Fat Embolism Syndrome And Orthopedic Surgery

Regardless of the exact mechanism for the development of fat embolism syndrome, in trauma it most commonly occurs when the medullary (bone marrow) cavity of a long bone is violated. This occurs first when the bone is fractured, and again when it is instrumented for fixation. The initial shower of emboli cannot be prevented. However, ongoing emboli can be reduced with early fixation. This can be in the form of a good splint, or surgical external or internal fixation.

One type of internal fixation, intramedullary (IM) nailing, has been associated with embolism and FES for some time. This technique was introduced 80 years ago and has been refined significantly since. Here is a picture of a femur with an IM nail.

The nail is inserted proximally near the greater trochanter. The marrow cavity is first reamed to make insertion of the nail easier. This causes a number of changes in the physiology of and pressures within the marrow cavity. Pressure increases during the initial reaming, and hits a peak when the reamer enters the distal fragment. Once complete, there are no further increases as the nail is inserted. However, these pressure changes alter medullary blood flow and allow emboli to enter the venous system.

Reaming is actually beneficial in several ways. It simplifies and shortens the surgical procedure. And in animal models there is evidence that bone debris from the reaming process collects at the fracture site, creating an autograft that may improve healing.

A surgical group in Ireland has been using a novel technique for lavaging the marrow cavity during fixation for several years. Once the bone is entered proximally, a cut piece of suction tubing is inserted into the end of the bone. Suction is then applied for 2-3 minutes. The procedure continues, including reaming, then the suction procedure is repeated. Unfortunately, FES is uncommon, so it is difficult to judge whether their technique really works. The authors believe it is safe, but recommend formal studies to prove efficacy.

Use of an additional venting hole between the trochanters has also been studied in a small randomized trial. This allows for drainage of marrow during the reaming process, reducing any pressure rise. The number of embolic events detected using transesophageal echo was significantly reduced in the vented group (20% vs 85% of patients).

Next, prevention and treatment of fat embolism syndrome.


  1. A Simple and Easy Intramedullary Lavage Method to Prevent Embolism During and After Reamed Long Bone Nailing. Cureus 9(8):e1609, Aug 2017.
  2. Relevance of the drainage along the linea aspera for the reduction of fat embolism during cemented total hip arthroplasty. A prospective, randomized clinical trial. Arch Ortho Trauma Surg 119:146, 1999

Diagnosis Of Fat Embolism Syndrome

A number of scoring systems have been developed to identify FES (Gurd’s and Wilson’s criteria, Schonfeld’s criteria, Lindeque’s criteria to name a few). Unfortunately, none of these are helpful. They were developed in the 1980s as part of the authors’ studies on the use of  steroids for treatment, and no one else has taken the time to study their sensitivity and specificity.

Diagnosis of FES is primarily clinical. It relies upon recognition of the principal findings on physical exam, and exclusion of more common conditions that may mimic it.

Here is a template for diagnosing FES:

Is your patient at risk? The vast majority of these patients will have fractures. One, or especially two or more long bone fractures (mostly the femur) are usually present. Other fractures that add risk are those involving the pelvis or bones that contain marrow, such as the ribs and sternum. Patients who have just undergone fracture repair are also at risk and will be discussed in the next section. Finally, patients who have had intraosseous lines placed are also at risk, regardless of the type of infusate.

What signs or symptoms have developed? Skin changes are very suggestive of FES if your patient is at risk. However, rashes are common manifestations of contact allergies, drug reactions, infectious diseases, and many other conditions. If those are ruled out, then the presence of risk factors plus a rash is sufficient to make the diagnosis.

Mental status changes are more difficult to pin on FES, even though it is a more common initial presentation than the rash. Since this is a trauma patient, you must rule out delayed manifestations of head trauma. Urgent CT of the head is required to do so. And typically, there will be no specific findings that point to FES. It is always a diagnosis of exclusion.

Pulmonary dysfunction requires a search for the usual suspects. A good physical examination of the chest coupled with a chest x-ray will help identify pneumothorax, hemothorax, or pneumonia. A chest CT may be indicated if pulmonary embolism is suspected.

Once other more common clinical problems have been eliminated, you are left with the diagnosis of FES. There are no specific lab tests to draw, and more invasive studies are neither helpful nor indicated. Fat embolism syndrome is a diagnosis of exclusion.

Next, the relationship of fat embolism and orthopedic surgery.

