Increased troponin is associated with poor survival in patients with amyloid light-chain (AL) amyloidosis with cardiac involvement (CAL). The purpose of this investigation was to define the relation between increased troponin and clinical, morphologic, and functional features. The comparative utility of clinical, echocardiographic, and biochemical measurements in predicting survival in CAL was also investigated. One hundred seventeen patients with CAL were divided into 2 groups: normal troponin I (<0.06 ng/ml, n = 42) or increased troponin I (≥0.06 ng/ml, n = 75). Patients in the high troponin I group were older (63 vs 58 years, p = 0.04), with higher B-type natriuretic peptide levels (1,417 vs 936 pg/ml, p = 0.0004). The high troponin I group also had higher echocardiography-determined early/late mitral inflow velocity ratio (2.2 vs 1.4, p = 0.005) and myocardial performance index (0.59 vs 0.45, p = 0.04) and lower stroke index (28 vs 38 ml/beat/m 2 , p <0.0001) and left atrial systolic force (5.9 vs 8.4 k-dynes, p = 0.037) than the normal troponin group. Median survival was significantly shorter in the high troponin group (11 vs 45 months, p <0.001). At time of CAL diagnosis, univariate predictors of all-cause mortality included increased troponin, older age, male gender, New York Heart Association class III to IV, >2 organs involved, higher B-type natriuretic peptide, lower creatinine clearance, greater ventricular septal thickness, and higher myocardial performance index. However, by multivariate Cox survival analysis, only increased troponin was a significant predictor for all-cause mortality (hazard ratio 3.1, p = 0.002). In conclusion, increased troponin is associated with worse left ventricular and left atrial functions by echocardiography in patients with CAL. Among baseline variables, it is the strongest predictor of all-cause mortality in multivariate analysis. Troponin is a powerful tool in clinical and prognostic assessments of patients with CAL.
Cardiac amyloidosis is a myocardial disease characterized by extracellular amyloid infiltration throughout the heart. The heart is pathologically affected in up to 90% of patients with systemic amyloid light-chain (AL) amyloidosis, and cardiac AL amyloidosis (CAL) is the most common cause of death in this disease. Once symptoms of congestive heart failure develop, median survival is <6 months in untreated patients, and sudden cardiac death occurs with frequency in patients with CAL despite therapy. Predictors of outcome have included clinical and echocardiographic parameters. Biomarkers have also recently been identified as powerful prognostic indicators. Dispenzieri et al reported that increased troponin might surpass symptomatic congestive heart failure and 2-dimensional echocardiography as a predictor of survival. However, although increased troponin may reflect myocyte damage because of extracellular amyloid deposition, little is known about the relation between increased troponin and cardiac structure and function. In this retrospective study, the relation of clinical and echocardiographic variables to increased troponin was determined in patients with systemic CAL. The contribution of clinical and echocardiographic variables and troponin levels to the prediction of survival in CAL was also investigated.
Methods
Records of 292 patients with AL amyloidosis who were treated at Memorial Sloan-Kettering Cancer Center from May 2001 through May 2008 were reviewed. Amyloidosis was diagnosed histologically and by demonstrable clonal cell dyscrasia. Cardiac involvement was present in 119 of 292 patients. Twenty-five patients had myocardial biopsy-proved cardiac involvement. The remaining patients had cardiac involvement defined by echocardiographic mean left ventricular wall thickness >12 mm in the absence of other potential causes of left ventricular hypertrophy or low voltage on 12-lead electrocardiogram defined by QRS complex amplitude <0.5 mV in each limb lead or <1 mV in each precordial lead in the presence of clinical, laboratory, and histologic evidence of AL amyloidosis according to standard criteria. Echocardiographic and laboratory blood tests including troponin and brain natriuretic peptide (BNP) levels were measured within 1 month of AL amyloidosis diagnosis. Of 119 patients, 2 with significant valvular heart disease were excluded from analysis; therefore, 117 patients with CAL were included in the study. All echocardiograms were centrally reviewed and interpreted at Memorial Sloan-Kettering Cancer Center.
For this analysis, patients were divided into 2 groups based on serum cardiac troponin I level at diagnosis. Group 1 consisted of 42 patients (36%) with baseline normal troponin I (<0.06 ng/ml) and group 2 comprised 75 patients (64%) with increased troponin I (≥0.06 ng/ml). Measurement of troponin I was determined in heparinized plasma samples using an AIA-PACK troponin I second-generation immunoenzyme metric assay (Tosoh Medics, Inc.). Minimum detectable concentration of cardiac troponin I at the Memorial Sloan-Kettering Cancer Center clinical chemistry laboratory is set at 0.06 ng/ml. Patient demographics and clinical information were obtained from chart review. Vital status of each patient was confirmed by review of medical records and Social Security Death Index. The primary end point was all-cause mortality. The study was approved by the Memorial Sloan-Kettering Cancer Center institutional review board.
