Menstrual Abnormalities and Reproductive Lifespan in Females with CKD

Visual Abstract

Abstract

Background and objectives

Menstrual abnormalities and shortened reproductive lifespan are associated with shorter life expectancy and higher cardiovascular and osteoporosis risk in the general population, although the magnitude of these reproductive factor irregularities in females with CKD is unclear. This systematic review and meta-analysis aimed to summarize the current knowledge regarding menstrual abnormalities and reproductive lifespan among females with CKD.

Design, setting, participants, & measurements

A comprehensive bibliographic search (MEDLINE, Embase, and Cumulative Index to Nursing and Allied Health Literature [CINAHL]) was completed from database inception to February 2022 to identify all original articles reporting on females of reproductive age with nondialysis-dependent/nonkidney transplant CKD, dialysis-dependent CKD, or kidney transplantation and menstruation patterns, age of menarche, and/or menopause. Data extraction and study quality assessment were completed in duplicate. Random effects meta-analyses were used to derive pooled proportions estimates.

Results

Forty-six studies were identified, and 35 were meta-analyzed, stratified by KRT modality and reported outcome. Menstrual abnormalities were present in 19%–47% of patients on hemodialysis and 75% of patients on peritoneal dialysis. Kidney transplantation was associated with a 7%–30% decrease in menstrual abnormalities. Reproductive lifespan was 32 years (95% confidence interval, 30 to 34 years). Although significant heterogeneity was present, study quality ranged from fair to good, and no evidence of publication bias was noted.

Conclusions

Menstrual abnormalities and shorter reproductive lifespan are common in females with CKD, although kidney transplantation may improve menstrual health.

Introduction

Globally, CKD is more prevalent in women (1), with over 40% affected in the reproductive years (2). Although considerable attention has been placed on improving fertility rates and complications of pregnancy in the CKD population (3), menstrual abnormalities have received notably less consideration despite menarche, menstruation, and menopause being fundamental parts of female reproductive biology. Despite the fact that they are commonly experienced in individuals with CKD (4), studies exploring menstrual abnormalities, shorter reproductive lifespan, and early menopause are often limited by sample size and combine results from populations across CKD stages, including those with kidney transplants (5). Menstrual abnormalities, premature menopause, and shortened reproductive lifespan are associated with lower life expectancy, higher risk of cardiovascular disease (6,7), and osteoporosis (8) in the general population. However, the prevalence of these reproductive factors in the CKD population is unclear. There is increasing recognition that premenopausal-aged women treated with dialysis or kidney transplantation are at high risk of both cardiovascular- and noncardiovascular-related mortality compared with the general population (9) as well as age-matched men treated with dialysis (10). As female reproductive health is associated with important nonreproductive health outcomes, including mortality (7), this underscores the importance of understanding the prevalence of menstrual abnormalities and the reproductive lifespan in the CKD population. Therefore, the objective of this systematic review and meta-analysis was to determine the prevalence of menstrual abnormalities and quantify the reproductive lifespan of females with CKD.

Materials and Methods

Protocol

This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (11), and the protocol was registered with the Prospective Systematic Review Register.

Information Sources

Published studies were identified through searching electronic databases (MEDLINE, Embase, and Cumulative Index to Nursing and Allied Health Literature [CINAHL]) from inception until February 2022 without language restrictions. Relevant text words and medical subject headings were used to develop a search strategy (Supplemental Table 1) on two broad themes encompassing the population (females with CKD; theme 1) and outcomes (reports on menstrual abnormalities and/or reproductive lifespan; theme 2). The two themes were combined using the Boolean operator “AND,” and the search was limited to original, peer-reviewed human studies. References from selected papers were scanned to ensure completeness. The search strategy was reviewed by research librarians within the University of Calgary. Database-specific subject headings and medical subject heading words were adapted for each database as necessary. Non-English articles were translated using Google Translate.

Study Eligibility

Study Article Identification.

Studies were eligible for inclusion if they (1) reported females with CKD, including those undergoing/who underwent KRT, such as hemodialysis (HD), peritoneal dialysis (PD), or kidney transplantation (if a study included kidney transplantation recipients, menstrual patterns both before and after transplantation were required to be reported for inclusion); (2) reported the menstrual status of females, including the number of individuals with regular menstruation, individuals with irregular menstruation and/or those with menstrual abnormalities, and/or reproductive lifespan and/or reproductive lifespan–associated conditions (e.g., primary ovarian insufficiency); and (3) were an observational study design (i.e., cohort, cross-sectional, or case-control).

