106
Surg Neurol 1991;36:106-11
Editorial A Perspective on the Care of Gliomas from the United Kingdom Since the time when Sir Hugh Cairns spent a year with Harvey Cushing and returned to London to do neurosurgery, the surgery of gliomas has been similar in the United Kingdom and in North America. There are many changes occurring in neurosurgery today, but the care of supratentorial gliomas in the elderly has undergone relatively little change. This is not because of the lack of intense investigation into many fields with many modes of treatment. In addition, our colleagues in the United Kingdom are under considerable duress in terms o f costs and equipment for patient care. T h e y are also subject to m o r e frequent audits of their work and have far fewer imaging centers than those o f us in N o r t h America. Cost effectiveness, which has been such a byword in the United Kingdom for over a generation, is becoming quite a c o m m o n p r o b l e m in N o r t h America. T h e contribution by Dr. Whittle and his colleagues is unique in that it gives a very detailed, frank assessment
of the patients that have been cared for at an outstanding neurosurgical center. The fact that the o u t c o m e of most of these patients was generally appalling has not been glossed over. As Dr. Whittle points out, the results o f intensively treated subgroups in the past have given the ultimate results in malignant gliomas o f the brain in adults; although these patients have been cared for by several outstanding neurosurgeons, there has not been a standard of controlled care under protocol. It is, therefore, "the run of the mill" and forms a background for future audits. The editor is pleased that these authors have seen fit to submit this manuscript for review and that Surgical Neurology can publish it for the edification o f those in N o r t h America as well as in the United Kingdom. E B E N A L E X A N D E R , JR., M.D.
Management of Patients Aged Over 60 Years with Supratentorial Glioma: Lessons from an Audit Ian R. Whittle, M.D., Ph.D., F.R.A.C.S., F.R.C.S.E. (S.N.), Stuart W. Denholm, F.R.C.S.E., and Anna Gregor, M.R.C.P., F.R.C.R. Neuro-oncology Clinic, Departments of Clinical Neuroscience and Clinical Oncology, Western General Hospital, Edinburgh, Scotland
Whittle IR; Denholm SW; Gregor A. Management of patients aged over 60 years with supratentorial glioma: lessons from an audit. Surg Neurol 1991;36;106-11. This audit of clinical management confirmed the poor prognosis for patients (n = 80) aged over 60 years with a diagnosis of supratentorial glioma. The median survival time after diagnosis was 9 weeks following steroids and 7 weeks after biopsy and steroids. Cytoreductive surgery Address reprint requests to," Ian R. Whittle, M.D., Ph.D., F.R.A.C.S., F.R.C.S.E. (S.N.), Department of Clinical Neuroscience, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, Scotland. Received November 19, 1990; accepted January 13, 1991. © 1991 by ElsevierSciencePublishingCo., Inc.
and radiotherapy improved median survival time by a maximum of 16 weeks, but there was significant morbidity in some patients undergoing craniotomy. Although 92% of the biopsied lesions were either glioblastoma or anaplastic astrocytoma, the median survival of these patients was similar to the 8% of patients confirmed as having intermediate grade astrocytoma. Patients presenting with minimal functional deficit (WHO grade I or II) had longer median survival times than those presenting in poor condition (WHO grade I I I or IV). In this series there was no relationship between management undertaken and clinical status of the patient. The 9% of the cohort that survived 1 year were treated in a variety of ways. This audit, 0090-3019/91/$3.50
Surg Neurol 1991;36:107-11
together with results from other studies, suggests that a prospective clinical trial of different management regimes in elderly patients with supratentorial gliomas is needed. Since median survival time in the most intensively treated patients will be around 6 months, treatment evaluation must consider the quality of life provided. KEY WORDS: Audit; Glioma; Stereotactic biopsy; Radiotherapy; Chemotherapy
107
were right lobar (33 patients), left lobar (31 patients), and callosal-diencephalic (six patients). Histological diagnoses were anaplastic astrocytoma or glioblastoma (63 patients) and intermediate-grade astrocytoma (five patients). Twelve patients had a diagnosis of malignant glioma made on CT scan alone. The median number of patients managed each year was 12, but this ranged from 8 (1986) to 17 (1987). Methods
The current management of patients with malignant gliomas ranges from therapeutic nihilism to aggressive multidisciplinary approaches [12,13,19]. The principal reason for differences in management stems from the poor prognosis associated with the disease [ 12,19] and uncertainties about the best management option [ 1,3,4,8,19]. The most important adverse prognostic factor in patients with malignant glioma is advanced age, with older patients surviving for significantly shorter periods than those aged below 45 years [4,12,13,19]. Since 8 5 % - 9 2 % of gliomas in the elderly are malignant [11,16], patients aged over 60 years with a diagnosis of glioma based on computed tomography (CT) scan present a difficult management problem [11,16]. Since we try to optimize management for brain tumor patients in a combined neurooncology clinic, we have reviewed the practice in the Department of Surgical Neurology and Radiation Oncology, during the period of 1983-1989, to determine (a) the impact of treatment on outcome, (b) factors influencing decision making, and (c) if there had been a consistent management policy for these patients. The increasing number of elderly people in the United Kingdom and a greater emphasis on costeffective management [9] make this type of analysis particularly relevant.
