Sample Summaries

April 2006

The Impact of Size of the Adnexal Mass on the Accuracy of Frozen Section Diagnosis

Geomini PM, Zuurendonk LD, Bremer GL et al.
Gynecol Oncol 2005(Nov); 99(2):362-6

Key Points

Frozen section diagnosis of adnexal masses generally achieves a specificity of >99%; the sensitivity of frozen section varies
Frozen section diagnosis fails to detect cancer more often in tumors >10 cm; in the current study 11% of benign frozen sections in tumors >10 cm were diagnosed as malignant or borderline on final paraffin section
Frozen section is less likely to correctly diagnose malignancy in mucinous tumors: Frozen section correctly diagnosed malignancy in 96% of non-mucinous tumors, but only 73% of mucinous tumors
In cases of large adnexal masses, especially those of mucinous histology, additional frozen section slices may improve the accuracy of frozen section

Summary

Intraoperative frozen section is an essential tool in the assessment and treatment of the adnexal mass. The authors’ previous meta-analysis showed an overall sensitivity of 65-100% of detecting ovarian cancer, but a specificity of >99%. This retrospective study was designed to determine if the size of the adnexal mass was related to the accuracy of intra-operative frozen section.

The study included all women who had undergone frozen section evaluation of an adnexal mass over a three-year period in three teaching hospitals in the Netherlands. A total of 257 consecutive frozen section evaluations were included. The final pathology demonstrated benign disease in 55%, malignancy in 34%, and borderline tumors in 11%. Menopausal women were more likely than pre-menopausal women to have a final diagnosis which was malignant (38% vs. 26%).

All malignant frozen sections were confirmed to be malignant at paraffin section. Among those with benign frozen sections, there was a difference in size between tumors subsequently found to be benign, borderline, and malignant, with a mean diameter of 13.1, 17.2, and 19.6 cm respectively. In women with tumors less than 10 cm, one out of 50 benign frozen sections subsequently turned out to be a malignancy (i.e., there was only one out of 50 false negative results); in the group with tumors >10 cm, the 97 with benign frozen sections included six women with malignant tumors and 5 with borderline tumors (11 false negatives out of 97). Put another way, 1 out of 9 (11%) of women with benign frozen sections and masses >10 cm will ultimately have a malignancy on paraffin section.

Sensitivity and specificity were calculated first considering tumors with a final borderline diagnosis as benign, and then by counting this group as malignant. The sensitivity of frozen section was related to the histologic type of tumor. In mucinous tumors, the sensitivity was 73% and the specificity was 95%. In non-mucinous tumors, the sensitivity was 96% and the specificity was 95%. Overall, 12 patients out of 257 with a benign frozen section ultimately had a malignant or borderline lesion; 9 out of these 12 had mucinous tumors. Since mucinous cystadenomas tend to be large and heterogenous, frozen section was especially prone to miss malignancy in large mucinous tumors. CA125 levels were not useful, since CA125 is generally expressed in low levels by mucinous tumors.

It is not surprising that the sensitivity of frozen section in detecting malignancy declines with increasing size of the adnexal mass: Because intra-operative frozen section is expected to provide a rapid result, the pathologist must take limited samples for frozen section. Other authors have shown that more numerous slices improve the accuracy of frozen section diagnosis, though this was not evaluated in the present study. It has been suggested that at least one slice for every 10 cm of tumor should be recommended.

The results of this study will make surgeons circumspect about benign frozen section results in adnexal masses >10 cm, especially if they are mucinous.

Ovarian Remnant Syndrome

Magtibay PM, Nyholm JL, Hernandez JL et al

Am J Obstet Gynecol 2005(Dec); 193(6): 2062-6

Key Points:

Ovarian remnant syndrome (ORS) is characterized by pain or a pelvic mass containing ovarian tissue in a patient who has previously undergone bilateral salpingo-oophorectomy
Surgical treatment of ORS is an often radical procedure; the basic principles involve high ligation and resection of the gonadal vessels, wide excision of the pelvic sidewall peritoneum containing the ovarian remnant, and ureterolysis
Surrounding structures (bowel, bladder, ureter) frequently must be compromised (and of course repaired) to ensure that all ovarian tissue is removed
Complications of surgery for ORS are not infrequent; in this series the most common complication was adynamic ileus or partial small bowel obstruction
With proper surgical treatment, symptom resolution was seen in >90% of patients; only 1% of patients had definite residual ovarian tissue after the radical operation described by the authors

Summary

Ovarian remnant syndrome (ORS) refers to the occurrence of pain or pelvic mass associated with residual ovarian tissue in a patient previously treated with bilateral salpingo-oophorectomy (BSO). The recommended treatment for ORS is complete surgical excision by laparotomy or laparoscopy.