Study Selection.

After conducting a calibration exercise, titles and abstracts from the literature search were screened independently and in duplicate by two reviewers (C.L.R. and G.S.K.). The same process was followed for the screening of full-text articles. When data were missing or clarification of published data was needed, the study authors were contacted. Percentage agreement was quantified at each screening stage, and any disagreements in study inclusion were resolved by discussion with a third reviewer (L.S.).

Data Extraction.

Study identifiers (authors and location and year of publication), study design characteristics (sample size and inclusion/exclusion criteria), participant characteristics (age and comorbidities), CKD characteristics (year/age of diagnosis, KRT, and etiology), and menstruation characteristics (hormone therapy use, menstrual abnormalities related to regularity, frequency, duration, amount, age of menarche, and age of menopause) were abstracted using a standardized form. Individuals with primary amenorrhea were not included in counts for amenorrhea. Assessment of study quality was completed using a modified version of the National Institutes of Health Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies and Case-Control Studies (12). Study quality for observational and cross-sectional studies was rated as poor (zero to three of 12 questions), fair (four to seven), or good (eight to 12). Case-control studies were rated as poor (zero to five of 17), fair (six to 11), or good (12 to 17).

Data Synthesis and Analysis.

The proportion of patients with menstrual abnormalities and the mean reproductive lifespan were the primary measures of interest. Denominator sample sizes for menstrual disturbance calculations excluded nonreporting, individuals with primary amenorrhea, postmenopausal individuals, and those with menstrual abnormalities prior to CKD diagnoses. Reporting of both age of menarche and age of menopause was required to calculate the reproductive lifespan. When two values were presented for either age of menarche or menopause (e.g., reporting stratified by KRT modality type), the ages were combined using a sample size weighted mean within each study to produce one overall number. When possible, measures were pooled using DerSimonian Laird random effects models (13). Pooled estimates for the proportions of menstrual abnormalities were assessed separately by KRT. For dialyzed individuals, pooled proportion estimates were further stratified by HD and PD. Pooled mean ages were calculated for reported ages of menarche, menopause, and reproductive lifespan separately. Pooled estimates were used to compare changes in menstrual abnormalities in transplanted females pre- and postkidney transplantation, stratified by menstrual abnormalities type. Heterogeneity was determined using Cochran Q and I² statistics. Publication bias was determined through visual assessments of funnel plots. A two-sided P=0.05 was considered statistically significant. All analyses were performed using STATA (V16; StataCorp LP, College Station, TX).

Results

Study Selection

Of 4485 potentially relevant citations identified within the electronic search, 3439 titles were assessed, and 166 full-text publications were retrieved for further review, of which 48 studies met inclusion criteria (Figure 1). Two sets of two studies (set 1 [14,15] and set 2 [16,17]) represented the same study population and were merged to ensure that there was no duplication. Therefore, 46 studies (27 cohort, 17 cross-sectional, and two case-control) were retained representing 5636 individuals. Sixteen articles reported menstrual abnormalities of dialyzed patients, 11 reported menstrual abnormalities of transplanted patients, two studies reported menstrual abnormalities in females treated with dialysis and kidney transplantation, and six studies reported reproductive lifespan. Percentage agreement values between reviewers (C.L.R. and G.S.K.) were 91% and 77% for title and full-text review, respectively.

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow chart. CINAHL, Cumulative Index to Nursing and Allied Health Literature.

Study Characteristics

Characteristics of included studies are shown in Table 1. The mean age of study cohorts ranged from 15 to 70 years old. Study publication dates ranged from 1975 to 2021, with the majority (n=32) published after 2000. Three articles (18–20) included individuals with nondialysis-dependent CKD (n=1165), and three (5,21,22) included those receiving conservative kidney care (n=92). The majority of individuals were receiving HD (n=3039) or PD (n=229), and 1084 kidney transplant recipients had undergone HD (n=223), PD (n=37), or both (n=8) before transplantation. Thirteen articles did not specify which kind of dialysis, if at all, was undertaken before transplantation (5,22–34).

Table 1.

