Twenty-five years of the trauma audit and research network: a continuing evolution to drive improvement

Epidemiology and outcomes of severe injury patients: Nationwide community-based study in Korea

INTRODUCTION

The aim of this study was to estimate the annual incidence of various severe injuries, case fatality rates, and case disability rates for patients in Korea.

MATERIALS AND METHODS

This community-based observational study involved all residents in Korea between Jan. 1, 2018, and Dec. 31, 2018. Nationwide data were retrieved from the patient care report of the National Fire Agency and hospital medical records review of the Korea Disease Control and Prevention Agency. Severe injury was defined as any injury that resulted in emergency medical service (EMS) use and any field-based physiologic abnormality: systolic blood pressure ≤90 mm Hg, respiratory rate <10 or >29 per min, or AVPU scale nonalert. Crude and adjusted incidence rates were calculated without any exclusion. Outcomes including the Glasgow Outcome Scale (GOS) at hospital discharge were assessed according to a medical review. Case-fatality and case-disability rates were calculated after excluding patients with unknown vital status.

RESULTS

During the one-year period, 36,363 severe injuries occurred, yielding a crude annual incidence rate of 70.9 per 100,000 population; 57.8% were traumatic injuries, and 42.2% were nontraumatic injuries. The adjusted annual incidence rates were 61.7/100,000 for all severe injuries, 35.3/100,000 for traumatic injuries, and 26.4/100,000 for nontraumatic injuries. The case-fatality rate was 24.3% for all severe injuries, 27.0% for traumatic injuries, and 21.0% for nontraumatic injuries. The case-disability rate of GOS 3,4 was 6.0%, and the case-disability rate of GOS 2-4 was 16.4% for all severe injury patients.

CONCLUSIONS

This nationwide community-based study revealed incidence rates and outcomes of severe injury patients in Korea.

Healthcare system impacts of the 2017 Manchester Arena bombing: evidence from a national trauma registry patient case series and hospital performance data

Introduction

In response to detonation of an improvised explosive device at the Manchester Arena on 22 May 2017, we aimed to use detailed information about injured patients flowing through hospital healthcare to objectively evaluate the preplanned responses of a regional trauma care system and to show how routinely collected hospital performance data can be used to assess impact on regional healthcare.

Methods

Data about injury severity, management and outcome for patients presenting to hospitals were collated using England’s major trauma registry for 30 days following hospital attendance. System-wide data about hospital performance were collated by National Health Service England’s North West Utilisation Management Unit and presented as Shewhart charts from 15 April 2017 to 25 June 2017.

Results

Detailed information was obtained on 153 patients (109 adults and 44 children) who attended hospital emergency departments after the incident. Within 6 hours, a network of 11 regional trauma care hospitals received a total of 138 patients (90%). For the whole patient cohort, median Injury Severity Score (ISS) was 1 (IQR 1–10) and median New ISS (NISS) was 2 (IQR 1–14). For the 75 patients (49%) attending a major trauma centre, median ISS was 7.5 (IQR 1–14) and NISS was 10 (IQR 3–22). Limb and torso body regions predominated when injuries were classified as major life threatening (Abbreviated Injury Scale>3). Ninety-three patients (61%) required hospital admission following emergency department management, with 21 (14%) requiring emergency damage control surgery and 24 (16%) requiring critical care. Three fatalities occurred during early resuscitative treatment and 150 (98%) survived to day 30. The increased system-wide hospital admissions and care activity was linked to increases in regional hospital care capacity through cancellations of elective surgery and increased community care. Consequently, there were sustained system-wide hospital service improvements over the following weeks.

Conclusions

The systematic collation of injured patient and healthcare system data has provided an objective evaluation of a regional major incident plan and provided insight into healthcare system resilience. Hospital patient care data indicated that a prerehearsed patient dispersal plan at incident scene was implemented effectively.

Retrograde intramedullary nail fixation compared with fixed-angle plate fixation for fracture of the distal femur: the TrAFFix feasibility RCT

BACKGROUND

Fractures of the distal femur are an increasingly common injury; the optimal management of these injuries remains controversial. The two interventions used in UK practice are intramedullary fixation, with a locked retrograde nail, and extramedullary fixation, with a fixed angle plate.

