Page 32 - ANZCP Gazette May 2023

Jabur et al.
5
Figure 2. Pre–post aortic cross-clamp removal difference in Mx (right hemisphere) with mean and 95% CI. Comparison is of mean pre-post change between groups.
vasoreactivity and autoregulation in humans. Our pri- mary hypothesis was that, when compared to patients undergoing CCS, patients undergoing OCS would be exposed to more cerebral arterial emboli and, conse- quently, would exhibit a significantly greater increase in correlation of MAP and CBFv (implying greater auto- regulation impairment) after removal of the aortic cross-clamp. Patients undergoing OCS were indeed exposed to more cerebral emboli than patients undergo- ing CCS, particularly after the release of the aortic cross- clamp (Table 2). The substantially greater emboli distribution to the right cerebral hemisphere in both forms of surgery was unexpected and the underlying reason is not known. The brachiocephalic artery is the first and largest vessel arising from the aortic arch, and it is possible that the flow of blood from the tip of the arterial cannula (a potential source of some emobli) is
Figure 3. Pre–post aortic cross-clamp removal difference in Mx (left hemisphere) with mean and 95% CI. Comparison is of mean pre-post change between groups.
directed toward this vessel.20 A study of neurologic defi- cits after cardiac surgery reported predominant left sided hemiparesis.21 If this asymmetry is attributed to the distribution of emboli, this could reflect the higher right sided embolic count. Greater distribution of larger bubbles may be affected by their buoyancy but it is dif- ficult to appreciate why this would be relevant in a supine patient. A detailed appraisal of the emboli detec- tion results appears elsewhere.17
In our linear mixed model analyses, there was no sig- nificant overall effect of group, so, as defined, our pri- mary hypothesis was not confirmed. However, there was a significant interaction between group and side, and a significantly greater increase in Mx on the right side after removal of the cross-clamp in the OCS group than in the CCS group. Given that the majority of emboli in both groups were seen on the right side, this finding is consistent with the research construct underpinning our primary hypothesis: namely, that greater emboliza- tion could be associated with a greater increase in Mx reflecting a greater reduction in autoregulation.
One possible interpretation of the failure to see a sig- nificant difference between the groups overall is that the study was underpowered. Our sample size was pragmatic, the interpatient variation was considerable, and we did not anticipate the marked difference seen between the left and right sides. On the basis of our data, we have now estimated that 105 patients per group would be needed to show a difference of the magnitude seen here with α = 0.05 and β = 0.8. However, given what we now know about the distribution of emboli, an alternative view would be that the hypothesis of a future study should focus on the right side, and we would have been justified in accepting such a hypothesis on the basis of our current findings had we stated it in those terms at the outset.
The view that our results support a relationship between emboli counts and autoregulation is confluent with the findings of the only other study that explicitly evaluated the effect of arterial embolization on cerebral autoregula- tion. In that study, latex microsphere exposure in a canine model of CPB resulted in greater dependence of CBFv on MAP.15 We are unaware of other studies that have evalu- ated this issue using emboli exposures more typical of clinical cardiac surgery (where the majority of emboli are very small bubbles),9 or in humans. However, there are studies that demonstrate endothelial damage,11,12,22,23 arte- rial inflammation,24 and impaired vasoreactivity12,13 in animals, and deranged peripheral vasoreactivity in humans,14 after intravascular bubble exposure. Collecti- vely, these studies provide a credible basis for suggesting that cerebral arterial bubble exposure could impair cer- ebral vasoreactivity and autoregulation in humans, although the matter cannot yet be taken as fully settled.
Resolving this question is important because the answer would inform debate about whether continued or enhanced attention to minimizing arterial emboli
29 MAY 2023 | www.anzcp.org