Study demographics

Authors (yr)	Study Design	Country	Study Size	Age Range, yr	Mean Age, yr	Ethnicity/Race, % White	Stage of CKD	Months since Diagnosis (SD)	Hemodialysis, n	Peritoneal Dialysis, n	Months on Dialysis (SD)	Kidney Transplantation, n	Months since Kidney Transplantation (SD)	Age at Kidney Transplantation, yr (SD)	Dialysis before Kidney Transplantation	Months on Dialysis Prekidney Transplantation (SD)
Ølgaard et al. (1975) (38)	Cross-sectional	Denmark	20	14–58	HD group: 36; PD group: 40	NR	5d	—	10	10	19 (range, 0.5–50)	—	—	—	—	—
Bierman and Nolan (1977) (36)	RCS	United States	23	13–43	24	NR	5d	4.4	—	—	9.5 (NR)	23	—	23.7 (NR)	Yes, n=23 HD	—
Morley et al. (1979) (24)	RCS	South Africa	22	17–37	27	NR	5d	—	17	—	—	5	24.2 (NR)	—	Unclear	—
Lim et al. (1980) (58)	Cross-sectional	United States	24	18–59	—	NR	5d	—	24	—	Range, 6–48	—
Gómez et al. (1980) (41)	Cross-sectional	Switzerland	23	19–66	44	NR	5d	—	23	—	16 (2)	—	—	—	—	—
Weizman et al. (1983) (51)	Cross-sectional	Israel	21	20–53	37	NR	5d	—	21	—	21.6 (9.8)	—	—	—	—	—
El-Sheikh et al. (1987) (28)	RCS	United Kingdom	48	20–66	46	NR	5d	—	—	—	—	48	—	—	Yes, unclear what type
Schover et al. (1990) (59)	Cross-sectional	United Kingdom	36	—	40	NR	5d	—	—	—	—	36	38.4 (NR)	—	Yes; n=31 HD, n=5 PD	HD group: 19.0 (22.2); PD group: 10.0 (5.6)
Bianchi et al. (1992) (21)	Cross-sectional	Italy	36	—	47	NR	Stage 3 (n=13), stage 4 (n=13), stage 5 (n=10)	—	—	—	—	—	—	—	—	—
Otieno et al. (1993) (22)	PCS	Kenya	40	—	29	NR	5d	—	13	1	—	—	—	—	—	—
Milde et al. (1996) (78)	Cross-sectional	United States	56	25–86	53	95	5d	n=56 <36 (NR)	41	15	n=38 >36 (NR)	—	—	—	—	—
Cochrane and Regan (1997) (5)	PCS	United Kingdom	100	17–70	39	NR	5d	—	17	9	—	44	—	—	Unclear	—
Holley et al. (1997) (42)	RCS	United States	76	15–59	42	59	5d	—	52	24	54 (50.4)	—	—	—	—	—
Kim et al. (1998) (31)	RCS	South Korea	72	14–54	38	NR	5d	—	—	—	—	72	40.2 (29.7)	34.6 (8.8)	Yes, unclear what type	27.0 (31.9)
Kawashima et al. (1998) (55)	RCS	Japan	25	24–48	38	NR	5d	—	25	—	53.4 (range, 7–144)	—	—	—	—	—
Rush et al. (2000) (39)	RCS	United States	102	26–88	56	15	5d	—	102	—	49.2 (4.8)	—	—	—	—	—
Weisinger et al. (2000) (47)	Cross-sectional	Venezuela	74	—	37	NR	5d	86.9 (69.4)	74	—	56.9 (43.3)	—	—	—	—	—
Jang et al. (2001) (43)	Cross-sectional	Australia	48	20–84	56	NR	5d	—	48	—	46.8 (range, 1–204)	—	—	—	—	—
Shanmugavadivoo and Shaariah (2003) (40)	RCS	Malaysia	137	—	—	0	5d	—	72	65	33.3 (range, 2–216)	—	—	—
Kramer et al. (September 2003) (48)	RCS	United States	41	—	37	76	5d	—	—	—	—	41	n=24 >60 (NR); n=17 <60 (NR)	30.1 (9.6)	Unclear	—
Matuszkiewicz-Rowińska et al. (2004) (17)	Cross-sectional	Poland	75	—	35	NR	5d	—	75	—	32 (30)	—	—	—	—	—
Kramer et al. (March 2003) (37)	Cross-sectional	United States	238	47–89	70	66	5d	—	238	—	38.4 (39.6)	—	—	—	—	—
Di Iorio et al. (2004) (57)	Cross-sectional	Italy	1256	25–84	63	NR	5d	—	1256	—	62.2 (61.4)	—	—	—	—	—
