Abstract
Background
Red blood cell (RBC) membrane fatty acid composition may reflect early metabolic disturbances linked to chronic kidney disease (CKD) and type 2 diabetes (T2D), offering insights into shared pathophysiological mechanisms. This study aimed to characterise RBC membrane fatty acid profiles in individuals with early-stage CKD, comparing those with and without T2D to identify disease-specific patterns.
Methods
The cross-sectional analysis comprised 893 participants (290 with T2D and 603 deemed at high risk of T2D by the African Diabetes Risk Score), aged ≥ 18 years, recruited from 16 communities in Cape Town, South Africa, between 2017 and 2019. RBC membrane fatty acids were extracted and analysed using gas chromatography. CKD was defined as an estimated glomerular filtration rate < 60 ml/min/1.73 m² and/or albumin-to-creatinine ratio > 3 mg/mmol.
Results
In total, 25.9% presented with CKD, 36.1% with T2D, and 15.6% presented with comorbid CKD and T2D. A higher lipogenic index was associated with an increased odds of prevalent CKD [OR (95% CI): 2.73 (1.22–6.12); p = 0.015], with 18:2n-6 (linoleic acid) [OR (95% CI): 0.81 (0.67–0.99); p = 0.035], total n-6 polyunsaturated fatty acids (PUFA) [OR (95% CI): 0.86 (0.76–0.98); p = 0.025], and total PUFA [OR (95% CI): 0.88 (0.77–0.99); p = 0.041] associated with reduced odds of prevalent CKD, with the associations between the PUFAs modified by T2D status (interaction p < 0.070).
Conclusion
Our findings showed that comorbid CKD and T2D exacerbates changes in PUFAs, suggesting complex metabolic interactions. These fatty acids may be of clinical value adding to already existing biomarkers to improve early detection or monitoring of these diseases. Furthermore, promoting appropriately balanced dietary intake of n-6 and n-3 PUFAs, could support metabolic health, contributing to more personalized patient care.
Keywords: Chronic kidney disease, Type 2 diabetes mellitus, Red blood cell membrane fatty acids
Subject terms: Biomarkers, Diseases, Endocrinology, Medical research, Nephrology
Background
Chronic kidney disease (CKD) affects more than 840 million individuals globally and increases the risk of cardiovascular disease (CVD) mortality significantly1,2. Type 2 diabetes mellitus (T2D) is one of the major risk factors for CKD with the coexistences of the two conditions exerting synergistic negative effects on metabolic and cardiovascular health3,4.
CKD and T2D are characterized by complex metabolic abnormalities, including dyslipidaemia, insulin resistance, and altered glucose metabolism, with dyslipidaemia playing a pivotal role in the progression of CVD and kidney dysfunction5,6. Indeed, as the kidneys regulate the synthesis, transportation, and excretion of lipids and lipoproteins7, changes in kidney function disrupts lipid metabolism, leading to changes in fatty acid profiles, dyslipidaemia and increased cardiovascular risk. Similarly, T2D disrupts lipid metabolism through insulin resistance, increased hepatic lipogenesis, altered lipoprotein metabolism, and inflammatory processes; all of which increases the risk of cardiovascular complications associated with T2D8. These changes often results in hydrocarbon-chain saturation variation, chain length, and omega-3 (n-3)/omega-6 (n-6) fatty acid ratios, which in turn influences membrane fluidity, stability, and signalling properties9. Red blood cell (RBC) membrane fatty acid profiles, which reflect long-term dietary intake, endogenous synthesis, and metabolic processing10, have been associated with CVD and T2D11. However, despite the recognized interplay between CKD and T2D, the specific profile of RBC membrane fatty acid composition in cases of CKD, especially those with concomitant T2D, remains underexplored.
Given the link between lipid metabolism, inflammation, and cellular function, understanding the complex relationship between CKD, T2D, and RBC membrane fatty acid composition could provide valuable insights into shared pathophysiological mechanisms and identify novel therapeutic targets for mitigating cardiovascular risk and kidney-related complications. Exploring the intersection of CKD and T2D through the lens of RBC membrane fatty acid composition may uncover markers that may be used in addition to currently existing biomarkers to improve early detection, monitoring, and treatment of these intertwined diseases. Indeed, changes in RBC membrane fatty acids may serve as early indicators of metabolic imbalances associated with CKD and T2D, allowing for earlier diagnosis and better disease tracking. This study aims to characterise the profile of RBC membrane fatty acids in individuals with CKD, comparing those with and without T2D, to identify specific patterns linked to these conditions.
Methods
Study population and setting
The South African Diabetes Prevention Programme (SA-DPP) is an open-labelled cluster randomized control trial, conducted across 16 resource-poor communities in Cape Town, South Africa12. The SA-DPP sample size was calculated assuming a 13.6% cumulative diabetes incidence over 2–3 years, based on the Bellville South cohort13, with a relative risk of 0.51 reflecting the pooled effect of lifestyle interventions versus usual care14. An intra-cluster correlation coefficient (ICC) for fasting glucose of 0.02 was assumed15. The calculation used a 5% significance level, 80% power, and accounted for an anticipated 20–25% loss to follow-up over 36 months. Participants included men and women from black and mixed ancestry populations, aged ≥ 18 years, with known T2D or deemed at high-risk of T2D. Participants were excluded if they were outside the eligible age range, bedridden, pregnant or breastfeeding, or had received cancer or tuberculosis treatment within three months prior to recruitment. These individuals were recruited by self-selection approaches. The African Diabetes Risk Score16 was used to identify adults at high-risk for T2D. This validated African screening tool17 comprises non-laboratory-based variables such as age, waist circumference and prevalent hypertension. The data included in this analysis was collected between 2017 and 2019. The study was conducted in accordance with the Declaration of Helsinki and approved by the Research Ethics Committee of the South African Medical Research Council (SAMRC) (approval no. EC018-7/2015). All participants voluntary signed written informed consent after the procedures were fully explained in the language of their choice.
Questionnaires and physical examination
Data were collected by eight trained fieldworkers, who were employed study staff. They conducted all anthropometric and blood pressure measurements using standardized methodology18, as well as administering a questionnaire to ascertain participant age, sex, and population group. Body weight (to the nearest 0.1 kg) was measured with a calibrated Omron digital scale (Omron Health Care, Hamburg, Germany), with the participant in light clothing and without shoes. Participant height was measured with a stadiometer (to the nearest 1.0 cm), with the participant standing in an upright position, on a flat surface. Waist and hip circumference were measured using a non-elastic tape measure at the level of the umbilicus and the widest part of the hips, respectively. Blood pressure was measured after five minutes of seated rest. The systolic and diastolic blood pressures (SBP and DBP, respectively) were recorded with an automated blood pressure monitor (Omron 711, Omron Health Care, Hamburg, Germany) on three occasions at three-minute intervals, using appropriately sized cuffs. An average of the last two readings was used in the current analyses.
Biochemical analyses
All biochemical analyses were conducted at an ISO accredited laboratory (PathCare Laboratories, Cape Town, SA). An oral glucose tolerance test (OGTT) was conducted on participants not previously diagnosed with T2D, as per the World Health Organization’s (WHO) guidelines19. Briefly, a qualified nurse collected blood samples after a 10-hour overnight fast which was followed by the oral administration of 75 g anhydrous glucose dissolved in 250 ml water and a blood sample collection two hours later. For all participants, the following concentrations were determined: plasma glucose by the glucose oxidase method (Glucose Analyzer 2, Beckman Instruments, Fullerton, CA, USA); serum insulin by a Microparticle Enzyme Immunoassay (AxSym Insulin Kit, Abbot, IL, USA); glycated haemoglobin (HbA1c) by high-performance liquid chromatography (Biorad Variant Turbo, BioRad, Johannesburg, SA); serum high-sensitivity C-reactive protein (hsCRP) by automated immunoturbidimetric assays; serum triglycerides (TG), total cholesterol (TC), and high-density lipoprotein cholesterol (HDL-C) by the Roche Modular auto analyzer and enzymatic colorimetric assays; low-density lipoprotein cholesterol (LDL-C) calculated using the Friedewald formula20; serum and urinary creatinine concentration by means of the modified Jaffe-Kinetic method (calibrated to isotope dilution mass spectrometry standards) (Beckman AU, Beckman Coulter, SA); urine albumin by colorimetric (using bromocresol purple) method (Beckman AU, Beckman Coulter, SA) and a full blood count was determined on a Coulter LH 750 haematology analyser (Beckman Coulter, South Africa).
