Baseline Frailty Measured by the Risk Analysis Index and 30-Day Mortality After Surgery for Spinal Malignancy: Analysis of a Prospective Registry (2011–2020)

Article information

Neurospine. 2024;21(2):404-413

Publication date (electronic) : 2024 June 30

doi : https://doi.org/10.14245/ns.2347120.560

Rachel Thommen^,¹^,²

, Christian A. Bowers ²

, Aaron C. Segura ²

, Joanna M. Roy ³

, Meic H. Schmidt ²

¹School of Medicine, New York Medical College, Valhalla, NY, USA

²Bowers Neurosurgical Frailty and Outcomes Data Science Lab, Sandy, UT, USA

³Topiwala National Medical College, Mumbai, India

Corresponding Author Rachel Thommen Bowers Neurosurgical Frailty and Outcomes Data Science Lab, Sandy, UT 84070, USA Email: meic.schmidt@gmail.com

Received 2023 October 26; Revised 2024 February 22; Accepted 2024 February 28.

Abstract

Objective

To evaluate the prognostic utility of baseline frailty, measured by the Risk Analysis Index (RAI), for prediction of postoperative mortality among patients with spinal malignancy (SM) undergoing resection.

Methods

SM surgery cases were queried from the American College of Surgeons – National Surgical Quality Improvement Program database (2011–2020). The relationship between preoperative RAI frailty score and increasing rate of primary endpoint (mortality or discharge to hospice within 30 days, “mortality/hospice”) were assessed. Discriminatory accuracy was assessed by computation of C-statistics (with 95% confidence interval [CI]) in receiver operating characteristic (ROC) curve analysis.

Results

A total of 2,235 cases were stratified by RAI score: 0–20, 22.7%; 21–30, 11.9%; 31–40, 54.7%; and ≥ 41, 10.7%. The rate of mortality/hospice was 6.5%, which increased linearly with increasing RAI score (p < 0.001). RAI was also associated with increasing rates of major complication, extended length of stay, and nonhome discharge (all p < 0.05). The RAI demonstrated acceptable discriminatory accuracy for prediction of primary endpoint (C-statistic, 0.717; 95% CI, 0.697–0.735). In pairwise ROC comparison, RAI demonstrated superiority versus modified frailty index-5 and chronological age (p < 0.001).

Conclusion

Preoperative frailty, as measured by RAI, is a robust predictor of mortality/hospice after SM surgery. The frailty score may be applied in clinical settings using a user-friendly calculator, deployed here: https://nsgyfrailtyoutcomeslab.shinyapps.io/spinalMalignancyRAI/.

Keywords: Frailty; Risk Analysis Index; Spinal tumor; Metastatic; Spinal oncology; National Surgical Quality Improvement Program

INTRODUCTION

Spinal malignancies (SMs) are often associated with significant morbidity and reduced quality of life from neurological deficits and/or fracture-associated pain. These patients commonly present with pain and/or focal neurologic deficits. Spine metastases were found to be the most common osseous destination for secondary malignancy (70%) [1,2]. Recent literature reported 15.67% of patients diagnosed with a solid tumor have spinal metastasis; 9.6% of patients with SM will develop epidural spinal cord compression and 12.6% will experience pathologic compression fractures [3]. Although primary central nervous system malignancies are less common, 2.9% occur in the spinal cord or cauda equina regions [4]. Despite a low incidence of primary spinal neoplasms (0.77 per 100,000), their incidence peaks at the 75–84 year age group (1.80 per 100,000 in individuals), which is relevant to research studying frailty’s impact on SM surgery [5].

There is an increased risk of developing primary malignancy with increasing age, and people are living longer thanks to advancements in medical science and cancer management. Due to improved rates of survival of cancer patients, there is a higher incidence of older patients presenting with a greater cancer disease burden with spinal metastasis, in addition to the peak incidence of primary SMs that accompanies increasing patient age until the 75–84 year group [6]. Therefore, it is becoming even more important to accurately measure preoperative risk and quantify frailty as a measure of physiologic reserve, as surgical outcomes prediction cannot be based on age alone for risk stratification [7-10]. The majority of previous frailty studies in the neurosurgical literature have used the 11- or 5-modified frailty index (mFI), the mFI-11 or mFI-5, respectively. The Risk Analysis Index (RAI) is a robust frailty index originally developed and validated in surgical populations [11-13]. The RAI is uniquely versatile for both clinical prospective application with a patient-centered questionnaire and for large retrospective database analysis. However, the generalizability of RAI to neurosurgical spinal oncology is currently unknown.

