Session 15: Trauma

20. November
09:00 - 10:30
Lokale: 01+02
Chairmen: Michael Brix & Peter Tengberg

112. Completeness and validity of the Danish Fracture Database
Anders Bo Rønnegaard, Per Hviid Gundtoft, Peter Toft Tengberg, Bjarke Viberg
Department of Orthopedic Surgery and Traumatology, Kolding Hospital – part of Hospital Lillebaelt; Department of Orthopedic Surgery, Hvidovre Hospital; Department of Orthopedic Surgery and Traumatology, Odense University Hospital

Background: The Danish Fracture Database (DFDB) has delivered data for numerous studies and currently there are more than 75,000 fracture- related surgical procedures registered. However, the validity of the data is only known for two hospitals.
Aim: To determine completeness and validity of the Danish Fracture Database.
Materials and Methods: Patients registered with a primary or planned secondary procedure in the DFDB in 2016 were included. For validity, patient medical records were assessed using a sample of 10% of the study population. The sample was randomly selected and stratified according to anatomical region and hospital. We calculated validity as positive predictive value (PPV) for both non-dichotomous and dichotomous variables. Key variables were date of surgery, operated side, and type of surgery. For type of surgery, validity was defined as the presence of the correct type of surgery in DFDB, regardless of other types registered. Observations unavailable in patient medical records or containing missing values were excluded on an individual variable level. For completeness all primary or planned- secondary procedures in the DFDB in 2016 were compared with cases in the Danish National Patient Registry (DNPR) in 2016. Cases were matched on CPR number, date for surgery and hospital. Completeness was calculated as sensitivity of the DFDB overall and stratified by hospital volume.
Results: For validity, the sample population consisted of 1,541 patients. The mean age was 61 years (2-102 years) and 45.2% were male. PPV for key variables was 97% for date of surgery, 98% for operated side, and 99% for type of surgery. PPV ranged from 81% to 100% with ASA score and trauma status as lowest at 81% and 86% respectively. Pathological fracture (n=8) and location of periprosthetic fracture (n=19) were both at 100%. For variables other than ASA score and trauma status, PPV was more than 94%. For completeness, the sample population consisted of 16,225 cases in the DFDB and 23,397 cases in the DNPR. Sensitivity was 55% overall, ranging from 54% to 60% for large and small volume hospitals.
Interpretation / Conclusion: This indicates that the DFDB in general has high validity, however, it also suggests that the DFDB is lacking in completeness.

113. Patient-related risk factors for nonunion following surgically managed, traumatic, diaphyseal fractures: A systematic review and meta-analysis
Signe Steenstrup, Niels Martin Jensen, Per Gundtoft, Søren Kold, Robert Zura, Bjarke Viberg
Dept. of Orthopedic Surgery and Traumatology, Kolding Hospital, Denmark; Dept. of Orthopedic Surgery and Traumatology, Kolding Hospital, Denmark; Dept. of Orthopedic Surgery and Traumatology, Aarhus University Hospital, Denmark; Dept. of Orthopedic Surgery, Aalborg University Hospital, Denmark; Dept. of Orthopaedic Surgery, Louisiana State University Medical Center, New Orleans, LA, United States; Dept. of orthopaedic surgery and traumatology, Odense University hospital, Denmark