Patient Cohort The clinical course of a consecutive series of 80 patients, aged over 60 years with a CT diagnosis of supratentorial glioma and managed by one of six consultant neurosurgeons at the Department of Surgical Neurology between 1983 and 1989, was reviewed. Information on these patients was retrieved from the computerized departmental records, with additional histological and surgical data being abstracted from case note and neuropathological review. There were 53 men and 27 women. The median age was 63 years (range 60-74 years). Confusion (62%), headache (59%), and seizures (14%) were the most common presenting symptoms, while sensorimotor deficits (84%), dysphasia (44%), and papilledema (25%) were the most common clinical findings. Tumor locations
Patients' admission clinical performance grades were classified using the five-point WHO scale [ 18]. Management regimes were defined as permutations of steroid therapy, biopsy, cytoreductive surgery, chemotherapy, and postoperative radiotherapy. Outcome was determined by length of hospitalization and case fatality and morbidity at 30 days. Surgical morbidity was defined as postoperative intracranial hematoma, iatrogenic neurological deficit, pulmonary embolus, myocardial infarction, and sepsis at the surgical site. For survival analysis, day 0 was defined as the first day of the admission that led either to CT or tissue diagnosis of glioma. The influence of various management regimes and initial performance status on median survival time and inpatient stay were analyzed using the Mann-Whitney-Wilcoxon test. Confidence intervals were calculated using standard methods [6].
Results The five management regimes used by the clinicians in this study were (1) corticosteroids alone, (2) steroids followed by biopsy, (3) steroids followed by craniotomy and tumor debulking (cytoreductive surgery), (4) steroids followed by biopsy with postoperative radiotherapy, and (5) steroids followed by cytoreductive surgery and postoperative radiotherapy. Four patients in the latter category also had TCNU chemotherapy; in two patients TCNU was given during surgery, and in the other two following relapse. Steroid dosage was initially 16 mg/day of dexamethasone and decreased after 1 week. Biopsies were performed either freehand (through a burr hole or craniotomy) (n = 11) or using the Brown-Roberts-Wells CT image-guided stereotactic system (n -- 13) [ 14]. Craniotomy technique and resection ranged from traditional resection incorporating lobectomy to CT image stereotactic microsurgical craniotomy [10,17]. Patients with tumors that had central, callosal, or basal ganglionic extension generally had either a biopsy or no surgical intervention. Radiotherapy consisted of 47.5 Gy, in 20 fractions, of whole brain irradiation before 1987, and thereafter 60 Gy in 30 fractions to tumor and peritumoral brain.
108
Surg Neurol 1991;36:107-11
Whittle et al
Table 1. Impact of Various Management Strategies on the 30-Day Surgical Morbidity and Case Fatality, Median Duration of Neurosurgical Unit Hospitalization, and Median Survival Time in Patients Aged over 60 Years with a Supratentorial Glioma 30 Day
Management
No. of patients
Surgical morbidity (95% CI)
Steroids only Biopsy Cytoreduction Biopsy + RT Cytoreduction + RT
12 24 13 5 26
N/A 8% (0-19%) 61% (34%-83%.) 0 0
Case fatality (95% CI)
Median hospital stay (days)
Median survival time (weeks)
25% (1%-50%) 33% (14%-52%) 0 0 0
5 9 21 15 19
9 7 15 16 23
Abbreviation:CI, confidenceinterval;N/A, not available;RT, radiotherapy.
The effects o f the five management regimens on median hospital stay (excluding inpatient admission for radiotherapy), 30-day case fatality, surgical morbidity, and median survival time (weeks) are listed in Table 1. Length o f inpatient hospitalization was significantly (p 0.01) prolonged following craniotomy when compared with either biopsy or steroids alone. Patients having cytoreductive surgery followed by postoperative radiotherapy had a significantly (p < 0.001) longer survival than those having either biopsy or steroids alone. Otherwise, there were no significant survival advantages among different treatment regimens. The survival curves for the various treatment modalities, except the six patients having biopsy and radiotherapy, are shown in Figure 1. T h e r e was no consistent relationship between the preoperative performance grade o f the patient and the selected management regimen (Table 2). Patients with better clinical grades ( W H O grade I or II) at the time o f presentation had, regardless o f treatment, significantly
longer survival times (p < 0.05) than those with W H O grades III or IV (Table 3). The median survival time for patients with either biopsy-proven intermediate-grade astrocytoma or malignant glioma was 16 weeks, while those patients diagnosed from CT scan alone as having a malignant supratentorial glioma had a median survival time of 9 weeks. Seven patients (9% o f the total) survived 12 months or longer, and the clinical details of these patients are listed in Table 4. There was no morbidity associated with stereotactic biopsy; however, after cytoreductive surgery, eight of 39 (21%) patients had either a postoperative hematoma or iatrogenic neurological deficit. There was no postoperative wound sepsis.
