key: cord-0848117-pa32qhlf authors: Antonelli, Michele; Donelli, Davide; Firenzuoli, Fabio title: Ginseng integrative supplementation for seasonal acute upper respiratory infections: a systematic review and meta-analysis date: 2020-06-20 journal: Complement Ther Med DOI: 10.1016/j.ctim.2020.102457 sha: fdd1b2953144b8c99c33ec1f32f87b818ae2992a doc_id: 848117 cord_uid: pa32qhlf BACKGROUND: The aim of the review was to assess whether ginseng can be a useful supplementation for seasonal acute upper respiratory infections (SAURIs). METHODS: All clinical studies investigating ginseng efficacy for the treatment or prevention of SAURIs were included in the review. Medline, EMBASE, Web of Science, Scopus, Cochrane Library, Google Scholar were systematically screened for relevant articles up to May 26th, 2020. The risk of bias was assessed with the Cochrane tool (RoB 2). RESULTS: Nine articles (describing ten trials about P. ginseng or P. quinquefolius) were included in the review. Evidence globally indicated some useful activity of intervention when administered in adjunct to influenza vaccination. The results of our quantitative synthesis suggested a significant effect on SAURIs incidence (RR = 0.69 [95% C.I. 0.52 to 0.90], p < 0.05), as well as a significant reduction of their duration if only studies with healthy individuals were included in the analysis (MD=-3.11 [95% C.I. -5.81 to -0.40], p < 0.05). However, the risk of bias was high-to-unclear for most included trials, and publication bias couldn't be excluded. DISCUSSION: Limitations of existing evidence don’t allow to draw conclusions on the topic. Nevertheless, it is not excluded that ginseng supplementation in adjunct to influenza vaccination and standard care might be useful for SAURIs prevention and management in healthy adult subjects, but further high-quality trials are needed to support this hypothesis. OTHER: This research was not funded. The protocol was registered in PROSPERO under the following code: CRD42020156235. Seasonal acute upper respiratory infections (SAURIs) refer to infectious conditions involving the upper respiratory tract which mostly occur during cold months of the year, especially in winter 1, 2 . Common symptoms of SAURIs often include cough, sore throat, runny nose, nasal congestion, sneezing, headache, fever, malaise, and myalgias 2 . The etiology of SAURIs is mostly viral, with bacteria approximately accounting for only 15% of all cases 1,2 . In particular, over 200 different viruses can cause acute upper respiratory infections, and such viruses generally belong to one of the following six microbial families: orthomyxoviruses (influenza), paramyxoviruses (respiratory syncytial virus), parainfluenza viruses, coronaviruses, picornaviruses (common cold), herpes viruses, and adenoviruses 1 . From an epidemiological point of view, the most relevant ones are picornaviruses like rhinoviruses, often responsible for the common cold, and flu viruses, which can cause influenza 2 . Despite several similarities, these two diseases show slightly different epidemic trends: influenza exhibits the typical seasonal incidence during wintertime, whereas the common cold can potentially occur all the year long but its incidence only peaks in cold months of the year 3 . Although usually self-limiting, SAURIs can be sometimes followed by severe respiratory, cardiovascular or general complications J o u r n a l P r e -p r o o f with poor clinical outcomes especially in elderly subjects, fragile individuals or patients with important comorbidities 4 . According to the Centers for Disease Control and Prevention (CDC), over the last influenza season in the United States (2018-2019), it was estimated that flu-related hospitalizations were around 810.000, with 61 .000 flu-associated deaths 5 . From a socio-economic perspective, the average annual total economic burden of influenza to the US healthcare system and society was calculated to be around $11.2 billion 6 , and the total impact of non-influenzarelated viral respiratory tract infections was estimated to be approximately $40 billion annually 7 , with the common cold being responsible for 40% of all time lost from working activities 3 . Therefore, taken together, all these data highlight that SAURIs have a relevant impact not only on individual wellbeing and quality of life, but also on public health, community resources and productivity. To date, the common treatment of SAURIs is mostly symptomatic (with a limited array of antiviral medications to be used in selected conditions) and, beyond basic hygienic rules employed for the prevention of airborne infections, prophylactic strategies involving a specific vaccination are only available for a few microorganisms like influenza viruses 8, 9 . Among all medicinal herbs, ginseng is a complementary herbal remedy with a long-standing tradition, especially in some Asian and American countries, where it is still used for the treatment and prevention of various diseases, including SAURIs [10] [11] [12] [13] . Surveys have also shown that physicians, pharmacists, and