Taulant Muka,1,2 Jana Nano,1 Loes Jaspers,1 Cindy Meun,3 Wichor M. Bramer,4
Albert Hofman,1,2 Abbas Dehghan,1 Maryam Kavousi,1 Joop S.E. Laven,2 andOscar H. Franco1
Associations of Steroid Sex Hormonesand Sex Hormone–Binding GlobulinWith the Risk of Type 2 Diabetes inWomen: A Population-Based CohortStudy and Meta-analysisDiabetes 2017;66:577–586 | DOI: 10.2337/db16-0473
It remains unclear whether endogenous sex hormones(ESH) are associated with risk of type 2 diabetes (T2D) inwomen. Data of 3,117 postmenopausal women partici-pants of the Rotterdam Study were analyzed to examinewhether ESH and sex hormone–binding globulin (SHBG)were associated with the risk of incident T2D. Addition-ally, we performed a systematic review and meta-analysisof studies assessing the prospective association of ESHand SHBG with T2D in women. During a median follow-upof 11.1 years, we identified 384 incident cases of T2D inthe Rotterdam Study. No association was observed be-tween total testosterone (TT) or bioavailable testosterone(BT) with T2D. SHBG was inversely associated with therisk of T2D, whereas total estradiol (TE) was associatedwith increased risk of T2D. Similarly, in the meta-analysisof 13 population-based prospective studies involvingmore than 1,912 incident T2D cases, low levels of SHBGand high levels of TE were associated with increased riskof T2D, whereas no associations were found for otherhormones. The association of SHBG with T2D did notchange by menopause status, whereas the associationsof ESH and T2D were based only in postmenopausalwomen. SHBG and TE are independent risk factors forthe development of T2D in women.
Menopause is an important transition in a woman’s life,not only for marking the end of reproductive life but also
for being accompanied by an increased risk of cardiovas-cular disease and type 2 diabetes (T2D) (1,2). Changes inhormonal patterns in menopause, including the decline inendogenous estradiol (E) levels and the relative androgenexcess, contribute to an increase in visceral adiposity thatis associated with glycemic traits and therefore may influ-ence the risk of T2D (3,4). Furthermore, polycystic ovarysyndrome, a common disorder among women character-ized by hyperandrogenism, has been identified as a signif-icant nonmodifiable risk factor associated with T2D (5).
Although the relation between sex hormone–bindingglobulin (SHBG) and T2D has long been recognized (6,7),literature on the associations of steroid sex hormones, suchas endogenous E and testosterone (T), with T2D is scarce.SHBG, T, and E have been associated with glucose me-tabolism and development of insulin resistance (6–9).Few epidemiological studies investigating the relationbetween sex hormones and T2D have yielded conflictingresults (10–12). These studies were limited by their cross-sectional design, selected samples, or insufficient adjust-ment for diabetes risk factors. To date, no large prospectivecohort study has examined the association of T2D withSHBG, T, and E in healthy postmenopausal women.Thus, we aimed to investigate the association betweenSHBG, sex hormones, and T2D in postmenopausal women.Furthermore, to clarify the contradictory results, we sys-tematically reviewed and meta-analyzed studies evaluating
1Department of Epidemiology, Erasmus MC, Rotterdam, the Netherlands2Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston,MA3Department of Obstetrics and Gynecology, Erasmus MC, Rotterdam, theNetherlands4Medical Library, Erasmus MC, Rotterdam, the Netherlands
the association between SHBG, sex hormones, and T2Din women.
RESEARCH DESIGN AND METHODS
The Rotterdam StudyThe Rotterdam Study is a prospective cohort study whichstarted since 1990 in the Ommoord district, in the city ofRotterdam, the Netherlands. Details regarding the design,objectives, and methods of the Rotterdam Study havebeen described in detail elsewhere (13). In brief, in 1990,all inhabitants of a well-defined district of Rotterdamwere invited, of whom 7,983 agreed (78.1%). In 2000,an additional 3,011 participants were enrolled (RS-II),consisting of all people living in the study district whohad become 55 years of age. Follow-up examinations wereperformed periodically, approximately every 3–5 years (13).There were no eligibility criteria to enter the RotterdamStudy cohorts except the minimum age and residentialarea based on ZIP codes. The Rotterdam Study has beenapproved by the medical ethics committee according to thePopulation Screening Act: Rotterdam Study, executed bythe Ministry of Health, Welfare and Sport of the Nether-lands. All participants in the present analysis providedwritten informed consent to participate and to obtaininformation from their treating physicians.
Ascertainment of T2DThe participants were followed from the date of baselinecenter visit onwards. At baseline and during follow-up, casesof T2D were ascertained through active follow-up usinggeneral practitioners’ records, glucose hospital discharge let-ters, and glucose measurements from Rotterdam Study vis-its, which take place approximately every 4 years (14). T2Dwas defined according to recent World Health Organizationguidelines, as a fasting blood glucose $7.0 mmol/L, a non-fasting blood glucose $11.1 mmol/L (when fasting sampleswere absent), or the use of blood glucose–lowering medica-tion (15). Information regarding the use of blood glucose–lowering medication was derived from both structured homeinterviews and linkage to pharmacy records (14). At base-line, .95% of the Rotterdam Study population was coveredby the pharmacies in the study area. All potential eventsof T2D were independently adjudicated by two studyphysicians. In case of disagreement, consensus was soughtwith an endocrinologist. Follow-up data were completeuntil 1 January 2012.
