Predictors of Oxygenator Exchange in Patients Receiving Extracorporeal Membrane Oxygenation.

Thrombosis within the membrane oxygenator (MO) during extracorporeal membrane oxygenation (ECMO) can lead to sudden oxygenator dysfunction with deleterious effects to the patient. The purpose of this study was to identify predictors of circuit exchange during ECMO. This is a single-center, retrospective study of all patients who received ECMO at our institution from January 2010 to December 2015. Changes in potential markers were compared on Day 3 vs. Day 0 before MO exchange. Of the 150 patients who received ECMO, there were 58 MO exchanges in 35 patients. Mean ECMO duration was 21.1 (±12.7) days. D-dimer (DD) (µg/mL) (mean difference -2.6; 95% confidence interval [CI]: -4.2 to -1.1; p = .001) increased significantly in the 3 days leading up to MO exchange, whereas fibrinogen (mg/dL) (mean difference 90.7; 95% CI: 41.8-139.6; p = .001), platelet (PLT) count (1,000/µL) (mean difference 23.3; 95% CI: 10.2-36.4; p = .001), and heparin dose (units/h) (mean difference 261.7; 95% CI: 46.3-477.1; p = .02) decreased. Increasing DD or decreasing fibrinogen, PLT count, or heparin dose may indicate an impending need for MO exchange in patients receiving ECMO. Early identification of these changes may help prevent sudden MO dysfunction.

Flow dynamics of different adult ECMO systems: a clinical evaluation.

Membrane oxygenator (MO) failure is a known hazard during venovenous extracorporeal membrane oxygenation (v-v ECMO) therapy. Knowledge about technical and performance details of different ECMO systems (Maquet, Rastatt, Germany; Medos, Stolberg, Germany; Sorin, Modena, Italy) licensed for adults with acute lung failure might improve their handling. This retrospective study comprises 186 adult patients (Regensburg ECMO Registry) treated with v-v ECMO. Flow dynamic data were used to analyze the performance of different blood pumps, cannula types, and MOs to maintain an adequate blood flow (1-5 L/min). Usage of the Medos ECMO system in critically ill patients required a higher pump speed and generated a higher pressure drop across the MO (dpMO), however, without an increase in free plasma hemoglobin. The dpMO depended on the type of MO and increased with blood flow as expected. Type-specific normal values are reported. A distinct increase in dpMO above normal values within 1 day required an immediate MO exchange. This was an infrequent technical complication (3%). Finally, pressure-flow performance of single dual-lumen cannulas (27 Fr) was comparable with small single-lumen cannulas (15 Fr), without an increased risk of technical-induced hemolysis. Despite different performances, all current commercially available adult v-v ECMO systems produce adequate blood flow without an increased risk in technical-induced hemolysis. Familiarity with the specific properties of individual systems allows early detection of technical complications. Additionally, the choice of an adequate cannula requires a closer consideration of the individual patient situation.

Early predictors of oxygenator exchange during veno-venous extracorporeal membrane oxygenation: A retrospective analysis.

INTRODUCTION: Oxygenator exchange during extracorporeal membrane oxygenation (ECMO) is a life-threatening procedure. D-dimer has been used to predict oxygenator failure, but it is a parameter used a few days before oxygenator exchange. This study investigated parameters before and immediately after ECMO initiation that predict oxygenator exchange. METHODS: This was a single-center, retrospective study of 28 patients who received veno-venous ECMO with heparin/silicone-coated polypropylene oxygenator (NSH-R HPO-23WH-C; Senko Medical Inc., Tokyo, Japan), due to acute respiratory failure, from April 2015 to March 2020. Clinical data before ECMO initiation and during the first 3 days on ECMO were compared between the patients with oxygenator exchange (exchange group) and those without oxygenator exchange (non-exchange group). RESULTS: Nine (32%) patients required oxygenator exchange. The exchange group had significantly higher white blood cell count (WBC) (16,944 ± 2423/µL vs 10,342 ± 1442/µL, p < 0.05) and Acute Physiology and Chronic Health Evaluation (APACHE) II score (31 ± 5 vs 25 ± 8, p < 0.05) before ECMO initiation than the non-exchange group. The partial pressure of oxygen at the outlet of the oxygenator (PO2 outlet) and activated partial thromboplastin time (aPTT) during the first 3 days on ECMO were significantly lower in the exchange group than in the non-exchange group. CONCLUSIONS: High WBC and APACHE II score before ECMO initiation, low PO2 outlet, and aPTT during the first 3 days on ECMO were associated with oxygenator exchange during veno-venous ECMO. These parameters could be used to avoid unexpected oxygenator exchange.

Comparison between clinical indicators of transmembrane oxygenator thrombosis and multidetector computed tomographic analysis.

This study aims to assess whether multidetector computed tomography (MDCT) could accurately confirm the clinical suspicion of transmembrane oxygenator thrombosis (MOT) during extracorporeal membrane oxygenation (ECMO). Twenty-seven oxygenators were examined using MDCT at the end of patient treatment. Transmembrane oxygenator thrombosis was suspected in 15 of them according to the presence of at least 2 of the following clinical indicators: (1) increase in d-dimer, (2) decrease in platelet count, (3) decrease in oxygenator performance, and (4) presence of clots on the surface of the oxygenator. Transmembrane oxygenator thrombosis was confirmed by MDCT in 5 (33%) of them. Transmembrane oxygenator thrombosis was unexpectedly found in 5 (41%) of the remaining 12 oxygenators not suspected for MOT. Eight (80%) of these oxygenators had clots accounting for less than 1% of total volume. Clots were mainly detectable at the apical corner of the oxygenator, most likely due to greater blood stasis. We found a significant increase in d-dimer and in membrane oxygenator shunt and a decrease in platelet count from the start to the discontinuation of ECMO. Hemostatic abnormalities significantly reverted 48 hours after oxygenator removal, suggesting the role of ECMO in activation of the coagulation cascade. Multidetector computed tomographic scan could not accurately confirm the clinical suspicion of MOT.

D-dimers as an early marker for oxygenator exchange in extracorporeal membrane oxygenation.

PURPOSE: Early markers of oxygenator dysfunction during prolonged use of extracorporeal membrane oxygenation (ECMO) are important for timely exchange to avoid sudden loss of function due to clot formation within the system. The measurement of D-dimers (DDs) in plasma might be a marker for early diagnosis of thrombus formation and dysfunction of heparin-coated membrane oxygenators (MOs). METHODS: This is a retrospective study on prospectively collected data of 24 adult acute respiratory distress syndrome patients requiring long-term veno-venous ECMO with at least 1 MO exchange. Kinetics of coagulation, inflammation, and oxygenator function were analyzed before and after MO exchange. RESULTS: Median (interquartile range) support duration is 20 (15-29) days. Thirty-four MOs had to be replaced. Exchange occurred due to visible thrombus formation in the MO (n=16), worsening gas exchange (n=11), increased blood flow resistance (n=1), and activation of coagulation with diffuse bleeding (n=6). In 15 cases, DDs were continuously elevated and, therefore, not suitable as marker for MO exchange. In the remaining 19 cases, DDs increased significantly within 3 days before exchange from 15 (9-20) to 30 (21-35) mg/dL (P=.002) and declined significantly within 1 day thereafter to 13 (7-17) mg/dL (P=.003). CONCLUSIONS: An increase in plasma DD concentration in absence of other explaining pathology can be helpful in predicting an MO exchange in miniaturized heparin-coated ECMO systems.