Renal impairment (RI) is a common and life-threatening complication of cirrhosis, affecting one in every five patients [1]. While liver transplantation (LT) offers a critical lifeline for patients with cirrhosis who have impaired renal function, postoperative acute RI frequently occurs and is influenced by multiple factors [2], resulting in poor outcomes [3].

Studies aimed at identifying risk factors for postoperative RI have found that these include female sex, body mass index (BMI), Child-Pugh score, diabetes, the number of units of blood or fresh frozen plasma (FFP) transfused during the perioperative period, and nonalcoholic steatohepatitis as the etiology of end-stage liver disease [4]. Moreover, some studies have suggested that preoperative renal dysfunction and a high Model for End-Stage Liver Disease (MELD) score increase the risk for the development of post-LT RI [5,6].

Although research has been conducted on this topic, the risk factors for RI following LT are still not fully understood. Mongolia has the highest incidence rate of hepatocellular carcinoma globally, at 54.1 cases per 100,000 population, primarily due to the high prevalence of hepatitis B or C virus infections [7,8]. Therefore, the number of patients who need LT remains high. In Mongolia, the first LT was performed at the First Central Hospital of Mongolia (FCHM) in September 2011 in collaboration with South Korea’s Asan Medical Center. LTs continue to be performed by the national team. However, there is an increasing number of patients on the waiting lists due to a shortage of donor organs.

Therefore, identifying risk factors for postoperative RI is crucial for developing preventive and treatment strategies to improve the outcomes of liver transplant recipients. Proper management of RI can significantly improve patient survival. This study aimed to investigate the incidence of postoperative RI and the factors contributing to RI following LT in our patient cohort.

This retrospective, single-center study was approved by the Ethical Committee of the Mongolian National University of Medical Sciences (No. 2021/3-01). Informed consent was waived due to the retrospective nature of the study.

Patients

The study involved collecting data from all patients who underwent LT at FCHM from September 2011 to December 2022. The patients were categorized into two groups based on their postoperative renal function, as determined by the glomerular filtration rate (GFR) measured post-LT: those with RI and those with no RI (NRI). Patients under 18 years of age or with incomplete data for calculating the estimated GFR (eGFR) were excluded.

Clinical and Laboratory Data

Pre-LT data included routine recipient demographic variables such as age, sex, ABO blood type, body surface area (BSA), BMI, comorbidities, Child-Turcotte-Pugh (CTP) and MELD with sodium (MELD-Na) scores, diagnosis, waiting list duration, previous LT, ABO compatibility, bridge therapies, and preoperative RI. Perioperative data encompassed graft type and weight, graft-to-recipient weight ratio (GRWR), percentage of graft fatty change, inferior vena cava (IVC) cross-clamping, operation duration, cold and warm ischemic times, ascites volume, portosystemic shunts, perioperative continuous renal replacement therapy (CRRT), transfusion of blood products (leukocyte-reduced red blood cells [LR-RBC], platelets [PLT], FFP, and cryoprecipitate), and hemodynamic variables including heart rate, mean arterial pressure (MAP), the difference between baseline and lowest MAP, lowest central venous pressure (CVP), and urine output. The postoperative data included early complications within the first month post-LT, immunosuppression maintenance regime, transfusion of blood products (LR-RBC, PLT, FFP, and cryoprecipitate), intubation duration, and total days of hospital stay. Renal function was assessed by measuring the GFR using the Cockcroft-Gault creatinine clearance formula at pretransplantation, and at 24 hours, 72 hours, 7 days, 14 days, and 28 days post-LT. RI was defined as an eGFR <60 mL/min at any point during the first 28 days after LT.

Statistical Analysis

We performed a statistical analysis using STATA ver. 15.0 (StataCorp). To assess the normality of the data, we employed both Shapiro-Wilk test and a histogram. Continuous parametric variables were presented as means with standard deviations and compared using the Student t-test. Continuous nonparametric variables were presented as medians with interquartile ranges and compared using the Mann-Whitney U-test. Categorical variables were expressed as frequencies and percentages, and differences were evaluated using the chi-square and Fisher exact tests. Logistic regression was used to explore the association between independent and dependent variables in patients with and without RI. Survival rates at 3 years were calculated using the Kaplan-Meier method and compared across groups with renal dysfunction using log-rank tests. All statistical tests were two-sided, and a P-value of less than 0.05 was considered statistically significant.

