OBJECTIVES: The objectives of this study were to study with Power Doppler US (PDUS) the SI joints (SIJs) of patients with suspected active sacroiliitis, to describe SIJ flows with spectral wave analysis (SWA) on Doppler US, and to correlate US data with both clinical characteristics and presence of SIJ bone marrow oedema (BME) in subsequent MRI. METHODS: A total of 42 patients (32 females and 10 males, mean age 46.8 years) with recent onset of inflammatory back pain (IBP) were included. Every patient underwent US examination with a convex 1-8 MHz probe [scoring PDUS signals with a three-point scale and describing flows in SWA calculating the mean Resistive Index (RI)] and subsequent MRI of the SIJs. RESULTS: PDUS signals were detected in 34 patients and 62 SIJs. In 29 patients and 56 SIJs, MRI revealed BME. A definite diagnosis of SpA was made in 32 patients. PDUS signals were more frequent (P < 0.0001) in patients with a final diagnosis of SpA, yielding a higher PDUS score (P = 0.0304). PDUS grading correlated with both BME grading (r = 0.740, P = 0.0001) and AS DAS (ASDAS) (r = 0.6257, P = 0.0004), but not with inflammatory reactants nor anthropometric data. Mean RI were, respectively, 0.60 and 0.73 (P < 0.0001) in patients with or without diagnosis of active sacroiliitis. The most inclusive RI cut-off resulted <0.70 [positive predictive value (PPV) 94%, accuracy 90%, P = 0.0001]. The best Likelihood Ratio (5.471) for RI to detect pathologic cases was obtained with a cut-off of <0.60 (PPV 96%). CONCLUSIONS: PDUS and SWA of SIJs demonstrate good diagnostic accuracy for active sacroiliitis compared with MRI.
OBJECTIVES: The objectives of this study were to study with Power Doppler US (PDUS) the SI joints (SIJs) of patients with suspected active sacroiliitis, to describe SIJ flows with spectral wave analysis (SWA) on Doppler US, and to correlate US data with both clinical characteristics and presence of SIJ bone marrow oedema (BME) in subsequent MRI. METHODS: A total of 42 patients (32 females and 10 males, mean age 46.8 years) with recent onset of inflammatory back pain (IBP) were included. Every patient underwent US examination with a convex 1-8 MHz probe [scoring PDUS signals with a three-point scale and describing flows in SWA calculating the mean Resistive Index (RI)] and subsequent MRI of the SIJs. RESULTS: PDUS signals were detected in 34 patients and 62 SIJs. In 29 patients and 56 SIJs, MRI revealed BME. A definite diagnosis of SpA was made in 32 patients. PDUS signals were more frequent (P < 0.0001) in patients with a final diagnosis of SpA, yielding a higher PDUS score (P = 0.0304). PDUS grading correlated with both BME grading (r = 0.740, P = 0.0001) and AS DAS (ASDAS) (r = 0.6257, P = 0.0004), but not with inflammatory reactants nor anthropometric data. Mean RI were, respectively, 0.60 and 0.73 (P < 0.0001) in patients with or without diagnosis of active sacroiliitis. The most inclusive RI cut-off resulted <0.70 [positive predictive value (PPV) 94%, accuracy 90%, P = 0.0001]. The best Likelihood Ratio (5.471) for RI to detect pathologic cases was obtained with a cut-off of <0.60 (PPV 96%). CONCLUSIONS: PDUS and SWA of SIJs demonstrate good diagnostic accuracy for active sacroiliitis compared with MRI.