Kyphoscoliotic Ehlers-Danlos syndrome due to a novel homozygous PLOD1 variant: severe vascular involvement and molecular diagnosis Authors: Yan Yu, Yimo Zeng, Yingdi Liu, Yaning Liu, Qingxin Shi, Huimin Zhu, Juan Wen, Desheng Liang, Zhuo Li, Lingqian Wu Published: 2026 ABSTRACT AbstractBackgroundKyphoscoliotic Ehlers-Danlos syndrome (kEDS, OMIM: #225400) is a rare subtype of Ehlers-Danlos syndrome characterized by joint hypermobility and spinal deformity, with occasional vascular complications. The rarity of vascular phenotypes in kEDS often leads to low clinical suspicion, contributing to delayed diagnosis and poor outcomes.Case presentationWe report a 13-year-old boy who presented with joint laxity, scoliosis, blue sclerae, and pectus excavatum, and who ultimately succumbed to aortic rupture. Initial whole-exome sequencing (WES) and copy number variation (CNV) analysis failed to identify a causative variant, and genes associated with Marfan syndrome were excluded by WES and multiplex ligation-dependent probe amplification (MLPA). Upon re-analysis of the WES data, however, a homozygous deletion spanning exons 15–16 of PLOD1 was detected and subsequently confirmed by quantitative PCR.ConclusionsThis case demonstrates that small exon deletions in PLOD1 may not be reliably detected by routine WES/CNV pipelines, leading to diagnostic delay, and underscores the underrecognized vascular risk in kEDS. Clinicians should maintain vigilance for vascular complications in patients with kEDS and consider re-analysis strategies to ensure timely diagnosis and appropriate genetic counseling.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12920-026-02391-w.Keywords: Kyphoscoliotic Ehlers-Danlos syndrome, Aortic rupture, Novel homozygous PLOD1 variant, Whole-exome sequencing reanalysis Background Ehlers-Danlos syndrome (EDS) comprises a group of heritable connective tissue disorders (HCTDs) with marked clinical and genetic heterogeneity. The Ehlers-Danlos syndromes have a combined prevalence of at least 1/5000 [1]. The cardinal clinical manifestations include generalized joint hypermobility, skin hyperextensibility, and connective tissue fragility [1]. In 2017, the International Classification of the Ehlers-Danlos Syndromes redefined EDS into 13 molecular subtypes based on causative genes, inheritance patterns, and distinct clinical features, and provided standardized diagnostic criteria [2]. Most EDS subtypes result from pathogenic variants in genes encoding fibrillar collagens (types I, III, and V) or in genes involved in their post-translational modification. Easy bruising, reflecting capillary and perivascular connective tissue fragility, occurs to varying degrees across all subtypes [3]. Because connective tissue plays a critical role in vascular wall integrity, some EDS subtypes may present with vascular involvement such as arterial aneurysms or aortic dissections, which substantially affect survival. Among these, arterial aneurysms and dissections are hallmark features of the vascular subtype (vEDS) [4–6]. By contrast, classical EDS (cEDS), cardiac-valvular EDS (cvEDS), dermatosparaxis EDS (dEDS), and kyphoscoliotic EDS (kEDS) exhibit a much lower incidence of vascular events, although rare cases with arterial complications have been reported [7, 8]. Fatal arterial rupture in non-vEDS subtypes is exceptionally rare and may therefore be underrecognized in clinical practice. Kyphoscoliotic EDS is a rare autosomal recessive subtype of EDS [9]. PLOD1-related kyphoscoliotic EDS is extremely rare, with an estimated incidence of approximately 1 in 100,000 live births [9]. It is caused by pathogenic variants in either PLOD1 (procollagen-lysine, 2-oxoglutarate 5-dioxygenase 1; MIM *153454) or FKBP14 (FKBP prolyl isomerase 14; MIM *614505). Patients harboring homozygous or compound heterozygous variants in PLOD1 are specifically classified as the kEDS-PLOD1 subtype [10]. PLOD1 is located on chromosome 1p36.22, comprises 19 exons, and encodes lysyl hydroxylase 1 (LH1), an enzyme responsible for hydroxylating specific lysine residues during collagen biosynthesis and for subsequent glycosylation of hydroxylysyl residues [11, 12]. Loss of LH1 activity leads to uniform under-hydroxylation of lysine residues and under-glycosylation of hydroxylysine residues in type I procollagen, disrupting normal collagen crosslinking [13]. Clinically, kEDS-PLOD1 