The effects of KD on cardiac health have been widely investigated, but researches concerning the effects of KD on cardiac functions provided a few relatively controversial data. Studies generally suggest that KD intake benefited cardiac metabolic efficiency and acted as a cardioprotective antioxidant. Selvaraj et al. (44) reviewed current evidence surrounding the use of therapeutic ketosis including KD in heart failure (HF) and pointed out its potential benefit in HF, particularly in HF with reduced ejection fraction. Further, Balietti et al. (45) found that an 8-week supplementation of medium-chain triglycerides KD (MCT-KD) to late-adult rats partly restored age-related decrease of succinic dehydrogenase (SDH) activity and metabolically active mitochondria, which might offset senescent alterations leading to apoptosis-induced myocardial atrophy and failure. Another study with a similar conclusion indicated that a 19-week low carbohydrate KD following global ischemic injury significantly increased the numbers of mitochondria in cardiac muscles and the reperfusion recovery of coronary flow (46). As such, the two studies demonstrated that KD was cardio-protective in terms of regulating cardiac energy metabolism including mitochondrial capability. However, some studies suggested that KD might be just not harmful to cardiac functions. A study utilizing KD for at least 12 months on cardiac functions in intractable epilepsy patients suggested that the KD used appeared to have no negative impact on ventricular functions in epileptic children in the midterm (47). Similarly, a 6-month KD therapy didn’t affect electrocardiogram outcomes in the drug-resistant children with epilepsy (48). The subjects in these two studies are both epileptic children, which cannot represent all the patients who might use KD therapy. Thus, we can still stay optimistic about the effects of KD on cardiac functions.

Studies have also been conducted concerning the mechanism of how KD might affect cardiac health. Abnormal substrate metabolism is one of the major changes of insulin resistance and diabetic myocardium (49). Given this, changes in the regulation of myocardial ketone body metabolism appear to be a novel diagnostic biomarker of altered ketolytic capacity. Wentz et al. (50) utilized ketogenic nutritional mouse models (24 h of fasting and a very low carbohydrate ketogenic diet) to demonstrate that cardiac muscle engages a metabolic response that limits ketone body utilization. Specifically, the results revealed that unmetabolized substrate concentrations were higher within the hearts of ketogenic diet-fed mice. Furthermore, a recent study suggested that a KD or a high-fat diet could reverse the structural, metabolic and functional remodeling of non-stressed cMPC1-/- (cardiomyocyte-restricted deletion of subunit 1 of mitochondrial pyruvate carrier) mouse hearts (51). A KD of 3 weeks before transverse aortic constriction was already enough to rescue cMPC1-/- hearts from rapid decompensation and early mortality after pressure overload. Another study also indicated that a high-fat, low-carbohydrate KD could completely reverse progressively developed cardiac dilation and contractile dysfunction in mice with cardiac-specific deletion of Mpc2 (CS-MPC2-/-) (52). Accordingly, KD therapy might be promising in improving cardiac fat metabolism to prevent or reverse cardiac dysfunction and remodeling in MPC deficiency.

As mentioned above, KDs are generally cardioprotective, which might be attributable to the effects of KDs on cardiac metabolism, such as ketone body metabolism and energy metabolism including mitochondrial capability ( Figure 2 ). Despite the evidence supporting the cardioprotective effect of KDs, another study utilizing KD on cardiac remodeling in spontaneously hypertensive rats (SHRs) suggested that KD might deteriorate cardiac remodeling in the hypertensive heart and warranted fully evaluation of its reliability before clinical use (53). The different pathogenesis backgrounds of hypertension might account for the different results. More studies with larger samples, longer follow-up duration, and standardized basic health status can be conducted to further clarify the role of KDs in cardiac functions and other potential mechanisms.

