Adherence to the Mediterranean diet (Dietmed) exerts protective effects on cardiovascular disease (CVD). In the Lower Extremity Peripheral Arterial Disease (PAD) there are fewer studies that analyze these data.
ObjectiveTo determine adherence to Dietmed and dietary habits in patients with PAD, according to a history of CVD (coronary and/or cerebral ischaemic pathology) and according to the ankle-brachial index (ABI ≥ or < 0,5).
Material and methodsCross-sectional analytical study carried out in a tertiary hospital. The sample was collected consecutively. Sociodemographic and clinical history, ankle-brachial index (ABI) and a 14-point Dietmed adherence dietary questionnaire were included. The analysis of categorical variables was carried out using the Pearson’s Chi-Square test, the T-Student’s statistic test for independent samples was used for parametric variables and the U. Mann-Whitney test for non-parametric variables.
ResultsOf the 97 patients, 87,6% had low adherence to Dietmed, with no differences according to the severity of PAD. However, when we analysed the data according to whether or not they had a history of CVD, we observed a high adherence to some items included in Dietmed, specifically, in the CVD group, the consumption of lean meat (95,5% vs 64%; p = 0.004). In addition, we observed a significant difference in the consumption in the group without a history of CVD (32% vs 9,1%; p = 0.033).
ConclusionIn our population, patients with PAD, regardless of the stage of the disease and whether they had associated coronary or cerebral ischaemic pathology, had low adherence to Dietmed. Therefore, it is important to implement nutritional education programmes in patients with PAD in all stages, as well as in those patients who have already suffered a vascular event, so that they maintain adherence to healthy dietary habits in the long term.
La adherencia a la dieta mediterránea (Dietmed) ejerce efectos protectores sobre la enfermedad cardiovascular (ECV). En la Enfermedad Arterial Periférica (EAP) de miembros inferiores (MMII) existen menos estudios que analicen estos datos.
ObjetivoDeterminar la adherencia a la Dietmed y los hábitos dietéticos en pacientes con EAP, según antecedentes de ECV (patología isquémica coronaria y/o cerebral) y según cifras en el índice tobillo-brazo (ITB ≥ ó < 0,5).
MetodologíaEstudio transversal analítico realizado en un hospital terciario. La recogida de la muestra se realizó consecutivamente. Se incluyeron los antecedentes sociodemográficos, clínicos, ITB y se cumplimentó un cuestionario dietético de adherencia a la Dietmed de 14 ítems. El análisis de las variables categóricas se llevó a cabo mediante la prueba Chi-Cuadrado de Pearson, se usó la prueba estadística T-Student para muestras independientes para variables paramétricas y la prueba de U. Mann-Whitney para las no paramétricas.
ResultadosDe los 97 pacientes, el 87,6% presentaban una baja adherencia a la Dietmed, sin diferencias en función de la severidad de la EAP. Sin embargo, cuando analizamos los datos en base a si presentaban o no asociados antecedentes de ECV, observamos una alta adherencia a algunos ítems incluidos en Dietmed, concretamente, en el grupo con ECV, el consumo de carne magra (95,5% frente 64%; p = 0,004). Además, observamos una diferencia significativa en el consumo de frutos secos, presentando una mayor frecuencia de consumo en el grupo sin antecedentes de ECV (32% vs 9,1%; p = 0,033)
ConclusiónEn nuestra población, los pacientes con EAP de MMII, independientemente del estadio de la enfermedad y de si tenían asociada patología isquémica coronaria o cerebral, presentaban una baja adherencia a la Dietmed. Por lo que, es importante implementar programas de educación nutricional en pacientes con EAP en MMII en todos sus estadios, así como en aquellos pacientes que ya han sufrido un evento vascular, para que mantengan la adherencia a hábitos dietéticos saludables a largo plazo.
Peripheral arterial disease (PAD) is the most common clinical entity of atherosclerotic aetiology, affecting non-coronary arteries, most frequently the arteries of the lower limbs (LL).1 The most frequent form of manifestation of PAD or chronic ischaemia in the lower limbs is usually asymptomatic,2,3 which makes its early diagnosis difficult, and as a consequence, it is underdiagnosed, under recognised and undertreated.2 Between 20% and 59% of patients with objectively proven PAD do not report symptoms in the lower limbs and the prevalence of asymptomatic PAD varies depending on whether patients are recruited in a primary or community care setting (lower%) or in a vascular laboratory (higher%).4 Its diagnosis is confirmed when the Ankle-brachial index (ABI) is less than .93,5,6 ABI values between .41 and .9 are considered mild to moderate lower limb ischaemia,5 and ABI values below .5 (ABI < .5) indicate severe ischaemia2 or greater disease progression. Patients with peripheral arterial disease have higher cardiovascular (CV) and cerebrovascular morbidity and mortality,2,7 compared to the population without PAD.
