Osteogenesis imperfecta type 2 (OI type II) is the most severe form of osteogenesis imperfecta, a group of genetic disorders that weaken bones. It is almost always fatal during the perinatal period, meaning most affected infants die before, during, or shortly after birth, typically from respiratory failure linked to a severely underdeveloped chest and rib cage. The condition results from defective type I collagen, the main structural protein of bone, and in most cases arises from a new (de novo) dominant mutation rather than being inherited from a visibly affected parent.
What Causes OI Type II at the Genetic Level
The vast majority of OI type II cases trace back to mutations in one of two genes, COL1A1 or COL1A2, which together encode the two chains that twist into the type I collagen triple helix. In a study that sequenced these genes in 63 individuals with perinatal lethal OI, researchers found mutations in 60 of them, with 43 of those mutations never previously reported. About 5% of the cohort instead carried recessive mutations in genes called CRTAP and LEPRE1.1Oxford Academic. Mutation and polymorphism spectrum in osteogenesis imperfecta type II: implications for genotype–phenotype relationships
These recessive forms involve a different mechanism. The proteins encoded by CRTAP and LEPRE1 form a complex that chemically modifies a single amino acid on the collagen chain. When that complex is missing or defective, collagen folds abnormally even though the collagen genes themselves are fine. A screen of 78 individuals with OI type II or III identified 3 with CRTAP mutations and 16 with LEPRE1 mutations, all showing multiple fractures and extremely low bone density.2PubMed Central. CRTAP and LEPRE1 mutations in recessive osteogenesis imperfecta
For the dominant forms, the type of glycine substitution matters. Collagen’s triple helix depends on glycine, the smallest amino acid, appearing at every third position. When a larger amino acid replaces glycine, the helix destabilizes. Laboratory experiments with synthetic collagen-like peptides have shown that the identity of the replacement amino acid directly predicts how much disruption occurs, with bulkier replacements causing greater destabilization and correlating with more severe disease.3PubMed. Destabilization of osteogenesis imperfecta collagen-like model peptides correlates with the identity of the residue replacing glycine The position of the substitution along the collagen chain also matters: a computational model that accounts for the structural environment around each mutation site predicted lethality with roughly 90% accuracy.4PubMed. Predicting the clinical lethality of osteogenesis imperfecta from collagen glycine mutations
How OI Type II Differs from Other Forms
The Sillence classification, first introduced in 1979, divides OI into types based on clinical severity. Type I is the mildest, type III is severely deforming but survivable, and type IV falls somewhere in between. Type II stands apart because it is defined by lethality in the perinatal period. Infants with type II typically have shortened and deformed limbs despite normal birth weight and length, along with a chest so compromised that breathing is inadequate from birth.5PubMed Central. Modern classification and molecular-genetic aspects of osteogenesis imperfecta
Within type II itself, three subtypes have been described. Type IIA produces thick, short, squared-off long bones with a furrowed appearance. Type IIC results in thin, twisted tubular bones. Type IIB overlaps with type III and is characterized by short tubular bones with multiple fractures; infants in this subgroup sometimes survive birth but usually die within the first few years from respiratory failure. One distinguishing feature is the ribs: types IIA and IIB lack the beaded appearance along the ribs that characterizes type IIC.
The boundary between types II and III can be blurry. Infants born with fractures and deformity who survive the perinatal period are reclassified as type III. In practice, the distinction is often made retrospectively: an infant who was expected to have lethal disease but lives past the newborn period ends up categorized differently.
What Happens to the Bones
At a microscopic level, the bones of infants with OI type II look fundamentally immature. Normal bone develops through a sequence where a rough “woven” scaffold is laid down first by one population of bone-forming cells, and then more organized “lamellar” bone is built on top of it. In lethal perinatal OI, this maturation sequence stalls. The bones consist almost entirely of woven bone, the most structurally weak arrangement, because the abnormal collagen cannot provide an adequate scaffold for the transition to stronger lamellar bone.6PubMed Central. Histopathology of osteogenesis imperfecta bone. Supramolecular assessment of cells and matrices in the context of woven and lamellar bone formation using light, polarization and ultrastructural microscopy.
When researchers examined the aggregated structure of collagen-mimicking peptides carrying different glycine substitutions, they found that bulkier replacement amino acids produced more disorganized assemblies with defective shapes, mirroring how different mutations lead to different degrees of clinical severity.7PubMed. Morphology of Osteogenesis Imperfecta Collagen Mimetic Peptides Assemblies Correlates with the Identity of Glycine-Substituting Residue In other words, the problem goes beyond just “weak bones.” The collagen molecules themselves cannot assemble properly, which cascades into abnormal tissue structure from the molecular level up.
