During the 1970’s and even later a shotgun approach to pre-anesthetic assessment was routine. Reasons for this attitude included the argument that broad based testing was good screening for patients who might not otherwise seek medical advice on a regular basis. Indeed, it was suggested that the information thus gained might well negate the need for an annual physical examination. Moreover, obtaining a battery of information seemed to be medico- legally necessary should anything untoward happen later. Many hospital policies mandated such testing and, as insurance companies usually paid in full, additional revenues could be collected at little cost by health care facilities. But perhaps the most compelling reason given for shotgun evaluation was simply that it was felt to be “best for the patient and doctors.”
Broad spectrum pre-anesthetic testing was not considered necessary in the early days of anesthetic administration. John Snow in his treatise on chloroform administration, writing under the heading of “Preparations for the inhaling of chloroform,” noted that “the only direction which is usually requisite to give beforehand to the patient who is to inhale chloroform, is to avoid taking a meal previous to the inhalation.” However, two pages further on in the book, he remarked on the importance of a physical examination… "On feeling the pulse of a gentleman, about 21 years of age in March 1855, who had just seated himself in the chair to take chloroform, I found it to be small, weak and intermitting and it became more feeble as I was feeling it. I told the patient that he would feel no pain, and that he had nothing whatever to apprehend. His pulse immediately improved. He inhaled the chloroform, woke up, and recovered without any feeling of depression. Now, if the inhalation had been commenced without inquiry or explanation, the syncope which seemed approaching would probably have taken place and it would have had the appearance of being caused by the chloroform, although not so in reality. In effect, Dr. Snow had realized and emphasized the importance of the physical examination.
In those early days, anesthesia and surgical procedures were frequently carried out in the home – a bedroom or kitchen was converted to operating and recovery areas. Only after the beginning of the last century did it become routine to admit patients to hospitals. A prominent authority on hospital management emphasized that only 13% of the sick were admitted to hospitals as there was no need or provision of beds for every sick person. Rather, the wage earner should not be obliged to leave work and become an inmate of a hospital ward.
Between the World Wars, hospital stays became more fashionable and much longer, perhaps because of the prolonged recovery and repeated surgical and reconstructive needs of so many war casualties. Municipal hospitals in cities such as New York often had 1-2 thousand beds and even more. Hospital costs were escalating. As early as 1949, Green and Howalt suggested that a preadmission evaluation clinic was feasible. Some 25 years later, the ability of such an arrangement to decrease costs was demonstrated. This author (EAMF) developed an outpatient assessment clinic that processed 3,500 patients over 3 years between 1972 and 1975 and showed significant savings in hospital days (up to 50% decrease in length of stay).6 At that time abnormalities that necessitated rescheduling of cases were demonstrated in about 5% of patients. The majority of situations uncovered could be revealed by history and physical examination (e.g. pregnancy, colds, uncontrolled diabetes). But the clinic was not generally enthusiastically received. Insurance carriers in 1975 paid $205 per in patient day and only $41 for outpatient evaluation and nothing for physician assessment. Hospital administrators preferred to fill beds with patients who were having “light” days (e.g. blood tests, chest X-rays, EKGs) rather than “full service” days (e.g. operating room, recovery area, intensive care nursing etc).
The freestanding ambulatory surgery movement was initiated in the United States in the 1970’s. As anesthesia became safer because of improved and increased monitoring, and surgical procedures could be more easily performed with new (and expensive) equipment (e.g. microscopes, lasers, laparoscopes etc), ambulatory practice grew such that now 70% of all surgery performed in the United States is on an outpatient basis. Although at first outpatient surgery was restricted to young, healthy individuals, age is no longer a deterrent to the practice.
If one considers that health care costs cannot simply continue to increase as new drugs and technology are added, then cuts must be made. Questions are raised. “What are the priorities?” “What can we do without?” “What makes a difference?” “What are the risk/benefit ratios?”