Yücel et al. (2004) (53)	RCS	Turkey	32	25–65	44	NR	5d	—	32	—	86.4 (range, 24–231)	—	—	—	—
Pezeshki et al. (2004) (25)	Case-control	Iran	50	—	28	NR	5d	—	—	—	—	50	—	—	Unclear	—
Guazzelli et al. (2008) (29)	RCS	Brazil	197	28–44	36	NR	5d	—	—	—		197	108	32.7 (8.8)	Yes, unclear what type	—
Song et al. (2008) (69)	Case-control	South Korea	38	—	46	NR	5d	—	38	—	98 (range, 12–264)	—	—	—	—	—
Filocamo et al. (2009) (54)	PCS	Italy	39	—	36	NR	5d	—	—	—	—	39	16 (range, 13–46)	—	Yes, n=39 HD	18 (range, 12–36)
Karayalcin et al. (2010) (30)	RCS	Turkey	50	18–52	34	NR	5d	—	—	—	—	50	52.8 (50.9)	—	Yes, unclear what type	23.5 (11.2)
Arikan et al. (2011) (44)	RCS	Turkey	30	15–45	34	NR	5d	—	30	—	75.6 (182.2)	—	—	—	—	—
Koca et al. (2012) (32)	RCS	Turkey	117	—	HD group: 41; PD group: 40; KT group: 37	NR	5d	—	39	43	HD group: 50.4 (39.6); PD group: 56.4 (52.8)	35	36.0 (45.6)	—	Yes, unclear what type	38.4 (30)
van Eps et al. (2012) (79)	PCS	Australia	7	—	41	NR	5d	—	7	—	—	—	—	—	—	—
Yu et al. (2013) (26)	RCS	China	42	25–45	29	NR	5d	—	—	—	—	42	—	—	Yes, unclear what type	16.0 (range, 1–27 mo)
Al-Turki et al. (2014) (35)	Cross-sectional	Saudi Arabia	40	19–90	52	NR	5d	—	40	—	>60	—	—	—	—	—
Cheung et al. (2015) (18)	RCS	United States	1007	—	69	50	Unclear	—	—	—	—	—	—	—	—	—
Chakhtoura et al. (2015) (23)	RCS	France	129			NR	5d	—	—	—	—	102	—	—	Yes, unclear what type	—
Teuwafeu et al. (2016) (50)	Cross-sectional	Cameroon	52	18–69	38	NR	5d	—	52	—	Median: 36.0 (IQR, 16–60)	—	—	—	—	—
Lin et al. (2016) (34)	Cross-sectional	China	165	14–44	35	NR	5d	HD group: 57 (21); PD group: 54 (18); KT group: 56 (19)	86	43	HD group: 52 (8.9); PD group: 48.5 (11.3)	36	—	—	Yes, unclear what type	KT group: 49.8 (10.2)
Sikora-Grabka et al. (2018) (15)	PCS	Poland	10	31–37	34	NR	5d	167 (95% CI, 74 to 258)	—	—	—	—	—	—	Yes, n=8 HD, n=2 PD	Median: 28.6 (95% CI, 11 to 46)
Gonçalves et al. (2019) (68)	Cross-sectional	Portugal	31	20–72	49	NR	5d	156 (NR)	—	—	—	31	Range, 6–60 (NR)	—	Yes, n=105 HD, n=20 PD, n=8 both HD and PD	52.8 (NR)
Yaprak et al. (2019) (52)	RCS	Turkey	33	—	34	NR	5d	—	—	—	—	33		29.7 (6.3)	Yes, n=17 HD, n=10 PD	Median: 21.5 (IQR, 6–105)
Fayed et al. (2019) (45)	PCS	Egypt	60	20–35	30	NR	5d	79 (42)	60	—	26.9 (8.8)	—	—	—		—
Ramesh et al. (2020) (60)	PCS	Canada	476	—	60	75	5d	—	476	—	—	—	—	—	—	—
Serret-Montoya et al. (2020) (20)	PCS	Mexico	57	12–17	15 (median)	NR	Stage 4 (n=13), stage 5 (n=44); n=34/44 receiving dialysis	Median: 14 (range, 1–144)	25	19	Median: 16 (range, 3–60)	—	—	—	—	—
Kim et al. (2020) (19)	PCS	United States	145	1–16	NR	0	G2, G3a, G3b, unclear n for each	Median: 168 (IQR, 96–180)	—	—	—	—	—	—	—	—
Dines et al. (2021) (27)	Cross-sectional	United States	190	18–44	NR	78	5d	—	—	—	—	190	—	—	Unclear	Unclear

HD, hemodialysis; PD, peritoneal dialysis; NR, not reported; —, not reported; RCS, retrospective cohort study; PCS, prospective cohort study; KT, kidney transplantation; IQR, interquartile range; 95% CI, 95% confidence interval.

