PERIOPERATIVE VOLUME AND BLOOD MANAGEMENT
Key Clinical Questions
What is the best perioperative volume and blood management in orthopedic patients?
Which evidence based strategies prevent orthopedic surgical site infections?
What are the best strategies for perioperative pain management in these patients?
What is the best way to diagnose and manage compartment syndrome?
What is the best way to recognize fat embolism syndrome?
Exposed cancellous bone presents a hemostatic challenge, and consequently, nearly all patients undergoing major orthopedic procedures have some degree of acute blood loss anemia. The estimated blood loss for a total hip replacement is approximately 3.2 units or 4.07 g of hemoglobin, while total knee replacement patients may lose 1 to 1.5 L of blood or 3.85 g of hemoglobin. Blood loss after bilateral or revision joint replacement may be significantly more.
Historically, 10% to 38% of total joint-replacement patients received a postoperative blood transfusion, usually 1 to 2 units for primary arthroplasties and 3 to 4 units for revisions. While blood transfusions may restore oxygen-carrying capacity, replace fluid volume, and increase vigor, they expose patients to risks, including transfusion reactions, transfusion-related lung injury, antigen exposure, higher mortality, disease transmission, immunosuppression, and infection, as well as higher costs and length of stay.
While the most significant risk factor for acute blood loss anemia is the magnitude of the surgical procedure, several other issues must be considered. Patients on chronic anticoagulation are at high risk for development of postoperative anemia. If possible, medications such as warfarin, rivaroxaban, clopidogrel, aspirin, and even nonsteroidal anti-inflammatory drugs should be stopped preoperatively. Therapeutic anticoagulation should ideally be held postoperatively until adequate hemostasis has been assured and the wound has stabilized. Intraoperative bleeding complications are increased by 1.5 times if preoperative aspirin is not stopped, but discontinuation may increase the risk of postoperative cardiac and vascular complications. Thus, cardioprotective doses of aspirin can be continued for most orthopedic surgery patients who are at increased cardiac risk. Transfusion requirements increase by 50% if clopidogrel is continued perioperatively. Bridging anticoagulant therapy with therapeutic low-molecular-weight heparin is associated with a 92% incidence of bleeding complications, 69% occurrence of hematoma, and 15% development of a prosthetic joint infection; the perioperative use of these drugs in therapeutic doses should therefore be undertaken with caution.
Preoperative anemia should be identified, evaluated, and treated. Anemia is present in 21% of elderly patients who are undergoing elective orthopedic surgery. It is also common in patients with thrombocytopenia, chronic disease, and in menstruating females. Total joint-replacement patients with a preoperative hemoglobin <10 g/dL have a 90% risk of needing a transfusion postoperatively. Of note, preoperative autologous donation (PAD) of blood has declined in popularity because of cost, waste from overcollection, and increased transfusions. Current recommendations for the treatment of preoperative anemia include a thorough hematologic workup, correction of vitamin B12 or iron deficiencies, if present, and evaluation for other sources of ongoing blood loss, such as the gastrointestinal tract.
EVALUATION AND MANAGEMENT OF BLOOD LOSS
Intraoperative measures to limit blood loss and resultant anemia include the use of acute normovolemic hemodilution, tourniquets, hypotensive anesthesia, regional anesthesia, avoidance of hypothermia, blood salvage, meticulous hemostasis, topical hemostatic agents, and intravenous antifibrinolytics.
The use of antifibrinolytics has increased dramatically in the past decade. Tranexamic acid and aminocaproic acid are both lysine analogues that inhibit fibrinolysis by reversible competitive blockade of lysine binding sites on plasminogen and plasmin. By impeding conversion of plasminogen to plasmin, which is an enzyme that degrades fibrin clots, and interfering with the action of plasmin, tranexamic acid inhibits clot breakdown and reduces postoperative bleeding. By disturbing the physiologic balance between clot formation and clot dissolution, there is a theoretical concern that these agents may increase the risk of thromboembolic disease but to date the data have not borne out this concern. Numerous studies have demonstrated decreased blood loss, lower incidence of postoperative anemia, and fewer transfusions in total joint, spine, and orthopedic trauma patients. In some of these studies, high-risk patients with ischemic heart disease or previous thromboembolic disease were excluded, but there is increasing evidence that anti-fibrinolytic therapy does not increase risk of thrombosis in these patients. Nevertheless, rare case reports have documented cerebral thrombosis, arterial thrombosis, ARF, coronary graft occlusion, and PA catheter thrombosis. In our institution, the use of tranexamic acid has decreased the number of patients requiring a transfusion by 44%, with a 55% reduction in the number of units transfused per patient without an increase in thrombotic complications.
Intravenous crystalloid given intraoperatively quickly leaves the pressurized intravascular space and accumulates in the interstitial (or “third”) space. Intravascular normovolemia must be maintained perioperatively, but significant postoperative weight gain is a marker of excessive extravascular fluid and is frequently a poor prognostic sign. Pathologic interstitial fluid shifts occur with infection, large procedures, major trauma, and inflammation. Isotonic crystalloid for fluid maintenance and replacement decreases the interstitial shift by maintaining an intravascular osmotic gradient. Colloids (hetastarch, dextran, and albumin) are more effective in maintaining this gradient and are preferred to replace blood loss in major surgical procedures. While the majority of blood loss occurs during operation or in the first 24 hours thereafter, decreases in hemoglobin and hematocrit may not stabilize until postoperative day 2 or 3, as interstitial fluid becomes mobilized. Often the decline in lab values is dilutional during this period and does not necessarily indicate ongoing blood loss or create hemodynamic instability. Incremental postoperative blood loss can be reduced by compressive dressings, minimizing phlebotomy, avoidance of continuous passive motion machines, optimized nutrition, and judicious use of chemoprophylaxis for deep vein thrombosis. Therapeutic anticoagulation should be avoided.
