Endocrine Effects of Coffee Consumption Kahvenin Endokrin Etkileri

Review Turk J Endocrinol Metab

Marmara University Pendik Research and Training Hospital, Department of Endocrinology and Metabolism, İstanbul, Turkey

Caffeine has been found to exert various biological effects Kafein; antiproliferatif, antianjiyogenik, antimetastatik etki- including, antiangiogenic, antiproliferative, antimetastatic leri, artmış yağ oksidasyonu, glikojenin kas içinde mobili- activity, increased fat oxidation and mobilization of glyco- zasyonu, artmış lipoliz gibi çeşitli biyolojik etkileri gen in muscle, increased lipolysis, and reduction of body fat. gösterilmiştir. Bu çalışmada amacımız kahve tüketiminin en- The aim of this review is to analyze the endocrine effects of dokrin sistem üzerine olan etkilerini incelemektir. Mayıs coffee consumption. A systematic literature search was con- 2019'a kadar yayınlanan ve kahve tüketimi ve diyabet, os- ducted on PubMed and Web of Science databases seeking teoporoz, tiroid hastalıkları, adrenal ve gonad fonksiyonları articles published until May 2019, dealing with coffee con- hakkında yayınlanan makaleler PubMed ve Web of Science sumption and diabetes, osteoporosis, thyroid gland, adre- veritabanlarında sistematik olarak tarandı. Epidemiyolojik nal, and gonads. The results of the most epidemiologic çalışmalar, kahve tüketiminin tip 2 diyabet riskini azaltmada studies reported that coffee consumption has positive ef- olumlu etkisi olduğunu, kemik mineral yoğunluğu üzerinde fects on combating type 2 diabetes risk, has no significant anlamlı bir etkisinin olmadığını ancak yüksek kahve tüketen effects on BMD (Bone Mineral Density) levels but fracture grupta kırık riskinin daha yüksek olduğunu, tiroid kanseri risk was shown to be higher in the high coffee consumer üzerinde anlamlı etkisinin olmadığı, SHBG seviyelerini art- group. Coffee intake has no significant effect on thyroid can- tırdığı, doğurganlık üzerine etkisi olmadığını ancak spontan cer, increases SHBG levels, has no effect on fertility but hig- abortus riskini artırdığını göstermiştir. Kahve tüketimi ve en- her consumption was related to spontaneous abortion. dokrin etkiler üzerine yapılan çalışmaların çelişkili sonuçları Studies pertaining to coffee consumption and endocrine ef- olup net etkilerini belirleyebilmek için uzun takip süreli kli- fects have contrary results. More randomized clinical stu- nik çalışmalar gereklidir. dies with a long term follow up period are required.

Keywords: Coffee; diabetes mellitus; fertility; Anahtar kelimeler: Kahve; diabetes mellitus; fertilite; gonadal hormones; osteoporosis; gonadal hormonlar; osteoporoz; thyroid diseases tiroid hastalıkları

Introduction North Europeans. Boiled coffee is fre- Coffee is one of the most popular beverages quently consumed by the inhabitants of the consumed worldwide, thereby enhancing its Balkan area, Ireland, North Africa, and market demand (1). Turkey. Statistics reveal that the average con- Scientists have extracted numerous bioac- sumption of coffee ranges from 2-4 cups tive compounds, such as chlorogenic acids per day in western societies. Western (CGA), polyphenols, diterpenes, caffeine, countries prefer to consume brewed coffee and caffeine metabolites (1,2) from this while instant coffee is popular among the complex beverage.

Address for Correspondence: Tuğçe APAYDIN, Marmara University Pendik Research and Training Hospital, Department of Endocrinology and Metabolism, İstanbul, Turkey Phone: :+90 554 4036978 E-mail: [email protected]

Peer review under responsibility of Turkish Journal of Endocrinology and Metabolism.

Received: 18 Oct 2019 Received in revised form: 21 Jan 2020 Accepted: 30 Jan 2020 Available online: 13 Feb 2020

This is an open access article under the CC BY-NC-SA license (https://creativecommons.org/licenses/by-nc-sa/4.0/) DOI: 10.25179/tjem.2019-71878 Owing to the presence of these bioactive coffee on developing type 2 diabetes. components, there is an increased scientific Epidemiological studies conducted among interest in the potential health benefits of different populations throughout the world regular coffee consumption. Recent studies highlighted the positive effects of coffee in have reported that coffee exerts functional combating the risk of type 2 diabetes, in a effects on human health. The effect of coffee dose-dependent manner. Studies suggested intake on malignancy has already been re- that 3-4 cups of coffee per day were capable ported in various studies. Presently, re- of reducing the risk of development of T2DM search is mostly focused on exploring the (Type 2 Diabetes Mellitus) by approximately scientific aspect of the effect of coffee on 25%, as compared to those consuming less type 2 diabetes, cardiovascular, and cere- than 2 cups per day. The relative risk of brovascular diseases. All these medicinal ef- T2DM for the highest level of coffee intake fects of coffee are attributed to the presence (>6 cups/day) was estimated to be 0.71 of various bioactive compounds like caffeine, (0.67-0.76) for caffeinated coffee and 0.79 chlorogenic acid, and cafestol. Caffeine has (0.69-0.91) in case of decaffeinated coffee been known to exhibit several biological ef- (1,4). In addition, increasing coffee con- fects, such as antiproliferative, antiangio- sumption +1 cup/day incurred a decrease of genic, antimetastatic activities, increased fat 11% on T2DM risk in the following four oxidation, and mobilization of glycogen in years. On the other hand, reducing coffee muscles, increased lipolysis, and decreased consumption by more than 1 cup/day in- body fat (3). creased the risk of development of T2DM in Recent clinical and laboratory data implicate four years by 17% (1,5). Both caffeinated that endocrine tissues could also be the tar- and decaffeinated coffee, filtered and instant gets of coffee compounds. The present re- coffee were associated with reduced risk of view aims to analyze the clinical trials that T2DM (6–8). Studies also suggested that evaluated the consequences of coffee con- this inverse association was applicable to sumption on endocrine disorders. very high levels of coffee consumption. Dose-response analysis suggested that 12% Methods (0.88 (0.86-0.90)) reduced risk of T2DM A systematic literature search was con- was attributed to the intake of every 2 ducted on PubMed and Web of Science data- cups/day of caffeinated coffee. On the other bases seeking articles published until May hand, intake of every 2 cups/day of decaf- 2019 using a combination of the following feinated coffee was associated with an 11% Medical Subject Headings terms and key- (0.89 (0.82-0.98)) reduced risk (9). A re- words: coffee ‘AND’ (diabetes ‘OR’ type 2 di- cent study from a low coffee consumer abetes ‘OR’ thyroid ‘OR’ goiter ‘OR’ country, Iran also concluded a lower risk of hypothyroid ‘OR’ hyperthyroid ‘OR’ thyroid diabetes or pre-diabetes among the coffee cancer ‘OR’ infertility ’OR’ osteoporosis ‘OR’ consumer group (10). vitamin D ‘OR’ adrenal). However, few epidemiologic studies failed to Constraints were used for advanced search: report any correlation between coffee con- adults, human, clinical trial, and search sumption and type 2 diabetes risk (11). fields: title/abstract. Additionally, we scruti- Q 2: Is it the caffeine in coffee that causes nized references within identified papers as diabetes? well as articles that had come to our attention through other means. Published reports opined that both caffeinated and decaffeinated coffee consump- Coffee and Type 2 Diabetes tion lowered the risk of type 2 diabetes (7,10,12,13). Q1: Does coffee consumption reduce type 2 There are controversial studies about the ef- diabetes risk? fect of decaffeinated coffee on type 2 dia- Although the correlation of coffee with the betes risk. A prospective cohort study pathogenesis of diabetes is still controver- reported a negative relationship between sial, special attention has been given to es- caffeinated coffee and type 2 diabetes risk. timate the possible beneficial effects of However, this study failed to establish any

