The effects of salvia hispanica (chia seeds) on insulin sensitivity, hematological inflammatory indices and liver function in obese patients with non-alcoholic fatty liver disease (NAFLD)

Abstract

Objective

To examine the effects of chia seeds on insulin sensitivity, hematological inflammatory indices (HIIs), liver function, and steatosis in obese patients with non-alcoholic fatty liver disease (NAFLD).

Methods

This single-blinded randomized controlled clinical trial was carried out on 44 newly diagnosed obese patients with NAFLD. Following liver ultrasonography and dietary assessment, patients randomly allocated into two groups: “Control group” (n = 22) receiving only calorie-restricted diet (CRD) and “Chia group” (n = 22) receiving chia seeds (40 g/day) plus CRD for 8 weeks. After an overnight fasting, serum glucose, insulin, hemoglobin A1c (HbA1c), alanine aminotransferase (ALT), aspartate aminotransferase (AST), ferritin, and complete blood cell count were assessed. Then, insulin sensitivity indices, NAFLD fibrosis score (NFS), and HIIs were estimated.

Results

After 8 weeks, significant inter-group reductions were observed in energy (p = 0.015) and carbohydrate (p = 0.029) intakes, in favor of chia. Moreover, chia plus CRD noticeably improved insulin sensitivity by reducing serum insulin (p = 0.016) and increasing homeostatic model assessment of insulin sensitivity (p < 0.001), after adjusting for confounders. Marked attenuations in NFS (p = 0.020) and AST to ALT ratio (p = 0.038), as well as marginal amelioration in serum aminotransferases was observed in the chia group, while among HIIs, only platelet to lymphocyte ratio (p = 0.027) significantly enhanced in favor of chia.

Conclusion

Chia seeds supplementation could reduce food intake, ameliorate insulin sensitivity, and NAFLD fibrosis score in obese patients with NAFLD.

Introduction

Non-alcoholic fatty liver disease (NAFLD)- the abnormal deposition of fat droplets within hepatocytes in the absence of alcohol abuse or viral hepatitis- is the most prevalent chronic liver disorder affecting up to 38% of the general population [1, 2]. This number is expected to reach approximately 55.4% by 2040, with a notable increase in Asia [2]. Despite the involvement of genetic factors, the “Multiple-hits” theory has presupposed an interplay between insulin resistance (IR), inflammation, oxidative stress, and gut microbiome dysbiosis in the pathogenesis of NAFLD [3]. IR, the predominant trigger of hepatic steatosis, is followed by chronic low-grade inflammation, which in turn, emerges as the progression of disease spectrum [4].

According to the “multi-metabolic” nature of the disease, nomenclature has been replaced with metabolic dysfunction-associated fatty liver disease (MAFLD) and subsequently to metabolic dysfunction-associated steatotic liver disease (MASLD) as novel terminologies [5]. These acronyms highlight the bi-directional association between NAFLD and metabolic disorders, specifically metabolic syndrome components and presents a more precise definition and diagnosis as well as emphasizing the urgent need for proper adjuvant therapies based on targeting insulin sensitivity, inflammation, and liver metabolic regulation [3, 5].

These acronyms highlight the bi-directional association between NAFLD and metabolic disorders, specifically metabolic syndrome components and present a more precise definition and diagnosis of the disease and related complications.

Despite the lack of an approved pharmaceutical approach for NAFLD treatment, lifestyle modifications, including increased physical activity and adherence to a hypocaloric diet, have been highly recommended [6]. Furthermore, a combination of caloric restriction and conjugated therapies, specifically herbal products, has recently been recommended [7].

Salvia hispanica L. (Chia seeds) is an egg-shaped, gray seed with black and white stains, with a width of about 2 mm, belonging to the mint family (Lamiaceae, Labiatae class), originating from Mexico and Guatemala [8, 9]. From a nutritional perspective, chia seeds are composed of about 42% carbohydrate, 30% fat, and 16% protein, with energy up to 486 kcal/100 g [9]. It is a potent plant source of omega-3 fatty acids, dietary fiber (mucilaginous polysaccharide gum), antioxidants, protein, vitamins, and minerals [10]. In this regard, dietary chia intervention has been applied in a number of non-communicable diseases because of its favorable cardio-protective features, including glucose- and lipid-lowering effects as well as anti-oxidative and anti-inflammatory activities [11, 12].

Previous experimental studies on rats have shown that chia intervention could reduce hepatic fat accumulation and triglyceride content of the liver [13, 14]. The omega-3 fatty acids of chia seeds, specifically alpha-linolenic acid and dietary fiber, are the main two ingredients that exert hepato-protective activities by regulating molecular mediators, immune cells, adipocyte function, in addition to modulating inflammatory cytokines, enterocyte proliferation, appetite control, and insulin sensitivity [11, 12]. Results of the only single-arm human trial by Medina-Urrutia et al. [15] on 25 patients with NAFLD revealed that 25 g/day of milled chia in the diet could reduce body weight, waist circumference, and metabolic abnormalities. The current body of evidence seems to be insufficient and unclear regarding the exact clinical outcomes of chia intervention in patients with NAFLD. Therefore, the present study was designed to investigate the effects of chia seeds on insulin sensitivity, hematological inflammatory indices (HIIs), liver function, and hepatic steatosis in obese patients with NAFLD.

