- Email: [email protected]
The role of arterial reconstruction in spontaneous renal artery dissection
The role of arterial reconstruction in spontaneous renal artery dissection L i n d a M. Reilly, M D , C h r i s t o p h e r G. C u n n i n g h a m , M D , R o b e r t M a g g i s a n o , M D , William K. Ehrenfeld, M D , and R o n a l d J. Stoney, M D , San Francisco, Calif. Spontaneous renal artery dissection is an uncommon cause of renovascnlar hypertension, usually associated with fibromuscular dysplasia. Among reported nonautopsy cases (N = 80), arterial reconstruction has seldom been attempted (N = 21) and the outcome has frequently been poor (48% clinical failure rate). This is attributed in part to the frequent involvement of renal artery branches by the dissection. Furthermore, the report of spontaneous reversion to normoteusion among patients treated medically has also clouded the role of surgery in this disease. Since progress in the technique of renal artery repair now allows successful treatment of anatomically complex lesions, we reviewed our experience with arterial reconstruction in the management of spontaneous renal artery dissection to determine the frequency of and factors correlating with cure after operative repair. Ten patients (eight men, two women; mean age, 39.3 --. 5.9 years) were admitted with severe hypertension (10/10), often associated with neurologic symptoms, hematuria, or flank pain (8/10). Serum creatinine was elevated in only two patients. Angiography demonstrated changes consistent with fibromuscular dysplasia in 7 of 10 patients and evidence of dissection in 6 of 10. Bilateral disease was present in three patients. Only five patients had a single renal artery on the involved side. The dissection extended into the primary branches in 8 of 10 patients and involved both renal arteries in four of the five patients with two arteries. Histologic study confirmed fibromuscular dysplasia in six and intramural dissection in all operative specimens. Five patients underwent revascularization (in one case requiring the ex vivo technique), with use of hypogastric artery as a conduit in four of five or resection and primary reanastomosis in one of five. Three patients became normotensive, and two returned to their previous level of blood pressure control. Follow-up averaged 14.5 years. Two patients underwent nephrectomy after exploration demonstrated nonrecoustructible vessels, and two underwent nephrectomy when intraoperative assessment of the kidney showed that revascularization had failed to adequately reverse extensive renal ischemia. After a mean follow-up of 14.6 years these patients remain normotensive, although two require antihypertensive medications. One patient was treated medically and is currently hypertensive offall medications. Nine of 10 patients have maintained a normal serum creatinine during follow-up. We conclude that renal revascularization is frequently successful in spontaneous renal artery dissection (five of seven, 71.4%) and results in sustained relief of hypertension with maximal conservation of renal tissue. This is important because of the young age at onset and the not infrequent occurrence of bilateral fibromuscular dysplasia, and even of dissection. Successful reconstruction is facilitated by early treatment, before the dissection extends into the secondary renal artery branches. Because extension of the dissection is a continued threat and because the hypertension associated with this event is usually severe and difficult to control, we do not think that patients should uniformly be treated medically, waiting to see if spontaneous reversion to normotension will occur. (J VAsc SuRG 1991;14:468-79.)
Spontaneous dissection o f the renal artery is a rare phenomenon, representing less than one fourth of all From the Department of Surgery, University of California, San Francisco. Presented at the Sixth Annual Meeting of the Western Vascular Society, Palm Springs, Calif., Jan. 13-16, 1991. Reprint requests: Linda M. Reilly,MD, Moffitt 488, Department of Surgery, University of California, San Francisco, CA 94143. 24/6/30621 468
renal artery dissections. 1 It occurs in the absence o f trauma and commonly in the presence of mural disease, such as fibromuscular dysplasia (FMD), atherosclerosis, or aneurysmal degeneration. A m o n g reported cases, arterial reconstruction has seldom been attempted, and the outcome has frequently been poor. 2-12 Furthermore, spontaneous reversion to normotension has been reported a m o n g patients
Volume 14 Number 4 October 1991
treated medically. 4"5"7"s'13-15 As a result controversy exists about the role of operative reconstruction in the treatment of this disease. Since progress in the technique of renal artery repair now allows successful treatment of anatomically complex lesions, we reviewed our experience with arterial reconstruction in the management of spontaneous renal artery dissection to determine the frequency of and factors correlating with cure after operative repair. C L I N I C A L MATERIAL Ten patients (eight men, two women) with a mean age of 39.3 + 5.9 years have been treated for spontaneous renal artery dissection at the University of California, San Francisco. Three patients had preexisting hypertension, one had previously undergone a nephrectomy for unknown reasons, and none had experienced any antecedent trauma. All patients were admitted with either new onset of severe, intractable hypertension or marked exacerbation of their chronic hypertension. The mean blood pressure at admission was 205 + 36/134 + 18 mm Hg. In four patients this severe hypertension was associated with neurologic symptoms, including occipital headache, diplopia, blurred vision, and, in one case, hemiparesis. Microscopic hematuria, flank pain, and an abdominal or flank bruit were less common (Table I). Serum creatinine was usually normal as was intravenous pyelography (IVP). Split renal function tests and renal vein renins were infrequently performed but always showed tmilateral reduction of function and lateralization to the appropriate side. Arteriography demonstrated changes consistent with FMD in most patients; however, evidence of dissection was only present on slightly more than one half of the studies (Table II). Other angiographic interpretations included stenosis (N = 1), aneurysm formation (N = 2), or both changes (N = 3). Bilateral FMD, but not dissection, was present in three patients. Five patients had two renal arteries supplying the involved kidney, and in four of these patients both arteries were involved by the dissection. The main renal artery or arteries were always involved by the dissection, and extension into the primary renal artery branches was very common (Table II). TREATMENT
Seven patients underwent revascularization. A single renal artery was repaired in two cases, one of two renal arteries in three cases, and both of two involved renal arteries in the remaining two patients. Extensive branch renal artery involvement necessi-
Renal artery dissection 469
tated use of the ex vivo technique twice, 16 while the remaining arteries were repaired in situ. Arteries in two patients were reconstructed by primary reanastomosis after excision of the involved segment of renal artery. Arteries in five were reconstructed with a bypass or interposition graft of autogenous artery (hypogastric artery). Two patients tmderwent primary nephrectomy without any attempt at repair. In one case, exploration revealed that the dissection extended beyond the main renal artery bifurcation. At the time this patient was treated (1962), this was deemed an unreconstructible situation. In the second case, the preoperative arteriograrn showed an extensive intraparenchymal dissection that precluded repair. The final patient was treated with antihypertensive medication without undergoing any operative intervention. SELECTED CASE HISTORIES
Patient 3. A 36-year-old man was referred for evaluation of refractory hypertension of recent onset. He had been normotensive until 6 months before admission when he complained of blurred vision and pulsatite occipital headaches. Blood pressure then was 220/160 mm Hg. Treatment with propranolol, hydralazine, and prazosin afforded poor control (170/115 mm Hg). No history of trauma was elicited. On physical examination grade III hypertensive retinopathy was found. The peripheral pulses were normal without bruits. Urinalysis, serum creatinine, and IVP were normal, but renal vein renins lateralized markedly to the left side. On complete bed rest, salt restriction, and maximal medical therapy, there was no further improvement in blood pressure control (165/114 mm Hg). Arteriography was performed and demonstrated FMD with a dissection in the lower left renal artery (Fig. 1). At operation a thrill could be felt over the vessel. Significant periarterial fibrosis was present; however, when the artery was exposed, it showed a bluish discoloration typical of dissection. The involved segment was excised and repaired with a hypogastric artery interposition graft. Examination of the resected arterial segment showed an obvious dissection with compression of the true lumen by the aneurysmal false channel (Fig. 2). After operation the patient became asymptomatic and normotensive on no medications. A follow-up arteriogram obtained I year later shows •the normally functioning repair (Fig. 3). Currently, 13 years after operation, the patient remains cured of his hypertension. Patient 4. A 39-year-old man was referred for evaluation of new onset hypertension (270/130 mm
Journalof VASCULAR SURGERY
470 Reilly et al.
Table I. Patient profile Patient no.
Age
1 2 3 4 5 6 7 8 9 10
45 50 36 39 32 33 35 36 42 45
Gender Hypertension
M F M M M M M M F M
170/130* 240/150~ 220/160 270/130 ?~ 190/150 180/125 ? 200/110 170/115
Neurologic symptoms
Flank pain
Hematuria
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes
Serum creatinine
Normal Normal Normal Elevated Elevated Normal Normal Normal Normal Normal
I~
Normal Normal Normal Normal Smallright kidney Normal Small right kidney Normal Reducedfunction Normal
Renal vein renins
Not done Abnormal Abnormal Not done Not done Abnormal Not done Not done Not done Not done
IVP, Intravenous pyelography. ~Exacerbationof chronic hypertension.
Table II. Arteriographic findings No. of renal arteries
Fibromuscular dysplasia
1
2
Yes*
2 3 4 5 6 7
2 2 1 1 2 1
Yes~ Yes
8
2
9 10
1 1
Patient no.
