Hemorrhagic complications during extracorporeal membrane oxygenation: prevention and treatment

Hemorrhage related to systemic heparinization is the major complication of extracorporeal membrane oxygenation (ECMO). Intracranial hemorrhage (ICH) is the most devastating complication. ICH developed in 13 of our 25 ECMO patients (52%). Six died, six survived with normal neurologic function, and one is severely impaired. In nine of 13 patients (69%) ECMO was discontinued when serial cranial ultrasounds showed progressive ICH. Seizures developed in six infants while receiving ECMO, and ICH developed in all. There is a correlation between hypertension and ICH. A hypertension index (hours systolic BP greater than 90/hours receiving ECMO) was 0.1 +/- 0.12 for infants without ICH and 0.37 +/- 0.28 for infants with ICH (P less than .05). ICH developed in 79% of the patients with an index greater than 0.1. Twenty neck explorations were required in the first 20 patients for incisional bleeding (mean blood loss, 21.9 +/- 18.0 mL/kg/d). We now use fibrin glue following cannulation and have done only one neck exploration in the last five patients (mean blood loss, 2.8 +/- 2.2 mL/kg/d, P less than .05). Endobronchial bleeding has responded to phenylephrine lavage and increased positive end-expiratory pressure. We have controlled pleural space bleeding with topical thrombin. None of the hemorrhagic complications encountered correlate with the activated clotting time or the amount of heparin used. There is an increased risk of hemorrhage associated with platelet counts less than 100,000/microL for 75% of a day (P less than .05) so that aggressive platelet transfusion remains important in preventing hemorrhagic complications during ECMO.     

Nesiritide during extracorporeal membrane oxygenation

Nesiritide is a recombinant formulation of B-type natriuretic peptide (BNP). Preliminary experience in the adult population has shown nesiritide to be an effective agent in the treatment of decompensated congestive heart failure (CHF) in adults. Given its physiological effects, it may be an effective agent in other clinical scenarios. We report the use of nesiritide in two infants during extracorporeal membrane oxygenation (ECMO). In one patient, nesiritide in doses up to 0.09 microg.kg(-1).min(-1) were used to control mean arterial pressure while in the other patient, doses of 0.01-0.03 microg.kg(-1).min(-1) were used to augment urine output. The potential applications of nesiritide and dosing regimens for this agent in the ECMO population are discussed.     

Mediastinal hemorrhage during extracorporeal membrane oxygenation

A case of mediastinal hemorrhage along with hemorrhage into a pneumatocele while on extracorporeal membrane oxygenation (ECMO) is presented. Computerized tomography of the chest was utilized to support the diagnosis. Barotrauma to the lungs best explains the inciting event that allowed the hemorrhage to occur once the patient was heparinized for ECMO. This complication serves to point out the importance of commencing early ECMO support before widespread pulmonary and mediastinal barotrauma develops.     

Tension pneumothorax during extracorporeal membrane oxygenation

Life-threatening tension pneumothorax in neonates on extracorporeal membrane oxygenation (ECMO) has been associated with an increase in arterial oxygen tension and a decrease in peripheral perfusion, followed by a decrease in ECMO flow with progressive hemodynamic deterioration. To investigate this triad, chest tubes were placed bilaterally in 9 dogs to allow injection of air to produce tension pneumothorax. Six dogs were subsequently placed on standard venoarterial ECMO before the reinduction of tension pneumothorax. Measured values included arterial pulse pressure, inferior vena cava pressure, systemic arterial blood gases, peripheral arterial oxygen saturation, mixed venous oxygen saturation, and left heart cardiac output. Oxygen delivery was calculated from directly measured values. Each of the 6 dogs on ECMO demonstrated the triad of increased arterial oxygen tension (92 +/- 7 to 325 +/- 20 mm Hg; p less than 0.05), decreased peripheral perfusion (as evidenced by a decrease in pulse pressure from 55 +/- 4 to 31 +/- 5 mm Hg; p less than 0.05), and decreased mixed venous oxygen saturation (71% +/- 3% to 22% +/- 2% saturation; p less than 0.05) followed by a lower ECMO flow with progressive hemodynamic deterioration (oxygen delivery decreased from 285 +/- 11 to 111 +/- 12 mL/min; p less than 0.05). Aspiration of the intrathoracic air allowed return to baseline ECMO flow and hemodynamic stability in all dogs. The triad of increased arterial oxygen tension and decreased peripheral perfusion (as evidenced by a lower arterial pulse pressure and lower mixed venous oxygen saturation) followed by decreased ECMO flow with progressive hemodynamic deterioration consistently appears when tension pneumothorax occurs on ECMO.     

