- Email: [email protected]
Mortality in candidates waiting for combined liver-intestine transplants exceeds that for other candidates waiting for liver transplants
Mortality in Candidates Waiting for Combined Liver– Intestine Transplants Exceeds That for Other Candidates Waiting for Liver Transplants Jonathan Fryer,* Suzanne Pellar,* Debra Ormond,† Allan Koffron,* and Michael Abecassis* The United Network for Organ Sharing (UNOS) reports indicate that mortality on the intestine transplant waiting list is higher than on other transplant waiting lists. The goals of this study were (1) to determine whether most of the intestinal transplant candidate deaths have occurred in those who also need liver transplants, and (2) to compare the waiting list mortality in the liver–intestine candidate subset with the overall liver transplant candidate population. We found that 90% of intestine transplant waiting list deaths have occurred in candidates who also needed liver transplants. Since 1994, annual mortality has been higher in liver–intestine transplant candidates than in the overall liver transplant candidate population, and these differences have been statistically significant since 1996. These mortality differences applied to all age groups. Also, status 2B, 3, and 7 candidate mortality was significantly higher in liver–intestine candidates than in the overall liver transplant candidate population. Because there were so few liver–intestine transplant candidates listed as status 1 or 2A, a meaningful comparison was not possible in these statuses. These data indicate that liver– intestine transplant candidates are a unique subset of liver transplant candidates with a significantly higher risk of dying on the waiting list. Recent changes in UNOS liver allocation policy that gives higher priority to liver–intestine candidates may help to reduce this discrepancy. However, further research into the etiology of liver disease in patients on long-term parenteral nutrition and earlier referral of high-risk short bowel syndrome patients to centers with special expertise in their management are needed for an ultimate solution to this problem. (Liver Transpl 2003;9:748-753.)
patients will develop significant complications, including line-sepsis, recurrent dehydration, loss of available sites for vascular access, and progressive liver disease.2 Medicare recognizes these complications as being indicators of TPN failure that warrant proceeding to intestinal transplantation (Medicare Coverage Policy Decisions, Intestinal and Multivisceral Transplantation, CAG-00036, October 4, 2000). Of these complications, liver disease may have the most devastating and predictable consequences. End-stage liver disease (ESLD) can ultimately develop in over 50% of intestinal failure patients in high-risk subsets, such as those with ⬍50 cm of small bowel.3 Those who develop ESLD have essentially 0% survival at 5 years.4 Therefore, in SBS patients with ESLD, combined liver–intestine transplantation (LIT) is the only option for survival.5 United Network for Organ Sharing (UNOS) data indicate that waiting list mortality for intestinal transplant candidates greatly exceeds that for all other organ candidates.6 However, a subset analysis to determine whether the high waiting list mortality in the intestine candidate group is skewed by a specific candidate subset has not previously been performed. The purpose of this study was to define the waiting list mortality in the LIT candidate subset of intestinal transplant candidates and compare it with the waiting list mortality of the overall liver transplant candidate population.
P
Experimental Procedures
atients with short bowel syndrome (SBS) or other forms of intestinal failure require long-term total parenteral nutrition (TPN) to survive.1 Some of these From the *Department of Surgery, Northwestern University Medical School, Chicago, IL, and the †United Network for Organ Sharing, Richmond, VA. Address reprint requests to Jonathan P. Fryer, MD, Northwestern University Medical School, Department of Surgery, Division of Transplantation, 675 N. St. Clair, Galter Pavilion, Suite 17-200, Chicago, IL 60611. Telephone: 312-695-2125; FAX: 312-695-9194; E-mail: [email protected] Copyright © 2003 by the American Association for the Study of Liver Diseases 1527-6465/03/0907-0016$30.00/0 doi:10.1053/jlts.2003.50151
748
Data were obtained from the UNOS database. Waiting list data for potential intestinal transplant recipients since 1993 was collected, and was first reported in 2000 and updated in 2001.6 For the purposes of this study, further analyses were performed on the UNOS data to provide more detailed information on the intestine transplant candidate subsets, particularly those that also needed liver transplants. These data were compared with UNOS data on all liver transplant candidates. For each comparison group, the mortality rate was calculated as the number of deaths were divided by the total number of patient days of follow-up in that group. Mortality rates were compared using Fisher’s exact test,
Liver Transplantation, Vol 9, No 7 ( July), 2003: pp 748-753
749
High Mortality in Liver–Intestine Candidates
Table 1. Waiting List Mortality in Intestine and Liver Transplantation Candidate Subsets Before and After 1997 UNOS Status Modifications 01/01/93 to 06/30/97 Transplantation Candidate Type
Total Patients Listed*
Total Deaths
Liver and intestine only Intestine only Intestine/multivisceral All intestine All liver
249 76 38 363 29,514
69 5 7 81 3,299
07/01/97 to 12/31/01
Mortality Rate†
Total Patients Listed*
Total Deaths
Mortality Rate†
333.63 104.21 129.70 284.05 166.16
429 198 108 735 48,447
154 14 13 181 6,943
493.66‡ 118.12 128.51 323.75 121.06‡
*Total number of patients on the list at any time during the designated time period. †Number of deaths per 1,000 patient years waiting for a transplant. ‡P ⬍ .05 versus 01/01/93 to 06/30/97.
and differences were considered significant if P values were less than .05.