often manifests in the perinatal period with congenital hypotonia, generalized joint hypermobility, progressive kyphoscoliosis, and increased scleral fragility with risk of spontaneous ocular rupture. The clinical diagnosis is based on major and minor diagnostic criteria (Table 1) [1], and can be established when three major criteria or two major plus at least three minor criteria are present. However, kEDS shares considerable phenotypic overlap with other EDS subtypes. Moreover, sporadic cases of arterial dissections and ruptures in kEDS have been reported [11, 14–16], resembling features of Marfan syndrome, which complicates early clinical recognition. Table 1Diagnostic criteria for kEDS [1]CategoryClinical featuresMajor criteriaCongenital muscle hypotoniaCongenital or early onset kyphoscoliosis (progressive or non-progressive)GJH with dislocations/subluxations (shoulders, hips, and knees in particular)Minor criteriaSkin hyperextensibilityEasy bruisable skinRupture/aneurysm of a medium-sized arteryOsteopenia/osteoporosisBlue scleraeHernia (umbilical or inguinal)Pectus deformityMarfanoid habitusTalipes equinovarusRefractive errors (myopia, hypermetropia) Diagnostic criteria for kEDS [1] Here, we report a patient with kEDS-PLOD1 harboring a novel homozygous deletion encompassing exons 15–16 of PLOD1. Initial whole-exome sequencing (WES) failed to identify a pathogenic variant, but longitudinal follow-up and progressive clinical manifestations prompted reanalysis of exome data, which led to a definitive diagnosis. The patient experienced a fatal aortic rupture at the age of 13 years, underscoring that life-threatening vascular events, though considered rare, can occur in kEDS-PLOD1 and warrant heightened clinical vigilance. This case highlights the importance of long-term follow-up, timely reanalysis of genomic data, and heightened awareness of vascular complications in patients with rare connective tissue disorders. Methods Clinical data collection Clinical information of the proband was retrospectively collected, including demographic characteristics, medical and family history, physical examination, and results of ancillary investigations. Genetic testing and bioinformatics analysis Peripheral blood samples (5 mL) were collected from the proband and his parents using EDTA anticoagulant tubes. Genomic DNA (gDNA) was extracted from peripheral blood using the KURABO DNA extraction kit (Kurabo Industries, Osaka, Japan) according to the manufacturer’s instructions. Initial genome-wide CNV analysis was performed using high-throughput DNA sequencing-based CNV analysis. For whole-exome sequencing (WES), gDNA was fragmented by sonication, followed by exome capture using the xGen Exome Research Panel (IDT, Coralville, IA, USA). Sequencing was performed on an Illumina NovaSeq 6000 platform (Illumina) with an average coverage depth of 100×. Raw sequencing data were subjected to quality control using FastQC. Reads were aligned to the human reference genome (hg19/GRCh37) using BWA. Variant calling and annotation were performed using the Verita Trekker® analysis system and the Enliven platform. For reanalysis of the existing WES data, copy number variation (CNV) detection was performed using the XHMM pipeline, in which exon-level read depth was normalized and copy number states were inferred using a hidden Markov model. Validation studies Quantitative PCR (qPCR) was performed to validate the copy number status of PLOD1 exons. Primer sets covering all 19 coding exons were designed and validated by Berry Genomics (Beijing, China), with specificity confirmed by in silico BLAST analysis. The primer sequences are listed in Supplementary Table S1. Data analysis and figure generation were also conducted by Berry Genomics. Case presentation Clinical findings The proband was a 7-year-old boy referred to Jiahui Genetic Specialty Hospital of Hunan Province and the Genetic Diagnosis Center of Xiangya Hospital, Central South University, for evaluation of joint hypermobility and scoliosis. He was the second child of parents who reported that they were non-consanguineous (Fig. 1). Both parents were from the same village, with households approximately 10 km apart. The mother’s first child, a girl, also presented with joint hypermobility at birth and died one hour postpartum due to hypoxia (specific details unavailable). Subsequently, the