Skip to main content
NCBI home page
Diagnostics logoLink to Diagnostics
. 2026 Jul 22;16(14):2293. doi: 10.3390/diagnostics16142293

Diagnostic Yield of Genetic Disorders in Children with Hip Dysplasia Mimicking Bilateral Legg-Calvé-Perthes Disease

Beyhan Tüysüz 1,2,*, Nilay Güneş 2, Timur Yıldırım 3, Hasan Karakaş 4, Büşra Kasap 2, Hilal Onur 2, Dilek Uludağ Alkaya 2, Sezgin Şahin 5, Mehmet Müfit Orak 6, Gazi Zorer 7, Sebuh Kuruğoğlu 8, Özgür Kasapçopur 5
Editor: Ki-Mo Jang
PMCID: PMC13409581  PMID: 42510156

Abstract

Background: Pathogenic variants in genes that cause skeletal dysplasias may, instead of producing classic findings, present in children with a phenotype whose hip radiographs resemble bilateral Legg-Calvé-Perthes disease (LCPD). Objectives: This study aims to investigate the efficacy of genetic diagnosis in children with waddling gait or joint pain and radiological evidence of hip dysplasia mimicking bilateral LCPD. Methods: Forty children with bilateral femoral head dysplasia from 36 families were included in the study. Exome sequencing was performed, and all identified variants were confirmed within the families by Sanger sequencing. Results: Twelve pathogenic or likely pathogenic variants were identified: six in COL2A1, two in COL9A1, and one each in RPL13, EIF2AK3, DNAJC21, and ARSK; six are novel. The diagnostic yield was 33.3% (12/36) in 12 families. Additionally, variants of uncertain significance (VUS), proposed as causative, were detected in five families (5/36:13.9%): two in COL11A1 and one each in COL9A3, COL11A2, and ARSK. Based on bilateral epiphyseal dysplasia of the femoral head, it was observed that seven families may be compatible with mild spondyloepiphyseal dysplasia and six families may have Stickler syndrome. Notably, among these, three children carrying closely localized pathogenic/likely pathogenic variants in COL2A1 shared a novel phenotype characterized by short stature and bilateral irregular femoral heads. In four families, EIF2AK3, DNAJC21, and ARSK were also responsible for the ultra-rare disorders Wolcott-Rallison syndrome, bone marrow failure syndrome 3, and mucopolysaccharidosis 10, respectively. Conclusions: This study, for the first time, investigated the frequency of associated genes in a pediatric cohort with bilateral hip dysplasia resembling LCPD, providing important information for pathogenesis and differential diagnosis.

Keywords: COL2A1, COL9A1, COL9A3, COL11A1, COL11A2, RPL13, EIF2AK3, DNAJC21, ARSK, Legg-Calvé-Perthes disease

1. Introduction

Joint pain and fatigue during walking or running, or a waddling gait, are common reasons for presentation to pediatric outpatient clinics. If changes are detected on hip joint radiographs, the underlying causes may include rheumatic, infectious, or oncological conditions, as well as Legg-Calvé-Perthes disease (LCPD) and genetic disorders [1,2]. LCPD is avascular necrosis of the femoral head epiphysis that typically occurs between the ages of 4 and 8 years, is three to four times more common in boys than in girls, and is bilateral in 10–25% of cases. The radiologically diagnostic appearance is characterized by small, irregular, fragmented femoral head epiphyses [3,4,5,6]. Clinical and radiological findings of bilateral LCPD may overlap with disorders affecting epiphyseal growth, such as avascular necrosis of the head of the femur (ANFH), and skeletal dysplasias including, Beukes hip dysplasia, multiple epiphyseal dysplasia (MED), Stickler syndrome, mild forms of spondyloepiphyseal dysplasia (SED), and mucopolysaccharidosis (MPS) [7,8].

The pathogenesis of LCPD and ANFH is complex; some patients with heterozygous recurrent missense variants (p.Gly1170Ser or p.Gly630Ser) in COL2A1 have been reported [3,4,5,6]. Although MIM (#150600) classifies LCPD as a disease caused by COL2A1 mutations, the role of COL2A1 in LCPD remains controversial [3]. Kannu et al. [9] identified two different mutations (p.Gly672Cys and p.Gly213Asp) in COL2A1 in two children with bilateral LCPD. One of these children had additional findings such as short stature and midface hypoplasia, and the other had cleft palate; they, explicitly used the term “LCPD-like presentation” to distinguish them from idiopathic isolated LCPD. Studies examining the clinical phenotypes of patients with COL2A1 mutations have shown that these patients may exhibit a wide range of clinical phenotypes such as early osteoarthritis of the hip, Stickler syndrome or mild type MED or SED congenita (SEDC) [10,11]. Type II collagen is the major protein of cartilage and, together with types IX and XI collagen, forms large fibrillar structures that are vital for the structure and function of cartilage [12]. The genetic pathogenesis of clinical phenotypes caused by COL2A1, as well as COMP, COL9, COL11, and MATN3 which are involved in the extracellular matrix of cartilage was investigated in a comprehensive study. In this study, the role of variable expression, allelic heterogeneity, and other genetic and epigenetic factors was highlighted, suggesting that individuals carrying pathogenic variants in these genes may present with a phenotype predominantly affecting the proximal femoral heads, resembling LCPD, while showing only mild involvement of other joints and the spine, rather than the classic findings of the defined diseases; these phenotypes were termed MED and related disorders [13].

It is very difficult to clinically determine whether patients with bilateral hip dysplasia have underlying LCPD or other disorders. Isolated idiopathic LCPD and its mimicking phenotypes share features such as a waddling gait, joint pain, and similar radiological hip involvement. However, in other phenotypes there may also be additional skeletal or systemic findings that could be overlooked which require different follow-up and management approaches [3,4,14]. To the best of our knowledge, there are no studies on molecular yield rates in children with this group of disorders using exome sequencing approaches. This study aims to investigate the genetic diagnostic efficacy of exome sequencing in a well-defined cohort of children with hip and/or leg pain or waddling gait and radiological evidence of bilateral hip joint involvement mimicking LCPD.

