Inflammatory bowel disease can reduce ovarian reserve function in women: A systematic review and meta-analysis

Abstract

Purpose: 

We conducted a systematic review and meta-analysis to examine the role of inflammatory bowel disease (IBD) in ovarian reserve functions.

Methods: 

The PECO strategy was employed. Women of reproductive age (Population) and with IBD (Exposure) were compared with healthy women of reproductive age (Comparison) to evaluate the ovarian reserve function (Outcome). Two reviewers searched three databases as well as relevant gray literature. After following the PRISMA 2020 guidelines, RevMan 5.0 software and Newcastle–Ottawa Scale (NOS) scoring were used to analyze and summarize the data included in the studies. The protocol was registered on PROSPERO (CRD42021267804).

Results: 

The search yielded 367 studies, out of which 13 were selected for full-text evaluation, and finally, seven studies were included in our research. An analysis of ovarian reserve function in IBD women of reproductive age and healthy women revealed that the ovarian reserve function was lower in IBD women of reproductive age than in healthy women (P < .01, I² = 81%); the ovarian reserve function was significantly lower in women with IBD in remission than in healthy women (P < .01, I² = 0%), and ovarian reserve function was lower in IBD women of reproductive age taking thalidomide than in healthy women (P < .01, I² = 18%).

Conclusion: 

IBD could reduce ovarian reserve function in women of reproductive age, and patients should plan for conception as soon as possible under permissible conditions.

1. Introduction

Inflammatory bowel disease (IBD), including Crohn’s disease (CD) and ulcerative colitis, is a chronic and recurrent autoimmune disease characterized by chronic intestinal inflammation. The increasing incidence and prevalence of IBD have made it a global disease. The peak age of onset is 20 to 40 years old, and about half of patients are diagnosed with IBD before 32.1 years old, which characterizes the best reproductive years.^[1] Therefore, most IBD patients face IBD-related fertility problems.

The incidence and severity of IBD are reported to be higher in women than in men.^[2] Women who want to get pregnant are often very concerned about the effects of illness on pregnancy and childbirth.^[3] According to the latest consensus from the European Crohn’s and Colitis Organisation (ECCO), the decline in fertility is particularly pronounced in women having active IBD.^[4] IBD causes inflammation of the fallopian tubes, ovaries, and perianal disease that causes dyspareunia, leading to reduced fertility, as well as tubal adhesion caused by surgical treatment.^[5–9] Some authors believe that patients are afraid of the heritability of IBD, congenital fetal abnormalities, and teratogenicity of drugs, so they voluntarily delay pregnancy or do not have children.^[10,11] Little information is available on the factors that influence IBD patients’ decision to become pregnant, but lower pregnancy rates in celiac patients may reflect a lack or delay in conceiving rather than infertility. However, delayed delivery is associated with age-related changes in ovarian reserve and subsequent infertility in women with IBD.

The assessment of fertility in young women with IBD is controversial.^[10] Some studies suggest that IBD is associated with low fertility,^[12] while others state that fertility in such patients is similar to or close to that of the general population.^[11,13] Considerable attention should be paid to checking whether the ovarian reserve function of IBD patients has diminished or not.

There are no studies to date on various changes in ovarian reserve in young women with IBD. Ovarian reserve is determined by the number of primary follicles in the ovarian cortex, which tend to undergo aperiodic and periodic recruitment, resulting in monthly ovulation by a selected dominant follicle. It is also an indicator of female fertility and can be assessed by age, follicle-stimulating hormone (FSH), luteinizing hormone, estradiol, statin B, anti-Mullerian hormone (AMH) concentrations, ovarian volume, antral follicle count (AFC), and ovarian biopsy.^[14] It is shown that AFC and AMH are good predictors of ovarian reserve.^[15] AMH is a member of the transforming growth factor-β (TGF-β) family and is secreted by pre-sinus and early-sinus follicles,^[16] and its level reflects the size of the original follicular pool. In mature women, AMH levels decrease over time due to a reduction in the primordial follicular pool with advancing age.^[17]

To date, data on serum AMH levels in women of reproductive age with IBD are limited, with inconsistent results. Additionally, the risk factors associated with reduced ovarian reserve remain unclear.^[12,18,19] This study aimed to assess the changes in ovarian reserve function in women of reproductive age with IBD and to analyze risk factors for reduced ovarian reserve function.