Discussion Therapeutic decision making in elderly patients with a CT diagnosis of malignant supratentorial glioma is diffi-
100 I
80
0
} "
Steroids only Biopsy only Craniotomy + Debulk Debulk + RT
60
Figure 1. Survival curvesforfour of thefive treat"~
ment strategies used in the management of patients aged over 60 years with a diagnosis of supratentorial glioma. R_TT,radiotherapy.
40
20
•
0
,
2
•
,
4
•
,
6
.
,".
8
,--.-.-..~ 10
Months
12
14
~
16
•
,
18
.
,
20
Management of Patients
Surg Neurol 1991;36:107-11
109
Table 2. Clinical Status (WHO Grade) and Management Regimen Selectedfor Patients Aged Over 60 years with a Supratentorial Glioma WHO Grade Management
No. of patients
1
2
3
4
Steroids only Biopsy Cytoreduction Biopsy + RT Cytoreduction + RT
12 24 13 5 26
-1 1 -2
5 2 6 2 8
4 14 6 3 12
3 7 --4
Abbreviation: RT, radiotherapy. Steroids were prescribed in all cases, and craniotomy was performed for all cytoreductive procedures.
cult. These patients tolerate impairment of brain function poorly, and usually die several months after the diagnosis [3,4,12,19]. In patients aged under 45 years with a malignant glioma, the most successful treatment, which produces a median survival time of 24 months, has been with cytoreductive surgery followed by irradiation (60 Gy) and single-agent nitrosourea therapy [12,13]. The median survival time reported after similar treatment in patients aged over 65 years is only 27 weeks [13]. This nonrandomized observational study, which showed a similar median survival time in aggressively treated patients, suggests that such interventional management has only a small, temporal influence on disease progression and is associated with significant surgical morbidity. If it is assumed that all patients in the group subjected to craniotomy and tumor debulking were initially scheduled to go on to radiotherapy, the postoperative status of 30% of these patients would be bad enough to preclude cerebral irradiation. This figure is almost identical to the incidence of postoperative complications described in a series of 80 patients aged over 65 years who had a spectrum of brain tumors [ 16]. Paradoxically, although morbidity may be high, 30-day case fatality after craniotomy in this and other series is below 5% [11,16]. The failure of surgery to ameliorate clinical status in many patients
Table 3. Survival Times of Patients Aged Over 60 Years with Supratentorial Malignant Glioma in Relation to Their Presenting Clinical Status Survival Time (weeks) WHO Grade
No. of patients
Median
Range
1 2 3 4
4 23 39 14
32 20 14 8
24-36 7-84 1-90 1-49
will not only reflect postoperative complications but also ill-advised patient selection. Highly sophisticated, technically innovative surgery, which is associated with a low morbidity even in the elderly, does not alone contribute substantially to median survival times, although it may lead to transitory improvements in quality of life and less problems during radiotherapy [3,7,10,16]. The use of high-dose radiotherapy in some elderly patients may be inappropriate since, due to limited cerebral reserve [5], transient confusion and disorientation are common following such therapy (Whittle, Gregor, unpublished data). Lower doses (30 Gy, administered in 10 fractions) seem to be as effective and less toxic [8]. Cerebral irradiation, whether following biopsy [3] or cytoreductive surgery, seems integral to producing a definitive survival advantage [1,4,8,12], but the effectiveness of this therapy decreases with advancing age [13, 19]. If aggressive multimodality treatment is not planned, should even a biopsy be performed? A C T scan diagnosis is not infallible, and the diagnostic error rate for this condition will be at least 5 % - 10% [2,14,15 ]. This study supports the contention that CT image-guided stereotactic biopsy is a safe and effective method of establishing a histological diagnosis in the elderly [ 14]. Lesion biopsy, particularly if performed without CT-imaged targeting, is subject to the problem of tumor heterogeneity, and this may explain why in this series the median survival time for patients with histologically diagnosed intermediate-grade glioma was the same as a malignant glioma. Nonetheless, the incidence (92%) of malignant tumors among the biopsy-confirmed gliomas is remarkably similar to other series [11,16], and it may be that in the elderly, even intermediate-grade astrocytomas progress more rapidly than similar tumors in younger patients. The lack of either a consensus about optimal management or a correlation between treatment and patients' clinical status among a group of consultant neurosurgeons highlights the difficulties in deciding treatment for these patients. Although this may reflect differences in management philosophy, it is apparent that definite goals
110
Surg Neurol 1991;36:107-11
Whittle et al
T a b l e 4. Clinicopathological Details of the Seven Patients Who Survived 12 Months or Longer Survival time (months)
Sex/age (yr)
WHO Grade
Pathology
Management regimen
12 12 14 16 20 21 27
F/66 M/61 M/62 F/60 M/60 F/63 M/63
2 2 3 3 2 3 2
Glioblastoma Glioblastoma Glioblastoma Unknown AA Glioblastoma AA
Biopsy Cytoreduction + RT Cytoreduction + RT Steroids Biopsy + RT Cytoreduction + RT Biopsy
Abbreviations: AA, anaplastic astrocytoma; F, female; M, male; RT, radiotherapy. WHO grade refers to presenting neurological status.
