Survival After Liver Transplantation in Patients With Hepatic Iron Overload: The National Hemochromatosis Transplant Registry

Background & Aims: Previous uncontrolled studies have suggested that patients with hepatic iron overload have a poor outcome after liver transplantation. We examined the effect of HFE mutations on posttransplantation survival in patients with hepatic iron overload. Methods: Two hundred sixty patients with end-stage liver disease and hepatic iron overload were enrolled from 12 liver transplantation centers. Hepatic iron concentration (HIC), hepatic iron index (HII), HFE mutation status, and survival after liver transplantation were recorded. Results: HFE-associated hemochromatosis (HH) defined as homozygosity for the C282Y (n = 14, 7.2%) mutation or compound heterozygosity for the C282Y/H63D (n = 11, 5.6%) mutation was identified in 12.8% of patients. Survival postliver transplantation was significantly lower among patients with HH (1-, 3-, and 5-year survival rates of 64%, 48%, 34%, respectively) compared with simple heterozygotes (C282Y/wt or H63D/wt) or wild-type patients. Patients with HH had a hazard ratio for death of 2.6 (P = .002) after adjustment for age, United Network for Organ Sharing status, year of transplantation, and either elevated HII or HIC. Non-HH patients with hepatic iron overload also had significantly decreased survival when compared with the overall population undergoing liver transplantation (OR = 1.4, 95% CI: 1.15–1.61, P < .001). Conclusions: One- and 5-year survivals after liver transplantation are significantly lower among patients with HFE-associated HH. Our data also suggest that hepatic iron overload may be associated with decreased survival after liver transplantation, even in patients without HH. Early diagnosis of hepatic iron overload using HFE gene testing and iron depletion prior to liver transplantation may improve posttransplantation survival, particularly among patients with HH. Background & Aims: Previous uncontrolled studies have suggested that patients with hepatic iron overload have a poor outcome after liver transplantation. We examined the effect of HFE mutations on posttransplantation survival in patients with hepatic iron overload. Methods: Two hundred sixty patients with end-stage liver disease and hepatic iron overload were enrolled from 12 liver transplantation centers. Hepatic iron concentration (HIC), hepatic iron index (HII), HFE mutation status, and survival after liver transplantation were recorded. Results: HFE-associated hemochromatosis (HH) defined as homozygosity for the C282Y (n = 14, 7.2%) mutation or compound heterozygosity for the C282Y/H63D (n = 11, 5.6%) mutation was identified in 12.8% of patients. Survival postliver transplantation was significantly lower among patients with HH (1-, 3-, and 5-year survival rates of 64%, 48%, 34%, respectively) compared with simple heterozygotes (C282Y/wt or H63D/wt) or wild-type patients. Patients with HH had a hazard ratio for death of 2.6 (P = .002) after adjustment for age, United Network for Organ Sharing status, year of transplantation, and either elevated HII or HIC. Non-HH patients with hepatic iron overload also had significantly decreased survival when compared with the overall population undergoing liver transplantation (OR = 1.4, 95% CI: 1.15–1.61, P < .001). Conclusions: One- and 5-year survivals after liver transplantation are significantly lower among patients with HFE-associated HH. Our data also suggest that hepatic iron overload may be associated with decreased survival after liver transplantation, even in patients without HH. Early diagnosis of hepatic iron overload using HFE gene testing and iron depletion prior to liver transplantation may improve posttransplantation survival, particularly among patients with HH. Hereditary hemochromatosis (HH) is characterized by iron overload in multiple organs, in particular the liver, heart, pancreas, joints, and endocrine glands.1Whittington C.A. Kowdley K.V. Review article haemochromatosis.Aliment Pharmacol Ther. 2002; 16: 1963-1975Crossref PubMed Scopus (56) Google Scholar Most cases of HH are associated with mutations in the HFE gene, which encodes a nonclassical HLA class I protein. Two