George D Chloros1, Nikolaos K Kanakaris1, James S H Vun1, Anthony Howard1, Peter V Giannoudis2,3. 1. Academic Department of Trauma and Orthopaedics, School of Medicine, University of Leeds, Clarendon Wing, Floor D, Great George Street, Leeds General Infirmary, Leeds, LS1 3EX, UK. 2. Academic Department of Trauma and Orthopaedics, School of Medicine, University of Leeds, Clarendon Wing, Floor D, Great George Street, Leeds General Infirmary, Leeds, LS1 3EX, UK. pgiannoudi@aol.com. 3. NIHR Leeds Biomedical Research Center, Chapel Allerton Hospital, Leeds, UK. pgiannoudi@aol.com.
Abstract
PURPOSE: To evaluate the available tibial fracture non-union prediction scores and to analyse their strengths, weaknesses, and limitations. METHODS: The first part consisted of a systematic method of locating the currently available clinico-radiological non-union prediction scores. The second part of the investigation consisted of comparing the validity of the non-union prediction scores in 15 patients with tibial shaft fractures randomly selected from a Level I trauma centre prospectively collected database who were treated with intramedullary nailing. RESULTS: Four scoring systems identified: The Leeds-Genoa Non-Union Index (LEG-NUI), the Non-Union Determination Score (NURD), the FRACTING score, and the Tibial Fracture Healing Score (TFHS). Patients demographics: Non-union group: five male patients, mean age 36.4 years (18-50); Union group: ten patients (8 males) with mean age 39.8 years (20-66). The following score thresholds were used to calculate positive and negative predictive values for non-union: FRACTING score ≥ 7 at the immediate post-operative period, LEG-NUI score ≥ 5 within 12 weeks, NURD score ≥ 9 at the immediate post-operative period, and TFHS < 3 at 12 weeks. For the FRACTING, LEG-NUI and NURD scores, the positive predictive values for the development of non-union were 80, 100, 40% respectively, whereas the negative predictive values were 60, 90 and 90%. The TFHS could not be retrospectively calculated for robust accuracy. CONCLUSION: The LEG-NUI had the best combination of positive and negative predictive values for early identification of non-union. Based on this study, all currently available scores have inherent strengths and limitations. Several recommendations to improve future score designs are outlined herein to better tackle this devastating, and yet, unsolved problem.
PURPOSE: To evaluate the available tibial fracture non-union prediction scores and to analyse their strengths, weaknesses, and limitations. METHODS: The first part consisted of a systematic method of locating the currently available clinico-radiological non-union prediction scores. The second part of the investigation consisted of comparing the validity of the non-union prediction scores in 15 patients with tibial shaft fractures randomly selected from a Level I trauma centre prospectively collected database who were treated with intramedullary nailing. RESULTS: Four scoring systems identified: The Leeds-Genoa Non-Union Index (LEG-NUI), the Non-Union Determination Score (NURD), the FRACTING score, and the Tibial Fracture Healing Score (TFHS). Patients demographics: Non-union group: five male patients, mean age 36.4 years (18-50); Union group: ten patients (8 males) with mean age 39.8 years (20-66). The following score thresholds were used to calculate positive and negative predictive values for non-union: FRACTING score ≥ 7 at the immediate post-operative period, LEG-NUI score ≥ 5 within 12 weeks, NURD score ≥ 9 at the immediate post-operative period, and TFHS < 3 at 12 weeks. For the FRACTING, LEG-NUI and NURD scores, the positive predictive values for the development of non-union were 80, 100, 40% respectively, whereas the negative predictive values were 60, 90 and 90%. The TFHS could not be retrospectively calculated for robust accuracy. CONCLUSION: The LEG-NUI had the best combination of positive and negative predictive values for early identification of non-union. Based on this study, all currently available scores have inherent strengths and limitations. Several recommendations to improve future score designs are outlined herein to better tackle this devastating, and yet, unsolved problem.
Authors: Kevin O'Halloran; Max Coale; Timothy Costales; Timothy Zerhusen; Renan C Castillo; Jason W Nascone; Robert V O'Toole Journal: Clin Orthop Relat Res Date: 2016-06 Impact factor: 4.176
Authors: Leo Massari; Francesco Benazzo; Francesco Falez; Ruggero Cadossi; Dario Perugia; Luca Pietrogrande; Domenico Costantino Aloj; Antonio Capone; Michele D'Arienzo; Matteo Cadossi; Vincenzo Lorusso; Gaetano Caruso; Matteo Ghiara; Luigi Ciolli; Filippo La Cava; Marco Guidi; Filippo Castoldi; Giuseppe Marongiu; Alessandra La Gattuta; Dario Dell'Omo; Michelangelo Scaglione; Sandro Giannini; Mattia Fortina; Alberto Riva; Pier Luigi De Palma; Antonio Pompilio Gigante; Biagio Moretti; Giuseppe Solarino; Francesco Lijoi; Giovanni Giordano; Pier Giorgio Londini; Danilo Castellano; Giuseppe Sessa; Luciano Costarella; Antonio Barile; Mariano Borrelli; Attilio Rota; Raffaele Fontana; Alberto Momoli; Andrea Micaglio; Guido Bassi; Rossano Stefano Cornacchia; Claudio Castelli; Michele Giudici; Mauro Monesi; Luigi Branca Vergano; Pietro Maniscalco; M'Putu Bulabula; Vincenzo Zottola; Auro Caraffa; Pierluigi Antinolfi; Fabio Catani; Claudio Severino; Enrico Castaman; Carmelo Scialabba; Venceslao Tovaglia; Pietro Corsi; Paolo Friemel; Marco Ranellucci; Vincenzo Caiaffa; Giovanni Maraglino; Roberto Rossi; Antonio Pastrone; Patrizio Caldora; Claudio Cusumano; Pier Bruno Squarzina; Ugo Baschieri; Ettore Demattè; Stefano Gherardi; Carlo De Roberto; Alberto Belluati; Antonio Giannini; Ciro Villani; Pietro Persiani; Silvio Demitri; Bruno Di Maggio; Guglielmo Abate; Francesca De Terlizzi; Stefania Setti Journal: Biomed Res Int Date: 2018-04-30 Impact factor: 3.411