Livestock-associated methicillin-resistant Staphylococcus aureus responsible for human colonization and infection in an area of Italy with high density of pig farming

Methicillin-resistant Staphylococcus aureus (MRSA) emerged in hospital fifty years ago and swiftly became one of the most important hospital-associated (HA) pathogens being responsible for serious infections such as pneumonia and sepsis [1], and commonly characterized by a multi-drug resistant phenotype. Distinct lineages of HA-MRSA emerged in the years and became established with different success in different geographical locations [2].

Starting from the last decade, MRSA epidemiology largely changed, due to the emergence of new MRSA lineages, responsible for infections occurring in the community among patients without known risks factors for the acquisition of MRSA and without previous hospital contacts [3]. These strains, defined community acquired (CA)-MRSA, are mainly responsible for skin and soft-tissue infections (SSTI), although deep-seated infections such as necrotizing pneumonia, sepsis and meningitis, have also been reported [3, 4]. CA-MRSA is characterized by the presence of the Panton-Valentine Leukocidin (PVL), a toxin that causes polymorphonuclear lysis and tissue necrosis [5] although its contribution to disease is still debated [6, 7].

More recently, the emergence of an MRSA clone colonizing pigs and, more rarely, other farm animals (cattle and poultry) have been reported in Europe [8‐10]. These strains, designated livestock-associated (LA)-MRSA and belonging to sequence type (ST)398, were mostly found in countries with high density of pig farming such as the Netherlands, Denmark and Germany [11, 12]. These strains were able to colonize persons working in close contact with pigs, such as farmers and veterinarians. In Netherland and in Germany, 26.5% and 24% of pig farmers were colonized by LA-MRSA, respectively [12, 13]. Wulf et al. found that 12.5% of veterinarians attending an international conference were colonized by MRSA of which 92.5% were LA-MRSA [14]. Farmers’ family members, who are not in direct contact with pigs, were colonized at a lower frequency than the farm workers, indicating that inter-human transmission may occur [15].

Since LA-MRSA was found in dust from pig holdings [16], the environment could represent a vehicle for the transmission of LA-MRSA strains from animals to humans, allowing colonization of subjects who do not work directly with animals.

Infections due to LA-MRSA have been reported occasionally in farm workers and their family members [10]. In the Netherlands and Denmark where the percentage of HA-MRSA is very low (<1%), LA-MRSA strains represent an important reservoir for human MRSA infections [10, 11]. In the Netherland, by the end of 2008, LA-MRSA ST398 accounted for 42% of all new detected MRSA [12].

In Italy, LA-MRSA ST398 prevalence in farm workers is unknown. In the study previously cited, MRSA carriage in Italian veterinarians was found to be 54% [14]. Two cases of serious infections due to LA-MRSA were described in Italian farmers in the Lombardy region: a man working on a pig farm who developed cellulitis and pyomiositis of the buttock and a man working on a dairy farm who developed necrotizing fasciitis of the neck [17, 18].

Other MRSA lineages different from ST398 were found in a European survey on the prevalence of MRSA in pig holdings. One lineage, characterized by spa type t127 and belonging to ST1 was predominantly reported from Italy [16], as shown also by other studies [19]. Franco et al. recently demonstrated that t127/ST1 isolates can be assigned to two genetically different clusters (porcine and human) and hypothesized that t127/ST1 strains could represent another lineage of LA-MRSA [20].

This study was undertaken to investigate the prevalence of LA-MRSA colonization and the occurrence of LA-MRSA infections in subjects living in an area of the Lombardy region, where livestock farming is a prominent activity.

In the colonization study, 9 out of 879 patients (1%) subjected to nasal screening prior to hospitalization were positive for MRSA. The age of patients carrying MRSA ranged from 17 to 71 years (median age 58); 4 patients were females and 5 were males. Out of the 9 patients colonized, 5 (56%) carried LA-MRSA ST398. The overall LA-MRSA colonization rate was 0.56%.

The characteristics of the MRSA isolates are shown in Table 1. The 5 LA-MRSA strains were t899/SCCmec type IVa (2 isolates), t899/SCCmec IVc (1 isolate), t108/SCCmec type V (1 isolate) and t2922/SCCmec type V (1 isolate). The other isolates obtained from carriage belonged ST8 (3 isolates) and ST5 (1 isolate). None of the colonizing strains was positive for PVL genes.

