Predictors of Prolonged Length of Stay After Lumbar Interbody Fusion: A Multicenter Study

Discharge within 1 day following elective single-level transforaminal lumbar interbody fusion: a propensityscore-matched analysis of predictors, complications, and readmission

Study Design

This was a retrospective case-control study using 8 years of data from a nationwide database of surgical outcomes in the United States.

Purpose

This study aimed to improve our understanding of the risk factors associated with a length of stay (LOS) >1 day and aid in reducing postoperative hospitalization and complications.

Overview of Literature

Despite the proven safety of transforaminal lumbar interbody fusion (TLIF), some patients face prolonged postoperative hospitalization.

Methods

Data were collected from the American College of Surgeons National Surgical Quality Improvement Program dataset from 2011 to 2018. The cohort was divided into patients with LOS up to 1 day (LOS ≤1 day), defined as same day or next-morning discharge, and patients with LOS >1 day (LOS >1 day). Univariable and multivariable regression analyses were performed to evaluate predictors of LOS >1 day. Propensity-score matching was performed to compare pre- and postdischarge complication rates.

Results

A total of 12,664 eligible patients with TLIF were identified, of which 14.8% had LOS ≤1 day and 85.2% had LOS >1 day. LOS >1 day was positively associated with female sex, Hispanic ethnicity, diagnosis of spondylolisthesis, American Society of Anesthesiologists classification 3, and operation length of >150 minutes. Patients with LOS >1 day were more likely to undergo intraoperative/postoperative blood transfusion (0.3% vs. 4.5%, p<0.001) and reoperation (0.1% vs. 0.6%, p=0.004). No significant differences in the rates of postdischarge complications were found between the matched groups.

Conclusions

Patients with worsened preoperative status, preoperative diagnosis of spondylolisthesis, and prolonged operative time are more likely to require prolonged hospitalization and blood transfusions and undergo unplanned reoperation. To reduce the risk of prolonged hospitalization and associated complications, patients indicated for TLIF should be carefully selected.

Utilizing a comprehensive machine learning approach to identify patients at high risk for extended length of stay following spinal deformity surgery in pediatric patients with early onset scoliosis

PURPOSE

Early onset scoliosis (EOS) patient diversity makes outcome prediction challenging. Machine learning offers an innovative approach to analyze patient data and predict results, including LOS in pediatric spinal deformity surgery.

METHODS

Children under 10 with EOS were chosen from the American College of Surgeon's NSQIP database. Extended LOS, defined as over 5 days, was predicted using feature selection and machine learning in Python. The best model, determined by the area under the curve (AUC), was optimized and used to create a risk calculator for prolonged LOS.

RESULTS

The study included 1587 patients, mostly young (average age: 6.94 ± 2.58 years), with 33.1% experiencing prolonged LOS (n = 526). Most patients were female (59.2%, n = 940), with an average BMI of 17.0 ± 8.7. Factors influencing LOS were operative time, age, BMI, ASA class, levels operated on, etiology, nutritional support, pulmonary and neurologic comorbidities. The gradient boosting model performed best with a test accuracy of 0.723, AUC of 0.630, and a Brier score of 0.189, leading to a patient-specific risk calculator for prolonged LOS.

CONCLUSIONS

Machine learning algorithms accurately predict extended LOS across a national patient cohort and characterize key preoperative drivers of increased LOS after PSIF in pediatric patients with EOS.

High health care use prior to elective surgery for osteoarthritis is associated with poor postoperative outcomes: A Canadian population-based cohort study

BACKGROUND

The characterization and influence of preoperative health care use on quality-of-care indicators (e.g., readmissions) has received limited attention in populations with musculoskeletal disorders. The purpose of this study was to characterize preoperative health care use and examine its effect on quality-of-care indicators among patients undergoing elective surgery for osteoarthritis.

METHODS

Data on health care use for 124,750 patients with elective surgery for osteoarthritis in Ontario, Canada, from April 1, 2015 to March 31, 2018 were linked across health administrative databases. Using total health care use one-year previous to surgery, patients were grouped from low to very high users. We used Poisson regression models to estimate rate ratios, while examining the relationship between preoperative health care use and quality-of-care indicators (e.g., extended length of stay, complications, and 90-day hospital readmissions). We controlled for covariates (age, sex, neighborhood income, rural/urban residence, comorbidities, and surgical anatomical site).

RESULTS

We found a statistically significant trend of increasing worse outcomes by health care use gradients that persisted after controlling for patient demographics and comorbidities. Findings were consistent across surgical anatomical sites. Moreover, very high users have relatively large numbers of visits to non-musculoskeletal specialists.

CONCLUSIONS

Our findings highlight that information on patients' preoperative health care use, together with other risk factors (such as comorbidities), could help decision-making when benchmarking or reimbursing hospitals caring for complex patients undergoing surgery for osteoarthritis.

Impact of American Society of Anesthesiologists' Classification on Postoperative Satisfaction and Clinical Outcomes Following Lumbar Decompression: Cohort-Matched Analysis

STUDY DESIGN

Retrospective cohort.

OBJECTIVE

The aim was to compare patient-reported outcomes (PROMs), minimum clinically important difference (MCID) achievement, and postoperative satisfaction following minimally invasive lumbar decompression in patients stratified by American Society of Anesthesiologists (ASA) classification.

SUMMARY OF BACKGROUND DATA

Some guidelines recommend against performing elective procedures for patients with an ASA score of 3 or greater; however, long-term postoperative outcomes are not well described.

METHODS

Primary, single-level, minimally invasive lumbar decompression procedures were identified. PROMs were administered at preoperative, 6-week, 12-week, 6-month, 1-year, 2-year timepoints and included Patient-Reported Outcomes Measurement Information System-Physical Function, visual analog scale (VAS) back/leg, Oswestry disability index (ODI), and 12-item short form physical component score. Satisfaction scores were collected postoperatively for VAS back/leg, ODI, and individual ODI subcategories. Patients were grouped (ASA<3, ASA≥3), and propensity scores were matched to control for significant differences. Demographic and perioperative characteristics were compared using χ 2 and the Student's t test. Mean PROMs and postoperative satisfaction were compared at each time point by a 2-sample t test. Postoperative PROM improvement from the preoperative baseline within each cohort was calculated with a paired t test. MCID achievement was determined by comparing ΔPROMs to established thresholds and comparing between groups using simple logistic regression.

RESULTS

One hundred and twenty-nine propensity-matched patients were included: 99 ASA<3 and 30 ASA≥3. No significant demographic differences were observed between groups. ASA≥3 patients experienced significantly increased length of stay and postoperative narcotic consumption on surgery day ( P <0.048, all). Mean PROMs and MCID achievement did not differ. The ASA<3 cohort significantly improved from the preoperative baseline for all PROMs at all postoperative time points. ASA<3 patients demonstrated higher levels of postoperative satisfaction at 6 weeks for VAS leg, VAS back, ODI, sleeping, lifting, walking, standing, sex, travel, and at 6 months for VAS back ( P <0.045, all).

