Liver transplantation for familial amyloidotic polyneuropathy: Impact on Swedish patients' survival



Liver transplantation (LTx) for familial amyloidotic polyneuropathy (FAP) is an accepted treatment for this fatal disease. However, the long-term outcome with respect to that of nontransplanted patients has not been fully elucidated. The aim of this study was to compare the long-term survival of Swedish LTx FAP patients with that of historical controls, especially with respect to the age at onset of the disease and gender. In order to evaluate the outcome of LTx as a treatment for FAP, survival was calculated from the onset of disease. One hundred forty-one FAP patients, 108 transplanted and 33 not transplanted, were included in the study. Significantly increased survival was noted for LTx patients in comparison with controls. The outcome was especially favorable for those with an early onset of the disease (age at onset < 50 years) in comparison with early-onset controls (P < 0.001). In contrast, no significant difference for late-onset cases (≥50 years) was found. Transplanted late-onset females had significantly improved survival in comparison with transplanted late-onset males (P = 0.02). We were unable to find significant differences in survival between patients with long (≥7 years) or short (<7 years) disease duration at transplantation. The survival of male patients with late-onset disease appeared not to improve with LTx. LTx is an efficacious treatment for improving the survival of early-onset FAP patients. Further studies are needed to analyze the cause of the poorer outcome for late-onset male patients. Liver Transpl 15:1229–1235, 2009. © 2009 AASLD.

Familial amyloidotic polyneuropathy (FAP) is an inherited, fatal, systemic amyloidosis that is caused by a point mutation in the transthyretin (TTR) gene. The most prevalent amyloidgenic transthyretin (ATTR) variant is that with a substitution of methionine for valine at residue 30 (ATTR Val30Met). The disease is found worldwide, with endemic areas in Portugal, Sweden, and Japan. The course of the disease is steadily progressive, and life expectancy has been reported to be approximately 9 to 13 years from the symptomatic onset of disease.1–3

Because the liver synthesizes more than 90% of circulating TTR, liver transplantation (LTx) should halt the production of the amyloidogenic variant TTR and thus cease the production of amyloid. The first LTx procedures in the world for FAP were performed in Sweden in 1990, and the favorable clinical outcome for the first 4 transplanted patients prompted several centers to perform LTx on FAP patients.4 LTx for FAP ATTR Val30Met is now widely performed to halt the progression of the clinical manifestations of the disease. According to the Familial Amyloidotic Polyneuropathy World Transplant Register (, approximately 100 FAP patients undergo LTx each year.

LTx is still the only available treatment proven to halt the progression of the disease. We previously reported increased survival for FAP patients who underwent transplantation at an early stage of their disease and who had good nutritional status.5, 6 In our previous study, the age of the transplanted patients had a significant impact on long-term survival (>6 months), whereas malnutrition at the time of transplantation was related to short-term survival (<6 months).

The aim of the present study is to elucidate the long-term outcome in Swedish LTx FAP patients, especially with respect to the impact of gender, age at onset of the disease, and duration of the disease at LTx. For that purpose, the survival from onset of disease is compared with that of historical nontransplanted controls.


ATTR, amyloidgenic transthyretin; FAP, familial amyloidotic polyneuropathy; LTx, liver transplantation; mBMI, modified body mass index; TTR, transthyretin.


One hundred forty-one patients were included in the study: 108 transplanted patients and 33 nontransplanted historical controls. In all cases, the diagnosis was based on typical clinical symptoms and findings of amyloid deposits in biopsy specimens and further confirmed by identification of an ATTR mutation by genetic testing. One hundred five of the transplanted patients and all of the controls carried the ATTR Val30Met mutation; the remaining 3 LTx patients carried other mutations (ATTR Phe33Leu, Ala45Ser, or Leu55Gln).

All LTx procedures were performed from April 1, 1990 to June 1, 2008; thus, all surviving patients were followed after the transplantation procedure for more than 6 months. As in our previous studies, 1 transplanted recipient was excluded because she had an unusual history of more than 25 years of disease before she underwent LTx, and she died a few days after the procedure. Otherwise, all Swedish transplanted patients who had been evaluated at, or in collaboration with, our center (2 patients) were included in the study. Fewer than 5 transplanted Swedish patients were not evaluated at our center; therefore, our study is representative of the Swedish transplant results.

LTx was performed at Karolinska University Hospital (Huddinge, Sweden) or Sahlgrenska University Hospital (Göteborg, Sweden). The patients' clinical details are listed in Table 1. All patients had been evaluated at the Department of Medicine of Umeå University Hospital (Umeå, Sweden), except for 2 patients who were evaluated at their local hospital in collaboration with the Department of Medicine of Umeå University Hospital.

