Although it has been considered a safe procedure, computed tomography scanning uses high doses of radiation and can cause malfunctioning in those patients with ICD when the radiation is directly incident on the device. We present a case of ventricular oversensing during a thoracic computed tomography.
A 65-year-old woman with a history of mixed connectivopathy (systemic scleroderma and lupus erythematosus) since 1984 and coronary artery disease with myocardial infarct since 1994 suffered a syncopal episode in March 2004.
During the electrophysiological study, a syncopal monomorphic ventricular tachycardia with a 240 milliseconds cycle length was induced. Ejection fraction was 45%.
On 12 March 2004 an ICD was implanted (Vitality 1871 DR; Boston Scientific, St Paul, MN, USA) with a right ventricular pace/sense and defibrillation lead (Endotak model 0148; Boston Scientific); values measured in at implant: capture threshold of 1.2 V at 0.5 milliseconds, measured
Detection of VT was programmed at 180 beats/min, and VF at 220 beats/min. Ventricular sensitivity was programmed at 0.27 mV.
In January 2008 the elective replacement indicator (ERI) was reached and the device was replaced with Vitality 2 EL-T167 (Boston Scientific), maintaining the same programming. No ventricular arrhythmia had been detected by the device.
On 5 November 2008, as part of a study of pulmonary involvement by her rheumatic disorder, she underwent a high-resolution thoracic CT scan.
This exam was conducted with a CT Volume Access Somaton (Siemens Medical Solutions, Forchheim, Germany). The study began at
CT topogram. Initial topogram of the scan, showing the implanted device in the left pectoral zone and both electrodes. Arrows: date and time of scan.
On 26 November 2008 the patient underwent ICD followup. All measurements of the battery and the leads remained within normal ranges; however, measurement of the intrinsic amplitude of the ventricle was impossible because the patient was pacing dependent.
The interrogation of the device-stored electrograms reveals an episode of nonsustained ventricular fibrillation which started capacitors charge that was diverted 3 seconds later (Figure
Therapy history: episode 460. Detection of ventricular fibrillation on November 5, 2008 at
Analysis of the stored electrograms of the episode, which coincided with the time of the thoracic CT scan, reveals an oversensing in the right ventricular channel interpreted by the device as ventricular activity. The oversensing provokes the charge of the capacitors and at the same time causes significant pauses due to inhibition of ventricular pacing (Figure
Electrograms of episode 460. Oversensing due to artifacts that start capacitors charge and provoke ventricular pacing inhibition.
Diaphragmatic or other usual causes of oversensing were excluded from the beginning by performing extreme breathing movements, left arm movements, and pocket manipulation in different patient positions.
Stability and integrity of both leads was confirmed with several followups during 6 months.
The causes of interference affecting electronic cardiac devices are numerous and of diverse origin [
CT did not appear in lists of potential sources of electromagnetic interference [
In order to confirm these results a prospective study was set up to analyse 13 pacemakers and 8 defibrillators, manufactured by Medtronic, subjected to ionizing radiation using CT systems with the production of images in spiral and dynamic mode at different dosage levels [
Oversensing was detected in the majority of the units, with inhibition of up to 4 seconds in some devices. It is not specified whether episodes of oversensing occur in the defibrillators that cause a diagnosis of tachycardia and the activation of a therapy.
A recent study [
Short episodes of oversensing were observed on 3 pacemakers, but none of the ICDs appeared to be affected. In addition, there were two curious observations. The units affected were those of most recent manufacture, and the 3 pacemakers that were affected remained undamaged during the repetition of the test when protected by a thin layer of copper.
Clinically implanted devices have shown the influence of CT on neurostimulators.
Three patients showed unexpected electrical stimulation when the radiation was directly incident on the device [
None of these reports show examples of electrograms with oversensing.
In July 2008, the [US]FDA collected all this information and issued a preliminary notification [
Both the FDA and ECRI Institute [
Use the lowest possible dose of radiation and during the shortest possible time.
Programme the device to “Off” (ICD) or in safety mode (pacemakers).
Have a pacemaker/defibrillator programmer, an external defibrillator and emergency medication available.
After the examination, reprogramme the device to “On” or back to the initial programming. Check the full and correct functioning of the device.
Due to the coincidence in time of both and the absence of other known interference causes, we can conclude that the oversensing is caused by the CT scan.
This is the first case demonstrating that an ICD implanted in a human subject may be affected during a CT scan; oversensing may occur, causing pacing inhibition and starting a VF episode detection with charge of the capacitors of the high-voltage circuit.
Although it has been considered a safe procedure, computed tomography scanning uses high doses of radiation and can cause malfunctioning in those patients with ICD when the radiation is directly incident on the device.
Radiologists need to be aware of this possibility so that they may avoid this radiation as far as possible and will have to be prepared to treat those patients if complications arise.
In addition, the patient should be examined by his/her doctor after the scan to check that the device has not been damaged.