Twiddler’s Syndrome: Utilizing New ICD Technology to Help Treat an Old Problem

A 59-year-old male presented with an out-of-hospital cardiac arrest and documented ventricular fibrillation. Coronary angiography revealed severe multivessel coronary artery disease, and he subsequently underwent surgical revascularization. Because of continued ventricular arrhythmias, a dual-chamber implantable cardioverter-defibrillator (ICD) was placed. He had no further arrhythmias, and subsequent routine device checks were normal. However, nearly two years after implant, an audible alert and “twitching” of his left upper extremity and upper abdomen prompted a device interrogation. At the device check, there was no ventricular or atrial sensing or pacing, and the device had reverted to a high output mode. Left-sided muscle twitching was also noted during the check. A chest x-ray was obtained (Figure 1), which revealed that the atrial and ventricular leads had been dislodged from the heart and were coiled around the pulse generator. Twiddler’s syndrome was suspected. After discussing the radiograph with the patient, he admitted to frequently “massaging” the pulse generator pocket, likely dislodging the leads and wrapping them around the pulse generator. Given the secondary prevention nature of the device, the patient was fitted for an external ICD and scheduled for reimplantation. 

 

Prior to reimplantation, high-output pacing was performed with A-V sequential pacing, causing the pectoral muscle and diaphragm, respectively, to contract (Figure 2, also see video). The ICD pocket itself was somewhat swollen, and when the pocket was opened, a large amount of chronic inflammatory tissue was seen. The atrial lead tip was overlying the pectoralis muscle, and the RV lead tip was in the region of the phrenic nerve, explaining the “A-V sequential pacing” of the left chest and diaphragm. Left axillary venous access was attempted, but unsuccessful. A left upper extremity venogram was then done, which showed complete occlusion of the left subclavian vein (Figure 3). The patient did not have a pacing indication and had no history of monomorphic ventricular tachycardia; therefore, he was thought to be candidate for a completely subcutaneous implantable cardioverter-defibrillator (S-ICD). A subcutaneous ICD was implanted using the two-incision technique, and the patient recovered unremarkably. At follow-up, the subcutaneous ICD was sensing appropriately and the pocket was healing well.

Bayliss et al first described Twiddler’s syndrome in 1968 as failure to pace due to pacemaker lead retraction from manipulation of the pulse generator.¹ Manipulation of the pulse generator can cause lead dislodgment and coiling of leads around the pulse generator, which can result in poor sensing as well as poor stimulation of the brachial plexus, phrenic nerve, and pectoral muscle. This syndrome has also been reported in ICDs. Most cases occur during the first year of implantation; however, cases have been reported as early as 48 hours and as late as two years after implant, as is the case with this patient.² Twiddler’s syndrome has also been documented in patients with chemotherapy infusion devices and deep brain stimulators.² ICD pulse generators implanted in the abdomen are more vulnerable to manipulation.³

 

Twiddler’s syndrome is uncommon. Though the precise incidence is unknown, it has been reported to be 0.07-7%.^2,4 In 2003, Fahreaus and Höijer reported their rate of incidence in 10 years was 0.07% (12 cases of Twiddler’s syndrome in 17,000 follow-ups).⁵ 

 

Risk factors for Twiddler’s syndrome include the female gender, obesity, an elderly age, increased laxity of subcutaneous tissue,⁴ or psychiatric illness. It can also be due to pulse generator/pocket size mismatch,² as an overly large pocket can cause excess device movement and ease of manipulation.

 

Clinical manifestations of Twiddler’s syndrome vary depending on the location of the dislodged leads. Stimulation of the phrenic nerve causing involuntary respiration or hiccups, stimulation of the brachial plexus causing involuntary twitching of the arm, pectoral twitching, and sharp pain in the neck have all been reported. This patient had stimulation of the pectoralis muscle, brachial plexus stimulation with upper extremity twitching and, most interestingly, sequential “pectoralis-phrenic” pacing during high-output A-V device pacing. Damage to ICD leads can cause oversensing and result in inappropriate therapy.⁴ Diagnosis can be suspected with Holter monitoring and device interrogation, but is confirmed with chest radiography.⁴ 

 

Treatment involves device reimplantation, with care taken to uncoil and inspect the leads for damage prior to reimplantation. Fahraeus and Höijer discussed using non-absorbable suture to fix the generator to the fascia or subpectoral implantation to prevent subsequent manipulation.⁵ Minimizing the size of the pocket or placing the device in a Dacron-woven pouch can also decrease the incidence of the syndrome.² However, the most effective way to prevent this syndrome is to provide proper patient education and counseling.⁴ 

 

The subcutaneous ICD has been shown to be effective for both the primary and secondary prevention of sudden cardiac death.⁶ This device offers some theoretical advantages in preventing Twiddler’s syndrome, as the pocket for the pulse generator is in a more difficult spot to manipulate and the electrode is less likely to dislodge. Kooiman et al was the first to report using this device for Twiddler’s syndrome.⁷ 

Twiddler’s syndrome is an infrequent though well-described complication of cardiac device therapy. Here we describe a unique presentation of this syndrome manifesting as “A-V sequential pacing” of the left pectoralis muscle and diaphragm. Reimplantation of the chronic device was attempted, but the left subclavian vein was completely destroyed by the leads as they were torn from the vasculature. Options for this patient included a right-sided endocardial device or a subcutaneous ICD, as was done here. This is a unique application of the subcutaneous ICD, which offers several theoretical advantages over a traditional ICD, including a more difficult to access pulse generator (mid clavicular line, inferior to left axilla) and the absence of an endocardial lead to dislodge. Given this patient’s young age, lack of a pacing indication, and absence of known monomorphic VT, he was an ideal candidate for the subcutaneous ICD, though it was not available at our institution at the time of initial device implant. At follow-up, the patient was doing well with no signs of pocket manipulation of the subcutaneous ICD.