Giant cell myocarditis (GCM) is a rare disorder in which survival beyond 1 year without heart transplantation is uncommon. Long-term follow-up data on those with such survival are lacking. Twenty-six patients with biopsy-proved GCM who survived for >1 year without heart transplantation were identified from a multicenter GCM registry. The incidence of death, transplantation, ventricular assist device placement, and histologically proved disease recurrence was ascertained retrospectively. The rates of recurrent heart failure, ventricular arrhythmias, renal failure, and infectious complications were calculated. The mean age of the cohort was 54.6 ± 13.9 years (65% women). The mean follow-up duration was 5.5 years starting 1 year after diagnosis. There were 3 deaths (12%), 5 heart transplantations (19%), and 1 ventricular assist device placement (4%). Three histologically confirmed recurrences of GCM (12%) occurred between 1.5 and 8 years after diagnosis. Thirteen of 26 patients experienced a total of 30 heart failure episodes ≥1 year after initial diagnosis. There were 23 episodes of elevated creatinine in 12 patients, 41 infectious events in 13 patients, and 19 episodes of ventricular arrhythmias in 6 patients with a total of 144 years of follow-up. Starting 1 year after GCM diagnosis, the combined rate of death, transplantation, ventricular assist device placement, and GCM recurrence was 47% at 5 years. In conclusion, the risk for GCM recurrence continues to ≥8 years after diagnosis.
Giant cell myocarditis (GCM) is an uncommon but frequently fatal form of acute myocarditis that has been shown to respond to cyclosporine-based immunosuppressive therapy. In cases proved by biopsy, the most frequent outcome is death or transplantation within the first year after diagnosis. Even with transplantation, GCM recurrence in the donor heart has been cited as high as 20% to 25%. As patients survive longer without transplantation because of efficacious medical therapy, questions of long-term survival and of recurrence risk in the native heart beyond the first year have not been answered. Likewise, the optimal duration and intensity of immunosuppression beyond 1 year are not known. Furthermore, extracardiac morbidity in these patients has not been evaluated. No single center has accumulated experience to catalog the long-term risks for disease recurrence and extracardiac morbidity in patients with GCM. To assess survival and recurrence risk, and to characterize the clinical picture of a prolonged course of GCM, we followed patients from multiple sites in the United States and Germany with biopsy-proved GCM who survived for >1 year on immunosuppression without cardiac allotransplantation.
Methods
Records from Mayo Clinic (Rochester, Minnesota) and Charite Hospital (Berlin, Germany) were systematically searched to gather all cases of histologically confirmed GCM with survival beyond 1 year without heart transplantation. Additional known cases, in our consultative files, were also included if sufficient patient data and research consent were available. Investigators completed a form on each case, requesting historical data on the medical history, presenting symptoms, cardiac rhythm, immunosuppressive regimen during and after year 1 after diagnosis, and heart failure treatment.
Adverse events were followed after the first year. These included new or worsening heart failure, defined as a change in New York Heart Association class of ≥1, a ≥10% decrease in the ejection fraction (by any method) compared with the year 1 value or subsequent peak before measure, or a new hospitalization or emergency department visit for heart failure; new or worsening renal insufficiency, defined as elevated serum creatinine higher than the end of year 1 baseline by 0.3 mg/dl; new or worsening ventricular arrhythmias; and infection. Outcomes including death, heart transplantation, ventricular assist device (VAD) placement, biopsy-proved GCM recurrence, and date of last follow-up were obtained.
Available slides from patients with GCM were retrieved from institutional archives. Three separate biopsies were considered for scoring: initial GCM diagnosis; postdiagnostic biopsy, usually at 1 month after initial diagnosis; and at the time of suspected recurrence. A single cardiovascular pathologist (JJM) reviewed all available material. Histologic features assessed included the presence of giant cells, the character of inflammation (lymphocytes, eosinophils, macrophages), the presence of granulomas, the presence of myocyte hypertrophy, and interstitial fibrosis (particularly replacement type). The diagnosis of active GCM ( Figure 1 ) was made by established criteria. Likewise, recurrence was established on the basis of the same criteria ( Figure 1 ). The finding of replacement-type fibrosis, in the absence of coronary artery disease, was interpreted as representing myocarditis-related injury ( Figure 1 ). If histologic features did not allow definitive discrimination among GCM, lymphocytic myocarditis, and/or cardiac sarcoidosis, the case was excluded from the study.
Overall event rates were estimated using the Kaplan-Meier method. These events included overall survival as well as event-free survival for the combined events of infection, heart failure, and ventricular arrhythmia. Comparisons of the cumulative event rates between groups were completed using log-rank tests. Event rates were evaluated starting 1 year after diagnosis.
