Journal of Neuro-Ophthalmology, September 1993, Volume 13, Issue 3

Consecutive oculomotor nerve palsy from a de novo cerebral aneurysm.

A 26-year-old woman presented with a partial left oculomotor nerve palsy and an acute subarachnoid hemorrhage. Bilateral angiography showed a left posterior communicating artery aneurysm and a normal right anterior circulation. The aneurysm was successfully clipped. At 4 1/2 years later, she had an acute, nearly complete, right oculomotor nerve palsy. Repeat bilateral angiography showed a right posterior communicating artery aneurysm, in a previously angiographically normal region, and continued closure of the original left-sided aneurysm. New aneurysms may develop from angiographically normal locations. It is possible that certain patients are more susceptible than average to aneurysm formation or that aneurysm repair leads to a higher rate of new aneurysm occurrence. The previous cases of de novo aneurysms are summarized and the implications for repeat aneurysm screening, using current technology, are discussed.

Journal of Clinical Neuro-ophthalmology 13(3): 181-187, 1993. © 1993 Raven Press, Ltd., New York Consecutive Oculomotor Nerve Palsy from a De Novo Cerebral Aneurysm Gerald G. Striph, M.D. A 26-year-old woman presented with a partial left ocu­lomotor nerve palsy and an acute subarachnoid hemor­rhage. Bilateral angiography showed a left posterior communicating artery aneurysm and a normal right an­terior circulation. The aneurysm was successfully clipped. At 41/2 years later, she had an acute, nearly com­plete, right oculomotor nerve palsy. Repeat bilateral an­giography showed a right posterior communicating ar­tery aneurysm, in a previously angiographically normal region, and continued closure of the original left-sided aneurysm. New aneurysms may develop from angiographically normal locations. It is possible that certain patients are more susceptible than average to aneurysm formation or that aneurysm repair leads to a higher rate of new an­eurysm occurrence. The previous cases of de novo an­eurysms are summarized and the implications for repeat aneurysm screening, using current technology, are dis­cussed. Key Words: Cerebral aneurysm-Cerebral angiogra­phy- Oculomotor palsy-Subarachnoid hemorrhage. From the Section of Ophthalmology, Department of Surgery, The Toledo Hospital, Toledo, Ohio, U.S.A. Address correspondence and reprint requests to Dr. Gerald G. Striph, 2121 Hughes Drive, Suite 970, Toledo, OH 43606, U.S.A. 181 CASE REPORT A 26-year-old woman presented to the emer­gency room in 1987 with a severe left-sided frontal headache. She had been previously healthy and carried six pregnancies to term. Ophthalmologic findings were limited to a sluggish, partially­dilated left pupil and minimal conjunctival edema. She complained of neck stiffness, but her neuro­logic examination was otherwise unremarkable. A computed tomographic (CT) scan showed a sub­arachnoid hemorrhage, which was confirmed by lumbar puncture. Cerebral angiography demon­strated a 6 x 8 mm saccular aneurysm of the left posterior communicating artery (Fig. 1). The right internal carotid system was normal (Fig. 2). No spasm was noted on the arteriogram. The aneu­rysm was clipped and she completely recovered. She returned to the emergency center once in 1990 and again in 1991 with a right-sided severe headache. In each case, a CT scan did not reveal an aneurysm or subarachnoid hemorrhage. In the lat­ter episode, a lumbar puncture did not have xan­thochromic cerebrospinal fluid 24 hours after the onset of the headache. There were no other neu­rologic findings. She was diagnosed with hyper­tension during the 1991 visit and started on Pro­cardia. She returned to the emergency center in 1992, complaining of right retro-orbital pain. She was found to be hypertensive (171/113) and, upon con­trol of her blood pressure, the pain resolved. She was discharged but returned 48 hours later com­plaining of a right-sided headache. At that time, she had a nearly complete right oculomotor nerve palsy. The remainder of her ophthalmic and neu­rologic examinations were unremarkable. A CT scan with contrast showed the previous aneurysm clip but