The e ﬀ ect of hormonal contraceptive therapy on clinical laboratory parameters: a literature review

: Hormonal contraceptives (HC) are widely used among women in reproductive ages. In this review, the effects of HCs on 91 routine chemistry tests, metabolic tests, and tests for liver function, hemostatic system, renal function, hormones, vitamins and minerals were evaluated. Test parameters were di ﬀ erently a ﬀ ected by the dosage, duration, composition of HCs and route of administration. Most studies concerned the e ﬀ ects of combined oral contraceptives (COC) on the metabolic, hemostatic and (sex) steroids test re-sults. Although the majority of the e ﬀ ects were minor, a major increase was seen in angiotensinogen levels (90 – 375 %) and the concentrations of the binding proteins (SHBG [ ∼ 200 %], CBG [ ∼ 100 %], TBG [ ∼ 90 %], VDBP [ ∼ 30 %], and IGFBPs [ ∼ 40 %]). Also, there were signi ﬁ cant changes in levels of their bound molecules (testosterone, T3, T4, cortisol, vitamin D, IGF1 and GH). Data about the e ﬀ ects of all kinds of HCs on all test results are limitedand sometimes inconclusive due to thelarge variety in HC, administration routes and dosages. Still, it can be concluded that HC use in women mainly stimulates the liver production of binding proteins. All biochemical test results of women using HC should be assessed carefully and unexpected testresultsshouldbefurtherevaluatedforbothmethodological and pre-analytical reasons. As HCs change over time, future studies are needed to learn more about the e ﬀ ects of other types


Introduction
The importance of the oral contraceptive pill, also known as "the pill," cannot be overstated.It was developed at the beginning of the 1960s, and for the first time in human history, women had the power to regulate their own fertility.The pill was referred to as "one of the seven wonders of the modern world" and "the one invention that historians will look back on and say: That characterized the 20th century" [1].
The combined oral contraceptive pill (COC) was developed more or less by chance.The first progestin drugs were tainted with mestranol, a synthetic estrogen, during the first human study in Puerto Rico in 1956 [2].After breakthrough bleeding occurred as a result of the products' purification and lower estrogen content, it was decided to keep the estrogen for cycle control, establishing the idea behind combined estrogenprogestin oral contraceptives [3].
Immediately following the pill's debut, a sizable number of women started using it; however, over time, many health issues emerged, along with evidence of additional health benefits.The first COCs (first generation) on the market contained high doses (100-150 µg) of ethinylestradiol (EE) which were reduced since the mid-1960s due to the association with increased risks of venous thromboembolisms (VTE) and cardiovascular events.While in de 1970s 50 µg was the most prescribed dose, the switch to <50 µg was made from the 1980s since no loss of potency besides less clinical side effects were reported.Currently, the EE concentration in prescribed COC can be variable in which 20 µg is determined as the lowest effective dose.
In order to increase safety, in addition to the decrease of EE levels, other estrogen molecules, new progestins, new treatment regimens, and different delivery methods were added to hormonal contraceptive (HC) formulations.For decades, only the estrogenic molecule EE was utilized in HC.Recently, the novel compounds estradiol (E2), its valerate derivative (E2V), and estetrol (E4) have entered the market.New progestins derived from different compounds were also added to HC formulations, as three generations of progestins derived from testosterone were created.Progestins can be categorized by their activity in addition to the molecule from which they originate as (anti-)estrogenic, (anti-)androgenic, glucocorticoid, or anti-mineralocorticoid.The types and classifications of sex steroids that are being used in HC formulations are listed in Table 1.
In 2019, HCs were used by about 250 million women globally [4].The use varies greatly over the world, with Northern and Western Europe, and the United States, having the highest rates.In the US, 17.5 % of women between the ages of 15 and 45 y use COC, which equates to 27.5 % of all contraceptive users.Similarly in the Netherlands nearly 35 % of contraceptive methods are COCs.
There are countless potential COC combinations for both chemicals and dosages.According to Stridham Hall and Trussell's analysis of COC use in the US, there were over 80 different COC formulations in use [5].Between nations as well as between different regions of the world, the prescription pattern varies.Different safety data interpretations are a contributing factor, but traditions can have an impact as well.The most popular COCs in the US are those containing EE and norgestimate or drospirenone (DRSP); however, triphasic tablets are also widely used [3].Due to data showing a decreased risk of VTE in this group of users compared to those on thirdgeneration pills, in 2001, the European Medicines Agency (EMA) suggested second-generation COCs as the first choice (COC containing EE and levonorgestrel [LNG], desogestrel [DSG] or gestodene).
The use of an implant, an intrauterine device (IUD), injections, pills, vaginal rings, and skin patches are other examples of hormonal approaches.HC could be classified as combined (estrogen and progestin) or progestin only according to their contents.They can also be divided as long acting (depot), medium acting (monthly) and short acting (mini-pills).
Because of its wide usage among fertile women, its effects on biochemical test results should strongly be considered while assessing patients' laboratory results.In the 1970s reviews were published focusing on the effects of COCs on laboratory tests, but the dosages of hormonal contraceptives used at that time were much higher than currently used [6][7][8].
Table : Types and classifications of sex steroids used in hormonal contraception formulations.HC formulations consist of synthetic estrogens or natural estrogens and progestins (the androgenic progestins and the anti-androgenic progestins).