and guidelines are required to rationalize management. A parallel study within our Regional Health Authority [15] and a study from another British unit [19] both suggested that clinicians have an unpredictable attitude toward referral of elderly patients--whose CT scan shows a brain tumor--to neurosurgeons. It is, therefore, likely that the patients in this series represent about 50% of patients with this CT diagnosis. This study confirms that clinical status at the time of presentation is an important prognostic parameter [1,4,12,13]. This parameter should probably be used when deciding clinical management. Nonetheless, some patients who present in poor clinical condition (WHO grade III) will survive between 1 and 2 years. Given this variability in natural outcome and the uncertainty as to which treatment strategy represents optimal management, it would seem appropriate that a clinical trial to address this question should be undertaken in elderly patients with glioma. A fundamental question that such a trial would also have to address is the quality of life provided by various treatments [1]. With the relative ineffectiveness of current therapy, the provision of a good quality of survival, albeit limited, may provide some justification for intervention in selected patients. At present, the cost benefit of neurosurgery for brain tumors is two orders of magnitude worse than surgery for aneurysms [9]; thus, resource and economic considerations also provide further impetus for a more objective approach to management.
The authors would like to thank Professor J. Douglas Miller and Dr. Peter Sandercock for reviewing the manuscript.
References 1. BradaM. Backto the future--radiotherapyin highgradegliomas. Br J Cancer 1989;60:1-4. 2. Choksey MS, Valentine A, Shawdon H. Computerised tomography in the diagnosis of malignant brain tumors. Do all require biopsy? J Neurol Neurosurg Psychiatry 1989;52:821-5. 3. Coffey RJ, Dunsford LD, Taylor FH. Survival after stereotaxic biopsy of malignant gliomas. Neurosurgery 1988;22:465-73. 4. Cohadon F. Indications for surgery in the management ofgliomas. Adv Tech Stand Neurosurg 1990;17:189-234. 5. Galbraith S. Head injuries in the elderly. Br MedJ 1987;294:325. 6. Gardener MJ, Altman DG. Statistics with confidence. Br Med 1989;137. 7. Kelly PJ, Kall BA, Goerss S, et al. Computer assisted sterotaxic laser resection of intraaxial brain neoplasms. J Neurosurg 1986;69:423-39. 8. Newall J, RansohoffJ, Kaplan B. Glioblastoma in the older patient: how long a course of radiotherapy is necessary? J Neurooncol 1988;6:325-7. 9. Pickard JD, Bailey S, Sanderson H, et ai. Steps towards cost benefit analysis of regional neurosurgical care. Br Med J 1990;301:629-35. 10. Sabin HI, Whittle IR. Stereotaxic craniotomy and intraoperative lesion locaiisation; technical note. Acta Neurochir (Wien) 1990;102:137-40. 11. Salcman M. Brain tumors and the geriatric patient. J Am Geriatr Soc 1982;30:501-8. 12. Shapiro WR. Therapy of adult malignant brain tumors. What have the clinical trials taught us. Semin Oncol 1986;13:38-45. 13. Shapiro WR, Green SB, Burger PC, et al. Randomized trial of three chemotherapy regimens and two radiotherapy regimens in postoperative treatment of malignant glioma. J Neurosurg 1989;71:1-9. 14. Thomas DGT, Nouby RM. Experience in 300 cases of CT-directed stereotaxic surgery for lesion biopsy and aspiration of hematoma. J Neurosurg 1989;3:337-42. 15. Todd NV, McDonagh T, Miller JD. What follows diagnosis by computer tomography of solitary brain tumor? Lancet 1987;1:611-2.
Management of Patients
Surg N e u r o l 1991;36:107-11
111
16. Tomita T, Raimondi AJ. Brain tumors in the elderly. JAMA 1981;246:53-5.
18. WHO handbook for reporting results of cancer treatment. Geneva, Switzerland: WHO, 1979:45 pp.
17. Waters CL. Surgical management of intracranial gliomas. In: Schmidek HH, Sweet WH, eds. Operative neurosurgical technique, ed 2. New York: Grune and Stratton, 1982:44780.
19. Wroe SJ, Foy PM, Shaw MDM, et al. Neurological and neurosurgical approachs in management of malignant brain tumors. Br Med J 1986;293:1015-8.
