Familial amyloidosis (FAP*), identified just over 40 years ago, is the rarest type of systemic amyloidosis (1). The exact prevalence of FAP is unknown but, based on referral patterns to major centers, it probably represents less than 10% of the patients seen with amyloidosis. It is estimated that there are between 1000 and 2225 new cases of primary amyloidosis in the United States each year (2, 3) and therefore one might predict between 100 and 200 newly diagnosed cases of FAP per year.
FAP is inherited as an autosomal dominant protein deposition disease, which in the majority of cases is associated with production of variants of the protein transthyretin (1). Mutant forms of transthyretin usually have a single amino acid substitution resulting from a DNA point mutation that is inherited within each affected kinship (1). Transthyretin is deposited in extracellular sites throughout the body where the fibrillar protein matrix results in organ dysfunction.
Symptoms begin in middle life with peripheral neuropathy, autonomic dysfunction, cardiomyopathy, and occasionally renal failure; death results between 7 and 15 years after disease onset. Disorders of conduction necessitating permanent pacing are the predominant manifestation of cardiovascular disease (4, 5), although myocardial infiltration is also frequently present (6, 7). Treatment of FAP depends on the pattern of organ involvement but until recently has been limited to supportive measures including a cardiac pacemaker, renal dialysis, and parenteral nutrition.
The major source of production of mutant transthyretin is from the liver, and in 1990 it was proposed that liver transplantation might offer a definitive cure for these patients by removing the source of abnormal protein synthesis (8). After liver transplantation, serum mutant protein transthyretin becomes unmeasurable (9, 10), and there has been objective evidence of improvement in neurological disease, particularly autonomic neuropathy (8, 11-13). We describe the longitudinal cardiac status of a series of patients with FAP treated with liver transplantation.
PATIENTS AND METHODS
Between August 1986 and August 1995, 19 patients with FAP were evaluated by the Amyloid Treatment and Research Unit at Boston University Medical Center and underwent liver transplantation. Of these 19 patients, 11 had transplants performed in Boston and 8 had transplants performed elsewhere in the United States. Five of the patients had a survival of less than 1 year after transplantation and therefore could not be included in any serial long-term follow-up assessment. Of the remaining 14 patients, complete clinical evaluations were available in 11, who were followed for a mean (±SD) of 41±34 months before and 31±14 months after liver transplantation (Table 1). After the initial evaluation patients were prospectively followed by written contact and annual visits to their regional transplant units. All patients were seen by a cardiologist and in addition underwent electrocardiography and transthoracic echocardiography with M-mode, 2D, and Doppler evaluation. Heart failure was considered to be present on physical examination in patients with elevated jugular venous pressure and/or radiologic evidence of pulmonary venous congestion. Severity of heart failure was classified using the New York Heart Association criteria (14). No patient was hypertensive, had obstructive valvular disease, or a history of previous myocardial infarction.
Among the 11 patients with FAP, 3 different mutations of the protein transthyretin existed among 6 kindreds (Table 1). Total transthyretin was separated from other serum proteins by polyacrylamide gel electrophoresis and subjected to isoelectric focusing to separate normal and mutant forms. Postoperatively, sera from all patients showed only normal transthyretin (10). Blood concentrations of tacrolimus (Prograf, formerly FK506; Fujisawa USA, Deerfield, IL) were measured using a microparticle enzyme immunoassay.
A total of 64 echocardiographic and Doppler studies were analyzed; each patient had between 3 and 10 studies (mean of 6). The studies were analyzed with the echocardiographer unaware of the date of liver transplantation, the kindred mutation, or the identity of the patient. Echocardiographic features common to primary amyloidosis (15, 16) and familial (17, 18) cardiac amyloidosis were recorded. Left ventricular mass was calculated from M-mode measurements (19). Right ventricular wall thickening was defined as thickness >0.7 cm. Left ventricular diastolic and systolic volumes were calculated from tracings of the endocardial border in the fourchamber view, and ejection fractions were determined. Impairment of left and right ventricular function were defined as an ejection fraction of <50% and fractional area change <35%, respectively (20). Mitral Doppler studies were performed at the leaflet tips, and flow velocity waveforms were analyzed to assess the presence of restrictive ventricular filling patterns (21, 22). A low voltage electrocardiogram was defined as a tracing with a mean QRS voltage amplitude in the limb leads of ≤0.5 mV. Pseudo-infarction was defined as a recording with QS waves in the anteroseptal or inferior leads. Mean 12 lead and precordial voltage (Sokoloff index) were also measured (15).
