Search Results

You are looking at 1 - 5 of 5 items for

  • Author or Editor: Bethany Hung x
  • Refine by Access: all x
Clear All Modify Search
Free access

Jeff Ehresman, Zach Pennington, James Feghali, Andrew Schilling, Andrew Hersh, Bethany Hung, Daniel Lubelski, and Daniel M. Sciubba

OBJECTIVE

More than 8000 patients are treated annually for vertebral column tumors, of whom roughly two-thirds will be discharged to an inpatient facility (nonroutine discharge). Nonroutine discharge is associated with increased care costs as well as delays in discharge and poorer patient outcomes. In this study, the authors sought to develop a prediction model of nonroutine discharge in the population of vertebral column tumor patients.

METHODS

Patients treated for primary or metastatic vertebral column tumors at a single comprehensive cancer center were identified for inclusion. Data were gathered regarding surgical procedure, patient demographics, insurance status, and medical comorbidities. Frailty was assessed using the modified 5-item Frailty Index (mFI-5) and medical complexity was assessed using the modified Charlson Comorbidity Index (mCCI). Multivariable logistic regression was used to identify independent predictors of nonroutine discharge, and multivariable linear regression was used to identify predictors of prolonged length of stay (LOS). The discharge model was internally validated using 1000 bootstrapped samples.

RESULTS

The authors identified 350 patients (mean age 57.0 ± 13.6 years, 53.1% male, and 67.1% treated for metastatic vs primary disease). Significant predictors of prolonged LOS included higher mCCI score (β = 0.74; p = 0.026), higher serum absolute neutrophil count (β = 0.35; p = 0.001), lower hematocrit (β = −0.34; p = 0.001), use of a staged operation (β = 4.99; p < 0.001), occurrence of postoperative pulmonary embolism (β = 3.93; p = 0.004), and surgical site infection (β = 9.93; p < 0.001). Significant predictors of nonroutine discharge included emergency admission (OR 3.09; p = 0.001), higher mFI-5 score (OR 1.90; p = 0.001), lower serum albumin level (OR 0.43 per g/dL; p < 0.001), and operations with multiple stages (OR 4.10; p < 0.001). The resulting statistical model was deployed as a web-based calculator (https://jhuspine4.shinyapps.io/Nonroutine_Discharge_Tumor/).

CONCLUSIONS

The authors found that nonroutine discharge of patients with surgically treated vertebral column tumors was predicted by emergency admission, increased frailty, lower serum albumin level, and staged surgical procedures. The resulting web-based calculator tool may be useful clinically to aid in discharge planning for spinal oncology patients by preoperatively identifying patients likely to require placement in an inpatient facility postoperatively.

Restricted access

Zach Pennington, Jeff Ehresman, Andrew Schilling, James Feghali, Andrew M. Hersh, Bethany Hung, Eleni N. Kalivas, Daniel Lubelski, and Daniel M. Sciubba

OBJECTIVE

Patients with spine tumors are at increased risk for both hemorrhage and venous thromboembolism (VTE). Tranexamic acid (TXA) has been advanced as a potential intervention to reduce intraoperative blood loss in this surgical population, but many fear it is associated with increased VTE risk due to the hypercoagulability noted in malignancy. In this study, the authors aimed to 1) develop a clinical calculator for postoperative VTE risk in the population with spine tumors, and 2) investigate the association of intraoperative TXA use and postoperative VTE.

METHODS

A retrospective data set from a comprehensive cancer center was reviewed for adult patients treated for vertebral column tumors. Data were collected on surgery performed, patient demographics and medical comorbidities, VTE prophylaxis measures, and TXA use. TXA use was classified as high-dose (≥ 20 mg/kg) or low-dose (< 20 mg/kg). The primary study outcome was VTE occurrence prior to discharge. Secondary outcomes were deep venous thrombosis (DVT) or pulmonary embolism (PE). Multivariable logistic regression was used to identify independent risk factors for VTE and the resultant model was deployed as a web-based calculator.