2. Materials and Methods

2.1. Patients

The study included 40 children from 36 families who presented with joint pain or a waddling gait and bilateral femoral head epiphyseal involvement on radiographs, which could not be attributed to rheumatological or infectious causes. Patients with metaphyseal and vertebral involvement were excluded, while those with negligible metaphyseal or vertebral endplate irregularities were included. In addition, patients with typical clinical features and pathogenic variants in genes associated with MED (COMP, COL9A1–A3 and MATN3, heterozygous; SLC26A2 and CANT1, homozygous) were excluded from the study group. After exome sequencing analysis, reverse phenotyping was performed in all patients, regardless of the presence of a disease-causing variant, through joint evaluation at a single center by a pediatric orthopedist (TY), pediatric radiologists (SK), and pediatric geneticists (BT).

Physical examination findings, anthropometric measurements (weight, height, head circumference, arm span, sitting height and hand length), and skeletal deformities were recorded at initial and at annual follow-up examinations. The standard deviation score (SDS) for all anthropometric measurements was calculated using a national pediatric calculator based on national standards (https://www.ceddcozum.com). X-ray examinations of the skeleton, magnetic resonance imaging (MRI) of the hip, biochemical blood analysis, urinalysis, eye examination, hearing test, echocardiography and abdominal ultrasonography were performed. MRI findings suggestive of bilateral LCPD included proximal femoral epiphyseal abnormalities, subchondral linear low-signal intensity on T1-weighted images, and the presence of a double-line sign on T2-weighted images. Detailed histories and radiographs were also obtained from affected relatives.

2.2. Molecular Genetic Analysis

After comprehensive clinical characterization, exome sequencing was performed in 36 probands. Exome capture was carried out using the QIAseq Human Exome Kit (QIAGEN, Venlo, The Netherlands), followed by 101-base paired-end sequencing on the Illumina NovaSeq platform (Illumina, San Diego, CA, USA). Data processing included quality control using FastQC v0.12.1, alignment to the reference genome with BWA-MEM v0.7.17, variant calling with GATK v4.2.0.0, and annotation with ANNOVAR (version 2023 Mar 15). The mean exome read depth across the cohort was 129× (median: 118×; range: 74–233×). Between 94% and 98% of targeted regions achieved a minimum coverage depth of 20×. Variant calling focused on single nucleotide variants (SNVs) and small insertions and deletions (indels). Variant filtering was performed across the entire exome without restricting analysis to a predefined disease-specific gene panel. Variants demonstrating concordance with each patient’s clinical phenotype were prioritized for further interpretation. Rare variants were prioritized based on a minor allele frequency of less than 1% in public databases, a minimum quality score of 30, and a read depth of at least 10×. Variant classification was performed according to the American College of Medical Genetics and Genomics (ACMG) guidelines by integrating genetic, computational, and clinical evidence [15]. Population allele frequencies were assessed using in-house and Turkish genome databases (https://tgd.tuseb.gov.tr/en/; accessed on 16 March 2026), as well as gnomAD (v3) [16]. In silico prediction tools included Franklin Genoox (version 93), Varsome (version 13.17.1.0), SIFT v2.4, PolyPhen-2 (version 2.2.3), SpliceAI v1.3.1.0, CADD v1.7, and REVEL v1.3. For missense variants, a CADD Phred score of ≥15 and/or a REVEL score of ≥0.5 was considered supportive evidence for pathogenicity. Selected variants, including pathogenic, likely pathogenic variants and variants of uncertain significance (VUS), were confirmed by Sanger sequencing using an ABI PRISM 3500 genetic analyzer (Applied Biosystems, Foster City, CA, USA). After all variants were confirmed by Sanger sequencing in the proband, and in parental and family studies when available, variant inheritance was determined. Because five patients from five families had VUSs (variants of uncertain clinical significance) with very low population allele frequencies and clinical phenotypes consistent with those caused by related genes, these families were included in the uncertain diagnosis group based on overall clinical and genetic evidence.

3. Results

The study cohort comprised 40 children (31 males, nine females) from 36 unrelated families. The age of disease onset ranged from 1.5 to 12 years (median 4.5 years), and the initial findings in most patients were a waddling gait and hip and/or leg pain. Nine patients had mild to moderate short stature. Detailed clinical findings are presented in Supplementary Tables S1 and S2.

3.1. Molecular Findings

Twelve families had pathogenic or likely pathogenic variants; six of these variants were novel (Table 1). The diagnostic yield for disease-causing variants was 33.3% (12/36; 95% confidence interval (CI):18.6–51.0). Six of the 12 families had one child affected by a monoallelic COL2A1 mutation (MIM#120140). Two different families had a biallelic mutation in COL9A1, one with two affected children. One family had a monoallelic mutation in RPL13 (MIM#113703), while the other three families had biallelic mutations in EIF2AK3 (MIM#604032), ARSK (MIM#610011; affecting two siblings), and DNAJC21 (MIM#617048) (Figure 1).

Table 1.

Molecular findings and associated phenotypes in 17 families.

