Myocardial oxygen supply depends on both myocardial blood flow and its oxygen content, as reflected by blood hemoglobin levels. However, there is little data on the impact of hemoglobin levels on myocardial blood flow as captured by positron emission tomography (PET). Consecutive patients undergoing Rb-82 PET myocardial perfusion imaging (MPI) using regadenoson stress testing and hemoglobin levels available within 90 days of imaging were included. Patients with significant perfusion defects were excluded. Pearson correlation coefficients and multivariable Gamma regression models were used to study the association between myocardial flow reserve (MFR) and hemoglobin levels. The consistency of the prognostic value of MFR across different levels of hemoglobin was assessed using multivariable Andersen-Gill Cox models with robust standard errors. A total of 2,676 (median age: 67 years; 54.6% females) patients were included. Hemoglobin levels were negatively correlated with rest MBF (-0.33; P < 0.001), positively correlated with stress MBF (0.04; P : 0.04) and positively correlated with MFR (0.38; P < 0.001). In multivariable models, hemoglobin levels were more strongly associated with rest MBF than with stress MBF. The prevalence of microvascular dysfunction (MFR < 2) was significantly less common at higher hemoglobin levels, with an adjusted prevalence of 52.0%, 35.6%%, and 24.6% at hemoglobin levels of 10, 12, and 14 g/dL respectively. MFR was prognostic across the spectrum of hemoglobin, with no statistically significant evidence of an interaction ( P for interaction = 0.269). Hemoglobin levels are negatively associated with rest MBF, weakly positively associated with stress MBF, and positively associated with MFR. The prevalence of microvascular dysfunction, reflected by an MFR < 2, is significantly higher in patients with lower hemoglobin levels. However, the prognostic value of MFR is maintained across the spectrum of hemoglobin levels.
Sayed et al. (Sat,) studied this question.