Travis P. Webb

Safety and Efficacy of Prophylactic Anticoagulation in Patients with Traumatic Brain Injury

BACKGROUND Patients with traumatic brain injury (TBI) are at high risk for venous thromboembolism (VTE), but physicians are cautious with chemical prophylaxis in these patients because of concern about exacerbating intracranial hemorrhage. We hypothesized that early use of chemical thromboprophylaxis would reduce VTE incidence without increasing intracranial hemorrhage. STUDY DESIGN Records of all patients admitted with a TBI to a Level I trauma center from 2006 to 2008 were reviewed. TBI was defined as intracranial hemorrhage, hematoma, contusion, or diffuse axonal injury with a head Abbreviated Injury Scale score >2. Patients were excluded if they were discharged or died within 72 hours of admission. Chemical prophylaxis was defined as subcutaneous or intravenous unfractionated heparin or low molecular weight heparin before any VTE diagnosis. Progression of TBI was defined by worsening CT findings. VTE was defined as deep venous thrombosis or pulmonary embolus confirmed by radiology reports. Primary outcomes were progression of hemorrhage and VTE events. RESULTS Eight hundred and twelve of the 1,258 patients admitted to the trauma center with a TBI met study criteria. Chemical thromboprophylaxis was given to 49.5% (n = 402). Mean head Abbreviated Injury Scale score was 3.4 in both groups. One hundred and sixty-nine patients started prophylaxis within 48 hours and 242 patients began within 72 hours. Patients receiving chemical prophylaxis had a lower incidence of VTE (1% versus 3%; p = 0.019). Although not statistically significant, they also had a lower rate of injury progression, 3% versus 6% (p = 0.055). CONCLUSIONS Use of chemical thromboprophylaxis in TBI patients with a stable or improved head CT after 24 hours substantially reduces the incidence of VTE and does not increase the risk of progression of intracranial hemorrhage.

Intraabdominal vascular injury: are we getting any better?

BACKGROUND Intraabdominal vascular injury (IAVI) as a result of penetrating and blunt trauma carries a high mortality rate. This study was performed to compare current mortality rates with a previously reported historic control. METHODS The experience at our institution from 1970 to 1981 was previously reported with an overall mortality rate of 32% in 112 patients with penetrating IAVI. In a retrospective analysis, this historic cohort was compared with 248 patients with penetrating and blunt IAVI during a 138-month interval ending in June 2007. RESULTS Overall mortality rate was 28.6%. The most commonly injured arteries were the iliac artery, aorta, and superior mesenteric artery. The most commonly injured veins were the inferior vena cava, iliac vein, and portal vein. Injury to the aorta, IVC, and portal vein had the highest mortality rates of 67.8%, 42.1%, and 66.6%, respectively. One hundred forty-four patients with one vessel injured had a mortality rate of 18.7%, whereas those with more than one vessel injured had a mortality rate of 48.7% (p < 0.001). A total of 46% of 117 patients in shock died compared with 9.6% of 104 patients not in shock (p < 0.001). Patients with a base deficit of less than -15 had a mortality rate of 72%, whereas those with a base deficit of 0 to -15 (p < 0.001) had a mortality rate of 18.9%. There was no difference in the overall mortality rate for penetrating trauma compared with the previous study. CONCLUSIONS Although over 20 years have passed, no significant changes have occurred in the mortality associated with IAVI. Patients presenting in shock with IAVI continue to have a high mortality rate.

Structured teaching versus experiential learning of palliative care for surgical residents

BACKGROUND Previous end-of-life and palliative care curricula for surgical residents have shown improved learner confidence, but have not measured cognitive knowledge or skill acquisition. METHODS A nonrandomized trial evaluated a structured palliative care curriculum for 7 postgraduate year 2 surgical residents (intervention group) compared with 6 postgraduate year 5 surgical residents (comparison group). Outcomes were measured using an 18-item knowledge test, a 20-minute objective structured clinical examination simulating an intensive care unit family conference, and a survey measuring self-confidence. RESULTS The mean knowledge test scores for the intervention group, both before and after undergoing the structured palliative care curriculum, were no different from the comparison group. There was also no difference in objective structured clinical examination scores between the 2 groups. The intervention group felt less comfortable managing pain, breaking bad news, or addressing ethical issues. CONCLUSIONS Junior surgical residents have similar palliative care knowledge to senior residents without a palliative care curriculum. After participating in a palliative care curriculum, they have simulated skills that are similar to chief residents. However, self-confidence is lower among junior residents despite undergoing a palliative care curriculum.

Characteristics of patient encounters that challenge medical students' provision of patient-centered care.

