Efficacy of nutraceuticals (probiotics or prebiotics or synbiotics) in the prevention or treatment of COVID -19: a systematic review and meta-analysis

Abstract

Nutraceuticals have generated an interest among clinicians for their applicability in the prevention and treatment of many ailments. Literature suggests the possible role of nutraceuticals in COVID-19. However, substantial uncertainty related to their safety and efficacy still exists. The aim of this study is to assess the efficacy and safety of nutraceuticals in preventing or treating COVID-19. We searched electronic databases, registries, websites, and e-libraries of development agencies. We included randomized controlled trials (RCTs), quasi-experiments, pre-post studies, and other experimental study designs. We assessed the risk of bias (RoB) using the RoB2 tool for RCTs and the ROBINS-I tool for non-RCTs. We assessed the overall certainty of the evidence using GRADEpro GDT. The outcomes assessed were the number of COVID-19 cases, change in disease severity, days of hospitalization, deaths, and adverse events. We performed the last search on 13th December 2023. After screening 481 studies, our analysis of five revealed one study with ‘low risk’ and four studies with ‘some concerns’ regarding bias. Our findings indicated that nutraceuticals did not significantly reduce COVID-19 mortality; their effect on hospitalization duration was uncertain. Furthermore, our assessment found no substantial evidence of increased adverse effects with probiotic use (RR 0.11, 95% CI 0.01 to 1.89, P = 0.13). Importantly, none of the studies investigated nutraceuticals’ preventive effects or their role in mitigating COVID-19 severity. There is limited evidence available on the use of nutraceuticals (probiotics, prebiotics, or synbiotics) for the treatment and prevention of COVID-19. Conducting methodologically robust RCTs with large population samples is recommended.PROSPERO REGISTRATION ID CRD42021284923

Keywords:

• COVID-19
• Probiotics
• nutraceuticals
• mortality
• days of hospitalization

REVIEWING EDITOR:

• M. Luisa Escudero-Gilete, Universidad de Sevilla, Spain

SUBJECTS:

• Nutrition
• Nutraceuticals & Functional Foods

Introduction

Literature suggests the possible role of nutraceutical intervention in boosting immunity against COVID-19 (Akour, 2020; Infusino et al., 2020; Khaled, 2021; Morais et al., 2020; Olaimat et al., 2020; Sundararaman et al., 2020). Probiotics improve the mucosal-barrier function, antagonize pathogens, inhibit pathogen adherence, and enhance the immunity angulating the central nervous system (Stavropoulou & Bezirtzoglou, 2020). When the microbiota is disrupted, infectious diseases, autoimmune diseases, allergic conditions, and other diseases may arise. Probiotics can prevent and limit the symptoms of such diseases when administered as an adjuvant by conserving the intestinal microbiota and boosting the immune system (Stavropoulou & Bezirtzoglou, 2020).

In a recent research conducted by McCarty and colleagues, it has been suggested that certain nutraceuticals could potentially alleviate the effects of encapsulated RNA viruses in infected individuals, such as influenza and coronavirus, by boosting immune responses (McCarty & DiNicolantonio, 2020). Probiotics are viable, non-pathogenic bacteria that, when consumed, improve host health or physiology (George Kerry et al., 2018; Maldonado Galdeano et al., 2019). Ingestion of probiotics has been linked to a variety of positive impacts on human health as well as changes in the physiological homeostasis of the gut flora (Ballan et al., 2020; Bermudez-Brito et al., 2012; Markowiak & Śliżewska, 2017). Prebiotics are ideal fermented components that offer the potential for specific biological modifications, influencing the microbiota’s composition and/or function in the gastrointestinal (GI) tract, thereby promoting health benefits for the host (Khaled, 2021). The beneficial impacts of prebiotics on the gastrointestinal (GI) tract, including pathogen inhibition and immune system stimulation, stem from their ability to modulate the composition and function of the human microbiota (Olaimat et al., 2020).

On the other hand, the term ‘synbiotic’ describes a combination of probiotics and prebiotics that collaboratively enhance each other’s effects. They possess properties of both probiotics and prebiotics and were developed to address potential challenges in the survival of probiotics within the GI tract. Incorporating nutraceuticals (probiotics, prebiotics, or synbiotics) probiotics, prebiotics, or synbiotics into the human diet is advantageous for the intestinal microbiota (Markowiak & Śliżewska, 2017). The finest evidence for the effectiveness of certain probiotic strains derived from RCTs is there for treating or preventing antibiotic-associated disorders, gastroenteritis, acute diarrhoea, and relieving lactose intolerance (Aponte et al., 2020; Nazir et al., 2018). Probiotic microbial agents and their components defend the body through several mechanisms. Probiotic organisms either directly or indirectly inhibit the growth of traditional or potential microorganisms, along with their epithelial adhesion and invasion, by secreting antimicrobial compounds or by encouraging host expression of defensive molecules (Plaza-Diaz et al., 2019). Furthermore, increasing probiotic levels may induce a ‘barrier’ effect against common pathogens (Maldonado Galdeano et al., 2019). They can influence the host production of immunosuppressive molecules that reduce the inflammatory response in the host or, conversely, activate host protective immunologic processes that can prevent or hasten the clearance of pathogenic infections (Ewald & Sumner, 2018).

Prebiotics are resistant to digestion by host enzymes. They reach the digestive system largely unchanged and undergo fermentation by saccharolytic bacteria like those from the Bifidobacterium genus in the colon (Markowiak & Śliżewska, 2017). The intake of prebiotics has a notable impact on the composition and metabolic functions of the intestinal microbiota. The main goal is to support the growth and functionality of beneficial bacteria, or probiotics, in the gastrointestinal (GI) tract, ultimately bestowing health advantages upon the host (Markowiak & Śliżewska, 2017). Commonly include prebiotics are fructans, oligosaccharides, isomaltooligosaccharides, arabinooligosaccharides, xylooligosaccharides, lactosucrose, lactobionic acid, resistant starch, psyllium, galactomannan, polyunsaturated fatty acids, and polyphenols (Markowiak & Śliżewska, 2017; Olaimat et al., 2020). The health advantages that prebiotics bring to the GI tract, such as pathogen inhibition and immune system stimulation, result from their capacity to regulate the composition and activity of the human microbiota (Olaimat et al., 2020). Prebiotics may also have an excellent potential effect against COVID-19 by enhancing probiotics’ growth and survivability. Furthermore, prebiotics could have a direct effect on GI symptoms caused by COVID-19 by blocking the Angiotensin-converting enzyme (ACE) enzymes (Olaimat et al., 2020).

On the other hand, the impact of synbiotics on metabolic health remains uncertain, and it is worth noting that the health effects of synbiotics are likely linked to the specific combination of a probiotic and prebiotic (Markowiak & Śliżewska, 2017). Among synbiotic products, combinations featuring Bifidobacterium or Lactobacillus genus bacteria with fructooligosaccharides appear to be particularly popular (Markowiak & Śliżewska, 2017).