Determination of red blood cell membrane fatty acid composition
RBC analyses were conducted at the fatty acid research laboratory of the Non-Communicable Diseases Research Unit of the SAMRC. The fatty acid composition in RBC membranes were assessed by subjecting aliquots of saline-washed RBCs (300µL) to total lipid extraction with chloroform (amylene stabilized): methanol (2:1; vol: vol; containing 0.01% butylated hydroxytoluene) by using a modification of the method of Folch et al.21. Thin-layer chromatography (TLC) was applied to isolate the total phospholipid (TPL) fraction in the RBC total lipid extract22. The TLC-isolated RBC-TPL fractions was trans-methylated using methanol: sulphuric acid (95:5; vol: vol) at 70 °C for 2 h to yield fatty acid methyl esters (FAMEs) which, after cooling, were extracted with water and n-hexane. The organic layer containing the FAMEs was evaporated, redissolved in a small volume of n-hexane, and analyzed by gas-liquid chromatography (GLC); applied GLC conditions were as described23. The sample FAMEs were identified by comparison of the retention times with those of a standard FAME mixture (27 FAMEs; Nu-Chek Prep Inc., MN, USA). The relative percentage of a FAME was calculated by taking the area under the curve (AUC) of a given FAME as a percentage of the total area count of all the FAMEs identified in the sample (%, wt: wt). Product-to-precursor fatty acid ratios of the samples were used as a proxy index to reflect estimated desaturase enzyme activities (desaturase indices) as 20:3n-6/18:2n-6 for D6D and 20:4n-6/20:3n-6 for D5D activity. Stearoyl-Coenzyme A desaturase-1 (SCD1) index was estimated by the ratios of 16:1n-7/16:0 (for SCD1-16) and 18:1n-9/18:0 (for SCD1-18). This method has been well established as an approach to estimate the activity of desaturase enzymes in humans24. The elongase index was estimated by the ratio 18:0/16:0, the lipogenic index by 16:0/18:2n-6 and the omega-3 index was calculated as the sum of 20:5n-3 + 22:6n-3. Additional ratios were calculated and included arachidonic acid (AA)/linoleic acid (LA), (20:4n-6/18:2n-6); AA/eicosapentaenoic acid (EPA), (20:4n-6/20:5n-3); AA/EPA + docosahexaenoic acid (DHA), (20:4n-6/(20:5n-3 + 22:6n-3)); docosapentaenoic acid (DPA)n-6/DHA, (22:5n-6/22:6n-3) and total highly unsaturated fatty acids (tHUFA)n-6/tHUFAn-3.
Classification of kidney function and co-morbidities
Kidney function was estimated using the serum creatinine-based CKD Epidemiology Collaboration 2009 (CKD-EPI) equation25, with the race correction factor omitted. CKD was defined as an estimated glomerular filtration rate (eGFR) of < 60 ml/min/1.73 m2 and/or urinary albumin-to-creatinine ratio (uACR) > 3 mg/mmol26. CKD staging was based on the Kidney Disease Improving Global Outcomes (KDIGO) guidelines27, as stage 1 (eGFR ≥ 90 ml/min/1.73m2 and uACR ≥ 3 mg/mmol), stage 2 (eGFR 60–89 ml/min/1.73m2 and uACR ≥ 3 mg/mmol), stage 3 (eGFR 30–59 ml/min/1.73m2), stage 4 (eGFR 15–29 ml/min/1.73m2) and stage 5 (eGFR < 15 ml/min/1.73m2). Body mass index (BMI) was calculated as weight in kilograms divided by height in metres squared (kg/m2). Weight status was categorised as normal weight (BMI ≤ 24.9 kg/m2), overweight (BMI 25.0–29.9 kg/m2) and obese (BMI ≥ 30 kg/m2). Hypertension was defined as SBP ≥ 140 mmHg and/or DBP ≥ 90 mmHg28, or taking anti-hypertensive medications. T2D was defined as fasting blood glucose (FBG) ≥ 7.0 mmol/L and/or 2-h glucose ≥ 11.1 mmol/L or taking glucose-lowering medications19. Participants were classified as having no disease if they had neither CKD nor T2D.
Statistical analysis
Participant characteristics were summarised as median (25th–75th percentile), due to the non-Gaussian distribution of most variables, or counts and percentages. Group comparisons were analysed by X2 tests for categorical variables and Kruskal-Wallis test for continuous variables. In instances where the Kruskal-Wallis test was rejected, the Dunn’s test, with Bonferroni adjustment, was used as non-parametric pairwise multiple-comparison post-hoc test. To assess the independent association between measured fatty acids and kidney function (i.e., eGFR), crude and adjusted multivariable linear regression models were used. Variables adjusted for included those found to be significantly associated with eGFR, namely age, sex, waist circumference, BMI, SBP, HbA1c, TG, LDL-C, CRP, and prevalent T2D. To assess the independent association between measured fatty acids and prevalent CKD, multivariable logistic regression models were used, while adjusting for confounding variables (age, sex, waist circumference, HbA1c, TG, and prevalent T2D. To assess potential multicollinearity among covariates included in the multivariable regression models, we calculated Variance Inflation Factor (VIF) values for all predictors. A VIF threshold of < 5 was used to indicate absence of concerning multicollinearity. To investigate whether the associations between individual fatty acids and the outcomes differed by T2D status, we evaluated effect modification by including an interaction term between T2D status and each fatty acid in the regression models. In addition to the interaction term, we conducted stratified analyses by T2D status to further clarify the nature and direction of any potential effect modification. Separate regression models were therefore fitted for participants with and without T2D, using the same covariate structure applied in the main analyses. To visually investigate the interaction between T2D and the various fatty acids, predictive margins were estimated, and graphs plotted for all fatty acids with a significant interaction term. All statistical analyses were performed using STATA version 18 (Statcorp, College Station, TX). Statistical significance was defined as a p-value < 0.05 for main effects, whereas a threshold of p < 0.10 was applied for interaction tests. This slightly more liberal alpha was chosen given the exploratory nature of the interaction analyses and the inherently lower statistical power of such tests. This approach facilitates the detection of potential effect modifications that may warrant further investigation, while maintaining a reasonable balance between sensitivity and the risk of Type I error.
Results
General characteristics of the study population
Of the 2,173 individuals screened in the community, 690 participants who were deemed at high-risk of T2D based on the African Diabetes Risk Score, and 346 participants with known T2D presented at our research clinic for further assessments. Of the 1,036 participants received at the clinic, 893 had complete kidney function and fatty acid data and were included in the current analysis. Among the 893 participants, 82.1% were female, with a group median age of 55 years. In this sample, 25.9% (n = 231) had CKD, 36.1% had T2D, and 15.6% had comorbid CKD and T2D. Of the participants with CKD, most were in CKD stage 1 (61.5%) and stage 2 (25.1%), with 18.2% presenting with a moderately elevated uACR (3–30 mg/mmol) and 3.8% with a markedly elevated uACR (> 30 mg/mmol). Table 1 presents a summary of the sociodemographic, clinical, and biochemical characteristics of the study participants categorized as having no disease (participants with no CKD or T2D), CKD only, T2D only and comorbid CKD and T2D (CKD + T2D). Participants with CKD + T2D were significantly older than their counterparts (p = 0.0001), with 62.6% being of mixed ancestry. CKD + T2D was associated with lower body weight, waist circumference, hip circumference, and BMI compared to their counterparts (p = 0.0001 for all). Those with CKD only had significantly higher SBP and DBP compared to their counterparts (p < 0.0009 for both). T2D only and CKD + T2D, was associated with higher FBG, 2 h glucose and HbA1c concentrations, compared to groups without T2D (p = 0.0001 for all). The group with no disease had significantly lower TG and higher LDL-C concentrations compared to those in the T2D only and CKD + T2D groups (p = 0.0001 for both). The group with CKD + T2D had a lower eGFR, but similar uACR, compared to those with CKD only (p = 0.0039 and p = 0.382, respectively).
Table 1.
