Fluorescence-guided resection of glioblastoma multiforme utilizing 5-ALA-induced porphyrins: a prospective study in 52 consecutive patients

Full access

Object. It has been established that 5-aminolevulinic acid (5-ALA) induces the accumulation of fluorescent porphyrins in glioblastoma multiforme (GBM), a phenomenon potentially exploitable to guide tumor resection. In this study the authors analyze the influence of fluorescence-guided resection on postoperative magnetic resonance (MR) imaging and survival in a series of patients who underwent surgery in the authors' department.

Methods. Fifty-two consecutive patients with GBM received oral doses of 5-ALA (20 mg/kg body weight) 3 hours before induction of anesthesia. Intraoperatively, tumor fluorescence was visualized using a modified operating microscope. Fluorescing tissue was removed whenever it was considered safely possible. Residual enhancement on early postoperative MR imaging was quantified and related to each patient's characteristics to determine which factors influenced resection. Survival was analyzed using the Kaplan—Meier method and multivariate analysis was performed in which the Karnofsky Performance Scale (KPS) score, residual fluorescence, patient age, and residual enhancement on MR images were considered.

Intraoperatively, two fluorescence qualities were perceived: solid fluorescence generally reflected coalescent tumor, whereas vague fluorescence mostly corresponded to infiltrative tumor. Complete resection of contrast-enhancing tumor was accomplished in 33 patients (63%). Residual intraoperative tissue fluorescence left unresected for safety reasons predicted residual enhancement on MR images in 18 of the 19 remaining patients. Age, residual solid fluorescence, and absence of contrast enhancement in MR imaging were independent explanatory factors for survival, whereas the KPS score was significant only in univariate analysis. No perioperative deaths and one case of permanent morbidity were encountered.

Conclusions. The observations in this study indicate the usefulness of 5-ALA—induced tumor fluorescence for guiding tumor resection. The completeness of resection, as determined intraoperatively from residual tissue fluorescence, was related to postoperative MR imaging findings and to survival in patients suffering from GBM.

Abstract

Object. It has been established that 5-aminolevulinic acid (5-ALA) induces the accumulation of fluorescent porphyrins in glioblastoma multiforme (GBM), a phenomenon potentially exploitable to guide tumor resection. In this study the authors analyze the influence of fluorescence-guided resection on postoperative magnetic resonance (MR) imaging and survival in a series of patients who underwent surgery in the authors' department.

Methods. Fifty-two consecutive patients with GBM received oral doses of 5-ALA (20 mg/kg body weight) 3 hours before induction of anesthesia. Intraoperatively, tumor fluorescence was visualized using a modified operating microscope. Fluorescing tissue was removed whenever it was considered safely possible. Residual enhancement on early postoperative MR imaging was quantified and related to each patient's characteristics to determine which factors influenced resection. Survival was analyzed using the Kaplan—Meier method and multivariate analysis was performed in which the Karnofsky Performance Scale (KPS) score, residual fluorescence, patient age, and residual enhancement on MR images were considered.

Intraoperatively, two fluorescence qualities were perceived: solid fluorescence generally reflected coalescent tumor, whereas vague fluorescence mostly corresponded to infiltrative tumor. Complete resection of contrast-enhancing tumor was accomplished in 33 patients (63%). Residual intraoperative tissue fluorescence left unresected for safety reasons predicted residual enhancement on MR images in 18 of the 19 remaining patients. Age, residual solid fluorescence, and absence of contrast enhancement in MR imaging were independent explanatory factors for survival, whereas the KPS score was significant only in univariate analysis. No perioperative deaths and one case of permanent morbidity were encountered.

Conclusions. The observations in this study indicate the usefulness of 5-ALA—induced tumor fluorescence for guiding tumor resection. The completeness of resection, as determined intraoperatively from residual tissue fluorescence, was related to postoperative MR imaging findings and to survival in patients suffering from GBM.

Among factors determining the prognosis in patients suffering from malignant gliomas, the extent of resection remains controversial. Investigators in many studies acknowledge a benefit from extensive tumor resection,1–3,7,24,28,30,38–41 whereas others do not.10,17,18,20,23,26,31 However, many studies in which the concept of extensive cytoreductive surgery is challenged are based on the surgeon's personal impression,8,18,20,23,26,31 which may be prone to a large degree of inaccuracy.1,23 On the other hand, in studies incorporating postoperative imaging it has been reported that radical removal of contrast-enhancing tumor is achieved in fewer than 20% of patients,1,17,29,40 probably because of the uncertainty involved in identifying viable tumor margins during surgery.

We recently introduced a new method in which we use 5-ALA for intraoperative visualization of malignant glioma tissue, which is now being routinely used in our department.33–35 A metabolic precursor, 5-ALA leads to accumulation of fluorescent porphyrins in malignant glioma tissue, a phenomenon exploitable for improving tumor resection.35 We analyze the data obtained in a series of 52 consecutive patients with GBM who were treated according to a prospective protocol to determine whether the procedure, besides improving the completeness of resection, may afford a survival benefit for patients.

Clinical Material and Methods
Patient Population

Local ethical committee approval was granted for this study and the patient's informed consent was obtained in each case. From December 1995 to December 1998 all patients presenting at our department with MR studies suggestive of unifocal supratentorial malignant glioma were considered for inclusion in the study. All individuals 18 years of age or older were eligible if open surgery was the selected treatment option according to traditional criteria established at our department. This decision depended on clinical condition (KPS score ≥ 60), location of tumor (90% resection anticipated without endangering eloquent functions), and the absence of preexisting medical conditions that would have precluded use of general anesthesia or surgery. Patients who did not meet these requirements were treated with radiotherapy after tumor biopsy procedures. Exclusion criteria also included medical conditions prohibiting the administration of 5-ALA—that is, porphyria or renal (plasma creatinine > 2 mg/dl) or hepatic disease (plasma bilirubin > 3 mg/dl, prothrombin time < 60%, γ-glutamyl-transferase > 70 U/L).

Of a total of 66 patients, GBM (World Health Organization Grade IV) was diagnosed in 52. These patients form the basis of the present analysis. In the remaining 14 patients, Grade III tumors were diagnosed (11 anaplastic astrocytomas and one each of anaplastic oligodendroglioma, anaplastic astrooligodendroglioma, and anaplastic pleomorphic xanthoastrocytoma). Because of their small number and the diversity of diagnoses, patients suffering from Grade III tumors were not included in the present analysis. Table 1 contains a summary of the characteristics of patients with GBM who underwent surgery in which 5-ALA was used.

TABLE 1

Characteristics in 52 patients with GBMs

CharacteristicValue
no. of patients52 
sex 
 male36 
 female16 
age (yrs) 
 range31–69 
 average53.8 ± 10.1 
 median57 
KPS score 
 range60–100 
 median90 
 60–8023 
 90–10029 
tumor in eloquent 
 location 
  no28 
  yes24 

All patients were maintained on three 4-mg doses of dexamethasone per day for at least 3 days before surgery. Three hours before induction of anesthesia, patients were given oral doses (20 mg/kg body weight) of 5-ALA dissolved in 80 to 100 ml of tap water. Tumor resection usually began within 5 hours after drug administration.

