A Questionnaire-based Survey of Clinical Neuro-oncological Practice in India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.319199
Source of Support: None, Conflict of Interest: None
Keywords: Glioma, metastases, neuro-oncology, neurosurgery, protocols, subspecialties, surveyKey Message: Neuro-oncology has emerged as a specialty in neurosurgery, yet our survey reveals that practice patterns are variable even in high-volume centers. Standardized protocols and formal systematized training modules for neuro-oncology are the need of the hour.
Until recently, neuro-oncology and brain tumor surgery were not considered to be viable subspecialties. However, with the advent of molecular diagnostics and advances in surgical technology and techniques, neuro-oncology has blossomed into a major sub-specialty of neurosurgery.,
The Society for Neuro-Oncology was established in 1996 and the Indian Society of Neuro-oncology (ISNO) came into existence in 2008. Neurosurgeons in India have been slow to accept and recognize the need for neuro-oncology as a separate subspecialty. In the 10th anniversary year of the founding of ISNO, we conducted a survey among Indian neuro-oncologists to assess the knowledge of, attitudes towards, and practice of (KAP) neuro-oncology in India.
This online survey was designed to collect data regarding the various domains of relevance to neuro-oncology as a subspecialty. The survey comprised 45 questions and was administered on the SurveyMonkeyTM platform. All clinicians who deal with brain tumors in their practice setting (neurosurgeons, neurologists, medical, and radiation oncologists) and were practicing in India at the time the survey was conducted were considered eligible to participate in the survey. Emails containing the link to the survey were sent out to all members of the Neurological Society of India and ISNO. At the end of the survey period, the responses were collated and analyzed.
The survey was mailed to 2447 individual email IDs; 53 (2.2%) email IDs were incorrect and 46 invitees (1.9%) opted out of the survey. Overall, 319 responses were received, 191 were complete, and 128 were partial; the response rate was 13%. The specialty distribution of the respondents was – 282 (88.4%) neurosurgeons, 5.6% were neurologists, 1.9% were radiation oncologists, and 1 response (0.3%) was received from a medical oncologist. The number of responses received to each question varied since no question was compulsory and some questions were not open to respondents of certain specialties. Therefore, the number of responses (denominator) are mentioned as appropriate.
Demographics (n = 226)
South India accounted for 48% of the responses, followed by 26.5% from North India. Seventy respondents (31%) were employed in the government sector, 23 (10%) in not-for-profit or charitable institutions, and 131 (58%) in the private sector. The majority (n = 158, 70%) worked in a setting where MCh/DM/DNB courses were conducted. Among neurosurgeons (n = 211), most respondents had less than 5 years of experience (43%) whereas 26% had >10 years of experience, 15% had 5–10 years of experience, and 17% were residents.
The operative load of the institutions where respondents worked is shown in [Figure 1]a and [Figure 1]b. Although 37% of the institutions had reasonable surgical volumes (>1 procedure/day), only about half of these had high volumes of malignant brain tumor surgery.
Medical aspects of neuro-oncological management (n = 226)
Drugs frequently prescribed to patients with brain tumors include steroids, anti-epileptic drugs (AEDs), and mannitol [Figure 2]. Prophylactic AEDs were routinely prescribed to all patients with brain tumors by 87% (n = 197), while 84% reported the routine use of steroids and 42% reported the routine use of mannitol. Other drugs such as analgesics, antibiotics, and 3% normal saline are routinely prescribed by 7% of the respondents.
When steroids are indicated, there are a variety of options and routes of administration. Oral dexamethasone was used by 28% of the respondents, 63% reported using intravenous dexamethasone, and 3% reported using methylprednisolone as the preferred steroid. Overall, 11% reported not using steroids routinely. Of the others, 30% used steroids for 1 week post-surgery, 55% used a 3-3-3-day taper, and the remaining 5% reported various other durations of use. When AEDs were used, phenytoin was the drug of choice in 47%, levetiracetam in 48%, valproate in 3%, and carbamazepine in 0.1%. If a patient had not had a seizure prior to surgery, 3% respondents reported not using routine postoperative AEDs, 29% reported using AEDs for 1–3 weeks post-op, and 68% reported using it for longer [Figure 3].
General surgical policies (n = 195)
Hair preparation was most commonly performed (80%) using a chlorhexidine or betadine hair wash on the evening prior to or on the morning of the surgery. Full head shave was preferred by 42%, 20.5% perform a partial head shave, 26% perform incision-line hair clipping, 26% take the patient's preference into account, and only 3% perform surgery without shaving or clipping. As per 47% of responses, hair removal was done in the operating room after induction, whereas in 51% cases it was done outside the operating room (evening prior to or the morning of surgery).
Most respondents (84%) perform a watertight dural closure for all cases (with duraplasty if required), 6% use watertight duraplasty only for infratentorial tumors, and 8% use only tacking sutures in all cases [Figure 4]a. Bone flap is replaced and fixed using sutures by 41.5% respondents, 40.5% use mini-plates and screws for fixing the bone flap, while 12% do not anchor the bone [Figure 4]b.
