Can we predict neurological complications in patients with metastatic spinal tumors?

doi:10.21203/rs.3.rs-2063122/v1

Download PDF

Research Article

Can we predict neurological complications in patients with metastatic spinal tumors?

https://doi.org/10.21203/rs.3.rs-2063122/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Purpose

This study aimed to identify predictors of the occurrence and severity of neurological symptoms in patients with metastatic spinal tumor who still had no neurological symptoms.

Methods

We retrospectively examined patients with metastatic spinal tumor (regardless of tumor type) who visited the Department of Rehabilitation Medicine between 2017 and 2021. Using the patients’ magnetic resonance imaging and clinical data, we investigated whether epidural spinal cord compression (ESCC) scale, spinal instability neoplastic score (SINS), and pain character were related to the occurrence and severity of neurological symptoms within 1 year.

Results

Of the 70 included patients, 40 showed neurological symptoms, and 30 did not. Multivariable analysis showed that ESCC (2 or 3) (P = 0.002), SINS (P = 0.007), and radiating pain (P = 0.015) were associated with neurological complications. None of the factors was associated with neurological symptom severity. Notably, neurological symptoms progressed rapidly in patients with high ESCC (P = 0.003).

Conclusion

ESCC, SINS, and radiating pain were related to the occurrence of neurological symptoms, but not to severity. Only ESCC was associated with the rapid progression of neurological symptoms. This retrospective study provides useful insights in predicting a patient's functional prognosis and determining management using radiology and clinical data.

metastatic spinal tumor

epidural spinal cord compression

spinal instability neoplastic score

Metastatic spinal tumors are common, of which 5–14% of cases cause spinal cord compression [1]. With the development of treatment techniques such as stereotactic radiosurgery and minimal invasive surgery, research interest has shifted from survival rates to patients’ quality of life [2]. Spinal cord compression, in particular, has devastating consequences for patients. The overall survival and quality of life are reduced in patients with neurological complications [3, 4]. However, the impact of neurological complication on patients who have not yet developed neurological symptoms cannot be overemphasized. In these patients, treatment decision making is difficult [5]. The choice of treatment is important in patients who have not yet developed neurological complication. Recently, decision making using the NOMPRS framework, which incorporates six fundamental assessments (neurologic, oncologic, medical, pain, rehabilitation, and support status) has been suggested [6].

In the case of instability or in the case of spinal cord compression with neurological symptoms, surgery is commonly performed [7, 8]. Stability is measured by spinal instability neoplastic score (SINS), which uses radiologic and clinical data in patients with metastatic spinal disease to provide clinicians with information necessary for treatment selection [9, 10]. Patients with low-grade SINS have a good prognosis after radiotherapy treatment, whereas those with high-grade SINS have a higher risk for radiotherapy failure [11].

In addition, the degree of spinal cord compression can be evaluated by epidural spinal cord compression (ESCC) scale, which assesses the degree of impingement of the cerebrospinal fluid (CSF) space and spinal cord compression through magnetic resonance imaging (MRI) [12].

Both scoring systems have been reliable for treatment decision making. However, reports investigating the relationship between neurological symptom occurrence and the two scoring systems in patients who have not yet developed neurological complication are still few.

Clinically, 90–95% of patients with metastatic spinal tumor experience pain as the first presenting symptom [13]. Various types of pain can be caused by various reasons, including vertebral body fracture, pressure on the periosteum, and nerve root compression [14]. The clinical manifestation of pain can be a key sign for predicting patient’s functional prognosis.

Thus, this study aimed to determine whether SINS, ESCC, and clinical factors are related to the occurrence of neurological complication within 1 year and whether they are also related to severity.

Study design and participants

We retrospectively reviewed the medical records of patients with metastatic spinal tumor who visited the Department of Rehabilitation Medicine in Asan Medical Center between January 1, 2017 and May 31, 2021. These patients were divided into two groups: (1) group with newly developed neurological symptoms and (2) group without neurological symptoms at that time of visit. We compared the past records within the previous year from the time of visit between the two groups.