Probiotics have recently earned the interest of physicians or clinicians due to their utilization for treating and preventing several disorders. The existing understanding indicates a connection between the gut microbiota and the liver, recognized as the ‘microbiota-gut-liver axis.’ Additionally, there is an acknowledged interaction between the intestinal microbiota and the central nervous system, referred to as the ‘microbiota-gut-brain axis.’ Moreover, there is emerging evidence supporting bidirectional communication between the lungs and the gut, termed the ‘gut-lung axis. Some people suffering from COVID-19 can present fatigue, such as ‘brain fog’, pain, even after several months, signifying chronic association of the CNS in COVID-19. In addition, gut microbiome disturbances can severely affect the brain, and its functions and restoration could have a beneficial effect. Probiotic supplementation has been observed to improve metabolic homeostasis and prevent fatigue development by enhancing the metabolites concerned with the utilization of glucose and energy (Santinelli et al., 2021). The significance of the ‘gut-brain’ axis has been connected to overall mental health (Dhar, 2021). Probiotics can have anxiolytic effects via alteration of gut microbiota (Dhar, 2021). Thus, gut microbiome disturbances can have a severe adverse impact on the brain and its function, and their restoration can benefit people suffering from these conditions.

Despite the elevated prevalence of COVID-19, effective medicines or therapies remain few, and no definite pharmaceutical treatment is available to target the disease’s relevant components. Several studies have demonstrated the positive and encouraging effects of probiotics, prebiotics, or synbiotics in treating or preventing COVID-19 (Akour, 2020; Giannoni et al., 2020; Infusino et al., 2020; Mak et al., 2020; Morais et al., 2020; Sundararaman et al., 2020). The positive effect of nutraceuticals (probiotics, prebiotics, or synbiotics) cannot be disregarded, but substantial uncertainty related to their safety and efficacy in clinical practice still exists (Pandey et al., 2022). Furthermore, the safety and effectiveness of the same in COVID-19 patients have not been well studied or systematically reviewed. There is a need to synthesize the information or evidence so that patients, practitioners, and policymakers may determine if these nutraceuticals can be used to prevent or treat COVID-19 and, if data permits, investigate the best pharmacological program for this group of patients. With this background, the objective of this systematic review is to assess the safety and efficacy of nutraceuticals (probiotics, prebiotics, or synbiotics) in preventing or treating COVID-19.

Methods

The systematic review was conducted using a standard methodology suggested in the Cochrane Handbook of Systematic Reviews (Higgins et al., 2021). CRD42021284923 is the registration number of the proposed protocol submitted in PROSPERO. The International Life Sciences Institute (ILSI), India, funded this work.

Inclusion and exclusion criteria

We included randomized controlled trials (RCTs), quasi-experiments, pre-post studies, and other experimental study designs reporting the efficacy of nutraceuticals (probiotics, prebiotics, or synbiotics) to prevent or treat COVID-19 and excluded case reports, observational studies, clinical observations, systematic reviews, and narrative reviews. We did not include grey literature, reports, abstracts, or conference proceedings. We restricted the language of publishing and included articles published only in English.

We included studies investigating participants affected/exposed to COVID-19 infection, regardless of age, gender, ethnicity, and setting (inpatients and outpatients), and excluded severely ill patients.

We included studies on supplementation of nutraceuticals (probiotics, prebiotics, or synbiotics) as an adjunct to standard treatments, irrespective of the form, dose, duration, or frequency. Comparators could consist of the standard of care without supplementation of probiotics, prebiotics, synbiotics, or any active comparator (such as appetizers, nutritional supplements, etc) and placebo. We also included studies that administered other adjunct interventions to the standard treatment (such as ozone oxygen therapy).

Following outcomes were considered in the review:

Primary outcomes

1. Number of cases of COVID-19

2. Change in disease severity

3. Days of hospitalization

Secondary outcomes

1. Number of deaths

2. Adverse events

3. Free of fatigue (Post-Covid fatigue)

Search methods for identification of studies

We searched for any probiotics, prebiotics, or synbiotics intervention provided to COVID-19. We searched Cochrane Central Register of Controlled Trials (CENTRAL) (via the Cochrane Library; 2021, issue 9), MEDLINE (via PubMed), Web of Science, WHO COVID-19 database (https://search.bvsalud.org/global-literature-on-novel-coronavirus-2019-ncov/), WHO International Clinical Trials Registry (ICTRP) (apps.who.int/trialsearch/), clinicaltrials.gov (www.clinicaltrials.gov) for ongoing trials. We also searched Google Scholar, the WHO Regional Journals databases from Latin America, Africa, and South-East Asia, websites, and e-libraries of development agencies (WHO Department of Nutrition for Health and Development, UNICEF, Nutritional International, International Food Policy Research Institute (IFPRI). Additionally, we checked the reference lists of reviews and retrieved articles.

As the COVID-19 pandemic emerged in December 2019, we considered including studies between 2019 to 2023 and the last search was performed on 13th December 2023. The search included relevant keywords, free-text terms, and Medical Subject Headings (MeSH). The search strategy for PubMed, CENTRAL, and WHO COVID-19 databases are presented in Appendix 1:

Selection of studies

After removing duplicates, two investigators (SU and MW) independently screened all the articles retrieved from the searches using the Covidence screening tool (Covidence systematic review software, 2019) in two phases. In the first phase, we screened all articles against the inclusion criteria based on titles and abstracts. The studies found eligible in the first screening phase were then screened based on full texts by two investigators. Disagreements were resolved by consensus. We recorded the flow of studies in the systematic reviews in the form of a PRISMA-flow diagram as suggested in ‘Cochrane Handbook for Systematic Reviews of Intervention’ (Higgins & Green, 2011).

Data extraction and management

Two reviewers (SU and MW) extracted data from each eligible study using a pilot-tested data extraction form, including study design, country, sample size details, characteristics of study participants (age, sex, race/ethnicity, health status), intervention details, dependent variables, and their measurement, and funding sources. One investigator reviewed all the extracted data for accuracy.

Assessment of risk of bias in included studies

Two investigators (APS and MNK) assessed the risk of bias for RCTs using Cochrane’s Risk of Bias Tool 2 (RoB2) and Non-RCTs using the ROBINS-I tool (Higgins & Green, 2011). Discrepancies amongst investigators were resolved through consensus and in consultation with a third reviewer.

Measures of treatment effect

One reviewer inputted the results of each trial into Review Manager 5 (Review Manager, 2014) for the computation of treatment effects. We intended to perform meta-analyses selectively, focusing on situations where combining results made sense, specifically when participants, interventions, and the underlying clinical question were sufficiently similar to justify pooling.

We classified the studies based on the outcomes reported. For dichotomous outcomes (deaths and adverse events), we expressed the results as risk ratios (RRs) with 95% confidence intervals (CIs) for analyses. We had planned to use mean differences (MDs) and 95% CIs or standardized mean differences (SMDs) with 95% CIs to express continuous variables (days of hospitalization). We provided a narrative description of the results when the data were inadequate. For each outcome, we examined the reported effect sizes and the heterogeneity (I² statistic) across studies. We designated an I² statistic ranging from 0% to 40% as indicative of low heterogeneity, 41% to 60% as moderate heterogeneity, 61% to 90% as substantial heterogeneity, and values exceeding 91% as representing considerable heterogeneity.

We had planned to use fixed‐effect analysis unless there was significant heterogeneity (P < 0.1) as assessed by the Chi² test, where we had planned to use random‐effect analysis. Summary of the result of dichotomous and continuous outcomes was mentioned in Appendix 2 as A2 Table 1 and A2 Table 2. Individual participants in each clinical trial were the unit of analysis.

Regarding strategies for dealing with missing data, we adapted the Cochrane Handbook for Systematic Reviews of Interventions recommendations (Higgins et al., 2021). We contacted the investigators for clarification when the missing data was not explicit. If the outcome data for standard deviations (SD) was missing, we calculated it from 95% confidence intervals as suggested in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins et al., 2021).

