We examined 5 recent infections of humans with Diphyllobothrium spp. (Table, cases 12, 16–19) that occurred in China. Each case had been originally reported as a D. latum infection on the basis of morphologic identification only. Case 12 was reported in Harbin City, Heilongjiang Province, in 2009 (13). The 4 cases reported in Shanghai were diagnosed at the National Institute for Parasitic Diseases, Shanghai, on the basis of morphologic features of passed strobila. Case-patient 16 lived in Japan, but it was suggested that he acquired the tapeworm in Shanghai where he had frequently eaten raw salmon. Case-patient 17 was a 10-year-old girl from Japan. Whether she became infected in Shanghai or Japan was unclear because of lack of information. Case-patients 18 and 19 acquired the infection in Shanghai because they had never been abroad.

Figure 1

Figure 1. Diphyllobothriid samples examined in the present study, China, 2008–2012A) Proglottids stained with acetic acid–carmine from case-patient 12B–E) Sagittal sections of proglottids stained with hematoxylin-eosin from case-patients 16–19cs, cirrus sac; ut, uterus;...

Because all patients in Shanghai had eaten raw salmon, we decided to re-examine how the causative Diphyllobothrium spp.were identified. D. latum infection is associated with consumption of freshwater fish such as perch (Perca spp.), not Pacific salmon (Oncorhynchus keta, O. masou) and Atlantic salmon (Salmo salar) in the Northern Hemisphere (1–5). To expand diagnostic parameters and clarify the point of misidentification, we re-identified Diphyllobothrium spp. by examining the tapeworms’ morphologic features and using a molecular marker. In a sample from case-patient 12, only proglottids stained with acetic acid–carmine were available for testing by both methods (Figure 1, panel A). Proglottids obtained from 4 case-patients in Shanghai were preserved in either 10% formalin (case-patient 16) or 70% ethanol (case-patients 17–19) after collection (Table). Parts of the proglottids were embedded in paraffin, and sagittal sections were prepared for morphologic observation.

For molecular identification of the Diphyllobothrium spp., genomic DNA samples were extracted from specimens by using a DNeasy Blood & Tissue Kit (QIAGEN, Hilden, Germany). In specimens from case-patients 17–19, the mitochondrial cytochrome c oxidase subunit 1 gene (cox1, 1,566 bp) was amplified by PCR by using Ex Taq DNA polymerase (Takara Bio, Shiga, Japan) (7). In formalin-fixed samples of proglottids from case-patients 12 and 16, DNA degradation caused by the fixative meant that only shorter cox1 fragments (249 bp, corresponding to sites 880–1128 of cox1) could be amplified successfully by PCR by using KOD FX DNA polymerase (Toyobo, Osaka, Japan). DNA sequencing of amplicons was performed with a 3100-Advant Genetic Analyzer or 3730 xl DNA Analyzer (Life Technologies, Foster City, CA, USA). Phylogenetic analysis was performed by the maximum likelihood method (MEGA 5.05, http://megasoftware.net/mega.php) and Bayesian inference (MrBayes version 3.1.2, http://mrbayes.sourceforge.net/). Clades were assessed by bootstrap resampling (1,000 replicates) and a posterior probability (10⁶ generations) for the maximum likelihood and Bayesian inference trees, respectively. Diphyllobothrium spp. isolated from case-patients 12 and 16 were identified on the basis of sequence identity (%) by performing a BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) analysis of a DNA Data Bank of Japan (http://www.ddbj.nig.ac.jp).

Accurately identifying the Diphyllobothrium spp. isolated from case-patient 12 on the basis of morphologic features alone was difficult (Figure 1, panel A). In Figure 1, panels B–E show the sagittal sections of the proglottids from case-patients 16–19. The angle formed by the cirrus sac and the anterior–posterior axis of the proglottids was used as a criterion for differentiating D. latum from D. nihonkaiense (1), even though this criterion is not considered definitive: the angle is usually horizontal in D. latum, but oblique in D. nihonkaiense. Nonetheless, in this study, on the basis of morphologic criteria, tapeworms from case-patients 16, 17, and 19 were identified as D. latum (Figure 1, panels B, C, and E) and the tapeworm found in case-patient 18 was identified as D. nihonkaiense (Figure 1, panel D).

Figure 2

Figure 2. Phylogenetic tree constructed by using the maximum likelihood algorithm (Kimura’s 2-parameter model) on the basis of the complete cox1 sequences of isolates from Diphyllobothrium species found in persons in China and...

Phylogenetic trees based on the complete cox1 nucleotide sequences showed the same topologies in maximum likelihood and Bayesian inference analyses, implying that the 3 isolates from persons in China (case-patients 17–19; GenBank accession numbers AB684621–AB648623) are D. nihonkaiense (Figure 2). The 2 isolates (AB684625 and AB684624) from case-patients 12 and 16, respectively, were excluded from the analysis because they produced smaller PCR products, but they were identified as D. nihonkaiense on the basis of their 99%–100% sequence identity to D. nihonkaiense.

The 5 Diphyllobothrium spp. tapeworms examined in this study were previously identified as D. latum on the basis of morphologic characteristics, as were 3 of the 5 when we re-examined their morphologic characteristics. However, the 5 etiologic agents were confirmed as D. nihonkaiense by molecular analysis. This discrepancy in the identity of these agents may be attributed to the morphologic similarities between the species and the century-long confusion between the parasite D. latum and the parasite that caused human diphyllobothriosis associated with the consumption of Pacific salmon in Japan (1–3). Diphyllobothriosis caused by D. nihonkaiense has also been reported in South Korea (8) and in the Far Eastern Federal District of Russia (9) and is considered to be autochthonous and linked to the consumption of wild Pacific salmon in these regions. Therefore, some cases of diphyllobothriosis reported in mainland China were probably caused by infections with D. nihonkaiense; case-patient 12 (13) in this study is considered to have had such a case. However, a recent report stating that the causative species of 2 diphyllobothriosis cases in northeastern China was D. latum which suggests that D. latum is also indigenous to mainland China (14).

We confirmed human diphyllobothriosis caused by D. nihonkaiense in mainland China by using a mitochondrial DNA marker. Reassessment of a case in Harbin revealed that some, if not all, of the autochthonous diphyllobothriosis cases were likely initially misdiagnosed as D. latum infection because of morphologic similarities between D. nihonkaiense and D. latum tapeworms. Consequently, molecular analysis is indispensable not only for avoiding diagnostic confusion among Diphyllobothrium spp., but also for facilitating the acquisition of reliable epidemiologic and epizootic information and improving clinical relevance and preventive controls for diphyllobothriosis.

Information on diphyllobothriosis and warnings of the potential risks associated with infection by its local species should be disseminated to food handlers, restaurant owners, physicians, and consumers. Because we cannot determine with certainty whether previous diphyllobothriosis cases in mainland China were caused by D. latum or D. nihonkaiense, identification of Diphyllobothrium spp. should be performed with care. In addition, studies on the distribution and sources of infection of D. latum and D. nihonkaiense on mainland China should be undertaken.