Results A more detailed analysis of UNOS data regarding intestinal waiting list mortality6 shows that since 1994, 86% of the deaths occurred in candidates listed for combined LITs (Table 1). An additional 8% of the deaths occurred in candidates listed for multivisceral transplants, most of whom needed both a liver and an intestine as part of their multivisceral graft. Only 7% of the intestine waiting list deaths occurred in candidates who needed isolated intestine transplants only. In every year since 1993, candidates needing LITs have never made up more than 1% of the annual liver transplant waiting list (Table 1). Since 1997, when significant changes in the UNOS status definitions for liver transplant candidates were implemented, the mortality rate has increased significantly in the LIT candidate subset. Conversely, a significant decrease in the overall liver transplant waiting list mortality has occurred since 1997 (Table 1). The annual mortality rate with LIT candidates has consistently exceeded that of the overall liver transplant waiting list since 1993 (Fig. 1). Although most liver transplantations have been performed in adults, the majority of intestinal transplantations have been performed in pediatric patients.5 Accordingly, most candidates on the intestine waiting list are pediatric.6 Therefore, to determine whether the overall waiting list mortality differences may be skewed by a higher mortality in the pediatric LIT candidate subset, in which donor–recipient size discrepancies and other issues severely limit access to both cadaveric and
living donor organs, we evaluated adult and pediatric patients independently. When the total numbers of waiting list deaths were evaluated for 1993 to 2001, 83% of the LIT candidate deaths occurred in the pediatric population, whereas only 8% of deaths in the overall liver candidate population were pediatric. When the pediatric candidate population (0 to 17 years) was evaluated independently, annual waiting list mortality rates were significantly higher for the LIT candidates for 1995 to 2001 inclusive (Fig. 2). In the adult candidate population (⬎17 years), annual waiting list mortality rates were significantly higher for the LIT candidates for
Figure 1. Comparison of annual waiting list mortality rates in all liver transplant (Liver), and combined Liver– intestine transplant (LIT) candidates for 1993 to 2001. Mortality rates were calculated as the total annual number of deaths occurring in patients on the waiting list divided by the combined waiting times (in years) for all patients on the list during that year ⴛ 1,000 (*, P < .05 compared with liver for the same year). Liver, LIT.
750
Fryer et al
Figure 2. Comparison of annual waiting list mortality rates in adult (>17 years) and pediatric (Ped)(0 to 17 years) liver transplant (Liver) and combined liver–intestine transplant (LIT) candidates for 1993 to 2001. Mortality rates were calculated as the total annual number of deaths occurring in patients on the waiting list divided by the combined waiting times (in years) for all patients on the list during that year ⴛ 1,000 (#, P < .05 compared with adult liver, and *, P < .05 compared with pediatric liver for the same year). 䊐 Liver-Adult, Liver-Ped, LIT-Adult, LIT-Ped.
1997 to 1999 inclusive (Fig. 2). However, when specific age groups were evaluated based on the candidate’s age at the time of listing, waiting list mortality in LIT candidates significantly exceeded that seen in the overall liver candidate population in all age groups (Fig. 3). These differences were most significant in neonates (⬍1 year) and adults (35 to 65 years). Mortality rates were further analyzed based on the candidate’s final UNOS status. UNOS mortality data analyzed on the basis of status were unavailable for candidates listed before 1995. Because there were several modifications made in the UNOS status definitions in 1997, data from that year were excluded and results were analyzed separately based on whether the status was assigned before (1995 to 1996 inclusive) or after (1998 to 2001 inclusive) 1997. Because of differences in the definition of status 1 in adult and pediatric patients, mortality rates for status 1 adult and pediatric candidate populations were evaluated separately. Almost all status 1 LIT waiting list deaths occurred in the pediatric candidate population, and this reflected a very low use of the status 1 category in adult LIT candidates. In fact, the total cumulative waiting time for all status 1 adult LIT candidates was 0.6 years. With the pediatric patients, after 1997, the mortality rates were significantly lower in the LIT can-
didates than in the overall liver transplant candidate population (Fig. 4). Otherwise, the data available for status 1 patients were insufficient for statistical analysis. For all other UNOS statuses (2A, 2B, 3, and 7), cumulative waiting list mortality data are shown for the time period following the 1997 modifications (i.e., 1998 to 2001; Fig. 5). In status 2A patients, mortality rates in the overall liver transplant candidate population exceeded those seen in LIT candidates, although inadequate numbers of LIT candidates precluded statistical analysis. Because the 2A status category was not available to pediatric patients, these data reflect only adult candidates. As with status 1, the status 2A category was used infrequently in adult LIT candidates, as is reflected by the total cumulative waiting time of 1.7 years for all adult LIT 2A candidates. In status 2B, 3, and 7 candidates, mortality rates in LIT candidates significantly exceeded that of the overall liver transplant candidate population.
Discussion Intestinal transplantation is the only therapeutic option for patients who are unable to sustain themselves nutritionally via enteral routes and are unable to tolerate
Figure 3. Comparison of waiting list mortality rates in liver transplant (Liver) and combined liver–intestine transplant (LIT) candidates in specific age groups from 1993 to 2001. Age group assignment was determined based on the candidate’s age at the time of listing. Agespecific mortality rates were calculated as the total annual number of deaths occurring in candidates on the waiting list in a specific age group divided by the combined waiting times (in years) for all candidates in that age group that were on the waiting list during the designated time period ⴛ 1,000 (* ⴝ P < .05 compared with liver for the same age group). Liver, LIT.
