Food-borne salmonellosis continues to be an important public health problem both in developed and developing countries [1] -[3] . Gastroenteritis is the most frequent illness derived from this infection and the use of stool or rectal swab samples are the main specimens used in diagnostic clinical laboratories. However, the presence of saprophytic flora, especially other coliforms in stool samples, has traditionally made isolation of Salmonella spp. difficult. Consequently, there is no ideal method for diagnosis of Salmonella infection and the development of highly selective media continues to be critical importance for diagnostic and treatment purposes. Conventional media base identification relies on the production of H₂S and pH indicator to detect lactose fermentation while bile salts and dyes inhibit saprophytic flora. However, these media types are inadequate to differentiate Salmonella spp. from other members of the Enterobacteriaceae, such as Proteus spp. and Citrobacter spp. [4] . In decreasing order of selectivity, MacConkey agar (MAC), Hektoen Enteric (HE) agar, Salmonella-Shigella (SS) agar, and Xylose-Lysine- Desoxycolate (XLD) agar are some of the more commonly used plating media in clinical laboratories.

Recently, several agar media containing chromogenic substrates for Salmonella-specific enzymes have been developed, i.e. Rambach agar [5] , Salmonella Detection and Identification Medium (SM-ID; bioM’erieux, France), and ABC medium (Lab M. Ltd., Bury, United Kingdom). However, chromogenic agar media are less sensitive, but have greater specificity when compared to conventional media [6] [7] . Analternative method implementing enriched semisolid motility medium for isolating Salmonella from stool samples has been previously investigated. Jones & Handley used a semisolid medium containing cacotheline as a selective medium after initial culture of suspected material in a broth containing hydroquinone [8] . Ino & Graber used passage through semisolid medium in a “U” tube to recover Salmonella from cultures contaminated with Pseudomonas aeruginosa [9] . Stuart and Pivnick used selective motility (SM) medium modified with selenite F broth to make a semisolid medium contained in motility tubes to isolate Salmonella spp. from stool specimens. Data showed that the SM technique was equal or superior to those standard routine procedures employed in clinical laboratories and primary isolation of Salmonella from SM enrichment were relatively free from competing floras. Disadvantages to using the SM medium were the media was not shelf stable, could not be autoclaved and the migration time of Salmonella through the medium were lengthy [10] .

Harper and Shortidge modified Shigella-Salmonella medium to a semisolid agar contained in Craigie tube to isolate Salmonella from stool specimens and found that 90% of specimens from which Salmonella were isolated were positive. Over 40% of these specimens would have been reported negative if the method had not been in use [11] . Modified Semisolid Rappaport-Vassiliadis (MSRV) medium is a modification of Rappaport-Vassiliadis enrichment broth for detecting motile Salmonella spp. in food products and feces [12] [13] . The original study on MSRV medium revealed a semisolid could be used as a rapid and sensitive method for isolation of Salmonella spp. from food products following pre-enrichment or selective enrichment [14] [15] . The efficacy of MSRV medium is based on the ability of Salmonella spp. to migrate through the selective medium ahead of other competing motile bacteria, thus producing opaque halos of growth surrounding the inoculating area. MSRV can be used as a plating medium for isolating Salmonella spp. (other than S. Typhi and S. Paratyphi A) from stool specimens with high sensitivity and specificity relative to chromogenic and conventional agar media [16] [17] .

In order to evaluate the performance of MSRV, XLD agar media, and enrichment broths (e.g., BPW and RV) with our standard method for Salmonella spp. detection (MAC and HE agars and SB enrichment broth), a comparative study was conducted by direct plating for the isolation of Salmonella spp. Additionally, samples were enriched in SB broth (routine standard medium), BPW and RV broths, and prior to plating.

