Table 1.
Summary of the results of the selected studies about the role of cholesterol and fatty acids in the pathogenesis of ASDs.
| Study | Design | Sample | Intervention | Biomarkers | Findings in ASDs |
|---|---|---|---|---|---|
| Tierney et al., 2001 [38] | Cross-sectional study | 28 SLOS children | / | Plasma TC | SLOS, which presents with hypocholesterolemia, is associated with the presence of autism. |
| Vancassel et al., 2001 [39] | Cross-sectional study | 15 ASD subjects: Mean age 8.4 years 73.33% males 18 ID subjects: Mean age 8.8 years 72.22% males |
/ | Plasma FA | A marked reduction in the levels of DHA (23%) was shown in ASD, resulting in significantly lower levels of n − 3 PUFA (20%; p = 0.032) and a significant increase in the ω6/ω3 ratio (25%; p = 0.039). |
| Goldenberg et al., 2002 [40] | Cross-sectional study | 45 SLOS subjects | / | Plasma TC, 7-DHC, 8-DHC | Plasma hypocholesterolemia correlates significantly with clinical severity (1–2.5% increased risk). |
| Bell et al., 2004 [41] | Cross-sectional study | ASD children HC |
/ | Plasma and RBC membranes FA composition | Higher frequency of FAD was found in patients with autism and Asperger syndrome compared to HC. Patients with regressive autism had higher percentages of stearic acid, linoleic acid and total saturates in their RBC membranes than HC, as well as higher lignoceric acid, docosapentaenoic acid, nervonic acid and AA/EPA ratio; these latter characteristics are shared by regressive autism and Asperger syndrome. Patients with regressive autism showed lower oleic acid and AA values and patients both with regressive autism and with Asperger syndrome presented lower docosapentaenoic acid and total n − 3 PUFAs compared to HC. |
| Dziobek et al., 2006 [42] | Cross-sectional study | 22 Asperger children: Mean age 40.8 ± 10.8 years 77.3% male 22 HC: Mean age 44.6 ± 14.8 years 77.3% male |
/ | Plasma TC, LDL-C, TG | Asperger subjects showed statistically significant elevated levels of TC, LDL-C and TG and significant lower levels of HDL compared with HC. After controlling for physical activity, group differences remained significant for TC (p = 0.016) and LDL-C (p = 0.017), but not for HDL-C (p = 0.323) and TG (p = 0.217). |
| Sikora et al., 2006 [43] | Clinical trial | 14 SLOS children: Mean age 7.1 ± 3.5 years |
2-years cholesterol supplementation | Plasma TC, 7-DHC, 8-DHC | Plasma TC, 7-DHC and 8-DHC at baseline and after supplementation did not correlate with the presence or severity of autistic symptoms. |
| Sliwinski et al., 2006 [44] | Cross-sectional study | 16 high-functioning ASD subjects aged 12–18 years: 100% males 22 HC |
/ | Plasma FA | In ASD there was a significant increase in the fraction of DHA and an increase in the total ω3/ω6 ratio. |
| Tierney et al., 2006 [45] | Cross-sectional study | 100 ASD children | / | Plasma TC, 7-DHC | Of 19% of subjects with low TC (<100 mg/dL), 31.5% met criteria for ASD. |
| Meguild et al., 2008 [46] | Clinical trial | 30 ASD children aged 3–11 years: 60% males 30 age- and sex-matched HC |
3-months fish oil supplementation | Plasma FA | First assessment: In ASD group, linolenic acid showed a significant reduction (71%), followed by DHA (65%) and AA (45%), while linoleic acid was the least affected PUFA (32%), compared to HC. Second assessment: 66% of autistic children showed clinical and biochemical improvement (decrease in Childhood Autism Rating Scale scores: p < 0.0001). Plasma PUFA showed increase in plasma levels after supplementation (all p < 0.0001). |
| Bell et al., 2010 [47] | Cross-sectional study | 49 ASD children: Mean age 7.5 ± 3.5 years 89.80% males 39 DD children: Mean age 6.0 ± 3.3 years 89.74% males 52 HC: Mean age 7.5 ± 3.6 years 94.23% males |
