Five-Year Follow-up of Patients After Thromboendarterectomy of the Internal Carotid Artery
Relevance of Baroreceptor Sensitivity
Background and Purpose In patients after myocardial infarction, baroreceptor sensitivity has been identified as a factor of prognostic relevance. This study was designed to assess the effects of an increased baroreceptor sensitivity in patients after surgery in the area of the internal carotid artery with respect to blood pressure variability, therapeutic interventions, and vascular events during a 5-year follow-up.
Methods Receptor sensitivity before and immediately after carotid surgery was measured in 84 patients. Blood pressure variability, carotid artery status, and echocardiographic findings were assessed before and after surgery and at the end of follow-up. Vascular events as well as changes in blood pressure therapy during the follow-up period were evaluated.
Results Significant negative correlations between an increase of baroreceptor sensitivity after surgery and the range of systolic (r=−.47; P<.001) and diastolic (r=−.33; P<.01) blood pressure were found for the immediate postoperative period. For the range of systolic blood pressure, this relation persisted (4.3 to 7 years after surgery) at the end of the observation period (r=−.38; P<.001). An inverse relation to the increase in baroreceptor function was also found for the average annual number of therapeutic interventions during follow-up (r=−.38; P<.001). Furthermore, the subgroup of patients without a postoperative increase of receptor sensitivity was characterized by a significantly higher risk of major vascular events (log-rank test, P<.018).
Conclusions Because an improvement of receptor sensitivity after carotid surgery is related to a long-lasting reduction of blood pressure levels and variability, baroreceptor function may be considered an indirect indicator for the later postoperative course.
The importance of the carotid sinus baroreceptors for the regulation of human blood pressure has been known since the publication of Hering's1 work in 1924. Evidence for the relationship between receptor sensitivity and blood pressure variability has been reported in a variety of studies, mostly of an experimental nature.2 3 4 5 6 7 8 9 Practical implications have been described for the receptor sensitivity of patients after myocardial infarction. A low receptor sensitivity is closely associated with the occurrence of ventricular arrhythmias and sudden cardiac death.10 11 12 13
This relationship is of particular interest for surgery on the internal carotid artery because such an intervention involves a direct manipulation in the immediate receptor region. A variety of reports describing characteristic blood pressure changes in the immediate postoperative period support this hypothesis.14 15 16 17 18 19 20 Surgery in other vascular areas (eg, on peripheral vessels) is not associated with any such phenomenon.
Furthermore, it has been demonstrated that changes in baroreceptor sensitivity affect blood pressure in the immediate postoperative phase and during a 6-month follow-up.21 22 The higher the baroreceptor sensitivity, ie, the better the baroreflex function, the higher the stability of the observed blood pressure values.
Patients with stenoses of the extracranial pathways constitute a multimorbid population who frequently show several vascular risk factors, including arterial hypertension. Adequate secondary prevention, including well-adjusted antihypertensive therapy, may be of critical importance for the future course in such patients.
To our knowledge, there are no reports describing to what extent baroreceptor sensitivity after carotid endarterectomy will affect further outcome in such patients or whether a poor receptor sensitivity will be associated with a risk similar to that seen for patients after myocardial infarction.
Thus, our prospective study investigated the possible correlation between postoperative baroreceptor sensitivity and the progression of atherosclerosis in both the local areas of the “operated” carotid artery and vasculature as a whole (occurrence of vascular events) and determined whether blood pressure variability, and thus therapeutic blood pressure control, was affected by the quality of postoperative baroreceptor function.
Subjects and Methods
A total of 101 patients were enrolled in the study. The criteria for inclusion were the availability of the complete preoperative and postoperative data and the regular recording of control measurements and examinations for the full time of the follow-up period until completion of the study or until the death of the patient. The average postoperative follow-up time for our patients was 5.0±1.3 years (maximum 7 years, minimum 4.3 years).
