Relationship between Carotid-Femoral Pulse Wave Velocity and Diet-Induced Weight Loss

: The global epidemic of obesity has increased over the past two decades, which has been attributed to a lack of physical activity and increased energy intake. Given the fact that obesity increases the risk of cardiovascular diseases, nutritional approaches to prevent cardiovascular diseases often target weight loss. Although many strategies are used to induce weight loss, the most common ones involve either total energy restriction, carbohydrate restriction, or dietary fat restriction. This report addresses the impact of each diet on improving carotid-femoral pulse wave velocity, a technique that is used to measure arterial stiffness—a surrogate marker of cardiovascular disease risk. Fourteen studies were included in the analysis and revealed that diet-induced weight-loss plateaus with increasing duration of the intervention ( p = 0.174, r = 0.455). Weight loss was signiﬁcantly associated with a reduction in cf-PWV ( p = 0.012, r = 0.591). Further, when macronutrient composition was considered, weight loss was achieved through a low-carbohydrate diet ( p = 0.097, r = 0.626), not total energy restriction ( p = 0.660, r = 0.204) or low-fat diet ( p = 0.142, r = 0.975), tended to reduce cf-PWV. These data suggest that weight loss achieved through a low-carbohydrate diet may have a greater impact on PWV over total energy restriction. More studies are needed to test the effect of weight loss achieved through a low-fat diet on cf-PWV. to make calculations because of missing baseline body weight. A quadratic regression analysis was performed to test the relationship between study duration and percent weight loss achieved.


Introduction
The global epidemic of obesity has increased over the past two decades [1], which has been attributed to poor lifestyle, i.e., lack of physical activity and increased energy intake [2]. Epidemiological data suggest that the prevalence of obesity would continue to increase in Australia, Canada, European countries, the United Kingdom, and the United States [3][4][5][6][7][8]. Importantly, obesity and cardiovascular diseases, which were the leading cause of death in Canada and other parts of the world [9,10], often occur concurrently. Indeed, Hingorani et al. suggest that obesity is the main cause of developing cardiovascular diseases [2]. The potential mechanism involves the increased flux of fatty acids and dyslipidemia, which eventually results in the accumulation of lipids in arterial walls, leading to atherosclerosis [11]. Identifying atherosclerosis at an early stage has been a beneficial strategy to prevent the risk of cardiovascular disease-related mortality.
The diagnosis of atherosclerosis at early stages has been challenging. Measuring arterial stiffness appears to be a promising tool to identify atherosclerosis noninvasively during early stages. Although the direct mechanism through which atherosclerosis influences arterial stiffness is unknown, several studies have shown a strong relationship between arterial stiffness and atherosclerosis and, therefore, arterial stiffness has been used as a surrogate marker for atherosclerosis [12][13][14]. Arterial stiffness can be measured using a technique called pulse wave velocity (PWV). PWV can be used to measure peripheral vascular stiffness, using brachial-ankle (ba-PWV), carotid-radial (cr-PWV), femoral-ankle (fa-PWV), and carotid-ankle vascular index (CAVI); and central vascular stiffness, using carotid-femoral (cf-PWV). Although several factors such as blood pressure and body weight Obesities 2022, 2 149 have been shown to affect cf-PWV in children [15], cf-PWV is still considered as a goldstandard measurement of arterial stiffness and has shown to predict mortality in end-stage renal disease patients [16,17].
The primary cause of the obesity-induced increased risk of atherosclerosis-related mortality appears to be excess energy intake, which is why the prevention strategy includes energy restriction. Several dietary interventions have been considered [18], and they primarily focus on either reducing total energy intake or reducing the intake of a particular macronutrient. Reducing total nutrient intake or dietary fats has been a classical approach; however, reducing carbohydrate intake has been relatively new due to the number of metabolic benefits, i.e., reduction in lipids such as triglycerides (TG), total cholesterol (TC), and low-density lipoproteins (LDL), and cardiovascular benefits (arterial stiffness) observed with low or very low-carbohydrate diets [19][20][21][22]. However, limited data have been published to differentiate between the effects of each macronutrient restriction (lowfat or low-carbohydrate diet). Indeed, regardless of the type of macronutrient restricted, all diets appear to induce weight loss effectively [19]. However, whether similar effects are observed with weight loss on cf-PWV or restriction of specific macronutrient-induced weight loss can have differential effects on cf-PWV is unknown. Therefore, the purpose of this brief report was to demonstrate the changes in cf-PWV in response to diet-induced weight loss. Furthermore, the secondary purpose of this report was to identify whether weight loss-induced changes in cf-PWV were related to any particular macronutrient restriction.

