➢ Hemostatic agents are widely utilized in spine surgery to reduce blood loss and the need for allogenic blood transfusions.
➢ Several hemostatic agents are available for the spine surgeon and can be categorized as active agents (via activation of the coagulation cascade) and passive agents (via contact activation and promotion of platelet activation).
➢ Intravenous administration of tranexamic acid has been consistently demonstrated in numerous clinical trials to reduce blood loss in spine surgery.
➢ Several case reports and case series have suggested that the use of hemostatic agents may cause adverse events in the perioperative period. Understanding these associated adverse events and the means of avoiding them may greatly improve patient safety.
➢ There is insufficient and inconclusive evidence to advocate for or against the use of many of the available hemostatic agents. In turn, there is a need for more clinical trials evaluating the effectiveness and safety of hemostatic agents in spine surgery.
Blood loss during spine surgery is one of the most difficult challenges facing the surgeon, with effects ranging from decreased visualization of the operative field to life-threatening hemorrhage. Specific considerations for intraoperative bleeding include the potential for development of acute stroke from vertebral artery transection to spinal cord ischemia. Additionally, the need for perioperative transfusion of blood products has been associated with longer hospital stays and increased cost1. Management of blood loss begins in the preoperative period with discontinuation of medications known to increase the risk of bleeding, including aspirin, other nonsteroidal anti-inflammatory medications, and anticoagulants. The reasons for taking these medications and their discontinuation should be discussed with both the patient and the patient’s primary care physician or cardiologist. At the time of the operation, a collaborative effort between the anesthesiologist and the surgeon is undertaken to manage appropriate hemostasis. This collaboration includes blood pressure management by the anesthesia team as well as meticulous dissection and management of hemostasis by the surgical team. Surgical hemostasis is the product of a combination of strict subperiosteal dissection, liberal use of electrocautery, and utilization of hemostatic agents. This collaborative care becomes especially important during the course of care for the patient with multiple medical comorbidities.
Although hemostatic agents remain an invaluable resource for the surgical team, numerous case reports and case series have suggested that the use of such agents may cause adverse events in the perioperative period2-14. Understanding the adverse events associated with these hemostatic agents may greatly improve patient safety. This article provides a comprehensive review of the most commonly utilized intraoperative modalities.
Topical Hemostatic Agents
The use of topical hemostatic agents remains the frontline treatment for intraoperative blood loss in major and minor spine surgery. Multiple agents are available for use and can be broadly divided into active and passive agents. Active agents have innate biological activity and promote fibrin clot formation through activation of the coagulation cascade. These agents include thrombin and products combining thrombin with a passive hemostatic agent. Passive hemostatic agents exert their effects through contact activation and promotion of platelet activation. These products include collagen-based, cellulose-based, and gelatin-based products.
Bone wax is one of the oldest hemostatic agents available. The earliest reports of the use of wax for the purposes of the hemostasis of bleeding bone surfaces date to 175715, although the first report of the use of a beeswax-based formula was that of Horsley in 189216. Bone wax is composed of beeswax and petroleum jelly, and its hemostatic effect is based primarily on mechanical intercalation within the vascularized trabecular bone. To date, there have been no clinical trials that we are aware of that have evaluated the effectiveness of bone wax as a hemostatic agent in spine surgery.
In several animal models, bone wax has been demonstrated to inhibit bone healing and osteogenesis17,18. The use of bone wax was found to inhibit healing in a rat model of tibial fracture17. Its use has similarly been found to induce resorption of cancellous bone and inhibit osteogenesis in goats that have undergone a sternotomy procedure18. Even though, to our knowledge, there have been no clinical studies evaluating the effects of bone wax in arthrodesis or spine surgery, the results of these nonclinical studies suggest that bone wax should not be applied to arthrodesis sites.
Although largely free from substantial adverse events, bone wax has been associated with allergic reaction, intracranial granuloma, epistaxis, and granulomatous infection2-4 (Table I). There is a reported case of iatrogenic quadriplegia following bone-wax use that was postulated to be a result of compression from the bone wax or epidural bleeding by detachment of epidural veins5. We recommend that only a minimal amount of bone wax be used and that excess wax be removed.