Estrogens
Ethinyl In addition to oral HC, vaginal, subdermal, transdermal and injectable HC are on the market now and their use is growing.Therefore, a literature review is needed to understand the effects of modern HCs on clinical chemistry test results.The aim of this review is to create a reference for physicians and laboratory specialists in case of a woman using HC and showing deviating biochemical test results.

Metabolism of hormonal contraceptives
Figure 1 illustrates the metabolism of hormonal contraceptives (HC) in the female body, which varies depending on the administration method.Orally taken HCs undergo significant first-pass metabolism in the intestinal wall and liver, resulting in a 40-60 % loss of the oral dose [9].This results in the synthesis of their metabolites [10].Enterohepatic circulation plays an important role in the metabolism of contraceptive steroids like ethinyl estradiol (EE) and estradiol (E2).
Most EE is metabolized before entering the bloodstream, while the majority of E2 is excreted in urine [10].E2 has a plasma half-life of 2.5 h, but its ultimate half-life, including metabolites, is 13-20 h.However, there is considerable intersubject variability in pharmacokinetics [11].Drug interactions may also influence the metabolism and efficacy of oral contraceptives.Progestins are generally well absorbed orally, reaching peak concentration in 1-3 h after first-pass metabolism.
In the circulation, contraceptive steroids are mainly bound (>95 %) to binding proteins, making them vulnerable to influences that might just slightly alter the binding.Therefore, a negligible 2 % reduction in the amount bound could result in an important 50 % increase in the unbound fraction, which is what determines the biologic activity of the steroids.EE exclusively binds to albumin.Despite the existence of albumin variants or deviating albumin concentrations in specific populations, no abnormalities in interaction between HC and albumin have been reported in literature.However, the gestagens have a complex equilibrium between the various fractions and bind to both albumin and sex hormone binding globulin (SHBG), with the extent and tightness of the binding to SHBG depending on their structure [9].

Metabolic parameters
COCs are suggested to have a possible deleterious impact on metabolic markers like lipids and carbohydrates [16].Clinical investigations have demonstrated that the doses, ratios, and potencies of the estrogens and progestogens found in COC have a variety of effects on blood lipid and lipoprotein concentrations [17][18][19][20][21][22][23]; see Table 2 for a summary of all published effects.
There is no noticeable difference in the lipid profiles between the cyclic and long period regimens (more than 1 year), according to four randomized controlled trials (RCT) [40][41][42][43].Over the course of the one-year study period Wiegratz et al. discovered a rise in total cholesterol, high-density lipoprotein, triglycerides, and very-low-density lipoprotein and a decrease in low-density lipoprotein in both groups [40].Progestins and  estrogens can both have an impact on lipid metabolism.The final impact is influenced by the estrogen-progestin dosage and the relative balance between the progestins' androgenic activity and estrogenic potency.LNG containing combinations that are more androgenic compared to HCs with less androgenic progestins, such as dienogest or nomegestrol acetate mixed with natural estrogen, are demonstrated to have a favorable effect on the lipid profile [44,45].
Effects of HCs on other metabolic parameters are summarized in Table 3   metabolism (i.e. lowered glucose tolerance and elevated insulin) in case of COC usage were not significant due to availability of low and ultra-low EE containing COCs.EE specifically has a higher impact on hepatic metabolism, and metabolic alterations, as compared to E2 or its valerate [45].
Although earlier progestins (norethindrone, MPA and CPA) could also affect metabolism [45], the latest progestin generations (NOMAC, DRSP and NES) essentially have no androgenic, estrogenic, or glucocorticoid-related adverse effects and no effects on fasting glucose or insulin levels [16,43,46].Contrary to COCs containing EE, COCs containing natural estrogens do not adversely impact the metabolism of glucose and insulin [29,48].
A previous systematic review showed also no effect of HC use on glucose metabolism [49].Due to the small number of studies, methodological flaws, and studies that compared the same forms of contraception, no firm conclusions can be made [49].
There is limited data about growth hormone (GH), insulin-like growth factor-1 (IGF-1) and IGF binding proteins concentrations upon HC usage.Some studies showed that GH levels were elevated more than 100 % upon COC (EE/DRSP), whereas IGF-1 levels were 20-50 % reduced.These changes seem to be caused by EE in the COCs, leading to a decrease in liver synthesis of IGF-1 that causes increased GH levels due to feedback interaction [50].Increased insulin-like growth factor binding protein-1 (IGFBP1) or insulin-like growth factor binding protein-3 (IGFBP3) synthesis in the liver could also be an effect of EE containing COC usage leading to up to 191 and 16 % increase in these levels, respectively [30,47].
In summary, contraceptives containing less androgenic progestins, i.e. drospirenone (DRSP), lead to improved lipid profiles compared to those containing androgenic progestins, i.e. levonorgestrel (LNG).This also depends on route of administration and dosage.Natural estrogens E2 and E2V cause fewer metabolic alterations than the synthetic EE.Despite statistical significance, except for GH, IGF-1 and IGFBPs changes in laboratory parameters related to metabolism are either small and results stay within the reference interval or data are not consistent (Table 3).Unfortunately, no studies are available to show how fast levels return to baseline after discontinuation of HC usage.Based on the half-life of listed parameters, it could be assumed that all changes might be reversed within 4 weeks after stopping HC.Thus, in case of uncertainty about the influence of HC on metabolic parameters in a woman using HC, these should be re-evaluated at least 4 weeks after HC discontinuation.