OBJECTIVES

This study assessed the feasibility of a definitive trial and included a process evaluation to understand the generalisability and likely success of a future trial.

DESIGN

A multicentre, parallel, two-arm, randomised controlled feasibility trial with an embedded process evaluation. Treatment with a plate or nail was allocated in a 1 : 1 ratio, stratified by centre and chronic cognitive impairment. Surgeons were not blinded, but participants were not told their allocation.

SETTING

Seven NHS hospitals.

PARTICIPANTS

Patients aged ≥ 18 years with a fracture of the distal femur who the attending surgeon believed would benefit from internal fixation were potentially eligible. Patients were excluded if they had a loose arthroplasty requiring revision or a femoral deformity or arthroplasty that precluded nail fixation. The sample was recruited between 29 September 2016 and 31 August 2017. Consent was obtained from the patient or appropriate consultee before enrolment.

INTERVENTIONS

Patients were randomised to receive fixation of their distal femur fracture with either a proximally and distally locked retrograde nail that spanned the diaphysis of the femur or an anatomical distal femoral locking plate with at least one locked screw distal to the fracture. Reduction and supplemental fixation were at the surgeon's discretion.

OUTCOMES

The primary outcome measures for this study were the recruitment rate and the completion rate of the EuroQol-5 Dimensions, five-level version (EQ-5D-5L), at 6 weeks and 4 months. Additional measurements included baseline characteristics, measures of social support and self-efficacy, disability rating index, dementia quality-of-life measures and a radiographical assessment of any malunion. Participants' and staff views were obtained, at interview, for the process evaluation.

RESULTS

The process evaluation showed that surgeon-related factors, principally confidence with both technologies and a lack of individual equipoise, were key barriers to recruitment. A total of 23 participants were randomised and analysed (nail, n = 11; plate, n = 12). The recruitment rate was estimated as 0.42 [95% confidence interval (CI) 0.27 to 0.62] participants per centre per month, lower than the prespecified feasibility threshold of 1.0 participants per centre per month. Data completeness of the EQ-5D-5L was estimated at 65% (95% CI 43% to 83%).

CONCLUSIONS

This feasibility study has challenged many of the assumptions that underpinned the development of proposed definitive trial protocol. A modified protocol is proposed that would be feasible given the recruitment rate observed here, which is equal to that reported in the similar FixDT trial [Health Technology Assessment (HTA) 11/136/04: Costa ML, Achten J, Hennings S, Boota N, Griffin J, Petrou S, et al. Intramedullary nail fixation versus locking plate fixation for adults with a fracture of the distal tibia: the UK FixDT RCT. Health Technol Assess 2018;22(25)], which delivered to target and budget.

FUTURE WORK

A definitive trial with a modified design is recommended, including an internal pilot to confirm initial recruitment rate assumptions.

REGISTRATION

Current Controlled Trials ISRCTN92089567.

FUNDING

This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 23, No. 51. See the NIHR Journals Library website for further project information. Funding was also supported by the NIHR Oxford Biomedical Research Centre.

Age and the distribution of major injury across a national trauma system

Background

Trauma places a significant burden on healthcare services, and its management impacts greatly on the injured patient. The demographic of major trauma is changing as the population ages, increasingly unveiling gaps in processes of managing older patients. Key to improving patient care is the ability to characterise current patient distribution.

Objectives

There is no contemporary evidence available to characterise how age impacts on trauma patient distribution at a national level. Through an analysis of the Trauma Audit Research Network (TARN) database, we describe the nature of Major Trauma in England since the configuration of regional trauma networks, with focus on injury distribution, ultimate treating institution and any transfer in-between.

Methods

The TARN database was analysed for all patients presenting from April 2012 to the end of October 2017 in NHS England.

Results

About 307,307 patients were included, of which 63.8% presented directly to a non-specialist hospital (trauma unit (TU)). Fall from standing height in older patients, presenting and largely remaining in TUs, dominates the English trauma caseload. Contrary to perception, major trauma patients currently are being cared for in both specialist (major trauma centres (MTCs)) and non-specialist (TU) hospitals. Paediatric trauma accounts for <5% of trauma cases and is focussed on paediatric MTCs.

