Session 7: Hip Fracture

18. November
11:00 - 12:00
Lokale: Auditorium
Chairmen: Bjarke Viberg & Rikke Thorninger

52. Feasibility and preliminary effect of strength training, nutritional supplement and anabolic steroids in rehabilitation of patients with hip fracture: A randomized controlled pilot trial (HIP-SAP1 trial)
Signe Hulsbæk, Thomas Bandhom, Ilija Ban, Nicolai Bang Foss, Jens-Erik Beck Jensen, Henrik Kehlet, Morten Tange Kristensen
Physical Medicine and Rehabilitation Research – Copenhagen (PMR-C), Department of Physiotherapy, Copenhagen University Hospital, Amager-Hvidovre; Department of Orthopedic Surgery, Copenhagen University Hospital, Amager-Hvidovre; Department of Clinical Research, Copenhagen University Hospital, Amager-Hvidovre; Department of Anesthesiology, Copenhagen University Hospital, Amager-Hvidovre; Department of Endocrinology, Copenhagen University Hospital, Amager-Hvidovre; Department of Clinical Medicine, University of Copenhagen; Section for Surgical Pathophysiology, Copenhagen University Hospital, Rigshospitalet.

Background: Anabolic steroid has been suggested as a supplement during hip fracture rehabilitation. A Cochrane Review evaluating the effect of anabolic steroids after hip fracture was inconclusive and recommended further trials.
Aim: To determine feasibility and preliminary effect of a 12-week multimodal intervention consisting of physiotherapy, nutritional supplement and anabolic steroid on knee-extension strength and function after hip fracture surgery.
Materials and Methods: Patients were randomized (1:1) during acute care to: 1. Anabolic steroid or 2. Placebo. Both groups received identical physiotherapy (with strength training) and a nutritional supplement. Primary outcome was change in maximal isometric knee- extension strength from the week after surgery to 14 weeks. Secondary outcomes were physical performance, patient reported outcomes and measures of body composition.
Results: 717 patients were screened, and 23 (target:48) randomized (mean age 73.4 years, 78% women). Main limitations for inclusion were “not home-dwelling” (18%) and “cognitive dysfunction” (16%). Among eligible, the main reason for declining participation was “Overwhelmed and stressed by situation (37%). Adherence to interventions was: Anabolic steroid 87%, exercise 91% and nutrition 61%. Addition of anabolic steroid showed a non-significant between-group difference in knee-extension strength in the fractured leg of 0.11 (95%CI -0.25;0.48) Nm/kg in favor of the anabolic group. Correspondingly, a non-significant between- group difference of 0.16 (95%CI -0.05;0.36) Nm/Kg was seen for the non-fractured leg. No significant between-group differences were identified for the secondary outcomes. 18 adverse reactions were identified (anabolic=10, control=8).
Interpretation / Conclusion: Early inclusion after hip fracture surgery to this trial seemed non-feasible, primarily due to slow recruitment. Although inconclusive, positive tendencies were seen for the addition of anabolic steroid. Trial registration: NCT03545347

53. Comorbidity and quality of in-hospital care for hip fracture patients
Christine Krogsgaard Schrøder, Thomas Johannesson Hjelholt, Morten Madsen, Alma Becic Pedersen, Pia Kjær Kristensen
Klinisk Epidemiologisk afdeling, Århus Universitetshospital, Århus Universitet

