Why does delay exist in the diagnosis of intradural spinal cord tumor despite the availability of MRI?

Automated Detection and Diagnosis of Spinal Schwannomas and Meningiomas Using Deep Learning and Magnetic Resonance Imaging

Spinal cord tumors are infrequently identified spinal diseases that are often difficult to diagnose even with magnetic resonance imaging (MRI) findings. To minimize the probability of overlooking these tumors and improve diagnostic accuracy, an automatic diagnostic system is needed. We aimed to develop an automated system for detecting and diagnosing spinal schwannomas and meningiomas based on deep learning using You Only Look Once (YOLO) version 4 and MRI. In this retrospective diagnostic accuracy study, the data of 50 patients with spinal schwannomas, 45 patients with meningiomas, and 100 control cases were reviewed, respectively. Sagittal T1-weighted (T1W) and T2-weighted (T2W) images were used for object detection, classification, training, and validation. The object detection and diagnosis system was developed using YOLO version 4. The accuracies of the proposed object detections based on T1W, T2W, and T1W + T2W images were 84.8%, 90.3%, and 93.8%, respectively. The accuracies of the object detection for two spine surgeons were 88.9% and 90.1%, respectively. The accuracies of the proposed diagnoses based on T1W, T2W, and T1W + T2W images were 76.4%, 83.3%, and 84.1%, respectively. The accuracies of the diagnosis for two spine surgeons were 77.4% and 76.1%, respectively. We demonstrated an accurate, automated detection and diagnosis of spinal schwannomas and meningiomas using the developed deep learning-based method based on MRI. This system could be valuable in supporting radiological diagnosis of spinal schwannomas and meningioma, with a potential of reducing the radiologist's overall workload.

Lumbar Schwannoma as a Rare Cause of Radiculopathy in the Chiropractic Office: A Case Report

Cauda equina tumors are rare, slow-growing, and typically benign. These tumors present with low back pain resembling disc displacement with radiculopathy and thus may go undiagnosed for months. A 52-year-old, otherwise healthy man presented to a chiropractor with a one-year history of worsening low back pain radiating to the right lower extremity, rated an 8/10 in severity and aggravated by recumbency. Previously, his primary care physician had ordered radiographs revealing mild lumbar degenerative changes, prescribed a non-steroidal anti-inflammatory medication, and referred him to an orthopedist and physical therapist. There had been no change in symptoms. Upon examination by the chiropractor, the patient had neurologic deficits, and due to progressive worsening, the chiropractor recommended magnetic resonance imaging (MRI), which the patient deferred due to cost. The chiropractor initiated a trial of care, with initial success; however, the patient's symptoms recurred, and he consented to an MRI. MRI revealed an intradural extramedullary lumbar tumor, and the chiropractor referred the patient to an oncologist, who referred the patient to a neurosurgeon. The neurosurgeon surgically removed the mass, with a biopsy confirming a schwannoma. The patient had significantly improved six weeks after surgery. This case highlights a patient with chronic low back pain for whom a chiropractor identified a cauda equina tumor and referred him for further evaluation and surgery. Clinicians should consider night pain and persistent symptoms, despite conservative care, as red flags warranting further investigation in those with low back pain. Providers should refer for neurosurgical evaluation when clinical and radiological findings suggest a cauda equina tumor.

Automated Detection of Spinal Schwannomas Utilizing Deep Learning Based on Object Detection From Magnetic Resonance Imaging

STUDY DESIGN

A retrospective analysis of magnetic resonance imaging (MRI) was conducted.

OBJECTIVE

This study aims to develop an automated system for the detection of spinal schwannoma, by employing deep learning based on object detection from MRI. The performance of the proposed system was verified to compare the performances of spine surgeons.

SUMMARY OF BACKGROUND DATA

Several MRI scans were conducted for the diagnoses of patients suspected to suffer from spinal diseases. Typically, spinal diseases do not involve tumors on the spinal cord, although a few tumors may exist at the unexpectable level or without symptom by chance. It is difficult to recognize these tumors; in some cases, these tumors may be overlooked. Hence, a deep learning approach based on object detection can minimize the probability of overlooking these tumors.

METHODS

Data from 50 patients with spinal schwannoma who had undergone MRI were retrospectively reviewed. Sagittal T1- and T2-weighted magnetic resonance imaging (T1WI and T2WI) were used in the object detection training and for validation. You Only Look Once version3 was used to develop the object detection system, and its accuracy was calculated. The performance of the proposed system was compared to that of two doctors.

RESULTS

The accuracies of the proposed object detection based on T1W1, T2W1, and both T1W1 and T2W1 were 80.3%, 91.0%, and 93.5%, respectively. The accuracies of the doctors were 90.2% and 89.3%.

CONCLUSION

Automated object detection of spinal schwannoma was achieved. The proposed system yielded a high accuracy that was comparable to that of the doctors.Level of Evidence: 4.