Methods
As discussed above, arterial stiffness measured using cf-PWV has been used as a gold-standard technique and has shown to be the strongest predictor of CVD-related mortality [16,17]; only studies that measured arterial or vascular stiffness using cf-PWV published until the year 2021 were included in this brief analysis. Studies utilizing exercise, drugs, or surgical intervention (alone or in combination with dietary intervention) were excluded from the analysis. Review articles were screened to confirm that all the studies were included in the analysis. Studies were searched using the following term "("weight loss" [Title/Abstract]) AND (pwv[Title/Abstract] OR "arterial stiffness" [Title/Abstract] OR "aortic stiffness" [Title/Abstract] OR "vascular stiffness" [Title/Abstract])" in PubMed/NLM. A regression analysis was performed using IBM SPSS (v. 27, Armonk, NY, USA, IBM Corp.) to test the relationship between percent weight loss and study duration using study duration as the independent variable (x-axis) and weight loss as the dependent variable (y-axis). For the relationship between percent weight loss and change in cf-PWV, the percent change in weight loss as an independent variable (reported on the y-axis) and change in cf-PWV as a dependent variable (x-axis).

Results
As shown in Figure 1, 112 studies were identified using the search and, of those, fourteen studies (five studies had two different dietary interventions and were reported separately) were selected for the analysis based on the criteria mentioned above. Information related to studies included in this report is presented in Table 1. Of the fourteen studies, one study was conducted on men (two different arms), three studies were conducted only on women, and eleven studies were conducted on both sexes. All the studies included middle-aged participants with an average age above 35 years who were overweight or obese. Two studies investigated low-fat diet intervention (one study had two arms) [23,24], four studies conducted low-carbohydrate diet intervention [24][25][26][27], four studies were very low-carbohydrate diet intervention [20,24,27,28], one study had two arms (one with low glycemic and one with high glycemic index diet) [29], and the remaining six studies had total energy restriction [30][31][32][33][34][35]. Study duration ranged from as short as two weeks to as long as 52 weeks. Given the limited number of studies available in the literature to identify differences between macronutrient-related effects on cf-PWV, average weight loss from each intervention was recorded in kilograms (kg) and percent weight loss either was reported directly from the study itself or was estimated via calculations. Similarly, change in cf-PWV was recorded from each publication (not shown in table).
