Acute pulmonary embolism (PE) is classified into risk categories to better assess the therapeutics options. We assessed the prognostic power of the revised Geneva score (rGs) in patients with anatomically extensive PE.
We studied 119 patients admitted with anatomically important PE (defined as bilateral extensive thrombi in contrast-enhanced chest computer tomography - CT) of whom 85.9% were treated with fibrinolysis. All patients were afterwards classified using the Pulmonary Embolism 2008 ESC guidelines risk-score, using markers of ventricular dysfunction (positive troponins or right ventricular dilation/dysfunction (RVDD) in low-risk and intermediate risk patients. Patients with shock/hypotension were excluded, as they form a risk-group themselves. Population was classified using the rGs into low (0–3), intermediate (4–10) and high-risk (>=11) of having PE.
No differences were found regarding clinical and laboratory variables among groups of the rGs, namely age, body mass index, systolic pressure or troponin levels. As expected, patients with higher scores had higher heart rates. Although all patients analyzed on this study had anatomically important PE on CT, on admission 14.7% of them were classified by the rGs in the low probability strata, whereas 81.1% were classified as intermediate-probability and 4.2% within high-probability. Regarding the 30-day mortality risk, 50% of low-probability patients on admission based on rGs belonged to the intermediate-risk category, whereas 74% of intermediate-probability were classified in intermediate-risk category. No patient with a high-probability rGS was classified in the low-risk category. The risk of having RVDD was higher with higher scores in the rGs: 75% in the lower, 83.7% in the intermediate and 100% in the high-probability group (p=NS). Regarding 30-day mortality, none of the patients with low-probability rGs died. The mortality rate in patients with scores >= 4 was 6.5%.
Although the rGs is a robust score, in this population with extensive PE on CT, almost 15% of these patients were classified on the low-probability strata.
The rGs performed well identifying the lower risk patients, as none of them died in the 30-day follow-up.
Rui Baptista, No Financial Disclosure Information; No Product/Research Disclosure Information