Other methods, including assessment of heterogeneity, reporting biases, subgroup analysis, sensitivity analysis and investigation of heterogeneity, have been taken from the Cochrane Handbook for Systematic Reviews of Interventions (Higgins et al., 2021).

Two review authors (SZQ and MNK) independently assessed the overall certainty of the evidence for three outcomes and graded using the GRADE approach (GRADEpro GDT, 2021). We reported an overall assessment of the certainty of the evidence for the number of deaths, number of adverse events, and days of hospitalizations in Summary of Findings (SoF) Table using GRADE pro-GDT (GRADEpro GDT, 2021).

Results

We identified 481 potential studies from different electronic databases. After removing duplicates, two reviewers independently examined 416 studies and, after initial screening (based on title and abstract), removed 376 studies. We then independently assessed the full texts of 40 potentially relevant studies and excluded 35 studies based on different study designs (n = 17), participants other than those with COVID-19 (n = 13) and outcomes not of interest (n = 5). We identified five studies for potential inclusion (Araimo et al., 2021; Gutiérrez-Castrellón et al., 2022; Ivashkin et al., 2023; Rathi et al., 2021; Reino-Gelardo et al., 2023). All five studies (Araimo et al., 2021; Gutiérrez-Castrellón et al., 2022; Ivashkin et al., 2023; Rathi et al., 2021; Reino-Gelardo et al., 2023) were included for quantitative synthesis (meta-analysis). We have presented a flow diagram detailing the selection of studies in Figure 1.

Figure 1. PRISMA Flow diagram detailing the selection of studies.

We included three single-centric and one multicentric RCT published between 2019 and 2023 year (Table 1). The studies included were two open labelled (Araimo et al., 2021; Ivashkin et al., 2023), one double-blinded (Rathi et al., 2021), one quadruple-blinded (Gutiérrez-Castrellón et al., 2022), and one non-blinded (Reino-Gelardo et al., 2023). The studies were conducted in Spain (Reino-Gelardo et al., 2023), Mexico (Gutiérrez-Castrellón et al., 2022), Italy (Araimo et al., 2021), Moscow (Ivashkin et al., 2023), and India (Rathi et al., 2021).

Table 1. Characteristics of included studies.

Review ID (Country)	Study design	Sample size details	Inclusion criteria	Exclusion criteria	Characteristics of study participants	Intervention details	Comparison group	Outcomes	Conclusion	Comments
(Reino-Gelardo et al., 2023) (Spain)	Prospective, non-blinded, RCT (NCT04666116)	Total: N = 139 IG:n = 70 CG: n = 69	1. RT-PCR-positive COVID-19 patients 2. Patients required >48 hours of hospital admission	1. age <18 years 2. Hospital discharge in the first 48 hours.	Age (Mean ± SD): IG: 70 ± 16 years CG: 69 ± 17 years Male/Female ratio: IG: 40/30 CG: 38/31	Standard treatment + Probiotics and prebiotics (Gasteel Plus), one stick daily (300 mg), dissolved in water	Standard treatment	Outcomes of interest:1. Days of hospitalization (days) Outcomes not of interest:1. Duration of gastrointestinal symptoms (days).	Gasteel Plus showed protective benefits in severe COVID-19 cases, facilitating faster recovery from digestive symptoms and reducing hospital stays.	Gasteel Plus contains Bifidobacterium lactis BPL1, Bifidobacterium longum ES, Lactobacillus rhamnosus CNCM I-4036, in 1:1:1 ratio and FOS (200 mg)
(Gutiérrez-Castrellón et al., 2022) (Mexico)	single-centre, parallel, quadruple-blinded, RCT (NCT04517422)	Total: N = 293 IG: n = 147 CG: n = 146	1. Symptomatic COVID-19 outpatient 2. Age 18 to 60 years 2. Positive RT-qPCR for the SARS-CoV2 RdRp gene 4. SpO2 ≥ 90% 5. Ability to understand study procedures.	1. Uncontrolled diabetes and systolic hypertension 2. BMI ≥40 Kg/m^2. 3. Coagulopathy and acute pancreatitis. 4. Chronic constipation and diarrhoea or IBD 7. Currently immunosuppressed 8. Severe respiratory disease 9. Probiotic or antibiotic use within 2 days before entering the study 10. Pregnancy or lactation.	Age (Median, IQR): IG: 34 (26–45) years CG: 39 (27–49) years Male/Female ratio: IG: 68/82 CG: 71/79	AB21 probiotic capsules (Lactiplantibacillus plantarum KABP033, L. plantarum KABP022, L. plantarum KABP023 and Pediococcus acidilactici KABP021 at a dose of ≥2 × 10^9 total CFU, with a maltodextrin carrier for 30 days	Placebo capsules containing the maltodextrin carrier	Outcomes of interest:1. No of Death at Day 30 2. Hospitalized (moderate/severe) n,% 3. Adverse events Outcomes not of interest:1. Complete remission (n,%) 2. Days of ICU stay	Probiotic supplementation was found to be well-tolerated compared to placebo.	Future studies should replicate these findings and elucidate its mechanism of action.
(Araimo et al., 2021) (Italy)	Single centric, Open-label, RCT. (NCT04366089)	Total: 28 IG: 14 CG:14	1. Age > 18 years 2. Nasopharyngeal swab positive for COVID-19 3.COVID-19 stages III 4.Hospitalized in wards	1.COVID-19 StagIV-VI–VI 2.Hospitalization in ICUs 3.Pregnancy 4. G6PD Deficiency 5.Patients who deny consent 6.Contra- indications oxygen-ozoneone therapy	Age (Mean ± SD): IG: 63.3 ± 12.2 years CG: 60.1 ± 14.4 years Male/Female ratio: IG: 9/5 CG: 7/7	Standard care (referred to as ‘ad interim BAT: antibacterial + antiinflammatoryanti-cytokineine) + Probiotics (SivoMixx* one sachet every 12 h for 7 days) + Oxygen-ozone therapy (5 × 10^3 mcg of ozone for 7 days)	Standard care (referred to as ‘ad interim BAT: antibacterial, anti-inflammatory + anticytokine	Outcomes of interest: 1.No of Deaths: Day 7, 14, 30 2. Adverse events Outcomes not of interest: 1.Requiring orotracheal intubation	Systemic ozone therapy is safe in critically‐ill COVID‐19 patients Efficacy of ozone autohemotherapy including probiotic supplements as an adjunct in COVID‐19 remains unclear.	There is a discrepancy in data reported in the abstract and in the result section. However, we have taken data from the result section
(Ivashkin et al., 2023) Moscow, Russia Federation	Randomized, controlled, single-center, open-label trial (NCT04854941)	Total: N = 200 IG:n = 99 CG:n = 101	1.Age: 18–75 years 2.Confirmed COVID-19 by PCR on nasopharyngeal and oropharyngeal swabs 3. Admitted patients	1. Age: > 75 years or < 18 years 2. Consumption of probiotics for 3 months prior to admission 3. History of intolerance to probiotics or their components 4.Refusal to participate. 5.Pregnant women 6.Cancer or mental illness 7. Severe renal dysfunction (GFR < 50 mL/ min) 8.Discontinued consumption of probiotics.	Age IG: 65 (59–71) years CG: 64 (54–70) years Male/Female ratio: IG: 44/55 CG: 48/53	Standard Care (Dexamethasone + antiviral, antibacterial + anticoagulant + anticytokine) + Probiotics Probiotics Dosage: Tds for no more than 14 days)	Standard Care (Dexamethasone + antiviral, antibacterial + anticoagulant + anticytokine	Outcomes of interest: 1.Days of hospitalization: 2.No. of deaths n 3.No. of adverse events Outcomes not of interest: 1.Admission to 2.Need for mechanical ventilation 3.Oxygen support 4.Duration of oxygen support 5.Biomarkers	No effect on mortality Effective in preventing hospital-acquired diarrhea and treating diarrhea associated with COVID-19	Probiotics include Lacticasei­bacillus rhamnosus PDV 1705, Bifdobacterim bifdum PDV 0903, Bifdobacterim longum subsp. infantis PDV 1911, Bifdobacterium longum subsp. longum PDV 2301
(Abhijit Rathi et al., 2021) (India)	Randomized, multicentric, double-blind placebo-controlled trial. (CTRI/2021/05/033576)	Total 200 IG: 100 CG: 100	1.Aged ≥18 and ≤75 years 2.Not suffering from active SARS-CoV-2 with a complaint of post-COVID fatigue/ muscle weakness. 3.History of COVID-19 in past 4.Non-pregnant, non-lactating women 5.Able to take the drug orally	1.Critical condition 2.FOS: ≤90% 3. PaO2/ FiO2:≤300 mmHg 4.Respiratory failure 6.Viral pneumonia 7.Unable to take food or drugs 8Consumption of another oral probiotic during the trial 9.Pregnancy 10.Allergic conditions 11.Hb < 8 mg/dL	Age (years); SD; Range- IG: 41.29 ± 13.0 (20–75) CG: 41.17 ± 12.9 (20–75) Males: Females (%)- IG: 65:35 CG: 62:38. Subjects with co-morbidities (%)- IG: 17 CG: 12 Days to enrolment post negative COVID test; Range- IG: 18.5 (2 to 48) CG: 20.6 (4 to 87)	ImmunoSEB (500 mg/capsule) + ProbioSEB CSC3 (5 billion CFUs /capsule) ProbioSEB CSC3 contains Bacillus coagulans LBSC, B.subtilis PLSSC and B.clausii 088AE	Placebo as maltodextrin for 14 days	Outcomes of interest: Nil Outcome not of interest Free of fatigue (91% vs. 15%)	Supplementation of ImmunoSEB + ProbioSEB CSC3 for 14 days resolves post-COVID-19 physical and mental fatigue. It is also effective in the recovery of COVID-19 patients if presc ribed early.	ImmunoSEB (multi-enzyme formulation of Peptizyme SP, bromelain, amylase, lysozyme, peptidase, catalase, papain, glucoamylase and lactoferrin) Fatigue assessment was done by CFQ-11 (self-report instrument).

