Evans et al.(16) (2017), USA Perfusion Qualis: B2 FI: 0.65 |
Cross-Sectional /High |
41 oxygenators were analyzed after treatment of 27 patients with cardiogenic shock and acute myocardial infarction / VV, VA ECMO |
To quantify thrombus volume in the oxygenator and correlate it with demographic data, flow characteristics, and anticoagulation parameters |
The mean volume of thrombi in oxygenators was 11.4 cm3. Increase of 1L/min in the flow (p=0.038); VA / VV ECMO modality (p=0.026); each 1 cm3 of visible thrombus (p<0.001), R2 Coefficient adjusted from linear regression=0.39. |
The combination of median flow (L/min), VA modality of ECMO, and visible increase of thrombi in the oxygenator are predictors of internal volume in the thrombus. |
Lorusso et al.(17) (2016), Holland / Crit Care Med Qualis: A1 FI: 3.88 |
Retrospective Cohort / High |
4522 adults with cardiac, respiratory and CPR indications. Data collected from 230 ECMO centers from ELSO / VA ECMO records |
To identify predictors of neurological complications |
Neurologic complications occurred in 15.1% of patients. Predictors in respiratory indication: pre-ECMO CPR [OR: 2.44; 95% CI (1.48-4.02);p<0.0001]; hemolysis (plasma free Hb>50mg/dL) [OR:2.1; 95%IC(1.12-4.02);p=0.0001]; inotropes in ECMO [OR:1.74; 95%IC(1.06-2.84;p=0,027]. Predictors in cardiac indication: age [OR:0.99;95%IC(0.98-0.99); p=0.001]; pre-ECMO CPR (OR:2.34;95% CI(1.84-2.97); p<0.0001; inotropes in ECMO (OR:1.49; 95% IC (1.13-1.96);p=0.005; tamponade (OR: 1.73; 95%CI (1.14-2.64) p=0.01; DIC (OR: 1.70;9 5%IC (1.06-2.72); p=0.026); hypoglycemia (OR:2.50; 95% CI (1.42-4.40); p=0.001; Cr>3mg/dL (OR:1.77; 95% CI(1.32-2.37); p<0.0001; 1.5mg/dL<Cr<3mg/dL (OR:1.49; 95% CI(1.14-1.96); p=0.004. Predictors in E-CPR: age (OR:0.99; 95%CI(0.98-0.99); p=0.039; hypoglycemia (OR:4.81; 95%CI(1.46-15.87);p=0.010; need for hemodialysis (OR:2.01; 95%CI(1.29-3.13), p=0.002; Cr>3mg/dL (OR:1.66; 95%CI(1.13-2.42);p=0.009); 1.5mg/dL<Cr<3mg/dL (OR:2.19; 95%CI(1.49-3.00), p<0.0001. |
Predictors in respiratory indication: pre-ECMO CPR, plasma free Hb> 50mg / dL, use of inotropes during ECMO. Predictors in cardiac indication: age (39-53 years), pre-ECMO CPR, use of inotropes ECMO,ˌ tamponade, DIC, glucose <40mg/dL, Cr>3mg/dL and 1.5mg/dL<Cr<3mg/dL. Predictors in the E-CPR: age (39-53 years), glucose <40mg/dL, need of hemodialysis, 1.5mg/dL<Cr<3mg/dL. |
Lorusso et al.(18) (2017), USA/ Crit Care Med Qualis: A1 FI: 3.88 |
Retrospective Cohort / High |
4988 adults on IR. Data collected from 350 ECMO centers from ELSO / ECMO VV registries |
Investigate factors associated with neurological complications |
Neurological injuries occurred in 7.1% of cases, such as intracranial hemorrhage, brain death, ischemic stroke, and seizure pre-ECMO CPR [OR:3.12; 95%CI(1.78-5.46); p<0.001]; Hyperbilirubinemia [OR:2.37; 95%CI(1.44-3.88); p=0.001]; Hemodialysis [OR:2.33; 95%CI(1.28-4.42); p=0.006]. |
Pre-ECMO CPR, dialysis, and hyperbilirubinemia during ECMO are predictors of neurological complications. |
Abrams et al.(19) (2016) USA/ Intensive Care Med Qualis: A1 FI: 4.92 |
Retrospective Cohort / High |
A hundred adults in severe AKI/ VV, VA ECMO |
To verify the relationship between ECMO duration and other clinical characteristics during cannulation with the development of thrombocytopenia |
APACHE II score (increase of 5 points) [OR: 1.35; 95% CI (0.94-1.94)], platelet count at cannulation <188,000 / µL (decrease of 25,000 / µL) [OR: 1.35; 95% CI (1.10-1.64)]. |
Lower platelet count in cannulation and higher APACHE II score are predictors of severe thrombocytopenia. |
Kim et al.(20) (2017), Korea/ J Korean Med Sci Qualis: S/A FI: 1.18 |
Retrospective Cohort / High |
61 adults in cardiogenic shock / VA ECMO |
To investigate the risk factors for nosocomial infection |
18 infections occurred (23.0%) in 14 individuals, with bloodstream infection prevalence Preoperative Cr [OR: 2.17, 95% CI (1.06-4.44), p = 0.033]; Time in ECMO [OR: 1.40; 95% CI (1.08-1.81); p = 0.011]. |
