Outpatient treatment of pulmonary embolism

Pulmonary embolism (PE) is traditionally treated in hospital. Growing evidence from non randomized prospective studies suggests that a substantial proportion of patients with non-massive PE might be safely treated in the outpatient setting using low molecular weight heparins. Based on this evidence, professional societies started to recommend outpatient care for selected patients with non-massive PE. Despite these recommendations, outpatient treatment of non-massive PE appears to be uncommon in clinical practice. The major barriers to PE outpatient care are, firstly, the uncertainty as how to identify low risk patients with PE who are candidates for outpatient care and secondly the lack of high quality evidence from randomized trials demonstrating the safety of PE outpatient care compared to traditional inpatient management. Also, although clinical prognostic models, echocardiography and cardiac biomarkers accurately identify low risk patients with PE in prospective studies, the benefit of risk stratification strategies based on these instruments should be demonstrated in prospective management studies and clinical trials before they can be implemented as decision aids to guide PE outpatient treatment. Before high quality evidence documenting the safety of an outpatient treatment approach is published, outpatient management of non-massive PE cannot be generally recommended.


Introduction
This article was partly supported by grant 33CSCO-122659 from the Swiss National Science Foundation.

Benefits of PE outpatient care
The potential benefits of PE outpatient care over traditional inpatient care include an improvement in health-related quality of life and increased physical activity and social functioning.[11,12] Moreover, the implementation of outpatient treatment strategies is likely to reduce the length of hospital stay and may result in substantial cost savings.[13] We used projections from a prior cost-effectiveness analysis to estimate the potential economic impact of outpatient treatment of PE in Switzerland.Assuming a cost difference of $ 4500 between inpatient and outpa-tient treatment of PE and an annual PE incidence of 3,133 cases, over $ 7 million per year could be saved in Switzerland if 50% of PE patients were treated as outpatients.[13,14] Many hospitals could readily employ the existing infrastructure used to treat patients with DVT as outpatients in low risk patients with PE (e.g., anticoagulation clinics and outpatient treatment protocols).The projected cost savings as a result of outpatient care of PE is likely to far outweigh the implementation costs for any outpatient treatment strategy.

Barriers to PE outpatient care
To our knowledge, there are no published data on the utilisation of PE outpatient care but reports from Europe and Australia suggest that outpatient treatment of non-massive PE is uncommon.[15][16][17] Two major barriers exist to PE outpatient care.Firstly, there is uncertainty as to how to identify low risk patients with non-massive PE who may be safely treated in the outpatient setting.The eligibility criteria of previous studies of PE outpatient treatment were heterogeneous and vague (e.g., "comorbid conditions that necessitate hospitalisation") and are difficult to reproduce with uniformity or confidence (table 1).Also, practice guidelines recommending PE outpatient care fail to specify the details as to how to select low risk patients with PE who could be safely treated as outpatients.[7][8][9] The lack of prognostic criteria to identify low risk patients with PE and physicians' insecurity in assessing baseline risk to the patient is reflected by the large variation in length of hospital stay for PE and may have a negative impact on patient outcomes.[18] A retrospective cohort study demonstrated that after adjustment for patient and hospital factors, patients with a relatively short hospital stay of four days or less had a significantly higher 30-day mortality that those with a typical length of stay of five to eight days (OR 1.55, 95% CI: 1.22-2.00).[18] These results suggest that physicians may inappropriately select patients with PE for early discharge who are at increased risk of complications.

Risk stratification of patients with PE
As the subjective judgment of the physician may fail to identify patients with PE who have a good prognosis and who may be safely treated in the outpatient setting, several objective prognostic instruments may help physicians to identify low risk patients with PE who are potential candidates for outpatient care: clinical prognostic models, imaging procedures and cardiac biomarkers.

Clinical prognostic models
The Geneva Prognostic Score (GPS) is based on six clinical, laboratory and ultrasonographic variables to predict the combined adverse outcome of death, recurrent venous thromboembolism and major bleeding episodes during the first three months following the index PE (table 2).[21] Low risk patients based on the GPS (≤2 points) have a low rate of adverse outcomes (2.2-5.0%), with a sensitivity of 58-85% and a negative predictive value of 95-98% for predicting adverse outcomes.[21,22] The most extensively validated clinical prognostic model is the Pulmonary Embolism Severity Index (PESI) that accurately stratifies patients into five risk classes (I-V) with increasing risk of all cause short term mortality, ranging from 1.1% in class I to 24.5% in class V (table 3).[23] The PESI comprises eleven routinely available clinical parameters without any need for ultrasonography or laboratory studies.[23] Patients in risk classes I and II have a 30-day all cause mortality of 0.9-2.6%only and are considered low risk.[23,24] When dichotomized as low (classes I/II) versus higher risk (classes III-V), the PESI has a sensitivity of ≥90% and a negative predictive value of 98-100% for predicting mortality.[24,25] Jiménez et al. retrospectively compared the prognostic accuracy of the GPS and the PESI in 599 patients with PE. [25] The GPS identified a significantly higher proportion of patients with PE as low risk than the PESI (84% vs 36%, P <0.001).However, low-risk patients based on the GPS had a significantly higher 30 day mortality (5.6% vs 0.9%, P <0.001), resulting in a lower sensitivity for overall mortality compared to the PESI (35% vs 95%).The proportions of low risk patients who had any adverse outcome at 30 days (death or non fatal recurrent venous thromboembolism or major bleeding) were similar for both scores.Because both scores were primarily devel-

Table 2
The Geneva Prognostic Score.

