In-hospital infection (IHI) after transcatheter aortic valve implantation (TAVI) has received little attention, although it may have a significant effect on outcomes and costs because of prolonged hospital stay. Therefore, the aim of this study was to determine the incidence, type, predictors, and prognostic effects of IHI after TAVI. This study included 298 consecutive patients from 2 centers who underwent TAVI from November 2005 to November 2011. IHI during the hospital stay was defined on the basis of symptoms and signs assessed by the attending physician in the cardiac care unit or medium care unit in combination with all technical examinations performed to confirm infection. IHI after TAVI was observed in 58 patients (19.5%): urinary tract infections in 25 patients (43.1%), pneumonia in 12 patients (20.7%), and access-site infections in 7 patients (12.1%). In 12 patients (20.7%), the site of infection could not be determined, and 2 patients (3.4%) had multiple infection sites. Multivariate analysis revealed that surgical access through the femoral artery was the most important determinant of infection (odds ratio [OR] 4.18, 95% confidence interval [CI] 1.02 to 17.19), followed by perioperative major stroke (OR 3.21, 95% CI 1.01 to 9.52) and overweight (body mass index ≥25 kg/m 2 ; OR 2.27, 95% CI 1.12 to 4.59). The length of hospital stay in patients with IHIs was 15.0 days (interquartile range 8.0 to 22.0) compared with 7.0 days (interquartile range 4.0 to 10.0) in patients without infections (p <0.0001). Kaplan-Meier estimates of survival at 1 year were 76.6% and 74.4% (log-rank, p = 0.61), respectively. Unadjusted and adjusted OR analysis revealed that IHI did not predict mortality at 30 days (OR 1.27, 95% CI 0.49 to 3.30) or at 1 year (hazard ratio 1.24, 95% CI 0.68 to 2.25). In conclusion, IHI occurred in 19.5% of the patients. Patient-related and, more important, procedure-related variables play a role in the occurrence of infection, indicating that improvements in the execution of TAVI may lead to a reduction of this complication.
Transcatheter aortic valve implantation (TAVI) is increasingly used to treat patients with aortic stenosis and prohibitive risk for surgical valve replacement. Although conferring obvious benefits, TAVI is associated with a number of complications, including infection. The latter has received little attention although it may have a significant effect on outcomes and costs because of the need for additional treatment and prolonged hospital stay. The occurrence of in-hospital infection (IHI) after TAVI was anecdotally reported by Cribier et al in 2002. During the subsequent period, IHI, including sepsis, has been reported to occur in 3% to 24% of all patients. The Valve Academic Research Consortium was established to develop uniform end point criteria and definitions but did not publish criteria for infection after TAVI, except for prosthetic valve endocarditis. This is noteworthy given that along with the increase in TAVI procedures, infectious complications are frequently encountered in patients who undergo TAVI, because these patients are at particular risk because of age, co-morbid conditions, and eventually frailty. We sought to explore in more detail the frequency and determinants of infectious complications after TAVI because this information may help improve outcomes. We also sought to explore the prognostic effects of infection after TAVI on mortality at 30 days and during follow-up.
Methods
The study population consisted of 298 consecutive patients with symptomatic aortic valve stenosis who underwent TAVI from November 2005 to November 2011 at the Erasmus Medical Center (Rotterdam, The Netherlands; n = 230 patients) and Angiografia de Occidente (Cali, Colombia; n = 68 patients). At the 2 institutions, similar processes of patient and procedural planning were set up at the initiation of TAVI at each institution, as 1 of the investigators (P.P.T.d.J.) helped initiate the program in Cali and was present during all procedures from 2008 (first implantation) to 2010. This also holds for the database and data collection during the hospital stay, as previously described.
One hour before the procedure and upon completion of the procedure, prophylactic antiobiotic therapy was administered according to local practice guidelines (cefazoline 1 g at both times). If needed, antibiotic therapy was continued postprocedurally by the attending physician. All patients underwent transfemoral (n = 287), subclavian TAVI (n = 9) or transapical TAVI (n = 2) under general anesthesia (Rotterdam) or deep sedation (50% of all patients in Cali) with the 18Fr third-generation Medtronic CoreValve System (MCS; Medtronic CV Luxembourg S.a.r.l., Luxembourg) except for the first 5 patients treated in Rotterdam in 2005 and 2006, in whom the 21Fr second-generation MCS was implanted.
All data were prospectively collected and entered in a dedicated database. Source verification of the baseline data and clinical events was performed by the first (R.M.A.v.d.B.) and second (R.-.J.N.) investigators for the patients treated in Rotterdam and Cali, respectively.