During scalp closure, a subgaleal suction drain is used routinely by 72%, 13% use a drain only if the dura was closed watertight, and 6% if the dura was not closed watertight, only for posterior fossa surgeries (1.5%), for skull base surgeries (3%) or in case if inadequate homeostasis (12%).
The most widely available intraoperative tool was neuronavigation (47%), followed by intraoperative ultrasonography (34%), and 3D ultrasound in 4%. Fluorescence-guided resection is used by 20%. Other intraoperative adjuncts used include intraoperative MRI (7%) and intraoperative CT (4%).
Awake craniotomy is routinely utilized by 43.5%; 33% of respondents routinely use cranial nerve ENMG monitoring and 29% use multimodality electrophysiology systems. Interestingly, 45.5% individuals use none of the abovementioned modalities. As per 46% of responses, IONM/awake craniotomy is used in less than 50% cases of eloquent or perieloquent area gliomas, while 21% of those surveyed use it in all such cases [Table 4].
In all, 211 responses were received regarding perioperative antibiotic policies. The majority (n = 141, 67%) reported starting antibiotics at surgery and continuing it for 1–3 days post-op. Only 19% reported using a single preoperative dose of prophylactic antibiotics. There was no uniformity in the choice of drug; third-generation cephalosporins were the most commonly used antibiotics.
Protocols regarding venous thromboembolism (VTE) prophylaxis elicited 195 responses; 58% mentioned the regular use of nonpharmacological measures for all patients in the immediate postoperative period, whereas 26% use the same only for patients who have limb paresis or are unconscious. Twenty-four percent of the respondents reported using prophylactic anticoagulation in patients with limb paresis. In most cases, anticoagulation was reportedly started 24–48 hours after surgery and 14% reported the lack of a specific policy for VTE prophylaxis.
Fifty-two percent of 195 respondents perform a routine postoperative plain CT brain in the early postoperative period for all patients who undergo surgery for brain tumor, contrast-enhanced CT brain is used routinely by 22%, and contrast-enhanced MRI brain is used by 10% and 14.5% of those surveyed do not perform a routine postoperative imaging, unless warranted by the clinical condition.
Neuro-oncology as a subspecialty (n = 166)
Gliomas are routinely operated on by senior consultants as per 64% of responses. Nearly two-thirds of the respondents (63%, n = 104) were of the opinion that specialist neuro-oncologists are required in India and 61% (n = 101) of those surveyed also expressed interest in pursuing neuro-oncology as their primary practice specialty.
This survey reveals wide variations in the practice of neuro-oncology across India.
Drugs used in neuro-oncology
Steroids have had an assured place in neuro-oncological practice since the 1960s; however, there are considerable variations in their continuation after surgery. A recent cohort-study showed that patients operated for glioblastoma who were not steroid dependent had a lower risk of mortality. The study also opined that in most instances, steroids can be stopped within 2 weeks after surgery. Our survey shows a similar pattern of practice with most responders continuing steroids for 1 week after surgery or using a 3-3-3 day taper.
AEDs have traditionally been prescribed to all patients with brain tumors. Several recent studies suggest that AEDs are not routinely indicated in patients with brain tumors undergoing craniotomy, in the absence of de-novo seizures., However, the debate continues; a recent Cochrane review concluded that available evidence regarding this issue is all of low-quality. The majority of practitioners we surveyed (87%) routinely prescribe AEDs for all patients with brain tumors, with phenytoin being the preferred drug for 47%. It is now well accepted that levetiracetam is as effective as phenytoin for seizure control or prophylaxis and has a better adverse effect profile, making it the AED of choice. The continuing popularity of phenytoin presumably stems from its significantly lower cost.
A large proportion of practitioners in India perform routine pre-operative hair shaving for cranial surgeries. Concerns regarding hair-preserving surgery arise from the presumed risk of infection and poor patient hygiene. However, current literature supports either completely avoiding hair removal or at most incision line clipping; these have been shown to have similar rates of postoperative surgical site infection with better esthetic results. In fact, shaving hair is considered to predispose to infections by causing microabrasions that allow translocation of the scalp flora. Since previous studies have shown that scalp flora among Indian patients is similar to that described in literature, these results should be applicable to our practice as well. Difficulty in identifying surface landmarks for marking the scalp flap is another issue in hair-preserving surgery. The availability of neuronavigation has greatly aided in marking precise scalp flaps and should be used to avoid hair removal.
Venous thromboembolism (VTE) prophylaxis must be considered in neurosurgical patients who are in altered sensorium, have limb paresis, or are not ambulant. This can be either via mechanical devices or pharmacological means. While traditional recommendations provided by NICE and the American Association of Hematology advise preferential use of mechanical devices in low risk patients, there is mounting evidence in favor of anticoagulant usage. As of now, the issue remains unresolved and at least mechanical prophylaxis is recommended. Developing an institutional policy can help in further research in this area.
