Shlomo M. Stemmer, MD, MS, MBA, Eli Mordechai, Ph.D., Martin E. Adelson, PhD,
Scott E. Gygax, PhD, David W. Hilbert, PhD
PII: S0002-9378(17)32481-X
DOI: 10.1016/j.ajog.2017.12.006
Reference: YMOB 11975
Please cite this article as: Stemmer SM, Mordechai E, Adelson ME, Gygax SE, Hilbert DW, Trichomonas
vaginalis is most frequently detected in women at the age of peri-/pre-menopause–an unusual pattern
for a sexually transmitted pathogen, American Journal of Obstetrics and Gynecology (2018), doi:
10.1016/j.ajog.2017.12.006.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
PT
4 Shlomo M. Stemmer, MD, MS, MBA,a, b; Eli Mordechai, Ph.D. b, Martin E. Adelson, PhD b;
RI
6
SC
a
7 Virtua Hospital, Department of Obstetrics and Gynecology, Voorhees, NJ 08043
U
b
8 Medical Diagnostic Laboratories, Hamilton, NJ 08690
9
AN
M
10 Running title: Detection rates of T. vaginalis infection by age and state
D
11
TE
13
15
16 Conflict of interest and financial disclosure: This study was funded by Medical Diagnostic
18 Laboratories (MDL).
1
ACCEPTED MANUSCRIPT
19 Abstract
21 infection (STI). However, as it is not a reportable disease in the United States, there is limited
PT
22 information on the age of infected individuals and their geographic distribution.
RI
24 Chlamydia trachomatis by age and state in a commercial laboratory setting .
SC
25 Methods: Quantitative Real-time PCRs (qPCRs) were used to detect the presence of T. vaginalis
U
27 1,554,966 and 1,999,077 samples from females age 10 to 79 were retrospectively analyzed for
28
AN
the presence of T. vaginalis and C. trachomatis, respectively.
M
29 Results: The highest detection rate of an infection with T. vaginalis was between ages 47 to 53.
30 For C. trachomatis the highest detection rate was between ages 14 to 20. T .vaginalis detection
D
31 rate distribution by age shows a bimodal pattern with first peak at ages 21-22 (4.0-4.1%) and a
TE
32 higher second peak at ages 48 to 51 (5.4 -5.8%). C. trachomatis prevalence distribution by age
shows a maximum peak of 8.6% at age 17 and a rapid decline thereafter. In general, the detection
EP
33
34 rates of both pathogens were higher in the southeast and in states along the Mississippi River
C
35 Valley than in other parts of the country. A nucleotide polymorphism associated with T.
AC
36 vaginalis metronidazole resistance (ntr6TVK80STOP) was not associated with age and was found
38 Conclusions: The detection rate of T. vaginalis does not appear to decrease with age as observed
39 for C. trachomatis and reaches maximum rates in women ages 48 to 51 years old. The
2
ACCEPTED MANUSCRIPT
40 geographic distribution of T. vaginalis appears to broadly similar to that of other STIs. The
41 ntr6TVK80STOP polymorphism did not have a specific association with age or geography.
42
PT
43
RI
44
SC
45
46
U
47
AN
48
M
49
D
50
TE
51
EP
52
C
53
AC
54
55
56
57
3
ACCEPTED MANUSCRIPT
58 Introduction
59 Trichomoniasis, caused by the protozoan Trichomonas vaginalis, is the world’s most common
60 non-viral sexually transmitted infection (STI), with an estimated 143 million new cases per year
PT
61 globally 1. In the United States (US) it is estimated that there are 3.7 million new infections per
2
62 year and the overall prevalence among reproductive-aged women is 3.1% 3. Populations with
RI
63 an elevated prevalence of T. vaginalis infection include African American women (13%) 3,
women attending STI clinics (26%) 4 and incarcerated women (14 - 32%) 5, 6. In comparison, the
SC
64
65 prevalence among non-Hispanic white women is only 1.3% ³. Symptoms in women include a
U
66 malodorous green-yellow vaginal discharge, vulvar irritation and colpitis macularis (i.e.
67 AN
"strawberry cervix"). However, up to 85% of infected women are asymptomatic
4
7
and 6.5% of
68 asymptomatic women at STI clinics were infected with T. vaginalis , indicating that
M
69 asymptomatic colonization is common. Complications associated with T. vaginalis infection
8
70 include adverse pregnancy outcomes (e.g. pre-term labor and low birth weight infants) and
D
72 culture, molecular methods, such as quantitative real-time polymerase chain reaction (qPCR)
10, 11
73 have proven to be much more sensitive . Most infections can be successfully treated with
EP
74 metronidazole 7, although resistance has been reported in 4.3% - 9.6% of isolates 12, 13.
C
75 Most STIs have the highest prevalence among younger individuals and decrease in frequency
AC
76 with increasing age. For example, the prevalence of chlamydia is 3.4% among women aged 20-
77 24 years and decreases to 1.2% at ages 25-30 and to < 1% thereafter 14. Similarly, the prevalence
78 of gonorrhea is the highest between the ages 15-24 (≥ 0.5%) and decreases to ≤ 0.2% thereafter
14
79 . However, three studies evaluating the prevalence or detection rate of T. vaginalis have found
80 a very different pattern. Sutton et al. found that women 30-40 years of age had a prevalence of ≥
4
ACCEPTED MANUSCRIPT
81 3.6% compared to ≤ 2.3% for women 14-29 years of age 3. Ginocchio et al. found the highest
15
82 rate (12%) in women ≥ 40 years of age compared to ≤ 8.5% in women < 40 years of age .
