Study Design. Retrospective comparison study of patients who had a delayed infection following a posterior spinal fusion and instrumentation (PSFI) for adolescent idiopathic scoliosis (AIS). Objective. To define risk factors for the development of delayed infections following PSFI for AIS by comparing those patients who developed this complication to a randomly selected group of patients who did not. Summary of Background Data. Despite studies reporting the incidence and treatment of delayed infection following PSFI for AIS, there are no studies analyzing risk factors for its occurrence. Methods. All patients who required treatment for delayed infections following PSFI for AIS were identified (infection group, n = 36). A random selection of patients who did not develop a delayed infection (no infection, n = 90) was made in a ratio of 3:1 (no infection/infection). The 2 groups were compared using statistical methods. Results. Parameters associated with the infection group included: presence of a significant medical history, surgeon, less surgical time, a more distal fusion level (16% infection rate with a thoracic LIV vs. 33% infection rate with a lumbar LIV), not using postoperative drains, and increased drainage when drains were used. Other factors associated with infection were use of a blood transfusion and when increasing units of transfusion were used. Multivariate logistic regression analysis identified 3 factors that remained statistically significant: 1) significant medical history, 2) receiving a blood transfusion, and 3) not using a postoperative drain. Factors that were not associated with delayed infection included body mass index, the number of anchor points used, use of allograft bone, and the total number of levels instrumented and antibiotic regimen. Conclusion. The occurrence of a delayed infection is most likely multifactorial and is related to a positive past medical history and the use of blood transfusions. Postoperative use of a drain may be important to avoid delayed infection.

Case series retrospective review. To identify what factors predict successful eradication of infection after I&D of an infected posterior spinal fusion with instrumentation. The treatment of infection of instrumented spine fusions in children has few clear guidelines in the literature. The medical records of patients who required a surgical irrigation and debridement (I&D) for infection after posterior spinal fusion and instrumentation for scoliosis from 1995 to 2002 were retrospectively reviewed. Fifty-three patients were identified with the following underlying diagnoses: 21 patients (40%) idiopathic scoliosis, 10 patients (23%) cerebral palsy, 3 patients (6%) spina bifida, 1 patient (2%) congenital scoliosis, and 17 patients (32%) other. There were 31 patients (58%) with surgery 6 months. Of the 43 patients with implant retained at the time of the first I&D, 20 patients required a second I&D (47%). Of the 10 patients with complete implant removal, 2 patients required a second I&D (20%). Coagulase-negative Staphylococcus was the most prevalent organism, growing in 25 (47%) of the cultures. Of patients with idiopathic scoliosis, 8 of 21 (38%) required a second I & D; of the patients with other diagnoses, 14 of 32 (44%) required a second I&D, which was not a significant difference (P > 0.05). To the best of our knowledge, this is the largest reported series of spinal implant infections. When children with an infection after posterior spinal fusion with instrumentation undergo irrigation and debridement, there is a nearly 50% chance that the infection will remain if all spinal implants are not removed. As nearly 50% of the infections were caused by coagulase-negative Staphylococcus, we recommend that prophylactic antibiotic coverage for this organism is used at the time of the initial spinal fusion.

Flexible intramedullary nails are now routinely used for stabilization of pediatric femur fractures. Few data are available regarding patients' postoperative range of motion, weight-bearing status, activity levels, use of immobilization, and radiographic leg length discrepancy measured via routine scanograms. Patients who underwent placement of flexible intramedullary nails for a pediatric femur fracture at a single institution from 1998 to 2003 were identified retrospectively. Ninety-one patients were identified with 94 femur fractures. The complication rate was 17% for the 94 fractures, with 8 patients requiring an unplanned return to surgery. The complication rate was significantly higher for patients aged 10 years or older (34%) as compared with that for younger patients (9%). Average time to full weight bearing was 10 weeks, time to radiographic union averaged 10.7 weeks, and time to return to preoperative level of activity averaged 4.9 months. Immediate postoperative weight bearing status was nonweight bearing in 57%. Immobilization or support was used postoperatively in 60% of the patients. Postoperatively, patients had minimal loss of range of motion in hip internal and external rotation and knee extension. Hip and knee flexion rapidly improved postoperatively with an average loss of hip flexion of 0 degree by 3 months and an average loss of knee flexion of 4 degrees by 6 months. Postoperatively, limb length discrepancy was greater than 1 cm in 7 patients at 6 months, 11 patients at 12 months, 3 patients at 18 months, and 2 patients at 2 years. Two patients had persistent limb length discrepancy of greater than 2 cm, but only one patient required an epiphysiodesis for his limb length inequality. Although the end results are favorable, complications are relatively frequent, particularly in older children.

