Preoperative cerebral oxygenation in high-risk noncardiac surgical patients: an observational study on postoperative mortality and complications

Abstract

Near Infrared Spectroscopy (NIRS) has become widely accepted to evaluate regional cerebral oxygen saturation (rScO2), potentially acting as a surrogate parameter of reduced cerebral oxygen delivery or increased consumption. Low preoperative rScO2 is associated with increased postoperative complications after cardiac surgery. However, its universal potential in pre-anesthesia risk assessment remains unclear. Therefore, we investigated whether low preoperative rScO2 is indicative of postoperative complications and associated with poor outcomes in noncardiac surgical patients. We prospectively enrolled 130 patients undergoing high-risk noncardiac surgery. During pre-anesthesia evaluation, baseline rScO2 was recorded with and without oxygen supplementation. The primary endpoint was 30-day mortality, while secondary endpoints were postoperative myocardial injury, respiratory complications, and renal failure. We further evaluated the impact of body position and preoperative hemoglobin (Hb) concentration on rScO2. Of the initially enrolled 130 patients, 126 remained for final analysis. Six (4.76%) patients died within 30 postoperative days. 95 (75.4%) patients were admitted to the ICU. 32 (25.4%) patients suffered from major postoperative complications. There was no significant association between rScO2 and 30-day mortality or secondary endpoints. Oxygen supplementation induced a significant increase of rScO2. Furthermore, Hb concentration correlated with rScO2 values and body position affected rScO2. No significant association between rScO2 values and NYHA, LVEF, or MET classes were observed. Preoperative rScO2 is not associated with postoperative complications in patients undergoing high-risk noncardiac surgery. We speculate that the discriminatory power of NIRS is insufficient due to individual variability of rScO2 values and confounding factors.

Keywords: Monitoring perioperative; Near infrared spectroscopy; Preoperative anesthetic assessment; Risk evaluation.

Fig. 1

Histograms of rScO2 values in a seated position a under ambient air and b under administration of 2 L per minute of oxygen. Histogram showing the distribution of rScO2 values and binned frequency counts

Fig. 2

Receiver operating curve analysis of preoperative cerebral oxygen saturation rScO2 and the combined endpoint of all evaluated complications. Blue ROC curve breathing ambient air (p = 0.05) and red ROC curve breathing 2 L per minute oxygen (p = 0.047)

Fig. 3

Boxplots of rScO2 distribution comparing a breathing ambient air or b breathing 2 L per minute oxygen regarding increasing NYHA Class. The box represents 25th and 75th percentiles; the range between them is the interquartile range. Within the box, the bold line represents the median. The whiskers (extensions from the box) indicate the lowest and highest value no further than 1.5 times the interquartile range. Outliers (values beyond whiskers) are shown as dots

Fig. 4

Boxplots of rScO2 distribution comparing a breathing ambient air or b breathing 2 L per minute oxygen regarding LVEF Class [31]. Left ventricular function Class (1) Normal = LVEF 50–70% (midpoint 60%), (2) Mild dysfunction = LVEF 40–49% (midpoint 45%), (3) Moderate dysfunction = LVEF 30–39% (midpoint 35%), (4) Severe dysfunction = LVEF less than 30%. The box represents 25th and 75th percentiles; the range between them is the interquartile range. Within the box, the bold line represents the median. The whiskers (extensions from the box) indicate the lowest and highest value no further than 1.5 times the interquartile range. Outliers (values beyond whiskers) are shown as dots

Fig. 5

Boxplots of rScO2 distribution comparing a breathing ambient air or b breathing 2 L per minute oxygen regarding body position. On the left side of the diagram the boxplot represents the rScO2 distribution of the patient in the seated position, subsequently the measurement was repeated in the supine position, shown on the right side of the diagram. The box represents 25th and 75th percentiles; the range between them is the interquartile range. Within the box, the bold line represents the median. The whiskers (extensions from the box) indicate the lowest and highest value no further than 1.5 times the interquartile range

Fig. 6

Correlation of rScO2 to preoperative hemoglobin value in a seated position and room air b in seated position under oxygen supplementation. R² Pearson correlations coefficient, p level of significance