Frequency of Abnormal and Critical Laboratory Results in Older Patients Presenting to the Emergency Department With Syncope

Syncope is a common and costly chief complaint among patients presenting to the emergency department (ED), accounting for 740,000 ED visits annually with an estimated annual cost of $2.4 billion per year in the United States.1, 2 Syncope presents a diagnostic dilemma for clinicians in the ED since differentiating serious and benign causes of syncope can be challenging, particularly in the older adult. Routine laboratory testing with complete blood count (CBC) and basic metabolic panel (BMP) is commonly ordered for patients presenting to the ED with syncope.3 The 2017 American Heart Association (AHA), American College of Cardiology (ACC), and Heart Rhythm Society (HRS) guidelines on the evaluation of syncope in the ED recommend targeted laboratory testing based on history and physical examination, such as prior history of peptic ulcer disease or dark tarry stools.4 Despite these guidelines, it is common practice to obtain broad panel laboratory testing such as CBC or BMP for ED evaluation of patients with syncope regardless of level of suspicion for abnormalities.3, 4 Broad panel laboratory testing with CBC and BMP could identify potentially reversible causes of syncope in the older adult including anemia, dehydration, and electrolyte abnormalities. It is not known how frequently routine laboratory testing identifies normal, abnormal, and critical laboratory values in older patients with syncope. The objective of this study was to evaluate the frequency of normal, abnormal, and critical laboratory results from CBC and BMP ordered in older patients presenting to the ED with syncope or near syncope. We conducted a secondary analysis of a large, multicenter, prospective cohort study (ClinicalTrials.gov identifier NCT01802398) to determine the frequency of abnormal and critical routine laboratory testing at the primary ED visit. The study was approved by the institutional review boards at all sites, and written informed consent was obtained from all participating subjects. Eligible patients were ≥60 years of age with a complaint of syncope or near syncope at 11 academic EDs across the United States. Exclusion criteria were intoxication, medical or electrical intervention to restore consciousness, or inability or unwillingness to provide informed consent or follow-up information. Patients with a presumptive cause of loss of consciousness due to seizure, stroke, or transient ischemic attack or hypoglycemia were also excluded. For this analysis, we also excluded patients who did not have a CBC or BMP performed. All patients underwent history and physical examination, laboratory testing at the discretion of the treating physician, and 12-lead EKG testing. Patient disposition was directed by the treating clinical providers. We evaluated the following routine laboratory tests: hematocrit, glucose, sodium, potassium, bicarbonate, blood urea nitrogen, and creatinine. We defined abnormal ranges for the BMP using standard reference ranges.5 We determined abnormal hematocrit ranges based on preexisting syncope literature.6 We determined critical values by an institutional laboratory protocol.7 The primary outcome is the proportion of abnormal and critical laboratory values obtained from CBC and BMP laboratory testing. We present data as means with standard deviations (SD) or medians with interquartile ranges. We tested independence between categorical variables with chi-square or Fisher’s exact test. We performed statistical analyses using the R package. All p-values are two-sided and we considered those ≤ 0.05 to be significant. There were 6,930 subjects that met eligibility criteria, of whom 3,686 (53.2%) consented and were enrolled into the primary study. Of the 3,686 enrolled subjects, 105 subjects had no serious outcome and were lost to follow-up or withdrawn from the study, resulting in a base cohort of 3,581 subjects. There were 3,557 patients (99.3%) who had CBC or BMP laboratory testing performed in the ED, representing the cohort for this study. Twenty-four subjects (0.7%) with a mean (±SD) age of 68.4 (±6.8) years (p = 0.015), but otherwise similar baseline demographics to the study cohort, had missing laboratory values and were excluded from the analysis (Data Supplement S1, available as supporting information in the online version of this paper, which is available at http://onlinelibrary.wiley.com/doi/10.1111/acem.13096/full) . Study subjects had a mean (±SD) age of 72.8 (±9) years, 1,835 (51.6%) were male, and 1,722 (48.4%) female. Compared to patients who had normal laboratory testing, those who had