Detection of cardiac sarcoidosis using cardiac markers and myocardial integrated backscatter

International Journal of Cardiology 102 (2005) 259 – 268 www.elsevier.com/locate/ijcard

Detection of cardiac sarcoidosis using cardiac markers and myocardial integrated backscatter Hiroko Yasutake*, Yoshihiko Seino, Mutsumi Kashiwagi, Hiroshi Honma, Tsuyako Matsuzaki, Teruo Takano First Department of Medicine, Nippon Medical School, 1-1-5 Sendagi, Bunkyo-ku, Tokyo 113-8603, Japan Received 24 December 2003; received in revised form 12 April 2004; accepted 5 May 2004 Available online 28 August 2004

Abstract Backgrounds: It is not known whether cardiac markers and cyclic variations of integrated backscatter can be used to detect cardiac sarcoidosis. Methods: We studied 62 patients with sarcoidosis affecting the lung, eyes, skin, or heart (27 patients with cardiac involvement and 35 patients without). The cyclic variation of integrated backscatter and wall thickening was evaluated in the left ventricular anterior septum and posterior wall. Plasma A-type natriuretic peptide (ANP) and B-type natriuretic peptide (BNP) concentrations and serum cardiac troponin T were also determined. Results: Plasma natriuretic peptide concentrations were higher in the cardiac involvement group (ANP: 15.5 [interquartile range (IQR) 2.5– 34.0] vs. 12.0 [10.0–16.5] pg/ml, P=0.25; BNP: 28.6 [5.9–141] vs. 10.1 [4.8–15.4] pg/ml, P=0.049). However, cardiac troponin T concentration was b0.01 ng/ml in all patients. Receiver–operator characteristic (ROC) analysis showed that both ANP and BNP could identify patients with high-degree atrioventricular block, ventricular tachyarrhythmias, or symptomatic heart failure (the areas under the ROC curve were 0.94 and 0.97, respectively). The cardiac involvement group could be distinguished from the noninvolvement group by combining cutoff values for the magnitude of integrated backscatter cyclic variation (5.5 dB) and wall thickening (30%), albeit only for the posterior wall. Conclusion: Both ANP and BNP are useful markers for identifying patients with sarcoidosis and cardiac complication(s). Moreover, evaluation of integrated backscatter cyclic variation combined with wall thickening may be of help in detecting cardiac involvement in the posterior wall. D 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Cardiac sarcoidosis; Integrated backscatter; Natriuretic peptide; Echocardiography

1. Introduction Sarcoidosis is a systemic granulomatous disease of unknown etiology, which commonly affects the lungs, eyes, or skin with a favorable prognosis. However, once the heart is involved, unfavorable cardiac complications occur in many patients. These complications include high-degree atrioventricular block, ventricular tachyarrhythmias, and congestive heart failure, which often result in sudden * Corresponding author. Tel.: +81 3 3822 2514; fax: +81 3 5685 0987. E-mail address: [email protected] (H. Yasutake). 0167-5273/$ - see front matter D 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2004.05.028

cardiac death [1]. Accordingly, earlier detection of cardiac involvement is essential to reduce such cardiac complications because the disease is treatable with corticosteroids and other immunosuppressive agents [1]. Holter monitoring [2], two-dimensional echocardiography [3,4], and radionuclide myocardial imaging [5,6] are generally used to detect cardiac involvement, but the cardiac lesions are still often overlooked because of their subclinical disease progression. It has recently been shown that more sophisticated diagnostic techniques, such as gadolinium diethylene triamine pentaacetic acid (DTPA)-enhanced magnetic resonance imaging [7–9] and fluorine-18-deoxyglucose positron

260

H. Yasutake et al. / International Journal of Cardiology 102 (2005) 259–268

emission tomography [10,11] may be used in both diagnosing cardiac involvement and evaluating the disease activity in response to steroid therapy. However, these methods are technically demanding and expensive, and simpler and more cost-effective methods are needed. One possible method is the measurement of integrated ultrasonic backscatter—a relatively new, noninvasive measure of the acoustic properties of the myocardium. This technique provides a novel approach for defining the physical state of cardiac muscle tissue that complements the assessment of ventricular wall motion and chamber dimensions by conventional two-dimensional echocardiography [12]. Recent studies have demonstrated that myocardial tissue characterization with integrated backscatter can be used to evaluate various pathologic conditions, such as ischemic/reperfusion injury [12–14], hypertrophied myocardium [15,16], dilated cardiomyopathy [17,18], diabetic heart disease [19], or cardiac amyloidosis [20]. We previously reported that the magnitude of integrated backscatter cyclic variations was significantly smaller in a patient with cardiac sarcoidosis [21]. Interestingly, it normalized in response to steroid therapy much earlier than the systolic function improved. There are several cardiac markers that may be useful in diagnosing cardiac sarcoidosis. A-type natriuretic peptide (ANP) and B-type natriuretic peptide (BNP) have been widely used as simple and useful markers of left ventricular dysfunction and prognosis in patients with chronic heart failure [22]. Likewise, serum concentration of cardiac troponin T, a specific marker for myocardial cell injury, has been shown to reflect the severity and prognosis of heart failure [23–25]. Moreover, there is evidence that these cardiac markers may be useful for the early detection of cardiac damage or chronic heart failure in patients who are asymptomatic [26–29]. In the present study, we determined whether echocardiographic parameters based on the integrated backscatter cyclic variation and cardiac markers may be used to detect myocardial involvement of sarcoidosis.