Participants characteristics presented by CKD and T2D status.
| Sociodemographic variables | No disease (n = 479) | CKD only (n = 92) | T2D only (n = 183) | CKD + T2D (n = 139) | p-value |
|---|---|---|---|---|---|
| Age (years) | 52.0 (45.0–59.0) | 53.5 (46.5–60.0) | 58.0 (50.0–66.0)b, d | 61.0 (55.0–68.0)c, e,f | 0.0001 |
| Sex (n,% female) | 395 (82.5) | 74 (80.4) | 155 (84.7) | 109 (78.4) | 0.505 |
| Race (n, % Mixed Ancestry) | 200 (41.8) | 33 (35.9) | 106 (57.9) | 87 (62.6) | 0.0001 |
| Anthropometry | |||||
| Weight (kg) | 91.8 (80.3-105.2) | 89.0 (77.9-101.1) | 80.1 (65.8–91.4)b, d | 79.7 (66.2–91.9)c, e | 0.0001 |
| Height (m) | 159.6 (155.0-165.0) | 158.3 (154.9-163.6) | 159.2 (154.3-164.8) | 161.2 (155.2-167.2) | 0.330 |
| Waist circumference (cm) | 103.8 (95.9-111.1) | 104.2 (96.2-111.1) | 96.3 (84.1-105.1)b, d | 94.3 (83.4-104.2)c, e | 0.0001 |
| Hip circumference (cm) (n = 844) | 113.5 (104.3-123.1) | 108.9 (101.6-117.2)a | 106.6 (98.1-115.3)b | 104.2 (96.6-112.4)c, e | 0.0001 |
| Body mass index (kg/m2) | 36.1 (30.7–41.5) | 34.1 (30.1–40.2) | 30.4 (25.9–36.8)b, d | 30.3 (26.6–35.1)c, e | 0.0001 |
| Body mass index categories (n, %) | 0.0001 | ||||
| Underweight | 0 (0.0) | 0 (0.0) | 3 (1.6) | 4 (2.9) | |
| Normal | 15 (3.1) | 4 (4.4) | 40 (21.9) | 21 (15.4) | |
| Overweight | 88 (18.4) | 18 (19.8) | 41 (22.4) | 40 (29.4) | |
| Obese | 376 (78.5) | 69 (75.8) | 99 (54.1) | 71 (52.2) | |
| Blood pressure | |||||
| Systolic blood pressure (mmHg) | 121.0 (112.0-133.5) | 128.5 (114.0-150.5)a | 119.5 (109.0-135.0)d | 120.0 (108.0-130.0)e | 0.0008 |
| Diastolic blood pressure (mmHg) | 82.5 (76.5–90.0) | 87.0 (79.0–97.0)a | 80.5 (74.5–90.0)d | 78.8 (72.2–85.0)c, e | 0.0001 |
| Biochemical | |||||
| Fasting blood glucose (mmol/L) | 4.9 (4.6–5.4) | 4.9 (4.6–5.5) | 8.8 (6.9–12.2)b, d | 9.5 (6.5–12.9)c, e | 0.0001 |
| 2-hour glucose (mmol/L) (n = 599) | 5.8 (4.8–6.9) | 6.3 (5.2–7.3) | 16.5 (13.9–22.0)b, d | 17.3 (16.8–17.8)c, e | 0.0001 |
| HbA1c (%) | 5.8 (5.6–6.1) | 5.8 (5.6–6.1) | 8.7 (7.1–10.5)b, d | 8.9 (7.2–10.8)c, e | 0.0001 |
| Fasting insulin (IU/L) | 8.4 (5.5–12.3) | 9.7 (5.9–14.1) | 9.1 (6.3–13.7)b | 9.2 (6.0-13.8) | 0.0187 |
| C-reactive protein (mg/l) | 8.3 (4.2–13.9) | 8.7 (3.8–12.7) | 7.3 (3.2–13.4) | 6.6 (3.8–14.1) | 0.354 |
| Total cholesterol (mmol/l) | 4.8 (4.3–5.7) | 5.0 (4.2–5.8) | 4.8 (4.0-5.6) | 4.6 (3.8–5.4)c | 0.0293 |
| Triglycerides (mmol/l) | 1.2 (0.9–1.6) | 1.3 (1.0-1.8)a | 1.6 (1.2-2.0)b | 1.8 (1.2–2.5)c, e | 0.0001 |
| High-density lipoprotein cholesterol (mmol/l) | 1.2 (1.1–1.4) | 1.2 (1.1–1.4) | 1.2 (1.0-1.4) | 1.2 (1.0-1.4) | 0.294 |
| Low-density lipoprotein cholesterol (mmol/l) | 3.2 (2.6–3.9) | 3.3 (2.6–4.2) | 2.9 (2.3–3.6)b, d | 2.8 (2.0-3.6)c, e | 0.0001 |
| Kidney function measures | |||||
| eGFR (ml/min/1.73m2) | 103.0 (95.0-114.0) | 99.0 (89.5–111.0)a | 99.0 (87.0-110.0)b | 93.0 (73.0-106.0)c, e,f | 0.0001 |
| CKD stages (n, %) | 0.0001 | ||||
| Stage 1 | 0 (0) | 69 (75.0) | 0 (0) | 73 (52.5) | |
| Stage 2 | 0 (0) | 15 (16.3) | 0 (0) | 43 (30.9) | |
| Stage 3 | 0 (0) | 7 (7.6) | 0 (0) | 20 (14.4) | |
| Stage 4 | 0 (0) | 1 (1.1) | 0 (0) | 3 (2.2) | |
| Albumin-to-creatine ratio | 1.0 (0.0–1.0) | 5.0 (3.0–12.0)a | 1.0 (1.0–1.0)b, d | 6.0 (4.0–21.0)c, f | |
| Albuminuria severity (n, %) | 0.0001 | ||||
| Moderately increased (3–30 mg/mmol) | 0 (0) | 65 (70.6) | 0 (0) | 94 (67.6) | |
| Severely increased (> 30 mg/mmol) | 0 (0) | 9 (9.8) | 0 (0) | 24 (17.3) | |
Data are presented as median (25th–75th percentiles), count and percentages. The p-values represent the differences in participant characteristics between groups with no CKD nor T2D (denoted as without disease), with CKD only, with T2D only and with CKD and T2D using the Dunn’s test (with Bonferroni adjustment) when the Kruskal-Wallis test was rejected as aWithout disease vs. CKD only; bWithout disease vs. T2D only; cWithout disease vs. CKD + T2D; dCKD only vs. T2D only; eCKD only vs. CKD + T2D; fT2D only vs. CKD + T2D. Abbreviations: CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin.
Red blood cell membrane fatty acid profile
Table 2 presents the fatty acid profiles of study participants stratified by disease status (no disease, CKD only, T2D only and CKD + T2D), while Fig. 1 graphically displays only those fatty acids that differed significantly across groups. The RBC membrane contained the highest relative percentage of saturated fatty acids (SFAs) (43.9% to 44.4% across the groups), followed by PUFAs (39.1% to 39.9%) and MUFAs (16.2% to 16.6%) with the lowest relative percentage. The relative percentage of 18:0 (stearic acid), 20:0 (arachidic acid), total SFA, 16:1n-7 (palmitoleic acid), 18:1n-7 (cis-vaccenic acid), 22:1n-9 (erucic acid), total n-7 MUFA, and 20:3n-3 (eicosatrienoic acid) were lower and 24:0 (lignoceric acid), 18:1n-9 (oleic acid), total n-9 MUFA, and 22:6n-3 (docosahexaenoic acid) were higher in those with T2D only and CKD + T2D compared to participants with no disease or CKD only (p < 0.042 for all). The desaturase index of SCD1-16 and total n-6/total n-3 ratio were lowest, and the desaturase index of SCD1-18 and omega-3 index highest in the group with CKD + T2D compared to those with no disease (p < 0.0007 for all). The ratios AA/(EPA + DHA), DPAn-6/DHA and tHUFAn-6/tHUFAn-3 were all lower in the groups with T2D, including those with T2D only and CKD + T2D (p = 0.0001 for all).
Table 2.
