A middle-aged man who had a biventricular pacemaker inserted 3 weeks ago presented with fever as well as pain over the left upper chest wall. Clinical examination revealed erythema and tenderness over the pacemaker body insertion site, with just a small amount of pus expressed from the incision scar. Blood cultures and a swab of the pus from the wound were both positive for methicillin-resistant Staphylococcus aureus with the following drug susceptibility profile (obtained from VITEK2):

• Vancomycin MIC = 1.5 (S)
• Gentamicin = S
• Trimethoprim/sulfamethoxazole = S
• Erythromycin = R
• Clindamycin = R
• Ciprofloxacin = R
• Rifampicin = S
• Linezolid = S

Chest X-ray of the man showing a biventricular pacemaker.

Question: How should this patient be managed?

[Updated 17 Oct 2015]

This is a particularly difficult type of infection to treat. It is likely that the pacemaker generator/pocket site has been infected, and given the presence of bacteraemia, the infection is unlikely to be superficial/confined to the surgical site alone. A trans-oesophageal echocardiogram (TEE) should be performed to determine if there is any accompanying endocarditis of the heart valves or lead wires, especially given that the bacteraemia and pocket site infection are caused by MRSA. A trans-thoracic echocardiogram – often performed because it is relatively more straightforward a test – lacks sensitivity. The updated American Heart Association recommendations are clear – regardless of whether there is any endocarditis, in this situation, the entire device (including wires and generator) should be removed. Treatment failure is high if the infected foreign body is left in place.

The choice of antibiotics is not straightforward. Vancomycin remains the most cost-effective antibiotic against severe MRSA infections, especially in the local/regional setting. However, the organism has a vancomycin MIC of 1.5 via Etest, which is associated with higher treatment failure rates if the drug is used. Daptomycin would be a good alternate option (without factoring in the cost of the drug, which is about 7 times more than vancomycin, it would probably be the drug of choice). There is little high quality data with regards to other drugs such as linezolid or ceftaroline with regards to treating MRSA bacteraemia especially in the presence of a device infection or potential endocarditis.Trimethoprim/sulfamethoxazole should be considered inferior to vancomycin for the treatment of such infections.

The duration of antibiotic therapy depends on the site/extent of infection. If the infection only involves the pocket/generator (as confirmed by TEE), then 14 days of therapy should be adequate provided the entire pacemaker has been removed and the bacteraemia is rapidly cleared. If there is endocarditis or infection of other sites such as bone, or if the bacteraemia is persistent, the patient may required up to 4-6 weeks of antibiotic therapy.

The other question to address is when should a new pacemaker be inserted. Here, the conventional wisdom is to implant another new pacemaker only after the infection has been brought under control.