Technical Principles

Tumor-related porphyrin fluorescence was visualized using a modified surgical microscope, as previously described.33 The prototype was equipped with a 300-W short-arc xenon light source that was switchable to violet-blue illumination by activating a foot pedal, which introduced a specially designed dielectric 440-nm short-pass filter into the illumination path. During white-light illumination, an attenuator reduced irradiance to approximately 50% to minimize the need for accommodation. A liquid light guide (diameter 4.8 mm, length 2.5 m) coupled to the microscope light inlet enabled low-loss light transmission. An electromagnetic filter switcher was used to introduce dielectrically coated 440-nm long-pass filters into the two observer light paths of the stereomicroscope. Filters were designed to transmit red porphyrin fluorescence as well as a fraction of backscattered blue excitation light necessary for distinguishing nonfluorescing tissue. Dimming the room lights helped to distinguish tissue porphyrin fluorescence because ambient light in the tumor cavity was observed to interfere with the fluorescence signal. In situ tissue porphyrin spectra were obtained with an optical multichannel analyzer in the spectral range of 550 to 750 nm.

For documentation, the camera outlet of the microscope was equipped with a three-chip color charge-coupled device camera optimized for red porphyrin fluorescence visualization by enhanced sensitivity in the wavelength range beyond 600 nm. Target integration increased image brightness.

Fluorescence-Guided Resection

We began tumor removal by using a conventional microsurgical procedure under white-light illumination, first removing tumor portions that were easily identified, including areas of necrosis. Whenever desired, the surgeon was able to change the illumination mode to identify porphyrin fluorescence and to continue the operation either under violet-blue light or by switching back to conventional illumination. With violet-blue illumination, necrotic tumor usually displayed no or only inhomogenous red fluorescence. Adjoining, perfused, and viable tumor tissue was distinguished by its deep red fluorescence, which we call “solid” in this report, whereas normal brain tissue was colored blue. Usually, a transition zone with clearly discernible, but lighter, that is, pink fluorescence, was encountered between solidly fluorescing tumor and non-fluorescing blue brain tissue. We call this fluorescence impression “vague” (Fig. 1).

Fig. 1.
Fig. 1.

Intraoperative photographs demonstrating tumor cavity viewed under conventional white light (left) and violet-blue illumination (right). Note different qualities of tumor fluorescence (necrosis, solid, vague).

The resection procedure depended on individual preference and included simple suction, bipolar cautery, monopolar cautery, or ultrasound aspiration. When we could no longer confidently distinguish residual tumor from normal brain tissue, violet-blue illumination was regularly used for resecting residual fluorescing tissue.

To clarify the differences between vaguely and solidly fluorescing tissue, cell morphology was studied in a subset of 60 hematoxylin and eosin—stained smear preparations. These specimens were examined for the presence of solid tumor tissue characterized by coalescent tumor cells, or infiltrating tumor characterized by large, abnormal nuclei with little or no cytoplasm within discernible neural parenchyma. In addition, cell density was classified as high, intermediate, or low. A total of 264 biopsy specimens were obtained in both fluorescent and nonfluorescent tissue regions at the tumor margins and fixed in buffered formaldehyde for correlative histological examination.

The aim of the operation was removal of all visible porphyrin fluorescence. For safety reasons, however, resection was discontinued when the surgeon was concerned about risks to eloquent functions (that is, speech, motor, or visual functions). In these cases, the anatomical location as well as the quality of residual fluorescence (solid or vague) were assessed and recorded by at least one of two authors (W.S. or C.G.), to ensure consistency of the data.

Postoperative Care

During surgery and within 24 hours of drug administration, patients were maintained in darkened surroundings to minimize the risk of phototoxic skin reactions. Within 72 hours after surgery, an MR image was obtained for each patient. Areas of residual contrast enhancement were identified by comparing T1-weighted noncontrastenhanced and contrast-enhanced axial sections (matrix 256 × 256 pixels, 5-mm slices, 1-mm gap; 0.1 mmol gadolinium-diethylenetriamine pentaacetic acid/kg body weight). Contrast-enhanced coronal and sagittal sections were available for clarifying inconclusive cases. Tumor volume was determined by fitting irregular tumor shapes with geometric figures in T1-weighted axial sections for calculation of tumor areas. These were then multiplied by slice thickness and summed to obtain the final volume of residual contrast enhancement. All patients were maintained on three 4-mg doses of dexamethasone per day, at least until early postoperative MR images had been obtained.

All patients were referred for standard radiotherapy (> 55 Gy) following the procedure with the exception of two individuals who were entered into the European Collaborative Boron Neutron Capture Therapy Trial. No restrictions were imposed regarding subsequent modes of therapy in case of tumor progression. These included repeated operation in nine patients, various regimens of salvage chemotherapy in 19, intracavitary radioimmunotherapy in 11, and interstitial brachytherapy in one. Nine patients received a combination of therapies.

Statistical Analysis

To identify factors influencing the degree of resection as determined from postoperative MR imaging, regression analysis was performed, with residual enhancement on MR imaging counting as a dependent variable. The following regressors and their possible influence on resection were tested: date of operation, that is, the chronological point within the series at which the operation was performed; proximity to or extension of tumor into eloquent regions; patient age; and KPS score. The influence of each factor was analyzed separately and together in a multivariate regression model.

The relationship between residual intraoperative fluorescence and residual contrast enhancement on early postoperative MR images was tested with the chi-square test, using contingency tables. These tables were also used for statistical corroboration of data obtained from morphological examination of tissue smears. Differences in residual tumor volumes on postoperative MR images in the 5-ALA—enhanced group were tested by analysis of variance. Changes in laboratory values were tested using the paired t-test.

For analysis of survival, time from surgery to death was analyzed as the primary endpoint. Surviving patients were censored as of the last date of contact (September 1999). Curves for survival time were constructed using the method of Kaplan and Meier.14 Apparent differences in survival time were tested by the log-rank and the Wilcoxon tests. For the analysis of age, patients were stratified according to whether they were older or younger than the median age of 57 years. Multivariate analysis was performed using the Cox proportional hazards model.6 The following covariates were considered: patient age, sex, KPS score (subgroups 70–80 or 90–100), residual intraoperative fluorescence (no, vague, or solid fluorescence), residual contrast enhancement on MR images (residual enhancement or no residual enhancement), volume of enhancement, tumor proximity to eloquent brain regions (subgroups eloquent or noneloquent), reoperation, and radioimmunotherapy. The different modes of chemotherapy were not tested for an influence on survival because of the lack of uniformity on which to base the analysis. The strategy adopted for model selection was similar to that proposed by Collett,5 including inspection of deviance and martingale residual plots for survival to test the proportional hazards assumption. Relative risks were calculated for each covariate that was included in the model.

An α-error probability not exceeding 5% was considered significant. All values are given as averages ± standard deviation. Statistical analyses were performed using commercially available software.

Sources of Supplies and Equipment

The 5-ALA was obtained from Medac, Hamburg, Germany. A prototype of the modified Zeiss OPMI-CS-NC surgical microscope was developed in a collaborative effort between the Laser Research Laboratory at the Klinikum Grosshadern, Carl Zeiss, Inc., Oberkochen, Germany, and Karl Storz GmbH, Tuttlingen, Germany. The xenon light source (D-Light 2013220), liquid light guide (495 FR), long-pass filters, and charge-coupled device camera (Tricam SL PDD PAL prototype) were all provided by Karl Storz GmbH. The optical multichannel analyzer (O-SMA II) was acquired from Spectroscopy Instruments GmbH, Gilching, Germany. The ultrasound aspirator (Sonoca) was purchased from Söring GmbH, Quickborn, Germany. The MR imager (1.5-tesla Siemens Vision) was obtained from Siemens, Munich, Germany. The Statview statistical software was purchased from SAS Institute, Cary, NC.