Intraoperative Neuromonitoring in Patients with Intradural Extramedullary Spinal Cord Tumor: A Single-Center Case Series

BACKGROUND

Intradural extramedullary spinal cord tumors (ID-EMSCT) make up 40% of all spinal neoplasms. Resection of these tumors is mostly conducted using intraoperative neuromonitoring (IONM). However, the literature shows heterogenous data on its added value for ID-EMSCT. The aim of this study is to define sensitivity and specificity of IONM in ID-EMSCT resection and to study possible correlations between preoperative, intraoperative, and postoperative variables and neurologic outcomes after ID-EMSCT resection.

METHODS

Data of patients undergoing ID-EMSCT surgeries with IONM from January 2012 until July 2019 were examined. Using neurologic status 6 weeks and 1 year postoperatively, sensitivity and specificity for IONM were calculated. IONM test results and neurologic outcomes were paired to preoperative, intraoperative, and postoperative parameters.

RESULTS

Data of 78 patients were analyzed. 6 weeks postoperatively, 14.10% of patients had worse neurologic status, decreasing to 9.84% 1 year postoperatively. Multimodal IONM showed a sensitivity of 0.73 (95% confidence interval [CI], 0.39-0.94) and a specificity of 0.78 (95% CI, 0.66-0.87) after 6 weeks, and a sensitivity of 1.00 (95% CI, 0.54-1.00) and a specificity of 0.71 (95% CI, 0.57-0.82) after 1 year.

CONCLUSIONS

IONM yielded high to perfect sensitivity and high specificity. However, IONM signals did not always determine the extent of resection, and false-positive results did not always result in incomplete tumor resections, because of surgeons overruling IONM. Therefore, IONM cannot fully replace clinical judgment and other perioperative information.

Two Cases with Incidental Finding of Large Asymptomatic Intradural Lumbar Tumors

Benign lumbar intradural tumors are statistically uncommon and usually present with complaints of back pain with or without radicular neurological complaints. This report involves two separate patients that were found incidentally to have large intradural tumors without any neurologic complaints. In both cases the tumors were discovered when having magnetic resonance imaging (MRI) after minor auto accidents. Neither patient had any pre-existing lumbar or neurologic complaints. The report will review the different regions and types of incidental findings commonly seen on lumbar MRI scans and the need for close follow-up in patients with incidental lesions such as tumors.

Inexplicable Abdominal Pain due to Thoracic Spinal Cord Tumor

Chronic, refractory abdominal pain without a metabolic or structural gastroenterological etiology can be challenging for diagnosis and management. Even though it is rare, it has been reported that such a recurrent abdominal pain associated with radicular pattern can be derived from structural neurologic lesion like spinal cord tumor. We experienced an unusual case of chronic recurrent abdominal pain that lasted for two years without definite neurologic deficits in a patient, who has been harboring thoracic spinal cord tumor. During an extensive gastroenterological workup for the abdominal pain, the spinal cord tumor had been found and was resected through surgery. Since then, the inexplicable pain sustained over a long period of time eventually resolved. This case highlights the importance of taking into consideration the possibility of spinal cord tumor in differential diagnosis when a patient complains of chronic and recurrent abdominal pain without other medical abnormalities.

Delay in diagnosis of primary intradural spinal cord tumors

Background:

It has been our impression in recent years that there is a significant delay in diagnosis (DID) of patients in Israel harboring intradural spinal cord tumors (IDSCTs). DID can lead to irreversible deficits and unnecessary suffering. Our goal was to identify the incidence and the specific reasons for DID of IDSCTs in patients operated upon at our institution.

Methods:

A retrospective record review, with additional telephone survey, of 101 patients operated upon at our institute between the years 1996 and 2009 was conducted. The patients who were not diagnosed locally and those who were diagnosed during routine spinal imaging studies as part of their basic disease check-up were excluded. Accordingly, neurofibromatosis and medical tourist patients were excluded.

Results:

The clinical presentation of IDSCTs in our study was similar to the descriptions given in previous reports. The average age was 41.9 ± 23.3 years. Most tumors were ependymomas, astrocytomas, and schwannomas. The most common symptoms were motor or sensory disturbance, back pain, walking disturbance, and sphincter control deficit. The median time to diagnosis was 12.0 ± 37.0 months (range 3 days to 20 years). We found DID in 82.2% of the cases. 62.4% of the cases were defined as “unreasonable delay.” The most common reasons for DID were “classical symptoms with a wrong diagnosis” and “delayed imaging.”

Conclusions:

Based on the results of this study, the incidence of unreasonable delays in diagnosis of primary IDSCTs in Israel is very high. In order to shorten the time to diagnosis, primary and secondary care physicians need to increase their awareness of symptoms that may be associated with these lesions. We hereby offer feedback for care providers, relevant to the diagnostic workup of these patients. Such a feedback must be delivered by neurosurgeons to the community they are serving.