Circadian differences among 4,096 emergency department patients with acute asthma.

INTRODUCTION: In patients with asthma, airways narrow during the night. The clinical implications of a nocturnal presentation of patients with acute asthma to the emergency department (ED) are uncertain. OBJECTIVE: Our objective was to determine whether patients with asthma who had ED visits during the night (midnight to 7:59 am) vs. other times were more severe, responded less well to ED therapy, and had worse clinical outcomes. DESIGN AND SETTING: We performed a cohort study, as part of the Multicenter Airway Research Collaboration (n = 77 sites). ED patients with acute asthma, ages 2-54 yrs, underwent a structured interview in the ED. Chart review of missed/refusal patients created a truly consecutive case series. MEASUREMENTS AND MAIN RESULTS: Among 1,602 children, 19% presented at night Nighttime patients were more likely to be younger, male, and have a shorter duration of symptoms; there were no other clinical differences noted. Among 2,494 adults, 20% presented at night, and they were more likely to be female and to have a history of steroid use for asthma. Nighttime adults also had a shorter duration of symptoms and slightly lower peak flows (mean, 45% vs. 49% of predicted; p = .006) and were more likely to receive steroids. They were more likely to be intubated (2.0% vs. 0.2%; p < .001), but, overall, they were equally likely to be admitted or relapse after ED discharge. In contrast to objective measures of acute asthma severity, both nighttime children and adults were significantly less likely to report their asthma symptoms as severe. CONCLUSION: Except for endotracheal intubation (in adults only), circadian differences minimally affect ED presentation, therapy, or the outcomes of acute asthma. Nighttime asthmatics may be relatively insensitive to the symptoms of severe asthma.

Serum markers in the emergency department diagnosis of acute myocardial infarction.

No currently used cardiac-specific serum marker meets all the criteria for an "ideal" marker of AMI. No test is both highly sensitive and highly specific for acute infarction within 6 hours following the onset of chest pain, the timeframe of interest to most emergency physicians in making diagnostic and therapeutic decisions. Patients presenting to the ED with chest pain or other symptoms suggestive of acute cardiac ischemia therefore cannot make a diagnosis of AMI excluded on the basis of a single cardiac marker value obtained within a few hours after symptom onset. The total CK level is far too insensitive and nonspecific a test to be used to diagnose AMI. It retains its value, however, as a screening test, and serum of patients with abnormal total CK values should undergo a CK-MBmass assay. Elevation in CK-MB is a vital component of ultimate diagnosis of AMI, but levels of this marker are normal in one fourth to one half of patients with AMI at the time of ED presentation. The test is highly specific, however, and an abnormal value (particularly when it exceeds 5% of the total CK value) at any time in a patient with chest pain is highly suggestive of an AMI. There have been several improvements of CK-MB assay timing and subform quantification that appear highly useful for emergency physicians. Rapid serial CK-MB assessment greatly increases the diagnostic value of the assay in a timeframe suitable for ED purposes but unfortunately still misses about 10% of patients ultimately diagnosed with acute MI. Assays of CK-MB subforms have very high sensitivity, and, although unreliable within 4 hours of symptom onset, have excellent diagnostic value at 6 or more hours after chest pain begins. Automated test assays recently have become available and could prove applicable to ED settings. The cardiac troponins are highly useful as markers of acute coronary syndromes, rather than specifically of AMI, and abnormal values at any time following chest pain onset are highly predictive of an adverse cardiac event. The ED applicability of the troponins is severely limited, however, because values remain normal in most patients with acute cardiac events as long as 6 hours following symptom onset. Myoglobin appeared promising as a marker of early cardiac ischemia but appears to be only marginally more sensitive than CK-MB assays early after symptom onset and less sensitive than CK-MB at 8 hours or more after chest pain starts. Rapid serial myoglobin assessment, however, appears highly useful as an early marker of AMI. The marker has a very narrow diagnostic window. The clinician is left with several tests that are highly effective in correctly identifying patients with AMI (or at high risk for AMI), but none that can dependably exclude patients with acute coronary syndromes soon after chest pain onset. A prudent strategy when assessing ED patients with chest pain and nondiagnostic ECGs is to order CK-MB and troponin values on presentation in the hope of making an early diagnosis of AMI or unstable coronary syndrome. Although it is recognized that normal values obtained within 6 hours of symptom onset do not exclude an acute coronary syndrome, patients at low clinical risk and having normal cardiac marker tests could be provisionally admitted to low-acuity hospital settings or ED observation. After 6 to 8 hours of symptom duration has elapsed, the cardiac-specific markers are highly effective in diagnosing AMI, and such values obtained can be used more appropriately to make final disposition decisions. At no time should results of serum marker tests outweigh ECG findings or clinical assessment of the patient's risk and stability.

