Faculty Bibliography

The aim of this study was to investigate the cost utility of treating non-frail versus frail or severely frail adult spinal deformity (ASD) patients. 79 surgical ASD patients >18 years with available frailty and ODI data at baseline and 2-years post-surgery (2Y) were included. Utility data was calculated using the ODI converted to the SF-6D. QALYs utilized a 3% discount rate to account for decline to life expectancy (LE). Costs were calculated using the PearlDiver database. ICER was compared between non-operative (non-op.) and operative (op.) NF and F/SF patients at 2Y and LE. When compared to non-operative ASD, the ICER was $447,943.96 vs. $313,211.01 for NF and F/SF at 2Y, and $68,311.35 vs. $47,764.61 for NF and F/SF at LE. Frail and severely frail patients had lower cost per QALY compared to not frail patients at 2Y and life expectancy, and had lower ICER values when compared to a non-operative cohort of ASD patients. While these results support operative correction of frail and severely frail patients, it is important to note that these patients are often at worse baseline disability, which is closely related to frailty scores, and have more opportunity to improve postoperatively. Furthermore, there may be a threshold of frailty that is not operable due to the risk of severe complications that is not captured by this analysis. While future research should investigate economic outcomes at extended follow up times, these findings support the cost effectiveness of ASD surgery at all frailty states.

STUDY DESIGN/METHODS:Retrospective analysis of outcomes in single-level spine and primary hip and knee arthroplasty patients. OBJECTIVE:Compare baseline and post-operative outcomes in patients undergoing spine surgery procedures with total hip arthroplasty (THA) and total knee arthroplasty (TKA) to further define outcomes in orthopedic surgery. SUMMARY OF BACKGROUND DATA/BACKGROUND:Computer-adaptive Patient Reported Outcome Information System (PROMIS) allows for standardized assessment of the Health Related Quality of Life across different disease states. METHODS:Patients who underwent spine surgery (anterior cervical discectomy and fusion, cervical disc replacement, lumbar laminectomy, microscopic lumbar discectomy, transforaminal lumbar interbody fusion or adult reconstruction surgery (THA, TKA) were grouped. Mean Charlson Comorbidity Index (CCI), Baseline (BL) and 6-month (6 M) PROMIS scores of Physical Function, Pain Interference, and Pain Intensity were determined. Paired t-tests compared differences in CCI, BL, 6 M, and change in PROMIS scores for spine and adult reconstruction procedures. RESULTS:304 spine surgery patients (Age=58.1 ± 15.6; 42.9% Female) and 347 adult reconstruction patients (Age=62.9 ± 11.8; 54.1% Female) were compared. Spine surgery groups had more disability and pain at baseline than adult reconstruction patients according to Physical Function [(21.0, 22.2, 9.07, 12.6, 10.4) vs (35.8, 35.0), respectively, p < .01], Pain Interference [(80.1, 74.1, 89.6, 92.5, 90.6) vs (64.0, 63.9), respectively, p < .01] and Pain Intensity [(53.0, 53.1, 58.3, 58.5, 56.1) vs (53.4, 53.8), respectively, p < .01]. At 6 M, spine surgery patients remained more disabled and had more pain compared to adult reconstruction patients. Over the 6-month timespan, spine patients experienced greater improvements than adult reconstruction patients in terms of Physical Function [(+8.7, +22.2, +9.7, +12.9, +12.1) vs (+5.3, +3.9), respectively, p < .01] and Pain Interference scores [(-15.4, -28.1, -14.7, -13.1, -12.3) vs (-8.3, -6.0), respectively, p < .01]. CONCLUSIONS:Spinal surgery patients had lower BL and 6 M PROMIS scores, but greater relative improvement in PROMIS scores compared to adult reconstruction patients. LEVEL OF EVIDENCE/METHODS:3.

BACKGROUND CONTEXT: This study evaluates the necessity of a multi-rod construct to help prevent a pseudarthrosis in ASD surgery of lesser magnitude. Regardless of an interbody at L5/S1, a multi-rod construct does not decrease the risk of developing a pseudarthrosis. If a pseudarthrosis develops in patients who undergo an interbody fusion at L5/S1, it is more likely to occur above the L5/S1 level and the number of rods used does not decrease this risk. PURPOSE: To examine the necessity of IBF and multi-rod use to prevent pseudarthrosis in these lesser magnitude ASD surgeries. STUDY DESIGN/SETTING: Retrospective review of a multicenter database. PATIENT SAMPLE: A total of 483 patients met our inclusion criteria of pelvic fixation, > 2 year follow up, >5 levels fused, no 3-column osteotomy (3CO), and no multi-level interbody fusions. OUTCOME MEASURES: Pseudarthrosis rate and location in ASD patients.
