Association between the infections of Trichomonas vaginalis and uterine cervical human papillomavirus: a meta-analysis

Abstract

Trichomonas vaginalis (TV) may have an impact on other reproductive tract infections. Studies on the connection between the infection of TV and human papillomavirus (HPV) have been inconsistent. We performed a systematic review of the relevant articles through keywords that satisfy the criteria and filtered the articles according to the inclusion and exclusion criteria. A total of 16 eligible studies were screened for the meta-analysis, involving a total of 150,605 women. RevMan 5.4 software was used for meta-analysis of the selected literatures. The results showed that the papers included in this study had good homogeneity and no significant publication bias was found in the current analysis. The pooled estimates using a fixed-effects model showed that TV was more prevalent in HPV-infected women than in non-infected women [odds ratio (OR): 1.51, 95% confidence interval (CI): 1.29–1.75]; In turn, HPV was more widespread in TV-infected women than in uninfected women (OR: 3.62, 95% CI: 2.71–4.85). Moreover, the interaction between TV and HPV infection was insensitive to the deletion of some studies and correlation coefficients, consequently, the results were robust and reliable. These results suggested that TV is positively associated with HPV infection, and HPV is also a risk factor for TV infection.

Keywords:

• Trichomonas vaginalis
• HPV
• infection
• correlation

Introduction

Cervical cancer is one of the deadliest gynecological cancers in the world. Globally, more than 500,000 new cases and 274,000 deaths are reported each year (Siegel et al. 2016). Persistent infection with human papillomavirus (HPV) is a major factor in the development of cervical cancer (Spriggs and Laimins 2017), with HPV16 and 18 subtypes alone causing more than 70% of cervical cancer (Lsche et al. 2021, Zhang et al. 2022). It is estimated that women have a >80% chance of being infected with HPV in their lifetime, but most are cleared spontaneously by the host immune response (Hirth 2019). Local cervical factors may determine the outcome of HPV. For this reason, there is a lot of interest in researching factors that predispose to the acquisition and persistence of this virus.

Trichomonas vaginalis (TV), a flagellate parasite in the urethra and urogenital tract is the most common non-viral pathogen in the world. According to statistics, there are about 156 million new cases of TV, and more than 276 million TV patients in the world every year (Rowley et al. 2019, Mabaso and Abbai 2021). TV infection can cause vaginitis, urethritis and prostatitis, and other diseases (Schwebke and Burgess 2004). In addition, it is an important source of reproductive diseases, a promoter of HIV transmission and acquisition (Masha et al. 2019, Raffone et al. 2021), associated with infertility, premature labour, and possibly a higher incidence of cervical cancer (Hirt and Sherrard 2015). Studies have found that TV infection might determine an increased risk of HPV persistence by inducing inflammatory response in the cervico-vaginal epithelium, impairing vaginal epithelial cells, degrading cervical mucus, and cleaving immunoglobulin A (Raffone et al. 2021). The risk of HPV16 infection increased 6.5 times in the presence of TV (Lazenby et al. 2014). Although there is a lot of evidence pointing that TV increases the risk of HPV infection, some studies still claim that there is no correlation between the infections of TV and HPV. Due to the inconsistency between numerous research reports, whether the TV infection is associated with HPV infection in women remains a controversial issue.

To search a theory closer to the truth in chaos, this study conducted a meta-analysis on TV infection in HPV patients and HPV infection in TV, and analysed the publication bias and heterogeneity of the selected literatures. The purpose of this study was to analyse the relationships between TV and HPV infection in females, and to provide a new perspective for the prevention and treatment of TV and HPV co-infection.

Methods

Literature search

The study was registered at PROSPERO (www.crd.york.ac.uk/PROSPERO) as CRD42022343402. Relevant studies on the association between TV and HPV infection were identified through an extensive search of PubMed, Ovi-Mediline medical literature according to the preferred reporting items for systematic reviews and meta-analysis(PRISMA) extended statement. The keywords used for the literature search were ‘Trichomonas vaginalis’ or ‘TV’, in combination with ‘human papillomavirus’ or ‘HPV’.

Literature inclusion and exclusion criteria

The inclusion criteria were as follows: (1) the statistical data of HPV rate among people with or without TV infection were included in the article; (2) the statistical data of TV infection rate among people with or without HPV infection were included in the article; (3) the data of TV infection rate among people with or without HPV infection were compared in the article; (4) the article type belongs to the original study, and the full text is available; (5) the article was a case-control study; (6) the latest and most comprehensive articles should be selected for articles with the same statistical items.

The exclusion criteria were as follows: (1) only HPV rate in TV patients was reported, while there was no data related to non-TV patients in the article; (2) only the infection of TV in HPV-positive people was reported, but no data were available on HPV-negative people in the article; (3) the article was a non-population study; (4) the article published too early (before 2000) and the data was not of reference significance; (5) only TV infection or HPV infection was reported in the article.

