Secondary Logo

Journal Logo

Clinical Science

Prolonged Amenorrhea and Low Hip Bone Mineral Density in Women Living With HIV—A Controlled Cross-sectional Study

King, Elizabeth M. MDa; Nesbitt, Ariel MDa; Albert, Arianne Y.K. PhDb,c; Pick, Neora MDa,b,c; Cote, Helene C.F. PhDb,d; Maan, Evelyn J. RNb,c; Prior, Jerilynn C. MDb,e; Murray, Melanie C.M. MD, PhDa,b,c; the CIHR team on Cellular Aging and HIV Comorbidities in Women and Children (CARMA)

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: April 15, 2020 - Volume 83 - Issue 5 - p 486-495
doi: 10.1097/QAI.0000000000002282

Abstract

INTRODUCTION

Globally, more than 50% of people living with HIV are women.1 In the era of combined antiretroviral therapy (cART), because women with HIV are living longer, healthier lives, it is increasingly important to understand the chronic health complications associated with HIV. Prolonged secondary hypothalamic (sometimes called “functional”) amenorrhea is one such condition.2 Amenorrhea is an adaptive, reversible response to numerous physiological, nutritional, psychosocial, and illness-related stressors3; it is more commonly experienced by women living with HIV (WLWH) than HIV-negative populations and reported to occur at rates of approximately 5% or almost double that of the general population.4–7

In the general population, amenorrhea can be a result of multiple causes, including but not limited to hypogonadotropic hypogonadism, premature ovarian insufficiency, dysregulation of other hormones, including prolactin and thyroid-stimulating hormone, uterine abnormalities, and polycystic ovarian syndrome (PCOS).3 Although the exact cause of increased amenorrhea in HIV is not fully elucidated, it may relate in part to comorbidities that disproportionately affect WLWH, such as opioid, illicit substance and antipsychotic use, extremes of weight, and psychosocial stresses.4–6,8 Interestingly, in a recent analysis that controlled for many of these factors, including medication and substance use, HIV remained an independent predictor for amenorrhea, drawing to question whether the virus itself may directly affect reproductive pathways.4 Importantly, most underlying causes of amenorrhea, such as premature ovarian failure and hypogonadotropic hypogonadism, are accompanied by hormonal dysregulation characterized by low estrogen and progesterone, the notable exception being PCOS, which can be accompanied by normal or high levels of estrogen.3,9

Among HIV-negative women, amenorrhea and low estradiol/progesterone states have been linked to a wide range of negative health effects, including accelerated atherosclerotic disease, increased osteoporosis and fracture risk, and infertility, and are also associated with adverse mental health.10–12 In addition, evidence is emerging that estrogen may also play regulatory roles in preventing progression of renal decline in those with chronic kidney disease and progression of hepatic fibrosis in patients with HIV/hepatitis C virus (HCV) coinfections.13–15 Taken together, the downstream effects of amenorrhea and its associated ovarian hormonal changes have the potential to be profound when considered in the context of HIV-associated aging and comorbidities.

Low areal bone mineral density (BMD) is common in persons living with HIV (PLWH). PLWH are 6.8 times more likely to have osteopenia compared with HIV-negative controls, with 67% and 16% meeting criteria for osteopenia and osteoporosis, respectively.16 In PLWH, BMD loss is twice as rapid among women as in men and occurs regardless of the initiation of cART.17,18 In PLWH, the rate of BMD loss after cART initiation is approximately 2%–6% per year during the first 2 years.17,19 The cause of low BMD among PLWH is likely influenced by numerous factors, including direct viral effects, chronic inflammatory states, co-occurrence of osteoporotic risk factors, and cART side effects.20,21 Clinically, altered bone metabolism is correlated with increased fracture rates in PLWH compared with uninfected controls.22–24 Taken together, low BMD and rapid bone loss are particularly common in WLWH and important sources of morbidity.

Amenorrhea is associated with low BMD in HIV-negative populations, a finding related to the low estradiol and progesterone levels in most amenorrheic states.25 Normal and stable premenopausal estradiol levels are needed to prevent increased bone resorption and preserve bone architecture.20 Progesterone acts to increase bone formation through progesterone receptor–mediated osteoblastic activity.25 Similarly, population studies of HIV-negative women have shown an association between amenorrhea and decreased BMD at both the lumbar spine26 and total hip/femoral neck sites.27 In WLWH, although Dolan et al showed an association of oligomenorrhea and osteopenia, no studies have yet evaluated the effects of lifetime history of prolonged amenorrhea (ie, ≥1 year without menses) on BMD among WLWH, a condition we postulate may further negatively impact BMD.11 Identifying factors that reduce BMD in WLWH is important, especially when considering their already compromised skeletal health. Such a characterization is an important step toward early recognition and prevention of osteoporosis and fracture within this high-risk group.

