Citrus juices

Citrus fruits and their juices are rich in micronutrients such as vitamin C, folate, potassium and magnesium, as well as in bioactive (poly)phenols^{(Reference Visvanathan and Williamson34,Reference Miles and Calder35,Reference Martí, Mena and Cánovas36)} . Citrus are the main dietary source of flavanones, a flavonoid subclass. Flavanones in citrus fruit occur primarily in a glycosylated form, and the dominant flavanone glycosides are hesperidin in sweet orange and tangerine, naringin and neohesperidin in sour orange, naringin in grapefruit, and hesperidin and eriocitrin in lemon and lime^{(Reference Peterson, Dwyer and Beecher37,Reference Peterson, Beecher and Bhagwat38)} . Citrus juices have been associated with different protective features, as they may have a role in reducing the risk of T2D^{(Reference Visvanathan and Williamson34)} and in the inflammatory response^{(Reference Miles and Calder35)}. They are among the most studied FVJ as they also represent the major group of FVJ consumed by Western populations, orange juice being the most consumed 100% juice worldwide.

Orange juice

Most of the literature on the health effects of juices is related to orange juice, and the number of targets addressed is quite broad. Some representative studies have been presented in Table 1.

Table 1. Characteristics of some representative studies investigating the health effects of 100% orange juice

Juices were 100% juice unless otherwise stated. Abbreviations: 11-dehydro-TXB2, 11-Dehydrothromboxane B2; Apo A, apolipoprotein A; Apo B, apolipoprotein B; BDNF, brain-derived neurotrophic factor; BMI, body mass index; BP, blood pressure; BW, body weight; C_max, maximal glucose concentration; CDH, hesperidin control drink; CDP, placebo control drink; CO, crossover; CVR, cardiovascular risk; DB, double-blind; DBP, diastolic blood pressure; EOJ, enriched orange juice; F, female; FMD, flow-mediated dilation; GAE, gallic acid equivalent; HDL-c, HDL-cholesterol; HOMA-IR, homeostasis model assessment-insulin resistance; HPJ, high concentrations of (poly)phenols; hs-CRP, high-sensitivity CRP; IL-6, interleukin-6; LDL-c, LDL-cholesterol; M, male; MDA, malondialdehyde; NO, nitric oxide; NPJ, normal concentrations of (poly)phenols; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OJ, orange juice; OL, open label; OPF, orange pomace fibre; OW, overweight (BMI: 25–30 kg/m²); PAC, A-type proanthocyanidins; PP, pulse blood pressure; RCT, randomised controlled trial; RT, randomised trial; SB, single-blind; SBP, systolic blood pressure; SCFAs, short-chain fatty acids; SOD, superoxide dismutase; T_max, time to reach the maximal glucose concentration; TAG, triglycerides; TC, total cholesterol; TNF-α, tumour necrosis factor-α; WC, waist circumference; WOF, whole orange fruit; ↓, decreased level; ↑ increased level

Regarding body weight, Rangel-Huerta et al. (2015)^{(Reference Rangel-Huerta, Aguilera and Martin39)}, in a randomised crossover double-blind 12-week study, investigated the effects of consuming 500 ml/d of orange juice containing either normal (589 mg/l) or high (1490 mg/l) concentrations of flavanones (normal-polyphenol juice (NPJ) and high-polyphenol juice (HPJ), respectively) in 100 non-smoking adults who were overweight or obese. In both juices, hesperidin was the main flavanone, accounting for 89% and 78% of the polyphenol content for NPJ and HPJ, respectively. Potassium (460 and 1065 mg/500 ml, in NPJ and HPJ, respectively) and vitamin C (210 and 235 mg/500 ml, in NPJ and HPJ, respectively) were the main minerals and vitamins. Results showed a reduction in body weight, BMI and waist circumference following consumption of both juices. These reductions could be explained by the decrease in energy intake during the study, but not by the flavanone content of each treatment, which did not entail differences in the urinary excretion of flavanone metabolites between treatments^{(Reference Rangel-Huerta, Aguilera and Martin39)}. A similar experimental protocol (12-week intervention, 500 ml/d, 100% orange juice, 324 mg/l hesperitin), where orange juice was supplemented as part of a reduced-calorie diet, did not inhibit weight loss in seventy-eight patients with obesity, as compared with the control group (a reduced-calorie diet without orange juice)^{(Reference Ribeiro, Dourado and Cesar40)}. Another 12-week study showed no effect of 250 ml/d of orange juice (542 mg/l hesperidin) on body weight in men who were overweight^{(Reference Simpson, Mendis and Macdonald41)}. Together, these studies showed no adverse effect of orange juice on body weight management, in agreement with two recent meta-analyses of RCT on the effect of orange juice on anthropometric measures in healthy or unhealthy/at-risk adults^{(Reference Djafari, Shahavandi and Amini42,Reference Motallaei, Ramezani-Jolfaie and Mohammadi43)} .

Considering vascular health, daily consumption of flavanone-rich orange juices may be able to reduce BP, while red orange juice might affect endothelial function (contrary to ‘blonde’ or common orange, red or ‘blood’ orange contains anthocyanins). Consumption of an orange juice with normal levels of flavanones (NPJ, 589 mg/l) for 12 weeks reduced both systolic and diastolic BP in adults who were overweight/obese, while the decrease after consuming a juice with higher flavanone content (HPJ, 1490 mg/l) was not significant^{(Reference Rangel-Huerta, Aguilera and Martin39)}. Contrarily, an effect of the flavanone content on BP has been described in a 12-week randomised parallel double-blind placebo-controlled trial evaluating the effects of hesperidin in orange juice on BP in 129 mildly hypertensive individuals^{(Reference Valls, Pedret and Calderón-Pérez44)}. Consumption of 500 ml/d of orange juice containing 690 mg/l of hesperidin, or enriched orange juice containing 1200 mg/l of hesperidin, decreased both systolic BP and pulse pressure in a dose-dependent manner with the hesperidin content of the beverage administrated and in comparison to a control drink. Furthermore, a single dose of enriched orange juice (500 ml) but no other treatments reduced systolic BP. These functional changes were backed by changes in the plasma concentrations of hesperetin metabolites^{(Reference Valls, Pedret and Calderón-Pérez44)}. Similarly, Morand et al. (2011)^{(Reference Morand, Dubray and Milenkovic45)} showed that lower levels of diastolic BP were reached after consumption of both orange juice, naturally rich in hesperidin, and an hesperidin-enriched control drink. Nevertheless, other authors have reported a lack of effect of orange juice on BP. Hollands et al. (2018) compared 4-week consumption of 500 ml/d of ‘blood’ orange juice (134 mg/l of hesperidin + 100 mg/l of anthocyanins) or ‘blonde’ orange juice (208 mg/l of hesperidin, without anthocyanins) in 41 predominantly individuals who were overweight and found no effect of any juice on BP. In a 2-week study, consumption of 400 ml/d of ‘blood’ orange juice (802 mg/l of hesperidin + 24 mg/l of anthocyanins) did not change BP in 15 non-smoking adults who were overweight or obese and who had baseline pressure within a healthy range^{(Reference Li, Lyall and Alberto Martinez-Blazquez47)}. The low amounts of flavanones provided^{(Reference Hollands, Armah and Doleman46)}, the intervention duration, and the baseline values in the study population^{(Reference Li, Lyall and Alberto Martinez-Blazquez47)} might be behind the lack of effect reported. The inter-individual variability in the metabolism of flavanones may also condition the effect of orange juice consumption on systolic BP^{(Reference Fraga, Coutinho and Rozenbaum48)}. On the other hand, in the case of endothelial function,^{(Reference Li, Lyall and Alberto Martinez-Blazquez47)} a favourable effect (2% increase in flow-mediated dilation (FMD)) of blood orange juice (400 ml/d for 2 weeks) was demonstrated compared with a control drink. The result correlated with a higher urinary excretion of hesperetin metabolites^{(Reference Li, Lyall and Alberto Martinez-Blazquez47)}. A similar improvement in FMD was reported following 7-d consumption of red orange juice (500 ml/d, 319 mg/l of hesperidin and 71 mg/l of anthocyanins) in 19 non-diabetic subjects with increased cardiovascular risk, while there was no effect in healthy subjects^{(Reference Buscemi, Rosafio and Arcoleo49)}. The result of the ongoing HESPER-HEALTH study, which evaluates the role of orange juice in vascular health^{(Reference Verny, Milenkovic and Macian50)}, will help to better establish the effect of orange juice intake on FMD.

Regarding lipid profile, a recent SRMA of fifteen RCT in healthy or unhealthy at-risk adults investigated the effectiveness of orange juice intake on primary cardiometabolic markers, including lipid profile^{(Reference Motallaei, Ramezani-Jolfaie and Mohammadi43)}. The results suggested that orange juice intake might be associated with reduced serum total cholesterol (TC) concentration. But, despite this extensive evidence, several works have failed to find an effect of orange juice consumption on lipid profile. Hollands et al. (2018)^{(Reference Hollands, Armah and Doleman46)}, comparing 4-week consumption of 500 ml/d of blood orange juice (containing anthocyanins) or blonde orange juice (without anthocyanins), found no effect of the orange juice pre- and post-supplementation on the lipid profile of forty-one predominantly overweight individuals. In particular, the hypothesised improvement in LDL-C with blood orange juice (because of its anthocyanin content) did not occur, nor was there any change in TC, HDL-cholesterol (HDL-C) and triacylglycerols (TG). As suggested by the authors, the study population was only minimally hyperlipidaemic (5·1 mmol/l TC), and an effect in individuals with higher TC (for example, >6·0 mmol/l) cannot be precluded. However, 12-week consumption of 250 ml/d of orange juice did not affect circulating lipids in thirty-six men with overweight and elevated fasting serum TC (5–7 mmol/l)^{(Reference Simpson, Mendis and Macdonald41)}. Nevertheless, in the orange juice group, those with the highest TG concentration pre-intervention showed the greatest reduction after 12 weeks of supplementation^{(Reference Simpson, Mendis and Macdonald41)}. Li et al. (2020)^{(Reference Li, Lyall and Alberto Martinez-Blazquez47)} also reported that the lipid profile (TC, LDL-C, HDL-C, and TGs) was unaffected by blood orange juice (400 ml/d for 2 weeks) in subjects with normal lipid levels. The same conclusions were found following either consumption of NPJ or HPJ (normal or high-polyphenol concentration orange juice, respectively), except for TG and apoB (the major component of LDL-C), which decreased significantly only after the intake of NPJ when compared with baseline^{(Reference Rangel-Huerta, Aguilera and Martin39)}. Differently, when orange juice was consumed concomitantly to a reduced-energy diet for 12 weeks, it helped to reduce TC and LDL-C in individuals with obesity and a normal lipid profile compared with the control group^{(Reference Ribeiro, Dourado and Cesar40)}. Therefore, orange juice consumption may impact the lipid profile but only for specific target populations, particularly in subjects with high TG concentrations.

The effect of orange juice supplementation on glucose–insulin homeostasis and the prevention of T2D onset has been broadly studied. Dietary supplementation of 500 ml/d of blood or blonde orange juice for 4 weeks did not change glycaemia in subjects with overweight^{(Reference Hollands, Armah and Doleman46)}. Similarly, orange juice, as part of a reduced-calorie diet, did not increase glucose levels over time in patients with obesity, whereas it improved insulin sensitivity^{(Reference Ribeiro, Dourado and Cesar40)}. The decrease in insulin levels and the reduction of homeostasis model assessment-insulin resistance (HOMA-IR) were noted only after 8 weeks of intervention^{(Reference Ribeiro, Dourado and Cesar40)}. However, Simpson et al. (2016)^{(Reference Simpson, Mendis and Macdonald41)} and Rangel-Huerta et al. (2015)^{(Reference Rangel-Huerta, Aguilera and Martin39)} found no change in HOMA-IR index after a 12-week orange juice intervention in individuals who were overweight and with elevated fasting serum cholesterol and individuals who were overweight/obese, respectively, despite the amounts of orange juice administered were different (250 ml versus 500 ml per day). On the other hand, orange juice consumption may have positive effects on reducing uric acid levels in healthy adults^{(Reference Morand, Dubray and Milenkovic45)}.

The SRMA of RCT by Motallaei et al. (2021)^{(Reference Motallaei, Ramezani-Jolfaie and Mohammadi43)} suggested that orange juice intake might be associated with improved insulin sensitivity, although the quality of the evidence was low-to-moderate and further well-designed studies are needed to confirm this finding. Indeed, the contradictory nature of the available evidence has been highlighted and investigated by Prof. Williamson’s group^{(Reference Visvanathan and Williamson34)}. In a recent review of epidemiological and intervention studies on the effect of orange juice consumption on risk of developing T2D, it was stated that orange juice might improve fasting glucose, fasting insulin and insulin sensitivity after 4 to 12 weeks of orange juice consumption. Inter-individual variability in the metabolism of flavanones seems to be one of the main drivers affecting the physiological responses to chronic consumption in humans^{(Reference Visvanathan and Williamson34)}. This review also concluded that the acute effect of orange juice consumption on post-prandial glycaemic response is relatively small^{(Reference Visvanathan and Williamson34)}, as it may depend on the hesperidin and sugar content of the juice^{(Reference Kerimi, Gauer and Crabbe51)}. Fibre content may also play a role in the post-prandial glycaemic response. Dong et al. (2016)^{(Reference Dong, Rendeiro and Kristek52)} found that consuming 240 ml of orange juice added with pomace fibre (5·5 g) significantly reduced the maximal change in glucose concentrations reached after meal ingestion in men with increased cardiometabolic risk. Similar results have been recently shown by Guzman et al. (2021)^{(Reference Guzman, Xiao and Liska53)}.

Motallaei et al. (2021)^{(Reference Motallaei, Ramezani-Jolfaie and Mohammadi43)} also investigated how orange juice intake might be associated with other cardiometabolic markers such as inflammatory markers (C-reactive protein (CRP), interleukin-6 (IL-6), and vascular cell adhesion molecule-1 (VCAM-1)), although no significant associations were found. However, two recent meta-analyses have accounted for the beneficial effect of 100% orange juice and hesperidin in orange juice on inflammation^{(Reference Tadros and Andrade54,Reference Cara, Beauchesne and Wallace55)} . In this sense, some works indicated positive effects of orange juice consumption on inflammatory markers^{(Reference Ribeiro, Dourado and Cesar40)} and genes^{(Reference Milenkovic, Deval and Dubray56)}. On the other hand, markers related to oxidative stress have also been considered. Both NPJ and HPJ protected against DNA damage and lipid peroxidation and modified several antioxidant enzymes in non-smoking subjects who were overweight/obese ^{(Reference Rangel-Huerta, Aguilera and Martin39)}. In contrast, no change in plasma protein carbonyl concentrations, measured as a biomarker of oxidative stress, was observed following red orange juice intake in subjects with increased cardiovascular risk^{(Reference Buscemi, Rosafio and Arcoleo49)}. Similarly, consumption of 600 ml/d of blood orange juice for 21 d did not modify plasma antioxidant status and lipid peroxidation, but it improved lymphocyte DNA resistance to oxidative stress in healthy women^{(Reference Riso, Visioli and Gardana57)}. The 600 ml portion provided approximately 450 mg of vitamin C, 21 mg of cyanidin-3-glucoside, 0·4 mg of β-cryptoxanthin, 0·12 mg of lutein, 0·11 mg of zeaxanthin and 0·1 mg of lycopene, and higher plasma concentrations of these components were recorded after juice intake^{(Reference Riso, Visioli and Gardana57)}. The recent meta-analysis by Cara et al. (2022)^{(Reference Cara, Beauchesne and Wallace55)} also suggested a positive effect of 100% orange juice on malondialdehyde (MDA) levels, although the results were not statistically significant.