mutations account for the overwhelming majority (85%) of cases of phenotypic HH among American patients: the homozygous C282Y mutation (C282Y+/+) or the compound heterozygous C282Y/H63D mutation (C282Y/H63D).2Morrison E.D. Brandhagen D.J. Phatak P.D. Barton J.C. Krawitt E.L. El-Serag H.B. Gordon S.C. Galan M.V. Tung B.Y. Ioannou G.N. Kowdley K.V. Serum ferritin level predicts advanced hepatic fibrosis among U.S. patients with phenotypic hemochromatosis.Ann Intern Med. 2003; 138: 627-633Crossref PubMed Scopus (154) Google Scholar Patients with cirrhosis because of HH have an increased risk of hepatocellular carcinoma and decompensated liver disease.1Whittington C.A. Kowdley K.V. Review article haemochromatosis.Aliment Pharmacol Ther. 2002; 16: 1963-1975Crossref PubMed Scopus (56) Google Scholar, 3Kowdley K.V. Hassanein T. Kaur S. Farrell F.J. Van Thiel D.H. Keeffe E.B. Sorrell M.F. Bacon B.R. Weber Jr, F.L. Tavill A.S. Primary liver cancer and survival in patients undergoing liver transplantation for hemochromatosis.Liver Transpl Surg. 1995; 4: 237-241Crossref Scopus (102) Google Scholar Survival among HH patients with cirrhosis is significantly decreased, even with iron depletion therapy via phlebotomy, primarily because of an increased risk of hepatocellular carcinoma.4Niederau C. Fischer R. Sonnenberg A. Stremmel W. Trampisch H.J. Strohmeyer G. Survival and causes of death in cirrhotic and in noncirrhotic patients with primary hemochromatosis.N Engl J Med. 1985; 313: 1256-1262Crossref PubMed Scopus (941) Google Scholar, 5Adams P.C. Speechley M. Kertesz A.E. Long-term survival analysis in hereditary hemochromatosis.Gastroenterology. 1991; 101: 368-372Abstract PubMed Google Scholar, 6Niederau C. Fischer R. Purschel A. Stremmel W. Haussinger D. Strohmeyer G. Long-term survival in patients with hereditary hemochromatosis.Gastroenterology. 1996; 110: 1107-1119Abstract Full Text Full Text PDF PubMed Scopus (758) Google Scholar Liver transplantation has been used to treat end-stage liver disease since the initial report from Pittsburgh of 6 patients with hemochromatosis. All patients had undergone iron depletion prior to transplantation, and all patients had survived 6 months or more after liver transplantation.7Pillay P. Tzoracoleftherakis E. Tzakis A.G. Kakizoe S. Van Thiel D.H. Starzl T.E. Orthotopic liver transplantation for hemochromatosis.Transplant Proc. 1991; 23: 1888-1889PubMed Google Scholar A subsequent report from a large health care financing administration database, however, suggested that hemochromatosis was associated with a significantly lower survival after liver transplantation.8Kilpe E. Krakauer H. Wren R.E. An analysis of liver transplant experience from 37 transplant centers as reported to Medicare.Transplantation. 1993; 56: 554-561Crossref PubMed Scopus (142) Google Scholar One- and 5-year survivals among patients with a pretransplantation diagnosis of “hemochromatosis” were 54% and 43%, respectively, which was lower than for patients with alcoholic or viral cirrhosis. A major limitation of the study was the lack of a definitive diagnosis of HLA-linked hemochromatosis in this cohort, possibly because there is only 1 ICD-9 code available to describe all causes of iron overload, regardless of etiology. This study included pediatric patients and patients with secondary iron overload in the hemochromatosis group. A subsequent study enrolled 37 patients from 5 liver transplantation centers with a presumed diagnosis of hereditary hemochromatosis.3Kowdley K.V. Hassanein T. Kaur S. Farrell F.J. Van Thiel D.H. Keeffe E.B. Sorrell M.F. Bacon B.R. Weber Jr, F.L. Tavill A.S. Primary liver cancer and survival in patients undergoing liver transplantation for hemochromatosis.Liver Transpl Surg. 1995; 4: 237-241Crossref Scopus (102) Google Scholar The diagnosis was restricted to patients with unexplained hepatic iron overload in a pattern consistent with hemochromatosis or a known pretransplantation diagnosis of hemochromatosis based on hepatic iron concentration (HIC) >4000 μg/g dry weight (approximately 70 μmol/g) or hepatic iron index (HII) (HII = HIC [μmol/g]/age (years)) >1.9. One- and 5-year survivals in this cohort3Kowdley K.V. Hassanein T. Kaur S. Farrell F.J. Van Thiel D.H. Keeffe E.B. Sorrell M.F. Bacon B.R. Weber Jr, F.L. Tavill A.S. Primary liver cancer and survival in patients undergoing liver transplantation for hemochromatosis.Liver Transpl Surg. 1995; 4: 237-241Crossref Scopus (102) Google Scholar were disappointingly similar to that reported by Kilpe et al.8Kilpe E. Krakauer H. Wren R.E. An analysis of liver transplant experience from 37 transplant centers as reported to Medicare.Transplantation. 