patients have a favorable general attitude towards the integrative clinical use of medicinal plants, but a lack of knowledge regarding actual properties and limits of such interventions has been reported even among health professionals, thus underscoring the necessity to study the topic in depth and to J o u r n a l P r e -p r o o f better disseminate relevant information for the promotion of evidence-based practices [14] [15] [16] . The main ginseng species used in clinical practice are Korean ginseng (Panax ginseng), American ginseng (Panax quinquefolius), and Chinese ginseng (Panax notoginseng) 17, 18 . Some of the main constituents of these three ginseng species are polysaccharides and saponins like dammarane ginsenosides, which are transformed by the drug preparation procedure (steaming and heating) into ginsenosides Rg3, Rg5, Rk1, later converted by the intestinal microflora into bioactive substances such as compound K, ginsenoside Rh1, and protopanaxatriol (PPT) 18, 19 . Although similar, the chemical composition of P. ginseng, P. notoginseng, and P. quinquefolius shows some differences with regard to the type and quantity of specific ginsenosides 18 . In particular, the ratio between ginsenosides Rg1/Rb1 varies among the three species, and the majority of ginsenosides found in Korean ginseng are Rb1, Rg1, Rb2, whereas in American ginseng they are Rb1, Re, Rd, and in Chinese ginseng they are Rb1, Rg1, Ra3, and R1 18 . Differences have been also observed in the volatile composition of the three species, mainly characterized by the presence of various sesquiterpenes 20 . Additionally, regardless of biochemical differences due to specific botanical origins, both the preparation method 21 and the individual characteristics of enteric microflora 18 can influence the type and quantity of bioactive compounds that are absorbed through the intestine. In order to minimize such variability, controlled preparation methods have been developed, including fermentation with enzymes or microorganisms for the reduction of the impact of metabolization by the intestinal microflora 18 , and standardized extracts with a minimum amount of ginsenosides have been proposed for clinical uses 17 . J o u r n a l P r e -p r o o f The aim of the review was to assess whether ginseng can be a useful integrative supplementation for the prevention and/or treatment of seasonal acute upper respiratory infections (SAURIs). This review was conducted in accordance with the PRISMA guidelines 22 . The protocol was registered both in Open Science Framework (link: https://osf.io/rw369, DOI: 10.17605/OSF.IO/RW369), and in PROSPERO (code: CRD42020156235). A copy of the review protocol can be also found in the Appendix 1 of this article for a rapid consultation. The Appendix 2 contains the 27-item PRISMA checklist. All articles describing the efficacy of ginseng for the treatment or prevention of seasonal acute upper respiratory infections (SAURIs) were included in the review. The following PICOS criteria for inclusion and exclusion of studies in the systematic review were applied: • Population. Inclusion: patients (any age) with SAURIs (e.g.: influenza or common cold), reporting at least a respiratory symptom like runny nose, sneezing, cough, sore throat, nasal or sinus congestion, in combination with at least a systemic symptom like fever, chills, myalgia, fatigue, headache. All relevant studies were included regardless of their participants' comorbidities. Details about article screening and study selection process were reported in a flowchart ( Figure 1 The following PICOS criteria for inclusion and exclusion of studies in the metaanalysis were applied: • Population. Adult subjects with SAURIs (e.g.: influenza or common cold) and no relevant comorbidities. In order to maximize retrievable evidence on the topic and to reduce the risk of publication bias, data from studies involving sub-healthy participants were also included. Sub-healthy subjects were defined as individuals affected by a stable and mild or early-stage chronic condition, taking no drugs and not affected by any other relevant disease. Additional analyses were performed to evaluate the impact of studies not involving healthy subjects on the overall result of our quantitative synthesis. • Intervention. The oral administration of any extract obtained from ginseng (Panax ginseng, Panax notoginseng, or Panax quinquefolius) at any dosage over a well defined period (minimum: 8 weeks). • Control. The oral administration of placebo pills. • Outcomes. Outcome 1: the risk ratio for being infected throughout the study period. Outcome 2: the duration of disease symptoms (measured in days) after being infected. • Study design. Randomized Controlled Trials (RCTs). When article full-texts or essential details of included studies were missing, authors were contacted both by email and through ResearchGate®. However, no additional useful information was collected in this way, and for one study it was only possible to retrieve the article abstract 23 . Despite this, considering that the study