Cardiovascular Surgery, Vol. 9, No. 4, pp. 334–338, 2001 2001 The International Society for Cardiovascular Surgery. Published by Elsevier Science Ltd All rights reserved. Printed in Great Britain 0967-2109/01 $20.00
PII: S0967-2109(01)00019-9
www.elsevier.com/locate/cardiosur
Carotid endarterectomy in awake patients with contralateral carotid artery occlusion R. Karmeli, N. Lubezky, M. Halak, Z. Loberman, B. Weller and S. Fajer Vascular Surgery Department, Carmel Medical Center, Neurology Department, Bnai Zion Medical Center, Haifa, Israel Objectives: Patients with severe stenosis of an internal carotid artery with contralateral occlusion (ICO) are at an increased risk for stroke, and therefore surgical treatment is usually recommended. Carotid endarterectomy (CEA) under regional anesthesia enables constant monitoring of neurologic status and selective shunting in cases of clinically evident cerebral ischemia. In this study, we assess the selective use of shunts based solely on changes in neurological status in awake patients with ICO undergoing CEA as well as their complication rates. Methods: During 1996–1998, we studied intraoperative findings and results of CEA under regional anesthesia with clinical monitoring of neurological status in two groups: (1) patients with stenosis (>70% by NASCET) and contralateral occlusion (n = 50) and (2) patients with stenosis and no contralateral occlusion (n = 94). Results: Shunt insertion was required in 42% of group 1, and 6% in group 2. All of the patients in group 1 requiring shunts had stump pressures ⬍50 torr. The average stump pressure of group 1(40 torr) was significantly lower than that of group 2 (75 torr), and was also lower than that of patients with severe contralateral stenosis (35 patients, 76 torr). Perioperative stroke rates were identical in both groups (2.1%). Conclusion: Since ICO patients are at a high risk for brain ischemia during ICA clamping, they require shunt insertion frequently. Patients with no contralateral occlusion require shunting at a much lower rate — even in the presence of severe contralateral stenosis. Regional anesthesia allows for early detection of brain ischemia and therefore, the perioperative results in both groups are similar. 2001 The International Society for Cardiovascular Surgery. Published by Elsevier Science Ltd. All rights reserved Keywords: internal carotid artery occlusion, carotid endarterectomy, regional anesthesia, selective shunting
Introduction Most of the current literature supports carotid endarterectomy (CEA) for patients suffering from severe internal carotid artery stenosis of >70% and contralateral internal carotid artery occlusion (ICO)
Correspondence to: Ron Karmeli, M.D., Vascular Surgery Department, Carmel Medical Center, 7 Michal Street, Haifa 34362, Israel. Tel.: +972-4-825-0267; Fax: +972-4-826-1013
334
[1–4]. The natural history of patients with ICO and contralateral stenosis treated medically vs. surgically differs significantly [3, 5]. Nicholls published a combined TIA and stroke rate of 8% in the first year and 33% in the second year [3]. The 5 yr stroke rate is significantly lower in patients undergoing CEA — 9– 11% [3, 5]. During carotid endarterectomy of patients with contralateral ICO with impaired collateral circulation, cerebral perfusion may decrease with the possible development of neurological symptoms. CARDIOVASCULAR SURGERY
AUGUST 2001 VOL 9 NO 4
Carotid endarterectomy in awake patients: R. Karmeli et al.
Common available methods of cerebral protection include shunt use and the maintenance of higher systemic blood pressure [4, 5]. An ongoing debate exists concerning the safest method of anesthesia, intraoperative monitoring, and strategy for shunt use. The current options include: (1) regional or general anesthesia with nonselective shunting [2, 6–9], (2) general anesthesia with selective shunting according to stump pressures and/or EEG signs of ischemia [5, 10, 11], and (3) regional anesthesia with selective shunting in patients suffering from intraoperative clinical signs of ischemia [4, 9]. Our vascular surgery department performs most CEA’s under regional anesthesia with selective shunting in patients demonstrating clinical signs of ischemia during cross clamping of the ICA. The aim of our study was to assess CEA in awake patients (under regional anesthesia) with ICA stenosis (>70%) and contralateral occlusion in comparison to those without contralateral occlusion with regard to the following parameters: 1. intraoperative findings such as stump pressure and changes in neurological status, 2. shunt use, 3. perioperative complication rates.
