ABSTRACT
BACKGROUND: Although acute kidney injury (AKI) has been established as an independent risk factor for in-hospital mortality for patients on veno-arterial (V-A) extracorporeal membranous oxygenation (ECMO), the impact of Kidney Disease Improving Global Outcomes (KDIGO) stages of AKI has yet to be elucidated as a risk factor. METHODS: We conducted a retrospective analysis of patient outcomes based on KDIGO stages of AKI at a single institution. The analysis was a cohort of 179 patients; 66 without AKI, 19 with stage 1 AKI, 18 with stage 2 AKI, and 76 with stage 3 AKI. RESULTS: Every 1-year increase in age was associated with 4% increased odds of mortality at 30 days (95% confidence interval [CI] 1.01, 1.07; p = 0.004). The presence of AKI at any stage was associated with 59% increased odds of 30-day mortality (95% CI 0.81, 3.10; p = 0.176). The presence of stage 1 AKI was associated with a 5% decreased odds of 30-day mortality (95% CI 0.32, 2.89). The presence of stage 2 AKI (odds ratio [OR] 2.29, 95% CI 0.69, 7.55; p = 0.173) and stage 3 AKI (OR 1.68, 95% CI 0.81, 3.46; p = 0.164) was associated with increased odds of 30-day mortality. CONCLUSION: Based on our single-center study, higher KDIGO stages of AKI likely have increased odds of mortality at 30 days. Larger studies are needed to confirm these findings.
ABSTRACT
Preoperative cardiopulmonary exercise testing (CPET) is well validated for prognostication before advanced surgical heart failure therapies, but its role in prognostication after LVAD surgery has never been studied. VE/VCO2 slope is an important component of CPET which has direct pathophysiologic links to right ventricular (RV) performance. We hypothesized that VE/VCO2 slope would prognosticate RV dysfunction after LVAD. All CPET studies from a single institution were collected between September 2009 and February 2019. Patients who ultimately underwent LVAD implantation were selectively analyzed. Peak VO2 and VE/VCO2 slope were measured for all patients. We evaluated their association with hemodynamic, echocardiographic and clinical markers of RV dysfunction as well mortality. Patients were stratified into those with a ventilatory class of III or greater. (VE/VCO2 slope of ≥ 36, n = 43) and those with a VE/VCO2 slope < 36 (n = 27). We compared the mortality between the 2 groups, as well as the hemodynamic, echocardiographic and clinical markers of RV dysfunction. 570 patients underwent CPET testing. 145 patients were ultimately referred to the advanced heart failure program and 70 patients later received LVAD implantation. Patients with VE/VCO2 slope of ≥ 36 had higher mortality (30.2% vs. 7.4%, p = 0.02) than patients with VE/VCO2 slope < 36 (n = 27). They also had a higher incidence of clinically important RVF (Acute severe 9.3% vs. 0%, Severe 32.6% vs 25.9%, p = 0.03). Patients with a VE/VCO2 slope ≥ 36 had a higher CVP than those with a lower VE/VCO2 slope (11.2 ± 6.1 vs. 6.0 ± 4.8 mmHg, p = 0.007), and were more likely to have a RA/PCWP ≥ 0.63 (65% vs. 19%, p = 0.008) and a PAPI ≤ 2 (57% vs. 13%, p = 0.008). In contrast, peak VO2 < 12 ml/kg/min was not associated with postoperative RV dysfunction or mortality. Elevated preoperative VE/VCO2 slope is a predictor of postoperative mortality, and is associated with postoperative clinical and hemodynamic markers of impaired RV performance.