Gelatin Sponge and Powder
Absorbable gelatin sponges were first introduced in the 1940s for use as hemostatic agents during surgery19. Gelfoam (Pfizer, New York, NY) is made from animal-skin gelatin that is processed into sponge form. When soaked in thrombin, Gelfoam acts via a swelling mechanism, producing a compressive force at the site of application as well as functioning to activate the coagulation cascade by cleaving fibrinogen into an active fibrin product. This product is also available with thrombin included in the packaging (Gelfoam Plus; Baxter, Deerfield, Illinois). Gelfoam sponges are often used to fill the blood-soaked laminectomy field in order to achieve hemostasis. In contrast to bone wax, Gelfoam has not been shown to inhibit osteogenesis20. To date, there have been no clinical studies that we are aware of that have evaluated the effectiveness of Gelfoam as a hemostatic agent in spine surgery.
While Gelfoam is widely used in spine surgery, there have been reports of adverse events, largely as a result of mass effect6,7 (Table I). In one case, authors reported worsening cervical myelopathy following posterior cervical decompression for moderate myeloradiculopathy caused by an engorged Gelfoam sponge that was found to compress the underlying dura6. Another reported case involved quadriparesis following anterior cervical corpectomy with strut grafting, in which a swollen mass of Gelfoam was found between the fibular strut graft and the spinal cord7.
In contrast to gelatin sponges, Surgifoam powder (Ethicon, Bridgewater, New Jersey) comes in a paste form that can be spread or shaped to conform to irregular surfaces. As with Gelfoam, the efficacy of Surgifoam as a hemostatic agent has not been evaluated in clinical trials. Unlike the properties of gelatin sponges, the properties of Surgifoam are such that creation of mass effect is unlikely. To date, there have been no case reports of neurological complications associated with Surgifoam powder, to our knowledge. However, one case report illustrated that unintended intravascular introduction of Surgifoam can result in multiple thromboemboli, which in the report led to right ventricular failure and disseminated intravascular coagulation in an eighteen-year-old female undergoing posterior spinal arthrodesis for the treatment of scoliosis8. Therefore, Surgifoam powder should be removed via irrigation and suction prior to closure21.
First approved for use in the U.S. by the Food and Drug Administration (FDA) in 199922, FloSeal (Baxter) comprises two major components: a bovine-derived gelatin matrix and a human or bovine-derived thrombin-powder component that is mixed as a semiliquid in a calcium chloride solution23. FloSeal’s composition and mechanism of action are similar to those of Gelfoam Plus, combining a physical compression effect, a matrix for fibrin clot formation, and the biochemical reaction triggered by the thrombin component, which initiates the innate coagulation cascade. FloSeal’s gelatin matrix has a consistency different from that of other agents, such as Gelfoam, and contains microparticles that incorporate into the clot formed. These microparticles are eventually cleared by the immune system over a six to eight-week period.
One landmark study of FloSeal (branded as Proceed at the time) was conducted in 2001 by Renkens et al.24. In the study, 127 spinal surgery patients were prospectively randomized to receive either FloSeal or Gelfoam-thrombin for monitored use on bleeding sites with evaluation at one, two, three, six, and ten minutes after application24. Hemostasis was defined as the stopping of bleeding and oozing within ten minutes. FloSeal demonstrated 98% efficacy at ten minutes compared with 90% efficacy for the Gelfoam-thrombin control, and hemostasis was achieved within three minutes in 97% of patients who received FloSeal compared with just 71% of patients who received the Gelfoam-thrombin control24. Another prospective study of FloSeal, in 214 patients undergoing cranial, craniospinal, or spinal procedures, found that hemostasis was achieved in all but eleven of the patients in less than three minutes; of those eleven cases, reapplication achieved effective hemostasis in all but four25.
The thrombin component of FloSeal is derived from human plasma, and the gelatin matrix is derived from dermis of cattle raised in the U.S. The origin of the thrombin was cause for concern because of the risk for transmission of infectious entities, and this concern initially delayed the FDA’s release of the agent. To date, there have been no reported cases of such infectious transmission, to our knowledge.
