The effect of medroxyprogesterone acetate plus conjugated equine estrogens on lipoprotein(a) and apolipoprotein concentrations in postmenopausal women: a systematic review and meta-analysis of randomized controlled trials.

The primary search of literature provided 8,316 publications. Rejection of duplicate studies resulted in 5,580 articles. Then title and abstract screening identified 81 publications for further assessment. Finally, this meta-analysis ends up with 27 articles with 37 arms [ 17 , 18 , 24 – 48 ], encompassing 24 arms on ApoA1, 22 on ApoB, and lipoprotein(a) (Fig.  1 ). Fig. 1 Flow chart for study search and inclusion into the meta-analysis Flow chart for study search and inclusion into the meta-analysis The characteristics of the included studies are presented in Table  1 . The included RCTs enrolled exclusively postmenopausal female and were studied from 1993 to 2018 in different countries, including Italy, the United States of America (USA), Denmark, Korea, Spain, Canada, Turkey, Finland, Greece, Australia, New Zealand, Chile, Hong Kong, Japan, and the Netherlands. The mean age of the included subjects for each study varied between 47.3 years and 68.4 years, while the baseline BMI ranged from 20.7 kg/m² to 32.3 kg/m². MPA/CEE treatment duration ranging from 2 months to 36 months, with daily dosages varied from 1.5 to 10 mg/day for MPA and from 0.3 to 2 mg/day for CEE. The RCTs enrolled solely postmenopausal women, naturally postmenopausal or surgically-induced menopause post-hysterectomy, either healthy, symptomatic, or with hypercholesterolemia, coronary heart disease, or type 2 diabetes mellitus. The assessment of the methodological quality and risk of bias for the examined studies is detailed in Supplementary Table 2 . Table 1 Characteristics of the eligible studies Author Publications years Country Population Participants age (years) Sample size Medroxyprogesterone acetate + E2/placebo Duration Baseline BMI(kg/m 2) Outcome Intervention dose (mg) Control group Pickar, J. H. [ 24 ] 2018 USA Symptomatic postmenopausal women 54 35/36 3 24.8 Lipoprotein (a) Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg Placebo Tuomikoski, P. [ 25 ] 2010 Finland Healthy symptomatic postmenopausal women 52.2 34/37 6 22.8 ApoA-I, ApoB, Lipoprotein (a) Oral medroxyprogesterone acetate 5 mg/day + oral estradiol valerate 2 mg/day Placebo Kooperbergl, C. [ 26 ] 2007 USA Postmenopausal women 68.4 249/178 12 28.3 Lipoprotein (a) Christodoulakos, G. E. [ 27 ] 2006 Greece Healthy postmenopausal women 53.7 110/76 6 24 ApoA-I, ApoB, Lipoprotein (a) Oral medroxyprogesterone acetate 5 mg/day + conjugated equine estrogens 0.625 mg/day No treatment Lamon-Fava, S. [ 32 ] 2006 USA Healthy postmenopausal women 56.4 8/8 2 28.37 ApoA-I Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg/day Placebo Christodoulakos, G. E. [ 29 ] 2004 Greece Healthy postmenopausal women 50.3 80/45 12 24.4 ApoA-I, ApoB Oral medroxyprogesterone acetate 5 mg/day + conjugated equine estrogens 0.625 mg/day No treatment Sumino, H.(a) [ 28 ] 2004 Japan Postmenopausal women without hypercholesterolemia 53.3 43/26 12 24 ApoA-I, ApoB, Lipoprotein (a) Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg/day No treatment Sumino, H.(b) [ 28 ] 2004 Japan Postmenopausal women with hypercholesterolemia 54.6 17/14 12 26 ApoA-I, ApoB, Lipoprotein (a) Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg/day No treatment Lamon-Fava, S. [ 17 ] 2003 USA Healthy postmenopausal women 56.7 14/14 2 27.26 ApoA-I, ApoB, Lipoprotein (a) Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg/day Placebo Woo, J. [ 30 ] 2003 Hong Kong Healthy postmenopausal women 56.2 43/39 3 23.8 ApoA-I, ApoB Oral medroxyprogesterone acetate 5 mg/day for the last 14 days of each 28-day cycle + conjugated equine estrogens 0.625 mg/day No treatment Ossewaarde, M. E. [ 31 ] 2003 Netherlands Healthy postmenopausal women 58.5 34/33 3 24.3 Lipoprotein (a) Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg Placebo Ulloa, N. [ 33 ] 2002 Chile Healthy postmenopausal women 53.8 17/11 2.5 28.6 ApoA-I, ApoB Oral medroxyprogesterone acetate 5 mg/day during the last 14 days + conjugated equine estrogens 0.625 mg/day Placebo Lobo, R. A.