Stunned myocardium during extracorporeal membrane oxygenation

The "stunned myocardium" is a syndrome of reversible myocardial dysfunction that may be mediated by oxygen-derived free radicals. This phenomenon has been seen in some neonates undergoing extracorporeal membrane oxygenation. We performed echocardiograms and measured creatine phosphokinase isoenzymes and lipid peroxide levels in 16 neonates before, during, and after extracorporeal membrane oxygenation. Infants who developed stunned myocardia by echocardiography did so shortly after initiation of bypass and exhibited concurrent elevations of the MB fraction of creatine phosphokinase. Lipid peroxide levels did not simultaneously rise. These data suggest that oxygen-derived free radicals may not cause the stunned myocardium seen in neonates undergoing extracorporeal membrane oxygenation.     

Antithrombin replacement during extracorporeal membrane oxygenation

Heparin remains the predominant anticoagulant during extracorporeal membrane oxygenation (ECMO). Heparin acts by potentiating the anticoagulant effect of antithrombin (ATIII). Acquired ATIII deficiency, common in pediatric patients requiring ECMO, may result in ineffective anticoagulation with heparin. ATIII replacement may result in increased bleeding. Our objective is to determine ATIII's effect on anticoagulation and blood loss during ECMO. A retrospective chart review was performed of all patients at Children's Hospital of Wisconsin who received ATIII while supported on ECMO in 2009. ATIII activity levels, heparin drip rate, and activated clotting times (ACT) were compared before, 4, 8, and 24 h after ATIII administration. Chest tube output and packed red blood cell (pRBC) transfusion volume were compared from 24 h before ATIII administration to 24 h after. Twenty-eight patients received ATIII as a bolus dose during the course of 31 separate times on ECMO support. The median age of these patients was 0.3 years (range 1 day-19.5 years). ATIII activity increased significantly at 8 and 24 h after administration. No significant difference was noted in heparin drip rate, ACT levels, chest tube output, or pRBC transfusion volume. ATIII administration resulted in higher ATIII activity levels for 24 h without a significant effect on heparin dose, ACT, or measures of bleeding.     

Experience with renal failure during extracorporeal membrane oxygenation: treatment with continuous hemofiltration

We use extracorporeal membrane oxygenation (ECMO) to treat respiratory and cardiac failure in children who are unresponsive to standard ventilator and pharmacologic management. All patients have cardiac and abdominal ultrasonography prior to ECMO to identify major structural anomalies and anatomically normal kidneys. Despite this, oliguric renal failure is seen in a number of patients. Acute renal failure (ARF) developed in two of the first 20 patients we placed on ECMO and both of these patients died. Six of the last 27 patients (22%) also developed ARF and were treated with continuous hemofiltration (CH) placed in-line with the extracorporeal circuit. The technique of CH removes plasma water and dissolved solutes while retaining proteins and cellular components of the intravascular space. The duration of CH ranged from 9 to 112 hours (mean 57.5 hours). Indications for CH were hypervolemia, hyperkalemia, and azotemia. The mean serum potassium prior to CH was 5.6 (range 4.3 to 7.0) compared with 4.5 after filtration. We filtered 5 to 10 mL/kg/h and replaced it with crystalloid chosen on the basis of serum and filtrate electrolytes. These six patients had a 33% mean weight gain prior to CH. We were able to remove as much as 2,200 g in the most edematous patient with significant improvement in cardiopulmonary status. Four of the patients on CH died of their primary pulmonary or cardiac disease without specific problems related to ARF. The other two patients were successfully weaned from ECMO, extubated, and have not needed further therapy for renal failure. We conclude that CH is useful in managing the complications of oliguric renal failure during ECMO.     