CONCLUSION

ASA≥3 patients may achieve similar long-term clinical outcomes to ASA<3 patients, though they may show poorer short-term satisfaction for disability, leg pain, and back pain, which could be related to differing preoperative expectations.

MRI-based central sarcopenia negatively impacts the therapeutic effectiveness of single-segment lumbar fusion surgery in the elderly

Central sarcopenia is associated with the prognosis of various orthopedic surgeries in the elderly. This study aims to investigate its impact on the outcomes of single-segment lumbar fusion surgery in elderly patients. Retrospective analysis was conducted on 314 patients aged 60 to 80 who underwent single-segment posterior lumbar fusion surgery due to degenerative lumbar diseases. Patients were categorized into high psoas and L4 vertebral index (PLVI) and low PLVI groups according to the MRI-measured PLVI for central sarcopenia. Basic patient data, surgery-related parameters, functional assessments at preoperative and postoperative 3, 6, and 12 months, and X-ray-based fusion status were compared. The basic data of the two groups showed no significant differences. Parameters including the operative segment, preoperative hemoglobin levels, surgical duration, and intraoperative blood loss exhibited no significant variances. However, notable differences were observed in postoperative initial hemoglobin levels, transfusion requirements, and length of hospital stay between the two groups. During the postoperative follow-ups at 3, 6, and 12 months, the VAS scores for lower back pain and ODI scores in the lower PLVI group were significantly higher compared to the high PLVI group. Additionally, the EuroQoL 5D scores were notably lower in the low PLVI group. There were no significant differences between the groups in terms of leg pain VAS scores at each time point and the fusion status at 12 months postoperatively. MRI-based central sarcopenia has a negative impact on the therapeutic effectiveness following single-segment lumbar fusion surgery in elderly patients.

Artificial intelligence-based analysis of associations between learning curve and clinical outcomes in endoscopic and microsurgical lumbar decompression surgery

PURPOSE

A common spine surgery procedure involves decompression of the lumbar spine. The impact of the surgeon's learning curve on relevant clinical outcomes is currently not well examined in the literature. A variety of machine learning algorithms have been investigated in this study to determine how a surgeon's learning curve and other clinical parameters will influence prolonged lengths of stay (LOS), extended operating times (OT), and complications, as well as whether these clinical parameters can be reliably predicted.

METHODS

A retrospective monocentric cohort study of patients with lumbar spinal stenosis treated with microsurgical (MSD) and full-endoscopic (FED) decompression was conducted. The study included 206 patients with lumbar spinal stenosis who underwent FED (63; 30.6%) and MSD (118; 57.3%). Prolonged LOS and OT were defined as those exceeding the 75th percentile of the cohort. Furthermore, complications were assessed as a dependent variable. Using unsupervised learning, clusters were identified in the data, which helped distinguish between the early learning curve (ELC) and the late learning curve (LLC). From 15 algorithms, the top five algorithms that best fit the data were selected for each prediction task. We calculated the accuracy of prediction (Acc) and the area under the curve (AUC). The most significant predictors were determined using a feature importance analysis.

RESULTS

For the FED group, the median number of surgeries with case surgery type at the time of surgery was 72 in the ELC group and 274 in the LLC group. FED patients did not significantly differ in outcome variables (LOS, OT, complication rate) between the ELC and LLC group. The random forest model demonstrated the highest mean accuracy and AUC across all folds for each classification task. For OT, it achieved an accuracy of 76.08% and an AUC of 0.89. For LOS, the model reached an accuracy of 83.83% and an AUC of 0.91. Lastly, in predicting complications, the random forest model attained the highest accuracy of 89.90% and an AUC of 0.94. Feature importance analysis indicated that LOS, OT, and complications were more significantly affected by patient characteristics than the surgical technique (FED versus MSD) or the surgeon's learning curve.

CONCLUSIONS

A median of 72 cases of FED surgeries led to comparable clinical outcomes in the early learning curve phase compared to experienced surgeons. These outcomes seem to be more significantly affected by patient characteristics than the learning curve or the surgical technique. Several study variables, including the learning curve, can be used to predict whether lumbar decompression surgery will result in an increased LOS, OT, or complications. To introduce the provided prediction tools into clinics, the algorithms need to be implemented into open-source software and externally validated through large-scale randomized controlled trials.

Performance of artificial intelligence-based algorithms to predict prolonged length of stay after head and neck cancer surgery

Background

Medical resource management can be improved by assessing the likelihood of prolonged length of stay (LOS) for head and neck cancer surgery patients. The objective of this study was to develop predictive models that could be used to determine whether a patient's LOS after cancer surgery falls within the normal range of the cohort.

Methods

We conducted a retrospective analysis of a dataset consisting of 300 consecutive patients who underwent head and neck cancer surgery between 2017 and 2022 at a single university medical center. Prolonged LOS was defined as LOS exceeding the 75th percentile of the cohort. Feature importance analysis was performed to evaluate the most important predictors for prolonged LOS. We then constructed 7 machine learning and deep learning algorithms for the prediction modeling of prolonged LOS.

Results

The algorithms reached accuracy values of 75.40 (radial basis function neural network) to 97.92 (Random Trees) for the training set and 64.90 (multilayer perceptron neural network) to 84.14 (Random Trees) for the testing set. The leading parameters predicting prolonged LOS were operation time, ischemia time, the graft used, the ASA score, the intensive care stay, and the pathological stages. The results revealed that patients who had a higher number of harvested lymph nodes (LN) had a lower probability of recurrence but also a greater LOS. However, patients with prolonged LOS were also at greater risk of recurrence, particularly when fewer (LN) were extracted. Further, LOS was more strongly correlated with the overall number of extracted lymph nodes than with the number of positive lymph nodes or the ratio of positive to overall extracted lymph nodes, indicating that particularly unnecessary lymph node extraction might be associated with prolonged LOS.

Conclusions

The results emphasize the need for a closer follow-up of patients who experience prolonged LOS. Prospective trials are warranted to validate the present results.

Specialty Impact on Patient Outcomes: Paving a Way for an Integrated Approach to Spinal Disorders

Spinal surgical procedures are steadily increasing globally due to broad indications of certain techniques encompassing a wide spectrum of conditions, including degenerative spine disorders, congenital anomalies, spinal metastases, and traumatic spinal fractures. The two specialties, neurosurgery (NS) and orthopedic surgery (OS), both possess the clinical adeptness to perform these procedures. With the advancing focus on comparative effectiveness research, it is vital to compare patient outcomes in spine surgeries performed by orthopedic surgeons and neurosurgeons, given their distinct approaches and training backgrounds to guide hospital programs and physicians to consider surgeon specialty when making informed decisions. Our review of the available literature revealed no significant difference in postoperative outcomes in terms of blood loss, neurological deficit, dural injury, intraoperative complications, and postoperative wound dehiscence in procedures performed by neurosurgeons and orthopedic surgeons. An increase in blood transfusion rates among patients operated by orthopedic surgeons and a longer operative time of procedures performed by neurosurgeons was a consistent finding among several studies. Other findings include a prolonged hospital stay, higher hospital readmission rates, and lower cost of procedures in patients operated on by orthopedic surgeons. A few studies revealed lower sepsis rates unplanned intubation rates and higher incidence of urinary tract infections (UTIs) and pneumonia postoperatively among patient cohorts operated by neurosurgeons. Certain limitations were identified in the studies including the use of large databases with incomplete information related to patient and surgeon demographics. Hence, it is imperative to account for these confounding variables in future studies to alleviate any biases. Nevertheless, it is essential to embrace a multidisciplinary approach integrating the surgical expertise of the two specialties and develop standardized management guidelines and techniques for spinal disorders to mitigate complications and enhance patient outcomes.

