The natural history of carotid fibromuscular dysplasia

Transcription

1 The natural history of carotid fibromuscular dysplasia Mark T. Stewart, M.D., Mark W. Moritz, M.D., Robert B. Smith III, M.D., J. Timothy Fulenwider, M.D., and Garland D. Perdue, M.D., Atlanta, Ga. Although graduated internal dilatation has proved to be an effective, safe, and durable operation for the treatment of symptomatic patients with fibromusofiar dysplasia of the extracranial internal carotid artery, the role of surgical treatment in this entity remains unclear because the natural history is not well defined. Forty-nine patients, aged 29 to 82 years (mean, 58.5 years), with angiographically proven fibromusoalar dysplasia of 88 internal carotid arteries have been evaluated since Twenty patients showed symptoms of focal cerebral or retinal ischemia, 10 patients had nonlateralizing neurologic symptoms, three patients sustained intracerebral hemorrhage, five patients complained of nonischemic symptoms, and 11 patients were asymptomatic. The three patients with intracranial hemorrhage and one person who suffered a massive stroke after angiography died within weeks of admission; no surgical therapy was performed. Initial management of the other patients included four internal carotid endarterectomies in four patients for associated atherosclerosis, one with simultaneous graduated internal dilatation; seven graduated internal dilatations in five patients; and one extracranial-to-intracranial bypass in a patient with occlusion occurring after graduated internal dilatation. Seventy-three nondilated arteries in 42 patients have been followed for up to 16 years (mean, 6.8 years). During this time only three patients have undergone surgical therapy; one carotid endarterectomy was done for an asymptomatic atherosclerotic lesion and two graduated internal dilatations in patients with nonfocal ischemia. Through follow-up of all 49 patients, none has had a new neurologic deficit. Fourteen patients who initially presented with focal ischemia were not treated surgically and all but one are now asymptomatic. No instances of carotid dissection or hemorrhage from intracranial aneurysm rupture have been observed during follow-up. Carotid fibromuscular dysplasia appears to be a benign problem that only rarely warrants operative intervention. (J VAsc SURG 1986; 3: ) Fibromuscular dysplasia (FMD) is a rare arteriopathy of uncertain etiology. Leadbetter and Burkland 1 7~ported the first case of a patient with renal artery stenosis and hypertension caused by FMD in FMD was: ~ubsequently shown to affect many medium-sized ~essels, including carotid, celiac, hepatic, superior mesenteric, and external iliac arteries. 2'3 The first histologically proven case of carotid FMD was published in 1965 by Connett and Lansche. 4 Subsequently, many clinical syndromes were attributed to these lesions, including transient hemispheric ischemia, cerebral infarction, nonfocal cerebral ischemia, extracranial and intracranial aneurysms, and carotid dissection, s From the Department of Surgery, Emory University School of Medicine. Presented at the Thirty-ninth Annual Meeting of the Society for Vascular Surgery, Baltimore, Md., June 7-8, Reprint requests: Mark T. Stewart, M.D., Department of Surgery, 69 Butler St., Atlanta, GA Surgical treatment of symptomatic patients has been advocated with the use of a variety of techniques, including arterial resection, bypass with saphenous vein, excision of fibromuscular tissue with patch grafting, and graduated internal dilatation (GID). 4"6"s GID has become the standard surgical approach and good results have been reported with its application from several centers. 912 However, despite the success of surgical treatment, the role of surgery in altering the natural history of this disease remains poorly defined. Recent reports of angiographically proven carotid FMD indicate that the prognosis of untreated lesions is relatively be. nign with few subsequent episodes of cerebral ischemia) 31s In many cases the angiographic finding of carotid fibromuscular dysplasia is only incidental with no symptomatic correlation. We have reviewed our experience with patients who had carotid FMD to better define the natural history and prognosis of this disorder in the unoperated patient. 305