was otherwise unremarkable. A cerebral angiogram showed successful dip­ping of the previous left-sided aneurysm (Fig. 3) 182 G. G. STRIPH FIG. 1. Angiogram of the left internal carotid system in 1987. The arrow points to a left posterior communi­cating artery aneurysm. and a new 5 x 8 mm bilobed aneurysm (Fig. 4) at the origin of the right posterior communicating ar­tery (which was angiographically normal in 1987). No spasm was noted during the arteriogram. She underwent surgical clipping of the new aneurysm FIG. 2. Angiogram of the right internal carotid system in 1987. The arrow points to the right posterior com­municating artery. No aneurysm or vasospasm is seen. / Gin Neuro-ophthalmol, Vol. 13, No.3, 1993 FIG. 3. Angiogram of the left internal carotid system in 1992. The arrow points to the clip at the site of the previous aneurysm. It has not recurred. with complete resolution of the oculomotor nerve palsy. DISCUSSION Patients who are successfully treated for intra­cranial aneurysms are at continued risk for the de­velopment and rupture of new aneurysms. The case presented describes a woman who developed a second aneurysm at a previously angiographi­cally normal site. Although in anyone patient, one FiG. 4. Angiogram oftlie right internal carotid system in 1992. The arrow points to a new right posterior communicating artery aneurysm. OCULOMOTOR PALSY FROM DE NOVO ANEURYSM 183 might assume that the second aneurysm was sim­ply not visualized at the time of the first arterio­gram, there are several reasons, discussed below, why this is probably not the case. In either event, this clinical situation, although not commonly known, has been demonstrated several times in the neurosurgical literature. It has implications in the theory of aneurysm development and in the continuing care of postoperative aneurysm pa­tients. This review is limited to those published cases with documentation of the development of an an­eurysm at an angiographically normal site. Table 1 summarizes 27 such reported patients. Twenty­four patients presented with subarachnoid hemor­rhage from an intracranial aneurysm. Of these, the average patient age when the first aneurysm be­came symptomatic was 38.6 years and there was about a 2:1 female preponderance. The average in-terval between treatment of the first aneurysm and discovery of the second aneurysm was 8.6 years. The treatment of the initial aneurysm was by ca­rotid ligation in 13, clipping of the aneurysm in 10, and unreported in 1. Three patients had baseline normal angiography for other reasons and subse­quently developed a subarachnoid hemorrhage from an apparently new aneurysm. Although Wright and Sweet (1) probably re­ported the first case of this entity, it was Graf and Hamby (2) who proposed the term de novo aneu­rysm to describe it. In addition to single aneurysm development, van Alphen and Yang-Zhong (15) reported two cases of multiple de novo aneurysm development in mirror lateral locations after repair of original midline aneurysms. Miller and col­leagues (12) suggested that de novo aneurysms may be a special case of multiple aneurysms in which the lesions appear in series rather than in TABLE 1. Summary of reported angiographically documented de novo aneurysm patients Age at Ref. presentation Sex Location 1 39 F Rt. supraclinoid carotid 2 39 F Rt. middle cerebral 3 32 M Rt. internal carotid-posterior communicating 4 58 F Lt. carotid siphon 5 45 F Lt. posterior communicating 50 M Lt. internal carotid-posterior communicating 6 24 F Lt. posterior communicating 7 41 F Rt. internal carotid-posterior communicating 8 ?-2 patients ? ? 