Clinical Manifestations Of Fat Embolism Syndrome

There are three organ systems that are classically involved in FES: pulmonary, CNS, and skin. Manifestations generally begin between 24 and 72 hours after injury. In rare cases, symptoms can begin within 12 hours. In my experience, these tend to be the ones that become the most severe and are frequently life-threatening.

Pulmonary (95% of cases): This is the most common manifestation of FES, and may occur without other signs and symptoms. Nearly all patients develop some degree of hypoxia. Progressive tachypnea and mild tachycardia may provide the first clinical clue if oxygen saturation is not being monitored.

Chest x-ray is usually unremarkable early on. And once the syndrome has developed, it is generally not helpful. CT scan is useful for defining the extent of pulmonary injury, but lags the clinical picture by several days. Findings are non-specific, usually consisting of small, ground-glass opacities in the periphery.

In the example above, the opacities are very small and difficult to see.

But they’re a little more obvious here!

Other CT findings include small pulmonary nodules in the upper lobes or along peripheral pulmonary vessels. These are thought to be areas of obstruction caused by the emboli. Nonspecific pleural effusions may be seen, and bronchial thickening has also been described. Rarely, fat globules may be seen in the lower extremity veins or IVC, and should immediately raise suspicion for developing FES even before symptoms develop.

CNS (60% of cases): If they occur, CNS changes generally crop up after the pulmonary manifestations begin. Generally, they start as mild confusion, but can progress to decreasing level of consciousness and even coma. Focal neurologic deficits are occasionally seen, and seizures can occur.

The actual mechanism behind this appears to be very similar to the skin changes which will be described in the next section. Emboli occur in vessels predominantly in the white matter of the brain. This leads to petechial hemorrhages, which are likely due to the inflammatory mechanisms previously described.

Note the numerous dark petechiae visible in the white matter in this specimen.

Retinal exam can also show evidence of fat embolism. Fat globules may actually be seen in the retinal vessels early.

Note the fat globules at the 9:30 and 2:00 positions to the optic nerve in the image above.

Skin (33% of cases): The most recognizable sign of FES is the petechial skin rash. This rash usually involves the torso, and axillary petechiae are very common. It can spread to involve the head and neck, and occasionally the extremities. Subconjunctival hemorrhages are sometimes seen. The rash tends to be transient and usually lasts only a few days. Here is an example of the classic petechial rash.

Other findings: Fat globules may be found in the urine in patients with FES. However, they are commonly present in patients with long bone fractures, so their presence is not helpful or predictive. Nonspecific findings such as fever, leukocytosis, anemia, and thrombocytosis are also relatively common. In severe cases, cardiac dysfunction, hypotension, and peripheral hypoperfusion can occur. I have personally seen necrosis of fingers and toes from a very severe case.

Unfortunately, the “classic” triad of mental status changes, skin rash, and pulmonary insufficiency are seen in only a small minority of patients. Typically, only one or two signs and symptoms appear at the same time, making diagnosis a bit challenging.

In the next post, making the diagnosis of fat embolism syndrome.

Fat Embolism vs Fat Embolism Syndrome

It’s fat embolism week! I’ll cover this uncommon, yet very important clinical condition in my next four posts.

Fat embolism syndrome (FES) is one of those clinical problems that trauma professionals read about during their training, then rarely ever see. Although the clinical manifestations are frequently mild, they can progress rapidly and become life-threatening. Over the next five days, I’ll try to  help you better understand this condition, and provide details on diagnosis and treatment.

Fat embolism syndrome (FES) is a constellation of findings that arise from a single, unified cause: the escape of fat globules into the circulation (fat embolism). The ultimate resting places of those globules determine the specific manifestations of FES seen in clinical practice. When it occurs, it typically becomes apparent 24 to 72 hours after injury.

Simple fat embolism occurs to some degree any time tissues containing fat are manipulated or injured. It has been demonstrated during plastic surgical injections for cosmetic purposes and lipid infusions. It is more frequently seen with orthopedic injuries, especially those involving the femurs and pelvis. And it makes sense that the more fractures that are present, the more likely fat embolism will occur. Embolism is also known to occur when performing orthopedic procedures, particularly those involving the marrow cavity (intramedullary nailing), but has also been reported in total knee and hip procedures.

Fat embolism syndrome has a generally reported incidence of 1 – 10%, although I believe that is on the high side. I see a case every 3 – 4 years in a predominantly blunt, fracture-laden practice. Fat embolism without symptoms occurs much more frequently. A study from 1995 using transesophageal echo found evidence of emboli in 90% of patients with long bone fractures.

