Journal of Neuro-Ophthalmology, December 2001, Volume 21, Issue 4

Patterns of Extraocular Muscle Weakness in Vasculopathic Pupil-Sparing, Incomplete Third Nerve Palsy

OBJECTIVE: To determine the pattern of extraocular muscle (EOM) paresis in incomplete vasculopathic third nerve palsies (3NP) that have normal pupillary function. METHODS: A retrospective study in a private practice and academic neuro-ophthalmic practice. Patients diagnosed with vasculopathic 3NP within 4 weeks of symptom onset were identified. The chart of each patient was reviewed to determine pupillary function and the pattern and degree of EOM and levator palpebrae paresis at the time of presentation. RESULTS: Of 55 patients with vasculopathic 3NP, 42 (76%) had normal pupillary function. Of these 42, 23 (55%) demonstrated an incomplete EOM palsy, defined as partially reduced ductions affecting all third nerve-innervated EOMs and levator (diffuse pattern) or partially reduced ductions that involved only some third nerve-innervated EOMs and levator (focal pattern). Twenty (87%) of these 23 patients showed a diffuse pattern of paresis; only three (13%) showed a focal pattern of paresis, one that affected only the superior rectus and levator muscles (superior division weakness). CONCLUSIONS: Based on our series, most patients with EOM/levator involvement in pupil-sparing, incomplete 3NP of vasculopathic origin have a diffuse pattern of paresis. In contrast, our review of the literature suggests that pupil-sparing 3NP of aneurysmal origin usually have a focal pattern of paresis. We propose that distinguishing these two patterns of EOM paresis may be helpful in differentiating between vasculopathic and aneurysmal 3NP. Future studies will be needed to confirm the clinical utility of this hypothesis.

Journal of Neuro- Ophthalmology 21( 4): 256- 259, 2001. • 2001 Lippincott Williams & Wilkins, Inc., Philadelphia Original Contribution Patterns of Extraocular Muscle Weakness in Vasculopathic Pupil- Sparing, Incomplete Third Nerve Palsy Scott Sanders, MD, Aki Kawasaki, MD, and Valerie A. Purvin, MD Objective: To determine the pattern of extraocular muscle ( EOM) paresis in incomplete vasculopathic third nerve palsies ( 3NP) that have normal pupillary function. Methods: A retrospective study in a private practice and aca­demic neuro- ophthalmic practice. Patients diagnosed with vas­culopathic 3NP within 4 weeks of symptom onset were iden­tified. The chart of each patient was reviewed to determine pupillary function and the pattern and degree of EOM and levator palpebrae paresis at the time of presentation. Results: Of 55 patients with vasculopathic 3NP, 42 ( 76%) had normal pupillary function. Of these 42, 23 ( 55%) demonstrated an incomplete EOM palsy, defined as partially reduced duc-tions affecting all third nerve- innervated EOMs and levator ( diffuse pattern) or partially reduced ductions that involved only some third nerve- innervated EOMs and levator ( focal pattern). Twenty ( 87%) of these 23 patients showed a diffuse pattern of paresis; only three ( 13%) showed a focal pattern of paresis, one that affected only the superior rectus and levator muscles ( superior division weakness). Conclusions: Based on our series, most patients with EOM/ levator involvement in pupil- sparing, incomplete 3NP of vasculopathic origin have a diffuse pattern of paresis. In con­trast, our review of the literature suggests that pupil- sparing 3NP of aneurysmal origin usually have a focal pattern of pare­sis. We propose that distinguishing these two patterns of EOM paresis may be helpful in differentiating between vasculopathic and aneurysmal 3NP. Future studies will be needed to confirm the clinical utility of this hypothesis. Key Words: Third nerve- Pupil sparing. - Aneurysm- Vasculopathic palsy- Manuscript received May 10, 2001; accepted September 6, 2001. From Associated Vitreo- Retinal and Uveitis Consultants Indianapo­lis, Indiana ( SS); Hopital Ophtalmique lules Gonin, Lausanne, Swit­zerland ( AK); Midwest Eye Institute and Department of