Cardiovascular risk factors (CVRF) in patients with PAD are less well controlled compared to patients with cardiovascular or cerebrovascular disease.8 Maintaining a healthy lifestyle is considered essential to mitigate progression and achieve better control of cardiovascular disease (CVD). Changing dietary habits is essential for preventing and progressing atherosclerosis. The Mediterranean diet (MedDiet) is one of the best-known and most studied dietary patterns related to health. It is considered the most effective dietary model and has been tested by numerous investigations, demonstrating its benefits for both the primary and secondary prevention of CVD.9–11 It is considered an additional non-pharmacological strategy to reduce the incidence of atherosclerosis and the progression of vascular complications.12 Furthermore, this dietary model improves the management of other CVRF and comorbidities that affect patients with peripheral arterial disease (poor glycaemic control, high blood pressure, and weight control) and incorporates many of the nutritional recommendations suggested for patients with PAD.12 The MedDiet emphasizes an abundance of plant-based foods, the use of olive oil as the main source of fat, limited consumption of dairy products, moderate consumption of fish, poultry, and wine, low amounts of red meat, and a daily consumption of fresh fruit, vegetables, and greens.9 The main mechanisms that explain the CV protection of the MedDiet are related to the effects of specific dietary nutrients, either by decreasing endothelial activation or by improving endothelium-dependent vasodilation,8 thereby minimising the harmful consequences of maladaptive and deregulated inflammation in atherosclerosis by restoring the balance between proinflammatory mediators and anti-inflammatory responses.12
Most studies that include a Mediterranean-type dietary pattern or habit contribute to a reduced risk of CV events; however, they focus on CVD, such as myocardial infarction or ischaemic stroke.9–11 Compared to patients with coronary artery or cerebrovascular disease, CVRFs, which contribute to the development of PAD, are less well controlled or are less well managed.8,13 There is a paucity of clinical research demonstrating the contribution of the MedDiet to PAD. No studies have been found that relate dietary adherence to the MedDiet with the degree or severity of atherosclerotic disease in the lower limbs, based on ABI values. Therefore, in this study, the objective was to determine adherence to the MedDiet and dietary habits in patients with PAD of the lower limbs. This study analysed whether patients exhibited different eating habits depending on: 1) whether they had a history of CVD (ischaemic heart disease or ischaemic brain disease) and 2) their ABI values, i.e., whether the ABI was greater or less than .5 (less advanced or more advanced disease, respectively). In other words, we sought to determine whether dietary habits in patients who had previously suffered from CVD and who had therefore received nutritional recommendations, were healthier than in those without a history of CVD. Furthermore, we also investigated whether patients with milder or less advanced PAD (ABI ≥ .5) had better dietary habits than those with more advanced PAD (ABI < .5), to assess any possible contribution of dietary behaviours to the development of PAD.
Materials and methodsDesign, scope and study populationThis is a cross-sectional analytical study conducted in a hospital setting (tertiary care hospital). The study project was previously submitted to the Ethics Committee on Drug Research (CEIM) of the same hospital for approval. All examinations and procedures were performed by the principal investigator, thus avoiding interobserver error. The study was conducted without interfering with the functioning of the department involved.
Based on their medical history and physical examination, patients attended the hospital's vascular diagnostic laboratory (VDL) by appointment to undergo any diagnostic examination requested by their vascular surgeon for suspected or monitored PAD. Candidates who met the inclusion criteria were recruited through consecutive sampling over a period of 6 months. The sample size was calculated, and a total of 97 patients were recruited.
After explaining the objective of the study, the patients who voluntarily agreed to participate signed the informed consent form. The following inclusion criteria were considered: men and women aged over 60 years, due to the increased prevalence of this disease in those over this age.13 They had attended scheduled, not emergency consultations, and had been assessed in the VDL. The clinical diagnosis recorded was PAD, peripheral arterial disease, or chronic ischaemia in the lower limbs, and the ABI values included in the study were between .41 and .9. Patients with an ABI under .4 (severe stages of PAD) were excluded, since, to date, the contribution of dietary habits alone in patients with severe PAD (ABI < .4) is still not entirely clear, with the ideal being a combination with other changes, such as stopping smoking or physical exercise.14 Patients with an ABI greater than .91 were also excluded, as were those with trophic lesions (vascular ulcer or diabetic foot) or any lower limb amputation; those who had undergone prior revascularisation of the lower extremity arteries through surgery or percutaneous transluminal angioplasty; chronic kidney disease being treated with or without haemodialysis; patients with liver disease; previous pancreatitis, or gastrointestinal inflammatory diseases of any aetiology; patients with gastrectomy or colostomy; those receiving immunosuppressive medication; those with cognitive impairment; those who had suffered from any previous cancer, or patients who had recently undergone major surgery (within 30 days).
Data collection and information sourcesAn initial interview was conducted. Personal data and medical history were reflected in a spreadsheet, and an identifier was assigned for this purpose in the order in which they were referred. A history of CVD was defined as a history of coronary and/or cerebral atherosclerosis. The following data were included: age, civil status (with or without a partner), educational level (no studies, primary or secondary studies) and economic level (according to pharmaceutical contribution regimen: low-medium and medium-high), CVRF (arterial hypertension, diabetes mellitus, dyslipidaemias), ischaemic heart disease and cerebral ischaemia, lipid-lowering treatment, oral antidiabetics, insulin, main medical diagnosis, physical, anthropometric and biochemical examination (blood glucose (mg/dl), total cholesterol (mg/dl), plasma triglycerides (mg/dl) and C-reactive protein (CRP) (mg/dl)), previous ABI performed in the Vascular Laboratory and data on lifestyle (smoking in the last 6 months, alcohol consumption and physical activity).
A new ABI was performed on both lower extremities to determine the degree of PAD at the time and confirm that they met the criteria for mild-moderate ischaemia (ABI between .41 and .9).5 For its determination, the Consensus Document proposed by March-García et al.15 for the Non-invasive Study of Chronic Lower Limb Ischaemia in the Chapter on Vascular Diagnosis was followed. A calibrated ERKA manual aneroid sphygmomanometer for adults with an accuracy of 2 mmHg was used with a pressure cuff measuring 50 cm long × 14 cm wide. It was replaced by an adapted cuff of larger diameter (60 cm long × 16 cm wide) in patients with obesity. A Hadeco ES-100 handheld bidirectional continuous Doppler ultrasound (8 mHz) device, equipped with ultrasound gel, was used. The patient remained supine for 10 min prior to the examination, and systolic blood pressures were recorded in both arms (brachial artery). Systolic blood pressures were then taken in the anterior tibial and posterior tibial arteries of both ankles. To calculate the ABI for each lower extremity, the highest systolic pressure in the ankle was divided by the highest systolic pressure in the arm, which is normally .9 or slightly greater than 1. ABIs were recorded from both lower extremities, with the extremity with the lowest ABI values being recorded.
