The effect of spirulina on type 2 diabetes: a systematic review and meta-analysis

Abstract

Purpose

The aim of the present study was to investigate the effect of spirulina on lipid profiles and glycemic related markers in type 2 diabetes patients.

Methods

PubMed, Scopus, Cochrane Library, ISI Web of Science, and Google Scholar were searched from inception to August 2020. All clinical trials which investigated the effect of spirulina supplementation on glycemic related markers and lipid profile among type 2 diabetes patients were included. Random effects modeling was utilized for pooling analysis to compensate for the between-study heterogeneity.

Results

Eight studies (9 arms) were included in the meta-analysis. We found a significant reduction in fasting blood glucose (−17.88 mg/dl; 95% CI: −26.99, −8.78; I²: 25%), triglyceride (−30.99 mg/dl; 95% CI: −45.20, −16.77; I²: 50%), total-cholesterol (−18.47 mg/dl; 95% CI: −33.54, −3.39; I²: 73%), LDL-C (−20.04 mg/dl; 95% CI: −34.06, −6.02; I²: 75%), VLDL (−6.96 mg/dl; 95% CI: −9.71, −4.22; I²: 33%), in addition to a significant increase in HDL-C (−6.96 mg/dl; 95% CI: −9.71, −4.22; I²: 33%), after spirulina administration. No significant effect was observed on HbA1C or post prandial blood sugar following spirulina consumption.

Conclusion

The present study suggests that spirulina supplementation can elicit beneficial effects on fasting blood glucose and blood lipid profiles.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40200-021-00760-z.

Background

Type 2 diabetes is a non-communicable disease, manifest via impairment in glucose metabolism, affecting both developed and developing countries [1]. Although many strategies have been suggested for ameliorating or treating diabetes, the incidence of this disease is growing rapidly [2]. Such high incidence imposes a critical burden on health care system utilization, which consequently confers a large economic cost annually [3]. Thus, any viable alternative, complementary, or adjunct therapy that may alleviate some economical and/or health care burden represents an issue of high importance.

In contemporary practice, lifestyle modification, including change in diet and physical activity, is the first step to the treatment of type 2 diabetes [3, 4]. However, many patients find it difficult to adhere to dietary restrictions [5]. On the other hand, many pharmacological agents cause adverse side-effects which limit their palatability and success [6, 7]. In this case, the efficacy of functional food and natural medicines as adjuvant therapies, concomitant with pharmacological agents, has become an interesting area for many researchers [8, 9].

Spirulina (Arthrospira maxima) is a microalga, belonging to the family of cyanobacteria with the most curative and prophylactic components of nutrition [10]; possessing cardioprotective and antioxidant activity due to high amount of phycocyanins, polyphenols, carotenoids, vitamins, essential fatty acids and protein [11]. The beneficial effect of spirulina in many non-communicable diseases has been shown previously [12–14]. Furthermore, animal studies indicated that spirulina can improve metabolic parameters related to glycemic status and lipid profile in diabetic mice [14, 15]. However, there is a lack of consensual evidence from clinical trials. Therefore, the present systematic review and meta-analysis was performed to summarize the current evidence and investigate the effect of spirulina supplementation on glycemic related markers and blood lipid profiles in type 2 diabetes patients.

Methods

The present systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) as a standard guideline [16].

Search strategy

A systematic literature search was carried out through Medline, Scopus, and ISI Web of Science, Cochrane Library, and Google Scholar from inception to August 2020. The comprehensive electronic search was performed by using the following keywords in combination with wildcard ‘*’ and medical subject headings (MeSH): (“Spirulina” OR “Arthrospira”) AND (“diabetes” OR “diabetic” OR “diabetes mellitus” “blood glucose” OR “glucose metabolism disorders” OR “hyperglycemia” OR “Hemoglobin A, Glycosylated”). The references of related clinical trials and pertinent review articles were also hand-searched to identify any additional studies of interest, which might have been missed during the electronic search.

Study selection

To identify eligible studies, two authors independently screened the studies which were included by the primary search. After excluding duplicates, studies were reviewed first by title/abstract and was articles obviously irrelevant were excluded. Subsequently, the full-texts of remaining studies were scanned. All clinical trials which investigated the effect of spirulina supplementation on glycemic related markers and lipid profile among type 2 diabetes patients were included. Studies were excluded if the duration of studies was <1 weeks, spirulina was administrated as part of a complex substance, spirulina was compared with an active agent/component, and the outcome of interest was not reported in the studies. Any discrepancy was settled by the third author. Supplemental Table 1 shows the PICOS (participants, intervention/exposure, comparisons, outcomes, and study design) criteria which is used to define the research question.