Attention has also been paid to the effects of KD in hypertension. Most studies showed positive effects of KD in hypertension. Castellana et al. (54) suggested that very-low-calorie ketogenic diet (VLCKD) manifested improvements in hypertension, type 2 diabetes and dyslipidemia, apart from being a promising lifestyle intervention for overweight and obesity. Another study incorporating 377 patients across Italy drew a similar conclusion that VLCKD could significantly lower SBP in three months (55). Even a short-term 4-week KD with micronutrient supplementation could result in improved hypertension control and in a reduction for the usage of hypertension medications in patients with preoperative T2DM and hypertension (56). Increasing ketone bodies by nutritional interventions of ketone bodies or their precursors, such as 1,3-Butanediol, was also reported to attenuate hypertension (57). However, Guo et al. (58) revealed that subjecting spontaneously hypertensive rats (SHRs) to KD for 4 weeks aggravated hypertension, increased the expression of IL1-β and TNF-α, impaired endothelium-dependent relaxation and decreased CD31 and eNOS expression in mesenteric arteries. This finding is opposite to the previous results; thus, it remind us to be cautious in treating hypertension with KD, and perform more studies to explore the effects of KD on hypertension in human studies and animal models.

The Dietary Approaches to Stop Hypertension (DASH) diet is a classic dietary approach that has been endorsed for patients with elevated blood pressure (BP). Besides, the addition of exercise and weight loss to the DASH diet resulted in even larger BP reductions, greater improvements in vascular and reduced left ventricular mass for obese people with elevated BP (59). The main drawback of DASH might be the difficulty in long-term adherence to this diet. Considering the different components of DASH and KD, KD might become another choice for those people who love a high-fat diet, although we suggest cautious application since the antihypertensive efficacy and side effects of KD under the background of hypertension remain unclear. Interestingly, a review suggested that intermittent fasting could lower both systolic and diastolic blood pressure in human studies and animal studies, possibly through reducing oxidative stress, syncing with circadian rhythm, and inducing a ketogenic state (60). The less consumption of fats is assumed as the reason why intermittent fasting appears to be more beneficial than KD in treating hypertension. Thus, the type of fats consumed in KD therapy is crucial to be considered both in treating hypertension and evaluating its effects.

A study by Keogh et al. (61) has indicated that a very-low-carbohydrate, high-saturated-fat weight-loss diet did not impair FMD. How would the actual KD impact on vascular function and vascular blood flow?

In some studies, KD appears to play a protective role in vascular functions. Ischemic tolerance can reduce brain injury and neurological dysfunction after brain ischemia. Additional to the cardiovascular effects such as higher reperfusion recovery of coronary flow, KD also can enhance brain vascular function. As supported by the results from a study upon feeding a KD to young healthy mice, KD intervention enhanced neurovascular function through reducing mTOR protein expression and increasing eNOS levels (62). Yang et al. (63) discovered that feeding mice with KD-fed mice could remarkably decrease infarct volume and elevate regional cerebral blood flow in both ischemic and reperfusion phases. Besides, while investigating the effects of KB level on HMEC-1 endothelial cells, one study indicated that KB activated transcription factor Nrf2 and elevated the expression of cell antioxidant defending genes via inducing moderate oxidative stress (64). Thus, the increased KB level by KD might also lead to these protective effects.

However, as for big vessels such as carotid and aortic artery, the effects of KD remain controversial. For instance, after observing the effect of KD on the vascular structure and functions for at least one year, it was found that KD notably elevated the serum levels of lipids but didn’t significantly affect carotid intima-media thickness, aortic and carotid strain, the stiffness index, distensibility, and elastic modulus (65). Another study by Doksoz et al. (66) also demonstrated that a 6-month KD didn’t affect carotid intima-media thickness and elastic properties of the carotid artery and the aorta. In contrast, in the research of Coppola et al. (67), participants prescribed with KD had higher arterial stiffness parameters, including AIx and beta-index and higher serum levels of cholesterol or triglycerides. Another study revealed that a high-fat KD notably elevated atherogenic apolipoprotein B (apoB)-containing lipoproteins and decreased antiatherogenic HDL cholesterol and urged further researches to investigate whether this diet deteriorates endothelial function and facilitates inflammation and formation of atherosclerotic lesions (68). However, a clinical study involving 26 children after one year and 13 children after two years of KD suggested that the initial influences on arterial function observed within the first year of KD-treatment were reversible and were no longer significant after 2 years of the therapy (69). Therefore, the effects of KD on big vessels such as carotid and aortic artery were reversible and were no longer significant after 1-2 years, which might explain the above results.