Materials and methods

Study design and patients

The present study was a single-blind randomized controlled clinical trial to examine the effects of chia seed intervention on insulin sensitivity, HIIs, and hepatic function in 44 obese patients with NAFLD. The patients were recruited through public printed advertisements and referral by a gastroenterologist. After pre-screening the initial study criteria through telephone interview, the subjects were precisely screened for the study inclusion and exclusion criteria at the first visit. Participants from both genders were included in the study if they were aged 20–50 years with a body mass index (BMI) between 30 and 40 Kg/m² and a diagnosis of mild and moderate hepatic steatosis (Grade I and Grade II). Exclusion criteria included cigarette smoking, alcohol drinking, pregnancy, lactation or menopause, hormone therapy or oral contraceptive use, any history of metabolic diseases such as cardiovascular, renal, liver and autoimmune diseases, hypertension, diabetes mellitus (T2DM) and thyroid disease as well as taking dietary supplements, glucose- or lipid-lowering medications and following a specific diet for at least three months prior the study. Participants who experienced any side effects or failed to follow the prescribed diet and consume chia seeds were withdrawn from the study.

Liver echogenicity of the newly diagnosed obese patients with NAFLD was evaluated in a fasting state according to ultrasound findings by an expert single ultrasonographist (Sonoace X4 Medisio, South Korea). The severity of hepatic steatosis was assessed based on hepatorenal echo contrast, vessel blurring, and deep attenuation defined by Hamaguchi et al. [16] and classified into four grades as follow: Grade zero (without steatosis), Grade I (mild steatosis as a slight increase in the liver echogenicity or the presence of bright liver), Grade II (moderate steatosis as impaired visualization of intrahepatic vessels and increased liver echogenicity) and Grade III (severe steatosis as a marked discrepancy between hepatic and renal echoes, poor or no visualization of hepatic vessels, and the presence of severe bright liver).

Before the commencement of the study, the objectives and protocol of the study were fully explained to the participants, and they were asked to sign a written informed consent. This trial was carried out in accordance with the Declaration of Helsinki. The study protocol was approved by the Ethics Committee of Tabriz University of Medical Sciences, Tabriz, Iran (reference number: TBZMED.REC.1402.886) and registered in the website of Iranian Registry of Clinical Trials (available at: https://www.irct.ir. ID: IRCT20100209003320N23).

Sample size

Considering the limited RCTs on the effects of chia intervention in patients with NAFLD and the potent role of triglyceride accumulation in the pathogenesis and progression of NAFLD, we used the changes in mean (standard deviation (SD)) of serum triglyceride levels reported by Alwosais et al. [17] to calculate the sample size. Considering a power of 90% and a confidence level of 95% (errors of β = 10% and α = 0.05) in two-sided tests using G-power software (Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany) and taking into account a probable 10% drop-out rate was estimated 19 patients with NAFLD for each arm.

Study procedures, randomization, and intervention

The eligible patients with NAFLD were randomly allocated into the: “Chia” (n = 22) or “control” (n = 22) groups with a ratio of 1:1, using a block randomization procedure of size 3 and were stratified by age (≤ 35 years vs. > 35 years), sex (male vs. female), and BMI (30–34.9 vs. 35–40 Kg/m²). To generate a random sequence, the Random Allocation Software was used by a statistician. The random sequences were kept in a secure location and administered by an assistant who had no involvement in the study or any contact with the patients. The researchers were blinded to group assignment.

Participants in both studied arms received a calorie-restricted diet (CRD), whereas those in the chia group received chia seeds accompanied by CRD for 8 weeks. The daily energy requirement was individually estimated based on Mifflin formula and gender by an experienced dietitian and the total energy expenditure was calculated by adding up individual estimated resting energy expenditure, physical activity level (approximated based on the data obtained from International physical activity questionnaire-short form), and thermic effect of food (10% of TEE). Individualized CRD was planned 500 kcal less than the TEE in the control group, while in the chia group, CRD was set 700 kcal/day less than the estimated TEE, as 40 g of Brazilian chia contains 200 kcal energy. Macronutrient distribution was 55%, 30% and 15% of energy from carbohydrates, fat, and protein, respectively. The CRDs were designed according to the Food Guide Pyramid and the recommendations of the National Institutes of Health Obesity Education Initiative Expert Panel. Meal plans for CRD were also prepared based on the mentioned calculations and the food-based dietary guidelines for Iranians (available at http://www.fao.org/nutrition/education/food- based dietary- guide lines/regions/countries/iran/fr/). To each patient, the food group exchange list as well as the food album were delivered with full explanations on how to use food exchange lists for replacing foods did not have access with those providing similar calories from the corresponding food groups.

The patients in the control group received only CRD, whereas those in the chia group received Berazilian chia (Giahineh Co., Isfahan, Iran) containing sacchets (each 20 g) twice a day, accompanied by CRD for 8 weeks. Each Chia sachet (20 g) was consumed 30 min before lunch and dinner by mixing with a glass of warm water. All patients were monitored fortnightly during the study period for adherence to CRD as well as Chia consumption. In addition, the patients were asked to return the unconsumed sachets every two weeks, and by counting them, the noncompliance rate was considered if the unconsumed sachet count was more than 10% of the administered ones.