Methods During the years 1996–1998, 94 consecutive patients without contralateral occlusion were compared to 50 patients with contralateral occlusion who underwent CEA in our hospital. A total of 306 carotid endarterectomies were performed during this period. Preoperative evaluation included duplex scans, CTA, Brain CT, and angiography in selected cases. Patients with internal carotid artery occlusion were defined as those with ICO diagnosed by duplex and confirmed by CTA [12]. The degree of stenosis by duplex was determined by criteria based on the Washington University, Seattle, WA data. Stenosis greater than 70% was defined by the following: 1. A peak systolic velocity (PSV) of 170–200 cm/s, 2. An end diastolic velocity (EDV) of 110–140 cm/s, 3. The presence of spectral broadening, 4. A PSV ratio, (i.e. PSA in the ICA /PSV in the CCA) >4
Group 1 — patients with severe stenosis (>70% as defined by NASCET) and contralateral occlusion, Group 2 — patients with severe stenosis and varying degrees of contralateral stenosis but no contralateral occlusion. The endarterectomy was performed under regional cervical anesthesia with clinical monitoring of neurological status and selective shunting solely according to changes in neurological status during cross clamping, after maximal manipulation of blood pressure, and independent of stump pressure values. Statistical analysis Comparison between groups was performed using the chi square test and t-test of independent variables. Statistical significance was defined by P⬍ 0.05.
Results Epidemiology Most patients in both groups aged over 60. In group 1, 26% aged 60–69 yr old, whereas 36% aged 70– 79 yr old. In group 2, 23% aged 60–69 yr old, whereas 52% aged 70–79 yr old. The average age of the patients in both groups was 71 yr. The gender distribution among both groups was similar: approximately 70% male patients compared to 30% females. Almost all patients underwent surgery under regional anesthesia (92%). Eight percent underwent general anesthesia in both groups due to poor patient compliance or failed cervical block. Clinical presentation Ipsilateral clinical presentation varied significantly between those patients with and without contralateral occlusion (Table 1). Although the majority of operated patients in group 1 were asymptomatic ipsilateral to the stenosis, the presence of the contralateral occlusion and ipsilateral severe stenosis in itself was an indication for CEA. In comparison, in those patients without contralateral occlusion (group 2), a total of 60% of the patients were symptomatic on the operated side. Table 1 Clinical presentation ipsilateral to CEA (P = 0.000) Group 1
In our study published in 1998, CTA has a positive predictive value of 95% in diagnosing ICO without the complications of angiography [12]. Our two groups of patients were defined as follows: CARDIOVASCULAR SURGERY
AUGUST 2001 VOL 9 NO 4
Asymptomatic TIA CVA Total symptomatic
76% 11% 13% 24%
Group 2 40% 52% 8% 60%
335
Carotid endarterectomy in awake patients: R. Karmeli et al.
Evaluation of contralateral stenosis in group 2 with regard to intraoperative results
Figure 1 Intraoperative findings
Surgery (CEA) was performed on a relatively large amount of asymptomatic patients as was indicated randomly in those participating in an international asymptomatic study, in those with severe bilateral stenosis, and in cases awaiting cardiac bypass surgery. Intraoperative findings Significant differences were found with regard to stump pressure and shunt use between groups with and without contralateral occlusion (Figures 1 and 2). Shunting was indicated in the presence of neurological changes during cross clamping independent of stump pressures. In group 1 with contralateral occlusion, stump pressures averaged 41.6 mmHg with shunt use in 42% of the cases. In contrast, in patients without contralateral occlusion, stump pressures averaged 74.6 mmHg and shunts were inserted in only 6.4% of the cases (P = 0.001). In addition, we assessed shunt use with regard to stump pressure. In group 1 with contralateral occlusion, 76% of those shunted had stump pressures ⱕ40 mmHg, while 24% had >40 mmHg (but less than 50). Thirty one percent of those not needing shunts had stump pressures less than 40 mmHg compared to those who needed shunts with stump pressures less than 40 mmHg — 76% (P = 0.001).