Neurology and Ophthalmology, Clarian Hospitals of Indiana and Indiana University Medical Center, Indianapolis, Indiana ( VAP). Address correspondence and reprint requests to Valerie Purvin, MD, 201 Pennsylvania Parkway, Indianapolis, IN 46280, USA; E- mail: vpurvin@ iupui. edu The two foremost diagnostic considerations in the evaluation of acute isolated third nerve palsy ( 3NP) are aneurysmal compression and ischemic injury ( 1). The former diagnosis is an acute medical emergency because of the possibility of aneurysmal rupture and intracranial hemorrhage. Urgent radiographic and neurosurgical evaluation is necessary. The latter diagnosis, also called vasculopathic 3NP, is a benign condition in which simple observation is sufficient because spontaneous recovery is the typical clinical outcome. Therefore, it is important to have a rapid and reliable clinical means of distinguishing between aneurysmal and vasculopathic 3NP. In 1958, Rucker ( 2) provided one such clinical tool to differentiate these two entities when he wrote that " pu­pillary reactions yield valuable information. If they are normal the paralysis is probably due to occlusive vascu­lar disease; if there is iridoplegia, paralysis is more likely due to an aneurysm." In the setting of an acute isolated 3NP, the presence of a dilated or poorly reactive pupil warns the clinician of the possibility of an aneurysm and thus the need for acute evaluation. This came to be known as " the rule of the pupil," and it has proved useful over the years ( 2- 4). However, exceptions to the rule subsequently emerged in which patients with pupil-sparing, in fact, harbored an aneurysm ( 5- 11). On closer inspection of these cases, it was noted that when the pupil was spared, other extraocular muscles were spared as well. Thus, the concept of ' the exception to the rule' was proposed, stating that when the external oph­thalmoplegia and ptosis related to 3NP are not complete, the presence of normal pupillary reactivity does not re­liably exclude aneurysmal compression ( 12,13). In this setting, it has been suggested that a full neuroradiologic investigation including angiography be undertaken to exclude aneurysm. We sought to identify other clinical parameters that might assist the clinician in differentiat­ing aneurysmal from vasculopathic 3NP when the palsy is incomplete. We hypothesized that ischemic third nerve injury might produce a pattern of incomplete extraocular 256 EXTRAOCULAR WEAKNESS IN THIRD NERVE PALSY 257 muscle ( EOM) weakness that was distinctive from that caused by aneurysmal compression. To this end, we ret­rospectively investigated the distribution and degree of EOM and levator palpebrae paresis among patients with incomplete, pupil- sparing vasculopathic 3NP. SUBJECTS AND METHODS The records of all patients diagnosed with acute vas­culopathic 3NP at a single neuro- ophthalmology clinic over the past 14 years were reviewed. Patients who were not seen within 4 weeks of onset of symptoms were excluded because after that time interval, spontaneous recovery of function from ischemic injury may occur. Patients who had had strabismus surgery or mechanical iris dysfunction or were using topical ophthalmic medi­cations that might alter pupillary function were excluded. Based on the initial examination, the following three as­pects of the clinical examination were documented: 1) the pupil size and reactivity; 2) paresis of the superior rectus, inferior rectus, medial rectus, inferior oblique, and levator palpebrae muscles; and 3) the degree of pare­sis of affected EOMs. For purposes of this study, EOMs refer to all third nerve- innervated striated muscles in­cluding the levator palpebrae. We defined pupillary involvement as a poor pupillary light reaction on the side of the 3NP compared with the contralateral side and/ or anisocoria that was notably worse under bright than dim light and in which the larger pupil was on the side of the 3NP. We defined levator palpebrae involvement as upper lid ptosis of 2 mm or more in the primary position