Prevention and management of device infection

Current practice guidelines

Cardiovascular device therapy has become increasingly commonplace and is now applied not only to patients with manifest rhythm disturbances but also in anticipation of events such as implantable cardioverter-defibrillators (ICDs) for primary prevention, and for improvement in cardiac function in the absence of cardiac arrhythmia (cardiac resynchronization therapy, or CRT.) Approximately 400,000 cardiac devices are implanted yearly in the U.S., and it is estimated that more than 3 million patients have implanted devices.

Presently, one of the most vexing and morbid complications of device therapy is the development of infection. The incidence of device infection is low but has increased.

"Until 2004, the rate of infection was constant at approximately 1.5 percent per year, but then it steadily increased to a rate of 2.5 percent per year as reported in 2008," according to Larry M. Baddour, M.D., chair of the Division of Infectious Diseases at Mayo Clinic in Rochester, Minn. This increase coincides with the expanded use of ICDs and CRT and is likely related to longer procedure times, the increased bulk of larger generators and multiple leads, all undermining wound and skin integrity.

Other factors known to be associated with the risk of device infection include:

  • Chronic renal insufficiency
  • The presence of chronic intravenous lines (including dialysis catheters)
  • Chronic anticoagulation
  • Multiple device leads (≥ 3)
  • Number of device-related procedures (≥ 3)
  • Immunosuppression, either related to an underlying disease process or therapy with corticosteroids

Early device reoperation (especially within days) is associated with the greatest risk of infection, increasing it by fifteenfold. Also contributing are the increasing longevity of patients with devices and the subsequent need for generator replacement and lead revision.

Appropriate antibiotic prophylaxis at the time of device procedure is a crucial step in device infection prevention. In most centers, cefazolin is administered one hour prior to device placement and is continued for less than 24 hours. In contrast, there are no data to support secondary prophylaxis for patients with devices who undergo dental and other invasive procedures and, therefore, it is not recommended.

Pocket infection

In the majority of patients (60 percent), a device infection will involve the device pocket. "The most obvious sign is a pocket abscess, but the presenting symptoms may be subtle, such as pocket erythema, induration or pain," says Michael J. Osborn, M.D., electrophysiologist at Mayo Clinic in Rochester. "The latter symptom should not be overlooked, as pocket pain persisting for more than a few days postprocedure or developing after a symptom-free period is extremely unusual and should be vigorously evaluated."

A draining sinus is a very common manifestation of pocket infection. Those that develop shortly after a device procedure may represent stitch abscesses, which may respond to antibiotics and careful removal of protruding suture material. Any more aggressive therapy should be avoided.

A chronic draining sinus, especially one that develops late after a procedure, is worrisome. If the sinus has clear communication with the pocket or if more pronounced pocket erosion occurs, the entire pocket should be considered contaminated, and the device and leads should be removed. If there is no clear communication with the pocket, a trial of appropriate antibiotic therapy is reasonable as long as there are no signs of systemic infection, such as:

  • Chills
  • Fever
  • Valve or lead endocarditis seen on an echocardiogram

If the sinus does not heal or recurs after therapy, the device should be managed as an infection.

In about 40 percent of patients, bloodstream infection occurs as the only sign of device infection. Bloodstream infection due to staphylococci or relapsing bloodstream infection should prompt concern for device infection, even when the pocket site appears normal.

"Echocardiography should be done to determine if there is evidence for complicating valve infection. The finding of a mobile mass on an intracardiac lead is much less reliable as a sign of lead infection, as many leads develop an irregular fibrinous coating over time in noninfected patients," says Dr. Osborn. Device infection requires complete system removal for attempted cure. Antibiotic therapy without complete device and lead removal is associated with an increase in 30-day mortality.