(cardiac involvement group, n=27) and the other without it (noncardiac involvement group, n=35). Of the 53 asymptomatic patients, 18 patients developed cardiac complications and three of them eventually required treatment for the complications—two patients required pacemaker implantation for complete atrioventricular block and one patient with a pacemaker implanted before entering the study required steroid therapy for heart failure. All the nine symptomatic patients met the criteria for cardiac sarcoidosis: five of them required pacemaker implantation for complete atrioventricular block, two of them needed implantable cardioverter– defibrillator implantation for ventricular tachycardia, and seven of them received steroid therapy (Table 1). We used the diagnostic criteria for cardiac sarcoidosis developed by the Specific Diffuse Pulmonary Disease Research Group, Sarcoidosis Division of the Japanese Ministry of Health, Labor, and Welfare [7,30]. The criteria include (category 1) histologic confirmation of sarcoidosis in the heart or (category 2) a histologic diagnosis of extracardiac sarcoidosis with (a) the presence of electrocardiographic abnormalities as an essential item plus one or more of the following clinical findings: (b) left ventricular asynergy, local wall thinning, or hypertrophy on echocardiograms; (c) abnormal tracer uptake or perfusion defects on radionuclide examinations; (d) intracardiac pressure abnormalities during cardiac catheterization, or wall motion abnormalities or decreased left ventricular ejection fraction on left ventriculography; and (e) interstitial fibrosis or cell infiltration noted in endomyocardial biopsies, even if the findings are nonspecific. The electrocardiographic abnormalities could include right bundle branch block, left axis deviation, atrioventricular block, ventricular tachycardia, ventricular premature contractions (zLown grade 2), abnormal Q waves, or ST–T changes on standard 12-lead tracing or Holter monitor. Category 1 findings were present in six patients (five based on endomyocardial biopsy and one based on postmortem examination). Informed consent was obtained from all subjects, and they were confirmed to be free from renal insufficiency.

2. Methods

2.2. Measurement of cardiac markers

2.1. Patient selection and diagnosis of cardiac sarcoidosis

Venous blood samples were obtained on the first visit to our clinic for the measurement of ANP and BNP and cardiac troponin T. Plasma ANP and BNP concentrations were determined with a specific immunoradiometric assay for alpha-human ANP and human BNP, respectively, using commercial kits (Shionoria kit; Shionogi and Kyowa Medex, Tokyo, Japan). The performance characteristics of Shionoria ANP kit are %CV=5.2–8.5% (n=5) with analytical range of 18–490 pg/ml, and the corresponding value for Shionoria BNP kit is 2.5–4.3% (n=10) with analytical range of 4–2000 pg/ml. Cross-reactivities with other natriuretic peptides were: ANP assay, human ANP 100% and human BNP 0.001%; and BNP assay, human BNP 100% and human ANP 0.001%. The serum concentration of

We studied 62 consecutive patients (15 men and 47 women, age 53F16 years, meanFS.D.) who were recruited between January 1995 and September 2000. Fifty-three patients had either pulmonary, cutaneous, or ocular sarcoidosis, and were referred to our clinic for further examination of possible cardiac involvement although they had no cardiac symptoms. The remaining nine patients who visited our clinic had cardiac symptoms, with one or more complaints due to atrioventricular block, ventricular arrhythmias, or congestive heart failure. All patients met diagnostic criteria (detailed below) for the presence of cardiac sarcoidosis and were divided into two groups: one with cardiac involvement

H. Yasutake et al. / International Journal of Cardiology 102 (2005) 259–268

261

Table 1 Demographics and clinical characteristics of patients with cardiac sarcoidosis

Age (years) Male sex Involved organ Heart Lung Eye Skin Cardiothoracic ratio Use of corticosteroids Composite of cardiac complications Atrioventricular block requiring hospitalization Ventricular tachyarrhythmias requiring hospitalization Pacemaker implantation Congestive heart failure requiring hospitalization Cardiac death a

Patients without cardiac involvement (n=35) [number (%) or meanFS.D.]

Patients with cardiac involvement (n=27) [number (%) or meanFS.D.]

