Case volume at liver transplantation (LT) centers has been classically associated with posttransplant outcomes. Compared to low-volume centers (LVCs), high-volume centers (HVCs) are known to have better outcomes, including lower in-hospital mortality after LT [1], simultaneous liver and kidney transplantation (SLKT) [2], and other gastrointestinal surgical procedures [3]. How case volume affects posttransplant outcomes is not clear, although case volume can be associated with a protocolized approach, surgical experience, postoperative critical care, and the availability of radiologic and endoscopic expertise, all of which affect the transplant outcomes. However, some studies suggest that transplant center case volume is no longer a significant predictor of posttransplant survival in the model for end-stage liver disease (MELD) era [4].

Although the term “HVC” is used in the literature, the definition of HVC varies widely in different countries and even within different geographic regions of a country. In Korean studies, HVCs were defined as those conducting 30 to 50 transplants per year [5], and in the West the definition of 65 to 175 cases per year has been used [6]. The criteria for high- or low-volume transplant centers have not yet been defined in India, however, centers like ours that perform >100 LTs per year are considered HVCs [7].

The coronavirus disease 2019 (COVID-19) pandemic caused tremendous morbidity and mortality worldwide and reduced transplant work in India, with only the sickest patients being operated on. The volume of transplants reduced by half at our center with the onset of the COVID-19 pandemic in 2020. This unique situation gave us an opportunity to study the effect of case volume on posttransplant outcomes at our center, which may not have been possible otherwise. Our study analyzed the impact of case volume reduction during the COVID-19 pandemic on posttransplant outcomes at our center by comparing results with the previous year.

This study was approved by the Institutional Review Board of Max Super Speciality Hospital (IRB No. BHR/RS/MSSH/MHIL/SKT-1/MHEC/ANES/22-05). The requirement for informed consent was waived by the institutional ethics committee in view of retrospective nature of study.

This was a retrospective analysis of a prospectively maintained departmental database. Patients undergoing LT between April 1, 2020, and September 30, 2020 were included in the pandemic group (PG). The PG was further subdivided into the adult PG (APG; ≥18 years of age) and the pediatric PG (PPG; <18 years of age). Similarly, patients undergoing LTs between April 1, 2019, and March 31, 2020, were grouped in the pre-PG (P-PG), which was further subdivided according to age into the adult P-PG (AdP-PG) and the pediatric P-PG (PedP-PG). The data of the APG was compared with the AdP-PG and the PPG was compared with the PedP-PG. The collected data included the transplant recipients’ demographics, the details and extent of liver disease and other ailments, preoperative laboratory parameters, donor details, and details of the recipients’ preoperative intensive care unit (ICU) stay. The patients’ alcohol-related liver diseases included decompensated cirrhosis, acute-on-chronic liver failure (ACLF), and severe alcoholic hepatitis. All patients with alcohol-related liver disease underwent comprehensive assessment for posttransplant relapse of drinking behaviors as per our center’s protocol. ACLF in adult patients was defined according to the European Association for the Study of Chronic Liver Failure criteria [8]. The intraoperative details noted were graft-to-recipient weight ratio, cold and warm ischemia time, graft steatosis (fat estimation on imaging), intraoperative blood transfusion, duration of transplant surgery, and type of donor surgery (open or laparoscopic). Postoperative data included the duration of ICU stay and hospital stay; postoperative complications including biliary and vascular complications (i.e., hepatic artery thrombosis, portal vein thrombosis (PVT), middle hepatic vein/right hepatic vein thrombosis, bile leak) and sepsis; postoperative interventions performed (e.g., dialysis, plasmapheresis); and surgical, endoscopic, and radiological interventions conducted prior to discharge. The mortality rates at 30 days, 3 months, and 6 months were the primary outcomes of the study and complication rates were the secondary outcomes.

