From the Battlefield to the Lab: Heterotopic Ossification

(Air Force photo by Tech. Sgt. Tyrona Lawson)

By Lt. Benjamin M Wheatley, USN, MD, Dr. Devaveena Dey, and Dr. Thomas A Davis

TRAUMATIC AMPUTATIONS HAVE LONG BEEN A TERRIBLE SCOURGE OF THOSE INJURED IN WAR, AS HAS THE LESSER-KNOWN BUT PERSISTENT PROBLEM OF EXCESSIVE BONE GROWTH IN THE AFTERMATH OF AMPUTATIONS. THIS OLD CONDITION SUFFERED BY WOUNDED WARRIORS IS GETTING NEW ATTENTION AT THE NAVAL MEDICAL RESEARCH CENTER.

Advances in medicine and technology often followed times of great conflict and war. The modern ultrasound can trace its roots back to World War I when it was developed to track and sink German U-boats. That same war saw the development of plastic surgery because of a large number of war causalities with facial wounds requiring reconstruction. Synthetic rubber, blood banks, ambulances, improved antiseptics, and vaccinations were all developed during times of war.

In our research efforts at the Naval Medical Research Center, in collaboration with university partners, we have seen an old problem, heterotopic ossification (HO), come to the forefront at an unprecedented rate. Written descriptions of HO can be found in medical notations from World War I, and even as far back as the American Civil War, in which excessive bone growth could be found following traumatic amputations. Heterotopic ossification is the formation of bone where it is not supposed to be, such as in muscles or other soft tissues. This aberrant bone commonly forms following severe trauma including crush injuries, burns, and blast injuries. More recent conflicts in Iraq and Afghanistan have seen a relative increase in extremity injuries because of the advances in body armor and its association with blast injuries. Heterotopic ossification has plagued the wounded and the doctors who treat them over the past 16 years. In a pivotal clinical research study conducted at the Walter Reed National Military Medical Center, the Journal of Bone and Joint Surgery (Potter, 2013) reported as many as 65 percent of combat-injured service members who sustained an amputation went on to develop HO.

Heterotopic ossification, at times, can be severely debilitating and painful during the rehabilitative phase of a patient’s care. Since bone is forming where it is not supposed to be, it can crowd or press against other structures such as nerves and muscles in their normal position. It can also create wound healing problems and difficulties with prosthesis fit and wear. If it forms near a joint it may lead to contractures or a decreased range of motion. All of these can make it difficult, if not impossible, for patients to regain their independence.

Prior to the current conflicts, HO was not a common problem in military medicine and it was not seen nearly as often in civilian trauma. The recent wars generated a great deal of interest in understanding the mechanisms behind HO development so we could develop effective treatments.

A two-pronged approach was developed by researchers at the Naval Medical Research Center in collaboration with doctors at Walter Reed National Military Medical Center and university researchers. This included clinical research studies of patients who developed HO and bench-level laboratory research developing a model to recreate the injury patterns commonly seen in combat. Initially, two injury patterns were identified that were commonly seen in the clinical experience.

Heterotopic ossification, at times, can be severely debilitating and painful during the rehabilitative phase of a patient’s care. Since bone is forming where it is not supposed to be, it can crowd or press against other structures such as nerves and muscles in their normal position.

The first was a blast overpressure model, which is the concussive force experienced in proximity to an explosion (such as an improvised explosive device). The second was a severe extremity injury model of a femur fracture followed by a prolonged crush injury and finally an amputation through the “zone of injury.” The zone of injury refers to an area surrounding a wound that may appear normal at first but is actually damaged by the initial injury.

These two injury models were tested in isolation and in combination to determine their effect on the amount of HO formed. We found the combination injury led to a significantly greater amount of bone formation than either injury model alone, and it did so consistently. This gave us the model we would use to further characterize the pathways in HO on a molecular level.

Once the laboratory model was created, our next goal was to characterize the cell signaling pathways and gene expression that ultimately produces HO. Gene expression is the process by which various portions of DNA are turned on or off to produce, or stop the production of, various proteins. These proteins direct the function of a variety of cells, including those that form bone. Genes may be differentially expressed during times of growth, stress, or injury as the body compensates to the change in its environment to maintain homeostasis.

By microscopic analysis of the cellular composition, we know the bone formed by HO is normal bone in its appearance and cellular structure, closely resembling the long bones in the body such as the femur and tibia. What makes it abnormal is the location. Long bones in the body are typically formed by a process called “endochondral ossification,” when bone is formed by first laying down a cartilage scaffold that is slowly replaced by bone cells through a complex cellular mechanism. Interestingly, the bone formed during HO follows the same mechanism.

Following the injury, early cartilage cells could be found propagating in the injured tissues. Many of the genes involved in the production of inflammatory molecules, cartilage, and bone were expressed at levels that were orders of magnitude higher compared to the uninjured controls. The variable expression of these genes changed over time with some initially up-regulated and later down-regulated and vice versa. By understanding these early cellular and molecular changes occurring in HO, we can better understand the disease process and progression. This in turn will help us to develop therapies directed to specific cellular and molecular targets to interrupt the process early on.

