Faculty Bibliography

OBJECTIVE:History of previous cervical spine surgery is a frequently cited cause of worse outcomes after cervical spine surgery. The purpose of this study was to determine any differences in clinical outcomes after cervical spine surgery between patients with and without a history of previous cervical spine surgery. METHODS:A multicenter prospective database was reviewed retrospectively to identify patients with cervical spondylosis undergoing surgery with a minimum 2-year follow-up. Patients were divided into 2 groups: patients with (W) or without (WO) previous history of cervical spine surgery. Statistical analyses of Health-Related Quality of Life scores were analyzed with statistical software to fit linear mixed models for continuous longitudinal outcome. RESULTS:A total of 1286 patients (377 W, 909 WO) met criteria for inclusion. Overall, patients in both groups experienced an improvement in their Health-Related Quality of Life scores. However, patients in the W group had significantly decreased improvement compared with WO patients in the Neck Disability Index score and the following SF-36 domain scores: Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Health Transition, and Physical Component Summary at all time points (P < 0.05). There was no statistically significant difference between the W and WO groups in operative time, estimated blood loss, length of stay, or complications (P > 0.05). CONCLUSIONS:Patients with a history of previous cervical spine surgery had inferior improvement in quality of life outcome scores. Patients with a history of previous surgical intervention who elect to undergo subsequent surgeries should be appropriately counseled about expected results.

STUDY DESIGN/METHODS:A retrospective study. OBJECTIVE:To examine the link between major complications, surgical techniques, and perioperative care in the intralesional spondylectomy of the upper cervical spine. SUMMARY OF BACKGROUND DATA/BACKGROUND:Spondylectomy has been demonstrated to prolong cancer-free survival in many patients with locally aggressive spinal tumors. However, the challenging nature of this surgical procedure and the potential for severe complications often limit its application in the upper cervical spine. METHODS:Nineteen patients with primary upper cervical tumors were treated with spondylectomy from March 2005 to August 2009, using either the anterior-posterior or posterior-anterior approach. Anterior procedures were transmandibular, transoral, or high retropharyngeal. Anterior reconstructions were performed in plates with iliac crest strut grafts, plates with mesh cages, and Harms mesh cages alone. Occipitocervical fixation was performed with Halo-vest application for postoperative immobilization. RESULTS:Vertebral artery injuries occurred unilaterally in 5 cases intraoperatively: 4 occurred in the anterior approach of anterior-posterior procedures. Fusion was achieved in 9 patients with intact internal instrumentation. Fusion with the anterior construct in a tilted position occurred in 3 patients, all of whom underwent anterior-posterior procedures with Halo-vest immobilization for less than 1 month. Nonunion occurred in 3 cases after the posterior-anterior procedure because of anterior bone graft absorption. Prolonged Halo-vest immobilization maintained postoperative stability. Failure of internal instrumentation occurred in 3 cases. Anterior construct dislocation and severe tilting occurred in 2 cases after the anterior-posterior procedure. Five patients had a local recurrence. All recurrent lesions were malignant tumors and occurred in regions where surgical exposure was inadequate with incomplete excision. CONCLUSION/CONCLUSIONS:The order of the surgical approach is a critical determinant of complications, fusion rates, choice of surgical technique, and reconstruction methods. The postoperative use of a Halo-vest is recommended. Local recurrence is associated with tumor malignancy and inadequate excision margin. LEVEL OF EVIDENCE/METHODS:4.

PURPOSE: Reciprocal mechanisms for standing alignment have been described in thoraco-lumbar deformity but have not been studied in patients with primary cervical deformity (CD). The purpose of this study is to report upper- and infra-cervical sagittal compensatory mechanisms in patients with CD and evaluate their changes post-operatively. METHODS: Global spinal alignment was studied in a prospective database of operative CD patients. Inclusion criteria were any of the following: cervical kyphosis (CK) > 10 degrees , cervical scoliosis > 10 degrees , cSVA (C2-C7 Sagittal vertical axis) > 4 cm or CBVA (Chin Brow Vertical Angle) > 25 degrees . For this study, patients who had previous fusion outside C2 to T4 segments were excluded. Patients were sub-classified by increasing severity of cervical kyphosis [CL (cervical lordosis): < 0 degrees , CK-low 0 degrees -10 degrees , CK-high > 10 degrees ] and cSVA (cSVA-low 0-4 cm, cSVA-mid 4-6 cm, cSVA-high > 6 cm) and were compared for pre- and 3-month post-operative regional and global sagittal alignment to determine compensatory recruitment. RESULTS: 75 CD patients (mean age 61.3 years, 56% women) were included. Patients with progressively larger CK had a progressive increase in C0-C2 (CL = 34 degrees , CK-low = 37 degrees , CK-high = 44 degrees , p = 0.004), C2Slope and T1Slope-CL (p < 0.05). As the cSVA increased, there was progressive increase in C2Slope, T1Slope and TS-CL (p < 0.05) and patients compensated through increasing C0-C2 (cSVA-low = 33 degrees , cSVA-mid = 40 degrees , cSVA-high = 43 degrees , p = 0.007) and pelvic tilt (cSVA-low = 14.9 degrees , cSVA-mid = 24.1 degrees , cSVA-high = 24.9 degrees , p = 0.02). At 3 months post-op, with significant improvement in cervical alignment, there was relaxation of C0-C2 (39 degrees -35 degrees , p = 0.01) which positively correlated with magnitude of deformity correction. CONCLUSIONS: Patients with cervical malalignment compensate with upper cervical hyper-lordosis, presumably for the maintenance of horizontal gaze. As cSVA increases, patients also tend to exhibit increased pelvic retroversion. Following surgical treatment, there was relaxation of upper cervical compensation.