Assessment of Mediterranean diet adherence and eating habitsAfter completing the above assessments, participants were administered a quantitative questionnaire validated for the Spanish population16 (Predimed Mediterranean Diet Adherence Screener (MEDAS) and used by the Mediterranean Diet Prevention group (Predimed).10,17 It consists of 14 items and assesses the intake of certain foods, as well as the patient's level of adherence to the MedDiet. For the score, each item was assigned a point if it had a positive connotation regarding adherence to the MedDiet. The total score obtained from the 14 questions determined the level of adherence to the MedDiet. Two categories were established: a total score of less than 9, low adherence, and a score greater than or equal to 9, high adherence.
Data processing and statistical analysisAll data from the personal interview and the physical, anthropometric, and biochemical examinations were collected electronically in a spreadsheet, except for the MEDAS, which was completed on paper and subsequently entered into the database. For data analysis, patients were grouped into two study groups: 1) based on their history of CVD (without CVD: n = 75 and with CVD: n = 22) and 2) based on the degree or severity of PAD in the lower limbs (ABI ≥ .5: n = 83 and ABI < .5: n = 14).
Statistical analysis was performed using IBM SPSS® Statistics v. 25.0 software. Continuous variables were normally distributed using the Kolmogorov-Smirnov goodness-of-fit test with Lilliefors correction. Normality was rejected if the two-sided p < .05 was statistically significant. Homogeneity of variance was assessed using Levene's test. For descriptive analysis of the sample, the mean, standard deviation, and/or 95% confidence interval were calculated for parametric continuous variables, and the median and interquartile range (IQR) for non-parametric variables. Categorical variables were described by calculating the absolute (n) and relative (%) frequencies of each of their categories. Categorical variables were analysed using Pearson's chi-square test. A comparison of means between groups was performed using the Student’s t test for independent samples for variables with a normal distribution (parametric) and homoscedasticity. For variables that did not comply with a normal distribution (non-parametric), the Mann-Whitney U test was used as a hypothesis test. For all the tests described above, a statistical significance level of p<.05 was established.
ResultsA total of 102 patients were recruited in this study. Five of these patients were excluded for not fully meeting the selection criteria. The final sample (n = 97) consisted of 86 men and 11 women with claudication and a clinical diagnosis of PAD, peripheral arterial disease, or chronic ischaemia in the lower limbs. Men represented 88.7% of the total sample.
Regarding the descriptive study, the sociodemographic data, medical history, and lifestyle factors of the sample are described in Table 1. The mean age was 68.4 years (95% CI: 67.1–69.7). All participants were Spanish. It was observed that 77.3% of the entire sample was free of CVD (without ischaemic heart disease or previous atherothrombotic stroke). A greater number of cases with heart disease were recorded than with ischaemic stroke (13.4% vs. 9.3%).
Descriptive analysis of the sociodemographic data, medical history, and lifestyle of the total sample.
| Characteristics | Total (n = 97) |
|---|---|
| Socio-demographic data | |
| Sex: | |
| Men | 86 (88.7) |
| Women | 11 (11.3) |
| Civil status: | |
| Without a partner | 26 (26.8) |
| With a partner | 71 (73.2) |
| Educational level: | |
| No education | 22 (22.7) |
| Primary level education | 45 (46.4) |
| Secondary level education | 30 (30.9) |
| Economic level: | |
| Low-medium | 60 (61.9) |
| Medium-high | 37 (38.1) |
| Medical background | |
| High blood pressure: | |
| No | 34 (35.1) |
| Yes | 63 (64.9) |
| Diabetes mellitus: | |
| No | 55 (56.7) |
| Yes | 42 (43.3) |
| Dyslipidaemias: | |
| No | 35 (36.1) |
| Hypercholesterolemia | 39 (40.2) |
| Hypertriglyceridaemia | 23 (23.7) |
| CVD: | |
| No | 75 (77.3) |
| Ischaemic heart disease | 13 (13.4) |
| Ischaemic stroke due to atherosclerosis | 9 (9.3) |
| Ankle-brachial indices (ABI): | |
| ABI ≥ .5 (mild PAD) | 83 (85.6) |
| ABI < .5 (more advanced PAD) | 14 (14.4) |
| Lifestyle | |
| Smoking: | |
| Never | 7 (7.2) |
| Ex-smoker (> 6 months | 44 (45.4) |
| Active smoker (<10 cigarettes/day) | 20 (20.6) |
| Active smoker (≥10 cigarettes/day) | 26 (26.8) |
| Alcohol consumption: | |
| Never or ex drinker >6 months | 29 (29.9) |
| Active (<2 drinks/week) | 18 (18.6) |
| Active (2−5 drinks/week) | 9 (9.3) |
| Active (≥5 drinks/week | 41 (42.3) |
| Physical exercise by walking: | |
| No | 36 (37.1) |
| Walks < 1 h/day | 30 (30.9) |
| Walks ≥1 h/day | 31 (32.0) |
The results are expressed as absolute (n) and relative (%) frequencies.
ABI values were collected according to the subjects' risk of progression to more severe ischaemia.15 As a result, ABI values greater than or equal to .5 (mild ischaemia, less severe PAD progression) and ABI values less than .5 (moderate ischaemia, greater PAD progression) were grouped, representing 85.6% and 14.4% of subjects, respectively (Table 2). The median ABI result was .60 (IQR: .19).
