Table 2 Associations reported between calprotectin levels, levels of autistic traits, GI symptoms and other biological markers
GI symptoms | ||
6-Item Gastrointestinal Severity Index (6-GSI) questionnaire [58] | Azouz et al. [39] | A moderate correlation was found between the faecal calprotectin and gastrointestinal symptoms as indicated by parent-reported 6-GSI severity (r = 0.471, p = 0.002) in a sample of 40 autistic children aged between 3 to 12 years (Mage = 6.53 ± 2.10) |
Laghi et al. [35] | No significant differences were found in the calprotectin levels of autistic preschoolers (average age: 4.14 ± 1.01 years) without GI symptoms (n = 52, median = 79.27 μg/g, IQR = 131.15 μg/g) and those with GI symptoms (n = 28, median = 69.50 μg/g, IQR = 131.21 μg/g) | |
Full GSI [59], an adaptation of a modified Truelove and Witts Severity Index used for ulcerative colitis clinical trials [60] | Alookaran et al. [38] | No significant differences were found in calprotectin levels in comparisons of controls (n = 20, Mage = 9.8, SD = 3.8), autistic children and adolescents without GI symptoms (parent-reported GSI score of < 7) (n = 10, Mage = 8.1, SD = 3.1) and autistic children with GI symptoms (n = 20, Mage = 9.3, SD = 3.8) aged between 4–16 years |
Rome III criteria for constipation [61] | Strati et al. [43] | No difference in calprotectin levels was reported between 5 constipated (based on the Rome III criteria) and 29 non-constipated autistic participants (Mage = 11.1 ± 6.8). No significant differences were also found in comparisons of calprotectin levels between 11 constipated and 29 non-constipated control participants (Mage = 9.2 ± 7.9) |
Other GI measures | Tomova et al. [44] | A moderate positive correlation was found between calprotectin levels and parent-reported GI scores based on 5 types of GI symptoms (r = 0.36, p < 0.05) in a sample of 63 autistic boys aged 2.8–9.2 years (Mage = 5.00, SEM = 0.20) |
Bramati-Castellarin et al. [57] | A negative association was found between calprotectin levels and an item in a study specific parent-report measure which asked whether or not their child experienced constipation (β = − 1.58, SE = 0.69, p = 0.022) in a sample of 49 autistic children (46 male and 3 female) aged between 3.5 and 8 years | |
Total level of autistic traits | ||
Autism Diagnostic Observation Schedule, Second Edition (ADOS-2) [62] | Tomova et al. [44] | A positive weak correlation was found between ADOS-2 total scores and calprotectin (r = 0.29; p < 0.05) in a sample of 63 autistic boys aged 2.8-9.2 years (Mage = 5.00, SEM = 0.20) |
Childhood Autism Rating Scale (CARS) [63] | Azouz et al. [39] | No significant correlation was found between CARS total scores and calprotectin (r = 0.175, p = 0.280) in 40 autistic children aged between 3 to 12 years (Mage = 6.53 ± 2.10). 57.5% of the children were categorised as having mild to moderate scores and 42.5% were categorised as having severe CARS scores |
Iovene et al. [41] | No significant correlation was found in univariate analyses of calprotectin and total CARS scores (r = 0.278, p = 0.075) in a sample of 47 autistic children (Mage = 6.0 ± 2.8) A significant correlation was found between CARS scores and calprotectin levels in a multivariate analysis that adjusted for sex and the presence or absence of GI symptoms, polymorphic leukocytes and specific strains in cultured samples (Lactobacilli, Anaerobic bacteria and Clostridium) (Rpartial = 0.622; p = 0.0044) | |
Approach Withdrawal Problem Composite-subtest of the Pervasive Developmental Disorder Behaviour Inventory (PDDBI) [64] | Pusponegoro et al. [42] | There was no difference found between the median calprotectin in the control group (108.1 µg/g faeces, range 5.3–3595), the group categorised with ‘mild maladaptive’ autism (100.54 µg/g faeces, range: 5.3–3536) and in the group categorised as having ‘severe maladaptive’ autism (56.37 µg/g faeces, range: 5.3–2778) (p = 0.80) in a sample of 60 healthy controls, 102 children with ‘mild maladaptive’ autism, and 44 with ‘severe maladaptive’ autism aged between 2 to 10 years |
Social affect and reciprocal social interaction challenges | ||