Conclusions

Within adult major trauma patients in England, mechanism of injury is dominated by low level falls, particularly in older people. These patients are predominately cared for in TUs. This work illustrates the reality of current care pathways for major trauma patients in England in the recently configured regional trauma networks.

[Data content of the TraumaRegister DGU® : Results of a random sample control]

BACKGROUND

Registries are becoming increasingly more important in clinical research. The TraumaRegister DGU® of the German Society for Trauma Surgery plays an excellent role with respect to the care of severely injured patients.

AIM

Within the framework of this investigation the quality of data provided by this registry was to be verified.

MATERIAL AND METHODS

Certified hospitals participating in the TraumaNetzwerk DGU® of the German Society for Trauma Surgery are obliged to submit data of treated severely injured patients to the TraumaRegister DGU®. Participating hospitals have to undergo a re-certification process every 3 years. Within the framework of this re-audit, data from 5 out of 8 randomly chosen patient cases included in the registry are controlled and compared to the patient files of the certified hospital. In the present investigation discrepancies concerning data provided were documented and the pattern of deviation was analyzed.

RESULTS

The results of 1075 re-certification processes carried out in 631 hospitals including the documentation of 5409 checked patient cases from 2012-2017 were analyzed. The highest number of discrepancies detected concerned the documented time until initial CT (15.8%) and the lowest concerned the discharge site (3.2%). The majority of data sheets with discrepancies showed deviations in only one out of seven checked parameters. Interestingly, large trauma centers with a high throughput of severely injured patients showed the most deviations.

CONCLUSION

The present investigation underlines the importance of standardized checks concerning data provided for registries in order to be able to guarantee an improvement in entering data.

Intramedullary nails versus distal locking plates for fracture of the distal femur: results from the Trial of Acute Femoral Fracture Fixation (TrAFFix) randomised feasibility study and process evaluation

OBJECTIVES

This feasibility study and process evaluation assessed the likely success of a definitive trial of intramedullary fixation with locked retrograde nails versus extramedullary fixation with fixed angle plates for fractures of the distal femur.

DESIGN & SETTING

A multicentre, parallel, two-arm, randomised controlled feasibility study with an embedded process evaluation was conducted at seven NHS hospitals in England. Treatment was randomly allocated in 1:1 ratio, stratified by centre and chronic cognitive impairment. Participants, but not surgeons or research staff, were blinded to the allocation.

PARTICIPANTS

Patients 18 years and older with a fracture of the distal femur, who their surgeon believed would benefit from internal fixation, were eligible to take part.Participants were allocated to receive either a retrograde intramedullary nail or an anatomical locking plate.

OUTCOMES

The primary outcomes for this feasibility study were the recruitment rate and completion rate of the EQ-5D-5L at 4 months post-randomisation. Baseline characteristics, disability rating index, quality of life scores, measurements of social support and self-efficacy, resource use and radiographic assessments were also collected. The views of patients and staff were collected during interviews.

RESULTS

Recruitment and data completion were lower than expected. 23 of 82 eligible patients were recruited (nail, 11; plate, 12). The recruitment rate was estimated as 0.42 (95% CI 0.27 to 0.62) participants per centre-month. Data completeness of the EQ-5D-5L at 4 months was 61 per cent (95% CI 43% to 83%). The process evaluation demonstrated that the main barriers to recruitment were variation in treatment pathways across centres, lack of surgeon equipoise and confidence in using both interventions and newly formed research cultures that lacked cohesion.

CONCLUSIONS

A modified trial design, with embedded recruitment support intervention, comparing functional outcome in cognitively intact adults who have sustained a fragility fracture of the distal femur is feasible.

ETHICS APPROVAL

The Wales Research Ethics Committee 5 approved the study (ref: 16/WA/0225).

TRIAL REGISTRATION NUMBER

ISRCTN92089567; Pre-results.

Disparities in the management of paediatric splenic injury

BACKGROUND

The non-operative management of splenic injury in children is recommended widely, and is possible in over 95 per cent of episodes. Practice appears to vary between centres.