Descriptive analysis of socio-demographic data, medical history and lifestyle of the total sample studied and according to the degree of severity of PAD.
| Characteristics | Total (n = 97) | ITB ≥ .5 (n = 83) | ITB < .5 (n = 14) |
|---|---|---|---|
| Socio-demographic data | |||
| Sex | |||
| Men | 86 (88.7) | 73 (75.3) | 13 (13.4) |
| Women | 11 (11.3) | 10 (10.3) | 1 (1) |
| Civil status | |||
| Without a partner | 26 (26.8) | 22 (22.7) | 4 (4.1) |
| With a partner | 71 (73.2) | 61 (62.9) | 10 (10.3) |
| Educational level | |||
| No education | 22 (22.7) | 18 (18.6) | 4 (4.1) |
| Primary level education | 45 (46.4) | 39 (40.2) | 6 (6.2) |
| Secondary level education | 30 (30.9) | 26 (26.8) | 4 (4.1) |
| Economic level | |||
| Low-medium | 60 (61.9) | 50 (51.5) | 10 (10.3) |
| Medium-high | 37 (38.1) | 33 (34) | 4 (4.1) |
| Medical background | |||
| High blood pressure | |||
| No | 34 (35.1) | 28 (28.9) | 6 (6.2) |
| Yes | 63 (64.9) | 55 (56.7) | 8 (8.2) |
| Diabetes mellitus | |||
| No | 55 (56.7) | 49 (50.5) | 6 (6.2) |
| Yes | 42 (43.3) | 34 (35.1) | 8 (8.2) |
| Dyslipidaemias | |||
| No | 35 (36.1) | 30 (30.9) | 5 (5.2) |
| Hypercholesterolemia | 39 (40.2) | 34 (35.1) | 5 (5.2) |
| Hypertriglyceridaemia | 23 (23.7) | 19 (19.6) | 4 (4.1) |
| CVD | |||
| No | 75 (77.3) | 66 (68) | 9 (9.3) |
| Ischaemic heart disease | 13 (13.4) | 11 (11.3) | 2 (2.1) |
| Ischaemic stroke due to atherosclerosis | 9 (9.3) | 6 (6.2) | 3 (3.1) |
| Lifestyle | |||
| Smoking | |||
| Never | 7 (7.2) | 6 (6.2) | 1 (1) |
| Ex-smoker (> 6 months | 44 (45.4) | 35 (36.1) | 9 (9.3) |
| Active smoker (<10 cigarettes/day) | 20 (20.6) | 19 (19.6) | 1 (1) |
| Active smoker (≥10 cigarettes/day) | 26 (26.8) | 23 (23.7) | 3 (3.1) |
| Alcohol consumption | |||
| Never or ex drinker >6 months | 29 (29.9) | 22 (22.7) | 7 (7.2) |
| Active (< 2 drinks/week) | 18 (18.6) | 15 (15.5) | 3 (3.1) |
| Active (2−5 drinks/week) | 9 (9.3) | 7 (7.2) | 2 (2.1) |
| Active (≥5 drinks/week | 41 (42.3) | 39 (40.2) | 2 (2.1) |
| Physical exercise by walking | |||
| No | 36 (37.1) | 30 (30.9) | 6 (6.2) |
| Walks <1 h/day | 30 (30.9) | 25 (25.8) | 5 (5.2) |
| Walks ≥1 h/day | 31 (32.0) | 28 (28.9) | 3 (3.1) |
Results are expressed as absolute (n) and relative (%) frequencies.
PAD: peripheral arterial disease; CVD: cardiovascular disease; ABI: ankle-brachial index (ABI≥.5: mild or less advanced PAD; and ABI<.5: more advanced PAD).
Regarding lifestyle, it is worth noting that almost half of the sample had been ex-smokers for more than 6 months (45.4%) and 47.4% were current smokers, of whom 26.8% smoked more than 10 cigarettes per day. Strict adherence to the medical guidelines of walking for one hour per session, including rest periods,18 was observed in 32% of the subjects. Thirty-seven point one per cent of the sample did not walk daily.
All patients diagnosed with PAD in our department were being treated with atorvastatin 40 mg/24 h, and 23.7% were receiving concomitant fenofibrate. 43.3% of the subjects were diabetic, 14% were being treated with insulin and oral antidiabetics, and 29.3% were receiving oral antidiabetics.
After administering the 14-item Mediterranean diet adherence questionnaire (MEDAS-14), the level of adherence to the MedDiet was calculated. Almost the entire study sample (87.6%) had low adherence to the MedDiet (score less than 9). The median score obtained on the administered questionnaire was 6.0 (IQR: 2.5).
Regarding the MEDAS questionnaire, as shown in Table 3, more than half of the respondents used olive oil (OO) as their primary cooking fat, instead of other oils. However, only 13.4% of them exceeded the recommended intake of 4 or more tablespoons (Tbsp) per day, making this the lowest-consumption item in the entire questionnaire. Regarding vegetable consumption, meaning that at least one of the vegetables had to be raw or in a salad, and the other cooked, only 27.8% of the total sample consumed 2 or more servings per day. Only 40.2% of all patients ate more than 3 pieces of fruit per day. Regarding the consumption of legumes, fish/seafood, and nuts, the reported consumption rates (≥ 3 servings/week) were also low. Regarding meat consumption, 76.3% consumed less than one daily serving of red meat, hamburgers, sausages, or cured meats, and 71.1% preferred lean meat (chicken, turkey, or rabbit) over beef, pork, or processed meat. Red wine consumption (≥7 glasses of wine per week) was reported by 26.8% of the total sample.