Sex Steroid MeasurementsAll blood samples were drawn in the morning (#11:00 A.M.)and were fasting. Total estradiol (TE) levels were mea-sured with a radioimmunoassay and SHBG with the Im-mulite platform (Diagnostic Products Corporation,Breda, the Netherlands). The minimum detection limitfor E was 18.35 pmol/L. Undetectable E was scoredas 18.35. Serum levels of total testosterone (TT) weremeasured with liquid chromatography–tandem massspectrometry. The corresponding interassay coefficientsof variations for TE, SHBG, and TT are ,7%, ,5%,
and ,5%. The free androgen index (FAI), calculated as(T/SHBG)*100, is used as a surrogate measure of bio-available testosterone (BT) (16).
Population of AnalysisThe current study used data from the third visit of thefirst cohort (RSI-3) and the baseline examinations of thesecond (RSII-1) cohort. Overall, there were 3,683 post-menopausal women eligible for blood measurements.Among them, 122 women did not come for a bloodmeasurement at the research center and 32 did not haveT2D follow-up data and were excluded from the analysis.Furthermore, 412 women with prevalent T2D wereexcluded, leaving 3,117 for our final analysis. Potentialconfounding variables are described in detail in Supple-mentary Appendix 1.
Statistical AnalysisPerson-years of follow-up were calculated from studyentrance (March 1997 to December 1999 for RSI-3 andFebruary 2000 to December 2001 for RSII-1) to the dateof diagnosis of T2D, death, or the censor date (date oflast contact of the living), whichever occurred first.Follow-up was until 1 January 2012. Cox proportionalhazards modeling was used to evaluate whether SHBG, TT,TE, and BT were associated with T2D. Relative risks (RRs)and 95% CIs were reported. All sex hormone variables wereassessed in separate models, continuously and in tertiles.For E, first tertile included all women with levels of Elower than the detection limit (n = 992). To study therelations across increasing tertiles, trend tests werecomputed by entering the categorical variables as con-tinuous variables in multivariable Cox proportional hazardsmodels. To achieve approximately normal distribution,skewed variables (SHBG, TT, BT, plasma triglyceride,LDL cholesterol [LDL-C], C-reactive protein [CRP],thyroid-stimulating hormone [TSH], and insulin) werenatural log transformed. In the base model (model 1),we adjusted for age, cohort (1,2), and fasting status (fast-ing sample vs. nonfasting sample). To examine whetherthe relations of sex hormones and SHBG with risk of T2Dwere independent of established risk factors for T2D,model 2 included the terms of model 1, BMI (continuous),glucose (continuous), and insulin (continuous). BMI andwaist circumference were highly correlated (Pearson cor-relation coefficient = 0.81, P , 0.001), so only BMI wasused as a measure of adiposity, consistent with previousstudies (10,12). Model 3 included all covariates in model2 and further potential intermediate factors, includingmetabolic risk factors (total cholesterol, systolic bloodpressure [continuous], indication for hypertension [yesvs. no], and use of lipid-lowering medications [yes vs.no]), lifestyle factors (alcohol intake [continuous] andsmoking status [current vs. former/never]), prevalent cor-onary heart disease (yes vs. no), age of menopause, hor-mone replacement therapy (yes vs. no), CRP (continuous),and sex hormones for each other. Effect modifications ofsex hormones by BMI and years since menopause were
578 Steroid Sex Hormones, SHBG, and T2D Diabetes Volume 66, March 2017
tested in the final multivariable model in addition to per-forming stratified analysis. We also performed a series ofsensitivity analyses. Since waist circumference is a bettermeasure of visceral adiposity, an important risk factor fordiabetes and of sex hormone levels after menopause, weperformed the analysis substituting it with BMI. To ac-count for the specific effects of lipid particles on diabetes,we substituted total cholesterol with HDL cholesterol,triglycerides, and LDL-C. TSH, physical activity, numberof pregnancies, and type of menopause (nonnatural vs.natural) are associated with sex hormone levels and/orrisk of T2D; therefore, the models were further adjustedfor these factors. To explore potential reverse causation,we reran the analysis by excluding the first 3 years offollow-up. Multiple imputation procedure was used (n =5 imputations) to adjust for potential bias associated withmissing data. Rubin method was used for the pooledregression coefficients (b) and 95% CI (17). A P valueof ,0.05 was considered statistically significant. All anal-yses were done using SPSS statistical software (SPSS, ver-sion 21.0; SPSS, Inc., Chicago, IL).
