Analysis of factors delaying the surgical treatment of patients with neurological deficits in the course of spinal metastatic disease

Development of a natural language processing algorithm for the detection of spinal metastasis based on magnetic resonance imaging reports

Background

Metastasis to the spinal column is a common complication of malignancy, potentially causing pain and neurologic injury. An automated system to identify and refer patients with spinal metastases can help overcome barriers to timely treatment. We describe the training, optimization and validation of a natural language processing algorithm to identify the presence of vertebral metastasis and metastatic epidural cord compression (MECC) from radiology reports of spinal MRIs.

Methods

Reports from patients with spine MRI studies performed between January 1, 2008 and April 14, 2019 were reviewed by a team of radiologists to assess for the presence of cancer and generate a labeled dataset for model training. Using regular expression, impression sections were extracted from the reports and converted to all lower-case letters with all nonalphabetic characters removed. The reports were then tokenized and vectorized using the doc2vec algorithm. These were then used to train a neural network to predict the likelihood of spinal tumor or MECC. For each report, the model provided a number from 0 to 1 corresponding to its impression. We then obtained 111 MRI reports from outside the test set, 92 manually labeled negative and 19 with MECC to test the model's performance.

Results

About 37,579 radiology reports were reviewed. About 36,676 were labeled negative, and 903 with MECC. We chose a cutoff of 0.02 as a positive result to optimize for a low false negative rate. At this threshold we found a 100% sensitivity rate with a low false positive rate of 2.2%.

Conclusions

The NLP model described predicts the presence of spinal tumor and MECC in spine MRI reports with high accuracy. We plan to implement the algorithm into our EMR to allow for faster referral of these patients to appropriate specialists, allowing for reduced morbidity and increased survival.

Systematic Review and Meta-Analysis on Optimal Timing of Surgery for Acute Symptomatic Metastatic Spinal Cord Compression

Introduction: Symptomatic acute metastatic spinal epidural cord compression (MSCC) is an emergency that requires multimodal attention. However, there is no clear consensus on the appropriate timing for surgery. Therefore, to address this issue, we conducted a systematic review and meta-analysis of the literature to evaluate the outcomes of different surgery timings. Methods: We searched multiple databases for studies involving adult patients suffering from symptomatic MSCC who underwent decompression with or without fixation. We analyzed the data by stratifying them based on timing as emergent (≤24 h vs. >24 h) and urgent (≤48 h vs. >48 h). The analysis also considered adverse postoperative medical and surgical events. The rates of improved outcomes and adverse events were pooled through a random-effects meta-analysis. Results: We analyzed seven studies involving 538 patients and discovered that 83.0% (95% CI 59.0-98.2%) of those who underwent urgent decompression showed an improvement of ≥1 point in strength scores. Adverse events were reported in 21% (95% CI 1.8-51.4%) of cases. Patients who underwent emergent surgery had a 41.3% (95% CI 20.4-63.3%) improvement rate but a complication rate of 25.5% (95% CI 15.9-36.3%). Patients who underwent surgery after 48 h showed 36.8% (95% CI 12.2-65.4%) and 28.6% (95% CI 19.5-38.8%) complication rates, respectively. Conclusion: Our study highlights that a 48 h window may be the safest and most beneficial for patients presenting with acute MSCC and a life expectancy of over three months.

Preoperative Performance Status Threshold for Favorable Surgical Outcome in Metastatic Spine Disease

BACKGROUND AND OBJECTIVES

Surgical treatment is an integral component of multimodality management of metastatic spine disease but must be balanced against the risk of surgery-related morbidity and mortality, making tailored surgical counseling a clinical challenge. The aim of this study was to investigate the potential predictive value of the preoperative performance status for surgical outcome in patients with spinal metastases.

METHODS

Performance status was determined using the Karnofsky Performance Scale (KPS), and surgical outcome was classified as "favorable" or "unfavorable" based on postoperative changes in neurological function and perioperative complications. The correlation between preoperative performance status and surgical outcome was assessed to determine a KPS-related performance threshold.

RESULTS

A total of 463 patients were included. The mean age was 63 years (range: 22-87), and the mean preoperative KPS was 70 (range: 30-100). Analysis of clinical outcome in relation to the preoperative performance status revealed a KPS threshold between 40% and 50% with a relative risk of an unfavorable outcome of 65.7% in KPS ≤40% compared with the relative chance for a favorable outcome of 77.1% in KPS ≥50%. Accordingly, we found significantly higher rates of preserved or restored ambulatory function in KPS ≥50% (85.7%) than in KPS ≤40% (48.6%; P < .001) as opposed to a significantly higher risk of perioperative mortality in KPS ≤40% (11.4%) than in KPS ≥50% (2.1%, P = .012).

