Pulmonary embolism (PE) is a condition characterised by an obstruction of the pulmonary arterial system by one or more emboli (National Institute of Health and Care Excellence (NICE), 2019a). In the authors' experience, patients often present with non-specific symptoms, which makes diagnosis of a PE difficult. In addition, the assessment and diagnosis of PE can vary across specialties and clinicians. In this discussion the authors will critically analyse and discuss the evidence around the diagnosis and management of PE.
Advanced clinical practitioners (ACPs) are often faced with ruling out a diagnosis of PE in patients with non-specific symptoms such as dyspnoea and pleuritic chest pain, which can be fairly mild, and therefore a diagnosis of PE easily missed (Geersing et al, 2012). In addition, patients may also present with symptoms such as haemoptysis, fever and tachycardia (European Lung Foundation, 2019). Pulmonary embolism can be a challenge to diagnose, especially in elderly people, since it can be difficult to differentiate their symptoms from other less serious illnesses (Ma et al, 2017). These may include upper or lower viral respiratory tract infection, undiagnosed or exacerbation of chronic obstructive pulmonary disease (COPD) and bronchiectasis.
Every 37 seconds someone in the Western world dies from a VTE, which is recognised as a leading cause of preventable deaths in hospitals (Thrombosis UK, 2019). In the UK 55–60% of VTE cases occur during or following hospitalisation (Thrombosis UK, 2019).
Risk factors
Risk factors for pulmonary embolism include pregnancy, cancer, obesity, previous deep vein thrombosis (DVT) or PE, varicose veins, recent surgery, hospitalisation and immobilisation or lower limb trauma within the previous 12 weeks (NICE, 2019a). An assessment of the clinical probability of PE should be the first step in the assessment of a patient suspected to have a PE. Family history of VTE has also been identified as a risk factor for patients developing VTE (Kelly et al, 2018). Therefore, for patients with a family history of VTE presenting to the emergency department (ED) with symptoms such as chest pain and dyspnoea, an assessment for VTE should be mandatory (Kelly et al, 2018).
Diagnostic tools
The following tools—Wells scoring system, D-dimer test, computed tomography pulmonary angiography (CTPA) and the VQ (ventilation perfusion) scan—are frequently used by ACPs and other health professionals to diagnose VTE and PE. However, deciding what test and in which order often depends on local guidelines and clinical acumen.
Wells scoring system
The Wells score is the most widely used pre-test clinical probability indicator of PE in the UK (Wells et al, 1997). It scores the patient's probability of having a PE based on their risk factors (Kline, 2018). The nine-component Wells score for DVT was originally developed by Wells et al in 1997 followed by a seven-component risk assessment score for PE in 1998. The two-tier score is divided into ‘PE likely’ (a score of more than four points), or ‘PE unlikely’ (a score of four points or less) (Kline, 2018; NICE, 2019a). The Wells score has been shown to be accurate in predicting PE in hospitalised patients (Posadas-Martínez et al, 2014). Furthermore, the Wells score has been found to be more effective in predicting PE than the Geneva score, another risk assessment tool (Ma et al, 2016). However, that study was only a small single-centre study and should therefore be replicated in a large multi-centre environment to increase validity of the results.
D-dimer test
The D-dimer test is a useful and relatively simple investigation to rule out VTE. Fibrin D-dimer is a degradation product of cross-linked fibrin that increases in the presence of VTE (Goldhaber and Bounameux, 2012). It has been found to have 95% sensitivity (Goldhaber and Bounameux, 2012). However, the D-dimer is not specific enough to diagnose a PE in isolation, since it can be elevated in a range of other conditions such as pregnancy, peripheral vascular disease, cancer and inflammatory diseases (British Thoracic Society, 2003).