9 19 M Rt. posterior communicating 10 44 M Rt. carotid siphon 35 M Rt. carotid siphon 11 50 M None: angiography done following head trauma 12 31 F Lt. posterior communicating 49 F Lt. internal carotid 42 F Rt. middle cerebral 40 M Rt. middle cerebral 51 F Rt. posterior communicating 49 F Rt. posterior communicating 39 F None: tumor 13 23 M Lt. middle cerebral 26 M None: angiography done for headache 14 49 F Rt. carotid-ophthalmic 15 37 F Basilar 27 F Lt. anterior cerebral This case 26 F Lt. posterior communicating Rt, right; Lt, left. Treatment Carotid ligation Clipped Common carotid ligation Common carotid ligation Common and external carotid ligation Common carotid ligation Carotid ligation Rt. common carotid ligation Carotid ligation Common carotid ligation Carotid ligation Common carotid ligation None Clipped Clipped incompletely Clipped Clipped Clipped ? Pituitary surgery Clipped and wrapped None Wrapped. superficial temporal-middle cerebral anastamosis, carotid ligation Clipped Clipped Clipped Interval until discovery of 2nd aneurysm (years) Location 9.5 Anterior communicating 3 Lt. middle cerebral 10 Lt. supraclinoid carotid 8 Rt. carotid siphon 8 Rt. parasellar carotid 7 Rt. anterior cerebral-anterior communicating 10 Internal carotid bifurcation 9 Lt. carotid-posterior communicating and 2 daughter lesions ? 1 vertebral 1 anterior communicating 3.7 Lt. posterior communicating 9 Lt. supracavernous carotid 15 Lt. posterior communicating 8 Anterior communicating and rt. pericallosal 3 Rt. posterior communicating 11 Rt. internal carotid-posterior communicating and rt. vertebral 5 Basilar 4 Anterior communicating 8 Anterior communicating and rt. middle cerebral 20 Basilar 9 Rt. posterior communicating 9 Anterior communicating and rt. middle cerebral 4 Lt. internal carotid bifurcation 6 Anterior communicating 8 Rt. internal carotid and both middle cerebrals 18 Both middle cerebrals 4.5 Rt. posterior communicating , eli" Neuro-ophthnlmol, Vol. 13, No.3, 1993 184 G. G. STRIPH parallel. Given an incidence of multiplicity of an­eurysms of about 20% (16,17), it would not be sur­prising to find some patients at a point in time between having a single aneurysm and several. The theories regarding the development of sac­cular aneurysms are beyond the scope of this dis­cussion, but have been the subject of review arti­cles (18-20). Aneurysms are usually acquired le­sions that generally appear at locations of weakness of both the internal elastic lamina and tunica media; these weak areas may be congenital or acquired or both. These areas are more fre­quently located at circulatory branch points, in mirror locations to other aneurysms, and with in­creasing age, atherosclerosis and hypertension. Familial (there are some families with an appar­ently inherited tendency toward aneurysm forma­tion) and congenital factors (e.g., collagen abnor­malities) may playa role in some cases (21), but there is no direct evidence of this in these patients. However, it is interesting to note that the mean age of aneurysm rupture of these de novo patients (38.6 years) is much closer to the average age of aneurysm rupture in familial cases (42.3 years) (21) than the typical 6th decade presentation of non­familial cases (22,23). Stehbens (24) described fun­nel- shaped dilatations (or infundibula) at apical angles of the cerebral arteries pathologically. These may be the enlarging "blebs" and infundibula, which have been visualized angiographically to later develop into aneurysms (in addition to the de novo aneurysms from "normal" sites) in many of these cases (3,11,13,15) and may represent a link between the pathologically based theory and clin­ical disease. Changes in hemodynamics may also influence aneurysm development. Areas that are structur­ally weak may develop aneurysms when exposed to greater pressure. This has been supported by both clinical and experimental evidence (25-27). About half of the de novo aneurysms presented here occurred after carotid ligation. Although this may suggest a higher rate of aneurysm formation with the marked change in contralateral perfusion, it also may be an artifact of the prevalence of that surgical procedure in a retrospective review. Two large surveys of carotid ligation patients failed to find an increased rate of new aneurysm formation (28,29). Clipping or balloon occlusion of smaller aneurysms may not cause as dramatic a change in contralateral perfusion pressure as carotid ligation, but it still may cause a sudden, rather than a grad­ual, change in the hemodynamics of the circle of Willis or nearby arteries. However, some of the clipped first aneurysms were relatively small and J Gin Neuro-ophlhalmol. Vol. 13. No.3. 