The objective of the present study is to evaluate the predictive ability of the RAI for 30-day mortality and other adverse postoperative outcomes in patients undergoing surgical intervention for a SM using data from a large multicenter, clinical surgical registry representing over 700 hospitals across 49 U.S. states and 11 countries.

MATERIALS AND METHODS

1. Study Design

The present study, designed as a secondary analysis of a prospective surgical registry, was conducted in accordance with STROBE (Strengthening the Reporting of Observational studies in Epidemiology) guidelines.

2. Data Source

The patient population was derived from the American College of Surgeons-National Surgical Quality Improvement Program (ACS-NSQIP) database. The ACS-NSQIP is a prospective, peer-controlled, validated database for quantifying 30-day surgical outcomes. NSQIP data are prospectively entered at each participating hospital by ACS-trained surgical clinical reviewers to improve consistency and reliability. The present study was performed under the data user agreement of the ACS with our institution and was approved by Institutional Review Board of University of New Mexico (Study ID 21-315). Given the deidentified nature of the information in NSQIP database, patient consent was neither sought nor required.

3. Participants

The malignant spinal tumor patient population was queried from the NSQIP (2011–2020) using a combination of Current Procedural Terminology (CPT) codes and International Classification for Disease (ICD), 9th and 10th Revision, Clinical Modification codes (Table 1). Initial query using exclusively primary CPT codes yielded 7,851 records. The postoperative diagnoses fields in NSQIP (“PODIAG”) further narrowed the search to 2,235 records using prespecified ICD codes (Table 1).

Table 1.

List of CPT, ICD-9-CM, and ICD-10-CM codes used to identify SM surgery cases from the NSQIP database 2011–2020

4. Predictor Variables

The primary predictor of interest was preoperative frailty, measured by the RAI score. The RAI was originally developed and validated for compatibility with the ACS-NSQIP database [12]. RAI has been previously utilized to predict outcomes in patients undergoing spine surgery [14,15]. The revised RAI score, as described by Arya et al. [12], was computed using standard NSQIP variables: age, sex, disseminated cancer (“DISCANC”), weight loss (“WTLOSS”), renal failure (“RENAFAIL; DIALYSIS”), congestive heart failure (“HXCHF”), shortness of breath (“DYSPNEA”), functional status (“FNSTATUS2”), and living status (“TRANST”=nursing home or chronic care facility). The mFI-5, a readily utilized risk index in neurooncological literature, was computed and analyzed for comparative purposes alongside RAI [7,16-18]. mFI-5 has a 5-point maximum score, with 1 point given to each positive variable. mFI-5 score was calculated from NSQIP using the cumulative score of nonindependent functional status (“FNSTATUS2”), diabetes mellitus (“DIABETES”), chronic obstructive pulmonary disease (“HXCOPD”), and congestive heart failure (“HXCHF”).

5. Primary and Secondary Outcomes

The primary outcome for this study was mortality within 30 days of surgery and/or discharge to hospice, which was calculated using the NSQIP variables “DOPERTOD” (days from operation to death for mortality) and “DISCHDEST” (hospital discharge destination). The secondary outcomes were nonhome discharge disposition (NHD), extended length of hospital stay (ELOS), major complication occurrence, and Clavien-Dindo grade IV complication occurrence (CD IV). CD IV complications include life-threatening complications that require intensive care unit management. Major complication occurrences were defined by one or more instances of prolonged intubation (defined as greater than 48 hours), unplanned reintubation, sepsis, septic shock, pneumonia, deep venous thrombosis/thrombophlebitis, pulmonary embolism, cerebrovascular accident/stroke with neurological deficit, acute renal failure, myocardial infarction, cardiac arrest requiring cardiopulmonary resuscitation, superficial surgical site infection (SSI), deep incisional SSI, organ space SSI, or wound dehiscence.