Data abstraction and assessment of the quality

Eligible clinical trials were separately reviewed by two authors (A.H and M.P) and the following information was recorded from each study: the first author’s last name, years of publication, country of origin, the total number of participants in each arm as well as their characteristics (mean age, gender), study design, duration of intervention, details of intervention and control groups, the dose of spirulina supplementation and outcomes of interest which reported.

Cochrane Risk of Bias Tool for Randomized Controlled Trials was used to detect the potential risk of bias in included studies (20). This scale included several criteria to evaluate the adequacy of random sequence generation, allocation concealment, blinding as well as detection of incomplete outcome data, reporting selective outcome, and other potential sources of bias. Based on recommendations of the Cochrane Handbook, the judgment of each item appears by “Low”, “High” and “Unclear” risk of bias. Any disagreement in data extraction and quality assessment judgment was resolved by discussion with a third investigator.

Statistical analysis

The whole process of statistical analyses was conducted using the Cochrane Program Review Manager Version 5.3 and STATA software (version 11.0; Stata Corporation). To estimate pooled effect size, data from all variables, including fasting blood glucose (FBS), postprandial blood sugar (PPBS), glycated hemoglobin A1C (HbA1C), triglyceride (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C) and very-low-density lipoprotein cholesterol (VLDL), which were reported in three or more studies, were extracted as mean difference and standard deviation (SD). In any instance where mean change and SD of change were not reported directly for intervention and control groups, it was calculated following a suggested formula [SD = square root [(SD pre-treatment)² + (SD post-treatment)² - (2 r × SD pre-treatment × SD post-treatment)] [17]. To take probable between-study heterogeneity into account, we applied the random-effects model in our analyses to estimate the overall effect size [18, 19]. Sensitivity analysis was conducted by eliminating each study, one at a time, to determine the influence of each study on the overall result. Egger’s regression asymmetry test and Begg’s rank-correlation methods were also performed to explore potential publication bias [20, 21]. To evaluate the potential influence of putative moderators such as baseline measures and duration of administration on changing variables in response to spirulina supplementation, the meta-regression was applied. Results were assumed as statistically significant when P < 0.05.

Results

The study selection process and the reason for study exclusion at each step is illustrated in Fig. 1. The electronic selection process yielded 907 unduplicated trials, in which 896 of them were excluded by title/abstract screening, and 11 studies remained for full-text assessment. Three studies were omitted due to not being conducted on diabetes patients (n = 2) or not reporting outcomes of interest (n = 1). One study administrated 2 different doses of spirulina and was considered as 2 separate arms. Therefore, 8 studies comprising 9 arms met eligibility criteria and were included in the present meta-analysis.

Fig. 1.

Flow chart of the process of the study selection

Studies’ characteristics

The characteristics of included studies are detailed in Table 1. In brief, 8 clinical trials (9 arms) [22–29] comprising 334 patients with diabetes, with a mean age of 51 years old, were included in the meta-analysis. The studies were conducted in various countries including India [22, 23, 25, 27, 28], Iran [24], Romania [29] and South Korea [26], and published between 2001 and 2017. The baseline BMI of participants was only reported in 4 trials [26–29]. Duration of intervention ranged from 45 to 90 days, and the dose of spirulina administration varied between 0.8 and 8 g/day. Only one trial [29] reported that participants received a placebo as a control group. Two trials recruited only male participants [23, 25], whilst of the remaining studies enrolled patients of both genders.

Table 1.

The main characteristics of included studies.