Background: Nonunion is a severe complication in the treatment of fractures and evidence on patient-related risk factors for nonunion following surgically managed fractures is limited. The risk of nonunion is possibly associated with several biomechanical and epidemiological factors.
Aim: To systematically review and assess patient-related risk factors for nonunion following surgically managed, traumatic, diaphyseal fractures.
Materials and Methods: We searched Medline, Embase, Scopus, and Cochrane using a search string developed with aid from a scientific librarian. No date restrictions were made and languages included: English, French, German, Danish, Norwegian, and Swedish. The studies were screened independently by the two main authors using Covidence. Eligible study data was extracted. We performed random-effects-meta-analyses for risk factors found in five or more studies including smoking, sex, open/closed fracture, Gustilo, AO/OTA classification, infection, diabetes, and obesity. Nonunion and risk factors were assessed as a binary outcome, and a 2x2 contingency table was made for each risk factor. The evidence synthesis was performed using odds ratio (OR) as effect measure.
Results: Of 11.738 records screened, 30 were eligible and included a total of 38,465 patients. The age ranged from 16 to 100 years. One study was prospective, one was uncertain, and the remaining 28 were retrospective. Twenty-five studies were eligible for meta-analyses. Sex and obesity were not a significant risk factors for nonunion, however, smokers had a 70% increased likelihood of nonunion, 95% CI 1.2 to 2.4. Open fractures, diabetes, infection, higher Gustilo, and increasing AO were also significant risk factors for nonunion (p<0.05).
Interpretation / Conclusion: Establishing compelling evidence on risk factors for nonunion is challenging due to the heterogeneity and complexity of nonunion, and because the current studies are predominantly small and retrospective. However, with this study, we can conclude that smoking, open fractures, diabetes, infection, increasing Gustilo and AO are significant risk factors for nonunion following surgically managed, traumatic, diaphyseal fractures.

114. Complications of Fitbone and Precice intramedullary bone lengthening nails: a systematic review with 782 patients
Markus Frost, Ole Rahbek, Jens Trærup, Adriano Ceccotti, Søren Kold
Department of Orthopaedics, Aalborg University Hospital; Interdisciplinary Orthopaedics, Aalborg University Hospital

Background: Intramedullary lengthening nails were introduced to reduce complications in limb lengthening. More than 16000 intramedullary Precice and Fitbone lengthening nails have been implanted worldwide. Complications are so far only heterogeneously reported in small case series, and no systematic overview of complications exists.
Aim: To perform a systematic literature review of complications with Fitbone and Precice bone lengthening nails in lower extremity bone lengthening.
Materials and Methods: In PubMed, EMBASE, Cochrane Library a systematic search, with no limits concerning study design, date or language, was performed with search string of medical subject headings: Bone Nails, Bone Lengthening and “Word” Fitbone and Precice nail. One author selected the articles. The first and last author assessed complications. Complications were severity graded (Black et al. 2015) and categorized into subgroups based on origin.
Results: The search found 952 articles, 116 were full text screened and 41 included. 983 segments were lengthened in 782 patients (age: 8-74 years). Number of patients: 208 congenital, 305 acquired limb shortening, 111 short status, 158 unidentified etiology. We identified 332 complications (34% of segments). Type I (minimal intervention): 11% of segments; Type II (substantial change in treatment plan): 15% of segments; Type IIIA (failure to achieve goal): 5% of segments; Type IIIB (new pathology or permanent sequelae): 3% of segments. Joint contracture/subluxation/luxation was the most frequent type IIIB complication. The two most frequent origins of complications were related to device (13% of segments) and bone (9% of segments).
Interpretation / Conclusion: In this first systematic review of complications with intramedullary bone lengthening nails, the overall risk of complication was high with one complication for every three segments lengthened. In one of every four segments, the complication had a major impact on treatment: substantial change in treatment, such as unplanned surgery (15%), failure of achieving lengthening goal (5%) or introduction of a new pathology or permanent sequelae (3%).

115. Radiographs of 366 removed limb lengthening nails reveal differences in bone abnormalities between different nail types: FITBONE, PRECICE and STRYDE.
Markus Winther Frost 1, Christopher Iobst 2, Jan Duedal Rölfing 3, Ole Rahbek 1, Anirejuoritse Bafor 2, Molly Duncan 2, Søren Kold 1
1. Department of Orthopedic Surgery Interdisciplinary Orthopaedics, Aalborg University Hospital Aalborg, Denmark 2. Department of Orthopaedics Center for Limb Lengthening and Reconstruction Nationwide Children’s Hospital Columbus, USA 3. Orthopaedic Reconstruction and Children’s Orthopaedics Aarhus University Hospital Aarhus, Denmark