1993; 56: 554-561Crossref PubMed Scopus (142) Google Scholar This cohort had a higher than expected prevalence of HCC; however, infections and cardiac complications accounted for most deaths.3Kowdley K.V. Hassanein T. Kaur S. Farrell F.J. Van Thiel D.H. Keeffe E.B. Sorrell M.F. Bacon B.R. Weber Jr, F.L. Tavill A.S. Primary liver cancer and survival in patients undergoing liver transplantation for hemochromatosis.Liver Transpl Surg. 1995; 4: 237-241Crossref Scopus (102) Google Scholar, 9Tung B.Y. Farrell F.J. McCashland T.M. Gish R.G. Bacon B.R. Keeffe E.B. Kowdley K.V. Long-term follow-up after liver transplantation in patients with hepatic iron overload.Liver Transpl Surg. 1999; 5: 369-374Crossref PubMed Scopus (43) Google Scholar These early studies examining outcomes after liver transplantation among patients with hepatic iron overload presumed secondary to hemochromatosis were completed prior to the era of HFE gene testing. Subsequent studies, albeit with small sample sizes, have demonstrated that most patients with hepatic iron overload in the setting of end-stage cirrhosis do not appear to have HLA-linked or classic hereditary hemochromatosis because the majority of such patients lack the characteristic C282Y homozygous or C282Y/H63D compound heterozygous genotype.10Deugnier Y. Turlin B. le Quilleuc D. Moirand R. Loreal O. Messner M. Meunier B. Brissot P. Launois B. A reappraisal of hepatic siderosis in patients with end-stage cirrhosis practical implications for the diagnosis of hemochromatosis.Am J Surg Pathol. 1997; 21: 669-675Crossref PubMed Scopus (109) Google Scholar, 11Brandhagen D.J. Alvarez W. Therneau T.M. Kruckeberg K.E. Thibodeau S.N. Ludwig J. Porayko M.K. Iron overload in cirrhosis-HFE genotypes and outcome after liver transplantation.Hepatology. 2000; 31: 456-460Crossref PubMed Scopus (81) Google Scholar, 12Stuart K.A. Fletcher L.M. Clouston A.D. Lynch S.V. Purdie D.M. Kerlin P. Crawford D.H. Increased hepatic iron and cirrhosis no evidence for an adverse effect on patient outcome following liver transplantation.Hepatology. 2000; 32: 1200-1207Crossref PubMed Scopus (47) Google Scholar Hepatic iron overload in the setting of end-stage liver disease appears most common among patients with chronic hepatitis C and alcoholic liver disease but has also been described among patients with other types of end-stage liver disease.13Cotler S.J. Bronner M.P. Press R.D. Carlson T.H. Perkins J.D. Emond M.J. Kowdley K.V. End-stage liver disease without hemochromatosis associated with elevated hepatic iron index.J Hepatol. 1998; 29: 257-262Abstract Full Text PDF PubMed Scopus (79) Google Scholar, 14Ludwig J. Hashimoto E. Porayko M.K. Moyer T.P. Baldus W.P. Hemosiderosis in cirrhosis a study of 447 native livers.Gastroenterology. 1997; 112: 882-888Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar Some authors have suggested that hepatic iron overload is associated with a poor outcome after liver transplantation regardless of the presence or absence of HFE mutations.12Stuart K.A. Fletcher L.M. Clouston A.D. Lynch S.V. Purdie D.M. Kerlin P. Crawford D.H. Increased hepatic iron and cirrhosis no evidence for an adverse effect on patient outcome following liver transplantation.Hepatology. 2000; 32: 1200-1207Crossref PubMed Scopus (47) Google Scholar, 14Ludwig J. Hashimoto E. Porayko M.K. Moyer T.P. Baldus W.P. Hemosiderosis in cirrhosis a study of 447 native livers.Gastroenterology. 1997; 112: 882-888Abstract Full Text Full Text PDF PubMed Scopus (242) Google Scholar Based on these data, some liver transplantation programs have been excluding such patients from liver transplantation, particularly if they have evidence of cardiac iron overload.15Brandhagen D.J. Liver transplantation for hereditary hemochromatosis.Liver Transpl. 