Red blood cell membrane fatty acid composition (%) presented by CKD and T2D status.
| No disease (n = 479) | CKD only (n = 92) | T2D only (n = 183) | CKD + T2D (n = 139) | p-value | |
|---|---|---|---|---|---|
| Saturated fatty acids | |||||
| 14:0 (myristic acid) | 0.20 (0.17–0.22) | 0.20 (0.17–0.23) | 0.19 (0.16–0.22) | 0.19 (0.16–0.21) | 0.070 |
| 16:0 (palmitic acid) | 21.66 (20.93–22.36) | 21.97 (20.94–22.54) | 21.58 (20.89–22.42) | 21.59 (20.94–22.35) | 0.571 |
| 18:0 (stearic acid) | 15.30 (14.78–15.89) | 15.45 (14.75–15.98) | 15.00 (14.45–15.62)b, d | 15.04 (14.56–15.53)c, e | 0.0001 |
| 20:0 (arachidic acid) | 0.34 (0.31–0.38) | 0.34 (0.30–0.38) | 0.33 (0.30–0.36)b | 0.32 (0.29–0.37)c | 0.0001 |
| 22:0 (behenic acid) | 1.59 (1.43–1.75) | 1.53 (1.40–1.70) | 1.58 (1.44–1.73) | 1.49 (1.35–1.69) | 0.128 |
| 24:0 (lignoceric acid) | 5.12 (4.71–5.54) | 5.16 (4.66–5.51) | 5.30 (4.83–5.77)b | 5.23 (4.73–5.56) | 0.0419 |
| Total SFA | 44.15 (43.39–44.87) | 44.38 (43.76–44.99) | 43.92 (43.23–44.66)d | 43.80 (43.20-44.47)c, e | 0.0003 |
| Monounsaturated fatty acids | |||||
| 16:1n-7 (palmitoleic acid) | 0.22 (0.17–0.29) | 0.23 (0.18–0.31) | 0.22 (0.17–0.26)d | 0.20 (0.16–0.27)c, e | 0.0011 |
| 18:1n-7 (cis-vaccenic acid) | 0.95 (0.85–1.05) | 0.95 (0.87–1.09) | 0.88 (0.81-1.00)b, d | 0.88 (0.81–0.99)c, e | 0.0001 |
| 18:1n-9 (oleic acid) | 10.74 (10.10-11.47) | 11.00 (10.45–11.70) | 11.09 (10.40-11.72)b | 11.15 (10.63–11.64)c | 0.0001 |
| 20:1n-9 (eicosenoic acid) | 0.160 (0.140–0.170) | 0.155 (0.140–0.170) | 0.160 (0.150–0.180) | 0.160 (0.150–0.190) | 0.060 |
| 22:1n-9 (erucic acid) | 0.07 (0.06–0.08) | 0.07 (0.06–0.08) | 0.06 (0.06–0.08)b | 0.06 (0.06–0.07)c, e | 0.0004 |
| 24:1n-9 (nervonic acid) | 4.03 (3.68–4.38) | 4.09 (3.72–4.51) | 4.00 (3.65–4.33) | 3.89 (3.54–4.29) | 0.125 |
| Total n-7 MUFA | 1.18 (1.06–1.33) | 1.21 (1.08–1.40) | 1.11 (1.00-1.23)b, d | 1.08 (0.99–1.23)c, e | 0.0001 |
| Total n-9 MUFA | 15.02 (14.32–15.71) | 15.34 (14.53–16.07) | 15.32 (14.55–15.87)b | 15.24 (14.72–15.90)c | 0.0048 |
| Total MUFA | 16.21 (15.44-17.00) | 16.61 (15.77–17.28) | 16.41 (15.63–17.05) | 16.41 (15.92–17.12) | 0.0531 |
| Polyunsaturated fatty acids | |||||
| 18:2n-6 (linoleic acid) | 10.33 (9.46–11.27) | 10.11 (9.15–11.13) | 10.31 (9.20-11.08) | 10.13 (9.19–11.09) | 0.092 |
| 18:3n-3 (alpha-linolenic acid) | 0.09 (0.07–0.10) | 0.09 (0.07–0.10) | 0.08 (0.07–0.09) | 0.09 (0.07–0.10) | 0.131 |
| 18:3n-6 (gamma-linolenic acid) | 0.07 (0.06–0.08) | 0.07 (0.07–0.08) | 0.07 (0.06–0.08) | 0.07 (0.06–0.08) | 0.756 |
| 20:2n-6 (eicosadienoic acid) | 0.30 (0.26–0.33) | 0.29 (0.26–0.33) | 0.28 (0.26–0.32)b | 0.29 (0.27–0.33) | 0.0371 |
| 20:3n-3 (eicosatrienoic acid) | 0.09 (0.08–0.10) | 0.09 (0.08–0.10) | 0.08 (0.08–0.09)b, d | 0.08 (0.07–0.09)c, e | 0.0001 |
| 20:3n-6 (dihomo-gamma-linolenic acid) | 1.63 (1.44–1.83) | 1.58 (1.44–1.75) | 1.58 (1.41–1.80) | 1.59 (1.39–1.84) | 0.159 |
| 20:4n-6 (arachidonic acid) | 15.12 (14.22–16.09) | 14.94 (14.21–15.66) | 15.06 (14.17–16.12) | 15.05 (14.30-15.95) | 0.715 |
| 20:5n-3 (eicosapentaenoic acid) | 0.47 (0.38–0.61) | 0.49 (0.40–0.59) | 0.50 (0.41–0.64) | 0.51 (0.41–0.64) | 0.362 |
| 22:2n-6 (docosadienoic acid) | 0.09 (0.07–0.10) | 0.08 (0.07–0.10) | 0.08 (0.07–0.10) | 0.08 (0.06–0.10) | 0.567 |
| 22:4n-6 (adrenic acid) | 3.36 (2.98–3.71) | 3.27 (2.83–3.57) | 3.28 (2.89–3.69) | 3.26 (2.86–3.68) | 0.214 |
| 22:5n-3 (docosapentaenoic acid, DPAn-3) | 2.15 (1.94–2.34) | 2.13 (2.00-2.34) | 2.22 (2.00-2.40) | 2.24 (1.97–2.49) | 0.079 |
| 22:5n-6 (docosapentaenoic acid, DPAn-6) | 0.64 (0.54–0.72) | 0.63 (0.55–0.72) | 0.60 (0.51–0.71) | 0.61 (0.52–0.72) | 0.132 |
| 22:6n-3 (docosahexaenoic acid) | 4.93 (4.34–5.62) | 5.04 (4.50–5.77) | 5.40 (4.76–6.01)b | 5.44 (4.79–6.18)c, e | 0.0001 |
| Total n-3 PUFA (Tn-3) | 7.79 (7.02–8.55) | 7.97 (7.26–8.61) | 8.22 (7.58–9.03)b | 8.32 (7.56–9.23)c, e | 0.0001 |
| Total n-6 PUFA (Tn-6) | 31.68 (30.42–32.87) | 31.31 (29.74–32.09)a | 31.44 (30.11–32.51) | 31.22 (29.95–32.31) | 0.0048 |
| Total PUFA | 39.63 (38.45–40.57) | 39.08 (37.84–40.01)a | 39.85 (38.69–40.65)d | 39.74 (38.69–40.72)e | 0.0052 |
| Estimated desaturase indices, elongase indices and fatty acid ratios | |||||
| Delta-5 desaturase (D5D) | 9.25 (7.99–10.66) | 9.57 (8.14–10.55) | 9.60 (8.05–11.13) | 9.62 (8.17–11.03) | 0.252 |
| Delta-6 desaturase (D6D) | 0.16 (0.14–0.18) | 0.16 (0.14–0.18) | 0.16 (0.14–0.18) | 0.16 (0.14–0.18) | 0.979 |
| Stearoyl-Coenzyme A desaturase-1-16 (SCD1-16) | 0.010 (0.008–0.013) | 0.011 (0.008–0.015) | 0.010 (0.008–0.012)d | 0.009 (0.008–0.012)c, e | 0.0007 |
| Stearoyl-Coenzyme A desaturase-1-18 (SCD1-18) | 0.70 (0.65–0.76) | 0.73 (0.66–0.77) | 0.73 (0.68–0.79)b | 0.74 (0.69–0.79)c | 0.0001 |
| Elongase | 0.70 (0.67–0.75) | 0.70 (0.66–0.76) | 0.69 (0.65–0.74) | 0.69 (0.65–0.73) | 0.054 |
| Omega-3 index | 5.44 (4.78–6.14) | 5.55 (4.93–6.21) | 5.87 (5.25–6.55)b | 6.00 (5.27–6.79)c, e | 0.0001 |
| Lipogenic index | 2.08 (1.89–2.29) | 2.14 (1.96–2.39) | 2.16 (1.96–2.35) | 2.15 (1.93–2.35) | 0.0548 |
| Tn-6/Tn-3 | 4.06 (3.64–4.60) | 3.90 (3.53–4.34) | 3.80 (3.40–4.23)b | 3.75 (3.32–4.25)c | 0.0001 |
| AA/LA | 1.46 (1.28–1.65) | 1.44 (1.30–1.69) | 1.47 (1.30–1.67) | 1.49 (1.30–1.69) | 0.268 |
| AA/EPA | 32.13 (24.40-40.31) | 30.50 (24.98–38.25) | 30.62 (22.66–38.47) | 29.39 (22.50-37.68) | 0.547 |
| AA/(EPA + DHA) | 2.79 (2.39–3.23) | 2.71 (2.30–3.08) | 2.60 (2.18–3.02)b | 2.56 (2.11-3.00)c | 0.0001 |
| DPAn-6/DHA | 0.13 (0.10–0.16) | 0.12 (0.10–0.15) | 0.11 (0.09–0.14)b | 0.12 (0.09–0.14)c | 0.0001 |
| tHUFAn-6/tHUFAn-3 | 2.70 (2.40–3.08) | 2.63 (2.29–2.92) | 2.58 (2.19–2.92)b | 2.52 (2.15–2.88)c | 0.0001 |
Fatty acids are presented as relative percentages of total fatty acids (%, wt: wt) and median (25th–75th percentiles). Differences in fatty acid profiles among the four groups, namely participants without disease, with CKD only, with T2D only, and with both CKD and T2D, were assessed using Dunn’s post-hoc test with Bonferroni adjustment whenever the Kruskal-Wallis test indicated a significant overall difference. Pairwise comparisons are denoted as aWithout disease vs. CKD only; bWithout disease vs. T2D only; cWithout disease vs. CKD + T2D; dCKD only vs. T2D only; eCKD only vs. CKD + T2D; fT2D only vs. CKD + T2D. Exact p-values for all comparisons are provided in Fig. 1. Abbreviations: CKD, chronic kidney disease; SFA, saturated fatty acid; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; D5D, ratio of 20:4n-6/20:3n-6; D6D, ratio for 20:3n-6/18:2n-6; omega-3 index, sum of 20:5n-3 + 22:6n-3; lipogenic index, ratio of 16:0/18:2n-6; SCD1-16, ratio of 16:1n-7/16:0; SCD1-18, ratio of 18:1n-9/18:0; AA/LA, ratio of 20:4n-6/18:2n-6; AA/EPA, ratio of 20:4n-6/20:5n-3; AA/(EPA + DHA), ratio of 20:4n-6/(20:5n-3 + 22:6n-3); DPAn6/DHA, ratio of 22:5n-6/22:6n-3; tHUFAn-6/tHUFAn-3, ratio of total highly unsaturated n-6 fatty acids/total highly unsaturated n-3 fatty acids.