(a) [ 18 ] 2001 USA Healthy postmenopausal women 51.5 86/94 12 24.3 ApoA-I, ApoB Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg/day Placebo Lobo, R. A.(b) [ 18 ] 2001 USA Healthy postmenopausal women 51.5 96/94 12 24.3 ApoA-I, ApoB Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.45 mg/day Placebo Lobo, R. A.(c) [ 18 ] 2001 USA Healthy postmenopausal women 51.1 94/94 12 24.7 ApoA-I, ApoB Oral medroxyprogesterone acetate 1.5 mg/day + conjugated equine estrogens 0.45 mg/day Placebo Lobo, R. A.(d) [ 18 ] 2001 USA Healthy postmenopausal women 51.3 98/94 12 24.4 ApoA-I, ApoB, Oral medroxyprogesterone acetate 1.5 mg/day + conjugated equine estrogens 0.3 mg/day Placebo Manning, P. J. [ 36 ] 2001 New Zealand Postmenopausal women with type 2 diabetes mellitus 64 48/48 6 31.6 ApoA-I, ApoB, Lipoprotein (a) Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg/day Placebo Ozsener, S. [ 35 ] 2001 Turkey Postmenopausal women with hypercholesterolemia 50.8 11/35 4 29.7 ApoA-I, ApoB, Lipoprotein (a) Oral medroxyprogesterone acetate 5 mg/day + conjugated equine estrogens 0.625 mg/day + Pravastatin 20 mg/day Pravastatin 20 mg/day Sutherland, W. H. F. [ 34 ] 2001 New Zealand Postmenopausal women with type 2 diabetes mellitus 64 28/19 6 32.3 ApoA-I, ApoB, Lipoprotein (a) Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg/day Placebo Park, J. S. [ 37 ] 2000 Korea Postmenopausal women on maintenance hemodialysis 57 33/32 12 20.7 Lipoprotein (a) Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg Placebo Darling, G. M. [ 38 ] 1999 Australia Postmenopausal women with hypercholesterolemia 61 21/23 4 27.1 Lipoprotein (a) Oral medroxyprogesterone acetate 5 mg/day + conjugated equine estrogens 0.3 mg/day Simvastatin 10 mg/day Espeland, M. A.(a) [ 42 ] 1998 USA Healthy postmenopausal women 55.8 70/68 36 - Lipoprotein (a) CEE 0.625 daily + medroxyprogesterone acetate 10 mg, days 1–12 Placebo Espeland, M. A.(b) [ 42 ] 1998 USA Healthy postmenopausal women 55.8 69/68 36 - Lipoprotein (a) CEE 0.625 daily + MPA 2.5 mg daily Placebo Meschia, M. [ 41 ] 1998 Italy Naturally postmenopausal women 51 44/19 12 24 Lipoprotein (a) Oral medroxyprogesterone acetate 10 mg/day for 12 days every 3 months + daily con No treatment Sbarouni, E.(a) [ 40 ] 1998 Greece Postmenopausal women with hypercholesterolemia and coronary artery disease 66 16/16 2 NR ApoA-I, ApoB, Lipoprotein (a) Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg/day Placebo Sbarouni, E.(b) [ 40 ] 1998 Greece Postmenopausal women with hypercholesterolemia and coronary artery disease 66 16/16 2 NR ApoA-I, ApoB, Lipoprotein (a) Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg/day + simvastatin 20 mg/day Simvastatin 20 mg/day Walsh, B. W. [ 39 ] 1998 USA Healthy postmenopausal women 60 78/88 6 26 ApoA-I, Lipoprotein (a) Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg/day Placebo Lemay, A. [ 44 ] 1995 Canada Healthy naturally postmenopausal women 52.3 15/11 24 24.7 ApoA-I, ApoB Oral medroxyprogesterone acetate 5 mg/day on days 14–25 + conjugated equine estrogens 0.625 mg/day on days 1–25 + no treatment on days 26–30 No treatment Castelo-Branco, C.(a) 1993 Spain Surgically postmenopausal women post-hysterectomy 47.9 16/19 12 NR ApoA-I, ApoB Oral medroxyprogesterone acetate 2.5 mg/day for the last 12 days of each cycle + conjugated equine estrogens 0.625 mg/day for 25 days/month No treatment Castelo-Branco, C.