Liver function tests
The effect of sex hormones on liver function and the implications for the patient with liver disease are often of concern to obstetricians and gynecologists.The published effects on the liver function tests are shown in Table 4.It is generally known that estrogen receptors exist in the liver [51].Normal liver tissue, hepatic adenomas, hepatocellular carcinomas, and hepatoblastomas all include estrogen receptors.For sufficient systemic levels to be available after hepatic metabolism,  substantial concentrations of EE and a progestin should enter the portal circulation.There have been numerous reports of the hepatic side effects of COC, including cholestasis and the growth of hepatic adenomas.Some estrogen and progestin combinations damage the liver's ultrastructure and affect a variety of metabolic processes, including protein biosynthesis, energy production, and increased cellular catabolism [52].
Even though these side effects are uncommon and dependent on the dosage, duration and type of OC treatment, the possibility of liver damage makes managing contraception in people with liver illness challenging.Depending on the severity of liver damage due to HC use, effects might not always be represented as elevated alanine aminotransferase (ALT) or aspartate aminotransferase (AST) enzyme levels, since these markers are quite insensitive to small changes.The effects of HC on liver function are less when the hormones are delivered by parenteral route without liver transit.For instance, it was shown that subdermal implants after three years of use did not significantly impact hepatic enzymes [31].
The unusually elevated liver enzymes and bilirubins related with the development of intrahepatic cholestasis following COC usage is typically attributed to the estrogen component of COC pills [58].Additionally, bile salt export pump (BSEP) and ATP-binding cassette subfamily B11 (ABC B11) member gene-associated estrogen-induced cholestasis can be found, particularly in women with a history of idiopathic cholestasis of pregnancy or hereditary bilirubin metabolism defects like Dubin Johnson syndrome [59,60].Studies on rats have led researchers to hypothesize that canalicular bile transporters (such as multidrug resistance protein 2, which is in charge of biliary secretion of bilirubin glucuronides) may be involved in the estrogen-induced cholestasis [61].
In contrast to estrogens, progestogens rarely contribute to cholestatic liver test changes [62,63], and can therefore be a good alternative to prevent hepatic test fluctuations, although intrahepatic cholestasis has also been shown after high doses of progesterone treatment in patients with breast cancer [64].On the other hand, progesterone metabolites may sometimes play a more significant role than estradiol metabolites in the pathogenesis of obstetric cholestasis.Patients with obstetric cholestasis have high concentrations of progesterone metabolites in serum and sulfated progesterone metabolites in urine [65,66].Recently, a 49-year-old woman taking progestin only pill (norethisterone) with extremely higher liver parameters related with intrahepatic cholestasis was reported [67].Therefore, in vulnerable patients with cholangiopathy brought on by steatosis and/or genetic predispositions, progestogens may cause high liver enzymes and bilirubins.
The effects of HCs on the hematology parameters are shown in Table 5. Users of traditionally administered COCs have higher serum transferrin levels than nonusers [68].Instead of being an indication of iron shortage, the greater transferrin concentration could be the result of the contraceptive pill's general impact on the levels of transport proteins.Several studies showed that COC users had no differences in hemoglobin vs. non-COC users and that cyclic vs. continuous use of COC also did not influence hemoglobin levels [69,70].Thus, it appears that ongoing OC use has little to no effect on the body's iron status and hemoglobin levels [68].In summary, current COC formulations have not been connected to liver enzyme elevations, despite early COC formulations being linked to frequently elevated serum enzyme levels.In people with hereditary types of bilirubin metabolism like Dubin Johnson syndrome, COCs can produce a modest suppression of bilirubin excretion, resulting in jaundice.The ability of COCs to cause aberrant test results with liver damage is more significant in specific cases often having a history of cholestasis of pregnancy (with jaundice and/or pruritus), and genetic variations in bile acid transporter genes (ABC B4, B11 and C2).The effects of HCs on the liver are reversible and most of liver tests could be reassessed 4 weeks after stopping HCs [59,60].

