Korean J Anesthesiol Search

CLOSE


Korean J Anesthesiol > Volume 79(4); 2026 > Article
Running title Intraoperative hypotension in LT
Authors Initials A Katayama, E Abuelkasem, LC Su
Katayama, Abuelkasem, Su, and Wang: Intraoperative hypotension during liver transplantation is associated with long-term mortality

Abstract

Background

Intraoperative hypotension (IOH) is common during liver transplantation (LT); however, its impact on long-term mortality remains unclear. We investigated the association between IOH and 1-year mortality using various mean arterial pressure (MAP) thresholds and exposure durations.

Methods

This retrospective observational cohort study included 1063 patients who underwent LT. We defined IOH as an MAP < 65 mmHg and examined each threshold in 5 mmHg increments to < 40 mmHg. Our primary outcome was 1-year mortality after LT. Multivariate logistic regression was used to control for confounding variables while assessing the association between IOH and 1-year mortality.

Results

The 1-year mortality after LT was 7.1%. Only moderate-to-severe IOH showed independent associations with 1-year mortality: MAP < 55 mmHg for ≥ 43 minutes (hazard ratio [HR] = 2.48, 95% CI [1.05–5.86], P = 0.038), MAP < 50 mmHg for ≥ 14 min (HR = 2.60, 95% CI [1.01–6.65], P = 0.047), MAP < 45 mmHg for ≥ 13 min (HR = 4.98, 95% CI [1.18–21.0], P = 0.029), and MAP < 40 mmHg for ≥ 5 min (HR = 6.72, 95% CI [1.56–28.9], P = 0.010), whereas longer exposure durations at MAP < 65 or < 60 mmHg were not independently associated with increased mortality. Furthermore, both a lower MAP and longer exposure duration for each threshold increased the predicted probability of 1-year mortality.

Conclusions

Moderate-to-severe hypotension was independently associated with increased 1-year mortality after LT. These findings underscore the importance of intraoperative blood pressure management in LT recipients.

Introduction

Liver transplantation (LT) is the only therapeutic option for patients with end-stage liver disease (ESLD). Although advancements in surgical techniques and perioperative management over the past decades have dramatically improved patient outcomes, the 1-year mortality rate remains as high as 6.5%–17% [1–3]. Previous studies have identified several risk factors for post-LT mortality including age, obesity, diabetes mellitus (DM), renal insufficiency, etiology of liver failure, model for end-stage liver disease (MELD) score, sarcopenia, preoperative hemoglobin level, preoperative mechanical ventilation, and intraoperative blood loss [4–12]. In addition, donor age and small-for-size grafts are donor-related risk factors for mortality [13,14]. Although these findings are important for preoperative risk stratification, most of these variables are not easily modifiable.
Intraoperative hypotension (IOH) is a common complication during LT owing to its surgical complexity, severe associated coagulopathy, and potentially massive intraoperative blood loss. This is concerning as IOH has been associated with postoperative mortality in cardiac and noncardiac surgeries [15,16]. Patients with ESLD have unique hyperdynamic hemodynamics characterized by high cardiac output, low systemic vascular resistance (SVR), and relatively low systemic blood pressure. Therefore, baseline hemodynamic differences should be considered before extrapolating the results from other surgical settings to the LT population. Although one study identified a relationship between IOH and 30-day mortality post-LT [17], the relationship between IOH and long-term mortality has yet to be elucidated.
This retrospective study aimed to investigate the relationship between post-LT 1-year mortality and IOH of different severities and durations. We hypothesized that IOH, defined by both the threshold and duration, is independently associated with 1-year mortality after LT.

Materials and Methods

Ethical approval

Ethical approval was obtained from the University of Pittsburgh Institutional Review Board (STUDY20050148) in Pennsylvania, USA. This study was conducted in accordance with the principles of the Declaration of Helsinki. The need for informed consent was waived owing to the retrospective study design.

Study design and cohort

We conducted a retrospective observational cohort study of all adult patients aged ≥ 18 years who underwent LT, including both living and deceased donor transplantations, at the University of Pittsburgh Medical Center between January 2012 and June 2023. Patients who died intraoperatively, underwent simultaneous transplantation, or had chronic kidney disease requiring hemodialysis were excluded. Patients with missing intraoperative mean arterial pressure (MAP) data were excluded from the analysis. No imputation was performed for missing data, and all analyses were conducted using complete-case data.

Anesthetic management

Anesthetic management was performed according to institutional practices as described previously [18]. Briefly, after the induction of general anesthesia by endotracheal intubation, transesophageal echocardiography, pulmonary artery catheterization, and two arterial lines were used as standard intraoperative monitors. Red blood cell transfusions were performed during LT to maintain a hematocrit concentration of 25%–30%. Freshly frozen plasma, platelet, and cryoprecipitate transfusions were performed using thromboelastography. The maintenance of intraoperative hemodynamics was left to the discretion of each faculty anesthesiologist. The timing, dose, and type of vasopressor administration also differed among anesthesiologists.