The current study was undertaken to examine the surgical management of patients with ORS. One hundred eighty six patients with pathologically confirmed ORS were examined over an 18 year period at the Mayo Clinic. All patients in the current series were treated by laparotomy. Similar surgical principles were followed in each case: High religation and resection of the gonadal vessels; bilateral excisional stripping of the pelvic peritoneum; and wide excision of the tissue surrounding the ovarian remnant.

The surgical technique described follows that described by Webb in 1989. The peritoneum is opened widely lateral and parallel to the ovarian vessels. The pararectal and paravesical spaces are dissected. The anterior division of the internal iliac is ligated. The ovarian vessels are ligated above the pelvic brim after identification of the ureter. The pelvic peritoneum between the bifurcation of the iliacs to the entrance of the ureter into the bladder is completely excised, as this is likely to contain ovarian tissue. If there are adhesions to the vaginal vault, bladder or intestine, segmental resection is carried out to ensure that no ovarian tissue remains.

Eighty-five percent of the patients studied underwent their initial oophorectomy by laparotomy, with the remainder undergoing BSO by laparoscopy or, less commonly, via a vaginal route. The initial indication for BSO was endometriosis in 57%, pelvic mass in 9%, and pelvic inflammatory disease in 7%. About a third of the patients studied (35%) had undergone previous surgery for treatment for ORS; some had been treated with Danazol, GnRH agonists, oral contraceptives, or radiation for their ORS symptoms. Presenting symptoms included constant pain (84%), pelvic mass (66%), dyspareunia (26%), cyclic pain (9%), cyclic bleeding (8%), dysuria (7%), and painful defecation (6%). Many of the patients had no symptoms of estrogen deprivation despite BSO; in those who underwent hormonal evaluation, most showed pre-menopausal levels of LH and FSH. A pelvic mass was demonstrated by ultrasound or CT in >90% of patients studied.

Serious complications were not infrequent. In 186 patients, enterotomy or colostomy occurred in 10 patients (3 intentionally); cystotomy occurred in 3 patients (1 intentionally), and ureteral injury occurred in 2 patients (1 intentionally). Blood transfusion was required in 12% of patients (22), and 3 patients returned to the OR within 30 days (2 for hemorrhage and 1 for a bowel injury). Other complications seen included pulmonary embolism (2), DVT (1), ileus (14), small bowel obstruction (1), wound infections (7), intra-abdominal abscess (2), sepsis (3), ureteral obstruction (3), and small bowel obstruction (1).

Histology most commonly demonstrated a corpus luteum (42%) or endometriosis (29%).

The mean follow-up for this study was 1.2 years. During this time, 12 patients were re-explored for ORS; only one had an ovarian remnant histologically identified. The recurrence rate with the surgical technique described was thus <1%.

Fundamentally, ORS results from the incomplete removal of ovarian tissue at BSO. Risk factors for ORS include endometriosis, pelvic inflammatory disease, multiple prior pelvic surgeries, and pelvic adhesive disease. Ovarian tissue adherent to the pelvic peritoneum may undergo neovascularization, remaining viable even after ligation of the ovarian vessels. For this reason, the authors advocate opening the peritoneum widely and ligating the ovarian vessels high, cephalad to the ovaries. This principle applies to BSO both by laparotomy and by laparoscopy.

The pre-operative evaluation of patients with suspected ORS should include evaluation of LH and FSH levels. Pre-menopausal levels strongly suggest, but are not necessary for, the diagnosis of ORS. Imaging by ultrasound, CT, or MRI may detect a mass. Clomiphene citrate may be used to stimulate and enlarge the ovarian remnant before imaging or surgery.

The ovarian remnant is often densely scarred near the angle of the vaginal vault. The authors emphasize that all remaining tissue should be removed, even if the bowel, bladder, ureter, or other organs must be compromised (and repaired) in doing so. A suitably experienced surgeon is a necessary prerequisite.

This is the largest published series of ORS cases at a single institution, treated with a uniform surgical approach. Adequate follow-up, however, was obtained in only 76% of patients. Most patients (91%) reported resolution of symptoms after definitive surgery.

