Gestational diabetes mellitus (GDM) is a disease in which pregnant women who did not have diabetes before pregnancy develops glucose intolerance because of the interference of pregnancy hormones with the action of insulin (1, 2). It has been estimated that roughly 15 to 20% of pregnant women were affected by GDM (3). High gestational age, obesity, polycystic ovary syndrome, ethnicity, glycosuria, family history of diabetes, and previous history of GDM are underlying risk factors of GDM (4). Approximately 70% of women with GDM will be affected by type 2 diabetes mellitus (T2DM) later in their life (5). Infants born to mothers with GDM are at increased risk of macrosomia, hypoglycemia, jaundice, and epigenetic changes (6). In the long term, they are prone to being obese or diabetic in childhood (7). Therefore, the identification of preventive strategies to reduce GDM has great importance.

Diet and physical activity are fundamental lifestyle interventions to control GDM (7). Using foods with a low glycemic index, high antioxidants, and also decrease in intake of energy, distribution of carbohydrates, and intake of fat or protein modification share some dietary recommendations for women with GDM (8–11).

Coffee is commonly drunk among women aged 20–50 years, and thus their possible effects on GDM absorb many interests (12). A meta-analysis of prospective cohort studies exhibited both the caffeinated and decaffeinated coffee was associated with reduced diabetes risk (13). It appears that the influence of intake of coffee on GDM may vary from T2DM in non-pregnant women. Metabolism of caffeine as the main phytochemical in coffee decreases during pregnancy (14), and subsequently, a high amount of caffeine levels in blood was related to elevating insulin resistance (15). It seems that the favorable effects of coffee are not linked to caffeine and its metabolites during pregnancy, and originated from antioxidants and prebiotic compounds, including phenolic components and micronutrients, which can improve insulin sensitivity and glycemic response (16).

The findings of studies regarding the association of intake of coffee with GDM are conflicting. A double-blind, randomized crossover study indicated that acute caffeine ingestion impairs insulin sensitivity in women with GDM (17). Furthermore, some observational studies displayed no relationship between intake of coffee and GDM (18–21), while two studies showed inverse associations (22, 23).

Based on our knowledge, there is no systematic review and meta-analysis to clarify the association between the consumption of coffee and the risk of GDM. Therefore, the purpose of this study is to achieve a solid response to this question: Is consumption of coffee associated with the risk of GDM in pregnant women?

The protocol of this study has been established based on Preferred Reporting Items of Systematic Reviews and Meta-Analysis (PRISMA) criteria (24).

The flowchart of study selection was displayed in Figure 1. The title and abstract of 162 records identified through the initial search were screened following inclusion and exclusion criteria. After deleting unrelated papers, the full-texts of a total of 17 remained articles were checked. In this stage, 10 articles were omitted because of the following reasons: irrelevant articles (n = 9) and consideration of both tea and coffee as exposure (n = 1). Finally, seven epidemiologic studies possessed eligibility to this study (18–23, 30).

The brief information of each selected study has been described in Table 1. Three cohort studies (18, 19, 30), two case-control studies (20, 22), and two cross-sectional studies (21, 23) were imported to meta-analysis. The studies were published between 2007 and 2021, and these were conducted in the US (18), Denmark (19), Indonesia (22), Oman (20), Spain (21), Malaysia (30), and Ethiopia (23). Women with GDM were aged between 13 and 49 years. A total of 75,607 pregnant women participated, 1,625 of whom were diagnosed with GDM. Three studies used a food frequency questionnaire (FFQ) (18, 21, 30), one 24-h food recall (23), and others used a dietary questionnaire (19, 20, 22). The method of GDM diagnosis was oral glucose tolerance test (OGTT) for four studies (18, 21, 23, 30) and medical records (22) or questionnaire (19, 20) for the others. Four studies controlled the covariates (18, 19, 23, 30), and others did not adjust. Three studies had high-methodological quality (score ≥ 7) (18, 19, 23) (Table 1).

This meta-analysis of seven observational studies demonstrated that intake of coffee is not linked with GDM. However, a significant inverse association was indicated between intake of coffee and GDM in studies conducted in non-Asia countries. To date, no systematic review and meta-analysis addressing a correlation between intake of coffee and GDM.

