Comparison of Ponte Osteotomies and 3-Column Osteotomies in the Treatment of Congenital Spinal Deformity.

BACKGROUND: Congenital spinal deformity (CSD) has traditionally been treated with 3-column osteotomies [hemivertebrectomy (HV) or vertebral column resection (VCR)] to address rigid deformities. Alternatively, multiple Ponte osteotomies (PO) may provide correction while minimizing risk. The purpose of this study was to compare safety and outcomes of patients undergoing surgical treatment for CSD with these 3 procedures. METHODS: Retrospective review of CSD patients treated with posterior spinal fusion between 1996 to 2013. Patients treated with multiple Ponte osteotomies (PO group) were compared with those managed with 3-column osteotomies (HV/VCR group). Patients with previous instrumentation, isolated cervical deformity, growing spine instrumentation, or <2 year follow-up were excluded. Deformity angular ratio (DAR) was calculated as curve magnitude divided by number of levels of the deformity. RESULTS: There were 49 patients [17 PO, 32 HV/VCR (26 HV, 6 VCR)]. For the PO group, mean age was 14 years, and they had an average of 4 ponte osteotomies and 11 levels fused. Mean total DAR was 25 and mean number of congenital anomalies was 1.8 in the PO group. The HV/VCR group had a mean age of 7 years and 5 levels fused. Mean total DAR was 28 and mean number of congenital anomalies was 2.1 in the HV/VCR group. Patients had a mean of 54.1% correction of coronal deformity in the PO group and 54.4% in the HV/VCR group (P=0.78). Signal changes were observed less frequently with PO (1/17) and HV (1/26) than with VCR (4/6), P=0.001. Revision rates were 17.6% (3/17) in the PO group and 37.5% (12/32) in the HV/VCR group (P=0.35). CONCLUSIONS: Patients with CSD and a mean total DAR of 25 treated with multiple PO and long fusions had correction comparable with the HV/VCR group. Patients treated with VCR had the highest incidence of signal changes and postoperative neurologic deficits. LEVEL OF EVIDENCE: Level III.

Removal of Infected Posterior Spinal Implants: Be Prepared to Transfuse.

STUDY DESIGN: Single-center retrospective review of spinal deformity patients undergoing removal of infected posterior spinal fusion implants over a 10-year period. OBJECTIVE: To evaluate the intraoperative blood loss and perioperative complications of implant removal in posterior spinal fusions. SUMMARY OF BACKGROUND DATA: To our knowledge, no studies examine blood loss or complications associated with removal of infected spinal implants in spinal deformity. METHODS: A retrospective review of 28 consecutive cases of infected posterior spinal fusion implant removal from 2003 to 2012 was performed. Exclusion criteria were patients with ≤6 levels of instrumentation, a partial removal of implants or a bleeding disorder. RESULTS: The average estimated blood loss was 465 mL (range 100-1,505 mL). Average estimated blood volume was 3,814 mL (range 1,840-9,264 mL). The average percentage of estimated blood loss was 14.2% (range 1.9%-43.5%). On postoperative labs obtained at the conclusion of the procedure, there was an average loss in hematocrit of 6.6 from preoperative values. Seventy-one percent of patients (20/28) received a blood transfusion; 39% (11/28) of these received a transfusion intraoperatively and 54% (15/28) received a transfusion postoperatively. Forty-six percent of patients (13/28) experienced an associated medical complication in the postoperative period. Among these 13, there were 16 total complications, with the most common being seizures (4/16), pneumonia (2/16), and sepsis (2/16). Average hospital stay was 14 days (range 4-52). CONCLUSION: Seventy-one percent of patients undergoing removal of infected spinal implants received a blood transfusion. We recommend having blood products available when removing posterior spinal instrumentation >6 levels. Patients and families should be counseled on the high risk of complications and expected hospital stay in these cases. LEVEL OF EVIDENCE: Level III.

Variability of the Anterior Humeral Line in Normal Pediatric Elbows.

BACKGROUND: The anterior humeral line (AHL) is considered a valuable radiographic tool in the assessment of pediatric elbow sagittal plane alignment following injury. However, few studies exist that examine the validity of the AHL. The purpose of this study is to report the variability of the AHL in skeletally immature children with normal elbows. METHODS: A total of 124 true lateral elbow radiographs of normal pediatric elbows were retrospectively identified and examined for the relationship of the AHL to the capitellum. The percentage of AHLs falling outside the middle third of the capitellum was compared among different age and sex groups using Fisher exact tests. RESULTS: In 100% of patients, the AHL touched the ossific nucleus of the capitellum. In 100% (52/52) of patients ≥5 years the AHL goes through the middle third of the capitellum, but this is significantly different from patients less than 5 years of age in whom 25% (18/72) of patients the AHL fell outside of the middle third of the capitellum (P<0.001). In children less than 2 years of age, the AHL was in the anterior third in 30% of the cases. CONCLUSIONS: Traditional teaching that the AHL touches the capitellum on a lateral radiograph of a normal elbow in a child is correct, so if the AHL does not touch the capitellum it is appropriate to look for pathology. Similarly, in children 5 years and older the AHL goes through the middle third of the capitellum in all patients, so if it does not, it is appropriate to look for pathology. However, with decreasing age variability increases, with the AHL touching the anterior third of the capitellum in almost 1/3 of children. LEVEL OF EVIDENCE: Level III.

