Evaluation of Transcutaneous Non-Invasive Blood Gas Analysis for Monitoring Gas Exchange in Pediatric Cardiac Surgical Patients Post Extubation

TL;DR

This study evaluates non-invasive transcutaneous blood gas monitoring in young children after heart surgery, finding it reliable for carbon dioxide but not oxygen levels.

Contribution

The study demonstrates that transcutaneous CO2 monitoring can reliably replace arterial measurements in post-cardiac surgery pediatric patients.

Findings

Transcutaneous CO2 (TcPCO2) strongly correlates with arterial CO2 (PaCO2) with an r-value of 0.95.

Transcutaneous O2 (TcPO2) shows moderate correlation with arterial O2 (PaO2) with an r-value of 0.8942.

Bland-Altman analysis indicates TcPCO2 is interchangeable with PaCO2 but TcPO2 has wider limits of agreement.

Abstract

Pediatric cardiac surgery patients need close post-extubation monitoring for
ventilation. Non-invasive transcutaneous partial pressure of oxygen
(TcPO2) and transcutaneous partial pressure of carbon dioxide
(TcPCO2) offer continuous insights and in improving care.

To investigate the correlation of transcutaneous blood gases
(TcPO2, TcPCO2) with arterial blood gases
i.e. arterial partial pressure of oxygen
(PaO2) and arterial partial pressure of carbon dioxide
(PaCO2).

We conducted a study on 30 pediatric post-cardiac surgery patients (four
months to three years old) who were extubated and exhibited stable
hemodynamics (inotropic score ≤ 5), normal sinus rhythm, and no
respiratory or heart failure signs. Continuous transcutaneous and
intermittent arterial blood gas monitoring started one hour after
extubation, with recordings every 30 minutes for four hours. A single
observer conducted probe calibration and data recording to minimize
variability, while analysis of 240 paired samples included correlation
coefficient, linear regression, Bland-Altman analysis, and Mountain
plot.

The r-value between PaCO2 and TcPCO2 was 0.95,
r2-value of 0.9060
(P<0.001). Bland-Altman showed a bias of 2.579, and 95%
limits of agreement were -6.4 to 1.3. The r-value between
PaO2 and TcPO2 was 0.8942,
r2-value of 0.7996
(P<0.001); bias of 20.171 and 95% limit of agreement of
-0.5 to 40.9. The Mountain plot revealed a median of 2.57 for
PaCO2
vs. TcPCO2 and 20.17 for PaO2
vs. TcPO2.

Transcutaneous carbon dioxide values are interchangeable with arterial
PaCO2 in our population study, acting as a surrogate in
postoperative pediatric cardiac surgery. Confirmation with arterial blood
gases is needed if discrepancies occur.

Figures