Review article
The case against routine preoperative laboratory testing

https://doi.org/10.1016/S0025-7125(02)00147-5Get rights and content

Section snippets

Normal and abnormal test results

It is important for physicians to understand how laboratories define normal and abnormal test results. For many blood test results, a continuous distribution of results is possible. For example, a hemoglobin level may theoretically be measured from 0 through infinity with any value in between. Other tests may have only ordinal results. For example, the degree of proteinuria reported on a dipstick determination may be reported only in a small number of discrete values (1+, 2+, etc.). Finally,

Rationale for preoperative testing

There are several theoretic reasons why clinicians might order routine preoperative tests. These include: (1) to detect unsuspected abnormalities that might influence the risk of perioperative morbidity or mortality; (2) to establish a baseline value for a test that has a high likelihood of being monitored and changing after the surgical procedure is complete; and (3) for medical-legal reasons.

For the first rationale, three actions are possible regarding an abnormal test result. First,

Hemoglobin

Blood loss as a result of surgery is common and, in many types of surgery, perioperative transfusion is necessary. Severe anemia during surgery risks tissue hypoxia from impaired oxygen delivery. Thus, unsuspected severe anemia theoretically may predispose the patient to tissue hypoxia in the perioperative period if not corrected preoperatively. For this reason, a case can be made on the basis of physiology for routine hemoglobin (or hematocrit) measurements for patients in whom significant

White blood cell count

The prevalence of unanticipated elevations of white blood cell counts is very low. In 4 of the 5 available studies, the prevalence was less than 1% (Table 2). In the two studies in which authors assessed management changes related to abnormal white blood cell counts, no patient underwent a management change [11], [12]. In the two studies that allow the calculation of likelihood ratios for abnormal white blood cell counts, the findings suggest that an unanticipated elevated white blood cell

Platelet count

Ten investigations have been published regarding the usefulness of platelet counts (Table 3). The incidence of abnormalities of platelet counts is 0.9%. One study reported a yield of 8%, but all patients in this investigation had slightly elevated counts rather than thrombocytopenia [11]. Management changes on the basis of unanticipated abnormalities in platelet count are rare. In our article, only 0.02% of all measurements of platelet count were abnormal and influenced surgical management. For

Coagulation tests

Many different coagulation tests are available to the clinician. Theoretically, these tests might detect a predisposition to perioperative hemorrhage. In this article, we discuss only the prothrombin time and partial thromboplastin time, as they are the most commonly ordered laboratory tests to screen for coagulation disorders. We have chosen not to discuss the bleeding time as it is infrequently used as a routine screening test before surgery. Furthermore, the studies that have investigated

Electrolytes

A theoretic rationale for measuring serum electrolytes routinely before surgery would be to identify patients at risk for adverse postoperative events including arrhythmia. Previous studies have primarily evaluated abnormalities of serum sodium and potassium. Clinical lore suggests that hypokalemia would be the most clinically important preoperative electrolyte abnormality to detect. The evidence does not, however, suggest a relationship between this laboratory abnormality and postoperative

Renal function tests

Over the past decade, a growing body of literature suggests that preoperative renal insufficiency is one of the most important risk factors for postoperative complications in both cardiac and noncardiac surgery. For example, in the revised cardiac risk index of Lee et al, preoperative serum creatinine >2.0 mg/dL was 1 of 6 predictors of risk for postoperative cardiac complications in a validated multifactorial analysis of patients undergoing noncardiac surgery [22]. The recently revised

Serum glucose

Established diabetes requiring treatment is a risk factor for postoperative cardiac complications. In the recent revised cardiac risk index by Lee et al, diabetes requiring insulin therapy was 1 of 6 independent risk factors for postoperative cardiac complications; the relative risk associated with this factor was 3.0 [22]. Diabetes also increases morbidity and mortality among patients undergoing coronary artery bypass surgery [24], [27]. In addition, diabetes increases the risk of sternal