Evicel (Ethicon) and Tisseel (Baxter) are fibrin sealants. They consist of two components: human fibrinogen and thrombin. Evicel is a tranexamic acid-free formulation of Quixil (Omrix Biopharmaceuticals, Brussels, Belgium) or Crosseal (Omrix Biopharmaceuticals), formulated this way because tranexamic acid may act as a potential neurotoxin when applied topically. Additionally, Evicel comprises 100% human products. While Tisseel is composed of the same fundamental elements of thrombin and fibrinogen, it also contains a synthetic aprotinin that may result in a hypersensitivity reaction. When applied to a source of bleeding, thrombin cleaves fibrinogen to create a fibrin fiber lattice structure for coagulation and directly activates factors XIII and XIIIa, stabilizing the clot. While Evicel contains a relatively large amount of factor XIII, levels of the protein are undetectable in Tisseel26. As postulated to be a result of this, fibrin clots formed with Evicel demonstrated significantly greater resistance to stretching and tensile strength when compared with clots formed with Tisseel (p < 0.001)27. The clot-formation kinetics of Evicel, as assessed with thromboelastography, were found to be faster than those of Tisseel; however, this difference did not reach significance27. Nonetheless, other clot properties—for example, the time to initial clot formation, clot strength, and elasticity constant—have not been found to differ significantly between the two fibrin sealants27. These mechanical differences have been shown to impact efficacy, as demonstrated in a rabbit model of partial resection of the liver28. Evicel yielded a shorter duration of bleeding with use of a lower volume of sealant to achieve hemostasis when compared with Tisseel. To date, there have been no large studies that we are aware of comparing efficacy between fibrin sealants used in the spine.
Oxidized Regenerated Cellulose
Oxidized regenerated cellulose, first described in 1947 for the treatment of existaxis, can be utilized as a topical hemostatic agent in spinal surgery29. An advantage of oxidized regenerated cellulose-based products is the ability to conform the products to the recipient surface, allowing for a favorable three-dimensional structure for clot organization. Additionally, oxidized regenerated cellulose-based products have a direct mechanism of action via interaction with proteins, platelets, and intrinsic and extrinsic pathway activation30. To our knowledge, there have been no published clinical trials to date that have evaluated the effectiveness of oxidized regenerated cellulose-based products.
Surgicel (Ethicon), an oxidized regenerated cellulose-based product, has been associated with a number of possible neurologic complications. There are several reported cases of paraplegia following thoracic and lumbar surgeries resulting from the compressive effects of Surgicel on the underlying dura9-11. The removal of Surgicel after hemostasis is therefore recommended in order to avoid complications related to mass effect10.
Antifibrinolytic agents have been found to effectively decrease blood loss during cardiovascular surgery and in some orthopaedic procedures31,32. In spine surgery, tranexamic acid and epsilon-aminocaproic acid have been shown to be effective at minimizing intraoperative blood loss, with no substantial morbidity or increased rate of thromboembolic events31. Additionally, aprotinin has been shown to be effective in reducing blood loss during pediatric scoliosis surgery31,32.
Aprotinin is a serine-protease inhibitor extracted from bovine lung tissues that works to decrease bleeding by acting on the coagulation cascade, antifibrinolytic pathway, inflammatory response, and the platelet membrane. Aprotinin affects the coagulation cascade by inhibiting several anticoagulatory enzymes, including kallikrein, plasmin, plasminogen activator, and thrombin. Aprotinin has also been shown to inhibit platelet aggregation and thrombosis. High-dose aprotinin has been found to decrease blood loss and transfusion requirements following major cardiac surgery33. Further, aprotinin has been utilized in other surgical procedures, including liver transplantation and vascular surgery, as well as in various orthopaedic cases34-37.