Systematic Review and Meta-analysisData Sources and Search Strategy. The review wasconducted using a predefined protocol and in accordancewith the Preferred Reporting Items for Systematic Reviewsand Meta-analyses (PRISMA) (18) and Meta-analysis OfObservational Studies in Epidemiology (MOOSE) (19)guidelines (Supplementary Appendices 2 and 3). Medline,Embase.com, Web of Science, the Cochrane Library,PubMed, and Google Scholar were searched from inceptionuntil 2 November 2015 (date last searched) with the assis-tance of an experienced biomedical information specialist.The computer-based searches combined terms related tothe exposure (e.g., sex hormone binding globulin, T, andE) with outcomes (e.g., T2D), without any language re-striction. Details on the search strategy are provided inSupplementary Appendix 4.Study Selection and Eligibility Criteria. Studies wereincluded if they 1) were observational cohort, case-cohort,or prospective nested case-control studies; 2) had reportedon at least one of the sex hormones as exposures (SHBG,TT, BT, TE, and bioavailable estradiol [BE]); and 3)had assessed associations with risk of T2D in women(pre- and postmenopausal). Two independent reviewersscreened the titles and abstracts of all initially identifiedstudies according to the selection criteria. Full texts wereretrieved from studies that satisfied all selection criteria.Data extraction, quality assessment, and data synthesisand analysis are described in detail in SupplementaryAppendix 5.
RESULTS
Table 1 summarizes the baseline characteristics of the par-ticipants included in the analysis. Of the 3,117 postmeno-pausal women without diabetes at baseline, 384 womendeveloped diabetes over a median follow-up of 11.1 years.
Sex Hormones and the Risk of Developing T2DIn models adjusted for age, cohort effect, and fastingstatus, lower SHBG levels (third vs. first tertile: RR 0.33[95% CI 0.25–0.43], P trend ,0.001) and higher levelsof BT (third vs. first tertile: RR 2.01 [95% CI 1.55–2.60],P trend ,0.001) and TE (third vs. first tertile: RR 2.02[95% CI 1.50–2.70], P trend ,0.001) were associated withan increased risk of T2D (Table 2). Further adjustmentsfor BMI, insulin, and glucose attenuated but did not abol-ish the association between SHBG (third vs. first tertile:RR 0.56 [95% CI 0.41–0.77], P trend ,0.001) or TE andincident T2D (third vs. first tertile: RR 1.39 [95% CI1.004–1.93], P trend = 0.07). On the other hand, adjust-ment for obesity and glycemic traits weakened the associ-ations of BT with T2D such that they were no longerstatistically significant (Table 2). Controlling for metabolic
Table 1—Selected characteristics of study participants, theRotterdam Study
Women(n = 3,117)
%missingvalues
Age (years) 69.7 6 8.7 0
Years since menopause (years) 20.9 6 10.0 4.4
Age of menopause (years) 48.9 6 5.2 4.4
Number of pregnancies of at least6 months 2.3 6 2 12.4
Natural menopause, n (%) 2,433 (78.1) 0
Current smokers, n (%) 218 (9.2) 1.8
Alcohol intake (g/day) 1.3 (10)a 26.5
BMI (kg/m2) 27.0 6 4.3 2.3
Waist circumference (cm) 89.4 6 11.6 5.8
Prevalent coronary heartdisease, n (%) 86 (2.8) 0.06
E (pmol/L) 34.2 (41.62)a 0
TT (nmol/L) 0.8 (0.56)a 0
SHBG (nmol/L) 69.6 6 33.0 0
FAI 1.3 (1.1)a 0
TSH (mU/L) 1.95 (1.7)a 0.03
Hormone replacement therapy, n (%) 159 (5.3) 4.8
Insulin (pmol/L) 67 (47)a 0.26
Glucose (mmol/L) 5.5 6 0.6 1.3
CRP (mg/mL) 1.7 (2.93)a 3.7
Total cholesterol (mmol/L) 6.0 6 1.0 1.3
LDL-C (mmol/L) 4.2 (1.22)a 2.5
HDL cholesterol (mmol/L) 1.5 6 0.4 2.3
Statin use, n (%) 681 (14) 4.8
Triglycerides (mmol/L) 1.27 (0.74)a 0.26
Systolic blood pressure (mmHg) 142.0 6 21.1 1.03
Indication for hypertension, n (%) 794 (25.5) 1.03
Incident T2D, n (%) 384 (12.3) 0
Plus/minus values are mean6 SD. aMedian (interquartile range).
diabetes.diabetesjournals.org Muka and Associates 579
risk factors, lifestyle factors, inflammatory markers, and prev-alent coronary heart disease did not materially affect theseassociations (Table 2). No association was found between TTand incident T2D in any of the models (Table 2).
Because associations of continuous hormone variableswith T2D in model 1 appeared linear, RRs stratifiedand sensitivity analyses were expressed per unit log orunit increase in hormone biomarkers. In the sensitivityanalyses, substituting BMI with waist circumference as ameasure of adiposity, substituting total cholesterol forother blood lipids, adjusting further for serum TSH,physical activity, number of pregnancies of at least6 months, or menopause type, and excluding the first3 years of follow-up did not affect any of the associations(Supplementary Table 1). Also, in the stratification anal-ysis, no significant interactions were found for SHBGand TE with BMI or years since menopause (Supplemen-tary Table 1). Significant interaction terms were foundfor TT (P interaction = 0.019) and FAI (P interaction =0.03) with years since menopause. However, no associa-tion was found between these hormones and T2D afterstratification for time since menopause (Supplementary
Table 1). Also, no effect modification by BMI was foundfor TT and BT (Supplementary Table 1).
