Corruption of the Blood-Brain Barrier Literature Identified by ‘TorturedPhrases’: A Case Study for Bibliometric Neuroethics

Patients living with a neurological disease depend on the integrity of the neuroscience literature to improve the probability of effective treatments becoming available to them. The BloodBrain Barrier (BBB) is one of the key components of the nervous system, and its dysfunction is implicated in different neurological diseases. Non-standardized terms (or ‘tortured phrases’; TP) to describe the BBB have emerged in the indexed and non-indexed literature, and their use is suggestive of low-quality science or even misconduct. A total of 13 variants of BBB TPs were initially discovered on Google Scholar on 18-20 April 2024, followed by new variants and cases on 21-30 May 2025. In total, 260 documents (220 journal papers, 26 book chapters, eight conference proceedings, and six preprints), with and without a DOI, were identified. The three most common variants (i.e., TPs) of BBB were blood-brain obstruction, blood-brain boundary, and blood-cerebrum boundary/hindrance, identified in 84, 58, and 31 documents, respectively. Only two variants (bloodmind boundary and blood-brain obstruction) were found in the Tortured Phrase Detector of the Problematic Paper Screener, while a total of four and one documents in Scopus and Web of Science Core Collections databases, respectively, contained any of these BBB TPs. The existence of these 19 variants of TPs suggests the corruption of the associated BBB literature. This study provides a methodological case study for neuroethicists wishing to use bibliometric methods to identify problematic instances of low-quality or insufficiently vetted neuroscience research via an approach that we term “bibliometric neuroethics.”