Surg Neurol 1991;36:106-11
Editorial A Perspective on the Care of Gliomas from the United Kingdom Since the time when Sir Hugh Cairns spent a year with Harvey Cushing and returned to London to do neurosurgery, the surgery of gliomas has been similar in the United Kingdom and in North America. There are many changes occurring in neurosurgery today, but the care of supratentorial gliomas in the elderly has undergone relatively little change. This is not because of the lack of intense investigation into many fields with many modes of treatment. In addition, our colleagues in the United Kingdom are under considerable duress in terms o f costs and equipment for patient care. T h e y are also subject to m o r e frequent audits of their work and have far fewer imaging centers than those o f us in N o r t h America. Cost effectiveness, which has been such a byword in the United Kingdom for over a generation, is becoming quite a c o m m o n p r o b l e m in N o r t h America. T h e contribution by Dr. Whittle and his colleagues is unique in that it gives a very detailed, frank assessment
of the patients that have been cared for at an outstanding neurosurgical center. The fact that the o u t c o m e of most of these patients was generally appalling has not been glossed over. As Dr. Whittle points out, the results o f intensively treated subgroups in the past have given the ultimate results in malignant gliomas o f the brain in adults; although these patients have been cared for by several outstanding neurosurgeons, there has not been a standard of controlled care under protocol. It is, therefore, "the run of the mill" and forms a background for future audits. The editor is pleased that these authors have seen fit to submit this manuscript for review and that Surgical Neurology can publish it for the edification o f those in N o r t h America as well as in the United Kingdom. E B E N A L E X A N D E R , JR., M.D.
Management of Patients Aged Over 60 Years with Supratentorial Glioma: Lessons from an Audit Ian R. Whittle, M.D., Ph.D., F.R.A.C.S., F.R.C.S.E. (S.N.), Stuart W. Denholm, F.R.C.S.E., and Anna Gregor, M.R.C.P., F.R.C.R. Neuro-oncology Clinic, Departments of Clinical Neuroscience and Clinical Oncology, Western General Hospital, Edinburgh, Scotland
Whittle IR; Denholm SW; Gregor A. Management of patients aged over 60 years with supratentorial glioma: lessons from an audit. Surg Neurol 1991;36;106-11. This audit of clinical management confirmed the poor prognosis for patients (n = 80) aged over 60 years with a diagnosis of supratentorial glioma. The median survival time after diagnosis was 9 weeks following steroids and 7 weeks after biopsy and steroids. Cytoreductive surgery Address reprint requests to," Ian R. Whittle, M.D., Ph.D., F.R.A.C.S., F.R.C.S.E. (S.N.), Department of Clinical Neuroscience, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, Scotland. Received November 19, 1990; accepted January 13, 1991. © 1991 by ElsevierSciencePublishingCo., Inc.
and radiotherapy improved median survival time by a maximum of 16 weeks, but there was significant morbidity in some patients undergoing craniotomy. Although 92% of the biopsied lesions were either glioblastoma or anaplastic astrocytoma, the median survival of these patients was similar to the 8% of patients confirmed as having intermediate grade astrocytoma. Patients presenting with minimal functional deficit (WHO grade I or II) had longer median survival times than those presenting in poor condition (WHO grade I I I or IV). In this series there was no relationship between management undertaken and clinical status of the patient. The 9% of the cohort that survived 1 year were treated in a variety of ways. This audit, 0090-3019/91/$3.50
Surg Neurol 1991;36:107-11
together with results from other studies, suggests that a prospective clinical trial of different management regimes in elderly patients with supratentorial gliomas is needed. Since median survival time in the most intensively treated patients will be around 6 months, treatment evaluation must consider the quality of life provided. KEY WORDS: Audit; Glioma; Stereotactic biopsy; Radiotherapy; Chemotherapy
107
were right lobar (33 patients), left lobar (31 patients), and callosal-diencephalic (six patients). Histological diagnoses were anaplastic astrocytoma or glioblastoma (63 patients) and intermediate-grade astrocytoma (five patients). Twelve patients had a diagnosis of malignant glioma made on CT scan alone. The median number of patients managed each year was 12, but this ranged from 8 (1986) to 17 (1987). Methods
The current management of patients with malignant gliomas ranges from therapeutic nihilism to aggressive multidisciplinary approaches [12,13,19]. The principal reason for differences in management stems from the poor prognosis associated with the disease [ 12,19] and uncertainties about the best management option [ 1,3,4,8,19]. The most important adverse prognostic factor in patients with malignant glioma is advanced age, with older patients surviving for significantly shorter periods than those aged below 45 years [4,12,13,19]. Since 8 5 % - 9 2 % of gliomas in the elderly are malignant [11,16], patients aged over 60 years with a diagnosis of glioma based on computed tomography (CT) scan present a difficult management problem [11,16]. Since we try to optimize management for brain tumor patients in a combined neurooncology clinic, we have reviewed the practice in the Department of Surgical Neurology and Radiation Oncology, during the period of 1983-1989, to determine (a) the impact of treatment on outcome, (b) factors influencing decision making, and (c) if there had been a consistent management policy for these patients. The increasing number of elderly people in the United Kingdom and a greater emphasis on costeffective management [9] make this type of analysis particularly relevant.