Dissection
Yes Yes~ Yes Yes
Yes Yes~ Yes Yes
Involvement of renal artery branches
Yes (just at bifurcation) Yes Yes Yes
Yes (just at bifurcation) Yes Yes
Yes Yes
~Bilateral involvement. ~Lower renal artery only.
Hg), associated with throbbing occipital headaches and vague chest pain. N o history o f trauma was elicited, but the patient had undergone a left nephrectomy at the age o f 14 for reasons unknown to him. Results o f a CT brain scan were normal, as were IVP and urinary assay for catecholamines. Antihypertensive therapy was unsuccessful. At this point physical examination detected a right flank bruit, and the patient's serum creatinine was noted to be elevated (2.1 mg/dl). An arteriogram was obtained and showed dissection of the right renal artery, involving both o f the primary branches (Fig. 4). At operation the diagnosis o f dissection was confirmed by the intramural bluish discoloration o f the vessels. Microscopy o f the excised specimen clearly showed the dissection (Fig. 5). Ex vivo reconstruction was required because o f the branch involvement. The right hypogastric artery was used as a conduit, and its two primary branches were individually anastomosed to the two primary renal artery branches at the renal hilum. After operation the patient became normotensive and his creatinine fell to 1.6 mg/dl at the time o f
discharge from the hospital. His creatinine and blood pressure remain normal off all medications, 51/2 years later. A follow-up arteriogram obtained at that time demonstrates the normal functioning reconstruction (Fig. 6). Patient 7. A 35-year-old man was admitted for evaluation o f a 1-year history of hypertension (180/125 m m Hg), detected on routine physical examination. H e had been well except for longstanding chronic headaches. His physical examination was otherwise entirely unremarkable. Urinalysis demonstrated hematuria. Serum creatinine and urea nitrogen were normal. The IVP showed a small right kidney. Catecholamine levels were normal. An arteriogram was remarkable for right renal artery stenosis and changes suggestive o f FMD bilaterally. Dissection was not suggested by the arteriogram. The patient underwent exploration which showed dense periarterial inflammation. O n the right side the middle third o f the main renal artery was diffusely narrowed, while the distal third was aneurysmal. There was a palpable thrill and a 100 m m H g pressure
Volume 14 Number 4 October 1991
Renal artery dissection
471
i N~,%,,:!
}
~
Fig. 1. A, Arteriogram shows normal right renal artery and two left renal arteries with the lower one being abnormal. B, Early phase of selective injection shows aneurysmally dilated false channel. C, Late phase of selective injection shows delayed perfusion of lower pole branches.
A
~o~
1
B
inJ'-h 4
a
Fig. 2. Specimen of aneurysmal segment of resected artery from patient 3. A, The apparent lumen of the opened artery is actually the false channel of the dissection. B, This higher power view shows that the false lumen (a) occupies most of the vessel diameter. The original vessel wall (b) is collapsed such that the true lumen is a mere slit. Underneath the plane of dissection (c) organized hemorrhage separates the old wall from the false channel. (original magnification × 12.)
gradient across these lesions. Resection o f the abnormal artery revealed a double lumen that extended beyond the bifurcation o f the main renal artery. Reconstruction was judged to be impossible, and a
nephrectomy was performed. It was noted that the distal half o f the left renal artery was dilated and thickened, but had no measurable pressure gradient. After operation the patient had a transient period o f
Journal of VASCULAR SURGERY
472 Reilly et al.
Fig. 3. Follow-up arteriogram shows the normal appearing hypogastric artery bypass graft.
Fig. 4. Midstream aortogram (A) and selective right renal artery injection (B) show long, tapering stenosis of most of the main renal artery produced by compression of the true lumen by the dissection. The dilated segment of the distal renal artery is the aneurysmal false channel, and it extends into the primary bifurcation. renal insufficiency not requiting dialysis. Thereafter he became normotensive and remained that way for many years. Almost 29 years after operation mild hypertension developed and is easily controlled with minimal medication. Patient 10. A 45-year-old man, previously normotensive, was referred for evaluation of the sudden onset of left flank pain that radiated into his groin. Physical examination was entirely unremark-
able except that his blood pressure was now noted to be 170/115 mm Hg. Urinalysis revealed microscopic hematuria, and an IVP was normal. A renal scan demonstrated a defect in the upper pole of his left kidney. Arteriography showed extensive branch involvement by the dissection (Fig. 7). Treatment with antihypcrtensive medication was chosen, and the patient was begun on propanolol, apresoline, and thiazide diuretics. His blood pressure was reasonably
Volume 14 Number 4 October 1991
Renal artery dissection 473
A
Fig. 5. Cross-section of specimen from patient 7. A, The false lumen (a) compresses the smaller, but patent, true lumen (b). B, Higher power view shows that the outer elastic lamina (c) is separated from the adventitia (d) by a wedge of old organized hematoma (e) that has undergone fibroplasia. (original magnification × 120.)
Fig. 6. Late follow-up arteriogram (5 years after operation) shows the widely patent ex vivo repair, using a branched hypogastric artery interposition graft. Note stump of the left renal artery as a result of previous nephrectomy.
474
lournal of VASCULAR SURGERY
Reilly et al.
Fig. 7. Arteriogram shows diffuse involvement of the primary branches by dissection, with formation of multiple stenoses and aneurysms.
well controlled on this regimen (130/90 mm Hg). Subsequently he stopped taking all antihypertensive medications and adopted a rigorous program of diet and exercise. Currently, 10 years later, his diastolic blood pressure remains elevated at 100 mm Hg. RESULTS Five of seven (71.4%) attempted revascularizations were completed successfully. In two cases renal artery reconstruction failed to reverse extensive renal parenchymal ischemia, as determined by the gross appearance of the kidney and renal biopsy. In each case the surgeon then proceeded directly to nephrectomy. One of the two failed revascularizations was an attempted ex vivo repair. Of the five patients with successful revascularizations, three became normotensive and two returned to their previous level of blood pressure control. None of the five patients experienced any deterioration of blood pressure status or serum creatinine during a mean follow-up interval of 14.5 years (Table III). Furthermore, all five patients had postoperative arteriograms, and three of these patients had follow-up angiograms after a mean interval of 4.7 years. All studies showed patent, undiseased, functioning repairs (Figs. 3 and 6). The four patients who underwent nephrect o m y - t w o primarily and two after unsuccessful reconstruction attempts-initially became normotensire, and two have remained so. The third patientwho originally had bilateral renal artery involvement by F M D - remained normotensive for many years.
Now in his mid-sixties, he has begun to require medication (two agents) for the treatment of mild hypertension. Immediately after operation he had a significant rise in his creatinine (7.8 mg/dl), but this returned toward normal and currently is 1.5 to 1.9 mg/dl. He has not had repeat angiography to assess the status of his remaining kidney and renal vasculature. The fourth patient is also on antihypertensive medication (three agents) to maintain a normal blood pressure and has a normal creatinine. The mean follow-up interval for these patients is 14.6 years. Finally, the one patient in this group who did not undergo any operation was initially rendered normotensive by antihypertensive therapy (three agents). However, subsequent to stopping his medications, he became hypertensive with a diastolic blood pressure of 100 mm Hg. Ten years later he remains hypertensive with a normal creatinine and has not experienced a spontaneous remission. Histologic study of the involved renal arteries confirmed FMD in six cases and intramural dissection in all (N = 9) operative specimens. The dissection extended into the primary branches in eight patients and involved both renal arteries in four of the five patients with two renal arteries. DISCUSSION The natural history of spontaneous renal artery dissection is unknown. However, the most consistent clinical sequence has been the abrupt onset of severe, poorly controlled, persistent hypertension in previously normotensive young men. Among the approx-
Volume 14 Number 4
October i991
Renal artery dissection 475
Table III. Treatment and results Patient no.
