Varidh Katiyar

Microvascular decompression versus stereotactic radiosurgery as primary treatment modality for trigeminal neuralgia: A systematic review and meta-analysis of prospective comparative trials

Objective: The current opinion among neurosurgeons regarding the selection between microvascular decompression (MVD) and gamma knife radiosurgery for trigeminal neuralgia is not based on clear evidence. In this meta-analysis, we have attempted to synthesize the findings of the prospective trials comparing the efficacy and complications of the two procedures as primary treatment modality for medically refractory trigeminal neuralgia. Materials and Methods: The authors performed a systematic review of PubMed for manuscripts comparing the efficacy or complications of MVD and stereotactic radiosurgery for medically refractory trigeminal neuralgia. The data of the identified studies was pooled and a meta-analysis was done. Results: Five prospective studies fulfilling the eligibility criteria were identified. The mean age of the patients subjected to gamma knife therapy (GKT) was more than those who underwent MVD. The initial success rate in the pooled data with MVD was 96% (95% confidence interval [C.I.] 93.3%–98.6%) as compared to GKT which was 71.8% (95% C.I. 64.9%–78.7%) with the ratio of 1.309 (95% C.I. 1.217–1.409; P= <0.001). This superiority was sustained till the last follow up available in all the studies. Out of the complications common to both procedures, MVD had a lower rate of facial numbness, with a risk ratio of 0.481 (95% C.I. 0.297–0.778); and dysesthetic pain, with a risk ratio of 0.470 (95% C.I. 0.172–1.286). Conclusions: MVD seems to be more efficacious than GKT as a first line treatment for trigeminal neuralgia immediately as well as on a long term basis. However, the dilemma regarding the choice of treatment to be adopted still remains for special subgroups of patients, like the elderly patients and those in whom no vascular compression has been found during surgery. Further studies are needed for elucidating the unequivocal treatment plan under these circumstances.

Basic Physics for a Surgeon

Dear editor, We read with keen interest the article regarding the physical principles important for a surgeon titled BRajkumar, J., Chopra, P., and Chintamani. (2015). Basic Physics Revisited for a Surgeon.^ The Indian Journal of Surgery, 77(3), 169–175 [1]. We would like to commend the authors for highlighting the importance of knowledge of basic principles of physics for surgeons, an inspiration that drives all the major breakthroughs in scientific disciplines. However, we would also like to point out certain issues, e.g., gastro-esophageal reflux disease (GERD) and the role of Poiseulli’s theorem. The compression of lower esophagus by raised intra-abdominal pressure post prandially plays a major role in preventing reflux. The physics behind compressibility of compliant tubular structures under external pressure explained by soft straw experiment has been wrongly attributed to Poiseulli’s equation. For this, the authors must have explained how the pressures inside a fluid increase in direct proportion to the depth. Even this fails to capture what actually happens in this situation as the compression of esophagus due to intra-abdominal pressure is not akin to greater compression at increasing depth in a fluid. Another example of Poiseulli’s equation is for correcting arterial narrowing at two or more sites. We would like to explain this concept further using hypothetical values as it might initially seem counter intuitive to some of the intended readers that how a similar increase in radius may result in greater increase in the flow rate in larger artery though the relative increase in radius is lesser for the larger artery. Let us suppose there are two arteries with radii of 3 units and 4 units, then the flow rate should be 3^4 = 81 and 4^4 = 256 units. On increasing the radii of both by 1 unit, the flow rates become 256 units and 5^4 = 625 units. An increase of 175 units in the smaller and 369 units in the larger artery though there was 33% relative increase in radius of smaller artery compared to only 25% in the larger artery. Bernoulli’s principle is primarily about the radial pressure at the site of high-velocity flow rather than proximal rise in pressure due to increased resistance distally. In nonrigid tubular structures, it does not explain in entirety the concept of treatment of leak after sleeve gastrectomy with stent placement. The authors also seemed to have simplified the principles of diathermy and have not explained the role of the different frequency and duration of stimulation in addition to current density in deciding the relative amounts of cutting and coagulation of the tissues. Concluding, we agree with the authors on the importance of understanding basic principles of physics. We have highlighted certain aspects that the authors have either missed or might have omitted due to the intended readership belonging to a biological science domain.