High Mortality in Liver–Intestine Candidates
Figure 4. Comparison of waiting list mortality rates in status 1 adult (>17 years) and pediatric (0 to 17 years) liver transplant (Liver) and combined liver–intestine transplant (LIT) candidates before (1995 to 1996) and after (1998 to 2001) 1997 modifications in UNOS status 1 criteria. Mortality rates were calculated as the total number of waiting list deaths occurring in a candidate subset divided by the combined waiting times (in years) for all patients in that subset ⴛ 1,000. All data was determined based on a candidate’s final UNOS status before death, transplantation, or year’s end. The total number of waiting list deaths occurring in each candidate subset is indicated in parentheses in the corresponding data bar (*, P < .05 compared with liver in the same time period). Liver (95-96), LIT (95-96), 䊐 Liver (98-01), LIT (98-01).
long-term TPN therapy. According to Medicare policy, impending liver failure is one of the criteria that defines failed TPN therapy, and therefore warrants consideration of intestinal transplantation. Although the development of irreversible liver disease is relatively uncommon in the entire population of patients who receive home TPN, in patients with very short guts (i.e., ⬍50 cm) who are destined to be permanently TPN-dependent, it occurs in up to 50%.3 The etiology of the liver disease in these patients has not been clearly established, but is likely multifactorial.3,7-10 Several factors that correlate with its occurrence include duration of TPN therapy,9 length of residual small bowel,3 and the presence of residual colon.3 Other factors that are implicated include bacterial overgrowth,8 composition of TPN macronutrients (i.e., lipids, carbohydrates),3 missing nutritional factors,7 and hepatotoxic factors in TPN.10 In SBS patients on TPN in whom ESLD develops, the 5-year survival is essentially 0%.4 Therefore, a com-
751
bined LIT is the only option for survival for these patients. Based on the overall world experience thus far, patient survival after combined liver–intestine transplantation is 63% at 1 year.5 If ESLD has not developed, simultaneous liver transplantation is not needed. In these circumstances, when the intestine is transplanted alone, 1-year patient survival is significantly better (70% to 80%).5 This survival discrepancy may be partly because life-threatening graft- or immunosuppression-related complications can be salvaged by graft removal and withdrawal of immunosuppressive therapy with isolated intestinal transplants, whereas this is not an option with LIT. Another factor is that LIT candidates are, by definition, much sicker going into their transplantation. Although it is therefore preferable to perform isolated intestinal transplantations before ESLD develops, many patients are not referred for intestinal transplantation evaluation until irreversible liver disease is already present. These patients who need LITs are placed on both the intestine and liver waiting lists, although their priority on the liver waiting list
Figure 5. Comparison of waiting list mortality rates in liver transplant (Liver) and combined liver–intestine transplant (LIT) candidates in all UNOS status categories (except status 1) after 1997 (1998 to 2001). Status assignment was based on the candidate’s final UNOS status before death, transplantation, or year’s end. Status-specific mortality rates were calculated as the total annual number of waiting list deaths occurring in a specific status category (i.e., 2A, 2B, 3, or 7) divided by the combined waiting times (in years) for all candidates in that status category that were on the list during the designated time period ⴛ 1,000. The total number of waiting list deaths occurring in each candidate subset is indicated in parentheses in the corresponding data bar (*, P < .05 compared with liver for the same UNOS status). Liver, LIT.
752
Fryer et al
Table 2. One-Year Patient Survival in Liver/Intestine Transplant Recipients Based on Their UNOS Status at the Time of Transplantation Liver-Intestine Transplant status 1-yr patient survival
Status 1
Status 2
Status 2A
Status 2B
Status 3
49%
56%
*
67%
60%
*Insufficient data available.
ultimately determines when they will be allocated organs. Until recently, all candidates, including those requiring LIT, were prioritized on the liver waiting list based on criteria derived from the Childs-TurcottePugh (CTP) score.11 Concerns regarding the effectiveness of these criteria to appropriately prioritize all candidates waiting for liver transplants led to their abandonment and replacement by the Model for EndStage Liver Disease/Pediatric End-Stage Liver Disease (MELD/PELD) system. However, the data presented here indicate that CTP-based UNOS prioritization was a much a better predictor of waiting list mortality in candidates waiting only for liver transplants than it was in the subset who needed an LIT. This underscores the fact that, in general, criteria used to predict waiting list mortality for liver transplant candidates may be irrelevant to the LIT candidate subset because they do not address the unique risk factors inherent in this population. The most important factor contributing to high mortality in LIT candidates is sepsis. Although all ESLD patients are at risk for fatal episodes of sepsis,12 SBS patients on TPN are particularly susceptible.13,14 Several factors may contribute to this. First, unlike most ESLD patients, LIT candidates need long-term central venous catheters (CVCs). Because of constant diarrhea and/or the need to maintain stomas and/or enteral tubes, SBS patients must be especially meticulous in their technique to avoid contamination of their CVCs with enteral organisms. Although their CVCs are susceptible to infection in the best of circumstances,15 they are even more likely to become infected as their clinical status deteriorates and their ability to provide meticulous CVC maintenance deteriorates. For the same reasons, the risk of urinary tract infections caused by contamination with enteral organisms is increased. Another factor that likely contributes to increased sepsis in LIT candidates is bacterial overgrowth and translocation.16,17 Although graft and patient survival results after LITs are currently inferior to the results seen after isolated
