Decreased Right Ventricular Function after Coronary Artery Bypass Grafting

Abstract

Decreased right ventricular function after coronary artery bypass grafting is a common and well-known (if not well-understood) phenomenon.

We prospectively evaluated right ventricular function via echocardiographic tricuspid annular motion, tricuspid annular velocity, and right ventricular strain analysis before and after coronary artery bypass grafting. We also evaluated the effect of right coronary artery disease and revascularization on post-coronary artery bypass grafting, right ventricular function, and interventricular septal motion.

We performed baseline echocardiography in 250 candidates for coronary artery bypass grafting, and we repeated echocardiography in 240 of those patients 1 year after coronary artery bypass grafting. We evaluated right ventricular function via tricuspid annular motion, tricuspid annular velocity, and right ventricular strain analysis, all measured at the right ventricular free wall.

Right ventricular function as evaluated by tricuspid annular motion showed a significant reduction 1 year after coronary artery bypass grafting (21.7 vs 12.1 mm; P < 0.001) compared with preoperative measurements. Right ventricular tissue velocity (14.0 vs 7.0 cm/s; P < 0.001) and right ventricular strain (20.3% vs 11.6%; P < 0.001) were also significantly reduced after coronary artery bypass grafting. Interventricular septal motion was paradoxical in 97% of the patients 1 year after coronary bypass.

Right ventricular function remained depressed for as long as 1 year after coronary artery bypass grafting. These findings were independent of the state of the right coronary artery and the graft. It is likely that the interventricular septum is recruited to maintain right ventricular stroke volume after coronary artery bypass grafting.

A decrease in right ventricular (RV) function is an event known to occur after coronary artery bypass grafting (CABG). Right ventricular dysfunction can be seen during and immediately after cardiac surgery. Although the mechanism of this phenomenon is not well understood, cardiopulmonary bypass, perioperative myocardial ischemia, intraoperative myocardial damage, cardioplegia, and pericardial disruption or adhesion have been suggested as probable causes.^1,2

The quantitative assessment of RV function, historically difficult due to the complex geometry of the RV and to poor endocardial definition on chest radiography, has in more recent years been facilitated by the application of such newer imaging techniques as 2-dimensional (2-D) echocardiography, Doppler tissue imaging (DTI), magnetic resonance imaging, transesophageal echocardiography, and thermodilution pulmonary-artery catheterization.^3-5 The most common methods used in these studies are the measurement of tricuspid annular motion (TAM)—which itself provides an estimation of RV ejection fraction—and the measurement of tricuspid annular velocity (TAV); the measurement of RV strain has received less emphasis.^1,2

Doppler tissue imaging tends to be the most favored of all these techniques because of its easy application, cost-effectiveness, and suitability for the measurement of long-axis ventricular function. Indeed, DTI could become an important tool in routine echocardiography, because it can detect the impairment of longitudinal myocardial fiber motion, which is a sensitive marker of early myocardial dysfunction and ischemia. Of more interest in specific application to the evaluation of RV function is the fact that DTI enables the truly quantitative measurement of regional myocardial function.⁶

We used 2-D echocardiography and DTI in an effort to evaluate, respectively, the efficacy of TAM and TAV in detecting RV dysfunction 1 year after CABG. We also measured RV strain, in order to compare its efficacy in the evaluation of RV function with that of TAM and TAV. As an offshoot of our study, we inquired into the postulate that right coronary artery (RCA) revascularization could favorably affect post-CABG RV function. It was our goal to recruit a higher number of patients than had been recruited for previous studies in this field of enquiry.

Patients and Methods

We recruited 250 patients who had been referred to our hospital from May 2002 through May 2004 for elective CABG. Of this total, 4 patients died and 6 were lost to follow-up; the study population, therefore, comprised 190 men and 50 women. The mean age was 58.3 years (range, 30–81 yr); 36.7% of the patients were ≥65 years old. Sixty-eight (28.3%) patients had diabetes mellitus, and 139 (57.9%) were hypertensive. The patients' characteristics are summarized in Table I.