Infection during the hospital stay was defined on the basis of the assessment of symptoms and signs during the daily visits of the attending physician in the cardiac care unit or medium care unit. The site of infection was categorized according to the presence of positive culture and/or clinical signs of inflammation as follows: access site, urinary tract, pneumonia, or other (undetermined) origin. The causative agent was determined using the culture report of the microbiology department. Treatment of IHI was left to the discretion of the attending physician, who was in charge of the postoperative care of TAVI patients, in consultation with the microbiologist.
All end points were selected and defined according to the Valve Academic Research Consortium. In addition, the length of hospital stay (LOS) was recorded and defined as the period between the day of the procedure and the day of discharge or in-hospital death, excluding patients who died during the procedure. In case a patient was transferred to the referring hospital after having undergone TAVI, LOS was defined as the total time spent at the treating and the referring centers. All patients, except a few, were admitted to the treating center 1 day before TAVI. The time of hospital stay before TAVI was not included in the definition of LOS.
A full blood and chemistry sample was taken before and up to 3 days after the procedure. Data on red blood cell transfusions were recorded by the institutions’ blood bank laboratories. Because packed red blood cell transfusions influence post-TAVI hemoglobin level, the modified Landefeld equation was used to estimate the corrected nadir hemoglobin level and the net hemoglobin decrease after the procedure. The definition of anemia of the World Health Organization was adopted, which defines anemia as a serum hemoglobin level <13 g/dl in men and <12 g/dl in women. Furthermore, to assess the effect of the severity of anemia, patients were divided into tertiles to assess the number of patients with mild (12.99 to 11.81 g/dl in men, 11.99 to 11.31 g/dl in women), moderate (11.80 to 10.71 g/dl in men, 11.30 to 10.51 g/dl in women), and severe (≤10.70 g/dl in men, ≤10.50 g/dl in women) anemia.
Follow-up information on the patients treated at Erasmus Medical Center was collected by first checking vital status in the civil registries every 6 months. Patients who survived were sent questionnaires to assess symptoms, (cardiac) events, and readmissions. Also, surviving patients were contacted by telephone to confirm hospital readmission and reason, after which events were verified with the treating hospital. All medical records were revised, and general practitioners were contacted when necessary. Follow-up was complete for all patients. Follow-up information on the patients treated in Colombia was obtained by regular office visits and/or telephone contact (by a dedicated local research nurse [L.C.] or doctor) with the treating physician and/or general practitioner and/or patient or family members, followed by verification of the event with the treating hospital. Follow-up was complete for all patients as previously described.
Categorical variables were compared using Pearson’s chi-square or Fisher’s exact tests, as appropriate, and are expressed as frequencies and percentages. Continuous variables were compared use Student’s t tests or Mann-Whitney U tests and are expressed as mean ± SD (in case of a normal distribution) or median (interquartile range [IQR]) (in case of a skewed distribution). Normality of distributions was assessed using the Shapiro-Wilk test. Survival curves were constructed using Kaplan-Meier estimates and compared using the log-rank test.
To assess the determinants of IHI, a univariate logistic regression analysis was first performed comparing the baseline patient and procedural characteristics between patients with and without IHI. Unadjusted odds ratios (ORs) were then calculated for all variables with p values <0.10. To study the independent predictors of 30-day mortality, logistic regression was performed. All characteristics that were significant on univariate analysis and those judged to be clinically relevant were included in the multivariate logistic regression model, taking into account the restricted number of variables. The same method was applied for the calculation of the unadjusted and adjusted odds of mortality at follow-up using Cox regression analyses. A 2-sided α level of 0.05 was used for all superiority testing. All statistical analysis were performed using SPSS version 17.0 (SPSS, Inc., Chicago, Illinois).
Results
The baseline characteristics, procedural details, and outcomes of the total population and patients with and without infection after TAVI are listed in Tables 1 to 3 . A total of 58 patients (19.5%) had IHIs. Of these infections, 43.1% were urinary tract infections, 20.7% were pneumonia, 20.7% were of undetermined origin, and 12.1% were access-site infections; 3% of patients with IHIs (n = 2) had multiple infection sites ( Figure 1 ). Clinically, patients with IHIs were more often overweight (body mass index ≥25 kg/m 2 ) compared with patients without IHIs (69.0% vs 55.0%, p = 0.05), and they also had lower hemoglobin levels before the procedure (median 11.9 g/dl [IQR 10.8 to 13.1] vs 12.4 g/dl [IQR 11.3 to 13.5], p = 0.03). There was no difference in leukocyte count at baseline between the 2 groups (median 7.1 × 10 9 cells/L [IQR 6.0 to 8.3] vs 7.0 × 10 9 cells/L [IQR 5.8 to 8.2], p = 0.76).
