Hospital Mortality and Current Trends in Liver Transplantation in Germany
A Systematic Analysis of Standardized Hospital Discharge Data, 2008–2017
; ; ; ; ; ; ; ;
Background: Liver transplantation (LT) has undergone dynamic developments in recent decades. In Germany, the Federal Joint Committee (G-BA) recently tightened the guidelines regarding the minimum number of transplantations a center should perform annually. The aim of the study presented here, was to analyze recent trends in hospital mortality due to LT in Germany.
Methods: Standardized hospital discharge data (2008–2017) from the Federal Statistical Office of Germany were used to establish hospital mortality after LT and case volume distribution among centers performing <20 LT annually (low volume centers, LVC), 20–49 LT (medium volume centers, MVC), and ≥ 50 LT (high volume centers, HVC).
Results: Data from 9254 LT procedures were evaluated. The annual frequency of LT fell from n = 984 (2008) to n = 747 (2017), and over the same period the hospital mortality for all LT procedures went down from 15.8% to 11.0%. Hospital mortality was associated with age (<16 years: 5.3% to 60–69 years: 17.4%); however, there was no further increase in patients ≥ 70 years (16.5%). Univariate analysis revealed association of increased hospital mortality with liver disease etiology, the necessity for relaparotomy, and prolonged mechanical ventilation. The proportion of LT procedures performed in LVC and MVC increased and that in HVC decreased. LVC had higher hospital mortality than MVC/HVC, but this effect was dependent on patient age and disease etiology.
Conclusion: Our study showed that differences in mortality rate after LT among centers (LVC vs. MVC/HVC) were dependent on patient age and disease etiology. This should be taken into account when discussing the overall organization of LT in Germany.
Liver transplantation (LT) has evolved as a standard therapeutic approach for patients with end-stage liver disease, acute liver failure and early-stage liver cancer (1). While 5-year survival rates have constantly increased to over 80% (2, 3), organ shortage still represents the biggest obstacle to this curative treatment (4). In 2019, 868 patients were actively awaiting LT on the Eurotransplant waiting list and 317 patients were removed from the list because of death or being unfit for LT (5). Given the existing shortage of donor organs, hospital mortality following LT, which is specified between 10 and 17% (6, 7), is a critical factor that should be closely monitored to further optimize outcome of LT. In addition, the German LT field faced a major scandal in 2012 when irregularities over liver organ allocation at individual German LT centers became public. Following intense discussion about reducing the number of transplant centers in order to further improve the quality of LT in Germany (8), the German Federal Joint Committee (G-BA) has recently tightened the minimum quantity regulation for the performance of LT (20 LT/year) (9).
Standardized data sets represent a unique opportunity to understand current developments and identify areas with a potential for improvement. In this study, we used hospital discharge data to evaluate hospital mortality rates and other trends of LT in Germany between 2008 and 2017.
Study design and statistical analysis
Patients who received LT were identified using the OPS codes 5–504.0 and 5–504.1 (OPS, Classification of Operations and Procedures, i.e. the German modification of the International Classification of Procedures in Medicine). Due to complete anonymization of data no additional ethics approval was necessary. All analyses were performed via remote data access at the German Federal Statistical Office (Wiesbaden, Germany) using SPSS (IBM Corporation, Armonk, USA). Statistical tests and corresponding p-values are presented for descriptive purposes, only. The analyses are essentially hypothesis-generating and the results need to be confirmed on further data. Detailed information on eligibility criteria and statistics are given in the Supplementary material.
Liver transplantation in Germany between 2008 and 2017
A total of 9254 liver transplantation (LT) procedures performed in Germany between 2008 and 2017 were included in this study (Table, eTable). While the annual number of LT in Germany increased until 2010, it has since shown a decreasing trend with a fall of 19.8% from 2011 to 2013 (Figure 1), which correlated well with the occurrence of the liver transplant scandal in 2012 (10).