liver transplantations, they have steadily improved as the experience with the intestinal transplantation increases,5 and the early progress of LITs is not unlike that encountered with other solid organ transplantations.18,19 Since 1997, the vast majority of LIT candidate deaths (85%) have occurred in status 2B, 3, or 7 candidates (Fig. 2) and status 2B candidates accounted for 50% of these deaths. These data indicate that the high waiting list mortality risk in these LIT candidates was not fully recognized. However, the status 2B candidates who did survive long enough to receive a LIT did better posttransplant than LIT candidates in all other status categories (Table 2). This suggests that the best results may be achieved with more timely transplantation of LIT candidates, before they become too sick. The great discrepancy in mortality between LIT candidates and other liver transplant candidates clearly distinguishes this candidate subset from other liver transplant candidates. These data show that the recently abandoned UNOS liver prioritization system failed in appropriately prioritizing LIT candidates on the liver transplant waiting list. Although the recently implemented MELD/PELD system20 has not been in place long enough to allow a valid assessment, there is little reason to believe that it will do better, because it does not incorporate criteria that recognize the unique risk factors of the LIT subset. LIT candidates are a small but unique subset of liver transplant candidates who have had a disproportionately high mortality on the waiting list in all age groups. Changes in UNOS liver allocation policy that would increase their priority on the liver transplant waiting list will be an important step to rectifying this discrepancy. However, ultimate resolution of this problem will require additional measures, including increased research to determine the etiology of ESLD in these patients and increased awareness by physicians who manage TPN patients that high-risk (i.e., ⬍100 cm of short bowel) SBS patients need early referral to centers with advanced expertise in SBS management, including intestinal transplantation, to avoid the need for LIT.
High Mortality in Liver–Intestine Candidates
Acknowledgement The authors thank Renee Poland for her assistance in preparation of this manuscript, and Fred Rademaker, Sarah Fung, and Eric Edwards for their statistical analyses of the data. The data and analyses reported in the 2001 Annual Report of the U.S. Organ Procurement and Transplantation Network and the Scientific Registry of Transplant Recipients have been supplied by UNOS and University Renal Research and Education Association (URREA) under contract with HHS. The authors alone are responsible for reporting and interpreting these data.
References 1. Vanderhoof J, Langnas A. Short bowel syndrome in children and adults. Gastroenterology 1997;113:1767-1778. 2. Fryer J. The role of intestinal transplantation in the management of intestinal failure. Curr Gastroenterol Rep 3:334. 3. Cavicchi M, Beau P, Crenn P, Degott C, Messing B. Prevalence of liver disease and contributing factors in patients receiving home parenteral nutrition for permanent intestinal failure. Ann Intern Med 2000;132:525. 4. Chan S, McCowen K, Bistrian BR, Thibault A, Keane-Ellison M, Forse RA, et al. Incidence, prognosis, and etiology of endstage liver disease in patients receiving home parenteral nutrition. Surgery 1999;126:28. 5. Grant D. Intestinal transplantation: 1997 report of the international registry. Transplantation 1999;67:1061-1064. 6. 2001 Annual Report of the U.S. Organ Procurement and Transplantation Network and the Scientific Registry for Transplant Recipients. Transplant Data 1991–2000. Department of Health and Human Services, Health Resources and Services Administration, Office of Special Programs, Division of Transplantation, Rockville, MD; United Network for Organ Sharing, Richmond, VA; University Renal Research and Education Association, Ann Arbor, MI. 7. Buchman AL, Dubin MD, Moukarzel AA, Jenden DJ, Roch M, Rice KM, et al. Choline deficiency: A cause for hepatic steatosis during parenteral nutrition that can be reversed with intravenous choline supplementation. Hepatology 1995;22:1399-1403.
753
8. Meehan JJ, Georgeson KE. Prevention of liver failure in parenteral nutrition– dependent children with short bowel syndrome. J Pediatr Surg 1997;32:474-475. 9. Messing B, Crenn P, Beau P, Boutron-Ruault MC, Rambaud JC, Matuchansky C. Long-term survival and parenteral nutrition dependence in adult patients with short bowel syndrome. Gastroenterology 1999;117:1043-1050. 10. Iyer KR, Spitz L, Clayton P. New insight into mechanisms of parenteral nutrition–associated cholestasis: Role of plant sterols. J Pediatr Surg 1998;33:1-6. 11. Holman JM, Rikkers LF. Success of medical and surgical management of acute variceal hemorrhage. Am J Surg 1980;140:816. 12. Cirera I, Bauer TM, Navasa M, Vila J, Grande L, Taura P, et al. Bacterial translocation of enteric organisms in patients with cirrhosis. J Hepatol 2001;34:32-37. 13. Terra RM, Plopper C, Waitzberg DL, Cukier C, Santoro S, Martins JR, et al. Remaining small bowel length: Association with catheter sepsis in patients receiving home parenteral nutrition: evidence of bacterial translocation. World J Surg 2000;24: 1537-1541. 14. Okada Y, Klein NJ, Pierro A. Neutrophil dysfunction the cellular mechanism of impaired immunity during total parenteral nutrition in infancy. J Pediatr Surg 1999;34:242-245. 15. Moukarzel AA, Haddad I, Ament ME, Buchman AL, Reyen L, Maggioni A, et al. 230 patient years experience with home long-term parenteral nutrition in childhood: Natural history and live of central venous catheters. J Pediatr Surg 1994;29:13231327. 16. Pierro A, Van Saene HK, Donnell SC, Hughes J, Ewan C, Nunn AJ, et al. Microbial translocation in neonates and infants receiving long-term parenteral nutrition. Arch Surg 1996;131:176179. 17. Wells CL, Maddaus MA, Simmons RL. Proposed mechanisms for the translocation of intestinal bacteria. Rev Infec Dis 1998; 10:958-979. 18. Calne RY, Williams R. Orthotopic liver transplantation: the first 60 patients. Br Med J 1977;1:471-476. 19. Jamieson SW, Stinson EB, Shumway NE. Cardiac transplantation in 150 patients at Stanford University. Br Med J 1979;1:9395. 20. Pagliaro L. MELD: The end of the Child-Pugh classification? J Hepatol 2002;36:141-142.