2 effect of decaffeinated coffee on the devel- receive any coffee for 16 weeks. Second- opment of T2DM (14). hour glucose levels were found to be lower for the caffeinated group. When adjusted Q3: Does coffee consumption affect according to waist circumference, caf- glycemic control? feinated and decaffeinated groups were as- There are a few studies reflecting the out- sociated with a modest reduction in glucose come of coffee consumption on glucose lev- levels post 2-hour glucose load (19). els in type 2 diabetic patients. Studies Randomized controlled trials were also con- suggested that exaggerated postprandial ducted estimating the short-moderate term glucose and altered insulin responses owing effects of coffee on glucose metabolism and to the presence of caffeine might lead to im- insulin sensitivity. In a trial, the glucose and paired glycemic control (15). This contradic- insulin levels were recorded during the sec- tory effect might result from the acute effect ond week of the regular coffee intake period of coffee. However, habitual consumption and two weeks of abstinence period in 26 might indulge in the positive effects of this healthy volunteers. Fasting plasma glucose beverage. Therefore, it is difficult to com- levels were similar but fasting insulin levels pare the studies investigating acute meta- were higher after the coffee intake period bolic effects. (20). Another study among euglycemic individu- Q4: Are there any effects of coffee als compared the insulin sensitivity among consumption on glucose metabolism and 5 cups/day caffeinated, 5 cups/day decaf- insulin sensitivity in healthy individuals? feinated, and no coffee consumers for a Oral glucose tolerance test (OGTT) was con- period of eight weeks. Though insulin sen- ducted among euglycemic individuals where sitivity was reduced due to short term ef- subjects were provided with a 75-gram glu- fects of coffee, positive effects on cose load and the blood glucose level was adipocyte and liver functions were wit- estimated, henceforth. Such cross-sectional nessed (21). studies among healthy populations revealed that second-hour glucose levels, following Q5: What is the main mechanism of the the glucose load, were lower in the coffee beneficial effects of coffee? consumer group (16). Another study also The main mechanism behind the beneficial reported second-hour glucose levels in a effects of coffee on glucose metabolism has coffee consumer group to be lower among not been elucidated so far. Some in vivo and 1328 healthy individuals (17). in vitro studies have been conducted to Randomized controlled studies in healthy in- screen the pharmacological activity of the dividuals observed reduced insulin sensitiv- biological compounds present in coffee. In a ity attributed to acute caffeine load. cross-over trial, 12 gram decaffeinated cof- However, the plasma glucose levels re- fee, 1 gram chlorogenic acid, 500 mg mained unaffected. Contrary to these find- trigonelline, and 1 gram mannitol used as ings, one study indicated a disruption in the placebo were compared. Chlorogenic acid second-hour glucose levels in healthy vol- and trigonelline ingestion were associated unteers with a single dose of coffee (18). with significantly lower glucose and insulin Contradictory reports are available regard- concentrations, 15 min after an oral glucose ing the effect of coffee consumption on post- load was given. However, OGTT insulin and prandial glucose levels. Some data indicated glucose under the curve areas were not re- that tolerance to the adverse consequences duced when compared with the placebo. So of coffee on insulin sensitivity may occur as it was claimed that the positive effects on a long term effect. glucose metabolism may be due to chloro- In a randomized controlled clinical trial from genic acid and trigonelline, present in coffee Japan, overweight men were randomly clas- (22). Studies involving animal models pos- sified into three groups: one group receiv- tulated that the reduction of blood glucose ing five cups of caffeinated coffee, the concentration by chlorogenic acid was me- second group consumed five cups of decaf- diated by the activation of adenosine feinated coffee, and the third group did not monophosphate-activated protein kinase

3 (AMPK), affecting fat and glucose metabo- tained from clinical and observational stud- lism at the cellular level (1,23). Evaluation ies. of the effect of chronic consumption of caffeine by animals was found to restore insulin Q2: How coffee consumption affects BMD sensitivity and revert age-induced insulin re- (Bone Mineral Density) levels? sistance in the rat. The effect on Glut 4 Studies pertaining to the effect of coffee on transporters present in the skeletal muscles BMD have contrary results. Studies reported and AMPK activity along with the anti-oxi- positive, negative or neutral effects of cof- dant and anti-inflammatory effects of coffee fee consumption on BMD, as shown in Table were hypothesized to contribute to glucose 1. Most of the studies measured BMD with metabolism (1). QUS (quantitative ultrasound) or with DXA The results of observational and clinical (Dual-energy x-ray absorptiometry) in pre- studies about the effects of coffee on glu- menopausal, postmenopausal women or cose metabolism are contrary. Clinical trials male population. mostly focus on the short term effects of Coffee consumption was negatively corre- coffee on glucose and insulin parameters, lated with the prevalence of osteoporosis in long term randomized controlled studies are postmenopausal and premenopausal women essential to establish the preventive effect (26). of coffee from type 2 diabetes. Yu et al. evaluated the prevalence of osteoporosis among 992 Chinese men, aged be- Conclusion tween 30-90 years. Osteoporosis was The results of the most epidemiologic stud- recorded according to T scores and the ies establish the dose-dependent beneficial prevalence of osteoporosis was found to be effect of coffee consumption on reducing less frequent in Chinese men with moderate type 2 diabetes risk. coffee intake (27). Yang et al. (2015) eval- Although the number of patients was high uated the association between osteoporosis and the follow-up period was long in epi- and coffee consumption in postmenopausal demiological studies, the coffee drinking and women. Osteoporosis in 1,817 participants dietary habits of the participants were eval- was investigated by calcaneus quantitative uated with a validated self-reported ques- ultrasound and T scores were recorded (26). tionnaire. However, lengthy follow up period A study from the Korean premenopausal of these cohort studies, often result in miss population of 1,761 participants failed to as- assessments. Differences in cup sizes and certain the significant correlation between brewing types also need to be taken into BMD and coffee consumption (28). Choi et consideration. al. (2016) explored 4,066 postmenopausal In randomized-controlled studies, there may women according to BMD levels measured be bias as most of the studies in the litera- by DXA and coffee consumption. The partic- ture are not double-blinded. ipants in the highest quartile for coffee con- Coffee consumption failed to portray signif- sumption had 36% lower odds for icant beneficial effects on glucose control in osteoporosis (29). França et al. from Brazil, diabetic patients. found adverse effects of coffee on BMD, in a study population of 156 postmenopausal Coffee and Osteoporosis women (30). A study from Turkey, in 2005, analyzed a study population of 200 post- Q1: Does coffee consumption has an menopausal women and found no associa- association with increased osteoporosis risk? tion between BMD levels and coffee As coffee is one of the most consumed bev- consumption (31). A recent study from the erages, its outcomes on bone health were Taiwanese population examined the associ- also studied extensively. In experimental ation between coffee consumption and T studies, high doses of caffeine were shown scores. The study revealed higher T scores to suppress osteogenesis (24) and increase in the coffee consumer group but half of the osteoclastic formation (25). However, these women in the study were premenopausal, unfavorable consequences on bone health so the evaluation of this study should be were in contradiction with the results ob- done carefully (32).