Descriptive analysis of the MEDAS questionnaire items with a positive response score (+1 point).
| MEDAS components | Positive response (+1 point) | Total (n = 97) |
|---|---|---|
| OO as the main cooking fat | Yes | 66 (68.0) |
| Tbs of OO daily | ≥4 Tbs | 13 (13.4) |
| Daily S of vegetables | ≥2 S | 27 (27.8) |
| Daily pieces of fruit | ≥3 pieces | 39 (40.2) |
| Daily S of red meat, hamburgers, sausages, or cured meats | <1 S | 74 (76.3) |
| Daily S of butter, margarine, or cream | <1S | 90 (92.8) |
| Daily carbonated and/or sugary beverages | <1 drink | 66 (68.0) |
| Glasses of wine per week | ≥7 glasses | 26 (26.8) |
| Weekly S of legumes | ≥3 S | 33 (34.0) |
| Weekly S of fish/seafood | ≥3 S | 25 (25.8) |
| Weekly frequency of commercial pastries consumption | 34 (35.1) | |
| Weekly frequency of nuts consumption | ≥3 times | 26 (27.1) |
| Preferred consumption of chicken, turkey, or rabbit meat | Yes | 69 (71.1) |
| Weekly frequency of consumption of tomato sauce with OO | ≥Twice | 19 (19.6) |
| Total points | 6.0 (5.0−7.5) |
Results are expressed as absolute (n) and relative (%) frequencies. The total questionnaire score is shown as the median (p25-p75).
OO: olive oil; Tbs: tablespoons; MEDAS: MedDiet adherence questionnaire; S: servings.
The questions with a negative connotation were: daily consumption of butter, margarine, or cream, daily consumption of red or processed meat, daily consumption of carbonated and/or sugary beverages, and the number of occasions of consumption of non-homemade commercial pastries. These items were analysed considering the response option indicating consumption of less than one serving per day, thus obtaining a positive score (+1 point) for each item. Commercial pastry consumption (<2 times per week, 35.1%) had the lowest score and, therefore, was the item with the highest consumption frequency.
The descriptive analysis of the biochemical variables is shown in Table 4. The only biochemical variable that met the normal range was total cholesterol.
Descriptive analysis of the biochemical parameters of the sample.
| Biochemical parameters | Total (n = 97) |
|---|---|
| Fasting plasma glucose (mg/dL) | 105.0a (93.0−129.1) |
| Total cholesterol (mg/dL) | 176.9 (168.9−184.9) |
| Plasma triglycerides (mg/dL) | 127.7a (90.9−176.6) |
| C-reactive protein (mg/dL) | 2.9a (.7−5.8) |
Parametric continuous variables are expressed as mean (95% CI).
95% CI: 95% confidence interval.
Regarding the bivariate analysis, we analysed whether there were differences in adherence to the items included in the MedDiet between patients with PAD in the lower limbs who had a history of coronary and/or cerebral atherosclerosis versus patients without this history (Table 5). It was observed that patients with a history of CVD and PAD in the lower limbs had higher consumption frequencies of wine, legumes, fish, and shellfish. Moreover, more than 95% preferred to consume lean meat (chicken, turkey or rabbit), compared to 64% of the group without a history of CVD and PAD in the lower limbs, this difference being statistically significant (p = .004; OR: 11.8; 95% CI: 1.5–92.97). However, we observed that they had not incorporated other items from the MedDiet, mainly highlighting the low frequency of nut consumption (less than 3 servings per week of nuts), registering 9.1 vs. 32% in the group without a history of CVD, this difference being statistically significant (p = .033; OR: .2; 95% CI: .46−.9).
Bivariate Chi-square analysis of the MEDAS questionnaire items according to history of CVD (without and with CVD) and according to the degree of severity of PAD (ABI ≥ .5 and <.5).
| MEDAS | Without CVD (n = 75) | With CVD (n = 22) | p-value | OR (95% CI) | ABI ≥.5 (n = 83) | ABI <.5 (n = 14) | p-value | OR (95% CI) |
|---|---|---|---|---|---|---|---|---|
| OO as the main cooking fat | 52 (69.3) | 14 (63.6) | .614 | .7 (.2−2.0) | 56 (67.5) | 10 (71.4) | 1.0* | .8 (.2−2.8) |
| Calculators of OO daily | 10 (13.3) | 3 (13.6) | 1.0* | 1.0 (.25−4.1) | 10 (12.0) | 3 (21.4) | .597* | .5 (.1−2.1) |
| Daily S of vegetables | 21 (28.0) | 6 (27.3) | .947 | .9 (.3−2.7) | 25 (30.1) | 2 (14.3) | .368* | 2.5 (.5−12.4) |
| Daily pieces of fruit | 31 (41.3) | 8 (36.4) | .676 | .8 (.3−2.1) | 34 (41.0) | 5 (35.7) | .711 | 1.2 (.3−4.0) |
| Daily S of red meat. Hamburgers, Sausages,or cured meats | 57 (76.0) | 17 (77.3) | .902 | .9 (.3−2.8) | 63 (75.9) | 11 (78.6) | 1.0* | 1.1 (.2−4.5) |
| Daily S of butter, margarine, or cream | 71 (94.7) | 19 (86.4) | .393* | 2.8 (.5−13.6) | 76 (91.6) | 14 (100.0) | .569* | .8 (.7−.9) |
| Daily carbonated and/or sugary beverages | 53 (70.7) | 13 (59.1) | .306 | 1.6 (.6−4.4) | 57 (68.7) | 9 (64.3) | .987* | .8 (.2−2.6) |
| Glasses of wine per week | 18 (24.0) | 8 (36.4) | .250 | 1.8 (.6−5.0) | 23 (27.7) | 3 (21.4) | .869* | 1.4 (.3−5.4) |
| Weekly S of legumes | 23 (30.7) | 10 (45.5) | .198 | 1.8 (.7−4.9) | 27 (32.5) | 6 (42.9) | .653* | .6 (.2−2.0) |
| Weekly S of fish/seafood | 18 (24.0) | 7 (31.8) | .461 | 1.4 (.5−4.1) | 22 (26.5) | 3 (21.4) | .943* | 1.3 (.3−5.1) |
| Weekly frequency of commercial pastries consumption | 26 (34.7) | 8 (36.4) | .883 | .9 (.3−2.5) | 31 (37.3) | 3 (21.4) | .394* | .4 (.1−1.7) |
| Weekly frequency of nuts consumption | 24 (32.0) | 2 (9.1) | .033 | .2 (.46−.9) | 23 (27.7) | 3 (21.4) | .869* | 1.4 (.3−5.4) |
| Preferred consumption of chicken, turkey, or rabbit | 48 (64.0) | 21 (95.5) | .004 | 11.8 (1.5−92.7) | 58 (69.9) | 11 (78.6) | .730* | .6 (.1−2.4) |
| Weekly frequency of tomato sauce with OO | 15 (20.0) | 4 (18.2) | 1.0* | .8 (.2−3.0) | 18 (21.7) | 1 (7.1) | .366* | 3.6 (.4−29.3) |
The results are expressed as absolute (n) and relative (%) frequencies. Statistical test used: Pearson's chi-square test.