/ | Plasma and RBC membranes FA composition | RBC and plasma FA in ASD had an increased AA/EPA ratio. Decreased levels of lignoceric acid and nervonic acid were found in children with DD with respect to ASD. ASD subjects consuming fish oil showed reduced erythrocyte AA, adrenic acid, docosapentanaeoic acid and total n − 6 FA and increased EPA, DHA and total n − 3 FA along with reduced n − 6/n − 3 and AA/EPA ratios. |
| Wiest et al., 2009 [48] | Cross-sectional study | 153 ASD children: Mean age 3.75 years 88.89% males 97 HC: Mean age 3.42 years 78.36% males |
/ | Plasma lipidomics | Levels of CE, FS, PC and FA did not differ between ASD and HC. Total TG, lysophosphatidylcholine, PE and diglyceride levels were higher in the ASD than the HC group (respectively, p = 0.006; p = 0.001; p = 0.007; p = 0.005). DHA was significantly decreased in ASD with respect to HC (p < 0.05). |
| Kim et al., 2010 [49] | Cross-sectional study | 29 ASD children: Mean age, 10.1 ± 1.3 years 100% male 29 age- and sex-matched HC |
/ | Plasma TC, HDL-C, LDL-C, LDL/HDL ratio, TG | The mean TG level was significantly higher, whereas the mean HDL-C level was significantly lower in cases as compared to controls (respectively, p = 0.01; p = 0.003). There was no significant difference in TC and LDL-C levels between ASD and HC. The LDL/HDL ratio was significantly higher in ASD compared to HC (p = 0.03). Autism was associated with plasma TG, HDL and the LDL/HDL ratio (respectively, p = 0.01; p = 0.0003; p = 0.04). |
| El-Ansary et al., 2011 [50] | Cross-sectional study | 25 ASD children aged 4–12 years 16 HC aged 4–11 years |
/ | Plasma FA, PE, PS and PC | ASD patients showed higher LA/AA, ALA/DHA, AA/DHA, EPA/AA ratios compared to HC (respectively, p = 0.034; p = 0.004; p = 0.000; p = 0.000). ASD patients showed higher plasma PE, PS and PC compared to HC (respectively, p = 0.002; p = 0.000; p = 0.000). |
| El-Ansary et al., 2011 [51] | Cross-sectional study | 26 ASD children aged 4–12 years 26 age- and sex-matched HC |
/ | Plasma FA | ASD patients showed an increase in acetic, valeric, hexanoic and stearidonic acid compared to HC (respectively p = 0.000; p = 0.000; p = 0.000; p = 0.009). ASD subjects show different percentage decrease of saturated acids (propionic, butyric, caprylic, decanoic, lauric, palmitic and stearic) together with mono (oleic) and polyunsaturated fatty acids (arachidic, α-linolenic, DHA, linoleic, γ-linolenic, AA, elaidic) compared to HC (respectively, p = 0.000; p = 0.028; p = 0.000; p = 0.000; p = 0.000; p = 0.0037; p = 0.000; p = 0.000; p = 0.000; p = 0.0045; p = 0.000; p = 0.023; p = 0.000; p = 0.000; p = 0.000). |
| Schengrund et al., 2012 [52] | Cross-sectional study | 16 ASD children: Mean age 5.13 years 93.75% male 20 HC: Mean age 6.0 years 40% male |
/ | RBC membranes composition: GM1 and cholesterol | ASDs children have less cholesterol and more GM1 in their RBC membranes than HC (respectively, p = 0.012; p = 0.019). |
| Fong et al., 2013 [53] | Cross-sectional study | 102 ASD children: Mean age 3.5 years 52.0% male 102 HC: Mean age 3.9 years 52.0% male |
/ | Plasma CS | Comparison of normal and autistic children showed no statistically significant difference in plasma CS level. |
| Ghezzo et al., 2013 [54] | Cross-sectional study | 25 ASD children: Mean age 7.8 ± 2.23 years 80.95% males 23 HC: Mean age 7.6 ±1.96 years 70% males |
/ | RBC membranes FA composition | Alteration in RBC FA membrane profile (increase in monounsaturated fatty acids, decrease in EPA and DHA with a consequent increase in ω6/ω3 ratio) were found in ASD compared to HC (respectively, p < 0.01; p < 0.05; p < 0.01). |
| Brown et al., 2014 [55] | Cross-sectional study | 19 ASD children 23 HC (siblings) |
/ | Plasma FA | Those infants not breastfed (with colostrum) within the first hour of life and who had a history of FAD symptoms were more likely to have an ASD diagnosis. |
| Moses et al., 2014 [56] | Cross-sectional study | 80 adults with ASD 77 adults with ID 828 HC |
/ | Plasma TC | TC levels of people with ASD and ID were significantly lower than those of HC (p < 0.001) but after adjusting for gender, age and BMI and using Bonferroni correction, the significance was lost. |
