Staphylococci account for the bulk of cardiovascular implantable electronic device (CIED) infections. However, a detailed analysis of clinical features and outcomes of CIED infections due to staphylococcal species has not been published. We retrospectively reviewed all cases of CIED infection seen at the Mayo Clinic from 1991 through 2008. Differences in device and host factors, clinical features, and patient outcomes were compared between cases of early and late Staphylococcus aureus and coagulase-negative staphylococci (CoNS) CIED infections. Of 280 cases of staphylococcal CIED infections, 43.9% were due to S. aureus and 56.0% were due to CoNS. Staphylococcus aureus CIED infection cases more frequently involved initially implanted devices. Late S. aureus CIED infection cases compared to late CoNS cases were associated with corticosteroid therapy, hemodialysis, implanted catheters, prosthetic valves, and remote sources of bacteremia. Cases of S. aureus endovascular infections had longer duration of bacteremia (56.0% vs 20.3% ≥3 days), longer hospitalization (37.4% vs 15.2% >20 days), and increased mortality (25.2% vs 9.5%) compared to cases of CoNS endovascular infections (p <0.001 for all comparisons). Overall, CoNS CIED infections compared to S. aureus were associated with a history of multiple device revisions and a higher number of total and abandoned leads at presentation (p <0.001 for all comparisons). In conclusion, CIED infections due to S. aureus and CoNS have distinct clinical features and outcomes.
An aging population with multiple co-morbid conditions and expanding indications have led to an increased implantation of cardiovascular implantable electronic devices (CIEDs) during the previous 2 decades in the United States. Benefits of CIED therapy, however, have been partly eclipsed by increased rates of CIED infection that are out of proportion to implantation rates. CIED infections can present as generator pocket infection and/or endovascular infection with or without CIED-related infective endocarditis. Staphylococcus aureus and coagulase-negative staphylococci (CoNS) account for the bulk of these infections. However, a detailed analysis of staphylococcal CIED infections based on syndrome presentation or coagulase designation has not been previously done. In the present investigation, we characterized distinct clinical syndromes and outcomes of CIED infection due to S. aureus and CoNS.
Methods
We retrospectively reviewed all cases of CIED infection at the Mayo Clinic (Rochester, Minnesota) from January 1, 1991 through December 31, 2008. Cases were identified from the Mayo Clinic Heart Rhythm Device database, surgical index, and computerized central diagnostic index. All patients previously consented to the use of their medical records for research. The Mayo Foundation institutional review board approved the study protocol.
CIED infection was defined and further classified as pocket infection or endovascular infection (bloodstream infection or CIED-related infective endocarditis) as previously described by our group. We classified source of infection in all patients according to a classification scheme developed by Friedman et al. We also recorded oxacillin susceptibility results in all cases. Species identification was not routinely done for CoNS isolates. Demographics, device characteristics, host co-morbid conditions, clinical presentation, laboratory parameters, and outcomes were compared between cases of S. aureus and CoNS. Cases that presented with CIED infection within 1 year of device implantation or revision were classified as early, and those presenting within ≥1 year after device-related procedure were classified as late. Clinical outcomes included duration of bloodstream infection, duration of hospitalization, infection relapse rate, and short- and long-term mortalities.
Frequency (percentage) and mean ± SD were used to characterize patients’ demographics, device characteristics, host co-morbid conditions, clinical presentation, and laboratory parameters. Bivariate comparisons were performed between cases of early and late S. aureus and CoNS CIED infections using Student’s t test for continuous variables and Pearson chi-square or Fisher’s exact test for categorical variables. Effect modification was evaluated using a 2-way interaction term between relevant covariates using PROC GENMOD. Kaplan–Meier curves and log-rank test were used to evaluate differences in timing of device infection between cases of S. aureus and CoNS pocket versus endovascular infection. Thirty-day and 1-year mortalities were confirmed by the Social Security Index using Accurint (San Francisco, California) in all 280 cases in December 2010, which was 2 years from the date of hospital discharge for the most recently enrolled case in the database. A 2-sided p value ≤0.05 was considered to indicate statistical significance. All analyses were conducted using SAS 9.1.3 (SAS Institute, Cary, North Carolina).