CONCLUSION

Our results underline the predictive value of the KPS in metastatic spine patients for counseling and decision-making. The study suggests an overall clinical benefit of surgical treatment of spinal metastases in patients with a preoperative KPS score ≥50%, while a high risk of unfavorable outcome outweighing the potential clinical benefit from surgery is encountered in patients with a KPS score ≤40%.

Enhancing the referral process for surgical management of spinal metastases: insights from a 12-year, bi-institutional study of 533 patients

Introduction

Delayed surgical management of spinal metastases (SMs) can have detrimental effects on patient survival and quality of life, leading to pain and potential neurological impairment. This study aimed to assess the impact of delayed referral for SMs on clinical outcomes by analyzing patients managed in emergency situations.

Methods

We retrospectively reviewed the data of all patients admitted on either emergency or elective basis who underwent surgery for the treatment of neoplastic spine lesions at our two institutions (tertiary referral neurosurgical units) between January 2008 and December 2019.

Results

We analyzed 210 elective (EGp) and 323 emergency patients (UGp); emergencies increased significantly over the 12-year period, with a Friday peak (39.3%) and frequent neurological impairment (61.6% vs. 20%). Among the UGp patients, 186 (7.5%) had a previously monitored primitive cancer, including 102 (31.6%) with known SMs. On admission, 71 of the 102 (69.9%) patients presented with neurological deficits. UGp patients were more likely to undergo a single decompression without fixation. Outcomes at the 3-month follow-up were significantly worse for UGp patients ([very] poor, 29.2 vs. 13.8%), and the median overall survival for UGp patients was statistically lower. Risk factors for patients with SM undergoing emergency management included short delay between onset of symptoms and first contact with a spine surgeon, and an initial motor deficit.

Conclusion

Many patients with previously identified metastases, including those with neurological deficits, are urgently referred. Optimization is needed in the oncology pathway, and all stakeholders must be made aware of the factors contributing to the improvement in the clinical and radiological identification of potential complications affecting patient survival and quality of life.

Time to Surgical Treatment for Metastatic Spinal Disease: Identification of Delay Intervals

STUDY DESIGN

Retrospective cohort study.

OBJECTIVES

Minimizing delays in referral, diagnosis and treatment of patients with symptomatic spinal metastases is important for optimal treatment outcomes. The primary objective of this study was to investigate several forms of delay from the onset of symptoms until surgical treatment of spinal metastases for patients with and without a known preexisting known malignancy.

METHODS

All patients receiving surgical treatment for spinal metastases in a single tertiary spine center were identified. Referral patterns were reconstructed and the total delay was divided into 4 categories: patient delay (onset of symptoms until medical consultation), diagnostic delay (medical consultation until diagnosis), referral delay (diagnosis until referral to spine surgeon) and treatment delay (referral spine to surgeon until treatment). These intervals were compared between patients with and without a known preexisting malignancy.

RESULTS

The median total delay was 99 days, patient delay 19 days, diagnostic delay 21,5 days, referral delay 7 days, treatment delay 8 days and diagnosis and treatment delay combined 18,5 days. No difference in total delay was observed between patients with and without a known preexisting malignancy. Total delay was not significantly associated with patient age, sex, oncological history, tumor prognosis and spinal level of the tumor.

CONCLUSIONS

Patients with symptomatic spinal metastases experience considerable delays, even after metastatic spinal disease has been diagnosed, regardless of a preexisting malignancy. By identifying and eliminating the causes of these delays, diagnosis, referral and treatment may be expedited leading to improved patient outcome.

Deep learning assessment compared to radiologist reporting for metastatic spinal cord compression on CT

Introduction

Metastatic spinal cord compression (MSCC) is a disastrous complication of advanced malignancy. A deep learning (DL) algorithm for MSCC classification on CT could expedite timely diagnosis. In this study, we externally test a DL algorithm for MSCC classification on CT and compare with radiologist assessment.

