Anticoagulant and antiplatelet treatments have well defined indactions, with a clearly proved benefit, respectivly for prevention of arterial and venous emblism and for prevention of athermo-related arterial thrombosis.

Bleeding risk represents the main adverse effect of these antithrombotic medications. Then benefit-risk ratio is sometimes difficult to evaluate, especially for elderly patients prone to fall (incidence of falls estimated to 30% per year for patients over 65), exposed on the one hand to thromboembolic risk and on the other hand to bleeding risk.

Associations between falls and antithrombotic-related bleeding risk had already been evaluated in several studies :

• Concerning anticoagulant treatments in patients at high risk of falls, retrospective studies shown a overrated risk of intracranial hemorrhage and mortality, but those results remain discordant wtih 3 major prospective studies on larger populations.
• Concerning antiplatelet treatments in patients at high risk of falls, majority of retrospective studies reported an overrated risk of major bleeding, intracranial bleeding and mortality, but datas remain fewer than for anticoagulant and results are as well discordant with prospective studies.
• No difference of morbi-mortality is clearly estalblished depending of antithrombotic treatment class (anticoagulant versus antiplatelet), however there is a cumulative risk in case of association of both anticoagulant and antiplatelet.
• Main factors associated with fall-related bleeding for patients on anticoagulant include age, female sex, anemia, chronic kidney disease, dementia and polymedication.

Thus, the purpose of this study is to specify whether occurrence of falls justify to reconsider prescription of antithrombotic treatments in patients having an indication of antiplatelet or anticoagulant therapy.

Retrospective, descriptive and non-interventional study over a period of 12 months (from 01/01/2017 to 31/12/2017), from unwanted events database of Rennes University Hospital.

Patients meeting criterias receive an information mail. Without opposition over a period of 3 weeks, datas will be treated, focusing on the event and its consequencies until hospital outing.

Baseline characteristics :

• Demographic datas : age, sex
• Antithrombotic treatement, treatment class and molecule
• Last INR dosage before the fall for patients of vitamin K antagonists
• Antithrombotic treatment indication : supra-ventricular arythmia, heart valve mecanical prothesis, severe left ventricle hypokinesia, venous thromboembolism, anti-phospholipid syndrome, myeloproliferative disorder, atheroma (in primary or secondary prevention)
• Comorbidities : history of bleeding (including hemorrhagic stroke), haemostasis disorder, thrombocytopenia, anemia, chronic kidney disease, chronic liver disease, alcohol abuse, ischemic stroke or transient ischemic stroke, cognitive disoreder, active neoplasm, HAS-BLED score, HEMORRHAGE score
• Other undergoing treatments : total number of medications, antihypertensive treatment, psychotropic treatment, non-steroid anti-inflamatory treatment

Description of fall-related bleeding events rate (until hospital outing) according to the severity :

• Any severity bleedings
• Major bleedings (hemorrhagic shock, deglobulisation, red blood cells transfusion required, vital or functional-threatening bleeding spot, uncontrolled hemorrhage requiring specialized haemostatic intervention)
• Intracranial bleeding
• Fatal bleeding

Modification of the antithrombotic prescription in the aftermath of the fall :

- Discontinuation or suspension of antithrombotic treatment

Sub-group analysis for each previous item, according to undergoing antithrombotic treatment at the moment of the fall :

• Curative-dose anticoagulant in monotherapy
• Preventive-dose anticoagulant in monotherapy
• Antiplatelet in monotherapy
• Curative-dose anticoagulant and antiplatelet in association
• Preventive-dose anticoagulant and antiplatelet in association
• No antithrombotic treatement

Inclusion Criteria:

• Hospitalized patients in internal medicine unit in Rennes University Hospital, from 1/01/2017 to 31/12/2017
• Age over 18
• In-hospital fall reported in unwanted event database
• Separated inclusion for each fall in the same patient

Exclusion Criteria:

• Adults on legal protection (safeguarding justice, trusteeship, guardianship) or deprived of liberty
• Patient opposed to inculsion

Hospitalized patients in internal medicine unit for whom an in-hospital fall was reported