We collected the following clinical and radiologic data: age, sex, cancer type, treatment type, neurologic severity scale according to the American Spinal Injury Association (ASIA) classification, pain characteristics (localized back pain, mechanical pain, and radicular pain), SINS (Appendix 1), and ESCC (Appendix 2).

The inclusion criteria were (1) age of more than 18 years, (2) radiologically confirmed spinal metastasis regardless of cancer type, and (3) available whole-spine MRI within 1 year at the time of hospital visit. The exclusion criteria were (1) brain metastasis, (2) whole-spine MRI taken after a neurological complication has already occurred, and (3) impossible to score an ESCC (e.g., the level of spinal metastasis below L2 or no axial cut in whole-spine MRI).

Clinical evaluation

Pain was categorized into three: localized back pain, radicular pain, and mechanical back pain. Radicular pain was defined as a lancinating quality of pain traveling into the limb along the certain dermatome [15]. Movement-related pain with specific spinal level involvement indicated mechanical pain [7].

The severity of neurologic deficit in the neurological complication group was evaluated by the ASIA classification. This classification consists of four grades: (A) complete (no sensory or motor function is preserved in the sacral segments S4–S5); (B) incomplete (sensory but not motor function is preserved below the neurological level, and includes the sacral segments S4–S5); (C) incomplete (motor function is preserved below the neurological level, and more than half of key muscles below the neurological level have a muscle grade < 3); (D) incomplete (motor function is preserved below the neurological level, and at least half of key muscles below the neurological level have a muscle grade ≥ 3) [16]. In this study, the neurological complication group was further divided into the mild group (ASIA D) and the moderate to severe group (ASIA A, B, and C).

SINS and ESCC

Stability was evaluated using SINS, which included six parameters: location, pain, alignment, osteolysis, vertebral body collapse, and posterior element involvement. Patients were categorized into three according to the total scores: stable (0–6), potentially unstable (7–12), and unstable spines (13–18) [9]. We analyzed it in two ways: analysis using the numerical value itself and analysis using categorical variables separately.

The extent of epidural tumor invasion was evaluated using the ESCC scale, which is also known as the Bilsky scale [12]. This scale is graded 0–3 on the basis of the MRI (0, bone involvement only; 1, epidural impingement; 2, spinal cord compression with visible CSF; and 3, spinal cord compression without visible CSF). Both the SINS and ESCC scores were assessed by a radiologist.

The relationship between neurological complication occurrence and factors such as ESCC, numerical SINS, categorized SINS, and pain character was analyzed. The period from MRI performance to the onset of neurological symptoms was also analyzed. Patients with neurological complications were further divided into two subgroups: those with exacerbated SINS score and those without. Thereafter, we evaluated which factor of the initial SINS category was associated with SINS deterioration.

Statistical analysis

We express baseline characteristics as mean and standard deviation (SD) for continuous variables and percentage for categorical variables. Comparisons of neurological complication occurrence and subgroup analysis were conducted using Fisher’s exact test for the categorical variables and two-sample t-test for the continuous variables. For the comparisons of neurological complication severity, we employed Mantel–Haenszel correlation and one-way ANOVA for the categorical and continuous variables, respectively.

In the multivariate analysis, we examined the variables that showed a statistically significant difference in the univariate analysis (P < 0.05). A logistic regression model was built to predict neurological complications. The correlation with the onset of neurological symptoms according to the ESCC and SINS was analyzed by the Jonckheere–Terpstra test. A P-value of less than 0.05 was considered statistically significant. We used SPSS Statistics 25.0 for Windows (IBM Corp, Armonk, NY) for all statistical analyses.

