Segmental Early Relaxation Phenomenon

Segmental Early Relaxation Phenomenon* Incidence, Clinical Characteristics, and Significance in Stress Echocardiography Omar Obeidat, MD; Muhammed Arida, MD; Mouaz Al-Mallah, MD; Mohsin Alam, MD; and Karthik Ananthasubramaniam, MD

Study objectives: To evaluate the incidence, patient characteristics, and clinical significance of segmental early relaxation phenomenon (SERP) in stress echocardiography. Design: Retrospective interpretation of digitized rest/stress echocardiographic images of 244 consecutive patients undergoing exercise or dobutamine echocardiography with subsequent patient follow-up for outcomes. Setting: Tertiary care referral center. Patients: Two hundred forty-four consecutive patients referred for stress echocardiography (exercise or dobutamine) for various clinical indications. Interventions: None. Measurements and results: SERP was diagnosed as a sudden outward motion of a portion of the left ventricle during early diastole (using frame-by-frame analysis) after peak systole prior to mitral valve opening at rest and after stress. Stress-associated SERP was observed in 71 patients (29.1%), with 25 patients having SERP in more than one segment. Five patients had resting SERP, with two persisting during stress. Ninety-six of 3,658 analyzed segments were positive for SERP. The apical septum and midseptum were most commonly involved in 49% and 18%, respectively. Only 5 of 96 patients (5.2%) had new hypokinesis and SERP in the same segment. No significant differences existed in demographic, clinical, or echocardiographic variables in patients with and without SERP. Follow-up revealed no significant differences in event rates in those with and without SERP. Conclusions: This is the first stress echocardiographic study demonstrating that SERP is a distinct and relatively common stress echocardiographic phenomenon occurring in early diastole regardless of type of stress. SERP occurs predominantly in apical and midseptum in the distribution of the left anterior descending coronary artery. It should not be mistaken for atypical septal motion, ischemia, or dyskinesia, and does not seem related solely to the presence of underlying coronary disease or stress-induced ischemia. No adverse long-term outcomes are seen in patients with SERP and no inducible ischemia. (CHEST 2004; 125:1218 –1223) Key words: asynchronous relaxation; diastole; ischemia; segmental early relaxation; stress echocardiography Abbreviations: ASE ⫽ American Society of Echocardiography; CAD ⫽ coronary artery disease; IVRP ⫽ isovolumic relaxation phase; LAD ⫽ left anterior descending coronary artery; LV ⫽ left ventricular; SERP ⫽ segmental early relaxation phenomenon

abnormalities in ventricular contraction S egmental are known to be associated with ischemic heart disease and have been well studied with regards to prognostic implications.1,2 However, regional asynergy of ventricular relaxation, although observed during angiographic studies nearly 30 years ago,3,4

still has unclear implications. These diastolic phenomena represent a sudden localized outward motion of a portion of the left ventricular (LV) wall that has been observed to occur during the isovolumic relaxation phase (IVRP) of the cardiac cycle. Various descriptive terms, such as segmental early relaxation

*From Henry Ford Heart and Vascular Institute, Detroit, MI. This work was presented in part at the American College of Cardiology Scientific Sessions 2002, Atlanta, GA. Manuscript received May 20, 2003; revision accepted October 2, 2003.

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: [email protected]). Correspondence to: Karthik Ananthasubramaniam, MD, Henry Ford Hospital, Heart and Vascular Institute, 2799 West Grand Blvd, K-14, Detroit MI 48202; e-mail: [email protected]

