Among the analyzed samples, one sample showed a positive result in the RT-PCR targeting hsp65 and three samples showed a positive result in the RT-PCR targeting Dig15. However, all four of these samples were confirmed to be T. whipplei-negative by sequencing (Table 2). By integrating the findings of RT-PCR and sequencing, we established that T. whipplei was not detected in any of the 252 included samples. Thus, the prevalence of T. whipplei in duodenal tissues of Koreans biopsied from visible lesions on EGD was estimated to be less than 0.4%. This is a lower estimate than the previously reported prevalence of T. whipplei in the stools of Korean patients with diarrhea (1.4%) [3]. Considering the high sensitivity and low detection limits of RT-PCRs as previously identified, our present findings are highly convincing [3].

It is possible that T. whipplei DNA was present in the samples at levels below the detection limit of the RT-PCRs used in this study. Alternatively, various factors may have reduced the sensitivity of RT-PCR. Since the previous and present studies differed in that the samples were stools and duodenal biopsy tissues, respectively, we used a kit appropriate for DNA extraction from each sample. Nevertheless, the use of different kits to extract DNA may influence the sensitivity of the test, especially for low bacterial biomass samples [17]. In addition, while the stools in the previous study were fresh materials, the duodenal biopsy tissues was the paraffin-embedded materials. The PCR assays on paraffin-embedded materials require optimization because the detection rates of PCR may be reduced [18, 19]. DNA extracted from T. whipplei-infected tissue would be appropriate as a positive control for PCR. However, this study has the limitation that recombinant plasmids were used as a positive control because it was difficult to obtain a standard strain of T. whipplei or T. whipplei-infected tissue. Instead, the quality of DNA extracted from the tissue was verified by parallel RT-PCR targeting the human β-actin gene, and the target gene was detected in all samples.

Four samples showed false-positive results by RT-PCRs. One sample with a false-positive result for hsp65 showed mixed peaks on sequencing, for which no significant similarity was found in the BLAST search. Among the three samples with false-positive results for Dig15, one was confirmed as a noise signal and one as mixed peaks of Palisada intermedia with 86.0% identity and Laurencia catarinensis with 85.6% identity. The last one was confirmed as Homo sapiens with 97.3% identity. These sequencing results obtained from non-specific PCR products offer insights for refining primer designs to further increase the specificity of the PCR assay. In addition, the discrepancy between RT-PCR results highlights the credibility of using multiple PCR targets rather than a single target.

Our previous and present findings collectively suggest that T. whipplei exists in the intestine of Koreans, but may be merely a passenger. In carriers, T. whipplei has been detected at a high rate in stool samples, but only a relatively low rate in intestinal biopsy samples [20]. Due to differences in sample types, the prevalence in this study using duodenal biopsy tissues could be different than in our previous study using stool samples. In addition, the obtained results could be attributed to sampling bias resulting from the uneven distribution of the bacterium within the intestine [1, 3]. To address potential sampling bias, it is recommended to obtain multiple samples from different sites of the intestine [1]. The samples in this study were limited in that biopsy tissue could only be obtained from the duodenum and not from other parts of the small intestine. Although the samples were not obtained from multiple sites, they were obtained from sites that had a relatively high probability of bacterial presence, specifically from abnormal lesions that were determined to require biopsy based on endoscopic findings. Our findings indicate that T. whipplei is by no means a common cause of duodenal infection/inflammation in Korea. However, there is still the possibility of infection in parts of the small intestine other than the duodenum.

In this study, the majority of samples were obtained from adults. The intestinal colonization of T. whipplei has been shown to have a higher prevalence in children and adolescents than in adults, indicating an age-dependent presence [3, 8, 21]. It has been reported that the prevalence of antibodies against T. whipplei was higher in adults than in children, indicating the possibility of acquired immunity to T. whipplei infection [22]. Repeated reinfection and/or other host factors (e.g. immunity) appear to be required to maintain T. whipplei colonization in adults [21]. The host’s immune system is an important predisposing factor for T. whipplei infections, and the bacterial colonization alone is not sufficient to cause WD [1]. Because this study included subjects who visited the CAUMC for a routine medical check-up and was limited in obtaining information on their medical records, host factors such as underlying health conditions and drug use could not be considered. Further studies are needed to detect T. whipplei in a larger number of intestinal biopsy samples from Korean patients across a wider age range, taking host factors into account.

In conclusion, this is the first study to attempt to detect T. whipplei in duodenal biopsy tissues of Koreans and estimate its prevalence. T. whipplei, which had been detected at a prevalence of 1.4% in the stools of Koreans, was not detected in the duodenal biopsy tissue of the Koreans analyzed in this study. Our findings infer that while T. whipplei carriers exist in Korea, the incidence of duodenal infection/inflammation caused by T. whipplei is extremely rare, which makes it reasonable that WD has not been reported in Korea.