Perioperative Acute Pulmonary Embolism: A Concise Review with Emphasis on Multidisciplinary Approach

Abstract

Perioperative acute pulmonary embolism represents a relatively rare complication; however, it could be very serious and devastating in some cases. Its diagnosis could be particularly challenging, especially in the intraoperative period. Herein, we emphasize some key concepts with the aim to perform an early and appropriate risk stratification, diagnostic and therapeutic approach in a multidisciplinary fashion, a brief overview on thromboprophylaxis, with the main objective to improve outcomes and survival in these challenging patients.

Venous thromboembolism (VTE), which encompasses deep vein thrombosis (DVT) and acute pulmonary embolism (PE), remains a relatively common and catastrophic event in the Western countries, leading to an estimated 100,000 to 200,000 deaths per year. ^{1 2} Acute PE is the leading cause of preventable deaths among hospitalized patients, including those who underwent major surgical procedure. ^{3 4 5 6} Perioperative acute PE represents a potentially fatal disease in particular clinical scenarios; thus, it represents a unique challenge for the clinicians, given the complex clinical scenario in the perioperative setting. Given the scarce data in the appropriate approach for the diagnosis and treatment of perioperative acute PE, herein we analyze the value of multidisciplinary teams for the execution of prompt, appropriate risk stratification, diagnostic and therapeutic strategies in these complex subgroup of patients with perioperative acute PE, with the aim of improving clinical outcomes, particularly morbidity and mortality in these challenging patients.

Epidemiology and Risk Factors

Surgical population has a fivefold increased risk of developing VTE, especially during the transoperative and postoperative periods, with an overall incidence between 0.3 and 1.6% in the general surgery population. ^{7 8 9} The incidence is higher in major orthopedic procedures, ranging from 0.7 to 30% and specifically, hip fracture repair, hip hemiarthroplasty, and knee replacement/arthroplasty, carrying an incidence of 5 to 25%. One of the main reasons of such high incidence in the orthopedic population is perhaps because of the nature of such procedures, occurring in anatomical areas of the body in which joints are in closer proximity of rich vascularized territories, including the large caliber deep venous system. ¹⁰ This incidence decreases to 0.25% when adequate thromboprophylaxis is used routinely. ¹¹ In addition, acute PE is associated with a perioperative mortality of up to 12.5%. ¹⁰

The modality of anesthesia also has been shown to have an impact in the development of perioperative acute PE. A meta-analysis by Rodgers et al showed a significant reduction in the rates of perioperative acute PE of 55%, when using epidural regional anesthesia compared with general balanced anesthesia. ¹²

Surgical patients share unique characteristics that may lead toward the development of acute PE such as a proinflammatory environment, activation of the coagulation cascade caused by tissue trauma, prolonged immobilization, and venous stasis. ^{7 8 9}

The Caprini risk assessment model for the prediction of perioperative VTE has been arguably one of the most utilized tools worldwide. ^{13 14 15} It has been validated in >250,000 patients across more than100 studies, demonstrating to be consistent, thorough, and efficient while risk stratifying surgical patients during their perioperative period. ^{13 14 15}

In addition, male sex, age >80 years, prolonged immobilization, chronic heart disease, and a lack of perioperative thromboprophylaxis are independent risk factors for early perioperative acute PE-related mortality in a study of 294 patients. ¹³

Classification and Risk Stratification

Appropriate acute PE classification and risk stratification remain the cornerstone to dictating clinical decision-making and management. The 2019 European Society of Cardiology (ESC) guidelines classify PE in three categories: high, intermediate, and low-risk PE. ^{5 16 17} High-risk PE (or massive PE), defined by hemodynamic instability manifested by persistent hypotension with a systolic blood pressure of <90 mm Hg or cardiogenic shock, carries the highest mortality rate exceeding up to 50%. ¹⁶ Patients without hemodynamic instability but presence of right ventricular (RV) dysfunction, either documented by echocardiography or computed tomographic pulmonary angiography (CTPA) and/or elevated biomarkers of myocardial injury (troponin and/or brain natriuretic peptide), are classified as intermediate-risk PE (or submassive PE). ^{15 16 17} The ESC 2019 guidelines further subdivided the intermediate-risk group into intermediate-high and intermediate-low risk. The intermediate-high risk PE has both objective evidence of RV dysfunction by imaging and positive biomarkers, having a mortality of up to 25%, while the intermediate-low risk PE may have none or only one of these present, carrying a mortality between 2 and 8%. ^{18 19} Finally, low-risk PE patients are those without hemodynamic involvement, RV dysfunction, or elevation of biomarkers, and have <1% mortality. ^{18 19}

