Hospital learning curves for robot-assisted surgeries: a population-based analysis

BACKGROUND

Robot-assisted surgery has been rapidly adopted. It is important to define the learning curve to inform credentialling requirements, training programs, identify fast and slow learners, and protect patients. This study aimed to characterize the hospital learning curve for common robot-assisted procedures.

STUDY DESIGN

This cohort study, using administrative health data for Ontario, Canada, included adult patients who underwent a robot-assisted radical prostatectomy (RARP), total robotic hysterectomy (TRH), robot-assisted partial nephrectomy (RAPN), or robotic portal lobectomy using four arms (RPL-4) between 2010 and 2021. The association between cumulative hospital volume of a robot-assisted procedure and major complications was evaluated using multivariable logistic models adjusted for patient characteristics and clustering at the hospital level.

RESULTS

A total of 6814 patients were included, with 5230, 543, 465, and 576 patients in the RARP, TRH, RAPN, and RPL-4 cohorts, respectively. There was no association between cumulative hospital volume and major complications. Visual inspection of learning curves demonstrated a transient worsening of outcomes followed by subsequent improvements with experience. Operative time decreased for all procedures with increasing volume and reached plateaus after approximately 300 RARPs, 75 TRHs, and 150 RPL-4s. The odds of a prolonged length of stay decreased with increasing volume for patients undergoing a RARP (OR 0.87; 95% CI 0.82-0.92) or RPL-4 (OR 0.77; 95% CI 0.68-0.87).

CONCLUSION

Hospitals may adopt robot-assisted surgery without significantly increasing the risk of major complications for patients early in the learning curve and with an expectation of increasing efficiency.

Machine learning in the prediction of massive transfusion in trauma: a retrospective analysis as a proof-of-concept

PURPOSE

Early administration and protocolization of massive hemorrhage protocols (MHP) has been associated with decreases in mortality, multiorgan system failure, and number of blood products used. Various prediction tools have been developed for the initiation of MHP, but no single tool has demonstrated strong prediction with early clinical data. We sought to develop a massive transfusion prediction model using machine learning and early clinical data.

METHODS

Using the National Trauma Data Bank from 2013 to 2018, we included severely injured trauma patients and extracted clinical features available from the pre-hospital and emergency department. We subsequently balanced our dataset and used the Boruta algorithm to determine feature selection. Massive transfusion was defined as five units at 4 h and ten units at 24 h. Six machine learning models were trained on the balanced dataset and tested on the original.

RESULTS

A total of 326,758 patients met our inclusion with 18,871 (5.8%) requiring massive transfusion. Emergency department models demonstrated strong performance characteristics with mean areas under the receiver-operating characteristic curve of 0.83. Extreme gradient boost modeling slightly outperformed and demonstrated adequate predictive performance with pre-hospital data only, as well as 4-h transfusion thresholds.

CONCLUSIONS

We demonstrate the use of machine learning in developing an accurate prediction model for massive transfusion in trauma patients using early clinical data. This research demonstrates the potential utility of artificial intelligence as a clinical decision support tool.

Surgical Stabilization of Rib Fractures for Flail Chest: Analysis of Center-Based Variability in Practice and Outcomes

BACKGROUND

Given the lack of high-quality data on patient selection for surgical stabilization of rib fractures (SSRF), significant variability in practice likely exists across trauma centers. We aimed to determine whether centers with a more liberal approach to SSRF had improved outcomes.

METHODS

We performed a retrospective cohort study of adult patients with flail chest admitted to level I or II trauma centers participating in the American College of Surgeons' Trauma Quality Improvement Program. The primary outcome was hospital mortality; secondary outcomes included discharge status, tracheostomy, duration of mechanical ventilation, and hospital length of stay. Logistic regression was performed to calculate center-level observed/expected rates of SSRF and centers were grouped into quintiles from "most liberal" to "most restrictive". Multivariable regression was used to determine the association between these quintiles and outcomes. We also used an instrumental variable analysis to evaluate the association between SSRF and mortality at the patient level.

RESULTS

Among 23,619 patients with flail chest across 354 centers, 22% underwent SSRF. Center rates of fixation ranged from 0-88%. Higher rates of SSRF were not associated with lower mortality overall (highest vs. lowest quintile OR 0.86, 95% CI 0.63-1.17). However, centers with a more liberal approach to SSRF had lower rates of independent status at discharge, higher tracheostomy rates, longer duration of mechanical ventilation, and longer hospital and ICU length of stay. The patient level analysis demonstrated that SSRF as was associated with a 25% lower risk of death.

CONCLUSIONS

Overall, centers with a liberal approach to SSRF do not show improved outcomes among patients with a flail chest, but have higher resource utilization. Results at the patient level suggest that there is a population likely to benefit but these patients remain to be identified through further research.

LEVEL OF EVIDENCE

Prognostic/epidemiological, Level III.

Firearm injuries in Missouri

Firearm deaths continue to be a major public health problem, but the number of non-fatal firearm injuries and the characteristics of patients and injuries is not well known. The American College of Surgeons Committee on Trauma, with support from the National Collaborative on Gun Violence Research, leveraged an existing data system to capture lethal and non-lethal injuries, including patients treated and discharged from the emergency department and collect additional data on firearm injuries that present to trauma centers. In 2020, Missouri had the 4th highest firearm mortality rate in the country at 23.75/100,000 population compared to 13.58/100,000 for the US overall. We examined the characteristics of patients from Missouri with firearm injuries in this cross-sectional study. Of the overall 17,395 patients, 1,336 (7.7%) were treated at one of the 11 participating trauma centers in Missouri during the 12-month study period. Patients were mostly male and much more likely to be Black and uninsured than residents in the state as a whole. Nearly three-fourths of the injuries were due to assaults, and overall 7.7% died. Few patients received post-discharge services.

Firearm Injuries treated at Trauma Centers in the United States

BACKGROUND

While firearm injuries and deaths continue to be a major public health problem, the number of non-fatal firearm injuries and the characteristics of patients is not well known. The American College of Surgeons (ACS) Committee on Trauma leveraged an existing data system to collect additional data on fatal and non-fatal firearm injuries presenting to trauma centers. This report provides an overview of this initiative and highlights the challenges associated with capturing actionable data on firearm-injured patients.

METHODS

128 trauma centers that are part of the ACS Trauma Quality Improvement Program (TQIP) collected data on individuals of any age arriving alive between March 1, 2021 and February 28, 2022 with a firearm injury. In addition to the standard data collected for TQIP, abstractors also extracted additional data specific to this study. We linked data from the Distressed Community Index (DCI) to patient records using zip code of residence.