The severity of preoperative anemia escalates risk of poor short-term outcomes after lumbar spine fusion

OBJECTIVE

To evaluate how preoperative anemia severity affects 90-day outcomes of spinal fusion surgery.

METHODS

A retrospective cohort study was conducted on adult lumbar fusion patients at a tertiary medical center. Patients were classified by World Health Organization anemia severity definitions for comparisons. Multivariate regression models were created to control for confounding variables, for all primary outcomes of transfusion requirements, non-home discharge, readmissions, complications, and length of stay.

RESULTS

A total of 2582 patients were included: 2.7% with moderate-severe anemia, 11.0% with mild anemia, and 86.3% without anemia. Moderate-severe patients had the longest hospital stay (5.03 days vs 4.14 and 3.59 days, p < 0.001) and highest risk of transfusion (52.2% vs 13.0% vs 2.69%, p < 0.001), non-home discharge (39.1% vs 27.8% vs 15.4%, p < 0.001), readmission (7.25% vs 5.99% vs 3.36%, p = 0.023), and complications (13.0% vs 9.51% vs 6.20%, p = 0.012). On multivariable logistic regression, both patients with mild and moderate-severe anemia had an increased risk of transfusion (OR: 37.3, p < 0.001; OR: 5.25, p < 0.001, respectively) and non-home discharge (OR: 2.00, p = 0.021; OR: 1.71, p = 0.001, respectively) compared to patients without anemia. Anemia severity was not independently associated with complications or 90-day readmission. On multivariable linear regression, mild anemia (β: 0.37, p = 0.001) and moderate-severe anemia (β: 1.07, p < 0.001) were independently associated with length of hospital stay.

CONCLUSION

Patients with moderate-severe preoperative anemia are at increased risk for longer length of stay, transfusions, and non-home discharge. Improved optimization of preoperative anemia may significantly reduce healthcare utilization, and surgeons should consider these risks in preoperative planning.

LEVEL OF EVIDENCE

III.

Preoperative predictors of prolonged hospitalization in patients undergoing lateral lumbar interbody fusion

PURPOSE

We aim to examine the preoperative factors associated with increased postoperative length of stay in patients undergoing LLIF in the hospital setting.

METHODS

Patient demographics, perioperative characteristics, and patient-reported outcome measures (PROMs) were collected from a single-surgeon database. Patients undergoing LLIF in the hospital setting were separated into postoperative LOS <48 h (H) and LOS ≥ 48H. Univariate analysis for preoperative characteristics was utilized to determine covariates for multivariable logistic regression. Multivariable logistic regression was then utilized to determine significant predictors of extended postoperative length of stay. Secondary univariate analysis of inpatient complications, operative, and postoperative characteristics were calculated to determine postoperative factors associated with prolonged hospitalization.

RESULTS

Two-hundred and forty patients were identified with 115 patients' LOS ≥ 48H. Univariate analysis identified age/Charlson Comorbidity Index (CCI) score/gender/insurance type/number of contiguous fused levels/preoperative PROMs of Visual Analog Scale (VAS) back/VAS leg/Patient-Reported Outcomes Measurement Information System (PROMIS-PF)/Oswestry Disability Index (ODI)/degenerative spondylolisthesis diagnoses/foraminal stenosis/central stenosis for multivariable logistic regression. Multivariable logistic regression calculated significant positive predictors of LOS ≥ 48H to be age/3-level fusion/preoperative ODI scores. Negative predictors of LOS ≥ 48H were the diagnosis of foraminal stenosis/preoperative PROMIS-PF/male gender. The secondary analysis determined that patients with longer operative time/estimated blood loss/transfusion/postoperative day 0 and 1 pain and narcotic consumption/complications of altered mental status/postoperative anemia/fever/ileus/urinary retention were associated with prolonged hospitalization.

CONCLUSION

Older patients undergoing LLIF with greater preoperative disability and 3-level fusion were more likely to require prolonged hospitalization. Male patients with higher preoperative physical function and who were diagnosed with foraminal stenosis were less likely to require prolonged hospitalization.

Incidence of Surgical Site Infection Following Lateral Lumbar Interbody Fusion Compared With Posterior/Transforaminal Lumbar Interbody Fusion: A Propensity Score-Weighted Study

STUDY DESIGN

Retrospective cohort study.

OBJECTIVE

The objective of this study was to compare the incidence of surgical site infection (SSI) after lateral lumbar interbody fusion (LLIF) and posterior/transforaminal lumbar interbody fusion ( P /TLIF).

SUMMARY OF BACKGROUND DATA

Previous studies have shown that LLIF can improve neurological symptoms to a comparable degree to P /TLIF. However, data on the risk of SSI after LLIF is lacking compared with P /TLIF.

MATERIALS AND METHODS

The study was conducted under a retrospective cohort design in patients undergoing LLIF or P /TLIF for lumbar degenerative diseases between 2013 and 2020 using a hospital administrative database. We used propensity score overlap weighting to adjust for confounding factors including age, sex, body mass index, comorbidities, number of fusion levels, hospital size, and surgery year. We estimated weighted odds ratios (ORs) and 95% CIs for SSI within 30 days postoperatively.

RESULTS

We compared the risk of SSI between 2874 patients who underwent LLIF and 24,245 patients who received P/TLIF Patients who had received LLIF were at significantly less risk of experiencing an SSI compared with those receiving P/TLIF (0.7% vs. 1.2%; weighted OR: 0.57; 95% CI: 0.36 -0.92; P=0.02). As a secondary outcome, patients who had received LLIF had less risk of transfusion (7.8% vs. 11.8%; weighted OR: 0.63; 95% CI:0.54 -0.73; P <0.001).

CONCLUSIONS

In this large retrospective cohort study of adults undergoing lumbar interbody fusion, LLIF was associated with a significantly lower risk of SSI than P /TLIF. The small, but significantly, decreased risk of SSI associated with LLIF may inform decisions regarding the technical approach for lumbar interbody fusion.