A placebo-controlled randomized study by Lentschener et al. demonstrated significantly lower combined intraoperative and postoperative blood loss (1935 ± 873 mL compared with 2839 ± 993 mL; p = 0.007), lower combined intraoperative and postoperative transfusion requirements (forty-two units compared with ninety-five units; p = 0.001), and a decreased percentage of patients requiring transfusion (40% compared with 81%; p = 0.02) of those receiving aprotinin38. In addition, intraoperative activation of fibrinolysis (assessed through postoperative fibrinogen and D-dimer levels) was significantly less pronounced in the aprotinin group than in the placebo group (p < 0.0001)38. A prospective placebo-controlled randomized controlled trial evaluating aprotinin in pediatric patients undergoing spine surgery also found that aprotinin significantly decreased intraoperative blood loss (545 ± 311.9 mL compared with 929.9 ± 771.7 mL; p = 0.039) and transfusion requirements (1.1 ± 1.0 units compared with 2.2 ± 1.7 units; p = 0.016)39. Similarly, in a retrospective review of the cases of forty-one patients receiving aprotinin and forty-one controls, Tayyab et al. showed decreased blood loss (1324 ± 608 mL compared with 2113 ± 1296 mL; p = 0.005) and decreased mean transfusion requirements (2.73 ± 2.45 units compared with 5.02 ± 4.02 units; p = 0.003) in the patients receiving aprotinin40. Further, a 2008 meta-analysis demonstrated the efficacy of aprotinin as a hemostatic agent in spine surgery41.
Aprotinin, while effective in reducing blood loss and transfusion requirements, has been associated with a number of adverse outcomes, including perioperative myocardial infarction, stroke, anaphylaxis, and acute renal failure12,13 (Table I). Authors of a retrospective matched cohort study found that while aprotinin reduced blood loss and transfusion requirements in adult patients with spinal deformity undergoing spinal arthrodesis, four (10%) of the forty patients receiving the drug developed acute renal failure and one developed symptomatic deep vein thrombosis4. All of the patients with acute renal failure required dialysis, with one requiring chronic dialysis14. Subsequently, in late 2007, the FDA suspended the use of aprotinin except for investigational use14.
Epsilon-aminocaproic acid is a synthetic antifibrinolytic agent that acts by competitively inhibiting plasminogen activation, thereby stabilizing fibrin clots42-44. A prospective randomized controlled trial found a significant decrease in perioperative blood loss (1391 ± 212 mL compared with 1716 ± 513 mL; p = 0.036) and the need for postoperative autologous blood transfusion (0.11 ± 0.3 units compared with 0.7 ± 0.6 units; p = 0.002) in patients undergoing posterior spinal arthrodesis for idiopathic scoliosis45. There were no perioperative or postoperative thromboembolic events45. Authors of another study evaluated the use of epsilon-aminocaproic acid in same-day anterior and posterior spinal arthrodeses for the treatment of adolescent idiopathic scoliosis and found that aminocaproic acid was effective in reducing total perioperative blood loss (3442.8 ± 1344 mL for controls, 2089.8 ± 684 mL for patients managed with epsilon-aminocaproic acid for the posterior portion of the operation only, and 2098.4 ± 1061.6 mL for patients managed with epsilon-aminocaproic acid for the posterior and anterior portions of the operation; p = 0.0005) and transfusion requirements (1537.1 ± 905.1 mL for controls, 485.2 ± 349.8 mL for patients managed with epsilon-aminocaproic acid for the posterior portion of the operation only, and 531.5 ± 510.5 mL for patients managed with epsilon-aminocaproic acid for the posterior and anterior portions of the operation; p = 0.0001)46. As noted by the authors, analysis of the data suggests that the main hemostatic effect of epsilon-aminocaproic acid is through the reduction of estimated intraoperative blood loss during posterior spinal arthrodesis with segmental instrumentation, while its effect during anterior spinal arthrodesis was minor. In turn, the authors concluded that epsilon-aminocaproic acid is effective in anterior and posterior spinal arthrodesis surgeries for scoliosis but should be used only in posterior spinal arthrodesis with segmental instrumentation.
In contrast, Berenholtz et al., in a placebo-controlled randomized trial that evaluated the use of epsilon-aminocaproic acid in adult patients undergoing major spinal surgery, found that the mean total (intraoperative and postoperative) number of allogeneic red blood cell units transfused (5.9 ± 4.7 units compared with 6.9 ± 5.4 units; p = 0.18) and the mean estimated intraoperative blood loss (2938 ± 2315 mL compared with 3273 ± 2195 mL; p = 0.32) were not significantly different between the groups47. Notably, however, the mean number of postoperative red blood cell units transfused in the epsilon-aminocaproic acid group was significantly lower compared with the group that received the placebo (2.0 ± 1.8 units compared with 2.8 ± 2.8 units; p = 0.03). The rate of thromboembolic events was similar between the two groups (2.2% in the epsilon-aminocaproic acid group compared with 6.6% in the group that received the placebo; p = 0.15)47.