summary provided sufficient information to meet the PICOS criteria, it was decided to include the trial all the same in order to maximise retrievable evidence on the topic. The following data were extracted: participants' demographics and baseline characteristics (including their influenza vaccination status), details regarding intervention (e.g.: ginseng type, dose, duration of administration) and comparison type, outcome measures (duration, severity, and symptoms of SAURIs; incidence of SAURIs during the study period; microbial etiology of respiratory infections; reported adverse events), information about study design, funding sources and country where the trial was performed. End-of-study significant differences between groups in any efficacy or safety outcome were also reported. The most relevant data were summarized in a table (Table 1) Version 2 of the Cochrane risk-of-bias tool for randomized trials (RoB 2) was used for the quality assessment of included RCTs 24 . Analyzed domains were the following ones: risk of bias arising from the randomization process, risk of bias due to deviations from the intended interventions, missing outcome data, risk of bias in measurement of the outcome, risk of bias in the selection of the reported result, and the overall risk of bias. Results of the assessment was adequately considered to inform the qualitative data synthesis in the discussion section of the review. Two reviewers were involved in the quality assessment of included studies (M.A.; D.D.). Disagreements between reviewers' judgements were resolved by discussing any relevant issue with a third reviewer (F.F.). All details regarding the risk of bias final assessment were displayed in Figure 2 . In the first meta-analysis, the chosen measure of effect size was the relative risk (RR) for being infected throughout the study period. The Mantel-Haenszel method was used to weight each trial and, when necessary, the treatment arm continuity correction (TACC) was applied 25 . Results of this meta-analysis were graphically displayed with a L'Abbé plot, a dedicated scatter plot for binary data ( Figure 3 ). In the second meta-analysis, the chosen measure of effect size was the duration of disease symptoms (measured in days) after being infected. The mean difference (MD) was adopted to combine data of all includible studies and the inverse variance J o u r n a l P r e -p r o o f method was used to weight each included trial. Results of this meta-analysis were graphically displayed with a forest plot ( Figure 4 ). Data synthesis was conducted per trial arm, thus using aggregated data rather than individual participant data. Trials with three arms (two interventional/ginseng-based and one control/placebo arm) were considered as if they were two different studies: the first one comparing one intervention with control, while the second one comparing the other intervention with control. A random-effects model was adopted for both meta-analyses. The Hartung-Knapp-Sidik-Jonkman adjustment for randomeffects models was applied, since it is demonstrated that it outperforms the standard DerSimonian-Laird method 26 . The threshold for significance of the overall effect size was set at p<0.05. I² was used as a measure of consistency, and I² values of 25%, 50%, and 75% were interpreted as representing small, moderate and high levels of heterogeneity, respectively 27 . Statistical analysis was performed with "R-Studio" software by two authors (D.D.; M.A.), and, in cases of disagreement, a third author (F.F.) was consulted to reach consensus. Following the Cochrane recommendations, publication bias was assessed with a dedicated funnel plot, the Egger's test and the trim-and-fill method in the first metaanalysis (where the number of trials was close to 10), but this approach was not feasible in the second meta-analysis, due to the limited number of included studies J o u r n a l P r e -p r o o f 28 . In particular, first of all, the funnel plot was visually assessed, and asymmetry as well as an irregular arrangement of points (representing included studies) were considered suggestive for publication bias 29 . Then, the Egger's test was performed and, as recommended by its authors, a statistically significant result was interpreted as an indication of publication bias 30 . Afterwards, if previous tests were positive, the trim-and-fill method was applied as a sensitivity analysis in order to provide an estimated effect of intervention after adjusting it for the publication bias 28, 31, 32 . The p-curve method was adopted for both meta-analyses to further assess the risk of bias across studies and to detect any potential "p-hacking" 33, 34 . The p-curve method was used to test if the sets of included studies were, on average, powered enough to detect a true effect of studied intervention, and to correct for the potentially inflated estimates that arise from the publication of results intentionally modified to be significant ("p-hacking") 33, 34 . All these analyses, aimed at assessing the potential risk of bias across studies, were performed