The median age at LT was 53 years and remained constant during the observation period (p = 0.291, eTable 1). 35.2% of patients were female and 64.8% were male. We identified 1652 hepatocellular carcinoma (HCC) patients, 1953 patients with alcoholic liver disease, 752 patients with acute or subacute liver failure, 385 patients with transplant rejection, 867 patients with biliary disease (primary biliary cholangitis [PBC], primary sclerosing cholangitis [PSC], secondary sclerosing cholangitis [SSC]) and 1989 patients with liver cirrhosis (not specified), based on their main diagnosis. The remaining 1656 patients were classified as “others”.
Hospital mortality rates after liver transplantation in Germany
The average hospital mortality rate of LT in Germany between 2008 and 2017 was 14.3%. Hospital mortality was highest in 2010 (17.0%) and decreased over time to 11.0% in 2017 (p<0.001, Figure 2). Hospital mortality was comparable between female (14.9%) and male patients (14.0%, odds ratio [OR]: 0.932, 95% confidence interval [CI]: 0.826–1.052). Moreover, hospital mortality was associated with recipients’ age.
When comparing six different age groups, hospital mortality ranged from only 5.2% in patients <16 years of age to 17.4% in patients between 60 and 69 (p<0.001, Figure 3). The small subgroup of patients aged 70 years or older (n = 170) showed a slightly lower hospital mortality rate (16.5%) compared to patients aged 60–69 years (p = 0.834, Figure 3). While the median age of liver donors has constantly increased from 50 years (2007) to 55 years (2017) (11), the subgroup of LT recipients aged ≥70 years has remained unchanged during this period (eTable).
In terms of disease etiologies, HCC patients showed the lowest hospital mortality rate (8.8%) followed by biliary disease patients (11.9%), while patients with acute or subacute liver failure as well as acute and chronic transplant rejection had a higher mortality rate (20.7% and 28.1%, respectively) (eFigure a). Patients with alcoholic liver disease (15.3%) and liver cirrhosis (not specified, 13.9%), had a hospital mortality rate comparable to the overall mortality rate of 14.3%.
We subsequently evaluated the influence of relaparotomy as a surrogate for postoperative complications and the influence of duration of mechanical ventilation on hospital mortality. The results were as follows: 2764 (29.9%) patients underwent relaparotomy. Relaparotomy rates were lowest in 2008 (26.3%) and highest in 2013 (33.1%, p = 0.061, eFigure b). Patients who required relaparotomy had a higher hospital mortality rate (26.3%) compared to those with an uncomplicated postoperative course (hospital mortality rate, 9.2%) (OR: 3.536, 95% CI: 3.138–3.985, p<0.001). 6482 patients required mechanical ventilation (MV) during their hospital stay (not including surgery itself). The median duration of MV was 76 h and was comparable between 2008 and 2017 (p = 0.380).
Hospital mortality rates showed a stepwise increase from 2.3% in patients without postoperative MV (n = 2772), 6.1% in patients with 1–48 h of MV (n = 2594), 14.6% in patients with 48 h to 168 h of MV (n = 1703), to 39% in patients with more than 168 h of MV (n = 2185) (p<0.001).
Liver transplantation at high, medium, and low case volume transplant centers
We stratified the German transplant centers into three groups according to the number of LT performed annually:
- Low case volume center (LVC): <20 LT/year
- Medium case volume center (MVC): 20–49 LT/year
- High case volume center (HVC): >50 LT/year.
In total, 60.2% of patients were transplanted at a HVC, 32.1% at a MVC, and 7.7% received LT at a LVC (Table). Hospital mortality was highest among patients transplanted at LVCs (17.0%) compared to patients who received LT at a MVC (13.4%) or HVC (14.4%) (OR: 0.799, 95% CI: 0.652–0.980, p = 0.031).
However, multivariate regression analysis revealed that the higher mortality rate at LVCs was not independent of the patients’ age and disease etiology (OR: 0.863, 95% CI: 0.701–1.063, p = 0.166). When looking at patients under 16 years of age only, hospital mortality was comparable between transplant centers that performed more or less than 20 LT annually (5.7 vs. 4.8%, OR: 0.784, 95% CI: 0.421–1.462).
