(Stroke. 1998;29:1068-1069.)
© 1998 American Heart Association, Inc.
Transcranial Doppler and Stump Pressure During Carotid Endarterectomy
Simona Zannetti, MD;
Giuseppe Giordano, MD;
Piergiorgio Cao, MD
Unità Operativa di Chirurgia Vascolare,
Policlinico Monteluce,
Perugia, Italy
To the Editor:
We read with great interest the article by Finocchi et
al.1 on the role of transcranial Doppler
(TCD) and stump pressure (SP) during carotid
endarterectomy (CEA). Having recently published a
paper on the same subject,2 we would like to offer some
considerations as an adjunct to the issues raised by Finocchi et al.
We understand that it was the intention of the authors to assess the
usefulness of SP as an indicator of hemodynamic changes
predicting intraoperative cerebral ischemia. In order to do so,
112 patients who underwent CEA for symptomatic and
asymptomatic severe carotid stenosis under general
anesthesia were monitored by TCD and SP measurement. After
examining duration of clamping, values of TCD flow reduction and of SP
at clamping, microembolic signals, and
perioperative complications, the authors concluded that
the major complications of CEA may result from
hemodynamic factors and that SP alone is not a reliable
indicator of hemodynamic changes that predict clamping
ischemia.
In our study, we evaluated a series of 175 CEAs monitored by SP
measurement and TCD and performed under local anesthesia.
The need for shunting was compared between SP/TCD flow velocity
reduction and the awake response. Since there is no general consensus
on the appropriate SP cutoff value that indicates the need for
shunting,3 4 5 we constructed a receiver operating
characteristic (ROC) curve to determine the relationship between
specificity and sensitivity of SP and TCD. Values that combined the
highest sensitivity with the highest specificity for both SP and TCD,
using the ROC curve, were 
50 mm Hg (100% sensitivity, 83%
specificity) and 
70% flow velocity reduction from baseline (83%
sensitivity, 96% specificity), respectively. In our experience, both
SP and TCD showed limitations, because they overestimate or
underestimate CEAs in need of a shunt. We believe that sensitivity is
more important than specificity in CEA and thus concluded that SP is a
more dependable indicator of cerebral perfusion than TCD. We
acknowledge the usefulness of intraoperative monitoring with TCD for
testing adequacy of shunt flow and detection of embolism.
We believe that drawing conclusions about reliability of SP while not
being able to actually document the neurological status of patients at
clamping is speculative. In our opinion, a gold standard should be used
to test the effectiveness of a monitoring technique. In our study,
patients were operated on under local anesthesia, which
allowed us to effectively test the response of patients to carotid
cross-clamping and thus assess the reliability of the monitoring
techniques. Furthermore, we believe that the arbitrarily chosen cutoff
value of 40 mm Hg is rather low. This value appears to be
associated with a low sensitivity. According to our analysis, a
cutoff value of 40 mm Hg implies an undershunting rate of 18%.
In other words, using this cutoff value, 18 patients of 100 who
exhibited either focal or global ischemia at clamping would not
have been shunted (sensitivity 82%, specificity 95%). The opposite is
true with a 50 mm Hg cutoff: there is an overshunting rate of
17%, but all of the patients with neurological deficits at clamping
would have been shunted (sensitivity 100%, specificity 83%).
With respect to postoperative neurological complications, the authors
state in their discussion that the major complications of CEA may be
related to hemodynamic factors. This estimate is in
conflict with other reports,6 7 and in our opinion this
conclusion is not supported by the findings on postoperative cerebral
CT scans: in the 5 nonshunted patients who developed neurological
postoperative complications, 3 postoperative CT scans remained
unchanged, and 2 showed new "small deep infarcts." Based on these
findings, it is reasonable to believe that the "small deep
infarcts" are lacunar infarctions, the pathophysiology and causes of
which are still controversial. Therefore, we believe that it is
hypothetical to consider the hemodynamic changes that
occur at clamping the only factors responsible for these two
ischemic events. Likewise, the authors suggest that in the two
postoperative large cerebral infarctions detected on CT scan (which
occurred in 2 shunted patients), a relevant role for embolism cannot be
excluded. According to the high SP values in those 2 patients (52 and
60 mm Hg), shunt and potential embolism from shunt might have
been avoided.
Finally, we believe it would have been interesting if the authors had
further investigated independent predictors of postoperative
neurological complications (eg, clamping times, hypertensive brunts,
microemboli detected at TCD, contralateral occlusion, etc) through
multivariate analysis.
References
1.
Finocchi C, Gandolfo C, Carissimi T, Del Sette M,
Bertoglio C. Role of transcranial Doppler and
stump pressure during carotid
endarterectomy. Stroke. 1997;28:2448–2452.[Abstract/Free Full Text]
2.
Cao P, Giordano G, Zannetti S, De Rango P, Maghini M,
Parente B, Simoncini F, Moggi L. Transcranial
Doppler monitoring during carotid
endarterectomy: is it appropriate for selecting
patients in need of a shunt? J Vasc Surg. 1997;26:973–980.[Medline]
[Order article via Infotrieve]
3.
