Orthobiologics in Foot and Ankle Applications


In recent years, as orthobiologic interventions for a variety of sports-related foot and ankle conditions have become more readily available, results are beginning to appear in the literature. The notion that we can affect the biology of the musculoskeletal tissues with these treatments and trigger a healing response is generally accepted. What conditions can be affected, how these orthobiologics are delivered, the mechanism of action, and the posttreatment modalities are being defined. Since these preparations can be delivered percutaneously, applied topically, and used in surgery as stand-alone or adjuvant approaches, there are often confounding variables that affect their assessment.

By the U.S. Food and Drug Administration (FDA) definition, biologics are biological products made from a variety of natural sources (human, animal, or microorganism) or the final products of advanced biotechnologies for medical applications ( www.Fda.gov/aboutfda ). The most common autologous orthobiologics, biologics for orthopaedic applications, in use today are platelet-rich plasma (PRP) and bone marrow concentrate (BMC). There is an increasing frequency of use of autologous fat tissue-derived treatment preparations, including stromal vascular fraction (SVF) and “fractured fat.” There also are allogeneic (donor tissue, cell, or growth factor) products obtained from donated placentas, amniotic membranes, or fluids. These products contain growth factors with or without viable cells, depending on the specific product. Because biologics are often a mixture of multiple biological components that are difficult to quantify in many cases, their mechanisms of action through cellular signaling may not be the same as that in vitro or are still unknown. Recombinant growth factors (e.g., recombinant human bone morphogenetic protein [rhBMP] and recombinant human platelet-derived growth factor [rhPDGF) and synthetic compounds (hyaluronic acid, calcium phosphate, or calcium sulfate) are regulated by the FDA as devices or drugs and are released with specific indications for orthobiologic use.

Often these biologic products are used in an “off-label” fashion for specific conditions “within the standard of care” (i.e., within the standards of appropriate medical practice without committing malpractice), even though they are not approved for that situation. Off-label use is not against the law, since the FDA does not regulate the practice of medicine. The companies that provide the products must comply with the FDA and cannot advertise or encourage an off-label use. Practitioners determine treatment on the basis of their clinical and scientific experience in accordance with their medical community’s reasonable standards, given the patient’s specific condition and setting.

From a regulatory perspective, human cells and tissues that comply with standards of the American Association of Tissue Banks are minimally manipulated in processing, and are marketed for homologous use are in compliance with US FDA regulation 21 CFR 1271 on human cells, tissues, and cellular and tissue products (HCT/P). Examples given in the regulation are “amniotic membrane when used alone or without added cells, bone, cartilage, cornea, fascia, ligament, pericardium, peripheral or umbilical cord blood stem cells for autologous use or use in a first or second degree blood relative), sclera, skin, tendon, vascular graft, heart valves, dura mater, reproductive cells, and tissues (e.g., semen, oocytes, embryos). On the other hand, if an HCT/P product is “dependent upon the metabolic activity of living cells for its primary function,” it is not regulated as a section 361 product according to an FDA guidance document on minimal manipulation and homologous use of HCT/Ps, and requires FDA clearance. HCT/Ps that are not minimally manipulated or used for homologous applications need FDA approval, since they are considered to be biological drugs. PRP and BMC are fully compliant with the existing regulatory framework established by the FDA, when used on an autologous basis and with homologous use. Interestingly SVF, although widely used in medical practice, is accepted as within the standard of care but falls outside the current standards established by the FDA for point-of-care (POC) therapeutic agents (e.g., 21 CFR 1271) as indicated in the FDA Guidance issued in late 2017. Placental products vary broadly and often fall into a gray zone, in which the tissue may be treated as an allograft but may only be marketed for specific uses, or the material contains living cells that require FDA clearance. The only noncontroversial placental-derived products on the market are sheet-forms used as a covering for wounds, or amniotic fluid without cells or micronized tissue. Neither of these types of products are “cleared” by the FDA, although the agency has taken action against companies who advertise the sheet-form for promoting healing or reducing pain. Use of the word “indication” implies a formal clearance by the FDA. Manufacturers can state what the material is used for, but not as an “indication for use” in the classic sense. Companies that produce these products must register and comply with current good manufacturing procedures (cGMP) and donor screening and tracking to qualify for regulatory clearance. In general, further definition and oversight for these products are expected to expand.

