In this cross-sectional multicenter study of HD patients, we found that the participants who consumed more frequently dietary fiber sources were older, had a higher educational level, had less time on HD treatment, and were overweight. Also, the frequency of intake of these food groups was not associated with hyperkalemia.
Our participants reported a low frequency of intake of all the food groups included in the assessment. Although they had been counseled to control their intake of specific food groups to treat or decrease hyperkalemia risk, the frequency reported was lower than what is usually recommended. This dietary pattern is similar to results obtained by epidemiological studies performed in our country in the general population and in patients with CKD: low consumption of fruits, vegetables, and whole grains, leading to inadequate fiber intake [15–18].
Though different methods for evaluating and reporting food groups and dietary fiber intake have been used in the literature, it is still possible to make comparisons with our findings. A low intake of these food groups was also found in other HD populations worldwide . In a multinational HD cohort study, the usual intake in servings per week of fruits was (5.5 [1.0–7.0]), vegetables (3.0 [1.0–5.5]), and legumes and nuts (1.0 [0.5–3.0]) . In a study conducted in the United States, only 3% of the study population consumed at least four servings of fruits and vegetables per day .
Patients undergoing HD have very low dietary fiber intake compared to the daily recommendations for adults (25 g/day for women and 38 g/day for men ). In some American cohorts, the daily fiber intake ranged from 10 to 12 g/day, or around 50% lower than in a large American cohort [4, 20]. In Turkey, fiber intake of 12.6 ± 4.7 g/day was reported by 128 participants , while in Brazil, it varied from 9 to 19 g/day [12, 21, 22].
We found associations between the food groups and total scores with demographic, nutritional, and clinical variables. Of note, educational level and BMI influenced the frequency of intake of four out of the five food groups. Participants with higher educational levels consumed fruits, vegetables, whole grains, and seed sources more frequently, reflecting a higher level of information and monetary access since schooling directly influences family income . Overweight participants had a higher score in fruits, vegetables, and whole grain and a lower score in legumes. We did not find other similar studies that assessed this relationship. We speculate that this pattern might be due to the higher overall quantity of food consumed among overweight participants. The lack of association between the frequency of fiber sources intake and hyperkalemia corroborates current evidence that diet is not the main cause of this common disorder in CKD [5, 9, 24]. It is well known that serum potassium is regulated not only by an external balance (relationship among nutrient intake and its excretions) but also by many clinical factors that influence the shift between intra- and extracellular cites, such as diabetes and metabolic acidosis . In a study with a Mexican HD cohort that investigated the factors associated with serum potassium, diabetes was an important hyperkalemia predictor . Nevertheless, we cannot discard the potential bias of the amount of potassium ingested since the present, and most of the mentioned studies report low potassium intakes. Although we have not evaluated potassium intake, the association between dietary potassium and fiber is known  as those nutrients share important food sources, such as fruits, vegetables, legumes, and nuts.
Besides the amount, the bioavailability of potassium in food matrix may also influence its relationship with hyperkalemia. Potassium from plant foods seems to have less impact on potassemia than animal sources or potassium salts . Carbohydrate-containing foods, such as fruits, stimulate insulinemia, which in turn induces the shift of potassium into the intracellular space . At the same time, as sources of fiber, fruits and vegetables favor intestinal transit time and the consequent fecal excretion of potassium [27, 28]. In fact, a multinational cohort of 8078 HD patients found similar serum predialysis potassium levels across the tertiles of fruit and vegetable intake. More importantly, lower consumption of these food groups was associated with higher mortality . Moreover, in a prospective cohort study, lower dietary potassium intake was associated with higher mortality risk, suggesting that excessive dietary potassium restriction may be deleterious in HD patients .
Independent determinants of total score showed that the frequency of intake of dietary fiber sources increased with age, education level, and BMI and was lower among patients on HD treatment for longer periods. In general, a higher intake of dietary fiber sources in older and higher education levels is evidenced in national surveys and HD populations [3, 29, 30]. Regarding BMI, although this parameter can reflect an overall higher quantity of food consumed, this might confer an advantage in terms of healthy nutrient intake and may benefit overweight patients. This aspect has not been discussed among the theories on the reasons behind the survival advantages consistently found in HD patients with increased BMI , but we believe it deserves attention.
Interestingly, HD vintage was also an independent determinant of the frequency of dietary fiber sources. It is currently unknown why patients on HD for shorter periods reported a higher intake of these food groups. Similarly, Saglimbene et al. reported that patients on HD for a shorter time had a higher fruit and vegetable intake . It is possible that patients who had been on treatment for a longer time experienced more hyperkalemia episodes, influencing their food choices.
Some limitations of the study should be mentioned. Causal relationships cannot be established due to the cross-sectional nature of the study. Second, the dietary investigation was qualitative, limiting our ability to quantify the amount of fiber and potassium consumed. Finally, clinical factors associated with potassemia, which could further explain the results, were not captured. However, this exploratory analysis adds to the growing body of observational research on the relationship between diet and hyperkalemia and the importance of promoting a healthy diet among patients in HD.
In conclusion, the usual frequency of fiber source intake was low, and its independent determinants were age, HD vintage, education, and BMI. The lack of association with hyperkalemia suggests that other dietary sources and clinical factors should be considered when managing hyperkalemia in this population.