A Systematic Review of Ultrasound Imaging As a Tool for Evaluating Hemophilic Arthropathy in Children and Adults

PURPOSE:

To semi-quantitatively assess the evidence on the value of ultrasound (US) for assessment of hemophilic arthropathy (HA) in children and adults. We sought to provide the answer to the following questions: (1) Are currently available US techniques accurate for early diagnosis of pathological findings?  (2)  Can treatment reduce the incidence of US-detectable findings in HA? (3) Do US scores correlate with clinical/radiological constructs in the evaluation of HA?  (4) Are US findings associated with functional status of joints?

METHODS:

Articles were screened using MEDLINE (n= 519), EMBASE (n= 493), and the Cochrane Library (n=24) (1946-2015). Two independent reviewers assessed the reporting quality and the methodological quality of articles by using the Standards for Reporting of Diagnostic Accuracy (STARD) and the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tools respectively. 4 different US scanning protocols for assessment of hemophilic joints were compared based on scanning times and anatomical landmarks. 6 US scoring systems were also compared according to number of soft-tissue and osteochondral parameters evaluated.

RESULTS:

Out of 16 full-text articles, 9 diagnostic accuracy studies (417 patients with hemophilia A, B and von Willebrand’s disease) were evaluated for reporting and methodological quality using STARD and QUADAS-2 assessment tools respectively. Seven studies were of moderate reporting quality and 2 of low reporting quality. When using QUADAS-2, 1 study was of high, 3 of moderate, 2 of low, and 3 of very low methodological quality. Out of 9 diagnostic accuracy studies, 1 evaluated HA in ankles, knees, elbows, and shoulders while 3 evaluated ankles, knees, and elbows and only 2 evaluated ankles and knees. 2 more studies focused solely on knees and 1 on shoulders.

Six US interpretation scores were reviewed and compared. All 6 articles included synovial hypertrophy in their evaluation. 5 articles incorporated cartilage modification while only 2 articles assessed hemosiderin deposition in their evaluation. Among these 6 scores, 4 were of moderate reporting quality, 1 of low and 1 of very low reporting quality. When using QUADAS-2 however, only 1 study was of high, 1 of moderate, 2 of of low, and 2 of very low methodological quality.  

Four US scanning protocols were also assessed, all evaluated the joints in both prone and supine positions. 3 suggested an extended scanning procedure of up to 30 minutes per joint, while 1 proposed a more simplified procedure. Two scanning protocols evaluated knees and ankles, while one focused on elbows. Only one scanning protocol included ankles, knees, and elbows in its assessment.

CONCLUSIONS:

There is insufficient evidence (Grade I) to recommend US as an accurate technique for early diagnosis of pathologic findings, to demonstrate that US scores correlate with clinical/radiological constructs, that treatment can reduce the incidence of US-detectable findings in HA, and to prove an association between US findings and the functional status of the joint. Further studies are required to establish standardized US scanning protocols and scoring systems and to determine US as a valuable tool for early diagnosis of hemophilic arthropathy (HA) in children and adults.

ACKNOWLEDGEMENTS:

This work was funded by Novo Nordisk Health Care AG.