- Top of page
- PATIENTS AND METHODS
Essential thrombocythaemia (ET) is a chronic myeloproliferative disorder characterized by the occurrence of thromboembolic episodes, particularly in patients aged > 60 years or with a previous history of thrombosis, and/or by haemorrhages in patients with an exceedingly high platelet count. In these subgroups of patients the use of cytoreductive therapy is beneficial in terms of risk/benefit ratio. Only limited anecdotal data are available on the thrombotic or haemorrhagic risk and survival in young asymptomatic ET patients with a platelet count < 1500 × 109/l. Therefore the optimal management of these patients is unknown.
To assess the incidence of thrombosis and haemorrhages in this group of patients we carried out a prospective observational study in a cohort of 65 patients with ET, aged < 60 years, with no history of thrombosis or haemorrhage and platelet count < 1500 × 109/l, and in 65 age- and sex-matched controls. Patients were not treated with cytoreductive therapy until the occurrence of thrombosis or haemorrhage. Arterial or venous thrombotic events were objectively documented both in cases and in controls.
The median follow-up was 4.1 years, with an incidence of thrombosis in patients and controls of 1.91 and 1.50 cases/100 patient-years, respectively. The age- and sex-adjusted risk rate ratio was 1.43 (95% CI 0.37–5.4). Only three minor haemorrhagic episodes occurred in patients, with an incidence of 1.12 cases/100 patient-years.
Pregnancy and surgery were not associated with thrombosis in these patients.
We conclude that the thrombotic risk in young ET patients, with no thrombotic history and a platelet count < 1500 × 109/l, is not increased compared to the normal population and that a conservative therapeutic approach should therefore be considered in these patients.
Essential thrombocythaemia (ET) is a clonal myeloproliferative disorder characterized by chronic peripheral thrombocytosis and by an abnormal proliferation of marrow megakaryocytes ( Laslo, 1975). Although the majority of the patients are asymptomatic at diagnosis ( Barbui et al, 1996 ) and the overall survival is comparable to that of general population ( Rozman et al, 1991 ), thrombotic and haemorrhagic complications represent the major cause of morbidity ( Schafer, 1984). Therefore the main challenge for the clinician is to select patients who ultimately will benefit from a cytoreductive or antiplatelet therapy, notwithstanding the potentially life-threatening side-effects of these treatments ( Weinfeld et al, 1994 ).
In a retrospective study Cortelazzo et al (1990 ) identified age > 60 years and previous thrombotic events as major risk factors for thrombosis. In such high-risk patients, treatment with hydroxyurea was found to be beneficial by significantly reducing the rate of thrombotic events ( Cortelazzo et al, 1995 ). Cytoreduction has also been suggested for patients with a platelet count > 1000 × 109/l who are at greater risk of haemorrhagic complications ( Van Genderen et al, 1993 ).
In contrast, no prospective data, apart from anecdotal evidence ( Barbui et al, 1996 ), are available on the thrombotic and haemorrhagic risk in young asymptomatic ET patients who do not have an exceedingly high platelet count. Therefore there is no clear indication as to the use of cytoreductive therapy, antiplatelet drugs or no therapy at all in this group, which represents a substantial proportion of ET patients ( Barbui et al, 1996 ).
This prospective observational study was designed to estimate the incidence of thrombotic and haemorrhagic complications in a cohort of asymptomatic ET patients aged < 60 years of age who had a platelet count < 1500 × 109/l, and to compare it with that of age- and sex-matched normal controls. Apart from aspirin for erythromelalgia, no treatment was administered until the occurrence of a significant clinical event.
- Top of page
- PATIENTS AND METHODS
Table 1. Table I. Characteristics of patients and controls. * Presence of at least one of the following conditions: hypertension (requiring treatment), smoking, diabetes, hyperlipidaemia, obesity. Table 2. Table II. Incidence of thrombosis in patients and controls. * Cases/100 patient-years.† Incidence rate ratio (CI 95%).
Among cardiovascular risk factors, obesity (body mass index > 30 kg/m2) was the most predictive of thrombosis in the patients in a multivariate Poisson regression model (age- and sex-adjusted incidence ratio for obese subjects: 12.3, 95% CI 1.2–120.8, P = 0.031).
Only three minor haemorrhagic episodes occurred (1.12 cases/100 patients-years), in three patients taking aspirin, with a platelet count of 750, 1000 and 1142 × 109/l respectively, 14, 48 and 55 months from diagnosis.
During the study, nine patients underwent surgical intervention, with a platelet count < 1000 × 109/l, without any complication ( 3 Table III). Three patients became pregnant, with a spontaneous reduction of platelet counts during gestation and delivered three healthy newborns after uneventful pregnancies.
Table 3. Table III. Surgery in ET patients.
Eighteen patients (27.6%) were censored, a median of 34 months from diagnosis (range 7–79) because of the inception of cytoreductive therapy. The reasons for cytoreduction were: thrombosis (n = 5), haemorrhage and platelet levels > 1000 × 109/l (n = 2), thrombocytosis > 1500 × 109/l (n = 5), microvascular symptoms resistant to aspirin (n = 5) and age > 60 years (n = 1).
No leukaemic and myelodysplastic transformations or fatalities were recorded during the study.
- Top of page
- PATIENTS AND METHODS
The present study shows for the first time in a prospective manner that asymptomatic ET patients aged < 60 years and a platelet count <1500 × 109/l had a thrombotic risk comparable with that of a normal control group. Therefore this study suggests that clinical observation alone could be appropriate in these patients.