| Brigandi et al., 2015 [57] | Cross-sectional study | 121 ASD subjects aged 3–17 years 110 HC aged 3–17 years |
/ | Plasma and RBC membranes FA composition | The percentage of total PUFA was lower in ASD than in HC; levels of AA and DHA were particularly decreased (p < 0.001). |
| Esparham et al., 2015 [58] | Cross-sectional study | 7 ASD children aged 7–18 years: 71.43% males |
/ | Plasma FA | An abnormal level of α-linolenic, linoleic acid and high levels of DHA were found, as well as an elevated ω6/ω3 ratio. |
| Jory, 2015 [59] | Cross-sectional study | 11 ASD children: Mean age 3.05 ± 0.79 years 72.73% males 15 HC: Mean age 3.87 ± 1.06 years 40% males |
/ | Plasma and RBC membranes FA composition | Children with ASD demonstrated lower RBC DHA, EPA, AA and ω3/ω6 ratios (respectively, p < 0.0003; p < 0.03; p < 0.002; p < 0.001). They also demonstrated lower plasma DHA, AA and linoleic acid levels (respectively, p < 0.02; p < 0.05; p < 0.02). |
| Mostafa and Al-Ayadi, 2015 [60] | Cross-sectional study | 100 ASD children: Mean age 6.22 ± 2.1 years 78% males 100 HC: Mean age 5.96 ± 2 years 78% males |
/ | Plasma FA and carnitine | Reduced levels of plasma carnitine and plasma DHA, AA, linolenic and linoleic acids were found in 66%, 62%, 60%, 43% and 38%, respectively of ASD children. 54% of ASD patients had elevated ω6/ω3 ratio. ASD patients with GI manifestations had significantly increased percentage of reduced serum carnitine (91.7%) and plasma DHA levels (87.5%) than HC (respectively, 42.3%; 38.5%), (respectively, p < 0.001; p < 0.001). |
| Yui et al., 2016 [61] | Cross-sectional study | 28 ASD subjects: Mean age 13.5 ± 4.6 years 71.43% males 21 HC: Mean age 13.9 ± 5.7 years 71.43% males |
/ | Plasma FA | Plasma EPA, DHA and arachidic acid levels and plasma DHA/AA and EPA/AA ratios were significantly higher in ASD compared to HC (respectively, p = 0.02; p = 0.03; p = 0.04; p = 0.0002; p = 000). Plasma AA and adrenic acid were significantly lower in ASD compared to HC (respectively, p = 0.05; p = 0.04). |
| Yui et al., 2016 [62] | Cross-sectional study | 30 ASD subjects: Mean age 13.6 ± 4.3 years 33.33% males 20 HC: Mean age, 13.2 ± 5.4 years 30% males |
/ | Plasma FA | The plasma levels of EPA and the plasma ratios of EPA/AA and DHA/AA were significantly higher (respectively, p = 0.1; p = 000; p = 0.000), while the plasma levels of AA and metabolites, such as adrenic acid, were significantly lower in ASD compared to HC (respectively, p = 0.01; p = 0.0; p = 0.004). |
| Parletta et al., 2016 [63] | Cross-sectional study | 85 ASD children 401 ADHD children 79 HC |
/ | Plasma FA | Children with ADHD and ASD had lower DHA, EPA and AA, higher AA/EPA ratio and lower ω3/ω6 than controls (p < 0.001 except AA between ADHD and controls: p = 0.047). Children with ASD had lower DHA, EPA and AA than children with ADHD (p < 0.001). Childhood Autism Rating Scale scores correlated significantly with DHA, EPA and AA (respectively, p = 0.002; p = 0.038; p = 0.021). |
| Puig-Alcatraz et al., 2016 [64] | Cross-sectional study | 26 ASD children aged 4–13 years 23 HC aged 4–12 years |
/ | Urinary adipic acid, suberic acid | No increase in the concentration of adipic acid or suberic in children with ASD compared to HC. The increase in adipic acid concentration was significantly and indirectly correlated with the severity of the deficit in socialization and communication skills in ASD children. |
| Wang et al., 2016 [65] | Cross-sectional study | 73 ASD children: Mean age 4.6 ± 0.8 years 80.82% males 63 HC: Mean age 4.1 ± 0.7 years 80.95% males |
/ | Plasma metabolomics | ASD was associated with 2 metabolites: sphingosine 1-phosphate and DHA (respectively, p < 0.001; p < 0.001). |
| Yui et al., 2016 [66] | Cross-sectional study | 30 ASD subjects: Mean age 13.0 years 20 sex-matched HC: Mean age 13.6 years |