Background: Limb lengthening nails have largely replaced external fixation in limb lengthening and reconstructive surgery. However, the adverse events and high prevalence of radiographic changes recently noted with the STRYDE lengthening nail have raised concerns about the use of internal lengthening nails.
Aim: The aim was, therefore, to compare the prevalence of radiographic bone abnormalities between STRYDE, PRECICE, and FITBONE nails prior to nail removal
Materials and Methods: This study was performed as a retrospective case series from three centers. Patients were included if they had either of the three limb lengthening nails (FITBONE, PRECICE, or STRYDE) removed. Standard orthogonal radiographs immediately prior to nail removal were examined for bone abnormalities at the junction of the telescoping nail parts.
Results: In total, 306 patients (168 males, 138 females) had 366 limb lengthening nails removed. The mean (SD, min-max) from nail insertion to radiographic evaluation was 434 days (SD 381, 36 – 3015 days). 77% (20/26) STRYDE nails had bone abnormalities at the interface compared with only 2% (4/242) FITBONE and 1% (1/98) PRECICE nails (P<0.0001). In addition, the extent of bone abnormalities was more pronounced in the STRYDE nails compared with the other nails.
Interpretation / Conclusion: Bone abnormalities at the interface of telescoping nail parts were seen in the majority of STRYDE nails, but only very rarely with FITBONE or PRECICE nails. Of clinical importance, the low prevalence of radiographic changes at the junctional interface of 242 evaluated FITBONE and 98 evaluated PRECICE nails at the time of nail removal does not warrant clinical concerns.

116. The quality of tension band wiring for olecranon fractures is related to complications. A multicentre cohort study
Kia Cirkeline Møller Hansen, Mustafa Mahamoud, Stefan Jensen, Bjarke Viberg
Department of Orthopaedic Surgery and Traumatology, Hospital Lillebaelt, University Hospital of Southern Denmark; Department of Orthopaedic Surgery and Traumatology, Hospital of South West Jutland, Denmark; Department of Orthopaedic Surgery and Traumatology, Odense University Hospital, Denmark

Background: Tension band wiring (TBW) is the most frequently used fixation for displaced olecranon fractures. TBW is generally known as a simple method that can be performed by most orthopaedic surgeons and has satisfying results.
Aim: The aim of this study was to determine if the quality of the TBW osteosynthesis for displaced olecranon fractures in adult patients was associated with complications.
Materials and Methods: This is a multicentre retrospective cohort study with data on adult patients with an operatively treated displaced olecranon fracture. From 2013 to 2018, eligible patients were retrieved using diagnosis codes for elbow and olecranon fractures from the hospitals administration databases in the Region of Southern Denmark. Patients’ health care files were reviewed for operation details and complications. Major complication was defined as any reoperation within 8 weeks or deep infection. Any loss of fixation was added to define surgical complications. Pre- operative x-rays were classified using the Mayo classification and early postoperative x-rays were assessed for quality of the TBW osteosynthesis based on the 10 imperfection from Schneider et al. yielding 10 points if no imperfections were present. STATA was used for descriptive statistics with median and inter quartile range. Chi square test was used for group comparison.
Results: 307 patients were included, 68% females, median age was 64 (IQR=24), and 20% were ASA=3. There were 76% Mayo type 2A and 20% type 2B. A postoperative cast applied for 96% of the patients for 2 (IQR=1) weeks. The TBW’s were rated with a median score of 8 points (IQR=2). There were 12% with major complications. No complications were seen if given 10 points, 6% if given 9 points, and 11-17% if given between 5-8 points, thereby resulting in a clear relation between a higher score and less complications (p<0.0000). There were 30% with surgical complications demonstrating similar relation between higher points and less complications (p<0.001).
Interpretation / Conclusion: There were 12% with major complications and we found a clear relation between the quality of osteosynthesis and complications. Surgeons need to be very thorough when performing TBW in order to reduce the postoperative complications.