Hormonal Therapy Use

Fourteen studies (18,23,24,27,29,33,35–43) reported hormone therapy/contraceptive use, although numerous (15,34,44–47) considered hormone therapy as exclusion criteria. Eight dialysis-specific articles (24,39–43,48) reporting on 585 individuals reported low use of hormone replacement therapy (n=363), oral contraceptives (n=6), and intrauterine devices (n=2) among study participants. Five transplant-specific articles (23,27,29,33,36) reported 242 and 138 individuals using hormonal contraceptives before and after transplantation, respectively. The most common type was oral contraceptives both before (n=205; 53%) and after (n=124; 34%) transplantation. However, few studies reported on the hormonal composition (e.g., combined or progestogen) of the contraceptive used (23,24,27).

Medication Use

There were 27 studies (14,17,24,26,28–30,33,34,36,38,39,41,43,44,49–56) reporting a combined total of 3534 medications used in 2706 study participants (1.25:1 ratio). The most commonly reported medications included those for anemia treatment (e.g., erythropoietin and ferrous sulfate; n=1612; 46%), immunosuppressant medications (e.g., cyclosporin, tacrolimus, prednisone, mycophenolate mofetil, and mycophenolate sodium; n=1282; 36%), and antihypertensive medications (e.g., propranolol, β-blockers, and calcium channel blockers; n=154; 4%).

Study Outcomes

Menstrual Abnormalities Definitions.

Definitions and descriptions of menstrual conditions are listed in Supplemental Tables 2 and 3 for dialysis and transplantation patients, respectively. Amenorrhea was defined as the absence of menstrual bleeding for 3 (57) to ≥6 months (14,17,23,26,28,33,34,47,55,56), oligomenorrhea was defined as menstruation intervals of >32 days (17) or >35 days apart (34,41,55), polymenorrhea was defined as cycles lasting <28 days (26,54), and irregular menses were defined as excessive (bleeding >5 days), short (<20 days) or long (>45 days) cycles (29), or ranging from occasional spotting to frequent dysfunctional uterine bleeding (58). Several articles failed to provide definitions of oligomenorrhea (22,24,44,50), polymenorrhea (24,50), amenorrhea (22,24,31,35,36,41,44,50,51,58,59), or irregular/sporadic menses (14,15,31,33,35,36,38,40,43,59). Other articles listed definitions of menstrual abnormalities as questions or statements (27,30). One study (20) listed definitions for each menstrual abnormality but did not stratify results by KRT.

Menstrual Abnormalities in Nondialysis-Dependent, Nontransplanted Females.

Only three articles (18–20) included individuals with CKD not treated with dialysis or kidney transplant (n=1165), and three articles (5,21,22) included individuals receiving conservative kidney care (n=92). The ages of menarche and menopause were reported by Kim et al. (19) and Bianchi et al. (21), and only menstrual abnormalities in eight of 13 (61%) individuals who were predialysis were reported by Serret-Montaya et al. (20). Otieno et al. (22) reported menstrual abnormalities in four of 26 (15%) individuals treated with conservative care both before and after treatment initiation. The combined prevalence of menstrual abnormalities among all participants (i.e., conservative treatment, HD, and PD) was reported by Cochrane and Regan (5). Thus, we were unable to report a global assessment of menstrual abnormalities in populations with CKD not treated with dialysis or kidney transplantation.

Menstrual Abnormalities in Dialyzed Females.

