Bhavani Shankar Kodali MD

Value of capnography during major vascular surgery

It has been well explained in this website that end tidal carbon dioxide values (ETCO2) has a very strong relationship to cardiac output. In my opinion, the value of this relationship is grossly underestimated in clinical practice. For a given ventilation, abrupt decreases in ETCO2 reflects decreases in cardiac output. Utilizing this principle, Boccara et al, from France, demonstrated that there is a correlation between the decreases in ETCO2 on cross-clamping and hypotension ensuing unclamping. When aortic clamp was released, systolic hypotension (>20%) occurred in those subjects who had a decrease in ETCO2 greater than 15% during aortic cross-clamping.

To assess the variations in end-tidal CO2 in response to aortic cross-clamping and the relationship with systolic arterial pressure (SAP) changes induced by unclamping. Boccara et al studied thirty-three patients undergoing infrarenal aortic abdominal aneurysm repair by aorto-aortic prothetic bypass prospectively. All patients were anesthetized with i.v. midazolam (0.05 mg x kg(-1)), thiopentone (3-5 mg x kg(-1)), fentanyl (5 microg x kg(-1)), pancuronium (0.1 mg x kg(-1)) and the maintainance of anesthesia used was 1-1.5% end-tidal isoflurane and i.v. fentanyl. The perioperative management was standardized. ETCO2 and SAP were measured 5 min before (Pre-XAA), 15 min after infrarenal aortic cross-clamping (XAA), 5 min before (Pre-UXAA) and immediately after unclamping (UXAA).

A total of 16 (48.5%) from 33 patients presented decrease in SAP following aortic unclamping, and 13 out of these patients had arterial hypotension defined as SAP<90 mmHg. End-tidal CO2 variation (PreXAA-PreUXAA) induced by aortic clamping was correlated with SAP variation (PreUXAA-UXAA) induced by unclamping (r=0.763; P=0.0001). An end-tidal CO2 reduction above 15% after aortic cross-clamping was found to have a 100% sensitivity to detect a SAP decrease greater than 20% after unclamping, with a 100% specificity and a negative predictive value of 1.0. Complete aortic occlusion duration was not correlated to SAP unclamping variation. Intraoperative characteristics (fluid loading, hematocrits, urinary output) were comparable, although blood loss was higher in patients experiencing ΔSAP>20%.

The authors concluded that end-tidal CO2 variation monitoring during aortic cross-clamping may provide a reliable and non-invasive method to predict unclamping hypotension. When the aortic clamp was released, systolic hypotension (>20%) occurred in those subjects who had a decrease in end-tidal CO2 greater than 15% during aortic cross-clamping.

A second study from Greece by Vretzakis et al also evaluated capnography during aortic cross-clamping in a different prospective. The aim of this study was to evaluate the alterations of CO2 release during abdominal aortic surgery. Seventeen patients undergoing abdominal aortic aneurismal (AAA) repair and 8 patients undergoing repair of aortoiliac occlusive disease were the study subjects. The authors recorded PaCO2, PetCO2, PECO2, VD, VD (alveolar), and VCO2. Patients with aneurysms were randomly divided to have constant ventilation (group AA) or modified ventilation to preserve normocapnia (group AB) during clamping. Patients with AAA showed a significant decrease of VCO2 during clamping and an elevation after unclamping in both groups (AA and AB), with no statistical difference between them. During clamping, ETCO2 and PaCO2 values decreased and VD (alveolar) was increased especially in group AA, while unclamping produced the opposite effect.


1. Boccara G, Jaber S, Eliet J, Mann C, Colson P. Monitoring of end-tidal carbon dioxide partial pressure changes during infrarenal aortic cross-clamping: a non-invasive method to predict unclamping hypotension. Acta Anaesthesiol Scand. 2001 Feb;45(2):188-93.

2. Vretzakis G, Papadopoulos G, Koutsias S, Papaziogas B, Antoniadou E, Pitoulias G, Papadimitriou D. Alterations in carbon dioxide release during abdominal aortic clamping for aneurysmal or occlusive repair. Minerva Anestesiol. 2001 Sep;67(9):629-36.