Hemostasis parameters
The use of COCs has an impact on a number of hemostatic variables that are associated with thrombin formation and procoagulant (Table 6), anticoagulant, or fibrinolytic pathways (Table 7).Despite that their concentrations often stay within the reference range, changes in D-Dimer, antitrombin III, factor II activity and plasminogen levels (Tables 6 and 7) may have additive or antagonistic effects on the risk of venous thromboembolism (VTE) [72].Activated protein C resistance (APCr) is a relatively new marker found to be important in the pathogenesis of VTE [73,74].Increased APCr is seen in COC users, and COCs containing DSG, gestodene (GSD), or DRSP exhibit a higher normalized APC sensitivity ratio (nAPCsr) comparable to those of heterozygous Factor V Leiden carriers [75,76].In COC users minimal increases of coagulation factors (such as prothrombin and factor VIII), small decreases of anticoagulant proteins (such as tissue factor pathway inhibitor [TFPI], protein S, protein C and antithrombin), and acquired APCr were reported [77,78].
In the seventies, the effects of estrogen on blood coagulation, fibrinolysis, and platelets were compared between COCs containing 50 and 30 µg of EE.With both preparations, higher levels of platelet activity, factor II, VII, VIII, IX, and X, fibrinogen, and soluble fibrin were seen, as well as lower levels of antithrombin (AT) and vessel wall fibrinolytic activator.The effects were EE dose dependent and less pronounced with EE concentrations with 30 µg in COC preparations [89][90][91][92].Yet, in a more recent meta-analysis the VTE risk remained higher with the so-called third generation COCs (EE + DSG, GSD, or norgestimate) compared to second generation COCs (EE + LNG or norethisterone) [93].The difference in VTE risk for each COC could be attributed to a unique modulation of the procoagulant action of EE, exerted by the progestogens, as progestin-only contraceptives do not interfere with the synthesis of coagulation protein [94,95].
EE combined with DSG, GSD or DRSP cause more alterations in coagulation markers and a greater observed VTE risk than EE/LNG [78,85].This is probably due to the less potent antagonistic interactions between the EE-induced alterations in hemostasis variables and DSG, GSD, and DRSP as compared to LNG [80].COCs that contain E2 seem to have less of an impact on hemostasis parameters [94].A recent study found that changes in hemostasis measures following six cycles of a natural estrogen (E4) with DRSP were either less pronounced or comparable to those seen with EE/LNG [83].The observation of larger changes compared to EE/DRSP is consistent with the theory that the estrogenic component of COCs primarily mediates the effect on hemostasis measures.
Regardless of their limited effects on routinely used hemostasis parameters, high estrogenicity is associated with an increased risk of VTE [96].A COC's estrogenicity is calculated as the sum of its progesterone and estrogen contributions.The estrogenicity levels of either alone or combined hormonal contraceptives are depicted in Figure 2.
The hemostatic measures during traditional COC administration and the values with continuous administration were examined in three RCTs [42,43,46].The pro-coagulatory factors were increased, the coagulation inhibitors were decreased, and the fibrinolysis parameters were enhanced in both regimens.This might mean that there is no difference between the two regimens, but that both traditional and continuous use of OCs have increased coagulation activity.
In summary, the effects of HCs on parameters related to the coagulation system are heterogeneous.While increased APCr is seen in some COCs (containing DSG, GSD or DRSP) exhibiting a higher normalized APC sensitivity ratio (nAPCsr), most changes are within reference intervals and are generally not related with the composition of HC preparations.Although these changes in coagulation are non-explanatory for VTE risk related with HCs, it is still a clinically important issue and estrogen contraception is usually contraindicated in women with a personal history of VTE.The underlying mechanisms should also be further investigated.