Other health outcomes such as cognitive function have also been investigated. Kean et al. (2015)^{(Reference Kean, Lamport and Dodd58)}, in a randomised double-blind placebo-controlled trial in thirty-seven healthy older adults (66·7; SD 5·3 years), showed a significantly better global cognitive function after 8 weeks of 500 ml/d of high-flavanone 100% orange juice (610 mg/l flavanones, 90% hesperidin) compared with an isoenergetic low-flavanone control drink. In another study, consumption of 240 ml of orange juice with added orange pomace fibre containing 917 mg/l hesperidin and 5·5 g fibre was associated with acute improvements in cognitive function and subjective alertness up to 6 h post-consumption in healthy middle-aged males (51; SD 6·6 years), relative to an energy-matched placebo drink^{(Reference Alharbi, Lamport and Dodd59)}. In young adults (18–30 years), 500 ml/d of citrus juice (mainly orange including grapefruit juice, containing 140 mg/l flavonoids, of which 84 mg/l was hesperidin) enhanced cerebral blood flow. However, the results did not show a clear association between increased cerebral blood flow and behavioural benefits^{(Reference Lamport, Pal and Macready60)}. Moreover, a recent review on the impact of citrus (poly)phenols, mainly from orange juice, on brain functions accounted for positive effects on reduction of depression risk and cognitive ability linked to schizophrenia^{(Reference Głąbska, Guzek and Groele31)}.

Orange juice could play a role in gut microbiota modulation. In a randomised crossover 7-d study, Brasili et al. (2019)^{(Reference Brasili, Hassimotto and Del Chierico61)} found that 500 ml/d consumption of two different orange juices belonging to ‘Bahia’ or ‘Cara Cara’ cultivars, significantly changed the gut microbiota composition in twenty-one healthy individuals. Interestingly, the authors observed that the two juices, characterised by different vitamin C content, flavanones and carbohydrates, affected the modulation of the microbiota profile differently. Changes in metabolome were also cultivar-specific^{(Reference Brasili, Hassimotto and Del Chierico61)}. In another study carried out in ten healthy young women that consumed 300 ml of commercial pasteurised orange juice for 2 months, orange juice positively modulated the composition and metabolic activity of gut microbiota, increasing the population of Bifidobacterium spp. and Lactobacillus spp.^{(Reference Delgado, Cecatti and Priscila62)}. Changes in microbiota composition following flavonoid-rich orange juice (600 mg flavonoids/380 ml daily dose, for 8 weeks) were also linked to a potential improvement in depression status in young adults^{(Reference Park, Choi and Lee63)}.

In conclusion, orange juice shows no adverse effect on body weight and other anthropometric markers, as supported by recent meta-analyses. It could lower systolic BP and improve endothelial function, notably due to the hesperidin content. It did not adversely affect the lipid profile and may help reduce high plasma TG concentrations. Similarly, it did not modify glycaemia and insulin sensitivity, neither could it improve the latter. The role of orange juice in oxidative stress and inflammation is not clear. Last, its effect on cognitive function and microbiota modulation showed interesting prospects, although these observations are still preliminary. Further studies should investigate the potential role of hesperidin and other bioactives in orange juices. Indeed, more attention should be paid on characterisation of the carotenoid and fibre content in orange juices. For instance, it has been reported that the bioavailability of β-cryptoxanthin is greater from pasteurised orange juice than from fresh oranges, while higher fibre amounts may limit carotenoid bioavailability^{(Reference Aschoff, Rolke and Breusing64)}. To draw more robust conclusions, a detailed phytochemical profiling is needed, and future research must consider the numerous confounding factors that may condition the physiological response, from the orange juice characteristics to the drivers of the inter-individual variability both in the metabolism and the response (responders/non-responders) to phytochemicals^{(Reference Visvanathan and Williamson34,Reference Tadros and Andrade54)} .

Grapefruit, mandarin and lemon juice

In comparison to orange juice, only a few studies have addressed the impact of grapefruit, mandarin and lemon juices on human subject health (Table 2). Some works have investigated the potential interaction of grapefruit juice with medication pharmacokinetics, particularly calcium channel blockers as antihypertensive therapy, but these were not reported as they go beyond the aim of this review.

Table 2. Characteristics of some representative studies investigating the health effects of grapefruit, mandarin and lemon juices

Juices were 100% juice unless otherwise stated. Abbreviations: BMI, body mass index; BP, blood pressure; BW, body weight; CO, crossover; DB, double-blind; FG, fasting blood glucose; FMD, flow-mediated dilation; FRAP, ferric reducing ability of plasma; GGT, &x01B4;-glutamyl transpeptidase; GSH, glutathione; HOMA-IR, homeostasis model assessment-insulin resistance; hs-CRP, high-sensitivity CRP; ICAM-1, intercellular adhesion molecule 1; IL-6, interleukin-6; MDA, malondialdehyde; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OL, open label; OW, overweight (BMI: 25–30 kg/m²); PWV, pulse wave velocity; RCT, randomised controlled trial; RT, randomised trial; SB, single-blind; vWF, von Willebrand factor; ↓, decreased level; ↑ increased level

Habauzit et al. (2015)^{(Reference Habauzit, Verny and Milenkovic65)}, in a 6-month randomised controlled crossover double-blind trial in forty-eight healthy post-menopausal women, found that 340 ml/d of either blonde grapefruit juice (626 mg/l of naringenin glycosides) or isoenergetic control drink matched for the macro- and micronutrients of the juice, but without naringenin glycosides, did not affect body weight and other anthropometric measures. Considering vascular function, no effect on BP and endothelial function (for example, FMD) was observed following grapefruit juice consumption. However, grapefruit flavanones significantly lowered arterial stiffness compared with the control drink^{(Reference Habauzit, Verny and Milenkovic65)}. This effect has been recently attributed to the ability of grapefruit juice flavanones to modulate the expression of genes regulating inflammation, cell interactions and vascular function^{(Reference Krga, Corral-Jara and Barber-Chamoux66)}. Considering glucose metabolism, inflammatory biomarkers and oxidative stress, they were not affected by grapefruit juice^{(Reference Habauzit, Verny and Milenkovic65)}. A similar lack of effect on body composition, BP and glucose homeostasis was achieved by Silver et al. (2011)^{(Reference Silver, Dietrich and Niswender67)} on subjects with obesity, although higher increases in serum HDL-C concentrations in the grapefruit juice group relative to the grapefruit group were observed. Anyway, the evidence on grapefruit is still limited.

There is also a scarcity of available information on the health effects of mandarin or clementine juices in human subject settings. To date, just a few studies have been published. One dealt with children with obesity^{(Reference Codoñer-Franch, López-Jaén and De La Mano-Hernández68)}, where mandarin juice consumption (35 mg of vitamin C, 30 mg/l of low flavanone content, and 700 μg/l of carotenoids) within a 4-week hypoenergetic diet significantly decreased fasting insulin and HOMA-IR index when compared with those who were not supplemented. Treatment reduced some markers of oxidative stress (MDA and carbonyl groups) while increasing the level of circulating antioxidants (vitamin C, α-tocopherol and glutathione)^{(Reference Codoñer-Franch, López-Jaén and De La Mano-Hernández68)}. Another intervention assessed the effect of β-cryptoxanthin-rich Satsuma mandarin juice supplementation (4 mg) in comparison to a β-cryptoxanthin-deprived Satsuma mandarin juice (0 mg) on pulse wave velocity^{(Reference Nakamura, Sugiura and Shibata69)}. A total of 117 participants completed this 12-week parallel intervention and, although serum β-cryptoxanthin concentration increased in the treatment group, there were no differences between the treatment and control groups. Nonetheless, supplementation of both juices led to decreases in brachial ankle pulse wave velocity (PWV) and the levels of oxidative stress biomarkers, reducing the cardiovascular risk^{(Reference Nakamura, Sugiura and Shibata69)}. Despite the lack of effect of β-cryptoxanthin in Satsuma mandarin juice at the cardiovascular level, a previous work showed that the 8-week intake of juice fortified with β-cryptoxanthin might have stimulatory effects on bone formation and inhibitory effects on bone resorption in humans and, in particular, in post-menopausal women^{(Reference Yamaguchi, Igarashi and Uchiyama70)}.

Unlike orange, grapefruit and mandarin, lemon juice is not consumed alone, and there are no 100% lemon juices commercially available. Nevertheless, lemon juice can be used to prepare lemon-based drinks or to acidulate some other juices to create 100% juices, being a way to increase the level of some bioactive compounds in the final beverage^{(Reference González-Molina, Domínguez-Perles and Moreno71,Reference Mena, Martí and Saura72)} . Freitas et al. (2021)^{(Reference Freitas, Boué and Benallaoua73)}, in a randomised crossover trial, investigated the effect of drinking 250 ml lemon juice (50% lemon juice, 50% spring water), black tea or spring water (control) with 100 g of bread on glycaemia and subsequent energy intake. No beverage affected the energy intake, but results showed that lemon juice significantly delayed and reduced peak post-prandial blood glucose concentrations compared with water. Authors suggested that lemon juice, which lowered the pH of the meal, slowed down starch digestion through premature inhibition of salivary α-amylase^{(Reference Freitas, Boué and Benallaoua73)}. Therefore, lemon juice may help to control the glycaemic response, although further research is needed to confirm this aspect. In addition, the role of lemon flavanones should be explored in the light of the evidence collected for orange flavanones. Last but not least, a recent prospective randomised controlled open trial assessed the effects of fresh lemon juice supplementation (60 ml twice per day) to a standard diet on time to stone recurrence in 203 patients with recurrent idiopathic calcium oxalate nephrolithiasis^{(Reference Ruggenenti, Caruso and Cortinovis74)}. Results suggested that lemon juice supplementation might prevent stone recurrence in patients with calcium-oxalate nephrolithiasis. Nevertheless, adherence to the treatment was an issue as it also increased the frequency of gastrointestinal disorders^{(Reference Ruggenenti, Caruso and Cortinovis74)}.

Apple juice

Apples provide good amounts of fibre, in particular pectin, and a variety of (poly)phenols. Juice processing, specifically clarification, notably reduces the fibre and pectin content. In addition, the concentration of (poly)phenols varies considerably between the peel and flesh of the apple. The compounds most found in apple peel are chlorogenic acids, flavan-3-ols (catechin, epicatechin and proanthocyanidins), phloridzin and quercetin derivatives. While quercetin derivatives are found exclusively in apple peel, the other compounds are also in the apple flesh but at much lower concentrations, except for chlorogenic acids, which are higher in the flesh than in the peel^{(Reference Boyer and Liu75)}. Apple pomace has been widely associated with cardiometabolic health benefits^{(Reference Antonic, Jancikova and Dordevic76)}, while the information on apple juice is relatively scarce considering the commercial importance of apple juices. A recent review has tried to bridge this gap^{(Reference Vallée Marcotte, Verheyde and Pomerleau77)} and some relevant articles are presented in Table 3.

Table 3. Characteristics of some representative studies investigating the health effects of 100% apple juice

Juices were 100% juice unless otherwise stated. Abbreviations: BP, blood pressure; BMI, body mass index; BW, body weight; CLJ, clear apple juice; CO, crossover; DB, double-blind; F, female; FRAP, ferric reducing ability of plasma; GSH, glutathione; hs-CRP, high-sensitivity CRP; ICAM-1, intercellular adhesion molecule 1; IGF1, insuline-like growth factor; IGFBC3, Insulin-like growth factor-binding protein 3; IL-6, interleukin-6; HDL-c, HDL-cholesterol; LDL-c, LDL-cholesterol; M, male; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OL, open label; OW, overweight (BMI: 25–30 kg/m²); PBMCs, peripheral blood mononuclear cell; PR, (poly)phenol-rich juice; RCT, randomised controlled trial; RT, randomised trial; SB, single-blind; T2D, type 2 diabetes; TAG, triglycerides; TC, total cholesterol; TNF-α, tumour necrosis factor-α; VCAM-1, vascular cell adhesion molecule; VCR, vitamin C-rich juice; WC, waist circumference; WHR, waist-to-hip ratio; ↓, decreased level; ↑ increased level

Barth et al. (2012)^{(Reference Barth, Koch and Watzl78)}, in a randomised parallel study, investigated the effect of a (poly)phenol-rich cloudy apple juice (total phenol content: 1070 mg/l) or an isoenergetic control drink (without (poly)phenols) on obesity-associated metabolic and endocrine parameters in males with obesity. After 4 weeks at 750 ml/d, cloudy apple juice had no significant effect on BMI and waist circumference, while it significantly reduced percent body fat compared with the control drink. This effect was related to the IL-6–174 G/C polymorphism as body fat reduction was detectable only in C/C carriers but not in G/C or G/G ones^{(Reference Barth, Koch and Watzl78)}. Genetic polymorphisms were considered key to the better understanding of the effect of cloudy apple juice on fat reduction; although, it was not related to other obesity-related parameters. In agreement with this study, a randomised five-arm crossover study reported no changes in body weight and waist-to-hip ratio in healthy individuals who followed a 4-week restricted diet supplemented with 500 ml/d of either cloudy or clear apple juice compared with whole apples (550 g/d), apple pomace (22 g/d) and the control (diet without supplementation)^{(Reference Ravn-Haren, Dragsted and Buch-Andersen79)}. The cloudy and clear apple juice contained 290 and 216 mg polyphenols per litre, respectively, where chlorogenic acid, procyanidin dimers and epicatechin were the major compounds in both juices; in addition, the cloudy juice had a pectin content of 0·94 g/l while it was almost absent in clear juice^{(Reference Ravn-Haren, Dragsted and Buch-Andersen79)}. Therefore, although the literature shows promising results coming from animal studies^{(Reference Asgary, Rastqar and Keshvari80)}, the available information through human subject interventions with apple juice do not account for improvements in anthropometric parameters and body composition, except when specific genotypic differences are considered.