mother delivered twin boys with similar joint manifestations, both of whom died shortly after birth from unknown causes. At birth, the proband exhibited hypermobility of the wrists and metacarpophalangeal joints and scoliosis (Fig. 2C-F). At 7 years of age, his height was 126 cm (above average for age), weight was 19 kg (near the lower percentiles for age), head circumference was 53 cm (within the upper-normal range for age), and arm span was 112 cm. He had a slender build with characteristic craniofacial features, including a broad nasal bridge, blue sclerae, bulbous nose, scattered facial nevi, prominent forehead, triangular face, midface hypoplasia, flat nasal root, mildly prominent chin, epicanthal folds, and protruding ears with normal positioning and aligned dentition. Pedigree of the family. The arrow indicates the proband Clinical phenotype of the patient at 7 years of age. A, B Atrophic scarring observed on the back (A) and lower extremities (B). C, D Joint hypermobility of the hands. E Pectus excavatum. F Standing whole-spine radiograph showing S-shaped scoliosis of the spine Physical examination revealed pectus excavatum and progressive scoliosis with right thoracic and left lumbar curvature deviating 4–5 cm from the midline. Marked hypermobility was noted in the fingers (passive extension up to 90°) and wrists (flexion up to 125°). A formal Beighton score was not documented in the archived clinical records and could not be reliably reconstructed retrospectively. He exhibited mild generalized hypotonia with reduced muscle bulk but preserved muscle strength. Multiple atrophic scars were present on the limbs (Fig. 2A-B). Abdominal ultrasound showed no abnormalities, and according to the available clinical record, echocardiography revealed no structural or functional cardiac defects. Cognitive development was normal. The external genitalia were normal; however, the patient had undergone orchiopexy for cryptorchidism. During follow-up, he developed ocular fragility and experienced spontaneous globe rupture after minor trauma. At 13 years of age, he suffered a fatal aortic rupture. Due to the retrospective nature of this case and the patient’s death, some original imaging materials and detailed quantitative measurements were unavailable for review, including the Cobb angle, the sinus of Valsalva Z-score, specific aortic diameter measurements, echocardiographic images, and ophthalmologic image records. Genetic findings Initial whole-exome sequencing (WES) did not identify any candidate pathogenic variants, and CNV-seq detected no aneuploidies or pathogenic CNVs > 100 kb (Supplementary Figure S1). Following the occurrence of vascular complications during follow-up, reanalysis of WES data identified a homozygous deletion spanning exons 15–16 of PLOD1. Quantitative PCR confirmed the proband’s homozygous deletion and demonstrated that both parents were heterozygous carriers. DNA from the deceased twin brothers revealed the same homozygous deletion (Fig. 3). qPCR-based copy number analysis. A Proband: a homozygous deletion of exons 15 and 16 in the PLOD1 gene was detected (red bars, copy number close to zero). B Deceased sibling: the same homozygous deletion of exons 15 and 16 was identified. C Father: heterozygous deletion of exons 15 and 16. D Mother: heterozygous deletion of exons 15 and 16 Variant interpretation Based on the ACMG/AMP guidelines, the PLOD1 exon 15–16 deletion variant fulfilled the criteria PVS1_Strong, PM2_Supporting, and PP1, and was therefore classified as likely pathogenic. The clinical presentation aligned with the characteristics of kEDS, leading to a final diagnosis of kEDS. Discussion In this study, we report a rare case of kEDS-PLOD1 caused by a homozygous deletion encompassing exons 15–16 of the PLOD1 gene. The proband presented with pronounced generalized joint hypermobility, scoliosis, cryptorchidism, hypotonia, pectus excavatum, craniofacial dysmorphism, and atrophic scarring of the extremities. Based on the clinical features of the proband and the overlapping manifestations observed in his sister and twin brothers, an initial diagnosis of Ehlers-Danlos syndrome (EDS) was considered. However, WES and CNV-seq failed to identify any pathogenic variants. At the age of 13 years, the proband died from massive hemorrhage due to an aortic rupture, indicative of severe vascular fragility. Given