D-dimer concentration also increases with age and therefore its specificity for PE decreases (Douma et al, 2010). This has prompted numerous researchers to investigate whether an age-adjusted D-dimer should be used as a method to exclude PE rather than the standard cut off value of 500 commonly used in most laboratories in the UK (Douma et al, 2010; Kline, 2018). It has been suggested that an age-adjusted D-dimer would be more specific in diagnosing PE in older adults, reducing the number of patients investigated for PE and potentially reducing costs and harm to patients (Dutton et al, 2018). However, the exact age cut off figure has yet to be decided. There have been many studies that have measured D-dimer assays as 10 times the patient's age, in those aged over 50 years, which have supported the age-adjusted D-dimer as a safe method to exclude PE in those assessed as having low or moderate probability (Douma et al, 2010; Penaloza et al, 2012). In these studies, a higher percentage of older adults had a normal D-dimer concentration when the adjusted D-dimer assay was used, reducing the need for further harmful investigations. In contrast, a more recent study investigated whether five times the patient's age would be a more appropriate target range (Dutton et al, 2018). However, this was a small-scale study and therefore further large-scale randomised control trials (RCTs) are required to investigate this further.
The British Thoracic Society (BTS) (2003) guidelines advise that a D-dimer should be checked only once the probability of PE has been assessed. Furthermore, they suggest that the D-dimer should not be performed in those with a high probability of PE—these patients should proceed straight to imaging (BTS, 2003). This is also supported by the American College of Physicians, since a negative D-dimer in a patient with a high probability of PE should not remove the need for imaging (McCarthy, 2015).
In the authors' experience, many clinicians will often request a D-dimer for a patient without first assessing the Wells score and therefore risk assessing the patient for the likelihood of a PE. Obviously, the D-dimer is often found to be elevated in older patients, which prompts further investigation with imaging.
Computed tomography pulmonary angiography
CTPA is regarded as the gold standard imaging modality for investigation of acute PE, but should be undertaken after an assessment of probability of PE (BTS, 2003). Studies have shown that a combination of risk assessment, D-dimer testing and CTPA is the most preferred diagnostic method for diagnosis of a PE (Gao et al, 2018). CTPA remains far superior to alternative imaging modalities such as the VQ scan, as demonstrated in the literature (Anderson et al, 2007). Furthermore, CTPA has been shown to detect more pulmonary emboli than the VQ scan (van Es et al, 2015).
Studies have shown that clinicians will often request CTPA for patients deemed low risk, leading to exposure to hazardous radiation, increasing the risk of breast cancer, especially in those aged under 40 years, in addition to the risk of contrast-induced nephropathy (McCarthy, 2015; Takach et al, 2015; Ma et al, 2017). Ma et al (2017) established that only 10–15% of elderly patients were confirmed to have a PE following a CT scan, demonstrating the amount of unnecessary CTPAs performed, especially in older people. It is recognised that older patients are at risk of developing age-related nephropathy, and this can be compounded further by contrast being administered for the CT scan, thereby increasing their risk of developing a contrast-induced nephropathy.
VQ scans
VQ scans be used as an alternative imaging technique for diagnosing PE where CTPA is contraindicated such as those with chronic kidney disease, elderly people and pregnant women (Grippi and Elias et al, 2015). A VQ scan is a nuclear medicine scan that uses radioactive material to examine ventilation and perfusion of the lungs (Jong, 2018). During the first part of the scan, radioactive material is inhaled and the ventilation of the lungs is analysed. During the second part, radioactive material is injected into a vein and the perfusion of the lungs is studied (Jong, 2018).
A VQ scan should not be used in those who are critically ill, obese or have underlying pulmonary disease since the test involves the patient lying supine and can take significantly longer than a CTPA (Jong, 2018). Furthermore, CTPA would be preferable to a VQ scan in a patient who may be suitable for an embolectomy as it allows mapping of the location and extent of clot burden in the pulmonary arteries, which a VQ scan does not (Jong, 2018). However, an additional limitation of the VQ scan is that around 50% of VQ scans do not give a definitive diagnosis and therefore other tests, such as a CTPA, would be required (van Es et al, 2015). The clinician would then be faced with the decision of whether to further investigate with a CTPA if PE was still clinically a possible diagnosis, increasing the risks to the patient and subjecting the patient to additional investigations. However, since the VQ scan exposes the patient to a significantly lower dose of radiation with no side effects this is in its favour and therefore a safer alternative to CTPA in pregnancy and younger patients (van Es et al, 2015).
The overuse of investigations such as D-dimer testing and CT scanning has been shown to not actually improve patient outcomes, but rather increase the risk of harm to the patient and expense to the hospital trust and the NHS as an organisation (McCarthy, 2015, Raja et al, 2015). CTPA remains the gold standard imaging modality to diagnose a PE despite the risks. However, an assessment of those risks should be calculated by the clinician and a VQ scan considered if the risks are deemed too high. This would be with the knowledge that a VQ scan has a lower specificity than the CTPA.