6. Statistical Methods

Statistical analyses were performed with a combination of R Project for Statistical Computing ver. 4.2.1 (The R Foundation, Vienna, Austria) (https://www.R-project.org/), IBM SPSS Statistics ver. 28.0 (IBM Co., Armonk, NY, USA), GraphPad Prism ver. 9.0 (GraphPad Software Inc., La Jolla, CA, USA), and MedCalc for Windows ver. 19.4 (MedCalc Software, Ostend, Belgium). Statistical significance was set a priori to alpha of 0.05. The R gtsummary package was utilized to generate descriptive statistics for the spine tumor surgery cohort [19]. RAI was analyzed as a continuous variable and categorized into a base “nonfrail” category of 0–20 and subsequent ascending 5-point bins (e.g., 21–25, 26–30). The binning scheme was defined a priori and arbitrary in nature to avoid scoring biases associated with tiers [20]. The Cochran-Armitage trend test was utilized to determine significance of proportional trends (frailty score bin vs. binary outcomes). Discriminatory accuracy of frailty score for primary outcome was quantified using C-statistics (with 95% confidence interval [CI]) and interpreted using standard criteria per Hosmer-Lemeshow [19]. A C-statistic > 0.7 was indicative of adequate discriminatory accuracy [21,22]. The relative discriminatory superiority of RAI (vs. increasing chronological age and mFI-5) was assessed using pairwise receiver operating characteristic (ROC) comparison (DeLong) tests. The R, packages rms and shiny were used to generate an interactive Risk Analysis Index and Malignant Spinal Tumor Surgery Outcomes calculator found at the following website: https://nsgyfrailtyoutcomeslab.shinyapps.io/spinalMalignancyRAI/ [23-25].

RESULTS

1. Participants and Descriptive Statistics

There were 2,235 patients that underwent SM surgery and 507 were RAI 0–20, 266 were RAI 21–30, 1,222 were RAI 31–40, and 240 were RAI ≥ 41. Baseline demographics, clinical characteristics, and baseline frailty status are summarized in Table 2. The median (interquartile range) age was 61 (52–70) and 39% were female.

Table 2.

Demographics, clinical characteristics, and baseline frailty in patients undergoing surgery for SM, subgroup by primary endpoint (mortality/hospice), ACS-NSQIP 2011–2020 (N=2,235)

2. RAI and Mortality Within 30 Days of Operation

The rate of primary outcome (30-day mortality/hospice) was 6.1% (N = 136). This was significantly associated with increasing frailty by RAI (Fig. 1). RAI demonstrated adequate discriminatory accuracy for prediction of the primary endpoint in ROC analysis (C-statistic, 0.717; 95% CI, 0.697–0.735) (Fig. 2). On pairwise ROC comparison, RAI score had superior discriminatory accuracy compared to mFI-5 and increasing chronological age (both p < 0.001, DeLong test, Fig. 2).

Fig. 1.

Incidence rate of mortality within 30 days of operation stratified by preoperative RAI frailty score, ACS-NSQIP 2011-2020 (N=2,235). RAI, Risk Analysis Index; ACS-NSQIP, American College of Surgeons – National Surgical Quality Improvement Program.

Fig. 2.

Risk Analysis Index (RAI) with superior discriminatory accuracy for primary endpoint of 30-day mortality compared to age and mFI-5 (DeLong pairwise comparison, p < 0.001), ACS-NSQIP, 2011–2020 (N=2,235). ACS-NSQIP, American College of Surgeons – National Surgical Quality Improvement Program; mFI-5, modified frailty index-5; CI, confidence interval.

3. Complications, ELOS, and NHD

There were 181 major complication occurrences (8.1%), 126 CD IV occurrences (5.6%), 937 NHDs (42%), 571 ELOS (26%), within 30 days. Increasing RAI score (in stepwise 5-point bins) was statistically significantly associated with increasing occurrence rate of major complication, CD IV, ELOS, and NHD (Table 3, Fig. 3).

Table 3.

Postoperative complications, discharge outcomes, and 30-day outcomes after surgery for SM stratified by RAI frailty score, ACS-NSQIP 2011–2020 (N=2,235)

Fig. 3.

Postoperative outcomes stratified by RAI frailty tier, ACS-NSQIP 2011–2020 (N=2,235). RAI, Risk Analysis Index; ACS-NSQIP, American College of Surgeons – National Surgical Quality Improvement Program.

DISCUSSION

In this analysis of 2,235 SM patients undergoing surgical treatment in a prospective surgical registry, increasing frailty, as measured by the RAI, provided discriminatory accuracy for prediction of postoperative mortality/hospice discharge. Increasing RAI scores were also associated with the other secondary outcomes consisting of major complications, CD IV, NHD, and ELOS. Comparative ROC curve analysis demonstrated the superior discrimination of the RAI compared to the mFI-5 and chronological age in prediction of the primary endpoint. These results may improve informed consent and surgical decision-making by enabling more accurate preoperative surgical risk prediction.