Obesities 2022, 3, FOR PEER REVIEW 3 as two weeks to as long as 52 weeks. Given the limited number of studies available in the literature to identify differences between macronutrient-related effects on cf-PWV, average weight loss from each intervention was recorded in kilograms (kg) and percent weight loss either was reported directly from the study itself or was estimated via calculations. Similarly, change in cf-PWV was recorded from each publication (not shown in table).    As expected, shown in Figure 2 is the weight loss over 52 weeks, a trend commonly observed, i.e., the weight loss is greatly achieved with shorter studies; whereas, when the intervention is long, weight loss plateaus (p = 0.067, r = 0.454). The average weight loss with energy-restricted diet was 7.8 kg (7.5%), low-carbohydrate diet was 8.0 kg (7.9%), and low-fat diet was 6.4 kg (6.0%) with no differences between them. The average reduction in cf-PWV with energy-restricted diet was 4.3 cm/s, low-carbohydrate diet was 8.3 cm/s, and low-fat diet was 4.5 cm/s with no significant differences between them. Regression analysis revealed no relationship between change in cf-PWV and change in percent weight loss (p = 0.152, r = 0.363) or absolute weight loss (p = 0.140, r = 0.351) for all studies combined (data not shown). Similar results were observed for studies that restricted total energy intake (percent weight loss: p = 0.978, r = 0.015, absolute weight loss: p = 0.957, r = 0.023), and low-fat diets (percent weight loss: p = 0.434, r = 0.777, absolute weight loss: p = 0.426, r = 0.784, data not shown). However, for studies that utilized carbohydrate-restricted diets, a trend was observed (percent weight loss: p = 0.098, r = 0.625, absolute weight loss: p = 0.101, r = 0.620, data not shown). However, the duration of these studies was highly variable; therefore, to control for the effect of study duration, both the percent weight loss and change in cf-PWV were divided by the number of weeks to obtain the percent change in body weight per week and change in cf-PWV per week. Interestingly, as shown in Figure 3, a significant relationship was observed between percent weight loss per week and change in cf-PWV (black line in Figure 3, p = 0.015, r = 0.579). When the regression analysis was performed for studies that utilized total energy restriction (orange line, p = 0.683, r = 0.215) or low-fat diet (yellow line, p = 0.143, r = 0.975), no relationships were observed. However, when studies that included only low-carbohydrate diets were included, a trend was observed (grey line, p = 0.099, r = 0.623). Lack of significance was likely due to smaller sample size. Similar results were observed with absolute change in weight loss per week and change in cf-PWV per week (data not shown). Two studies were excluded because they did not report percent weight loss / were not able to make calculations because of missing baseline body weight. A quadratic regression analysis was performed to test the relationship between study duration and percent weight loss achieved. Two studies were excluded because they did not report percent weight loss / were not able to make calculations because of missing baseline body weight. A quadratic regression analysis was performed to test the relationship between study duration and percent weight loss achieved. Figure 2. Percent weight loss achieved and duration of dietary intervention. Legend: Each data point represents each study included in the analysis. Two studies were excluded because they did not report percent weight loss / were not able to make calculations because of missing baseline body weight. A quadratic regression analysis was performed to test the relationship between study duration and percent weight loss achieved.

Figure 3.
Relationship between diet-induced percent weight loss and arterial stiffness calculated per week. Legend: Each data point represents each study included in the analysis. Two studies were excluded because they did not report percent weight loss/were not able to make calculations because of missing baseline body weight. The change in cf-PWV was divided by the duration of the study to estimate the cf-PWV per week. Similar calculations were performed for weight loss. Blackline represents all studies included in the analysis. Orange line and dots represent studies with total energy restriction; grey line, and dots represent studies that restricted only carbohydrates, darker grey dots represent moderate restriction and lighter grey dots represent very low-carbohydrate diets. Yellow line and dots represent studies that restricted dietary fat. Abbreviations: cf-PWV, carotid-femoral pulse wave velocity. week. Legend: Each data point represents each study included in the analysis. Two studies were excluded because they did not report percent weight loss/were not able to make calculations because of missing baseline body weight. The change in cf-PWV was divided by the duration of the study to estimate the cf-PWV per week. Similar calculations were performed for weight loss. Blackline represents all studies included in the analysis. Orange line and dots represent studies with total energy restriction; grey line, and dots represent studies that restricted only carbohydrates, darker grey dots represent moderate restriction and lighter grey dots represent very low-carbohydrate diets. Yellow line and dots represent studies that restricted dietary fat. Abbreviations: cf-PWV, carotid-femoral pulse wave velocity.