One of the first large studies to challenge the usefulness of preoperative laboratory screening was published as long ago as 1985. The authors assessed the usefulness of routine laboratory screening of 2,000 preoperative patients over a four month period. The tests ordered included complete and differential blood cell counts, prothrombin time and partial prothrombin time, platelet counts, six- factor automated multiple analysis, and glucose level. Sixty percent of these routinely ordered tests would not have been performed if testing had only been done for recognizable indications, and only 0.22% of these revealed abnormalities that might influence perioperative management. Chart review indicated that these few abnormalities were not acted on nor did they have any adverse anesthetic or surgical consequences. The conclusions drawn 22 years ago were that in the absence of specific indications, routine preoperative laboratory tests contribute little to patient care, and in most instances can reasonably be eliminated. Even when an irregular result is revealed, therapy is rarely changed.
In a review article published a few years later, Roizen, who was one of the authors of the study cited above, took the conclusions a little further. Pointing out that more than $40 billion a year was spent in the United States in the late 1980’s on preoperative testing and evaluation, he noted that 60% is wasted. Likening it to a rather absurd statement that, “If a little epinephrine is good, more is better,” he noted that extra testing causes iatrogenic disease by pursuit and treatment of borderline and false positive results, thereby increasing medicolegal risks and decreasing the efficiency of practice. He wrote that anesthesiologists (and also other physicians by implication), could turn such inefficiency to advantage by showing the patient and the bureaucrat (the watchdog), that we can use inexpensive technology to reduce costs substantially and improve the quality of care. Unfortunately, at this point, hospitals were still reimbursed handsomely for unnecessary testing. Laboratory departments fiercely clung to their territory. Further studies were undertaken.
Attempts were made to design studies that would accurately identify essential preoperative tests. One French report considered the basic concepts concerning classification of studies evaluating diagnostic procedures and the specific problem of assessment of routine preoperative tests.9 The study included 3,866 consecutive patients and the protocol led to the performance of only 33% of all predicted tests. There were no adverse effects on the quality of care. These conclusions agreed with the findings of the Chicago group that 60% of tests routinely performed preoperatively are useless. Another general investigation from the Letterman Army Medical Center in San Francisco published in 1991, evaluated the ability of preoperative laboratory testing to predict postoperative complications.10 Data collected from 520 patients undergoing elective surgery included American Society of Anesthesiologists (ASA) classification, ponderal index, electrolyte values (including glucose), blood/urea nitrogen/creatinine values, complete blood counts, coagulation studies, total protein/albumin/lymphocyte count, EKG, chest X-ray, urinalysis, pulmonary function tests, type of anesthesia and operation. The authors found that postoperative complications were strongly associated with the ASA classification, type of anesthesia (surgical complexity) and procedure. The only laboratory tests that might indicate postoperative complications were EKG, chest X-ray and nutritional status. The conclusions drawn were that preoperative laboratory testing should only be undertaken for specific indications and might be determined by age. In other words, more abnormalities are found in sicker patients.
Adams and Weigelt studied 169 adults scheduled for elective hernia repair. Again they considered the usefulness of performing routine tests and also divided the patients to 2 groups. Group 1 had no disease except the hernia and group 2 had evidence of another disease process. Abnormal results not predicted by history were found in 1% of group 1 patients and in 1.4% of group 2 patients. In only 2% of patients was treatment altered by the findings. The authors felt that routine preoperative testing in this patient population was of little value. In a similar study of preoperative testing of patients prior to elective surgery, Macpherson found that the frequency of unanticipated abnormalities was too low to justify the practice pattern of routine testing. Little or no evidence pointed to an association between test abnormalities and perioperative morbidity.
