Mark Erickson

Chiari I Malformation Associated with Syringomyelia and Scoliosis : A Twenty-Year Review of Surgical and Nonsurgical Treatment in a Pediatric Population

Study Design. Retrospective review of patients with Chiari I malformation with or without associated scoliosis. Objectives. Determine the effect of decompression of Chiari I malformation with syringomyelia on stabilization or improvement of associated scoliosis. Summary of Background Data. Chiari malformations are often associated with spinal deformities, including scoliosis. Studies have suggested a causal relation between syringomyelia and scoliosis. Methods. Patients with Chiari I malformation and syringomyelia with or without scoliosis treated over the last 20 years were reviewed. Patients with any other anomalies were excluded. Scoliotic curves were classified by magnitude and curve type. All patients were treated with surgical decompression of the Chiari malformation with or without drainage of the syringomyelia. Results. Twenty-five patients were identified, ranging in age from 19 months to 16.5 years. Nineteen patients (76%) had associated scoliosis. The majority of the patients with scoliosis (13 of 19) sought treatment for spinal deformity, and only 6 had for pain or neurologic symptoms. Eleven of 19 patients with scoliosis (58%) underwent fusion. Eight of 19 (42%) patients have not undergone fusion: 3 have experienced progress, 1 remains in a stable condition, and 4 have experienced improvement of curvature since undergoing decompression. The mean age of patients who experienced progress after decompression was 14.5 years, compared to 6 years for patients who experienced improvement. Conclusion. Early decompression of Chiari I malformation with syringomyelia and scoliosis resulted in improvement or stabilization of the spinal deformity in 5 cases. Each of these patients underwent decompression before 8 years of age and before the curve was severe. However, this series represents a few patients demonstrating this trend, and further follow-up and investigation are warranted.

Non-Neurologic Complications Following Surgery for Adolescent Idiopathic Scoliosis

BACKGROUND The reported prevalence of non-neurologic complications following corrective surgery for adolescent idiopathic scoliosis ranges from 0% to 10%. However, most studies were retrospective evaluations of treatment techniques and did not focus solely on complications. The purpose of this study was to determine the prevalence of non-neurologic complications following surgery for adolescent idiopathic scoliosis and to identify preoperative and operative factors that can increase this risk. METHODS The demographic data, medical and surgical histories, and prevalence of non-neurologic complications were reviewed in a prospective cohort of 702 patients who had undergone corrective surgery for adolescent idiopathic scoliosis and were consecutively enrolled in a multicenter database. RESULTS There were 556 female and 146 male patients. The mean age at the time of surgery was 14.25 years (range, eight to eighteen years). Five hundred and twenty-three patients had only posterior spinal surgery, 105 had only anterior spinal surgery, and seventy-four had a combined anterior and posterior procedure. There was a total of 108 complications in eighty-one patients, for an overall prevalence of 15.4%. There were ten respiratory complications (1.42%), six cases of excessive bleeding (0.85%), five wound infections (0.71%), and five cases of wound hematoma, seroma, or dehiscence (0.71%). Five patients, two with an early infection and three with late failure of the implant, required a reoperation. Factors that did not correlate with an increased prevalence of complications were age, body mass index, presence of cardiac or respiratory disease, previous surgery, pulmonary function, surgical approach, number of levels fused, graft material, use of a diaphragmatic incision, Lenke curve type, or region of the major curve. Although the number of patients with renal disease was small, these patients were 7.90 times more likely to have a non-neurologic complication. Increased blood loss as well as prolonged operative and anesthesia times were associated with a higher prevalence of non-neurologic complications. CONCLUSIONS The prevalence of non-neurologic postoperative complications following surgery for correction of adolescent idiopathic scoliosis in this study was 15.4%. The few factors noted to significantly increase the rate of complications include a history of renal disease, increased operative blood loss, prolonged posterior surgery time, and prolonged anesthesia time.

Building consensus: development of a Best Practice Guideline (BPG) for surgical site infection (SSI) prevention in high-risk pediatric spine surgery.