Considering cardiometabolic biomarkers, BP, TC and HDL-C were not affected by cloudy or clear apple juice in healthy subjects^{(Reference Ravn-Haren, Dragsted and Buch-Andersen79)}. However, cloudy apple juice showed a similar – not significant (p = 0·064) – trend to the whole apple and apple pomace in lowering TC and LDL-C, while these parameters increased with clear apple juice compared with whole apple and pomace, but not to the control^{(Reference Ravn-Haren, Dragsted and Buch-Andersen79)}. Similarly, Barth et al. (2012)^{(Reference Barth, Koch and Watzl78)} found no changes in blood lipids upon cloudy apple juice consumption. Results from a meta-analysis of the effects of flavan-3-ol-containing products on blood lipids and including the two previous studies indicated that intake of apple products was associated with reduced TC and LDL-C levels^{(Reference González-Sarrías, Combet and Pinto81)}. Therefore, as the impact of apple juice on blood lipids is still contradictory, further research would be appreciated.

In the case of glucose–insulin homeostasis, cloudy or clear apple juice did not show any effect on glucose metabolism markers (including insulin), in line with the consumption of whole apples or apple pomace^{(Reference Ravn-Haren, Dragsted and Buch-Andersen79)}. On the other hand, Soriano-Maldonado et al. (2014)^{(Reference Soriano-Maldonado, Hidalgo and Arteaga82)} conducted a randomised crossover study in healthy adults for 4 weeks comparing two cloudy apple juices: a vitamin C-rich juice (vitamin C and total polyphenol content: 60 and 510 mg/l, respectively) and a polyphenol-rich juice (22 and 993 mg/l, respectively). An increase in insulin and HOMA index was observed after polyphenol-rich cloudy apple juice consumption but not in the vitamin C-rich cloudy apple juice group. Glucose and blood lipids did not change between treatments^{(Reference Soriano-Maldonado, Hidalgo and Arteaga82)}.

Considering inflammatory markers, no significant changes were found in a panel of systemic and vascular inflammation markers and adipokines after cloudy apple juice consumption by individuals with obesity^{(Reference Barth, Koch and Watzl78)}. Apple consumption, regardless of form (whole, pomace, clear or cloudy juice), did not change high-sensitivity CRP (hs-CRP) levels in healthy volunteers^{(Reference Ravn-Haren, Dragsted and Buch-Andersen79)}. The same results were reported by Soriano-Maldonado et al. (2014)^{(Reference Soriano-Maldonado, Hidalgo and Arteaga82)}, where the vitamin C-rich or polyphenol-rich cloudy apple juice did not change inflammation-related parameters compared to the baseline. Examining biomarkers of oxidative stress, clear and cloudy apple juices did not modify the antioxidant status with respect to the control in healthy individuals^{(Reference Ravn-Haren, Dragsted and Buch-Andersen79)}. These results were in line with the data collected for the polyphenol-rich cloudy apple juice compared to the vitamin C-rich one, as no modifications in plasma antioxidant activity were registered^{(Reference Soriano-Maldonado, Hidalgo and Arteaga82)}. On the contrary, antioxidant activity significantly increased after vitamin C-rich juice, but this increase was not related to the plasma vitamin C levels. In addition, consuming the polyphenol-rich juice reduced total glutathione levels in peripheral blood mononuclear cells^{(Reference Soriano-Maldonado, Hidalgo and Arteaga82)}.

A thorough acute intervention study was conducted to assess the effect of fructose from fruit sources in the increase in uric acid concentration, as it is well known that the intake of large amounts of fructose raises circulating urate^{(Reference White, Carran and Reynolds83)}. Participants ingested two servings (small or large) of apple segments and apple juice, or a glucose and a fructose control beverage. Plasma uric acid levels increased after consumption of all fructose-containing treatments, without differences between apple segments and juices^{(Reference White, Carran and Reynolds83)}. More attention should be paid to this point, as hyperuricemia (high levels of uric acid) represents the main risk factor for gout, the most common inflammatory form of arthritis in men^{(Reference Scanu, Luisetto and Ramonda84)}.

Finally, no changes in microbiota composition were observed after consumption of apple juice, clear or cloudy, by healthy volunteers for 4 weeks^{(Reference Ravn-Haren, Dragsted and Buch-Andersen79)}. Similar results were achieved after cloudy apple juice intake by patients with T2D for the same time period: although small decreases were found in numbers of Enterococci and Firmicutes, the overall microbiota profile did not change when compared with the isoenergetic control beverage^{(Reference Cho, König and Seifert85)}.

Few high-quality studies have been conducted on apple juice despite its market share. No significant effects of apple juice on human subject health have been discovered. Apple juice only showed a moderate effect in reducing body fat depending on gene polymorphisms. This is a point worth mentioning as inter-individual variability may be key to better evaluating the beneficial effects of apple juice. Indeed, as a notable inter-individual variability has been reported in the metabolism of apple flavan-3-ols^{(Reference Anesi, Mena and Bub86)}, leading to different profiles of biologically active metabolites, individual profiles of phenolic metabolites, as well as genotypic differences, should be investigated in further interventions studies with apple juice. Further efforts to assess the impact of compositional differences would favour further mechanistic research.

Grape juice

Red (or purple) and white grape juices from Vitis vinifera cultivars are predominant in the market, and most of the scientific literature on the health properties of grape juices is related to these two juices. Nevertheless, other purple grape juices made from Concord (V. labrusca) and Muscadine (V. rotundifolia) cultivars have received much attention in recent years. Grapes are typically good sources of (poly)phenols, being rich in flavan-3-ols (both catechins and proanthocyanidins), flavonols, phenolic acids, stilbenes and, in the case of red/purple cultivars, anthocyanins. The cultivar used for juice production substantially influences the phenolic profile and content, as well as environmental factors and cultivation methods^{(Reference Granato, de Magalhães Carrapeiro and Fogliano87)}. As a general indication, white grape juice shows higher contents in phenolic acids and resveratrol and lower contents in flavan-3-ols, flavonols, and anthocyanins^{(Reference Blumberg, Vita and Oliver Chen88)}. Although their health effects may change notably due to the grape type (Table 4), all the grape juices are discussed together to provide a global overview and since the evidence in humans is not so broad as for grape wines.

Table 4. Characteristics of some representative studies investigating the health effects of 100% grape (white, red/purple and Concord) juice

Juices were 100% juice unless otherwise stated. Abbreviations: AC, abdominal circumference; Apo A, apolipoprotein A; Apo B, apolipoprotein B; BA, brachial artery; BP, blood pressure; CAD, coronary artery disease; CO, crossover; DB, double-blind; DBP, diastolic blood pressure; F, female; FMD, flow-mediated dilation; GAE, gallic acid equivalent; GJ, grape juice; HDL-c, HDL-cholesterol; HR, heart rate; ICAM-1, intercellular adhesion molecule 1; JG, juice group; LDL-c, LDL-cholesterol; Lp(a), lipoprotein (a); M, male; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OL, open label; OW, overweight (BMI: 25–30 kg/m²); NO, nitric oxide; NTGD, nitroglycerin-mediated vasodilation; PEH, post-exercise hypotension; PWV, pulse wave velocity; RCT, randomised controlled trial; RT, randomised trial; SB, single-blind; SBP, systolic blood pressure; SOD, superoxide dismutase; TAG, triglycerides; TC, total cholesterol; TRAP, plasma total radical-trapping antioxidant potential; VCAM-1, vascular cell adhesion molecule; WC, waist circumference; ↓, decreased level; ↑ increased level

Zuanazzi et al. (2019)^{(Reference Zuanazzi, Maccari and Beninca89)} investigated the health effects of 100% white grape juice from V. labrusca, characterised by a total phenolic content of 267·9 mg/l, where caffeic acid (13·9 mg/l) and (+)-catechin (11·3 mg/l) were the most abundant compounds. Over a 30-d intervention at 7 ml/kg/d (490 ml/d for a 70-kg woman), white grape juice supplementation to twenty-five non-smoking women (eleven eutrophic and fourteen who were overweight or obese) reduced BMI and waist and abdominal circumference. Nevertheless, the lack of a control arm precludes solid conclusions. Indeed, in contrast to this work, Dohadwala et al. (2010)^{(Reference Dohadwala, Hamburg and Holbrook90)} reported no change in body weight following consumption of 7 ml/kg/d of 100% Concord grape juice (total phenols: 1970 mg/l) or placebo beverage for 8 weeks in sixty-four otherwise healthy patients (with prehypertension and with stage one hypertension, but taking no anti-hypertensive medications).

Considering vascular function, BP did not change after consumption of white (Zuanazzi et al. 2019)^{(Reference Zuanazzi, Maccari and Beninca89)} or Concord grape juice^{(Reference Dohadwala, Hamburg and Holbrook90)}. Similarly, no effect on BP was observed after 12-week daily consumption of 355 ml Concord grape juice (total phenols: 2188 mg/l, 21% anthocyanins and 43% proanthocyanidins) in healthy middle-aged working mothers compared with a placebo^{(Reference Lamport, Lawton and Merat91)}. Nevertheless, some works have reported a positive effect of grape juice consumption on both systolic and diastolic BP in healthy individuals (Concord grape juice, 5·5 ml/kg/d, 8 weeks)^{(Reference Park, Lee and Park92)}, hypertensive men (Concord grape juice, 5·5 ml/kg/d, 8 weeks)^{(Reference Park, Kim and Kang93)} and hypercholesterolemic patients (purple grape juice, 500 ml/d, 2 weeks)^{(Reference Coimbra, Lage and Brandizzi94)}, so that a beneficial effect of grape juice on BP cannot be ruled out. While BP effects are contradictory, the evidence on arterial function assessed using FMD seems more promising. Coimbra et al. (hypercholesteraemic patients)^{(Reference Coimbra, Lage and Brandizzi94)}, Stein et al. (coronary patients, Concord grape juice, 8 ml/kg/d, 2 weeks)^{(Reference Stein, Keevil and Wiebe95)}, Chou et al. (coronary patients, Concord grape juice, 4 and 8 ml/kg/d, 8 weeks)^{(Reference Chou, Keevil and Aeschlimann96)}, and Siasos et al. (healthy smokers, Concord grape juice, 7 ml/kg/d, 2 weeks; juice phenolics: flavan-3-ols 434 μmol/l, anthocyanins 296 μmol/l, hydroxycinnamates 162 μmol/l, and flavonols 76 μmol/l)^{(Reference Siasos, Tousoulis and Kokkou97)} demonstrated that short-term ingestion of purple/Concord grape juice could improve FMD in different population settings.

Regarding lipid profile, white^{(Reference Zuanazzi, Maccari and Beninca89)}, purple^{(Reference Coimbra, Lage and Brandizzi94)}, and Concord^{(Reference Chou, Keevil and Aeschlimann96)} grape juice did not affect blood lipids, except for a 16% increase in HDL-C in women consuming white grape juice for 30 d^{(Reference Zuanazzi, Maccari and Beninca89)}. In the case of glucose metabolism, grape juice supplementation did not modify glycaemia or insulin levels^{(Reference Zuanazzi, Maccari and Beninca89,Reference Dohadwala, Hamburg and Holbrook90)} , unless for a minimal but significant reduction in glucose levels after 8-week Concord grape juice consumption^{(Reference Dohadwala, Hamburg and Holbrook90)}. Taking into account other cardiometabolic biomarkers, no major changes in blood markers of inflammation and platelet activity are reported in the literature.

Examining biomarkers of oxidative stress, Zuanazzi et al. (2019)^{(Reference Zuanazzi, Maccari and Beninca89)} reported no change in nitric oxide levels or markers of oxidative damage following consumption of white grape juice in women. No change in lipid peroxidation was observed following a single dose of 10 ml/kg of 100% purple grape juice in recreational runners^{(Reference de Lima Tavares Toscano, Silva and de França98)}. Differently, O’Byrne et al. (2002)^{(Reference O’Byrne, Devaraj and Grundy99)} found that consumption of 10 ml/kg/d 100% Concord grape juice for 2 weeks increased serum antioxidant capacity and protected LDL against oxidation in thirty-two healthy adults, to an extent similar to that obtained with 268 mg/d α-tocopherol, while decreased native plasma protein oxidation significantly more than α-tocopherol. A reduction in LDL susceptibility to oxidation was also reported by Stein et al. ^{(Reference Stein, Keevil and Wiebe95)}. Last, protective effects on lymphocyte DNA damage have also been reported^{(Reference Park, Lee and Park92)}.

The association between grape juice and cognitive function has been widely studied. A recent critical review of epidemiological and randomised-controlled human subject trials regarding the role of grapes and their derivatives in modulating cognitive decline was conducted by Restani et al. (2021)^{(Reference Restani, Fradera and Ruf100)}. All reviewed studies investigating the effect of grape juice consumption reported improved cognitive function in the intervention group versus the control group after both single dose and long-term (up to 6 months) supplementation. Remarkably, the most encouraging results included reaction times, verbal skills, degree of orientation, learning, and memory. Cognitive improvement was observed in both healthy young adults (21·05 ± 0·89 years)^{(Reference Haskell-Ramsay, Stuart and Okello101)}, healthy middle-aged women (40–50 years)^{(Reference Lamport, Lawton and Merat91)}, and older adults with mild cognitive decline (78·2; SD 5·0 years)^{(Reference Krikorian, Boespflug and Fleck102)}. Restani et al. (2021)^{(Reference Restani, Fradera and Ruf100)} concluded that Concord and purple grape juice consumption (200–500 ml/d) was generally associated with improved cognitive performance. Similar conclusions have been presented in a recent systematic review on the topic^{(Reference Wang, Haskell-Ramsay and Gallegos103)}.

Grape juice is considered a potentially ergogenic food for sports performance. De Lima Tavares Toscano et al. (2020)^{(Reference Toscano, Tavares and Toscano104)} demonstrated that drinking a single dose of 100% purple grape juice (10 ml/kg, total phenols: 3106 mg/l) increased run time to exhaustion by almost 19% in fourteen male recreational runners. In a previous study, Toscano et al. (2015)^{(Reference Leal, Leal and Leal105)} also found a 15% increased physical performance following 28-d supplementation with a similar juice (10 ml/kg/d, same grape cultivar). Authors also reported some benefits in inflammatory markers in these runners. Last, contrary to the evidence on BP upon chronic consumption of grape juice by non-sport individuals, grape juice supplementation may also help BP management after aerobic exercise^{(110, 111)}.

In conclusion, homogeneous results were observed for vascular and cognitive function, showing moderate improvements after red or Concord grape juice consumption. Some benefits in specific markers of oxidative stress are also well documented. The benefits on exercise conditions are of interest, but they cannot be extrapolated to all physical activity contexts. More discordant results for other outcomes were achieved. In general, the evidence to date is quite limited, considering the importance of this juice at the market level. Although some health benefits might be similar to those investigated for grape wines^{(Reference Mena, Calani and Dall’Asta108)}, further high-quality intervention studies are needed. In addition, emphasis should also be paid to white grape juice, as the evidence is quite limited. Comparisons among different classes of grape juice (white, red or Concord) may help to better understand the potential role of grape juices in disease prevention and what are the responsible compounds backing the physiological effects observed.

