Recovery of Third, Fourth, and Sixth Cranial Nerve Palsies in Pituitary Adenoma and Meningioma Patients

Supplemental Digital Content is Available in the Text.


Patients
This retrospective cohort study included meningioma and PA patients from the outpatient clinic of the Department of Ophthalmology at the Leiden University Medical Center (LUMC), a tertiary referral center for patients with these pathologies, from January 1, 1978 to January 31, 2021.The following inclusion criteria were used: patients $18 years with surgery-or tumor-induced OMP(s) of CN III, CN IV, and/or CN VI.Notably, multiple OMPs occurring in the same patient were counted separately.Exclusion criteria were: (I) incomplete documentation of orthoptic/ ophthalmologic examination (,2 orthoptic follow-up examinations and/or incomplete follow-up because of loss to follow-up), (II) mechanical restriction of extraocular muscles because of tumor size, orbital involvement, or surgery-related edema, (III) a history of diplopia or strabismus (operations) unrelated to the tumor, and (IV) other diseases or complications causing OMPs (e.g., severe complicated diabetes mellitus [DM]).The study was approved by the LUMC Medical Ethics Committee, and a verbal declaration of nonobjection was obtained from all patients.

Study Design
Variables and outcomes were assessed at different time points (See Supplemental Digital Content, Figure 1, http://links.lww.com/WNO/A775).Summarizing, patients underwent follow-up until OMP recovery, or for a minimum duration of 18 months (3-4 months intervals).If no recovery was achieved within 18 months, follow-up was extended until recovery or last moment of data collection, January 31, 2021.All ophthalmic and orthoptic assessments were performed by an ophthalmologist and orthoptist during (follow-up) examinations.

Variables
Data were retrospectively collected, and included baseline characteristics (e.g., age, gender), tumor diagnosis (including type, CS involvement, and presence of apoplexy [radiological diagnoses]), relevant medical history (e.g., DM, hypertension 11 ), and treatment modalities.Notably, patients were rarely treated for the OMP specifically, because patients were treated for the cause of the OMP, that is, the meningioma or PA.Ophthalmologic evaluation included best-corrected visual acuity (BCVA) using a Snellen chart followed by logMAR conversion, and orthoptic assessment at baseline and during follow-up.

Orthoptic Assessment
Standard orthoptic evaluation consisted of ocular deviation and movement assessment (ODMA), duction movement assessment, and Hess screen testing.ODMAs were manually performed using the cover test in the 9 gaze directions.Duction movements were measured using a synoptophore (Clement Clarke, model 2003, Haag-Streit UK Ltd, Harlow, United Kingdom).Normal duction movements are abduction and adduction of $ 240 and 40°, respectively, and elevation and depression of $30°.In case of diplopia and adequate BCVA, Hess screen testing (Clement Clarke, Haag-Streit, UK Ltd) was manually performed and compared with the previous Hess chart.Patients were questioned for subjective diplopia.If eyes were blind/had significantly decreased vision (Snellen BCVA , 0.1), diplopia could not be assessed.

Outcomes
A 5-tier recovery scale was created to evaluate OMP recovery, as shown in Table 1.All categories were defined using the extent of subjective diplopia, duction restrictions, restrictions on Hess charts, and misalignment of the eye(s).In case of CN III palsies, mydriasis, ptosis, and accommodative disorder were not taken into account, because eye motility has been demonstrated to be more sensitive for (prolonged) CN III recovery evaluation. 8MP recovery was defined as recovery scale's Category 1 and 2 combined (Table 1).Recovery rates were determined at 6, 12, and 18 months, and 24 months of follow-up in 18 patients (11 meningioma and 7 PA patients, respectively).Recovery time was calculated from the day of OMP diagnosis until the date of the first follow-up appointment at which recovery was reported.In the absence of orthoptic examinations, clinical reports were used to evaluate OMP recovery.

