Journal of Neuro-Oncology

, Volume 116, Issue 2, pp 381–386

Skull base meningiomas: neurological outcome after microsurgical resection

  • Judith Scheitzach
  • Karl-Michael Schebesch
  • Alexander Brawanski
  • Martin A. Proescholdt
Clinical Study

DOI: 10.1007/s11060-013-1309-x

Cite this article as:
Scheitzach, J., Schebesch, KM., Brawanski, A. et al. J Neurooncol (2014) 116: 381. doi:10.1007/s11060-013-1309-x

Abstract

Microsurgical resection is the primary treatment of skull base meningiomas. Maximal resection provides the best tumor control rates but can be associated with high surgical morbidity. To understand the relation between extent of resection (EOR) and functional outcome we have analyzed the neurological improvement and recurrence rate in a large consecutive series of skull base meningioma patients. In addition, we defined anatomical and biological factors predictive for recurrence and overall outcome. We investigated 226 skull base meningioma patients receiving tumor resection in our institution. The most frequent location was the medial sphenoid ridge (29.6 %). EOR was rated according to the Simpson scale. Overall performance was measured by the Karnofsky performance score (KPS); neurological deficits were quantified using the Medical Research Council Neurological Severity Score (MRC-NPS). Complete resection was achieved in 62.8 % and the EOR was significantly correlated to tumor location. The morbidity and mortality rate was 32.1 and 2.7 % respectively, new permanent neurological deficits occurred in 3.5 % of all patients. From all patients with focal neurological deficits, 60.1 % experienced significant improvement. Both the MRC-NPS and the KPS significantly improved from the preoperative status to discharge, however the improvement rate was dependent on the tumor location. Recurrence rate was 15.5 %; tumor size, bone- and venous sinus infiltration, WHO grade, poor EOR but not MIB-1 labeling index were independent factors predictive for recurrence. Microsurgical resection of skull base meningiomas improves neurological impairment in the majority of patients. Specific risk factors for recurrence require consideration for postoperative management.

Keywords

Skull base Meningioma Recurrence Prognosis Surgery Radiosurgery 

Introduction

Meningiomas are mostly benign tumors that originate from the arachnoidal coverings of the brain [1] and account for 13–26 % of all intracranial tumors [2]. About 30 % of all meningiomas arise from the skull base [3], these tumors frequently lead to serious and potentially lethal consequences due to their intracranial location [4]. The primary treatment modality for skull base meningiomas is surgical resection, with radiation being the only accepted form of adjuvant therapy [5, 6]. Several studies have demonstrated a strong correlation between the extent of resection and the frequency of recurrence in skull base meningiomas [7, 8, 9, 10]. However, complete surgical resection of skull base meningiomas is frequently limited due to their intimate relationship to the brain stem, neurovascular structures and cranial nerves [11, 12]. Radical surgical strategies can cause high morbidity and poor overall outcome in these patients [13, 14]. To understand the impact of surgical resection on the functional outcome, we have analyzed the neurological improvement and recurrence rate in a large consecutive series of skull base meningioma patients. In addition, we determined factors predicting early tumor recurrence in this patient population.

Materials and methods

We retrospectively analyzed 226 patients (170 female/56 male) with skull base meningiomas receiving microsurgical tumor resection between 2003 and 2011 in our institution. The mean age was 58.7 years (range 10.4–86.3), the median preoperative Karnofsky Score was 90 %, (range 40–100). Follow up was completed up to March 2012 by reviewing outpatient records and contacting the patient’s primary physician. The mean follow up time was 29.9 months; 17 patients (7.5 %) were lost to follow up. The study was conducted in accordance to the ethical standards of the Helsinki Declaration and approved by the local ethics review board (protocol 11-101-0307). All resections were performed by a team member of staff neurosurgeons specialized in skull base surgery. To assess intraoperative findings, the surgery reports were reviewed and the extent of resection was graded using the Simpson classification. To evaluate the soft tissue extension pattern, preoperative MRI scanning following a standardized protocol was performed. Bone infiltration was analyzed on high resolution CT scans at bone window levels. After surgery, patients received a CT scan within 24 h to rule out intracranial hemorrhage or severe brain edema. Follow up imaging consisted of a base—line control MRI scan 3 months after surgical resection which was performed in 93.8 % of all patients. The extent of resection based on the 3 months MRI imaging was performed by an independent neuro—radiologist. Follow up consisted of control MRI scans as well as a standardized neurological exam using the Medical Research Council Neurological Performance Status Scale (MRC-NPS, Table 1) on an annual basis. In addition, visual acuity, hemiparesis, aphasia and signs of increased intracranial pressure (ICP) were recorded separately using a four-step rating scale (resolved, improved, stable, worsened) and reported at three time points: preoperative, at discharge and at the last follow up exam. A new postoperative neurological deficit was defined as permanent if it was not fully dissipated at last follow-up. Histologic diagnosis of the tumor samples was performed by an independent pathologist. The histology as well as grade of malignancy according to the WHO grading system are described in Table 2. The MIB-1 labeling index was recorded as the percent of all positively stained tumor cell nuclei of 1,000 tumor cell nuclei evaluated in at least four high magnification (400×) vision fields and in the areas with the greatest degree of immunostaining. In a situation of incomplete resection (Simpson grade III–V) recurrence has been defined as significant enlargement of the residual lesion, whereas after complete resection (Simpson grade I and II) major changes in MRI scanning indicative for recurrent tumour growth were viewed as recurrence. All patients with subtotal resection or biopsy corresponding to Simpson grade III–V received adjuvant conventional radiotherapy utilizing intensity modulated radiation therapy technology with a median dosage of 54 Gy, fractionated into a daily dosage of 1.8 Gy; no patient received chemotherapy. Comparative statistical analysis was performed for rates and proportions using Chi square analysis or a two tailed Fisher Exact test. Differences in time to progression was calculated by the Kaplan–Meier method. For univariate analysis, Mann–Whitney Rank Sum test and ANOVA on ranks was computed and multivariate statistics of independent factors predicting tumor recurrence was performed by calculating a multiple logistic regression analysis (SigmaStat Version 3.5, Systat Software Inc. Chicago, IL, USA)
Table 1