Oblique lumbar interbody fusion versus minimally invasive transforaminal lumbar interbody fusion for the treatment of degenerative disease of the lumbar spine: a systematic review and meta-analysis

This meta-analysis compared the efficacy of oblique lumbar interbody fusion (OLIF) and minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF) in the treatment of lumbar degenerative diseases. A computer search for the published literature on OLIF and MIS-TLIF for the treatment of lumbar degenerative diseases in the PubMed, Web of Science, Embase, CINAHL, MEDLINE, Cochrane Library, and other databases was performed, from which 522 related articles were retrieved and 13 were finally included. Two reviewers independently extracted data from the included studies and analyzed them using RevMan 5.4. The quality of the studies was assessed using the Cochrane systematic analysis and the Newcastle-Ottawa scale. Meta-analysis showed that the blood loss [95% confidence intervals (CI) (- 121.01, - 54.56), [Formula: see text]], hospital stay [95% CI (- 1.98, - 0.85), [Formula: see text]], postoperative fusion rate [95%CI (1.04, 3.60), [Formula: see text]], postoperative disc height [95% CI (0.50, 3.63), [Formula: see text]], and postoperative foraminal height [95% CI (0.96, 4.13), [Formula: see text]] were all better in the OLIF group; however, the complication rates were significantly lower in the MIS-TLIF group [95% CI (1.01, 2.06), [Formula: see text]]. However, there were no significant differences between the two in terms of surgery time, patient satisfaction, or postoperative functional scores. The OLIF group had the advantages of lower blood loss, a shorter hospital stay, a higher postoperative fusion rate, and better recovery of the disc and foraminal heights, whereas MIS-TLIF had a relatively lower complication rate.

Factors contributing to a longer length of stay in adults admitted to a quaternary spinal care center

BACKGROUND

Longer hospital length of stay (LOS) has been associated with worse outcomes and increased resource utilization. However, diagnostic and patient-level factors associated with LOS have not been well studied on a large scale. The goal was to identify patient, surgical and organizational factors associated with longer patient LOS for adult patients at a high-volume quaternary spinal care center.

METHODS

We performed a retrospective analysis of 13,493 admissions from January 2006 to December 2019. Factors analyzed included age, sex, admission status (emergent vs scheduled), ASIA grade, operative vs non-operative management, mean blood loss, operative time, and adverse events. Specific adverse events included surgical site infection (SSI), other infection (systemic or UTI), neuropathic pain, delirium, dural tear, pneumonia, and dysphagia. Diagnostic categories included trauma, oncology, deformity, degenerative, and "other". A multivariable linear regression model was fit to log-transformed LOS to determine independent factors associated with patient LOS, with effects expressed as multipliers on mean LOS.

RESULTS

Mean LOS for the population (SD) was 15.8 (34.0) days. Factors significantly (p < 0.05) associated with longer LOS were advanced patient age [multiplier on mean LOS 1.011/year (95% CI: 1.007-1.015)], emergency admission [multiplier on mean LOS 1.615 (95% CI: 1.337-1.951)], ASIA grade [multiplier on mean LOS 1.125/grade (95% CI: 1.051-1.205)], operative management [multiplier on mean LOS 1.211 (95% CI: 1.006-1.459)], and the occurrence of one or more AEs [multiplier on mean LOS 2.613 (95% CI: 2.188-3.121)]. Significant AEs included postoperative SSI [multiplier on mean LOS 1.749 (95% CI: 1.250-2.449)], other infections (systemic infections and UTI combined) [multiplier on mean LOS 1.650 (95% CI: 1.359-2.004)], delirium [multiplier on mean LOS 1.404 (95% CI: 1.103-1.787)], and pneumonia [multiplier on mean LOS 1.883 (95% CI: 1.447-2.451)]. Among the diagnostic categories explored, degenerative patients experienced significantly shorter LOS [multiplier on mean LOS 0.672 (95%CI: 0.535-0.844), p < 0.001] compared to non-degenerative categories.

CONCLUSION

This large-scale study taking into account diagnostic categories identified several factors associated with patient LOS. Future interventions should target modifiable factors to minimize LOS and guide hospital resource allocation thereby improving patient outcomes and quality of care and decreasing healthcare-associated costs.

Transdiscal instrumentation in single-level lumbosacral fusion for high-grade isthmic pediatric spondylolisthesis: Technical note and review of the literature

Pediatric spondylolisthesis is a common cause of back pain in children, typically managed conservatively with bracing and non-steroidal anti-inflammatory drugs. When posterolateral fusion is performed for refractory pain, pseudarthrosis and implant failure may occur, necessitating reoperation. To improve patient outcomes, there is a need for alternative surgical techniques to effectively manage high-grade isthmic slips. Here, the authors report the case of a child with Meyerding grade III anterolisthesis of L5 on S1 who was treated with a single-level, instrumented fusion using bilateral S1-L5 transdiscal screws, supported with L5-S1 posterolateral instrumentation and arthrodesis. Postoperatively, there was improvement in the patient's symptoms with good clinical and radiographic outcomes. The patient continues to be symptom free with radiographic evidence of hardware stability and bony fusion across the segment. The authors detail a novel surgical technique in children as well as a review of lumbosacral transdiscal screw fixation. Further evidence is required to definitively establish the safety, outcomes, and biomechanical strength of this technique.

How Much Blood Loss Is Too Much for a 1-Level Open Lumbar Fusion?

BACKGROUND

Despite the known association between increased estimated blood loss (EBL) and suboptimal perioperative outcomes, the exact threshold of EBL that impacts outcomes following elective spine surgery remains unknown. In a cohort of patients undergoing elective 1-level open posterior lumbar fusion, we sought to identify EBL thresholds associated with: (1) prolonged length of stay (LOS), (2) postoperative complications, and (3) patient-reported outcomes (PROs).

METHODS

A retrospective, single-center study was performed of patients undergoing elective, 1-level open posterior lumbar fusion with and without interbody fusion between October 2010 and April 2021. The primary exposure variable was EBL. Primary outcomes included: (1) LOS, (2) 30-day complications, and (3) 3-month PROs. Minimum clinically important difference was set at 30% improvement from baseline. For purposes of receiver-operating characteristic curves, LOS was dichotomized as 1 vs ≥2 days.

RESULTS

Of the 2028 patients undergoing posterior lumbar fusion surgery, 1183 underwent 1-level fusions, 763 (64.5%) with interbody fusion and 420 (35.5%) without. With interbody fusion: Median (interquartile range [IQR]) EBL was 350 mL (200-600), and median (IQR) LOS was 2 days (2-3). A positive linear association was found between EBL and LOS (P < 0.001) but not with PROs. EBL above 275 mL was associated with LOS beyond postoperative day 1 (POD1) (area under the curve [AUC] = 0.73, 95% CI 0.68-0.78, P < 0.001), with no significant association with overall complications or PROs. Without interbody fusion: Median EBL (IQR) was 300 mL (150-500), and median (IQR) LOS was 3 days (2-4). A positive linear association was found between EBL and LOS (P < 0.001) but not with PROs. EBL above 238 mL was associated with LOS beyond POD1 (AUC = 0.78, 95% CI 0.71-0.85, P < 0.001), with no impact on overall complications or PROs.