Data collection

At baseline, demographic characteristics including age, sex, marital status, educational and occupational status, medical history, and following any special diets were obtained from each patient. Body weight and height were measured with minimal clothes and no shoes on using a calibrated stadiometer (Seca, Hamburg, Germany) with an accuracy of 100 g and 0.1 cm, respectively, and then, BMI was calculated by dividing the weight (Kg) by height squared (m²).

Food intake was assessed by completing a 3-day 24-hour food recall (including two non-consecutive weekdays and one weekend) by face to face interview, before and after the intervention for estimating energy and nutrient intakes using the Nutritionist IV software program modified for Iranian foods (First Databank; Hearst, San Bruno, CA, USA). Moreover, a validated international physical activity questionnaire-short form was used for the estimation of physical activity level. Responses were converted to metabolic equivalent of task minutes per week (MET-min/week) at the beginning and end of the study [18].

Lab assay

Pre- and post- intervention and after a 12–14 h overnight fasting, 10 ml blood samples were taken from each subject and divided into two tubes: 5 ml whole blood in EDTA-containing vacuum blood collection tubes (Vacutainer K2E) and 5 ml blood in vacuumed gel separator tubes to be centrifuged to obtain serum and stored at −70^◦c until the end of the intervention.

To assess total white blood cell count (WBC), neutrophil, lymphocyte, and monocyte count per mm3 as well as platelet count in whole blood, an automated and daily-calibrated Coulter CBCH1 counter was applied. Then HIIs, as novel inflammatory markers, were also estimated based on the following formula [19]: SIRI=(neutrophil count*monocyte count)/lymphocyte count.

Serum fasting blood sugar (FBS), albumin, alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were determined using enzymatic-colorimetic commercial kits (Pars-Azmoon Co., Tehran, Iran) and an auto-analyzer (Hitachi 902, Tokyo, Japan) and the ratio of AST to ALT (AST/ALT) was estimated. Hemoglobin A1c (HbA1c) was assessed using photometry in whole blood using Pars Azmoun Company kit (Pars Azmoun, Iran) and Hitachi auto analyzer (Hitachi-917, Tokyo, Japan).

Serum insulin and ferritin concentrations were evaluated using an enzyme-linked immunosorbent assay kit (Monobind, Lake Forest, CA, USA). Then, insulin sensitivity was addressed by the estimation of indices, including: as follows [20]:

HOMA-IR = [fasting insulin (µIU/mL) × fasting glucose (mg/dL)]/405

HOMA-B = 360 × fasting insulin (µU/mL)/(fasting glucose (mg/dL) − 63)

HOMA-S (%)= [1/HOMA-IR] × 100%

Moreover, the was calculated as below [21]:

NFS = −1.675 + 0.037 × age (year) + 0.094 × BMI (kg/m²) + 1.13 × IFG/diabetes (yes = 1, no = 0) + 0.99 × AST/ALT ratio − 0.013 × platelet count (×10⁹/L) − 0.66 × albumin (g/dL).

Then, the score was classified into: [1] “high probability of advanced fibrosis” (≥ 0.676) [2], “intermediate probability of advanced fibrosis” (between − 1.455 and 0.676), and [3] “low probability of advanced fibrosis” (≤ − 1.455) [21].

Study outcomes

In the present trial, changes in energy and nutrient intakes, insulin sensitivity, liver function, hepatic steatosis, and fibrosis were considered as the primary outcomes, while changes in HIIs were defined as secondary outcomes.

Statistical analysis

Data was analyzed using IBM Statistical Package for Social Analysis (IBM SPSS Statistics, Armonk, USA, version 23). To check the normal distribution of the variables, Kolmogorov-Smirnov test was applied, and then numeric and categorical variables were presented as mean [22] and number (percentage), respectively. Moreover, an intention-to-treat approach was followed for data analysis. Within-group changes were addressed using paired samples t-test and Wilcoxon signed-rank test, after 8 weeks. Differences in symmetric and asymmetric continuous variables were done using two-independent sample t- and Mann-Whitney U-tests between the two groups, respectively. Chi-square test and Fisher’s exact test were also applied for categorical variables. At the end of the trial, between-group changes of numeric variables were tested using the analysis of covariance and Quantile regression by adjusting for baseline values and changes in energy intake as the confounders. To assess the clinical effectiveness of treatment with Chia, the number needed to treat (NNT) was calculated as: NNT = 1/absolute risk reduction. p < 0.05 was considered statistically significant level.

Results

Of the total 85 recruited obese patients with NAFLD, 44 patients were enrolled in the present clinical trial (n = 22 in both groups) (Fig. 1). During the trial, three patients in the chia group and one patient in the control group were excluded because of improper chia consumption and not following the prescribed diet. Consequently, 19 patients in the chia group and 21 patients in the control group completed the intervention. There were no adverse effects related to the intervention.

Fig. 1.