Utilizing a multimodality approach by combining clinical judgment, clinical prognostic scores, such as the pulmonary embolism severity index or its simplified version, cardiovascular imaging techniques and cardiac biomarkers for prediction of early outcomes and mortality in acute PE, is now strongly encouraged and stressed to be utilized not only by the ESC/ESR (European Society of Radiology), but by many other clinical societal guidelines. ^{5 15 20}

It is important to emphasize that the above classification and risk assessment of acute PE also apply to all patients in which acute PE is being suspected, including surgical and/or perioperative patients, since there is a lack of adequate prospective studies that have validated different classifications and/or risk assessments pertaining exclusively to the surgical population.

Pathophysiology

Acute PE produces several pathophysiologic derangements within the pulmonary vasculature. The initial insult provoked by blood flow obstruction to the pulmonary vasculature, leads to abrupt increase of the RV afterload and impedance. Subsequent pulmonary vasoconstriction is biochemically mediated by hypoxemia and platelet activation of complement factors 3 and 5, humoral release of serotonin, histamine, thromboxane-A2 within the platelet-rich emboli. This leads to a significant rise in the pulmonary vascular resistance. ^{7 21} The abrupt rise in RV pressure decreases RV stroke volume, causing increased preload, RV dilation, and tachycardia. Finally, progressive RV failure causes ischemia, systemic hypotension, hypoperfusion, and cardiovascular collapse. ^{21 22} Additionally, surgical patients undergoing general anesthesia have a higher risk to developing hypotension caused by intraoperative systemic vasodilation. ^{7 9 12}

Diagnosis

The diagnosis of perioperative acute PE remains a challenge due to the absence of typical signs and symptoms of acute PE in anesthetized patients. High index of clinical suspicion is often required for early diagnosis. Abrupt and unexplained tachycardia, hypotension, hypoxia, and/or decreased expired end-tidal carbon dioxide (EtCO ₂ ) levels, may be the signs of acute PE and should prompt further diagnostic work-up. ^{20 21 22}

Visnjevac and colleagues compared the utility of four perioperative monitoring tools in 146 high-risk intraoperative PE patients: EtCO ₂ levels, central venous pressure, echocardiography, and standard vital signs monitoring. ²³ Changes in vital signs were the most predominant finding and were associated with increased mortality (72%). Increase in EtCO ₂ as an alerting tool was associated with improved survival compared with hemodynamic changes ( p  < 0.0001), corresponding to a decreased mortality (30.2%). Echocardiographic signs of RV strain (such as RV wall motion abnormality, RV dilatation, and increased tricuspid regurgitation jet) were associated with worse outcomes, but direct thrombus visualization was not. ²³

Transthoracic echocardiography (TTE) is frequently used in the diagnosis of acute PE, especially in the setting of hemodynamic instability and inability to move to radiology for a CTPA. Presence of RV dilation, increased RV to left ventricular (RV/LV) diameter ratio, interventricular septal compression, paradoxical septal motion, and RV hypokinesis may be identified in acute PE and indicate right heart dysfunction or RV strain. McConnell sign, which refers to a pattern of RV hypokinesis with relative apical sparing of the RV free wall, has a sensitivity and specificity of 77 and 94%, respectively. ^{24 25} Additional RV parameters to consider include RV end-diastolic ratio, right atrial and inferior vena cava dilation, tricuspid annular plane systolic excursion, RV index of myocardial performance, and Doppler estimation of pulmonary arterial systolic pressures. ²⁵

Transesophageal echocardiogram (TEE) may allow direct visualization of PE in the pulmonary arteries, in addition to the obstructive physiologic changes seen on TTE. TEE might be particularly useful in the surgical setting, as occasionally cardiac surgeries utilize this intraoperative technology already to quantify improvement in LV and/or valvular function after its repair. Additionally, intraoperatively the patient will already be sedated and possibly also intubated, facilitating the esophageal probe insertion. However, the sensitivity for direct visualization for PE in TEE was only 26% in a study performed on 46 patients immediately before surgical pulmonary embolectomy (SPE). ²⁴ In the same study, TEE did have a sensitivity of 96% for RV dysfunction, 50% for moderately severe tricuspid regurgitation, and 98% for interventricular paradoxical septal motion. In TTE, the McConnell's sign had a sensitivity of 77% and a specificity of 94% for acute PE. ²⁵ Despite CTPA being the preferred test for acute PE diagnosis, in the complex and possibly chaotic intraoperative period, TEE may be the diagnostic tool of choice, especially in those patients with hemodynamic collapse, to avoid interrupting the surgical procedure.