that aortic dilation and dissection are hallmark complications of Marfan syndrome [8], MLPA was subsequently performed to assess CNVs in the Marfan syndrome-related genes FBN1 and TGFBR2, but no pathogenic deletions or duplications were detected. These negative findings prompted further consideration of pathogenic variants that might have escaped detection by conventional methods. Considering the inherent limitations of conventional methods in detecting small- to medium-sized CNVs, it was suspected that the proband might harbor pathogenic variants that had escaped prior detection. Specifically, WES, particularly routine short-read analysis pipelines, may have limited sensitivity for detecting small intragenic exon-level deletions [17, 18]. CNV-seq is useful for genome-wide CNV screening, but its resolution is insufficient for detecting small intragenic exon-level deletions [19]. SNP microarray is another useful alternative for genome-wide CNV screening, but its sensitivity for small exon-level deletions is also limited by probe density and resolution. MLPA, although highly sensitive for targeted exon-level deletions/duplications, is probe-dependent and lacks genome-wide coverage. To overcome these challenges and improve the detection of exon-level CNVs, CNV reanalysis was performed based on the existing WES data using the eXome-Hidden Markov Model (XHMM) algorithm. XHMM leverages read depth information from WES data, removes systematic biases via principal component analysis, and applies a hidden Markov model to identify statistically significant CNV events, making it particularly suitable for detecting exon-level deletions/duplications [20]. Re-analysis identified a homozygous deletion of PLOD1 exons 15–16 in both the proband and his twin brothers, while both parents were identified as heterozygous carriers. Based on the proband’s clinical features and the ACMG interpretation of this variant, the proband in this case was definitively diagnosed with kEDS caused by a homozygous deletion of exons 15–16 of the PLOD1 gene. Since the initial report of a pathogenic PLOD1 variant in 1993 [21], numerous variants associated with kEDS have been recorded in ClinVar (https://www.ncbi.nlm.nih.gov/clinvar) and LOVD (https://databases.lovd.nl/shared/genes). The majority of variants annotated as pathogenic are frameshift, nonsense, or canonical splice-site (± 1/2) variants. Reports of large exon deletions are relatively rare. The patient described in this study carries a homozygous deletion of exons 15 and 16 of PLOD1, a specific event not previously reported. To better understand the genetic basis of such severe phenotypes, we performed a literature review of reported kEDS cases with vascular complications (Table 2). As shown in Fig. 4A, PLOD1 variants are distributed throughout the gene, affecting all four major structural domains of lysyl hydroxylase 1 (LH1) [22]. Notably, variants within the prolyl 4-hydroxylase domain were most frequently reported. However, mapping of exons to protein domains revealed no exon-specific clustering, suggesting that the risk of vascular complications is not restricted to specific gene regions. Among reported cases, frameshift variants and multi-exon deletions were the most frequent (36%) (Fig. 4B). Our finding further supports the observation that life-threatening vascular complications can occur across diverse genotypic backgrounds in kEDS-PLOD1. Table 2Review of kEDS with vascular eventsPatient PLOD1 GenotypeMutationTypeSexVascularAbnormalityAge(years)Type of Vascular EventP1[11]c.77_168del/In-frameM13Celiac artery aneurysm with ruptureP2[11]c.77_168delDeletionM10Celiac artery aneurysmP3[12]c.1756-13C>A/p.Glu272Leufs*9Splice site/FrameshiftF29Multiple lower-limb arterial lesions: pseudoaneurysm of right anterior tibial artery with mural thrombosis; dissection of posterior tibial artery; dissection with localized aneurysm and mural hematoma of tibioperoneal trunk; occlusion of posterior tibial artery with dysplastic features; dissection and mural hematoma of left anterior tibial artery with moderate stenosis; saphenofemoral junction reflux on duplexP4[22]p.L85P/p.L85PMissenseM27Spontaneous coronary artery dissection during angiographyP5[22]p.Trp419Leufs*48/p.Trp419Leufs*48FrameshiftF8Hydrocephalus secondary to multiple extensive intracranial and intraventricular hemorrhagesP6[23]p.Arg319XNonsenseF2Neonatal