Few studies have similarly examined the use of frailty tools for outcome prediction in SM patients [7]. Many studies have described the importance of capturing frailty for preoperative risk assessment, moreover accurate frailty measurement and prediction is exceedingly crucial for SM patients who are often among the most frail due to their metastatic disease or SMs, advanced chronologic age, prior radiotherapy, and the necessity of surgical intervention despite these factors [8,26-28]. Hersh et al. [9] established that the mFI-5 and Metastatic Spinal Tumor Frailty Index (MSTFI) were superior to the widely used Charlson Comorbidity Index, supporting the well-established tenet that increasing frailty, and not just comorbidity accounting, robustly predict outcomes. A recent meta-analysis determined mFI-5 correlates better with postoperative complication prediction, while MSTFI predicts mortality better [8]. However, other scholars found the MSTFI had poor discrimination in predicting mortality and postoperative complications for SM patients in a single center machine learning study [29]. Another retrospective cohort study from a single institution similarly concluded that the mFI-5 did not accurately predict outcomes in surgically treated SM patients [30]. In a series of surgical SM patients treated at a single comprehensive cancer center, Ehresman et al. [31] analyzed specific preoperative comorbidities to create a web-based calculator predicting nonhome discharge disposition. Preoperative albumin values, emergency status, and increasing complexity of the surgical procedure are established variables known to affect postoperative outcomes [32-36]. In addition to functional dependence, knowledge of the primary site of cancer, number, and site of metastases play a role in risk stratification in patients with metastatic spinal tumors [37,38]. However, these variables do not quantify the baseline physiological reserve of a patient (i.e., frailty) even in conjunction with the mFI-5 [33,39-41]. Hall and colleagues adapted the Minimum Data Set Mortality Risk Index-Revised as a tool for the prospective or retrospective calculation of frailty. The 14-variables captured by RAI to create a frailty score have been recalibrated and validated across multiple surgical specialties [11,42-45].

Prior studies have robustly demonstrated the predictive value of mFI-5 for poor postoperative outcomes in various surgical populations, including SM [8,46]. However, the mFI-5 is limited as a unidimensional scale, merely consisting of 4 comorbidities plus a binary functional assessment of independent versus dependent status (fully dependent or partially dependent). The RAI incorporates multiple domains of frailty by assessing comorbidities, functional status, nutritional baseline, mental status, and transfer status. The RAI’s more comprehensive assessment of physiologic reserve in patients with SM is reflected by its superior discrimination in mortality/hospice prediction (compared to mFI-5) in patients with SMs. This study also adds support to the previous literature that also refutes the notion that frailty simply increases in proportion with chronological age [47,48]. Rather, frailty and chronological age need to be independently assessed preoperatively to accurately predict risk associated with any potential surgical intervention [47,49]. As such, the predictive ability of RAI plays a vital role in this cohort where the prevalence of primary SMs increases with age and improved rates of cancer survival predispose to the risk of future metastasis [6,49]. This study complements the growing body of literature validating the use of RAI in predicting adverse postoperative outcomes in neurosurgical spine patient populations [50,51].

Frailty assessment at clinic visits could guide preoperative counseling and prehabilitation efforts to prevent adverse outcomes in elective spine surgery procedures [52]. The ease of calculating frailty scores could play a role in the extent of their incorporation into routine clinical practice [53]. For instance, scores like the Hospital Frailty Risk Score (HFRS) are calculated from over a 100 ICD-10 codes [54]. Although it is unclear as to how well HFRS is currently being utilized in clinical practice, one might imagine that calculating frailty scores based on HFRS might prove challenging [55]. RAI, on the other hand, can be calculated using a 14-item questionnaire by the caregiver. Prior literature has described its ease of calculation in prospective cohorts of patients undergoing spine surgery [50,56].

While not the primary endpoint, the RAI was also associated with ELOS, NHD, major complications, and CD IV complications. These adverse outcomes impact the patient and their family, but also increase the cost and burden on the healthcare system [57]. The elevated risk of poor outcomes with higher RAI scores highlights the importance of preoperative counseling with patients and families for more precise risk-benefit estimation [58,59]. As our analysis demonstrates the superiority of RAI over mFI-5 in terms of discriminatory accuracy for predicting mortality and since increasing RAI is associated with all the other adverse outcomes, clinicians may consider implementing RAI scoring during the preoperative surgical evaluation of SM patients.