FOS: Fructoligosaccharides, IBD: Inflammatory bowel disease, BAT: Best Available Therapy, G6PD: glucose-6-phosphate dehydrogenase, HCQS: hydroxychloroquine, CRP: C-Reactive protein, VSS: Ventilatory support system, PCT: procalcitonin, KFT: Kidney function test, LFT: Liver function test, PCR: Polymerase Chain Reaction, GFR: glomerular filtration rate, FOS- Finger Oxygen Saturation, PaO2- Pressure in arterial blood, Hb- Hemoglobin Concentration, FiO2- Fraction of inspired oxygen, *SivoMixx: Streptococcus therphilus, DSM322245, Bifidobacterium,lactis DSM32246, Bifidobacterium lactis DSM32247, Lactobacillus acidophilus DSM32241, Lactobacillus helveticus DSM32242, Lactobacillus para-casei DSM32243, Lactobacillus plantarum DSM32244, Lactobacillusbrevis DSM2796.

The total number of participants included in the five trials were 860, including both males and females (Araimo et al., 2021; Gutiérrez-Castrellón et al., 2022; Ivashkin et al., 2023; Rathi et al., 2021; Reino-Gelardo et al., 2023). The age of the participants ranged between 18 to 75 years. All the 860 participants were diagnosed with COVID-19. Further, Araimo et al.(Araimo et al., 2021) included participants of COVID stage III admitted to the hospital wards andexcluded participants who were suffering from Stage IV-VI COVID-19 and were in ICU.

All the studies included had two treatment arms. In three trials (Araimo et al., 2021; Ivashkin et al., 2023; Reino-Gelardo et al., 2023), probiotics were used as an adjuvant treatment to the standard care compared with standard treatment alone. Bacterial strains used as probiotics in these trials were species of Bifidobacterium (Bifi.), lactobacillus, and Streptococcus. In another two trials (Gutiérrez-Castrellón et al., 2022; Rathi et al., 2021), probiotics were given as asupplement or as an adjuvant to the immuno booster therapy compared to a placebo containing maltodextrin.

In a trial conducted by Reino-Gelardo et al., (Reino-Gelardo et al., 2023), food supplement (Gasteel Plus that includes Bifidobacterium. Lactis BPL1, lactobacillus rhamnosus CNCM I-4036, Bifi. longum ES1) was provided along with the standard treatment that was based on the international guidelines available during the study considering the severity and complexities of the disease. In one trial (Araimo et al., 2021), participants were given standard care consisting of antibiotics (azithromycin), anti-inflammatory (Hydroxychloroquine sulphate), and anti-cytokines (Tocilizumab) along with probiotics supplementation of SivoMixx 200 billion six sachets twice a day. However, in the other trial (Ivashkin et al., 2023), standard care encompassed steroids (Dexamethasone), antiviral (Favipiravir), and anticoagulant (enoxaparin) with probiotics supplementation comprising Florasan-D, Bifidobacterium bifidum PDV 0903, Lacticaseibacillus rhamnosus PDV 1705, Bifidobacterium longum PDV 2301, Bifidobacterium longum subsp, Infantis PDV 1911 for three times a day for 14 days. A trial conducted by Rathi et al. (Rathi et al., 2021) evaluated the efficacy of ImmunoSEB (multi-enzyme formulation of Peptizyme SP, bromelain, amylase, lysozyme, peptidase, catalase, papain, glucoamylase and lactoferrin) and ProbioSEB CSC3 (5 billion CFUs/capsule) that included Bacillus coagulans LBSC, B. subtilis PLSSC and B. clausii 088AE) prescribed twice daily in patients previously suffering from COVID-19 compared with placebo. a Further, another trial conducted by Gutierrez-Castrellon et al. (Gutiérrez-Castrellón et al., 2022) assessed the efficacy of AB21 probiotic capsules plus maltodextrin as a carrier compared with placebo containing only maltodextrin in symptomatic COVID-19 patients. We did not find any study where nutraceuticals (probiotics, prebiotics, or synbiotics) were given to normal subjects to prevent COVID-19. In addition, no studies reported data about the efficacy of nutraceuticals (probiotics, prebiotics, or synbiotics) in changing the severity of COVID-19. The outcome measures mainly addressed in the included studies were the number of deaths, adverse events, and days of hospitalization. Only Ivashkin et al. and Reino-Gelardo et al. (Ivashkin et al., 2023; Reino-Gelardo et al., 2023) reported the number of hospitalization days. Similarly, Araimo et al. (Araimo et al., 2021) and Gutiérrez-Castrellón (Gutiérrez-Castrellón et al., 2022) reported adverse events. Ivashkin et al. (Ivashkin et al., 2023) and Araimo et al. (Araimo et al., 2021) reported the number of deaths due to COVID-19, whereas Rathi et al. (Rathi et al., 2021) reported data on post-covid fatigueAdverse events were reported in three trials (Araimo et al., 2021; Gutiérrez-Castrellón et al., 2022; Ivashkin et al., 2023).