Results
Our cohort was comprised predominately of elderly white men ( Table 1 ). Of 416 cases of CIED infection during the study period, 280 (67.3%) were due to staphylococci. Of these, 123 (43.9%) were due to S. aureus and 157 (56.0%) were due to CoNS ( Figure 1 ). One hundred thirty-four cases (47.8%) were community-acquired, 112 (40.0%) health care-associated, and 34 (12.1%) nosocomial. Oxacillin resistance was seen in 29.2% of S. aureus and 47.7% of CoNS isolates; it was higher among CoNS isolates causing nosocomial and health care–associated infections (p <0.001) compared to those causing community-acquired infections ( Figure 2 ).
| Variable | All Staphylococcal CIED Infections (n = 280) | Early CIED Infection | Late CIED Infection | ||||
|---|---|---|---|---|---|---|---|
| S. aureus (n = 70) | CoNS (n = 66) | p Value | S. aureus (n = 53) | CoNS (n = 91) | p Value | ||
| Demographics | |||||||
| Age (years), mean ± SD | 69.4 ± 14.9 | 66.4 ± 16.2 | 68.8 ± 15.0 | 0.366 | 72.6 ± 13.7 | 70.3 ± 14.3 | 0.343 |
| Men | 211 (75%) | 56 (80%) | 51 (77%) | 0.698 | 39 (74%) | 65 (71%) | 0.780 |
| White | 257 (92%) | 65 (93%) | 63 (95%) | 0.520 | 48 (91%) | 81 (89%) | 0.768 |
| Body mass index (kg/m 2 ), mean ± SD | 28.5 ± 6.2 | 29.9 ± 7.5 | 28.9 ± 5.9 | 0.436 | 28.5 ± 6.0 | 27.0 ± 5.2 | 0.150 |
| Device characteristics | |||||||
| Device type | 0.172 | 0.674 | |||||
| Permanent pacemaker | 173 (62%) | 33 (46%) | 41 (62%) | 37 (70%) | 63 (69%) | ||
| Implantable cardioverter–defibrillator | 84 (30%) | 30 (43%) | 18 (27%) | 12 (23%) | 24 (26%) | ||
| Cardiac resynchronization therapy defibrillator | 22 (8%) | 7 (10%) | 7 (11%) | 4 (8%) | 4 (4%) | ||
| Cardiac resynchronization therapy pacemaker | 1 (1%) | 1 (1%) | 0 (0%) | 0 (0%) | 0 (0%) | ||
| Dual chamber ⁎ | 180 (70%) | 34 (55%) | 43 (73%) | 0.039 | 33 (67%) | 70 (80%) | 0.086 |
| Implanted at Mayo Clinic, Rochester | 89 (32%) | 37 (53%) | 19 (29%) | 0.004 | 17 (32%) | 16 (18%) | 0.046 |
| Indication for cardiovascular implantable electronic device implantation | 0.245 | 0.916 | |||||
| Heart block | 94 (34%) | 14 (20%) | 21 (32%) | 21 (40%) | 38 (42%) | ||
| Sinus node disease | 82 (29%) | 19 (27%) | 21 (32%) | 15 (28%) | 27 (30%) | ||
| Ventricular arrhythmia | 64 (23%) | 24 (34%) | 16 (24%) | 11 (21%) | 13 (14%) | ||
| Hypertrophic cardiomyopathy | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | ||
