Author + information
- Elena Arbelo, MD, PhD and
- Josep Brugada, MD, PhD∗ ()
- Department of Cardiology, Thorax Institute, Hospital Clinic of Barcelona, University of Barcelona, Barcelona, Spain
- ↵∗Reprint requests and correspondence:
Dr. Josep Brugada, Department of Cardiology, Thorax Institute, Hospital Clínic de Barcelona, C/Villarroel 170, 6° - Escala 3, 08036 Barcelona, Spain.
In the past few decades, the use of cardiac rhythm management (CRM) devices has increased worldwide due to the increase in life expectancy and technological developments (e.g., cardiac resynchronization therapy) that have extended their indications (1–4). When imaging is required in patients with CRM devices, computed tomography (CT) is the preferred modality because magnetic resonance imaging has been generally contraindicated (5–7).
CT scanning technology is ubiquitous and over the past 15 years has become routine in the evaluation of neurologic, respiratory, cardiovascular, gastrointestinal, and other diseases. Therefore, an increase in the number of patients with CRM devices undergoing CT scanning at least once in their lifetime is also expected.
As the use of CT scans has skyrocketed, some experts have raised concerns over the potential risk of patients' increased exposure to radiation. Between 2006 and 2007, several reports of CT interference with CRM device functioning were published (8–10).
This led the U.S. Food and Drug Administration (FDA) to issue in 2008 an advisory informing of possible malfunction of CRM devices, such as inappropriate shocks from implanted cardioverter-defibrillators (ICDs) and transient changes in permanent pacemaker (PPM) output pulse rate that could result in failure to pace.
Was the FDA overreacting, or was this public health notification justified?
The FDA is responsible for overseeing the research, development, and marketing of various foods, drugs, and medical devices. However, some concern has been raised about the possibility of a tendency toward over-regulation by this institution, and the use of CT scans in patients with CRM devices might very well be a good example.
To answer this question, we have reviewed the scientific data on which the FDA 2008 recommendations were based, finding that there are scarce data about potential interference on CRM device functioning from CT imaging. In fact, there are only two peer-reviewed publications with limited evidence of direct causality (8,9). In the study by Yamaji et al. (8), 6 of 11 patients with an implanted PPM showed transient lead oversensing during chest CT scanning. Of note, none of these were associated with any clinically significant event. Both that report (8) and one by McCollough et al. (9) used in vitro phantom models to demonstrate the interaction of ionized radiation with CRM devices, but studies of confirmed events in humans were lacking. Notably, as shown by McCollough et al. (9), only some CRM devices were sensitive to radiation exceeding certain dose rates, and these interactions terminated immediately after the radiation beam was moved out of the region of the device. Those investigators could not provoke any permanent changes in the device programming (9). We therefore believe that the FDA may have been overzealous when issuing the 2008 advisory, on the basis of the limited scientific information available at the time. The main problem is that the results of these studies may not be generalizable to the everyday clinical setting.
Since 2008, two additional reports have evaluated the interaction of CT imaging with CRM functioning in vitro (11,12). As in the previous studies, these experimental studies suggested that the interference is dependent on the radiation dose and the duration of the exposure and recommended lower x-ray doses and short periods of radiation.
In this issue of the Journal, Hussein et al. (13) assess the effect on CRM devices of direct exposure to radiation beams from CT imaging to further address these concerns in a clinical setting. Data were drawn from the medical records of 386 patients who underwent a total of 516 CT scans. The primary outcome was a composite endpoint of death, unplanned hospitalization, bradycardia or tachycardia requiring termination of the CT scan or an immediate intervention, inappropriate ICD shock, resetting or reprogramming of the device, or device replacement/revision attributable to radiation exposure. As a secondary endpoint, the authors analyzed the occurrence of significant changes in device parameters. Their main finding was that direct exposure to radiation from CT imaging was not associated with any clinically relevant adverse event, nor did it induce significant variations in battery voltage or electrode parameters.