Recently, point of care TTE has emerged as a powerful and rapid noninvasive diagnostic tool in acute PE. It is particularly useful in the critically ill patient during the perioperative period and in the operating room for diagnosis and monitoring of the response to hemodynamic interventions in patients with suspected perioperative acute PE. ²⁶

Fig. 1 provides a proposed algorithm for perioperative PE diagnosis and therapeutic approach. It also includes a VTE risk assessment using the Caprini score, which might be useful to categorize patients in a high or very high-risk groups, and such patients must have an appropriate thromboprophylaxis strategy perioperatively. ¹⁴

Fig. 1.

Schematic flow diagram suggesting multidisciplinary assessment, diagnostic, and therapeutic approach in perioperative acute PE. CTPA, computed tomography pulmonary angiogram; EtCO2, end tidal carbon dioxide; LV, left ventricle; PE, pulmonary embolism; PERT, pulmonary embolism response team; SBP, systolic blood pressure; SCI, spinal cord injury; TEE, transesophageal echocardiogram; TTE, transthoracic echocardiogram; VTE, venous thromboembolism.

Thromboprophylaxis

The latest 2019 American Society of Hematology (ASH) clinical practice guidelines recommend that in patients undergoing general or abdominal-pelvic surgery in which the risk of developing VTE is very low (<0.5%), no specific pharmacological or mechanical thromboprophylaxis is recommended other than early ambulation. ²⁷ In patients at moderate risk for VTE (approximately 3%, e.g., previous history of VTE between 3 and 12 months) with low bleeding risk, the ASH guidelines recommend the use of either low molecular weight heparins (LMWH), low-dose unfractionated heparin (UFH), or intermittent pneumatic compression (IPC), over no thromboprophylaxis. For those with high risk for VTE (>6%, e.g., very recent VTE <3 months) and low bleeding risk, the ASH recommends the use of LMWH or UFH over IPC. ²⁷

There is now strong compelling evidence-based studies for the use of direct oral anticoagulants (e.g., rivaroxaban 10 mg PO every 24 hours or apixaban 2.5 mg PO every 12 hours) rather than warfarin or aspirin in selected orthopedic population undergoing elective major hip or knee arthroplasty; thus, the current 2019 ASH clinical practice guidelines currently recommend the use of direct oral anticoagulants beyond their hospital length of stay. ²⁷

In a recent meta-analysis of 4,807 patients undergoing major abdominal-pelvic oncological surgery, extended pharmacological thromboprophylaxis (4–6 weeks post-operatively) significantly reduced the risk of VTE and proximal DVT by approximately 50% when compared with conventional duration of thromboprophylaxis (<2 weeks). ²⁸

Therapeutic Approach

The perioperative management of acute PE may be challenging and complex, especially with an increased risk of major bleeding. Risks and benefits must carefully be considered as well as the severity of the PE prior to the initiation of anticoagulation and/or reperfusion strategies. A multidisciplinary pulmonary embolism response team (PERT) may be useful in assessing each clinical scenario, and together with the surgical and anesthesia teams, provide individualized recommendations. ^{5 29 30}

Anticoagulation

Current practice guidelines recommend the use of subcutaneous LMWH, rather than UFH, as initial therapy for acute PE, particularly for intermediate-low and low-risk acute PE. However, in the surgical population, the use of UFH might represent a better option due to a shorter half-life, the possibility of adjusting the dose in a short time period, and its reversal with protamine, if major bleeding is encountered. ^{29 30 31} Anticoagulation should be initiated as soon as PE is suspected, even prior to its confirmation, if the bleeding risk is considered low.

Systemic Thrombolysis

Its use during the perioperative period remains controversial given the known increased bleeding risk, which is high in the postoperative period. Major trauma or surgery within 3 weeks is considered an absolute contraindication to the use of systemic thrombolysis (ST). ⁵ However, its use may be considered in high-risk PE with cardiopulmonary collapse and/or cardiac arrest, where other reperfusion strategies are not readily available.