Pomegranate juice

Regarding main bioactive compounds, pomegranate juice contains anthocyanins, ellagitannins and ellagic acid derivatives, among other phenolics^{(Reference Fischer, Carle and Kammerer109)}. Although the edible part of pomegranate (arils) contains almost exclusively anthocyanins, commercial pomegranate juices also present high amounts of ellagitannins and ellagic acid coming from the peel and inner parts (mesocarp) as a consequence of the juicing process^{(Reference Fischer, Carle and Kammerer110)}. In addition, besides cultivars, agronomical and harvest conditions, processing may pretty much alter the phytochemical composition of pomegranate juices^{(Reference Mena, Martí and Saura72,Reference Mena, García-Viguera and Navarro-Rico111,Reference Mena, Gironés-Vilaplana and Moreno112,Reference Kandylis and Kokkinomagoulos113)} . Several works have reviewed the evidence behind the potential health benefits of pomegranate juice^{(Reference Mena, Gironés-Vilaplana and Moreno112,Reference Kandylis and Kokkinomagoulos113,Reference Giménez-Bastida, Ávila-Gálvez and Espín114,Reference Dall’Asta, Angelino and Del Rio115,Reference Moazzen and Alizadeh116)} , so here only a brief overview is presented. Of note, many studies have focused on the role of pomegranate juice in patients rather than in the general population (Table 5).

Table 5. Characteristics of some representative studies investigating the health effects of 100% pomegranate juice

Juices were 100% juice unless otherwise stated. Abbreviations: CML, carboxy methyl lysine; CO, crossover; DB, double-blind; DBP, diastolic blood pressure; F, female; FG, fasting blood glucose; HDL-c, HDL-cholesterol; HOMA-IR, homeostasis model assessment-insulin resistance; hs-CRP, high-sensitivity CRP; LDL-c, LDL-cholesterol; M, male; MDA, malondialdehyde; MetS, metabolic syndrome; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OL, open label; OW, overweight (BMI: 25–30 kg/m²); ox-LDL, oxidised-LDL; PON1, paraoxonase-1; PP, pulse blood pressure; PWV, pulse wave velocity; RCT, randomised controlled trial; RT, randomised trial; SB, single-blind; SBP, systolic blood pressure; T2D, type 2 diabetes; TAC, total antioxidant capacity; TAG, triglycerides; TC, total cholesterol; VLDL, very low density lipoproteins; ↓, decreased level; ↑ increased level

Moazzen & Alizadeh (2017)^{(Reference Shema-Didi, Kristal and Sela118)}, in a double-blinded randomised crossover controlled trial, investigated the effects of 1-week supplementation with 500 ml/d of pomegranate juice on cardiovascular risk factors in thirty patients with metabolic syndrome. The juice contained 100 mg/l of anthocyanins, while the composition in ellagitannins was not reported. Results showed a reduction in systolic and diastolic BP in the intervention group compared with placebo^{(Reference Sohrab, Roshan and Ebrahimof117)}. Similarly, after 6 weeks at 200 ml/d of pomegranate juice (total polyphenols: 2125 mg/l), a significantly reduced systolic and diastolic BP was observed in sixty patients with T2D in comparison with the control group (no juice supplementation)^{(Reference Lynn, Hamadeh and Leung119)}. In another double-blind randomised controlled study in 101 haemodialysis patients, Shema-Didi et al. (2014)^{(Reference Sahebkar, Ferri and Giorgini120)} found that systolic BP decreased with respect to placebo after one year of consuming 100 ml pomegranate juice (total polyphenols: about 1190 mg/l) three times per week. However, improvements in BP or PWV were not observed in fifty-one healthy subjects (middle-aged normotensive adults) consuming 330 ml/d of pomegranate juice (total polyphenols: 3162 mg/l; potassium: 1711 mg/l) for 4 weeks^{(Reference Sohrab, Ebrahimof and Sotoudeh121)}. Cumulative evidence indicates that pomegranate juice consumption may lead to consistent benefits on BP, as stated by a SRMA of eight RCT^{(Reference Sohrab, Angoorani and Tohidi122)}. Nevertheless, these BP improvements happen mainly in hypertensive patients, regardless of the presence/absence of antihypertensive medication, that may also suffer from other cardiometabolic diseases. Further information on healthy subjects or individuals at risk of hypertension may help to draw preventive strategies taking pomegranate juice into account.

Evaluating the effect of pomegranate juice on lipid profile, 500 ml/d supplementation for 1-week increased blood TG and very low-density lipoprotein cholesterol (VLDL-C) levels in individuals with metabolic syndrome, as opposed to the placebo, whereas TC, HDL-C, and LDL-C did not change^{(Reference Sohrab, Roshan and Ebrahimof117)}. Authors hypothesised that this negative effect could disappear in a long-term supplementation so, rather than as a negative outcome, this data should be regarded as part of the inconclusive evidence on the role of pomegranate juice in the lipid profile. For instance, contrary to the previous work, TC, LDL-C, HDL-C and TG concentrations were not different compared with the control group after 6-week pomegranate juice consumption in patients with T2D^{(Reference Lynn, Hamadeh and Leung119)}. Shema-Didi et al. (2014)^{(Reference Sahebkar, Ferri and Giorgini120)} found that 1-year pomegranate juice consumption, compared with the placebo, improved TG and HDL-C levels in haemodialysis patients, with a greater effect in subjects with hypertension, low HDL-C and high blood TGs (higher cardiovascular risk)^{(Reference Sahebkar, Ferri and Giorgini120)}. Overall, it seems that pomegranate juice is not able to modify the lipid profile consistently. Differences may be due to the heterogeneous evidence available, as well as due to the different physiological responses of individuals metabolising differently pomegranate ellagitannins^{(Reference Dall’Asta, Angelino and Del Rio115)}.

In the case of glucose–insulin metabolism, pomegranate juice did not change fasting blood glucose, insulin and HOMA-IR in patients with metabolic syndrome on 1 week of consumption^{(Reference Sohrab, Roshan and Ebrahimof117)}. Similarly, no adverse effect on fasting blood glucose level was found after 6 weeks at 200 ml/d^{(Reference Kerimi, Nyambe-Silavwe and Gauer123)} and 12 weeks at 250 ml/d^{(Reference Sahebkar, Gurban and Serban124)} in T2D patients. In the case of acute responses, Kerimi et al. (2017)^{(Reference Asgary, Karimi and Joshi125)} found that 200 ml pomegranate juice (71·5 mg punicalin, 12·4 mg punicalagin, 4·8 mg ellagic acid and 2·8 mg ellagic acid hexose), consumed with approximately 109 g of white bread (50 g available carbohydrate), significantly attenuated the post-prandial glycaemic response in healthy subjects, compared with a control solution containing the equivalent amount of sugars. These authors proposed that the effect was primarily due to the ability of punicalagin to inhibit human subject salivary α-amylase, whereas microbial metabolites of pomegranate ellagitannins (namely urolithins) may also modulate glucose metabolism after the acute post-prandial period. Interestingly, a pomegranate polyphenol-rich extract was ineffective on glycaemic response, and the beneficial effects were only observed after juice intake. So, while pomegranate juice does not seem able to modify glucose metabolism in chronic conditions, it may be able to reduce acute post-prandial glycemic response in healthy subjects^{(Reference Asgary, Karimi and Joshi125)}.

In the context of inflammation, Moazzen and Alizadeh (2017)^{(Reference Shema-Didi, Kristal and Sela118)} found that pomegranate juice (500 ml/d, 1 week), but not the placebo, decreased hs-CRP levels in patients with metabolic syndrome. However, when pooling evidence, a meta-analysis of five RCT did not indicate a significant effect of pomegranate juice in lowering plasma CRP levels^{(Reference Siddarth, Li and Miller126)}. Similarly, a recent SRMA of six RCT showed no significant effect of pomegranate juice on vascular adhesion factors, intercellular adhesion molecule 1 (ICAM-1), VCAM-1 and E-selectin compared with the control group, but a significant effect in reduction of IL-6^{(Reference Ammar, Bailey and Chtourou127)}.

Examining biomarkers of oxidative stress, the information is not so robust, but still promising. For instance, Sohrab et al. (2017)^{(Reference Kerimi, Nyambe-Silavwe and Gauer123)} found that consumption of 200 ml/d of pomegranate juice for 6 weeks significantly decreased oxidised LDL (ox-LDL) and anti-oxidised LDL antibodies, as well as increased total serum antioxidant capacity and arylesterase activity of paraoxonase-1 (PON-1) in T2D patients in comparison with the control group. An increase in total antioxidant capacity in diabetic patients had already been observed in a previous study^{(Reference Sahebkar, Gurban and Serban124)}; however, reductions in any biomarkers of oxidative stress were not reported. Detailed information on the potential antioxidant effects of pomegranate juice, particularly in the context of LDL oxidation, is broadly discussed in some comprehensive reviews^{(Reference Mena, Gironés-Vilaplana and Moreno112,Reference Dall’Asta, Angelino and Del Rio115,Reference Moazzen and Alizadeh116)} .

Taking into account health outcomes beyond cardiometabolic diseases, some information has been published with regard to memory and exercise performance. In a recent randomised double-blind placebo-controlled study, Siddarth et al. (2020)^{(Reference Nile and Park128)} investigated the memory effects of pomegranate juice in non-demented middle-aged and older adults. The intervention group did not show any change in the ability to learn visual information after daily consumption of 236 ml (8 oz.) of pomegranate juice (368 mg punicalagin, 93 mg anthocyanins, 29 mg ellagic acid and 98 mg other tannins) for 12 months, whereas the placebo group showed a significant decline. Other visual and verbal memory measures were not significantly different between groups. On the other hand, pomegranate juice has also been evaluated for its effect on exercise performance and post-exercise recovery. Ammar et al. (2018)^{(Reference Skrovankova, Sumczynski and Mlcek129)}, in a systematic review of eleven studies, nine on pomegranate juice and two on pomegranate extract, concluded that pomegranate could enhance exercise performance and expedite recovery from intensive exercise in healthy adults. However, positive effects were more likely when pomegranate juice contained >1·4 g/l total (poly)phenols, when it was consumed at least 60 min before exercise, and when large muscle mass exercise was engaged.

In conclusion, pomegranate juice may significantly improve BP and, to a limited extent, post-prandial glycaemic response, markers of oxidative stress, some cognitive markers and exercise performance. Further well-designed experiments are needed to confirm these insights^{(Reference Dall’Asta, Angelino and Del Rio115)}. Attention should be paid to the accurate characterisation of the phytochemical profile (most of the studies present a scarce characterisation of the juice bioactives) and the inter-individual variability in the metabolism of their bioactive compounds (the existence of diverse metabolic phenotypes in the metabolism of ellagitannins is well-known). Of note, while ellagitannins, in particular punicalagins, have attracted much attention in nutrition research, the amount of anthocyanins in some pomegranate juices may also play a role in their beneficial effect, and they should be considered^{(Reference Mena, Gironés-Vilaplana and Moreno112,Reference Moazzen and Alizadeh116)} .

Berry juices

Berries are a broad family of species including cranberries, blueberries, strawberries, raspberries, blackberries, chokeberries, blackcurrants, elderberries and bilberries, among others. They have attracted considerable attention owing to their potential benefits to human subject health^{(Reference Zhao, Liu and Gu130)}. Besides vitamins, minerals, and fibre, they present high amounts of different bioactives, in particular (poly)phenols. Berries are rich in flavonoids, such as coloured anthocyanins, flavonols, and flavan-3-ols (both catechins and proanthocyanidins), phenolic acids (both hydroxybenzoic and hydroxycinnamic acids) and hydrolysable tannins such as ellagitannins. Their phenolic composition depends on the botanical species, cultivar, growing location, environmental and growing conditions, ripeness stage, time of harvest, and subsequent storage conditions and processing methods^{(Reference Javid, Maghsoumi-Norouzabad and Ashrafzadeh131)}. To date, most of the evidence on the health effects of berries comes from dietary interventions considering whole fruits or extracts, while the insights derived from dietary interventions with berry juices are scarce. This review will focus on major berry juices such as cranberry and chokeberry, but will also consider other interesting outcomes associated with other berry juices.

Cranberry juice

Cranberries (Vaccinium macrocarpon) have been widely associated with the prevention of urinary tract infections and cardiometabolic diseases. Their beneficial effects have been associated with their unique phenolic profile, rich in anthocyanins, proanthocyanidins (both A- and B-type), flavonols and hydroxycinnamic acids. However, evidence is inconclusive and further research is needed^{(Reference Zare Javid, Maghsoumi-Norouzabad and Ashrafzadeh132)}. The following lines will provide an overview of the state-of-the-art, taking into account, when possible, 100% cranberry juices, although most of the evidence is provided for beverages prepared with cranberry juice concentrate (Table 6). Cranberry beverages prepared with powders/extracts were not considered.

Table 6. Characteristics of some representative studies investigating the health effects of cranberry juices

Juices were 100% juice unless otherwise stated. Abbreviations: BMI, body mass index; BP, blood pressure; CO, crossover; DB, double-blind; DBP, diastolic blood pressure; F, female; FG, fasting blood glucose; FMD, flow-mediated dilation; GSH, glutathione; HbA1c, hemoglobin A1c; HDL-c, HDL-cholesterol; HOMA-IR, homeostasis model assessment-insulin resistance; hs-CRP, high-sensitivity CRP; ICAM-1, intercellular adhesion molecule 1; IL-1β, interleukin-1β; IL-6, interleukin-6; IL-10, interleukin-10; LDL-c, LDL-cholesterol; M, male; MDA, malondialdehyde; MetS, metabolic syndrome; NO, nitric oxide; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OL, open label; OW, overweight (BMI: 25–30 kg/m²); ox-LDL, oxidised-LDL; PD, pocket depth; PWV, pulse wave velocity; RCT, randomised controlled trial; RT, randomised trial; SB, single-blind; SOD, superoxide dismutase; TAG, triglycerides; TC, total cholesterol; TNF-α, tumour necrosis factor-α; UTI, urinary tract infection; VCAM-1, vascular cell adhesion molecule; VLDL, very low density lipoproteins; WC, waist circumference; ↓, decreased level; ↑ increased level

Regarding anthropometric parameters, the information available is quite limited. Javid et al. (2017)^{(Reference Basu, Betts and Ortiz133)} indicated that daily supplementation of 400 ml cranberry juice (230 mg/l vitamin C, 40 mg/l anthocyanins and 535 mg/l proanthocyanidins, 13 g/l fructose, 4 g/l glucose and 2 g/l sucrose) for 8 weeks in diabetic patients with periodontal disease did not condition BMI or waist circumference. Similarly, 480 ml/d of low-calorie cranberry beverage (27% cranberry juice, 250 mg/l vitamin C, 52 mg/l anthocyanins, 496 mg/l proanthocyanidins, 20 g/l fructose, 8 g/l glucose, 0·4 g/l sucrose) for 8 weeks did not significantly affect waist circumference in women with metabolic syndrome^{(Reference Pourmasoumi, Hadi and Najafgholizadeh135)}.

