Abstract
Among 345 persons who underwent indirect hemagglutination (IHA) serological assays and assays of specific amebic antigens in their stool samples, 24 of 36 (66.7%) who were seropositive for Entamoeba histolytica had intestinal amebiasis as determined by antigen assays compared with 2 of 309 (0.2%) who were seronegative (odds ratio, 307; 95% confidence interval, 64.9 to 1,451). The estimated cost to detect a case of intestinal amebiasis by serology followed by antigen assays ($52) could be reduced by 74.3% and 69.9%, respectively, compared with the costs of the concurrent use of both assays ($202) and the antigen assays alone ($173). Our finding suggests that IHA assays followed by specific-amebic-antigen assays can be cost-effective in the diagnosis of intestinal amebiasis among persons with or without human immunodeficiency virus infection who are at risk for E. histolytica infection.
In East Asia, Entamoeba histolytica infection (amebiasis) has become an emerging parasitic infection among men who have sex with men (MSM), especially those who are human immunodeficiency virus (HIV) infected (4-6, 9-14, 16). In Japan, 80% of the annually reported cases of amebiasis occurred in MSM (8), and the seroprevalence of amebiasis in MSM was estimated to be 13.4% to 20.4%, compared with 1.0% in heterosexuals and 0.8% in prostitutes (12-14). In Taiwan, the prevalence and incidence of amebiasis are also significantly higher among HIV-infected MSM than among healthy controls and heterosexuals (5, 6, 16). Therefore, it is important for health care providers to diagnose amebiasis and provide appropriate treatment and counseling to MSM to prevent the development of invasive amebiasis and the transmission of E. histolytica.
The tests used to diagnose amebiasis include microscopy of clinical samples, serology, specific-amebic-antigen assays, and molecular methods of testing stool and liver aspirate samples (3, 15). Of those tests, specific-amebic-antigen assays of stool samples have the highest sensitivity and specificity in the diagnosis of intestinal amebiasis (2, 15); however, their cost is significantly higher than microscopy and serology (2, 15). How to maximize the cost-effectiveness of serological tests and amebic-antigen assays for the diagnosis of intestinal amebiasis has not been investigated before. In this study, we aimed to compare the costs and diagnostic yields of serological tests combined with amebic-antigen assays among persons at risk for HIV infection and amebiasis.
(Preliminary analyses of these data were presented at the 18th European Congress of Clinical Microbiology and Infectious Diseases held in Barcelona, Spain, on 19 to 22 April 2008 [1].)
Between January 2006 and February 2008, 376 HIV-infected persons who sought HIV care and 2,511 non-HIV-infected persons who sought anonymous HIV testing at the National Taiwan University Hospital were enrolled in the study. At least one stool sample and one blood sample from each person were obtained for specific-amebic-antigen assays and indirect hemagglutination (IHA) assays, respectively (Fig. 1). IHA assays to detect anti-E. histolytica antibodies were performed by following the instructions of the manufacturer (Cellognostics; Boehhringer Diagnostics GmbH, Marburg, Germany). Seropositivity was defined as a titer of 128 or greater. The purchase cost for each test was estimated to be $1.7; three additional tests ($5.1) are needed to confirm an IHA titer of 128 or greater. To detect E. histolytica in stool samples, specific-antigen assays were performed using commercial kits (Entamoeba test; TechLab, Branchburg, NJ). The purchase cost for each antigen test was estimated to be $13. To confirm the presence of E. histolytica, PCRs were performed using primers specific for E. histolytica as described previously (6). The primer sets for a multiplex nested PCR were based upon the variable regions between 16S-like ribosomal DNAs of E. histolytica (GenBank accession no. X56991) and E. dispar (GenBank accession no. Z49256).
FIG. 1.
Results for the study population of men with amebiasis, whose sera were subjected to IHA assays and whose stool samples were subjected to specific-amebic-antigen assays.
The characteristics of the study populations are shown in Table 1 and Fig. 1. The seroprevalence of amebiasis was 3.8% (47/1,237) (95% confidence interval [CI], 2.9 to 5.0) among MSM compared with 0.36% (6/1,650) (95% CI, 0.15 to 0.75) among individuals not in the MSM group (non-MSM individuals) and those with unknown risk factors (P < 0.001). HIV-infected MSM had the highest seroprevalence of E. histolytica infection (10.1%; 28/277) compared with non-HIV-infected MSM (2.1%; 19/960) and non-HIV-infected non-MSM individuals (0.2%; 3/1,551).
