In this multicenter randomized dietary trial of the outpatient adults with COVID-19, we demonstrated that the patients in the probiotic group had lower global, respiratory and constitutional symptom scores on the day 10th. Selection of the 10th day time-point for the outcome measures was made upon the reports that it was critical to the further fate of the patient, improvement or deterioration. The median time from onset of illness to acute respiratory distress was 8–12 days, and admission to the intensive care unit was 9.5–12 days allowing a period of approximately 10 days that the participant theoretically could be observed as an outpatient [21,22,23,24]. However, this time point was based on clinical observation of the more aggressive type of SARS-CoV-2 (delta) that was prevalent at the beginning of the pandemics when the study protocol was constructed [25]. According to the local health department updates, the most likely variant of coronavirus circulating in the communities during the study was Omicron B.1.1, with milder manifestation of COVID-19 [26, 27]. Indeed, none of the patients experienced emergencies resulted from respiratory failures (pulse oximetry values varied within normal ranges, data not shown) or in hospitalizations due to other complications. Daily global symptom scores varied within mild disease limits, and progressively decreased for the first 2 weeks with better resolution in the probiotic group.
Along with global symptom score, it`s respiratory and constitutional subsets were lower in the probiotic group on the 10th day of the disease. On the same day, due to their shorter duration, gastrointestinal symptoms were almost absent with no statistical significance between the groups. However, analysis of proportion of patients with gastrointestinal symptoms for the first 10 days was variably lower in the probiotic group, while duration of symptoms only tended to be significant. In the study of hospitalized patients with severe COVID-19 the duration of diarrheal episodes was -2.41 days shorter in patients receiving B. longum, L. bulgaricus and S. thermophilus in a dose of > 6 × 107 CFU a day for 7 days [28]. Along with this finding, the prospective open-label trial evaluating the impact of one-month oral intake of Kluyveromyces marxianus B0399 (1 × 109 CFU/day) and L. rhamnosus CECT 30579 (1 × 108 CFU/day) showed a benefit in decreasing number of patients without abdominal pain as assessed on the 30th day of observation [29]. The mentioned studies varied in bacterial strains used, type and timing of endpoints for gastrointestinal symptoms, that together with statistically insufficient proportion of patients challenges a direct comparison of the results of our study.
Use of the eight-point WHO’s ordinal severity scale showed greater proportion of patients with no limitation (score 1) and reciprocally fewer numbers with limitation (score 2) of activity in the probiotic group. Reasonably that this finding may be explained by milder disease in the probiotic group according with data retrieved from the Respiratory Illness Diary.
At time of construction of this article, reports on the use of probiotics in clinical trials in COVID -19 patients were sparse. We have located a large randomized clinical study with similar design, prospectively observing clinical course of COVID-19 using a self-reported electronic diary as the data collection form [30]. As with our findings, the authors reported a significant shortening of most COVID-19 symptoms in ambulatory patients taking a mixture L. plantarum KABP022, KABP023, KAPB033 and Pediococcus acidilactici KABP021 strains in a dose of 2 billion CFU daily for the 30-days starting within first 7 days of the disease. Depending on the particular symptom, use of probiotics was associated with faster improvement in 2.5 through 7.0 days [16]. In our study, the median difference in time to recovery was 3 days for the global symptom score, which falls within the range of the cited report. In the retrospective study of the hospitalized COVID-19 patients, dietary supplementation with Bifidobacterium, Lactobacillus and Enterococcus 1.0 × 107 CFU (strains not specified) each ingredient 3 times daily for 30 days resulted in 3 days faster clinical improvement than in patients without supplementation [31]. Using the mean baseline-follow-up difference of total symptoms per the probiotic and control group on the day 30th of observation, it was shown significant change in proportion of asymptomatic patients favoring the use of probiotics [16].
Other researchers found no probiotic effect in COVID-19 manifestation when evaluating the throat spray containing L. casei AMBR2, L. plantarum WCFS1, and L. rhamnosus GG. The spray was administered in multiple doses, two puffs each, containing approximately 9.5 × 108 CFU, and started within the first 4 days of the confirmed SARS-CoV-2 infection and continued for 14 days [32]. The lack of effect was attributed to the small sample size, the highly variable course of disease in a given patient, and the substantial variability of symptoms among patients. The same study showed a trend toward faster clearance of the virus, with more than three times fewer positive cases in the probiotic group after 3 weeks [19]. Further analysis of microbiome amplicon sequence variants of lactobacilli, indicated significant negative association between the strains and the acute symptom score suggesting that the application of these lactobacilli could result in less acute symptoms in the verum exposure group [19]. An earlier real-life hospital-based observation showed faster recovery from fever, asthenia, headache, myalgia, dyspnea, and an 8-fold lower risk of progression to respiratory failure in patients receiving oral probiotics compared with controls [33]. Intranasal irrigation with L. lactis W136 twice daily for 14 days in a small sample of 23 patients aged 18–59 years without concomitant diseases was associated with significantly less fatigue, olfactory dysfunction and dyspnea [34]. The use of B. longum ES1, B. lactis CBP-001010, and L. rhamnosus CNCM I-4036 (> 109 CFU per stick) enriched with zinc, selenium, and vitamin D and administered once daily to hospitalized patients with COVID-19 led to a shorter duration of digestive symptoms and a shorter hospital stay in the cohort of patients with milder pulmonary involvement as evidenced by the chest X-ray [35]. Decrease in calprotectin and C-reactive protein levels, IL-6 in hospitalized patients supplemented with probiotics corroborated their beneficial role in COVID-19 [36, 37]. However, the results of studies exploring upper airway probiotic application or the results of hospital-based studies may not be directly compared with our ones due to the different outcomes measures (recovery vs. improvement, active inflammation markers), severity of the disease (outpatients vs. inpatients), formulations (unprotected probiotics in liquid vs. powder in intestine-soluble capsules), routes of administration (local vs. oral), and areas of action sites (nose/throat vs. small/large intestines).
