Preoperative Radiologic Studies

The preoperative clinician should target ordering of preoperative radiology studies to specific issues raised by the patient’s history and physical examination. For example, concern over the status of current heart failure or active pulmonary infection may prompt the preoperative clinician to order chest radiographs. In addition, radiologic studies may be indicated to define cervical spine or tracheal anatomy of concern so that safe airway management can be provided. In these instances the ordering preoperative clinician needs to ensure that accountability for review of the results of these studies exists in the perioperative workflow.

There needs to be clear definition between radiologic studies ordered by the surgeon to define indications for the operation and studies ordered by the preoperative clinician for the purpose of preoperative assessment and management. For example, surgeons may order chest radiographs as part of a general screening in patients undergoing procedures for cancer diagnosis. The ordering physician is responsible for reading and acting on the results of the test. If systems to ensure accountability are not adequate, patients may have abnormal chest radiograph results present in the system that have not been reviewed and acted on by the ordering clinician. Special attention needs to be paid when there are short intervals between surgical evaluation and procedure date, in which all test results may not have been adequately reviewed. It is prudent for institutions to develop standards to clearly delineate accountability for preoperative test review; for example, at our institution it is reinforced with a documented policy that the clinician who orders the study is responsible for any result. These measures should be taken to avoid the unfortunate circumstance in which, for example, a nodule is present on a preoperative chest radiograph that was ordered but not reviewed, and the patient returns later with a cancer diagnosis.

The lack of value of screening radiographs has been documented in a number of studies. In existing pulmonary conditions such as chronic obstructive pulmonary disease (COPD), it is unlikely the expected abnormalities revealed on a preoperative chest radiograph will affect perioperative management. In a literature review of articles published between 1966 and 2004, an association between preoperative screening with chest radiographs and a decrease in perioperative morbidity and mortality could not be established. Up to 65% of the changes seen were associated with chronic disorders and had little impact on management. Postoperative pulmonary complications did not differ between patients who had preoperative screening chest radiographs and those who did not. These authors concluded that, although the prevalence of chest radiograph abnormalities increases with age and risk factors, most abnormalities found were chronic and were not shown to affect anesthetic management or perioperative outcome. Chest films ordered because of concern about the possibility of acute heart failure or acute pneumonia were the only possible exceptions, which led to the authors’ recommendation that asymptomatic patients do not warrant screening chest radiographs, regardless of age.

In contrast, the American College of Physicians considers that chest radiographs may be helpful in patients older than 50 years who are undergoing abdominal aortic aneurysm (AAA), upper abdominal, or thoracic surgery. The American Heart Association suggests that patients with severe obesity (body mass index > 40 kg/m ² ) also have chest radiographs performed preoperatively. The thought in these cases is that screening radiographs may reveal undiagnosed heart failure or abnormalities suggestive of significant pulmonary hypertension. However, there are no studies supporting the fact that these recommendations have been correlated with a change in perioperative outcomes. It is our recommendation based on this review that the preoperative anesthesiologist only order chest radiographs when suspicion of an acute process exists. The surgeon may decide to order a preoperative chest radiograph for other reasons, including as part of an overall screening for metastatic disease, but should be responsible for reviewing and acting on the results.

The Canadian Anesthesiologist Society guidelines recommending that preoperative chest radiographs not be done in asymptomatic patients is supported by a systematic review noting that most abnormalities found are chronic and the majority are cardiomegaly and COPD. Abnormalities, with the possible exception of acute heart failure, were not found to affect anesthetic or surgical management or perioperative outcome. The Task Force of the American Society of Anesthesiology has reviewed the evidence on preoperative chest radiographs. This group states that although chest radiograph abnormalities may be more frequent in patients who are older, have stable COPD, have stable cardiac disease, smoke, or have resolved recent upper respiratory infections, there is no evidence that chest radiograph results in these patients will affect outcome or management.

Preoperative Pulmonary Function Testing

In specific cases, the anesthesiologist might find the results of spirometry helpful for discussing the complete risk–benefit of surgery with the patient, planning perioperative management, and anticipating potential pulmonary complications. For example, in severe scoliosis, studies have shown that poor preoperative pulmonary function test (PFT) results were correlated with a high incidence of postoperative pulmonary complications.

Similarly, patients with degenerative neurologic diseases with a restrictive pulmonary component may also benefit from preoperative PFTs. For example, in patients with multiple sclerosis severe enough to result in an inability to ambulate, PFT results may help to assess the ability of the patient to wean successfully from the ventilator postoperatively. In patients with myasthenia gravis, PFTs are part of the algorithm used to predict the probability of extended postoperative ventilation. In one study, the results of preoperative values for forced vital capacity (FVC), forced expiratory flow (FEF) _25-75% , and midexpiratory flow (MEF) _50% , along with patient gender, successfully predicted the actual ventilatory outcomes in 88.2% of patients.

For some specific surgeries, preoperative spirometry can help predict long-term mortality. For example, patients with AAAs frequently are smokers with COPD. Lower FEV ₁ and lower FVC values preoperatively were independently associated with an increased risk of long-term mortality after endovascular AAA repair. This suggests that evaluation of lung function should be considered in patients scheduled for AAA repair suspected of having significant COPD.

