Identification of patients who derive mortality benefit from ICD implantation by EPS with concurrent revascularization

Dali Fan, Eric Kessler, Patrick Lam, Darryl Hoffman, Charles Geller, Loren Harris, Jamie Stelzer, and Paul Schweitzer
The Heart Institute, Beth Israel Medical Center, New York, NY 10003

Abstract: It has been unclear on how to optimally manage asymptomatic patients with low EF, who were found on routine monitoring to have NSVT immediately after undergoing re-vascularization. Use of EPS has been our standard practices to risk stratify these patients. However, its value in patients with ischemic heart disease, non-sustained ventricular arrhythmia and low EF is limited in general. In the MUSTT study, patients with inducible ventricular tachycardia on EPS obtained defined benefits from placement of ICDs, while those who were non-inducible on EPS still suffered relatively high rates of mortality.

We retrospectively studied 92 consecutive patients who were revascularized by PTCA or CABG within 4 weeks of subsequent EPS.  Those with a positive EPS underwent ICD placement. Those with negative EPS were treated medically.  The primary endpoint was all cause mortality. The follow-up was a total of 75,751 days for the 26 patients in the ICD group and the 66 patients in the medically treated group.  The mortality rate amongst the ICD group was 11.5% and 12.5% in the medically group.

In our small series, EPS seems to be a valid risk stratifier in patients who had maximal reduction in their ischemic burden immediately prior to testing.

INTRODUCTION

It has been unclear on how to optimally manage asymptomatic patients with low EF, who were found on routine monitoring to have NSVT after undergoing re-vascularization. Use of EPS has been our standard practice to risk stratify these patients. However, its value in patients with ischemic heart disease, non-sustained ventricular arrhythmia and low EF is limited in general. The MUSTT study established the mortality benefit with use of ICDs in patients CAD, LVEF less than 40% and documented NSVT who had sustained VT induced by EPS[1]. Unfortunately, while those who had positive EPS had their two-year mortality reduced from 28% to 11% via ICD implantation, those who had negative EPS studies still had a 21% mortality rate when treated medically over the same period of time. This implies that some EP negative patients could have benefited from ICD placement. Ischemia is the main reversible risk factor in this patient population and it can be modified by current techniques like PTCA or CABG[2,3].

Our goal was to determine if our standard practice is valid in terms of mortality reduction by using EPS guided risk stratification for ICD implantation immediately after revascularization by CABG or PTCA. We hoped to determine the effect of our strategy on the mortality gap between the EPS positive ICD protected patients and the EPS negative medically treated group.  We studied patients who underwent EPS testing to determine the need for ICD therapy within one month of revascularization with either PTCA or CABG. These patients would be at the apex of attainable ischemic control, and would thus present the most optimally managed substrate for EPS. Our desire was to determine if our standard practice is valid from mortality point of view.

METHODS

We retrospectively evaluated a series of 92 consecutive patients who underwent EPS testing within 30 days of subsequently undergoing either CABG or PTCA procedures at Beth Israel Medical Center in New York City between January 1998 and May 2001. Subjects were identified from the databases of CABG, PTCA and EPS. Each patient in the study must have an US social security number. Patients who underwent EPS more than 30 days post revascularization were excluded. Patients who had sudden death were likewise excluded.  All patients who underwent PTCA had at least one lesion of >70% stenosis intervened on. Charts were reviewed to glean various data including patient age, sex, ethnicity, and medications at discharge. Data on ejection fraction were taken from either ventriculogram results in the catheterization reports or echocardiography studies performed no more than 6 months before or after revascularization. The US social security registry determines death.

Positive EPS is defined as sustained, monomorphic VT induced by programmed ventricular stimulation up to three extrastimuli. The cumulative mortality experience of each treatment group was summarized as a survival curve, which was estimated using the Kaplan-Meier method (Figure 1.). The survival curves were compared using a Mantel-Haenszel test. Both study groups had similar characteristics of age, sex, use of angiotensin blockade and beta blockade, ejection fraction and follow-up. However, less than half of the patients had a complete baseline characteristics profile from our chart review.

