New real-world evidence shows substantial variation in ambulatory cardiac monitoring strategy.¹⁻²

Ambulatory cardiac monitoring (ACM) services provide electrocardiograms (ECGs), which record and analyze the heart’s electrical activity to help physicians determine heart rhythms and diagnose a patient’s medical condition. There are several types of ACM services:  

• Holter monitoring services provide short-term continuous monitors that are worn for up to 48 hours.¹  
• Long-term continuous monitoring (LTCM) services provide monitors that are worn from 48 hours to 14 days.¹
• Event monitoring services provide noncontinuous monitors that are worn for up to 30 days.¹
• Mobile cardiac telemetry (MCT) monitoring services provide data transmission and the device is worn for up to 30 days.¹ 

JUMP TO A TOPIC:

• Detecting specified arrhythmias. 
• Can monitor selection reduce retesting?

Some professional societies have offered guidance on which ACM services physicians should consider, but evidence-based recommendations on monitoring service selection is not currently available from professional societies.¹  

At this year’s American College of Cardiology’s Annual Scientific Session Together with the World Congress of Cardiology, the study “Comparative effectiveness of ambulatory monitors for arrhythmia diagnosis: A retrospective analysis of Medicare beneficiaries” (CAMELOT) was presented. CAMELOT is the first and largest real-world evidence study of diagnostic-naïve Medicare beneficiaries managed with external cardiac monitoring services.^1,3 The researchers compared ACM types to gain an understanding of variation in monitoring strategy, clinical outcomes, and health care utilization. The cohort was derived from the full data set of 1,582,784 Medicare fee-for-service claims between January 1, 2016 and December 31, 2019; it comprised 287,789 Medicare beneficiaries age 65 and older who had not previously received ACM. 

We spoke with two of the study’s authors, Matthew R. Reynolds, MD, Cardiac Electrophysiologist at Lahey Health in Burlington, Massachusetts and Suneet Mittal, MD, Director of Cardiac Electrophysiology at Valley Hospital in Paramus, New Jersey.  

Dr. Reynolds and Dr. Mittal discussed why the CAMELOT study is an important step forward for the field: “There is a need for the clinical community to have some objective data to help them choose between monitors, to know which one is most likely to answer the question for which they are prescribing the monitor. Secondarily, there is interest among insurers and payers to know they are getting good value for the different types of diagnostic testing they are paying for,” Dr. Reynolds commented.  

Dr. Mittal added, “Despite the fact that so many of these monitors are used, many clinicians do not have a precise understanding of the differences behind these monitors, and they may not understand what the implications of diagnosis, retesting, and health care utilization are as a function of these monitors.”  

Holter monitors, LTCMs, external ambulatory event monitors (AEMs), and MCTs were the devices of interest in CAMELOT. Ambulatory monitoring strategy by ACM brand was evaluated within each device category. The iRhythm Zio® XT patch-based device was included in the LTCM category. 

Members of the cohort had no arrhythmia diagnosis during the twelve-month period prior to ACM monitoring as well as no evidence of an intervention for a conduction disturbance during that time.  

Dr. Reynolds noted that sophisticated multivariate analysis was conducted to adjust for baseline differences between the patients in each of the monitoring groups. He explained, “The baseline characteristics of the patients do differ in some respects, and this is not necessarily surprising, for the different types of monitors that were prescribed. For example, the patients in the Holter group were less likely to have a history of stroke than the patients in the long-term continuous monitoring group or the event monitoring group.”  

Historically, multi-lead Holter monitors were commonly used for up to 48 hours. Today, technological advances have contributed to an evolution of ACM technology. “We are now using patch-based systems and we’re using longer-term recordings,” Dr. Mittal explained.   

CAMELOT revealed how often each type of monitoring service was used, the differences in diagnostic yield based on monitor type, and how often retesting occurred with each type of monitor. 

Additionally, the study provided new insight into whether “the monitor vendors were tied to the performance of the monitors, based on outcome variables, such as diagnostic yield, need for retesting, and health care utilization,” Dr. Mittal mentioned. 

Dr. Reynolds offered some insight to help explain the principal finding, “Not knowing what's going on often leads to repeat testing, inefficient repeat testing, but also, emergency department visits and hospitalizations because a diagnosis was not made, and a treatment was not instituted, and so there was uncertainty on the part of the patients and their providers as to what the treatment plan should be. It doesn't surprise me that much, that the diagnostic strategy with the highest yield had the least amount of increased health care costs.” 

Clinical endpoints including 90-day diagnostic yield, 180-day retest (i.e., another ACM), annualized health care utilization, and difference-in-difference of health care utilization were evaluated. 

“When we look at these endpoints . . . there was a big difference between the types of monitors and the choice of vendors for these monitors,” said Dr. Mittal. “There are differences across monitors, even within the same category based on vendor,” Dr. Reynolds agreed.   

Detecting specified arrhythmias. 

In CAMELOT, the following specified arrhythmias were of interest because each one qualifies as a comorbidity that could increase future health care utilization based on Medicare Hierarchical Condition Code (HCC) 96: atrial fibrillation (paroxysmal, persistent, long-standing persistent, chronic, permanent, unspecified), atrial flutter (typical, atypical, unspecified), SVT, VT, AV block, sick sinus syndrome, and junctional premature depolarization. 

“ECG monitoring is a core component of clinical practice,” Dr. Mittal noted. Patients are typically monitored to detect arrhythmias due to symptoms including palpitations, syncope, or a palpitations and syncope.  

Based on the study, compared to Zio XT, other long-term continuous monitoring services are 20% less likely to detect a specified arrhythmia.^1,2 

• Holter monitoring services are 49% less likely than Zio XT to detect a specified arrhythmia.^1,2
• Event monitoring services are 42% less likely than Zio XT to detect a specified arrhythmia.^1,2
• MCT monitoring services are 32% less likely than Zio XT to detect a specified arrhythmia.^1,2 

Can monitor selection reduce retesting? 

Compared to Zio XT, other LTCM services are 3.51 times more likely to result in a retest, Holter monitoring services are 1.35 times more likely to result in a retest, event monitoring services are 4.27 times more likely to result in a retest, and MCT monitoring services are 2.83 times more likely to result in a retest.^1,2 

Dr. Reynolds described an observation from the data that raises new questions, “A lot of the patients who had retesting done underwent retesting with the same monitor.” Future research could be conducted to try to understand why physicians would prescribe the same monitor type for a retest if that monitoring service did not provide answers on the first test.  

See the CAMELOT poster.

Matthew R. Reynolds, MD, Cardiac Electrophysiologist at Beth Israel Lahey Health 

Suneet Mittal, MD, Director of Cardiac Electrophysiology at Valley Health System