Underlying Mechanisms of Therapeutic Benefit

There are a number of biologically derived materials that have been used for regenerative therapy, including autogenous (e.g., PRP, BMC, and SVF) and allogeneic materials (e.g., placental/amniotic tissue–derived fluid and membranes).

A key distinction between PRP and BMC is the presence of progenitor cells in the bone marrow, including multipotent mesenchymal stromal cells (MMSC) formerly known as mesenchymal stem cells (MSC). In the lab, those cells can be isolated from bone marrow and demonstrate two critical behaviors: 1) they can differentiate into adult tissue cells under specific conditions (e.g., becoming chondrocytes, osteoblasts, and adipocytes); and 2) they can replicate without undergoing differentiation. Therefore, harvesting these stem cells does not create a deficiency given their ability to replenish their population. These cells are, in fact, present in all tissues of the body and produce trophic factors that help to coordinate/organize repair and regeneration of pathologic tissue through the production of growth factors and cytokines. The cells can immunomodulate and help control inflammation. They home to the site of injury and reside in the local injured tissues around blood vessels.

The distinction between the classical term “MSC” and the newer iteration “MMSC” must not be overlooked, as this distinction is central to understanding the origin of these cells and how they can be applied for therapeutic benefit. Historically, mesenchymal stem cells were aptly named because they were thought to derive from the connective tissue (stroma) of bone marrow and other tissue subtypes. The publication of a landmark article by Crisan and colleagues in 2008 marked a significant paradigm shift with the realization that MSCs are not stromal in nature, but rather derived from perivascular cells (pericytes). Indeed, Crisan’s data were so compelling that Arnold Caplan himself—a pioneer of the MSC—actually went so far as to conclude that all MSCs are pericytes.

This change in dogma is particularly significant because it has had such a profound effect upon the way we view the functionality of MMSCs and, consequently, the way we view their therapeutic capacity. With the characterization of MMSCs as pericytes, the therapeutic focus of these cells has shifted from their multipotent differentiation to their secretory ability, which leads to immunomodulatory and trophic effects. This shift has paved the way for a “new era” of cell-based orthobiologics to arise over the past decade, with applications ranging broadly from tendon healing to fracture union to cartilage regeneration.

The increased emphasis upon the secretory function of MMSCs is perhaps best illustrated by MMSC-derived exosomes, which have recently gained traction in biomedical research for their paracrine therapeutic potential. The exosome is a type of extracellular vesicle that is released by the MMSC and, much like its cellular counterpart, helps to maintain tissue homeostasis. While the precise mode of action is not completely understood, exosomes are believed to deliver mRNA, miRNA, and proteins to the target cell, thereby modulating the recipient cell function through complex cellular signaling cascades. Furthermore, exosomes have several possible advantages over traditional cell-based therapy: eliminated risk of mutation, decreased likelihood of immune rejection, improved tissue migration, and circulation through small blood vessels (e.g., capillaries).

The therapeutic possibilities for MMSC-derived exosomes are extensive, with applications in orthopaedic surgery and several other medical disciplines. With regard to musculoskeletal biology, in vitro and animal studies have proposed a role for exosomes in fracture healing, osteoporosis, cartilage regeneration, intervertebral disc degeneration, and osteonecrosis of the femoral head. Although donor-derived exosomes currently remain in the preclinical phase, they have shown great promise in these early investigations and may prove to be an important component of nonoperative therapeutics for a variety of orthopaedic conditions in the not-so-distant future.

Platelet-Rich Plasma

PRP, being derived from whole blood, contains very few circulating stem cells under normal conditions. What gives PRP a therapeutic potential in treating certain types of pathologies is the concentrated level of platelets, which release vesicles like exosomes, which might contribute to their therapeutic potential. These exosomes are filled with a wide variety of growth factors and other biologically active proteins. The plasma component in PRP contains several proteins that have been shown to have a beneficial effect, including interleukin 1-receptor antagonist protein (IL-1RAP or so-called IRAP ) and α-2- macroglobulin (α-2-M). IL-1RAP directly blocks the pro-inflammatory action of IL-1, a cytokine released by macrophages and neutrophils, in particular, at sites of inflammation, and is responsible for pain and sustaining a pro-inflammatory environment. At the site of injection, platelets release growth factors that are thought to stimulate/interact with adult tissue cells, enabling those cells to initiate a repair sequence. Releasing growth factors in a small-volume tissue (i.e., intratendinous) creates a stimulatory gradient of growth factors as they diffuse away from the site of implantation. As a result of this gradient, adult stem cells in the adjacent tissue may migrate to the treatment site. Thus, the benefit of PRP is an indirect one, since the growth factors act on adult tissue cells and stem cells present in the adjacent tissue to begin the reparative process.

