A couple weeks back, I posted my algorithm regarding (not) evaluating patients with chest pain for pulmonary embolism. As has been written multiple times – most recently in this BMJ article – the evidence for overdiagnosis is rather overwhelming, and I’m trying to come up with strategies to do my small part to reduce it.
These Dutch authors, however, perform a retrospective analysis of prospectively-collected data and conclude “patients with symptomatic [subsegmental pulmonary embolism] appear to mimic those with segmental or more proximal PE as regards their risk profile and short term clinical course.” If correct, it would imply there are no “clinically insignificant” PE – which flies in the face of our evidence of overdiagnosis.
These authors found 116 SSPE, 632 proximal PE, and 2980 patients without PE. They found the folks with proximal and SSPE, by almost every measure, had significantly more thromboembolic risk factors – particularly malignancy – than the folks without PE. Unsurprisingly, given the identical VTE risk profile between proximal and SSPE, there was essentially no difference in rate of recurrent VTE in the proximal and SSPE groups during 3 month follow-up – 4 patients (3.6%) with SSPE and 14 patients (2.5%) for proximal PE. Given an absence of risk factors, the patients without PE at baseline had only 1.1% incidence of VTE during the follow-up period. For all-cause mortality, the risk was 10.3% for SSPE, 6.3% for proximal PE, and 5.4% for patients without PE at baseline. 1.6% of both the SSPE and proximal PE group suffered major bleeding complications from anticoagulation.
So, there is some truth to the authors’ conclusion that SSPE has a clinical course similar to proximal PE. However, the clinical significance of SSPE is almost certainly confounded by co-occurrence of comorbid conditions. It is reasonable to suggest true PE and SSPE are poor prognostic indicators of background disease burden, and not the salient pathologic diagnosis. Anticoagulation may not be the most important course of action; rather, identifying and treating the underlying cause, if present.
This article ought not support any argument regarding the necessity of diagnosis and treatment of sub-segmental PE, except in the context of a broader approach to a patient.
“Risk profile and clinical outcome of symptomatic subsegmental acute pulmonary embolism”
www.ncbi.nlm.nih.gov/pubmed/23736701
Great post. Very provocative article.
I would sub-classify SSPE into three groups:
(a) Multiple SSPE
(b) Single SSPE with positive DVT study
(c) Single SSPE with negative DVT study
I think most practitioners would consider (a) or (b) to be definite venous thromboembolic disease and treat it as such with anticoagulation. There is significant disagreement currently regarding group (c) however.
Unfortunately this study combines all three groups into a single group. As such, this study can’t answer the real question – what to do with group (c).
Josh
Great post. Very provocative article.
I would sub-classify SSPE into three groups:
(a) Multiple SSPE
(b) Single SSPE with positive DVT study
(c) Single SSPE with negative DVT study
I think most practitioners would consider (a) or (b) to be definite venous thromboembolic disease and treat it as such with anticoagulation. There is significant disagreement currently regarding group (c) however.
Unfortunately this study combines all three groups into a single group. As such, this study can’t answer the real question – what to do with group (c).
Josh
I agree. The fact is you don't die of the PE you've had,you die of the next one.
The critical factor for prognosis is residual clot load (or the propensity to form new clot if there is a thrombophilia,malignancy etc).
I agree. The fact is you don't die of the PE you've had,you die of the next one.
The critical factor for prognosis is residual clot load (or the propensity to form new clot if there is a thrombophilia,malignancy etc).
Agree with both above – PE is a heterogenous disease, and, it's really a "symptom" of sorts of the underlying disease process. Thrombophilia secondary to – short-term pro-coagulopathic state? Long-term malignancy with ultimate demise? Just as in cardiology, all elevated troponins are not ACS – we should be smart enough to tailor treatment to the individual, not just universally apply a textbook treatment.
Agree with both above – PE is a heterogenous disease, and, it's really a "symptom" of sorts of the underlying disease process. Thrombophilia secondary to – short-term pro-coagulopathic state? Long-term malignancy with ultimate demise? Just as in cardiology, all elevated troponins are not ACS – we should be smart enough to tailor treatment to the individual, not just universally apply a textbook treatment.