Corticosteroids, androgens, estrogens and their binding proteins
The published effects of HC usage on the corticosteroids, androgens, estrogens and their binding proteins are listed in Table 8.Synthetic estrogen-and progesterone-like chemicals in the HCs work by inhibiting the pituitary's ability to produce follicle stimulating hormone (FSH) and luteinizing hormone (LH).While the LH surge initiates mid-cycle ovulation, FSH promotes the development of folliculogenesis.After ovulation progesterone is secreted by the corpus luteum which inhibits endometrial proliferation, determines endometrial receptivity and gives negative feedback to the pituitary and hypothalamus.
LH and FSH testing gives important information about gonadal function.Follicular development and ovulation are stimulated by LH together with FSH resulting proper growth of follicles.These hormones are mostly normally suppressed in women using COCs.HC may also suppress endogenous estradiol and/or progesterone levels.If the HC contain EE, the estradiol measurements are decreased as EE do not cross react in estradiol immunoassays and are not measured using  estradiol LC-MS/MS assays.However, when HC contain E2, this exogenous estradiol is measured in the estradiol assays.
EE increases levels of SHBG [109].As testosterone is bound to SHBG, and the testosterone production in theca cells, that is promoted by LH, is decreased by OCs [98,110], the serum concentration of free testosterone is often described as being reduced by 20-40 % when low-dose HCs are used [111].The lowering of free testosterone levels during conventional OC administration is quickly reversed over the 7-day hormone-free period [112,113].Free testosterone is said to be returned to baseline levels as a result of this gain in total testosterone and the attenuation of the OC-stimulated rise in SHBG during the pill-free week [99].The adrenal glands are the primary source of the remaining amounts (25 %) of circulating testosterone, either directly or through conversion of precursors such androstenedione or dehydroepiandrosterone.Dehydroepiandrosterone sulfate (DHEAS) levels are moderately decreased while using OCs (and are partially restored during the hormone-free period), which may be due to a direct interaction between the contraceptive steroids and the production of adrenal steroid hormones [98,99,112,114].Recent studies have shown that COCs containing E4 have milder effects on SHBG, CBG, and gonadotropins (60-90 %) compared to those containing EE (200-400 %) [29,30].
Anti-androgen progestins combined with EE are not only used for contraception but also for the treatment of hirsutism and polycystic ovary syndrome.Levels of free and total testosterone, androstenedione and DHEA-S as well as FSH and LH were found lower in women using COCs containing EE and anti-androgen progestin CPA [101,115].
One of the drawbacks and shortcomings of testosterone immunoassays revealed that these methods are not accurate in the low testosterone concentrations range and are analytically influenced by high SHBG concentrations.The high SHBG concentration in women using HCs might therefore lead to falsely low testosterone concentrations measured using these immunoassays [116,117].Thus, the lower testosterone levels commonly seen in hormonal contraceptive usage, should be confirmed and studied by LC-MS/MS methods, which are not analytically influenced by SHBG concentrations, in future studies.
As estrogens regulate cortisol equilibrium by promoting hepatic CBG synthesis, both CBG and total cortisol levels rise when COCs are used [29,85,111].Increased CBG levels lead to a new, altered CBG-cortisol equilibrium followed by concurrent increases in cortisol production.EE-based COCs have more significant impact on adrenal steroids and CBG levels (∼two-fold higher than baseline) whereas different E2/EV/E4 combinations seem to be of less significance [29,81,85].Adrenal reactivity to adrenocorticotrophic hormone (ACTH) is similarly increased by COC usage due to an increase in CBG concentrations [118].After exogenous ACTH administration (synacthen test) higher cortisol levels in serum, but not in saliva, were found in women using three months low-dose COC [119].Moreover, COC use is also associated with inadequate suppression of cortisol by overnight-1 mg dexamethasone [120].Recently, a two-day low dose dexamethasone suppression test was found to be a better screening test for hypercortisolism in women using HC [121].Alternatively, the 1 mg dexamethasone suppression tests could be performed after a 1 month HC free period [121].The free cortisol index or salivary cortisol which are said not to be affected by EE might be an option for hypercortisolism screening in women using HC [122].
Anti-Mullerian hormone (AMH), which is a widely used parameter for the assessment of ovarian reserve, was found to be decreased by long term (>6 months) HC usage in studies having a larger number of participants while smaller studies did not find any change [104,105,107,108,123,124].This decline in AMH levels returned to baseline levels after two months of HC discontinuation [124].Although the exact reason why AMH levels were suppressed during HC treatment is currently not known, suppression of FSH and LH levels, and diminished formation of antral follicles by HC were suggested as a possible causative explanation [125].
In summary, HCs suppress LH and FSH and the endogenous estradiol production due to feedback on the pituitary gland and hypothalamus.Furthermore, lower levels of AMH during HC usage could reflect suppression of ovarian functions.HCs including EE have a major impact on SHBG and CBG concentrations.The impact varies and depend on the kind of progestin present in the COC, i.e. those having progestins with antiandrogenic qualities had the greatest impact, leading to increases between 200 and 400 %.Natural estrogens appear to have less effect on those hepatic proteins, i.e. an increase of only 60-90 %.All preparations of HCs have the ability to decrease androgen secretion since they inhibit the hypothalamus-pituitary-ovarian axis.Yet, the weaknesses of widely used immunoassay methods for total testosterone levels (and thereby the calculated free testosterone) should also be considered as an important analytical error source leading to falsely low androgen levels.

