Kabir Yadav

The utility of base deficit and arterial lactate in differentiating major from minor injury in trauma patients with normal vital signs

OBJECTIVES Early recognition and treatment of hemorrhagic shock after trauma limits multi-organ failure and mortality. Traditional vital signs (VS) although specific are not highly sensitive for hemorrhage detection. Metabolic parameters such as lactate and base deficit (BD) are highly sensitive indicators of blood loss by measuring tissue perfusion. Does adding information from BD and lactate to traditional VS improve the identification of trauma patients with major injuries? METHODS We conducted a retrospective study of a prospectively collected database at a Level I trauma center from January 2003 to September 2005. Patients >13 years, suspected of having significant injury by mechanism, were included. Abnormal VS were defined by heart rate >100 beats/min or systolic blood pressure <90 mmHg. Metabolic parameters from initial arterial blood gas were measured in all patients, abnormal defined by BD >-2.0 mMol/L or lactate >2.2 mMol/L. Our outcome variable, major injury, was defined as any trauma patient who received a blood transfusion, or dropped their hematocrit >10 points in the first 24 h, or had an Injury Severity Score (ISS) >15. RESULTS 1435 patients were enrolled, 242 (17%) had major injuries. Abnormal VS alone had a sensitivity of 40.9% (95% CI, 34.7-47.1%) for identifying major injury patients. When abnormal metabolic parameters were added, major injury detection increased significantly to a sensitivity of 76.4% (95% CI, 71.1-81.8%). CONCLUSIONS The addition of BD and lactate to triage vital signs increases the ability to distinguish major from minor injury.

Ultrasound-guided peripheral intravenous access program is associated with a marked reduction in central venous catheter use in noncritically ill emergency department patients.

STUDY OBJECTIVE We examine the central venous catheter placement rate during the implementation of an ultrasound-guided peripheral intravenous access program. METHODS We conducted a time-series analysis of the monthly central venous catheter rate among adult emergency department (ED) patients in an academic urban ED between 2006 and 2011. During this period, emergency medicine residents and ED technicians were trained in ultrasound-guided peripheral intravenous access. We calculated the monthly central venous catheter placement rate overall and compared the central venous catheter reduction rate associated with the ultrasound-guided peripheral intravenous access program between noncritically ill patients and patients admitted to critical care. Patients receiving central venous catheters were classified as noncritically ill if admitted to telemetry or medical/surgical floor or discharged home from the ED. RESULTS During the study period, the ED treated a total of 401,532 patients, of whom 1,583 (0.39%) received a central venous catheter. The central venous catheter rate decreased by 80% between 2006 (0.81%) and 2011 (0.16%). The decrease in the rate was significantly greater among noncritically ill patients (mean for telemetry patients 4.4% per month [95% confidence interval {CI} 3.6% to 5.1%], floor patients 4.8% [95% CI 4.2% to 5.3%], and discharged patients 7.6% [95% CI 6.2% to 9.1%]) than critically ill patients (0.9%; 95% CI 0.6% to 1.2%). The proportion of central venous catheters that were placed in critically ill patients increased from 34% in 2006 to 81% in 2011 because fewer central venous catheterizations were performed in noncritically ill patients. CONCLUSION The ultrasound-guided peripheral intravenous access program was associated with reductions in central venous catheter placement, particularly in noncritically ill patients. Further research is needed to determine the extent to which such access can replace central venous catheter placement in ED patients with difficult vascular access.

Management of traumatic occult pneumothorax

STUDY OBJECTIVE Occult pneumothorax (OPTX) is defined as a pneumothorax seen on computed tomography but not apparent on supine plain radiography. Though increasingly common, the acute management of OPTX after trauma remains controversial. This evidence-based review evaluates the existing evidence regarding the safety and efficacy of observation as compared to tube thoracostomy (TT) for management of OPTX in emergency department trauma patients. METHODS The authors searched MEDLINE, EMBASE, the Cochrane Library, and other databases. INCLUSION CRITERIA studies of adult or pediatric trauma victims at first presentation after blunt or penetrating injury (population), randomized to observation (intervention) or TT (comparison). Studies that enrolled patients on positive pressure ventilation were included but those that enrolled hemodynamically unstable patients were excluded. Outcomes of interest included progression of OPTX, mortality, complications (pneumonia, empyema), and length of stay in hospital and intensive care unit (ICU). RESULTS A total of 411 articles were identified. After applying the inclusion/exclusion criteria, 3 randomized trials enrolling a total of 101 patients were found to have acceptable quality standards suitable for analysis. The included studies did not reveal any significant difference between observation and TT in regards to progression of OPTX, risk of pneumonia, or length of stay in hospital or ICU. Mortality risk and empyema rate were also not different in the single studies that reported those outcomes. CONCLUSION The existing evidence leads to the conclusion that observation is at least as safe and effective as tube thoracostomy for management of occult pneumothorax.