4 men with moderate coffee men with moderate intake of the prevalence osteoporosis bone exert a negative effect on a negative exert BMD consumption and BMD premenopausal and nutral premenopausal and nutral in postmenopausal consumption no- 3/day effects on protective have 2/day, 3/day2/day, between BMD and coffee by quantitative USby quantitative less frequent in Chinese by quantitative USby quantitative sometimes, always correlated with negatively and osteoporotic 1.593 postmenopausal Study et al.,2016Yu China Country 992 men Sample size Sample type 30-90 years Design Cross-sectional BMD measured at calcaneus Seldom or moderate Osteoporosis diagnosis of osteoporosis Prevalence Coffee dose Result Yang et al., 2015 et al., Yang China 2016Choi et al., 1.817 women Postmenopausal Korea Cross-sectional 4.006 women BMD meaured at calcaneus Postmenopausal Seldom, Cross-sectional DXA L1-4 and femoral Coffee consumption was 9 categories almost Coffee consumption may França et al., 2015 et al., França Brazil 156 women Postmenopausal Cross-sectional DXA L1-4 and femoral food diary 3 day Caffeinated beverages Demirbag et al., 2005Demirbag et al., Turkey 2017Chang et al., 200 women Taiwan Postmenopausal 2.929 women Cross-sectional 1.366 premanopausal, Cross-sectional DXA L1-4 and femoral US Quantitative Cups/day Cup/day No relation between coffe effect in Positive Choi et al., 2014Choi et al., Korea 1.761 women Postmenopausal Cross-sectional DXA L1-4 and femoral 1/day, <1/day, No significant association Table 1. Coffee consumption and risk of osteoporosis. Table

5 Q3: Does coffee consumption have an consumption was found to have a positive association with increased fracture risk? association with a wrist fracture, whereas Conflicting results of clinical studies are there was no significant association with hip available in the literature regarding the ef- or vertebral fracture (41). fect of coffee consumption on the risk of A meta-analysis report documented no sig- fracture. There are large cohort studies that nificant increase in fracture risk with coffee claim increased fracture risk in the high cof- consumption; however, the relative risk in- fee consumer group (>4 cups/day). The creased from the highest to the lowest cate- studies pertaining to coffee consumption gory of coffee intake, when compared with and fracture risk are enlisted in Table 2. those who never drank coffee. The pooled RR A cohort study in the United States (US) for of hip fracture was found to be 1.13 (95%CI: a period of six years among middle-aged 0.86 to 1.48) for individuals with the highest women consuming more than 4 cups of cof- level of coffee consumption (42). fee per day was found to be associated with Q4: What are the possible mechanisms of a three-fold increase in the risk of low im- action of coffee on the bone? pact trauma hip fractures (33). Another cohort study with 31,527 Swedish As the studies in the literature have contra- middle-aged women found a modestly in- dictory results, possible mechanisms for the creased fracture risk in the 4 cups/day or negative effects of coffee were predicted to more coffee consumer group (34). Jokinen be due to increased urinary calcium excre- correlated the risk factors for cervical and tion and decreased intestinal calcium ab- trochanteric fractures with coffee consump- sorption (41). The estrogenic effects, tion of over 5 cups per day and he found in- antioxidant activity or anti-inflammatory creased trochanteric fractures among the property of coffee might contribute to its coffee consumers (35). A recent study from positive effects on bone. Allred et al. China also documented that the coffee in- demonstrated that trigonelline, a compound take of four or more cups was associated present in coffee, can function as an estro- with increased fracture risk in both men and gen receptor agonist in estrogen receptor- women (36). positive breast cancer (29,43). Studies have Literature also contains few studies that also highlighted prominent antioxidant prop- negate the effect of coffee on fracture risk. erty of coffee as compared to grape juice, Halstrom et al. (2012) explored the effects orange, raspberry (44). Chlorogenic acid, an of coffee in a cohort of 61,433 women and ingredient in coffee, was shown to exhibit an high coffee consumption was associated inhibitory effect on osteoclastogenesis with a small reduction in bone density. How- (29,45). The anti-inflammatory activity of ever, no significant association between high coffee was attributed to nitric oxide syn- coffee intake and an increase in fracture risk thase and COX–2 inhibition (46). was witnessed (37). Albrand et al. also failed Earlier studies have mostly been based on to correlate caffeine consumption with frac- food questionnaires such as the number of ture risk (38). cups per day, while randomized controlled Most of the studies focus on the risk of frac- trials are lacking. Therefore, further studies ture in postmenopausal women. Few stud- are needed with detailed information on ies also investigated the risk in men. confounding factors, coffee types, and Halstrom et al., in 2014, examined and longer follow-up periods. found no increase in fracture risk in the coffee consumer group comprising of a cohort Conclusion of 42, 978 middle-aged men (39). Another Cohort studies indicated increased fracture cohort study among the geriatric population risk in high coffee consumer group >4 failed to establish any association between cups/day coffee consumption. hip fractures and caffeine intake (40). The results of the studies should be inter- Another interesting observation was ob- preted carefully as some of the studies em- tained from a clinical study that scrutinized ployed QUS, which is not the gold standard the risk of fracture in postmenopausal for the assessment of BMD. Also, the coffee women based on the fracture site. Coffee intake was assessed with a self-reported

6 increased fracture risk increased fracture osteoporotic fracture risk osteoporotic fracture caffeine or more (>4 cups/day) is associated with modest increased risk of fracture associated with trochanteric fractures risk increased fracture by site risk varying and fracture risk associated with increased fracture risk consumption and fracture postmenopausal 35.298 women Study 1991Hernandez et al., USA 84.484 women Middle aged women Country Sample size Cohort Sample type 6 years Design Cup/day Follow up time Coffee dose increase >4 cups/day Result Hallstrom et al., 2006Hallstrom et al., Sweeden 31.527 women 40-76 aged 2010Jokinen et al., Halstrom et al. 2012 Finland 1.681 women SweedenHansen et al. Cohort 61,443 women Age 70-73 years 10.3 years Middle aged and older 2014Hallstrom et al., Sweeden Cohort USA 2003 et al., Albrand Cohort to 4 times Never 42.978 men 19.4 years of 330 mg Daily intake 10 years FrnaceZhaoli et al. 34.703 women men 45-79 years 55-69 years, 1.039 Cups/day Cups/day Cohort China 11.2 years in bone density but no Reduction 31-89 years Cohort coffee consumption is >5 cups/day 27.959 men, 6.5 years Cups/day 45-74 years Cohort Cups/day not High coffee consumption was 5.3±1.1 years Cohort Cups/day Modest association between coffee 5 years No association between caffeine Cups/day is associated with >4 cups/day Table 2. Coffee consumption and risk of fracture. Table

