Doma et al.: chronic total occlusion PCI in acute MI with cardiogenic shock

Summary

Research letter on chronic total occlusion (CTO) PCI in patients with acute MI complicated by cardiogenic shock. CTO-PCI was identified by CPT/HCPCS billing codes. The time criterion required a shock diagnosis on the day before or the same day as the CTO-PCI.

The cohort had 617 patients. 81.9% also had PCI of another vessel, described as preceding the CTO-PCI (culprit-vessel treatment). More than 60% received mechanical support (intra-aortic balloon pump, Impella or ECMO).

The authors compared 30-day mortality with the roughly 50% reported in large all-comer AMI-shock registries (Karami et al., Goyal et al.).

Strengths

• The first dedicated study of CTO-PCI in AMI with shock that I know of.
• A sample of 617 patients in a scenario with little data.

Methodological issues

• CTO not confirmed. The consensus definition requires a documented occlusion of at least 3 months. With no core lab, adjudication or angiographic confirmation of chronicity, classification depends on the operator’s code. The cohort likely includes acute occlusions coded as CTO. Treating an acute non-culprit occlusion in shock is a different scenario from opening a chronic, collateralized one.
• When the CTO was identified. The CTO did not have to be found during the index admission. A patient with a known right coronary CTO who presents with anterior MI and shock meets the criteria. The cohort mixes at least three groups: CTO found at the index angiogram, known CTO treated because of the shock (for example, difficulty weaning support after culprit PCI) and acute occlusion coded as CTO. No supplementary material allows separating them.
• No timing data. The time criterion is coarse. The actual interval and whether patients were stabilized before CTO-PCI are unknown. The roughly 18% without culprit PCI are the least clear subgroup: possibly NSTEMI with multiple CTOs or a vein graft culprit.
• No comparator. The study is single-arm. The cited registries include SCAI C to E, out-of-hospital arrest, mechanical complications and patients who never reached the cath lab. Without SCAI staging or hemodynamic data, it is unknown whether CTO-PCI was done in stable patients or as a last attempt.
• Selection bias. To undergo CTO-PCI in shock, a patient must survive the acute event, have culprit PCI (81.9%), stabilize on mechanical support (more than 60%) and be judged fit for a complex procedure. No one does a retrograde CTO in SCAI E. These are the survivors judged salvageable.
• Immortal time bias. The interval from shock diagnosis to CTO-PCI is immortal: the exposed group cannot die before the procedure. The comparison registries accumulate deaths from time zero. This alone can produce an apparent survival advantage.
• Descriptive analysis only. No multivariable adjustment, propensity matching or regression. With a mean follow-up of 1.68 ± 1.97 years, the 5-year mortality of 40.4% rests on few patients at risk after 3 years. There are no confidence intervals for cumulative incidence. There are no subgroups (staged versus same-sitting PCI, balloon pump versus Impella, culprit plus CTO versus CTO only).