Bone Marrow Concentrate

Bone marrow contains cells needed to continuously replenish the cellular components of blood: white blood cells (WBCs), red blood cells (RBCs), and platelets. These adult peripheral blood components are derived from hematopoietic stem cells (HSCs). In bone marrow, there are also MMSCs or MSCs, which don’t give rise to blood cells. They were initially identified in bone marrow stroma and believed to provide microenvironmental support to HSCs. In the 1960s, the pioneering work by Alexander Friedenstein proved that they are multipotent stem cells, like HSCs, being able to differentiate to osteoblasts, chondrocytes, and adipocytes.

In view of the high cellular content of BMC, which includes a wide variety of progenitor/stem cells, BMC provides a direct therapeutic benefit at the site of implantation. The plasma of the BMC preparation is a source of additional benefit, as mentioned for PRP, but with some differences. For example, BMC has been shown to contain 20-fold higher levels of IL-1RAP compared with patient-matched PRP preparations, while BMC will contain less a-2-M compared to PRP, since a-2-M is actively secreted into peripheral blood. One advantage of using BMC over PRP is that cellular activity in BMC treatment is initiated much more quickly, since the progenitor and stem cells are present in the injectate and don’t need to migrate to the treatment site, thereby shortening the time lag for initiation of repair.

Since MMSCs are able to differentiate into a wide variety of tissue cells, it was thought that MMSCs were therapeutically beneficial due to their ability to differentiate. However, after studies reported that MMSCs in most cases differentiated into adult tissue cells at a very low rate (< 2%–3%), research focus shifted to the fact that MMSCs are able to secrete a diverse set of cytokines and growth factors. Murphy and colleagues have reviewed the extensive list of biologically active molecules MMSCs produce when activated. For example, Murphy et al. indicated that there are five separate biochemicals that MMSCs secrete in a pro-inflammatory environment. This approach is called the paracrine effect and is similar to what platelets do when they degranulate and release growth factors. An important advantage of working with BMC is that the stem cells present do not disappear shortly after initiating their paracrine effect, but continue to respond to the environment, thereby modulating their response as tissue is being repaired over a longer period.

Dose–Response Relationships of PRP and BMC

As indicated above, the important components in PRP and BMC that are associated with a therapeutic benefit have been identified. Counting the number of platelets present in a PRP sample and assessing the levels of critical growth factors contained is possible, but very few of the clinical studies include these data. Furthermore, the use of RBC- and/or WBC-reduced PRP preparations creates more challenges. More work is needed to define these variables and their impact on healing.

On the other hand, mostly due to the pioneering work of Philippe Hernigou (Chief of Orthopaedic Surgery, Henri Mondor Hospital, East University [University of Paris], Créteil, France), BMC preparations have been characterized in terms of the levels of MMSCs present in the injectate. For example, Hernigou and his coworkers observed that study subjects receiving an autologous BMC treatment following a standard surgical repair of a full-thickness supraspinatus tendon tear were twice as likely to have an intact tendon at the 10-year milestone compared with patients receiving just the standard surgical repair. They showed that all of the BMC-treated patients who had more than 30,000 MMSCs (as determined by assessing the number of colony-forming unit fibroblasts [CFU-Fs] in the preparation) per injectate had intact tendons at the 10-year milestone, whereas those patients receiving fewer than 30,000 MMSCs all had tendon failures by the 10-year point. Hernigou and colleagues have also reported similar threshold effects in treating other pathologies, including tibial nonunion and avascular necrosis of the femoral head. The problem with trying to establish stem cell dosing guidelines is that each patient has their own level of adult stem cells in their bone marrow. Thus, an explicit knowledge of the number of MMSCs in a preparation is of retrospective benefit only due to the lag time in obtaining a CFU-F value, but it can help physicians refine how they aspirate bone marrow in order to maximize the potential to get a therapeutically beneficial level of adult stem cells. Dr. Hernigou’s group has published several papers on his bone marrow aspiration technique, including specific recommendations on the practical aspects of aspiration.

