Effect of menstrual cycle phase and hormonal treatments on evaluation of tubal patency in baboons

Abstract Background We evaluated whether menstrual cycle phase influences the assessment of tubal patency by hysterosalpingography (HSG) in baboons. Methods Retrospective analysis of baseline tubal patency studies and serum estradiol (E2) and progesterone (P4) values obtained from female baboons used as models for development of non‐surgical permanent contraception in women. The main outcome measure was bilateral tubal patency (BTP) in relationship with estradiol level. Results Female baboons (n = 110) underwent a single (n = 81), two (n = 26), or three (n = 3) HSG examinations. In 33/142 (23%) HSG examinations, one or both tubes showed functional occlusion (FO). The median E2 in studies with BTP (49 pg/mL) was significantly higher than in those studies with FO (32 pg/mL, P = .005). Among 18 animals with repeat examinations where serum E2 changed from <60 to ≥ 60 pg/mL, 13 results changed from FO to BTP (P = .0001). No sets showed a change from BTP to FO with an increase in estradiol. Conclusion In baboons, functional occlusion of the fallopian tube is associated with low estradiol levels, supporting a role for estrogen‐mediated relaxation of the utero‐tubal junction.


| INTRODUCTION
The American College of Radiology recommends performing evaluation of tubal patency by hysterosalpingogram (HSG) during the early proliferative phase of the menstrual cycle. 1 This timing prevents inadvertent disruption of a luteal phase pregnancy and may offer other advantages to imaging. More recently, HSG evaluation has been required as a verification step after transcervical hysteroscopic permanent contraception (PC) procedures. While women undergoing infertility evaluation generally are not using hormonal contraceptives, patients are required to continue using a contraception method after hysteroscopic PC until verification of tubal occlusion.
Functional tubal occlusion (FO) is the term used to describe a falsepositive assessment of tubal occlusion at the utero-tubal junction (UTJ) observed during HSG that is not explained by a true anatomic blockade or the presence of an obstructing material. 2 FO likely occurs due to muscular spasm at the UTJ that may be cycle dependent. 3 Few studies have evaluated whether FO is influenced by menstrual cycle timing or hormonal therapy. Lindhal and Helander performed HSG in 511 women of proven fertility and found the highest rates of tubal occlusion during the follicular (proliferative) phase and recommended the early luteal (secretory) phase as the best stage for tubal evaluation. 4 Menstrual cycle timing and treatments that relax the uterine musculature may also impact the success with non-surgical permanent contraception techniques such as quinacrine sterilization (QS) or hysteroscopic placement of Essure™ microinserts. In a review of QS efficacy, Sokal et al 5 found that pre-treatment with papaverine, a uterine relaxant, was associated with a lower failure rate. Sinha found that secretory phase timing reduced the success of placement of Essure™ microinserts. 6 In support of research evaluating This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2017 The Authors. Journal of Medical Primatology Published by John Wiley & Sons Ltd non-surgical approaches to PC, we performed HSGs in female baboons prior to other transcervical procedures. 7,8 During the course of these studies, we noted differences in the success of our tubal imaging based on cycle timing and hormonal treatment. To explore this hypothesis, we conducted a descriptive retrospective reanalysis of these baseline data to evaluate the association of the hormonal milieu and tubal patency. Here, we present descriptive results from this series, focusing on the correlation of bilateral tubal patency with serum estradiol.

| Animals and treatments
Between October 2012 and November 2016, we completed several studies investigating approaches to non-surgical permanent contraception in women at SNPRC and ONPRC using the baboon model.
Results from some of these studies have previously been reported and others are ongoing. 7,8 We performed one or more baseline hys- As baboons undergo significant sexual skin swelling during natural menstrual cycles, 10

| Hysterosalpingogram procedures
Females underwent a hysterosalpingogram (HSG) procedure to confirm tubal patency as previously described. 8

| Verification of tubal patency
In cases where we could not confirm bilateral tubal patency by HSG, we evaluated patency ex vivo and histologically as previously described. 8

| Data analysis
This study presents descriptive data on tubal patency from a series of examinations carried out to evaluate approaches to non-surgical permanent contraception carried out without an a priori hypothesis or formal power analysis. We report descriptive statistics of tubal patency results. As estradiol concentrations were not normally distributed, we compared the median level of E 2 in examinations with BTP to those with FO using the Mann-Whitney U-test. We stratified the studies according to estradiol concentration (E 2 ≥ 60, E 2 < 60 pg/mL) and compared the proportion of examinations with BTP using Fisher's exact test, and further evaluated this relationship by comparing repeat studies in the same animals using McNemar's test with continuity correction.

| RESULTS
We performed 142 HSG examinations on 110 baboons with single examinations in 81, two in 26, and three examinations in 3 females.
For the 26 females contributing two examinations, 16 had BTP observed in the first study. Among the three contributing three examinations, one had BTP only in the initial study, one only in the second examination, and one in none of the studies. BTP was confirmed in 85 (77%) of the females with the initial HSG, 12 (11%) with the second HSG, and by histologic evaluation in 10 (9%). Baseline tubal patency could not be confirmed in three females, as they received an active treatment following the initial HSG and histology demonstrated bilateral (n = 2) or unilateral (n = 1) occlusion changes consistent with the treatment effect. 8 As all three of these females had a history of a recent term pregnancy, we considered the HSG results consistent with FO.  Test P = .0001). The animal that remained FO participated in a contraceptive study as a control and did not get pregnant over 6 cycles of exposure to a fertile male. 7 Two other females showed FO with E 2 > 60 pg/mL; technical difficulties occurred with both examinations.

