A total of 2828 records have been identified after a systematic literature search from the abovementioned databases. Prior to the title and abstract screening, 1284 duplicate records were removed. Subsequently, 41 full-text articles were retrieved. Five articles were retrieved from a manual search as well. Full-text screening was performed on 46 articles based on our review eligibility criteria. Finally, 12 studies were incorporated into our review [11, 12, 14,15,16,17,18,19,20,21,22,23] (see Fig. 1).

PRISMA flow diagram of the screening and selection process

All 12 studies included were placebo-controlled randomized trials. The types of probiotic intervention used have been mentioned in Table 1.

Drago et al. 2022, reported a reported a significantly reduced proportion of patients in the probiotic group who experienced asthma exacerbations, with 24 out of 212 patients in the probiotic group and 67 out of 210 in the placebo group (11). However, the mean asthma exacerbation duration was similar in both groups, which were 3.3 ± 2.45 days and 3.3 ± 2.57 days in the probiotic and placebo groups, respectively. They have used a multi-strain probiotics formulation of Ligilactobacillus salivarius LS01 and Bifidobacterium breve B632.

Of all the included studies, Liu et al. 2021 and Li et al. 2022 reported asthma control test scores after the probiotic intervention [12, 14]. Both studies showed improved asthma control test scores after the probiotic intervention. Li et al. 2022 used Lactobacillus reuteri CCFM1040 as an intervention, whereas Liu et al. 2021 used Bifidobacterium lactis Probio-M8 as their study intervention [12, 14].

The fixed effect model was chosen due to the absence of statistical heterogeneity among studies (I2 ≤ 50% or P ≥ 0.10). The pooled analysis of two studies with 63 asthma patients showed a significant p-value of 0.0001 and an OR of 1.18 with a 95% CI: 1.18–3.64. (Fig. 2).

Forest plot for asthma control test

Out of all the studies included, we could extract the data on fractional exhaled nitric oxide (FeNO) from Liu et al. 2016 and Liu et al. 2021 [18, 24]. While Liu et al. 2016 did not conclude any significant impact of the probiotic intervention on FeNO, Liu et al. 2021 concluded that the probiotic-treated group improved FeNO compared to the placebo group [18, 24]. However, the pooled analysis revealed statistically non-significant (p-value of 0.46) (Fig. 3).

Liu et al. 2016 found lower eosinophils at six months of Clostridium butyricum intervention than the placebo group [18]. Likewise, Liu et al. 2021 observed with Bifidobacterium lactis Probio-M8 intervention [24]. However, from pooled analysis, we could not find a statistically significant (p-value of 0.29) impact of probiotics intervention on the eosinophilia condition (Fig. 4).

Forest plot for eosinophils

Out of all the studies included, we could extract the data on forced expiratory volume in 1 s (FEV1) from Liu et al. 2016, Liu et al. 2021, Chen et al. 2010, and Gutkowski et al. 2010, none of the studies concluded any significant impact of probiotic intervention on FEV1. Since the data units were not uniform across the studies a meta-analysis could not be performed. [12, 18, 20, 21]

We also extracted data on peak expiratory flow rate (PEFR) from Liu et al. 2016, Liu et al. 2021 and Chen et al. 2010. All three studies found no significant impact of probiotic intervention on PEFR. [12, 18, 20]

Out of all the studies included, we could extract data on adverse effects from Li et al. 2022, Drago et al. 2022, Satia et al. 2021, Huang et al. 2018, Smith et al. 2016, Chen et al. 2010. Only Satia et al. 2021 reported that one patient developed a mild upper respiratory tract infection during the treatment period and one subject developed mild transient nausea with no serious adverse events. However, no adverse effects were reported with probiotics intervention in other studies [11, 14,15,16, 19, 20]

We have graded the risk of bias assessment as low, high, and unclear by using the CROB tool (Table 2).