(b) [ 47 ] 1993 Spain Surgically postmenopausal women post-hysterectomy 48.1 15/19 12 NR ApoA-I, ApoB Oral medroxyprogesterone acetate 2.5 mg/day for the last 12 days of each cycle + conjugated equine estrogens 0.625 mg/day continuously No treatment Castelo-Branco, C.(c) [ 47 ] 1993 Spain Surgically postmenopausal women post-hysterectomy 47.3 20/19 12 NR ApoA-I, ApoB Oral medroxyprogesterone acetate 2.5 mg/day + conjugated equine estrogens 0.625 mg/day both continuously No treatment Haarbo, J. [ 48 ] 1991 Denmark Healthy early postmenopausal women 51.6 23/20 3 24.4 ApoA-I, ApoB Oral medroxyprogesterone acetate 10 mg/day from day 11 to 21 of each 28-day cycle + estradiol valerate 2 mg/day for the first 21 days of each cycle Placebo Kim, C. J.(a) [ 45 ] 1996 Korea Healthy postmenopausal women 50.9 97/71 12 23.9 Lipoprotein (a) Oral medroxyprogesterone acetate 5 mg/day on days 16–25 + conjugated equine estrogens 0.625 mg/day on days 1–25 + no treatment on days 16–30 No treatment Kim, C. J.(b) 1996 Korea Healthy postmenopausal women 51.1 109/71 12 24 Lipoprotein (a) Oral medroxyprogesterone acetate 10 mg/day on days 16–25 + conjugated equine estrogens 0.625 mg/day on days 1–25 + no treatment on days 16–30 No treatment Kim, C. J.(a) [ 45 ] 1994 Korea Healthy postmenopausal women 50.7 67/29 2 24.2 Lipoprotein (a) Oral medroxyprogesterone acetate 10 mg/day on days 16–25 + conjugated equine estrogens 0.625 mg/day on days 1–25 + no treatment on days 16–30 No treatment Kim, C. J.(b) [ 45 ] 1994 Korea Healthy postmenopausal women 50.8 65/29 2 24.2 Lipoprotein (a) Oral medroxyprogesterone acetate 5 mg/day on days 16–25 + conjugated equine estrogens 0.625 mg/day on days 1–25 + no treatment on days 16–30 No treatment Soma, M. R. [ 46 ] 1993 USA Healthy postmenopausal women 50.3 25/23 12 26.9 Lipoprotein (a) Oral medroxyprogesterone acetate 10 mg/day for 10 days per month + conjugated equine estrogens 1.25 mg/day No treatment Characteristics of the eligible studies The assessment of ApoA1 levels included 24 RCT arm analysis, encompassing 1982 subjects (1026 cases and 956 controls). With random effects model, the comprehensive analysis indicated an increase in the concentrations of ApoA1 after administration of MPA/CEE in postmenopausal female (WMD: 12.42 mg/dL, 95% CI: 9.31, 15.52, P  < 0.001). A notable heterogeneity was observed among the studies, with I² = 98% and P  < 0.001 (Fig.  2 ). Fig. 2 Forest plot of the randomized controlled trials evaluating the effects of oral medroxyprogesterone acetate plus conjugated equine estrogens administration on ApoA-1 concentrations. WMD, weighted mean difference. CI, confidence interval. ApoA-I, apolipoprotein A-I Forest plot of the randomized controlled trials evaluating the effects of oral medroxyprogesterone acetate plus conjugated equine estrogens administration on ApoA-1 concentrations. WMD, weighted mean difference. CI, confidence interval. ApoA-I, apolipoprotein A-I Subgroup analyses revealed a greater increase in ApoA1 levels in studies with doses of 2.5 mg/day of MPA (WMD: 16.58 mg/dL; 95% CI: 12.59, 20.58, P  < 0.001) compared with 5 mg/day doses (WMD: 8.07 mg/dL; 95% CI: -2.62, 18.78, P  = 0.139) and < 2.5 mg/day doses (WMD: 10.22 mg/dL; 95% CI: 5.55, 14.89, P  < 0.001), as well as in participants with a mean age ≥ 60 years (WMD: 18.96 mg/dL; 95% CI: 18.29, 19.62, P  < 0.001) when compared to a mean participant age < 60 years (WMD: 11.87 mg/dL; 95% CI: 8.35, 15.39, P  < 0.001), and RCTs with ≥ 12 months duration (WMD: 14.92 mg/dL; 95% CI: 11.05, 18.78, P  < 0.001) compared to < 12 months (WMD: 10.83 mg/dL; 95% CI: 3.24, 18.42, P  = 0.005), and participant’s mean BMI ≥ 25 kg/m² (WMD: 15.77 mg/dL; 95% CI: 12.14, 19.41, P  < 0.001) compared to < 25 kg/m² (WMD: 11.77 mg/dL; 95% CI: 7.51, 16.02, P  < 0.001) (Supplementary Fig.  1 ). The influence of MPA/CEE administration on the levels of ApoB included the assessment of 22 RCT arms with 1800 subjects (940 