Transthoracic echocardiography was performed using commercially available systems (IE33, Philips Medical Systems NA, Bothell, Washington; Acuson Sequoia, Siemens Medical Solution USA, Inc., Malvern, Pennsylvania; or Vivid 7, General Electric, Milwaukee, Wisconsin). Systolic and diastolic parameters were measured according to American Society of Echocardiography guidelines. Left atrial (LA) and left ventricular volumes were measured in standard fashion. Myocardial performance index (MPI) was calculated using a previously described and validated technique. Stroke index was calculated using the Doppler-based method. LA systolic force was calculated using the method of Manning et al.
Statistical analysis was performed using JMP 4.0 (SAS Institute, Cary, North Carolina) and SAS 9.1 (SAS Institute, Cary, North Carolina). The normal quintile plot method was used to determine the distribution of all variables. Variables not normally distributed were logarithmically transformed. For comparison of continuous variables between the 2 groups, an analysis of covariance model was used to adjust for potentially confounding variables. Dichotomous variables were compared by chi-square analysis using the Pearson correlation test. Results are reported as means with 95% confidence intervals or percentages. Two-tailed p value <0.05 indicates statistical significance. Survival analysis was performed using Kaplan–Meier estimates of risk.
Univariate Cox proportional hazards analysis was conducted for each baseline clinical, laboratory, and echocardiographic variable. Hazard ratios for all-cause mortality and 2-sided 95% confidence intervals were calculated separately. To assess for associations and possible interactions between baseline characteristics as predictors of death and to find independent predictors of mortality, a multivariate stepwise Cox proportional hazards regression model was used. In the multivariate model, variables with a p value <0.50 were entered into the model and those with a p value <0.10 were retained.
Results
Baseline demographic and clinical characteristics of 117 patients with CAL were compared in patients with normal versus increased troponin I values at time of diagnosis ( Table 1 ). Mean troponin I level in the latter group was 0.26 ng/ml (0.18 to 0.34). Patients with increased troponin I were older and more often men. All 6 patients with atrial fibrillation had increased troponin. The high troponin I group had significantly higher levels of BNP than the normal group, with no difference in creatinine clearance. Prevalence of New York Heart Association class III or IV heart failure or number of organs involved in AL amyloidosis at initial presentation did not differ between troponin groups.
| Characteristic | Troponin Level (ng/ml) | p Value | |
|---|---|---|---|
| <0.06 (n = 42) | ≥0.06 (n = 75) | ||
| Age (years) | 58 (55–62) | 63 (60–65) | 0.04 |
| Men | 21 (51%) | 52 (71%) | 0.02 |
| Body mass index (kg/m 2 ) | 25 (24–26) | 24 (23–26) | 0.5 |
| Heart rate (beats/min) | 75 (71–80) | 83 (80–87) | 0.02 |
| Atrial fibrillation | 0 | 6 | 0.07 |
| Brain natriuretic peptide (pg/ml) | 936 (562–1,310) | 1,417 (1,126–1,707) | 0.0004 † |
| Creatinine clearance (ml/min) | 63 (54–72) | 54 (47–61) | 0.1 |
| >2-organ involvement by amyloid ⁎ | 12 (29%) | 27 (37%) | 0.4 |
| New York Heart Association functional class III to IV | 16 (39%) | 23 (32%) | 0.4 |
⁎ More than 2 organs involved (including heart).
Table 2 lists echocardiographic variables in study patients stratified by troponin I groups. These variables were adjusted for age and gender. The high troponin I group had a higher MPI and a lower cardiac index than the normal troponin I group but left ventricular ejection fraction was similar between the 2 groups. LA systolic force, an index of LA function, was lower in the high troponin I group, with no significant difference in LA volume between the 2 groups. The high troponin I group had higher early mitral velocity and early/late mitral velocity ratio than the normal troponin I group, which may be due in part to a slightly higher heart rate.