Postoperative anemia results in fatigue, decreased vigor, slower recovery, and increased length of stay. More severe cases may lead to impaired cognition, hypotension, tachycardia, dyspnea, and decreased tissue oxygenation. American anesthesia guidelines suggest that transfusions are unnecessary if hemoglobin is >10 mg/dL and is recommended if <6 mg/dL. Decisions about transfusions for hemoglobin values between 6 and 10 should be based on a combination of factors including medical history, symptoms, laboratory studies, and an understanding of operative and ongoing postoperative blood loss. When given, packed red blood cells should be administered on a unit-by-unit basis rather than ordering the historic-standard of two units at a time.
Postoperative anemia substantially recovers between post-op day 7 and 28, but complete recovery may be delayed by iron and nutritional deficiency. Oral iron supplementation is frequently administered, but is of questionable benefit. Gastrointestinal side-effects are frequent, and randomized trials have not shown an increase in hemoglobin recovery in the first 3 weeks of FeSO4 administration, when return to 85% of the pre-op values occurs. There is a statistically-significant improvement of 3% to 3.5% in the next 3 weeks with iron administration, but the clinical importance of this change is questionable.
Surgical site infections are estimated to complicate 0.5% to 2.5% of clean orthopedic cases, including elective joint replacements. Currently, approximately 1 million hip and knee joint-replacement procedures are done in the United States annually. At a rate of 1%, approximately 10,000 post-op joint replacements will become infected annually. At an estimated cost of over $75,000 per case, the cost to the US health care system is approaching $1 billion. In most large centers, infection has replaced aseptic loosening, osteolysis, dislocation, and implant failure as the most frequent cause of revision total joint arthroplasty.
The most effective strategy for managing orthopedic infections is prevention, making risk stratification and modification essential. Development of a postoperative infection requires introduction of a bacterial load of sufficient quantity and virulence in a favorable environment, which often suggests an immunocompromised host. Knowledge of these factors helps us understand which patients are at risk for development of this serious complication.
There are numerous risk factors associated with the development of postoperative orthopedic infections (Table 65-1). These include factors related to the initial contamination of the wound, the patient’s intrinsic bacterial flora, the wound environment, and host factors which may limit the patient’s ability to eradicate potential contamination.
TABLE 65-1Risk Factors for Surgical Site Infection ||Download (.pdf) TABLE 65-1 Risk Factors for Surgical Site Infection
|Risk Factors for Infection |
- Inflammatory Arthritis (2-8%)
- Diabetes (3.1-13.5%)
Sickle cell disease
- Malnutrition (3-5× higher)
- ASE > 3
- Hemophilia (9-13%)
- Malignant tumors
- Tobacco use
- Renal failure (HD)
- Dental infections/hygiene
- Skin infections
- Chronic UTI’s
- Previous surgeries
- Vascular disease
- MRSA Colonization
- Obesity (6.7× higher TKA, 42× for THA)
- Atrial fibrillation
- Older patients
- Low income
- Male gender
- Hospital or surgeon with low volume
- Longer operations (>3 hours)
Assessment of potentially modifiable patient-associated risk factors is an important part of perioperative care of the elective orthopedic patient. It has been shown that 80% of primary total joint replacements and 93% of revision replacements have at least one modifiable risk factor:
Diabetes is a well-established risk factor for SSI. Perioperative hyperglycemia affects the microvascular circulation, impairs oxygen delivery, and inhibits chemotaxis, complement function, and phagocytosis. Each of these factors contribute to problems with wound healing and infection. Uncontrolled perioperative diabetes has been associated with 2.25 times increased risk of wound infection.
There is a strong association of postoperative urinary tract infections with prosthetic joint infection, but the significance of preoperative urinary tract infections is unknown. The most important factor is whether patients are symptomatic with obstructive symptoms in the preoperative setting; such patients should have urologic consultation prior to elective surgery. Pre- or postoperative urine cultures demonstrating bacterial counts of >100,000 colonies/mL require treatment even in the asymptomatic patient. Lower-bacterial counts may also warrant treatment in the presence of obstructive or irritative symptoms. Postoperatively, attention toward prevention of bladder distention and stasis is essential, and catheters should be removed as soon as possible to prevent colonization.
Malnutrition is identified by transferrin levels <200 mg/dL, albumin <3.5 g/dL, or a total lymphocyte count <1500 cells/mm3. Patients that are malnourished have a 5 to 7 times higher risk of major wound complications.
Patients with morbid obesity have a risk of deep infection that is more than twice that of normal controls, and in some cases the risks may outweigh the benefits of surgery.
Patients with rheumatoid arthritis have a 2 to 3 times increased risk of infection compared to patients with osteoarthritis due to a combination of innate autoimmune immunosuppression and the use of immunosuppressive medications. Anti-inflammatories, prednisone, methotrexate, and biologic agents are all associated with wound healing complications and infections. Perioperative cessation of methotrexate and biologic agents is controversial, but frequently recommended. Prednisone has a high association with immunosuppression and perioperative infection, but is not usually stopped because of the risk of disease flares and acute adrenal insufficiency.
Smoking is a frequently occurring modifiable risk factor. Smokers have 3 times the risk of wound healing complications with 3 to 4 times higher rates of nonunion in fractures and spinal fusions. Nicotine-containing products should be stopped 4 to 6 weeks prior to surgery, if possible.
EVALUATION AND MANAGEMENT OF SSI
There is no single reliable diagnostic test to rule-in or rule-out early postoperative infection, making evaluation difficult. Fever and elevated white cell count in the immediate postoperative period are common and often due to atelectasis and small amounts of embolized fat in patients undergoing orthopedic surgery. Disproportionate pain, prolonged wound drainage, or excessive erythema most commonly occur. Wounds with purulent drainage or a sinus tract should be considered infected. Serologic testing is nonspecific, and therefore only marginally useful in the perioperative period. Both erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) are typically elevated in the early postoperative setting; the CRP returns to normal in about 3 weeks and ESR in 6 weeks.
