups (Table 1). Moreover, no statistical significance was observed in postoperative complications such as stroke, postoperative myocardial infarction, cardiac arrest, ICU readmission, surgery-associated reoperation, RRT, IABP, and ECMO between the two groups (Table 3). Notably, the incidence of respiratory failure was significantly lower in the ulinastatin group than in the control group (0.74 vs. 5.40%, p = 0.02; Table 3).

The long-term survival analysis was assessed in the 411 patients who were discharged from the hospital. The 10-year follow-up (median: 9.37, 95% confidence interval [CI]: 9.17–9.57) survival rates were 87.80% in the ulinastatin group and 86.60% in the control group, with no significant difference tested by Kaplan-Meier curve (p = 0.76; Fig. 3). There was no significant difference between the ulinastatin and the control group in the rates of loss to follow-up (14.07% [19/135] vs. 14.03% [39/278]; p = 0.99). Statistical significance was observed by multivariable Cox regression analysis in operation time (hazard ratio [HR]: 1.01; 95% CI: 1.00–1.02; p = 0.044), NGAL (HR: 7.68; 95% CI: 4.43–13.3; p < 0.001), and preoperative SCr (HR: 2.97; 95% CI: 1.1–8.01; p = 0.031), which were associated with the long-term survival.

In the present study, we found that ulinastatin significantly reduced the incidence of CSA-AKI, which was consistent with the previous studies [12]. In addition, the two groups had no significant difference in RRT. In-hospital mortality, morbidity, and adverse outcomes were comparable between the two groups. However, a significantly lower incidence of respiratory failure was observed in the ulinastatin group compared with the control group. Moreover, the two groups’ long-term survival rate was not statistically significant.

According to the CSA-NGAL Score, the subclinical AKI was significantly lower in the ulinastatin group than in the control group. Evidence suggests that the KDIGO criteria are superior to other criteria, but they are less sensitive to detecting early-stage tubular damage. The KDIGO criteria are based on the SCr level to diagnose AKI. However, changes in SCr often occur 48 h to 7 days after the original damage, and SCr levels are influenced by many related factors [13]. Thus, the interventions to prevent patients from developing perioperative AKI may be too late [14]. These limitations drive us to search for new biomarkers that detect AKI in an earlier phase. For example, NGAL, a novel discovered biomarker, plays a crucial role in detecting AKI, and it was increased at an earlier stage than the elevation of SCr [2, 15, 16]. NGAL in plasma and urine is increased when acute tubular damage occurs, but pNGAL is detectable earlier than uNGAL. In a previous study, pNGAL was used to predict AKI based on the CSA-NGAL Score, and the results based on the CSA-NGAL Score were similar to those on the KDIGO Score [2]. Furthermore, in this study, according to the KDIGO criteria and the CSA-NGAL Score criteria, statistical significance between ulinastatin and the control group was detected by both criteria. Moreover, it is indispensable to prevent AKI from cardiac surgery damage. Early detection by NGAL may improve patient’s prognosis.

We also found that the serum IL-6 level was significantly lower in the ulinastatin group compared with the control group. The effect of ulinastatin on reducing CSA-AKI is considered to involve the inhibition of inflammation. Nakanishi et al. [17] demonstrated that ulinastatin administered before CPB could decrease the elevation of IL-6 and IL-8 postoperatively, and Wang et al. [18] reported that ulinastatin could attenuate inflammation and CPB-induced kidney injury in infant piglets dose-dependently. Moreover, a propensity score-matched study performed by Wan et al. [12] showed that the administration of ulinastatin during CPB was associated with a lower incidence of CSA-AKI. Meanwhile, other studies indicated that ulinastatin could inhibit polymorphonuclear neutrophils, which helps reduce the risk of bleeding and the requirement for blood transfusion after surgery [19, 20]. However, Song et al. [21] found that ulinastatin did not affect the attenuation of IL-6, tumor necrosis factor-α (TNF-α), and any improvements in renal function. For example, it was reported that SIRS arising after cardiac surgery played a pivotal role in the development of AKI [22, 23]. The pathogenesis of SIRS was associated with multiple factors, including surgical trauma, transfusion, hypothermia, blood loss, ischemia-reperfusion injury, immune system activation, and endotoxemia [24].

