2. Materials and Methods2.1. Source of SpecimensAll samples tested were obtained from the Multi-Center Study of Prevalence and Drug Resistant Pattern of Campylobacter and Other Enteric Pathogens in Thailand. Samples were approved by the Ethical Review Committee of Research in Human Subjects, Ministry of Public Health, Nonthaburi, Thailand and the Human Subjects Research Review Board, US Army Medical Research and Material Command, Fort Detrick, MD. After obtaining informed consent from a parent or guardian, a stool sample was collected from children aged 3 months to 5 years who presented to the hospital with diarrhea. All stools were collected in a clean wide mouth container by a nurse or a health care provider.
2.1.1. Study AFresh stool samples were collected from January 2003 to October 2006 from children who presented to the OPD ward of the Queen Sirikit National Institute of Child Health, Bangkok, Thailand. All samples were transported to the Armed Forces Research Institute of Medical Sciences (AFRIMS) laboratory, Bangkok, Thailand and processed for culture within 3 hours after collection. Transport medium was not used in this study.
2.1.2. Study BStool samples collected from the northeast region of Thailand during October 2004 to December 2006 originated from children who presented to the OPD of the Suppasitprasong Hospital. Stool samples collected within the southern region of Thailand from October 2004 to October 2006 originated from children who presented to the OPDs of the Trung Hospital and Patalung Hospital. In all cases, three polyester tipped swabs were used to harvest fecal material from each stool sample with subsequent inoculation in MCB transport medium. All inoculated MCB transport medium were refrigerated at 2˚C - 8˚C until transported in Styrofoam boxes with ice packs to the AFRIMS laboratory. Specimen processing occurred within 7 days of collection.
2.2. MediaSemi-solid, 0.16% (w/v), Modified Cary-Blair (MCB) transport medium [6] was used in this study. Brucella agar plates (Becton, Dickinson and Company, MD, USA) supplemented with 5% sheep blood (BAP), and mCCDA (Oxoid Ltd., Hampshire, England), supplemented with 32 mg/L cefoperazone and 10 mg/L Amphotericin B (Sigma-Aldrich.Inc, St. Louis, USA) were prepared according to the manufacturer’s instructions.
2.3. Specimen ProcessingUpon receipt at the laboratory, approximately 0.5 gram of each fresh stool sample or 3 fecal swabs were diluted 1:10 with an appropriate volume of sterile normal saline in sterile test tubes. Each fecal suspension was cultured as follows;
2.3.1. Direct PlatingFiltration technique as described by Steele and McDermott [7] was modified as follows; five to six drops of fecal suspension were separately applied on top of a 47-mm, 0.65 µm sterile cellulose acetate membrane filter (Millipore, Bedford, USA) centrally placed on the surface of a BAP, followed by membrane removal after 30 minutes of filtration under ambient conditions. Selective technique was performed by streaking a single drop of fecal suspension onto the mCCDA.
2.3.2. Enrichment and Plating after EnrichmentApproximately 0.5 mL of each fecal suspension was inoculated into 7 mL of Preston broth (Oxoid, UK), keeping caps loose during incubation. Every 18 - 24 hours a new set of BA filtration plates and mCCDA plates were inoculated.
2.3.3. IncubationInoculated BAP and mCCDA plates were incubated at 37˚C under microaerobic conditions (5% O2, 10% CO2, 85% N2), generated by a trigas-incubator (Nuaire, Plymouth, USA) for up to 72 hours and 18 - 24 hours for Preston enrichment broths.
2.4. Plate ReadingBAP and mCCDA plates were examined after 48 hours of incubation. Typical colonies resembling those of Campylobacter were picked to perform identification. Specimens were determined to be Campylobacter culture negative for both direct and enrichment plating after 72 hours of no growth.
2.5. Identification of CampylobacterInitial identification of Campylobacter was determined by colony morphology, Gram stain, and oxidase test. Confirmatory identification was determined through phenotypic tests [8], including catalase, hippurate hydrolysis, nitrate reduction, indoxyl acetate hydrolysis, H2S production in TSI, growth at 37˚C aerobically, ability to grow at 25˚C and 42˚C microaerobically, and susceptibility to Nalidixic Acid (30 µg) and cephalothin (30 µg) disks.
2.6. Data AnalysisIn each study, improvement of Campylobacter isolation by Preston broth were measured by comparing the numbers of Campylobacter-positive samples from enrichment culture to the number from direct plating. Non-parametric McNemar test was applied to each study’s results to assess efficacy of enrichment. A cricital level (α) of 0.05 was selected to denote statistical significance.
3. Results and DiscussionTable 1 shows the results of Campylobacter species isolated from direct and enrichment plating in studies A and B. Of the 900 fresh stool specimens collected from Study A, without using MCB transport medium, 158 were Campylobacter-positive, 145 isolates by direct plating, and 141 isolates through enrichment plating (P = 0.5839). From 1168 preserved stool samples in MCB from Study B, 184 were positive for Campylobacter spp. Of these, 139 were cultured by direct plating, while 168 isolates were cultured through enrichment plating (P = 0.0003). Numbers of species-specific Campylobacter isolates from the two studies are shown in Table 1. In both studies C. jejuni comprised the majority of isolates, while the remaining isolates comprised C. coli. There was no marked improvement, Study A, using enrichment broth for the recovery of Campylobacter species from fresh stool specimens (P = 0.5839). However, Study B methods demonstrated significant improvement in C. jejuni isolation from archived stool specimens utilizing enrichment medium (P = 0.0006), but not for other Campylobacter species (most likely due in part to low numbers of cells from other Campylobacter spp.).
