Search Results

You are looking at 1 - 5 of 5 items for

  • Author or Editor: Rishi Lall x
Clear All Modify Search
Free access

Anthony V. Nguyen, Elizabeth E. Blears, Evan Ross, Rishi R. Lall and Juan Ortega-Barnett

OBJECTIVE

Glioblastoma (GBM) and primary central nervous system lymphoma (PCNSL) are common intracranial pathologies encountered by neurosurgeons. They often may have similar radiological findings, making diagnosis difficult without surgical biopsy; however, management is quite different between these two entities. Recently, predictive analytics, including machine learning (ML), have garnered attention for their potential to aid in the diagnostic assessment of a variety of pathologies. Several ML algorithms have recently been designed to differentiate GBM from PCNSL radiologically with a high sensitivity and specificity. The objective of this systematic review and meta-analysis was to evaluate the implementation of ML algorithms in differentiating GBM and PCNSL.

METHODS

The authors performed a systematic review of the literature using PubMed in accordance with PRISMA guidelines to select and evaluate studies that included themes of ML and brain tumors. These studies were further narrowed down to focus on works published between January 2008 and May 2018 addressing the use of ML in training models to distinguish between GBM and PCNSL on radiological imaging. Outcomes assessed were test characteristics such as accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC).

RESULTS

Eight studies were identified addressing use of ML in training classifiers to distinguish between GBM and PCNSL on radiological imaging. ML performed well with the lowest reported AUC being 0.878. In studies in which ML was directly compared with radiologists, ML performed better than or as well as the radiologists. However, when ML was applied to an external data set, it performed more poorly.

CONCLUSIONS

Few studies have applied ML to solve the problem of differentiating GBM from PCNSL using imaging alone. Of the currently published studies, ML algorithms have demonstrated promising results and certainly have the potential to aid radiologists with difficult cases, which could expedite the neurosurgical decision-making process. It is likely that ML algorithms will help to optimize neurosurgical patient outcomes as well as the cost-effectiveness of neurosurgical care if the problem of overfitting can be overcome.

Free access

Rishi R. Lall, Rohan R. Lall, Jason S. Hauptman, Carlos Munoz, George R. Cybulski, Tyler Koski, Aruna Ganju, Richard G. Fessler and Zachary A. Smith

Spine surgery carries an inherent risk of damage to critical neural structures. Intraoperative neurophysiological monitoring (IONM) is frequently used to improve the safety of spine surgery by providing real-time assessment of neural structures at risk. Evidence-based guidelines for safe and efficacious use of IONM are lacking and its use is largely driven by surgeon preference and medicolegal issues. Due to this lack of standardization, the preoperative sign-in serves as a critical opportunity for 3-way discussion between the neurosurgeon, anesthesiologist, and neuromonitoring team regarding the necessity for and goals of IONM in the ensuing case. This analysis contains a review of commonly used IONM modalities including somatosensory evoked potentials, motor evoked potentials, spontaneous or free-running electromyography, triggered electromyography, and combined multimodal IONM. For each modality the methodology, interpretation, and reported sensitivity and specificity for neurological injury are addressed. This is followed by a discussion of important IONM-related issues to include in the preoperative checklist, including anesthetic protocol, warning criteria for possible neurological injury, and consideration of what steps to take in response to a positive alarm. The authors conclude with a cost-effectiveness analysis of IONM, and offer recommendations for IONM use during various forms of spine surgery, including both complex spine and minimally invasive procedures, as well as lower-risk spinal operations.

Restricted access

Timothy R. Smith, Rohan R. Lall, Rishi R. Lall, Isaac Josh Abecassis, Omar M. Arnaout, MaryAnne H. Marymont, Kristin R. Swanson and James P. Chandler

Object

Patients with systemic cancer and a single brain metastasis who undergo treatment with resection plus radiotherapy live longer and have a better quality of life than those treated with radiotherapy alone. Historically, whole-brain radiotherapy (WBRT) has been the mainstay of radiation therapy; however, it is associated with significant delayed neurocognitive sequelae. In this study, the authors looked at survival in patients with single and multiple intracranial metastases who had undergone surgery and adjuvant stereotactic radiosurgery (SRS) to the tumor bed and synchronous lesions.