CONCLUSIONS

In patients undergoing 1-level posterior lumbar fusion, EBL volumes greater than 275 and 238 mL in patients with and without interbody fusion, respectively, were associated with increased LOS beyond POD1. No effect was found regarding 30-day complications and 3-month PROs. Although EBL did not directly impact complications or PROs, surgeons may expect longer LOS when higher EBL is reported.

CLINICAL RELEVANCE

EBL above 275 mL with an interbody and 238 mL without an interbody were associated with prolonged LOS beyond POD1 in 1-level open lumbar fusion.

LEVEL OF EVIDENCE: 3

Intraoperative Hypotension Is Not Correlated With Acute Kidney Injury During Spinal Fusion Surgery

BACKGROUND

Intraoperative hypotension (IOH) has been found to be associated with organ damage, including cardiac injury and acute kidney injury (AKI). However, to our knowledge, this relationship has not been studied in a neurosurgery-specific patient population. In this report, we review our institutional experience to understand the magnitude of association between IOH in spinal fusion operations and incidence of postoperative AKI.

METHODS

This retrospective cohort study included 910 patients who underwent posterior spinal fusion procedures performed in the prone position. Intraoperative variables collected and analyzed include minute-by-minute mean arterial pressure (MAP) from an arterial catheter, intermittent blood pressure cuff readings, volume of administered intravenous fluids, urine output, and all relevant vitals and administered medications. The electronic medical record was queried for additional patient data. IOH was defined as MAP <65 mm Hg for greater than 10 minutes. The primary endpoints of the study were presence and staging of AKI ( [Kidney Disease: Improving Global Outcomes] consensus classification), postoperative ileus, and postoperative troponin leak.

RESULTS

Using a partial correlation analysis, no association was found between IOH metrics (IOH occurrence, IOH duration >10 minutes, and total IOH time) and any outcome metrics, including AKI, except for vasopressor usage and estimated blood loss. Patient age at surgery was not associated with any outcome variables. The lack of association between IOH and AKI contrasts with existing literature; this could be due to underlying differences in our patient population or could highlight a more complex relationship between IOH and AKI than previously understood.

CONCLUSION

Occurrence and duration of IOH were not associated with AKI, postoperative ileus, troponin leak, length of stay, or any other major outcome variables in spinal fusion patients.

CLINICAL RELEVANCE

These findings depart from previous literature showing a correlation between IOH and AKI and provide level 3 evidence clinically relevant to spinal surgery. Further research is needed to better understand the exact nature of this relationship.

LEVEL OF EVIDENCE: 3

Predicting prolonged length of stay in patients undergoing transforaminal lumbar interbody fusion

BACKGROUND

With growing emphasis on high-value care, many institutions have been working on improving surgical efficiency, quality, and complication reduction. Unfortunately, data are limited regarding perioperative factors that may influence length of stay (LOS) following transforaminal lumbar interbody fusion (TLIF). We sought to design a predictive algorithm that determined patients at risk of prolonged LOS after TLIF. The goal was to identify patients who would benefit from preoperative intervention aimed to reduce LOS.

METHODS

We conducted a review of perioperative data for patients who underwent TLIF between 2014 and 2019. Univariate and multivariate stepwise regression models were used to analyze risk factor effects on postoperative LOS.

RESULTS

Two hundred and sixty-nine patients were identified (57.2% women). Mean age at surgery was 61.7 ± 12.3 years. Mean postoperative LOS was 3.08 ± 1.54 days. In multivariate analysis, American Society of Anesthesiologists class (odds ratio [OR] = 1.441, 95% confidence interval [CI] 1.321-1.571), preoperative functional status (OR = 1.237, 95% CI 1.122-1.364), Oswestry Disability Index (OR = 1.010, 95% CI 1.004-1.016), and estimated blood loss (OR = 1.050, 95% CI 1.003-1.101) were independent risk factors for postoperative LOS ≥ 5 days. The final model had an area under the curve of 0.948 with good discrimination and was implemented in the form of an online calculator ( https://spine.shinyapps.io/TLIF_LOS/ ).

CONCLUSION

The prediction tool derived can be useful for assessing likelihood of prolonged LOS in patients undergoing TLIF. With external validation, this calculator may ultimately assist healthcare providers in identifying patients at risk for prolonged hospitalization so preoperative interventions can be undertaken to reduce LOS, thus reducing resource utilization.

Performance of Artificial Intelligence-Based Algorithms to Predict Prolonged Length of Stay after Lumbar Decompression Surgery

Background: Decompression of the lumbar spine is one of the most common procedures performed in spine surgery. Hospital length of stay (LOS) is a clinically relevant metric used to assess surgical success, patient outcomes, and socioeconomic impact. This study aimed to investigate a variety of machine learning and deep learning algorithms to reliably predict whether a patient undergoing decompression of lumbar spinal stenosis will experience a prolonged LOS. Methods: Patients undergoing treatment for lumbar spinal stenosis with microsurgical and full-endoscopic decompression were selected within this retrospective monocentric cohort study. Prolonged LOS was defined as an LOS greater than or equal to the 75th percentile of the cohort (normal versus prolonged stay; binary classification task). Unsupervised learning with K-means clustering was used to find clusters in the data. Hospital stay classes were predicted with logistic regression, RandomForest classifier, stochastic gradient descent (SGD) classifier, K-nearest neighbors, Decision Tree classifier, Gaussian Naive Bayes (GaussianNB), support vector machines (SVM), a custom-made convolutional neural network (CNN), multilayer perceptron artificial neural network (MLP), and radial basis function neural network (RBNN) in Python. Prediction accuracy and area under the curve (AUC) were calculated. Feature importance analysis was utilized to find the most important predictors. Further, we developed a decision tree based on the Chi-square automatic interaction detection (CHAID) algorithm to investigate cut-offs of predictors for clinical decision-making. Results: 236 patients and 14 feature variables were included. K-means clustering separated data into two clusters distinguishing the data into two patient risk characteristic groups. The algorithms reached AUCs between 67.5% and 87.3% for the classification of LOS classes. Feature importance analysis of deep learning algorithms indicated that operation time was the most important feature in predicting LOS. A decision tree based on CHAID could predict 84.7% of the cases. Conclusions: Machine learning and deep learning algorithms can predict whether patients will experience an increased LOS following lumbar decompression surgery. Therefore, medical resources can be more appropriately allocated to patients who are at risk of prolonged LOS.