Clinically meaningful changes in quantitative measures of asthma severity.

OBJECTIVE: To determine minimum clinically meaningful improvements in peak expiratory flow rate (PEFR) and dyspnea visual analog score (VAS) in patients with acute asthma exacerbation. METHODS: Patients presenting to the emergency department (ED) with acute asthma exacerbation were eligible. The PEFR and VAS were assessed at presentation and after initial asthma therapy. During reassessment, subjects were asked to describe their asthma symptoms as "much better," "a little better," "no change," "a little worse," or "much worse." Correspondence between self-reported improvement and changes in PEFR and VAS was assessed. The "minimum clinically significant change" in either index was defined as the difference between pre- and posttreatment measures in subjects reporting their symptoms "a little better." RESULTS: One hundred fifty-six subjects were included. Asthma symptoms were "much better" in 99 (64%), "a little better" in 41 (26%), and "unimproved" (composed of patients describing symptoms as "no change," "a little worse," or "much worse") in 16 (10%). The mean VAS change among the "a little better" subjects was 2.2 cm (95% CI = 1.1 to 3.4), significantly greater than the -0.4 cm (95% CI = -2.1 to 1.4) change in the "unimproved" subjects. The mean change in percent predicted PEFR among the "a little better" subjects was 11.9 (95% CI = 7.3 to 16.1), not statistically different from the change of 6.1 (95% CI = 1.1 to 11.3) in the "no change" subjects. The "much better" group showed significantly greater changes in both measures than either of the other groups. A VAS change of > or =0.5 cm reliably discriminated between subjects with and without symptom improvement. CONCLUSIONS: Improvements in VAS of 2.2 cm and in predicted PEFR of about 12 percentage points are minimal clinically significant improvements during ED asthma therapy. The dyspnea VAS is valid in assessing symptomatic changes and may detect small subjective improvements better than the PEFR.

Endogenous mediators in emergency department patients with presumed sepsis: are levels associated with progression to severe sepsis and death?

STUDY OBJECTIVE: We sought to determine whether levels of the endogenous mediators tumor necrosis factor (TNF)-alpha, interleukin (IL) 6, and nitric oxide (NO) measured in patients with presumed sepsis (systemic inflammatory response syndrome [SIRS] and infection) are different than levels in patients with presumed noninfectious SIRS, whether levels are associated with septic complications, and whether there are potential relationships between mediators. METHODS: A prospective, observational tricenter study of a convenience sample of adults presenting to the emergency department meeting Bone's criteria for SIRS (any combination of fever or hypothermia, tachycardia, tachypnea, or WBC count aberration) was performed. Mediator levels were determined and associated with deterioration to severe sepsis (hypotension, hypoperfusion, or organ dysfunction) and death in subjects admitted to the hospital with presumed sepsis. RESULTS: One hundred eighty subjects with SIRS were enrolled and classified into 3 groups: group 1 (SIRS, presumed infection, admitted; n=108), group 2 (SIRS, presumed infection, discharged; n=27), and group 3 (SIRS, presumed noninfectious, admitted; n=45). Group 1 TNF-alpha and IL-6 levels were significantly higher than those found in the other groups. NO levels for groups 1 and 2 were significantly lower than those for group 3. TNF-alpha and IL-6 levels were higher in the group 1 subjects who had bacteremia or progressed to severe sepsis or death. NO levels were not associated with these outcomes. CONCLUSION: ED patients admitted with presumed sepsis have elevated cytokine levels compared with patients with sepsis who are discharged and with those patients with presumed noninfectious SIRS. An association appears to exist between cytokines and subsequent septic complications in these patients. The importance of these measures as clinical predictors for the presence of infection and subsequent septic complications needs to be evaluated.

Absolute lymphocyte count as a predictor of CD4 count.

STUDY OBJECTIVE: To determine whether the absolute lymphocyte count (ALC) (white blood count x lymphocyte percentage) can be used to predict a low CD4 count. METHODS: We conducted a retrospective data analysis of consecutive CD4 count analyses performed between January 1, 1995, through December 1, 1995, at an urban university teaching hospital. Results of consecutive CD4 counts and simultaneously measured ALCs were analyzed from samples obtained in inpatient, clinic, and emergency department settings. The ability of ALC to predict a CD4 count less than 200 cells/mm3 was analyzed by calculating sensitivities, specificities, predictive values, and likelihood ratios for a range of ALC values. RESULTS: Among the 807 samples, 322 results (40%) had a CD4 count less than 200 cells/mm3. The ALC and CD4 count were correlated (r=.69, P<.0001). An ALC less than 1,000 cells/mm3 predicted CD4 counts less than 200 cells/mm3 with a sensitivity of .67 (95% confidence interval .62 to .72), specificity of .96 (.94 to .98), positive predictive value of .91 (.87 to .95), and a negative predictive value of .81 (.78 to .84). An ALC less than 2,000 cells/mm3 predicted CD4 counts less than 200 cells/mm3 with a sensitivity of .97 (.95 to .99), specificity of .41 (.37 to .45), positive predictive value of .52 (.48 to .56), and negative predictive value of .95 (.92 to .98). CONCLUSION: A reliable relationship exists between ALC and CD4 count. In a similar population, an ALC less than 1,000 cells/mm3 is predictive of a CD4 count less than 200 cells/mm3, and an ALC greater than or equal to 2,000 cells/mm3 is predictive of a CD4 count greater than or equal to 200 cells/mm3. Physicians may find these criteria useful in identifying patients with increased risk of opportunistic infection.