| Supraventricular tachycardia | 2 (1%) | 0 (0%) | 0 (0%) | 1 (2%) | 1 (1%) | ||
| Syncope | 8 (3%) | 2 (3%) | 3 (5%) | 1 (2%) | 2 (2%) | ||
| Other | 30 (11%) | 11 (16%) | 5 (8%) | 4 (8%) | 10 (11%) | ||
| Procedure type | <0.001 | 0.001 | |||||
| Initial implantation | 111 (40%) | 38 (54%) | 13 (20%) | 31 (58%) | 29 (32%) | ||
| Noninitial implantation | 169 (60%) | 32 (46%) | 53 (80%) | 22 (42%) | 62 (68%) | ||
| Site of generator | |||||||
| At original site of implantation | 183 (65%) | 53 (76%) | 35 (53%) | 0.005 | 43 (81%) | 52 (57%) | 0.003 |
| Left chest | 169 (60%) | 46 (66%) | 31 (47%) | 0.040 | 36 (68%) | 56 (62%) | 0.743 |
| Number of procedures † | <0.001 | 0.001 | |||||
| 1 | 107 (38%) | 33 (47%) | 12 (18%) | 33 (62%) | 29 (32%) | ||
| 2 | 87 (31%) | 23 (33%) | 26 (39%) | 9 (17%) | 29 (32%) | ||
| ≥3 | 86 (31%) | 14 (20%) | 28 (42%) | 11 (21%) | 33 (36%) | ||
| Number of leads | 2.2 ± 1.0 | 1.9 ± 0.8 | 2.6 ± 1.2 | <0.001 | 1.8 ± 0.6 | 2.5 ± 0.9 | <0.001 |
| Number of abandoned leads | <0.001 | 0.010 | |||||
| 0 | 206 (74%) | 62 (89%) | 36 (55%) | 48 (91%) | 60 (66%) | ||
| 1 | 40 (14%) | 4 (6%) | 21 (32%) | 2 (4%) | 13 (14%) | ||
| 2 | 25 (9%) | 3 (4%) | 4 (6%) | 3 (6%) | 15 (16%) | ||
| ≥3 | 9 (3%) | 1 (1%) | 5 (8%) | 0 (0%) | 3 (3%) | ||
| Recent device manipulation ‡ | 27 (10%) | 10 (14%) | 2 (3%) | 0.020 | 2 (4%) | 13 (14%) | 0.051 |
| Previous device infection | 33 (12%) | 2 (3%) | 8 (12%) | 0.038 | 4 (8%) | 19 (21%) | 0.036 |
| Co-morbid conditions | |||||||
| Charlson Co-Morbidity Index | 0.012 | 0.001 | |||||
| ≤3 | 205 (73%) | 44 (63%) | 57 (86%) | 29 (55%) | 75 (82%) | ||
| 4–5 | 41 (15%) | 15 (21%) | 5 (8%) | 10 (19%) | 11 (12%) | ||
| 6–7 | 25 (9%) | 8 (11%) | 4 (6%) | 9 (17%) | 4 (4%) | ||
| ≥8 | 9 (3%) | 3 (4%) | 0 (0%) | 5 (9%) | 1 (1%) | ||
| Autoimmune disease | 16 (6%) | 4 (6%) | 4 (6%) | 0.931 | 3 (6%) | 5 (5%) | 1.000 |
| Immunosuppressive therapy | 8 (3%) | 1 (1%) | 3 (5%) | 0.355 | 3 (6%) | 1 (1%) | 0.141 |
| Corticosteroid therapy | 20 (7%) | 4 (6%) | 1 (2%) | 0.366 | 10 (19%) | 5 (5%) | 0.011 |
| Hemodialysis | 19 (7%) | 3 (4%) | 3 (5%) | 1.000 | 9 (17%) | 4 (4%) | 0.015 |
| Long-term skin condition | 23 (8%) | 7 (10%) | 2 (3%) | 0.166 | 8 (15%) | 6 (7%) | 0.096 |
| Implanted central venous catheters | 19 (6%) | 4 (6%) | 3 (5%) | 1.000 | 8 (15%) | 4 (4%) | 0.031 |
| Prosthetic heart valve | 41 (15%) | 10 (14%) | 13 (20%) | 0.400 | 11 (21%) | 7 (8%) | 0.022 |