with a demonstrable de­crease in levator function. We defined EOM paresis as reduced ocular ductions, ocular misalignment, and sac­cadic slowing of the suspected paretic EOM. The degree of EOM paresis was classified as either complete ( duc­tions completely abolished) or incomplete ( ductions in­completely abolished). We subdivided incomplete palsy into two patterns: 1) diffuse, when all EOMs were paretic but incompletely so and 2) focal, when only some but not all the EOMs were paretic. A diagnosis of vasculopathic origin was given to pa­tients who had an acute isolated 3NP and complete spon­taneous recovery over 3 months and had a history of one or more of the following: age greater than 50 years, hypertension, diabetes mellitus, and hyperlipidemia. Other causes of 3NP were excluded by history, physical examination, blood tests, edrophonium test, and neu-roimaging in appropriate patients. No patient demon­strated aberrant regeneration of the oculomotor nerve acutely or at follow- up. RESULTS Fifty- eight instances of vasculopathic 3NP in 57 pa­tients were identified. Three instances ( in two patients) were excluded because of surgically altered pupils or the use of topical medications that could affect pupillary function. Therefore, 55 patients with vasculopathic 3NP constituted the population of our study. Ages ranged from 40 to 86 years with a mean of 66.5 years. Thirty-seven were men ( 67%) and 18 were women ( 33%). The interval from onset of symptoms to time of clinical pre­sentation ranged from 2 to 28 days with a mean of 13.4 days. Twenty patients underwent computed axial tomog­raphy, 28 patients had magnetic resonance imaging, and seven patients had both studies. Eleven patients had mag­netic resonance angiography, 13 patients had catheter angiography, and two patients had both. All neuroimag-ing tests were unremarkable except for one patient who had a remote occipital lobe infarction. At the time of initial examination, 42 patients ( 76%) had normal pupillary function and 13 patients ( 24%) had abnormal pupillary function. Of the 42 patients with pu­pil- sparing, 19 ( 45%) patients had complete EOM pa­ralysis and 23 patients ( 55%) had incomplete EOM pare­sis ( Table 1). These 23 patients with incomplete, pupil- sparing 3NPs constituted the specific subgroup of interest in this study. They accounted for 55% of our total population of vas­culopathic 3NPs. Twenty of these 23 patients ( 87%) had paresis of all the EOMs, but the weakness was incom­plete ( diffuse pattern of paresis). Three patients ( 13%) demonstrated paresis of only the SR and levator palpe­brae ( superior division) muscles with sparing of the re­maining EOMs ( focal pattern of paresis). Reexamination of these three patients in subsequent weeks showed no change in the pattern of EOM paresis until recovery. Because of concern regarding possible evolution of EOM weakness, the data were reanalyzed based on the timing of the first examination. Specifically, patients with partial, pupil- sparing 3NP seen within the first 10 days of onset of symptoms were analyzed separately, and their data were compared with the data from the entire group of similar patients. We found that 14 ( 61%) of 23 patients had been examined within 10 days of symptom onset; 12 ( 85%) had diffuse paresis, and two ( 15%) had focal paresis. COMMENT It is well established that in 3NP owing to compres­sion by a posterior communicating artery aneurysm, the pupil is typically involved ( 2^ t). Impairment of pupillary function is found in 86% to 95% of such patients, and, thus, pupillary examination and utilization of the " rule of the pupil" have been a critical step in the evaluation of an acute isolated 3NP ( 7,13). The high incidence of pupil­lary involvement is presumably related to the peripheral location of pupillary fibers within the oculomotor nerve, TABLE 1. Characteristics of vasculopathic third nerve palsy in 55 patients Pupil abnormal 13 ( 24%) Pupil normal 42 ( 76%) Complete EOM Palsy 19 ( 45%) Incomplete EOM Palsy 23 ( 55%) Diffuse Pattern 