The former study also assessed the effect of juice supplementation on BP, but no significant improvements following cranberry or placebo beverage were observed^{(Reference Pourmasoumi, Hadi and Najafgholizadeh135)}. Conversely, Novotny et al. (2015)^{(Reference Richter, Skulas-Ray and Gaugler136)} found a reduction in diastolic BP in healthy adults (predominantly subjects who were overweight/obese) after 8 weeks at 480 ml/d of low-calorie cranberry juice (its composition was almost identical to that previously reported for Basu et al., 2011, also for sugar content)^{(Reference Pourmasoumi, Hadi and Najafgholizadeh135)} compared with the placebo beverage. Nevertheless, pooled evidence from a recent meta-analysis indicated that cranberry juice could not reduce systolic or diastolic BP, while cranberry products as a whole may favour physiologically relevant decreases in systolic BP^{(Reference Rodriguez-Mateos, Feliciano and Boeres137)}. A recent work carried out in middle-aged adults with overweight/obesity and elevated brachial blood pressure also accounted for a lack of effect of cranberry juice (8 weeks, 500 ml/d, 27% cranberry juice; 27 g/l of total sugars of which 19 g/l are added sugars) on BP^{(Reference Favari, Mena and Curti138)}. In the case of acute studies on vascular function, the work by Rodriguez-Mateos et al. (2016)^{(Reference Mena, Bresciani and Brindani139)} is worth mentioning as it assessed whether cranberry juice consumption in a dose-dependent manner could improve vascular function in healthy men. The different doses (450 ml) were equivalent to 25, 48, 76, 94 and 117% of concentrated cranberry juice, having a total sugar-content of 78 g/l. Juice supplementation increased FMD in a dose-dependent way and the effect was correlated to a series of phenolic metabolites belonging to different classes of (poly)phenols^{(Reference Mena, Bresciani and Brindani139)}. Interestingly, cranberry juice consumption also increased the production of phenyl-γ-valerolactones in a dose-dependent way^{(Reference Dohadwala, Holbrook and Hamburg140)}, which are the main gut microbial metabolites derived from flavan-3-ols that may be behind some beneficial effects attributed to flavan-3-ol-rich sources such as cranberry juice^{(Reference Mathison, Kimble and Kaspar141)}. Improvements in vascular function (PWV) have also been reported upon chronic consumption of cranberry juice (54% juice, 835 mg total polyphenols and 94 mg anthocyanins; 3% glucose and 1% fructose) for 4 weeks in comparison with placebo^{(Reference Simão, Lozovoy and Simão142)}.

Regarding biochemical parameters associated with cardiometabolic risk, a recent SRMA concluded that neither cranberry products nor cranberry juice specifically improve TC, LDL-C, HDL-C, TG, fasting glucose, fasting insulin and HOMA-IR⁽¹³⁸⁾. Some controversy has been observed for fasting serum TG levels, as they decreased after 8-week cranberry juice consumption in the work by Novotny et al. (2015)^{(Reference Richter, Skulas-Ray and Gaugler136)}, while they increased significantly in the study conducted by Basu et al. (2011)^{(Reference Pourmasoumi, Hadi and Najafgholizadeh135)} in women with metabolic syndrome. The recent paper by Richter et al. (2021)^{(Reference Favari, Mena and Curti138)} confirmed the lack of effect in blood lipids of cranberry juice after 8 weeks at 500 ml/d in subjects at risk of CVD, while it reported an interesting shift in the lipoprotein profile.

Considering inflammation biomarkers, CRP did not change after cranberry juice consumption^{(Reference Rodriguez-Mateos, Feliciano and Boeres137)}, and a lack of effect on other biomarkers (IL-6, IL-10, IL-1β, tumour necrosis factor-alpha or TNF-α, soluble ICAM or sICAM, and soluble VCAM or sVCAM) has also been reported^{(Reference Richter, Skulas-Ray and Gaugler136)}. Cranberry juice significantly increased plasma antioxidant capacity and decreased ox-LDL and MDA at 8 weeks versus placebo^{(Reference Pourmasoumi, Hadi and Najafgholizadeh135)}. In agreement with these findings, other works have also reported improvements in the endogenous antioxidant status^{(Reference Maki, Kaspar and Khoo144)} and oxidative stress measurements^{(Reference Stapleton, Dziura and Hooton145)}.

Many studies have investigated the effect of cranberry juice on the prevention of recurrent urinary tract infections (UTI)^{(Reference Fu, Liska and Talan146)}. To date, the results are contradictory. Maki et al. (2016)^{(Reference Tambunan and Rahardjo147)} reported that consumption of a cranberry juice beverage (27% juice, 240 ml/d, 5 mg/l anthocyanins, 496 mg/l proanthocyanidins, 25 g/l sugars) for 24 weeks significantly reduced the number of UTI episodes in women (40·9 SD 1·1 years) with a history of ≥2 UTI in the previous year. Conversely, Stapleton et al. (2012) ^{(Reference Xia, Yang and Xu148)} did not observe a significant reduction in UTI risk after juice consumption juice (27% juice and sucralose, 120 or 240 ml/d) compared with placebo in premenopausal women with a history of ≥1 UTI in the previous year. To comprehensively assess the effect of cranberry on the risk of UTI recurrence in otherwise healthy women, Fu et al. (2017)^{(Reference Mena, González de Llano and Brindani149)} conducted a SRMA of seven randomised controlled trials. The meta-analysis revealed that cranberry products could effectively prevent UTI recurrence, but this protective effect was not significant when only juice-related studies were taken into account. Similar insights (protective effect of cranberry products, but not cranberry juice by itself) were reported for another, more updated meta-analysis for women and UTI^{(Reference Llano, Esteban-Fernández and Sánchez-Patán150)}. Nevertheless, the last meta-analysis published in this field, taking into account not only women with recurrent UTI but also children and patients using indwelling catheters, for a total of almost 4000 participants, concluded that cranberry juice could be considered an adjuvant therapy for preventing UTI in susceptible populations^{(Reference Mena, Favari and Acharjee151)}. Of note, all the authors of these meta-analyses emphasised the need for larger, high-quality studies to confirm these results. A call for caution in the design of new studies is thus needed as one of the reasons behind this inconsistent evidence may be related to the high inter-individual variability associated with the metabolism of cranberry phenolics^{(Reference Mena, Bresciani and Brindani139,Reference Dohadwala, Holbrook and Hamburg140)} . Some phenolic metabolites, in particular phenyl-γ-valerolactones, have demonstrated to exert a high anti-adhesive activity against uropathogenic Escherichia coli in bladder epithelial cells at concentrations achievable upon consumption of reasonable amounts of cranberry juice^{(Reference Li, Ma and Guo152,Reference Sidor, Drożdżyńska and Gramza-Michałowska153)} . Consequently, they may be the responsible compounds behind the preventive features of cranberry juice on UTI. However, not everybody metabolises these compounds equally^{(Reference Mena, Bresciani and Brindani139,Reference Dohadwala, Holbrook and Hamburg140)} and some individuals may benefit more from a dietary intervention with cranberry juice. Keeping in mind the existence of different metabolic phenotypes in the production of phenyl-γ-valerolactones and phenylpropanoic acids after cranberry intake^{(Reference Pokimica, García-Conesa and Zec154)} might be a winning strategy to demonstrate the effectiveness of this juice in UTI prevention.

The antimicrobial activity of cranberry juice components has also been investigated for Helicobacter pylori. H. pylori infection can induce peptic ulcers and increase the risk of developing gastric cancer. In a recent double-blind randomised placebo-controlled study, Li et al. (2021)^{(Reference Loo, Erlund and Koli155)} evaluated the effects of cranberry juice in 522 H. pylori-infected adults. Consumption of 240 ml of a beverage containing 35% cranberry juice (4·6% cranberry concentrate, 11·8% beet sugar and 83·6% water, 183 mg/l A-type proanthocyanidins) twice daily for 8 weeks (high-proanthocyanidin cranberry juice) decreased infection rate by 20% as compared with other dosages (low- and medium-proanthocyanidin cranberry juices) and placebo. Furthermore, an increase in the percentage of H. pylori-negative participants was observed. Contrary to juice, encapsulated cranberry powders were not significantly effective.

Overall, cranberry juice showed no effect on body weight, BP, blood lipids, glycaemic control and biomarkers of inflammation. Nevertheless, promising insights have been reported about the role of cranberry juice on vascular function (FMD), redox status, prevention of UTI recurrence in susceptible populations and suppression of H. pylori infection. Although most of the publications were conducted with juice made from concentrates and presenting low/medium (27–56%) amounts of juice, the evidence behind this juice is encouraging and may boost the development of 100% cranberry juices acceptable from a sensory point of view. Cranberry juices are preferably blended with other fruit juices or added with sweeteners to improve palatability affected by the tart taste. In any case, the amount of phenolic compounds provided by these diluted juices is high compared with other FVJ and this may, in fact, back the beneficial effects observed. To better understand cranberry juice effects, further research should consider the high inter-individual variability in the metabolism of cranberry phenolics.

Black chokeberry juice

Black chokeberry or aronia (Aronia melanocarpa) is known for its high content in anthocyanins and intense red/black colour, but it also contains high amounts of hydroxycinnamic acids and proanthocyanidins. This significant amount of (poly)phenols has been related to the potential benefits on human subject health^{(Reference Nowak, Gośliński and Wesołowska156)}. Nevertheless, the evidence of chokeberry juice in humans is limited as only a few studies have been conducted (Table 7). Most of the biological properties of aronia have been tested in human subject interventions with extracts or animal models, as recently reviewed^{(Reference Nowak, Gośliński and Wesołowska156)}.

Table 7. Characteristics of some representative studies investigating the health effects of some berries (chokeberry, blueberry, bayberry, bilberry, barberry, blackcurrant, sea buckthorn, açaí, juçara and noni) juices

Juices were 100% juice unless otherwise stated. Abbreviations: AJ, Açaí juice; BG, blood glucose; BJ, blackcurrant juice drink; BMI, body mass index; BW, body weight; CO, crossover; CVD, cardiovascular disease; DB, double-blind; DBP, diastolic blood pressure; F, female; GAE, gallic acid equivalent; HDL-c, HDL-cholesterol; HR, heart rate; hs-CRP, high-sensitivity CRP; ICAM-1, intercellular adhesion molecule 1; IL-6, interleukin-6; IL-7, interleukin-7; IL-8, interleukin-8; IL-10, interleukin-10; IL-13, interleukin-13; IL-15, interleukin-15; JJ, Juçara juice; LDL-c, LDL-cholesterol; M, male; MIG, monokine induce by INF-γ; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OL, open label; OSI, oxidative stress index; OW, overweight (BMI: 25–30 kg/m²); PCG, protein carbonyl groups; RCT, randomised controlled trial; RT, randomised trial; SB, single-blind; SBP, systolic blood pressure; T2D, type 2 diabetes; TAC, total antioxidant capacity; TAG, triglycerides; TC, total cholesterol; TNF-α, tumour necrosis factor-α; VCAM-1, vascular cell adhesion molecule; ↓, decreased level; ↑ increased level

In a randomised controlled double-blind study, Pokimica et al. (2019)^{(Reference Rahmani, Clark and Kord Varkaneh157)} found no significant change in BMI after a 4-week intervention with 100 ml/d of high- or low-dose polyphenol chokeberry juice (or a polyphenol-free placebo drink) in individuals with cardiovascular risk. The high-polyphenol juice consisted of a 100% chokeberry juice containing almost 12000 mg/l total phenols and 1100 mg/l anthocyanins; the low polyphenol dose was prepared by dilution (1:3) of the juice into the placebo drink. Similarly, in an 8-week intervention study in subjects with untreated mild hypertension, body weight remained unchanged after 300 ml/d of a 100% chokeberry hybrid juice mixed with chokeberry powder (3 g)^{(Reference Stote, Sweeney and Kean158)}. Chokeberry juice mixed with the powder provided a total of 7313 mg of polyphenols per litre, anthocyanins (3413 mg/l) and proanthocyanidins (2483 mg/l) being the most abundant compounds.

Considering cardiometabolic biomarkers, Pokimica et al. (2019)^{(Reference Rahmani, Clark and Kord Varkaneh157)} concluded that a 4-week intake of 100 ml/d of chokeberry juice could not be linked with a reduction in systolic and diastolic BP in individuals at cardiovascular risk. An acute study investigating the effect of chokeberry juice (three portions consumed at 1-h intervals, for a total of 600 ml) on BP did not yield significant improvements in young adults^{(Reference Sinclair, Bottoms and Dillon159)}. Differently, Loo et al. (2016)^{(Reference Krikorian, Shidler and Nash160)} found that 8-week consumption of 300 ml/d of chokeberry juice mixed with chokeberry powder decreased daytime ambulatory diastolic BP in patients with untreated mild hypertension. No conclusive results can be drawn from this information.

In the case of the lipid profile, chokeberry juice containing either low or high doses of polyphenols did not significantly change TC and LDL-C compared with the placebo drink^{(Reference Rahmani, Clark and Kord Varkaneh157)}. Similarly, Loo et al. (2016)^{(Reference Krikorian, Shidler and Nash160)} reported no effect of chokeberry juice on lipoproteins and TG. Indeed, although some positive effects on HDL-C have been reported for aronia products, benefits occurred in works carried out with supplements^{(Reference Guo, Zhong and Liu161)}. No significant impacts of chokeberry consumption on serum glucose concentration have been reported^{(Reference Rahmani, Clark and Kord Varkaneh157,Reference Sinclair, Bottoms and Dillon159,Reference Krikorian, Shidler and Nash160)} .

A reduction in the concentration of the inflammation biomarkers IL-10 and TNF-α and a downward trend for IL-4 and IL-5 have been reported^{(Reference Stote, Sweeney and Kean158)}. However, no significant effects on the serum levels of the other cytokines (IL-6, IL-7, IL-8, IL-13) and hs-CRP levels were observed in this study^{(Reference Stote, Sweeney and Kean158)}. In general, the meta-analysis by Rahmani et al. (2019)^{(Reference Guo, Zhong and Liu161)} did not show beneficial effects of aronia products on inflammatory markers.

In conclusion, chokeberry juice showed only limited benefits on HDL-C and systolic BP. Further studies are fully needed to increase the limited body of evidence available to date.