An interesting study compared the amount of tests ordered and length of stay when either a surgeon or an anesthesiologist was responsible for preoperative assessment. Over 3,000 patients were reviewed in each group. The percentage of patients subjected to laboratory tests, EKG or chest X-ray decreased from 90%, 55%, and 50% respectively when the surgeon was in charge to 53%, 43% and 10% when the patient was under the care of the anesthesiologist. Delay of admission until the day of surgery was more likely in the group handled by the anesthesiologist and length of hospital stay was also significantly reduced. No doubt the surgeon, who is the first line link to the patient, is influenced by the latter to some extent. The authors suggest that tests should only be ordered as indicated. Education of patients is also essential. In a similar study, surgeons were persuaded to delegate the ordering of tests to nurses and anesthesiologists. Reductions of 50%-60% in the first and second years respectively in the overall number of tests ordered per patient were demonstrated. The improvement in the appropriateness of tests indicated 81% and 86% over the study period. The cost savings approximated $75,000 per year in what was a small study size. Yet another study proved that the ordering of preoperative tests should better be managed by anesthesiologists. These authors compared costs, number of tests, rate of cancellation and outcome between a group of patients for whom tests were ordered primarily by surgeons and a second group which was mainly controlled by anesthesiologists while still allowing surgeons to order tests that they deemed necessary. In the group managed primarily by anesthesiologists, far fewer tests were ordered; there was no increase in cancellation rate and no adverse outcome. The average savings per patient was $21(in 1997 dollars). On average, 72.5% of tests ordered by surgeons were considered not indicated by anesthesiologists. This decrease represented an annual cost savings of at least $80,000 at that hospital.
The efficacy of routine testing has also been examined in the office setting. In 458 patients who received general anesthesia for dental surgery, about 3% had aberrant laboratory values (low Hct, hematuria, elevated white cell counts). In only two patients was surgery deferred (one was found to be pregnant and the other, a known diabetic, had glycosuria). Thus, in the office setting, history and physical examination and reassessment of key portions of the history were the major factors in the development of the anesthetic treatment plan. Laboratory data had little or no effect on the decision making process.
Several studies have examined the financial impact of reducing routine testing. Wattsman and Davies looked at the utility of routine preoperative testing over one year.18 They tried to determine whether tests indicated or not indicated by patient history or physical examination would identify abnormalities that might influence perioperative care or predict postoperative outcome. In only 4 cases out of 300 did an abnormal test which was in the indicated group result in cancellation. Patient charges were over $15,000 of which $8,500 was for non-indicated tests. The authors concluded that if tests were ordered as dictated by patient medical and physical examination only, patient charges could be reduced by $400,000 per year (1997 dollars). There were no adverse outcomes.
Similar studies pointing to the uselessness of routine testing have been made in several other countries. A multicenter survey of 60 hospitals in Italy examined the practices of surgeons and anesthesiologists and showed that prescription of preoperative investigations was driven more by personal experience than by evidence based scientific knowledge.19 The system led to ineffective and inefficient clinical practice. Healthy patients were forced to undergo useless, time consuming, costly and sometimes harmful procedures. A retrospective review of a preoperative investigation in a Karachi hospital showed that even in a developing country, where tendencies have leaned to advocating routine testing in asymptomatic, healthy patients to identify undocumented medical conditions, no useful information is obtained. A British study systematically looked at several components of routine testing. Although random preoperative chest X-rays may be reported as abnormal in 2.5-37% of cases, there was a change in clinical management in only 0-2% of patients. Chest X-ray as a baseline is of value in less than 9%. In most patients, lung disease has already been identified and is the reason for surgery. Preoperative EKGs are abnormal in 4.6-31.7% of cases but lead to a change in management in only 0-2.2% of patients. There is no evidence to support the value of recording a preoperative EKG as baseline in an asymptomatic patient. The proportion of abnormal tests rises with age and worsening ASA status.
The mean number of tests declined from 8 to 5.6 after new guidelines were established. There were no untoward events and savings totaled $34,000 for the 279 patients in the study.
Essential Tests: Are There Any?