TABLE 1.
Characteristics of the subjects who submitted stool and blood samples for specific-amebic-antigen assays and IHA assays
| Variable | Value for persons submittinga:
|
||
|---|---|---|---|
| Blood specimens for IHA assays | Stool specimens for antigen assays | Both specimens for antigen and IHA assays | |
| Total no. of subjects | 2,887 | 426 | 345 |
| Median age (range) (yr) | 28 (15-83) | 37 (15-93) | 36 (15-83) |
| Males | 2,453 (85) | 405 (95.1) | 334 (96.8) |
| MSM | 1,237 (42.8) | 291 (68.3) | 249 (72.2) |
| Non-MSM individuals | 1,632 (56.5) | 98 (23.0) | 96 (27.8) |
| Subjects of unknown sexual behavior | 18 (0.6) | 37 (8.7) | 0 |
| HIV infected | 376 (13.0) | 358 (84.0) | 299 (86.7) |
| MSM | 277 (73.7) | 256 (71.5) | 214 (71.6) |
| Non-MSM individuals or unknown | 99 (26.3) | 102 (28.5) | 85 (28.4) |
| Non-HIV infected | 2,511 (87.0) | 68 (16.0) | 46 (13.3) |
| MSM | 960 (38.2) | 35 (51.5) | 35 (76.1) |
| Non-MSM individuals or unknown | 1,551 (61.8) | 33 (48.5) | 11 (23.9) |
| Antigen positive | NA | 30/426 (7.0) | 26/345 (7.5) |
| MSM | NA | 27/291 (9.3) | 25/249 (10.0) |
| Non-MSM individuals or unknown | NA | 3/135 (2.2) | 1/96 (1.0) |
| IHA titer of ≥128 | 53/2,887 (1.8) | NA | 36/345 (10.4) |
| MSM | 47/1,237 (3.8) | NA | 34/249 (13.7) |
| HIV infected | 28/277 (10.1) | NA | 24/214 (11.2) |
| Non-HIV infected | 19/960 (2.0) | NA | 10/35 (28.6) |
| Non-MSM individuals or unknown | 6/1,650 (0.4) | NA | 0/96 (0) |
All values are numbers (percentages) of patients except where indicated otherwise. NA, not applicable.
A total of 426 persons submitted stool samples for amebic-antigen assays. Their clinical characteristics are shown in Table 1. The prevalence of intestinal amebiasis was 9.3% (27/291) (95% CI, 6.4 to 13.2) among MSM compared with 2.2% (3/135) (95% CI, 0.5 to 6.6) among non-MSM individuals and those with unknown risk factors (P = 0.004). The prevalence of intestinal amebiasis was 25.7% (9/35) among non-HIV-infected MSM, 7.0% (18/256) among HIV-infected MSM, and 0% (0/98) among non-MSM individuals or those with unknown risk factors. The higher prevalence of intestinal amebiasis among non-HIV-infected MSM than among HIV-infected MSM might be related to the fact that those non-HIV-infected MSM were more willing to submit stool samples for amebic-antigen assays once they learned that they were seropositive for E. histolytica.
During the same study period, specimens from a total of 345 persons were subjected to both IHA and amebic-antigen assays (Table 1 and Fig. 1); 36 of them (10.4%; 95% CI, 7.5 to 14.0) had high IHA titers. Seropositivity for E. histolytica was statistically significantly associated with the presence of E. histolytica in the stool samples. Among the 36 persons with high IHA titers, 24 (66.7%) also tested positive for E. histolytica by antigen assays compared with 2 of 309 (0.6%) who had low or negative IHA titers (odds ratio, 307; 95% CI, 64.9 to 1,451) (Fig. 1). Of the 26 persons who tested positive for E. histolytica by antigen assays (including 2 persons who were seronegative for E. histolytica), 21 submitted a second stool sample for PCR. Of the 21 antigen-positive stool samples, the presence of E. histolytica was confirmed by PCR in 17 (81.0%), and all 17 samples were from persons seropositive for E. histolytica.