Cherry juice

Cherries can be classified into sweet cherries (Prunus avium L.) and tart cherries (Prunus cerasus L.). They are rich in vitamin C, potassium and phenolic compounds, and especially rich in anthocyanins such as cyanidin and peonidin derivatives, with notable amounts of hydroxycinnamic acids and flavan-3-ols. Generally, tart cherries show higher concentrations of phenolics than sweet cherries^{(Reference Johnson, Navaei and Pourafshar177)} and evidence of the health effects of cherry juices comes predominantly from tart cherries. Many studies used tart cherry juice from the ‘Montmorency’ cultivar, which shows high amounts of anthocyanins^{(Reference Hillman, Trickett and Brodsky178)}. Melatonin is also present in cherries, and some sleep regulation-related biological activities have been attributed to this compound^{(Reference Johnson, Navaei and Pourafshar177)}. Here, the available literature on cherry juice is discussed even when no 100% chery juices, but only diluted ones, were provided to volunteers, as the number of intervention studies providing pure cherry juice is quite scarce, and the amounts of (poly)phenols provided by these diluted ones are quite high and might lead to potential beneficial effects (Table 8).

Table 8. Characteristics of some representative studies investigating the health effects of cherry (sweet and tart) juices

Juices were 100% juice unless otherwise stated. Abbreviations: 4HNE, 4-hydroxynonenal; 8-OHdG, 8-hydroxydeoxyguanosine; BP, blood pressure; BW, body weight; CMJ, countermovement jump-height; CO, crossover; CRP, C-reactive protein; DB, double-blind; DBP, diastolic blood pressure; F, female; FG, fasting blood glucose; FRAP, ferric reducing ability of plasma; HDL-c, HDL-cholesterol; HOMA-B%, homeostasis model assessment-beta cell function; HOMA-IR, homeostasis model assessment-insulin resistance; hs-CRP, high-sensitivity CRP; IL-6, interleukin-6; IL-10, interleukin-10; LDL-c, LDL-cholesterol; M, male; MCP-1, monocyte chemoattractant protein 1; MDA, malondialdehyde; MetS, metabolic syndrome; MS, muscle soreness; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OL, open label; OW, overweight (BMI: 25–30 kg/m²); ox-LDL, oxidised-LDL; PWV, pulse wave velocity; RCT, randomised controlled trial; RSI, reactive strength index; RT, randomised trial; SB, single-blind; SBP, systolic blood pressure; TAG, triglycerides; TC, total cholesterol; TCJ, tart cherry juice; TNF-α, tumour necrosis factor-α; TRAcP 5b, tartrate-resistant acid phosphatase type 5b; VCAM-1, vascular cell adhesion molecule; WHR, waist-to-hip ratio; ↓, decreased level; ↑ increased level

Chai et al. (2019)^{(Reference Chai, Davis and Wright180)}, in a randomised controlled trial, showed no change in body weight or BMI after 12-week supplementation of either 480 ml/d of Montmorency tart cherry juice (68 ml juice concentrate diluted in water, total phenols: 937 mg/l, potassium: 740 mg/l) or control drink in 34 older adults who were overweight. In this line, Johnson et al. ^{(Reference Kent, Charlton and Roodenrys181)} claimed that 480 ml/d of tart cherry juice for 12 weeks in metabolic syndrome patients did not affect body weight or composition. A lack of effect of Montmorency tart cherry juice from concentrate on anthropometric parameters has also been reported recently by two studies in healthy adults^(175,197). No effect on BP was also noted in these studies^(175,197), contrary to previous evidence in at-risk/diseased subjects. Indeed, in a randomised single-blind crossover trial, Desai et al. (2021)^{(Reference Kent, Charlton and Jenner183)} investigated the effects of 130 ml/d consumption of Montmorency tart cherry juice (30 ml juice concentrate diluted in water, 2076 mg/l anthocyanins) on BP in twelve metabolic syndrome patients, showing that 24-h ambulatory systolic, diastolic BP and mean arterial pressure were significantly lower than placebo after 7-d cherry juice consumption. Chai et al. (2018)^{(Reference Lynn, Mathew and Moore184)} reported a reduction in systolic BP (by 4·1 mmHg), but not in diastolic BP, after 12-week consumption of 480 ml/d of Montmorency tart cherry juice in older adults who were overweight. A significant reduction in systolic BP and only a trend (not significant) for diastolic BP reduction was also observed in older adults with dementia, following 12-weeks at 200 ml/d of anthocyanin-rich Bing sweet cherry juice (690 mg/l anthocyanins)^{(Reference Desai, Roberts and Bottoms185)}. Variations in BP upon acute conditions have also been assessed: a single dose of 160 ml of Montmorency tart cherry juice (60 ml juice concentrate diluted in water) significantly lowered systolic BP over a period of 3 h compared with a placebo drink in fifteen men with early hypertension^{(Reference Martin, Burrell and Bopp186)}. Similar results were found when 300 ml of the aforementioned anthocyanin-rich cherry juice was supplied to healthy volunteers^{(Reference Chen, Liu and Chien187)}. Functional improvements in both studies were related to the increase in circulating phenolic metabolites. Conversely, no significant differences in PWV, measured as a predictor of arterial stiffness, were observed in three different interventions^{(Reference Martin, Burrell and Bopp186,Reference Stamp, Chapman and Frampton188,Reference Dodier, Anderson and Bothwell189)} .

Considering other cardiometabolic markers, TC, LDL-C and TC:HDL-C ratio were significantly lower following 7-d consumption of 130 ml/d of Montmorency tart cherry juice compared with the placebo in twelve participants with metabolic syndrome, without changes in TG concentration^{(Reference Kent, Charlton and Jenner183)}. Nevertheless, an acute study by the same authors in the same population setting did not account for changes in blood lipids^{(Reference Lamb, Ranchordas and Johnson190)}, in line with two studies, one conducted in forty-seven healthy adults (30–50 years) with tart cherry juice for 6 weeks^{(Reference Stamp, Chapman and Frampton188)} and another in metabolic syndrome patients for 12 weeks^{(Reference Kent, Charlton and Roodenrys181)}. On the contrary, Chai et al. (2018)^{(Reference Lynn, Mathew and Moore184)} reported that, after the 12-week intervention at 480 ml/d, older adults in the tart cherry juice group had lower LDL-C than the control group. Neither tart cherry juice nor control altered HDL-C concentrations. Data available to date in the case of glucose metabolism is also contradictory. Some authors observed that tart cherry juice did not affect insulin and HOMA-IR levels, while significantly increasing glucose levels in older adults (65–80 years) who were overweight^{(Reference Lynn, Mathew and Moore184)}. Conversely, after 7-d consumption of Montmorency tart cherry juice (130 ml/d) by metabolic syndrome patients, fasting glucose concentrations decreased significantly without major changes in insulin levels^{(Reference Kent, Charlton and Jenner183)}, while only insulin changed when supplementing the same juice in acute conditions to these patients^{(Reference Lamb, Ranchordas and Johnson190)}.

Regarding inflammation, after 12 weeks at 480 ml/d, tart cherry juice significantly lowered CRP levels but not TNF-α levels compared with the control drink in older adults^{(Reference Desai, Roberts and Bottoms179)}. Conversely, Martin et al. (2018)^{(Reference Morehen, Clarke and Batsford191)}, in a randomised crossover study, showed no change in hs-CRP, IL-6, IL-10 and TNF-α levels after 4-week consumption of either 240 ml/d 100% tart cherry juice (total polyphenols: 1827 mg/l) or placebo in ten adults who were overweight/obese. Nevertheless, there was a significant decrease in pro-inflammatory monocyte chemoattractant protein 1 (MCP-1) compared with the placebo group^{(Reference Morehen, Clarke and Batsford191)}. Examining biomarkers of oxidative stress, Chai et al. (2019)^{(Reference Chai, Davis and Wright180)} showed that tart cherry juice significantly increased the DNA repair activity of 8-oxoguanine glycosylase compared with the control drink. In addition, plasma levels of MDA slightly decreased after 12 weeks of tart cherry juice consumption compared with the control drink. Other biomarkers such as 4-hydroxynonenal and 8-hydroxydeoxyguanosine were not affected by either tart cherry or control juice^{(Reference Desai, Roberts and Bottoms179)}. Tart cherry juice supplementation twice daily for 12 weeks reduced ox-LDL and VCAM-1, but not ICAM-1, in adults with metabolic syndrome^{(Reference Kent, Charlton and Roodenrys181)}. Interestingly, serum uric acid, a marker not so commonly assessed in studies evaluating the health properties of fruit juices, has been broadly studied in interventions with cherry juices. Evidence indicates that serum urate decreases consuming tart cherry juice, which may be beneficial for gout patients^{(Reference Abbott, Brashill and Brett192)}, but further studies are needed as recent data from a dose-dependent study in people with gout has pointed to a lack of effect of tart cherry juice in reducing serum urate levels and gout flares^{(Reference Jaiswal, Karaköse and Rühmann193)}.

The potential beneficial effect of cherry juice on cognitive function has also been evaluated. In a randomised controlled trial, Kent et al. (2017)^{(Reference Igwe and Charlton194)} investigated the effect of 200 ml/d of anthocyanin-rich Bing sweet cherry juice in older adults with dementia. After 12 weeks of intervention, improvements in verbal fluency, and short- and long-term memory were found only in the cherry juice group. Improvements in cognitive abilities have also been shown upon consumption of tart cherry juice by older adults (65–80 years)^{(Reference Chai, Davis and Wright180)}. Overall, cherry juice supplementation might improve psychomotor speed, as assessed in a recent meta-analysis^{(Reference Wang, Haskell-Ramsay and Gallegos103)}. Interestingly, the elderly population has also been addressed with regard to testing the effect of Montmorency tart cherry juice on bone metabolism, but no major benefits were observed in post-menopausal women, who were largely osteopenic at baseline (82%), after 90-d juice consumption^{(Reference Santhakumar, Kundur and Fanning195)}.

In the case of exercise performance/recovery, most studies have not found ergogenic effects. Lamb et al. (2019)^{(Reference Bhaswant, Brown and Mathai196)} showed that 500 ml/d of tart cherry juice for 9 d did not enhance recovery in non-resistance trained men after high-force eccentric exercise of the elbow flexors. A lack of effect of cherry juice has also been described for rugby and soccer players^{(Reference do Rosario, Fitzgerald and Broyd197,Reference Igwe, Charlton and Roodenrys198)} . Cherry juice had no effect on the sleep quality of rugby players^{(Reference do Rosario, Fitzgerald and Broyd197,Reference Igwe, Charlton and Roodenrys198)} or healthy adults^{(Reference Keane, George and Constantinou182)} despite the presence of melatonin in cherry juice.

In conclusion, tart cherry juice has attracted much more attention than sweet cherry juice. Cherry juice, in particular tart cherry juice, seems to improve BP and cognitive function, while it may also lead to some benefits at inflammation and oxidative stress level. Some studies have investigated the role of these juices in bone metabolism, exercise performance and gout, with no significant beneficial effects seen. In addition, most of the studies available were carried out using reconstituted concentrate juices, so human subject interventions with 100% juices would be helpful to increase the evidence behind cherry juice.

Plum juice

The term plum refers to a series of Prunus species, namely P. domestica (European plum), P. cerasifera (myrobalan or cherry plum) and P. salicina (Japanese plum). Plums provide phenolic compounds such as chlorogenic acids and other hydroxycinnamates, benzoates, anthocyanins, flavan-3-ol monomers and proanthocyanidins, flavonols, and coumarins. Among thirty-three plum varieties analysed, chlorogenic acids and proanthocyanidins were the major phenolics present in plums, but the qualitative and quantitative phenolic profiles showed high diversity even among closely related cultivars^{(Reference do Rosario, Chang and Spencer199)}. Regarding the health effects of plums (reviewed by Igwe and Charlton, 2016; including fresh, dried plums and juice)^{(Reference Ali, Ali and Sina200)}, most studies used Queen Garnet (QG) plum, an anthocyanin-rich Japanese plum cultivar developed in Australia^{(Reference Etminan, Takkouche and Caamaño-Isorna201)} (Table 9). Evidence on plum juice is provided below.

Table 9. Characteristics of some representative studies investigating the health effects of plum juices

Juices were 100% juice unless otherwise stated. Abbreviations: BP, blood pressure; BW, body weight; CO, crossover; CRP, C-reactive protein; DB, double-blind; DBP, diastolic blood pressure; DROM, derivatives of reactive oxidative metabolites; F, female; FG, fasting blood glucose; FMD, flow-mediated dilation; HDL-c, HDL-cholesterol; HFHE, high fat high energy; hs-CRP, high-sensitivity CRP; IL-1β, interleukin-1β; IL-6, interleukin-6; IL-13, interleukin-13; LDL-c, LDL-cholesterol; M, male; MCI, mild cognitive impairment; MDA, malondialdehyde; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OL, open label; OW, overweight (BMI: 25–30 kg/m²); QG, Queen Garnet; RCT, randomised controlled trial; RT, randomised trial; SB, single-blind; SBP, systolic blood pressure; TAG, triglycerides; TC, total cholesterol; TNF-α, tumour necrosis factor-α; WHR, waist-to-hip ratio; ↓, decreased level; ↑ increased level

In a randomised double-blinded placebo-controlled trial, Bhaswant et al. (2019)^{(Reference Luo, Ke and He202)} found that 250 ml/d of QG plum juice for 12 weeks (1020 mg/l anthocyanins – mainly cyanidin-3-glucoside; 360 mg/l quercetin derivatives) did not change body weight, waist-to-hip ratio and body composition measurements in twenty-nine mildly hypertensive subjects who were overweight or obese. A lack of effect on BMI has also been reported for two QG plum juices containing different anthocyanins levels (48 and 201 mg daily) after supplementation of 250 ml for 8 weeks in older adults with mild cognitive impairment^{(Reference Tierney, Rumble and Billings203)}. Considering BP, QG plum juice 12-week treatment decreased systolic and diastolic BP compared to baseline values and placebo drink (without flavonoids)^{(Reference Luo, Ke and He202)}, while do Rosario et al. (2021)^{(Reference Tierney, Rumble and Billings203)} did not find any effect on BP for any QG plum juice regardless of the anthocyanin amount in older adults after 8 weeks. Under acute conditions, a BP reduction effect was reported for an anthocyanin-rich QG plum juice in both young and older adults^{(Reference Michaličková, Belović and Ilić204)}, while 250 ml of QG plum juice, providing 200 mg anthocyanins and consumed in conjunction with a high fat-high energy meal, did not change the increased post-prandial BP compared with an apricot juice (control juice, no anthocyanins) in sixteen older adults who were overweight^{(Reference Rattanavipanon, Nithiphongwarakul and Sirisuwansith205)}. However, the authors observed that 2-h post-prandial FMD and some parameters of microvascular function were better in the QG plum juice than the apricot juice group.

