Surgical robots have captured the interest—if not the widespread acceptance—of spinal neurosurgeons. But successful innovation, scientific or commercial, requires the majority to adopt a new practice. “Faster, better, cheaper” products should in theory conquer the market, but often fail. The psychology of change is complex, and the “follow the leader” mentality, common in the field today, lends little trust to the process of disseminating new technology. Beyond product quality, timing has proven to be a key factor in the inception, design, and execution of new technologies. Although the first robotic surgery was performed in 1985, scant progress was seen until the era of minimally invasive surgery. This movement increased neurosurgeons’ dependence on navigation and fluoroscopy, intensifying the drive for enhanced precision. Outside the field of medicine, various technology companies have made great progress in popularizing co-robots (“cobots”), augmented reality, and processor chips. This has helped to ease practicing surgeons into familiarity with and acceptance of these technologies. The adoption among neurosurgeons in training is a “follow the leader” phenomenon, wherein new surgeons tend to adopt the technology used during residency. In neurosurgery today, robots are limited to computers functioning between the surgeon and patient. Their functions are confined to establishing a trajectory for navigation, with task execution solely in the surgeon’s hands. In this review, the authors discuss significant untapped technologies waiting to be used for more meaningful applications. They explore the history and current manifestations of various modern technologies, and project what innovations may lie ahead.
Karthik Madhavan, John Paul G. Kolcun, Lee Onn Chieng and Michael Y. Wang
John L. Go, Sandy C. Lee and Paul E. Kim
✓Primary central nervous system lymphoma (PCNSL) is an aggressive neoplastic process that occurs in both immuno-competent and immunocompromised patients. Over the past 30 years there has been a steady increase in the number of cases in both patient populations. The imaging features for the disease and demographic characteristics within these patient populations vary, and in this article the authors describe the salient features of these two groups.
Patrick Paullus, Taylor A. Wilson, Paul Lee, Arunprasad Gunasekaran and Noojan Kazemi
In this video, the authors demonstrate a minimally invasive approach and resection of a paraspinal schwannoma. Using an expandable retractor, the authors were able to identify important adjacent bony landmarks and hence visualize and remove this peripheral nerve sheath tumor. While a tubular retractor is commonly used for interbody fusion procedures, the location of the tumor allowed this minimally invasive approach resulting in excellent access, minimal soft-tissue injury, and a short hospital stay. The authors present this approach as a less invasive and yet effective technique for resection of otherwise difficult-to-access nerve lesions.
The video can be found here: https://youtu.be/89OY5wdMB_k.
Richard B. North, J. Paul McNamee, Lee Wu and Steven Piantadosi
Artificial neural networks are used increasingly in applications such as graphic pattern recognition, which are difficult to address with conventional statistical methods. In the management of chronic pain, graphic methods are used routinely; patients describe their patterns of pain using “pain drawings.” The authors have previously reported an automated, computerized pain drawing methodology, which has been used by patients with implanted spinal cord stimulators to represent a technical goal of the procedure, the overlap of pain by stimulation paresthesias. Standard linear discriminant statistical methods have shown associations between stimulation parameters and electrode positions as independent variables and technical outcome and relief of pain as dependent variables.
The authors have applied artificial neural networks to the problem of optimizing implanted stimulator adjustment. A data set of 3000 electrode combinations obtained in 41 patients was used to develop a linear discriminant statistical model on a mainframe computer and to train artificial neural networks on a personal computer. The performance of these two systems on a new data set obtained in 10 patients was compared with that of human “experts.” The best neural network model was marginally better than the linear discriminant model; the variance in patient ratings was predicted by these models to a degree that the human experts were unable to predict. The authors anticipate expanding the role of these models and incorporating them into expert sytems for clinical use.