RESULTS

A total of 17,395 patients were included, with mean (SD) age of 30.2 (13.5) years, 82.5% were male and the majority were Black and non-Hispanic. The mean proportion of variables with missing data varied among trauma centers, with a mean of 20.7% missing data. Injuries occurred most commonly in homes (31.2%) or on the street (26.6%); 70.4% of injuries were due to assaults. Nearly one-third of patients were discharged from the ED, 25.9% were admitted directly to the operating room, 10.9% to the ICU; 5.9% died in the ED and 10.3% died overall during their course of care. Nearly two-thirds of patients lived in the two highest distressed categories of communities; only 7.5% lived in the least distressed quintile.

CONCLUSIONS

Utilizing trauma center data can be a valuable tool to improve our knowledge of firearm injuries if clinical practices and documentation of patient risks and circumstances are standardized.

LEVEL OF EVIDENCE

III Level, epidemiological.

Predictors of hospital-acquired pressure injuries in patients with complete spinal cord injury: a retrospective case-control study

BACKGROUND

Despite current best practices, pressure injuries (PI) remain a devastating and prevalent hospital-acquired complication for patients with acute traumatic spinal cord injuries (SCIs). This study examined associations between risk factors for PI development in patients with complete SCI, such as norepinephrine dose and duration, and other demographic factors or lesion characteristics.

METHODS

This case-control study included adults with acute complete SCIs ASIA-A, who were admitted to a level-one trauma center between 2014-18. A retrospective review was implement using data on patient and injury characteristics, including age, gender, level of SCI (cervical vs. thoracic), Injury Severity Score (ISS), length of stay (LOS) and mortality; presence/absence of PI during their acute hospital stay; and treatment factors such as spinal surgery, mean arterial pressure (MAP) targets, and vasopressor treatment. Multivariable logistic regression evaluated associations with PI.

RESULTS

Eighty-two out of 103 eligible patients had complete data, and 30 (37%) developed PIs. Patient and injury characteristics, including age (Mean: 50.6; SD:21.3), location of SCI (48 cervical, 59%) and ISS (Mean 33.1; SD:11.8), did not differ between PI and non-PI groups. Logistic regression analysis revealed that male gender (OR:34.1; CI₉₅:2.3-506.5, p = 0.010) and increased LOS (log-transformed; OR:20.5, CI₉₅:2.8-149.9, p = 0.003) were associated with increased risk of PI. Having an order for a MAP > 80mmg (OR:0.05; CI₉₅:0.01-0.30, p = 0.001) was associated with a reduced risk of PI. There were no significant associations between PI and duration of norepinephrine treatment.

CONCLUSIONS

Norepinephrine treatment parameters were not associated with development of PI, suggesting that MAP targets should be a focus for future investigations for SCI management. Increasing LOS should highlight the need for high-risk PI prevention and vigilance.

The evolving use of robotic surgery: a population-based analysis

INTRODUCTION

Robotic surgery has integrated into the healthcare system despite limited evidence demonstrating its clinical benefit. Our objectives were (i) to describe secular trends and (ii) patient- and system-level determinants of the receipt of robotic as compared to open or laparoscopic surgery.

METHODS

This population-based retrospective cohort study included adult patients who, between 2009 and 2018 in Ontario, Canada, underwent one of four commonly performed robotic procedures: radical prostatectomy, total hysterectomy, thoracic lobectomy, partial nephrectomy. Patients were categorized based on the surgical approach as robotic, open, or laparoscopic for each procedure. Multivariable regression models were used to estimate the temporal trend in robotic surgery use and associations of patient and system characteristics with the surgical approach.

RESULTS

The cohort included 24,741 radical prostatectomy, 75,473 total hysterectomy, 18,252 thoracic lobectomy, and 4608 partial nephrectomy patients, of which 6.21% were robotic. After adjusting for patient and system characteristics, the rate of robotic surgery increased by 24% annually (RR 1.24, 95%CI 1.13-1.35): 13% (RR 1.13, 95%CI 1.11-1.16) for robotic radical prostatectomy, 9% (RR 1.09, 95%CI 1.05-1.13) for robotic total hysterectomy, 26% (RR 1.26, 95%CI 1.06-1.50) for thoracic lobectomy and 26% (RR 1.26, 95%CI 1.13-1.40) for partial nephrectomy. Lower comorbidity burden, earlier disease stage (among cancer cases), and early career surgeons with high case volume at a teaching hospital were consistently associated with the receipt of robotic surgery.

CONCLUSION

The use of robotic surgery has increased. The study of the real-world clinical outcomes and associated costs is needed before further expanding use among additional providers and hospitals.

Critically Appraising the Quality of Reporting of American College of Surgeons TQIP Studies in the Era of Large Data Research

BACKGROUND

The American College of Surgeons-Trauma Quality Improvement Program (ACS-TQIP) database is one of the most widely used databases for trauma research. We aimed to critically appraise the quality of the methodological reporting of ACS-TQIP studies.

STUDY DESIGN

The ACS-TQIP bibliography was queried for all studies published between January 2018 and January 2021. The quality of data reporting was assessed using the Strengthening the Reporting of Observational studies in Epidemiology-Reporting of Studies Conducted Using Observational Routinely Collected Health Data (STROBE-RECORD) statement and the JAMA Surgery checklist. Three items from each tool were not applicable and thus excluded. The quality of reporting was compared between high- and low-impact factor (IF) journals (cutoff for high IF is >90th percentile of all surgical journals).

RESULTS

A total of 118 eligible studies were included; 12 (10%) were published in high-IF journals. The median (interquartile range) number of criteria fulfilled was 5 (4-6) for the STROBE-RECORD statement (of 10 items) and 5 (5-6) for the JAMA Surgery checklist (of 7 items). Specifically, 73% of studies did not describe the patient population selection process, 61% did not address data cleaning or the implications of missing values, and 76% did not properly state inclusion/exclusion criteria and/or outcome variables. Studies published in high-IF journals had remarkably higher quality of reporting than those in low-IF journals.

CONCLUSION

The methodological reporting quality of ACS-TQIP studies remains suboptimal. Future efforts should focus on improving adherence to standard reporting guidelines to mitigate potential bias and improve the reproducibility of published studies.