Data are reported as mean values ± SD apart from blood tacrolimus concentrations, which are reported as medians. Comparisons of continuous variables on serial studies were compared using a paired Student t test. Significance was accepted as a probability of P<0.05.
On examination of the echocardiographic results of all 11 patients who underwent liver transplantation, two distinct patterns emerged that allowed the patients to be divided into two groups. Group 1 consisted of five patients in whom echocardiographic evidence of significant progression of wall thickening (defined as a ≥0.3 cm increase in mean left ventricular wall thickness) was found after liver transplantation. All four patients (same kindred) with the Glu 42 Gly transthyretin mutation were in this group, as was the single patient with the Ala 36 Pro mutation. At the time of surgery, all five patients already had evidence of cardiac involvement with ventricular wall and valve thickening. The remaining six patients formed group 2; they showed no wall thickening before liver transplantation and had no clinically significant progression of echocardiographically determined cardiac involvement after surgery. Of note, these six patients all had the same transthyretin mutation, Val 30 Met. Patients from group 1 were of similar ages and were studied for a similar follow-up period (40±4 years and 34±11 months, respectively) to patients from group 2 (41±6 years and 29±16 months, respectively).
In the 11 patients described there has been objective evidence of an improvement in polyneuropathy in nine patients. Details of these improvements in autonomic and sensorimotor deficit are beyond the scope of this report, however, eight of these nine patients have recently been described elsewhere (13). Despite clear evidence for heart involvement by amyloid in the group 1 patients, their ability to engage in physical activity was predominantly influenced by the presence of peripheral and autonomic neuropathies, with four of the five patients having experienced symptoms due to postural hypotension before liver transplantation. The combination of autonomic dysfunction and gastrointestinal involvement was particularly debilitating, with two patients requiring a prolonged period of parenteral nutrition after surgery. Mean arterial blood pressures did not change significantly from values measured before liver transplantation (91±10 mmHg) to those measured after surgery (95±11 mmHg, P=0.39). One patient from group 1 (Ala 36 Pro mutation) has since died; although no autopsy was performed, the death was sudden and unexpected and most probably had an underlying cardiac etiology. In addition, a second patient from group 1 (Glu 42 Gly mutation) has recently developed heart failure.
Within group 1 patients, serial echocardiograms showed a progressive increase of mean left ventricular wall thickness, with significant increases in values obtained before liver transplantation (1.28±0.29 cm) compared with the most recent assessments (1.71±0.32 cm, P<0.0001). In four of these patients there had been evidence for both left and right ventricular wall thickening before liver transplantation, whereas in one patient (patient 5) the echocardiogram before transplant was equivocal (Fig. 1). Concentric left ventricular wall thickening resulted in a significant increase in left ventricular mass from that before liver transplantation (202±38 g) compared with that calculated from the most recent study (317±46 g, P=0.001) (Fig. 2). Right ventricular wall thickening (>0.7 cm) was present in three of the group 1 patients before transplantation and in two additional patients after transplantation. Valve thickening, a feature common to cardiac amyloidosis, was present before surgery in four patients from group 1 and in all five group 1 patients after transplantation.
No significant differences were found when left ventricular wall thickness and mass measurements made before transplantation were compared with those after transplantation in group 2 patients (Val 30 Met mutation) (Figs. 1 and 2). After transplantation, four patients showed an increase in left atrial size, two patients developed right atrial dilatation, and three patients had pericardial effusions on echocardiography; all were from group 1.