Apart from the cardiometabolic effects mentioned above, the impact of KD on serum CVD biomarkers has also been investigated. Research on 20 normal-weight, normolipidemic men indicated that a 6-week KD notably decreased fasting serum triglyceride, postprandial lipemia, and fasting serum insulin concentrations, tended to increase HDL cholesterol, while not affecting fasting serum total and LDL cholesterol and oxidized LDL (70). These results revealed that short-term KD would not deteriorate CVD risk profile and, indeed, appeared to ameliorate lipid disorders that are characteristics of atherogenic dyslipidemia. Another research also indicated that changes in the ratio of protein to carbohydrate toward higher protein proportion could provide beneficial effects on serum lipids apart from lowering body weight (71).

However, Özdemir et al. (65) pointed out that prescribing patients with at least 12 months KD could significantly elevate serum total and LDL cholesterol and triglyceride at a median of 12.6 months while not affecting HDL level. Moreover, another research (72) found that a 6-month KD could notably increase median triglyceride, total cholesterol, LDL, and HDL. They suggested that classic KD was indeed efficient in treating refractory seizures in children but might give rise to hypercholesterolemia and hypertriglyceridemia.

As such, disputable voices concerning the impact of KDs on serum lipids remain to be settled by future work in this field.

Obesity is closely related to CVD, and complications of CVD are often witnessed in obese patients. Cicero et al. (55) evaluated the effect of a very low carbohydrate ketogenic diet (VLCKD) on overweight-related risk factors of CVD such as blood pressure, lipid levels, and glucose metabolism, and the study found that VLCKD intervention for 3 months was generally safe and found effective in inducing weight loss and improved CV risk factors levels.

Another study recruited a hundred obese patients and prescribed them a ketogenic diet for over six months showed significant improvement in patients’ cardiovascular status in addition to weight reduction (73). Moreover, a meta-analysis study by Bueno et al. (74) assessed the long-term effects of VLCKD on body weight and cardiovascular risk factors. The results indicated that under VLCKD, the participants had a significant reduction in body weight, TAG, and diastolic blood pressure, while increased HDL-C and LDL-C levels were observed. Apart from long-term studies, a study by Ministrini et al. (75) that treated obese patients with VLCKD for 25 days also concluded that VLCKD had positive effects on cardiovascular risk factors, and such a beneficial outcome in the short term is remarkable.

Other studies investigated the effects of modified KDs on cardiovascular risks in obese participants. Perez-Guisado et al. (76) carried out a prospective evaluation in 31 obese participants with “Spanish Ketogenic Mediterranean Diet” (incorporating virgin olive oil as a principal source of fat, moderate red wine intake, green vegetables and salads as the primary source of carbohydrates and fish as the main source of protein, SKMD). The SKMD was found safe and effective interventional approach for improving non-atherogenic lipid profiles and lowering blood pressure while lowering body weight. Similarly, Paoli et al. (77) applied another modified KD that incorporated phytoextracts and ingredients imitating the taste of carbohydrates (ketogenic Mediterranean with phytoextracts, KEMEPHY). The study recruited 106 overweight Rome council employees and revealed a remarkable reduction in body weight, BMI, percentage of fat mass, total cholesterol, LDL-C, TAG and blood glucose while displaying a significant increase in HDL-C after the intervention with KEMEPHY. In addition to good compliance, extra beneficial effects on cardiovascular risk markers and waist circumference were also achieved by the KEMEPHY diet.

Since the beneficial effects of KDs on metabolism and cardiovascular risk factors are similar to those seen after n-3 polyunsaturated fatty acids (omega-3) supplementation, Paoli’s team (78) modified the ketogenic Mediterranean diet with phytoextracts after their previous research through combining with omega-3 supplementation. The results suggested that this newly modified diet can further enhance the beneficial effects on cardiovascular risk factors and inflammation in overweight participants.