Aspiration or biopsy of a suspected orthopedic surgical site infection is the gold standard diagnostic test. Fluid obtained should be sent for cell count with differential, as well as aerobic and anaerobic cultures. In the early postoperative setting, the white blood cell count in a total joint replacement can be as high as 27,000 WBCs/µL, but should be below 3000 in the nonacute setting. A neutrophil count of greater than 89% is highly sensitive and specific for infection even in the first 6 weeks following surgery. Fluid culture has relatively low sensitivity at about 75%, but is 95% specific for infection. False negative cultures are frequently associated with administration of antibiotics, which should be held for at least 2 weeks prior to sending cultures for reliable results.
Treatment of orthopedic surgical site infection requires both operative and medical management. Since there are large amounts of surface area around implants that are not exposed to blood and vascularized tissue, antibiotics alone are ineffective in eradicating most hardware associated infections. Moreover, the biofilm layer that forms within days of acute infection and is associated with many bacteria is usually not cleared even with surgical debridement. Most acute postoperative or hematogenous infections warrant a trial of surgical irrigation, debridement, and implant retention followed by IV antibiotic therapy. Chronic infections usually require hardware removal, IV antibiotics, and a staged reconstruction for successful eradication of the infection.
The goals of treatment of orthopedic infections are listed in Table 65-2. It is important to realize that management may be variable depending on numerous patient factors, and achieving all of the stated goals may not be possible in all patients. This requires extensive collaboration between the orthopedic surgeon, the hospitalist, and the infectious disease consultants.
TABLE 65-2Goals of Treatment of Orthopedic Infections* ||Download (.pdf) TABLE 65-2 Goals of Treatment of Orthopedic Infections*
Prevention of life-threatening bacteremia and sepsis.
Prevention of local or hematogenous remote extension of the infection.
Salvage of the affected extremity and avoidance of potential amputation.
Infection suppression to allow for union of fractures or spinal fusion.
Preservation of function of the infected prosthesis or hardware.
Long-term eradication of the infection and prevention of recurrence.
Patients should maintain a sterile dressing over the wound until all serous drainage has stopped and the wound has become sealed. Persistent drainage beyond 2 weeks usually warrants operative investigation and debridement for early detection and eradication of a potential infection. Oral antibiotics for erythematous or draining wounds are rarely indicated and may complicate diagnosis via false-negative culture results as a deep infection progresses. Administration of oral antibiotics is associated with poor outcomes if debridement and implant retention is attempted as they often lead to a delay in diagnosis, establishment of a mature biofilm, and development of antibiotic resistance. Fever or an increase in pain or swelling may be the first signs of an infection, and should be thoroughly evaluated.
Long-term protection of previously implanted orthopedic hardware is also important. Patients should be informed of potential risks and understand the importance of prompt treatment of remote infections, particularly of the hands, feet, and mouth. Antibiotic prophylaxis for dental and other procedures is controversial but favored by most orthopedic surgeons.
PERIOPERATIVE PAIN MANAGEMENT
Acute pain following orthopedic surgery remains one of the most vexing challenges impacting patient experience and outcome. Historically, general anesthesia (GA) combined with either oral or parenteral narcotic medications have been the preferred strategies to address postoperative pain in the orthopedic patient. With classic opioid-centric postoperative pain regimens, the incidence of adverse drug events has been reported to be 8.5% among total joint-replacement (TJR) patients. This accounts for greater than half of all postoperative in-hospital complications and contributes to increased length of stay. In recent years, concerns regarding the side-effects of GA combined with high-dose opioids have prompted the evolution and popularity of multimodal analgesic regimens. Multimodal analgesia was first popularized among total joint-replacement patients, but has expanded to orthopedic trauma and spine patients as well.
Pain regimens need to be tailored to the individual patient, particularly with respect to opioids in the following situations:
Elderly patients as well as those with known hepatic or renal dysfunction may have impaired drug clearance, and therefore may be more susceptible to side effects if dosages are not properly adjusted.
Patients with obstructive sleep apnea have increased rates of cardiopulmonary complications, including arrest, following major surgery. Many analgesics, but most notably opioids, exacerbate hypoxia via reduction of respiratory drive, volume, and alertness, and may cause previously undiagnosed apnea to become clinically relevant in the postoperative setting.
Patients taking regular doses of opioids preoperatively may develop a tolerance to postoperative narcotic pain medications. These patients are likely to present a challenge in terms of management of acute pain and may require significantly increased doses compared to the opiate-naïve patient.
Obese patients can require dose adjustments of fat-soluble medications like opioids to avoid sequestration and effective dilution in excess adipose tissue.
EVALUATION AND MANAGEMENT OF PAIN
Acute pain that is poorly controlled postoperatively may lead to hypertension, tachycardia, increased tissue oxygen demand, myocardial ischemia, and decreased minute ventilation from respiratory depression with pulmonary complications (including pneumonia). Pain prevents patients from effectively mobilizing and participating in therapy, which may jeopardize the outcome of surgery. Additionally, acute pain affects two of the classic components of Virchow’s triad via indirect inhibition of fibrinolysis as well as limiting mobility and contributing to venous stasis. Safe and effective treatment of acute pain in the postoperative setting is of paramount importance for these reasons and also to improve the overall patient experience.
Now gaining popularity nationwide, the combination of regional plus multimodal anesthesia is replacing the historic strategy of general anesthesia plus opioid-centric analgesia as the new standard of care for orthopedic surgery patients. This concept aims to both prevent and treat acute pain by using a combination of agents with different mechanisms of action in order to maximize the analgesic effects while decreasing undesirable side effects of any single agent (Figures 65-1 and 65-2).
Multimodal analgesia pathways, agents, and targets. (From Kehlet H, Dahl JB. The value of “multimodal” or “balanced analgesia” in postoperative pain treatment. Anesth Analg. 1993;77:1048-1056.)