RAI score was modified according to criteria described in prior studies to compute the score without the preoperative cognitive decline variable since this variable is no longer available in the NSQIP database [20]. The NSQIP only supplies a single ICD diagnosis field called “PODIAG” (postoperative diagnosis) and thus ICD codes cannot be utilized to define diagnoses other than the primary surgical diagnosis. There are also limitations related to specific outcomes of interest for the surgical treatment of SM. The NSQIP does not record postoperative functional outcomes such as urinary/bowel incontinence or other neurological deficits. Spinal cord compression is typically an indication for emergent/urgent surgical decompression and restoration. Information on quality of life or symptom improvement after surgery would add value to future research as it relates to RAI frailty indexing preoperatively. Additionally, variables critical in neurooncology research are not recorded in the NSQIP such as tumor size, surgical approach, and extent of resection. Furthermore, ICD coding is inherently difficult in distinguishing primary from metastatic neoplasms (Table 1). Thus, the authors opted to include all “malignant” spinal tumors. Further prospective research using qualitative data may provide better insight into discrete outcomes between these 2 groups that was unable to be performed in our retrospective analysis. Similarly, the MSTFI is not available using NSQIP, which we could not compare in this study to retrospective RAI. However, further prospective RAI studies can compare MSTFI in this patient population. Despite these limitations, this study is robust, with a large sample size of 2,235 patients, providing the necessary statistical power to analyze the prognostic utility of the RAI score.

CONCLUSION

Preoperative frailty, measured by the RAI score, was a predictor of postoperative mortality and/or hospice discharge in patients undergoing resection of SMs. Increased RAI scores are not only highly predictive of 30-day mortality or discharge to hospice, but are also associated with major complications, CD IV complications, ELOS, and NHD. Of note, the RAI demonstrated superior discrimination as compared to mFI-5 and age for outcome prediction. The frailty analysis may be translated to the bedside with a user-friendly application. The present work provides a foundation for frailty research in patients with SMs, including ongoing prospective clinical studies at the authors’ institution.

Notes

Conflict of Interest

In order to comply with the Hospital Participation Agreement (HPA) that is agreed to between the ACS and participating sites, facility identifiers as well as geographic information regarding the case have been removed. The HPA stipulates that the ACS does not identify participating sites. Site identification could be possible even with blinded identifiers through advanced statistics. A stipulation of access to the PUF is completion of the Data Use Agreement that strictly prohibits attempts to identify hospitals, health care providers, or patients. The authors have nothing to disclose.

Funding/Support

This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Author Contribution

Conceptualization: RT, CAB, ACS, MHS; Data curation: RT, ACS; Formal analysis: RT, ACS; Methodology: CAB, ACS, JMR; Visualization: CAB, ACS; Writing – original draft: RT, CAB; Writing – review & editing: RT, ACS, JMR, MHS.

References

1. Wewel JT, O'Toole JE. Epidemiology of spinal cord and column tumors. Neurooncol Pract 2020;7(Suppl 1):i5–9.

2. Sutcliffe P, Connock M, Shyangdan D, et al. A systematic review of evidence on malignant spinal metastases: natural history and technologies for identifying patients at high risk of vertebral fracture and spinal cord compression. Health Technol Assess 2013;17:1–274.

3. Van den Brande R, Mj Cornips E, Peeters M, et al. Epidemiology of spinal metastases, metastatic epidural spinal cord compression and pathologic vertebral compression fractures in patients with solid tumors: a systematic review. J Bone Oncol 2022;35:100446.

4. Ostrom QT, Cioffi G, Gittleman H, et al. CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2012–2016. Neuro Oncol 2019;21(Suppl 5):v1–100.

5. Schellinger KA, Propp JM, Villano JL, et al. Descriptive epidemiology of primary spinal cord tumors. J Neurooncol 2008;87:173–9.

6. Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394–424.

7. Kazim SF, Dicpinigaitis A, Bowers C, et al. Frailty status is a more robust predictor than age of spinal tumor surgery outcomes: a NSQIP analysis of 4,662 patients. Neurospine 2022;19:53–62.

8. Veronesi F, Borsari V, Martini L, et al. The impact of frailty on spine surgery: systematic review on 10 years clinical studies. Aging Dis 2021;12:625–45.

9. Hersh AM, Pennington Z, Hung B, et al. Comparison of frailty metrics and the Charlson Comorbidity Index for predicting adverse outcomes in patients undergoing surgery for spine metastases. J Neurosurg Spine 2021;36:849–57.

10. Kweh B, Lee H, Tan T, et al. Spinal surgery in patients aged 80 years and older: risk stratification using the modified frailty index. Global Spine J 2021;11:525–32.

11. Hall DE, Arya S, Schmid KK, et al. Development and initial validation of the Risk Analysis Index for measuring frailty in surgical populations. JAMA Surg 2017;152:175–82.

12. Arya S, Varley P, Youk A, et al. Recalibration and external validation of the Risk Analysis Index: a surgical frailty assessment tool. Ann Surg 2020;272:996–1005.

13. George EL, Hall DE, Youk A, et al. Association between patient frailty and postoperative mortality across multiple noncardiac surgical specialties. JAMA Surg 2021;156e205152.

