Observer agreement of lower limb venous reflux assessed by duplex ultrasound scanning using manual and pneumatic cuff compression in patients with chronic venous disease and controls

Abstract

Objectives: The study aimed to evaluate observer agreement between two experienced ultrasound operators examining deep venous reflux assessed by duplex ultrasound (DU) using either manual or pneumatic cuff compression. In addition, the two methods were compared with each other with regard to immediate "eyeballing" and direct measurements of reflux time from Doppler flow curves.

Design: This was a case control study.

Material and methods: Cases were found among patients admitted to our department with deep venous thrombosis of the iliac, femoral or popliteal veins during the period 1999-2006. Controls were departmental staff. DU was used to assess valve function in the common femoral, femoral and popliteal veins in the standing position using manual and pneumatic cuff compression. The investigators were blinded to the other's observations. Observer agreement was assessed using the Rasch model for binary items.

Results: Twenty patients and 20 controls participated in the study and were analysed by the Rasch model. Quantitative measurement was found to be more reliable than "eyeballing", and cuff compression was more reproducible in identifying reflux than manual compression. We found that assessment by manual measurement by one investigator functioned at a lower level of expertise than for the other investigator.

Conclusions: Cuff measurements were more accurate in diagnosing deep venous reflux than manual measurements, and measurement was more accurate than "eyeballing". The fact that assessment by manual compression by one investigator functioned at a lower level of expertise suggests that cuff measurement might be the optimal assessment method, especially in the difficult cases.