3.1. Results of the Selection Studies
3.2. Result of the data extraction
3.3 Result of the risk bias assessment
The assessment of quality and risk of bias in the included studies [22, 23, 24, 25, 26] is based on seven key areas: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other biases. This assessment relies on established criteria outlined in the "Cochrane Handbook for Systematic Reviews of Interventions" to evaluate the likelihood of bias in each domain. The nature of risk is categorized as "High risk," "Low risk," or "Unclear risk" for each study, considering its methodological quality and potential biases.
Table 1
Data extracted for inclusion in the meta-analysis.
Study (Author, year) | Country | Study design | Participants | Intervention | Duration | Outcomes |
Li, Q., et al 2021[22] | China | Single-centre retrospective study | 311 adult patients (≥ 18 years). 123 patients were treated with probiotics, 67 males and 56 females. The remaining 188 patients, 83 males and 105 females, received other drugs, for example, Chloroquine Phosphate. | 4g of oral probiotics contain strains of: − 1.5 grams ofBifidobacterium, Lactobacillus, Enterococcus, Bacillus tablet, and Bifidobacterium infantis. − 2 grams of Bifidobacterium and Lactobacillus tablet (Bifidobacterium longum, Lactobacillus bulgaricus, Streptococcus thermophiles). - And finally 0.5 grams of Bacillus subtilis and Enterococcus Faecium soluble capsules (Enterococcus faecium, Bacillus subtilis). | 26 Days | 08 selected laboratory parameters: (IL-6, CRP, LDH, total T cells, NK cells, B cells, CD4 + T cells, CD8 + T cells, and CD4/CD8 ratio). |
Ceccarelli, G., et al 2021[23] | Italy | Observational and retrospective cohort study | The study involved 200 adult patients (≥ 18 years), consisting of 113 males and 87 females. − 112 patients received the BAT protocol without oral bacteriotherapy (probiotic). − 88 patients received the BAT protocol with oral bacteriotherapy. | Treatment by oral administration of SivoMixx formulations includes the probiotics : Lactobacillus brevis DSM 27961, Streptococcus thermophilus DSM 32245, Bifidobacterium lactis DSM 32246, Bifidobacterium lactis DSM 32247, Lactobacillus acidophilus DSM 32241, Lactobacillus helveticus DSM 32242, Lactobacillus paracasei DSM 32243, Lactobacillus plantarum DSM 32244, The formulation wasadministered in three equal doses per day, for a total of 24× 1011 CFU per day. | 51 Days | (CRP, LDH, T-cells, Albumin, Monocytes, BMI, PSI, Lymphocytes, White blood cells, Neutrophils, Duration of hospital stay (days), Deaths). |
Gobbi, M., et al 2021[24] | Italy | Observational study | 48 patients (26 males/22 females), 29 patients were treated with a nutritional intervention containing probiotics. | An oral intervention of fat-free milk powder contains vitamins, amino acids, minerals, and probiotics: Streptococcus thermophilus, Bifidobacterium breve, Bifidobacterium longum, Bifidobacteriuminfantis, Lactobacillus acidophilus, Lactobacillus plantarum, Lactobacillus paracasei, and Lactobacillus delbrueckii subsp. bulgaricus. A minimum of 1 to 1.3 g/day/kg of body weight. An optimal fat-to-carbohydrate energy ratio was 30:70, 50:50. | 25 Days | This study included laboratory parameters measured from day 1 to day 7 of administration: (IL-6, CRP, T-cells, albumin, ferritin, BMI, creatinine, serum iron), muscle strength (HG), performance test (TUG). |
Araimo, F. et al 2021[25] | Italy | An interventional, non-pharmacological, open-label, randomized, prospective, double-blind study | 85 patients (≥ 18 years) confirmed with COVID-19; only 28 patients were admitted to the study. - The first experimental group of 14 patients, known as the ozone group, was treated with the BAT protocol + O3-AHT therapy through the administration of oxygen-ozone, followed by a duration of a probiotic mixture. - The second group of14 patients was treated with the therapeutic protocol only. | Interventiontreatment of the administered SivoMixx product sachets contains probiotics: Streptococcus thermophilus DSM 32345, L. acidophilus DSM 32241, L.helveticus DSM 32242, L. paracasei DSM 32243, L. plantarum DSM 32244, L. brevis DSM 27961, B. lactis DSM 32246, and B. lactis DSM 32247. - Administration was only to the 14 patients, with one sachet every 12 hours for 7 days. - Each sachet of SivoMixx in this study contained a high concentration of 8×109 CFU, administered every 12 hours. | 07 Days | This study included laboratory parameters measured from day 1 to day 7 of administration: (IL-6, CRP, T-cells, albumin, ferritin, BMI, creatinine, serum iron), muscle strength (HG), performance test (TUG). |
d'Ettorre, G., et al 2020[26] | Italy | A retrospective cohort study | 70 COVID-19-positive patients - The first group of 48 patients received a BAT treatment protocol often containing hydroxychloroquine, antibiotics, and tocilizumab. - The second group of 28 patients received oral bacteriotherapy (probiotic formulation). | The SivoMixx formulation administered in this study contained: Streptococcus thermophilus DSM 32345, L. acidophilus DSM 32241, L. helveticus DSM 32242, L. paracasei DSM 32243, L. plantarum DSM 32244, L. brevis DSM 27961, B. lactis DSM 32246, and B. lactis DSM 32247. - This oral bacteriotherapy involved the use of 24×1011 CFU per day. - The formulation was administered in three equal doses per day. | 17 Days | This study includes respiratory parameters: (FiO2, SO2, pH, HCO3, PO2, PCO2), and BMI. |
Table 1: Data extracted for inclusion in the meta-analysis.
