How Fast Does Avascular Necrosis Progress?

Avascular necrosis, often called AVN or osteonecrosis, progresses at wildly different rates depending on how much bone has lost its blood supply, where the damage sits, and what caused it. In the hip, where AVN is most common, untreated disease leads to collapse of the femoral head in roughly a third of cases, with a mean time to collapse of about four years in some studies and as little as two years in others. But those averages obscure enormous variation: a small, centrally located lesion may never collapse at all, while a large one involving the weight-bearing surface can give way within months.

The General Timeline for Untreated Hip AVN

Most of what we know about how quickly AVN advances comes from studies tracking hips that were diagnosed but either left untreated or managed conservatively. One study following 81 asymptomatic hips in patients already diagnosed with AVN on the other side found that about a third eventually collapsed, with a mean interval from diagnosis to collapse of just over four years.¹ A larger pooled analysis of nearly 670 untreated hips found that about 59% eventually progressed to symptoms or collapse, but outcomes varied sharply by lesion characteristics: small, medially located lesions collapsed less than 10% of the time.²

On the faster end of the spectrum, a study specifically tracking “silent” hips (the opposite hip in patients with bilateral disease) found that every single involved hip eventually collapsed, at a mean of 23 months after being identified. The range was broad, from just three months to over five years.³ The takeaway is that “average time to collapse” is a misleading number on its own. Whether your case progresses in months or years depends on a handful of measurable factors.

Lesion Size and Location Are the Strongest Predictors

The single biggest predictor of how fast AVN progresses is how much of the femoral head is involved and where the dead bone sits. Researchers quantify this using imaging-based measurements of the necrotic area. In one study using MRI-based measurements of the “combined necrotic angle,” every hip with an angle above 240 degrees collapsed within three years, while none of those below 190 degrees did. Hips in the middle range collapsed about half the time.⁴ In plain terms, a large dead zone that extends into the weight-bearing dome of the femoral head is almost certain to collapse. A smaller dead zone tucked toward the center or the edge of the joint has a much better chance of staying intact.

A prediction model built from patients who had undergone core decompression surgery identified several factors that independently sped up collapse: a necrotic angle above 250 degrees, lesions located in the lateral weight-bearing zone, sudden rather than gradual onset of symptoms, and the presence of bone marrow edema on MRI before surgery.⁵ Bone marrow edema deserves special attention. When MRI shows swelling in the bone marrow around the necrotic zone, collapse follows the majority of the time. In one study, 85% of hips that developed bone marrow edema went on to advanced disease.⁶ If your MRI report mentions bone marrow edema alongside AVN, the disease is more likely to progress quickly.

How the Cause of AVN Affects the Timeline

AVN has several known triggers, and the underlying cause influences both how early it gets caught and how rapidly it worsens.

Steroid-induced AVN is one of the most studied forms. The onset is often sneaky: symptoms may not appear until weeks to a full year after steroid use ends.⁷ Some reports put the window at three weeks to three months after glucocorticoid treatment.⁸ The biological reason this damage is so insidious is that glucocorticoids trigger programmed cell death of osteocytes, the cells embedded in bone that sense mechanical load. Because the body cannot replace dead osteocytes once they are gone, the bone gradually loses its ability to sense and respond to stress, setting the stage for structural failure.⁹

COVID-19 brought renewed interest in steroid-related AVN. A pooled analysis of post-COVID cases found that the mean time from infection to AVN onset was about 126 days, roughly four months.¹⁰ The steroid doses used in COVID treatment were typically lower than those given during the SARS epidemic, but the risk was still real. If you received prolonged steroid treatment for COVID or any other condition, new hip or groin pain in the months afterward warrants imaging.

