Abstract 13585: Red Flag Comorbid Conditions That Are Easily Extracted From the Electronic Health Record Often Precede Diagnosis of Cardiac Amyloidosis by Years

Abstract

Introduction: With new and emerging disease-modifying therapies for cardiac amyloidosis (CA), there is a growing impetus for improved awareness and early diagnosis of this condition. Several groups have identified comorbid conditions, termed ‘red flag’ systemic symptoms and diagnostic findings, that occur commonly in patients with CA. To better understand the prevalence and temporal relationship of these ‘red flag’ comorbid conditions to a CA diagnosis, we used automated extraction of clinical data stored in the electronic health record (EHR) at a single institution. Methods: Patients with confirmed light-chain (AL) and transthyretin (ATTR) cardiac amyloidosis were identified at a single institution. The dates of diagnosis of ‘red flag’ comorbid conditions and for CA were extracted from the EHR via ICD-9 and ICD-10 codes, electrocardiogram intervals, and echocardiogram reports. Frequency and time between the diagnoses of red flags and CA were assessed. Results: Of 175 CA patients (85% ATTR), mean age was 72.7 years with majority (76.6%) male patients. Patients were most commonly black (52.6%) followed by Caucasian (42.3%). A diagnosis code for amyloidosis was found in 89.1% of patients, followed by heart failure in 84.6% of patients. ‘Red flag’ comorbid conditions were more likely to have been diagnosed before rather than after CA. When the ‘red flag’ was diagnosed prior to CA, the greatest average intervals between diagnoses were seen in carpal tunnel syndrome (8.3 years), spinal stenosis (5.9 years), and atrial fibrillation (2.8 years) (Table 1). Conclusions: EHR data from a single institution demonstrates that the diagnosis of CA is often delayed following the diagnosis of ‘red flag’ comorbid conditions. Our automated extraction of EHR data provides a foundational step in the ongoing development of an EHR-based warning score aimed at improving clinician awareness of the possibility of CA in appropriate patients.