patients will develop significant complications, including line-sepsis, recurrent dehydration, loss of available sites for vascular access, and progressive liver disease.2 Medicare recognizes these complications as being indicators of TPN failure that warrant proceeding to intestinal transplantation (Medicare Coverage Policy Decisions, Intestinal and Multivisceral Transplantation, CAG-00036, October 4, 2000). Of these complications, liver disease may have the most devastating and predictable consequences. End-stage liver disease (ESLD) can ultimately develop in over 50% of intestinal failure patients in high-risk subsets, such as those with ⬍50 cm of small bowel.3 Those who develop ESLD have essentially 0% survival at 5 years.4 Therefore, in SBS patients with ESLD, combined liver–intestine transplantation (LIT) is the only option for survival.5 United Network for Organ Sharing (UNOS) data indicate that waiting list mortality for intestinal transplant candidates greatly exceeds that for all other organ candidates.6 However, a subset analysis to determine whether the high waiting list mortality in the intestine candidate group is skewed by a specific candidate subset has not previously been performed. The purpose of this study was to define the waiting list mortality in the LIT candidate subset of intestinal transplant candidates and compare it with the waiting list mortality of the overall liver transplant candidate population.
P
Experimental Procedures
atients with short bowel syndrome (SBS) or other forms of intestinal failure require long-term total parenteral nutrition (TPN) to survive.1 Some of these From the *Department of Surgery, Northwestern University Medical School, Chicago, IL, and the †United Network for Organ Sharing, Richmond, VA. Address reprint requests to Jonathan P. Fryer, MD, Northwestern University Medical School, Department of Surgery, Division of Transplantation, 675 N. St. Clair, Galter Pavilion, Suite 17-200, Chicago, IL 60611. Telephone: 312-695-2125; FAX: 312-695-9194; E-mail: [email protected] Copyright © 2003 by the American Association for the Study of Liver Diseases 1527-6465/03/0907-0016$30.00/0 doi:10.1053/jlts.2003.50151
748
Data were obtained from the UNOS database. Waiting list data for potential intestinal transplant recipients since 1993 was collected, and was first reported in 2000 and updated in 2001.6 For the purposes of this study, further analyses were performed on the UNOS data to provide more detailed information on the intestine transplant candidate subsets, particularly those that also needed liver transplants. These data were compared with UNOS data on all liver transplant candidates. For each comparison group, the mortality rate was calculated as the number of deaths were divided by the total number of patient days of follow-up in that group. Mortality rates were compared using Fisher’s exact test,
Liver Transplantation, Vol 9, No 7 ( July), 2003: pp 748-753
749
High Mortality in Liver–Intestine Candidates
Table 1. Waiting List Mortality in Intestine and Liver Transplantation Candidate Subsets Before and After 1997 UNOS Status Modifications 01/01/93 to 06/30/97 Transplantation Candidate Type
Total Patients Listed*
Total Deaths
Liver and intestine only Intestine only Intestine/multivisceral All intestine All liver
249 76 38 363 29,514
69 5 7 81 3,299
07/01/97 to 12/31/01
Mortality Rate†
Total Patients Listed*
Total Deaths
Mortality Rate†
333.63 104.21 129.70 284.05 166.16
429 198 108 735 48,447
154 14 13 181 6,943
493.66‡ 118.12 128.51 323.75 121.06‡
*Total number of patients on the list at any time during the designated time period. †Number of deaths per 1,000 patient years waiting for a transplant. ‡P ⬍ .05 versus 01/01/93 to 06/30/97.
and differences were considered significant if P values were less than .05.
