Open discussion: Does frontal lobe oxygenation measured by near-infrared spectroscopy represent cerebral oxygenation ?

With accumulating evidence regarding an extracranial contamination of the near-infrared spectroscopy signal (here, here and here), it would be interesting to use this space to have a friendly debate around that issue…

So, does frontal lobe oxygenation measured by near-infrared spectroscopy represent cerebral oxygenation ?

What do you think ?

Update (14/03/2014 – 14h23): Initially, I wanted to have this debate with colleagues involved in studies related to vasopressors/exercise/whole body heating/etc. However, after a discussion on twitter with researchers involved in neurocognition, you guys are more than welcome to comment ! I am here to learn !

Utilization of norepinephrine and phenylephrine during cardiopulmonary bypass in diabetics: What is the impact on brain oxygenation ?

Regular readers will remember my interest in the influence of specific vasopressors on cerebral oxygenation. You can have a look at my postdoc work here and posts related to that topic here, here, here, here and here.

We have just published a new study regarding that issue, this time in diabetics undergoing cardiac surgery.  A potential reduction in brain oxygenation following administration of norepinephrine or phenylephrine could elevate the risk of cerebral ischemia in these patients, a clinical population already at risk for the development of vascular cognitive impairments, transient ischemic attack and stroke. Still, we did not know to date the influence of norepinephrine and phenylephrine on cerebral oxygenation measured by near-infrared spectroscopy during cardiopulmonary bypass in diabetics.

Accordingly, we decided to quantify changes in cerebral oxygenation during administration of norepinephrine and phenylephrine during cardiopulmonary bypass in diabetics and non-diabetics (our goal was not to compare the two vasopressors).

To do so, norepinephrine (6 diabetics; 8 non-diabetics) or phenylephrine (8 diabetics; 9 non-diabetics) was administered intravenously to maintain mean arterial pressure above 60 mmHg during cardiopulmonary bypass.

Mean arterial pressure, venous temperature, arterial oxygenation, and cerebral oxygenation were recorded before anesthesia induction (baseline) and continuously during cardiopulmonary bypass.

Interestingly, cerebral oxygenation was lowered to a greater extent in diabetics vs. non-diabetics with the administration of norepinephrine (-14±13% vs. 3±12%; p<0.05), and tended to be lowered to a greater extent in diabetics vs. non-diabetics with the administration of phenylephrine [-12±8% vs. -6±7%; p=0.1] during cardiopulmonary bypass.

Basically, the results of our study suggest that administration of norepinephrine to restore mean arterial pressure during cardiopulmonary bypass is associated with a reduction in cerebral oxygenation (measured by near-infrared spectroscopy) in diabetics but not in non-diabetics. Administration of phenylephrine is associated with a trend towards a greater reduction in cerebral oxygenation in diabetics compared to non-diabetics.

Although these findings are really interesting, we have highlighted some issues in our discussion:

1) Further studies are necessary in order to support our findings while strictly controlling for potential confounders;

2) Further studies are necessary in order to investigate which mechanism is responsible for the reduction in cerebral oxygenation during cardiopulmonary bypass with the utilization of norepinephrine and phenylephrine in these patients;

3) The small number of subjects precludes the generalization of these findings to the whole population of diabetics undergoing cardiac surgery;

4) 60 mmHg is not necessarily the lower limit of cerebral autoregulation for all patients undergoing cardiac surgery. Accordingly, maintaining mean arterial pressure at 60 mmHg represents a potential limitation of this study, since brain blood flow could be (or not) autoregulated at this mean arterial pressure in these patients;

5) The lowering in cerebral oxygenation, measured by near-infrared spectrocopy, following administration of norepinephrine, could partly be explained by changes in skin blood flow.


Brassard P, Pelletier C, Martin M, Gagné N, Poirier P, Ainslie PN, Caouette M, Bussières JS. Influence of norepinephrine and phenylephrine on frontal lobe oxygenation during cardiopulmonary bypass in patients with diabetes. J Cardiothorac Vasc Anesth. 2013 Dec 17. pii: S1053-0770(13)00513-2. doi: 10.1053/j.jvca.2013.09.006. [Epub ahead of print]

Why cerebral oxygenation doesn’t necessarily decrease during anesthesia-induced hypotension?

Regular readers know my interest for the influence of vasopressors, to correct mean arterial pressure during surgeries (whether it is during anesthesia or cardiopulmonary bypass), on cerebral oxygenation.

In some cases, vasopressors may not be necessary because evidence shows that cerebral oxygenation doesn’t always decrease during anesthesia-induced hypotension. We still don’t know the reason though.

In a recent study, Meng et al. tested the hypothesis that the reduction in brain perfusion, associated with hypotension during anesthesia, would be counterbalanced by a decrease in brain metabolism. As a consequence, frontal lobe oxygenation (or cerebral tissue oxygen saturation in this paper) measured by near-infrared spectroscopy would not be lowered.

To test this hypothesis, the authors recruited patients (data for 30 patients are presented) scheduled for elective non-neurological surgery under general anesthesia with tracheal intubation.

Anesthesia was induced with propofol and fentanyl. Cerebral tissue oxygen saturation, mean arterial pressure, heart rate, cardiac output and bispectral index were monitored before induction of anesthesia and after tracheal intubation.

Mean arterial pressure and cardiac output decreased from pre-anesthesia induction to tracheal intubation. Frontal lobe oxygenation after tracheal intubation remained similar to baseline (pre-anesthesia induction) in these patients.

Cerebral metabolism was not measured in this study. Based on previous findings showing that cerebral metabolism is decreased following induction of anesthesia, the authors inferred that brain oxygen demand and supply remained matched after induction of anesthesia with propofol and fentanyl, leading to an absence of significant change in cerebral tissue oxygen saturation in their patients.