METHODS

The Trauma Audit and Research Network (TARN) database was interrogated to determine the management of isolated paediatric splenic injuries in hospitals in England and Wales. Rates of non-operative management, duration of hospital stay, readmission and mortality were recorded. Management in paediatric surgical hospitals was compared with that in adult hospitals.

RESULTS

Between January 2000 and December 2015 there were 574 episodes. Children treated in a paediatric surgical hospital had a 95·7 per cent rate of non-operative management, compared with 75·5 per cent in an adult hospital (P < 0·001). Splenectomy was done in 2·3 per cent of children in hospitals with a paediatric surgeon and in 17·2 per cent of those treated in an adult hospital (P < 0·001). There was a significant difference in the rate of non-operative management in children of all ages. There was some improvement in non-operative management in adult hospitals in the later part of the study, but significant ongoing differences remained.

CONCLUSION

The management of children with isolated splenic injury is different depending on where they are treated. The rate of non-operative management is lower in hospitals without a paediatric surgeon present.

Assessing the Outcome of Hip Arthroscopy for Labral Tears in Femoroacetabular Impingement Using the Minimum Dataset of the British Non-arthroplasty Hip Register: A Single-Surgeon Experience

PURPOSE

The aim of this study was to assess changes in British Non-arthroplasty Hip Register (NAHR) minimum dataset (MDS) patient-reported outcome measures (PROMs) after hip arthroscopy for femoroacetabular impingement (FAI) and define the relation between these and patient satisfaction. Secondary aims included exploring the impact of patient characteristics (age, sex, and social deprivation status) on MDS PROMs and satisfaction and determining the Net Promoter Score for hip arthroscopy for FAI.

METHODS

Preoperative data were collected from the NAHR, and postoperative data were collected through the NAHR, by mail, and by telephone survey. Correlations between satisfaction, International Hip Outcome Tool 12 (iHOT-12), and EQ-5D scores were explored.

RESULTS

A consecutive series of 89 primary hip arthroscopy procedures for FAI in 88 patients is reported. Patients reported improvements in the iHOT-12 score (mean, 34.08; 95% confidence interval [CI], 27.88 to 40.28; P < .001), EQ-5D index score (+0.124; 95% CI, 0.063 to 0.185; P < .001), and EQ-5D visual analog scale (VAS) (+4.49; 95% CI, -1.56 to 10.54; P = .061) after hip arthroscopy for FAI. Satisfaction was predicted by both change in iHOT-12 score (Spearman r [r_s] = 0.54, P < .001) and absolute postoperative iHOT-12 score (r_s = 0.78, P < .001), change in EQ-5D index score (r_s = 0.42, P < .001) and absolute postoperative EQ-5D index score (r_s = 0.70, P < .001), and change in EQ-5D VAS score (r_s = 0.30, P = .012) and absolute postoperative EQ-5D VAS score (r_s = 0.59, P < .001); and the strength of correlation was greater with the absolute postoperative score than with the change in score for all 3. Sex, age, and social deprivation status did not predict postoperative PROMs (P ≥ .15) or satisfaction (P ≥ .32). The postoperative iHOT-12 score correlated strongly with EQ-5D index (r_s = 0.90, P < .001) and EQ-5D VAS (r_s = 0.81, P < .001) scores. The Net Promoter Score for hip arthroscopy for FAI was 70.31.

CONCLUSIONS

This study showed significant improvements in hip-specific function (iHOT-12) and health-related quality of life (EQ-5D), as measured by the NAHR MDS, in patients undergoing hip arthroscopy for FAI. Satisfaction rates were high (75.7%) and correlated strongly with hip-specific and general health PROMs. Satisfied patients were more likely to be willing to undergo similar surgery in the future. Self-reported postoperative hip function correlated very strongly with general health-related quality of life.

LEVEL OF EVIDENCE

Level IV, case series.

Management of blunt splenic injury in a UK major trauma centre and predicting the failure of non-operative management: a retrospective, cross-sectional study

PURPOSE

To review the management of patients >16 years with blunt splenic injury in a single, UK, major trauma centre and identify whether the following are associated with success or failure of non-operative management with selective use of arterial embolization (NOM ± AE): age, Injury Severity Score (ISS), head injury, haemodynamic instability, massive transfusion, radiological hard signs [contrast extravasation or pseudoaneurysm on the initial computed tomography (CT) scan], grade, and presence of intraparenchymal haematoma or splenic laceration.