1993 in locations that were unlikely to be hemodynam­ically related to the locations of subsequent aneu­rysms, so other factors are clearly involved. The central issue is whether de novo aneurysm patients are really a special case of aneurysm pa­tients, a chance occurrence, or simply a case of delayed or missed diagnosis. The incidence of un­diagnosed, asymptomatic aneurysms in the gen­eral population is unknown but may be estimated from autopsy and angiography series. Autopsy data suggest a prevalence of about 2-5% in adults but ranges considerably above and below these fig­ures depending on factors such as the patient age distribution, definition of aneurysm and tech­niques in locating them, and the pathologist's skill and interest (19,30-32). Angiography series sug­gest a prevalence of about 1% or less but also are inaccurate due to the types and ages of patients where angiograms are available, insufficient views, since aneurysms were not the original pur­pose of the study, and the other known problems of false-negative angiography (discussed below) (33,34). With variable estimates of prevalence, the exact risk of rupture is difficult to determine. The incidence of subarachnoid hemorrhage (Olmstead county, MN) has remained fairly constant at about 10 cases per 100,000 people per year over the last three decades, and 92% of these hemorrhages are due to aneurysms (23). This corresponds to the observed incidence of 20,000 ruptured aneurysms annually in the United States (35,36). Depending on the prevalence selected, the number of aneu­rysms, and the aneurysm size, the risk of rupture has been estimated from 1 to 10% (with 2-3% per year as a commonly accepted figure) (37-39). Pa­tients with more aneurysms than clinically sus­pected may have a higher rupture and mortality rate (16,22). Unless screening for some other rea­son, one would not find a de novo aneurysm prior to clinical symptoms so only the incidence of rup­tured de novo aneurysms can be observed directly. Miller (12) found at least a sixfold increase in the rate of symptomatic de novo aneurysms compared with the rate of ruptured aneurysms in the general population. Given a bleeding rate of unruptured aneurysms of 3% per year and the mean interval between first and second aneurysm discovery of nearly 9 years in these patients, it seems unlikely that the de novo cases represent a delayed diag­nosis of an aneurysm that was originally present. In the current case, as in all of the patients re­ported above, one possibility is that the second aneurysm was simply missed by the first angio­gram or angiograms. McKissock and colleagues (16) found in their series of 100 postsubarachnoid OCULOMOTOR PALSY FROM DE NOVO ANEURYSM 185 hemorrhage necropsies that 12% of patients thought to have one aneurysm actually had two or more at autopsy and another 8% had more aneu­rysms than angiographically suspected. Although cerebral arteriography remains the "gold stan­dard" for aneurysm diagnosis, most authors place the false-negative rate between 2% and 5%; how­ever, there is evidence, some recent, that it could be higher (22,40--44). The most common reasons for failing to visualize an aneurysm are vasospasm, observer error, inadequate views, and partial thrombosis with sluggish filling of the aneurysmal sac. In this particular case, and specifically noted in many of the other reported cases, the other ves­sels were widely patent and without evidence of vasospasm. Observer error in the reported cases is minimized by retrospective analysis of the original films after the de novo aneurysms were found. It is impossible to exclude thrombosis but the fre­quently long asymptomatic periods between symptoms make this unlikely in this many cases. The increasing number of these cases, and the demonstration over sequential arteriograms, of the development and enlargement of aneurysms, sug­gests that this is a real entity. There is no obvious connection between this pa­tient and the others in the literature beyond having had one treated aneurysm. It is possible that there is some unrecognized difference in these patients, making them a special case, as they seem to have bled at a younger than average age and the cases seem to occur and bleed at a higher than expected rate for the general