Breastfeeding was associated with lower risks of maternal postpartum hypertension and diabetes mellitus: A systematic review and meta-analysis.

This study was a systematic review and meta-analysis that followed the PRISMA 2009 statement[ 33 ]. Inclusion criteria: The study design was comparative study (breastfeeding vs non-breastfeeding). The study population comprised parous women. This study investigated the association between breastfeeding and the maternal risk of type 2 diabetes or hypertension. The risk of diabetes or hypertension in parous women was measured using the HR or OR with 95%CIs for a defined duration in months of breastfeeding vs no breastfeeding, or any breastfeeding vs no breastfeeding. Papers were published in English. Exclusion criteria: Articles that were not original, such as reviews, editorials, letters, and protocols were excluded. Articles that did not compare breastfeeding to non-breastfeeding were excluded ( e.g. , breastfeeding compared to partial breastfeeding). Articles that did not include women who have given birth as a part of the population were excluded. Articles that did not include HR or OR for diabetes or hypertension were excluded. The Medline, EMBASE, PubMed, Web of Science, and Cochrane databases were searched in 2024, with the search running until May 6, 2024. The following key terms were used in the search: “Breastfeeding”, “Lactation”, “Diabetes Mellitus” and “Hypertension”. The search strategy used to search the databases is detailed in the Supplementary material . All the studies identified in the search were downloaded from EndNote. After duplicates were eliminated, two independent reviewers (Zheng SW and Lin XY) screened the abstracts and titles of the remaining studies. They then examined the full texts of potentially relevant studies to select the final group of eligible studies. Disputes between the two independent reviewers were resolved through discussion and, if necessary, by consulting a third reviewer. A pretest data collection form was used to extract data from studies that were included by two independent reviewers (Zheng SW and Lin XY). The data collection included the following items: (1) Basic characteristics of the included studies, demographic information of the target population; (2) Disease reported (diabetes or hypertensions); (3) Breast feeding durations and accordingly HR/OR of diabetes/hypertensions; (4) Data of breastfeeding, data of outcome, adjustment factors (key covariates used in adjusted models). The discrepancies in data extraction were resolved through discussions within the research team, following a careful review of the original papers. If there is still a deficiency in information, we contacted the corresponding authors; and (5) All the OR-related data were cross-sectional data, even though the data were extracted from longitudinal design studies. In comparison, the HR-related data all came from longitudinal studies. The primary outcomes were the occurrence and incidence of diabetes and hypertension in women after giving birth. We collected data on diabetes/hypertension OR/HR for Breastfeeding vs not breastfeeding; diabetes/hypertension OR/HR for breastfeeding 12 months vs not breastfeeding. Risk of bias in non-randomized studies-of exposure (ROBINS-E) (2022; Launch version: https://www.riskofbias.info/welcome/robins-e-tool ) was used to assess bias in all included studies[ 34 ]. Two independent researchers (Zheng SW and Lin XY) reviewed these studies. Disputes between the two independent reviewers were resolved through discussion and, if necessary, by consulting a third reviewer. RevMan 5 was used for synthesis. We performed a meta-analysis to synthesize the OR/HRs with 95%CI for each category ( e.g. , 12 months, any breastfeeding). Adjusted OR/HRs were used in the analysis instead of non-adjusted ones. Heterogeneity between studies was quantified using I 2 and Q tests[ 35 ]. If heterogeneity existed [ I 2 ≥ 50% and P value < 0.1 ( Q test)], random effects models were applied. Otherwise, fixed models were applied to perform synthesis. A subgroup analysis was performed according to the lactation duration. A funnel plot was used to check for publication bias if more than ten studies were included in the analysis.