RESULTS

Three hundred fifty patients were included. The mean patient age was 57 years, 53% of patients were male, and 67% of surgeries were performed for spinal metastases. TXA use was not associated with increased VTE (14.3% vs 10.1%, p = 0.37). After multivariable analysis, VTE was independently predicted by lower serum albumin (odds ratio [OR] 0.42 per g/dl, 95% confidence interval [CI] 0.23–0.79, p = 0.007), larger mean corpuscular volume (OR 0.91 per fl, 95% CI 0.84–0.99, p = 0.035), and history of prior VTE (OR 2.60, 95% CI 1.53–4.40, p < 0.001). Longer surgery duration approached significance and was included in the final model. Although TXA was not independently associated with the primary outcome of VTE, high-dose TXA use was associated with increased odds of both DVT and PE. The VTE model showed a fair fit of the data with an area under the curve of 0.77.

CONCLUSIONS

In the present cohort of patients treated for vertebral column tumors, TXA was not associated with increased VTE risk, although high-dose TXA (≥ 20 mg/kg) was associated with increased odds of DVT or PE. Additionally, the web-based clinical calculator of VTE risk presented here may prove useful in counseling patients preoperatively about their individualized VTE risk.

Restricted access

Jeff Ehresman, Daniel Lubelski, Zach Pennington, Bethany Hung, A. Karim Ahmed, Tej D. Azad, Kurt Lehner, James Feghali, Zorica Buser, James Harrop, Jefferson Wilson, Shekar Kurpad, Zoher Ghogawala, and Daniel M. Sciubba

OBJECTIVE

The objective of this study was to evaluate the characteristics and performance of current prediction models in the fields of spine metastasis and degenerative spine disease to create a scoring system that allows direct comparison of the prediction models.

METHODS

A systematic search of PubMed and Embase was performed to identify relevant studies that included either the proposal of a prediction model or an external validation of a previously proposed prediction model with 1-year outcomes. Characteristics of the original study and discriminative performance of external validations were then assigned points based on thresholds from the overall cohort.

RESULTS

Nine prediction models were included in the spine metastasis category, while 6 prediction models were included in the degenerative spine category. After assigning the proposed utility of prediction model score to the spine metastasis prediction models, only 1 reached the grade of excellent, while 2 were graded as good, 3 as fair, and 3 as poor. Of the 6 included degenerative spine models, 1 reached the excellent grade, while 3 studies were graded as good, 1 as fair, and 1 as poor.

CONCLUSIONS

As interest in utilizing predictive analytics in spine surgery increases, there is a concomitant increase in the number of published prediction models that differ in methodology and performance. Prior to applying these models to patient care, these models must be evaluated. To begin addressing this issue, the authors proposed a grading system that compares these models based on various metrics related to their original design as well as internal and external validation. Ultimately, this may hopefully aid clinicians in determining the relative validity and usability of a given model.

Restricted access

Bethany Hung, Zach Pennington, Andrew M. Hersh, Andrew Schilling, Jeff Ehresman, Jaimin Patel, Albert Antar, Jose L. Porras, Aladine A. Elsamadicy, and Daniel M. Sciubba

OBJECTIVE

Previous studies have suggested the possibility of racial disparities in surgical outcomes for patients undergoing spine surgery, although this has not been thoroughly investigated in those with spinal metastases. Given the increasing prevalence of spinal metastases requiring intervention, knowledge about potential discrepancies in outcomes would benefit overall patient care. The objective in the present study was to investigate whether race was an independent predictor of postoperative complications, nonroutine discharge, and prolonged length of stay (LOS) after surgery for spinal metastasis.

METHODS

The authors retrospectively examined patients at a single comprehensive cancer center who had undergone surgery for spinal metastasis between April 2013 and April 2020. Demographic information, primary pathology, preoperative clinical characteristics, and operative outcomes were collected. Factors achieving p values < 0.15 on univariate regression were entered into a stepwise multivariable logistic regression to generate predictive models. Nonroutine discharge was defined as a nonhome discharge destination and prolonged LOS was defined as LOS greater than the 75th percentile for the entire cohort.