Thyroid hormones and related proteins
The published effects of HC usage on thyroid hormones and related proteins are listed in Table 9.It is widely accepted that thyroxine-binding globulin (TBG) concentration is increased during OC therapy [110,126].Initial studies suggest that estrogen can induce hepatic TBG protein synthesis [127].
Other studies showed that natural estrogen induces sialylation of TBG.Sialylation prolongs TBG half-life due to a decreased clearance rate resulting in higher serum TBG  concentrations upon COC therapy [128].The amount of COC induced TBG elevation is dependent on the prescribed type and concentration of the synthetic estrogens and progestogens [111].Since TBG is the major binding protein for thyroid hormones, its elevation is directly related to a rise in total T3 and T4.This is the reason why nowadays most laboratories changed thyroid functions test from total T4 and T3 into fT4 and fT3 as these are not affected by COC use, despite transient changes (depending on COC type)_ during the first cycle of COC use [111].The elevations in TBG, T3 and T4 are more prominent in dienogest-containing COC compared to LNG which can be explained by the lower estrogenic activity of LNG [99,110,111].rT3 remains unaffected after LNG or DSG-containing COC administration [110].TSH levels are also not affected, except following EE/E2V/DNG use which is suggested to result in higher TSH levels [111].
In addition to studies on the effects of COC therapy on thyroid parameters, other studies focused on the effect of different manners of administration of estrogens and progestogens on thyroid function parameters.Remarkably, transdermal administration of E2 did not have an effect on TBG concentration [130].A more recent study showed that the use of a contraceptive (transdermal) patch that delivers EE and LNG results in a significant TBG increase.This rise was even more prominent compared to the elevation upon COC administration [132].Another study showed that a single intramuscular progesterone administration has not been associated with elevation of TBG and T4 levels [129].Usage of depot medroxyprogesterone acetate as injectable contraceptive caused a minor but statistically significant increase in fT4 levels, which was not observed in patients using copper IUD [131].In accordance, administration of micronized progesterone increases fT4 concentration, which is still in the reference range, without affecting fT3 and TSH levels [133].
In conclusion, HC, depending on the type and the administration route, can lead to elevated TBG, T4 and T3 concentrations.It is therefore recommended not to measure total T4 and total T3 levels in women using HCs, but instead rely on fT4 and fT3 measurements which show at most minor changes during the usage of some of HCs.

Renin angiotensin system and renal function
The renin-angiotensin-aldosterone system is known to be affected by HC in two ways.The published data on the effects are shown in Table 10.The increased (95-375 %) formation of angiotensinogen is first and foremost highly stimulated by estrogens, which results in higher levels of angiotensin and aldosterone as well as sodium retention.Second, progesterone inhibits sodium retention by acting as a strong aldosterone antagonist on the mineralocorticoid receptor.Progestogens lacking anti-androgenic and anti-mineralocorticoid properties are unable to offset the sodium-retention impact of the EE component in COC.In turn, this could lead to increased fluid retention and the promotion of symptoms like edema and increased body weight [134,135].
Due to the significant hepatic effect of EE, oral contraceptives are also known to raise the liver derived plasma renin substrate angiotensinogen [144].Angiotensin II (the reninangiotensin system [RAS] effector substance), renin and aldosterone levels may also increase in conjunction in this way [145].
The impact of synthetic estrogens on RAS is also shown by exogenous [146] estrogen injections increasing plasma, hepatic, and renal concentrations of angiotensinogen, which has the ability to increase plasma concentrations of angiotensin II.This can be explained by the angiotensinogen gene having a promoter region responding to estrogen [147].When taken as part of a COC, EE (15 µg/day) causes a rise in plasma angiotensinogen that is only marginally smaller than that observed during pregnancy [148].Studies on exogenous estrogen administration have shown both stimulation [149,150] and inhibition of the circulating RAS [151].Studies evaluating the effects of OCs on aldosterone levels have also shown conflicting results [139,145,152,153].
The generation of angiotensinogen in the liver will not be stimulated by the contraceptive patch because of the absence of first-pass metabolism [154].Therefore, the transdermal application does not result in an increase in circulating RAS components [130,146,155] and similarly, the use of injectable contraceptives was not found to be linked to an increase in blood pressure [156].
The endogenous creatinine clearance was significantly raised in groups using COC.Also significantly higher rates of potassium salt excretion were found [136].This study demonstrated that oral contraceptives could raise glomerular filtration rate and some types have a protein catabolic effect, in addition to their different effects on renal tubular function.Numerous short-term investigations and cross-sectional studies have revealed a relationship between HC usage and urine albumin loss with regard to the effects of HC on urinary albumin excretion [139,140].In contrast to women who did not use HC, women using HC had a 90 % higher risk of microalbuminuria (30-300 mg/day) [140].These effects account for both normotensive and hypertensive women who used oral contraceptives [139].A significant reduction on albumin excretion rates was found after discontinuation of HC usage [137].
In summary, angiotensinogen levels can be extremely (even 4-fold) elevated by HC use.The RAS system could also be affected depending on the composition of HCs.Because of pre-analytical effects and also analytical variations, aldosterone levels show significant variations independent from the compositions of OCs.
Therefore, the diagnosis of primary hyperaldosteronism in women using HCs could be interfered by false positive aldosterone renin ratios due to affected parameters of reninangiotensin-aldosterone system.Despite insufficient evidence that using HC makes women more likely to develop renal illness, HCs may raise the risk of developing microalbuminuria regardless of blood pressure.The disturbances in RAS and renal functions due to HC usage are therefore advised to be re-evaluated after at least a 4 week treatment free period.

