Inclusion and exclusion criteria.

Patients aged >18 years who underwent CIED extraction for infection at MHC between 1^st January 2013 and 31^st March 2015. Patients who had the following device types were included: cardiac resynchronization therapy (CRT-D [with defibrillator], CRT-P [with pacemaker]), ICDs and permanent pacemakers (PPMs). Patients with implantable loop recorders and those with infection of temporary pacing systems were excluded.

Definitions.

For the purposes of this analysis, a CIED infection episode was considered to start on the date of admission to MHC and end on the day of completion of antibiotic therapy for that episode (or discharge following re-implantation of a new device). CIED infections were sub-categorized, according to current UK guidelines, into the following groups; CIED-generator pocket infection (CIED-GPI), CIED-lead infection (CIED-LI), CIED-infective endocarditis (CIED-IE).[13]

Relapse of infection in the case of CIED-LI/IE was considered to have occurred if blood cultures became positive following initial resolution of clinical signs of infection, with blood cultures/device growing the same microorganism as the previous infective episode (confirmed with species identification), within 12 months of completing therapy. In the case of CIED-GPI, relapse was based on clinical evidence of infection occurring within 12 months of the initial infection episode.

Management strategy.

All cases underwent “two-stage” management of their infection. Patients were classified into two groups: ER cases remained as inpatients for the duration of their treatment episode; DR cases had the infected device extracted and were discharged prior to re-implantation of a new device at a later date. In the DR group, the decision to receive outpatient antimicrobial therapy (OPAT), either parenteral or oral, was on advice of the microbiologists; guided by antimicrobial sensitivities and the trend in blood markers for infection.

Cost analysis.

The total cost of each CIED infection episode was calculated from the following formula:

The procedure cost was calculated by adding the cost of materials used during the case (bar code analysis of discarded packaging) to the total catheter lab staff cost for the procedure (unit of laboratory time (ULT)). At MHC, one ULT costs £175, where one ULT equates to 30 minutes of staff time (covering the cost of a cardiologist, an anaesthetist, 2 nurses, a cardiac physiologist and a radiographer). Concerning the re-implantation procedure, the cost of the new device was based on the price paid locally to the manufacturer for that particular model, which is also recorded in the electronic patient records. The price of the discarded device was not included in calculations.

The cost of inpatient stay was calculated by multiplying the LOS (days) during the CIED infection episode by a daily bed cost of £350 per day per bed. Because of the significant additional cost, use of a Lifevest device was included separately. Cost was calculated from the duration that the Lifevest was issued to the patient, based on a monthly cost of £4,000 per patient.

Actual cost of inpatient antibiotic therapy was not prospectively recorded in the electronic database and therefore not available. However, details of OPAT were recorded as a daily tariff based on the cost of antibiotic therapy.

CRT-D.

Of the 18 patients with an infected CRT-D, seven (38.9%) underwent an ER strategy and were re-implanted after a median of 14 days (IQR 8–20); eleven (61.1%) underwent a DR strategy and were re-implanted a median of 53 days later (IQR 36–91). Of these 11 DR patients, nine (81.8%) were discharged home with a Lifevest; none of these patients had any arrhythmia detected or received therapy. One patient who had an externalized pacemaker (and was thus not suitable for a Lifevest) requested to be managed as an outpatient pending device re-implant. Another, who had had an ICD for 13 years with no therapy, requested a downgrade to a CRT-P, and therefore did not receive a Lifevest.

ICD.

Twenty-four patients had an infected ICD. Ten (41.7%) underwent an ER strategy a median of 14 days following extraction (IQR 8–16). Fourteen (58.3%) patients underwent a DR strategy and with re-implantation a median of 56 days later (IQR 20–72) (Table 2). Of these, nine (81.8%) patients received a Lifevest. Four patients who required an externalised pacemaker requested outpatient management so were unsuitable for a Lifevest, while one patient was downgraded to a pacemaker at re-implant.

Pacemakers.

Twenty-four of 42 (57.1%) PPM patients underwent an ER strategy a median of 15 days later (IQR 12–16). An externalized temporary pacemaker was required in five (20.8%) of these cases following extraction. Eighteen (42.9%) PPM patients underwent a DRstrategy and were discharged with re-implantation a median 25 days later (IQR 15–33). An externalized temporary pacemaker was used in four (22.2%) cases.

Length of hospitalisation.

When all device types were considered, total LOS was significantly shorter in patients managed via a DR strategy compared to an ER strategy (median 5 vs. 18 days, p<0.001). Those with either ICD or PPM in the DR group had significantly shorter LOS compared to similar patients in the ER group, [(ICD 3 vs. 17 days, p = 0.006) (PPM 5 vs. 21 days, p<0.001)]. Median LOS for patients with CRT-D was not significantly different (6 vs. 22 days, p = 0.964), however ER CRT-D comprised only seven patients with LOS located at both extremes of the range. Cost of hospitalization, attributable to bed costs, was significantly higher in ER patients (£6,300.00 vs £1,750.00; p<0.001) (Table 2). No patients required an intensive care unit (ICU) admission in either group.

Extraction procedure-related costs.

Although the median cost of the CIED extraction procedure was higher for ER compared to DR, the difference was not statistically significant (£2,487.59 vs £1,339.85 respectively; p = 0.250) (Table 2). This cost takes into consideration the staff, equipment and facility-related costs for operative interventions.

Outpatient antibiotic therapy.

Twenty-two (51.2%) of DR cases received OPAT prior to CIED re-implantation. Of these, one patient received intravenous OPAT; the remainder received oral treatment (flucloxacillin, n = 12; linezolid n = 5; other antibiotic n = 4). Median duration of outpatient treatment was 14 (12.5–14.0) days. Median cost was £3.47 (1.96–20.66). The staffing costs associated with administering IV antibiotics in the community were not calculated.

Cost of CIED infection episode according to device type.

When all cases were considered, the cost of infection ranged £5,139 (PPM) to £24,318 (CRT-D).

The cost of the Lifevest increased the overall costs in the DR group by a median of £10,000.00; prior to the addition of Lifevest and OPAT costs, the cost of CIED infection episodes was significantly less in the ER group (£14,241.48 vs £12,252.68, p = 0.003). However, when the cost of the Lifevest and OPAT were taken into account, the total cost of CIED infection was similar in the ER and DR groups (£14,241.48 vs £14,741.70, respectively; p = 0.491). It should be noted that not all patients in whom a defibrillator was extracted (ICD or CRT-D) in the DR group received a Lifevest whilst awaiting re-implantation. Reasons for this included incompatibility with externalised pacemakers (1 CRT-D patient, 4 ICD patients), planned downgrade at re-implant (1 CRT-D patient, 1 ICD patient) and patient choice, i.e. patients unwilling to wear the Lifevest who accepted the short-term risk of untreated ventricular arrhythmias. When the groups were split according to type of CIED extracted (CRT-D, ICD or PPM), the total cost of CIED infection episode was significantly higher in ER cases compared to DR cases for PPM devices only (£11,213.20 vs £5,139.14; p = 0.001) (Table 2).