Comment on “Characterization of individuals with skeletal dysplasia at a referral center in Brazil”

Dear Editor,

We read with great interest the article by Meira et al., “Characterization of individuals with skeletal dysplasia at a referral center in Brazil” (Coelho et al. 2022). The authors deserve credit for addressing an important gap in regional data on hereditary skeletal dysplasias in Bahia, Brazil. Their observational, descriptive, cross-sectional study included 90 individuals with suspected skeletal dysplasia but no confirmed aetiological diagnosis, evaluated between December 2020 and May 2023, with clinical data extracted from medical records and PhenoTips and standardised using Human Phenotype Ontology (HPO) terms (Coelho et al. 2022). The study offers a valuable foundation, but a few points merit clarification before its findings are generalised. The first concerns the genomic framing of the cohort. Participants were recruited through the Rare Genomes Project with whole-genome sequencing (WGS) planned, yet the sequencing results remained under analysis at the time of publication, and no diagnostic yield, pathogenic variants, segregation data, or genotype–phenotype correlations were reported (Coelho et al. 2022). This matters because contemporary skeletal dysplasia classification increasingly links phenotype to molecular aetiology, and a confirmed genetic diagnosis can change recurrence-risk counselling, prognosis, surveillance, and treatment planning (Gargano et al. 2024; Meira et al. 2026; Priego Zurita et al. 2024). Statements suggesting that many cases were “indicative of osteogenesis imperfecta,” or that the cohort demonstrates a strong genetic influence, should therefore be read as provisional until the WGS findings are available. Future analyses combining diagnostic yield, variant interpretation, and genotype–phenotype correlations would strengthen these conclusions considerably. A second issue concerns the HPO data, which would benefit from clearer denominators. The manuscript reports 299 distinct HPO terms, including 98 terms cited 251 times in the “decreased bone density” subgroup and 288 terms cited 570 times across the remaining subgroups (Coelho et al. 2022). HPO is a powerful tool for standardised phenotyping and cohort analytics (Scocchia et al. 2021), but term-citation frequency is not the same as patient-level prevalence: patients with more thoroughly documented phenotypes will naturally contribute more terms, distorting the apparent frequency of fractures, pain, short stature, hearing impairment, ophthalmological abnormalities, or developmental delay. Reporting both term-level frequencies and patient-level denominators, alongside a transparent grouping strategy and an account of missing data, would make these findings more directly usable in clinical practice. Finally, the single-centre convenience sample, together with the deliberate exclusion of clinically obvious diagnoses such as achondroplasia and lethal forms, limits how far these results can be generalised (Coelho et al. 2022). They should not be read as representing the true epidemiology or subtype distribution of skeletal dysplasias in Bahia or Brazil. The geographic categories in the table and text also need sharper definition, since their current form complicates conclusions about rural access and underreporting. Recent rare-bone registry initiatives point the way forward, emphasising multicentre participation, longitudinal data capture, and standardised datasets (Sillence 2024; Unger et al. 2023). A Brazilian multicentre registry covering diagnosed, undiagnosed, prenatal, lethal, paediatric, and adult cases, with mutually exclusive geographic categories and clear referral denominators, would strengthen future public-health interpretation.

None of this diminishes the authors’ contribution. It points instead to where the field can go next: integrated phenotyping, molecular confirmation, patient-level prevalence reporting, and multicentre recruitment, together improving diagnosis, counselling, and equitable access to care.

Author contributions

AA wrote the main manuscritpt, edit and submit it.

Funding

No fund was received for this project.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.