I'm not sure if this article is publicly available, but it discusses the difference between an unblinded SSRA and a blinded SSRA. I won't copy and paste too much, but what I copy below gives the main differences. Seems that unblinded is more focused on treatment effect (i.e. % relative risk reduction in 3-point MACE), whereas blinded is more focused on control event rate and subject discontinuation rate. I think there are also partial blinded, or hybrid, analyses that are possible too. In any case, there are multiple variables that may go into the SSRA.

Unblinded Sample Size Re-estimation (SSR): In this type of adaptive design, the sample size is re-estimated at the interim analysis using unblinded data based on a prespecified criterion. In spite of results from previous clinical trials, the most crucial parameter for sample size estimation, the targeted treatment effect as an alternative for hypothesis testing, could be either unknown or known with a high level of uncertainty at the design stage. This uncertainty may result in an under- or overpowered study with potential to either fail to detect the treatment effect or lead to a large and inefficient trial. SSR allows more flexibility than a fixed trial by increasing the trial size if the initial assumed treatment effect to power the study is overly optimistic. It also gives an opportunity for staged investments in clinical research where the trial can start with a smaller sample size, assuming an optimistic scenario of the true treatment effect and then increase the sample size if the interim estimate is lower than the assumed one, but is still of clinical relevance.

Blinded Sample Size Re-estimation (SSR): Blinded SSR (bSSR) is a type of SSR that uses accumulating, aggregate data in the mid-course of the trial to re-estimate nuisance parameters that have an impact on the power calculation of the trial. Some examples of nuisance parameters are the variance of a continuous variable, the control group event rate, or subject discontinuation rate. When the sample size is adjusted using bSSR, the treatment effect in the ongoing study is not estimated and used in the algorithm of SSR.