Lee A. Tan, Carter S. Gerard, Vincent C. Traynelis and Paul A. Anderson
Paul D. Woolf, Robert W. Hamill, Louyse A. Lee, Christopher Cox and Joseph V. McDonald
✓ Because of the central role of the sympathetic nervous system in mediating the stress response, plasma norepinephrine (NE), epinephrine (E), and dopamine (DA) levels were measured in 61 traumatically brain-injured patients to determine whether catecholamine (CA) levels obtained within 48 hours after injury provide reliable prognostic markers of outcome. Patient outcome was determined at 1 week using the Glasgow Coma Scale (GCS) and at the time of discharge using the Glasgow Outcome Scale (GOS). Levels of NE, E, and DA correlated highly with the admission GCS score (NE: r = 0.58, p < 0.0001; E: r = 0.46, p < 0.0025; DA: r = 0.27, p < 0.04). Moreover, in the 21 patients with GCS scores of 3 or 4 on admission, NE levels predicted outcome at 1 week. All six patients with NE levels less than 900 pg/ml (normal level less than 447 pg/ml) improved to GCS scores of greater than 11, while 12 of 15 with NE values greater than 900 pg/ml remained with GCS scores of 3 to 6 or died. Levels of E and DA were not as useful. Catecholamine levels also increased significantly as the GOS score worsened. Levels of NE and E were significantly higher in patients who died or remained persistently vegetative than in those with better outcomes. In the 54 patients who survived beyond 1 week, significant correlations were present between the length of hospitalization and NE (r = 0.71, p < 0.0001) and E (r = 0.61, p < 0.0001) levels. Concentrations of NE (r = 0.61, p < 0.0004) and E (r = 0.48, p < 0.01) were also highly correlated with the duration of ventilatory assistance. Analysis of the interactions of CA levels and GCS scores, duration of ventilatory assistance, and length of hospitalization revealed that the CA's either enhanced the reliability of the GCS score or were independent predictors of outcome.
Thus, these findings indicate that alterations in circulating CA levels reflect the severity of the neurological insult and provide support for the use of CA measurements as a physiological marker of patient outcome in both the acute and chronic phases of traumatic brain injury.
Azeem O. Oladunjoye, Rudolph J. Schrot, Marike Zwienenberg-Lee, J. Paul Muizelaar and Kiarash Shahlaie
Decompressive craniectomy plays an important role in the management of patients with traumatic brain injury (TBI) and stroke. Risks of decompressive craniectomy include those associated with cranioplasty, and may be related to adhesions that develop between the brain surface and overlying scalp and temporalis muscle. The authors report their institutional experience using a multilayered technique (collagen and gelatin film barriers) to facilitate safe and rapid cranioplasty following decompressive craniectomy.
The authors conducted a retrospective chart review of 62 consecutive adult and pediatric patients who underwent decompressive craniectomy and subsequent cranioplasty between December 2007 and January 2011. Diagnoses included TBI, ischemic stroke, intraparenchymal hemorrhage, or subarachnoid hemorrhage. A detailed review of clinical charts was performed, including anesthesia records and radiographic study results.
The majority of patients underwent unilateral hemicraniectomy (n = 56), with indications for surgery including midline shift (n = 37) or elevated intracranial pressure (n = 25). Multilayered decompressive craniectomy was safe and easy to perform, and was associated with a low complication rate, minimal operative time, and limited blood loss.
Decompressive craniectomy repair using an absorbable gelatin film barrier facilitates subsequent cranioplasty by preventing adhesions between intracranial contents and the overlying galea aponeurotica and temporalis muscle fascia. This technique makes cranioplasty dissection faster and potentially safer, which may improve clinical outcomes. The indications for gelatin film should be expanded to include placement in the epidural space after craniectomy.
Jason P. Sheehan, Zhiyuan Xu, David J. Salvetti, Paul J. Schmitt and Mary Lee Vance
Cushing's disease is a challenging neuroendocrine disorder. Although resection remains the primary treatment option for most patients, the disease persists if there is residual or recurrent tumor. Stereotactic radiosurgery has been used to treat patients with persistent Cushing's disease after a prior resection. The authors report on the long-term risks and benefits of radiosurgery for Cushing's disease.
A retrospective review of a prospectively collected database of radiosurgery patients was undertaken at the University of Virginia. All patients with Cushing's disease treated with Gamma Knife surgery (GKS) were identified. Those without at least 12 months of clinical and radiological follow-up were excluded from this analysis. Rates of endocrine remission, tumor control, and adverse events were assessed. Statistical methods were used to identify favorable and unfavorable prognostic factors.
Ninety-six patients with the required follow-up data were identified. The mean tumor margin dose was 22 Gy. The median follow-up was 48 months (range 12–209.8 months). At the last follow-up, remission of Cushing's disease occurred in 70% of patients. The median time to remission among all patients was 16.6 months (range 1–165.7 months). The median time to remission in those who had temporarily stopped taking ketoconazole at the time of GKS was 12.6 months, whereas it was 21.8 months in those who continued to receive ketoconazole (p < 0.012). Tumor control was achieved in 98% of patients. New loss of pituitary function occurred in 36% of patients. New or worsening cranial neuropathies developed in 5 patients after GKS, with the most common involving cranial nerves II and III.