Study the flowchart

As shown in Table 1, most of the patients in both studied groups were females, married, and with a diploma or university degrees. Mean age and BMI were 37.0 years and 40.1 years, as well as 33.3 Kg/m² and 31.8 Kg/m² in the chia and control groups, respectively. The majority of the participants in both groups exhibited class-1 obesity (BMI: 30-34.9 Kg/m²). Most of the patients in the chia group had mild liver steatosis (52.6%), while moderate liver steatosis was more prevalent in the control group (57.1%) (p = 0.536) (Table 1).

Table 1.

Baseline characteristics of the participants

Variable	Chia (N=19)	Control (N=21)	p
	N (%)	N (%)
Sex
Female	10 (52.6)	13 (61.9)	0.796†
Male	9 (47.4)	8 (38.1)
Marital status			0.689††
Married	15 (78.9)	18 (85.7)
Single	4 (21.1)	3 (14.3)
Educational level			0.554††
Less than diploma	4 (21.0)	1 (4.8)
Diploma	6 (31.6)	12 (57.1)
University degree	9 (47.4)	8 (38.1)
BMI status (Kg/m2)			0.473††
30-34.9	13 (68.4)	17 (81.0)
35-39.9	6 (31.6)	4 (19.0)
NAFLD severity			0.536†
Mild	10 (52.6)	9 (42.9)
Moderate	9 (47.4)	12 (57.1)

	Mean (SD)	Mean (SD)
Age (yrs.)	37.0 (9.3)	40.1 (9.2)	0.291*
Weight (Kg)	94.3 (13.0)	89.2 (13.2)	0.212*
Height (cm)	168.4 (9.2)	167.3 (10.1)	0.727*
BMI (Kg/m2)	33.3 (3.9)	31.8 (4.2)	0.272*
AST (U/L)	25.2 (8.9)	23.0 (6.0)	0.649*
ALT (U/L)	29.6 (21.2)	32.2 (23.5)	0.872*

SFA, Saturated fatty acids; MUFA, Monounsaturated fatty acids; PUFA, Polyunsaturated fatty acids; PAL, Physical activity level; METs, Metabolic equivalents of task (MET-minutes/week)

Mean (standard deviation) is presented for the data

Bold values indicate statistically significant (p<0.05)

*p-value for paired-t test; ^**p-value for Independent samples t-test; ^***p-value for ANCOVA test or Quantile regression (adjusted for baseline values and energy intake); ^†p-value for Related samples Wilcoxon Rank test, ^†† p-value for Mann-Whitney U test

Table 2 demonstrates changes in daily dietary intake and physical activity. Significant decreases were observed in energy and carbohydrate intake in the studied groups (~ −444 Kcal/d and ~ −60 g/d in chia group vs. ~ −147 Kcal/d and ~ −7 g/d in control group, respectively), while reductions were markedly greater in the chia group (p = 0.015 for energy intake and p = 0.029 for carbohydrate intake, respectively). Despite the significant increases in protein and fat intake in the control group, between-group differences were not significant. In addition, patients in the chia group reported greater intake of polyunsaturated fatty acids and dietary fiber in comparison with the control group (p = 0.016 and p < 0.001, respectively). Between-group comparison of PAL did not reach statistically significant levels (p = 0.746) (Table 2). Despite significant intra-group reductions in body weight (−4.5 Kg in chia group (p < 0.001) vs. −4.6 Kg in control group (p = 0.010), respectively) and BMI (−0.6 Kg/m² in chia group (p < 0.001) and − 1.5 Kg/m² in control group (p = 0.005), respectively) in both studied groups after 8 weeks, inter-group differences failed to reach to statistically significant levels after adjusting for potential confounders.

Table 2.

Dietary intake and physical activity before and after the intervention

Variable	Chia (N=19) Mean (SD)	Control (N=21) Mean (SD)	p
Energy (Kcal/d)
Baseline	1910.8 (349.7)	1944.1 (284.0)	0.742**
End	1466.8 (240.2)	1796.7 (296.0)	0.015 ***
p *	<0.001	0.003
Carbohydrate (g/d)
Baseline	230.2 (106.0)	241.9 (42.5)	0.656**
End	170.1 (59.1)	235.3 (44.7)	0.029 ***
p *	0.008	0.220
Protein (g/d)
Baseline	70.9 (21.9)	75.7 (9.6)	0.373**
End	66.2 (19.2)	80.3 (11.8)	0.111***
p *	0.387	0.030
Fat (g/d)
Baseline	49.5 (24.4)	50.3 (8.6)	0.898††
End	56.0 (16.5)	62.2 (11.7)	0.802***
p †	0.229	<0.001
SFA (g/d)
Baseline	10.7 (5.6)	17.6 (10.8)	0.016††
End	12.4 (5.2)	18.1 (7.4)	0.169***
p †	0.134	0.816
MUFA (g/d)
Baseline	15.9 (11.1)	11.5 (5.7)	0.119††
End	21.9 (6.3)	21.8 (5.4)	0.744***
p †	0.021	<0.001
PUFA (g/d)
Baseline	15.1 (9.0)	15.3 (4.4)	0.953††
End	15.2 (6.9)	13.5 (5.7)	0.016 ***
p †	0.988	0.182
Dietary fiber (g/d)
Baseline	10.4 (4.9)	18.6 (6.3)	<0.001 **
End	22.8 (3.9)	17.9 (7.0)	<0.001 ***
p *	<0.001	0.548
PAL (MET-min/wk)
Baseline	786.6 (646.8)	668.2 (639.6)	0.573††
End	603.9 (430.1)	623.9 (598.9)	0.746***
p †	0.009	0.209