During the 6-month lockdown period, only 60 living donor LT (LDLT) procedures were performed in our unit, of which 43 were adult recipients (APG) and 17 were pediatric recipients (PPG) (Fig. 1). Five transplants were deferred during the pandemic when the donors tested positive during preoperative screening for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. Three of these patients underwent LDLT at a later date. One of the two patients with ACLF and SARS-CoV-2 infection underwent LDLT after recovery from COVID-19. In contrast, 228 LT procedures were performed in our unit during the pre-pandemic year, of which 178 were in adults (AdP-PG) and 50 were in pediatric patients (PedP-PG). Three of these were deceased donor LT (DDLT) procedures, and three involved SLKT.

The demographics, including baseline recipient and donor characteristics and the preoperative and intraoperative details of adult LT recipients (APG and AdP-PG) are depicted in Table 1, and those of the pediatric age group (PPG and PedP -PG) are depicted in Table 2. The age, sex, and body mass index (BMI) of adult and pediatric recipients were found to be comparable between the two groups (Tables 1 and 2). Foreign nationals seeking treatment decreased significantly during the lockdown due to travel restrictions (14% in APG vs. 37.1% in AdP-PG, P=0.003 and 53% in PPG vs. 72% in Ped-PG, P=0.024) (Tables 1 and 2).

The most common etiology for cirrhosis among adults was alcohol use in the APG (37.2%) and viral infection in the AdP-PG (hepatitis B and C; 28.6%). The number of recipients with comorbidities (stroke, seizure disorders, interstitial lung disease) was higher in the APG. The mean MELD score of patients in the APG was significantly higher (24.39±9.55 vs. 21.14±9.17, P=0.034) compared to the AdP-PG (Table 1). More APG patients than AdP-PG patients had complications of cirrhosis in the form of spontaneous bacterial peritonitis (SBP; 11.6% vs. 4.5%, P=0.073), PVT (8.4% vs. 2.3%, P=0.277), hepatic encephalopathy (HE; 46.5% vs. 34.27%, P=0.131), and hepatorenal syndrome (HRS; 13.95% vs. 7.3%, P=0.163), though these results were not statistically significant (Table 1). More patients with ACLF (30.2% vs. 10.2%, P=0.002) underwent transplantation in the APG, including ACLF-3 (i.e., three or more organs failing), compared to the AdP-PG (5.5% vs. 5%, P=0.001) (Table 1).

In the pediatric groups, the number of foreign nationals was significantly lower in the PPG than in the PedP-PG (53.57% vs. 72%, P=0.024). Patients in the PPG had higher pediatric end-stage liver disease (PELD) scores then the PedP-PG, but this difference was not statistically significant (24.93±16.54 vs. 18.21±17.59, P=0.477). The most common etiology was biliary atresia in both pediatric groups and all other parameters were comparable as well (Table 2).

Although duration of surgery was longer in the APG compared to the AdP-PG, mean intraoperative packed red blood cell transfusions were significantly less in the APG compared to the AdP-PG (6.18±4.65 vs. 7.84±9.45 units, P=0.044). The incidence of postoperative complications were similar in the APG and the AdP-PG (25.58% vs. 35.96%, P=0.197), with similar needs for radiological (35% vs. 37%, P=0.597), endoscopic, and surgical interventions (16.27% vs. 16.29%, P=0.994) in the early postoperative phase. While bile leaks were the most common early postoperative complication in the AdP-PG (14.6% vs. 4.7%, P=0.121), the incidence of postoperative bleeds (11.6% vs. 8.9%, P=0.596) and sepsis (9.3% vs. 7.3%, P=0.749) was similar in the APG and the AdP-PG. Postoperative ICU stays were significantly longer in the APG compared to the AdP-PG (16.06±10.09 vs. 13.54±11.82 days, P=0.045) (Table 3).

In both pediatric groups (PPG and PedP-PG), biliary atresia was the most common etiology. The mean PELD score (24.93±16.54 vs. 18.21±17.59, P=0.477) was higher in the PPG. Although not statistically significant, patients in the PPG had more postoperative complications then the PedP-PG (32.14% vs. 18.0%, P=0.502) (Table 4). Postoperative ICU stays and hospital stays were significantly longer in the PPG than in the PedP-PG (28.6±15.17 vs. 15.34±7.77 days, P=0.001) (Table 4).