Results A more detailed analysis of UNOS data regarding intestinal waiting list mortality6 shows that since 1994, 86% of the deaths occurred in candidates listed for combined LITs (Table 1). An additional 8% of the deaths occurred in candidates listed for multivisceral transplants, most of whom needed both a liver and an intestine as part of their multivisceral graft. Only 7% of the intestine waiting list deaths occurred in candidates who needed isolated intestine transplants only. In every year since 1993, candidates needing LITs have never made up more than 1% of the annual liver transplant waiting list (Table 1). Since 1997, when significant changes in the UNOS status definitions for liver transplant candidates were implemented, the mortality rate has increased significantly in the LIT candidate subset. Conversely, a significant decrease in the overall liver transplant waiting list mortality has occurred since 1997 (Table 1). The annual mortality rate with LIT candidates has consistently exceeded that of the overall liver transplant waiting list since 1993 (Fig. 1). Although most liver transplantations have been performed in adults, the majority of intestinal transplantations have been performed in pediatric patients.5 Accordingly, most candidates on the intestine waiting list are pediatric.6 Therefore, to determine whether the overall waiting list mortality differences may be skewed by a higher mortality in the pediatric LIT candidate subset, in which donor–recipient size discrepancies and other issues severely limit access to both cadaveric and
living donor organs, we evaluated adult and pediatric patients independently. When the total numbers of waiting list deaths were evaluated for 1993 to 2001, 83% of the LIT candidate deaths occurred in the pediatric population, whereas only 8% of deaths in the overall liver candidate population were pediatric. When the pediatric candidate population (0 to 17 years) was evaluated independently, annual waiting list mortality rates were significantly higher for the LIT candidates for 1995 to 2001 inclusive (Fig. 2). In the adult candidate population (⬎17 years), annual waiting list mortality rates were significantly higher for the LIT candidates for
Figure 1. Comparison of annual waiting list mortality rates in all liver transplant (Liver), and combined Liver– intestine transplant (LIT) candidates for 1993 to 2001. Mortality rates were calculated as the total annual number of deaths occurring in patients on the waiting list divided by the combined waiting times (in years) for all patients on the list during that year ⴛ 1,000 (*, P < .05 compared with liver for the same year). Liver, LIT.
750
Fryer et al
Figure 2. Comparison of annual waiting list mortality rates in adult (>17 years) and pediatric (Ped)(0 to 17 years) liver transplant (Liver) and combined liver–intestine transplant (LIT) candidates for 1993 to 2001. Mortality rates were calculated as the total annual number of deaths occurring in patients on the waiting list divided by the combined waiting times (in years) for all patients on the list during that year ⴛ 1,000 (#, P < .05 compared with adult liver, and *, P < .05 compared with pediatric liver for the same year). 䊐 Liver-Adult, Liver-Ped, LIT-Adult, LIT-Ped.
1997 to 1999 inclusive (Fig. 2). However, when specific age groups were evaluated based on the candidate’s age at the time of listing, waiting list mortality in LIT candidates significantly exceeded that seen in the overall liver candidate population in all age groups (Fig. 3). These differences were most significant in neonates (⬍1 year) and adults (35 to 65 years). Mortality rates were further analyzed based on the candidate’s final UNOS status. UNOS mortality data analyzed on the basis of status were unavailable for candidates listed before 1995. Because there were several modifications made in the UNOS status definitions in 1997, data from that year were excluded and results were analyzed separately based on whether the status was assigned before (1995 to 1996 inclusive) or after (1998 to 2001 inclusive) 1997. Because of differences in the definition of status 1 in adult and pediatric patients, mortality rates for status 1 adult and pediatric candidate populations were evaluated separately. Almost all status 1 LIT waiting list deaths occurred in the pediatric candidate population, and this reflected a very low use of the status 1 category in adult LIT candidates. In fact, the total cumulative waiting time for all status 1 adult LIT candidates was 0.6 years. With the pediatric patients, after 1997, the mortality rates were significantly lower in the LIT can-
didates than in the overall liver transplant candidate population (Fig. 4). Otherwise, the data available for status 1 patients were insufficient for statistical analysis. For all other UNOS statuses (2A, 2B, 3, and 7), cumulative waiting list mortality data are shown for the time period following the 1997 modifications (i.e., 1998 to 2001; Fig. 5). In status 2A patients, mortality rates in the overall liver transplant candidate population exceeded those seen in LIT candidates, although inadequate numbers of LIT candidates precluded statistical analysis. Because the 2A status category was not available to pediatric patients, these data reflect only adult candidates. As with status 1, the status 2A category was used infrequently in adult LIT candidates, as is reflected by the total cumulative waiting time of 1.7 years for all adult LIT 2A candidates. In status 2B, 3, and 7 candidates, mortality rates in LIT candidates significantly exceeded that of the overall liver transplant candidate population.
Discussion Intestinal transplantation is the only therapeutic option for patients who are unable to sustain themselves nutritionally via enteral routes and are unable to tolerate
Figure 3. Comparison of waiting list mortality rates in liver transplant (Liver) and combined liver–intestine transplant (LIT) candidates in specific age groups from 1993 to 2001. Age group assignment was determined based on the candidate’s age at the time of listing. Agespecific mortality rates were calculated as the total annual number of deaths occurring in candidates on the waiting list in a specific age group divided by the combined waiting times (in years) for all candidates in that age group that were on the waiting list during the designated time period ⴛ 1,000 (* ⴝ P < .05 compared with liver for the same age group). Liver, LIT.
High Mortality in Liver–Intestine Candidates
Figure 4. Comparison of waiting list mortality rates in status 1 adult (>17 years) and pediatric (0 to 17 years) liver transplant (Liver) and combined liver–intestine transplant (LIT) candidates before (1995 to 1996) and after (1998 to 2001) 1997 modifications in UNOS status 1 criteria. Mortality rates were calculated as the total number of waiting list deaths occurring in a candidate subset divided by the combined waiting times (in years) for all patients in that subset ⴛ 1,000. All data was determined based on a candidate’s final UNOS status before death, transplantation, or year’s end. The total number of waiting list deaths occurring in each candidate subset is indicated in parentheses in the corresponding data bar (*, P < .05 compared with liver in the same time period). Liver (95-96), LIT (95-96), 䊐 Liver (98-01), LIT (98-01).