Systematic Review and Meta-analysis
Literature Search, Characteristics, and Quality ofEligible StudiesThe initial search identified 3,209 potentially relevantcitations. After screening and detailed assessment, 15 ar-ticles based on 12 unique studies were included (Supple-mentary Fig. 1 and Supplementary Appendix 5). Therefore,we meta-analyzed estimates from 13 studies (including thecurrent study) involving a total of 14,902 pre- and post-menopausal women with 1,912 incident T2D cases, report-ing on the association between sex hormones and T2D risk.Detailed characteristics of these studies and quality assess-ment have been summarized in Supplementary Table 2. Allstudies were medium to high quality except one.
Sex Hormones and T2D in Pooled AnalysisThe meta-analyses for BT, TE, and BE are based only onstudies examining postmenopausal women; the meta-analysisfor TT is based on four studies including postmenopausal
Table 2—Associations of SHBG, TT, FAI, and TE with the risk of T2D in postmenopausal women, the Rotterdam Study(n = 3,117)
Significant association (P , 0.05) indicated by boldface type. Model 1: adjusted for age, cohort, fasting status; model 2: model 1 +insulin, glucose, and BMI; model 3: model 2 + alcohol intake, smoking status, coronary heart disease, serum total cholesterol, statin use, systolicblood pressure, treatment for hypertension, hormone replacement therapy, age of menopause, CRP, and sex hormones for each other.
580 Steroid Sex Hormones, SHBG, and T2D Diabetes Volume 66, March 2017
women and one study including pre- and postmenopausalwomen, whereas the findings for SHBG derive from studiesincluding premenopausal women (n = 2), postmenopausalwomen (n = 4), and combined (n = 3). The pooled RRs forT2D adjusted for several metabolic risk factors comparingthird versus first tertile of SHBG, TT, BT, TE, and BE were0.44 (95% CI 0.30–0.66, I2 = 77.9%, P, 0.001), 1.32 (95%CI 0.79–2.21, I2 = 53.8%, P = 0.07), 1.75 (95% CI 0.92–3.33, I2 = 80.7%, P = 0.001), 1.99 (95% CI 1.21–3.27, I2 =55.1%, P = 0.06), and 3.58 (95% CI 0.86–14.84, I2 = 81.0%,P = 0.02) (Figs. 1–3). There was evidence of between-studyheterogeneity for all these analyses, with the possible ex-ception of the meta-analysis on the association between TEand the risk of T2D (Figs. 1–3). For SHBG, heterogeneitywas not explained by any of the study-level characteristicsassessed, such as menopause status, location, and numberof participants (Supplementary Table 4). For TT, the levelof heterogeneity was largely explained by location (Supple-mentary Table 4). Five studies could not be included in themeta-analyses. Soriguer found that in pre- and postmeno-pausal women, per one unit log increase in SHBG, TT, andBT, the corresponding RRs were 0.23 (95% CI 0.1–0.53),1.04 (0.59–1.83), and 1.12 (0.59–2.13), respectively (20).Boyd-Woschinko et al. (21) reported a fivefold increase in T2Dincidence in the lowest quintile of SHBG. Similarly, Lindstedtet al. (22) found that among patients in the low SHBGtertile, 18% converted to T2D as compared with 5% in
the mid SHBG tertile and 2.5% in the high SHBG tertile.Okubo et al. (23) reported lower levels of SHBG in T2Dconverters (59.7 6 8.4 nmol/L) than nonconverters(69.5 6 2.5 nmol/L) during 3 years of follow-up, but thatwas not significantly different after adjusting for age, BMI,and waist-to-hip ratio. Sex steroids and SHBG were notassociated with diabetes outcomes in pre- and postmeno-pausal women in the study of Mather et al. (24).
Publication BiasThe appearance of funnel plots was asymmetrical for theanalysis on SHBG and T2D, and Egger test results weresignificant (P = 0.014) (Supplementary Fig. 2). This sug-gested that publication bias may be present. After exclu-sion of the four studies that included 50 or fewer casesubjects with T2D, findings were not statistically signifi-cant (Egger test, P = 0.93, data not shown). No evidenceof publication bias was observed for the analysis of TT orTE and T2D (Supplementary Fig. 2).
DISCUSSION
In this large population–based study of postmenopausalwomen free of T2D at baseline, we showed that that lowerlevels of SHBG and higher levels of TE were associatedwith the risk of T2D, independent of established riskfactors for T2D, including BMI, glucose, and insulin. Incontrast, the association between T and the risk of T2D
Figure 1—RRs of T2D comparing top vs. bottom thirds of baseline plasma SHBG. The summary estimates presented were calculatedusing random-effects models (D+L) and fixed effects (I-V). The sizes of the data markers are proportional to the inverse of the variance ofthe odds ratio; the CIs are represented by the bars. X2 = 36.2. I2 = 77.9%. P < 0.001.
diabetes.diabetesjournals.org Muka and Associates 581
Figure 2—RRs of T2D comparing top vs. bottom thirds of baseline plasma TT and FT levels. The summary estimates presented werecalculated using random-effects models (D+L) and fixed effects (I-V). The sizes of the data markers are proportional to the inverse of thevariance of the odds ratio; the CIs are represented by the bars. A: X2 = 8.6. I2 = 53.8%. P = 0.07. B: X2 = 15.5. I2 = 80.7%. P = 0.001.