With this in mind, we conducted a cross-sectional study examining the relationship between prolonged amenorrhea and BMD in WLWH compared with an HIV-negative but socioculturally similar control group.

METHODS

Cohort and Participants

The Children and women: AntiRetrovirals and Markers of Aging (CARMA) study is a prospective community cohort study of WLWH and HIV-negative control women that aims to investigate the effects of HIV and cART on cellular aging in women and children. Several substudies have stemmed from this main study, including CARMA-OSTEO, focused on examining BMD in WLWH, and CARMA-ENDO, examining endocrine, metabolic, and reproductive health variables. We included women who were enrolled in both the CARMA-ENDO and CARMA-OSTEO substudies for the present cohort–control study.

Study enrollment occurred from January 2013 to August 2017. Figure 1 outlines participant inclusion for this analysis. Participants were women aged 19 years and older and had completed questionnaires, including detailed review of menstrual patterns (see Endocrinology Questionnaire, Supplemental Digital Content, http://links.lww.com/QAI/B423) and bone risk factors (see Osteoporosis Questionnaire, Supplemental Digital Content, http://links.lww.com/QAI/B424). A dual x-ray absorptiometry (DXA) scan was performed as close to the CARMA-ENDO visit as possible. History of prolonged amenorrhea was taken by participant self-report from a cross-sectional survey at a single time point. Participants were asked to report on age of amenorrhea, perceived cause, and duration. Participants with a history of bilateral oophorectomies or primary amenorrhea (ie, never developed menstrual periods) were excluded. Demographic, clinical, laboratory, and substance exposure data were collected prospectively during CARMA visits. Medical history, including HCV coinfection and substance use, was by self-report and/or medical record data. Fracture history was by systematic questionnaire and included how the fracture occurred. For WLWH, visits took place during regular care visits at the Oak Tree Clinic in the British Columbia Women's Hospital; controls had data collected over 1 or 2 visits without longitudinal follow-up.

FIGURE 1.
FIGURE 1.:
Flow diagram of women included in analysis grouped by prolonged secondary amenorrhea experience.

Definition of Prolonged Secondary Amenorrhea

Women were classified as having either “prolonged secondary amenorrhea,” hereafter referred to simply as “amenorrhea,” or a more normal menstrual cycle pattern (Fig. 1). For the purpose of this analysis, prolonged secondary amenorrhea was defined as self-reported history of no menses for at least 1 year in the past or present occurring when the woman was younger than 45 years and not because of surgery, breastfeeding, pregnancy, or continuous combined hormonal contraception, a progestin-releasing intrauterine device or depot medroxyprogesterone injections.

For our definition, we selected a timeline of 1 year without menses based on previously published large population studies of amenorrhea in WLWH.5 We used an age cutoff of <45 years based on published data that define early menopause as lack of menses before age 45.28 Most women (>95%) experience menopause at ages ≥45 years,29 and we selected an age criterion that aimed to include women with a menstrual history that deviated from this norm. Women with a history of a hysterectomy or single oophorectomy were included and stratified in the amenorrhea group only if they had a history of prolonged amenorrhea before hysterectomy. Participants not meeting this definition were stratified as not having amenorrhea.

BMD and Laboratory Assessments

Areal BMD of the lumbar spine and total hip were measured by DXA scan using a Hologic QDR 4500 W instrument in the BC Women's Hospital and a similar machine at the Centre for Hip Health and Mobility.30 Data from the 2 instruments were adjusted to common values by a shared phantom as previously described.31 Z-scores were used to compare the BMD between age-, sex-, and ethnicity-matched cohorts. A Z-score of less than or equal to −2.0 is below the expected range for age as outlined in Canadian osteoporosis guidelines.32 Laboratory investigations included plasma FSH measured by ELISA (FSH ELISA Kit; Enzo Life Sciences, Farmingdale, NY). An FSH level of ≥25 IU/L in the setting of ongoing amenorrhea of ≥1 year was used to define menopause in this population as previously reported.5

