Background: Routine preoperative coagulation tests continue to be ordered despite the lack of evidence supporting their use or ability to predict bleeding complications. This may delay surgery, cause unnecessary concern if “abnormal”, and increase cost. It is also unclear whether prophylactic fresh frozen plasma (FFP) decreases bleeding risk. The hospitalists at our university preoperative assessment clinic (UPAC) developed guidelines in conjunction with our anesthesiologists in an attempt to minimize inappropriate testing, although several surgeons remain adamant about ordering them.
Methods: We retrospectively conducted a search of our EMR (EPIC) for a list of all tests for prothrombin/international normalized ratio and activated partial thromboplastin times (PT/INR/PTT) ordered in UPAC from January 1, 2013-March 31, 2014 to determine the rate of abnormal values as well as what action, if any, was taken, and whether abnormalities predicted postoperative bleeding. Our UPAC indications for these tests include: 1)clinical evidence or history of bleeding disorder (easy bruising, nose bleeds, bleeding gums from dental procedures); 2)family history of bleeding disorder; 3)history or presence of liver disease or alcoholism; 4)history of malabsorption/malnutrition; 5)anticoagulant use or drugs affecting coagulation; 6)neurosurgery and ortho spine patients (craniotomy; spine surgery).
Results: (See TABLE 1) 56/850 (6.6%) PT/INRs and 51/778 (6.6%) PTTs were abnormal, but only 12.5% and 27.5% of these abnormalities respectively were not predicted based on risk factors Management was affected in only 1/850 (0.1%) patients with an INR and 2/778 (0.26%) patients with a PTT performed. No patient with an abnormal test experienced major or clinically significant bleeding perioperatively.
Discussion:The low incidence of abnormalities and absence of clinical bleeding support the recommendation for restricted testing. There is also no reason to obtain these tests in patients on anticoagulants who are seen at least 5 days before surgery who will be stopping the drugs preoperatively. Only testing patients 1 day before or on the morning of surgery would be likely to affect management.
Conclusions: Despite narrowing the indications for preoperative coagulation testing, only 6.6% of these patients had an abnormal INR or PTT, resulting in management change in only 3 patients and no significant perioperative bleeding. Our site specific data can hopefully be used to eliminate coagulation testing except for patients with a bleeding history or those on anticoagulants undergoing surgery in the next 24 hours.