| Variable | Overall (n = 298) | IHI | p Value | |
|---|---|---|---|---|
| Yes (n = 58) | No (n = 240) | |||
| Age (yrs) | 80.0 ± 70.0 | 80.4 ± 7.1 | 79.9 ± 7.0 | 0.63 |
| Men | 157 (53%) | 29 (50%) | 128 (53%) | 0.65 |
| Height (cm) | 166.5 ± 9.3 | 167.3 ± 8.2 | 166.3 ± 9.5 | 0.46 |
| Weight (kg) | 72.14 ± 12.77 | 74.67 ± 13.34 | 71.53 ± 12.58 | 0.09 |
| Body mass index (kg/m 2 ) | 26.04 ± 4.21 | 26.71 ± 4.16 | 25.88 ± 4.21 | 0.18 |
| Body mass index ≥25 kg/m 2 | 126 (42%) | 40 (69%) | 132 (55%) | 0.05 |
| Body surface area (m 2 ) | 1.82 ± 0.19 | 1.86 ± 0.20 | 1.81 ± 0.19 | 0.06 |
| New York Heart Association class III or IV | 244 (82%) | 47 (81%) | 197 (82%) | 0.85 |
| Logistic European System for Cardiac Operative Risk Evaluation | 15.6 (9.2–21.9) | 16.4 (10.2–22.6) | 15.2 (8.8–21.5) | 0.66 |
| Antecedents | ||||
| Previous cerebrovascular accident | 56 (19%) | 14 (24%) | 42 (18%) | 0.25 |
| Previous myocardial infarction | 77 (26%) | 17 (29%) | 60 (25%) | 0.50 |
| Previous coronary artery bypass grafting | 76 (26%) | 13 (22%) | 63 (26%) | 0.55 |
| Previous percutaneous coronary intervention | 81 (27%) | 16 (28%) | 65 (27%) | 0.94 |
| Coronary artery disease | 162 (54%) | 32 (55%) | 130 (54%) | 0.89 |
| Diabetes mellitus | 82 (28%) | 18 (31%) | 64 (28%) | 0.50 |
| Hypertension | 193 (65%) | 37 (64%) | 156 (65%) | 0.86 |
| Glomerular filtration rate ≤60 ml/min | 203 (68%) | 36 (62.1) | 167 (70%) | 0.27 |
| Chronic hemodialysis | 12 (4%) | 2 (3%) | 10 (4%) | 0.80 |
| Chronic obstructive pulmonary disease | 102 (34%) | 19 (33%) | 83 (35%) | 0.79 |
| Peripheral vascular disease | 49 (16%) | 9 (16%) | 40 (17%) | 0.83 |
| Permanent pacemaker | 33 (11%) | 6 (10%) | 27 (11%) | 0.84 |
| Atrial fibrillation | 77 (26%) | 16 (28%) | 61 (25%) | 0.73 |
| Baseline echocardiography | ||||
| Aortic valve area (cm 2 ) | 0.67 ± 0.22 | 0.67 ± 0.20 | 0.67 ± 0.22 | 0.95 |
| Left ventricular ejection fraction (%) | 50.5 ± 14.9 | 49.8 ± 15.9 | 50.6 ± 14.7 | 0.70 |
| Aortic annular diameter (mm) | 22.39 ± 2.67 | 22.70 ± 2.83 | 22.30 ± 2.63 | 0.37 |
| Peak velocity (m/s 2 ) | 4.24 ± 0.77 | 4.28 ± 0.86 | 4.24 ± 0.75 | 0.72 |
| Peak gradient (mm Hg) | 75.0 ± 27.4 | 77.2 ± 31.7 | 74.4 ± 26.3 | 0.49 |
| Mean gradient (mm Hg) | 43.8 ± 17.0 | 45.0 ± 19.2 | 43.5 ± 16.5 | 0.59 |
| Aortic regurgitation grade ≥III | 54 (18%) | 15 (26%) | 39 (16%) | 0.09 |
| Mitral regurgitation grade ≥III | 31 (10%) | 7 (12%) | 24 (10%) | 0.64 |
| Baseline laboratory results | ||||
| Creatinin level (μmol/L) | 95.0 (72.8–117.3) | 88.0 (62.0–114.0) | 96.8 (76.8–116.8) | 0.30 |
| Hemoglobinl level (g/dl) | 12.2 (11.2–13.3) | 11.9 (10.8–13.1) | 12.4 (11.3–13.5) | 0.03 |
| Mild anemia ∗ | 52 (17%) | 11 (19%) | 41 (17%) | 0.74 |
| Moderate anemia † | 44 (15%) | 13 (22%) | 31 (13%) | 0.067 |
| Severe anemia ‡ | 54 (18%) | 9 (16%) | 45 (19%) | 0.57 |
| Leukocyte count (×10 9 /L) | 7.1 (5.9–8.30) | 7.1 (6.0–8.3) | 7.0 (5.8–8.2) | 0.76 |
∗ Mild anemia: 12.99 to 11.81 g/dl in men, 11.99 to 11.31 g/dl in women.
