BACKGROUND: Acute kidney injury is a common comorbidity for children placed on extracorporeal membrane oxygenation (ECMO) because of primary cardiac disease. Continuous venovenous hemofiltration (CVVH) can optimize fluid status and lessen inflammatory response during ECMO. However, published data are derived primarily from children without primary cardiac disease. METHODS: A retrospective analysis of our institutional ECMO database from 2002 to 2011 was performed. To limit the bias that CVVH initiation was after evidence of end-organ injury, we considered "early CVVH" to be instituted within 48 hours of ECMO initiation. Multivariate logistic regression was undertaken to adjust for covariates. RESULTS: Of 153 cardiac ECMO patients, 59 (39%) received early CVVH. Time from ECMO initiation to CVVH initiation was 1.7±0.7 days (median 1 day). Pre-ECMO and post-ECMO serum creatinine levels were similar in both groups. However, peak serum creatinine was 1.1±0.4 mg/dL (median 1.0 mg/dL) in the ECMO and CVVH group and 0.9±0.4 mg/dL (median 0.8 mg/dL) in the ECMO alone group (p=0.003). Patients who received CVVH had a higher mortality (p<0.0001), were less likely to have had ECPR (p=0.004), and had a longer duration on ECMO (p<0.0001). In multivariate analysis subjects receiving CVVH support within 48 hours of ECMO cannulation were 3 times more likely to die during their hospitalization (odds ratio 3.02; 95% confidence interval 1.32 to 6.9, p=0.009) after adjusting for other significant risk factors. CONCLUSIONS: Early CVVH in pediatric cardiac patients requiring ECMO is associated with increased mortality. Early CVVH in the cardiac ECMO population does not appear justified.
BACKGROUND: Acute kidney injury is a common comorbidity for children placed on extracorporeal membrane oxygenation (ECMO) because of primary cardiac disease. Continuous venovenous hemofiltration (CVVH) can optimize fluid status and lessen inflammatory response during ECMO. However, published data are derived primarily from children without primary cardiac disease. METHODS: A retrospective analysis of our institutional ECMO database from 2002 to 2011 was performed. To limit the bias that CVVH initiation was after evidence of end-organ injury, we considered "early CVVH" to be instituted within 48 hours of ECMO initiation. Multivariate logistic regression was undertaken to adjust for covariates. RESULTS: Of 153 cardiac ECMO patients, 59 (39%) received early CVVH. Time from ECMO initiation to CVVH initiation was 1.7±0.7 days (median 1 day). Pre-ECMO and post-ECMO serum creatinine levels were similar in both groups. However, peak serum creatinine was 1.1±0.4 mg/dL (median 1.0 mg/dL) in the ECMO and CVVH group and 0.9±0.4 mg/dL (median 0.8 mg/dL) in the ECMO alone group (p=0.003). Patients who received CVVH had a higher mortality (p<0.0001), were less likely to have had ECPR (p=0.004), and had a longer duration on ECMO (p<0.0001). In multivariate analysis subjects receiving CVVH support within 48 hours of ECMO cannulation were 3 times more likely to die during their hospitalization (odds ratio 3.02; 95% confidence interval 1.32 to 6.9, p=0.009) after adjusting for other significant risk factors. CONCLUSIONS: Early CVVH in pediatric cardiac patients requiring ECMO is associated with increased mortality. Early CVVH in the cardiac ECMO population does not appear justified.
Authors: Santiago Borasino; Yuvraj Kalra; Ashley R Elam; Lawrence Carlisle O'Meara; Joseph G Timpa; Kellen G Goldberg; J Leslie Collins Gaddis; Jeffrey A Alten Journal: J Extra Corpor Technol Date: 2018-12
Authors: Heidi J Murphy; John B Cahill; Katherine E Twombley; David J Annibale; James R Kiger Journal: J Artif Organs Date: 2017-10-30 Impact factor: 1.731
Authors: Heidi J Murphy; Meryle J Eklund; Jeanne Hill; Kristen Morella; John B Cahill; James R Kiger; Katherine E Twombley; David J Annibale Journal: J Artif Organs Date: 2019-07-25 Impact factor: 1.731
Authors: Javier J Lasa; Rachel S Rogers; Russell Localio; Justine Shults; Tia Raymond; Michael Gaies; Ravi Thiagarajan; Peter C Laussen; Todd Kilbaugh; Robert A Berg; Vinay Nadkarni; Alexis Topjian Journal: Circulation Date: 2015-12-03 Impact factor: 29.690