Meng L, Gelb AW, McDonagh DL. Changes in cerebral tissue oxygen saturation during anaessthetic-induced hypotension: an interpretation based on neurovascular coupling and cerebral autoregulation. Anaesthesia 2013; 68(7):736-41.

Phenylephrine and cerebral oxygenation in anesthetized patients: influence of carbon dioxide

Our research group, as well as others, have reported a reduction in cerebral oxygenation with administration of phenylephrine in healthy volunteers and anesthetized patients. You can read other posts related to this topic for further details (here, here, here and here).

Since carbon dioxide is a powerful modulator of cerebrovascular tone, Meng et al. were interested in determining whether the reduction in cerebral oxygenation (measured by near-infrared spectroscopy or NIRS) following administration of phenylephrine, was dissimilar at different arterial carbon dioxide tension (PaCO2) in healthy anesthetized patients undergoing surgery (14 patients [11 males, three females, age 44 (15) yr old, height 175 (10) cm, and weight 80 (14) kg].

Phenylephrine boluses were administered (to increase mean arterial pressure by 20-30% from baseline) during steady state normocapnia (normal PaCO2) and then during steaty state hypocapnia (reduced PaCO2) or hypercapnia (increased PaCO2) by manipulating minute ventilation. Of note, the same dose of vasopressor was administered at all CO2 levels, but doses varied between patients because of differences in body weight. Cerebral oxygenation, cerebral blood volume, systemic hemodynamics and arterial blood oxygen saturation were monitored during the protocol.

Reductions in cerebral oxygenation following phenylephrine boluses were observed during hypocapnia, normocapnia and hypercapnia. However, changes in cerebral oxygenation were significantly different between the three experimental conditions. More specifically,  the reduction is cerebral oxygenation was more important during hypocapnia and blunted during hypercapnia in these patients. Still, changes in mean arterial pressure, cardiac output, heart rate, arterial blood oxygen saturation and depth of anesthesia, were similar between different CO2 levels. Thus, results from this study suggest that the CO2 level could influence the extent of the reduction in cerebral oxygenation following administration of phenylephrine in anesthetized patients.

An important section of the discussion is dedicated to how phenylephrine could influence the NIRS signal. The signal monitored by NIRS represents a mix of arterial and venous blood per unit of cerebral tissue and changes in the cerebral arterial to venous blood volume ratio is known to modify the NIRS signal.

The authors suggest that phenylephrine could influence the arterial/venous blood volume ratio.

They propose that:

(…) the flow velocity in cerebral arterial bed (mainly arterioles) is increased by a phenylephrine-induced increase in perfusion pressure. At the same time, cerebral blood flow (CBF)-regulating vessels (mainly arterioles) constrict due to stretch or increased transmural pressure-mediated vasoconstriction. In accordance with autoregulation, CBF (velocity multiplied by cross-sectional area) is maintained because the increase in velocity is offset by the decrease in cross-sectional area. Cerebral vasoconstriction is an indirect autoregulation-mediated consequence, because phenylephrine does not cross the blood – brain barrier and it cannot constrict cerebral vessels directly.

This study also supports the notion that cardiac output is involved in these changes in cerebral oxygenation, since reductions in cardiac output were observed in the three CO2 conditions. However, since these changes in cardiac output following phenylephrine were similar between CO2 conditions, other mechanisms have to play a significant role in the reduction in cerebral oxygenation following phenylephrine administration.

An important issue is also addressed by the authors:

Even though phenylephrine causes a consistent decrease in SctO2 (cerebral oxygenation), the magnitude of the decrease is very small (1.5–3.5% absolute change depending on CO2 level) and seems unlikely to be clinically significant. Al-Rawi and colleagues found that a relative 13% decrease from baseline, which is approximately twice as great as that seen here, correlates with cerebral ischaemia in patients undergoing carotid artery procedures.

I agree that such a small reduction (and for a short amount of time) may be trivial for the brain. However, some studies have reported greater reductions in cerebral oxygenation following administration of this vasopressor. In addition, we still don’t know what is the impact of a reduction of, let’s say 10%, over longer periods of time (2 hours during cardiopulmonary bypass for example) vs. a few minutes during a controlled study in a laboratory. The latter could have a potentially negative impact on the brain…

Meng L, Gelb AW, Alexander BS, Cerussi AE, Tromberg BJ, Yu Z, Mantulin WW. Impact of phenylephrine administration on cerebral tissue oxygen saturation and blood volume is modulated by carbon dioxide in anaesthetized patients. Br J Anaesth 108 (5): 815–22 (2012). doi:10.1093/bja/aes023

Cerebrovascular physiology online journal club

I was wondering if anyone following this blog would be interested in joining an online journal club where we could discuss papers related to cerebrovascular physiology?

I haven’t thought about the details yet, but with positive answers from you, I could start working on a game plan…

We would also need to choose the best medium to do it. Twitter? Google + Hangout? Any other ideas?

Just let me know if you think it’s a good idea.

I’m looking forward to reading your comments!

Are you writing letters to the editor?

I have already discussed that issue in previous posts (here and here). Now that we have access to social media to discuss science, does the submission of letters to the editor is still worth?  Should we take more of our time to scan the literature related to our area of expertise and be active in post-publication discussion through the official route ? Or should we simply write posts on our respective blog or comments on twitter? Personally, I find unfortunate the fact that after having worked several months/years on a study and published a manuscript, I cannot further discuss the paper (response to a letter to the editor or ecomment) after its publication…

Are you writing letters to the editor ?

If the answer is no, why?

If the answer is yes, why?