1218

Clinical Investigations

phenomenon (SERP),4 diastolic asynergy,5 late systolic bulge,6 preinflow relaxation,7 and asynchronous relaxation8 have been used to describe these phenomena. Stress echocardiography is an increasingly used, widely available noninvasive modality for the diagnosis and prognosis of coronary artery disease (CAD).9 Atypical septal motion due to various cardiovascular conditions and left bundle-branch block are situations that can make interpretation of wall motion abnormalities difficult in daily practice. SERP is an abnormal localized bulge of a portion of the ventricle predominantly reported in the resting state.4 – 8 To our knowledge, the prevalence and clinical significance of SERP in stress echocardiography has not been well studied. The objectives of this study were to retrospectively study the prevalence, patient demographics, and clinical and echocardiographic features of SERP in patients undergoing stress echocardiography. We further wanted to identify if this phenomenon had any bearing on the presence of ischemic response on stress echocardiography. Unless specified, we will refer to all similar diastolic phenomena reported in other studies as SERP. Materials and Methods We retrospectively evaluated 244 consecutive patients who underwent stress echocardiography (exercise and dobutamine stress) at our echocardiographic laboratory between September 1999 and June 2000. Chest pain (48%), screening for CAD (14%), and miscellaneous indications (38%) were the main reasons for referral of study patients for stress echocardiography. Exclusion criteria included left bundle-branch block, poor baseline echocardiographic images (lack of visualization of three or more segments), and patients in whom IV contrast enhancement for endocardial border definition was needed. This was due to the discretionary use of contrast at the time of stress by the sonographer; hence, some study patients did not have contrast enhancement at rest, making comparisons of rest and stress images suboptimal. Baseline demographics and clinical features of all the study patients were obtained by review of their medical records. Two American Society of Echocardiography (ASE) level III-equivalent echocardiographers who were blinded to patient information and study results interpreted all digitized rest and stress images recorded on optical discs. Frame-by-frame analysis of all baseline and post-stress cine-loops were performed with side-by-side comparison using standard four-quadrant format. Each segment was graded for wall motion using the ASE 16-segment model (Fig 1)10 as follows: 1 ⫽ normal, 2 ⫽ hypokinetic, 3 ⫽ akinetic, 4 ⫽ dyskinetic, and 5 ⫽ aneurysmal. SERP was denoted as being present or absent in each segment if a portion or segment of the LV wall demonstrated a sudden outward bulge after peak systole prior to mitral valve opening, which would represent IVRP. Peak systole was defined as the frame during systole that demonstrated the maximum inward excursion with the smallest LV cavity. Since the mitral valve was not visualized adequately in parasternal short-axis views, SERP was predominantly evaluated in the parasternal long-axis, apical four-chamber and two-chamber views, although the short-axis www.chestjournal.org

Figure 1. The standard 16-segment model of the left ventricle as recommended by the ASE. Ant ⫽ anterior; AL ⫽ anterolateral; AS ⫽ anteroseptum; Inf ⫽ inferior; IL ⫽ inferolateral; IS ⫽ inferoseptum; Lat ⫽ lateral; Sep ⫽ septum.

was also assessed. In cases where there was disagreement of ⬎ 1 score in segment evaluation, a joint review and consensus between the two readers was obtained. Follow-up was done by the investigators from the electronic patient database available at our health system. Medical records were reviewed to look for emergency department visits, office visits, or hospitalization for cardiovascular events including angina, unstable angina, myocardial infarction, revascularization, stroke, congestive heart failure, and death. If no follow-up was documented in our electronic database, mortality through Michigan state vital records and health statistics and the United States vital records and health statistics were cross-referenced. The study protocol was approved by our Institutional Review Board. Statistical Analysis Baseline demographics and clinical characteristics were reported for patients with SERP and without SERP, including means ⫾ SD and proportions as appropriate. Univariate comparisons were carried out using two-sample t test and ␹2 test. A two-sided ␣ level of 0.05 has been used to determine clinical significance. Bivariate statistics were performed with ␹2 test, and odds ratios were reported with 95% confidence intervals and p values. Multivariate statistics were performed using primarily CHEST / 125 / 4 / APRIL, 2004

1219

conditional multiple logistic regression for the independent relationship of different factors of interest with SERP.

Table 2—Stress Echocardiographic Variables* Variables

Results Baseline characteristics of patients with and without SERP on stress imaging are shown in Table 1. No significant differences in baseline characteristics were seen except for a trend toward lesser incidence of hypertension in patients with stress-induced SERP (p ⫽ 0.058). Also, there were a trend toward decreased incidence of previous coronary artery disease and hyperlipidemia in patients with stressinduced SERP. Table 2 outlines the comparison of stress echocardiographic variables in patients with and without SERP. No significant differences were noted in both patient groups. A total of 3,658 segments were systematically analyzed in 244 patients. Figure 2 shows the distribution of SERP segments (percentage of total segments). Stress-associated SERP was observed in 96 segments in 71 patients (29.1%). Multiple segments were involved with SERP in 25 of 71 patients (35%). Only five patients had resting SERP (2%). The resting ECG in these five patients only revealed normal sinus rhythm with nonspecific ST-T wave abnormalities, except one patient who had also nonspecific intraventricular conduction delay. None of the five patients had a positive ECG finding suggestive of ischemia with stress. SERP persisted during stress only in two of five patients. The distal (apical) septum was the most commonly involved (49%) segment with the midseptum in 20% (Fig 2). Six patients had estimated ejection fraction ⬍ 40%, and none of them had SERP. Only 5 of 96 segments (5.2%) with SERP had stress-induced ischemia in the same segment. On the contrary, 15 of 71 patients (21%) had inducible ischemia in segments other than SERP segments and 55 of 173 patients (32%) had inducible ischemia without any SERP.

