To investigate the current prevalence of these infections, we conducted a pilot study to identify STH and other potentially endemic parasitic infections in convenience samples of specimens collected from patients in Mississippi. We deidentified fresh fecal samples submitted for diagnostic testing from patients at the University of Mississippi Medical Center (UMMC; Jackson, Mississippi, USA) during March 30, 2017–February 22, 2018, and serum samples submitted during October 28, 2017–March 29, 2018. This study was approved by the UMMC Institutional Review Board; the Centers for Disease Control and Prevention (CDC) was determined to be nonengaged and therefore did not undertake a separate institutional review board review.

We froze two 250-mg aliquots of feces for later DNA extraction. Where sample volume allowed, we performed microscopic examination using the saturated salt (specific gravity 1.2) passive flotation method as previously described (4). We extracted DNA by using the SurePrep Soil DNA isolation kit (ThermoFisher, https://www.thermofisher.com) after conducting initial bead beating for 3 minutes using zirconium beads. We stored DNA extracts at −80°C and sent them to the CDC for real-time PCR analysis. At CDC, each sample was initially tested for inhibition and poor DNA extraction by a real-time PCR assay targeting the human cytochrome B gene (5). Samples positive by this inhibition and extraction control were then tested by multiparallel real-time PCR for STH (6). A cycle threshold (C_t) <35 was considered to represent a positive result. Any positive PCR results were confirmed by duplicate testing.

We froze the deidentified serum samples at –80°C and sent them to CDC, where they were tested for antibodies to Toxocara spp., S. stercoralis, Cryptosporidium spp., and G. duodenalis using MAGPIX multiplex serology (ThermoFisher) (Appendix) to detect evidence of prior exposure. For statistical calculations, we used Excel (Microsoft, https://www.microsoft.com) and R version 3.3.1 (https://www.r-project.org).

A total of 650 fecal samples were obtained from UMMC patients. The median age of patients providing fecal samples for this analysis was 56 years (range 2–95 years). We obtained samples sufficient to perform saturated salt centrifugal flotation on 507 samples (80%). We found no samples to contain helminth eggs or larvae. Sufficient sample for DNA extraction was available for 631 (99.5%) samples. Of these fecal DNA extracts, a negative inhibition and extraction control excluded 37 samples. We tested 224 DNA extracts for Ancylostoma spp., N. americanus, S. stercoralis, A. lumbricoides, and T. trichiura by real-time PCR. (Table 1)

Because prior work in Alabama (3) detected only N. americanus and S. stercoralis infections, we screened an additional 370 DNA extracts for these helminths only. Of these 370 samples, 2 DNA extracts yielded positive amplicons for S. stercoralis (C_t 29.57 and 30.48). The first of these samples (C_t 29.57) yielded no amplification curve on repeat testing and was interpreted as representing an initial false-positive result. The second sample (C_t 30.48) was positive upon confirmatory retesting (C_t 28.52 and 30.49). (Table 1)

A total of 1,960 postdiagnostic serum samples from Mississippi residents were available for multiplex serologic testing. The median age of patients providing serum samples for this analysis was 38 years (range 0–94 years). Of the 1,960 samples, 646 (33.0%) reacted with the Cp17 antigen of C. parvum (range 87–48,448 mean fluorescence intensity [MFI]), and 1,076 (54.9%) reacted with Cp23 (range 377–56,727 MFI). Of those samples, 538 (27.4%) reacted with both C. parvum antigens (Figure 1, panel A), suggesting prior Cryptosporidium species infection. A total of 111 samples (5.7%) reacted with the G. duodenalis VSP3 antigen (range 84–48,547 MFI) (Figure 1, panel B). A total of 38 (1.9%) samples contained antibodies to the Cp17, Cp23 and VSP3 antigens (Figure 1, panel C), demonstrating prior exposure to both Cryptosporidium and G. duodenalis infections. A total of 172 (8.8%) samples contained antibodies to Toxocara spp. Tc-CTL-1 antigen (range 23.2–33,814 MFI) (Table 2; Figure 2, panel A). When Toxocara-seropositive participants <6 years of age were excluded, 167/1,814 (9.2%) of UMMC patient samples were seropositive. A total of 9 (0.4%) samples contained antibodies reacting with the recombinant S. stercoralis NIE-1 antigen (range 16.2–11248 MFI) in MAGPIX serologic testing, of which 4 (0.2%) were positive in the confirmatory S. stercoralis CrAg-ELISA (range 9.94–57.7 IU/mL) (Figure 2, panel B).

The results of this limited pilot study suggest a low prevalence of STH infections in Mississippi but that rare infections with S. stercoralis might be found in Mississippi residents. The single case confirmed by real-time PCR tests likely represents active infection. Because >80% of patients with strongyloidiasis serorevert within 18 months after successful treatment (7), the 4 confirmed antibody-positive serum samples also likely represent active cases of strongyloidiasis. No linked immigration or travel history data on patients providing these samples were available, so whether these infections were acquired within the United States is unknown. Combined with the recent finding of strongyloidiasis in a rural community from Alabama (3), these data should encourage more focused sampling of areas with poor sanitation and hygiene, high levels of poverty, and poor access to healthcare for potential residual foci of endemic STH and strongyloidiasis transmission in Mississippi and the wider southeastern United States.

The total Toxocara spp. seroprevalence in all participants in this study was 8.8%, which is higher than the average prevalence reported by the most recent National Health and Nutrition Examination Survey study (8). Although these results are not directly comparable because of different sampling methods, the potentially high Toxocara spp. seroprevalence in Mississippi warrants further investigation.

The seroprevalence results of this study suggest that prior exposure to Cryptosporidium spp. is common in Mississippi. Only 5.7% of the postdiagnostic serum samples were found to have serologic evidence of prior exposure to G. duodenalis infection. A small number of samples (1.9%) contained antibodies reacting with the 3 antigens Cp17, Cp23, and VSP3, indicating prior exposure to Cryptosporidium spp. and G. duodenalis infection. Further investigation of the epidemiology of waterborne protozoan infection in Mississippi, including determination of the actual prevalence and distribution using systematic sampling and determination of the species and subtypes infecting persons, is warranted.

The absence of any positive findings by microscopic examination or PCR for the STH suggests that such infections are uncommon in the general Mississippi population. We found high seroprevalence of antibodies to Toxocara spp. in Mississippi. Although this finding could indicate increased exposure to this infectious agent compared with the national average, our data do not enable determination of the sources of increased infection or overall annual incidence of disease. Further studies on the epidemiology and prevalence of parasitic diseases in the state of Mississippi are indicated.

In conclusion, this convenience sampling study did not find evidence of high STH prevalence in Mississippi. However, we did identify several likely current cases of strongyloidiasis and relatively high rates of Toxocara exposure. We recommend further investigation with larger sample sizes to more clearly define the true extent of STH infection in this region.

Additional information about parasitic disease surveillance, Mississippi, USA.