A breakdown of patients with varying contralateral stenosis on the operated side was performed in group 2. Forty-four patients suffered from stenosis ⱕ50%, eight patients had stenosis between 51 and 70%, and 35 patients suffered stenosis ⱖ71%. Overall, 40% of the patients in group 2 suffered severe contralateral stenosis of >71% (NASCET). We further studied the effect of contralateral stenosis on stump pressure during cross-clamping. No significant difference ( P = 0.58) was found among stump pressures with increasing stenosis (i.e. ⬍50%, 51–70%, and >71%) (Table 2). Significant contralateral stenosis did not affect stump pressures as demonstrated by an average stump pressure of 75.8 mmHg in those with > 70% stenosis as compared to the overall average stump pressure in group 2 of 74.6 mmHg. Six patients (6/35) suffered from tight stenosis: 91–99%. One of these patients was operated upon with general anesthesia with shunt; stump pressure was not measured. The average stump pressure of the five remaining patients was 57 mmHg, with none requiring shunts. In those patients with contralateral stenosis of 91–99%, 2/6 patients had stump pressures below 50 mmHg while 3/6 patients had stump pressure above 50 mmHg. Perioperative stroke rate Perioperative stroke rates were determined by an independent peer group. In both groups, a 2% stroke rate was reported. No stroke was debilitating, and no mortality was reported.
Discusssion Perioperative stroke rates in patients with ipsilateral severe stenosis and contralateral occlusion have been reported to vary between 1 and 17% [4, 5] with NASCET reporting 14.2% in a small group of patients. Imparato reported his results of perioperative stroke rates when CEA was performed under regional anesthesia to be 6.3% [4]. In studies where non-selective shunting was performed, lower stroke rates were observed — 4% [2]. Jacobowitz reported a 0.7% early perioperative stroke rate using regional anesthesia and selective shunting [9]. Table 2 Stump pressure and contralateral stenosis NS (P = 0.58)
Figure 2 Intraoperative findings
336
Stump ⱕ40 mmHg Stump >40 mmHg
0–50%
51–70%
71–99%
11%
0%
9%
89%
100%
91%
CARDIOVASCULAR SURGERY
AUGUST 2001 VOL 9 NO 4
Carotid endarterectomy in awake patients: R. Karmeli et al.
Regional anesthesia allows for constant neurological monitoring and therefore selective shunting is performed independently of stump pressure, since stump pressure does not always correlate reliably with the presence of cerebral ischemia [10, 13]. In addition, changes in neurological status under regional anesthesia has been shown to be more accurate than EEG monitoring as an indicator for clinical cerebral ischemia [13]. Under regional anesthesia, shunt use in high risk patients (s/p recent CVA or contralateral occlusion) has been reported to range between 6 and 28% [4, 9, 13–15], in comparison to overall shunt use under regional anesthesia ranging from 5 to 10% [7, 14]. We report an unusually high rate of shunt use (42%) under regional anesthesia in those with contralateral occlusion. Similar figures have been reported under general anesthesia: 46% shunt rate while using EEG changes under general anesthesia as indications for shunt use, while Cao et al. reported a 30% shunt use in CEA’s using regional and general anesthesia [7]. However under regional anesthesia, Lawrence et al., recently published a 19% (5/26) shunt use rate in a group of patients undergoing CEA under regional anesthesia with contralateral occlusion [14], while Harbaugh et al. recently reported a 16% shunt rate in 43 patients with contralateral ICO [15]. In our group of patients with contralateral occlusion, all patients who required shunting had stump pressures less than 50 mmHg (60%), however, 40% of those with stump pressures below 50 mmHg did not need a shunt as dictated by their stable neurological status. This supports the conclusion by Harada et al., who reported that stump pressures alone are not sensitive enough to accurately predict the need for shunt use. Their study showed EEG to be more sensitive in indicating shunt use whereas we have demonstrated the efficacy of awake patient monitoring [10]. Our results also show a considerable difference between patients with bilateral severe stenosis and those with contralateral occlusion with regard to intraoperative findings and the need for shunt. This seems to emphasize the fact the stenoses are not flow limited until tight stenosis >90% is present. In the literature, these two groups are often similarly identified as being high risk due to bilateral disease in both cases. However, a total occlusion proved to have a significant influence on cerebral hemodynamics intraoperatively, as demonstrated by the relatively higher average stump pressure in the nonoccluded group (74.6 mmHg) and their low shunt use (6.4%). Within group 2, 40% had contralateral stenosis > 70%, which did not reduce brain perfusion in most cases, as measured by an average stump pressure of 76 mmHg. According to these results, patients with bilateral CARDIOVASCULAR SURGERY
AUGUST 2001 VOL 9 NO 4
stenosis behave differently from those with contralateral occlusion and ipsilateral severe stenosis with regard to stump pressures and shunt use. We believe that in the presence of occlusion, cerebral hemodynamics are affected by an impaired collateral circulation and decreased vasodilatory reserve, resulting in decreased stump pressures and an increased need for shunts. Perioperative stroke rates in our patients under regional anesthesia with selective shunting were low — 2%, whether the patient suffered from bilateral stenosis or occlusive disease contralateral to severe stenosis. We conclude that regional anesthesia with neurological monitoring of awake patients to be safe for patients with severe stenosis and contralateral occlusion. We also suggest considering shunting all patients with ICO undergoing general anesthesia because of their relatively high rate of shunt use.
