Introduction

Metabolic acidosis is known to be one of the complications of chronic kidney disease (CKD), and also one of the main factors that contribute to its progression. Numerous studies have shown that acid base status is affected by dietary intake, and additionally may negatively influence progression of CKD [1,2,3].

Complement activation induced by increased ammonia production by remnant nephrons, as well as increased endothelin, angiotensin, aldosterone, and cortisol production are part of the adaptive response that may contribute to ongoing injury of surviving nephrons [4, 5].

Recent studies have demonstrated an association of dietary acid load with incident CKD in a population-based sample [6, 7]. These studies provide evidence that dietary acid load may represent a modifiable risk factor for the development of CKD in otherwise healthy individuals. On the other hand, high acid diets have also been associated with a higher incidence of pathologic conditions implicated in the pathogenesis of CKD, such as hypertension, insulin resistance, risk factors for type 2 diabetes mellitus and for cardiovascular disease in diabetic patients [8,9,10,11,12].

Western diet is characterized by a high content of acid forming elements provided by animal derived foods as compared to alkaline precursors contained in the group of fruits and vegetables [13, 14]. Dietary acid load has been measured by several methods. Frassetto and colleagues proposed the urinary protein/potassium ratio as an indicator of dietary acid base balance [15], while Remer and Manz developed the method for potential renal acid load (PRAL) calculation, which estimates endogenous acid production in excess of base generated by a given amount of food consumed daily [16, 17]. Fish, meats, dairy, and grain products have a high PRAL with acid-forming potential, while fruits and vegetables have a negative PRAL, which reflects base-forming potential.

Most of studies on diet acid load in patients with CKD have been published in adults and only recently the Cardiovascular Comorbidity in Children with Chronic Kidney Disease (4C) Study investigated the deleterious effects of metabolic acidosis in children with CKD [18]. The objective of the present study was to estimate PRAL and acid base status in children with CKD attending the Outpatient Clinic of the Department of Nephrology and the Department of Dietetics and Nutrition of the Children’s Hospital JM de los Ríos from April 2015 to February 2016.

Methods

The study was approved by the institution’s Bioethics Committee. Children’s parents were informed in detail about the objectives of the research. A written consent was obtained from parents or representatives and from children over 8 years of age by means of an informed or assent consent, respectively.

Children were classified with CKD stages 1–4 according to the Kidney Disease Improving Global Outcomes (KDIGO) guidelines for the Evaluation and Management of Chronic Kidney Disease [19]. Acute pathologies that could interfere with normal intake, such as fever or digestive disorders, or chronic pathologies such as diabetes mellitus, heart disease, hypothyroidism, severe malnutrition, or cancer were considered as exclusion criteria. Alkalizing treatment (sodium bicarbonate) dose was registered.

Dietary assessment

Dietary intake was assessed by a 24-h recall (R24H) and a food frequency questionnaire (FFQ) applied to the children’s parents. To improve reporting quality in accurate portion assumption, we used household measures like cups, glasses, plates and spoons. Three-dimensional wooden forms of different sizes were used to estimate food serving sizes. The semi-quantitative FFQ designed ad hoc evaluated the frequency of consumption per food group over a specific period of time (daily, weekly, or biweekly). The FFQ was validated among the Venezuelan population according to the specificities in dietary habits, culture, and food [20]. Quantities of food and beverages consumed, preparation ingredients, cooking methods, and brand names were considered. Nutrient intake was determined according to the Kidney Disease Outcome Quality Initiative (KDOQI) Pediatric Clinical Practice Guidelines for Nutrition in Chronic Renal Failure in its update published in 2009 [21]. Nutritional values that did not appear in the Venezuelan Food Composition Tables [22] were obtained through the nutrition facts data. Cut-off points for nutrient intake adequacy were: <85% low adequacy, 85–115% normal adequacy, and >115% high adequacy [23].