Data collection, intraoperative hypotension, and lactate levels

Patient characteristics, preoperative laboratory findings, and perioperative variables were retrospectively collected from the patients’ medical or anesthesia records. Detailed donor characteristics (e.g., degree of steatosis, use of machine perfusion, donor risk index, extended-criteria donor status, and degree of liver fibrosis) were not consistently available in the institutional database and were therefore not included in the primary analysis. The MAP was automatically recorded via the inserted arterial line at 1-min intervals throughout the surgery. Raw MAP values were extracted from the anesthesia records and filtered for obvious artifacts by removing MAP values < 20 or > 150 mmHg. We defined IOH as an MAP < 65 mmHg and examined the duration spent below various MAP thresholds, including < 65, < 60, < 55, < 50, < 45, and < 40 mmHg for each patient, based on recent literature for LT populations [19,20]. For descriptive purpose, IOH was also categorized by severity according to MAP thresholds: mild (MAP < 65 and < 60 mmHg), moderate (MAP < 55 and < 50 mmHg), and severe (MAP < 45 and < 40 mmHg). For each patient, the duration of IOH was calculated as the cumulative time the MAP remained below a pre-specified threshold during anesthesia. All hypotensive episodes occurring in the same patient were summed to obtain a single cumulative duration value for each MAP threshold, regardless of vasopressor administration. Each patient contributed one cumulative duration value per MAP threshold, and was counted once in each corresponding analysis. This cumulative duration-based definition of IOH exposure is a commonly used and well-established approach in the perioperative and LT literature [17,21]. Furthermore, we divided the patients into four groups according to duration in each threshold: < 5 min, 5–9 min, 10–14 min, and ≥ 15 min. For the MAP threshold < 40 mmHg, as the proportion of patients exposed for ≥ 5 min was expected to be extremely low, patients were divided into two groups: < 5 min and ≥ 5 min. These duration categories were defined using 5-min intervals to balance clinical interpretability and statistical stability and to facilitate the assessment of potential dose-response relationships between the duration of IOH and postoperative outcomes. This approach is consistent with prior studies in the LT population that categorized hypotension exposure at 5-min intervals [20].
The total intraoperative doses of norepinephrine (µg), epinephrine (µg), and vasopressin (units) were extracted from the anesthetic records. Based on previously published literature [22], these doses were converted to a single composite variable representing the total catecholamine dose using the following equation:
Total catecholamine dose (µg) = norepinephrine (µg) + epinephrine (µg) + 2.5 × body weight (kg) × vasopressin (units).
Lactate levels were collected, analyzed, and reported using intraoperative arterial blood gas analyses. Intraoperative lactate measurements for each patient were calculated as time-weighted average lactate levels for a more comprehensive assessment. This was calculated as the area under the curve divided by the time interval between the first and last measurement. At our institution, arterial blood gas analysis is routinely performed at least once per hour during LT with additional measurements obtained as clinically indicated. In this cohort, the median number of intraoperative arterial blood gas analyses per patient was 10 (9, 11). As sampling was not performed at fixed intervals but was supplemented by clinically indicated measurements, the median sampling interval could not be reliably calculated. This methodological approach has been previously described in perioperative LT research [23].

Outcomes

The primary outcome was all-cause mortality within 1 year of LT. The secondary outcome was intraoperative lactate levels calculated using the time-weighted average method.

Statistical analysis

Continuous variables are reported as the median (Q1, Q3) and were compared using the Student t or Mann-Whitney U test, as appropriate. Categorical data are reported as the count (percent [%]) and were compared using chi-square tests. Kaplan–Meier analyses for 1-year mortality were performed among the four groups for each threshold and log-rank tests were applied. Univariate and multivariate Cox proportional hazards models were used to calculate the unadjusted and adjusted hazard ratios (HRs). Multivariate analysis was performed using known risk factors for post-LT mortality, including age, sex, DM, MELD score, donor type (living or deceased), donor age, and intraoperative blood loss [4,6,8,11,12,24–26]. In addition, the total catecholamine dose was included as a covariate in the multivariate model because intraoperative blood pressure is closely related to vasopressor use. We performed further analyses to identify the minimum duration of hypotension that was independently associated with 1-year mortality for each MAP threshold. Exploratory analyses were performed without formal adjustments for multiple comparisons. Exploratory interaction analyses were also performed to assess the potential heterogeneity in the association between severe IOH and 1-year mortality stratified by the MELD score. Intraoperative lactate levels were also compared among the four groups for various MAP thresholds (< 65, < 60, < 55, < 50, and < 45 mmHg) using the Kruskal–Wallis test and between the two groups for the MAP threshold < 40 mmHg using the Mann–Whitney U test. Additionally, the association between IOH exposure and time-weighted average intraoperative lactate levels was examined using simple linear regression as an exploratory, unadjusted secondary analysis. Variance inflation factors (VIFs) were calculated to assess multicollinearity among the covariates included in the multivariate model. A VIF < 5 was considered acceptable.
The results of the Cox proportional hazard models are expressed as HRs and 95% CIs. All P values were two-sided, and values < 0.05 were considered statistically significant. All analyses were conducted using StataSE version 17.0 (College Station) and EZR (Saitama Medical Center), a graphical user interface for R version 4.4.1 (R Foundation for Statistical Computing).

