Abstract
In this retrospective cohort study, we evaluated the predictive value of methicillin-resistant Staphylococcus aureus (MRSA) nasal swabs for the development of MRSA infections in patients with left ventricular assist devices. In 106 patients, the MRSA nasal swab had a negative predictive value of 92.9% demonstrating a potential role in antibiotic de-escalation.
Heart failure remains the leading cause of morbidity and mortality in the United States. 1 The utilization of durable mechanical circulatory support (MCS) has provided a life-saving means for patients by providing hemodynamic support as a bridge to transplant, bridge to decision, or destination therapy. 2 Over the past decade, ∼26,688 continuous-flow left ventricular assist devices (LVAD) have been implanted in the United States. 3 Long-term complications of durable MCS devices include stroke, bleeding, and infection. Infections in these patients may ultimately lead to increased hospital admissions, morbidity, and mortality. 2,4–6 Nasal colonization with methicillin-resistant S. aureus (MRSA) is a risk factor for development of clinical MRSA infections. 7,8 Thus, 94% of LVAD centers perform nasal colonization surveillance swabs for MRSA and 92% use antiseptics to reduce colonizing bacteria before MCS implant surgery. 2 Despite the frequent surveillance of MRSA nasal colonization, limited studies have been conducted assessing its effectiveness and utility in the LVAD population. We evaluated the impact of MRSA nasal colonization on infectious outcomes in patients with an LVAD.
Methods
This single-center, retrospective, cohort study was performed at a large, academic medical center. Patients were screened for inclusion if they were ≥18 years of age with an LVAD implantation between August 1, 2014, to July 1, 2019. These patients were evaluated for study outcomes from time of LVAD implantation until July 1, 2020. Patients were excluded if they had a positive MRSA nasal swab within 7 days prior to LVAD implantation and received MRSA decolonization, if they died during index hospitalization, if they had a nondurable MCS, or if no MRSA nasal swab was obtained within 365 days of their LVAD implantation.
The primary outcome of this study was to determine the prognostic value of MRSA nasal swabs for the development of MRSA LVAD-specific or LVAD-related infections by evaluating sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). The secondary outcomes were to assess the prevalence of LVAD-specific or LVAD-related infections, time to infection, hospital readmission rate, and mortality rate. A comprehensive electronic medical record review was conducted and data collection was performed by manual chart review. Baseline characteristics were assessed during the index admission for LVAD implantation. Outcomes were assessed from the time of first positive MRSA nasal swab following LVAD implantation or first negative MRSA nasal swab if a positive was not detected.
Definitions
An infection was defined according to the International Society for Heart and Lung Transplantation (ISHLT) criteria as the development of a positive bacterial culture at our institution. 2 An LVAD-specific infection involved any aspect of the device, including pump, cannula, pocket, or driveline. An LVAD-related infection occurs in patients without MCS, including endocarditis, mediastinitis, bacteremia, and intra-abdominal infection. 2 MRSA nasal swab surveillance and routine culture collection were not obtained at regular intervals at our institution. Additionally, methicillin-sensitive S. aureus (MSSA) nasal swab surveillance was not available during the study period. MRSA decolonization was defined as the use of mupirocin ointment or chlorhexidine solution prior to LVAD implantation, and no standardized protocol was implemented. 2
Statistical analysis
Statistical analyses were conducted using SPSS version 22 software (IBM, Armonk, NY). Baseline characteristics were analyzed using descriptive statistics. Outcomes with dichotomous variables were analyzed using the Fisher exact test, and continuous variables were analyzed using the Mann-Whitney U test. For all analyses, a 2-sided P value <.05 was considered significant.
Results
In total, 125 patients were screened for enrollment, with 106 patients included in the analysis. Baseline characteristics assessed during the index admission for LVAD implantation are displayed in Table 1. Overall, 6.6% patients had a positive MRSA nasal swab for a median follow-up period of 1.94 years. The MRSA nasal swab test demonstrated the following diagnostic performance characteristics for detecting culture proven MRSA: sensitivity (22.2%), specificity (94.8%), PPV (28.6%), and NPV (92.9%). Comparing positive and negative MRSA nasal swabs, we detected similar rates of LVAD-related or LVAD-specific infections, and we detected no difference in additional secondary outcomes (Table 2). Moreover, 52 patients (49.1%) received MRSA decolonization prior to LVAD implantation, and among these patients, 5.6% had a subsequent positive MRSA nasal swab. We detected a higher prevalence of MRSA infection development in patients who did not receive MRSA decolonization (13.5%) compared to those who did (3.7%).
Table 1.
