Session 11: Trauma 2

Fredag den 25. oktober
09:00 – 10:30
lokale: Helsinki/Oslo
Chairmen: Michael Brix / Juozas Petruskevicius

105. Level of supervision in fracture-related surgery in Denmark. Experience from centres participating in the DFDB (Danish Fracture Database) collaboration.
Morten Jon Andersen, Kirill Gromov, Michael Brix, Anders Troelsen, DFDB Collaborators
Orthopaedics, Hvidovre Hospital

Background: The surgeon's level of experience and degree of supervision is an important factor in outcome following fracture-related surgery. No studies exist describing levels of supervision for fracture-related surgery in Denmark. The Danish Fracture Database (DFDB) was introduced as a quality-monitoring tool of fracture-related surgery and can be used to monitor the surgeon's experience level and degree of supervision at the participating centres.
Purpose / Aim of Study: To describe the levels of experience of the operating surgeons and the level of supervision for fracture-related surgeries performed at 8 participating centres.
Materials and Methods: Currently 9.765 surgical procedures are registered in DFDB. Data regarding experience level of the surgeon and the level of supervision was assessed for primary surgeries (n=7.958) and reoperations (n=576) separately. We describe the level of surgical expertise and amount of supervision for the most common fracture-related surgeries and the most frequently re- operated fracture types. We also investigate the changes in supervision outside regular working hours.
Findings / Results: Interns (IN) performed 18% and junior registrars (JR) 29% of all primary fracture- related procedures. 9,6% of primary procedures performed by IN and 32% by JR, were unsupervised. IN and JR combined operated 16% of all proximal femoral fractures (the most frequent fracture group reported to DFDB) unsupervised. Unsupervised surgeries performed by JR increased from 29,7% to 39,6% outside regular work hours (p<0.001).
Conclusions: While overall levels of supervision were generally high, we found that 1/3 of procedures performed by JR were unsupervised. Lower degree of supervision for surgeries performed outside regular work hours could be a matter of concern.

106. Routine blood tests indicate increased mortality risk in lower limb amputation patients
Steen Vigh Buch, Nikolaj Sode, Troels Riis, Søren Kring, Annette Sylvest, Benn Rønnow Duus
Orthopedic, Bispebjerg

Background: Non-traumatic lower limb amputation is associated with high mortality rates. Recent studies show mortality rates after 30 days of 50% for through knee amputation (TKA) and 31-36% for above knee amputation (AKA).
Purpose / Aim of Study: The aim of this study is to review the outcome of patients undergoing a primary TKA or AKA and identify factors predicting increased mortality rates.
Materials and Methods: All patients who underwent a primary TKA or AKA at Bispebjerg Hospital, Copenhagen in the period from February 2009 to February 2013 where identified using the hospital surgery database (Orbit). Patient records were reviewed retrospectively and additional data was obtained from the hospital biochemistry department and the national civil register. 154 patients (82 male, 72 female) with 69 TKA and 85 AKA were identified. Mean age for patients was 74.0 years. P<0.05 was considered significant.
Findings / Results: The mortality rates after 30 days (D30) were 14.5% for TKA and 23.5% for AKA, and after 90 days (D90) 36.2% for TKA and 40.0% for AKA. Comparison of mortality rates revealed that male versus female patients had a higher risk of D30 (OR 2.41, p<0.04) with no significant difference for D90 (OR 1.67, p<0.1). No further significant differences were identified. Regarding blood samples, serum creatinine (>100 ìmol/L), leucocytosis (>15.1 x 10-9/L) and CRP (>50 mg/L) were associated with significantly higher D90 (OR 3.27, p<0.001, OR 2.69, p<0.009 and OR 3.37, p<0.006, respectively).
Conclusions: In this study, mortality rates are in line with comparable studies. Male patients tend to have higher mortality rates postoperatively, though data is not conclusive. Leucocytosis, elevated levels of both serum creatinine and CRP correlate to increased risk of death postoperatively, and could help identify the group of patients in need of special attention.