METHOD(S): A multi-center database was evaluated to identify patients with pelvic fixation, > 2 year follow up, >5 levels fused, and no 3-column osteotomy (3CO). We define a lesser magnitude ASD surgery as surgeries that do not include 3CO or multi-level interbody fusions. In this cohort, ASD patients with and without an IBF at L5/S1 were assessed to evaluate if the number of rods had an impact on the pseudarthrosis rate. Pseudarthrosis was determined by intraoperative findings or record review. Fisher's exact tests, univariate and multivariate analysis were used to identify predictors of pseudo.
RESULT(S): A total of 483 patients met our inclusion criteria with an average follow-up of 48 months. Of those, 321 (66%) had interbody at L5/S1 (IBF). Twenty-three percent of IBF patients had a MR construct and 77% had a SR construct. In the non-IBF group, 33% had a MR construct. In the IBF group, the MR construct did not impact the pseudarthrosis rate compared to a SR construct, (6.8% vs 5.2% respectively, p=.35). In the IBF group, 29% (4) of MR patients and 75% (6) of SR patients with pseudo received BMP. However, regardless of the rod construct, ASD patients with an interbody at L5/S1 had an increased risk of having a pseudarthrosis above this level (OR.096, p=.045). Also, the MR construct did not decrease the risk of developing a pseudarthrosis in the non-IBF group (0% vs 4.5%, p=.18).
CONCLUSION(S): In ASD surgery of lesser magnitude, a multi-rod construct does not decrease the risk of developing a pseudarthrosis. Patients who have an interbody fusion at L5/S1 are more likely to develop a pseudarthrosis above this level, regardless of the rod construct. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Proximal junctional failure (PJF) is a common and particularly adverse complication of adult spinal deformity (ASD) surgery. There is evidence that the rigidity of posterior spinal constructs may impact risk of PJF. PURPOSE: We hypothesized that lower mean screws per level and decreased rod stiffness would be associated with lower incidence of PJF. STUDY DESIGN/SETTING: Retrospective cohort study. PATIENT SAMPLE: Patients with ASD and 2-year minimum follow-up were included. Only patients undergoing fusion of >=5 levels, and with LIV at the sacro-pelvis were included. Patients undergoing revision surgery were excluded. In total, 420 patients were analyzed. OUTCOME MEASURES: The primary outcome variable was PJF, defined using previously published radiographic criteria (PJ angle >28degreeand PJ angle >22degree, and >=8mm/>=3mm listhesis at upper thoracic / thoracolumbar levels, respectively).
METHOD(S): The primary independent variables were the mean number of screws per level fused analyzed with a cutoff of 1.8 (determined by ROC analysis) and rod material/diameter. Multivariable logistic regression was utilized to investigate confounding factors, including age, history of osteoporosis, BMI, gender, CCI, preoperative Schwab modifiers, preoperative TPA, postoperative change in lumbar lordosis, upper instrumented vertebra, osteotomy, approach, UIV type (ie, pedicle screw vs other), and number of levels fused.
RESULT(S): Of the total patients, 78.8% were female. PJF occurred in 14.1% of patients. The mean screws per level was 1.7 (SD 0.2), and 57.6% of patients had <1.8 screws per level. PJF occurred in 19.4% vs 9.8% of patients with >=1.8 vs <1.8 screws per level, respectively (p<0.01). In multivariable analysis, patients with <1.8 screws per level exhibited lower odds of PJF (OR 0.39, p<0.01). Rod material and diameter (both p>0.2) were not significantly associated with PJF. Screw density specifically adjacent to the UIV was not related to PJF (p>0.2).