cases and 860 controls). The observation using the random effects model depicted a decrease in ApoB levels with MPA/CEE administration in postmenopausal female (WMD: -6.84 mg/dL; 95% CI: -8.28, -5.39, P  < 0.001). The RCTs revealed a notable heterogeneity, with I²=92% and P  < 0.001 (Fig.  3 ). Fig. 3 Forest plot of the randomized controlled trials evaluating the effects of oral medroxyprogesterone acetate plus conjugated equine estrogens administration on ApoB concentrations. WMD, weighted mean difference. CI, confidence interval. ApoB, apolipoprotein B Forest plot of the randomized controlled trials evaluating the effects of oral medroxyprogesterone acetate plus conjugated equine estrogens administration on ApoB concentrations. WMD, weighted mean difference. CI, confidence interval. ApoB, apolipoprotein B Subgroup analyses revealed a decrease in the concentrations of ApoB, appeared to be more significant in studies with doses of < 2.5 mg/day (WMD: -7.64 mg/dL; 95% CI: -10.28, -5.01, P  < 0.001) or 2.5 mg/day (WMD: -7.03 mg/dL; 95% CI: -8.37, -5.68, P  < 0.001) compared with 5 mg/day doses (WMD: -2.01 mg/dL; 95% CI: -12.08, 8.05, P  = 0.695). A greater decrease in the levels of ApoB was also observed in participants with mean BMI ≥ 5 kg/m² (WMD: -8.52 mg/dL; 95% CI: -12.64, -4.41, P  < 0.001) compared to < 25 kg/m² (WMD: -6.68 mg/dL; 95% CI: -8.27, -5.09, P  < 0.001), as well as in participants with mean age < 60 years (WMD: -6.87 mg/dL; 95% CI: -8.34, -5.40, P  < 0.001) compared to a mean participant age ≥ 60 years (WMD: -5.69 mg/dL; 95% CI: -14.52, 3.12, P  = 0.206) (Supplementary Fig.  1 ). The examination of lipoprotein(a) concentrations involved the assessment of 22 RCT arms, encompassing 2208 subjects (1238 cases and 970 controls). Random effects model analysis depicted a decrease in the levels of lipoprotein(a) after treatment with MPA/CEE in postmenopausal female (WMD: -5.12 mg/dL; 95% CI: -6.58, -3.65, P  < 0.001). A notable heterogeneity was observed among the RCTs with I²=78% and P  < 0.001 (Fig.  4 ). Fig. 4 Forest plot of the randomized controlled trials evaluating the effects of oral medroxyprogesterone acetate plus conjugated equine estrogens administration on lipoprotein (a) concentrations. WMD, weighted mean difference. CI, confidence interval Forest plot of the randomized controlled trials evaluating the effects of oral medroxyprogesterone acetate plus conjugated equine estrogens administration on lipoprotein (a) concentrations. WMD, weighted mean difference. CI, confidence interval The subgroup analysis disclosed a higher decrease in RCTs with doses of MPA < 2.5 mg/day (WMD: -5.95 mg/dL; 95% CI: -7.33, -4.57, P  < 0.001) compared with 2.5 mg/day doses (WMD: -4.91 mg/dL; 95% CI: -6.03, -3.78, P  < 0.001) and 5 mg/day (WMD: -2.14 mg/dL; 95% CI: -5.67, 1.38, P  = 0.233) and in women with a mean BMI < 25 kg/m² (WMD: -6.84 mg/dL; 95% CI: -11.05, -2.62, P  = 0.001) compared to ≥ 25 kg/m² (WMD: -4.36 mg/dL; 95% CI: -5.37, -3.36, P  < 0.001). Furthermore, a more significant change in the levels of lipoprotein(a) was noted in studies with ≥ 12 months duration (WMD: -5.83 mg/dL; 95% CI: -7.01, -4.65, P  < 0.001) compared to < 12 months (WMD: -3.48 mg/dL; 95% CI: -5.82, -1.13, P  = 0.004), as well as in mean participant age < 60 years (WMD: -5.88 mg/dL; 95% CI: -8.21, -3.55, P  < 0.001) compared to ≥ 60 years (WMD: -4.13 mg/dL; 95% CI: -5.00, -3.27, P  < 0.001) (Supplementary Fig.  1 ). Sensitivity analysis does not indicate any notable impact on the overall effect sizes by any of the arms of the RCTs (Supplementary Fig.  2 ). Examination for publication bias, performed by visually inspecting funnel plots and employing Egger’s test, did not reveal any evidence of publication bias for ApoA1 and ApoB. Although the Egger’s test P for lipoprotein(a) equals 0.07, the funnel plot suggests a possibility of bias; therefore, non-reporting bias for lipoprotein(a) cannot be excluded (Fig.  5 ). Fig. 5 Funnel plot of the weighted mean difference (WMD) versus the standard error (s.e.) of the WMD. ApoA-I, apolipoprotein A-I. ApoB, apolipoprotein B Funnel plot of the weighted mean difference (WMD) versus the standard error (s.e.) of the WMD. ApoA-I, apolipoprotein A-I. ApoB, apolipoprotein B