We found 296 patients with metastatic spinal tumor who visited the Department of Rehabilitation Medicine in Asan Medical Center between January 1, 2017 and May 31, 2021. Among them, 226 were excluded because of the concomitant presence of brain metastases (n = 59), no spine MRI within 1 year (n = 114), ESCC scoring not possible (n = 32), and insufficient data (n = 21). ESCC measurement was impossible in those with lesions below L2 or without axial views. Thus, 70 patients were finally included, of which 40 had neurological complications and 30 had not. Table 1 describes the demographics, tumor type, and treatment type of the two groups. The most common histological diagnosis was breast cancer, followed by non–small-cell lung cancer. The “Others” type included the following: osteosarcoma, adenoid cystic cancer, neuroendocrine tumor, thymoma, cholangiocarcinoma, colorectal cancer, ureter cancer, lymphoma, and cervical cancer.

Neurological complication occurrence

Univariate analysis showed that ESCC (P = 0.004), categorized SINS (P = 0.017), and radicular pain (P = 0.024) were associated with neurological complication occurrence (Table 2). In the multivariate analysis, ESCC score 2 or 3, categorized SINS, and radicular pain were associated with neurological complication occurrence (Table 3). Although ESCC 1 showed no statistical significance in this study (P = 0.08), it had some marginal significance with neurological complication occurrence (OR = 3.46). Meanwhile, ESCC 2 and 3 demonstrated a high correlation with high odds ratio (OR = 12.60).

Neurological symptoms progressed rapidly in patients with high ESCC (P = 0.003) (Table 4). However, SINS score was not related to symptom progression.

Aggravation of stability of the spine

We identified 29 patients. Eleven patients were excluded because of missing SINS data at the time of neurological symptom occurrence. In particular, we could not measure the post-SINS of 9 patients because surgery was performed before the neurological symptoms occurred, and 2 patients because surgery was performed without MR imaging when neurological symptoms occurred. These 29 patients were divided into the SINS-worsened group (n = 15) and the nonexacerbated group (n = 14). Localized pain (P = 0.009) and bony lesion (P = 0.028) were associated with SINS aggravation (Table 6).

This study is the first to predict the occurrence of neurological symptoms by using ESCC, SINS, and clinical data in patients who had not yet developed neurological symptoms. This study also showed how neurological symptoms can occur in patients with spinal metastasis who have not yet manifested neurological symptoms. Neurological symptoms developed, mainly because of mechanical compression caused by the tumor or angulation caused by vertebral body collapse [7]. In this study, ESCC, a representative of mass effect, was associated with the occurrence and timing of neurological symptoms, but not with severity. SINS, which could predict angulation, was only related to the occurrence of neurological symptoms, but not to the onset and severity of these symptoms. In the subgroup whose stability deteriorated within the group with neurological symptoms, no patients actually deteriorated to the unstable category (SINS 13–18). Considering the multivariable analysis results, the contribution of ESCC to the occurrence of neurological symptoms may be greater than that of SINS. Therefore, predicting the occurrence of neurological symptoms with SINS alone is difficult, and ESCC must be considered as well.

As reported by Bilsky [17], ESCC, which has highly inter- and intra-rater reliability, has been a useful tool for treatment decision making [7, 12]. However, studies on whether ESCC can act as a predictor of the development of neurological symptoms are few. A previous study showed that a high-grade ESCC indicated a high risk for rapidly progressive neurological symptoms within 3 weeks [18]. But it did not target patients who have not yet developed neurological symptoms.

In our study, ESCC was not related to neurological complication severity, consistent with previous studies [18], This result suggests that the neurological symptoms are determined by various factors. The tumor growth rate is different for each tumor type, and factors such as vascular damage and mass effect may have had an effect [19]. Oshima et al. reported that neurological symptoms could be predicted according to the circumferential ratio of cord compression [20]. In fact, the choice of treatment, such as surgery, may be related to neurological symptom severity [21]. In this study, 4 out of 5 patients with unstable SINS underwent surgery early, and all showed mild neurological symptoms (ASIA D). Given that the sample size is small and the statistical relationship has not been revealed, further research considering treatment factors is needed..

