Bone marrow aspirate grafting entails mesenchymal stem cell-containing bone marrow harvesting and injection into a fracture site to promote bone formation. Although the use of bone marrow aspirate in orthopedic trauma is not widespread, an increasing number of studies are reporting clinical success. Advantages of using bone marrow aspirate are that it is readily obtainable, has low harvest morbidity, and can be easily and quickly injected. However, no universally accepted role for its use exists. Future studies directly comparing bone marrow aspirate with conventional techniques are needed to define its role in the treatment of orthopedic trauma patients.
Key points
- •
Bone marrow aspirate contains mesenchymal stem cells (MSCs) capable of promoting the formation of bone.
- •
Bone marrow is ideally aspirated from the iliac crest using small syringes in multiple small volume aliquots.
- •
Determining the iliac crest harvest location and whether to concentrate the aspirate is at the surgeon’s discretion; a greater number of injected MSCs is associated with better healing.
- •
Clinical case series have reported that bone marrow aspirate injection has a success of 75% to 90% in treating atrophic tibial nonunions.
Introduction
The adult skeleton possesses 2 types of bone marrow: yellow and red. Yellow marrow has undergone adipose involution and is inactive. The red marrow possesses hematopoietic cells as well as 2 known populations of adult stem cells. Hematopoietic stem cells give rise to all cell components of circulating blood, such as erythrocytes, lymphocytes, neutrophils, and thrombocytes. The other stem cell population consists of mesenchymal stem cells (MSCs), which are also known as marrow stromal cells. MSCs have the potential to differentiate into connective tissue cells such as osteoblasts, osteocytes, adipocytes, and chondrocytes. The ability of MSCs to differentiate into bone-producing cells has led to interest in their clinical use in orthopedic trauma to improve fracture healing and to treat bone defects. Numerous in vivo animal studies as well clinical human case series have reported the successful use of MSCs or MSC-containing bone marrow aspirate.
This review summarizes the basic science and clinical results of using MSCs and MSC-containing bone marrow aspirate to increase bone formation in the setting of a nonunion or a critical sized defect. It also reviews the technique of aspirating bone marrow and the various ways of increasing the MSC concentration of the aspirate.
Introduction
The adult skeleton possesses 2 types of bone marrow: yellow and red. Yellow marrow has undergone adipose involution and is inactive. The red marrow possesses hematopoietic cells as well as 2 known populations of adult stem cells. Hematopoietic stem cells give rise to all cell components of circulating blood, such as erythrocytes, lymphocytes, neutrophils, and thrombocytes. The other stem cell population consists of mesenchymal stem cells (MSCs), which are also known as marrow stromal cells. MSCs have the potential to differentiate into connective tissue cells such as osteoblasts, osteocytes, adipocytes, and chondrocytes. The ability of MSCs to differentiate into bone-producing cells has led to interest in their clinical use in orthopedic trauma to improve fracture healing and to treat bone defects. Numerous in vivo animal studies as well clinical human case series have reported the successful use of MSCs or MSC-containing bone marrow aspirate.
This review summarizes the basic science and clinical results of using MSCs and MSC-containing bone marrow aspirate to increase bone formation in the setting of a nonunion or a critical sized defect. It also reviews the technique of aspirating bone marrow and the various ways of increasing the MSC concentration of the aspirate.
Basic science
The critical component of bone marrow aspirate for use in orthopedic trauma is its population of osteoprogenitor cells. MSCs reside in bone marrow and are multipotent cells with the ability to differentiate into osteoprogenitor or chondroprogenitor cells based on molecular signals from their local environment. Osteoprogenitor cells can then differentiate into osteoblasts in response to surrounding cytokines and growth factors. The differentiation of osteoprogenitor cells and their ability to promote bone formation is augmented by the osteoinductive factors also contained within bone marrow aspirate.
Numerous basic science animal studies have investigated the ability of MSCs to stimulate bone formation. A recent systematic review by Gianakos reported the pooled results of 35 animal studies where bone marrow aspirate concentrate (BMAC) was used to treat critical sized bone defects. The reviewed studies used different animal models, including rabbits, rats, mice, goats, canines, sheep, and pigs. They found that, of the studies that reported their results statistically, 100% (14/14) found significantly greater radiographic evidence of osteogenesis, 81% (13/16) showed significantly increased mean bone volume using micro-CT scanning, and 90% (19/21) found significantly more bone formation based on histologic analysis versus control groups. The limitations of these studies are the wide variation in the animal and bone defect models used as well as the unknown translatability of their findings to human clinical scenarios. However, numerous clinical outcome studies have been published and their findings are comparable with these animal model results.