Hepatic tests

Patients with advanced liver disease and cirrhosis have a marked increase in the risk of postoperative complications and death. The risk increases with increasing severity of liver disease, as classified by the Child-Pugh criteria. For example, among patients with cirrhosis undergoing abdominal surgery, mortality rates for Child-Pugh class A, B, and C are 10%, 31%, and 76%, respectively [30]. But few data exist to suggest that asymptomatic elevations of transaminases or alkaline phosphatase in

Urinalysis

The rationale for routine preoperative testing would be to identify asymptomatic abnormalities that would modify preoperative care or predict an increased risk for postoperative complications. Potential relationships between abnormalities and adverse outcomes include pyuria and wound infection, and glucosuria and hyperglycemic complications. Table 10 summarizes the primary data sources on preoperative urinalysis. The incidence of abnormal urinalyses was 19.1%. Among the subset of studies that

Electrocardiogram

The potential value of a routine preoperative ECG would be to detect abnormalities that would increase the risk of postoperative cardiac complications or to serve as a baseline in the event that a postoperative ECG is required. Baseline findings that may modify risk include the finding of Q waves that confer risk according to the original Goldman cardiac risk index [35]. If one uses, however, the more recent revised risk index, which outperformed the original Goldman index, any clinical

Chest radiograph

Routine preoperative chest radiographs are more likely to be abnormal than are most other preoperative tests. For example, in a study of 3959 patients who received a preoperative chest radiograph, 23% of studies were abnormal [40]. Abnormalities included those of lung parenchyma (13%), heart (7%), pleura (2%), mediastinum (3%), and other abnormalities (2%). In our survey of 15 studies of routine preoperative chest radiographs (Table 12), 21.2% of all such studies were abnormal. Two potential

Outcomes of patients with no routine perioperative testing

Our review establishes the low predictive value of most commonly obtained preoperative laboratory tests. Several recent reports have reached the same conclusion through evaluation of the outcomes of patients subjected to surgery with no routine preoperative testing. Narr et al performed a retrospective review of 1044 patients who underwent elective surgery or diagnostic procedures [45]. The patients were disproportionately young and healthy. The median age was 21 years, and 97% of patients were

Recommendations

We have shown that routine preoperative testing before elective surgery, without regard to patient-related factors that increase the likelihood of abnormal test results, leads to a low incidence of abnormal results—most of which are either ignored by clinicians or are false-positive results that do not predict an increase in the risk of postoperative complications. Several tests, in particular, BUN or creatinine, and hemoglobin measurements, are more likely to identify a subset of patients at

Using protocols and policy to influence testing

Most institutions have established minimal criteria for routine preoperative testing. Thus, facility policies often require the physician caring for the patient to obtain a minimal set of tests before surgery. In many settings, assurance that the tests are complete is the responsibility of support staff for the surgeon or anesthesiologist. If tests are not obtained before surgery, the patient's surgery may be delayed; hence, most physicians are compliant with these policies.

The establishment of

Summary

In this article, we have shown that almost all “routine” laboratory tests before surgery have limited clinical value. Clinicians should order only a small number of routine tests based on age as noted in Table 13. Selective use of other preoperative tests should be based on history and physical examination findings that identify subgroups of patients who are more likely to have abnormal results. In general, clinicians should order tests only if the outcome of an abnormal test will influence

First page preview

First page preview
Click to open first page preview

References (83)

  • A.R. Tait et al.

    Evaluation and efficacy of routine preoperative electrocardiograms

    J Cardioth Vasc Anesth

    (1997)
  • B.J. Narr et al.

    Outcomes of patients with no laboratory assessment before anesthesia and a surgical procedure

    Mayo Clin Proc

    (1997)
  • A. Perez et al.

    Value of routine preoperative tests: a multicentre study in four general hospitals

    Br J Anaesth

    (1995)
  • R.H. Haug et al.