Results
Fluorescence Qualities

Under the operating microscope, the fluorescence intensity in all patients' lesions was strong enough to be readily perceived and was completely sufficient for confident tumor removal. No notable period of dark adaptation was necessary when switching to violet-blue illumination for continuing the operation. Blood was found to obscure the fluorescent signal but could easily be sucked away and did not interfere with resection.

However, tumor fluorescence was not homogenous. Rather, two clearly discernible fluorescence qualities could usually be distinguished. A vivid solid red fluorescence impression was given by viable, that is, perfused, tumor regions. This was surrounded by a transition zone into nonfluorescing brain tissue, characterized by less vivid, pink fluorescence (vague; Fig. 1). After removing solid, red-fluorescing tissue and the adjoining zone that featured vague fluorescence, normal blue brain tissue appeared. Areas of gross tumor necrosis showed no fluorescence. The zones with different tissue fluorescence were commonly and unequivocally identified. Their margins were easily detected with the operating microscope and there did not appear to be a gradual transition between zones.

Using in situ spectrography, solidly fluorescing tissue was distinguished by a strong protoporphyrin IX signal with peaks at 635 nm and 704 nm. The signal amplitude obtained in tissue with vague fluorescence was typically an order of magnitude lower (Fig. 2). Histologically, tissue with solid fluorescence was usually characterized by coalescent tumor cells and associated neovascularity, with cell cytoplasm accumulating reddish protoporphyrin IX fluorescence (Fig. 3 upper). Vaguely fluorescing tissue usually featured infiltrating tumor of medium to high density, with individual cells accumulating protoporphyrin IX fluorescence within their cytoplasm (Fig. 3 lower).

Fig. 2.
Fig. 2.

Graph showing intraoperative fluorescence spectra obtained from regions with solid and vague fluorescence. Peaks at 635 nm and 704 nm indicate protoporphyrin IX fluorescence.

Fig. 3.
Fig. 3.

Photomicrographs showing morphological features of tissue with solid and vague fluorescence. Upper Left: Fluorescence videomicroscopy image of smear preparation obtained from biopsy sample featuring solid fluorescence. Image shows confluent red protoporphyrin fluorescence with negatively contrasted nuclei and nonfluorescing proliferated capillaries. Upper Right: Periodic acid Schiff staining of the smear in upper left. Lower Left: Fluorescence videomicroscopy image of smear preparation obtained from biopsy sample, featuring vague fluorescence associated with various large abnormal nuclei in invaded brain tissue. Note interspersed white fluorescence of lipid droplets. Lower Right: Giemsa staining of the preparation in lower left.

Quantitative data obtained from the subset of 60 specimen smears demonstrated a preponderance of coalescent tumor tissue with medium to high cellularity in biopsy samples in lesions considered to display solid fluorescence, whereas vaguely fluorescing samples more often contained infiltrating tumor with medium to low cellularity (Fig. 4).

Fig. 4.
Fig. 4.

Bar graphs showing dependence of fluorescence accumulation on morphological features. Left: Coalescent tumor compared with infiltrating tumor (p = 0.0136, chi-square test, 60 patients). Right: Association of fluorescence impression with tumor cellularity (p = 0.0112, chi-square test, 60 patients).

Residual Intraoperative Fluorescence and Postoperative MR Imaging

In 17 of 52 patients, we were able to remove fluorescent tissue completely during surgery. In the other 35 patients some fluorescent tissue was left in situ. In these 35 patients, the quality of residual fluorescence was described as vague in 12 and solid in 23. However, on postoperative MR images, complete resection of contrast-enhancing tumor was observed in 33 (63%) of 52 patients. Of the 17 patients in whom complete removal of fluorescent tissue was achieved, 16 revealed no residual enhancement on postoperative MR images. In the remaining patient, a small region of residual enhancement was found. On the other hand, postoperative MR imaging was also devoid of residual enhancement in nine of the 12 patients with vague residual fluorescence but in only eight of the 23 patients with solid residual fluorescence (Fig. 5).

Fig. 5.
Fig. 5.

Pie chart illustrating the relationship between the presence and quality of residual, fluorescing tissue and residual enhancement on early postoperative MR images (MR+ = residual enhancement on MR imaging; MR− = no enhancement on MR imaging; for statistics see Table 2).

The anatomical location of residual contrast enhancement on postoperative MR images corresponded to the location documented intraoperatively for residual fluorescence in all cases, with the exception of one patient, in whom no residual fluorescence had been documented. In this patient, residual tumor on the MR image consisted of a small fold of enhancing tissue situated under an overlapping rim of cortical tissue, which may have been overlooked.

Volumes of residual contrast-enhancing tumor were greatest in patients with solid residual fluorescence (no residual fluorescence: 0.4 cm3, one patient; vague fluorescence: 1.23 ± 0.68 cm3, three patients; solid fluorescence: 4.43 ± 5.75 cm3, 15 patients).

To confirm statistically whether residual fluorescence predicted residual contrast enhancement on postoperative MR images, contingency tables (Table 2) were generated. The relationship between residual fluorescence (that is, both vague and solid) and residual contrast enhancement was highly significant (χ2 = 10.2, p = 0.0014). However, when residual solid fluorescence and residual enhancement were compared the relationship was stronger (χ2 = 14.6, p = 0.0001). Accordingly, predicting residual enhancement from residual fluorescence (vague and solid) was possible with 65% accuracy (sensitivity 95%, specificity 48%), whereas predicting residual enhancement from residual solid fluorescence was more accurate (accuracy 77%, sensitivity 79%, specificity 75%).

TABLE 2

Relationship between quality of tissue fluorescence and residual contrast enhancement on early postoperative MR images*

Fluorescence Quality
C1C2
VariableNoneVague or SolidNone or VagueSolidTotal
residual enhancement
no161725833
yes11841519
total1735292352

C1 = Contingency 1: no residual enhancement compared with vague or solid fluorescence; C2 = Contingency 2: no or vague residual enhancement compared with solid fluorescence.

p = 0.0014 according to chi-square analysis.

p = 0.0001 according to chi-square analysis.

Factors Influencing Degree of Resection

The volumes of residual contrast-enhancing tumor on postoperative MR images were analyzed to determine which factors might have influenced the decision to resect more or less radically. Therefore, multivariate regression analysis was performed, considering age, KPS score, tumor proximity to eloquent regions, and date of operation (that is, the chronological point at which the operation was performed). Tumor proximity to eloquent regions was identified as the only factor with independent influence on residual tumor volumes on postoperative MR images (Table 3). Residual volumes on MR images in patients with tumors in eloquent regions were significantly greater (2.75 ± 5 cm3) than in patients with tumors in noneloquent regions (0.22 ± 0.49 cm3, p = 0.011, analysis of variance).

