Introduction: Several systematic reviews and meta-analyses of primary studies have been published on the relationship between annual case load of carotid endarterectomy (CEA) and carotid artery stenting (CAS) performed at hospital level or by individual surgeons, and perioperative outcomes. Many studies on volume-outcome relationship have already been published and high-quality systematic reviews are crucial for further guideline development.
Evidence acquisition: Systematic reviews and meta-analyses on the relationship between hospital or surgeon CEA/CAS volume and periprocedural outcomes were identified through a systematic literature search of Medline, Web of Science, and the Cochrane Database of Systematic Reviews. Methodological quality of the systematic reviews was appraised using the AMSTAR2 tool independently by two authors. Systematic reviews were aggregated in their volume-outcome findings. Quantitative data from primary studies included in the systematic reviews were synthesized. Additionally, volume definitions and the time point of outcome assessment used in primary studies were analyzed.
Evidence synthesis: In total, five systematic reviews published between 2000 and 2018 were identified, each comprising 11-25 primary studies. Methodological quality appraisal of these reviews revealed high quality for only the most recent review, low quality for three reviews, and critically low quality in one review. Aggregation of the systematic reviews revealed a significant inverse relationship between hospital/operator volume and the periprocedural risk of death or stroke following CEA. For CAS, high operator volume was associated with lower outcome rates. Regarding hospital volume, an inverse but non-significant relationship between CAS hospital volume and outcome rate was found. In our synthesis of primary studies from these systematic reviews an inverse CEA hospital and operator volume relationship was present for stroke or death and for CAS for hospital volume, respectively. A high heterogeneity regarding the definitions of volume categories, and of time points assessing outcomes was apparent.
Conclusions: For CEA, high quality aggregated evidence revealed an inverse relationship between hospital/surgeon CEA volume and periprocedural rate of stroke or death. The same was true for operator linked CAS volume. Regarding hospital linked CAS volume, no unequivocal evidence was found. Additionally, heterogeneity was found regarding volume definition, and time of outcome assessment. Thus, future studies should aim to harmonize volume definitions and outcome time points.