107. High patient volume is associated with increased 30-day mortality after hip fracture.
Pia Kjær Kristensen, Theis Muncholm Thillemann, Søren Paaske Johnsen
Orthopaedic Surgery, Region Hospital Horsens; Ortopaedic Surgery, Region Hospital Horsens; Clinical Epidemiology, Aarhus University Hospital

Background: Hip fractures are associated with increased mortality. Arthroplasty procedures have demonstrated better clinical outcomes at high volume units, but the results after hip fracture are inconclusive.
Purpose / Aim of Study: We aimed to evaluate the association between patient volume in hip fracture units and 30-day mortality.
Materials and Methods: Using prospectively collected data from the Danish Hip Fracture Registry, we identified 12,065 patients ≥ 65 years that were admitted with a hip fracture from 2010 to 2011. Patient volume was divided in three groups (≤170 hip fracture admissions per year, 171 to 350 and ≥351 admissions per year). The primary outcome was 30-day mortality. Secondary outcome included quality of care assessed using six process indicators. Data was analyzed using regression techniques while controlling for potential confounders.
Findings / Results: The 30-day mortality was 10.5%, 11.0% and 13.2% for low, medium and high volume units, respectively. Admittance to high volume units was associated with higher 30-day mortality (adjusted odds ratio (OR) = 1.26, 95%CI: 1.01-1.58). Furthermore, patients who were admitted to high volume hip fracture units had lower odds for being mobilized within 24 hours postoperatively (OR=0.71, 95%CI: 0.61-0.82), for basic mobility assessment (OR=0.60, 95%CI: 0.50-0.73), and for receiving a post discharge rehabilitation program (OR=0.48, 95%CI: 0.38-0.60). After adjusting for different quality of care, mortality was comparable (OR=1.14, 95%CI: 0.81-1.60).
Conclusions: Patients with hip fractures admitted to high volume units have higher mortality rates and receive lower quality of care. Variations in quality of care could apparently explain variations in 30-day mortality between units with low and high patient volume.

108. External versus internal fixation of intra-articular distal tibial fractures - A systematic critical review
Peter Ivan Andersen, Bjarke Løvbjerg Viberg, Morten Schultz Larsen
Ortopædkirurgisk afdeling, Kolding Sygehus, SLB

Background: Intra-articular fractures of the distal tibia are among the most challenging of orthopaedic problems. The management of these fractures requires both an understanding of the delicacy of the soft tissue on the distal 1/3 of the tibia, comprehension of the current concepts of treatment and the expertise to apply this knowledge into the treatment of these fractures
Purpose / Aim of Study: The aim of this review was to evaluate literature comparing external fixation (EF) to open reduction and internal fixation (ORIF) of intra-articular distal tibial fractures with focus on complications and functional outcome
Materials and Methods: A search string was designed to search Pubmed, Embase and Cochrane Databases for the literature and revealed 13,096 articles (1993 dublets). 2 reviewers independently assessed the literature for relevance by title, abstracts and full text. Initially only level 2 evidence and above was accepted which gave 3 articles, and therefore level 3 evidence was included and gave 2 more articles. Extraction of data were done by 2 reviewers and sorted regarding to study aims. The quality of studies was assessed by both reviewers using CASP 2010 checklists
Findings / Results: The 5 articles covered 258 intra-articular distal tibia fractures, 118 managed by EF and 140 treated by ORIF. The patients treated with EF are more often subject to non-union, mal-union, and deep infection than patients with ORIF. Very few studies include good functional scores but there might be a better functional outcome after 6 months which even outs after 1 year
Conclusions: The literature at hand is still insufficient to make any definitive conclusions. There is not yet an agreement of which clinical scores to use in follow up, and the low level of evidence in study design makes confounding bias a great risk

109. Over- og undertriage ved modtagelse af multitraumatiserede patienter - En sammenligning af to triagesystemer
Torben Stryhn, Morten Schultz Larsen
Ortopædkirurgisk afdeling, OUH

