J Extra Corpor Technol 2020, 52, 112-117

Article

Optimal Sweep Gas to Blood Flow Ratio (V/Q) for Initiation of Cardiopulmonary Bypass in a Pediatric Patient Population: A Retrospective Analysis

^* Department of Cardiovascular Perfusion, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
^† Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, Ohio; and
^‡ Division of Biostatistics and Epidemiology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio

Address correspondence to: Sean Clingan, MS, CCP, The Heart Institute Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue, MLC 2004, Cincinnati, OH 45229. E-mail: sean.clingan@cchmc.org

Received: 4 December 2019
Accepted: 30 March 2020

Abstract

The optimal setting to achieve a suitable PaCO₂ value of 35–45 mmHg upon initiating cardiopulmonary bypass (CPB) in the pediatric population is undefined in the literature. Sweep gas is set upon initiating and modified throughout CPB to reduce potential complications related to compensatory metabolic acidosis or metabolic alkalosis and associated cerebral blood flow fluctuations. This study retrospectively examined 1,077 CPB cases for which PaCO₂ values were no less than 30 mmHg and no greater than 50 mmHg on the pre-CPB blood gas result. Through an observation of the results, we attempted to determine the optimal sweep gas setting upon initiating CPB to obtain a physiologic PaCO₂ value of 35–45 mmHg. The probability of achieving an optimal PaCO₂ value was modeled as a function of the average sweep gas to blood flow ratio during the period before the first blood gas on CPB. The median sweep gas to blood flow ratio (V/Q) was .64 (.51; .76), with a median first PaCO₂ value on CPB of 42 mmHg (38.8; 45). A .6 V/Q had an odds ratio (OR) of 1.57 of obtaining a PaCO₂ value between 35 and 45 mmHg on the first CPB blood gas when compared with a .4 V/Q (Figure 1Figure 1.

Bivariate associations between PaCO₂ and the V/Q ratio. (A) Spike histogram with loess curve showing the proportion of patients with a first PaCO₂ value on CPB between 35 and 45 mmHg according to the V/Q ratio. (B) Scatterplot and loess curve (gray line) for PaCO₂ on CPB according to the V/Q ratio. Dashed lines indicate the target range of 35–45 mmHg. (C) Model-based estimate of the predicted probability and 95% CI for PaCO₂ on CPB between 35 and 45 mmHg according to the V/Q ratio obtained from logistic regression. (D) Model-based estimate of the predicted PaCO₂ on CPB according to the V/Q ratio obtained from ordinal regression. Prop, proportion.

). A .9 V/Q had a 1.76 OR when compared with a .4 and a 1.12 OR when compared with .6. Using a .6 V/Q ratio achieved a PaCO₂ value within normal physiologic limits with no significant advantage to a higher V/Q ratio overall. However, younger or smaller patients required a higher V/Q to achieve similar probabilities of being within limits and similar PaCO₂ values when compared with the older or larger patients.