Higher serum levels of preoperative Cr (mg / dL) and longer ECMO (days) were predictors of infection. |
Omar et al.(21) (2016), USA/ J Crit Care Qualis: B1 FI: 2.48 |
Retrospective Cohort / High |
171 adults in cardiogenic shock, post-cardiotomy, CPR, post-tx, massive pulmonary embolism, AKI/ VV, VA ECMO |
To investigate the predictors of ischemic stroke |
Ten patients developed ischemic stroke (5.8%) while on ECMO. Pre-ECMO lactic acid> 10 mmol / L [OR: 7.58; 95% CI (1.39-41.22); p = 0.019). |
High serum level of pre-ECMO lactic acid was a predictor of ischemic stroke. |
Ryu et al.(22) (2015), South Korea/ BMC Anesthesiol Qualis: B2 FI: 1.78 |
Retrospective Cohort / High |
115 adults survived to E-CPR / VA ECMO |
To investigate the predictors of neurological outcomes |
41% had poor neurological outcomes and 24 evolved to brain death. Dependent variable on multivariate analysis = good neurological outcomes: Pre-ECMO lactic acid [OR: 0.76; 95% CI (0.66-0.88), p <0.001]; pre-ECMO Hb [OR: 1.50; 95% CI (1.07-2.10), p = 0.019]; interval from CA to ECMO setup [OR: 0.96; 95% CI (0.92-0.99) p = 0.042]. |
The highest pre-ECMO serum lactic acid level, the lowest pre-ECMO serum Hb level, and a longer CA interval before ECMO was established were predictors of poor neurological outcomes. |
Arachchillage et al. (23) (2018), UK/Semin Thromb Hemost Qualis: A2 FI: 3.12 |
Retrospective Cohort / High |
149 adults in severe AKI / VV ECMO |
To identify clinical and laboratory variables that predict intracranial hemorrhage |
The prevalence and incidence of intracranial bleeding were 10.7% and 5.2%, respectively. Thrombocytopenia [OR: 22.6; 95% CI (2.6-99.5), p = 0.001]; creatinine clearance [OR: 10.8; 95% CI (5.6-16.2), p <0.0001]. |
Thrombocytopenia and reduced creatinine clearance were predictors of intracranial hemorrhage. |
Aubron et al.(24) (2016), France/ Ann Intensive Care Qualis: A1 FI: 4.82 |
Retrospective Cohort / High |
147 adults with cardiovascular problems, post or pre-tx, with pneumonia, in CA, post cardiac surgery / VA , VV ECMO |
To identify the risk factors for bleeding |
The most common bleeds were: cannulation site (37%); hemothorax or cardiac tamponade (17%) aPTT ≥ 70sec on the previous day (OR: 3.0, 95% CI (1.64-5.47), p = <0.01), APACHE III score (OR: 1.01, 95% CI (1: 1, 95% CI: 1.64-5.47), p = <0.01) (P = 0.01), ECMO post-surgery [OR: 3.04, 95% CI (1.62-5.69), p <0.01). |
Elevated aPTT, high APACHE III score, and ECMO postoperatively were predictors of hemorrhagic complications. |
Austin et al.(25) (2017), Australia/ Crit Care Resusc Qualis: A2 FI: 2.01 |
Retrospective Cohort / High |
98 adults with primary graft dysfunction, HF, AMI, respiratory tract infection, severe asthma / VV, VA ECMO |
To evaluate the risk factors for infections |
Twenty-one (21.4%) patients presented infection: 8 developed bloodstream infection, 14 infections at the cannulation site, and two infections at the sternum. Immunosuppression [OR: 2.9; p = 0.04]; VA ECMO [OR: 14.7; p = 0.01]. |
Immunosuppression and treatment with VA ECMO were predictors of infection. |
Chang et al.(26) (2017), China/ Int J Clin Exp Med Qualis: B1 FI: 0.83 |
Retrospective Cohort / High |
71 adults with ARDS, post-cardiotomy or other cardiovascular or pulmonary problems / VV, VA ECMO |
Determine the risk factors for AKI |
Approximately 73% developed acute kidney injury. Length of ICU stay (<20 days vs. 20 days) [RR: 0.32; 95% CI (0.14-0.73); p <0.007]; infection [RR: 2.28; 95% CI (1.06-4.87); p <0.034]. |
Length of ICU stay and infection were predictors of AKI. |
Sandersjöo et al.(27) (2017), Switzerland/ J Intensive Care Qualis: S/A FI: S/A |
Retrospective Cohort / High |
253 adults with cardiac or pulmonary indication or CA/ VV, VA ECMO |
To identify predictors of intracranial hemorrhage |