The influence of a multi-step dietary program including different dietary patterns has also been evaluated. In an open-label study by Castaldo et al. (79), 73 obese patients entered a rehabilitative multi-step dietary program: a 3-week protein-sparing, very low-calorie KD (<500 kcal/day; Oloproteic Diet) and a 6-week hypocaloric (25–30 kcal/kg of ideal body weight/day), low glycemic index, Mediterranean-like diet (hypo-MD). In both phases, improved glucose and lipid metabolism and blood pressure were observed. Based on this, it was concluded that a dietary program consisting of a KD and a subsequent MD could decrease cardiovascular risks efficaciously in obese patients.

Diabetes mellitus is often associated with obesity (80), and recent estimates showed that 87.5% of T2DM patients are overweight or obese (24). Moreover, obese subjects are prone to developing hypertension, CVD, and strokes, and the risk is even higher if it co-exists with T2DM (81).

In patients with both T2DM and obesity, LC diets not only cause weight loss but also improve postprandial plasma glucose levels, glucose variability, serum triglycerides, and HDL-C levels (82). Similar results were observed by a 2-year randomized clinical trial study (39) that investigated the effect of an LC diet with high unsaturated fat and low saturated fat on glycemic control and CVD risk factors in overweight or obese patients with T2DM. KD is a low-carbohydrate (LC) and high-fat (HF) diet, which sort of belongs to one type of LC diets. However, one of the potential concerns of KDs is postprandial hyperlipidemia, which leads to significant cardiovascular risks (83).

Studies have found that individuals with pre-diabetes or diabetes who received an earlier LCHF diet revealed several beneficial outcomes, including weight loss, improved insulin sensitivity, glucose homeostasis, and lower fasting blood glucose levels. These outcome improvements also decreased the risks of cardiovascular diseases development (38, 84). Mobbs et al. (28) analyzed the evidence concerning the treatment of diabetes and diabetic complications with a KD. They revealed that a classic KD significantly reduced blood glucose in animal models of type 1 and 2 diabetes and reversed diabetic nephropathy without producing significant cardiovascular risks. Moreover, a study on db/db mice revealed that KD ameliorates cardiac dysfunction by inhibiting apoptosis via activating the PI3K-Akt pathway in type 2 diabetic mice and suggested KD as a promising lifestyle intervention against diabetic cardiomyopathy (85).

However, there also are studies with conflicting or controversial findings and opinions. Westman et al. (86) stated that LCHF diets gave rise to decreased appetite, thus the improved surrogate markers of cardiovascular disease resulted from weight loss but not from low carbohydrate intake itself. In an animal study, Abdurrachim et al. (87) investigated the effects of long-term KD on cardiac metabolism and diabetic cardiomyopathy status in lean diabetic Goto-Kakizaki (GK) rats. Upon KDs for 62 weeks, diabetic GK rats displayed decreased blood glucose, triglyceride, and insulin levels, revealing increased blood ketone body levels. Additionally, KDs decreased myocardial ketone body and glucose oxidation and induced cardiac hypertrophy. These results suggested that KDs might lead to maladaptive cardiac metabolic modulation and lipotoxicity and deteriorate diabetic cardiomyopathy in GK rats. Given this, the possible role of KDs in cardiovascular risks of DM remains controversial in rodent models and humans, which warrants more studies for elucidation.

As for patients with dyslipidemia, Westman et al. (88) investigated the effect of KDs on serum lipoprotein subclasses to address the concern of KDs on cardiovascular risks. The study was a randomized, two-arm clinical trial involving overweight and hyperlipidemic participants motivated to lose weight. After 6 months, the KD group displayed more significant decreases in medium VLDL, small VLDL, and medium LDL, and more significant increases in VLDL particle size, large LDL, and HDL particle size than the control group. Although the KDs did not decrease total LDL cholesterol, they shifted from small, dense LDL to large, buoyant LDL, thus decreasing CVD risks in these participants.

NAFLD contributes to CVD through various mechanisms. Weight loss has been commonly recommended for treating obesity-associated NAFLD; meanwhile, LCKD benefits weight loss. Recent studies have revealed an association between LC diets and NAFLD in both rodents and humans. In the study by Garbow et al. (28), mice fed a KD for 12 weeks were lean, euglycemic, ketotic, and hypo-insulinemic but were glucose intolerant and with NAFLD. Also, obese subjects on LC diets displayed enhanced weight loss, improved metabolic parameters and decreased intrahepatic triglyceride content. Nevertheless, long-term KDs led to NAFLD and systemic glucose intolerance in mice (89), negatively impacting CVD. As such, current evidence is insufficient to conclude, and more related studies are warranted to explore how KDs might influence NAFLD-related CVD in the long run.