Background: Perioperative surgical site infection (SSI) after pediatric spine fusion is a recognized complication with rates between 0.5% and 1.6% in adolescent idiopathic scoliosis and up to 22% in “high risk” patients. Significant variation in the approach to infection prophylaxis has been well documented. The purpose of this initiative is to develop a consensus-based “Best Practice” Guideline (BPG), informed by both the available evidence in the literature and expert opinion, for high-risk pediatric patients undergoing spine fusion. For the purpose of this effort, high risk was defined as anything other than a primary fusion in a patient with idiopathic scoliosis without significant comorbidities. The ultimate goal of this initiative is to decrease the wide variability in SSI prevention strategies in this area, ultimately leading to improved patient outcomes and reduced health care costs. Methods: An expert panel composed of 20 pediatric spine surgeons and 3 infectious disease specialists from North America, selected for their extensive experience in the field of pediatric spine surgery, was developed. Using the Delphi process and iterative rounds using a nominal group technique, participants in this panel were as follows: (1) surveyed for current practices; (2) presented with a detailed systematic review of the relevant literature; (3) given the opportunity to voice opinion collectively; and (4) asked to vote regarding preferences privately. Round 1 was conducted using an electronic survey. Initial results were compiled and discussed face-to-face. Round 2 was conducted using the Audience Response System, allowing participants to vote for (strongly support or support) or against inclusion of each intervention. Agreement >80% was considered consensus. Interventions without consensus were discussed and revised, if feasible. Repeat voting for consensus was performed. Results: Consensus was reached to support 14 SSI prevention strategies and all participants agreed to implement the BPG in their practices. All agreed to participate in further studies assessing implementation and effectiveness of the BPG. The final consensus driven BPG for high-risk pediatric spine surgery patients includes: (1) patients should have a chlorhexidine skin wash the night before surgery; (2) patients should have preoperative urine cultures obtained; (3) patients should receive a preoperative Patient Education Sheet; (4) patients should have a preoperative nutritional assessment; (5) if removing hair, clipping is preferred to shaving; (6) patients should receive perioperative intravenous cefazolin; (7) patients should receive perioperative intravenous prophylaxis for gram-negative bacilli; (8) adherence to perioperative antimicrobial regimens should be monitored; (9) operating room access should be limited during scoliosis surgery (whenever practical); (10) UV lights need NOT be used in the operating room; (11) patients should have intraoperative wound irrigation; (12) vancomycin powder should be used in the bone graft and/or the surgical site; (13) impervious dressings are preferred postoperatively; (14) postoperative dressing changes should be minimized before discharge to the extent possible. Conclusions: In conclusion, we present a consensus-based BPG consisting of 14 recommendations for the prevention of SSIs after spine surgery in high-risk pediatric patients. This can serve as a tool to reduce the variability in practice in this area and help guide research priorities in the future. Pending such data, it is the unsubstantiated opinion of the authors of the current paper that adherence to recommendations in the BPG will not only decrease variability in practice but also result in fewer SSI in high-risk children undergoing spinal fusion. Level of Evidence: Not applicable.

What's the evidence? Systematic literature review of risk factors and preventive strategies for surgical site infection following pediatric spine surgery.