As mentioned, SINS was analyzed in two ways. We found that SINS was not associated with neurological symptoms in continuous numbers but was associated with SINS as categorical variables. In one study, postoperative functional recovery was high in the potentially unstable and unstable groups when SINS was used as a categorical value [22]. Numerous studies have also confirmed the reliability of [23].

Regarding the strength of this study, both the clinical factors and radiologic findings were investigated. Pain was also classified into three categories, of which only radicular pain was associated with neurological symptoms. Radicular pain could be a typical symptom of instability because of neural foramen collapse [7].

Presently, brace application as a treatment for spinal metastatic tumor has not yet been confirmed to be effective, and no randomized control trials on brace effects have been conducted [24]. Nevertheless, it is widely used for stability or pain relief among clinicians. Identifying which factor exacerbates stability can be greatly beneficial to patient care. In this study, bony lesion and localized pain were associated with stability aggravation. Osteolytic bone metastases are more aggressive than osteoblastic metastases and are associated with a higher incidence of pathologic fractures [25]. These characteristics would have also affected SINS deterioration. Mechanical pain is a typical symptom that occurs when stability deteriorates. But in the present study, mechanical pain was irrelevant, whereas localized pain was highly relevant. One of the possible mechanisms is that pathologic fracture can be preceded by localized pain [26]. In addition, SINS could not be measured in six of our patients with mechanical pain because surgery was performed before the onset of symptoms. Hence, the effect of mechanical pain may not have been properly evaluated in this study. Therefore, when osteolytic lesion is suspected on the radiology along with the newly generated pain, the possibility of deterioration in stability should be carefully examined.

This study has certain limitations. First, treatments were not controlled between the two groups. In particular, the number of surgeries was largely different between the groups, probably because patients with high ESCC and SINS generally choose surgery as a treatment option. In addition, radiation therapy may have affected the frequency of complications because radiation-induced myelopathy is possible [27]. However, radiation-induced myelopathy is mainly a cumulative injury and occurs as a late onset. Therefore, the impact of radiation in this study may be minimal. Second, the period was not uniform. The time between the time of visit and spine MRI was limited to a period within 1 year, and the result may have been affected by the difference in each patient’s periods. Third, selection bias is possible. Given that this study was conducted on patients visiting the Department of Rehabilitation Medicine, the severity may have been higher than that of patients with general bone metastasis. However, this observation can be directly reflected in real-life clinical situations. Finally, given that this study is retrospective, important data such as pain information could be insufficient.

ESCC, SINS, and radiating pain are associated with the occurrence of neurological symptoms within 1 year but not with the severity of these symptoms. In addition, only ESCC was related to how quickly neurological symptoms developed. Therefore, these factors can be used as predictors of neurological symptom occurrence and can be helpful in managing patients without neurological symptoms.

Author contribution: Hee Tae Shin: Planning and development of the study and writing of the article.

Jae Yong Jeon: Planning of the study, critical review and approval of the final version.

Choong Guen Chee: Planning of the study, critical review and approval of the final version.

Cho Rong Bae: Critical review and approval of the final version.

Joon Hee Lee: Extracted the information and assisted in information collection and analysis.

Funding: No funding was received for conducting this study.

Data availability: Not applicable

Code availability: Not applicable

Ethics approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Consent to participate: Not applicable

Consent for publication: All authors contributed to the article and approved the submitted version

Competing Interests: The authors declare no competing interests.

