Session 9: Trauma

Torsdag den 23. oktober
13:00 – 14:30
Lokale: Helsinki/Oslo
Chairmen: Peter Toft Tengberg / Søren Kold

85. Orthogeriatric collaborative improves 30 day mortality and quality of care after hip fracture in the elderly: a nationwide cohort study
Pia Kjær Kristensen, Theis Muncholm Thillemann, Kjeld Søballe, Søren Paaske Johnsen
Department of Orthopaedics , Hospital Horsens; Orthopaedic surgery , University Hospital Aarhus; Orthopaedics , University Hospital Aarhus; Department of Clinical Epidemiology, University Hospital Aarhus

Background: Orthogeriatric collaborative has been linked with better clinical outcomes for hip fracture in clinical trials, however little is known about the impact of orthogeriatrics on quality of care and clinical outcomes among patients with hip fracture in a routine setting.
Purpose / Aim of Study: We examined the association between orthogeriatrics and 30-day mortality, quality of care, surgical delay and length of stay among patients with hip fracture.
Materials and Methods: Using prospectively collected data from the Danish Multidisciplinary Hip Fracture Registry, we identified 12,065 patients ≥65 years who were admitted with a hip fracture between 1. March 2010 to 30. November 2011. The exposure was divided in two groups; Orthopaedic ward with geriatric or medical consultant service on request and orthogeriatric collaborative where the geriatrician is integrated into the orthopaedic team with nurses, physiotherapist and the orthopaedic surgeon. The primary outcome was 30-day mortality. Secondary outcome included quality of care assessed using six process indicators, surgical delay and length of stay. Data was analyzed using regression techniques while controlling for potential confounders.
Findings / Results: The 30-day mortality was 9.4 % and 12.0 % for orthogeriatric and orthogeriatric ward. Admittance to an orthogeriatric ward was associated with decreased 30-day mortality (adjusted odds ratio = 0.69 (95% CI: 0.54-0.88)). Furthermore, patients had higher relative risk for systematic pain assessment, for receiving basic mobility assessment, for a post discharge rehabilitation program, for anti-osteoporotic medication and for prevention future fall accidents. Length of hospital stay and surgical delay were similar in the groups.
Conclusions: Admittance to an orthogeriatric ward was associated with decreased mortality rates and improved quality of care.

86. Tranexamic Acid (TA) Reduces Overall Blood Loss in Unstable Trochanteric Fractures Treated with Intramedullary Nailing (IMN)
Peter Toft Tengberg, Henrik Palm, Nicolai Foss, Thomas Kallemose, Anders Troelsen
CORH, Hvidovre Hospital; Anesthesiologic dept, Hvidovre Hospital

Background: Patients with unstable trochanteric fractures treated with intramedullary nailing (IMN) suffer a major overall blood loss. This is associated with risk of postoperative anemia resulting in increased morbidity, mortality and prolonged hospitalization. Tranexamic Acid (TA) has shown documented effect, with reduced blood loss, in other fields of orthopaedics, such as arthroplasty surgery.
Purpose / Aim of Study: The aim of the study was to test if TA can reduce overall blood loss in patients with unstable trochanteric fractures treated with a short IMN.
Materials and Methods: The study is a 2-arm, double blinded, randomized placebo controlled trial with two groups. The inclusion period was September 2011 to June 2014. Patients with unstable trochanteric fractures scheduled for a short IMN were eligible for inclusion. Patients in the intervention group were given a 1 g bolus of TA during initiation of anesthesia and 3 g of TA in 1 l of saline over 24 hours postoperatively. The placebo group was given a similar regime, without TA. 270 patients were approached for inclusion. Of these, 75 were included, with a final study group of 73 after 2 postoperative inclusions.
Findings / Results: There were 51 (71%) women. Mean age was 77.3 (SD: 12.3). Mean blood loss in the intervention group (33 patients) was 1410,9 ml (1001,4) compared to a mean blood loss of 2100,4 (1152,6) in the placebo group (p=0.008, t-test). There were no cases of in-hospital, thromboembolic events in either group.
Conclusions: We found a statistically significant reduction in the overall blood loss close to 700 mL for patients treated with TA. TA seems to be an effective blood saving strategy for this group of hip fracture patients and should be considered in future treatment regimens. Our data revealed no safety concerns. Further studies should focus on optimal timing and dose of TA.