long-term TPN therapy. According to Medicare policy, impending liver failure is one of the criteria that defines failed TPN therapy, and therefore warrants consideration of intestinal transplantation. Although the development of irreversible liver disease is relatively uncommon in the entire population of patients who receive home TPN, in patients with very short guts (i.e., ⬍50 cm) who are destined to be permanently TPN-dependent, it occurs in up to 50%.3 The etiology of the liver disease in these patients has not been clearly established, but is likely multifactorial.3,7-10 Several factors that correlate with its occurrence include duration of TPN therapy,9 length of residual small bowel,3 and the presence of residual colon.3 Other factors that are implicated include bacterial overgrowth,8 composition of TPN macronutrients (i.e., lipids, carbohydrates),3 missing nutritional factors,7 and hepatotoxic factors in TPN.10 In SBS patients on TPN in whom ESLD develops, the 5-year survival is essentially 0%.4 Therefore, a com-
751
bined LIT is the only option for survival for these patients. Based on the overall world experience thus far, patient survival after combined liver–intestine transplantation is 63% at 1 year.5 If ESLD has not developed, simultaneous liver transplantation is not needed. In these circumstances, when the intestine is transplanted alone, 1-year patient survival is significantly better (70% to 80%).5 This survival discrepancy may be partly because life-threatening graft- or immunosuppression-related complications can be salvaged by graft removal and withdrawal of immunosuppressive therapy with isolated intestinal transplants, whereas this is not an option with LIT. Another factor is that LIT candidates are, by definition, much sicker going into their transplantation. Although it is therefore preferable to perform isolated intestinal transplantations before ESLD develops, many patients are not referred for intestinal transplantation evaluation until irreversible liver disease is already present. These patients who need LITs are placed on both the intestine and liver waiting lists, although their priority on the liver waiting list
Figure 5. Comparison of waiting list mortality rates in liver transplant (Liver) and combined liver–intestine transplant (LIT) candidates in all UNOS status categories (except status 1) after 1997 (1998 to 2001). Status assignment was based on the candidate’s final UNOS status before death, transplantation, or year’s end. Status-specific mortality rates were calculated as the total annual number of waiting list deaths occurring in a specific status category (i.e., 2A, 2B, 3, or 7) divided by the combined waiting times (in years) for all candidates in that status category that were on the list during the designated time period ⴛ 1,000. The total number of waiting list deaths occurring in each candidate subset is indicated in parentheses in the corresponding data bar (*, P < .05 compared with liver for the same UNOS status). Liver, LIT.
752
Fryer et al
Table 2. One-Year Patient Survival in Liver/Intestine Transplant Recipients Based on Their UNOS Status at the Time of Transplantation Liver-Intestine Transplant status 1-yr patient survival
Status 1
Status 2
Status 2A
Status 2B
Status 3
49%
56%
*
67%
60%
*Insufficient data available.
ultimately determines when they will be allocated organs. Until recently, all candidates, including those requiring LIT, were prioritized on the liver waiting list based on criteria derived from the Childs-TurcottePugh (CTP) score.11 Concerns regarding the effectiveness of these criteria to appropriately prioritize all candidates waiting for liver transplants led to their abandonment and replacement by the Model for EndStage Liver Disease/Pediatric End-Stage Liver Disease (MELD/PELD) system. However, the data presented here indicate that CTP-based UNOS prioritization was a much a better predictor of waiting list mortality in candidates waiting only for liver transplants than it was in the subset who needed an LIT. This underscores the fact that, in general, criteria used to predict waiting list mortality for liver transplant candidates may be irrelevant to the LIT candidate subset because they do not address the unique risk factors inherent in this population. The most important factor contributing to high mortality in LIT candidates is sepsis. Although all ESLD patients are at risk for fatal episodes of sepsis,12 SBS patients on TPN are particularly susceptible.13,14 Several factors may contribute to this. First, unlike most ESLD patients, LIT candidates need long-term central venous catheters (CVCs). Because of constant diarrhea and/or the need to maintain stomas and/or enteral tubes, SBS patients must be especially meticulous in their technique to avoid contamination of their CVCs with enteral organisms. Although their CVCs are susceptible to infection in the best of circumstances,15 they are even more likely to become infected as their clinical status deteriorates and their ability to provide meticulous CVC maintenance deteriorates. For the same reasons, the risk of urinary tract infections caused by contamination with enteral organisms is increased. Another factor that likely contributes to increased sepsis in LIT candidates is bacterial overgrowth and translocation.16,17 Although graft and patient survival results after LITs are currently inferior to the results seen after isolated
liver transplantations, they have steadily improved as the experience with the intestinal transplantation increases,5 and the early progress of LITs is not unlike that encountered with other solid organ transplantations.18,19 Since 1997, the vast majority of LIT candidate deaths (85%) have occurred in status 2B, 3, or 7 candidates (Fig. 2) and status 2B candidates accounted for 50% of these deaths. These data indicate that the high waiting list mortality risk in these LIT candidates was not fully recognized. However, the status 2B candidates who did survive long enough to receive a LIT did better posttransplant than LIT candidates in all other status categories (Table 2). This suggests that the best results may be achieved with more timely transplantation of LIT candidates, before they become too sick. The great discrepancy in mortality between LIT candidates and other liver transplant candidates clearly distinguishes this candidate subset from other liver transplant candidates. These data show that the recently abandoned UNOS liver prioritization system failed in appropriately prioritizing LIT candidates on the liver transplant waiting list. Although the recently implemented MELD/PELD system20 has not been in place long enough to allow a valid assessment, there is little reason to believe that it will do better, because it does not incorporate criteria that recognize the unique risk factors of the LIT subset. LIT candidates are a small but unique subset of liver transplant candidates who have had a disproportionately high mortality on the waiting list in all age groups. Changes in UNOS liver allocation policy that would increase their priority on the liver transplant waiting list will be an important step to rectifying this discrepancy. However, ultimate resolution of this problem will require additional measures, including increased research to determine the etiology of ESLD in these patients and increased awareness by physicians who manage TPN patients that high-risk (i.e., ⬍100 cm of short bowel) SBS patients need early referral to centers with advanced expertise in SBS management, including intestinal transplantation, to avoid the need for LIT.