In the infection study, 20 CAI due to MRSA were detected in the span of one year. Thirteen patients were females and 7 were males. The age of patients ranged from 8 months to 85 years (median age 53). Out of the 20 isolates, 17 were obtained from SSTI, 2 from otitis externa and 1 from bloodstream infection. One of the isolates from otitis externa was t899/ST398/SCCmec IVc and was categorized as LA-MRSA. Eight strains were positive for PVL and were considered typical CA-MRSA. Of these, 3 were t008/ST8/SCCmec IVa and ACME-positive and therefore most likely corresponding to the USA300 clone. Another strain was t044/ST80/SCCmec IVc and belonged to the European clone, while 3 other strains belonged to other lineages (ST22, ST772 and ST1).

ST8 and ST1 were found among both PVL-positive and PVL-negative isolates. PVL-negative ST8 were negative for ACME, too. All ST1 isolates (4 strains), belonged to spa type t127 and harboured the sak gene; of these one carried also PVL. The PVL-positive t127 isolate carried SCCmec V while the 3 PVL-negative t127 isolates carried SCCmec IVa.

All the isolates, both from colonization and infections, were fully susceptible to vancomycin, teicoplanin, linezolid, tigecycline and mupirocin. All ST398 isolates were resistant to tetracycline. Resistance profiles to other non-beta-lactam antibiotics are shown in Table 1.

In the present study we examined MRSA strains obtained from nasal screening of subjects prior to hospital admission and MRSA strains from CAI in an Italian Region, Lombardy, with the highest density of pig farming in Italy (48.2% of all the pigs reared in the country) [21]. In particular, in the area where the study was conducted, a serious case of LA-MRSA infection was identified in 2008 [18].

Although the number of MRSA isolates obtained in this study was small, two different pictures emerged clearly: LA-MRSA, belonging to ST398, was prevalent (56%) among the admission screening MRSA isolates, while, among the infections typical PVL-positive CA-MRSA predominate and only one strain belonging to ST398 was found. The percentage of LA-MRSA out of all MRSA colonizing strains is similar to that found in patients at admission to hospital in a region with high density of animal farming in Germany [30].

Unfortunately we were unable to obtain precise information about the working activity of the colonized subjects or of the infected patients, therefore we do not know if any of them worked or lived in close contact with pigs or other farm animals. The overall percentage of ST398 colonization in our study (0.56%) is in the range (from <0.01 to 1.2%) described in the Netherlands and Germany, regarding non-exposed healthy population [12, 13].

Our study confirms that the presence of LA-MRSA in human infections is rare. LA-MRSA isolates have been reported to sporadically cause mild or serious infections and a hospital outbreak [17, 18, 31, 32]. According to data reported by Van Cleef et al., MRSA ST398 clone contributes only to a small fraction (<2%) of all MRSA isolates in humans [9]. In a framework of an European survey conducted in 2011 on invasive infections caused by S. aureus predominant clones, we found that in Italy 1 out of 97 MRSA (1.5%) strains belonged to a spa type indicative of ST398 (unpublished data), while in a previous survey carried out in 2006 no ST398 was found [2]. Therefore the ability of LA-MRSA to cause infections appears to be limited. This is probably due to the lack or rare occurrence of specific virulence factors, such as toxic shock syndrome toxic-1 (TSST-1) toxin and PVL, as shown by microarray and genomic studies [33, 34]. In addition, LA-MRSA lacks the cluster of genes for human-specific immune evasion encoded by Saϕ3 bacteriophage, and therefore is less virulent in humans than other staphylococcal clones [34, 35].

In our study on outpatient infections, a prominent role was played by typical CA-MRSA strains. The presence of a variety of CA-MRSA clones including USA300, has been observed, as already reported in other studies [4, 25, 36].

The recovery of isolates belonging to t127/ST1 is intriguing. t127/ST1 strains were recently isolated from pigs in Italy and the presence of a pig reservoir of this lineage has been hypothesized [20]. In the study by Franco et al., the t127 strains of porcine origin carried SCCmec type V, while the t127 isolates of human origin harboured SCCmec type IVa and carried the sak gene, an immune-evasion factor present almost exclusively in human strains [20]. In our study three t127 strains carried SCCmec type IVa and the sak gene, indicating human origin. The only strain carrying SCCmec type V also harbored sak and PVL and therefore its porcine or human origin cannot be hypothesized.

Other clones causing infections are likely of hospital origin, such as the PVL-negative ST8 clone, resembling the “Lyon clone”, a well adapted hospital clone that emerged in France and spread in other European countries, including Italy [37, 38]. The presence of HA-MRSA in infections in the community might be due to previous contact of the patients with health-care settings.

Italy is one of the European countries with the highest MRSA rate according to data from the EARS-Net (around 40%) and with the highest density of pig-livestock farming [16]. This is the first study evaluating the prevalence of LA-MRSA in the general population, in carriage and infections, in our country.

The presence of LA-MRSA in the general population confirms the importance of the animal reservoir and indicate the need for preventive measures to control the extent of LA-MRSA spread.