87. Validation of fracture treatment codes from the Danish National Patient Registry: Implications for The Danish Fracture Database
Morten Jon Andersen, Michael Kuhlman, Michael Brix, Kirill Gromov, Anders Troelsen
Orthopaedic Surgery, Copenhagen University Hospital Hvidovre

Background: More than 99% of all public discharges are recorded in the Danish National Patient Registry (NPR), however data have been proven to have a degree of uncertainty. The Danish Fracture Database (DFDB) was established with the goal of recording all fracture surgery in Denmark. With the growing amount of data in DFDB, a viable way of verifying completeness and validity is needed.
Purpose / Aim of Study: The aim of this study was to validate the NPR treatment codes regarding fracture surgery to allow for continuous monitoring of DFDB data.
Materials and Methods: At our institution the diagnosis and operation codes are reported directly from the operation-booking programme (OBP) to NPR. A list of 500 consecutive patients operated for fractures from January to April 2012 was created from the OBP. Patient charts and x-rays where reviewed by the authors. The treatment codes were considered correct when codes and the x- ray assessment matched exactly. The anatomic region of the diagnosis or operation was considered correct if the fracture or osteosynthesis was present in the region specified by the first four letters in the code.
Findings / Results: In 16 cases x-rays were missing, 484 cases was analysed further. The overall validity of data was 86% (417/484). In 94% (454/484) the NPR diagnosis code was correct and the NPR anatomic region was correct in all but one case. In 91% (440/484) the operation code was correct and the anatomic region for the operation was correct in all but two cases. The correct side (left/right) was given in 99% (477/484) of cases.
Conclusions: Diagnosis and operation codes regarding fracture surgery reported to the NPR showed very high validity. If the same data validity of data reported to NPR can be confirmed for other regions and hospitals in Denmark NPR data can serve as a way of verifying completeness and validity in DFDB.

88. Sensitivity and specificity of CT and MRI imaging in occult hip fractures
Martin Haubro, Camilla Stougaard, Trine Torfing, Søren Overgaard
Department of Orthopaedic Surgery and Traumatology, Odense University Hospital; Department of Radiology, Odense University Hospital

Background: The incidence of occult hip fracture has been estimated to 2-10% using standard X-ray imaging. It has been stated that MRI is superior to CT in detecting occult fracture of the hip however this statement is based on case reports and non-controlled studies of few patients.
Purpose / Aim of Study: To estimate sensitivity and specificity of CT and MRI examinations in patients with fractures of the proximal femur. To determine the interobserver agreement of the modalities among a senior consulting radiologist, a resident in radiology and a resident in orthopaedics surgery.
Materials and Methods: 67 patients seen in the emergency room with hip pain after fall, inability to stand and a primary X-ray without fracture were evaluated with both CT and MRI. The images were analysed by a senior consulting musculoskeletal radiologist, a resident in radiology and a resident in orthopaedic surgery. Sensitivity and specificity were estimated with MRI as the golden standard. Kappa value was used to assess level of agreement in both MRI and CT finding.
Findings / Results: 15 fractures of the proximal femur were found. Two fractures were not identified by CT and four changed fracture location. Among those, three patients underwent surgery. Sensitivity of CT was 0.87; 95% CI [0.60;0.98]. Kappa for interobserver agreement for CT were 0.46; 95% CI [0.23; 0.76] and 0.67; 95% CI [0.42; 0.90]. For MRI 0.67; 95% CI [0.43;0.91] and 0.69; 95% CI [0.45;0.92].
Conclusions: MRI was observed to have a higher diagnostic accuracy than CT in detecting occult fractures of the hip. Interobserver analysis showed high kappa values corresponding substantial agreement in both CT and MRI.