Treatment
Technique
Conduit
1
Revascularization
In situ
Hypogastric artery
2
Revascularization
In situ
3
Revascularization Revascularization Revascularization
In situ
Excision + 1° reanastomosis Hypogastric artery Hypogastric artery Hypogastric artery
4 5
Ex vivo In sire
6
Nephrectomy*
7
Nephrectomy
8
Nephrectomy ~ Ex vivo
9 10
In situ
Excision + 1° reanastomosis Hypogastric artery
Nephrectomy Medication
Blood pressure
Creatinine
Pathology
Follow-up Follow-up ar~eriogram Interval ~*
Controlled at previous level Normal
Normal
FMD
Postoperative 8 years
24.0
Normal
FMD
Postoperative
27.5
Normal
Normal
FMD
13.0
Normal
Normal
FMD
Controlled at previous level Normal on medication
Normal
FMD
Postoperative i year Postoperative 5 years Postoperative
Normal
? FMD
Not done
11.0
Normal on medication Normal
1.5 to 1.9
Aneurysm
Not done
28.5
Normal
? FMD
Not done
13.0
Normal Elevated
Normal Normal
FMD Not done No Specimen Not done
5.0 10.0
5.5 2.7
*At the same operation, after failed revascularization. **In years.
imately 90 reported cases of spontaneous renal artery dissection (not including autopsy reports), 2-15'~7-25 only eight instances have been documented in which reversion to normal blood pressure occurred after varying intervals (2 to 45 m o n t h s ) . 4'5'7's'13-15 This clinical improvement was accompanied by the conversion of abnormal renin studies to normal in some patients and by an improved arteriographic appearance in others. 4'7"8'~3-~s Reentry of the dissection is postulated as the mechanism for this resolution of hypertension. 4'7 In addition, since "silent" dissections are occasionally discovered incidentally on arteriography, it has been suggested that resolution of dissection may be more frequent than implied by the above cases. However, only one instance of a completely silent, isolated renal artery dissection has been reported so far. 6 Therefore it seems that the overall incidence of spontaneous cure of the hypertension associated with primary renal artery dissection may be about 10% at most. It is important to note that there has been no report of spontaneous normalization of the renal artery anatomy after dissection. Although dramatic improvement has been demonstrated in one case by interval angiography, z5 in general the anatomic defect remains. In contrast to a clinical course characterized by improvement, worsening of blood pressure during diagnostic evaluation or initial attempts at medical therapy has been reported almost as frequently? '13'~8-21 In addition, at least nine patients
(including one in the present series) had clinical histories notable for repeated episodes of the symptoms associated with the renal artery dissection, sometimes extending over many months.* In one case, distal progression of the dissection and increased aneurysmal dilation of the false channel was documented over a 3-month period after the patient initially decfined operative intervention.l~ The combined incidence of clinical deterioration and multiple symptomatic episodes suggesting extension of the dissection is at least equal to if not slightly greater than the incidence of spontaneous return to normal blood pressure. In summary, most of the patients with spontaneous renal artery dissection face a future of long-term, multidrug therapy to control their hypertension, a situation that is unlikely to improve and may actually worsen. In view of the young patient age at the onset of this disease, another approach would seem warranted. The most frequent operative approach to renal artery dissection has been nephrectomy or partial nephrectomy, which has been required in almost half of the reported cases (38 of 85, 44.7%). Although removing the ischemic renal tissue is very effective in curing hypertension, the significant incidence of bilateral FMD (30% to 50%) as well as of dissection (20% to 25%), s'17makes this approach less attractive. In many of these cases the patients were explored *References 3, 4, 11, 13, 17, 18, 20, 22.
476
Journal of VASCULAR SURGERY
Reilly et al.
Table IV. Outcome of operative repair Outcome
Success-clinical Failure-overall clinical anatomic No follow-up
UCSF N (%)
Literature N (%)
5 (71.5) 2 (28.6)
11 (52.4) 7 (33.3) 6 (28.6) 1 (4.8) 3 (14.3)
2 (28.6)
Overall N (%)
16 9 6 3 3
(57.1) (32.1) (21.4) (10.7) (10.7)
with the intention of revascularization and renal salvage. The most common finding necessitating a change to the ablative approach was irreversible renal ischemia or extension of the dissection into branches, rendering reconstruction impossible. Arterial reconstruction has been attempted in 28 patients (32 arteries) with spontaneous renal artery dissection, including the seven patients in this report. Overall 29 of 32 repairs (90.6%) were anatomically successful; that is, a new conduit was successfully constructed. However, clinical success has been less consistent (Table IV). Among the completed reconstructions reported in the literature, only eight patients (38.1%) were subsequently normotensive, one of whom requires antihypertensive medication and one of whom was, in fact, normotensive before operation. 4"6-9"1°'~2Three additional patients (14.3%) demonstrated improved blood pressure control.12 In contrast all of the patients who had successful reconstruction in this series returned to their predissection blood pressure levels, without requiring additional medication. Undoubtedly, the key determinants of such a successful outcome are identification of correctable anatomic lesions, selection of the optimal operative technique and intraoperative assessment of the status of the reconstruction. It is evident from reviewing the reported cases of spontaneous renal artery dissection, that at the time of initial presentation, a substantial proportion of these patients are not candidates for renal salvage. Successful treatment of this disease requires restoration of normal arterial anatomy and normal renal perfusion. Indentificaton of a reconstruction candidate therefore requires precise and accurate assessment of the extent of the dissection and the extent of parenchymal infarction. High quality selective renal angiography is the cornerstone of this assessment. Dissections extending into intraparenchymal arterial branches or producing multiple branch occlusions cannot be repaired. Isotope renography is also helpful in determining the extent and distribution of renal infarction. Once a patient has been determined to have reconstructible anatomy, the appropriate operative
technique must be selected. This, too, depends mainly on the extent of the dissection. For the most part, main renal artery involvement-out to and including the first branch p o i n t - c a n usually be repaired in situ. If primary or secondary branches are involved, as long as the dissection remains extraparenchymai, the ex vivo technique allows tmcompromised exposure to get beyond the disease, as well as adequate time to perform a meticulous repair without unduly risking renal function. One might also select the ex vivo technique for main renal artery repair if the exposure is particulary poor (obese patient), if it is a repeat operation, or if there is a solitary kidney. Among the reported repairs, the ex vivo technique had been used seven times in five patients, with a successful outcome in six reconstruct. tions (four patients). 11'12 The only failure resulted from the inability to reverse parenchymal damage from long-standing ischemia, not from an inability to reconstruct the conduit. In view of the young age of these patients and the frequent necessity for branch repair, we feel strongly that autologous artery is the optimal conduit. The experience with use of other conduits would support this approach, since four of 10 saphenous vein and Dacron conduits reported in the literature have failed clinically. 2'3's Two additional venous conduits showed aneurysmal dilation only 4 years after implantation and have not been further studied. 12 In contrast, among the 11 cases of renal artery dissection treated with some form of autologous arterial reconstruction (hypogastric artery graft-5, excision and primary reanastomosis-3, splenorenal bypass-i, autotransplantation-1, autotransplantation + ex vivo repair-i), eight have been successful and only one has failed. 3'9"H No follow-up is available for the other two patients. ~° Including the primary branches when harvesting the hypogastric artery facilitates size-matched branch repair, and use of autogenous tissue seems to correlate with a durable reconstruction with minimal risk of late stenosis or occlusion. Once the repair has been completed, it is mandatory to determine if it has been successful. It is essential to recognize that a successful repair must reverse renal ischemia as well as establish a normal conduit. Intraoperative duplex scanning of the reconstruction has become an important tool in assessing the functional status of the repair. If the intraoperative assessment shows poor flow through the new conduit, evidence of high resistance to flow through the kidney, or the appearance of the kidney shows persistent areas of ischemia or infarction, then partial or complete nephrectomy should be undertaken. It is almost certain that the high clinical failure
Volume 14 Number 4 October 1991
rate of the attempted reconstructions reported in the fiterature reflects a failure to recognize persistent ischemic renal parenchyma. Using the approach summarized above has allowed us to identify w h e n repair is n o t possible or n o t successful. Consequently we have been able to p e r f o r m successful revascularization frequently (five o f seven, 71.4%), with restoration o f predissection b l o o d pressure (five o f five, 100%), consistent preservation o f renal function as indicated by the serum creatinine (five o f five, 100%), yielding sustained and durable clinical benefit (14.5-year
follow-up). Successful reconstruction is clearly facilitated by early treatment, before the dissection has extended into multiple branches and before significant infarction has occurred. Because extension o f dissection is a continued threat that m a y have irreversible consequences, and because spontaneous reversion to norm o t e n s i o n is u n c o m m o n , patients should n o t be uniformly treated medically. I f the patient is a g o o d candidate for operation, arterial reconstruction should be considered at the outset. Ultimately a successful, durable repair m a y be m o r e cost-effective and provide a better quality o f life than 3 or 4 decades o f multiagent antihypertensive therapy and continued risk o f kidney loss. REFERENCES
1. Foord AG, Lewis RD. Primary dissecting aneurysms of peripheral and pulmonary arteries: dissecting hemorrhage of media. Arch Pathol 1959;68:553-7. 2. Acconcia A, Manganelli A. Dissecting aneurysm of renal artery owing to subadventitial angioma. J Urol 1978;119: 268-70. 3. Bakir AA, Patel SK, Schwartz MM, Lewis EJ. Isolated dissecting aneurysm of the renal artery. Am Heart J 1978; 96:92-6. 4. Cummings KB, Lecky JW, Kanfman JJ. Renal artery aneurysms and hypertension. J Urol 1973;109:144-8. 5. Edwards BS, Stanson AW, Holley KE, Sheps SG. Isolated renal artery dissection. Presentation, evaluation, management, and pathology. Mayo Clin Proc 1982;57:564-71. 6. Gewertz BL, Stanley JC, Fry WJ. Renal artery dissections. Arch Surg 1977;112:409-14.