Clinical outcomes
The clinical application of bone marrow aspirate in orthopedic trauma is a relatively new development. The majority of the published studies regarding the clinical outcomes of patients treated with bone marrow aspirate are case series treating nonunions, especially of the tibia. However, bone marrow may have a role not only in delayed or nonunited fractures, but also in certain acute fracture cases.
Acute Fractures
To our knowledge, no published studies have reported the use of bone marrow aspirate in the management of acute fractures. However, there is some evidence that bone marrow aspirate combined with allograft or a collagen scaffold may have the same osteogenic capability as autograft, and therefore could potentially be substituted in acute fracture cases with an osseous defect where primary bone grafting is indicated. Hernigou and colleagues evaluated 20 patients with acetabular defects secondary to osteolysis after total hip arthroplasty. Revision of the acetabular component included grafting the osteolytic defect with either autograft, allograft, or allograft with bone marrow aspirate. All patients later underwent a re-revision for femoral component failure after a mean of 10 years from acetabular grafting. At that time, the area of prior acetabular grafting was evaluated to quantify the MSC content of the graft as well as to perform a histologic analysis of the bone. They found that the cohort of allograft with bone marrow aspirate had a significantly higher concentration of MSCs and demonstrated increased new bone formation compared with the other 2 groups. They concluded that allograft combined with bone marrow aspirate is a reasonable alternative to autograft and had similar osteogenic capability.
We currently perform grafting of acute fractures using allograft and bone marrow aspirate for cases with small to medium bone defects either in patients who have evidence of significant metabolic bone disease causing reduced quality of their autograft (eg, renal osteodystrophy, history of prolonged bisphosphonate use) or in patients who cannot undergo an iliac crest bone harvest or reamer irrigator aspirator (DePuy Synthes, West Chester, PA) procedure for various reasons ( Fig. 1 ). The preferred clinical scenarios when to use the technique of using bone marrow aspirate alone or mixed with allograft in acute fracture treatment are continuing to evolve.
Delayed or Nonunited Fractures
Most literature about bone marrow aspirate use in orthopedic trauma has focused on the successful treatment of delayed or nonunited fractures of the upper and lower extremities. However, the personality of the nonunion and host need to be evaluated thoroughly to determine the potential contribution of infection, metabolic abnormalities, mechanical instability, and status of the surrounding soft tissues before choosing the appropriate treatment. Bone marrow aspirate injection has been shown to have a potential role in the treatment of aseptic, atrophic nonunions with acceptable alignment and minimal gap, or displacement between fracture fragments.
Tibial nonunion treatment with bone marrow aspirate has been well-documented and found to be successful in 75% to 90% of reported tibial nonunion case series ( Fig. 2 ). Connolly and colleagues published one of the first studies using bone marrow grafting to treat delayed unions and nonunions percutaneously. Their cohort included 20 tibial nonunions that were treated initially with various techniques, including casting, external fixation, and intramedullary nailing. A vast majority of cases were open fractures and one-half of their patients had evidence of infection after the index procedure. Bone marrow was aspirated from the posterior iliac crest a mean of 14.3 months from initial injury. It was obtained in small aliquots and not concentrated. The bone marrow aspirate was injected typically into the posterolateral aspect of the tibial nonunion. Additionally, the bone marrow aspirate was not the sole treatment; 50% of patients had intramedullary nailing performed at the time of injection. They reported a 90% union rate (18/20) and advocated for the use of bone marrow aspirate as a viable alternative to autologous grafting.
Hernigou and colleagues reported their results of 60 tibial nonunion patients treated exclusively with BMAC. The location of the nonunion was heterogenous with a majority (52%; 31/60) located within the midshaft. Additionally, 48 of the nonunions were initially open fractures and treated with external fixation. The remaining cases were closed and treated in a cast. All nonunions were considered aseptic and atrophic. BMAC was injected after a mean of 8 months from initial injury. They found that 88% of their patients obtained union with no further intervention. Similar success was reported by Braly and colleagues, who described the outcome of 11 distal metadiaphyseal tibial nonunions treated exclusively with bone marrow aspirate. All 11 patients were treated initially with plates and screws and were considered to have aseptic and atrophic nonunions. The aspirate was harvested from the posterior iliac crest and it was not concentrated. The mean time from injury to bone marrow aspiration injection was 8 months. They found that 82% of patients (9/11) had successful union.