Family/
Patient Number		 Gene
(Transcript)		 Variant
(Protein Change)		 ACMG Classification Zygosity Reference for
Variants		 MIM Phenotype/Number That May Be Associated with Clinical Features
F1/P1 COL2A1
(NM_001844.5)		 c.1510G>A
(p.Gly504Ser)		 Pathogenic
(PS4, PM1, PP2, PM2, PM5, PP3)		 Monoallelic Xu et al. [17] A mild form of SED congenita?
F2/P2 COL2A1
(NM_001844.5)		 c.1538G>C
(p.Gly513Ala)		 Likely pathogenic (PM1, PP2, PM2, PM5, PP3) Monoallelic This study
F3/P3 COL2A1
(NM_001844.5)		 c.1492G>C
(p.Gly498Arg)		 Likely pathogenic (PM1, PP2, PM2, PM5, PP3) Monoallelic This study
F4/P4 COL2A1
(NM_001844.5)		 c.823C>T
(p.Arg275Cys)		 Pathogenic
(PS4, PP1, PM2, PP3, PP2)		 Monoallelic Markova et al. [18] SED, metatarsal shortening/609162
F5/P5 COL2A1
(NM_001844.5)		 c.719G>T
(p.Gly240Val)		 Likely pathogenic (PM1, PP2, PM2, PM5, PP3) Monoallelic This study SED, Stanescu type/
616583
F6/P6 COL2A1
(NM_001844.5)		 c.2059G>A
(p.Gly687Ser)		 Pathogenic
(PS4, PP1, PM1, PM2, PM5, PP2, PP3)		 Monoallelic Barat Houari
et al. [19]		 Stickler syndrome I/
108300
F7/P7, P8 COL9A1
(NM_001851.6)		 c.1491delT
(p.Pro498Leufs Ter133)		 Likely pathogenic (PVS1, PM2) Biallelic This study Stickler syndrome IV/614134
F8/P9 COL9A1
(NM_001851.6)		 c.151C>T
(p.Gln51Ter )		 Likely pathogenic (PVS1, PM2) Biallelic This study Stickler syndrome IV/614134
F9/P10 COL9A3
(NM_001853.4)		 c.1081_1083del
(p.Gly361del)		 VUS
(PM2,PM4)		 Biallelic This study Stickler syndrome VI/
620022
F10/P11 COL11A1
(NM_080629.3))		 c.2644C>T
(p.Arg882Trp)		 VUS
(PM2, PP3)		 Monoallelic This study Stickler syndrome II/604841
F11/P12 COL11A1
(NM_001854.4)		 c.2768C>T
(p.Pro923Leu)		 VUS
(PM2,PP3)		 Monoallelic This study Stickler syndrome II/604841
F12/P13 COL11A2
(NM_080680.3)		 c.4312G>T
(p.Gly1438Cys)		 VUS
(PM2, PP3)		 Biallelic This study Mild form SED or developmental hip dysplasia?
F13/P14 RPL13
(NM_000977.4)		 c.477+1G>A
-		 Pathogenic
(PVS1, PM2, PS4)		 Monoallelic Reinsch et al. [20] SED, Isidor-Toutain type/618728
F14/P15 EIF2AK3
(NM_004836.7)		 c.1126G>T
(p.Glu376Ter)		 Likely pathogenic (PVS1, PM2) Biallelic This study Early-onset DM and MED (Wolcott-Rallison syndrome)/226980
F15/P16 DNAJC21
(NM_001012339.3)		 c.983+1G>A
-		 Pathogenic
(PVS1, PM2, PM3)		 Biallelic D’Amours
et al. [21]		 Bone marrow failure syndrome 3/617052
F16/P17 ARSK
(NM_198150.3)		 c.1254G>T
(p.Met418Ile)		 VUS
(PM2,PP3)		 Biallelic Alkaya et al.
[22]		 MPS10/619698
F17/P18, P19 ARSK
(NM_198150.3)		 c.427G>T
(p.Glu143Ter)		 Likely pathogenic (PVS1, PM2) Biallelic Alkaya et al.
[22]		 MPS10/619698
Open in a new tab

Abbreviations: F family, P patient, DM Diabetes mellitus, MED multiple epiphyseal dysplasia, MIM Mendelian inherited man, MPS mucopolysaccharidosis, SED spondyloepiphyseal dysplasia, VUS variants of uncertain significance.

Figure 1.

Figure 1

Open in a new tab

The pie chart shows the frequency of pathogenic/likely pathogenic (P/LP) variants and associated genes in 12/36 families. The diagnostic yield is 33.3%. The proportion of families with genes proposed as variants of uncertain significance (VUS) was 13.9% (5/36).

In addition, VUSs were identified in five families (5/36; 13.9%). These VUSs were monoallelic state in COL11A1 (MIM#120280) in two families, and biallelic in COL9A3 (MIM#120270), COL11A2 (MIM#120290) and ARSK in three families (Figure 1). In these families, the variants were considered proposed causal VUSs, resulting in an uncertain diagnosis because of their segregation within the family, their very low population frequencies, and clinical presentations consistent with these genes.

In 21 children from 19 families (52.8%), no pathogenic variant was found.

3.2. Clinical and Radiologic Features

3.2.1. Patients with Pathogenic/Likely Pathogenic Variants and VUSs

Clinical and radiological findings were examined in 14 children from 12 families with pathogenic/likely pathogenic variants, along with five children with VUS (Table 1 and Table S1).

Three patients (P1–P3) with monoallelic pathogenic/likely pathogenic variants (c.1492G>C, c.1510G>A, c.1538G>C) located in close proximity in COL2A1 exhibited similar features, including short stature, severe lumbar lordosis, waddling gait, and joint pain (Table S1, Figure 2A,D,G). Their radiographs showed irregularities or destruction of the femoral heads, while the vertebral bodies were ovoid or slightly irregular (Figure 2B,C,E,F,H,I). P4, who had the recurrent c.823C>T variant in COL2A1 associated with SED and metatarsal shortening, had normal height, joint pain, and ulnar deviation of the fingers (Figure 2J,K), as well as irregular vertebral endplates, femoral heads, and acetabulum (Figure 2L,M). P5, with a likely pathogenic missense variant in COL2A1, presented with finger stiffness, a waddling gait, leg pain, flattened proximal femoral epiphyses, and a broad femoral neck (Figure 2N,O). P6 and his affected father, with a known missense variant in COL2A1, had a waddling gait, short stature, severe myopia, arachnodactyly, and irregular proximal femoral epiphyses (Figure 2P–R).

Figure 2.

Figure 2

Open in a new tab

Patients with monoallelic disease-causing variants in COL2A1. P1 at age 4.5 (A–C), P2 at 7 (D–F), and P3 at 3.5 years of age (G–I). Note marked lumbar lordosis (A,D,G), mildly ovoid vertebrae (B,H), irregular vertebral endplates (E), and small, fragmented, irregular, or destroyed femoral heads with slightly broad and irregular proximal femoral metaphyses (C,F,I). P4, with SED and metatarsal shortening, at age 16: ulnar deviation of the second finger (J), normal metacarpal and phalangeal bones (K), slightly elongated vertebrae with irregular endplates (L), and irregularities of the femoral head and acetabulum with narrowing of the joint space and coxa valga (M). P5, with SED Stanescu type, at age 7, had mild platyspondyly (N), an irregular acetabulum, a slightly flattened femoral head, an irregular proximal femoral metaphysis, a broad femoral neck, and coxa valga (O) on radiographs. P6, with Stickler syndrome type I, at age 17, had long metacarpal bones (P), flat and irregular epiphyses of the femoral head (Q), and rectangular vertebral bodies (R).