83 Finally, Stemmer et al. found the highest detection rate (6.3%) was in women 46-55 years old 16.
84 These studies evaluated a relatively modest number of specimens—3,754; 7,593 and 78,428,
PT
85 respectively 3, 15, 16. Lastly, a recent study of 9,186 women in the United Kingdom found that age
RI
86 was an independent risk factor for T. vaginalis infection 17.
SC
87 According to the Centers for Disease Control and Prevention (CDC), the highest prevalence of
88 C. trachomatis is found in Alaska and the Southeastern US, with rates exceeding 600 cases per
U
89 100,000 individuals in Mississippi, Louisiana and Alabama 14. Lower rates were reported for the
90 AN
mid-west, western and northeastern states 14. Detection rates using nucleic acid amplification in a
91 commercial laboratory setting were also highest in specimens from the Southeastern US (7.9%)
M
92 in comparison to other regions (≤ 6.6 %) 15. Similar results were obtained for T. vaginalis, with a
93 14% detection rate in specimens from the Southeastern United States and ≤ 9.5% detection rates
D
95 The goals of this study were to evaluate the relationship between age and geography with T.
EP
97 evaluated C. trachomatis detection rates over a similar time frame. Lastly, we assessed the
C
99 resistance 18.
100 Methods
102
5
ACCEPTED MANUSCRIPT
103 Cervico-vaginal sample specimens sent to Medical Diagnostic Laboratories, L.L.C. (MDL), from
104 January 1, 2014, to December 31, 2014, by practicing obstetricians and gynecologists for
105 diagnostic testing for T. vaginalis (n = 1,554,966) and C. trachomatis (n = 1,999,077), and were
106 analyzed using quantitative real-time polymerase chain reactions. Specimens were processed
PT
107 using a computer-generated number to protect patient anonymity, and clinical information was
RI
108 not available for this retrospective study. Only one specimen was analyzed per patient. Age, state
109 of residence, and positivity status for T. vaginalis and C. trachomatis were the only information
SC
110 used for this analysis. The specimens were obtained from a mixture of urban, suburban and rural
111 practices. With regards to payment for T. vaginalis testing, 45.9% of patients had commercial
U
112 insurance and 31.1% had government-provided insurance. Patients’ dates of birth were obtained
113
AN
from their health care provider and confirmed by MDL before submission of payment claim to
third party payers or receipt of cash payment. Cervico-vaginal specimens were collected by the
M
114
115 evaluating physician using a OneSwab® (Medical Diagnostic Laboratories) and placed in 2 ml of
D
116 UTM™-RT transport medium (Copan Diagnostics) and were transported to the laboratory within
TE
118
EP
120
C
121 DNA was extracted from the specimens using the X-tractor Gene automated nucleic acid
AC
122 extraction system (Corbett Robotics) according to the manufacturer's instructions. C. trachomatis
123 and T. vaginalis were detected using qPCRs performed using a CFX384 real-time thermocycler
124 (Bio-Rad). These assays have been validated as having a sensitivity of ≤ 100 copies of template
125 and specificity against a broad panel of bacteria, fungi and viruses found in the lower female
6
ACCEPTED MANUSCRIPT
126 genital tract. All 384 well reaction plates included four separate no-template control wells and
127 six positive control wells with 102, 104 or 106 copies of the qPCR target template. Each assay
128 evaluated 400 nl of extracted DNA. The assay used to detect the ntr6TV(K80STOP)
129 polymorphism associated with T. vaginalis metronidazole resistance has been described
PT
130 previously 18.
RI
131
SC
133
134 Positivity rates, normality tests and correlation analysis were performed using GraphPad Prism
U
135 5.01. The D'Agostino and Pearson omnibus normality test was used to evaluate the distribution
136
AN
of positivity rates by age. Ninety-five percent confidence internals were determined using the
138 pairwise comparisons were performed by Fisher's exact tests with False Discovery Rate to
D
139 control for multiple comparisons using R 3.4.2 and the package "RVAideMemoire". An alpha of
TE
140 0.05 was used to determine significance. Heat maps of the United States were generated using
142
143 Results
C
AC
144 We first analyzed the positivity rates for T. vaginalis as a function of age (Fig. 1). We observed
145 that the positivity rate steadily increased from 10 years (0.5%) through 22 years (4.1%) and then
146 slowly decreased through 31 years (3.6%). After this point, the positivity rate begins to slowly
147 and steadily increase again until it reaches a peak at age 48 (5.8%) and then begins to drop,
148 although it remains > 5% through 53 years and > 4% through 60 years. We compared the
7
ACCEPTED MANUSCRIPT
149 detection rate for specimens from patients at each to every other age (Fig. 2). In general, the
150 detection rates for specimens from patients aged 40 - 55 were greater than those from patients
151 aged < 40 (upper left quadrant) and those from patients aged 60 - 80 were less than those of
152 patients aged < 60 (lower right quadrant). We next evaluated specimens for the positivity rate of
PT
153 C. trachomatis as a function of patient age (Fig. 3). From 10 years (the lowest age evaluated in
RI
154 the study) through approximately 27 years we observe a normal distribution (D'Agonisto and
155 Pearson omnibus normality test, P > 0.05), beginning with very low positivity rates (≤ 0.9% at
SC
156 10-11 years), peaking at 8.6% at 17 years and then dropping down to 2.0% by 27 years. After
157 this point we observe a steady slow decline from 27-46 years, with positivity rates decreasing
U
158 from 2.0% to 0.4%. From 47-79 years the positivity rate was ≤ 0.3%. We also compared the
159
AN
detection rate for specimens from patients at each age to each other (Fig. 4). Overall, the
detection rates were greater for specimens from patients aged 16 - 30 than every other age group
M
160
161 evaluated.