Statistical analysis

The Newcastle-Ottawa scale (NOS) was used to evaluate the quality of the epidemiological investigations, and articles that scored ≥ 6 were considered to be high-quality articles. A meta-analysis was performed on selected articles using RevMan 5.4 software, and a forest plot was drawn to calculate the p-value and conduct the heterogeneity test. Statistical testing with p < 0.05 was considered statistically significant. If p < 0.05 or inconsistency (I²) ≥ 50%, heterogeneity among the research results was confirmed, and the origin of the heterogeneity was analysed, the results were then analysed using a random-effect model. If p > 0.05 or I² ≤ 50%, good consistency among the research results was confirmed, and the results were analysed using a fixed-effect model along with the calculation of the OR and 95% confidence interval. The publication bias was excluded using NOS.

Sensitivity analyses

To assess the stability of our findings, we used an influence analysis to estimate the impact of each study on the results of this study. We deleted the studies individually and then recalculated the pooled estimate. The individual studies were considered to have little impact on the results of this study if the recalculated pooled estimate did not exceed the original 95% CI in which it was included.

Results

Literature inclusion

A total of 310 articles were retrieved (Figure 1). After removing 197 irrelevant articles, the remaining 113 articles were reviewed by reading through the content according to the inclusion and exclusion criteria. Sixteen relevant articles were selected after evaluation. Of these, twelve of the articles (Table 1) concerned the prevalence of TV infection among HPV-positive patients, and HPV-positive patients were assigned to the study group, while people who were not infected with HPV were assigned to the control group. The remaining four articles (Table 1) concerned the incidence of HPV among people with TV infection, in which people with TV infection were assigned to the study group, while people without TV infection were assigned to the control group. There were plenty of samples in these articles, which were deemed as high-quality articles based on the meta-analysis (NOS score ≥ 6).

Figure 1. Flow-chart of article selection for inclusion in meta-analysis.

Table 1. A detailed list of the articles included in this study.

Number	Title	First author	Year	Location	Total sample	The study group	Control group	NOS score
1	Incidence of Gardnerella vaginalis, Candida sp and human papilloma virus in cytological smears.	Murta et al.	2000	Brazil	786	9/390 (2.3%)	14/396 (3.5%)	6
2	Epidemiology of Trichomonas vaginalis and human papillomavirus infection detected by real-time PCR in Flanders.	Depuydt et al.	2010	Flanders	62944	58/9518 (0.61%)	176/53426 (0.33%)	7
3	Association of human papillomavirus infection with other microbial pathogens in gynaecology	Zheng et al.	2010	Chinese	857	3/266 (1.12%)	7/591 (1.18%)	7
4	Prevalence of human papillomavirus and co-existent sexually transmitted infections among female sex workers, men having sex with men and injectable drug abusers from eastern India.	Ghosh et al.	2012	Eastern India	45	8/35 (22.9%)	0/10 (0%)	6
5	Association of Trichomonas vaginalis and cytological abnormalities of the cervix in low risk women	Donders et al.	2013	Belgium	63251	58/9580 (0.61%)	178/53,671 (0.33%)	6
6	Human papillomavirus and other genital infections in indigenous women from Paraguay: a cross-sectional analytical study	Mendoza et al.	2013	Paraguay	181	5/42 (11.9%)	14/139 (10.1%)	6
7	Prevalence of microorganisms co-infections in human papillomaviruses infected women in Northern China	Liu et al.	2016	Northern China	4290	55/1452 (3.8%)	93/2838 (3.3%)	6
8	Prevalence and risk factors of Trichomonas vaginalis among female sexual workers in Nairobi, Kenya.	Lockhart et al.	2019	Kenya	348	11/97 (11.3%)	21/247 (8.5%)	6
9	Prevalence of human papillomavirus, human immunodeficiency virus and other sexually transmitted infections among female sex workers in Togo: a national cross-sectional survey.	Ferré et al.	2019	France	310	9/102 (8.8%)	11/208 (5.3%)	6
10	Association between vaginal infections and the types and viral loads of human papillomavirus: a clinical study based on 4,449 cases of gynaecologic outpatients.	Wang et al.	2020	China	4449	28/1115 (34.57%)	53/3334 (1.60%)	6
11	Prevalence of cervical HPV infection, sexually transmitted infections and associated antimicrobial resistance in women attending cervical cancer screening in Mali.	Jary et al.	2021	Mali	144	7/90 (8%)	3/54 (5%)	7
12	Detection of sexually transmitted pathogens and co-infection with human papillomavirus in women residing in rural Eastern Cape, South Africa.	Taku et al.	2021	South Africa	205	11/32 (34.4%)	55/173 (31.8%)	7
13	High prevalence of high-risk human papillomavirus infection among women with Trichomonas vaginalis infection on monolayer cytology.	Noël et al.	2010	Belgium	507	13/57 (22.8%)	47/450 (10.4%)	6
14	Epidemiological investigation of the relationship between common lower genital tract infections and high-risk human papillomavirus infections among women in Beijing, China.	Zhang et al.	2017	Beijing	954	3/16 (18.8%)	73/938 (7.8%)	6
15	Trichomonas vaginalis as a risk factor for human papillomavirus: a study with women undergoing cervical cancer screening in a northeast region of Brazil.	Belfort et al.	2021	Brazil	562	79/107 (7%)	184/455 (40%)	7
16	Association between common vaginal infections and cervical non-human papillomavirus (HPV) 16/18 infection in HPV-vaccinated women.	Hu et al.	2021	China	10776	26/97 (26.8%)	963/10679 (9.0%)	6