with "R" software. A qualitative subgroup analysis was performed with regard to the patient's specific Finally, given that in one included trial sub-healthy individuals with early-stage chronic leukemia were recruited 35 , a leave-one-out analysis was performed to estimate the effect size of intervention exclusively based on data of studies with healthy subjects. It was not possible to perform any meta-regression to find explanations for heterogeneity, due to the limited number of included studies. The search of electronic databases and trial registries globally yielded 1242 results, and 821 articles remained when duplicates were removed. After the screening and selection process, nine articles describing ten studies were included in the review 23, [35] [36] [37] [38] [39] [40] [41] [42] . In one article, two trials were reported, labeled as "Trail A: CVT-E002 9907" and "Trail B: CVT-E002 2000-1" respectively 36 . Details about the article screening and selection process were reported in a dedicated flow diagram ( Figure 1 ). Overall, 2058 patients were recruited in included studies, and the number of subjects ranged from a minimum of 43 to a maximum of 783 (median: 104,5) across trials comprised in the systematic review. Females were more represented than males, J o u r n a l P r e -p r o o f accounting, on average, for around 57% of study populations. In one study, participants were children with a mean age of 5 years old 40 , whereas in four trials (described in three articles) only elderly subjects, aged 65 and above, were recruited [36] [37] [38] . In one trial, study population was composed of patients with Chronic Lymphocytic Leukemia (CLL) 35 ; in all the other included RCTs, participants were healthy subjects with no relevant comorbidities. Influenza vaccination status of participants varied across included studies: in four trials, subjects were recruited only if not vaccinated against the flu in the past 3 40 or 6 months 39, 41, 42 ; in three studies, patients were all vaccinated 23, 37, 38 ; in the remaining included RCTs, flu vaccination status was heterogeneous with only some participants being vaccinated, but no significant differences between groups were detected 35, 36 . Lifestyle habits (tobacco smoking or alcohol drinking) of study subjects, when reported, were described in the Supplementary Material B. In seven studies P. quinquefolius was administered to participants [35] [36] [37] [38] [39] [40] , whereas in three trials P. ginseng was given to patients 23, 41, 42 . No included study investigated the effects of P. notoginseng on SAURIs. In two studies, one group of participants was administered the ginseng extract given to the main intervention group but at a low-dose regimen 38, 40 . In one trial, intervention groups were given two different types of ginseng extracts named "GS-3K8" and "GINST" respectively 42 . In all but one RCTs, intervention was administered daily for 8 to 16 weeks, whereas in the trial conducted by Vohra and colleagues, ginseng was only given to patients at the onset of respiratory symptoms for a few days, thus only testing its therapeutic but not its preventive efficacy 40 . All included trials were placebo-controlled and, as described by study authors, participants randomly assigned to control groups were given placebo pills seemingly indistinguishable from ginseng capsules. In three included studies, it was explicitly reported that placebo composition was formulated in such a way as to taste of ginseng when ingested in order to further conceal its inert composition [40] [41] [42] The main health condition of interest, namely the occurrence, length and severity of SAURIs, was defined according to symptomatic criteria in all included studies, as shown in Table 1 study investigators, patients were tested in order to find a laboratory confirmation of the specific microbial etiology of disease 36, 38 . With regard to the preventive efficacy of ginseng administration (percentage of patients who developed a SAURI at least once during the study period), in seven trials a significant result in favor of intervention was found 23, [36] [37] [38] 41, 42 , in one trial this outcome was not reported 40 , and in two trials the difference between groups was not significant 35, 39 . In the two trials in which the preventive efficacy of ginseng administration and the microbial etiology of SAURIs were analyzed together, pooled results showed a significant result in favor of intervention for a reduced incidence of laboratory-confirmed influenza illness 36 . When considering the therapeutic efficacy of ginseng administration (days of sickness), in two trials intervention was significantly associated with a decrease in the duration of disease 37, 39 , data regarding this outcome were not retrievable in one trial 23 , while in the other included studies the difference between groups was not significant, as reported in Table 1 . If the efficacy of ginseng administration in reducing the severity of symptoms was taken into account, in two trials a significant effect associated with