Influence of Preoperative Handgrip Strength on Length of Stay after Lumbar Fusion Surgery

Preoperative sarcopenic status can affect length of hospital stay and patient outcomes after surgery. The aim of this study was to investigate the impacts of preoperative handgrip strength (HGS) on length of stay (LOS) and outcomes after lumbar fusion surgery. HGS was measured preoperatively, and the cut-off value for low HGS was <28 kg for men and <18 kg for women. Perioperative patient outcomes were compared between patients with low and normal HGS. A total of 225 patients, consisting of 86 and 139 patients in the low and normal HGS groups, respectively, fully satisfied the study criteria for analysis. A longer LOS (median 10 vs. 8 days, p = 0.013) and a higher incidence of serious postoperative complications (15.1 vs. 3.6%, p = 0.002) were observed in the low HGS group. In the multivariate analysis, a low HGS (odds ratio (OR) = 1.917, 95% confidence interval (CI) = 1.046−3.513, p = 0.035) was significantly associated with a longer LOS after surgery. Preoperative HGS below the reference values by sex appeared to be an independent factor associated with longer LOS after lumbar fusion surgery.

Advanced Age Does Not Impact Outcomes After 1-level or 2-level Lateral Lumbar Interbody Fusion

STUDY DESIGN

This was a retrospective comparative study.

OBJECTIVE

The objective of this study was to assess the effect of increased age on perioperative and postoperative complication rates, reoperation rates, and patient-reported pain and disability scores after lateral lumbar interbody fusion (LLIF).

SUMMARY OF BACKGROUND DATA

LLIF was developed to minimize soft tissue trauma and reduce the risk of vascular injury; however, there is little evidence regarding the effect of advanced age on outcomes of LLIF.

METHODS

Patients who underwent LLIF from 2009 to 2019 at one institution with a minimum 6-month follow-up were retrospectively reviewed. Patients less than 18 years old with musculoskeletal tumor or trauma were excluded. The primary outcome was the preoperative to postoperative change in the Numeric Pain Rating Scale (NPRS) for back pain. Operative time, estimated blood loss, length of stay, perioperative and 90-day complications, unplanned readmissions, reoperations, and change in Oswestry Disability Index were also evaluated. Relationships with age were assessed both with age as a continuous variable and segmenting by age below 70 versus 70+.

RESULTS

In total, 279 patients were included. The median age was 65±13 years and 159 (57%) were female. Age was not related to improvements in back NPRS and Oswestry Disability Index. Operative time, estimated blood loss, length of stay, perioperative and 90-day complications, unplanned readmissions, reoperations, and radiographic fusion rate also were not related to age. After multivariable risk adjustment, increasing age was associated with greater improvements in back NPRS. The decrease in back NPRS was 0.68 (95% confidence interval: 0.14, 1.22; P=0.014) points greater for every 10-year increase in age. Age was not associated with rates of complication, readmission, or reoperation.

CONCLUSIONS

LLIF is a safe and effective procedure in the elderly population. Advanced age is associated with larger improvements in preoperative back pain. Surgeons should consider the benefits of LLIF and other minimally invasive techniques when evaluating elderly candidates for lumbar fusion.

LEVEL OF EVIDENCE

Level III.

Variations in LOS and its main determinants overtime at an academic spinal care center from 2006-2019

Objectives

Efforts to safely reduce hospital LOS while maintaining quality outcomes and patient satisfaction are paramount. The primary goal of this study was to assess trends in LOS at a high-volume quaternary care spine center. Secondary goals were to assess trends in factors most associated with prolonged LOS.

Methods

This is a prospective study of all consecutive patients admitted from January 2006 to December 2019. Data included demographics, diagnostic category (degenerative, oncology, deformity, trauma, other), LOS (mean, median, interquartile range, standard deviation, defined as days from admission to discharge), and in-hospital adverse events.

Results

A total of 13,493 patients were included. Overall LOS has not changed over time with an overall median of 6.3 days (p = 0.451). Median LOS significantly increased for patients treated for degenerative pathology from 2.2 days in 2006 to 3.2 days in 2019 (p = 0.019). LOS has not changed for patients treated for deformity (overall median 6.8 days, p = 0.411), oncology (overall median 11.0 days, p = 0.051), or trauma (overall median 11.8 days, p = 0.582). Emergency admissions increased 3.2%/year for degenerative pathologies (p =  < 0.001). Mean age has increased from 48.4 years in 2006 to 58.1 years in 2019 (p =  < 0.001). This trend was observed in the deformity, degenerative and trauma group, not for patients treated for oncological disease. More adverse events were significantly associated with increasing age.

Conclusion

This is the first North American study to comprehensively analyze trends in LOS for spinal surgery overtime in an academic center. Overall, LOS has not changed from 2006–2019. Various factors that influence LOS appear to have balanced each other. It may also be explained by the changing epidemiology of both elective and emergency surgeries. These findings provide opportunities for intervention and improvement, targeted at the geriatric population, to reduce length of hospitalization.

Single-level TLIF Versus LLIF at L4-5: A Comparison of Patient-reported Outcomes and Recovery Ratios

INTRODUCTION

Both transforaminal lumbar interbody fusion (TLIF) and lateral lumbar interbody fusion (LLIF) are suitable for achievement of lumbar arthrodesis. Comparative studies have observed complications and outcomes without stratification by lumbar level. This study aims to assess patient-reported outcome measures (PROMs) and recovery in TLIF and LLIF at L4-5.

METHODS

Patients undergoing primary, elective, single-level, TLIF or LLIF procedures at L4-5 were grouped. Demographics, perioperative characteristics, and postoperative complication rates were collected. PROMs included Patient-Reported Outcome Measurement Information System Physical Function, visual analog scale (VAS) back and leg, Oswestry Disability Index, and 12-Item Short-Form Physical Component Summary, and Mental Component Summary and were collected at preoperative, 6-week, 12-week, 6-month, 1-year, and 2-year time points. Delta values and recovery ratios (RRs) were calculated for all PROMs at all time points. Demographics, perioperative characteristics, and postoperative complications were compared using chi-squared and Student t-test for categorical and continuous variables, respectively. Differences in mean PROMs, delta values, and RR at each time point were evaluated using unpaired Student's t-test.

RESULTS

Three hundred sixty TLIF and 46 LLIF patients were included. Most (54.3%) were men, mean age 56.3 years, and mean body mass index 30.8 kg/m2. Body mass index and insurance significantly differed (P ≤ 0.045, all). TLIF showed significantly greater mean operative time, length of stay, and postoperative narcotic consumption (P < 0.033, all) and greater postoperative nausea/vomiting (P = 0.004). No preoperative PROMs significantly differed. TLIF cohort had significantly greater VAS back at 6 months and VAS leg at 12 weeks and 6 months (P < 0.034, all). No mean delta PROMs or RRs significantly differed.

DISCUSSION

LLIF demonstrated significantly reduced length of stay, postoperative narcotic consumption, and postoperative nausea/vomiting and significantly improved VAS back at 6 months and VAS leg at 12 weeks and 6 months versus TLIF. Although 2-year PROMs and RRs did not significantly differ, our findings may suggest improved midterm follow-up pain scores for LLIF patients.