The Impact of Size of the Adnexal Mass On the Accuracy of Frozen Section Diagnosis Ovarian Remnant Syndrome

July 2006

A New Era in Ovulation Induction

Holzer H, Casper R, Tulandi T
Fertil Steril 2006(Feb); 85(2):277-84

Key Points

Aromatase inhibitors such as letrozole inhibit the last step of estrogen synthesis, the conversion of androgens to estrogens
Aromatase inhibitors are useful in induction of ovulation and superovulation
Letrozole appears to be superior to clomiphene citrate (CC) for ovulation induction; unlike CC, aromatase inhibitors have no anti-estrogenic effects on the endometrium and cervical mucous
Letrozole in a dose of 5 mg/day from days 3 through 7 is effective
When used in conjunction with gonadotropins, letrozole reduces the gonadotropin requirement to achieve ovulation, thereby reducing the cost of treatment

Summary

Clomiphene citrate (CC) has been a cornerstone of ovulation induction and superovulation for more than 40 years. CC results in ovulation in most patients (60-90%) but pregnancy rates are disappointing (10-40%) and multiple pregnancy rates are 10-20%. This has been attributed to the anti-estrogenic effects of CC on the endometrium and cervical mucous. Gonadotropins are more effective than CC, but also more expensive, and are associated with hyperstimulation syndrome and a higher risk of multiple gestations. In view of the limitations of currently available agents for induction of ovulation, there is considerable enthusiasm for the newest available agents: Aromatase inhibitors.

Aromatase is an enzyme of the cytochrome P450 system which catalyzes the rate-limiting final step in production of estrogen (E), the aromatization of androgens to estrogens. The lack of estrogen results via feedback loop in an increase in FSH secretion from the anterior pituitary. Accumulation of unconverted androgens in the ovary in turn increases follicular sensitivity to FSH.

The current, third generation aromatase inhibitors letrozole and anastrazole decrease E production by over 97%. These agents are completely absorbed orally, and are cleared by the liver with a half life of about 45 hours. The main side effects observed with their use are GI disturbances, asthenia, hot flashes, headache, and back pain.

Studies of aromatase inhibitors have involved mostly women who failed treatment with CC, either because of resistance or due to a thinning of the endometrium. In this group, 70-80% ovulated with aromatase inhibitors; endometrial thickness of 7-9mm was observed, and pregnancy rates of 20-27% were achieved. Aromatase inhibitors appear to induce the growth of fewer dominant follicles, suggesting a lower multiple gestation rate. In head-to-head comparison with CC, aromatase-treated women had lower estradiol levels, fewer follicles, a thicker, better perfused endometrium, and higher pregnancy rates (16.7% vs 5.6%).

Aromatase inhibitors have also been studied as an adjunct to gonadotropin treatment. CC has been used in this capacity, with the hope of increasing the number of pre-ovulatory follicles and reducing the gonadotropin requirement; results have been discouraging. The addition of letrozole to gonadotropins does, indeed, decrease gonadotropin requirement and increases the number of pre-ovulatory follicles. Except in PCOS patients, however, an overall increase in pregnancy rates was not observed. In patients with a previous poor response to FSH, aromatase inhibitors decrease the amount of gonadotropins required for ovulation, and can thereby reduce the cost of treatment.

There is some evidence that use of aromatase inhibitors may increase the number of embryos obtained in women with breast cancer seeking to cryopreserve their eggs prior to chemotherapy.

A five day regimen of letrozole at 5 mg/day seems to be the optimal dose. Preliminary information suggests that a single 20 mg dose on day 3 may be an effective alternative.

Aromatase inhibitors are a valuable addition to available ovulation induction agents. They are easily administered orally, inexpensive, and have minor side effects. Letrozole can replace CC as the first line treatment for ovulation induction or superovulation at a dose of 5 mg/day from days 3 through 7. When added to gonadotropin regimens, letrozole reduces the gonadotropin requirement without a decrease in pregnancy rates.

[Editor’s note: Letrozole is commercially available as Femara®]

Failure of Uterine Fibroid Embolization

Huang JY, Kafy S, Dugas A, et al.
Fertil Steril 2006(Jan); 85(1):30-35

Key Points

In the current study, 9.4% of patients ultimately required hysterectomy or myomectomy after clinical failure of uterine fibroid embolization (UFE)
At six months, the mean reduction in uterine volume was 28.4% after UFE; the mean reduction in volume of the dominant fibroid was 39.7% after UFE
Clinical failure of UFE was seen more commonly in patients with prior myomectomy, and in those patients who achieved the the most uterine shrinkage after UFE
Patients who fail UFE are not well served by repeat UFE; in this study all three patients re-treated with UFE ultimately required hysterectomy
UFE was associated with the apparent new development of adhesions in 59% of patients in this study