Statistical Analysis

Demographic and clinical variables were compared between WLWH and HIV-negative women using Mann–Whitney U tests for continuous variables and Fisher exact tests for categorical variables. Significance level was α = 0.05. If data were missing for a given variable, the univariate analysis was performed among participants for whom data were available. We used linear regressions to compare hip and spine BMD Z-scores between WLWH and HIV-negative women controls and between the 2 “amenorrhea status” groups. There were too few HIV-negative women with prolonged amenorrhea to investigate whether the relationship between amenorrhea and BMD differed by HIV status (ie, an interaction effect). We also investigated associations between BMD and body mass index (BMI), HCV coinfection, and tobacco, alcohol use, and illicit drug use. Ultimately, we constructed a multivariable model, including HIV status and amenorrhea, and other variables that were associated with BMD in the univariate analysis or that were considered a priori to be associated with either BMD or amenorrhea and might moderate the relationship found; these variables in the model included BMI, illicit drug use, and HCV coinfection. Finally, a subanalysis was conducted to compare clinical and demographic characteristics of WLWH who had amenorrhea and those who did not. All analyses were performed in R version 3.5.3.

RESULTS

Participant Characteristics

A total of 258 women were included, of which 129 were WLWH and 129 were HIV-negative controls (Table 1). The flow of participants through this study is shown in Figure 1. Age and BMI (mean ± SD) were similar in both groups, with combined average of 45.7 ± 11.8 years and 26.8 ± 7.0, respectively (Table 1). Amenorrhea was significantly more common in WLWH at 21% compared with 9% in HIV-negative controls (P = 0.01). Average age of the first episode of amenorrhea was 26 years [interquartile range (IQR): 22–32]. Rates of employment and postsecondary education and income were lower among WLWH than controls. WLWH had higher rates of tobacco smoking and past or current opioid use, lower use of alcohol, and a similar rate of illicit substance use compared with HIV-negative controls (Table 1).

TABLE 1.
TABLE 1.:
Baseline Demographics of WLWH and Controls
TABLE 1-A.
TABLE 1-A.:
Baseline Demographics of WLWH and Controls

Most WLWH had HIV viral suppression <40 copies/mL (77.5%), with a median CD4 count of 560 cells/mm3. Nearly 80% of WLWH had a history of tenofovir disoproxil fumarate (TDF) exposure in the past, with a median (IQR) of 46.5 (6.8–95.0) months of TDF exposure (Table 1). Calcium supplementation was more common in WLWH than in HIV-negative controls (30% vs. 16%), but vitamin D supplementation was similar (38% vs. 30%). In WLWH, lifelong fracture prevalence was 5/129 (3.8%). Although most of these were sports or trauma related (sites: finger, ankle, knee, and humerus), one was a pelvic fracture because of a fall from a standing height at age 38. In HIV-negative women, 6/129 (4.7%) had experienced a prevalent fracture. All these were associated with trauma or sports (sites: forearm, toe, fibula, finger, and jaw).

Bone Mineral Density

Hip Z-Scores

The BMD Z-score at the total hip (Fig. 2A) was significantly lower in WLWH (−0.4 ± 0.9) vs. HIV-negative controls (0.3 ± 1.1; P < 0.001). HIV status remained significantly associated with low BMD by univariate analysis (P < 0.001) and after multivariate adjustment (P < 0.001; Table 2). Hip BMD was also significantly lower in women with a history of prolonged amenorrhea compared with those without, both univariately (P < 0.001) and after adjustment (P = 0.01). WLWH with amenorrhea had the lowest BMD at the total hip of all groups investigated (Z-score: −0.8 ± 0.9) compared with WLWH without amenorrhea (Z-score: −0.3 ± 0.8), HIV-negative controls with amenorrhea (Z-score: −0.04 ± 0.9), and HIV-negative controls without amenorrhea (Z-score: 0.3 ± 1.1; Fig. 2A). Univariate analysis suggested that HIV, prolonged amenorrhea, history of HCV, opioid use, and low BMI were all associated with lower hip BMD. When combined into a multivariate model, HIV status, BMI, and amenorrhea remained independently associated with BMD (Table 2).