Catheter-Directed Therapies

Suction thrombectomy, thrombus fragmentation, and catheter-directed thrombolysis are novel therapeutic options that may be considered in those with intermediate or high-risk PE, especially in those with higher bleeding risks. These techniques may not be available in all institutions. The specific role and clinical outcomes of catheter-directed therapies have not been sufficiency studied in postoperative patients, but they represent an important alternative to consider in those patients who are not candidates for ST due to increased risk of hemorrhagic complications. ^{32 33}

Surgical Pulmonary Embolectomy

Its use is recommended for critically ill patients in which there is contraindication to ST, failure of ST, and/or catheter thrombectomy, or in whom there is insufficient time for effective ST. ^{5 30} It may also be considered in the setting of intracardiac thrombi. The mortality associated with SPE might be as low as 5 to 10% nowadays in experienced high-volume centers. ^{34 35 36 37 38}

Extracorporeal Membrane Oxygenation

In hemodynamically unstable patients the use of extracorporeal membrane oxygenation (ECMO) has shown benefit while waiting for a definitive intervention. ³⁹ ECMO offers potential to stabilize severely compromised hemodynamics in high-risk PE patients with acute RV failure, refractory hypoxia, and cardiac arrest. The venoarterial-extracorporeal membrane oxygenation configuration (VA-ECMO), can be used as a bridge to surgical embolectomy, endovascular approaches, or ST. ^{39 40}

Inferior Vena Cava Filters

Filters should be considered in patients with acute PE with absolute contraindications to systemic anticoagulation or in those patients with history of recurrent VTE despite anticoagulation. ⁵ Placement of prophylactic temporary inferior vena cava (IVC) filters may be strongly considered in patients with a very recent venous thromboembolic event (e.g., within 3–4 weeks), in which they may need prolonged interruption of systemic anticoagulation (>12 hours in the postoperative period) due to a major surgery/procedure (e.g., major orthopedic procedures, bariatric surgery, and major spinal cord surgeries). ⁴¹ After prophylactic placement of retrievable IVC filters in the carefully selected, higher risk VTE surgical patients, it is highly recommended to follow them up, either in a designated IVC filter clinic, vascular medicine, or thrombosis clinic, to decide in a timely manner when to retrieve such IVC filters.

Emphasis on Multidisciplinary Approach: Development of Evolving Pulmonary Embolism Response Teams

A critically ill, perioperative patient diagnosed with an acute PE may need immediate consultations from one or more of the following specialties: pulmonary, critical care, interventional cardiology, interventional radiology, cardiothoracic surgery, thrombosis, and/or vascular medicine. Coordinating these consults and implementing their recommendations can be time consuming for the busy surgeon/anesthesiologist and may lead to redundant communication, confusing or conflicting treatment plans, and delays in emergent care. Additionally, in the hemodynamically unstable perioperative population, obtaining a CTPA may not be possible, and on occasions, treatment decisions need to be made without a definitive diagnosis.

The initiation, organization, and implementation of PERTs facilitate and streamline care. The aim is to construct a system that would allow a rapid multidisciplinary assessment, mobilization of resources, offering a course of action from diagnosis to therapy, with the main goal of improving outcomes. ^{42 43 44 45 46} The interactive and inclusive nature of PERT has led to its dissemination and adoption in hospitals worldwide. Education and advertisement of this innovative model, across disciplines is the key.

Kabrhel et al reported the initial 30-month experience of a PERT from the Massachusetts General Hospital, describing 394 unique PERT activations during such period. Interestingly, only 18 activations (5%) were originated from the surgical floors; however, no specific data was further described in detail in regards with perioperative acute PE cases (e.g., how many PERT cases were activated either intraoperatively or postoperatively). ⁴⁷

Conclusion

Perioperative acute PE represents a challenging population, both from the diagnostic and therapeutic perspective. A better understanding of significant perioperative risk factors, availability of diagnostic tools, and multimodality risk stratification are essential in making rapid therapeutic decisions for the management of acute PE. The worldwide adoption of PERT represents an attractive proposal for the contemporary multidisciplinary approach of perioperative acute PE. With the dissemination of PERT, evolving models that include surgical and anesthesia team members with skills and training in echocardiography, remain fundamental for the appropriate functionality of a PERT, especially while actively dealing with a potential case of perioperative or intraoperative acute PE.