Data on menstrual abnormalities in patients on dialysis were collected and summarized in Figures 2 and 3, Supplemental Figure 1, and Supplemental Table 2. The overall rate of menstrual abnormalities in patients on dialysis from 18 studies was 63% (95% confidence interval [95% CI], 51% to 75%) (Supplemental Figure 1); however, significant heterogeneity was noted (I²: 95%; P<0.001). Stratification by menstrual disturbance type within the HD group (Figure 2) showed proportions of irregular menses (47%; 95% CI, 38% to 56%; I²: 12%; n=6 studies), polymenorrhea (19%; 95% CI, 13% to 26%; I²: 23%; n=4), amenorrhea (41%; 95% CI, 30% to 52%; I²: 87%; n=13), and oligomenorrhea (19%; 95% CI, 14% to 25%; I²: 14%; n=9). The proportion of menstrual abnormalities in females undergoing PD (Figure 3) was 75% (95% CI, 51% to 98%; n=4) with notable heterogeneity (I²: 89%; P<0.001). There were too few studies including PD populations to stratify by menstrual disturbance type.

Figure 2.

Prevalence of menstrual abnormalities stratified by menstrual disturbance type in females undergoing hemodialysis. 95% CI, 95% confidence interval.

Figure 3.

Prevalence of menstrual abnormalities in females undergoing peritoneal dialysis.

Menstrual Abnormalities in Transplanted Females.

Menstrual abnormalities in kidney transplant recipients were reported in 13 studies (15,23,24,26–31,33,36,54,59), and both pre- and post-transplantation estimates were collected from studies and are summarized in Figure 4 and Supplemental Table 3. The overall prevalence of amenorrhea decreased post-transplantation (8.01; 95% CI, 4.41 to 11.62) as compared with pretransplantation (39.32; 95% CI, 31.26 to 47.38). No significant differences were observed in the prevalence of oligomenorrhea (pretransplant: 20.53; 95% CI, 8.81 to 32.25; post-transplant: 17.83; 95% CI, 12.39 to 23.27), polymenorrhea (pretransplant: 7.39; 95% CI, 1.73 to 13.04; post-transplant: 10.23; 95% CI, 3.67 to 16.79), or irregular menses (pretransplant: 38.37; 95% CI, 18.60 to 58.13; post-transplant: 32.42; 95% CI, 22.21 to 42.62) with kidney transplantation. High levels of between-study heterogeneity were noted across most subgroups (P=0.05).

Figure 4.

Prevalence of menstrual abnormalities stratified by menstrual disturbance type in females pre- and post-transplantation.

Menarche, Menopause, and Reproductive Lifespan.

The age of menarche was reported in 1116 individuals in 12 articles (19,20,23,27,29,31,32,36,40,42,45,53); it ranged from 11.9 to 15.6 years and did not appear to differ among nondialyzed (13.73; 95% CI, 13.36 to 14.10; n=1 study), dialyzed (12.90; 95% CI, 12.48 to 13.32; n=6), and transplanted females (13.68; 95% CI, 12.96 to 14.40; n=6) (Supplemental Figure 2). The age of menopause was reported in 1587 individuals in 14 articles (18,21,23,27,28,31,33,39,40,42,43,50,53,60); it ranged from 36 to 64 years and did not appear to differ among nondialyzed (47.62; 95% CI, 46.66 to 48.57; n=2), dialyzed (46.99; 95% CI, 44.83 to 49.14; n=8), and transplanted females (46.29; 95% CI, 43.19 to 49.40; n=3) (Supplemental Figure 3). Only two articles (39,40) reported when menopause was reached with respect to dialysis initiation, where the majority reached menopause before (n=97; 78%) as compared with after (n=28; 22%) dialysis initiation. Similarly, only three articles (23,28,31) reported that all individuals (n=107; 100%) reached menopause after kidney transplantation. Of the 12 and 14 articles reporting age of menarche and age of menopause, respectively, only six articles (23,27,31,40,42,53) reported both the age of menarche and the age of menopause and were thus used to calculate reproductive lifespan (Supplemental Figure 4). Reproductive lifespan length was found to be 32 (95% CI, 30 to 34) years, although there was significant heterogeneity (I²: 92%; P<0.001). Primary ovarian insufficiency was reported in 12 individuals from two studies (5,23), at least five of whom were transplant recipients (the remaining seven were from one study [5] reporting dialyzed and transplantation recipients combined). Early menopause (<45 years old, n=264 of 1007 [26%]; 45–50 years old, n=227 of 1007 [22%]) was reported in women with CKD from one study (18); however, the type of KRT undergone, if any, was not reported and was therefore not included in analyses.

Study Quality and Risk of Bias.