Left ventricular systolic function, determined from the ejection fraction, was normal for all patients before transplantation. However, within group 1 there was a significant decrease in left ventricular ejection fraction before transplantation (67±3%) compared with that at follow-up (59±5%, P=0.002). Group 2 patients had similar ejection fractions before their surgery (67±5%) and at follow-up (68±5%, P=NS). Within group 1, three patients had a reduction in right ventricular systolic function to <35%, resulting in a significant reduction from pretransplant measurements (43±5%) compared with those at recent evaluation (33±8%, P=0.02). Patients within group 2 showed no significant change in right ventricular systolic function after transplantation (Table 2).
In group 1, one patient satisfied criteria for restrictive physiology before transplantation, and an additional two patients developed Doppler features of restrictive ventricular physiology (21, 22). No patient from group 2 satisfied the criteria of a restrictive left ventricular physiology (Table 2).
Conduction system disease was present in two group 1 patients before transplantation and in an additional two patients after liver transplantation. Of the five patients of group 1, all of whom showed progressive wall thickening after transplantation, the electrocardiographic limb lead QRS voltage amplitude decreased in three patients, increased in one patient, and showed little change in the fifth patient after surgery (Fig. 3). In most instances the mean 12 lead voltage and precordial voltage (Sokoloff index) followed the same trends. The predominant electrocardiographic abnormalities in group 2 patients were those of conduction system disease. Before transplantation, three patients showed first degree atrioventricular block (AVB), and at subsequent follow-up a fourth patient had developed first degree AVB. In addition one patient had progressed to left bundle branch block and another had required a permanent pacemaker after progression of first to second degree AVB. QRS voltage amplitudes in the group 2 patients showed a tendency to increase on all three measurement parameters, although there were inconsistencies. Electrocardiograms from patients who developed left bundle branch block or required permanent pacing were uninterpretable with regard to voltage changes.
Immunosuppressive therapy consisted of cyclosporine in six patients and tacrolimus in the remaining five patients. All five patients taking tacrolimus had median blood concentrations of the drug that were less than 12 ng/ml. Of the five patients (group 1) who showed echocardiographic evidence of ventricular wall thickening, three were taking cyclosporine and two were taking tacrolimus. In addition, all 11 patients had taken, and the majority continue to take, prednisone.
We have shown that five patients with FAP (four with transthyretin Glu 42 Gly and 1 with Ala 36 Pro) show continued wall thickening after liver transplantation with a concomitant trend to deterioration in biventricular systolic function and left ventricular diastolic function. In contrast six patients with transthyretin Val 30 Met had normal wall thickness, which continued after transplantation.
In 1993, the first International Workshop on liver transplantation in FAP concluded that after liver transplantation, disease progression was halted and that some autonomic nervous system improvement occurs even with the brief period of follow-up available (11). The absence of the mutant protein in the posttransplant serum verifies the fact that the major source of mutant transthyretin has been removed (10). However, there remain more minor sources of transthyretin production, which include the choroid plexus and associated retinal epithelium. Furthermore, both progressive and denovo amyloid deposition in ocular tissues after liver transplantation in FAP have been described (23, 24). These sources account for less than 2% of the transthyretin production but the question arises as to whether amyloid fibrils already present, for example in the heart of the transplant recipient, may act as a nidus for nonhepatic sources of mutant transthyretin or for the deposition of normal transthyretin produced by the transplanted liver. The possibility that native transthyretin might continue to be deposited as amyloid is supported by the evidence that this occurs in senile amyloidosis (25). Furthermore, amyloid deposits in FAP consist of both native and variant transthyretin (26, 27), the former representing about one third of the molecules (26, 28, 29).
There was an overall improvement in neurological deficit after transplantation in 9 of these 11 patients and a general improvement in symptoms, although most of the patients studied still retain some symptoms of neuropathy. Of the patients with progressive wall thickening (group 1), one has since died at the age of 46, a second has recently developed heart failure, and a third remains symptomatic due to autonomic neuropathy and direct involvement of the gastrointestinal tract.