Background: Correct trauma team activation (TTA) is important. Undertriage may affect mortality, but overtriage is resource consuming and may affect awareness of the Trama Team
Purpose / Aim of Study: This study was done to evaluate the difference between two triage systems used by Odense University Hospital in the period 1/6-31/12; 2010 and 2011. Especially the ability to correctly identify the multitraume patient defined as ISS > 15
Materials and Methods: A score system based on basic observations were used in the first period 1/6 – 31/12 2010. In the second period a more clinically oriented system based on evaluation by trained health proffesionals were used. A retrospective comparative cohordstudie using UAG's trauma registry, and medical record review. For each period two groups was identified. 1) Patients received in the emergency room by TTA. Exclusion: Patients transferred from other hospitals. 2) All other admissions with the trauma codes T or S was reviewed. Deaths within 30 days after emergency room contact, was evaluated separately without any possibility of exclusion. Exclusion criteria: Isolated hip fracture or hospital admissions shorter then 3 days.
Findings / Results: In 2010 and 2011, 614 and 451 patients was received by TTA, 565 and 503 was admitted without. The groups were statistically homogeneous. Overtriage was reduced from 47% to 39%, undertriage from 7% to 6%. The reduction in overtriage was statistically significant. A substantial amount of undertriaged patients had severe head trauma
Conclusions: The introduction of a new triagesystem, with emphasis on the clinical assessment by trained health professionals, has resulted in a reduction of overtriage, without increasing the undertriage. The results are in accordance, with international recommendations and previous Scandinavian studies.

110. Patient volume in hip fracture units is associated with increased length of hospital stay and increased surgical delay.
Pia Kjær Kristensen, Theis Muncholm Thillemann, Søren Paaske Johnsen
Orthopaedic Surgery, Region Hospital Horsens; Ortopaedic Surgery, Region Hospital Horsens; Clinical Epidemiology, Aarhus University Hospital

Background: Hip fractures are associated with the largest use of bed days in hospitals in the Western World. It is unclear whether there are any scale advantages from treating a larger number of patients with hip fractures.
Purpose / Aim of Study: We examined on patient level whether patient volume in hip fracture units is associated with length of hospital stay and surgical delay.
Materials and Methods: In a nationwide prospective population-based cohort study, we identified 12,065 Danish patients aged >65 years with an incident episode of hip fracture admitted between 2010 to 2011. The patient volume was divided in three groups (≤170 hip fracture admissions per year, 171 to 350 and ≥351 admissions per year). The primary outcome was length of hospital stay. Secondary outcome included surgical delay. Data was analyzed using regression techniques while controlling for potential confounders.
Findings / Results: Length of hospital stay was 7.9 days, 8.1 days and 10.7 days for low, medium and high volume units, respectively. Admittance to high volume units was associated with a longer length of hospital stay (adjusted OR 1.29, 95%CI: 1.07-1.55 ). Surgical delay was 20.4 hours, 21.8 hours and 23.0 hours for low, medium and high volume units, respectively. Admittance to high volume hospital was associated with a longer surgical delay (adjusted OR 1.27, 95%CI: 1.02-1.57).
Conclusions: Patients with hip fractures admitted to high volume units have increased surgical delay and increased length of hospital stay compared to low volume units.

111. Demographic and short-term outcome changes within 10 years of a multimodal fast-track hip fracture program
Morten Tange Kristensen, Henrik Palm
Physical Medicine and Rehabilitation Research – Copenhagen (PMR-C), Department of Physiotherapy and , Copenhagen University Hospital Hvidovre; Department of Orthopaedic Surgery, Copenhagen University Hospital Hvidovre

Background: Hip fracture rates seem to have fallen within the last decade but whether patient demographics and short-term outcome also changed during this period is unknown.
Purpose / Aim of Study: To examine changes within a 10 year period in patient demographics, fracture type distribution, and outcome among patients admitted to a specialized hip fracture unit at a university hospital.
Materials and Methods: A consecutive series of 288 patients, median (IQR) age of 83 (77-88) years, admitted within 6 months of 2012 was compared with 288 patients, median age of 81 (73-87) years, admitted at the same hospital from September 2002, and with respectively 82% and 81% admitted from their own home. The pre-fracture functional level was evaluated with the New Mobility Score (NMS, 0-9 points).
Findings / Results: The percentage of men admitted with a hip fracture increased to 35% (101/288) in 2012, from 26 % (76/288) in the 2002 cohort (P=0.02), while no significant difference was seen over time for age, type of fracture, and cognitive status. Patients however had a higher pre-fracture functional level (median NMS, IQR) from 4 (3-9) in 2002 versus 6 (3-9) in 2012 (P=0.01). Time in hospital was unchanged, with a 2012 median LOS of 11 (7-16) versus 11 (6-23) in 2002, and corresponding with 79% and 77% discharged directly to their previous residence. Patients not discharged directly to their own home were more often referred to rehabilitation in 2012 as compared to nursing home in 2002 (P<0.001).
Conclusions: The rate of men sustaining a hip fracture increased significantly with 9% from 2002 to 2012 at our institution, while the pre-fracture function in general was improved. Whether this increase is representative for other parts of Europe, should be further examined. Also, more details of the male cohort should be examined for prophylactic treatment of risk patients.