Finally, we investigated a potential association between LT volume and the patients’ clinical course and found differences in terms of duration of hospitalization and mechanical ventilation (MV). As such, the median duration of hospitalization was numerically highest at LVCs (38 days, interquartile range [IQR]: 22–66) compared to MVCs/HVCs (34 days, IQR: 21–60) (p = 0.009). Moreover, patients transplanted at a LVC had a longer median duration of MV (98 hours, IQR: 30–410 hours) compared to patients transplanted at a MVC/HVC (75 hours, IQR: 25–283.5) (p = 0.002).
The percentage of patients receiving LT at a HVC showed a decreasing trend between 2008 and 2017 with a sudden drop from 69.0% to 45.4% in 2013, which was most likely caused by the German transplant scandal in 2012 (eFigure c).
Decreased hospital mortality
We identified an overall hospital mortality rate of 14.3% for all LT performed in Germany between 2008 and 2017, which is comparable to analyses e.g. from South Korea (6, 11). Our data show a decrease in hospital mortality over time, most likely caused by a variety of factors including a better matching of organ donors and recipients, decreased cold ischemia time combined with improved surgical techniques, and developments in immunosuppression, anesthesia and intensive care medicine (12, 13). However, the total number of LT has likewise steadily decreased from 2010 on e.g., due to a decline in postmortem organ donation associated with detection and reporting deficits (4). This trend was possibly aggravated by the liver transplant scandal that became public in 2012 (10).
Our study also provides evidence that various aspects of the individual clinical course such as the duration of MV or the necessity of relaparotomy are associated with increased hospital mortality. Similarly, data from the US suggest that a longer duration of MV is a risk factor for long-term mortality (14). Interestingly, it has recently been shown that early tracheostomy (≤3 days post-surgery) can result in shorter duration of MV following LT (16), representing a starting point to reduce the duration of MV to improve outcome.
The systematic analysis of mortality data is a fundamental aspect of patient safety. The time after the introduction of the MELD (Model for End-stage Liver Disease) system has shown that mortality increases with the severity and complexity of the disease (16). Therefore, it is imperative that the decision to list and transplant is made by an interdisciplinary team, as it has been established practice at all German transplant centers at the latest after the liver transplant scandal through several mechanisms of self- and third-party control (17). In this sense, the decision to transplant critically ill patients naturally involves the risk of increasing mortality at one’s own center and it cannot be excluded that there has been a bias due to selection of patients more likely to undergo a successful LT while other individuals may have been disadvantaged. In contrast, there is an increasing debate about minimum numbers and quality parameters in the German health care system with its large number of service providers, financial incentives for complex interventions, and a lack of central control of patient flows. In line, the G-BA has recently significantly reinforced the minimum quantity regulation for the performance of liver transplants (9).
Our study found a higher hospital mortality at low volume transplantation centers (<20 LT/year) compared to centers that perform ≥20 LT annually. In addition, we observed a longer duration of hospital stay and MV in patients transplanted at LVCs, which is in good agreement with data from Korea (6). Nevertheless, multivariate regression analysis shows that the higher mortality at LVCs is partially influenced by confounding variables such as the patients’ age and disease etiology. In addition, the higher mortality rate in LVCs was not observed when looking at patients under 16 years of age only.
Particularly in the wake of the German transplantation scandal in 2012, patient advocates, health insurance funds and doctors have called for a reduction in the number of German transplantation centers to simplify constant monitoring and to reduce the existing economic incentives associated with LT (8). Interestingly, our data show that the percentage of patients transplanted at LVCs increased from 2008 to 2013 and has remained around 10% since.
LT in older recipients has remained controversial (11, 18, 19). With only 170 LT during the 10-year observation period (1.2%), LT in patients >70 years has been an exception in Germany. Interestingly, our analysis revealed an increase in hospital mortality with recipients’ age but a hospital mortality rate of 16.5% in recipients older than 70 years, which was comparable to patients between 60 and 70 years. Thus, in times of an increasing donor age, LT in very old recipients may still be a reasonable treatment option in selected cases and a higher numerical recipient age should not per se be an exclusion criterion for LT.