First author (publication year)	Country	Number and gender (M/F)	Mean age	BMI	Duration of disease (years)/treatment	Clinical Trial design/randomized/Blinding	Duration of study (Days)	Comparison group	Amount Of Spirulina intake	Notes about participants	Outcomes of interest
Serban et al. (1982) [29]	Romania	Intervention: 15 Control: 15 (NR)	Range years: 30-70 Intervention: 61.7 ±6.85 Control: 61.6±8.90	Intervention: 36.6±6.05 Control: 36.2±6.93	> 2 / Metformin	Parallel/Yes/Yes	60 day	Metformin + Placebo	Metformin + Spirulina tablet 0.8 g/day	Type 2 Diabetes	FBS, HbA1C, TC, TG, HDL-C, LDL-C
Mani et al. (2000) [27]	India	Intervention: 15 Control: 7 (Both gender)	Intervention: 47.80±9.10 Control: 53.40±6.13	Intervention: 29.24±0.30 Control: 25.75±0.13	-/-	Parallel/NR/NR	60 day	-	Spirulina tablet 2 g/day	Non-insulin dependent diabetes mellitus	FBS, TC, TG, HDL-C, LDL-C, VLDL
Parikh et al. (2001) [28]	India	Intervention: 15 Control: 10 (Both gender)	Intervention: 53.8 ± 7.2 Control: 54.6 ± 5.4	Intervention: 25.22±5.4 Control: 25.1±2.7	12.7/oral hypoglycemic agent	Parallel/Yes/NR	60 day	-	Spirulina tablet 2 g/day	Type 2 Diabetes	FBS, HbA1C, BSPP, TC, TG, HDL-C, LDL-C, VLDL
Lee et al. (2008) [26]	Korea	Intervention: 19 Control: 18 (Both gender)	Range: years Intervention: 52.1 ± 10.02 Control: 54.5 ± 6.36	Intervention: 23.8±2.17 Control: 23.4±2.12	2.45/without drug use	Parallel/Yes/NR	84 day	-	Spirulina tablet 8.0 g/day	Type 2 Diabetes	FBS, HbA1C, TC, TG, HDL-C, LDL-C
Kaur et al. (2008) [25]	India	Intervention: (1) 20 (2) 20 Control: 20 (Male)	Range: 40-60 years Intervention: (1) 46.3±7.60 (2) 45.95±7.15 Control: 47.6±6.70	NR	7.52/ oral hypoglycemic agent	Parallel/NR/NR	60 day	-	Spirulina tablet (1) 1 g/day (2) 2 g/day	Non-insulin dependent diabetes mellitus	FBS, HbA1C, BSPP, TC, TG, HDL-C, LDL-C, VLDL
Anitha et al. (2010) [23]	India	Intervention: 40 Control: 40 (Male)	Range: 45 – 60 years Intervention: NR Control: NR	NR	-/-	Parallel/NR/NR	84 day	-	Spirulina tablet 1g/day	Non-insulin dependent diabetes mellitus	FBS, HbA1C, TC, TG, HDL-C, LDL-C, VLDL
Alam et al. (2016) [22]	India	Intervention: 30 Control: 10 (Both gender)	Range: 41-60 years Intervention: 45.07 ± 7.67 Control: 44.00 ± 9.39	NR	-/-	Parallel/Yes/NR	45 days	Metformin 500 mg/day	Spirulina powder 7g/day	Type 2 Diabetes	FBS, HbA1C, BSPP
Beihaghi et al. (2017) [24]	Iran	Intervention: 20 Control: 20 (Both gender)	Range years: 30-60 Intervention: NR Control: NR	NR	5-10 /-	Parallel/Yes/NR	90 day	-	Spirulina tablet 8 g/day	Type 2 Diabetes	FBS, HbA1C

Lipoprotein; HDL-C: High-Density Lipoprotein; NR: Not Reported.

Abbreviations: FBS: Fasting Blood Sugar; HbA1C: glycated hemoglobin; BSPP: Blood Sugar post prandial; TG: Triglyceride; TC: Total-Cholesterol; LDL-C: Low-Density

Risk of bias assessment

Five trials were randomized [22, 24, 26, 28, 29], however, only one study [22] sufficiently addressed information around allocation concealment. Only 2 studies [22, 29] were blinded. The data regarding attrition and reporting biases were well-addressed in all trials. Table 2 presents the risk in each item of bias among included studies in detail.

Table 2.

The summary of review authors’ judgments about each risk of bias item for included studies.

Study	Random sequence generation	Allocation concealment	Blinding	Incomplete outcome data	Selective reporting	Other bias
Serban et l. (1982) [29]	L	U	L	L	L	U
Mani et al. (2000) [27]	U	U	H	L	L	U
Parikh et al. (2001) [28]	L	U	H	L	L	L
Lee et al. (2008) [26]	L	U	H	L	L	L
Kaur et al. (2008) [25]	U	U	H	L	L	L
Anitha et al. (2010) [23]	U	U	H	L	L	U
Alam et al. (2016) [22]	L	L	L	L	L	L
Beihaghi et al. (2017) [24]	L	U	H	L	L	U

H: high risk of bias; L: low risk of bias; U: unclear or unrevealed risk of bias. Criteria defined for risk of bias assessment are according to the Cochrane guidelines.

