Racial and ethnical discrepancy in hypoxemia detection in patients on extracorporeal membrane oxygenation

Kalra, Andrew; Shou, Benjamin L.; Zhao, David; Wilcox, Christopher S.; Keller, Steven P.; Whitman, Glenn; Kim, Bo Soo; Cho, Sung Min; Calligy, Kate; Brown, Patricia M.; Alejo, Diane; Anderson, Scott; Acton, Matthew; Rando, Hannah; Chang, Henry

doi:10.1016/j.xjon.2023.02.011

Cited by 9 publications

(6 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In another study applying ML to predict ABI in VV-ECMO, ( 22 )ECMO duration was also a top 3 most important variable for ABI; however, a longer duration of ECMO support was associated with lower risk of ABI in VV-ECMO patients while it was associated with a higher risk in VA-ECMO. As VV-ECMO patients have been shown to be cannulated longer than VA-ECMO patients,( 23 – 25 ) the longer ECMO duration and lower risk of ABI associated may be attributed to the withdrawal of life-sustaining therapy for severely sick patients. ( 26 , 27 ) Accordingly, this may have created a selection bias for patients who did undergo ABI and survived on ECMO support for longer.…”

Section: Discussionmentioning

confidence: 99%

Predicting Acute Brain Injury in Venoarterial Extracorporeal Membrane Oxygenation Patients with Tree-Based Machine Learning: Analysis of the Extracorporeal Life Support Organization Registry

Kalra,

Bachina,

Shou

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Objective: To determine if machine learning (ML) can predict acute brain injury (ABI) and identify modifiable risk factors for ABI in venoarterial extracorporeal membrane oxygenation (VA-ECMO) patients. Design: Retrospective cohort study of the Extracorporeal Life Support Organization (ELSO) Registry (2009-2021). Setting: International, multicenter registry study of 676 ECMO centers. Patients: Adults (≥18 years) supported with VA-ECMO or extracorporeal cardiopulmonary resuscitation (ECPR). Interventions: None. Measurements and Main Results: Our primary outcome was ABI: central nervous system (CNS) ischemia, intracranial hemorrhage (ICH), brain death, and seizures. We utilized Random Forest, CatBoost, LightGBM and XGBoost ML algorithms (10-fold leave-one-out cross-validation) to predict and identify features most important for ABI. We extracted 65 total features: demographics, pre-ECMO/on-ECMO laboratory values, and pre-ECMO/on-ECMO settings. Of 35,855 VA-ECMO (non-ECPR) patients (median age=57.8 years, 66% male), 7.7% (n=2,769) experienced ABI. In VA-ECMO (non-ECPR), the area under the receiver-operator characteristics curves (AUC-ROC) to predict ABI, CNS ischemia, and ICH was 0.67, 0.67, and 0.62, respectively. The true positive, true negative, false positive, false negative, positive, and negative predictive values were 33%, 88%, 12%, 67%, 18%, and 94%, respectively for ABI. Longer ECMO duration, higher 24h ECMO pump flow, and higher on-ECMO PaO2 were associated with ABI. Of 10,775 ECPR patients (median age=57.1 years, 68% male), 16.5% (n=1,787) experienced ABI. The AUC-ROC for ABI, CNS ischemia, and ICH was 0.72, 0.73, and 0.69, respectively. The true positive, true negative, false positive, false negative, positive, and negative predictive values were 61%, 70%, 30%, 39%, 29% and 90%, respectively, for ABI. Longer ECMO duration, younger age, and higher 24h ECMO pump flow were associated with ABI. Conclusions: This is the largest study predicting neurological complications on sufficiently powered international ECMO cohorts. Longer ECMO duration and higher 24h pump flow were associated with ABI in both non-ECPR and ECPR VA-ECMO.

show abstract

Section: Discussionmentioning

confidence: 99%

Predicting Acute Brain Injury in Venoarterial Extracorporeal Membrane Oxygenation Patients with Tree-Based Machine Learning: Analysis of the Extracorporeal Life Support Organization Registry

Kalra,

Bachina,

Shou

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…One explanation is that aggressive oxygen therapy helps mitigate the effects of potential hypoxemia during ECMO support as hypoxemia in ECMO patients is well documented. 44,45 Furthermore, a higher PaO 2 may help increase regional brain oxygen tension 46 and accordingly improve the cerebral metabolic rate of oxygen. We also note we did not "bin" PaO 2 values (continuous variable) by groups as this previous study did which may increase their risk of bias.…”

Section: Discussionmentioning

confidence: 99%

Impact of Pulse Pressure on Acute Brain Injury in Venoarterial ECMO Patients with Cardiogenic Shock During the First 24 Hours of ECMO Cannulation: Analysis of the Extracorporeal Life Support Organization Registry