14. Kassicieh AJ, Rumalla K, Segura AC, et al. Endoscopic and nonendoscopic approaches to single-level lumbar spine decompression: propensity score-matched comparative analysis and frailty-driven predictive model. Neurospine 2023;20:119–28.

15. Covell MM, Rumalla K, Kassicieh AJ, et al. Frailty measured by risk analysis index and adverse discharge outcomes after adult spine deformity surgery: analysis of 3104 patients from a prospective surgical registry (2011-2020). Spine J 2023;23:739–45.

16. Khalafallah AM, Huq S, Jimenez AE, et al. The 5-factor modified frailty index: an effective predictor of mortality in brain tumor patients. J Neurosurg 2020;135:78–86.

17. Dicpinigaitis AJ, Kazim SF, Schmidt MH, et al. Association of baseline frailty status and age with postoperative morbidity and mortality following intracranial meningioma resection. J Neurooncol 2021;155:45–52.

18. Dicpinigaitis AJ, Hanft S, Cooper JB, et al. Comparative associations of baseline frailty status and age with postoperative mortality and duration of hospital stay following metastatic brain tumor resection. Clin Exp Metastasis 2022;39:303–10.

19. Sjoberg D, Whiting K, Curry M, et al. Reproducible summary tables with the gtsummary package. R J 2021;13:570–80.

20. Thommen R, Kazim SF, Rumalla K, et al. Preoperative frailty measured by risk analysis index predicts complications and poor discharge outcomes after Brain Tumor Resection in a large multi-center analysis. J Neurooncol 2022;160:285–97.

21. Hosmer D Jr, Lemeshow S, Sturdivant R. Applied logistic regression Hoboken (NJ): John Wiley & Sons Inc; 2013.

22. Swets JA. Measuring the accuracy of diagnostic systems. Science 1988;240:1285–93.

23. The R Project for Statistical Computing [Internet]. The R Foundation; [cited 2022 Aug 21]. Available from: https://www.r-project.org/.

24. Harrell FE Jr. rms: regression modeling strategies [Internet]. 2024. [cited 2022 Aug 22]. Available from: https://CRAN.Rproject.org/package=rms.

25. Chang W, Cheng J, Allaire JJ, et al. shiny: web application framework for R [Internet]. 2022. [cited 2022 Aug 21]. Available from: https://CRAN.R-project.org/package=shiny.

26. Bakhsheshian J, Shahrestani S, Buser Z, et al. The performance of frailty in predictive modeling of short-term outcomes in the surgical management of metastatic tumors to the spine. Spine J 2022;22:605–15.

27. Clegg A, Young J, Iliffe S, et al. Frailty in elderly people. Lancet 2013;381:752–62.

28. Ahmed AK, Goodwin CR, De la Garza-Ramos R, et al. Predicting short-term outcome after surgery for primary spinal tumors based on patient frailty. World Neurosurg 2017;108:393–8.

29. Massaad E, Williams N, Hadzipasic M, et al. Performance assessment of the metastatic spinal tumor frailty index using machine learning algorithms: limitations and future directions. Neurosurg Focus 2021;50:E5.

30. Mohd Rothi I, Deverall HH, Baker JF. The modified Frailty Index does not correlate with survival in surgically-treated patients with metastatic spine disease. J Clin Neurosci 2019;66:178–81.

31. Ehresman J, Pennington Z, Feghali J, et al. Predicting nonroutine discharge in patients undergoing surgery for vertebral column tumors. J Neurosurg Spine 2020;34:364–73.

32. Adogwa O, Elsamadicy AA, Mehta AI, et al. Preoperative nutritional status is an independent predictor of 30-day hospital readmission after elective spine surgery. Spine (Phila Pa 1976) 2016;41:1400–4.

33. Adogwa O, Martin JR, Huang K, et al. Preoperative serum albumin level as a predictor of postoperative complication after spine fusion. Spine (Phila Pa 1976) 2014;39:1513–9.

34. De la Garza Ramos R, Park C, McCray E, et al. Interhospital transfer status for spinal metastasis patients in the United States is associated with more severe clinical presentations and higher rates of inpatient complications. Neurosurg Focus 2021;50:E4.

35. Hu SS, Fontaine F, Kelly B, et al. Nutritional depletion in staged spinal reconstructive surgery. The effect of total parenteral nutrition. Spine (Phila Pa 1976) 1998;23:1401–5.

36. Zuckerman SL, Laufer I, Sahgal A, et al. When less is more: the indications for MIS techniques and separation surgery in metastatic spine disease. Spine (Phila Pa 1976) 2016;41 Suppl 20(Suppl 20):S246–53.

