A compression fracture shows up on a standard X-ray as a vertebra that has lost height compared to its neighbors. Instead of the normal rectangular block shape you’d expect, the fractured vertebral body appears squashed, sometimes uniformly, sometimes more on one side than the other. The change can be subtle or dramatic depending on the severity, and the specific pattern of height loss tells a radiologist quite a bit about what happened and when. But the X-ray appearance is more nuanced than just a “short bone,” and a surprising number of these fractures go unreported even when they’re sitting right there on the film.
The Three Shapes of a Compressed Vertebra
Compression fractures don’t all look the same. Radiologists recognize three main morphology types, each with a distinct silhouette on an X-ray. A wedge-shaped fracture loses more height at the front of the vertebral body than the back, so the bone looks like a doorstop when viewed from the side. This is the most common pattern, especially in the thoracic spine, because the natural forward curve of the upper back concentrates force on the anterior edge. A biconcave fracture, sometimes called a “codfish vertebra,” bows inward at the center of the top or bottom endplate while the front and back walls stay relatively intact, giving the bone an hourglass or fish-mouth look. A crush fracture loses height across the entire vertebral body more or less evenly, so the bone just looks flattened.
1PubMed. Radiomics classification of fresh and old vertebral compression fractures: Impact of compression grade and morphology on diagnostic performanceThese shapes matter beyond classification for its own sake. The morphology reflects how force was distributed through the bone. When the disc above a vertebra has degenerated and stiffened, compressive load concentrates on the rim of the vertebral body, favoring wedge or crush patterns. When the disc is healthier and more hydrated, it presses inward against the center of the endplate, which is structurally the weakest part of the bone, producing the biconcave shape. The condition of the neighboring discs plays a surprisingly large role in which fracture type develops, sometimes rivaling bone density itself as a determining factor.
2SpringerLink. Biomechanics of vertebral compression fractures and clinical applicationHow Severity Gets Graded
Radiologists don’t just note that a vertebra looks compressed; they measure how much height it has lost. The most widely used system looks at three points on the vertebral body in a lateral (side-view) X-ray: the anterior height (front wall), the middle height (center), and the posterior height (back wall). The fractured measurement is compared against either the expected normal height or the height of adjacent healthy vertebrae. A mild fracture involves less than roughly 25 percent height loss, a moderate fracture falls in the range of about 26 to 40 percent, and a severe fracture exceeds 40 percent.
3PubMed Central. Can a Deep-learning Model for the Automated Detection of Vertebral Fractures Approach the Performance Level of Human Subspecialists?In practice, grading on plain X-rays is trickier than it sounds. The image is a flat projection of a three-dimensional structure, so small changes in patient positioning or the angle of the X-ray beam can exaggerate or mask height loss. Mild fractures are the hardest to catch because the height difference can be just a few millimeters. Automated tools have been developed to help, but even computer-assisted grading shows only moderate agreement with expert readings in some evaluations.
4PubMed Central. Vertebral Body Compression Fractures and Bone Density: Automated Detection and Classification on CT ImagesAdditional measurements used in more severe injuries include the degree of vertebral wedging (the angular tilt between the top and bottom endplates) and the percentage of spinal canal narrowing if bone has been pushed backward. These details show up more clearly on CT than on plain film, but they can sometimes be estimated from a good-quality lateral X-ray.
5PubMed Central. Classifying thoracolumbar fractures: role of quantitative imagingFresh Versus Old Fractures on Plain Film
One of the most clinically important questions when a compression fracture appears on an X-ray is whether it happened recently or has been there for months or years. The distinction matters because a fresh fracture is treatable with bracing, pain management, or sometimes a cement-injection procedure, while an old, healed fracture usually requires no acute intervention. Unfortunately, X-rays alone are not great at answering this question. Both fresh and old fractures can look like the same wedge or crush deformity, and the bone doesn’t carry a visible timestamp.
That said, a few radiographic clues tilt the odds. A subendplate cleft, which appears as a thin dark line just beneath the top or bottom surface of the vertebral body, roughly doubled the odds of the fracture being acute in one study comparing X-ray signs to MRI confirmation. Similarly, subendplate density, a hazy white band in the same location representing impacted or healing bone, was also associated with acuteness. On the other hand, features like osteophytes (bone spurs) at the endplate or subendplate cysts didn’t reliably distinguish fresh from old injuries.
6PubMed. Radiographic assessment of acute vs chronic vertebral compression fracturesWhen it truly matters, MRI is the definitive test. A fresh fracture lights up on fluid-sensitive MRI sequences because of edema (swelling) in the bone marrow, while a healed fracture does not. This is one of the main reasons clinicians order an MRI after finding a compression fracture on X-ray, especially if the patient has new or worsening back pain and the fracture’s age is unclear.
7PubMed. A deep-learning model for identifying fresh vertebral compression fractures on digital radiographyWhy So Many Get Missed
Compression fractures are remarkably under-detected on routine imaging studies. A review of multiple studies found that roughly one in five patients in imaging populations had vertebral compression fractures visible on their films, yet only about 27 percent of those fractures were mentioned in the official radiology report.