Can diabetic ketoacidosis (DKA) precipitate posterior reversible encephalopathy syndrome (PRES)

Dear Editor: We read with keen interest the case report by Nao et al. [1] describing posterior reversible encephalopathy syndromewith spinal cord involvement (PRES-SCI) occurring as a complication of diabetic ketoacidosis. We would like to commend the authors for reporting a rare case of PRES-SCI in a patient with diabetic ketoacidosis (DKA). On reviewing the literature, we came across two similar case reports [2, 3] in which PRES occurred as a complication of DKA as discussed in Table 1. However, we would like to discuss a few issues in this report. The authors have reported that the patient developed numbness with paraplegia on day 5 along with headache and blurring of vision after resolution of DKA on day 4. The findings in magnetic resonance imaging (MRI) of brain and spinal cord were supportive of PRES-SCI. De Havenon et al. [4] in their series of eight patients reported that all patients had acute hypertension at the time of onset of neurological deficits, so it would have been interesting if authors had mentioned whether there was any evidence of systolic or diastolic hypertension at the time of visual deterioration and what were the fundus findings. Also, the authors did not describe the metabolic parameters of the patient in the ICU after neurological worsening as dyselectrolytemia may also interfere with cerebrovascular auto-regulation. Hypomagnesaemia has been reported to cause transient neurological symptoms associated with reversible T2 signal changes on MRI of PRES [5]. The authors should also have mentioned if there was any evidence of anemia or azotemia. Diabetic uremic syndrome has been shown to result in bilateral reversible lesions in the brain [6]. The authors have reported lower motor neuron features in their patient. However, this is in contrast to the previously reported cases of PRES in which the patients had upper motor neuron features with hyperreflexia and hypertonia [4]. Though the authors acknowledge that they could not rule out the possibility of myelitis and consequently started the patient on methylprednisolone, the rationale for this treatment is unjustified as it could further worsen glycemic control in a patient with history of DKA. Though the steroids might have helped in resolution of vasogenic edema in this case, however, most of the authors have reported gradual and complete resolution of the symptoms with adequate control of blood pressure and consider hyper-perfusion injury to be the main pathophysiological mechanism underlying PRES-SCI. * Hitesh Gurjar hiteshgurjar@gmail.com

Early tracheostomy in traumatic brain injury: conundrum continues…

We note with interest the article by Shibahashi et al. regarding the effect of tracheostomy performed within 72 hours following traumatic brain injury. We commend the authors for bringing this question in light, as there is no consensus regarding the optimal timing of doing tracheostomy after traumatic brain injury (TBI). Given the advantages of tracheostomy over endotracheal intubation, Plummer et al. recommended that tracheostomy should be performed if an artificial airway is anticipated for greater than 21 days. But in the current scenario, the decision regarding the timing of tracheostomy was left to the physician. Recent studies have advocated early tracheostomy after TBI because of advantages like decreased duration of mechanical ventilation and length of stay (LOS) in the intensive care unit (ICU). Unfortunately, the definition of early and late tracheostomy differs among studies with the earliest tracheostomy day chosen as early as 5th/6th day after injury in previous studies for group division. The authors in this study used the 72-hour cut-off for dividing into early and late groups. However, we would like to highlight issues in this article that need further consideration. The authors did not mention whether patients with baseline pulmonary diseases such as asthma, interstitial lung or COPD were excluded from the study as these could have confounded the results including the time taken for weaning off, LOS in ICU and associated morbidity and mortality. They have mentioned that tracheostomy was performed by the standardized technique using a low-pressure cuffed tracheostomy tube, but it is unclear whether it was done under GA or by percutaneous technique under local anesthesia. This could have a differential impact on the peri-procedural inflammatory markers that they have evaluated. The authors have not commented upon any contraindications for performing tracheostomy such as coagulopathy, which is quite common in the peri-traumatic period that could have resulted in a decision for delayed tracheostomy. The authors excluded 2 patients because of upper airway obstruction that required urgent tracheostomy but whether patients with maxillofacial trauma were excluded is unclear. The authors say that patients in the early group (n1⁄4 40) were expected to have a worse prognosis in terms of probability of survival, risk of mortality at 14 days and unfavorable outcome than those in the late group (n1⁄4 51) according to various prognostic models (CRASH and TRISS). However, the results seem quite surprising as the median total duration of mechanical ventilation [early group-5 (4–6); late group-8 (6–10)] and median LOS in ICU [early group-10 (7–13); late group-11 (10–15)] were significantly less in the early group. Also the time from tracheostomy to ventilator weaning was the same [2 (1–4) days] in both the groups. This median figure and the inter-quartile range are quite overwhelming considering the early subgroup had predicted mortality of more than 50% within the next 14 days. Even in the randomized controlled study by Bouderka et al, which used 5th/6th day as cut-off for deciding early versus late group, the duration of mechanical ventilation was never less than 14 days in the early tracheostomy group. Though the authors have acknowledged that these prognostic models took into account only the initial injury and effects of secondary injury were not considered, it might have been possible that patients with lower degrees of secondary brain were selected for the early tracheostomy group that resulted in early group fairing well; however, it is surprising that most of the patients were weaned off the ventilator within 2 days after being tracheostomised. Similar contradictory figures compared to initial prediction were also seen in the 30-day mortality with only one patient death in the early group, and the favorable GOS at time of discharge was similar i.e. 40% in both the groups. In addition, the decision to perform tracheostomy was on the basis of surgeon’s anticipation that the patient was going to need an artificial airway for more than 7 days; this introduces a selection bias in deciding between early and late groups. Moreover, the delay to perform tracheostomy because of family decision-making was not included in the study, which is a very common factor we see in ICU settings. The authors did not consider other independent factors, which may have played a role in patients’ outcomes such as APACHE and Simplified Acute Physiology Score (SAPE). In conclusion, we would like to commend the authors for the study highlighting the benefits of early tracheostomy but what is needed is a prospective study using standardized guidelines as to when to perform tracheostomy, removal of surgeons’ subjective bias and objectify it using a standardized scoring system. Otherwise, the debate for any specific time cut-off would remain unresolved.

