Subfertility and early menopause in women with glomerular disease

Subfertility, an inability to achieve desired conception for an extended period, affects up to one in six couples and leads to considerable psychological distress [1]. Due to an altered hypothalamic–pituitary–ovarian axis, women with advanced chronic kidney disease (CKD) are more likely to experience menstrual irregularities, subfertility and early menopause [2–6]. Autoimmune diseases, including systemic lupus erythematosus and antiphospholipid antibody syndrome, are associated with subfertility related to autoantibodies, autoimmune oophoritis and medical treatment [7, 8].

Given the immune-mediated origins of glomerular disease and its immunosuppressive treatment, subfertility may be common in this population. Beyond the gonadotoxicity of cyclophosphamide, limited research exists regarding subfertility in women with glomerular disease. More robust knowledge could help nephrologists counsel women regarding risks for subfertility and early menopause, encourage conception during disease quiescence and consider earlier referral to fertility specialists.

A women’s health survey was distributed via e-mail, mail and in person to women aged 18–65 years enrolled in the Glomerular Disease Collaborative Network, a longitudinal research registry in the southeastern USA. We used descriptive statistics to assess responses and the Chi-squared tests to assess cyclophosphamide use and disease type with subfertility. Subfertility was ascertained by asking ‘Have you ever tried to become pregnant for more than 6 months without success?’ and menopause by asking ‘Have your periods (menses) stopped permanently?’ Follow-up questions for menopause included age, reason for menopause, associated symptoms and whether menopause was diagnosed by a physician. Early menopause was defined as occurring before the age of 45 years.

The survey was completed by 189 women (with a response rate of 16%). Median [interquartile range (IQR)] age was 48 (34–57) years and median (IQR) body mass index was 28 (24–36) kg/m². Most were Caucasian (74%, n/N = 139/189), while 19% were African American (n/N = 35/189); 76% held a college degree. The most common diagnoses were IgA nephropathy/vasculitis (22%), lupus nephritis (20%), anti-neutrophil cytoplasmic antibody (ANCA) vasculitis (20%) and focal segmental glomerulosclerosis (FSGS) (17%). Over half of them reported hypertension (57%) and 10% had diabetes. At the time of response, participants had known disease for a median (IQR) of 6.7  (3.7–12.6) years and 7% (n/N = 13/189) required dialysis.

Table 1 describes self-reported menstrual cycle length and menopause by age group. In those without menopause, nearly one-third (n/N = 31/96, 32%) reported abnormal menstrual cycles (<24 days, >35 days or too irregular to say). Among women aged >45 years who reported permanent cessation of menses, 41% (n/N = 35/86) reported early menopause. Reasons cited were natural ending of menses (n/N = 33/93, 35%), oophorectomy or hysterectomy (n/N = 27/93, 29%), radiation/chemotherapy or medication-induced (n/N = 19/93, 20%), endometrial ablation (n/N = 8/93, 9%) or other (n/N = 6/93, 6%). Associated symptoms included hot flashes (n/N = 59/93, 63%) and vaginal dryness (n/N = 44/93, 47%), and 33% (n/N = 31/93) reported physician diagnosis of menopause.

Table 1.

Menstrual cycle length and cessation of menses by age group in women enrolled in the glomerular disease collaborative network

Age group (years)	18–25	26–35	36–45	46–55	56–65
	n = 10	n = 39	n = 32	n = 49	n = 59
Menstrual cycle lasting for 24–35 days, %	30	67	69	27	0
Abnormal menstrual cycle,a %	70	28	19	12	2
Permanent cessation of menses, %	0	5	13	59	98
Cessation of menses occurring at <45 years oldb	NA	NA	NA	62 (18/29)	30 (17/57)

^aAbnormal menstrual cycle defined as lasting <24 days, >35 days or too irregular to say.

^bReported as percentage and n/N of respondents. NA: not applicable.

