Accuracy of the intima-media thickness by ultrasound for the diagnosis of giant cell arteritis: a systematic review

Objectives

To evaluate the performance of quantitative intima-media thickness (IMT) measurement by ultrasound (US) for diagnosing GCA.

Methods

A systematic literature review of the following electronic databases was performed: PubMed, Embase, Web of Science, Scopus, Lilacs, and Google Scholar (no date and language restriction, last search June 3, 2024). Studies that tested IMT`s diagnostic accuracy as a primary outcome in GCA suspected patients were included. The quality of the studies was assessed using the QUADAS-2 tool.

Results

Among 2786 records screened by title and abstracts, 7 fulfilled the inclusion criteria. The proposed IMT cut-off values for vasculitis in between the studies varied between 0.4 and 0.44, and 0.81 and 1.2 mm for the temporal (TA) and the axillary artery (Axa), respectively. Most cut-off values were post hoc calculated. Diagnostic accuracy vis-à-vis qualitative judgement of the respective segments by US or MRI was high (sensitivities: 76 to 100 %, specificities: 85.7 to 100 %) despite the different cut-offs used. All studies have a high risk of bias in at least two QUADAS-2 domains due to patient selection issues, lack of US blinding, and use of the index test as part of the reference standard.

Conclusion

The seven studies had considerable drawbacks due to biases across several domains. This might explain the high reported diagnostic accuracy for defining vasculitis of the TA and Axa segments, despite different cut-off values and reference standards being used in these studies. Reliable IMT cut-off values of the TA and the Axa for diagnosing GCA are not yet available.