Bedside Ultrasound Reduces Diagnostic Uncertainty and Guides Resuscitation in Patients With Undifferentiated Hypotension.

Objectives:Utilization of ultrasound in the evaluation of patients with undifferentiated hypotension has been proposed in several protocols. We sought to assess the impact of an ultrasound hypotension protocol on physicians’ diagnostic certainty, diagnostic ability, and treatment and resource utilization. Design:Prospective observational study. Setting:Emergency department in a single, academic tertiary care hospital. Subjects:A convenience sample of patients with a systolic blood pressure less than 90 mm Hg after an initial fluid resuscitation, who lacked an obvious source of hypotension. Interventions:An ultrasound-trained physician performed an ultrasound on each patient using a standardized hypotension protocol. Differential diagnosis and management plan was solicited from the treating physician immediately before and after the ultrasound. Blinded chart review was conducted for management and diagnosis during the emergency department and inpatient hospital stay. Measurements and Main Results:The primary endpoints were the identification of an accurate cause for hypotension and change in physicians’ diagnostic uncertainty. The secondary endpoints were changes in treatment plan, use of resources, and changes in disposition after performing the ultrasound. One hundred eighteen patients with a mean age of 62 years were enrolled. There was a significant 27.7% decrease in the mean aggregate complexity of diagnostic uncertainty before and after the ultrasound hypotension protocol (1.85–1.34; –0.51 [95% CI, –0.41 to –0.62]) as well as a significant increase in the absolute proportion of patients with a definitive diagnosis from 0.8% to 12.7%. Overall, the leading diagnosis after the ultrasound hypotension protocol demonstrated excellent concordance with the blinded consensus final diagnosis (Cohen k = 0.80). Twenty-nine patients (24.6%) had a significant change in the use of IV fluids, vasoactive agents, or blood products. There were also significant changes in major diagnostic imaging (30.5%), consultation (13.6%), and emergency department disposition (11.9%). Conclusions:Clinical management involving the early use of ultrasound in patients with hypotension accurately guides diagnosis, significantly reduces physicians’ diagnostic uncertainty, and substantially changes management and resource utilization in the emergency department.

Does Ethanol Explain the Acidosis Commonly Seen in Ethanol-Intoxicated Patients?

Objective. Emergency physicians frequently treat ethanol-intoxicated trauma patients. In patients with apparently minor injuries, the presence of metabolic acidosis is often attributed to serum ethanol. We tested whether there is justification for the bias that ethanol reliably explains the acidosis commonly seen in alcohol-intoxicated patients. Methods. Prospective, observational. Inclusion criteria: Ethanol-intoxicated patients admitted to the emergency department (ED) following significant trauma mechanisms, in whom diagnostic evaluation revealed only minor injury. Exclusion criteria: Major trauma (blood transfusions, drop in Hct > 10 points over 24 h, or Injury Severity Score [ISS] > 5) or positive urine toxicology screen. Definitions: Ethanol Intoxication: (Blood Alcohol Level (BAL) ≥ 80 mg/dl), Acidosis: BD ≤ − 3.0 mMol/L; Lactic Acidosis (LAC > 2.2 mMol/L). Data were reported as mean ± SD. Data were compared by t-tests or Fishers exact test as appropriate (α = 0.05, 2 tails) and correlations by Pearson correlation coefficient. Results. 192 patients were studied (84% male) with a mean age of 31.7 ± 15.6 years. Acidosis was observed in 19.3% (CI 95%, 14.5% to 25.0%) of all study patients. We observed significant (p < 0.001) difference in prevalence of acidosis in ethanol intoxicated (42%) compared to nonintoxicated (1%) patients. Comparing the two study groups, patients with ethanol intoxication had lower BD ( − 2.24 ± 2.74 vs. − 0.05 ± 2.35, p < 0.001) and higher LAC (2.69 ± 1.48 vs. 2.00 ± 1.78, p = 0.02). However, ethanol levels did not correlate significantly with BD (p = 0.50) or LAC (p = 0.14). Conclusion. Ethanol intoxication is associated with acidosis, which does not correlate with BD or LAC. The complexity of pathogenesis of acidosis in ethanol intoxication justifies further diagnostic evaluation of these patients in order to rule out other causes of acidosis.