FIGURE 2.
FIGURE 2.:
Scatterplots of BMD Z-scores: (A) at the total hip and (B) at the spine by HIV status and history of prolonged secondary amenorrhea. The open circles indicate the raw data, the filled circles indicate the means, and the error bars extend to the 95% confidence interval of the means. Hip: Mean (SD) in WLWH = −0.4 (±0.9), HIV-negative controls = 0.3 (±1.1), WLWH with prolonged secondary amenorrhea = −0.8 (±0.9), WLWH with no amenorrhea = −0.3 (±0.8), HIV-negative controls with prolonged secondary amenorrhea = −0.04 (±0.9), and HIV-negative controls with no amenorrhea = 0.3 (±1.1). Spine: Mean (SD) in WLWH = −0.5 (±1.3), HIV-negative controls = 0.2 (±1.3), WLWH with prolonged secondary amenorrhea = −0.6 (±1.4), WLWH with no amenorrhea = −0.4 (±1.3), HIV-negative controls with prolonged secondary amenorrhea = −0.5 (±0.6), and HIV-negative controls with no amenorrhea = 0.2 (±1.3).
TABLE 2.
TABLE 2.:
Linear Regression Model, Both Raw and Adjusted, to Examine Determinants of Hip and Spine BMD Z-Scores in All Participating Women, Including WLWH and Similar Controls

Spine Z-Scores

BMD at the spine was significantly lower among WLWH (Z-score: −0.5 ± 1.3) compared with HIV-negative controls (Z-score: 0.2 ± 1.3; P < 0.0001; Fig. 2B). There was no difference between WLWH with prolonged amenorrhea (Z-score: −0.4 ± 1.4) and those with more normal menstrual patterns (Z-score: −0.6 ± 1.4; P = 0.74; Fig. 2B). Similarly, linear regression models showed that spine BMD Z-score was significantly associated with HIV infection (P < 0.001; Table 2) but had no relationship with a history of prolonged secondary amenorrhea (P = 0.15). Univariate modeling found that illicit substance use, HCV infection, opioid use, and low BMI were also associated with lower spine BMD. When these are combined into a multivariate analysis, only BMI and HIV remained significantly associated with reduced BMD at the spine (Table 2).

WLWH With and Without Prolonged Secondary Amenorrhea

We compared the demographic and clinical characteristics of WLWH who had experienced prolonged amenorrhea and those who had not (Table 3). Among WLWH with a history of amenorrhea, 7/27 (26%) reported having current amenorrhea. Among these, only 4 had an elevated FSH (≥25 IU/L) at the time of assessment, suggestive of early menopause or primary ovarian insufficiency (Table 3). Frequencies of extremes of weight, cirrhosis, and antipsychotic use were similar between WLWH with amenorrhea and more normal menstrual patterns. WLWH with amenorrhea had significantly higher rates of HCV coinfection, tobacco use, and opioid use. Opioid use was particularly high in this cohort, with 63% of WLWH experiencing prolonged amenorrhea indicating past or present opioid use/therapy compared with 22% of those without (Table 3). Finally, when WLWH were asked about what they perceived caused their amenorrhea, 11/27 (40.7%) said it was due to drug use or methadone, 4/27 (14.8%) due to menopause, and 4/27 (14.8%) due to both of these. For the remaining 8/27, perceived causes included PCOS (n = 1), low weight (n = 1), illness (n = 1), thyroid disease (n = 1), and other factors, such as medications, relationships, and family history (n = 4; Fig. 1).

TABLE 3.
TABLE 3.:
Subgroup Comparison in WLWH of Clinical and Demographic Variables and Those With and Without a History of Prolonged Secondary Amenorrhea
TABLE 3-A.
TABLE 3-A.:
Subgroup Comparison in WLWH of Clinical and Demographic Variables and Those With and Without a History of Prolonged Secondary Amenorrhea

As previously noted, hip BMD was significantly lower in the WLWH who had experienced amenorrhea compared with those without, but spine BMD data showed no difference (P = 0.01 and 0.74, respectively). Rates of calcium and vitamin D supplementation, glucocorticoid (steroid) use, TDF exposure, and menopausal hormone therapy were not significantly different between groups (Table 3). In the WLWH who had ever experienced amenorrhea, 7/27 (25.9%) had an importantly low total hip Z-score <−2.0 compared with 7/102 (6.9%) in WLWH with normal menstrual patterns and 4/129 (3.1%) in HIV-negative controls. Fracture prevalence was similar in WLWH regardless of amenorrhea history, with a prevalence of 1/27 (4%) in WLWH with amenorrhea and 4/102 (4%) in those without.