But how do these fat globules get into the circulation and produce such chaos? We know that they can be mechanically pushed into small venules when tissues containing fat cells or bone marrow are injured. In bone, there are numerous small venules located throughout that are anchored to it. When the bone is fractured, these venules tear and are held open so yellow (fatty) marrow can be pushed into them.

If enough emboli enter the blood stream, they may accumulate in the end vessels of tissues and block flow. Although this is a simple and appealing explanation, it may not be the full story. If the emboli primarily occur during and after injury, why does it take several days for the full-blown syndrome to develop?

A likely explanation is that the fat globules begin to degrade while in the circulatory system. Breakdown into free fatty acids results in the release of a cascade of cytokines and other mediators. The inflammatory response around the end vessels create the gross pathology that we associate with fat embolism syndrome.

In the next post, clinical manifestations of fat embolism syndrome.

Are Transfusing Too Much Blood During The MTP?

The activation of the massive transfusion protocol (MTP) for hypotension is commonplace. The MTP provides rapid access to large volumes of blood products with a simple order. Trauma centers each design their own protocol, which usually includes four to six units of PRBC per MTP “pack.”

This rapid delivery system, coupled with rapid infusion systems, allows the delivery of large volumes of blood and other blood products very quickly. But could it be that this system is too slick, and we are a bit too zealous, and could even possibly transfuse too much blood?

The trauma group at Cedars-Sinai in Los Angeles retrospectively reviewed their own experience via registry data with their MTP over a 2.5 year period for evidence of overtransfusion. All patients who received blood via the MTP were included. Patients who had a continuous MTP > 24 hours long, those who died within 24 hours, and those who had a missing post-resuscitation hemoglobin (Hgb) were excluded.

The authors arbitrarily defined overtransfusion as a Hgb > 11 at 24 hours. They also compared the Hgb at the end of the MTP and upon discharge with this threshold. They chose this Hgb value because it allows for some clinical uncertainty in interpreting the various endpoints to resuscitation.

Here are the factoids:

  • 240 patients underwent MTP during the study period, but 100 were excluded using the criteria above, leaving 140 study patients
  • Average injury severity was high (24) and 38% suffered penetrating injury
  • Median admission Hgb was 12.6
  • At the conclusion of the MTP, 71% were overtransfused using the study definition, 44% met criteria 24 hours after admission, and 30% did at time of discharge
  • Overtransfused patients were more likely to have a penetrating mechanism, lower initial base excess, and lower ISS (median 19)

The authors concluded that overtransfusion is more common than we think. This may lead to overutilization of blood products, which has become much more problematic during the COVID epidemic. They recommend that trauma centers track this metric and consider it as a quality of care measurement.

Bottom line: This is a nicely crafted and well-written study. It asks a simple question and answers it with a clear design and analysis. The authors critique their own work, offering a comprehensive list of limitations and a solid rationale for their assumptions and conclusions. They also offer a good explanation for their choice of Hgb threshold in defining overtransfusion.

I agree that overtranfusion truly does occur, and I have seen it many times first-hand. The most common reason is the lack of well-defined and reliable resuscitation endpoints. How do we know when to stop? What should we use? Blood pressure? Base excess? TEG or ROTEM values? There are many other possibilities, but none seem reliable enough to use in every patient. 

Patients with penetrating injury proceeding quickly to OR more commonly experience overtransfusion. This may be due to the reflexive administration of everything in each cooler and the sheer speed with which our rapid infuser technology can deliver products. The more product in the cooler, the more that is given, which may lead to the overtranfused condition. 

The authors suggest reviewing the makeup of the individual MTP packs, and this makes sense. Are there too many in it? This could be a contributing factor to overtransfusion. It might be an interesting exercise to do a quick registry review at your own center to obtain a count of the number of MTP patients with a final Hgb > 11. If you find that your numbers are high, consider reducing the number of red cell packs in the cooler to just four. But if you already only include four, don’t reduce it any further. And in any case, critically review the clinical indicators your  surgeons use to decide to end the MTP to see if, as a group, they can settle on one to use consistently. 

Reference: Overtransfusion of packed red blood cells during massive transfusion activation: a potential quality metric for trauma resuscitation. Trauma Surg Acute Care Open 7:e000896., July 26 2022.