Fig. 1.
Distribution of selected red blood cell membrane fatty acids (%) with significant differences across disease categories.
Relationship between red blood cell membrane fatty acids and kidney function
The unadjusted and adjusted associations between fatty acids and kidney function are presented in Table 3. All measured SFAs were independently associated with eGFR (p < 0.004 for all), with lower eGFR being associated with higher percentage of 14:0 (myristic acid), 16:0 (palmitic acid), 20:0 (arachidic acid), 22:0 (behenic acid), 24:0 (lignoceric acid) and total SFAs and lower percentage of 18:0 (stearic acid). eGFR was also positively associated with the 18:1n-7 (cis-vaccenic acid) (p = 0.018), 20:2n-6 (eicosadienoic acid) (p = 0.003), 20:3n-3 (eicosatrienoic acid) (p = 0.040), 20:5n-3 (eicosapentaenoic acid) (p = 0.024), 22:5n-3 (docosapentaenoic acid) (p = 0.003), total n-3 PUFAs (p = 0.009), total n-6 PUFAs (p = 0.026), total PUFA (p < 0.0001), elongase (p < 0.0001) and omega-3 index (p = 0.042) and inversely associated with 24:1n-9 (nervonic acid) (p = 0.001), 18:3n-6 (gamma-linolenic acid) (p = 0.042), 22:2n-6 (docosadienoic acid) (p = 0.016) and lipogenic index (p = 0.016). VIF values for covariates ranged from 1 to 3.
Table 3.
Unadjusted and adjusted association between red blood cell membrane fatty acids and measures of kidney function.
| Fatty acids | Unadjusted | Adjusted | ||||
|---|---|---|---|---|---|---|
| β | 95% CI | p | β | 95% CI | p | |
| Saturated fatty acids | ||||||
| 14:0 (myristic acid) | −106.22 | −155.27 to −57.17 | < 0.0001 | −121.78 | −179.11 to −64.46 | < 0.0001 |
| 16:0 (palmitic acid) | −2.60 | −4.75 to −0.45 | 0.018 | −4.54 | −7.18 to −1.90 | 0.001 |
| 18:0 (stearic acid) | 6.83 | 4.05 to 9.61 | < 0.0001 | 6.12 | 2.57 to 9.67 | 0.001 |
| 20:0 (arachidic acid) | −51.30 | −95.65 to −6.95 | 0.023 | −82.31 | −138.95 to −25.66 | 0.004 |
| 22:0 (behenic acid) | −20.78 | −30.90 to −10.66 | < 0.0001 | −28.93 | −41.36 to −16.51 | < 0.0001 |
| 24:0 (lignoceric acid) | −7.92 | −11.70 to −4.13 | < 0.0001 | −12.44 | −17.15 to −7.73 | < 0.0001 |
| Total SFA | −1.68 | −3.48 to 0.13 | 0.069 | −4.85 | −7.10 to −2.63 | < 0.0001 |
| Monounsaturated fatty acids | ||||||
| 16:1n-7 (palmitoleic acid) | −6.65 | −29.17 to 15.88 | 0.563 | −8.1 | −34.31 to 18.10 | 0.544 |
| 18:1n-7 (cis-vaccenic acid) | 28.19 | 12.99 to 43.40 | < 0.0001 | 21.49 | 3.66 to 39.31 | 0.018 |
| 18:1n-9 (oleic acid) | −2.60 | −5.08 to −0.13 | 0.039 | −0.31 | −3.43 to 2.82 | 0.846 |
| 20:1n-9 (eicosenoic acid) | −47.99 | −134.05 to 38.07 | 0.274 | −64.53 | −173.87 to 44.82 | 0.247 |
| 22:1n-9 (erucic acid) | −113.76 | −304.48 to 76.96 | 0.242 | −148.54 | −367.48 to 70.39 | 0.183 |
| 24:1n-9 (nervonic acid) | −3.56 | −8.02 to 0.90 | 0.117 | −9.21 | −14.71 to −3.71 | 0.001 |
| Total n-7 MUFA | 12.46 | 1.62 to 23.31 | 0.024 | 8.84 | −3.85 to 21.52 | 0.172 |
| Total n-9 MUFA | −3.14 | −5.38 to −0.90 | 0.006 | −2.71 | −5.45 to 0.03 | 0.053 |
| Total MUFA | −2.24 | −4.32 to −0.16 | 0.035 | −1.89 | −4.39 to 0.60 | 0.137 |
| Polyunsaturated fatty acids | ||||||
| 18:2n-6 (linoleic acid) | 1.31 | −0.48 to 3.09 | 0.151 | 1.40 | −0.81 to 3.62 | 0.214 |
| 18:3n-3 (alpha-linolenic acid) | 39.17 | −90.65 to 168.99 | 0.554 | 40.6 | −123.07 to 204.27 | 0.626 |
| 18:3n-6 (gamma-linolenic acid) | −93.02 | −231.88 to 45.84 | 0.189 | −158.8 | −311.66 to 5.94 | 0.042 |
| 20:2n-6 (eicosadienoic acid) | 81.55 | 31.57 to 131.54 | 0.001 | 92.79 | 31.12 to 154.47 | 0.003 |
| 20:3n-3 (eicosatrienoic acid) | 224.21 | 64.98 to 383.44 | 0.006 | 199.35 | 9.44 to 389.26 | 0.040 |
| 20:3n-6 (dihomo-gamma-linolenic acid) | −1.92 | −9.20 to 5.36 | 0.604 | −2.22 | −11.33 to 6.89 | 0.633 |
| 20:4n-6 (arachidonic acid) | 1.21 | −0.56 to 2.97 | 0.181 | 2.03 | −0.19 to 4.25 | 0.072 |
| 20:5n-3 (eicosapentaenoic acid) | 5.21 | −5.63 to 16.05 | 0.346 | 16.95 | 2.23 to 31.67 | 0.024 |
| 22:2n-6 (docosadienoic acid) | −91.72 | −194.57 to 11.13 | 0.080 | −167.63 | −304.02 to 31.25 | 0.016 |
| 22:4n-6 (adrenic acid) | 5.61 | 1.24 to 9.97 | 0.012 | 0.38 | −5.24 to 5.99 | 0.895 |
| 22:5n-3 (docosapentaenoic acid, DPAn-3) | 5.39 | −2.03 to 12.81 | 0.154 | 13.67 | 4.53 to 22.81 | 0.003 |
| 22:5n-6 (docosapentaenoic acid, DPAn-6) | 12.13 | −4.22 to 28.49 | 0.146 | 2.528 | −18.39 to 23.44 | 0.813 |
| 22:6n-3 (docosahexaenoic acid) | −1.47 | −3.85 to 0.91 | 0.224 | 2.90 | −0.26 to 6.05 | 0.072 |
| Total n-3 PUFA | −0.32 | −2.18 to 1.54 | 0.737 | 3.22 | 0.81 to 5.64 | 0.009 |
| Total n-6 PUFA | 1.74 | 0.50 to 2.98 | 0.006 | 1.81 | 0.21 to 3.40 | 0.0260 |
| Total PUFA | 1.72 | 0.43 to 3.01 | 0.009 | 3.22 | 1.63 to 4.81 | < 0.0001 |
| Estimated desaturase indices, elongase indices and ratios | ||||||
| Delta-5 desaturase (D5D) | 0.75 | −0.41 to 1.90 | 0.206 | 0.78 | −0.69 to 2.26 | 0.296 |
| Delta-6 desaturase (D6D) | −43.79 | −110.81 to 23.23 | 0.200 | −46.26 | −130.12 to 37.60 | 0.279 |
| Stearoyl-Coenzyme A desaturase-1-16 (SCD1-16) | −126.58 | −646.64 to 393.48 | 0.633 | −126.02 | −729.01 to 476.97 | 0.682 |