Results

In total, 1149 patients aged ≥ 18 years underwent LT during the study period. Of these, 8 patients who died intraoperatively, 33 who underwent simultaneous transplantation, 32 who had chronic kidney disease dependent on hemodialysis, and 13 who had missing data were excluded, leaving 1063 patients included in the analysis. The median duration of surgery was 469 min (388, 549 min), and the median duration of each MAP threshold was as follows: 59 min (24–137 min) for < 65 mmHg, 17 min (6–43 min) for < 60 mmHg, 4 min (1–12 min) for < 55 mmHg, 1 min (0–3 min) for < 50 mmHg, 0 min (0–1 min) for < 45 mmHg, and 0 min (0–0 min) for < 40 mmHg. Although the median duration of exposure at < 45 mmHg and < 40 mmHg was 0 min, 326 (30.7%) and 162 (15.2%) patients, respectively, experienced at least 1 min at these thresholds. The distribution of different durations of hypotension for each MAP threshold is shown in Fig. 1, with the MAP threshold < 40 mmHg divided into only two duration groups (< 5 min and ≥ 5 min). In our cohort, almost all patients (98.9%) were exposed to IOH for at least 1 min. The percentage of patients exposed to mild IOH for ≥ 15 min was high (84.2% < 65 mmHg and 54.3% < 60 mmHg), while the percentage of patients exposed to moderate IOH for ≥ 15 min was low (20.2% < 55 mmHg and 3.3% < 50 mmHg). The majority of patients were exposed to severe IOH for < 5 min (95.9% at < 45 mmHg and 99.1% at < 40 mmHg), and the proportion of patients exposed to severe IOH for a prolonged duration was small (1.2% at < 45 mmHg for ≥ 10 min and 0.9% at < 40 mmHg for ≥ 5 min).
Patient characteristics, preoperative laboratory findings, and intraoperative variables are shown in Table 1. The median age was 59 years (50–65 years), male sex was dominant (67.0%), the median MELD score was 22 (15–29), and the median donor age was 40 years (30–49 years). Venovenous bypass was performed in 766 of the 1063 patients (72.1%). Norepinephrine, epinephrine, and vasopressin were used intraoperatively in 82.8%, 82.5%, and 58.5% of patients, respectively (Supplementary Table 1).
Fig. 2 shows Kaplan–Meier analyses for 1-year mortality comparing the duration of hypotension for each MAP threshold. Overall, 76 patients (7.1%) died within one year of LT. Although no significant differences were found between MAP thresholds < 65, < 60, and < 55 mmHg (P = 0.245, P = 0.269, and P = 0.086, respectively), differences were found between MAP thresholds < 50, < 45, and < 40 mmHg (P = 0.007, P < 0.001, and P = 0.002, respectively). In addition, survival over the 1-year period was significantly worse for MAP thresholds < 50 and < 45 mmHg for ≥ 15 min and < 40 mmHg for ≥ 5 min.
The results of the univariate and multivariate analyses are shown in Table 2. The HRs and 95% CIs for all covariates included in the multivariate model are shown in Supplementary Table 2. All covariates had VIFs < 2.0, suggesting that multicollinearity was not a concern. Although no groups were independently associated with 1-year mortality for MAP thresholds < 65, < 60, and < 55 mmHg, MAP thresholds < 50 and < 45 mmHg for ≥ 15 min (HR = 3.19, 95% CI [1.22–8.35], P = 0.018 and HR = 6.76, 95% CI [1.60–28.6], P = 0.009, respectively) and < 40 mmHg for ≥ 5 min (HR = 6.72, 95% CI [1.56–28.9], P = 0.010) were independently associated with 1-year mortality. In an exploratory interaction analysis stratified by MELD score, no significant effect modification was observed for the association between MAP < 50 mmHg for ≥ 15 min and 1-year mortality (P for interaction = 0.927; Supplementary Table 3).
Exploratory analyses to investigate the minimum duration of IOH associated with 1-year mortality are presented in Table 3. The shortest duration for each MAP threshold that showed associations with 1-year mortality were as follows: ≥ 43 min for MAP < 55 mmHg (HR = 2.48, 95% CI [1.05–5.86], P = 0.038), ≥ 14 min for MAP < 50 mmHg (HR = 2.60, 95% CI [1.01–6.65], P = 0.047), ≥ 13 min for MAP < 45 mmHg (HR = 4.98, 95% CI [1.18–21.0] P = 0.029), and ≥ 5 min for MAP < 40 mmHg (HR = 6.72, 95% CI [1.56–28.9], P = 0.010). Prolonged durations for MAPs < 65 mmHg and < 60 mmHg were not independently associated with increased 1-year mortality. The predicted probability of 1-year mortality, estimated using MAP thresholds and durations, was assessed by univariate logistic regression analysis (Figs. 3A and B). Both lower MAPs and longer durations increased the predicted probability of 1-year mortality, and the interaction between the MAP and duration was significantly associated with 1-year mortality (P = 0.020).
Intraoperative lactate levels for each MAP threshold across durations of hypotension are shown in Fig. 4. Within each MAP threshold, lactate levels differed significantly across categories of hypotension duration (< 5 min, 5–9 min, 10–14 min, and ≥ 15 min) for all thresholds except < 40 mmHg (MAP < 65 mmHg, P < 0.001; MAP < 60 mmHg, P < 0.001; MAP < 55 mmHg, P < 0.001; MAP < 50 mmHg, P < 0.001; and MAP < 45 mmHg, P = 0.009). Simple linear regression analyses revealed that longer durations of IOH were significantly associated with higher lactate levels across all MAP thresholds (all P < 0.001). Furthermore, the β coefficients from the simple linear regression increased progressively as the MAP threshold decreased, suggesting that lower MAP levels were associated with greater elevations in intraoperative lactate levels (Fig. 5).