TABLE I
Baseline Characteristics of the study patients

Group	EPS+/ICD	EPS-/No ICD
Age	72	69
Male	67%	76%
Ejection Fraction (EF)	33%	36%
ACE	38%	39%
Beta Blockade	50%	48%
Underwent CABG pre EPS	18	35
Underwent PTCA pre EPS	8	31
Average F/U (days)	901	879

 

Figure 1.  Kaplan-Meier Survival Curves for EPS+/ICD group (diamond) and EPS-/no ICD Group (square)

 

TABLE 2. Mortality

Group	EPS+/ICD	EPS-/No ICD
Mortality	11.5%	12.1%

RESULTS

A total of 92 patients were identified by our criteria, of which 26 underwent ICD placement after a positive EPS test. The remaining 66 patients had a negative EPS study and were treated medically. Essential demographic characteristics for those with and without ICD placement were similar and are summarized in table 1.

The results for the endpoint of overall mortality are shown in Table 3: mortality in the ICD group was 11.5%, whereas the mortality in the group that did not qualify for ICD implantation was 12.1%. Kaplan-Meier graphical analysis of mortality data is displayed in table 2 and shows that mortality rates are relatively steady and temporally concordant between the two groups over the follow up period, which averaged 901 and 879 days for the two subject sets. The shortest follow up period not truncated by death was 349 days, and the longest follow up period was 1397 days.

DISCUSSION

It is not uncommon in our daily practice to see asymptomatic patients with non-sustained VT on routine monitoring post PTCA or CABG. This often presents a management dilemma in terms of interpreting the significance of NSVT in the post ischemic state. The standard practice in our medical center is to perform EPS on these patients, in broad accord with the MUSTT data. When sustained VT is induced with standard programmed stimuli in EPS, the patient receives an ICD. If sustained VT is not induced, the patient is treated medically. EPS has been evaluated in the past by testing reproducibility in different patient populations, and it has been noted that EPS results change over the long term, probably at least in part due tochanges in ischemic burden.  We believe that when ischemia is minimized as an operant factor at the time of EPS, its accuracy may be improved. Indeed, the data from our study showed that after PTCA for CABG, the mortality of the EPS negative, medically treated patients is the same as the EPS positive, ICD protected patients.

We chose all-cause mortality as our endpoint for several reasons. The main reason is it is difficult to judge the accuracy of the cause of death documented in charts and death certificates. Another reason is that almost thirty percent of the patients chose not to have routine follow-up in our medical center.  This made it impossible to track other proxy events, such as “arrhythmic death” or “appropriate ICD discharge events”. Yet another reason is by using all-cause mortality as our endpoint, it possible, at least from practical point of view, to use the all-cause mortality rate historical data from large trials such as the MUSTT study as a reference. This position is reflective of the recommendations put forth by the NASPE  in its guidelines on ICD placement[4].

Our results on the rates of mortality in the EP positive/ICD treated and the EP negative/no ICD group were very close, though the confidence intervals allowed for more significant disparity. The mortality in our treatment group (11.5%) is also very close to the historical control of MUSTT (about 11% for a similar group over a similar follow up period). But again, the spread of the confidence intervals for mortality allow for significant variation, pointing up the need for such observations to be confirmed by larger, preferably randomized studies.

In summary, our small, retrospective study validated our practice strategy of identifying high-risk patients for ICD implantation by EPS after peak revascularization. EPS negativity in a setting of recent revascularization may be adequate in identifying a cohort of patients who may be treated medically with mortality rates approximately in line with their ischemia free, EPS positive and ICD protected brethren. However, because of the nature and sample size of our study, these, conclusions must be confirmed with larger randomized studies.

REFERENCES

1) Buxton A, Lee K, Fisher J, et.al. A randomized study of the prevention of sudden death in patients with coronary artery disease. N Engl J Med 1999;341:1882-1890.

2) Moss A, Hall J, Cannom D, et.al.Improved survival with an implanted defibrillator in patients with coronary disease at high risk of ventricular arrhythmias. N Engl J Med 1996;335:1933-1940.

3) Bigger JT, et.al. Prophylactic use of implanted cardiac defibrillators in patients at high risk for ventricular arrhythmias after coronary artery bypass graft surgery. N Engl J Med 1997

4) Winters SL. Packer DL. Marchlinski FE. et.al. Consensus statement on indications, guidelines for use, and recommendations for follow-up of implantable cardioverter defibrillators. NASPE. Pacing & Clinical Electrophysiology. 24(2):262-9, 2001 Feb.