Despite over 50 years of silicate bioactive glass (SBG) research, commercial success, and 6000+ published articles, there remains a lack of understanding of how soluble silicate (Si) species released from SBGs influences cellular responses. Using a systematic approach, this article quantitatively compares the in vitro responses of cells to SBG dissolution products reported in the literature and determines if there is a Si concentration ([Si]) dependent effect on cell behaviour. Cell behavioural responses to SBGs [Si] in dissolution products included metabolic activity (reported in 52 % of articles), cell number (24 %), protein production (22 %), gene expression (22 %) and biomineralization (24 %). There was a difference in the [Si] reported to cause increased (desirable) cellular responses (median = 30.2 ppm) compared to the [Si] reported to cause decreased (undesirable) cellular responses (median = 52.0 ppm) (P ≤ 0.001). The frequency of undesirable outcomes increased with increasing [Si], with ∼3 times more negative outcomes reported above 52 ppm. We also investigated the effect of [Si] on specific cellular outcomes (e.g., metabolic activity, angiogenesis, osteogenesis), if cell type/species influenced these responses and the impact of other ions (Ca, P, Na) within the SBG dissolution media on cell behaviour. This review has, for the first time, quantitatively compared the cellular responses to SBGs from the literature, providing a quantitative overview of SBG in vitro practices and presents evidence of a range of [Si] where desirable cellular responses may be more likely (30-52 ppm). This review also demonstrates the need for greater standardisation of in vitro methodological approaches and recommends some minimum reporting standards.

Statement of significance

This systematic review investigates the relationship between the concentration of Si released from Si-bioactive glasses (SBG) and in vitro cellular responses. Si releasing materials continue to be of considerable scientific, commercial, and medical interest (with 1500+ articles published in the last 3 years) but there is considerable variation in the reported biologically effective Si concentrations and on the importance of Si on cell behaviour. Despite the variation in methodological approaches, this article demonstrated statistical commonalities in the Si concentrations that cause desirable and undesirable cellular behaviours, suggesting a window where positive cellular outcomes are more likely. This review also provides a quantitative analysis of in vitro practices within the bioactive glass field and highlights the need for greater standardisation.