P value

53F15 5 (14)

54F18 10 (36)

0.81 0.076

0 27 (77) 22 (63) 8 (23) 47F5 7 (20) 0 0

27 (100) 19 (70) 15 (56) 3 (11) 53F8 8 (30) 12 (44) 7 (26)

– 0.55 0.75 0.39 0.005 0.56 b0.0001 0.005

0

5 (19)

0.029

0 0

10 (37) 2a (7) 3 (11)

0.0003 0.36a 0.15

0

1 (4)

0.90

Implantable cardioverter–defibrillator.

cardiac troponin T was measured using a commercially available electrochemiluminescence immunoassay (Elecsys 1010/2010 Troponin T Stat; Roche Diagnostics, Mannheim, Germany). The minimum detectable concentration of cardiac troponin T was 0.01 ng/ml.

determined as the average in three consecutive cardiac cycles of the differences between the minimal and maximal values of the acoustic signals. The time for backscatter signal to recover from the nadir value by 2 dB corrected by the heart rate (bcorrected recovery timeQ; ms) was also calculated.

2.3. Determination of echocardiographic parameters

2.4. Radionuclide cardiac imaging for cardiac involvement

All patients were subjected to echocardiographic examination. Ultrasonic backscatter signal was measured with a commercially available two-dimensional imaging system equipped with an acoustic densitometry software (SONOS 2500; Philips Medical Systems, Japan). With this system, either conventional two-dimensional and M-mode echo amplitude or integrated backscatter images could be obtained. For analysis of wall motion, all four standard echocardiographic views (parasternal long and short axis; apical four- and two-chamber views) were imaged with a 2.5- or 3.5-MHz transducer. All segments were subjected to a visual scoring system. The left ventricular ejection fraction and percent wall thickening of the anterior septum and posterior wall were calculated using M-mode recordings of the parasternal long-axis view at the papillary muscle level. Integrated backscatter data were analyzed offline for each patient. An elliptical region of interest was placed in the midmyocardial region in each segment analyzed. The region of interest was tracked manually on a frame-by-frame basis for the entire loop of 60 frames. The analysis package displayed a mean power of the integrated backscatter radiofrequency signals in the region of interest in decibels (dB) against the entire loop. This allows assessing cardiac cycledependent variation of the integrated backscatter signals. The magnitude of integrated backscatter cyclic variation was

During the process of evaluating myocardial involvement, myocardial thallium-201 scintigraphy was performed in 31 patients. The tomographic vertical long-axis, horizontal long-axis, and short-axis images were reconstructed. Segmental uptakes were blindly graded by an experienced radiologist. Ten patients had focal areas of decreased thallium uptake. In eight of these patients, the finding was taken as supportive evidence for the presence of cardiac lesions. In the remaining two, it was not used because they did not have any electrocardiographic abnormalities (category 2a). When echocardiographic parameters were compared between the cardiac involvement group and the noncardiac involvement group, data from patients who had abnormal myocardial perfusion scans were segregated and used for comparison. Coronary artery lesions were excluded in all eight cardiac sarcoidosis patients with abnormal thallium uptake (seven by coronary angiography and the remaining one by postmortem examination). Gallium-67 scintigraphy was performed in 15 patients; however, it was not used to diagnose cardiac sarcoidosis. 2.5. Statistical analysis The concentrations of natriuretic peptides were expressed as median with interquartile range (IQR). Otherwise, data

262

H. Yasutake et al. / International Journal of Cardiology 102 (2005) 259–268

are expressed as the meanFS.D. Gaussian distribution variables were compared using Student’s t test. When data were not normally distributed, the Mann–Whitney U test was utilized. Binomially distributed variables were compared using a chi-square test for a 22 table. The diagnostic utility of ANP and BNP concentrations was analyzed with receiver–operator characteristic (ROC) curves. The appropriate cutoff points were selected such that the value for sensitivity was as close as possible to the value for specificity. The results are expressed as the area under the ROC curve and 95% confidence intervals for this area. A value of Pb0.05 was considered statistically significant.

3. Results 3.1. Natriuretic peptides and cardiac troponin T BNP was determined in 44 patients and ANP determined in 37 on their first visit. The plasma BNP concentration was

significantly higher in the cardiac involvement group than in the noncardiac involvement group. In contrast, there was no statistically significant difference in left ventricular ejection fraction estimated by echocardiography between the two groups (68.5F5.0% in the noncardiac involvement group vs. 63.5F12.8% in the cardiac involvement group, P=0.11). The plasma ANP concentration also tended to be higher in the cardiac involvement group, but the difference was not statistically significant (Fig. 1A). Left ventricular ejection fraction for these patients was 68.3F5.0% in the noncardiac involvement group and 65.2F11.9% in the cardiac involvement group ( P=0.32). The ROC analysis indicated that both ANP and BNP levels had a poor diagnostic accuracy for cardiac involvement [area under the curve: 0.61 (95% confidence interval: 0.45–0.77) for ANP, 0.67 (0.52–0.81) for BNP; Fig. 2A]. If a cutoff value for ANP concentration was set at 20 pg/ml, the sensitivity, specificity, positive predictive value, and negative predictive value for cardiac involvement were 38.9%, 94.7%, 87.5%, and 62.0%, respectively. Likewise, if a cutoff value for BNP concen-

Fig. 1. Plasma concentrations of ANP and BNP (pg/ml) in patients with (closed circles) or without (open circles) cardiac involvement of sarcoidosis at the time of the first visit (panel A). In panel B, both peptide concentrations were compared in patients with or without cardiac complications, such as advanced atrioventricular block, ventricular tachyarrhythmias, or congestive heart failure. Data are expressed as median with IQR (indicated by side bars).