Gilbert RW, Kim JH, Posner JB (1978) Epidural spinal cord compression from metastatic tumor: diagnosis and treatment. Ann Neurol 3:40–51. https://doi.org/10.1002/ana.410030107
Barzilai O, McLaughlin L, Amato MK, Reiner AS, Ogilvie SQ, Lis E, Yamada Y, Bilsky MH, Laufer I (2018) Minimal access surgery for spinal metastases: prospective evaluation of a treatment algorithm using patient-reported outcomes. World Neurosurg 120:e889-e901. https://doi.org/10.1016/j.wneu.2018.08.182
Prasad D, Schiff D (2005) Malignant spinal-cord compression. Lancet Oncol 6:15–24. https://doi.org/10.1016/S1470-2045(04)01709-7
Barzilai O, Versteeg AL, Goodwin CR, Sahgal A, Rhines LD, Sciubba DM, Schuster JM, Weber MH, Lazary A, Fehlings MG, Clarke MJ, Arnold PM, Boriani S, Bettegowda C, Gokaslan ZL, Fisher CG, Laufer I, AOSpine Knowledge Forum Tumor (2019) Association of neurologic deficits with surgical outcomes and health-related quality of life after treatment for metastatic epidural spinal cord compression. Cancer 125:4224–4231. https://doi.org/10.1002/cncr.32420
Barzilai O, Fisher CG, Bilsky MH (2018) State of the art treatment of spinal metastatic disease. Neurosurgery 82:757–769. https://doi.org/10.1093/neuros/nyx567
Bae CR, Gelvosa MN, Jeon JY (2022) Rehabilitation in advanced cancer patients with bone metastases and neural compromise: current status and future directions. Curr Oncol Rep 24:1023–1033. https://doi.org/10.1007/s11912-022-01229-9
Laufer I, Rubin DG, Lis E, Cox BW, Stubblefield MD, Yamada Y, Bilsky MH (2013) The NOMS framework: approach to the treatment of spinal metastatic tumors. Oncologist 18:744–751. https://doi.org/10.1634/theoncologist.2012-0293
Cofano F, Di Perna G, Zenga F, Ducati A, Baldassarre B, Ajello M, Marengo N, Ceroni L, Lanotte M, Garbossa D (2020) The neurology-stability-epidural compression assessment: a new score to establish the need for surgery in spinal metastases. Clin Neurol Neurosurg 195:105896. https://doi.org/10.1016/j.clineuro.2020.105896
Fisher CG, DiPaola CP, Ryken TC, Bilsky MH, Shaffrey CI, Berven SH, Harrop JS, Fehlings MG, Boriani S, Chou D, Schmidt MH, Polly DW, Biagini R, Burch S, Dekutoski MB, Ganju A, Gerszten PC, Gokaslan ZL, Groff MW, Liebsch NJ, Mendel E, Okuno SH, Patel S, Rhines LD, Rose PS, Sciubba DM, Sundaresan N, Tomita K, Varga PP, Vialle LR, Vrionis FD, Yamada Y, Fourney DR (2010) A novel classification system for spinal instability in neoplastic disease: an evidence-based approach and expert consensus from the Spine Oncology Study Group. Spine (Phila Pa 1976) 35:E1221-E1229. https://doi.org/10.1097/BRS.0b013e3181e16ae2
Dakson A, Leck E, Brandman DM, Christie SD (2020) The clinical utility of the Spinal Instability Neoplastic Score (SINS) system in spinal epidural metastases: a retrospective study. Spinal Cord 58:892–899. https://doi.org/10.1038/s41393-020-0432-8
Huisman M, van der Velden JM, van Vulpen M, van den Bosch MA, Chow E, Öner FC, Yee A, Verkooijen HM, Verlaan JJ (2014) Spinal instability as defined by the spinal instability neoplastic score is associated with radiotherapy failure in metastatic spinal disease. Spine J 14:2835–2840. https://doi.org/10.1016/j.spinee.2014.03.043
Bilsky MH, Laufer I, Fourney DR, Groff M, Schmidt MH, Varga PP, Vrionis FD, Yamada Y, Gerszten PC, Kuklo TR (2010) Reliability analysis of the epidural spinal cord compression scale. J Neurosurg Spine 13:324–328. https://doi.org/10.3171/2010.3.SPINE09459
Helweg-Larsen S, Sørensen PS, Kreiner S (2000) Prognostic factors in metastatic spinal cord compression: a prospective study using multivariate analysis of variables influencing survival and gait function in 153 patients. Int J Radiat Oncol Biol Phys 46:1163–1169. https://doi.org/10.1016/s0360-3016(99)00333-8
