Transthyretin Amyloid Cardiomyopathy Risk Evaluation in a Cohort of Patients With Heart Failure.

Introduction

Transthyretin amyloid cardiomyopathy (ATTR-CM) is a progressive, infiltrative form of heart failure (HF) that occurs when soluble transthyretin proteins, responsible for transporting retinol and thyroxine, misfold and deposit in the myocardium as insoluble amyloid fibrils. 1,2 Accumulation of these fibrils leads to left ventricular wall thickening and impaired diastolic filling. 3 Moreover, dysrhythmias caused by amyloid deposits can further impair diastolic filling, resulting in reduced stroke volume with elevated intracardiac pressures. 1,2

The exact prevalence of ATTR-CM is unknown, as the condition is largely underrecognized and often misdiagnosed. 4 Recent data suggest that ATTR-CM is more common than previously thought and increasing in prevalence because of a growing older population. 2 In a convenience sample of 120 Spanish patients aged 60 years and older admitted for HF with preserved ejection fraction (HF p EF) and left ventricular hypertrophy, 13% were identified with ATTR-CM. 5 Among 151 US patients undergoing transcatheter aortic valve replacement, 16% were identified with ATTR-CM. 6

Early diagnosis and treatment (eg, with tafamidis or tafamidis meglumine 7 ) of ATTR-CM is essential, as survival is short and mortality is high in untreated patients. 8 Several imaging procedures (eg, technetium-99m pyrophosphate cardiac imaging, 9 speckle-tracking echocardiography, 10 and cardiac magnetic resonance 11 ) have been used for targeted testing in patients believed more likely to have ATTR-CM. In addition, heart tissue biopsy can be used to confirm amyloid deposits. 12 More recently, genetic testing for a mutation in the TTR gene has been used to detect inherited ATTR-CM (hATTR-CM). 13 Nevertheless, the diagnosis of ATTR-CM continues to remain a challenge for clinicians because of the presence of nonspecific symptoms, more commonly diagnosed comorbidities, and a lack of awareness of ATTR-CM. 14

To identify phenotypes of patients with wild-type ATTR-CM (wtATTR-CM), Huda and colleagues derived and validated a machine learning model (MLM) using International Classification of Diseases codes (9th [ICD-9] and 10th [ICD-10] revisions), demographics, and health care utilization information from pooled US medical claims and electronic health record (EHR) data. 15 In a matched cohort of 1071 patients with and 1071 patients without diagnosed wtATTR-CM, the MLM algorithm from EHR data performed well in predicting patients with diagnosed wtATTR-CM (87% sensitivity, 88% positive predictive value) and identified patient characteristics not previously reported in the literature. 15 Although Huda and colleagues provided a methodology for the identification of wtATTR-CM, the MLM identified approximately 50 characteristics related to ATTR-CM, patients with hATTR-CM were not included in the MLM algorithm, and the patients in the MLM who were designated as having wtATTR-CM were identified only with a new and unvalidated diagnosis code (ie, patients did not have targeted ATTR-CM testing results). 15

Streamlining of the MLM algorithm in a real-world setting of patients with and without confirmed hATTR-CM and wtATTR-CM is needed to improve the ability to screen a large population of patients with HF. The purpose of this study was to use Huda and colleagues’ MLM algorithm 15 and supplement it with additional EHR data to develop a more efficient model for assessing the probability of ATTR-CM among patients with HF in a real-world clinical setting. A practical model for systematic screening of administrative data could allow for early identification of younger patients with a high probability of ATTR-CM for referral to the appropriate specialists.

Materials and Methods

Study design and setting

This report describes an observational cohort study of patients with HF from Kaiser Permanente Colorado. Patients diagnosed with confirmed hATTR-CM/wtATTR-CM (observation group) or with HF but without confirmed ATTR-CM (control group) between January 1, 2019, and July 1, 2021, were included. Patient demographic, cardiac, noncardiac, prescription medication, laboratory, health care utilization, and risk score characteristics were extracted from administrative and claims data and compared between the groups. A propensity to have ATTR-CM score was modeled. Samples of 50 control patients with the highest and lowest propensity scores were identified and combined in a dataset. Additional EHR data were collected for these patients through manual chart review. A board-certified cardiologist with experience diagnosing and treating ATTR-CM (MYJ) who was blinded to ATTR-CM status reviewed the dataset and adjudicated whether further workup was warranted for each patient. The index date was defined as the date of ATTR-CM confirmation for the observation group and as the date of the HF diagnosis entry in the EHR for the control group.