About 20% developed intracranial hemorrhage, with a mortality rate of 81% within one month. Antithrombotic therapy [p = 0.011; R2 = 0.037]; platelet count [p = 0.035; R2 = 0.074]. |
Pre-admission antithrombotic therapy and low platelet count are predictors of intracranial hemorrhage. |
Hoshino H et al.(28) (2018), Japan/ J Artif Organs Qualis: S/A FI: 0.61 |
Retrospective Cohort / High |
10 adults with acute RF / VV ECMO |
To identify the coagulation / fibrinolysis markers as predictors for ECMO circuit replacement |
Six circuit replacements were necessary. Soluble fibrin (10µg / mL) [OR: 1.20; 95% CI (10.6-1.36); p <0.01] |
Soluble fibrin is a predictor for ECMO circuit exchange. |
Lotz et al.(29)(2017), Germany/ ASAIO J Qualis: S/A FI: 0.55 |
Retrospective Cohort / High |
59 adults with RF or HF / VV, VA ECMO |
To identify risk factors for bleeding during ECMO |
Bleeding occurred in 60% of the patients in VA ECMO and in 80% in VV ECMO. Fungal pneumonia [RR: 4.38; 95% CI (1.15-16.71); p = 0.031]. |
Only fungal pneumonia remained a predictor of bleeding requiring therapeutic intervention. |
Luyt et al.(30) (2016) France/ Intensive Care Med Qualis: A1 FI: 4.92 |
Retrospective Cohort / High |
135 adults with indication for VV ECMO, but with a primary diagnosis of ARDS / VV ECMO |
To investigate the risk factors for neurological complications |
18 (13.3%) patients developed neurological complications; the most common was intracranial bleeding (10 individuals). Kidney insufficiency [RR: 6.13; 95% CI (1.29-28.57)]; PaCO2 <-27mmHg [RR: 6.02; 95% CI (1.28-28.57)]. |
Kidney insufficiency at ICU admission and low pre-ECMO PaCO2 were predictors of intracranial bleeding. |
Lyu et al.(31) (2015), China/ J Cardiothorac Vasc Anesth Qualis: B2 FI: 1.57 |
Retrospective Cohort / High |
84 adults post-cardiac, post-cardiotomy and in HF / VA ECMO |
To investigate whether the increased serum of plasma free hemoglobin level is associated with AKI |
The incidence of AKI was 48.8%, and 41.7% evolved to hemodialysis. Free hemoglobin [OR: 1.05; 95% CI (1.01-1.08); p = 0.005]. |
Free plasma hemoglobin increased during ECMO was a predictor for AKI. |
Otani et al.(32) (2017), Japan/ Am J Emerg Med Qualis: B2 FI:1.31 |
Retrospective Cohort / High |
102 adults admitted after extra-hospital cardiac arrest and treated with E-CPR / VA ECMO |
To verify the frequency of bleeding complications and determine their related factors |
70% had some type of bleeding, and the VA ECMO puncture site and the gastrointestinal tract were the most frequent bleeding events. Age (increase of one year) [OR: 1.053; 95% CI (1.00-1.10); p = 0.018]; platelet count (103 / µL increase) [OR: 0.984; 95% CI (0.97-0.99), p = 0.014]; D-dimer (1 µg / mL increase) [OR: 1.066; 95% CI (1.01-1.11); p = 0.006]. |
Older age, lower platelet count, and higher serum D-dimer level at admission were predictors of bleeding complications. |
Salna et al.(33) (2017), USA/ J Vasc Surg Qualis: A2 FI:1.40 |
Retrospective Cohort / High |
192 adults in refractory cardiogenic shock / femoral VA ECMO |
To evaluate the incidence and risk factors associated with lymphocele formation |
Lymphocele formation was identified in 16% of individuals. Primary dysfunction of cardiac graft [OR: 8.66; 95% CI (3.38-22.16); p <0.001]. |
Primary dysfunction of cardiac graft was a predictor of lymphocele formation. |
Trudzinski et al.(34) (2016), Germany/ Ann Intensive Care Qualis: A1 FI: 4.82 |
Retrospective Cohort / High |
102 adults in acute RF / VV ECMO |
To analyze the incidence and predictive factors of thromboembolism |
The highest incidence of thrombosis was related to cannulation and the incidence of pulmonary embolism was 11.1%. Time on ECMO [OR: 1.04; 95% CI (1.00-1.09); p = 0.026]; aPTT> 50s [OR: 0.97; 95% CI (0.95-0.99); p = 0.024]. |
Longer time on ECMO and higher aPTT were predictors of thrombosis and pulmonary thromboembolism. |