High Mortality in Liver–Intestine Candidates
Acknowledgement The authors thank Renee Poland for her assistance in preparation of this manuscript, and Fred Rademaker, Sarah Fung, and Eric Edwards for their statistical analyses of the data. The data and analyses reported in the 2001 Annual Report of the U.S. Organ Procurement and Transplantation Network and the Scientific Registry of Transplant Recipients have been supplied by UNOS and University Renal Research and Education Association (URREA) under contract with HHS. The authors alone are responsible for reporting and interpreting these data.
References 1. Vanderhoof J, Langnas A. Short bowel syndrome in children and adults. Gastroenterology 1997;113:1767-1778. 2. Fryer J. The role of intestinal transplantation in the management of intestinal failure. Curr Gastroenterol Rep 3:334. 3. Cavicchi M, Beau P, Crenn P, Degott C, Messing B. Prevalence of liver disease and contributing factors in patients receiving home parenteral nutrition for permanent intestinal failure. Ann Intern Med 2000;132:525. 4. Chan S, McCowen K, Bistrian BR, Thibault A, Keane-Ellison M, Forse RA, et al. Incidence, prognosis, and etiology of endstage liver disease in patients receiving home parenteral nutrition. Surgery 1999;126:28. 5. Grant D. Intestinal transplantation: 1997 report of the international registry. Transplantation 1999;67:1061-1064. 6. 2001 Annual Report of the U.S. Organ Procurement and Transplantation Network and the Scientific Registry for Transplant Recipients. Transplant Data 1991–2000. Department of Health and Human Services, Health Resources and Services Administration, Office of Special Programs, Division of Transplantation, Rockville, MD; United Network for Organ Sharing, Richmond, VA; University Renal Research and Education Association, Ann Arbor, MI. 7. Buchman AL, Dubin MD, Moukarzel AA, Jenden DJ, Roch M, Rice KM, et al. Choline deficiency: A cause for hepatic steatosis during parenteral nutrition that can be reversed with intravenous choline supplementation. Hepatology 1995;22:1399-1403.
753
8. Meehan JJ, Georgeson KE. Prevention of liver failure in parenteral nutrition– dependent children with short bowel syndrome. J Pediatr Surg 1997;32:474-475. 9. Messing B, Crenn P, Beau P, Boutron-Ruault MC, Rambaud JC, Matuchansky C. Long-term survival and parenteral nutrition dependence in adult patients with short bowel syndrome. Gastroenterology 1999;117:1043-1050. 10. Iyer KR, Spitz L, Clayton P. New insight into mechanisms of parenteral nutrition–associated cholestasis: Role of plant sterols. J Pediatr Surg 1998;33:1-6. 11. Holman JM, Rikkers LF. Success of medical and surgical management of acute variceal hemorrhage. Am J Surg 1980;140:816. 12. Cirera I, Bauer TM, Navasa M, Vila J, Grande L, Taura P, et al. Bacterial translocation of enteric organisms in patients with cirrhosis. J Hepatol 2001;34:32-37. 13. Terra RM, Plopper C, Waitzberg DL, Cukier C, Santoro S, Martins JR, et al. Remaining small bowel length: Association with catheter sepsis in patients receiving home parenteral nutrition: evidence of bacterial translocation. World J Surg 2000;24: 1537-1541. 14. Okada Y, Klein NJ, Pierro A. Neutrophil dysfunction the cellular mechanism of impaired immunity during total parenteral nutrition in infancy. J Pediatr Surg 1999;34:242-245. 15. Moukarzel AA, Haddad I, Ament ME, Buchman AL, Reyen L, Maggioni A, et al. 230 patient years experience with home long-term parenteral nutrition in childhood: Natural history and live of central venous catheters. J Pediatr Surg 1994;29:13231327. 16. Pierro A, Van Saene HK, Donnell SC, Hughes J, Ewan C, Nunn AJ, et al. Microbial translocation in neonates and infants receiving long-term parenteral nutrition. Arch Surg 1996;131:176179. 17. Wells CL, Maddaus MA, Simmons RL. Proposed mechanisms for the translocation of intestinal bacteria. Rev Infec Dis 1998; 10:958-979. 18. Calne RY, Williams R. Orthotopic liver transplantation: the first 60 patients. Br Med J 1977;1:471-476. 19. Jamieson SW, Stinson EB, Shumway NE. Cardiac transplantation in 150 patients at Stanford University. Br Med J 1979;1:9395. 20. Pagliaro L. MELD: The end of the Child-Pugh classification? J Hepatol 2002;36:141-142.