• embolia polmonare
• evoluzione a lungo termine
• follow-up

Background: Pulmonary embolism (PE) is potentially lethal acutely and prone to recur. After anticoagulant therapy discontinuation, the rate of recurrences increases with time and therefore decide when and how to stop secondary prophylaxis may be difficult in several patients.
Objective: to investigate the incidence of mortality and recurrence of PE and the time course of these events during a 5 year-follow-up in patients who discontinue and in those who continue secondary prophylaxis only according to 2001 ACCP Guide-lines. Secondary purposes were to evaluate the risk factors for recurrences and mortality of PE in patients who continue and in those who discontinue therapy and the incidence of bleeding in the former group of patients.
Design and methods: we evaluated consecutive patients with acute PE enrolled in a single University centre during a 5 year-period (2001-2005), and followed for 5 further years (2006-2011) after the decision to continue or stop anticoagulant treatment on the basis of 2001 ACCP Guide-lines.
Results: we considered 471 patients with PE that were followed for 5 year follow-up period. Among them, 361 (76.6 %) continued anticoagulant therapy, the remaining 110 (23.4 %) patients stopped it. Among all patients, 263 (55.8%) completed the 5 year-follow-up period while 122 (25.9%) patients died before completing it and 86 (18.3%) were lost to follow-up. Patients who continued anticoagulant therapy experienced 34 (72.3%) recurrences (incidence recurrence rate 2.58 events for 100 person-years), while patients who discontinued therapy developed 13 (27.7 %) recurrences (incidence recurrence rate 2.92 events for 100 person-years). In the group of patients who continued anticoagulant therapy, 109 (89.3 %) deaths (mortality rate 8 events for 100 person-years) occurred, while in the group who discontinued it 13 (10.7 %) deaths (mortality rate 2.84 events for 100 person-years) occurred; the difference between the two groups was statistically significant (RR 2.81, 95% CI 1.58-5.46). The group of patients who continued anticoagulant therapy experienced 13 fatal recurrences (incidence recurrence rate 0.95 events for 100 person-years) while patients who discontinued it had 4 fatal recurrences (incidence recurrence rate 0.8 events for 100 person-years). Among patients who continued anticoagulant therapy, those with unprovoked or recurrent PE show more than 3 times grater probability of dying for fatal recurrence than those who continued treatment since affected by uncontrolled cancer or disabling chronic illness (RR 3.60, 95% CI 001-1.84). Recurrence was significantly associated with age greater than 61 years and with the presence of DVT at the diagnosis of PE. Conditions independently associated with all-cause mortality, were age greater than 70 years, presence of cancer, and continuation of anticoagulant therapy. Fatal recurrences, were significantly associated with age greater than 78 years.
Conclusion: our findings suggest that, according to Guide-lines, most patients (80%) should be anticoagulated for long period of time; that prolonging anticoagulation markedly reduce the risk for fatal recurrence in all patients and, mostly, in those with unprovoked PE, while it does not decrease total recurrences or all-cause mortality. This, indeed, is influenced greatly by the presence of cancer, both in patients who continue and in those who stop anticoagulant prophylaxis. Whether or not the strategy of adhering to Guide-lines is correct remains not demonstrated.