Introduction 

   It is well known that aneurysms of the internal carotid posterior communicating artery junction cause oculomotor nerve palsy.⁵⁾²⁴⁾ Oculomotor nerve palsy related findings include ptosis, dilated pupils, and extraocular muscle weakness. Aneurysmal enlargement and rupture appear to be responsible for these findings. The majority of neurosurgeons are primarily concerned about the life-saving nature of intervention for such disastrous aneurysms, and subsequently address oculomotor palsy. However, few detailed studies have been performed on the various factors that influence the prognosis of oculomotor nerve palsy. The aim of this study was to identify prognostic factors for the recovery of oculomotor palsy in patients with posterior communicating artery aneurysms after surgical intervention. 

Materials and Methods 

   We retrospectively reviewed hospital medical records, operative reports, office follow-up records with a detailed ophthalmic examination, preoperative and postoperative angiogrphic studies of 16 patients with ruptured or unruptured posterior communicating artery aneurysms associated with third cranial nerve palsy who were surgically treated at our hospital between January 2002 and July 2006. Of these 16 patients, 15 were women, ages ranged between 38 and 75 years (mean 52.6 years), and follow up periods from 0.5 to 36 months (mean 8.75 months). Brain computed tomographic angiography (CTA) was checked in all patients, and additional 4-vessel angiography was performed on 11 patients preoperatively. All aneurysms were clipped under a microscope. Completeness of ptosis improvement, recovery of dilated and fixed pupils, and limitations of extraocular muscles were retrospectively reviewed. Preoperatively, we defined complete oculomotor nerve palsy as the presence of all components of palsy, i.e. ptosis, mydriasis, and extraocular muscle weakness. The criteria used to define complete recovery from oculomotor nerve palsy were as follows : 1) complete resolution of ptosis; 2) full range of movement in medial, inferior, and superior gaze; and 3) partial or complete recovery of pupillary reaction. The degree of oculomotor nerve palsy (i.e., complete or incomplete palsy), symptom duration, ruptured versus unruptured aneurysms, and aneurysmal sizes (≥10 mm or <10 mm) in the two groups were compared. To compare the two groups, the data was analyzed by cross table studies by generating probability value using either the chi-square test of Fisher's exact probability test. Data analysis was performed using SPSS for windows (SPSS Inc, Chicago, IL). 

Results 

Incidence 
   The total number of aneurysmal cases treated over the 5-year study period was 1135. 223 patients (19.6%) presented with posterior communicating artery aneurysms, and 16 (7.2%) of these presented with third cranial nerve palsy (ruptured : 12, unruptured : 4). 

Oculomotor nerve signs and other symptoms 
   Ptosis was presented by all 16 cases (100.0%), and extraocular muscle palsy was presented to some degree by 10 cases (62.5%). Affected pupils were dilated and fixed in all patients (100.0%), and thus, complete oculomotor nerve palsy was presented by 10 patients. Other symptoms or signs included; diplopia, retro-orbital pain, headache, unilateral headache, and eye swelling. All patients had a preoperative Hunt-Hess grade III or better. 

Characteristics of aneurysms 
   Based on imaging studies, all sixteen posterior communicating artery aneurysms were directed posterolaterally and had a saccular configuration, and these imaging characteristics corresponded with intraoperative findings. 

Recovery from oculomotor palsy 
   Ptosis was improved completely in 13 patients (81.3%), and incompletely in 3 (18.7) within 6 months. Dilated and fixed pupils were completely improved in 10 (62.5%) of the 16 cases at last follow-up. Of the 6 patients with incomplete dilated and fixed pupil recovery, incomplete improvement of ptosis was presented in three. This patient had an unruptured, 8mm sized aneurysm, and complete oculomotor palsy. Complete oculomotor palsy improvement was demonstrated by 13 patients. A summary of patient characteristics is provided in (Table 1). 

Duration of symptoms and oculomotor palsy recovery 
   The mean interval between onset of palsy and surgery was 6.2 days in complete recovery cases, but 22 days in incomplete recovery cases. However, recovery of pupillary function was unpredictable. 