Kaiser Permanente Colorado is a not-for-profit, integrated health care delivery system based in the United States that provides comprehensive medical services. Kaiser Permanente Colorado uses an EHR system (EPIC Systems Corp) in which pharmacy, office visit, laboratory, procedure, radiology, and practitioner encounter data are housed. In addition, coded and free-text emergency department, hospitalization, and membership information from within the delivery system, as well as from other contracted and affiliated facilities, are captured in Kaiser Permanente’s administrative and claims databases. All aspects of this study were reviewed and approved by Kaiser Permanente Colorado’s Institutional Review Board (IRB) before data collection (approval no. 1818323-1). The IRB approved a waiver of the requirement to obtain informed consent and a waiver of the requirement to obtain HIPAA Privacy Rule Authorization for the use of protected health information because of the retrospective nature of the study.

Patient population

Adults aged 18 years or older with a diagnosis of confirmed ATTR-CM or HF (ICD-10 codes available upon request) and had at least 180 days of Health Plan membership prior to the index date were included. Patients who had a hospice stay during the 6 months before or 3 months after the index date or were pregnant during the 9 months before the index date or became pregnant during the 3 months after the index date were excluded. ATTR-CM diagnoses were confirmed with targeted testing including technetium-99m pyrophosphate cardiac imaging, serum protein electrophoresis, serum-free light chains, and/or cardiac biopsy. Genetic testing for TTR mutation was performed as warranted.

Study outcomes

The primary outcome was a description and comparison of patient characteristics between the groups. Secondary outcomes included a comparison of patient characteristics from the propensity model between control patients who had high and low probabilities of ATTR-CM. In addition, the sensitivity and specificity of the propensity score to differentiate between patients without confirmed ATTR-CM who were candidates for further workup for ATTR-CM were calculated.

Data collection

Patients with ATTR-CM were identified from the Kaiser Permanente Colorado cardiology registry. Administrative electronic medical, pharmacy, membership, and laboratory records were used to identify control patients and patient characteristics. Patient information was collected during the 60 months before and 6 months after the index date. From the top clinical characteristics predictive of wtATTR-CM reported by Huda and colleagues, 15 ICD-9 and ICD-10 codes of these characteristics were assembled (codes available upon request) and used to extract data from administrative and claims databases.

Supplemental data to inform adjudication of the control patients with high and low propensity scores were collected from manual chart review of the EHR. These data included the following: 1) counts of HF hospitalizations during the 5 years prior to the index date; 2) primary or secondary diagnosis of HF during hospitalization; 3) status of diuretic use during hospitalization; 4) presence of left ventricular hypertrophy on most recent echocardiogram; 5) left ventricular hypertrophy size and grade (ie, severity of diastolic dysfunction); 6) presence of pericardial effusion on echocardiogram during the 5 years prior to the index date; 7) loop diuretic and metolazone dispensing statuses; 8) highest levels of B-type natriuretic peptide (BNP), N-terminal prohormone BNP (NT-proBNP), and troponin and lowest estimated glomerular filtration rate during the 5 years prior to index date; 9) type(s) of joint disorders (eg, arthritis, arthropathy, gout, degenerative disc disease); 10) spinal stenosis status; and 11) type(s) of refraction/accommodation disorder (eg, myopia, astigmatism, presbyopia, hyperopia, pseudophakia). Control patients who were to be adjudicated but were subsequently identified as deceased after the index date were not adjudicated owing to IRB determinations regarding protection of the confidentiality of the data.

