Malviya, Shobha MD; D'Errico, Celia DO; Reynolds, Paul MD; Huntington, John MD; Voepel-Lewis, Terri MS, RN; Pandit, Uma A. MD
A number of studies suggest that routine preoperative laboratory testing has limited diagnostic value and rarely detects clinically significant abnormalities that alter perioperative management [1-5]. These findings, in light of the demand for appropriate use of health care resources, have led to the elimination of several routine laboratory investigations without adversely affecting patient outcome [6,7]. The issue of routine preoperative pregnancy testing in adolescent and adult women remains unresolved. A recent retrospective review of mandatory pregnancy testing in 412 presurgical adolescent and adult patients ranging in age from 10.5 to 20 years reported a 1.2% incidence of pregnancy . In three of the five patients with positive pregnancy tests, surgery was deferred, while in two the anesthetic management was modified. Another study demonstrated that the incidence of unrecognized pregnancy in menstruating women of childbearing potential presenting for ambulatory surgery was 0.3% . All women in that study elected to cancel or defer surgery when they were informed of their positive pregnancy test results.
Statistics regarding teenage pregnancy suggest that one of every 10 women between 15 and 19 years of age becomes pregnant each year in the United States . In 1990, approximately 835,000 females in this age group became pregnant. Ninety-five percent of these pregnancies were unplanned . Conceivably, some of these patients may present for elective surgery with an unrecognized pregnancy. There are few data regarding the teratogenic effects of anesthetics on the human fetus . However, previous investigators have reported a significant risk of spontaneous abortion, prematurity, intrauterine growth retardation, and death within seven days after birth in the offspring of patients who had been anesthetized during early pregnancy [13,14]. These risks may make it prudent to defer elective surgery in pregnant patients. The practice of routine preoperative pregnancy testing is therefore an important issue in adolescents, who may be unable or unwilling to provide a reliable history regarding sexual activity, last menstrual period, and the possibility of being pregnant.
A recent nationwide survey found that 27% of practitioners routinely order a pregnancy test in adolescent patients prior to surgery . The most common reason for routine testing cited by the respondents was concern regarding the reliability of the history of sexual activity or of the last menstrual period in this patient population. On the other hand, cost was the most common reason cited for not testing patients routinely for pregnancy. In addition to cost, routine preoperative pregnancy testing raises issues related to patient autonomy, the ethics of patient confidentiality, and medicolegal concerns regarding informed consent . These issues may be avoided if a reliable preoperative history regarding pregnancy can be obtained. The following study was undertaken to determine whether the preoperative history obtained from adolescent patients regarding the possibility of pregnancy correlates with urine human chorionic gonadotrophin (HCG) testing results.
With approval from the institutional review board, all adolescent, postmenarchal female patients presenting for elective outpatient surgery from September 1993 to January 1996 were enrolled in this study. All patients were educated about the potential risks of anesthetics to the fetus and were questioned in a standardized fashion regarding the possibility of pregnancy both at the time of their preanesthesia screening clinic visit, which was typically 1 day to 3 wk prior to surgery, and on the day of surgery in the preoperative admitting area. The history was obtained by a female member of the nursing staff in the absence of family members whenever possible. This was accomplished by questioning the patient in the examination room during assessment of vital signs while parents waited in the waiting area. In addition, regardless of the history, a first morning urine sample was sent for qualitative HCG testing in accordance with standard institutional practice. Patients and their families were informed regarding institutional policy for routine pregnancy testing and the reasons for this policy. Urine pregnancy testing was performed on the day of surgery in all cases using an Abbott Registered Trademark test pack (Abbott Laboratories, Abbott Park, IL). This self-performing assay uses monoclonal and polyclonal antibodies to detect HCG in serum and urine. It detects HCG levels >or=to25 mIU/mL. This test has a sensitivity of 99.4% and a specificity of 99.5%.
The results of the test and their impact on the decision to proceed with surgery or on anesthetic management were prospectively documented. In addition, cancellation of surgery, delays in the operating room schedule, patient or parent refusal to be tested, and patient or parent reactions to the institutional policy regarding preoperative pregnancy testing were noted in a prospective manner. For all patients who were not tested prior to surgery during the study period, the reasons for not testing were recorded.
