Publications

STUDY OBJECTIVE: Glottic visualization with video is superior to direct laryngoscopy in controlled operating room studies. However, glottic exposure with video laryngoscopy has not been evaluated in the emergency department (ED) setting, where blood, secretions, poor patient positioning, and physiologic derangement can complicate laryngoscopy. We measure the difference in glottic visualization with video versus direct laryngoscopy.

METHODS: We prospectively studied a convenience sample of tracheal intubations at 2 academic EDs. We performed laryngoscopy with the Karl Storz Video Macintosh Laryngoscope, which can be used for conventional direct laryngoscopy, as well as video laryngoscopy. We rated glottic visualization with the Cormack-Lehane (C-L) Scale, defining "good" visualization as C-L I or II and "poor" visualization as C-L III or IV. We compared glottic exposure between direct and video laryngoscopy, determining the proportion of poor direct visualizations improved to good visualization with video laryngoscopy. We also determined the proportion of good direct visualizations worsened to poor visualization by video laryngoscopy.

RESULTS: We report data on 198 patients, including 146 (74%) medical, 51 (26%) trauma, and 1 (0.51%) unknown indications. All were tracheally intubated by emergency physicians. Postgraduate year 3 or 4 residents performed 102 (52.3%) of the laryngoscopies, postgraduate year 2 residents performed 60 (30.8%), interns performed 20 (10.3%), attending physicians performed 9 (4.6%), and operator experience and specialty were not reported in 4. Overall, good visualization (C-L grade I or II) was attained in 158 direct (80%) versus 185 video laryngoscopies (93%; McNemar's P<.0001). Of the 40 patients with poor glottic exposure on direct laryngoscopy, video laryngoscopy improved the view in 31 (78%; 95% confidence interval 62% to 89%). Of the 158 patients with good glottic view on direct laryngoscopy, video laryngoscopy worsened the view in 4 (3%; 95% confidence interval 0.7% to 6%).

CONCLUSION: Video laryngoscopy affords more grade I and II views than direct laryngoscopy and improves glottic exposure in most patients with poor direct glottic visualization. In a small proportion of cases, glottic exposure is worse with video than direct laryngoscopy.

PURPOSE: A large share of the information in electronic medical records (EMRs) consists of free-text compositions. From a computational point-of-view, the continuing prevalence of free-text entry is a major hindrance when the goal is to increase automation in EMRs. However, the efforts in developing standards for the structured representation of medical information have not proven to be a panacea. The information space of clinical medicine is very diverse and constantly evolving, making it challenging to develop standards for the domain. This paper reports a study aiming to increase automation in the EMR through the computational understanding of specific class of medical text in English, namely emergency department chief complaints.

METHODS: We apply domain-specific analytical modeling for the computational understanding of chief complaints. We evaluate the performance of this approach in the automatic classification of chief complaints, e.g., for use in automatic syndromic surveillance.

RESULTS: The evaluation in a multi-hospital setting showed that the presented algorithm was accurate in terms of classification correctness. Also, use of approximate matching in the algorithm to cope with typographic variance did not affect classification correctness while increasing classification completeness.

OBJECTIVES: Serum lactate levels are a useful tool in monitoring critically ill patients, especially those who are septic. However, lactate levels are often not routinely drawn or rapidly available in some institutions. The objective of this study was to determine if a readily available anion gap (AG) could be used as a surrogate marker for abnormal lactate level in Emergency Department (ED) patients at risk for sepsis.

METHODS: Prospective, observational cohort study of consecutive ED patients seen at an urban university tertiary care referral center with 46,000 annual ED visits. ED patients aged 18 years or older presenting with clinically suspected infection were eligible for enrollment if a serum chemistry and lactate levels were drawn during the ED visit. During the 9-month study period, 1419 patients were enrolled. The initial basic chemistry panels, calculated AG, and lactate levels drawn in the ED were collected. We defined, a priori, an AG > 12 and a lactate > 4 mmol/L to be abnormal. Analysis was performed with Student's t-test, operating characteristics with 95% confidence intervals, and logistic regression.

RESULTS: The mean AG was 11.8 (SD 3.6) and the mean lactate was 2.1 (SD 1.3). For an AG > 12, the mean lactate was 2.9 (SD 1.7), compared with 1.8 (SD 0.8) for an AG < 12. The sensitivity of an elevated AG (> 12) in predicting elevated lactate levels (> 4 mmol/L) was 80% (72-87%) and the specificity was 69% (66-71%). Patients with a gap > 12 had a 7.3-fold (4.6-11.4) increased risk of having a lactate > 4 mmol/L. The area under the curve was 0.84.

CONCLUSION: This study suggests that an elevated AG obtained in the ED is a moderately sensitive and specific means to detect elevated lactate levels in ED patients at risk for sepsis. This information may be somewhat helpful to Emergency Physicians to risk-stratify their patients to provide more aggressive early resuscitation.

OBJECTIVES: To determine the prevalence and associated characteristics of traumatic intracranial hemorrhage (ICH) in elderly fallers presenting to the emergency department (ED) without focal findings.

DESIGN: Retrospective cohort study.

SETTING: University-affiliated teaching hospital ED.

PARTICIPANTS: Patients aged 65 and older presenting with a fall to the ED and undergoing a head computed tomography (CT) scan.

MEASUREMENTS: Electronic medical records and CT scans of 404 consecutive patients were reviewed. Characteristics of patients with and without ICH were compared using unadjusted analyses. Patients taking warfarin, aspirin, or clopidogrel alone or in combination were compared with those not taking these medications. Multivariate logistic regression analyses were performed to determine variables independently associated with ICH.

