Cognitive impairment (CI) is chronic and progressive disablement of cognitive functions beyond what is expected in the normal aging process. CI may progress to dementia, and is associated with an increased risk of disability and health care expenditures.1 Chronic kidney disease (CKD) is an increased risk factor for CI and it has been demonstrated that there is a correlation between the severity of CKD and CI.2 Because, CI was detected in about 70% of middle-aged and elderly hemodialysis patients compared to the age-compatible general population.3 The probable causes of CI in this patient population are explained as vascular damage to brain tissue and direct neurotoxic effects of uremic toxins.4 Nowadays, the Montreal Cognitive Assessment (MoCA) can be used as a reliable method in the diagnosis and follow-up of CI in hemodialysis patients.5

l-Carnitine (3-hydroxy-4-N-trimethyl-aminobutyric acid), a low-molecular-weight amino acid derivative, is crucial for the-oxidation of long-chain fatty acids. It is the cofactor that mediates the entry of long-chain fatty acids into the mitochondria during the energy production process, and inhibits the accumulation of toxic fatty acids in the mitochondria in some pathological situations that can cause oxidative stres.6 Kidneys also play an important role in l-carnitine metabolism (expression, secretion, reabsorption).7 Carnitine hemostasis is disrupted in patients undergoing maintenance hemodialysis, and the carnitine level in plasma and tissues decreases due to both decreased protein intake and the removal of carnitine by dialysis.8

Omega-3 long chain polyunsature fatty acids are essential fatty acids, and their main source is sea foods. These fatty acids mediate a number of key biological processes such as eicosanoid production, the effects of cell membranes on signal transduction pathways, and modulation of inflammation. Long-chain omega-6 fatty acids are also essential fatty acids, are most abundant in meat and meat products, and have pro-inflammatory and pro-oxidant effects competitively against the anti-inflammatory effects of omega-3 fatty acids.9–11 It is emphasized that omega-3/omega-6 fatty acids balance is deteriorated in favor of omega-6 fatty acids due to decreases in some enzyme activities such as desaturase in hemodialysis patients,11 and this situation can be partially corrected by both omega-3 fatty acids12 and l-carnitine treatment.13 In previous studies on dialysis patients, calculations were made on the ratios of omega-3 and omega-6 fatty acids, especially due to their opposite effects on the inflammatory process.12–14 In addition, low plasma omega 3/omega-6 fatty acid ratios have been shown to be associated with CI.15

Recently, positive effects of l-carnitine16,17 and omega-3 fatty acids18,19 on cognitive functions have been reported. However, there is not enough data in the literature on the effect of carnitine and fatty acids on cognitive functions in patients with CKD. According to an experimental CKD rat model, it has been shown that administration of l-carnitine can provide a significant improvement in cognitive function.20 In a recent study in hemodialysis patients, it was emphasized that long-term l-carnitine treatment can reduce white matter microstructural damage and CI.21 We considered that carnitine and omega-3 fatty acids may have positive effects on CI observed in hemodialysis patients. Therefore, in the study, it was aimed to evaluate the relationship between serum total carnitine, free carnitine, and plasma omega-3 [eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA)] and omega-6 [arachidonic acid (AA) and dihomo-gamma-linolenic acid (DGLA)] fatty acid levels and ratios with cognitive functions that was determined according to the MoCA in patients undergoing maintenance hemodialysis.

The mean age of 62 hemodialysis patients (40 males, 64.5%) was 51±13 years. The number of patients with an education level of 12 years was 9 (14.5%), and the number of patients who received 1g of intravenous l-carnitine treatment after each hemodialysis session was 22 (35.5%). There was no statistically significant difference between total and free carnitine levels of patients who received l-carnitine treatment and those who did not [38.6 (14.4–93.4) nmol/mL vs. 33.1 (17.2–68.0) nmol/mL, P=0.971, and 26.4 (13.1–42.1) nmol/mL vs. 23.1 (15.3–40.1) nmol/mL, P=0.930, respectively]. The mean MoCA total score of all patients was 20.5. Details of the demographic, clinical and laboratory data of the patients are presented in Tables 1 and 2.

In the MoCA test, when the threshold value for determining CI was taken as 24 points, the total score was found to be ≤24 in 43 (69.4%) of the patients (Group 1), while the total score was found to be >24 in 19 (30.6%) of the patients (Group 2). In Group 1, the female gender was statistically significantly higher (P=0.006), and education levels were significantly lower (1–5 years, P<0.001; 6–11 years, P=0.019; ≥12 years, P=0.002, respectively.). While there was no statistically significant difference for l-carnitine treatment between the groups (P=0.669), serum free carnitine levels were found to be statistically significantly higher in Group 2 (P=0.040). However, although the serum total carnitine level was numerically higher in Group 2, it did not reach statistical significance (P=0.072). While plasma DHA and EPA+DHA levels were statistically significantly higher in Group 2 (P=0.032, P=0.032, respectively), plasma AA+DGLA/EPA+DHA ratio was found to be statistically significantly lower (P=0.008). In addition, all specific cognitive function scores of Group 1 were statistically significantly lower than Group 2 (P<0.05). Details of demographic and clinical data between groups are shown in Tables 3 and 4.