Table I. Patients' Characteristics

The subjects, who had a history of significant coronary artery disease, underwent CABG within 2 months after diagnostic coronary angiography. None of the patients had a history of recent myocardial infarction (4 weeks before preoperative angiography), atrial fibrillation, significant valvular heart disease, pulmonary hypertension, or previous CABG. Standard laboratory markers for myocardial infarction were obtained during and after CABG, and only 1 patient received a diagnosis of perioperative myocardial infarction (a small infarction). We recorded the patients' sex and age, comorbidities (diabetes mellitus and hypertension), and the total number of grafts for each patient, with special attention to whether or not the RCA was revascularized. The study was approved by the hospital's ethics committee, and written informed consent was obtained from all study participants. All patients were monitored for 1 year after CABG.

Surgical Technique

Standard hypothermic (approximately 32°C) cardiopulmonary bypass was used in all patients. Blood cardioplegic solution, delivered via both antegrade and retrograde routes, was used to ensure myocardial protection. All patients in the study underwent total revascularization. In the event that RCA stenosis reduced flow in both the posterior descending coronary artery and the posterolateral branch, 2 grafts were implanted.

Study Design

This was a prospective study in which all the patients had a baseline 2-D echocardiogram and DTI before CABG. Subsequently, study participants were monitored repeatedly throughout 1 year after CABG, in specific regard to their RV function. The measurements recorded in this study were 1) TAM, measured by M-mode echocardiography; 2) TAV, measured by DTI; and 3) RV strain, measured by DTI. All the measurements were recorded at the RV free wall. The paradoxical motion of the interventricular septum was analyzed either by visual detection or by using M-mode echocardiography from the parasternal and apical views.

One year after CABG, each of these tests was repeated, and each result was compared with the preoperative result of the same test. The expected decrement in RV function was measured by comparing the efficacy of each of the said parameters. The effects of sex, age, diabetes mellitus, hypertension, and number of grafts were also taken into account.

Because not all of our patients underwent RCA revascularization (which could have affected RV function), we divided our patients into 2 groups: those with RCA revascularization, and those without.

The RV function decrement (pre-CABG vs post-CABG) of patients in the 2 groups, measured via the above-mentioned methods, was compared to evaluate the effect of RCA revascularization on RV function.

Statistical Analysis

Statistical analysis was performed by use of the SPSS software package (SPSS Inc.; Chicago, Ill), version 13.0. Data are presented as mean ± SD, percentage, and total number when necessary. The paired t test helped in the comparison of pre- and post-CABG values for TAM, TAV, and RV strain. Linear regression analysis was performed not only to compare pre-and post-CABG values of TAM, TAV, and RV strain, but also to test the relationships between those same values and patients' ages in years. The independent-samples t test was used to find the association between qualitative data (such as sex, comorbidities, and history of RCA revascularization) and quantitative data (such as pre- and post-CABG values and the number of grafts). The χ² test was applied to compare categorical variables. A P value of less than 0.05 was considered statistically significant.

Results

Tricuspid annular motion and TAV revealed a significant decrease in RV function 1 year after CABG. The paired t test indicated a significant difference (P <0.001) between RV function as measured by TAM before CABG (21.7 ± 2.0 mm) and 1 year after CABG (12.1 ± 2.1 mm). Similarly, TAV showed a statistically significant decrement in RV function after CABG (14.0 ± 1.4 vs 7.0 ± 1.0 cm/s, P <0.001) (Table II). Tricuspid annular motion and TAV found no association between sex or age and RV dysfunction (P >0.05). Right ventricular dysfunction in diabetic and hypertensive patients did not differ from that in other patients (P >0.05).

Table II. Functional Data Before and One Year After CABG

Like TAM and TAV testing, RV strain testing was able to detect a decrement in RV function even 1 year after CABG (20.3 ± 2.4 vs 11.6 ± 1.7, P <0.001) (Table II). This finding was independent of sex (P >0.05) and age (P >0.05). The comorbidities (diabetes mellitus and hypertension) had no effect on RV dysfunction (P >0.05). Although the values of RV functional loss detected by TAM were higher than those detected by RV strain and TAV, no statistically significant difference was found between the 3 methods (P >0.05).