2. Materials and methods

2.1. Registration protocol and study design

This systematic review was registered on PROSPERO under registration number CRD42021267804 and was organized based on PRISMA 2020 guidelines (Preferred Reporting Items for Systematic Reviews and Meta-analyses).

2.2. Focused questions

All patients were women of reproductive age, and the intervention group (or risk factors) were women with IBD (inflammatory bowel disease), C for healthy women without IBD, and O for ovarian reserve function. The PECO format: Participants (P), Exposure (E), comparisons (C), and Outcomes (O) were used to formulate the focused question: “Is ovarian reserve function reduced in women of reproductive age with IBD when compared with healthy women?”

2.3. Eligibility criteria

1.2.3. Search strategy.

We searched three different electronic databases (Embase, PubMed, and Web of Science). The search period was from the establishment of the database to January 2022. We used the following keyword combinations:(1) Indexes related to ovarian reserve function: ① (anti-Mullerian hormone) OR (AMH); ② (ovarian reserve) OR (ovarian function); ③ (antral follicle count) OR (AFC); ④ (follicular stimulating hormone) OR (FSH); ⑤ (Estradiol) OR (E2); ⑥ (Inhibin-B) OR (INH-B); ⑦ (Serum gonadotropin); ⑧ (premature ovarian insufficiency) OR (POI); ⑨ (Decreasing ovarian reservation) OR (DOR); ⑩ (Premature Ovarian Failure) OR (POF); ⑪ Ovarian volume. And (2) (inflammatory bowel disease) OR (Crohn’s disease) OR (ulcerative colitis). Search keywords are entered in all combinations of (1) and (2); all databases followed a similar strategy. Furthermore, gray literature (Google Scholar) and manual searches on the references of the included studies were also evaluated. In all the cases, no restrictions on the publication date and languages were applied.

2.2.3. Inclusion criteria and study selection.

We only included observational studies in which the ovarian reserve function in women with IBD at reproductive age was compared to that of healthy women. The aim was to assess whether the ovarian reserve function was affected in women with IBD. Literature reviews, letters to Editors, and reference guides were excluded. We also ruled out animal experiments and few studies involving men or adolescent women. Criteria to assess ovarian reserve: AMH; AFC; FSH; E2; INH-B; Ovarian volume.

3.2.3. Articles selection and data extraction.

Two researchers separately conducted the database searches and independently reviewed all titles and abstracts using Excel. Studies that did not meet the eligibility criteria were excluded. Then, from the selected abstracts, the same investigators evaluated the full manuscripts based on these same criteria. The senior authors made the final judgment when a consensus could not be reached between the two reviewers.

The data were collected independently by the same reviewers, and the information was then cross-checked. The collected information included the authors, country, cohort design, year of publication, sample size, years, inclusion and exclusion criteria, IBD diagnosis, period of illness, and indicators of ovarian reserve function. In cases of incomplete or unclear information, the authors were contacted by email.

4.2.3. Methodological quality assessment.

The quality of the literature in retrospective case-control studies was assessed using the Newcastle-Ottawa Scale (NOS). The highest score for each study was 9; the exposure assessment and subject selection (4 points), comparability between study groups (2 points), and the adequacy of measurement and recording of study results and follow-up (3 points). Studies with a score of 7 to 9 were considered of high quality and at low risk of bias. Each study was rated independently by two reviewers. If any differences occurred in the results, a consensus was reached through consultation with a third investigator.

2.4. Data analysis

We analyzed the relationship between IBD and ovarian reserve decline using RevMan Software (5.4), while the Random-effects model was used in the meta-analysis. The summary of the effect measures was depicted in a forest plot containing I² and P values. The publication bias was quantitatively evaluated by funnel plots to identify and avoid asymmetry in the selected studies. According to the heterogeneity of methodology, IBD patients were then divided into active and in remission and whether they took thalidomide or not for subgroup analysis.