Of note, our analysis showed that two out of three livers were transplanted to men, which raises the question if there could be a gender bias in the allocation system. Allocating donated organs based on the MELD system disfavors females, due to the inherently lower creatinine values in women (20). However, there are also findings arguing against this hypothesis, such as lower incidences and severity of liver disease and lower alcoholism or HCC incidence rates in women. Importantly, a recent study found that women had no difference in risk of liver related death despite lower rates of listing and transplantation (21).
Patients transplanted for HCC had the lowest hospital mortality rate, corresponding to the fact that HCC patients usually have a clinically lower stage of liver cirrhosis. Of course, this difference relates exclusively to hospital mortality. In contrast, the risk of tumor recurrence, which often presents as systemic tumor spreading to multiple organs including lungs, liver, bones and lymph nodes, relativizes the overall outcome of HCC patients (12, 22).
Of note, the present German ICD coding system in combination with the remote access of data precluded identification of all etiology subgroups of patients such as those with non-alcoholic steatohepatitis (NASH) or hepatitis B and C, representing a disadvantage of nationwide multicenter analysis including ours compared to smaller single center studies. As e.g. the incidence of NASH is rising (23), a clearer attribution of NASH cirrhosis and other liver conditions to individual ICD codes could lead to a more adequate representation of this important entity in the health care system and the political discourse.
We acknowledge that our study has some important limitations. Our analysis only gave information on hospital mortality but mortality data for later time points such as transplant and overall survival were not available. Complete anonymization of data also prevented the identification/exclusion of specific subgroups such as patients who received more than one LT.
Furthermore, the pure analysis of mortality is limited in its informative value, as it does not allow a more in-depth analysis of the causes. Therefore, any conclusions must be drawn carefully and it must always be borne in mind that mortality has many causes. Such a discussion must not lead to a situation in which the doctor is discouraged to transplant even very seriously ill patients in order not to jeopardize the center’s LT success rate e.g., in terms of patient mortality rate.
Finally, due to the time latency with which the data is made available by the Federal Statistical Office, it was not possible to perform an analysis up to the present date.
Taken together, our data show that despite being a highly complex procedure that necessitates well organized interdisciplinary cooperation, hospital mortality rates of LT in Germany have decreased between 2008 and 2017. Despite these positive findings, efforts to improve the organization, coordination, and quality of organ donation should persist, to ensure that even more patients will benefit from LT in the future.
Work in the group of Tom Luedde was funded by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program through the ERC Consolidator Grant PhaseControl (Grant Agreement n° 771083). The group of Tom Luedde was further supported by the German Cancer Aid (Deutsche Krebshilfe 110043 and a Mildred Scheel Professorship) and the German Research Foundation (SFB-TRR57/P06, LU 1360/3–1, CRC1380/A01, and CA 830/3–1).
We would like to thank Melanie Heiliger at the German Federal Statistical Office for her professional assistance in the remote analysis of data. Moreover, we express our gratitude to the following colleagues for their intellectual input and the vivid discussion of the results: Theresa H. Wirtz (University Hospital RWTH Aachen), Tom F. Ulmer (University Hospital RWTH Aachen), Jan Bednarsch (University Hospital RWTH Aachen), Patrick H. Alizai (University Hospital RWTH Aachen), Matthijs Kramer (University Hospital RWTH Aachen), Pavel Strnad (University Hospital RWTH Aachen), Alexander Koch (University Hospital RWTH Aachen), Andreas A. Schnitzbauer (Frankfurt University Hospital), Jonel Trebicka (Frankfurt University Hospital), Frank Tacke (Charité Berlin), and Matthijs Kramer (MUMC Maastricht).
Conflict of interest statement
Dr. Loosen received reimbursement of travel costs and lecture fees from Roche, MSD und BMS.
Prof. Bock received counsel fees from Abbvie and lecture fees from Abbvie und Gilead.
Prof. Knoefel received payment for medicolegal evaluations relating to the topic of this article. Study support (third party funding) was made available to him by the Else-Kröner-Fresenius Foundation.
