Faculty Bibliography

BACKGROUND CONTEXT: Artificial intelligence (AI) has enhanced the orthopedic surgical tool kit by introducing a broad range of analytical advances that may offer enhanced preoperative planning, intraoperative robotic or navigational guidance and prediction of postoperative complications. However, there remains a paucity of literature in regards to the utility of AI in adult spinal deformity (ASD)-corrective surgery. PURPOSE: To access the impact of AI utilization on ASD-corrective surgery outcomes. STUDY DESIGN/SETTING: Retrospective cohort. PATIENT SAMPLE: A total of 689 ASD patients. OUTCOME MEASURES: Artificial intelligence, perioperative complications, radiographic outcomes METHODS: Operative ASD patients with complete baseline (BL) and 2-year (2Y) radiographic/HRQL data were stratified by AI-based utilization and robotic or navigational assistance in pre- and peri-operative course (AI+) or not (AI-). Corrections were based on AI models linked to age, proportional alignment and frailty status algorithms to predict outcomes, junctional failure and thoracic compensations. Means comparison tests and regression analysis assessed differences between patient groups.
RESULT(S): There were 158 patients included (57 AI+, 101 AI-). The cohort was 50% female, mean age of 58.8 yrs, BMI 31.6 kg/m2, CCI 3.9 and 6.6 levels fused. At baseline, patient groups were comparable in terms of BL radiographic parameters, all p <.05. Surgically, AI+ had significantly shorter operative times and EBL than AI-, both p <.05. AI+ had more combined approaches and less osteotomies overall, both p <.05. Postoperatively, AI+ patients were noted to have significantly improved segmental alignment in terms of decreased PT (p=0.006), and improved global alignment per decreased TPA and SVA by 2Y, both p < 0.05. Compared to AI-, AI+ patients had a lower overall complication rate by 2Y (28.1% vs 47.5%), p < 0.05. in a multivariate analysis controlling for age, CCI, and invasiveness, AI+ patients were 61.6% less likely to experience a perioperative complication (OR.384 [CI.149-.989], p=0.047).
CONCLUSION(S): This study demonstrates that when using artificial intelligence-based technologies, patients demonstrated lower intraoperative invasiveness, increased likelihood of reaching radiographic alignment targets, and decreased complication rates specifically in the perioperative period. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Evidence on long-term surgical recovery in adult spinal deformity (ASD) patients who are both frail and have an invasive procedure is limited. PURPOSE: This study aims to evaluate frail and invasive patients with 5-year recovery kinetics. We hypothesize that patients who are both frail and have invasive surgeries will have adverse postoperative recovery kinetics STUDY DESIGN/SETTING: Retrospective review. PATIENT SAMPLE: A total of 133 ASD patients with complete HRQOL data at preoperative, 1-year, 2-year, 5-year follow-up were included. OUTCOME MEASURES: Integrated Health State Scores (IHS) in Oswestry Disability Index (ODI), Short-form 36 (SF-36) physical (PCS) and mental (MCS) component score, and Scoliosis Research Society (SRS)-22r measures METHODS: ASD-FI scores were used to stratify non-frail (0.3) patients. ASD-SR scores were used to stratify low invasive (90) surgeries. Using ASD-FI and ASD-SR, patients were separated into four cohorts: non-frail low invasive (NFLI), frail low invasive (FLI), non-frail high invasive (NFHI), and frail high invasive (FHI). HRQOLs at 1 year, 2 years, and 5 years were normalized against preoperative values. AUC was calculated across time points to generate an integrated health state score (IHS). Multivariable linear regression was used to compare IHS scores of FLI, NFHI, and FHI to NFLI while controlling for age, gender, comorbidity, and radiographic alignment.