TABLE 3

Factors influencing degree of resection as determined on postoperative MR images

p Value
FactorUnivariate RegressionMultivariate Regression
eloquent area0.0110.029
date of op0.1630.076
age of patient0.3020.535
KPS score0.2030.47

Analysis of Survival

For analysis of survival, patient age, sex, KPS score, residual intraoperative fluorescence, tumor proximity to eloquent regions, residual enhancement on postoperative MR images, reoperation, and radioimmunotherapy were examined as potential prognostic factors. The mean overall survival for the entire group was 79.7 ± 7.6 weeks (Fig. 6A). Patient age (Fig. 6B), KPS score (Fig. 6C), and residual tumor enhancement on MR images (Fig. 6D) were significantly associated with survival, whereas tumor proximity to eloquent regions was not (Fig. 6E). No significant differences were noted when comparing patients in whom vague residual fluorescence was observed with patients in whom fluorescing tissue was removed completely, whereas the presence of residual solid fluorescence appeared to be a strong influence (Fig. 6F). In addition to age, KPS score, residual enhancement, and residual solid fluorescence, the volume of residual enhancement was significantly associated with survival time in univariate analysis.

Fig. 6.
Fig. 6.

Graph showing Kaplan—Meier survival curves. A: Overall survival (mean survival 79 ± 7 weeks). B: Survival stratified by age (< 57 years, 100 ± 12 weeks; ≥ 57 years, 58 ± 5 weeks). C: Survival stratified by KPS score (90–100, 98 ± 10 weeks; 60–80, 50 ± 3 weeks). D: Survival stratified by presence of residual enhancement on MR images (no enhancement, 103 ± 11 weeks; residual enhancement, 54 ± 5 weeks). E: Survival stratified by tumor proximity to eloquent brain regions (noneloquent, 86 ± 11 weeks; eloquent, 59 ± 3 weeks). F: Survival stratified by residual tissue fluorescence (no fluorescence, 101 ± 15 weeks; vague fluorescence, 79 ± 6 weeks; solid fluorescence, 51 ± 3 weeks).

For multivariate analysis, the colinearity between residual intraoperative fluorescence and residual tumor on MR images were taken into account (Table 2). Therefore, these variables were not included in the same model. Three models were fitted (Table 4). In the first model we considered the traditional prognostic factors: patient age, KPS score, and residual enhancement on MR images, the latter as a two-level variable (no enhancement compared with enhancement). In the second model, the presence or absence of residual enhancement was replaced by the volume of residual enhancement as a continuous variable. The third model—the treatment model—contained solid fluorescence as the treatment effect, adjusted by patient age and KPS score.

TABLE 4

Prognostic factors in 50 patients with GBMs in whom 5-ALA was used to guide extent of resection

Prognostic Factors IPrognostic Factors IITreatment§
Factor*Univariate p ValueMultivariate p ValueRelative RiskMultivariate p ValueRelative RiskMultivariate p ValueRelative Risk
age0.00660.02421.0520.02031.0530.04231.046
sex0.4708
KPS score0.00520.132.1040.02872.910.8521.111
 70–80
eloquent area0.3683
no flu0.1511
solid flu0.00030.01215.35
no resid postop0.00140.04470.4
 MR
MR volume0.03380.62751.026
reop0.8239
radioimmunotherapy0.6582

MR volume = volume of enhancing tissue on postoperative MR images; no flu = no residual intraoperative fluorescence; no resid postop MR = no residual enhancement on postoperative MR images; solid flu = residual solid fluorescence.

Multivariate analysis using age, KPS score, and absence of residual enhancement on postoperative MR images.

Multivariate analysis using age, KPS score, and volume of residual enhancing tissue on postoperative MR images.

Multivariate analysis using age, KPS score, and solid fluorescence.

Tested as a continuous variable.

Age was an independent prognostic variable in both models, whereas KPS score and volume of residual enhancement were not. Residual enhancement on MR images and residual solid fluorescence as nominal covariates were each independently significant in their respective models.

Adverse Events and Complications

No case of perioperative death was encountered. In three patients, preexisting symptoms were temporarily aggravated, but these had returned to preoperative levels or improved at hospital discharge, whereas only one patient suffered permanent right-sided hemiparesis as a consequence of intraoperative vessel damage. In one patient, unintentional exposure to daylight resulted in erythema of exposed skin, which resolved within 3 days. One patient developed leg pain on the 3rd postoperative day and underwent phlebography to rule out deep venous thrombosis; the pain resolved spontaneously 3 days later.

On a regular basis, there was a mild elevation of liver enzymes with no other signs of hepatic disorder. Glutamate-pyruvate transaminase, glutamate-oxaloacetic transaminase, and gamma-glutamyltransferase were found to increase in 85, 81, and 96% of patients, respectively, as determined on Day 7 after administration of 5-ALA. Average serum levels of glutamate-pyruvate transaminase increased from 22.7 ± 17.9 U/L preoperatively to 43.7 ± 43.3 U/L (p < 0.033), glutamate-oxaloacetic transaminase from 9.8 ± 4.7 U/L to 20.8 ± 13.7 U/L (p < 0.0001), and gamma-glutamyltransferase from 30.6 ± 33.9 U/L to 55.5 ± 60.2 U/L (p = 0.0182). In three cases, serum levels of glutamate-pyruvate transaminase and gamma-glutamyltransferase increased enough to qualify as Grade 3 toxicity on the common toxicity criteria score of the National Cancer Institute. In these patients repeated measurements on Day 14 after administration demonstrated resolution to Grade 1 or 2 (measurements were discontinued thereafter). No Grade 4 toxicity was encountered.

Histological Examination

A total of 264 biopsy specimens were obtained intraoperatively from fluorescing and nonfluorescing tumor margins. Of these, 211 biopsy samples displayed macroscopic fluorescence, and histological examination showed the presence of tumor. Normal brain was seen in 26 biopsy specimens without fluorescence. However, 26 other nonfluorescing biopsy specimens contained tumor. Of these 26 samples, 21 (81%) were found to contain diffusely infiltrating tumor cells and five (19%) contained solid tumor. One biopsy specimen showed macroscopic fluorescence but failed to show tumor.

Discussion

Cytoreductive surgery is accepted by many neurosurgeons as the best possible treatment for patients with malignant gliomas. Apart from providing surgical decompression and possibly prolonging survival, radical tumor removal appears desirable as a basis for adjuvant therapies (radiotherapy, immunotherapy, gene therapy). For these reasons, methods enabling better intraoperative discrimination of viable tumor borders may be valuable.

Attempts at contrast enhancement of malignant glioma tissue by intravenous administration of fluorescent markers are not new and have been described as early as 1948.21 Despite more recent attempts,13,22,25 these methods have not been widely used. No large series has been published demonstrating the efficacy of this approach for routine use or for prolonging survival. In earlier studies in brain tumors, apparent accumulation of intravenously administered dyes in malignant tissue relies on BBB disruption. Possible pitfalls may be leakage of dyes with accompanying edema into surrounding nontumorous tissue32 and contamination of the surgical cavity by blood containing the fluorescent agent, again jeopardizing selectivity.