Why Respiratory Failure Is the Primary Cause of Death
The immediate cause of death in OI type II is almost always respiratory failure. The rib cage is small, fractured, and poorly mineralized, which means the chest cannot expand enough for adequate breathing. But the lungs themselves may also be directly affected. A systematic review of respiratory problems in OI found evidence that reduced lung function is not solely a mechanical consequence of rib and spine deformities; intrinsic collagen abnormalities in the lung tissue itself may contribute to poor respiratory function as a primary pulmonary problem.8Taylor & Francis Online (Annals of Medicine). Pathophysiology of respiratory failure in patients with osteogenesis imperfecta: a systematic review This is a point that often goes unappreciated: type I collagen is not just in bone. It is a structural component of many tissues, and when it is defective, the effects extend beyond the skeleton.
Prenatal Detection
OI type II can often be identified on routine prenatal ultrasound during the second trimester. The hallmark ultrasound features include widespread poor mineralization of the skeleton, shortened and bowed long bones, and multiple fractures already present in utero.9PubMed Central. Perinatal lethal type II osteogenesis imperfecta: a case report In one reported case, ultrasound evaluation at around 22 weeks revealed bones well below the fifth percentile, fractures of the femur, tibia, fibula, ulna, and ribs, skull deformity under probe pressure, and cranial fractures.10PubMed Central. A prenatal diagnosis of osteogenesis imperfecta in a patient with a novel pathogenic variant in COL1A2
One challenge for prenatal diagnosis is distinguishing OI type II from other skeletal dysplasias, particularly thanatophoric dysplasia, which also produces severely shortened long bones. Radiographic analysis has shown that while both conditions result in significantly shortened bones compared to normal, OI type II tends to produce true angulation or visible displacement at fracture sites, a feature absent in thanatophoric dysplasia.11PubMed. Comparative X-ray morphometry of prenatal osteogenesis imperfecta type 2 and thanatophoric dysplasia: a contribution to prenatal differential diagnosis Genetic testing of fetal DNA, obtained through amniocentesis or chorionic villus sampling, can confirm the diagnosis and identify the specific mutation.
Recurrence Risk and Parental Mosaicism
Because most OI type II cases are caused by new dominant mutations, the standard reassurance given to parents is that the chance of another affected child is very low. This is often wrong, and the reason is mosaicism. A parent can carry the mutation in a fraction of their cells, including their egg or sperm cells, without showing any symptoms themselves. If the mutation is present in a meaningful percentage of their reproductive cells, the risk of having another affected baby can be substantial.
In a study of 37 families who had more than one child with perinatal lethal OI, researchers found collagen gene mutations in 70% of families and recessive mutations in genes like CRTAP, LEPRE1, or PPIB in 19%. Among the families with collagen mutations, the mosaic parent was identified in 15 out of 16 families, split nearly evenly between mothers and fathers.12Genetics in Medicine. Recurrence of perinatal lethal osteogenesis imperfecta in sibships: Parsing the risk between parental mosaicism for dominant mutations and autosomal recessive inheritance In one case, a mutation found in a father’s sperm directly demonstrated that germ-line mosaicism explained why two siblings were both affected despite neither parent having any signs of OI.13PubMed Central. Recurrence of lethal osteogenesis imperfecta due to parental mosaicism for a dominant mutation in a human type I collagen gene (COL1A1)
This has real consequences for genetic counseling. If a family has already had one child with OI type II, a verified finding of mosaicism in one parent significantly raises the recurrence risk above the background “negligible” figure that might otherwise be quoted. Genetic evaluation for mosaicism in apparently healthy parents is therefore an important part of counseling after a lethal OI pregnancy.14PubMed Central. Asymptomatic parental mosaicism for osteogenesis imperfecta associated with a new splice site mutation in COL1A2
Separately, the roughly 5-19% of families whose affected children carry recessive mutations face a straightforward 25% recurrence risk per pregnancy, since both parents are carriers. Distinguishing between these two genetic scenarios, dominant with mosaicism versus recessive, directly determines the risk figure families receive and the options they are offered for future pregnancies.
How Diagnosis Has Evolved
For decades, OI type II was believed to follow autosomal recessive inheritance, which would mean both parents had to carry a defective copy for the disease to appear. This assumption was upended when biochemical studies of collagen from affected infants’ skin cells revealed two populations of collagen molecules, one normal and one abnormal, a pattern consistent with a heterozygous dominant mutation rather than a recessive one.15PubMed. Studies of type I collagen in osteogenesis imperfecta This finding fundamentally changed genetic counseling. What had been categorized as a recessive disease with a 25% recurrence risk was reframed as a dominant condition arising from new mutations, with recurrence largely explained by parental mosaicism.