7 food questionnaire in most of the studies. Coffee Consumption and Thyroid Thus standardization of the cup sizes and brewing types was not considered in most of Q1: Does coffee consumption affect thyroid the studies. function in healthy people? Randomized controlled studies are required. Limited data indicated that coffee has no effect on thyroid function tests in healthy in- Coffee Consumption and Adrenal Hormones dividuals. Limited resources are available regarding the In a prospective study from the Nether- effects of coffee consumption on adrenal hor- lands, 2 gr/day Arabica and Robusta oils mone metabolism. A double-blind, random- were given to 11 healthy volunteers for ized, cross-sectional study investigated the three weeks and serum total and free thy- effect of caffeine ingestion on plasma renin roxine (T4), triiodothyronine (T3), and thy- activity and catecholamines. The results sug- roid-stimulating hormone (TSH) remained gested that caffeine ameliorated the plasma unaffected in all the study subjects (52). renin activity by 57%, plasma norepinephrine To date, no study was found in the literature by 75%, and plasma epinephrine by 207% that correlated the effect of coffee con- (47). Caffeine’s effect on the hypophysis ad- sumption on thyrotoxicosis, goiter or hy- renal axis was investigated previously. Caf- pothyroidism. feine was found to enhance the levels of ACTH (adrenocorticotropic hormone) and cor- Q2: Does coffee consumption affect tisol and remained for 3 h after caffeine in- levothyroxine treatment? gestion. This suggested that caffeine led to a Clinical and experimental data indicated that cortisol rise with the action of ACTH. The coffee may negatively affect levothyroxine blockage of adenosine receptors and interfer- absorption in hypothyroid patients. ence with cyclic adenosine monophosphate In a prospective study in 2008 from Italy, 8 were the possible mechanisms behind this ac- hypothyroid patients and 9 volunteers were tion of caffeine (48). given LT4 with only water or coffee, and A placebo-controlled study explored the ef- water one hour after coffee. Italian-style fects of caffeine on heart rate, blood pres- coffee physically interacted with LT4 and sure, and urinary excretion of reduced the intestinal absorption of LT4 catecholamines at work and home. The re- (53). sults portrayed increased levels of urinary epinephrine in the caffeine group. Blood Q3: Is there a relation between coffee pressure and heart rate were also elevated consumption and thyroid cancer? in this caffeine group (49). Clinical and epidemiological data provided A more recent study that investigated the controversial results regarding the effect of acute effects of caffeine on appetite, inflam- coffee consumption and differentiated thy- mation, and energy intake, reported no ef- roid cancer. fect on these parameters. However, this A total of 24 studies about thyroid cancer study reported high levels of cortisol in the and coffee were found, out of which a total morning in the caffeinated coffee group (50). of 10 studies were included after the exclu- In another randomized, double-blind, sion of duplications (n=8) and ‘not suitable’ placebo-controlled trial, caffeine was found (n=14) articles. to have an insignificant effect on DHEA-S, As documented in Table 3, no association androstenedione levels. Contrary to the pre- between coffee consumption and thyroid vious research, this study failed to display cancer could be extracted from the popula- the ameliorating effect of caffeine on cortisol tion-based cohort studies (54-56) and levels (51). meta-analyses of cohort and case-control studies (57,58). Conclusion While a few case-control studies reported Though coffee consumption failed to show contradictory results. Two case-control stud- promising effects on steroid metabolism, it ies detected coffee consumption reduced the was associated with elevated urine cate- risk of thyroid cancer (59,60) and two stud- cholamine levels. ies [a case-control (61), a cohort study

8 (62)] showed drinking only caffeinated coffee had a preventive effect on thyroid cancer. In all these studies coffee intake was assessed by a food-frequency questionnaire. The mechanism behind this effect was postulated to be increasing intracellular adenosine monophosphate levels. The antioxidant nature of this adenosine monophosphate yielded an inhibitory effect on tumor growth. Only two studies have explored the effect of caffeinated or decaffeinated coffee consumption on thyroid cancer (TC) separately. The first one was a case-control study from German and detected that drinking decaffeinated coffee was associated with an increased risk of TC, whereas drinking caffeinated coffee had a preventive effect on thyroid cancer (57,61). The

and the risk of total differentiated TC, papillary tumor and and the risk of total differentiated TC, follicular tumor in either sex cancer with thyroid intake cancer risk in both for men and women thyroid second one was a prospective cohort study from the USA with 167,720 non-white and white participants with 15.3 years follow-up period. This study also observed an inverse association of coffee consumption with thyroid cancer and TC was prevalent among participants who consumed caffeinated coffee, but not in those who consumed decaffeinated coffee. The associations with TC were mostly man- ifested in whites and it was a dose-dependent relationship (62). A meta-analysis of 7 seven studies (2 prospective cohorts, 5 case-control studies) also failed to detect any significant association of coffee consumption with TC (57). Similar findings were observed in another pooled 4776 controls220.816 controls cancer thyroid cancer thyroid meta-analysis of 14 case-control studies where no consistent association between coffee intake and thyroid cancer could be derived (58).

Conclusion • Results from most of the cohort studies and meta-analyses reported that coffee consumption does not affect the progression of thyroid cancer. European muticentic 2018 476.108 2015USA Japan 2011 2003USA 14 years 97.334 2017Korea 100.507 between coffee consumption observed No association was cancer cases, 2725 thyroid • -Only cancer, 1039 thyroid 10 years two 14.2 years - studies No association with caffeinated and decaffeinated coffee No association between coffee consumption and No association between coffee consumption and discriminated No association between coffee consumption and between caffeinated and decaffeinated coffee and detected

56 that caffeinated coffee had a preventive effect on 54

55 thyroid cancer (57).

Coffee Consumption and Gonads

58 57 We retrieved 159 studies from a literature search in PubMed and Web of Science. After removing duplicates (n=18) and not suitable (n=109), a total of 32 studies were considered to identify the relationship of coffee consumption with the A prospective Cohott study A prospective Cohort Population-based Meta-analyses Study type Cohort European Prospective Date and country Number of subjects Follow up time Results Meta-analyses fertility-fecundability. Table 3. Studies without any significant relationship between coffee consumption and thyroid cancer. significant relationship between coffee consumption and thyroid 3. Studies without any Table

9 Q 1: Does coffee consumption affect female in fecundability with coffee consumption. fertility? Male caffeine intake had no significant influ- Controversial results were reported. Numer- ence on fertilization, pregnancy or live birth ous studies recognized a nonsignificant as- delivery (86,87). sociation between coffee or total caffeine Q3: Does coffee consumption affect gonadal consumption and female fertility, as repre- hormone metabolism? sented in Table 4. Several studies showed coffee or total caf- Coffee consumption positively affected feine consumption led to reduced fertility (63- SHBG concentrations in women (88,89). Nu- 72). Most of this effect was found to be merous studies indicated conflicting results dose-dependent (63,64,66-68,70-72). Hakim between free estradiol level and coffee/caf- et al. (72) evaluated coffee consumption with feine intake. Significant associations be- a dietary diary, whereas, other studies used a tween caffeine consumption and sex questionnaire. Higher fecundability was re- hormone levels in women were detected in ported with moderate caffeine (400-700 some studies (88). Higher caffeine intake mg/day) intake. On the other hand, heavy (≥200 mg/d), which was evaluated by di- caffeine intake (> 700 mg/day) was inversely etary recalls, was inversely associated with proportional to fecundability (73). free estradiol concentrations among white women and positively associated among Q 2: Does coffee consumption affect male Asian women (90). Other studies estimated fertility? higher estrogen levels (88,91) and total A literature survey portrayed a limited num- testosterone (75) resulting from coffee con- ber of studies with men. All of these studies sumption. Kotsopoulos et al. (89) deter- documented coffee consumption by inter- mined the inverse correlation between views. Coffee intake was positively corre- increased caffeine intake and luteal free E2 lated with SHBG (sex hormone-binding levels. Reduced duration of the menstrual globulin) levels in men (74,75). Cross-sec- cycle was observed in subjects with a daily tional studies showed caffeine intake was caffeine consumption >300 mg (92). On the associated with reduced semen volume other hand, androgen or estrogen concen- (76), sperm concentration, total sperm trations remain unaffected as a result of caf- count, and higher (%14) testosterone levels feinated beverage consumption (74,93). (75,77). A pregnancy follow-up cohort Choi et al. (87) determined that higher caf- showed higher maternal coffee consumption feine intake inversely affected peak E2 in at the time of pregnancy was associated women, undergoing infertility treatment. El- with diminished semen volume and testos- evated or insufficient E2 concentrations can terone levels (77). Inversely, current male inhibit ovulation. However, altered E2 levels caffeine intake was associated with in- associated with moderate coffee intake creased (%14) testosterone levels, but with failed to modulate ovulatory function (90). normal semen quality (77). Besides, the caffeine levels in serum and fol- Two cross-sectional studies found no asso- licular fluid were correlated (94). Caffeine ciation between male coffee consumption levels were inversely related to the number and semen quality (78,79). Contrary to this, of eggs. However, the association was not Adelusi et al. (80), Marshburn et al. (81), established between the success rate of and Sobreiro et al. (82) found a positive as- pregnancy and caffeine consumption (94). sociation between coffee consumption and sperm motility. Nonetheless, Parazzini et al. Q4: Does coffee consumption affect (83) reported coffee intake increased the menopause time? propensity of poor semen quality. A cross-sectional study from Norway with Few prospective studies also reported that 2123 women found that coffee consumption, male caffeine consumption was associated as estimated by a questionnaire, remained with reduced fecundability (70,73). Coffee unassociated with early menopause (95). intake might indulge in DNA breaks and Kinney et al. (96), Cramer et al. (97), and aneuploidy (84). Curtis et al. (65), Wes- Nilsson et al. (98) also indicated no associa- selink et al. (85) also found a slight decrease tion between caffeine and age of menopause.