Recipients’ Baseline Characteristics

The study included 187 patients who had undergone LT at the FCHM. Of these, 181 patients (94 males and 87 females) met the inclusion criteria for the study, and 57 of them developed RI post-LT. The average age of the participants at the time of LT was 44±11 years. Recipients who developed RI constituted 31.5% of the total. Those in the RI group were significantly older (P<0.001) and more likely to be female compared to those in the NRI group (P=0.001). The table also presents detailed preoperative characteristics of the recipients. There was a statistically significant difference in BSA between the two groups (P=0.009), but no significant difference in BMI. According to the CTP scoring system, recipients with RI were significantly more likely to be categorized into group C than those with NRI (P=0.008). Additionally, recipients in the RI group had substantially higher MELD scores at the time of transplantation (P<0.001) (Table 1).

Table 1. Baseline characteristics of recipients

Variable	Total (n=181)	RI (n=57)	NRI (n=124)	P-valuea)
Age (yr)	44±11	49±10	42±10	<0.001b)
Sex				0.001
Male	94 (51.9)	19 (33.3)	75 (60.5)
Female	87 (48.1)	38 (66.7)	49 (39.5)
BMI (kg/m2)	25.2 (24.4–25.8)	25.2 (21.6–28.6)	25.1 (21.8–30.1)	0.711c)
BSA (m2)	1.8 (1.7–1.8)	1.7 (1.6–1.9)	1.8 (1.6–2.0)	0.009c)
Recipient blood type				0.430d)
O+	63 (34.8)	17 (29.8)	46 (37.1)
A+	48 (26.5)	13 (22.8)	35 (28.2)
B+	61 (33.7)	24 (42.1)	37 (29.8)
AB+	9 (5.0)	3 (5.3)	6 (4.8)
Child-Pugh score				0.008
A	23 (12.7)	2 (3.5)	21 (16.9)
B	87 (48.1)	25 (43.9)	62 (50.0)
C	71 (39.2)	30 (52.6)	41 (33.1)
MELD score	15 (14–16)	17 (12–23)	14 (10–20)	<0.001c)
Diagnosis				0.137d)
Liver cirrhosis	114 (63.0)	37 (64.9)	77 (62.1)
Hepatocellular carcinoma	58 (32.0)	17 (29.8)	41 (33.1)
Primary biliary cirrhosis	7 (3.9)	1 (1.8)	6 (4.8)
Secondary biliary cirrhosis	2 (1.1)	2 (3.5)	0
Comorbidities
Hypertension	16 (8.8)	2 (3.5)	14 (11.3)	0.980
Diabetes mellitus	16 (8.8)	4 (7.0)	12 (9.7)	0.270
Tuberculosis	7 (3.9)	1 (1.8)	6 (4.8)	0.070d)
Urinary tract infection	9 (5.0)	1 (1.8)	8 (6.5)	0.020d)
Preoperative renal dysfunction	8 (4.4)	8 (14.0)	0	<0.001d)
Waiting days	140 (122–186)	116 (42–346)	184 (70–319)	0.024c)
Previous liver transplantation	3 (1.6)	3 (5.3)	0	0.030
ABO incompatibility	3 (1.6)	1 (1.8)	2 (1.6)	0.940
Preoperative CRRT	3 (1.6)	3 (5.3)	0	0.030d)
Preoperative liver resection	4 (2.2)	0	4 (3.2)	0.310d)
Preoperative downstaging	42 (23.2)	11 (19.3)	31 (25.0)	0.390
Preoperative esophageal variceal ligation	49 (27.1)	13 (22.8)	36 (29.0)	0.380

Values are presented as mean±standard deviation, number (%), or median (interquartile range).

RI, renal impairment; NRI, non-RI; BMI, body mass index; BSA, body surface area; MELD, Model for End-Stage Liver Disease; CRRT, continuous renal replacement therapy.

^a)Chi-square test; ^b)Student t-test; ^c)Mann-Whitney U-test; ^d)Fisher exact test.

Patients diagnosed with liver cirrhosis comprised over 60% of the RI group; however, there was no significant difference in the overall distribution of reasons for LT. In the NRI group, there were no cases of previous LT, whereas three recipients in the RI group had a history of undergoing LT previously (P<0.05). In the RI group, three recipients (5.3%) underwent CRRT during the preoperative period, a significantly higher proportion than in the NRI group (P=0.03).