Bollen L, Dijkstra SPD, Bartels R, de Graeff A, Poelma DLH, Brouwer T, Algra PR, Kuijlen JMA, Minnema MC, Nijboer C, Rolf C, Sluis T, Terheggen M, van der Togt-van Leeuwen ACM, van der Linden YM, Taal W (2018) Clinical management of spinal metastases-The Dutch national guideline. Eur J Cancer 104:81–90. https://doi.org/10.1016/j.ejca.2018.08.028
Bogduk N (2009) On the definitions and physiology of back pain, referred pain, and radicular pain. Pain 147:17–19. https://doi.org/10.1016/j.pain.2009.08.020
Maynard FM, Jr., Bracken MB, Creasey G, Ditunno JF, Jr., Donovan WH, Ducker TB, Garber SL, Marino RJ, Stover SL, Tator CH, Waters RL, Wilberger JE, Young W (1997) International standards for neurological and functional classification of spinal cord injury. Spinal Cord 35:266–274. https://doi.org/10.1038/sj.sc.3100432
Bilsky MH, Boland PJ, Panageas KS, Woodruff JM, Brennan MF, Healey JH (2001) Intralesional resection of primary and metastatic sarcoma involving the spine: outcome analysis of 59 patients. Neurosurgery 49:1277–1287. https://doi.org/10.1097/00006123-200112000-00001
Uei H, Tokuhashi Y, Maseda M (2018) Analysis of the relationship between the epidural spinal cord compression (ESCC) scale and paralysis caused by metastatic spine tumors. Spine (Phila Pa 1976) 43:E448-E455. https://doi.org/10.1097/BRS.0000000000002378
Cole JS, Patchell RA (2008) Metastatic epidural spinal cord compression. Lancet Neurol 7:459–466. https://doi.org/10.1016/S1474-4422(08)70089-9
Oshima K, Hashimoto N, Sotobori T, Joyama S, Tanigami H, Nakanishi K, Araki N (2016) New magnetic resonance imaging features predictive for post-treatment ambulatory function: imaging analysis of metastatic spinal cord compression. Spine (Phila Pa 1976) 41:E422-E429. https://doi.org/10.1097/BRS.0000000000001308
Furstenberg CH, Wiedenhofer B, Gerner HJ, Putz C (2009) The effect of early surgical treatment on recovery in patients with metastatic compression of the spinal cord. J Bone Joint Surg Br 91:240–244. https://doi.org/10.1302/0301-620X.91B2.20894
Hussain I, Barzilai O, Reiner AS, DiStefano N, McLaughlin L, Ogilvie S, Bilsky M, Laufer I (2018) Patient-reported outcomes after surgical stabilization of spinal tumors: symptom-based validation of the Spinal Instability Neoplastic Score (SINS) and surgery. Spine J 18:261–267. https://doi.org/10.1016/j.spinee.2017.07.008
Fisher CG, Versteeg AL, Schouten R, Boriani S, Varga PP, Rhines LD, Heran MK, Kawahara N, Fourney D, Reynolds JJ, Fehlings MG, Gokaslan ZL (2014) Reliability of the spinal instability neoplastic scale among radiologists: an assessment of instability secondary to spinal metastases. AJR Am J Roentgenol 203:869–874. https://doi.org/10.2214/AJR.13.12269
National Collaborating Centre for Cancer (Great Britain) (2008) Metastatic spinal cord compression: diagnosis and management of patients at risk of or with metastatic spinal cord compression. National Collaborating Centre for Cancer Cardiff, UK
O'Sullivan GJ, Carty FL, Cronin CG (2015) Imaging of bone metastasis: an update. World J Radiol 7:202–211. https://doi.org/10.4329/wjr.v7.i8.202
Damron TA, Morgan H, Prakash D, Grant W, Aronowitz J, Heiner J (2003) Critical evaluation of Mirels' rating system for impending pathologic fractures. Clin Orthop Relat Res 415:S201-207. https://doi.org/10.1097/01.blo.0000093842.72468.73
Wong CS, Fehlings MG, Sahgal A (2015) Pathobiology of radiation myelopathy and strategies to mitigate injury. Spinal Cord 53:574–580. https://doi.org/10.1038/sc.2015.43