10 Contrary to this, another cross-sectional study found that a higher intake of coffee was inversely associated with later menopause after controlling for age, total energy, parity, menar- che age, and relative weight (99). A case-control study with postmenopausal women found a positive association between SHBG and caffeinated coffee. However, no relation was observed with caffeinated coffee and sex hormones (93). Similar results were also reported in a cross-sectional study with 2377 women (100). Nonselective blocking of adenosine receptors by caffeine, chlorogenic acid, and other phyto- chemicals, resulted in enhanced intracellular concentration of cyclic AMP. As a result of im- proving liver function by blocking liver adeno-

consumption and the risk of primary infertilty in women infertility coffee and risk of ovulatory and more than 8 cups of coffee consumers smokers showed increased infertility risk sine receptors, serum SHBG concentrations may increase. Caffeine and other bioactive substances in coffee are known to possess a high affinity for the estrogen receptor (101). Besides, both caffeine and estradiol are metabolized by the he- patic enzyme CYP1A2, thus a common metabolism pathway may be responsible for the effect of caffeine on estradiol levels (101).

Q5: What are the effects of coffee consumption on pregnancy? Caffeine can traverse the placental barrier and therefore, the fetus gets exposed to the same level as that of the mother. This results in an 662 male non-monotonically associated with fecundability was increased level of catecholamines and cyclic adenosine monophosphate that can impact placental blood flow. As a consequence, high dose coffee consumption (≥300 mg/d) is associated with an increased number of spontaneous abortions (94,101,102). Studies have shown that those who consumed more than 300 mg/day caffeine were associated with fetal growth restriction compared to those consuming below 100 mg/day (103). The European Food Safety Authority (EFSA, 2015) and the WHO recommend a daily caffeine intake below 200 mg and 300 mg, re- spectively.

Conclusion • Relation of coffee consumption with increased SHBG levels, testosterone levels, estradiol level results were conflicting. • Fecundability studies were also contradictory. However, higher consumption was related to A prospective population based Cohort study64A prospective Denmark 2018 7.574 women 20 yrs No association between coffee or total caffeine Study type Cohort study63A prospective 2016 USA 2.135 pregnancy planners, Date and country 12 m Number of subjects Follow up time Results among males, but not females, caffeine intake Total A prospective Cohort study65A prospective Cohort study66A prospective Cohort study67A prospective Cohort study68A prospective Cross sectional Denmark 2012 2011 USA 1998 USA 3.628 pregnancy planners 2009 USA 12 m 470 pregnancy planner 187 pregnancy planner 12 m 18.555 women No monotonic trend of coffee consumption on fecundability 12 m Denmark 1991 No effect of caffeine on fecundability 8 years 10.886 pregnant women No effect of caffeine on fecundability - of coffee or decaffeinated No association between intake No association with only low dose coffee intake, spontaneous abortus. Table 4. Studies without any significant relationship between coffee consumption and female fertility. 4. Studies without any Table

11 Summary 3. Ranheim T, Halvorsen B. Coffee consumption and human health--beneficial or detrimental?--Mecha- Coffee consumption may exert positive ef- nisms for effects of coffee consumption on different fects on reducing type 2 diabetes risk, may risk factors for cardiovascular disease and type 2 di- have negative effects on the bone with in- abetes mellitus. Mol Nutr Food Res. creased risk of fracture over 4 cups/day of 2005;49(3):274-84. 4. Huxley R, Lee CM, Barzi F, Timmermeister L, Czer- coffee consumption, may enhance abortion nichow S, Perkovic V, et al. Coffee, decaffeinated risk, and have no effect on thyroid cancer coffee, and tea consumption in relation to incident development. There are limited data about type 2 diabetes mellitus: a systematic review with the effects of coffee consumption on thyroid meta-analysis. Arch Intern Med. and adrenal hormones. 2009;169(22):2053-63. 5. Bhupathiraju SN, Pan A, Manson JE, Willett WC, van Long term randomized controlled studies are Dam RM, Hu FB. Changes in coffee intake and sub- essential to elucidate any possible biological sequent risk of type 2 diabetes: three large cohorts interactions of coffee on the endocrine sys- of US men and women. Diabetologia. tem. 2014;57(7):1346-54. 6. Carlstrom M, Larsson SC. Coffee consumption and reduced risk of developing type 2 diabetes: a sys- Source of Finance tematic review with meta-analysis. Nutr Rev. During this study, no financial or spiritual 2018;76(6):395-417. support was received neither from any phar- 7. van Dam RM, Willett WC, Manson JE, Hu FB. Cof- maceutical company that has a direct con- fee, caffeine, and risk of type 2 diabetes: a prospec- nection with the research subject, nor from tive cohort study in younger and middle-aged U.S. women. Diabetes Care. 2006;29(2):398-403. a company that provides or produces med- 8. Ding M, Bhupathiraju SN, Chen M, van Dam RM, Hu ical instruments and materials which may FB. Caffeinated and decaffeinated coffee consump- negatively affect the evaluation process of tion and risk of type 2 diabetes: a systematic re- this study. view and a dose-response meta-analysis. Diabetes Care. 2014;37(2):569-86. 9. Jiang X, Zhang D, Jiang W. Coffee and caffeine in- Conflict of Interest take and incidence of type 2 diabetes mellitus: a No conflicts of interest between the authors meta-analysis of prospective studies. Eur J Nutr. and / or family members of the scientific 2014;53(1):25-38. and medical committee members or mem- 10.Mirmiran P, Carlstrom M, Bahadoran Z, Azizi F. bers of the potential conflicts of interest, Long-term effects of coffee and caffeine intake on the risk of pre-diabetes and type 2 diabetes: Find- counseling, expertise, working conditions, ings from a population with low coffee consumption. share holding and similar situations in any Nutr Metab Cardiovasc Dis. 2018;28(12):1261-6. firm. 11. Hamer M, Witte DR, Mosdol A, Marmot MG, Brunner EJ. Prospective study of coffee and tea consumption Authorship Contributions in relation to risk of type 2 diabetes mellitus among men and women: the Whitehall II study. Br J Nutr. Idea/Concept: Dilek Gogas Yavuz; Design: 2008;100(5):1046-53. Dilek Gogas Yavuz; Control/Supervision: 12. Salazar-Martinez E, Willett WC, Ascherio A, Manson Dilek Gogas Yavuz; Data Collection and/or JE, Leitzmann MF, Stampfer MJ, et al. Coffee con- Processing: Dilek Gogas Yavuz, Ceyda sumption and risk for type 2 diabetes mellitus. Ann Dinçer, Tuğçe Apaydın; Analysis and/or In- Intern Med. 2004;140(1):1-8. 13.Pereira MA, Parker ED, Folsom AR. Coffee con- terpretation: Dilek Gogas Yavuz, Ceyda sumption and risk of type 2 diabetes mellitus: an Dinçer, Tuğçe Apaydın; Literature Review: 11-year prospective study of 28 812 post- Dilek Gogas Yavuz, Ceyda Dinçer, Tuğçe menopausal women. Arch Intern Med. Apaydın; Writing the Article: Ceyda Dinçer, 2006;166(12):1311-6. Tuğçe Apaydın; Critical Review: Ceyda 14. van Dieren S, Uiterwaal CS, van der Schouw YT, van der AD, Boer JM, Spijkerman A, et al. Coffee and Dinçer, Tuğçe Apaydın. tea consumption and risk of type 2 diabetes. Dia- betologia. 2009;52(12):2561-9. References 15. Lane JD, Barkauskas CE, Surwit RS, Feinglos MN. 1. Reis CEG, Dorea JG, da Costa THM. Effects of coffee Caffeine impairs glucose metabolism in type 2 dia- consumption on glucose metabolism: A systematic betes. Diabetes Care. 2004;27(8):2047-8. review of clinical trials. J Tradit Complement Med. 16. van Dam RM, Dekker JM, Nijpels G, Stehouwer CD, 2019;9(3):184-91. Bouter LM, Heine RJ, et al. Coffee consumption and 2. Cornelis MC, El-Sohemy A. Coffee, caffeine, and incidence of impaired fasting glucose, impaired glu- coronary heart disease. Curr Opin Clin Nutr Metab cose tolerance, and type 2 diabetes: the Hoorn Care. 2007;10(6):745-51. Study. Diabetologia. 2004;47(12):2152-9.