3. Results

3.1. Study selection and characteristics

The flowchart diagram of this review is outlined in Figure 1. After searching all databases, a total of 364 reports were initially identified. After removing duplicate articles, 141 studies had their titles and abstracts read and 13 potential references were appraised. Of these 13 articles, six studies were excluded for some reasons like studies on animals (n = 3), men (n = 2) and adolescent women (n = 1). The other three articles were found through a manual search, but they were excluded after reading the full text. The characteristics and results of the included studies are presented in Table 1.

Table 1 - Characteristics of studies.

Study	Country	Cohort design	Year of publication	Sample size	Years	Inclusion criteria	Exclusion criteria	Diagnosis of IBD	Period of illness	Indicators of ovarian reserve function
Kadirogullari	Turkey	Prospective case-controlled study	2021	145	18–40	Aged between 18–40 years, regular menstrual periods; no gynecological pathologies detected by gynecological or ultrasound examination; no history of infertility	History of ovarian surgery; PCOS;renal failure (serum creatinine levels > 1.2 mg/dL); suspected malignancy, hereditary or acquired hematologic disease; severe comorbid chronic diseases; abnormal thyroid function tests; serious psychological problems; alcohol abuse; women with first degree relatives diagnosed with inflammatory bowel disease (IBD); positive surgical histories	The diagnosis of CD confirmed with endoscopic;radiologic and histopathological findings	Active and in remission	AMH; Right ovarian volume; Left ovarian volume; Right ovarian AFC; Left ovarian AFC
Fréour	France	Retrospective case-control study	2011	213	2041	Reproductive age; both ovaries present; No previous exposure to cytotoxic drugs or pelvic radiation therapy; no known systemic, endocrine, or metabolic disease; Normal ovarian status according to basal and dynamic evaluations	Women with polycystic ovarian syndrome (PCOS) according to the Rotterdam criteria20 were not included	Clinical, radiological, endoscopic, and histological features	In remission.	AMH
Şenateş	Turkey	Observational, cross-sectional case–control study	2013	70	<40	Women with no known diseases; Had never received any drug or transfusion treatment; No inflammatory bowel disease diagnoses in any of their first-degree relatives; no surgical history and not be taking any	Having a previous ovarian resection; renal failure (serum creatinine levels > 1.2 mg/dL); A diagnosis or suspicion of malignancy; the presence of hereditary or acquired hematologic disease; pregnancy, current lactation, the presence of a serious comorbid chronic illness; chronic liver disease-induced cirrhosis; abnormal thyroid function tests; the presence of a known serious psychological problem; alcoholism and male gender	Endoscopic, radiologic and histopathologic findings	Active and in remission	AMH
Mao	China	Case-control crosssectional study 18–45	2020	117	18–45	Non-pregnancy and not taking oral contraceptive	Oral contraceptive taking history; a gynecological and endocrine relevant surgical history; cytotoxic drugs history; pelvic radiotherapy history; diagnosed or potential polycystic ovary syndrome; thyroid disease; hyperprolactinemia or other relevant	Clinical manifestations and endoscopic, radiologic, and histopathological findings	Active and in remission	AMH
Henes	Germany	Prospective case-control study	2015	60	18–40		Patients who had received previous systemic cytotoxic treatment, especially CYC, as well as patients who had undergone ovarian surgery or radiation, were excluded			AMH
Peng	China	Prospective case-control study	2017	151	18–40	Healthy subjects were selected from healthy women; nonpregnancy, not taking birth control pills, and in similar age to the patient’s group	Abnormal menstrual cycle or menstrual flow, ovarian function failure, the unrelated reproductive system diseases such as endocrine disorder, hysterectomy, pelvic radiotherapy, ovariectomy, polycystic ovary syndrome, endometriosis, ovarian granulosa cell cancer, taking birth control pills, and pregnancy 3 months before selection	Clinic symptoms, physical condition, endoscopy, radiology, and pathology		AMH
Rekawek	America	Retrospective cohort study	2018	340	–	–	–	–	–	AMH

AFC = antral follicle count, AMH = anti-muller’s hormone, CD = celiac disease, PCOS = PCOS polycystic ovary syndrome.

Figure 1.: The flowchart diagram of this review.