582 Steroid Sex Hormones, SHBG, and T2D Diabetes Volume 66, March 2017
Figure 3—RRs of T2D comparing top vs. bottom thirds of baseline plasma TE and free E levels. The summary estimates presented werecalculated using random-effects models (D+L) and fixed effects (I-V). The sizes of the data markers are proportional to the inverse of thevariance of the odds ratio; the CIs are represented by the bars. A: X2 = 8.91. I2 = 55.1%. P = 0.06. B: X2 = 5.26. I2 = 81.0%. P = 0.02.
diabetes.diabetesjournals.org Muka and Associates 583
was explained by BMI, glucose, and insulin. Pooled resultsfrom the systematic meta-analysis of 13 studies reinforcethe validity and generalizability of our findings, suggestingthat SHBG and TE are robust risk markers of T2D inwomen.
Unlike the previous meta-analysis by Ding et al. (25),which was based mainly on studies with cross-sectional de-sign and examined only mean differences between case sub-jects with T2D and control subjects without T2D, our currentpooled analysis is based on the findings from 13 prospectivestudies (only 2 studies included in the previous review wereeligible), including 14,902 participants with 1,912 case sub-jects with T2D. Therefore, our meta-analysis provides a moredetailed assessment of the nature and magnitude of theassociation between sex hormones and T2D in women.
SHBG levels have been associated with metabolicsyndrome, glucose, and insulin levels, established riskfactors for T2D (7,8,26). Also, women with polycysticovary syndrome, a condition of anovulation and hyper-androgenism, are at increased risk of T2D, and levels ofSHBG are decreased in these women (27). The complexbiological mechanisms that explain the association be-tween circulating SHBG levels and the risk for T2D arenot fully understood. Classically, the primary function ofSHBG was thought to be the binding of circulating hor-mones in order to regulate free sex hormone bioavailabil-ity to target tissues. Therefore, it has been hypothesizedthat the relation between SHBG and T2D may result fromthe indirect influence of alterations in SHBG on sex hor-mone bioavailability. However, in our study, the associa-tion between SHBG and T2D risk remains significant afteradjustment for TT, BE, and TE, implicating SHBG levels asa risk factor for T2D independent of serum androgenlevels. Additional evidence in support of an independenteffect of SHBG on T2D comes from recent studies thathave found several polymorphisms in the SHBG to asso-ciate with insulin resistance and T2D, suggesting thataltered SHBG physiology may be a primary defect in thepathogenesis of disease (28–31). Furthermore, a growingbody of evidence shows that SHBG may directly mediatecell-surface signaling, cellular delivery, and biological ac-tion of sex hormones via activation of a specific plasmareceptor (32–34). At the target tissue level, the fractionof SHBG that is not bound to sex steroid has the abilityto bind plasma membrane high-affinity receptors (RSHBG)(32). Sex steroids of variable biological potency can activatethe anchored SHBG-RSHBG complex, and the activatedcomplex can have either an agonist or antagonist effect.For example, SHBG-RSHBG complex can have direct cellu-lar antagonistic properties against estrogen; SHBG mayinteract with cellular estrogen receptors, which can triggera biological antiestrogenic response (32). Specific down-stream effects of the SHBG-receptor complex merit fur-ther investigation since they may help to clarify theunderlying mechanisms linking SHBG to T2D.
Our result for a positive relation between E and T2Dis in contrast with the results from previous large
randomized control trials of oral estrogen therapy, whichshowed a lower risk of T2D among postmenopausalwomen assigned to estrogen treatment (35–37). However,due to the observational design, our study does not pro-vide causality. Mendelian randomization experiments arewarranted to investigate the potential causal implicationsof E on T2D. Exogenous estrogen may have differentphysiological effects depending on type, route, duration,and dose of estrogen therapy (38–41). For example, op-posing effects of oral estrogen on fasting glucose versusglucose tolerance have been reported (38,39). Also, in arandomized trial of postmenopausal women, oral estro-gen elevated CRP levels up to 12 months of treatment butnot transdermal E (40). Moreover, a bimodal relationshipof estrogen dose may exist. In a clinical trial of postmen-opausal women, a lower dose of estrogen therapy increasedinsulin sensitivity whereas a higher dose had the oppositeeffect (42).
In postmenopausal women, endogenous E may beassociated with diabetes risk through its relation toglucose, insulin, obesity, and inflammation. Indeed, pre-vious cross-sectional studies have linked both BE and TEwith higher glucose and insulin resistance levels in post-menopausal women, independent of obesity (6,9,43). Also,whereas animal studies suggest that E regulates bodycomposition, many studies in postmenopausal womenhave failed to show a consistent beneficial role of E inweight loss and in the distribution of body fat (44). How-ever, in our study, the association between TE and T2D,although attenuated, remained significant after adjust-ment for plasma levels of glucose and insulin, BMI, andCRP, suggesting that E may play a direct role in thepathophysiology of T2D in postmenopausal women. Fur-thermore, additional adjustment for TT did not affectthis association, suggesting that E may be more thanjust a marker of increased aromatase conversion. Ex-plicit mechanisms of estrogen in relation to T2D requirefurther study.