population. The sex ratio of the patients is similar to aneurysm patients in general. Insufficient information is available to analyze pu­tative risk factors such as smoking and hyperten­sion. Treatment of asymptomatic aneurysms is con­troversial, but the trend is clearly toward prophy­lactic treatment (33,45--47), raising the question of screening higher risk groups (particularly those who have already had one subarachnoid hemor­rhage) (48). Evidence suggests that a higher pro­portion of asymptomatic aneurysms rupture than was previously thought; further, the morbidity from ruptured aneurysms is still exceedingly high (23) (about 50% die in the first 30 days) in the face of dropping morbidity from surgical and interven­tional radiologic care. Warning signs of aneurysm rupture may only appear about half of the time (49) and cannot, therefore, be the sole guide in screen­ing potential aneurysm patients or selecting surgi­cal candidates. For this reason, and since these pa­tients seem to be at higher risk, some advocate screening them periodically for aneurysm develop-ment. However, the cost-benefit ratio of various types of screening studies in this setting is still not established. Although the rate of new aneurysm formation may be higher in previously treated aneurysm pa­tients than in the general population, it is probably not high enough to justify screening asymptomatic patients with routine late arteriography. Angiog­raphy, even in a high-volume and -quality setting, still has risk of permanent neurologic damage (0.33%), transient neurologic complications (2.6%), and other lesser complications (8.5%) (50). In patients with lower suspicion of aneurysm, magnetic resonance angiography may be able to playa role (41,51,52). This technique seems to be able to detect aneurysms 3-4 mm in size. When combined with standard magnetic resonance im­aging of the brain (adding 15-20 minutes to that procedure), it may have a sensitivity of 95% in de­tecting patients with aneurysms (53). The addition of standard spin-echo views of the brain paren­chyma to the angiographic views adds signifi­cantly to the procedure's sensitivity in detecting aneurysms as well as screening for other intracra­nial processes that mimic aneurysms (and perhaps then removing the need for conventional arteriog­raphy). MR angiography and arteriography may occasionally be complimentary. In Lanfermann's retrospective comparison of selective angiography with MRI, magnetic techniques found 92% of the aneurysms detected by arteriography as well as two thrombosed aneurysms (6%) missed by the conventional angiograms (54). Aneurysms that cause oculomotor palsies (such as the one in the presented patient) are typically much larger than the stated 3- to 4-mm resolution of MR angiography. Soni (55) found an average posterior communicating artery aneurysm diame­ter greater than 11 mm in over half of his oculo­motor palsy patients and Bartleson and associates (56) data on patients with minimal oculomotor palsy suggests an average minimum aneurysm di­mension of greater than 9 mm. However, negative magnetic resonance imaging is probably not suffi­cient when clinical suspicion is high, such as with signs of subarachnoid hemorrhage or oculomotor nerve palsy involving the pupil. Ross' study (53) missed aneurysms 3-8 mm in size. Keane and Ah­madi (57) reported two patients with oculomotor nerve palsies involving the pupil in which mag­netic resonance imaging (MRI) of the brain with MR angiography missed aneurysms 5-6 mm in size. In patients presenting with signs strongly suggestive of subarachnoid hemorrhage, CT scan­ning followed, if positive, by cerebral angiography J Clin Neuro-ophthalmol, Vol. 13, No.3, 1993 186 G. G. STRIPH is still probably the evaluation method of choice (41). In patients with suggestive findings, screening for aneurysm development using a technique ap­propriate for the level of clinical suspicion, is man­datory. It may be reasonable to periodically screen patients with a history of aneurysm repair for re­currence or new aneurysm development; these cases suggest that the risk remains high for many years. Previously negative angiography should not deter the clinician from reevaluating a patient for an aneurysm-particularly in that subgroup of pa­tients that has had aneurysm repair in the past. REFERENCES 1. Wright RL, Sweet WHo Carotid or vertebral occlusion in the treatment of intracranial aneurysms; value of early and late readings of carotid and retinal pressures. Clin Neurosurg 1962;9:163-92. 