PurposeThis analysis compared the rate of deep wound infections in patients with open tibia fractures, treated with intramedullary nails, receiving additional locally-delivered antibiotics to those receiving standard care.MethodsTwo systematic literature searches identified studies reporting infection rates in patients treated with intramedullary nails for tibia fractures receiving systemic antibiotics only (search one) and in patients receiving adjunctive locally-administered antibiotics peri-operatively at the tissue-implant interface (search two). After applying inclusion and exclusion criteria, 14 and seven papers from searches one and two, respectively, were included in meta-analyses.ResultsThe absolute rate of infection was lower for all Gustilo-Anderson grades of tibia fractures when local antibiotics were administered as adjunctive prophylactic therapy. For severe fractures, classified as GAIII fractures, patients receiving systemic antibiotics only had an infection rate of 14.4 % [95 % CI: 10.5 %, 18.5 %]; adding local antibiotics reduced the rate to 2.4 % [0.0 %, 9.4 %], with an odds ratio of 0.17. Risk of deep wound infections increased with severity of fracture, rising to over 31 % in GIIIB&C fractures for patients receiving systematic antibiotics only, but to below 9 % with additional local antibiotics.ConclusionThe findings support consideration of augmenting the antibiotic prophylaxis regimen to include locally-delivered antibiotics. Patients with severe fractures will obtain greatest benefit from infections avoided. No trial directly compared the two treatments for open tibia fractures, limiting the ability to attribute the differences in observed infection rates directly to the treatments themselves. A large comparative study to improve the evidence on relative effect size is merited.Level of evidence: Level III.

Scoliosis—structural lateral curvature of the spine—affects around four children per 1,000. The MAGEC system comprises a magnetically distractible spinal rod implant and an external remote controller, which lengthens the rod; this system avoids repeated surgical lengthening. Rod implants brace the spine internally and are lengthened as the child grows, preventing worsening of scoliosis and delaying the need for spinal fusion. The Medical Technologies Advisory Committee at the National Institute for Health and Care Excellence (NICE) selected the MAGEC system for evaluation in a NICE medical technologies guidance. Six studies were identified by the sponsor (Ellipse Technologies Inc.) as being relevant to the decision problem. Meta-analysis was used to compare the clinical evidence results with those of one conventional growth rod study, and equal efficacy of the two devices was concluded. The key weakness was selection of a single comparator study. The External Assessment Centre (EAC) identified 16 conventional growth rod studies and undertook meta-analyses of relevant outcomes. Its critique highlighted limitations around study heterogeneity and variations in baseline characteristics and follow-up duration, precluding the ability to draw firm conclusions. The sponsor constructed a de novo costing model showing that MAGEC rods generated cost savings of £9,946 per patient after 6 years, compared with conventional rods. The EAC critiqued and updated the model structure and inputs, calculating robust cost savings of £12,077 per patient with MAGEC rods compared with conventional rods over 6 years. The year of valuation was 2012. NICE issued a positive recommendation as supported by the evidence (Medical Technologies Guidance 18).

This study was funded by Pfizer, which determined the research topic and paid York Health Economics Consortium to develop the analysis and conduct the research. York Health Economics Consortium has received consultancy fees from Pfizer. Gerber, Wallenstein, Mendelsohn, Bourret, Singh, and Moynagh are employees and shareholders of Pfizer. Editorial support was funded by Pfizer and was provided by Claxton, Jenks, and Taylor, who are employees of York Health Economics Consortium. Study concept and design were contributed primarily by Taylor, Jenks, Gerber, and Singh, along with the other authors. Gerber, Moynagh, and Singh collected the data, assisted by Bouret and Mendelsohn; data interpretation was performed by Claxton, Gerber, Bouret, and Mendelsohn. The manuscript was written primarily by Claxton, with assistance from the other authors, and revised by Claxton, Gerber, Bouret, and Mendelsohn, with assistance from the other authors.