Modifications of lipid profile and glucose metabolism by plum juice consumption have been scarcely investigated. QG plum juice for 12 weeks decreased fasting plasma LDL-C, but not HDL-C, TC and TG concentration, in mildly hypertensive subjects who were overweight or obese^{(Reference Luo, Ke and He202)}. A three-arm randomised double-blind crossover trial conducted with 200 ml anthocyanin-rich QG plum juice (1010 mg/l anthocyanins: 760 mg/l cyanidin-3-glucoside, 250 mg/l cyanidin-3-rutinoside; 437 mg/l quercetin derivatives), prune juice (no anthocyanins neither quercetin derivatives) and a control drink (matched for energy and macronutrients) for 4 weeks in twenty-one healthy adults did not lead to changes in blood lipids after consumption of any of the juices, which may be related to the adequate physiological conditions of the study population^{(Reference Etminan, Takkouche and Caamaño-Isorna201)}. Similarly, the post-prandial increases in TC and TGs due to a high fat-high energy meal were not altered by either QG plum juice or control juice^{(Reference Rattanavipanon, Nithiphongwarakul and Sirisuwansith205)}. Interestingly, Bhaswant et al. (2019)^{(Reference Luo, Ke and He202)} reported that QG plum juice decreased fasting plasma glucose and insulin compared with baseline and placebo. This effect was not observed when plum juice was tested in healthy adults^{(Reference Etminan, Takkouche and Caamaño-Isorna201)}. Platelet aggregation has also been considered in plum juice research: Santhakumar et al. (2015)^{(Reference Blood, Lowering and Trialists206)} indicated that QG plum juice but not prune juice had a significant effect on different markers of thrombosis, reducing platelet activation and hyper-coagulability.

In relation to inflammatory markers, Bhaswant et al. (2019)^{(Reference Luo, Ke and He202)} found a reduction in TNF-α and plasma interleukins such as IL-6 and IL-13 after QG plum juice in mildly hypertensive subjects who were overweight or obesity. Decreased concentrations of TNF-α were also observed in older adults after 8-week QG consumption^{(Reference Tierney, Rumble and Billings203)}. Conversely, in healthy adults, Santhakumar et al. (2015)^{(Reference Blood, Lowering and Trialists206)} reported that there were no significant changes in inflammatory markers regardless of the treatment (QG plum juice, prune juice or control). Under acute post-prandial conditions, anthocyanin-rich QG plum juice decreased hs-CRP levels compared with the apricot juice^{(Reference Rattanavipanon, Nithiphongwarakul and Sirisuwansith205)}. Do Rosario et al. (2021)^{(Reference Rattanavipanon, Nithiphongwarakul and Sirisuwansith205)} also observed a downtrend for IL-6 while no treatment altered TNF-α and IL-1β. Examining biomarkers of oxidative stress, 200 ml/d QG plum juice for 4 weeks decreased plasma MDA levels, while prune juice did not^{(Reference Etminan, Takkouche and Caamaño-Isorna201)}. A lack of effect of QG plum juice on oxidative stress status under acute conditions has also been reported^{(Reference Rattanavipanon, Nithiphongwarakul and Sirisuwansith205)}. A single dose of QG plum juice did not improve cognitive function in younger or older adults^{(Reference Michaličková, Belović and Ilić204)}.

Few studies have assessed the health effects of plum juice, in particular in the case of common European plum juices. Anthocyanin-rich QG plum juice has attracted almost all the attention in terms of plum juice research, and it might be able to have positive moderate effects on vascular function, LDL-C and inflammatory status. Its ability to attenuate some biomarkers related to cardiovascular risk was seen mainly in subjects at risk of disease, but not in healthy ones. On the other hand, the choice of apricot or prune juices as control drinks^{(Reference Etminan, Takkouche and Caamaño-Isorna201,Reference Rattanavipanon, Nithiphongwarakul and Sirisuwansith205)} could bias the results as both drupe juices share some bioactive compounds with plum juice. Although these juices can be good choices to exclude the role of anthocyanins, they did not allow a complete assessment of the effect of plum juice on health outcomes. Further research on these drupe juices may also be needed.

Tomato juice

Tomato is one of the most popular vegetables worldwide, and its juice is likely the predominant vegetable juice on the market. Lycopene is the main carotenoid in tomatoes and tomato-based products and, among phenolic compounds, quercetin, kaempferol, naringenin, luteolin and caffeic acid derivatives are the most common^{(Reference Silaste, Alfthan and Aro207)}. Tomato has been typically investigated for its lycopene content and has been related epidemiologically to cancer prevention, specifically for prostate cancer^{(Reference Li, Chen and Zhao208)}. However, the only meta-analysis that considered tomato juice for sub-group analysis did not find any significant association between juice consumption and the risk of prostate cancer, so further studies would be needed to clarify the potential chemopreventive effects of tomato juice^{(Reference Saito, Nitta and Imai209)}. Epidemiological studies have also emphasised the potential cardiometabolic benefits associated with tomato consumption, while the contribution of tomato juice is unknown^{(Reference Cámara, Fernández-Ruiz and Sánchez-Mata210)}. The results from key intervention studies with 100% tomato juice are discussed below and are focused mainly on blood lipids, inflammatory markers and oxidative stress status (Table 10).

Table 10. Characteristics of some representative studies investigating the health effects of tomato juice

Juices were 100% juice unless otherwise stated. Abbreviations: BP, blood pressure; CO, crossover; CRP, C-reactive protein; CVD, cardiovascular disease; CXCL10, CXC motif chemokine 10; DB, double-blind; F, female; FG, fasting blood glucose; GAE, gallic acid equivalent; HTD, high tomato diet; ICAM-1, intercellular adhesion molecule 1; IFN-γ, interferon-gamma; IL-6, interleukin-6; IL-8, interleukin-8; LDL-c, LDL-cholesterol; LTD, low tomato diet; M, male; MDA, malondialdehyde; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OL, open label; OW, overweight (BMI: 25–30 kg/m²); PON1, paraoxonase-1; PT, prothrombin time; RCT, randomised controlled trial; RT, randomised trial; SB, single-blind; T2D, type 2 diabetes; TAC, total antioxidant capacity; TC, total cholesterol; VCAM-1, vascular cell adhesion molecule; ↓, decreased level; ↑ increased level

Michaličková et al. (2019)⁽²¹¹⁾ assessed the effects of daily ingestion of tomato juice enriched in polyphenols using a tomato extract against a standard tomato juice on BP in subjects with stage one hypertension. Juice composition was similar in both products (16 mg/l lycopene) except for the phenolic content (486 versus 721 mg/l for the control juice and the enriched one, respectively). BP did not change significantly for any treatment after 4 weeks, although 5–10 mmHg reductions were reported as a consequence of both treatments. Although pooled data from a meta-analysis have demonstrated similar outcomes for tomato products^{(Reference Colmán-Martínez, Martínez-Huélamo and Valderas-Martínez212)}, the limited evidence of tomato juice effects on BP precludes from drawing robust conclusions. Nevertheless, a 5-mmHg lowering in BP could represent a reduction in the risk of cardiovascular events by about 10%, so this topic should be better explored^{(Reference Ghavipour, Saedisomeolia and Djalali213)}.

Considering the lipid profile, Silaste et al. (2007)^{(Reference Upritchard, Sutherland and Mann214)} observed a reduction in TC and LDL-C concentrations in healthy, normocholesterolaemic adults after a 3-week high-tomato diet (400 of ml/d tomato juice and 30 mg/d of tomato ketchup) compared with the 3-week low-tomato diet (no tomato products, or fruit and vegetables containing lycopene). The high-tomato diet provided approximately 27 mg lycopene/d, of which 23·6 mg lycopene per 400 ml was from juice and 3·7 mg lycopene per 30 mg was from ketchup, and blood lipid improvements were correlated to changes in serum concentrations of lycopene, β-carotene and γ-carotene. TC and LDL-C also decreased in the work by Michaličková et al. (2019)⁽²¹¹⁾, but only in the control group (tomato juice with no added polyphenols). In the case of glucose metabolism, this last work did not record differences between groups for fasting glucose after 3 weeks⁽²¹¹⁾, in line with pooled data for tomato products^{(Reference Ghavipour, Sotoudeh and Ghorbani215)}. However, under acute conditions, consuming 200 g of tomato juice 30 min before a carbohydrate-rich challenge ameliorated the post-prandial glucose response^{(Reference Briviba, Schnäbele and Rechkemmer216)}.

Platelet hyperaggregability is among the factors associated with CVD risk. A recent review of human subject intervention studies by Cámara et al. (2020)^{(Reference Bub, Barth and Watzl217)} concluded that consuming tomatoes and tomato products is a promising nutritional strategy for the prevention of platelet aggregation. Indeed, the European Food Safety Authority (EFSA) has assessed positively the beneficial effects of a water-soluble tomato concentrate in platelet aggregation^{(Reference Sharma, Karki and Thakur218)}. Nevertheless, the information related to tomato juice is scarce and, for instance, Michaličková et al. (2019)⁽²¹¹⁾ did not observe differences in prothrombin time between treatments.

In a randomised controlled crossover trial, Colmán-Martínez et al. (2017)^{(Reference Potter, Foroudi and Stamatikos219)} found that, in subjects at high cardiovascular risk after 4-week consumption of tomato juice at 200 ml/d or 400 ml/d (401 μmol/l of carotenoids, trans-lycopene and β-carotene accounting for about 48% and 47% of the total carotenoids, respectively; both juices contained 5% olive oil), the concentration of adhesion molecules ICAM-1 and VCAM-1 was significantly lower compared with the control group (water). Other inflammatory biomarkers (IL-8, CRP, eotaxin, interferon-γ, and CXC motif chemokine 10 -CXCL10-) were not significantly different in the intervention group compared with the control group. The lowering effect in inflammatory biomarkers was correlated with the trans-lycopene in circulation, while the other carotenoids in tomato juice showed a minor or no association^{(Reference Potter, Foroudi and Stamatikos219)}. In another study, using IL-6, IL-8, hs-CRP and TNF-α as biomarkers of inflammation, Ghavipour et al. (2013)^{(Reference Watzl, Bub and Briviba220)} found that 330 ml/d of tomato juice (112 mg lycopene/l) for 20 d significantly decreased serum concentrations of IL-8 and TNF-α in subjects who were overweight, while, among subjects with obesity, only serum IL-6 concentration decreased in the intervention group. Conversely, after 4-week consumption of 500 ml/d of tomato juice, no changes in inflammatory biomarkers (CRP, ICAM-1, and VCAM-1) in patients with T2D were observed^{(Reference Ramezani, Tahbaz and Rasouli221)}.

Many studies have investigated the effect of tomato juice on biomarkers of oxidative stress. Ghavipour et al. (2015)^{(Reference Ramezani, Yousefinejad and Javanbakht222)}, in a randomised controlled trial in females who were overweight, found that 330 ml/d of tomato juice for 20 d increased plasma total antioxidant capacity and erythrocyte antioxidant enzymes and decreased serum MDA levels compared with both baseline and the control group. No improvement was observed in subjects with obesity and, as hypothesised by the authors, this may be due to the need for a greater amount of lycopene or a longer duration of lycopene supplementation^{(Reference Ramezani, Yousefinejad and Javanbakht222)}. Upritchard et al. (2000)^{(Reference Ramezani, Tahbaz and Rasouli221)} reported that consumption of 500 ml/d of tomato juice for 4 weeks by T2D patients increased both plasma lycopene levels and the resistance of LDL to oxidation, almost as effectively as supplementation with a high dose of vitamin E (537 mg/d). Increased LDL resistance to oxidation was also observed by Silaste et al. ^{(Reference Upritchard, Sutherland and Mann214)} after 3 weeks of a high-tomato diet in healthy adults (27 mg lycopene/d). Conversely, in the study by Briviba et al. (2004)^{(Reference Zamani, de Joode and Hossein223)}, MDA levels in plasma and faeces and ex vivo LDL oxidation were not affected in healthy men by supplementation for 2 weeks of 330 ml/d of tomato juice (112 and 5 mg/l of lycopene and β-carotene, respectively) in comparison to carrot juice^{(Reference Zamani, de Joode and Hossein223)}. Nonetheless, most studies have found improvements in biomarkers of oxidative stress. In addition, Bub et al. (2002)^{(Reference Vanhatalo, Bailey and Blackwell224)} found that the changes in antioxidant status after tomato juice consumption could be genotype-dependent, precisely related to the PON-1-192 polymorphism. Indeed, their results showed that consumption of 330 ml/d of tomato juice for 8 weeks reduced LDL-oxidation in healthy elderly who were R-allele carriers (QR/RR) but not in the QQ wildtype^{(Reference Vanhatalo, Bailey and Blackwell224)}.

In conclusion, tomato juice may have favourable effects on lipid metabolism and glucose post-prandial response and could improve biomarkers of inflammation and oxidative stress. However, further studies are needed to demonstrate the effect of tomato juice on CVD risk factors and establish dose-response effects taking into account the responsible bioactive compounds. Although most of the evidence points to lycopene, the role of phenolic compounds on the health benefits of 100% tomato juice should not be neglected. Genotypic differences should also be considered.

Carrot juice

Carrot (Daucus carota) is a popular root vegetable and an important source of dietary carotenoids. Besides vitamins and minerals, carrot juice is rich in α- and β-carotene^{(Reference Kapil, Khambata and Robertson225)}. Both carotenes are vitamin A precursors (the pro-vitamin A activity of α- and β-carotene is 50% and 100%, respectively), and β-carotene has attracted much attention during the last decades due to its preventive features in different pathophysiological scenarios. Moreover, carrot juice is rich in caffeic acid, among other (poly)phenols, while black carrot juice may also present a high amount of anthocyanins^{(Reference Kapil, Khambata and Robertson225)}. Carrot juice and blends are important vegetable juices from a market point of view, while the level of evidence on their health properties is low due to the limited number of works published (Table 11).

Table 11. Characteristics of some representative studies investigating the health effects of 100% carrot juice

Juices were 100% juice unless otherwise stated. Abbreviations: Apo A, apolipoprotein A; Apo B, apolipoprotein B; CO, crossover; CRP, C-reactive protein; DB, double-blind; DBP, diastolic blood pressure; F, female; HDL-c, HDL-cholesterol; IL-1α, interleukin-1α; IL-6, interleukin-6; LDL-c, LDL-cholesterol; M, male; MDA, malondialdehyde; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OL, open label; OW, overweight (BMI: 25–30 kg/m²); RCT, randomised controlled trial; RT, randomised trial; SBP, systolic blood pressure; T2D, type 2 diabetes; TAG, triglycerides; TC, total cholesterol; ↓, decreased level; ↑ increased level

Potter et al. (2011)^{(Reference Bondonno, Liu and Croft226)} evaluated the effect of carrot juice on different cardiometabolic risk markers in seventeen individuals with elevated plasma cholesterol and TG levels. Treatment consisted of 470 ml (16 oz) of freshly squeezed carrot juice for 3 months without a control group. Carrot juice did not alter anthropometric parameters, body fat percentage, BP, lipid profile, glucose metabolism markers, and inflammatory markers, while it led to an increase in plasma antioxidant status and a decrease in MDA^{(Reference Bondonno, Liu and Croft226)}. Contrary to these benefits in the oxidative stress balance, Briviba et al. (2004)^{(Reference Zamani, de Joode and Hossein223)} studied lipid peroxidation in plasma and faeces of healthy men consuming a diet low in carotenoids supplemented with 330 ml/d of tomato juice, as previously reported, or carrot juice (82 and 39 mg/l of β-carotene and α-carotene, respectively) for 2 weeks. Carrot juice consumption raised the plasma levels of α- and β-carotene, but this did not lead to improvements in biomarkers of lipid peroxidation^{(Reference Zamani, de Joode and Hossein223)}. In addition, carrot juice did not modulate immune functions in comparison to tomato juice^{(Reference Gilchrist, Winyard and Aizawa227)}.