Table 1. Data for the Patients Included in the Study
 LTx PatientsControls
  • Abbreviation: LTx, liver transplantation.

  • *

    The early series consisted of patients transplanted before 1996.

  • P < 0.05 between transplanted and nontransplanted (controls) patients.

Number of patients (early series/late series)*108 (32/76)33
Gender (males/females)60/4816/17
Age at onset [years; median (range)]46 (22–68)53 (29–65)
Number with age at onset ≥ 50 (males/females)45 (30/14)20 (12/8)
Duration of disease at LTx [years; median (range)]3 (0–12)
Follow-up from onset of disease (years)10 (2–24)12 (5–20)†
Number of deceased2832

The nontransplanted control group consisted of 30 patients; all of them had been examined at the Department of Medicine of Umeå University Hospital prior to the establishment of our transplant program, and they were thus never considered for transplantation. In addition, 3 patients who were offered LTx but declined were included in the control group, which consisted of 33 patients (16 males and 17 females). Thus, none of the patients in the control group had been rejected for transplantation because of advanced FAP disease. Patients older than 65 years at the onset of the disease were also excluded because patients with an onset of disease at this age rarely are accepted for LTx. Clinical data for the controls are summarized in Table 1. One patient of the control is alive for 8 years from the onset of the disease. The causes of death for 30 of the 32 deceased controls were related to FAP disease (28 suffered from kidney failure, malnutrition and infection, and heart failure), accident (a severely disabled patient), and postoperative bleeding after heart surgery (1 of the patients who had been offered LTx); for the other 2 patients, the exact cause of death was unknown.

Medical records were scrutinized for information regarding the duration of disease and nutritional status. Nutritional status was evaluated by the modified body mass index (mBMI), which is calculated by multiplication of the body mass index by the serum albumin concentration in order to compensate for edema.1 The transplanted patients were divided into 2 series: early and late. The early series consisted of patients transplanted from 1990 until 1995. They were unselected with respect to FAP disease. Thus, severely incapacitated patients with a depleted nutritional status and severe autonomic disturbances were accepted for transplantation. However, it became apparent that survival for nutritionally depleted LTx patients was inferior to that for non-LTx patients,6 so patients in the late series were selected primarily according to their nutritional status as measured by the mBMI. The late series consisted of patients from 1996 to 2008 who generally had better nutritional status with an mBMI above 600 at the time of transplantation.

To investigate the impact of age at onset on survival, patients and controls were divided into 2 groups: those with an early onset of the disease (<50 years) and those with a late onset of the disease (≥50 years). Fifty years of age has been used to separate early-onset and late-onset cases in several previous studies.6–8 In addition, the impact of gender was analyzed.

Differences between groups were tested with the chi-square test and Mann-Whitney U test. Correlation was tested with Spearman's rank correlation coefficient. Survival analysis was performed by Kaplan-Meier analysis, and differences between group survivals were tested with the Cox-Mantel test. For statistical analysis, SPSS version 16.0 (SPSS, Inc., Chicago, IL) was used. P values less than 0.05 were accepted as statistically significant. The analysis of survival was calculated from the onset of disease and not from the date of transplantation.


Patients in the control group were significantly older than those in the LTx group (Table 1), and follow-up was significantly longer for controls (median, 12 years; range, 5-20 years) versus LTx patients (median, 10 years; range, 2-24 years; P < 0.02). The clinical differences between the early-onset and late-onset groups are shown in Table 2. The late-onset group's duration of symptomatic disease at LTx was shorter than that of early-onset cases, and their nutritional status was comparable with that of early-onset cases. The frequency of males in the late-onset group was higher than that in the early-onset group.

Table 2. Clinical Data for the Transplanted Patients
 Early-Onset Cases (<50 Years of Age)Late-Onset Cases (≥50 Years of Age)P
  1. Abbreviations: LTx, liver transplantation; mBMI, modified body mass index.

Age at onset (years)36 (22–49)58 (50–68) 
Duration of disease at LTx (years)3.5 (0–12)2.0 (1–9)0.02
Follow-up from onset of disease (years)11 (2–24)6.5 (2–16)<0.001
mBMI (BMI [kg/m2] × serum albumin [g/dL])901 (420–1335)948 (563–1401)0.14
Gender (males/females)30/3430/140.03

Among the 108 LTx patients, 28 patients died. The causes of death for 11 patients were related to the operation: 3 patients died during the operation or reoperation because of circulatory collapse, 2 patients died of heart failure soon after LTx, 5 died of sepsis and multiorgan failure (including 1 kidney-liver transplant patient), and 1 died because of a pulmonary embolism. Fifteen patients died from FAP disease and its complications: 2 patients suddenly died, probably because of heart arrhythmia; 2 patients with progressing neuropathy and cardiomyopathy after LTx died of sepsis and heart failure; and 9 patients were malnourished and had advanced disease at LTx and died from FAP-related infections and heart failure. Two patients died of a malignancy that may have been related to the immunosuppressant therapy (malignant melanoma and leukemia). The age at onset of the deceased transplanted patients (median, 51 years; range, 33-65 years) tended to be higher than that of the surviving patients (median, 43 years; range, 22-68 years; P = 0.051).