Adipose Stem Cells

As early as the turn of the 21st century, adipose tissue was identified as a potential autologous source of MMSCs. Much like bone marrow, adipose contains an easily isolated stroma-derived cell type that can differentiate into osteogenic, tenogenic, myogenic, and chondrogenic lineages. These unique properties led to the use of lipoaspiration as a means of harvesting adipose from fat-rich areas, with the end goal of applying this tissue for therapeutic benefit in other parts of the body. However, before the adipose tissue can be re-injected, it must first be processed in order to isolate an MMSC-rich product.

Several methods for processing adipose tissue have been proposed in the literature. Classically, the preferred technique has relied upon manipulation of the cells through enzymatic degradation. This method involves treating the lipoaspirate with collagenase to digest the extracellular matrix, then centrifuging the sample to obtain the product known as the stromal vascular fraction. Because the SVF contains a variety of cell types, cultures are required to isolate the adipose-derived stem cells (ASCs). The longer time required for culture has limited the clinical utility of this technique, since it is not feasible to perform during a single procedure in the operating room. Moreover, the FDA has yet to approve the use of enzymatically digested ASCs for musculoskeletal disorders in humans.

In an animal study, local injections of enzymatically digested ASCs significantly increased the tissue volume of the foot fat pad in rats. A few studies, all performed outside the US, have reported clinical outcomes following the injection of enzymatically digested ASCs or SVF for foot and ankle conditions. In a study of 49 patients with varus ankle osteoarthritis (OA) who underwent arthroscopic bone marrow stimulation with lateral calcaneal osteotomy, clinical outcome scores were significantly better in patients who received a supplemental injection of enzymatically digested ASCs compared with those who underwent the surgical procedure alone. Similarly, a study of 49 patients with osteochondral lesions of the talus revealed that bone marrow stimulation supplemented by SVF injections was associated with superior clinical and MRI outcomes compared with marrow stimulation alone. Further, SVF injections also showed benefit for Achilles tendinopathy in a randomized controlled trial, with significantly better outcome scores at 15 and 30 days postprocedure compared with PRP injections.

Recently, a novel processing method has been proposed that employs mechanical forces to create a reduced-volume preparation with viable progenitor cells, including ASCs. This technique, known by the trade name Lipogems (Lipogems International SpA, Milan, Italy), is a closed system that processes adipose tissue using microfragmentation as opposed to enzymatic digestion. Preclinical studies of microfragmented adipose tissue have demonstrated promising results when compared with the conventional processing methods. In two studies, microfragmentation produced a homogenous adipose tissue product with a significantly higher percentage of ASCs and a lower number of hematopoietic elements compared to enzymatic digestion. A separate study demonstrated that microfragmented adipose tissue releases more growth factors and cytokines involved in tissue repair and regeneration compared with enzymatically degraded adipose tissue. Finally, microfragmented lipoaspirate has also shown the ability to induce the production of connective tissue in a paracrine fashion, suggesting that this technique may have a therapeutic role in cartilage regeneration or repair.

To date, the authors are not aware of any clinical studies in the foot and ankle literature that have reported clinical outcomes following the injection of microfragmented fat. However, preliminary data from the knee literature do suggest that this may be a safe and effective therapeutic option for cartilage disorders of the foot and ankle. In a small study of 17 patients with knee OA, microfragmented fat injections were associated with no adverse events and a significant improvement in Knee Society Scores at 6 weeks and 12 months. Similarly, a larger study showed improvements in both International Knee Society scores and Visual Analogue Scale (VAS) scores in 52 patients with knee OA who underwent microfragmented fat injection plus arthroscopic debridement. Given the lack of robust clinical data in the literature, a prospective randomized controlled trial has recently been initiated to compare the efficacy of microfragmented fat and hyaluronic acid for mild-to-moderate knee OA. Further studies are necessary for foot and ankle applications before conclusions may be drawn regarding the utility of microfragmented fat injections in this patient population. Also, regulatory issues on the technique and technology are being clarified.