| Stearoyl-Coenzyme A desaturase-1-18 (SCD1-18) | −60.75 | −90.16 to −31.35 | < 0.0001 | −37.16 | −75.30 to 0.97 | 0.056 |
| Elongase | 83.86 | 46.42 to 121.28 | < 0.0001 | 99.67 | 52.79 to 146.56 | < 0.0001 |
| Omega-3 index | −0.94 | −3.04 to 1.16 | 0.378 | 2.90 | 0.11 to 5.70 | 0.042 |
| Lipogenic index | −9.51 | −17.55 to −1.48 | 0.020 | −12.22 | −22.19 to −2.25 | 0.016 |
| Tn-6/Tn-3 | 2.01 | −1.21 to 5.23 | 0.221 | −2.84 | −6.83 to 1.15 | 0.163 |
| AA/LA | −2.94 | −11.80 to 5.91 | 0.514 | −1.10 | −12.53 to 10.33 | 0.850 |
| AA/EPA | −0.01 | −0.21 to 0.19 | 0.898 | −0.21 | −0.47 to 0.05 | 0.108 |
| AA/(EPA + DHA) | 2.89 | −1.06 to 6.83 | 0.151 | −1.84 | −6.73 to 3.05 | 0.461 |
| DPAn-6/DHA | 50.12 | −5.50 to 105.73 | 0.077 | −18.41 | −87.84 to 51.02 | 0.603 |
| tHUFAn-6/tHUFAn-3 | 2.73 | −1.97 to 7.43 | 0.255 | −4.19 | −10.07 to 1.70 | 0.163 |
Data represents β-coefficient, 95% confidence interval and p-value. Models are adjusted for age, sex, waist circumference, BMI, SBP, HbA1c, TG, LDL-C, CRP, prevalent T2D, and the interaction between T2D and the various fatty acids. Abbreviations: Abbreviations: eGFR, estimated glomerular filtration rate; SFA, saturated fatty acid; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; D5D, ratio of 20:4n-6/20:3n-6; D6D, ratio of 20:3n-6/18:2n-6; omega-3 index, sum of 20:5n-3 + 22:6n-3; lipogenic index, ratio of 16:0/18:2n-6; SCD1-16, ratio of 16:1n-7/16:0; SCD1-18, ratio of 18:1n-9/18:0; AA/LA, ratio of 20:4n-6/18:2n-6; AA/EPA, ratio of 20:4n-6/20:5n-3; AA/(EPA + DHA), ratio of 20:4n-6/(20:5n-3 + 22:6n-3); DPAn6/DHA, ratio of 22:5n-6/22:6n-3; tHUFAn-6/tHUFAn-3, ratio of total highly unsaturated n-6 fatty acids/total highly unsaturated n-3 fatty acids.
Relationship between red blood cell membrane fatty acids and prevalent CKD
The unadjusted and adjusted associations between individual fatty acids and indices that were significantly associated with prevalent CKD are presented in Table 4, with the significant interactions by T2D status graphically presented in Fig. 2. Higher 18:2n-6 (linoleic acid), total n-6 PUFA, and total PUFA were associated with reduced odds of prevalent CKD (p < 0.041 for all), with the associations between total n-6 PUFA, and total PUFA modulated by T2D status (interaction: p < 0.070 for all). A higher lipogenic index was also associated with increased odds of prevalent CKD (p = 0.0015) (Table 4). In the stratified analyses by T2D status, among participants without T2D, higher percentages of total n-6 PUFA [OR (95%CI): 0.87 (0.76–0.98); p = 0.047] and total PUFA [OR (95%CI): 0.89 (0.77–0.99); p = 0.049] were associated with lower odds of prevalent CKD. In contrast, among participants with T2D, the percentages of total n-6 PUFA [OR (95%CI): 1.02 (0.90–1.17); p = 0.726] and total PUFA [OR (95%CI): 1.04 (0.90–1.21); p = 0.575] were not statistically significant.
Table 4.
Unadjusted and adjusted association between red blood cell membrane fatty acids and indices and prevalent CKD.
| Unadjusted | Adjusted | |||||
|---|---|---|---|---|---|---|
| Odds ratio | 95% CI | p | Odds ratio | 95% CI | p | |
| Polyunsaturated fatty acids | ||||||
| 18:2n-6 (linoleic acid) | 0.91 | 0.81 to 1.01 | 0.086 | 0.81 | 0.67 to 0.99 | 0.035 |
| Interaction (18:2n-6*T2D) | – | – | – | 1.22 | 0.94 to 1.58 | 0.129 |
| Total n-6 PUFA | 0.91 | 0.84 to 0.98 | 0.014 | 0.86 | 0.76 to 0.98 | 0.025 |
| Interaction (total n-6 PUFA*T2D) | – | – | – | 1.19 | 0.99 to 1.42 | 0.064 |
| Total PUFA | 0.99 | 0.91 to 1.07 | 0.712 | 0.88 | 0.77 to 0.99 | 0.041 |
| Interaction (total PUFA*T2D) | – | – | – | 1.19 | 0.99 to 1.43 | 0.070 |
| Indices | ||||||
| Lipogenic index | 1.66 | 1.01 to 2.74 | 0.046 | 2.73 | 1.22 to 6.12 | 0.015 |
| Interaction (lipogenic index*T2D) | – | – | – | 0.42 | 0.14 to 1.27 | 0.124 |
Data represents odds ratio (OR), 95% confidence interval and p-value. Models adjusted for age, sex, waist circumference, HbA1c, TG, prevalent T2D, and the interaction between T2D and the various fatty acids. Only the individual fatty acids and indices that were significantly associated with CKD or had a significant interaction with the glycaemic categories are presented in the table. Normoglycaemia was used as reference. Abbreviations: lipogenic index, ratio of 16:0/18:2n-6; PUFA, polyunsaturated fatty acids.
Fig. 2.
Adjusted association between red blood cell membrane fatty acids and estimated indices [(A total n-6 PUFA, and B) total PUFA] and prevalent CKD, by T2D status. Data are presented as predictive margins for those without (blue line) and with (red line) T2D.
Discussion
This study characterised the profile of RBC membrane fatty acids in individuals with CKD, both with and without T2D, and identified specific fatty acid profiles associated with these conditions. A higher content of PUFAs [in particular, 18:2n-6 (linoleic acid), total n-6 PUFA and total PUFA] were associated with reduced odds of prevalent CKD. A key finding from our study, however, was that T2D status modified the association between total n-6 PUFA, total PUFA and prevalent CKD, such that the “protective” effect of the PUFAs was only seen in those without T2D.