Marking tumors with 5-ALA is conceptually different from the foregoing investigations. Itself nonfluorescing, 5-ALA is metabolized into strongly fluorescing protoporphyrin IX by a number of malignant tumors in situ through enzymes of the heme-biosynthesis pathway.11,16,19,27 This observation has been extended to C6 glioma cells, both cultured and in experimental brain tumors,12,34 and to human malignant gliomas.35 Fluorescence detection technology has since been adapted to a standard surgical microscope to enable its simple and routine use.33

Residual Intraoperative Fluorescence and Postoperative MR Imaging

It is generally accepted that surgery for malignant gliomas should aim at resecting tumor portions that enhance on imaging studies. Early postoperative MR imaging is regarded as the best modality for assessing residual tumor.1 In this series, early postoperative MR imaging demonstrated a lack of residual enhancement in more than 60% of patients, which was more than three times the rate cited in different series in which postoperative imaging was used.1,3,29,40

However, in 17 of 33 patients with no residual enhancement, fluorescing tissue was not removed completely intraoperatively. Because biopsy specimens of fluorescing tissue almost always contained tumor, the tissue fluorescence method appeared to be more sensitive in delineating residual tumor than enhancement on MR images, marking tumor in some regions in which there was no radiological enhancement. According to serial biopsy specimens of Grade IV astrocytomas,9,15 areas of contrast enhancement on CT scanning or MR imaging most often correspond to tumor tissue without intervening parenchyma, but can also be observed in regions with infiltrating tumor cells. Rarely, solid tumor fails to enhance.15 If tissue fluorescence with 5-ALA is more sensitive than enhancement on MR or CT images, fluorescence-guided resection would consequently be more sensitive than methods relying on MR or CT studies to guide resection, for example intraoperative imaging or image-based intraoperative navigation.

On the other hand, because residual fluorescence was highly predictive of residual enhancement on postoperative MR images, more radical resection of fluorescing tissue may have increased the number of patients without residual enhancement. In this study, however, the primary aim of resection in which 5-ALA was used was to remove as much fluorescing tumor as possible without endangering neurological functions, and the decision to resect fluorescing tissue was left to the individual surgeon. To test whether the surgeon's concern for neurological function was truly an important factor governing the extent of resection, multivariate regression analysis was performed. In this analysis, tumor proximity to eloquent regions proved the most significant contributor to the finding of postoperative enhancement on MR images. The date of the procedure within the study period was found to marginally affect residual enhancement, thus demonstrating a learning curve over time. Importantly, patient age and KPS score did not influence the completeness of tumor removal.

Intraoperative Biopsy Procedures

In this investigation, biopsy specimens were obtained from tumor margins and correlated with fluorescence findings to confirm sufficient specificity for a method that encourages more radical resection. Because only one in 212 fluorescing biopsy samples obtained in the transition zone between nonfluorescing and fluorescing tissue failed to show tumor, specificity indeed appeared to be sufficiently high. On the other hand, there were limitations regarding the method's sensitivity, that is, its ability to delineate all of a tumor. This was evident from that fact that 26 nonfluorescing tumor biopsy specimens contained infiltrating and in some cases even coalescent tumor. The reasons why a few samples failed to accumulate fluorescence are unclear. In an earlier report35 we gave an example of an initially low-grade glioma containing a secondary focus of anaplasia. This tumor was found to accumulate fluorescence only in its transformed portion, which were characterized histologically by vascular proliferation. Because 5-ALA is a polar substance, it may require a sufficiently disrupted BBB to enter the brain. Therefore, the failure of a number of samples with infiltrating tumor to accumulate protoporphyrin IX macroscopically may be the consequence of a sufficiently intact BBB. This is likely to be the case in infiltrating tumor located beyond the zone of edema. With regard to the nonfluorescing samples containing solid tumor, one might speculate whether these represent remnants of initially low-grade tumor or areas of GBM with a low proliferation rate.

Nevertheless, the sensitivity of this method demonstrated by the present findings was sufficient for improving the extent of resection. In this context it must be borne in mind that fluorescence-guided resection is based on an optical method. Sensitivity might theoretically be increased by more sophisticated technology to visualize remote tumor cell clusters. However, unlimited visualization may be dangerous, because resection of infiltrated but otherwise intact neural tissue may result in neurological deficit.15

Survival Analysis

The presence of residual tumor on MR images was a significant predictor of survival time in multivariate analysis. In accordance with most studies in which multivariate analysis is used, age was the other significant variable. On the other hand, the KPS score was highly significant in univariate analysis, but not independently prognostic in the multivariate test. In many studies that include survival analysis, the KPS score has been identified as an independent factor3,7,17,28,38 with exceptions.4,18 The present failure of the KPS score to predict survival in the multivariate test may be the result of limited numbers of patients suffering few deaths within the study period, but may also reflect preselection of patients with a KPS score greater than or equal to 60 as candidates for surgical therapy.

Interestingly, the volume of residual enhancement on postoperative MR images was of significant influence only in the univariate test. This aspect is of importance, because from it we infer that only complete resection of contrast-enhancing tumor will influence prognosis. This observation may explain why some investigations fail to demonstrate an influence of resection, as in the study by Kowalczuk, et al.17 In their study the category gross-total removal was defined to include volumes of residual enhancement greater than null.

Leaving solidly fluorescing tissue unresected was of predominant influence on survival, whereas the prognosis of patients with residual vague fluorescence in this small series was not significantly different from that of patients in whom fluorescing tissue was removed completely. Histologically, solidly fluorescing tissue was mostly characterized by coalescent tumor cells, whereas vaguely fluorescing tissue usually represented infiltrating tumor of intermediate or low cellular density. Our analysis of tumor biopsy specimens revealed infiltrating tumor also to be present beyond vaguely fluorescing tissue portions. Probably, removing vaguely fluorescing tissue does not diminish the quantity of infiltrating tumor cells completely enough to influence survival further. In contrast, the short-term prognosis appears to be affected by removing rapidly proliferating, coalescent tumor, which is represented by solidly fluorescing tissue.

Risk Assessment

Using a method that potentially leads to more radical resection immediately raises concerns about increasing surgical morbidity. This issue is important in patients in whom tumor recurrence is inevitable and treatment is essentially palliative. Even if modern surgical methods enable resection to proceed with minimal trauma, remote and unanticipated tissue damage cannot be ruled out. Moreover, blood vessels traversing the tumor may supply functionally relevant tissue. For these reasons, we often terminated resection in proximity to eloquent brain regions despite residual fluorescence indicating tumor tissue.

In the present series of 52 patients, there was no case of perioperative death and only one case (2%) of permanent neurological deficit (severe hemiparesis). In three patients (6%) preexisting symptoms were temporarily aggravated; otherwise, patients' outcomes were unchanged or improved on discharge. Thus, despite more radical resection, outcome in our series compares favorably with that in others in which neurological deterioration was observed in 14 to 20% of patients.7,18,36,39

We encountered no major side effects related to the administration of 5-ALA that qualified as severe adverse events in the present investigation. This observation was in accordance with those reported by others,37 who describe only minor side effects after oral ingestion of higher doses of 5-ALA (30–60 mg/kg body weight).

Conclusions

Taken together, fluorescence-guided resection in which 5-ALA is used appears to be a safe method and may improve the completeness of tumor removal in patients who harbor contrast-enhancing malignant glioma by highlighting residual solid and infiltrating tumor. The data demonstrate that survival strongly depends on residual solid intraoperative fluorescence and residual contrast-enhancing tumor visualized on early postoperative MR images.