A search of the databases identified 13238 references that were downloaded and imported into EndNote. After duplicates were removed, 9170 references remained. After screening titles and abstracts, 62 studies remained. After screening the full text of the remaining references, 18 were retained. A flowchart of the study selection process is presented in Figure 1 . PRISMA flow diagram for article selection for meta-analysis. A total of 18 studies were included in the qualitative analysis[ 22 , 27 , 29 , 36 - 50 ], with five studies detailing the association of breastfeeding with hypertension, 12 studies detailing the association of breastfeeding with diabetes, and one study detailing both associations. Five of the six studies detailing the association between breastfeeding and hypertension reported significant results favoring breastfeeding. These studies featured samples from various settings in four countries (Korea, Sudan, the United Kingdom, and the United States). One study reported results that did not favor breastfeeding, but they were not statistically significant. Twelve of the 13 studies detailing the association between breastfeeding and breastfeeding reported significant results favoring breastfeeding. These studies featured samples from various settings in six countries (Canada, China, Germany, Mexico, Singapore, and the United States). A summary of the characteristics of the included studies is presented in Table 1 . The characteristics of the 18 included studies, mean ± SD HR: Hazard ratio; OR: Odds ratio; BMI: Body mass index; DIE: Deep infiltrative endometriosis; GDM: Gestational diabetes mellitus; IVF: In vitro fertilization. Overall, the risk of bias (RoB) for the studies included in the review was rated from ‘unclear’ to ‘low’. All domains of ROBINS-E were also rated as ‘unclear’ to ‘low.’ Of the studies included in the meta-analysis, four were rated as having a ‘low’ RoB and seven were rated as having an ‘unclear’ RoB (see Figures 2 and 3 ). Risk of bias graph: The review authors' judgments about each risk of bias item are presented as percentages across all included studies. Risk of bias summary: Review authors' judgments about each risk of bias item for each included study. Three studies[ 22 , 42 , 45 ] described the HR for diabetes among parous women who previously had gestational DM (GDM). Those who breastfed for any duration were compared to those who did not breastfeed. Ziegler et al [ 45 ] had a sample size of 304 participants, Martens et al [ 42 ] had 180107 participants, and Gunderson et al [ 22 ] had 1035 participants. The I 2 test yielded a result of 71% with a Q test P = 0.03, suggesting moderate heterogeneity and necessitating the use of a random-effects model. The meta-analysis of HR was found to be 0.64 (95%CI: 0.47-0.89) (see forest plot in Figure 4A ). Forest plot. A: The hazard ratio (HR) of lactation for the risk of diabetes among parous women who previously had gestational diabetes mellitus; B: The HR of lactation for the risk of diabetes among parous women; C: The odds ratio of lactation for the risk of hypertension among parous women. IV: Inverse-variance method; SE: Standard error. Eight studies reported the HR of diabetes risk compared with those who had breastfed and those who did not. Randolph random effects were applied ( I 2 = 82% with Q test P < 0.001), and the meta-analysis of HR of the seven studies was 0.82 (95%CI: 0.78-0.87) (Figure 4B ). All subgroups significantly decreased the risk of diabetes occurrence, except for the breastfeeding duration of < 6 months [HR = 0.88 (95%CI: 0.78-1.01)]. Three studies described the OR for diabetes among women who have given birth[ 37 , 44 , 46 ]. The meta-analysis of the three studies’ OR was 0.83 (95%CI: 0.75-0.91), applying random effects ( I 2 = 63% with Q test P = 0.01). Six studies provided data on the OR for hypertension in parous women by comparing those who breastfed with those who did not. According to the meta-analysis, the OR of the six studies was 0.92 (95%CI: 0.88-0.96) (see forest plot in Figure 4C ). However, all subgroups showed a reduced incidence of hypertension without significant differences, except for the breastfeeding duration of 6-12 months [OR = 0.88 (95%CI: 0.86-0.91)]. The literature included in this systematic review did not include studies that investigated the HR for hypertension among women who have given birth. A funnel plot was used to assess publication bias (Figure 5 ). No apparent publication bias was observed in this analysis. Funnel plots for publication bias assessment. A: Funnel plot to analyze the odds ratios for hypertension risk among parous women; B: Funnel plot for the analysis of risk for diabetes hazard ratio among parous women. SE: Standard error; OR: Odds ratios.