If repeated studies have indicated that routine testing is unnecessary and may even be harmful, what, if anything, should be done preoperatively in the otherwise healthy individual? Out of an electronic database search of 56,119 medical records, researchers at the Mayo Clinic identified 5,120 patients for whom no testing was done within 90 days of the procedure. From this group, 1,044 patients were selected randomly to document the absence of preoperative tests, the presence of preexisting disease (by organ system), the type of anesthetic agent and the outcomes and tests intraoperatively and postoperatively. The age range was 0 to 95 years. No deaths or major perioperative morbidity occurred. Although 10 patients underwent type and screening for antibodies immediately preoperatively, no blood transfusions were necessary. Intraoperatively, 17 laboratory tests and I EKG were obtained. Three results were abnormal. Postoperatively, 42 blood tests and 2 EKGs were obtained. Five of the blood tests were abnormal (Hb levels in 3, serum sodium in 1 and arterial blood gases in 1). One EKG showed premature ventricular contractions. No tests done either intraoperatively or postoperatively changed surgical or medical management. The findings concluded that patients who have been assessed by history and physical examination and determined to have no preoperative indication for laboratory tests can safely undergo anesthesia and surgery with tests then performed only as indicated, either intraoperatively or postoperatively.
Other studies have focused on the cost effectiveness of routine type and screening before endoscopic procedures. Usal et al found that the incidence of blood transfusion during laparoscopic cholecystectomy was 0.46% (data base of 2,589 patients) and that for open cholecystectomy was 5.47% (data base, 603 patients).24 Two of the 12 patients who required transfusion in the laparoscopic group sustained major vascular injury. The others had preexisting medical conditions such as sickle cell disease, end stage renal disease and iron deficiency anemia. The authors estimated that the elimination of routine blood type and screening could save $80,000 during a 6-year period. In another study of 7,529 women undergoing laparoscopic procedures, Ransom et al identified 57 who required blood transfusions. All transfusions were the result of previously identifiable problems, including ectopic pregnancy and preoperative anemia. No blood replacement was required emergently, thus allowing time for typing and cross matching. The investigators extended their study to women undergoing vaginal hysterectomy for non-cancerous conditions. Of 1,063 patients, 26 needed transfusions. Ten of the transfusions were given preoperatively because of anemia. Again there were no emergent situations. Thus, the authors conclude that in the absence of preoperative indications, routine preoperative type and screening does not enhance patient care and should be eliminated. Policy at Mount Sinai Medical Center in New York does not require that blood be made available in the operating room during donor kidney transplantation (2006).
In 2003, the National Institute for Clinical Excellence (NICE) in the United Kingdom released guidelines for preoperative testing for physicians and patients. Involvement by NICE was motivated by variability in testing and uncertainty as to the value of routine tests. Eleven tests were considered and color coded as to appropriateness. The following year a review of the impact of these guidelines was published. In essence, the clinician needs to know only 4 key facts about a patient: 1, the age band, 2, the complexity of surgery, 3, the ASA classification and 4, the nature of co-morbidities. Based on analysis of these items, tests may or may not be required. Similarly, in 2002 the ASA issued a practice advisory for pre-anesthesia evaluation28 and concluded that a pre-anesthesia history and physical evaluation must precede the ordering, requiring, or performance of specific pre-anesthesia tests.
It would thus appear that a careful history and physical examination and a properly executed informed consent plus an admonition not to eat or drink for 4-6 hours (as per John Snow, 1858 ref 2) constitute routine preoperative evaluation. After that, tests should be ordered only if the medical history indicates appropriateness. The argument that general testing could substitute for an annual evaluation is clearly absurd as the cancer detection methods (mammography, colonoscopy, prostatic cancer blood test) have no bearing on anesthetic management. No doubt laboratory tests can help to optimize a patient’s preoperative condition once a disease has been identified. But routine test have inherent shortcomings: they may fail to uncover pathologic conditions, the abnormalities that they uncover may not improve patient care or outcome, they are inefficient in screening for disease that has not been identified by history and physical examination, follow up may be inadequate, and false positives may lead to considerable patient distress and provoke further invasive testing that may have even more harmful consequences. When designing a "required labs’ policy the physician must consider local and state requirements in addition to those of the governing board of the facility (e.g. JCAHO, AAAHC etc.) It is important to remember that the requirements for testing in an outpatient facility may differ from those mandated when the patient is treated in an in-patient health center, even though the planned procedure is identical. However, the most difficult stumbling block remains the mind set of surgeons and nurses who have long been comfortable with a routine and lists that are completed. Even though physicians frequently do not review the results of laboratory tests and might even not be aware of abnormal values or how to interpret them (e.g. EKG), they often find a comfort in seeing a so-called completed chart with all check marks intact. Education is required as is dissemination of the knowledge obtained from these very large studies. Interestingly, surgeons can often be convinced that routine testing is unnecessary when they themselves are required to undergo a procedure and find themselves spending many hours waiting for a chest X-ray or a technician to be available to complete an EKG, only to find that all the studies are normal.