The high rate of radical resection demonstrated to be possible with 5-ALA is noteworthy and may give a new impetus to research on the question of whether resection influences survival. To assess this question on a firmer statistical basis as well as to confirm the safety and efficacy of fluorescence-guided resection in which 5-ALA is used, a phase III, multicenter, prospective randomized study has now commenced at 13 German centers with an anticipated enrollment of 270 patients.

Acknowledgments

We gratefully acknowledge the ongoing support of Dr. Reinhold Baumgartner and Susanne Stocker, Ph.D., Laser Research Laboratory, Ludwig-Maximilians-University. We thank the Carl Zeiss (Oberkochen, Germany) and Karl Storz (Tuttlingen, Germany) companies for their generous support with donations, including microscope filter and illumination equipment, throughout the course of this investigation.

Disclosure

None of the authors has a financial interest in the promotion of fluorescence-guided resection in which 5-ALA is used or the devices needed for this purpose.

References

  • 1.

    Albert FKForsting MSartor Ket al: Early postoperative magnetic resonance imaging after resection of malignant glioma: objective evaluation of residual tumor and its influence on regrowth and prognosis. Neurosurgery 34:45611994Neurosurgery 34:

  • 2.

    Ammirati MVick NLiao Yet al: Effect of the extent of surgical resection on survival and quality of life in patients with supratentorial glioblastomas and anaplastic astrocytomas. Neurosurgery 21:2012061987Neurosurgery 21:

  • 3.

    Barker FG IIIPrados MDChang SMet al: Radiation response and survival time in patients with glioblastoma multiforme. J Neurosurg 84:4424481996J Neurosurg 84:

  • 4.

    Coffey RJLunsford LDTaylor FH: Survival after stereotactic biopsy of malignant gliomas. Neurosurgery 22:4654731988Neurosurgery 22:

  • 5.

    Collett D: Modelling Survival Data in Medical Research. London: Chapman & Hall19948081Collett D: Modelling Survival Data in Medical Research.

  • 6.

    Cox DR: Regression models and life tables. J R Stat Soc (B) 34:1872201972Cox DR: Regression models and life tables. J R Stat Soc (B) 34:

  • 7.

    Devaux BCO'Fallon JRKelly PJ: Resection, biopsy, and survival in malignant glial neoplasms. A retrospective study of clinical parameters, therapy and outcome. J Neurosurg 78:7677751993J Neurosurg 78:

  • 8.

    Duncan GGGoodman GBLudgate CMet al: The treatment of adult supratentorial high grade astrocytomas. J Neurooncol 13:63721992J Neurooncol 13:

  • 9.

    Earnest F IIKelly PJScheitauer BWet al: Cerebral astrocytomas: histopathologic correlation of MR and CT contrast enhancement with stereotactic biopsy. Radiology 166:8238271988Radiology 166:

  • 10.

    Franklin CI: Does the extent of surgery make a difference in high grade malignant astrocytoma. Australas Radiol 36:44471992Franklin CI: Does the extent of surgery make a difference in high grade malignant astrocytoma. Australas Radiol 36:

  • 11.

    Grant WEHopper CMacRobert AJet al: Photodynamic therapy of oral cancer: photosensitisation with systemic aminolaevulinic acid. Lancet 342:1471481993Lancet 342:

  • 12.

    Hebeda KMSaarnak AEOlivo Met al: 5-Aminolevulinic acid induced endogenous porphyrin fluorescence in 9L and C6 brain tumours and in the normal rat brain. Acta Neurochir 140:5035131998Acta Neurochir 140:

  • 13.

    Kabuto MKubota TKobayashi Het al: Experimental and clinical study of glioma at surgery using fluorescent imaging by a surgical microscope after fluorescein administration. Neurol Res 19:9161997Neurol Res 19:

  • 14.

    Kaplan ELMeier P: Non-parametric estimation for incomplete observations. J Am Stat Assoc 53:4574811958J Am Stat Assoc 53:

  • 15.

    Kelly PFDaumas-Duport CKispert DBet al: Imaging-based stereotaxic serial biopsies in untreated intracranial glial neoplasms. J Neurosurg 66:8658741987J Neurosurg 66:

  • 16.

    Kennedy JCPottier RH: Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J Photochem Photobiol B 14:2752921992J Photochem Photobiol B 14:

  • 17.

    Kowalczuk AMacdonald RLAmidei Cet al: Quantitative imaging study of extent of surgical resection and prognosis of malignant astrocytomas. Neurosurgery 41:102810381997Neurosurgery 41:

  • 18.

    Kreth FWWarnke PCScheremet Ret al: Surgical resection and radiation therapy versus biopsy and radiation therapy in the treatment of glioblastoma multiforme. J Neurosurg 78:7627661993J Neurosurg 78:

  • 19.

    Loh CSVernon DMacRobert AJet al: Endogenous porphyrin distribution induced by 5-aminolaevulinic acid in the tissue layers of the gastrointestinal tract. J Photochem Photobiol B 20:47541993J Photochem Photobiol B 20:

  • 20.

    Miller PJHassanein RSGiri PGSet al: Univariate and multivariate statistical analysis of high-grade gliomas: the relationship of radiation dose and other prognostic factors. Int J Radiat Oncol Biol Phys 19:2752801990Int J Radiat Oncol Biol Phys 19:

  • 21.

    Moore GEPeyton WTFrench LAet al: The clinical use of fluorescein in neurosurgery. The localization of brain tumors. J Neurosurg 5:3923981948J Neurosurg 5:

  • 22.

    Murray KJ: Improved surgical resection of human brain tumors. Part I. A preliminary study. Surg Neurol 17:3163191982Murray KJ: Improved surgical resection of human brain tumors. Part I. A preliminary study. Surg Neurol 17:

  • 23.

    Nazaaro JMNeuwelt EA: The role of surgery in the management of supratentorial intermediate and high-grade astrocytomas in adults. J Neurosurg 73:3313441990J Neurosurg 73:

  • 24.

    Nitta TSato K: Prognostic implications of the extent of surgical resection in patients with intracranial malignant gliomas. Cancer 75:272727311995Cancer 75:

  • 25.

    Poon WSSchomacker KTDeutsch TFet al: Laser-induced fluorescence: experimental intraoperative delineation of tumor resection margins. J Neurosurg 76:6796861992J Neurosurg 76:

  • 26.

    Quigley MRMaroon JC: The relationship between survival and the extent of the resection in patients with supratentorial malignant gliomas. Neurosurgery 29:3853891991Neurosurgery 29:

  • 27.

    Regula JMacRobert AJGorchein Aet al: Photosensitisation and photodynamic therapy of oesophageal, duodenal and colorectal tumours using 5 aminolaevulinic acid induced protoporphyrin IX—a pilot study. Gut 36:67751995Gut 36:

  • 28.

    Rostomily RCSpence AMDuong Det al: Multimodality management of recurrent adult malignant gliomas: results of a phase II multiagent chemotherapy study and analysis of cytoreductive surgery. Neurosurgery 35:3783881994Neurosurgery 35:

  • 29.

    Shrieve DCAlexander E IIIBlack PMet al: Treatment of patients with primary glioblastoma multiforme with standard postoperative radiotherapy and radiosurgical boost: prognostic factors and long-term outcome. J Neurosurg 90:72771999J Neurosurg 90:

  • 30.