Data analysis

Age was defined as of the index date. Laboratory values of BNP ≥ 150 pg/mL and troponin > 0.05 ng/mL (using fourth-generation sensitivity analysis) were categorized as elevated. A chronic disease score, which is an indicator of a patient’s burden of chronic illness and a predictor of future health care utilization, was calculated from each patient’s baseline prescription dispensing history. 16 The presence of specific comorbidities was determined using the Quan adaptation of the Charlson comorbidity index. 17 Other comorbidities were determined from Huda and colleagues’ MLM algorithm. 15

The propensity score was determined using logistic regression modeling of confirmed ATTR-CM status on Black race, diuretic use, and elevated BNP level, as well as nonvalvular atrial fibrillation/flutter, cardiomegaly, carpal tunnel syndrome, chronic kidney disease, heart block, HF p EF, joint disorders, lumbar spinal stenosis, neuropathy, osteoarthritis, pericardial effusion, refraction and accommodation disorder, and synovitis and tenosynovitis comorbidities. Characteristics included in the model were chosen based on clinical judgment and statistical analysis. The model’s c-statistic was 0.875.

Patient characteristics are reported as means and standard deviations for continuous data (eg, age) and proportions for nominal data (eg, sex) and were compared with t-tests/nonparametric equivalents, as appropriate, and χ 2 tests of association, respectively. The adjudication of patients who were candidates for further workup used the collected administrative and claims data supplemented with the collected manual chart review data. Adjudication was based on clinical judgment. The proportion of control patients with a high probability of ATTR-CM who were determined to be candidates for further workup is reported as the sensitivity with Wald 95% confidence interval (CI). The proportion of control patients with a low probability of ATTR-CM who were determined to not be candidates for further workup is reported as the specificity with Wald 95% CI. The α value was set at 0.05.

Data sharing

The data used for this study contain protected health information; thus, individual-level data may not be made publicly available because of IRB, business, and privacy concerns. Data are available for researchers who meet the criteria for access to confidential data. Please contact the Kaiser Permanente Colorado IRB (kpco.irb@kp.org) for more information.

Results

A total of 7651 patients were included in the study, with 31 patients having confirmed ATTR-CM and 7620 patients having HF without known ATTR-CM ( Figure 1 ). In the overall cohort, patients were predominantly older and white and had a high burden of chronic disease ( Table 1 ). Patients in the ATTR-CM group were older (p = 0.028) more likely to be male (p = 0.002) and Black (p = 0.001, Table 1 ).

Patients with ATTR-CM had higher prevalences of cardiac features including cardiac amyloidosis (p < 0.001), cardiomegaly (p = 0.038), HF p EF (p = 0.014), light-chain amyloidosis (p = 0.044), and premature beats (p = 0.034, Table 1 ). In addition, patients with ATTR-CM had higher prevalences of noncardiac features including ascites (p = 0.016), carpal tunnel syndrome (p < 0.001), joint disorders (p = 0.019), lumbar spinal stenosis (p = 0.049), multiple myeloma (p < 0.001), nontraumatic bicep rupture (p = 0.046), osteoarthritis (p = 0.002), refraction and accommodation disorder (p = 0.020), and synovitis and tenosynovitis (p < 0.001). Furthermore, patients with ATTR-CM were more likely to have had an elevated BNP level (p < 0.001) and diuretic use (p < 0.001). Patients in the control group were more likely to have had a history of myocardial infarction (p = 0.046) and an HF-related emergency department visit during the 365 days prior to the index date (p = 0.008).

Among the 50 control patients with the highest probabilities of requiring further workup from the propensity score model and the 50 control patients with the lowest such probabilities, high-probability patients were more likely to be Black (p < 0.001) and to have had atrial flutter/fibrillation (p = 0.027), cardiomegaly(p < 0.001), HF p EF (p < 0.001), pericardial effusion (p < 0.001), carpal tunnel syndrome (p < 0.001), chronic kidney disease (p < 0.001), joint disorders (p < 0.001), lumbar spinal stenosis (p < 0.001), osteoarthritis (p < 0.001), refraction and accommodation disorder (p < 0.001), and synovitis and tenosynovitis (p < 0.001) comorbidities ( Table 2 ). In addition, high-probability patients were more likely to have had an elevated BNP level and diuretic use (both p < 0.001). Low-probability patients were more likely to have had a heart block comorbidity (p < 0.001).