Table 1

Demographics of patients
		Neurological complication (+) (n = 40)	Neurological complication (−) (n = 30)
Sex (%)	Male	23 (57.5)	11 (36.6)
	Female	17 (42.5)	19 (63.4)
Age (mean (SD))		56.97 (9.95)	61.46 (10.32)
Cancer type (%)	Breast cancer	7 (17.5)	11 (33.6)
	Non-small-cell-lung cancer	7 (17.5)	8 (26.6)
	Prostate cancer	4 (10.0)	3 (10.0)
	Renal cell carcinoma	5 (12.5)	3 (10.0)
	Multiple myeloma	3 (7.5)	0 (0.0)
	Hepatocellular carcinoma	2 (5.0)	3 (10.0)
	Adrenal cancer	2 (5.0)	0 (0.0)
	Others	10 (17.5)	2 (6.6)
Treatment type (%)	Chemotherapy	4	10
	Radiation therapy	10	5
	Concurrent chemo-radiation therapy	17	13
	Surgery	9	2

Table 2

Univariable analysis of neurological complication occurrence
		Neurological complication(+) (n = 40)	Neurological complication(−) (n = 30)	P-value*
ESCC (%)	0	10 (25.0)	18 (60.0)	0.004*
	1	12 (30.0)	8 (26.7)
	2	10 (25.0)	4 (13.3)
	3	8 (20.0)	0 (0.0)
SINS (mean (SD))		8.30 (3.64)	7.77 (2.30)	0.484
SINS categorized (%)	Stable	16 (40.0)	7 (23.3)	0.017*
	Potentially unstable	19 (47.5)	23 (76.7)
	Unstable	5 (12.5)	0 (0.0)
Localized pain (%)	No	24 (60.0)	19 (63.3)	0.809
	Yes	16 (40.0)	11 (36.7)	0.809
Mechanical pain (%)	No	27 (67.5)	24 (80.0)	0.287
	Yes	13 (32.5)	6 (20.0)	0.287
Radicular pain (%)	No	21 (52.5)	24 (80.0)	0.024*
	Yes	19 (47.5)	6 (20.0)	0.024*
For categorical variables, Fisher’s exact test was used, and n (%) was reported. For continuous variables, two-sample t-test was used, and mean (SD) was reported. ESCC, epidural spinal cord compression scale; SD, standard deviation; SINS, spinal instability neoplastic score.

Table 3

Multivariate logistic regression analysis of neurological complication occurrence
		OR (95% CI)	P-value
ESCC	0	Reference
	1	3.46 (0.86, 13.95)	0.080
	2 or 3	12.60 (2.59, 61.27)	0.002*
Categorized SINS	Stable	Reference
	Potentially unstable or unstable	1.13(1.03, 1.57)	0.007*
Radiating pain	Yes	Reference
	No	5.56 (1.40, 22.02)	0.015*
CI, confidence interval; ESCC, epidural spinal cord compression scale; OR, odds ratio; SINS, spinal instability neoplastic score.

Table 4

Period until neurological symptoms develop according to ESCC and SINS among patients with neurologic complications
		Period until neurological symptoms develop (mean (SD))	P-value
ESCC	0 (n = 10)	6.80 (4.13)	0.003*
	1 (n = 12)	4.60 (3.36)
	2 or 3 (n = 18)	3.11 (3.17)
Categorized SINS	Stable (n = 16)	4.62 (3.75)	0.575
	Potentially unstable (n = 19)	4.31 (3.46)
	Unstable (n = 5)	4.80 (5.21)
Jonckheere–Terpstra test was used, and mean (SD) was reported. ESCC, epidural spinal cord compression scale; SD, standard deviation; SINS, spinal instability neoplastic score.