Results
Twenty-six patients with GCM met the selection criteria of ≥1-year survival without heart transplantation. The demographics, presenting features, and medical treatment for all patients are described in Table 1 . The mean age was 54.6 ± 14.1 years at the time of diagnosis. The cohort was 65% female. The mean follow-up duration, starting from 1 year after diagnosis, was 5.5 years (range 1 to 16.6, median 4.8).
| Case | Age | Sex | Duration of Follow-up (years) | Primary presenting symptoms | Number of Immunosuppressive | Non-Infectious Adverse Outcomes | |
|---|---|---|---|---|---|---|---|
| First Year | > First Year | ||||||
| 1 | 32 | F | 15.0 | heart failure | steroid, azathioprine | – | – |
| 2 | 34 | F | 16.6 | chest pain | cyclosporine, steroid, azathioprine, OKT3 | cyclosporine, steroid, azathioprine | renal failure, heart failure |
| 3 | 35 | M | 5.9 | shortness of breath | cyclosporine, steroid | cyclosporine, steroid | – |
| 4 | 38 | M | 9.2 | chest pain | cyclosporine, steroid, OKT3 | cyclosporine, steroid, azathioprine, sirolimus | heart failure, ventricular arrhythmia |
| 5 | 42 | F | 1.4 | shortness of breath | cyclosporine, steroid, OKT3 | cyclosporine, steroid | – |
| 6 | 45 | F | 9.0 | presyncope | cyclosporine, steroid, OKT3 | cyclosporine, steroid | renal failure, heart failure |
| 7 | 46 | F | 1.0 | heart failure | cyclosporine, steroid, OKT3 | – | death |
| 8 | 47 | F | 2.7 | chest pain | cyclosporine, steroid, OKT3 | cyclosporine, steroid | renal failure |
| 9 | 47 | M | 12.5 | heart failure | cyclosporine, steroid, azathiprine, OKT3 | cyclosporine, steroid, azathioprine | renal failure, heart failure |
| 10 | 47 | M | 6.5 | heart failure | steroid, sirolimus, mycophenolate mofetil | steroid, mycophenolate mofetil | renal failure |
| 11 | 50 | F | 4.1 | heart failure | cyclosporine, steroid, sirolimus, mycophenolate mofetil | cyclosporine, steroid, sirolimus, mycophenolate mofetil | renal failure, heart failure, ventricular arrhythmia |
| 12 | 50 | F | 6.7 | presyncope | steroid, azathioprine | steroid, azathioprine | heart failure, ventricular arrhythmia |
| 13 | 51 | M | 1.4 | shortness of breath | steroid, azathioprine | azathioprine | – |
| 14 | 51 | F | 4.0 | shortness of breath | cyclosporine, steroid, azathioprine | cyclosporine, steroid, azathioprine, mycophenolate mofetil | – |
| 15 | 52 | M | 1.4 | heart failure | steroid | – | – |
| 16 | 52 | M | 7.6 | presyncope | cyclosporine, steroid, OKT3 | cyclosporine, steroid, mycophenolate mofetil | heart failure |
| 17 | 55 | F | 1.3 | heart failure | cyclosporine, steroid, sirolimus | cyclosporine, steroid, azathioprine, sirolimus | renal failure, heart failure |
| 18 | 58 | F | 4.8 | heart failure | cyclosporine, steroid, OKT3 | steroid, sirolimus, mycophenolate mofetil | renal failure, heart failure |
| 19 | 60 | F | 1.7 | heart failure | cyclosporine, steroid, sirolimus, mycophenolate mofetil | steroid, sirolimus, mycophenolate mofetil | renal failure, heart failure, ventricular arrhythmia |
| 20 | 60 | M | 2.6 | heart failure | cyclosporine, steroid, mycophenolate mofetil | steroid, mycophenolate mofetil | renal failure, heart failure, ventricular arrhythmia, death |
| 21 | 69 | F | 1.6 | heart failure | cyclosporine, steroid, mycophenolate mofetil | heart failure, death | |
| 22 | 72 | M | 5.7 | heart failure | cyclosporine | Cyc, Ster, Siro, MM | heart failure |
| 23 | 76 | F | 6.8 | heart failure | cyclosporine, steroid | cyclosporine, steroid, mycophenolate mofetil | heart failure |
| 24 | 78 | F | 1.2 | heart failure | cyclosporine, steroid, mycophenolate mofetil | steroid, sirolimus, mycophenolate mofetil | – |
| 25 | 78 | F | 7.9 | heart failure | cyclosporine, steroid, sirolimus, mycophenolate mofetil | cyclosporine, steroid | renal failure, heart failure, ventricular arrhythmia |
| 26 | 80 | F | 4.4 | shortness of breath | cyclosporine, steroid | steroid | – |
Detailed medication and device therapy data were available for 23 patients. Of these, 23 (100%) received angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, 19 (83%) received β blockers, 19 (83%) received antiarrhythmic medications, and 9 (39%) received digoxin. Fifteen subjects (65%) received implantable cardioverter-defibrillators. One subject (4%) received a pacemaker, and 1 subject (4%) required a temporary VAD for 53 weeks before undergoing heart transplantation but did not survive the transplantation surgery. No patients required permanent VAD support or intra-aortic balloon counterpulsation.