RESULT(S): Of 633 eligible ASD patients, 339 had 5-year follow-up. Of those, 125 patients with complete HRQOL data at preoperative, 1-year, 2-year, and 5-year visits were included; 27.2% (34) were NFLI, 20.0% (25) were FLI, 26.4% (33) were NFHI, and 26.4% (33) were FHI. Using NFLI as the referent, FLI and NFHI did not have differences in ODI, MCS, PCS or SRS-22r IHS scores (P >0.05). On multivariable analysis of integrated health scores, FHI had higher MCS (7.6 vs 5.47; P=0.0188), SRS activity (6.97 vs 5.67; P=0.0004), SRS pain (8.49 vs 6.4; P=0.001), SRS appearance (8.97 vs 6.81; P=0.0014), SRS satisfaction (11.71 vs 7.97; P=0.0033), and SRS total (7.49 vs. 6.09; P=0.0002), indicating more improved recovery over a 5-year period. Patients who were FHI had higher rates of complications (P <0.05).
CONCLUSION(S): Despite having more complications, patients who were frail and underwent more invasive surgeries were more likely to have greater overall improvement in activity, pain, and satisfaction over a 5-year period relative to preoperative baseline. Our results suggest that frailty in combination with invasiveness do not hinder long-term postoperative recovery kinetics, in comparison to frailty or invasiveness alone. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Adaptation of age-adjusted alignment has gained popularity for its correlation with clinical improvement and lowering rates of proximal junctional kyphosis. Age-adjusted parameters correlate with outcomes. However, frailty may be a better predictor of each following ASD surgery. PURPOSE: To adjust the Sagittal Age-Adjusted Score (SAAS) to accommodate frailty in alignment considerations will increase the predictability of clinical outcomes and junctional failure. STUDY DESIGN/SETTING: Retrospective. PATIENT SAMPLE: A total of 689 ASD Patients. OUTCOME MEASURES: Mechanical complications, PJF, PJK and ODI.
METHOD(S): Included: surgical ASD pts with 2Y data. Frailty assessed by ASD modified Frailty Index (ASD-mFI).
Outcome(s): proximal junctional kyphosis (PJK) and failure (PJF), major mechanical complications, and Smith et al Best Clinical Outcome (BCO), defined as ODI 4.5. Linear regression analysis established a 6W score based on SAAS component scores, frailty and ODI US norms per published by Lafage et al. Logistic regression followed by conditional inference tree (CIT) analysis generated categorical thresholds. Logistic regression analysis controlling for age, baseline deformity and revision status generated odds ratios for the continuous score. Thirty percent of the cohort was used as a random sample for internal validation.
RESULT(S): There were 412 pts included. BL frailty categories: 57% not frail, 30% frail and 14% severely frail. Overall, by 2Y, 39% of patients developed PJK, 8% PJF, 21% mechanical complications, 22% underwent reoperation and 15% met BCO. SAAS only correlated with development of PJF. The ASD-mFI demonstrated correlation with all outcomes except PJK (all p1.4, offset: 0.75-1.4, sseverely offset:.05). Internal validation saw these outcomes maintain significance between categories, with significant adjusted correlation to meeting BCO (OR: 3.8, 1.1-13.5; p=.037).
CONCLUSION(S): Consideration of physiologic age, in addition to chronological age, may be beneficial in management of operative goals to maximize clinical outcomes while minimizing junctional failure. This combination enables the spine surgeon to fortify a surgical plan for even the most challenging patients undergoing adult spinal deformity surgery. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: The Dubousset Functional Test (DFT) is a novel functional assessment tool. It has been shown that patients with spinal pathology take significantly longer to complete the DFT than do control patients. There has been a recent focus in the literature on using functional assessment to predict patient outcomes. PURPOSE: To investigate the relationship between patients' performance of the novel four-component functional test proposed by Dubousset, sagittal alignment and patient-reported outcome measures PROMs. STUDY DESIGN/SETTING: Prospective, single center study. PATIENT SAMPLE: Patients presenting to a single institution for lumbar degenerative disease or spinal deformity. OUTCOME MEASURES: The correlation between time to complete DFT, radiographic measures of sagittal alignment and PROMs.