Is decision-making easier post RESCUE ICP trial?

Dear editor, We studied with keen interest the article by Wettervik et al. regarding their experience of the role of decompressive craniectomy in patients with traumatic brain injury (TBI) as a primary or a stepwise procedure [4]. It is commendable that in this study, the authors have attempted to evaluate the role of decompressive craniectomy with a protocol that resembles the realworld scenario. The authors have correctly pointed out the artificiality of the decision-making process used in the only two large randomized controlled trials (RCTs), i.e., DECRA trial and RESCUE ICP [2, 3]. The findings of these RCTs have been a reason for major controversy since their publication. While DECRA included only bifrontal craniectomies done as an early intervention as a stage 2 treatment, RESCUE ICP evaluated unilateral decompressive craniectomies as well and only as the last tier treatment. However, there was a high crossover rate of 37.2% from medical to decompressive craniectomy group in the RESCUE ICP trial. The two RCTs have brought forth contrasting conclusions as well. While the DECRA trial shows that decompressive craniectomy results in a worse outcome at 6 months, RESCUE ICP shows the results to be similar between the two groups at 6 months and better outcome with decompressive craniectomy at 12 months. The differences in the inclusion criteria may be responsible for this difference in the findings. An important finding that was missing in both these trials was whether cranioplasty was done before the outcome assessment in some of the patients as cranioplasty has been shown to improve outcomes and thus may act as a confounding factor distorting the results [1]. By virtue of being randomized clinical trials, the selection criteria used seem to be unrelated to the routine clinical decision-making. By classifying the patients under no thiopental/no DC, no thiopental/No DC, thiopental/DC, and thiopental/No DC groups, the authors have attempted to capture the usual decision-making process in a clinical scenario. However, we would like to point out that since such classification was done nonrandomly on the basis of clinical features, it is capable of providing only descriptive results, as a control group with similar clinical features is not available. It is necessary to capture the merits of RCT and simulate clinical decision-making together; however, for that, a very large sample size with random allocation at each point of clinical decision-making will be needed. Also, there must be a clear protocol regarding the intensity and appropriateness of the medical treatment instituted. There must also be an attempt to evaluate decompressive craniectomy as the primary procedure which DECRA and RESCUE ICP have not discussed. Thus, we commend the effort of the authors to match the clinical scenario closely; however, since there is no clinically similar comparison group, this study just serves to describe authors’ experience. Due to the systematic nonrandom choice of treatment modality based on clinical status, it is not feasible to draw any conclusion regarding what might be the best management protocol in such patients. Thus, the need of the hour is to conduct RCTs with randomization at each objectively defined clinical decisionmaking point. Such a study might go a long way in putting forth guidelines to choose between the medical and surgical management in TBI according to the clinical scenario.