20 Focal Pattern 3 Total 55 J Neuro- Ophthalmol, Vol. 21, No. 4, 2001 258 S. SANDERS ETAL. making them particularly susceptible to extrinsic com­pression ( 14). In 1985, Trobe ( 12) pointed out that in previously reported cases of aneurysmal 3NP in which the pupil was normal, the EOM palsy was incomplete. In other words, in cases with an incomplete 3NP, pupil sparing could not safely rule out the possibility of aneurysm. In his follow-up editorial outlining several " footnotes to the pupil rule," Trobe advised: " do not apply the rule when the extraocu­lar palsy is incomplete. If some but not all of the oculo­motor- innervated muscles are completely impaired or if all of the muscles are impaired but only slightly, then pupil sparing may exist with aneurysm..." ( 13). How­ever, since then, there has been no further elaboration regarding any potential clinical significance to the dif­ference between these two patterns of incomplete 3NP. We wondered whether the pattern of EOM palsy might be helpful in determining the cause of a partial 3NP. More specifically, could the pattern of EOM palsy suggest whether aneurysmal compression or ischemic in­jury was the cause? To further investigate this, we stud­ied the pattern of EOM palsy in patients with vasculo-pathic incomplete 3NP. We found that 20 ( 87%) of 23 patients with pupil- sparing, partial 3NP demonstrated a diffuse pattern of EOM palsy in which all the EOMs were paretic but only partly so. Three patients demon­strated a focal pattern of paresis in which some of the EOMs were weak but others were spared. From our data, the diffuse pattern of EOM paresis appears to be the predominant form of pupil- sparing, incomplete 3NP of ischemic origin. In comparison, what is the predominant pattern of EOM weakness in patients with aneurysmal 3NP? To our knowledge, there is no comparable systematic analysis of patterns of EOM involvement in aneurysmal pupil-sparing 3NP. We therefore extracted individual case re­ports from the literature. Kissel et al. ( 7) described 51 patients in whom 3NP was the initial manifestation of a posterior communicating artery aneurysm. Of these pa­tients, seven ( 14%) had a pupil- sparing 3NP at presen­tation and in all seven, the 3NP was incomplete. These authors defined incomplete as " greater than 2 mm of ptosis, decreased range of motion in the appropriate di­rection with or without deviation of the eye in the pri­mary position, or a combination of these." There were no further specifics regarding the pattern of EOM paresis. Bartleson et al. ( 11) reported the ocular motor findings in 12 patients with incomplete 3NP owing to posterior com­municating artery aneurysm. Nine of the 12 patients had initial pupillary impairment, but three patients remained who had a pupil- sparing, incomplete 3NP. Two of these patients had only ptosis and symptomatic diplopia, and one patient was described as having " paresis of R N- III EOMs and ptosis." Good ( 9) and Greenspan and Reeves ( 10) have each reported a single case of ptosis as the sole sign of 3NP owing to a similarly located aneurysm. Be­cause the percentage of aneurysmal 3NP that spares the pupil is small, a large series is difficult to achieve. Based on the cases cited here, it appears that the focal pattern of EOM paresis may be the more predominant form of pu­pil- sparing, partial 3NP when owing to aneurysm. We examined some potential sources of error for our findings. First, we acknowledge that the diagnosis of a vasculopathic cranial neuropathy is a clinical one and thus, it is possible that our patient population inadver­tently contained patients with other causes of 3NP such as inflammatory disease. Second, because we included patients examined as late as 4 weeks from symptom on­set, we acknowledge that a focal pattern of EOM paresis might have been missed. Because Jacobson and Broste observed that the peak severity of ophthalmoplegia in patients with ischemic oculomotor palsies occurred at a median of 10 days, we reanalyzed our data by reviewing those patients who were