Tranexamic acid is a synthetic antifibrinolytic drug that, although considered similar to epsilon-aminocaproic acid in action, has been demonstrated to have tenfold higher potency48. The mechanism of action for tranexamic acid, similar to that of epsilon-aminocaproic acid, is competitive blockade of the lysine-binding sites of plasminogen, plasmin, and tissue plasminogen activator. It has both surgical and nonsurgical applications48. Nonsurgical uses of tranexamic acid include the management of bleeding associated with leukemia, ocular bleeding, sustained hemoptysis, and severe menorrhagia49, while the initial surgical uses of tranexamic acid involved high-risk cardiac surgery48. In several studies, tranexamic acid has been shown to reduce blood transfusion requirements and costs of cardiac surgery50,51. Hiippala et al. also demonstrated that tranexamic acid reduced estimated total blood loss as well as the mean number of postoperative blood transfusions for patients undergoing total knee arthroplasty52.
A recent retrospective review evaluated the effectiveness of tranexamic acid and aprotinin in patients who underwent surgery (anterior, posterior, and combined anterior and posterior) for the correction of degenerative scoliosis53. Patients receiving tranexamic acid or aprotinin had significantly less blood loss than those receiving no antifibrinolytic agents (738 mL in the tranexamic acid group, 710 mL in the aprotinin group, and 972 mL in the control group; p = 0.037). The difference in average blood loss between patients receiving either of the hemostatic agents was not significant (p = 0.085).
A prospective randomized placebo-controlled trial evaluating the hemostatic effect of tranexamic acid in cervical laminoplasty for cervical myelopathy found that tranexamic acid significantly reduced total blood loss compared with a placebo control54. Specifically, while intraoperative blood loss was similar between the tranexamic acid and placebo groups, the use of tranexamic acid was found to significantly reduce postoperative blood loss (p < 0.01) and thus total blood loss (353.9 ± 60.8 mL in the control group compared with 264.1 ± 75.1 mL in the tranexamic acid group; p < 0.01). No thromboembolic events or complications occurred in either group.
A meta-analysis of six prospective randomized placebo-controlled trials evaluating tranexamic acid found that administration of the agent significantly reduced both total blood loss (1288 ± 507 mL compared with 765 ± 239 mL; p < 0.001) and the volume of allogenic blood transfused per patient (258 ± 246 mL compared with 97.7 ± 192.1 mL; p < 0.001)55. There were no reported cases of deep vein thrombosis or pulmonary embolism.
Other Hemostatic Agents
Recombinant Activated Factor VII
Recombinant activated factor VII (rFVIIa) is indicated for the treatment of bleeding episodes and the reduction of blood loss during surgery in hemophilia patients with inhibitors to factors VIII or IX. Recombinant activated factor VII works to achieve hemostasis by enhancing thrombin generation on activated platelets, thereby facilitating the formation of a stabilized fibrin hemostatic plug that is resistant to premature fibrinolysis56. The efficacy of rFVIIa has been demonstrated in a number of situations, including traumatic brain injury, retroperitoneal bleeding, and postpartum hemorrhage56-58. However, there are limited data on the efficacy of rFVIIa in spinal and orthopaedic procedures in patients who do not have hemophilia.
Kaw et al. published a case series of four patients who had developed severe bleeding and coagulopathy while undergoing major spinal surgery59. Treatment with rFVIIa led to an improvement (decrease) in prothrombin time and partial thromboplastin time. Along with standard hemostatic techniques, this treatment helped to stem intraoperative blood loss. There were no thromboembolic events in any of the four patients.