Alcohol-induced AVN tends to present later and more severely than steroid-related cases. Patients with alcohol-related disease were significantly older on average, at about 49 years, and 90% already had collapsed femoral heads at first presentation.¹¹ This does not necessarily mean alcohol-related AVN progresses faster biologically. It more likely means that heavy drinkers are less likely to seek medical attention for hip pain early, so the disease is further along by the time it is diagnosed. When a separate study tracked asymptomatic hips by cause, alcohol-related cases developed symptoms at a slightly higher rate (about 64%) than steroid-related (57%) or idiopathic cases (46%).¹²

Post-traumatic AVN, which develops after a hip fracture or dislocation that disrupts blood supply, follows its own clock. Imaging studies using bone scans after femoral neck fracture repair have shown that persistent blood flow deficits in the femoral head predict AVN development with high accuracy. In patients whose imaging showed only partially recovered blood flow at a few months post-surgery, over 60% went on to develop AVN.¹³

Sickle Cell Disease Accelerates Everything

AVN in people with sickle cell disease is a different animal. It hits younger, progresses faster, and responds worse to treatment than in the general population. Half of sickle cell anemia patients develop hip AVN at some point, and without effective intervention, about half of those need hip surgery within two years of diagnosis.¹⁴

A long-term study following over 120 asymptomatic hips in sickle cell patients found that 91% eventually became painful and 77% collapsed. Even hips with the earliest radiographic stage of disease were not safe: 84% of those went on to become symptomatic and 61% collapsed. Hips caught at a slightly later asymptomatic stage were even worse off, with 95% developing pain within three years and 86% ultimately collapsing. Among those who became symptomatic, the average time from pain onset to collapse was just 11 months.¹⁵

Long-term outcomes remain poor even with treatment. A multicentre study found that most sickle cell patients with hip AVN had bilateral disease, over half eventually required total hip replacement (at a median age in the mid-30s), and a high rate of joint dysfunction and pain persisted ten years after diagnosis regardless of the treatment approach.¹⁶ If you have sickle cell disease and develop any hip or groin discomfort, the threshold for getting an MRI should be very low.

AVN in Joints Beyond the Hip

Though the hip gets the most attention, AVN can affect other joints, and the pace of progression differs. The shoulder is the second most common site of nontraumatic AVN.¹⁷ Shoulder AVN often progresses more slowly than hip AVN, partly because the shoulder is not a weight-bearing joint. The forces pushing toward structural collapse are simply smaller. That said, advanced shoulder AVN still causes significant pain and functional limitation, especially in patients who need their arms overhead for work or sport.

AVN of the lunate bone in the wrist, known as Kienböck’s disease, is an interesting contrast. A study tracking patients diagnosed after age 50 found remarkably little progression over nearly eight years of follow-up: the bone structure measurements stayed stable, arthritis did not develop, and only one patient showed any radiographic worsening.¹⁸ Kienböck’s in younger patients can be more aggressive, but the general point stands: AVN in non-weight-bearing or small joints tends to be more indolent than in the hip.

Children Versus Adults

When AVN of the femoral head strikes children, it goes by the name Legg-Calvé-Perthes disease and behaves differently than adult AVN. The condition typically affects children under 15, with the blood supply disruption and subsequent bone death following a broadly similar pattern to adult disease.¹⁹ The crucial difference is remodeling potential. Children’s bones are still growing, and the femoral head can reshape itself to a surprising degree as it heals. Younger children, especially those under six at diagnosis, have the best chance of the deformed head remodeling into a functional shape.²⁰

In a large prospective study of 368 children with Perthes disease, the strongest predictor of outcome was the extent of femoral head involvement: children with more than half the head affected did substantially worse. Age at diagnosis was the next strongest predictor, with older children facing worse outcomes. In children over six with extensive involvement, surgical correction led to significantly better results than bracing or physiotherapy alone, while in children under six there was no significant difference among treatment approaches.²¹ Adults lose this remodeling advantage entirely. In adult AVN, damaged bone structure essentially stays damaged unless it is surgically replaced, which is why the window for early intervention matters so much.

How Treatment Changes the Clock

The whole reason progression speed matters clinically is that catching AVN before the femoral head collapses opens up treatments that become ineffective afterward. Core decompression, a procedure where a surgeon drills into the femoral head to relieve pressure and stimulate healing, works best when done before any collapse has occurred.²² A prospective randomized trial found that combining core decompression with platelet-rich plasma injection reduced disease progression in early-stage cases compared to core decompression alone. Among patients with stage 2 disease, the progression rate dropped from 43% to 24% with the addition of platelet-rich plasma, and the overall rate of surviving without femoral head collapse was 84% in the combination group versus 68% with surgery alone.²³

Bisphosphonates, a class of drugs that slow bone breakdown, have shown promise in slowing AVN progression. A 20-year study reported that both oral bisphosphonate therapy alone and combination bisphosphonate regimens slowed disease progression, reduced the rate of collapse, and reduced the need for joint replacement surgery.²⁴ That said, the evidence base is still catching up with the enthusiasm. A systematic review noted that no controlled, double-blinded studies had established bisphosphonate efficacy for AVN at the time of publication, even though uncontrolled studies showed favorable results in reducing pain, improving mobility, and lowering collapse rates.²⁵ Bisphosphonates are not a guaranteed rescue, but they appear to buy time in early disease.

