Circulation 122,210 • Volume 31, No. 1 • June 2016   Issue PDF

Expert Clarifies Complexity of Unintended ICD Firing

Robert Addleman, MD; Annemarie Thompson, MD

Letter to the Editor:

To the Editor:

We had a problem with an ICD a couple weeks ago that caught us by surprise. I thought it might be of interest to you and your readership.

The procedure was an open reduction and internal fixation (ORIF) of the left proximal humerus (essentially the shoulder) in the beach chair position. The anesthesiologist placed the magnet over the St Jude ICD in our usual manner (centered over the ICD). The ICD was in the left subclavian position, approximately 6 inches from the surgical field. The St Jude ICD does not emit a tone or vibration when the magnet has made contact, so there is no way to know for sure it is correctly placed.

Early into the case, during a long cautery run, the ICD fired, in spite of the fact that the patient did not have a shockable rhythm and the magnet was securely placed over the ICD. Upon investigation, we learned that the correct magnet placement to shut off the St Jude ICD is slightly different from almost all other ICD models. I say “almost all” because during my investigation, I learned that the Boston Scientific “EMBLEM” model has the same requirement.

For these models, the magnet should be placed such that the metal arc of the magnet is centered over the center of the ICD body, with the “donut hole” of the magnet slightly off center. I have been told that the arc emits more magnetic force than the center. One excellent reference is the 2011 review article “Clinical Applications of Magnets on Cardiac Rhythm Management Devices” (our own Dr. Benzy Padanilam is a co-author).1

Two things struck me during this investigation. First, not all of the St Jude representatives were aware of these suggestions of magnet placement. Second, there is no mention of this idiosyncrasy in the most recent (2011) HRS/ASA Consensus Statement on the Perioperative Management of Patients with ICDs/PMs.

Thank you.

Robert Addleman, MD Chief of Anesthesia Chair, Department of Anesthesiology St. Vincent Indianapolis


  1. Jacob S, Panaich SS, Maheshwari R, Haddad JW, Padanilam BJ, John SK.Europace. Clinical applications of magnets on cardiac rhythm management devices. Europace. 2011;13:1222-30.

Dear Dr. Addleman,

You presented an interesting case of a patient undergoing an ORIF of the left proximal humerus in the beach chair position with an ipsilateral St. Jude ICD generator that fired during a prolonged period of electrocautery. This case merits discussion to examine the possible causes of the ICD firing inappropriately. By the term “firing inappropriately,” I am referring to electromagnetic interference, such as electrocautery, which is interpreted by the ICD as a tachyarrhythmia meeting the threshold for tachyarrhythmia therapy, and subsequently delivering a calculated shock.

Assuming a device check revealed no malfunction of the ICD, the most likely reason the ICD delivered inappropriate anti-tachycardia therapy is the magnet lost contact with the ICD generator, regardless of whether the magnet was initially centered over the generator or off-center. With a humeral fracture in a patient with an ipsilateral ICD, the chest muscles around the generator are in continuity with the shoulder. Movement of skin or muscle overlying the generator during manipulation of the ipsilateral arm could result in the magnet losing contact with the ICD generator. The non-supine position further increases the likelihood of disengagement of the magnet with the ICD generator.

It wasn’t noted in the report whether or not the magnet function was disabled in the patient’s device. Some manufacturers allow the device to be programmed to not have a typical response to magnet placement. For St. Jude devices, the magnet response can be turned off, meaning magnet application will not suspend anti-tachycardia therapy. Therefore, a provider who wasn’t aware that magnet function had been programmed “off” might place a magnet over the device thinking it would suspend tachyarrhythmia detection; however, when the magnet response is programmed “off,” a magnet will not suspend tachyarrhythmia therapy. In order to determine if the magnet function has been disabled, the device must be either interrogated using a programmer, confirmed with the manufacturer’s registry of devices, or confirmed with the physician who normally manages the patient’s device.

Another factor in magnet engagement is patient size. The exact location of the ICD generator can be difficult to ascertain in a very obese patient with submuscular generator placement. At times, even when the generator is located, a second magnet may need to be applied to the overlying skin to elicit the expected effect of the magnet on the ICD. The initial central placement over the St. Jude device rather than off-center is probably the least likely cause of the ICD inappropriately firing in this case. A ring magnet is typically at least 90 gauss, whereas a field of greater than 5-10 gauss is considered sufficient magnet strength to affect device function. Although off-center magnet placement over a St. Jude device is recommended by the company documents, the strength of the manufacturers’ ring magnet is sufficiently strong enough to ensure expected magnet functionality in most patients regardless of “center” versus “off center” placement. Engineers from other manufacturers who were consulted in preparation of this letter have also questioned the theory that “off center” placement gives significantly “better” magnetic field strength.

In patients with electrocautery occurring within 15 cm of the device and potential difficulties securing magnet engagement due to patient position, I would recommend arranging for device reprogramming to suspend tachyarrhythmia therapy during surgery. When that occurs, an alternative for external cardioversion/defibrillation should be immediately available, and the patient should remain on telemetry monitoring until the device is reprogrammed to its initial settings in the postoperative period.

Thank you for allowing me to contribute to the discussion.


Annemarie Thompson, MD Professor of Anesthesiology Director, Anesthesiology Residency Program Department of Anesthesiology Division of Cardiothoracic Anesthesiology Duke University Medical Center