Kaplan-Meier plots for LTx patients and controls are shown in Fig. 1. Survival rates differed significantly between the 2 groups, with increased survival for transplanted patients (P < 0.001; Fig. 1A). The median survival for controls was 12 years (range, 6-20 years), which is similar to what has been previously reported.1–3 The 10- and 15-year survival rates for the transplanted group were 83% and 60%, respectively, versus 62% and 19% for the controls. The majority of deceased patients were from the early series, in which 19 of 32 died, whereas only 9 of 76 died in the later series (P < 0.001). For transplanted early-onset cases, significantly improved survival was disclosed in comparison with that of the early-onset controls (P < 0.001; Fig. 1B). In contrast, no significantly increased survival was noted for transplanted late-onset cases in comparison with that of late-onset controls (P = 0.7; Fig. 1C).

Figure 1.

Kaplan-Maier plots of survival for transplanted patients and for nontransplanted historical controls. (A) Survival curves for all transplanted patients (n = 108) and for controls (n = 33). (B) Comparison of early-onset (<50 years) transplanted patients (n = 64) and controls (n = 13). (C) Comparison of late-onset (≥50 years) transplanted patients (n = 44) and controls (n = 20). Survival is from the onset of disease. The solid lines represent the LTx group, and the dashed lines represent the control group. Abbreviation: LTx, liver transplantation.

This difference between the survival of early-onset and late-onset transplanted cases became more pronounced in an analysis of the late series, in which all fatal transplanted cases were confined to the late-onset group. Thus, the improved survival for transplanted early-onset patients compared with that of early-onset controls was highly significant (P = 0.001; Fig. 2A), whereas no improved survival was found for late-onset cases (P = 0.5; Fig. 2B). The few deceased patients in the late series (n = 9) prevented multiple regression analysis for age at onset and gender as independent regressors.

Figure 2.

Kaplan-Maier plots of survival for late-series patients (1996-2008; n = 76) and controls (n = 33). (A) Comparison of early-onset (<50 years) transplanted patients from the late-series (n = 40) and early-onset controls (n = 13). (B) Comparison of late-onset (≥50 years) cases from the late series (n = 36) and late-onset controls (n = 20). Survival is from the onset of disease. The solid lines represent the late series, and the dashed lines represent the control group.

In Fig. 3, survival after LTx with respect to gender is shown. No significant difference was noted between male and female transplanted patients (P = 0.2; Fig. 3A). For early-onset cases, no significant difference in survival between male and female patients was noted (Fig. 3B). However, transplanted female late-onset cases had significantly improved survival in comparison with transplanted late-onset males (P = 0.02; Fig. 3C). In the age at onset of the late-onset group, there was no significant difference between male and female patients [60 (50-68) years for male patients versus 56 (50-64) years for female patients; P = 0.4].

Figure 3.

Influence of gender on patient survival. (A) All cases (n = 108). (B) Early-onset cases (<50 years; n = 64). (C) Late-onset cases (≥50 years; n = 44). Survival is from the onset of disease. The solid lines represent males, and the dashed lines represent females.

In our relatively small sample, no significant difference in survival was noted between patients with long disease duration (≥7 years) and patients with short disease duration (<7 years; P = 0.6; Fig. 4). However, only 14 patients with longstanding disease had been transplanted, and of those, 8 belonged to the early series. In addition, 7 of the patients in the early series with longstanding disease are dead, and all had an mBMI < 600. The mBMI of patients with longstanding disease (≥7 years) was significantly lower than that of patients with a shorter duration of disease [604.5 (420-975) versus 959.5 (550-1401); P < 0.001]. In addition, a significant correlation was found between longstanding disease and mBMI (rs = −3.0; P = 0.002). No difference in the duration of disease with respect to gender was noted (P = 0.31).

Figure 4.

Influence of disease duration at transplantation on survival (n = 108). The solid line represent <7 years, and the dashed line represents ≥7 years. Abbreviation: LTx, liver transplantation.