Vitamins and minerals
Studies have shown that the widespread use of oral contraceptives may have a negative impact on the body's levels of several vital vitamins and minerals [160][161][162][163].The published effects of HC usage on vitamins and minerals are listed in Table 11.According to the World Health Organization (WHO), the impact of nutritional depletion brought on by using oral contraceptives is of high clinical relevance.Folate, vitamins B1, B2, B6, B12, C, and E, as well as the minerals magnesium, selenium, and zinc, may be affected by oral contraceptives.Additionally, they may lower levels of the amino acid tyrosine and the antioxidant Coenzyme Q10.
Vitamin B6, B12, or folate levels were found to be lower in OC users compared to controls [164][165][166]182].According to a study by Green et al. [168], total vitamin B12 levels were 33 % lower in OC users than in non-users.Given that OCs affect the quantities of circulating proteins, it is probable that they change the transcobalamins (TCs), which are circulating B12 binding proteins.In the circulation, approximately 20 % of B12 is bound to TC2 and is available for tissues (holo-TC or active B12); 80 % is bound to TC1 (Haptocorrin) and is not available for tissues.Haptocorrin was found to be lower in OC users compared to nonusers, suggesting a decreased vitamin B12 binding ability rather than a functional deficiency [167,183].
There was no difference between OC users and controls in plasma MMA and plasma total homocysteine (tHcy) suggesting that OC does not lead to a functional B12 deficiency [167].It is still unclear whether the lower serum B12 levels associated with OC usage genuinely point to a biochemical B12 shortage, despite the fact that there is consistent data to support this association.The levels of active vitamin B12 during HC use should be further investigated.It is also unknown how people already at risk for B12 insufficiency, including vegetarians, may be impacted by a possible alteration in blood B12 binding ability brought on by OC usage.
The use of COCs has a detrimental effect on folate status by affecting the absorption pathway and thereby lowering serum levels.In comparison to a control group, women taking OCs had significantly decreased serum folate levels [184,185].Other investigations supported these findings revealing either a decrease in intestinal folate absorption in OC users [186] or increased folate metabolism and urine excretion [185].While other studies conducted during the same time period found no difference in the folate status of OC users and controls, a recent meta-analysis that included case-control, cohort studies, and clinical trials from 1970 to 2013 came to the conclusion that OC use is, in fact, associated with lower serum folate status [187].Confounding factors, such as dietary folate intake not being adjusted for, supplement use, smoking, and alcohol consumption may be the underlying explanation for this variation [188,189].The variable results even excluding dietary intake of folate [168,190,191] may be caused by inter-individual genetic variations in how folate is processed [188].
To investigate the effects of OCs of vitamin B6, one should take other metabolites of vitamin B6 into account because estrogens may affect tryptophan metabolism independently of vitamin B6 [188,192].They are urinary 4-pyridoxic acid (4-PA), plasma pyridoxal 5-phosphate (PLP), urinary B6, and erythrocyte aminotransferase or transaminase activity.Interestingly, PLP is considerably lower in OC users than non-users in both fasting and non-fasting plasma [165,193], supporting the presence of lower vitamin B6 status in these women.
Moreover, COCs also cause a vitamin E deficiency.In fact, OC users experience more oxidative stress, particularly in terms of lipid peroxidation, which is linked to reduced amounts of vitamin E [194][195][196] in the blood.The injection of contraceptive drugs dramatically decreased plasma levels of tocopherol and increased dietary needs for vitamin E [197], according to preclinical research in rats.In agreement with these findings, a clinical study found that COCs also lower plasma tocopherol levels in healthy women [198].OC users have also greater blood-clotting activity, which may be related to the participants' lower levels of tocopherol [199].
COC-based treatments, especially containing estrogens, have an impact on vitamin C status as indicated by lower levels of ascorbate in plasma leukocytes, platelets, and whole blood entities due to the increased rate of vitamin C metabolism [200][201][202].This effect is particularly amplified in women who have a disease of malabsorption, a poor diet, bad lifestyle, or both.In agreement with this, research has shown that a sufficient diet rich in ascorbic acid can prevent depletion brought on by OC usage over a seven-year period [203].
The usage of HC has been linked to statistically higher total 25-hydroxyvitamin D (25(OH)D) levels in adults and adolescents [176][177][178][179][180] which might be due to higher levels of vitamin D-binding protein (VDBP) [179,180,204,205].HC use did not change free 25(OH)D (if possible to measure accurately).Vitamin D status of HC users may therefore be overestimated [205].As a result, in a population that uses HC, free or bioavailable 25(OH)D might be a more helpful indicator of vitamin D status in theory.VDBP-bound 25(OH)D can however be used by several tissues, including the parathyroid gland, which may change the relevance of free 25(OH)D levels [206].However, as the majority of immunoassays produce 25(OH)D results that could be inaccurate due to being affected by VDBP concentration, further studies should be performed using LC-MS/MS methods before any conclusions can be drawn [207,208].
Women using HC usually show altered mineral status, particularly at the levels of magnesium, zinc, and selenium [198].OC users have lower serum magnesium levels than either noncontraceptive-users or women using alternative methods of contraception [160,173,209,210].In early studies, women using OCs had lower zinc levels than non-users [211,212].These findings were supported by other studies [172,175,213,214], which suggested that reduced serum zinc levels might be a reflection of a decline in the status of zinc in tissues as a result of changes in zinc absorption, excretion, or tissue turnover.Numerous studies have shown that OCs may hinder the absorption of selenium and result in insufficiency.According to a clinical investigation in 200 young women, taking low-dose OCs for at least three months causes a statistically significant drop in mean serum selenium levels when compared to control subjects [172,215].
In summary, HCs can affect blood levels of vitamins or minerals but probably without affecting their physiological functions.When assessing vitamin state in women using HCs, determination of functional markers (like homocysteine, and methylmalonic acid) could be beneficial to determine whether there is an actual vitamin deficiency or not.However, women who get HCs are at risk for a number of adverse effects linked to deficiency in a number of vitamins and minerals, including zinc, magnesium, selenium, vitamins B, C, and E which may request further investigations.25OH Vitamin D levels may be increased in women using HC due to increased VDBP concentrations, however similar to testosterone, analytical errors may lead to falsely low 25OH Vitamin D levels.