The 30-day, 3-month, and 6-month outcomes were not significantly different in both the adult and pediatric groups (Tables 3 and 4, Fig. 2). Fig. 2 shows the 1-year Kaplan Meier survival analysis of the adult (Fig. 2A) and pediatric (Fig. 2B) populations in the pandemic and pre-pandemic periods. In the pediatric cohort, the 30-day mortality rate was lower in the PPG, whereas the 3-month and 6-month mortality rates were similar in both groups (Table 4, Fig. 2B).

It has been classically observed that centers dealing with higher surgical volumes have better outcomes. Mortality at HVCs is lower than at centers with lower surgical volumes, perhaps due to better surgical techniques, established protocols, and team organization [5,9]. However, these same HVCs also operate on sicker patients and accept borderline grafts so that the outcomes in these centers eventually even out [10,11]. The cut-off for designating a center as high-volume varies across different geographical regions. The cut-off for HVCs in the Indian subcontinent has yet to be formally defined; however, a large center in India typically conducts >100 LT procedures per year [7]. Our center is one of the high HVCs in India, consistently performing >200 LDLT procedures annually.

Axelrod et al. [6] investigated how the volume of cases impacted liver and kidney transplant outcomes by analyzing data from the American Scientific Registry of Transplant Recipients over 4 years (1996–2000). They found that the unadjusted 1-year mortality rate following LT was significantly different in HVCs (defined as those conducting 93 transplants per year, 15.9%) compared to LVCs (defined as those conducting 21 transplants per year, 16.9%) and medium volume centers (those conducting 48 transplants per year, 14.7%). LVCs were associated with a significantly higher risk of death (adjusted odds ratio, 1.30; P=0.0036). The unadjusted rate of renal graft loss within 1 year was significantly lower at HVCs (167 per year, 8.6%) than at centers with very low (20 per year, 9.6%), low (58 per year, 9.9%), and medium (93 per year, 9.7%) volumes (P=0.0014) [6].

Macomber et al. [12] analyzed the Universal Health Coverage database in the US from 2009 to 2010 and identified 63 liver transplant centers and 5,130 transplants. Mortality was also found to be lower at HVCs (>76 cases in 2009 and >71 cases in 2010) than at LVCs (11–47 cases in 2009 and 13–43 cases in 2010); 2.9% and 3.4%, respectively. HVCs had a shorter median length of stay than LVCs (9 vs. 10 days, P<0.0001) and shorter median ICU stays than LVCs (2 vs. 3 days, respectively; P<0.0001). Their study showed that increased center volume resulted in improved long-term LT outcomes and more efficient use of hospital resources, thereby lowering the cost [12].

A study analyzing outcomes in patients who underwent liver retransplantation between 2007 and 2016 from the Korean National Healthcare Insurance Service database reported similar results. In-hospital mortality after liver retransplantation was lower in HVCs than LVCs (25% and 36%, respectively; P=0.069) [13]. LT centers were categorized as HVCs (≥64 LTs per year) or LVCs (<64 LTs per year) based on annual LT case volume. In India, with imposition of a lockdown in March 2020, healthcare workers geared up to cope with the largely unknown pandemic. An advisory issued by the Indian Council of Medical Research (ICMR) recommended operating in emergencies only, as all healthcare resources were being diverted to manage those affected by the pandemic [14].

For several reasons, referrals to our center decreased significantly, leading to a drop in the total number of transplants performed. Patients and their families were reluctant to visit a hospital for fear of contracting infection. The general public postponed health-related matters, and physician availability in the hospitals was reduced as the workforce was diverted to take care of patients affected by COVID-19. Interstate and international travel restrictions also prevented patients from reaching transplant centers. The transplant community updated its treatment guidelines [15], stating that only sick patients with advanced liver failure such as ACLF, acute liver failure, and hepatocellular carcinoma (HCC) should be scheduled for transplant since a delay in their surgery was perceived to be life-threatening. Only very sick patients were referred to our transplant unit during the pandemic and the number of LT procedures performed in our unit was reduced by half. Delman et al. [16] reported that the COVID-19 pandemic did not significantly affect overall transplant outcomes. However, those eligible patients who could not receive transplants due to the overwhelmed health care system during the pandemic should be considered.