婴幼儿颈部通路体外膜肺氧合相关性出血并发症分析

目的研究婴幼儿颈部通路体外膜肺氧合（ECMO）所致出血并发症及转归。 方法回顾性分析2009年9月至2021年1月期间在中山市人民医院行颈动静脉插管ECMO治疗的21例婴幼儿临床资料。观察ECMO运行期间和撤机后血小板、凝血功能和出血并发症情况。 结果ECMO建立后血小板计数明显下降,活化部分促凝血酶原激活时间（APTT）和活化全血凝固时间（ACT）快速升高,差异均有统计学意义（P<0.05）。出血并发症发生率为61.90%（13/21）,包括颈部切口出血8例（38.10%）、肺出血7例（33.33%）、消化道出血5例（23.81%）、颅内出血3例（14.29%）、泌尿系统出血2例（9.52%）。出血组死亡率为38.46%（5/13）,无出血组死亡率为25.00%（2/8）。 结论婴幼儿颈部通路ECMO相关出血并发症较多。颅内出血在上机期间影响患儿生存,但撤机后基本可以恢复正常,无严重后遗症。     

Complement activation during prolonged extracorporeal membrane oxygenation

Short-term cardiopulmonary bypass activates the complement system, possibly resulting in pulmonary dysfunction from granulocyte aggregation and pulmonary endothelial damage. These effects may be inhibited by steroids. Prolonged extracorporeal membrane oxygenation (ECMO) is used for newborn respiratory failure, but the effects of ECMO on complement activation are unknown. Twenty-one newborn infants with respiratory failure treated with ECMO were randomly assigned to group I (control, no steroids) or group II (30 mg/kg intravenous methylprednisolone before ECMO). Depletion assays of C3 and C5 were performed in each group at intervals before and during ECMO (declining values indicate complement activation). The groups were compared for complement levels, survival, time on ECMO and on the ventilator, and total hospitalization time. Steroids significantly shortened the time on ECMO and time on the ventilator after ECMO but did not affect survival or total hospitalization time. Steroids also enhanced activation of C3 and C5. Complement activation occurs during ECMO. Steroid administration paradoxically causes earlier complement activation but shortens ECMO and ventilator times. Complement activation during ECMO is of questionable significance. The benefits of steroids during ECMO may be mediated through other mechanisms.     

Interventional treatment of bleeding complications due to percutaneous cannulation for peripheral extracorporeal membrane oxygenation

PurposeThis study aimed to evaluate the safety and effectiveness of interventional techniques as a treatment for bleeding complications secondary to percutaneous cannulation for peripheral extracorporeal membrane oxygenation (PECMO).Materials and methodsOut of 1264 patients who underwent PECMO at our hospital between January 2009 and September 2018, we reviewed the clinical characteristics and outcomes of eight patients (4 men, 4 women; mean age, 54.9 years [range, 31–77 years]) who underwent percutaneous interventional treatment for bleeding complications secondary to percutaneous cannulation for PECMO.ResultsBoth hemodynamic instability and coagulopathy were present in 7 patients who had direct injury during PECMO insertion and absent in one patient with pseudoaneurysm at the PECMO removal site. Percutaneous ultrasound-guided thrombin injection was performed in three patients with pseudoaneurysm of the common or superficial femoral artery, and adjunctive embolization was combined with microcoils or n-butyl 2-cyanoacrylate in two of them. Stent graft was inserted in four patients with contrast extravasation (n = 3) from external iliac artery (n = 1) or common femoral or iliac veins (n = 2) ruptures or the fistula between the superficial femoral artery and vein (n = 1). N-butyl 2-cyanoacrylate and coil embolization was performed for pseudoaneurysm from the internal pudendal artery branch in the remaining one patient. Technical success was achieved in all eight patients. There were no procedure-related complications. There was no rebleeding during the follow-up.ConclusionInterventional treatment is a safe, technically feasible and therapeutically effective modality for treating bleeding complications secondary to a percutaneous cannulation for PECMO.     

Venovenous extracorporeal membrane oxygenation during high-risk airway interventions