D
162 The specimens analyzed in this study were obtained from all 50 states as well as from
TE
163 Washington D.C (Tables 1 and 2). The states from which the greatest number of specimens were
164 collected and analyzed for testing (% of all C. trachomatis tests, % of all T. vaginalis tests) were
EP
165 Texas (19.0, 19.5), Florida (13.9, 14.3), Pennsylvania (8.8, 8.6), New Jersey (7.4, 8.1), Georgia
C
166 (6.0, 6.4) and Arizona (5.6, 6.1). Kentucky, Louisiana, Ohio, Illinois, Maryland, Michigan,
AC
167 Alabama, Virginia, Tennessee, Missouri, California, Delaware and Oklahoma all accounted for
168 1-5% of total tests for both pathogens, and all other states each individually accounted for < 1%
169 of total tests performed. Positivity rates by state are listed in Tables 1 and 2. With regards to T.
170 vaginalis positivity (Fig. 5), the highest rates were found among specimens from Mississippi
171 (9.0%), Wisconsin (7.8%), South Carolina (7.4%), Illinois (7.0%), Arizona (6.7%), Alabama
8
ACCEPTED MANUSCRIPT
172 (5.7%), Louisiana (5.7%), Kansas (5.5%), Georgia (5.3%), Washington D.C. (5.2%) and
173 Tennessee (5.2%). The lowest T. vaginalis positivity rates were among specimens from
174 Arkansas, Hawaii and North Dakota at 0.0%, followed by New Hampshire (0.4%), Minnesota
175 (1.1%), Massachusetts (1.7%), Utah (1.8%), New York (1.8%), Maine (1.8%) and California
PT
176 (1.8%). With regards to state-by-state comparisons (Fig. 6), we found that Alabama, Florida,
RI
177 Mississippi, Wisconsin and South Carolina had significantly greater detection rates than nearly
178 every other state that could be evaluated. It is important to note that there were no significant
SC
179 comparisons for Alaska, Hawaii, Montana, North Dakota or South Dakota due to the low number
U
181 AN
The states with the highest positivity rates for C. trachomatis (Fig. 7) were Maine (6.4%), South
182 Dakota (4.6%), Louisiana (4.3%), Alabama (4.0%) and Wisconsin (4.0%). The states with the
M
183 lowest positivity rates for C. trachomatis were Arkansas, Hawaii and North Dakota at 0.0%,
184 followed by Delaware (1.1%), New York (1.2%), Minnesota (1.2%), New Hampshire (1.3%)
D
185 and Massachusetts (1.5%). With regards to state-by-state comparisons (Fig. 8), we found that
TE
186 detection rates in Alabama, Georgia, Louisiana, Mississippi South Carolina and Wisconsin were
187 significantly greater than most other states. Although the detection rate was greatest for Maine, it
EP
188 was only significantly greater than one other state, due to the small number of specimens
C
189 evaluated from this state (n = 73). We also evaluated the positivity rates of the pathogens for
AC
190 correlations at the state level (Fig. 9) and found a weak positive correlation (R-squared = 0.29, P
192 Last, we analyzed our detection rate for a specific polymorphism (K80STOP) in the ntr6TV gene
18
193 of T. vaginalis which is associated with metronidazole resistance . Overall, 876 out of 58,891
194 T. vaginalis-positive specimens were positive for this polymorphism, resulting in a positivity rate
9
ACCEPTED MANUSCRIPT
195 of 1.5%. We evaluated the positivity rate as a function of age (Fig. 10) but there was no clear
196 pattern, with rates seemingly fluctuating at random. There were no significant differences in
197 detection rates between specimens from patients of different ages. When evaluated for detection
198 rate by geography, we found the highest positivity rates for this polymorphism were found
PT
199 among specimens from Vermont (14.3%), New Mexico (11.1%), West Virginia (5.6%) and
RI
200 California (5.3%) (Table 3, Fig. 11). We failed to detect the ntr6TV (K80STOP) polymorphism in
201 specimens from the following states: Arkansas, Colorado, Hawaii, Massachusetts, Maine,
SC
202 Montana, New Hampshire, North Dakota, South Dakota, Washington Wisconsin or Wyoming.
203 The rates in West Virginia and California were significantly greater than that of many other
U
204 states, whereas the rate in Wisconsin was significantly lower than many other states (Fig. 12). It
205
AN
is important to note that although Vermont had the greatest detection rates, it was significantly
greater than only one other state (Wisconsin) due to the low number of specimens evaluated (n =
M
206
207 7). In contrast, New Mexico, with the second greatest detection rate, had a significantly greater
D
208 detection rate than nine other states due, in part, to the larger number of specimens evaluated (n
TE
209 = 63).