intracranial hemorrhageP7[24]p.Pro622Argfs*3FrameshiftF1.5Neonatal intracranial hemorrhage with periventricular leukomalacia; sequelae consistent with cerebral palsyP8[25]p.Tyr455fsFrameshiftM12Pseudoaneurysms of left brachial and radial arteriesP9[14]F4Mitral valve prolapse; neonatal hypotensionP10[26]c.1098_1328del/c.1471_1755delIn-frameDeletionF41Spontaneous arterial rupture, dissections, and dissecting aneurysmsP11[27]uknuknM10Internal jugular vein ectasia postcatheterismP12[28]uknuknFuknRight iliac artery rupture at deliveryP13[29]uknuknF8Rupture of the superior vena cava after extubationP14[15]uknuknM20Coeliac trunk occlusion after spinal surgery(mechanism undisclosed)P15[16]uknuknM32Profundal femoral artery rupture, femoral false aneurysm(access site), femoral venous rupture, splanchnic arteryrupture (30 years), stroke (12 years)(mechanism undisclosed)P16[30]uknuknM24Traumatic aortic dissection, popliteal artery aneurysm(age undisclosed)P17[31]uknuknuknuknVertebral artery ruptureP18[31]uknuknuknuknMultiple ruptures of the femoral artery and repeatedspontaneous intrathoracic arterial rupturesP19[32]uknuknuknuknAbdominal aorta and iliac artery rupture, common iliac veinrupture during spine surgeryP20[32]uknuknuknuknIntraoperative superior gluteal artery rupture(spine surgery) Review of kEDS with vascular events Overview of reported PLOD1 mutations in kEDS patients with vascular events. A Reported mutations are mapped according to the affected exons of the PLOD1 gene, with each dot representing a unique variant. The red rectangle highlights the exon deletion identified in the present patient, emphasizing its position among the reported variants. The color of the dots corresponds to mutation types indicated in the legend. B Distribution of mutation types among patients with vascular events, showing frameshift variants/in-frame deletions as the most frequent (36%), followed by missense (9%), nonsense (9%), and splice-site mutations (9%) The identification of a homozygous PLOD1 exon 15–16 deletion in this patient, who died of aortic rupture, highlights the need to reconsider vascular involvement in kEDS-PLOD1. While historically regarded as less severe than in vEDS [11], vascular fragility has been documented in animal models and in a subset of patients, suggesting that clinically significant vascular complications may arise in this subtype. Approximately 15% of Plod1−/− mice died between 1 and 4 months of age due to aortic dissections, most likely caused by degenerated smooth muscle cells and abnormal collagen fibrils in the aortic wall [33]. Consistent with these findings, case reports have shown that a subset of kEDS-PLOD1 patients develop major vascular complications [34]. Mechanistically, these complications are thought to result from abnormal type III collagen cross-linking and impaired structural stability of arterial smooth muscle cells caused by PLOD1 mutations [35]. Among the reported kEDS-PLOD1 patients (Table 2), approximately 30% presented with aneurysms in various locations, such as the left brachial artery, coronary artery, celiac artery, and tibiofibular artery. Other manifestations included cerebral hemorrhage (20%) and arterial dissection (15%). Vascular events have also included sudden intraoperative arterial rupture, particularly during orthopedic or vascular procedures, highlighting the risk of catastrophic and unpredictable vascular events. These observations warrant heightened clinical vigilance, especially in perioperative management of kEDS-PLOD1 patients. Notably, this patient died at the age of 13 due to aortic rupture, indicating potential severe vascular involvement. Notably, the early-onset aortic rupture in our patient contrasts with prior reports. For example, Rohrbach et al. described a female with a homozygous PLOD1:c.1651–2 A > G variant causing exon 16 skipping, but no aortic rupture was reported [36]. By contrast, some patients develop vascular events only in adulthood: Foy et al. reported a 29-year-old with compound heterozygous variants who remained free of vascular complications during childhood but later developed multiple lower-limb aneurysms in adulthood [12]; Henneton et al. reported a 41-year-old female with compound exon deletions who experienced arterial rupture and dissection [26]. Even among patients with identical mutations, vascular phenotypes can differ markedly: a 4-year-old female with homozygous c.1362delC showed only mitral valve prolapse [14], whereas a 12-year-old male with the same variant developed a brachial artery pseudoaneurysm [25]. Modifying sequence variants within the entire genome, epigenetic factors, or environmental factors may be responsible for this observation [36]. In contrast, patients sharing both identical mutations and genetic background—such as two brothers reported by Bhandari et al. who both developed celiac artery aneurysms—exhibit consistent vascular phenotypes [11]. Taken together, these reports underscore the variable expressivity of vascular involvement in kEDS-PLOD1. Currently, there are no unified guidelines for vascular screening and management in kEDS-PLOD1 patients. Expert consensus recommends echocardiographic surveillance every 5 years even when initial findings are normal, together with strict blood pressure control to mitigate rupture risk [7]. Clinical management should emphasize multidisciplinary collaboration, involving cardiovascular, genetic, neurological, and other relevant specialties for joint intervention. Regular monitoring of the vascular system using echocardiography, computed tomography angiography (CTA), or magnetic resonance angiography (MRA) is also advised. In summary, we report a kEDS-PLOD1 patient who died from aortic rupture, demonstrating that severe vascular complications can occur even in EDS subtypes traditionally considered low-risk. Enhanced recognition and follow-up management of vascular phenotypes in kEDS-PLOD1 patients are clinically imperative. Furthermore, this case demonstrates a complete diagnostic pathway where a phenotype-driven reanalysis of WES-CNV identified the causative variant following an initial negative WES screen, highlighting the dynamic interplay between clinical phenotype and genetic testing. For patients with strong clinical suspicion but negative routine testing, phenotype-driven reanalysis strategies may improve diagnostic yield in selected cases. Our findings expand the spectrum of PLOD1 variants and alert clinicians to the potential vascular risk in kEDS-PLOD1. Supplementary Information Supplementary Material 1: Supplementary Table S1. Supplementary Material 2: Supplementary Figure S1 Acknowledgements We thank the patient’s family for their cooperation and participation in this study. Authors' contributions YY conceived the study. YY and YZ developed the methodology. YZ performed the clinical investigation and case analysis. YY wrote the original draft. YY, YZ, LW and ZL reviewed and edited the manuscript. LW and ZL supervised the study. All authors have read and agreed to the published version of the manuscript. Funding This work was funded by grants from the National Key Research and Development Program of China (2022YFC2703700, 2024YFC2707100, 2025YFC2708200), the National Natural Science Foundation of China (82171711, 82371724), the Hunan Provincial Natural Science Foundation of China (2023JJ30725). Data availability All data generated or analyzed during this study are included in this published article and its supplementary information files. The raw sequencing data are not publicly available to protect patient privacy but are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate This study was approved by the Ethics Committee of The Third Xiangya Hospital, Central South University, and was conducted in accordance with the Declaration of Helsinki. Written informed consent to participate in this study was obtained from the patient’s parents. Consent for publication Written informed consent for publication of the clinical details and any accompanying images was obtained from the patient’s parents. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Yan Yu, Yimo Zeng and Yingdi Liu contributed equally to this work. Contributor Information Zhuo Li, Email: lizhuo@sklmg.edu.cn. Lingqian Wu, Email: wulingqian@sklmg.edu.cn. References Associated Data Supplementary Materials Supplementary Material 1: Supplementary Table S1. Supplementary Material 2: Supplementary Figure S1 Data Availability Statement All data generated or analyzed during this study are included in this published article and its supplementary information files. The raw sequencing data are not publicly available to protect patient privacy but are available from the corresponding author on reasonable request.