Our findings agree with previous observational studies which suggest that a higher PUFA content is associated with better kidney function and reduced prevalent CKD, mainly explained through its anti-inflammatory properties29,30. Although the role of n-3 PUFAs in CKD development and progression has been extensively studied31,32, the association between n-6 PUFA content and the preservation of kidney function and disease are lacking, with only a few published studies. In a study of elderly participants (≥ 65 years old), those with a higher n-3 and n-6 PUFA content, including 18:2n-6 (linoleic acid), had a superior preservation of kidney function33. Similar to our study, a Japanese study showed that lower dietary n-3 intake and higher Tn-6/Tn-3 ratio were associated with CKD in middle-aged and older people (≥ 40 years old), however, in this study the results was true for those with diabetes but not for those without diabetes34. Other investigators have shown the opposite to our findings, in that a higher n-6 PUFA intake is associated with increased CKD risk in participants without T2D at high CVD risk29. Furthermore, studies conducted in participants with known T2D showed that a lower intake of 18:2n-6 (linoleic acid) and 18:3n-3 (α-linolenic acid) was associated with CKD in this population35,36. Thus, the observed association between higher relative percentage of PUFAs and reduced odds of CKD in individuals without T2D found in our study may reflect differences in the underlying pathophysiology of CKD in those with and without T2D. PUFAs, particularly n-3 fatty acids, exhibit anti-inflammatory effects that can protect kidney function29,30; however, their impact may be more pronounced in individuals without T2D, where chronic low-grade inflammation is less severe compared to those with T2D32. Additionally, individuals without T2D tend to have more stable lipid metabolism, facilitating the effective incorporation of PUFAs into cell membranes to support vascular and kidney health, whereas dyslipidaemia in T2D may impair PUFA utilization37. PUFAs also mitigate oxidative stress, a key contributor to CKD, but in people with T2D, hyperglycaemia-driven oxidative stress may overwhelm these protective mechanisms38. Furthermore, the aetiology of CKD differs between populations with and without T2D. Indeed, for people without T2D, hypertension or primary kidney diseases are predominant causes, while in those with T2D, diabetic nephropathy is a major contributor. Diabetic nephropathy involves hyperglycaemia-induced metabolic changes and advanced glycation end-products that PUFAs may not adequately counteract39. Lastly, PUFAs are precursors to lipid mediators that resolve inflammation and support vascular health, but altered mediator activity in T2D could limit their protective effects40. Although speculative, as it was not evaluated in our study, these findings underscore the complex interplay between PUFAs and CKD pathophysiology and highlight the need for further research to elucidate the mechanisms underlying these differential associations.
Our study has limitations, including the high proportion of female-to-male participation, which precludes sex generalization. However, this is a common trend in South African population studies, and we adjusted for sex in all our analysis. Also, given that this study includes individuals at high-risk for T2D and those with known diabetes, the findings cannot be generalized across populations. The observational study design limited the assessment of the causal association between the RBC membrane fatty acids and CKD and T2D, and as such, further longitudinal studies should be undertaken to assess this in the local population. Furthermore, because interaction tests were evaluated using a more liberal significance threshold (p < 0.10) to improve sensitivity for detecting potential effect modification, these findings should be interpreted cautiously as exploratory and may be subject to an increased risk of Type I error. Although RBC membrane fatty acids reflect long-term dietary intake, the lack of dietary intake data precluded adjustment for potential nutritional factors. Another limitation is that CKD was defined based on a single time point serum and urinary creatinine and albumin assessment and not on repeated measurements, at least three months apart, as per KDIGO guidelines26. With that said, a major strength of our study was that CKD was defined using both eGFR and albuminuria, unlike most other population-based CKD studies in South Africa and Africa in general which rely on eGFR only for CKD classification.
Conclusion
The findings of this study provide insights into the metabolic imbalances associated with CKD and T2D. The observed changes in RBC membrane fatty acid profiles, particularly in MUFAs and PUFAs may be exacerbated by the co-existence of CKD and T2D. These findings highlight the need for further studies to investigate the clinical value of using fatty acid profiles in addition to already existing biomarkers to improve early detection of these conditions. Furthermore, early dietary interventions, such as a balanced intake of both n-6 and n-3 PUFAs to modify the metabolic imbalances associated with CKD and T2D, might improve disease management, facilitate personalised care and improve health outcomes and as such should be investigated further. Overall, addressing the shared metabolic pathways of these diseases through integrated care models and tailored nutritional education programmes can enhance disease management and improve long-term health outcomes, ultimately reducing the economic and health burdens on public health systems.
Acknowledgements
We would like to acknowledge the South African Medical Research Council (SAMRC) for infrastructure and support. The authors would also like to thank Ms Johanna van Wyk for conducting the RBC fatty acid analyses, under supervision of PJvJ, and the entire South African Diabetes Prevention Programme team and collaborators for their continued support.
Abbreviations
- ACR Albumin
to-creatinine ratio
- AUC
Area under the curve
- BMI
Body mass index
- CKD
Chronic kidney disease
- CKD-EPI
CKD Epidemiology Collaboration
- CVD
Cardiovascular disease
- DBP
Diastolic blood pressure
- eGFR
Estimated glomerular filtration rate
- FAMEs
Fatty acid methyl esters
- FBG
Fasting blood glucose
- GLC
Gas-liquid chromatography
- HbA1c
Glycated haemoglobin
- HDL-C
High-density lipoprotein
- hsCRP
High-sensitivity C-reactive protein
- HUFA
Highly unsaturated fatty acids
- LDL-C
Low-density lipoprotein
- n-3
Omega-3
- n-6
Omega-6
- OGTT
Oral glucose tolerance test
- PUFA
Polyunsaturated fatty acids
- RBC
Red blood cell
- SA-DPP
South African Diabetes Prevention Programme
- SAMRC
South African Medical Research Council
- SBP
Systolic blood pressure
- SCD
Stearoyl-Coenzyme A desaturase
- SFA
Saturated fatty acids
- T2D
Type 2 diabetes mellitus
- TC
Total cholesterol
- TG
Triglycerides
- TLC
Thin-layer chromatography
- TPL
Total phospholipid
- WHO
World Health Organization
Author contributions
CG: Conceptualization, Data curation, Formal analysis, Visualization, Writing – original draft, Writing – review and editing. JH: Data curation, Project administration, Writing – review and editing. NUN: Conceptualization, Writing – review and editing. DDM: Conceptualization, Writing – review and editing. NP: Funding acquisition, Resources, Writing – review and editing. APK: Conceptualization, Funding acquisition, Resources, Writing – review and editing. PJvJ: Conceptualization, Investigation, Supervision, Validation, Writing – review and editing.
Funding
The development of the SA-DPP has been supported by APK, from the Non-Communicable Diseases Research Unit of the South African Medical Research Council (2017–2021) [Baseline funded].
Data availability
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Declarations
Competing interests
The authors declare no competing interests.
Ethics approval and consent to participate
The study was conducted in accordance with the Declaration of Helsinki and approved by the Research Ethics Committee of the South African Medical Research Council (SAMRC) (approval no. EC018-7/2015). All participants voluntary signed written informed consent after the procedures were fully explained in the language of their choice.