Of the 100 control patients to be adjudicated, 18 (36.0%) and 7 (14.0%) with the highest and lowest probabilities, respectively, were deceased after the index date (p = 0.011). The sensitivity of the propensity model was 71.9% (23/32; 95% CI 54.6–85.3%) and the specificity was 93.0% (40/43; 95% CI 82.2–98.2%).

Discussion

This retrospective cohort study of patients with and without confirmed ATTR-CM identified 3 demographic characteristics, 4 cardiac-related characteristics, 9 non-cardiac-related characteristics, 1 laboratory characteristic, and 1 prescription medication characteristic associated with ATTR-CM. In addition, a propensity score for having confirmed ATTR-CM developed from 16 clinical and demographic characteristics was found to distinguish well between control patients with high and low probabilities of having ATTR-CM. Furthermore, the propensity score developed in this work had a good ability to distinguish patients who warranted further workup for ATTR-CM and a very good ability to distinguish those who did not. These findings are noteworthy as the study included patients with wtATTR-CM and hATTR-CM, identified novel characteristics (eg, diuretic use) and reinforced previously reported characteristics (eg, carpal tunnel syndrome) associated with ATTR-CM, and developed a more efficient model for detecting patients with a high probability of ATTR-CM who are candidates for further workup.

The findings of this study confirm previous investigations that identified carpal tunnel syndrome, 18 lumbar spinal stenosis, 18–20 nontraumatic bicep rupture, 21 and synovitis and tenosynovitis 22 as being associated with ATTR-CM. The 14 characteristics identified in this study mirror those reported by Huda and colleagues, although this study did not identify a number of characteristics (eg, atrial flutter/fibrillation, HF with reduced ejection fraction) that Huda and colleagues reported as being associated with ATTR-CM. 15 The identification of the associations of Black race and diuretic use with ATTR-CM in this work adds further information on potential characteristics leading to an increased chance of ATTR-CM.

That Black race was identified as being associated with ATTR-CM was not surprising, as all 3 of the cohorts in Huda and colleagues’ work for which patient race was reported had numerically higher proportions of “Black ethnicity” in the patients with diagnosed wtATTR-CM. 15 In addition, a hereditary form of transthyretin cardiac amyloidosis has been identified in Black individuals. 23 Although diuretics are a basic treatment for HF, there is a dearth of clinical trial data on their proper use. 24 That patients with ATTR-CM in this study were more likely to have been using a diuretic may have been because they were receiving more specialized cardiac care whereby prescribing of a diuretic was commonly practiced in response to volume overload from HF.

The performance of Huda and colleagues’ MLM algorithm 15 in predicting wtATTR-CM was evaluated in a retrospective study at 2 academic centers, and an accuracy of approximately 83% was reported. 25 Although such results support the use of the algorithm, its value in a nonacademic, real-world clinical setting is unclear, as the algorithm uses a large count of inputs that may not be gleaned readily from administrative and/or claims databases. An ATTR-CM predictive model reported recently by Davies and colleagues used 3 clinical characteristics (age, male sex, and hypertension diagnosis) and 3 echocardiographic characteristics (ejection fraction, posterior wall thickness, and relative wall thickness) to predict an increased chance of ATTR-CM in patients with HF p EF who consented to participate in the study. 26 This model had an accuracy of 89% but relied on clinical characteristics to be abstracted from medical charts. 26

Inputs relying upon ICD-9 and ICD-10 codes for identification can be problematic as incorrect or missing coding of events could lead to under- or overestimations of the true prevalences. An example of this is that 16% of the patients in this study with confirmed ATTR-CM did not have a diagnosis of cardiac amyloidosis documented in the EHR. The use of propensity score modeling reduced the count of inputs to 16 demographic, diagnosis, laboratory, and pharmacy inputs without the need for medical record abstraction. This model may provide a more simplified and efficient means of identifying patients with HF who are more likely to have ATTR-CM and would warrant further workup. Further investigation with actual targeted testing of patients with high probabilities of ATTR-CM is required to validate the model proposed in this work so as to develop a more refined predictive model.