89. Inter-rater reliability and agreement of the 6-minute walk test in females with hip fractures
Jan Arnholtz Overgaard, Camilla Marie Larsen, Morten Tange Kristensen
Department of Rehabilitation and Physical Medicine and Rehabilitation Research - Copenhagen (PMR-C), Municipality of Lolland; Research Unit of Musculoskeletal Function and Physiotherapy, Institute of Sports Science and Clinical Biomechanics, University of Southern Denmark; Physical Medicine and Rehabilitation Research - Copenhagen (PMR-C), Departments of Physiotherapy and, Hvidovre University Hospital

Background: The 6-minute walk test (6MWT) is widely used as a clinical outcome measure. However, the reliability and agreement of the 6MWT is unknown in individuals with hip fractures.
Purpose / Aim of Study: To evaluate the relative inter-rater reliability and agreement of the 6MWT in individuals with hip fractures.
Materials and Methods: Two senior physiotherapy students independently examined (randomized order) a convenient sample of 20 participants; their assessments were separated by two days, and testing followed instructions from the American Thoracic Society. Hip pain was assessed with the Verbal Ranking Scale.
Findings / Results: Participants (all women) with a mean (SD) age of 78.1 ± 5.9 years performed the test within a mean of 31.5 ± 5.8 days post-surgery; 10 had a cervical and 10 a trochanteric fracture. Excellent inter-rater reliability; ICC2.1 = 0.92 (95% CI, 0.81 - 0.97) was found, and the standard error of measurement (SEM) and smallest real difference (SRD) were calculated as 21.4 meters and 59.4 meters, respectively. Bland-Altman plots revealed no significant difference (mean of 3.2 ± 31.5 meters, P = 0.83) between the two raters, and no heteroscedasticity was seen (r = -0.196, P = 0.41). On the contrary, participants walked a mean of 21.7 ± 22.6 meters longer, at the second trial (P = 0.002). Participants with moderate hip fracture- related pain walked a shorter distance than those with no or light pain during the first test (P = 0.04), while this was not the case during the second (P = 0.25).
Conclusions: Excellent inter-rater reliability was found with a low level of measurement error, particularly for a group of participants with hip fractures, as a change of 22 meters can be considered a real change. The importance of measuring hip fracture-related pain during testing seems important when individuals with hip fracture perform the 6MWT.

90. Surgical delay increases early mortality for patients with proximal femoral fractures. A study from The Danish Fracture Database Collaborators.
Anne Marie Nyholm, KIrill Gromov, Henrik Palm, Michael Brix, Thomas Kallemose, Anders Troelsen
Department of Orthopaedics, Hvidovre Hospital

Background: Surgical delay (SD) of proximal femoral fracture (PFF) is of interest as some studies show SD >24 or >48 hours to significantly increase mortality (MT), while others find no correlation.
Purpose / Aim of Study: To show if a) SD or b) surgeon’s experience (SE) increases MT rates for following PFF.
Materials and Methods: We included trochanteric or femoral neck fractures (AO31A and 31B) from the Danish Fracture Database, excluding pathological and high-energy trauma fractures. Data included age, gender, American Society of Anesthesiologists (ASA) score, type of fracture and osteosynthesis, SE and SD. SE was defined as “attending or above” or “below attending”. SD was defined as hours (h) from radiological diagnostics until start of surgery. MT data was from The Civil Registration System. Multiple logistical regression analysis was used to calculate MT rates.
Findings / Results: 3595 fractures were included: mean age 81.0 years, 70% were female and 49% were trochanteric fractures (AO31A). SD was <12h in 21%, <24h in 70%, <48h in 92% and <72h in 95% of cases. SE was “Attending or above” in 49% of all cases. MT was 10.8% at day 30 and 17.4% at day 90. SD >12h increased adjusted risk of 30-day MT (OR 1.43, p=0.02). SD >24h increased adjusted risk of 90-day MT (OR 1.24, p=0.03). SE “below attending” increased MT risk at both day 30 (OR 1.27, p=0.04) and day 90 (OR 1.28, p=0.01). Increasing age and ASA score, male gender and type of osteosynthesis significantly increased both 30-day and 90-day MT.
Conclusions: In this study SD >12 h and SD >24 h significantly increased adjusted risk of MT at day 30 and day 90, respectively. The adjusted risk of both 30-day MT and 90-day mortality increased significantly when SE was “below attending”. The study findings challenge the departments to facilitate fast surgical treatment supported by attending orthopaedic surgeons.