Bone marrow aspirate and BMAC injection has also been reported to treat nonunions in the femur, humerus, and ulna successfully. One of the earliest studies, by Garg and colleagues, reported their outcomes of tibial, humerus, and ulnar nonunions treated exclusively with bone marrow aspirate. They reported injecting 15 to 20 mL of posterior iliac crest bone marrow aspirate after a mean of 10 months from initial injury. They found that 85% (17/20) healed with no further intervention.
Unfortunately, no comparative studies of bone marrow aspirate injection with other nonunion treatment techniques, such as intramedullary nail dynamization, exchange nailing, or compressive plating with autologous bone grafting has been performed. However, analyzing studies with similar cohorts can extrapolate generalized comparisons between these different nonunion treatment techniques. Guimaraes and colleagues recently reported their outcomes of femoral shaft nonunions that were initially treated with locked intramedullary nailing who subsequently underwent percutaneous BMAC injection. They found that 50% of patients (8/16) had successful union of their fracture. This number compares favorably to previous publications, which reported a 58% to 76% success with intramedullary nail dynamization, and 53% to 88% success with exchange nailing. It should be noted that the bone marrow aspirate cohort from Guimaraes and associates performed their secondary procedure at a mean of 41 months after initial injury and treatment. Likewise, comparing common treatment options for tibial shaft nonunions shows that bone marrow aspirate compares favorably as well. As mentioned, tibial unions treated with bone marrow aspirate achieved union in 75% to 90% of cases, whereas nail dynamization or exchange nailing has been found to be successful in 83% and 90% of cases, respectfully. However, caution must be exercised when directly comparing the different studies as numerous unaccounted for cofounding variables and potentially dissimilar treatment groups likely exist. The point of comparing the success of the different treatment techniques is to show that equipoise exists and a large, randomized study is needed to definitively compare treatments and better delineate the nonunion characteristics for when percutaneous bone marrow aspirate injection is the most efficacious treatment option.
Surgical technique
Owing to the varied clinical scenarios often being addressed with the use of bone marrow aspiration and the lack of robust clinical trials, no standardized technique for bone marrow aspiration and grafting exists. However, multiple aspects of the procedure require specific attention to maximize efficacy and minimize clinical complications, including the location of aspiration, the neurovascular structures at risk, the size of syringe, the volume aspirated from each area, and the necessity to concentrate the aspirate. Each of these factors deserves individual consideration.
Bone Marrow Harvest Location
Bone marrow aspiration from the iliac crest is considered the gold standard location for orthopedic trauma and extremity surgery. However, controversy exists as to the best location within the iliac crest and whether other anatomic locations can provide an equivalent number or concentration of MSCs. Pierini and colleagues compared the concentration of MSCs between bone marrow aspirated from the anterior and posterior iliac crests in 22 patients. They found that the mean number of MSCs from the posterior iliac crest was 60% greater than from the anterior iliac crest. This difference was significant. They concluded that harvesting bone marrow from the posterior iliac crest was preferable. Although the posterior crest may maximize the number of harvested MSCs, several factors may influence the site of bone marrow harvest, including the positioning of the patient for the subsequent grafting procedure, the inability to position the patient prone, surgeon familiarity with iliac crest anatomy, and the use of a point-of-care bone marrow concentrator.
Besides the location within the crest for bone marrow aspiration, Hyer and colleagues investigated the MSC yield from different anatomic sites. Their group harvested bone marrow from the calcaneus, distal tibia, and iliac crest from the same patient. A total of 40 patients were enrolled in the study. They found that MSCs were obtained from all 3 sites, but in significantly different concentrations. The distal tibia (32.4 MSCs/mL) and calcaneus (7.1 MSCs/mL) had a 96.4% and 99.2%, respectively, lower concentration of MSCs compared with the iliac crest (898.4 MSCs/mL). This study confirmed that aspirate from the iliac crest is preferred.