Once thought to be a rare condition, ATTR-CM is increasingly recognized as an important cause of HF, particularly HF p EF. 27 Unfortunately, ATTR-CM diagnostic delays can be multiple years or indefinite. 14 These delays can lead to disease progression, lack of or inappropriate treatment, and/or higher health care utilization in many patients. 28 Screening and targeted testing to enhance early diagnosis is critical, as improper or delayed diagnosis can have prognostic and therapeutic implications. Ensuring appropriate and timely diagnosis and treatment, especially in advanced disease, is paramount to allow patients to benefit from earlier access to new disease-modifying therapies. 7 The findings from this study build on and further the exploration of approaches for identifying patients who are more likely to have ATTR-CM and suggest that efforts to develop more refined methods will have a substantial impact on making early diagnosis of ATTR-CM practical.

This study was strengthened by the use of patients with confirmed wtATTR-CM/hATTR-CM in addition to a sample of over 7500 patients with HF. Administrative and claims data were used to identify patients with high and low probabilities of ATTR-CM, and those data were then supplemented with medical chart reviews to identify additional key characteristics that allowed for the adjudication of the merit of targeted testing. This study, however, had limitations. Only 31 patients with confirmed ATTR-CM were available for use in the comparisons and modeling. Although small, this sample size of patients with ATTR-CM is typical in real-world studies, 25 and it was sufficient for the construction of a propensity model with an excellent goodness of fit of the data. Only one referee was used to adjudicate control patients who warranted target testing; however, he was blinded patients’ probability of ATTR-CM status and was a board-certified cardiologist with experience diagnosing and treating ATTR-CM. Not all patients had laboratory values recorded. Although incomplete laboratory testing is typical in retrospective analyses, this difficulty might have been compounded by the COVID-19 pandemic. This study was conducted in one integrated health care delivery system, other institutions may have different experiences.

Conclusions

This retrospective study of patients with and without confirmed ATTR-CM identified 16 demographic, diagnostic, laboratory, and pharmacy inputs from electronic databases that could be used to detect patients with HF who are good candidates for targeted ATTR-CM testing. This study built upon previous modeling to identify novel characteristics that may increase suspicion of ATTR-CM and contribute to the growing body of literature characterizing the potential risk factors associated with ATTR-CM. Future studies should expand upon this methodology to incorporate different health systems and larger samples of patients with confirmed ATTR-CM to develop more refined predictive tools for targeted ATTR-CM testing.

Acknowledgments

This research did not receive any specific grants from funding agencies in the public, commercial, or nonprofit sectors. Preliminary results were presented at the 2022 Residency Conference of the Rockies, Estes Park, CO, on June 1, 2022.

Funding: None declared

Conflicts of Interest: None declared

Author Contributions: Angie A Suh, PharmD, designed the research, extracted information from medical charts, interpreted the analysis, drafted the initial version of the manuscript, and revised the manuscript. Paul B Shaw, PharmD, designed the research, interpreted the analysis, and revised the manuscript. Mark Y Jeong, MD, designed the research, interpreted the analysis, and revised the manuscript. Kari L Olson, PharmD, designed the research, interpreted the analysis, and revised the manuscript. Thomas Delate, PhD, designed the research, extracted information from electronic data sources, performed the statistical analysis, interpreted the analysis, and revised the manuscript. All authors approved the version of the manuscript for submission.