210
EP
211 Discussion
C
212 Previous studies have found the prevalence of T. vaginalis, unlike most STIs, to be higher in
AC
213 women older than 30 years of age 3, 15, 16. We have replicated these findings using a much larger
214 specimen collection evaluating 20-fold greater specimens than any prior study, thus allowing us
215 to determine the peak age of T. vaginalis detection rate more precisely as between 46 and 55
216 years of age. The reason for this later in life peak is unknown. One possible explanation is an
10
ACCEPTED MANUSCRIPT
217 increase in sex partners and unprotected sexual activity 19. However, one would expect to see a
14
218 similar pattern with other STIs, such as chlamydia and gonorrhea, which is not the case .
219 Therefore, we hypothesize that the age-related increase in T. vaginalis detection rate is due to
220 either increased susceptibility to T. vaginalis infection in these women or re-activation of a pre-
PT
221 existing latent T. vaginalis infection in asymptomatic women.
RI
222 T. vaginalis infection is associated with abnormal vaginal flora that is deficient in
SC
223
21
224 and thus reduced glycogen concentration in vaginal secretions , one may speculate that this
U
225 mechanism is responsible for the age-related increase we observed. However, a similar increase
226
22
AN
in susceptibility to gonorrhea and chlamydia has also been noted in women with abnormal
227 vaginal flora . Another possibility is altered mucosal immunology in the lower reproductive
M
228 tract. For example, menopause is associated with a reduction in B cells and CD8+ T cells in the
24
230 secretory leukocyte protease inhibitor (SLPI) , which protects mucosal tissues from serine
TE
231 proteases. As cysteine proteases produced by T. vaginalis degrade SLPI 25, we speculate that this
232 protein is protective against infection and its deficiency after menopause could contribute to our
EP
233 observed increase in detection rates. Although pregnant women also have altered mucosal
C
26
234 immunity in the lower genital tract , no association was found between pregnancy and T.
AC
235 vaginalis infection 3. An alternate explanation is that T. vaginalis can persist after an initial
236 episode in a dormant, undetected fashion and then is re-activated during or after menopause.
237 Although T. vaginalis does not form cysts, an environmentally-resistant pseudocyst form has
27
238 been described . We speculate that this form may persist in the lower reproductive tract and
239 differentiate into the infectious trophozoite form and proliferate in response to hormonal,
11
ACCEPTED MANUSCRIPT
240 epithelial and/or immunological changes associated with menopause. Lastly, the endocervical
241 atrophy that accompanies menopause 28 may provide protection against pathogens that infect this
243 vaginalis, which infects both the vagina and the cervix 31.
PT
244 With regards to geographic distribution, we observed that detection rates for both pathogens
RI
245 were greater in the Southeastern US (Mississippi, Louisiana, and Alabama) and along the
SC
246 Mississippi River Valley (Missouri, Indiana and Wisconsin) than in other parts of the country.
247 Our results for C. trachomatis are largely consistent with those of the CDC, which found that
U
248 Mississippi, Louisiana and Alabama are the states with the 2nd-4th greatest prevalence of this
249 AN
disease 14. Other states where our results were concordant with those of the CDC include South
250 Carolina, Oklahoma, Texas, South Dakota and Illinois. States where the CDC reports a high
M
251 prevalence where we found a low detection rate include New Mexico, Arkansas and New York.
252 Conversely, we found high detection rates for Wisconsin and Kansas whereas the CDC-reported
D
253 prevalence is low. One simple explanation for these discrepancies is that the CDC is reporting all
TE
254 cases state-wide whereas our detection rate determination is limited to specimens sent to our
255 laboratory. In states where our detection rate is higher than expected we may be receiving
EP
256 specimens from patients where the T. vaginalis prevalence is greater, such as among black
C
14
257 women or those with lower socioeconomic status 32. One possible explanation for this pattern
AC
33
258 is that the Southern US is the region with the highest frequency of douching , which is
259 associated with T. vaginalis infection 3. Conversely, in states where our detection rate is lower
260 than anticipated we may be receiving specimens from patients who are disproportionately white
12
ACCEPTED MANUSCRIPT
262 Strengths of our study include a very large sampling population from all states and ages 10 to 80
263 years old. Additionally, we used a sensitive and specific qPCR test and measured the detection of
265 However, it is important to note that this polymorphism is associated with in vitro resistance to
PT
266 metronidazole and its association with metronidazole treatment failure has not yet been
RI
267 demonstrated. Our analysis was limited by sampling that was not generally random. Women may
268 have been tested during annual gynecological examination, for newly diagnosed pregnancy or
SC
269 due to concerns of a STI. However, because most women infected with T. vaginalis are
270 asymptomatic 7, the use of clinical considerations for T. vaginalis detection is largely inaccurate.
U
271 As there is no routine screening for T. vaginalis, our testing population may be skewed towards
272
AN
symptomatic patients which would bias our detection rates upward. Another weakness is the lack
of demographic data on the women tested, specifically number of sexual partners and racial
M
273
274 distribution. In addition, we do not know if the infection was recurrent, asymptomatic or if there
D
276 In contrast to chlamydia, trichomoniasis is not a reportable disease and therefore we do not have
277 extensive data regarding its epidemiology in the US. In general, we found a similar pattern for T.