When the comparison was made by gender, statistically significant lower MoCA score, education level, serum creatinine level (P<0.001, P=0.003, P=0.005, respectively), and higher spKt/Vurea and URR were detected in women (P<0.001, P<0.001, respectively). The plasma AA+DGLA/EPA+DHA ratio was found to be significantly higher in women [4.12 (2.63–5.69) vs. 4.76 (4.29–5.81), P=0.050].

When correlation analysis of variables associated with MoCA total scores was performed, age (rs=−0.410, P=0.001), gender (rs=−0.538, P<0.001), education level (rs=0.694, P<0.001)), spKt/Vure (rs=−0.398, P=0.001), URR (rs=−0.390, P=0.002), albumin (rs=0.296, P=0.019) and plasma AA+DGLA/EPA+DHA ratio (rs=−0.284, P=0.026) were found to be correlated. There was no statistically significant correlation between MoCA total scores and taking l-carnitine treatment, serum total and free carnitine levels. (rs=−0.083, P=0.521; rs=0.176, P=0.171; rs=0.206, P=0.108, respectively). However, there was a statistically significant positive correlation between serum free carnitine levels and plasma EPA and EPA+DHA levels (rs=0.278, P=0.030 and rs=0.271, P=0.034, respectively). In addition, statistically significant negative correlation was found between serum total and free carnitine levels and plasma AA+DGLA/EPA+DHA ratio (rs=−0.363, P=0.004 and rs=−0.414, P=0.001, respectively). Details of the correlation analyzes mentioned above are presented in Tables 5 and 6.

When logistic regression analyses are performed, MoCA total score has a significant relationship with gender (P=0.013), education level (P<0.001), albumin (P=0.027), EPA (P=0.034), DHA (P=0.045), EPA+DHA (P=0.031) and AA+DGLA/EPA+DHA ratio (P=0.002) in univariate logistic regression analysis, whereas in multivariate logistic regression analysis MoCA total score was correlated with education level (P<0.001 and AA+DGLA/EPA+DHA ratio (P=0.009) (Table 7).

In this study, it was determined that CI was common according to the MoCA test in patients undergoing maintenance hemodialysis. This result was similar to previous studies in which cognitive function assessment was performed with the same method in hemodialysis patients.5,25 It was shown that CI was correlated with age, gender, education level, dialysis efficiency, albumin level and AA+DGLA/EPA+DHA ratio. However, it was found that the relationship between education level and AA+DGLA/EPA+DHA ratio with cognitive impairment continued in the other variable adjustments were done.

In our study, similar to previous studies, it was shown that low education level is associated with poor cognitive function.5,25 The significant gender difference between the groups can be explained by the fact that the education level of women in the study is lower than that of men, and this situation negatively affects the cognitive functions of women. In general population based studies, it is emphasized that CI is higher in women compared to men,26 and this may be related to the socioeconomic status.27 Although our study results show that education level has an important effect on cognitive functions, it suggests that there are other risk factors for CI in this patient population, especially in women.

Another result of our study is that the spKt/Vurea and URR values were higher in Group 1 and women, although they did not reach statistical significance. In addition, negative correlations were found between MoCA total scores and spKt/Vurea and URR values in all patients. These results show that increased dialysis dose is not associated with better cognitive function, and even similar to the results of previous studies, high Kt/Vurea value may be associated with worse cognitive function.28,29 The authors stated that this finding can be explained by the fact that patients with high Kt/Vurea were generally weaker than normal and had worse nutrition. In our study, it was determined that low serum albumin level, which is an important indicator of poor nutritional status in hemodialysis patients,30 may be associated with CI. It was also considered that wide fluctuations in fluid, osmolarity and uremic toxin levels that may occur due to more intense hemodialysis in these patients may adversely affect cognitive functions. This relationship between serum albumin level, dialysis dose and cognitive functions was not observed in logistic regression analyses. However, the data support that higher spKt/Vurea is not associated with better cognitive function.