SFA, Saturated fatty acids; MUFA, Monounsaturated fatty acids; PUFA, Polyunsaturated fatty acids; PAL, Physical activity level; METs, Metabolic equivalents of task (MET-minutes/week)

Mean (standard deviation) is presented for the data

Bold values indicate statistically significant (p<0.05)

*p-value for paired-t test; ^**p-value for Independent samples t-test; ^***p-value for ANCOVA test or Quantile regression (adjusted for baseline values and energy intake); ^†p-value for Related samples Wilcoxon Rank test, ^†† p-value for Mann-Whitney U test

Changes in biochemical parameters in both groups, before and after the intervention, are presented in Table 3. Chia consumption plus CRD resulted in significant reductions in serum insulin and HOMA-IR, as well as significant increases in HOMA-S, while there were significant increases in serum insulin, HOMA-IR, and HOMA-B along with decreases in HOMA-S in the control group. Inter-group comparison revealed significant improvements in glycemic profile in terms of insulin (p = 0.016) and HOMA-S (p < 0.001) after chia intervention by adjusting for baseline values and changes in energy intake. Regarding liver function, marked reductions were found in serum AST, ALT, and NFS in the chia group, while AST/ALT significantly increased in the control group. Moreover, after adjusting for the confounders, chia plus CRD significantly reduced NFS (p = 0.020) and inhibited the increases in AST/ALT (p = 0.038), compared with only CRD (Table 3).

Table 3.

Biochemical parameters before and after the intervention

Variable	Chia (N=19) Mean (SD)	Control (N=21) Mean (SD)	p
FBS (mg/dL)
Baseline	94.0 (8.9)	99.1 (13.4)	0.166**
End	91.5 (12.2)	96.4 (13.6)	0.584***
p *	0.385	0.347
Insulin (µIU/mL)
Baseline	18.8 (12.3)	12.0 (2.6)	0.093††
End	17.5 (12.1)	15.0 (6.5)	0.016 ***
p †	0.007	0.003
HbA1c (%)
Baseline	5.51 (0.41)	5.46 (0.49)	0.770**
End	5.36 (0.34)	5.40 (0.43)	0.463***
p *	0.158	0.248
HOMA-IR
Baseline	4.24 (2.48)	2.94 (0.77)	0.065††
End	3.88 (2.53)	3.68 (2.04)	0.056***
p †	0.043	0.005
HOMA-B (%)
Baseline	276.7 (327.0)	135.6 (58.2)	0.069††
End	321.2 (466.4)	178.8 (72.8)	0.162***
p †	0.601	0.005
HOMA-S (%)
Baseline	33.7 (22.7)	36.3 (9.9)	0.647††
End	44.3 (41.2)	32.9 (13.7)	<0.001 ***
p †	0.013	0.009
AST (U/L)
Baseline	25.2 (8.9)	23.0 (6.0)	0.649**
End	20.9 (6.8)	32.2 (6.0)	0.068***
p *	0.004	0.553
ALT (U/L)
Baseline	29.6 (21.2)	32.2 (23.5)	0.872††
End	22.6 (11.0)	23.3 (14.8)	0.844***
p †	0.001	0.001
AST/ALT
Baseline	1.06 (0.49)	0.96 (0.57)	0.587††
End	1.04 (0.38)	1.26 (0.67)	0.038 ***
p †	0.520	0.007
Ferritin (ng/mL)
Baseline	83.0 (69.8)	108.9 (114.3)	0.957††
End	86.3 (76.2)	94.4 (91.8)	0.420***
p †	0.243	0.170
NFS
Baseline	−2.74 (1.16)	−2.53 (0.94)	0.522**
End	−2.99 (1.16)	−2.34 (0.83)	0.020 ***
P *	0.049	0.259
Albumin (g/dL)
Baseline	4.49 (0.40)	4.61 (0.42)	0.338**
End	4.57 (0.37)	4.56 (0.36)	0.602 ***
p *	0.334	0.611

FBS, Fasting blood sugar; HbA1c, Glycosylated hemoglobin A1c; HOMA-IR, Homeostatic model assessment of insulin resistance; HOMA-B, Homeostasis beta-cell function; HOMA-S, Homeostatic model assessment of insulin sensitivity; AST, Aspartate aminotransferase; ALT, Alanine transaminase; AST/ALT, Aspartate aminotransferase to alanine transaminase ratio; NFS, NAFLD fibrosis score

Mean (SD) is presented for the data

Bold values indicate statistically significant (p<0.05)

*p-value for paired- t test; ^**p-value for Independent samples t-test; ^***p-value for ANCOVA test or Quantile regression (adjusted for baseline values and energy intake); ^†p-value for Related samples Wilcoxon Rank test, ^†† p-value for Mann-Whitney U test

Table 4 illustrates the changes in hematologic parameters and HIIs over the trial in both groups. Results show that although chia supplementation resulted in a significant reduction in lymphocyte count and therefore significant increases in NLR and PLR, intergroup differences in HII at the end of the trial reveal noticeable differences in neutrophil counts (p = 0.003), MLR (p = 0.001), NLR (p = 0.005), PLR (p = 0.027), and SIRI (p = 0.034), after adjusting for confounders.