Risk of bias in included studies

One trial (Gutiérrez-Castrellón et al., 2022) was at ‘low risk’ of bias, whereas the other four trials (Araimo et al., 2021; Ivashkin et al., 2023; Reino-Gelardo et al., 2023) were at ‘Some concerns’ for risk of bias. We have presented a review authors’ judgment for each risk of bias for each study in the ‘Risk of bias’ summary (Figure 2).

Figure 2. Risk of bias summary: review authors’ judgments about each risk of bias item for each included study.

Domain 1: D1: Randomisation process

There were ‘some concerns’ in the randomization process in three trials (Araimo et al., 2021; Ivashkin et al., 2023; Reino-Gelardo et al., 2023) and ‘low risk’ for two trials (Gutiérrez-Castrellón et al., 2022; Rathi et al., 2021). In three trials (Gutiérrez-Castrellón et al., 2022; Rathi et al., 2021; Reino-Gelardo et al., 2023) randomization method was mentioned whereas the only information about randomization methods in two trials (Araimo et al., 2021; Ivashkin et al., 2023) with ‘some concerns’ is a statement that the trial is randomized. Also, no information is available about the allocation concealment in three trials with ‘some concerns’ (Araimo et al., 2021; Ivashkin et al., 2023; Reino-Gelardo et al., 2023). Whereas two trials (Gutiérrez-Castrellón et al., 2022; Rathi et al., 2021) judged to be at ‘low risk’ of bias used randomization method, generated the allocation sequence and used concealed envelopes for treatment allocation.

Domain 2: Deviations from the intended effect

‘Some concerns’ were reported in this domain in two trials (Araimo et al., 2021; Ivashkin et al., 2023) and ‘low risk’ for three trials (Gutiérrez-Castrellón et al., 2022; Rathi et al., 2021; Reino-Gelardo et al., 2023). Two trials (Araimo et al., 2021; Ivashkin et al., 2023) had ‘some concerns,’ as no information was available on whether the participants, caregivers, and/or people delivering the interventions were aware of their assigned intervention during the trial. In another three trials, one trial (Rathi et al., 2021) with ‘low risk’ was a double-blind trial wherein the participants and people delivering the interventions were blinded to their assigned intervention during the trial. The investigators have undertaken intention-to-treat (ITT) analyses. Another trial (Gutiérrez-Castrellón et al., 2022) was quadruple-blinded, where all participants and caregivers were unaware of the treatment, whereas the other trial (Reino-Gelardo et al., 2023) with ‘low risk’.

Domain 3: Missing outcome data

All the five included trials (Araimo et al., 2021; Gutiérrez-Castrellón et al., 2022; Ivashkin et al., 2023; Rathi et al., 2021; Reino-Gelardo et al., 2023) were at ‘low risk’ for missing outcome data. A trial conducted by Reino-Gelardo et al. (Reino-Gelardo et al., 2023) reported a 5% loss to followup, and another trial (Gutiérrez-Castrellón et al., 2022) reported a < 10% loss to follow up. Only two participants in the intervention group were lost to follow-up in one (Ivashkin et al., 2023) of the included trials. The other two trials (Araimo et al., 2021; Rathi et al., 2021) reported the participants’ 100% follow-up compliance.

Domain 4: Measurement of the outcome

All the five included trials were at ‘low risk’ of bias for outcome measurement. The methods of measurement of outcomes included in this review (days of hospitalization, number of deaths, and number of adverse events) were appropriate. They probably may not have differed between intervention groups in all the included trials. The outcome assessors were not blinded in the open-labelled trial (Ivashkin et al., 2023) and were blinded in the two trials (Gutiérrez-Castrellón et al., 2022; Rathi et al., 2021). The trial by Araimo et al., (2021) did not provide information about the blinding status of the outcome assessors, whereas Reino-Gelardo et al. (Reino-Gelardo et al., 2023) conducted a non-blinded study.

Domain 5: Selection of the reported results

Two trials (Araimo et al., 2021; Rathi et al., 2021) had ‘some concerns’, while three trials (Gutiérrez-Castrellón et al., 2022; Ivashkin et al., 2023; Reino-Gelardo et al., 2023) were at ‘low risk’ for this domain. The pre-specified analysis plans reported in the protocols of the two included trials (Araimo et al., 2021; Rathi et al., 2021) were not available in sufficient detail to compare with those presented in the published report(s). In the other three trials (Gutiérrez-Castrellón et al., 2022; Ivashkin et al., 2023; Reino-Gelardo et al., 2023), the protocol was available with a registered ID.

Effects of interventions

Comparison 1: Nutraceuticals (Probiotics or prebiotics or synbiotics) as a adjunct to standard treatments verses Standard of care without supplementation of probiotics or prebiotics or synbiotics

We did not find any study that assessed, the number of cases of COVID-19, change in disease severity, and fatigue in this comparison group.

Days of hospitalization

Two studies (Ivashkin et al., 2023; Reino-Gelardo et al., 2023) reported data on days of hospitalization. The pooled analysis shows reduction in the days of hospitalization with probiotics and prebiotics with standard care as compared to only standard care (two studies, 339 participants, MD -1.49, 95% CI -4.34 to 1.35, P = 0.30, I²=73%) (Figure 3).

Figure 3. Forest plot of comparison: Standard care + Probiotics verses Standard care, outcome: Days of hospitalization Number of deaths.

In a trial by Araimo et al. (Araimo et al., 2021), assessing probiotics and ozone therapy alongside standard care versus standard care alone, no significant difference in the risk of death was observed between the intervention and comparison groups. Similarly, in a trial by Ivashkin et al. (Ivashkin et al., 2023), comparing probiotics as an adjunct to standard care with standard care alone, the risk of death in both groups was found to be nearly identical. The pooled analysis does not show any reduction in the number of deaths with prebiotics (two studies, 228 participants, RR 1.02, 95% CI 0.3 to 3.41, P = 0.98, I²=0%) (Figure 4).

Figure 4. Forest plot of comparison: Standard care + Probiotics verses Standard care, outcome: Deaths.

Adverse events

One trial (Araimo et al., 2021) compared the efficacy of probiotics and ozone therapy given as an adjunct to standard care verses standard care (without probiotics and ozone therapy). No adverse events were reported in the intervention group while four participants (30%) reported diarrhoea (one study, 28 participants, RR 0.11,95% CI 0.01 to 1.89, P = 0.13).

Comparison 2: Nutraceuticals (Probiotics orprebiotics or synbiotics)an adjunct to standard treatments versus Any active comparator (such as appetizers, nutritional supplements, etc.)

We did not find any study comparing nutraceuticals (probiotics, prebiotics, or synbiotics) adjuncts to standard treatments with an active comparator such as appetizers or nutritional supplements, etc.

Comparison 3: Nutraceuticals (Probiotics or prebiotics or synbiotics) as an adjunct to standard treatments verses placebo

Except for the adverse events, we did not find any study that assessed any of the other outcomes in this comparison group.

Adverse events

One study (Gutiérrez-Castrellón et al., 2022) compared the adverse events in probiotics and prebiotics versus placebo and found that the risk of events was 35% lesser with probiotics and prebiotics as compared to placebo (one study, 300 participants, RR 0.65, 95% CI 0.47 to 0.90, P = 0.009) Fatigue.