Iliac Crest Osseous Anatomy
Although bone marrow aspirate from the posterior crest has been found to have the highest concentrate of MSCs, it can be harvested from any site along the crest from the anterior superior iliac spine (ASIS) to the posterior superior iliac spine. The distance from the ASIS to the posterior superior iliac spine along the iliac crest has been estimated to be approximately 24 cm. Bone marrow aspiration can occur at any point along this distance. However, depending on the location, there are significant changes in the iliac wing width, which increases the potential for cortical penetration as well as alters the proximity to neurovascular structures. Hernigou and colleagues, divided the length of the iliac crest into 6 different sections starting at the ASIS that were each approximately 4 cm in length. The sections were pie-shaped, with each dividing line converging at the center of the hip. The thickness between the inner and outer iliac wing cortices was calculated for each of the sections, and the ability of each section to accommodate a hypothetical 3-mm trocar was determined. They found that sections 1, 4, and 5 had the thinnest areas of ilium resulting in an increased risk of cortical penetration. They observed that sections 2 and 3 (4–12 cm posterior to the ASIS) as well as section 6 (from the posterior superior iliac spine to a point 4 cm anterior to it) are the most amenable for trocar placement given their increased iliac wing width.
Neurovascular Structures at Risk
The proximity of important neurovascular structures such as the lateral femoral cutaneous nerve, external iliac artery, sciatic nerve, and superior gluteal vessels to the iliac cortex changes depending on the location of trocar insertion. Hernigou and colleagues studied 48 hemipelvic CT angiography scans to estimate whether a hypothetical 10-cm length trocar placed with up to a 20° deviation in the insertion angle could result in damage to various neurovascular structures. Using a similar ilium dividing scheme of 6 pie-shaped sections (as described), they found that the risk to the external iliac artery was the greatest for the anterior sections. The external iliac artery was found to be within a hypothetical area of a misplaced trocar in 45.8% and 62.5% of cases for sections 1 and 2, respectively. This risk decreased substantially as the trocar insertion location was moved posteriorly to sections 3 (18.8%) and 4 (14.6%). Additionally, the authors found that the sciatic notch was located a mean of 70 mm from the iliac crest. They concluded that any trocar inserted in sections 5 and 6 to a depth of greater than 60 mm and only 5° of deviation risked cortical penetration and possible injury to the either the sciatic nerve or superior gluteal vessels.
Syringe Size
Although the size of syringe used for bone marrow aspiration seems trivial, 1 study has demonstrated that it has a significant impact on the number of obtained MSCs. Hernigou and colleagues compared the MSC concentration in aspirate using a 10- or 50-mL syringe. Thirty patients had bone marrow aspiration harvest performed on both iliac wings. One side had multiple aspirations at standardized sites of different volumes using a 10-mL syringe and the other side had a similar protocol but using a 50-mL syringe. The 10-mL syringe aspirated 1-, 2-, 4-, and 10-mL volumes, whereas the 50-mL syringe aspirated 5-, 10-, 20-, and 50-mL volumes. The aspirate from each site using the 2 different syringes were then analyzed and compared. They found that the concentration of MSCs was approximately 300% higher in the 10-mL syringe cohort for similar volume aspirations. They hypothesized that if the same force is used to withdrawal the syringe plunger, the smaller diameter plunger would create a higher negative pressure (pressure = force/area) resulting in a greater MSC harvest. Their recommendation was to use a smaller volume syringe and perform aspirations at multiple sites.
Aspirate Volume
The volume of aspirate has also been found to influence the harvested MSC concentration. Muschler and colleagues first studied the effect of aspiration volume by comparing 1-, 2-, and 4-mL bone marrow aspirate samples. They found that although the total number of MSCs increased with greater aspirated volumes, so did the quantity of diluting peripheral blood. The harvested MSC concentration decreased 28% (1451–1051 MSCs/mL) between 1- and 2-mL volumes and 38% (1418–882 MSCs/mL) between 2- and 4-mL aspiration volumes. They recommended limiting the aspiration volumes to less than 2 mL from 1 site unless intraoperative processing to concentrate the sample was to be used. Similar findings were reported by Hernigou and colleagues, who calculated the concentration of MSCs obtained after aspirating different volumes using a similar sized syringe from a single site. They found that an increased concentration of MSCs was obtained with smaller aspirated volumes. For example, when using a 10-mL syringe, the concentration of MSCs decreased a mean of 82% from their 1- and 10-mL aspirated samples, or from 2062 MSCs/mL to 376 MSCs/mL. Their conclusion was that an aspiration of 10% to 20% of the syringe volume was ideal. With increasing aspiration volumes, they postulated that the sample may be diluted with peripheral blood thereby decreasing the MSC concentration.