EP
278 vaginalis detection as for C. trachomatis, with states in the Southeastern US and along the
C
279 Mississippi River Valley having greater detection rates. These results are generally consistent
AC
280 with those of a prior study examining the US distribution of T. vaginalis detection, which found
15
281 the highest rates in the Southeastern US . We also evaluated the detection rate of the
284 Mexico (11%). We also found high detection rates in several regions of the country, including
13
ACCEPTED MANUSCRIPT
285 the west coast (California and Oregon), the Midwest (north Dakota, Idaho and Oklahoma), and
286 the Northeast (new York and Rhode Island) along with Iowa and West Virginia. The higher
287 detection rates in these areas may indicate a greater degree of metronidazole resistance and
288 potential treatment failure with this agent. However, as routine testing for metronidazole
PT
289 susceptibility is not available to most physicians, there is no way at this time to test this
RI
290 hypothesis.
SC
291 In summary, we have evaluated a very large number of specimens (n = ~1.5 million for T.
292 vaginalis, n = ~1.9 million for C. trachomatis), and confirmed and extended prior analyses of the
U
293 relationship between T. vaginalis maximum detection rates in women aged 46 to 55 years old.
294 AN
Explanations for this pattern, biological and/or behavioral, should be investigated. We also
295 provide data regarding the geographical distribution of the detection rates, which have only been
M
296 reported in one prior study examining a relatively small number of specimens (n= 7,593) 15. We
297 found that the geographic distribution of detection rates for T. vaginalis in the US is broadly
D
298 similar to those of other STIs. Lastly, we provide the first report of the detection rate for a T.
TE
299 vaginalis nucleotide polymorphism associated with metronidazole resistance 18 and found higher
300 rates in several geographically discontinuous clusters. Overall, these data broaden our
EP
301 understanding of the epidemiology of the most common non-viral STI and can serve as a
C
303 Acknowledgements
304 This study was funded by Medical Diagnostic Laboratories. Shlomo M. Stemmer, MD, is a paid
305 consultant for Medical Diagnostic Laboratories (MDL). We would like to thank Janet Cohen for
14
ACCEPTED MANUSCRIPT
307 References
308 1. NEWMAN L, ROWLEY J, VANDER HOORN S, et al. Global Estimates of the Prevalence and
309 Incidence of Four Curable Sexually Transmitted Infections in 2012 Based on Systematic Review
PT
310 and Global Reporting. PloS one 2015;10:e0143304.
311 2. SATTERWHITE CL, TORRONE E, MEITES E, et al. Sexually transmitted infections among US
RI
312 women and men: prevalence and incidence estimates, 2008. Sexually transmitted diseases
313 2013;40:187-93.
SC
314 3. SUTTON M, STERNBERG M, KOUMANS EH, MCQUILLAN G, BERMAN S, MARKOWITZ L. The
315 prevalence of Trichomonas vaginalis infection among reproductive-age women in the United
U
316 States, 2001-2004. Clinical infectious diseases : an official publication of the Infectious Diseases
320 2013;40:865-9.
D
321 5. FREEMAN AH, KATZ KA, PANDORI MW, et al. Prevalence and correlates of Trichomonas
TE
322 vaginalis among incarcerated persons assessed using a highly sensitive molecular assay. Sexually
324 6. SUTCLIFFE S, NEWMAN SB, HARDICK A, GAYDOS CA. Prevalence and correlates of Trichomonas
325 vaginalis infection among female US federal prison inmates. Sexually transmitted diseases
C
326 2010;37:585-90.
AC
327 7. WORKOWSKI KA, BOLAN GA, CENTERS FOR DISEASE C, PREVENTION. Sexually transmitted
328 diseases treatment guidelines, 2015. MMWR Recommendations and reports : Morbidity and
329 mortality weekly report Recommendations and reports / Centers for Disease Control 2015;64:1-
330 137.
15
ACCEPTED MANUSCRIPT
331 8. SILVER BJ, GUY RJ, KALDOR JM, JAMIL MS, RUMBOLD AR. Trichomonas vaginalis as a cause
332 of perinatal morbidity: a systematic review and meta-analysis. Sexually transmitted diseases
333 2014;41:369-76.
334 9. HILBER AM, FRANCIS SC, CHERSICH M, et al. Intravaginal practices, vaginal infections and HIV
PT
335 acquisition: systematic review and meta-analysis. PloS one 2010;5:e9119.
336 10. CALIENDO AM, JORDAN JA, GREEN AM, INGERSOLL J, DICLEMENTE RJ, WINGOOD GM. Real-
RI
337 time PCR improves detection of Trichomonas vaginalis infection compared with culture using
SC
338 self-collected vaginal swabs. Infectious diseases in obstetrics and gynecology 2005;13:145-50.
339 11. PILLAY A, RADEBE F, FEHLER G, HTUN Y, BALLARD RC. Comparison of a TaqMan-based real-
U
340 time polymerase chain reaction with conventional tests for the detection of Trichomonas
342 12.
AN
KIRKCALDY RD, AUGOSTINI P, ASBEL LE, et al. Trichomonas vaginalis antimicrobial drug
343 resistance in 6 US cities, STD Surveillance Network, 2009-2010. Emerging infectious diseases
M
344 2012;18:939-43.
D
345 13. SCHWEBKE JR, BARRIENTES FJ. Prevalence of Trichomonas vaginalis isolates with resistance to
TE
348 15. GINOCCHIO CC, CHAPIN K, SMITH JS, et al. Prevalence of Trichomonas vaginalis and coinfection
349 with Chlamydia trachomatis and Neisseria gonorrhoeae in the United States as determined by the
C
350 Aptima Trichomonas vaginalis nucleic acid amplification assay. Journal of clinical microbiology
AC
351 2012;50:2601-8.