Methods

The authors retrospectively reviewed the records from an 8-year period at a single institution for consecutive patients with brain metastases treated via complete resection of dominant lesions and adjuvant radiosurgery. The cohort was analyzed for time to local progression, synchronous lesion progression, new intracranial lesion development, systemic progression, and overall survival. The Kaplan-Meier method (stratified by age, sex, tumor histology, and number of intracranial lesions prior to surgery) was used to calculate both progression-free and overall survival. A Cox proportional-hazards regression model was also fitted with the number of intracranial lesions as the predictor and survival as the outcome controlling for disease severity, age, sex, and primary histology.

Results

The median overall follow-up among the 150-person cohort eligible for analysis was 17 months. Patients had an average age of 46.2 years (range 16–82 years), and 62.7% were female. The mean (± standard deviation) number of intracranial lesions per patient was 2.5 ± 2.3. The mean time between surgery and stereotactic radiosurgery (SRS) was 3.2 ± 4.1 weeks. Primary cancers included lung cancer (43.3%), breast cancer (21.3%), melanoma (10.0%), renal cell carcinoma (6.7%), and colon cancer (6.7%). The average number of isocenters per treated lesion was 7.6 ± 6.6, and the average treatment dose was 17.8 ± 2.8 Gy. One-year survival for patients in this cohort was 52%, and the 1-year local control rate was 77%. The median (±standard error) overall survival was 13.2 ± 1.9 months. There was no difference in survival between patients with a single lesion and those with multiple lesions (p = 0.319) after controlling for age, sex, and histology of primary tumor. Patients with primary breast histology had the greatest overall median survival (22.9 ± 6.2 months); patients with colorectal cancer had the shortest overall median survival (5.3 ± 1.8 months). The most common cause of death in this series was systemic progression (79%).

Conclusions

These results confirm that 1-year survival for patients with multiple intracranial metastases treated with resection followed by SRS to both the tumor bed and synchronous lesions is similar to established outcomes for patients with a single intracranial metastasis.

Free access

Albert P. Wong, Rishi R. Lall, Nader S. Dahdaleh, Cort D. Lawton, Zachary A. Smith, Ricky H. Wong, Michael J. Harvey, Sandi Lam, Tyler R. Koski and Richard G. Fessler

OBJECT

Patients with symptomatic intradural-extramedullary (ID-EM) tumors may be successfully treated with resection of the lesion and decompression of associated neural structures. Studies of patients undergoing open resection of these tumors have reported high rates of gross-total resection (GTR) with minimal long-term neurological deficit. Case reports and small case series have suggested that these patients may be successfully treated with minimally invasive surgery (MIS). These studies have been limited by small patient populations. Moreover, there are no studies directly comparing perioperative outcomes between patients treated with open resection and MIS. The objective of this study was to compare perioperative outcomes in patients with ID-EM tumors treated using open resection or MIS.

METHODS

A retrospective review was performed using data collected from 45 consecutive patients treated by open resection or MIS for ID-EM spine tumors. These patients were treated over a 9-year period between April 2003 and October 2012 at Northwestern University and the University of Chicago. Statistical analysis was performed to compare perioperative outcomes between the two groups.

RESULTS

Of the 45 patients in the study, 27 were treated with the MIS approach and 18 were treated with the open approach. Operative time was similar between the two groups: 256.3 minutes in the MIS group versus 241.1 minutes in the open group (p = 0.55). Estimated blood loss was significantly lower in the MIS group (133.7 ml) compared with the open group (558.8 ml) (p < 0.01). A GTR was achieved in 94.4% of the open cases and 92.6% of the MIS cases (p = 0.81).

The mean hospital stay was significantly shorter in the MIS group (3.9 days) compared with the open group (6.1 days) (p < 0.01). There was no significant difference between the complication rates (p = 0.32) and reoperation rates (p = 0.33) between the two groups. Multivariate analysis demonstrated an increased rate of complications in cervical spine tumors (OR 15, p = 0.05).

CONCLUSIONS

Thoracolumbar ID-EM tumors may be safely and effectively treated with either the open approach or an MIS approach, with an equivalent rate of GTR, perioperative complication rate, and operative time. Patients treated with an MIS approach may benefit from a decrease in operative blood loss and shorter hospital stays.