examined within the first 10 days of symptom onset ( early group, n = 14) and com­pared the results with those from the entire group of patients with incomplete, pupil- sparing 3NPs seen within the first 4 weeks ( n = 23). Two of 14 patients ( 15%) of the early group had focal EOM paresis compared with 13% of the entire group. Furthermore, at subsequent fol­low- up, the pattern of EOM paresis in these two patients did not progress to involve other muscles before sponta­neous recovery occurred. Because the findings in the two groups were quite similar, we believe that it is valid to draw conclusions from the entire patient group. Third, we recognize that our hypothesis regarding two different patterns of partial EOM paresis based on two different etiologies lacks comparable data for patients with aneurysmal palsy. We remind the reader that our data for vasculopathic palsy represents just half of that hypothesis. The full implication of our data awaits a better description of the pattern of EOM paresis in an­eurysmal palsy. Where in the overall evaluation of an acute 3NP might our findings be helpful? In patients with a 3NP in whom the pupil is dilated or poorly reactive, the evaluation is generally straightforward; these patients should undergo timely and appropriate investigation for a possible aneu­rysm. Likewise, there is little controversy regarding ob­servational management of cases in which pupillary function is normal but EOM palsy is complete, as these are likely vasculopathic. These clinical scenarios con­form to the " rule of the pupil" ( 12,13). The problematic clinical scenario occurs when pupillary function is nor­mal and EOM palsy is incomplete. In such cases, the " rule of the pupil" has been found to be less reliable, and it is in this setting that we believe a diffuse pattern might, if future studies bear us out, tilt the physician toward suspecting a vasculopathic rather than an aneu­rysmal cause. REFERENCES 1. Richards BW, Jones FR Jr, Younge BR. Causes and prognosis in 4278 cases of paralysis of the oculomotor, trochlear and abducens cranial nerves. Am J Ophthalmol 1992; 113: 489- 96. 2. Rucker CW. Paralysis of the third, fourth and sixth cranial nerves. Am J Ophthalmol 1958; 46: 787- 94. 3. Rucker CW. The causes of paralysis of the third, fourth and sixth cranial nerves. Am J Ophthalmol 1966; 61: 1293- 8. / Neuro- Ophthalmol, Vol. 21, No. 4, 2001 EXTRAOCULAR WEAKNESS IN THIRD NERVE PALSY 259 4. Green WR, Hackett ER, Schlezinger NS. Neuro- ophthalmologic evaluation of oculomotor nerve paralysis. Arch Ophthalmol 1964; 72: 154- 67. 5. Keane JR. Aneurysms and third nerve palsies. Ann Neurol 1983; 14: 696- 7. 6. Kasoff I, Kelly DL Jr. Pupillary sparing in oculomotor palsy from internal carotid aneurysm. Case report. J Neurosurg 1975; 42: 713- 7. 7. Kissel JT, Burde RM, Klingele TG, et al. Pupil- sparing oculomotor palsies with internal carotid- posterior communicating artery aneu­rysms. Ann Neurol 1983; 13: 149- 54. 8. O'Connor PS, Tredici TJ, Green RP. Pupil- sparing third nerve palsies caused by aneurysm. Am J Ophthalmol 1983; 95: 395- 7. 9. Good EF. Ptosis as the sole manifestation of compression of the oculomotor nerve by an aneurysm of the posterior communicating artery. J Clin Neuroophthalmol 1990; 10: 59- 61. 10. Greenspan BN, Reeves AG. Transient partial oculomotor nerve paresis with posterior communicating artery aneurysm. A case re­port. J Clin Neuroophthalmol 1990; 10: 56- 8. 11. Bartleson JD, Trautmann JC, Sundt TM Jr. Minimal oculomotor nerve paresis secondary to unruptured intracranial aneurysm. Arch Neurol 1986; 43: 1015- 20. 12. Trobe JD. Isolated pupil- sparing third nerve palsy. Ophthalmology 1985; 92: 58- 61. 13. Trobe JD. Third nerve palsy and the pupil: footnotes to the rule. Arch Ophthalmol 1988; 106: 601- 2. 14. Kerr FWL, Hollowell OW. Location of pupillomotor and accom­modation fibers in the oculomotor nerve. Experimental observa­tions on paralytic mydriasis. J Neurol Neurosurg Psychiatry 1964; 27: 473- 81. J Neuro- Ophthalmol, Vol. 21, No. 4, 2001