Visual evaluation of left ventricular performance predicts volume responsiveness early after resuscitation from
Matthias Gruenewald1, Patrick Meybohm, Ole Broch
1Department of Anaesthesiology and Intensive Care Medicine, University Hospital Schleswig-Holstein, Campus Kiel, Schwanenweg 21, D-24119 Kiel, Germany. gruenewald@anaesthesie.uni-kiel.de
Purpose:
Haemodynamic optimisation is a fundamental goal of post-cardiac arrest therapy. Therefore, predicting volume responsiveness is a key issue in therapy of these high-risk patients and transoesophageal echocardiography (TEE) may provide helpful information. The aim of the present study was to evaluate the performance of visual evaluation (eyeballing) of standardised TEE-loops to predict volume responsiveness during post-cardiac arrest period.
Methods:
After approval of the local animal investigation committee, TEE mid-oesophageal long-axis views were recorded before a 5 ml/kg volume bolus at baseline and both 1 and 4h after return of spontaneous circulation (ROSC) from 8 min electrically induced cardiac arrest. Post-hoc, TEE loops were independently presented in randomized order to 7 blinded TEE-experts and 14 blinded TEE novices who were asked to predict whether the ventricle will increase stroke volume ≥ 15% after volume loading or not. Statistics were performed calculating sensitivity and specificity for the correct evaluation and agreement of raters.
Results:
14 out of 20 pigs were successfully resuscitated, and 924 ratings from 21 echocardiographers were included into analysis. Overall, we observed a sensitivity between 71 and 100%, whereas the specificity showed rather low values between 0 and 67% for prediction of volume responsiveness. Best prediction was recorded 1h after ROSC with median sensitivity (95% CI) of 100% (89-100%) and median specificity of 67% (61-72%). No significant difference was found between ratings of experienced and inexperienced echocardiographers. The concordance rate within the two groups was comparable.
Conclusions:
In post-cardiac arrest period, visual evaluation of long-axis TEE loops allows prediction of volume responsiveness with good sensitivity and reasonable specificity even by novice users, and may therefore be suitable for implementation into treatment protocols.