History of the Department of Surgery at the University of Toronto: celebrating a centennial of progress and innovation

Now in its centennial year since inauguration, the Department of Surgery at the University of Toronto lays claim to more than 500 faculty, 270 residents, and 250 clinical fellows. There are 7 direct entry residency training programs, and 4 subspecialty programs accredited by the Royal College of Physicians and Surgeons of Canada. There have been 10 chairs of the department since 1921. This article chronicles the life and times of the previous chairs in sequence; the success of the department originates from its many talented and luminary surgeons who have innovated and shaped their fields of surgery. In recent years, the department’s academic productivity has been characterized by more than 1400 peer-reviewed publications per year, and annual research grant capture in excess of $90 million. Since the time of William Gallie, surgical trainees have been enabled to develop careers in surgery and science through the Gallie Program and, more recently, the Surgeon Scientist Training Program (SSTP) to attain higher graduate degrees. Providing quaternary surgical care at multiple hospital sites in Toronto, the Department of Surgery takes great pride in its robust clinical fellowship programs across all specialties that continue to attract trainees from around the world.

Review of the requirements for effective mass casualty preparedness for trauma systems. A disaster waiting to happen?

Mass casualty incidents (MCIs) are diverse, unpredictable, and increasing in frequency, but preparation is possible and necessary. The nature of MCIs requires a trauma response but also requires effective and tested disaster preparedness planning. From an international perspective, the aims of this narrative review are to describe the key components necessary for optimisation of trauma system preparedness for MCIs, whether trauma systems and centres meet these components and areas for improvement of trauma system response. Many of the principles necessary for response to MCIs are embedded in trauma system design and trauma centre function. These include robust communication networks, established triage systems, and capacity to secure centres from threats to safety and quality of care. However, evidence from the current literature indicates the need to strengthen trauma system preparedness for MCIs through greater trauma leader representation at all levels of disaster preparedness planning, enhanced training of staff and simulated disaster training, expanded surge capacity planning, improved staff management and support during the MCI and in the post-disaster recovery phase, clear provision for the treatment of paediatric patients in disaster plans, and diversified and pre-agreed systems for essential supplies and services continuity. Mass casualty preparedness is a complex, iterative process that requires an integrated, multidisciplinary, and tiered approach. Through effective preparedness planning, trauma systems should be well-placed to deliver an optimal response when faced with MCIs.

Evaluating the Tactical Combat Casualty Care principles in civilian and military settings: systematic review, knowledge gap analysis and recommendations for future research

OBJECTIVES

The Tactical Combat Casualty Care (TCCC) guidelines detail resuscitation practices in prehospital and austere environments. We sought to review the content and quality of the current TCCC and civilian prehospital literature and characterize knowledge gaps to offer recommendations for future research.

METHODS

MEDLINE, EMBASE, CINAHL, and Cochrane Central Register of Controlled Trials were searched for studies assessing intervention techniques and devices used in civilian and military prehospital settings that could be applied to TCCC guidelines. Screening and data extraction were performed according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Quality appraisal was conducted using appropriate tools.

RESULTS

Ninety-two percent (n=57) of studies were observational. Most randomized trials had low risk of bias, whereas observational studies had higher risk of bias. Interventions of massive hemorrhage control (n=17) were wound dressings and tourniquets, suggesting effective hemodynamic control. Airway management interventions (n=7) had high success rates with improved outcomes. Interventions of respiratory management (n=12) reported low success with needle decompression. Studies assessing circulation (n=18) had higher quality of evidence and suggested improved outcomes with component hemostatic therapy. Hypothermia prevention interventions (n=2) were generally effective. Other studies identified assessed the use of extended focused assessment with sonography in trauma (n=3) and mixed interventions (n=2).

CONCLUSIONS

The evidence was largely non-randomized with heterogeneous populations, interventions, and outcomes, precluding robust conclusions in most subjects addressed in the review. Knowledge gaps identified included the use of blood products and concentrate of clotting factors in the prehospital setting.

LEVEL OF EVIDENCE

Systematic review, level III.

Antithrombotic choice in blunt cerebrovascular injuries: Experience at a tertiary trauma center, systematic review, and meta-analysis

BACKGROUND

Blunt cerebrovascular injuries (BCVIs) may occur following trauma and lead to ischemic stroke if untreated. Antithrombotic therapy decreases this risk; however, the optimal agent has yet to be determined in this population. The aim of this study was to compare the risk-benefit profile of antiplatelet (AP) versus anticoagulant (AC) therapy in rates of ischemic stroke and hemorrhagic complications in BCVI patients.

METHODS

We performed a retrospective review of BCVI patients at our tertiary care Trauma hospital from 2010 to 2015, and a systematic review and meta-analysis of the literature. The OVID Medline, Embase, Web of Science, and Cochrane Library databases were searched from inception to September 16, 2019. References of included publications were searched manually for other relevant articles. The search was limited to articles in humans, in patients 18 years or older, and in English. Studies that reported treatment-stratified clinical outcomes following AP or AC treatment in BCVI patients were included. Exclusion criteria included case reports, case series with n < 5, review articles, conference abstracts, animal studies, and non-peer-reviewed publications. Data were extracted from each study independently by two reviewers, including study design, country of origin, sex and age of patients, Injury Severity Score, Biffl grade, type of treatment, ischemic stroke rate, and hemorrhage rate. Pooled estimates using odds ratio (OR) were combined using a random-effects model using a Mantel-Hanzel weighting. The main outcome of interest was rate of ischemic stroke due to BCVI, and the secondary outcome was hemorrhage rate based on AC or AP treatment.

RESULTS

In total, there were 2044 BCVI patients, as reported in the 22 studies in combination with our institutional data. The stroke rate was not significantly different between the two treatment groups (OR, 1.27; 95% confidence interval, 0.40-3.99); however, the hemorrhage rate was decreased in AP versus AC treated groups (OR, 0.38; 95% confidence interval, 0.15-1.00).

CONCLUSION

Based on this meta-analysis, both AC and AP seem similarly effective in preventing ischemic stroke, but AP is better tolerated in the trauma population. This suggests that AP therapy may be preferred, but this should be further assessed with prospective randomized trials.

LEVEL OF EVIDENCE

Review article, level II.

Committed to Better Outcomes: Reducing Infection after Surgery Across the Ontario Surgical Quality Improvement Network

BACKGROUND

In 2015, the Ontario Surgical Quality Improvement Network was established to create a community of practice for Ontario hospitals to improve surgical quality. A provincial campaign to decrease postsurgical infections was launched in 2017.