Prof. Keitel received lecture fees from Falk, Abbvie, Albireo, Gilead, and Intercept.
Prof. Neumann received reimbursement of conference fees and travel expenses as well as lecture fees from Astellas, Roche, Merck and Novartis. Study support was made available to him by Astellas, Merck, Roche, Novatis, and Organ Assist.
Prof. Luedde received counsel fees from Roche, Eisai, BMS, MSD, Sanofi, Incyte, Merck, and Bayer. He was reimbursed for conference fees and travel expenses by EASL and received lecture fees from Falk.
The remaining authors declare that no conflict of interest exists.
Manuscript received on 28 July 2020, revised version accepted on 12 April 2021
Prof. Dr. med. Tom Luedde
PD Dr. med. Sven H. Loosen
Klinik für Gastroenterologie, Hepatologie und Infektiologie
Medizinische Fakultät der Heinrich-Heine-Universität Düsseldorf
Moorenstr. 5, 40225 Düsseldorf, Germany
Cite this as:
Loosen SH, Bock HH, Hellmich M, Knoefel WT, Trautwein C, Keitel V, Bode JG, Neumann UP, Luedde T: Hospital mortality and current trends in liver transplantation in Germany—a systematic analysis of standardized hospital discharge data, 2008–2017. Dtsch Arztebl Int 2021; 118: 497–502. DOI: 10.3238/arztebl.m2021.0210
eMethods, eTable, eFigures:
Liver Transplant Society Aachen–Düsseldorf–Maastricht: PD. Dr. med. Sven H. Loosen, Prof. Dr. med. Hans H. Bock, Prof. Dr. med. Wolfram T. Knoefel, Prof. Dr. med. Christian Trautwein, Prof. Dr. med. Verena Keitel, Prof. Dr. med. Johannes G. Bode, Prof. Dr. med. Ulf P. Neumann, Prof. Dr. med. Tom Luedde
Institute of Medical Statistics and Computational Biology, Faculty of Medicine, University of Cologne: Prof. Dr. rer. medic. Martin Hellmich
Department for General Visceral and Pediatric Surgery, Düsseldorf University Hospital, Medical Faculty of the Heinrich Heine University Düsseldorf: Prof. Dr. med. Wolfram T. Knoefel
Department of Medicine III, University Hospital RWTH Aachen: Prof. Dr. med. Christian Trautwein
Department of Visceral and Transplant Surgery, University Hospital RWTH Aachen: Prof. Dr. med. Ulf P. Neumann
Maastricht University Medical Centre+ (MUMC), Maastricht, Netherlands: Prof. Dr. med. Ulf P. Neumann
|1.||Durand F, Pavesi M, Cheung R: Liver transplantation for non-alcoholic steatohepatitis in Europe: Where do we stand? J Hepatol 2019; 71: 240–2 CrossRef MEDLINE|
|2.||Martin K, Allen E: England NHS: Annual report on liver transplantation: Report for 2014/2015; 2015.|
|3.||Adam R, Karam V, Cailliez V, et al.: 2018 Annual report of the European Liver Transplant Registry (ELTR)—50-year evolution of liver transplantation. Transpl Int 2018; 31: 1293–317 CrossRef MEDLINE|
|4.||Schulte K, Borzikowsky C, Rahmel A, et al.: Decline in organ donation in Germany—a nationwide secondary analysis of all inpatient cases. Dtsch Arztebl Int 2018; 115: 463–8 CrossRef MEDLINE PubMed Central|
|5.||Branger P, Vogelaar S: Annual report 2019 – Eurotransplant International Foundation; 2020.|
|6.||Yoo S, Jang EJ, Kim GH, et al.: Effect of institutional case volume on in-hospital mortality after deceased donor liver transplantation: a nationwide retrospective cohort study in Korea. Transplant Proc 2018; 50: 3644–9 CrossRef MEDLINE|
|7.||Nijboer A, Ulrich F, Bechstein WO, Schnitzbauer AA: Volume and outcome relation in German liver transplant centers: What lessons can be learned? Transplant Res 2014; 3: 5 CrossRef MEDLINE PubMed Central|