Table 1—Patient Demographics* Variables Age, yr Male gender Hypertension Diabetes Hyperlipidemia History of CAD Smoking History of CHF Mitral regurgitation Ejection fraction

SERP (n ⫽ 71) No SERP (n ⫽ 173) p Value 54 ⫾ 13 56 45 17 38 13 27 2.8 7.0 52

58 ⫾ 14 50 58 17 51 20 28 3.4 5.7 52

0.4 0.58 0.058 NS 0.068 0.20 NS NS NS NS

*Data are presented as mean ⫾ SD or %. NS ⫽ not statistically significant; CHF ⫽ congestive heart failure. 1220

SERP (n ⫽ 71)

No SERP (n ⫽ 173)

Exercise stress test 61 (86) 133 (77) Dobutamine stress test 10 (14) 40 (23) Mean ejection fraction 52 52 Exercise time, min 8.5 ⫾ 3 7.5 ⫾ 3 Double product 26,684 ⫾ 6,649 23,258 ⫾ 5,690 %PMHR 90 ⫾ 11 87 ⫾ 9 Positive EKG 18 12 Positive echo 28 32

p Value NS NS NS 0.06 0.12 0.10 0.16 0.68

*Data are presented as No. (%), mean ⫾ SD, or %. PMHR ⫽ predicted maximum heart rate; positive EKG ⫽ ⱖ 1 mm of new horizontal or downsloping ST depression at 80 ms beyond the J point in three consecutive QRS complexes; Positive echo ⫽ new or worsening regional wall motion abnormality in any ventricular segment except resting akinetic wall segments. See Table 1 for expansion of other abbreviation.

Patient follow-up was completed for 89% of the study subjects through our health system database, and no mortality data were available for the other 11% from the State of Michigan and US vital records and health statistics. The mean follow-up time was 29 ⫾ 9 months (mean ⫾ SD). Only one patient died because of sepsis from the group without SERP. There was no statistical difference of event occurrence between both groups (Table 3). Angiographic data were available in only 25 of 244 patients (10%). Only 6 of 25 patients had SERP, and all SERP was observed in the distal septal segment. The left anterior descending coronary artery (LAD) had significant atherosclerosis in four of six patients. The other two patients had nonobstructive epicardial CAD. Discussion The results of our study suggest that SERP is a commonly observed early diastolic stress echocardiographic phenomenon with a predominant occurrence in the distal and midseptal distribution supplied by the LAD. There seems to be no unique patient profile demonstrating this phenomenon during stress, and in particular there seems to be no definite relation between detection of SERP and stress-induced ischemia in the underlying segment or coronary distribution. Furthermore, the presence of SERP does not by itself seem to predict increased incidence of adverse outcomes (Table 3). Although 75% of patients with SERP in the distal septum had angiographic evidence of LAD disease (four of six patients) given the small patient numbers with angiographic correlation, firm conclusions with regards to occurrence of SERP during stress echocardiograClinical Investigations

Figure 2. The distribution of SERP segments. Distal and midseptum were most commonly involved. The apex and distal septum were analyzed as one segment (distal septum) when interpreting for presence or absence of SERP. Also, note the very low prevalence of SERP in basal LV segments. % ⫽ percentage of total SERP segments. See Figure 1 for expansion of other abbreviations.

phy as a sign of underlying CAD (in the absence of an ischemic response) cannot be made from current study although the possibility does exist (see discussion below). Gooch et al3 initially described SERP in patients with mitral valve prolapse and attributed it to functional cardiomyopathy. They noticed sudden outward movement of the anterior wall during IVRP and believed that this was due to mitral valve prolapse. Altieri et al4 studied 100 consecutive patients with chest pain undergoing cardiac catheterization. SERP was found in 83 patients, with 29 of 83 patients showing concomitant inward movements in other segments. Only 30 of 59 patients with obstructive coronary artery disease (CAD) had SERP, with SERP occurring predominantly in normally contracting segments of the left ventricle. They concluded that SERP was a normal variation of the LV relaxation pattern. Table 3—Clinical Outcomes After 29 ⴞ 9 Months of Follow-up* Variables Atypical chest pain Typical chest pain Myocardial infarction Congestive heart failure Cerebrovascular accident Death