We found only one study (Rathi et al., 2021) that addressed this outcome. In this study, fatigue evaluation utilized a validated Chalder Fatigue Scale (CFQ-11), an 11-item self-report tool (Rathi et al., 2021). A cumulative score of four or higher indicates meeting the fatigue criteria.

The supplemental administration of prebiotics and immune boosters made patients free of fatigue in a significantly greater percentage of subjects in the intervention group compared to the control arm on day 14 (one study, 200 participants, RR 6.07, 95% CI 3.79 to 9.71). Positive outcomes were observed at earlier time intervals, revealing that a higher percentage of patients in the experimental group experienced relief from fatigue on days 4 and 8 compared to those who received the intervention at later stages.

Quality of evidence

The quality of evidence (QoE) was ‘low’ for a number of deaths and adverse events, and ‘very low’ for Days of hospitalization for the comparison ‘Standard care + Nutraceuticals (Probiotics/Prebiotics) compared to Standard care for Covid-19’.The QoE was downgraded for the reasons mentioned in the explanatory notes of Table 2.

Table 2. Summary of findings table for comparison: standard care + nutraceuticals (probiotics/prebiotics/synbiotics) compared to standard care.

Standard care + nutraceuticals (probiotics/prebiotics/synbiotics) compared to standard care for Covid-19
Patient or population: Covid-19 Intervention: Standard care + nutraceuticals (probiotics/prebiotics/synbiotics) Comparison: Standard care only
Outcomes	Anticipated absolute effects*(95% CI)		Relative effect (95% CI)	№ of participants (studies)	Certainty of the evidence (GRADE)	Importance
	Risk with standard care only	Risk with standard care + nutraceuticals (probiotics/prebiotics/synbiotics)
Deaths	43 per 1,000	44 per 1,000 (13 to 148)	RR 1.02 (0.30 to 3.41)	228 (2 RCTs)	⨁⨁◯◯ Low^a,b,c	Important Outcome
Adverse events	286 per 1,000	31 per 1,000 (3 to 540)	RR 0.11 (0.01 to 1.89)	28 (1 RCT)	⨁⨁◯◯ Low^c,d	Important Outcome
Days of hospitalisation	The mean days of hospitalisation was 0	MD 1.62 lower (3.09 lower to 0.14 lower)	–	339 (2 RCTs)	⨁◯◯◯ Very low^,c,e	Important Outcome

* The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

CI: confidence interval; MD: mean difference; RR: risk ratio.

GRADE Working Group grades of evidence:

High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.

Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.

Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Explanations

a. The QoE was downgraded by one level as the included studies did not report the method of randomization and were open-labeled RCTs.

b. The QoE was downgraded by one level as both the individual included studies as well as the pooled analysis did not report significant benefit or harm.

c. The sample size is below the OIS.

d. The QoE was downgraded by one level as the included study did not report the method of randomization and was open-labeled RCT.

e. The QoE was downgraded by one level due to the presence of heterogeneity (I²=73%).

The quality of evidence (QoE) was ‘moderate’ for adverse events for the comparison of ‘Nutraceuticals (Probiotics/Prebiotics/Synbiotics) versus placebo for COVID-19’. The QoE is downgraded by one level as the sample size is less than the Optimal Information Size (OIS) (Table 3).

Table 3. Summary of findings table for comparison: Nutraceuticals (Probiotics/Prebiotics/Synbiotics) verses placebo.

Nutraceuticals (probiotics/prebiotics/synbiotics) compared to placebo for Covid-19
Patient or population: Covid-19 Intervention: Nutraceuticals (probiotics/prebiotics/synbiotics) Comparison: Placebo
Outcomes	Anticipated absolute effects*(95% CI)		Relative effect (95% CI)	№ of participant (studies)	Certainty of the evidence (GRADE)	Importance
	Risk with placebo	Risk with Nutraceuticals (Probiotics/Prebiotics/Synbiotics) verses Placebo
Adverse events	420 per 1,000	273 per 1,000 (197 to 378)	RR 0.65 (0.47 to 0.90)	300 (1 RCT)	⨁⨁⨁◯ Moderate^a	Important Outcome

* The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

CI: confidence interval; RR: risk ratio.

GRADE Working Group grades of evidence:

High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.

Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.

Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Explanations

^aThe QoE is downgraded by one level as the sample size is less than the Optimal Information Size (OIS).

Excluded studies

A total of 35 trials were excluded from this review. A total of 13 trials did not include participants of interest, and 17 studies had study designs other than RCTs, quasi-experiments, trials, pre-post studies, and other experimental study designs, and in five studies, outcomes not of interest were reported. We presented the list of excluded studies in Table 4.

Table 4. Characteristics of excluded studies.

SN	Study	Reason for exclusion
1.	(Akour, 2020)	The study design is not relevant
2.	(Giannoni et al., 2020)	Study design is not relevant
3.	(Angurana & Bansal, 2021)	Study design is not relevant
4.	(He et al., 2020)	Study design is not relevant
5.	(Gohil et al., 2021)	Study design is not relevant
6.	(Shahbazi et al., 2020)	Study design is not relevant
7.	(Paknahad & Moravejolahkami, 2021)	Study design is not relevant
8.	(Tateno et al., 2020)	Population not of interest
9.	(Peng et al., 2021)	Study design is not relevant
10.	(Louca et al., 2021)	Study design is not relevant
11.	(Gouda et al., 2021)	Study design is not relevant
12.	(Taghinezhad-S et al., 2021)	Study design is not relevant
13.	(Patra et al., 2021)	Study design is not relevant
14.	(Heidari et al., 2021)	Study design is not relevant
15.	(Kullar et al., 2021)	Study design is not relevant
16.	(Baindara et al., 2021)	Study design is not relevant
17.	(Darbandi et al., 2021)	Study design is not relevant
18.	(Li et al., 2021)	Study design is not relevant
19.	(Sevillano-Jiménez et al., 2022)	outcome not of interest
20.	(Blackett et al., 2022)	population not of interest
21.	(Nistor-Cseppento et al., 2022)	population not of interest
22.	(Forsgård et al., 2023)	population not of interest
23.	(De Boeck et al., 2022)	outcome not of interest
24.	(Hasslöf et al., 2022)	population not of interest
25.	(Slykerman & Li, 2022)	population not of interest
26.	(Vaezi et al., 2023)	outcome not of interest
27.	(Akbarzadeh et al., 2022)	population not of interest
28.	(Leal-Martínez et al., 2022)	outcome not of interest
29.	(Fernández-Ferreiro et al., 2022)	outcome not of interest
30.	(Slykerman et al., 2022)	population not of interest
31.	(Rayyan et al., 2023)	population not of interest
32.	(Wong et al., 2023)	population not of interest
33.	(Vázquez-Frias et al., 2023)	population not of interest
34.	(Nobile & Puoci, 2023)	population not of interest
35.	(Bangar et al., 2023)	population not of interest

Characteristics of ongoing studies

We came across 29 ongoing trials on nutraceuticals (probiotics, prebiotics, or synbiotics) for COVID-19. Recruitment status is completed in 9 trials, Recruiting in 12 trials, and not yet recruiting in 6 trials. Two trials are currently suspended. The trials are being conducted in Mexico (n = 1), Pakistan (n = 1), Canada (n = 3), Ukraine (n = 1), Japan (n = 1) India (n = 1), Austria (n = 1), Belgium (n = 1), China (n = 1), Italy (n = 1), New Zealand (n = 1), Iran (n = 3), Spain (n = 4), United States (n = 3), United Kingdom (n = 2), Indonesia (n = 1), Austria (n = 1), Hong Kong (n = 2). We documented the list of ongoing studies in Table 5: Characteristics of ongoing studies.

