Study selection
Electronic searches from all sources retrieved 493 citations (Fig. 1). Using titles and abstracts to screen content, 305 citations were excluded duplications. 126 articles were excluded due to non-clinical studies in humans or were reviews or opinion papers. Out of 62 clinical trials, 38 did not meet the inclusion criteria [measured other outcomes [9], follow up period variation [16], variation in RCT design [3], missing values of outcome [6], variation in index used for plaque assessment [1] and full text was not available for two articles] (See Additional file 1: Table S2 for list of excluded trials and reasons). Most studies originated from Southeast Asia and all were in the English language.
Study description
The 24 RCTs comprising 1597 adults (899 HTP participants and 698 HMR participants) for inclusion in the summary analyses. Selected characteristics of the included studies are shown in Table 1. There were 15 HTP and 9 HMR trials using non-herbal toothpaste (NHTP) or non-herbal mouth rinse (NHMR) as the control arm. Eleven HTP studies [9, 22,23,24,25,26,27,28,29,30,31] assessed short-term effects (4-weeks follow up) on dental plaque reduction whereas four studies [31,32,33,34] assessed for long-term effects (12-weeks follow up). Ten HTP studies [7, 9, 22,23,24,25,26,27, 30, 31] assessed short-term effects (4 weeks follow up) on gingival inflammation reduction whereas three studies [31,32,33] assessed long-term effects. Among the HTP studies, seven and eight studies assessed short-term effects on dental plaque reduction and gingival inflammation reduction, respectively, with fluoridated toothpaste as the control. Six HMR trials each assessed short-term [35,36,37,38,39] and long-term effects [36, 37, 40,41,42] on dental plaque reduction. Six studies assessed short-term effects on gingival inflammation [35,36,37,38,39] reduction whereas five studies assessed for long-term effects [36, 37, 41, 42].
There was clinical heterogeneity in the herbal ingredients present in toothpastes studied. Four studies [7, 22, 25, 31] assessed chamomile (Matricaria recutita), two studies evaluated neem (Azadirachta indica) [9, 30], Aloe vera (Aloe barbadensis) [23, 33] and calendula (Calendula officinalis) [26, 32] respectively. Individual studies for salvoadoral persica [29], chitosan [28], ajamoda satva (Apium graveolens) [24], lippia sidiodes (Pepper-rosmarin) [34] and vaikrantha bhasma (Dolichos biflorus) were also conducted [27]. Eight HTP studies [22,23,24,25, 27, 30,31,32] used fluoride as the control, whereas four studies [26, 28, 33, 34] used placebo with the rest using non-herbal, non-fluoride OTC toothpastes. Six studies [9, 23, 26,27,28, 32] assessed dental plaque using the Silness and Löe Plaque Index [43] whereas eight studies [22, 24, 25, 29,30,31, 33, 34] assessed dental plaque using the Turesky-Gilmore modification of the Quigley Hein Plaque [44] Index in HTP studies. All studies assessed gingival inflammation by Silness and Löe Gingival Index [45].
Every HMR study had a different herbal ingredient, with the exception of two trials which had Neem (Azadirachta indica) as the active ingredient [36, 37]. Eight of these had chlorhexidine as the control while one had placebo. Five studies assessed dental plaque using the Silness and Löe Plaque Index whereas four studies assessed dental plaque using the Turesky-Gilmore modification of Quigley Hein Plaque Index in HMR studies. All studies assessed gingival inflammation by Silness and Löe Gingival Index. Clinical outcomes in all studies were measured as continuous variables reported as mean ± SD.
Risk of Bias assessments
A synthesis of the assessment of the methodological quality items (authors’ judgement of risk of bias for each included study) is presented in Additional file 1: Figure. S1. Three studies showed low risk of bias [22, 29, 38], seven studies had unclear risk [23, 25,26,27, 34, 36, 40] and the remainder were high risk. Additional file 1: Figure. S2 depicts a risk of bias graph, illustrating the authors’ judgements about each risk of bias item presented as percentages across all included studies. Among all, allocation concealment or selection bias and blinding of the participants had higher proportions of bias across the studies.
Synthesis of results - effect of interventions
Herbal toothpaste
Overall, in 11 pooled studies involving 712 adults (Table 2), participants using HTP were more likely to experience a reduction in dental plaque scores during a four-week period compared to those using NHTP [SMD 1.95, 95% CI (0.97 to 2.93)], but there was substantial heterogeneity (95%) across studies (Fig. 2-2a). However, 4 trials studying long-term effects did not favour HTP for reduction in dental plaque [SMD 0.89, 95% CI (− 0.93 to 2.72)]. Regarding gingival inflammation, for both short-term [SMD 0.09, 95% CI (− 0.14 to 0.00), 10 studies] and long-term effects [SMD 0.07, 95% CI (− 0.23 to 0.36), 3 studies], the pooled results did not significantly favour HTP when compared to NHTP (Fig. 2-2b).
The significant difference in Plaque Index reduction that was found at 4 weeks between HTP and NHTP was investigated with a sub group analysis. The controls (NHTP) were divided into non-fluoride toothpaste and fluoridated toothpaste.
HTP was not superior over fluoride toothpaste (SMD 0.99, 95% CI − 0.14 to 2.13, 7 studies, short-term) in reducing dental plaque at 4 weeks (Fig. 3-3a). however, it was favoured to reduce dental plaque over non-fluoride toothpaste (SMD 4.64., 95% CI (2.23, 7.05), 4studies] (Fig. 3-3b). In another subgroup analysis, HTP was favoured over NHTP when short-term studies used the Silness and Löe Index [MD 0.37, 95% CI (0.14 to 0.59), 5 studies] in reducing dental plaque (Fig. 3-3c). There was significantly greater reduction in Plaque was observed for HTP compared to non-fluoride toothpastes, but not with fluoride toothpastes;
Herbal mouth rinse
There was no difference in mean reduction of dental plaque [SMD -2.93, 95% CI (− 6.43 to 0.58), 6 studies, 582 participants] by HMR compared to NHMR for short-term use (Fig. 4-4a). However, there was substantial evidence of mean reduction of dental plaque by users of NHMR compared to HMR in 6 studies [SMD -2.61, 95% CI (− 4.42 to − 0.80), 285 participants) at 12 weeks. Regarding gingival inflammation, for both short-term [SMD -0.15, 95% CI (− 0.32 to 0.01), 6 studies] and long-term effects [SMD -0.09, 95% CI (− 0.25 to 0.08), 6 studies], the pooled findings did not significantly favour NHMR when compared to HMR (Fig. 4-4b).