The results of the electrocardiographic measurements in our five patients with progressive wall thickening are compatible with a pattern of hypertrophy (increase in voltage) in one patient and with possible progression of amyloid deposition (decrease in voltage) in three patients. Although wall thickness was normal in group 2 patients, there was evidence of conduction system disease before liver transplantation as is commonly seen in this condition.
Both the particular transthyretin mutation type and the kindred association are recognized as governing the pattern of target organ involvement, the course of the disease, and the approximate age at which the function of a particular organ system might be affected by amyloid involvement. It is of interest that group 1 patients, defined by progressive left ventricular wall thickening after transplantation, showed an abnormal wall thickness before surgery in four of the five patients and a mild increase in the remaining patient. This suggests that amyloid deposition was present before transplantation and may be of significance in the etiology of its apparent progression. In contrast no patient with the Val 30 Met variant (forming group 2) had wall thickening before or after transplantation. In the Val 30 Met variant, in which cardiac involvement usually develops late in the course of the illness (1), the possibility exists that change in wall thickness may develop at a later date.
It has been suggested that immunosuppressive therapy (30, 31) and/or high-dose steroids (32) might be responsible for heart wall thickening in pediatric transplant recipients who receive tacrolimus. In adult patients the possible hypertrophic effects of tacrolimus are less consistent (33, 34); this may reflect substantially lower blood concentrations of tacrolimus in adult liver transplant patients.
Of the five patients that showed wall thickening in our study (group 1), two were taking tacrolimus and three were taking cyclosporine. One patient from this group took tacrolimus for 6 months before continuing with cyclosporine, which one might predict would have had a limited effect, particularly as the tacrolimus-associated hypertrophic cardiomyopathy seems to be reversible on discontinuation of the drug (30, 32). Additionally, none of our group 2 patients have developed hypertrophy despite the fact that three of these six patients continue to use tacrolimus. In comparison to the levels reported in studies that have shown hypertrophy (31, 33, 35) the blood levels of tacrolimus are low in our patients, with no patient exceeding a median blood concentration of 12 ng/ml.
Hypertension is a recognized complication accompanying immunosuppressive therapies in transplant recipients; it can also be occult due to disturbances in circadian patterns (35, 36), a feature that magnifies hypertensive target effects. The side effects of fluid retention and hypertension are shared by prednisone and the immunosuppressants, and either could be involved in the pathogenesis of cardiac hypertrophy. However, in our patients mean arterial blood pressure measurements made before liver transplantation and during immunosuppressive therapy were similar.
This study does not include a control group of patients who were not treated with liver transplantation and thus it is not known whether the rate of progression of the echocardiographic changes would be less, similar, or increased in such a group. In an ideal control group such subjects should have the same transthyretin mutation, be age matched, and be of the same kindred. The rarity of FAP and the limited number of patients who have undergone liver transplantation would make this extremely difficult to accomplish. The magnitude of the results are enhanced by the longitudinal study method, which used all available echocardiographic and electrocardiographic examinations to demonstrate trends in particular features over time. We would emphasize that four of the five patients in group 1 who showed progression of wall thickening are from the same family, and one should be cautious in extrapolating their apparent progression to other kindreds even of the same mutation. However, the fifth patient in group 1, with a different mutation and who has since died, had similar progression of wall thickening, which suggests that this may be generally applicable.
In conclusion, patients with FAP who have undergone liver transplantation may continue to develop wall thickening and increases in left ventricular mass in addition to electrocardiographic abnormalities. We believe this probably represents continued amyloid deposition in the heart, and we base this on the observation that progressive wall thickening occurred only in patients with amyloid deposition present before transplantation. In such patients this may represent a seeding effect in which nucleation-dependent polymerization is the underlying mechanism (37). Because mutant transthyretin is no longer detected we suggest that this amyloid deposition in the heart may be due to native transthyretin. Progression of wall thickness was associated with sudden death in one patient and new-onset heart failure in another. As a result of these findings we suggest that all patients with transthyretin mutations associated with cardiomyopathy be followed by annual assessments that should include electrocardiography and echocardiography.