Specific clinical scenarios
The currently available options for management of cerebral metastasis include surgical excision, stereotactic radio surgery (SRS), and whole brain radiotherapy (WBRT). WBRT used to be the only treatment option for cerebral metastasis prior to the RTOG trial conducted by Patchell et al., which provided level 1 evidence for the superiority of surgery followed by WBRT over WBRT alone in solitary brain metastasis. This has been the mainstay of management ever since. In addition to improving survival, surgery also provides the benefit of reducing mass effect from the tumor and improved local control. Lesions greater than 3 cm in size in surgically accessible areas with or without mass effect are typically subjected to excision. The Japanese Radiation Oncology Study Group (JROSG) trial and the European Organization for Research Treatment of Cancer (EORTC) trial showed that the addition of WBRT after either surgery or SRS for an oligometastatic disease (1–4 metastasis) does not improve the overall survival, although local and distant relapses inside the brain are significantly lessened.,
SRS in the treatment of cerebral metastasis has been a subject of great interest in recent years. Lesions less than 3 cm in size or located in deep areas of the brain are ideal candidates for SRS. In addition to focused radiation and short treatment duration, SRS does not require the interruption of systemic chemotherapy for the primary disease and can treat multiple lesions simultaneously. The use of SRS boost after WBRT in patients with multiple metastasis has also been shown to improve local control rates. In patients with an intermediate (3–12 months) or poor prognosis (< 3 months), the use of WBRT is controversial because of its acute side effects and detrimental effect on quality of life. For these patients, palliative care, primary systemic treatment, and SRS may be preferred over WBRT on an individualized patient basis.
The value of technology to optimize resection and minimize morbidity has been well established. Of all the available adjuncts, neuro-navigation seems to have had the best penetrance. A couple of worrying trends emerged regarding the use of adjuncts for safe surgery. Among the respondents, 45.5% reported using neither navigation nor IONM. This may be considered unacceptable in today's practice milieu. Again, per 46% of responses, IONM/awake craniotomy was used in less than 50% cases of eloquent or near eloquent gliomas, while only 21% of those surveyed use it in all such cases. High equipment cost could be one issue that is responsible for the low utilization rates of these proven technologies. Indigenous low-cost alternatives are the need of the hour. Another issue could be the lack of familiarity and exposure to these technologies, especially during the residency phase. This issue needs to be remedied and certain basic minimum equipment should be identified when a center is recognized to train residents.
Neuro-oncology as a specialty
This survey demonstrated the significant heterogeneity in the practice of neuro-oncology, even in domains where evidence-based guidelines exist. This emphasizes the need to develop standardized protocols for various conditions, as the ISNO has been doing., There is also a need to disseminate these guidelines so that they inform the standard of care. The institution of a journal dedicated to neuro-oncology by the ISNO is a step in this direction.
There is now data that shows that patients with gliomas fare better when treated by surgical neuro-oncologists with specialist training. Extent of resection, surgical mortality, and survival times were all better in the group treated by specialist surgical neuro-oncologists vis a vis those treated by neurosurgeons not specialized in neuro-oncology. This data argues for the involvement of multispecialty tumor boards in the clinical decision-making process for all recurrent or residual brain tumors and for uncommon pathologies. Since it may not be feasible to set up physical tumor boards in many centers, virtual tumor boards could serve this important need.
In this context, it is heartening that nearly two thirds of the respondents recognize the need for neuro-oncology as a specialty and are willing to embrace it and are ready to train in it. Routine employment of advanced operative technology for safe maximal resection of brain tumors mandates familiarity with these techniques in this day and age. Shorter and long-term training programmes, therefore, appear to be the need of the hour.
Limitations of the data
One major limitation of a survey is perception bias, both on the part of the surveyor and those being surveyed. We attempted to overcome this issue in two ways—first, by framing questions around clinical scenarios and second, by allowing free text answers to several questions. The other major limitation of Internet-based surveys is the response rate. Whereas to get an intuition regarding the opinion of the majority of the surveyed population, a >50% response rate would be required, this is often impossible to achieve, despite multiple reminders. We were able to achieve a response rate of 13% after three reminders were sent out. Mathematical modeling studies have been carried out to establish the minimum acceptable response rate for various population sizes, so as to have statistically valid information obtained from a survey. For a population size of 2447 (number of emails sent out), the number of responses would need to be between 322–341 to have an alpha < 0.05 for the responses. We received 319 responses, and therefore, our response rate could be considered statistically close to the ideal rate. We had very few responses from non-neurosurgeon neuro-oncologists and therefore the practice patterns reflected in this survey are largely from a neurosurgical perspective.
Neuro-oncology needs to be developed as a separate subspecialty in the same vein as vascular, functional, or skull base neurosurgery. The insights gained from this survey can identify the potential areas of research and investment, promote standardized evidence-based neurosurgical practice, and help in the institution of specialized neuro-oncology training programmes.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
[Table 1], [Table 2], [Table 3], [Table 4]