RESULTS: Forty-seven of 404 elderly fallers (11.6%) without focal findings had an ICH. Unadjusted analyses in these pilot data showed that warfarin was not significantly associated with ICH. Multivariate analyses indicated that elderly people living in at home were more likely to have ICH than those living in nursing homes or assisted living facilities (odds ratio (OR)=3.2, 95% confidence interval (CI)=1.30-8.13) and that those with head trauma were more likely to have ICH than those without (OR=3.9, 95% CI=1.25-7.80). Aspirin was found to be protective (OR=0.49, 95% CI=0.24-0.98).

CONCLUSION: ICH is common in elderly fallers presenting to the ED without focal findings. Anticoagulation alone did not appear to increase the risk of ICH, and aspirin was found to be protective, but prospective studies are needed to better assess this relationship.

An approach coined as "reflective information management" is presented as a technique for the management of emergency department chief complaint information. The architecture of a system integrating principles from this approach is described and its performance is evaluated in providing categorical information from free-text chief complaints for use, e.g., in automated syndromic surveillance.

The objective of this study was to determine the incidence of septic arthritis in the presence of joint crystals. A retrospective study was conducted at a university tertiary care referral center. The study population included all patients with synovial fluid crystals in the joint aspirate sent to the laboratory during the 7-year study period. Septic arthritis was defined as a positive synovial culture. Of the 265 joint aspirates containing crystals, 183 (69.0%) contained gout crystals, 81 (30.6%) contained pseudogout crystals, and 1 (0.4%) contained both. Four (1.5%) of the aspirates had positive cultures. The mean synovial WBC of the 4 samples with concomitant crystals and septic arthritis was 113,000 (95% confidence interval [CI] 72,700-153,200), which was significantly higher than the entire population at 23,200 (95% CI 19,400-27,000; p < 0.01). Of note, all 4 patients with concomitant disease had significant co-morbidities and synovial WBC counts greater than 50,000. Septic arthritis and acute crystal-induced arthritis can occur simultaneously; there were 4 cases (1.5%) of concomitant disease in our study population. The presence of crystals cannot exclude septic arthritis with certainty.

There is a need for an efficient and effective means of categorizing Emergency Department Chief Complaints. We improved the performance of our previously described learning normalizer algorithm by broadening its training set to include data from multiple hospitals. We also achieved a statistically significant additional improvement by incorporating a spell-checking algorithm.

OBJECTIVE: This study was conducted to determine the sensitivity of the current standard for synovial fluid leukocytosis analysis in diagnosing infectious arthritis or a septic joint. How accurate is the standard synovial fluid white blood cell (WBC) cutoff of 50,000 WBC/mm3 to rule out septic arthritis?

METHODS: We conducted a retrospective study at an urban tertiary care medical center with 50,000 adult emergency department visits per year. The study population consisted of patients with infectious arthritis confirmed by synovial fluid culture growth of a pathogenic organism. The study period lasted from January 1996 to December 2002. Extracted data included synovial fluid leukocyte count, Gram's stain, culture, past medical history, and discharge diagnosis. Fisher exact test was used to compare proportions. Sensitivity and means were calculated with 95% confidence intervals (CI).

RESULTS: There were 49 culture-positive synovial fluid aspirates in the 6-year study period. Nineteen (39%) of 49 patients (95% CI, 25%-52%) had a synovial WBC of less than 50,000/mm3 and 30 (61%) of 49 patients (95% CI, 48%-75%) had a synovial WBC of more than 50,000/mm3. The sensitivity of the 50,000 synovial WBC/mm3 cutoff was 61% (95% CI, 48%-75%). Twenty-seven (55%) of 49 patients had a negative Gram's stain (95% CI, 41%-69%) and 15 (56%) of 27 patients (95% CI, 37%-74%) with negative Gram's stain had a synovial WBC of less than 50,000/mm3.

CONCLUSION: A synovial fluid WBC cutoff of 50,000/mm3 lacks the sensitivity required to be clinically useful in ruling out infectious arthritis.

STUDY OBJECTIVE: Little is known about risk-stratification biomarkers in emergency department (ED) patients with suspected infection, and lactate is a biologically plausible candidate. We determine whether a serum venous lactate is associated with an increased risk of death in ED patients with infection.

METHODS: This was a prospective cohort study in an urban, academic medical center with 50,000 annual ED visits. A total of 1,278 consecutive patient visits met enrollment criteria between July 24, 2003, and March 24, 2004, and all patients were enrolled. Inclusion criteria were age 18 years or older, serum lactate level obtained, and admission to the hospital with an infection-related diagnosis. The main outcome measure was all-cause 28-day inhospital mortality and death within 3 days of presentation.

RESULTS: Among 1,278 patient visits, there were 105 (8.2%) deaths during hospitalization, with 55 (4.3%) of 1,278 deaths occurring in the first 3 days. Mortality rates increased as lactate increased: 43 (4.9%) of 877 of patients with a lactate level between 0 and 2.5 mmol/L died, 24 (9.0%) of 267 patients with a lactate level between 2.5 and 4.0 mmol/L died, and 38 (28.4%) of 134 patients with a lactate level greater than or equal to 4.0 mmol/L died. Lactate level greater than or equal to 4.0 mmol/L was 36% (95% confidence interval [CI] 27% to 45%) sensitive and 92% (95% CI 90% to 93%) specific for any death; it was 55% (95% CI 41% to 68%) sensitive and 91% (95% CI 90% to 93%) specific for death within 3 days.

CONCLUSION: In this cohort of ED patients with signs and symptoms suggestive of infection, our results support serum venous lactate level as a promising risk-stratification tool. Multicenter validation, as well as comparison of the lactate level with clinical predictors, needs to be done before widespread implementation.