Preoperative deficit and recovery from oculomotor palsy 
   Six patients had preoperatively incomplete third nerve palsy, and all achieved complete recovery. Of the ten patients with complete third nerve palsy, seven patients achieved complete recovery, and three incomplete recovery (p=0.252). The correlation between recovery of third nerve palsy and preoperative degree of palsy is presented in (Table 2). 

Aneurysmal size and recovery from oculomotor palsy 
   Aneurysms with a diameter of larger than 10mm were present in 6 patients. In these patients, five (83.3%) achieved complete recovery of third nerve palsy within 6 months. Dilated and fixed pupils were complete recovery in 4 (66.7%) of the 6 patients at last follow-ups. Of the 10 patients with an aneurysm of less than 10mm in diameter, 8 (80.0%) achieved complete improvement, and 6 (60%) patients showed complete improvement of dilated and fixed pupil at last follow-ups (Table 3). 

Presence of subarachnoid hemorrhage (SAH) and recovery from oculomotor palsy 
   There were 12 ruptured aneurysm cases and 4 unruptured cases. In the 12 ruptured cases, 10 patients (83.3%) showed complete recovery of third nerve palsy and 2 (16.7%) showed incomplete recovery. Three of the 4 unruptured cases achieved complete recovery, and the remaining patient, incomplete recovery (Table 4). 