Background: Despite relatively high rates of surgical site infections (SSIs) after pediatric spine surgery, practice guidelines are absent. We performed a systematic review of the literature, determining the level of evidence for risk factors for SSIs and prevention practices to reduce SSIs following pediatric spine surgery. Methods: The search utilized the root search words “spine,” “scoliosis,” and “infection” resulting in 9594 abstracts. Following removal of duplicate abstracts, those that assessed only SSI rates, SSI treatment, nonoperative spine infections, or adult populations, 57 relevant studies were rated for level of evidence and graded using previously validated scales. Results: Very few studies lead to grade A (good evidence) or grade B (fair evidence) recommendations. Ceramic bone substitute did not increase the risk of SSIs when compared with autograft (grade A). Comorbid medical conditions, particularly cerebral palsy or myelodysplasia; urinary or bowel incontinence; nonadherence to antibiotic prophylaxis protocols; and increased implant prominence increase the risk of SSIs (grade B). SSIs caused by gram-negative bacilli were more frequent in neuromuscular populations and first-generation stainless steel implants increased the risk of delayed infection compared to newer generation titanium implants (grade B). Evaluations of other risk factors for SSIs yielded conflicting or poor-quality evidence (grade C); these included malnutrition or obesity; number of levels fused or fusion extended to the sacrum/pelvis; blood loss; and use of allograft. Insufficient evidence (0 to 1 published studies) was available to recommend numerous practices shown to reduce SSI risk in other populations such as chlorhexidine skin wash the night before surgery, preoperative nasal swabs for Staphylococcus aureus, chlorhexidine skin disinfection, perioperative prophylaxis with intravenous vancomycin, vancomycin, or gentamicin powder in the surgical site or graft. Conclusions: Few studies have evaluated risk factors and preventive strategies for SSIs following pediatric spine surgery. This systematic review documents the relative lack of evidence supporting SSI prevention practices and highlights priorities for research. Level of Evidence: Level III therapeutic study.

Correlation of preoperative deformity magnitude and pulmonary function tests in adolescent idiopathic scoliosis

Study Design. Clinical study correlating preoperative pulmonary function tests (PFTs) to radiographic measures of thoracic deformity severity in adolescent idiopathic scoliosis (AIS) patients. Objective. To determine (1) the incidence of clinically relevant (<65% predicted value) pulmonary impairment in AIS patients; (2) if patients with more severe deformity have greater impairment of PFTs than those with lesser deformity; (3) the effect, if any, of juvenile onset deformity (onset <age 10) on preoperative PFTs. Summary of Background Data. Patients with late-onset (adolescent) spinal deformity are generally believed to have no respiratory morbidity except in severe curves exceeding 100°. Methods. A large multicenter database of surgically treated AIS patients with Lenke 1 to Lenke 4 curves was queried to report preoperative PFTs and correlate them with radiographic measures of coronal, sagittal, and axial plane deformities. Results. Nineteen percent of 858 patients had <65% predicted forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC) before surgery, and had larger main thoracic (MT) curves and greater axial rotation than those with predicted PFT values >65%. Patients with MT curves >70°, and especially >80°; proximal thoracic (PT) curves which were >30° or structural; or T5–T12 kyphosis <10° had significantly (P ⩽ 0.001) lower FEV1 or FVC compared to those with less deformity. Axial plane deformity did not correlate with PFT impairment. Juvenile-onset patients had greater PFT impairment than AIS patients, along with slightly larger MT curves. Patients who were braced before surgery had worse PFTs than those had no treatment before surgery. Conclusion. Preoperative PFTs are clinically impaired in 19% of AIS patients, and correlate significantly with the MT and sagittal plane deformity severity, and with PT curve severity to a lesser degree. PFTs do not correlate with degree of axial deformity. From a purely pulmonary standpoint, attention directed to coronal and sagittal plane deformity correction appears warranted, to address the specific deformities which are associated with PFT impairment.

The role of serial casting in early-onset scoliosis (EOS).