Statistics
Data-collected in Castor (Castor EDC, Amsterdam, NL)were analyzed using IBM SPSS version 25 (SPSS Inc., Chicago, IL).Data were reported as number of patients (N), number of eyes, or palsies (n), with percentages (%) for categorical variables.Continuous variables were reported as a mean with an SD, or median with an interquartile range (IQR).Chi-square test or Fisher exact test (categorical variables), and independent-samples t test or Mann-Whitney U test (continuous variables) were used to compare patient groups (all results presented as meningioma vs PA patients).Mixed model analysis was used to determine differences in BCVA between meningiomas and PAs for OMP diagnosis and recovery.Kaplan-Meier curves were used to analyze mean OMP recovery time (expressed as median ± SEM).Differences between tumor types were calculated using a Logrank test (Mantel-Cox).To analyze predictive factors, multivariable logistic regression (recovery rate at 18 months), and Cox proportional hazards regression models (recovery time) were used.Statistical significance was set at P , 0.007 (false discovery rate adjusted P-value). 12

Recovery Rates of Ocular Motor Nerve Palsies
Recovery rates for the different follow-up periods are shown in Table 3.At 18 months, recovery rates for all OMPs were significantly lower in treated meningioma patients compared with treated PA patients: CN III 37.5% vs 95.8%, P = 0.0015; CN IV 0% vs 100%; and CN VI 40% vs 100%, P = 0.007, respectively.No differences between the 2 untreated tumor groups were observed.
Recovery rates (both tumor types combined) were observed to be highest for CN VI palsies (84.1%) and lowest for CN III palsies (65.1%).Recovery after 18 months was still observed in 2 meningioma patients' eyes (one with a CN III palsy, and one with a CN VI palsy), compared with one eye in a PA patient with a CN III palsy.Recovery rates of tumor-induced OMP did not differ significantly from surgery-related OMP (P = 0.460).Moreover, in PA patients, no differences in recovery rates were observed between the nonapoplexy and apoplexy group (P = 1.00).

Original Contribution
Prognostic Factors for Recovery Rate Following correction for age, sex, hypertension, and DM, no prognostic factors were found for CN III and CN VI palsies, as reported in Table 4. Within meningioma patients with CN IV palsy, only one patient recovered, and therefore no comparative and regression analyses were performed.

Recovery Time of Ocular Motor Nerve Palsies
Kaplan-Meier curves showing OMP recovery of meningioma and PA patients are presented in Figure 1.Median recovery time of all OMPs combined was significantly longer in meningioma patients compared with PA patients (37.9 ± 14.3 vs 3.3 ± 0.1 months, P , 0.001, Fig. 1A).No significant difference in median recovery time was observed between the 3 cranial nerves, as shown in Figure 1B (CN III: 5.1 ± 3.0, CN IV: 3.3 ± 0.5, CN VI: 3.8 ± 0.3 months, P = 0.339).Furthermore, palsies of treated PAs recovered faster than palsies of treated meningiomas (P , 0.001, Fig. 1C), whereas no significant difference was observed in the untreated group (P = 0.015, Fig. 1D).Median recovery time of tumor-induced OMPs did not differ substantially from surgery-related OMPs (3.8 ± 0.7 vs 4.0 ± 2.5 months, P = 0.823), and no differences in recovery time was observed between the nonapoplexy and apoplexy group of PA patients (P = 0.798).As shown in Table 4, no prognostic factors were found for recovery time of CN III and CN VI palsies.