Note: The experimental group in studies 1–12: the incidence of TV infection among HPV-positive patients; the control group in studies 1–12: the incidence of TV infection among people who were not infected with HPV. The experimental group in studies 13–16: the incidence of HPV among people with TV infection; the control group in studies 13–16: the incidence of HPV among people without TV infection.

Associations between TV and HPV infection

These twelve articles on the incidence of TV infection among HPV-positive patients included a total of 137,806 research subjects, with 22,719 cases in the study (HPV-positive patients) groups and 115,087 cases in the control (people who were not infected with HPV) groups (Murta et al. 2000, Depuydt et al. 2010, Zheng et al. 2010, Ghosh et al. 2012, Donders et al. 2013, Mendoza et al. 2013, Liu et al. 2016, Ferré et al. 2019, Lockhart et al. 2019, Wang et al. 2020, Jary et al. 2021, Taku et al. 2021). The heterogeneity test yielded Chi² = 11.89, p = 0.37, and I² = 7%, which indicated that heterogeneity among the results cannot be considered (p > 0.05 or I² ≤ 0%) (Figure 2A). Consequently, the fixed-effect model was employed to perform the analysis, and the results showed that the OR and 95% confidence interval were calculated as 1.51 and 1.29–1.75, and z and p-values of the combined effect size were 5.28 and <0.00001, respectively. Based on these results, the incidence of TV infection among people with or without HPV infection was significantly different (p < 0.05).

Figure 2. Forest map of estimate of association between TV and HPV infection. (A) The articles concerned the incidence of TV among people with or without HPV infection. (B) The articles concerned the incidence of HPV among people with or without TV infection.

In addition, four studies on the infection rate of HPV among people with TV infection were included in this work (Table 1) (Noël et al. 2010, Zhang et al. 2017, Belfort et al. 2021, Hu et al. 2021). These four articles included a total of 12,799 research subjects, with 277 cases in the study groups (with TV infection) and 12,522 cases in the control groups (without TV infection). The heterogeneity test yielded Chi² = 1.56, p = 0.67, and I² = 0, which indicated that heterogeneity among the results cannot be considered (p > 0.05 or I² ≤ 0%) (Figure 2B). Therefore, the fixed-effect model was employed to perform the analysis, and the results showed that the OR and 95% confidence interval were calculated as 3.62 and 2.71–4.85, and z and p-values of the combined effect size were 8.68 and <0.00001, respectively. Based on these results, the infection rate of HPV between people with or without TV infection was significantly different (p < 0.05).

Publication bias

As shown in Figure 3, the horizontal axis was the OR of the effect size, and the vertical axis was the standard error. The accuracy of the study increased with increasing sample size, and most of the twelve epidemiological investigations were based on large samples, which were concentrated in the middle and top part of the funnel plot. However, the symmetry of the funnel plot was not good enough (Figure 3A). So egger test was further carried out, and the result showed egger plot was symmetrical (p = 0.338) (Figure 4A). The other four articles included have decent symmetry (Figure 3B). However, the number of articles included in this analysis was small. So egger test was further carried out, and the result showed egger plot was symmetrical (p = 0.308) (Figure 4B), These results suggested that publication bias had no meaningful effect on the results of the analysis.

Figure 3. Funnel diagram to assess publication bias. (A) The articles concerned the incidence of TV among people with or without HPV infection. (B) The articles concerned the incidence of HPV among people with or without TV infection.

Figure 4. Egger plot to assess publication bias. (A) The articles concerned the incidence of TV among people with or without HPV infection. (B) The articles concerned the incidence of HPV among people with or without TV infection.