Background: Serial casting has demonstrated efficacy for idiopathic early-onset scoliosis (EOS). Results of casting in nonidiopathic (syndromic and congenital) EOS patients have not previously been well described. Methods: A total of 53 patients underwent serial casting for EOS from 2005 to 2010 at a single institution. Deformity was classified as idiopathic or nonidiopathic. Diagnosis, time in cast, number of casts, use of bracing, complications, and outcomes were recorded. Radiographic measures included Cobb angle and thoracic height (T1-T12). Thoracic height velocity was calculated and compared with established norms. Results: A total of 36 patients, 19 idiopathic and 17 nonidiopathic (14 syndromic, 3 congenital), completed cast treatment and had >6-month follow-up and were therefore included. Of those, 17% (6/36) experienced resolution of their deformity, 53% (19/26) are currently in braces, and 31% (11/36) had undergone surgery. Surgery occurred on average at age 5.6 years and was delayed by an average of 2.1 years from time of first cast. A 19% complication was observed. There was no statistical difference in the rate of resolution of deformity between idiopathic (5/19) and nonidiopathic (1/17) patients (P=0.182), although there exists a trend toward greater curve correction in idiopathic patients. Surgery occurred in fewer patients (2/19) in the idiopathic group compared with the nonidiopathic group (9/17) (P=0.006). Significant improvements in Cobb angle was observed in the idiopathic group (12.2 degrees) during casting (P=0.003). Nonidiopathic patients did not maintain the correction gained during casting at the time of final follow-up. T1-T12 height increased across all study patients regardless of etiology during the period of casting at similar velocity to established norms of 1.4 cm/y for this age group. Conclusions: Serial casting offers modest deformity correction in idiopathic deformities compared with nonidiopathic deformities. Thoracic height growth continued throughout the casting period at normal velocity. Serial casting maintained normal longitudinal thoracic growth in all patients with EOS in this cohort. Although many required surgery, the increased thoracic height may have positive implications on ultimate pulmonary function. Level of Evidence: Therapeutic level III.

Thoracoscopic anterior spinal procedures in children

METHODS From February 1996 through July 1997, 20 patients were determined to be appropriate for a thoracoscopic approach for the anterior portion of their spinal surgery. Ages ranged from 8 to 17 years of age and weight from 20 to 70 kg. The diagnosis included severe idiopathic and neurogenic scoliosis or kyphosis (n = 20) and congenital hemivertebra (n = 4). The procedures performed included diskectomy and anterior release from two to nine vertebral levels, hemivertebrectomy, and epiphysiodesis. Four patients also had an open lumbar exposure and all had a posterior fusion. RESULTS All procedures were completed successfully endoscopically. Surgical times for the thoracoscopic portion of the procedure averaged 106 minutes and total procedure times were comparable with the standard open technique. Three patients were extubated at the end of the procedure with the majority extubated on postoperative day 1. Average intensive care unit stay was 1.8 days, and chest tubes were removed between postoperative day 1 and 5 (average, 2.2 days). In follow-up, surgical correction was deemed to be acceptable and equivalent to open techniques in all cases. CONCLUSION This technique has proven to be safe and effective in children and appears to be associated with less pain and morbidity as evidenced by earlier extubation and chest tube removal and shorter ICU stay.

Use and outcomes of wound drain in spinal fusion for adolescent idiopathic scoliosis.

Study Design. A multicenter retrospective analysis. Objective. To evaluate outcomes of closed-suction wound drainage after posterior spinal fusion with instrumentation for adolescent idiopathic scoliosis and to identify surgeon patterns of drain use in this cohort. Summary of Background Data. There is little evidence on the use of drains in spinal surgery, particularly for repair of adolescent idiopathic scoliosis. Studies on hip and knee arthroplasty suggest no advantage to draining. There are few published reports on surgeon technique and rationale for drain use in spinal surgery. Methods. Patients were divided into drain and no drain cohorts and followed for 2 years. Primary outcome was complication rate. A separate survey was conducted from surgeons in the Spinal Deformity Study Group to evaluate drain practice patterns. Results. There were 324 drained and 176 undrained patients. Complication rate did not differ between the drain and no drain cohorts in any of the 4 categories (wound infection, neural injury, other infection, and other complication) at any time (all P > 0.1). More drained patients received postoperative transfusions compared with those without a drain (43% vs. 22%, P < 0.001). Of the 50 surgeons in the group, 36 used drains. Half of these did so out of habit. Surgeons tended to place deep drains with bulb suction, without drain manipulation. Half removed drains on the basis of output, whereas half removed them after 1 to 3 days. Conclusion. More patients tended to receive wound drains than not receive wound drains. Drains did not impact complication rate and drained patients received more blood product. There are no universal criteria for draining and practice patterns vary widely.

A multidisciplinary approach improves infection rates in pediatric spine surgery.