AO: olive oil; CS: tablespoons; CVD: cardiovascular disease; CI: confidence interval; ABI: ankle-brachial index; MEDAS: adherence to the MedDiet questionnaire; OR: odds ratio; S: servings.
To determine whether there were differences in adherence to the MEDAS items among patients with PAD of the lower limbs according to the severity of the disease, a bivariate analysis was performed. In this case, the independent variable was ABI ≥ .5 (less advanced PAD) vs. ABI < .5 (more advanced PAD). It was observed that the consumption of fruits, vegetables, fish, and nuts was lower in subjects with ABI < .5 (more advanced PAD) compared to the group with ABI ≥ .5 (less advanced PAD). However, no statistically significant differences could be demonstrated between the consumption of any food group in the MEDAS questionnaire and the severity of atherosclerotic disease in the lower limbs according to their ABI values.
As shown in Table 6, we analysed whether patients with a history of CVD had greater adherence to the MedDiet and, conversely, whether those with more advanced PAD (ABI < .5) had lower adherence to the MedDiet. Based on the history of CVD, in subjects with CVD, we recorded fewer cases with low adherence to the MedDiet compared to those without known CVD (20.6 vs. 67%), with no differences between groups (p > 0.05). Furthermore, we observed that ABI values were lower than .5, and therefore presented more advanced PAD, in patients with low adherence to the MedDiet compared to those with high adherence (14.4 vs. 0%). However, the difference between adherence to the MedDiet and ABI values also did not reach statistical significance (p > .05).
Bivariate chi-square analysis of the degree of adherence to the MedDiet according to a history of CVD (with and without CVD) and according to the level of severity of CVD (ABI ≥ .5 and <.5).
| Adherence according to groups | Low adherence (score < 9) | High adherence (score ≥ 9) | p-value | OR (95% CI) |
|---|---|---|---|---|
| According to CVD history | .870 | .6 (.1−3.2) | ||
| Without CVD | 65 (67.0) | 10 (10.3) | ||
| With CVD | 20 (20.6) | 2 (2.1) | ||
| According to figures in the ABI | .280 | ― | ||
| ABI ≥ .5 | 71 (73.2) | 12 (12.4) | ||
| ABI < .5 | 14 (14.4) | 0 (0.0) | ||
| Total adherence to MedDiet | 85 (87.6) | 12 (12.4) |
Results are presented as absolute (n) and relative (%) frequencies. P = Pearson's chi-square test. Significant differences were defined as p-values <.05.
CVD: cardiovascular disease; CI: confidence interval; ABI: ankle-brachial index; OR: odds ratio.
Regarding blood biochemical parameters (Table 7), higher and statistically significant levels were found in basal glycaemia in subjects with ABI < .5 (more advanced disease: p = .034) and in plasma triglycerides (p = .000) in the group of patients with a history of CVD.
Bivariate analysis of the biochemical parameters of the sample according to history of CVD (without and with CVD) and according to the degree of severity of PAD (ABI ≥ .5 and <.5).
| Biochemical parameters by group | Without CVD (n = 75) | With CVD (n = 22) | p-value | ABI ≥ .5 (n = 83) | ABI < .5 (n = 14) | p-value |
|---|---|---|---|---|---|---|
| Fasting plasma glucose (mg/dL) | 111.0* (90.0−129.6) | 101.0* (96.2−122.7) | .870 | 103.0* (90.6−125.0) | 121.2* (98.7−153.6) | .034 |
| Total cholesterol (mg/dL) | 179.5 (170.4−188.5) | 168.1 (150.5−185.8) | .239 | 176.0 (167.1−184.8) | 182.3 (162.9−201.8) | .581 |
| Plasma triglycerides (mg/dL) | 111.1* (81.1−155.0) | 175.5* (146.6–211.0) | .000 | 127.7* (91.0−179.0) | 125.0* (86.7−151.2) | .696 |
| C-reactive protein (mg/dL) | 3.1* (.8−5.7) | 1.6* (.5−7.5) | .787 | 3.1* (.8−5.7) | 1.3* (.3−7.7) | .478 |
Parametric continuous variables expressed as mean (95% CI), analysed using the Student’a t test.
PAD: peripheral arterial disease; CVD: cardiovascular disease; ABI: ankle-brachial index.
Dietary modifications are considered essential for improving traditional CVRFs. A large-scale cohort study by Wan et al.,19 attributed poor diet quality in patients with established CVD and/or PAD to an increased risk of new adverse cardiovascular and peripheral events.
Lifestyle plays an important role in the progression of chronic lower limb ischaemia.20–23 Clinical trials derived from the Predimed Study24 and more recent prospective cohorts25 demonstrate the importance of maintaining healthy lifestyles (smoking cessation, physical exercise, moderate alcohol consumption, and adherence to the MedDiet), revealing a lower incidence of PAD with greater adherence to each healthy lifestyle factor.