Statistical methodology: II. Reliability and validity assessment in study design, Part B.

Validity measures the correspondence between a test and other purported measures of the same or similar qualities. When a reference standard exists, a criterion-based validity coefficient can be calculated. If no such standard is available, the concepts of content and construct validity may be used, but quantitative analysis may not be possible. The Pearson and Spearman tests of correlation are often used to assess the correspondence between tests, but do not account for measurement biases and may yield misleading results. Techniques that measure interest differences may be more meaningful in validity assessment, and the kappa statistic is useful for analyzing categorical variables. Questionnaires often can be designed to allow quantitative assessment of reliability and validity, although this may be difficult. Inclusion of homogeneous questions is necessary to assess reliability. Analysis is enhanced by using Likert scales or similar techniques that yield ordinal data. Validity assessment of questionnaires requires careful definition of the scope of the test and comparison with previously validated tools.

Statistical methodology: II. Reliability and variability assessment in study design, Part A.

Assessment of test reliability and validity is often complex. Although tests of correlation are frequently used to measure interest agreement, such indexes measure only the strength of the linear relationship between variables and may not provide an accurate assessment of the correspondence between test results. Inspection of interest differences, either visually or using the r1, may provide a better indicator of the correspondence between test results and accounts for measurement biases. Strength of association between categorical variables can be measured using related tests such as the kappa statistic. Test reliability may be assessed by retesting, but this is not practical in many cases when subject memory or learning may confound the results of repeated examinations. Several methods exist for determining reliability from a single test administration and for assessing the correspondence between answers to homogeneous test questions. In the continuation article (Part B) on this subject, the concept and assessment of validity will be examined in more detail, and techniques for maximizing the reliability and validity of questionnaires will be discussed.

Poisoning from "Spanish fly" (cantharidin).

Cantharidin, known popularly as Spanish fly, has been used for millennia as a sexual stimulant. The chemical is derived from blister beetles and is notable for its vesicant properties. While most commonly available preparations of Spanish fly contain cantharidin in negligible amounts, if at all, the chemical is available illicitly in concentrations capable of causing severe toxicity. Symptoms of cantharidin poisoning include burning of the mouth, dysphagia, nausea, hematemesis, gross hematuria, and dysuria. Mucosal erosion and hemorrhage is seen in the upper gastrointestinal (GI) tract. Renal dysfunction is common and related to acute tubular necrosis and glomerular destruction. Priapism, seizures, and cardiac abnormalities are less commonly seen. We report four cases of cantharidin poisoning presenting to our emergency department with complaints of dysuria and dark urine. Three patients had abdominal pain, one had flank pain, and the one woman had vaginal bleeding. Three had hematuria and two had occult rectal bleeding. Low-grade disseminated intravascular coagulation, not previously associated with cantharidin poisoning, was noted in two patients. Management of cantharidin poisoning is supportive. Given the widespread availability of Spanish fly, its reputation as an aphrodisiac, and the fact that ingestion is frequently unwitting, cantharidin poisoning may be a more common cause of morbidity than is generally recognized. Cantharidin poisoning should be suspected in any patient presenting with unexplained hematuria or with GI hemorrhage associated with diffuse injury of the upper GI tract.

Chest pain associated with cocaine: an assessment of prevalence in suburban and urban emergency departments.