| Vascular graft | 8 (3%) | 3 (4%) | 3 (5%) | 1.000 | 1 (2%) | 1 (1%) | 1.000 |
| Remote infection | 44 (16%) | 10 (14%) | 4 (6%) | 0.114 | 21 (40%) | 9 (10%) | <0.001 |
| Clinical presentation | |||||||
| Fever | 124 (44%) | 44 (63%) | 18 (27%) | <0.001 | 38 (72%) | 24 (26%) | <0.001 |
| Tachycardia | 30 (11%) | 16 (23%) | 6 (9%) | 0.029 | 6 (11%) | 2 (2%) | 0.051 |
| Chills | 92 (33%) | 33 (47%) | 13 (20%) | <0.001 | 27 (51%) | 19 (21%) | <0.001 |
| Diaphoresis | 44 (16%) | 17 (24%) | 6 (9%) | 0.018 | 9 (17%) | 12 (13%) | 0.533 |
| Malaise | 100 (36%) | 36 (51%) | 16 (24%) | 0.001 | 28 (53%) | 20 (22%) | <0.001 |
| Nausea | 18 (6%) | 5 (7%) | 2 (3%) | 0.442 | 8 (15%) | 3 (3%) | 0.018 |
| Anorexia | 36 (13%) | 14 (20%) | 4 (6%) | 0.021 | 10 (19%) | 8 (9%) | 0.077 |
| Hypotension | 40 (14%) | 19 (27%) | 2 (3%) | <0.001 | 15 (28%) | 4 (4%) | <0.001 |
| Congestive heart failure symptoms | 80 (29%) | 26 (37%) | 17 (25%) | 0.153 | 20 (38%) | 17 (19%) | 0.011 |
| Laboratory parameters | |||||||
| Leukocytosis | 114 (40%) | 37 (53%) | 22 (33%) | 0.021 | 31 (58%) | 24 (26%) | <0.001 |
| Creatinine (mg/dl) | 0.015 | 0.007 | |||||
| ≤1 | 67 (24%) | 19 (27%) | 12 (18%) | 7 (13%) | 29 (32%) | ||
| 1.1–1.2 | 80 (29%) | 18 (26%) | 27 (41%) | 10 (19%) | 25 (27%) | ||
| 1.21–1.55 | 60 (22%) | 11 (16%) | 18 (27%) | 13 (25%) | 18 (20%) | ||
| ≥1.56 | 73 (26%) | 22 (31%) | 9 (14%) | 23 (43%) | 19 (21%) | ||
| Outcomes | |||||||
| Duration of bloodstream infection (days) | <0.001 | <0.001 | |||||
| 0 | 136 (49%) | 24 (34%) | 44 (67%) | 8 (15%) | 60 (66%) | ||
| 1–2 | 43 (15%) | 12 (17%) | 11 (17%) | 10 (19%) | 10 (11%) | ||
| 3–5 | 62 (22%) | 24 (34%) | 8 (12%) | 17 (32%) | 13 (14%) | ||
| ≥6 | 39 (14%) | 10 (14%) | 3 (5%) | 18 (34%) | 8 (9%) | ||
| Duration of hospitalization (days) | 0.052 | 0.001 | |||||
| <9 | 73 (26%) | 10 (14%) | 15 (23%) | 13 (25%) | 35 (38%) | ||
| 9–14 | 67 (24%) | 22 (31%) | 22 (33%) | 4 (8%) | 19 (21%) | ||
| 15–20 | 70 (25%) | 15 (21%) | 20 (30%) | 13 (25%) | 22 (24%) | ||
| >20 | 70 (25%) | 23 (33%) | 9 (14%) | 23 (43%) | 15 (16%) | ||
| Cardiovascular implantable electronic device infection relapse rate § | 10 (4%) | 5 (8%) | 2 (3%) | 0.440 | 2 (4%) | 1 (1%) | 0.250 |
| Mortality (days) | |||||||
| 30 | 17 (6%) | 4 (6%) | 2 (3%) | 0.681 | 7 (13%) | 4 (4%) | 0.099 |
| 365 | 46 (16%) | 12 (17%) | 5 (8%) | 0.091 | 19 (36%) | 10 (11%) | <0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