DISCUSSION

In this cross-sectional analysis of 258 WLWH and HIV-negative controls, prolonged secondary amenorrhea was significantly associated with decreased BMD at the hip, an effect that was not seen at the spine. Our data suggest that amenorrhea may have an additive adverse effect on hip BMD in WLWH who already have decreased BMD. Among WLWH, those with amenorrhea also had higher rates of illicit substance use, chronic opioid therapy, HCV coinfection, and less effective HIV control than those with more normal menstrual patterns, suggesting that a complex interplay of variables contribute to the observed low BMD in WLWH who also have amenorrhea. Taken together, this study recognizes a subgroup of WLWH easily identified by screening of menstrual cycle history who are at increased risk for hip fracture.

This cohort is representative of WLWH in British Columbia in the cART era. It reflects the typical demographics of the population, including age, ethnic distribution, socioeconomic status, rate of illicit substance use, and HCV coinfection.33 In contrast, the rate of amenorrhea observed in this cohort of WLWH was much higher than in previous reports (21% vs. ∼5%)5,6,34 in part due to our inclusion of a lifetime experience of amenorrhea rather than solely current experience. Because amenorrhea is potentially reversible, and this is increasingly recognized among WLWH, we believe that its inclusion in our analysis enables a more complete and pragmatic evaluation of the relationship between HIV, this profound menstrual cycle disturbance, and BMD.4

Our cohort of WLWH had lower total hip and spine Z-scores than HIV-negative controls, a finding in line with previous reports.30 Among WLWH, hip BMD was further reduced in women with amenorrhea compared with those with more normal menstrual patterns, an effect that we postulate relates at least in part to dysregulated estrogen/progesterone levels. PCOS is a cause of amenorrhea that is exceptional in that it is accompanied by normal to high levels of estrogen and hence does not adversely affect bone architecture.9 We suspect that PCOS was not a common cause of amenorrhea in our cohort because PCOS more typically causes oligomenorrhea and is a rare cause of prolonged amenorrhea for ≥1 year.9 In addition, only 1 participant self-identified PCOS as the perceived cause of amenorrhea in our study. As such, we expect that low estrogen levels likely accompanied episodes of amenorrhea in most cases. The extent of diminished BMD we saw observed in our high-risk group was profound because hip BMD was 1 SD lower in WLWH with a history of amenorrhea than in HIV-negative controls. This observed difference may be clinically important because a 1 SD decrease in femoral neck BMD has been shown to increase risk of hip fracture by 2.6 times in an HIV-negative population.35 In keeping with these findings, a higher proportion of WLWH reporting amenorrhea met criteria for a low BMD (Z-score: <−2.0) than those with more normal cycles (25.9% vs. 6.9%). Although we did not see a difference in prevalent fracture rates between WLWH with a history of amenorrhea and those without, our ability to see these differences was limited by sample size and the relatively young age of our cohort; incident fractures require further study.

As a whole, the frequency with which we observed low BMD in WLWH who had ever experienced prolonged amenorrhea suggests that this group may benefit from earlier fracture risk investigation than is currently guideline recommended (ie, DXA after menopause or an incident fragility fracture).21 Prompt recognition of abnormally low BMD values may allow early initiation of effective therapy25 or preventative measures, such as avoiding HCV infection, weight loss/cycling, and undernutrition, facilitating smoking cessation, considering discontinuation of opioid therapy, and reassessment of tenofovir-based cART and usual strategies (adequate calcium, vitamin D, and weight-bearing exercise) to mitigate further BMD loss.

Interestingly, the reduced hip BMD in WLWH with a history of amenorrhea was not observed at the spine. There are a number of possible explanations for the site-specific variation in bone density. First, because we collected information on all episodes of amenorrhea, it is likely that women with resumption of menses had recovery of BMD more rapidly at the spine than at the hip. The spine is a predominantly cancellous bone, which has higher rates of bone turnover and metabolism compared with the more prominent cortical bone at the hip, and thus, it may have a faster response to hormonal changes once estradiol and progesterone levels normalize.36 Second, because the cohort with menstrual abnormalities also had higher rates of substance use, we postulate that they may also have had lower rates of healthy nutrition and weight-bearing exercise compared with the remaining group of WLWH. Finally, vitamin D deficiency is common in Canadian women and predominantly affects BMD at the hip, and hence, this may have also preferentially affected bone loss at the hip.37–39 Despite the aforementioned discrepancies in BMD at different sites, we suggest that the observed reductions in hip BMD are clinically important given that hip fractures represent the largest burden of disability and mortality associated with any osteoporotic fracture.40,41