2 306 Stewart et al. Journal of VASCULAR SURGERY Table I. Presentation of 49 patients with carotid fibromuscular dysplasia TIA Asymptomatic Nonhemispheric ischemic Vascular tinnitus Amaurosis fugax Intracerebral hemorrhage Hemispheric stroke Headache Postoperative occlusion ( +TEA TIA = transient ischemic attack. Table II. Angiographic findings in 49 patients with carotid fibromuscular dysplasia (FMD) Findings No. of patients % No. of patients Bilateral FMD 39 No Other lesion 31 Bilateral atherosclerosis 4 Unilateral atherosderosis 4 Unilateral FMD 10 No other lesion 5 Nonvisualized opposite side 2 Contralateral atherosclerosis 1 Bilateral atherosderosis 1 Contralateral occlusion s 1 ~Previous graduated internal dilatation. PATIENTS The records of all patients evaluated at Emo~ University Hospital from January 1969 through January 1985 with angiographically proven carotid FMD were reviewed. There were 49 patients, only one of whom was male, aged 29 to 82 years (mean, 58.1 years), who had involvement of 88 internal carotid arteries. Data after hospitalization were collected from medical records, referring physicians, patients, or family members. No patient was lost to follow-up. Three patients with intracranial hemorrhages resulting from ruptured aneurysms died within 2 months of their initial hospitalizations. One additional patient admitted with transient hemispheric ischemia sustained a major stroke following angiography and died within 2 months of discharge. Excluding these four patients, the remaining 45 patients had angiographically proven FMD of 81 internal carotid arteries. Two were identical twins. These patients have been followed up for as long as 196 months (mean, 77.9 months). The results of follow-up largely reflect the natural history of carotid FMD, since operations for this disorder were not performed at Emory University before 1981 L~ Presenting problems are listed in Table I. Clinical findings included a carotid bruit on at least one side in 35 of the 49 patients. Bruits were bilateral in 19 patients. The absence of a bruit carried little significance, however, as no bruit was audible in 33 diseased vessels. Nineteen patients complained of headache mad 22 were hypertensive. High blood pressure was easily controlled in most patients. Few have undergone angiographic investigation for renovascular disease and none has been treated surgically for that problem. The one patient with a postoperative carotid occlusion was referred to Emory University Hospital 1 month after undergoing GID of the left internal carotid artery elsewhere. She came to our institution with symptoms of left hemispheric i,~% - emia, was treated by extracranial-to-intracranial (EC- IC) bypass, and is now asymptomatic. ~ The angiographic spectrum of disease is presented in Table'IL Both internal carotid arteries were visualized in all but two patients. Atherosclerotic disease was present at the bifurcation of the internal carotid artery in 10 patients. Intracranial aneurysms were identified in seven patients (14%), three of whom had shown evidence of intracranaial hemorrhage and died. Two nonruptured aneurysms have been treated successfully surgically. Repeat angiography was available in eight patients at 3 to 101 months (mean, 43.1 months) after the initial studies. Only two patients showed significant change. One patient who initially presented with unilateral FMD had developed FMD in the contralateral artery 67 months later. An atherosclerotic (80% diameter) stenosis developed in an internal carotid artery of another patient with bilater~a~ FMD at 101 months. "~' Beginning in 1981 some patients if our series were subjected to operative intcrventiofi Jbr carotid FMD, chiefly for persistent neurologic symptoms. Operations within 3 months of diagnosis have been classified as early. Ten patients were thought to warrant early operation. Four carotid endartcrectomies (one with simultaneous GID) were performed in four patients with associated atherosclerotic stenosis. In addition, five patients underwent a total of seven GIDs and one patient required an EC-IC bypass for transient ischemic attacks (TIAs) following occlusion after dilatation. All of these patients who underwent early operation recovered without complication and are currently withdut symptoms at a mean of 23.4 months from diagnosis. Thirty-five patients did not undergo early opcr-

3 Volume 3 Number 2 February Natural history of carotid fibromuscular dysplasia 307 ation (Table III). Follow-up in these patients average&91.1 months. None has had a stroke during observation. Three patients in this group have died: one at 42 months of congestive heart failure, one at 102 months of lung cancer, and one at 96 months following a myocardial infarction. All of these persons were neurologically asymptomatic at the time of death. Three patients have undergone late operation. Persistent symptoms of nonfocal cerebral ischemia led to operative therapy in two patients at 24 and 66 months after diagnosis. One was completely relieved of episodic syncope, whereas the other continues to experience paroxysms of dizziness and bilateral blurred vision. One other patient had an asymptomatic 80% atherosclerotic stcnosis at the bifurcation c ~ the left common carotid artery and was treated by Carotid endartcrectomy without GID of the fibromuscula~':disease present. All initially asymptomatic patients remained free of symptoms during follow-up. Only one patient who initially presented with TIAs and is now followed up elsewhere continued to experience ischemic events at 99 months. Of nine patients evaluated for symptoms of nonfocal ischemia and not treated surgically, only one remains symptomatic, although two of this group were treated by late GID. From the perspective