In accordance with standard institutional practice, the surgeon and anesthesiologist were jointly responsible for informing the patient of a positive preoperative pregnancy test result. Additionally, in the event of a positive test, the patient was given the option to be present when the family was informed of the results and the options regarding proceeding with the surgery, alterations in anesthetic management, and risks of anesthesia in a pregnant patient were presented. Data are presented in a descriptive fashion.
Four hundred forty-four patients who met eligibility criteria underwent 525 elective surgical procedures during the study period. The patients ranged in age from 10 to 17 yr (mean +/- SD = 14.5 +/- 1.6 yr, median age 15 yr). For the fiscal year ended June 1995, the payor mix of our population was composed of Medicaid 35%, Blue Cross 27%, commercial insurance 32%, and managed care 6%. Approximately 7000 anesthetics are administered to patients under 18 yr of age each year at this institution. For 57% of the procedures, the preoperative history was obtained in the absence of family members, and for the rest, a family member was present during the preoperative interview. Patients presenting for 508 procedures denied the possibility of pregnancy. Eight patients stated there was a possibility that they could be pregnant, while in another six, the parents responded for the patient and denied the possibility of pregnancy. None of the eight patients who reported a possibility of pregnancy were with their parents at the time of disclosure. Three patients were unwilling to respond to the question, and in each of these cases the parents were present at the time of the preoperative interview.
Pregnancy testing was not performed in 17 patients due to patient/parent refusal (n = 9) or inability to void (n = 8). The 9 patients (mean age 13.4 +/- 1.6 yr, median 13 yr) who refused the test included 4 mentally impaired patients, 2 patients who were menstruating at the time of surgery, and a 12-yr-old patient whose mother stated that the patient had only menstruated twice and there was no possibility that she could be pregnant. The remaining two of these nine patients (aged 13 and 15 yr old) stated that they were sure they were not pregnant and refused the test. Of the eight patients who were unable to void (mean age 13.5 +/- 1.7 yr, median 13 yr), two were mentally impaired. The decision to proceed with anesthesia without a preoperative pregnancy test was made by the attending anesthesiologist on an individual basis in each of these 17 cases.
All pregnancy test results were negative except one that was questionably negative. This patient had denied the possibility of pregnancy and had been anesthetized prior to test results. A follow-up phone call after surgery revealed that the patient was not pregnant and had menstruated in the interval between surgery and follow-up. An additional six patients were anesthetized prior to test results becoming available. Five patients underwent two pregnancy tests for the same surgical procedure, 11 were tested even though they were menstruating at the time of surgery, and two premenarchal patients were tested. Four patients were taking birth control pills and were tested. Ten surgical procedures were delayed while awaiting pregnancy test results. After the first 158 tests, the preanesthesia personnel were trained to perform urine HCG testing and the test kits were stocked in the preanesthesia screening area. Eight of 10 delays occurred prior to this change and two after this change due to lack of availability of test kits. Seven families were offended by the policy of pregnancy testing.
The history obtained from the study population was in agreement with urine HCG results in all cases except the patient with the questionably negative test. It was later determined that even this patient's history should have correlated with the test result. Anesthetic and surgical management was not altered by pregnancy test results in any patient in this study population. These data could not be subjected to correlation and/or other statistical analysis due to the lack of a positive test.
The issue of routine preoperative pregnancy testing in adolescent patients remains controversial. Previous investigators have reported that, in their population, history and the results of pregnancy testing do not always agree . Yet other investigators contend that the possibility of a false-positive result, the low yield of positive results in this population, and the potential for emotional upheaval do not support the practice of routine preoperative pregnancy testing . Our data suggest that in the adolescent presurgical patient, history taking may be a reliable method of determining the possibility of pregnancy prior to surgery. The importance of obtaining the history in the absence of family members is underscored by the three patients who were apparently unwilling to provide a history in the presence of their family regarding the possibility of their being pregnant.
The results of the present study, however, do not support the observations of previous investigators who have evaluated the reliability of history in ruling out pregnancy in presurgical and emergency department patients [8,9,17]. The discrepancies among the results of the current study and those of previous studies may be explained in part by age and other differences in patient populations, the clinical situations studied, and study design. On the basis of a retrospective review, Azzam et al.  reported a 1.2% incidence of pregnancy in a cohort of 412 young women scheduled for elective surgery. It is unclear from their report whether a specific history related to the possibility of pregnancy was sought in every case. Of the five patients with a positive HCG test, sexual history/last menstrual period information was available in only three patients. It is difficult to determine from these data whether a detailed, specific history would have indicated the need for a pregnancy test. Manley et al.  reported a 0.3% incidence of unrecognized pregnancy in 2056 women undergoing outpatient surgery. The ages of the study group or of the women with positive pregnancy tests are unclear from their data. However, since several of their patients were scheduled to undergo fertilization and other gynecologic procedures, it would be safe to assume that some of their patients were adults. Additionally, one could assume that women who are scheduled to undergo fertilization procedures are actively pursuing pregnancy and may have a higher incidence of pregnancy than an adolescent population.