The causes underlying increased amenorrhea among WLWH are still a matter of much debate. The pathophysiology connecting these clinical entities is clearly multifactorial and associated with “threats” of physiological, nutritional, illness, and psychosocial dimensions.3 Most women with amenorrhea in our study did not have a high FSH, suggesting that their amenorrhea was because of causes other than premature ovarian insufficiency. Of other possible contributors to amenorrhea, rates of opioid use and tobacco smoking were significantly higher among WLWH with amenorrhea compared with those with more expected menstrual patterns; both opioid and tobacco use have been associated with amenorrhea and may have increased the frequency with which we saw it in the HIV arm.42,43 Other variables associated with amenorrhea, such as cirrhosis, antipsychotic use, and extremes of BMI, were infrequent or well-matched between WLWH and HIV-negative control groups and are unlikely to be main factors in the observed difference in rates of amenorrhea. Finally, the role of immune dysfunction and HIV-related chronic inflammation on amenorrhea is not fully elucidated. Although some studies suggest an association between amenorrhea and very low CD4 counts (CD4 <35 cells/mm3), these are difficult to tease apart from other confounders, such as AIDS-associated wasting illness or concurrent illness.44,45 We observed significantly lower current CD4 counts and nadirs in WLWH with prolonged secondary amenorrhea than in those without this history; however, all these women had current CD4 counts >50 cells/mm3; we suspect low CD4 counts alone are insufficient to explain their amenorrhea. Furthermore, we could not control for unmeasured variables associated with low CD4 counts, such as physical, nutritional, and psychosocial stresses that could also have predisposed to amenorrhea.46 Because prolonged secondary amenorrhea occurs at high rates among WLWH, further characterization of its multifactorial causes and downstream consequences is paramount.

The present study should be interpreted within the context of its strengths and limitations. Its strengths are a representative North American cohort of WLWH compared with controls. It is also the first study to our knowledge that has examined BMD in WLWH with a history of prolonged amenorrhea. As a cross-sectional analysis, we are limited by the possible impact of unmeasured variables on the relationships observed and cannot draw conclusions on causation. Although we showed that the relationship between hip BMD and amenorrhea remained significant after controlling for common comorbidities in HIV, such as opioid use and HCV, we note that the influence of overlapping comorbidities on hormonal function is likely more complex than the current model. In addition, the single study site may limit the generalizability of our findings, particularly to parts of the world where comorbidities in WLWH differ from those seen in North America. Moving forward, longitudinal bone health data would offer supporting evidence on how BMD is affected by prolonged amenorrhea and its recovery. Finally, the risk for fragility hip and other fractures related to WLWH with prolonged amenorrhea is key for future investigation.

In conclusion, in this cross-sectional cohort–control study of demographically and socioculturally similar groups of WLWH and HIV-negative controls, we found lower Z-scores for both hip and spine BMD among WLWH. We further discovered that a fifth of WLWH had experienced prolonged amenorrhea that was associated with significantly and clinically importantly lower total hip BMD. These findings reinforce the importance of routine screening of WLWH for amenorrhea to allow for prompt recognition of this important reproductive disturbance so early interventions can mitigate morbidity.