Consent for publication
Not applicable.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Jager, K. J. et al. A single number for advocacy and communication-worldwide more than 850 million individuals have kidney diseases. Nephrol. Dial Transpl.34, 1803–1805. 10.1093/ndt/gfz174 (2019). [DOI] [PubMed] [Google Scholar]
- 2.Global Burden of Diseases Chronic Kidney Disease Collaboration. Global, regional, and National burden of chronic kidney disease, 1990–2017: a systematic analysis for the global burden of disease study 2017. Lancet395, 709–733. 10.1016/S0140-6736(20)30045-3 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Jha, V. et al. Chronic kidney disease: global dimension and perspectives. Lancet382, 260–272. 10.1016/S0140-6736(13)60687-X (2013). [DOI] [PubMed] [Google Scholar]
- 4.Kahn, S. E., Cooper, M. E. & Del Prato, S. Pathophysiology and treatment of type 2 diabetes: perspectives on the past, present, and future. Lancet383, 1068–1083. 10.1016/S0140-6736(13)62154-6 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Vaziri, N. D. & Norris, K. Lipid disorders and their relevance to outcomes in chronic kidney disease. Blood Purif.31, 189–196. 10.1159/000321845 (2011). [DOI] [PubMed] [Google Scholar]
- 6.Tsimihodimos, V., Dounousi, E. & Siamopoulos, K. C. Dyslipidemia in chronic kidney disease: an approach to pathogenesis and treatment. Am. J. Nephrol.28, 958–973. 10.1159/000144024 (2008). [DOI] [PubMed] [Google Scholar]
- 7.Herman-Edelstein, M., Scherzer, P., Tobar, A., Levi, M. & Gafter, U. Altered renal lipid metabolism and renal lipid accumulation in human diabetic nephropathy. J. Lipid Res.55, 561–572. 10.1194/jlr.P040501 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Krauss, R. M. Lipids and lipoproteins in patients with type 2 diabetes. Diabetes Care. 27, 1496–1504. 10.2337/diacare.27.6.1496 (2004). [DOI] [PubMed] [Google Scholar]
- 9.Calder, P. C. N-3 polyunsaturated fatty acids and inflammation: from molecular biology to the clinic. Lipids38, 343–352. 10.1007/s11745-003-1068-y (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Brenna, J. T., Plourde, M., Stark, K. D., Jones, P. J. & Lin, Y. H. Best practices for the design, laboratory analysis, and reporting of trials involving fatty acids. Am. J. Clin. Nutr.108, 211–227. 10.1093/ajcn/nqy089 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Dougherty, R. M., Galli, C., Ferro-Luzzi, A. & Iacono, J. M. Lipid and phospholipid fatty acid composition of plasma, red blood cells, and platelets and how they are affected by dietary lipids: a study of normal subjects from Italy, Finland, and the USA. Am. J. Clin. Nutr.45, 443–455. 10.1093/ajcn/45.2.443 (1987). [DOI] [PubMed] [Google Scholar]
- 12.Hill, J. et al. Findings from Community-Based screenings for type 2 diabetes mellitus in at risk communities in cape Town, South africa: A pilot study. Int. J. Environ. Res. Public Health. 1710.3390/ijerph17082876 (2020). [DOI] [PMC free article] [PubMed]
- 13.Matsha, T. E. et al. Three-year’s changes in glucose tolerance status in the Bellville South cohort: rates and phenotypes associated with progression. Diabetes Res. Clin. Pract.99, 223–230. 10.1016/j.diabres.2012.10.018 (2013). [DOI] [PubMed] [Google Scholar]
- 14.Gillies, C. L. et al. Pharmacological and lifestyle interventions to prevent or delay type 2 diabetes in people with impaired glucose tolerance: systematic review and meta-analysis. BMJ (Clinical Res. ed.). 334, 299. 10.1136/bmj.39063.689375.55 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Parker, D. R., Evangelou, E. & Eaton, C. B. Intraclass correlation coefficients for cluster randomized trials in primary care: the cholesterol education and research trial (CEART). Contemp. Clin. Trials. 26, 260–267. 10.1016/j.cct.2005.01.002 (2005). [DOI] [PubMed] [Google Scholar]
- 16.Mayige, M. Derivation and Validation of a Simple Risk Score for Undiagnosed Diabetes for Tanzania and Other African Populations. PhD thesis, Newcastle University, (2014).
- 17.Royce, N. et al. HbA1c comparable to fasting glucose in the external validation of the African diabetes risk score and other established risk prediction models in black South Africans. BMC Endocr. Disord. 24, 213. 10.1186/s12902-024-01735-w (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Alberti, K. G., Zimmet, P. & Shaw, J. Metabolic syndrome–a new world-wide definition. A consensus statement from the international diabetes federation. Diabet. Medicine: J. Br. Diabet. Association. 23, 469–480. 10.1111/j.1464-5491.2006.01858.x (2006). [DOI] [PubMed] [Google Scholar]
- 19.Alberti, K. G. & Zimmet, P. Z. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet. Medicine: J. Br. Diabet. Association. 15, 539–553. 10.1002/(SICI)1096-9136(199807)15:7%3C539::AID-DIA668%3E3.0.CO;2-S (1998). [DOI] [PubMed] [Google Scholar]
- 20.Friedewald, W. T., Levy, R. I. & Fredrickson, D. S. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin. Chem.18, 499–502 (1972). [PubMed] [Google Scholar]
- 21.Folch, J., Lees, M. & Sloane Stanley, G. H. A simple method for the isolation and purification of total lipides from animal tissues. J. Biol. Chem.226, 497–509 (1957). [PubMed] [Google Scholar]
- 22.Chimhashu, T. et al. Sensitivity of fatty acid desaturation and elongation to plasma zinc concentration: a randomised controlled trial in Beninese children. Br. J. Nutr.119, 610–619. 10.1017/S000711451700366X (2018). [DOI] [PubMed] [Google Scholar]
- 23.Mentoor, I., Engelbrecht, A. M., van de Vyver, M., van Jaarsveld, P. J. & Nell, T. The paracrine effects of adipocytes on lipid metabolism in doxorubicin-treated triple negative breast cancer cells. Adipocyte10, 505–523. 10.1080/21623945.2021.1979758 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kroger, J. et al. Erythrocyte membrane phospholipid fatty acids, desaturase activity, and dietary fatty acids in relation to risk of type 2 diabetes in the European prospective investigation into cancer and nutrition (EPIC)-Potsdam study. Am. J. Clin. Nutr.93, 127–142. 10.3945/ajcn.110.005447 (2011). [DOI] [PubMed] [Google Scholar]
- 25.Levey, A. S. et al. A new equation to estimate glomerular filtration rate. Ann. Intern. Med.150, 604–612. 10.7326/0003-4819-150-9-200905050-00006 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Kidney Disease. Improving global Outcomes. KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int.3, 19–62. 10.1038/kisup.2012.64 (2013). [DOI] [PubMed]
- 27.Kidney Disease: Improving global Outcomes. KDIGO 2024 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int.105, S117–S314. 10.1016/j.kint.2023.10.018 (2024). [DOI] [PubMed]
- 28.World Health Organization. A global brief on Hypertension: Silent killer, global public health crisis. (2013).
- 29.Diaz-Lopez, A. et al. Cross-sectional associations between macronutrient intake and chronic kidney disease in a population at high cardiovascular risk. Clin. Nutr.32, 606–612. 10.1016/j.clnu.2012.10.013 (2013). [DOI] [PubMed] [Google Scholar]
- 30.Tofte, N. et al. Plasma metabolomics identifies markers of impaired renal function: A Meta-analysis of 3089 persons with type 2 diabetes. J. Clin. Endocrinol. Metab.105, 2275–2287. 10.1210/clinem/dgaa173 (2020). [DOI] [PubMed] [Google Scholar]
- 31.Syren, M. L. et al. The polyunsaturated fatty acid balance in kidney health and disease: A review. Clin. Nutr.37, 1829–1839. 10.1016/j.clnu.2017.11.019 (2018). [DOI] [PubMed] [Google Scholar]
- 32.Ong, K. L. et al. Association of Omega 3 polyunsaturated fatty acids with incident chronic kidney disease: pooled analysis of 19 cohorts. BMJ (Clinical Res. ed.). 380, e072909. 10.1136/bmj-2022-072909 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Lauretani, F. et al. Plasma polyunsaturated fatty acids and the decline of renal function. Clin. Chem.54, 475–481. 10.1373/clinchem.2007.095521 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hara, A. et al. Distinct associations between dietary omega-3 and omega-6 fatty acids intake with chronic kidney disease in adults with and without diabetes: A cross-sectional study. Nutr. (Burbank Los Angeles Cty. Calif). 115, 112156. 10.1016/j.nut.2023.112156 (2023). [DOI] [PubMed] [Google Scholar]
- 35.Cardenas, C. et al. Polyunsaturated fatty acid consumption May play a role in the onset and regression of microalbuminuria in well-controlled type 1 and type 2 diabetic people: a 7-year, prospective, population-based, observational multicenter study. Diabetes Care. 27, 1454–1457. 10.2337/diacare.27.6.1454 (2004). [DOI] [PubMed] [Google Scholar]
- 36.Dos Santos, A. L. T. et al. Low linolenic and Linoleic acid consumption are associated with chronic kidney disease in patients with type 2 diabetes. PloS One. 13, e0195249. 10.1371/journal.pone.0195249 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Jin, Q. et al. Circulating metabolomic markers linking diabetic kidney disease and incident cardiovascular disease in type 2 diabetes: analyses from the Hong Kong diabetes biobank. Diabetologia67, 837–849. 10.1007/s00125-024-06108-5 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Han, E. et al. Effects of Omega-3 fatty acid supplementation on diabetic nephropathy progression in patients with diabetes and hypertriglyceridemia. PloS One. 11, e0154683. 10.1371/journal.pone.0154683 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Chewcharat, A., Chewcharat, P., Rutirapong, A. & Papatheodorou, S. The effects of omega-3 fatty acids on diabetic nephropathy: A meta-analysis of randomized controlled trials. PloS One. 15, e0228315. 10.1371/journal.pone.0228315 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Fassett, R. G., Gobe, G. C., Peake, J. M. & Coombes, J. S. Omega-3 polyunsaturated fatty acids in the treatment of kidney disease. Am. J. Kidney Dis.56, 728–742. 10.1053/j.ajkd.2010.03.009 (2010). [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.