As therapy for glomerular disease, cyclophosphamide use was reported by one-third of the respondents (n/N = 63/189, 33%), though total cumulative dose was not assessed and was unavailable. Few of these patients consulted with a fertility specialist (n/N = 3/63, 5%). Fertility preservation methods in cyclophosphamide-exposed women included leuprolide (n/N = 6/63, 10%) and oocyte cryopreservation (n/N = 1/63, <2%). Subfertility was reported by 21% (n/N = 13/63) of these women compared with 31% (n/N = 35/114) of women who did not receive cyclophosphamide.

One in four women (n/N = 49/189, 26%) reported subfertility; 31% (n/N = 13/42) with IgA, 29% (n/N = 11/38) with lupus nephritis, 16% (n/N = 6/38) with ANCA and 33% with FSGS (n/N = 11/33). Most women (n/N = 26/49, 53%) reported the reason for their unsuccessful conception as unknown, while 31% (n/N = 15/49) stated it was due to a personal medical problem. Medical support of conception (e.g. clomiphene) was used by 27% (n/N = 13/49), and 16% (n/N = 8/49) utilized intrauterine insemination, in vitro fertilization or received egg/sperm donation.

From our cross-sectional study, women with glomerular disease and/or vasculitis reported high rates of menstrual irregularities, subfertility and early menopause. Neither disease type nor cyclophosphamide therapy was significantly associated with subfertility. While early menopause affects ∼5–10% of the general population [9, 10], we found that 41% of the eligible women reported menopause before the age of 45 years. Because early menopause is associated with both ischemic cardiovascular disease and increased mortality, our findings may have significant clinical implications [10, 11].

While our response rate and sample size were low, this study is the first to our knowledge to assess subfertility and early menopause across glomerular diseases. Women with an affirmative history may have been more responsive, leading to a positive response bias. With these survey data, we could not evaluate the impact of glomerular filtration rate (GFR), which is known to have an inverse relationship with subfertility and anovulation. Recall bias and lack of biochemical confirmation may have inflated the number reporting early menopause. A higher proportion of nonresponders had lupus nephritis, potentially increasing the percentage reporting subfertility and early menopause due to potential autoantibodies (e.g. antiovarian antibodies) and greater cyclophosphamide treatment in this subgroup.

Cyclophosphamide can reduce fertility in both an age- and a dose-dependent manner [6, 12]. Treatment with gonadotropin-releasing hormone (GnRH) agonists during cyclophosphamide therapy may reduce premature ovarian failure in women with lupus [13]. Leuprolide utilization by our respondents was low, and further research is needed to understand contemporary prescribing practices and efficacy of GnRH agonists in women with glomerular disease. Surprisingly, only one in four women with subfertility reported cyclophosphamide exposure, highlighting that other disease- or treatment-related mechanisms are potentially at play. Future prospective studies in women with glomerular disease attempting conception should examine the influence of auto-antibody status, proteinuria and disease activity in addition to GFR, medications and comorbidities. The safety of fertility preservation strategies and assisted reproductive technology, as well as their efficacy in women with CKD and glomerular disease, requires evaluation.

Globally, infertility is rising [14] and a growing trend to delay childbirth has led to increasing numbers of women entering pregnancy with chronic disease [15]. Uniquely, women with glomerular disease are often advised to delay pregnancy until their disease is well-controlled, potentially contributing to older maternal age. Nephrologists must recognize their pivotal role in counseling their patients regarding family planning early and often in their disease course. Prompt referral for effective subfertility treatment could increase the likelihood of a successful pregnancy. Finally, women with glomerular disease should be habitually assessed for early menopause as it may increase their already elevated risk for future cardiovascular disease.

FUNDING

M.L.R. was supported by National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases T32DK007750-20.

AUTHORS’ CONTRIBUTIONS

M.L.R. and V.K.D. were involved in conception/design, analysis and interpretation of data, and drafting and revising of study; conception/design, analysis of data and revising of study were done by C.J.P.; analysis of data and revising were carried out by L.N.B.; conception/design and revising were done by S.L.H. and R.J.F.

DATA AVAILABILITY STATEMENT

The data underlying this article cannot be shared publicly due to the privacy of individuals that participated in the study. The data will be shared on reasonable request to the corresponding author.

CONFLICT OF INTEREST STATEMENT

None declared. The results presented in this article have not been published previously in whole or part, except in abstract format.