Baseline Characteristics
| Characteristic | Total (n= 106) |
|---|---|
| Age, median y (IQR) | 53 (44–60) |
| Sex, male, no. (%) | 70 (66.0) |
| Weight, median kg (IQR) | 93 (77–108) |
| Height, median cm (IQR) | 175 (168–183) |
| BMI, median kg/m2 (IQR) | 29 (26–35) |
| Ethnicity, no. (%) | |
| African American | 59 (55.7) |
| White | 44 (41.5) |
| Other | 3 (2.8) |
| Diabetes, no. (%) | 48 (45.3) |
| WBC, median K/mm3 (IQR) | 7.5 (5.9–9.8) |
| Serum creatinine, median mg/dL (IQR) | 1.4 (1.1–1.9) |
| Renal replacement therapy, no. (%) | 17 (16.0) |
| eGFR, median mL/min/1.73 m2 (IQR) | 53 (32–72) |
| CKD category, no. (%) | |
| Stage 1 | 12 (11.3) |
| Stage 2 | 27 (25.5) |
| Stage 3a | 18 (17.0) |
| Stage 3b | 3 (2.8) |
| Stage 4 | 29 (27.4) |
| Stage 5 | 17 (16.0) |
| Cause of heart failure, no. (%) | |
| Ischemic cardiomyopathy | 27 (25.5) |
| Nonischemic cardiomyopathy | 74 (69.8) |
| Postpartum cardiomyopathy | 5 (4.7) |
| LVAD type, no. (%) | |
| HeartMate II | 46 (43.4) |
| HeartMate 3 | 41 (38.7) |
| HVAD | 19 (17.9) |
| Destination therapy | 86 (81.1) |
| Bridge to transplant | 20 (18.9) |
| Time to extubation, median d (IQR) | 0 (1–2) |
| MRSA decolonization, no. (%) | 52 (49.1) |
| Any culture ordered, no. (%) | 69 (65.1) |
| Blood culture ordered, no. (%) | 59 (55.7) |
| Positive blood culture (n = 59), no. (%) | 6 (10.2) |
| MRSA | 0 (0) |
| MSSA | 1 (16.7) |
| Other | 5 (83.3) |
| Discharged on oral antibiotic, no. (%) | 7 (6.6) |
| Discharged on intravenous antibiotic, no. (%) | 2 (1.9) |
| Follow-up period, median y (IQR) | 1.94 (1.05–3.07) |
Note. BMI, body mass index; WBC, white blood cell count; eGFR, estimated glomerular filtration rate; CKD, chronic kidney disease; LVAD, left ventricular assist device; MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-sensitive Staphylococcus aureus, IQR, interquartile range.
All baseline characteristics were obtained from index hospitalization for LVAD implantation.
Table 2.
Secondary Outcomes
| Outcome | Positive MRSA swab (n = 7) | Negative MRSA swab (n = 99) | P Value |
|---|---|---|---|
| LVAD-related or LVAD-specific infection, no. (%) | 4 (57.1) | 57 (57.6) | >.99 |
| Time to infection, median d (IQR) | 37 (5–205) | 226 (121–437) | .122 |
| Infection type, no. (%)a | |||
| Driveline | 2 (50.0) | 36 (63.2) | .628 |
| 0 (0) | 1 (1.8) | >.99 | |
| Blood | 1 (25.0) | 11 (19.3) | >.99 |
| Other | 1 (25.0) | 9 (15.8) | .521 |
| MRSA infection, no. (%) | 2 (28.6) | 7 (7.1) | .108 |
| MSSA infection, no. (%) | 0 (0) | 24 (24.2) | .346 |
| Surgical intervention, no. (%) a | 0 (0) | 4 (7.0) | >.99 |
| Discharged on oral antibiotics, no. (%) a | 1 (25.0) | 33 (57.9) | .313 |
| Discharged on intravenous antibiotics, no. (%) a | 2 (50.0) | 17 (29.8) | .582 |
| Subsequent infection-related hospital readmissions median (IQR) | 1 (0–4) | 0 (0–2) | .209 |
| Hospital readmissions, median (IQR) | 6 (3–8) | 4 (2–8) | .333 |
| Heart transplant, no. (%) | 1 (14.3) | 23 (23.2) | >.99 |
| Mortality, no. (%) | 3 (42.9) | 26 (26.3) | .389 |
| Lost to follow up, no. (%) | 0 (0) | 3 (3.0) | >.99 |
Note. MRSA, methicillin-resistant Staphylococcus aureus; LVAD, left ventricular assist device; MSSA, methicillin-sensitive Staphylococcus aureus; IQR, interquartile range
In the analysis, 4 patients had a positive MRSA swab and 57 patients had a negative MRSA swab.