H. Yasutake et al. / International Journal of Cardiology 102 (2005) 259–268

263

Fig. 2. ROC curves using ANP or BNP to detect cardiac involvement of sarcoidosis (panel A) or cardiac event(s), including advanced atrioventricular block, ventricular tachyarrhythmias, or congestive heart failure (panel B). Appropriate cutoff values are reported in picograms per milliliter. The numbers in parenthesis are the 95% confidence intervals for the area under the ROC curve (AUC).

tration was set at 20 pg/ml, the corresponding values were 52.4%, 91.3%, 84.6%, and 67.7%, respectively. When the patients in the cardiac involvement group were subgrouped in terms of cardiac complications, including high-grade atrioventricular block, ventricular tachyarrhythmias, or heart failure, the plasma concentrations of both peptides were significantly higher in patients with the cardiac complications than in those without (Fig. 1B). In the ROC analysis, both ANP and BNP concentrations were found to have an excellent ability to discriminate patients with one or more cardiac complications from those without [area under the curve: 0.94 (95% confidence interval: 0.85– 1.04) for ANP, 0.97 (0.90–1.03) for BNP; Fig. 2B]. With a cutoff value of 20 pg/ml for ANP concentration, the sensitivity, specificity, positive predictive value, and negative predict value for those complications were 100%, 91.7%, 85.7%, and 100%, respectively. Likewise, the corresponding values for BNP concentration at a cutoff value of 75 pg/ml were 87.5%, 100%, 100%, and 92.9%, respectively. In addition, the left ventricular ejection fraction of the patients with such complications was significantly lower than in those without them (55F16.7% vs. 69.0F8.7%, P=0.0094). Serum concentration of cardiac troponin T was determined in 31 patients on their first visit; however, they were all lower than the detection limit (b0.01 ng/ml). One patient

developed severe progressive heart failure and died. Elevation of this myofibril marker was noted only in this patient in the late stage. 3.2. Echocardiographic parameters Echocardiographic parameters for the anterior septum and the posterior wall are compared independently in Table 2. The values for the magnitude of integrated backscatter cyclic variation and percent wall thickening were lower in the cardiac involvement group than in the noncardiac involvement group, but the differences were not significant. Corrected recovery time for the posterior wall was significantly longer in the cardiac involvement group. However, this was not the case in the anterior septum. Because thallium-201 myocardial imaging is known to be useful for evaluating spatial distribution of cardiac sarcoidosis, we subgrouped the data of patients with perfusion defects in the cardiac involvement group and compared those data with data for the noncardiac involvement group (Fig. 3). Based on this analysis, the difference in the corrected recovery time of backscatter signal in the posterior wall became more discrete (267.7F105.0 vs. 443.8F128.4 ms, P=0.0012), despite no significant improvement in the discriminating power of the magnitude of cyclic variation. Moreover, the difference in the percent wall thickening of the posterior wall became statisti-

264

H. Yasutake et al. / International Journal of Cardiology 102 (2005) 259–268

Table 2 Echocardiographic parameters of patients with cardiac sarcoidosis

Left ventricular ejection fraction (%) Anterior septum of the left ventricle Percent wall thickening (%) Magnitude of cyclic variation in integrated backscatter (dB) Corrected recovery time of integrated backscatter (ms) Posterior wall of the left ventricle Percent wall thickening (%) Magnitude of cyclic variation in integrated backscatter (dB) Corrected recovery time of integrated backscatter (ms)

Patients without cardiac involvement [meanFS.D. (n)]

Patients with cardiac involvement [meanFS.D. (n)]

P value

67.9F6.0 (35)

62.5F14.4 (27)

0.076

24.8F12.8 (18) 5.4F1.6 (32)

21.7F10.9 (21) 5.0F1.6 (27)

0.42 0.32

370.2F124.1 (31)

397.3F132.5 (24)

0.44

38.7F11.3 (18) 6.1F1.8 (32)

31.7F15.9 (21) 5.9F2.0 (27)

0.12 0.66

267.7F105.0 (32)

344.5F153.1 (27)

0.027

cally significant (38.7F11.3% vs. 15.5F9.1%, P=0.003). In contrast, the parameters for the anteroseptal wall did not show any significant difference between the groups. The relationships between percent wall thickening and integrated backscatter parameters (magnitude of cyclic variation and corrected recovery time) are shown separately for the anterior septum or posterior wall in Fig. 4. In the

ventricular septum, there was a significant correlation between the magnitude of integrated backscatter cyclic variation and percent wall thickening in the cardiac involvement group (r=0.510, P=0.018), but no significant correlation was found between the corrected recovery time and percent wall thickening. Because the data distributions for the two groups were similar along both the horizontal and

Fig. 3. Comparison between patients without cardiac sarcoidosis and those with cardiac involvement detected by thallium-201 myocardial imaging. Open circles indicate patients who are free of myocardial perfusion defects; closed circles represent those with decreased uptake. Panel A shows the data from the anterior septum; panel B shows the data from the posterior wall of the left ventricle.