Conclusions and recommendations
In this review, the impact of currently used HCs on clinical chemistry test results was summarized.HCs were defined as all hormonal contraceptive types and combinations.The effects of HCs differ by type of contraceptives, administration route and duration.The results of studies that included the definition of the composition of HC (76 studies) were summarized in tables.Unfortunately, a lot of studies provided no information about the HC composition used by their patient groups, consequently these studies were not included in the tables of this review.In addition, the number of publications about newer (e.g., those containing naturel estrogens) and non-oral HCs (vaginal ring, IUD, subcutaneous, transdermal and parenteral) and their effects on biochemical test results are limited.Despite the scarce information, the data about binding proteins suggest that natural estrogens have less impact than EE.
Even though the changes shown in the tables were statistically significant, most of them were rather small and concentrations therefore mostly did not end up outside the reference interval.Although HCs itself did not interfere the laboratory analyses as an analytical bias, increased binding protein concentrations like SHBG or VDBP did analytically influence the determination of the total levels (like testosterone and 25(OH)D) leading to mostly falsely low total hormone concentrations using several immunoassays.Thus, we suggest to measure these total hormones using more reliable methods like LC-MS/MS in order not to be analytically influenced by the concentration of the binding proteins.This phenomenon is especially important as HCs strongly affect the synthesis of transport or binding proteins produced in the liver (like TBG, SHBG, CBG and VDBP).In addition to the analytical consequences this might have, the hormones bound to these proteins (T4, T3, testosterone, cortisol and 25(OH)D) can show concentrations that need to be interpreted with care.
An important question is: what to do with women on HCs and affected clinical chemistry tests?One possibility is stopping with the HC and performing the tests again later.This possibility leads to the next question: if a woman would stop, after what time frame can she be tested again?This is especially important for dynamic endocrine tests as hormones like cortisol are heavily influenced.Although there are not enough data available for a firm conclusion, based on the half-life of measured analytes (i.e., for SHBG, TBG and CBG biological half-life is about 1 week) it might be suggested discontinuation of HC for at least 1 month could be enough for measuring clinical chemistry laboratory parameters reliably.Future studies need to focus on when levels return to baseline after discontinuation of HCs.Discontinuation of HC, however, has a drawback on contraception and physicians and patients should be well aware of this.An alternative for stopping HC could theoretically be establishing specific reference intervals for women using HCs.However, the different effects between HC types, dosages and in-cycle dose on clinical chemistry parameters make it impossible to set up specific reference intervals for HC users.
A third option for some of these tests, are free hormone measurements, such as 24 h urine cortisol or late night salivary cortisol instead of a total serum cortisol after dexamethasone or to use functional markers (i.e., tHcy and MMA) instead of vitamins and minerals to understand the existence of an actual deficiency.
In conclusion, many clinical chemistry tests show small changes upon HC use.We found that only some of these tests require special attention when measured in women using HCs: the binding proteins TBG, SHBG, CBG and VDBP and total measurements of the hormones binding to these proteins; the RAS system, aldosterone, renin and angiotensinogen; AMH; and some vitamins and minerals.Natural estrogens seem to have less effects on these clinical chemistry tests yet effects of non-oral HCs need to be studied.Alternative measurements such as free hormones or functional markers are preferred for optimal interpretation, otherwise HCs should be stopped for at least one month while taking care for anticonception in another way.

Table  :
Effects of hormonal contraceptives on lipids and lipoproteins.

Table  :
(continued) en will be discussed in this paragraph.Sitruk-Ware et al. emphasized that changes in carbohydrate

Table  :
Effects of hormonal contraceptives on metabolic parameters.

Table  :
Effects of hormonal contraceptives on routine chemistry tests.

Table  :
Effects of hormonal contraceptives on hematology tests.

Table  :
Effects of HC on markers of thrombin formation and anticoagulant proteins.

Table  :
Effects of HCs on the parameters related with renal functions and renin-angiotensin-aldosterone system.

Table  :
Effects of HCs on vitamin and mineral parameters.