352 16. STEMMER SM, ADELSON ME, TRAMA JP, DORAK MT, MORDECHAI E. Detection rates of
353 trichomonas vaginalis, in different age groups, using real-time polymerase chain reaction. Journal
355 17. NICHOLLS JE, TURNER KME, NORTH P, et al. Cross-sectional study to evaluate Trichomonas
356 vaginalis positivity in women tested for Neisseria gonorrhoeae and Chlamydia trachomatis,
16
ACCEPTED MANUSCRIPT
357 attending genitourinary medicine and primary care clinics in Bristol, South West England.
359 18. PAULISH-MILLER TE, AUGOSTINI P, SCHUYLER JA, et al. Trichomonas vaginalis metronidazole
360 resistance is associated with single nucleotide polymorphisms in the nitroreductase genes ntr4Tv
PT
361 and ntr6Tv. Antimicrobial agents and chemotherapy 2014;58:2938-43.
362 19. STEMMER S. Reply to letter: increased prevalence of Trichomonas vaginalis in mid-aged women
RI
363 is linked to sexual activity and not to hormonal changes. Journal of lower genital tract disease
SC
364 2013;17:e33-4.
365 20. BALKUS JE, RICHARDSON BA, RABE LK, et al. Bacterial vaginosis and the risk of trichomonas
U
366 vaginalis acquisition among HIV-1-negative women. Sexually transmitted diseases 2014;41:123-
367 8.
368 21.
AN
PABICH WL, FIHN SD, STAMM WE, SCHOLES D, BOYKO EJ, GUPTA K. Prevalence and
369 determinants of vaginal flora alterations in postmenopausal women. The Journal of infectious
M
370 diseases 2003;188:1054-8.
D
371 22. WIESENFELD HC, HILLIER SL, KROHN MA, LANDERS DV, SWEET RL. Bacterial vaginosis is a
TE
372 strong predictor of Neisseria gonorrhoeae and Chlamydia trachomatis infection. Clinical
373 infectious diseases : an official publication of the Infectious Diseases Society of America
EP
374 2003;36:663-8.
375 23. TRIFONOVA RT, LIEBERMAN J, VAN BAARLE D. Distribution of immune cells in the human
C
376 cervix and implications for HIV transmission. American journal of reproductive immunology
AC
377 2014;71:252-64.
378 24. SHIMOYA K, ZHANG Q, TEMMA K, et al. Secretory leukocyte protease inhibitor levels in
380 25. DRAPER D, DONOHOE W, MORTIMER L, HEINE RP. Cysteine proteases of Trichomonas vaginalis
381 degrade secretory leukocyte protease inhibitor. The Journal of infectious diseases 1998;178:815-
382 9.
17
ACCEPTED MANUSCRIPT
383 26. WIRA CR, FAHEY JV, RODRIGUEZ-GARCIA M, SHEN Z, PATEL MV. Regulation of mucosal
384 immunity in the female reproductive tract: the role of sex hormones in immune protection against
PT
387 Protist 2003;154:313-29.
388 28. TAY SK, SINGER A. The effects of oral contracdeptive steroids, menopause and hormonal
RI
389 replacement therapy on the cervical epithelium. In: Jordan J, ed. The Cervix. USA: Wiley-
SC
390 Blackwell, 2006.
391 29. BRUNHAM RC, PAAVONEN J, STEVENS CE, et al. Mucopurulent cervicitis--the ignored
U
392 counterpart in women of urethritis in men. The New England journal of medicine 1984;311:1-6.
393 30. BHATTACHARYYA MN, JEPHCOTT AE, MORTON RS. Diagnosis of gonorrhoea in women:
394
AN
comparison of sampling sites. British medical journal 1973;2:748-50.
395 31. BOEKE AJ, DEKKER JH, PEERBOOMS PG. A comparison of yield from cervix versus vagina for
M
396 culturing Candida albicans and Trichomonas vaginalis. Genitourinary medicine 1993;69:41-3.
D
398 factors and other sources of variation in the prevalence of genital chlamydia infections: A
400 33. ARAL SO, MOSHER WD, CATES W, JR. Vaginal douching among women of reproductive age in
401 the United States: 1988. American journal of public health 1992;82:210-4.
C
402
AC
403
404
405
18
ACCEPTED MANUSCRIPT
407
PT
409 Figure 2. Year-by-year comparison of T. vaginalis positivity rates. Colored spaces indicate
410 comparisons where a significant different (P < 0.05) between ages was observed. The color of
RI
411 the space corresponds to the state with the greater positivity rate (blue for the state listed on the
412 X-axis and red for the state listed on the Y-axis).
SC
413 Figure 3. C. trachomatis positivity as a function of age
U
414 Figure 4. Year-by-year comparison of C. trachomatis positivity rates. Colored spaces indicate
415
AN
comparisons where a significant different (P < 0.05) between ages was observed. The color of
416 the space corresponds to the state with the greater positivity rate (blue for the state listed on the
M
417 X-axis and red for the state listed on the Y-axis).
D
419 Figure 6. State-by-state comparison of T. vaginalis positivity rates. Colored spaces indicate
EP
420 comparisons where a significant different (P < 0.05) between states was observed. The color of
421 the space corresponds to the state with the greater positivity rate (blue for the state listed on the
C
422 X-axis and red for the state listed on the Y-axis).
AC
424 Figure 8. State-by-state comparison of C. trachomatis positivity rates. Colored spaces indicate
425 comparisons where a significant different (P < 0.05) between states was observed. The color of
19
ACCEPTED MANUSCRIPT
426 the space corresponds to the state with the greater positivity rate (blue for the state listed on the
427 X-axis and red for the state listed on the Y-axis).
428 Figure 9. Correlation of positivity rates for C. trachomatis and T. vaginalis testing by state. The
PT
429 red line represents a linear regression of the two positivity rates.