91. Surgical delay and early mortality in patients with distal femoral fractures
Bozo Jian, Thomas Kallemose, Henrik Palm, Michael Brix, Anders Troelsen, Kirill Gromov
Department of Orthopaedic Surgery, Copenhagen University Hospital Hvidovre

Background: While fast surgical treatment of proximal femoral fractures is supported by previous studies and is encouraged by national clinical guideline in several countries, no consensus exists on optimal timing for surgical treatment of the equally frail patient group with distal femoral fractures.
Purpose / Aim of Study: The purpose of this study was to investigate whether increased surgical delay (SD) in patients with distal femoral fractures is associated with risk for early mortality.
Materials and Methods: 189 patients surgically treated for distal femoral fractures in 2008-2013 were identified at two university hospitals. Only patients > 50 years old were included for analysis (n=170). Recorded parameters included age, gender, American Society of Anesthesiologists (ASA) score, AO diagnosis, osteosynthesis type, presence of total knee arthroplasty (TKA) and SD (hours). SD was defined as time from admission to the hospital to beginning of surgery and divided into groups: <12h, <24h, <48h, <96h and >96h. 30 and 90 day mortality was recorded. Multiple logistical regression analysis was performed to calculate adjusted risk for 30-day and 90-day mortality.
Findings / Results: Median age was 78, range (51-99). 80% were female, 96% were treated with plate fixation and 21% had a TKA at the fracture site. 7% were operated within 12 hours, 37% within 24 hours, 62% within 48 hours and 86% within 96 hours. 30- and 90-day mortality rate was 5% and 11%, respectively. Patients with surgical delay > 24 hours had a significantly increased 90- day mortality risk (OR 5.76; p=0.03)
Conclusions: Delay of surgery > 24 hours for patients with distal femoral fractures is associated with increased risk for early mortality. This should be taken into consideration when planning surgery for this group of patients.

92. RSA-measured rotation across SHS-treated trochanteric fractures - is the anti-rotation screw obsolete?
Maiken Stilling, Torben Bæk-Hansen, Bøvling Søren, Steffen Jacobsen, Kim Holck, Henrik Palm
Department of Orhopaedics, Regional Hospital Holstebro, Hospital Unit West

Background: Rotation across trochanteric fractures treated by a sliding hip screw (SHS) remains controversial, and measurement is difficult. Some surgeons add a so-called anti-rotation screw, but the indication is questionable.
Purpose / Aim of Study: To investigate the amount of, and possible predictors for, rotation across SHS-treated stabile trochanteric fractures by use of RSA.
Materials and Methods: 24 patients (21 female) at mean age 76 (56-91) years with SHS-treated stabile trochanteric fractures were followed postop, 1.5, 3 and 6 months after surgery with marker-based RSA and measurement of fracture migration between the femoral head/neck and the trochanter/shaft fragment. EGS- RSA was used to transfer the rotation axis in line with the sliding screw. Screw position within the femoral head was assessed into thirds as high/central/low in AP and posterior/central/anterior in LA post- operative radiographs.
Findings / Results: No cut-outs or revisions were observed. Average rotation across the fracture around the sliding screw was 0.6 (sd 7.1) degrees, and rotation stopped at 6 weeks (p>0.24). 7 patients with non-central screw position (3 with a low AP screw position, 4 with an LA anterior or posterior screw position) had higher rotation compared with 17 patients with central screw placement in two planes (mean 5.3 (sd 4.5) vs. 1.5 (sd 7.1) degrees, p=0.01). Low positioned screws with mean rotation of 8.8 (sd 3.8) degrees had the most rotation (p=0.02). Other translations and rotations were not correlated with screw position.
Conclusions: Overall rotation across the trochanteric fracture was small, but depended on SHS position, with most rotation following a low-positioned screw. If the SHS is correctly positioned in the central third of the femoral head in both AP and LA radiographs, adding an anti-rotation screw appears obsolete.