Study quality is described in Supplemental Table 4. All studies were considered of either fair (n=32) or good (n=14) quality. The most common study quality factors reported included temporality of exposure and outcome measures and clearly stated research question/objective and population. Study quality factors that were least reported included sample size justifications and measurement of confounding variables. There was no evidence of publication bias through visual assessment of funnel plots (Supplemental Figure 5).

Discussion

This systematic review and meta-analysis represents the first comprehensive review, to our knowledge, summarizing menstrual abnormalities and reproductive lifespan among females with nondialysis-dependent CKD without kidney transplantation, women with dialysis-dependent CKD, and women with kidney transplantation. The findings from this systematic review and meta-analysis are three-fold. (1) The prevalence of menstrual abnormalities is high (63% overall) in dialyzed females, specifically irregular menses (47%) and amenorrhea (41%). (2) Kidney transplantation is associated with a reduction in menstrual abnormalities, particularly amenorrhea (approximately 30% reduction), although between-study heterogeneity was high. (3) The length of reproductive lifespan in females with CKD is 32 years. Thus, these results have important implications for the gynecologic and kidney care for this patient population (3).

Menstrual abnormalities are common among the general population of females (61), ranging from 6% to 40%, and they vary by disturbance type, socioeconomic position, and lifestyle factors (62). Menstrual abnormalities have significant public health ramifications (63), including missing work or school, which exacerbates sex inequity on a global level. The high prevalence of menstrual disorders in those with CKD is a concern and underscores reproductive health as an important aspect in the care of this population. However, nearly two thirds of nephrologists report not feeling confident managing (64) menstrual abnormalities in females with CKD, and almost two thirds of patients reported being unaware that menstruation may change with kidney transplantation (31).

Individuals with kidney failure treated with HD have abnormal hypothalamic-pituitary-ovarian axis function, with a reduction in the pulsatile release of gonadotropin-releasing hormone, resulting in lowered levels of follicle-stimulating hormone and luteinizing hormone (58) and no concurrent change in estradiol levels (60). However, the CKD stage at which disruption of the hypothalamic-pituitary gonadal axis occurs is unknown. Similarly, higher levels of prolactin may contribute to anovulation through negative feedback inhibition on the hypothalamic-pituitary-ovarian axis (65). Additionally, autoimmune disorders are more commonly a cause of CKD in premenopausal females (66), and thus, menstrual abnormalities may also be attributable to gonadotoxic medications (i.e., cyclophosphamide) in the CKD population (67) or the disease process itself. However, studies on menstrual health in the setting of CKD have previously been limited by small sample sizes, considerable exclusion criteria, and lack of stratification by CKD stage as well as biased survey techniques. For example, numerous studies considered inclusion of only married females (26,51,68,69) or interviewed individuals other than the patient (e.g., spouse or child) on menstrual pattern history (40). Thus, the current understanding of menstrual health in the CKD population warrants additional exploration.

KRT modalities, such as dialysis and transplantation, may have differing effects on regulating menstrual cycles. It has been noted that both HD and PD rarely improve the hypothalamic-pituitary-ovarian axis, sustaining abnormal hormonal milieus (17), although menstrual abnormalities can improve marginally. In contrast, successful kidney transplantation may improve hypothalamic-pituitary-ovarian axis irregularities (70), causing gradual restoration of luteinizing hormone, prolactin, and estradiol levels similar to those of healthy controls (71), although estradiol increases may be caused by immunosuppressive agents that interfere with estrogen receptor binding (72). However, irrespective of KRT modality, menstrual abnormalities are not completely alleviated. As noted in this systematic review, it is unclear why menstrual abnormalities other than amenorrhea are not improved following kidney transplantation. We speculate that nonsignificant differences may be due to between-study heterogeneity, differences in study cohorts, and a wide-spanning range of publication dates, as well as potential differences in the definitions, descriptions, and collection of menstrual abnormalities in each study population. Similarly, confounding factors, such as medication use, may have affected results. Taken together, these factors underscore the implications for fertility and overall health in females with CKD (3).

Although our findings indicate that the pooled average age of menarche of 13 years is similar to the general population, the pooled average age of menopause of 47 years and the reproductive lifespan of 32 years are considerably shorter (73). Reproductive lifespan has recently emerged as an important sex-specific factor that is inversely proportional to kidney and cardiovascular risk (6,74). Whether these associations represent a causal relationship or simply reflect a shared pathophysiology with cardiovascular disease in individuals with CKD is unclear. Similarly, earlier onset of menopause has been associated with lower life expectancy, greater risk for osteoporosis and cardiovascular disease, and lower risk of breast and gynecologic cancers in the general population (75). However, it has not been investigated if these same associations apply to the population with CKD, and therefore, they represent an unmet need in nephrology research and care.