Discussion

In this study, IOH was significantly associated with 1-year mortality after LT, independent of known perioperative confounders. Mild IOH, represented by MAP thresholds < 65 and < 60 mmHg, was not associated with increased 1-year mortality, whereas moderate-to-severe IOH, represented by MAP thresholds < 55, < 50, < 45, and < 40 mmHg, were correlated with poor post-LT outcomes, even with short durations of hypotension. However, because this study evaluated all-cause mortality and was not designed to determine cause-specific death, these findings should be interpreted as prognostic associations rather than evidence of a direct causal mechanism.
IOH is a common complication in patients undergoing LT surgery. Although the present study did not evaluate preoperative hemodynamic parameters, patients with ESLD typically exhibit a hyperdynamic circulatory state characterized by low SVR due to splanchnic and peripheral vasodilation and high cardiac output. This preoperative hemodynamic profile, combined with vasodilation caused by general anesthesia, is thought to frequently lead to IOH during LT. Furthermore, massive intraoperative blood loss, severe coagulopathy, acidosis or electrolyte disorders, cross-clamping of the portal vein and inferior vena cava, and volume shifts to the allograft after reperfusion may contribute to hemodynamic instability [27]. In fact, IOH was found to be common and prolonged during LT in a large retrospective study [17]. In an exploratory interaction analysis stratified by the MELD score, the association between severe IOH and 1-year mortality was directionally consistent across subgroups, with no significant evidence of effect modification.
The relationship between IOH and postoperative mortality has been investigated in patients undergoing noncardiac surgery. Zhao et al. [16] identified that an MAP ≤ 65 mmHg was highly associated with postoperative mortality in their cohort. Another meta-analysis found that an MAP < 60 mmHg for ≥ 5 min was correlated with an increased risk of postoperative 30-day mortality in non-cardiac surgery [28]. However, in our cohort, MAP thresholds < 65 and < 60 mmHg were not associated with post-LT mortality. One explanation may be the characteristic hemodynamic profile of patients with ESLD, who often exhibit chronic hypotension due to persistently low SVR accompanied by hyperdynamic circulation characterized by high cardiac output. These hemodynamic features may be associated with a leftward shift in the lower limit of cerebral and systemic autoregulation, allowing for the maintenance of tissue perfusion despite arterial pressures that would typically be considered hypotensive in other surgical populations [29]. Other studies have suggested that LT recipients may tolerate a lower MAP for a longer duration, which supports the theory of autoregulatory shift [17,20].
Limited data are available regarding the relationship between IOH and post-LT mortality. Wang et al. [17] performed a retrospective study to determine whether an MAP threshold was associated with 30-day mortality in LT. They reported that exposure to MAPs < 45 mmHg for ≥ 6 min, < 50 mmHg for ≥10 min, or < 55 mmHg for ≥ 25 min showed a significant association with postoperative 30-day mortality, whereas exposure to MAP < 60 mmHg up to 120 min was not associated with mortality. Although our findings that exposure to MAPs < 45 mmHg for ≥ 13 min, < 50 mmHg for ≥ 14 min, or < 55 mmHg for ≥ 43 min were associated with post-LT mortality is similar to the findings of that study, the hypotension duration thresholds associated with increased mortality were not identical. This variation may be attributable to differences in patient characteristics and intraoperative management strategies. Their cohort had higher preoperative MELD scores, suggesting more severe ESLD. Furthermore, differences in the fluid, transfusion, and hemodynamic management protocols may have contributed to the observed discrepancies. In fact, the proportion of patients exposed to each MAP threshold for ≥ 15 min in their cohort was approximately 95% for < 65 mmHg, 80% for < 60 mmHg, 55% for < 55 mmHg, 28% for < 50 mmHg, and 12% for < 45 mmHg, while the proportion in our cohort was 84.2% for < 65 mmHg, 54.3% for < 60 mmHg, 20.3% for < 55 mmHg, 3.3% for < 50 mmHg, and 0.7% for < 45 mmHg, which is a major contrast between these studies. Overall, our higher MAP threshold findings may be informative for anesthesiologists involved in clinical practice.
We also observed that longer durations of IOH were significantly associated with higher lactate levels, and this association became more pronounced at lower MAP thresholds, suggesting a graded association between the depth and duration of hypotension and lactate elevation. Elevated lactate levels may reflect a combination of excessive production via anaerobic metabolism and impaired clearance, both of which have been proposed for LT [30]. Although IOH has been suggested to cause tissue hypoperfusion and subsequent anaerobic metabolism [31], our study was not designed to establish a causal relationship between IOH and hyperlactatemia. Previous reports have shown an association between intraoperative hyperlactatemia and adverse post-LT outcomes, including primary graft dysfunction and mortality [23,32,33]. Although we observed a correlation between IOH exposure and intraoperative lactate levels, these findings should be interpreted with caution. Further studies are needed to determine whether lactate levels can serve as a surrogate marker for IOH exposure or as a predictor of postoperative prognosis in this population.
This study has several limitations that should be considered when interpreting the findings. First, its single-center observational study design inhibited the establishment of a causal relationship, potentially reducing the generalizability of the findings to other institutions. Second, residual confounding may have been introduced as several important variables could not be included in the analysis due to inconsistent availability. These included detailed donor-related characteristics, such as the donor risk index, graft steatosis, and use of machine perfusion, as well as key intraoperative surgical factors, including anhepatic time and type of caval anastomosis (piggyback vs. side-to-side). Although we adjusted for the type of transplantation (living vs. deceased donor), this variable cannot fully account for differences in graft quality, such as graft steatosis, which is more prevalent in deceased donor LT and is an established risk factor for adverse post-transplant outcomes. In addition, data on the specific causes of death were not available for this cohort, precluding a cause-specific mortality analysis. Therefore, although IOH was associated with 1-year all-cause mortality, the underlying mechanisms linking hypotension and mortality could not be examined directly. Deaths may have resulted from heterogeneous causes, including graft-related complications, infection, malignancy recurrence, or other non-ischemia-related factors. In addition, lactate levels were analyzed as an exploratory, unadjusted secondary outcome, and residual confounding factors related to graft function and differential vasoactive agent use, particularly epinephrine, could not be excluded. Third, no vasopressor selection or administration protocol exists at our institution; thus, these decisions were made at the discretion of each anesthesiologist, which may have introduced practice variability. Fourth, our analysis did not separate the pre- and post-reperfusion phases, which are physiologically distinct in terms of SVR, cardiac output, and vasopressor requirements. Future investigations should assess phase-specific blood pressure thresholds to clarify their prognostic impact. Finally, although the absolute blood pressure was used to identify IOH, relative changes from baseline blood pressures have been reported to be associated with postoperative outcomes [34]. However, we could not apply relative thresholds because the baseline blood pressure could not easily be determined retrospectively. However, since relative thresholds have been reported to be less sensitive than absolute thresholds in detecting an association with postoperative outcomes [35], we believe that absolute blood pressure is more applicable to clinical practice.
In conclusion, we found that IOH was independently associated with 1-year mortality after LT. Moderate-to-severe IOH was associated with poor outcomes, even with short durations, whereas mild IOH showed no association with post-LT mortality. These findings highlight the importance of careful intraoperative blood pressure management and warrant further studies to determine optimal blood pressure thresholds and duration limits to improve postoperative outcomes in LT recipients.