Seventeen patients had to be excluded from evaluation: 2 patients who died in the immediate postoperative phase, 6 with incomplete postoperative data, and 9 with incomplete follow-up records due to poor compliance. Thus, a total of 84 patients, 23 women and 61 men, with an average age of 66.5±7.7 years were evaluated. Twelve patients died during the follow-up period, leaving 72 for the final evaluation at the end of the study.
The study protocol was approved by a local ethics committee, and written consent was obtained from all patients after detailed information was given.
All patients invariably presented with unilateral (39 right, 45 left) symptomatic stenosis of the internal carotid artery (degree of stenosis at the symptomatic side: 79±14% and 38±33% at the contralateral side; 6 patients presented with contralateral occlusion). All patients had a history of hypertension (71% were receiving antihypertensive therapy at the time of enrollment). The average duration of hypertension as documented by the medical histories of our patients (time between the initial diagnosis of hypertension, according to World Health Organization criteria, and the first preoperative assessment) was 9±6 years.
Preoperative diagnostic assessment was made using duplex sonography with additional angiography as indicated. Status after transient ischemic attack (66 hemispheric ischemic events and 18 amaurosis fugax) was the indication for surgery. Surgery was performed with general anesthesia as open thromboendarterectomy at the carotid bifurcation with placement of an intraluminal shunt and subsequent patch graft.
Baroreflex sensitivity testing was performed after placement of an intra-arterial catheter (12 hours preoperatively) and before its removal (36 hours postoperatively). The tests were performed following the instructions of Goldstein et al.23 Three maneuvers were performed: Valsalva, vasoconstrictor-induced hypertension with angiotensin, and vasodilator-induced hypotension with nitroglycerin. As described by Smyth et al,24 baroreflex sensitivity was expressed as the change in pulse interval in milliseconds per millimeter of mercury change in blood pressure.
For each patient and measurement period, the mean value for the results obtained in the three test maneuvers was calculated and designated as preoperative and postoperative baroreceptor sensitivity. According to Goldstein et al,23 averaging leads to a more reliable index of arterial baroreflex function than relying on one technique alone.
For illustrative purposes, two patient groups were differentiated according to the postoperative change in baroreceptor sensitivity: (1) The group of patients with unchanged or reduced postoperative baroreceptor sensitivity comprised 14 women and 39 men (average age, 66±9 years). The average postoperative baroreceptor sensitivity for this group was 1.6±0.9 ms/mm Hg blood pressure change. This group was designated as “nonresponders.” (2) Patients with a postoperative increase of baroreceptor sensitivity of at least 0.5 ms/mm Hg included 9 women and 22 men (average age, 67±6 years). For these patients, designated as “responders,” a postoperative receptor sensitivity of 3.2±1.2 ms/mm Hg blood pressure change was determined.
These groups were comparable with regard to age and sex, risk factor profile, and preexisting vascular disorders (see Tables 1⇓ and 3).
Blood Pressure Variability
Perioperative blood pressure measurements according to Riva-Rocci were taken from day 4 before to day 4 after surgery at hourly intervals during the day and every 3 hours during the night. Before the termination of the study (4 to 6.7 years after surgery), blood pressure was determined by the subjects over a period of 3 months by means of a self-measurement technique, using automatic blood pressure self-measurement devices (Rivatron comfort and Rivatest automatic [Boehringer Mannheim] and Visoton DS-145, KA-We [Bständig]). The measured blood pressure levels were recorded by the patients on data sheets. Patients' records were examined by a physician at 1-month intervals. The study protocol called for twice-daily (morning and evening) measurements of blood pressure (median of 10 entries per week).
The mean values, as well as the ranges of systolic and diastolic blood pressures (the difference between the highest and lowest systolic and diastolic values, respectively), measured during the preoperative, postoperative, and follow-up observation periods were computed. The blood pressure ranges served as indicators for blood pressure variability.