Renal artery dissection 477
7. Kaufman JJ, Coulson WF, Lecky JW, Popjak G. Primary dissecting aneurysm of renal artery: report of a case causing reversible renal hypertension. Ann Surg 1973;177:259-63. 8. Perry MO. Spontaneous renal artery dissection. I Cardiovasc Surg 1982;23:54-8. 9. Poutasse EF. Renal artery aneurysms. J Urol 1975;113: 443-9. 10. Tham G, Ekelund L, Herrlin K, Lindstedt EL, Olin T, Bergentz S-E. Renal artery aneurysms. Natural history and prognosis. Ann Surg 1983;197:348-52. 11. Javadpour N, Thorpe WP, Williams GM. Technique for extracorporeal resection of dissecting renal aneurysm with renal autotransplantation. Urology 1977; 10:145-7. 12. Smith BM, Holcomb GW, Richie RE, Dean RH. Renal artery dissection. Ann Surg 1984;200:134-46. 13. Hare WSC, Kincaid-Smith P. Dissecting aneurysm of the renal artery. Radiology 1970;97:255-63. 14. Paul M, Bear RA. Dissecting aneurysm of renal artery. Reversible cause of high-renin hypertension. Urology 1984; 24:483-4. 15. Mori H, Hayashi I, Tasald T, Hori T, YamasakiT, Amamoto Y. Spontaneous resolution of bilateral renal artery dissection: a case report. J Urol 1986;135:114-6. 16. Kent KC, Salvatierra O, Reilly LM, Ehrenfeld WK, Goldstone l, Stoney RI. Evolving strategies for the repair of complex renovascular lesions. Ann Surg 1987;206:272-8. 17. Rao CN, Blaivas JG. Primary renal artery dissecting aneurysm: a review. J Urol 1977;118:716-9. 18. GilfillanRS, Smart WR, Bostick WL. Dissecting aneurysm of the renal artery. Arch Surg 1956;73:737-40. 19. Meyers DS, Grim CE, Keitzer WF. Fibromuscular dysplasia of the renal artery with medial dissection. A case simulating polyarteritis nodosa. Am J Med 1974;56:412-6. 20. Khanna OP, Nedwich A, Gonick P. Accelerated hypertension due to intramural dissection of accessoryrenal artery. Urology i973;1:130-3. 21. Goldfarb R, Pool JL, Wheeler T. Isolated renal artery dissection secondary to medial degeneration. J Urol 1988; 139:346-7. 22. Guthrie W, Maclean H. Dissecting aneurysms of arteries other than the aorta. I Pathol 1972;108:219-35. 23. Neuman HW, Sahin AF. Unilateral dissecting aneurysm of the renal artery. Can ]" Surg 1965;8:291-5. 24. Hasday JD, Sterns RH, Karch FE. Renal infarction due to renal artery dysplasia with dissection. Report of a case in a normotensive patient. Am J Med 1984;76:943-6. 25. Bumpus HC lr. A case of renal hypertension. J Urol 1944;52:295-9. Submitted Feb. 7, 1991; accepted May 3, 1991.
DISCUSSION
Dr. Wesley S. Moore (Los Angeles, Calif.). Dr. Cunningham and his colleagues have addressed a very rare problem encountered by vascular surgeons, spontaneous renal artery dissection. As such, a review of the UCSF experience with 10 patients will be particularly helpful to those of us who will face the management of the occasional
patient that will be seen in our practice. Their review of the literature has clearly documented that the natural history of this disease is unsatisfactory. Patients will either require multiple drug therapy for blood pressure control or nephrectomy. In contrast, the five patients in their series who underwent direct revascularization ended up with an
478
Reilly et al.
excellent clinical result that was sustained over a long follow-up. An additional four patients underwent exploration and immediate nephrectomy when it was deemed that the vessd was not reconstructible. These patients also had an excellent result and are normotensive for a follow-up of more than 14 years. Finally, the last patient who was treated medically remains hypertensive at the time of most recent follow-up. I suspect that the data presented may also be extrapolated to traumatic renal artery dissection. Perhaps one of the more common mechanisms for renal artery dissection seen today is a complication of balloon angioplasty. Implicit in the authors' presentation was the importance of the technique used for renal artery repair. Four of the five patients underwent renal artery replacement by means of the hypogastric artery, and one patient had a resection with primary anastomosis. Although this experience reflects the preference of the UCSF group for the hypogastric artery, it may tend to be written off by others in favor of their own preferences such as vein grafting or patch angioplasty. That may be a mistake. I would like to offer one anecdotal case of my own to emphasize this point. Approximately 3 years ago I had a patient transferred to my service who had a typical lesion of FMD involving her right renal artery that was treated by balloon angioplasty at another hospital. She did not have a good blood pressure response, and therefore a repeat angiogram was performed. The repeat angiogram demonstrated a dissection in the dilated segment, and she was transferred to my service. We explored her artery and elected to treat this by removing the dissected intima by open intimectomy and patch angioplasty. We had an excellent technical result with return of normal Doppler signals in primary and secondary branches. The patient had prompt return to normal blood pressure without the need for any further pharmacologic control. She remained normotensive for approximately 6 months and then had a return of her hypertension, We brought her back fbr repeat angiogram and were disappointed to find that the segment treated by intimectomy had now strictured down to involve the entire primary renal artery as well as the orifices of all of the secondary branches. It appeared to be a typical myointimal hyperplastic response. She ultimately underwent nephrectomy with cure of her hypertension. While I do not want to generalize from a single case because it may well be that the primary segment of renal artery damaged by spontaneous dissection will not respond well to anything other than resection and replacement. I suspect that the excellent results achieved by the UCSF group are based on their decision to treat these with resection and hypogastric artery repair. Whether or not there would be any difference with a vein graft is speculative. However, I am quite sure that repair of the damaged renal artery is ill advised. In closing I would like to ask two questions: (1) What clinical parameters do you use to suspect this lesion and recommend arteriography? (2) Have you looked at your own experience with traumatic dissection, and are the
Journal of VASCULAR SURGERY
surgical results comparable to those achieved with spontaneous dissection? Dr. Chris G. Cunningham. Thank you Dr. Moore. One of the aspects of the natural history of this illness is that it generally is of such a sudden onset and generates such a marked hypertension that even our medical colleagues are alerted to these patients. I suspect that it is the magnitude and severity of the hypertension in an otherwise previously healthy young patient that is the key. We have not reviewed the cases of traumatic renal artery dissection at our institution, because this type of case is rarely seen at our hospital. Dr. Anthony J. R o o n (Everett, Wash.). I'd like to ask the authors a question about the surgical management of the renal artery. I have had the oppornmity to treat three traumatic renal artery dissections, and in examining these patients, one often finds that the hematoma extends on to the branch arteries. However, conventional wisdom has been that if you repair the primary area of dissection and sew the distal intima to the media and adventitia with the repair, the remaining hematoma will disperse and heal. This has been our experience. When you say that certain arteries are not reconstructible by extension of the dissection onto the branches or a certain distance onto the branches, I would like to have you explain to the physicians in this audience who may treat this problem how you decide exactly which lesions are reconstructible and which are not. My experience has been that if yon get rid of the initial tear in traumatic dissections and reconstruct it appropriately that it will heal. Dr. Cunningham. We believe that as long as dissections extend into primary and secondary branch arteries, they are reconstructible. Those that we think are not reconstructible involve extension into the intraparenchymal renal arteries. Dr. Linda M. Reilly. I am actually not certain that this is a setting in which you can limit your repair only to eliminating the site of the tear. I have a suspicion that that is probably one of the reasons why other repairs reported in the literature failed. It may be quite different to repair a large vessel with a tear in it as opposed to repairing a 2 mm or 3 mm branch that has a tear. And the elimination of the false channel in a durable manner is not as predictable. The operative experience with these is often that the septum is very fragile and difficult to sew down. Therefore, trying to work in the diseased vessel may very well be condemning the patient to a less optimal result subsequently. Dr. Edward E. Levine (Orange, Calif.). I support Dr. Reilly's comments. I have experience with one case in which the end point was tacked down or there was a small flap still present, and we thought it was adequate. On the table the artery redissected and went out into all the primary and secondary branches. Dr. Allan D. Callow. (St. Louis, Mo.). I have two questions: the first is, knowing the proclivity for FMD to
Volume 14 Number 4 October 1991
Renal artery dissection 479
be or to become bilateral, I wonder about the long-term follow-up concerning the second side. What were your findings on the second and presumably asymptomatic side? What do you recommend concerning it when present? Was the false hmaen demonstrated in the enlarged version of the specimen you showed us typical of all of your lesions? In some carotid dissections a hematoma may develop deep within the media or in a subadventitial plane. It may, in other instances, be in a subintimal plane, and the method of treatment may be different based on differing etiologies, for example, traumatic versus spontaneous. Was the lesion which you showed us always characteristic in your series? Dr. Ctmningham. I do not know if all the dissections occurred in precisely the same plane. However, this appearance was very consistent in the five or six specimens available for review. None of our patients with bilateral FMD have shown clinical or radiographic evidence of
subsequent dissection in the contralateral renal artery. However, the number of these patients (N = 3) is very small. Dr. Ronald J. Stoney (San Francisco, Calif.). I think Dr. Callow is reminding us about, the complexity of an arterial dissection. These were very complex dissections. They seemed to go between layers, and the dissection was not always contiguous in one layer in the media. The dissection may form a fissure in the media and develop in several different planes making it difficult for a local repair which uses end point tacking as the primary objective. Our surgical objective has been to get beyond the dissection to a normal branch, to transect that, and then attach to that the replacement arterial graft. The graft is then anastomosed to the aorta restoring flow to the kidney.