Carnitine, an essential cofactor for normal fatty acid metabolism, has shown beneficial effects on brain tissue. l-carnitine is known to contribute to the formation of acetyl carnitine and acetylcholine by mediating the passage of acetyl particles from mitochondria to cytosol in brain tissue.16 Acetyl carnitine shows neurobiological effects with synaptic morphology, synaptic transmission of many neurotransmitters, phospholipid metabolism, and brain energy modulation.31 It has been demonstrated that l-carnitine treatment can affect the central nervous system, and slow down CI by increasing serum concentrations of both l-carnitine and acetyl carnitine.32 It has been shown that serum carnitine levels of patients undergoing maintenance hemodialysis are significantly lower.8 In a study of Zhang et al.,33 it was emphasized that low total and free carnitine levels in hemodialysis patients can increase with l-carnitine treatment, and dialysis related complications can be reduced by determining the appropriate plasma free carnitine level. The data of our study suggest that low serum free carnitine level may be associated with poor cognitive functions. Also, the fact that carnitine levels were not different between those who received and did not receive l-carnitine therapy, suggesting that those who received l-carnitine therapy had lower carnitine levels at the beginning of treatment. In an experimental study on this subject, it was shown that administration of l-carnine to rats with CKD provided significant improvement in cognitive functions, and it was stated that this effect could be achieved by the antioxidant property, the energy need of the cells, and regulating the changes in membrane phospho lipid metabolism.20 In a recent study, it was revealed that l-carnitine may exert a neuroprotective effect by suppressing neuroinflammation against white matter microstructural damage and cognitive deterioration in hemodialysis patients.21 Although the available data support the use of l-carnitine in the treatment of CI in this patient population, the mechanisms of the effect of l-carnitine on cognitive functions should be revealed in detail.

Nowadays, it is stated that EPA and DHA can have neuroprotective effects. In brain tissue, DHA, main long-chain fatty acid of phospholipids, may prevent or correct CI when they are taken with diet for a long time.18,19 In hemodialysis patients, it was found that plasma omega-3/omega-6 fatty acid distribution changes abnormally, and plasma/cell phospholipids have lower EPA and DHA concentration.17,34 It was thought that this situation may be related to the scarcity of sea food taken in the diet of hemodialysis patients, the decrease in the functions of elongase and desaturase enzymes, malabsorption, and omega-3 fatty acid loss from dialysate.35,36

In our study, a significant relationship was found between lower MoCA scores and high AA+DGLA/EPA+DHA ratio in hemodialysis patients. This finding supports that there may be a relationship between the decrease in omega-3 fatty acid ratios and CI in hemodialysis patients. In addition, higher AA+DGLA/EPA+DHA ratio in females compared to males in our study may contribute causally to higher CI in females. Possible reasons for this situation are, as stated in a recent study, lower consumption of food containing omega-3 fatty acids in women,36 worse nutritional status of female patients lower serum albumin levels or greater omega-3 fatty acid loss with higher dialysis dose related dialysate. The positive correlation of serum free carnitine level with EPA and DHA, and negative correlation with AA+DGLA/EPA+DHA ratio show that l-carnitine can also be effective on the regulation of omega-3 fatty acid metabolism and the preservation of cognitive functions.

Our study has some limitations. It is difficult to generalize according to our findings, since it is a single-center, non-randomized study with a relatively small sample size. In addition, due to its cross-sectional design, cause and effect association should be done carefully. In our study, it was accepted that cognitive functions were impaired in patients with a MoCA total score of 24 points, and no distinction was made between mild-moderate CI and dementia, since patients with dementia were excluded from the study. Although there was no statistical difference in the proportions of patients receiving l-carnitine therapy between the groups, receiving this therapy likely affected serum total and free carnitine levels. Finally, since plasma fatty acid distribution may be affected by diet and dietary fatty acid intake, this could be further evaluated with nutritional intake analyzes. Despite these limitations, the specificity of the study group constitutes the strength of our study.

As a result, the prevalence of CI according to the MoCA was found to be high in hemodialysis patients. Low education level and high plasma AA+DGLA/EPA+DHA ratio were determined to be independent risk factors associated with CI. It has been shown that serum carnitine levels may have positive effects on plasma EPA+DHA and AA+DGLA distributions in these patients. According to our findings, it can be said that low omega-3 fatty acid levels in hemodialysis patients may be associated with the onset and progression of CI, and that low carnitine levels may partially contribute to this process. Assessment of cognitive functions in these patients especially those with low education level can be included in routine clinical practice, and symptoms associated with CI can be improved with omega-3 fatty acid and carnitine supplementation in patients with CI. In addition, data support that cognitive education programs should be applied to prevent or reduce cognitive impairment in hemodialysis patients with low education levels. However, our findings should be confirmed by prospective studies with large participation.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The authors declare that they have no conflict of interests.