of unoperated arteries with FMD, 73 nondilated arteries in 42 patients have been followed for an average of 81.6 months. Two arteries during this time were treated by late GID, both in patients with persistent symptoms of nonfocal cerebral ischemia. No patient has suffered cerebral or retinal infarction in a territory supplied by an artery with FMD. No instances ofintracerebral hemorrhage Juring follow-up have been observed, although two ~patients have undergone surgical treatment. Spontaneous co,-otid dissection has not occurred in any patient in ~tfiis series. DISCUSSION Subsequent to the first description of FMD of the internal carotid artery in 1964, cerebral ischemia and a variety of other pathologic entities were linked to this lesion including stroke, nonfocal cerebral ischemia, carotid dissection, and TIAs.S With recognition of these associations, surgical procedures were developed to correct the anatomic defect. GID is now the standard operative approach and results with this procedure in selected centers have been good, with morbidity rates at or slightly above that expected for carotid endarterectomy. 912 Indications for operation in carotid FMD are Table III. Presentation of nonoperated patients (n = 35) No. of Problem patients (%) Transient ischemic attack l0 (29) Nonfocal ischemia 9 (26) Asymptomatic 7 (20) Vascular tinnitus 3 (9) Stroke 3 (9) Amaurosis fugax ] (3) Headache 2 (6) Table IV. Symptoms of surgically treated patients previously reported Author Krupski Starr Collins Symptom et al. n et al. w et al. v Transient ischemic attack 41% 44% 38% Nonfocal ischemia 31% 72% 46% Asymptomatic 8% 24% -- Vascular tinnitus 23% -- 8% Stroke 22% Amaurosis fugax 23% 16% 23% controversial. In reported surgical series, as in our experience, the spectrum of presenting symptoms is similar to that of patients with atherosclerosis undergoing carotid endarterectomy (Tables III and IV). ~ The assumption that patients with carotid FMD have a prognosis similar to patients with atherosclcrosis and therefore require corrective surgery is, however, unproved. The natural history of carotid FMD has remained unclear, although recent reports indicate that it may be extremely benign. 1 -~2 At Emory University Hospital, GID for carotid FMD has been performed only during the last 4 years. Most patients wcrc investigated for complaints specific to extracranial cerebrovascular disease but, with the finding of carotid FMD, were not treated surgically. These nonoperated patients, symptomatic or not, have fared very well over the years and none has sustained a stroke during observation. Is there any explanation for this difference from athcrosclerotic disease? First, the finding of carotid FMD on an arteriogram does not necessarily imply that it is the cause of the patient's symptoms. In a series of 79 patients reported by Corrin, Sandok, and House? 3 from the Mayo Clinic, carotid FMD was an indidental finding in the majority and few had or developed ischemic symptoms. None of our initially asymptomatic patients became symptomatic during follow-up.

4 308 Stewart et al. Journal of VASCULAR SURGERY Table V. Late ischemic cerebral infarction in nonoperated patients Authors No. of patients No. of strokes (%) Corrin et al. ~ (4) Wells, Smith ~4 15 2~(13) Current study 35 0 (0) *Both had carotid atherosclerosis. Second, carotid FMD may be associated with other entities that may cause cerebral ischemia. Atherosclerosis at the carotid bifurcation has been identiffed in 50% of patients with focal neurologic events and carotid FMD. ~2 Thirty-nine percent of the patients treated by GID in the series reported by Collins et al.9 also required a concomitant carotid bifurcation endarterectomy. Carotid atherosclerosis was visualized in 20% of the patients in the present series. If atherosclerosis is not present, other causes of cerebral ischcmia such as cardiac emboli, dysrhythmias, small vessel thrombosis, and postural alterations in blood pressure may be implicated. Next, patients with atherosclerosis, especially those who experience focal cerebral ischemia, are thought to be at high risk for recurrence and/or major stroke. Is this true for carotid FMD? Operation on symptomatic carotid FMD presupposes that recurrent symptoms and/or cerebral infarction are likely and that surgical treatment will reduce this risk. Our experience does not support this concept as none of our patients suffered cerebral infarction with or without operation during follow-up. Moreover, the majority of symptomatic patients not operated on had spontaneous resolution of their symptoms during observation. Other authors who have followed up patients with carotid FMD also report a low incidence of cerebral infarction, and if such does occur, an alternate cause is commonly identified (Table V).