Study Population

This retrospective cohort study was approved by the Cleveland Clinic Institutional Review Board. We used our CP database to identify patients with CP who underwent echocardiography performed between January 2007 and December 2010 using Vivid 7 or E9 (GE Medical Systems, Milwaukee, WI) ultrasound machines at our institution. Out of 98 patients identified, one was excluded because of significant valvular disease (severe aortic stenosis), four were excluded because of significant coronary artery disease, and one was excluded because of prior pericardiectomy. Out of the remaining 92 patients, nine were excluded because of poor image quality. Among the 83 patients remaining, the diagnosis of CP was established by a combination of imaging and invasive techniques, including echocardiography in 83 patients, tomographic imaging (computed tomography or magnetic resonance imaging [MRI]) in 77 of 83 patients, and dedicated invasive hemodynamic study that included simultaneous measurements of right ventricular and LV pressures in 69 of 83 patients ( Appendix ).

Out of the 83 patients, 50 were surgically treated, while 33 patients were treated medically ( Supplemental Figure 1 ). In all surgically treated patients, pericardiectomy was performed through a sternotomy or left thoracotomy incision. The standard pericardial resection at our institution is an on-pump comprehensive pericardiectomy with removal of the diaphragmatic component, anterior pericardium from phrenic nerve to phrenic nerve, and posterior pericardium to the left phrenic nerve. Radical pericardiectomy was performed in most patients, but when it was not achievable, as much pericardium was resected as possible. The visceral pericardium was also removed as required. Twenty-seven out of 50 surgically treated patients underwent postsurgical studies with a GE machine, and this was used for comparison before and after surgery. The median period between the presurgical echocardiographic examination and surgery was 8 days (interquartile range, 5–37 days), and the median period from surgery to follow-up echocardiography was 29 days (interquartile range, 7–54 days). Out of 33 medically treated patients, 16 were considered to have mild constriction, 11 had ongoing inflammation by MRI, and six were deemed poor surgical candidates because of concomitant systemic disease or cancer.

We randomly selected 20 control subjects >55 years of age from our database of 117 healthy volunteers free of any known cardiovascular disease, with normal results on physical examination and normal electrocardiographic results and not taking any cardioactive medications (except aspirin). The Institutional Review Board of Cleveland Clinic approved the database search. All of these control subjects gave written informed consent for participation in respective prospective studies.

Standard Echocardiography

Echocardiography was performed using standard commercial echocardiographic systems. Sector size and depth were adjusted to achieve optimal visualization of all LV myocardium at the highest possible frame rate. Acquisition was obtained at end-expiration. Multiple consecutive cardiac cycles of the standard views were acquired and stored digitally in a Digital Imaging and Communications in Medicine–based echocardiographic viewer (syngo Dynamics; Siemens Healthcare, Malvern, PA) for subsequent analysis. Ejection fraction was calculated using the biplane method of disks. Tissue Doppler–derived indices were measured using the apical four-chamber view. Peak systolic (s′) and early diastolic (e′) mitral annular velocities were calculated by averaging septal and lateral mitral annular velocities. All pulsed-wave and tissue Doppler measurements were averaged over three cardiac cycles.

Measurement of Deformation Parameters

Images were saved in digital format and analyzed offline using commercially available software (EchoPAC PC version 11.0.0; GE Medical Systems) by a single operator who was blinded to clinical information. The apical four-chamber view was used to obtain longitudinal displacement (LD) during one cardiac cycle at 56 ± 17 frames/sec. Special care was taken to avoid the pericardium when the width for the region of interest was determined. For LD, the basal segments were used for the analysis. After decortication, LD became different between the septum and lateral wall, so we compared the change in contraction of each segment using strain. Peak longitudinal strain (LS) was measured in the three apical views. The apical, middle, and basal segments were averaged separately in the septal and lateral walls for LS. Global LS (GLS) was calculated averaging the 18 segmental strain values from three apical views. Global circumferential strain (GCS) and global radial strain (GRS) were calculated by averaging the six segments from short-axis images at the level of the mid papillary muscle level.