Supervised Machine Learning for Predicting Length of Stay After Lumbar Arthrodesis: A Comprehensive Artificial Intelligence Approach

INTRODUCTION

Few studies have evaluated the utility of machine learning techniques to predict and classify outcomes, such as length of stay (LOS), for lumbar fusion patients. Six supervised machine learning algorithms may be able to predict and classify whether a patient will experience a short or long hospital LOS after lumbar fusion surgery with a high degree of accuracy.

METHODS

Data were obtained from the National Surgical Quality Improvement Program between 2009 and 2018. Demographic and comorbidity information was collected for patients who underwent anterior, anterolateral, or lateral transverse process technique arthrodesis procedure; anterior lumbar interbody fusion (ALIF); posterior, posterolateral, or lateral transverse process technique arthrodesis procedure; posterior lumbar interbody fusion/transforaminal lumbar interbody fusion (PLIF/TLIF); and posterior fusion procedure posterior spine fusion (PSF). Machine learning algorithmic analyses were done with the scikit-learn package in Python on a high-performance computing cluster. In the total sample, 85% of patients were used for training the models, whereas the remaining patients were used for testing the models. C-statistic area under the curve and prediction accuracy (PA) were calculated for each of the models to determine their accuracy in correctly classifying the test cases.

RESULTS

In total, 12,915 ALIF patients, 27,212 PLIF/TLIF patients, and 23,406 PSF patients were included in the algorithmic analyses. The patient factors most strongly associated with LOS were sex, ethnicity, dialysis, and disseminated cancer. The machine learning algorithms yielded area under the curve values of between 0.673 and 0.752 (PA: 69.6% to 80.1%) for ALIF, 0.673 and 0.729 (PA: 66.0% to 81.3%) for PLIF/TLIF, and 0.698 and 0.749 (PA: 69.9% to 80.4%) for PSF.

CONCLUSION

Machine learning classification algorithms were able to accurately predict long LOS for ALIF, PLIF/TLIF, and PSF patients. Supervised machine learning algorithms may be useful in clinical and administrative settings. These data may additionally help inform predictive analytic models and assist in setting patient expectations.

LEVEL III

Diagnostic study, retrospective cohort study.

Trends in the numbers of spine surgeries and spine surgeons over the past 15 years

The purpose of this study is to examine trends in spine surgeries at ten facilities over 15 years, and to analyze relationships with the number of spine surgeons at these facilities. The subjects were patients who underwent spine surgery at the ten facilities from 2003 to 2017. Data were collected every year via a questionnaire designed to obtain clinicopathological and surgical information. There were 37,601 spine surgeries (60.2% male) recorded in the registry at 9 facilities in the Nagoya Spine Group (NSG) between 2003 and 2017, with an increase in the annual number of surgeries by 2.4 times over 15 years. On the other hand, the number of spine surgeons has increased by just under 1.5 times. Instrumentation surgeries increased from 959 in 2003 to 2,276 in 2017 (2.3 times). There was a particularly marked increase in surgeries for spinal degenerative disease from 1,075 in 2003 to 2,821 in 2017 (2.6 times). The number of surgeries performed per surgeon increased from 61.4 in 2003 to 102.8 in 2017, while the average number of spine surgeons per hospital increased from 2.6 in 2003 to 3.7 in 2017. In conclusion, with heavier burden on spine surgeons and the major changes in the spine surgery environment, training and increasing surgeons with advanced expertise and skills will become increasingly important.

Minimally invasive transforaminal lumbar interbody fusion using the biportal endoscopic techniques versus microscopic tubular technique

BACKGROUND CONTEXT

Minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF) with microscopic tubular technique is an established surgical procedure with several potential advantages, including decreased surgical-related morbidity, reduced length of hospital stay, and accelerated early rehabilitation. A recently introduced biportal endoscopic technique for spine surgery presents familiar surgical anatomy and can be conducted using a conventional approach with a minimal footprint; it is also applicable to TLIF.

PURPOSE

To compare the clinical and radiological outcomes of biportal endoscopic technique transforaminal lumbar interbody fusion (BE-TLIF) and microscopic tubular technique transforaminal lumbar interbody (MT-TLIF) in patients with single- or two-segment lumbar spinal stenosis with or without spondylolisthesis.

STUDY DESIGN

A retrospective cohort study.

PATIENT SAMPLE

One hundred two participants with neurogenic intermittent claudication or lumbar radiculopathy with single- or two-level lumbar spinal stenosis with or without spondylolisthesis.

OUTCOME MEASURES

Clinical outcomes were assessed using the visual analog scale (VAS) score for the back and leg pain, Oswestry Disability Index (ODI), and the Short Form-36 health survey Questionnaire (SF-36). Demographic data, operative data (total operation time, estimated blood loss, amount of surgical drain, postoperative transfusion, and length of hospital stay), and laboratory results (plasma hemoglobin, serum creatine phosphokinase, and C-reactive protein) were also evaluated. The fusion rate was assessed using the Bridwell interbody fusion grading system. Postoperative complications were also noted.

METHODS

Patients were divided into two groups: group A (BE-TLIF) and group B (MT-TLIF). The clinical outcomes, including VAS-Back and VAS-Leg, ODI, and SF-36 scores, were evaluated at 1 month, 6 months, and 1 year after surgery. Differences in demographics, operative data, and the laboratory and radiological results were assessed between the two groups. The fusion rate was assessed using standard standing lumbar radiographs and computed tomography scans conducted 1 year after surgery.

RESULTS

Seventy-nine patients were analyzed in this study, 47 from group A and 32 from group B. Demographic and operative data were comparable for both the groups. The VAS-Back and SF-36 scores were more significantly improved in group A than in group B at 1 month after surgery. However, there were no significant differences between groups for the mean VAS-Back, VAS-Leg, ODI, and SF-36 scores at 1year after the surgery. Although the total operation time was significantly longer in group A, the estimated blood loss and the amount of surgical drainage was significantly higher in group B (p < .001). There were no between-group differences for the fusion rate and postoperative complications.

CONCLUSION

Both BE-TLIF and MT-TLIF provided equivalent and favorable clinical outcomes and fusion rates. Further large-scale, randomized, controlled trials with long-term follow-ups are warranted.

The Impact of Nurse Navigator-Led Preoperative Education on Hospital Outcomes Following Posterolateral Lumbar Fusion Surgery

Beyond the spine-specific pathology, patient factors such as associated medical and psychosocial conditions, understanding of the treatment process, and the degree of patient activation-defined as the ability of the individual to utilize the available information and actively engage in making their healthcare decisions-can influence outcomes after posterolateral lumbar fusion (PLF) surgery. A retrospective observational cohort study of 177 patients undergoing PLF at a single institution was conducted. Patient demographics, medical and psychosocial risk factors, and outcomes were compared between patients who attended a nurse navigator-led group preoperative education course and those who did not. Patients attending the course were younger, more likely to undergo one-level fusion, less likely to undergo 5- or more-level fusion, and had less comorbidity burden as measured by the hierarchical condition categories score. No differences in psychosocial risk factors were observed between groups. Course attendees had a significantly shorter length of stay (2.12 vs. 2.60 days, p = .042) and decreased average hospital cost (U.S. $10,149 vs. U.S. $14,792, p < .001) than those who did not attend; no differences in other outcomes were observed. After controlling for differences in risk factors, patients enrolled in a preoperative education course demonstrated a statistically significant reduction in hospital cost (β=-4,143, p < .001). Preoperative education prior to PLF surgery may reduce hospital cost, possibly through increased patient activation. Given the relatively high prevalence of psychosocial risk factors in this and similar patient populations, optimizing patient activation and engagement is important to achieve high value care. Based on our findings, nurse navigator-led preoperative education appears to be valuable in this patient population and should be included in enhanced recovery protocols.