3.2. Results for individual studies

In the study of Kadirogullari et al,^[16] serum AMH levels were significantly reduced in patients with CD, especially in patients with active CD. There was no significant correlation between serum AMH level and surgical resection history. Although the serum AMH level of smokers was lower than that of nonsmokers, the difference was not statistically significant. There was no significant difference in serum AMH levels between patients who received the drug and those who were devoid of medications (no thalidomide). The duration of the disease was negatively correlated with serum AHM levels.

In a study by Freour et al,^[12] although there was no significant difference in AMH between CD patients in remission and normal women, age was independently associated with a reduced risk of ovarian reserve changes after the regression analysis. Serum AMH level of female CD patients ≥ 30 years old was significantly lower than that of the control group. Additionally, the colonic disease was independently associated with a higher risk of ovarian reserve changes. The authors also noted that changes in ovarian reserve function were independent of disease duration. In the second article, there was no significant difference in serum AMH levels between smokers and nonsmokers.

Regression analysis conducted in a study by Senates et al^[18] showed that the level of the CD activity index (CDAI) was negatively correlated with serum AMH levels. Even in the absence of organic changes in the reproductive system, the fertility of female patients with CD was significantly lower than that of normal healthy women, especially during the active disease phase. The manuscript compared serum AMH levels in patients with and without surgical resection and found no significant difference between them.

In Zhao et al study,^[20] the data were analyzed according to the activity degree of the disease, the behavioral characteristics of the disease, the presence of perianal disease, the location of the disease, and medication history. Analysis of risk factors associated with ovarian reserve deficiency: age > 25 years; lasting for > 6 years; stenosis (type B2); perianal disease; active disease status, and use of thalidomide. It was also suggested that lesion location and behavior were not associated with insufficient ovarian reserve function.

The conclusion of Henes et al study^[21] was that the AMH level of IBD patients was significantly reduced. Although some patients had a severe reduction in AMH levels, 50% of IBD patients had reduced AMH levels (<1 ng/mL) when compared with normal women. Since other autoimmune diseases such as rheumatoid arthritis (RA) and spinal arthritis (SpA) were included in the article, further subgroup analysis was not performed.

In the study by Peng et al,^[19] Crohn’s Disease Endoscopic Index of Severity (CDEIS) and thalidomide usage were considered independent risk factors for decreased ovarian reserve. Simultaneously, this manuscript also used 30 years as the age demarcation line for subgroup comparison, as AMH levels decreased significantly after age > 30 years.

In Rekawek et al study,^[22] IBD was not associated with impaired ovarian reserve function. In addition, although CD and ulcerative colitis have different inflammatory pathways, these IBD subtypes do not differ in their impact on ovarian reserve function. Future studies should examine endometrial receptivity in patients with IBD as well as the correlation between the levels of inflammatory markers and the effect on ovarian reserve function.

3.3. Bias of risk and methodological quality assessments

In all the studies that were included, only one study was in the form of a poster with some incomplete content, so the NOS score was 6, while the scores of the other studies were all ≥ 7 (Table 2).

Table 2 - Newcastle–Ottawa scale scores and quality assessment of included studies.

Study	Selection					Outcome
	Representativeness	Selection	Ascertainment	Outcome	Comparability	Assessment	Follow-up	Adequacy	Quality*
Fréour[22]	*	*	*	*	*	*	-	*	Good (7)
Şenateş[8]	*	*	*	*	*	*	-	*	Good (7)
Henes[10]	*	*	*	*	*	*	-	*	Good (7)
Rekawek[23]	*	*	*	*	*	*	-	-	Good (6)
Peng[18]	*	*	*	*	*	*	*	*	Good (8)
Kadirogullari[19]	*	*	*	*	*	*	*	*	Good (8)
Zhao[20]	*	*	*	*	*	*	-	*	Good (7)

3.4. Meta-analysis results

The AMH level of IBD women of reproductive age was significantly lower than that of healthy women (P < .01). Since the heterogeneity among all the studies was significant (I² = 81%), random-effects models were preferred (Fig. 2A). There was no obvious publication bias in the study (Fig. 2B). As few studies did not clearly distinguish patients in active or remission states, three articles that explicitly recruited patients in remission were selected for subgroup analysis. The AMH level of IBD women of reproductive age in remission was lower than that of healthy women (P < .01, Fig. 3A). There was no obvious publication bias in that study (Fig. 3B). Two studies specifically described the effect of thalidomide on ovarian reserve function and were analyzed as a subgroup. The ovarian reserve function was significantly lower in reproduction-age IBD women taking thalidomide than in healthy women (Fig. 4A) without any obvious publication bias (Fig. 4B).