Regarding dietary habits, it has been shown that an abundance of plant-based foods (typical of a Mediterranean dietary pattern) ensures the intake of polyphenols contained in olive oil, nuts, fruits, vegetables, and legumes. These polyphenols, combined with a low intake of processed foods and red meat, promote a reduction in atherosclerosis, especially in patients with comorbidities.12
The MedDiet is considered an effective dietary strategy for patients with PAD. The results of the Predimed (Prevention with a Mediterranean Diet10 have had a significant impact on the primary prevention of CVD. In this multicentre randomised trial, the incidence of CV events after the follow-up period was lower in the groups assigned to the MedDiet (supplemented with nuts or extra virgin olive oil) compared to the control group (low-fat diet). There is only one randomised clinical trial derived from the Predimed study, with patients without prior PAD,26 evaluating the primary prevention of atherosclerotic disease in the lower limbs. This study showed that the groups assigned to the MedDiet were associated with a lower risk of PAD than the control group.
The benefits of the MedDiet are undeniable, attributed by the Scoditti group27 to "the high content of polyphenols with antioxidant functions found in a diet rich in fruits, vegetables, virgin olive oil, and red wine (resveratrol and quercetin), which reduce inflammatory angiogenesis in endothelial cells." The MedDiet is rich in polyunsaturated fats, due to the use of olive oil, nuts, and seeds. Furthermore, fibre intake from vegetables is included in almost every meal, while red and processed meats are replaced with fish, lean meats, and vegetable proteins.28 The main source of fat in the MedDiet is olive oil, which, thanks to its polyphenolic compounds (oleuropein and hydroxytyrosol), can generate greater attenuation of endothelial damage and improve endothelial regenerative capacity.29 In the present study, the recommendations regarding the amount of olive oil consumption (≥4 Cs per day) were very low: 13.4% of the total subjects, 13.6% in subjects with CVD, and 12% in patients with ABI ≥ .5.
There are studies that demonstrate an inverse association between fruit and vegetable consumption and a lower prevalence of PAD.30 However, other publications do not reveal a significant association between their consumption and a lower risk of PAD.31,32 In the present study, a low frequency of fruit consumption (≥3 pieces/day: 40.2%) and vegetables (≥2 servings/day: 27.8%) was recorded.
In the present study, it was found that 26.8% of the total sample consumed ≥7 glasses of wine per week, with this consumption being slightly higher in patients with a history of CVD and an ABI ≥ .5, but without significant differences. Similarly, previous research has not reported statistically significant differences in consumption between patients with or without PAD.22 Gómez et al.33 found lower wine and alcohol consumption in subjects without premature vascular damage, and neither Ciccarone et al.,22 nor Chen et al.34 showed an association between alcohol consumption and a lower incidence of PAD.
Nuts, especially walnuts, and fish oil are foods rich in omega-3 polyunsaturated fatty acids, the consumption of which has been associated with a decreased risk of developing PAD.35 In the cross-sectional study by Heffron et al.,36 an inverse association was found between daily nut consumption and the prevalence of PAD. Similarly, in our study, the protective effect of this dietary habit was also observed, since patients who had a higher frequency of nut consumption had a lower history of CVD (p = .033; OR: .2; 95% CI: .46−.9) and ABI values ≥.5 (non-significant difference).
Red meat and processed meat consumption has been linked to a higher risk of developing PAD34 and more premature vascular damage.33 In our study, we found a low frequency of red and processed meat consumption (<1 daily serving, in 76.3%) and a greater preference for consuming lean meat (chicken, turkey, or rabbit) instead of beef, pork, or processed meat (71.1% of the total sample). This consumption preference was higher in subjects with CVD (p = .004; OR: 11.8; 95% CI: 1.5–92.7), results that are in line with those found by Lasota et al.,37 who found fewer cases of disease when replacing red or processed meat with fatty fish, although without a statistically significant association ("borderline" values). These differences in the frequency of consumption of certain food groups found in our study between groups could be due to the fact that patients with PAD in the lower limbs and a history of CVD had likely already received nutritional education and, as a result, had incorporated a more heart-healthy diet into their lifestyles. These findings demonstrate that some items from the MEDAS questionnaire are more easily incorporated into the dietary habits of some patients than others.
All these findings allowed us to demonstrate that patients with an ABI ≥ .5 (less advanced PAD) did not have healthier dietary behaviours than patients with an ABI < .5 (more advanced PAD). Cacoub et al.13 demonstrated that CVRFs in patients with PAD are less well controlled than in patients with coronary artery or cerebrovascular disease; therefore, this is one of the reasons why we assumed that those subjects who have suffered a previous CV event would have better dietary habits and lifestyles. However, in our study, a higher frequency of lean meat consumption was only confirmed in subjects with a history of CVD, and of nuts in subjects without CVD. These observations lead us to believe that, in our population in general, there is a lack of dietary and nutritional recommendations implemented in clinical or hospital settings. Similar to Ciccarone et al.,22 we observed that dietary counselling can still be very useful in populations traditionally accustomed to a Mediterranean lifestyle.
