Methicillin-resistant Staphylococcus aureus (MRSA) catheter-related bacteraemia in haemodialysis patients

Study period and patients

This was post-hoc analysis of a prospective multicentre study of all consecutive episodes of MRSA bacteraemia in hospitalized adult patients occurring from June 2008 to December 2009 at 21 Spanish hospitals. The following information was collected from medical records using a standardized protocol: demographic characteristics, comorbidities, clinical features, antibiotic therapy, and outcomes. Patients with catheter-related bacteraemia (CRB) were selected for the study, and those undergoing haemodialysis (HD-CRB) were compared with those who were not (non-HD-CRB).

Definitions

MRSA bacteraemia was defined as the presence of at least one positive blood culture for MRSA in a patient with clinical signs and symptoms of sepsis [15]. Vascular catheter-related bacteraemia was diagnosed using clinical and microbiological criteria defined by the guidelines of the Infectious Diseases Society of America [16]. It was considered when MRSA grew from at least one percutaneous blood sample culture and from a culture of the catheter tip, or when it grew in two blood samples for culture (one from a catheter hub and the other from a peripheral vein) that met criteria for quantitative blood cultures or differential time to positivity in a patient with accompanying clinical signs of sepsis and no other apparent source of infection. Complicated bacteraemia was defined as those episodes with positive follow-up blood cultures performed at 2–4 days after the beginning of adequate therapy and/or with evidence of metastatic infection or endocarditis [17]. Comorbidity was measured by the Charlson score, as described elsewhere [18]. Patients were classified into three categories on the McCabe and Jackson scale [19] according to their prognosis of survival before the MRSA bacteraemia: rapidly fatal (death expected within the following year), ultimately fatal (death expected within a period of 1 to 5 years), and non fatal (life expectancy of > 5 years). Severity of sepsis in the acute condition was assessed by the Pitt score [20]. Three acquisition categories were considered according to the Friedman criteria [21]: nosocomial, healthcare-related, and community-acquired bacteraemia. However, since all cases had healthcare exposure (i.e., dialysis) prior to culture, no community-associated cases were included. Distant extension was diagnosed in the presence of at least one distant infection secondary to blood spread seeding. The empirical antibiotic was defined as the antibiotic administered in the first 48 h after a positive blood culture was drawn, and it was considered appropriate if the strain was susceptible to at least one of the antibiotics administered according to the current CLSI breakpoints [22]. All patients were followed up to 4 weeks after completion of antibiotic treatment whether they were hospitalized or not. Recurrence within this period of time was defined as the isolation of MRSA in blood cultures after documented negative blood cultures or a newly diagnosed metastatic focus of the bacteremic past infection. Overall mortality was defined as death from any cause occurring in the 30 days and early mortality (EM) was defined for patients who died within the first two days after the onset of MRSA bacteraemia.

Susceptibility testing and molecular epidemiology of MRSA isolates

MRSA strains were identified in each hospital, where preliminary susceptibility tests were performed. Isolates were then sent to a central reference laboratory. All S. aureus were identified by latex agglutination (Pastorex Staph-plus, Bio-Rad Laboratories, Madrid, Spain) and DNase production (DNasE-test Agar, BioMérieux, Marcy l’Etoile, France). Antimicrobial susceptibility was tested by the disc diffusion method according to the CLSI guidelines [22]. MICs were determined by the microdilution method in accordance with CLSI criteria by using commercial panels (ESTEN 2009, SensititreTM, Izasa, Barcelona, Spain) read visually. Vancomycin E-test (BioMérieux) MICs were determined using a 0.5 McFarland inoculum streaked evenly with a swab onto Mueller-Hinton agar plates [23]. PFGE was performed after SmaI restriction of chromosomal DNA [24]. Restriction patterns were interpreted in accordance with criteria published elsewhere [25]. Representative isolates of each PFGE type and subtype were studied to determine the Multilocus Sequence Type (MLST) [26] and the Staphylococcal Chromosome Cassette mec (SCCmec) types [27]. MLSTs and SCCmec types were further inferred for all the strains. The agr polymorphism and the presence of genes encoding class S (lukS-PV) and class F (lukF-PV) proteins for Panton–Valentine Leukocidin (PVL) were studied by PCR in all the isolates, following the methodology described elsewhere [28, 29].

Statistical analysis

Continuous variables were compared using the Student’s t-test or the Mann–Whitney U-test as appropriate. Categorical data were compared using Fisher’s exact or Chi-squared tests. Analyses were performed using SPSS v15 (SPSS Inc., Chicago, IL, USA).