Background: Inequalities in healthcare are a persistent challenge. Previous studies have shown patient-related disparity in the quality of in- hospital care. However, it is unknown whether recommended in-hospital care is provided equally to patients with and without known comorbidity.
Aim: We examined whether comorbidity is associated with the quality of in-hospital care among hip fracture patients.
Materials and Methods: From the Danish Multidisciplinary Hip Fracture Registry, we included 31,443 hip fracture patients (2014-2018). Comorbidity was measured using the Charlson Comorbidity Index (CCI). Quality of in- hospital care was defined as fulfillment of process performance measures including preoperative optimization, early surgery, early mobilization, assessment of pain, basic mobility, nutritional risk, need for anti- osteoporotic medication, fall prevention and a post-discharge rehabilitation program, reflecting guideline-recommended in- hospital care for hip fracture patients. The outcomes were 1) all-or-none composite measure defined as fulfillment of all relevant process performance measures and 2) fulfillment of the individual process performance measures. Using binary regression, we calculated relative risk (RR) for the association between CCI and outcomes.
Results: The overall proportion of hip fracture patients, who fulfilled the all-or-none measure, was 31%. Among patients with no comorbidity, 34% fulfilled all-or-none measure vs 29% among patients with high comorbidity (CCI > 3), which corresponds to a 15% lower chance. The impact of comorbidity varied slightly with calendar year. Increasing comorbidity was also associated with lower fulfillment of the individual process performance measures. Preoperative optimization, early surgery and early mobilization appeared to be most difficult to provide patients with comorbidity.
Interpretation / Conclusion: Increasing level of comorbidity was associated with lower quality of in-hospital care among hip fracture patients. Our results highlight the need for tailored clinical interventions to ensure that comorbid patients also benefit from the positive progress in hip fracture care in recent years.

54. Comorbidity in patients with hip fracture; current trends in prevalence and association with 30-day mortality – a population-based cohort study – A population-based cohort study
Pia Kjær Kristensen, Thomas Johannesson Hjeltholt, Alma Becic Pedersen
Department of Clinical Epidemiology, Aarhus University Hospital and Aarhus University

Background: Treatment of hip fracture patients is challenging due to their high comorbidity burden, and mortality. Due to the aging population, we would expect an increasing trend in comorbidity burden of hip fracture patients. It is unclear if use of different comorbidity indices have impact on comorbidity trend and subsequent mortality.
Aim: To examine the current trend in prevalence of comorbidity measured with different indices and the magnitude of the association between comorbidity and 30-day mortality.
Materials and Methods: From the Danish Multidisciplinary Hip Fracture Registry we included 31,443 hip fracture patients (2014-2018). As a measure of comorbidity we used two diagnosed-based indices; Charlson Comorbidity Index (CCI) and Elixhauser, and a medicine-based index; RxRisk. We categorized patients as having no -, moderate -, severe - or very severe comorbidity. We calculated sex and age adjusted odds ratios (aORs) for 30-day mortality with 95% confidence intervals (CI).
Results: Measured with the CCI, 38% of the hip fracture population had no comorbidity, compared to 44% and 28% with the Elixhauser and RxRisk index. The CCI measured 21% with very severe comorbidity whereas Elixhauser and RxRisk index measured 9% and 19 %. The prevalence of patient categories with no, moderate, severe and very severe comorbidity within each index did not change from 2014 to 2018. Compared to patients with no comorbidity, patients with very severe comorbidity had aORs for 30-day mortality of 2.7 (CI: 2.4-2.9) using CCI, 2.6 (CI: 2.4-3.1) using Elixhauser, and 3.1 (CI: 2.7-3.4) using the RxRisk index.
Interpretation / Conclusion: More than 50% of the hip fracture patients has comorbidity, but the prevalence of comorbidity depends on the index used. However, the prevalence of comorbidity burden was stable during the study period irrespective of the index used. All indices had a dose-response association between comorbidity level and 30- day mortality, and the magnitude of the association was unrelated to the index used.

55. Loss of pre-fracture basic mobility status at hospital discharge for hip fracture is associated with 30-day post-discharge risk of infections - A four-year nationwide cohort study of 23,309 Danish patients.
Jeppe Vesterager, Morten Tange Kristensen, Alma Becic Pedersen
Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark; Departments of Physiotherapy and Occupational Therapy - Copenhagen University Hospital – Bispebjerg and Frederiksberg; Department of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark

Background: The loss of prefracture basic mobility status is associated with increased mortality and any readmission after hip fracture. However, it is less known if the loss of prefracture mobility has impact on acquiring a post-discharge infection.
Aim: To examine if the loss of prefracture basic mobility status at hospital discharge was associated with hospital- or community-treated infections within 30-days of hospital discharge after hip fracture.
Materials and Methods: Using the nationwide Danish Multidisciplinary Hip Fracture Registry from January 2014 through November 2017, we included 23,309 patients undergoing surgery for a first-time hip fracture. The Cumulated Ambulation Score (CAS, 0-6 points) was recorded using questionnaire at admission (prefracture CAS) and objectively assessed at discharge. The loss of any CAS-points at discharge compared with prefracture CAS was calculated and dichotomized (yes/no). Using Cox regression analyses, we estimated the hazard ratio (HR) with 95% confidence interval (CI) of any hospital treated infection, hospital-treated pneumonia or community-treated infection adjusted for sex, age, body mass index, Charlson Comorbidity Index, residential status, type of fracture, and length of stay.
Results: Total of 12,046 (62%) patients lost their prefracture CAS status at discharge. Among patients with los of CAS status, 6.0% developed a hospitaltreated infection compared to 4% of those who did not lose their prefracture CAS. Correspondingly, 9.2% versus 6.2% developed a community-treated infection. The risk of 30-day post-discharge infection increased with increasing loss of any CAS points. The adjusted HRs for patients who had lost their prefracture CAS status, compared to patients who did not, was 1.34 (CI: 1.16-1.54) for hospital-treated infection, 1.35 (CI:1.09 – 1.67) for pneumonia and 1.36 (CI: 1.21-1.52) for community-treated infection.
Interpretation / Conclusion: In this large national cohort study, we found that loss of pre-fracture basic mobility status upon hospital discharge was strongly associated with 30-day postdischarge risk of developing infection. This emphasise the clinical importance of carefully focusing on regaining the prefracture basic mobility before discharging the patient.

56. The incidence of hip fractures, amongst elderly aged 70+, continues to decrease.
Tine Nymark, Niels Dieter Röck, Jens Lauritsen
Department of Orthopaedics, Odense University Hospital

Background: Hip fracture patients, constitutes one of the largest groups of patients in most of the orthopedic departments, and represents a substantial burden to the health system. It is therefore of interest to study the development of the number of fractures, and the incidence rates. From studies conducted in our department we know, that the annual number of fractures on Funen in the 70’ties was approximately 500 a year and the incidence rates were increasing. Around the millennium, the number was approximately 800 a year and the rates were decreasing. We have found it of interest to follow up on the above mentioned studies.
Aim: To compare the incidence rates of hip fractures on Funen from the periods 2000-2003 and 2017- 2019 for patients aged 70+, and see if there has been a significant change.
Materials and Methods: Data from the first period are from a study conducted in our department by Nymark et al. Data from the second period are calculated from our local Register, where all patients from Funen (except from the municipalities of Ærø and Middelfart fractures) are included.
Results: For men aged 70+ the incidence rates for the two periods were 0,83 (C.I. 0,72-0,95) and 0,56 (C.I. 0,48-0,65) and for women aged 70+ 1,86 (C.I. 1,72-2,01) and 1,01 (C.I. 0,91-1,11) in both cases a significant reduction of 32% and 46%. If the data are split into four age groups 70-79 80-84 85-90 and 90+the same tendency is seen. The age- and sex specific rates have decreased between 18,6 and 54,9% in the 8 groups. The number of patients in the period 2017-19 was 519, 568 and 515 and corrected for inhabitants in the two excluded municipalities: 570, 631 and 572.
Interpretation / Conclusion: The age- and sex specifik incidence rates for hip fractures have fallen significantly from the period 2000-2003 to 2017-2019 in all age groups older than 70. The patient group is still a large group, but the absolute number of fractures has decreased as well, which means that the expected rise, due to an increasing elderly population, has not been seen. It seems as if the general health among the elderly population has improved, and therefore compensated for the forecasted rise in the number of hip fractures.

57. Development and Validation of a Model for Predicting Mortality in Patients with Hip Fracture: Population-Based Cohort Study
Thomas Hjelholt, Søren Johnsen, Peter Brynningsen, Jakob Knudsen, Daniel Prieto-Alhambra, Alma Pedersen
Department of Clinical Epidemiology, Aarhus University Hospital; Danish Center for Clinical Health Services Research, Department of Clinical Medicine, Aalborg University; Department of Geriatrics, Aarhus University Hospital; Department of Clinical Pharmacology, Aarhus University Hospital; Pharmaco- and Device Epidemiology, Centre for Statistics in Medicine, NDORMS, University of Oxford