    Simpson JRHorton JScott Cet al: Influence of location and extent of surgical resection on survival of patients with glioblastoma multiforme: results of three consecutive Radiation Therapy Oncology Group (RTOG) clinical trials. Int J Radiat Oncol Biol Phys 26:2392441993Int J Radiat Oncol Biol Phys 26:

  • 31.

    Sneed PKPrados MDMcDermott MWet al: Large effect of age on the survival of patients with glioblastoma treated with radiotherapy and brachytherapy boost. Neurosurgery 36:8989041995Neurosurgery 36:

  • 32.

    Stummer WGötz CHassan Aet al: Kinetics of photofrin II in perifocal brain edema. Neurosurgery 33:107510811993Neurosurgery 33:

  • 33.

    Stummer WStepp HMöller Get al: Technical principles for protoporphyrin-IX-fluorescence guided microsurgical resection of malignant glioma tissue. Acta Neurochir 140:99510001998Acta Neurochir 140:

  • 34.

    Stummer WStocker SNovotny Aet al: In vitro and in vivo porphyrin accumulation by C6 glioma cells after exposure to 5-aminolevulinic acid. J Photochem Photobiol B 45:1601691998J Photochem Photobiol B 45:

  • 35.

    Stummer WStocker SWagner Set al: Intraoperative detection of malignant gliomas by 5-aminolaevulinic acid-induced porphyrin fluorescence. Neurosurgery 42:5185261998Neurosurgery 42:

  • 36.

    Vecht CJAvezaat CJJvan Putten WLJet al: The influence of the extent of surgery on the neurological function and survival in malignant glioma. A retrospective analysis in 243 patients. J Neurol Neurosurg Psychiatry 53:4664711990J Neurol Neurosurg Psychiatry 53:

  • 37.

    Webber JKessel DFromm D: Side effects and photosensitization of human tissues after aminolevulinic acid. J Surg Res 68:31371997J Surg Res 68:

  • 38.

    Winger MJMacdonald DRCairncross JG: Supratentorial anaplastic gliomas in adults. The prognostic importance of extent of resection and prior low-grade glioma. J Neurosurg 71:4874931989J Neurosurg 71:

  • 39.

    Wisoff JHBoyett JMBerger MSet al: Current neurosurgical management and the impact of the extent of resection in the treatment of malignant gliomas of childhood: a report of the Children's Cancer Group Trial No. CCG-945. J Neurosurg 89:52591998J Neurosurg 89:

  • 40.

    Wood JRGreen SBShapiro WR: The prognostic importance of tumor size in malignant gliomas: a computed tomographic scan study by the Brain Tumor Cooperative Group. J Clin Oncol 6:3383431988J Clin Oncol 6:

  • 41.

    Yoshida JKajita YWakabayashi Tet al: Long-term follow-up results of 175 patients with malignant glioma: importance of radical tumour resection and postoperative adjuvant therapy with interferon, ACNU and radiation. Acta Neurochir 127:55591994Acta Neurochir 127:

Article Information

Address reprint requests to: Walter Stummer, M.D., Department of Neurosurgery, Klinikum Grosshadern, Ludwig-Maximilians-University, 81366 Munich, Germany. email: wstummer@nc.med.uni-muenchen.de.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Intraoperative photographs demonstrating tumor cavity viewed under conventional white light (left) and violet-blue illumination (right). Note different qualities of tumor fluorescence (necrosis, solid, vague).

  • View in gallery

    Graph showing intraoperative fluorescence spectra obtained from regions with solid and vague fluorescence. Peaks at 635 nm and 704 nm indicate protoporphyrin IX fluorescence.

  • View in gallery

    Photomicrographs showing morphological features of tissue with solid and vague fluorescence. Upper Left: Fluorescence videomicroscopy image of smear preparation obtained from biopsy sample featuring solid fluorescence. Image shows confluent red protoporphyrin fluorescence with negatively contrasted nuclei and nonfluorescing proliferated capillaries. Upper Right: Periodic acid Schiff staining of the smear in upper left. Lower Left: Fluorescence videomicroscopy image of smear preparation obtained from biopsy sample, featuring vague fluorescence associated with various large abnormal nuclei in invaded brain tissue. Note interspersed white fluorescence of lipid droplets. Lower Right: Giemsa staining of the preparation in lower left.

  • View in gallery

    Bar graphs showing dependence of fluorescence accumulation on morphological features. Left: Coalescent tumor compared with infiltrating tumor (p = 0.0136, chi-square test, 60 patients). Right: Association of fluorescence impression with tumor cellularity (p = 0.0112, chi-square test, 60 patients).

  • View in gallery

    Pie chart illustrating the relationship between the presence and quality of residual, fluorescing tissue and residual enhancement on early postoperative MR images (MR+ = residual enhancement on MR imaging; MR− = no enhancement on MR imaging; for statistics see Table 2).

  • View in gallery

    Graph showing Kaplan—Meier survival curves. A: Overall survival (mean survival 79 ± 7 weeks). B: Survival stratified by age (< 57 years, 100 ± 12 weeks; ≥ 57 years, 58 ± 5 weeks). C: Survival stratified by KPS score (90–100, 98 ± 10 weeks; 60–80, 50 ± 3 weeks). D: Survival stratified by presence of residual enhancement on MR images (no enhancement, 103 ± 11 weeks; residual enhancement, 54 ± 5 weeks). E: Survival stratified by tumor proximity to eloquent brain regions (noneloquent, 86 ± 11 weeks; eloquent, 59 ± 3 weeks). F: Survival stratified by residual tissue fluorescence (no fluorescence, 101 ± 15 weeks; vague fluorescence, 79 ± 6 weeks; solid fluorescence, 51 ± 3 weeks).

References

1.

Albert FKForsting MSartor Ket al: Early postoperative magnetic resonance imaging after resection of malignant glioma: objective evaluation of residual tumor and its influence on regrowth and prognosis. Neurosurgery 34:45611994Neurosurgery 34:

2.

Ammirati MVick NLiao Yet al: Effect of the extent of surgical resection on survival and quality of life in patients with supratentorial glioblastomas and anaplastic astrocytomas. Neurosurgery 21:2012061987Neurosurgery 21:

3.

Barker FG IIIPrados MDChang SMet al: Radiation response and survival time in patients with glioblastoma multiforme. J Neurosurg 84:4424481996J Neurosurg 84:

4.

Coffey RJLunsford LDTaylor FH: Survival after stereotactic biopsy of malignant gliomas. Neurosurgery 22:4654731988Neurosurgery 22:

5.

Collett D: Modelling Survival Data in Medical Research. London: Chapman & Hall19948081Collett D: Modelling Survival Data in Medical Research.

6.

Cox DR: Regression models and life tables. J R Stat Soc (B) 34:1872201972Cox DR: Regression models and life tables. J R Stat Soc (B) 34:

7.

Devaux BCO'Fallon JRKelly PJ: Resection, biopsy, and survival in malignant glial neoplasms. A retrospective study of clinical parameters, therapy and outcome. J Neurosurg 78:7677751993J Neurosurg 78:

8.

Duncan GGGoodman GBLudgate CMet al: The treatment of adult supratentorial high grade astrocytomas. J Neurooncol 13:63721992J Neurooncol 13:

9.

Earnest F IIKelly PJScheitauer BWet al: Cerebral astrocytomas: histopathologic correlation of MR and CT contrast enhancement with stereotactic biopsy. Radiology 166:8238271988Radiology 166:

10.