The study by Hussein et al. (13) is of particular relevance for several reasons: first, it addresses a specific public health problem; second, it was designed and performed in a timely manner; and third, it is of major importance not only in the clinical community but also in the regulated industry and the FDA.
The few limitations of the reported analysis, acknowledged by the authors, mainly derive from the retrospective nature of the study. First, relevant variables such as radiation exposure (dose and duration, particularly over the device) were not recorded. With regard to the evaluation of adverse events, recording in the medical record could likely overlook “milder” events (e.g., transient dizziness). However, “stronger” outcomes (e.g., syncope or an inappropriate shock) would doubtfully be left out. Also, device parameters were not assessed immediately before and/or after CT scanning; therefore, some transient changes may have been overlooked. Finally, some clinically relevant events may have been overlooked due to the small sample size (in relation to the hundreds of thousands of patients with CRM devices undergoing CT scanning). The authors are aware of this limitation, but as they rightly stress, these would likely occur at a very low rate (0.2%).
From a practical perspective, the observational design of the study by Hussein et al. (13) may actually make the study results more, rather than less, important. The authors have validated in an everyday setting that PPM and ICD responses to radiation are transient and benign and that routine CT scanning in patients with CRM devices should not be restricted.
The authors are to be commended for having undertaken such a study to clarify a clinical situation of major importance.
To date, hundreds of thousands of patients with some type of CRM device have undergone CT scanning, considering the widespread use of both, particularly in older patients. The risks associated with imaging procedures are small compared with other risks. For instance, of patients with a PPM or ICD, some may require CT imaging for the evaluation of lung tumor or for possible internal bleeding after motor vehicle accidents. The relative risk from the natural course of any disease requiring CT evaluation compared to the risks of the imaging itself is remarkable.
Government institutions like the FDA are necessary to ensure that effective therapies are made available to patients as quickly as possible whereas ensuring that those therapies are safe over the life cycle of a product. However, it is necessary to be rigorous in order to avoid misleading and confounding. In medicine, we continuously live with risk/benefit ratios, and we make reasoned judgments accordingly. We understand that, in alleviating one problem, we may increase the risk for another (e.g., anticoagulation for stroke prevention).
Unfortunately, these agencies, when trying to guarantee the safety of a product, often neglect the potential of benefit. Furthermore, it is our belief that in this particular case, the FDA may have been somehow reckless. Since the first CT scan on a patient in practice in 1971 (14), several hundred thousand (probably millions) have undergone this test while having an implanted PPM or ICD. Very few CT examination types would have the radiation constant over the device beyond 2 s. Thus, common sense tells us that device interactions may, in fact, be quite frequent but benign because the effect is transient, lasting at most a few seconds. Indeed, 3 decades after that first CT scan, no cases of clinically major adverse events in patients in practice have been described to the authorities or published in the data. To answer our question, we believe that the FDA was overreacting when issuing the 2008 advisory.
We count on these regulatory agencies to provide us with a well-defined profile of both potential toxicity and efficacy, on the basis of solid evidence. For that purpose, it is of major importance that the FDA and other similar institutions work hand in hand with clinicians to develop a post-marketing structure that addresses new and ongoing concerns and identifies early signals of potential problems in a sensible fashion, always weighing the risk/benefit to patients and taking into account the accumulated experience of professionals.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Both authors report that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- (2012) The EHRA white book. Europace 14:ii1–55.
- Asia Pacific Heart Rhythm Society (APHRS). The APHRS White Book 2013. APHRS Web site. Available at: http://aphrs.asia/pdf/white_book2013.pdf 2013. Accessed January 3, 2014.
- Zhang S.
- Martin E.T.,
- Coman J.A.,
- Shellock F.G.,
- Pulling C.C.,
- Fair R.,
- Jenkins K.
- Roguin A.,
- Zviman M.M.,
- Meininger G.R.,
- et al.
- Levine G.N.,
- Gomes A.S.,
- Arai A.E.,
- et al.
- Hussein A.,
- Abutaleb A.,
- Jeudy J.,
- et al.
- Beckmann E.C.