LIEBIG F O U N D A T I O N AWARD FOR VASCULAR SURGICAL RESEARCH, 1992 The Liebig Foundation announces the tenth annual competitive award of $5000 for the best essay on a problem in general vascular surgery. The investigative work shall be: 1. Clinical or experimental research 2. Original and unpublished (nor submitted elsewhere for publication) 3. Performed by a house officer in the United States, Canada, or Mexico with senior collaborators acting in a consultive capacity 4. Submitted in English (6 copies of typed manuscript and 6 copies of glossy prints of illustrations) 5. Accompanied by a signed letter from the essayist's superior confirming the status of the essayist and complying with "Instructions to Authors" of the JOURNAL Previous winners were: 1990, Scott A. Berceli, MD, University of Pittsburgh School of Medicine, Pittsburgh, Pa.; 1989, Michael L. Marin, MD, College of Physicians and Surgeons, Columbia University, N.Y. The winning essay will be judged for presentation at the June 1991 meeting of the North American Chapter, International Society for Cardiovascular Surgery and will be considered for publication in the JOURNAL Further inquiries may be directed to the same address to which the essays must be sent, postmarked before Dec. 31, 1991: Dr. Richard Turner Award Committee Secretary 112 Bauer Drive Oakland, NJ 07436 USA (201) 337-6126
Spontaneous dissection o f the renal artery is a rare phenomenon, representing less than one fourth of all From the Department of Surgery, University of California, San Francisco. Presented at the Sixth Annual Meeting of the Western Vascular Society, Palm Springs, Calif., Jan. 13-16, 1991. Reprint requests: Linda M. Reilly,MD, Moffitt 488, Department of Surgery, University of California, San Francisco, CA 94143. 24/6/30621 468
renal artery dissections. 1 It occurs in the absence o f trauma and commonly in the presence of mural disease, such as fibromuscular dysplasia (FMD), atherosclerosis, or aneurysmal degeneration. A m o n g reported cases, arterial reconstruction has seldom been attempted, and the outcome has frequently been poor. 2-12 Furthermore, spontaneous reversion to normotension has been reported a m o n g patients
Volume 14 Number 4 October 1991
treated medically. 4"5"7"s'13-15 As a result controversy exists about the role of operative reconstruction in the treatment of this disease. Since progress in the technique of renal artery repair now allows successful treatment of anatomically complex lesions, we reviewed our experience with arterial reconstruction in the management of spontaneous renal artery dissection to determine the frequency of and factors correlating with cure after operative repair. C L I N I C A L MATERIAL Ten patients (eight men, two women) with a mean age of 39.3 + 5.9 years have been treated for spontaneous renal artery dissection at the University of California, San Francisco. Three patients had preexisting hypertension, one had previously undergone a nephrectomy for unknown reasons, and none had experienced any antecedent trauma. All patients were admitted with either new onset of severe, intractable hypertension or marked exacerbation of their chronic hypertension. The mean blood pressure at admission was 205 + 36/134 + 18 mm Hg. In four patients this severe hypertension was associated with neurologic symptoms, including occipital headache, diplopia, blurred vision, and, in one case, hemiparesis. Microscopic hematuria, flank pain, and an abdominal or flank bruit were less common (Table I). Serum creatinine was usually normal as was intravenous pyelography (IVP). Split renal function tests and renal vein renins were infrequently performed but always showed tmilateral reduction of function and lateralization to the appropriate side. Arteriography demonstrated changes consistent with FMD in most patients; however, evidence of dissection was only present on slightly more than one half of the studies (Table II). Other angiographic interpretations included stenosis (N = 1), aneurysm formation (N = 2), or both changes (N = 3). Bilateral FMD, but not dissection, was present in three patients. Five patients had two renal arteries supplying the involved kidney, and in four of these patients both arteries were involved by the dissection. The main renal artery or arteries were always involved by the dissection, and extension into the primary renal artery branches was very common (Table II). TREATMENT
Seven patients underwent revascularization. A single renal artery was repaired in two cases, one of two renal arteries in three cases, and both of two involved renal arteries in the remaining two patients. Extensive branch renal artery involvement necessi-
Renal artery dissection 469
tated use of the ex vivo technique twice, 16 while the remaining arteries were repaired in situ. Arteries in two patients were reconstructed by primary reanastomosis after excision of the involved segment of renal artery. Arteries in five were reconstructed with a bypass or interposition graft of autogenous artery (hypogastric artery). Two patients tmderwent primary nephrectomy without any attempt at repair. In one case, exploration revealed that the dissection extended beyond the main renal artery bifurcation. At the time this patient was treated (1962), this was deemed an unreconstructible situation. In the second case, the preoperative arteriograrn showed an extensive intraparenchymal dissection that precluded repair. The final patient was treated with antihypertensive medication without undergoing any operative intervention. SELECTED CASE HISTORIES
Patient 3. A 36-year-old man was referred for evaluation of refractory hypertension of recent onset. He had been normotensive until 6 months before admission when he complained of blurred vision and pulsatite occipital headaches. Blood pressure then was 220/160 mm Hg. Treatment with propranolol, hydralazine, and prazosin afforded poor control (170/115 mm Hg). No history of trauma was elicited. On physical examination grade III hypertensive retinopathy was found. The peripheral pulses were normal without bruits. Urinalysis, serum creatinine, and IVP were normal, but renal vein renins lateralized markedly to the left side. On complete bed rest, salt restriction, and maximal medical therapy, there was no further improvement in blood pressure control (165/114 mm Hg). Arteriography was performed and demonstrated FMD with a dissection in the lower left renal artery (Fig. 1). At operation a thrill could be felt over the vessel. Significant periarterial fibrosis was present; however, when the artery was exposed, it showed a bluish discoloration typical of dissection. The involved segment was excised and repaired with a hypogastric artery interposition graft. Examination of the resected arterial segment showed an obvious dissection with compression of the true lumen by the aneurysmal false channel (Fig. 2). After operation the patient became asymptomatic and normotensive on no medications. A follow-up arteriogram obtained I year later shows •the normally functioning repair (Fig. 3). Currently, 13 years after operation, the patient remains cured of his hypertension. Patient 4. A 39-year-old man was referred for evaluation of new onset hypertension (270/130 mm
Journalof VASCULAR SURGERY
470 Reilly et al.
Table I. Patient profile Patient no.
Age
1 2 3 4 5 6 7 8 9 10
45 50 36 39 32 33 35 36 42 45
Gender Hypertension
M F M M M M M M F M
170/130* 240/150~ 220/160 270/130 ?~ 190/150 180/125 ? 200/110 170/115
Neurologic symptoms
Flank pain
Hematuria
Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes
Serum creatinine
Normal Normal Normal Elevated Elevated Normal Normal Normal Normal Normal
I~
Normal Normal Normal Normal Smallright kidney Normal Small right kidney Normal Reducedfunction Normal
Renal vein renins
Not done Abnormal Abnormal Not done Not done Abnormal Not done Not done Not done Not done
IVP, Intravenous pyelography. ~Exacerbationof chronic hypertension.
Table II. Arteriographic findings No. of renal arteries
Fibromuscular dysplasia
1
2
Yes*
2 3 4 5 6 7
2 2 1 1 2 1
Yes~ Yes
8
2
9 10
1 1
Patient no.
Dissection
Yes Yes~ Yes Yes
Yes Yes~ Yes Yes
Involvement of renal artery branches
Yes (just at bifurcation) Yes Yes Yes
Yes (just at bifurcation) Yes Yes
Yes Yes
~Bilateral involvement. ~Lower renal artery only.
Hg), associated with throbbing occipital headaches and vague chest pain. N o history o f trauma was elicited, but the patient had undergone a left nephrectomy at the age o f 14 for reasons unknown to him. Results o f a CT brain scan were normal, as were IVP and urinary assay for catecholamines. Antihypertensive therapy was unsuccessful. At this point physical examination detected a right flank bruit, and the patient's serum creatinine was noted to be elevated (2.1 mg/dl). An arteriogram was obtained and showed dissection of the right renal artery, involving both o f the primary branches (Fig. 4). At operation the diagnosis o f dissection was confirmed by the intramural bluish discoloration o f the vessels. Microscopy o f the excised specimen clearly showed the dissection (Fig. 5). Ex vivo reconstruction was required because o f the branch involvement. The right hypogastric artery was used as a conduit, and its two primary branches were individually anastomosed to the two primary renal artery branches at the renal hilum. After operation the patient became normotensive and his creatinine fell to 1.6 mg/dl at the time o f
discharge from the hospital. His creatinine and blood pressure remain normal off all medications, 51/2 years later. A follow-up arteriogram obtained at that time demonstrates the normal functioning reconstruction (Fig. 6). Patient 7. A 35-year-old man was admitted for evaluation o f a 1-year history of hypertension (180/125 m m Hg), detected on routine physical examination. H e had been well except for longstanding chronic headaches. His physical examination was otherwise entirely unremarkable. Urinalysis demonstrated hematuria. Serum creatinine and urea nitrogen were normal. The IVP showed a small right kidney. Catecholamine levels were normal. An arteriogram was remarkable for right renal artery stenosis and changes suggestive o f FMD bilaterally. Dissection was not suggested by the arteriogram. The patient underwent exploration which showed dense periarterial inflammation. O n the right side the middle third o f the main renal artery was diffusely narrowed, while the distal third was aneurysmal. There was a palpable thrill and a 100 m m H g pressure
Volume 14 Number 4 October 1991
Renal artery dissection
471
i N~,%,,:!