 Open in a separate windowOBJECTIVESPractice patterns for the use of extracorporeal membrane oxygenation (ECMO) during high-risk airway interventions vary, and data are limited. We aim to characterize our recent experience using ECMO for procedural support during whole-lung lavage (WLL) and high-risk bronchoscopy for central airway obstruction (CAO).METHODSWe performed a retrospective cohort study of adults who received ECMO during WLL and high-risk bronchoscopy from 1 July 2018 to 30 March 2020. Our primary end point was successful completion of the intervention. Secondary end points included ECMO-associated complications and hospital survival.RESULTSEight patients received venovenous ECMO for respiratory support during 9 interventions; 3 WLLs for pulmonary alveolar proteinosis were performed in 2 patients, and 6 patients underwent 6 bronchoscopic interventions for CAO. We initiated ECMO prior to the intervention in 8 cases and during the intervention in 1 case for respiratory decompensation. All 9 interventions were successfully completed. Median ECMO duration was 17.8 h (interquartile range, 15.9–26.6) for the pulmonary alveolar proteinosis group and 1.9 h (interquartile range, 1.4–8.1) for the CAO group. There was 1 cannula-associated deep vein thrombosis; there were no other ECMO complications. Seven patients (87.5%) and 4 (50.0%) patients survived to discharge and 1 year postintervention, respectively.CONCLUSIONSUse of venovenous ECMO to facilitate high-risk airway interventions is safe and feasible. Planned preprocedural ECMO initiation may prevent avoidable respiratory emergencies and extend therapeutic airway interventions to patients otherwise considered too high-risk to treat. Guidelines are needed to inform the utilization of ECMO during high-risk bronchoscopy and other airway interventions.     

Major surgical intervention during extracorporeal membrane oxygenation.

Of 135 patients treated with extracorporeal membrane oxygenation (ECMO) between January 1987 and December 1989, 19 (14.0%) patients underwent surgical procedures while on ECMO. Thirteen (68%) patients had operations related to hemorrhage, including cannula site (6), mediastinal (1), hemoperitoneum (3), and hemothorax (3). Six of 13 patients required repetitive operations for bleeding; 4 of 6 died. Six (35%) patients had operations for congenital pathology including patent ductus (PDA) ligation (2), repair of transposition of the great vessels (2), repair of coarctation (1), and repair of congenital diaphragmatic hernia (3). One patient had multiple simultaneous procedures performed. Of these 6 patients, 4 were decannulated immediately and 2 were decannulated within 28 hours following surgery without any bleeding complications. Fifteen of 19 patients were operated on in the neonatal intensive care unit. The 4 remaining patients required transport on ECMO to the surgical suite. Thirteen of the 19 patients requiring surgical intervention on ECMO survived. In the 13 survivors, the mean time to decannulation postoperative was 45 hours, and in those that died it was 90 hours. Our experience suggests that surgical intervention while on ECMO is technically feasible with the best results achieved when rapid discontinuation of ECMO can be accomplished postoperatively. Due to this fact major surgical intervention should be postponed if possible until near the conclusion of the ECMO therapy.     

Severe myocardial dysfunction during extracorporeal membrane oxygenation.

Of the 102 neonates with respiratory failure supported with extracorporeal membrane oxygenation (ECMO) at this institution between 1984 and 1987, 8 patients developed severe myocardial dysfunction that was noted shortly after onset of bypass. The neonates in the cardiac dysfunction group were more hypoxic (average PaO2 = 26 +/- 8 mm Hg v 41 +/- 19 mm Hg, P less than .01) in the immediate pre-ECMO period. Seventy-five percent were unstable hemodynamically (6 hypotensive, 3 bradycardic, 2 sustained cardiac arrest, 4 required epinephrine pressor support). On ECMO, 5 of the 8 neonates developed an ischemic cardiomyopathy that lasted for less than 24 hours and resolved without therapeutic intervention. In the other 3 cases, prolonged periods of dysfunction were noted and afterload reduction through administration of tolazoline or hydralazine was beneficial. These 8 patients serve to demonstrate the reversible nature of postischemic cardiac dysfunction in patients on ECMO and in the neonatal population in general.     

体外膜式氧合相关并发症分析

目的 分析体外膜式氧合(ECMO)辅助过程中相关并发症情况,以期对提高ECMO辅助抢救成功率.方法 回顾2005年3月至2008年6月117例接受ECMO辅助者的临床资料,其中静脉-静脉转流2例,静脉-升主动脉转流5例,股静脉-股动脉转流110例.结果 ECMO平均辅助时间61h.死亡48例,病死率41.0%.74例治疗过程中发生各种并发症,发生率为63.2%.主要并发症为感染32例次、肾功能衰竭需要透析29例次、氧合器血浆渗漏29例次、二次开胸止血24例次潲化道出血14例次、溶血7例次、肢体血栓5例次、神经系统并发症4例次、离心泵故障1例次.结论 出血是ECMO早期最常见的并发症,随辅助时间延长,感染、肾功能衰竭及氧合器血浆渗漏等并发症明显增加.积极预防、治疗并发症对提高ECMO病人抢救成功率非常重要.