Comparison of Perioperative Parameters

The perioperative parameters are presented in Table 2. The distribution of graft types was similar across both groups, with the right-sided graft being the most common among recipients (P>0.05). The graft weight was significantly higher in the RI group than in the NRI group (P=0.026). The median GRWR was 0.9 in the NRI group and 1.1 in the RI group, showing a significant difference (P<0.001). There was no significant difference in the incidence of IVC clamping between the RI and NRI groups. The median operation time was 852 minutes for the NRI group and 872 minutes for the RI group (P=0.846). The average total ischemic time was 190 minutes in the NRI group and 195 minutes in the RI group, which was not a statistically significant difference (P=0.983). The median volume of ascites was 150 mL in the NRI group and 800 mL in the RI group; this difference reached statistical significance (P<0.001).

Table 2. Comparison of perioperative parameters

Variable	Total (n=181)	RI (n=57)	NRI (n=124)	P-valuea)
Graft type				0.670b)
Right	167 (92.3)	51 (89.5)	116 (93.5)
Left	2 (1.1)	1 (1.8)	1 (0.8)
Whole	12 (6.6)	5 (8.8)	7 (5.6)
Graft weight (g)	728 (686–756)	760 (640–898)	694 (583–859)	0.026c)
GRWR (%)	1.03 (0.99–1.07)	1.1 (0.9–1.4)	0.9 (0.8–1.24)	<0.001c)
Graft liver fatty change	5 (5–10)	5 (5–20)	5 (5–15)	0.521c)
Porto-systemic shunt	85 (47.0)	22 (38.6)	63 (50.8)	0.150
IVC clamp				0.520
Partial	169 (93.4)	52 (91.2)	117 (94.4)
Total	12 (6.6)	5 (8.8)	7 (5.6)
Bio-pump usage	6 (3.3)	3 (5.3)	3 (2.4)	0.380b)
Operation time (recipient)	859 (834–886)	872 (735–1,065)	852 (758–986)	0.846c)
Cold ischemic time (min)	126±62	132±77	123±55	0.370d)
Warm ischemic time (min)	80±23	79±22	82±23	0.370d)
Total ischemia time (min)	193 (182–207)	195 (157–248)	190 (162–249)	0.983c)
Ascites (mL)	200 (150–400)	800 (950–4,500)	150 (160–1,500)	0.003c)
Intraoperative LR-RBC (unit)	6 (6–8)	8 (4–18)	5 (2–10)	<0.001c)
Intraoperative PLT (unit)	5 (5–10)	5 (5–14)	5 (4–15)	0.018c)
Intraoperative FFP (unit)	10 (10–12)	14 (10–26)	10 (6–18)	<0.001c)
Intraoperative cryoprecipitate (unit)	6 (3–10)	6 (4–20)	4 (5–15)	0.168c)
Baseline MAP (mmHg)	76 (75–80)	80 (69–105)	76 (66–95)	0.161c)
Lowest MAP (mmHg)	53 (52–55)	53 (44–60)	53 (45–59)	0.529c)
MAP difference (mmHg)	25 (24–29)	28 (16–51)	25 (12–47)	0.225c)
Lowest CVP (mmHg)	2 (2–3)	1 (1–5)	2 (1–4)	0.848c)
Perioperative CRRT	1 (1.9)	1 (1.9)	0	0.310b)
Intraoperative urine output (mL)	2,438 (1,390–1,605)	1,385 (924–2,050)	1,523 (1,081–2,225)	0.051c)

Values are presented as number (%), median (interquartile range), or mean±standard deviation.

RI, renal impairment; NRI, non-RI; GRWR, graft-to-recipient weight ratio; IVC, inferior vena cava; LR-RBC, leukocyte-reduced red blood cells; PLT, platelet; FFP, fresh frozen plasma; MAP, mean arterial pressure; CVP, central venous pressure; CRRT, continuous renal replacement therapy.

^a)Chi-square test; ^b)Fisher exact test; ^c)Mann-Whitney U-test; ^d)Student t-test.

Recipients in the RI group received significantly more transfusions of LR-RBC, PLT, and FFP units during LT compared to those in the NRI group (P<0.001, P=0.018, and P<0.001, respectively). There was no significant difference in the intraoperative transfusion of cryoprecipitate between the two groups (P=0.168). However, the median intraoperative urine output was significantly lower in the RI group than in recipients without RI (P=0.051).