Discussion 

   Isolated third nerve palsy can be caused by diabetes mellitus, trauma, or neoplasm, but usually it is caused by an aneurysm.²¹⁾²²⁾ It has been long known that oculomotor nerve palsy is induced by cerebral aneurysms, especially by posterior communicating artery aneurysms.⁶⁾²³⁾ In fact, the incidence of palsy among cases of posterior communicating artery has been reported to range from 34% to 56%.⁸⁾¹⁴⁾ 
   Oculomotor nerve palsy is common in cases with a postero-lateral aneurysmal direction, whereas aneurysms with a purely posterior or medial direction are rarely associated with oculomotor disturbances.¹⁵⁾ The frequency of ocular paresis was reported to be considerably increased when aneurysms were multilocular and larger than 11mm diameter.¹⁶⁾ Moreover, only 15% of aneurysms below this size induced oculomotor palsy.¹⁶⁾ In the present study, we used 10mm as yardstick for aneurysmal size. Other investigators have failed to find a significant correlation between radiologically determined aneurysmal dimensions and clinical features.¹⁸⁾ However, it should be borne in mind that angiographic findings do not always reflect true aneurysmal size. Furthermore, dynamic factors, such as, pulsations and dilatation of the aneurismal sac, must also be taken into consideration.⁷⁾
   Oculomotor palsy may occur either immediately with the onset of other aneurismal symptoms or with a delay of some days. Ptosis is usually the first symptoms of an initially incomplete ocular paresis.¹⁾⁴⁾ The probability that an initially incomplete ocular palsy will progress to complete external and internal opthalmoplegia within some days or weeks is very high. It is obvious that the less the preoperative damage to the cranial nerve the better are the chances for rapid improvement after surgery. And Hamer⁷⁾ reported the prognosis in cases with incomplete oculomotor palsy and mild preoperative deficit is very good. 
   Several investigators have emphasized that the best results were seen only after early operation, i.e in the cases which were operated on within two weeks, Botterell et al. and Soni¹⁶⁾ expressly stated that full recovery was only observed when the operation was carried out within 10 days. This agrees with the findings of Hamer who said that the prognosis in cases with incomplete palsy and mild preoperative deficit is very good.⁷⁾ 
   The manifestation of oculomotor palsy as a warning episode without simultaneous rupture of the aneurysm is not exceptional.^1)13)19) The assumption that a symptomatic aneurysm has not ruptured is usually based on the absence of three common signs: disturbance of consciousness, meningism, and bloody lumbar cerebrospinal fluid. However, one should bear in mind the difference between major bleeding and minor leakage with limited diapedesis of red blood cells into a single basal cistern. Based on numerous microsurgical observations, the arachnoid fibers and membranes that usually separate the basal cisterns are "regularly thicker and tougher where the arteries pass through the trabeculated wall from one cisternal compartment to another." Thus, it is possible that minimal monocisternal bleeding may occur without producing the typical symptoms of SAH.²⁴⁾ Accordingly, oculomotor nerve palsy in patients with a posterior communicating artery aneurysm without a major hemorrhage must be considered an important warning sign of SAH.¹³⁾¹⁹⁾ 
   The initial symptoms of oculomotor nerve palsy among our cases were ptosis and/or blurred vision, regardless of the presence of SAH. In those with a SAH, sudden onset severe headache preceded oculomotor palsy, although in the group without SAH, retrobulbar pain and a unilateral headache were also frequently present.⁵⁾ 
   Although rare, pupillo-constrictor function can remain intact in patients with incomplete third nerve palsy.^3)4)10)12) This can be explained by the microscopic topography of the third nerve. According to anatomical investigations, parasympathetic fibers run peripherally in the dorso-medial arc, although a few fibers may also run in the lateral portion of the nerve. Thus, pupillo-constrictor axons may be spared if the nerve is only displaced postero-laterally and not lifted and pressed against the tentorial edge.^10)11)17) Summarizing, cerebral aneurysm cannot be excluded, even when a patient experiences oculomotor palsy with pupil sparing.¹²⁾ 
   Some investigators have suggested that oculomotor nerve palsy degree is related to direct damage to the nerve caused by aneurysm rupture, and have concluded that oculomotor nerve palsy outcome is poorer in cases with SAH.²²⁾ However, the degree of ultimate recovery from oculomotor nerve palsy achieved by our cases was unaffected by the presence of hemorrhage. Based on out findings it appears that prompt surgical treatment favors functional nerve recovery. In a previous study, full recovery from third nerve palsy was only observed when intervention was undertaken within 10 days of onset.¹⁾¹⁶⁾ In addition, several other investigators have emphasized that best results are achieved only after early operations, i.e. in those cases that are operated on within two weeks.^1)2)9)16)18) 
   It is self-evident that reduced preoperative damage to the cranial nerve improves the likelihood of rapid improvement after surgery, and the prognosis of cases with incomplete oculomotor palsy and mild preoperative deficit has been reported to be good.⁷⁾ 
   The return of oculomotor nerve function shows a consistent course. The levator palpabrae muscle improves first, followed by the medial rectus muscle and the inferior or superior rectus muscles. However, the recovery of pupillary function varies greatly.¹⁾²⁵⁾ Some authors suggest the use of decompressive procedures for the relief of third nerve palsy,¹⁵⁾ whilst authors have concluded that for nongiant aneurysms decompressive surgery is not always needed to improve oculomotor nerve palsy.⁵⁾ Possibly, the shrinking of an aneurysm after aneurysmal sac clipping and the associated abolition of pulsatile arterial transmission are essential and sufficient to produce neurovascular decompression.⁷⁾ 

Conclusion 

   Although our study does do not stand statistical scrutiny, this small group of sixteen patients demonstrates that symptom duration and degree of preoperative deficit should be regarded as prognostic factors for the postoperative recovery of oculomotor nerve palsy in patients with a posterior communicating artery aneurysm. 
   Our results suggest that early surgical management for a posterior communicating artery aneurysm accompanying oculomotor palsy offers the best chance of achieving a satisfactory outcome. A mullticenter study is needed to validate these findings. 

REFERENCES

1. Botterell EH, Lloyd LA, Hoffman HJ. Oculomotor palsy due to supraclinoidal internal carotid artery berry aneurysm. Amer J Ophthalmol 54 : 609-16, 1962. 