CONCLUSION(S): Among ASD patients undergoing long-segment primary fusion to the pelvis, the risk of PJF was lower among patients with <1.8 screws per level. Rod diameter and material, however, were not significantly associated with PJF. This finding may be related to construct rigidity. It is possible that residual confounding by other patient and surgeon-specific characteristics may exist. Further biomechanical and clinical studies exploring the relationship between screw density and PJK/PJF are warranted. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Frailty has been utilized in cervical deformity (CD) to characterize the influence of preop health state on postop outcomes. Frailty severity may be an important determinant for impaired recovery after CD corrective surgery. No prior studies have examined the associations between increasingly frail states and the trajectory of recovery in CD realignment. PURPOSE: Evaluate postop clinical recovery among CD patients between frailty states undergoing primary procedures STUDY DESIGN/SETTING: Retrospective review of a prospective CD database PATIENT SAMPLE: A total of 106 CD patients. OUTCOME MEASURES: Baseline (BL) to 1-Year(1Y) HRQL Instruments: NDI, mJOA, EQ5D METHODS: Patients>18yrs undergoing surgery for CD (C2-7 Cobb>10degree, coronal Cobb>10degree, cSVA>4cm or TS-CL>10degree, or CBVA>25degree) with HRQL data at BL, 3M and 1Y postop intervals were identified. Patients were stratified by the modified CD frailty index (mCD-FI, Passias et al.) scale from 0-1 (no frailty:<0.3[NF], mild/severe: >0.3[F]). Patients in NF and F groups were propensity score matched (PSM) for TS-CL to control for baseline deformity. Demographics, alignment and Ames-ISSG deformity modifiers were assessed using chi-squared and paired t-tests to compare HRQL outcomes. HRQL outcomes were normalized by dividing BL and postop (3M, 1Y) outcomes by BL for each patient. Normalized scores (y-axis) were plotted against duration of follow-up (x-axis). AUC was calculated for follow-up time intervals; total area for each follow-up interval was divided by cumulative follow-up, determining overall normalized, time-adjusted HRQL outcomes (Integrated Health State [IHS]). IHS was compared between NF and F groups.
RESULT(S): A total of 106 CD patients included (61.7yrs, 66% F, 27.7kg/m2). By frailty group: 52.8% NF, 47.2% F. After PSM for TS-CL (mean: 38.1degree), 38 patients remained in each of the NF and F groups. By surgical approach, 46.1% underwent posterior approach, 19.7% anterior, and 34.2% combined; which was not significant between the frailty groups(p>0.05). At baseline, cervical and spinopelvic radiographic parameters were not significant, except for the C7-S1 SVA (NF: -26.5mm vs F: 15.2mm, p=0.007). According to BL HRQLs, F patients displayed significantly worse NDI scores (NF: 36.8 F: 55.4; p<0.001), mJOA (NF: 14.7, F: 12.7; p=0.002), and EQ5D (NF: 0.78, F: 0.69; p<0.001). All frailty groups exhibited BL to 1Y improvement in NDI, EQ5D and NRS Neck Pain (all p<0.001). After HRQL normalization, F patients had more improvement in mJOA scores at 3M (p=0.065) as well as NDI (p=0.096) and EQ5D (p=0.016). IHS-adjusted HRQL outcomes from BL to 1Y showed a significant difference in EQ5D scores (NF: 1.02, F: 1.07, P=0.016). No significant differences were found in the IHS NDI and mJOA between frailty groups (p>0.05). F patients had more postop major complications (31.3%) compared to the NF (8.9%), p=0.004, though DJK occurrence and reoperation between the groups was not significant.
CONCLUSION(S): While all groups exhibited improved postop disability/pain scores, frail patients recovered better in overall health state. Despite frail patients having more complications, they seem to have overall better patient-reported outcomes, signifying that with frailty severity, patients have more room for improvement postop compared to baseline quality of life. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: With the rise of health care costs and a focus on value-based outcomes, hospitals have become more cognizant on cost of revisions and complications. However, literature on the effect of distal junctional kyphosis in driving up health care costs is scarce. PURPOSE: To evaluate the effect of distal junctional kyphosis on the cost effectiveness of corrective cervical deformity surgery. STUDY DESIGN/SETTING: Retrospective review of a single surgeon database. PATIENT SAMPLE: This study included 123 cervical deformity patients. OUTCOME MEASURES: Cost per QALY.
METHOD(S): Cervical deformity patients with minimum 1-year HRQL follow-up were included. Means comparison tests assessed differences in baseline demographic and clinical data. Utility data was calculated using published conversion methods to convert NDI to SF-6D. QALYs utilized a 3% discount rate to account for residual decline to life expectancy (78.7 years). Costs was calculated using the PearlDiver database incorporating complications, comorbidities (CC), major complications, and comorbidities (MCC) were assessed according to CMS.gov manual definitions. Reimbursement consisted of a standardized estimate using regression analysis of Medicare pay-scales for all services rendered within a 30-day window, including estimates regarding costs of postoperative complications, outpatient health care encounters, revisions and medical-related readmissions. After accounting for CC, MCC, length of stay (LOS) and death, cost per QALY at 2Y was calculated for revisions due to distal junctional kyphosis.