117. Injuries among Motorcycle riders – did imposed legislation for young riders modify the age and severity composition of patients seen at the A&E department in Odense University Hospital from 2003-2020?
Henriette Nelsson, Niels Dieter Röck , Jens Lauritsen
Henriette Nelsson, Odense University Hospital, Dept. Ortopedics, Accident Analysis Group Niels Dieter Röck, Odense University Hospital, Dept. Ortopedics, Accident Analysis Group Jens Lauritsen, Odense University Hospital, Dept. Ortopedics, Accident Analysis Group & Dept. of Clinical Medicine, University of Southern Denmark, Odense, Denmark

Background: In the 2010s stricter age limits for motorcycles were imposed intended to reduce number of young victims,
Aim: To investigate changes in severity, age composition and epidemiology of injured Motorcycle accident victims during the period 2003–2020 in relation to imposed age restrictions on size of Motorcycles in 2013.
Materials and Methods: All acute patients treated at the Accident and Emergency Dept (A&E) at Odense University Hospital in the period 2003-2020 were eligible for study if driving a two wheeled motorized vehicle (n=6043). Following scrutiny of descriptive texts patients were excluded if contained in at least one of: Passengers (n=236), Use of Scooter 45 (n=2158), Moped (n=2728), Sports and other (n=42), Age < 18 (n=1291). The remaining 1099 patients were verified as MC riders and included in the analysis. Data on drivers killed at the injury scene were obtained from Statistics Denmark (Routine Road Injury Statistics). Statistical analysis was preformed using EpiData Analysis (version 3.0.0.1).
Results: Among the 1099 injured treated in hospital 42 % (95% CI 39 - 45) suffered severe injury. Of the injured included 38 % (95% CI 35 – 41) were injured after 2013, 62 % (95% CI 59 – 65) before - a significant reduction of 24 %. The age profile of injured showed an increase in the 50+ year-olds after 2013, OR 2.06 (95% CI 1.58-2,68). No significant change vas found for 20+ or 25+ year-olds (OR 1.10 (95% CI 0.78-1,55), 1.05 (95% CI 0.73, 1.52) respectively). Severity were significantly associated with age, injuries becoming more severe, the older the victim (p<0.0001). Over the period we saw no difference in overall severity (p=0.618). 12 % (95% CI 10 – 14) of the injured were women. Women were less severely injured (p=0.017), Men and women did not differ significantly in age profile or ratio of injured over the period. 31 were killed at the scene, 10 % were women.
Interpretation / Conclusion: The legislation changes of 2013, imposing stricter age-limits for riding the big motorcycles, did not significantly modify the severity composition or ratio of injured in the affected age groups. The women and the young riders of both sexes were less severely injured. Men 45-64 years old were most frequent in the death- statistics.

118. Risk of Reoperation in Simple Ankle Fracture Surgery when Comparing Locking Plate with Non-Locking Plate
Gudrun Holm Jacobsen, Mads Holm Gude, Bjarke Løvbjerg Viberg, Per Hviid Gundtoft
Gudrun Holm Jacobsen: Department of Orthopaedic Surgery and Traumatology, Lillebaelt Hospital, University Hospital of Southern Denmark, Sygehusvej 24, 6000 Kolding, Denmark Mads Holm Gude: Department of Orthopaedic Surgery and Traumatology, Lillebaelt Hospital, University Hospital of Southern Denmark, Sygehusvej 24, 6000 Kolding, Denmark Bjarke Løvbjerg Viberg: Department of Orthopaedic Surgery and Traumatology, Lillebaelt Hospital, University Hospital of Southern Denmark, Sygehusvej 24, 6000 Kolding, Denmark Per Hviid Gundtoft: Department of Orthopaedic Surgery and Traumatology, Lillebaelt Hospital, University Hospital of Southern Denmark, Sygehusvej 24, 6000 Kolding, Denmark