Emergency Department Patient Volume and Troponin Laboratory Turnaround Time

OBJECTIVES Increases in emergency department (ED) visits may place a substantial burden on both the ED and hospital-based laboratories. Studies have identified laboratory turnaround time (TAT) as a barrier to patient process times and lengths of stay. Prolonged laboratory study results may also result in delayed recognition of critically ill patients and initiation of appropriate therapies. The objective of this study was to determine how ED patient volume itself is associated with laboratory TAT. METHODS This was a retrospective cohort review of patients at five academic, tertiary care EDs in the United States. Data were collected on all adult patients seen in each ED with troponin laboratory testing during the months of January, April, July, and October 2007. Primary predictor variables were two ED patient volume measures at the time the troponin test was ordered: 1) number of all patients in the ED/number of beds (occupancy) and 2) number of admitted patients waiting for beds/beds (boarder occupancy). The outcome variable was troponin turnaround time (TTAT). Adjusted covariates included patient characteristics, triage severity, season (month of the laboratory test), and site. Multivariable adjusted quantile regression was carried out to assess the association of ED volume measures with TTAT. RESULTS At total of 9,492 troponin tests were reviewed. Median TTAT for this cohort was 107 minutes (interquartile range [IQR] = 73-148 minutes). Median occupancy for this cohort was 1.05 patients (IQR = 0.78-1.38 patients) and median boarder occupancy was 0.21 (IQR = 0.11-0.32). Adjusted quantile regression demonstrated a significant association between increased ED patient volume and longer times to TTAT. For every 100% increase in census, or number of boarders over the number of ED beds, respectively, there was a 12 (95% confidence interval [CI] = 9 to 14) or 33 (95% CI = 24 to 42)-minute increase in TTAT. CONCLUSIONS Increased ED patient volume is associated with longer hospital laboratory processing times. Prolonged laboratory TAT may delay recognition of conditions in the acutely ill, potentially affecting clinician decision-making and the initiation of timely treatment. Use of laboratory TAT as a patient throughput measure and the study of factors associated with its prolonging should be further investigated.

Is all pain is treated equally? A multicenter evaluation of acute pain care by age

Summary Older patients receive less analgesics than younger patients, yet had greater reductions in acute pain scores. These differences may be driven by type of pain. ABSTRACT Pain is highly prevalent in health care settings; however, disparities continue to exist in pain care treatment. Few studies have investigated if differences exist based on patient‐related characteristics associated with aging. The objective of this study was to determine if there are differences in acute pain care for older vs younger patients. This was a multicenter, retrospective, cross‐sectional observation study of 5 emergency departments across the United States evaluating the 2 most commonly presenting pain conditions for older adults, abdominal and fracture pain. Multivariable adjusted hierarchical modeling was completed. A total of 6,948 visits were reviewed. Older (≥65 years) and oldest (≥85 years) were less likely to receive analgesics compared to younger patients (<65 years), yet older patients had greater reductions in final pain scores. When evaluating pain treatment and final pain scores, differences appeared to be based on type of pain. Older patients with abdominal pain were less likely to receive pain medications, while older patients with fracture were more likely to receive analgesics and opioids compared to younger patients. Differences in pain care for older patients appear to be driven by the type of presenting pain.

Adult asthma exacerbations and environmental triggers: a retrospective review of ED visits using an electronic medical record

BACKGROUND Despite familiarity with triggers for asthma, there is little recent study on the association of triggers with the emergency department (ED) presentation of adult asthma exacerbation. METHODS Retrospective electronic chart review of adult patients treated in an urban teaching hospital ED with chief complaint and diagnostic coding related to asthma and upper respiratory tract infection (URI) was conducted. Monthly aeroallergen data and environmental conditions were obtained from a local allergen extract laboratory and local government sources. Data analysis was performed using Newey-West time series regression modeling with adjustment for autocorrelation or ordinary least squares linear regression modeling using outcome variables of asthma visits and admissions. RESULTS There were 56, 747 visits, with 554 asthma visits and 1,514 URI visits. Asthma visits (R(2) = 0.631) were positively correlated with tree pollen counts (correlation coefficient = 0.458 [0.152-0.765]) and average humidity (correlation coefficient = 1.528 [0.296-2.760]). Asthma admissions (R(2) = 0.480) were negatively correlated with average temperature (correlation coefficient = -0.557 [-1.052 to -0.061]) when adjusting for confounding by fine particulate matter. CONCLUSIONS The ED acute asthma exacerbation presentation is positively correlated with tree pollen and humidity, whereas need for admission is associated with dropping temperatures. These results reinforce the need for vigilance during periods of increased risk and perhaps focused preventative strategies.