SLRD was analyzed in the apical four-chamber view as previously described. In brief, the region of interest was applied over the LV myocardium in an apical four-chamber view. The software automatically tracked down the rotational rate of myocardial motion with reference to the center of gravity of the region of interest. In accordance with engineering notation, a negative sign indicated clockwise rotation, and a positive sign signified counterclockwise rotation. The adequacy of tracking was verified manually, and the region of interest was readjusted to achieve optimal tracking. Figure 1 shows an example of LD, LS, and SLRD of a patient with CP.

A case of CP with LD and longitudinal rotation before and after pericardiectomy. (A) LD. Before pericardial surgery, the basal septum ( yellow curve; top left ) displaced greater than the basal lateral segment ( red curve ). This is also clearly shown in color-coded curved anatomic M-mode images of LD ( arrows; bottom left, purple area in the basal septum in the segment before surgery; bottom right, purple area in the basal lateral in segment after surgery). (B) LS. Before surgery, strain in the lateral segments was reduced compared with strain in the septum. The difference lessened after the surgery. (C) SLRD. The discrepancy between similar strain values of opposing walls and the greater displacement of the lateral wall after surgery is demonstrated by the counterclockwise longitudinal rotation of the whole heart, which is induced by an exaggerated heart swinging motion because of the loss of pericardial support. Arrows indicate longitudinal rotations curves.

Outcomes

Our prespecified outcomes were change in dose and/or number of diuretics used, and NYHA class. Specifically, a reduction in diuretics was defined as (1) a reduction in the dose of a diuretics (eg, furosemide 40 to 20 mg), (2) discontinuation of a diuretic (eg, furosemide 20 mg plus spironolactone 25 mg to furosemide 20 mg only), or both. Median reduction of furosemide was 40 mg (range, 20–160 mg), and reduction of spironolactone was 25 or 50 mg.

Interobserver and Intraobserver Variability

To assess interobserver and intraobserver variability of LD, LS, and SLRD, we assessed data from 10 randomly selected subjects. Absolute (and relative) values of intraobserver and interobserver variability for SLRD were 0.8 ± 0.5° (18 ± 13%) and 0.8 ± 0.7° (17 ± 14%). Absolute and relative values of intraobserver variability for LD were 1.2 ± 1.3 mm and 9 ± 11%, whereas they were 2.1 ± 2.1% and 9 ± 8% for LS. Absolute and relative values of interobserver variability for LD were 1.5 ± 1.3 mm and 10 ± 11%, whereas they were 2.4 ± 2.1% and 11 ± 10% for LS.

Statistical Analysis

Continuous data are presented as mean ± SD, unless otherwise noted. Categorical data are expressed as percentages. Student’s t tests and Mann-Whitney U tests were used to compare continuous variables when appropriate. For the comparison of paired samples of deformation indices, Wilcoxon signed-rank tests were used. Chi-square and Fisher exact tests were applied to compare categorical variables. Receiver operating characteristic curve analyses were performed to determine the utility of SLRD change to distinguish patients with reduced diuretic use after surgery from those without reduced diurect use and to determine optimal cutoff values. The best cutoff value was defined as the point with the highest sum of sensitivity and specificity. Statistical analyses were performed with SPSS version 20.0 (SPSS, Inc, Chicago, IL), and P values < .05 were considered to indicate statistical significance.