12 17. Gao F, Zhang Y, Ge S, Lu H, Chen R, Fang P, et al. 31.Demirbag D, Ozdemir F, Ture M. Effects of coffee Coffee consumption is positively related to insulin consumption and smoking habit on bone mineral secretion in the Shanghai High-Risk Diabetic Screen density. Rheumatol Int. 2006;26(6):530-5. (SHiDS) Study. Nutr Metab (Lond). 2018;15:84. 32.Chang HC, Hsieh CF, Lin YC, Tantoh DM, Ko PC, 18.Robertson TM, Clifford MN, Penson S, Chope G, Kung YY, et al. Does coffee drinking have beneficial Robertson MD. A single serving of caffeinated coffee effects on bone health of Taiwanese adults? A lon- impairs postprandial glucose metabolism in over- gitudinal study. BMC Public Health. weight men. Br J Nutr. 2015;114(8):1218-25. 2018;18(1):1273. 19. Ohnaka K, Ikeda M, Maki T, Okada T, Shimazoe T, 33. Hernandez-Avila M, Colditz GA, Stampfer MJ, Ros- Adachi M, et al. Effects of 16-week consumption of ner B, Speizer FE, Willett WC. Caffeine, moderate caffeinated and decaffeinated instant coffee on glu- alcohol intake, and risk of fractures of the hip and cose metabolism in a randomized controlled trial. J forearm in middle-aged women. Am J Clin Nutr. Nutr Metab. 2012;2012:207426. 1991;54(1):157-63. 20. van Dam RM, Pasman WJ, Verhoef P. Effects of cof- 34. Hallstrom H, Wolk A, Glynn A, Michaelsson K. Cof- fee consumption on fasting blood glucose and in- fee, tea and caffeine consumption in relation to os- sulin concentrations: randomized controlled trials in teoporotic fracture risk in a cohort of Swedish healthy volunteers. Diabetes Care. women. Osteoporos Int. 2006;17(7):1055-64. 2004;27(12):2990-2. 35. Jokinen H, Pulkkinen P, Korpelainen J, Heikkinen J, 21. Wedick NM, Brennan AM, Sun Q, Hu FB, Mantzoros Keinanen-Kiukaanniemi S, Jamsa T, et al. Risk fac- CS, van Dam RM. Effects of caffeinated and decaf- tors for cervical and trochanteric hip fractures in feinated coffee on biological risk factors for type 2 elderly women: a population-based 10-year follow- diabetes: a randomized controlled trial. Nutr J. up study. Calcif Tissue Int. 2010;87(1):44-51. 2011;10:93. 36. Dai Z, Jin A, Soh AZ, Ang LW, Yuan JM, Koh WP. Cof- 22. van Dijk AE, Olthof MR, Meeuse JC, Seebus E, Heine fee and tea drinking in relation to risk of hip fracture RJ, van Dam RM. Acute effects of decaffeinated cof- in the Singapore Chinese Health Study. Bone. fee and the major coffee components chlorogenic 2018;112:51-7. acid and trigonelline on glucose tolerance. Diabetes 37. Hallstrom H, Byberg L, Glynn A, Lemming EW, Wolk Care. 2009;32(6):1023-5. A, Michaelsson K. Long-term coffee consumption in 23. Tajik N, Tajik M, Mack I, Enck P. The potential ef- relation to fracture risk and bone mineral density in fects of chlorogenic acid, the main phenolic compo- women. Am J Epidemiol. 2013;178(6):898-909. nents in coffee, on health: a comprehensive review 38.Albrand G, Munoz F, Sornay-Rendu E, DuBoeuf F, of the literature. Eur J Nutr. 2017;56(7):2215-44. Delmas PD. Independent predictors of all osteo- 24.Su SJ, Chang KL, Su SH, Yeh YT, Shyu HW, Chen porosis-related fractures in healthy postmenopausal KM. Caffeine regulates osteogenic differentiation women: the OFELY study. Bone. 2003;32(1):78-85. and mineralization of primary adipose-derived stem 39. Hallstrom H, Wolk A, Glynn A, Michaelsson K, By- cells and a bone marrow stromal cell line. Int J Food berg L. Coffee consumption and risk of fracture in Sci Nutr. 2013;64(4):429-36. the Cohort of Swedish Men (COSM). PLoS One. 25. Liu SH, Chen C, Yang RS, Yen YP, Yang YT, Tsai C. 2014;9(5):e97770. Caffeine enhances osteoclast differentiation from 40. Cauley JA, Cawthon PM, Peters KE, Cummings SR, bone marrow hematopoietic cells and reduces bone Ensrud KE, Bauer DC, et al. Risk Factors for Hip mineral density in growing rats. J Orthop Res. Fracture in Older Men: The Osteoporotic Fractures 2011;29(6):954-60. in Men Study (MrOS). J Bone Miner Res. 26.Yang P, Zhang XZ, Zhang K, Tang Z. Associations 2016;31(10):1810-9. between frequency of coffee consumption and os- 41. Hansen SA, Folsom AR, Kushi LH, Sellers TA. Asso- teoporosis in Chinese postmenopausal women. Int ciation of fractures with caffeine and alcohol in post- J Clin Exp Med. 2015;8(9):15958-66. menopausal women: the Iowa Women's Health 27. Yu Q, Liu ZH, Lei T, Tang Z. Subjective evaluation of Study. Public Health Nutr. 2000;3(3):253-61. the frequency of coffee intake and relationship to 42. Li S, Dai Z, Wu Q. Effect of coffee intake on hip frac- osteoporosis in Chinese men. J Health Popul Nutr. ture: a meta-analysis of prospective cohort studies. 2016;35(1):24. Nutr J. 2015;14:38. 28. Choi EJ, Kim KH, Koh YJ, Lee JS, Lee DR, Park SM. 43.Allred KF, Yackley KM, Vanamala J, Allred CD. Coffee consumption and bone mineral density in ko- Trigonelline is a novel phytoestrogen in coffee rean premenopausal women. Korean J Fam Med. beans. J Nutr. 2009;139(10):1833-8. 2014;35(1):11-8. 44.Halvorsen BL, Carlsen MH, Phillips KM, Bohn SK, 29. Choi E, Choi KH, Park SM, Shin D, Joh HK, Cho E. Holte K, Jacobs DR, Jr., et al. Content of redox-ac- The Benefit of Bone Health by Drinking Coffee tive compounds (ie, antioxidants) in foods con- among Korean Postmenopausal Women: A Cross- sumed in the United States. Am J Clin Nutr. Sectional Analysis of the Fourth & Fifth Korea Na- 2006;84(1):95-135. tional Health and Nutrition Examination Surveys. 45. Tang QY, Kukita T, Ushijima Y, Kukita A, Nagata K, PLoS One. 2016;11(1):e0147762. Sandra F, et al. Regulation of osteoclastogenesis by 30. de Franca NA, Camargo MB, Lazaretti-Castro M, Pe- Simon extracts composed of caffeic acid and related ters BS, Martini LA. Dietary patterns and bone min- compounds: successful suppression of bone de- eral density in Brazilian postmenopausal women struction accompanied with adjuvant-induced with osteoporosis: a cross-sectional study. Eur J Clin arthritis in rats. Histochem Cell Biol. Nutr. 2016;70(1):85-90. 2006;125(3):215-25.

