Liver Transplantation in Singapore: Challenges and Strategies of Low- to Mid-volume Centers

LEGISLATION FOR ORGAN TRANSPLANTATION

Singapore, located in Southeast Asia, has a developed economy, a multiethnic population of 5.7 million, and spends 2.1% of its gross domestic product on health care. The Medical Therapy, Education and Research Act passed in 1972 allows people to pledge their organs for transplant, research, or education upon death. A low sign-up rate despite an extensive campaign led to the Human Organ Transplant Act in 1987, essentially an opt-out system that presumes consent to procure organs for transplantation upon death. Initially, Human Organ Transplant Act applied only to traumatic causes of death among Singapore permanent residents and citizens (21–60 y), non-Muslims, restricted to the procurement kidneys. This Act was amended in 2004 to include all causes of death and retrieval of heart, liver, cornea while also covering the regulation of living donor transplants. The amendment in 2008 included Muslims, and in 2009, the upper age limit was removed and paired matching was permitted.¹

OVERALL LIVER TRANSPLANT ACTIVITY

The first deceased donor liver transplantation (DDLT) in Singapore was performed in 1990 and the first living donor liver transplant (LDLT) in 1996 at the National University Hospital (NUH) Singapore.

Since then, >500 LDLT and DDLT have been completed in Singapore. At present, there are 4 liver transplantation centers in Singapore, with 2 centers in the public sector—National University Centre for Organ Transplantation at NUH and the SingHealth Transplant Center at the Singapore General Hospital that was commissioned for liver transplants in 2005; 2 additional centers are in the private sector. National University Centre for Organ Transplantation accounted for the majority of liver transplants in Singapore. Figure 1 shows the timelines of key events including liver transplant volume, number of actualized deceased solid donors (defined as those referred for donation after conformation of brain death), liver procured from deceased donors (defined as percent of actualized donors), and removal from liver transplant waitlist (defined as percent of patients removed from the waitlist because death or disease progression). Despite efforts in revising legislations, donation rates have been relatively static, whereas utilization rate has improved dramatically.

FIGURE 1.: Liver transplantation in Singapore. DDLT, deceased donor liver transplantation; HOTA, Human Organ Transplant Act; LDLT, living donor liver transplantation.

LDLT TO SUPPORT SHORTAGE OF DECEASED DONORS—WAYS TO ADDRESS THE LOW-VOLUME CHALLENGE

Singapore remains to have very low deceased donation rates at around 6 per million population per year, with “only” 2.4% of the population listed in the Objector’s Register as being opposed to organ donation. Possible reasons for the low donation rate in Singapore may be linked to not moving ahead with terminating life support on the basis of poor prognosis if there are concerns on family objection²; a large discrepancy in donor referral and actualization among hospitals,³ an insufficient awareness of organ donation, in addition to conflicting cultural and religious beliefs may represent additional reasons. An organ procurement case in 2007, during which organs were retrieved from a brain-dead donor, despite strong objections from the next of kin, not only made national news but also captured international attention. Such publicity is counterproductive with a detrimental impact on donation rates in addition to supporting pushback of intensive care unit and other medical personnel in supporting organ donation.⁴ Of note, those opposed to organ donation listed in the “Objector’s Register” increased from 2526 in 2006 to 9326 in 2007 and 34 494 in 2008 and only started dropping thereafter with 1778 individuals listed in 2019.

LDLT continues to play a major role in overcoming the deceased donor shortage in Singapore. Although discussions on LDLT are usually initiated early in the evaluation process, patients continue to hope for a deceased donor organ, and often, LDLT becomes a default option when the recipient’s condition worsens, which increases the ethical dilemma of balancing outcomes with donor risk. Despite LDLT flourishing in Asia, the majority of living donor hepatectomies have traditionally been limited to high-volume centers. Given the complexity of live donor hepatectomy, it is expected that the volume–outcome relation will favor higher volume centers.⁵ At the same time, mid- to low-volume centers will continue to play a role due to geographic constraints, population density, country demographics, political, economic, and local healthcare policies.

RESULTS OF LDLT IN NUH

Donor Outcomes

Two hundred four living donor hepatectomies were performed by December 2019 at NUH, of which 107 (52.5%) were from male donors. The mean age and body mass index of donors were 35.6 ± 8.5 and 24.5 ± 4.2 kg/m², respectively. The majority of donations were from parents to their children (68.6%). 14.2%, 30.9%, 15.2%, and 39.7% of donations were extended right lobe, right lobe, left lobe, and left lateral section, respectively. A total of 12 minimally invasive donor hepatectomies were performed at NUH until 2019.

The first 102 cases were performed from June 1996 to May 2014 (period 1); the subsequent 102 cases were performed in a much shorter period from June 2014 to December 2019 (period 2). Right lobe donations increased from 28.4% to 61.7%, reflecting a proportional increase of adult LDLT. There were 12 minimally invasive hepatectomies in period 2. There was a decrease in total operative time (481 ± 20.7 to 423 ± 71.8, P < 0.001), intraoperative blood loss (566 ± 417 to 350 ± 221, P < 0.001), and median hospital stay (6.5 ± 2.5 to 5.3 ± 1.5, P < 0.001), which is expected with increasing experience. The series included altruistic 13 donors following appeals in social media.