intervention was found 35, 39 , in four trials this outcome was not assessed 23, 37, 40, 42 , while in the remaining included studies no significant difference between groups was detected, as displayed in Table 1 . In general, ginseng administration appeared safe and well tolerated by patients involved in included studies, with no significant differences between intervention and placebo groups in terms of analyzed safety outcomes, such as the frequency, severity or type of adverse effects (Supplementary Material B) . In four trials, no differences in main hematological parameters, including blood markers of liver and kidney function, were detected 23, 41, 42, 45 . Laboratory safety data of one trial 39 were J o u r n a l P r e -p r o o f retrieved from another article 45 in which, in a subgroup of 42 study subjects whose blood was analyzed, intervention was associated with a significant increased proportion of CD4 and NK cells. All studies included in the review were randomized double-blind placebo-controlled trials. Two RCTs were pilot studies principally aimed at assessing intervention safety and the feasibility of larger trials on the topic 40, 42 . Follow-up duration ranged from 8 weeks to 6 months across included studies, as reported in the Supplementary Material B. The overall risk of bias of individual studies was rated as low for one trial 41 , high for three studies 23, 36, 38 , and some concerns were raised for the remaining RCTs. The most relevant concerns regarded the patients' self-reporting modality of SAURIsrelated symptoms and the participants' dropout rates. All details of the risk of bias assessment were reported in Figure 2 . The overall result of the first meta-analysis, which included 9 trials involving 1550 participants, significantly favored ginseng-based interventions in terms of relative risk With regard to the first meta-analysis, the funnel plot visually showed some degree With regard to the second meta-analysis, the p-curve referred to the C.8 subanalysis including only studies with healthy subjects was significantly right skewed (p<0.05), the test for flatness was not significant (p=0.63), and the power estimate was 45% (C.I. 5% to 88%) (Supplementary Material C). The p-curve referred to the C.9 sub-analysis including only studies at non-high risk of bias with healthy subjects was significantly right skewed (p<0.05), the test for flatness was not significant (p=0.88), and the power estimate was 73% (C.I. 17% to 97%) (Supplementary Material C). Pre-clinical laboratory studies underscore that ginseng extracts have antimicrobial properties against viruses usually involved in SAURIs such as rhinoviruses, influenza viruses, and respiratory syncytial virus [46] [47] [48] . Based on available data, it has been hypothesized that ginseng extracts can synergically exert their antimicrobial effects through different mechanisms of action, including a direct antiviral activity (inhibition of virus penetration and replication) and the enhancement of host immunity, to which the majority of ginseng effects are attributed 46, 47, 49 . Furthermore, laboratory studies have shown that the antiviral activity of ginseng against a broad range of influenza viruses appears dose-dependent 47 , and that the administration of ginseng extracts to mice can boost the immune response to influenza vaccination, thus acting as a vaccine adjuvant 46 and humoral components of the immune system have been suggested both for P. ginseng 50 and for P. quinquefolius 51 . Overall, the mechanisms of action of all ginseng species have been mostly studied on the basis of pre-clinical studies and, for infectious diseases, are hypothesized to be a general boost of the immune system, including an adjuvation of influenza vaccination. In a previous systematic review of clinical studies published up to December 2007, it was concluded that P. quinquefolius seemed effective in shortening the duration of acute respiratory infections in healthy adults, although it was unclear whether it could reduce the incidence or severity of common colds 52 . The findings of our qualitative synthesis suggested that, with regard to the overall preventive or therapeutic efficacy J o u r n a l P r e -p r o o f of each ginseng subspecies, the most relevant supporting evidence was about P. ginseng and P. quinquefolius. The included trial characterized by the highest methodological quality, thus being the only one with a low risk of bias, indicated that ginseng may be useful to reduce the incidence of acute respiratory infections, although no significant difference compared to placebo was found with regard to a potential reduction of disease duration and severity 41 . Among RCTs which remained after the exclusion of pilot studies and trials characterized by a high risk of bias, the use of ginseng was demonstrated to have a significant action even on the reduction of SAURIs severity and duration 35, 37, 39 . However, some concerns were raised about their methodological quality. Overall, in the majority of included RCTs analyzing the preventive efficacy of ginseng, a