Patient Cohort The clinical course of a consecutive series of 80 patients, aged over 60 years with a CT diagnosis of supratentorial glioma and managed by one of six consultant neurosurgeons at the Department of Surgical Neurology between 1983 and 1989, was reviewed. Information on these patients was retrieved from the computerized departmental records, with additional histological and surgical data being abstracted from case note and neuropathological review. There were 53 men and 27 women. The median age was 63 years (range 60-74 years). Confusion (62%), headache (59%), and seizures (14%) were the most common presenting symptoms, while sensorimotor deficits (84%), dysphasia (44%), and papilledema (25%) were the most common clinical findings. Tumor locations
Patients' admission clinical performance grades were classified using the five-point WHO scale [ 18]. Management regimes were defined as permutations of steroid therapy, biopsy, cytoreductive surgery, chemotherapy, and postoperative radiotherapy. Outcome was determined by length of hospitalization and case fatality and morbidity at 30 days. Surgical morbidity was defined as postoperative intracranial hematoma, iatrogenic neurological deficit, pulmonary embolus, myocardial infarction, and sepsis at the surgical site. For survival analysis, day 0 was defined as the first day of the admission that led either to CT or tissue diagnosis of glioma. The influence of various management regimes and initial performance status on median survival time and inpatient stay were analyzed using the Mann-Whitney-Wilcoxon test. Confidence intervals were calculated using standard methods [6].
Results The five management regimes used by the clinicians in this study were (1) corticosteroids alone, (2) steroids followed by biopsy, (3) steroids followed by craniotomy and tumor debulking (cytoreductive surgery), (4) steroids followed by biopsy with postoperative radiotherapy, and (5) steroids followed by cytoreductive surgery and postoperative radiotherapy. Four patients in the latter category also had TCNU chemotherapy; in two patients TCNU was given during surgery, and in the other two following relapse. Steroid dosage was initially 16 mg/day of dexamethasone and decreased after 1 week. Biopsies were performed either freehand (through a burr hole or craniotomy) (n = 11) or using the Brown-Roberts-Wells CT image-guided stereotactic system (n -- 13) [ 14]. Craniotomy technique and resection ranged from traditional resection incorporating lobectomy to CT image stereotactic microsurgical craniotomy [10,17]. Patients with tumors that had central, callosal, or basal ganglionic extension generally had either a biopsy or no surgical intervention. Radiotherapy consisted of 47.5 Gy, in 20 fractions, of whole brain irradiation before 1987, and thereafter 60 Gy in 30 fractions to tumor and peritumoral brain.
108
Surg Neurol 1991;36:107-11
Whittle et al
Table 1. Impact of Various Management Strategies on the 30-Day Surgical Morbidity and Case Fatality, Median Duration of Neurosurgical Unit Hospitalization, and Median Survival Time in Patients Aged over 60 Years with a Supratentorial Glioma 30 Day
Management
No. of patients
Surgical morbidity (95% CI)
Steroids only Biopsy Cytoreduction Biopsy + RT Cytoreduction + RT
12 24 13 5 26
N/A 8% (0-19%) 61% (34%-83%.) 0 0
Case fatality (95% CI)
Median hospital stay (days)
Median survival time (weeks)
25% (1%-50%) 33% (14%-52%) 0 0 0
5 9 21 15 19
9 7 15 16 23
Abbreviation:CI, confidenceinterval;N/A, not available;RT, radiotherapy.
The effects o f the five management regimens on median hospital stay (excluding inpatient admission for radiotherapy), 30-day case fatality, surgical morbidity, and median survival time (weeks) are listed in Table 1. Length o f inpatient hospitalization was significantly (p 0.01) prolonged following craniotomy when compared with either biopsy or steroids alone. Patients having cytoreductive surgery followed by postoperative radiotherapy had a significantly (p < 0.001) longer survival than those having either biopsy or steroids alone. Otherwise, there were no significant survival advantages among different treatment regimens. The survival curves for the various treatment modalities, except the six patients having biopsy and radiotherapy, are shown in Figure 1. T h e r e was no consistent relationship between the preoperative performance grade o f the patient and the selected management regimen (Table 2). Patients with better clinical grades ( W H O grade I or II) at the time o f presentation had, regardless o f treatment, significantly
longer survival times (p < 0.05) than those with W H O grades III or IV (Table 3). The median survival time for patients with either biopsy-proven intermediate-grade astrocytoma or malignant glioma was 16 weeks, while those patients diagnosed from CT scan alone as having a malignant supratentorial glioma had a median survival time of 9 weeks. Seven patients (9% o f the total) survived 12 months or longer, and the clinical details of these patients are listed in Table 4. There was no morbidity associated with stereotactic biopsy; however, after cytoreductive surgery, eight of 39 (21%) patients had either a postoperative hematoma or iatrogenic neurological deficit. There was no postoperative wound sepsis.