Systemic inflammatory response in pediatric central nervous system tumors

Dear editor, We read with great interest the article by Wilson et al. [4] titled Pre-operative neutrophil count and neutrophillymphocyte count ratio (NLCR) in predicting the histological grade of paediatric brain tumours: a preliminary study. We commend the authors for undertaking this study to evaluate the feasibility of using a common laboratory parameter as a marker for grade of intra-axial brain tumor. In this study, the authors have systematically excluded those with recurrent tumor, metastasis, and bone marrow transplant recipients. These factors may act as confounders, and thus, their exclusion was warranted. However, with a marker such as differential leucocyte count, which is commonly perturbed by conditions much more prevalent than brain tumors, these conditions must have been excluded to realize the true value of this marker. Acute infectious conditions may alter these values randomly and result in dilution of the observed association. Also, other diseases like congenital heart disease, past history of infection and drugs may alter the NLR [2]. Use of steroids or a hypercortisolic state due to stress reaction may also significantly impact the NLR, and thus, cortisol levels must have been measured and those with abnormal levels must have been either excluded or results have been statistically adjusted. Similarly, the other probable confounders as discussed above were neither excluded nor systematically analyzed by the authors. Since, the proposed hypothesis for association of NLR with grade of tumor is systemic inflammatory activity, it seems plausible that other markers of inflammation like erythrocyte sedimentation rate (ESR), C reactive protein (CRP), or procalcitonin may also be related to the grade of pediatric brain tumors. It would have been interesting if these parameters were evaluated and compared as markers [1]. The authors have excluded patients of craniopharyngioma in this study as they considered it extra-axial tumor. However, Chen et al. [3] found that preoperative inflammatory markers including white blood cells (WBC), neutrophil, and NLR + platelet lymphocyte ratio (PLR) were to be related to the pathogenesis of craniopharyngioma. Thus, the reason for exclusion of craniopharyngioma seems unsatisfactory, and it would have been informative to know their findings regarding this subset as well. Also, authors have not separately analyzed their findings for different types of pediatric brain tumors. It would be of greater clinical significance if the role of NC, NLR, and other inflammatory markers is known for each type of tumor rather than a group as a whole. We would like to commend the authors for discussing an important topic; however, there were some shortcomings in the study as we have discussed. Thus, there is a need for a prospective study, which takes into account all the possible confounders, and with a greater sample size to detect the intended association with adequate power.

Post-traumatic hydrocephalus following decompressive craniectomy: how well can it be predicted?

Dear Editor, We studied with keen interest the article by Nasi et al. regarding the risk factors for post-traumatic hydrocephalus (PTH) after decompressive craniectomy in traumatic brain injury patients and their 6-month clinical outcomes [2]. We commend the authors on undertaking an evaluation of these risk factors as early prediction of hydrocephalus may facilitate early anticipatory management and result in a better long-term outcome vis-à-vis a late responsive intervention. In this study, the authors have found radiological finding of interhemispheric hygroma and delayed cranioplasty to be significantly associated with development of post-traumatic hydrocephalus. The predictive potential of interhemispheric hygroma has been reported by other authors as well. According to these findings, subdural hygroma preceded the development of PTH and can be a good predictor for the same. One of the studies by Kaen et al. [1] reported a sensitivity of 94% and specificity of 96%. However, Kaen et al. [1] also observed the appearance of PTH after the resolution of subdural hygroma. The authors failed to acknowledge this interesting finding and have not commented on the status of subdural hygroma in their surgical series at the time of development of PTH. There have been several studies evaluating different aspects of debate between early and late cranioplasty. It is traditionally classified at an arbitrary cutoff of 3 months. The authors have used this cutoff of 3 months only which seems to be counterintuitive for the purpose of this study because the assessment of post-traumatic hydrocephalus was done at 1 month after decompressive craniectomy in this study. Thus, it must have been classified at 1 month to understand the impact cranioplasty has on the development of PTH. An early cranioplasty done between 1 and 3 months after decompressive craniectomy is not expected to benefit a patient in terms of development of PTH at 1 month vis-à-vis delayed cranioplasty. In this study, there were 37 patients who developed PTH out of whom 3 underwent early cranioplasty. The authors have described 3 patients who developed PTH and were managed by cranioplasty and lumbar drain insertion. It must have been made clear that these 3 patients are the same or not. If they are, then cranioplasty is more of a therapeutic intervention done for PTH rather than a risk factor for the same. It does not preclude the predictive potential of early cranioplasty but to assess that either the association must be evaluated with PTH at >3 months or the cutoff time for cranioplasty must be reduced accordingly. The choice between ventriculoperitoneal shunt and urgent cranioplasty and lumbar drain insertion as the therapeutic modality should also have been explained.