Preoperative Urine Analyses and Culture

Routine urinalysis is not generally recommended for most surgical procedures and is not necessary for preanesthesia assessments in asymptomatic patients. The concern is that in cases with urinary tract infections there is a risk of bacteremia. Therefore a relationship may exist between undiagnosed and untreated urinary tract infection and postsurgical infections, particularly in surgery in which a prosthesis is placed. However, the literature on this point is controversial. In addition, although this is a relatively inexpensive test, it is done in such high volumes that the aggregate costs may outweigh the clinical benefits. For example, in a study published in 1989, given the best estimate of increase in risk of wound infection related to the presence of urinary tract infection, the cost was $1.5 million per wound infection prevented. The ASA Task Force concluded that preanesthesia urinalysis is not recommended, except for specific procedures such as prosthesis implantation and urologic procedures or when urinary tract symptoms exist.

Preoperative Coagulation Studies

Review of the current literature suggests that preoperative routine screening coagulation studies should not be performed because of the lack of significant impact on preoperative management and outcome. If a good preoperative history is taken, unexpected coagulation defects are extremely infrequent. If the patient has a low risk of bleeding by history and physical examination, it is very unlikely that excessive surgical bleeding will result from an inherent abnormality. A systematic review of the literature from 1966 to 2005 was done in an attempt to provide a rational approach to the use of bleeding history and coagulation tests before procedures and summarized some key recommendations. Firstly, indiscriminate coagulation screening before procedures to predict the risk of bleeding in unselected patients is not recommended. Secondly, a bleeding history that includes family history of coagulation issues, history of excessive bleeding with previous procedures, and current use of prescription antithrombotic or antiplatelet agents should be taken in all patients before invasive procedures. In addition, clinical conditions that predispose patients to bleeding (e.g., significant liver disease) should be noted. If the patient’s history is negative for these factors, no further coagulation testing is needed. If this history is positive, coagulation testing should be targeted for the type of clinical features present. A recent study focused on a comparison of an assessment of patient history versus preoperative hemostasis screening in adult neurosurgical patients supports these recommendations. The study found that patient history was as predictive as laboratory testing for all outcomes and had higher sensitivity. In addition, these authors estimate that hemostatic screening limited to neurosurgical patients with a positive history would save an estimated $81 million annually in the United States, on the basis of approximately 2.1 million neurosurgical procedures performed.

Of note, the anesthesiologist needs to be aware that the anticoagulant effect of some agents, such as enoxaparin, is not adequately assessed by routine coagulation studies. In addition, patients may be taking nonprescription substances not regulated by the Food and Drug Administration (FDA) that could potentially have an impact on coagulation, although no definitive data exist on the effects of these nonprescription agents.

It must be recognized that unnecessary ordering of coagulation studies will also result in wasting resources dealing with insignificant abnormal results. For example, one of the most commonly seen abnormalities after routine screening is an elevated prothrombin time (PT) or partial thromboplastin time (PTT). Appropriate interpretation of this test requires knowledge that the in vitro result may not reflect the in vivo response, as outlined in a systematic review. For example, normal biologic variation, with definition of the normal range as above two standard deviations from the mean, means that 2.5% of healthy patients will have an abnormal result. Unnecessary further investigation may result in excess cost and potential delay of the procedure. In addition, some clinically important bleeding disorders, such as von Willebrand’s disease, will be missed if the presence of normal routine coagulation studies is assumed to ensure appropriate hemostasis.

The volume and age of the blood sample tested has a major impact on the reliability of results. An inadequate sample size, prolonged storage, or excessively traumatic venipuncture will result in an inaccurate result. Finally, the presence of certain conditions, such as the presence of a lupus anticoagulant, will falsely prolong the results and is not indicative of excessive bleeding.

In view of the aforementioned issues, when an abnormal coagulation result is obtained, the study should be repeated and the sample analyzed before any additional workup is undertaken. In many cases no abnormality is identified on repeated testing. A study of 1603 prospective routine screening tests in preoperative tonsillectomy patients demonstrated 35 abnormal test results; of these, only 15 remained abnormal on retesting. A total of 11 patients in this study were shown to have inhibitors, one had mild hemophilia A, and several had no determined etiology. No relationship with the predictability of postoperative bleeding was demonstrated. These authors note that the large number of false-positive results and the absence of an impact on surgical bleeding raise doubts about the value of routine preprocedure coagulation testing.

There are no studies supporting the use of preoperative coagulation testing before the use of regional anesthesia, and the Preanesthesia Task Force did not have a recommendation on this issue.