Growing Rods Versus Shilla Growth Guidance: Better Cobb Angle Correction and T1-S1 Length Increase But More Surgeries.

STUDY DESIGN: Retrospective comparison. OBJECTIVES: To compare treatment of early-onset scoliosis (EOS) with Shilla growth guidance versus distraction-based dual growing rods (GR). SUMMARY OF BACKGROUND DATA: We are not aware of any prior studies comparing the Shilla procedure with other surgical procedures in the treatment of EOS. METHODS: The authors performed a multicenter case-matched comparison of patients with EOS treated with Shilla versus dual spine-spine GR from 1995 to 2009. A total of 36 Shilla patients from 3 centers were matched with 36 GR patients from the database by age at index surgery (±1 year), preoperative Cobb angle (±15°), and diagnosis (neuromuscular, congenital, idiopathic, or syndromic). Average follow-up was similar between groups (GR, 4.3 years; Shilla, 4.6 years; p = .353). RESULTS: Average Cobb angle improvement preoperatively to latest follow-up was 36° (range, 72° to 36°) in the GR group versus 23° (range, 69° to 45°) in the Shilla group (p = .0124). T1-S1 length increased 8.8 cm in patients treated with GR, compared with 6.4 cm in Shilla patients (p = .0170). Shilla patients had fewer surgeries (2.8) than patients in the GR group (7.4) (p < .001) but had a higher rate of unplanned surgeries for implant complications (Shilla, 1.3; GR, 0.5; p = .0151). When revisions for implant complications done at the time of scheduled lengthenings and revisions for construct maintenance were included, the groups did not differ significantly in the number of procedures for implant complications (Shilla, 1.4; GR, 1.5; p = .9451). The overall complication rate did not differ significantly between groups (Shilla, 1.9 [range, 0-7]; GR, 1.3 [range, 0-9]; p = .2085). CONCLUSIONS: The GR group had a greater improvement in Cobb angle and a greater increase in T1-S1 length than Shilla. The GR patients had more surgeries but Shilla patients had more unplanned procedures. The rate of complications overall did not differ significantly between groups.

Insurance status does not predict curve magnitude in adolescent idiopathic scoliosis at first presentation to an orthopaedic surgeon.

BACKGROUND: Previous studies have identified that children with public insurance have limited access to orthopaedic care. The purpose of this study was to explore the relationship between insurance status and curve magnitude at the time of presentation to an orthopaedic surgeon and time to treatment at a tertiary pediatric medical center. METHODS: This study was retrospective review of all patients with idiopathic scoliosis over 10 years, who have not had previous spine surgery. Data were collected on demographics, insurance type, curve magnitude at presentation, source of referral, treatment initiated, and time from recommendation for surgery to surgical intervention. RESULTS: Of the 642 patients included in this study, 53% were publicly insured and 45% were privately insured. Privately insured patients were significantly more likely to be seen as a second opinion (30% vs. 10%, P<0.001), and were significantly more likely to have received previous treatment (8% vs. 4%, P=0.011). Publicly insured patients were significantly more likely to be referred by their primary care doctor (64% vs. 50%, P=0.001) or as a part of school screening program (20% vs. 13%, P=0.036). At the time of presentation, there was no significant difference detected in major Cobb angles in the privately insured group [(private=28.7 (±15.4) degrees vs. public=26.4 (±16.8) degrees, P=0.076)]. There was no significant difference between the 2 groups in the number of patients who were recommended for operative treatment (public=11% vs. private 16%, P=0.072). However, in a multivariate regression analysis, publicly insured patients waited an average of 2.6 months longer for surgery than privately insured patients (P=0.010). CONCLUSIONS: Patients with private insurance presenting for evaluation of idiopathic adolescent scoliosis were significantly more likely to present as a second opinion than those with government insurance. In this group of 642 patients, no significant differences were found in major Cobb angle at presentation or eventual need for surgery. LEVEL OF EVIDENCE: Level III.

Iatrogenic nerve injuries in the treatment of supracondylar humerus fractures: are we really just missing nerve injuries on preoperative examination?