Revascularization of the RCA played no role in RV dysfunction after CABG, for the independent-samples t test showed the difference between the 2 groups of patients (with and without RCA revascularization) only in terms of post-CABG RV function as measured by TAV and RV strain (P >0.5)—not in terms of pre- and post-CABG (Table III).

Table III. Functional Data in Patients with (Group I) and without (Group II) RCA Revascularization

In 60% of our patients with RCA stenosis, the site of the stenosis was proximal; in the remaining 40%, it was only distal. When TAM was used to measure right ventricular function preoperatively (21.6 vs 21.8 mm) and again 1 year postoperatively (12.1 vs 12.1 mm), there was no difference between patients with proximal or distal RCA stenosis. Similarly, when TAV was used preoperatively (14.0 vs 13.9 mm) and 1 year postoperatively (6.9 vs 6.9 mm), there was no significant difference between patients with proximal or distal RCA stenosis. Nor was there any significant difference between these 2 groups when RV strain was used preoperatively (20.2% vs 20.4%) and in follow-up (11.1% vs 11.9%) (Table IV).

Table IV. Functional Data Before and One Year After CABG, in Specific Regard to Site of RCA Stenosis

The number of grafts did not affect RV dysfunction after CABG, as determined by our techniques of measurement (P >0.5) (Table III). Neither was there any significant difference between the 2 groups in pump time or cross-clamp time (P >0.5) (Table III).

One year after CABG, interventricular septal motion was paradoxical in 97% of the patients. In the remaining patients, interventricular septal motion was normal: in all of them, TAM was more than 18 mm; TAV was more than 12 cm/s; and RV strain was more than 16%.

New York Heart Association functional class was significantly improved, in comparison with functional class at baseline (3.3 vs 1.3, P <0.05).

Discussion

Using the 3 techniques of TAM, TAV, and RV strain, our study reconfirmed that post-CABG RV function is depressed, compared with pre-CABG RV function. The variations in RV functional loss were not statistically significant between the 3 techniques. On the other hand, diminished RV function proved to be independent of the number of grafts and RCA revascularization.

Decreased RV function after CABG is a frequently observed phenomenon,^7,8 the leading cause of which has yet to be found; nonetheless, intraoperative ischemia, intraoperative myocardial damage, cardioplegia, and pericardial disruption have been suggested as probable causes.^9-13

There are studies maintaining that exercise performance after CABG improves despite unchanged left ventricular function at rest, but the significance of the RV in regard to postoperative exercise testing remains to be explained.¹⁴ Postoperative exercise testing in the Hedman study⁸ was associated with a higher workload and rate-pressure product, despite depressed RV function after CABG. We found that our patients were asymptomatic, with good functional class despite echocardiographic evidence of reduced RV function.

Of all the imaging methods for the evaluation of RV function, recording the amplitude of TAM by means of conventional M-mode or 2-D echocardiography is the simplest, because it is free of such complications as trabeculae and myocardial dropouts and interobserver and intraobserver variations in recording and calculating.

Tricuspid annular motion has shown good correlation with RV ejection fraction calculated via radionuclide angiography.¹⁵ Moreover, previous studies10,14,16,17 have pointed to the usefulness of TAM in evaluating RV function in patients with acute and chronic ischemic heart disease, in those with congestive heart failure, and in patients after CABG. In our study, TAM decreased significantly 1 year after CABG. These preliminary data in our study are in agreement with those in recent studies that report the perioperative occurrence of RV dysfunction and lingering dysfunction 6 months after CABG.^14,18 Another study,⁸ with a smaller number of patients at 1-year follow-up after CABG, also showed the persistence of depressed RV function.