Our study showed no association between TT and therisk of T2D, whereas a suggestive positive association wasobserved between BT and T2D. The lack of associationbetween free testosterone (FT) and the risk of T2D in ourstudy might be due to the lack of a direct measure of BTin the blood, which could have biased our results towardthe null. These findings are in line with previous studiesreporting higher levels of insulin resistance with in-creasing levels of BT in postmenopausal women, whereasno association has been observed between TT and insulinresistance (6,14). Similarly, BT has been related to in-creased odds of having impaired fasting glucose (14).
The strengths of our study include its prospectivedesign, the long follow-up, and adequate adjustments fora broad range of possible confounders. We also performedseveral sensitivity analyses, such as excluding the first3 years of follow-up to avoid potential bias of undiagnoseddisease at baseline. Furthermore, our study included, inaddition to an analysis of primary data, a systematic review
584 Steroid Sex Hormones, SHBG, and T2D Diabetes Volume 66, March 2017
of all available published prospective cohorts, which is thefirst-ever quantitative synthesis of these associations thusfar in women. Also, most of the studies included in ourmeta-analysis adjusted for potential confounding. How-ever, there are several limitations that need to be takeninto account. First, we did not have measures of BE inthe Rotterdam Study, which could have strengthenedour results. Also, TE was measured using an immuno-assay with a detection limit of 18.35 pmol/L, which isconsidered suboptimal, particularly in postmenopausalwomen. However, the observed effect remained thesame while analyzing TE continuously and categorically.Second, FT levels were not measured directly in theblood and therefore have to be interpreted with caution.Nevertheless, FT levels in this study were derived fromthe ratio of T to SHBG, which is considered a preciseproxy for BT (45). Third, we observed a moderate to highlevel of heterogeneity across the included studies. Dif-ferent assays (Supplementary Table 3) used to assess thelevels of sex hormones and SHBG contributed to theobserved heterogeneity. However, since the number ofavailable studies included in each meta-analysis was gen-erally small, it precluded our ability to conduct subgroupanalyses involving various study-level characteristics(such as age). Fourth, there was evidence of publicationbias for the association between SHBG and the risk ofT2D, so it is possible that our results constitute an over-estimation of the performance of the test. However,when we excluded small studies, differences were notstatistically significant, and therefore, the effect of pub-lication bias may be only minor. Fifth, except for SHBG,the other findings come from studies conducted mainlyin postmenopausal women, and thus, these results can-not be extended to pre- or perimenopausal women. Fi-nally, contrary to the results of random-effects models,the fixed-effects models showed a significant associationof both BE and BT with the risk of T2D. The differencesin random- versus fixed-effects models might be explainedby the substantial heterogeneity observed between studies(for example, in the association of BT), which could bebetter captured under the random-effects model (46). ForBE, the small size of the studies might undermine theprecision of the estimate under a fixed-effects model. How-ever, in light of these observations, the overall results ofthis study should be interpreted with caution.
In conclusion, lower levels of SHBG and higher levelsof TE are independently associated with risk of T2D inpostmenopausal women. Further studies are needed toestablish hormone thresholds at which diabetes risk isincreased, because this may aid in identifying high-riskpostmenopausal women in the clinical setting.
Acknowledgments. The authors thank Dr. Wanes Kazanjian (ErasmusMC) for reviewing the abstract.Funding. This study was sponsored and funded by Metagenics Inc.
Metagenics Inc. had no role in the design or conduct of the study; collection,management, analysis, or interpretation of the data; or preparation, review, or
approval of the manuscript. The funder/sponsor did not have the ability to vetothe publication of study results.Duality of Interest. T.M., L.J., and O.H.F. work at ErasmusAGE, a centerfor aging research across the life course funded by Nestlé Nutrition (Nestec Ltd.),Metagenics Inc., and AXA. T.M. and L.J. reported receiving research support fromMetagenics Inc. J.N. has been financially supported by Erasmus Mundus WesternBalkans (ERAWEB), a project funded by the European Commission. M.K. issupported by the AXA Research Fund. O.H.F. reported receiving grants or re-search support from Metagenics Inc. No other potential conflicts of interestrelevant to this article were reported.
These funding sources had no role in the design or conduct of the study;collection, management, analysis, or interpretation of the data; or preparation,review, or approval of the manuscript.Author Contributions. T.M. conceived and designed the study, ran theanalysis, screened the title and abstract, obtained the full text, determined theeligibility of articles, participated in data extraction, participated in data synthesis,analysis, and interpretation, drafted the final manuscript, and contributed to thecritical revision of the manuscript and approved the final version. J.N. screenedthe title and abstract, obtained the full text, determined the eligibility of articles,participated in data extraction, and contributed to the critical revision of the man-uscript and approved the final version. L.J., C.M., A.H., A.D., M.K., and J.S.E.L.contributed to the critical revision of the manuscript and approved the final version.W.M.B. designed and executed the search strategies and contributed to the criticalrevision of the manuscript and approved the final version. O.H.F. conceived anddesigned the study, drafted the final manuscript, and contributed to the criticalrevision of the manuscript and approved the final version. T.M. is the guarantor ofthis work and, as such, had full access to all the data in the study and takesresponsibility for the integrity of the data and the accuracy of the data analysis.