Biallelic likely pathogenic frameshift and nonsense variants in COL9A1 were detected in two siblings (P7, P8), and in P9, respectively. A biallelic VUS (in-frame deletion) in COL9A3 was identified in P10 (Table 1). These patients had hip pain or a waddling gait, increased lumbar lordosis, and joint hypermobility, as well as mild scoliosis, myopia, and deafness compatible with Stickler syndrome types IV and VI (Table S1). Radiological features included an irregular acetabulum, flat and irregular femoral heads, and short femoral necks with coxa valga, observed in P7 and P9 at ages 9 and 12, respectively (Figure 3A,C). Radiographs taken at age 18 during follow-up of P7 revealed narrowed joint spaces, a slightly smaller femoral head, a shortened femoral neck, and a more superiorly positioned greater trochanter (Figure 3B). In P10, involvement of the femoral head epiphysis and acetabulum was more pronounced (Figure 3D).

Figure 3.

Figure 3

Open in a new tab

Radiographs of patients with biallelic variants in COL9A1 (P7, P9) and COL9A3 (P10) with Stickler types IV and VI are shown in the first line. P7 at 9 years (A) and P9 at 12 years (C) display bilateral, slightly irregular acetabula, flattened and slightly irregular proximal femoral epiphyses, and broad, shortened femoral necks with coxa valga. P7 at age 18 (B) shows a narrowed joint space, a slightly small femoral head and a shortened femoral neck with a high greater trochanter. In P10, at age 9 (D), an irregular acetabulum and a hypoplastic and irregular femoral head epiphysis, more pronounced on the right side and coxa vara, are noted. In the second line, pelvic radiographs of P11 and P12, who have a monoallelic variant in COL11A1 are shown. Note the fragmentation of the proximal femoral epiphysis in 7-year-old P11 (E), and hypoplasia of the medial part of the proximal femoral epiphysis in 9-year-old P12 (F). P13, who has a biallelic variant in COL11A2, is shown in pelvic radiographs at ages 3 and 8. Note the oblique acetabulum and small, hypoplastic femoral head (G), as well as the coxa vara and straight femoral head (H).

Two monoallelic missense VUSs in COL11A1 were identified in P11 and P12. Both children had a waddling gait and either bilateral fragmentation (Figure 3E) or hypoplasia of the medial part (Figure 3F) of the proximal femoral epiphysis, but no visual or hearing problems and midface hypoplasia. P13, with a biallelic missense VUS in COL11A2, had underdeveloped femoral epiphyses at the age of 3 (Figure 3G); by 8 years, flattened and irregular femoral heads had developed (Figure 3H).

A heterozygous mutation of RPL13 was detected in P14, who presented with a waddling gait and limb and hip pain at the age of 8 years; radiographs showed severe destruction of the epiphysis of the femoral head and a short, wide femoral neck (Figure 4A,B), while the vertebral bodies and metaphyses of the lower limb were normal (Figure 4C,D). A biallelic likely pathogenic variant in EIF2AK3, which causes early-onset diabetes mellitus (DM) and MED (MIM#226980), was detected in a girl (P15). This patient, who had bilateral proximal femoral epiphyseal hypoplasia, had neonatal DM in her history (Figure 4E,F). P16 had delayed bone age with an LCPD-like appearance of the right femoral head and a small, flattened femoral epiphysis on the left side (Figure 4G,H). In this patient, a biallelic mutation in DNAJC21 leading to the very rare disease “bone marrow failure syndrome 3” (MIM#617052) was identified, and bone marrow failure developed during follow-up. Biallelic ARSK variants were found in three patients (P17-P19) from two families with bilateral epiphyseal irregularities in the femoral heads and metaphyseal striations. Three children with these gene variants leading to the very rare MPS type 10 (MIM#619698), did not have joint restriction, organomegaly, or radiologically specific dysostosis multiplex that would suggest MPS. Detailed clinical findings for these patients have been previously published by our group [22].

Figure 4.

Figure 4

Open in a new tab

P14, with a monoallelic mutation in RPL13, was examined at ages 10 (A,C,D) and 17 (B) years. Note the slanted acetabulum and flat femoral head with a short femoral neck (A,B), while the metaphyses of the lower extremities and the vertebral bodies are normal (C,D). P15, with a biallelic EIF2AK3 variant at age 7.5 years showed hypoplasia of the medial part of the capital femoral epiphysis, a horizontal acetabulum, coxa valga, and normal vertebral bodies (E,F). P16, with a biallelic DNAJC21 mutation at age 10 showed an LCPD-like appearance of the right femoral head and a small, flattened femoral epiphysis on the left, with mild bilateral metaphyseal irregularity (G) and delayed bone age (H).

3.2.2. Patients Without Variants Related to Hip Dysplasia Genes or Other Disorders

Of the 21 patients from 19 families without variants causing the disease, 18 were male, and the mean age at disease onset was 6 years (Table S2). Deep phenotyping of clinical findings, hip radiographs, and MRI scans was performed, revealing features associated with hip dysplasia ranging from mild to severe, inconsistent with a specific skeletal dysplasia phenotype (Figure 5A–G). The remaining nine patients showed a phenotype consistent with bilateral LCPD (Figure 5H–K), one with ANFH (Figure 5L), and three with features of Stickler syndrome.

Figure 5.

