it is recommended to be started depending in risk of tromboembolisem and ICH recurrence after date 10-14 days the European Stroke initiative recommends 

Here is some information to consider in response to your question, courtesy of "UpToDate":

●Anticoagulants — Very few studies have studied the risks and benefits of prophylactic anticoagulation after a prior episode of ICH [16,24-26]. A meta-analysis of four studies, two randomized, that compared anticoagulation therapy (unfractionated heparin, low-molecular-weight heparin, heparinoids) with other treatments in patients with acute ICH found that anticoagulation therapy was associated with a significant reduction in PE (1.7 versus 2.9 percent); while rates of DVT and mortality were nonsignificantly decreased, and rates of hematoma expansion were nonsignificantly increased [24]. Because study designs varied in quality and very few PE occurred; these results should be considered cautiously.

The overall benefit versus risks of anticoagulant and/or antiplatelet therapy for prevention of VTE remains uncertain [27]. Extrapolating from patients with acute ischemic stroke, it is likely that anticoagulant therapies reduce the incidence of VTE in patients with an acute ICH. However, the safety of these agents in a setting of acute ICH has yet to be adequately established, and the absolute benefit for VTE prevention is likely to be low if IPC prophylaxis is properly used.

An American Heart Association/American Stroke Association Stroke Council guideline has recommended that after documentation of cessation of bleeding, low-dose LMW or unfractionated heparin may be considered in patients at high risk (mostly bed bound) as early as the first day after ICH onset [28]. In choosing whether to start anticoagulant therapy, we also consider the risk of hematoma expansion (eg, presence of SPOT sign, blood pressure control, presenting hematoma volume).

Specific regimens for anticoagulant therapy for this indication are discussed separately. (See "Prevention of venous thromboembolic disease in acutely ill hospitalized medical adults", section on 'Pharmacologic thromboprophylaxis'.)

Treatment — There are neither good data nor expert consensus about the optimal treatment of venous thromboembolism (VTE) in patients with acute ICH [29-32].

One review estimated the risk of fatal PE in patients with acute ICH to be about 25 percent for those who were untreated for proximal DVT or nonfatal PE [30]. Anticoagulant therapies are highly efficacious for reducing fatal and nonfatal PE in nonstroke patients with proximal DVT and/or PE. (See "Overview of the treatment of lower extremity deep vein thrombosis (DVT)" and "Treatment, prognosis, and follow-up of acute pulmonary embolism in adults".)

As discussed above, DVT and PE are most likely to occur after the period of highest risk of hematoma expansion. While the risks of worsening or recurrence of an ICH with anticoagulation therapy during this relatively subacute period are not fully known it has been estimated to be approximately 3 to 5 percent, ie substantially less than the risk of fatal PE. (See 'Risk of hemorrhage recurrence' below.)

These constructs, while not confirmed by clinical studies, favor the use of anticoagulation in this setting. As a result, in patients with acute ICH and proximal DVT or nonfatal PE and who are >4 days from ICH onset, we recommend treatment with heparin, LMW heparin, or heparinoid, followed by the use of reduced intensity of warfarin anticoagulation (ie, a target INR of 2.0) or low-molecular weight heparin for a period of three months. For ICH patients with DVT confined to below the knee, withholding anticoagulation is reasonable, providing cardiorespiratory reserve is adequate and serial monitoring to detect proximal extension of the thrombosis can be carried out. (See "Overview of the treatment of lower extremity deep vein thrombosis (DVT)" and "Treatment, prognosis, and follow-up of acute pulmonary embolism in adults".)

While the use of inferior vena cava (IVC) filters in this situation has been recommended by some authorities, including the 2007 American Heart Association guidelines [21,33], there are no direct comparisons between these two options (ie, anticoagulation versus IVC filter). The higher rate of long-term complications and lesser efficacy of IVC filters may therefore support the use of anticoagulants for patients who are more than one week from ICH onset or when IVC placement is not feasible [30,32].

PRIOR OR RECENT INTRACEREBRAL HEMORRHAGE — Use of long-term antiplatelet or anticoagulant drugs in ICH survivors involves a challenging risk/benefit decision that considers the risk of late hemorrhage expansion and hemorrhage recurrence as well as the risk of thromboembolism [34-37].

Only those ICH survivors with an exceptionally high risk for atherothrombotic events should be treated with antiplatelet drugs, and only those with very high risk for cardioembolic stroke (ie, prosthetic valves, dilated cardiomyopathy, documented intracardiac thrombus) should be treated with anticoagulation [21].

Regardless of whether or not the patient is treated with antithrombotic agents, control of hypertension substantially reduces the risk of recurrent ICH.

Risk of hemorrhage recurrence — ICH recurs in survivors at a rate of 2 to 3 percent per year; this represents an approximate 10-fold increase in relative risk and a 2 percent per year absolute risk increase over the general population [38-41].

During the first three months following an acute ICH, the reported rate of recurrent ICH, in the absence of anticoagulation, is about 1 percent (range 0.4 to 3.0 percent) [8,39,42]. It is uncertain how many of these "recurrences" are actually extensions of initial bleeding versus bleeding from a separate site.

Factors associated with a higher risk of recurrence include the following:

●Uncontrolled hypertension is an important modifiable risk factor for ICH recurrence. (See "Spontaneous intracerebral hemorrhage: Treatment and prognosis", section on 'Recurrence'.)

●Lobar ICH, involving the cerebral cortex, is often related to cerebral amyloid angiopathy (CAA) in the elderly, and appears to carry a higher risk of recurrence (21 percent in two years, according to one cohort study) than deep hemispheric hemorrhage [38,39]. (See "Cerebral amyloid angiopathy", section on 'Prognosis'.)

●Anticoagulant therapies may pose a risk for recurrent ICH. Patients taking anticoagulant therapies who have a recurrent ICH are also likely to have larger hematomas and worse outcomes than patients not taking these agents. (See 'Risk associated with antithrombotic therapy' below.)

●More variably associated with recurrence risk are older age, male gender, microbleeds or leukoaraiosis on MRI. (See "Spontaneous intracerebral hemorrhage: Treatment and prognosis", section on 'Recurrence'.)

While MRI gradient-echo techniques to detect clinically silent microbleeds may stratify patients according to their risk of recurrent ICH, the sensitivity and specificity of these techniques have not been adequately defined to apply to individual patients at the present time.

Both the absolute risk of ICH recurrence and the presence of risk factors are considered when weighing decisions to resume or initiate antithrombotic therapy in patients who have had an ICH. As an example, anticoagulation is avoided in patients with lobar hemorrhage, particular if there are associated microhemorrhages consistent with CAA, unless the indication for antithrombotic therapy is compelling [43]. (See "Cerebral amyloid angiopathy", section on 'Avoiding anticoagulants and antiplatelet agents'.)

Some excellent points raised here. Risk vs benefit decision-making often rests on indication for anticoagulation (AF/VTE), presence of CAA and perhaps most importantly the level of hypertensive control. Retrospective studies have not associated restarting as per guidelines with worsening HT or neurological status. We need better prospective data ultimately.

Hi Genc and Bruce,

Further to my comment above on pragmatism when posed with such a position of equipose. There was a reference I wanted to share that helped me recently in a very similar situation. Timing perhaps is a method of improving safety:

Article Optimal Timing of Anticoagulant Treatment After Intracerebra...

Acute ICH patients were included but there is crossover with HT populations.