Background

Postreperfusion hyperkalemia (PRHK) is a well-known and potentially life-threatening complication during liver transplantation (LT). PRHK may induce severe postreperfusion syndrome (PRS) [1], and its most serious consequence is cardiac arrest [1,2,3,4,5,6,7], which is associated with higher intraoperative mortality and poorer graft and patient survival after transplantation [4,5,6,7]. Therefore, identifying risk factors for PRHK, especially novel, modifiable predictors, may help clinicians develop targeted preventive strategies.

Several published studies have examined predictive risk factors for PRHK in adult LT. In 2000, Nakasuji and colleagues first demonstrated that the cardiac index, serum lactate levels, and serum potassium (sK+) levels during the anhepatic stage are independently associated with the peak sK+ levels immediately following reperfusion [8]. One large-sample study in 2007 by **; 3) immediately before reperfusion; 4) at one minute following reperfusion; 5) at 5 min following reperfusion; 6) at one hour following reperfusion; 7) at 2 h following reperfusion; and 8) at the end of the surgery. Notably, most blood samples were collected from the arterial lines except the blood sample one minute following reperfusion, which was collected from the pulmonary artery catheters.

Data collection

Our institutional database was used to collect baseline patient characteristics, intraoperative variables, and postoperative outcomes retrospectively.

  • Baseline patient characteristics included age, sex, weight, height, primary diagnosis, Child–Pugh score, and Model for End-Stage Liver Disease (MELD) score.

  • The intraoperative variables collected included graft weight, graft-to-recipient weight ratio (GRWR), cold ischemia time (CIT), warm ischemia time (WIT), inferior vena cava clam** time, duration of the anhepatic phase, prereperfusion amount of RBCs transfused, presence of an ECD liver graft (DCD donors, donor age of > 65 years, a sodium level of > 155 mmol/L, aspartate aminotransferase or alanine aminotransferase levels of > 100 IU/L, macrosteatosis > 30%, CIT > 16 h, and WIT > 90 min), the eK+ concentration, the prereperfusion sK+ (sK+0), sK+1, and sK+ levels at five minutes following reperfusion (sK+5), and incidences of postreperfusion significant arrhythmias, cardiac arrest, and severe PRS.

  • Data on postoperative outcomes, including re-operation and in-hospital death within the first 30 days post-LT, early allograft dysfunction (EAD), acute kidney injury (AKI), duration of mechanical ventilation, intensive care unit (ICU) admission, and length of hospital stay, were also collected.

Study outcomes

The primary outcome was the development of PRHK, defined as the sK+ levels during the postreperfusion period exceeding 5.5 mmol/L. Postreperfusion significant arrhythmias and severe PRS were defined according to the Peking criteria (Table 1) [16]. AKI was diagnosed according to the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines [17]. EAD was defined as meeting one or more of the following criteria: 1) aspartate aminotransferase or alanine aminotransferase levels > 2000 IU/L within the first seven days post-LT, 2) total bilirubin ≥ 10 mg/dL on postoperative day 7, or 3) an international normalized ratio ≥ 1.6 on postoperative day 7 [18].

Table 1 Peking criteria for the definition of severe postreperfusion syndrome in liver transplantation
Full size table

Statistical analysis

The normality of the distribution of all continuous variables was assessed using the Kolmogorov–Smirnov test. Normally distributed variables were expressed as the mean ± standard deviation, and non-normally distributed variables were expressed as the median (interquartile range). Data were compared using the Student’s t-tests (normal distribution) or Mann–Whitney U tests (non-normal distribution). Categorical variables were described as the number and percentage of patients. Data were compared using the χ2 test or Fisher’s exact test. Univariate logistic regression analyses were used to screen factors associated with the occurrence of PRHK. Potentially significant factors, which had a P-value < 0.10 in univariate analyses, were further enrolled in a binary logistic regression model using a forward (conditional) stepwise procedure. Receiver operator characteristic (ROC) curve analysis was used to assess the predictive accuracy and optimal cut-off value of the eK+ concentration. The sensitivity, specificity, and positive and negative predictive values (PPV and NPV) were determined at the optimal cut-off points. In addition, a multiple linear regression model was used to identify independent predictors of the maximum percentage increase in the sK+ levels following reperfusion. All statistical tests were 2-sided, and a P-value < 0.05 was considered statistically significant. Statistical analyses were performed using SPSS for Windows software (version 22.0; IBM SPSS, Inc., Chicago, IL, USA) and MedCalc for Windows software (version 15.2; MedCalc Software, Ostend, Belgium).