Concentrating Aspirate
There is no consensus about whether the bone marrow aspirate should be concentrated. Studies have shown that BMAC has a greater concentration of MSCs than unconcentrated aspirate. Hernigou and colleagues measured the concentration of MSCs in aspirate before and after concentration. In their technique, bone marrow aspirate was centrifuged to separate the heavier polynuclear cell layer, which was then isolated and analyzed. An initial aspirate volume of 300 mL was typically reduced to 60 mL after the concentrating process, and the aspirate concentration increased from 612 MSCs/mL before concentrating to 2579 MSCs/mL after concentrating. Furthermore, after injection of the BMAC into 60 tibial nonunions, they reported a significant difference in the MSC concentration and overall number of injected MSCs between the patients who subsequently achieved union and those that did not. The patients successfully treated with BMAC had a mean aspirate concentrate of 2835 MSCs/mL, whereas the persistent nonunion patients had 634 MSCs/mL. All of the patients who were unsuccessfully treated had aspirate containing less than 1000 MSCs/mL or fewer than 30,000 total MSCs injected. Age, sex, and medical comorbidities were not associated with treatment outcome. This study was instrumental in confirming that the concentration and total number of MSCs injected are 2 of the most important factors.
Although the concentration process can optimize the success of bone marrow aspirate grafting in treating nonunions, the need for additional equipment and possible increased operative time can create obstacles in implementing this step. A handful of published case series have described successful results when bone marrow aspirate concentrating was not performed. It must be emphasized that if a concentration process is not to be used, other steps to maximize the concentration and total number of obtained MSCs should be used such as aspirating multiple small volume aliquots with small syringes as well as obtaining aspirate from the posterior iliac crest.
Clinical Complication Rate
As mentioned, there are numerous neurovascular and visceral structures potentially at risk when performing an iliac crest bone marrow aspiration. This is true whether aspirating from the anterior or posterior iliac crest. However, the reported clinical morbidity is low. Bain surveyed members of the British Society of Haematology and collected information regarding the number or bone marrow biopsy procedures performed and the number of “biopsy-related misadventures.” She found that of approximately 55,000 procedures, only 26 adverse events were reported, an 0.05% incidence. The most frequent adverse event was hemorrhage and one death resulted from the procedure. Other authors have also described cases of large volume hemorrhage with bone marrow biopsy or aspiration. In the orthopedic literature, Hernigou and colleagues reviewed 523 bone marrow aspiration cases over a 16-year period and compared them with a separate cohort of 435 patients who underwent a standard iliac crest bone graft harvest. These investigators found that the rate of complication with bone marrow aspiration was approximately 7.6%. Complications included patients who experienced anemia not requiring transfusion, early or persistent pain at the site of aspiration, neuralgia, hematoma or seroma formation, superficial wound infection, ossification at the aspiration site, and harvest site fracture. Of note, none of the patients required surgical treatment of the complications and all were managed successfully with observation. In comparison, summation of the same complications in the iliac crest bone graft harvest cohort revealed a significantly higher rate of 80.2%. The authors concluded that the rate of complication with bone marrow aspirate was 10 times less frequent as with standard iliac crest bone graft harvest and is a relatively safe procedure.
One other hypothetical concern with the application of bone marrow aspirate into any extremity site is the possibility of an increased cancer risk. This theoretic risk is due to the ability of pluripotent stem cells to either differentiate into or stimulate a tumorigenic process. Hernigou and colleagues reviewed 1873 cases of patients treated with bone marrow aspirate to see if any tumors developed either locally or elsewhere and compared that with general population data. They found that no tumors formed at the treatment site and only 53 cancers developed elsewhere, which was lower than the incidence in the general population. They concluded that bone marrow aspirate injection did not increase a patient’s propensity to develop cancer.
Technique
Phillippe Hernigou and his group from Paris, France, have the largest published case volume regarding iliac crest bone marrow aspiration. Through meticulous research analyzing and improving their techniques, they have developed several methods to obtain the highest concentrate MSC aspirate possible. The typical steps of a bone marrow aspiration are as follows.
- •
The patient is positioned supine, lateral decubitus, or prone depending on the planned harvest site and patient position for subsequent injection.
- •
The affected extremity and one or both of the iliac crests are prepped and draped in the standard sterile fashion.
- •
A disposable or reusable trocar needle with a beveled edge (ie, Jamshidi or Lee-Lok needle; Fig. 3 ) is inserted through a small skin incision along the iliac crest.