H. Yasutake et al. / International Journal of Cardiology 102 (2005) 259–268

265

Fig. 4. Relationship between percent wall thickening and magnitude of integrated backscatter cyclic variation (dB) or corrected recovery time of backscatter signal in the anterior septum (panel A) and posterior wall of the left ventricle (panel B). Closed circles represent data from patients with cardiac involvement; open circles represent data from those without. E identifies patients with cardiac event(s); E+T identifies patients with cardiac event(s) after steroid therapy had been started for noncardiac reasons.

vertical axes, these parameters could not identify cardiac sarcoidosis patients (Fig. 4A). In contrast, data from the noncardiac involvement group for the posterior wall localized to the right upper quadrant partitioned by the cutpoints of 5.5 dB (median) for magnitude of cyclic variation and 30% (median) for percent wall thickening (Fig. 4B). Accordingly, the sensitivity and specificity for detecting cardiac involvement using either parameter or their combination were compared (Table 3). The combination of the variables appeared to improve the diagnostic ability in a cooperative manner, such that the union of magnitude of integrated backscatter cyclic variation b5.5 dB and percent wall thickening b30% gave a sensitivity of 75% and the intersection of those a specificity of 100%. A similar Table 3 Diagnostic value for cardiac involvement of percent wall thickening and/or magnitude of integrated backscatter cyclic variation in the left ventricular posterior wall Parameters (n=38)

Sensitivity (%)

Percent wall thickening (b30%) Magnitude of cyclic variation in integrated backscatter (b5.5 dB) Union of both parameters Intersection of both parameters

60 50

83 72

75 35

56 100

discrimination can be achieved using the corrected recovery time and percent wall thickening if the cutoff point is set at 300 ms (median) for corrected recovery time of backscatter signal. No significant correlation was found between the integrated backscatter parameters and percent wall thickening in either group. 3.3. Relation between natriuretic peptide concentrations and echocardiographic parameters We examined the relationships between plasma concentrations of ANP or BNP and each of the three echocardiographic parameters for the anteroseptal or posterior wall independently. No significant correlation was found between the plasma concentration of ANP and any of the three parameters. The plasma concentration of BNP was significantly correlated with percent wall thickening for the posterior wall (r= 0.48, P=0.010), but not with any other parameters (data not shown).

Specificity (%)

4. Discussion The present study is the first to identify a role for cardiac markers (ANP, BNP, and cardiac troponin T) and tissue

266

H. Yasutake et al. / International Journal of Cardiology 102 (2005) 259–268

characterization by integrated backscatter in discriminating cardiac involvement of sarcoidosis. Our data suggest that plasma concentrations of ANP and BNP may be useful in spotting sarcoidosis patients with unfavorable cardiac complications, such as atrioventricular block, severe ventricular arrhythmias, or congestive heart failure, rather than in diagnosing cardiac involvement of sarcoidosis per se. In contrast, the determination of serum cardiac troponin T was of no clinical use in our patients studied. The data also suggest that both left ventricular percent wall thickening and tissue characterization using integrated backscatter cyclic variation may play a role in detecting cardiac involvement of sarcoidosis, but only in the posterior wall. 4.1. Role of natriuretic peptides It has been shown that the plasma concentrations of ANP and BNP are higher than normal in patients with overt heart failure and, to a lesser extent, in patients with asymptomatic cardiac impairment [22,26,27,31]. Because cardiac involvement of sarcoidosis may affect cardiac contractility, it is reasonable to assume that these peptides would be useful markers for diagnosing cardiac sarcoidosis. However, the diagnostic utility of ANP and BNP for detecting cardiac sarcoidosis was not favorable, as shown by the ROC analysis in our study. The high specificity and positive predictive value, albeit low sensitivity and negative predictive value, of ANP and BNP concentrations in diagnosing cardiac sarcoidosis meant that the presence of cardiac involvement was highly likely if the plasma concentration of those peptides was N20 pg/ml. However, the reverse was not true because 50–60% of the patients with cardiac sarcoidosis would be misdiagnosed even if peptide concentrations were V20 pg/ml. The reason for this may be related to the patient population. In this study, 85% of the patients were asymptomatic and their left ventricular ejection fraction was generally within the normal range. As a result, there was no significant difference in left ventricular ejection fraction between the cardiac involvement and noncardiac involvement groups. There also was no significant correlation between plasma concentrations of either peptide and left ventricular ejection fraction, although a weak significant correlation was found between BNP concentration and left ventricular ejection fraction only in patients with a left ventricular ejection fraction b0.50 (r= 0.373, P=0.013; data not shown). However, if patients were subgrouped based on the presence of cardiac complications, such as high-grade atrioventricular block, ventricular tachyarrhythmias, or symptomatic heart failure, both peptides were of great value in differentiating patients with one or more of those complications from the other group. Left ventricular ejection fraction was significantly lower in those patients, and the plasma concentrations of ANP and BNP were elevated. Interestingly, of all 12 patients with cardiac event(s), three were symptom-free on their first visit, but their plasma