The findings of this review agree with the existing evidence and recommendations. Breastfeeding has consistently been associated with a lower incidence of maternal hypertension and diabetes later in life. Although the evidence is limited, lactation can decrease the risk of diabetes among pregnant mothers with GDM. Lactation for more than 6 months seems to play a critical role in diabetes and hypertension prevention. When advising their patients, healthcare providers should continue to recommend breastfeeding for at least six months.

This review found that breastfeeding is likely to protect against diabetes among parous women who experienced GDM (HR = 0.64, 95%CI: 0.47-0.89). Protection against hypertension is also possible in parous women. However, only 6-12 months of breastfeeding was statistically significant in preventing the occurrence of hypertension. We found that breastfeeding can reduce both the incidence [OR = 0.83 (95%CI: 0.75-0.91)] and risk [HR = 0.82 (95%CI: 0.78-0.87)] of diabetes in parous women. However, the reduced risk for breastfeeding duration of < 6 months [HR = 0.88 (95%CI: 0.78-1.01)] was not statistically significant. We found that breastfeeding significantly reduced the risk of diabetes in women with GDM during pregnancy. This finding was consistent with a prospective cohort study by Nguyen et al [ 51 ], which indicated that breastfeeding for ≥ 3 months reduced the risk of diabetes and delayed the development of diabetes by a further 10 years compared with breastfeeding for < 3 months. However, a multicenter study reported that GDM can reduce the breastfeeding duration[ 51 ]. Therefore, as breastfeeding protects GDM mothers against diabetes later in life, encouraging them to breastfeed is vital for their health. According to our findings, breastfeeding and breastfeeding duration may reduce the risk of diabetes in parous women. This finding is consistent with evidence suggesting that breastfeeding protects against type 2 diabetes[ 52 , 53 ]. As breastfeeding can prevent the development of type 2 diabetes, it is vital to encourage new mothers to breastfeed. However, we found that the reduced risk of breastfeeding duration < 6 months was not statistically significant. Studies have shown that breastfeeding for six months or less is associated with a higher risk of developing type 2 diabetes in mothers compared to those who breastfeed for longer durations[ 22 , 39 ]. The mechanisms underlying this phenomenon refer to the fact that lactation for more than 6 months can improve glucose metabolism[ 54 ], reduce insulin resistance[ 55 ], and lower HbA1c levels[ 29 ]. On the other hand, longer than 12 months of breastfeeding has similar effects on decreasing the risk of diabetes. Our findings differ from those of a recently published systematic review, which did not detect a significant relationship between long-term lactation and the risk of type 2 diabetes[ 56 ]. However, those studies have potential confounding factors like pre-pregnancy BMI, lifestyle choices, and family history of diabetes. Despite these factors, clinical guidelines generally recommend breastfeeding for at least six months due to its numerous health benefits for both mother and child[ 57 ]. Previous studies have suggested that breastfeeding babies was associated with a lower prevalence of hypertension in their later life[ 58 - 60 ]. Perrine et al [ 32 ] detailed the effect of breastfeeding on hypertension in a descriptive analysis and found that breastfeeding protected against hypertension. Magnus et al [ 61 ] also found that breastfeeding has protective benefits for the mother’s blood pressure later in life. While the exact mechanisms are still being researched, lactation appears to play a significant role in improving long-term cardiovascular health. Some possible mechanisms include: (1) Hormonal influences: Lactation is associated with changes in hormones like oxytocin, prolactin, cortisol, estrogen, and progesterone, which can affect blood pressure[ 62 ]; (2) Cardiovascular benefits: Lactation may reverse some of the metabolic changes that occur during pregnancy, potentially reducing the risk of developing hypertension[ 30 , 62 ]; and (3) Improved lipid profiles: Lactation has been linked to a lower risk of hypercholesterolemia and other lipid-related issues, which can contribute to hypertension[ 55 , 63 , 64 ]. According to our findings, 6-12 months of breastfeeding likely protected against hypertension in parous women, whereas other analyses (comparing breastfeeding vs no breastfeeding, 1-6 months of breastfeeding vs no breastfeeding, and > 12 months of breastfeeding vs no breastfeeding) failed to yield significant results. Although our results regarding 6-12 months of breastfeeding align with those in the published literature, the analyses