Methods

Retrospective collection of CT and corresponding MRI from patients with suspected MSCC was conducted from September 2007 to September 2020. Exclusion criteria were scans with instrumentation, no intravenous contrast, motion artefacts and non-thoracic coverage. Internal CT dataset split was 84% for training/validation and 16% for testing. An external test set was also utilised. Internal training/validation sets were labelled by radiologists with spine imaging specialization (6 and 11-years post-board certification) and were used to further develop a DL algorithm for MSCC classification. The spine imaging specialist (11-years expertise) labelled the test sets (reference standard). For evaluation of DL algorithm performance, internal and external test data were independently reviewed by four radiologists: two spine specialists (Rad1 and Rad2, 7 and 5-years post-board certification, respectively) and two oncological imaging specialists (Rad3 and Rad4, 3 and 5-years post-board certification, respectively). DL model performance was also compared against the CT report issued by the radiologist in a real clinical setting. Inter-rater agreement (Gwet's kappa) and sensitivity/specificity/AUCs were calculated.

Results

Overall, 420 CT scans were evaluated (225 patients, mean age=60 ± 11.9[SD]); 354(84%) CTs for training/validation and 66(16%) CTs for internal testing. The DL algorithm showed high inter-rater agreement for three-class MSCC grading with kappas of 0.872 (p<0.001) and 0.844 (p<0.001) on internal and external testing, respectively. On internal testing DL algorithm inter-rater agreement (κ=0.872) was superior to Rad 2 (κ=0.795) and Rad 3 (κ=0.724) (both p<0.001). DL algorithm kappa of 0.844 on external testing was superior to Rad 3 (κ=0.721) (p<0.001). CT report classification of high-grade MSCC disease was poor with only slight inter-rater agreement (κ=0.027) and low sensitivity (44.0), relative to the DL algorithm with almost-perfect inter-rater agreement (κ=0.813) and high sensitivity (94.0) (p<0.001).

Conclusion

Deep learning algorithm for metastatic spinal cord compression on CT showed superior performance to the CT report issued by experienced radiologists and could aid earlier diagnosis.

Diagnostic Accuracy of CT for Metastatic Epidural Spinal Cord Compression

Background: Early diagnosis of metastatic epidural spinal cord compression (MESCC) is vital to expedite therapy and prevent paralysis. Staging CT is performed routinely in cancer patients and presents an opportunity for earlier diagnosis. Methods: This retrospective study included 123 CT scans from 101 patients who underwent spine MRI within 30 days, excluding 549 CT scans from 216 patients due to CT performed post-MRI, non-contrast CT, or a gap greater than 30 days between modalities. Reference standard MESCC gradings on CT were provided in consensus via two spine radiologists (11 and 7 years of experience) analyzing the MRI scans. CT scans were labeled using the original reports and by three radiologists (3, 13, and 14 years of experience) using dedicated CT windowing. Results: For normal/none versus low/high-grade MESCC per CT scan, all radiologists demonstrated almost perfect agreement with kappa values ranging from 0.866 (95% CI 0.787–0.945) to 0.947 (95% CI 0.899–0.995), compared to slight agreement for the reports (kappa = 0.095, 95%CI −0.098–0.287). Radiologists also showed high sensitivities ranging from 91.51 (95% CI 84.49–96.04) to 98.11 (95% CI 93.35–99.77), compared to 44.34 (95% CI 34.69–54.31) for the reports. Conclusion: Dedicated radiologist review for MESCC on CT showed high interobserver agreement and sensitivity compared to the current standard of care.

Evaluation of prognostic preoperative factors in patients undergoing surgery for spinal metastases: Results in a consecutive series of 81 cases

Background:

Surgical treatment of spinal metastases should be tailored to provide pain control, neurological deficit improvement, and vertebral stability with low operative morbidity and mortality. The aim of this study was to analyze the predictive value of some preoperative factors on overall survival in patients undergoing surgery for spinal metastases.

Methods:

We retrospectively analyzed a consecutive series of 81 patients who underwent surgery for spinal metastases from 2015 and 2021 in the Clinic of Neurosurgery of Ancona (Italy). Data regarding patients’ baseline characteristics, preoperative Karnofsky Performance Status Score (KPS), and Frankel classification grading system, histology of primary tumor, Tokuhashi revised and Tomita scores, Spine Instability Neoplastic Score, and Epidural Spinal Cord Compression Classification were collected. We also evaluated the interval time between the diagnosis of the primary tumor and the onset of spinal metastasis, the type of surgery, the administration of adjuvant therapy, postoperative pain and Frankel grade, and complications after surgery. The relationship between patients’ overall survival and predictive preoperative factors was analyzed by the Kaplan–Meier method. For the univariate and multivariate analysis, the log-rank test and Cox regression model were used. P ≤ 0.05 was considered as statistically significant.