concentrations of those peptides were similar to those in the nine symptomatic patients. This may suggest that both ANP and BNP are useful predictors for those cardiac complications in patients with cardiac sarcoidosis. The cutoff values used for the discrimination were 20 pg/ml for ANP and 75 pg/ml for BNP, which are fairly comparable to the values used in previous studies of heart failure [32–34]. Therefore, elevations of these peptide concentrations appear to reflect left ventricular systolic dysfunction, whether the patients were symptomatic or asymptomatic. Moreover, of the 12 patients with those complications, seven of them received steroid therapy to treat heart failure and/or severe ventricular arrhythmias. This supports the idea that determining plasma concentrations of these peptides could be useful in detecting high-risk patients who require steroid therapy. In light of this, a retrospective multicenter study showed that New York Heart Association functional class, left ventricular enddiastolic diameter, and sustained ventricular tachycardia were independent predictors of mortality for patients with cardiac sarcoidosis, and that initiating corticosteroid therapy before the development of systolic dysfunction results in an excellent clinical outcome [35]. 4.2. Role of integrated backscatter parameters As mentioned previously, cyclic variation of integrated backscatter has been shown to be significantly reduced under various pathophysiologic conditions [12–20]. Although the mechanisms responsible for the phenomenon have not been fully elucidated, several factors, including regional myocardial contractile function, orientation of myocardial fibers, and structural changes of the myocardium, have been suggested to contribute to the changes in integrated backscatter cyclic variation. In the present study, it was anticipated that the percent wall thickening and magnitude of cyclic variation would be smaller and corrected recovery time would be longer (as a result of a reduction in the magnitude of cyclic variation) in patients with cardiac sarcoidosis. However, a significant difference was found only in the corrected recovery time in the posterior wall. Although the difference became greater when the patients with thallium-201 perfusion defects were compared to those without cardiac involvement (Fig. 3), these findings were valid only for the posterior wall, not the anterior septum. The reason for this is uncertain. It might be that the posterior wall had been affected by sarcoid more than the anterior septum, but it was unlikely because thallium perfusion defects occurred equally in both areas. Differences in the anisotropy of the myocardium between the septal area and the posterior wall may provide a possible clue to our understanding. Cardiac lesions of sarcoidosis, including sarcoid granuloma, disarrangement, myocyte hypertrophy, fragmentation of muscle bundles, interstitial edema, large mononuclear cell infiltration, and myocardial interstitial fibrosis, may well increase anisotropy of myocardial fiber orientation [1], thereby reducing

H. Yasutake et al. / International Journal of Cardiology 102 (2005) 259–268

cyclic variation of integrated backscatter. Because the anisotropy of the myocardium may be greater in the anterior septum than in the posterior wall [36], in keeping with the smaller magnitude of integrated backscatter cyclic variation of the anterior septum [15,16,19,20,36], the ability to detect changes in integrated backscatter cyclic variation might be less for the anterior septum than for the posterior wall. It is interesting to note that, in patients with cardiac amyloidosis, integrated backscatter cyclic variation of the left ventricular posterior wall (not at the interventricular septum) is a powerful predictor of clinical outcome and is superior to standard echocardiographic/Doppler flow indexes [20]. There was a weak correlation between the parameter for tissue characterization (magnitude of integrated backscatter cyclic variation) and that for regional contractile function (percent wall thickening), in keeping with previous studies [15,37,38]. This implies that regional myocardial function contributed, at least in part, to the changes in integrated backscatter cyclic variation in our patients. As illustrated in Fig. 4B and Table 3, the combination of these parameters improved the diagnostic accuracy for detecting cardiac involvement in the posterior wall. It is noteworthy that the patients who received steroid therapy (E+T in Fig. 4) when they had the cardiac event(s) appeared to have higher values for percent wall thickening and magnitude of cyclic variation (or lower values for corrected recovery time) than those who did not receive treatment (E), suggesting the importance of steroid therapy in patients with sarcoidosis having cardiac event(s). Taken together with our previous report [21], it is necessary to determine whether these parameters in such patients would improve with steroid therapy and therefore may be used to adjust the dose of corticosteroids. This clinical application of these parameters also seems to be limited to the posterior wall for unclear reasons. As discussed above, clinical implications of plasma concentrations of the natriuretic peptides and the echocardiographic parameters appear to be different, in keeping with the finding of no significant correlations between the two diagnostic approaches, except for a weak correlation between the plasma concentration of BNP and percent wall thickening for the posterior wall. Both peptides would be useful for monitoring disease activities and echocardiographic parameters for early detection of posterior involvement. 4.3. Study limitations and implications Since there is no gold standard for the diagnosis of cardiac sarcoidosis, we adopted the diagnostic criteria developed by the Specific Diffuse Pulmonary Disease Research Group, Sarcoidosis Division of the Japanese Ministry of Health, Labor, and Welfare, which have been commonly used in previous studies [7,30]. Accordingly, 21 patients were diagnosed as having cardiac sarcoidosis without definitive evidence from endomyocardial biopsy

