90. Deimplementation: Use of Electronic Clinical Decision Support to Reduce Unnecessary Erythrocyte Sedimentation Rate (ESR) Ordering

Background In recent years, several de-implementation initiatives have focused on diagnostic testing. One such initiative, the Choosing Wisely campaign, recommends against routine use of erythrocyte sedimentation rate (ESR) for assessment of acute undiagnosed inflammation or infection. With the development of newer biomarkers of inflammation, particularly C-reactive protein (CRP), there is a decreasing role for ESR in screening for acute-onset conditions; however, ESR continues to be commonly ordered. Methods We examined ESR and CRP ordering practices at the Children’s Hospital of Philadelphia (CHOP) from July 2019 to July 2020 and found that 80% of ESR orders were placed concurrently with an order for CRP. We aimed to reduce ESR ordering by 20% at CHOP by using electronic clinical decision support in the form of embedded order guidance for ESR orders placed in the Emergency Department (ED) and inpatient setting. We examined the effect of the clinical decision support by assessing ESR ordering rate, defined by ESR orders per monthly patient days for the inpatient setting and ESR orders per monthly ED visits for the ED setting. We then examined differences in ordering rates using a quasi-experimental model with a concurrent control (basic metabolic panel). ESR Electronic Clinical Decision Support Intervention Inpatient and ED versions of the embedded electronic clinical decision support for ESR orders. Results Prior to implementation of the electronic decision support intervention, the median monthly rate of ESR orders was 13.6 per 1000 patient days and 70.3 per 1000 ED visits. During the initial month after implementation, we found that ESR ordering was 12.5 in inpatient and 46.4 in ED, reflecting decreased rates of ordering. The median monthly rate of basic metabolic panel orders (concurrent control) was 194.5 per 1000 patient days and 110.0 per 1000 ED visits. This was unchanged after intervention. Conclusion We conclude that electronic clinical decision support is a potentially effective deimplementation method for improving diagnostic test utilization, even with non-disease specific testing such as inflammatory markers. However, continued post-implementation data monitoring and analysis is needed to determine if this is a true difference and sustainable trend. Disclosures All Authors: No reported disclosures

The multicenter European Biological Variation Study (EuBIVAS): a new glance provided by the Principal Component Analysis (PCA), a machine learning unsupervised algorithms, based on the basic metabolic panel linked measurands

Objectives The European Biological Variation Study (EuBIVAS), which includes 91 healthy volunteers from five European countries, estimated high-quality biological variation (BV) data for several measurands. Previous EuBIVAS papers reported no significant differences among laboratories/population; however, they were focused on specific set of measurands, without a comprehensive general look. The aim of this paper is to evaluate the homogeneity of EuBIVAS data considering multivariate information applying the Principal Component Analysis (PCA), a machine learning unsupervised algorithm. Methods The EuBIVAS data for 13 basic metabolic panel linked measurands (glucose, albumin, total protein, electrolytes, urea, total bilirubin, creatinine, phosphatase alkaline, aminotransferases), age, sex, menopause, body mass index (BMI), country, alcohol, smoking habits, and physical activity, have been used to generate three databases developed using the traditional univariate and the multivariate Elliptic Envelope approaches to detect outliers, and different missing-value imputations. Two matrix of data for each database, reporting both mean values, and “within-person BV” (CVP) values for any measurand/subject, were analyzed using PCA. Results A clear clustering between males and females mean values has been identified, where the menopausal females are closer to the males. Data interpretations for the three databases are similar. No significant differences for both mean and CVPs values, for countries, alcohol, smoking habits, BMI and physical activity, have been found. Conclusions The absence of meaningful differences among countries confirms the EuBIVAS sample homogeneity and that the obtained data are widely applicable to deliver APS. Our data suggest that the use of PCA and the multivariate approach may be used to detect outliers, although further studies are required.

Budget Constrained Machine Learning for Early Prediction of Adverse Outcomes for COVID-19 Patients

Background: Machine learning (ML) based risk stratification models of Electronic Health records (EHR) data may help to optimize treatment of COVID-19 patients, but are often limited by their lack of clinical interpretability and cost of laboratory tests. We develop a ML based tool for predicting adverse outcomes based on EHR data to optimize clinical utility under a given cost structure. This cohort study was performed using deidentified EHR data from COVID-19 patients from ProMedica Healthcare in northwest Ohio and southeastern Michigan. Methods: We tested performance of various ML approaches for predicting either increasing ventilatory support or mortality and the set of model features under a budget constraint was optimized via exhaustive search across all combinations of features. Results: The optimal sets of features for predicting ventilation under any budget constraint included demographics and comorbidities (DCM), basic metabolic panel (BMP), D-dimer, lactate dehydrogenase (LDH), erythrocyte sedimentation rate (ESR), CRP, brain natriuretic peptide (BNP), and procalcitonin and for mortality included DCM, BMP, complete blood count, D-dimer, LDH, CRP, BNP, procalcitonin and ferritin. Conclusions: This study presents a quick, accurate and cost-effective method to evaluate risk of deterioration for patients with SARS-CoV-2 infection at the time of clinical evaluation.