The incidence of thrombosis observed during the follow-up in our ET patients was low (1.91 cases/100 patient-years), compared with that reported in the literature (9–15 cases 100/patient-years in non-selected series: Bellucci et al, 1986 ; Van de Pette et al, 1986 ; Hehlmann et al, 1988 ; Grossi et al, 1988 ; Fenaux et al, 1990 ; Cortelazzo et al, 1990 ; Colombi et al, 1991 ), probably due to the stringent criteria adopted for selection of this study. On the other hand, the rate of events in the control group (1.50 cases/100 patient-years, Table II), is in keeping with that observed in the normal population. In fact, the rate of non-fatal myocardial infarction plus transient ischaemic attack reported in British male doctors not treated with aspirin was 0.70% subject/year ( Peto et al, 1988 ) and, in a recent study, the incidence of peripheral arterial disease was 1.5% subject/year ( Leng et al, 1996 ). The incidence rate ratio shows that our cohort of ET patients has only a slight increase of risk (1.43 after age and sex adjustment), which is neither statistically nor clinically significant.
For several reasons, the finding that this group of ET patients is at low risk has an important clinical impact.
An asymptomatic clinical picture at diagnosis was present in a large proportion of patients in many published series ( Bellucci et al, 1986 ; van de Pette et al, 1986 ; Hehlmann et al, 1988 ; Grossi et al, 1988 ; Fenaux et al, 1990 ; Cortelazzo et al, 1990 ; Colombi et al, 1991 ), and in about half of these cases the platelet count was < 1000 × 109/l and age was < 60 years. Thus, our study highlights that abstention from treatment is justifiable for approximately 25–30% of all ET patients. Therefore a significant proportion of patients could be spared the side-effects of drugs currently used in ET. Alkylating agents may increase the risk of leukaemia ( Sedlacek et al, 1986 ; Shibata et al, 1994 ). At the moment, the mutagenic potential of hydroxyurea is still a matter of debate after the report of some cases of acute leukaemia in ET patients treated with this non-alkylating agent ( Weinfeld et al, 1994 ; Nand et al, 1996 ; Murphy et al, 1997 ; Sterkers et al, 1998 ). Other treatments, administered with the aim of reducing the platelet count, such as interferon-alpha ( Gisslinger et al, 1989 ) or anagrelide ( Anagrelide Study Group, 1992), may have significant adverse effects and their efficacy in reducing the thrombotic complication in ET has not yet been proved.
The efficacy and safety of use of aspirin in ET has never been assessed by prospective trials and its use is limited to relieve microvascular disturbances ( Michiels et al, 1985 ). Two of the five thrombotic episodes occurred in patients taking aspirin for erythromelalgia, but the number of events is too small to assess the risk of major thrombotic complications associated with the presence of vasomotor symptoms. In five patients, vasomotor symptoms which occurred during follow-up were resistant to antiplatelet therapy, a finding also observed by other authors ( Fenaux et al, 1990 ; Colombi et al, 1991 ; Regev et al, 1997 ) and all three haemorrhagic episodes observed in our study occurred during aspirin therapy. In the general population aspirin reduces the risk of myocardial infarction (a rare event in ET), but slightly increases the risk of haemorrhagic stroke ( Antiplatelet Trialists' Collaboration, 1994). Therefore formal prospective studies will be needed to establish the indication of use of aspirin in asymptomatic ET patients and the optimal effective dose for the long-term prophylaxis of microvascular symptoms.
Our study suggests that conservative management of low-risk ET patients should also be considered during pregnancy and surgery, although the validity of these findings is limited by the small number of events. In pregnant ET patients this suggestion is in keeping with previously reported data ( Beressi et al, 1995 ; Frezzato & Rodeghiero, 1986), although the use of aspirin or interferon has been advocated ( Pardini et al, 1993 ; Vianelli et al, 1994 ; Williams et al, 1994 ; Thornley et al, 1994 ; Pagliaro et al, 1996 ). The effect of surgery as a trigger for thrombosis or haemorrhage is rarely reported in the literature ( Ravich et al, 1970 ); in our study, only minor surgical procedures, all in patients with platelet levels < 1000 × 109/l, were carried out. Therefore the management of pregnancy and surgery in ET patients remains controversial and multicentric trials are needed before definitive conclusions.
We emphasize the importance of the concurrent presence of all the above-mentioned clinical characteristics to define the low-risk profile. In particular, young age is not sufficient protection from thrombotic complications ( Mitus et al, 1990 ; Randi et al, 1990 ; Millard et al, 1990 ; McIntyre et al, 1991 ; Frezzato et al, 1993 ). Previous evidence of thrombosis and/or extreme thrombocytosis are clear indications for active therapy (cytoreduction and/or antiaggregation) in spite of the young age. Moreover, we recommend obtaining a clinical history using validated questionnaires, given that previous thrombotic events represent a strong risk factor of vascular occlusive episodes ( Cortelazzo et al, 1990 ).
Finally, this study provided data regarding the incidence of thrombotic and haemorrhagic complications in untreated ET patients, and valuable information for estimating the future clinical efficacy of prophylactic drugs. Based on our data, about 900 patients should be followed for 10 years in each of two arms of a randomized trial to detect a 50% decrease of thrombotic risk.
Pending the results of these larger studies on active therapy in low-risk thrombocythaemia, our study suggests that abstention from cytoreductive and antiplatelet treatment (apart from the relieve of erythromelalgic circulation disturbancies) could be the best option for these patients.