References1. Atsma F, Bartelink ML, Grobbee DE, van der Schouw YT. Postmenopausalstatus and early menopause as independent risk factors for cardiovasculardisease: a meta-analysis. Menopause 2006;13:265–2792. Gambacciani M, Ciaponi M, Cappagli B, et al. Body weight, body fat dis-tribution, and hormonal replacement therapy in early postmenopausal women.J Clin Endocrinol Metab 1997;82:414–4173. Liu Y, Ding J, Bush TL, et al. Relative androgen excess and increasedcardiovascular risk after menopause: a hypothesized relation. Am J Epidemiol2001;154:489–4944. Brand JS, van der Schouw YT, Onland-Moret NC, et al. Age at menopause,reproductive life span, and type 2 diabetes risk: results from the EPIC-InterActstudy. Diabetes Care 2013;36:1012–10195. Gambineri A, Patton L, Altieri P, et al. Polycystic ovary syndrome is a riskfactor for type 2 diabetes: results from a long-term prospective study. Diabetes2012;61:2369–23746. Kalish GM, Barrett-Connor E, Laughlin GA, Gulanski BI; PostmenopausalEstrogen/Progestin Intervention Trial. Association of endogenous sex hormonesand insulin resistance among postmenopausal women: results from the Post-menopausal Estrogen/Progestin Intervention Trial. J Clin Endocrinol Metab 2003;88:1646–16527. Brand JS, van der Schouw YT. Testosterone, SHBG and cardiovascularhealth in postmenopausal women. Int J Impot Res 2010;22:91–1048. Brand JS, van der Tweel I, Grobbee DE, Emmelot-Vonk MH, van der SchouwYT. Testosterone, sex hormone-binding globulin and the metabolic syndrome: asystematic review and meta-analysis of observational studies. Int J Epidemiol2011;40:189–2079. Golden SH, Dobs AS, Vaidya D, et al. Endogenous sex hormones andglucose tolerance status in postmenopausal women. J Clin Endocrinol Metab2007;92:1289–129510. Ding EL, Song Y, Manson JE, Rifai N, Buring JE, Liu S. Plasma sex steroidhormones and risk of developing type 2 diabetes in women: a prospective study.Diabetologia 2007;50:2076–2084
diabetes.diabetesjournals.org Muka and Associates 585
11. Oh JY, Barrett-Connor E, Wedick NM, Wingard DL; Rancho Bernardo Study.Endogenous sex hormones and the development of type 2 diabetes in older menand women: the Rancho Bernardo study. Diabetes Care 2002;25:55–6012. Kalyani RR, Franco M, Dobs AS, et al. The association of endogenous sexhormones, adiposity, and insulin resistance with incident diabetes in post-menopausal women. J Clin Endocrinol Metab 2009;94:4127–413513. Hofman A, Darwish Murad S, van Duijn CM, et al. The Rotterdam Study:2014 objectives and design update. Eur J Epidemiol 2013;28:889–92614. Leening MJG, Kavousi M, Heeringa J, et al. Methods of data collection anddefinitions of cardiac outcomes in the Rotterdam Study. Eur J Epidemiol 2012;27:173–18515. World Health Organization. Definition and Diagnosis of Diabetes Mellitus andIntermediate Hyperglycemia: Report of a WHO/IDF Consultation. Geneva, WorldHealth Org., 2006, p. 1–5016. Rosner W, Auchus RJ, Azziz R, Sluss PM, Raff H. Position statement: utility,limitations, and pitfalls in measuring testosterone: an Endocrine Society positionstatement. J Clin Endocrinol Metab 2007;92:405–41317. Rubin BD. Multiple imputation for nonresponse in surveys. Invest Radiol198718. Rubin RB. Multiple Imputation for Nonresponse in Surveys. New York, JohnWiley & Sons, 198719. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observationalstudies in epidemiology: a proposal for reporting. Meta-analysis Of ObservationalStudies in Epidemiology (MOOSE) group. JAMA 2000;283:2008–201220. Soriguer F, Rubio-Martin E, Fernandez D, Valdes S, et al. Testosterone,SHBG and risk of type 2 diabetes in the second evaluation of the Pizarra cohortstudy. Eur J Invest 2012;42:79–8521. Boyd-Woschinko G, Kushner H, Falkner B. Androgen excess is associatedwith insulin resistance and the development of diabetes in African Americanwomen. J Cardiometab Syndr 2007;2:254–25922. Lindstedt G, Lundberg PA, Lapidus L, Lundgren H, Bengtsson C, Bjorntorp P.Low sex-hormone-binding globulin concentration as independent risk factor fordevelopment of NIDDM. 