Dietary PRAL assessment

Dietary PRAL was calculated for each type and amount of food consumed as reported in the R24H and then derived from PRAL values established for each type of food. Dietary PRAL was estimated by the amount of acid equivalents (positive values) or alkali equivalents (negative values) in each of the consumed foods based on 100 g of cooked or ready to eat foods.

Anthropometric assessment

Weight and height were obtained according to the guidelines of the International Biological Program [24]. By means of these two measurements, the indicator of body mass index-for-age (BMI-for-age) was constructed and categorized by percentiles. Growth was evaluated according to the anthropometric indicators weight for age (W/A) and height for age (H/A). The reference value was based on the growth package recommended by WHO AnthroPlus®. (licensed by World Health Organization. Av Appia 20. 1211 Geneva 27. Switzerland). BMI data were also compared with reference data of healthy children and adolescents from the Venezuelan population [25]. The categories of BMI-for-age were: excess ≥ p85, normal > p15–<p85, and deficit ≤ p15 [26].

Laboratory tests

Creatinine, serum electrolytes and venous gases were measured. Peripheral venous blood was obtained from any easily accessible peripheral vein. Creatinine was measured by the Jaffe reaction, serum electrolytes by ion-selective electrodes and venous gases with a Radiometer ABL 555 series blood gas analyzer (Radiometer, Copenhagen, Denmark). Ceatinine clearance was estimated according to Schwartz formula [27].

Statistical analysis

Basic descriptive (represented with mean and standard deviation) and bivariate statistics, such as Pearson correlation were applied. Associations between dietary PRAL and energy, macronutrients and daily food servings were established. Student’s t test was applied for comparison between means. A p value < 0.05 was considered significant in all statistical tests.

Results

Twenty-seven patients were evaluated, 14 boys and 13 girls: 18 patients (66.6%) were between 2 and 9 years (mean age 5.62 ± 2.0) and 9 patients (33.4%) were between 10 and 17 years (mean age 13.86 ± 2.3 years). Patient distribution according to CKD stages was as follows: stage 4, 7 patients (25.9%); stage 3, 13 patients (48.2%); stage 2, 4 patients (14.8%); and stage 1, 3 patients (11.1%).

Thirteen patients (48.2%) had normal weight and 13 (48.2%) were below the 10th percentile. Only one boy was overweight (3.6%). Height for age was adequate (percentiles 15–85) in 13 patients (48.2%), and below the 15th percentile in 14 patients (51.8%). BMI-for-age was normal in 14 (53.6%), low in 9 (32.1%), and high in 4 (14.3%) patients.

Dietary intake and nutritional adequacy

Boys had higher nutritional intake, especially in regard to calories and fat, although this difference was not significant (p = 0.5 and 0.18, respectively). Daily energy and macronutrients intake, PRAL and daily servings of basic food groups are specified in Table 1 for both age groups. Significant difference between age groups was evident for protein intake expressed as g/kg/day and for PRAL expressed as mEq/day/1.73m2 (p < 0.0001 and 0.04, respectively).

Table 1 Daily energy, macronutrients intake, PRAL, and daily servings of basic food groups.
Full size table

Diet adequacy for energy, protein, fat, and carbohydrate is specified in Table 2. Energy and protein intake were above recommended range in 14 patients (50%) and 22 patients (78.6%), respectively.

Table 2 Diet adequacy for energy and macronutrients intake.
Full size table

Cereals were one of the most consumed foods among starches. Precooked maize flour, as “arepa” (traditional maize bread), followed by rice and pasta were the most frequent choices among cereals. Plantains were the most consumed food with an alkaline load. Whole cow’s milk and white cheese were consumed daily by 13 patients, while yogurt and yellow cheese were the least consumed foods in this group. Preferred animal derived proteins were chicken meat, followed by turkey ham and eggs. Beef and fish were less frequent choices.