}
~
Fig. 1. A, Arteriogram shows normal right renal artery and two left renal arteries with the lower one being abnormal. B, Early phase of selective injection shows aneurysmally dilated false channel. C, Late phase of selective injection shows delayed perfusion of lower pole branches.
A
~o~
1
B
inJ'-h 4
a
Fig. 2. Specimen of aneurysmal segment of resected artery from patient 3. A, The apparent lumen of the opened artery is actually the false channel of the dissection. B, This higher power view shows that the false lumen (a) occupies most of the vessel diameter. The original vessel wall (b) is collapsed such that the true lumen is a mere slit. Underneath the plane of dissection (c) organized hemorrhage separates the old wall from the false channel. (original magnification × 12.)
gradient across these lesions. Resection o f the abnormal artery revealed a double lumen that extended beyond the bifurcation o f the main renal artery. Reconstruction was judged to be impossible, and a
nephrectomy was performed. It was noted that the distal half o f the left renal artery was dilated and thickened, but had no measurable pressure gradient. After operation the patient had a transient period o f
Journal of VASCULAR SURGERY
472 Reilly et al.
Fig. 3. Follow-up arteriogram shows the normal appearing hypogastric artery bypass graft.
Fig. 4. Midstream aortogram (A) and selective right renal artery injection (B) show long, tapering stenosis of most of the main renal artery produced by compression of the true lumen by the dissection. The dilated segment of the distal renal artery is the aneurysmal false channel, and it extends into the primary bifurcation. renal insufficiency not requiting dialysis. Thereafter he became normotensive and remained that way for many years. Almost 29 years after operation mild hypertension developed and is easily controlled with minimal medication. Patient 10. A 45-year-old man, previously normotensive, was referred for evaluation of the sudden onset of left flank pain that radiated into his groin. Physical examination was entirely unremark-
able except that his blood pressure was now noted to be 170/115 mm Hg. Urinalysis revealed microscopic hematuria, and an IVP was normal. A renal scan demonstrated a defect in the upper pole of his left kidney. Arteriography showed extensive branch involvement by the dissection (Fig. 7). Treatment with antihypcrtensive medication was chosen, and the patient was begun on propanolol, apresoline, and thiazide diuretics. His blood pressure was reasonably
Volume 14 Number 4 October 1991
Renal artery dissection 473
A
Fig. 5. Cross-section of specimen from patient 7. A, The false lumen (a) compresses the smaller, but patent, true lumen (b). B, Higher power view shows that the outer elastic lamina (c) is separated from the adventitia (d) by a wedge of old organized hematoma (e) that has undergone fibroplasia. (original magnification × 120.)
Fig. 6. Late follow-up arteriogram (5 years after operation) shows the widely patent ex vivo repair, using a branched hypogastric artery interposition graft. Note stump of the left renal artery as a result of previous nephrectomy.
474
lournal of VASCULAR SURGERY
Reilly et al.
Fig. 7. Arteriogram shows diffuse involvement of the primary branches by dissection, with formation of multiple stenoses and aneurysms.
well controlled on this regimen (130/90 mm Hg). Subsequently he stopped taking all antihypertensive medications and adopted a rigorous program of diet and exercise. Currently, 10 years later, his diastolic blood pressure remains elevated at 100 mm Hg. RESULTS Five of seven (71.4%) attempted revascularizations were completed successfully. In two cases renal artery reconstruction failed to reverse extensive renal parenchymal ischemia, as determined by the gross appearance of the kidney and renal biopsy. In each case the surgeon then proceeded directly to nephrectomy. One of the two failed revascularizations was an attempted ex vivo repair. Of the five patients with successful revascularizations, three became normotensive and two returned to their previous level of blood pressure control. None of the five patients experienced any deterioration of blood pressure status or serum creatinine during a mean follow-up interval of 14.5 years (Table III). Furthermore, all five patients had postoperative arteriograms, and three of these patients had follow-up angiograms after a mean interval of 4.7 years. All studies showed patent, undiseased, functioning repairs (Figs. 3 and 6). The four patients who underwent nephrect o m y - t w o primarily and two after unsuccessful reconstruction attempts-initially became normotensire, and two have remained so. The third patientwho originally had bilateral renal artery involvement by F M D - remained normotensive for many years.
Now in his mid-sixties, he has begun to require medication (two agents) for the treatment of mild hypertension. Immediately after operation he had a significant rise in his creatinine (7.8 mg/dl), but this returned toward normal and currently is 1.5 to 1.9 mg/dl. He has not had repeat angiography to assess the status of his remaining kidney and renal vasculature. The fourth patient is also on antihypertensive medication (three agents) to maintain a normal blood pressure and has a normal creatinine. The mean follow-up interval for these patients is 14.6 years. Finally, the one patient in this group who did not undergo any operation was initially rendered normotensive by antihypertensive therapy (three agents). However, subsequent to stopping his medications, he became hypertensive with a diastolic blood pressure of 100 mm Hg. Ten years later he remains hypertensive with a normal creatinine and has not experienced a spontaneous remission. Histologic study of the involved renal arteries confirmed FMD in six cases and intramural dissection in all (N = 9) operative specimens. The dissection extended into the primary branches in eight patients and involved both renal arteries in four of the five patients with two renal arteries. DISCUSSION The natural history of spontaneous renal artery dissection is unknown. However, the most consistent clinical sequence has been the abrupt onset of severe, poorly controlled, persistent hypertension in previously normotensive young men. Among the approx-
Volume 14 Number 4
October i991
Renal artery dissection 475
Table III. Treatment and results Patient no.
Treatment
Technique
Conduit
1
Revascularization
In situ
Hypogastric artery
2
Revascularization
In situ
3
Revascularization Revascularization Revascularization
In situ
Excision + 1° reanastomosis Hypogastric artery Hypogastric artery Hypogastric artery
4 5
Ex vivo In sire
6
Nephrectomy*
7
Nephrectomy
8
Nephrectomy ~ Ex vivo
9 10
In situ
Excision + 1° reanastomosis Hypogastric artery
Nephrectomy Medication
Blood pressure
Creatinine
Pathology
Follow-up Follow-up ar~eriogram Interval ~*
Controlled at previous level Normal
Normal
FMD
Postoperative 8 years
24.0
Normal
FMD
Postoperative
27.5
Normal
Normal
FMD
13.0
Normal
Normal
FMD
Controlled at previous level Normal on medication
Normal
FMD
Postoperative i year Postoperative 5 years Postoperative
Normal
? FMD
Not done
11.0
Normal on medication Normal
1.5 to 1.9
Aneurysm
Not done
28.5
Normal
? FMD
Not done
13.0
Normal Elevated
Normal Normal
FMD Not done No Specimen Not done
5.0 10.0
5.5 2.7
*At the same operation, after failed revascularization. **In years.
imately 90 reported cases of spontaneous renal artery dissection (not including autopsy reports), 2-15'~7-25 only eight instances have been documented in which reversion to normal blood pressure occurred after varying intervals (2 to 45 m o n t h s ) . 4'5'7's'13-15 This clinical improvement was accompanied by the conversion of abnormal renin studies to normal in some patients and by an improved arteriographic appearance in others. 4'7"8'~3-~s Reentry of the dissection is postulated as the mechanism for this resolution of hypertension. 4'7 In addition, since "silent" dissections are occasionally discovered incidentally on arteriography, it has been suggested that resolution of dissection may be more frequent than implied by the above cases. However, only one instance of a completely silent, isolated renal artery dissection has been reported so far. 6 Therefore it seems that the overall incidence of spontaneous cure of the hypertension associated with primary renal artery dissection may be about 10% at most. It is important to note that there has been no report of spontaneous normalization of the renal artery anatomy after dissection. Although dramatic improvement has been demonstrated in one case by interval angiography, z5 in general the anatomic defect remains. In contrast to a clinical course characterized by improvement, worsening of blood pressure during diagnostic evaluation or initial attempts at medical therapy has been reported almost as frequently? '13'~8-21 In addition, at least nine patients
(including one in the present series) had clinical histories notable for repeated episodes of the symptoms associated with the renal artery dissection, sometimes extending over many months.* In one case, distal progression of the dissection and increased aneurysmal dilation of the false channel was documented over a 3-month period after the patient initially decfined operative intervention.l~ The combined incidence of clinical deterioration and multiple symptomatic episodes suggesting extension of the dissection is at least equal to if not slightly greater than the incidence of spontaneous return to normal blood pressure. In summary, most of the patients with spontaneous renal artery dissection face a future of long-term, multidrug therapy to control their hypertension, a situation that is unlikely to improve and may actually worsen. In view of the young patient age at the onset of this disease, another approach would seem warranted. The most frequent operative approach to renal artery dissection has been nephrectomy or partial nephrectomy, which has been required in almost half of the reported cases (38 of 85, 44.7%). Although removing the ischemic renal tissue is very effective in curing hypertension, the significant incidence of bilateral FMD (30% to 50%) as well as of dissection (20% to 25%), s'17makes this approach less attractive. In many of these cases the patients were explored *References 3, 4, 11, 13, 17, 18, 20, 22.