An example of a multimodal analgesia pathway for total joint replacement patients. CSE, combined spinal-epidural anesthesia; FNB, femoral nerve block; GA, general anesthesia; IV, intravenous; LA, local anesthetic; PCA, patient-controlled analgesia; PO, oral. (From Parvizi J, et al. J Bone Joint Surg. 2011;93(11):1075-1084.)
One of the principles of multimodal analgesia is the preemptive treatment of acute pain, and therefore medications are initiated prior to surgery and continued postoperatively. Though substantial variability exists between institutions, a typical multimodal protocol will include some combination of opioids, acetaminophen, peripheral as well as central nerve blockade, local anesthetics, steroids, NSAIDs, and anticonvulsants. The synergistic effects of combination agents allows each to be dosed more conservatively than with single medication therapy. Though the severity and likelihood of side effects of any individual agent may be diminished, the spectrum of possible concerns expands with the additional medications.
Regional as opposed to general anesthesia offers advantages including a decreased incidence of postoperative nausea and confusion, improved early postoperative comfort, and facilitates early mobilization. Several potential complications specifically related to neuraxial and peripheral regional anesthesia need to be considered:
Epidural hematoma is a rare though potentially devastating complication of combined spinal-epidural (CSE) anesthesia, occurring in <1/100,000 patients who receive epidural or spinal anesthesia. Orthopedic patients commonly receive anticoagulants to prevent VTE in the postoperative period, theoretically increasing the risk of hematoma formation.
Epidural abscess is also rare but can be devastating; it is likely to present with systemic signs of infection, including fever, chills, nausea, and malaise.
Abnormal motor blockade in a postoperative CSE patient warrants close monitoring and likely requires further evaluation. Lumbar spine MRI is the most informative and appropriate diagnostic study.
CSE may be associated with persistent hypotension in the early postoperative period. Anesthetic medications delivered to the lumbar epidural space can affect the lumbosacral sympathetic chain, thereby producing an effective transient sympathectomy. Resulting vasodilation produces hypotension, particularly with the patient erect and standing; it is generally responsive to IV fluid support until the medication wears off.
Urinary retention is more likely following CSE than after GA; patients receiving CSE may benefit from bladder catheterization postoperatively.
CSE anesthesia is commonly supported by local and/or regional nerve blockade. Local anesthetic toxicity is a particular risk if all three modalities are employed. Depending on the specific type of local anesthetic agent, a mixture of cardiac (hypotension) and neurologic (seizures) symptoms may be present, with bupivacaine being the most common offending agent.
The side-effect profiles of individual components of a multimodal analgesia regimen are well established. A few relevant orthopedic considerations deserve mention here:
NSAIDs are a mainstay of nonoperative treatment of many orthopedic conditions, but also play an important role in the management of acute postoperative pain. Aside from nephrotoxicity and gastrointestinal effects, animal models and basic science studies suggest interference of NSAIDs with both fracture healing and osseointegration of implants. These biologic pathways are critical not only in the setting of trauma, but also following many joint-replacement procedures, the long-term durability of which often requires ingrowth of native bone into and around the porous surface of metal prostheses. Though unsupported by clinical trials, most orthopedic surgeons prefer to avoid NSAIDs in the setting of either fracture or cementless arthroplasty despite their analgesic benefits. Similarly, NSAIDs inhibit platelet aggregation and contribute to risk of bleeding complications, particularly in patients who are also receiving prophylactic antithrombotic agents.
Anticonvulsants such as pregabalin and gabapentin provide effective analgesia and reduce opioid requirements in postoperative orthopedic patients.
Acetaminophen has long been utilized as an antipyretic as well as a central analgesic agent. Recently, intravenous formulations have been approved for use and proven efficacious as an adjunct for the management of postoperative pain.
Compartment syndrome in the extremities is most commonly encountered in the acute trauma patient. Tibial fracture remains the specific injury most commonly associated with compartment syndrome, and open fractures have a rate of compartment syndrome five times that of closed fractures. Patients with a vascular injury are far more likely to have compartment syndrome than those without. The consequences of unrecognized compartment syndrome may be devastating, including joint contractures, permanent loss of function, loss of limb, and even death.
It is important to know which patients are at risk for developing compartment syndrome, as well as those at risk for failure to recognize and diagnose compartment syndrome in a timely fashion. Patients incapable of providing an accurate history secondary to intubation, intoxication, or severe dementia can present a diagnostic dilemma. In these circumstances the clinician must rely solely upon physical examination and objective findings. Similarly, patients who have received a peripheral or central nerve blockade for the purpose of postoperative pain control may be unable to accurately report severity of pain, and this may delay the diagnosis of compartment syndrome. Postoperative orthopedic patients commonly experience swelling and pain, and occasionally neurologic symptoms in the recently operated extremity. Differentiation between expected symptomatology and those concerning for compartment syndrome can be challenging.
EVALUATION AND MANAGEMENT
An understanding of the pathophysiology of compartment syndrome is critical to performance of a thorough physical examination and achievement of an accurate diagnosis. Within each anatomic segment of the extremity, the soft tissue contents traversing the limb are contained within compartments bordered and contained by tough, nonexpansile fascial connective tissue. Upon traumatic injury, or other myriad events that increase the fluid content of a compartment, the fixed space within the compartment is unable to accommodate expanding volume and results in increased interstitial pressure. If the interstitial pressure exceeds the capillary perfusion pressure, oxygen-rich arterial blood cannot diffuse to the tissues, and relative hypoxemia occurs. This results in a positive feedback cycle by which the resulting hypoxemia exacerbates local inflammation and increases edema, which again increases volume and raises interstitial pressure, and further impairs oxygen delivery to tissues (Figure 65-3).
The positive feedback loop and pathophysiology of compartment syndrome. (From Aguirre JA, et al. Case scenario: compartment syndrome of the forearm in patient with an infraclavicular catheter: breakthrough pain as indicator. Anesthesiology. 2013;118:1198-1205.)