430 Figure 10. Positivity for ntr6TV(K80STOP) as a function of age. Ages for which there were no
RI
431 positive specimens are displayed as having a value of 0 for presentation purposes on a
SC
432 logarithmic scale.
433 Figure 11. Map of ntr6TV(K80STOP) positivity rates (%) among T. vaginalis-positive
U
434 specimens.
AN
435 Figure 12. State-by-state comparison of ntr6TV(K80STOP) positivity rates. Colored spaces
M
436 indicate comparisons where a significant different (P < 0.05) between states was observed. The
437 color of the space corresponds to the state with the greater positivity rate (blue for the state listed
D
438 on the X-axis and red for the state listed on the Y-axis).
TE
439
EP
440
C
441
AC
20
ACCEPTED MANUSCRIPT
State Positive Specimens (n) Negative Specimens (n) Total Specimens (n) Positivity Rate (%) 95% Confidence Interval (%)
MS 577 5,861 6,438 9.0 8.3 - 9.7
WI 77 907 984 7.8 6.1 - 9.5
SC 147 1,852 1,999 7.4 6.3 - 8.5
IL 3,678 48,839 52,517 7.0 6.8 - 7.2
AR 301 4,197 4,498 6.7 6 - 7.4
AL 2,123 35,343 37,466 5.7 5.5 - 5.9
PT
LA 3,191 52,880 56,071 5.7 5.5 - 5.9
KS 725 12,484 13,209 5.5 5.1 - 5.9
GA 5,246 93,978 99,224 5.3 5.2 - 5.4
DC 321 5,861 6,182 5.2 4.6 - 5.8
RI
TN 1,558 28,336 29,894 5.2 4.9 - 5.5
MO 1,306 25,499 26,805 4.9 4.6 - 5.2
DE 731 14,751 15,482 4.7 4.4 - 5.0
NC 510 10,818 11,328 4.5 4.1 - 4.9
SC
OH 1,642 35,887 37,529 4.4 4.2 - 4.6
IN 420 9,781 10,201 4.1 3.7 - 4.5
KY 2,094 48,767 50,861 4.1 3.9 - 4.3
OK 879 20,520 21,399 4.1 3.8 - 4.4
MI 1,310 31,630 32,940 4.0 3.8 - 4.2
U
AK 2 50 52 3.8 0.0 - 9.0
NE 55 1,376 1,431 3.8 2.8 - 4.8
SD 3 77 80 3.8 0.0 - 8
WY
MD
TX
9
1,889
10,596
233
50,838
292,725
AN 242
52,727
303,321
3.7
3.6
3.5
1.3 - 6.1
3.4 - 3.8
3.4 - 3.6
PA 4,619 129,516 134,135 3.4 3.3 - 3.5
FL 6,853 216,096 222,949 3.1 3 - 3.2
M
VA 1,036 32,234 33,270 3.1 2.9 - 3.3
VT 7 222 229 3.1 0.9 - 5.3
NJ 3,628 122,747 126,375 2.9 2.8 - 3.0
WA 98 3,402 3,500 2.8 2.3 - 3.3
D
State Positive Specimens (n) Negative Specimens (n) Total Specimens (n) Positivity Rate (%) 95% Confidence Interval (%)
ME 5 73 78 6.4 1 - 11.8
SD 5 104 109 4.6 0.7 - 8.5
LA 3,212 70,666 73,878 4.3 4.2 - 4.4
MS 599 13,382 13,981 4.3 4 - 4.6
AL 1,702 40,820 42,522 4 3.8 - 4.2
WI 96 2,299 2,395 4 3.2 - 4.8
PT
KS 725 17,744 18,469 3.9 3.6 - 4.2
IL 2,286 59,414 61,700 3.7 3.6 - 3.8
OK 913 24,426 25,339 3.6 3.4 - 3.8
SC 84 2,278 2,362 3.6 2.9 - 4.3
MO 1,332 37,270 38,602 3.5 3.3 - 3.7
RI
TX 13,334 365,520 378,854 3.5 3.4 - 3.6
GA 4,096 115,805 119,901 3.4 3.3 - 3.5
NE 60 1,753 1,813 3.3 2.5 - 4.1
IN 539 16,598 17,137 3.1 2.8 - 3.4
SC
TN 1,218 38,437 39,655 3.1 2.9 - 3.3
IA 129 4,198 4,327 3 2.5 - 3.5
AR 174 5,767 5,941 2.9 2.5 - 3.3
NM 89 3,072 3,161 2.8 2.2 - 3.4
NV 488 17,094 17,582 2.8 2.6 - 3