Acknowledgments. The authors thank Julie Dellot, Department of Surgery, New England Deaconess Hospital, for providing the technical data on these patients.
This study was supported by NIH grants 40414, 20613, and RR 533, the Amyloid Research Fund of Boston University, and the Sue Sellors Finley Cardiac Amyloid Research Fund.
Abbreviations: Ala, alanine; AVB, atrioventricular block; FAP, familial amyloid polyneuropathy; Glu, glutamate; Gly, glycine; Met, methionine; Pro, proline; Val, valine.
1. Benson MD, Amyloidosis. In: Scriver CR, Beaudet AK, Sly WS, Valle D, ed. The metabolic and molecular bases of inherited disease. New York: McGraw-Hill, 1995: 4159.
2. Simms RW, Prout MN, Cohen AS. The epidemiology of AL and AA amyloidosis. Balliere's Clin Rheumatol 1994; 8: 627.
3. Kyle RA, Gertz MA. Primary systemic amyloidosis: clinical and laboratory features in 474 cases. Semin Hematol 1995; 32: 45.
4. Falcao De Freitas A, Barbedo A. Conduction disturbances in 190 patients with familial amyloidotic polyneuropathy (Andrade's type). Adv Cardiol 1978; 21: 206.
5. Olofsson BO, Andersson R, Furberg B. Atrioventricular and intraventricular conduction in familial amyloidosis with polyneuropathy. Acta Med Scand 1980; 208: 77.
6. Eriksson P, Backman C, Bjerle P, Eriksson A, Holm S, Olofsson BO. Non-invasive assessment of the presence and severity of cardiac amyloidosis: a study in familial amyloidosis with polyneuropathy. Br Heart J 1984; 52: 321.
7. Hongo M, Ikeda S. Echocardiographic assessment of the evolution of amyloid heart disease: a study with familial amyloid polyneuropathy. Circulation 1986; 73: 249.
8. Holmgren G, Ericzon BG, Groth CG, et al. Clinical improvement and amyloid regression after liver transplantation in hereditary transthyretin amyloidosis. Lancet 1993; 341: 1113.
9. Holmgren G, Steen L, Ekstedt J, et al. Biochemical effect of liver transplantation in two Swedish families with familial amyloidotic polyneuropathy (FAP-met30
). Clin Genet 1991; 40: 242.
10. Skinner M, Lewis WD, Jones LE, et al. Liver transplantation as a treatment for familial amyloidotic polyneuropathy. Ann Intern Med 1994; 120: 133.
11. Steen L, Holmgren G, Suhr O, Wikstrom L, Ericzon BG, Groth CG. World wide survey of liver transplantations in patients with familial amyloidotic polyneuropathy. Amyloid Int J Exp Clin Invest 1994; 1: 138.
12. Parrilla P, Lopez-Andreu FR, Ramirez P, et al. Familial amyloidotic polyneuropathy type 1 (Andrade's disease): a new indication for liver transplantation. Transplantation 1994; 57: 473.
13. Bergethon PR, Sabin TD, Lewis D, Simms RW, Cohen AS, Skinner M. Improvement in the polyneuropathy associated with familial amyloid polyneuropathy after liver transplantation. Neurology 1996; 47: 944.
14. Criteria Committee, New York Heart Association, Inc. Diseases of the heart and blood vessels: nomenclature and criteria for diagnosis, 6th ed. Boston: Little, Brown and Co., 1964: 114.
15. Falk RH. Cardiac amyloidosis. Prog Cardiol 1989; 2: 143.
16. Kyle RA, Gertz MA. Primary systemic amyloidosis: clinical and laboratory features in 474 cases. Semin Hematol 1995; 32: 45.
17. Backman C, Olofsson BO. Echocardiographic features in familial amyloidosis with polyneuropathy. Acta Med Scand 1983; 214: 273.