There are many sub-specialties within surgery and many subsets of patients. Several studies have addressed some of the unique characteristics of each group. General indications for identification of appropriate tests within each patient population are noted in Tables 1-8.
Routine preoperative testing of children has historically included complete blood count and urinalysis (UA) in addition to history and physical examination and pregnancy testing in adolescents. Questioning the usefulness of this approach, O’Connor and Drasner retrospectively reviewed 486 elective surgeries in children to determine the role of abnormal laboratory test results in preoperative management. Anemia or microcytosis was identified in 17% and abnormal UA results in 15%. More that 80% of the abnormal UA results were known historically, clinically insignificant or false-positives. Surgery was cancelled in 5 children due to abnormal findings: two due to anemia, two for abnormal UA, and in one because of a prolonged partial prothrombin time. The children with anemia were treated with iron and underwent uneventful surgery. Of the abnormal UA, one was contaminated and the cancellation of surgery resulted in a complication that required emergency surgery. The data indicated that a routine UA adds little to the preoperative evaluation of a healthy child and should be omitted.
Another study considered the value of routine Hb measurement.29 Two thousand patients ranging in age from one month to 18 years were studied. Eleven patients, all of whom were older than 5 years, were found to be anemic. Surgery was postponed in 3 patients (27%) pending iron therapy. In the other 8 patients (73%), anesthesia and surgery were completed without incident. The authors conclude that the low incidence of anemia and the low deferral rate of anemic children question the value of preoperative Hb testing in this age group.
Other investigators have challenged the routine use of pregnancy testing. In a study of 444 patients who underwent 525 procedures, Malviya et al found that the preoperative history obtained from adolescent patients was in complete agreement with the pregnancy test results. Although only 8 patients in the survey stated that they might be pregnant, all urine tests were negative. A false positive was obtained in one patient who was anesthetized before the test result was available. Subsequently she was shown not to be pregnant. The authors suggest that a detailed history regarding last menstrual period, contraceptive practices, sexual activity and the possibility of pregnancy be obtained in all postmenarchal patients presenting for surgery. Human chorionic gonadotrophin testing is reliable but appears to be necessary only if indicated by history.
The authors of another study to evaluate pregnancy testing protocol in adolescents undergoing surgery reached a different conclusion. A total of 801 girls were tested and a positive pregnancy test was found in 6. Two of these were false positives: four were pregnant (0.49%). Sexual activity was disclosed by two, one of whom admitted to the chance of pregnancy. The other three all denied the possibility of pregnancy. The surgical procedure was postponed in all 4. The authors felt that the current protocol contributed to enhanced care and can identify adolescents who need counseling before surgical procedures are performed. However, one might argue that less than half a percent is a very low return, delaying surgery may have adverse effects and moreover the risk of anesthesia to the fetus is probably very low.
It seems, however, that the area of pregnancy testing is one where anesthesiologists are more prone to seek assurance. Questionnaires were distributed to physicians at the Society of Obstetrical Anesthesia and Perinatology. About one third mandated testing through hospital policy. Anesthesiologists mandated routine pregnancy testing of all elective (30%) versus all emergency (17%) surgical patients. Sixty six percent would require rather than simply offer (20%) testing when the history suggested pregnancy. Twenty percent indicated that elective surgery would be cancelled if the patient were pregnant or refused testing. Interestingly, while 98% of respondents recognized their legal requirement to report child abuse and 82% believed pregnancy in a juvenile to constitute child abuse by definition, fewer than 4% would report the information to the police, even if the impregnating male was known to be an adult. Policy at Mount Sinai Medical Center, New York in 2006 requires pregnancy testing of all women of child bearing age.