Once the positive feedback cycle has been established, it is difficult to reverse without operative intervention. Surgery for compartment syndrome entails extensile incisions and complete release of the restrictive fascial boundary of the affected compartment. Upon release, the liberated muscles within the compartment expand, often precluding primary closure of the wound. Open wounds pose risk for secondary infection and necessitate additional surgery for wound closure. Depending on the timeliness of initial diagnosis, serial debridement of necrotic muscle and soft tissues within the involved compartment(s) may also be necessary.
The goal of evaluation and management of the at-risk patient is to avoid if possible, but at the very least recognize in a timely fashion, the need for surgical treatment of compartment syndrome. This process starts with identification of the at-risk patient. Given the drastic consequences of a missed diagnosis, even low-energy trauma patients with relatively modest symptoms deserve to be screened for compartment syndrome via a careful physical examination. Those patients with more concerning mechanism of injury (high energy, crush injury, delayed discovery), or concerning initial physical exam should be aggressively treated with noninvasive measures to reduce inflammation and swelling in the affected extremity and subjected to serial examinations. Aggressive application of ice along with strict elevation can prevent development of compartment syndrome, yet elevation in the setting of an established or evolving compartment syndrome can exacerbate ischemia. Treatment of concomitant hypotension can maintain capillary perfusion pressure greater than interstitial pressure and ensure continued tissue perfusion.
Other injuries and comorbidities may conspire to conceal the diagnosis of compartment syndrome. A high index of suspicion is necessary to avoid missed diagnoses. Classic medical teaching refers to the six “P’s” for the diagnosis of compartment syndrome: pain, pallor, pulselessness, poikilothermia, paralysis, and paresthesias. With the exception of pain and perhaps paresthesias, all are late findings of compartment syndrome. If the diagnosis has not been established prior to the development of pulselessness and paralysis, the prognosis becomes dire as tissue necrosis is likely already ongoing.
Pain out of proportion to the injury is the hallmark physical exam finding of compartment syndrome. By definition, compartment syndrome cannot occur without swelling although it may be difficult to appreciate even significant edema contained within the deep posterior compartment in the lower leg. Compartments may be described as, in order of increasing concern, soft, full, tense or firm. The possibility of evolving or early compartment syndrome should be considered in the presence of severe pain despite less worrisome objective exam findings. In such instances, the patient should be reexamined frequently at regular intervals by the same physician who may therefore recognize trends and changes in serial examinations.
Certainly a worsening examination in a patient at risk for compartment syndrome is concerning and warrants further action. Pain with passive stretch of muscles in the suspect compartment is the most ominous physical finding. Compartment syndrome is a clinical diagnosis, and this scenario would indicate the need for emergent surgical fasciotomies. In instances where the patient and/or the exam is unreliable or equivocal—such as the obtunded or intubated patient—direct measurement of compartment pressures is indicated. This may be accomplished via the use of commercially available devices specifically designed for this purpose, or a modified arterial line setup (Figure 65-4). An absolute pressure value >30 mm Hg is consistent with a diagnosis of compartment syndrome. Of more precise diagnostic value is an absolute pressure within 20 mm Hg of the diastolic pressure, that is, the delta (diastolic blood pressure minus compartment pressure). A compartment with a pressure of 30 mm Hg may remain well-perfused in a normotensive patient with a diastolic blood pressure of 70 mm Hg, as opposed to a scenario where that same patient were hypotensive with a diastolic pressure of only 40 mm Hg. Early consultation with orthopedics is advisable if there is any question or concern of compartment syndrome.
Pressure measurement of the anterior compartment of the lower leg with a commercially available device.
VENOUS THROMBOEMBOLIC DISEASE
Venous thromboembolism (VTE), specifically pulmonary embolism (PE), is the most common cause of readmission and death after elective hip and knee replacement. In the absence of prophylaxis, 80% to 85% of THA and TKA patients will develop deep venous thrombosis (DVT), as will patients following hip fracture. Polytraumatized patients are known to be at high risk for DVT; pelvic (60%), femoral (80%) and tibial (77%) shaft fractures, and spinal cord injury (81%) are associated with the highest risk. Compared to patients with head, chest, or abdominal injury, those with isolated femur or tibial shaft fracture have a 5 times greater DVT risk; spinal cord injury imparts an 8.5 times greater DVT risk. Following TKA, in excess of 90% of thrombi occur below the trifurcation of the popliteal vein, and “proximal” clots occur almost exclusively by contiguous extension of primary calf thrombi to the popliteal vein. In contrast, after THA approximately half of thrombi occur primarily in the proximal (popliteal and common femoral) veins, are typically not contiguous with calf clots, and are discontinuous and segmental in nature. Segmental femoral vein clots after THA typically occur in the region of the lesser trochanter and are thought to result from torsional injury to the intima incurred with positioning of the lower limb. Nearly 20% of postoperative calf clots will extend proximally and, once in the thigh, nearly 50% will embolize to the lung. Accordingly, conventional wisdom holds that patients undergoing THA and hip fracture repair have the greatest risk of PE by virtue of their high frequency of proximal thrombi.
Contemporary prophylaxis against VTE is considered standard of care subsequent to a 1986 NIH consensus conference on the subject. Venographic clot rates with potent new anticoagulants are <5% after THA and <20% after TKA. Reduction in DVT after TKA is relatively more refractory to anticoagulant prophylaxis than after THA; symptomatic VTE is now twice as common in-hospital (1% vs 0.5%) and symptomatic PE is twice as common after discharge (0.27% vs 0.14%) following TKA than THA, respectively. Despite these advancements, fatal PE occurs after 0.1% to 0.5% of total joint replacements, accounting for more than 1000 deaths each year. Appropriately, considerable effort has been directed at identifying the optimal prophylaxis regimen. However, use of potent anticoagulants to mitigate activation of thrombogenesis after orthopedic operations must necessarily be tempered by consideration of the bleeding risk following these procedures, where hemostasis is imperfect in the setting of exposed bony surfaces. It must represent a risk-benefit balance between the fear of fatal PE weighed against the morbidity of persistent wound drainage and hematoma, secondary infection, and reoperation resulting from bleeding associated with perioperative anticoagulant use.