Based on the finding described above, we estimated that the cost to detect a case of intestinal amebiasis by IHA assays [($1.70 × 345 tests) + ($1.70 × 3 × 36 tests) = $770] followed by amebic-antigen assays of the stool samples for those who were seropositive for E. histolytica ($13 × 36 tests = $468) was $52 [($770 + $468)/24]. By concurrent use of serology and amebic-antigen assays and by amebic-antigen assays alone, the cost was estimated to be $202 [($770 + 13 × 345)/26] and $173 [(13 × 345)/26], respectively. Therefore, the cost to detect a case of intestinal amebiasis by serological tests followed by specific-amebic-antigen assays could be reduced by 74.3% and 69.9% compared with the costs of concurrent use of serology and amebic-antigen assays and the use of amebic antigen assays alone, respectively.
In this study, we found that MSM were at a higher risk for E. histolytica infection either by serology or amebic-antigen assays than persons who were not MSM in Taiwan, probably because of infrequent use of condoms during oral-anogenital sexual contact among MSM (7), which may increase the risk for E. histolytica transmission. Therefore, screening and counseling for E. histolytica infection among MSM should be provided to prevent infection and the development of invasive diseases, because those carriers of E. histolytica may potentially serve as a reservoir for the transmission of E. histolytica to other MSM via oral-anal sex.
Screening using stool samples can be cumbersome compared with screening using blood samples because stool samples may not be as readily available as blood samples in the clinical-care setting. Our finding that seropositivity for E. histolytica is highly associated with intestinal E. histolytica infection among subjects consisting of HIV-infected MSM, non-HIV-infected MSM, and non-HIV-infected non-MSM individuals suggests that the serological test may be used as a complementary tool and may provide a more cost-effective way to understand the epidemiology of intestinal E. histolytica infection among high-risk populations (6).
In this study of a population with a high seroprevalence of E. histolytica infection (10.1%), we estimated that to identify one case of intestinal amebiasis, the cost can be reduced by 69.9% to 74.3% if the algorithm is to test with serological assays first and then with amebic-antigen assays for those who are seropositive instead of concurrently using serology and amebic-antigen assays or using antigen assays only. Based on two assumptions, namely, that among those persons who are seropositive for E. histolytica, two-thirds will test positive for E. histolytica by antigen assays and that the prevalence of intestinal E. histolytica infection as determined by antigen assays in persons who are seronegative is approximately 0.6%, as demonstrated in this study, we estimate that the cost can be reduced by approximately 30% by using sequential assays rather than a combination of serological and antigen assays or antigen assays alone in a population with a low seroprevalence for E. histolytica infection (0.2% in non-HIV-infected MSM) and by 80% in a population with an intermediate seroprevalence (2.1% in non-HIV-infected MSM) (data not shown).
There are several limitations of our study, and interpretations of the results should be cautious. First, the prevalence of E. histolytica infection is higher among our subjects, who are mainly HIV-infected MSM living in an area that used to have a higher prevalence of amebiasis, than among the general population (16). The results of the cost-effectiveness analysis may not be generalizable to a population consisting mainly of non-HIV-infected persons or to other areas showing a different epidemiology of E. histolytica and HIV infection. Second, our subjects had asymptomatic intestinal colonization with E. histolytica, and the results may not be generalizable to subjects with amebic colitis or liver abscess. Third, we did not include microscopy as a diagnostic tool for intestinal amebiasis. However, microscopy still has its role in the detection of other intestinal infections that are transmitted via the fecal-oral route (2). Therefore, our finding can be considered generalizable only in the context of intestinal amebiasis among MSM. Fourth, in the estimation of cost, we did not take into consideration costs related to equipment, personnel, other administrative issues, or PCR. Last, the shedding of E. histolytica may be intermittent, which may reduce the sensitivity of antigen assays if only one stool sample is tested.
In conclusion, our finding suggests that IHA assays followed by specific-amebic-antigen assays can be cost-effective in the diagnosis of intestinal amebiasis among persons at risk for HIV infection in an area of a high prevalence of E. histolytica infection among MSM.
Acknowledgments
We declare that no competing interests exist.
Footnotes
Published ahead of print on 2 July 2008.
REFERENCES
- 1.Chang, S.-Y., H.-Y. Sun, D.-D. Ji, Y.-C. Lo, C.-H. Wu, P.-Y. Wu, W.-C. Liu, C.-C. Hung, and S.-C. Chang. 2008. Abstr. 18th Eur. Congr. Clin. Microbiol. Infect. Dis., abstr. 742, Barcelona, Spain, 19 to 22 April 2008.