Comparison of Postoperative Outcomes

No significant difference was observed between the two groups in terms of early complications following LT, including vascular and biliary complications and rejection (Table 3). The postoperative mortality rate was 17.5% in recipients who developed RI and 9.7% in the NRI group; this difference was not statistically significant (P=0.130). Postoperatively, recipients in the RI group received significantly more units of LR-RBC and cryoprecipitate transfusions than those in the NRI group (P<0.005 and P<0.05, respectively). However, the mean units of postoperative PLT and FFP transfusions did not differ significantly between the two groups (P=0.160 and P=0.179). The median intubation time was significantly longer in the RI group compared to the NRI group (P=0.002). There was no significant difference in the length of hospital stay between the two groups (P=0.408). We assessed the impact of immunosuppression doses on postoperative renal dysfunction and found no significant differences between the two groups (P=0.250).

Table 3. Comparison of postoperative outcomes

Variable	Total (n=181)	RI (n=57)	NRI (n=124)	P-valuea)
Vascular complication	20 (11.0)	6 (10.5)	14 (11.3)	0.879
Rejection	5 (2.8)	2 (3.5)	3 (2.4)	0.678b)
Biliary complication	33 (18.2)	8 (14.0)	25 (20.2)	0.321
Early return to operating room	18 (9.9)	7 (12.3)	11 (8.9)	0.476
Postoperative mortality	22 (12.2)	10 (17.5)	12 (9.7)	0.130
Starting maintenance				0.250
Cyclosporin	8 (4.4)	4 (7.0)	4 (3.2)
Tacrolimus	173 (95.6)	53 (93.0)	120 (96.8)
Postoperative LR-RBC (unit)	3 (2–4)	4 (2–11)	2 (2–6)	0.005c)
Postoperative PLT (unit)	15 (15–20)	15 (10–30)	15 (5–35)	0.160c)
Postoperative FFP (unit)	6 (6–8)	6 (4–20)	6 (2–12)	0.179c)
Postoperative cryoprecipitate (unit)	1 (0–12)	2 (0–30)	1 (0–15)	0.020c)
Incubated hours	31 (30–32)	33 (29–53)	31 (27–35)	0.002c)
Hospital stay (day)	30 (29–32)	30 (21–44)	30 (24–39)	0.408c)

Values are presented as number (%) or median (interquartile range).

RI, renal impairment; NRI, nonrenal impairment; LR-RBC,-leukocyte-reduced red blood cells; PLT, platelet; FFP, fresh frozen plasma.

^a)Chi-square test; ^b)Fisher exact test; ^c)Mann-Whitney U-test.

Risk Factors Associated with Renal Impairment Post-liver Transplantation

We conducted logistic regression models to identify factors associated with RI following LT (Table 4). Factors that increased the risk of RI among recipients post-LT included female sex (odds ratio [OR], 3.06; 95% confidence interval [CI], 1.58–5.91), CTP scores of B and C (OR, 4.23; 95% CI, 0.92–19.41 and OR, 7.68; 95% CI, 1.67–35.30, respectively), preoperative CRRT (OR, 5.86; 95% CI, 1.1–31.21), and high GRWR (OR, 3.45; 95% CI, 1.23–9.63).

Table 4. Risk factors associated with renal impairment

Variable	cORa) (95% CI)	aORb) (95% CI)
Recipient age	1.07 (1.03–1.11)	1.07 (1.03–1.11)
Recipient sex
Male	1	1
Female	3.06 (1.58–5.91)	2.87 (1.45–5.71)
MELD score	1.09 (1.03–1.16)	1.15 (1.08–1.22)
Child-Pugh score
A	1	1
B	4.23 (0.92–19.41)	3.98 (0.82–19.39)
C	7.68 (1.67–35.30)	9.48 (1.93–46.40)
GRWR (%)	3.45 (1.23-9.63)	5.70 (1.85–17.52)
Preoperative CRRT	5.86 (1.10–31.21)	10.2 (1.46–71.59)
Intraoperative LR-RBC (unit)	1.09 (1.04–1.14)	1.09 (1.04–1.14)
Intraoperative PLT (unit)	1.06 (1.00–1.12)	1.07 (1.01–1.14)
Intraoperative FFP (unit)	1.07 (1.02–1.11)	1.09 (1.04–1.14)
Postoperative LR-RBC (unit)	1.09 (1.04–1.14)	1.10 (1.03–1.17)
Postoperative cryoprecipitate (unit)	1.10 (1.01–1.19)	1.11 (1.01–1.22)
Intubated hours	1.04 (1.01–1.07)	1.04 (1.01–1.07)

cOR, crude odds ratio; CI, confidence interval; aOR, adjusted odds ratio; MELD, Model for End-Stage Liver Disease; GRWR, graft-to-recipient weight ratio; CRRT, continuous renal replacement therapy; LR-RBC,-leukocyte-reduced red blood cells; PLT, platelet; FFP, fresh frozen plasma.

^a)Logistic regression; ^b)Adjusted for recipients’ age and sex.

Moreover, recipients’ age (OR, 1.07; 95% CI, 1.03–1.11), MELD score (OR, 1.09; 95% CI, 1.03–1.16), intraoperative LR-RBC (OR, 1.09; 95% CI, 1.04–1.14), PLT (OR, 1.06; 95% CI, 1.00–1.12), FFP (OR, 1.07; 95% CI, 1.02–1.11) and postoperative LR-RBC (OR, 1.09; 95% CI, 1.04–1.14), cryoprecipitate transfusion (OR, 1.1; 95% CI, 1.01–1.19) and intubated hours (OR, 1.04; 95% CI, 1.01–1.07) were associated with slight increases in the risk of RI following LT.

Kaplan-Meier survival analysis revealed that the 1-year and 3-year survival rates in patients with RI were 93.4% and 78.1%, respectively. However, the difference in survival rates between recipients with and without RI was not statistically significant (P=0.224) (Fig. 1).

Figure 1. Kaplan-Meier survival curve.

This is the first study to investigate the risk factors for RI after LT in patients at a single center in Mongolia. Our study found that female sex, high CTP scores, preoperative CRRT, and high GRWR were risk factors for developing RI after LT in Mongolia. However, significant differences were observed in age, sex, BSA, CTP score, MELD score, preoperative CRRT, GRWR, transfusion requirements, and duration of intubation between the RI and NRI groups.

Despite the challenges posed by varying definitions and methods of calculating RI, the literature consistently reports high rates of RI following LT [9–11]. A recent review of 67 observational studies spanning three decades indicates that the rate of early post-LT RI exceeds 50% [4,12]. Feldkamp et al. [3] reported an incidence of RI post-LT at 65.8%. In contrast, using the Cockcroft-Gault formula, our study identified a lower incidence rate of 31.5%, which reflects the specific characteristics of our cohort. This formula is considered the most accurate for determining the GFR based on a patient's sex, age, BMI, and serum creatinine level [13–15].

Known risk factors for RI after LT include a high MELD score, hypovolemia, infection, hepatorenal syndrome, reperfusion injury, and nephrotoxic drugs such as calcineurin inhibitors [6,16–18]. In line with this literature, we identified a MELD score of 20 or higher as a significant independent risk factor for RI. Additionally, our findings support the increased susceptibility of female recipients to RI following LTs [19–21]. While factors such as age, hypertension, dyslipidemia, diabetes, and high BMI were associated with a decline in renal function post-LT, body weight, surprisingly, was not.

Guitard et al. [22] reported that higher volumes of intraoperative blood transfusions, especially those exceeding 10 units, were associated with an increased risk of renal insufficiency when urine output is less than 100 mL/hr. It is believed that hypotension resulting from blood loss triggers a proinflammatory response and nephrotoxicity. Our data are consistent with these findings, underscoring the importance of blood loss as a risk factor.

Survival analysis in our cohort did not reveal a significant difference between patients with varying degrees of RI, which contrasts with studies suggesting that severe RI or the need for renal replacement therapy significantly reduces survival rates [6]. It is noteworthy that the survival rates at 1 and 3 years for our patients were higher than those reported in referenced studies, potentially influenced by our modest sample size and stringent selection criteria for both donors and recipients.

This study is limited by its reliance on data from a single center. As Mongolia is a lower-middle-income country with only two LT centers, the statistical power is constrained by the small sample size. Additionally, cultural practices have influenced our transplant program, resulting in the exclusive performance of living donor transplants.

As a country that has only recently developed LT capabilities, Mongolia is facing the challenge of incorporating advanced medical procedures into its healthcare system. The findings from this study offer important insights into the prevalent risk factors among its patient population and lay the groundwork for creating targeted strategies to reduce risks of rejection and enhance the success rates of LT. Given the existing constraints, there is a definite need for broader, multicentric research. This research is crucial for establishing a strong evidence base that will inform Mongolia’s LT and postoperative care strategies, with the goal of meeting international standards and achieving comparable outcomes.

Conflict of Interest

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

Author Contributions

Conceptualization: BB, CB. Data curation: BB, SA, CB. Formal analysis: FLV, JN, AT, SO. Visualization: BB, SA. Writing–original draft: all authors. Writing–review & editing: all authors. All authors read and approved the final manuscript.