Acknowledgments

We would like to express our sincere gratitude to the attending transplant surgeons at the University of Pittsburgh Medical Center for performing all liver transplant surgeries during the study period.
The authors confirm that no generative artificial intelligence (AI) or AI-assisted technologies were used in the preparation, writing, or editing of this manuscript.

Funding

None.

Conflicts of Interest

No potential conflict of interest relevant to this article was reported.

Data Availability

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

Author Contributions

Akira Katayama (Conceptualization; Data curation; Formal analysis; Methodology; Visualization; Writing – original draft; Writing – review & editing)

Ezeldeen Abuelkasem (Conceptualization; Methodology; Supervision; Writing – review & editing)

Leon C. Su (Formal analysis; Methodology; Writing – review & editing)

David W. Wang (Conceptualization; Methodology; Supervision; Writing – original draft; Writing – review & editing)

Supplementary Materials

Supplementary Table 1.
Intraoperative vasoactive agent use in the study cohort.
kja-25965-Supplementary-Table-1.pdf
Supplementary Table 2.
Multivariable cox regression analysis including intraoperative hypotension (MAP <50 mmHg) for 1-year mortality.
kja-25965-Supplementary-Table-2.pdf
Supplementary Table 3.
Exploratory subgroup analysis stratified by MELD score.
kja-25965-Supplementary-Table-3.pdf

Fig. 1.
Distribution of hypotension duration across MAP thresholds. Stacked bar graphs illustrate the proportion of patients exposed to IOH at each MAP threshold (< 65 to < 40 mmHg), categorized by duration (< 5, 5–9, 10–14, and ≥ 15 min). Although IOH occurred in 98.9% of the patients, exposure to MAP < 50 mmHg for ≥ 5 min was observed in < 20% of the patients. MAP: mean arterial pressure, IOH: intraoperative hypotension.
kja-25965f1.jpg
Fig. 2.
Kaplan–Meier survival analysis based on intraoperative MAP thresholds and exposure durations. Kaplan–Meier curves show the 1-year survival rates stratified by the duration of intraoperative hypotension at each MAP threshold. Log-rank tests showed significant differences among the groups at MAP thresholds < 50, < 45, and < 40 mmHg. (A) MAP < 65 mmHg; (B) MAP < 60 mmHg; (C) MAP < 55 mmHg; (D) MAP < 50 mmHg; (E) MAP < 45 mmHg; and (F) MAP < 40 mmHg. MAP: mean arterial pressure.
kja-25965f2.jpg
Fig. 3.
Relationship between IOH and predicted 1-year mortality. (A) The curves depict the relationship between the duration of IOH and predicted probability of 1-year mortality stratified by MAP threshold. Predicted mortality tended to increase with longer exposure duration and lower MAP levels. (B) Contour plot illustrating the predicted probability of 1-year mortality according to the MAP and duration of IOH. Lower MAP levels and longer durations were associated with higher predicted mortality rates. MAP: mean arterial pressure, IOH: intraoperative hypotension.
kja-25965f3.jpg
Fig. 4.
Intraoperative lactate levels stratified by MAP threshold and duration of hypotension. Box plots illustrating intraoperative lactate levels stratified by MAP threshold and duration of hypotension (< 5, 5–9, 10–14, and ≥ 15 min). Lactate levels tended to increase with both a lower MAP and longer duration of hypotension. MAP: mean arterial pressure.
kja-25965f4.jpg
Fig. 5.
Association between duration of hypotension and intraoperative lactate level across MAP thresholds. Simple linear regression analyses were performed for each MAP threshold to assess the relationship between hypotension duration and intraoperative lactate levels. Longer duration of hypotension was associated with higher lactate levels, and the strength of this association was greater at lower MAP thresholds. MAP: mean arterial pressure.
kja-25965f5.jpg
Table 1.
Preoperative and Intraoperative Characteristics
Variable
Age (yr) 59 (50, 65)
Sex (F) 351 (33.0)
BMI (kg/m2) 28.4 (25, 33.1)
Etiology
 Alcoholic 254 (23.9)
 NASH 240 (22.6)
 Malignancy 186 (17.5)
 PSC 91 (8.6)
Comorbidity
 Hypertension 542 (51.0)
 DM 364 (31.6)
Coronary artery disease 154 (14.5)
MELD score 22 (15, 29)
Donor age (yr) 40 (30, 49)
Living donor liver transplantation 559 (52.6)
AST (U/L) 47 (33, 72)
ALT (U/L) 29 (19. 47)
Albumin (g/dL) 3.3 (2.9, 3.7)
Bilirubin (mg/dL) 2.7 (1.4, 6.2)
Creatinine (mg/dL) 1.0 (0.8, 1.4)
Platelet count (×103/µL) 8.3 (5.7, 12.7)
PT-INR 1.6 (1.4, 2.1)
Sodium (mEq/L) 136 (132, 138)
HbA1c (%) 5.1 (4.5, 6.1)
Duration of surgery (min) 469 (388, 549)
Veno-venous bypass 766 (72.1)
CIT (min) 161 (110, 416)
WIT (min) 30 (25, 36)
Crystalloid (mL) 6599 (4927, 8766)
Colloid (mL) 1250 (750, 2000)
Red blood cell (mL) 550 (0, 1200)
Fresh frozen plasma (mL) 0 (0, 800)
Platelets (mL) 0 (0, 291)
Intraoperative blood loss (mL) 800 (500, 1500)
Total catecholamine dose (μg) 2101 (758, 4703)