267

(category 2). In addition, this was an observational study carried out at a single hospital with a small sample size. Besides, the data might have been biased in several ways (referral and selection biases). Therefore, one should interpret the results with caution. Cardiac sarcoidosis has a substantial diversity of disease activity and spatial distribution (bpatchy involvementQ), which makes early detection difficult. It occasionally takes serious clinical course with a recognizable latent phase that has an effective treatment known to decrease both mortality and progression. When cardiac involvement is suspected, regardless of cardiac symptoms, measurement of plasma concentrations of the both natriuretic peptides and echocardiographic tissue characterization as well as regional wall motion should be performed. Whether these diagnostic tests can reflect the disease activity and thereby be a useful marker for treatment is worthy of further investigation.

Acknowledgements The authors thank the physicians in the Departments of Pulmonology, Dermatology, and Ophthalmology for referring patients, and Dr. Masahiro Yasutake for many valuable discussions and assistance in manuscript preparation.

References [1] Sekiguchi M, Yazaki Y, Isobe M, Hiroe M. Cardiac sarcoidosis: diagnostic, prognostic, and therapeutic considerations. Cardiovasc Drugs Ther 1996;10:495 – 510. [2] Suzuki T, Kanda T, Kubota S, Imai S, Murata K. Holter monitoring as a noninvasive indicator of cardiac involvement in sarcoidosis. Chest 1994;106:1021 – 4. [3] Lewin RF, Mor R, Spitzer S, Arditti A, Hellman C, Agmon J. Echocardiographic evaluation of patients with systemic sarcoidosis. Am Heart J 1985;110:116 – 22. [4] Burstow DJ, Tajik AJ, Bailey KR, DeRemee RA, Taliercio CP. Twodimensional echocardiographic findings in systemic sarcoidosis. Am J Cardiol 1989;63:478 – 82. [5] Yamamoto N, Gotoh K, Yagi Y, et al. Thallium-201 myocardial SPECT findings at rest in sarcoidosis. Ann Nucl Med 1993;7:97 – 103. [6] Okayama K, Kurata C, Tawarahara K, Wakabayashi Y, Chida K, Sato A. Diagnostic and prognostic value of myocardial scintigraphy with thallium-201 and gallium-67 in cardiac sarcoidosis. Chest 1995;107: 330 – 334. [7] Shimada T, Shimada K, Sakane T, et al. Diagnosis of cardiac sarcoidosis and evaluation of the effects of steroid therapy by gadolinium-DTPA-enhanced magnetic resonance imaging. Am J Med 2001;110:520 – 7. [8] Vignaux O, Dhote R, Duboc D, et al. Detection of myocardial involvement in patients with sarcoidosis applying T2-weighted, contrast-enhanced, and cine magnetic resonance imaging: initial results of a prospective study. J Comput Assist Tomogr 2002;26: 762 – 767. [9] Skold CM, Larsen FF, Rasmussen E, Pehrsson SK, Eklund AG. Determination of cardiac involvement in sarcoidosis by magnetic resonance imaging and Doppler echocardiography. J Intern Med 2002;252:465 – 71.