Ramoska et al.  evaluated the reliability of history in excluding pregnancy in emergency department patients ranging in age from 12 to 49 years (mean age 23 years). These investigators reported three historical variables that were statistically less likely to be associated with pregnancy, including last menstrual period that was on time, the patient thinking that she was not pregnant, and the patient denying any chance that she could be pregnant (P < 0.001). Contrary to the results of the present study, however, they found that even with these historical criteria there was a 10%-15% chance of the patient being pregnant. It is difficult to determine from these data the number of adult patients included in the study or whether any of the patients who had positive pregnancy tests were adolescents.
Differences in clinical situations may also account in part for the differences in the results of the studies. Ramoska et al.  studied a population of women presenting to the emergency department with abdominal pain, vaginal bleeding, irregular or missed periods, or concerns about possible pregnancy. The prevalence of pregnancy in such a population may be greater than that in a population of presurgical adolescents. Additionally, it may be difficult to elicit a history in private regarding delicate issues, such as sexual activity or pregnancy, in a busy emergency department with patients of varying degrees of acuity. Variability in study design may also have contributed to differences in findings among the studies. Prior to eliciting the history regarding the possibility of pregnancy, all patients included in our study were made aware that the adverse effects of anesthetics on the unborn fetus were not clearly defined. Provided our patients understood these explanations, they may have been more likely to provide an accurate history regarding their pregnancy status. It is unclear whether the patients included in the Azzam and Manley studies [8,9] were educated about the potential adverse effects of anesthetics on the fetus prior to questioning them about pregnancy.
Most routine preoperative laboratory tests, such as blood counts, chest radiographs, and electrocardiograms, do not alter perioperative management even when abnormalities are detected [3,7,19]. Manley et al.  have demonstrated that positive preoperative pregnancy tests, however, resulted in deferral or cancellation of surgery, with patient desire for cancellation being the major determining factor in all cases. The actual cost of pregnancy testing in this study was $2879 per pregnancy discovered. Although the positive tests resulted in cancellation of surgery, these authors concluded that the cost-effectiveness of routine preoperative pregnancy testing was not clear. Azzam et al.  reported that the average hospital charge for pregnancy testing was $25-$30 per test. Based on this charge, the cost per pregnancy discovered in their population was approximately $2060. In our study, the charge to the third-party payors was $21 per completed pregnancy test for a total of $10,290. In addition to this expense, there were unknown hospital costs related to delays resulting in loss of operating room time and duplication of tests. The costs of subjecting the entire female adolescent population to routine preoperative pregnancy testing must therefore be carefully weighed against the potential risk of anesthetizing an adolescent patient with an undisclosed pregnancy. These costs could be greatly reduced if preoperative pregnancy tests were ordered only if indicated by history. When testing is indicated, delays related to waiting for results may be eliminated by performing the tests in the presurgical area. In our setting, implementation of on-site testing also eliminated the unacceptable practice of anesthetizing patients prior to test results being available. Coordination between surgery and anesthesia services would minimize unnecessary costs related to duplication of tests.
It is possible that the history obtained from our patients may not reflect the true incidence of pregnancy among all presurgical adolescents. However, our data suggest that history may be reliable in excluding pregnancy in some adolescent populations since we found 100% agreement between history and HCG results. One could make the argument that refusal of pregnancy testing in 17 of our patients was due to fear of disclosure and that, if tested, some of these patients may have been found to be pregnant. In all 17 cases, however, either the patient or her parent firmly refused testing after a detailed discussion of institutional policy and the potential risks of anesthetizing a pregnant patient. In most of these cases, insistence on performing the test may have led to cancellation of surgery by the parent. We believe that the extenuating circumstances that resulted in these patients not undergoing testing would have been present at any institution regardless of the policy of preoperative pregnancy testing.