RESULT(S): A total of 123 cervical deformity patients met inclusion criteria (57.22yrs, 54%F, 29.0kg/m2). At baseline, patients presented radiographically as: PT (16.9+/-9.7), PI (54.7+/-11.4), PI-LL (-2.9+/-12.6), SVA (80.5+/-49.8), cSVA (28.4+/-20.6), TS-CL (26.6+/-14.4). Surgical details: EBL of 708 mL, operative time of 438.6 min, with 29.51% undergoing an anterior approach, 50.82% posterior-only approach, and 19.67% combined approach. Overall, 7.9% of patients developed DJK within two years postoperatively. Average cost of revision surgery due to DJK within 2years of index surgery was $50,736 +/- 31,467. Patients that developed DJK within 2years of index surgery trended toward having a greater baseline NDI (62.8 vs 55.47, p>0.05) and showed less improvement in NDI at 2 years (4 vs 16.6). Overall, cost per QALY was higher for patients developing DJK at 2 years ($28,483 vs $20,989).
CONCLUSION(S): Cervical deformity revisions due to distal junctional kyphosis had a cost per QALY of $28,483. Efforts to limit postoperative DJK after surgical intervention can further limit additional costs associated with revisions and complications. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Frailty is a baseline measure of disability that transcends age alone and has been determined a strong predictor of outcomes following adult spinal deformity (ASD) surgery. This postop impact calls for investigation of unique adjustment of Global Alignment and Proportion (GAP) scores accounting for frailty. This adjustment in spinal proportion may help surgical planning for individualized, optimal postop outcomes. PURPOSE: Modify the GAP score with frailty to optimize outcomes in surgical ASD patients. STUDY DESIGN/SETTING: Retrospective review of a single-surgeon comprehensive ASD database PATIENT SAMPLE: A total of 140 ASD patients OUTCOME MEASURES: Frailty-adjusted GAP scores; Health Related Quality of Life scores (HRQLs): ODI, SRS-22 METHODS: Surgical ASD patients (SVA>=5cm, PT>=25degree, or TK >=60degree, >3 levels fused) >=18 years old with available baseline (BL) radiographic data were isolated in the single-center Comprehensive Spine Quality Database (Quality). Patients were dichotomized by the ASD frailty index, F (Not Frail, Frail). Linear regression analysis established radiographic equations for frailty-adjusted GAP Scores at baseline and 2-years involving relative pelvic version, relative lumbar lordosis, lordosis distribution index, relative spinopelvic alignment, and an age factor to formulate a sagittal plane score. Patients were restratified into frailty-adjusted proportionality groups: Proportional (<5.8), Moderately Disproportional (MD) (5.8-7), Severely Disproportional (SD) (>7). Frailty-adjusted GAP proportionality at 2-years were compared to adjusted-BL to determine whether patients improved, deteriorated or remained the same in their spine proportion.
RESULT(S): A total of 140 patients were included (55.5+/-16.4 yrs, 77.5% female, 25.2+/-4.7 kg/m2). BL frailty: 32.8% not frail, 67.2% frail. Primary analyses demonstrated correlation between BL frailty score and BL and 2-year GAP scores(P<0.001). Linear regression analysis(p<0.001) developed a frailty-adjusted GAP threshold equation: 4.4 + 0.93*(frailty score). Adjusted-baseline scores were taken and re-stratified based distribution and placed 26.4% of patients in Proportional, 26.6% MD, and 44% SD. BL adjusted GAP scores by frailty group: 5.3 Not Frail, 7.5 Frail; p<0.001. At 2-years, GAP scores were grouped into the frailty-adjusted proportionality groups: 66.2% Proportional, 10.8% MD, and 23.1% SD. Patients who were 2-year MD/SD underwent significantly more reoperations (>33.5%) compared to Proportional (12.8%), p=0.015. SD 2-year patients developed increased PJK at the 1-year mark (40%, Proportional: 13.9%, MD:7.1%, p=0.003), as well as had worse 2-year ODI and SRS-22 satisfaction scores(p<0.050). 47.5% improved in GAP (63.4% of frail patients), 12.3% deteriorated, and 40.2% remained in the same proportionality group at 2-year follow up.