Machine Learning Prediction of Length of Stay in Adult Spinal Deformity Patients Undergoing Posterior Spine Fusion Surgery

(1) Background: Length of stay (LOS) is a commonly reported metric used to assess surgical success, patient outcomes, and economic impact. The focus of this study is to use a variety of machine learning algorithms to reliably predict whether a patient undergoing posterior spinal fusion surgery treatment for Adult Spine Deformity (ASD) will experience a prolonged LOS. (2) Methods: Patients undergoing treatment for ASD with posterior spinal fusion surgery were selected from the American College of Surgeon's NSQIP dataset. Prolonged LOS was defined as a LOS greater than or equal to 9 days. Data was analyzed with the Logistic Regression, Decision Tree, Random Forest, XGBoost, and Gradient Boosting functions in Python with the Sci-Kit learn package. Prediction accuracy and area under the curve (AUC) were calculated. (3) Results: 1281 posterior patients were analyzed. The five algorithms had prediction accuracies between 68% and 83% for posterior cases (AUC: 0.566-0.821). Multivariable regression indicated that increased Work Relative Value Units (RVU), elevated American Society of Anesthesiologists (ASA) class, and longer operating times were linked to longer LOS. (4) Conclusions: Machine learning algorithms can predict if patients will experience an increased LOS following ASD surgery. Therefore, medical resources can be more appropriately allocated towards patients who are at risk of prolonged LOS.

Application of Cooperative Game Theory Principles to Interpret Machine Learning Models of Nonhome Discharge Following Spine Surgery

STUDY DESIGN

Retrospective analysis of prospectively acquired data.

OBJECTIVE

The aim of this study was to identify interaction effects that modulate nonhome discharge (NHD) risk by applying coalitional game theory principles to interpret machine learning models and understand variable interaction effects underlying NHD risk.

SUMMARY OF BACKGROUND DATA

NHD may predispose patients to adverse outcomes during their care. Previous studies identified potential factors implicated in NHD; however, it is unclear how interaction effects between these factors contribute to overall NHD risk.

METHODS

Of the 11,150 reviewed cases involving procedures for degenerative spine conditions, 1764 cases (15.8%) involved NHD. Gradient boosting classifiers were used to construct predictive models for NHD for each patient. Shapley values, which assign a unique distribution of the total NHD risk to each model variable using an optimal cost-sharing rule, quantified feature importance and examined interaction effects between variables.

RESULTS

Models constructed from features identified by Shapley values were highly predictive of patient-level NHD risk (mean C-statistic = 0.91). Supervised clustering identified distinct patient subgroups with variable NHD risk and their shared characteristics. Focused interaction analysis of surgical invasiveness, age, and comorbidity burden suggested age as a worse risk factor than comorbidity burden due to stronger positive interaction effects. Additionally, negative interaction effects were found between age and low blood loss, indicating that intraoperative hemostasis may be critical for reducing NHD risk in the elderly.

CONCLUSION

This strategy provides novel insights into feature interactions that contribute to NHD risk after spine surgery. Patients with positively interacting risk factors may require special attention during their hospitalization to control NHD risk.Level of Evidence: 3.

Surgeon experience influences patient characteristics and outcomes in spine deformity surgery

PURPOSE

To characterize differences in patient demographics and outcomes by surgeon experience in a cohort of patients undergoing adult spinal deformity surgery.

METHODS

Patients undergoing degenerative spinal deformity were included. Patients whose surgeons graduated from fellowship ≤ 5 years prior to surgery versus > 5 years were compared. Multivariable linear and logistic regression, controlling for age, sex, comorbidity burden, number of segments fused, blood loss and operative time were used to evaluate differences in outcomes. Characteristics of operative invasiveness were plotted against surgeons' level of experience, and trends in these measures were assessed with univariate linear regression.

RESULTS

Three-hundred sixty-three patients were included. 147 patients' surgeons had ≤ 5 years of experience. Patient demographics were evenly matched. Patients with junior surgeons had more pre-existing medical complications, and senior surgeons were less likely to take care of patients with Medicare/Medicaid (p < 0.001). Junior surgeons were more likely to operate on non-elective patients (p < 0.001). Patients of junior surgeons received larger fusions (9.6 vs. 7.6 segments fused, p < 0.001). There were no differences in complication rates or death. Patients of junior surgeons had longer overall length of stays (p = 0.037) and higher rates of nonhome discharge (OR 2.0, p < 0.001), 30- and 90-day (p < 0.005) ED visits, and higher costs (+ $8548, 95% CI: $1596 to $15,502; p = 0.016).

CONCLUSION

Junior surgeons tend to perform more extensive deformity operations on more medically complex patients compared to senior surgeons, associated with higher costs and more resource utilization than senior surgeons.

Drivers of Prolonged Hospitalization Following Spine Surgery: A Game-Theory-Based Approach to Explaining Machine Learning Models

BACKGROUND

Understanding the interactions between variables that predict prolonged hospital length of stay (LOS) following spine surgery can help uncover drivers of this risk in patients. This study utilized a novel game-theory-based approach to develop explainable machine learning models to understand such interactions in a large cohort of patients treated with spine surgery.

METHODS

Of 11,150 patients who underwent surgery for degenerative spine conditions at a single institution, 3,310 (29.7%) were characterized as having prolonged LOS. Machine learning models predicting LOS were built for each patient. Shapley additive explanation (SHAP) values were calculated for each patient model to quantify the importance of features and variable interaction effects.

RESULTS

Models using features identified by SHAP values were highly predictive of prolonged LOS risk (mean C-statistic = 0.87). Feature importance analysis revealed that prolonged LOS risk is multifactorial. Non-elective admission produced elevated SHAP values, indicating a clear, strong risk of prolonged LOS. In contrast, intraoperative and sociodemographic factors displayed bidirectional influences on risk, suggesting potential protective effects with optimization of factors such as estimated blood loss, surgical duration, and comorbidity burden.

CONCLUSIONS

Meticulous management of patients with high comorbidity burdens or Medicaid insurance who are admitted non-electively or spend clinically indicated time in the intensive care unit (ICU) during their hospitalization course may be warranted to reduce their risk of unanticipated prolonged LOS following spine surgery.

LEVEL OF EVIDENCE

Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.