STUDY DESIGN

Thirty hospitals implemented activities related to the campaign from April 2018 to March 2019. The community of practice was used to disseminate suggested change ideas in each area. Self-reported data from participating hospitals and collaborative-wide aggregate risk-adjusted data from the American College of Surgeons NSQIP were reviewed to determine the impact of the campaign on the rates of postoperative surgical site infections (SSIs), urinary tract infections (UTIs), and pneumonia.

RESULTS

A total of 24, 8, and 2 hospitals selected SSIs, UTIs, and pneumonia, respectively, as their targets for improvement. Three hospitals selected both SSIs and UTIs, 1 hospital selected SSIs and pneumonia, and 1 hospital selected all 3 indicators as targets. Self-reported data demonstrated that the rates of SSIs and UTIs decreased significantly post campaign from 4.87% to 3.99% (p < 0.0001) and from 3.65% to 1.25% (p = 0.007), respectively. Pneumonia rates also decreased from 1.27% to 1.05%. Overall rates of SSIs, UTIs, and pneumonia across all Ontario Surgical Quality Improvement Network hospitals were reduced from 3.4%, 1.29%, and 0.88% to 3.37%, 1.14%, and 0.84%, respectively.

CONCLUSIONS

The 1-year campaign resulted in a clinically significant reduction in the rates of SSIs and UTIs, as well as a trend for decrease in pneumonia incidence among participating hospitals. Using a flexible approach with priority setting and leveraging the community of practice for dissemination of change ideas is an effective way of sustaining quality improvement activities.

Freeze-dried plasma for major trauma - Systematic review and meta-analysis

BACKGROUND

Treatment of acute trauma coagulopathy has shifted toward rapid replacement of coagulation factors with frozen plasma (FP). There are logistic difficulties in providing FP. Freeze-dried plasma (FDP) may have logistical advantages including easier storage and rapid preparation time. This review assesses the feasibility, efficacy, and safety of FDP in trauma.

STUDY DESIGN AND METHODS

Studies were searched from Medline, Embase, Cochrane Controlled Trials Register, ClinicalTrials.gov, and Google Scholar. Observational and randomized controlled trials (RCTs) assessing FDP use in trauma were included. Trauma animal models addressing FDP use were also included. Bias was assessed using validated tools. Primary outcome was efficacy, and secondary outcomes were feasibility and safety. Meta-analyses were conducted using random-effect models. Evidence was graded using Grading of Recommendations Assessment, Development, and Evaluation profile.

RESULTS

Twelve human studies (RCT, 1; observational, 11) and 15 animal studies were included. Overall, studies demonstrated moderate risk of bias. Data from two studies (n = 119) were combined for meta-analyses for mortality and transfusion of allogeneic blood products (ABPs). For both outcomes, no difference was identified. For mortality, pooled odds ratio was 0.66 (95% confidence interval, 0.29-1.49), with I2 = 0%. Use of FDP is feasible, and no adverse events were reported. Animal data suggest similar results for coagulation and anti-inflammatory profiles for FP and FDP.

CONCLUSION

Human data assessing FDP use in trauma report no difference in mortality and transfusion of ABPs in patients receiving FDP compared with FP. Data from animal trauma studies report no difference in coagulation factor and anti-inflammatory profiles between FP and FDP. Results should be interpreted with caution because most studies were observational and have heterogeneous population (military and civilian trauma) and a moderate risk of bias. Well-designed prospective observational studies or, preferentially, RCTs are warranted to answer FDP's effect on laboratory (coagulation factor levels), transfusion (number of ABPs), and clinical outcomes (organ dysfunction, length of stay, and mortality).

LEVEL OF EVIDENCE

Systematic review and meta-analysis, level IV.

Survey of major trauma centre preparedness for mass casualty incidents in Australia, Canada, England and New Zealand

BACKGROUND

Mass casualty incidents (MCIs) are increasing. Trauma centres play a key role in MCIs due to their readiness and expansive multidisciplinary expertise for injury management. Previous studies have shown deficiencies in trauma centre disaster preparedness. The aim of this study was to describe the current disaster preparedness of Major Trauma Centres (MTCs) in Australia, Canada, England and New Zealand.

METHODS

A cross-sectional survey of all (n = 82) MTCs was undertaken. The anonymous survey collected data about disaster preparedness in nine key areas. Respondents were encouraged to consult appropriately at their centre to provide an accurate representation of their centre's preparedness.

FINDINGS

Responses were received from 69 (84%) centres; 61 completed all questions. 91% had a disaster preparedness committee and 80% had an all-hazards emergency plan. 79% had held an MCI drill in the past 2 years. 54% reported a system in place to calculate maximum capacity, but testing of surge capacity was uncommon. 55% reported the presence of stored resources for an MCI and 58% had a database of staff trained in Emergency Management. 74% had a training and education plan available for staff involved in an MCI and a plan for professional debriefing of staff post-MCI, while 62% had a post-disaster employee assistance programme. Most centres had appropriate back-up communication, safety and security plans.

INTERPRETATION

The disaster preparedness of MTCs was high for communication, safety and security but there was clear need for improvement in other areas including surge capacity, human resources and post-disaster recovery.

Knee dislocation and associated injuries: an analysis of the American College of Surgeons National Trauma Data Bank

PURPOSE

Knee dislocations (KDs) are potentially devastating injuries, leading to loss of function or limb in often young patients. This retrospective database review aims to determine the relative incidence and risk factors for KDs presenting to North American Level I and II trauma centers.

METHODS

The National Trauma Data Bank (NTDB) was retrospectively interrogated using ICD-9-CM codes to identify KDs between 2010 and 2014 to derive KD incidence. KDs were stratified by age, sex, Injury Severity Score (ISS), Glasgow Coma Scale (GCS), drug and alcohol use, injury mechanism, open vs. closed KD, vascular injury and fracture. Each co-variate was tested against different mechanisms of injury, using Chi-squared tests and risk adjusted analyses to derive risk factors for KD. The same calculations were done for secondary outcomes (vascular and neurological injuries, compartment syndrome, amputation, and mortality).