OBJECTIVE: The etiologies and epidemiology of traumatic cervical spine fracture have not been described with sufficient power or recency. Our goal is to describe demographics, incidence, etiology, spinal cord injuries (SCIs), concurrent injuries, treatments, and complications of traumatic cervical spine fractures. METHODS: Retrospective review of the Nationwide Inpatient Sample. ICD-9 E-Codes identified trauma cases from 2005-2013. Patients with cervical fracture were isolated. Demographics, incidence, etiology, fracture levels, concurrent injuries, surgical procedures, and complications were analyzed. T-tests elucidated significance for continuous variables, chi-square for categorical variables. Level of significance P<0.05. RESULTS: 488,262 patients isolated (age:55.96, male:60.0%, white:77.5%). Incidence (2005:4.1% vs 2013:5.4%), Charlson-Comorbidity-Index (2005:0.6150 vs. 2013:1.1178), and total charges (2005:$71,228.60 vs. 2013:$108,119.29) have increased since 2005, while length of stay (LOS) decreased (2005:9.22 vs. 2013:7.86) (all P<0.05). Most common etiologies were MVA (29.3%), falls (23.7%) and pedestrian accidents (15.7%). Most frequent fracture types were closed at C2 (32.0%) and C7 (20.9%). Concurrent injury rates have significantly increased since 2005 (2005:62.3% vs. 2013:67.6%). Common concurrent injuries included fractures to the rib/sternum/larynx/trachea (19.6%). Overall fusion rates have increased since 2005 (2005:15.7% vs 2013:18.0%), while decompressions and halo insertion rates have decreased (all P<0.05). SCIs have significantly decreased since 2005, except for upper-cervical central cord syndrome. Complication rates have significantly increased since 2005 (2005:31.6% vs. 2013:36.2%). Common complications included Anemia (7.7%), Mortality (6.6%), and ARDS (6.6%). CONCLUSION: Incidence, complications, concurrent injuries, and fusions have increased since 2005. LOS, SCIs, decompressions and halo insertions have decreased. Indicated trends should guide future research in management guidelines.

BACKGROUND: Hospital length of stay (LOS), 30-day readmission rate, and other metrics are increasingly being used to evaluate quality of surgical care. The factors most relevant to cervical spondylotic myelopathy (CSM) are not yet established. OBJECTIVE: To identify peri-operative factors associated with extended LOS and/or 30-day readmission following elective surgery for CSM. METHODS: Surgical CSM patients at institutions represented by the American-College-of-Surgeons-National-Surgical-Quality-Improvement-Program (ACS-NSQIP) from 2010-2012 were included. Patients with fracture, >/=9 levels fused, or cancer were excluded. Extended LOS was defined as 75th percentile of the cohort. Univariate analysis and multivariate logistic regression identified predictors for extended LOS, 30-day readmission, and reoperation. Linear regression modeling was used to evaluate variables. RESULTS: 3057 surgical CSM cases were isolated. Age (OR-1.496), diabetes (OR-1.691), American Society of Anesthesiologists (ASA) class (OR-2.081), posterior surgical approach (OR-2.695), and operative time (OR-1.008) were all positive predictors (P<0.05) for extended LOS (>/=4 days). 32% of the cohort (976 patients) had 30-day readmission data. Among these, 915 patients were not readmitted (93.8%), while 61 (6.2%) were. Diabetes (OR-1.460) and ASA-class (OR-2.539) were significant positive predictors for hospital readmission. Age (OR-0.918) was a negative predictor of re-operation in readmitted patients, and pulmonary comorbidities (OR-4.584) were a positive predictor (P<0.05). CONCLUSIONS: Patients with diabetes and higher ASA-Class were at increased risk for extended LOS and readmission within 30-days. Patients with increased operative time have greater risk for extended LOS. Pre-operative pulmonary comorbidities increased reoperation risk, while increased age reduced the risk. Attention to these factors may benefit CSM patients.