SERP (n ⫽ 71) 6 1 0 1 0 0

*See Table 1 for expansion of abbreviation. www.chestjournal.org

No SERP (n ⫽ 173)

p Value

22 6 1 1 1 1

NS 0.38 NS NS NS NS

SERP and CAD Prior studies with angiograpic6,11 and LV wall motion12 correlations during experimental ischemia have suggested that SERP-like phenomena was seen frequently in the distribution of arteries subtending significant stenosis. Hamby and colleagues6 suggested that SERP was probably related to LAD disease, as 15 of 21 patients in their series had significant LAD disease in segments demonstrating SERP. Abe and Tomotsune8 studied various subgroups of patients with mitral valve disease and CAD with and without history of myocardial infarction. They found a high incidence of SERP in the subgroup of patients with angina pectoris and myocardial infarction, and concluded that localized myocardial ischemia definitely plays an etiologic role in SERP, which could be an early manifestation of ischemia. Our study confirms and extends previous observations regarding SERP. The distribution of SERP was noted predominantly in normal segments and was absent in severely hypokinetic or akinetic segments. Furthermore, SERP was observed in patients with and without underlying CAD; although statistically insignificant, there was actually a trend toward a less prevalence of previous history of CAD in patients with SERP (Table 1). Unlike the study by Gibson et al,12 we did not observe a significant inward motion of the left ventricle in ischemic segments during stress echocardiography. New contributions from our study include the CHEST / 125 / 4 / APRIL, 2004

1221

Figure 3. Left, A: peak systole. Right, B: SERP of distal septum after peak systole during IVRP prior to mitral valve opening.

incidence, segmental distribution, and significance of SERP in the setting of stress echocardiography, which to our knowledge has not been reported so far. Since only 5% of patients had stress-induced SERP and ischemia in the same segment, SERP is probably not unique to CAD. Our study also indicates that SERP by itself in the absence of ischemia does not predict high cardiac event rates. Interestingly, despite the fact that firm conclusions cannot be made from our angiographic data due to small numbers, SERP as a phenomenon linked to CAD cannot be excluded, as four of six patients with stress-induced SERP who had angiographic data had significant underlying LAD. Diastolic Function in Patients With SERP Previous angiographic studies have suggested that alteration of viscoelastic properties of the left ventricle due to ischemia may be a cause of SERP. Alam et al5 noted prolongation of isovolumic relaxation phase in patients with SERP. Other studies13,14 have described significant correlation between SERP (identified angiographically) and disturbed echo-Doppler ventricular filling dynamics. Proposed Etiologies for SERP The exact etiology precipitating SERP has still not been clarified, but CAD with underlying ischemia,8,11 loss of erectile effect associated with early coronary blood flow,15 noncoronary causes such as valvular disease, atrial septal defect and nonischemic 1222

cardiomyopathy,11 ventricular pacing,16 and regional fibrosis or regional autonomic denervation have all been implicated. Limitations Given that our study was retrospective, it has its inherent limitations. Correlation of occurrence of SERP to diastolic LV indexes changes was not possible, as a complete diastolic study was not done in many patients as a part of the initial resting study. Furthermore angiographic correlation was not possible for the obvious reasons in those patients with normal stress echo findings and SERP did not undergo angiography, and not all patients with abnormal stress echocardiographic findings and SERP underwent cardiac catheterization. Nevertheless, to our knowledge ours is the first stress echocardiographic study to address SERP with patient followup, thus contributing more to the existing literature that is largely angiographic, with most patients being assessed in resting state. The very low incidence of SERP in resting images in our study (2%) contrary to all previous angiographic studies is unclear, but could be related to different patient demographics, neurohormonal and hemodynamic alterations at time of identification of SERP (angiography vs echocardiography), unknown effect of cardiac medications in modifying this phenomenon, and different methodologies and definitions used in identifying this phenomenon in different studies. Clinical Investigations