Background: Surgical site infections (SSI) associated with elective pediatric spinal surgery are a commonly reported complication, increasing hospital length of stay, readmissions, operations, and financial costs. In July 2007, a multidisciplinary task force, designated Target Zero, was created to address this issue and establish prevention protocols at our institution. Methods: A consecutive series of 394 patient charts from April 2006 to September 2008 were retrospectively reviewed to identify patients who developed an SSI secondary to elective spinal surgery. Four cohorts were evaluated; high-risk (HR) and low-risk (LR) patients who underwent surgery before (April 2006 to June 2007) and after (July 2007 to September 2008) Target Zero initiation. The definition of HR included diagnoses of cerebral palsy, spina bifida, muscle disease, paralytic deformities, and vertebral column resections. Patients were followed for 1 year to meet The Center for Disease Control-National Health Safety Network’s definition of an SSI with an implantable device. Overall infection rates were determined for each group and compared statistically. Results: A total of 192 patients (70 HR and 122 LR) underwent surgery before, and 202 patients (92 HR and 110 LR) underwent surgery after Target Zero initiation. Overall infection rates were reduced from 7.8% to 4.5% (P=0.203), 12.9% to 6.5% (P=0.183), and 4.9% to 2.7% (P=0.505) for all patients, HR patients, and LR patients, respectively. The relative risk reduction was 43.0% for all patients, 49.3% for HR patients, and 44.6% for LR patients. Conclusions: Although decreases in overall infection rates were not statistically significant, the results from Target Zero were shown to be clinically meaningful with a relative risk reduction approaching 50% overall and in defined subgroups. Based on the number needed to treat analysis, 1 infection in every 16 patients within the HR group, and 1 in 30 overall, was prevented up to 1 year postoperatively. This study is the first to document the effectiveness of a multidisciplinary team implementing protocols for decreasing infection rates in pediatric spine surgery.

Biomechanical Assessment of Conventional Unit Rod Fixation : A Human Cadaver Study versus : A Human Cadaver Study a Unit Rod Pedicle Screw Construct: A Human Cadaver Study

Study Design. Fresh frozen human cadaver specimens were used to perform a biomechanical assessment focusing on spinal-pelvic fixation. Unit rod fixation alone was compared with that of unit rod fixation supplemented with L5 pedicle screws. Objectives. To investigate the biomechanical effects of adding L5 pedicle screw fixation to a unit rod construct. Summary of Background Data. The treatment of neuromuscular scoliosis associated with pelvic obliquity continues to evolve. Spinal fusion using fixation extending from the thoracic spine to the pelvis supplemented with segmental instrumentation has become the standard of care. The instrumentation of choice used at this institution has been a unit rod construct. Traditional Luque-Galveston technique with sublaminar wiring from T2 to L5 has been used. A noted complication has been a loss of distal fixation, the emphasis for this study. We have subsequently modified this technique, replacing wires at L5 with pedicle screws. This was done to improve the strength of the distal fixation, with the goal of decreasing the rate of instrumentation failure before fusion. Methods. Twelve fresh-frozen human cadavers, between the ages of 45 and 60 years, were instrumented from T4 to the pelvis with a Unit Rod (Sofamor Danek). The specimens were randomized into two groups. Segmental sublaminar wiring was performed from T4 to L5 in one group of six spines. The remaining six spines underwent sublaminar wiring from T4 to L4, with pedicle screws used for L5 fixation. The spines were then potted from T4 to T7 in preparation for biomechanical testing. The spines were tested in a randomized fashion for subfailure stiffness in anteroposterior, oblique, and lateral planes with an Instron 1321 servo-hydraulic load frame (Instron Corp., Canton, MA). Each spine was then loaded to failure in forward flexion. Results. Our results demonstrate a statistically significant increase in lateral stiffness (P = 0.0006) and oblique stiffness (P = 0.003) associated with the use of L5 pedicle screws. Statistical significance was also achieved with load to failure testing (P = 0.0007). The mean load to failure was 1377 Newtons for the pedicle screw construct and 467 Newtons for the wire construct. Conclusions. We conclude that the addition of L5 pedicle screws increases the stiffness and strength of the unit rod construct. Clinically, this should help to avoid complications associated with loss of fixation in this area.