Discussion
This study was designed to determine the utility of MRSA nasal colonization as a risk factor for the development of LVAD-related or LVAD-specific infections. The use of MRSA nasal screening to rule out MRSA respiratory infections has led to improvements in early antimicrobial de-escalation. However, its clinical utility in other infection types remains unclear. Thus far, no studies assessed the predictive value of MRSA infections in patients with LVADs. 7,8 In 82 patients who received an LVAD, Papathanasiou et al 9 found that colonization with a multidrug-resistant bacteria prior to implantation was associated with higher likelihood of death from infection in 34.1% of patients. However, only 7 of these patients were colonized with MRSA. 9 Nurjadi et al 10 evaluated nasal colonization with S. aureus in 49 patients with durable ventricular assist devices and found that S. aureus nasal colonization was associated with an increased incidence of infection by S. aureus during the first year after implantation. In comparison to these studies, our study focused on determining the clinical utility of MRSA nasal swabs for predicting specifically MRSA infections. We included a larger sample size, and our study had a longer follow-up period. Although the MRSA nasal swab demonstrated low sensitivity, its high NPV supports the use of MRSA nasal swabs as a tool for early antimicrobial de-escalation in a patient population at high risk for developing drug resistance. In our study, 7 patients with a negative MRSA nasal swab went on to develop a MRSA infection; however, the average time from collection to MRSA infection was 178 days. This finding highlights the importance of implementing standardized surveillance.
The 2015 ISHLT guidelines for prevention and management strategies for MCS infection recommend considering utilization of mupirocin and chlorhexidine before MCS implantation for patients with MRSA colonization. 2 Our study showed that patients who received MRSA decolonization had a lower prevalence of development of a MRSA infection. MRSA decolonization was implemented with the use of either mupirocin ointment or chlorhexidine solution, which are relatively inexpensive interventions, suggesting a benefit of implementing universal decolonization protocols.
This study had several limitations. Due to the retrospective study design, our results were limited by its observational nature. Microbiologic data were also limited to our institution, allowing for the possibility of unidentified infection development and outcomes. MRSA nasal swab surveillance was not consistent during the study period, and MSSA nasal swab surveillance was not available. Future studies should be conducted to address these limitations, including implementation of MSSA nasal swab surveillance, routine nasal swab screenings, and standardization of decolonization.
For patients with an LVAD, our study demonstrated that MRSA nasal swabs have a high NPV for detecting LVAD-specific and LVAD-related MRSA infections. Overall, the utilization of MRSA nasal swabs for de-escalation of antimicrobial therapy for LVAD-associated infections should be considered. Additionally, universal MRSA decolonization may have a role in improving outcomes in this patient population. More studies are needed to definitively assess the utilization of MRSA nasal swabs in patients with LVADs.
Acknowledgments
Financial support
No financial support was provided relevant to this article.
Conflicts of interest
All authors report no conflicts of interest relevant to this article.
References
- 1. Parch J, Powell C. No longer failing to treat heart failure: a guideline update review. JAAPA 2019;32:11–15. [DOI] [PubMed] [Google Scholar]
- 2. Kusne S, Mooney M, Danziger-Isakov L, et al. An ISHLT consensus document for prevention and management strategies for mechanical circulatory support infection. J Heart Lung Transpl 2017;36:1137–1153. [DOI] [PubMed] [Google Scholar]
- 3. Shah P, Yuzefpolskaya M, Hickey GW, et al. Twelfth Interagency Registry for Mechanically Assisted Circulatory Support Report: readmissions after left ventricular assist device. Ann Thorac Surg 2022;113:722–737. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Hannan MM, Xie R, Cowger J, et al. Epidemiology of infection in mechanical circulatory support: a global analysis from the ISHLT Mechanically Assisted Circulatory Support Registry. J Heart Lung Transpl 2019;38:364–373. [DOI] [PubMed] [Google Scholar]
- 5. Ali JM, Abu-Omar Y. Complications associated with mechanical circulatory support. Ann Transl Med 2020;8:835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kirklin JK, Pagani FD, Kormos RL, et al. Eigth annual INTERMACS report: special focus on framing the impact of adverse events. J Heart Lung Transpl 2017;36:1080–1086. [DOI] [PubMed] [Google Scholar]
- 7. Dangerfield B, Chung A, Webb B, Seville MT. Predictive value of methicillin-resistant Staphylococcus aureus (MRSA) nasal swab PCR assay for MRSA pneumonia. Antimicrob Agents Chemother 2014;58:859–864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Carr AL, Daley MJ, Merkel KG, Rose DT. Clinical utility of methicillin-resistant Staphylococcus aureus nasal screening for antimicrobial stewardship: a review of current literature. Pharmacotherapy 2018;38:1216–1228. [DOI] [PubMed] [Google Scholar]
- 9. Papathanasiou M, Pohl J, Janosi RA, et al. Colonization with multiresistant bacteria: impact on ventricular assist device patients. Ann Thorac Surg 2018;105:557–563. [DOI] [PubMed] [Google Scholar]
- 10. Nurjadi D, Last K, Klein S, et al. Nasal colonization with Staphylococcus aureus is a risk factor for ventricular assist device infection in the first year after implantation: a prospective, single-centre, cohort study. J Infect 2020;80:511–518. [DOI] [PubMed] [Google Scholar]