This study has strengths and limitations. First, there was a lack of clarity and unanimity across studies for definitions of the types of menstrual abnormalities that may affect the synthesis of results. To the best of our abilities, definitions presented by authors were used verbatim to summarize menstrual abnormalities. Second, there was significant heterogeneity noted in results, which may be due to differing diagnostic criteria, definitions, and sampling processes; as such, our results are to be interpreted with caution. Further, because of the limited reporting, we were unable to adjust analyses by hormonal contraceptive use, which is known to induce amenorrhea and/or regulate menstrual cycles, although contraceptive use is low in this population (76,77). Similarly, we were unable to adjust analyses by immunosuppressive use or rates of fertility, which can also contribute to menstrual abnormalities (3). However, this review highlights the high prevalence of menstrual abnormalities and shorter reproductive lifespan in the CKD population and underscores the urgency of addressing this important component of overall health.

Menstrual disorders are common in females with CKD and are more than a reproductive health issue (63), with implications for quality of life and socioeconomic status (63). These sex-specific factors of women are associated with cardiovascular (6,7) and bone health (8) as well as malignancy risk (9) in the general population; whether these same associations exist in the CKD population is unknown. This lack of data precludes guidance in this key aspect of care of female patients across the stages of CKD and represents an important knowledge gap in nephrology.

Disclosures

S.B. Ahmed reports serving as an Advisory Board Member of the Canadian Institutes of Health Research Institute (CIHR) of Gender and Health (volunteer position), as a Canadian Medical Association Journal Governance Council Member (elected position), and as the President-Elect of the Organization for the Study of Sex Differences (elected volunteer position) and reports research funding from the Canadian Institutes of Health Research. M. Robert reports research funding from Can Health West Network, a CIHR Operation Grant, Cumming Medical Research Fund (CMRF), and a Women's Health Clinical Mentorship Grant; serving as the Medical Director of the Calgary Chronic Pain Program; and serving as a cochair of the Alberta Pain Strategy Outcomes Subcommittee (not paid). L. Skeith reports research funding from CSL Behring and honoraria from Leo Pharma and Sanofi. All remaining authors have nothing to disclose.

Funding

None.

Author Contributions

S.B. Ahmed, G.S. Kochaksaraei, C.L. Rytz, and L. Skeith conceptualized the study; S.M. Dumanski, G.S. Kochaksaraei, and C.L. Rytz were responsible for data curation; G.S. Kochaksaraei and C.L. Rytz were responsible for investigation; P.E. Ronksley and C.L. Rytz were responsible for formal analysis; M. Robert, P.E. Ronksley, C.L. Rytz, and L. Skeith were responsible for methodology; P.E. Ronksley and C.L. Rytz were responsible for software; S.B. Ahmed, P.E. Ronksley, and C.L. Rytz were responsible for validation; P.E. Ronksley and C.L. Rytz were responsible for visualization; S.B. Ahmed, S.M. Dumanski, P.E. Ronksley, and L. Skeith provided supervision; C.L. Rytz wrote the original draft; and S.B. Ahmed, S.M. Dumanski, G.S. Kochaksaraei, M. Robert, P.E. Ronksley, C.L. Rytz, and L. Skeith reviewed and edited the manuscript.

Data Sharing Statement

All data used in this study are available in this article.

Supplemental Material

This article contains the following supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.07100622/-/DCSupplemental.

Supplemental Figure 1. Prevalence of menstrual disturbances in dialyzed females.

Supplemental Figure 2. Pooled age of menarche stratified by KRT type in females with CKD.

Supplemental Figure 3. Pooled age of menopause stratified by KRT type in females with CKD.

Supplemental Figure 4. Pooled reproductive lifespan in females with CKD.

Supplemental Figure 5. Funnel plot for assessment of publication bias of pooled risk difference of menstrual disturbance in females with CKD pre- and postkidney transplantation.

Supplemental Table 1. MEDLINE search strategy.

Supplemental Table 2. Menstrual disturbances in dialyzed females living with CKD.

Supplemental Table 3. Menstrual disturbances in transplanted females living with CKD.

Supplemental Table 4. Study quality assessment.