Table 4.

Hematologic parameters and hematological inflammatory indices before and after the intervention

Variable	Chia (N=19) Mean (SD)	Control (N=21) Mean (SD)	p
WBC (*109/L)
Baseline	7.58 (1.45)	7.32 (2.40)	0.694**
End	7.55 (1.41)	6.79 (1.81)	0.067***
p *	0.933	0.020
Monocytes (*109/L)
Baseline	0.41 (0.11)	0.37 (0.13)	0.305**
End	0.37 (0.13)	0.35 (0.14)	0.876***
p *	0.115	0.310
Lymphocytes (*109/L)
Baseline	2.53 (0.75)	2.38 (0.98)	0.602**
End	2.22 (0.66)	2.39 (0.75)	0.092***
p *	0.033	0.925
Neutrophils (*109/L)
Baseline	4.05 (0.81)	4.17 (1.55)	0.779**
End	4.40 (1.03)	3.66 (1.20)	0.003 ***
p *	0.196	0.004
Platelets (*109/L)
Baseline	280.3 (54.6)	249.1 (58.1)	0.089**
End	282.1 (62.4)	248.9 (55.4)	0.658***
p *	0.673	0.971
MLR
Baseline	0.13 (0.04)	0.13 (0.06)	0.815**
End	0.13 (0.06)	0.10 (0.05)	0.001 ***
p *	0.796	0.114
NLR
Baseline	1.71 (0.54)	2.06 (1.44)	0.326**
End	2.10 (0.62)	1.58 (0.50)	0.005 ***
p *	0.015	0.107
PLR
Baseline	118.39 (36.18)	123.54 (71.50)	0.779††
End	134.13 (39.10)	110.85 (35.31)	0.027 ***
p †	0.031	0.399
SIRI
Baseline	0.71 (0.28)	0.75 (0.50)	0.713††
End	0.78 (0.36)	0.56 (0.32)	0.034 ***
p †	0.379	0.092

WBC, White blood cell count; MLR, Monocyte to lymphocyte ratio; NLR, Neutrophil to lymphocyte ratio; PLR, Platelets to lymphocyte ratio; SIRI, Systemic inflammation response index

Mean (SD) is presented for the data

Bold values indicate statistically significant (p<0.05)

*p-value for paired- t test; ^**p-value for Independent samples t-test; ^***p-value for ANCOVA test or Quantile regression (adjusted for baseline values and energy intake); ^†p-value for Related samples Wilcoxon Rank test, ^†† p-value for Mann-Whitney U test

The clinical effectiveness of chia supplementation plus CRD on liver steatosis and fibrosis severity is represented in Table 5. One grade reduction in hepatic steatosis was found in 52.6% and 47.6% in the chia group and control group, respectively. Moreover, 15.8% of the patients receiving chia plus CRD exhibited significant improvement in liver fibrosis score, whereas only one patient (4.8%) in the control group showed improved liver fibrosis. The estimated NNT for a 1-grade reduction in liver steatosis and relative improvement in liver fibrosis severity were 20 and 9, respectively. In other words, of every 20 and 9 patients with NAFLD who supplemented with chia plus CRD for 8 weeks, one patient experienced one one-grade reduction in liver steatosis and relative improvement in liver fibrosis, respectively (Table 5).

Table 5.

Improvements in liver steatosis and fibrosis severity

	N(%)	ARR (%)	NNT	p *
One grade reduction in NAFLD severity
Chia	10 (52.6)	5.0	20	0.752
Control	10 (47.6)
Improvement in liver fibrosis status
Chia	3 (15.8)	11.1	9	0.331
Control	1 (4.8)

ARR, Absolute risk reduction; NNT, Number needed to treat

^∗p-value for the chi-square test

Discussion

In the recent scientific landscape, functional foods, as the combination of nutrition and medicine have gained great interest [23]. Chia seed is a functional food that comprises protein, lipids, carbohydrates, and dietary fiber as well as phytochemicals (antioxidants, phenolic compounds, myricetin, quercetin, and kaempferol) and exerts several therapeutic effects in chronic diseases [24].

Findings of our clinical trial revealed greater reductions in energy and carbohydrate intakes in chia group compared with control group at the end of the study, whereas no inter-group differences were observed in body weight and BMI. Vuksan et al. [25] in a RCT on overweight and obese patients with T2DM reported that Salba-chia (30 g/1000 kcal/day) significantly reduced the desire to eat and body weight. Furthermore, Khan et al. [26] in 2024, by combining chia seeds and Fennel seeds for three months, showed markedly reduced BMI and improved lipid profile in subjects with obesity. These features have been attributed to the high fiber content of chia seeds [27]. It seems that prescribing 40 g/day chia seeds for 8 weeks in this study compared with previous trials with longer durations and higher chia doses, is the reason for non-significant changes in obesity status. The observed reduced food intake could reach to weight-lowering properties in more longer periods.