Discussion

We identified 481 potential studies from different electronic databases. Despite undertaking a comprehensive search strategy, only five trials with 860 participants were included in this review. Participants included both adult males and females. In three trials (Araimo et al., 2021; Ivashkin et al., 2023; Reino-Gelardo et al., 2023), probiotics were used as an adjuvant treatment to the standard care, and in other two trials (Gutiérrez-Castrellón et al., 2022; Rathi et al., 2021), probiotics (with or without prebiotics) were given as a supplement or as an adjuvant to the immunotherapy compared with placebo containing maltodextrin. One trial (Gutiérrez-Castrellón et al., 2022) was at ‘low risk’ of bias, whereas the other four trials (Araimo et al., 2021; Ivashkin et al., 2023; Reino-Gelardo et al., 2023) were at ‘some concerns’ for risk of bias. The total number of participants included in the five trials was 860. None of the included studies reported the number of cases of COVID-19 and changes in the disease severity. The outcome measures were mainly focused on the number of deaths (Araimo et al., 2021; Ivashkin et al., 2023), adverse events (Araimo et al., 2021; Gutiérrez-Castrellón et al., 2022), days of hospitalization (Ivashkin et al., 2023; Reino-Gelardo et al., 2023) and fatigue (Rathi et al., 2021). There is limited evidence available on using nutraceuticals (probiotics/prebiotics/synbiotics) for the treatment and prevention of COVID-19, as the observed effect in minimizing deaths and the duration of hospitalization in COVID-19 patients was uncertain. Therefore, there is presently no evidence to support the use of nutraceuticals (probiotics, prebiotics, or synbiotics) in the treatment of COVID-19.

Additionally, there was no evidence that using nutraceuticals would cause more negative side effects. There are currently 29 ongoing trials (Table 5) that can change the findings in the update of this review. Several uncertainties persist regarding using nutraceuticals for the prevention and treatment of COVID-19. These include the following: a small number of published studies with relatively few participants, no data on the efficacy of nutraceuticals in the prevention of COVID-19, and limited data on the number of deaths, adverse events, and days of hospitalization.

Table 5. Characteristics of ongoing studies.

SN	Recruitment status	Study Id	Date of Registration	Country
1.	Recruitment status completed	NCT04399252	May 22, 2020	Durham, North Carolina, United States
2.	Recruitment status completed	NCT04621071	November 9, 2020	Canada, Quebec
3.	Recruitment status completed	NCT04734886	February 2, 2021	Sweden
4.	Recruitment status completed	NCT04507867	August 11, 2020	Mexico
5.	Recruitment status completed	NCT04458519	July 7, 2020	Canada, Quebec
6.	Recruitment status completed	IRCT20161206031255N4	December 7, 2020	Iran
7.	Recruitment status completed	NCT05043376	September 14, 2021	Pakistan
8.	Recruitment status completed	IRCT20101020004976N6	2020-07-18	Iran
9.	Recruiting	NCT05080244	October 15, 2021	Canada, Quebec
10.	Recruiting	NCT04847349	April 19, 2021	United States, New Jersey
11.	No longer in EU/EEA	2020-001597-30	April 9, 2020	United Kingdom
12.	Recruiting	NCT04937556	June 24, 2021	Spain
13.	Not yet recruiting	NCT04907877	June 1, 2021	Ukraine
14.	Recruiting	JPRN-jRCTs021200020	August 25, 2020	Japan
15.	Not Yet Recruiting	CTRI/2021/03/031720	March 4, 2021	India
16.	Recruiting	NCT04813718	March 24, 2021	Austria
17.	Suspended	NCT04793997	March 11, 2021	Belgium
18.	Recruiting	ChiCTR2000029974	31 October 2020	China
19.	Recruiting	NCT04366089	April 28, 2020	Italy
20.	Not yet recruiting	ACTRN12620000480987	April 16, 2020	New Zealand
21.	Recruitment status completed	IRCT20200318046812N3	Jan 12, 2021	Iran
22.	Not yet recruiting	NCT04756466	February 16, 2021	Spain
23.	Recruiting	NCT04366180	April 28, 2020	Spain
24.	Recruiting	NCT04941703	June 28, 2021	United States, Tennessee
25.	Not yet recruiting	NCT04877704	May 7, 2021	United Kingdom
26.	Not yet recruiting	NCT04979065	July 27, 2021	Indonesia
27.	Recruiting	NCT04950803	July 6, 2021	Hong Kong
28.	Recruiting	NCT04420676	June 9, 2020	Austria
29.	Recruiting	NCT04884776	May 13, 2021	Hong Kong

Overall completeness and applicability of evidence

Some issues impede the completeness and applicability of the evidence in this review. The number of studies and participants included are limited. Studies included only adults. The findings of this review apply only to currently prescribed probiotic strains such as Lactobacillus, Bifidobacteria and Streptococcus species as a single strain or in probiotic mixtures and at different doses, frequencies, and durations. No studies were reported on non‐lactic acid bacteria. Prebiotics that were included in this review are fructooligosaccharides and maltodextrin (Gutiérrez-Castrellón et al., 2022; Reino-Gelardo et al., 2023). The lack of evidence regarding the efficacy of probiotics, prebiotics or synbiotics in preventing COVID-19 and changing the severity of COVID-19 (the primary outcomes for this review) is an issue with the completeness of the evidence.

There is insufficient evidence to confirm whether nutraceuticals (probiotics, prebiotics or synbiotics) can provide a therapeutic advantage in COVID-19 patients. Hence, it is hard to draw a definitive conclusion about their effects on the prevention and treatment of COVID-19.

Quality of the evidence

The quality of evidence (QoE) was ‘low’ for a number of deaths and adverse events, and ‘very low’ for Days of hospitalization for the comparison ‘Standard care + Nutraceuticals (Probiotics/Prebiotics/Synbiotics) compared to Standard care for Covid-19’. The QoE was downgraded by two for the risk of bias for some concerns in Domain 1 and Domain 2 of RoB2 Tool. The QoE was downgraded as the studies did not report the method of randomization and were open-labeled RCTs and also sample size was below the Optimal Information Size (OIS). The QoE was downgraded by one level due to the presence of heterogeneity (I²=73%) (Table 2). The quality of evidence (QoE) was ‘moderate’ for adverse events for the comparison of ‘Nutraceuticals (Probiotics/Prebiotics/Synbiotics) versus placebo for COVID-19’. The QoE was downgraded by one level as the sample size is less than the Optimal Information Size (OIS) (Table 3).

Potential biases in the review process

We conducted comprehensive searches to identify all the relevant studies. However, some trials may have been missed. We did not search the grey literature, thereby missing any conference proceedings/abstracts presented during the rapidly evolving pandemic situation. Three of the included trials were open-label, thus affecting the quality of the evidence. To avoid bias during the review process, two authors independently undertook screening for the selection of studies, extraction of data, and assessment of the risk of bias in the included studies. Additionally, disagreements amongst these reviewers were resolved initially through consensus and a third reviewer. None of the reviewers was involved in any of the included trials. As only five studies were analyzed in this review, it was impossible to detect the publication bias. Also, the power to identify the overall effect estimate pattern was inadequate, so a chance of coincidental findings cannot be omitted. There were no other potential biases.