Kalra,

Kang,

Wilcox

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Background: Low pulse pressure (PP) in venoarterial-extracorporeal membrane oxygenation (VA-ECMO) is a marker of cardiac dysfunction and has been associated with acute brain injury (ABI) as continuous-flow centrifugal pump may lead to endothelial dysregulation. Methods: We retrospectively analyzed adults (≥18 years) on “peripheral” VA-ECMO support for cardiogenic shock in the Extracorporeal Life Support Organization Registry (1/2018-7/2023). Cubic splines were used to establish a threshold (PP≤10 mmHg at 24 hours of ECMO support) for “early low” PP. ABI included central nervous system (CNS) ischemia, intracranial hemorrhage, brain death, and seizures. Multivariable logistic regressions were performed to examine whether PP≤10 mmHg was associated with ABI. Covariates included age, sex, body mass index, pre-ECMO variables (temporary mechanical support, vasopressors, cardiac arrest), on-ECMO variables (pH, PaO2, PaCO2), and on-ECMO complications (hemolysis, arrhythmia, renal replacement therapy). Results: Of 9,807 peripheral VA-ECMO patients (median age=57.4 years, 67% male), 8,294 (85%) had PP>10 mmHg vs. 1,513 (15%) had PP≤10 mmHg. Patients with PP≤10 mmHg experienced ABI more frequently vs. PP>10 mmHg (15% vs. 11%, p<0.001). After adjustment, PP≤10 mmHg was independently associated with ABI (adjusted odds ratio [aOR]=1.25, 95% confidence interval [CI]=1.06-1.48, p=0.01). CNS ischemia and brain death were more common in patients with PP≤10 mmHg vs. PP>10 mmHg (8% vs. 6%, p=0.008; 3% vs. 1%, p<0.001). PP≤10 mmHg was associated with CNS ischemia (aOR=1.26, 95%CI=1.02-1.56, p=0.03) but not intracranial hemorrhage (aOR=1.14, 95%CI=0.85-1.54, p=0.38). Conclusions: Early low PP (≤10 mmHg) at 24 hours of ECMO support was associated with ABI, particularly CNS ischemia, in peripheral VA-ECMO patients.

show abstract

“…Receiver-operating characteristic curve (ROC) analyses were performed to determine the accuracy of on-ECMO SpO 2 in predicting SaO 2 . Based on previous literature, [19][20][21][22] the following on-ECMO SaO 2 thresholds were tested: 88%, 92%, and 95%. All SpO 2 values were tested to detect the speci c SaO 2 threshold at or below its value.…”

Section: Discussionmentioning

confidence: 99%

Characterizing the Racial Discrepancy in Hypoxemia Detection in VV-ECMO: An ELSO Registry Analysis

Kalra,

Wilcox,

Holmes

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Importance: Skin pigmentation influences peripheral oxygen saturation (SpO2) measured by pulse oximetry compared to the arterial saturation of oxygen (SaO2) measured via arterial blood gas analysis. However, data on SpO2-SaO2 discrepancy are limited in venovenous-extracorporeal membrane oxygenation (VV-ECMO) patients. Objective: To determine whether there is racial/ethnical discrepancy between SpO2 and SaO2 in patients receiving VV-ECMO. We hypothesized VV-ECMO cannulation, in addition to race/ethnicity, accentuates the SpO2-SaO2 discrepancy due to significant hemolysis. Design: Retrospective cohort study of the Extracorporeal Life Support Organization Registry from 1/2018-5/2023. Setting: International, multicenter registry study including over 500 ECMO centers. Participants: Adults (≥ 18 years) supported with VV-ECMO with concurrently measured SpO2 and SaO2 measurements. Exposure: Race/ethnicity and ECMO cannulation. Main outcomes and measures: Occult hypoxemia (SaO2 ≤ 88% with SpO2 ≥ 92%) was our primary outcome. Multivariable logistic regressions were performed to examine whether race/ethnicity was associated with occult hypoxemia in pre-ECMO and on-ECMO SpO2-SaO2 calculations. Covariates included age, sex, temporary mechanical circulatory support, pre-vasopressors, and pre-inotropes for pre-ECMO analysis, plus single-lumen versus double-lumen cannulation, hemolysis, hyperbilirubinemia, ECMO pump flow rate, and on-ECMO 24h lactate for on-ECMO analysis. Results: Of 13,171 VV-ECMO patients (median age = 48.6 years, 66% male), there were 7,772 (59%) White, 2,114 (16%) Hispanic, 1,777 (14%) Black, and 1,508 (11%) Asian patients. The frequency of on-ECMO occult hypoxemia was 2.0% (N = 233). Occult hypoxemia was more common in Black and Hispanic versus White patients (3.1% versus 1.7%, P < 0.001 and 2.5% versus 1.7%, P = 0.025, respectively). In multivariable logistic regression, Black patients were at higher risk of pre-ECMO occult hypoxemia versus White patients (adjusted odds ratio [aOR] = 1.55, 95% confidence interval [CI] = 1.18–2.02, P = 0.001). For on-ECMO occult hypoxemia, Black patients (aOR = 1.79, 95%CI = 1.16–2.75, P = 0.008) and Hispanic patients (aOR = 1.71, 95%CI = 1.15–2.55, P = 0.008) had higher risk versus White patients. Furthermore, higher pump flow rate (aOR = 1.29, 95%CI = 1.08–1.55, P = 0.005) and higher on-ECMO 24h lactate (aOR = 1.06, 95%CI = 1.03–1.10, P < 0.001) significantly increased the risk of on-ECMO occult hypoxemia. Conclusions and Relevance: Hispanic and Black VV-ECMO patients experienced occult hypoxemia more than White patients. SaO2 should be carefully monitored during ECMO support for Black and Hispanic patients especially for those with high pump flow and lactate values at risk for occult hypoxemia.

show abstract

Racial and ethnical discrepancy in hypoxemia detection in patients on extracorporeal membrane oxygenation

Cited by 9 publications

References 40 publications

Predicting Acute Brain Injury in Venoarterial Extracorporeal Membrane Oxygenation Patients with Tree-Based Machine Learning: Analysis of the Extracorporeal Life Support Organization Registry

Predicting Acute Brain Injury in Venoarterial Extracorporeal Membrane Oxygenation Patients with Tree-Based Machine Learning: Analysis of the Extracorporeal Life Support Organization Registry

Impact of Pulse Pressure on Acute Brain Injury in Venoarterial ECMO Patients with Cardiogenic Shock During the First 24 Hours of ECMO Cannulation: Analysis of the Extracorporeal Life Support Organization Registry

Characterizing the Racial Discrepancy in Hypoxemia Detection in VV-ECMO: An ELSO Registry Analysis

Contact Info

Product

Resources

About