Results:

After surgery, the median survival time was 13 months. In our series, the histology of the primary tumor (P < 0.001), the Tomita (P < 0.001) and the Tokuhashi revised scores (P < 0.001), the preoperative KPS (P < 0.001), the adjuvant therapy (P < 0.001), the postoperative Frankel grade (P < 0.001), and the postoperative pain improvement (P < 0.001) were significantly related to overall survival in the univariate analysis. In the multivariate analysis, the Tomita (P < 0.001), Tokuhashi revised scores (P < 0.001), and the adjuvant therapy were confirmed as independent prognostic factors.

Conclusion:

These data suggest that patients with limited extension of primitive tumor and responsive to the adjuvant therapy are the best candidates for surgery with better outcome.

Deep Learning Model for Grading Metastatic Epidural Spinal Cord Compression on Staging CT

Background: Metastatic epidural spinal cord compression (MESCC) is a disastrous complication of advanced malignancy. Deep learning (DL) models for automatic MESCC classification on staging CT were developed to aid earlier diagnosis. Methods: This retrospective study included 444 CT staging studies from 185 patients with suspected MESCC who underwent MRI spine studies within 60 days of the CT studies. The DL model training/validation dataset consisted of 316/358 (88%) and the test set of 42/358 (12%) CT studies. Training/validation and test datasets were labeled in consensus by two subspecialized radiologists (6 and 11-years-experience) using the MRI studies as the reference standard. Test sets were labeled by the developed DL models and four radiologists (2−7 years of experience) for comparison. Results: DL models showed almost-perfect interobserver agreement for classification of CT spine images into normal, low, and high-grade MESCC, with kappas ranging from 0.873−0.911 (p < 0.001). The DL models (lowest κ = 0.873, 95% CI 0.858−0.887) also showed superior interobserver agreement compared to two of the four radiologists for three-class classification, including a specialist (κ = 0.820, 95% CI 0.803−0.837) and general radiologist (κ = 0.726, 95% CI 0.706−0.747), both p < 0.001. Conclusion: DL models for the MESCC classification on a CT showed comparable to superior interobserver agreement to radiologists and could be used to aid earlier diagnosis.

The "Friday peak" in surgical referrals for spinal metastases: lessons not learned. A retrospective analysis of 201 consecutive cases at a tertiary center

BACKGROUND

Spinal cord compression and fracture are potential complications of spine metastasis (SM). Rapid management by an expert team can reduce these adverse developments. Delays in seeking therapeutic advices, which lead to the need for sub-optimal emergency procedures, were already demonstrated nearly 20 years ago. We aimed to analyze the current weak points of referrals for vertebral metastasis so as to improve the care pathways.

METHODS

We retrospectively reviewed the data of all patients admitted on an emergency or elective basis who underwent palliative surgery for the treatment of neoplastic spine lesions in our institution (tertiary referral neurosurgical unit) between January 2009 and December 2016.

RESULTS

This retrospective study included 201 patients, 121 men and 80 women (mean age 65.1 years ± 10.9). Cancer was known for 59.7% of cases. Patients were neurologically asymptomatic in 52.7% of cases (Frankel E), and 123 (60.7%) were hospitalized for emergency reasons, including 51 (41.5% of emergencies) on a Friday (p < 0.0001). A significant increase in emergencies occurred over the studied period (p = 0.0027). The "emergency" group had significantly unfavorable results in terms of neurological status (p < 0.001), the occurrence of complications (p = 0.04), the duration of hospitalization (p = 0.02), and the clinical evolution (p = 0.04). Among 123 patients hospitalized for emergency reasons, 65 (52.8%) had known cancers, of which 33 had an identified SM, including 22 with neurological deficits (Frankel A-D), without prior surgical assessment (17.8% of emergencies).

CONCLUSION

Too many patients with previously identified metastases are referred for emergency reasons, including with a neurological deficit. Optimizing upstream pathways and referrals is imperative for improving the management of these patients. Involving a spine surgeon at the slightest symptom or an abnormal image is critical for defining the best treatment upstream. The use of telemedicine and the development of dedicated tumor boards are ways of improving this involvement.