There is extensive scientific evidence30,31,35,36,38 showing inverse associations between the intake of specific nutrients (antioxidant vitamins A, C, and E, folate, vitamins B6 and B12, dietary fibre, and omega-3 polyunsaturated fatty acids) and a lower incidence of PAD. Other studies show statistically significant associations between the consumption of saturated fatty acids and a higher risk of PAD22 or lower ABI values (more advanced PAD).38 However, there is increasing data suggesting that a Mediterranean-style diet should be recommended for patients with PAD,14 to improve prognosis and CV morbidity and mortality. However, when separately evaluating the food groups included in the MedDiet, it has been found that there is no protective effect against PAD when analysing each item included in isolation.22 These results suggest that nutritional changes or interventions with isolated foods or specific nutrients alone are not associated with the risk of suffering from PAD, since treatments based on nutritional interventions should not be limited to simple recommendations on which foods should be included in the diet.12 Therefore, we believe that global dietary patterns or total score are more important, probably due to the interactions between all the nutrients that comprise a dietary pattern. The score obtained from the MEDAS-14 questionnaire in our study was 6.0 points (IQR: 2.5). Adherence to the MedDiet in patients with PAD is associated with a reduction in CV risk,22,39 and high adherence (higher scores) in patients with type 2 DM conferred greater protection against PAD in a prospective Italian cohort, obtaining a 56% reduction in the risk of developing PAD.22 Studies with subjects without CVD33 suggest that greater adherence to the MedDiet decreases the likelihood of developing premature vascular aging. However, the Predimed subproject by López-Laguna CVD33 did not demonstrate a significant association between high adherence to the MedDiet and a lower risk of PAD, yet another cross-sectional study, also developed in Spain with the general population, reported a direct association between greater adherence and a higher risk of PAD.21
In our study, although all subjects with more advanced PAD (ABI < .5) had low adherence to the MedDiet, we found no significant differences compared to the group of patients with milder or less advanced PAD, probably due to the low number of patients with ABI values <.5 (14.4%) or the low number of patients in the high-adherence group (12.4%). Furthermore, smoking is the CVRF most strongly associated with the development and progression of PAD and its complications,2 and we believe that the isolated contribution of nutritional habits in patients with severe PAD would have very little impact on the disease. Similarly, although most subjects with low adherence to the MedDiet did not have a previous CVD (67%) and, therefore, may not have received specific nutritional recommendations from the hospital setting, no differences were found compared to those who suffered a previous CV event and, therefore, received nutritional education implemented from the clinical setting, derived from their hospital episode due to their CV event. 43.3% of the sample had diabetes, and therefore, the ABI values of these patients were falsely elevated, due to calcification of the arterial media. Perhaps, for this reason, many of the patients with ABI ≥ .5 (less advanced PAD) are actually < .5 (more advanced PAD), and diabetes probably acts as a confounding factor, making it impossible to clearly see that there are patients with low adherence to the MedDiet whose ABI should be < 0.5 (more advanced PAD).
PAD is considered an atherosclerotic condition strongly associated with cerebrovascular and coronary atherosclerosis. Healthy dietary patterns are a primary approach to increasing control of CVRF and disease development or progression. Therefore, the emphasis should be on improving overall diet quality through a comprehensive nutritional approach, rather than targeting specific foods or nutrients. Recommendations should therefore focus on following a Mediterranean-style diet, which incorporates increased consumption of fruits, vegetables, fish, nuts, legumes, whole grains, lean meat, eggs, and dairy products. For this reason, any change or optimisation in dietary and lifestyle habits in subjects with PAD, however modest, can contribute to a reduction in the symptoms of intermittent claudication and mitigate disease progression.8 Numerous publications reveal the association of the MedDiet with the prevention of PAD,22,26,39 the majority of which are observational studies.40 Only one randomised controlled trial was found in the scientific literature,26 and there are few large-scale studies on dietary patterns and PAD, and few systematic reviews on the topic.14,19,40 The findings were highly heterogeneous in the data and contradictory results.21,31 These studies emphasize the need for further research on the topic.
This study is considered the first to analyse the MedDiet in patients with established PAD, using the ABI test as a quantitative measure, with values within a range to define the disease (ABI < or ≥.5).
Finally, it is worth highlighting the low adherence to the MedDiet in the analysed sample and the high prevalence of subjects with milder PAD (ABI ≥ .5) and without CVD. This situation indicates that there is a great opportunity for improvement to minimise disease progression by including nutritional education within the patient care episode, constituting a valuable tool to improve the quality of life in patients with CVD. Furthermore, the counterproductive effect of underestimating dietary habits has been demonstrated as part of the comprehensive patient management to counteract the global health burden represented by atherosclerotic diseases.12 Diet quality should be considered an important CVRF when defining CV risk, since its impact is significant on the incidence, progression and complications of PAD.12 In practical terms, it would be interesting to invest, as soon as possible, in imparting patients with nutritional education aiming at achieving greater control of CVRFs, especially dietary habits, since the studied sample consisted of many patients free of CVD (77.3%), and whose progression or peripheral vascular complications may depend, in part, on lifestyle habits. It would also be very interesting to integrate supervised exercise programmes into our healthcare institutions as another fundamental pillar of treatment for patients with intermittent claudication. This would likely further reduce our patients' functional disability and limit the progression of PAD.
Regarding the study's limitations, first, the study subjects who attended the hospital were included consecutively. We therefore cannot rule out selection bias affecting the overall conclusions, as well as information bias (memory or attention) during individual interviews. Second, the study was designed as a cross-sectional study. Therefore, although a possible association between variables was demonstrated in the bivariate analysis, it did not allow us to report any causal relationship between the MedDiet and the degree of severity of PAD.
Although our results are consistent with previously published studies,21,36,39 we emphasize the importance of high overall adherence to Mediterranean-style dietary habits with a holistic approach. However, more rigorous, large-scale randomised intervention studies would be needed to demonstrate a causal relationship between variables, assess the contribution of Mediterranean-style dietary factors to disease progression, and thus confirm our findings, providing us with compelling evidence to justify the recommendations we make in daily clinical practice.
ConclusionsBased on our findings, we conclude that, in patients with lower limb PAD, adherence to the MedDiet was low, with low adherence to healthy dietary habits. Patients with a history of CVD reported a higher frequency of lean meat consumption and a lower frequency of nut consumption. Patients with a history of CVD and an ABI < .5 (more advanced PAD) had a worse biochemical profile than those without a history of CVD and an ABI ≥ .5 (less advanced PAD). To improve the prognosis of patients with PAD, more comprehensive monitoring and more effective management of all CVRFs are necessary. As part of comprehensive CV risk management, health systems should urgently integrate a dietary approach through individualised eating habits tailored to the needs and comorbidities of the patient, helping to improve disease control.
The authors have no conflict of interests to declare.