13 46. Kim JY, Kim DH, Jeong HG. Inhibitory effect of the among women in Tokai, Japan. J Epidemiol. coffee diterpene kahweol on carrageenan-induced 1996;6(3):140-7. inflammation in rats. Biofactors. 2006;26(1):17- 61. Frentzel-Beyme R, Helmert U. Association between 28. malignant tumors of the thyroid gland and exposure 47. Robertson D, Frolich JC, Carr RK, Watson JT, Holli- to environmental protective and risk factors. Rev field JW, Shand DG, et al. Effects of caffeine on Environ Health. 2000;15(3):337-58. plasma renin activity, catecholamines and blood 62. Park SY, Freedman ND, Haiman CA, Le Marchand L, pressure. N Engl J Med. 1978;298(4):181-6. Wilkens LR, Setiawan VW. Prospective Study of Cof- 48. Lovallo WR, Al'Absi M, Blick K, Whitsett TL, Wilson fee Consumption and Cancer Incidence in Non- MF. Stress-like adrenocorticotropin responses to White Populations. Cancer Epidemiol Biomarkers caffeine in young healthy men. Pharmacol Biochem Prev. 2018;27(8):928-35. Behav. 1996;55(3):365-9. 63. Wilcox A, Weinberg C, Baird D. Caffeinated bever- 49.Lane JD, Pieper CF, Phillips-Bute BG, Bryant JE, ages and decreased fertility. Lancet. 1988;2(8626- Kuhn CM. Caffeine affects cardiovascular and neu- 8627):1453-6. roendocrine activation at work and home. Psycho- 64. Hassan MA, Killick SR. Negative lifestyle is associ- som Med. 2002;64(4):595-603. ated with a significant reduction in fecundity. Fertil 50. Gavrieli A, Yannakoulia M, Fragopoulou E, Margari- Steril. 2004;81(2):384-92. topoulos D, Chamberland JP, Kaisari P, et al. Caf- 65. Curtis KM, Savitz DA, Arbuckle TE. Effects of ciga- feinated coffee does not acutely affect energy rette smoking, caffeine consumption, and alcohol intake, appetite, or inflammation but prevents intake on fecundability. Am J Epidemiol. serum cortisol concentrations from falling in healthy 1997;146(1):32-41. men. J Nutr. 2011;141(4):703-7. 66. Grodstein F, Goldman MB, Ryan L, Cramer DW. Re- 51. MacKenzie T, Comi R, Sluss P, Keisari R, Manwar S, lation of female infertility to consumption of caf- Kim J, et al. Metabolic and hormonal effects of caffeinated beverages. Am J Epidemiol. feine: randomized, double-blind, placebo-controlled 1993;137(12):1353-60. crossover trial. Metabolism. 2007;56(12):1694-8. 67. Hatch EE, Bracken MB. Association of delayed con- 52. Mensink RP, Lebbink WJ, Lobbezoo IE, Weusten-Van ception with caffeine consumption. Am J Epidemiol. der Wouw MP, Zock PL, Katan MB. Diterpene com- 1993;138(12):1082-92. position of oils from Arabica and Robusta coffee 68. Stanton CK, Gray RH. Effects of caffeine consump- beans and their effects on serum lipids in man. J In- tion on delayed conception. Am J Epidemiol. tern Med. 1995;237(6):543-50. 1995;142(12):1322-9. 53. Benvenga S, Bartolone L, Pappalardo MA, Russo A, 69. Williams MA, Monson RR, Goldman MB, Mittendorf Lapa D, Giorgianni G, et al. Altered intestinal absorption of L-thyroxine caused by coffee. Thyroid. R, Ryan KJ. Coffee and delayed conception. Lancet. 2008;18(3):293-301. 1990;335(8705):1603. 54. Hashibe M, Galeone C, Buys SS, Gren L, Boffetta P, 70. Jensen TK, Henriksen TB, Hjollund NH, Scheike T, Zhang ZF, et al. Coffee, tea, caffeine intake, and the Kolstad H, Giwercman A, et al. Caffeine intake and risk of cancer in the PLCO cohort. Br J Cancer. fecundability: a follow-up study among 430 Danish 2015;113(5):809-16. couples planning their first pregnancy. Reprod Tox- 55.Michikawa T, Inoue M, Shimazu T, Sasazuki S, icol. 1998;12(3):289-95. Iwasaki M, Sawada N, et al. Green tea and coffee 71. Bolumar F, Olsen J, Rebagliato M, Bisanti L. Caffeine consumption and its association with thyroid cancer intake and delayed conception: a European multi- risk: a population-based cohort study in Japan. center study on infertility and subfecundity. Euro- Cancer Causes Control. 2011;22(7):985-93. pean Study Group on Infertility Subfecundity. Am J 56.Zamora-Ros R, Alghamdi MA, Cayssials V, Epidemiol. 1997;145(4):324-34. Franceschi S, Almquist M, Hennings J, et al. Coffee 72. Hakim RB, Gray RH, Zacur H. Alcohol and caffeine and tea drinking in relation to the risk of differenti- consumption and decreased fertility. Fertil Steril. ated thyroid carcinoma: results from the European 1998;70(4):632-7. Prospective Investigation into Cancer and Nutrition 73. Florack EI, Zielhuis GA, Rolland R. Cigarette smok- (EPIC) study. Eur J Nutr. 2018. ing, alcohol consumption, and caffeine intake and 57. Han MA, Kim JH. Coffee Consumption and the Risk fecundability. Prev Med. 1994;23(2):175-80. of Thyroid Cancer: A Systematic Review and Meta- 74. Frey T, Platz EA, Kanarek N, Bradwin G, Dobs AS, Analysis. Int J Environ Res Public Health. Rohrmann S. Consumption of caffeinated beverages 2017;14(2). and serum concentrations of sex steroid hormones 58. Mack WJ, Preston-Martin S, Dal Maso L, Galanti R, in US men. Cancer Causes Control. Xiang M, Franceschi S, et al. A pooled analysis of 2018;29(1):157-66. case-control studies of thyroid cancer: cigarette 75.Svartberg J, Midtby M, Bonaa KH, Sundsfjord J, smoking and consumption of alcohol, coffee, and Joakimsen RM, Jorde R. The associations of age, tea. Cancer Causes Control. 2003;14(8):773-85. lifestyle factors and chronic disease with testos- 59.Linos A, Linos DA, Vgotza N, Souvatzoglou A, terone in men: the Tromso Study. Eur J Endocrinol. Koutras DA. Does coffee consumption protect 2003;149(2):145-52. against thyroid disease? Acta Chir Scand. 76. Jensen TK, Swan SH, Skakkebaek NE, Rasmussen 1989;155(6-7):317-20. S, Jorgensen N. Caffeine intake and semen quality 60. Takezaki T, Hirose K, Inoue M, Hamajima N, Kuroishi in a population of 2,554 young Danish men. Am J T, Nakamura S, et al. Risk factors of thyroid cancer Epidemiol. 2010;171(8):883-91.