REFERENCES

1. UNAIDS Data 2017 [Internet]. 2017. Available at: https://www.unaids.org/sites/default/files/media_asset/20170720_Data_book_2017_en.pdf. Accessed August 10, 2019.
2. Gordon CM, Ackerman KE, Berga SL, et al. Functional hypothalamic amenorrhea: an endocrine society clinical practice guideline. J Clin Endocrinol Metab. 2017;102:1413–1439.
3. Liu AY, Petit MA, Prior JC. Exercise and the Hypothalamus. Ovulatory Adaptations. 3rd ed. New York, NY: Springer Science: Endocrinology of Physical Activity and Sport; 2019.
4. Cejtin HE, Evans CT, Greenblatt R, et al. Prolonged amenorrhea and resumption of menses in women with HIV. J Womens Health 2018;27:1441–1448.
5. Cejtin HE, Kalinowski A, Bacchetti P, et al. Effects of human immunodeficiency virus on protracted amenorrhea and ovarian dysfunction. Obstet Gynecol. 2006;108:1423–1431.
6. King EM, Albert AY, Murray MCM. HIV and amenorrhea: a meta-analysis. AIDS Lond Engl. 2019;33:483–491.
7. Cicconi P, Ammassari A, Ladisa N, et al. Prevalence of prolonged amenorrhea in HIV-infected women: results from the Italian DIDI study. J Acquir Immune Defic Syndr. 2012;61:e19–e21.
8. Grinspoon S, Corcoran C, Miller K, et al. Body composition and endocrine function in women with acquired immunodeficiency syndrome wasting. J Clin Endocrinol Metab. 1997;82:1332–1337.
9. McBreairty LE, Zello GA, Gordon JJ, et al. Women with polycystic ovary syndrome have comparable hip bone geometry to age-matched control women. J Clin Densitom. 2018;21:54–60.
10. Karim R, Mack WJ, Kono N, et al. Gonadotropin and sex steroid levels in HIV-infected premenopausal women and their association with subclinical atherosclerosis in HIV-infected and -uninfected women in the women's interagency HIV study (WIHS). J Clin Endocrinol Metab. 2013;98:E610–E618.
11. Dolan SE, Carpenter S, Grinspoon S. Effects of weight, body composition, and testosterone on bone mineral density in. J Acquir Immune Defic Syndr. 2007;45:161–167.
12. van der Stege JG, Groen H, van Zadelhoff SJN, et al. Decreased androgen concentrations and diminished general and sexual well-being in women with premature ovarian failure. Menopause N Y N. 2008;15:23–31.
13. Sarkar M, Lai JC, Sawinski D, et al. Sex hormone levels by presence and severity of cirrhosis in women with chronic hepatitis C virus infection. J Viral Hepat. 2019;26:258–262.
14. Valdivielso JM, Jacobs-Cacha C, Soler MJ. Sex hormones and their influence on chronic kidney disease. Curr Opin Nephrol Hypertens. 2019;28:1–9.
15. Gluhovschi G, Gluhovschi A, Anastasiu D, et al. Chronic kidney disease and the involvement of estrogen hormones in its pathogenesis and progression. Rom J Intern Med. 2012;50:135–144.
16. Brown TT, Qaqish RB. Antiretroviral therapy and the prevalence of osteopenia and osteoporosis: a meta-analytic review. AIDS Lond Engl. 2006;20:2165–2174.
17. Brown TT, McComsey GA, King MS, et al. Loss of bone mineral density after antiretroviral therapy initiation, independent of antiretroviral regimen. J Acquir Immune Defic Syndr. 2009;51:554–561.
18. Erlandson KM, Lake JE, Sim M, et al. Bone mineral density declines twice as quickly among HIV-infected women compared with men. J Acquir Immune Defic Syndr. 2018;77:288–294.
19. Duvivier C, Kolta S, Assoumou L, et al. Greater decrease in bone mineral density with protease inhibitor regimens compared with nonnucleoside reverse transcriptase inhibitor regimens in HIV-1 infected naive patients. AIDS Lond Engl. 2009;23:817–824.
20. Kruger MJ, Nell TA. Bone mineral density in people living with HIV: a narrative review of the literature. AIDS Res Ther. 2017;14:35.
21. McComsey GA, Tebas P, Shane E, et al. Bone disease in HIV infection: a practical review and recommendations for HIV care providers. Clin Infect Dis. 2010;51:937–946.
22. Triant VA, Brown TT, Lee H, et al. Fracture prevalence among human immunodeficiency virus (HIV)-infected versus non-HIV-infected patients in a large U.S. healthcare system. J Clin Endocrinol Metab. 2008;93:3499–3504.
23. Collin F, Duval X, Le Moing V, et al. Ten-year incidence and risk factors of bone fractures in a cohort of treated. AIDS Lond Engl. 2009;23:1021–1024.