112. Venous thrombosis following fractures below the knee, a nationwide cohort study
Liv Riisager Wahlsten, Henrik Eckardt, Gunnar Hilmar Gislason, Jonas Bjerring Olesen, Christian Torp-Pedersen
Orthopedic, Rigshospitalet; Cardiology, Gentofte ; Institute of Healt, Science and Technology, Aalborg

Background: Only few studies have investigated the risk of deep venous thrombosis (DVT) or pulmonary embolism (PE) after osteosynthesis of fractures below the knee. Antithrombotic treatment following discharge is currently not recommended in international literature.
Purpose / Aim of Study: The aim of this study was to investigate the incidence of clinical significant DVT/PE in patients undergoing osteosynthesis of fractures below the knee, and to identify specific risk factors associated with the event.
Materials and Methods: Using individual linkage of nationwide registries, we included all patients undergoing osteosynthesis below the knee, 1999-2011. Event rates of DVT/PE were calculated and significant risk factors were identified using cox regression analyses. Patients were followed 180 days from discharge.
Findings / Results: We included 37,853 patients, from these 314 (0.8%) had DVT/PE that demanded hospitalization within 180 days. The event rate was markedly increased the first weeks after discharge, and decreased with time, stabilizing 12-14 weeks after discharge. Oral anticonceptives (Hazard Ratio [HR] 3.58), former DVT/PE (HR 6.XX), and peripheral artery disease (HR 3.1X) were the risk factors associated with the highest incidence of postoperative DVT/PE. Also obesity was associated with an increased risk of DVT/PE.
Conclusions: The overall risk of DVT/PE after osteosynthesis of fractures below the knee was lower than the risk of DVT/PE after hip or knee replacement. However, the risk of DVT/PE in patients with one or more of the risk factors identified above is similar or higher than the risk of DVT/PE after hip and knee replacement, and our study thus suggest that these patients could benefit from the prolonged anticoagulation therapy after discharge.

113. The management of anticoagulant therapy in hip fracture patients in Denmark
Peter Toft Tengberg, Nicolai Bang Foss, Henrik Palm, Anders Troelsen
Ortopædkirurgisk afd. 333, Hvidovre Hospital; Anæstesiologisk afd. , Hvidovre Hospital

Background: There is no consensus in the literature regarding the management of hip fracture patients who are receiving oral anticoagulant therapy on admission. The concept of cessation of treatment to prevent increased blood loss during operation is currently being debated in the literature. Recent studies suggest that this practice results in unnecessary delay of surgery, and increased risk of thromboembolic events.
Purpose / Aim of Study: We investigated current practice in the management of hip fracture patients in oral anticoagulant therapy in Denmark and compared the current practice with the latest findings in the literature.
Materials and Methods: We made a web based survey of the current practice concerning patients in oral anticoagulant therapy in 24 Danish orthopedic departments treating hip fractures. Contact was made by e-mail to the head of the traumatology team. We made a systematic search of the literature on the field in PubMed.
Findings / Results: We found that there was some discrepancy in the management of these patients in Danish orthopedic departments. Some departments do not have a clinical guideline on the subject, but rely on the guidance of other departments in the management of these patients. Some departments delay surgery in order to manage the risk of blood loss. Other departments have a more aggressive approach with no delay in surgery. We found that the literature on this field is very limited and characterized by low level of evidence.
Conclusions: The management of hip fracture patients who are receiving oral anticoagulants on admission shows markedly variance between centers in Denmark. This could in part be explained by the small amount and low level of evidence of studies on this subject. Studies investigating the controversies of the existing, “surgery- postponing” management are warranted.