2001; 7: 663-672Crossref PubMed Scopus (40) Google Scholar However, others have argued that hepatic iron overload per se is not associated with a poor outcome after liver transplantation in patients without hemochromatosis.12Stuart K.A. Fletcher L.M. Clouston A.D. Lynch S.V. Purdie D.M. Kerlin P. Crawford D.H. Increased hepatic iron and cirrhosis no evidence for an adverse effect on patient outcome following liver transplantation.Hepatology. 2000; 32: 1200-1207Crossref PubMed Scopus (47) Google Scholar A recent study of 22 subjects with suspected HH defined as homozygosity for the C282Y mutation (n = 17) or hepatic iron overload associated with either the HLA A3B7 haplotype or a family history of iron overload showed relatively poor outcome after liver transplantation with 1-, 3-, and 5-year survivals of 72%, 62%, and 55%, respectively.16Crawford D.H. Fletcher L.M. Hubscher S.G. Stuart K.A. Gane E. Angus P.W. Jeffrey G.P. McCaughan G.W. Kerlin P. Powell L.W. Elias E.E. Patient and graft survival after liver transplantation for hereditary hemochromatosis implications for pathogenesis.Hepatology. 2004; 39: 1655-1662Crossref PubMed Scopus (67) Google Scholar Increased risk of posttransplantation mortality appeared to be related to recurrent HCC in many subjects. However, this study did not control for other risk factors for decreased survival after transplantation such as United Network for Organ Sharing (UNOS) status, age, year of transplantation, or severity of hepatic iron overload. The primary goals of the current study were (1) to examine the prevalence of HFE mutations in a large multicenter cohort of United States liver transplant recipients with known or suspected hepatic iron overload and (2) to examine the relationship between HFE genotype and survival after liver transplantation. Secondary goals were to examine the causes of death after liver transplantation among patients with hepatic iron overload and to examine the relationship between HIC, HII, and survival in this cohort. Twelve liver transplantation centers participated in this study. The complete list of centers and participating investigators is shown in Appendix 1. Patients who had undergone their first liver transplantation prior to 1996 were eligible to enter this study. To be included, patients were required to have one of the following: a known diagnosis of hemochromatosis or hepatic iron overload prior to liver transplantation based on one of the following criteria: compatible hepatic iron stain (as ≥2+ diffuse, multifocal, iron staining in hepatocytes with or without stainable iron in reticuloendothelial cells); HIC ≥4000 μg/g dry weight (approximately 70 μmol/g) or HII ≥1.9; ≥4 grams of iron removed by quantitative phlebotomy or previously unsuspected hepatic iron overload (as defined above) found in the liver explant in the absence of a known cause of iron overload, such as blood transfusion, iron-loading anemia, homozygous β-thalassemia, or iatrogenic iron supplementation (in the absence of iron deficiency). The subjects in the study were ascertained as follows: known or suspected diagnosis of hemochromatosis prior to transplantation based on HII or HIC (n = 46, 17.7%) or iron overload discovered at liver transplantation without previous suspicion of iron overload (n = 214, 82.3%). Because the inclusive time period of the study was prior to the identification of the HFE gene, HFE mutation status was unknown in the study subjects at the time of transplantation. Paraffin blocks from liver explants were sent to the University of Washington, and HIC and HII measurement and HFE genotyping were performed using previously described techniques.17Emond M.J. Bronner M.P. Carlson T.H. Lin M. Labbe R.F. Kowdley K.V. Quantitative study of the variability of hepatic iron concentrations.Clin Chem. 1999; 45: 340-346PubMed Google Scholar, 18Raaka S. Huehnergarth K.V. Kowdley K.V. Bronner M.P. Tissue typing for HFE mutations.Hepatology. 