Patients who have undergone disruption of the medullary canal of the long bones of the lower limb, either by virtue of traumatic injury or instrumentation for fracture fixation or prosthetic joint replacement, have the highest risk of venous thromboembolism (VTE) of any in the hospital. Intravasation of marrow fat contents is now known to be a potent thrombogenic stimulus and is largely responsible for the peculiar high risk of venous thromboembolism in the orthopedic patient population. Indeed, this risk is appreciated to be so high that all patients undergoing any lower-limb orthopedic procedure are considered “high risk” in the Surgical Care Improvement Project (SCIP) guidelines for VTE risk stratification and all are considered as automatic recipients of directed anticoagulant prophylaxis unless there exist specific contraindications to same. While heritable hypercoagulable conditions such as Factor V Leiden have been shown to account for a majority of spontaneous VTE events in the community, such markers have not been able to identify those among the joint-replacement population at greatest risk for VTE. Neuraxial (spinal or epidural) anesthesia is widely acknowledged to reduce the risk of lower-limb DVT by approximately 50% compared to general anesthesia, independent of the type of anticoagulant VTE prophylaxis employed in the perioperative period.
Orthopedic practitioners typically opt for one of three main regimens for perioperative VTE chemoprophylaxis; aspirin, low-intensity warfarin, or one of the newer anticoagulants such as fractionated heparins or direct factor X or II inhibitors. Choice of anticoagulant is driven by some approach to risk stratification among this group of high-risk patients; most surgeons agree that a past history of VTE warrants more intensive anticoagulation than provided by aspirin alone.
Guideline reconciliation finally occurred between the American Academy of Orthopedic Surgeons and the American College of Chest Physicians in February 2012, largely as a result of agreement that clinically important PE and VTE was the relevant endpoint and insufficient data existed to endorse any one specific prophylaxis regimen. Concurrently, the ACCP placed greater value on avoidance of bleeding events based on acknowledgement of patient preferences and aversion to surgical complications. Likewise, the exhaustive AHRQ Comparative Effectiveness Review on Venous Thromboembolism Prophylaxis in Orthopedic Surgery concluded, “The balance of benefits to harms is favorable for providing prophylaxis … and to extend the period of prophylaxis beyond the standard 7 to 10 days.” They noted that “interclass comparisons could not be made due to lack of data, similarities between classes on outcomes, or offsetting effects where benefits on efficacy were tempered by an increased risk of bleeding.” In 2014, SCIP added aspirin to its list of acceptable agents for VTE prophylaxis, making the list of appropriate agents uniform and all-inclusive among all groups, with no endorsement of any agent or regimen as best practice.
Chemoprophylaxis (Figure 65-5) can by grouped into three main categories; warfarin, aspirin, and selective heparinoids or direct factor Xa or IIa inhibitors.
Warfarin. Low-intensity warfarin (INR target 2.0), despite its variable dosing and need for monitoring, is a time-honored option for orthopedic thromboprophylaxis, likely because it has an acceptably low-bleeding risk (1%-2%) coupled with efficacy in preventing clinically important PE. In 2008, it was the VTE prophylaxis of choice for nearly 50% of US orthopedic arthroplasty surgeons prior to the release of direct Xa and thrombin inhibitors. Warfarin remains popular in the US, despite the inconvenience and expense of monitoring, because many feel it represents the best available compromise of efficacy in preventing clinical VTE and safety in minimizing bleeding.
Critics have been most concerned with its erratic effect, need for monitoring, and associated bleeding. Early studies reported bleeding rates of 10% with a prothrombin time index (PTI) of 1.8 to 2.0, but efficacy is now proven with a PTI of 1.3 to 1.5 (INR 2.0) and reduced bleeding to <2%. Historically, low-intensity warfarin prophylaxis (INR 2.0) results in venographic DVT of 9% to 26% and proximal clot rates of 2% to 5% after THA compared with overall DVT of 35% to 55% and proximal clot rates of 2% to 14% after TKA performed with general anesthesia. Warfarin preferentially reduces proximal compared with distal DVT after THA and, when combined with epidural anesthesia/analgesia, the residual DVT rate dropped below 10%. Most importantly, extended low-intensity warfarin continued for 6 weeks after THA or TKA in 3293 patients was associated with readmission for clinical VTE in 0.3% in THA and 0.2% in TKA with a major bleed rate of 0.1%. In the aggregated THA and TKA groups, 6 weeks of warfarin after hospital discharge eliminated the risk of PE, and reduced VTE-related readmissions 0.2% versus 1.6%.
Optimal use of warfarin remains challenging and labor intensive; it ideally begins the night before surgery because of its 48-hour latency to anticoagulant effect. Indeed, this lag in activity is likely responsible for its popular safety margin with low-perioperative bleeding rates.
Aspirin. Aspirin, acetylsalicylic acid, is traditionally considered an arterial drug but has been used in conjunction with mechanical compression devices for venous thromboembolism prophylaxis by a consistent minority (10%-20%) of surgeons. Conventional wisdom holds that aspirin reduces arterial thrombosis through inhibition of platelet cyclooxygenase-1, which decreases synthesis of thromboxane A2 (platelet-activating eicosanoid) and related platelet activation, and has demonstrated substantial arterial thrombotic event reduction as well as survival benefit relative to stroke, myocardial infarction, and related deaths in high-risk patients. Conversely, clotting experts often regard aspirin as an inferior and inconsistent agent in mitigating venous thrombosis.
