RESUMO
Background: We aimed to clarify the predictive factors for left ventricular (LV) function after aortic valve replacement (AVR) in patients with aortic regurgitation (AR). Methods and results: Among 555 patients who underwent AVR at our institution from January 2015 to December 2020, we enrolled 44 patients for whom only AVR (or AVR + aortic replacement) was performed. We defined LV dysfunction under any of the following criteria: LV ejection fraction (LVEF) <50 %, LV diastolic dimension >65 mm, LV systolic dimension (LVDs) >50 mm, or LVDs/body surface area > 25 mm/m2. Multivariable logistic regression analysis revealed high natural logarithm (ln) C-reactive protein (CRP) and low LVEF in the pre-AVR period significantly associated with LV dysfunction after AVR (ln CRP: odds ratio [OR] 4.15, 95 % confidence interval [CI] 1.44-11.98, p < 0.01; LVEF: OR 0.79, 95%CI 0.65-0.97, p < 0.05). Receiver-operating characteristic analysis revealed an area under curve of CRP and LVEF in the pre-AVR period for LV dysfunction after AVR of 0.84 and 0.83, respectively. Upon dividing the patients into four groups according to cutoff values of CRP (0.13 mg/dL) and LVEF (50 %) in the pre-AVR period, no patients (0/19) had LV dysfunction in the low CRP (<0.13 mg/dL) and high LVEF (≥50 %) group, and all patients (5/5) in the high CRP (≥0.13 mg/dL) and low LVEF (<50 %) group had LV dysfunction after AVR. Conclusion: High CRP level was significantly and independently associated with LV dysfunction after AVR. Combination of CRP and LVEF values might be useful for predicting improvement in LV function after AVR.