METHODS

Retrospective, cross-sectional study undertaken between April 2012 and October 2015. Paediatric patients, penetrating splenic trauma, and iatrogenic injuries were excluded. Follow-up was for at least 30 days.

RESULTS

154 patients were included. Median age was 38 years, 77.3% were male, and median ISS was 22. 14/87 (16.1%) patients re-bled following NOM in a median of 2.3 days (IQR 0.8-3.6 days). 8/28 (28.6%) patients re-bled following AE in a median of 2.0 days (IQR 1.3-3.7 days). Grade III-V injuries are a significant predictor of the failure of NOM ± AE (OR 15.6, 95% CI 3.1-78.9, p = 0.001). No grade I injuries and only 3.3% grade II injuries re-bled following NOM ± AE. Age ≥55 years, ISS, radiological hard signs, and haemodynamic instability are not significant predictors of the failure of NOM ± AE, but an intraparenchymal or subcapsular haematoma increases the likelihood of failure 11-fold (OR 10.9, 95% CI 2.2-55.1, p = 0.004).

CONCLUSIONS

Higher grade injuries (III-V) and intraparenchymal or subcapsular haematomas are associated with a higher failure rate of NOM ± AE and should be managed more aggressively. Grade I and II injuries can be discharged after 24 h with appropriate advice.

Predictors of Outcome in Traumatic Brain Injury: New Insight Using Receiver Operating Curve Indices and Bayesian Network Analysis

Background

Traumatic brain injury remains a global health problem. Understanding the relative importance of outcome predictors helps optimize our treatment strategies by informing assessment protocols, clinical decisions and trial designs. In this study we establish importance ranking for outcome predictors based on receiver operating indices to identify key predictors of outcome and create simple predictive models. We then explore the associations between key outcome predictors using Bayesian networks to gain further insight into predictor importance.

Methods

We analyzed the corticosteroid randomization after significant head injury (CRASH) trial database of 10008 patients and included patients for whom demographics, injury characteristics, computer tomography (CT) findings and Glasgow Outcome Scale (GCS) were recorded (total of 13 predictors, which would be available to clinicians within a few hours following the injury in 6945 patients). Predictions of clinical outcome (death or severe disability at 6 months) were performed using logistic regression models with 5-fold cross validation. Predictive performance was measured using standardized partial area (pAUC) under the receiver operating curve (ROC) and we used Delong test for comparisons. Variable importance ranking was based on pAUC targeted at specificity (pAUC_SP) and sensitivity (pAUC_SE) intervals of 90–100%. Probabilistic associations were depicted using Bayesian networks.

Results

Complete AUC analysis showed very good predictive power (AUC = 0.8237, 95% CI: 0.8138–0.8336) for the complete model. Specificity focused importance ranking highlighted age, pupillary, motor responses, obliteration of basal cisterns/3rd ventricle and midline shift. Interestingly when targeting model sensitivity, the highest-ranking variables were age, severe extracranial injury, verbal response, hematoma on CT and motor response. Simplified models, which included only these key predictors, had similar performance (pAUC_SP = 0.6523, 95% CI: 0.6402–0.6641 and pAUC_SE = 0.6332, 95% CI: 0.62–0.6477) compared to the complete models (pAUC_SP = 0.6664, 95% CI: 0.6543–0.679, pAUC_SE = 0.6436, 95% CI: 0.6289–0.6585, de Long p value 0.1165 and 0.3448 respectively). Bayesian networks showed the predictors that did not feature in the simplified models were associated with those that did.

Conclusion

We demonstrate that importance based variable selection allows simplified predictive models to be created while maintaining prediction accuracy. Variable selection targeting specificity confirmed key components of clinical assessment in TBI whereas sensitivity based ranking suggested extracranial injury as one of the important predictors. These results help refine our approach to head injury assessment, decision-making and outcome prediction targeted at model sensitivity and specificity. Bayesian networks proved to be a comprehensive tool for depicting probabilistic associations for key predictors giving insight into why the simplified model has maintained accuracy.