Autism diagnostic interview schedule-revised (ADI-R) Reciprocal social interaction subscale [65] | Babinská et al. [40] | A moderate correlation was found between calprotectin levels and ADI-R reciprocal social interaction subscale (r = 0.30, p < 0.01) in a sample of 87 autistic children and adolescents aged between 2 and 17 (Mage = 7.2 ± 3.8 years) |
Tomova et al. [44] | A weak correlation was found between calprotectin levels and the ADI-R reciprocal social interaction subscale (r = 0.26, p < 0.05) in a sample of 63 autistic boys aged 2.8–9.2 years (Mage = 5.00, SEM = 0.20) | |
ADI-R communication [65] | Babinská et al. [40] | A moderate correlation was found between calprotectin levels and the ADI-R communication subscale (r = 0.38, p < 0.001) in a sample of 87 autistic children and adolescents aged between 2 and 17 (Mage = 7.2 ± 3.8 years) |
ADOS-2 Social Affect subscale [62] | Tomova et al. [44] | A moderate correlation was reported between the social affect subscale of the ADOS-2 and faecal calprotectin (r = 0.32, < 0.010) in a sample of 63 autistic boys aged 2.8–9.2 years (Mage = 5.00, SEM = 0.20) |
ADOS-2 Reciprocal Social interaction subscale [62] | Tomova et al. [44] | A moderate correlation was reported between the reciprocal interaction subscale of the ADOS-2 and faecal calprotectin (r = 0.32, < 0.010) in a sample of 63 autistic boys aged 2.8–9.2 years (Mage = 5.00, SEM = 0.20) |
Restricted and repetitive behaviours and interests | ||
ADI-R restrictive, repetitive behaviours [65] | Babinská et al. [40] | A moderate correlation between calprotectin levels and the ADI-R restricted repetitive behaviours (r = 0.33, p < 0.01) in a sample of 87 autistic children and adolescents aged between 2 and 17 (Mage = 7.2 ± 3.8 years) |
Other measures of autistic traits | Bramati-Castellarin et al. [57] | A positive association was found between levels of calprotectin and parent-reported need for fixed routines (β = 3.23, SE = 0.81, p 0.00009) in a sample of 49 autistic children (46 male and 3 female) aged between 3.5 and 8 years |
Other biological measures | ||
Serum markers | Tomova et al. [44] | A moderate positive correlation was identified between MIP-1β levels and calprotectin levels (r = 0.38, p < 0.05) in a sample of 63 autistic boys aged 2.8–9.2 years (Mage = 5.00, SEM = 0.20) |
Tomova et al. [66] | A weak positive correlation was identified between plasma levels of S100B and faecal calprotectin (r = 0.21, p < 0.05) in 93 autistic children and adolescents aged between 2 and 16 years (Mage = 6.22, SEM = 0.30) | |
Bacterial populations | Tomova et al. [44] | A strong positive correlation was identified between Costridiacae populations and calprotectin (r = 0.5; p < 0.001), in a sample of 63 autistic boys aged 2.8–9.2 years (Mage = 5.00, SEM = 0.20) |
Laghi et al. [35] | The amount of Akkermansia muciniphila was found to have a moderate negative correlation with calprotectin levels above 50 μg/g (r = − 0.32; p = 0.041) in a sample of 80 preschoolers (Age = 4.14 ± 1.01) The amount of Prevotella was strongly associated with calprotectin levels above 200 μg/g (r = 0.75, p = 0.003) The study did not report the proportion of children who had calprotectin levels above 50 μg/g or above 200 μg/g | |
Fungal populations | Iovene et al. [41] | No significant correlation was found in univariate analysis of calprotectin and levels of Candida (r = − 0.019, p = 0.90) in a sample of 47 autistic children (Mage = 6.0 ± 2.8) No significant correlation was found in the subsequent multivariate analyses which adjusted for sex and the presence or absence of GI symptoms, polymorphic leukocytes and specific strains in cultured samples (Lactobacilli, Anaerobic bacteria and Clostridium) |
Intestinal permeability | Iovene et al. [41] | No significant correlations were found in univariate analyses of calprotectin and intestinal permeability as assessed with the lactulose/mannitol test (r = − 0.010, p = 0.593) in a sample of 47 autistic children (Mage = 6.0 ± 2.8) No significant correlation was found in the subsequent multivariate analysis which adjusted for sex and the presence or absence of GI symptoms, polymorphic leukocytes and specific strains in cultured samples (Lactobacilli, Anaerobic bacteria and Clostridium) |