Research council-neurological performance status scale (MRC-NPS)

Grade

Performance

1

No neurological deficit

2

Some neurological deficit but function adequate for useful work

3

Neurological deficit causing moderate functional impairment e.g. ability to move limbs only with difficulty, moderate dysphasia, moderate paresis, some visual disturbance

4

Neurological deficit causing major functional impairment e.g. inability

to use limbs, gross speech or visual disturbances

5

No useful function-inability to make conscious responses

Table 2

Extent of resection expressed by Simpson grade and improvement rate of the MRC-NPS score of the analyzed skull base cases stratified by anatomical localization

Location

n

%

Simpson 1

Simpson 2

Simpson 3

Simpson 4

Simpson 5

Improvement rate (%)

Medial sphenoid ridge

67

29.6

31.3

28.3

10.5

28.4

1.5

4.5

Olfactory groove

66

29.2

36.4

43.9

19.7

0

0

12.1

Petroclival

32

14.2

12.5

28.1

12.5

40.6

6.3

6.2

Tentorial

31

13.7

22.6

29.0

9.7

35.5

3.2

6.5

Tuberculum sellae

11

4.9

18.2

27.2

27.3

18.2

9.1

9.1

Foramen magnum

11

4.9

9.1

27.2

18.2

45.5

0

0.0

Lateral sphenoid ridge

8

3.5

62.5

25

12.5

0

0

62.5

Results

The anatomical location of all tumors is described in Table 2; the most frequent locations were the medial sphenoid ridge (29.6 %). The majority of the cases (93.9 %) were classified as grade I WHO, 4.8 % as grade II and only 1.3 % as grade III (Table 3). Correspondingly, the mean MIB-labeling index of the entire population was low with an average of 2.8 % (range 1.0–20.0 %). The mean tumor diameter was 3.5 cm (range 0.8–10.3 cm). Infiltration of the venous sinuses was recorded in 21.2 % and infiltration of the adjacent bone structures in 48.2 % of all cases. Two patients (2.7 %) died in the immediate postoperative phase due to myocardial infarction and stroke, respectively. Significant surgical morbidity occurred in 32.3 % of all cases, the most frequent cause being intracranial hemorrhage (10.6 %) requiring evacuation in two cases followed by csf fistula (8.4 %), wound infection (6.6 %), pulmonary embolism (4.0 %), stroke (1.8 %) and myocardial infarction (0.9 %). New neurological deficits occurred in 14.2 % of all patients, which remained permanent in 3.5 %. In detail, new visual deficit, hemiparesis and aphasia occurred in 6.5, 5.3 and 2.2 % and remained permanent in 2.3, 1.8 and 0.4 % respectively. Complications of adjuvant radiotherapy were encountered in two patients (2.4 %) displaying enhanced white matter edema in the follow-up MRI scans, however without worsening of the neurological status. To test for the influence of age on the morbidity rate, we stratified our patient population based on age older than 70 years. Neither the surgical complications (p = 0.65) nor the rate of postoperative neurological worsening (p = 0.54) was significantly different in the two age groups. Complete resection corresponding to Simpson grade I and II was achieved in 62.8 % of all cases. Subtotal resection (grade III and IV) was performed in 34.5 %, with only 2.7 % of all patients receiving a biopsy corresponding to grade V. We found a strong correlation between the Simpson grading and post-operative imaging as an objective EOR measure. In patients receiving a complete resection corresponding to Simpson grading I and II, the initial postoperative MRI scanning failed to show residual tumor mass, confirming the surgeon’s impression in all cases. Conversely, in all cases corresponding to Simpson grade III–V, residual tumor formation was detected. Further, the EOR differed significantly within the anatomical locations. The best results were obtained in lateral sphenoid ridge and olfactory groove tumors, whereas meningiomas of the petroclival region and the foramen magnum showed the worst Simpson grades (p < 0.001; Table 2). In contrast, neither size nor consistency of the tumor was significantly correlated to the extent of resection (p = 0.131 and 0.144 respectively). The MRC-NPS