significant result in favor of intervention was found 23, [36] [37] [38] 41, 42 , whereas in two trials the difference between groups was not significant 35, 39 . In one of these two studies, patients with chronic leukemia were recruited, and their haematological health condition might have weakened the immune boosting effect of ginseng, possibly due to an insufficient drug dose or to the impairment of toll-like receptor pathways in such patients 35 . In the other trial, although no difference between intervention and control groups was observed in the number of subjects who had at least one cold during the study period, a significant difference between groups was reported when analyzing the proportion of participants who experienced two or more colds, as well as the severity of symptoms 39 . Here, the exclusion of many potentially eligible subjects from the study before randomization due to missing information, along with a consistent drop-out rate during the trial period (exceeding 20%), might have influenced the results. Nevertheless, it is interesting to notice that in both trials, study subjects were not vaccinated against influenza 35, 39 . Furthermore, in those RCTs in which the preventive efficacy of Therefore, on the basis of available evidence included in our quantitative synthesis and on their risk-of-bias assessment, it is not possible to affirm that ginseng supplementation can significantly reduce the incidence of SAURIs because the true effect might be different from the estimated effect. Nevertheless, considering both the overall result of the first meta-analysis and the above mentioned pre-clinical findings, existing data don't exclude that ginseng supplementation in adjunct to vaccination might have some preventive effects on SAURIs, and more high-quality RCTs are advocated to better study this potential activity. The overall result of the second meta-analysis indicated that ginseng supplementation cannot significantly reduce the duration of SAURIs symptoms if compared with placebo ( Figure 4) . However, when the study with sub-healthy individuals was removed from the analysis 35 , the result favoring intervention became significant, thus suggesting a potential effect of ginseng supplementation to reduce the duration of SAURIs by around 3 days on average (Supplementary Material C). In this meta-analysis, it was not possible to quantitatively assess the risk of publication bias with a funnel plot plot and the Egger's test, due to the limited number of included studies. The p-curve method didn't demonstrate a potential risk of "phacking" (Supplementary Material C). If the study at high risk of bias was excluded 38 , the p-curve shape further ameliorated, thus suggesting a higher average power J o u r n a l P r e -p r o o f estimate of the set of included studies. It is possible that, by conducting more highquality trials on the topic with healthy subjects, the result in favor of intervention may be confirmed. Globally, data from included trials suggested that studied ginseng extracts were relatively safe and well-tolerated by recruited subjects. In two systematic reviews investigating the safety of P. ginseng, it was concluded that this ginseng type shows a safe profile in the limited number of available RCTs on the topic, involving both healthy subjects and patients with various clinical conditions, and its use is generally associated with a low incidence of adverse effects 53, 54 . Based on available data, the safety profile of P. quinquefolius appears equally good, even on a relatively longterm (up to 12 weeks) 55, 56 . The oral consumption of all ginseng species has been reported to be sometimes responsible for adverse effects like hypertension, tachycardia, dry mouth, gastrointestinal disturbances, insomnia, and nervousness 56, 57 . Three cases of manic psychosis associated with ginseng consumption have been reported in predisposed individuals 58, 59 . A possible, although controversial, estrogenic effect has been also described 56, 60, 61 , as well as a potential increased risk of operatory bleeding following its high-dose oral intake 62 . Therefore, ginseng administration is contraindicated in patients who are expected to undergo surgery, or affected by psychotic disorders, mania, estrogen-dependent diseases, hormonal dysfunctions, hypertension, or hyperthyroidism 63 . Additionally, possible interactions with several medicinal drugs have been described, including anticoagulants, monoamine oxidase inhibitors, anti-diabetic agents, antiretroviral compounds, diuretics, and cytochrome P450-3A4 substrates, as well as caffeine-based and other J o u r n a l P r e -p r o o f stimulating substances 56, 61, 63, 64 . However, the ginseng-drugs interaction profile is not still fully clear to date, and, for example, with regard to warfarin, some authors suggest a potential inhibition of its anticoagulant effect 65, 66 , whereas others underscore no significant interaction in experimental settings 67, 68 . Based on results of vitro studies 61 , ginseng administration is to be avoided in pregnant