Nineteen patients (70%) reported that they consumed vegetables only as part of the preparation of chicken or ground beef, in very small quantities. At least 12 patients (44.4%) never consumed any type of vegetables. According to the 24-h recall, the average vegetable intake was 0.4 servings per day.

Fruits were consumed mainly as natural juices or shakes: 11 patients (39.3%) reported daily intake and 13 (46.6%), 2–4 times per week. Whole fruits intake was very low: only 5 patients (17.8%) reported eating whole fruits daily, while 17 patients (60%) consumed them 2–4 times per week. The 24-hour intake recall yielded an average intake of 1.5 servings per day, including fruit juices and shakes.

PRAL and dietary pattern

Most patients had a high PRAL (16 ± 10.7 mEq/day and 34.3 ± 19.3 mEq/day/1.73 m2 for the whole group). None of the children had a negative PRAL and 24 of them (88.8%) had a PRAL > 5 mEq/day or > 10 mEq/1.73 m2/day. There were no significant differences between PRAL among CKD stages. PRAL was higher in boys, although this difference was not significant. Correlations between food servings and PRAL are summarized in Table 3. Dairy, meats, fats, and cereals daily servings had a significant positive correlation with PRAL. On the other hand, alkaline foods (fruits and vegetables) had a negative correlation with PRAL values, although this correlation was significant only for vegetables. A significant positive correlation was found between PRAL and energy, protein and fat intake.

Table 3 Pearson coefficient between PRAL (mEq/day) and daily servings and between PRAL (mEq/day/173 m2) and energy and macronutrients intake.
Full size table

Acid base status: Serum pH and bicarbonate were 7.3 ± 0.08 and 20.46 ± 4.5 mEq/L, respectively for all patients. Values distributed according to CKD stages are specified in Table 4. Twenty-one patients (80.7%) were treated with sodium bicarbonate with an average dose of 3.42 mEq/kg/day. Serum sodium and potassium were 142.37 ± 10.6 and 3.89 ± 3 mEq/L, respectively. There was no correlation between PRAL and serum pH or bicarbonate (p = 0.85 and 0.77, respectively).

Table 4 Serum pH and HCO3 according to CKD stage.
Full size table

Discussion

Most of the 27 children included in this study had a dietary acid base imbalance with predominance of acidifying diets, possibly related to high intake of acid-forming foods such as meats and cereals, and very low intake of base-forming foods such as fruits and vegetables. A significant positive correlation was found between PRAL and daily servings of acid precursors containing foods (dairy, meats, and cereals), and also with energy, protein and fat intake. On the other hand, there was a negative correlation with vegetable and fruit intake, although it was only statistically significant with vegetables. These results are in agreement with findings reported in 52 healthy Venezuelan children [28] and in a series of 720 children and adolescents aged 8–18 years published in Germany [29]. The mean PRAL in both studies was positive for all ages and genders, coinciding with the present study, in which, not only mean PRAL was positive, but none of the Venezuelan patients had a negative PRAL underlining the main lack of base-forming foods in their diet.

Protein intake in this series was high when compared to protein requirements recommended for children with CKD. The dietary pattern of our patients, with high protein, fat, and carbohydrate intake, together with low fruit and vegetable intake, is very similar to that reported by local and international population studies in healthy children [30, 31]. Our findings are also similar to those of a previous study in Venezuelan children with CKD [32] and confirm the association of this intake pattern with elevated PRAL previously reported by several investigators in healthy children and adults.

The acidifying effect of this type of diet on net endogenous acid production could be especially deleterious in patients with CKD due to the activation of renal buffering mechanisms that may accelerate progression of the disease towards terminal stages [1,2,3,4]. In this regard, it is noteworthy that recent studies by Goraya and Moe have shown that the most important determinant of dietary protein on the progression of CKD is its quality above its quantity [33, 34].