During the first year after diagnosis, all patients received cyclosporine-based immunosuppression. At the time of diagnosis, 11 of the 26 patients (42%) received muromonab-CD3, antithymocyte globulin, equine antithymocyte globulin, or a comparable antilymphocytic agent. One patient received a combination of antilymphocyte therapies. Seven patients (27%) were weaned to corticosteroid therapy only, and the remaining 19 (73%) received a combination of cyclosporine and corticosteroids during the first year. At varying times in the first year, 6 patients (23%) received azathioprine, 6 (23%) received mycophenolate mofetil, and 5 (19%) were switched to a sirolimus-based regimen for cyclosporine intolerance. The average minimum number of immunosuppressive medications used in the first year after diagnosis was 1.91 ± 0.65. The average maximum number of immunosuppressive medications used was 2.7 ± 0.46 ( Table 1 ).
After the first year after diagnosis, the average minimum number of immunosuppressive medications decreased to 1.25 ± 0.85, while the average maximum number of immunosuppressive medications decreased to 2.12 ± 0.90. At any point in time, 26 patients (100%) received corticosteroids, 21 (81%) received cyclosporine, 10 (38%) received mycophenolate mofetil, 8 (31%) received azathioprine, and 8 (31%) received sirolimus. The indications and combinations of these drugs were dictated by clinical course, tolerance of the medications, and episodes of renal insufficiency or heart failure.
In general, the dose of cyclosporine was adjusted to achieve 11-hour trough levels of 150 to 200 μg/L during the first 6 months, 100 to 150 μg/L between 6 and 12 months, and 75 to 100 μg/L after 1 year. Of those with histopathologic recurrence, 2 patients had recently discontinued or decreased immunosuppression. There was no definite correlation between arrhythmia risk and level of immunosuppression.
All patients, as a matter of selection, met histologic criteria for GCM at the time of initial biopsy ( Figure 1 ). Tissue slides were available for review in 17 patients (65%). The remaining 9 patients were all evaluated at outside academic institutions with experienced cardiovascular pathologists rendering unequivocal diagnostic interpretations. Well-formed granulomas were not present in any case.
Throughout follow-up, additional histologic material was available in 13 patients (ranging from 1 to 11 biopsy procedures) beyond their diagnostic biopsies, either for therapeutic monitoring or suspected disease recurrence. Three cases of biopsy-proved recurrence were identified within the cohort ( Figure 1 ), 2 of which were available for histologic review (the third had only a detailed pathology report available). These recurrences occurred 1.5 to 8 years after initial diagnosis.
Changes in myocardial infiltrate were noted in all patients for whom initial and follow-up biopsy samples were available. The 2 cases of recurrence available for review showed increased interstitial (replacement-type) fibrosis (there was no mention of such in the report of the third case, so it is not possible to determine whether such was present or not). Although 2 cases contained giant cells, the third exhibited only a lymphocytic-eosinophilic inflammatory infiltrate with associated myocyte injury. The remaining 10 cases, sampled after the diagnostic biopsy, uniformly lacked giant cells in the subsequent biopsies. The 3 most extensively sampled cases (with 11, 6, and 4 endomyocardial biopsies), showed a stepwise progression from fulminant GCM, to a myocarditis characterized by a lymphocytic-eosinophilic infiltrate, to a smoldering lymphocytic myocarditis, to interstitial (replacement-type) fibrosis over the course of 2 years, 5 months, and 2 years (respectively). Two cases showed unremarkable myocardium on repeat biopsy, and 1 showed only mild interstitial fibrosis. One case showed persistent lymphocytic-eosinophilic myocarditis. The remaining 4 cases all showed borderline lymphocytic myocarditis with associated mild to moderate interstitial (pericellular- and replacement-type) fibrosis.
Adverse events were dichotomized into noninfectious and infectious events. In this sick population, 84 total noninfectious adverse events were noted over 144 patient-years of follow-up after the first year after diagnosis ( Table 1 ). New or worsening renal insufficiency was noted in 12 patients, with an event rate of 20% per year. Ventricular arrhythmias were noted in 6 patients, all of whom presented initially with some form of arrhythmia, for a total of 19 episodes during the observed follow-up period and an event rate of 17% per year. No new ventricular arrhythmias were noted in patients who did not initially have arrhythmias at presentation. New or worsening heart failure, suggesting possible disease recurrence, was noted in 13 patients, including 30 total distinct events with a 26% event rate per year. There were 23 episodes of elevated creatinine (renal failure) in 12 patients, resulting in an event rate of 20% per year.
Infectious adverse events were observed in 13 patients, totaling 41 separate events ( Table 2 ). These included 2 episodes of pneumonia, 2 cases of herpes (including oral and genital forms), and 1 case each of atypical mycobacterial infection, Epstein-Barr virus, and herpes zoster reactivation. Examples of urinary tract infections, upper respiratory infections, and sinusitis were also noted.
