Methods

We considered eligible for this retrospective work all patients diagnosed with new-onset CS from December 2010 to November 2014 in 6 Finnish hospitals participating in the Myocardial Inflammatory Diseases in Finland (MIDFIN) research network. To be included, the patient had to (1) have newly diagnosed histologically proved CS in conformity with our earlier criteria, (2) be treatment naive, (3) have undergone measurements of hs-cTnT or hs-cTnI at the time diagnosis and after the start of treatment, and (4) have an estimated glomerular filtration >60 ml/min/1.73 m ² by the Modification of Diet in Renal Disease (MDRD) Study formula. A total of 62 patients fulfilled these criteria and were included in our analyses. Histologic confirmation of sarcoidosis had been made from samples of endomyocardium in 32 patients and from biopsy samples of lymph nodes in 28 patients and of skin and liver each in 1 patient. Of the 30 patients with an extracardiac histology of sarcoidosis, totally 28 had abnormalities compatible with myocardial involvement either in Gd-MRI (myocardial late enhancement, nonischemic pattern) or in F-18 FDG PET (focally increased myocardial FDG activity). The remaining 2 patients had not undergone either Gd-MRI or F-18 FDG PET but had LV ejection fraction (EF) <35% at echocardiography in the absence of signs of ischemic heart disease.

Plasma hs-TnT/I measurements were introduced into the evaluation and care of CS in Finland in December 2010. Four of the 6 contributing centers used plasma hs-cTnT determinations, whereas 2 centers measured plasma hs-cTnI. All patients included here had hs-cTnT/I measured as part of their diagnostic process at presentation, but the schedule of the measurements after the start of treatment varied depending on the local practice and on the course of each individual case. During the later follow-up of stable patients, hs-TnT/I measurements were usually repeated 2 to 3 times a year. All initial and early follow-up measurements were analyzed as part of clinical routine, and the results were available to the attending cardiologist. Some of the hs-cTnT measurements taken at outpatient visits in 1 center (Helsinki University Hospital) were analyzed later from plasma samples stored at −80°C.

The treatment of CS followed the same principles in each of the participating hospitals. Corticosteroid therapy was started once the diagnosis was established. The initial dose of prednisone was 0.5 to 1 mg/kg/day, wherefrom the dose was tapered down to 10 to 20 mg/day within 6 months. After 12 months, prednisone could be slowly discontinued if LV function was stable and there were no obvious signs of active myocardial disease. Any manifest relapses were treated by restarting prednisone or increasing its dose or by adding either azathioprine (1 to 2 mg/day) or methotrexate (10 to 20 mg/week) to low-dose prednisone therapy. The treatment response was evaluated at 3- to 6-month intervals by clinical examination, 2-dimensional echocardiographic assessment of LV structure and EF, assessment of arrhythmias and cardiac conduction, and laboratory measurements including plasma N-terminal brain natriuretic propeptide (NT-proBNP) in addition to plasma hs-cTnT/I. Repeat studies with cardiac Gd-MRI or F-18 FDG PET were not done routinely.

The details of patient characteristics, clinical CS manifestations, cardiac imaging studies, and diagnostic biopsy specimens and the results of serial measurements of plasma hs-TnT/I and NT-proBNP were collected in retrospect from hospital charts. The details and timing of serious adverse cardiac events were noted up to the end of December 2014. The MIDFIN registry study has been approved by the national ethical review board (STM/1219/2009) and by the participating hospitals’ local ethical committees. Each patient was included after an informed consent.

Plasma hs-cTnT was measured in 4 hospitals using the same commercial assay (Elecsys 2010 analyzer; Roche Diagnostics GmbHL, Mannheim, Germany). The 99th percentile value in healthy subjects was 13 ng/L, which was also the upper limit of the normal reference range by this method. Lower normal concentrations (5 to 12 ng/L) were not specified in all laboratory reports of the participating hospitals, and therefore, for consistency of the scale, all these values were included as <13 ng/L (i.e., “normal”) in statistical analyses. Plasma hs-cTnI was measured in 2 hospitals using 1 commercial assay (TnI-Ultra assay on the ADVIA Centaur XP immunoanalyzer; Siemens Healthcare Diagnostics, Erlangen, Germany). The lowest reported concentration was 0.01 ng/ml, and the upper normal limit, representing the ninety-ninth percentile value, was 0.04 ng/ml. Plasma NT-proBNP was determined by a commercially available assay (Elecsys 2010 analyzer; Roche Diagnostics).

Continuous variables are presented as mean ± SD or as median (minimum to maximum). Comparisons between groups were performed using the chi-square test or Fisher’s exact test for categorical data and analysis of variance or Mann-Whitney test for continuous variables. Within-group comparisons were made using Wilcoxon’s signed-ranks test and McNemar’s test. The predefined end point for the assessment of time-dependent outcome was a composite of cardiac death, aborted sudden cardiac death (ventricular fibrillation treated by external or internal defibrillation), cardiac transplantation, sustained ventricular tachycardia (VT), and new high-grade atrioventricular conduction block. Survival free of the composite end point was analyzed by the Kaplan-Meier method using the log-rank test and the Cox regression analysis. The follow-up time was calculated from the hs-TnT/I measurement taken at presentation to the occurrence of an end point event or to the end of data collection. All tests were 2 sided with a level of significance set at p <0.05. The analyses were performed with SPSS, version 22 for Windows (SPSS Inc., Chicago, Illinois).

Results

Plasma hs-cTnT, measured in 50 patients, ranged from ≤13 to 240 ng/L at presentation (median, 15 ng/L) and was abnormally elevated (>13 ng/L) in 26 patients (52%). Hs-cTnI, measured in the remaining 12 patients, ranged from 0.01 to 1.30 ng/ml (median, 0.05 ng/ml) and was elevated (>0.04 ng/ml) in 7 patients (58%). Table 1 summarizes the characteristics of the study population and compares the subgroups of normal versus elevated hs-cTnT/I. The data show that elevated hs-cTnT/I was associated with poorer LV function and a tendency toward less high-grade heart block and more heart failure as the presenting manifestation.

Stay updated, free articles. Join our Telegram channel

Join