This meta-analysis conforms to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement guidelines [ 19 ]. Two researchers independently performed an extensive literature search in Scopus, PubMed/Medline, EMBASE, and Web of Science, for peer-reviewed articles published in English until April 2025. The aim was to determine the studies examining the impact of MPA/CEE on ApoA1, Apolipoprotein A2 (ApoA2), ApoB, and lipoprotein(a) levels in postmenopausal female. This process involved employing Medical Subject Headings (MeSH) and non-MeSH keywords combination. A comprehensive outline of our search methodology is provided in Supplementary Table 1 . We included studies that were structured as RCTs, the intervention involved the administration of MPA/CEE in comparison with control; the cohort population comprised postmenopausal females; the evaluations provided adequate data on the specific outcomes related to our review, particularly ApoA1, ApoB, and/or lipoprotein(a) levels, pre and post-intervention, for MPA/CEE and also the control cohorts, and the publication language was English. For this meta-analysis, we excluded studies lacking sufficient data on the relevant outcomes, and unpublished studies, commentaries, correspondence letters, literature reviews, concise communications, observational studies, meta-analyses, studies published in languages other than English, and studies on animal subjects. Two investigators independently extricated pertinent information, while a third superior investigator dealt with disparities. Information obtained from each RCT encompassed the publication year, primary author’s name, health condition of the participant, MPA/CEE dosage (mg/day), relevant outcomes (ApoA1, ApoB, and lipoprotein(a) before and after the intervention), number of participants, their age and BMI, and the duration of intervention. The risk of bias in the included studies was assessed using the Cochrane ROB2 tool [ 19 , 20 ]. This evaluation considered biases related to the randomization process, deviations from planned interventions, missing outcome data, outcome measurement, selective reporting, and overall bias. Each study was then categorized as having a low, some concerns, or high risk of bias. Furthermore, the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) framework was employed to evaluate the certainty of the evidence, classifying it into very low, low, moderate, or high levels [ 20 ]. STATA software version 14 was utilized to compute the aggregated difference in mean and standard deviation, examining the effect of MPA/CEE with regard to the control group for every outcome. Results were depicted as weighted mean difference (WMD) along with a 95% confidence interval (CI), determined using random effects model. In situations where such information were not available, change in mean and standard deviation were obtained from the available information [ 21 , 22 ]. A p-value of less than 0.05 was considered statistically significant for WMD. Statistical heterogeneity was assessed using I² statistic and Cochran’s Q test, values exceeding 50% and at P  < 0.10 as significant, respectively. Subgroup analyses were conducted to investigate the possible sources of heterogeneity. Sensitivity analysis was used to evaluate the magnitude of impact of decision on certain studies, and examination of publication bias was performed via inspection of funnel plots visually and the application of Egger’s tests [ 23 ].