93. Patient safe pain management of acute hip fracture patients requires focus on the kidney function
Morten Baltzer Andersen, Beata Malmqvist, Henrik Palm
Department of Orthopaedic Surgery & The Hospital Pharmacy, Copenhagen University Hospital Hvidovre, Denmark; Department of Orthopaedic Surgery, Copenhagen University Hospital Hvidovre, Denmark

Background: Standard medical treatment including painkillers are often used for hip fracture patients in Denmark. Painkillers and the frequent co-morbidity Chronic Kidney Disease (CKD) are a known as risk factors for the serious complication including Acute Kidney Injury (AKI). AKI are often reversible but increase mortality and hospitalization.
Purpose / Aim of Study: The purpose was to identify the amount of CKD, AKI and use of painkillers among hip fracture patients.
Materials and Methods: 202 consecutive hip fracture patients (71% female, mean age 78 (range 22- 97)) admitted through the ER from Sept 2012 to March 2013 were included. Painkillers were defined as NSAID and/or Morphine, both included in our standard medicine package. The patient’s upper and lower estimated Glomerular Filtration Rate (eGFR) was recorded between admission and discharge. The stage of CKD was based on the patients' highest eGFR during hospitalization and the AKI definition was based on the eGFR change by the RIFLE (R-risk, I-Injury, F- failure, L-loss, E-end stage) classification.
Findings / Results: 91 (45%) patients had a normal kidney function, while it was reduced mildly in 69 (34%), moderately in 38 (20%), severe in 3 (1%) and a single patient had terminal renal insufficiency. AKI occurred in 20 cases (10%) and patients with reduced kidney function had four times greater risk of developing AKI (p=0.02). Among these cases, use of NSAID was not stopped in 8 and oral Morphine in 7 of the 20 patients after they had developed AKI.
Conclusions: Use of standardized prescribed painkillers appear secure for 90% of hip fracture patients. However, as half of the patients have reduced kidney function and thereby increased risk of AKI, continued focus on kidney function seems necessary for increasing patient safety.

94. Complications and Functional Outcome after Locking Plate Fixation of Distal Tibial Fractures in the Region of Southern Denmark 2007-2011
Silje Kleven, Ellen Hamborg Petersen, Bjarke Viberg, Ole Skov
Dept. of Orthopaedic Surgery and Traumatology, Odense University Hospital

Background: Distal tibia fractures often present a challenge for the orthopaedic surgeon and the best treatment remains controversial. Previous research on locking plates has shown promising results, but we do not know enough about complications and functional outcome
Purpose / Aim of Study: To assess the rate of complication and the long-term functional outcome in patients with distal tibial fractures treated with a low- profile locking plate
Materials and Methods: A historical case-series of 70 patients with 71 distal tibial fractures treated with low-profile locking plate between January 2007 and April 2011 was retrieved. Patient-, injury- and treatment characteristics as well as information of post-operative complications were retrieved from electronic health records and patient interviews. Complications were classified as minor complications and major complications. Long-term functional outcome was assessed by EQ5D-5L, AOFAS score, and return to pre-injury job function through patient interview and examination
Findings / Results: There were 32 43A, 5 43B and 34 43C- fractures, 12 open and 10 high-energy fractures. 49 cases (69%) experienced complications during the follow-up time, of which 34 were minor and 15 were major complications. The median (IQR) EQ5D-5L index was 0.76 (0.65-0.84), health VAS- score 80 (60-90), and AOFAS score 73 (60- 87). Logistic regression analyses showed that smoking increased the risk of minor complications (OR 23, p<0.018). All other variables showed no statistical significance for minor or major complications. 33 % of working patients had not returned to work as a result of the fracture
Conclusions: Our study suggests that treatment of distal tibial fractures with low-profile locking plates might have higher rates of complications and worse functional outcome than previously reported