2002; 35: 977-978Crossref PubMed Scopus (2) Google Scholar Data were collected at the participating centers using a customized Microsoft Access database (Microsoft Corporation, Redmond, WA) and sent electronically to the central database at the University of Washington. The following data were collected on all subjects: age at the time of transplantation, sex, UNOS status (using the old system of 1, 2a, 2b, and 3), HFE genotype, date of last follow-up, year of transplantation, and transplantation center. The following key data were collected wherever possible (given that this was a retrospective study, data were not available in all subjects): HIC and HII, presence or absence of hepatocellular carcinoma, phlebotomy therapy pretransplantation, pretransplantation diagnosis of iron overload, and cause of death. Exclusion criteria included the following: retransplantation, inability to perform HFE genotyping (lack of tissue or inability to obtain DNA from paraffin tissue [<5% of subjects]), or incomplete key data provided by the participating center. Inability to perform HFE genotyping most commonly occurred because subjects were deceased and because the retrospective nature of the study precluded HFE mutation analysis in some cases. The study was approved by the institutional review boards of all participating centers that enrolled patients into the study. Cumulative survival estimates and survival curves for the entire group and subgroups were calculated using the Kaplan—Meier method.19Kaplan E.L. Meier P. Nonparametric estimation from incomplete observations.J Am Stat Assoc. 1958; 53: 457-481Crossref Scopus (46749) Google Scholar The curves were compared between groups using the log-rank test. The 1, 3, and 5- year survivals in the groups were tabulated along with their standard errors based on Greenwood’s formula.20Hosmer D.W. Lemeshow S. Applied survival analysis. John Wiley & Sons, New York2002Google Scholar We compared mean HII and HIC among the genetic groups using normal-based ANOVA. Before carrying out the ANOVA, HII and HIC were transformed as log (1 + HII) and log (30 + HIC), respectively, to improve normality of distribution. The proportion of cancer cases in the genetic groups was compared using Fisher exact test. We developed a multivariate model for survival using the Cox proportional hazards method. Because the Kaplan—Meier survival curves for the UNOS groups suggested nonproportionality of hazards over time, we stratified on the UNOS categories in the Cox models rather than including UNOS as a covariate. We also compared the survival among the non-HH patients in this study (defined below as groups 3 + 4 + 5, n = 235) to survival among all adult patients undergoing a first liver transplantation at the 12 participating transplantation centers between 1990 and 1996 (n = 5493) to examine whether survival in the non-HH patients with hepatic iron overload was lower than the remaining patients undergoing a first liver transplantation. The statistical significance of the differences in survival rates between groups 3 + 4 + 5 (n = 235) and the remaining patients (n = 5493) at each time point (1, 3, 5 years) is based on the χ2 test, using the baseline cohort sample sizes and the estimated Kaplan—Meier survival proportions in the 2 samples. Odds ratios and confidence intervals are based on binomial proportions. This approach was chosen because of lack of information on standard errors of survival proportions in the cohort. The P values and confidence intervals are apt to be slightly anticonservative because of censoring in both groups. The characteristics of the study population are shown in Table 1. The mean age of the patients was 51 years, and 80% were male. HFE genotyping was available in 195 of the 260 patients (75%). The prevalence of HFE mutations in those with genotype available was 46%. Twenty-five patients (13% of those genotyped) had HH, as defined by presence of homozygous C282Y (n = 14) or compound heterozygous C282Y/H63D mutations (n = 11). Fifty-nine (30%) patients were “simple” heterozygotes, and 106 were wild type for HFE (n = 106, 54%). Among HFE heterozygotes, 23 (12%) were C282Y heterozygotes, and 36 (18%) were H63D heterozygotes; 4 (2%) were H63D homozygotes. Among patients with available data on pretransplantation diagnoses, 46 patients (18%) were classified as having hemochromatosis, 32 (12%) as having hepatitis C, and 18 (7%) as having alcoholic liver disease; the etiology of liver disease was unspecified or listed as