Background: Locking plates were initially designed to provide improved stability to ankle fractures with poor bone quality but are currently widely used. The literature is sparse regarding the benefit of locking plate in ankle fractures.
Aim: The aim is to compare the risk of reoperation for locking plate compared with non-locking plate in patients with ankle fractures. Secondary, to investigate the distribution of locking plate use in Denmark.
Materials and Methods: The study was a population-based register study. Data regarding patients with AO type 44A1/2 and 44B1/2 who were treated with either locking or non- locking plates were obtained from the Danish Fracture Database from March 15, 2012 to December 31, 2016. The follow-up period was 24 months. Major complications were defined as complications requiring surgical intervention, with the exception of simple hardware removal 6 weeks after primary surgery, which was defined as a minor complication. Multivariate regression analysis was performed to determine relative risk (RR), adjusted for age, sex, American Society of Anesthesiologists Classification (ASA) score and the level of the surgeon’s experience. All results are reported with 95% confidence interval.
Results: A total of 2,177 ankle fractures were included, among which 718 (33%) were treated with locking plates, and 1,459 (67%) were treated with non- locking plates. The mean age was higher in the locking plate group (p<0.001) and locking plate was used more often in women (p=0.018), in patients with higher ASA-score (p<0.001), and in patients operated by consultants (p=0.018). In both groups, the risks for major and minor complications were 3% and 22%, respectively, resulting in adjusted RRs of 1.00 (95% CI 0.66;1.66) for major reoperation comparing locking with non-locking plates and 0.92 (95% CI 0.76;1.11) for minor reoperations. The proportion of locking plate use varied widely between departments, ranging from 6% to 61%.
Interpretation / Conclusion: We conclude that there is no significant difference in association to reoperation when comparing locking plates with non-locking plates among patients with surgically treated ankle fractures. The indication of locking plate use should be evaluated on all hospitals.

119. The Positive Predictive Value of Ankle Fracture Diagnoses and Surgical Procedure Codes in the Danish National Patient Registry
Per Hviid Gundtoft, Frederik Borup Danielssson, Michael Houlind Larsen, Simon Oksbjerre Mortensen, Yasemin Corup, Nicholas Bonde, Michael Brix, Jeppe Barckman, Mads Terndrup, Bjarke Viberg
Per Hviid Gundtoft Department of Orthopaedics, Aarhus University Hospital Department of Orthopaedics, Kolding Hospital Frederik Borup Danielsson Department of Orthopaedics, Kolding Hospital Michael Houlind Larsen Department of Orthopaedics, Kolding Hospital Simon Oksbjerre Mortensen Department of Orthopaedics, Aarhus University Hospital Yasemin Corup Department of Orthopaedics, Odense University Hospital Nicholas Bonde Department of Orthopaedics, Hvidovre Hospital Michael Brix Department of Orthopaedics, Odense University Hospital Jeppe Barckman Department of Orthopaedics, Aarhus University Hospital Mads Terndrup Department of Orthopaedics, Hvidovre Hospital Bjarke Viberg Department of Orthopaedics, Odense University Hospital Department of Orthopaedics, Kolding Hospital

Background: Knowledge about the validity of data is essential for understanding the precision of studies using data from the Danish National Patient Registry (DNPR).
Aim: To validate the quality of ankle fracture registry data by estimating the positive predictive value (PPV) of the diagnosis and procedure codes regarding ankle fractures reported to the DNPR.
Materials and Methods: We identified all patients, with an ankle fracture diagnosis code and/or procedural codes for surgical fixation of ankle fractures in the period 01.01.2018 to 31.12.2018 from the health care database of four hospitals: two university and two regional centers located in Zealand, Funen and two in Jutland. The primary outcome was the PPV of a random sample of 10% from the group of patients with both an ankle fracture diagnosis and an ankle fracture procedure code. Secondarily, patients with only an ankle fracture diagnosis code or only an ankle fracture surgical procedural code were extracted for validation. Data from medical records, as well as radiographs including radiological descriptions were extracted for each study subject. Two consultants independently reviewed radiographs including radiological descriptions in order to validate the ankle fracture diagnosis and procedure codes registered in the DNPR.
Results: From the four centers 651 patients with both an ankle fracture diagnosis and an ankle fracture surgical procedure codes were included in the source population, of which 65 (10%) were extracted for validation. The Positive Predictive Value (PPV) of registry data reported to the DNPR with both ankle fracture diagnosis and procedure code was 93%. Furthermore, 56 patients were identified with either an ankle fracture diagnosis or an ankle fracture surgical procedure code and extracted for validation. In these cases, the PPV was 91% if only an ankle fracture diagnosis was reported and 82%, if only a surgical procedure code was reported.
Interpretation / Conclusion: The study shows that the ankle fracture diagnosis and procedure codes registered in the DNPR are of high quality and therefore a valuable data source for research on ankle fractures.