RESULTS

A total of 6454 KDs met the inclusion criteria (18/10,000 admissions). KDs occurred most commonly amongst men, aged 20-39, with an ISS score 1-14 and following motor vehicle collision (MVC). A vascular investigation was performed in 29%, with injury documented in 15% of KDs and 10.8% receiving a vascular procedure. Associated fractures were observed in 41.4% of KDs. Open injuries in 13.6%. Neurological injury documented in 6.2%, compartment syndrome in 2.7%, amputation in 3.8% (> 50% had vascular injury) and 2.8% died. MVC was the most common mechanism of injury (p < 0.001), significantly more common in young, male patients, associated with higher ISS and lower GCS, especially when drugs or alcohol were involved (p < 0.0001). Being male, having a vascular injury or open KD were all risk factors for compartment syndrome, amputation and neurological injuries.

CONCLUSIONS

KDs are rare injuries, but their relative incidence may be increasing. Young, male patients involved in MVCs are risk factors for KDs and their associated injuries, such as neurological injuries, amputations and compartment syndrome. Vascular injury occurs at a frequency of around 15%. The findings of the current study may guide future research and help to inform clinicians on the expected rates of associated injuries in patients identified to have KD in a trauma center population. It informs regarding risk factors for KD, which may improve diagnosis rates of spontaneously reduced knee dislocations by increasing index of suspicion in high-risk patients and identifies specific links with impaired driving.

LEVEL OF EVIDENCE

IV.

Causes of Delay During Interfacility Transports of Injured Patients Transported by Air Ambulance

Background: Many severely injured patients are initially brought to a non-trauma centers for initial assessment and stabilization. Air ambulance services are commonly used to expedite interfacility transport of injured patients to trauma centers. Little is known of the types of delays experienced during interfacility transports. The purpose of this study was to identify specific causes of modifiable delays and estimate the attributable time associated with each of these delays.Methods: This was a retrospective cohort study of injured patients undergoing interfacility transfer to a trauma center who were transported by a provincial air ambulance service between January 1, 2014 and December 31, 2016. Electronic patient care records were screened and then manually reviewed to identify causes of delay during the interfacility transport process. The attributable time for each of these delays was also estimated.Results: There were 932 injured patients emergently transported by air ambulance from a community hospital to a trauma center over the 3-year study period from which 458 unique causes of delay that were identified. The most frequent cause of delays to sending facility were refueling (38%), waiting for land emergency medical services escort (25%) and weather (12%). The most common in-hospital delays included waiting for documentation (32%), delay to intubate (15%), medically unstable patient (13%) and waiting for diagnostic imaging (12%). The most frequent delays to receiving/handover included waiting for land EMS escort (31%), trauma team not assembled (24%) and weather (17%). In-hospital delays with the longest average length of delay included chest tube insertion (53 minutes), intubation (49 minutes) and delays for diagnostic imaging (46 minutes).Conclusions: In conclusion, we identified numerous modifiable causes of delay during interfacility transport. Efforts to reduce these delays can be made at both the air ambulance and hospital levels.

Population-based study of the prevalence and management of self-reported high pain scores in patients with non-resected pancreatic adenocarcinoma

BACKGROUND

Pain is a common debilitating symptom in pancreatic adenocarcinoma. This cohort study examined the use of, and factors associated with, pain-directed interventions for a high pain score in patients with non-curable pancreatic adenocarcinoma.

METHODS

Administrative databases were linked and patients with non-resected pancreatic adenocarcinoma diagnosed between 2010 and 2016, who reported one or more Edmonton Symptom Assessment System (ESAS) score, were identified. A high pain score was defined as an ESAS score of at least 4. Outcomes were pain-directed interventions: opiates (in patients aged 65 years or more with universal drug coverage), nerve block and radiation therapy for a high pain score. Reduction in pain score of at least 1 point after pain-directed intervention was also evaluated. Modified Poisson regression was used to examine factors associated with pain-directed intervention.

RESULTS

Among 2623 patients with a median age of 67 years, 1223 (46·6 per cent) were women, and 1621 (61·8 per cent) reported a high pain score at a median of 38 days after diagnosis. Of those with a high pain score, 75·6 per cent (688 of 910) received opiates, 13·5 per cent (219 of 1621) radiation and 1·2 per cent (19 of 1621) nerve block. The pain score decreased in 62·1 per cent of patients after administration of opiates, 73·4 per cent after radiation and all patients after nerve block. In multivariable analysis, no patient factor (age, sex, co-morbidity burden, rurality, income quintile) was associated with receipt of non-opiate pain-directed intervention for a high pain score. In patients aged at least 65 years, advanced age was associated with lower odds of opiate use.

CONCLUSION

Opiates are the most common pain-directed intervention for non-curable pancreatic adenocarcinoma, whereas radiation therapy and nerve blocks are seldom used. The lack of association between pain-directed interventions and patient factors points toward practice-driven patterns.

Mental health outcomes after major trauma in Ontario: a population-based analysis

BACKGROUND

Major injury continues to be a common source of morbidity and mortality; improving the functional recovery of survivors of major trauma requires a better understanding of the mental health outcomes that may occur in this population. We assessed the association between major trauma and the development of a new mental health diagnosis or death by suicide.

METHODS

We completed a population-based, self-controlled, longitudinal cohort analysis using linked administrative data on patients treated for major trauma in Ontario between 2005 and 2010. All survivors were included and composite rates of mental health diagnoses during inpatient admissions were compared between the 5 years after injury and the 5 years before injury, using Poisson regression with generalized estimating equations. The incidence of suicide was calculated for the 5 years after injury. Risk factors for suicide were calculated using Cox proportional hazard regression analyses.

RESULTS

The analysis included 19 338 patients, predominantly men (70.7%) from urban areas (82.6%), with unintentional (89%), blunt injuries (93.4%). Overall, trauma was associated with a 40% increase in the postinjury rate of mental health diagnoses (incidence rate ratio [IRR] 1.4, 95% [confidence interval] CI 1.1 to 1.8). The suicide rate was 70 per 100 000 patients per year, substantially higher than the population average. Risk factors for completing suicide were prior inpatient diagnosis of mood disorder (hazard ratio [HR] 4.3, 95% CI 2.1 to 8.8) and self-inflicted injury (HR 7.8, 95% CI 3.9 to 15.4).

INTERPRETATION

Survivors of major trauma are at a heightened risk of developing mental health conditions or death by suicide in the years after their injury. Patients with pre-existing mental health disorders or who are recovering from a self-inflicted injury are at particularly high risk.