Patient Characteristics

All patients with CP met at least one echocardiographic criterion for CP ( Appendix ). Also, out of 77 patients with tomographic imaging (computed tomography or MRI), 76 patients met at least one criterion for CP, while 64 of 69 patients who underwent dedicated cardiac catheterization met at least one criterion for CP. In all 50 surgical patients, the diagnosis was surgically confirmed. Clinical characteristics of CP patients and control subjects are summarized in Table 1 . There were no differences in age and gender between the control and CP groups. The most common etiology of CP was idiopathic ( n = 50) followed by postsurgical ( n = 20). LV diastolic function was characterized by higher E velocity, E/A ratio, and E/e′ ratio and shorter deceleration time in the CP group. Surgically treated patients were in higher NYHA classes and had shorter deceleration times than medically treated patients. There were 14 patients (17%) with atrial fibrillation (four [12%] medically treated and 10 [20%] surgically treated). Although there was no difference in gender ( P = .46), the patients with atrial fibrillation were older than the patients in sinus rhythm (64 ± 12 vs 57 ± 12 years, P = .04). There were four patients with myocardial involvement (two confirmed by MRI and two by computed tomography).

Deformation Parameters

LV Displacement

Patients with CP had lower LD than control subjects ( P < .001). LD values were similar between the opposing walls in both patients with CP and control subjects ( Figure 2 A). After the pericardiectomy, there was heterogeneity of change in LD in individual LV walls. Lateral LD became larger ( P = .008) but septal LD smaller ( P < .001) ( Figure 2 B). There was no significant correlation between end-diastolic volume and septal LD ( r = −0.06, P = .58), but there was a weak trend between end-diastolic volume and lateral LD ( r = 0.18, P = .08) in the entire population.

LD change and LS by walls. LD values of opposing walls were similar but both decreased in patients with CP (A) . After the surgery, septal LD decreased but lateral LD increased (B) . Peak LS during one cardiac cycle was decreased in the lateral wall of patients with CP (C) . However, the difference between the LS of the opposing walls lessened (D) . (A,B) LD; (C,D) LS.

LV Strain

Patients with CP had lower absolute values of GLS and GCS ( P < .01 for both), with no significant difference in GRS ( P = .07) compared with control subjects ( Table 2 ). Similar results were seen after adjusting these for end-diastolic volume index, except for GRS, for which it became significant after the adjustment ( P = .031). Although septal LS was similar between control subjects and patients with CP ( P = NS), lateral LS was lower in the CP group than the control group ( P < .001) ( Figure 2 C). There were no changes in GLS and GRS after pericardiectomy (GLS, P = .88; GRS, P = .73), whereas GCS increased ( P = .008). Septal LS decreased significantly (from −20.3 ± 5.0% to −17.7 ± 4.6 %, P = .032) after surgery, but lateral LS did not (from −14.7 ± 5.8% to −15.2 ± 3.4%, P = .51) ( Figure 2 D). There was less difference in LS between the opposing walls after surgery (before, 5.6 ± 5.3%; after, 2.5 ± 4.2%; P = .008).

Table 2

GLS, strain rate, and longitudinal rotation

Variable	Control ( n = 20)	CP ( n = 83)	P ∗	Before ( n = 27)	After ( n = 27)	P †
GLS	−20.1 ± 1.9	−16.2 ± 3.3	<.001	−15.6 ± 3.9	−15.8 ± 3.2	.99
GCS	−20.7 ± 5.1	−14.7 ± 5.0	<.001	−13.5 ± 5.7	−17.6 ± 5.5	.032
GRS	50.4 ± 16.2	40.8 ± 18.8	.07	37.4 ± 18.9	39.1 ± 16.5	.99
SLRD	−2.3 ± 3.3	−0.6 ± 3.0	.07	−0.8 ± 3.3	2.1 ± 3.0	.008

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Usefulness of Two-Dimensional Strain Parameters to Diagnose Acute Rejection after Heart Transplantation
2. Discrepancies in Left Ventricular Mass Calculation Based on Echocardiography and Cardiovascular Magnetic Resonance Measurements in Patients with Left Ventricular Hypertrophy
3. Longitudinal Strain by Automated Function Imaging Detects Single-Vessel Coronary Artery Disease in Patients Undergoing Dipyridamole Stress Echocardiography
4. Three-Dimensional Echocardiography: A Powerful New Tool in the Evaluation of Mitral Annular Structure and Dynamics

Stay updated, free articles. Join our Telegram channel

Join