cryptogenic in 60 patients (23%). Only 20 of 46 patients diagnosed as having hemochromatosis prior to liver transplantation (43%) were confirmed to have HH based on HFE genotyping. Data on presence or absence of liver cancer were present in 252 of 260 patients (97%). Liver cancer was present in 29 of 252 patients (11.5%). There was a higher prevalence of liver cancer among C282Y homozygotes (29%) compared with simple heterozygotes (13%) or wild-type patients (10%), although the difference was not statistically significant (P = .2).Table 1Characteristics of the Study PopulationVariableNumberMean NSD or %Sex254F5220%M20280%Age at transplantation25950.610.2Age group2590–39 years3413%40–49 years8533%50–59 years9839%60+ years4216%Mean HII (μmol/g/age[y])1972.32.1HII ≤1.99749%HII >1.910051%Mean HIC (μmol/g dry weight)196117111HIC ≤70 μmol/g7036%HIC >70 μmol/g12664%Status260Alive at last follow-up14756.5%Deceased11343.5%Years post transplantation2603.93.2UNOS23212310%2a3716%2 or 2b7934%3 or 49340%HFE mutation status195C282Y homozygote147%C282Y/H63D CPD Het116%C282Y heterozygote2312%H63D heterozygote3619%H63D homozygotes42%Wild type10654%NOTE. Total N = 260. Open table in a new tab NOTE. Total N = 260. Data on HIC or HII were available in 197 of 260 (76%) patients. Multiple measurements of HIC were available in 11%; the mean weight per sample was 2.77 mg (92% of samples were >1-mg dry weight. HIC was >70 μmol/g dry weight in 126 of 197 (64%) patients; HII was >1.9 in 100 of 197 (51%) patients. The actual 1-year survival by HFE genotype is shown in Table 2. Cumulative survival differed significantly among genotypes (P = .02) with 1-, 3-, and 5-year survivals among C282Y homozygotes (64%, 43%, 32%, respectively) and C282Y/H63D compound heterozygotes (64%, 55%, 33%, respectively) lower than among HFE heterozygotes (84%, 74%, 61%, respectively) and wild-type patients (78%, 69%, 65%, respectively). There was no significant difference in cumulative survival between C282Y heterozygotes and C282Y/H63D compound heterozygotes (P = .99). There was also no significant difference in cumulative survival between C282Y heterozygotes (n = 23; 1-, 3-, and 5-year rates of 87%, 77%, 66%, respectively), H63D heterozygotes (n = 36; 81%, 72%, 56%, respectively) and H63D homozygotes (n = 4; 100%, 75%, 75%, respectively; P = .8). Therefore, the C282Y homozygous and C282Y/H63D compound heterozygous genotypes were combined in the survival analysis as were the HFE heterozygous and wild-type groups. The overall cumulative survival for the entire cohort is shown in Figure 1. The survival curves for the HH, simple HFE heterozygote, and wild-type groups are shown in Figure 2. The combined survival curve for the C282 homozygote and C282Y/H63D compound heterozygote group compared with simple HFE heterozygote and wild-type groups is shown in Figure 3. Patients in whom HFE genotype was unavailable (classified as missing in Table 2) had 1-, 3-, and 5-year survivals that were comparable with simple heterozygotes and wild-type patients.Table 2Relationship Between HFE Mutations Status and Post Transplant SurvivalHFE statusN%1 year (%)3 years (%)5 years (%)SurvivalSESurvivalSESurvivalSEGroup 1: C282Y homozygoteaP = .02.145641343133214Group 2: Compound hetaP = .02.114641555153315Group 3: Other HFEbC282/wt, H63D/wt, H63D/H63D.6324845746617Group 4: Wild typeaP = .02.10741784695655Group 5: MissingaP = .02.6525815716617Group 1 + 2cP = .0007.2510641048103410Group 3 + 4cP = .0007.17065803714644Comparison to Group 1 + 2:dP = 0008.Group 3 + 4 + 5dP = 0008.23590813713633Group 3a: C282Y heteP = .8.2398777796611Group 3b: H63D heteP = .8.3614817728569Group 3c: H63D homozygoteeP = .8.42100075227522NOTE. Significant differences occur because of the contrast in survival between groups 1 and 2 vs 3, 4, and 5.