There would appear to be good evidence that routine preoperative testing in children is not indicated. To find out current practices in the United States, a survey was done of members of the Society of Pediatric Anesthesia. Six hundred and eighty five questionnaires were reviewed. Results indicate that Hb testing is routine in 27%-48% of children depending on age. UA is ordered in less than 15%. Pregnancy testing is routine in 43%. Hemostatic tests prior to tonsillectomy are conducted by 45% of anesthesiologists. The results seem to indicate that the practice of routine preoperative testing of children is still prevalent in spite of the many studies that indicate no specific benefits in healthy children.
Patients with Cardiovascular Disease
Much has been written about the special needs of the patient undergoing major cardiac and vascular surgery and how complications can be minimized by better preoperative preparation. Perioperative cardiac morbidity affects more than 1.5 million people in the United States and costs more than $20 billion annually in health care resources.35 Solutions to this problem have been addressed using routine history and physical examinations and risk indices. Non-routine cardiovascular testing has been recommended including exercise and Holter electrocardiography, radionuclide ventriculography, stress echocardiography, and dipyridamole thallium scintigraphy (DPT). These tests all have strengths and weaknesses. Although large scale studies are needed to fully assess the cost effectiveness of total cardiac evaluation, several studies have already indicated that more is not always better.
Carliner et al. conducted a prospective study of preoperative exercise testing in 200 patients over 40 years of age, scheduled for elective major noncardiac surgery under general anesthesia.36 The exercise test response was positive in 32 (16%), equivocal in 11 (5.5%) and negative in 157 (78.5%). Patients were followed with serial EKGs and determination of serum creatine kinase (CK) and CK-MB. Six patients (3%) had primary endpoints: 3 died and 3 had definite postoperative myocardial infarctions. Secondary endpoints of suspected myocardial infarction occurred in 27 patients (14%). Endpoint events were more common in those over 70 years and in those with an abnormal preoperative exercise stress test (27% vs 14%). Nevertheless, preoperative exercise results were not statistically significant independent predictors of cardiac risk. In fact, the only significant risk factor by multivariate analysis was the preoperative EKG. Endpoint events were more common in patients with an abnormal tracing than in those with a normal record (27% vs 7%). Because the results of exercise testing do not appear to add substantially to the risk separation provided by the EKG at rest, exercise testing is not recommended as a routine preoperative method to assess risk in older patients before major non-cardiac surgery.
In fact most patients with coronary artery disease (CAD) who are screened are found suitable for surgery without additional invasive intervention and cardiac revascularization.37 Good functional capacity (defined as the ability of a 50-70 year old patient to achieve 6-8 METS of activity without significant signs of dyspnea or exertion) and absence of cardiac risk factors (angina, congestive heart failure, diabetes, previous myocardial infarction, ventricular ectopy, and the elderly) should allow elective surgery without further evaluation in 30-40% of the time. The findings of high-risk perfusion scan abnormalities is limited to about 15-20% of patients recommended for further noninvasive testing. Clinical evaluation combined with exercise testing and dobutamine stress echocardiography may also reliably assess cardiac function before aortic reconstruction and be used as a reliable alternative to routine coronary arteriography. Selective screening may be beneficial. In one large study, patients were divided into 4 groups: Group 1 proceeded directly to surgery; Group 2 was screened with DPT; Group 3 patients all underwent coronary angiography; in group 4, selective screening was used. Patients were divided to high, intermediate and low risk using clinical criteria. High risk patients had perioperative angiography, intermediate patients had noninvasive screening and low risk patients had no further testing. Patients were followed for 5 years after abdominal aortic aneurysm repair. Group 1 had the poorest 5 year survival (77.4%) as compared to the other groups (all had survival rates of 86-87%). The incremental cost effectiveness ratio for selective testing was significantly lower than for routine angiography ($44,880/years of life saved vs $93,300/years of life saved). Thus, selective screening before vascular surgery may improve 5 year survival and be cost effective. Neither routine noninvasive testing nor routine angiography is cost effective when compared with currently accepted medical therapies. Along the same lines, another study examined the cost implications of selective preoperative risk screening in patients undergoing peripheral vascular operations.40 Routine versus selective screening strategies were compared for patients with an intermediate likelihood of CAD on the basis of history and symptoms. In-hospital costs were 11% higher for routine screening by cardiac catheterization ($27,760) than for routine pharmacologic stress imaging ($24,826). The total cost of a do-nothing course (i.e. no preoperative testing) was 5.9% less than that of a routine pharmacologic stress test and 15.9% lower than cardiac catheterization. Selective screening resulted in further reduction of cost to a level similar to the do-nothing group. Use of noninvasive testing was cost effective for patients 60-80 years of age. Cost per life saved ranged from $21,790 to $33,338. Coronary revascularization after an abnormal noninvasive test was cost effective only for patients older than 70 years. With a selective screening approach, the economic impact of initial diagnostic testing may be minimized without adverse patient effects.