Preoperative Hematocrit and Complete Blood Count

The evidence would suggest that a targeted history and physical examination should determine whether a preprocedure hematocrit level and/or complete blood count should be done. (See Chapter 23 for a complete discussion on preoperative hemoglobin.) Laboratory tests not targeted by a history and physical examination rarely affect care or outcome and can unnecessarily increase costs. For example, a study of 142 general surgery patients showed that if laboratory tests, including hematocrit, had been ordered only as dictated by patient history and physical examination, patient charges could have been reduced by more than $400,000 in one year. Anemia has been shown to be present in about 1% of asymptomatic patients, but surgically significant anemia in unselected patients is rare. However, there are data in male veterans correlating 30-day postoperative mortality rates after major noncardiac surgery with abnormal preoperative hematocrit levels. Nonetheless, it is unclear whether it actually is the comorbidity or the low hematocrit level that contributes to the increase in mortality. The Anesthesia Task Force concluded that routine hematocrit testing is not warranted and that characteristics such as type and invasiveness of procedure, extremes of age, and history of liver disease, anemia, bleeding, and other hematologic disorders be considered in determining the need for this testing.

In view of the evidence just mentioned, individual institutions have generally established protocols regarding indications for preoperative hematocrit testing. These may be based on age, as well as on the invasiveness of surgery and potential for blood loss. In considering the very low possibility of revealing significant white blood cell and platelet abnormalities on routine screening with complete blood counts, these are generally not included as part of these protocols.

Preoperative Serum Chemistry and Glucose

Preoperative blood testing for serum chemistry values should be specifically targeted to clinical characteristics. Significant electrolyte abnormalities noted on routine screening are extremely rare. The Anesthesia Task Force notes that the presence of endocrine abnormalities, extremes of age, renal dysfunction, liver dysfunction, and the use of certain medications or therapies should be considered when making the decision to order analysis of serum chemistry.

It is important to note, however, that renal insufficiency (creatinine > 2.0 mg/dL) is one of the independent risk factors that was correlated with an increased risk of postoperative cardiac complications. The current American College of Cardiology/American Heart Association (ACC/AHA) algorithm defines this as one of the clinical risk factors that should be used in determining the need for further cardiac evaluation in patients with low functional status undergoing moderate- to high-risk procedures. Because the incidence of renal dysfunction increases with age, some institutional protocols may include age requirements for renal function testing in patients having more invasive procedures, particularly if additional cardiac risk factors are present.

Similarly, the literature indicates that insulin-dependent diabetes is an independent risk factor for postoperative cardiac complications in patients with low functional status undergoing moderate- to high-risk surgery. Non–insulin-dependent diabetes has not been correlated with increases in postoperative cardiac complications. Previous work has suggested that there is no correlation between routine screening blood glucose levels and significant changes in perioperative management or outcome. The degree of long-term glucose control in known diabetic patients is likely better determined by obtaining results of hemoglobin A _1c testing rather than random glucose testing. Better control of perioperative glucose management in known diabetic patients has been correlated with fewer wound infections and less mortality after cardiac bypass surgery. Therefore preoperative testing using hemoglobin A _1c and fasting glucose measurements may be of help in planning appropriate insulin management in these patients.

More recent work indicates that increased preoperative prediabetes glucose levels in patients having noncardiac, nonvascular surgery were associated with a 1.7-fold increased cardiovascular mortality risk compared with normoglycemic preoperative glucose levels. These authors noted that prediabetes glucose levels in patients without a history of diabetes were associated with increased risk of cardiovascular complications even after adjustment for a broad range of comorbidities. They suggest that screening for glucose abnormalities in surgical patients should be considered to identify patients at risk for postoperative cardiovascular events. However, no data exist on whether appropriate treatment of these patients when identified preoperatively would have prevented these complications or whether there is benefit to delaying elective surgery to achieve better preoperative glucose control. A medical record study of about 3000 patients undergoing noncardiac surgery showed that patients without a known history of diabetes who had perioperative hyperglycemia experienced worse outcomes and higher mortality at a glucose level similar to that of those with known diabetes. These authors suggest that perhaps there is a lack of adaptation to hyperglycemia, and they recommend presurgical screening and the need to address glycemic control in these patients.

Urine Toxicology Screen

The significant prevalence of substance abuse in the general population and the potential dangerous interactions with perioperative medications prompt consideration of screening for at-risk patients. Cocaine use, which has particularly concerning implications for anesthesia, can be found in all sociodemographic groups. A careful history, paying special attention to habits regarding illicit substance use, should be taken by the clinician to guide the need for preoperative screening.

Screening tests for illicit drug use generally involve urine testing. Urine testing for toxic substances is simple to perform, can yield rapid results, and provides information about many of the drugs of concern during the perioperative period. Of note, depending on the amount and type of drug taken, a preoperative urine test may be positive for several days after use of a substance. Anesthesiologists should be familiar with the particular type of urine drug testing done at their institutions and which drugs are screened for with a routine test. Those most commonly screened for include opioids, alcohol, cocaine, phencyclidine, and amphetamines. If suspicion of illicit drug use exists, the clinician should consider the timing of the preoperative assessment relative to the surgery to decide whether testing is warranted during the preoperative visit or on the day of the procedure. A positive urine toxicology screen is an indication of drug use within the past few days but will not indicate if drug use is short- or long-term. These patients may be unreliable historians. Therefore preoperative urine screening may be required immediately before the procedure so that the absence of an interaction of these agents with perioperative medications is ensured.