METHOD(S): This is a prospective, single-center study including primary patients who presented for evaluation of lumbar degenerative disease and spinal deformity. DFT is a test which assesses 4 domains: Up and Walking Test (UWT), Steps Test (ST), Down and Sitting Test (DST), Dual Tasking Test (DTT). The UWT asks the patient to sit-to-stand unassisted, walk 5 m, and sit unassisted. ST asks the patient to ascend 3 steps, turn, and descend 3 steps. DST asks the patient to sit from standing, and then stand from sitting, both unassisted. Finally, DTT asks the patient to walk 5 m and back while counting down from 50 by 2. Each test was timed, and performance was scored in seconds required to finish the test. Spinopelvic radiographs and PROMs (EQ5D, ODI, SF12) were collected. DFT performance was stratified by 40th (Fast) and 60th (Slow) percentile cutoffs. Radiographic sagittal parameters were compared between groups. Regression models were built to predict DFT domains performances using alignment parameters.
RESULT(S): A total of 55 patients were included (52y, 66%F, BMI 25.6). Patients in all DFT Slow domains had significantly worse ODI (all above 40), EQ5D (all below 0.5), and worse SF12_PCS (all below 32) except DTT Slow patients who had a comparable PCS to Fast. UWT Slow patients had significantly greater PI-LL (15 vs 2), lower LL (45 vs 60), and greater PT (22 vs 12); DTT Slow had greater PT (22 vs 11); DST Slow had greater PT (23 vs 12), and greater PI-LL (13 vs 1); ST and DTT slow patients had comparable radiographic parameters to ST and DTT Fast; however, they were older (59 vs 43 ys) (all p <0.05). Greater PT correlated with longer time to perform UWT and DST (r=0.451, r=0.488, respectively, p < 0.05). Greater PI-LL correlated with longer UWT (r=0.349) and loss of LL with longer ST (r=0.416), all p <0.05. Notably, loss of L4-S1 correlated with slower UWT (r=0.377, p <0.05). Regression models revealed that PT, PI-LL and SVA together predict UWT with r=0.472, DST with r=0.370, DTT with r=0.310, and ST with r=0.149.
CONCLUSION(S): The Dubousset Functional Test correlates with sagittal radiographic parameters and PROMs. PT, SVA, and PI-LL were able to predict up to 25% of patients' performance on the functional testing. Although radiographic parameters are helpful in guiding ASD treatment, they should be supplemented with other forms of patients' assessment which may include functional testing. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Prophylaxis usage has been established in literature as an important component of minimizing the risk of proximal junctional kyphosis (PJK) and proximal junctional failure (PJF) development. However, literature is scarce on the effects of prophylaxis in patients who have achieved adequate postoperative alignment and those who maintained poor alignment postoperatively. PURPOSE: To investigate how PJK prophylaxis impacts rates of PJK and PJF with and without ideal alignment and the associated cost/cost-effectiveness. STUDY DESIGN/SETTING: Retrospective cohort study of a prospective adult thoracolumbar deformity database. PATIENT SAMPLE: This study included 1,541 patients. OUTCOME MEASURES: Radiographic alignment, patient-reported outcome measures (ODI), cost per QALY.
METHOD(S): Operative adult spinal deformity patients (scoliosis >20degree, SVA>5cm, PT>25degree, or TK>60degree) with an UIV at L1 or below and available baseline (BL) and 2-year (2Y) radiographic and HRQL data were included. "Matched" and "unmatched" alignment refers to the age-adjusted alignment criteria. PJK prophylaxis was defined by usage of cement, hooks or tethers. PJF was defined as PJK with reoperation. Costs were calculated using the PearlDiver database, accounting for additional costs of prophylaxis when applicable, through estimates from Medicare pay scales for services within a 30-day window, including estimates regarding costs of postoperative complications, outpatient healthcare encounters, revisions and medical related readmissions. QALY was calculated using SF6D.