Agreements and disagreements with other studies or reviews

No previous systematic review investigated the use of nutraceuticals (Probiotics/Prebiotics/Synbiotics) for COVID-19.

Conclusions

There is limited evidence available on using nutraceuticals (Probiotics/Prebiotics/Synbiotics) for the treatment and prevention of COVID-19, as the observed effect in minimizing deaths and the duration of hospitalization in COVID-19 patients was uncertain. In addition, there was no evidence of increased adverse effects with nutraceuticals use. Therefore, for the treatment of COVID-19, the use of nutraceuticals is uncertain, and it should not be considered conclusive for any implication for practice. Future studies should report/measure outcomes such as the number of cases of COVID-19 and the change in disease severity. Researchers should also consider standardizing doses/concentrations of nutraceuticals given. Methodologically robust randomized controlled trials needed to be undertaken in large samples of populations so that evaluating their therapeutic potential gives us quality evidence for their efficacy and safety in clinical practice.

Authors’ contributions

Concept and ideation: APS; Protocol development: APS and MNK; Literature search: AG and SZQ; Screening and selection of studies: SU and MW; Data extraction: SU and MW; Assessment of risk of bias of included studies: APS and MNK; Meta-analysis: AG; Assessment of Quality of Evidence: SZQ; Overall preparation of the manuscript: APS, MNK and SU with inputs from all the other reviewers.

Differences between protocol and review

In the protocol, we had planned to assess the risk of bias for RCTs using Cochrane’s Risk of Bias Tool. However, we used the RoB 2 tool, the revised tool for assessing the risk of bias in randomized trials (Sterne et al., 2019). We had not kept ‘Comparison 3: Nutraceuticals (Probiotics or prebiotics or synbiotics) as an adjunct to standard treatments verses placebo’ as a Comparison in the protocol. However, one trial (Rathi et al., 2021) reported a ‘Free of fatigue’ outcome for this comparison group. We felt that readers might be interested in understanding the efficacy of probiotics in reducing fatigue. Hence, we have added ‘Free of fatigue’’ as a secondary outcome under this comparison group.

Acknowledgement

Dr. Rekha Sinha, International Life Sciences of India (ILSI).

Disclosure statement

The authors report there are no competing interests to declare.

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.

Additional information

Funding

This SR was funded by the International Life Sciences Institute (ILSI), India under Grant 5/7/ILSI-India/2021-RBMCH, Special Grant Project No.252.

Appendices

Appendix 1

Search strategy for PubMed

The search strategy for PubMed: ((‘probiotic*’[Title/Abstract] OR ‘pro biotic*’[Title/Abstract] OR ‘prebiotic*’[Title/Abstract] OR ‘pre biotic*’[Title/Abstract] OR ‘synbiotic*’[Title/Abstract] OR ‘lactobacill*’[Title/Abstract] OR ‘lacto bacill*’[Title/Abstract] OR ‘bifidobacteri*’[Title/Abstract] OR ‘bifidus*’[Title/Abstract] OR ‘streptococcus thermophil*’[Title/Abstract] OR ‘saccharomyce*’[Title/Abstract] OR ‘probiotics’[MeSH Terms] OR ‘lactobacillus’[MeSH Terms] OR ‘bifidobacterium’[MeSH Terms] OR ‘streptococcus thermophilus’[MeSH Terms] OR ‘saccharomyces’[MeSH Terms]) AND (‘covid*’[Title/Abstract] OR ‘corona-virus’[Title/Abstract] OR ‘corona-virus’[Title/Abstract] OR ‘cov’[Title/Abstract] OR ‘ncov*’[Title/Abstract] OR ‘n cov*’[Title/Abstract] OR ‘sars*’[Title/Abstract] OR ‘covid*’[Title/Abstract] OR ‘covid 19’[MeSH Terms])) AND (2019:2023[pdat])

Search Strategy for CENTRAL

CENTRAL via Cochrane Library, issue 9, 2023

ID Search

#1 covid*

#2 corona*

#3 n-cov*

#4 ncov*

#5 sars*

#6 MeSH descriptor: [COVID-19] explode all trees

#7 #1 OR #2 OR #3 OR #4 OR #5 OR #6

#8 probiotic*

#9 ‘pro biotic*’

#10 pro-biotic*

#11 prebiotic*

#12 ‘pre biotic*’

#13 pre-biotic*

#14 synbiotic*

#15 lactobacill*

#16 bifidobacteri*

#17 bifidus*

#18 ‘streptococcus thermophil*’

#19 saccharomyce*

#20 MeSH descriptor: [Probiotics] explode all trees

#21 MeSH descriptor: [Prebiotics] explode all trees

#22 MeSH descriptor: [Synbiotics] explode all trees

#23 MeSH descriptor: [Lactobacillus] explode all trees

#24 MeSH descriptor: [Bifidobacterium] explode all trees

#25 MeSH descriptor: [Streptococcus thermophilus] explode all trees

#26 MeSH descriptor: [Saccharomyces] explode all trees

#27 #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20 OR #21 OR #22 OR #23 OR #24 OR #25 OR #26

#28 #7 AND #27

Search Strategy for WHO Covid-19 Databasetw:(probiotic* OR prebiotic* OR synbiotic* OR lactobacill* OR bifidobacteri* OR bifidus* OR ‘streptococcus thermophil*’ OR saccharomyce*) AND type_of_study:(‘clinical_trials’) AND la:(‘en’)

Appendix 2

Summary of results of dichotomous and continuous outcomes reported in individual studies

A2 Table 1: Summary of results of dichotomous outcomes reported in individual studies.

	Study ID	No of events in IG	Total No. of participants in IG	No of events in CG	Total No. of participants in CG	Follow up period
Comparison 1: Nutraceuticals (probiotics or prebiotics or synbiotics) as a adjunct to standard treatments verses Standard of care without supplementation of probiotics or prebiotics or symbiotics
Number of cases of COVID-19	–	–	–	–	–	–
Change in disease severity	–	–	–	–	–	–
Number of deaths	(Araimo et al., 2021)	0	14	0	14	7 to 14 days
		01	14	01	14	30 days
	(Ivashkin et al., 2023)	4	99	4	101	Lasted from the time of the inclusion until recovery or death
Adverse events	(Araimo et al., 2021)	No	14	Diarrhoea: 4 (30%)	14	21 days
Comparison 3: Nutraceuticals (probiotics or prebiotics or synbiotics) as an adjunct to standard treatments verses placebo
Number of cases of COVID-19	–	–	–	–	–	–
Change in disease severity	–	–	–	–	–	–
Adverse events	Gutiérrez-Castrellón et al., 2022 (Gutiérrez-Castrellón et al., 2022)	41	150	63	150	30 days p = 0.008
Free of fatigue	(Rathi et al., 2021)	Day 4: Day 11:	Day 4: Day 11:	Day 4: Day 11:	Day 4: Day 11:

A2 Table 2: Summary of results of continuous outcomes reported in individual studies.

Comparison 1: Nutraceuticals (probiotics or prebiotics or symbiotics) as a adjunct to standard treatments) verses Standard of care without supplementation of probiotics or prebiotics or symbiotics
	Study Id	Mean of IG	Total no. of participants in IG	Mean of CG	Total no. of participants in CG	Duration of follow up
Days of hospitalization	(Ivashkin et al., 2023)	11 (10–14) days	99	11 (9–14) days	101	Lasted from the time of the inclusion until recovery or death
	(Reino-Gelardo et al., 2023)	8.7 ± 3.9 days	70	11.6 ± 7.4 days	69		p = 0.007