Risk Factors for Non-optimal Resource Utilization for Emergent Interfacility Transfers by Air Ambulance in Ontario

Background: The use of air ambulance to facilitate interfacility transfer has been associated with improved mortality; however, air ambulance is a limited resource and sometimes the optimal resource to transport a patient is unavailable. When a non-optimal resource is used there is an inherent delay and critically unwell patients may deteriorate as a result. This study aimed to identify risk factors associated with non-optimal resource utilization for adult patients undergoing emergent interfacility transport by air ambulance in Ontario, Canada. A secondary objective was to determine if non-optimal resource utilization was associated with deterioration in clinical status by measuring a delta rapid emergency medicine score (REMS). Methods: This was a retrospective cohort study of all emergent, adult interfacility transfers transported by air ambulance over a 5-year period in Ontario, Canada. Determination of optimal resource use was based on distances and historic time data for all sending-receiving facility pairs. A logistic regression model was used to explore patient, provider and institutional risk factors for non-optimal resource use. To explore the secondary objective a linear regression model was used to explore impact of non-optimal resource use on deltaREMS. Results: There were a total of 9,687 patients included in the study cohort, with 4,984 having an optimal resource use and 4,703 having non-optimal resource. The median delay in interfacility transfer caused by a non-optimal transfer strategy was 35.7 minutes. Patients who required mechanical ventilation (OR 1.13, p = 0.031) and or were transferred out of nursing stations had higher odds of non-optimal resource use (OR 2.84, p = 0.019). Paramedic level of care of advanced (OR 0.37, p = < 0.001) and critical care (OR 0.28, p = < 0.001) as well as spring season (OR 0.75, p = < 0.001) had lower odds of non-optimal resource utilization. Optimal resource utilization did not significantly affect delta REMS (beta coefficient 0.002, p = 0.64). Conclusions: Patients who required mechanical ventilation and were transferred out from a nursing station had higher odds of non-optimal resource utilization while patients that required advanced or critical care level of care and spring season had lower odds of non-optimal resource use. Additionally, non-optimal resource use for air ambulance interfacility transfers did not result in patient deterioration as measured by a delta REMS score.

Use of performance reports among trauma medical directors and programme managers in the American College of Surgeons' Trauma Quality Improvement Program: a qualitative analysis

BACKGROUND

The American College of Surgeons' Trauma Quality Improvement Program (TQIP) provides trauma centres with performance reports on their processes and outcomes of care relative to their peers. This study explored how performance reports are used by trauma centre leaders to engage in performance improvement and perceived barriers to use.

STUDY DESIGN

Qualitative focus group study with trauma medical directors (TMDs) and trauma programme managers (TPMs) in US trauma centres. Consistent with qualitative descriptive analysis, data collection and interpretation were inductively and iteratively completed. Major themes were derived using a constant comparative technique.

RESULTS

Six focus groups were conducted involving 22 TMDs and 22 TPMs. Three major themes were captured: (1) technical uses of performance reports; (2) cultural uses of performance reports; (3) opportunities to enhance the role and value of TQIP. First, technical uses included using reports to assess data collection procedures, data quality and areas of poor performance relative to peers. In this domain, barriers to report use included not trusting others' data quality and challenges with report interpretation. Second, reports were used to influence practice change by fostering inter-specialty discussions, leveraging resources for quality improvement, community engagement and regional collaboratives. Perceived lack of specialist engagement was viewed as an impediment in this domain. Lastly, identified opportunities for TQIP to support report use involved clarifying the relationship between verification and performance reports, and increasing partnerships with nursing associations.

CONCLUSION

Trauma centre improvement leaders indicated practical and social uses of performance reports that can affect intention and ability to change. Recommendations to optimise programme participation include a focus on data quality, adequate resource provision and enhanced support for regional collaboratives.

Accelerating surgical quality improvement in Ontario through a regional collaborative: a quality-improvement study

BACKGROUND

The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) collaborative in Ontario, the Ontario Surgical Quality Improvement Network (ON-SQIN), was launched in January 2015. We describe its approaches to support surgical quality improvement and examine its early impact on member hospitals.

METHODS

All Ontario hospitals that participated in the ON-SQIN and NSQIP were included in this quality-improvement study. The primary intervention was the introduction of the ON-SQIN, and the secondary interventions included a community of practice and access to quality-improvement resources and tools. Outcome measures included the level of quality-improvement capacity, collaborative-wide aggregate data on postoperative complications, and self-reported rates of surgical site and urinary tract infections.

RESULTS

Eighteen hospitals that enrolled in the ON-SQIN in 2015 reported an increase in their capacity for quality improvement after 18 months. Analysis of the collaborative-wide aggregate data in a 6-month period (14 748 surgical cases) revealed a substantial reduction of acute renal failure (relative risk 0.48, 95% confidence interval 0.25-0.95) and urinary tract infection (relative risk 0.77, 95% confidence interval 0.61-0.97). Most hospitals that targeted prevention of surgical site infection and urinary tract infection reported reduction of these occurrences during a 1-year period.

INTERPRETATION

The ON-SQIN supported the uptake of the NSQIP in Ontario hospitals and promoted targeted surgical quality-improvement initiatives, resulting in increased quality-improvement capacity and development of the community of practice. Furthermore, our early experience suggests that improvements in surgical care are being realized.

Canceled to Be Called Back: A Retrospective Cohort Study of Canceled Helicopter Emergency Medical Service Scene Calls That Are Later Transferred to a Trauma Center

INTRODUCTION

In our trauma system, helicopter emergency medical services (HEMS) can be requested to attend a scene call for an injured patient before arrival by land paramedics. Land paramedics can cancel this response if they deem it unnecessary. The purpose of this study is to describe the frequency of canceled HEMS scene calls that were subsequently transferred to 2 trauma centers and to assess for any impact on morbidity and mortality.

METHODS

Probabilistic matching was used to identify canceled HEMS scene call patients who were later transported to 2 trauma centers over a 48-month period. Registry data were used to compare canceled scene call patients with direct from scene patients.

RESULTS

There were 290 requests for HEMS scene calls, of which 35.2% were canceled. Of those canceled, 24.5% were later transported to our trauma centers. Canceled scene call patients were more likely to be older and to be discharged home from the trauma center without being admitted.

CONCLUSION

There is a significant amount of undertriage of patients for whom an HEMS response was canceled and later transported to a trauma center. These patients face similar morbidity and mortality as patients who are brought directly from scene to a trauma center.

Systematic Review of Human and Animal Studies Examining the Efficacy and Safety of N-Acetylcysteine (NAC) and N-Acetylcysteine Amide (NACA) in Traumatic Brain Injury: Impact on Neurofunctional Outcome and Biomarkers of Oxidative Stress and Inflammation

Background

No new therapies for traumatic brain injury (TBI) have been officially translated into current practice. At the tissue and cellular level, both inflammatory and oxidative processes may be exacerbated post-injury and contribute to further brain damage. N-acetylcysteine (NAC) has the potential to downregulate both processes. This review focuses on the potential neuroprotective utility of NAC and N-acetylcysteine amide (NACA) post-TBI.

Methods

Medline, Embase, Cochrane Library, and ClinicalTrials.gov were searched up to July 2017. Studies that examined clinical and laboratory effects of NAC and NACA post-TBI in human and animal studies were included. Risk of bias was assessed in human and animal studies according to the design of each study (randomized or not). The primary outcome assessed was the effect of NAC/NACA treatment on functional outcome, while secondary outcomes included the impact on biomarkers of inflammation and oxidation. Due to the clinical and methodological heterogeneity observed across studies, no meta-analyses were conducted.

Results

Our analyses revealed only three human trials, including two randomized controlled trials (RCTs) and 20 animal studies conducted using standardized animal models of brain injury. The two RCTs reported improvement in the functional outcome post-NAC/NACA administration. Overall, the evidence from animal studies is more robust and demonstrated substantial improvement of cognition and psychomotor performance following NAC/NACA use. Animal studies also reported significantly more cortical sparing, reduced apoptosis, and lower levels of biomarkers of inflammation and oxidative stress. No safety concerns were reported in any of the studies included in this analysis.

Conclusion

Evidence from the animal literature demonstrates a robust association for the prophylactic application of NAC and NACA post-TBI with improved neurofunctional outcomes and downregulation of inflammatory and oxidative stress markers at the tissue level. While a growing body of scientific literature suggests putative beneficial effects of NAC/NACA treatment for TBI, the lack of well-designed and controlled clinical investigations, evaluating therapeutic outcomes, prognostic biomarkers, and safety profiles, limits definitive interpretation and recommendations for its application in humans at this time.

Toward late career transitioning: a proposal for academic surgeons

SUMMARY

In the absence of a defined retirement age, academic surgeons need to develop plans for transition as they approach the end of their academic surgical careers. The development of a plan for late career transition represents an opportunity for departments of surgery across Canada to initiate a constructive process in cooperation with the key stakeholders in the hospital or institution. The goal of the process is to develop an individual plan for each faculty member that is agreeable to the academic surgeon; informs the surgical leadership; and allows the late career surgeon, the hospital, the division and the department to make plans for the future. In this commentary, the literature on the science of aging is reviewed as it pertains to surgeons, and guidelines for late career transition planning are shared. It is hoped that these guidelines will be of some value to academic programs and surgeons across the country as late career transition models are developed and adopted.

A retrospective cohort study of the relationship between quality indicator measurement and patient outcomes in adult trauma centers in the United States

BACKGROUND

Improving care is a key strategy for reducing the burden of injuries, but it is unknown whether the use of quality indicators (QI) is associated with patient outcomes. We sought to evaluate the association between the use of QIs by trauma centers and outcomes in adult injury patients.

METHODS

We identified consecutive adult patients (n=223,015) admitted to 233 verified trauma centers January 1, 2007 to December 31, 2010 that contributed data to the National Trauma Data Bank and participated in a survey of QI practices. Generalized Linear Mixed Models were employed to evaluate the association between the intensity (number of QIs) and nature (report cards, internal and external benchmarking) of QI use and survival to hospital discharge, adjusting for patient and hospital characteristics.

RESULTS

There were no significant differences in the odds of survival to trauma center discharge according to the number of QIs measured (quartiles; odds ratio{OR} [95% confidence interval{CI}] 1.00 vs. 1.08 [0.90-1.31] vs. 1.00 [0.82-1.22] vs. 1.21 [0.99-1.49]), or whether centers used reports cards (OR 1.07, 95%CI 0.94-1.23), internal (OR 1.06, 95%CI 0.89-1.26) or external (OR 1.09, 95%CI 0.92-1.31) benchmarking. The duration (geometric mean) of mechanical ventilation (4.0days), ICU stay (4.6days), hospital stay (7.7days) and proportion of patients with a complication (13.6%) did not significantly differ according to the intensity or nature of QI use.

CONCLUSIONS

The intensity and nature of the QIs used by trauma centers was not associated with outcomes of patient care. Alternative quality improvement strategies may be needed.

Early prediction of outcome after severe traumatic brain injury: a simple and practical model

Background

Traumatic brain injury (TBI) is a heterogeneous syndrome with a broad range of outcome. We developed a simple model for long-term outcome prognostication after severe TBI.

Methods

Secondary data analysis of a large multicenter randomized trial. Patients were grouped according to 6-month extended Glasgow outcome scale (eGOS): poor-outcome (eGOS ≤ 4; severe disability or death) and acceptable outcome (eGOS > 4; no or moderate disability). A prediction decision tree was built using binary recursive partitioning to predict poor or acceptable 6-month outcome. Comparison to two previously published and validated models was made.

Results

The decision tree included the predictors of head Abbreviated Injury Scale (AIS) severity, the Marshall computed tomography score, and pupillary reactivity. All patients with a head AIS severity of 5 were predicted to have a poor outcome. In patients with head AIS severity < 5, the model predicted an acceptable outcome for (1) those with Marshall score of 1, and (2) those with Marshall score above 1 but with reactive pupils at admission. The decision tree had a sensitivity of 72.3 % (95 % CI: 66.4–77.6 %) and specificity of 62.5 % (95 % CI: 54.9–69.6 %). The proportion correctly classified for the comparison models was similar to our model. Our model was more apt at correctly classifying those with poor outcome but more likely to misclassify those with acceptable outcome than the comparison models.

Conclusion

Predicting long-term outcome early after TBI remains challenging and inexact. This model could be useful for research and quality improvement studies to provide an early assessment of injury severity, but is not sufficiently accurate to guide decision-making in the clinical setting.

Creating a Powerful Platform to Explore Health in a Correctional Population: A Record Linkage Study

We used record linkage to create a data repository of health information of persons who were federally incarcerated in Ontario and Canada. We obtained records from 56,867 adults who were federally incarcerated between January 1, 1998 and December 31, 2011 from the Correctional Service of Canada; 15,248 records belonged to individuals residing in Ontario, Canada. We linked these records to the Registered Persons Database (RPDB) which contained records from 18,116,996 individuals eligible for health care in Ontario. Out of 56,867 OMS records, 22,844 (40.2%) were linked to the RPDB. Looking only at those incarcerated in Ontario, 98%, (14 953 of 15248) records were linked to RPDB. Most records of persons in Ontario-based facilities were linked deterministically. Linkage rates were lower for women, minority groups, and substance users. In conclusion, record linkage enabled the creation of a valuable data repository: there are no electronic medical records for correctional populations in Canada, making it more difficult to profile their health.