95. A review of deep wound infection after hip-fracture surgery
Frederik Stensbirk, Henrik Palm
Orthopaedic Surgery, Copenhagen University Hospital Hvidovre

Background: Deep wound infection is one the most severe and costly complications following hip-fracture surgery. The few previously published series have shown multiple re- operations, longer hospitalization, massive antibiotic treatment and increased mortality and morbidity. For improving results, surgeons should besides performing surgical debridement be aware, which organisms are responsible, the relevant antibiotics, and the consequences for patient and hospital.
Purpose / Aim of Study: To investigate rate, bacteriology and antibiotic treatment of patients developing deep wound infection following hip-fracture surgery.
Materials and Methods: 4189 consecutive hip fracture patients admitted and treated at our hospital from September 2002 to June 2013 were included. All patients were treated with arthroplasty, intramedullary nailing, dynamic hip screw or parallel screws/pins. Retrospectively, deep wound infections, bacteriology, antibiotic treatment, hospitalization, number of re-operations and 1-year mortality were investigated.
Findings / Results: 73 patients reoperated due to deep wound infection were identified (1.7%). 80 positive bacterial cultures were identified. 45 displayed S. aureus (63%), 13 coag. neg. staph. (18%), 7 E. coli (10%), 15 distributed on 10 other bacteria, while 6 patients had negative bacteria cultures (8%). 17 cultures were resistant to cefuroxime (21%). 29 patients were treated with >3 antibiotics (40%). Hospitalization was 52 days. 37 patients were re-operated more than once (51%). Within the first year after re- operation 30 patients had died (41%).
Conclusions: A deep wound infection rate at 1.7% is relatively low and comparable to previously published series, but when it occurs, consequences are devastating. Continuous focus on infection registration and optimal prophylactics and treatment is therefore important to improve outcome.

96. Nailing of unstable trochanteric fractures with and without circumferential wires – a study with focus on complications and reoperations within two years.
Lasse Birkelund, Michael Brix, Ilija Ban, Henrik Palm, Anders Troelsen
Ortopædkirurgisk, Aabenraa Sygehus

Background: Open reduction and circumferential wires have long been controversial in trochanteric fractures treated with an intramedullary nail (IMN) because of concerns about compromising the periosteal blood supply leading to bone necrosis. We hypothesize that wires may instead facilitate optimal implant position and fracture reduction
Purpose / Aim of Study: To compare per- and postoperative results in patients with an unstable trochanteric fracture treated with IMN with or without wires, within 2 year from surgery
Materials and Methods: 51 consecutive patients with trochanteric fractures treated with IMN and wires were identified in from two prospective databases. It was compared to, a by age and fracture subtype, matched control group of 51 patients treated with IMN without wires, selected in a third database. ASA, operation-time, bleeding, fracture reduction postoperative, tip-apex distance (TAD) and reoperations within two years were assessed from records and radiographs
Findings / Results: Due to the match method, the 2 groups were alike regarding age and fracture classification. ASA and TAD was NS. Application of wires resulted in significantly longer operation time (p<0.001) and increased bleeding (p<0.001), but a superior reduction (p<0.001). 4/51 patients with wires were reoperated: 1 due to technical failure during osteosynthesis, 1 screw cut out, 1 new fracture after a fall and 1 hardware removal. 9/51 patients without wires were reoperated: 2 deep infections, 3 new fractures after a fall, 1 screw cut out, 1 nail removal and 2 non- unions
Conclusions: It appears that open reduction and application of circumferential wires facilitates a superior fracture reduction, probably worth the increased bleeding and longer operation time. Future level-1 studies are warranted including outcome parameters such as mobilization and pain on short and longer term