Values are presented as median (Q1, Q3) or number (%). BMI: body mass index, NASH: non-alcoholic steatohepatitis, PSC: primary sclerosing cholangitis, DM: diabetes mellitus, MELD: model for endstage liver disease, AST: aspartate aminotransferase, ALT: alanine transaminase, PT-INR: prothrombin international normalized ratio, CIT: cold ischemic time, WIT: warm ischemic time.

Table 2.
Univariate and Multivariate Analysis of 1-Year Mortality by Threshold and Duration
MAP threshold (mmHg) Duration (min) Patients (n [%]) Univariate analysis Multivariate analysis
Hazard ratio 95% CI P value Hazard ratio 95% CI P value
MAP < 65 < 5 63 (5.9) (Reference) (Reference)
5–9 47 (4.4) - - - - - -
10–14 58 (5.5) 0.65 0.16–2.72 0.556 0.72 0.17–3.04 0.653
≥ 15 895 (84.2) 0.96 0.39–2.38 0.930 0.79 0.31–2.02 0.623
MAP < 60 < 5 219 (20.6) (Reference) (Reference)
5–9 143 (13.5) 0.98 0.38–2.52 0.962 1.09 0.42–2.82 0.864
10–14 124 (11.7) 1.46 0.61–3.53 0.397 1.00 0.37–2.75 0.993
≥ 15 577 (54.3) 1.70 0.88–3.27 0.111 1.38 0.70–2.73 0.353
MAP < 55 < 5 543 (51.1) (Reference) (Reference)
5–9 201 (18.9) 1.05 0.55–2.00 0.874 0.97 0.48–1.95 0.934
10–14 104 (9.8) 0.93 0.39–2.21 0.864 1.02 0.42–2.47 0.961
≥ 15 215 (20.2) 1.87 1.10–3.16 0.020 1.41 0.77–2.57 0.262
MAP < 50 < 5 870 (81.8) (Reference) (Reference)
5–9 125 (11.8) 1.57 0.84–2.93 0.159 1.41 0.71–2.82 0.328
10–14 33 (3.1) 1.52 0.48–4.87 0.479 0.58 0.08–4.26 0.590
≥ 15 35 (3.3) 3.55 1.62–7.81 0.002 3.19 1.22–8.35 0.018
MAP < 45 < 5 1019 (95.9) (Reference) (Reference)
5–9 32 (3.0) 0.42 0.06–3.00 0.385 0.49 0.07–3.52 0.474
10–14 5 (0.5) - - - - - -
≥ 15 7 (0.7) 8.36 2.63–26.6 <0.001 6.76 1.60–28.6 0.009
MAP < 40 < 5 1053 (99.1) (Reference) (Reference)
≥ 5 10 (0.9) 5.25 1.65–16.7 0.005 6.72 1.56–28.9 0.010

Univariate and multivariate analyses could not be performed for patients with MAPs < 65 mmHg for 5–9 min and < 45 mmHg for 10–14 min because the mortality rate was 0 for these groups. MAP: mean arterial pressure.