Background: One-year mortality following hip fracture surgery is 30% on average. However, user-friendly prediction tools to guide clinicians and patients on appropriate targeted preventive measures are needed.
Aim: We aimed to develop a user-friendly chart displaying one-year mortality of hip fracture patients.
Materials and Methods: Using the population-based Danish Hip Fracture Registry, we identified all patients with a first-time hip fracture in 2011-2017 (N=28,791). We assessed patient-related prognostic factors available at the time of admission as potential predictors of mortality: Nursing home residency, comorbidity (Charlson Comorbidity Index), frailty (Hospital Frailty Risk Score), basic mobility (Cumulated Ambulation Score), atrial fibrillation, fracture type, Body Mass Index, age, and sex. We examined the association with one-year mortality by determining the cumulative incidence, applying univariable logistic regression and assessing discrimination (area under the ROC curve [AUROC]). We fitted a decision tree model on a development cohort (70% of patients) and plotted the relative variable importance of each predictor. We then selected relevant predictors for the final model (logistic regression). We subsequently assessed discrimination and calibration based on the validation cohort (remaining 30% of patients).
Results: All predictors showed an association with one-year mortality, but discrimination was moderate; age and Charlson Comorbidity Index had the best AUROC of 0.65 and 0.61, respectively. The final model included nursing home residency, Charlson Comorbidity Index, Cumulated Ambulation Score, Body Mass Index, and age. It had an acceptable discrimination (AUROC 0.74) and calibration, and predicted one- year mortality risk spanning from 5% to 91% depending on the combination of predictors in the individual patient.
Interpretation / Conclusion: Using information obtainable at the time of admission, one-year mortality among patients with hip fracture can be predicted. We present a user- friendly chart for daily clinical practice and provide new insight regarding the interplay between prognostic factors.

58. Quality of in-hospital care and postoperative complications and mortality among hip fracture patients with Parkinson's disease.
Peter Nguyen, Thomas Johannesson Hjelholt, Alma Becic Pedersen
Department of Clinical Epidemiology, Aarhus University Hospital

Background: Patients with Parkinson’s disease (PD) have a high risk of sustaining fractures. They are also less likely to regain their previous functional status after hip fracture, and have a higher risk of complications than patients without PD. It remains uncertain if mortality is affected by PD and in addition, no studies have investigated if quality of care is equal for PD and non-PD patients.
Aim: To investigate the association between PD and quality of in-hospital care, postoperative complications, and mortality in patients with hip fracture.
Materials and Methods: We included patients aged 65+ with an incident hip fracture from 2004-2017, registered in Danish Multidisciplinary Hip Fracture Registry who had not been treated with antibiotics or admitted with an infection 7 days prior to hip surgery. Patients with PD were defined with ICD-10 “G20” prior to hip fracture. Using log- binomial regression, we calculated both 30- day crude and adjusted risk ratios (aRR) with 95% confidence intervals (CIs) for the following outcomes: hospital-treated infections as well as pneumonia, urinary tract infection, sepsis, community-treated infections, cardiovascular events, quality of in-hospital care (measured by fulfilment of quality indicators) and mortality. Analyses were adjusted for age, sex and Charlson comorbidity score.
Results: We identified 75.635 patients without and 1915 patients with PD at the time of hip fracture. Compared to non-PD, presence of PD was associated with higher risk of any hospital-treated (aRR = 1.27 (CI: 1.10-1.45) and community-treated infection (aRR = 1.28 (CI: 1.20-1.37)), pneumonia (aRR = 1.38 (1.11-1.69)), urinary tract infection (aRR of 1.58 (CI: 1.28-1.92)) and sepsis (aRR = 1.18 (CI: 0.67-1.89)), but a reduced aRR for cardiovascular events of 0.59 (CI: 0.41-0.82). PD was associated with increased risk of 30-day mortality (aRR = 1.11 (CI: 0.97-1.27)). aRRs for fulfillment of all quality indicators was found to be approximately 1.
Interpretation / Conclusion: Hip fracture patients with PD have a higher risk of infections and mortality within 30 days after surgery. They do however receive equal quality of in-hospital care after hip fracture compared to non-PD patients.