Discussion
The primary purpose of this report was to identify studies and analyze the changes in cf-PWV and body weight in response to dietary restrictions. The secondary purpose was to investigate the relationship between change in weight loss and change in cf-PWV and whether this effect was related to particular macronutrient restrictions. The analysis from this report showed that the reduction in arterial stiffness is associated with dietinduced weight loss. Wildman et al. conducted a longitudinal study in 152 participants with an average of 30 years, and measured arterial PWV annually [36]. The goal was to observe changes in PWV with changes in body weight observed in the absence of dietary intervention. Interestingly, the increase in PWV was strongly associated with an increase in body weight; similarly, a decrease in body weight was associated with a decrease in PWV [36]. In the present report, similar findings were observed when weight loss was achieved through dietary intervention. These data suggest that reduction in arterial stiffness is primarily associated with weight loss alone. In contrast, in a study conducted by Desamericq et al., no relationship was observed between BMI and cf-PWV [37]. In this study, the authors measured cf-PWV in 2304 patients referred for ambulatory blood pressure monitoring. No differences were observed in cf-PWV with regard to body weight or BMI. However, Desamericq et al. did not consider sex differences that could play a significant role in cf-PWV. Indeed, Zuo et al. conducted a study on 834 participants and reported a relationship between cf-PWV and BMI in females only [38]. With an increment of 10 units of BMI, only females but not males showed a significant increase in PWV. However, in the present study, sex differences were not evaluated due to a lack of studies on both sexes. Furthermore, the secondary purpose of this study was to identify whether it would make a difference to the cf-PWV if weight loss was achieved through different dietary interventions (i.e., total energy restriction, low-fat diet, or low-carbohydrate diet). Indeed, when total energy-restricted studies were included in the analysis, no differences were observed even though weight loss was achieved. In contrast, when low-carbohydrate diets were utilized to achieve weight loss, cf-PWV tended to relate to weight loss. The lack of significance was likely due to the smaller sample size. These findings suggest that low-carbohydrate diet-induced weight loss may have a greater impact on improving cf-PWV compared to weight loss achieved through total energy restriction. Given that low-carbohydrate diets can reduce systolic blood pressure [39]; systolic blood pressure has been previously associated with PWV [15]. The improvement in cf-PWV observed with a low-carbohydrate diet could be either a consequence of or the product of decreased cf-PWV. The other major mechanism through which the low-carbohydrate diets may affect cf-PWV is likely through an established insulin resistance pathway. Further, individuals with Type 2 diabetes (who are hyperglycemic and insulin resistant) have exhibited increased cf-PWV compared to non-diabetic individuals [40,41]. Insulin resistance is a common feature of obesity and has been shown to induce pathological vascular stiffening [42]. Indeed, in a study conducted by Grunewald et al. [43], mice fed with high sucrose and high fructose diets resulted in increased body weight. The diet-induced obesity resulted in the increased insulin resistance of the vasculature and worsening of arterial stiffness through endothelial dysfunction. In contrast, when carbohydrates were restricted in the diet of patients taking statins, insulin sensitivity was improved as well as vascular endothelial function (a marker of vascular stiffness) [44]. This is likely due to the nature of the low-carbohydrates diets that improve glycemic control and insulin sensitivity-factors that may eventually lead to improved arterial stiffness [45].
With regard to the low-fat diet, only two studies were found that met the inclusion criteria, and, therefore, it is difficult to speculate the effects of low-fat diet on cf-PWV. Findings from the report present new insights into weight loss-induced improvements in cf-PWV. Despite the fact that this report indicated a stronger relationship between lowcarbohydrate diet-induced weight loss and cf-PWV compared to total energy-restricted dietinduced weight loss, more studies should be conducted to establish a definitive conclusion.
One of the limitations of this study was that only fourteen studies were included in the analysis due to stringent inclusion criteria. The goal was to identify if diet-induced weight loss changes cf-PWV, and if these changes are associated with particular macronutrient restriction. Therefore, studies including exercise, drugs, or in combination with diet were excluded. Lastly, despite other valid methods of measuring PWV are available, only studies that utilized cf-PWV were included in the analysis. The primary reason was that cf-PWV is considered a gold-standard technique over the other techniques and the secondary reason was to be able to compare studies without the effect of confounding factors.
In summary, the analysis from this report revealed a trend with diet-induced weight loss and the duration of the intervention; that was expected. The novel findings from this report presents a relationship between weight loss achieved though low-carbohydrate diets and cf-PWV which was not observed when weight loss was achieved though total energy intake. Mechanistic studies are needed to identify the mechanism through which dietary carbohydrate restriction improves arterial stiffness. Data Availability Statement: Data will be available from the corresponding author upon reasonable request.