References

1. Yamamoto H , Yokochi T . Transthyretin cardiac amyloidosis: An update on diagnosis and treatment . ESC Heart Fail . 2019 ; 6 ( 6 ): 1128 – 1139 . 10.1002/ehf2.12518 31553132 PMC6989279

2. Ruberg FL , Grogan M , Hanna M , Kelly JW , Maurer MS . Transthyretin amyloid cardiomyopathy: JACC state-of-the-art review . J Am Coll Cardiol . 2019 ; 73 ( 22 ): 2872 – 2891 . 10.1016/j.jacc.2019.04.003 31171094 PMC6724183

3. Rapezzi C , Merlini G , Quarta CC , et al . Systemic cardiac amyloidoses: Disease profiles and clinical courses of the 3 main types . Circulation . 2009 ; 120 ( 13 ): 1203 – 1212 . 10.1161/CIRCULATIONAHA.108.843334 19752327

4. Donnelly JP , Hanna M . Cardiac amyloidosis: An update on diagnosis and treatment . Cleve Clin J Med . 2017 ; 84 ( 12 Suppl 3 ): 12 – 26 . 10.3949/ccjm.84.s3.02 29257735

5. González-López E , Gallego-Delgado M , Guzzo-Merello G , et al . Wild-type transthyretin amyloidosis as a cause of heart failure with preserved ejection fraction . Eur Heart J . 2015 ; 36 ( 38 ): 2585 – 2594 . 10.1093/eurheartj/ehv338 26224076

6. Castaño A , Narotsky DL , Hamid N , et al . Unveiling transthyretin cardiac amyloidosis and its predictors among elderly patients with severe aortic stenosis undergoing transcatheter aortic valve replacement . Eur Heart J . 2017 ; 38 ( 38 ): 2879 – 2887 . 10.1093/eurheartj/ehx350 29019612 PMC5837725

7. Maurer MS , Schwartz JH , Gundapaneni B , et al . Tafamidis treatment for patients with transthyretin amyloid cardiomyopathy . N Engl J Med . 2018 ; 379 ( 11 ): 1007 – 1016 . 10.1056/NEJMoa1805689 30145929

8. Gillmore JD , Damy T , Fontana M , et al . A new staging system for cardiac transthyretin amyloidosis . Eur Heart J . 2018 ; 39 ( 30 ): 2799 – 2806 . 10.1093/eurheartj/ehx589 29048471

9. Castano A , Haq M , Narotsky DL , et al . Multicenter study of planar technetium 99m pyrophosphate cardiac imaging: Predicting survival for patients with ATTR cardiac amyloidosis . JAMA Cardiol . 2016 ; 1 ( 8 ): 880 – 889 . 10.1001/jamacardio.2016.2839 27557400

10. Phelan D , Collier P , Thavendiranathan P , et al . Relative apical sparing of longitudinal strain using two-dimensional speckle-tracking echocardiography is both sensitive and specific for the diagnosis of cardiac amyloidosis . Heart . 2012 ; 98 ( 19 ): 1442 – 1448 . 10.1136/heartjnl-2012-302353 22865865

11. Martinez-Naharro A , Treibel TA , Abdel-Gadir A , et al . Magnetic resonance in transthyretin cardiac amyloidosis . J Am Coll Cardiol . 2017 ; 70 ( 4 ): 466 – 477 . 10.1016/j.jacc.2017.05.053 28728692

12. Maurer MS , Elliott P , Comenzo R , Semigran M , Rapezzi C . Addressing common questions encountered in the diagnosis and management of cardiac amyloidosis . Circulation . 2017 ; 135 ( 14 ): 1357 – 1377 . 10.1161/CIRCULATIONAHA.116.024438 28373528 PMC5392416

13. Brown EE , Lee YZJ , Halushka MK , et al . Genetic testing improves identification of transthyretin amyloid (ATTR) subtype in cardiac amyloidosis . Amyloid . 2017 ; 24 ( 2 ): 92 – 95 . 10.1080/13506129.2017.1324418 28494620

14. Rozenbaum MH , Large S , Bhambri R , et al . Impact of delayed diagnosis and misdiagnosis for patients with transthyretin amyloid cardiomyopathy (ATTR-CM): A targeted literature review . Cardiol Ther . 2021 ; 10 ( 1 ): 141 – 159 . 10.1007/s40119-021-00219-5 33877591 PMC8126532