476
Journal of VASCULAR SURGERY
Reilly et al.
Table IV. Outcome of operative repair Outcome
Success-clinical Failure-overall clinical anatomic No follow-up
UCSF N (%)
Literature N (%)
5 (71.5) 2 (28.6)
11 (52.4) 7 (33.3) 6 (28.6) 1 (4.8) 3 (14.3)
2 (28.6)
Overall N (%)
16 9 6 3 3
(57.1) (32.1) (21.4) (10.7) (10.7)
with the intention of revascularization and renal salvage. The most common finding necessitating a change to the ablative approach was irreversible renal ischemia or extension of the dissection into branches, rendering reconstruction impossible. Arterial reconstruction has been attempted in 28 patients (32 arteries) with spontaneous renal artery dissection, including the seven patients in this report. Overall 29 of 32 repairs (90.6%) were anatomically successful; that is, a new conduit was successfully constructed. However, clinical success has been less consistent (Table IV). Among the completed reconstructions reported in the literature, only eight patients (38.1%) were subsequently normotensive, one of whom requires antihypertensive medication and one of whom was, in fact, normotensive before operation. 4"6-9"1°'~2Three additional patients (14.3%) demonstrated improved blood pressure control.12 In contrast all of the patients who had successful reconstruction in this series returned to their predissection blood pressure levels, without requiring additional medication. Undoubtedly, the key determinants of such a successful outcome are identification of correctable anatomic lesions, selection of the optimal operative technique and intraoperative assessment of the status of the reconstruction. It is evident from reviewing the reported cases of spontaneous renal artery dissection, that at the time of initial presentation, a substantial proportion of these patients are not candidates for renal salvage. Successful treatment of this disease requires restoration of normal arterial anatomy and normal renal perfusion. Indentificaton of a reconstruction candidate therefore requires precise and accurate assessment of the extent of the dissection and the extent of parenchymal infarction. High quality selective renal angiography is the cornerstone of this assessment. Dissections extending into intraparenchymal arterial branches or producing multiple branch occlusions cannot be repaired. Isotope renography is also helpful in determining the extent and distribution of renal infarction. Once a patient has been determined to have reconstructible anatomy, the appropriate operative
technique must be selected. This, too, depends mainly on the extent of the dissection. For the most part, main renal artery involvement-out to and including the first branch p o i n t - c a n usually be repaired in situ. If primary or secondary branches are involved, as long as the dissection remains extraparenchymai, the ex vivo technique allows tmcompromised exposure to get beyond the disease, as well as adequate time to perform a meticulous repair without unduly risking renal function. One might also select the ex vivo technique for main renal artery repair if the exposure is particulary poor (obese patient), if it is a repeat operation, or if there is a solitary kidney. Among the reported repairs, the ex vivo technique had been used seven times in five patients, with a successful outcome in six reconstruct. tions (four patients). 11'12 The only failure resulted from the inability to reverse parenchymal damage from long-standing ischemia, not from an inability to reconstruct the conduit. In view of the young age of these patients and the frequent necessity for branch repair, we feel strongly that autologous artery is the optimal conduit. The experience with use of other conduits would support this approach, since four of 10 saphenous vein and Dacron conduits reported in the literature have failed clinically. 2'3's Two additional venous conduits showed aneurysmal dilation only 4 years after implantation and have not been further studied. 12 In contrast, among the 11 cases of renal artery dissection treated with some form of autologous arterial reconstruction (hypogastric artery graft-5, excision and primary reanastomosis-3, splenorenal bypass-i, autotransplantation-1, autotransplantation + ex vivo repair-i), eight have been successful and only one has failed. 3'9"H No follow-up is available for the other two patients. ~° Including the primary branches when harvesting the hypogastric artery facilitates size-matched branch repair, and use of autogenous tissue seems to correlate with a durable reconstruction with minimal risk of late stenosis or occlusion. Once the repair has been completed, it is mandatory to determine if it has been successful. It is essential to recognize that a successful repair must reverse renal ischemia as well as establish a normal conduit. Intraoperative duplex scanning of the reconstruction has become an important tool in assessing the functional status of the repair. If the intraoperative assessment shows poor flow through the new conduit, evidence of high resistance to flow through the kidney, or the appearance of the kidney shows persistent areas of ischemia or infarction, then partial or complete nephrectomy should be undertaken. It is almost certain that the high clinical failure
Volume 14 Number 4 October 1991
rate of the attempted reconstructions reported in the fiterature reflects a failure to recognize persistent ischemic renal parenchyma. Using the approach summarized above has allowed us to identify w h e n repair is n o t possible or n o t successful. Consequently we have been able to p e r f o r m successful revascularization frequently (five o f seven, 71.4%), with restoration o f predissection b l o o d pressure (five o f five, 100%), consistent preservation o f renal function as indicated by the serum creatinine (five o f five, 100%), yielding sustained and durable clinical benefit (14.5-year
follow-up). Successful reconstruction is clearly facilitated by early treatment, before the dissection has extended into multiple branches and before significant infarction has occurred. Because extension o f dissection is a continued threat that m a y have irreversible consequences, and because spontaneous reversion to norm o t e n s i o n is u n c o m m o n , patients should n o t be uniformly treated medically. I f the patient is a g o o d candidate for operation, arterial reconstruction should be considered at the outset. Ultimately a successful, durable repair m a y be m o r e cost-effective and provide a better quality o f life than 3 or 4 decades o f multiagent antihypertensive therapy and continued risk o f kidney loss. REFERENCES
1. Foord AG, Lewis RD. Primary dissecting aneurysms of peripheral and pulmonary arteries: dissecting hemorrhage of media. Arch Pathol 1959;68:553-7. 2. Acconcia A, Manganelli A. Dissecting aneurysm of renal artery owing to subadventitial angioma. J Urol 1978;119: 268-70. 3. Bakir AA, Patel SK, Schwartz MM, Lewis EJ. Isolated dissecting aneurysm of the renal artery. Am Heart J 1978; 96:92-6. 4. Cummings KB, Lecky JW, Kanfman JJ. Renal artery aneurysms and hypertension. J Urol 1973;109:144-8. 5. Edwards BS, Stanson AW, Holley KE, Sheps SG. Isolated renal artery dissection. Presentation, evaluation, management, and pathology. Mayo Clin Proc 1982;57:564-71. 6. Gewertz BL, Stanley JC, Fry WJ. Renal artery dissections. Arch Surg 1977;112:409-14.
Renal artery dissection 477
7. Kaufman JJ, Coulson WF, Lecky JW, Popjak G. Primary dissecting aneurysm of renal artery: report of a case causing reversible renal hypertension. Ann Surg 1973;177:259-63. 8. Perry MO. Spontaneous renal artery dissection. I Cardiovasc Surg 1982;23:54-8. 9. Poutasse EF. Renal artery aneurysms. J Urol 1975;113: 443-9. 10. Tham G, Ekelund L, Herrlin K, Lindstedt EL, Olin T, Bergentz S-E. Renal artery aneurysms. Natural history and prognosis. Ann Surg 1983;197:348-52. 11. Javadpour N, Thorpe WP, Williams GM. Technique for extracorporeal resection of dissecting renal aneurysm with renal autotransplantation. Urology 1977; 10:145-7. 12. Smith BM, Holcomb GW, Richie RE, Dean RH. Renal artery dissection. Ann Surg 1984;200:134-46. 13. Hare WSC, Kincaid-Smith P. Dissecting aneurysm of the renal artery. Radiology 1970;97:255-63. 14. Paul M, Bear RA. Dissecting aneurysm of renal artery. Reversible cause of high-renin hypertension. Urology 1984; 24:483-4. 15. Mori H, Hayashi I, Tasald T, Hori T, YamasakiT, Amamoto Y. Spontaneous resolution of bilateral renal artery dissection: a case report. J Urol 1986;135:114-6. 16. Kent KC, Salvatierra O, Reilly LM, Ehrenfeld WK, Goldstone l, Stoney RI. Evolving strategies for the repair of complex renovascular lesions. Ann Surg 1987;206:272-8. 17. Rao CN, Blaivas JG. Primary renal artery dissecting aneurysm: a review. J Urol 1977;118:716-9. 18. GilfillanRS, Smart WR, Bostick WL. Dissecting aneurysm of the renal artery. Arch Surg 1956;73:737-40. 19. Meyers DS, Grim CE, Keitzer WF. Fibromuscular dysplasia of the renal artery with medial dissection. A case simulating polyarteritis nodosa. Am J Med 1974;56:412-6. 20. Khanna OP, Nedwich A, Gonick P. Accelerated hypertension due to intramural dissection of accessoryrenal artery. Urology i973;1:130-3. 21. Goldfarb R, Pool JL, Wheeler T. Isolated renal artery dissection secondary to medial degeneration. J Urol 1988; 139:346-7. 22. Guthrie W, Maclean H. Dissecting aneurysms of arteries other than the aorta. I Pathol 1972;108:219-35. 23. Neuman HW, Sahin AF. Unilateral dissecting aneurysm of the renal artery. Can ]" Surg 1965;8:291-5. 24. Hasday JD, Sterns RH, Karch FE. Renal infarction due to renal artery dysplasia with dissection. Report of a case in a normotensive patient. Am J Med 1984;76:943-6. 25. Bumpus HC lr. A case of renal hypertension. J Urol 1944;52:295-9. Submitted Feb. 7, 1991; accepted May 3, 1991.