|8.||Deutsches Ärzteblatt: Verringerung von Transplantationszentren gefordert. 2013. www.aerzteblatt.de/nachrichten/52893/Verringerung-von-Transplantationszentren-gefordert. kna/dapd/afp/aerzteblatt.de (last accessed on 5 July 2021).|
|9.||Gemeinsamer Bundesausschuss (G-BA): Beschluss des Gemeinsamen Bundesausschusses über eine Änderung der Mindestmengenregelungen: Änderung der Nr. 1 der Anlage und weitere Änderungen (BAnz AT B1). Berlin 2020; 1–2. www.g-ba.de/downloads/39-261-4392/2020-07-16_Mm-R_Lebertransplantation_BAnz.pdf.|
|10.||Shaw D, Neuberger J, Murphy P: Lessons from the German organ donation scandal. J Intensive Care Soc 2013; 14: 200–1 CrossRef|
|11.||Gil E, Kim JM, Jeon K, et al.: Recipient age and mortality after liver transplantation: a population-based cohort study. Transplantation 2018; 102: 2025–32 CrossRef MEDLINE PubMed Central|
|12.||Wallace D, Cowling TE, Walker K, et al.: Short- and long-term mortality after liver transplantation in patients with and without hepatocellular carcinoma in the UK. Br J Surg 2020; 107: 896–905 CrossRef MEDLINE|
|13.||van der Meulen JHP, Lewsey JD, Dawwas MF, Copley LP, UK and Ireland Liver Transplant Audit: Adult orthotopic liver transplantation in the United Kingdom and Ireland between 1994 and 2005. Transplantation 2007; 84: 572–9 CrossRef MEDLINE|
|14.||Yuan H, Tuttle-Newhall JE, Chawa V, et al.: Prognostic impact of mechanical ventilation after liver transplantation: a national database study. Am J Surg 2014; 208: 582–90 CrossRef MEDLINE PubMed Central|
|15.||Cammann S, Timrott K, Vondran FWR, et al.: Early tracheostomy reduces time of mechanical ventilation in respiratory high-risk patients after liver transplant. Exp Clin Transplant 2018; 16: 631–4.|
|16.||Weismüller TJ, Negm A, Becker T, et al.: The introduction of MELD-based organ allocation impacts 3-month survival after liver transplantation by influencing pretransplant patient characteristics. Transpl Int 2009; 22: 970–8 CrossRef MEDLINE|
|17.||Tacke F, Kroy DC, Barreiros AP, Neumann UP: Liver transplantation in Germany. Liver Transplant 2016; 22: 1136–42 CrossRef MEDLINE|
|18.||Wilson GC, Quillin RC, Wima K, et al.: Is liver transplantation safe and effective in elderly (≥ 70 years) recipients? A case-controlled analysis. HPB (Oxford) 2014; 16: 1088–94 CrossRef MEDLINE PubMed Central|
|19.||Aduen JF, Sujay B, Dickson RC, et al.: Outcomes after liver transplant in patients aged 70 years or older compared with those younger than 60 years. Mayo Clin Proc 2009; 84: 973–8 CrossRef CrossRef|
|20.||Mindikoglu AL, Regev A, Seliger SL, Magder LS: Gender disparity in liver transplant waiting-list mortality: The importance of kidney function. Liver Transplant 2010; 16: 1147–57 CrossRef MEDLINE PubMed Central|
|21.||Mazumder NR, Celaj S, Atiemo K, et al.: Liver-related mortality is similar among men and women with cirrhosis. J Hepatol 2020; 73: 1072–81 CrossRef MEDLINE|
|22.||Filgueira NA: Hepatocellular carcinoma recurrence after liver transplantation: risk factors, screening and clinical presentation. World J Hepatol 2019; 11: 261–72 CrossRef MEDLINE PubMed Central|
|23.||Wong RJ, Aguilar M, Cheung R, et al.: Nonalcoholic steatohepatitis is the second leading etiology of liver disease among adults awaiting liver transplantation in the United States. Gastroenterology 2015; 148: 547–55 CrossRef MEDLINE|