Extracranial Carotid Artery Status
Duplex sonographic assessments of the carotid pathways (Ultramark 8- to 7.5-MHz sector scanner, Acuson 128×P/10m 5- and 7-MHz linear scanner) were performed preoperatively, within the first 2 postoperative weeks, and 5 years postoperatively (controls during follow-up were not included in this report). Duplex sonography was performed in the usual manner described in the literature.25 The usual sonomorphological and hemodynamic criteria26 27 were used for the diagnosis of the degree of stenosis. The parameters evaluated for our study included the percent degree of stenosis on the operated and contralateral sides as determined preoperatively, postoperatively, and at the end of the study.
Transthoracic echocardiography in the parasternal and apical standard projections with two-dimensional and M-mode imaging (Hewlett-Packard Sonos 500 ultrasound unit, 2.5-MHz transducer) was performed within 2 weeks after surgery and at the end of follow-up. Echocardiographic procedures and evaluation of results were performed according to the guidelines issued by the American Society of Echocardiography.28 The parameters evaluated for our study included the left ventricular thickness in millimeters (average of free wall and septum) and the ejection fraction in percentage. The methods described are particularly suitable for a long-term follow-up.29 30 31 32
Vascular Events During Follow-up
At the routine examinations performed at 3-month intervals, the following findings were defined and documented as vascular events: death from cardiac or cerebral causes, transient ischemic attack, prolonged reversible ischemic neurological deficit or nonlethal stroke, nonlethal myocardial infarction, development of symptomatic or worsening of existing peripheral arterial occlusive disease (progression of peripheral arterial occlusive disease), and development of anginal symptoms or worsening of existing anginal symptoms (progression from stable to unstable angina [progression of coronary heart disease]). The type and time point of such events were recorded. Major vascular events were defined as death from vascular causes, myocardial infarction, and stroke.
Blood Pressure Therapy During Follow-up
The patients were instructed to keep a record of their medications for the complete time of the follow-up; patients' entries were checked by a physician at intervals of 3 months, except for the last 3 months of follow-up (see “Blood Pressure Variability,” above). A change of blood pressure therapy was defined as change of therapy for more than 14 days, discontinuation of a medication, or prescription of a new medication. For the purposes of our study, the average annual number of all therapy changes required during the follow-up period was computed.
To investigate the relationship between baroreceptor sensitivity and circulation parameters, Spearman's rank-correlation coefficients were computed. In addition, patients were subdivided into those showing a preoperative-to-postoperative increase of receptor sensitivity (responders) and those without such an increase (nonresponders). These two groups were compared using Mann-Whitney U tests.
For the comparison of responders and nonresponders with regard to vascular events, Kaplan-Meier plots of the probability of remaining event-free as a function of time after carotid surgery were produced, and a log-rank test was performed. Additionally, the predictive value of mean follow-up blood pressure and preoperative-to-postoperative increase of reflex response with respect to major vascular events was assessed by multiple Cox's regression analysis. For all tests, values of P<.05 were considered significant.
Parameters of Blood Pressure Variability, Extracranial Carotid Artery Status, Echocardiographic Findings, and Criteria of Blood Pressure Therapy: Correlation With Postoperative Values and Preoperative-to-Postoperative Increase in Baroreceptor Sensitivity
Table 2⇓ shows the Spearman rank-correlation coefficients of postoperative values and of the preoperative-to-postoperative increase in baroreceptor sensitivity (sensitivity change) to parameters of blood pressure variability, the degree of carotid stenosis, echocardiographic findings, and characteristics of blood pressure therapy.
Generally, the correlations to the preoperative-to-postoperative increase in baroreceptor sensitivity were higher than those obtained for postoperative values. For the immediate postoperative period, values as well as increase in baroreceptor sensitivity were inversely related to systolic and diastolic blood pressures, especially to the ranges of systolic and diastolic blood pressure, and also to left ventricular thickness. Fig 1⇓ shows the inverse relationship between preoperative-to-postoperative change in baroreceptor sensitivity and systolic blood pressure ranges.
However, the preoperative-to-postoperative increase in baroreceptor sensitivity also showed significant negative correlations to systolic blood pressure values and ranges, as well as to left ventricular thickness, measured approximately 5 years after the operation. Furthermore, the lower the postoperative receptor sensitivity or increase in receptor sensitivity, the more frequent were the changes and adjustments of blood pressure therapy required during the follow-up period.