18. Hongo M, Ikeda S. Echocardiographic assessment of the evolution of amyloid heart disease: a study with familial amyloid polyneuropathy. Circulation 1986; 73: 249.
19. Devereux RB, Alonso DR, Lutas EM, et al. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol 1986; 57: 450.
20. Grose R, Strain J, Yipintosoi T. Right ventricular function in valvular heart disease: relation to pulmonary pressure. J Am Coll Cardiol 1983; 2: 225.
21. Klein AL, Cohen GI. Doppler echocardiographic assessment of constrictive pericarditis, cardiac amyloidosis, and cardiac tamponade. Cleve Clin J Med 1992; 59: 278.
22. Xie G-Y, Berk MR, Smith MD, DeMaria AN. Relation of Doppler transmitral flow patterns to functional status in congestive heart failure. Am Heart J 1996; 131: 766.
23. Benson MD, Uemichi T. Transthyretin amyloidosis. Amyloid Int J Exp Clin Invest 1996; 3: 44.
24. Ando Y, Ando E, Tanaka Y, et al. De novo amyloid synthesis in ocular tissue in familial amyloidotic polyneuropathy after liver transplantation. Transplantation 1996; 62: 1037.
25. Westermark P, Sletten K, Johansson B, Cornwell GG. Fibril in senile systemic amyloidosis is derived from normal transthyretin. Proc Natl Acad Soc USA 1990; 87: 2843.
26. Dwulet FE, Benson MD. Primary structure of an amyloid prealbumin and its plasma precursor in a heredofamilial polyneuropathy of Swedish origin. Proc Natl Acad Sci USA 1984; 81: 694.
27. Thylen C, Wahlquist J, Haettner E, Sandgren O, Holmgren G, Lundgren E. Modifications of transthyretin in amyloid fibrils: analysis of amyloid from homozygous and heterozygous individuals with the Met 30 mutation. EMBO J 1993; 12: 743.
28. Wallace MR, Dwulet FE, Conneally M, Benson MD. Biochemical and molecular genetic characterization of a new variant prealbumin associated with hereditary amyloidosis. J Clin Invest 1986; 78: 6.
29. Saraiva MJM, Birken S, Costa PP, Goodman DS. Amyloid fibril protein in familial amyloidotic polyneuropathy Portuguese type. J Clin Invest 1984; 74: 104.
30. Atkison P, Joubert J, Barron A, et al. Hypertrophic cardiomyopathy associated with tacrolimus in paediatric transplant patients. Lancet 1995; 345: 894.
31. Hibi S, Misawa A, Tamai M, et al. Severe rhabdomyolysis associated with tacrolimus. Lancet 1995; 346: 702.
32. Dwahan A, Mack DR, Langnas AN, Shaw BW, Vanderhoof JA. Immunosuppressive drugs and hypertrophic cardiomyopathy. Lancet 1995; 345: 1644.
33. Dollinger MM, Plevris JM, Chauhan A, MacGilchrist AJ, Finlayson NDC, Hayes PC. Tacrolimus and cardiotoxicity in adult liver transplant recipients [Letter]. Lancet 1995; 346: 507.
34. Natazuka T, Ogawa R, Kizaki T, et al. Immunosuppressive drugs and hypertrophic cardiomyopathy. Lancet 1995; 345: 1644.
35. Lipkin GW, Tucker B, Giles M, Raine AE. Ambulatory blood pressure and left ventricular mass in cyclosporin and non cyclosporin treated renal transplant recipients. J Hyperten 1993; 11: 439.
36. Textor SC, Canzanello VJ, Taler SJ, et al. Cyclosporin induced hypertension after transplantation. Mayo Clin Proc 1994; 69: 1182.
37. Jarrett JT, Lansbury PT Jr. Seeding “one-dimensional crystallization” of amyloid: a pathogenic mechanism in Alzheimer's disease and scrapie. Cell 1993; 73: 1055.