BACKGROUND: Proximal junctional kyphosis (PJK) following adult spinal deformity (ASD) surgery is a well-documented complication, but associations between radiographic PJK and cervical malalignment onset remain unexplored. OBJECTIVE: To study cervical malalignment in ASD surgical patients that develop PJK. METHODS: Retrospective review of prospective multicenter database. Inclusion: primary ASD patients (>/=5 levels fused, upper instrumented vertebra [UIV] at T2 or above, and 1-yr minimum follow-up) without baseline cervical deformity (CD), defined as >/=2 of the following criteria: T1 slope minus cervical lordosis < 20 degrees , cervical sagittal vertical axis < 4 cm, C2-C7 cervical lordosis < 10 degrees . PJK presence (<10 degrees change in UIV and UIV + 2 kyphosis) and angle were identified 1 yr postoperative. Propensity score matching between PJK and nonPJK groups controlled for baseline alignment. Preoperative and 1-yr postoperative cervical alignment were compared between PJK and nonPJK patients. RESULTS: One hundred sixty-three patients without baseline CD (54.9 yr, 83.9% female) were included. PJK developed in 60 (36.8%) patients, with 27 (45%) having UIV above T7. PJK patients had significantly greater baseline T1 slope in unmatched and propensity score matching comparisons (P < .05). At 1 yr postoperative, PJK patients had significantly higher T1 slope (P < .001), C2-T3 Cobb (P = .04), and C2-T3 sagittal vertical axis (P = .02). New-onset CD rate in PJK patients was 15%, and 16.5% in nonPJK patients (P > .05). Increased PJK magnitude was associated with increasing T1 slope and C2-T3 SVA (P < .05). CONCLUSION: Patients who develop PJK following surgical correction of ASD have a 15% incidence of development of new-onset CD. Patients developing PJK following surgical correction of ASD tend to have an increased preoperative T1 slope. Increased progression of C2-T3 Cobb angle and C2-T3 SVA are associated with development of PJK following surgical correction of thoracolumbar deformity.

BACKGROUND: Prior studies have observed similar health-related quality of life (HRQL) in revisions and nonrevision (NR) patients following adult spinal deformity (ASD) correction. However, a novel comparison approach may allow better comparisons in spine outcomes groups. OBJECTIVE: To determine if ASD revisions for radiographic and implant-related complications undergo a different recovery than NR patients. METHODS: Inclusion: ASD patients with complete HRQL (Oswestry Disability Index, Short-Form-36 version 2 (SF-36), Scoliosis Research Society [SRS]-22) at baseline, 6 wk, 1 yr, 2 yr. Generated revision groups: nonrevision (NR), revised-complete data (RC; with follow-up 2 yr after revision), and revised-incomplete data (RI; without 2-yr follow-up after revision). In a traditional analysis, analysis of variance (ANOVA) compared baseline HRQLs to follow-up changes. In a novel approach, integrated health state was normalized at baseline using area under curve analysis before ANOVA t-tests compared follow-up statuses. RESULTS: Two hundred fifty-eight patients were included with 50 undergoing reoperations (19.4%). Rod fractures (n = 15) and proximal joint kyphosis (n = 9) were most common. In standard HRQL analysis, comparing RC index surgery and RC revision surgery HRQLS revealed no significant differences throughout the 2-yr follow-up from either the initial index or revision procedure. Using normalized HRQL/integrated health state, RI displayed worse scores in SF-36 Physical Component Score, SRS activity, and SRS appearance relative to NR (P < .05), indicating less improvement over the 2-yr period. RC were significantly worse than RI in SF-36 Mental Component Score, SRS mental, SRS satisfaction, and SRS total (P < .05). CONCLUSION: ASD patients indicated for revisions for radiographic and implant-related complications differ significantly in their overall 2-yr recovery compared to NR, using a normalized integrated health state method. Traditional methods for analyzing revision patients' recovery kinetics may overlook delayed improvements.

PURPOSE: While there is a consensus that pelvic incidence (PI) remains constant after skeletal maturity, recent reports argue that PI increases after 60 years. This study aims to investigate whether PI increases with age and to determine potential associated factors. METHODS: 1510 patients with various spinal degenerative and deformity pathologies were enrolled, along with an additional 115 asymptomatic volunteers. Subjects were divided into six age subgroups with 10-year intervals. RESULTS: PI averaged 54.1 degrees in all patients. PI was significantly higher in the 45-54-year age group than 35-44-year age group (55.8 degrees vs. 49.7 degrees ). There were significant PI differences between genders after age 45. Linear regression revealed age, gender and malalignment as associated factors for increased PI with R 2 of 0.22 (p < 0.001). CONCLUSIONS: PI is higher in female patients and in older patients, especially those over 45 years old. Spinal malalignment also may have a role in increased PI due to increased L5-S1 bending moment.