Observational study of the value of continuous SCVO2 analysis in perioperative management of

John F Fraser, Graeme Hart, Dan Mullany, John Dunning,Kimble Dunster, Janelle Johnstone, David Leishman

Critical Care Research Group, Prince Charles Hospital,Brisbane, QLD

Introduction

Organ dysfunction and multiple organ failure are causes of prolonged hospital stay after cardiac surgery (CS). CS patients are at risk of inadequate perioperative oxygen delivery. Perioperative volume optimisation has been shown to improve outcome in CS patients, whereas inadequate oxygen delivery immediately after CS is associated with prolonged intensive care unit stay. We examined the relationship between central venous oxygen saturation (ScvO2) measured with the PreSep catheter (Edwards LifeSciences, Irvine, Cal) and predetermined clinical measures of haemodynamic adequacy in high-risk CS patients.

Methods

Twenty high-risk CS patients were studied. High-risk CS was defined as presence of any two of the following: age >75 years; valve and coronary artery surgery or double valve surgery; emergency surgery; creatinine concentration >0.2 mmol/L; pulmonary hypertension (mean pulmonary artery pressure >30mmHg); re-do sternotomy; or preoperative intra-aortic balloon pump. The ScvO2 catheter was inserted at the beginning of the operation, and co-oximetry was initiated on the patient’s return to the ICU. All routine haemodynamic data (heart rate, arterial pressure, central venous pressure, SaO2, respiratory rate and ST segment), and ScvO2 were then automatically recorded. Hourly urine output and inotrope requirement, and 4-hourly arterial blood gas analysis were recorded, along with duration of ICU stay and ventilation, and organ dysfunction scores.

Results

ScvO2 and haemodynamic data were obtained in all patients. The study group comprised 15 men and five women, with median age, 75 years (interquartile range, 65–78 years). Mean preoperative creatinine concentration was 164 mmol/L, with mean cross-clamp and bypass time of 105 and 132 minutes, respectively. Mean duration of ventilation and ICU stay were 72 h and 75 h, respectively. There was a trend to prolonged ICU stay and inotrope requirement associated with ScvO2 <65% (not significant). Low ScvO2 predicted cardiac tamponade in two cases.

Conclusions

Continuous ScvO2 measurements give additional useful information with no additional risk in high-risk CS patients. In this small study, this did not reach statistical significance. Intraoperative central venous co-oximetry may assist in haemodynamic anipulations in high-risk patients.