12-yr follow-up of population study of women inGothenburg, Sweden. Diabetes 1991;40:123–12823. Okubo M, Tokui M, Egusa G, Yamakido M. Association of sex hormone-binding globulin and insulin resistance among Japanese-American subjects.Diabetes Res Clin Pract 2000;47:71–7524. Mather KJKC, Kim C, Christophi CA, et al.; Diabetes Prevention Program.Steroid sex hormones, sex hormone-binding globulin, and diabetes incidencein the Diabetes Prevention Program. J Clin Endocrinol Metab 2015;100:3778–378625. Ding EL, Song Y, Malik VS, Liu S. Sex differences of endogenous sexhormones and risk of type 2 diabetes: a systematic review and meta-analysis.JAMA 2006;295:1288–129926. Weinberg ME, Manson JE, Buring JE, et al. Low sex hormone-bindingglobulin is associated with the metabolic syndrome in postmenopausal women.Metabolism 2006;55:1473–148027. Nestler JE, Powers LP, Matt DW, et al. A direct effect of hyperinsulinemia onserum sex hormone-binding globulin levels in obese women with the polycysticovary syndrome. J Clin Endocrinol Metab 1991;72:83–8928. Ding EL, Song Y, Manson JE, et al. Sex hormone-binding globulin and riskof type 2 diabetes in women and men. N Engl J Med 2009;361:1152–116329. Wang Q, Kangas AJ, Soininen P, et al. Sex hormone-binding globulinassociations with circulating lipids and metabolites and the risk for type 2 di-abetes: observational and causal effect estimates. Int J Epidemiol 2015;44:623–637
30. Zhao JL, Chen ZJ, Zhao YR, et al. Study on the (TAAAA)n repeat poly-
morphism in sex hormone-binding globulin gene and the SHBG serum levels in
putative association with the glucose metabolic status of Chinese patients suf-
fering from polycystic ovarian syndrome in Shandong province. Zhonghua Yi Xue
Yi Chuan Xue Za Zhi 2005;22:644–647 [in Chinese]31. Perry JR, Weedon MN, Langenberg C, et al.; MAGIC. Genetic evidence that
raised sex hormone binding globulin (SHBG) levels reduce the risk of type 2 di-
abetes. Hum Mol Genet 2010;19:535–54432. Rosner W, Hryb DJ, Khan MS, Nakhla AM, Romas NA. Sex hormone-binding
globulin mediates steroid hormone signal transduction at the plasma membrane.
J Steroid Biochem Mol Biol 1999;69:481–48533. Catalano MG, Frairia R, Boccuzzi G, Fortunati N. Sex hormone-binding
globulin antagonizes the anti-apoptotic effect of estradiol in breast cancer cells.
Mol Cell Endocrinol 2005;230:31–3734. Fortunati N, Catalano MG, Boccuzzi G, Frairia R. Sex hormone-binding globulin
(SHBG), estradiol and breast cancer. Mol Cell Endocrinol 2010;316:86–9235. Kanaya AM, Herrington D, Vittinghoff E, et al.; Heart and Estrogen/progestin
Replacement Study. Glycemic effects of postmenopausal hormone therapy: the
Heart and Estrogen/Progestin Replacement Study. A randomized, double-blind,
placebo-controlled trial. Ann Intern Med 2003;138:1–936. Margolis KL, Bonds DE, Rodabough RJ, et al.; Women’s Health Initiative
Investigators. Effect of oestrogen plus progestin on the incidence of diabetes in
postmenopausal women: results from the Women’s Health Initiative Hormone
Trial. Diabetologia 2004;47:1175–118737. Bonds DE, Lasser N, Qi L, et al. The effect of conjugated equine oestrogen
on diabetes incidence: the Women’s Health Initiative randomised trial. Dia-
betologia 2006;49:459–46838. Espeland MA, Hogan PE, Fineberg SE, et al. Effect of postmenopausal
hormone therapy on glucose and insulin concentrations. PEPI Investigators.
Postmenopausal Estrogen/Progestin Interventions. Diabetes Care 1998;21:1589–
159539. Zhang Y, Howard BV, Cowan LD, et al. The effect of estrogen use on levels
of glucose and insulin and the risk of type 2 diabetes in American Indian post-
menopausal women: the Strong Heart Study. Diabetes Care 2002;25:500–50440. Strandberg TE, Ylikorkala O, Tikkanen MJ. Differing effects of oral and
transdermal hormone replacement therapy on cardiovascular risk factors in
healthy postmenopausal women. Am J Cardiol 2003;92:212–21441. Decensi A, Omodei U, Robertson C, et al. Effect of transdermal estradiol and
oral conjugated estrogen on C-reactive protein in retinoid-placebo trial in healthy
possible bimodal effect of estrogen on insulin sensitivity in postmenopausal women
and the attenuating effect of added progestin. Fertil Steril 1993;60:664–66743. Goodman-Gruen D, Barrett-Connor E. Sex differences in the association of
endogenous sex hormone levels and glucose tolerance status in older men and
women. Diabetes Care 2000;23:912–91844. Davis SR, Castelo-Branco C, Chedraui P, et al.; Writing Group of the In-
ternational Menopause Society for World Menopause Day 2012. Understanding
weight gain at menopause. Climacteric 2012;15:419–42945. Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple
methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab
1999;84:3666–367246. Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. Introduction to Meta-Analysis. West Sussex, U.K., Wiley, 2009
586 Steroid Sex Hormones, SHBG, and T2D Diabetes Volume 66, March 2017
![The application of Reporter Gene Assays for the … et al...the female sex hormones, are responsible for growth, development and function ofthe female reproductive system[1,2]. Androgens,](https://static.documents.pub/doc/80x56/5f7ae2387dafaa6ff2071582/the-application-of-reporter-gene-assays-for-the-et-al-the-female-sex-hormones.jpg)