When it comes to children, metabolic acidosis induced by a diet with an elevated acid load brings along its additional negative effect on growth and nutritional status. CKD is characterized by growth retardation with a progressive decrease in height-z-score during progression of the disease and failure to reach final adult height estimated by the genetic potential [35]. The cause of growth failure in CKD is believed to be multifactorial, but one of the main factors is metabolic acidosis with its negative consequences on bone mineral metabolism, nutritional status, and growth hormone–insulin-like growth factor-1 axis [36].

Almost two-thirds of children in this study had serum pH and bicarbonate under 7.35 and 22 mEq/l, respectively, despite the fact that they had been prescribed sodium bicarbonate as alkalizing treatment. This frequency of metabolic acidosis in this cohort is higher than that reported by the 4C Study which found an average prevalence of 43%, 61% and 45% in CKD stages 2, 3 and 4, respectively [18]. The inadequate control of metabolic acidosis in our patients may be due to the poor compliance with the alkalizing treatment, probably because of very low family income and scarce availability of medications.

The lack of correlation between dietary PRAL and the acid base status could be explained by the various factors that may have modified the patient’s serum pH and bicarbonate, including the etiology and stage of the CKD, as well as the alkalizing treatment they were receiving.

In terms of nutritional status, one third of the patients in our cohort were under the 10th percentile for BMI and nearly half of them were under the 15th percentile for height. These latter results are in agreement with the Report from the Chronic Kidney Disease in Children Study in which poor linear growth was associated with moderate to severe metabolic acidosis [37]. In this collaborative study, children with serum bicarbonate <18 mEq/L had height and weight that were significantly lower than children with serum bicarbonate ≥22 mEq/l.

Several studies in adults with CKD have reported higher morbidity and mortality rates in patients with serum bicarbonate <22 mEq/L [38, 39], as well as the beneficial effects of alkalizing treatment in patients with low plasma HCO3 in slowing decline of renal function [40,41,42,43,44]. The first study to assess the role of metabolic acidosis in the progression of CKD in a pediatric population was the Cardiovascular Comorbidity in Children with Chronic Kidney Disease (4C) Study, which found that serum bicarbonate under 18 mMol/l was associated with a higher risk of CKD progression compared to serum bicarbonate of 22 mMol/l or more [18]. This evidence confirms the importance of an adequate correction of metabolic acidosis in children with CKD in order, not only to optimize growth, but also to minimize the potential for other negative consequences of metabolic acidosis, including bone disease and progressive decline in renal function. In line with this objective, recent studies by several investigators have agreed on the encouraging results obtained with high fruits and vegetables diets as part of the alkalizing treatment of patients with CKD [45, 46]. Additional studies have shown that dietary intervention with alkali-inducing fruits and vegetable can reduce acid load and prevent CKD progression without producing hyperkalemia [47].

Today’s evidence in regard to the renal interstitial damage induced by buffering mechanisms activated by the kidney to increase acid excretion stresses the importance of the maintenance of serum bicarbonate above 22 mEq/l in patients with CKD. Dietary interventions to achieve alkaline loads derived from a higher intake of fruits and vegetables may be an adequate strategy to substitute treatments with alkaline salts like sodium bicarbonate, or at least to lower the dose. This food-based approach has theoretical benefits over supplement-based approaches, due to the other health benefits of fruits and vegetables and to the sodium sparing effect which may facilitate blood pressure control.

Although the low number of patients in this study could make the generalizability of the results difficult, the predominance of high acid diets among Venezuelan children with CKD and the inadequate correction of metabolic acidosis in spite of alkalizing treatment should alert physicians in regard to the importance of a close monitoring of acid base status in children with CKD. Appropriate correction of metabolic acidosis should be achieved by every available media, including alkalizing salts, dietary manipulation, or both.

Dietary acid load assessment should be included as part of the nutritional evaluation of all children with CKD. Future studies should address the impact that dietary interventions to achieve lower acid loads could have on the progression of CKD in children.