CONCLUSION(S): Significant associations exist between frailty and spinal proportion. By adjusting the GAP proportionality groups accounting for baseline frailty contributed to improved outcomes and minimized reoperations. The adjusted GAP groups appeal for less rigorous spine proportion goals in severely frail patients. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Cervical deformity (CD) severity has been linked to poor quality of life and severe neck pain. However, little is known of the relationship between surgical invasiveness accounting for cervical deformity severity and frailty status. PURPOSE: Investigate the outcomes of CD surgery by invasiveness, frailty status and baseline degree of deformity. STUDY DESIGN/SETTING: Retrospective cohort study. PATIENT SAMPLE: Ninety-six CD patients. OUTCOME MEASURES: HRQL measurements: NDI, EQ5D, mJOA.
METHOD(S): Inclusion criteria was defined as operative CD patients (C2-C7 Cobb>10degreeor CK>10degree, cSVA>4cm or CBVA>25degree) >18yr with follow up (1-year) radiographic and HRQL scores, NDI, mJOA and EQ5D. Patients were stratified by severity of deformity by TS-CL, categorized by the new-mJOA based modifiers: Low/Mod <45degree (Low-Mod); Severe >45degree(Sev). Frailty scores were calculated based upon the modified CD frailty index by Passias et. Al and categorized into severely frail (SF) >=0.5 and not SF (N-SF)<0.5. Patients were categorized by their frailty and deformity status (Low-Mod/N-SF; Low-Mod/SF; Sev/N-SF; Sev/SF). Logistic regression analysis assessed the relationship between increasing invasiveness and outcomes (DJK, complications). Within the significant frailty/deformity risk groups, decision tree analysis assessed thresholds for an invasiveness severity cut-off point, below which experiencing a reoperation, complication, DJK occurrence and poor mJOA were higher.
RESULT(S): Ninety-six CD patients met inclusion criteria (62.2+/-10.2 years, 66% female, 28.4+/-7.4 kg/m2). By approach: 19.8% anterior-only, 47.9% posterior-only, 32.3% combined (levels fused: 7.7+/-3.9). By deformity severity: 23.7% Low, 40.9% Moderate, 34.4% Severe, while frailty assessment placed 32.3% in N-SF and 67.7% SF. This categorized the patients into deformity/frailty groups as follows: 19.8% Low-Mod/N-SF(19), 13.5%(13) Sev/N-SF, 44.8%(43) Low-Mod/SF, 21.9%(21) Sev/SF. Logistic regression analysis found a significant relationship between increasing deformity severity and occurrence of severe postop DJK(1.053 [1.016-1.093], p= 0.005), complications(1.045 [1.012 - 1.080], p=0.007), revision by 1-year(1.059 [1.000-1.122], p= 0.049). Additionally, increasing invasiveness and occurrence of severe DJK (1.030 [1.007-1.054], p= 0.024) and revision (1.026 [1.008 - 1.044], p=0.005). Invasiveness increased with deformity and frailty severity: 53.6 Low-Mod/N-SF, 81.4 Sev/N-SF, 56.4 Low-Mod/SF, 79.8 Sev/SF; p=0.002. After defining a favorable outcome as no occurrence of severe DJK, no major complications and no revisions, and 1Y mJOA improvement (28.1%), invasiveness scores were compared within deformity/frailty groups between patients who met/did not meet the favorable outcome. For the NSF deformity groups, those with a favorable outcome had larger invasiveness scores (Low-Mod: 58.7 vs 48.5; Sev: 77.7 vs 89.6). For the SF deformity groups, the favorable outcome had significantly lower invasiveness scores for the Low-Mod deformity group (38.1 vs 62.9, p=0.008), while the Sev/SF deformity favorable outcome group remained larger (86.8 vs 79.4), though this was not significant. For the Low-Mod/SF group an invasiveness cutoff score of <48 where achieving a favorable outcome was 3x higher (3.08[1.2-7.9], p=0.019).
CONCLUSION(S): For SF patients, when deformity severity is low to moderate, surgeons may limit the invasiveness of their procedures in order to account for the patient's susceptibility to poor outcomes. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: Complications occur commonly following adult spinal deformity surgery. Length of stay may be influenced by the type of surgical intervention, and/or by postoperative complication and the intervention required to treat that complication. PURPOSE: To understand the impact of complications and their intervention severity on the incremental increase in length of stay. STUDY DESIGN/SETTING: Retrospective cohort study of prospective ASD database. PATIENT SAMPLE: The study included 1418 adult spinal deformity patients. OUTCOME MEASURES: Neurologic complications, HRQL scores.