RESULT(S): A total of 738 ASD patients or below met inclusion criteria (59.9yrs+/-14.0, 79%F, BMI: 27.7 kg/m2 +/-6.0, CCI: 1.8 +/-1.7). Surgically, patients had a mean level fused of 11.1+/-4.4, LOS of 7.9 days+/-4.4, EBL of 1577 mL, operative time of 377 min, with 63% undergoing an osteotomy. Forty percent of patients had PJK prophylaxis. Controlling for age, CCI, BL osteoporosis, levels fused, usage of 3CO, UIV, BL SVA and BL PI-LL, patients who were matched postoperatively in PT, SVA, or PI-LL had lowered PJF rates (OR:.5, 95% CI:.28-.86, p=.01) with prophylaxis. Among those unmatched in either SVA, PILL, or PT by 6W, prophylaxis significantly reduced the rates of PJK and PJF as well (p <0.05). ANCOVA controlling for age, CCI, BL osteoporosis, levels fused, usage of 3CO, UIV, BL SVA and BL PI-LL shows patients with ideal age-adjusted alignment and prophylaxis resulted in a lower cost per QALY by 2Y ($399,948 vs $514,228, p <.001). Similarly, in unmatched patients, prophylaxis resulted in a substantially lower cost per QALY by 2Y ($466,409 vs 672, 024, p <.001), primarily due to decreased costs of reoperation and greater improvements in QALY among prophylaxis cohorts.
CONCLUSION(S): Despite additional surgical cost, optimization of radiographic realignment in conjunction with utilization of proximal junctional failure prophylactic techniques achieves ideal cost utility, predominately due to the minimization of mechanical failure related reoperations. Even among those not achieving optimal alignment, junctional prophylactic measures improved cost utility, emphasizing its critical role of minimization of junctional failures to achieve cost efficiency in adult spinal deformity surgery. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Despite multiple reports showing favorable long-term functional outcomes following adult spinal deformity (ASD) correction, it is still unclear which patient factors may impact the sustainability of these outcomes. PURPOSE: To assess the sustainability of functional gains following ASD correction and determine predictors for postoperative loss of patient functionality. STUDY DESIGN/SETTING: Retrospective cohort. PATIENT SAMPLE: A total of 689 ASD patients. OUTCOME MEASURES: Sustained clinical benefit, Oswestry Disability Index (ODI), reoperation, mechanical failure, proximal junctional failure (PJF) METHODS: ASD patients with baseline (BL) and three-year (3Y) follow-up data were considered for analysis. Only patients who met the definition of sustained clinical benefit (SCB) at one-year (1Y) were included: 1)Passias et al's threshold for ODI (ODI<31.3), 2) did not develop PJF, and 3) did not have a reoperation for a mechanical failure. Patient groups were created by those who met SCB at 3Y, and those who did not. SCB outcomes were also reported at five-years (5Y) postoperatively. Means comparison tests assessed differences in patient groups. Factors predicting SCB were identified using adjusted regression and conditional inference tree (CIT) analysis.
RESULT(S): A total of 157 patients met inclusion criteria and had SCB at 1Y. Patient breakdown for SCB was 75.0% at 3Y. Gender (78% vs 81%), BMI (25.5 vs 26.4), levels fused (10.6 vs 10.5), operative time (345 vs 362 min), and surgical invasiveness were similar between groups, all P>.05. Following surgery, patient groups were similarly aligned to Lafage et al's age-adjusted PT, PI-LL, and SVA, all P>.05. Regression and CIT analysis adjusting for BL deformity determined that patients who met SCB at 3Y were 95.4% less likely to have an ASD-mFI score > 11 (.046[.005-.420], p=0.006), 73.1% less likely to be older than 55 years at DOS (.269[.091-.798], p=.018), 69.4% less likely to have a CCI score > 0(.306[.116-.811], p=.017), and 86.9% less likely to have a BL EQ5D score <.723 (.131[.029-.585], p=.008). Adjusted for patients without follow-up, sustained clinical benefit at 5Y was 57.9%. Predictors of SCB from 3Y to 5Y included age, CCI and EQ5D, all P <.05.