3.4 The rationale for the application of probiotics as a potential prevention and alternative treatment strategy against COVID-19
The meta-analysis was conducted to analyze existing studies in the literature. Our focus was on probiotic treatment for respiratory infections. Q-statistics were employed to evaluate whether effect sizes across individual studies were homogeneous, indicating the degree of heterogeneity. Additionally, heterogeneity of individual outcome estimates in the meta-analysis was assessed using forest plots. The efficacy of probiotics varies based on specific strains, age groups, clinical dosages, and mode of administration. Clinical studies, including randomized, double-blind, intervention-controlled, or experimental human trials of probiotics, were planned. The overall effects of probiotics were quantified using values such as standardized mean differences (SMD) and confidence intervals (CI) [27].
3.5 Selection of parameters or outcomes in the analysis
The five included studies comprise human clinical trials involving COVID-19 patients experiencing respiratory difficulties due to inflammation. A total of 657 patients were divided into two groups, each following specific therapeutic protocols that included probiotics of various strains (such as Bifidobacterium, Lactobacillus, Streptococcus) sourced from products like SivoMixx, capsules, and milk powder. Following the treatment period in each trial, post-treatment monitoring of patients' immune and health status was conducted. This assessment involved various parameters, including blood tests and respiratory status evaluations, administered by specialists across all studies. Among these parameters, CRP, BMI, T Cells, Albumin, IL-6, LDH, and Ferritin were identified as common outcomes across all five studies for subsequent analysis.
Probiotics and C-Reactive Protein (CRP)
The CRP level serves as an indicator of inflammation, commonly associated with respiratory infections caused by COVID-19 [28]. It is a protein of the acute phase of inflammation, whose levels rise in serum or plasma during responses to infections or early inflammatory events [29]. In regard to this parameter, a meta-analysis of four studies based on five RCTs [22, 23, 24, 25] involving 587 participants (254 in the experimental group and 333 in the control group) revealed an overall significant effect (p = 0.002) with a standardized mean difference (SMD) of 0.26 and a confidence interval (CI) of 0.10 to 0.43. The analysis showed a statistically significant increase in CRP levels (p = 0.002 < 0.05), indicating significant heterogeneity.
Probiotics and Body Mass Index (BMI)
The Body Mass Index (BMI) is a widely used measure for assessing individuals' corpulence and tracking changes in body composition over time. It is calculated based on height and body mass, serving as an indicator of overweight and obesity according to WHO guidelines, with the formula BMI = P/T2 or BMI = m×h− 2 [30].
In the analysis of BMI across four studies from five RCTs [23, 24, 25, 26], involving 346 patients (159 in the test group and 187 in the control group), there was an overall indicative increase in BMI levels (SMD 0.28; CI 0.07 to 0.50; p = 0.01). This indicates a significant increase (p = 0.01 < 0.05) in BMI. Notably, the included studies exhibited low heterogeneity, which was not statistically significant (I2 = 0%; p = 0.40 > 0.05).
Probiotics and T-cells
T cell count serves as an indicator of the adaptive immune response, offering insight into the cellular immunity status, particularly in coronavirus-affected lung cells [31]. T cells, including subsets such as CD4 + T cells, CD8 + T cells, B cells, and natural killer (NK) cells, are crucial for maintaining immune system function post-viral infection [32].
