Invasive, conservative strategies confer similar long-term mortality in stable ischemia
Click Here to Manage Email Alerts
CHICAGO — In stable patients with ischemia, there was no difference between an invasive strategy and a conservative strategy of optimal medical therapy in mortality at 5.7 years, according to results of the ISCHEMIA-EXTEND study.
However, CV mortality was higher in the conservative group and non-CV mortality was higher in the invasive group, Judith S. Hochman, MD, senior associate dean for clinical sciences, co-director of the Clinical and Translational Science Institute, Harold Snyder Family Professor, associate director of cardiology and director of the Cardiovascular Clinical Research Center at the NYU School of Medicine, said during a presentation at the American Heart Association Scientific Sessions.
ISCHEMIA-EXTEND is an extended follow-up study of 4,825 patients (median age, 64 years; 78% men; 66% white; 29% Asian) from the ISCHEMIA trial. As Healio previously reported, in the main results of ISCHEMIA, an invasive strategy and a conservative strategy of optimal medical therapy yielded similar CV outcomes in stable patients with moderate or severe ischemia.
The outcomes of interest were all-cause mortality, CV mortality and non-CV mortality; deaths of undetermined cause were counted as CV mortality. Hochman presented interim results representing a median of 5.7 years of follow-up; the cohort will be followed out to 10 years.
“What matters most to patients is living longer and feeling better,” Hochman told Healio. “There is very valuable, rich potential data to follow these patients long term, to actually have power and precision around the ... hazard ratio for survival. We have funding from the NHLBI to report final results in 2026, but we believed that this was important information to get out.”
At 5.7 years, 557 of the patients in ISCHEMIA-EXTEND had died, almost twice as many as during the original analysis at 3.2 years, Hochman said during the presentation.
All-cause mortality was virtually identical in both groups (7-year rates: conservative, 13.4%; invasive, 12.7%; adjusted HR = 1; 95% CI, 0.85-1.18; log-rank P = .741), she said.
CV mortality was higher in the conservative group (7-year rates: 8.6% vs. 6.4%; aHR = 0.78; 95% CI, 0.63-0.96; Fine-Gray P = .008), but non-CV mortality was higher in the invasive group (7-year rates: 4.4% vs. 5.6%; aHR = 1.44; 95% CI, 1.08-1.91; Fine-Gray P = .016), Hochman and colleagues found.
“The apparent divergence in cardiovascular mortality that we saw and published in The New England Journal of Medicine, those curves have progressively diverged, so there is now actually a significant reduction in cardiovascular mortality with an initial invasive strategy,” Hochman told Healio. “That is exactly offset by an increase in non-cardiovascular death in the invasive strategy, largely, as we have previously reported, driven by malignancy. We have published on the causes of death and have already shown there was excess in non-cardiovascular death. Now, both differences are significant and essentially cancel each other out.”
She said the curves for non-CV death start to converge about 6 months later than the curves for CV death, and the death rate in the cohort is about 2% per year.
“If this were a competing risk issue, with that timing and that low death rate, you’d have to hypothesize that every patient saved from a cardiovascular death went on in short order to die of cancer, but that is pretty unlikely,” Hochman told Healio. “So we don’t think it’s a competing risk issue. Maybe it’s chance. It would be odd to have differences in the way deaths are reported, but that’s a possibility. We don’t know the answer.”
A previous analysis found that dual antiplatelet therapy, used more often in the invasive group, was not associated with CV death, she said, which rules out the non-CV mortality difference being driven by excess bleeding from DAPT.
However, she said, “we did see an association with use of procedures: angiography and fluoroscopy in the cath lab, and with an increased number of procedures that had radiation involved. But we had two radiation experts on that analysis, and it is implausible that malignancies would be induced over that short of a time period by any level of radiation.”
The results did not vary across prespecified subgroups, according to the researchers.
When the researchers stratified the analyses by whether patients had multivessel disease, defined as stenosis of 70% or more in at least two vessels, they found the treatment effect for all outcomes did not differ by multivessel disease status (P for interaction for all > .05).
A Bayesian analysis of posttest probability that there is a difference between the groups indicated that the probability that one strategy is superior to the other is extremely low, Hochman said.
“We now have much better data on which to base our discussion with patients. What’s important for patients is that all-cause mortality is the same, so the upshot is the importance of patient engagement and shared decision-making,” Hochman told Healio. “If they have angina and place a high priority on being angina-free, then an invasive strategy is the way to go. Some patients prefer to have everything ‘fixed.’ Other patients don’t want anything put in their body, or their chest opened for surgery.”
References:
- Maron MJ, et al. N Engl J Med. 2020;doi:10.1056/NEJMoa1915922.
- Sidhu MS, et al. Am Heart J. 2022;doi:10.1016/j.ahj.2022.01.017.
Perspective
Back to Top
C. Noel Bairey Merz, MD, FACC, FAHA
The take-home message is the same as for the original findings: With an invasive strategy, all-cause mortality is not reduced, CV mortality is comparably reduced and non-CV mortality is consistently elevated.
It is interesting that the curves remained the same over years despite an uncertainty about use of optimal medical therapy. Given that patients are less adherent, about 60%, when they are not part of a clinical trial, we might assume that adherence to optimal medical therapy was reduced over time, the results were durable. If that turns out to be true, that tells us that the 3 to 5 years of treatment altered their trajectory, similar to what we saw in the statin trials. Even if the patient does not take the medication another day in their life, you did them some good. That’s an important message.
Hochman’s conclusions are appropriate: This is information to be discussed between the physician and patient in a shared decision-making process.
We are all curious about the increased malignancy mortality in the invasive group. This may be a canary-in-the-mine signal that these invasive procedures and these repetitive studies using ionizing radiation may be contributing to cancer, whereas optimal medical therapy mostly does not use radiation. This is a large-enough sample size that you can see relatively small effects that are cumulative over time. Medical radiation is now the highest contributor to an individual’s lifetime radiation exposure. There may finally be some data that support what radiation experts have been telling us, that they can extrapolate when these radiation thresholds are going to start causing cancer. This has always been an estimate, but we may now be able to see it.
Perspective
Back to Top
I am pleased that funding has continued for what is probably the most landmark cardiology trial of this decade. The present ISCHEMIA-EXTEND analysis occurs at a median of 5.7 years, but patients will be followed for a full 10 years. There have been trials like STICH that have not shown interim differences between treatments at 5 years but have shown a significant difference at 10 years.
There are certain groups, such as those with multivessel disease and those who are interested in having better angina-related quality of life, who do seem to benefit from invasive management.
I look forward to the updates from this study in years to come, so we can finally answer the question about whether or not conservative management is equal to invasive management for our patients with stable ischemic heart disease.
Sanger Heart and Vascular Institute/Atrium Health
Clinical Professor
Wake Forest University School of Medicine
Vice President, American College of Cardiology
Collapse
Hochman JS, et al. LBS.06. Drugs and Strategies in ACS and Revascularization. Presented at: American Heart Association Scientific Sessions; Nov. 5-7, 2022; Chicago (hybrid meeting).
Click Here to Manage Email Alerts