Summary

About 40% of reproductive-aged women have fibroids, and fibroids account for up to 1/3 of all hysterectomies. UFE, introduced in 1995, has gained popularity as a safe and effective alternative to hysterectomy and myomectomy. Over 50,000 UFE procedures have been performed worldwide. After UFE, the mean reduction in fibroid volume is 50-60%, and improvement in menorrhagia and pressure symptoms is seen in 90% of patients. About 5% of patients suffer from perioperative morbidity. Little, however, is known regarding the long term efficacy of UFE. This study provides information on the longer-term outcomes of UFE, and to evaluate factors associated with clinical failure of UFE.

The study consisted of a retrospective review of 233 consecutive patients who underwent UFE between 1997 and 2004 at McGill. UFE was performed as an outpatient procedure by three experienced interventional radiologists, using polyvinyl alcohol particles. Patients underwent sonography at 3 months and 6 months after UFE, and examined annually thereafter by their gynecologists.

Failure of UFE was defined as persistent or recurrent pain, bleeding, or bulk symptoms requiring hysterectomy, myomectomy, or repeat UFE. "Short term failures" occurred within six months of initial embolization; "long term failures" occurred after six months.

Overall, 9.4% of patients failed UFE, i.e., ultimately required hysterectomy or myomectomy. The mean reduction in uterine volume was 28.4% at six months, and the mean reduction in volume of the dominant fibroid was 39.7% at six months.

Failure of UFE occurred more frequently (13% of the time) in women with a previous myomectomy (including hysteroscopic myomectomy) than in those without prior myomectomy (2.5%). Interestingly, women who failed UFE demonstrated greater uterine shrinkage at six months than those whose symptoms resolved with UFE (57.1% vs. 25.2%); the authors suggest that reduction of uterine or fibroid volumes do not correlate well with clinical improvement.

Of 22 UFE failures, three were initially treated with repeat UFE. All three subsequently required hysterectomy.

Of those who failed UFE, the average time to definitive surgical treatment was 13.2 months. Of those who ultimately underwent abdominal surgery, adhesions were seen in 58.8%; none of these patients had previously undergone an abdominal operation. In patients who ultimately underwent a hysterectomy following failed UFE, adenomyosis was seen in 25% of pathology specimens.

January 2007

ACOG Committee Opinion: Human Papillomavirus Vaccination

Obstet Gynecol 2006(Sep); 108(3 Pt 1): 699-705

Key Points:

HPV vaccination is recommended for females aged 9 to 26, regardless of past exposure
To facilitate vaccination of this age group, ACOG recommends that girls have their first reproductive health visit between ages 13 and 15
The quadrivalent vaccine [Gardasil^®] is 100% effective at preventing CIN 2 and CIN 3 caused by HPV types 16 and 18
HPV vaccination is contraindicated in pregnancy, but permissible in breastfeeding women
Physicians are central to the acceptance of the vaccine by parents/guardians; in one study, 80% would support vaccination based on physician recommendation

Summary:

Cervical cancer is the second largest worldwide cause of cancer mortality in women. Eighty percent of deaths occur in areas of limited resources and screening. Even in this country, many women do not have regular periodic screening: In 2003, 86% of insured women from 18-64 reported screening in the prior 3 years, while only 67% of uninsured women were screened.

The relationship between HPV and cervical cancer is well established. Of the over 100 genotypes of HPV that have been discovered, only 15 have been associated with cervical cancer. Types 16 and 18 cause about 70% of cervical cancers, and types 6 and 11 cause 90% of cases of genital warts.

The FDA recently licensed a quadrivalent vaccine aimed at genotypes 6, 11, 16, and 18. This vaccine, given at months 0, 2, and 6, protects against cervicovaginal and vulvar dysplasia and cancer, as well as genital warts. A second bivalent vaccine aimed at genotypes 16 and 18 is soon to be released. The quadrivalent vaccine is 100% effective in preventing CIN 2 and CIN 3 as well as condyloma associated with the specific genotypes it targets. In women who already have HPV, the vaccine does not offer protection from disease from the existing HPV type in the host; it does however offer protection from the other vaccine genotypes. Ideally, the vaccine should be given before exposure to HPV.

The ACOG Committee Recommendations are as follows:

Vaccination of Girls, Adolescents, and Young Women: ACOG recommends vaccination of all women between the ages of 9 and 26. Towards this end, ACOG recommends that the first adolescent reproductive health visit should occur between the ages of 13 and 15. HPV vaccination should be discussed and offered at this first visit.