This is the first meta-analysis of RCTs analyzing the effects of MPA/CEE on apolipoprotein and lipoprotein(a) levels in postmenopausal female. The quality of the evidence provided by our results is improved by the inclusion of solely RCTs into our study, as well as by the large number of RCTs included. Moreover, the exclusion of unpublished studies, commentaries, correspondence letters, literature reviews, concise communications, observational studies, meta-analyses maintain the high-quality evidence of our meta-analysis. Heterogeneity is inevitable in any meta-analysis, and we performed subgroup analysis to find out the heterogeneity sources. Furthermore, random effects models were employed in the quantitative analysis to estimate effect size in RCTs. However, some very old studies were included, which could have affected the study results, and significant heterogeneity was found in all the parameters, limiting the application of our results. Furthermore, publication bias cannot be discarded for our study on the effects of lipoprotein(a), meaning that the effect on lipoprotein(a) may be overestimated. Since we included healthy women and women with comorbidities in this meta-analysis, the effect of MPA/CEE on lipoprotein(a) and apolipoprotein concentrations based on the disease condition was not estimated. Only 3 of the included studies had longer intervention duration of more than a year and other studies have intervention duration of 1 year or less, which could have affected the study results of the subgroup analysis.

Our meta-analysis of RCTs shows an improvement in the levels of ApoA1 and a reduction in the concentrations of ApoB and lipoprotein(a) after MPA/CEE treatment in postmenopausal female, suggesting a beneficial change in the levels of atherogenic lipoproteins which may be correlated with a reduction in CVD risk. However, more longitudinal studies should be included in future meta-analysis, to derive at a conclusive report on long term effects of MPA/CEE on CVD.

In this meta-analysis, an increase in ApoA1 concentrations was observed subsequent to MPA/CEE administration in female. ApoA1 is a component of high-density lipoprotein cholesterol (HDL-C) particles, and hence was associated with anti-atherogenic and cardioprotective effects through its role in the pathway of reverse cholesterol transport, by facilitating the transport of surplus cholesterol to the liver for elimination [ 17 , 49 , 50 ]. Previous studies have published results similar to ours, including a case-control study reported that MPA/CEE administration resulted in an increase of ApoA1 levels [ 16 ] and an RCT demonstrated that ApoA1 was raised 27% from baseline by HRT [ 33 ]. Brinton et al. also supported the finding that HRT increased ApoA1 levels in postmenopausal women [ 51 ]. This increase in ApoA1 is thought to be the effect of estrogen, since oral estrogen increases ApoA1 synthesis in the liver. Our findings also indicate a decrease in the concentrations of ApoB post MPA/CEE administration in postmenopausal female. ApoB molecules can be found in low-density lipoprotein cholesterol (LDL-C), very low density lipoprotein cholesterol and lipoprotein(a), indicating a direct relation between ApoB levels and circulating atherogenic lipoproteins [ 52 ]. Estrogen modifies the atherosclerotic coronary artery vasomotion, and this can be proposed as a mechanism that this modulation may decreased the risk of CVD in postmenopausal women [ 53 ]. In our study, MPA/CEE administration decreased the levels of lipoprotein(a) in women, which is consistent with results from a previous study on HRT [ 14 ]; furthermore, Shlipak et al. reported that combined estrogen and progestogen therapy reduces lipoprotein(a) levels [ 54 ]. The effect of MPA/CEE in decreasing lipoprotein(a) could be due to reduction in ApoB bioavailability in the liver at the time of lipoprotein(a) assembly [ 55 ]. The recent meta-analysis with MPA alone demonstrated a significant rise in lipoprotein(a) in healthy postmenopausal female [ 56 ]. Previous studies have shown that, when compared with women with normal lipoprotein(a) levels, women with increased levels of lipoprotein(a) were more susceptible to suffer cardiac events [ 57 ], including heart failure [ 58 ], stroke, peripheral arterial disease, coronary artery disease [ 59 ], and thromboembolism [ 60 ]. This associations are due to the fact that small, dense LDL-C hampers the endothelium, increasing atherogenesis [ 61 ]. Higher lipoprotein(a) levels have also been associated with increase in age, the female sex [ 62 ], and the postmenopausal period [ 63 ]. Also, Pornel et al. concluded that combination HRT caused more decline in the levels of lipoprotein(a) after 1 year of therapy [ 64 ]. Similarly, our subgroup analysis also revealed a higher reduction in the levels of lipoprotein(a) in RCTs with ≥ 12 months duration than < 12 months; however, our subgroup analysis results revealed only a non-significant decrease. Our subgroup analysis disclosed a further decrease with < 2.5 mg/day doses compared to 2.5 mg/day and 5 mg/day doses. In contrast, another study by Lamon-fava indicated that low dose MPA does not strengthen the estrogen’s impact on lipoprotein(a) reduction [ 32 ]; however, only healthy postmenopausal women were included in Lamon-Fava’s study, while women with comorbidities were included into our meta-analysis, which could be the reason for this conflicting result. Even though our result showed the beneficial effects of MPA/CEE on apolipoproteins and lipoprotein(a), it should be clarified that it cannot be recommended as a primary therapy for dyslipidemia or for decreasing CVD risk [ 65 ], and that HRT should be tailored to the individual needs and characteristics of each patients. Qualitative assessments should be made to confirm our findings.