Figure 5

Open in a new tab

Radiographs of 12 patients without disease-causing variants. Patients with radiological features inconsistent with a specific skeletal dysplasia (A–G): severe irregularities of the proximal femoral epiphysis in P20 at 6 years of age (A); destruction of the femoral head in P21 at 16 years of age (B); flattening and irregularity of the bilateral femoral epiphyses in P22 at 9 years of age (C); a deformed femoral head and short femoral neck in P26 at 5 years of age (D); an irregular acetabulum and proximal femoral epiphyses in P33 at 11 years of age (E); a very small, irregular femoral head in P34 at 6 years of age (F); and irregular proximal femoral epiphyses in P38 at 8 years of age (G). Patients whose radiological features are consistent with LCPD (H–L): P23 at 3.5 years of age (H), P24 at 3 years of age (I), P32 at 4 years of age (J), and P36 at 4 years of age (K). Note the bilateral fragmented, small, irregular, or destroyed proximal femoral epiphyses. P28 at 15 years of age, with an ANFH phenotype, had irregular, flat proximal femoral epiphyses (L).

4. Discussion

In this study, we performed exome sequencing on a cohort of pediatric patients with bilateral femoral head epiphyseal dysplasia resembling LCPD who did not have a specific clinical diagnosis. Disease-causing variants were detected in 33.3% of the families. In five families, VUS with a high likelihood of causing the disease were identified and proposed as causative, and these patients were also included in the discussion. Following molecular diagnosis, deep phenotyping was undertaken through multidisciplinary examination, to determine which genetic disorder the phenotype might be associated with.

The most common disease-causing variants were in COL2A1, found in six families. Heterozygous variants in COL2A1 represent a large group of skeletal dysplasias, ranging from lethal forms to isolated osteoarthritis, and are listed as sixteen different phenotypes in MIM. Missense variants were most frequently observed (60%), including glycine substitutions within a Gly-X-Y repeat [19]. The three boys from unrelated families (P1–P3) presented here had missense variants (p.Gly498Arg, p.Gly504Ser, p.Gly513Ala) located close to each other in COL2A1, all involving glycine substitutions. The patients exhibited similar phenotypes, including mild- to moderate short stature (−2.7 to −4.2 SDS) with onset in early childhood, radiologically severe involvement of the femoral head epiphysis, and mild irregularities of the vertebral endplates. The p.Gly504Ser variant identified in P1 has previously been reported in a patient with SEDC [17]. The 2023 revision of the Genetic Skeletal Disorders nosology states that mild cases of SEDC within the group of type 2 collagenopathies may resemble MED, as well as mild chondrodysplastic osteoarthritis (Namaqualand-type hip dysplasia; MIM#604864) and Stanescu-type SED (MIM#616583) [8]. Although the recurrent p.Arg519Cys mutation reported in Namaqualand-type hip dysplasia (MIM#604864) is close to the variants found in our patients, their short stature and early childhood onset were not consistent with this dysplasia. They also differed from Stanescu-type SED by the absence of joint stiffness and the presence of short stature [23]. Our patients did not have vertebral or epiphyseal abnormalities as severe as those seen in SEDC, but all had short stature, which was inconsistent with the MED phenotype. We considered these patients to have a novel phenotype, a mild form of SEDC with similar clinical features and closely localized variants.

Stanescu-type SED (MIM#616583) is a very rare condition characterized by stiffness and pain in the hip and finger joints, a flattened acetabulum, a flattened and irregular femoral head, widening of the proximal femur, coxa valga, and metaphyseal irregularity [23]. We suspected that P5, in whom we found the novel variant (c.719G>T) in COL2A1, might have Stanescu-type SED because of stiffness in the finger joints, a waddling gait, leg pain, and similar pelvic radiological findings. However, a c.719G>A variant at the same protein position has also been reported in Stickler syndrome [24]. The recurrent Arg275Cys variant in COL2A1 is known to cause SED with metatarsal shortening, formerly Czech-type SED (MIM#609162), which is characterized by symmetric spondyloarthropathy with symptoms that develop gradually during adolescence and primarily affect the hip joint [18]. This variant was identified in our patient (P4) who exhibited only severe hip involvement and subtle irregularities in the vertebral endplates on radiographs. There was ulnar deviation of the second finger, but no metacarpal shortening. Interestingly, the same variant has been reported in four patients from the same family, in whom only hip involvement and a short fourth fingers were described [25]. We also identified the pathogenic COL2A1 variant, previously reported in two SEDC patients, in a girl (P6) and her father [19]. However, this patient and her father had severe hip involvement, myopia, and arachnodactyly compatible with Stickler syndrome type I (MIM#108300); the father underwent hip surgery at the age of 30 years. A recently published article stated that patients with genetically confirmed type 1 Stickler syndrome and a previous diagnosis of LCPD had retinal detachment, drawing clinicians’ attention to the possibility that patients with symptoms suggestive of LCPD may have underlying Stickler syndrome [14]. These observations highlight that the same or closely located variants in COL2A1 can cause similar or variable clinical presentations, resulting in cases where hip dysplasia is prominent and other skeletal features are absent, instead of the classic SED findings.

Stickler syndrome is caused by pathogenic variants in monoallelic COL2A1 (80%) and COL11A1 (20%), and less frequently in biallelic COL9A1, COL9A2, and COL9A3. It is characterized by midface hypoplasia, myopia, hearing loss, and mild vertebral and epiphyseal involvement. In particular, the hip joint is affected, and LCPD-like changes leading to early-onset degenerative joint disease are typical [26].

A monoallelic disease-causing variant results in functional impairment of COL9A1–A3, causing MED types 2, 3, and 6, whereas biallelic loss-of-function variants in COL9A1–A3 lead to a complete absence of collagen type IX and cause autosomal recessive Stickler syndrome types IV, V, and VI [13,26]. In this study, we detected novel biallelic nonsense or frameshift variants in COL9A1, resulting in complete absence of type IX collagen in three children from two unrelated families. One child also had a biallelic in-frame deletion VUS in COL9A3. These four patients had bilateral involvement of the proximal femoral epiphysis, as well as vision and hearing loss consistent with Stickler syndrome types IV (MIM#614134) and VI (MIM#620022) [26,27]. Radiographs of patients with COL9A1 variants showed hip dysplasia including an irregular acetabulum, flattened and irregular proximal femoral epiphyses, and a short, wide femoral neck, as previously reported [27]. In the patient with the COL9A3 VUS, the hip radiological findings were more pronounced than in both the patients with COL9A1 presented here and those reported previously [28,29].