Cervical Cytology Screening and HPV Testing: Current recommendations for cytology screening remain unchanged: Cervical cytology screening should begin approximately 3 years after onset of vaginal intercourse, or no later than age 21. Women under 30 should have annual screening. The vaccine does not replace cervical cytology screening, as it does not offer protection against all genotypes potentially causing warts, dysplasia, and cancer. HPV testing is NOT recommended prior to vaccination.

Vaccination of Sexually Active Women/Women with Previous CIN: Women who are already sexually active, or who have preexisting HPV infection, abnormal cytology, or genital warts can all receive the vaccine; the vaccine is protective against those vaccine genotypes with which they have not already been infected. Patients in this instance should be counseled that the vaccine may be less effective in preventing disease than in HPV-naïve women. Physicians must be careful not to instill a false sense of security in patients, and must continue cervical cytology screening according to current ACOG recommendations.

Vaccination is NOT treatment: The HPV vaccine is not intended to treat women with warts or dysplasia; women must be evaluated and treated as usual. Note that many early cytologic abnormalities regress, especially in young women, and can be followed conservatively.

Vaccination of Pregnant and Lactating Women: Although the vaccine is Category B, vaccination is not recommended in pregnancy. If vaccination should occur in pregnancy, the remainder of the vaccination series should be deferred, and the manufacturer’s pregnancy registry should be contacted [see below for contact information]. Lactating women can receive the vaccine.

Vaccination of Immunosuppressed Patients: Immunosuppression, as might be seen in HIV, is not a contraindication to vaccination, although immune response may be blunted.

Vaccination of Women Over 26 and Males: Research is underway concerning these groups; currently there are no recommendations for vaccination.

Other Methods of Preventing HPV Infection: Abstinence is the most effective way to prevent acquisition of STDs, including HPV. Limiting the number of sexual partners and using latex condoms reduces the risk of acquiring the virus.

Research & Advocacy: ACOG supports research on the need for booster vaccination, vaccination of women over 26, and vaccination for males. ACOG also encourages third-party payors and government agencies to assist in covering the cost of vaccination. ACOG encourages pharmaceutical companies to offer the vaccine at a discount to the Vaccines for Children (VFC) program; VFC provides free vaccines to children who are uninsured, underinsured, Medicaid-eligible, or Native American. Healthcare providers are encouraged to register to be providers for the VFC program at http://www.cdc.gov/nip/vfc/.

Educational Efforts: The quadrivalent vaccine is a major breakthrough, but widespread use depends on educational efforts by physicians. Patients—and parents—should be counseled about the prevalence of HPV and its potential consequences, as well as the benefits (and limitations) of the vaccine. In a recent study, 80% of parents/guardians would support vaccination of their daughters; physician recommendation is central to acceptance by patients and their parents.

Consent for HPV Vaccination: For children and adolescents under 18, consent of a parent is generally required for medical care. In all states, minors are allowed to consent for diagnosis and treatment of STDs; many state laws however do not mention vaccination. State medical societies can be helpful in determining the particular consent requirement for a given state and circumstance.

The quadrivalent vaccine is marketed under the name Gardasil^® by Merck. A bivalent vaccine against types 16 and 18 (Cervarix™; GlaxoSmithKline) is awaiting FDA approval as of January 2007. Should a patient inadvertently receive the vaccine in pregnancy, Merck’s pregnancy registry should be contacted at 1-800-986-8999.

Note that recommendations for initial screening have recently changed: Whereas ACOG recommends the “first adolescent reproductive health visit” between ages 13 and 15, cervical cytology is now deferred until 3 years after sexual debut or at age 21, whichever comes first. Previous recommendations were for cervical cytology at sexual debut or at age 18.