Gamma Knife surgery offers a high rate of tumor control and a reasonable rate of endocrine remission in patients with Cushing's disease. The cessation of cortisol-lowering medications around the time of GKS appears to result in a more rapid rate of remission. Delayed hypopituitarism and endocrine recurrence develop in a minority of patients and underscore the need for long-term multidisciplinary follow-up.
Paul Klimo Jr., Anne Matthews, Sean M. Lew, Marike Zwienenberg-Lee and Bruce A. Kaufman
Various surgical interventions have been described to evacuate chronic subdural collections (CSCs) of infancy. These include transfontanel percutaneous aspiration, subdural drains, placement of bur hole(s) with or without a subdural drain, and shunting. Shunt placement typically provides good long-term success (resolution of the subdural fluid), but comes with well-known early and late complications. Recently, the authors have used a mini–osteoplastic craniotomy technique with the goal of definitively treating these children with a single surgery while avoiding the many issues associated with a shunt. They describe their procedure and compare it with the traditional bur hole technique.
In this single-institution retrospective study, the authors evaluated 26 cases involving patients who underwent treatment for CSC. Preoperative, intraoperative, and postoperative data were reviewed, including radiographic findings (density of the subdural fluid and ventricular and subarachnoid space size), neurological examination findings, and intraoperative fluid description. The primary outcome was treatment failure, defined as the patient requiring any subsequent surgical intervention after the index procedure (minicraniotomy or bur hole placement).
Fifteen patients (10 male and 5 female; median age 5.1 months) collectively underwent 27 minicraniotomy procedures (each procedure representing a hemisphere that was treated). In the bur hole group, there were 11 patients (6 male and 5 female; median age 4.6 months) with 18 hemispheres treated. Both groups had subdural drains placed. The average follow-up for each treatment group was just over 7 months. Treatment failure occurred in 2 patients (13%) in the minicraniotomy group compared with 5 patients (45%) in the bur hole group (p = 0.09). Furthermore, the 2 patients who had treatment failure in the minicraniotomy group required 1 subsequent surgery each, whereas the 5 in the bur hole group needed a total of 9 subsequent surgeries. Eventually, 80% of the patients in the minicraniotomy group and 70% of those in the bur hole group had resolution of the subdural collections on the last imaging study.
The minicraniotomy technique may be a superior technique for the treatment of CSCs in infants compared with bur hole evacuation. The minicraniotomy provides greater visualization of the subdural space and allows more aggressive evacuation of the fluid, better irrigation of the space, the ability to fenestrate any accessible membranes safely, and continued egress of fluid into the subgaleal space. Although this preliminary report has obvious limitations, evaluation of this technique may be worthy of a prospective, multiinstitutional collaborative effort.
Lee A. Tan, Carter S. Gerard, Paul A. Anderson and Vincent C. Traynelis
Iatrogenic foraminal stenosis is a well-known complication in cervical spine surgery. Machined interfacet allograft spacers can provide a large surface area, which ensures solid support, and could potentially increase foraminal space. The authors tested the hypothesis that machined interfacet allograft spacers increase cervical foraminal height and area.
The C4–5, C5–6, and C6–7 facets of 4 fresh adult cadavers were exposed, and the cartilage was removed from each facet using customized rasps. Machined allograft spacers were tamped into the joints. The spines were scanned with the O-arm surgical imaging system before and after placement of the spacers. Two individuals independently measured foraminal height and area on obliquely angled sagittal images.
Foraminal height and area were significantly greater following placement of the machined interfacet spacers at all levels. The Pearson correlation between the 2 radiographic reviewers was very strong (r = 0.971, p = 0.0001), as was the intraclass correlation coefficient (ICC = 0.907, p = 0.0001). The average increase in foraminal height was 1.38 mm. The average increase in foraminal area was 18.4% (0.097 mm2).
Modest distraction of the facets using machined interfacet allograft spacers can increase foraminal height and area and therefore indirectly decompress the exiting nerve roots. This technique can be useful in treating primary foraminal stenosis and also for preventing iatrogenic foraminal stenosis that may occur when the initially nonlordotic spine is placed into lordosis either with repositioning after central canal decompression or with correction using instrumentation. These grafts may be a useful adjunct to the surgical treatment of cervical spine disease.