The incident involving a questionably negative HCG test did not alter patient management in any way, since the patient had already been anesthetized prior to test results being available. Previous reports have suggested that between 30% and 60% of all unexpected abnormalities found on preoperative laboratory testing are not noted preoperatively . An abnormality that is detected but not appropriately acted upon poses a greater liability risk than one that is not detected preoperatively . With professional liability considerations, it may have posed less of a medicolegal risk to the physician had the pregnancy test not been done in this patient than to have proceeded with the anesthetic before the results were available. This case highlights the risk of subjecting an entire population to a laboratory test on the basis of institutional policy that the individual practitioner may not consider important.
In summary, our data suggest that adolescent patients who have been educated about the potential fetal and maternal risks of anesthetics may provide a reliable history regarding the possibility of pregnancy. Routine preoperative HCG testing of the entire adolescent population may not be necessary. In our population, it may have been a reasonable alternative to perform a pregnancy test only in the eight patients who stated that they may have been pregnant or the three patients who were unable to provide a history. Extrapolation of our data to other geographic regions, however, requires assurance of similar demographic and sociologic factors . We recommend that, as with all other preoperative investigations, the cost of subjecting an entire population to a preoperative pregnancy test with a small likelihood of a positive result be balanced against potential benefit to the patient.
1. Roizen MF. The compelling rationale for less preoperative testing. Can J Anaesth 1988;35:214-8.
2. Kaplan EB, Sheiner LB, Boeckmann AJ, et al. The usefulness of preoperative laboratory screening. JAMA 1985;253:3576-81.
3. Gold BS, Young ML, Kinman JL, et al. The utility of preoperative electrocardiograms in the ambulatory surgical patient. Arch Intern Med 1992;152:301-5.
4. Johnson H Jr, Knee-Ioli S, Butler TA, et al. Are routine preoperative laboratory screening tests necessary to evaluate ambulatory surgical patients? Surgery 1988;104:639-45.
5. Turnbull JM, Buck C. The value of preoperative screening investigations in otherwise healthy individuals. Arch Intern Med 1987;147:1101-5.
6. Larocque BJ, Maykut RJ. Implementation of guidelines for preoperative laboratory investigations in patients scheduled to undergo elective surgery. Can J Surg 1994;37:397-401.
7. Charpak Y, Blery C, Chastang C, et al. Prospective assessment of a protocol for selective ordering of preoperative chest x-rays. Can J Anaesth 1988;35:259-64.
8. Azzam FJ, Padda GS, DeBoard JW, et al. Preoperative pregnancy testing in adolescents. Anesth Analg 1996;82:4-7.
9. Manley S, de Kelaita G, Joseph NJ, et al. Preoperative pregnancy testing in ambulatory surgery. Anesthesiology 1995;83:690-3.
10. Trussell J. Teenage pregnancy in the United States. Fam Plann Perspect 1988;20:262-72.
11. State-specific pregnancy and birth rates among teenagers--United States, 1991-1992. MMWR Morb Mortal Wkly Rep 1995;44:677-84.
12. Levinson G, Shnider SM. Anesthesia for surgery during pregnancy. In: Shnider SM, Levinson G, eds. Anesthesia for obstetrics. Baltimore: Williams & Wilkins, 1993:264-72.
13. Duncan PG, Pope WDB, Cohen MM, Greer N. Fetal risk of anesthesia and surgery during pregnancy. Anesthesiology 1986;64:790-4.
14. Mazze RI, Kallen B. Reproductive outcome after anesthesia and operation during pregnancy: a registry study of 5405 cases. Am J Obstet Gynecol 1989;161:1178-85.
15. Malviya S, Reynolds P, D'Errico C, et al. Should pregnancy tests be routine prior to surgery in adolescent patients? [abstract]. Anesthesiology 1994;81:A1385.
16. Duncan PG, Pope WD. Medical ethics and legal standards. Anesth Analg 1996;82:1-3.
17. Ramoska EA, Sacchetti AD, Nepp M. Reliability of patient history in determining the possibility of pregnancy. Ann Emerg Med 1989;18:48-50.
18. Orr RJ. Pregnancy testing in adolescent presurgical patients [letter]. Anesth Analg 1995;81:661.
19. Nigam A, Ahmed K, Drake-Lee AB. The value of preoperative estimation of haemoglobin in children undergoing tonsillectomy. Clin Otolaryngol 1990;15:549-51.