At our center, only 60 LT procedures were conducted over 6 months during the pandemic, compared to 228 patients in the pre-pandemic year. Since the caseload at our HVC was reduced by half during the pandemic, we decided to analyze the effect of the caseload reduction on patient outcomes. At our center, only sick cirrhotic patients with various types of decompensation, high MELD scores, and ACLF were admitted. LT was performed in those selected patients who, despite all medical management, were too sick to be discharged from the hospital. Although the PGs had higher rates of postoperative bleeding, sepsis, and longer ICU stays, the posttransplant outcomes were comparable in the pandemic and pre-pandemic patients for both adult and pediatric groups. We saw a higher percentage of patients with alcoholic hepatitis and alcohol-induced ACLF (Table 2), perhaps with the prevailing lockdown, uncertainty, and economic slowdown leading to excessive alcohol consumption. More patients were also referred with unresectable HCC (beyond Milan criteria), whom we managed with oral chemotherapy, chemo-embolization, or radio-embolization.

The COVID-19 pandemic and the national lockdown period presented new challenges. India’s blood banks were tasked with arranging plasma therapy for COVID-19 patients. Thus, in the absence of blood donation camps and the inability of the family and friends of patients undergoing LT to reach the hospital for blood donation, arranging blood and blood components for surgery became a major challenge. During the pandemic, all patients were screened for infection with SARS-CoV-2 before admission to the hospital and within 48 hours of surgery. Strict personal protective equipment policies were followed by healthcare personnel. The hospital did not permit visitors.

Despite all precautions, five elective transplants were deferred the day prior to elective LDLT when their asymptomatic donors tested positive for SARS-CoV-2 infection on routine mandatory preoperative screening. The surgeries were deferred for 6 weeks to ensure full recovery from COVID-19 as per the Liver Transplant Society of India guidelines. Three of the surgeries were subsequently rescheduled after ensuring the absence of COVID-19 sequelae in the donors, while two of the patients with liver disease died while awaiting transplant in the absence of another potentially healthy donor in the family. Two patients presented to our unit with COVID-19 induced ACLF. One of these underwent LDLT 4 weeks after recovery from the SARS-CoV-2 infection. Five of the chronic liver disease patients admitted for LT also tested positive for SARS-CoV-2 prior to their scheduled surgery. All five were managed with oxygen therapy and other supportive care and, over the course of a year, we were able to transplant three of these five patients after their full recovery. Two of the five died due to sepsis and multi-organ failure and could not be transplanted.

Our study was the first to compare the high and low case volume outcomes at the same center, thereby avoiding many confounding factors such as differences in the operative team and geographical factors. However, there were some limitations to our study. First, the time periods for the pandemic and PedP-PGs were not similar, which may have affected the number and type of patients included in both groups. The follow-up time was relatively short and some long-term complications like biliary stricture may have affected long-term outcomes. We also need to consider those patients who did not receive a transplant and died while the healthcare system was overwhelmed due to COVID-19. That loss is difficult to assess.

Despite these challenges, the 30-day, 3-month, and 6-month outcomes in adult patients during the pandemic were non-inferior to the pre-pandemic calendar year (Tables 3 and 4). This reflects how a reduction in the number of cases at an HVC led to the highly focused care of those patients who received LT despite adverse circumstances, which led to acceptable outcomes and graft survival in sicker patients. The COVID-19 pandemic affected transplant activity during the national lockdown in India, allowing only acutely ill patients to receive transplants at our facility. Although sicker patients were eventually transplanted at our center, the outcomes were not inferior. This indicates that a reduction in the usual number of operations at an otherwise busy HVC, allowed better individualized patient care for very sick transplant patients, which translated into equally good outcomes. Although HVCs traditionally have better outcomes, a reduction in their usual volume of operations need not result in inferior outcomes.