Insulin sensitivity is also a potent therapeutic target in the management of NAFLD [3]. Based on our findings, chia plus CRD resulted in significant reductions in serum insulin and HOMA-IR, as well as marked increases in HOMA-S values. Indeed, improvements in serum insulin and HOMA-S were obvious in favor of chia, after adjusting for the confounders. Different in vivo and in vitro models have highlighted the effects of chia (in the forms of seeds, flour, oil, and extract) on ameliorating glucose tolerance, postprandial glucose levels, and reducing gluconeogenic enzymes [22, 28].

The current body of evidence on the hypoglycemic effects of chia remains controversial. One clinical trial reported no effects of chia flour (35 g/day for 12 weeks) on glucose among participants with overweight and obesity [29]. Alwosais et al. [17] in 2021 also failed to observe any changes in FBS, HbA1c, and insulin after 12 weeks of chia seeds supplementation (40 g/day) in patients with T2DM, while some studies have suggested the positive aspects of chia supplementation on insulin function in metabolic diseases [22, 28, 30–32]. For example, Vuksan et al. [30] reported improved postprandial glucose following Salba chia ground Salba chia (25 g/day) administration in healthy participants. The same researcher has also reported marginal improvements in HbA1c levels in a crossover study comparing Salba chia with wheat bran on 20 patients with T2DM after 12 weeks [31]. Another trial on patients with IR and NAFLD also demonstrated improved adipose tissue, endothelial, and beta cell function, and thereby fine-tuned oral glucose tolerance test, subsequent to 8 weeks of ground chia supplementation (25 g/day) [32]. Collectively, a recent meta-analysis in 2024 (on 8 RCTs and 362 participants) suggested that chia seeds do not significantly alter glycemic status [33].

In spite of the controversy between these results and findings on metabolic diseases, our results are mostly in accordance with the evidence on NAFLD in animal models and human trials. Hence, far from the prescribed dose and duration of the intervention, it seems that chia seed exerts a better insulin sensitizing effect in patients with NAFLD compared with other diseases. As larger effect on HbA1c need a duration around three months, it appears that supplementation for longer times may also be able to significantly affect FBS and HbA1c levels which were unaffected in this trial. In this context, chia as a superfood contains high concentrations of omega-3 fatty acids and ALA which favors a functional return of pancreatic cells, as well as a generally healthier adipose tissue [30, 32]. Another mechanism of action is explained by the increases in tyrosine phosphorylation of insulin receptor substrate-1 and subsequent phosphorylation of protein kinase-B, emerging as the translocation of the glucose transporter-4 to the plasma membrane after chia supplementation [34]. Moreover, up-regulated hepatic AMP-activated protein kinase (AMPK) is followed by reduced adipocyte deposition and increased insulin sensitivity [34]. Chia also contains high values of soluble fiber, which increases intestinal viscosity and reduces glucose absorption [35].

Liver function is closely influenced by chronic low-grade inflammation [36]. Regarding the changes in HIIs, only PLR markedly increased in the chia arm compared with control arm.

To our knowledge, the evidence on the effects of chia on inflammation in NAFLD is limited. An in vitro study introduced the ability of protein-rich fractions of chia seeds to mitigate nitric oxide and interleukin-6 in HMC3 microglial cells [37]. These findings were strengthened by Chan-Zapata et al. [38] who documented the efficacy of chia seeds’ protein fractions in reducing the secretion of nitric oxide, H2O2, TNF-α, and pro-inflammatory interleukins in in vitro and in vivo settings. Furthermore, adding chia seeds to the diet of Wistar rats on a high-fat diet for 35 days resulted in decreased TNF-α and nuclear factor-κB levels [39]. Mohamed et al. [40] also reported the favorable effects of chia oil and mucilage on TNF-α and lipid peroxidation in a rat model of arthritis with obesity.

In this regard, increases in PLR, as the suggested preventive marker of inflammation and liver fibrosis in our findings is partially in line with previous evidence, while we failed to observe further anti-inflammatroy effects in this trial. It may be attributed not only to the form of prescribed chia (such as chia seeds vs. protein fractions), but also to the effects of CRD which imrproves inflammatory state by itself. It should be noted that the investigated patients were subjects with grade I and II hepatic steatosis with mild pro-inflammatory situation. The anti-inflammatory properties of chia seeds could be explored in patients with grade III steatosis along with steatohepatitis. In conclusion, further research is needed to confirm the effects of various chia forms on inflammation in NAFLD.

Taken together, the therapeutic anti-inflammatory function of chia is attributed to the rich sources of proteins (such as albumin and glutelin), polyphenols, and ALA, which inhibit oxidative stress and the production of pro-inflammatory cytokines [41]. The synchronized regulation of molecular targets, including AMPK, mitogen-activated protein kinases, nuclear factor kappa B, and PPAR-γ by chia seeds ingredients is the main molecular anti-inflammatory mechanism [36, 42, 43].