More recently, Sachs et al. reported on forty-nine patients undergoing posterior arthrodesis of three or more thoracic or lumbar vertebral segments60. Patients were randomized intraoperatively after losing 10% of their estimated total blood volume. Thirteen patients were randomized to receive three doses of the placebo at two-hour intervals, while thirty-six patients were randomized to receive 30, 60, or 120 µg/kg rFVIIa. The authors showed a significant decrease in the mean intraoperative blood loss (adjusted for the number of levels fused, duration of surgery, and preoperative blood volume) in the patients receiving rFVIIa compared with those receiving the placebo (2536 mL in the placebo group [95% confidence interval (CI), 1869 to 3441 mL] compared with 1120 mL in the group receiving 30 µg/kg rFVIIa [95% CI, 647 to 1938 mL], p = 0.001; and 400 mL in the group receiving 60 µg/kg rFVIIa [95% CI, 151 to 1059 mL] compared with 824 mL in the group receiving 120 µg/kg rFVIIa [95% CI, 435 to 1558 mL], p < 0.001). The mean adjusted total transfusion volume was significantly reduced, by 81% in the group receiving 120 µg/kg and 95% in the group receiving 60 µg/kg, with rFVIIa treatment. There was no significant difference in the rate of serious adverse events in either arm of the study.
Desmopressin (also referred to as DDAVP or deamino-8-D-arginine vasopressin), is a synthetic analog of the natural hormone L-arginine vasopressin61. Desmopressin primarily works by binding to V2 receptors in the renal collecting tubule epithelium, thereby limiting the amount of water excreted in the urine61. The exact mechanism as a hemostatic agent is unknown, although it is thought to increase the secretion of factor VIIIc and von Willebrand factor while also acting to enhance platelet adhesion to the blood vessel endothelium61. The use of desmopressin in scoliosis and other spine surgery has been well documented. In a randomized controlled trial (n = 35), Kobrinsky et al. demonstrated that desmopressin reduced blood loss by 32.5% (547 mL; 95% CI, 19 to 1075 mL; p = 0.015) and reduced the need for concentrated erythrocyte transfusions by 25.6% (0.86 unit; 95% CI, 0.08 to 1.65 units; p = 0.022)61.
In contrast, the authors of another randomized controlled trial found that blood loss per kilogram of body weight, blood loss per surgically treated spinal level, urinary output per kilogram of body weight, serum levels of fibrinogen, von Willebrand factor activity, tissue type plasminogen activator activity, and plasminogen activator inhibitor activity were not sensitive to the administration of desmopressin at any time interval during surgery or at twenty-four hours after surgery62.
Hemostatic agents have served a vital role in spine surgery since the advent of bone wax nearly 170 years ago. The exploration of the various ways of altering the coagulation process through both physical and chemical methods has resulted in multiple options for the surgeon to help decrease perioperative bleeding and thus transfusion requirements. Although there is anecdotal evidence supporting the use of topical agents, we did not find any randomized clinical studies evaluating the effectiveness of these topical agents in spine surgery and therefore do not have sufficient evidence to advocate for or against their use. While one randomized study demonstrated that FloSeal induced greater hemostasis when compared with gelatin-thrombin matrix sealants, there have been no randomized clinical trials that we are aware of evaluating the effectiveness of FloSeal compared with a naïve (untreated) or placebo control (Table II).
Results from several randomized clinical studies have suggested that perioperative administration of systemic antifibrinolytic compounds (epsilon-aminocaproic acid, rFVIIa, and tranexamic acid) decreases perioperative blood loss and the need for transfusion in spine surgery (grade of recommendation: A [Level-I studies with consistent findings]). Nevertheless, since these agents are systemically administered and therefore bear a systemic hemostatic effect, it is theoretically possible that utilization of the agents may be associated with an increased rate of postoperative thromboembolic events. Further research is needed in order to truly explore the safety of routine use of antifibrinolytic agents and to determine indications for their use in spine surgery (Table II).
Overall, while much data exist regarding the use of individual hemostatic agents in surgery, there is a general lack of data relating specifically to spine surgery. Although we as spine surgeons may continue to use these agents to aid in hemostasis, we should remain cognizant of the general paucity of large-scale randomized trials of these agents in the context of spine surgery.
Source of Funding: The authors did not receive grants or outside funding in support of their research for or preparation of this manuscript.
Investigation performed at the Leni and Peter W. May Department of Orthopaedics, Mount Sinai Medical Center, New York, NY
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. None of the authors, or their institution(s), have had any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, no author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
- Copyright © 2015 by The Journal of Bone and Joint Surgery, Incorporated