Discussion Therapeutic decision making in elderly patients with a CT diagnosis of malignant supratentorial glioma is diffi-
100 I
80
0
} "
Steroids only Biopsy only Craniotomy + Debulk Debulk + RT
60
Figure 1. Survival curvesforfour of thefive treat"~
ment strategies used in the management of patients aged over 60 years with a diagnosis of supratentorial glioma. R_TT,radiotherapy.
40
20
•
0
,
2
•
,
4
•
,
6
.
,".
8
,--.-.-..~ 10
Months
12
14
~
16
•
,
18
.
,
20
Management of Patients
Surg Neurol 1991;36:107-11
109
Table 2. Clinical Status (WHO Grade) and Management Regimen Selectedfor Patients Aged Over 60 years with a Supratentorial Glioma WHO Grade Management
No. of patients
1
2
3
4
Steroids only Biopsy Cytoreduction Biopsy + RT Cytoreduction + RT
12 24 13 5 26
-1 1 -2
5 2 6 2 8
4 14 6 3 12
3 7 --4
Abbreviation: RT, radiotherapy. Steroids were prescribed in all cases, and craniotomy was performed for all cytoreductive procedures.
cult. These patients tolerate impairment of brain function poorly, and usually die several months after the diagnosis [3,4,12,19]. In patients aged under 45 years with a malignant glioma, the most successful treatment, which produces a median survival time of 24 months, has been with cytoreductive surgery followed by irradiation (60 Gy) and single-agent nitrosourea therapy [12,13]. The median survival time reported after similar treatment in patients aged over 65 years is only 27 weeks [13]. This nonrandomized observational study, which showed a similar median survival time in aggressively treated patients, suggests that such interventional management has only a small, temporal influence on disease progression and is associated with significant surgical morbidity. If it is assumed that all patients in the group subjected to craniotomy and tumor debulking were initially scheduled to go on to radiotherapy, the postoperative status of 30% of these patients would be bad enough to preclude cerebral irradiation. This figure is almost identical to the incidence of postoperative complications described in a series of 80 patients aged over 65 years who had a spectrum of brain tumors [ 16]. Paradoxically, although morbidity may be high, 30-day case fatality after craniotomy in this and other series is below 5% [11,16]. The failure of surgery to ameliorate clinical status in many patients
Table 3. Survival Times of Patients Aged Over 60 Years with Supratentorial Malignant Glioma in Relation to Their Presenting Clinical Status Survival Time (weeks) WHO Grade
No. of patients
Median
Range
1 2 3 4
4 23 39 14
32 20 14 8
24-36 7-84 1-90 1-49
will not only reflect postoperative complications but also ill-advised patient selection. Highly sophisticated, technically innovative surgery, which is associated with a low morbidity even in the elderly, does not alone contribute substantially to median survival times, although it may lead to transitory improvements in quality of life and less problems during radiotherapy [3,7,10,16]. The use of high-dose radiotherapy in some elderly patients may be inappropriate since, due to limited cerebral reserve [5], transient confusion and disorientation are common following such therapy (Whittle, Gregor, unpublished data). Lower doses (30 Gy, administered in 10 fractions) seem to be as effective and less toxic [8]. Cerebral irradiation, whether following biopsy [3] or cytoreductive surgery, seems integral to producing a definitive survival advantage [1,4,8,12], but the effectiveness of this therapy decreases with advancing age [13, 19]. If aggressive multimodality treatment is not planned, should even a biopsy be performed? A C T scan diagnosis is not infallible, and the diagnostic error rate for this condition will be at least 5 % - 10% [2,14,15 ]. This study supports the contention that CT image-guided stereotactic biopsy is a safe and effective method of establishing a histological diagnosis in the elderly [ 14]. Lesion biopsy, particularly if performed without CT-imaged targeting, is subject to the problem of tumor heterogeneity, and this may explain why in this series the median survival time for patients with histologically diagnosed intermediate-grade glioma was the same as a malignant glioma. Nonetheless, the incidence (92%) of malignant tumors among the biopsy-confirmed gliomas is remarkably similar to other series [11,16], and it may be that in the elderly, even intermediate-grade astrocytomas progress more rapidly than similar tumors in younger patients. The lack of either a consensus about optimal management or a correlation between treatment and patients' clinical status among a group of consultant neurosurgeons highlights the difficulties in deciding treatment for these patients. Although this may reflect differences in management philosophy, it is apparent that definite goals
110
Surg Neurol 1991;36:107-11
Whittle et al
T a b l e 4. Clinicopathological Details of the Seven Patients Who Survived 12 Months or Longer Survival time (months)
Sex/age (yr)
WHO Grade
Pathology
Management regimen
12 12 14 16 20 21 27
F/66 M/61 M/62 F/60 M/60 F/63 M/63
2 2 3 3 2 3 2
Glioblastoma Glioblastoma Glioblastoma Unknown AA Glioblastoma AA
Biopsy Cytoreduction + RT Cytoreduction + RT Steroids Biopsy + RT Cytoreduction + RT Biopsy
Abbreviations: AA, anaplastic astrocytoma; F, female; M, male; RT, radiotherapy. WHO grade refers to presenting neurological status.
