Validation of the Radiographic Global Impression of Change (RGI-C) score to assess healing of rickets in pediatric X-linked hypophosphatemia (XLH)

Ruth Lim; Randheer Shailam; Rebecca Hulett; Alison Skrinar; Annabel Nixon; Angela Williams; Mark Nixon; Tom D Thacher

doi:10.1016/j.bone.2021.115964

Validation of the Radiographic Global Impression of Change (RGI-C) score to assess healing of rickets in pediatric X-linked hypophosphatemia (XLH)

Bone. 2021 Jul:148:115964. doi: 10.1016/j.bone.2021.115964. Epub 2021 Apr 18.

Authors

Ruth Lim¹, Randheer Shailam², Rebecca Hulett³, Alison Skrinar⁴, Annabel Nixon⁵, Angela Williams⁶, Mark Nixon⁵, Tom D Thacher⁷

Affiliations

¹ Massachusetts General Hospital, Harvard Medical School, MA, USA. Electronic address: rlim@mgh.harvard.edu.
² Massachusetts General Hospital, Harvard Medical School, MA, USA.
³ Washington University School of Medicine, St Louis, MO, USA.
⁴ Ultragenyx Pharmaceutical Inc., Novato, CA, USA.
⁵ Chilli Consultancy, Salisbury, UK.
⁶ Kyowa Kirin International, Marlow, UK.
⁷ Mayo Clinic, Rochester, MN, USA.

PMID: 33878504
DOI: 10.1016/j.bone.2021.115964

Abstract

Background: Rickets is a primary manifestation of pediatric X-linked hypophosphatemia (XLH) - a rare progressive hereditary phosphate-wasting disease. Severity is quantified from radiographs using the Rickets Severity Scale (RSS). The Radiographic Global Impression of Change (RGI-C) is a complementary assessment in which a change score is assigned based on differences in the appearance of rickets on pairs of radiographs compared side by side.

Objective: The current study evaluated the reliability, validity, and sensitivity to change of the RGI-C specifically in pediatric XLH.

Methods: The reliability, validity, and sensitivity to change of the RGI-C were evaluated using data from two studies in pediatric XLH (113 children aged 1-12 years) in which burosumab treatment significantly improved rickets severity. Intra-rater and inter-rater reliability were assessed by three pediatric radiologists.

Results: Intra-rater reliability for RGI-C global score was >90% for agreement within 1 point, with weighted kappa values >0.5, indicating moderate to almost perfect agreement. Inter-rater reliability was also >90% (0.47-0.52 for all reader pairs; moderate agreement). The RGI-C global score showed significant relationships with changes from baseline to week 64 in serum phosphorus (r = -0.397), alkaline phosphatase (-0.611), total RSS (-0.672), standing height (0.268), and patient-reported global functioning (0.306) and comfort/pain functioning (0.409). Based on standardized response means, RGI-C global scores were sensitive to change in RSS, differentiating between those considered improved and greatly improved. Results for validity and sensitivity to change were similar for the RGI-C wrist, knee, and standing long leg scores.

Conclusion: The RGI-C is a reliable, valid, and sensitive measure in pediatric XLH, and complementary to the RSS.

Keywords: Clinical outcome assessment (COA); Metabolic bone disease; Pediatric radiology; Reliability; Sensitivity to change; Treatment.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Alkaline Phosphatase
Antibodies, Monoclonal
Child
Familial Hypophosphatemic Rickets* / diagnostic imaging
Familial Hypophosphatemic Rickets* / drug therapy
Humans
Immunoglobulin G
Reproducibility of Results

Substances

Antibodies, Monoclonal
Immunoglobulin G
Alkaline Phosphatase