Once collapse has already occurred, the question shifts from “can we save the joint?” to “when do we replace it?” A meta-analysis of conservative treatment studies found that in patients with more advanced disease (stage III), about 37% ultimately needed total hip replacement, compared to just 8% of those treated at stage II.²⁶ The progression from stage II to stage III is where the biggest jump in joint replacement rates occurs, which is why so much clinical energy focuses on catching disease before that threshold.

Stem Cell and Regenerative Approaches

Regenerative strategies are the most active frontier in slowing AVN progression. A meta-analysis comparing core decompression with added stem cell therapy versus core decompression alone found that the combination was more effective at preventing collapse, slowing radiographic progression, and avoiding the need for total hip replacement.²⁷ A scoping review of core decompression combined with concentrated bone marrow aspirate reached a similar conclusion, finding that the combination improved functional scores and reduced both radiographic progression and the need for hip replacement in pre-collapse disease.²⁸ The underlying idea is that delivering concentrated stem cells or growth factors directly into the necrotic zone can kick-start the formation of new blood vessels and new bone, essentially rebooting a repair process that stalled.²⁹

These treatments are promising but still maturing. Most studies involve relatively small numbers of patients and follow-up periods under ten years. Nobody yet has definitive evidence that stem cell augmentation prevents joint replacement over a lifetime. But for patients caught early, the data so far suggest these approaches meaningfully slow the disease down.

Timing of Hip Replacement and What Happens After Collapse

For patients whose femoral head has already collapsed, the remaining question is when to proceed with hip replacement and how the duration of collapse affects outcomes. A recent study found a somewhat surprising pattern: patients who had lived with a collapsed femoral head for one to three years before surgery reported more pain after hip replacement than those who had surgery within a year of collapse or those who waited more than three years. The authors described this as a “U-shaped” trend in post-surgical pain, though all groups eventually reached similar functional outcomes.³⁰ The practical implication is that living with a collapsed femoral head for a moderate period does not necessarily make the eventual replacement work any less well functionally, but it may affect the pain experience in the intermediate term.

AVN patients who need hip replacement tend to be younger than typical hip replacement recipients, often in their 30s, 40s, or 50s. That means the prosthetic joint needs to last through decades of active life. Modern implants last 20 years or more in the majority of patients, but for someone who gets a replacement at 35, the possibility of a revision surgery later in life is real. This is another reason the medical community pushes so hard for early-stage intervention: not to avoid replacement forever, but to delay it as long as possible so fewer lifetime surgeries are needed.

When AVN Is Found Early by Accident

An increasingly common scenario is AVN found incidentally on an MRI done for something else, such as a back problem or an unrelated injury. These asymptomatic cases present a genuine dilemma. The disease is there on imaging, but you feel fine. Will it progress?

The data suggest roughly half of asymptomatic hips will eventually develop symptoms. In one study, about 56% became symptomatic at an average of just over two years after diagnosis, with alcohol-related cases progressing at slightly higher rates than steroid-related or idiopathic ones.¹² But the other half stayed stable. The lesion characteristics described earlier, particularly size and location, are the best guide to whether watchful waiting is reasonable or early intervention makes more sense. A small, centrally located lesion in someone with no ongoing risk factors (not currently on steroids, not drinking heavily) is a reasonable candidate for monitoring with serial MRI. A large lesion in the weight-bearing zone, especially with bone marrow edema, is not.

The fact that AVN can sit silently for months or even years before causing problems has practical consequences. If you have known risk factors, particularly recent high-dose steroid use, sickle cell disease, heavy alcohol use, or a recent hip fracture, and you develop any aching in the groin, thigh, or buttock, do not wait for it to get bad enough to limp. An MRI is the only way to catch AVN before collapse, and the entire treatment landscape is better before that threshold is crossed.