According to Cochrane criteria, study consider as a poor quality if it had high risk of bias in ≥2 items or unclear risk of bias in ≥3 criteria.

Meta-analysis

The effect of spirulina supplementation on glycemic related markers

The result of our meta-analysis suggested a significant effect of spirulina supplementation on FBS levels (−17.88 mg/dl; 95% CI: −26.99, −8.78; I²: 25%). However, no notable influence was detected for HbA1C (−0.12%, 95% CI: −0.70, 0.46; I² = 84%) or PPBS (−15.03%, 95% CI: −44.99, 14.92; I² = 0%) after spirulina intervention (Fig. 2).

Fig. 2.

Forest plot detailing mean difference and 95% confidence intervals (CI) for the effect of spirulina consumption on fasting blood sugar, hemoglobin A1C, blood sugar post-prandial

The effect of spirulina on blood lipid profiles

Pooled effect sizes revealed a significant reduction in TG (−30.99 mg/dl, 95% CI: −45.20, −16.77; I² = 50%), TC (−18.47 mg/dl, 95% CI: −33.54, −3.39; I² = 73%), LDL-C (−20.04 mg/dl, 95% CI: −34.06, −6.02; I² = 75%) and VLDL (−6.96 mg/dl, 95% CI: −9.71, −4.22; I² = 45%) following spirulina supplementation. In addition, the result indicated that spirulina supplementation yielded a significantl increase in HDL-C serum concentration (4.18 mg/dl, 95% CI: 1.67, 6.69; I² = 33%) (Fig. 3).

Fig. 3.

Forest plot detailing mean difference and 95% confidence intervals (CI) for the effect of spirulina consumption on lipid profiles

Meta-repression

The meta-regression revealed that the effect of spirulina supplementation on TC and LDL-C was inversely associated with baseline values (TC: coefficient = −0.89, P = 0.01; LDL-C: coefficient = −1.13, P = 0.005). In addition, the change in TG blood concentrations in response to spirulina intervention was related to the duration of the intervention (TG: coefficient = −0.87, P = 0.03). However, the change in of the remaining variables were independent of the dose of spirulina supplementation, duration of intervention, and baseline measures, respectively.

Sensitivity analysis and publication bias

Sensitivity analysis, by removing each RCT one by one, indicated that the pooled effect size of TG was non-significant after excluding Mani et al. (−16.77 mg/dl; 95% CI: −33.97, 0.43; I²: 77%). In addition, by removing Anitha et al. from TG overall effect size, the heterogeneity was altered from 50% to 7%, while the results remained significant (−24.51 mg/dl; 95% CI: −38.62, −10.39). The overall results of the remaining variables were not influenced by individual studies.

No evidence of publication bias was detected according to Egger’s regression asymmetry test and Begg’s rank-correlation methods in FBS (Begg’s test P = 0.67; Egger’s test P = 0.86), HbA1C (Begg’s test P = 0.85; Egger’s test P = 0.77), BSPP (Begg’s test P = 1.0; Egger’s test P = 0.99), TG (Begg’s test P = 0.65; Egger’s test P = 0.10), HDL-C (Begg’s test P = 0.54; Egger’s test P = 0.12), VDLD (Begg’s test P = 0.99; Egger’s test P = 0.12). Although Egger’s regression asymmetry test indicated significant evidence of publication bias in TC (Egger’s test: P = 0.005) and LDL-C (Egger’s test: P = 0.01); however, these results were not confirmed by Begg’s rank-correlation methods (TC: P = 0.45; LDL-C: P = 0.65).

Discussion

The present systematic review and meta-analysis suggests that spirulina supplementation can elicit a beneficial impact on metabolic parameters including FBS, TG, TC, LDL-C, HDL-C and VLDL. However, no favorable effect was observed in HbA1C and PPBS; which might be due to the low number of included studies that reported on these parameters. In addition, as HbA1C levels change over longer periods of time, it might that the duration of the included studies was not sufficient to truly reflect the efficacy of spirulina on decreasing in HbA1C. The meta-regression indicated that the change in blood TC and LDL-C concentrations was associated with baseline values; so that, higher baseline measures of TC or LDL-C led to greater reductions in blood concentration of these parameters. In addition, a greater reduction in TG was observed when the duration of spirulina supplementation was longer.