Hunter GC, Sieffert G, Malone JM, Moore WS. The
accuracy of carotid back pressure as an index for shunt
requirements. Stroke. 1982;13:319–26.[Abstract/Free Full Text]
4.
Harada RN, Comerota AJ, Good JM, Hashemi HA, Hulihan
JF. Stump pressure, electroencephalographic changes and the
contralateral carotid artery: another look at selective
shunting. Am J Surg. 1995;170:148–153.[Medline]
[Order article via Infotrieve]
5.
Brewster DC, O'Hara PJ, Darling C, Hallett JW Jr.
Relationship of intraoperative EEG monitoring and stump pressure
measurements during carotid endarterectomy.
Circulation. 1980;62(suppl 1):1–7.
6.
Riles TS, Imparato AM, Jacobowitz GR, Lamparello PJ,
Giangola G, Adelman MA, Landis R. The causes of
perioperative stroke after carotid
endarterectomy. J Vasc Surg. 1994;19:206–216.[Medline]
[Order article via Infotrieve]
7.
Whitney EG, Brophy CM, Kahn EM, Whitney DG.
Inadequate cerebral perfusion is an unlikely cause of
perioperative stroke. Ann Vasc Surg. 1997;11:109–114.[Medline]
[Order article via Infotrieve]
Response
Cinzia Finocchi, MD;
Carlo Gandolfo, MD;
Massimo Del Sette, MD
Department of Neurological Science and Neurological
Rehabilitation,
University of Genoa,
Genoa, Italy
Tiziana Carissimi, MD;
Carlo Bertoglio, MD
Department of Vascular Surgery,
Imperia Hospital,
Imperia, Italy
In this response we comment on the discrepancies between the
results of two different studies1 2 on the same subject:
the usefulness of SP and TCD monitoring as indicators of
hemodynamic changes predicting cerebral
ischemia during CEA.
There is an important difference in methods between the two studies:
the surgical procedure was performed under local anesthesia
in one1 and under general anesthesia in the
other.2 Cao et al1 founded that TCD had a
greater specificity but a lower sensitivity than SP with use of the
awake response under local anesthesia as the gold standard
for the need for shunt. Of course, it is only speculative that the same
results might be extended to patients under general
anesthesia, a condition in which the tolerance to cerebral
hypoperfusion is quite different.
In our study,2 we founded that 5 patients in the
nonshunted group developed a cerebral ischemic complication.
The findings on postoperative CT scan cannot help us determine the
pathogenesis of these complications, because only two new lesions, a
lacunar infarction and an internal borderzone infarction, whose
pathogenesis is controversial, were detected. However, in the latter
type, a hemodynamic pathogenesis is often
involved.3 4 The combined evaluation of percent reduction
of TCD mean velocity and clamping duration allowed us to separate
patients with and without cerebral ischemic complications. This
finding is a strong indicator of a hemodynamic
pathogenesis. We were not able to differentiate between patients with
and without cerebral ischemic complications using an SP of
either 40 mm Hg or 50 mm Hg. SP furnishes a "point"
evaluation of clamping ischemia and is a reliable indicator of
critical cerebral hypoperfusion, but its usefulness may be limited in
patients with moderate cerebral hypoperfusion in whom the developing of
ischemic complications may be time dependent. We think that SP
alone measures blood pressure value far from the brain time, whereas
TCD can give dynamic information on blood velocity directly in the
middle cerebral artery, which is the result of collateralization after
clamping, through the circle of Willis. Inadequate collateralization,
together with moderate but prolonged hypoperfusion, can bring the
development of cerebral ischemia, and continuous TCD monitoring
can contribute to the identification of this apparently not-at-risk
condition. The revision of our data through
multivariate analysis (see the
Table
)
does not offer further indications.
Finally, we would like to underline a methodological problem: some
surgeons measure the pressure directly within the internal carotid
artery; others, concerned about putting the needle into the internal
carotid artery, measure the blood pressure in the common carotid
artery. Cao et al1 used the first method, and
we2 used the second. This may cause discrepancies in the
measure of SP, but to the best of our knowledge, there is no specific
study in the literature that has addressed this methodological
problem.
References
1.
Cao P, Giordano G, Zanetti S, De Rango P, Maghini
M, Parente B, Simoncini F, Moggi L. Transcranial
Doppler monitoring during carotid
endarterectomy: is it appropriate for selecting
patients in need of a shunt? J Vasc Surg. 1997;26:973–980.
2.
Finocchi C, Gandolfo C, Carissimi T, Del Sette M,
Bertoglio C. Role of transcranial Doppler and
stump pressure during carotid
endarterectomy. Stroke. 1997;28:2448–2452.
3.
Hupperts RMN, Lodder J, Wilmink J, Boiten J, Heuts Von
Rook EPM. Haemodynamic mechanism in small subcortical borderzone
infarcts? Cerebrovasc Dis. 1993;3:231–235.
4.
Gandolfo C, Del Sette M, Finocchi C, Calautti C, Loeb
C. Internal borderzone infarcts in patients with
ischemic stroke. Cerebrovasc Dis. In press.