Clinical Implications Identifying SERP might be important for correct interpretation of exercise and pharmacologic stress echocardiograms (Fig 3). SERP might be confused with segmental dyskinesis, which is a paradoxical systolic outward motion of the LV wall. This is more likely if quick outward movements of apex and distal septum are not re-evaluated by frame-by-frame analysis. Frame-by-frame assessment with QRS reference or mitral valve visualization will clarify that dyskinesis occurs earlier than SERP during systole, and either involves segments with previously severely impaired contraction or severely ischemic segments. In contrast, stress-induced SERP almost exclusively occurs in normal or hypokinetic segments as shown in our study (95% of patients with SERP had normal underlying wall motion with stress). Furthermore, dyskinesia occurs during inward excursion of the rest of the ventricular wall, whereas SERP is seen after maximum systolic excursion is completed and prior to mitral valve opening, indicating occurrence during IVRP. SERP may also be confused with atypical or paradoxical septal motion. These are usually seen in conditions associated with conduction abnormalities such as left bundle-branch block, right ventricular volume overload, or after cardiothoracic surgery. Again, frame-by-frame analysis identifies normal septal thickening and motion pattern in systole with localized outward bulging of septum during IVRP in segments with SERP in contrast to systolic motion abnormalities in the above conditions. Based on existing literature,5,13,14 identification of SERP should prompt more thorough analysis of diastolic function particularly in patients with normal systolic function. Conclusion SERP is a distinct diastolic phenomenon frequently seen in the early phase of relaxation by stress echocardiography regardless of the type of the stress. It is predominantly seen in middistal septal distribution of the LAD, and is not seen in severely hypokinetic or akinetic segments. It should not be mistaken for atypical septal motion or ischemic response to stress, does not seem to be solely related to previous history of coronary disease or active ischemia, and is not associated with adverse patient outcomes. A prospective angiographic evaluation of patients dem-

www.chestjournal.org

onstrating this phenomenon with stress echocardiography and correlation with diastolic parameters and/or flow reserve indexes is needed to firmly address its relation to CAD and diastolic function. References 1 Hermann MV, Heinle RA, Klein MD, et al. Localized disorders in myocardial contraction: asynergy and its role in congestive heart failure. N Engl J Med 1967; 277:222–232 2 Baxley WA, Reeves JT. Abnormal regional myocardial performance in coronary artery disease. Prog Cardiovasc Dis 1971; 13:405– 421 3 Gooch AS, Vicencio F, Maranhao V, et al. Arrhythmias and left ventricular asynergy in the prolapsing mitral leaflet syndrome. Am J Cardiol 1972; 29:611– 620 4 Altieri PI, Wilt SM, Leighton RF. Left ventricular wall motion during the isovolumic relaxation period. Circulation 1973; 48:499 –505 5 Alam S, Tansey W, Cameron A, et al. Asynchronous ventricular relaxation: an angiographic temporal analysis of asynchronous left ventricular relaxation in man. Am J Cardiol 1979; 43:41– 46 6 Hamby RI, Aintablian A, Tabarah F, et al. Late systolic bulging of the left ventricle in patients with angina pectoris: a form of asynchronous contraction. Chest 1974; 65:169 –175 7 Ruttley MS, Adams DF, Cohn PF, et al. Shape and volume changes during isovolumetric relaxation in normal and asynergic ventricles. Circulation 1974; 50:306 –316 8 Abe H, Tomotsune K. Asynchronous relaxation of the ischemic left ventricle. Jpn Circ J 1982; 46:103–112 9 Elhendy A, Mahoney DW, Khandheria BK, et al. Significance of the location of wall motion abnormalities during exercise echocardiography. J Am Coll Cardiol 2002; 40:1623–1629 10 Schiller NB, Shah PM, Crawford M, et al. Recommendations for quantitation of the left ventricle by two-dimensional echocardiography: American Society of Echocardiography Committee on Standards, Subcommittee on Quantitation of Two-Dimensional Echocardiograms. J Am Soc Echocardiogr 1989; 2:358 –366 11 Wilson C, Krueger S, Forker AD, et al. Correlation between segmental early relaxation of the left ventricular wall and coronary occlusive disease. Am Heart J 1975; 89:474 – 479 12 Gibson D, Prewitt T, Browa D. Analysis of left ventricular wall movement during isovolumic relaxation and its relation to coronary artery disease. Br Heart J 1976; 38:1010 –1019 13 Vanoverschelde JL, Wijns W, Michel X, et al. Asynchronous (segmental) early relaxation impairs left ventricular filling in patients with coronary artery disease and normal systolic function. J Am Coll Cardiol 1991; 18:1251–1258 14 Ludbrook PA, Byrne JB, Tiefenbrunn AJ. Association of asynchronous protodiatolic segmental wall motion with impaired left ventricular relaxation. Circulation 1981; 64:1201–1211 15 Gaasch WH, Blaustein AS, Bing OHL. Asynchronous (segmental early) relaxation of the left ventricle. J Am Coll Cardiol 1985; 5:891– 897 16 Hood WB Jr, Joison J, Abelmann WH, et al. Asynchronous contraction during late systolic bulging at left ventricular pacing sites. Am J Physiol 1969; 217:215–221

CHEST / 125 / 4 / APRIL, 2004

1223