Preoperative Laboratory Testing

Preoperative Laboratory Testing

Routine laboratory testing when patients are fit and asymptomatic is not recommended. Testing should be guided by the history and physical examination. “Routine” testing is expensive and rarely alters peri-operative management; moreover, abnormal values often are overlooked or if recognized may result in unnecessary delays. Nonetheless, despite the lack of evidence of benefit, many physicians order a hema-tocrit or hemoglobin concentration, urinalysis, serum electrolyte measurements, coagulation stud-ies, an electrocardiogram, and a chest radiograph for all patients, perhaps in the misplaced hope of reduc-ing their exposure to litigation.To be valuable, preoperative testing must dis-criminate: there must be an increased periop-erative risk when the results are abnormal (and unknown when the test is not performed), and there must be a reduced risk when the abnormality is not detected (or it has been corrected). This requires that the test have a very low rate of false-positive andfalse-negative results. The utility of a test depends on its sensitivity and specificity. Sensitive tests have a low rate of false-negative results and rarely fail to identify an abnormality when one is present, whereas specific tests have a low rate of false-positive results and rarely identify an abnor-mality when one is not present. The prevalence of a disease or of an abnormal test result varies with the population tested. Testing is therefore most effective when sensitive and specific tests are used in patients in whom the abnormality will be detected frequently enough to justify the expense and inconvenience of the test procedure. Accordingly, laboratory testing should be based on the presence or absence of under-lying diseases and drug therapy as detected by the history and physical examination. The nature of the proposed surgery or procedure should also be taken into consideration. Thus, a baseline hemoglobin or hematocrit measurement is desirable in any patient about to undergo a procedure that may result in extensive blood loss and require transfusion, particu-larly when there is sufficient time to correct anemia preoperatively (eg, with iron supplements).

Testing fertile women for an undiagnosed early pregnancy is controversial and should not be done without the permission of the patient; pregnancy testing involves detection of chorionic gonadotropin in urine or serum. Routine testing for HIV antibody is not indicated. Routine coagulation studies and uri-nalysis are not cost-effective in asymptomatic healthy patients; nevertheless, a preoperative urinalysis is required by state law in at least one U.S. jurisdiction.