/ | Plasma FA | ASD had significantly higher plasma DHA/AA and EPA/AA ratios compared to HC. The plasma ceruloplasmin levels in ASD were significantly reduced compared to HC. Multiple linear regression demonstrated that plasma DHA/AA ratio was a fitting model for distinguishing ASD from the HC. |
| Cariou et al., 2018 [67] | Cross-sectional study | 839 adult psychiatric patients: group 1: hypobetalipoproteinemia (HBL) Mean age 35 ± 10 years 65% male group 2: non-HBL Mean age 44 ± 14 years 59% male |
/ | Plasma TC, HDL-C, LDL-C, TG | Psychiatric patients with HBL were characterized by a higher frequency of specific developmental disorders (including autism) (p = 0.011). |
| Howsmon et al., 2018 [68] | Cross-sectional study | 63 ASD children: Median age 7.8 years: 49 HC: Median age 10.0 years |
/ | RBC membranes FA composition | FA do not allow for classification at the individual level. |
| Toscano et al., 2018 [69] | Clinical trial | 64 ASD children aged 6–12 years: experimental group: n = 46 control group: n = 18 |
48-week exercise-based intervention | Plasma TC, HDL, LDL | The experimental group showed beneficial effects on metabolic indicators (TC, HDL, LDL), autism traits and parent-perceived quality of life. |
| Benachenhou et al., 2019 [70] | Cross-sectional study | 79 ASD children: Mean age 19.4 ± 12.1 years 81% male 79 HC: Mean age 19.4 ± 12.0 years 81% male |
/ | Plasma TC, HDL-C, TG, LDL-C | TC levels below the 10th centile were associated with a higher rate of ASD-associated ID (OR = 3.33; 95% CI: 1.26–8.00) and anxiety/depression (OR = 4.74; 95% CI: 1.40–15.73). |
| Hassan et al., 2019 [71] | Cross-sectional study | 63 ASD children 63 age- and sex-matched HC |
/ | Plasma TC | The serum levels of TC was significantly lower among ASD when compared with HC (p < 0.05). |
| Blazewicz et al., 2020 [33] | Clinical trial | 57 ASD children: 100% male group 1: low-fat diet (LFD), n = 14 Mean age 16.6 years group 2: gluten-casein-free diet (GF-CF), n = 10 Mean age 17.8 years group 3: regular diet (RD), n = 35 Mean age 17.3 years 36 HC: Mean age 17.6 years 100% male RD |
Different type of diet: LFD, GF-CF, RD | Plasma CRP, TC, HDL-C, TG | First assessment: in ASD subjects compared to HC: increased BMI, CRP and TC/HDL and decreased HDL-C for all types of diets (p < 0.05) increased TG in the group of LFD (p = 0.003) and RD individual (p < 0.001) increased non-HDL-C in the group of GF–CF (p = 0.008) and RD subjects (p < 0.001) Second assessment: increased levels of TC, non HDL-C and TC/HDL and decreased level of HDL-C for all ASD individuals regardless of diets used (p < 0.05) BMI and CRP increased only for individuals on LFD (BMI: p = 0.029; CRP: p < 0.001) and RD (BMI: p < 0.001; CRP: p > 0.001) |
| Usui et al., 2020 [72] | Cross-sectional study | 152 ASD children 122 HC |
/ | Plasma FA, lipoprotein analysis | 48 metabolites were identified in the plasma of ASD children by lipidomics (linoleic acid: p = 0.0133; linolenic acid: p = 0.0141; EPA: p = 0.0147; oleic acid: p = 0.0284; EPA: p = 0.0327; AA: p = 0.0395). Among these, increased FA, such as ω3 and ω6, showed correlations with clinical social interaction score and ASD diagnosis (p < 0.05). Specific reductions of plasma VLDL and APOB in ASD were found by large-scale lipoprotein analysis. |
| Yui et al., 2020 [73] | Cross-sectional study | 11 ASD subjects: Mean age 12.3 ± 5.4 years 27.27% males 7 HC: Mean age 10.0 ± 4.1 years 57.14% males |
/ | Plasma FA, MDA-LDL, superoxide dismutase | Plasma levels of MDA-LDL, EPA, DHA and DHA/AA ratios were significantly higher, while plasma superoxide dismutase levels were significantly lower in ASD than in HC (respectively, p = 0.034; p = 0.000; p = 0.000; p = 0.000; p = 0.006). Multiple linear regression and adaptative Lasso analysis revealed association of increased plasma DHA levels with the Aberrant Behavior Checklists scores and increased plasma MDA-LDL levels. |