METHOD(S): ASD patients (>18yrs, scoliosis >=20degree, SVA >=5cm, PT >=25degree and/or TK >60degree). Complications were classified based on complication type, and intervention severity (grade 0, no intervention, Grade 1 minimal intervention (medication change, etc), Grade 2 moderate (cardioversion, chest tube, etc), grade 3 major intervention (return to OR).
RESULT(S): Of the 1418 patients included, 901 had at least one complication event, with 568 with a complication before discharge. Three hundred and six had an event that required a reoperation and 275 were readmitted for their event. When determining LOS, a baseline was established that was influenced by the type of surgery and placed into three groups based on osteotomy (O) and interbody use (I); Group 1 (n408): 0% O, 75% I, LOS=7.9, Group 2 (n374): 100% O, 0% I, LOS=11.9, Group 3 (n595) 100% O and I, LOS=11.2. Having a complication event prior to discharge led to an increase of 1-4 days for each group (p<0.01). This was true for every complication, regardless of type, but was worse for renal and pneumonia. The increase in LOS was most driven by the intervention required. No intervention had an LOS of 6.9, Grade 1 +1.2 days, Grade 2, +1.8 days, Grade 3 +2.5 days. Using the sum of a weighted intervention score (SUM=1*#grade1 + 2*#grade2 + 3*#grade3), we can even more accurately predict the incremental increase in LOS. This occurs until a weighted score of 7 which accurately predicts an increase of 7.5 days.
CONCLUSION(S): Length of stay following ASD is predicted by the type of surgical intervention, and postoperative complication profile. Additional days are most predicted by the type of complication and by intervention severity. A weighted intervention score can predict the additional LOS and is useful to help understand the impact of complications. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright

BACKGROUND CONTEXT: AI algorithms have shown substantial promise in medical image analysis. Previous studies of ASD clusters have analyzed alignment metrics - this study sought to complement these efforts by analyzing images of anatomical landmarks. PURPOSE: We hypothesized that a neural-network-based artificial intelligence (AI) algorithm would cluster preoperative lateral radiographs of into groups with distinct morphology. STUDY DESIGN/SETTING: Retrospective cohort study. PATIENT SAMPLE: A total of 915 patients with adult spinal deformity and preoperative lateral radiographs. OUTCOME MEASURES: Schwab modifiers for SVA and PI-LL, three-column osteotomy, upper instrumented vertebrae, baseline Oswestry Disability Index, and 2-year likelihood of reaching MCID in ODI (set at -12.8). Proximal junctional kyphosis and proximal junctional failure were defined using previously published radiographic criteria.
METHOD(S): Vertebral locations for C3-L5, sacral endplate, and femoral heads were measured on lateral radiographs. Pixel locations were used to create a black-and-white overlay to the image, which was subsequently standardized for size and position using the femoral heads and sacral endplate. These images were used to train a self-organizing map (SOM). SOMs are a form of artificial neural network frequently employed in unsupervised classification tasks.
RESULT(S): In total, 915 preoperative lateral radiographs were analyzed. A 2 x 3, toroidal SOM was trained. The mean spine shape was plotted for each cluster. Alignment, surgical characteristics, and outcomes were compared between clusters. Clusters C and D exhibited a particularly high proportion of patients with optimal (ie, modifier 0) values of PI-LL (65.0% and 68.5%) and SVA (72.8% and 53.1%). Conversely, clusters B, E, and F tended to have poor (ie, modifier ++) PI-LL (74.8%, 66.9%, and 74.6%) and SVA (75.5%, 48.6%, and 58.7%). 3-CO was most common among cluster A (26.8%), cluster B (32.6%), and cluster F (32.7%). UIV at T7-T12 was most common among cluster B (51.1%) and cluster F (60.3%). ODI <30 was most prevalent among cluster D (31.4%). There was little difference, however, between groups in likelihood of reaching MCID in ODI at 2-year follow-up. PJK and PJF were particularly prevalent among clusters A (51.2% and 15.5%) and E (50.4% and 18.7%).
CONCLUSION(S): This study developed a self-organizing map that clustered preoperative lateral radiographs of ASD patients into groups with highly distinct overall spinal morphology. These clusters predicted alignment, surgical characteristics, and HRQOL. Further studies of this classification approach will expand to compare pre- and postoperative radiographs. FDA DEVICE/DRUG STATUS: This abstract does not discuss or include any applicable devices or drugs.
Copyright