The term “osteogenesis imperfecta” itself dates to the 1840s, when Willem Vrolik, a professor of anatomy in Amsterdam, described a newborn with numerous fractures and used the Latin phrase in his published anatomical plates. A re-examination of Vrolik’s original specimen in 1998 confirmed it as OI type II.16PubMed. Aspects of the history of osteogenesis imperfecta (Vrolik’s syndrome) So the very first described case of the condition that would eventually lend its name to the entire group of OI disorders was, in fact, the lethal type.
Experimental Approaches and Emerging Research
Because OI type II is lethal so early, conventional postnatal treatments like bisphosphonates or surgical rodding are not applicable. Research interest has instead focused on prenatal interventions, particularly stem cell transplantation. In a landmark case, fetal mesenchymal stem cells from a donor were transplanted into a fetus diagnosed with severe OI at 32 weeks of gestation. A bone biopsy at 9 months of age showed that donor cells had engrafted and differentiated into bone, making up a small but detectable percentage of cells, and the bone had regularly arranged structural elements.17Transplantation. Fetal Mesenchymal Stem-Cell Engraftment in Bone after In Utero Transplantation in a Patient with Severe Osteogenesis Imperfecta A related case report described a child who did well in terms of bone mineral content and fracture rate following prenatal transplantation of donor mesenchymal stem cells, with long-term persistence of the donor cells at the bone site.18American Journal of Obstetrics & Gynecology. In utero transplantation of fetal mesenchymal stem cells in a case of severe osteogenesis imperfecta
These are encouraging proofs of principle, but they remain isolated cases, not evidence that stem cell therapy can rescue an infant with the truly lethal type II form. What they demonstrate is that transplanted cells can survive and produce bone in a developing fetus, even when the donor and recipient are immunologically mismatched. Fetal mesenchymal stem cells are considered especially promising because of their high capacity to form bone tissue.19PubMed Central. Stem Cell Therapy as a Treatment for Osteogenesis Imperfecta
Research also relies on animal models. At least 20 animal models, including mice, dogs, and fish, have been developed to represent the various OI types and test treatments ranging from drug therapies to mechanical loading. These models remain essential for understanding disease mechanisms and testing interventions that cannot be trialed directly in affected human pregnancies.
When a Diagnosis Is Made After Birth
Not every pregnancy with OI type II is detected prenatally. When an infant is born with multiple fractures, severe limb shortening, a soft skull, and a tiny chest, the clinical picture is often immediately recognizable to neonatologists. Radiographs show diffuse poor mineralization, crumpled long bones, and beaded or fractured ribs. At that point, the focus of care shifts rapidly.
Multidisciplinary teams involving genetics, orthopedics, endocrinology, and palliative care are often assembled. In one documented case of a late-preterm infant initially diagnosed with type III OI but whose condition continued to worsen, the family and medical team chose to limit aggressive interventions and focus on comfort care. The infant eventually died of respiratory failure in the neonatal intensive care unit. The case report emphasized how much support and education bedside staff needed to manage the emotional difficulty of providing routine care to an infant at constant risk of fractures from normal handling.20PubMed Central. Osteogenesis Imperfecta: Multidisciplinary and Goal-Centered Care.
This experience reflects a broader reality for families facing OI type II. The medical conversation moves quickly from diagnosis to goals-of-care discussions. Some families choose comfort-focused care from the outset; others may want every possible intervention. There is no standard protocol because the condition itself leaves little room for treatment that could change the outcome. What families consistently need, across these different decisions, is accurate information about what is happening and why, time to process it, and a care team willing to follow their lead.
Collagen Beyond Bone
One aspect of OI type II that surprises people unfamiliar with collagen biology is how far the effects reach beyond the skeleton. Type I collagen is the most abundant protein in the human body. It is a major structural component of skin, tendons, ligaments, the whites of the eyes, blood vessel walls, and the organic framework of teeth. In milder forms of OI, this shows up as blue sclerae (the whites of the eyes appearing bluish), thin skin, easy bruising, loose joints, and dental problems. In type II, these features are present but largely academic, because the skeletal and respiratory problems are so immediately overwhelming that the softer-tissue effects never become the clinical focus.
What is less obvious, and what researchers have increasingly highlighted, is that collagen defects in the lungs may contribute independently to the respiratory failure that kills these infants. If the lung tissue itself is structurally compromised, then even a perfectly formed rib cage might not be enough to sustain breathing. This dual hit, a chest wall that cannot expand and lungs that cannot properly function even if they could expand, helps explain why OI type II is so consistently lethal despite the wide range of specific mutations that can cause it.