No significant correlation between preoperative receptor sensitivity and blood pressure variability or any other parameter of postoperative status was observed.
For an illustration of the effects of preoperative-to-postoperative increase in baroreceptor sensitivity, patients were grouped into those showing an increase in sensitivity (responders) and those without such an increase (nonresponders). Table 3⇓ presents means and standard deviations of parameters of blood pressure variability, extracranial carotid artery status, echocardiographic findings, and criteria of blood pressure therapy for these groups.
For the preoperative period, no significant differences between responders and nonresponders with respect to blood pressure values and ranges, as well as degree of stenosis, were found. In the immediate postoperative period, nonresponders had statistically significant higher systolic blood pressure values and higher variability (ranges) of systolic and diastolic blood pressure compared with the group of responders. These differences were also seen at the end of follow-up; additionally, left ventricular thickness was higher in nonresponders, who also needed significantly more therapeutic interventions.
Influence of the Duration of Hypertension
Preoperative and postoperative receptor sensitivity were the only parameters showing a relation to the duration of hypertension as established by the patients' medical histories. The longer the hypertensive history of the patient, the lower were the preoperative (r=−.42, P<.01) and postoperative sensitivities (r=−.35, P<.05). The extent of the changes in baroreceptor sensitivity, as well as all other parameters evaluated, was unaffected by the duration of hypertension. Moreover, no statistically significant difference could be found between responders and nonresponders with regard to the duration of hypertension.
Vascular Events During Follow-up: Comparison of Responders and Nonresponders
Three patients died of extravascular causes during the follow-up period (1 death due to prostate cancer, 1 due to bronchial cancer, 1 after multiple trauma); in 1 other patient the cause of death could not be established. Restenosis (defined as more than 50% stenosis) of the operated carotid artery was observed in 9 patients (4 responders [1 symptomatic stenosis] and 5 nonresponders [3 symptomatic]). The vascular events have been listed in Table 4⇓.
Fig 2⇓ shows the cumulative proportion of patients without major vascular events (event-free interval). Differences between responders and nonresponders can already be found after the second postoperative year. This difference reached the level of significance after 33 months. After 5 years, the life-table estimate for the cumulative rate of responders with major vascular events approached 7%, whereas the corresponding rate for nonresponders was 28%. The difference between responders and nonresponders was significant (P<.018). Cox's regression analysis revealed that the increase in baroreceptor reactivity was no longer significant as a predictor for major vascular events when corrected for mean follow-up blood pressure level.
Relationships between surgery on the internal carotid artery and blood pressure alterations have been described in various instances. Both hypertensive and hypotensive dysregulation have been reported.15 18 19
Results of a prior study21 show that correlations between changes in receptor sensitivity and blood pressure therapy can already be found in the immediate postoperative period. In hypertensive patients with a postoperative increase of receptor sensitivity (ie, with an improvement of receptor sensitivity), antihypertensive therapy could be reduced in the immediate postoperative phase. Prolonged hypertension persisting for several years is associated with a general depression of the baroreceptor system, as described in various reports.23 33 It has been hypothesized that circulatory reintegration of the receptors located in the area of the “operated” carotid artery, which had been reset to reduced pressure levels due to stenosis of the carotid artery, may play a major role in at least some of the patients. Such receptors that have not been subject to depression due to prolonged and persistent hypertension may effect an overall improvement of the circulatory regulation system.
In a further study involving a small number of patients, a threshold for the increase of postoperative receptor sensitivity associated with prolonged changes of blood pressure and blood pressure therapy beyond the immediate postoperative phase was identified.22
These findings have now been confirmed in a long-term follow-up for periods of up to 7 years. Significant correlations and differences related to postoperative receptor sensitivity were found not only for the immediate postoperative period but also for periods extending far beyond this initial postoperative phase. Thus, evaluation revealed significant correlations with the absolute values of postoperative receptor sensitivity, but the preoperative-to-postoperative increase in receptor sensitivity showed higher correlations and better predictions of long-term effects.