for 1-6 months did not align. One possible reason for this discrepancy may be that the limited data cannot detect significant results. More rigorous studies are needed to explore the hypothesis that different breastfeeding times have different impacts on maternal hypertension later in life. Breastfeeding is associated with a reduced risk of both diabetes and hypertension in mothers. This protective effect is likely due to several factors, including improved glucose metabolism, increased insulin sensitivity, and hormonal changes that occur during and after lactation[ 55 , 65 , 66 ]. According to our findings, lactation for longer than six months appears to play a significant role in the development of diabetes and hypertension. In practice, it is helpful to include relevant information for patient education. This review included studies with large sample sizes conducted in various settings. The risk of bias assessment conducted using ROBINS-E also showed that all studies had a low to moderate risk of bias. By examining the data provided by these studies using both qualitative analysis and meta-analysis, this review was able to address the research question comprehensively. This study had several limitations. As breastfeeding duration was quantified differently across many studies, a small number of studies were included for meta-analysis in each separate category. Owing to various factors, most analyses showed moderate to high heterogeneity. Besides, meta-analyses with a small number of studies (typically less than 5) are indeed prone to inflated effect sizes and instability in their results[ 67 ], for instance, the meta-analysis for the GDM subgroup analysis.

Diabetes mellitus (DM) and hypertension are metabolic disorders that have become increasingly prevalent in recent years[ 1 ]. The World Health Organization (WHO) estimates that for adults aged 18 years and older with diabetes increased from 7% in 1990 to 14% in 2022[ 2 ]. While hypertension is even more common than DM, with an estimate of 1.39 billion cases worldwide currently[ 3 ]. Diabetes is a debilitating disease that results in severe complications and requires lifelong management and support. Similarly, hypertension is one of the leading causes of premature death[ 4 - 6 ]. In postnatal periods, increased diabetes and hypertension can stem from a combination of factors, including gestational diabetes, pre-existing conditions, and lifestyle choices[ 7 , 8 ]. Fluctuations in blood glucose and hypertension significantly impact quality of life by increasing the risk of cardiovascular complications, nerve damage, and kidney problems[ 9 - 11 ]. Several signaling pathways, such as the renin-angiotensin-aldosterone system (RAAS)[ 12 - 14 ], oxidative stress[ 9 , 15 , 16 ], and inflammation[ 9 , 15 , 17 ], are implicated in the development and progression of these conditions. Every mother must decide whether to breastfeed their children. In 2012, the WHO recommended exclusive breastfeeding in the Comprehensive Implementation Plan on Maternal, Infant, and Young Child Nutrition at least 6 months, citing various long-term benefits for both child and mother[ 18 ]. Among these benefits is the reduced likelihood of the mother developing non-communicable diseases later in life[ 19 ]. Various studies have investigated the possible association between breastfeeding and maternal diabetes as well as breastfeeding and hypertension. Most of them indicated that breastfeeding is favorable for the prevention of DM and hypertension[ 20 - 29 ]. However, those studies investigated these relationships in various ways. Some studies characterized breastfeeding duration, some used breastfeeding intensity, and others used whether breastfeeding was performed or not. Diabetes and hypertension are also characterized differently; some studies have reported the incidence, while others have reported hazard ratios (HR) or odds ratios (OR). Some recent reviews on this topic are as follows: Zachou et al [ 30 ] discussed the effects of lactation on cardiovascular diseases. Pérez-Roncero et al [ 31 ] discussed the association between breastfeeding duration and carotid intima-media thickness in later life. Perrine et al [ 32 ] qualitatively assessed the association between breastfeeding and various chronic diseases. In these reviews, breastfeeding protected mothers against relevant chronic diseases, including diabetes and hypertension. While reviews have been written on the relationship between breastfeeding and non-communicable diseases[ 19 ], there is no systematic review to explore the relationship between breastfeeding and diabetes/hypertension by consolidating all available evidence on the topic. This review aimed to investigate the relationship between breastfeeding and maternal diabetes and hypertension.