Our overall morbidity rate within the same admission at 3 mo and 6 mo was 3.9% (8), 9.3% 1(9), and 9.8% (20), respectively. Our overall major morbidity rate within the same admission at 3 mo and 6 mo was 0.5% (1), 1.5% (3), and 1.5% (3), respectively. Of the 3 patients with Clavien-Dindo grade 3b complications, 1 patient had biliary leakage, which resulted in abdominal biloma requiring percutaneous drainage and loculated thoracic collection that required video-assisted thoracoscopic surgery, another patient had a biliary leakage, which resulted in subphrenic biloma requiring computed tomography–guided drainage, and finally, a donor with portal vein narrowing detected by ultrasound Doppler in the recovery room required surgical repair. There has been no donor mortality at NUH.

Recipient Outcomes

Hepatitis B cirrhosis (34.4%) and HCC (41%) were the most common indications in adults. Eleven percent LDLTs were done for acute liver failure and 20% for acute on chronic liver failure. Sixty-three percent recipients were males, the mean age was 55.0 ± 11.7 y, mean body mass index was 25.4 ± 4.7, and the mean model for end-stage liver disease score was 19.2% ± 9.2. 3.3% adults required retransplantation. The most common recipient surgical complications were biliary strictures/leaks (34%), bleeding requiring reoperation (11.4%), hepatic artery thrombosis (7%), outflow obstruction (5%), and portal vein thrombosis (4%). Overall survival of adult recipients at 1, 5, 10 y post-LDLT was 88%, 83%, and 82%, respectively. In comparison, the overall survival rate of adult recipients by 1, 5, and 10 y post-DDLT was 87%, 78%, and 74%, respectively.

The most common indication for pediatric liver transplant was biliary atresia (65%), metabolic disorders (9%), and acute liver failure (7%). Most frequent complications included biliary stricture/leak (15.5%), portal vein stenosis/thrombosis (8.9%), hepatic artery thrombosis (2.9%), and hepatic vein stenosis (2.9%). The overall survival rate of pediatric recipients by 1, 5, and 10 y post-LDLT was 93%, 83%, and 83%, respectively.

CHALLENGES OF LOW- TO MID-VOLUME CENTERS

Problems and biases associated with measuring transplant outcomes may be exaggerated in low- and mid-volume centers. On the one hand, there may be a tendency of declining high-risk grafts or offers for high-risk recipients who may then die while waiting. On the other hand, with competition for grafts and patients across centers, there will be a higher pressure for risk-taking behavior. Nevertheless, competition (for patient referrals and grafts) among low-volume centers may result in a “zero-sum game”: a gain for one center will be a loss for another. On the flip side, transplant activities may continue despite suboptimal outcomes linked to inherent challenges in establishing a transplant center and avoiding to lose capital.

Opportunities for training and adapting to minimally invasive and robotic surgery in addition to other technologies represent additional challenges with limited volumes. Our approach shows an overall donor morbidity rate of 10.3% by 6 mo that compares favorably with other larger volume transplant centers and reflects the involvement of a core group of experts with the advantages of implementation of a robust quality assurance framework to address the volume–outcome barrier^6-10 (Table 1). Although dedicated moderate-sized centers can achieve comparable results to large centers for LDLT, this fact should not detract from the goal to centralize LDLT, notwithstanding geographic constraints (country population, cost, and convenience of travel), financial considerations, or local health policies.

TABLE 1. - Donor outcomes in NUH compared with high-volume centers

Parameters	NUH	Benchmark study	Korean Organ Transplantation Registry study 7	Toronto program: (largest western world experience) 8	Italian multicentre survey 6	A2ALL study 9
	Outcomes	Benchmark outcomes determined at 75th percentile of study outcomes 10				Retrospective era	Prospective era
Follow-up period	Within hospital stay/after 3 mo/after 6 mo	Within hospital stay/after 3 mo/after 6 mo	Median, 19 mo (range, 10–31)	Median, 39 mo (IQR, 12–72)	Mean 112 mo (range, 6–169)	Median 40.8 mo (range, 0–124.8)	Median 21.6 y (range, 0.12–82.8)
Length of hospital stay (d)	Mean 5.9 ± SD 2.1	Mean 11.7 ± SD 5.3	NA	Median 6 (IQR, 6–7)	Median 8 (range, 3–45)	NA
Overall complication rate (%)	3.9/9.3/9.8	26.9/31.2/31.2	9.3	23.1	33.3	40.0
Major complication rate (%)	0.5/1.5/1.5	6.0/8.1/9.2	1.9	9.9	12.6	1.1
Minor complication rate (%)	3.4/7.8/7.8	18.9/22.6/22.6	7.4	13.2	20.7	38.9
Median comprehensive complication index	20.9/20.9/20.9	27.9/32.6/32.7	NA	NA	NA	NA

IQR, interquartile range; NA, not available; NUH, National University Hospital.

CONCLUSION

The first successful liver transplantation has been performed 30 y back in Singapore. Today, through our nationalized liver transplant program, we have been able to achieve outcomes comparable with international high-volume centers. This experience may provide valuable information for other developing liver transplant programs worldwide facing shortages of deceased liver donations and a low to moderate transplant volume.

ACKNOWLEDGMENTS

The authors would like to thank Ms Sally Kong from National Organ Transplant Unit, Singapore, for providing the Annual reports on organ transplantation.