Coffee is a common drink all over the world and consists of a combination of antioxidants and micronutrients, which possess favorable effects on cardiovascular disease (31, 32). In this review, we saw no significant reduced risk of GDM among high-coffee consumers compare to those with low intake. Inline with our finding, a cross-sectional study conducted on 785 adult pregnant women in São Paulo illustrated no association between consumption of coffee and tea and GDM (25). In 2018, one meta-analysis investigated the association of poly-phenol-rich foods and the risk of gestational diabetes. This study captured no connection between non-alcoholic beverages (coffee, tea, and juice) and GDM (33). On the other hand, the evidence shows a relationship between GDM and T2DM, and women with GDM are prone to T2D in the future (7). A recent meta-analysis of 30 cohort studies depicts that the risk of T2D lowered by 6% for each cup-per-day increment in consumption of coffee (34). Furthermore, one cup increase of caffeinated and decaffeinated consumption of coffee in a day leads to a 7 and 6% reduced risk of diabetes, respectively (13). Moreover, a clinical trial study found that intake of caffeine was related to impaired insulin sensitivity in women with GDM early in their third trimester but not in controls without GDM (17).

Some potential reasons can be considered for these inconsistencies. First, intake of caffeine during pregnancy has been related to adverse outcomes including low birth weight, elevated risk of delivering, impaired fetal length growth, and an infant with small-for-gestational-age (35–37). According to the guidelines of the American College of Obstetrics and Gynecology (ACOG), daily intake of up to two cups of moderate-strength coffee may be safe for pregnant women (38). It has been demonstrated that pregnant women significantly decline their intakes of caffeine, particularly coffee during pregnancy (39, 40). The low intake of coffee among pregnant women in the most included studies may explain finding any significant association. This study showed an inverse association between intake of coffee and risk of GDM among pregnant women who lived in the western countries. People in the western countries consume higher coffee as compared with Asia countries. In the United States, about 25% of women aged between 20 and 29 years, and 46% of women aged between 30 and 39 years, drink coffee daily (41).

Furthermore, studies reported the amount of coffee used as “cups” without expressing the volume of the cup. Therefore, the lack of standardization in the measurement of coffee must be taken into account. Besides, coffee is commonly drunk along with sugar, creamer, or milk, which attenuates the beneficial impact of coffee on GDM since fructose worsens hepatic insulin resistance (42). Finally, the degree of roasting and the method of preparation of coffee including coffee-grind setting and brew type may be different in varied populations.

Caffeine and its metabolites such as paraxanthine accumulate in the blood during pregnancy because of a decrease in the metabolism of caffeine (14). High concentrations of coffee and its metabolites have been associated with elevated insulin resistance during pregnancy (15). Therefore, it appears that the favorable effect of coffee on GDM may originate from components of non-caffeine, including micronutrients and phenolic compounds, which have antioxidant and prebiotic traits (16). The evidence indicated that coffee constitutes 60%, or nearly 11.1 mmol, of the total intake of daily antioxidant (43). Chlorogenic acid is the main polyphenol in coffee and is assumed to be the main antioxidant of coffee (44). Coffee also contains potassium, magnesium, niacin, and other antioxidants, which may have beneficial effects on glucose metabolism and insulin resistance (45). Furthermore, habitual intake of coffee declines subclinical inflammation and augments adiponectin levels (46, 47) that may improve insulin sensitivity (48).

This study has some limitations that should be considered once we interpret the findings. First, the small number of studies and a moderate amount of heterogeneity among them are the main limitations. We tried to find the sources of heterogeneity through subgroup analysis. Second, four of the seven included studies have case-control or cross-sectional designs. These kinds of studies possess some biases such as recall and selection biases. Third, measurement errors may occur when estimating consumption of coffee, particularly for those with low intake. Fourth, studies applied different methods for the diagnosis of GDM. Fifth, some studies did not control covariates, or they may ignore to adjust some underlying residual confounding. For example, only three studies controlled energy intake, which is an important risk factor for GDM. Moreover, most studies did not have any information on the intake of coffee of women before pregnancy. Finally, a lack of control for these dietary factors could disable us to identify a firm finding.

In conclusion, this systematic review and meta-analysis of seven observational studies have depicted no relationship between extreme intake of coffee and the risk of GDM. Furthermore, prospective cohort studies with a large sample size are obligatory to understand the relationship between coffee and GDM. Given that, the circulating level of caffeine and its metabolites are different in patients with GDM, future studies are required to examine biomarkers of coffee and its metabolites, including serum caffeine and paraxanthine.

JN and PW designed the work and extracted the data. TZ and LL analyzed the data. HP wrote the first draft of the manuscript. All authors critically read and approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.