and guidelines are required to rationalize management. A parallel study within our Regional Health Authority [15] and a study from another British unit [19] both suggested that clinicians have an unpredictable attitude toward referral of elderly patients--whose CT scan shows a brain tumor--to neurosurgeons. It is, therefore, likely that the patients in this series represent about 50% of patients with this CT diagnosis. This study confirms that clinical status at the time of presentation is an important prognostic parameter [1,4,12,13]. This parameter should probably be used when deciding clinical management. Nonetheless, some patients who present in poor clinical condition (WHO grade III) will survive between 1 and 2 years. Given this variability in natural outcome and the uncertainty as to which treatment strategy represents optimal management, it would seem appropriate that a clinical trial to address this question should be undertaken in elderly patients with glioma. A fundamental question that such a trial would also have to address is the quality of life provided by various treatments [1]. With the relative ineffectiveness of current therapy, the provision of a good quality of survival, albeit limited, may provide some justification for intervention in selected patients. At present, the cost benefit of neurosurgery for brain tumors is two orders of magnitude worse than surgery for aneurysms [9]; thus, resource and economic considerations also provide further impetus for a more objective approach to management.
The authors would like to thank Professor J. Douglas Miller and Dr. Peter Sandercock for reviewing the manuscript.
References 1. BradaM. Backto the future--radiotherapyin highgradegliomas. Br J Cancer 1989;60:1-4. 2. Choksey MS, Valentine A, Shawdon H. Computerised tomography in the diagnosis of malignant brain tumors. Do all require biopsy? J Neurol Neurosurg Psychiatry 1989;52:821-5. 3. Coffey RJ, Dunsford LD, Taylor FH. Survival after stereotaxic biopsy of malignant gliomas. Neurosurgery 1988;22:465-73. 4. Cohadon F. Indications for surgery in the management ofgliomas. Adv Tech Stand Neurosurg 1990;17:189-234. 5. Galbraith S. Head injuries in the elderly. Br MedJ 1987;294:325. 6. Gardener MJ, Altman DG. Statistics with confidence. Br Med 1989;137. 7. Kelly PJ, Kall BA, Goerss S, et al. Computer assisted sterotaxic laser resection of intraaxial brain neoplasms. J Neurosurg 1986;69:423-39. 8. Newall J, RansohoffJ, Kaplan B. Glioblastoma in the older patient: how long a course of radiotherapy is necessary? J Neurooncol 1988;6:325-7. 9. Pickard JD, Bailey S, Sanderson H, et ai. Steps towards cost benefit analysis of regional neurosurgical care. Br Med J 1990;301:629-35. 10. Sabin HI, Whittle IR. Stereotaxic craniotomy and intraoperative lesion locaiisation; technical note. Acta Neurochir (Wien) 1990;102:137-40. 11. Salcman M. Brain tumors and the geriatric patient. J Am Geriatr Soc 1982;30:501-8. 12. Shapiro WR. Therapy of adult malignant brain tumors. What have the clinical trials taught us. Semin Oncol 1986;13:38-45. 13. Shapiro WR, Green SB, Burger PC, et al. Randomized trial of three chemotherapy regimens and two radiotherapy regimens in postoperative treatment of malignant glioma. J Neurosurg 1989;71:1-9. 14. Thomas DGT, Nouby RM. Experience in 300 cases of CT-directed stereotaxic surgery for lesion biopsy and aspiration of hematoma. J Neurosurg 1989;3:337-42. 15. Todd NV, McDonagh T, Miller JD. What follows diagnosis by computer tomography of solitary brain tumor? Lancet 1987;1:611-2.
Management of Patients
Surg N e u r o l 1991;36:107-11
111
16. Tomita T, Raimondi AJ. Brain tumors in the elderly. JAMA 1981;246:53-5.
18. WHO handbook for reporting results of cancer treatment. Geneva, Switzerland: WHO, 1979:45 pp.
17. Waters CL. Surgical management of intracranial gliomas. In: Schmidek HH, Sweet WH, eds. Operative neurosurgical technique, ed 2. New York: Grune and Stratton, 1982:44780.
19. Wroe SJ, Foy PM, Shaw MDM, et al. Neurological and neurosurgical approachs in management of malignant brain tumors. Br Med J 1986;293:1015-8.