2. Graf q, Hamby WB. Report of a case of cerebral aneurysm in an adult developing apparently de novo. J Neurol Neu­rosurg Psychiatry 1964;27:153-6. 3. du Boulay GH. Some observations on the natural history of intracranial aneurysms. Br JRadiol 1965;38:721-57. 4. Gurdjian ES, Linder OW, Thomas LM. Experiences with ligation of the common carotid artery for treatment of an­eurysms of the internal carotid artery. J Neurosurg 1965;23: 311-18. 5. Somach FM, Shenkin HA. Angiographic end-results of ca­rotid ligation in the treatment of carotid aneurysms. JNeu­rosurg 1966;24:96~74. 6. Heiskanen 0, Marttila. I. Risk of rupture of a second an­eurysm in patients with multiple aneurysms. J Neurosurg 1970;32:29.>-9. 7. Klemme WM. Hemorrhage from a previously undemon­strated intracranial aneurysm as a late complication of ca­rotid ligation. J Neurosurg 1977;46:654-8. 8. Miller JD, Jawad K, Jennett B. Safety of carotid ligation and its role in the management of intracranial aneurysms. J Neurol Neurosurg Psychiatry 1977;40:64-72. 9. Winn HR, Richardson AE, Jane JA. Late morbidity and mortality of common carotid ligation for posterior commu­nicating aneurysms. JNeurosurg 1977;47:727-36. 10. Salar G, Mingrino S. Development of intracranial saccular aneurysms: report of two cases. Neurosurgery 1981;8:462-5. 11. Nukui H, Nagaya T, Miyagi 0, et al. Development of new aneurysm and enlargement of small aneurysm: report of seven cases. [English abstract]. Neurol Med Chir (Tokyo) 1982;22:437-45. 12. Miller CA, Hill SA, Hunt WE. De novo aneurysms: a clin­ical review. Surg Neurol 1985;24:173-80. 13. Misra BK, Whittle IR, Steers A}W, Sellar RJ. De novo sac­cular aneurysms. Neurosurgery 1988;23:1~15. 14. Dyste GN, Beck OW. De novo aneurysm formation follow­ing carotid ligation: case report and review of the literature. Neurosurgery 1989;24:81>-92. 15. van Alphen HAM, Yong-Zhong G. Multiple cerebral de novo aneurysms. Clin Neurol Neurosurg 1991;93:13--18. 16. McKissock W, Richardson A, Walsh L, Owen E. Multiple intracranial aneurysms. Lancet 1964;1 :623-6. 17. Inagawa T, Hirano A. Autopsy study of unruptured inci­dental aneurysms. Surg NeuroI1990;34:361-5. 18. Suzuki}, O'Hara H. Clinicopathological study of cerebral aneurysms: origins, rupture, repair, and growth. JNeuro­surg 1978;48:50.>-14. / Clin Neuro--ophthalmol, Vol. 13. No.3. 1993 19. Sekhar LN, Heros RC. Origin, growth, and rupture of sac­cular aneurysms: a review. Neurosurgery 1981;8:2~. 20. Yong-Zhong G, van Alphen HAM. Pathogenesis and his­topathology of saccular aneurysms: review of the literature. Neurol Res 1990;12:249-55. 21. ter Berg HWM, Dippel OW}, Limburg M, et aI. Familial intracranial aneurysms: a review. Stroke 1992;23:1024-30. 22. Locksley HB. Natural history of subarachnoid hemorrhage, intracranial aneurysms, and arteriovenous malformations based on 6368 cases in the Cooperative Study: report of the Cooperative Study of Intracranial Aneurysms and Sub­arachnoid Hemorrhage: Section V, Part 1. JNeuTosurg 1966; 25:219-39. 23. Phillips LH II, Whisnant ]P, O'Fallon WM, Sundt 1M Jr. The unchanging pattern of subarachnoid hemorrhage in a community. Neurology 1980;30:1034-40. 24. Stehbens WE. Histopathology of cerebral aneurysms. Arch Neurol 1963;8:272-85. 25. Hassler O. Experimental carotid ligation followed by aneu­rysmal formation and other morphological changes in the circle of Willis. JNeurosurg 1963;20:1-7. 26. Portnoy HD, Avellanosa A. Carotid aneurysm and contra­lateral carotid stenosis with successful surgical treatment of both lesions. JNeurosurg 1970;32:476-82., 27. Waga S, Ohada M, Kojima T. Saccular aneurysms associ­ated with absence of the left cervical carotid arteries. Neu­rosurgery 1978;3:201>-12. 28. Nishioka H. Report on the cooperative study of intracranial aneurysms and subarachnoid hemorrhage: results of the treatment of intracranial aneurysms by occlusion of the ca­rotid artery in the neck. JNeurosurg 1966;25:660-83. 