CONCLUSION(S): Sustained clinical benefit was maintainable for 75.0% of the operative ASD cohort at 3Y. Similarly, a 57.9% maintenance rate of sustained clinical benefit at 5Y reflects the longevity of ASD correction. Independent predictors of sustained clinical benefit included age, frailty, comorbidities and the patient health state. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Distal junctional kyphosis (DJK) is one of the most common complications in adult cervical deformity (ACD) correction. The utility of radiographic alignment alone in predicting and minimizing DJK occurrence warrants further study. PURPOSE: To investigate the impact of postoperative radiographic alignment on development of DJK in ACD patients. STUDY DESIGN/SETTING: Retrospective cohort study of a multicenter prospective ACD database. PATIENT SAMPLE: A total of 450 ACD patients. OUTCOME MEASURES: DJK after cervical deformity surgery.
METHOD(S): ACD patients (18 yrs) with complete baseline (BL) and two-year (2Y) radiographic data were included. DJF was defined as DJK greater than 15 or DJK with reop. Multivariable logistic regression (MVA) identified 3-month predictors of DJK. Conditional inference tree (CIT) machine learning analysis determined threshold cutoffs. Radiographic predictors were combined in a model to determine predictive value using area under the curve (AUC) methodology. "Match" refers to ideal age-adjusted alignment.
RESULT(S): A total of 140 cervical deformity patients met inclusion criteria (61.3yrs, 67%F, BMI 29kg/m², CCI 0.96+/-1.3). Surgically, 51.3% had osteotomies, 47.1% had a posterior approach, 34.5% combined approach, 18.5% anterior approach, with an average 7.6+/- 3.8 levels fused and EBL of 824 mL. Overall, 33 patients (23.6%) developed DJK, and 11 patients (9%) developed DJF. MVA controlling for age, and baseline deformity, followed by CIT found 3M cSVA <3.7 cm (OR.2, 95% CI.06-.6), and TK T4-T12 <50 (OR.17, 95% CI.05-.5, both p <.05) were significant predictors of a lower likelihood of DJK. Receiver operator curve AUC using age, T1S match, TS-CL match, LL-TK match, cSVA <3.7 cm, and T4-T12 <50 predicted DJK with an AUC of.91 for DJK by 2 years, and.88 for DJF by 2 years.
CONCLUSION(S): These findings suggest postoperative radiographic alignment is strongly associated with distal junctional kyphosis. When utilizing age-adjusted realignment in addition to newly developed thresholds, a suggested postoperative cSVA target of 3.7 cm and thoracic kyphosis less than 50, it is possible to substantially reduce the occurrence of distal junctional kyphosis and distal junctional failure. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Thoracic decompensation (TD) represents a distinct radiographic complexity from proximal junctional kyphosis (PJK). Few studies exist on the occurrence of TD following adult spinal deformity (ASD) corrective surgery. PURPOSE: To assess the incidence of TD following ASD-corrective surgery in comparison with the occurrence of PJK. STUDY DESIGN/SETTING: Retrospective cohort. PATIENT SAMPLE: A total of 689 ASD patients. OUTCOME MEASURES: ASD, TD, PJK, thoracic compensation.
METHOD(S): ASD patients with complete baseline (BL) and two-year (2Y) followup were included. Patients were divided into groups: those who developed TD: T4-T12 >54.2degree and those who developed PJK. Further analysis assessed outcomes among patients with both TD and PJK (TDPJK). Thoracic compensation was defined as expected thoracic kyphosis minus BL thoracic kyphosis. Means comparison tests and multivariable logistic regression analysis assessed differences between patient groups.