A total of 1168 fecal swab samples were received in MCB transport medium after an average of 6 days in transit prior to being processed for Campylobacter isolation; Table 2. An overall isolation rate by direct and enrichment plating was 15.8% (184/1168). Utilizing Preston enrichment medium resulted in a 2.5% (14.4% versus 11.9%) higher isolation rate of Campylobacter spp. in contrast to direct inoculation. Of 477 stool specimens examined within 3 days, 56 Campylobacter-positive cultures were assayed; 6 isolates by direct plating only, 10 isolates through enrichment plating only, and 40 isolates through both direct and enrichment plating; Table 2. Only 1% (4/477) more Campylobacter-positive specimens were detected from preserved human stool in MCB by enrichment plating than by direct plating (P = 0.4545). This data suggests that enrichment broth did not significantly enhance recovery of any Campylobacter species, though these specimens had been held in MCB at 2˚C - 8˚C for 1 to 3 days. Of 691 fecal swabs held in MCB transport medium for 4 - 7 days, 128 Campylobacter
Table 1. Campylobacter isolates from stool specimens using direct and enrichment plating, Study A and Study B.
a.Specimens positive by direct plating but negative by enrichment plating. bSpecimens positive by enrichment plating but negative by direct plating. NT, not tested.
Table 2. Campylobacter isolates using and duration of delay in transport before processing, Study B.
aMedian = 6 days, bSpecimens positive by direct plating but negative by enrichment plating, cSpecimens positive by enrichment plating but negative by direct plating.
positive cultures were assayed; 10 isolates by direct plating only, 35 isolates through enrichment plating only, and 83 isolates through both direct and enrichment plating; Table 2. These results suggest an increase in Campylobacter positive cultures by enrichment plating in comparison to direct plating. Suggesting Preston enrichment broth did significantly, (P = 0.0003), improve recovery of Campylobacter species from fecal specimens with delayed processing times from 4 to 7 days.
Isolation of Campylobacter is hampered by its fastidious nature and may be overgrown by normal flora present in stool specimens. Although isolation techniques have been improved, isolation of this organism is still difficult due to delays in transit to reference laboratories. Appropriate methods for transportation, storage of clinical samples, and enrichment are critical for recoverability of pathogens in field epidemiological studies. Particularly important when immediate processing of fecal specimens is not always possible. Tanner and Bullin [9] compared storage of 7 aliquots of positive human fecal specimens kept at +4˚C and at room temperature, after 2 days only one of seven samples stored at room temperature was positive, whereas all seven samples stored at +4˚C were still positive. Blaser et al. [10] reported that C. fetus subsp. jejuni was recoverable from 5 positive human stools from 9 to 22 days when refrigerated at 4˚C, but for only 2 to 8 days when left at room temperature. Wang et al. [5] reported 90% survival of C. jejuni in human fecal specimens stored in MCB after 5 - 8 days, (with decreased agar concentration, 1.6 g/liter), at 4˚C without enrichment and 50% survival after 2 weeks. Enrichment media has been used for many years to improve isolation of Campylobacter spp. from human, animal, food and environmental samples. Martin and others [11] reported a significant increase of C. jejuni after enrichment in Campylobacter enrichment broth. SjÖgren and coworkers [12] used semisolid motility test medium as enrichment for Campylobacter spp. from in/out patients whose samples were processed within 24 hours. Differences in cultured isolates were only seen in outpatients and sample qualities were suggested to be the variable factors. Bolton and Robertson [13] observed that Preston enrichment broth was suitable for recovering Campylobacter species from human stool specimen and other biological sources (animal, avian and environmental). They also found that enrichment was not necessary for laboratory diagnosis of acute infections in human when great numbers of Campylobacter were present in stool, but was necessary when the number of C.jejuni was lower than 104 organisms/gram of stool.
In the present study, it is probable that most specimens contained viable Campylobacter organisms in sufficient numbers to be easily detected by one or both agar media in the primary plating step. Of 158 positive Campylobacter isolates, 13 were recovered only after enrichment (Table 1). This suggests that even though the diarrheal stool specimens were processed for culture within 3 hours of collection, some of them may have contained insufficient numbers of Campylobacter to be detected by direct plating. However, no significant difference in the total number of Campylobacter isolated from direct plating and after enrichment in Preston broth were detected (P = 0.5839) (Table 1). Additionally, 11 C. jejuni, 3 C. coli and 3 C. upsaliensis isolates were recovered when plated directly, but were negative after enrichment (Table 1). This suggests that these Campylobacter species might be susceptible to at least one antimicrobial in the enrichment broth, and therefore could not be recovered in secondary plating. Though all specimens were received and processed, on average, 6 days after collection, 11.9% (139/1,168) of these specimens yielded Campylobacter in the direct plating step (Table 1). When Preston enrichment broth was incorporated, the isolation rate increased from 11.9% to 14.4% and the overall isolation rate by combining direct plating and enrichment plating was 15.8% (Table 2).
Successful isolation is dependent upon the quality of collection, storage and processing of the clinical specimens. For decades, Cary-Blair (with 0.5% agar) cottontipped applicators have been routinely used for transport/ storage and collection of human stool samples collected at AFRIMS field-sites. These preserved stools were processed for culture both at the field-site laboratory and at AFRIMS’s laboratory. Commonly, it is observed that fastidious enteric bacterial pathogens (Shigella and Campylobacter species) were routinely isolated at field-site laboratories. However, after extended storage of 1 week there was a decrease in colony forming units (CFU). One possible factor are the fatty acids present in cotton which, are known growth inhibitors of Shigella [14] and may also inhibit Campylobacter species. Implementation of MCB and polyester-tipped swabs resulted in comparable isolation rates of these pathogens in our laboratory to those of field-site laboratories. However, culture dependent variables, still need to be ascertained.