Ramezani et al. (2010) conducted a randomised controlled double-blind study with 200 ml/d β-carotene-enriched carrot juice (active group) and carrot juice (placebo) for 8 weeks in forty-four patients with T2D^{(Reference Ocampo, Paipilla and Marín228)}. Although serum levels of β-carotene increased, both treatments had no effect on markers of glycaemic homeostasis (glucose and insulin)^{(Reference Ocampo, Paipilla and Marín228)} or inflammation (CRP and IL-6)^{(Reference Bahadoran, Mirmiran and Kabir229)}.

Overall, despite the content of β-carotene and other bioactives, carrot juice does not seem to exert any significant effects on human subject health.

Beetroot juice

Beetroot (Beta vulgaris) juice has been primarily investigated for its high concentration of dietary nitrate (NO₃ ⁻), which is partially converted to NO after consumption and may exert vasodilation-related benefits associated with vascular function, cardiorespiratory endurance and exercise performance. Zamani et al. (2021)^{(Reference Benjamim, Porto and Valenti230)} recently conducted a systematic review of the benefits and risks of beetroot juice consumption in healthy subjects, and it can be useful to deepen the knowledge on beetroot juice (Table 12).

Table 12. Characteristics of some representative studies investigating the health effects of 100% beetroot juice

Juices were 100% juice unless otherwise stated. Abbreviations: BP, blood pressure; CO, crossover; COPD, chronic obstructive pulmonary disease; DB, double-blind; DBP, diastolic blood pressure; F, female; FMD, flow-mediated dilation; M, male; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OL, open label; OW, overweight (BMI: 25–30 kg/m²); ox-LDL, oxidised-LDL; PWV, pulse wave velocity; RCT, randomised controlled trial; RT, randomised trial; SBP, systolic blood pressure; T2D, type 2 diabetes; ↓, decreased level; ↑ increased level

Beetroot juice could help lower systolic and diastolic BP in healthy young adults, likely due to the vasodilatory effects of NO, whereas results in the elderly were inconclusive^{(Reference Benjamim, Porto and Valenti230)}. Several studies have reported a reduction in BP within 3 h after a single dose of beetroot juice, the effect lasting for several hours. For instance, Vanhatalo et al. (2010)^{(Reference Velmurugan, Gan and Rathod231)} observed that 500 ml/d of beetroot juice for 15 d (10·4 mmol nitrate per litre) reduced both systolic and diastolic BP at different time points between 2·5 h and 15 d, suggesting that the effect of dietary nitrate may be maintained over time if supplementation continues. However, further studies investigating the beneficial effect of long-term beetroot juice in healthy people are needed. Regarding otherwise healthy populations, Kapil et al. (2015)^{(Reference Shepherd, Gilchrist and Winyard232)} demonstrated that consumption of dietary nitrate from 250 ml/d of beetroot juice (∼25·6 mmol nitrate per litre) for 4 weeks significantly reduced BP in hypertensive patients with hypertension at trial inception, with no evidence of tachyphylaxis over the 4 weeks, in comparison to a nitrate-free beetroot juice (control). BP was measured as clinic BP, 24-h ambulatory BP and home BP and significant reductions in both systolic and diastolic BP were shown^{(Reference Shepherd, Gilchrist and Winyard232)}. Conversely, despite increased plasma, salivary, and urinary nitrite and nitrate, Bondonno et al. (2015)^{(Reference Shepherd, Wilkerson and Dobson233)} observed no differences in home BP or 24-h ambulatory BP after 1 week of 70 ml of concentrated beetroot juice twice daily (∼50 mmol nitrate per litre) compared with the placebo (nitrate-depleted beetroot juice) in treated hypertensive individuals. Similarly, 250 ml/d of beetroot juice (30 mmol nitrate per litre) for 2 weeks led to an increase in plasma nitrite and nitrate concentration but did not reduce 24-h mean ambulatory BP in T2D patients with antihypertensive therapy, compared with placebo (nitrate-depleted beetroot juice)^{(Reference Silva, Costa and Gomes234)}. It has been hypothesised that antihypertensive medication may limit the potential benefits of the dietary nitrates present in beetroot^{(Reference Berends, Van Den Berg and Guggeis235)}, explaining the absence of significant results in the two previous studies. Nevertheless, pooled evidence accounts for the beneficial effect of beetroot juice on BP reduction, and a greater reduction in both systolic and diastolic BP has been reported after beetroot juice supplementation in at-risk subjects compared with healthy participants^(251,252).

In a randomised double-blind placebo-controlled 6-week study, Velmurugan et al. (2016)^{(Reference Crowe-White, Nagabooshanam and Dudenbostel238)} reported that consumption of 250 ml/d of beetroot juice (∼24 mmol nitrate per litre) significantly increased the FMD response in untreated hypercholesterolemic individuals, with a worsening in the placebo group (nitrate-depleted beetroot juice). Dietary nitrates were thus associated with an improvement in vascular function. Nitrate-rich beetroot juice also led to a small but significant reduction in platelet–monocyte aggregates (a marker of platelet activation) as well as a reduction in P-selectin expression^{(Reference Crowe-White, Nagabooshanam and Dudenbostel238)}. The authors also observed a modest improvement in measures of arterial stiffness (aortic PWV and augmentation index) compared with the control group. In line with these results, Kapil et al. (2015)^{(Reference Shepherd, Gilchrist and Winyard232)} reported an improvement in endothelial function and a reduction in arterial stiffness in hypertensive patients who consumed beetroot juice, with no change after control treatment. However, dietary nitrate from beetroot juice did not improve endothelial function in patients with T2D^{(Reference Silva, Costa and Gomes234)}.

Many studies have evaluated the effects of beetroot juice on exercise and sport performance in healthy subjects^{(Reference Benjamim, Porto and Valenti230)}. Beetroot juice could improve sport performance through several mechanisms, such as reducing oxygen consumption in skeletal muscle and accelerating the transition between anaerobic and aerobic metabolism in muscle cells. In the latter case, the reduced accumulation of metabolites produced during anaerobic respiration (such as lactate) can delay the onset of fatigue and increase power output and force. In addition, due to the vasodilatory effect of NO, beetroot juice could increase muscle and cerebral blood flow^{(Reference Benjamim, Porto and Valenti230)}. Consumption of a single dose of beetroot juice has led to inconclusive results on training performance, although most of the studies in recreationally active or well-trained women suggested positive effects. Short-term beetroot juice consumption (more than one dose daily or multiple days) showed positive effects in recreationally active men (for example, by improving time to exhaustion or recovery), whereas results for well-trained men were inconclusive^{(Reference Benjamim, Porto and Valenti230)}. When taking into consideration subjects with underlying health conditions, Shepherd et al. (2015)^{(Reference Vincellette, Losso and Early239)} found that 4-d 70 ml/d beetroot juice (92 mmol nitrate per litre) did not reduce the O₂ cost of walking in individuals with T2D nor increase the distance covered in the 6-min walk test compared with the control juice (nitrate-depleted), despite plasma nitrate and nitrite concentration increased. The same results were obtained in individuals with chronic obstructive pulmonary disease consuming the same juice (70 ml/d, 97 mmol nitrateper litre) twice a day for 2·5 d, with the final serving 3 h before the activity; in this case, the O₂ cost was measured by a moderate intensity cycling^{(Reference Blohm, Beidler and Rosen240)}. Nevertheless, the research on this topic is continuously evolving, and a more in-depth analysis would be needed to better understand the role of beetroot juice on exercise and sport performance.

Examining biomarkers of oxidative stress, Velmurugan et al. (2016)^{(Reference Crowe-White, Nagabooshanam and Dudenbostel238)} did not find differences in ox-LDL in untreated hypercholesterolemic subjects who consumed 250 ml/d of nitrate-rich beetroot or a control nitrate-depleted beetroot juice. These authors also indicated that beetroot juice could influence microbiota composition. The salivary microbiome was altered after beetroot juice but not after the control, and the shift in the oral microbiome was in favour of organisms capable of nitrate reduction^{(Reference Crowe-White, Nagabooshanam and Dudenbostel238)}. In fact, the effect of beetroot juice on exercise performance may be mediated by oral microbiota^{(Reference Shanely, Nieman and Perkins-Veazie241)}, a topic that deserves further research.

Besides beneficial effects, Zamani et al. (2021)^{(Reference Benjamim, Porto and Valenti230)} also considered the potential risks of consuming beetroot juice. As a source of nitrate, beetroot juice could lead to the formation of potentially carcinogenic N-nitroso compounds. For example, Berends et al. (2019)^{(Reference Martínez-Sánchez, Ramos-Campo and Fernández-Lobato242)} found a significant increase in urinary apparent total N-nitroso compounds after a 70 ml dose of beetroot juice (∼92 mmol nitrate per litre) and a further increase after seven consecutive doses. Thus, although beetroot juice has shown several beneficial effects, further studies should also investigate the link between its intake and the formation of N-nitroso compounds.

In conclusion, dietary nitrate from beetroot juice has been shown to improve BP in healthy individuals or untreated hypertensive subjects, but not in treated hypertensive patients or T2D patients. Improvements in vascular function have also been reported for untreated hypercholesterolemic and hypertensive patients, but not for patients with T2D. In general, pooled results from meta-analyses point to the beneficial effects of beetroot juice on BP and vascular function, whereas the baseline characteristics of the populations seem to be critical to benefit from juice consumption^{(Reference Shanely, Zwetsloot and Jurrissen236)}. The benefits of beetroot juice on exercise performance were seen in some populations, but they depended very much on the dose and type of exercise, among other factors. The effect of beetroot juice on other common cardiometabolic risk factors beyond cardiovascular function has been scarcely investigated and deserves further research. Last, attention has been paid to nitrate, but other beetroot bioactives such as betalains may also play a role and, once again, additional research is needed.

Other juices

Watermelon juice

Watermelon (Citrullus lanatus) is a rich source of the non-essential amino acid L-citrulline, a precursor of L-arginine, which is a substrate for nitric oxide (NO) synthase. Additionally, it is a source of lycopene and other carotenoids^{(Reference Duarte, Milenkovic and Borges243)}. As previously stated, NO is a vasodilator molecule, and it is key to vascular endothelial function. In exercise/sport physiology, NO has received much interest because of its ergogenic effect and, indeed, watermelon juice has been studied primarily related to sport activity.

Shanely et al. (2020)^{(Reference Duarte, Milenkovic and Borges243)} observed that 6-week supplementation of 710 ml/d of 100% watermelon puree (1·87 g L-citrulline per litre, 0·39 g L-arginine per litre, 45 mg lycopene per litre) improved sVCAM-1 levels, a marker connected to atherogenesis, in post-menopausal women who were overweight/obese, whereas fasting blood glucose, insulin and HOMA-IR did not change (Table 13). In another study conducted on post-menopausal women, two daily 360 ml servings of 100% watermelon juice for 4 weeks did not affect BP or arterial stiffness despite an increase in circulating lycopene being recorded^{(Reference Prasertsri, Roengrit and Kanpetta244)}. No effect on inflammation and oxidative stress markers was observed^{(Reference Prasertsri, Roengrit and Kanpetta245)}, while fasting blood glucose slightly increased although changes in glucose homeostasis lacked clinical relevance^{(Reference Prasertsri, Roengrit and Kanpetta244)}. Some post-prandial beneficial effects of watermelon juice have recently been observed in healthy adults⁽²⁴⁶⁾.

Table 13. Characteristics of some representative studies investigating the health effects of other 100% juices like watermelon, wild passionfruit and cashew apple

Juices were 100% juice unless otherwise stated. Abbreviations: BMI, body mass index; BP, blood pressure; CO, crossover; DB, double-blind; DBP, diastolic blood pressure; F, female; FG, fasting blood glucose; FMD, flow-mediated dilation; HDL-c, HDL-cholesterol; HOMA-IR, homeostasis model assessment-insulin resistance; LDL-c, LDL-cholesterol; M, male; NW, normal weight (BMI: 18,5–24,9 kg/m²); OB, obesity (BMI: >30 kg/m²); OL, open label; OW, overweight (BMI: 25–30 kg/m²); PWV, pulse wave velocity; RCT, randomised controlled trial; RT, randomised trial; SBP, systolic blood pressure; TAG, triglycerides; TC, total cholesterol; VCAM-1, vascular cell adhesion molecule; ↓, decreased level; ↑ increased level

Regarding the role of watermelon juice on exercise, Blohm et al. (2020)⁽²⁴⁷⁾ found that a single dose pre-exercise of 355 ml of watermelon juice (2·2 g L-citrullineper litre; 3·1 g potassium per litre) prevented increased post-exercise systolic and diastolic BP in fourteen healthy non-athletic females, but not in thirteen males. It was thus suggested to examine the effect of sex when assessing the efficacy of watermelon juice on BP. Authors found no effect on post-exercise muscle soreness, blood lactate levels or exercise performance⁽²⁴⁷⁾. A lack of effect of watermelon juice on exercise performance in comparison with a placebo carbohydrate beverage was reported when twenty trained cyclists consumed 980 ml/d of watermelon puree for 2 weeks and during 75 km cycling time trials, despite watermelon increased plasma L-citrulline and L-arginine concentrations and total nitrates⁽²⁴⁸⁾. In another randomised crossover study, Martínez-Sánchez et al. (2017)⁽²⁴⁹⁾ reported a lower muscle soreness perception from 24 to 72 h after a half marathon race and lower plasma lactate concentrations in amateur runners who consumed 500 ml L-citrulline-enriched watermelon juice (6·91 g L-citrulline per litre; 13·98 mg lycopene per litre) 2 h before the marathon race compared to runners in the placebo group (no L-citrulline and lycopene).

Few studies were available on watermelon juice to draw major conclusions. The effect of 100% watermelon juice consumption at cardiometabolic level has been studied mainly in post-menopausal women, and no relevant benefits were reported. Although it has been hypothesised that watermelon juice may have a significant effect on exercise performance, no consistent effects have been seen in studies, perhaps due to heterogeneity in populations and exercise protocols.