Sensitivity analyses

As shown in Figure 5, the solid lines at 1.51 and 3.62 represented the primary OR values, and the solid lines at 1.30, 1.76, 2.71, and 4.85 represented the original 95% CI for all these studies. The open circles indicated that for the named study removed, the pooled estimate was still within the original 95% CI, implying that no study had a considerable effect on the results of the study.

Figure 5. Sensitivity analyses. (A) The articles concerned the incidence of TV among people with or without HPV infection. (B) The articles concerned the incidence of HPV among people with or without TV infection.

Discussion

Studies have shown that there is a connection between genital tract inflammation and cervical cancer. The association between self-reported abnormal vaginal secretions and cervical intraepithelial neoplasia (CIN) in HPV-infected women further suggested a link between genital tract abnormalities and cervical cancer (Discacciati et al. 2006, Venegas et al. 2011). The relationship between TV infection and cervical cancer has become a research hot spot in recent years. On the one hand, TV infection was considered to be a risk factor for the persistence or progression of low-grade cervical precancerous lesions (Raffone et al. 2021). On the other hand, TV was believed to promote HPV infection and induce cervical cancer. At present, there have been many studies on the relationship between TV and HPV infection, but the results were inconsistent.

To the best of our knowledge, this meta-analysis of 150,605 women was the first systematic assessment of the association between TV and HPV infection. According to the collected literatures, we analysed the TV infection rate on 137,806 women with or without HPV, and the overall odds ratio was 1.51. While the overall odds ratio of HPV infection among 12,799 women carrying TV or not was 3.62. There was no heterogeneity and publication bias in these results. Our results showed evidence of a positive association between TV and HPV.

Several hypotheses have been postulated, supporting this association. One theory is that parasites can be potential catalysts for the development of secondary infections, including HPV, through the production of a variety of enzymes associated with cytotoxicity and degradation of basement membrane components, such as cysteine proteases (Rodriguez-Cerdeira et al. 2012). In addition, some studies have shown that 82% of TV carry double-stranded RNA viruses, which may be associated with different expressions of enzymes that influence parasite virulence (Wang et al. 1987, Arroyo and Alderete 1995, Rughooputh and Greenwell 2005). Thus, TV also has the potential to alter the transmission of various sexually transmitted diseases, particularly HPV, by increasing virulence.

Another hypothesis proposes that TV infection may determine the increased risk of HPV persistence by inducing an inflammatory response in cervical-vaginal epithelial cells, damaging vaginal epithelial cells, degrading cervical mucus, and lyzing immunoglobulin A (Barrington et al. 1997, Schwebke and Burgess 2004, Costa et al. 2005). Provenzano and Alderete found that TV lysis products have the ability to degrade IgA and disrupt the protective mucosal barrier (Provenzano and Alderete 1995), thereby promoting HPV adhesion and invasion. In addition, TV infection induced an inflammatory response of vaginal epithelial cells (Lin et al. 2015), affected the vaginal environment of patients, and reduces the content of vaginal lactic acid bacteria (Yuan et al. 2021), which may also provide rationality for the increased risk of HPV infection.

Currently, there are still few studies on the mechanism of TV influencing HPV infection. Although there are already vaccines against HPV16 and 18 subtypes, they play an important role in the prevention of cervical cancer. But if a prospective study shows a cause-and-effect relationship between TV and HPV, then TV needs to be given more attention in the fight against cervical cancer than vaccination.

Conclusion

This meta-analysis revealed a positive association between TV and cervical HPV infection. TV is a risk factor for HPV infection and in turn, HPV infection is significantly associated with TV infection. Most of the inferences drawn in our study rely on a limited number of studies, which may compromise the generality of our findings. Randomised clinical trials may be considered to determine the impact of trichomoniasis control measures on HPV infection. Given how common these conditions are among women worldwide, further research in this area is imperative. In addition, more data from prospective studies are needed to accurately assess the time series of these two conditions to determine cause and effect and to identify specific subgroups with stronger associations between TV and HPV.

Ethical approval

The study was reviewed and approved by the Ethics Review Committee of Xinxiang Medical University (Reference No. 2015016).

Author contributions

X.M. and Z.Z. designed the study and critically revised the paper. R.Z., D.L., X.X., Y.Y., M.G., L.Z., and S.Z. collected the relevant articles and performed the meta-analysis. X.M., Z.Z., X.T., Z.Y., and S.W. analysed the results and wrote the manuscript. All authors contributed to the article and approved the final version.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The original contributions presented in the study are included in the article. Further inquiries can be directed to the corresponding author.

Additional information

Funding

This study was financially supported by the National Natural Science Foundation of China (No. 81802028), the Science and Technology Planning Project of Xinxiang City (No. GG2021011), and the Doctoral Scientific Research Activation Foundation of Xinxiang Medical University (Nos. XYBSKYZZ202140 and XYBSKYZZ201631).