The safety of delayed surgical treatment of severe supracondylar elbow fractures in children remains debated. No large studies have evaluated complications of injury and surgery evaluating only type 3 fractures. Our aim was to review the results of our experience treating children with severe supracondylar elbow fractures at various time points after injury. All children treated operatively for supracondylar humerus fractures from 2004 to 2007 at a single pediatric trauma center were identified. A total of 1296 children had operative treatment, of which 872 had type 3 fractures. Clinical records were reviewed to identify time to surgery from presentation at our institution. Patients were grouped into 4 cohorts [24 h (n=28)]. Emergency, operative, inpatient, and outpatient records were reviewed to determine morbidity at presentation as well as operative and postoperative complications. There was no difference in sex, age, or energy mechanism between children in the various time groups. An absent pulse was found in 54 children (6%) at presentation, of which only 5 ultimately required a vascular intervention. Nerve injury occurred in 105 patients (12%). Use of a medial entry pin was not associated with ulnar nerve injury. Increased time from presentation to surgery was not associated with increased morbidity from the injury or treatment complications. In contrast, there was a trend to steady decrease in morbidity and complication rates with increased time to surgery. This is the largest single-center study of severe supracondylar humerus fractures and describes rates of vascular compromise, nerve injury, infection, and other complications of these injuries. Most children with type 3 supracondylar humerus fractures can be treated safely in a delayed manner. Appropriate clinical judgment is imperative to optimize outcomes. Level III--retrospective comparative study.

N consecutive, operatively treated supracondylar humeral fractures in patients presenting to a level-I pediatric trauma center from January 2003 through December 2007 were studied retrospectively. Clinical records were reviewed to determine vascular and neurological examination findings, Gartland classification, timing of surgery, and postoperative complications. Results: One thousand two hundred and sixty-six patients had a documented radial pulse examination at the time of arrival in the emergency room; fifty-four (4%) of those patients lacked a palpable radial pulse. All fifty-four patients had type-3 fractures. Five (9%) of the fifty-four patients underwent open exploration of vascular structures on the basis of clinical findings of a pale hand, sluggish capillary refill, and/or weak or no pulse detected with use of Doppler ultrasound after closed reduction and percutaneous pinning. All five underwent vascular surgery to restore blood flow (two primary repairs, three saphenous vein grafts). Twenty (37%) of the fifty-four patients had a pulse documented with use of Doppler ultrasound and a pink hand after closed reduction and percutaneous pinning, but the radial pulse remained nonpalpable. These patients were observed in the hospital for signs of ischemia; one of the twenty patients required vascular repair after developing a pale hand nine hours after closed reduction and percutaneous pinning, and the other nineteen patients were also observed while they were in the hospital, and they all regained a palpable pulse either prior to discharge or by the time of the first postoperative visit. When compared with the group of patients with type-3 fractures for whom data regarding nerve examination were available, patients with type-3 fractures who lacked a palpable radial pulse had a higher rate of nerve palsy postoperatively (31% versus 9%, p < 0.0001). Conclusions: In this cohort, nearly 10% of patients who presented with a type-3 supracondylar humeral fracture and no palpable radial pulse underwent immediate vascular repair to restore blood flow following closed reduction and percutaneous pinning. However, in our series, the lack of a palpable radial pulse after closed reduction and percutaneous pinning was not an absolute indication to proceed with vascular exploration if clinical findings (i.e., Doppler signal and capillary refill) suggested that the limb was perfused. Careful inpatient monitoring of these patients postoperatively is mandatory to identify late-developing vascular compromise. Level of Evidence: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence....

Approximately 5% of supracondylar humerus fractures in children are associated with an ipsilateral forearm fracture, often referred to as a floating elbow when both injuries are displaced. Historically, these patients have higher complication rates than patients with an isolated supracondylar humerus fracture. The purpose of this study was to review the acute neurologic and vascular injuries in patients with ipsilateral, operative supracondylar humerus and forearm fractures and compare the findings with a cohort of isolated, operative supracondylar humerus fractures. We performed an IRB-approved, retrospective review of all pediatric patients with ipsilateral, operative supracondylar humerus and forearm fractures from a single institution and compared our findings to a cohort of isolated, operative supracondylar humerus fractures. A total of 150 patients with operative supracondylar humerus and ipsilateral forearm fractures were compared with 1228 patients with isolated, operative supracondylar humerus fractures. Twenty-two of the 150 (14.7%) floating elbow patients had documented pretreatment nerve palsies compared with 96/1228 (7.8%) of isolated injury patients (P=0.006). Eighteen of 22 nerve palsies were in patients with forearm fractures that required reduction. The overall incidence of nerve palsy was 18.9% (18/95) when a forearm fracture required reduction compared with only 7.3% (4/55) in a forearm fracture that was not reduced (P=0.05). We did not find a significant difference in the rate of pulseless extremities when comparing the ipsilateral (6/150 4%) and isolated (50/1228 4.1%) injury patients. No compartment syndromes were identified in any patient with an ipsilateral injury. The rate of acute neurologic injury in ipsilateral supracondylar humerus and forearm fractures is almost twice than that found in patients with isolated supracondylar humerus fractures. This rate increases further when the forearm fracture requires a manipulative reduction. The likelihood of a pulseless extremity was not dependent upon the presence of a forearm injury in our study. The presence of an ipsilateral forearm fracture should alert the surgeon to carefully assess the preoperative neurovascular status of patients with supracondylar humerus injuries. Level III.