*P value based on the log-rank test. The null hypothesis is that all of the HFE groups compared have an identical survival rate over time. Survival rates at 1, 3, and 5 years are presented as illustrative, but P values are based on the continuous survival experience throughout the entire observation period.a P = .02.b C282/wt, H63D/wt, H63D/H63D.c P = .0007.d P = 0008.e P = .8. Open table in a new tab Figure 2Kaplan—Meier estimates for survival after liver transplantation among patients based on HFE genotype. C282Y +/+, C282Y homozygous; CPD Het, C282Y/H63D compound heterozygote; Simple het, C282Y/wt or H63D/wt; Wt: wt/wt.View Large Image Figure ViewerDownload (PPT)Figure 3Kaplan—Meier estimates for survival after liver transplantation among patients with HH (defined as C282Y homozygous or C282Y/H63D compound heterozygous [group 1 and group 2]) compared with HFE heterozygotes (C282Y/wt or H63D wt) and patients wild type for HFE [group 3 and group 4].View Large Image Figure ViewerDownload (PPT) NOTE. Significant differences occur because of the contrast in survival between groups 1 and 2 vs 3, 4, and 5. *P value based on the log-rank test. The null hypothesis is that all of the HFE groups compared have an identical survival rate over time. Survival rates at 1, 3, and 5 years are presented as illustrative, but P values are based on the continuous survival experience throughout the entire observation period. We also compared the estimated 1-, 3-, and 5-year survivals between the non-HH patients in this study (groups 3 + 4 + 5 [n = 235]) and the remaining patients undergoing a first liver transplantation at the participating 12 centers during the time period of the study (n = 5493). As shown in Table 3, non-HH patients with hepatic iron overload undergoing liver transplantation appear to have significantly decreased 5-year survival after liver transplantation (63% vs 72%, respectively P = .003, OR = 1.51) compared with the remaining patients.Table 3Comparison of the Survival Among non-HH Patients With Iron Overload to Survival of Other Patients at the 12 Transplant Centers During the Period of 1990–1996N1 year (%)3 years (%)5 years (%)SurvivalSESurvivalSESurvivalSEGroups 3 + 4 + 5235813713633Whole population minus 260 patients in the NHTR study5493847672OR (95% CI)aOdds ratio for death for the 3 + 4 + 5 group relative to the group of the whole population minus 260 in the study.1.25 (0.90–1.75)1.30 (0.97–1.73)1.51 (1.56–1.99)P valuebTesting the null-hypothesis that the survival rates are the same in both groups..2.08.003a Odds ratio for death for the 3 + 4 + 5 group relative to the group of the whole population minus 260 in the study.b Testing the null-hypothesis that the survival rates are the same in both groups. Open table in a new tab The effect of hepatic iron content on survival after liver transplantation was also examined in detail. HIC and HII were not significantly different among C282Y homozygotes and C282Y/H63D compound heterozygotes compared with patients in other genotype groups, as shown in Table 4. There was no significant difference in survival between patients with HIC >70-μmol/g dry weight compared with those with HIC ≤70-μmol/g dry weight or between patients with HII >1.9 compared with those with HII ≤1.9, although overall survival was marginally better among patients with HIC ≤70 μmol/g and HII ≤1.9 (Figure 4A and 4B). HIC was also not significantly associated with survival when examined as a continuous variable (data not shown).Table 4Hepatic Iron Concentration and Hepatic Iron Index Among Different HFE Genotype Groups (for Patients With Available HFE Genotype)HFE statusHII (HIC/age[y])HIC (μmol/g dry wt)NMeanSDMedianMeanSDMedianP = .9aP value based on an F test comparing log(1+HII) or log(30+HIC) between HFE groups.P = .7aP value based on an F test comparing log(1+HII) or log(30+HIC) between HFE groups.Group 1: C282Y homozygote133.42.93.3196168212Group 2: Compound het102.31.22.610961131Group 3: Other HFEb(C282/wt, H63D/wt, H63D/H63D).572.42.22.111710497Group 4: Wild type98c97 observations available for HIC (age missing for one of the subjects).2.42.11.911911493P = .5aP value based on an F test comparing log(1+HII) or log(30+HIC) between HFE groups.P = .4aP value based on an F test comparing log(1+HII) or log(30+HIC) between HFE groups.Group 1 + 2232.92.42.8158118132Group 3 + 41552.42.22.013811095P = .3aP value based on an F test comparing log(1+HII) or log(30+HIC) between HFE groups.P = .3aP value based on an F test comparing log(1+HII) or log(30+HIC) between HFE groups.Group 3a: C282Y het213.23.22.2157148106Group 3b: H63D het332.01.32.1965897Group 3c: H63D homozygote31.50.71.3704065NOTE. D