Patients with type 2 diabetes suffer from innumerable complications and are at risk of several additional diseases, such as non-alcoholic fatty liver [30] and cardiovascular disease [31]. The current study revealed a significant reduction in FBS and lipid profile following spirulina consumption. Although the mechanisms underlying the beneficial activity of spirulina are not well-understood, although several putative pathways are attributed to spirulina’s hypoglycemic and hypolipidemic activity. One of the bioactive components of spirulina is C-phycocyanin, a protein which can inhibit lipid peroxidation, scavenge free radicals, as well as enhance GSH peroxidase and superoxide dismutase activity [32, 33]. Spirulina can inhibit pancreatic lipase activity via glycolipid H-b2 [34, 35], and regulate cholesterol and prostaglandin synthesis via its gamma-linolenic acid components [36, 37]. Spirulina also possesses hypoglycemic properties via stimulation of insulin secretion from β-cell, or elevation of blood glucose transport to peripheral tissues by its protein and amino acid constituents [38].

The beneficial effect of spirulina on type 2 diabetes is not only related to the aforementioned parameters, but also associated with body weight and inflammatory factors, where both are involved with this disease [39–41]. Increases in body weight, especially abdominal obesity and inflammation, are associated with insulin resistance [42, 43], such that spirulina can also improve diabetes by weight loss activity and alleviation of inflammation through suppressing the NF-KB activity and reducing pro-inflammatory cytokines production [44, 45].

Spirulina is regarded to be generally safe in commonly-used doses and only rare cases of unwanted effect have been reported [46, 47]. Except for one study [24], none of the included studies reported evidence of adverse effects attributed to spirulina consumption. Beihaghi et al. [24] reported that participants experienced side-effects such as abdominal discomfort and diarrhea after 8 g/d spirulina consumption, which was alleviated in many of them after a few days. In addition, it has been shown that feeding mice for 7 days with 10 g/kg and 30 g/kg of body weight of dried and fresh spirulina, respectively, did not cause any form of toxicity [48]. However, there is a concern about contamination with low levels of mercury and other heavy metals from open water sources [47, 49], which should be avoided by controlling the growth and processing of spirulina.

There are several limitations which should be acknowledged in the present study. First, the number of included studies was somewhat low, and the duration of studies was relatively short. Second, there are several factors related to type 2 diabetes, such as insulin levels, insulin resistance, and homeostatic model assessment of insulin resistance (HOMA-IR), which are essential in the etiology of this disease. However, none of the included studies reported on these factors, and future studies should be conducted to investigate the effect of spirulina on these parameters. Finally, the quality of methodology of the included studies was low, and they had a significant risk of bias in several items. In this case, although the overall results indicated a promising effect on metabolic parameters in diabetic patients, these findings are not conclusive enough to utilize in clinical practice, and more clinical trials, with high-quality methodology are needed to affirm the efficacy of spirulina in diabetic treatment.

Conclusion

The present meta-analysis highlights that spirulina supplementation can yield improvements in FBS as well as lipid profiles. This study summarizes the currently available information from clinical trials and provides better insight into the effect of spirulina supplementation on type 2 diabetes. Spirulina is a natural functional agent, and generally safe supplement with a low cost, along with a beneficial impact on improving metabolic abnormalities manifest in type 2 diabetes. The favorable effects of spirulina suggest it may be a beneficial adjuvant therapy in conjunction with conventional medicine. However, the results of the present study should be considered as primary findings and further studies are needed to confirm the veracity of the results.

Abbreviations

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

FBS

fasting blood glucose

PPBS

postprandial blood sugar;

HbA1C

glycated hemoglobin A1C

TG

triglyceride; total cholesterol: total cholesterol

LDL-C

low-density lipoprotein cholesterol

HDL-C

high-density lipoprotein cholesterol

VLDL

very-low-density lipoprotein cholesterol

SD

standard deviation

Authors’ contributions

M.P., A.N. and A.H. carried out the concept, design and drafting of this study. A.S., S.GH. and M.P. searched databases, screened articles and extracted data. E.H., M.A., and F.J. performed the acquisition, analysis, and interpretation of data. C.C. and F.M-G. critically revised the manuscript. All authors approved the final version of the manuscript. F.J. and F.M-G. are the guarantors of this study.

Data availability

Not applicable.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no conflict of interest.