~3-~ This is not to Say that carotid FMD cannot cause cerebral infarction but rather that the incidence is probably very low. Moreover, surgical treatment is not without risk, since an operative morbidity between 3% and 6% has been reported Including perioperative cerebral infarction, patients followed up after GID by Krupski, Effeney, and Ehrenfeld H and Collins et al. 9 had a stroke rate of 5% and 8%, respectively. If operative morbidity exceeds the expected stroke rate in nonoperated patients, obviously operation should not be performed. Repeat angiography has rarely been documented in patients with carotid FMD and the incidence of progressive disease has not been clarified. Only 15 patients with carotid FMD undergoing seria~ngiography have previously been described in the literature. Of these, progressive FMD was observed in five and a new lesion in two. s Of eight patients in the present series, only one developed a new lesion of FMD in the contralateral vessel and atherosclerotic stenosis evolved in one other. The significance of angiographic progression is uncertain as these changes were not correlated with new clinical symptoms. What role, if any, should surgery play in patients with carotid FMD? Clearly, no surgical treatment is indicated for asymptomatic persons. Patients with symptoms of focal ischemia will probably have spontaneous cessation of these symptoms and thus do not require immediate operation. Most patients in present series with focal neurologic events who were not operated on have been treated empiri~3.11y with antiplatelet agents. There will continue to be an occasional candidate with persistent symptoms of focal ischemia, recurrent nonfocal ischemic changes associated with objective hemodynamic abnormalities, or rarely intolerable vascular tinnitus who may benefit from direct operative intervention. However, the risk of operation should be carefully considered in light of the prognosis of the nonopcrated patient. CONCLUSIONS Carotid FMD is a benign, often incidental finding, which rarely causes ischemic infarction during prolonged observation. Indeed, coincidental vascular lesions may serve as the cause of cerebral ischemia. Certainly symptoms of TIA caused by carotid FMD are not associated with the same incidence of morbidity as with atherosclerotic carotid occlusive dk ease. When neurologic or visual complaints can b~" attributed to carotid FMD, these gener~lly resolve without operation and recurrence is e"~ceptional. Medical therapy is unproved, but antiplatelet agents seem appropriate and may be prescribed. We appreciate the assistance of JeWelry A. Landman, M.D. in reviewing the angiograms. REFERENCES 1. Leadbetter WF, Burkland CE. Hypertension in unilateral renal disease. J Urol 1938; 39: Palubinskas AJ, Ripley HR. Fibromuscular hyperplasia in extrarenal arteries. Radiology 1964; 82: Stanley JC, Gewertz BL, Bove EL, Sottiurai VS, Fry WJ. Arterial fibrodysplasia, histopathologic character, and current etiologic concepts. Arch Surg 1975; 110: Connett MC, Lansche JM. Fibromuscular hyperplasia of the

5 Volume 3 Number 2 Februaqr 1986 Natural history of carotid fibromuscular dysplasia 309 internal carotid artery: Report of a case. Ann Surg 1965; ~[ ~2: Sandok BA. Fibromuscular dysplasia of the internal carotid artery. Neurol Clin 1983; 1: Morris GC, Lechter A, DeBakey ME. Surgical treatment of fibromuscular disease of the carotid arteries. Arch Surg 1968; 96: Harrington OB, Crosby G, Nicholas L. Fibromuscular hyperplasia of the internal carotid artery. Ann Thorac Surg 1970; Perry MO. Fibromuscular disease of the carotid artery. Surg Gynecol Obstet 1972; 134: Coffins GJ, Rich NM, Clagett GP, Spebar MJ, Salander JM. Fibromuscular dysplasia of the internal carotid arteries: Clinical experience and follow2up. Ann Surg 1981; 194: Starr DS, Lawrie GM, Morris GC. Fibromuscular disease of carotid arteries: Long-term results of graduated internal dilatation. Stroke 1981; 12: Krupski WC, Effeney DJ, Ehrenfeld WK. Fibromuscular dys- plasia, aneurysms, and spontaneous dissection of the carotid artery. In: Bergan JJ, Yao JST, eds. Cerebrovascular insufficiency. New York: Grune & Stratton, Inc, 1983: So EL, Toole JF, Dalai P, Moody DM. Cephalic fibromuscular dysplasia in 32 patients. Arch Neurol 1981; 38: Corrin LS, Sandok BA, Houser OW. Cerebral ischemic events in patients with carotid artery fibromuscular dysplasia. Arch Neurol 1981; 38: Wells RP, Smith RR. Fibromuscular dysplasia of the internal carotid artery: A long-term follow-up. Neurosurgery 1982; 10: Patman RD, Thompson JE, Talkington CM, Garrett WV. Natural history of fibromuscular dysplasia of the carotid artery. Stroke 1980; 11: Effeney DJ, Ehrenfeld WK, Stoney RJ, Wylie EJ. Why operate on carotid fibromusctdar dysplasia? Arch Surg 1980; 115: DISCUSSION Dr. Bert Smith (Dallas, Tex.). We agree that fibromuscular dysplasia of the internal carotid artery is generally a benign lesion. In 1980 we reviewed 25 patients with 47 vessels involving disease of the internal carotid artery, and we treated seven of those patients surgically and safely with either a combination of carotid endarterectomy or graduated internal dilatation. We noted that 18 patients were asymptomatic up to 10 years of follow-up. We believed then and