Placental/Amniotic Products

The human placenta supports fetal development and has been used since 1910 as a source of allogeneic tissue to assist healing. The placenta is delivered after a birth and consists of the chorion, the amnion, the umbilical cord, and the amniotic sac. These tissues are rich in cells and growth factors; they can be donated by consenting mothers and tested for viruses (human immunodeficiency virus, hepatitis, cytomegalovirus, human T-lymphotropic virus), then processed according to standards established by the American Association of Tissue Banks. Placental tissue allografts have a role in the healing of conjunctival lesions, ulcers, and wounds.

Currently, placental or amniotic-derived tissue products work as a delivery system of growth factors and as a scaffold that play a unique role in wound healing. In some products the presence of cells may provide additional benefit. Also, the cells may not be fully immune-privileged and may incite a host response. While the allograft material for creating patches (nonmicronized) from placental tissues contains viable stem cells and adult tissue cells, most products that are FDA-compliant are those processed using techniques that kill any viable cells present in the tissue. Both the growth factors and the matrix itself are thought to be responsible for the therapeutic benefit of using placental tissue–derived membrane patches in wound healing. It is thought that the growth factors migrate from the tissue implant into the underlying wound bed to activate local stem and adult tissue cells residing there. These products also act as a physical barrier, helping to reduce the chance of infection. Finally, the matrix of the tissue has been shown to promote the activity of cells that have migrated into the tissue implant itself. This in-migration is thought to promote the melding of the tissue patch with the wound bed, thereby aiding in re-epithelization of the wound.

Cellular products such as the Pluristem allogeneic cell (Pluristem Therapeutics Inc., Haifa, Israel) can be harvested and expanded for use as growth factor delivery agents. Studies have been initiated to establish a role for these cells in orthopaedics. Currently, the use of this biologic in the musculoskeletal system is still investigational in the United States. Another cellular product associated with tissues that contain growth factors is available, Grafix (Osiris Therapeutics, Inc., Columbia, Maryland). Currently it is in Phase 2/3 clinical studies for approval by the FDA as a biological drug product, per an agreement reached between Osiris and the FDA in 2015. This human-derived, viable wound matrix, composed of cryopreserved amniotic membrane, was applied weekly to treat diabetic foot ulcers (n = 50), resulting in a wound closure rate of 62.0% at 12 weeks, compared to a 21.3% closure rate with standard of care (n = 47).

Amniotic fluid also has been found to contain MMSCs, but applications of these cells in healing are largely experimental. There is debate about the viability of the cells in vivo and their engraftment into local tissues. However, these fluids may contain growth factors that may provide symptomatic relief in musculoskeletal conditions. In one study of 44 patients with plantar fasciitis and Achilles tendinopathy, good resolution of symptoms was demonstrated using an injection of cryopreserved amniotic membrane and amniotic fluid–derived cells.

Umbilical cord–derived cells are currently banked as cord blood but do not yet have an indicated use in orthopaedics. The umbilical cord consists of an umbilical vein and two arteries surrounded by a collagenous, hyaluronan matrix called Wharton jelly. The cord components can be processed and used to produce growth factors, but there are no studies to support their efficacy.

Lyophilized human umbilical cord is decellularized and may be a source of growth factors for certain applications.

Most of the placental products available are decellularized and either dehydrated or cryopreserved. They are typically derived from the chorion, which is on the maternal side of the placenta, or the amnion, which is on the fetal side. These tissues are immune-privileged and, with proper processing, can be used as sheets or particles of growth factors. In general, the dehydrated tissues have been shown to decrease tissue fibrosis, enhance soft-tissue healing, and modulate immune function.

In addition to their widespread use for corneal lesions, placental products have also shown benefit in the musculoskeletal system for wound healing.

The various products are not equal due to procurement, processing, and preservation procedures. They are not all well studied, but in general have been used for diabetic ulcers and to promote healing. The best-studied products are EpiFix and AmnioFix by MiMedx (MiMedx Group, Inc., Marietta, GA), which has characterized over 226 growth factors.

AmnioFix is one of the new regenerative medicine advanced therapy (RMAT) category products that the FDA created after the 21st Century Cures Act, so it is in an investigational new drug clinical study premarket approval program. AmnioFix and EpiFix have been used in clinical studies and have shown consistent safety and efficacy in diabetic wounds, vascular wounds, and plantar fasciitis. A case study of 22 patients has also shown benefit in Achilles tendonitis. The benefit of this path of delivery is that it is relatively cost effective and requires minimal processing by the physician to inject. Future studies may show benefits in a greater number of indications including articular cartilage defects and osteoarthritis.