STUDY OBJECTIVE: Chest pain and myocardial infarction following the use of cocaine have been well documented. We assessed the prevalence of cocaine use in patients who presented to the emergency department with chest pain of possibly ischemic origin. DESIGN: During times of research assistant availability, consecutive adults with the chief complaint of chest pain unexplained by trauma or radiographic abnormality were questioned about cocaine use in the preceding week. Urine was tested for the presence of cocaine or cocaine metabolites with a highly accurate bedside urine test kit (specificity, 100%; sensitivity 98%). Anonymous unlinked data-collection methods were used. Therefore we could not determine whether the patients who used cocaine had sustained myocardial infarctions. SETTING: One suburban and three urban EDs. RESULTS: We enrolled 359 patients with a mean age of 51 years, 8% of whom sustained myocardial infarctions. Sixty patients (17%) had cocaine or cocaine metabolites in urine. The likelihood of testing positive for cocaine varied by age group: 18 to 30 years, 29%; 31 to 40 years, 48%; 41 to 50 years, 18%; 51 to 60 years, 3%; 61 years or older, 0% (P < .0001). Of the 60 patients who tested positive for cocaine, only 43 (72%) admitted recent use. CONCLUSION: Many ED patients with chest pain have recently used cocaine. Because the recent use of cocaine is not uncommon in patients with chest pain up to 60 years old, such patients should be questioned about cocaine use. When treatment or disposition may be altered, consideration should be given to objective assessment of cocaine use because patient self-report does not appear reliable.

Is circadian variation in asthma severity relevant in the emergency department?

STUDY OBJECTIVE: Asthmatic patients have marked circadian variation in disease severity, with bronchospasm far worse between midnight and 8 AM than at other times of day. In this study we sought to determine whether circadian variation in asthma severity is clinically relevant for purposes of emergency management. DESIGN: Prospective, observational cohort study. SETTING: The emergency department of a large urban university hospital. PARTICIPANTS: Asthmatic patients 16 years and older who presented with the complaint of asthma exacerbation. RESULTS: Peak expiratory flow rate (PEFR), respiratory rate, and objective and subjective assessments of dyspnea severity were determined at the time of each patient's presentation to the ED and at the time of disposition decision. Time of symptom onset, duration of symptoms, response to therapy, and final disposition were recorded. Nocturnal presentation and onset were defined as arrival in the ED or onset of symptoms between midnight and 8 AM. We enrolled 188 asthmatic patients in the study; 52 (28%) presented nocturnally. Comparison of asthmatic patients who presented nocturnally and those who presented at other times revealed no differences in initial or final PEFR or respiratory rate, change in PEFR, subjective or objective dyspnea, or rate of hospitalization, although nocturnal patients did have a significantly shorter ED treatment time. Analysis by time of symptom onset also revealed no differences between nocturnally and nonnocturnally presenting subjects in disease severity or outcome, with the exception that those with nocturnal symptom onset had less subjective dyspnea than other asthmatic patients. Repeat analysis with different "nocturnal" time periods did not alter the results. CONCLUSION: In our study group, asthmatic patients who presented nocturnally to the ED did not differ from other asthmatic patients in disease severity and therefore do not appear to represent a unique population for purposes of ED management or disposition decisions.

Comprehension of discharge instructions by patients in an urban emergency department.

STUDY OBJECTIVE: To assess patients' comprehension of their emergency department discharge instructions and to determine if inner-city patients' literacy levels are adequate to comprehend written discharge instructions. DESIGN: Prospective, observational study. SETTING: The ED of an inner-city university hospital. PARTICIPANTS: Two hundred seventeen patients consecutively discharged from the ED during 12 separate time slots. INTERVENTIONS: Patients were interviewed after discharge from the ED and asked to state their diagnosis, medication instructions, and follow-up instructions. Comparisons between patient recall and instructions as written in the chart were assessed by independent raters and scored from poor to excellent. Patients were administered a standardized test of reading ability. RESULTS: Overall comprehension rates were judged to be good, although 23% of patients exhibited no understanding of at least one component of their discharge instructions. Mean reading ability of the patients was at the sixth-grade level. The ED's printed discharge instructions were written at an 11th-grade reading level. Patients with low literacy scores were more likely to have poor comprehension of instructions. CONCLUSION: Overall comprehension rates in this population were good despite the fact that ED instruction sheets were written at an inappropriately high reading level. Verbal instructions given by the discharging physician likely have a significant effect on patients' comprehension of instructions.

Severe low back pain secondary to acute interstitial nephritis following administration of ranitidine.

Acute interstitial nephritis is a disease characterized by renal inflammation and is thought to be secondary to a hypersensitivity reaction. Although the causes of acute interstitial nephritis are numerous, adverse reactions to many common drugs, particularly antibiotics and nonsteroidal anti-inflammatory agents, are important etiological factors. Acute interstitial nephritis has many clinical manifestations, most notably fever and rash. Flank pain is an uncommon presentation. A case of acute, severe, low-back pain and rash in a healthy woman found to be secondary to acute interstitial nephritis is reported. The etiology of acute interstitial nephritis in this patient's case is suspected to be ranitidine (Zantac; Glaxo Pharmaceuticals, Research Triangle Park, NC), which has not been previously associated with this syndrome.