|
Correction
to
Carson and Buster,
|
| |||||||||||||||||||||||||||||||||
Abnormal serum progesterone levels or serial values for the beta subunit of human chorionic gonadotropin (
-hCG) are not specific to ectopic pregnancy. According to the data of Stovall and Ling,2 only 24 percent of patients with progesterone levels below 5 ng per milliliter had ectopic pregnancies, and only 56 percent of the patients with such pregnancies had levels below 5 ng per milliliter. Although Figure 1 of Carson and Buster implies that patients with intermediate progesterone levels have either viable intrauterine pregnancies or ectopic pregnancies, this clearly omits failing intrauterine pregnancies, which account for over 70 percent of pregnancies in this group. In other reports,3 it seems that the best use of serum progesterone measurement has been to eliminate the need for sonography in the 57 percent of patients whose initial levels are 25 ng per milliliter or more. Sonography is readily available in most centers and may provide valuable data at the first visit or within 24 hours.
For the majority of centers in which rapid progesterone measurement is not available, we propose the following. (1) In patients in stable condition, transvaginal ultrasonography or a -hCG measurement of 50,000 mIU per milliliter or more (International Reference Preparation) will reveal a viable intrauterine pregnancy2. (2) In the absence of a clear diagnosis of an ectopic or intrauterine pregnancy by ultrasonography, serial observations and measurement of hCG are indicated. (3) If the -hCG level falls or rises by less than 50 percent in two days, the possibility of an ectopic pregnancy can be excluded in many patients by performing dilation and curettage, which will identify a failed intrauterine pregnancy. (4) If the diagnosis is still unclear, serial observations and aggressive follow-up are indicated.
Critical to early diagnosis of ectopic pregnancy is consideration of the diagnosis in any woman with vaginal bleeding or pain in the first trimester, careful instructions to the patient, and the use of modern sonographic techniques correlated with measurement of -hCG. The benefit of routinely adding serum progesterone measurement to the algorithm has yet to be established.
Jean Abbott, M.D.
University of Colorado School of Medicine
Denver, CO 80262
Jedd Roe, M.D.
Denver General Hospital
Denver, CO 80204
References
Brian Budenholzer, M.D.
Group Health Northwest
Spokane, WA 99206
References
-hCG measurements within the "discriminatory zone" -- that is, values above the level at which all intrauterine pregnancies would be diagnosed on transvaginal ultrasonography -- should be presumed to have ectopic pregnancies. Women whose values fall below the discriminatory zone remain the most difficult cohort with respect to diagnosis. In their study of vaginal ultrasonography, in which they used the value of 1000 mIU of -hCG per milliliter (International Reference Preparation) as the cutoff for the discriminatory zone, Cacciatore et al. found that 26 percent of women with indeterminate ultrasonograms (5 of 19) had ectopic pregnancies and, of these, 40 percent (2 of 5) had -hCG values of less than 1000 mIU per milliliter1.
Contrary to the algorithm presented, the possibility of ectopic pregnancy cannot be completely excluded when the serum progesterone level exceeds 25 ng per milliliter or the -hCG level is greater than 100,000 mIU per milliliter. Stovall et al. reported that 3.1 percent of ectopic pregnancies (5 of 161) were associated with progesterone levels over 25 ng per milliliter2. In our collective experience at San Francisco General Hospital and Boston City Hospital, 0.7 percent of ectopic pregnancies (2 of 297) involved -hCG levels greater than 100,000 mIU per milliliter.
Tom Scaletta, M.D.
Beth C. Kaplan, M.D.
University of California, San Francisco
San Francisco, CA 94110
References
Marc R. Salzberg, M.D.
Baystate Medical Center
Springfield, MA 01199
References
James W. Kendig, M.D.
Fred M. Howard, M.D.
JoAnn S. Janas, M.D.
Rochester General Hospital
Rochester, NY 14621
References
To the Editor: The point of Kendig et al. is well taken. We perhaps mistakenly assumed that readers would know that Rh-negative patients require RhoD immune globulin after an ectopic pregnancy. The American College of Obstetricians and Gynecologists has widely distributed this information in its technical bulletins.
Oops! We regret using "sensitivity" for "specificity," and thank Dr. Budenholzer for pointing out our error.
A single measurement of progesterone will identify patients at risk for ectopic pregnancy earlier than will screening with ultrasonography. Thus, the figure of 90 percent to which Scaletta and Kaplan refer is not applicable. They describe data from studies in which the ultrasonography was used as the instrument of selection -- not progesterone measurement. Indeed, they state that when the -hCG level is 1000 mIU per milliliter or less, ultrasonography gives indeterminate results and other techniques such as curettage are required. Of course, no test is perfect. In our initial study, only 5 of 1136 patients (0.4 percent) with progesterone levels above 25 ng per milliliter had ectopic pregnancies. A single serum progesterone measurement costs about $20; an ultrasonogram costs approximately 10 times more. The radioimmunoassay for progesterone is available in commercial kits and takes four hours to perform; there is no reason that it cannot be made available daily. It is simply a matter of using it with sufficient frequency to justify the cost.
Drs. Abbott and Roe misunderstood our point regarding the use of screening progesterone levels. The single progesterone measurement is a screening test that alerts the physician to whether a patient needs diagnostic testing; the measurement is not in itself a diagnostic test. Sonography is 10 times more expensive than progesterone measurement. Furthermore, an abnormally low progesterone level alerts the physician to the possibility of an ectopic pregnancy at a time in gestation when such a pregnancy would be undetectable by ultrasonography; this has enabled our institution to reduce the rate of rupture in ectopic pregnancy from 79.2 percent to 38.8 percent.1
Although Dr. Salzberg's points are true, our review was written for the general medical community, which has found heterotopic pregnancy to be a very rare event, occurring once in 30,000 pregnancies; therefore, establishing the presence of an intrauterine pregnancy by ultrasonography will correctly rule out an ectopic pregnancy 29,999 times in 30,000. A patient undergoing gonadotropin therapy or in vitro fertilization will probably be followed by her reproductive endocrinologist until the location of the pregnancy is defined.
Sandra A. Carson, M.D.
John E. Buster, M.D.
University of Tennessee
Memphis, TN 38163
References
| |||||||||||||||||||||||||||||||||
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