In the context of T-cell analysis across four studies from five RCTs [22, 23, 24, 25], involving 587 participants (254 in the experimental group and 333 in the control group) included in this meta-analysis, each individual received probiotic intake. The overall effect revealed a non-significant increase (p = 0.26 > 0.05) with a standardized mean difference (SMD) of 0.09 and a confidence interval (CI) of -0.07 to 0.26 (p = 0.26). Notably, the included studies exhibited complete lack of heterogeneity, indicating a low and statistically non-significant level (I2 = 0%; p = 0.73).
Probiotics and Albumin
Albumin levels serve as an indicator of enzyme, drug, and hormone transport, as well as blood circulation stabilization [33]. As an acute phase reactant with antioxidant properties, plasma albumin is crucial in scavenging reactive oxygen species (ROS) under normal physiological conditions [34]. In cases of oxidation, ROS accumulation in neutrophils triggers extracellular neutrophil traps, which can accumulate in the lungs, as seen in COVID-19 cases [34].
Regarding albumin levels, analysis of three studies from five RCTs [23, 24, 25], involving a total of 276 participants (131 in the experimental group and 145 in the control group), showed an increase in albumin levels following probiotic supplementation. The overall effect of this increase was significant (p = 0.02), with a standardized mean difference (SMD) of 0.28 and a confidence interval (CI) of 0.04 to 0.52, as the p-value (p = 0.02 < 0.34) indicated significance
Probiotics and IL-6
Only two out of the five RCTs [22, 25], involving 339 patients (137 in the first group and 202 in the second group), demonstrated a significant increase in interleukin-6 levels following the administration of probiotic formulas. Interleukin-6 is produced by the immune system as part of the defensive response against SARS-CoV2, involving the intervention of various white blood cells, including macrophages, T cells, and B cells [35].
The meta-analysis revealed a pooled effect of a significant increase (p = 0.00001) (SMD 0.67; CI, 0.45 to 0.90), indicating statistical significance (p = 0.00001 < 0.05). Furthermore, a high level of heterogeneity (I2 = 94%) was observed, which is considered significant, given the p-value (p = 0.0001) being less than 0.05.
Probiotics and LDH
LDH levels serve as a non-specific indicator of cell death in various respiratory diseases, including COVID-19 [36]. Lactate dehydrogenase (LDH) is an enzyme crucial for the conversion of lactate to pyruvate in most body tissues, and its levels typically rise following tissue breakdown, reflecting elevated serum LDH in numerous clinical conditions [37].
In this context, only two out of five RCTs [22, 23] involving 511 confirmed COVID-19 participants (211 in one group and 300 in another) included in the meta-analysis demonstrated a decrease in LDH levels among patients receiving probiotics. However, the overall association between probiotic intervention and LDH levels was non-significant, with a p-value of 0.17, indicating no statistical significance (SMD 0.12; CI -0.05 to 0.30; p = 0.17). Additionally, the included studies exhibited moderate heterogeneity (I2 = 55%; p = 0.13), which was not statistically significant (p = 0.13 < 0.05).
Probiotics and Ferritin
Ferritin serves as a vital protein for iron storage, facilitating iron availability for crucial cellular processes while safeguarding DNA, lipids, and proteins from the potential toxic effects of iron. Monitoring ferritin levels through blood sampling is essential, given its involvement in various diseases, including inflammatory, neurodegenerative, and malignant conditions [38].
In this analysis, only two out of five RCTs [24, 25] involving a total of 76 participants contributed to the meta-analysis, revealing an increase in ferritin levels following probiotic intake. However, the overall association between probiotic usage and ferritin levels did not demonstrate significance, with a p-value of 0.41 (SMD 0.19; CI -0.27 to 0.66; p = 0.41 < 0.05). Notably, the included studies exhibited a higher level of heterogeneity, which was statistically significant (I² = 77%; p = 0.04).
3.6. Result of the publication bias assessment
The funnel plot, depicted in Fig. 3.11, illustrates minimal indications of publication bias across the five selected studies. The plot's funnel traces graphically represent the curves constructed for the analyzed parameters: CRP, BMI, T-cells, Albumin, IL-6, LDH, and ferritin, facilitating the assessment of bias in each included study. Of particular interest are the p-values obtained from the Egger test, which were 0.0001, 0.04, 0.13, 0.34, 0.40, 0.40, and 0.73 for the respective parameters. Notably, most of these values were statistically insignificant, as they exceeded the threshold of 0.05, except for IL-6 (p = 0.0001), indicating a small study effect. Consequently, only one parameter, IL-6, exhibited bias, while the remaining six parameters showed no signs of publication bias.