| DISCUSSION
In this study, we performed HSG examinations in females during various stages of normal menstrual cycles and following hormonal treatments. Although bilateral patency was seen with most (76.7%) studies, we found a significantly lower median concentration of estradiol in the group with functional occlusion with 90% of these results associated with an estradiol concentration < 60 pg/mL. We also found low estradiol associated with a subjective impression of high uterine pressure, difficult examinations, and extravasation of contrast media into the vascular system. Furthermore, when females shown to have FO in an examination with an estradiol < 60 pg/mL underwent a repeat examination with an estradiol > 60 pg/mL, BTP was typically observed. In a limited group of studies performed in the luteal phase, we observed no independent effect of progesterone; FO occurred only in association with low levels of estradiol in the late luteal phase.
Strengths of our study include the large number of animals studied, and the subset that underwent repeat examinations allowing us to evaluate the effect of different hormonal conditions. Additionally, we verified bilateral tubal patency either by HSG or by histology in all but three of the subjects. However, these 3 each had a history of a recent confirmed pregnancy, strongly suggestive of tubal patency.
Major limitations include the retrospective nature of our data analysis, the lack of strict timing of hormonal treatments, and the absence of a true experimental approach to evaluate estrogen action. As we did not perform examinations at well-controlled time points in the menstrual cycle, we cannot provide information as to the sensitivity and specificity of the HSG examination in baboons in the follicular or luteal phase.
The baboon provides an excellent model for reproductive research as the reproductive tract is similar to women and influenced by the same hormonal events. 11 Unlike most macaque species, the baboon has a cervix with a straight path similar to women that permits routine dilation and intrauterine procedures including hysterosalpingography.
However, a notable species difference in baboons is dramatic perineal sex skin swelling that peaks with ovulation prior to regressing. 10 In an effort to standardize and facilitate our evaluation of animals, we treated some females with DMPA or a GnRH agonist/antagonist prior to the HSG examination. Difficultly confirming tubal patency in these females led us to explore the hypothesis that low concentrations of estradiol may contribute to functional occlusion.
F I G U R E 2 Estradiol (E 2 ) levels at the time of hysterosalpingogram examination in 26 animals that underwent two, and 3 females that underwent 3 repeated studies. Closed (solid black) markers indicates examination with bilateral tubal patency (BTP), and open (white) markers an examination with functional occlusion (FO). An increase in E 2 from < 60 pg/mL to ≥ 60 pg/mL typically resulted in a change from bilateral or unilateral occlusion to bilateral tubal patency. Similarly, a drop in E 2 in this range resulted in a change from bilateral tubal patency to functional occlusion in some animals supporting that the change in patency was not a function of the repeat examination F I G U R E 3 Repeat examinations in 18 females where E2 changed from <60 to ≥ 60 pg/mL. Of these, 13 results changed from functional occlusion (FO) to bilateral tubal patency (BTP). In contrast, no sets showed a change from bilateral patency to occlusion with an increase in estradiol (McNemar's Test P = .0001) T A B L E 2 Estradiol (E2) and progesterone (P) concentrations in females examined in the luteal phase of natural cycles Cycle phase categorized as periovulatory (PO, E 2 > 100 pg/mL, P < 2 ng/mL), early-to-mid luteal (EML, E 2 40-100 pg/mL, P 1-10 ng/mL), or late luteal/ premenstrual (LL, E 2 < 40 pg/mL, P < 2 ng/mL). Functional Occlusion (FO) = unilateral or bilateral tubal occlusion. Bilateral tubal patency (BTP).
Although GnRH agonists/antagonists resulted in the lowest concentrations of estradiol, a higher proportion of females pre-treated with DMPA developed FO. This suggests that a progesterone receptormediated effect contributes to regulation of the UTJ, perhaps through downregulation of estrogen receptor. 12,13 We observed FO in all of the females pre-treated with DMPA >7 day prior to HSG, but BTP with shorter exposure. These results are similar to those observed with progesterone exposure during natural cycles, with FO seen only in the late luteal phase.
Although we observed a correlation between an estradiol concentration of < 60 pg/mL and FO, most HSG examinations performed even with low levels of estradiol showed BTP and three females with In conclusion, our results demonstrate an association between functional occlusion of the fallopian tube and low serum estradiol concentration in baboons, with 90% of cases of FO seen with E 2 < 60 pg/mL.
Furthermore, among females with findings of FO during an examination with E 2 < 60 pg/mL, a repeat HSG examination conducted with E 2 ≥ 60 pg/mL resulted in BTP. Together, these findings suggest that hysterosalpingography or transcervical permanent contraception procedures should be performed during the mid-to-late follicular phase to ensure adequate estradiol exposure to avoid a false diagnosis of functional fallopian tube occlusion or incomplete procedures. Some animals received pre-treatment with a GnRH agonist/antagonist, DMPA, or a COC followed by a hormone-free interval. Functional Occlusion (FO) = unilateral or bilateral tubal occlusion. Bilateral tubal patency (BTP). a Mann-Whitney U-test.