37. Tokuhashi Y, Uei H, Oshima M, et al. Scoring system for prediction of metastatic spine tumor prognosis. World J Orthop 2014;5:262–71.

38. Schoenfeld AJ, Le HV, Marjoua Y, et al. Assessing the utility of a clinical prediction score regarding 30-day morbidity and mortality following metastatic spinal surgery: the New England Spinal Metastasis Score (NESMS). Spine J 2016;16:482–90.

39. Paulino Pereira NR, Ogink PT, Groot OQ, et al. Complications and reoperations after surgery for 647 patients with spine metastatic disease. Spine J 2019;19:144–56.

40. Bisson EF, Dimar J, Harrop JS, et al. Congress of neurological surgeons systematic review and evidence-based guidelines for perioperative spine: preoperative nutritional assessment. Neurosurgery 2021;89(Suppl 1):S26–32.

41. van Tol FR, Choi D, Verkooijen HM, et al. Delayed presentation to a spine surgeon is the strongest predictor of poor postoperative outcome in patients surgically treated for symptomatic spinal metastases. Spine J 2019;19:1540–7.

42. Varley PR, Borrebach JD, Arya S, et al. Clinical utility of the risk analysis index as a prospective frailty screening tool within a multi-practice, multi-hospital integrated healthcare system. Ann Surg 2021;274:e1230–7.

43. Shah R, Borrebach JD, Hodges JC, et al. Validation of the Risk Analysis Index for evaluating frailty in ambulatory patients. J Am Geriatr Soc 2020;68:1818–24.

44. Shinall MC Jr, Arya S, Youk A, et al. Association of preoperative patient frailty and operative stress with postoperative mortality. JAMA Surg 2020;155e194620.

45. Hall DE, Arya S, Schmid KK, et al. Association of a frailty screening initiative with postoperative survival at 30, 180, and 365 days. JAMA Surg 2017;152:233–40.

46. Elsamadicy AA, Havlik JL, Reeves B, et al. Assessment of frailty indices and Charlson Comorbidity Index for predicting adverse outcomes in patients undergoing surgery for spine metastases: a national database analysis. World Neurosurg 2022;164:e1058–70.

47. Kweh BTS, Lee HQ, Tan T, et al. Risk stratification of elderly patients undergoing spinal surgery using the modified frailty index. Global Spine J 2023;13:457–65.

48. Ostrom QT, Cioffi G, Waite K, et al. CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2014-2018. Neuro Oncol 2021;23(12 Suppl 2):iii1–105.

49. Watanabe T, Kanayama M, Takahata M, et al. Perioperative complications of spine surgery in patients 80 years of age or older: a multicenter prospective cohort study. J Neurosurg Spine 2020;32:622–30.

50. Agarwal N, Goldschmidt E, Taylor T, et al. Impact of frailty on outcomes following spine surgery: a prospective cohort analysis of 668 patients. Neurosurgery 2021;88:552–7.

51. Conlon M, Thommen R, Kazim SF, et al. Risk Analysis Index and its recalibrated version predict postoperative outcomes better than 5-factor modified frailty index in traumatic spinal injury. Neurospine 2022;19:1039–48.

52. Mohamed B, Ramachandran R, Rabai F, et al. Frailty assessment and prehabilitation before complex spine surgery in patients with degenerative spine disease: a narrative review. J Neurosurg Anesthesiol 2023;35:19–30.

53. Roy JM, Bowers CA, Rumalla K, et al. Frailty indexes in metastatic spine tumor surgery: a narrative review. World Neurosurg 2023;178:117–22.

54. Gilbert T, Neuburger J, Kraindler J, et al. Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study. Lancet 2018;391:1775–82.

55. Covell MM, Roy JM, Rumalla K, et al. The limited utility of the hospital frailty risk score as a frailty assessment tool in neurosurgery: a systematic review. Neurosurgery 2024;94:251–62.

56. Link RL, Rumalla K, Courville EN, et al. Prospective application of the risk analysis index to measure preoperative frailty in spinal tumor surgery: a single center outcomes analysis. World Neurosurg 2023;19:100203.

57. Shahrestani S, Bakhsheshian J, Solaru S, et al. Inclusion of frailty improves predictive modeling for postoperative outcomes in surgical management of primary and secondary lumbar spine tumors. World Neurosurg 2021;153:e454–63.

58. Aebi M. Spinal metastasis in the elderly. Eur Spine J 2003;12 Suppl 2(Suppl 2):S202–13.

59. Yang L, Wang F, Zhang H, et al. Patient characteristics following surgery for spinal metastases: a multicenter retrospective study. Orthop Surg 2019;11:1039–47.