Table 5

Univariable analysis of neurological complication severity
		None (n = 30)	Mild (n = 22)	Moderate to severe (n = 18)	P-value
ESCC (%)	0	18 (64.3)	4 (14.3)	6 (21.4)	0.078
	1	8 (40.0)	5 (25.0)	7 (35.0)
	2	4 (28.6)	7 (50.0)	3 (21.4)
	3	0 (0.0)	6 (75.0)	2 (25.0)
Categorized SINS (%)	Stable	7 (30.4)	6 (26.1)	10 (43.5)	0.13
	Potentially unstable	23 (54.8)	11 (26.2)	8 (19.0)
	Unstable	0 (0.0)	5 (100.0)	0 (0.0)
Localized pain (%)	No	19 (44.2)	14 (32.6)	10 (23.3)	0.624
	Yes	11 (40.7)	8 (29.6)	8 (29.6)	0.624
Mechanical pain (%)	No	24 (47.1)	13 (25.5)	14 (27.5)	0.679
	Yes	6 (31.6)	9 (47.4)	4 (21.1)	0.679
Radicular pain (%)	No	24 (53.3)	10 (22.2)	11 (24.4)	0.106
	Yes	6 (24.0)	12 (48.0)	7 (28.0)	0.106
For categorical variables, Mantel-Haenszel correlation test was used, and n (row %) was reported.
For continuous variables, one-way ANOVA was used, and mean (SD) was reported.

Table 6

Subgroup analysis: univariable analysis of aggravation of SINS
		Aggravation of SINS (-)	Aggravation of SINS (+)	P-value
Localized pain (%)	No	12 (80.0)	4 (28.6)	0.009*
	Yes	3 (20.0)	10 (71.4)
Mechanical pain (%)	No	10 (66.7)	12 (85.7)	0.390
	Yes	5 (33.3)	2 (14.3)
Radicular pain (%)	No	10 (66.7)	8 (57.1)	0.710
	Yes	5 (33.3)	6 (42.9)
Location (%)	Semirigid	12 (80.0)	7 (50.0)	0.197
	Mobile spine	1 ( 6.7)	1 (7.1)
	Junctional	2 (13.3)	6 (42.9)
Bone lesion (%)	Blastic	7 (46.7)	1 (7.1)	0.028*
	Mixed	2 (13.3)	1 (7.1)
	Lytic	4 (26.7)	10 (66.6)
Radiographic spinal alignment (%)	Normal alignment	12 (80.0)	14 (100.0)	0.224
	De novo Deformity	3 (20.0)	0 (0.0)
	Subluxation / translation present	0 (0.00)	0 (0.00)
Vertebral body collapse (%)	None of the above	4 (26.7)	5 (35.7)	1.000
	No collapse with > 50% body involved	4 (26.7)	4 (28.6)
	< 50% collapse	4 (26.7)	3 (21.4)
	> 50% collapse	3 (20.0)	2 (14.3)
Posterolateral involvement (%)	None of the above	4 (26.7)	5 (35.7)	0.213
	Unilateral	4 (26.7)	7 (50.0)
	Bilateral	7 (46.7)	2 (14.3)
Fisher’s exact test was used, and n (%) was reported.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Can we predict neurological complications in patients with metastatic spinal tumors?

Status:

Version 1

Abstract

Purpose

Methods

Results

Conclusion

Introduction

Materials And Methods

Study design and participants

Clinical evaluation

SINS and ESCC

Statistical analysis

Results

Neurological complication occurrence

Aggravation of stability of the spine

Discussion

Limitations

Conclusion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1