Recommended Tools for Joint Chronic Graft-Versus-Host Disease: Results From the Chronic Gvhd Consortium

Assessing joint chronic GVHD needs to be accomplished reliably, simply and in a clinically meaningful way. To determine the optimal tool for assessing joint GVHD, we evaluated 2 NIH recommended joint tools (Table), a photographic range of motion (P-ROM) scale, and 7 other NIH recommended tools (Lee symptom scale, 10-point overall symptoms, FACT-G, SF36, Human Activities Profile, walk test and grip test).

Patients ≥ age 2 with systemically treated chronic GVHD ≤ 3 years after hematopoietic cell transplantation were eligible for a prospective multicenter observational study. Incident and prevalent cases were included. At follow-up visits every 3–6 months, the clinician (MD) and patient (PT) rated separately their perception of change in joint GVHD on an 8-pt scale, which was collapsed into improved, stable or worse categories. Linear mixed models were used to correlate change in each tool with MD or PT-perceived change (improved vs. stable or worse vs. stable) in joint GVHD status.

Nine sites in the Consortium enrolled 567 participants through December 2011. Joint involvement, as defined by NIH joint/fascia score ≥1, was present at enrollment in 164 (29%) patients and included wrists (64%), ankles (47%), shoulders (35%) and elbows (30%). Joint involvement at enrollment was associated with longer duration of chronic GVHD, high-dose total body irradiation, higher symptom burden, lower quality of life (QOL), similar activity profile and similar physical function, compared to those without joint involvement. Change in joint GVHD status was examined for 652 paired visits when joint involvement was documented in the previous or current visit. In the later visits, both MDs and PTs more often reported improvement (44% and 45%) than worsening (5% and 11%). Tools that correlated with both MD and PT-perceived joint improvement were NIH joint/fascia score, Hopkins fascia score and SF36-PCS. Tools that correlated with both MD and PT-perceived joint worsening were P-ROM total score, NIH joint/fascia score, Hopkins fascia score, Lee muscle/joint subscale, Lee symptom overall score, 10-point overall symptoms and FACT-G. Among the 3 joint/fascia tools (Figure), for MD-perceived improvement, estimated change in the NIH score and P-ROM score was slightly larger than in the Hopkins score. For PT-perceived improvement, estimated change was similar for NIH and Hopkins scores. In contrast, for both MD and PT-perceived worsening, estimated change for the P-ROM score was significantly larger than in the other tools.

Joint involvement with chronic GVHD is frequent and associated with increased symptom burden and decreased QOL. Our results support the combined use of NIH joint/fascia score and P-ROM scale to assess joint GVHD. The NIH score better reflects joint improvement and the P-ROM scale better reflects joint worsening. The more objective P-ROM scale is insensitive to PT-perceived joint improvement possibly because unlike the other two joint assessment tools, it does not incorporate tightness with or without activities of daily living.

Figure.

View largeDownload slide

Estimated change in tools and 95% confidence interval associated with clinician or patient-perceived improvement or worsening (vs. stable). Grey color = association not statistically significant.

Figure.

View largeDownload slide

Estimated change in tools and 95% confidence interval associated with clinician or patient-perceived improvement or worsening (vs. stable). Grey color = association not statistically significant.

Close modal

No relevant conflicts of interest to declare.