In the present RCT, chia plus CRD administration led to significant reductions in serum AST, ALT, and NFS after 8 weeks, while noticeable inter-group differences were observed in AST/ALT and NFS in favor of chia, resembling improvements in liver function. Moreover, according to ultrasonography findings and fibrosis score, of every 20 and 9 patients in the chia arm, one patient experienced one-grade reduction in hepatic steatosis and relative improvement in liver fibrosis, respectively (changes were statistically non-significant).

According to our extensive search, studies on the hepatoprotective aspects of chia seem insufficient, while preliminary findings have been reported [15, 22, 32, 35, 44, 45]. For instance, an animal study in 2022 examined the effects of chia seeds and sprout-derived bioactive compounds on high-fat and fructose diet-induced obesity in rats, reporting the attenuation of IR, hepatic triglyceride accumulation, and dyslipidemia [44]. Subsequently, Maturana et al. [45] in another study on obese mice, observed reduced size and number of hepatic lipid droplets following the administration of chia leaf ethanolic extract. Regarding human clinical trials, Medina-Urrutia et al. [15] in a single-arm trial addressed the effects of 8 weeks supplementation with milled chia (25 g/day) on 25 subjects with NAFLD and reported weight loss and improved liver steatosis in almost 52% of the participants. Hence, our results seem to be in line with previous evidence on liver function and highlight the anti-fibrotic feature of chia seeds plus CRD in patients with NAFLD for the first time. Our findings highlight hepato-protective effects of chia plus CRD, while due to the synergystic favorable effects of both chia and CRD, further studies could examine the effects of only chia supplementation versus a proper placebo. In addition, longer treatment period could possibly strenghten these positive effects.

From a mechanistic view, these effects could be explained by the phytochemical property of chia, including high omega-3 fatty acids, particularly, high eicosapentaenoic acid and docosahexaenoic acid content, as well as polyphenols such as caffeic acid and chlorogenic acid [35]. The mentioned ingredients up-regulate PPAR-α expression and thereby reduce lipid peroxidation, liver enzymes, and intra-hepatic fat content by interactions with carnitine palmitoyltransferase-1 and fatty acid oxidase enzymes [35]. Indeed, ALA, which is abundant in chia seeds, modulates sterol regulatory element-binding protein and then regulates hepatic de novo lipogenesis and oxidative-inflammatory state as the key mechanisms involved in the management of NAFLD [35].

Limitations & strengths

Relatively small sample size, short intervention duration, lack of proper placebo, and lack of assessing inflammatory and oxidative biomarkers could be considered as the limitations of the current study. Indeed, ultrasonography and NFS which were used as the non-invasive tests for hepatic steatosis and fibrosis evaluation are not gold standard methods. However, controlling possible confounding factors, and assessing HIIs for chronic inflammatory status are considered the strengths of this study.

Conclusion

Chia seeds supplementation (40 g/day) plus CRD for 8 weeks could reduce food intake, improve insulin sensitivity, liver function, and hepatic steatosis and fibrosis in obese patients with NAFLD. Nevertheless, more research is required to confirm these findings on the effectiveness of chia seeds as an adjunctive therapy in NAFLD.

Abbreviations

ALT

Alanine aminotransferase

AMPK

AMP-activated protein kinase

AST

Aspartate aminotransferase

AST/ALT

Aspartate aminotransferase to alanine aminotransferase ratio

BMI

Body mass index

FBS

Fasting blood sugar

HbA1c

Hemoglobin A1c

HIIs

Hematological inflammatory indices

HOMA-B

Homeostasis beta-cell function

HOMA-IR

Homeostatic model assessment of insulin resistance

HOMA-S

Homeostatic model assessment of insulin sensitivity

IR

Insulin resistance

MAFLD

Metabolic dysfunction- associated fatty liver disease

MASLD

Metabolic dysfunction-associated steatotic liver disease

MLR

Monocytes to lymphocyte ratio

NAFLD

Non-alcoholic fatty liver disease

NFS

NAFLD fibrosis score

NLR

Neutrophil to lymphocyte ratio

NNT

Number needed to treat

PLR

Platelet to lymphocyte ratio

PPAR-α

Peroxisome proliferator activated receptor-α

RCT

Randomized controlled clinical trial

SD

Standard deviation

SIRI

Systemic inflammation response index

T2DM

Type 2 diabetes mellitus

TEE

Total energy expenditure

TNF-α

Tumor necrosis factor-α

WBC

White blood cell count

WC

Waist circumference

Author contributions

The authors’ responsibilities were as follows: S.A.and M.P. collected data, S.A. wrote the original paper; M.E.M. contributed to the statistical analysis; H.T. and M.E.M. contributed to the final revision of the manuscript and the conception of the article. All authors read and approved the final version of the manuscript.

Funding

This study was funded by the ‘Research Vice-Chancellor’ of Tabriz University of Medical Sciences, Tabriz, Iran (Grant no.73118).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

All procedures performed in this study were in accordance with the ethical standards of the Ethics Committee of Tabriz University of Medical Sciences (IR.TBZMED.REC.1402.886), IRCT code: IRCT20100209003320N23, and all eligible patients signed an informed written consent form.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.