References 1. Faught, W.E., van Bemmelen, P.S., Mattos, M.A. et al., Presentation and natural history of internal carotid artery occlusion. Journal of Vascular Surgery, 1993, 18, 512–524. 2. Coyle, K.A., Smith, R.B., Salam, A.A. et al., Carotid endarterectomy in patients with carotid occlusion: review of a 10-year experience. Cardiovascular Surgery, 1996, 4(1), 71–75. 3. Nicholls, S.C., Bergelin, R. and Strandness, D.E., Neurologic sequelae of unilateral carotid artery occlusion: immediate and late. Journal of Vascular Surgery, 1989, 10, 542–548. 4. Friedman, S.G., Riles, T.S., Lamparello, P.J. et al., Surgical therapy for the patient with internal carotid artery occlusion and contralateral stenosis. Journal of Vascular Surgery, 1987, 5, 856–861. 5. Mackey, W.C., O’Donnell, T.F. Jr and Callow, A.D., Carotid endarterectomy contralateral to an occluded carotid artery: Perioperative risk and late results. Journal of Vascular Surgery, 1990, 11, 778–785. 6. Frawley, J.E., Hicks, R.G., Gray, L.J. et al., Carotid endarterectomy without a shunt for symptomatic lesions associated with contralateral severe stenosis or occlusion. Journal of Vascular Surgery, 1996, 23, 421–427. 7. Cao, P., Giordano, G., De Rango, P. et al., Carotid endarterectomy contralateral to an occluded carotid artery: a retrospective case-control study. European Journal of Vascular and Endovascular Surgery, 1995, 10, 16–22. 8. McCarthy, W.J., Wang, R., Pearce, W.H. et al., Carotid endarterectomy with an occluded contralateral artery. American Journal of Surgery, 1993, 166, 168–172. 9. Jacobowitz, G.R., Adelman, M.A., Riles, T.S. et al., Long-term follow-up of patients undergoing carotid endarterectomy in the presence of a contralateral occlusion. American Journal of Surgery, 1995, 170, 165–167. 10. Harada, R.N., Comerota, A.J., Good, G.M. et al., Stump pressure, electroencephalographic changes, and the contralateral carotid artery: another look at selective shunting. American Journal of Surgery, 1995, 170, 148–153. 11. Julia, P., Chemla, E., Mercier, F. et al., Influence of the status of the contralateral carotid artery on the outcome of carotid surgery. Annals of Vascular Surgery, 1998, 12, 566–571. 12. Lubezky, N., Fajer, S., Barmeir, E. et al., Duplex scanning and CT angiography in the diagnosis of carotid artery occlusion: a prospective study. European Journal of Vascular and Endovascular Surgery, 1998, 16, 133–136. 13. Benjamin, M.E., Silva, M.B. Jr, Watt, C. et al., Awake patient monitoring to determine the need for shunting during carotid endarterectomy. Surgery, 1993, 114, 673–681.
337
Carotid endarterectomy in awake patients: R. Karmeli et al. 14. Lawrence, P.F., Alves, J.C., Jicha, D. et al., Incidence, timing, and causes of cerebral ischemia during carotid endarterectomy with regional anesthesia. Journal of Vascular Surgery, 1998, 27, 329–337. 15. Magnadottir, H.B., Lightdale, N. and Harbaugh, R.E., Clinical
338
outcomes for patients at high risk who underwent carotid endarterectomy with regional anesthesia. Neurosurgery, 1999, 45, 786–792. Paper accepted 20 February 2001
CARDIOVASCULAR SURGERY
AUGUST 2001 VOL 9 NO 4