14 77. Ramlau-Hansen CH, Thulstrup AM, Bonde JP, Olsen take and reproductive hormones among pre- J, Bech BH. Semen quality according to prenatal menopausal women in the BioCycle Study. Am J Clin coffee and present caffeine exposure: two decades Nutr. 2012;95(2):488-97. of follow-up of a pregnancy cohort. Hum Reprod. 91.Lucero J, Harlow BL, Barbieri RL, Sluss P, Cramer 2008;23(12):2799-805. DW. Early follicular phase hormone levels in relation 78. Oldereid NB, Rui H, Purvis K. Lifestyles of men in to patterns of alcohol, tobacco, and coffee use. Fer- barren couples and their relationships to sperm til Steril. 2001;76(4):723-9. quality. Eur J Obstet Gynecol Reprod Biol. 92. Fenster L, Quale C, Waller K, Windham GC, Elkin EP, 1992;43(1):51-7. Benowitz N, et al. Caffeine consumption and men- 79. Horak S, Polanska J, Widlak P. Bulky DNA adducts in strual function. Am J Epidemiol. 1999;149(6):550- human sperm: relationship with fertility, semen 7. quality, smoking, and environmental factors. Mutat 93. Goto A, Song Y, Chen BH, Manson JE, Buring JE, Liu Res. 2003;537(1):53-65. S. Coffee and caffeine consumption in relation to 80. Adelusi B, al-Twaijiri MH, al-Meshari A, Kangave D, sex hormone-binding globulin and risk of type 2 di- al-Nuaim LA, Younnus B. Correlation of smoking and abetes in postmenopausal women. Diabetes. coffee drinking with sperm progressive motility in 2011;60(1):269-75. infertile males. Afr J Med Med Sci. 1998;27(1- 94. Al-Saleh I, El-Doush I, Grisellhi B, Coskun S. The 2):47-50. effect of caffeine consumption on the success rate 81.Marshburn PB, Sloan CS, Hammond MG. Semen of pregnancy as well various performance parame- quality and association with coffee drinking, ciga- ters of in-vitro fertilization treatment. Med Sci rette smoking, and ethanol consumption. Fertil Monit. 2010;16(12):CR598-605. Steril. 1989;52(1):162-5. 95. Mikkelsen TF, Graff-Iversen S, Sundby J, Bjertness 82.Sobreiro BP, Lucon AM, Pasqualotto FF, Hallak J, E. Early menopause, association with tobacco Athayde KS, Arap S. Semen analysis in fertile pa- smoking, coffee consumption and other lifestyle fac- tients undergoing vasectomy: reference values and tors: a cross-sectional study. BMC Public Health. variations according to age, length of sexual absti- 2007;7:149. nence, seasonality, smoking habits and caffeine in- 96.Kinney A, Kline J, Levin B. Alcohol, caffeine and take. Sao Paulo Med J. 2005;123(4):161-6. smoking in relation to age at menopause. Maturi- 83.Parazzini F, Marchini M, Tozzi L, Mezzopane R, tas. 2006;54(1):27-38. Fedele L. Risk factors for unexplained dyspermia in 97. Cramer DW, Harlow BL, Xu H, Fraer C, Barbieri R. infertile men: a case-control study. Arch Androl. Cross-sectional and case-controlled analyses of the 1993;31(2):105-13. association between smoking and early menopause. 84. Ricci E, Vigano P, Cipriani S, Somigliana E, Chiaf- Maturitas. 1995;22(2):79-87. farino F, Bulfoni A, et al. Coffee and caffeine intake 98. Nilsson P, Moller L, Koster A, Hollnagel H. Social and and male infertility: a systematic review. Nutr J. biological predictors of early menopause: a model 2017;16(1):37. for premature aging. J Intern Med. 85.Wesselink AK, Wise LA, Rothman KJ, Hahn KA, 1997;242(4):299-305. Mikkelsen EM, Mahalingaiah S, et al. Caffeine and 99.Nagata C, Takatsuka N, Inaba S, Kawakami N, caffeinated beverage consumption and fecundability Shimizu H. Association of diet and other lifestyle in a preconception cohort. Reprod Toxicol. with onset of menopause in Japanese women. Ma- 2016;62:39-45. turitas. 1998;29(2):105-13. 86.Klonoff-Cohen H, Bleha J, Lam-Kruglick P. A 100. Pihan-Le Bars F, Gusto G, Boutron-Ruault MC, prospective study of the effects of female and male Fagherazzi G, Bonnet F. Cross-sectional association caffeine consumption on the reproductive endpoints of coffee and caffeine consumption with sex hor- of IVF and gamete intra-Fallopian transfer. Hum Re- mone-binding globulin in healthy nondiabetic prod. 2002;17(7):1746-54. women. Clin Endocrinol (Oxf). 2017;87(5):475- 87. Choi JH, Ryan LM, Cramer DW, Hornstein MD, Miss- 83. mer SA. Effects of Caffeine Consumption by Women 101. Lyngso J, Ramlau-Hansen CH, Bay B, Ingerslev and Men on the Outcome of In Vitro Fertilization. J HJ, Hulman A, Kesmodel US. Association between Caffeine Res. 2011;1(1):29-34. coffee or caffeine consumption and fecundity and 88.Ferrini RL, Barrett-Connor E. Caffeine intake and fertility: a systematic review and dose-response endogenous sex steroid levels in postmenopausal meta-analysis. Clin Epidemiol. 2017;9:699-719. women. The Rancho Bernardo Study. Am J Epi- 102. Greenwood DC, Thatcher NJ, Ye J, Garrard L, demiol. 1996;144(7):642-4. Keogh G, King LG, et al. Caffeine intake during 89. Kotsopoulos J, Eliassen AH, Missmer SA, Hankinson pregnancy and adverse birth outcomes: a system- SE, Tworoger SS. Relationship between caffeine in- atic review and dose-response meta-analysis. Eur J take and plasma sex hormone concentrations in Epidemiol. 2014;29(10):725-34. premenopausal and postmenopausal women. Can- 103. Group CS. Maternal caffeine intake during preg- cer. 2009;115(12):2765-74. nancy and risk of fetal growth restriction: a large 90.Schliep KC, Schisterman EF, Mumford SL, Pollack prospective observational study. BMJ. AZ, Zhang C, Ye A, et al. Caffeinated beverage in- 2008;337:a2332.