Article information Continued

This is an open access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Coding system	Code	Description
CPT	63275-63280	Laminectomy for biopsy/excision of extradural spinal neoplasm
	63280-63283	Laminectomy for biopsy/excision of intradural extramedullary spinal neoplasm
	63285-63287	Laminectomy for biopsy/excision of intradural intramedullary spinal neoplasm
	63290	Laminectomy for biopsy/excision of combined intradural/extradural spinal neoplasm
	63300-63303	Vertebral corpectomy for excision of extradural spinal neoplasm
	63304-63307	Vertebral corpectomy for excision of intradural spinal neoplasm
ICD-9-CM	170.2	Malignant neoplasm of vertebral column excluding sacrum and coccyx
	170.6	Malignant neoplasm of pelvic bones, sacrum, and coccyx
	192.2	Malignant neoplasm of spinal cord
	192.3	Malignant neoplasm of spinal meninges
	198.3-198.5	Secondary malignant neoplasm of brain and spinal cord
	733.13	Pathological fracture of vertebrae
	239.7	Neoplasm of uncertain behavior other parts of central nervous system
ICD-10-CM	C41.2	Malignant neoplasm of vertebral column
	C72	Malignant neoplasm of spinal cord
	C79.49	Malignant neoplasm of other parts of central nervous system
	C79.51	Secondary malignant neoplasm of bone
	D43.4	Neoplasm of uncertain behavior of spinal cord
	M48.50X	Pathological fracture in neoplastic disease, unspecified site
	M84.68	Pathological fracture in other disease, other site

Variable	All cohort	Primary endpoint (mortality/hospice)		p-value
Variable	All cohort	No	Yes	p-value
Total patients	2,235	2,099	136
Age (yr), median (IQR)	61 (52–70)	61 (52–70)	66 (58–73)	0.003
Female sex (biological)	865 (39)	822 (39)	43 (32)	0.080
Race				0.640
White	1,476 (66)	1,381 (66)	95 (70)
Black	218 (9.8)	204 (9.7)	14 (10)
Asian	98 (4.4)	94 (4.5)	4 (2.9)
Other	443 (20)	420 (20)	23 (17)
Hispanic ethnicity	162 (7.2)	156 (7.4)	6 (4.4)
Body mass index (kg/m²), median (IQR)	27.0 (23.5–30.5)	27.0 (23.5–30.5)	26.0 (22.0–30.0)	0.04
Acute care transfer	423 (19)	385 (18)	38 (28)	0.006
RAI, median (IQR)	34 (26–42)	34 (26–42)	37 (34–41)	< 0.001
RAI categorical bins				< 0.001
0–20	507 (23)	498 (24)	9 (6.6)
21–25	173 (7.7)	169 (8.1)	4 (2.9)
26–30	93 (4.2)	90 (4.3)	3 (2.2)
31–35	635 (28)	605 (29)	30 (22)
36–40	587 (26)	540 (26)	47 (35)
41–45	172 (7.7)	145 (6.9)	27 (20)
46–50	51 (2.3)	41 (2.0)	10 (7.4)
≥ 51	17 (0.8)	11 (0.5)	6 (4.4)
mFI-5				< 0.001
0	1,079 (48)	1,031 (49)	48 (35)
1	807 (36)	759 (36)	48 (35)
2	301 (13)	272 (13)	29 (21)
3	44 (2.0)	37 (1.8)	7 (5.1)
4	4 (0.2)	0 (0)	4 (2.9)

Variable	All cohort	RAI category				p-value
Variable	All cohort	0–20	21–30	31–40	≥ 41	p-value
Total patient	2,235	507	266	1,222	240
Postoperative major complication	181 (8.1)	29 (5.7)	19 (7.1)	104 (8.5)	29 (12.1)	0.022
Clavien-Dindo IV complication	126 (5.6)	17 (3.4)	15 (5.6)	69 (5.6)	25 (10.4)	0.002
Extended length of stay	571 (25.5)	89 (17.6)	63 (23.7)	318 (26.0)	101 (42.1)	< 0.001
Nonhome discharge disposition	937 (41.9)	146 (28.8)	119 (44.7)	521 (42.6)	151 (62.9)	< 0.001
Mortality within 30 days of operation	136 (6.1)	9 (1.8)	7 (2.6)	77 (6.3)	43 (17.9)	< 0.001
Mortality or hospice	146 (6.5)	9 (1.8)	8 (3.0)	85 (7.0)	44 (18.3)	< 0.001