Intrathecal morphine (IM) is a common adjunct in paediatric spinal deformity surgery. We previously demonstrated with idiopathic scoliosis it provides safe and effective analgesia in the immediate postoperative period. This study represents our 25 year experience with IM in all diagnostic groups.

Our prospective Pediatric Orthopaedic Spine Database (1993–2018) was reviewed to identify all patients undergoing spinal deformity surgery who received IM and who did not. Patients 21 years of age or less who had a posterior spinal fusion (PSF) with segmental spinal instrumentation (SSI), and received 9–19 mcg/kg (up to 1 mg) of IM were included. Early onset scoliosis surgical patients were excluded. We assessed demographics, pain scores, time to first dose of opioids, diagnoses, surgical time, paediatric intensive care unit (PICU) admission and IM complications (respiratory depression, pruritus, nausea/ vomiting).

There were 986 patients who met inclusion criteria. This included 760 patients who received IM and 226 who did not. IM was not used for short procedures (< 3 hrs), respiratory concerns, unsuccessful access of intrathecal space, paraplegia, and anesthesiologist decision. Both groups followed the same strict perioperative care path. The patients were divided into 5 diagnostic groups (IM / non IM patients): idiopathic (578/28), neuromuscular (100/151), syndromic (36/17), and congenital scoliosis (32/21) and kyphosis (14/9). Females predominated over males (697/289). The first dose of opioids after surgery was delayed for a mean of 10.6 hrs in IM group compared to 2.3 hrs in the non-IM group (p=0.001). The postoperative pain scores were significantly lower in the IM groups in the Post Anesthesia Care Unite (p=0.001). Only 17 IM patients (2%) were admitted to the PICU for observation secondary to respiratory depression, none required re-intubation. None of the IM group were re-intubated. Forty-nine patients (6%) experienced pruritus in the IM group compared to 4 of 226 patients (2%) in the non IM group. There were 169 patients (22%) of the IM patients and 21 patients (9%) of the non IM had nausea and vomiting postoperatively. Three patients (0.39%) had a dural leak from the administration of IM but did not require surgical repair. There were no other perioperative complications related to the use of IM. There were no significant group differences.

Pre-incision IM is a safe and effective adjunct for pain management in all diagnostic groups undergoing spinal surgery. The IM patients had lower pain scores and a longer time to first administration of post-operative opioids. Although there is an increased frequency of respiratory depression, pruritus, and nausea/vomiting in the IM group, there were no serious complications.

Unstable slipped capital femoral epiphysis (SCFE) has an increased incidence of avascular necrosis (AVN). The purpose of this study was to determine if early identification and intervention for AVN may help preserve the femoral head.

We retrospectively reviewed 48 patients (50 hips) with unstable SCFE managed between 2000 and 2014. Based on two different protocols during the same time period, 17 patients (17 hips) had a scheduled MRI between 1 and 6 months from initial surgery, with closed bone graft epiphysiodesis (CBGE) or free vascularised fibular graft (FVFG) if AVN was diagnosed. Thirty-one patients (33 hips) were evaluated by plain radiographs. Outcomes analysed were Steinberg classification and subsequent surgical intervention. We defined Steinberg class IVC as failure in treatment because all of the patients referred for osteotomy, arthoplasty, or arthrodesis in our study were grade IVC or higher.

Overall, 13 hips (26%) with unstable SCFE developed AVN. MRI revealed AVN in 7 of 17 hips (41%) at a mean of 2.5 months postoperatively (range, 1.0 to 5.2 months). Six hips diagnosed by MRI received surgical intervention (4 CBGE, 1 FVFG, and 1 repinning due to screw cutout) at a mean of 4.1 months (range, 1.3 to 7.2 months) postoperatively. None of the 4 patients treated with CBGE within two months postoperatively progressed to stage IVC AVN. The two patients treated after four months postoperatively both progressed to stage VC AVN. Radiographically diagnosed AVN occurred in 6 of 33 hips (18%) at a mean of 6.8 months postoperatively (range, 2.1 to 21.1 months). One patient diagnosed with stage IVB AVN at 2.4 months had screw cutout and received CBGE at 2.5 months from initial pinning. The remaining 5 were not offered surgical intervention. Five of the 6 radiographically diagnosed AVN, including the one treated with CBGE, progressed to stage IVC AVN or greater.

None of the 4 patients with unstable SCFE treated with CBGE within 2 months post pinning developed grade IVC AVN, while all patients treated with other procedures after 2 months developed IVC or greater AVN. Early detection and treatment of AVN after SCFE may alter the clinical and radiographic progression.

Unicompartmental knee arthroplasty (UKA) is often considered to be attractive alternate surgical technique to total knee arthroplasty (TKA) and high tibial osteotomy (HTO), in particular young patients. In addition, it is recently reported that preservation of joint line in UKA is crucial factor for positive long-term outcome, especially in revision case for UKA. However, the role of this joint line has neither been invested nor is it consciously bothered during surgical implantation.

Validated finite element (FE) analysis was introduced in this study to investigate the effects of maximum contact stress on polyethylene (PE) insert and maximum compressive stress in opposite compartments for joint line in fixed-type UKA. As suggested by Weber et al., FE model for joint line was developed by means of determination of the angle between the pre-operative joint line and the reference line from lateral cortical is of the femur. Based on the method above, joint lines were modeled in −3, −2, −1, 0, +1, +2, and +3 mm cases and these seven FE models were compared and analyzed (Fig. 1). All implant components were modeled as linear elastic isotropic materials. However, the model was considered to have plastic characteristics of PE insert. FE analysis was performed using high kinematics displacement and rotation inputs, which were based on the kinematics of the natural knee. ISO standards were used for axial load and flexion (Fig. 2).

The FE model was subjected to validation based on cadaveric experimental data available in the literature by Sohn et al. and from previous cadaveric tests conducted by current investigators. The maximum contact stress was found at around 43 % of the gait cycle in 0 mm case. There were no difference between ± 1 and 0 mm cases, but maximum contact stress on PE insert becomes greater in ± 3 mm cases. The maximum compressive stress of the lateral meniscus in 0 mm case occurred at 62 % of the gait cycle. There were no difference in positive joint line cases in maximum compressive stress, however maximum compressive stress of the lateral meniscus becomes greater in - 3 mm cases.

This study emphasized the importance of joint line preservation after implantation of UKA. It would be critical to determine the joint line in UKA surgery in future based on the result showing that there has been no remarkable difference in stress but changed rapidly from the position beyond the joint line. In future study, it would be valuable study to compare between joint lines of fixed- and mobile-type UKA.