| Characteristics | Troponin Level (ng/ml) | p Value ‡ | |
|---|---|---|---|
| <0.06 (n = 42) | ≥0.06 (n = 75) | ||
| Ventricular septum (cm) | 1.32 (1.23–1.4) | 1.42 (1.35–1.49) | 0.07 |
| Early mitral velocity (cm/s) | 59 (77.1–94.3) | 97 (91–103.3) | 0.03 |
| Late mitral velocity (cm/s) ⁎ | 69 (60.3–77.5) | 56 (50.4–64.6) | 0.03 § |
| Early/late mitral inflow velocity ratio | 1.4 (1.04–1.81) | 2.2 (1.58–2.49) | 0.005 |
| Early diastolic tissue Doppler velocity at the lateral mitral annulus (cm/s) † | 5.9 (4.71–7.01) | 6.2 (5.36–7.11) | 0.5 |
| Early mitral velocity/Early diastolic tissue Doppler velocity at the lateral annulus | 17 (13.2–20.1) | 18 (14.8–20.7) | 0.7 |
| Deceleration time (cm/s) | 166 (148–181) | 148 (135–160) | 0.1 |
| Left atrial volume indexed to body surface area (ml/m 2 ) | 36.5 (32.3–40.6) | 37.7 (34.7–38.8) | 0.6 |
| Ejection fraction (%) | 61 (57.3–64.5) | 57 (54.6–60.2) | 0.09 |
| Myocardial performance index | 0.45 (0.35–0.56) | 0.59 (0.51–0.67) | 0.04 |
| Stroke index (ml/beat/m 2 ) | 38 (33.9–42.2) | 28 (24.7–31.1) | <0.0001 |
| Cardiac index (L/min/m 2 ) | 2.9 (2.51–3.20) | 2.3 (2.03–2.56) | 0.002 |
| Left atrial systolic force (k-dynes), number (range) | 8.4 (6.8–10.4) | 5.9 (4.2–7.6) | 0.037 § |
| Stage 2/3 diastolic dysfunction † | 12 (71%) | 25 (80%) | 0.5 |
⁎ Excluding 6 Patients with atrial fibrillation.
† Data available in only 48 patients (40% with normal troponin I and 60% with increased troponin I).
Survival was significantly shorter in the high troponin I group than in the normal troponin I group (median 11 vs 45 months, p <0.001) by Kaplan–Meier estimates ( Figure 1 ). Results of univariate and multivariate Cox proportional hazard analyses are presented in Table 3 . Among individual baseline characteristics, the strongest predictor of mortality expressed by hazard ratios for all-cause mortality was increased troponin I. Other individual predictors of mortality by univariate analysis were involvement of >2 organs with AL amyloidosis, higher MPI, New York Heart Association functional class III to IV, greater ventricular septal thickness, increased BNP, decreased creatinine clearance, male gender, and older age. However, when the multivariate stepwise Cox proportional hazards model was used to evaluate interactions among all characteristics as predictors of overall mortality, only troponin I remained a significant predictor of survival (hazard ratio 3.1, p = 0.002). All other univariate predictors of outcome entered into the model were not retained by the final multivariate Cox proportional hazards model.
| Mortality Predictors | Univariate Analysis ⁎ | Multivariate Analysis | ||
|---|---|---|---|---|
| HR (95% CI) † | p Value | HR (95% CI) † | p Value | |
| Raised troponin, number (range) ‡ | 4.77 (3.12–7.29) | <0.001 | 3.10 (1.41–6.83) | 0.002 |
| Myocardial performance index | 3.36 (1.52–7.42) | 0.003 | — | — ⁎⁎ |
| >2 organs involved | 3.14 (2.09–4.69) | <0.001 | — | — ⁎⁎ |
| New York Heart Association class III to IV | 2.83 (1.84–4.35) | <0.001 | — | — ⁎⁎ |
| Ventricular septum (cm) | 2.52 (1.20–5.28) | 0.015 | — | — ⁎⁎ |
| Brain natriuretic peptide § | 2.17 (1.74–2.72) | <0.001 | — | — ⁎⁎ |
| Low creatinine clearance ∥ | 2.07 (1.39–3.07) | <0.001 | — | — ⁎⁎ |
| Male gender | 1.65 (1.12–2.43) | 0.011 | — | — ⁎⁎ |
| Older age (by decades) | 1.25 (1.04–1.51) | 0.019 | — | — ⁎⁎ |
| Stroke index (ml/beat/m 2 ) | 0.89 (0.69–1.09) | NS | — | — |
| Left atrial systolic force (k-dynes) | 1.39 (0.81–2.38) | NS | — | — |
| Lower cardiac output index ¶ | 1.37 (0.63–2.989) | NS | — | — |
| Lower deceleration time # | 0.76 (0.47–1.23) | NS | — | — |
| Early/late mitral velocity ratio | 1.26 (0.98–1.62) | NS | — | — |
| Left atrial volume indexed to body surface area (ml/m 2 ) | 1.10 (0.99–1.03) | NS | — | — |
| Ejection fraction (%) | 0.98 (0.96–1.00) | NS | — | — |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