Patients with Respiratory Disease
Patients with both acute and chronic respiratory disease present special problems, in regards to both intraoperative management and to the anticipation and prevention of complications. Risk factors for the development of postoperative pulmonary complications include surgical site, smoking and the presence of pulmonary disease. While pulmonary function testing (including measurement of arterial blood gases) can predict pulmonary function after lung resection, they are of no value in predicting postoperative pulmonary complications. For example, the degree of airway obstruction as assessed by the forced expiratory volume in 1 sec is not a significant independent risk factor for the development of postoperative respiratory failure after major abdominal surgery, even in smokers with severe lung disease. Although spirometry or peak flows may help to monitor the status of patients with asthma, pulmonary function tests should be viewed as a management tool to optimize preoperative lung function and not as a means to assess risk. These tests should be ordered only if the results will be used to modify surgical, anesthetic or postoperative pain management. Routine preoperative pulmonary function testing is probably a waste of resources. However, some 20 million Americans are affected with asthma and many require some type of surgery. A recent report notes that preoperative assessment should include first a specialized history and physical examination which should then determine what if any pulmonary function tests are indicated. Treatment with steroids and bronchodilators may be indicated to prevent inflammation associated with intubation. The search for useful noninvasive methods of assessing lung function continues. Surface echocardiographic imaging may be a feasible tool to identify patients with cor pulmonale, suggesting pulmonary hypertension who are scheduled for lung reduction surgery.
As noted above, routine laboratory testing of children prior to anesthesia has been or should be, largely discarded. However, in the presence of an upper respiratory infection (URI), a complete blood count, including differential may be indicated although the usefulness of this test has been questioned. White blood cell counts over 10,000/cu mm should raise the suspicion of superimposed bacterial infection that might benefit from a course of antibiotics preoperatively. Little or no information can be obtained from a chest X-ray – especially if chest sounds are clear. Even in the child with a history of asthma who has a superimposed URI, a chest X-ray is rarely warranted. Also, pulmonary function tests, including preoperative arterial blood gas analyses will not offer any useful information and will probably only create a very unhappy child and anxious parents. Auscultation of the chest and careful inspection of the upper airway, ears, and eyes should be made. Notation of erythema is important. Findings of marked inflammation, bulging, tender eardrums, productive cough, purulent nasal secretions, and temperature over 38.5°C, chest rales or bronchospasm indicate that postponement of the case is probably prudent; however, wheezing is a common manifestation of viral upper respiratory tract disease in infants. Should this be the single finding, cancellation may not be necessary especially if intubation can be avoided. From a pediatric database of 29,220, Cohen and Cameron identified 1283 children with URIs. All patients had clear nasal secretions, may or may not have been febrile, but all had normal chest examinations and white cell counts. Throughout the perioperative period, children with an URI were at a 2-7 fold increased risk of a respiratory related event. Children with an URI but who were not intubated were almost 9 times more likely to develop problems than control patients. Intubation increased the risk of a critical respiratory incident by over 5 times, even in otherwise healthy children. Intubation in the presence of an URI was associated with an 11 fold risk increase. In a study of 3,585 cases, Tait and Knight found no increase in complications of anesthesia in symptomatic patients but a 3 fold increase in bronchospasm and laryngospasm in asymptomatic patients with a history of recent URI. Recognizing that patient selection may have been skewed towards selection of patients with chronic symptoms of nonviral origin, the authors performed a prospective cohort study of 489 patients undergoing myringotomies between the ages of 1-12 to investigate the prevalence of symptomatology and perioperative respiratory complications. Not only did they demonstrate no increase in morbidity, but, in mostly nonintubated patients, administration of general anesthesia to 78 children (combined, of course, with surgery to relieve chronic ear infection) appeared to decrease the appearance and duration of several respiratory symptoms. These same authors also showed that different anesthetic agents may attenuate the histopathologic response to influenza viral infections in animals (ferrets) and halothane may even inhibit viral replication. This evidence would support the clinical impression that a recent viral infection is a risk factor for pulmonary complications, albeit, usually of a temporary nature. Morbidity and mortality rates were not increased by an URI over control figures in any of the studies mentioned and were not influenced by preoperative testing.