Figure 2.: (A) The forest plot of comparison of ovarian reserve function between IBD women of reproductive age and healthy women. (B) The Funnel plot of comparison of ovarian reserve function between IBD women of reproductive age and healthy women. IBD = inflammatory bowel disease.

Figure 3.: (A) The forest plot of comparison of ovarian reserve function between IBD women in remission of reproductive age and healthy women. (B) The Funnel plot of comparison of ovarian reserve function between IBD women in remission of reproductive age and healthy women. IBD = inflammatory bowel disease.

Figure 4.: (A) The forest plot of comparison of ovarian reserve function between IBD women taking thalidomide of reproductive age and healthy women. (B) The Funnel plot of comparison of ovarian reserve function between IBD women taking thalidomide of reproductive age and healthy women. IBD = inflammatory bowel disease.

4. Discussion

In this meta-analysis, we found that the AMH level of IBD women of reproductive age was significantly lower than that of healthy women. Furthermore, the AMH level of IBD women of reproductive age in remission was lower than that of healthy women and the ovarian reserve function was significantly lower in reproduction-age IBD women taking thalidomide.

Despite few studies analyzing fertility in women with IBD, a thorough evaluation of fertility has not been widely assessed to date.^[5,23] Although some studies have assessed the reproductive potential of women with IBD, the conclusions remain inconsistent.^[10,11,24]

Several studies have shown that IBD patients’ disease activity, bowel resection procedures, and perianal lesions are associated with reduced female fertility.^[1,25,26] Disease activity mainly affects the fertility of female IBD patients by the following mechanisms: Firstly, active intestinal inflammation produces an adverse effect on reproductive organs in the pelvic cavity. IBD transmural inflammation of the intestinal tract may cause local inflammation of the oviduct and the ovaries. Active perianal lesions can directly affect sexual feelings, thus, reducing fertility.^[26] Secondly, the systemic inflammatory state caused by IBD can also affect ovarian reserve function and reduce fertility in women.^[5] Additionally, another study suggested that active disease can also lead to depression, malnutrition, and anemia, which can affect the fertility of women with IBD.^[27]

It has been reported that there is no significant difference in ovarian function or fertility between IBD women in remission and healthy women.^[10,13] However, in this study, the AMH of women with IBD in remission was compared with that of healthy women, which proved that the ovarian reserve function of women with IBD in remission was decreased. However, the reasons for the decline in ovarian reserve function in women with IBD in remission need further evaluation.

Thalidomide, as an immunosuppressant, has a strong regulatory effect on cytokine activation and cell-mediated immune response and can affect early apoptosis and proliferation of T lymphocytes. It also inhibits the cascade reactions of other cytokines, such as cyclooxygenase-2, IL1-β, TGF-β, and IL-6, through its effect on TNF-α.^[28] Few studies suggested^[19,20] that thalidomide usage caused a reduction in ovarian reserve function when taken for one month and eight weeks, respectively. The immunosuppressive effects of thalidomide may also inhibit TGF-α, TGF-β, or interleukin-1 β functions, which are critical for the growth and development of primordial follicles and oocytes.^[29]

Apart from confirming that the remission of disease and thalidomide use was closely associated with reduced ovarian reserve function, several other factors that might affect ovarian reserve function were also mentioned in the included articles. Although the in-depth analysis was not conducted due to various reasons, it provided some directions for our future clinical research.

In a study by Kadirogullari et al,^[16] there was a significant negative correlation between the duration of the disease and serum AMH levels. Furthermore, Freour et al study^[12] suggested that the disease duration was not associated with changes in ovarian reserve function, whereas Zhao et al study^[20] suggested that >6 years of disease was an independent risk factor for ovarian reserve decline. However, due to different cutoff points and incomplete original data, integration analysis could not be conducted. The fourth article was a retrospective analysis and was not suitable for demonstrating that AMH concentration decreased more rapidly with the duration of the disease. A longer follow-up is needed to make any conclusive finding.