score significantly improved from the preoperative status to discharge (2.02 vs. 1.86; p < 0.001; Fig. 1b), and remained stable at follow up (p = 0.110). In 64.6 % of all patients a significant improvement of the MRC-NPS was achieved at follow up. The KPS also improved significantly both between preoperative and status at discharge as well as at follow up (81.0 vs. 85.5 vs. 87.4; p < 0.001; Fig. 1a). From all patients presenting with focal neurological deficit not including increased ICP (83.1 %), 60.1 % experienced a significant improvement at follow up. More specifically, at follow up cranial nerve palsy, which occurred in 121 patients (53.5 %; CN I: 9; CNII: 45; CN III: 5; CNIV: 3; CN V: 11; CN VI: 4; CNVII: 2; CN VIII: 24; CN IX: 5; multiple: 13) was improved in 87 patients (71.9 %). 52 patients (23.0 %) had visual field deficits, which remained stable in 27 (51.9 %), improved in 15 (40.4 %) and worsened in 4 (7.7 %) patients. Hemiparesis was detected in nine patients preoperatively (4.0 %) and improved significantly in seven patients (77.7 %). In the six patients (2.7 %) presenting with aphasia, two remained unchanged (33.3 %) and 4 (66.6 %) improved significantly. Signs of increased ICP, recorded in 24 patients (10.6 %), were resolved in all patients at discharge. The extent of resection had no influence on the improvement rate of both the MRC-NPS score (p = 0.688) and the KPS (p = 0.174). However, the improvement of the MRC-NPS and KPS scores differed significantly with tumor location, with olfactory groove and lateral sphenoid wing tumors showing the best and foramen magnum meningiomas the worst functional results at follow up (p < 0.001, Table 2). Tumor recurrence was detected based on serial control MRI scanning as described in the methods section. Recurrence was observed in 35 patients (15.5 %); the mean time to recurrence was 24.8 months (range 2.4–54.4 months). The extent of resection correlated significantly with the time to recurrence (p = 0.009; Fig. 2). All patients with recurrent tumors underwent second resection and 20 % received additional radiotherapy. Univariate analysis revealed the extent of resection (Table 4), tumor size, bone and venous sinus infiltration, as well as WHO grade and a high MIB-1 labeling index as factors associated with a higher risk of recurrence (p < 0.05). In contrast, age (p = 0.629) and tumor location (p = 0.301) was not predictive for recurrence. Multivariate analysis confirmed the univariate results except for the MIB-1 labeling index, which was not significantly associated with recurrence in the multivariate model (Table 5).
Table 3

Histological classification of the analyzed skull base meningioma cases

Histology

WHO grade

n

%

Meningothelial

I

146

64.6

Fibroblastic

I

37

16.4

Transitional

I

29

12.9

Atypical

II

11

4.8

Anaplastic

III

3

1.3

Fig. 1

Improvement of KPS (A) and MRC-NPS (B) in patients receiving microsurgical resection of skull base meningiomas. The KPI improves both immediately after surgery and at follow-up, the MRC-NPS indicates good recovery after surgery and remains stable at follow-up (Bar graph shows mean values plus standard error of the mean; * p < 0.001, repeated measure ANOVA on ranks)

Fig. 2

Kaplan - Meier curves illustrating the time to recurrence stratified by resection grade according to the Simpson classification. Log rank analysis demonstrated a significant correlation between extent of resection and time to recurrence (p < 0.009)

Table 4

Relation between extent of resection expressed by Simpson grading and recurrence rate

Simpson grade

Extent of resection

Recurrence rate (%)

I

Complete resection including underlying bone and associated dura

6.8

II

Complete removal with coagulation of dural attachment

9.6

III

Complete removal without resection or coagulation of dural attachment

19.2

IV

Partial resection

28.8

V

Biopsy

50.0

Table 5

Uni- and multivariate analysis of factors predictive for tumor recurrence

Variable

Univariate (p)

Multivariate (p)