DISCUSSION
Our study demonstrates that OMP recovery was more favorable in PA compared with meningioma patients.Virtually, all OMP in patients with PA recovered partially or completely after 18 months, compared with less than half of OMP in patients with meningioma.The difference between these 2 patient groups was achieved in the treated patients.Clinical recovery was occasionally identified after 18 months, reflecting the potential of OMP recovery after prolonged follow-up.
The observed higher OMP recovery rates and concomitant BCVA recovery in patients with PA compared with patients with meningioma are in line with previous studies, 5,6,[8][9][10]13,14 which could hypothetically be explained by apoplexy, or differences in treatment strategies. As excted, apoplexy was observed solely in patients with PA, because these tumors have high metabolic demands and poor vascular densities.15 Meningiomas are often slow-growing and highly vascularized tumors, often requiring more invasive surgery compared with PAs.16 Moreover, OMP in apoplectic patients with PA has been reported to be more likely to recover than OMP in nonapoplectic patients.13 Apoplexy could, therefore, explain the higher recovery rates in patients with PA. Surprisingly, recovery rates were not affected by the presence of apoplexy in this study.
A significant disparity in treatment modalities for the tumor types was observed.Total resection rates have been

Original Contribution
reported to be similar in patients with meningioma and PA. 17,18However, 80%-90% of the meningiomas in those studies were not complicated by CS invasion, whereas most meningioma patients had CS invasion in this study (i.e., 68%). 6,16Treated meningioma patients had lower recovery rates than treated patients with PA in this study.Unexpectedly, treatment did not influence recovery rates after correction for potential confounding factors.However, regarding recovery rate and recovery time, a distinction should be made between the tumor-induced OMPs and surgery-related OMPs, because treatment would only benefit tumor-induced OMPs.Early treatment in tumor-induced OMPs has been identified as favorable for recovery rate in CN III palsies. 13Early treatment in meningiomas is often not achievable because of the insidious disease onset, resulting in treatment delay. 16Consequently, treatment timing in meningiomas remains to be systematically investigated.
Next to greater recovery rates, substantially shorter recovery times were observed in patients with PA compared with meningioma.Most OMP recoveries occurred within 6 months, with the prospect of recovery thereafter being rare in patients with meningioma, as published previously. 6,8,14Several differences between patients with PA and meningioma should be considered regarding the disparity in recovery time: tumor growth rate and patterns, affected cranial nerves, and OMP etiology.First, meningiomas tend to cause prolonged cranial nerve compression compared with PAs, resulting in delayed recovery, or no recovery at all, despite tumor treatment. 6,7Second, because CN IV has an anatomical predisposition to be affected by tumor compression (long, thin nerve), and CN VI is most likely to be damaged during surgery (because of its location close to skull base), we would expect that these nerves would have the longest recovery times. 6,19However, no statistically significant differences in FIG. 1. Recovery time of OMPs.OMP recovery differences between cranial nerves and between meningiomas and pituitary adenomas are shown in the Kaplan-Meier curves (data were shown as median ± SEM). A. OMP recovery probability between all meningioma and pituitary adenoma patients independent of affected ocular motor nerve (meningioma: 37.9 ± 14.3 vs pituitary adenoma: 3.3 ± 0.1 months, P , 0.001).B. OMP recovery for the individual cranial nerves showed a median recovery time of 5.1 ± 3.0 for CN III, 3.3 ± 0.5 for CN IV, and 3.7 ± 0.3 months for CN VI, respectively (P = 0.339).C. OMP recovery between treated meningiomas and pituitary adenomas is shown.Median recovery time in meningioma patients was 37.9 ± 24.1 months compared with 3.2 ± 0.1 months in pituitary adenoma patients (P , 0.001).D. OMP recovery between untreated meningiomas and pituitary adenomas is shown.Median recovery time of meningiomas was 37.9 ± 19.5 months and 3.5 ± 0.1 months in pituitary adenomas (P = 0.015).CN III, third cranial nerve; CN IV, fourth cranial nerve; CN VI, sixth cranial nerve; OMP, ocular motor nerve palsy; SEM, standard error of median.recovery times were observed between the different cranial nerves, because the number of patients in this study was not large enough to reach statistical significance.Moreover, apoplexy-induced OMPs (solely occurring in PA patients) would recover faster with adequate surgical intervention compared with other OMPs because of the acute onset, 8 a finding not observed in the present study.
Several limitations of the present study must be considered.First, the sample size of our study was limited, because OMPs are uncommon manifestations in patients with meningioma and PA. 4,8Second, previous studies adopted different criteria and definitions for recovery, hampering comparison between studies. 6,20In future studies, clinical recovery, in addition to complete and partial recovery, should be included, that is, using the proposed recovery scale (Table 1).Despite patients with meningioma being less likely to recover completely, a significant percentage of these patients reached the clinical recovery state, implying that OMP diagnosis no longer resulted in impairments in daily life in these patients.In addition, in-depth analyses (e.g., recovery time, prognostic factors) of CN IV palsies in patients with meningioma were not performed, because only one patient's OMP recovered.
In conclusion, this retrospective study showed that OMP recovery rates were more favorable in PA patients compared with meningioma patients independent of OMP etiology, especially in treated patients.Moreover, OMPs in PA patients recovered faster than OMPs in patients with meningioma.With these new insights in OMP recovery rates and times, physicians can provide more accurate prognoses, and therefore more appropriate follow-up strategies for patients with OMP caused by meningioma or PA.