U
AZ 3,040 107,972 111,012 2.7 2.6 - 2.8
KY 2,150 77,113 79,263 2.7 2.6 - 2.8
OH 1,738 61,908 63,646 2.7 2.6 - 2.8
OR
ID
CO
NC
62
85
335
483
2,238
3,182
12,845
18,483
AN 2,300
3,267
13,180
18,966
2.7
2.6
2.5
2.5
2.0 - 3.4
2.1 - 3.1
2.2 - 2.8
2.3 - 2.7
UT 52 2,036 2,088 2.5 1.8 - 3.2
M
VA 1,006 39,919 40,925 2.5 2.3 - 2.7
MI 1,054 42,149 43,203 2.4 2.3 - 2.5
MT 4 165 169 2.4 0.1 - 4.7
WA 98 4,358 4,456 2.2 1.8 - 2.6
WY 6 272 278 2.2 0.5 - 3.9
D
AK 0 64 64 0 0-0
ND 0 9 9 0 0-0
Total 58,891 1,496,075 1,554,966 2.8 2.8 - 2.8
ACCEPTED MANUSCRIPT
PT
WV 3 51 54 5.6 0.0 - 11.7
CA 20 358 378 5.3 3.0 - 7.6
NY 5 111 116 4.3 0.6 - 8.0
IA 2 46 48 4.2 0.0 - 9.9
ID 1 23 24 4.2 0.0 - 12.2
RI
OK 37 842 879 4.2 2.9 - 5.5
OR 1 23 24 4.2 0.0 - 12.2
AR 12 289 301 4.0 1.8 - 6.2
SC
RI 1 27 28 3.6 0.0 - 10.5
UT 1 27 28 3.6 0.0 - 10.5
KY 56 2038 2,094 2.7 2.0 - 3.4
AZ 41 1824 1,865 2.2 1.5 - 2.9
AL 42 2081 2,123 2.0 1.4 - 2.6
U
KS 13 712 725 1.8 0.8 - 2.8
LA 56 3135 3,191 1.8 1.3 - 2.3
NE 1 54 55 1.8 0.0 - 5.3
IL
TX
FL
62
177
100
3616
10419
6753
AN 3,678
10,596
6,853
1.7
1.7
1.5
1.3 - 2.1
1.5 - 1.9
1.2 - 1.8
TN 24 1534 1,558 1.5 0.9 - 2.1
M
IN 6 414 420 1.4 0.3 - 2.5
MS 8 569 577 1.4 0.4 - 2.4
SC 2 145 147 1.4 0.0 - 3.3
CT 1 83 84 1.2 0.0 - 3.5
D
AK 0 2 2 0.0 0-0
CO 0 173 173 0.0 0-0
MA 0 64 64 0.0 0-0
AC
ME 0 1 1 0.0 0-0
MN 0 6 6 0.0 0-0
NH 0 1 1 0.0 0-0
SD 0 3 3 0.0 0-0
WA 0 98 98 0.0 0-0
WI 0 77 77 0.0 0-0
WY 0 9 9 0.0 0-0
HI 0 0 0 NA NA
MT 0 0 0 NA NA
ND 0 0 0 NA NA
Total 879 58012 58,891 1.5 1.4 - 1.6
ACCEPTED MANUSCRIPT
Figure 1.
Total Tested
Total Positive
5 8
PT
Positivity Rate
4
# Samples (Log)
% Positive
RI
3
4
2
SC
2
1
U
0 0
10 20 30 40 50 60 70 80
Age AN
M
D
TE
C EP
AC
ACCEPTED MANUSCRIPT
Figure 2
PT
RI
U SC
AN
M
D
TE
EP
C
AC
ACCEPTED MANUSCRIPT
Figure 3.
Total Tested
Total Positive
5 Positivity Rate 10
PT
4 8
# Samples (Log)
% Positive
3 6
RI
2 4
SC
1 2
0 0
10 20 30 40 50 60 70 80
U
Age
AN
M
D
TE
C EP
AC
ACCEPTED MANUSCRIPT
Figure 4
PT
RI
U SC
AN
M
D
TE
EP
C
AC
ACCEPTED MANUSCRIPT
Figure 5.
PT
RI
U SC
AN
M
D
TE
C EP
AC
ACCEPTED MANUSCRIPT
Figure 6.
PT
RI
U SC
AN
M
D
TE
EP
C
AC
ACCEPTED MANUSCRIPT
Figure 7.
PT
RI
U SC
AN
M
D
TE
C EP
AC
ACCEPTED MANUSCRIPT
Figure 8.
PT
RI
U SC
AN
M
D
TE
EP
C
AC
ACCEPTED MANUSCRIPT
Figure 9.
10 r2 = 0.29
T. vaginalis (% Positivity)
9 P < 0.0001
PT
8
7
6
RI
5
4
3
SC
2
1
0
0 1 2 3 4 5 6 7 8
U
C. trachomatis (% Positivity)
AN
M
D
TE
C EP
AC
ACCEPTED MANUSCRIPT
Figure 10.
ntr6TV (K80STOP)
PT
Positivity Rate
Total Tested
4 8
Total Positive
RI
# Samples (Log)
3 6
% Positive
SC
2 4
1 2
U
0
10 20 30 40 50 60
AN
70 80
0
Age
M
D
TE
C EP
AC
ACCEPTED MANUSCRIPT
Figure 11.
PT
RI
U SC
AN
M
D
TE
C EP
AC
ACCEPTED MANUSCRIPT
Figure 12
PT
RI
U SC
AN
M
D
TE
EP
C
AC