| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1997;64:891
© 1997 The Society of Thoracic Surgeons
Department of Anesthesiology Mayo Clinic 200 First St SW Rochester, MN 55905
To the Editor:
We read with great attention and interest the article by Trivedi and associates entitled "Relative Changes in Cerebral Blood Flow During Cardiac Operations Using Xenon-133 Clearance Versus Transcranial Doppler Sonography" [1].
We congratulate them for their work, but note that they did not discuss other studies of relevance. Previous studies did not find a good correlation of transcranial Doppler sonography-derived middle cerebral artery blood flow velocity with cerebral blood flow during cardiopulmonary bypass. The study of Weyland and associates [2] and our own [3] used the Kety-Schmidt technique to determine cerebral blood flow. These values were compared with simultaneous middle cerebral artery blood velocity before, during, and after hypothermic cardiopulmonary bypass, and both investigations reported an r value of approximately 0.3. Like Trivedi and associates, the more recent abstract by Grocott and colleagues [4] used the xenon-133 clearance technique to determine cerebral blood flow and a comparison with simultaneous middle cerebral artery blood flow velocity was made at three time points during mild hypothermic cardiopulmonary bypass. Like Weyland's group and our own, they found that transcranial Doppler sonography did not reliably indicate cerebral blood flow. It is incumbent upon Trivedi and associates to explain why their results differed from previously published reports. Although the simplicity of transcranial Doppler sonography offers advantages over measurements of cerebral blood flow, the validity of this technology under the condition of hypothermic cardiopulmonary bypass has not been clearly demonstrated.
References
Departments of Cardiac Surgical Research and Cardiothoracic Surgery
The Rayne Institute St. Thomas' Hospital London SE1 7EH, UK
To the Editor:
We are grateful to Drs Nuttall and Cook for their interest in our study and for their kind comments. We welcome the opportunity to discuss, in relation to our work, the other studies they mention. First, the abstract by Grocott and coworkers [1] was clearly published after our article was in press; consequently, we believe that it is unreasonable for us to be expected to discuss these results and await with great interest their publication in a formal article. The other two articles will be discussed in further detail.
Weyland and colleagues [2] and Nuttall and coworkers [3] have, like ourselves, compared the relationship between cerebral blood flow (CBF) and middle cerebral artery flow velocity (
MCA) as measured by transcranial Doppler sonography (TCD) during cardiac operations. These two studies used the Kety-Schmidt technique for CBF with only the alpha-stat method of acid-base management during cardiopulmonary bypass (CPB) but were performed in a small number of patients (15 and 16, respectively). They report that there is a poor correlation between relative changes in CBF and MCA and conclude that TCD is not a suitable noninvasive technique for CBF monitoring during CPB. In contrast, we examined patients undergoing CPB with both pH-stat and alpha-stat acid-base management regimens and demonstrated a good correlation between relative changes in CBF and MCA.
Both the studies cited have defects and their conclusions are not as obvious as stated. In Weyland and colleagues' study, results were obtained from only 14 patients, with time points for comparisons made before skin incision, at hypothermic CPB, and after chest closure. There are many variables that change between induction of anesthesia and the hypothermic phase of CPB such as hematocrit, effect of lifting the pericardium, and a degree of systemic cooling of the patient. All these factors may influence CBF and, had comparisons been made just before and after CPB, the results may have been different. To examine relative changes, Weyland and colleagues compared the change from the immediately preceding value rather than from a true baseline value, as was done in our study. It was also reported that there was an overall increase in MCA during the hypothermic phase of CPB. However, examination of the graph (Figure 3 in the article) shows that the MCA increased in only 6 of the 14 patients. By reporting a mean increase in MCA (by only 4 cm/s) they are overestimating the true picture.
Nuttall and colleagues had results from only 12 patients, and these had an average age of 70 years compared with a much younger group (58 years) in our study. The combination of an older and much smaller study group could easily bias the results if only 1 patient had undetected cerebrovascular disease. Examination of their scatter plot (Figure 3 in their article) shows three points that could reasonably be treated as outliers; exclusion of these may have provided a significant correlation (we prefer to refer to it as linear regression as one is trying to predict changes in flow from changes in velocity), especially as their p value just fails to reach significance (p = 0.06). Nuttall and colleagues also use the Bland and Altman technique for their analysis of the relative changes in CBF and TCD velocity. In our view this analysis is inappropriate. In the original Bland and Altman article [4] all the examples given are comparing instruments that measure the same parameter. Here the comparison is between flow and velocity—two different parameters—and conversion of the relative changes to percentages does not make them the same in the context of the Bland and Altman analysis.
Both these studies have used the Kety-Schmidt technique for measuring CBF, and this has been criticized by Young and colleagues [5]. The Kety-Schmidt technique has not been proved to be the "gold standard" for CBF measurement in CPB. In our view there is a major problem in that, during CPB, the diffusion of the tracer takes place across the oxygenator rather than the patient's lungs. It is also known that there is a degree of pulmonary edema after CPB, which may alter comparisons made before and after CPB using inhalation tracers. We are not of the opinion that the Kety-Schmidt technique is the gold standard for CBF determination during CPB. In both the studies cited, a great deal of attention is paid to the association between the direction of changes in CBF and MCA, and changes in the same direction could only be demonstrated in 29% to 50% of measurements. Again, the problem of small numbers of patients needs to be considered as each patient had four measurements and so three changes in CBF and velocity. If any 1 patient had a poor relationship between flow and velocity then three data points would be included. Our data found that in only 10% of cases was the directional change not the same between flow and velocity. In all of the instances the discrepancy was not during CPB but in the change from 37°C CPB to after CPB. The reason for our better association in directional changes may be due to the fact that TCD measures the flow velocity in the basal cerebral arteries (usually the middle cerebral) and we used direct intraarterial injection of xenon 133 to determine CBF. When the arterial route is used to determine CBF it is likely that any other index of arterial flow would provide a better correspondence between velocity and flow. This may also explain why other investigators using intraarterial injections have found a good relationship between changes in CBF and TCD velocity [6, 7].
Consequently, we believe that TCD is a valid technique for monitoring changes in CBF during CPB and has the additional benefit of detecting emboli.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