Table 3.
Multivariate Analysis for Independent Association with 1-Year Mortality
MAP threshold, mmHg Duration, min Patients, n (%) Multivariate analysis
Hazard ratio 95% CI P value
MAP < 65 ≥ 60 528 (49.7) 1.28 0.77–2.14 0.340
≥ 120 289 (27.2) 1.57 0.94–2.62 0.085
≥ 180 133 (12.5) 1.24 0.64–2.42 0.520
MAP < 60 ≥ 60 195 (18.3) 1.37 0.78–2.42 0.276
≥ 120 56 (5.3) 1.86 0.79–4.35 0.154
≥ 180 20 (1.9) 2.81 0.86–9.19 0.088
MAP < 55 ≥ 40 54 (5.1) 2.23 0.94–5.26 0.068
≥ 42 51 (4.8) 2.32 0.98–5.47 0.055
≥ 43 47 (4.4) 2.48 1.05–5.86 0.038
MAP < 50 ≥ 10 68 (6.4) 1.70 0.72–4.01 0.223
≥ 13 44 (4.1) 2.40 0.94–6.13 0.068
≥ 14 40 (3.8) 2.60 1.01–6.65 0.047
MAP < 45 ≥ 10 12 (1.1) 3.52 0.84–14.7 0.084
≥ 12 10 (9.4) 4.04 0.96–17.0 0.057
≥ 13 9 (0.8) 4.98 1.18–21.0 0.029
MAP < 40 ≥ 3 28 (2.6) 2.01 0.63–6.44 0.238
≥ 4 17 (1.6) 2.52 0.61–10.4 0.202
≥ 5 10 (0.9) 6.72 1.56–28.9 0.010

The duration categories represent the cumulative time spent below each MAP threshold. Patients with higher duration thresholds were also included in analyses using lower thresholds. MAP: mean arterial pressure.

