Hemophilia Joint Health Score Poorly Predicts the Need for Musculoskeletal Referrals in Routine Clinical Practice

• Published in: Blood 2016-12, Vol.128 (22), p.3784-3784
• Main Authors: Cattivelli, Kim, Al-Sayegh, Hasan, Boyer, Amanda, Mirek, Elizabeth, Ma, Clement, Neufeld, Ellis J, Croteau, Stacy E.
• Format: Article
• Language: English
• ISSN: 0006-4971; 1528-0020
• DOI: 10.1182/blood.V128.22.3784.3784

Introduction Early detection of joint problems and musculoskeletal (MSK) abnormities is a priority in hemophilia care worldwide; however, there is no standard approach to screening. The World Haemophilia Federation Hemophilia Joint Health Score (HJHS) is well validated but requires time-intensive evaluation by a MSK expert (e.g. physical therapist, PT). Joint screening popularized for the US Center for Disease Control Universal Data Collection (UDC) focuses on range of motion (ROM) assessment only but is quick and easy to implement with modest staff training of non-experts. The approach used at our center, here dubbed UDC-plus, combines ROM measurements with pain and functional evaluation. In 2012 our center found that HJHS was sensitive but non-specific (high false positive rate) and correlated poorly with need for MSK referral or imaging [Morrison et al, Haemophilia, 18 suppl. 3 p.120]. The aim of this study was to evaluate the utility of using HJHS in clinical practice to predict need for MSK referral within 5 years of initial assessment. We also evaluated the ability of UDC-plus to predict actionable MSK findings at 5 years. Methods This is a single center, retrospective, observational study. Patients diagnosed with hemophilia A or B of any severity, a HJHS assessment between 2009 and 2011 and a PT evaluation 5 years later (between 2014-2016) during annual comprehensive visits at Boston Children's Hospital were eligible for the study. HJHS includes evaluation of 6 joints (elbows, knees, ankles) with score range from 0 to 124 points, considered positive if greater than 0. UDC-plus score includes the same joints with score range from 0 to 3, considered positive when 2 or greater. Outcomes included need for MSK-related referral during the 5-year observation period. Clinical and demographic characteristics were summarized using descriptive statistics; logistic regression was used to determine whether baseline joint scores were prognostic of referral. Sensitivity and specificity of joint scores were calculated. Results Seventy-one male patients met inclusion. Median observation time was 5.7 years (range: 3.2-6.8); median baseline age was 10.3y (3-21). Thirty-six (51%) had severe, 8 (11%) moderate and 27 (38%) mild hemophilia; 92% of severe and 25% of moderate patients were on regular prophylaxis. Nine (13%) patients had at least one TJ, reflecting that even for young patients in the prophylaxis era (after 1996 at our center), TJs are not eliminated. An abno