D Perfusion abnormalities in hypertrophic cardiomyopathy: mechanisms and prognostic importance. (June 2018)

• Record Type: Journal Article
• Authors: Raphael, Claire E; Cooper, Robert; Mitchell, Frances; Liew, Alphonsus; Newsome, Simon; O'Hanlon, Rory; Parker, Kim H; Vassiliou, Vassilis; Pennell, Dudley J; Keegan, Jennifer; Frenneaux, Michael; Stables, Rod H; Mario, Carlo Di; Prasad, Sanjay K
• Is Part Of: Heart. Volume 104(2018)Supplement 6
• Page Start: A112
• Page End: A113
• Publication Date: 2018-06
• Subjects: hypertrophic cardiomyopathy -- perfusion -- prognosis; Heart -- Diseases -- Treatment -- Periodicals; Cardiology -- Periodicals; 616.12
• DOI: 10.1136/heartjnl-2018-BCS.156
• Languages: English
• ISSNs: 1355-6037

Abstract

Abstract : Introduction: Heart failure develops in 3%–5% of patients with hypertrophic cardiomyopathy (HCM) and is associated with a ten-fold increase in mortality. Myocardial perfusion abnormalities using positron emission tomography (PET) predict patients at increased risk of heart failure and death, however the mechanism of these perfusion defects has not been demonstrated, nor has the added prognostic value of perfusion imaging over late gadolinium imaging been established. Methods: We measured simultaneous pressure and flow in the proximal left anterior descending artery in 33 HCM and 20 control patients at rest and during hyperemia, allowing calculation of wave intensity (WIA). Patients underwent quantitative first-pass perfusion cardiovascular magnetic resonance (CMR). We prospectively recruited a further 328 HCM patients for CMR, including adenosine first pass perfusion, LV and LA volumes and late gadolinium enhancement (LGE). We followed patients for development of heart failure, with a composite endpoint of heart failure death, cardiac transplantation and unplanned heart failure hospitalisation. Results: Patients with HCM had a lower coronary flow reserve than controls (1.9±0.8 versus 2.7±0.9, p=0.001). Coronary hemodynamics in HCM were characterised by a very large backward compression wave during systole (38%±11% versus 21±6%, p<0.001) and a proportionately smaller backward expansion wave (33%±6% versus 27±8%, p=0.006) compared to controls. Patients with severe left ventricular outflow tract (LVOT) obstruction had a bisferiens pressure waveform resulting in an additional proximally originating deceleration wave during systole. The proportion of waves acting to accelerate coronary flow increased with hyperemia and the magnitude of change was proportional to the myocardial perfusion reserve. (r=0.62, p<0.001). During a median follow up of 3.2 years, 32 patients met the heart failure endpoint. Baseline left atrial volume indexed to body surface area, LAVi, (HR 1.38 (1.2–1.6), p<0.001), %late gadolinium enhanceme, t%LGE (HR 1.03 (95% CI 1.01–1.05), p=0.03), LV ejection fraction (EF, 0.93 (0.9–0.97) p<0.001) and NYHA class at baseline were all univariate predictors of heart failure outcomes. On multivariable analysis, LAVi, %LGE, NYHA class, LVESVi and MR remained predictive. Presence or extent of inducible perfusion defect did not predict outcome (p=0.94, p=0.61 respectively). Discussion: Coronary flow in patients with HCM is deranged. Distally, compressive deformation of intra-myocardial blood vessels during systole result in an abnormally large backward compression wave, while proximally, severe LVOT obstruction is associated with an additional deceleration wave. Perfusion abnormalities in HCM are not simply a consequence of supply/demand mismatch or remodelling of the intra-myocardial blood vessels but represent a dynamic interaction with myocardial mechanics which may be amenable to treatment. In our prognostic study, we identified that while replacement fibrosis was a likely mechanism of progression to heart failure, LAVi as a more global measure of ventricular dysfunction was the strongest prognostic marker in HCM. Although perfusion using PET predicted development of heart failure in HCM, perfusion imaging with CMR did not predict outcome. Perfusion defects in HCM are complex and multifactorial, however do not correlate directly with outcome in large, prospective studies and should not be included in risk stratification scores.