Results

Patients

Of 147 adults who underwent deceased LT within the study period, we excluded 55 patients without sK+1 level measurements, five patients diagnosed with hyperkalemia during the anhepatic stage, and one patient with incomplete datasets, leaving 86 patients to be included in the analyses. The clinical characteristics of the study population are listed in Table 2. The average age of the study population was 49.6 ± 10.9 years, and the majority (68.6%) were male. The leading indication for LT was hepatitis B cirrhosis (50.0%). The median (interquartile range) MELD score was 15 (9–19) points. The mean graft weight was 1302.9 ± 270.5 g, and the mean sK+0 level was 4.17 ± 0.55 mmol/L. The median (interquartile range) eK+ concentration was 6.65 (5.38–9.73) mmol/L.

Table 2 Baseline characteristics of 86 liver transplantations (LTs)
Full size table

Postreperfusion hyperkalemia

Fifty-four (62.8%) patients experienced PRHK (Table 2). Among the primary diagnoses, only hepatitis B cirrhosis was significantly different between PRHK and non-PRHK patients (22 of 54 vs. 21 of 32, 40.7% vs. 65.6%, P = 0.026). Patients who developed PRHK were more often transplanted with an ECD graft (13 of 54 vs. 1 of 32, 24.1% vs. 3.1%, P = 0.011). Liver grafts for patients develo** PRHK had a higher graft weight (1362.7 ± 251.4 vs. 1201.9 ± 275.4 g, P = 0.007) and a larger GRWR (1.98 [1.79–2.45] vs. 1.63 [1.43–2.28] %, P = 0.045) compared to patients not develo** PRHK. Patients who developed PRHK had a higher sK+0 level (4.27 ± 0.56 vs. 4.00 ± 0.49 mmol/L, P = 0.030) and a higher eK+ concentration (7.65 [5.68–12.20] vs. 5.90 [4.40–6.88] mmol/L, P = 0.003).

Predictors for postreperfusion hyperkalemia and the maximum percentage increase in the sK+ level following reperfusion

Table 3 shows an association between perioperative patient and graft variables and PRHK in patients who underwent deceased LT. After an analysis of the potentially significant predictors by multivariate logistic regression, the following three variables were independently associated with the presence of PRHK: graft weight (odds ratio [OR] 1.283; 95% confidence interval [CI] 1.029–1.599 per 100 g; P = 0.027), the eK+ concentration (OR 1.291; 95% CI 1.068–1.561 per mol/L; P = 0.008), and the sK+0 level (OR 4.459; 95% CI 1.543–12.884 per mol/L; P = 0.006). Based on the areas under the ROC curves (AUROCs), the eK+ concentration had the best predictive ability for the presence of PRHK (AUROC, 0.694), followed by graft weight (AUROC, 0.645) and the sK+0 level (AUROC, 0.640) (Fig. 2). Table 4 shows the sensitivity, specificity, PPV, NPV, and diagnostic accuracy at the cut-off point that provides the best Youden index for each variable. The best cut-off point for the eK+ concentration was more than 6.9 mmol/L, with a sensitivity of 59.26%, specificity of 78.12%, PPV of 82.05%, and NPV of 53.19%. In addition, multiple linear regression was performed to predict the maximum percentage increase in the sK+ level following reperfusion based on graft weight, GRWR, prereperfusion amount of RBCs transfused, the sK+0 level, the eK+ concentration, and the presence of an ECD graft. A significant regression equation was found (F = 10.832, P < 0.001), with an R2 of 0.672 (Table 5). The eK+ and sK+0 levels were significant predictors of the maximum percentage increase in the sK+ level following reperfusion.