DISCUSSION
Dr. Wesley S. Moore (Los Angeles, Calif.). Dr. Cunningham and his colleagues have addressed a very rare problem encountered by vascular surgeons, spontaneous renal artery dissection. As such, a review of the UCSF experience with 10 patients will be particularly helpful to those of us who will face the management of the occasional
patient that will be seen in our practice. Their review of the literature has clearly documented that the natural history of this disease is unsatisfactory. Patients will either require multiple drug therapy for blood pressure control or nephrectomy. In contrast, the five patients in their series who underwent direct revascularization ended up with an
478
Reilly et al.
excellent clinical result that was sustained over a long follow-up. An additional four patients underwent exploration and immediate nephrectomy when it was deemed that the vessd was not reconstructible. These patients also had an excellent result and are normotensive for a follow-up of more than 14 years. Finally, the last patient who was treated medically remains hypertensive at the time of most recent follow-up. I suspect that the data presented may also be extrapolated to traumatic renal artery dissection. Perhaps one of the more common mechanisms for renal artery dissection seen today is a complication of balloon angioplasty. Implicit in the authors' presentation was the importance of the technique used for renal artery repair. Four of the five patients underwent renal artery replacement by means of the hypogastric artery, and one patient had a resection with primary anastomosis. Although this experience reflects the preference of the UCSF group for the hypogastric artery, it may tend to be written off by others in favor of their own preferences such as vein grafting or patch angioplasty. That may be a mistake. I would like to offer one anecdotal case of my own to emphasize this point. Approximately 3 years ago I had a patient transferred to my service who had a typical lesion of FMD involving her right renal artery that was treated by balloon angioplasty at another hospital. She did not have a good blood pressure response, and therefore a repeat angiogram was performed. The repeat angiogram demonstrated a dissection in the dilated segment, and she was transferred to my service. We explored her artery and elected to treat this by removing the dissected intima by open intimectomy and patch angioplasty. We had an excellent technical result with return of normal Doppler signals in primary and secondary branches. The patient had prompt return to normal blood pressure without the need for any further pharmacologic control. She remained normotensive for approximately 6 months and then had a return of her hypertension, We brought her back fbr repeat angiogram and were disappointed to find that the segment treated by intimectomy had now strictured down to involve the entire primary renal artery as well as the orifices of all of the secondary branches. It appeared to be a typical myointimal hyperplastic response. She ultimately underwent nephrectomy with cure of her hypertension. While I do not want to generalize from a single case because it may well be that the primary segment of renal artery damaged by spontaneous dissection will not respond well to anything other than resection and replacement. I suspect that the excellent results achieved by the UCSF group are based on their decision to treat these with resection and hypogastric artery repair. Whether or not there would be any difference with a vein graft is speculative. However, I am quite sure that repair of the damaged renal artery is ill advised. In closing I would like to ask two questions: (1) What clinical parameters do you use to suspect this lesion and recommend arteriography? (2) Have you looked at your own experience with traumatic dissection, and are the
Journal of VASCULAR SURGERY
surgical results comparable to those achieved with spontaneous dissection? Dr. Chris G. Cunningham. Thank you Dr. Moore. One of the aspects of the natural history of this illness is that it generally is of such a sudden onset and generates such a marked hypertension that even our medical colleagues are alerted to these patients. I suspect that it is the magnitude and severity of the hypertension in an otherwise previously healthy young patient that is the key. We have not reviewed the cases of traumatic renal artery dissection at our institution, because this type of case is rarely seen at our hospital. Dr. Anthony J. R o o n (Everett, Wash.). I'd like to ask the authors a question about the surgical management of the renal artery. I have had the oppornmity to treat three traumatic renal artery dissections, and in examining these patients, one often finds that the hematoma extends on to the branch arteries. However, conventional wisdom has been that if you repair the primary area of dissection and sew the distal intima to the media and adventitia with the repair, the remaining hematoma will disperse and heal. This has been our experience. When you say that certain arteries are not reconstructible by extension of the dissection onto the branches or a certain distance onto the branches, I would like to have you explain to the physicians in this audience who may treat this problem how you decide exactly which lesions are reconstructible and which are not. My experience has been that if yon get rid of the initial tear in traumatic dissections and reconstruct it appropriately that it will heal. Dr. Cunningham. We believe that as long as dissections extend into primary and secondary branch arteries, they are reconstructible. Those that we think are not reconstructible involve extension into the intraparenchymal renal arteries. Dr. Linda M. Reilly. I am actually not certain that this is a setting in which you can limit your repair only to eliminating the site of the tear. I have a suspicion that that is probably one of the reasons why other repairs reported in the literature failed. It may be quite different to repair a large vessel with a tear in it as opposed to repairing a 2 mm or 3 mm branch that has a tear. And the elimination of the false channel in a durable manner is not as predictable. The operative experience with these is often that the septum is very fragile and difficult to sew down. Therefore, trying to work in the diseased vessel may very well be condemning the patient to a less optimal result subsequently. Dr. Edward E. Levine (Orange, Calif.). I support Dr. Reilly's comments. I have experience with one case in which the end point was tacked down or there was a small flap still present, and we thought it was adequate. On the table the artery redissected and went out into all the primary and secondary branches. Dr. Allan D. Callow. (St. Louis, Mo.). I have two questions: the first is, knowing the proclivity for FMD to
Volume 14 Number 4 October 1991
Renal artery dissection 479
be or to become bilateral, I wonder about the long-term follow-up concerning the second side. What were your findings on the second and presumably asymptomatic side? What do you recommend concerning it when present? Was the false hmaen demonstrated in the enlarged version of the specimen you showed us typical of all of your lesions? In some carotid dissections a hematoma may develop deep within the media or in a subadventitial plane. It may, in other instances, be in a subintimal plane, and the method of treatment may be different based on differing etiologies, for example, traumatic versus spontaneous. Was the lesion which you showed us always characteristic in your series? Dr. Ctmningham. I do not know if all the dissections occurred in precisely the same plane. However, this appearance was very consistent in the five or six specimens available for review. None of our patients with bilateral FMD have shown clinical or radiographic evidence of
subsequent dissection in the contralateral renal artery. However, the number of these patients (N = 3) is very small. Dr. Ronald J. Stoney (San Francisco, Calif.). I think Dr. Callow is reminding us about, the complexity of an arterial dissection. These were very complex dissections. They seemed to go between layers, and the dissection was not always contiguous in one layer in the media. The dissection may form a fissure in the media and develop in several different planes making it difficult for a local repair which uses end point tacking as the primary objective. Our surgical objective has been to get beyond the dissection to a normal branch, to transect that, and then attach to that the replacement arterial graft. The graft is then anastomosed to the aorta restoring flow to the kidney.
LIEBIG F O U N D A T I O N AWARD FOR VASCULAR SURGICAL RESEARCH, 1992 The Liebig Foundation announces the tenth annual competitive award of $5000 for the best essay on a problem in general vascular surgery. The investigative work shall be: 1. Clinical or experimental research 2. Original and unpublished (nor submitted elsewhere for publication) 3. Performed by a house officer in the United States, Canada, or Mexico with senior collaborators acting in a consultive capacity 4. Submitted in English (6 copies of typed manuscript and 6 copies of glossy prints of illustrations) 5. Accompanied by a signed letter from the essayist's superior confirming the status of the essayist and complying with "Instructions to Authors" of the JOURNAL Previous winners were: 1990, Scott A. Berceli, MD, University of Pittsburgh School of Medicine, Pittsburgh, Pa.; 1989, Michael L. Marin, MD, College of Physicians and Surgeons, Columbia University, N.Y. The winning essay will be judged for presentation at the June 1991 meeting of the North American Chapter, International Society for Cardiovascular Surgery and will be considered for publication in the JOURNAL Further inquiries may be directed to the same address to which the essays must be sent, postmarked before Dec. 31, 1991: Dr. Richard Turner Award Committee Secretary 112 Bauer Drive Oakland, NJ 07436 USA (201) 337-6126