Pre-Endoscopic Rockall and Blatchford Scores to Identify Which Emergency Department Patients with Suspected Gastrointestinal Bleed Do Not Need Endoscopic Hemostasis

BACKGROUND The pre-endoscopic Rockall Score (RS) and the Glasgow-Blatchford Scores (GBS) can help risk stratify patients with upper gastrointestinal bleed who are seen in the Emergency Department (ED). The RS and GBS have yet to be validated in a United States patient population for their ability to discriminate which ED patients with upper gastrointestinal bleed do not need endoscopic hemostasis. OBJECTIVE We sought to determine whether patients who received a score of zero on either score (the lowest risk) in the ED still required upper endoscopic hemostasis during hospitalization. METHODS Retrospective electronic medical record chart review was performed during a 3-year period (2007-2009) to identify patients with suspected upper gastrointestinal bleed by ED final diagnosis of gastrointestinal hemorrhage and related terms at a single urban academic ED. The RS and GBS were calculated from ED chart abstraction and the hospital records of admitted patients were queried for subsequent endoscopic hemostasis. RESULTS Six hundred and ninety patients with gastrointestinal bleed were identified and 86% were admitted to the hospital. One hundred and twenty-two patients had an RS equal to zero; 67 (55%; 95% confidence interval [CI] 46-63%) of these patients were admitted to the hospital and 11 (16%; 95% CI 9-27%) received endoscopic hemostasis. Sixty-three patients had a GBS equal to zero; 15 (24%; 95% CI 15-36%) were admitted to the hospital and 2 (13%; 95% CI 4-38%) received endoscopic hemostasis. CONCLUSIONS Some patients who were identified as lowest risk by the GBS or RS still received endoscopic hemostasis during hospital admission. These clinical decision rules may be insufficiently sensitive to predict which patients do not require endoscopic hemostasis.

Automated outcome classification of emergency department computed tomography imaging reports.

BACKGROUND Reliably abstracting outcomes from free-text electronic health records remains a challenge. While automated classification of free text has been a popular medical informatics topic, performance validation using real-world clinical data has been limited. The two main approaches are linguistic (natural language processing [NLP]) and statistical (machine learning). The authors have developed a hybrid system for abstracting computed tomography (CT) reports for specified outcomes. OBJECTIVES The objective was to measure performance of a hybrid NLP and machine learning system for automated outcome classification of emergency department (ED) CT imaging reports. The hypothesis was that such a system is comparable to medical personnel doing the data abstraction. METHODS A secondary analysis was performed on a prior diagnostic imaging study on 3,710 blunt facial trauma victims. Staff radiologists dictated CT reports as free text, which were then deidentified. A trained data abstractor manually coded the reference standard outcome of acute orbital fracture, with a random subset double-coded for reliability. The data set was randomly split evenly into training and testing sets. Training patient reports were used as input to the Medical Language Extraction and Encoding (MedLEE) NLP tool to create structured output containing standardized medical terms and modifiers for certainty and temporal status. Findings were filtered for low certainty and past/future modifiers and then combined with the manual reference standard to generate decision tree classifiers using data mining tools Waikato Environment for Knowledge Analysis (WEKA) 3.7.5 and Salford Predictive Miner 6.6. Performance of decision tree classifiers was evaluated on the testing set with or without NLP processing. RESULTS The performance of machine learning alone was comparable to prior NLP studies (sensitivity = 0.92, specificity = 0.93, precision = 0.95, recall = 0.93, f-score = 0.94), and the combined use of NLP and machine learning showed further improvement (sensitivity = 0.93, specificity = 0.97, precision = 0.97, recall = 0.96, f-score = 0.97). This performance is similar to, or better than, that of medical personnel in previous studies. CONCLUSIONS A hybrid NLP and machine learning automated classification system shows promise in coding free-text electronic clinical data.