References

1. Hwang WJ, Jeon JP, Kang SH, Chung HS, Kim JY, Park CS. Sluggish decline in a post-transplant model for end-stage liver disease score is a predictor of mortality in living donor liver transplantation. Korean J Anesthesiol 2010; 59: 160-6.
crossref pmid pmc
2. Baganate F, Beal EW, Tumin D, Azoulay D, Mumtaz K, Black SM, et al. Early mortality after liver transplantation: defining the course and the cause. Surgery 2018; 164: 694-704.
crossref pmid
3. Kwong AJ, Kim WR, Lake JR, Schladt DP, Handarova D, Howell J, et al. OPTN/SRTR 2023 Annual Data Report: Liver. Am J Transplant 2025; 25: S193-287.
crossref pmid pmc
4. Chen HP, Tsai YF, Lin JR, Liu FC, Yu HP. Recipient age and mortality risk after liver transplantation: a population-based cohort study. PLoS One 2016; 11: e0152324.
crossref pmid pmc
5. Gil E, Kim JM, Jeon K, Park H, Kang D, Cho J, et al. Recipient age and mortality after liver transplantation: a population-based cohort study. Transplantation 2018; 102: 2025-32.
crossref pmid pmc
6. Watt KD, Pedersen RA, Kremers WK, Heimbach JK, Charlton MR. Evolution of causes and risk factors for mortality post-liver transplant: results of the NIDDK long-term follow-up study. Am J Transplant 2010; 10: 1420-7.
crossref pmid pmc
7. Nagai S, Collins K, Chau LC, Safwan M, Rizzari M, Yoshida A, et al. Increased risk of death in first year after liver transplantation among patients with nonalcoholic steatohepatitis vs liver disease of other etiologies. Clin Gastroenterol Hepatol 2019; 17: 2759-68.e5.
crossref pmid
8. Beckmann S, Drent G, Ruppar T, Nikolić N, de Geest S. Body weight parameters are related to morbidity and mortality after liver transplantation: a systematic review and meta-analysis. Transplantation 2019; 103: 2287-303.
crossref pmid
9. Jiang MJ, Wu MC, Duan ZH, Wu J, Xu XT, Li J, et al. Prevalence and clinical impact of sarcopenia in liver transplant recipients: a meta-analysis. World J Gastroenterol 2024; 30: 956-68.
crossref pmid pmc
10. Huang CT, Lin HC, Chang SC, Lee WC. Pre-operative risk factors predict post-operative respiratory failure after liver transplantation. PLoS One 2011; 6: e22689.
crossref pmid pmc
11. Zhang XM, Fan H, Wu Q, Zhang XX, Lang R, He Q. In-hospital mortality of liver transplantation and risk factors: a single-center experience. Ann Transl Med 2021; 9: 369.
crossref pmid pmc
12. Jóźwik A, Karpeta E, Nita M, Łągiewska B, Pacholczyk M. Impact of blood loss and intraoperative blood transfusion during liver transplantation on the incidence of early biliary complications and mortality. Transplant Proc 2020; 52: 2477-9.
crossref pmid
13. Macshut M, Kaido T, Yao S, Yagi S, Ito T, Kamo N, et al. Older donor age is a risk factor for negative outcomes after adult living donor liver transplantation using small-for-size grafts. Liver Transpl 2019; 25: 1524-32.
crossref pmid pdf
14. Del Prete L, Quintini C, Diago Uso T. The small-for-size syndrome in living donor liver transplantation: current management. Updates Surg 2025; 77: 1797-806.
crossref pmid pmc pdf
15. de la Hoz MA, Rangasamy V, Bastos AB, Xu X, Novack V, Saugel B, et al. Intraoperative hypotension and acute kidney injury, stroke, and mortality during and outside cardiopulmonary bypass: a retrospective observational cohort study. Anesthesiology 2022; 136: 927-39.
crossref pmid pdf
16. Zhao B, Chen X, Chen Q, Li G, Chen Z, Yang Z, et al. Intraoperative hypotension and related risk factors for postoperative mortality after noncardiac surgery in elderly patients: a retrospective analysis report. Clin Interv Aging 2021; 16: 1757-67.
crossref pmid pmc pdf
17. Wang L, Bui CM, Hofer I, Gabel E, Wray C, Xia VW. Intraoperative hypotension and 30-d mortality after liver transplantation. Transplant Direct 2022; 8: e1380.
crossref pmid pmc
18. Sakai T, Matsusaki T, Marsh JW, Hilmi IA, Planinsic RM. Comparison of surgical methods in liver transplantation: retrohepatic caval resection with venovenous bypass (VVB) versus piggyback (PB) with VVB versus PB without VVB. Transpl Int 2010; 23: 1247-58.
crossref pmid
19. Shah NJ, Mentz G, Kheterpal S. The incidence of intraoperative hypotension in moderate to high risk patients undergoing non-cardiac surgery: a retrospective multicenter observational analysis. J Clin Anesth 2020; 66: 109961.
crossref pmid
20. Caragata R, Emerson S, Santema ML, Selzner N, Sapisochin G, Wang S, et al. Intraoperative hypotension and the risk of acute kidney injury following liver transplantation. Clin Transplant 2023; 37: e15053.
crossref pmid
21. Mizota T, Hamada M, Matsukawa S, Seo H, Tanaka T, Segawa H. Relationship between intraoperative hypotension and acute kidney injury after living donor liver transplantation: a retrospective analysis. J Cardiothorac Vasc Anesth 2017; 31: 582-9.
crossref pmid
22. Goradia S, Sardaneh AA, Narayan SW, Penm J, Patanwala AE. Vasopressor dose equivalence: a scoping review and suggested formula. J Crit Care 2021; 61: 233-40.
crossref pmid
23. Katayama A, Abuelkasem E, Ligon MM, Longhitano Y, Wang DW. Intraoperative lactate level and lactate clearance is associated with long-term mortality in liver transplant recipients. Anesth Pain Med (Seoul) 2026; 21: 72-82.
crossref pmid pmc pdf
24. Serrano MT, Sabroso S, Esteban LM, Berenguer M, Fondevila C, Lorente S, et al. Mortality and causes of death after liver transplantation: analysis of sex differences in a large nationwide cohort. Transpl Int 2022; 35: 10263.
crossref pmid pmc
25. Hoofnagle JH, Lombardero M, Zetterman RK, Lake J, Porayko M, Everhart J, et al. Donor age and outcome of liver transplantation. Hepatology 1996; 24: 89-96.
crossref pmid
26. Humar A, Ganesh S, Jorgensen D, Tevar A, Ganoza A, Molinari M, et al. Adult living donor versus deceased donor liver transplant (LDLT versus DDLT) at a single center: time to change our paradigm for liver transplant. Ann Surg 2019; 270: 444-51.
crossref pmid
27. Rudnick MR, Marchi LD, Plotkin JS. Hemodynamic monitoring during liver transplantation: a state of the art review. World J Hepatol 2015; 7: 1302-11.
crossref pmid pmc
28. An R, Pang QY, Liu HL. Association of intra-operative hypotension with acute kidney injury, myocardial injury and mortality in non-cardiac surgery: a meta-analysis. Int J Clin Pract 2019; 73: e13394.
crossref pmid pdf
29. Bezinover D, Mukhtar A, Wagener G, Wray C, Blasi A, Kronish K, et al. Hemodynamic instability during liver transplantation in patients with end-stage liver disease: a consensus document from ILTS, LICAGE, and SATA. Transplantation 2021; 105: 2184-200.
crossref pmid
30. Kraut JA, Madias NE. Lactic acidosis. N Engl J Med 2014; 371: 2309-19.
crossref pmid
31. Levy B. Lactate and shock state: the metabolic view. Curr Opin Crit Care 2006; 12: 315-21.
crossref pmid
32. Golse N, Guglielmo N, El Metni A, Frosio F, Cosse C, Naili S, et al. Arterial lactate concentration at the end of liver transplantation is an early predictor of primary graft dysfunction. Ann Surg 2019; 270: 131-8.
crossref pmid
33. Coeckelenbergh S, Drouard L, Ickx B, Lucidi V, Germanova D, Desebbe O, et al. Arterial lactate concentration at the end of liver transplantation is independently associated with one-year mortality. Transplant Proc 2023; 55: 147-52.
crossref pmid
34. Panwar R, Tarvade S, Lanyon N, Saxena M, Bush D, Hardie M, et al. Relative hypotension and adverse kidney-related outcomes among critically Ill patients with shock. A multicenter, prospective cohort study. Am J Respir Crit Care Med 2020; 202: 1407-18.
crossref pmid
35. Gregory A, Stapelfeldt WH, Khanna AK, Smischney NJ, Boero IJ, Chen Q, et al. Intraoperative hypotension is associated with adverse clinical outcomes after noncardiac surgery. Anesth Analg 2021; 132: 1654-65.
crossref pmid pmc


ABOUT
ARTICLE CATEGORY

Browse all articles >

BROWSE ARTICLES
AUTHOR INFORMATION
Editorial Office
101-3503, Lotte Castle President, 109 Mapo-daero, Mapo-gu, Seoul 04146, Korea
Tel: +82-2-792-5128    Fax: +82-2-792-4089    E-mail: journal@anesthesia.or.kr                
Business Name: Korean Society of Anesthesiologists
Business Registration: 106-82-07194
Representative: Young-Tae Jeon

Copyright © 2026 by Korean Society of Anesthesiologists.

Developed in M2PI

Close layer
prev next