In Reply:--Lewis and Cooper raise a valid concern of positive and negative predictive values. Although briefly addressed in an earlier version of our original manuscript, it was dropped because of editorial considerations. However, we are concerned about their use of data from a preliminary study [1]presented in abstract form to calculate specificity and the positive predictive value of preoperative pregnancy testing. Inclusion of the abstract by Malviya et al. [2]in our discussion was not an endorsement of their data to serve a role in calculation of the specificity of pregnancy testing. Their data, which examined 179 patients, included a single case of a "questionably positive" test result in a patient who was not pregnant. Had this result been interpreted as a true negative rather than a false positive, their specificity would have been the same as that calculated from our much larger data: 100%. Because of small sample size, the single questionable result had a major impact on specificity and, thus, the positive predictive value of the test. In our series of 2,056 patients, clinical laboratory pregnancy testing of UCG in either urine or blood yielded neither false nor ambiguous results, with a 95% confidence interval of false result between 0.0 and 0.0018.
The scope of our study was to examine the incidence and impact of previously unsuspected pregnancy in patients scheduled for ambulatory surgery discovered during routine preoperative pregnancy testing. We understand Lewis and Cooper's concerns about the specificity of pregnancy testing in a population with an overall low prevalence of pregnancy, and perhaps further studies are warranted to examine the predictive value of routine preoperative pregnancy testing in ambulatory surgery patients.
The letters from Rosenberg and Zeig and Herschman raise the issue of performing pregnancy testing as close to the time of surgery as possible, preferably on the same day. In our study, pregnancy testing often was performed at the time of their laboratory testing, albeit often several days in advance, to lessen inconvenience to the patient. We agree that "closing the window as tightly as possible" is preferable, especially considering the available testing kits, which can accurately and quickly be used in the ambulatory surgery unit. Zeig and Herschman raise the provocative suggestion regarding the use of home testing kits. We are not confident that this is an acceptable alternative because test performance and interpretation would be unmonitored.
Steven Manley, M.D., Attending Anesthesiologist, Illinois Masonic Medical Center, Clinical Assistant Professor of Anesthesiology, University of Illinois College of Medicine.
Ninos J. Joseph, B.S., Research Associate, Illinois Masonic Medical Center.
M. Ramez Salem, M.D., Chair, Illinois Masonic Medical Center, Clinical Professor of Anesthesiology, University of Illinois College of Medicine.
Harold J. Heyman, M.D., Attending Anesthesiologist, Illinois Masonic Medical Center, Clinical Associate Professor of Anesthesiology, University of Illinois College of Medicine.
Grace de Kelaita, Research Assistant, Department of Anesthesiology, Illinois Masonic Medical Center, 836 West Wellington Avenue, Chicago, Illinois 60657.