abnormal laboratory testing or critical laboratory testing were more likely to have a history of congestive heart failure, coronary artery disease, or arrhythmia; be on diuretics, beta blockers, or calcium channel blockers; and have an abnormal ECG or hypotension on initial vital signs and to be hospitalized. Of all laboratory results collected, 69.9% were normal, 29.9% were abnormal, and 0.7% were in the critical range (Table 1). We found 323 patients (9.1%) had entirely normal laboratory results, 3,080 patients (86.6%) had at least one abnormal, but no critical, laboratory results, and 154 patients (4.3%) had at least one critical laboratory result (Table 1). Abnormal laboratory values were most commonly elevated blood glucose (72.9%), elevated blood urea nitrogen (44.4%), hyponatremia (21.7%), hypokalemia (16.7%), and anemia (8.4%; Table 1). The most common critical laboratory results were serum creatinine (2.5%), elevated serum potassium (0.7%), elevated blood urea nitrogen (0.6%), and low hematocrit (0.4%; Table 1). We found that in a large cohort of older adults with syncope, abnormal laboratory values were frequently identified in laboratory screening of this population with CBC and BMP. Critical laboratory values on laboratory testing with CBC and BMP occurred in 4.3% of this population. Current AHA/ACC/HRS guidelines indicate routine and comprehensive laboratory testing is not useful in the evaluation of syncope.4 Our findings suggest otherwise in a cohort of older adults. In patients ≥ 60 years of age, with syncope or near syncope, critical laboratory values occurred at a rate above the generally accepted risk threshold for ED presentations such as pulmonary embolism.8 In this population, laboratory testing with CBC and BMP is likely to provide abnormal results with minimal increased risk to the patient. Professional society guidelines recommend a minimal role for diagnostic testing in the evaluation and management of syncope.4 However, these guidelines are primarily based on literature of small study cohorts and included young, healthy individuals with mean age of 41.3 and 63.1 years, respectively.9, 10 Young, healthy patients are much more likely to have a vasovagal syncopal event and it is possible that inclusion of these patients diminished the value of broad-panel blood testing in the evaluation of syncope in these studies. Further, the small cohort sizes and wide age range of patients suggests that these early syncope studies may not accurately present the utility of laboratory screening in older adults with syncope. We limited enrollment to patients ≥ 60 years of age, or older, and our mean (±SD) patient age was 72.8 (±9) years. Our data suggest that patients ≥ 60 years have frequent abnormal laboratory findings on CBC and BMP and occasional (4.3%) critical laboratory findings. Our research demonstrates that routine laboratory testing in older adults with syncope may affect disposition decisions in the ED or identify needs for ED intervention. The major limitation of this study is that laboratory tests occurred at an index ED visit without comparison to historical laboratory results for each patient. Second, our study cohort had a small percentage of subjects with missing laboratory values that were not included in the analysis. Third, laboratory processing errors or spurious laboratory results due to hemolysis or dilution may have contributed to our high proportion of abnormal tests. Fourth, further stratification of laboratory values may increase the clinical significance of laboratory results. We selected abnormal and critical laboratory value ranges defined by existing literature and institutional guidelines, respectively. In some instances, such as glucose, our abnormal laboratory values may include a range that is not clinically relevant. Fifth, although we found that increased medication use, medical comorbidities, ECG abnormalities, and hypotension were associated with abnormal and critical laboratory results, this descriptive study did not assess outcomes based on laboratory results. Sixth, our study included only adults ≥ 60 years and may not be generalizable to younger patients. Finally, our population was predominately from academic medical centers, possibly resulting in higher medical complexity and comorbidities than the general population. In conclusion, in a large cohort of adults ≥ 60 years of age with syncope or near syncope, laboratory testing with CBC and BMP commonly identified nonnormal laboratory values and identified critical laboratory values in one in 25 patients. Data Supplement S1. Differences in Characteristics of Cohort. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.