Monoallelic pathogenic COL11A1 variants causing Stickler syndrome type II (MIM#604841) are mostly splicing and missense variants. We identified two different heterozygous missense variants, proposed as causative VUS in COL11A1 in P11 and P12 from unrelated families. This syndrome presents with midfacial hypoplasia; 15–20% of patients have no visual or hearing problems, and mild SED findings are present in 25% of cases [26]. In our patients (P11 and P12), the hip joints were affected, but there was no flat midface or visual or hearing problems. A mild form of Stickler syndrome type II caused by COL11A1 mosaicism was reported in the latest nosology of skeletal dysplasias [8]. Biallelic disease-causing variants in COL11A2 cause autosomal recessive deafness (MIM#609706) or otospondylomegaepiphyseal dysplasia (MIM#215150), characterized by short stature, joint contractures and pain, premature osteoarthritis, platyspondyly, and large epiphyses [30]. A boy (P13) in our study with a proposed causative biallelic VUS in COL11A2 had neither hearing loss nor platyspondyly with large epiphyses. He underwent surgery for developmental hip dysplasia in infancy; at age 6, his height was −1.9 SDS and he had only irregular proximal and distal femoral epiphyses. Recent studies on the genetics of hip dysplasia have suggested that the association between COL11A1 and COL11A2 and early-onset osteoarthritis is supported by evidence linking these genes to hip dysplasia [31,32].

Monoallelic pathogenic variants in RPL13, which encodes ribosomal protein L13 expressed in the growth plate, cause spondyloepimetaphyseal dysplasia, Isidor-Toutain type (MIM#618728), characterized by short stature, platyspondyly, and irregularities of the metaphyses and epiphyses of the lower extremities [33]. Interestingly, a patient with a mild phenotype showing only LCPD-like femoral epiphyseal changes and carrying the same pathogenic variant in RPL13 as our patient (P14) has been reported [20].

Our results support the disease mechanisms proposed by Dennis et al. [13]. They emphasized that individuals carrying the same variants in genes involved in the extracellular matrix of bone, such as COMP, COL2, COL9, COL11, and MATN3, may exhibit clinical findings ranging from mild to severe due to reduced penetrance or other factors. We observed that epiphyseal dysplasias in long bones other than the proximal femur, and spinal involvement, which should be present in SED, are so mild that they can be overlooked or may be absent.

We also identified biallelic variants in EIF2AK3, DNAJC21, and ARSK that cause extremely rare syndromes in four families with hip dysplasia, instead of the typical syndromic features. Biallelic mutations in EIF2AK3 cause “Early-onset DM and MED” (MIM#226980; Wolcott-Rallison syndrome), which is clinically variable and in some patients also affects the epiphyses [34]. In a girl (P15) with gait abnormalities and bilateral epiphyseal involvement of the proximal femur on hip radiograph, we detected a biallelic likely pathogenic variant in EIF2AK3; her history stated that she had DM. A biallelic DNAJC21 variant causing “bone marrow failure syndrome 3” (MIM#617052) was detected in our patient, who had an LCPD-like appearance of the right femoral head and a small, flat femoral epiphysis on the left side, and who developed bone marrow failure during follow-up. One patient with the same variant as our patient, and with similar hip dysplasia, has been reported [21]. In three patients with bilateral epiphyseal and acetabular irregularity of the proximal femur, we identified biallelic ARSK variants (one classified as a VUS) causing type X MPS (MIM#619698), an extremely rare form of the disease. These patients, whose clinical findings we previously published [22], did not exhibit prominent signs of dysostosis multiplex; additionally, they had distinctive metaphyseal striations. Based on our experience with this group of extremely rare syndromes, patients presenting with hip involvement must be carefully screened for endocrine, hematological, and lysosomal diseases.

A limitation of this study is that twenty-one patients in our cohort could not be genetically diagnosed and genome sequencing was not performed in this group. Clinical findings in half of these patients were consistent with bilateral LCPD or ANFH. Another limitation of this study is the inability to conduct functional studies to evaluate gene or protein expression that would indicate the pathogenicity of VUSs or clinical phenotype variability.

In conclusion, variants causing the disease were identified in 33.3% of families with bi-lateral femoral head epiphyseal dysplasia without a specific clinical diagnosis. This study demonstrates that disorders resulting from mutations in COL2A1, COL9A1, COL9A3, COL11A1, COL11A2, RPL13, EIF2AK3, DNAJC21, and ARSK can initially present with hip dysplasia, without the classic findings of the diseases associated with these genes. Based on this observation, early genetic diagnosis may be crucial in preventing other life-threatening problems associated with these disorders. Furthermore, our results highlight the role of these genes in the differential diagnosis of patients with bilateral LCPD-like presentation and underscore their importance in the pathogenesis of morphological hip abnormalities.

Acknowledgments

We would like to thank all patients and their families for participating in this study.

Abbreviations

The following abbreviations are used in this manuscript:

ANFH Avascular necrosis of the head of the femur
CI Confidence interval
DM Diabetes mellitus
LCPD Legg-Calvé-Perthes disease
MED Multiple epiphyseal dysplasia
MPS Mucopolysaccharidosis
MRI Magnetic resonance imaging
OSMED Otospondylomegaepiphyseal dysplasia
SDS Standard deviation score
SED Spondyloepiphyseal dysplasia
SEDC Spondyloepiphyseal dysplasia congenita
VUS Variants of uncertain significance
Open in a new tab

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/diagnostics16142293/s1, Table S1: Clinical and radiological features of the patients with found pathogenic and likely pathogenic variants, and variants of uncertain significance; Table S2: Clinical and radiological features of the patients without disease-causing variants or VUS.

diagnostics-16-02293-s001.zip (398.3KB, zip)

Author Contributions

Conception and design of the study: B.T.; Acquisition of data: T.Y., M.M.O., G.Z., H.K., N.G., S.Ş. and Ö.K.; Analysis and interpretation of data: D.U.A., H.O., N.G., B.K., S.K., T.Y. and B.T.; Writing—original draft: B.T., N.G. and T.Y.; Supervision: B.T. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

The study was carried out in accordance with the Declaration of Helsinki of the World Medical Association. The study was approved by the local ethics committee (Istanbul University-Cerrahpaşa, Cerrahpaşa Medical Faculty, (Date: 14 April 2020, No: 53771).