Of the 1096 stool samples processed by direct plating, a total of 102 stools were positive for Salmonella on at least one plating medium in which 87 (85.3%), 35 (34.3%), 30 (29.4%), and 35 (34.3%) were isolated from MSRV, MAC, HE, and XLD, respectively (Figure 1). A total 299/1297 stools samples were Salmonella-positive on at least one plating medium after enrichment procedure (Figure 2). The number of samples containing Salmonella spp. using RV, BPW, and SB Enrichment broths were 197 (86.0%), 180 (78.6%), and 177 (77.3%), respectively. All Salmonella isolates were nontyphi Salmonella.

Direct plating on MSRV yielded significantly more Salmonella isolates when compared to MAC, HE, and XLD (p-value < 0.0001) (Table 1). There was no statistical significance in numbers of Salmonella-positive stools when comparing MAC, HE and XLD. Enrichment with RV yielded significantly more Salmonella isolates when compared to BPW (p-value = 0.0405) and SB (p-value = 0.0126) (Table 1). Use of MSRV in combination with MAC and HE when compared to MAC + HE increased recoverability of Salmonella isolates by 3.9%, 55.5%, and 30.6% after enrichment cultures with RV, BPW, and SB respectively (Table 2).

Improving isolation of Salmonella from human stool samples by using MSRV medium in clinical laboratories

has been described previously. Aspinal et al. (1992) reported a 21.3% increase in isolates recovered by direct plating using MSRV compared to cultivable numbers utilizing selenite F broth enrichment prior to sub-culturing on brilliant green agar, while a 29% increase was indicated using Rappaport-Vassiliadis broth in conjunction with brilliant green agar. Furthermore, Dusch and Altwegg (1995) reported direct plating sensitivities of 63.4% for MSRV in contrast to 34.1% and 32.8% for HE and Xylose-Lysine-Tergitol 4 (XLT4) respectively [17] . Ruiz et al. (1996) reported the number of Salmonella isolates recovered by direct plating on MSRV was also greater than those obtained on SS, Salmonella identification detection agar medium (SM-ID), Rambach agar, novobiocin-brilliant green-glycerol-lactose (NBGL) agar, and selenite-SS combination [6] . More recently, Ruiz et al. (1998) found a clear performance of MSRV over SS by direct plating with 24.7% greater sensitivity. They did not obtain significant differences when MSRV was compared to the selenite-SS combination, but the sensitivity of the selenite-MSRV combination was 8.2% greater than that of selenite-SS combination (p-value < 0.01) [18] .

Presently, we intended to compare the yields of Salmonella from MSRV to MAC, HE and XLD only by DP method because it is known that number of Salmonella contained in stool samples to prior enrichment is much lower than after enrichment and is suitable for comparison. The performance of MSRV medium provided a 51.0% increase in recoverability of non-typhoid Salmonella spp., when compared to MAC and XLD, while a 59.9% increase when compared to HE agar. Although RV broth yielded the highest number of Salmonella-positive samples, there are several disadvantages of the medium as it is not suitable for detection of non-motile strains of Salmonella, S. Typhi and S. Paratyphi A [16] [19] . The yield obtained from BPW was highest when MSRV was used in conjunction with MAC and HE. In order to isolate all Salmonella serotypes, BPW should be used for enrichment followed by using MSRV in combination with MAC, HE. An additional advantage of BPW is that it enhances the growth of other enteric bacteria pathogens to include Shigella, Aeromonas, Plesiomonas shigelloides, and the diarrheagenic Escherichia coli. For these reasons, our laboratory has instituted using MSRV medium in direct plating and in combination with our standard media after enrichment culture in BPW for the improvement of isolating Salmonella spp. from human stool samples.

We thank Dr. Brett E. Swierczewski and Dr. Trent J. Peacock for reviewing the manuscript.

ApichaiSrijan,BoonchaiWongstitwilairoong,LadapornBodhidatta,CarlMason, (2015) Efficiency of Plating Media and Enrichment Broths for Isolating Salmonella Species from Human Stool Samples: A Comparison Study. Open Journal of Medical Microbiology,05,231-236. doi: 10.4236/ojmm.2015.54028