Since 2006, three large observational studies and the British joint-replacement registry have rekindled the notion that aspirin can prevent clinical PE after total joint arthroplasty, especially in conjunction with regional anesthesia. In nearly 10,000 THA and TKA patients managed with neuraxial anesthesia, in-hospital pneumatic compression, and 6 weeks of aspirin (325 mg bid or 150 md qd), overall VTE-related readmission was 3.2% after THA with a clinical PE rate of 0.6% compared with overall VTE-related readmission after TKA of 0.5% with a clinical PE rate of 0.36%. In 2009, National Joint Registry data matched nearly 23,000 patients receiving aspirin or LMWH and revealed no difference in 90-day outcomes related to clinical PE (0.7%), DVT (0.95%), stroke or gastrointestinal bleeding (0.75%), or reoperation (0.35%). Yet, patients receiving LMWH had a survival advantage over those receiving aspirin (all-cause mortality 0.49% vs 0.65%; p = 0.02).
In two studies of >1200 patients randomized to aspirin or placebo for prevention of recurrent VTE after standard therapy, aspirin was associated with a 32% reduction in recurrent VTE and a 34% reduction in major adverse vascular events, without an accompanying increase in bleeding. Given its apparent efficacy without a compromise in safety, the role of aspirin in VTE prophylaxis and prevention of embolization of existing venous thrombi is being reassessed.
Newer Agents. Many new selective anticoagulants have been introduced since the 1986 NIH conference, but the risk of fatal PE after THA and TKA has changed very little. Despite venographic clot rates up to 5 times greater than with these newer agents, low-intensity (INR 2.0) warfarin and aspirin offer comparable clinical PE rates and two- to three-fold less major bleeding complications.
Low-molecular-weight (fractionated) heparins and pentasaccharide are incrementally smaller sugar molecules with increasing specificity in binding antithrombin III (AT-III) as the critical intermediary prior to deactivating factor X or II (thrombin). Indeed, synthetically derived pentasaccharide exactly corresponds to the five sugar AT-III binding site and enjoys enhanced specificity to bind factor Xa. Due to its more proximal location in the coagulation cascade, this affinity for factor Xa greatly augments the potency of LMW heparins and pentasaccharide. Accordingly, LMWH reduced overall venographic DVT from 44% to 31% after THA compared with unfractionated heparins; DVT reduction was also significant after TKA, but less dramatic. Fondaparinux (pentasaccharide) further reduced venographic DVT by 50% compared with enoxaparin after THA, TKA, and hip fracture. It was the first agent to reduce DVT to <20% after TKA and the most extensively studied agent after hip fracture, but its bleeding rate of 3% to 6% exceeded even that seen with enoxaparin. One meta-analysis demonstrated a two- to three-fold increase in bleeding after THA and TKA with LMWH compared with warfarin prophylaxis.
New direct factor X and II inhibitors require no monitoring and are administered orally. Rivaroxaban, a direct factor Xa inhibitor, has been most extensively studied in two THA and two TKA trials. Compared with enoxaparin, pooled analysis of the four studies demonstrated a 58% reduction in all-cause mortality and symptomatic VTE (0.6% vs 1.3%; p < 0.001) but aggregate bleeding data revealed an increase in major plus clinically relevant nonmajor bleeding events (3.2% vs 2.6%; p = 0.039). Rivaroxaban has enjoyed rapid growth in popularity in the VTE prophylaxis market, but an increase in bleeding events has dampened enthusiasm for its widespread adoption by the orthopedic community. One observational study demonstrated all-cause mortality after total hip and knee replacement in patients who had been given potent anticoagulants was more than twice that in patients receiving aspirin, pneumatic compression, and regional anesthesia. Another report specifically noted the “failure” of low-molecular-weight heparin compared with warfarin prophylaxis; symptomatic DVT (3.8%), nonfatal PE (1.3%), persistent wound drainage resulting in readmission (4.7%) and reoperation (3.4%) all occurred at rates exceeding prior experience with low-intensity warfarin. Finally, one retrospective review of 1048 consecutive THA and TKA reported a two-fold increase in 30-day reoperation for wound complications with rivaroxaban compared with tinzaparin prophylaxis. Not surprisingly, the orthopedic community has been slow to adopt these newer agents in favor of a more balanced strategy that offers less bleeding risk with comparable protection against clinical VTED.
INPATIENT MANAGEMENT OF ACUTE VENOUS THROMBOEMBOLISM
Management of perioperative DVT requires more vigilant treatment than a spontaneous event in an ambulatory patient. Typically, routine prophylaxis is continued for 35 days after operation and the occurrence of a symptomatic clot may further prolong anticoagulation to mitigate the risk of clot propagation and embolization.
With the advent of multidetector computed tomography for definitive diagnosis of PE, the prevalence of PE has increased 10-fold in the past decade, from 0.2% to >2%, without any explicable change in the operation or perioperative management. It is likely that this increment is largely explained by the recognition of very small peripheral subsegmental filling defects with this more sensitive CT technology. The clinical significance of these lesions has been questioned and it has been proposed that they represent embolization of marrow elements from instrumentation of the medullary canal rather than a conventional PE. When clinically silent hypoxemia is recognized with pulse oximetry, a 10-minute 2 L/min oxygen challenge has been proposed; resolution of hypoxemia has been suggested to indicate the absence of a meaningful PE and no need for further intervention. Persistence of hypoxemia deserves further workup. The observed frequency of PE has returned to less than 1% with this pragmatic clinical protocol.
Acute postoperative PE is conventionally managed with more intensive anticoagulation. There is increasing interest in outpatient management with the newer oral agents, however the substantial risk of bleeding in the early period after hip and knee replacement should temper the enthusiasm for outpatient management of PE within 5 to 10 days of the index operation. Indeed, when initiating therapeutic anticoagulation with an IV heparin bolus for clinically significant PE within 5 days of operation, the risk of major wound hemorrhage is 50% and within 1 week the bleeding rate decreases to 15%. In general, intravenous heparinization should be initiated with a constant continuous infusion without bolus therapy in the first week after total joint replacement.