women, especially during the first trimester, due to potential risks to the fetus 63 . Although ginseng has been reported to be well tolerated if administered to children at a proper dose and for a short time period 40, 69 , data are still very limited in this specific category of patients: therefore, extreme caution is required in the pediatric population. Furthermore, it has to be reported that some adverse effects wrongly attributed to ginseng, like androgenization, have been eventually discovered to be caused by adulterants 61 , thus urging the need for stricter controls by health authorities over ginseng production and marketing. Overall, provided that clinical safety data of ginseng consumption are scant, further studies are advised and medical supervision is required for its safe and proper use. Nevertheless, as shown by the results of included trials, its short-term administration can be considered quite safe in healthy adults taking no drugs. Evidence base on the topic is limited. Among included RCTs, two studies were pilot trials involving a small number of participants 40, 42 , and it was not possible to retrieve the full-text version of a relevant article 23 . For most included RCTs some concerns were raised with regard to their overall risk of bias, especially when considering missing information, drop-out rates, and the symptoms self-reporting modality. Our analysis also individuated a potential risk of publication bias, thus indicating a J o u r n a l P r e -p r o o f possible over-representation in the scientific literature of under-powered small trials yielding positive results. Finally, "p-hacking" couldn't be totally excluded. Limitations of existing evidence don't allow to draw conclusions on the topic. Nevertheless, it is not excluded that ginseng supplementation in adjunct to influenza vaccination and standard care might be useful for SAURIs prevention and management in healthy adult subjects, but further high-quality trials are needed to support this hypothesis. FUNDING: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. No conflict of interest has to be disclosed by the authors. Caption: The first meta-analysis aimed to assess the relative risk for developing a seasonal acute upper respiratory infection at least once during the study period (winter seasons). Intervention was defined as taking ginseng and all trials were placebo-controlled. Each point represented a study included in the quantitative synthesis (red: P. ginseng; blue: P. quinquefolius). The X axis indicated the event rate in the control group, whereas the Y axis displayed the event rate in the experimental intervention group. Preventive efficacy (infected patients: %) = percentage of patients who developed a seasonal acute respiratory infection at least once during the study period (significant -p < 0.05 -difference between groups). Therapeutic efficacy (duration of disease: days) = duration of disease in days (see Supplementary Table A. 3 for further details) (significant -p < 0.05 -difference between groups). Therapeutic efficacy (symptoms severity) = report of any significant (p < 0.05) difference in disease symptoms severity favoring ginseng intervention groups (yes/no/not reported) (significant -p < 0.05 -difference between groups). (*) = statistically significant (p < 0.05). (ns) = non-statistically significant (p ≥ 0.05); when p = 0.05, it was explicitly indicated in the table. -= not investigated. ? = irretrievable data. An explanation for the seasonality of acute upper respiratory tract viral infections Common cold Complications of viral influenza Estimated Influenza Illnesses Economic burden of seasonal influenza in the United States The economic burden of non-influenza-related viral respiratory tract infection in the United States Prevention and Control of Seasonal Influenza with Vaccines: Recommendations of the Advisory Committee on Immunization Practices -United States, 2019-20 Influenza Season flu and coughing: a review of over-the-counter cold and flu medicines Complementary medicine for treating or preventing influenza or influenza-like illness A comparison of the ancient use of ginseng in traditional Chinese medicine with modern pharmacological experiments and clinical trials Prevention and Treatment of Influenza, Influenza-Like Illness, and Common Cold by Herbal, Complementary, and Natural Therapies Potential of Complementary and Alternative Medicine in Preventive Management of Novel H1N1 Flu (Swine Flu) Pandemic: Thwarting Potential Disasters in the Bud Complementary treatment of the common cold and flu with medicinal plants--results from two samples of pharmacy customers in Estonia Pharmacists' patterns of use, knowledge, and attitudes toward complementary and alternative medicine Physicians' attitudes toward complementary and alternative medicine and their knowledge of specific therapies: a survey at an academic medical center FITOTERAPIA: Guida all'uso clinico delle piante medicinali Chemical Diversity of Panax ginseng, Panax quinquifolium, and Panax notoginseng Chemical and pharmacological studies