There has been debate as to what tests are appropriate for elderly patients undergoing cataract extraction. As background, a national survey of ophthalmologists, anesthesiologists and internists was conducted in 1991. Fifty percent of ophthalmologists, 40% of internists and 33% of anesthesiologists frequently or always obtained chest X-rays, while 20% of ophthalmologists, 27% of internists and 37% of anesthesiologists never obtained chest X-rays unless dictated by another medical condition. Similarly, 70-90% of ophthalmologists, 73-79% of internists and 41-79% of anesthesiologists usually obtained a complete blood count, electrolyte panel and EKG but 4-11% of ophthalmologists, 14-17% of internists and 9-28% of anesthesiologists never obtained these tests. The majority of responders (up to 80%) believed tests were unnecessary but cited many reasons for doing them (e.g. medicolegal, institutional policy). Preoperative evaluation was also viewed as a screening opportunity and each specialist suggested that the other two physicians required the tests. Nine years later, a study of 19,557 patients showed that routine testing is of no value or importance in this patient population. Patients were randomly assigned to have a complete battery of tests or none at all. The most frequent medical events in both groups were hypertension and bradycardia. The overall complication rate was the same at 31 events per 1000 operations. There were no differences in the incidence of intraoperative events or postoperative problems. Therefore, routine medical testing before cataract surgery does not measurably increase the safety of surgery.
Screening for HIV Infection
Universal precautions are officially recommended to prevent HIV transmission to healthcare workers. For elective surgery, some practitioners advocate routine preoperative HIV testing. Lawrence et al performed an economic evaluation to compare both strategies using a simple approach to determine if routine testing was less expensive than universal precautions. They determined that routine testing at that time cost $57 and universal precautions cost $36 per procedure, indicating the economic feasibility of the latter method.
Thus, assessment of blood sugar levels should be made preoperatively in all known diabetic patients. Levels above 150-180mg/dl should be treated. Most hospital labs now consider 120mg/dl to be the upper limit of normal. Not uncommonly, patients who usually achieve good glycemic control with oral agents may require insulin therapy perioperatively.
The “How” of Pre-Anesthetic Assessment
There are several commonly used approaches for screening patients preoperatively. The patient may visit the hospital or facility prior to surgery, or a visit may be made to the office. A telephone interview only may be conducted. A health survey may be reviewed, again without a visit. Screening may be conducted on the day of surgery. Finally, information may be gathered by a computer assisted system. A video presentation has been used as an additional component of the preoperative interview. However and wherever the evaluation is conducted the ability to obtain an accurate history and clear information of a physical examination is key. Good communication and preparation benefit everyone.
We are continuing to order too many tests preoperatively. Preoperative assessment is a complex and multidisciplinary task. The most important factor in pre-anesthetic assessment is a careful history and physical examination and based on findings at that time, appropriate tests may be ordered. Surgical and anesthetic management is rarely changed by the results of routine preoperative tests. Many patients require no testing at all, even patients with significant heart and lung disease and the elderly. Education of the medical community is essential and long overdue in this area. Also, as the public becomes increasingly aware that routine testing and the inconvenience, pain and wasted time and money that it generates is of no value, it will, too, push for change.