8PubMed Central. Incidental vertebral compression fractures in imaging studies: Lessons not learned by radiologistsThe reporting gap was especially wide on CT scans, where radiologists noted fractures in only about 8 percent of cases, compared to around 41 percent on dedicated spine X-rays. This sounds counterintuitive since CT produces higher-resolution images, but it makes sense when you consider context: a CT scan of the chest or abdomen is usually ordered for something else entirely, like a lung nodule or abdominal pain, and the radiologist may not be focused on the spine. In one study looking specifically at routine chest X-rays, prevalence of moderate-to-severe vertebral fractures was about 22 percent, and only a little over half of them made it into the official report.
9PubMed. Underreporting of vertebral fractures on routine chest radiographyThis underreporting has real consequences. A missed compression fracture is a missed signal that someone’s bones are fragile enough to break under relatively minor loads, and that information should trigger a conversation about osteoporosis treatment. Without it, the patient goes home none the wiser, and the opportunity for prevention slips away.
Telling a Benign Fracture From a Malignant One
Not every compression fracture comes from osteoporosis or trauma. Cancer that has spread to the spine (metastatic disease) or primary bone tumors like multiple myeloma can weaken a vertebra enough to cause collapse. On X-ray, malignant and benign compression fractures can look disturbingly similar, which is why distinguishing between them is one of the more anxiety-producing tasks in spine imaging.
Certain X-ray features nudge the diagnosis one way or the other. A benign osteoporotic fracture tends to preserve the posterior wall of the vertebral body, shows no soft-tissue mass around the spine, and may involve multiple vertebrae in a pattern consistent with generalized bone loss. A malignant fracture is more likely to destroy a pedicle (the bony bridge connecting the vertebral body to the back of the spine), show convex bulging of the posterior wall into the spinal canal, or have an associated soft-tissue mass. But there is genuine overlap in appearances, and imaging features alone cannot always make the call.
10PubMed Central. Review of the Imaging Features of Benign Osteoporotic and Malignant Vertebral Compression FracturesMRI is far more useful than plain X-ray for this distinction. Advanced MRI techniques can measure the fat content within the fractured bone, and this turns out to be a powerful differentiator. Malignant fractures replace the normal fatty marrow with tumor cells, so their fat fraction drops dramatically compared to osteoporotic fractures. In one study, the measured fat fraction in malignant fractures averaged under 3 percent, versus about 14 percent in acute osteoporotic fractures, with extremely high diagnostic accuracy.
11PubMed. Differentiation of Acute Osteoporotic and Malignant Vertebral Fractures by Quantification of Fat Fraction With a Dixon MRI SequenceThe Vacuum Cleft Sign and Kümmell’s Disease
One distinctive X-ray finding worth knowing about is the vacuum cleft sign: a thin, dark, horizontal line within a compressed vertebral body that represents a gas-filled space inside the bone. On a standard X-ray, it shows up as a sharp linear lucency (a dark stripe) within the collapsed vertebra. This sign is characteristic of a condition called Kümmell’s disease, where a vertebra that initially seemed to heal after a fracture gradually undergoes internal osteonecrosis (bone death) and progressive collapse weeks or months later.
12PubMed Central. Kummell’s disease: delayed post-traumatic osteonecrosis of the vertebral bodyThe vacuum cleft can also appear in ordinary acute compression fractures, but it behaves differently. In Kümmell’s disease, MRI often shows a “double-line sign” within the vertebral body, which was found in about 27 percent of Kümmell’s cases in one study but not in any of the acute osteoporotic fractures examined. The vacuum cleft tends to change shape with position: it may open up when the patient extends the spine and compress when they flex forward. This dynamic behavior, visible on standing versus supine films, is a helpful diagnostic clue.
13PubMed Central. Comparison of Intravertebral Clefts between Kümmell Disease and Acute Osteoporotic Vertebral Compression Fracture: A Radiological StudyRecognizing this sign matters because Kümmell’s disease represents ongoing vertebral instability. The bone is not healing properly and may continue to collapse, sometimes requiring surgical stabilization rather than the conservative treatment that works for most osteoporotic compression fractures.
14PubMed Central. Vertebral compression fracture with intravertebral vacuum cleft sign: pathogenesis, image, and surgical interventionWhen X-ray Is Not Enough
Plain X-rays remain the first-line imaging study for suspected compression fractures because they’re fast, inexpensive, and widely available. But they have clear limitations. X-rays show bone outlines well but cannot visualize bone marrow, ligaments, or the spinal cord, and they perform poorly at detecting subtle fractures or distinguishing fracture age.
CT scans give much finer bony detail and can reveal features like posterior wall fracture fragments or canal compromise that are invisible on plain film. However, CT has its own blind spots. In one study of acute traumatic spinal fractures, CT’s sensitivity for detecting stable compression fractures was under 50 percent at all spinal levels when compared against MRI as the reference standard.
15Journal of Orthopaedic Reports. Assessment and comparison of the diagnostic value of CT-scan and MRI in acute traumatic spinal compression fracturesMRI changed the clinical game for compression fracture workup. In a study of thoracolumbar trauma patients, MRI discovered 18 occult injuries that were invisible on the initial X-rays and CT, modified the diagnosis in 40 percent of patients, and changed the treatment plan in about 16 percent.