Background Medical educators need to effectively engage and teach medical students to provide patient-centered care (PCC). There is limited appreciation for the issues that clinical students identify as challenges in providing PCC. Method As part of a required half-day PCC workshop in 2007, medical students authored critical incident scenarios on patient encounters where PCC was difficult. The authors analyzed 131 scenarios using qualitative memo technique to identify features associated with these encounters. Categories and themes were identified using constant comparative methodology. Results Commonly cited PCC challenges were student’s/patient’s emotional responses (63%/44%), patient’s/family’s perception of the care plan (54%), conflicting expectations (35%), communication barriers (30%) and patient’s social circumstances (29%). Sixty-three percent of incidents identified PCC-appropriate responses to these challenges. Conclusions Student-authored critical incidents regarding difficult patient encounters can be analyzed to identify key features that students perceive as challenges to providing PCC and can inform curriculum development.

Surgery Residency Curriculum Examination Scores Predict Future American Board of Surgery In-Training Examination Performance

IMPORTANCE A protected block curriculum (PBC) with postcurriculum examinations for all surgical residents has been provided to assure coverage of core curricular topics. Biannual assessment of resident competency will soon be required by the Next Accreditation System. OBJECTIVE To identify opportunities for early medical knowledge assessment and interventions, we examined whether performance in postcurriculum multiple-choice examinations (PCEs) is predictive of performance in the American Board of Surgery In-Training Examination (ABSITE) and clinical service competency assessments. DESIGN Retrospective single-institutional education research study. SETTING Academic general surgery residency program. PARTICIPANTS A total of 49 surgical residents. INTERVENTION Data for PGY1 and PGY2 residents participating in the 2008 to 2012 PBC are included. Each resident completed 6 PCEs during each year. MAIN OUTCOME MEASURES The results of 6 examinations were correlated to percentage-correct ABSITE scores and clinical assessments based on the 6 Accreditation Council for Graduate Medical Education core competencies. Individual ABSITE performance was compared between PGY1 and PGY2. Statistical analysis included multivariate linear regression and bivariate Pearson correlations. RESULTS A total of 49 residents completed the PGY1 PBC and 36 completed the PGY2 curriculum. Linear regression analysis of percentage-correct ABSITE and PCE scores demonstrated a statistically significant correlation between the PGY1 PCE 1 score and the subsequent PGY1 ABSITE score (p = 0.037, β = 0.299). Similarly, the PGY2 PCE 1 score predicted performance in the PGY2 ABSITE (p = 0.015, β = 0.383). The ABSITE scores correlated between PGY1 and PGY2 with statistical significance, r = 0.675, p = 0.001. Performance on the 6 Accreditation Council for Graduate Medical Education core competencies correlated between PGY1 and PGY2, r = 0.729, p = 0.001, but did not correlate with PCE scores during either years. CONCLUSIONS AND RELEVANCE Within a mature PBC, early performance in a PGY1 and PGY2 PCE is predictive of performance in the respective ABSITE. This information can be used for formative assessment and early remediation of residents who are predicted to be at risk for poor performance in the ABSITE.

Assessing Competency in Practice-Based Learning: A Foundation for Milestones in Learning Portfolio Entries

BACKGROUND Graduate medical education is undergoing a dramatic shift toward competency-based assessment of learners. Competency assessment requires clear definitions of competency and validated assessment methods. The purpose of this study is to identify criteria used by surgical educators to judge competence in Practice-Based Learning and Improvement (PBL&I) as demonstrated in learning portfolios. METHODS A total of 6 surgical learning and instructional portfolio entries served as documents to be assessed by 3 senior surgical educators. These faculty members were asked to rate and then identify criteria used to assess PBL&I competency. Individual interviews and group discussions were conducted, recorded, and transcribed to serve as the study dataset. Analysis was performed using qualitative methodology to identify themes for the purpose of defining competence in PBL&I. The assessment themes derived are presented with narrative examples to describe the progression of competency. RESULTS The collaborative coding process resulted in identification of 7 themes associated with competency in PBL&I related to surgical learning and instructional portfolio entries: (1) self-awareness regarding effect of actions; (2) identification and thorough description of learning goals; (3) cases used as catalyst for reflection; (4) reconceptualization with appropriate use and critique of cited literature; (5) communication skills/completeness of entry template; (6) description of future behavioral change; and (7) engagement in process--identifies as personally relevant. CONCLUSIONS The identified themes are consistent with and complement other criteria emerging from reflective practice literature and experiential learning theory. This study provides a foundation for further development of a tool for assessing learner portfolios consistent with the Accreditation Council for Graduate Medical Educations Next Accreditation System requirements.

Establishing a Conceptual Framework for Handoffs Using Communication Theory

BACKGROUND A significant consequence of the 2003 Accreditation Council for Graduate Medical Education duty hour restrictions has been the dramatic increase in patient care handoffs. Ineffective handoffs have been identified as the third most common cause of medical error. However, research into health care handoffs lacks a unifying foundational structure. We sought to identify a conceptual framework that could be used to critically analyze handoffs. METHODS A scholarly review focusing on communication theory as a possible conceptual framework for handoffs was conducted. A PubMed search of published handoff research was also performed, and the literature was analyzed and matched to the most relevant theory for health care handoff models. RESULTS The Shannon-Weaver Linear Model of Communication was identified as the most appropriate conceptual framework for health care handoffs. The Linear Model describes communication as a linear process. A source encodes a message into a signal, the signal is sent through a channel, and the signal is decoded back into a message at the destination, all in the presence of internal and external noise. The Linear Model identifies 3 separate instances in handoff communication where error occurs: the transmitter (message encoding), channel, and receiver (signal decoding). CONCLUSIONS The Linear Model of Communication is a suitable conceptual framework for handoff research and provides a structured approach for describing handoff variables. We propose the Linear Model should be used as a foundation for further research into interventions to improve health care handoffs.