Table 3 Logistic regression analysis of predictors for postreperfusion hyperkalemia in 86 consecutive deceased liver transplantation recipients
Full size table
Fig. 2
figure 2

ROC curve analysis to predict the occurrence of postreperfusion hyperkalemia in 86 consecutive deceased liver transplant recipients. eK+, effluent potassium concentration; sK+0, serum potassium concentration before reperfusion; ROC, receiver operating characteristic

Full size image
Table 4 Prediction of postreperfusion hyperkalemia in 86 consecutive deceased liver transplantation recipients
Full size table
Table 5 Linear regression analysis of predictors for the maximum percentage increase in postreperfusion serum potassium concentration
Full size table

Postreperfusion hyperkalemia and associated outcomes

Patients with PRHK had higher sK+1 (6.36 ± 0.57 vs. 4.77 ± 0.39 mmol/L, P < 0.001) and sK+5 (4.20 ± 0.96 vs. 3.36 ± 0.46 mmol/L, P < 0.001) levels, and therefore suffered more often from postreperfusion significant arrhythmias (36/54 vs. 6/32, 66.7% vs. 18.8%; P < 0.001) and severe PRS (43/54 vs. 18/32, 79.6% vs. 56.3%; P = 0.021). The occurrence of cardiac arrest after reperfusion did not differ between patients with and without PRHK. Furthermore, patients with PRHK suffered more often from postoperative EAD (32/54 vs. 5/32, 59.3% vs. 15.6%; P < 0.001). There were no significant differences for the other outcomes, including ventilation time, ICU and hospital stay lengths, AKI incidence, in-hospital mortality, and re-operation rate (Table 6).

Table 6 Comparison of postreperfusion and postoperative outcomes in patients with and without postreperfusion hyperkalemia
Full size table

Discussion

Elevated sK+ concentrations and PRHK are generally anticipated during deceased LT, especially following reperfusion of ECD liver grafts. However, little or no information is found in the literature concerning the roles of effluent fluid compositions in PRHK development. The most important finding of this study was that an elevated eK+ concentration was another modifiable yet potentially ignored risk factor for PRHK in deceased LT, together with the prereperfusion sK+ level and a large graft weight. These findings may be of significant clinical importance for PRHK prevention in deceased LT.

PRHK is associated with several intraoperative and postoperative complications, including but not limited to PRS, cardiac arrest, intraoperative death, and postoperative mortality [1,2,3,4,5,6,7]. Risk identification for PRHK is challenging, yet its prediction is the target of several investigations. In accordance with the present results, previous studies have demonstrated that the baseline or prereperfusion sK+ level was the most common modifiable factor associated with PRHK during deceased LT [8,27,28,29], and anterograde venting via the portal vein [13, 30], can be performed to reduce the eK+ concentrations before reperfusion; 3) portal vein speed-control reperfusion strategy [31] should be adopted to control the timing, severity, and duration of PRHK; and 4) aggressive preemptive therapies, such as calcium chloride, epinephrine, sodium bicarbonate, and atropine, should be initiated upon reperfusion.

Several limitations are worth acknowledging. First, the retrospective, single-center design limits the study’s generalizability and may generate inevitable selection bias. Second, due to the lack of real-time, continuous monitoring methods, the sK+1 level was generally regarded as the peak sK+ level following reperfusion, which may underestimate the actual incidence of PRHK. Third, although multivariate analyses have demonstrated the association between the eK+ concentrations and PRHK, the exact mechanism of action of eK+ in PRHK remains to be fully elucidated. Fourth, the effectiveness of existing intervention strategies for PRHK remains unclear; therefore, further studies are warranted to address this issue definitively.

Conclusions

A high eK+ concentration before reperfusion was a significant predictor of PRHK and the maximum percentage increase in the sK+ level following reperfusion in deceased LT. Further research is needed to evaluate whether proactive interventions can reduce the risk of PRHK during LT from a high eK+ concentration or ECD liver grafts.