2. Cantu RC, Souders Th, Hepler RS. Effects of carotid ligation on aneurysm-induced oculomotor palsy. J Neurosurg 31:528-32, 1969 

3. Cogan DG, Mount HTJ. Intracranial aneurysms causing ophthalmoplegia. Arch Ophthalmol 70 : 757-71, 1963 

4. Daily EJ, Holloway JA, Murto RE, Schlezinger NS. Evaluation of ocular signs and symptoms in cerebral aneurysms. Arch Ophthalmol 71 : 463-74, 1964 

5. Fujiwara S, K Fujii, S Nishio, T Matsushima, M Fukui. Oculomotor nerve palsy in patients with cerebral aneurysms. Neurosurg Rev 12:123-32, 1989 

6. Gliobini S, Ferraesi S, Pluchino F. Reversal of oculomotor disoders after intracranial aneurysm surgery. Acta Neruochir (Wien) 112 : 19-24, 1991 

7. Hamer J. Prognosis of oculomotor palsy in patients with aneurysms of the posterior communicating artery. Acta Neurochir 66 : 173-85, 1982 

8. Hamilton JG, Falconer MA. Immediate and late results of surgery in cases of saccular intracranial aneurysms. J Neurosurg 16 : 514-41, 1959 

9. Hepler RS, Cantu RC. Aneurysms and third nerve palsies. Ocular status of survivors. Arch Ophthalmol 77 : 604-8, 1967 

10. Kasoff I, Kelly LD. Pupillary sparing in oculomotor palsy from internal carotid aneurysm. J Neurosurg 42 : 713-17, 1975 

11. Kerr FWL, Hollowell OW. Location of pupillomotor and accommodation fibers in the oculomotor nerve: Experimental observations on paralytic mydriasis. J Neurol. Neurosurg Psychiat 27 : 473-81, 1964 

12. Kissel JT, Burde RM, Klingele TG, Zeiger HE. Pupil-sparing oculomotor palsies with internal carotid-posterior communicating artery aneurysms. Ann Neurol 13 : 149-54, 1983 

13. Okawara SH. Warning signs prior to rupture of an intracranial aneurysm. J Neurosurg 38 : 575-80, 1973 

14. Paterson A. Direct surgery in the treatment of posterior communicating aneurysms. Lancet 2 : 808-11, 1968 

15. Sano K. Internal carotid-posterior communicating anterior choroidal region aneurysms. Microsurgical treatment. In : Cerebral aneurysms. Advances in diagnosis and therapy (Pia HW, Langmaid C, Zierski J, eds.), Berlin-Heidelberg-New York : Springer : 252-9, 1979. 

16. Soni SR. Aneurysms of the posterior communication artery and oculomotor paresis. J Neurol Neurosurg Psychiat 37 : 475- 84, 1974 

17. Sunderland S, Hughes ESR. The pupilloconstrictor pathway and the nerves to the ocular muscles in man. Brain 69 : 301-9, 1946 

18. Shigeru F, Kiiyotaka F, Shuunji N, Toshio M, Masahi F. oculomotor nerve palsy in patients with cerebral aneurysmsd. Neurosurg Rev 12 : 123-32, 1989 

19. Waga S, Ohtusubo K, Handa H. Warning signs in intracranial aneurysms. Surg Neurol 3 : 15-20, 1975 

20. Walsh FB. Third nerve regeneration. A clinical evaluation. Brit. J. Opthalmol. 41: 577-98, 1957 

21. Walsh FB, King AB. Ocular signs of intracranial saccular aneurysms. Arch Ophthalmol 27 : 1-33, 1942 

22. Watanabe A, Ishii R, Tanaka R. Postoperative course of involved cranial nerves due to intracranial aneurysms. Neurol Med Chir (Tokyo) 25 : 633-9, 1985 

23. Weir B, Wilkins RH, Rengachary SS. Intracranial aneurysms and subarachnoid hemorrhage : an overview. Neurosurgery, McGraw-Hill, New York : 1308-29, 1985 

24. Yasargil MG, Kasdaglis K, Jain K, Weber HP. Anatomical observations of the subarachnoid cisterns of the brain during surgery. J. Neurosurg 44 : 298-302, 1976