RETURN TO ABSTRACT LISTING


Objective:
Left ventricular assist devices (LVAD) are increasingly being used for cardiac support in patients with end-stage heart failure. Some of these patients will eventually get a heart transplant as definitive therapy. There is significant morbidity and mortality attributable to these devices and the underlying cardiac condition including drive line infections.

Materials and Methods:
We studied a case series of 10 patients at our institution diagnosed with FDG avid LVAD infections by FDG-PET/CT to determine the site and extent of LVAD infection(s) with clinical follow-up to learn the clinical impact of management decision on the patient outcome. EMRs were reviewed for patients demographics (age, sex, ethnicity), inpatient vs. outpatient status, inpatient length of stay, culture data, and treatments decisions (antibiotic, surgical antiinfective intervention, LVAD replacement), deaths and heart transplant were recorded. Additionally, the length of required hospital stays, microbiology culture reports, treatments rendered (Antibiotics, surgical anti-infective intervention, LVAD replacement, and heart transplant). Data were analyzed for infection sites at four LVAD components, including driveline exit point, subcutaneous driveline pathway, pump pocket, and outflow tract for each patient.

Results:
Of the 10 patients, seven had FDG avid infection involving three components of LVAD (driveline, pump pocket, and/or outflow tract) and associated FDG avid reactive mediastinal lymphadenopathy. All seven patients were put on an accelerated list for orthotopic heart transplant, with two successfully obtaining the heart transplant. Hospitalization, surgical washout, and debridement were required in six of the infected patients, with longer lengths of stay and morbidity, including osteomyelitis, and septic emboli in patients with LVAD infections. The median length of stay in the hospitalized patients was 90 days (10 days - 110 days), with the caveat that five patients were outpatients at the time of PET scan and were eventually admitted. There were three patients with driveline infection only, of which one required LVAD replacement, but ultimately died secondary to uncontrolled driveline infection. The remaining two were stabilized on chronic outpatient antibiotics. The organisms involved included Pseudomonas aeruginosa (5/10 patients), Staphylococcus aureus (MSSA and MRSA, 3/10 patients), Serratia (2/10 patients), and one patient each with Corynebacterium, Achromobacter, and Candida. All patients required long duration of parenteral antibiotics followed by chronic suppressive therapy.

Conclusion:
FDG PET/CT plays an important role in diagnosing the site and extent of LVAD infections and in guiding/evaluating the effectiveness of clinical management/interventions. The prognosis and outcomes are dependent on the extent of LVAD driveline infection and concomitant pocket abscesses, with a higher need of surgical washouts and transplant in the more involved cases.