24. Prior J, Burdge D, Maan E, et al. Fragility fractures and bone mineral density in HIV positive women: a case-control population-based study. Osteoporos Int. 2007;18:1345–1353.
25. Prior JC. Progesterone for the prevention and treatment of osteoporosis in women. Climacteric J Int Menopause Soc. 2018;21:366–374.
26. Goshtasebi A, Berger C, Barr SI, et al. Adult premenopausal bone health related to reproductive characteristics-population-based data from the Canadian multicentre osteoporosis study (CaMos). Int J Environ Res Public Health. 2018;15:E1023.
27. Ouyang F, Wang X, Arguelles L, et al. Menstrual cycle lengths and bone mineral density: a cross-sectional, population-based study in rural Chinese women ages 30-49 years. Osteoporos Int. 2007;18:221–233.
28. Sullivan SD, Sarrel PM, Nelson LM. Hormone replacement therapy in young women with primary ovarian insufficiency and early menopause. Fertil Steril. 2016;106:1588–1599.
29. Society of Obstetricians and Gynecologists of Canada. What is menopause? 2019. Available at: https://www.menopauseandu.ca/what-is-menopause/. Accessed August 10, 2019.
30. Kalyan S, Pick N, Mai A, et al. Premature spinal bone loss in women living with HIV is associated with shorter leukocyte telomere length. Int J Environ Res Public Health 2018;15:E1018.
31. Berger C, Goltzman D, Langsetmo L, et al. Peak bone mass from longitudinal data: implications for the prevalence, pathophysiology, and diagnosis of osteoporosis. J Bone Miner Res. 2010;25:1948–1957.
32. Lentle B, Cheung AM, Hanley DA, et al. Osteoporosis Canada 2010 guidelines for the assessment of fracture risk. Can Assoc Radiol J. 2011;62:243–250.
33. Kerkerian G, Kestler M, Carter A, et al. Attrition across the HIV cascade of care among a diverse cohort of women living with HIV in Canada. J Acquir Immune Defic Syndr. 2018;79:226–236.
34. Massad LS, Evans CT, Minkoff H, et al. Effects of HIV infection and its treatment on self-reported menstrual abnormalities in women. J Womens Health. 2006;15:591–598.
35. Cummings SR, Black DM, Nevitt MC, et al. Bone density at various sites for prediction of hip fractures. The Study of Osteoporotic Fractures Research Group. Lancet Lond Engl. 1993;341:72–75.
36. Almeida M, Laurent MR, Dubois V, et al. Estrogens and androgens in skeletal physiology and pathophysiology. Physiol Rev. 2017;97:135–187.
37. Moayyeri A, Soltani A, Tabari NK, et al. Discordance in diagnosis of osteoporosis using spine and hip bone densitometry. BMC Endocr Disord. 2005;5:3.
38. Woodson G. Dual X-ray absorptiometry T-score concordance and discordance between the hip and spine measurement sites. J Clin Densitom. 2000;3:319–324.
39. Rucker D, Allan JA, Fick GH, et al. Vitamin D insufficiency in a population of healthy western Canadians. CMAJ. 2002;166:1517–1524.
40. Stone KL, Seeley DG, Lui LY, et al. BMD at multiple sites and risk of fracture of multiple types: long-term results from the Study of Osteoporotic Fractures. J Bone Miner Res. 2003;18:1947–1954.
41. Johnell O, Kanis JA. An estimate of the worldwide prevalence and disability associated with osteoporotic fractures. Osteoporos Int J. 2006;17:1726–1733.
42. Brennan MJ. The effect of opioid therapy on endocrine function. Am J Med. 2013;126(3 suppl 1):S12–S18.
43. Harlow SD, Cohen M, Ohmit SE, et al. Substance use and psychotherapeutic medications: a likely contributor to menstrual disorders in women who are seropositive for human immunodeficiency virus. Am J Obstet Gynecol. 2003;188:881–886.
44. Watts DH, Spino C, Zaborski L, et al. Comparison of gynecologic history and laboratory results in HIV-positive women with CD4+ lymphocyte counts between 200 and 500 cells/microl and below 100 cells/microl. J Acquir Immune Defic Syndr Hum Retrovirol. 1999;20:455–462.
45. Scherzer R, Bacchetti P, Messerlian G, et al. Impact of CD4+ lymphocytes and HIV infection on Anti-Mullerian Hormone levels in a large cohort of HIV-infected and HIV-uninfected women. Am J Reprod Immunol N Y N. 2015;73:273–284.
46. Fourman LT, Fazeli PK. Neuroendocrine causes of amenorrhea—an update. J Clin Endocrinol Metab. 2015;100:812–824.
Keywords:

amenorrhea; osteoporosis; HIV; bone mineral density

Supplemental Digital Content

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.