15. Huda A , Castaño A , Niyogi A , et al . A machine learning model for identifying patients at risk for wild-type transthyretin amyloid cardiomyopathy . Nat Commun . 2021 ; 12 ( 1 ): 2725 . 10.1038/s41467-021-22876-9 33976166 PMC8113237

16. Clark DO , Von Korff M , Saunders K , Baluch WM , Simon GE . A chronic disease score with empirically derived weights . Med Care . 1995 ; 33 ( 8 ): 783 – 795 . 10.1097/00005650-199508000-00004 7637401

17. Quan H , Sundararajan V , Halfon P , et al . Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data . Med Care . 2005 ; 43 ( 11 ): 1130 – 1139 . 10.1097/01.mlr.0000182534.19832.83 16224307

18. Aus dem Siepen F , Hein S , Prestel S , et al . Carpal tunnel syndrome and spinal canal stenosis: Harbingers of transthyretin amyloid cardiomyopathy? Clin Res Cardiol . 2019 ; 108 ( 12 ): 1324 – 1330 . 10.1007/s00392-019-01467-1 30953182

19. Carr AS , Shah S , Choi D , et al . Spinal stenosis in familial transthyretin amyloidosis . J Neuromuscul Dis . 2019 ; 6 ( 2 ): 267 – 270 . 10.3233/JND-180348 30856118

20. Westermark P , Westermark GT , Suhr OB , Berg S . Transthyretin-derived amyloidosis: Probably a common cause of lumbar spinal stenosis . Ups J Med Sci . 2014 ; 119 ( 3 ): 223 – 228 . 10.3109/03009734.2014.895786 24620715 PMC4116761

21. Geller HI , Singh A , Alexander KM , Mirto TM , Falk RH . Association between ruptured distal biceps tendon and wild-type transthyretin cardiac amyloidosis . JAMA . 2017 ; 318 ( 10 ): 962 – 963 . 10.1001/jama.2017.9236 28898370 PMC5818850

22. Sperry BW , Reyes BA , Ikram A , et al . Tenosynovial and cardiac amyloidosis in patients undergoing carpal tunnel release . J Am Coll Cardiol . 2018 ; 72 ( 17 ): 2040 – 2050 . 10.1016/j.jacc.2018.07.092 30336828

23. Quarta CC , Buxbaum JN , Shah AM , et al . The amyloidogenic V122I transthyretin variant in elderly black Americans . N Engl J Med . 2015 ; 372 ( 1 ): 21 – 29 . 10.1056/NEJMoa1404852 25551524 PMC4382209

24. Felker GM , Ellison DH , Mullens W , Cox ZL , Testani JM . Diuretic therapy for patients with heart failure . J Am Coll Cardiol . 2020 ; 75 ( 10 ): 1178 – 1195 . 10.1016/j.jacc.2019.12.059 32164892

25. Heitner S , Elman MR , Masri A , et al . Performance evaluation of a machine learning model for systematic identification of wild-type transthyretin amyloid cardiomyopathy at two academic medical centers . Journal of Cardiac Failure . 2020 ; 26 ( 10 ). 10.1016/j.cardfail.2020.09.114

26. Davies DR , Redfield MM , Scott CG , et al . A simple score to identify increased risk of transthyretin amyloid cardiomyopathy in heart failure with preserved ejection fraction . JAMA Cardiol . 2022 ; 7 ( 10 ): 1036 – 1044 . 10.1001/jamacardio.2022.1781 36069809 PMC9453635

27. Griffin JM , Rosenthal JL , Grodin JL , Maurer MS , Grogan M , Cheng RK . ATTR amyloidosis: Current and emerging management strategies: JACC: CardioOncology state-of-the-art review . JACC CardioOncol . 2021 ; 3 ( 4 ): 488 – 505 . 10.1016/j.jaccao.2021.06.006 34729521 PMC8543085

28. Lauppe R , Liseth Hansen J , Fornwall A , et al . Healthcare resource use of patients with transthyretin amyloid cardiomyopathy . ESC Heart Fail . 2022 ; 9 ( 3 ): 1636 – 1642 . 10.1002/ehf2.13913 35365974 PMC9065857