RESULT(S): A total of 373 patients met inclusion criteria. Patient breakdown by radiographic outcome was: TD (N=31), PJK (N=223) and TDPJK (N=119). Age, gender, and ASD-mFI were similar between TD and PJK patients. TD patients were more likely to be osteoporotic than PJK patients, p < 0.05. Procedures on TD patients were less invasive and utilized a shorter construct (9.3 vs 11.5 levels; both p < 0.05). TD patients had significantly greater cervical lordosis, thoracic kyphosis, and lumbar lordosis than PJK patients at BL and 2Y follow-up, all p < 0.05. Thoracic compensation was significantly associated with TD (OR 1.07 [CI 1.04-1.09], p < 0.001) controlling for age, ASD-mFI, and invasiveness. TDPJK had the highest complication rate (84.9%), significantly greater than PJK patients (70.9%), and TD patients (61.3%), both p < 0.05. PJK patients were 78.7% less likely to develop PJF than TDPJK patients (OR 0.213 [CI 0.101-0.453], p < 0.001).
CONCLUSION(S): Patients who developed thoracic decompensation were more likely to present with osteoporosis, but had less invasive procedures and levels fused than patients with proximal junctional kyphosis. The tradeoff between fusing too much resulting in proximal junctional kyphosis and fusing too little predisposing to thoracic decompensation can serve as the basis of future studies to determine optimal construct length to balance these two risks. Thoracic compensation was predictive of postoperative thoracic decompensation in adult spinal deformity-correction. Patients who developed TDPJK had the highest rate of complications and greater odds of junctional failure. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Adult spinal deformity is associated with severe pain and disability. Recent literature has shown that surgical intervention can significantly improve patient's quality of life and lessen disease burden. As many patients requiring spine surgery are elderly and often frail, restoration of alignment targets may differ. There is paucity in literature on whether different frailty and deformity states may warrant a different level of surgical invasiveness. PURPOSE: To investigate the effect of frailty and deformity severity on appropriate surgical invasiveness. STUDY DESIGN/SETTING: Retrospective cohort study from multicenter database. PATIENT SAMPLE: This study included 381 adult spinal deformity (ASD) patients. OUTCOME MEASURES: Complications; health-related quality of life (HRQL); Oswestry Disability Index (ODI).
METHOD(S): Operative ASD patients (scoliosis >20, SVA>5cm, PT>25, or TK>60) with available baseline (BL) and 2-year (2Y) radiographic and HRQL data were included. The ASD-FI was used to stratify patients into 2 categories: Not Frail (NF): 20. Four groups were formed by frailty and deformity severity: NF Low, F Low, NF Sev, F Sev. ANCOVA determined differences in Surgical Invasiveness (SI) among F/D groups. Logistic regression determined significant association between SI and reaching Ideal Clinical outcome (ICO), defined as: no major complications, no reoperations, and reaching MCID for ODI. Surgical Invasiveness (SI) was normalized to a scale of 0-100 using the SI range. Thresholds for invasiveness and achieving ICO were found through conditional inference tree (CIT) analysis.
RESULT(S): A total of 381 ASD patients (59.3+/-15yrs, 26.9+/- kg/m2, CCI: 1.52+/-1.6, FI: 3.09+/-1.6) were included and underwent surgery (11.6+/-4 lvls fused, EBL: 1522+/-1275 mL, op time: 337+/-124 min, LOS: 7.3+/-3.9 days), with 62.2% having a posterior-only approach, 36.5% combined approach, 46% having a decompression and 65% an osteotomy. By deformity severity: 60% Low-Mod, 40% Severe, while frailty assessment was 49% NF, and 51% F. This categorized the patients into deformity/frailty groups as follows: 35% NF Low, 13% NF Sev, 22% F Low, 25% F Sev. Patients had a significantly different degree of SI: 35 NF Low/Mod, 38 F Low/mod, 48 NF Sev, 50 F Sev (p<0.001). CIT analysis found NF Sev patients had a higher likelihood of reaching ICO with an SI below 40 (OR: 3.6, 95% CI: 1.1-12.1, p=.037), and F Sev with a SI below 30 (OR: 4.6, 95% CI: 1.7-12.6, p=.003). Low/mod groups revealed no significant association between invasiveness and achieving ICO.