120. The First COVID Lockdown’s Impact on the Number, Distribution and Severity of Hand Injuries at an Accident and Emergency Department.
Emil Ainsworth Jochumsen, Søren Larsen, Jens Lauritsen
Department of Orthopaedics, Odense University Hospital

Background: During the lockdown of the COVID-19 pandemic in Denmark in March 2020, the trauma Accidents and Emergency (A&E) Department of Odense University Hospital, experienced a marked drop in visits. We had the impression that there was a larger decline in visits for minor injuries compared to severe injuries.
Aim: The aim of this study was: 1) to present how the lockdown during the 1. wave of the COVID-19 pandemic affected the number and distribution of hand injury related visits in the A&E at OUH. 2) to analyze the distribution of injuries during the COVID-19 lockdown, in order to examine patients’ ability of self-assessing the need for a visit to the A&E.
Materials and Methods: We analyzed the Incident Rate Ratio (IRR) between the first month of the lockdown with 1 month up to the lockdown. We compared this IRR with the IRR for the equivalent periods of 2019. The outcomes were: Diagnoses (or diagnosis-groups), activity during injury (ADI), mechanism of injury, age-group and gender. Secondarily, certain specific diagnoses (or diagnosis- groups) were analyzed with regards to ADI and mechanism of injury.
Results: We observed a significant (95% confidence interval) reduction of 30,6% for all hand injuries in 2020. We observed a significant reduction in IRR for minor injuries of 58,7% and a non-significant reduction of 15,3% for severe injuries. The IRR in 2020 was significantly lower in 2020 for ages: 0-14, 15-17 and 25-49 years. The relative reduction in IRR was markedly larger for men (40,8%) than women (22,3%).
Interpretation / Conclusion: We observed a far greater reduction in minor injuries than severe injuries compared to 2019. This result could imply the patients’ ability to assess their injury severity is perhaps better than previously presumed. It is unclear if this result, could be recreated during different circumstances. Further research on this subject is needed.

121. Physical function, Quality of life and complications following a distal intra-articular tibia fracture; External Fixation (EF) or Open Reduction Internal Fixation (ORIF); A prospective cohort/observational study
Julie Erichsen, Frank Damborg, Morten Schultz-Larsen, Carsten Jensen, Bjarke Viberg


Background: Physical function and Quality of life (Qol) following a distal intra-articular tibia fracture (DIATF) has been demonstrated low regardless of operation method but only few studies have reported on this. The evidence is dominated by retrospective studies.
Aim: To perform a prospective cohort study with focus on physical function, Qol and complications evaluating Danish patients with a DIATF operated by either EF or ORIF.
Materials and Methods: From December 2015 to December 2019 patients with a DIATF (AO 43B1-3 and C1-3), and operated with either EF or ORIF, were prospectively included in a multicentre study with 7 departments. Primary outcome was comparison of 12 months physical function evaluated by the Self-reported Foot and Ankle Score (SEFAS) and EQ5D. Secondary were SEFAS and EQ5D at 1.5, 3, and 6 months follow- up as well as reoperation. Group comparison statistics were performed using the chi square for categorical data and the students t-test for continuous data. Descriptive data consists of mean with standard deviation.
Results: There were 59 patients with a DIATF included, 29 treated with EF. The mean age was 46 (1.9), 57% were males with no statistical difference between groups including for AO classification, comorbidities, smoking, and alcohol abuse. Mean SEFAS score 12 months postoperatively was 23 (9.4) in the EF group compared to 29 (11.2) in the ORIF group (p= 0.77). At 1.5 month the SEFAS in the EF group was 15 (7.9), at 3 months 20 (9.2), and 22 (7.3) after 6 months compared to 20 (8.2) (p=0.02), 20 (8.5) (p=0.9), and 27 (10.6) (p=0.01) in the ORIF group. The mean EQ5D index at 12 months was 0.68 (0.12) in the EF group compared to 0.78 (0.12) in the ORIF group (p= 0.22). Similar patterns for the 1.5, 3, and 6 months EQ5D follow- ups were seen as with the SEFAS data. In total, there were 46% reoperations with no difference between groups (p=0.5).
Interpretation / Conclusion: There is no significant difference between the EF and ORIF group in terms of 12 months physical function and Qol but there was statistical difference after 1.5 and 6 months in favour of ORIF. No statistical difference was seen in reoperations.