In general, it is accepted that the risk of VTE after orthopedic surgical procedures extends far beyond the time of operation and hospital discharge. Clinical guidelines endorse 35 days of prophylaxis after THA and 10 to 35 days after TKA.
The fat embolism syndrome (FES) was initially described by Zenker in 1861 in a patient following a thoracoabdominal crush injury. It may be practically defined as a complex alteration in coagulation homeostasis that occurs as an infrequent complication of fracture of the pelvis and long bones and is manifest clinically as acute respiratory insufficiency secondary to the filtering function of the lung. The full blown clinical syndrome is evident in 0.5% to 2% of patients after isolated long bone fractures and approaching 5% to 10% of patients with multiple fractures and pelvic injuries following polytrauma. In contrast, fat embolization occurs as a subclinical event following all fractures as well as instrumentation of the medullary canal during total joint replacement. The likelihood that this event results in the clinical manifestations characterized by florid respiratory failure is determined by the quantity of fat intravasated into the systemic circulation and the ability of the patient’s cardiopulmonary system to withstand the collection of this material in the lung.
Conditions that increase the size and fatty content of the marrow cavity, such as collagen vascular diseases and osteoporosis, increase the risk of FES. Children develop clinical FES nearly 100 times less commonly than adults, secondary to persistence of hematopoietic marrow and a paucity of fatty marrow. Instrumentation of the medullary canal during THA and TKA always results in embolization of marrow fat to the lung; concurrent bilateral TKA under the same anesthetic increases this risk and has been largely curtailed. Whether it is clinically manifest as FES is largely determined by the magnitude of the embolic load, the intrinsic cardiopulmonary reserve, and the ability to maintain oxygenation in the presence of pulmonary capillary occlusion.
EVALUATION AND MANAGEMENT
The initial insult after embolic fat to the lungs is characterized by increased right heart pressures and cardiovascular collapse, evidenced by hypotension, hypoxemia, confusion, and bradycardia, and is infrequently clinically evident in humans. It has, however, been reported after cementation of the femoral component during total hip arthroplasty, especially in the elderly osteoporotic patient treated for hip fracture and after intramedullary nailing of the femur for an impending pathologic fracture with concurrent filling of the medullary canal with methylmethacrylate. Autopsy evaluation typically demonstrates embolic marrow elements in the capillary bed of the brain, suggesting a lung filter overloaded with embolic material that leaks into the systemic circulation. Transient aphasia and other central neurologic deficits have been observed after bilateral total knee arthroplasty in patients with a patent foramen ovale; embolic intracerebral events can be seen by magnetic resonance imaging.
The more typical fat embolism syndrome is “delayed” in its appearance; clinical signs and symptoms typically develop in 85% of patients within 48 hours after fracture or instrumentation of the medullary canal. This temporal delay is thought to result from the evolving effects of vasoactive substances in the lung and the resulting decrement in gas exchange. Clinical manifestations are primarily cardiopulmonary in nature and include arterial hypoxemia; oxygen desaturation, sinus tachycardia, and fever comprise the classic triad of early findings. Nonspecific changes of pulmonary congestion may be seen on the chest radiograph and the EKG often exhibits ST segment elevation from ischemia or right heart strain. Alterations in cognition are common and may include lethargy, delirium, and/or seizures secondary to hypoxemia or fat embolization to the brain. Petecchiae develop in more than half of patients and typically occur in the axilla (Figure 65-6), over the chest and base of the neck, and in the conjunctivae; they often appear after a 24- to 48-hour delay and are evanescent in nature. Skin biopsy demonstrates embolic fat in capillaries with local hemorrhage. A variety of laboratory coagulation abnormalities are present but clinical bleeding is rare. While this constellation of findings is uncommon, the mortality rate approaches 10% to 15% after full-blown FES. A high index of suspicion is necessary for early diagnosis and proper treatment.
Axillary petechiae in a patient with clinically evident fat embolism syndrome.
No specific treatment exists for FES. Rather, acute management is predicated upon provision of supportive pulmonary care, often with intubation, oxygen, and mechanical ventilation with airway pressure support in an effort to mitigate the effects of the adult respiratory distress syndrome. In many circumstances high-dose steroids are empirically utilized as a general measure to blunt the lung reaction, stabilize the pulmonary capillary bed, and improve gas exchange.
VD Jr. Fat embolism syndrome: a review of the pathophysiology and physiological basis of treatment. Clin Orthop Rel Res. 1982;165:68–82.
CL. Prevention of deep periprosthetic joint infection. Instr Course Lect
J. Preventing infection in total joint arthroplasty. J Bone Joint Surg Am
. 2010;92(Suppl 2):36–46.
MR. Multimodal pain management in orthopaedics: implications for joint arthroplasty surgery. Orthopaedics. 2013;36(2 Suppl):7–14.
VD Jr. Prophylaxis against venous thromboembolism after total hip and knee arthroplasty: a critical analysis review. J Bone Joint Surg. 2015;3(9):e1.
CS. Prevalence of modifiable surgical site infection risk factors in hip and knee joint arthroplasty patients at an urban academic hospital. J Arthroplasty
A. Acute compartment syndrome in orthopaedics: causes, diagnosis, and management. Adv Orthop
WD. Diagnostic techniques in acute compartment syndrome of the leg. J Orthop Trauma
DA. Perioperative fluid and volume management: physiological basis, tools and strategies. Ann Intensive Care
KD. Reducing perioperative blood loss and allogeneic transfusion in patients undergoing major spine surgery. J Bone Joint Surg. 2011;93A:1268–1277.
LD. Effectiveness and safety of tranexamic acid in reducing blood loss in total knee arthroplasty: a meta-analysis. J Bone Joint Surg Am