A similar product by Amniox Medical, Inc., a TissueTech company in Miami, Florida, has been evaluated in a retrospective study of 124 foot and ankle patients for several conditions including open repair of peroneal and Achilles tendinopathy with an overall wound complication rate of 5.64% and a reoperation rate of 1.6% (2/124).

In a randomized, controlled, double-blind pilot study of 23 patients with plantar fasciitis treated with either steroid or cryopreserved human amniotic membrane, the amniotic membrane was found to be safe and effective relative to the steroid injection.

Recombinant Growth Factors

Across the field of orthopedics, recombinant growth factors such as BMP and PDGF have become increasingly popular adjuncts to promote bone healing. Although BMP is only FDA-approved for lumbar fusions and tibial nonunions, there is a mounting body of evidence to support its use in foot and ankle surgery. Several articles in the literature have reported increased fusion rates with the addition of BMP-2 and BMP-7 in high-risk patients undergoing arthrodesis of the ankle or hindfoot. However, given the retrospective design of these studies, the overall level of evidence regarding the efficacy of BMP in foot and ankle fusion remains low. Randomized prospective data must be presented if BMP is to become ubiquitous in this patient population.

In contrast to BMP, multiple randomized prospective trials have compared PDGF to autogenous bone graft (ABG); the current gold standard in foot and ankle fusion. A review article by Sun and colleagues aggregated clinical, radiographic, and safety data from the 634 patients included in these three high-quality studies. No significant difference was observed between PDGF and ABG with regard to safety or radiographic outcomes. Comparative analysis of the clinical data also yielded similar outcomes between groups, with the exception of the long-term SF-12 Physical Component scores, which were superior in the ABG cohort. Based upon these results, the authors concluded that PDGF is a viable alternative to ABG in foot and ankle fusion.

In addition to the value of PDGF as an adjunct to enhance bony fusion, emerging studies have begun to expand the scope of this growth factor in foot and ankle surgery. Novel arthroscopic procedures have utilized PDGF in the treatment of talar osteochondral defect (OCD) lesions, with encouraging preliminary results in a proof-of-concept trial and a case report of a professional rugby player. Likewise, a recent animal study used a chicken foot model to propose a role for PDGF as a bioinductive platform for flexor tendon repair. These innovative applications—including arthroscopic and soft-tissue procedures—suggest that growth factors may prove to be extremely versatile resources in the future of foot and ankle sports medicine.


Calcium Sulfate and Calcium Phosphate

In foot and ankle surgery, calcium sulfate and calcium phosphate have long been used to augment the fixation of displaced calcaneus fractures. In a randomized prospective study of 90 patients with displaced intra-articular calcaneus fractures, percutaneous screw fixation with calcium sulfate cement grafting was associated with less blood loss, better range of motion, lower infection rate, and superior functional outcome scores compared with open reduction and internal plate fixation. A separate prospective study reported a shorter time to union for displaced intra-articular calcaneus fractures augmented with demineralized bone matrix calcium sulfate compared to those treated with open reduction and internal fixation alone. Similarly, there is a substantial body of literature supporting the use of calcium phosphate bone cement to augment calcaneal fixation constructs. With the addition of calcium phosphate bone cement, studies have demonstrated superior compressive strength and earlier return to full weight bearing postoperatively, with comparable clinical outcome scores to conventional bone grafting.

Aside from their well-documented utility in calcaneus fracture fixation, calcium phosphate and calcium sulfate have also been used to improve screw purchase in a biomechanical model of osteoporotic ankles. Additionally, there may be a role for calcium phosphate cement in patients with symptomatic bone cysts of the foot and ankle. In a case series of 16 young athletes with symptomatic unicameral cysts of the calcaneus, endoscopic curettage with injection of calcium phosphate bone substitute was associated with significantly improved pain and functional outcome scores, as well as an early return to play (< 8 weeks) and zero instances of recurrence or pathologic fracture at 2 years postoperatively. However, given the paucity of literature investigating the use of calcium sulfate and calcium phosphate outside the realm of calcaneus fracture, further high-quality studies are needed to establish them as a mainstream therapeutic option in foot and ankle sports medicine.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here