29. Odom GL, Tindall GT. Carotid ligation in the treatment of certain intracranial aneurysms. Clin Neurosurg 1968;15:101­16. 30. McCormick WF, Acosta-Rua GJ. The size of intracranial sac­cular aneurysms: an autopsy study. J Neurosurg 1970;33: 422-7. 31. Housepian EM, Pool}L. A systematic analysis of intracra­nial aneurysms from the autopsy file of the Presbyterian Hospital. J Neuropathol Exp Neurol 1958;17:409-23. 32. Stehbens WE. Aneurysms and anatomical variation of ce­rebral arteries. Arch Pathol 1963;75:4.>-64. 33. Atkinson }LD, Sundt TM Jr., Houser OW, Whisnant JP. Angiographic frequency of anterior circulation intracranial aneurysms. JNeurosurg 1989;70:551-5. 34. Wakai S, Fukushima T, Furihata T, Sano K. Association of cerebral aneurysm with pituitary adenoma. Surg Neurol 1979;12:503--7. 35. Sypert GW. Intracranial aneurysm: natural history and sur­gical management. Compr Ther 1978;4:64-73. 36. Drake CG. Management of cerebral aneurysm. Stroke 1981; 12:273-83. 37. ~ane JA, Winn HR, Richardson AE. The natural history of mtracramal aneurysms: rebleeding rates during acute and long-term period and implication for surgical management. Clin Neurosurg 1977;24:176-84. 38. Heiskanen O. Risk of bleeding from unruptured aneurysm In cases with multiple intracranial aneurysms. JNeurosuTg 1981;55:524-{,. 39. Wiebers DO, Whisnant]p, O'Fallon WM. The natural his­tory of unruptured intracranial aneurysms. N Engl JMed. 1981;304:696-8. 40. Perrett LV, Bull JWD. Some aspects of subarachnoid hem­orrhage: a symposium III. The accuracy of radiology in demonstrating ruptured intracranial aneurysms. BT JRadiol 1959;32:8.>-92. 41. Wolpert SM, CaplanLR. Current role of cerebral angiogra­phy In the dIagnOSIs of cerebrovascular diseases. Am J Roentgenol 1992;159:191-7. 42. Forster DMC, Steiner L, Hakanson 5, Bergvall U. The value of repeat pan-angiography in cases of unexplained sub­arachnoid hemorrhage. JNeurosurg 1978;48:712-16. 43. Iwanaga H, Wakai 5, Ochiai C, et al. Ruptured cerebral OCULOMOTOR PALSY FROM DE NOVO ANEURYSM 187 aneurysms missed by initial angiographic study. Neurosur­gery 1990;27:45-51. 44. Juul R, Fredriksen TA, Ringkjob R. Prognosis in subarach­noid hemorrhage of unknown etiology. J Neurosurg 1986; 64:359-62. 45. Rice BJ, Peerless SI, Drake CG. Surgical treatment of un­ruptured aneurysms of the posterior circulation. JNeuro­surg 1990;73:165-73. 46. Ausman JL, Diaz FG, Malik GM, et al. Management of cerebral aneurysms: further facts and additional myths. Surg Neurol 1989;32:21-35. 47. Wirth FP, Laws ER Jr., Piepgras 0, Scott RM. Surgical treat­ment of incidental intracranial aneurysms. Neurosurgery 1983;12:507-11. 48. Wiebers DO, Torres VE. Screening for unruptured intracra­nial aneurysms in autosomal dominant polycystic kidney disease. N Engl JMed 1992;327:953-5. 49. Okawara S-H. Warning signs prior to rupture of an intra­cranial aneurysm. JNeurosurg 1973;38:575-80. 50. Earnest F IV, Forbes G, Sandok BA, et al. Complications of cerebral angiography: prospective assessment of risk. Am J NeuroradioI1983;4:1191-7. 51. Huston I, Rufenacht DA, Ehman RL, Wiebers DO. Intra­cranial aneurysms and vascular malformations: comparison of time-of-flight and phase-contrast MR angiography. Ra­diology 1991;181:721-30. 52. Edelman RR, Mattle HP, Atkinson OJ, Hoogewoud HM. MR angiography. Am JRoentgenol 1990;154:937-46. 53. Ross JS, Masaryk TI, Modic MT, et al. Intracranial aneu­rysms: evaluation by MR angiography. Am J Roentgenol 1990;155:159-65. 54. Lanfermann H, Gross-Fengels W, Steinbrich W. Intracrani­al aneurysms. A comparison between magnetic resonance tomography and arteriography [English abstract]. Roto Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1992;157:11B­23. 55. Soni SR. Aneurysms of the posterior communicating artery and oculomotor paresis. JNeurol Neurosurg Psychiatry 1974; 37:474-84. 56. Bartleson JD, Trautmann JC, Sundt TM Jr. Minimal oculo­motor nerve paresis secondary to unruptured intracranial aneurysm. Arch Neurol 1986;43:1015-20. 57. Keane JR, Ahmadi J. Third-nerve palsies and angiography. Arch NeuroI1991;48:470. 1Clin Neuro-ophthalmol, Vol. 13, No.3, 1993