TABLE 1 .
Recovery scale for ocular motor nerve palsiesThe recovery scale categories were used to evaluate OMP recovery in meningioma and pituitary adenoma patients.Normal ductions are 240°, 40°, 30°and 30°for abduction, adduction, elevation and depression, respectively.For further analyses, OMP recovery was defined as Category 1 and 2 combined.
*Objective restrictions without subjective symptoms interfering with daily activities.† Patients with decreased vision or blind eyes did not necessarily suffer from diplopia despite substantial duction restrictions.OMP, ocular motor nerve palsy.408 Lu et al: J Neuro-Ophthalmol 2024; 44: 406-413 Original Contribution

TABLE 2 .
Baseline characteristics of meningioma and pituitary adenoma patients with ocular motor nerve palsiesBaseline characteristics of both patient groups are shown.Because several variables differ between 2 eyes of one included patient, several characteristics are reported for of all included eyes (28 eyes in the meningioma patients vs 36 eyes in the pituitary adenoma patients).
*Differences between patient groups were analyzed with Pearson Chi-square.†Differences between patient groups were analyzed with independent samples t test.‡Differences between patient groups were analyzed with Fisher exact Test.§Differences between patient groups were analyzed with Mann-Whitney U test.kOf the 18 PA patients with apoplexy, 16 patients were treated with surgery, and 2 were treated conservatively.¶This patient received adjuvant pharmacologic treatment with temozolomide because of an aggressive pituitary adenoma (Cushing).BCVA, best-corrected visual acuity; CN III, third cranial nerve; CN IV, fourth cranial nerve; CN VI, sixth cranial nerve; N, number of patients; n, number of eyes; NA, not assessed; OMP, ocular motor nerve palsy; RT, radiotherapy.

TABLE 3 .
Recovery rates of ocular motor nerve palsies in meningioma and pituitary adenoma patients Recovery rates of OMPs at 6, 12, 18, and 24 months in meningioma and pituitary adenoma patients are shown.*Treated indicates treated for the tumor with palsy.† P-value of recovery rates between treated meningiomas and pituitary adenomas with palsy.‡ P-value of recovery rates between untreated meningiomas and pituitary adenomas with palsy.§ Differences between tumor groups and treatment groups were analyzed with Fisher exact test for the main outcome point (i.e., 18 months).CN III, third cranial nerve; CN IV, fourth cranial nerve; CN VI, sixth cranial nerve; NA, not applicable; OMP, ocular motor nerve palsy.

TABLE 4 .
Prognostic factors for recovery rate and recovery time in the study population at 18 months Prognostic factors for recovery rate and recovery time were assessed in CNII and CN VI ocular motor nerve palsy in the total study population.Logistic multivariate regression model and Cox regression model was used to analyze the significance of variables.*Referencesex was male sex.†Tumor type was defined as meningioma vs pituitary adenoma patients.‡