RESULTS

Three hundred twenty-eight patients who had undergone 348 operations were included: 240 (69.0%) White and 108 (31.0%) Black. On univariable analysis, cohorts significantly differed in age (p = 0.02), marital status (p < 0.001), insurance status (p = 0.03), income quartile (p = 0.02), primary tumor type (p = 0.04), and preoperative Karnofsky Performance Scale (KPS) score (p < 0.001). On multivariable analysis, race was an independent predictor for nonroutine discharge: Black patients had significantly higher odds of nonroutine discharge than White patients (adjusted odds ratio [AOR] 2.24, 95% confidence interval [CI] 1.28–3.92, p = 0.005). Older age (AOR 1.06 per year, 95% CI 1.03–1.09, p < 0.001), preoperative KPS score ≤ 70 (AOR 3.30, 95% CI 1.93–5.65, p < 0.001), preoperative Frankel grade A–C (AOR 3.48, 95% CI 1.17–10.3, p = 0.02), insurance status (p = 0.005), being unmarried (AOR 0.58, 95% CI 0.35–0.97, p = 0.04), number of levels (AOR 1.17 per level, 95% CI 1.05–1.31, p = 0.004), and thoracic involvement (AOR 1.71, 95% CI 1.02–2.88, p = 0.04) were also predictive of nonroutine discharge. However, race was not independently predictive of postoperative complications or prolonged LOS. Higher Charlson Comorbidity Index (AOR 1.22 per point, 95% CI 1.04–1.43, p = 0.01), low preoperative KPS score (AOR 1.84, 95% CI 1.16–2.92, p = 0.01), and number of levels (AOR 1.15 per level, 95% CI 1.05–1.27, p = 0.004) were predictive of complications, while insurance status (p = 0.05), income quartile (p = 0.01), low preoperative KPS score (AOR 1.64, 95% CI 1.03–2.72, p = 0.05), and number of levels (AOR 1.16 per level, 95% CI 1.05–1.30, p = 0.004) were predictive of prolonged LOS.

CONCLUSIONS

Race, insurance status, age, baseline functional status, and marital status were all independently associated with nonroutine discharge. This suggests that a combination of socioeconomic factors and functional status, rather than medical comorbidities, may best predict postdischarge disposition in patients treated for spinal metastases. Further investigation in a prospective cohort is merited.

Restricted access

Andrew M. Hersh, Zach Pennington, Bethany Hung, Jaimin Patel, Earl Goldsborough, Andrew Schilling, James Feghali, Albert Antar, Siddhartha Srivastava, David Botros, Aladine A. Elsamadicy, Sheng-Fu Larry Lo, and Daniel M. Sciubba

OBJECTIVE

Frailty—the state defined by decreased physiological reserve and increased vulnerability to physiological stress—is exceedingly common in oncology patients. Given the palliative nature of spine metastasis surgery, it is imperative that patients be healthy enough to tolerate the physical insult of surgery. In the present study, the authors compared the association of two frailty metrics and the widely used Charlson Comorbidity Index (CCI) with postoperative morbidity in spine metastasis patients.

METHODS

A retrospective cohort of patients who underwent operations for spinal metastases at a comprehensive cancer center were identified. Data on patient demographic characteristics, disease state, medical comorbidities, operative details, and postoperative outcomes were collected. Frailty was measured with the modified 5-item frailty index (mFI-5) and metastatic spinal tumor frailty index (MSTFI). Outcomes of interest were length of stay (LOS) greater than the 75th percentile of the cohort, nonroutine discharge, and the occurrence of ≥ 1 postoperative complication.

RESULTS

In total, 322 patients were included (mean age 59.5 ± 12 years; 56.9% of patients were male). The mean ± SD LOS was 11.2 ± 9.9 days, 44.5% of patients had nonroutine discharge, and 24.0% experienced ≥ 1 postoperative complication. On multivariable analysis, increased frailty on mFI-5 and MSTFI was independently predictive of all three outcomes: prolonged LOS (OR 1.67 per point, 95% CI 1.06–2.63, p = 0.03; and OR 1.63 per point, 95% CI 1.29–2.05, p < 0.01, respectively), nonroutine discharge (OR 2.65 per point, 95% CI 1.74–4.04, p < 0.01; and OR 1.69 per point, 95% CI 1.36–2.11, p < 0.01), and ≥ 1 complication (OR 1.95 per point, 95% CI 1.23–3.09, p = 0.01; and OR 1.41 per point, 95% CI 1.12–1.77, p < 0.01). CCI was found to be independently predictive of only the occurrence of ≥ 1 postoperative complication (OR 1.45 per point, 95% CI 1.22–1.72, p < 0.01).

CONCLUSIONS

Frailty measured with either mFI-5 or MSTFI scores was a more robust independent predictor of adverse postoperative outcomes than the more widely used CCI. Both mFI-5 and MSTFI were significantly associated with prolonged LOS, higher complication rates, and nonroutine discharge. Further investigation in a prospective multicenter cohort is merited.