GLUT1DS is an inherited but treatable disease concerning cerebral energy metabolism (90). KDs are currently a treatment option for GLUT1DS from infancy into adulthood, raising concerns about long-term cardiovascular risks (43, 60, 69). To address this problem, Heussinger et al. (91) performed a 10-year follow-up study on cardiovascular risk of KDs in GLUT1DS and revealed that dyslipidemia caused by KDs might be transient; and carotid intimal wall thickness (CIMT), BMI and blood pressure parameters remained normal after 10 years. Because of this, Heussinger et al. suggested that cardiovascular risks of KDs in some previous studies appeared to be attributable to inadequate follow-up. Also, a period of at least five years appears to be necessary for evaluating the effect of KDs on lipid parameters. Moreover, the authors recommended KDs as a treatment of choice for GLUT1DS. Another study by Alter et al. (90) also characterized the long-term course of GLUT1DS and followed up for an average of 14.2 years (range = 8.9-23.6). The results indicated that earlier introduction of KDs correlated with better long-term outcomes and KDs seemed to be protective of vital organs. However, GLUT1DS is a rare disease, and therefore, the study cohort’s size and external validity are limited. Long-term follow-up studies are warranted to confirm the above findings further.

One study found KD treatment ameliorates cardiac dysfunction by inhibiting apoptosis via activating the PI3K-Akt pathway in type 2 diabetic mice, suggesting that the KD is a promising lifestyle intervention offering protection against diabetic cardiomyopathy (85). In contrast, a ketogenic diet may lead to adverse effects on the remodeling in the hypertensive heart via mechanisms involving increased mTOR signaling, and they underscore the necessity to evaluate its reliability before clinical use (53). Preclinical studies results indicate that KD also has a potential safety concern, although much evidence suggests that KD is a promising approach for managing CVD.

As a key factor that triggers or exacerbates CVD risk, liver inflammation is a potential safety concern related to KD. In support of this, a study observed that mice fed with KD sustained unimpaired insulin-induced hepatic Akt phosphorylation and whole-body insulin responsiveness but ultimately developed hepatic endoplasmic reticulum stress, steatosis, cellular injury, and macrophage accumulation (28).

KD is enriched in lipid contents, and it’s natural to speculate the potential risk of elevated levels of lipids. Apart from studies indicating the beneficial effects of KD, concerns regarding the elevated level of lipids, including serum total and LDL cholesterol and triglyceride, are subjective while prescribing KD (65, 72). It is reported that KDs are likely to deteriorate levels of total, high-density lipoprotein (HDL) and low-density lipoprotein (LDL) cholesterol, and triglycerides (27, 34, 92) in rodents while doing the opposite in humans (39). These contradictory results might be attributable to the different composition of diets since animal researches generally employ diets higher not only in total fat but also in saturated fat (93). Considering this, it is necessary to compare the fat composition, e.g., content of saturated fat versus unsaturated fat in KDs in long-term studies involving both rodents and humans.

As described above, while considering KD as an exciting approach for managing CMDs, it also is important to be cautious about the potential safety concern associated with KD. While future studies are warranted to confirm and elucidate whether and how KD causes potential safety concerns, it would also be important to consider to modifying KD or combining KD with other healthy diets for managing CMDs.

Because of the irreconcilable options on the therapeutic use of KD, several studies concerning modified KD and cardiovascular risks have been performed. The “Spanish Ketogenic Mediterranean Diet” carried out by Perez-Guisado et al. (76) and two modified KDs (KEMEPHY (77) and KEMEPHY with omega-3 supplementation (78)) employed by Paoli et al. have all displayed beneficial effects on cardiovascular risk factors. A combined diet consisting of KD and a subsequent Mediterranean-like diet has been proven to decrease cardiovascular risks in patients (79). Therefore, a modified KD or multi-step dietary program including different diet patterns is promising in resolving the safety concerns associated with KDs.