TY - JOUR
T1 - Perioperative acute upper gastrointestinal haemorrhage in older patients with hip fracture
T2 - Incidence, risk factors and prevention
AU - Fisher, L.
AU - Fisher, A.
AU - Pavli, P.
AU - Davis, M.
PY - 2007/2
Y1 - 2007/2
N2 - Background: No specific preventive strategy exists for acute gastrointestinal haemorrhage in hip fracture patients. Aims: To determine the effectiveness of prophylactic use of proton pump inhibitors in patients with risk factors for acute gastrointestinal haemorrhage. Methods: Prospective two-stage study of 822 consecutive older (≥60 years) hip fracture patients. Results: Acute gastrointestinal haemorrhage occurred in 16 (3.9%) of 407 patients and was associated with increased length of hospital stay (28.7 vs. 15.9; P = 0.0027) and mortality (18.8% vs. 4.3%; P = 0.043). Multiple analysis identified five independent risk factors for acute gastrointestinal haemorrhage: pre-existing peptic ulcer (OR 4.3; P = 0.043), current smoking (OR 3.1; P = 0.023), post-operative use of an antiplatelet agent (OR 6.5; P = 0.046), post-operative use of non-steroidal anti-inflammatory drug/cyclo-oxygenase-2 inhibitor (OR 4.9; P = 0.06) and blood group O (OR 1.7; P = 0.046). These risk factors were highly sensitive and had a negative predictive value of 99.8%. Prophylactic use of proton pump inhibitors in patients with risk factor for acute gastrointestinal haemorrhage significantly reduced the incidence of this complication (0.72% in treated patients vs. 13.4% in untreated; P < 0.001); the number needed to treat was 7.9. Conclusions: In older hip fracture patients perioperative acute gastrointestinal haemorrhage occurs in 3.9% and is associated with poor outcome. Preventive proton pump inhibitor therapy in patients at risk of acute gastrointestinal haemorrhage is effective and safe.
AB - Background: No specific preventive strategy exists for acute gastrointestinal haemorrhage in hip fracture patients. Aims: To determine the effectiveness of prophylactic use of proton pump inhibitors in patients with risk factors for acute gastrointestinal haemorrhage. Methods: Prospective two-stage study of 822 consecutive older (≥60 years) hip fracture patients. Results: Acute gastrointestinal haemorrhage occurred in 16 (3.9%) of 407 patients and was associated with increased length of hospital stay (28.7 vs. 15.9; P = 0.0027) and mortality (18.8% vs. 4.3%; P = 0.043). Multiple analysis identified five independent risk factors for acute gastrointestinal haemorrhage: pre-existing peptic ulcer (OR 4.3; P = 0.043), current smoking (OR 3.1; P = 0.023), post-operative use of an antiplatelet agent (OR 6.5; P = 0.046), post-operative use of non-steroidal anti-inflammatory drug/cyclo-oxygenase-2 inhibitor (OR 4.9; P = 0.06) and blood group O (OR 1.7; P = 0.046). These risk factors were highly sensitive and had a negative predictive value of 99.8%. Prophylactic use of proton pump inhibitors in patients with risk factor for acute gastrointestinal haemorrhage significantly reduced the incidence of this complication (0.72% in treated patients vs. 13.4% in untreated; P < 0.001); the number needed to treat was 7.9. Conclusions: In older hip fracture patients perioperative acute gastrointestinal haemorrhage occurs in 3.9% and is associated with poor outcome. Preventive proton pump inhibitor therapy in patients at risk of acute gastrointestinal haemorrhage is effective and safe.
UR - http://www.scopus.com/inward/record.url?scp=33846311360&partnerID=8YFLogxK
U2 - 10.1111/j.1365-2036.2006.03187.x
DO - 10.1111/j.1365-2036.2006.03187.x
M3 - Article
SN - 0269-2813
VL - 25
SP - 297
EP - 308
JO - Alimentary Pharmacology and Therapeutics
JF - Alimentary Pharmacology and Therapeutics
IS - 3
ER -