268

H. Yasutake et al. / International Journal of Cardiology 102 (2005) 259–268

[10] Okumura W, Iwasaki T, Ueda T, et al. Usefulness of 18F-FDG PET for diagnosis of cardiac sarcoidosis. Kaku Igaku 1999;36:341 – 8. [11] Takeda N, Yokoyama I, Hiroi Y, et al. Positron emission tomography predicted recovery of complete A–V nodal dysfunction in a patient with cardiac sarcoidosis. Circulation 2002;105:1144 – 5. [12] Milunski MR, Mohr GA, Perez JE, et al. Ultrasonic tissue characterization with integrated backscatter. Acute myocardial ischemia, reperfusion, and stunned myocardium in patients. Circulation 1989; 80:491 – 503. [13] Naito J, Masuyama T, Yamamoto K, et al. Myocardial integrated ultrasonic backscatter in patients with old myocardial infarction: comparison with radionuclide evaluation. Am Heart J 1996;132: 54 – 60. [14] Hancock JE, Cooke JC, Chin DT, Monaghan MJ. Determination of successful reperfusion after thrombolysis for acute myocardial infarction: a noninvasive method using ultrasonic tissue characterization that can be applied clinically. Circulation 2002;105:157 – 61. [15] Masuyama T, St. Goar FG, Tye TL, Oppenheim G, Schnittger I, Popp RL. Ultrasonic tissue characterization of human hypertrophied hearts in vivo with cardiac cycle-dependent variation in integrated backscatter. Circulation 1989;80:925 – 34. [16] Ueda K, Murata K, Tanaka N, et al. Ultrasonic myocardial tissue characterization in patients with hypertrophic cardiomyopathy and pressure-overloaded hypertrophy by backscattered energy temporal analysis. Circ J 2002;66:729 – 34. [17] Vered Z, Barzilai B, Mohr GA, et al. Quantitative ultrasonic tissue characterization with real-time integrated backscatter imaging in normal human subjects and in patients with dilated cardiomyopathy. Circulation 1987;76:1067 – 73. [18] Suwa M, Ito T, Kobashi A, et al. Myocardial integrated ultrasonic backscatter in patients with dilated cardiomyopathy: prediction of response to beta-blocker therapy. Am Heart J 2000;139:905 – 12. [19] Perez JE, McGill JB, Santiago JV, et al. Abnormal myocardial acoustic properties in diabetic patients and their correlation with the severity of disease. J Am Coll Cardiol 1992;19:1154 – 62. [20] Koyama J, Ray-Sequin PA, Falk RH. Prognostic significance of ultrasound myocardial tissue characterization in patients with cardiac amyloidosis. Circulation 2002;106:556 – 61. [21] Honma H, Kishida H, Tsuchida T, et al. A 68-year-old man with complete atrioventricular block and congestive heart failure. J Cardiol 1999;33:111 – 5. [22] Boomsma F, van den Meiracker AH. Plasma A- and B-type natriuretic peptides: physiology, methodology and clinical use. Cardiovasc Res 2001;51:442 – 9. [23] Setsuta K, Seino Y, Takahashi N, et al. Clinical significance of elevated levels of cardiac troponin T in patients with chronic heart failure. Am J Cardiol 1999;84:608 – 11 A9.

[24] Missov E, Mair J. A novel biochemical approach to congestive heart failure: cardiac troponin T. Am Heart J 1999;138:95 – 9. [25] Ishii J, Nomura M, Nakamura Y, et al. Risk stratification using a combination of cardiac troponin T and brain natriuretic peptide in patients hospitalized for worsening chronic heart failure. Am J Cardiol 2002;89:691 – 5. [26] McDonagh TA, Robb SD, Murdoch DR, et al. Biochemical detection of left-ventricular systolic dysfunction. Lancet 1998;351:9 – 13. [27] Suzuki T, Yamaoki K, Nakajima O, et al. Screening for cardiac dysfunction in asymptomatic patients by measuring B-type natriuretic peptide levels. Jpn Heart J 2000;41:205 – 14. [28] Meinardi MT, van Veldhuisen DJ, Gietema JA, et al. Prospective evaluation of early cardiac damage induced by epirubicin-containing adjuvant chemotherapy and locoregional radiotherapy in breast cancer patients. J Clin Oncol 2001;19:2746 – 53. [29] Setsuta K, Seino Y, Ogawa T, Arao M, Miyatake Y, Takano T. Use of cytosolic and myofibril markers in the detection of ongoing myocardial damage in patients with chronic heart failure. Am J Med 2002;113:717 – 22. [30] Yoshida Y, Morimoto S, Hiramitsu S, Tsuboi N, Hirayama H, Itoh T. Incidence of cardiac sarcoidosis in Japanese patients with high-degree atrioventricular block. Am Heart J 1997;134:382 – 6. [31] Cowie MR, Struthers AD, Wood DA, et al. Value of natriuretic peptides in assessment of patients with possible new heart failure in primary care. Lancet 1997;350:1349 – 53. [32] Talwar S, Downie PF, Ng LL, Squire IB. Towards a blood test for heart failure: the potential use of circulating natriuretic peptides. Br J Clin Pharmacol 2000;50:15 – 20. [33] Cowie MR, Mendez GF. BNP and congestive heart failure. Prog Cardiovasc Dis 2002;44:293 – 321. [34] Maisel AS, Krishnaswamy P, Nowak RM, et al. Rapid measurement of B-type natriuretic peptide in the emergency diagnosis of heart failure. N Engl J Med 2002;347:161 – 7. [35] Yazaki Y, Isobe M, Hiroe M, et al. Prognostic determinants of longterm survival in Japanese patients with cardiac sarcoidosis treated with prednisone. Am J Cardiol 2001;88:1006 – 10. [36] Finch-Johnston AE, Gussak HM, Mobley J, et al. Cyclic variation of integrated backscatter: dependence of time delay on the echocardiographic view used and the myocardial segment analyzed. J Am Soc Echocardiogr 2000;13:9 – 17. [37] Wickline SA, Thomas III LJ, Miller JG, Sobel BE, Perez JE. The dependence of myocardial ultrasonic integrated backscatter on contractile performance. Circulation 1985;72:183 – 92. [38] Wickline SA, Thomas III LJ, Miller JG, Sobel BE, Perez JE. A relationship between ultrasonic integrated backscatter and myocardial contractile function. J Clin Invest 1985;76:2151 – 60.