Doppler tissue imaging, increasingly favored for quantitative measurement of left ventricular wall motion, serves also as a measure of RV systolic function through the determination of TAV and, therefore, enables quantitative evaluation of RV function.^3,19 Peak systolic velocity less than 11.5 cm/s indicates the presence of RV dysfunction with a sensitivity and specificity of 90% and 85%, respectively.²⁰ According to this criterion, not only did TAV reveal a significant decrease in post-CABG RV function in our study, but 98.8% of our patients were in the range of RV dysfunction.

In 2003, Alam and colleagues¹ reported an initial significant decrease in systolic TAV after CABG, followed by partial recovery 1 year after CABG; this is in stark contrast to our findings of persistent reduction in RV function 1 year after CABG. However, further studies with long-term follow-up are necessary to determine whether post-CABG RV function will recover and, if so, when recovery will occur.

Despite encouraging clinical data derived from the analysis of myocardial velocity profiles as a method to detect abnormal regional function, the use of motion to represent function has 2 important drawbacks: 1) peak velocity measurement is dependent on the angle at which the region of interest is studied, and 2) overall heart motion, cardiac rotation, and contraction in adjacent segments can influence regional velocity estimates. To overcome these problems, the measurement of regional velocity by ultrasonic RV strain, which is relatively independent of overall heart motion, has been developed; for reviews of this work, see D'hooge²¹ and Sutherland²² and their associates. Recent reports²³ have shown that using RV strain is not only feasible for the quantification of regional RV function but is usually applicable even in cases in which the acoustic window is poor, as is often true in patients with chronic obstructive pulmonary disease. To our knowledge, despite the wide application of RV strain to the study of various RV diseases, this investigative technique has never before been used in the evaluation of post-CABG RV dysfunction.

Using RV strain in the current study enabled us to see a significant loss in RV function 1 year after CABG. Although the value of the decrease was not statistically significant compared with the values shown by TAM and TAV, the reduction revealed by RV strain is perhaps more reliable and more probably is a true RV dysfunction: regional peak systolic values (RV strain) have a close correlation with intrinsic contractility. Notably, ischemia-related changes have also been shown to be detectable earlier in systolic RV strain indices than have changes either in tissue velocity or in visual detection of regional wall-motion abnormalities.²⁴ In addition, there are the practical limitations on velocity measurement that we mention above.

Given that ischemia is 1 of the causes posited for post-CABG RV dysfunction,^11,12 we compared post-CABG RV function in patients whose RCA had been revascularized with that same function in patients whose RCA had not been revascularized. The 3 different quantitative approaches to the evaluation of RV function (TAM, TAV, and RV strain) revealed no difference between the 2 groups. This finding agrees with that of a previous study, which also showed that TAM was equally depressed in both groups after CABG.⁸ An earlier study had reported RV ejection fraction to be higher postoperatively for patients who had undergone RCA revascularization.¹²

We also evaluated pump time and cross-clamp time with the intention of evaluating their impact on postoperative RV function but found no relation. Moreover, we evaluated the effect of the site of RCA stenosis (proximal or distal) on postoperative RV function but could find no difference.

Septal motion in our study remained paradoxical for as long as 1 year after CABG, which is in accord with the findings of Hedman and his group.⁸

Our findings suggest that functional class may be related to changes in both left ventricular and RV performance, rather than to changes in resting cardiac function. In addition, paradoxical interventricular septal motion might compensate for decreased RV systolic function. It is likely that the septum is recruited to maintain RV stroke volume after CABG.

Conclusions

One year after CABG, RV function measured via TAM, TAV, and RV strain remained highly depressed in our patients. We suggest that RV strain be applied as an alternative to TAM and TAV in the assessment of RV function, particularly when there are limitations in the application of TAM and TAV.

Despite reduced RV function, functional class 1 year after CABG was improved in our patients; it is worthy of note that the measurement of RV function by echocardiography after CABG probably has no clinical significance. We also conclude that reduction in RV function is independent of RCA revascularization.

A precise evaluation of RV function, notwithstanding the mechanisms already postulated for post-CABG RV dysfunction, requires further studies via various investigative methods.