Sub-segmental PE
There appears to be a lack of consensus in clinical practice on whether a sub-segmental PE, localised in the distal branches (sub-segmental) pulmonary arteries, should be treated with anticoagulants. It has been suggested that treatment should be commenced for sub-segmental PE only if the patient has symptoms, an elevated D-dimer, risk factors for VTE or a previous VTE (Kline, 2018).
Treatments
The following treatments are used in patients with PE:
Thrombolysis
European guidelines advise that thrombolysis is indicated in clinically unstable patients presenting with a submassive or massive PE on admission and in the absence of any contraindications (Konstantinides et al, 2014). A submassive PE is an acute PE without systemic hypotension (systolic BP ≥90 mmHg) and can be classed as intermediate risk. A massive PE involves sustained hypotension, with a SBP of ≤90 mmHg for at least 15 minutes or requiring inotropic support, not due to a cause other than PE, or persistent profound bradycardia. (Malik et al, 2016). Thrombolysis is, however, underutilised in clinical practice due to the fear of adverse bleeding events (Zuin et al, 2019). The optimal therapeutic window for administration of thrombolysis has not yet been established (Zuin et al, 2019). A review of the data from 374 patients deemed at high risk of PE over a 7-year period concluded that thrombolysis administered within 8.5 hours from onset of symptoms may be associated with a reduction in the 30-day cardiovascular mortality. However, owing to the small population size, further large-scale RCTs are required to evaluate this (Zuin et al, 2019).
Anticoagulant therapy
Patients without a massive or submassive PE are treated with anticoagulant therapy, usually commencing with subcutaneous low-molecular-weight heparin (LMWH) and switching over to a direct oral anticoagulant (DOAC). There has been a shift away from treatment with warfarin for prevention and treatment of VTE over the last decade, which was previously the only oral anticoagulant approved for prevention and treatment of VTE (Rosovsky and Merli, 2017). In contrast to warfarin, the newer oral anticoagulants only inhibit one component in the clotting cascade. Dabigatran affects prothrombin, factor II and edoxaban, rivaroxaban and apixaban affect factor Xa (Rosovsky and Merli, 2017). In addition to increased safety with the DOACs, the patient requires no monitoring with these drugs in comparison to warfarin, which requires regular blood tests and monitoring of their international normalised ratio (INR) level (Rosovsky and Merli, 2017). Warfarin also has many other limitations; it interacts with other medications and foods and has a narrow therapeutic window that increases the risk of bleeding and can therefore cause catastrophic bleeding in patients who are not compliant with their treatment or in those patients in whom maintaining a steady stable INR is difficult (E-Medicines Compendium (EMC), 2017).
A meta-analysis that systematically reviewed 10 clinical trials comparing the DOACs with standard VTE treatment of parenteral anticoagulants and vitamin K antagonists, found that DOACs were associated with fewer recurrent VTEs, fewer major and fatal bleeds and reduced mortality (Gómez-Outes et al, 2015). However, DOACs have an important limitation; warfarin can easily be reversed with vitamin K in the event of an overdose, but there is currently no standard reversal agent for the DOACs available except for dabigatran (Joint Formulary Committee, 2019a). Another advantage over warfarin in that DOACs have a shorter half-life, as once the last dose has been taken the drug will be out of their circulation after 12 to 24 hours. This is in contrast to warfarin, which can take up to 7 days. DOACs are thus generally safer than warfarin (Joint Formulary Committee, 2019b). Unfortunately, since DOACs are renally excreted, they cannot be used in patients who have severe renal impairment (Joint Formulary Committee, 2019a).
