Long-term prognostic implications of myocardial perfusion imaging in octogenarians: an all-comer,cohort study

Purpose

Evaluation of the long-term prognostic value of myocardial perfusion imaging (MPI) in octogenarians.

Methods

Six hundred and twenty-nine octogenarians [51% previous myocardial infarction (MI) or revascularization] who underwent single-isotope MPI (78% ²⁰¹Tl, 22% ^99mTc-tetrofosmin) with exercise (38% Bruce, 2% leg ergometry) or pharmacologic (58% adenosine, 2% dobutamine) stress were studied. All patients had LVEF determined by echocardiography within 1 month from MPI. Myocardial perfusion scoring was performed on a 17-segment LV-model with a 5-point grading system and three summed stress score (SSS)-based risk categories were formed [high-(SSS > 12), low-(SSS < 4), medium]. Prospective follow-up was performed to document all-cause (ACD), cardiac death (CD), MI, and revascularization. Revascularization was used to censor follow-up in survival analysis regarding ACD, CD, and CD/MI. For analysis of the CD, MI, or late revascularization (LR) composite, only revascularizations within 3 months from MPI (early revascularizations) were used for censoring.

Results

After 9.3 years there were 187 ACDs, 86 CDs, 28 MIs, and 77 revascularizations, including 28 early revascularizations. Adjusting for LVEF and stress-modality type, SSS was identified as an independent predictor of ACD [HR 1.03 (1.01–1.05)], CD [HR 1.05 (1.03–1.08)], CD,MI [HR 1.05 (1.02–1.07)], and CD,MI or LR [HR 1.05 (1.03–1.07)] (p ≤ 0.001 in all cases). Increased lung uptake had independent prognostic value only for the CD, MI, or LR end-point [HR 3 (1.2–7.7), p = 0.02]. Survival modeling demonstrated that LVEF and SSS, but not non-perfusion scintigraphic data provided incremental prognostic value over pre-test available clinical and historical information for all end-points. Differences between Kaplan-Meier survival curves of SSS-based risk groups were significant for all end-points (p < 0.001 in all cases).

Conclusions

In octogenarians, MPI provides effective long-term risk stratification, regardless of stress type used or underlying cardiac function.