Of our 3 patients with non–ATTR Val30Met mutations, 1 (ATTR Phe33Leu) with an early onset of a neuropathic disease has had a favorable outcome with no progression of his disease, although kidney complications with kidney failure caused by immunosuppression and circulatory disturbances during and after LTx have been a problem. The patient is alive 10 years after the onset of his disease and 7 years after LTx. For the remaining 2 patients, heart complications and neurological disturbances appeared to progress despite LTx. The female ATTR Ala45Ser patient had a late onset of a predominantly neuropathic disease, whereas the female Leu55Gln patient had an early onset of a neuropathic disease. The 2 patients are alive 4 and 8 years, respectively, after the onset of disease, and 3 and 2 years, respectively, after LTx.


It could be questioned if LTx for FAP is an acceptable treatment because no improvement in the patients' symptoms is expected after the procedure, and the survival compared to that of nontransplanted patients may not be prolonged. Since the initial favorable case report of LTx for FAP emerged,4 it has been impossible to conduct a controlled study of LTx for this fatal disease because no comparative treatment can be offered. The only possible way to investigate if LTx improves FAP patients' survival is to perform a retrospective study using historical controls, and the present study has all the limitations associated with this type of study. Because our results depend on our historical controls, we have selected them primarily from the pretransplant period in order to avoid controls with more severe or rapidly progressing disease compared with that of our transplanted patients. It should be noted that the survival for the control patients was similar to that previously reported for non-LTx FAP patients.1–3

In an earlier study, we found increased survival for transplanted FAP patients when the procedure was carried out early after onset and before the patients became malnourished.5 This has been confirmed in this follow-up study, in which improved survival is noted for our LTx FAP patients, who now have been followed for a median of 10 years after the onset of their disease. It is therefore apparent that LTx has a place in the treatment of FAP, but it is obvious that careful selection of the patients is needed. In addition, more knowledge of the special complications experienced with FAP patients during the operation, especially circulatory complications, has improved survival after our early series.9

According to data from the Familial Amyloidotic Polyneuropathy World Transplant Register, survival after transplantation depends on good nutritional status and a short duration of the disease (<7 years).10 Besides, the outcome for non–ATTR Val30Met patients is inferior to that of ATTR Val30Met patients, probably because of continuous amyloid formation from wild-type TTR, especially in the heart.11–13

In the present study, longstanding disease (>7 years) was not associated with increased mortality. This finding was also reported by Parrilla et al.14 and Adams et al.15 In a previous study,6 we found that the duration of disease correlates with the nutritional status; thus, patients with longstanding disease will generally have a low mBMI. However, a patient with well-preserved nutritional status, despite more than 7 years of disease, may be an acceptable candidate for LTx.

It is apparent from our data that late onset of the disease is an independent risk factor for survival after LTx.6, 16 However, the fact that the survival for transplanted late-onset FAP patients was similar to that for nontransplanted controls was unexpected. It is known that there are phenotypic variations between early-onset and late-onset FAP patients, especially with respect to heart complications.7, 17 The clinical presentation of late-onset FAP resembles in many aspects that of senile systemic amyloidosis, in which cardiomyopathy dominates the clinical presentation. Senile systemic amyloidosis has a strong association with aging and is predominantly found in male patients.18 Similarly, in a recent study by Rapezzi et al.,19 cardiomyopathy was predominantly found in male patients with ATTR amyloidosis caused by mutated TTR. We note that Swedish female late-onset cases had better survival than males; thus, female late-onset patients may be less prone to develop cardiomyopathy and may be better candidates for transplantation than males.

We have shown that cardiomyopathy is related to the late onset of the disease.17 Thus, cardiac involvement may precede other symptoms in late-onset cases, and a lead-time bias in late-onset patients may be present.

In a recent study, we found 2 different types of ATTR amyloid deposits in Swedish FAP ATTR Val30Met patients: deposits composed of only full-length TTR and deposits with a mixture of full-length and truncated TTR. Full-length TTR amyloid deposits were associated with early onset, and a mixture of full-length and truncated TTR was associated with late onset and signs of cardiac involvement.8 It is tempting to suspect that the outcome after LTx is related to the composition of the amyloid deposits: late onset and cardiomyopathy are related to a senile type of mixed fibril composition.

The progression of the disease after LTx noted in 2 of the 3 non–ATTR Val30Met patients underlines the difficulties in selecting non–ATTR Val30Met patients for transplantation. Thus, in our experience, neither ATTR Ala45Ser patients nor ATTR Leu55Gln patients appear to be candidates for LTx. To our knowledge, no other reports on the outcome after LTx for these 3 mutations have been published.

In conclusion, LTx for FAP improves overall survival. However, the survival rate for male patients with a late onset of the disease (>50 years of age) was not different from that of nontransplanted historical controls. Further studies are needed to analyze the cause of the poorer outcome for these patients. At present, we are reluctant to submit late-onset cases with echocardiographic evidence of amyloid cardiomyopathy with signs of heart failure for LTx; for those cases, combined liver and heart transplantation or LTx with the possibility of subsequent heart transplantation may be considered.