continue to believe that operative intervention in asymptomatic patients is unwarranted and are happy to see our thoughts substantiated by this article. However, unremitting hemispheric symptoms in patients receiving antiplatelet therapy continue to be a problem for us and we would still consider surgical intervention in that instance. Regarding technique, is balloon angioplasty, which is so appropriate in treatment of these lesions elsewhere in the arterial ~ ~stem, efficacious therapy intraoperatively in the internal dhrotid artery, and has this technique been used at Emory? We know of no instance of this being used in our practice at Baylor, but we do have knowledge of a patient who had a balloon dilatation performed in whom a carotid-cavernus sinus fistula developed. Dr, David J. Effeney (Brisbane, Australia). In our experience in San Francisco, 101 patients operated on for alleged symptomatic fibromuscular disease presented with a constellation of neurologic symptoms. It is difficult for me to accept the conclusion that symptomatic fibromuscular hyperplasia is a benign disease. When patients are asymptomatic, we have continually said they should not undergo operation. However, 25 patients have had completed strokes by the time of presentation. Some of these patients did have atherosclerosis at the carotid bulb, but atherosclerosis is not sufficient to explain all of the symptoms once they develop. The problem with this disease is that it is unpredictable. The first symptom may be a stroke. Even the data that the authors present to us here confirm that the complex of pathologic features that we find with fibromuscular disease--the carotid involvement, the presence of an excess ofintracranial aneurysms, and known hypertension--make this disease a far from benign disorder. It is of some interest that when I presented our data in Chicago, I offered to make a registry of fibromuscular disease. The response to that from the membership was underwhelming. Two surgeons.responded, each contributing one extra case. Although the membership, in the survey presented by Dr. DeWeese, indicated that this is not a role for the societies, I wonder whether, in the specific case of fibromuscular dysplasia, we would not learn a lot more if we all pooled our resources. I submit to you that operation in symptomatic patients can be performed with satisfactory early and late stroke morbidity but that patients without symptoms should not undergo operation. Dr. Martin M. Fisher (New York, N.Y.). In symptomatic carotid patients, I have been using the intravenous digital subtraction angiogram as a screening test only with ambulatory patients similar to what we would use with an intravenous pyelogram. This, with the history and noninvasive studies, sometimes will aid in the diagnosis. However, the conventional four-vessel angiogram, does not always confirm ulcerated plaques. However, I wonder.why in this conference the intravenous digital subtraction angiogram has been overlooked, and I would appreciate comments on that source.

6 310 Stewart et al. Journal of VASCULAR SURGERY Dr. Stewart (closing). With regard to intravenous digital subtraction angiography, we have had considerable experience with that at Emory and found that the technical quality of these studies was not really adequate to base decisions regarding operation. We have used intra-arterial digital studies and found these to be of acceptable quality. We have no experience with percutaneous transluminal angioplasty of the internal carotid artery for fibromuscular dysplasia, although there are a number of case reports. If the purpose of the procedure is to prevent further embolization to the brain, I wonder about the advisability of that procedure, and we have not employed it. Regarding the incidence of stroke, the group from San Francisco certainly illustrated that a number of clinical problems can occur in relation to carotid fibromuscular dysplasia. Once these patients present with a complication of fibromuscular dysplasia, is the risk of recurrence signif- icant enough to justify proceeding with surgical therapy? I think in patients who do have persistent problems~lated to cerebral ischemia, that is, persistent transient hemispheric ischemia, amaurosis fugax, or relatively severe and continuing episodes of nonfocal hemispheric ischemia, dilatation is certainly justified. However, in a relatively large number of patients that we have followed up without surgical therapy, we have not observed any subsequent strokes even in those patients who presented with evidence of focal hemispheric ischemia. Furthermore, all but one of the patients who were not operated on who had symptoms of focal hemispheric ischemia ceased to have these symptoms. Our review demonstrates that neurologic deterioration is not the natural history of carotid fibromuscular dysplasia. Instead, it is usually a stable lesion and thus we will continue to urge caution in recommending surgery if the indication for operation is to prevent stroke in these patients.