114. Reliability of a Scoring System for Measurement of Implant Position after Internal Fixation of Undisplaced Femoral Neck Fractures
Marie-Louise Lervad Bartholin, Kolja Weber, Rune Dueholm Bech, Henrik Palm, Bjarke Viberg, Morten Schultz Larsen
Orthopedic Surgery and Traumatology, Odense University Hospital; Orthopaedic Surgery,, Hvidovre University Hospital; Orthopaedic Surgery, Hvidovre University Hospital,

Background: Implant position may be an important predictor of failure after internal fixation (IF) of undisplaced femoral neck fractures (uFNF), but the use of scoring systems for measurement of implant position have been somewhat unreliable in previous studies.
Purpose / Aim of Study: The aim of this study was to evaluate the reliability of a scoring system for measurement of implant positioning after IF of uFNF.
Materials and Methods: 102 patients admitted with an uFNF treated with IF at one hospital between 01.05.2005 and 02.04.2007 were retrospectively included. Implant position on the first postoperative anterior-posterior and axial radiographs were both assessed visually and objectively measured according to a scoring system including screw tip distance, screw-shaft angle, screw-calcar distance, and screw positioning in the femoral head. Three raters (one medical student and two residents) each made the assessments twice with minimum 14 days interval, blinded for each other’s results. An independent person performed unweighted kappa statistics.
Findings / Results: Visually assessed implant position gave intra-rater kappa results at 0.40-0.75 (rater 1), 0.75-0.98 (rater 2) and 0.69-0.81 (rater 3), with inter-rater kappa results at 0.18-0.80 (combined range). Objectively measured implant position gave intra-rater kappa results at 0.56-0.88 (rater 1), 0.74-0.87 (rater 2) and 0.81-0.94 (rater 3), with inter-rater kappa results at 0.48-0.85 (combined range).
Conclusions: Implant position simply assessed visually confirmed to be unreliable but this appeared improvable by objectively measurement. Thus, measurements should be studied for relevance in future formal scoring systems for predicting failure after surgery.

115. Reoperation rates on proximal femoral fractures
Pernille Nygaard Vedel, Troels Riis, Annette Sylvest, Henrik Løvendahl Jørgensen, Benn Duus
Department of Orthopedic Surgery, Bispebjerg; Departement of Clinical Biochemistry, Bispebjerg

Background: Reoperation rates on patients with proximal femoral fractures (PFF) are reported to be between 5-20%. This depends on the pattern of fracture. However, other factors including surgical experience and choice of implants have been shown to influence the reoperation rate as well.
Purpose / Aim of Study: To study the association of reoperation rate among PFF at Bispebjerg Hospital with fracture type, surgical experience and type of implant.
Materials and Methods: A retrospective evaluation of 459 patients with operated PFF in a one-year consecutive period (01.09.2011 - 31.08.2012) with 6 months postoperative follow up. Fractures were stratified as Garden I-II, III-IV femoral neck fractures or extracapsular fractures. Surgeons were grouped as residents, senior residents or specialist. Following implants were registered; parallel screws, four-hole sliding hip screw, intramedullary nail and hemiprosthesis.
Findings / Results: Overall 8,7% underwent reoperation within 6 months. 15,1% of Garden I-II, 17,3% of Garden III- IV and 3,0% of the extracapsular fractures were reoperated. Multivariate Cox regression analysis incorporating fracture type, type of surgery and surgical experience, showed no significant influence of variance in surgical experience (p=0.8) or type of fracture (p=1), but significant difference in frequency of reoperation related to type of surgery (p=0.02). Parallel screws had the highest reoperation rate at 18,9%. For parallel screws in Garden III-IV fractures the reoperation rate was 26,9%, for hemiprosthesis it was 15,1%.
Conclusions: The reoperation rate for femoral neck fractures is higher than for extracapsular fractures. The Garden III – IV treated with parallel screws had the highest reoperation rate (26,9%) compared to 15.1% for those treated with hemiprosthesis and 15.1% for the Garden I – II treated with parallel screws.