The mechanism of ulinastatin in different clinical practices has been widely explored [25‒32]. A recent study reported that ulinastatin could attenuate the ischemia-reperfusion kidney injury by its capability of regulating the microenvironment, such as decreased acute inflammatory response, oxidative stress damage, and apoptosis. Above all, ulinastatin attenuates the incidence of CSA-AKI, which may be attributed to its anti-inflammatory properties, which inhibit the activation of various inflammatory pathways [19, 33, 34].

We concluded the potential protective effect of ulinastatin on CSA-AKI. Nevertheless, the survival rate after 10 years of follow-up did not differ significantly between the two groups. It was consistent with a previous meta-analysis, which demonstrated that ulinastatin could reduce inflammatory cytokines but did not affect hospital mortality [35]. Generally, patients with AKI have a poor prognosis, which can lead to chronic kidney disease. However, limited studies have evaluated the effect of ulinastatin on long-term survival [36]. According to previous studies, the impact of ulinastatin may be diminished depending on the duration of administration [37]. For example, ulinastatin was administered through a bolus injection rather than a continuous infusion [6]. Moreover, the half-life of ulinastatin is about 34 min in healthy adults [38], while the duration of CPB is often longer than 40 min [39]. However, more investigations need to be conducted.

It was worth noting that we found that ulinastatin could reduce the incidence of respiratory failure, which was consistent with the previous studies [40]. Xu et al. [41] suggested that ulinastatin could inhibit the release of proinflammatory cytokines and polymorphonuclear leukocyte elastase (PMNE) to reduce pulmonary injury and improve pulmonary function after CPB, thereby shortening the intubation time and length of ICU stay. In addition, the lung-protective effect of ulinastatin was dose-dependent, and a lower than effective dose could lead to negative results. Furthermore, a recent study reported that ulinastatin could decrease lung damage in diabetic sepsis rats [26]. However, large randomized controlled trials are needed to investigate the effect of ulinastatin on protecting the respiratory system. Moreover, the relationship between respiratory failure and AKI needs more investigation.

There are some limitations to the present study. First, the study was designed as a single-center prospective cohort with less power than RCTs. The rationale evidence remains sparse concerning the effect of ulinastatin on renal outcomes, and this was a pilot study. Therefore, it was in accordance with ethical considerations to conduct an observational rather than an interventional study. However, some confounding biases may have arisen. Second, there were relatively high follow-up loss rates, comprising 14.20% (19/134) in the ulinastatin group and 14.10% (39/277) in the control group. Achieving follow-up for as long as 10 years was a great challenge. Nevertheless, the rate of loss to follow-up was comparable between the two groups. Third, we only evaluated SCr for the AKI criteria in the present study. The urine output in the KDIGO criteria requires a state of oliguria or anuria lasting more than 6, 12, or 24 h, which was infeasible in postoperative ICU settings. For ethical considerations, oliguria and anuria would not be observed for such a long period and should be promptly treated according to the institutional protocol. Furthermore, the assessment of AKI by SCr alone has been widely accepted in studies such as the BART study [42] and the ATACAS study [43]. Finally, in terms of the sample size in this study, there were numerous emerging cardiac surgery centers across the country. The patients with the slight condition were cured in local hospitals, so the cases in our center, a National Center for Cardiovascular Disease, were mainly centralized in more severe patients, which may contribute to the relatively small sample size. Furthermore, the indications that ulinastatin could protect renal function were not reported before but deserved to be explored. Thus, it was not appropriate to conduct a pilot study that enrolled a large number of patients. Above all, further investigations are supposed to examine the effect of ulinastatin on renal function and long-term survival rate. In summary, ulinastatin can significantly reduce the incidence of early postoperative AKI during cardiac surgery with CPB without affecting the long-term survival rate.

The authors thank Alison Sherwin, Ph.D., from Liwen Bianji, Edanz Group China (www.liwenbianji.cn/ac), for editing the English text of a draft of this manuscript.

The study was approved by the Ethical Review Board of Fuwai Hospital (ethical approval no. 2008–366). Written informed consent was not required.

All the authors declare that they have no competing interests.

All authors declare that there were no funding sources or relevant funding.

Huanran Lv and Qian Li have given substantial contributions to the conception or the design of the manuscript. Yuda Fei, Peng Zhang, Lihuan Li, and Jia Shi contributed to acquisition, analysis, and interpretation of the data. All authors participated in drafting the manuscript, and Hong Lv revised it critically. All authors contributed equally to the manuscript and read and approved the final version of the manuscript.

Huanran Lv and Qian Li contributed equally to this work.

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