MONTREAL -- Gender-specific nuclear cardiology norms for known or suspected cardiovascular disease can help assure that men don't receive treatment appropriate for women, and vice versa, according to a review reported here.
MONTREAL, Sept. 13 -- Gender-specific nuclear cardiology norms for known or suspected cardiovascular disease can help assure that men don't receive treatment appropriate for women, and vice versa, researchers reported here.
Using gender-specific norms to identify which patients have abnormal perfusion and function values on single emission photon computed tomography (SPECT) imaging can help reduce the gender bias associated with cardiac care, found a team from the University of Rochester (N.Y.).
Women with known or suspected ischemic disease receive less treatment, are referred less often for cardiac catheterization, and have worse outcomes than men, said the group.
In the report presented at the American College of Nuclear Cardiology meeting, the investigators set out to determine whether this discrepancy may be at least partly explained by the fact that normal limits on SPECT perfusion and function parameters should be adjusted for gender.
"The point was to determine whether gender differences in left-ventricular volume and perfusion were prognostically important, and whether we should be using different values for men and women for risk stratification," said Orren Wexler, a medical student. Ronald G. Schwartz, M.D., was senior author.
The investigators reviewed gated SPECT perfusion studies performed on 891 consecutive patients, 43% of whom were women. They determined the gender-specific limits that optimally predict cardiac events for stress and rest hypoperfusion defect scores (SHS and RHS) as well as end-systolic and diastolic volume indices (ESVI and EDVI) using receiver operator curve (ROC) analyses.
They then employed Cox proportional hazard and Kaplan-Meier analyses to evaluate the predictive values of the new gender specific limits.
Among the 839 patients for whom follow-up was available, 6% of men and 7% of women went on to have a cardiac event, defined as non-fatal myocardial infarction and cardiac death. These proportions were statistically identical. "At baseline, women had smaller, less dilated hearts, even when corrected for body size," the investigators said.
"So, they had smaller systolic volumes, smaller diastolic volumes, and smaller ejection fractions. They also had smaller rest and stress hypoperfusion defects. Despite what would seemingly be more favorable baseline characteristics, at follow-up they still had the same rates of non-fatal MIs and cardiac death."
Overall, 30% of the men underwent cardiac catheterization, compared with 17% of the women. Similarly, 16% of the men underwent revascularization, compared with 7% of the women. These last two gender comparisons were both significant at P<0.001.
On the basis of their ROC analyses, the normal limits that optimally predicted coronary events were different for men and women. The normal limit for ESVI was 53 mL/m2 for men and 37 mL/m2 for women. The normal limit for EDVI was 53 mL/m2 96 mL/m2 for men and 80 mL/m2 for women. Normal limit for SHS was 1% for men and 4% for women, while normal limit for RHS was 1% for men and 3% for women. The gender specific limits of hypoperfusion, LV indices, and LV ejection fraction (EF) were all predictive of coronary events.
Had referral rates for angiography been based on gender-specific limits of ESVI, rates of referral would have been similar for both groups. ESVI was abnormal in 32% for men and 37% for women using these new limits (P=0.115). Similarly, the proportion of abnormal EDVI scores was 42% in men and 36% in women using the gender specific limits. These only trended toward statistical difference (P=0.059). SHS and RHS scores were still more commonly abnormal in men, even when gender specific limits were used.
This study demonstrates that, "we should be using different prognostic values for men and women," said the Rochester team, although additional research is required to confirm these findings.