Is It Time to Remove Total Calcium from the Basic and Comprehensive Metabolic Panels? Assessing the Effects of American Medical Association–Approved Chemical Test Panels on Laboratory Utilization

Background The necessity of individual tests within the most commonly used disease-oriented test panels has not been well established. We evaluated test-ordering practices for total calcium, both before and after implementation of American Medical Association (AMA)–approved panels (basic metabolic panel [BMP] and comprehensive metabolic panel [CMP]) in our electronic ordering system. Methods We performed a retrospective review of all total calcium orders placed during April and June 2018, before and after implementation of the panels. Orders from inpatient, outpatient, and emergency department (ED) care units were totaled, and the percentage of abnormal test results was calculated. We then queried institutional databases to determine the number of unique patients with calcium-related diagnoses and compared the rates from a 5-month period both before and after implementation of the panels. Results Total test volumes and tests per unique patient increased by more than 3-fold after implementation of calcium-containing AMA-approved panels, with the majority of those orders coming from BMPs and CMPs. The rate of low calcium values increased because of the shift toward more inpatient testing; however, the percentage of abnormal results within each patient population (inpatient, outpatient, ED) decreased. The prevalence of hypo- and hypercalcemia-related diagnoses among patients in the 5 months after implementation did not change significantly (1.29% before implementation vs 1.27% after implementation). Conclusions Implementation of BMPs and CMPs dramatically increased total calcium testing volumes without changing the rate of calcium-related diagnoses. The results suggest that the increase in total calcium orders associated with panel-based testing largely constitutes excess or unnecessary testing.

Pseudohyperkalemia: Hyperkalemia Cocktail or Alternative Diagnosis

Introduction. Hyperkalemia is a commonly encountered clinical problem. Pseudohyperkalemia is believed to be an in vitro phenomenon that does not reflect in vivo serum potassium and therefore should not be treated. Here, we present a case who unfortunately underwent unnecessary treatment because of failure to detect the common lab abnormality of pseudohyperkalemia. Case Presentation. A 91-year-old female with a history of chronic lymphocytic leukemia presented to the emergency with nausea and vomiting 24 hours after her first chemotherapy with chlorambucil. Physical examination was overall unremarkable. She had a leukocytosis of 210 × 103/µL with 96% lymphocytes along with chronic anemia with hemoglobin of 8.1 g/dL. Her initial sodium and potassium levels were normal. During the clinical course, her potassium progressively worsened and failed to improve despite standard medical treatment. Patient ultimately underwent dialysis. Conclusions. Differentiating true hyperkalemia from pseudohyperkalemia is very important in selected group of patients to avoid unnecessary medications, higher level of care, and unnecessary procedure including dialysis. We want to emphasize the importance of simple yet profound knowledge of technique of blood draws and basic metabolic panel processing for every clinician in day-to-day practice.

Pseudohyperkalemia in a Patient with T-Cell Acute Lymphoblastic Leukemia and Hyperleukocytosis

A 7-year-old girl presented with lymphadenopathy and bruising suggestive of leukemia. Complete blood count was significant for white blood cell count of 479,000/mm3. Basic metabolic panel sent via pneumatic tube system was significant for potassium > 10 mEq/L. The stat venous blood gas potassium level was 4.6 mEq/L. A 12-lead-ECG showed sinus tachycardia without peaked T-waves. It was determined that this was pseudohyperkalemia associated with significant hyperleukocytosis. This brief report discusses rare causes of pseudohyperkalemia that can be overlooked, and details the postulated mechanisms for pseudohyperkalemia in the setting of hyperleukocytosis.

Topiramate-Induced Chest Pain: A Case Report

Background: Topiramate, an anticonvulsant used for prophylaxis of migraines and epilepsy, is commonly associated with adverse effects of cognitive dulling and fatigue. Chest pain is a potential adverse effect that to our knowledge has not been reported with the use of topiramate. Case Presentation: We present the case of a 38-year-old female with a seizure disorder who experienced chest pain after the first dose of topiramate. On day 1, she presented to the emergency department, was admitted, and over the course of 3 days had a chest X-ray, electrocardiogram (ECG), and echocardiogram, and her vitals, basic metabolic panel, complete blood counts, troponin, and d-dimer levels were monitored. The chest pain improved when the topiramate was held. No identifiable causes of chest pain were apparent, other than the topiramate. Discussion: The Naranjo probability scale was utilized to determine the causality of topiramate. The resulting score of 3 indicates that it is possible that the chest pain was due to the topiramate. Conclusion: This report demonstrates an example of a patient who experienced chest pain possibly caused by the initiation of topiramate. The objective of this case report is to increase the awareness of chest pain as an adverse effect of topiramate.

Gemcitabine associated pseudocellulitis: A missed diagnosis

We report a case of a 51-year-old male with past medical history significant for cholangiocarcinoma presented with two weeks of worsening bilateral lower extremity swelling and erythema. Patient has been on active chemotherapy for his cholangiocarcinoma with Gemcitabine weekly infusions. Physical exam was significant for bilaterally petechial rash coalescing into ecchymoses over the dorsum of the feet, sparing soles and toes, which dissipated into thinning petechiae more proximally. On labs he did not have any leukocytosis, his platelet count was 50 × 103/µL and basic metabolic panel was benign. Patient was started on Vancomycin for presumed cellulitis. Lower extremity ultrasound Doppler ruled out deep venous thrombosis. Patient did have biopsies bilaterally on his legs, which showed hypersensitivity reaction consistent with the diagnosis of pseudocellulitis. His Vancomycin was discontinued and his symptoms improved. Our case further supports that pseudocellulitis is underrecognized and underreported, potentially leading to unnecessary antibiotic exposure and unnecessary diagnostic work-up as seen unfortunately in our patient. Unnecessary antibiotic exposure is increasing the risk for clostridium difficile and or antibiotic resistance, therefore awareness of this reaction is critical, as to avoid unnecessary antibiotics, and costly diagnostic workups.