The changes of receptor sensitivity as a result of the surgical procedure appear to be the decisive factor. The primary question is whether the receptors reintegrated into the circulatory regulation system will be capable of improving overall sensitivity. In other words, the local postoperative responses of the receptors in the operated region will be more essential than the absolute overall sensitivity of the complete receptor system.
Relationships between blood pressure variability and receptor sensitivity have been demonstrated beyond doubt,2 3 4 5 6 7 8 9 and those between receptor sensitivity and blood pressure therapy have at least been suggested in the literature.34 35 36 Hypertensive patients with well-adjusted (and effective) antihypertensive therapy appear to show advantages with regard to receptor sensitivity. Because ethical considerations prohibited comparative measurements of baroreceptor sensitivity in our patient population at the end of the study, receptor sensitivity at 5 years after surgery could only be indirectly assessed in terms of blood pressure variability. Thus, a correlation between improved receptor sensitivity, as indicated by indirect assessment, and successful blood pressure therapy cannot be excluded. However, with the obvious correlation between postoperative sensitivity and blood pressure parameters remaining virtually unchanged throughout the follow-up over 5 years, postoperative receptor sensitivity appears to be the major determinant. This was also confirmed by the fact that no correlation whatsoever could be seen between preoperative receptor sensitivity and postoperative parameters. With respect to the risk of vascular events, baroreceptor response seems to have an indirect influence, since its significant contribution to the prediction of major vascular events vanishes when corrected for mean follow-up blood pressure levels.
Relationships between echocardiographic parameters (eg, in patients with cardiomyopathy) and receptor functions have also been discussed. Various reports describe that successful therapy of patients with left ventricular hypertrophy was related to changes in receptor sensitivity.35 37 These findings are highly consistent with our results. Patients with poor postoperative receptor function reveal increased left ventricular thickness immediately after surgery. This also explains the correlation between the absolute values of receptor function and the duration of hypertension. A prolonged history of established hypertension will usually be associated with reduced preoperative and postoperative receptor sensitivity. In the course of the follow-up, the improved effects of treatment in responders will eventually come to bear. Five years after the operation, patients with at least partially restored postoperative receptor sensitivity (responders) show significant differences with regard to left ventricular thickness.
Although the relationship between cause and effect for this aspect remains to be clarified, it may be hypothesized that postoperatively improved receptor sensitivity may also be responsible for the improved blood pressure variability (facilitated and improved blood pressure adjustment) and may thus effect a regression of secondary consequences of hypertension (lower incidence of cardiovascular deaths, improvement of echocardiographic parameters).
The importance of efficient blood pressure control after carotid surgery has been confirmed in some recent studies. A direct positive relationship between successful therapy and progression/regression of carotid stenoses has been described.38 In addition, certain drugs, such as metoprolol, have been associated with beneficial effects on blood flow characteristics.39
The results presented certainly also have some practical implications. Although routine measurements of receptor sensitivity will not be possible for each carotid intervention, knowledge of postoperative blood pressure behavior will allow for significant conclusions, not only with regard to receptor reflex sensitivity but also long-term blood pressure variability and occurrence of vascular events.
Reprint requests to Mirko Hirschl, MD, Hanuschkrankenhaus, Interne Gefäßambulanz, Heinrich Collinstraße 30, A-1140 Vienna, Austria.
- Received December 22, 1995.
- Revision received March 6, 1996.
- Accepted March 18, 1996.
- Copyright © 1996 by American Heart Association
Hering HE. Der Sinus caroticus an der Ursprungsstelle der Carotis interna als Ausgangspunkt eines hemmenden Herzreflexes und eines depressorischen Gefässreflexes. Münch Med Wochenschr. 1924;72:701-704.
Kasting GA, Eckberg DL, Fritsch JM, Birkett CL. Continuous resetting of the human carotid baroreceptor-cardiac reflex. Am J Physiol. 1987;252:R732-R736.