The studies by Kadirogullari et al^[16] and Freour et al^[12] mentioned that there was no difference in AMH levels between smokers and nonsmokers. On the contrary, it is shown that smoking is associated with longer exacerbations, excessive complications, greater need for steroids, and postoperative recurrence.^[30] Further studies in larger populations are needed to identify potential tobacco-induced changes in ovarian reserve in women with IBD.

The studies by Kadirogullari et al^[16] and Senates et al^[18] suggested that pelvic surgery is not significantly associated with ovarian reserve function, whereas another study has shown an association between surgical intervention and reduced fertility in IBD patients.^[10] Reduced fertility can also result from surgical changes in pelvic anatomy that impede ovum excretion, sperm-ovum bonding, and fertilized ovum transport, especially after ileal pouch-anal anastomosis, without any relation to ovarian reserve function.

Freour et al study^[12] suggested that colonic disease was independently associated with a higher risk of ovarian reserve alteration. Since the distance between the colon and the pelvis is less, the colonic lesions are more likely to cause a higher degree of chronic pelvic inflammation than other sites. A recent study^[31] suggested that endometriosis is also a chronic inflammatory condition in which serum AMH levels are lower than in normal controls, and the severity of endometriosis is associated with changes in ovarian reserve. IBD is characterized by chronic intestinal inflammation, which releases large amounts of pro-inflammatory mediators into the bloodstream. Recent evidence indicates that IBD and intestinal endometriosis can coexist.^[32] Chronic low-grade inflammation might be one of the mechanisms affecting fertility in these patients.^[33]

AMH is a member of the transforming growth factor-β (TGF-β) family and is produced by granulosa cells in pre-sinus and small sinusoid follicles.^[34] Since AMH remains stable throughout the menstrual cycle, it has replaced traditional FSH, luteinizing hormone, and ovarian estrogen tests in assisted reproductive technology and chemotherapy for malignant tumors and rheumatic diseases such as systemic lupus erythematosus.^[35,36] AMH has also been successfully used to assess pathological or therapeutic drug damage to the ovaries.^[37] Hence, it is feasible to use AMH for assessing ovarian reserve function in women with IBD.^[12,17]

It is suggested that a woman’s fertility declines significantly after the age of 30 due to the loss of follicles in her ovaries.^[38] This follicle loss can be accelerated in some cases, leading to premature changes in ovarian reserve.^[12] In mature women, the number of follicles in the primordial follicular pool decreases with age, while AMH levels decrease over time.^[16] As age is an important factor affecting AMH levels, some articles were grouped according to age, but due to inconsistent grouping criteria and incomplete data, so the subgroup analysis was not carried out in-depth.

Our study had few limitations, especially the relatively small number of patients, and the use of only one ovarian reserve parameter displayed heterogeneous results due to the differences in AMH reagents. Previous studies indicated that the results were heterogeneous due to the differences in AMH reagents.^[39] A study by Mont’Alverne et al^[40] ruled out this possibility by comparing two different kits so that the detection results of AMH were homogeneous. Another study limitation was that the cutoff limit of AMH is not indicated in some articles, which also affected the homogeneity of the results to a certain extent.

Fertility is an important issue for patients with IBD. As long as their disease course and medications allow pregnancy, these patients should not wait too long to conceive. Since IBD is associated with reduced ovarian reserve function, ovarian reserve function is affected even in remission. Women of reproductive age with IBD have lower ovarian reserve when compared to healthy ones. It is recommended that IBD patients should consider pregnancy during their 20s. We believe that the results of this study will be beneficial for dispensing clinical treatment, especially for women of reproductive age who have recently been diagnosed with IBD.

Author contributions

Conceptualization: Xuewen Shi.

Data curation: Lin Guo, Yi Liu, Jiansheng Li.

Supervision: Xuewen Shi.

Writing – original draft: Lin Guo, Yi Liu, Quan Liu, Bing Liu.