CONCLUSION(S): Analysis of increasing surgical invasiveness in patients stratified by frailty and baseline deformity showed frail severely deformed patients, compared with non-frail severely deformed patients, had a lower invasiveness threshold above which there was a higher risk of major complications, reoperations and failure to reach clinically important improvement in ODI. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Bone Morphogenetic Protein-2 (BMP-2) has not shown superior benefit in terms of overall cost-effectiveness after implementation during adult spinal deformity (ASD) surgery. However, it remains to be shown whether certain discrete populations do obtain cost-utility from use of BMP-2. PURPOSE: Generate a score to determine usage of BMP-2 and correlate with rates of pseudarthrosis. STUDY DESIGN/SETTING: Retrospective cohort study of a single-center ASD database. PATIENT SAMPLE: A total of 689 ASD patients. OUTCOME MEASURES: Cost utility, pseudarthrosis.
METHOD(S): ASD patients with BL and 2-year(2Y) data included. BMP-2 kit size and cost: small - 4.2 mg ($21,800), medium - 8.4 mg ($23,667), large - 12 mg ($25,617). Published methods converted ODI to SF-6D. QALYs utilized a 3% discount rate for residual decline. Cost was calculated using the PearlDiver database and assessed for Complications and Comorbidities and Major Complications and Comorbidities according to CMS.gov definitions. Binary logistic regression analyses determined significant predictors for development of pseudarthrosis. Established weights were generated for predictive variables via back-step logistic regression for a risk score to predict development of pseudarthrosis. Risk score was then validated via Receiver Operating Characteristic (ROC) curve method analysis. Categories via conditional inference tree (CIT) analysis-derived thresholds were tested for cost-utility of BMP-2 usage. Marginalized means for utility gained and Cost per QALY were calculated within each risk score category, controlling for age, history of prior fusion, and baseline deformity and disability.
RESULT(S): Included: 387 ASD patients. Of 387, 64% received BMP-2 (1% small, 4% medium, 59% large). There were 17 (4.4%) of patients that developed pseudarthrosis by two years, 9 (2.3%) of which underwent reoperation. BMP-2 use, regardless of kit size, did not significantly lower pseudarthrosis rates overall (OR: 0.4, [0.2-1.04]). A predictive risk score for development of pseudarthrosis was formed by the following preoperative variables: age, frailty, history of diabetes, osteoporosis, depression, ASA grade, and baseline L4-S1 and T1PA. Via ROC method, this predictive risk score generated an AUC of 0.87. Following CIT machine learning, thresholds for the BMP Risk Score were derived: >5 No Risk (NoR), 3-5 Low Risk (LowR), 2-3 Moderate Risk (ModR), and <2 High Risk (HighR). The rates of pseudarthrosis for each category were: NoR - 0%; LowR - 1.6%; ModR - 9.3%; HighR - 24.3%. When assessing BMP-2 use and its cost-utility within each group, patients receiving BMP-2 had similar QALYs to those that did not receive BMP-2 (0.163 vs 0.171, p=.65). BMP-2 usage had significantly worse cost-utility in both NoR and LowR cohorts (both p<.05). In ModR patients, BMP-2 usage had equivocal cost-utility ($680,532.35 vs $580,380.21, p=.14). In the HighR cohort, the cost-utility difference narrowed even further (BMP-2 use: $743,155.21 vs $719,628.79, p=.82).
CONCLUSION(S): Our study shows BMP-2 has equivocal cost-utility within those at moderate and high risk for developing pseudarthrosis within two years following spinal deformity correction. The generated predictive score can better aid spine surgeons assess risk and enhance justification for the use of BMP-2 during surgical intervention for adult spinal deformity. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright