Optimized allocation of donor organs is a matter of concern for patients on the waiting list for deceased donor liver transplantation (DDLT). In June 2016, a new Korean liver organ allocation system was implemented, using the model for end-stage liver disease (MELD) score [1-4]. This system includes five categories, with status 1 indicating acute liver failure and early graft failure and status 2–5 assigned according to MELD scores [5]. Although this approach has been successfully applied in patients awaiting DDLT, but the MELD score cutoff for organ allocation is very high, primarily due to a serious shortage of organ donors.

The MELD score was originally developed to assess the prognosis of end-stage liver disease patients receiving best supportive care, and was subsequently adopted as a reliable predictor of waiting-list mortality [6,7]. It also provides an indication of survival outcomes after liver transplantation (LT) [8-13]. The upper limit of the MELD score is intentionally set at a cutoff of 40 for organ allocation in both the United Network for Organ Sharing and the Korean Network for Organ Sharing (KONOS). However, some patients on the waiting list would progressively deteriorate beyond a MELD score of 40, particularly as a result of organ shortages [13].

This study aimed to determine whether a MELD score exceeding 40 has any additional prognostic impact on the posttransplant outcomes. The incidence and prognosis of DDLT recipients with MELD scores >40 were determined through the analysis of the recent DDLT data from a high-volume LT center.

Study Design

This study was a retrospective analysis of DDLT data from the Asan Medical Center. To assess the posttransplant prognosis of patients with high MELD scores, three study groups were defined: KONOS status 3 (MELD score 31–37), KONOS status 2 (MELD score 38–40), and MELD-over-40 (MELD >40) groups, without application of a cutoff at 40. To reflect the real-world situation for DDLT in Korea, the study period for patient selection was set as 42 months between June 2016 and November 2019. To avoid unnecessary bias from patient selection, this study included only primary DDLT cases and excluded retransplantation cases who had undergone primary DDLT or living donor liver transplantation (LDLT) before the study period or primary LDLT during the study period. The study protocol was approved by the institutional review board (IRB No. 2019-1347).

Korean MELD Score-Based Liver Allocation System

The Korean MELD score-based liver allocation system is based on the following original formula: (9.57×loge [creatinine, mg/dL]+3.78×loge [total bilirubin, mg/dL]+11.2×loge [INR]+6.43). There are five KONOS status categories: status 1 (acute liver failure and early graft failure), status 2 (MELD score 38–40), status 3 (MELD score 31–37), status 4 (MELD score 21–30), and status 5 (MELD score ≤20) [4,5]. A cutoff of 40 in the MELD score was not applied to identify patients with MELD >40 in this study.

Statistical Analysis

Numerical data are presented as mean±standard deviation. Continuous variables were compared using Student’s t-test. The incidence variables were compared using the chi-square test and Fisher’s exact test. Survival rates were estimated using the Kaplan–Meier method and compared using a log-rank test. A p-value <0.05 was considered statistically significant. Statistical analyses were performed using SPSS version 22 (IBM, New York, NY).

Patient Grouping and Profiles

During the study period, a total of 168 DDLT operations were performed in 160 patients aged ≥18 years with MELD scores ≥31. Retransplantation after primary DDLT during the study period was performed in eight cases (5.0%). The MELD scores ranged from 31–57 with a mean value of 37.7±3.8. Patients were categorized as KONOS status 3 (n=77, 48.1%), KONOS status 2 (n=65, 40.6%), and MELD >40 (n=18, 11.3%). The characteristics of the three groups are summarized in Table 1.

Table 1 .. Comparison of patient profiles.

Group MELD score range	All patients (≥31)	KONOS status 3 (31-37) A	KONOS status 2 (38-40) B	MELD-over-40 (≥41) C	p-value
					A vs. C	B vs. C	A+B vs. C
Patient No.	160	65	77	18
Recipient sex (Male:Female)	119:41	48:17	56:21	15:3	0.27	0.40	0.41
Recipient age (years)	53.2±11.1	53.1±11.5	54.1±10.3	49.8±13.1	0.29	0.13	0.30
Primary disease					0.63*	0.38*	0.46*
HBV-LC	57 (35.6%)	22 (33.8%)	30 (39.0%)	5 (27.8%)
HCV-LC	2 (1.3%)	0	1 (1.3%)	1 (5.6%)
ALD	78 (48.8%)	30 (46.2%)	39 (50.6%)	9 (50.0%)
Others	23 (14.4%)	13 (20.0%)	7 (9.1%)	3 (16.7%)
Recipient ABO blood group					0.22**	0.53**	0.32**
A	61 (38.1%)	26 (40.0%)	29 (37.7%)	6 (33.3%)
B	44 (27.5%)	20 (30.8%)	20 (26.0%)	4 (22.2%)
O	33 (20.6%)	8 (12.3%)	20 (26.0%)	5 (27.8%)
AB	22 (13.8%)	11 (16.9%)	8 (10.4%)	3 (16.7%)
Preoperative laboratory finding
Total bilirubin (mg/dL)	28.3±12.1	25.2±12.5	29.7±11.9	33.7±8.4	0.008	0.18	0.008
Serum creatinine (mg/dL)	2.47±1.52	1.89±1.13	2.74±1.67	3.37±1.34	<0.001	0.14	<0.001
Prothrombin time (INR)	4.3±6.4	4.1±6.3	4.3±7.1	4.6±2.5	0.31	0.32	0.31
Actual MELD score	37.7±3.8	34.0±1.8	39.5±0.7	43.7±3.8	<0.001	<0.001	<0.001
Pretransplant ventilator support	30 (18.8%)	6 (9.2%)	19 (24.7%)	5 (27.8%)	0.040	0.78	0.30
Pretransplant renal replacement	69 (43.1%)	16 (24.6%)	41 (53.2%)	12 (66.7%)	<0.001	0.30	0.032
HCC in explant liver	36 (22.5%)	17 (26.2%)	18 (23.4%)	1 (5.6%)	0.061	0.089	0.067
Donor sex (Male:Female)	99:61	43:22	44:33	12:6	0.97	0.46	0.66
Donor age (years)	46.3±15.8	45.8±16.2	46.4±15.6	48.0±15.9	0.62	0.69	0.62
Graft type					NA	NA	NA
Whole liver	154 (96.3%)	60 (92.3%)	76 (98.7%)	18 (100%)
Split right liver	6 (3.8%)	5 (7.7%)	1 (1.3%)	0
Graft weight (g)	1484.1±334.9	1526.1±378.1	1459.1±310.5	1439.5±261.9	0.37	0.81	0.37
Graft-recipient weight ratio	2.32±0.68	2.37±0.73	2.30±0.64	2.19±0.61	0.34	0.53	0.42
Donor anti-HBc IgG	33 (20.6%)	11 (16.9%)	19 (24.7%)	3 (16.7%)	0.57	0.47	0.66
Ischemic time (min)
Cold	317.2±114.1	311.4±119.4	325.0±116.3	305.2±84.5	0.84	0.50	0.84
Warm	56.6±88.4	65.7±137.3	47.3±16.5	45.8±9.1	0.54	0.71	0.54
Total	371.4±160.2	377.1±206.3	372.3±124.9	346.9±97.5	0.55	0.42	0.55
Retransplantation					0.76	0.52	0.91
Early (<3months)	4 (2.5%)	3 (4.6%)	1 (1.3%)	0
Late (>3 months)	4 (2.5%)	2 (3.1%)	1 (1.3%)	1 (5.6%)

Data are expressed as number with percentage or mean value with standard deviation..

MELD score, model for end-stage liver disease score; KONOS, Korean Network for Organ Sharing; HBV-LC, hepatitis B virus-associated liver cirrhosis; HCV-LC, hepatitis C virus-associated liver cirrhosis; ALD, alcoholic liver disease; INR, international normalization ratio; HCC, hepatocellular carcinoma; anti-HBc IgG, hepatitis B virus core antibody immunoglobulin G; NA, not available..

*Comparison between HBV-LC and other groups..

**Comparison between blood group A and B versus O and AB..

Comparison of Posttransplant Survival

The graft survival rates of the primary DDLT were 84.0% at 1 year, 77.0% at 2 years and 70.7% at 3 years (Fig. 1A). The causes of retransplantation were early graft dysfunction or primary non-function (n=4) and chronic graft failure (n=4).

Figure 1. Patient survival (A) and graft survival (B) curves after liver transplantation.

The overall patient survival rates were 85.2% at 1 year, 77.1% at 2 years and 70.7% at 3 years (Fig. 1B). The causes of patient death were sepsis (n=19), hepatocellular carcinoma recurrence (n=5), various brain diseases including brain death (n=4), chronic graft failure (n=2), and hepatic artery occlusion, intractable bleeding, bowel infarct, sudden cardiac arrest, and unknown cause (all n=1).

The 1-, 2-, and 3-year graft survival rates of the primary DDLT were 85.9%, 76.9% and 74.2% in the KONOS status 3 group; 83.9%, 81.4% and 74.6% in the KONOS status 2 group; and 77.8%, 66.2% and 52.7% in the MELD-over-40 group, respectively (p=0.23 among all three groups; and p=0.089 between the status 2 versus MELD-over-40 group, Fig. 2A).

Figure 2. Comparison of the patient survival (A) and graft survival (B) curves according to the model for end-stage liver disease (MELD) scores.

The 1-, 2- and 3-year overall patient survival rates were 87.2%, 77.1% and 74.4% in the KONOS status 3 group; 85.1%, 82.6% and 75.7% in the KONOS status 2 group; and 77.8%, 66.2% and 52.7% in the MELD-over-40 group, respectively (p=0.19 among all three groups; p=0.78 between the status 2 versus status 3 group; and p=0.048 between the status 2 versus MELD-over-40 group).

Risk Factor Analysis for Patient Survival

The primary diseases for DDLT were categorized as hepatitis B virus-associated liver cirrhosis (n=57 [35.6%]), hepatitis C virus-associated liver disease (n=2 [1.3%]), alcoholic liver disease (n=78 [48.8%]), and subacute liver failure (n=5 [3.1%]) and other diseases (n=18 [11.3%]). Hepatitis C virus-associated liver disease was associated with inferior outcomes (p=0.001, Fig. 3A), but no statistically significant differences in survival were seen in the other four disease groups (p=0.61).

Figure 3. Comparison of the patient survival curves according to the primary liver diseases (A), pretransplant ventilator support (B), pretransplant renal replacement therapy (C), and retransplantation (D). ALD, alcoholic liver disease; SALF, subacute liver failure; HBV-LC, hepatitis B virus-associated liver cirrhosis; HCV-LC, hepatitis C virus-associated liver disease.

A requirement for pretransplant ventilator support (n=30 [18.8%]) was associated with significantly inferior patient survival outcomes (p=0.043, Fig. 3B), but pretransplant renal replacement therapy (RRT; n=69 [43.1%]) showed no prognostic difference (p=0.58, Fig. 3C). Retransplantation (n=8 [5.0%]) showed a highly significant prognostic difference (p<0.001, Fig. 3D).

Although multivariate analysis for overall patient survival showed that pretransplant ventilator support and retransplantation were significant prognostic factors, a MELD score >40 per se was not seen to be an independent risk factor (Table 2).

Table 2 .. Results of univariate and multivariate analyses for overall patient survival.

		Case No.	Univariate analysis			Multivariate analysis
			1-year patient survival rate	p-value		Hazard ratio	95% confidence interval	p-value
Pretransplant ventilatory support	No	30	87.1%			1
	Yes	130	76.7%	0.043		2.11	0.99–4.46	0.48
Retransplantation	No	152	88.0%			1
	Yes	8	33.3%	<0.001		7.87	3.22–19.2	<0.001
MELD score >40	No	142	86.0%			1
	Yes	18	77.8%	0.19		1.93	0.78–4.72	0.15

MELD, model for end-stage liver disease..

The high demand for DDLT in Korea is attributed to the prevalence of hepatitis B virus-associated liver diseases and a recent increase in the incidence of alcoholic liver diseases [2-7,14-16]. The annual number of deceased organ donors temporarily exceeded 10 per million in 2016 [1], but has since decreased because of certain medical and social issues, including the Life Insurance Decision Act regulating termination of life-sustaining treatment. As a result, the annual numbers of deceased organ donors and those undergoing DDLT in Korea decreased from 573 and 508 in 2016 to 515 and 450 in 2017, and 449 and 369 in 2018, respectively [16].

In the past 5 years, the total annual number of LT cases in Korea remained largely unchanged, suggesting a reciprocal relationship between number of DDLT and LDLT operations [2,3,17]. After the introduction of the Korean MELD score-based allocation system, the LT candidate pool awaiting DDLT with very high MELD scores has been expanding because the number of deceased organ donors cannot meet the demand. Of the DDLT recipients with a MELD score ≥31, half were patients with alcoholic liver disease, as reflected in the current study. In reality, these cases involving alcohol abuse were usually neglected by the patients themselves and family members until reaching the terminal stage of liver disease and finally being enrolled on the DDLT waiting list. The outcome in these patients would be death whilst waiting for LT or DDLT after a progressive rise in their MELD score [18].

Our previous study demonstrated that the Korean MELD score-based liver allocation system was greatly influenced by a serious shortage of deceased donors [1,19]. The mean MELD score of DDLT recipients exceeded 36, suggesting that patients with very high MELD scores could be allocated to DDLT due to a serious shortage of donor organs. Prior to the adoption of the MELD score-based allocation system, the new KONOS status 2 and 3 had been expected to account for approximately 50% of DDLT cases. However, the proportion of status 2 and 3 patients was seen to be much greater than anticipated in recent real-world practice.

In Korea, the majority of adult DDLT recipients are enrolled at KONOS status 2 and a considerable proportion of these patients will progress to MELD scores >40, as seen in the current study. Although the posttransplant survival outcomes of the MELD-over-40 group were inferior to those of the KONOS status 2 and 3 groups, very high MELD scores per se were not seen to be an independent risk factor for patient survival. However, the requirement for pretransplant ventilator support and transplantation of marginal-quality liver grafts requiring retransplantation were independent prognostic factors. This is consistent with the results of an international multicenter study, which reported that mechanical ventilation and use of marginal organs were associated with an increased risk of patient death [10]. It is generally recommended that ventilator support be initiated early if it is indicated during the LT waiting period because concurrent lung problems are known to increase posttransplant mortality. It is also reasonable to avoid marginal grafts for DDLT candidates with very high MELD scores, although serious shortage of deceased organ donors in Korea does not permit such prudent selection as any delay in DDLT surgery will increase the risk of waiting-list mortality.

Organ donor shortages also persist in the United States, particularly in regions where competition is highest and DDLT recipients frequently attain MELD scores of 40 or higher prior to transplantation. In a single-center retrospective analysis of 207 LT recipients who had MELD scores ≥40, implementation of Share 35 policy was identified as the strongest predictor of graft survival in LT recipients with MELD score ≥40, and recipients with diabetes at the time of listing was the strongest predictor of posttransplant patient mortality. However, almost 40% of patients remained hospitalized for more than 4 weeks after transplantation surgery, and 20% were discharged to an acute care facility [13]. An analysis of the Organ Procurement and Transplantation Network/UNOS database showed that Share 35 policy has had a positive impact on survival after transplantation in patients with MELD ≥35 [20]. In Korea, there are three regions in practice, but they are not separate independently and the concept of Share 35 policy has already been applied.

Uniquely, the requirement for pretransplant RRT did not worsen the posttransplant course in the current study. We have previously analyzed in-hospital mortality in relation to the pre- and posttransplant application of RRT. This study showed no incidence of in-hospital mortality in the pretransplant RRT group, but the incidence increased to 20.8% in patients who underwent de novo posttransplant RRT. We suggested that pretransplant RRT appears to improve posttransplant survival outcomes and therefore recommend that pretransplant RRT be performed if indicated while awaiting DDLT [18].

The current study has some limitations of note. This is a single-center study with a relatively small number of patients, which could potentially introduce selection bias. It was also a retrospective study and the MELD cutoff was arbitrarily set as exceeding 40. High-volume multicenter studies will be required to validate the results of this study.

In conclusion, the results of this study revealed that very high MELD score (>40) appear to confer additional risk in KONOS status 2 patients, although this was not an independent prognostic factor.

The study received no financial support.

All authors have no conflicts of interest to declare.

Conceptualization: SH. Data curation: GCP, GWS, DHJ, TYH, CSA, DBM, KHK, YIY. Methodology: WHK, HDC, MK, SHK, SGL. Visualization: SH. Writing - original draft: SH, BGN. Writing - review & editing: SH.