Diagnosis and management of abdominal vascular injuries

Intra-abdominal vascular trauma is an injury that continues to challenge the general surgeon and leads to high morbidity and mortality even at busy urban trauma centers. The majority of patients arrive in hemorrhagic shock; therefore, rapid evaluation, resuscitation, and transfer to the operating room are necessary to treat these patients. In the operating room, the surgeon must have a well-planned surgical approach to the variety of potential injuries encountered in the retroperitoneal space. This article reviews the incidence, diagnostic strategies, treatment principles, and expected outcomes when dealing with intra-abdominal vascular injury.

Evaluating handoffs in the context of a communication framework

Background. The implementation of mandated restrictions in resident duty hours has led to increased handoffs for patient care and thus more opportunities for errors during transitions of care. Much of the current handoff literature is empiric, with experts recommending the study of handoffs within an established framework. Methods. A prospective, single‐institution study was conducted evaluating the process of handoffs for the care of surgical patients in the context of a published communication framework. Evaluation tools for the source, receiver, and observer were developed to identify factors impacting the handoff process, and inter‐rater correlations were assessed. Data analysis was generated with Pearson/Spearman correlations and multivariate linear regressions. Rater consistency was assessed with intraclass correlations. Results. A total of 126 handoffs were observed. Evaluations were completed by 1 observer (N = 126), 2 observers (N = 23), 2 receivers (N = 39), 1 receiver (N = 82), and 1 source (N = 78). An average (±standard deviation) service handoff included 9.2 (±4.6) patients, lasted 9.1 (±5.4) minutes, and had 4.7 (±3.4) distractions recorded by the observer. The source and receiver(s) recognized distractions in >67% of handoffs, with the most common internal and external distractions being fatigue (60% of handoffs) and extraneous staff entering/exiting the room (31%), respectively. Teams with more patients spent less time per individual patient handoff (r = −0.298; P = .001). Statistically significant intraclass correlations (P ≤ .05) were moderate between observers (r ≥ 0.4) but not receivers (r < 0.4). Intraclass correlation values between different types of raters were inconsistent (P > .05). The quality of the handoff process was affected negatively by presence of active electronic devices (&bgr; = −0.565; P = .005), number of teaching discussions (&bgr; = −0.417; P = .048), and a sense of hierarchy between source and receiver (&bgr; = −0.309; P = .002). Conclusion. Studying the handoff process within an established framework highlights factors that impair communication. Internal and external distractions are common during handoffs and along with the working relationship between the source and receiver impact the quality of the handoff process. This information allows further study and targeted interventions of the handoff process to improve overall effectiveness and patient safety of the handoff.

Clinical significance of computed tomography contrast extravasation in blunt trauma patients with a pelvic fracture.

INTRODUCTION Blunt pelvic fractures can be associated with major pelvic bleeding. The significance of contrast extravasation (CE) on computed tomography (CT) is debated. We sought to update our experience with CE on CT scan for the years 2009–2014 to determine the accuracy of CE in predicting the need for angioembolization. METHODS This is a retrospective review of the trauma registry and our electronic medical record from a Level I trauma center. Patients seen from July 1, 2009, to September 7, 2014, with blunt pelvic fractures and contrast-enhanced CT were included. Standard demographic, clinical, and injury data were obtained. Patient records were queried for CE, performance of angiography, and angioembolization. Positive patients were those where CE was associated with active bleeding requiring angioembolization. All other patients were considered negative. RESULTS There were 497 patients during the study time period with blunt pelvic fracture meeting inclusion criteria, and 75 patients (15%) had CE. Of those patients with CE, 30 patients (40%) underwent angiography, and 17 patients (23%) required angioembolization. The sensitivity, specificity, positive predictive value, and negative predictive value of CE on CT were 100%, 87.9%, 22.7%, and 100%, respectively. Two patients without CE underwent angiography but did not undergo embolization. Patients with CE had higher mortality (13 vs. 6%, p < 0.05) despite not having higher ISS scores. CONCLUSIONS This study reinforces that CE on CT pelvis with blunt trauma is common, but many patients will not require angioembolization. The negative predictive value of 100% should be reassuring to trauma surgeons such that if a modern CT scanner is used, and there is no CE seen on CT, then the pelvis will not be a source of hemorrhagic shock. All of these findings are likely due to both increased comfort with observing CEs and the increased sensitivity of modern CT scanners. LEVEL OF EVIDENCE Therapeutic/care management study, level IV.