- 2.Fotedar, R., D. Stark, N. Beebe, D. Marriott, J. Ellis, and J. Harkness. 2007. Laboratory diagnostic techniques for Entamoeba species. Clin. Microbiol. Rev. 20511-532. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Haque, R., C. D. Huston, M. Hughes, E. Houpt, and W. A. Petri, Jr. 2003. Amebiasis. N. Engl. J. Med. 3481565-1573. [DOI] [PubMed] [Google Scholar]
- 4.Hung, C. C., P. J. Chen, S. M. Hsieh, J. M. Wong, C. T. Fang, S. C. Chang, and M. Y. Chen. 1999. Invasive amebiasis: an emerging parasitic disease in patients with HIV infection in an endemic area of amebic infection. AIDS 132421-2428. [DOI] [PubMed] [Google Scholar]
- 5.Hung, C. C., H. I. Deng, W. H. Hsiao, S. M. Hsieh, C. F. Hsiao, M. Y. Chen, S. C. Chang, and K. E. Su. 2005. Invasive amoebiasis is an emerging parasitic infection in patients with HIV infection. Arch. Intern. Med. 165409-415. [DOI] [PubMed] [Google Scholar]
- 6.Hung, C. C., D. D. Ji, H. Y. Sun, Y. T. Lee, S. Y. Hsu, S. Y. Chang, C. H. Wu, Y. H. Chan, C. F. Hsiao, W. C. Liu, and R. Colebunders. 2008. Increased risk for Entamoeba histolytica infection and invasive amebiasis in HIV seropositive men who have sex with men in Taiwan. PLoS Negl. Trop. Dis. 2E175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hung, C. C., N. Y. Ko, W. C. Ko, H. C. Lee, D. D. Ji, W. C. Liu, and S. C. Chang. 8 July 2008. Amebiasis among patrons visiting gay saunas in Taiwan. HIV Med. doi: 10.1111/j.1468-1293.2008.00609.x. [DOI] [PubMed]
- 8.Nozaki, T., S. Kobayashi, T. Takeuchi, and A. Haghighi. 2006. Diversity of clinical isolates of Entamoeba histolytica in Japan. Arch. Med. Res. 37277-279. [DOI] [PubMed] [Google Scholar]
- 9.Ohnishi, K., and M. Murata. 1997. Present characteristics of symptomatic amebiasis due to Entamoeba histolytica in the east-southeast area of Tokyo. Epidemiol. Infect. 119363-367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ohnishi, K., Y. Kato, A. Imamura, M. Fukayama, T. Tsunoda, Y. Sakaue, M. Sakamoto, and H. Sagara. 2004. Present characteristics of symptomatic Entamoeba histolytica infection in the big cities of Japan. Epidemiol. Infect. 13257-60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Park, W. B., P. G. Choe, J. H. Jo, S. H. Kim, J. H. Bang, H. B. Kim, N. J. Kim, M. D. Oh, and K. W. Choe. 2007. Amebic liver abscess in HIV-infected patients, Republic of Korea. Emerg. Infect. Dis. 13516. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Takeuchi, T., Y. Miyahira, S. Kobyash, T. Nozaki, S. R. Motta, and J. Matsuda. 1990. High seropositivity for Entamoeba histolytica infection in Japanese homosexual men: further evidence for occurrence of pathogenic strains. Trans. R. Soc. Trop. Med. Hyg. 84250-251. [DOI] [PubMed] [Google Scholar]
- 13.Takeuchi, T., S. Kobayashi, K. Asami, and N. Yamaguchi. 1987. Correlation of positive syphilis serology with invasive amebiasis in Japan. Am. J. Trop. Med. Hyg. 36321-324. [DOI] [PubMed] [Google Scholar]
- 14.Takeuchi, T., E. Okuzawa, T. Nozaki, S. Kobayashi, M. Mizokami, M. Minoshima, M. Yamamoto, and S. Isomura. 1989. High seropositivity of Japanese homosexual men for amebic infection. J. Infect. Dis. 159808. [DOI] [PubMed] [Google Scholar]
- 15.Tanyuksel, M., and W. A. Petri, Jr. 2003. Laboratory diagnosis of amebiasis. Clin. Microbiol. Rev. 16713-729. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Tsai, J. J., H. Y. Sun, L. Y. Ke, K. S. Tsai, S. Y. Chang, S. M. Hsieh, C. F. Hsiao, H. J. Yen, C. C. Hung, and S. C. Chang. 2006. Higher seroprevalence of Entamoeba histolytica infection is associated with human immunodeficiency virus type 1 infection in Taiwan. Am. J. Trop. Med. Hyg. 741016-1019. [PubMed] [Google Scholar]