Informed Consent Statement

Informed consent was obtained from all participants or their parents involved in the study. Written informed consent has been obtained from the patients or their parents to publish this paper.

Data Availability Statement

The data presented in this study are available on request from the corresponding author due to privacy and ethical restrictions related to patient data.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding Statement

This study was funded by Istanbul University-Cerrahpaşa, BAPSIS under project number TYO-2020-34927.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

References

  • 1.Bukowiec L.G., Kaji E.S., Koch J.A., Saniei S., Girod Hoffmann M.M., Sinnwell J.P., Wyles C.C. Genetics of morphological hip abnormalities and their implications for osteoarthritis: A scoping review. J. Hip Preserv. Surg. 2025;12:202–216. doi: 10.1093/jhps/hnaf020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Jana M., Nair N., Gupta A.K., Kabra M., Gupta N. Pelvic radiograph in skeletal dysplasias: An approach. Indian J. Radiol. Imaging. 2017;27:187–199. doi: 10.4103/ijri.IJRI_367_16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Spasovski V., Srzentic Dražilov S., Nikcevic G., Bašcarevic Z., Stojiljkovic M., Pavlovic S., Spasovski D. Molecular Biomarkers in Perthes Disease: A Review. Diagnostics. 2023;13:471. doi: 10.3390/diagnostics13030471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Rodríguez-Olivas A.O., Hernández-Zamora E., Reyes-Maldonado E. Legg-Calvé-Perthes disease overview. Orphanet J. Rare Dis. 2022;17:125. doi: 10.1186/s13023-022-02275-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Basit S., Khoshhal K.I. Clinical and genetic characteristics of Legg-Calve-Perthes disease. J. Musculoskelet. Surg. Res. 2022;6:1–8. doi: 10.25259/JMSR_123_2021. [DOI] [Google Scholar]
  • 6.Pavone V., Chisari E., Vescio A., Lizzio C., Sessa G., Testa G. Aetiology of Legg-Calvé-Perthes disease: A systematic review. World J. Orthop. 2019;10:145–165. doi: 10.5312/wjo.v10.i3.145. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Mendelsohn N.J., Wood T., Olson R.A., Temme R., Hale S., Zhang H., Read L., White K.K. Spondyloepiphyseal dysplasias and bilateral legg-calve-perthes disease: Diagnostic considerations for mucopolysaccharidoses. JIMD Rep. 2013;11:125–132. doi: 10.1007/8904_2013_231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Unger S., Ferreira C.R., Mortier G.R., Ali H., Bertola D.B., Calder A., Cohn D.H., Cormier-Daire V., Girisha K.M., Hall C., et al. Nosology of Genetic Skeletal Disorders: 2023 Revision. Am. J. Med. Genet. Part A. 2023;191:1164–1209. doi: 10.1002/ajmg.a.63132. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Kannu P., Irving M., Aftimos S., Savarirayan R. Two novel COL2A1 mutations, associated with a Legg-Calvé-Perthes disease-like presentation. Clin. Orthop. Relat. Res. 2011;469:1785–1790. doi: 10.1007/s11999-011-1850-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Dasa V., Eastwood J.R.B., Podgorski M. Exome sequencing reveals a novel COL2A1 mutation implicated in multiple epiphyseal dysplasia. Am. J. Med. Genet. Part A. 2019;179:534–541. doi: 10.1002/ajmg.a.61049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Terhal P.A., Nievelstein R.J., Verver E.J., Topsakal V., van Dommelen P., Hoornaert K., Le Merrer M., Zankl A., Simon M.E., Smithson S.F., et al. A study of the clinical and radiological features in a cohort of 93 patients with a COL2A1 mutation causing spondyloepiphyseal dysplasia congenita or a related phenotype. Am. J. Med. Genet. Part A. 2015;167:461–475. doi: 10.1002/ajmg.a.36922. [DOI] [PubMed] [Google Scholar]
  • 12.Briggs M.D., Dennis E.P., Dietmar H.F., Pirog K.A. New developments in chondrocyte ER stress and related diseases. F1000Research. 2020;9:290. doi: 10.12688/f1000research.22275.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Dennis E.P., Greenhalgh-Maychell P.L., Briggs M.D. Multiple epiphyseal dysplasia and related disorders: Molecular genetics, disease Mechanisms, and therapeutic Avenues. Dev. Dyn. 2021;250:345–359. doi: 10.1002/dvdy.221. [DOI] [PubMed] [Google Scholar]
  • 14.Wang A., Nixon T., Martin H., Richards A., McNinch A., Alexander P., Pujari R., Bale P., Shenker N., Bearcroft P., et al. Legg-Calve-Perthes’ disease: An opportunity to prevent blindness? Arch. Dis. Child. 2023;108:789–791. doi: 10.1136/archdischild-2022-325059. [DOI] [PubMed] [Google Scholar]
  • 15.Richards S., Aziz N., Bale S., Bick D., Das S., Gastier-Foster J., Grody W.W., Hegde M., Lyon E., Spector E., et al. Standards and guidelines for the interpretation of sequence variants: A joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet. Med. 2015;17:405–424. doi: 10.1038/gim.2015.30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Chen S., Francioli L.C., Goodrich J.K., Collins R.L., Kanai M., Wang Q., Jessica Alföldi J., Watts N.A., Vittal C., Gauthier L.D., et al. A genomic mutational constraint map using variation in 76,156 human genomes. Nature. 2024;625:92–100. doi: 10.1038/s41586-023-06045-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Xu Y., Li L., Wang C., Yue H., Zhang H., Gu J., Hu W., Liu L., Zhang Z. Clinical and molecular characterization and discovery of novel genetic mutations of chinese patients with COL2A1-related dysplasia. Int. J. Biol. Sci. 2020;16:859–868. doi: 10.7150/ijbs.38811. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Markova T., Kenis V., Melchenko E., Osipova D., Nagornova T., Orlova A., Zakharova E., Dadali E., Kutsev S. Clinical and genetic characteristics of COL2A1-associated skeletal dysplasias in 60 Russian patients: Part I. Genes. 2022;13:137. doi: 10.3390/genes13010137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Barat-Houari M., Dumont B., Fabre A., Them F.T., Alembik Y., Alessandri J.L., Amiel J., Audebert S., Baumann-Morel C., Blanchet P., et al. The expanding spectrum of COL2A1 gene variants in 136 patients with a skeletal dysplasia phenotype. Eur. J. Hum. Genet. 2016;24:992–1000. doi: 10.1038/ejhg.2015.250. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Reinsch B., Grand K., Lachman R.S., Kim H.K.W., Sanchez-Lara P.A. Expanding the phenotypic spectrum of RPL13-related skeletal dysplasia. Am. J. Med. Genet. Part A. 2021;185:2776–2781. doi: 10.1002/ajmg.a.61965. [DOI] [PubMed] [Google Scholar]
  • 21.D’Amours G., Lopes F., Gauthier J., Saillour V., Nassif C., Wynn R., Alos N., Leblanc T., Capri Y., Nizard S. Refining the phenotype associated with biallelic DNAJC21 mutations. Clin. Genet. 2018;94:252–258. doi: 10.1111/cge.13370. [DOI] [PubMed] [Google Scholar]
  • 22.Uludag-Alkaya D., Taner H.E., Yildirim T., Akpinar E., Tuysuz B. Further characterization of ARSK-related mucopolysaccharidosis type 10. Am. J. Med. Genet. Part A. 2024;194:e63635. doi: 10.1002/ajmg.a.63635. [DOI] [PubMed] [Google Scholar]
  • 23.Travessa A.M., Díaz-González F., Mirco T., Olivera-Ramos F., Parron-Pajares M., Heath K.E., Sousa A.B. Spondyloepiphyseal dysplasia type Stanescu: Expanding the clinical and molecular spectrum of a very rare type II collagenopathy. Am. J. Med. Genet. Part A. 2020;182:2715–2721. doi: 10.1002/ajmg.a.61817. [DOI] [PubMed] [Google Scholar]
  • 24.Richards J.A., McNinch A., Martin H., Oakhill K., Rai H., Waller S., Treacy B., Whittaker J., Meredith S., Poulson A., et al. Stickler syndrome and the vitreous phenotype: Mutations in COL2A1 and COL11A1. Hum. Mutat. 2010;31:E1461–E1471. doi: 10.1002/humu.21257. [DOI] [PubMed] [Google Scholar]
  • 25.Lankisch P., Honscheid A., Schaper J., Borkhardt A., Laws H.J. COL2A1 mutation as a cause of premature osteoarthritis in a 13-year-old child. Jt. Bone Spine. 2014;81:83–85. doi: 10.1016/j.jbspin.2013.06.007. [DOI] [PubMed] [Google Scholar]
  • 26.Mortier G. Stickler Syndrome. In: Adam M.P., Bick S., Mirzaa G.M., Pagon R.A., Wallace S.E., Amemiya A., editors. GeneReviews® [Internet] University of Washington; Seattle, WA, USA: 2000. [PubMed] [Google Scholar]
  • 27.Nixon T.R.W., Richards A.J., Martin H., Alexander P., Snead M.P. Autosomal recessive Stickler syndrome. Genes. 2022;13:1135. doi: 10.3390/genes13071135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Rad A., Najafi M., Suri F., Abedini S., Lohum S., Karamiani E.G., Dahtarian N., Murhdy D., Doosti M., Moghaddasi A., et al. Identification of three novel homozygous variants in COL9A3 causing autosomal recessive Stickler syndrome. Orphanet J. Rare Dis. 2022;17:97. doi: 10.1186/s13023-022-02244-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Markova T., Sparber P., Borovikov A., Nagornova T., Dadali E. Clinical and genetic characterization of autosomal recessive stickler syndrome caused by novel compound heterozygous mutations in the COL9A3 gene. Mol. Genet. Genom. Med. 2021;9:e1620. doi: 10.1002/mgg3.1620. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Selvam P., Singh S., Jain A., Atwal H., Atwal P.S. Novel COL11A2 pathogenic variants in a child with autosomal recessive otospondylomegaepiphyseal dysplasia: A review of the literature. J. Pediatr. Genet. 2020;9:117–120. doi: 10.1055/s-0039-1698446. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Jacobsen K.K., Lene Bjerke Laborie L.B., Kristiansen H., Schäfer A., Gundersen T., Zayats T., Rosendahl K. Genetics of hip dysplasia—A systematic literature review. BMC Musculoskelet. Disord. 2024;25:762. doi: 10.1186/s12891-024-07795-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Yoshino S., Chen S., Yamaguchi R., Kurakazu T., Hatzikotoulas K., Koike Y., Inoue D., Kohno Y., Sasaki K., Choe H., et al. Genetic study identifies novel genes in developmental dysplasia of the hip. Bone Res. 2026;14:34. doi: 10.1038/s41413-026-00514-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Díaz-González F., Parrón-Pajares M., Lucas-Castro E., Modamio-Høybjør S., Sentchordi-Montané L., Seidel V., Prieto P., Tarraso-Urios G., Codina-Sola M., Cueto-González A.M., et al. Evolution of clinical and radiological presentations of spondyloepimetaphyseal dysplasia, RPL13-related: Description of 11 further cases. Clin. Genet. 2023;104:100–106. doi: 10.1111/cge.14351. [DOI] [PubMed] [Google Scholar]
  • 34.Aldrian D., Bochdansky C., Kavallar A.M., Mayerhofer C., Deeb A., Habeb A., Rabasa A.R., Khadilkar A., Uçar A., Knoppke B. Natural history of Wolcott-Rallison syndrome: A systematic review and follow-up study. Liver Int. 2024;44:811–822. doi: 10.1111/liv.15834. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

diagnostics-16-02293-s001.zip (398.3KB, zip)

Data Availability Statement

The data presented in this study are available on request from the corresponding author due to privacy and ethical restrictions related to patient data.

Articles from Diagnostics are provided here courtesy of Multidisciplinary Digital Publishing Institute (MDPI)

RESOURCES