    A prospective evaluation of the value of preoperative laboratory testing for office anesthesia and sedation

    J Oral Maxillofac Surg

    (1999)
  • A.D. Hamdan et al.

    Renal insufficiency and altered postoperative risk in carotid endarterectomy

    J Vasc Surg

    (1999)
  • O. Schein et al.

    The value of routine preoperative testing before cataract surgery

    N Engl J Med

    (2000)
  • M.F. Roizen

    More preoperative assessment by physicians and less by laboratory tests

    N Engl J Med

    (2000)
  • Y. Charpak et al.

    Usefulness of selectively ordered preoperative tests

    Med Care

    (1988)
  • D.S. Macpherson et al.

    Preoperative screening: value of previous tests

    Ann Intern Med

    (1990)
  • R.H. Fletcher et al.

    Clinical epidemiology

    The essentials

    (1996)
  • D.L. Sackett et al.

    Clinical epidemiology

    A basic science for clinical medicine

    (1991)
  • S. Dzankic et al.

    The prevalence and predictive value of abnormal preoperative laboratory tests in elderly surgical patients

    Anesth Analg

    (2001)
  • D.P. Sanders et al.

    Clinical evaluation and cost effectiveness of preoperative laboratory assessment on patients undergoing total hip arthroplasty

    Orthopedics

    (1989)
  • P.M. Faris et al.

    The predictive power of baseline hemoglobin for transfusion risk in surgery patients

    Orthopedics

    (1999)
  • M. Rohrer et al.

    A prospective evaluation of the efficacy of preoperative coagulation testing

    Ann Surg

    (1988)
  • J.M. Turnbull et al.

    The value of preoperative screening investigations in otherwise healthy individuals

    Arch Intern Med

    (1987)
  • P. Peterson et al.

    The preoperative bleeding time test lacks clinical benefit. College of American Pathologists' and American Society of Clinical Pathologists' position article

    Arch Surg

    (1998)
  • R. Rodgers et al.

    A critical appraisal of the bleeding time

    Sem Thromb Hemost

    (1990)
  • A. Suchman et al.

    How well does the activated partial thromboplastin time predict postoperative hemorrhage?

    JAMA

    (1986)
  • A. Borzotta et al.

    Value of the preoperative history as an indicator of hemostatic disorders

    Ann Surg

    (1984)
  • R.J. Gross et al.

    Evaluation of healthy patients and ambulatory surgical patients

  • S. Rappaport

    Preoperative hemostatic evaluation: which tests, if any?

    Blood

    (1983)
  • I.A. Hirsch et al.

    The overstated risk of preoperative hypokalemia

    Anesth Analg

    (1988)
  • E.B. Kaplan et al.

    The usefulness of preoperative laboratory screening

    JAMA

    (1985)
  • T. Lee et al.

    Derivation and prospective validation of a simple index for prediction of cardiac risk of major noncardiac surgery

    Circulation

    (1999)
  • T.L. Higgins et al.

    Stratification of morbidity and mortality outcome by preoperative risk factors in coronary artery bypass patients. A clinical severity score

    JAMA

    (1992)
  • A.P. Furnary et al.

    Continuous intravenous insulin infusion reduces the incidence of deep sternal wound infections in diabetic patients after cardiac surgical procedures

    Ann Thorac Surg

    (1999)
  • J. Gibbs et al.

    Preoperative serum albumin level as a predictor of operative morality and morbidity

    Arch Surg

    (1999)
  • K. Kroenke et al.

    The admission urinalysis: impact on patient care

    J Gen Intern Med

    (1986)
  • L. Goldman et al.

    Multifactorial index of cardiac risk in noncardiac surgical procedures

    N Engl J Med

    (1977)
  • T. Dorman et al.

    Bundle-branch block as a risk factor in noncardiac surgery

    Arch Intern Med

    (2000)
  • Cited by (0)

    View full text