Mortality risk

Despite proper therapy, there is still a significant mortality risk with infection, especially when associated with staphylococcal bacteremia. The in-hospital mortality rate following successful extraction of an infected system ranges from 3 to 11 percent with a posthospital (up to two-year) mortality rate of 10 to 20 percent. The risk of death is two to three times higher in patients with staphylococcal bacteremia compared with infection limited to the pocket. This significant mortality risk is related to the virulence of the organism and also frequently to delays in delivery of appropriate therapy because of slow recognition of the infection, delays in device removal and attempts at more conservative therapy.

"Patients with devices who present with staphylococcal bacteremia have a high incidence of associated device infection and should be evaluated promptly by physicians with expertise in infected device removal," says Dr. Baddour. In contrast, patients with gram-negative bacteremia have associated device infection much less frequently and, as a result, device removal can be delayed to allow assessment of appropriate antibiotic therapy if there is no other evidence of device infection. Obvious signs of pocket site infection, lead endocarditis, valve endocarditis, or relapsing or sustained bloodstream infection despite antibiotic therapy should be managed aggressively with complete device removal.

The Heart Rhythm Society and American College of Cardiology have established appropriateness criteria for lead extraction in patients with infections based on prior experience managing patients in a more conservative fashion. Complete extraction is recommended in any definite device infection:

  • Valve or lead endocarditis
  • Bloodstream infection and associated pocket infection (abscess erosion or chronic draining sinus)
  • An occult gram-positive bacteremia

Extraction is reasonable in patients with persistent gram-negative bacteremia. Extraction is not recommended when there is no clear evidence of pocket infection or in the rare patients who have a significant underlying morbidity that would preclude aggressive therapy.

Lead extraction

There is a standard approach to lead extraction in patients with infection. Blood cultures are obtained when a diagnosis of device infection is suspected. The pocket is opened and inspected. If there is obvious infection, the pocket is extensively debrided, and tissue is sent for microbiologic studies. The leads are dissected free of all fibrous material within the pocket and from the sleeves that anchor them to the pectoralis muscle. Stylets are inserted, and extraction by traction is attempted.

If this approach is not entirely successful, the proximal portions of the leads are amputated, and a locking stylet is inserted into the lead and advanced under fluoroscopic control to the distal tip of the lead and locked in place. Traction is attempted again, and if not successful, a powered sheath (typically a laser sheath) is advanced over the lead to the first area of resistance or binding scar. The laser is activated until the fibrous tissue is ablated, and the sheath is then advanced to the next area of binding scar, with the process repeated until the lead is freed and removed from the heart. The extracted device and leads are sent for microbiologic studies.

Power extraction tools have improved the rate of successful lead removal. In the past 20 years at Mayo Clinic in Rochester, 97.3 percent of all leads have been completely removed. Approximately 1 percent of leads have been incompletely removed, leaving a very small segment firmly attached to the endocardium. In 1.5 percent of patients, leads could not be removed transvenously and required surgical intervention.

"Virtually all leads that have been in place for less than five years have been completely removed successfully," says Dr. Osborn. "For leads that have been in place for more than five years, the success rate for complete lead removal is approximately 92 percent."

The rate of potentially life-threatening complications, however, remains between 2 and 3 percent. These complications include:

  • Innominate vein or superior vena cava laceration
  • Cardiac perforation
  • Requirement for emergent (rescue) surgery
  • Pericardial effusion requiring intervention
  • Blood loss requiring transfusion
  • Deep venous thrombosis
  • Tricuspid valve damage resulting in significant regurgitation
  • Ventricular arrhythmias requiring intervention
  • Pulmonary embolism
  • Cerebrovascular accident
  • Transient respiratory failure
  • Transient renal failure
  • Axillary or subclavian venous bleeding requiring surgical intervention

Laceration of the superior vena cava accounts for more than 60 percent of procedural mortality. Cardiac perforation and consequent emergent surgery accounts for the majority of the other deaths.

For more information

Baddour LM, et al. Update on cardiovascular implantable electronic device infections and their management: A scientific statement from the American Heart Association. Circulation. 2010;121:458.