OR

5 % CI lower

95 % CI upper

WHO grade

0.033

0.022

3.532

1.197

10.424

MIB labeling index

0.020

0.550

0.929

0.731

1.182

Size

0.001

0.001

1.880

1.358

2.603

EOR

0.001

0.008

1.773

1.162

2.704

Sinus infiltration

0.002

0.027

3.584

1.152

11.146

Bone invasion

0.001

0.001

15.030

3.624

62.324

Discussion

Choosing the adequate treatment strategy for skull base meningiomas represents a formidable challenge. Compared to supratentorial meningiomas, these tumors grow slower and display a less aggressive biology with a significantly lower risk of malignant degeneration [15, 16, 17]. A subset of these lesions remains clinically silent [18], and can behave rather indolently with minimal symptoms even when left untreated [19]. However, skull base meningiomas frequently cause severe neurological morbidity due to compression of neurovascular structures, which requires immediate and definite treatment [20]. Surgery plays an indispensable role in the management of skull base meningioma, with a clear correlation between extent of resection and both recurrence rate and overall survival [7]. This has caused increasingly aggressive attempts to remove skull base meningiomas [21] frequently leading to unacceptable high mortality and morbidity rates [22]. Our results demonstrate that neurological impairment improved in the majority of the patients, with a new permanent postoperative deficit in only 3.5 % of all patients. Interestingly, we did not find a correlation between the extent of resection and the neurological improvement rate. Although several studies have not detected a correlation between extent of resection and neurological outcome [23, 24], Adachi et al. [4] have found a negative effect of extensive resection on neurological outcome in a subgroup of patients. Our results indicate that careful and complete decompression of eloquent structures is sufficient for the relief of clinical symptoms rather than attempting to radically resect the entire tumor shell [12, 25]. In addition, the postoperative improvement rate differed significantly between tumor locations, emphasizing the impact of location on the functional recovery rates in skull base meningiomas. The extent of resection differed significantly between tumor locations with the worst results in tumors of the petroclival region and the foramen magnum, which is in accordance to previous studies [26, 27]. In our patient cohort, extensive infiltration of neurovascular structures dictated a limited resection in those cases. The recurrence rate in our study population was 15.5 %, which is within the range reported in the literature [26]. Factors correlating with recurrence in the univariate analysis were poor Simpson grade, tumor size, bone and venous sinus infiltration, WHO grade and a high MIB-1 labeling index. Surprisingly, in the multivariate model, MIB-1 labeling index was not found to be an independent factor predicting recurrence. Although a number of studies have found a significant correlation of high MIB-1 labeling indices with tumor recurrence [24, 28, 29], others have not [30, 31, 32]. Besides the definition of positive staining which varies between different institutions [29] a significant regional heterogeneity of cell proliferation in meningiomas has been demonstrated [33]. The choice of representative tumor areas for MIB-1 index calculation therefore becomes an important source of interobserver variability. In our study, we have analyzed the MIB-index in a multivariate model with stratification for the WHO grade. Within the WHO grade I group, invariably showing low indices, it becomes difficult to distinguish tumors at risk for recurrence. This has led to the formulation of a threshold value of 3 % for grade I tumors with a higher risk of recurrence [24, 27], although a larger meta analysis has determined exactly 3 % as the average for all grade I tumors [29]. Our results indicate that MIB-1 labeling undoubtedly has the potential to differentiate between benign and malignant meningiomas, but needs cautious interpretation in the individual tumor. Radiosurgery plays an increasing role in the multidisciplinary treatment of skull base meningiomas [5]. In our study, all patients treated with radiotherapy only received conventional external beam radiation. Although conventional radiotherapy has been reported as an efficient and safe adjuvant treatment for skull base meningiomas [34], it is conceivable that radiosurgery will improve the tumor control rates and clinical outcome especially in patients of older age, poor overall performance stage or harboring tumors in unfavorable locations [35]. Finally, we need to interpret our data concerning tumor recurrence carefully since the follow-up time has been relatively short for this tumor entity In conclusion, microsurgical resection of skull base meningiomas can lead to the improvement of neurological symptoms, however the functional recovery rate is depending on tumor location rather than the EOR. Bone and sinus infiltration as well as WHO grade and EOR are independent predictors for tumor recurrence. In order to define the optimal treatment strategy, a combination of microsurgical resection with radiation therapy and adequate follow up needs to be tailored to the individual patient in a multidisciplinary setting.

Conflict of interest

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  • Judith Scheitzach
    • 1
  • Karl-Michael Schebesch
    • 1
  • Alexander Brawanski
    • 1
  • Martin A. Proescholdt
    • 1
  1. 1.Department of NeurosurgeryUniversity of Regensburg Medical CenterRegensburgGermany

Personalised recommendations