16PubMed Central. Impact of magnetic resonance imaging on decision making for thoracolumbar traumatic fracture diagnosis and treatmentFor patients who are getting a bone density scan anyway, there is a middle-ground option. Modern DXA machines (the same devices used for osteoporosis screening) can produce a lateral spine image called vertebral fracture assessment (VFA). This low-dose technique has been shown to reach about 70 percent sensitivity and over 98 percent specificity for detecting vertebral fractures on a per-vertebra basis, with lower radiation exposure and cost than a full set of spine X-rays.
17PubMed. Vertebral fracture assessment by new dual-energy X-ray absorptiometry The sensitivity is lower for mild fractures, but for moderate and severe fractures, VFA compares favorably to standard radiographs and has the advantage of being done during the same visit as the bone density test.18The Journal of Clinical Endocrinology & Metabolism. Clinical Applications of Vertebral Fracture Assessment by Dual-Energy X-Ray Absorptiometry
Look-Alikes in Children and Adolescents
Reading pediatric spine X-rays for compression fractures is a minefield of false alarms. Growing vertebrae go through a series of developmental stages that can closely mimic the appearance of fractures. Vertebral ring apophyses, the growth plates along the top and bottom edges of each vertebral body, have not yet fused to the main bone in children and adolescents. Their absence can make a vertebra look anteriorly wedged, as if the front wall has been crushed, when in reality the bone simply hasn’t finished growing.
19PubMed Central. Common normal variants of pediatric vertebral development that mimic fractures: a pictorial review from a national longitudinal bone health studySeveral other normal variants can cause confusion. Schmorl’s nodes, which are small herniations of disc material into the vertebral endplate, create a concave divot that looks like the central endplate depression seen in biconcave fractures. An anterior vertebral artery groove can leave a notch in the front of a vertebra that resembles a buckle fracture. Even a simple parallax artifact from the angle of the X-ray beam can create the illusion of an endplate interruption. If an unexpected vertebral shape is spotted on a child’s X-ray, the location within the spine, the exact morphology, and comparison with previous films can usually clarify whether it’s a genuine fracture or just a growing bone doing its thing.
In younger patients without osteoporosis or a history of significant trauma, mild-to-moderate wedging should be interpreted cautiously. One alternative to the standard height-loss approach is an algorithm-based method that looks for specific morphologic clues of true fracture, including cortical step defects and posterior wall retropulsion, which normal developmental variants generally don’t produce.
20PubMed Central. Schmorl’s nodes and vertebral fractures: A diagnostic dilemmaThe Fracture Cascade and Why One Fracture Leads to More
One of the most clinically relevant things about spotting a compression fracture on X-ray is what it predicts for the future. A single vertebral fracture substantially raises the risk of additional fractures, a phenomenon known as the vertebral fracture cascade.
21PubMed. The vertebral fracture cascade in osteoporosis: a review of aetiopathogenesisThe reasons are partly mechanical. When one vertebra collapses, the spine’s normal alignment shifts. The vertebrae above and below the fracture now bear load at different angles, concentrating stress on their endplates in ways they were not designed for. Over time, the spine develops increased forward curvature (kyphosis), which moves the body’s center of gravity forward and puts even more bending force on the thoracic and upper lumbar vertebrae. Combine that with the muscle weakness and reduced mobility that follow a painful fracture, and you get a self-reinforcing cycle of bone loss and further collapse.
22PubMed. The Vertebral Fracture Cascade: Etiology and Clinical ImplicationsThis cascade helps explain why compression fractures tend to cluster in two zones of the spine: the mid-thoracic region (around T7-T8) and the thoracolumbar junction (around T12-L1). These are the areas where the spine’s natural curves create the highest compressive and bending loads. Once a fracture occurs at either location, the adjacent vertebrae are biomechanically primed to follow.
23PubMed. Biomechanics of vertebral fractures and the vertebral fracture cascadeAI-Assisted Detection on Plain X-ray
Given how often compression fractures slip past human eyes, especially on studies ordered for other reasons, there has been growing interest in using artificial intelligence to flag them automatically. Deep learning models trained on large sets of lateral spine X-rays have shown promising results. One neural network achieved about 86 percent accuracy, with sensitivity around 85 percent and specificity around 87 percent. Its sensitivity was significantly higher than that of orthopedic residents, and it performed on par with experienced orthopedic surgeons.
24Scientific Reports. Artificial intelligence for the detection of vertebral fractures on plain spinal radiographyThese tools aren’t replacing radiologists. Their most likely role is as a safety net, running in the background on chest X-rays and CT scans, and flagging vertebrae that look suspiciously short or wedged so that a human can take a closer look. Given that reporting rates for incidental compression fractures on CT scans have been found to hover in the single digits, even an imperfect algorithm could catch fractures that would otherwise vanish into an unremarked corner of the image.
8PubMed Central. Incidental vertebral compression fractures in imaging studies: Lessons not learned by radiologists