Medroxyprogesterone acetate (MPA) mimics the molecular structure of progesterone. The main difference is that a methyl group is present in carbon 6 and an acetate group is present in carbon 17; this different molecular structure is responsible for high progestational activity of MPA [ 1 ]. MPA is most commonly used as a contraceptive and as hormone replacement therapy (HRT) to manage menopausal manifestations, including hot flashes, insomnia, mood changes, and vaginitis; however, some other uses include the treatment of various disorders like amenorrhea, dysmenorrhea, endometriosis, and abnormal uterine bleeding [ 2 ]. Conjugated equine estrogens (CEE) are one of the estrogens commonly utilized in HRT, both as estrogen monotherapy or combined with a progestin [ 3 ]. When choosing a HRT, estrogen used to be prescribed as monotherapy in women with menopausal manifestations; however, the use of unopposed estrogens in women has been proved to increase endometrial carcinoma risk and, therefore, combination of estrogen and progestogen treatment is now suggested in women with an undamaged womb [ 4 , 5 ]. Lipid metabolism is crucial in the progression of atherosclerosis, and elevated concentrations of lipoprotein(a) along with specific apolipoproteins have been associated with a greater risk of cardiovascular disease (CVD) [ 6 ]. However, estrogen and other HRT regimes, including CEE, have demonstrated a cardioprotective effect in women in many studies [ 7 , 8 ]. The proposed mechanism involved in this cardioprotective effect is their capacity to block the formation of oxidized low density lipoproteins, reducing their serum concentrations [ 9 ], which in turn has been correlated with a decrease in the occurrence of cardiovascular events [ 10 ]. Research has also revealed that the estrogen receptors present in the coronary arteries may be another factor involved in the anti-atherogenic properties of HRT; however, aging has also been inversely correlated to the function of estrogen receptors, suggesting a reduction in the cardioprotective effects of estrogen as age increase [ 11 ]. Recent meta-analyses have shown that combined HRT with 17 beta estradiol and norethisterone acetate combination had beneficial impacts on the serum concentrations of low density lipoprotein cholesterol (LDL-C), total cholesterol (TC), triglycerides (TG) [ 12 , 13 ], lipoprotein(a) and Apolipoprotein B (ApoB) [ 14 ], as well as a reduction of CVD risk [ 15 ]. Nevertheless, studies conducted using HRT with MPA plus CEE (MPA/CEE) provided conflicting results, with studies showing a decline in the levels of TC, LDL-C, and ApoB, but an elevation in TG levels [ 16 ]. A randomized controlled trial (RCT) reported an increase in Apolipoprotein A1 (ApoA1) levels with estrogen monotherapy, with the opposite effect observed after the addition of a progestogen [ 17 ], although another RCT reported that low MPA/CEE doses caused favorable alterations in the lipids and lipoproteins [ 18 ]. Hence, to address these inconsistencies, we performed this meta-analysis to evaluate the influence of MPA/CEE on lipoprotein(a) and apolipoprotein concentrations in postmenopausal female.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2 Supplementary Material 3 Supplementary Material 3 Supplementary Material 4 Supplementary Material 4