† Moderate anemia: 11.80 to 10.71 g/dl in men, 11.30 to 10.51 g/dl in women.
| Variable | IHI | p Value | |
|---|---|---|---|
| Yes (n = 58) | No (n = 240) | ||
| Vascular access | |||
| Surgical: femoral artery | 5 (9%) | 5 (2%) | 0.013 |
| Percutaneous: femoral artery | 50 (86%) | 227 (95%) | 0.025 |
| Surgical: subclavian artery | 1 (2%) | 6 (3%) | 0.72 |
| Percutaneous: subclavian artery | 1 (2%) | 1 (0.4%) | 0.27 |
| Surgical: transapical | 1 (2%) | 1 (0.4%) | 0.27 |
| Circulatory support | |||
| Extracorporal membrane oxygenation | 1 (2%) | 2 (1%) | 0.54 |
| Left ventricular assist device | 5 (9%) | 8 (3%) | 0.08 |
| Intra-aortic balloon pump | 0 | 8 (1%) | 0.49 |
| None | 52 (90%) | 227 (95%) | 0.17 |
| Additional interventions during TAVI | |||
| Percutaneous transluminal angioplasty, iliac artery | 1 (2%) | 5 (2%) | 0.86 |
| Percutaneous coronary intervention | 5 (9%) | 15 (6%) | 0.52 |
| Prosthesis size | |||
| Medtronic CoreValve 26 mm ∗ | 22 (38%) | 88 (37%) | 0.86 |
| Medtronic CoreValve 29 mm ∗ | 32 (55%) | 140 (58%) | 0.66 |
| Medtronic CoreValve 31 mm ∗ | 2 (3%) | 3 (1%) | 0.24 |
| Edwards Sapien 23 mm ∗ | 1 (2%) | 3 (1%) | 0.78 |
| Edwards Sapien 26 mm ∗ | 1 (2%) | 3 (1%) | 0.78 |
| Therapy-specific results | |||
| Postimplantation balloon dilatation | 5 (9%) | 44 (18%) | 0.073 |
| Valve-in-valve implantation | 2 (3%) | 12 (5%) | 0.62 |
| Ventricular perforation | 1 (2%) | 2 (1%) | 0.54 |
| Conversion to surgical aortic valve replacement | 0 | 0 | 1.00 |
| Procedure time (minutes) | 230 ± 84 | 202 ± 75 | 0.014 |
| Amount of contrast (ml) | 176 ± 87 | 167 ± 77 | 0.49 |
∗ Three patients did not undergo TAVI, 1 died during induction (anesthesia), and 1 died as a result of balloon valvuloplasty–induced left ventricular outflow tract rupture.
| Variable | IHI | p Value | |
|---|---|---|---|
| Yes (n = 58) | No (n = 240) | ||
| In-hospital clinical outcomes | |||
| 30-day or in-hospital death | |||
| All-cause | 8 (14%) | 20 (8%) | 0.20 |
| Cardiovascular | 3 (5%) | 19 (8%) | 0.47 |
| Myocardial infarction | |||
| Periprocedural (<72 h) | 1 (2%) | 2 (1%) | 0.54 |
| Spontaneous (>72 h) | 0 | 1 (0.4%) | 0.62 |
| Cerebrovascular complications | |||
| Major stroke | 6 (10%) | 11 (5%) | 0.09 |
| Minor stroke | 1 (2%) | 1 (0.4%) | 0.27 |
| Transient ischemic attack | 1 (2%) | 3 (1%) | 0.78 |
| Vascular complications | |||
| Major | 7 (12%) | 22 (9%) | 0.50 |
| Minor | 5 (9%) | 18 (8%) | 0.77 |
| Bleeding complications | |||
| Life threatening | 8 (14%) | 18 (8%) | 0.13 |
| Major | 5 (9%) | 29 (12%) | 0.46 |
| Minor | 7 (12%) | 19 (8%) | 0.32 |
| Acute kidney injury | |||
| Stage I | 14 (24%) | 30 (13%) | 0.025 |
| Stage II | 3 (5%) | 3 (1%) | 0.056 |
| Stage III | 1 (2%) | 3 (1%) | 0.78 |
| Reintervention in hospital | 0 | 2 (1%) | 0.49 |
| LOS | |||
| Total hospitalization | 15.0 (8.0–22.0) | 7.0 (4.0–10.0) | <0.001 |
| Prosthetic valve–associated complications | |||
| New permanent pacemaker requirement | 13 (22%) | 53 (22%) | 0.97 |
| Combined end points | |||
| Composite safety end point | 22 (38%) | 49 (20%) | 0.005 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