BACKGROUND: Recent studies report the rate of iatrogenic nerve injury in operatively treated supracondylar humerus (SCH) fractures is 3% to 4%. A reliable neurological examination can be difficult to obtain in a young child in pain. We hypothesized that nerve injuries may be missed preoperatively, later noted postoperatively in a more compliant patient, and then falsely considered an iatrogenic injury. METHODS: A prospective study was conducted on patients who presented between April 2011 and April 2013 with an extension-type SCH fracture that was managed surgically. A neurological examination was performed preoperatively, postoperatively, and at follow-up visits by a fellowship-trained attending pediatric orthopaedic surgeon. Only patients in whom the attending surgeon felt a reliable neurovascular examination was obtained were included in this study. RESULTS: Of the 100 patients, 16% had a nerve injury recognized on preoperative examination and 3% had a new nerve injury on postoperative examination (1 anterior interosseous, 1 median sensory, and 1 radial motor). The Gartland type (P=0.421), type of reduction (open vs. closed; P=0.720), and number of lateral-entry (P=0.898) or medial-entry (P=0.938) pins used were not associated with patients who had a new nerve injury found postoperatively. A trend was seen between fracture severity and rate of a preoperative nerve injury: type II 7% (2/28), type III 19% (9/58), and type IV 36% (5/14) (P=0.058). Preoperatively, nerve injuries were noted at the following rates: median 12% (12/100) (including 8 anterior interosseous nerve injuries), radial 8% (8/100), ulnar 3% (3/100). CONCLUSIONS: In this prospective study, in patients who were able to comply with a preoperative neurological examination done by an attending pediatric orthopaedic surgeon, the rate of iatrogenic nerve injury after operative treatment of SCH fractures is 3%. We conclude that this finding is true, and not a result of inadequate preoperative neurological examinations. LEVEL OF EVIDENCE: Level I prognostic study.

Screening for hip dysplasia in congenital muscular torticollis: is physical exam enough?

PURPOSE: An association between congenital muscular torticollis (CMT) and developmental dysplasia of the hip (DDH) has been established in the literature; however, whether the screening of patients with CMT for DDH requires hip imaging remains controversial. The purpose of this study is to determine (1) the coexistence rate of DDH requiring treatment in individuals with CMT and (2) if physical exam alone is sufficient screening. METHODS: A single-center retrospective chart review was performed among 97 consecutive patients between 1/1/2003 and 9/1/2012 with CMT who had hip imaging performed. RESULTS: 12 % (12/97) of patients with CMT had DDH, all requiring treatment. 75 % (9/12) of the patients with DDH had an abnormal clinical exam. Of the three patients with DDH and a normal clinical exam, two patients were presenting for a second opinion after being treated for DDH prior to evaluation. 90 % (9/10) of patients with DDH at the time of presentation had an abnormal hip exam. All 12 patients with hip dysplasia were referred for DDH or DDH with CMT. There were no patients who were referred for CMT alone that had DDH. CONCLUSIONS: In the care of a patient with CMT, it is important that the clinician remains vigilant about screening for DDH. An ultrasound or radiograph of the hips should be strongly considered as part of the evaluation of a child with CMT. LEVEL OF EVIDENCE: IV.

Mechanisms and risk factors of brachial plexus injury in the treatment of early-onset scoliosis with distraction-based growing implants.

BACKGROUND: Brachial plexus injuries have been reported in association with distraction-based instrumentation for early-onset scoliosis. The purpose of this study was to describe brachial plexus injuries associated with distraction-based spine instrumentation with rib anchors and the mechanisms and risk factors responsible. METHODS: We performed a retrospective single-center review of a consecutive series of forty-one patients with early-onset scoliosis who underwent distraction-based instrumentation with rib anchors from 2000 to 2011. RESULTS: Four (10%) of the forty-one patients experienced an intraoperative brachial plexus injury. Three mechanisms of brachial plexus injuries were identified: (1) injury of the brachial plexus by the first rib being pushed superiorly by rib-anchored growing instrumentation, (2) direct injury to the brachial plexus by the superior pole of the retracted scapula, and (3) injury of the brachial plexus when the scapula was moved inferiorly during Sprengel deformity reconstruction. The last two mechanisms are independent of spinal instrumentation. Two patients had neurological symptoms or neuromonitoring signal changes when the arm was in the adducted position but not when the arm was abducted. All patients had complete neurological recovery. CONCLUSIONS: Patients with Sprengel deformity appear to be at increased risk for brachial plexus injury when undergoing distraction-based spine instrumentation with rib anchors. Injury to the brachial plexus can occur with scapular elevation alone, presumably by direct compression of the superior end of the scapula on the brachial plexus. Brachial plexus injuries may be "hidden" during monitoring of an arm in shoulder abduction but symptomatic with shoulder adduction, as the brachial plexus is draped over the elevated first rib. LEVEL OF EVIDENCE: Therapeutic level IV. See Instructions for Authors for a complete description of levels of evidence.