Our laboratory model was still too simple at this point. The blast injuries service members sustain are complex, with open injuries highly contaminated with dirt and debris. These injuries require multiple surgeries to control infections and promote healing, all of which can increase the inflammation in the wound and promote the formation of HO. Our initial injury model was fine-tuned to more closely recreate a typical injury pattern by introducing a bacterial infection into the wound.

Returning to the clinical research, a study of the bacteria in wound infections found that nearly 25 percent involved a bacterium called methicillin-resistant Staphylococcus aureus, or MRSA. Our studies of the early characterization of HO, as well as other HO studies available in the literature, point to increased inflammation as a potential driving factor in the development of heterotopic bone. We hypothesized that by including a bacterium, which is commonly found in wound infections, we would increase the local inflammation and thereby facilitate HO development. We did, in fact, find the introduction of MRSA resulted in a significant increase in the amount of bone formed compared to those without bacterial infection. With the new laboratory model finalized and the early characterization of gene expression complete, we next focused on treatment. In addition to blast injuries, HO also develops following some major surgeries such as hip and knee replacements. This type of HO has been studied extensively in the civilian population by universities and researchers across the country to improve outcomes following surgery. Two treatments have been developed in these patients to prevent the formation of HO following surgery. They are nonsteroidal anti-inflammatory drugs, such as Ibuprofen or Motrin, and local radiation therapy, similar to that used in cancer treatments but at lower doses. While these treatments are effective in a patient undergoing surgery, neither is practical in combat-injured service members returning from the battlefield to military treatment facilities in the United States.

Once HO is established in an injured extremity, it can only be treated with surgery in which the offending bone is removed. Surgery to remove HO can lead to set backs in healing and rehabilitation and carry the risk of recurrence. An ideal therapy would be easy to implement in an austere environment and selective to prevent unwanted side effects. To date, we have studied three different therapies in the laboratory model, each of which act on different parts of the disease process.

The first substance tested was Palovarotene. This drug was originally developed for the treatment of chronic obstructive pulmonary disease, but it was later found to block new bone formation. This led to its use in a rare, genetic form of HO called fibrodysplasia ossificans progressiva, or FOP. Both HO and FOP are similar in the formation of abnormal bone in soft tissues, but bone growth in FOP is triggered either spontaneously or after a relatively minor trauma, such as vaccination, where bone is formed at the site of injection. Palovarotene works by preventing the formation of the cartilage scaffold which later turns into bone. When applied to our laboratory model, we found this drug did lead to a significant reduction of HO as measured by the total volume of bone formed. As expected, the gene expression for cartilage and bone producing cells was also reduced. There was a trend, however, toward increased wound healing complications or delays in the laboratory model studies with Palovarotene.

Once HO is established in an injured extremity, it can only be treated with surgery in which the offending bone is removed. Surgery to remove HO can lead to set backs in healing and rehabilitation and carry the risk of recurrence. An ideal therapy would be easy to implement in an austere environment and selective to prevent unwanted side effects.

The next drug tested was Rapamycin, which is typically used to prevent organ rejection after transplants. It has both immune suppressive and anti-inflammatory effects. It also has a myriad of other effects, including preventing the migration of progenitor cells and inhibiting the formation of blood vessels necessary for the development of HO. As seen in the Palovarotene study, Rapamycin also resulted in significant reduction in heterotopic bone volume, and lower expression of genes related to cartilage and bone formation. Rapamycin is a powerful immune suppressant that could be contraindicated in a patient who is also battling infection, as it may prevent their body from mounting an adequate response to the infection. Unlike Palovarotene, Rapamycin has been associated with minimal to no wound healing complications.

The third drug we tested is an antibiotic, Vancomycin. This antibiotic was chosen for its effectiveness against MRSA in particular. We hypothesized that if the addition of MRSA worsened the inflammation and HO formation, then treating the infection should reduce both. This turned out to be true but with some surprising results. First, in the injury model with MRSA infection, the addition of Vancomycin was able to completely eradicate the infection and significantly reduce the amount of HO formed. Interestingly, when the injury model was applied without the addition of MRSA, Vancomycin still resulted in significant reductions in the amount of HO formed. This indicates Vancomycin has other effects on HO formation than only clearing infection. This also means Vancomycin may have a broader application than we initially thought. We are currently investigating the mechanisms through which Vancomycin may be acting.

We are now in the process of designing experiments for additional therapies. These will include testing novel drugs developed specifically to inhibit HO formation as well as testing other drugs in combination. By using Rapamycin and Palovarotene in combination, we may be able to lower the dose of each drug and reduce the risk of harmful side effects while maintaining the effectiveness.

Our understanding and knowledge of heterotopic ossification has expanded dramatically over the past decade. More recent research has focused on treatment options that are safe and less invasive than surgical excision. Questions about side effects, practicality, and efficacy still remain. Even with all that we have learned, there is a long way to go before we will be able to transition the lessons we have learned in the laboratory to the men and women of our military.

About the authors:

Lt. Wheatley is an orthopedic surgery resident at the Naval Medical Research Center. Dr. Dey is a stem cell biologist, and Dr. Davis is scientific director, both in the regenerative medicine department at the Naval Medical Research Center.