There is growing number of cell culture and animal studies supporting clinical effects of probiotics in respiratory tract infections, which evolved into a separate concept of the gut-lung axis [14]. Within the concept, gut commensals can distantly signal lungs and influence respiratory system through circulatory transportation of soluble microbial components (peptidoglycans, PG; lipopolysaccharide, LPS) and/or metabolites (short-chain fatty acids, SCFA). In mice model of respiratory syncytial virus infection, PG1505 purified from lactobacteria enhanced innate respiratory antiviral immune response and increased antiviral resistance via activation of Toll-like receptor-3 (TLR-3) [15]. Rectal introduction of bacterial LPS and PG, the TLR agonists, rapidly restored both antibody and T-cell responses to influenza infection in lungs of antibiotically treated mice, further supporting the idea that colonic bacterial products can distantly prime the lungs [38]. Gut commensals synthesizing butyrate (SCFA) from dietary fibers may have an additional potential for sharpening of immunity in respiratory infection. Feces of influenza-infected mice on the high fiber diet, contained almost 140-fold greater concentration of butyrate than controls, which was accompanied with accumulation of alternatively activated macrophages in the lungs, enhanced hematopoiesis of pulmonary anti-inflammatory macrophage precursors (Ly6c– monocytes) in the bone marrow, decreased pulmonary concentration of myeloperoxidase, improvement of symptomatic and survival rates [39]. Furthermore, in mice experimental models lactobacilli elicited systemic effect such as antigen presenting cell migration, enhanced TNF-α and interferon production in airways in response to influenza infection [22, 40,41,42] Clinical trials evaluating the role of probiotics in respiratory viral infections have demonstrated a probiotic-dependent increase in serum concentration of interferon-γ, NK cell activity, Th1 cell activation, and an increased number of T-helper and T-killer cells [43,44,45].
In our study, we found that the probiotics increased serum concentrations of anti-RBD/S1 IgG by more than twofold when compared to the placebo. It may account for a faster clinical resolution of COVID-19 symptoms as these antibodies possess neutralizing activity and prevent re-entry of SARS-CoV-2 into host cells upon receptor binding and membrane fusion [46]. Change in median concentration of antibodies against NCP, an important protein which participate in RNA package and virus particle release, showed only a trend toward difference between the study groups probably due to small number of cases tested (22 vs. 22 subjects) and possibly difference of molecular weights of RBD/S1 and NCP antigens. It is academically accepted that larger macromolecules represent better immunogens, as they are more easily processed by macrophages for presentation to lymphocytes, and therefore, can evoke more potent immune responses [47]. Sum of molecular weight of RBD [48] and Spike antigens [49] is higher than that in NCP [50] making theoretically stronger humoral response to RBD/S1complex. Under additional stimulation of the adaptive immune system by probiotics [51], the difference in production of antibodies to antigens with different molecular weights can be even more noticeable and can persist up to the 7th month after onset of the disease[52], suggesting that clinical benefits of probiotics can extend beyond the acute COVID-19 period and influence post-COVID-19 symptoms.
Probiotics decreased duration of post-COVID-19 fatigue and anxiety but did not change their incidence. In the probiotic group, the reduction in fatigability can be linked to the reduction in anxiety, a condition leaving the mind in a constant state of internal tension, mental and physical exhaustion. The importance of anxiety in the post-COVID-19 syndrome is difficult to overestimate due to its high prevalence and even its increasing proportion after acute phase of the disease [53]. After analyzing PCFS data, we also found that probiotics improved patients’ general functional ability. Among others, the PCFS scale contains an “anxiety” variable, that probably played a role in reducing the PCFS score. A meta-analysis of clinical trials on the anxiolytic effects of probiotics revealed a minor but substantial improvement when compared to controlled therapies [54]. Mechanisms behind the effect may be explained within the microbiota-gut-brain axis concept, where probiotics can interact with brain via modification of hypothalamic–pituitary–adrenal pathways [55], synthesis of neurotransmitters (aminobutyric acid, serotonin, dopamine, noradrenaline, melatonin, histamine and acetylcholine) [56, 57], interaction with the nervus vagus [58]. However, the exact anxiolytic role of probiotics in our study remains a field for further exploration.
The strength of our study was its multicenter, randomized, double-blind, placebo-controlled, prospective, parallel group design. Regular phone calls allowed a good patient retention, on-going discussion of the patient self-reported records and thereby obtaining better quality of the source data. The assessment of antibody production revealed one of the potential pathways of probiotic therapeutic effects. However, the failure to evaluate fecal specimens for microbial ecology precluded analysis of the role of the microbiome in the observed effects of probiotics. Neither at baseline nor during the study, the participant had controlled diets that might influence the gut microbiome and influence the outcomes. However, all patients were advised to avoid foods with labelled content of pre- or probiotics, major potential confounders. Another limitation included the use of an invalidated RID and the PCQ-19. However, the endpoints in the mentioned data collection forms were simple, directly related to the disease, and a wording easily comprehended by study participants.
In summary, the results of our study indicate that a short-term use of probiotics plays a role in attenuation of symptoms caused by the SARS-CoV-2 infection, stimulate virus-neutralizing humoral responses, reduce duration of post-COVID-19 anxiety and fatigue. In future studies, we plan to assess a role of the present probiotic strains in prevention of viral respiratory disease as well as their adjuvant properties in anti-viral vaccines.