Table 3
The Pulmonary Embolism Severity Index.oped to identify low risk patients with PE, their positive predictive value for predicting mortality was low (<20%).

Imaging
A meta-analysis including five prospective studies of haemodynamically stable patients with PE demonstrated that patients without echocardiographic right ventricular (RV) dysfunction, defined by a RV wall hypokinesis, RV dilatation or an increased right/left ventricular end diastolic diameter ratio, had a short term all cause mortality of 3% only.[26] Although patients without RV dysfunction (56% of patients with PE) appear to have a low mortality, whether such patients can be safely treated as outpatients has not been prospectively evaluated.Moreover, the practical use of echocardiography for risk stratification is limited by its operator dependence, cost, and lack of availability 24 hours a day in many hospitals.
Whether RV dilatation on spiral computed tomography (CT) [26][27][28][29] and CT-based pulmonary artery obstruction indices [30][31][32][33][34] are independent predictors of all cause mortality and adverse events in patients with PE is still controversial.The safety and efficiency of these CTbased measures to identify low risk patients with PE who are candidates for outpatient care has never been prospectively validated.A recent study demonstrated that concomitant deep vein thrombosis shown by ultrasonography is associated with a fourfold increase in short-term overall mortality.[35] Whether ultrasonography can be used to risk stratify patients with PE must be assessed in further studies.

Cardiac biomarkers
The likely explanation for the release of cardiac biomarkers in patients with more severe PE is the development of RV microinfarctions (troponins) or cardiomyocyte stretch (brain natriuretic peptides).A meta-analysis including seven prospective studies demonstrated that elevated troponin I or T levels were significantly associated with short term all cause mortality in haemodynamically stable patients with PE (OR 5.9, 95% CI: 2.7-13.0)and that patients with normal troponin levels (21% of patients with PE) had a mor-tality of only 2.3%.[36] A meta-analysis of 13 studies demonstrated that elevated brain natriuretic peptide levels (BNP or NT-pro-BNP) were significantly associated with short term all cause mortality in haemodynamically stable and unstable patients with PE (OR 7.6, 95% CI: 3.4-17.1)[37].Patients with normal BNP/NT-pro-BNP levels (49% of patients with PE) had a mortality of 1.7% only.[37] A meta-analysis including haemodynamically stable patients showed similar results.[26] Overall, patients with PE who have normal cardiac biomarkers appear to have a low risk of overall short term mortality.However, the practical use of cardiac biomarkers in therapeutic decision making is currently limited by a lack of test standardisation (multiple assays and cut-off points used) and the absence of clinical studies demonstrating the safety of PE outpatient care among patients with normal biomarkers levels.Whether other novel cardiac biomarkers such heart-type fatty acid binding protein or growth differentiation factor-15 may be useful in identifying low risk patients with PE who may be candidates for outpatient care must be further evaluated.[38,39]

Combination of prognostic models, echocardiography and cardiac biomarkers
Several authors proposed outpatient care for low risk patients identified using risk stratification algorithms based on haemodynamic status, the PESI, biomarkers and/or echocardiography.[40][41][42] According to these algorithms, haemodynamically stable low risk patients based on the PESI (or those with normal echocardiographic RV function or BNP/NT-pro BNP values) who have normal troponin levels should be considered for outpatient care.However, the safety and efficiency of algorithms using echocardiography and cardiac biomarkers to identify low risk patients with PE has never been prospectively validated.Whether outpatient treatment of haemodynamically stable low risk patients based on the PESI is as safe and efficient as inpatient care is currently being evaluated in the international, randomized Outpatient Treatment of Pulmonary Embolism (OTPE) trial (NCT00425542).Results from this trial will become available by late 2010.

688
Home care for pulmonary embolism

Conclusion
Outpatient care of patients with non-massive PE using low-molecular weight heparins is logistically feasible.However, it remains uncertain how to best identify low risk patients with PE who are candidates for outpatient care and whether outpatient treatment for non-massive PE is really as safe as traditional treatment in hospital.Although clinical prognostic models, echocardiography and cardiac biomarkers accurately identified low risk patients with PE in prospective studies, the clinical impact of these prognostic measures on the safety and efficiency of outpatient care remains unclear.Any outpatient treatment strategy based on these risk stratification tools should be evaluated in prospective management studies and clinical trials before such a strategy can be implemented.Before high quality evidence documenting the safety of an outpatient treatment approach becomes available, outpatient management of non-massive PE must be decided on an individual basis and cannot be generally recommended.The site of treatment decision must also consider psychosocial contraindications to outpatient care.For example, patients who use intravenous drugs or who are alcoholic or unreliable or have severe psychiatric conditions may require hospitalisation to ensure adherence to treatment regardless of the severity of their illness.
73%) of these emergency physicians reported a willingness to consider this treatment option if high quality empiric data supporting the effectiveness and safety of the outpatient management of PE were available.The historical example of DVT demonstrates that evidence from randomized trials has the potential to change clinical practice.Before 1996, DVT was mostly managed in hospital.After the publication of clinical trials demonstrating the safety of DVT outpatient care compared to traditional inpatient care,

Points assigned
A total point score for a given patient is obtained by summing the patient's age in years and the points for each applicable predictor.Points assignments correspond with the following risk classes: ≤65 class I; 66-85 class II; 86-105 class III; 106-125 class IV; and >125 class V. Patients in risk classes I and II are defined as low-risk.*Defined as a history of cancer or active cancer.†Defined as disorientation, lethargy, stupor, or coma.‡ With and without the administration of supplemental oxygen.