of saponins with a focus on American ginseng Differences in the volatile compositions of ginseng species (Panax sp.) Characterization of Korean Red Ginseng (Panax ginseng Meyer): History, preparation method, and chemical composition Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement Efficacy and safety of the standardised Ginseng extract G115 for potentiating vaccination against the influenza syndrome and protection against the common cold RoB 2: a revised tool for assessing risk of bias in randomised trials What to add to nothing? Use and avoidance of continuity corrections in meta-analysis of sparse data The Hartung-Knapp-Sidik-Jonkman method for random effects meta-analysis is straightforward and considerably outperforms the standard DerSimonian-Laird method Cochrane Handbook for Systematic Reviews of Interventions Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials Bias in meta-analysis detected by a simple, graphical test The trim-and-fill method for publication bias: practical guidelines and recommendations based on a large database of meta-analyses Performance of the trim and fill method in the presence of publication bias and between-study heterogeneity P-curve: a key to the file-drawer p-Curve and Effect Size: Correcting for Publication Bias Using Only Significant Results A randomized, controlled trial of Panax quinquefolius extract (CVT-E002) to reduce respiratory infection in patients with chronic lymphocytic leukemia A placebo-controlled trial of a proprietary extract of North American ginseng (CVT-E002) to prevent acute respiratory illness in institutionalized older adults Efficacy of COLD-fX in the prevention of respiratory symptoms in community-dwelling adults: a randomized, double-blinded, placebo controlled trial Efficacy and Safety of CVT-E002, a Proprietary Extract of Panax quinquefolius in the Prevention of Respiratory Infections in Influenza-Vaccinated Community-Dwelling Adults: A Multicenter, Randomized, Double-Blind, and Placebo-Controlled Trial Efficacy of an extract of North American ginseng containing poly-furanosyl-pyranosyl-saccharides for preventing upper respiratory tract infections: a randomized controlled trial Safety and tolerability of North American ginseng extract in the treatment of pediatric upper respiratory tract infection: a phase II randomized, controlled trial of 2 dosing schedules Preventive effect of Korean red ginseng for acute respiratory illness: a randomized and double-blind clinical trial A randomized, double-blind, placebo-controlled pilot study to assess the effects of protopanaxadiol saponin-enriched ginseng extract and pectinase-processed ginseng extract on the prevention of acute respiratory illness in healthy people Transmission of the common cold to volunteers under controlled conditions. I. The common cold as a clinical entity Canadian Acute Respiratory Illness and Flu Scale (CARIFS) Immune Modulating Effects of Daily Supplementation of COLD-fX (a Proprietary Extract of North American Ginseng) in Healthy Adults the natural effectual antiviral: Protective effects of Korean Red Ginseng against viral infection Antiviral Activity of Fermented Ginseng Extracts against a Broad Range of Influenza Viruses Ginseng alleviates microbial infections of the respiratory tract: a review Antiviral activity of ginseng extract against respiratory syncytial virus infection Immunity Boost": The Effects of Panax ginseng on Immune System Immunomodulating activity of CVT-E002, a proprietary extract from North American ginseng (Panax quinquefolium) North American (Panax quinquefolius) and Asian Ginseng (Panax ginseng) Preparations for Prevention of the Common Cold in Healthy Adults: A Systematic Review Systematic Review of Randomized Controlled Trials Evaluating the Efficacy and Safety of Ginseng American Ginseng Extract (Panax quinquefolius L.) Is Safe in Long-Term Use in Type 2 Diabetic Patients Efficacy and safety of ginseng The Pharmacology of Chinese Herbs Manic psychosis associated with ginseng: a report of two cases and discussion of the literature A case of ginseng-induced mania Panax ginseng: a newly identified cause of gynecomastia Safety and efficacy of panax ginseng during pregnancy and lactation Over-the-counter natural products in cardiac surgery: a case of ginseng-related massive perioperative bleeding Interazioni fra erbe, alimenti e farmaci Elevated liver enzymes resulting from an interaction between Raltegravir and Panax ginseng: a case report and brief review Probable interaction between warfarin and ginseng Interactions of warfarin with garlic, ginger, ginkgo, or ginseng: nature of the evidence Interaction between warfarin and Panax ginseng in ischemic stroke patients Effect of St John's wort and ginseng on the pharmacokinetics and pharmacodynamics of warfarin in healthy subjects Effects of Korean red ginseng extract on behavior in children with symptoms of inattention and hyperactivity/impulsivity: a doubleblind randomized placebo-controlled trial This research work was not funded and no conflict of interest has to be disclosed by the authors.