Franklin CI: Does the extent of surgery make a difference in high grade malignant astrocytoma. Australas Radiol 36:44471992Franklin CI: Does the extent of surgery make a difference in high grade malignant astrocytoma. Australas Radiol 36:

11.

Grant WEHopper CMacRobert AJet al: Photodynamic therapy of oral cancer: photosensitisation with systemic aminolaevulinic acid. Lancet 342:1471481993Lancet 342:

12.

Hebeda KMSaarnak AEOlivo Met al: 5-Aminolevulinic acid induced endogenous porphyrin fluorescence in 9L and C6 brain tumours and in the normal rat brain. Acta Neurochir 140:5035131998Acta Neurochir 140:

13.

Kabuto MKubota TKobayashi Het al: Experimental and clinical study of glioma at surgery using fluorescent imaging by a surgical microscope after fluorescein administration. Neurol Res 19:9161997Neurol Res 19:

14.

Kaplan ELMeier P: Non-parametric estimation for incomplete observations. J Am Stat Assoc 53:4574811958J Am Stat Assoc 53:

15.

Kelly PFDaumas-Duport CKispert DBet al: Imaging-based stereotaxic serial biopsies in untreated intracranial glial neoplasms. J Neurosurg 66:8658741987J Neurosurg 66:

16.

Kennedy JCPottier RH: Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J Photochem Photobiol B 14:2752921992J Photochem Photobiol B 14:

17.

Kowalczuk AMacdonald RLAmidei Cet al: Quantitative imaging study of extent of surgical resection and prognosis of malignant astrocytomas. Neurosurgery 41:102810381997Neurosurgery 41:

18.

Kreth FWWarnke PCScheremet Ret al: Surgical resection and radiation therapy versus biopsy and radiation therapy in the treatment of glioblastoma multiforme. J Neurosurg 78:7627661993J Neurosurg 78:

19.

Loh CSVernon DMacRobert AJet al: Endogenous porphyrin distribution induced by 5-aminolaevulinic acid in the tissue layers of the gastrointestinal tract. J Photochem Photobiol B 20:47541993J Photochem Photobiol B 20:

20.

Miller PJHassanein RSGiri PGSet al: Univariate and multivariate statistical analysis of high-grade gliomas: the relationship of radiation dose and other prognostic factors. Int J Radiat Oncol Biol Phys 19:2752801990Int J Radiat Oncol Biol Phys 19:

21.

Moore GEPeyton WTFrench LAet al: The clinical use of fluorescein in neurosurgery. The localization of brain tumors. J Neurosurg 5:3923981948J Neurosurg 5:

22.

Murray KJ: Improved surgical resection of human brain tumors. Part I. A preliminary study. Surg Neurol 17:3163191982Murray KJ: Improved surgical resection of human brain tumors. Part I. A preliminary study. Surg Neurol 17:

23.

Nazaaro JMNeuwelt EA: The role of surgery in the management of supratentorial intermediate and high-grade astrocytomas in adults. J Neurosurg 73:3313441990J Neurosurg 73:

24.

Nitta TSato K: Prognostic implications of the extent of surgical resection in patients with intracranial malignant gliomas. Cancer 75:272727311995Cancer 75:

25.

Poon WSSchomacker KTDeutsch TFet al: Laser-induced fluorescence: experimental intraoperative delineation of tumor resection margins. J Neurosurg 76:6796861992J Neurosurg 76:

26.

Quigley MRMaroon JC: The relationship between survival and the extent of the resection in patients with supratentorial malignant gliomas. Neurosurgery 29:3853891991Neurosurgery 29:

27.

Regula JMacRobert AJGorchein Aet al: Photosensitisation and photodynamic therapy of oesophageal, duodenal and colorectal tumours using 5 aminolaevulinic acid induced protoporphyrin IX—a pilot study. Gut 36:67751995Gut 36:

28.

Rostomily RCSpence AMDuong Det al: Multimodality management of recurrent adult malignant gliomas: results of a phase II multiagent chemotherapy study and analysis of cytoreductive surgery. Neurosurgery 35:3783881994Neurosurgery 35:

29.

Shrieve DCAlexander E IIIBlack PMet al: Treatment of patients with primary glioblastoma multiforme with standard postoperative radiotherapy and radiosurgical boost: prognostic factors and long-term outcome. J Neurosurg 90:72771999J Neurosurg 90:

30.

Simpson JRHorton JScott Cet al: Influence of location and extent of surgical resection on survival of patients with glioblastoma multiforme: results of three consecutive Radiation Therapy Oncology Group (RTOG) clinical trials. Int J Radiat Oncol Biol Phys 26:2392441993Int J Radiat Oncol Biol Phys 26:

31.

Sneed PKPrados MDMcDermott MWet al: Large effect of age on the survival of patients with glioblastoma treated with radiotherapy and brachytherapy boost. Neurosurgery 36:8989041995Neurosurgery 36:

32.

Stummer WGötz CHassan Aet al: Kinetics of photofrin II in perifocal brain edema. Neurosurgery 33:107510811993Neurosurgery 33:

33.

Stummer WStepp HMöller Get al: Technical principles for protoporphyrin-IX-fluorescence guided microsurgical resection of malignant glioma tissue. Acta Neurochir 140:99510001998Acta Neurochir 140:

34.

Stummer WStocker SNovotny Aet al: In vitro and in vivo porphyrin accumulation by C6 glioma cells after exposure to 5-aminolevulinic acid. J Photochem Photobiol B 45:1601691998J Photochem Photobiol B 45:

35.

Stummer WStocker SWagner Set al: Intraoperative detection of malignant gliomas by 5-aminolaevulinic acid-induced porphyrin fluorescence. Neurosurgery 42:5185261998Neurosurgery 42:

36.

Vecht CJAvezaat CJJvan Putten WLJet al: The influence of the extent of surgery on the neurological function and survival in malignant glioma. A retrospective analysis in 243 patients. J Neurol Neurosurg Psychiatry 53:4664711990J Neurol Neurosurg Psychiatry 53:

37.

Webber JKessel DFromm D: Side effects and photosensitization of human tissues after aminolevulinic acid. J Surg Res 68:31371997J Surg Res 68:

38.

Winger MJMacdonald DRCairncross JG: Supratentorial anaplastic gliomas in adults. The prognostic importance of extent of resection and prior low-grade glioma. J Neurosurg 71:4874931989J Neurosurg 71:

39.

Wisoff JHBoyett JMBerger MSet al: Current neurosurgical management and the impact of the extent of resection in the treatment of malignant gliomas of childhood: a report of the Children's Cancer Group Trial No. CCG-945. J Neurosurg 89:52591998J Neurosurg 89:

40.

Wood JRGreen SBShapiro WR: The prognostic importance of tumor size in malignant gliomas: a computed tomographic scan study by the Brain Tumor Cooperative Group. J Clin Oncol 6:3383431988J Clin Oncol 6:

41.

Yoshida JKajita YWakabayashi Tet al: Long-term follow-up results of 175 patients with malignant glioma: importance of radical tumour resection and postoperative adjuvant therapy with interferon, ACNU and radiation. Acta Neurochir 127:55591994Acta Neurochir 127:

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 7 7 7
Full Text Views 219 219 83
PDF Downloads 88 88 54
EPUB Downloads 0 0 0

PubMed

Google Scholar