Ethical considerations

This observational study was conducted in accordance with the Declaration of Helsinki and was approved by the Spanish Network for Research in Infectious Diseases (REIPI) and by the Institutional Review Board at each participating centre: H. Arnau de Vilanova, Lleida; H. Universitari de Bellvitge, Barcelona; H. de Burgos, Burgos; H. Clìnic, Barcelona; H. Universitario 12 de Octubre, Madrid; H. de Cruces, Barakaldo; H. de Donostia, Donostia; H. General Universitario Gregorio Marañón, Madrid; H. Universitari Joan XXIII, Tarragona; H. Universitari del Mar, Barcelona; H. Universitario Marques de Valdecilla, Santander; H. Universitari Mutua de Terrassa, Terrassa; H. del Parc Taulí, Sabadell; H. Universitario Ramon y Cajal, Madrid; H. Universitario Reina Sofía, Córdoba; H. San Pedro de la Rioja, Logroño; H. de la Santa Creu i Sant Pau, Barcelona; H. Universitari Son Espases, Palma de Mallorca; H. Universitario Virgen Macarena, Sevilla; H. Universitario Virgen del Rocío, Sevilla; H. Universitari Vall d’Hebrón, Barcelona. To protect personal privacy, identifying information of each patient in the electronic database was encrypted. Informed consent was waived by the Clinical Research Ethics Committee because no intervention was involved and no information able to identify the patient was included.

Characteristics of patients

From a total of 579 episodes of MRSA bacteraemia, 218 (37.7 %) were catheter-related (Fig. 1): 34 (15.6 %) occurred in the HD-CRB group and 184 (84.4 %) in the non-HD-CRB group. The clinical and microbiological characteristics of patients are shown in Table 1. The comparison between the groups did not find significant differences in gender or age distribution. All HD-CRB patients acquired the infection at dialysis centres (chronic ambulatory haemodialysis), while the acquisition in non-HD-CRB patients was nosocomial in 85.3 % of cases (p < .001). Comorbidities measured by the Charlson score were higher in the HD-CRB group (Charlson ≥4: n = 25, 73.5 % vs. n = 85, 46.2 %, p = .003) while the severity of the underlying disease according to the McCabe scale was lower in this group (McCabe ≥2: n = 13, 38.2 % vs. n = 97, 52.7 %, p = .120). No differences were observed between groups in the severity of bacteraemia measured by the Pitt score.

	HD-CRB n = 34 (100 %)	Non-HD-CRB n = 184 (100 %)	p
Male Sex	23 (67.6)	113 (61.4)	.491
Age > 70 y	16 (47.1)	92 (50)	.753
Acquisition
Health-care	34 (100)	27 (14.7)	<.001
Nosocomial	0	157 (85.3)	<.001
Charlson score > 4	25 (73.5)	85 (46.2)	.003
McCabe scale > 2	13 (38.2)	97 (52.7)	.120
Pitt score > 3	6 (17.6)	55 (29.9)	.144
Microbiology
agr typea I	3 (10)	44 (24.9)	.082
II	27 (90)	132 (74.6)	.354
PFGE typea 2	25 (83.3)	107 (60.5)	.091
4	2 (6.7)	18 (10.2)	.688
Other	3 (10)	52(29.3)	.046
Clonal Complexb
5	28 (93.3)	128 (74.4)	.128
8	0	14 (8.1)	.199
22	2 (6.7)	17 (9.9)	.759
Other	0	13 (7.6)	.774
PVLa	0	2 (1.1)	1
Microdilution Vancomycin MICa > = 1.5 mg/L	0	(4.9)	.363
E-test Vancomycin MICa > = 1.5 mg/L	19 (63.3)	78 (44.1)	.051

^aData available in 207 patients

^bData available in 202 patients. Within Clonal Complex 5, 124 isolates belonged to the ST125 clone (HD-CRB: 25/83.3 % vs. Non-HD-CRB: 99/57.6 %; p = .013) whereas 9 isolates belonged to the ST146 clone (HD-CRB: 2/6.7 % vs. Non-HD-CRB: 7/4 %; p = .875)

Abbreviations: HD-CRB Haemodialysis catheter-related bacteraemia, Non-HD-CRB Non-haemodialysis catheter-related bacteraemia, agr Accessory gene regulator, PFGE Pulsed field gel electrophoresis, PVL Panton Valentine Leukocidin, MIC Minimum inhibitory concentration

Microbiologic studies

A dominant PFGE genotype (pulse-type 2) was found in both groups (Table 1), with all isolates belonging to Clonal Complex (CC) 5 (ST125 and ST146). They carried a SCCmec element type IV and agr type 2. PVL was positive in only two isolates in the non-HD-CRB group, both belonging to the USA300 clone. No isolates had vancomycin MIC ≥1.5 mg/L tested with the broth microdilution method in the HD-CRB group, compared with seven isolates (5 %) in the non-HD-CRB group (p = .60). However, when MIC was assessed by the E-test, 19 patients (63.3 %) in the HD-CRB group had vancomycin MIC ≥1.5 mg/L, and 78 (44.1 %) in the non-HD-CRB (p = .051).

Treatment and outcomes