Clinical practice guideline: management of acute pancreatitis

ABSTRACT

There has been an increase in the incidence of acute pancreatitis reported worldwide. Despite improvements in access to care, imaging and interventional techniques, acute pancreatitis continues to be associated with significant morbidity and mortality. Despite the availability of clinical practice guidelines for the management of acute pancreatitis, recent studies auditing the clinical management of the condition have shown important areas of noncompliance with evidence-based recommendations. This underscores the importance of creating understandable and implementable recommendations for the diagnosis and management of acute pancreatitis. The purpose of the present guideline is to provide evidence-based recommendations for the management of both mild and severe acute pancreatitis as well as the management of complications of acute pancreatitis and of gall stone-induced pancreatitis.

Effects of closure of an urban level I trauma centre on adjacent hospitals and local injury mortality: a retrospective, observational study

OBJECTIVE

To determine the association of the Martin Luther King Jr Hospital (MLK) closure on the distribution of admissions on adjacent trauma centres, and injury mortality rates in these centres and within the county.

DESIGN

Observational, retrospective study.

SETTING

Non-public patient-level data from the state of California were obtained for all trauma patients from 1999 to 2009. Geospatial analysis was used to visualise the redistribution of trauma patients to other hospitals after MLK closed. Variance of observed to expected injury mortality using multivariate logistic regression was estimated for the study period.

PARTICIPANTS

A total of 37 131 trauma patients were admitted to the five major south Los Angeles trauma centres from the MLK service area between 1999 and 2009.

MAIN OUTCOME MEASURES

(1) Number and type of trauma admissions to trauma centres in closest proximity to MLK; (2) inhospital injury mortality of trauma patients after the trauma centre closure.

RESULTS

During and after the MLK closure, trauma admissions increased at three of the four nearby hospitals, particularly admissions for gunshot wounds (GSWs). This redistribution of patient load was accompanied by a dramatic change in the payer mix for surrounding hospitals; one hospital's share of uninsured more than tripled from 12.9% in 1999 to 44.6% by 2009. Overall trauma mortality did not significantly change, but GSW mortality steadily and significantly increased after the closure from 5.0% in 2007 to 7.5% in 2009.

CONCLUSIONS

Though local hospitals experienced a dramatic increase in trauma patient volume, overall mortality for trauma patients did not significantly change after MLK closed.

A Geospatial Analysis of Severe Firearm Injuries Compared to Other Injury Mechanisms: Event Characteristics, Location, Timing, and Outcomes

OBJECTIVES

Relatively little is known about the context and location of firearm injury events. Using a prospective cohort of trauma patients, we describe and compare severe firearm injury events to other violent and nonviolent injury mechanisms regarding incident location, proximity to home, time of day, spatial clustering, and outcomes.

METHODS

This was a secondary analysis of a prospective cohort of injured children and adults with hypotension or Glasgow Coma Scale score ≤ 8, injured by one of four primary injury mechanisms (firearm, stabbing, assault, and motor vehicle collision [MVC]) who were transported by emergency medical services to a Level I or II trauma center in 10 regions of the United States and Canada from January 1, 2010, through June 30, 2011. We used descriptive statistics and geospatial analyses to compare the injury groups, distance from home, outcomes, and spatial clustering.

RESULTS

There were 2,079 persons available for analysis, including 506 (24.3%) firearm injuries, 297 (14.3%) stabbings, 339 (16.3%) assaults, and 950 (45.7%) MVCs. Firearm injuries resulted in the highest proportion of serious injuries (66.3%), early critical resources (75.3%), and in-hospital mortality (53.5%). Injury events occurring within 1 mile of a patient's home included 53.9% of stabbings, 49.2% of firearm events, 41.3% of assaults, and 20.0% of MVCs; the non-MVC events frequently occurred at home. While there was geospatial clustering, 94.4% of firearm events occurred outside of geographic clusters.

CONCLUSIONS

Severe firearm events tend to occur within a patient's own neighborhood, often at home, and generally outside of geospatial clusters. Public health efforts should focus on the home in all types of neighborhoods to reduce firearm violence.

Intracranial pressure monitoring among children with severe traumatic brain injury

OBJECT Well-designed studies linking intracranial pressure (ICP) monitoring with improved outcomes among children with severe traumatic brain injury (TBI) are lacking. The main objective of this study was to examine the relationship between ICP monitoring in children and in-hospital mortality following severe TBI. METHODS An observational study was conducted using data derived from 153 adult or mixed (adult and pediatric) trauma centers participating in the American College of Surgeons (ACS) Trauma Quality Improvement Program (TQIP) and 29 pediatric trauma centers participating in the pediatric pilot TQIP between 2010 and 2012. Random-intercept multilevel modeling was used to examine the association between ICP monitoring and in-hospital mortality among children with severe TBI ≤16 years of age after adjusting for important confounders. This association was evaluated at the patient level and at the hospital level. In a sensitivity analysis, this association was reexamined in a propensity-matched cohort. RESULTS A total of 1705 children with severe TBI were included in the study cohort. The overall in-hospital mortality was 14.3% of patients (n = 243), whereas the mortality of the 273 patients (16%) who underwent invasive ICP monitoring was 11% (n = 30). After adjusting for patient- and hospital-level characteristics, ICP monitoring was associated with lower in-hospital mortality (adjusted OR 0.50; 95% CI 0.30-0.85; p = 0.01). It is possible that patients who were managed with ICP monitoring were selected because of an anticipated favorable or unfavorable outcome. To further address this potential selection bias, the analysis was repeated with the hospital-specific rate of ICP monitoring use as the exposure. The adjusted OR for death of children treated at high ICP-use hospitals was 0.49 compared with those treated at low ICP-use hospitals (95% CI 0.31-0.78; p = 0.003). Variations in ICP monitoring use accounted for 15.9% of the interhospital variation in mortality among children with severe TBI. Similar results were obtained after analyzing the data using propensity score-matching methods. CONCLUSIONS In this observational study, ICP monitoring use was associated with lower hospital mortality at both the patient and hospital levels. However, the contribution of variable ICP monitoring rates to interhospital variation in pediatric TBI mortality was modest.