Mancia G, Bertinieri G, Gavalazzi A, Di Rienzo M, Parati G, Pamidossi G, Ramirez AJ, Zanchetti A. Mechanisms of blood pressure variability in man. Clin Exp Hypertens. 1985;7:167-178.
Simpson DM, Wicks R. Spectral analysis on heart rate indicates reduced baroreceptor-related heart rate variability in elderly persons. Gerontology. 1988;43:21-24.
Sleight P. The relevance of baroreflex mechanisms to the control of ambulatory and exercise blood pressure in humans. Clin Exp Pharmacol. 1989;16(suppl 15):65-69.
Takabatake T, Yamamoto Y, Ohta H, Nakamura S, Hara H, Ishida Y, Hashimoto N, Hattari N. Blood pressure variability and hemodynamic response to stress in patients with paroxysmal elevation of blood pressure. Clin Exp Hypertens. 1985;7:235-242.
Farrell TG, Odemuyiwa O, Bashir Y, Cripps TR, Malik M, Ward DE, Camm AJ. Prognostic value of baroreflex sensitivity testing after acute myocardial infarction. Br Heart J. 1992;67:129-137.
Wade JG, Larson CP Jr, Hickey RF, Ehrenfeld WK, Severinghaus JW. Effect of carotid endarterectomy on carotid chemoreceptor and baroreceptor function in man. N Engl J Med. 1970;282:823-828.
Hirschl M, Kundi M, Hirschl MM, Liebisch B, Magometschnigg D. Blood pressure responses after carotid surgery: relationship to postoperative baroreceptor sensitivity. Am J Med. 1993;33:463-468.
Goldstein DS, Horwitz D, Keiser HR. Comparison of techniques for measuring baroreflex sensitivity in man. Circulation. 1982;66:432-440.
Smyth HS, Sleight P, Pickering GW. Reflex regulation of arterial pressure during sleep in man. Circ Res. 1969;24:109-120.
Grant EG. Duplex sonography of the cerebrovascular system. In: Grant EG, White D, eds. Duplex Sonography. New York/Berlin: Springer Verlag; 1988:8-65.
Landwehr P. Extracranielle hirnversorgende Arterien. In: Wolf KJ, Fobbe F, eds. Farbcodierte Duplexsonographie. Stuttgart/New York: Georg Thieme Verlag; 1993:45-68.
Eckmann A, Kuhn FP, Groenninger J. Postoperative follow-up after thromboendarterectomy of the carotid artery using duplex sonography. ROFO Fortschr Geb Röntgenstr Nuklearmed. 1986;145:639-641.
Floras JS, Hassan MO, Jones JV, Osikowska BA, Sever PS, Sleight P. Factors influencing blood pressure and heart rate variability in hypertensive humans. Hypertension. 1988;11:273-281.
Grassi G, Giannattasio C, Cleoux J, Cuspidi C, Sampieri L, Bolla GB, Mancia G. Cardiopulmonary reflex before and after regression of left ventricular hypertrophy in essential hypertension. Hypertension. 1988;12:227-237.
Little WA, Young WA. The effect of nicardipine on blood pressure, its variability and reflex cardiac control. Br J Clin Pharmacol. 1985;20:1155-1195.
Sopher SM, Smith ML, Eckberg DL, Fritsch JM, Dibner-Dunlap ME. Autonomic pathophysiology in heart failure: carotid baroreceptor-cardiac reflexes. Am J Physiol. 1990;259:H689-H696.
Sutton-Tyrrell K, Wolfson SK Jr, Kuller LH. Blood pressure treatment slows the progression of carotid stenosis in patients with isolated systolic hypertension. Stroke. 1994;25:44-50.
Spence JD. Effects of antihypertensive drugs on flow disturbances: nifedipine, captopril, and metoprolol evaluated by quantitative spectral analysis of Doppler flow patterns in patients with carotid stenosis. Clin Invest. 1994;17:319-325.