Clinical and Radiological Differentiation of Septic Arthritis and Transient Synovitis of the Hip

Full Article

TWY Chin, KS Tse

Hong Kong J Radiol 2017;20:41-6

DOI: 10.12809/hkjr1716813

Objective: To distinguish septic arthritis (SA) and transient synovitis (TS) of the hip based on clinical, laboratory, and imaging parameters.

Methods: Hospital records of all paediatric patients with painful hip effusion referred for ultrasound-guided joint aspiration between January 2008 and February 2016 in a regional hospital in Hong Kong were retrospectively reviewed. Confirmed SA was defined as positive Gram’s stain or joint-aspirate culture, and presumed SA was defined as positive blood culture and a high joint-aspirate white cell count (>50,000 /mm3). TS was defined as negative joint-aspirate analysis. Multivariate analysis of the clinical, laboratory, and imaging parameters of the SA and TS groups was performed using logistic regression.

Results: Of 32 patients included, seven and 25 were confirmed or presumed to have SA and TS, respectively. Predictors for SA were body temperature of ≥38.5°C (odds ratio [OR] = 18, p = 0.02), serum C-reactive protein (CRP) level of ≥10 mg/l (OR = 7.9, p = 0.03), and an ultrasound finding of predominant synovial (capsular) thickening relative to joint effusion thickness at the anterior femoral recess (OR = 19, p = 0.01). Increasing effusion thickness had borderline significance in predicting SA (OR = 18.6 for each cm, p = 0.095). The receiver operating characteristic curve evaluating the test performance of effusion thickness in predicting SA had an area under the curve of 0.691. At an effusion thickness cutoff of 7.5 mm, sensitivity and specificity in prediction of SA was 71% and 56%, respectively.

Conclusion: Ultrasonography is useful for predicting SA among paediatric patients with acute hip pain and joint effusion. Predominant synovial (capsular) thickening relative to joint effusion thickness at the anterior femoral recess, increasing effusion thickness, high fever, and high serum CRP level are predictive of SA of the hip. Treatment should be prompt to avoid complications.


Authors’ affiliation:
TWY Chin, KS Tse: Department of Radiology and Imaging, Queen Elizabeth Hospital, Jordan, Hong Kong








方法:回顧2008年1月至2016年2月期間所有患有髖關節疼痛及積液並接受超聲引導抽取關節液及接受關節液分析的兒童。確認的SA被定義為革蘭氏染色或關節液細菌培養呈陽性,而假設的SA被定義為血培養呈陽性和高關節液白血球數(>50,000 / mm3)。TS被定義為關節液分析呈陰性。SA和TS 組的臨床、實驗室和醫學影像參數作多變量邏輯回歸分析。

結果:32例中,7例確診或推定有SA,25例確診或推定有TS。與SA有關聯的因素包括體溫≥38.5°C(比值比[OR] = 18,p = 0.02)、血清C-反應蛋白(CRP)水平≥10 mg/l(OR = 7.9,p = 0.03)和超聲發現前股凹處滑膜增厚(OR = 19,p = 0.01)。關節積液厚度增加可以粗略預測SA(OR為每厘米18.6,p = 0.095),接收者操作特徵曲線下面積為0.691。關節積液厚度值為7.5 mm時,預測SA的靈敏度和特異性分別為71%和56%。