Rivaroxaban tends to be the drug of choice for treatment and prevention of recurrent VTE in the authors' local trust. Usually rivaroxaban is administered orally at a dose of 15 mg twice daily for 21 days then 20 mg once daily thereafter (Bauersachs et al, 2010). This is supported by NICE guidance, which recommends treatment of DVT and PE and prevention of recurrence of VTE with rivaroxaban (NICE, 2013). The exact duration of treatment is not specified but must be continued for at least 3 months for those with transient risk factors such as trauma, but potentially for longer duration for those with more permanent risk factors or unprovoked DVT or PE (NICE, 2013). A large RCT, which was shown by the manufacturers to demonstrate the effectiveness of rivaroxaban, was the Einstein Investigators PE study (Büller et al, 2012; NICE, 2013). This study demonstrated that rivaroxaban is as effective for prevention of recurrent DVT and PE compared with LMWH and warfarin therapy, with the added advantage of significantly less bleeding (Büller et al, 2012). This study demonstrated that a single DOAC was as effective as combination treatment with LMWH and warfarin, which is likely to improve patient concordance and improve patient outcomes (Büller et al, 2012). Rivaroxaban has to be taken with food and, as yet, does not have a reversal agent but does still offer a safer alternative to warfarin (Rosovsky and Merli, 2017). Potential reversal agents are currently still undergoing clinical trials.
The DOACs have not yet been compared to each other in clinical trials, but clearly offer a safer and more effective treatment and prevention of recurrent VTEs than previous conventional treatments such as warfarin. Rivaroxaban is the recommended DOAC to treat and prevent recurrence of VTE but further research in the long-term effects of rivaroxaban are required (NICE, 2013).
VTE risk assessment
In addition to the assessment and management of patients presenting to hospital with symptoms of VTE, another vital role of the ACP is assessment and prevention of VTE in those patients admitted to hospital as an elective admission or with symptoms of another disease process. The authors have experience of completing these risk assessments and ACPs are often the role models in the medical team by ensuring these risk assessments are completed correctly and in a timely manner. NICE (2018a) advocate that all medical, surgical and trauma patients should have their VTE and bleeding risk assessed using a recognised national tool as soon as possible after admission to hospital. Furthermore, those patients assessed as at risk of VTE and not at increased risk of bleeding should be offered VTE prophylaxis (NICE, 2018). If assessed as requiring pharmacological VTE prophylaxis then this should be administered within 14 hours of admission (NICE, 2019b).
NICE (2019b) also advocates reassessment if the patient's condition changes—in the authors' experience this is poorly performed. For example, a patient may present with a possible gastrointestinal (GI) bleed, and after risk assessment it is decided that pharmacological VTE prophylaxis would not be appropriate. However, if that patient is later found to not have had a GI bleed and there is no other reason not to receive VTE prophylaxis, and they are not reassessed, then they could potentially develop a hospital-acquired VTE. Therefore, risk assessment is important on admission and reassessment is needed if their condition changes. Although a mandatory requirement, VTE risk assessment can go uncompleted and may not be completed correctly. The National VTE risk assessment tool (NICE, 2019b) is recommended for use in the UK. However, there are concerns that this tool has not been evaluated against other tools to assess its efficacy and accuracy (NICE, 2019b). It is widely recognised that further research is needed in this area.
If a patient's risk of bleeding is too high to receive pharmacological prophylaxis, they can be offered mechanical prophylaxis in the form of compression stockings. However, there needs to be further assessment of the patient before these can be applied (NICE, 2019b). Unfortunately, anti-embolism stockings cannot be offered to a patient with peripheral arterial disease, local tissue damage, extensive oedema or peripheral neuropathy (NICE, 2019b). The decisions not to provide a patient with VTE prophylaxis must be clearly documented and explained to the patient (NICE, 2019b).
Conclusion
ACPs are expected to autonomously clinically assess a patient admitted with non-specific symptoms such as pleuritic chest pain and dyspnoea. They should follow local and national guidance on assessing the probability of PE, performing D-dimer if indicated and then deciding if imaging is required. Assessment of patients with possible diagnosis of PE is often done poorly in practice and diagnosis can be difficult due to the often non-specific symptoms with which patients present. In the authors' opinion, ACPs are best placed to promote evidence-based practice and therefore assess each patient carefully and perform only the most appropriate and clinically indicated investigations. ACPs should be confident in the processes to follow in diagnosing and managing patients suspected of, and diagnosed with, a PE, always acting within their scope of professional practice. However, due to the difficulty of diagnosis, there may be times that further advice from the wider multidisciplinary team is needed to ensure the correct management of the patient. There is a requirement to move away from subjecting patients to needless and invasive investigations that are not clinically indicated, can cause patients harm and increase costs to the NHS.