Major Congenital Malformations After First-Trimester Exposure to ACE Inhibitors

Cooper WO, Hernandez-Diaz S, Arbogast PG

N Engl J Med(Jun); 354(23): 2443-51;

ACE Inhibitors and Congenital Anomalies

Friedman JM

N Engl J Med 2006(Jun); 354(23):2498-509

Key Points

Use of ACE inhibitors in late pregnancy is associated with oligohydramnios, IUGR, hypocalvaria, renal dysplasia, fetal oliguria, and fetal death
Use of ACE inhibitors in the first trimester was previously assumed to be safe; the current study demonstrates a nearly 3-fold increase in congenital anomalies with first trimester exposure to ACE inhibitors
Use of ACE inhibitors in the first trimester is associated with cardiovascular, CNS, and renal abnormalities in the fetus
Patients exposed to ACE inhibitors in early pregnancy should be offered targeted sonography and fetal echocardiography
Since half of pregnancies are unplanned, women of childbearing age taking ACE inhibitors should be counseled about the risk of first trimester exposure

Summary:

In utero exposure to ACE inhibitors in the second and third trimesters of pregnancy is known to be associated with an ACE-inhibitor fetopathy characterized by oligohydramnios, IUGR, hypocalvaria, renal dysplasia, anuria, and fetal death. The spectrum of fetal effects associated with use of ACE inhibitors in late pregnancy was thought to be mediated by impairment of fetal renal function, with resultant anuria and oligohydramnios. ACE inhibitors have not been thought to have teratogenic effects in the first trimester, as urine production starts later in pregnancy and increases gradually.

This epidemiologic study was undertaken to investigate the safety of ACE inhibitors in early pregnancy. Data was obtained retrospectively using Medicaid data in Tennessee, birth and death certificates, hospital records, and pharmacy records. Women who used ACE inhibitors in the first trimester, but not in later pregnancy, were identified by pharmacy records of filled prescriptions within 90 days of the last menstrual period. Because diabetes is strongly associated with congenital abnormalities, and because women with diabetes often take ACE inhibitors, pharmacy databases were used to exclude women being treated for diabetes. Women being treated with other known teratogens (e.g., warfarin; anticonvulsants; others) were similarly excluded.

A total of 29,507 infants born between 1985 and 2000 were studied. The presence of congenital anomalies was determined by blinded review of vital records and hospital records of delivery and the infants’ first year of life. Infants born to women using ACE inhibitors in the first trimester only were compared to infants who used other antihypertensives in the first trimester, and to women who used no antihypertensives in pregnancy.

Of the 29,507 infants studied, 209 had been exposed to ACE inhibitors in the first trimester only, and 202 had been exposed to other antihypertensives in the first trimester only. No increased risk of congenital anomalies was observed with the use of other antihypertensives; women exposed to ACE inhibitors in the first trimester, however, had major congenital anomalies in 7.12% of births, or 2.71 times the rate of anomalies in women exposed to no antihypertensives. Overall, the exposed group had 3.72 times the rate of cardiovascular anomalies (9 out of 209 patients) and 4.39 times the risk of central nervous system anomalies (3 out of 209 patients). Initial analysis showed no significant increase in all other types of anomalies combined (6 out of 209 patients), but post hoc analysis demonstrated a 9-fold increase in renal anomalies, especially renal dysplasia.

Of the 18 infants born to mothers exposed to ACE inhibitors in the first trimester, 7 had more than one anomaly. Atrial septal defects and/or patent ductus arteriosus (after 36 weeks) predominated, and were seen in 8 infants. All were diagnosed immediately at birth or shortly thereafter.

The use of ACE inhibitors in women of childbearing age is increasing nationally, with 4.4% of women between 18-44 years old using ACE inhibitors in 2002. The current study demonstrates that the use of ACE inhibitors in the first trimester—previously thought to be safe—is in fact associated with an increased risk of congenital anomalies, especially cardiac, CNS, and renal. Use of ACE inhibitors in the first trimester of pregnancy should be avoided.

Despite 25 years of market experience with 10 different ACE inhibitors, remarkably little is known about the teratogenicity of this class of drugs. Until the current study, there was practically no published data on which to evaluate the safety of ACE inhibitors in pregnancy.

When drugs receive FDA approval, little is typically known about their teratogenicity. No active surveillance or further clinical studies of teratogenicity are subsequently required. Many women will nonetheless require medical treatment of conditions like hypertension in pregnancy, and many more will become pregnant inadvertently while taking medication. Physicians cannot adequately counsel patients and perform risk/benefit evaluations when risks to the fetus are unknown—as they are for more than 90% of the prescription drugs approved in the U.S. between 1980 and 2000.

In view of the results of the Cooper study, a discussion of risk is warranted in all reproductive-aged women taking ACE inhibitors. Patients who become pregnant while taking these medications should be switched to another antihypertensive medication as early as possible in pregnancy. If exposure occurs during the first trimester, detailed sonography and fetal echocardiography is indicated.

ACE inhibitors include captopril (Capoten®), enalapril (Vasotec®), lisinopril (Prinivil® and Zestril®), quinapril (Acupril®) and ramipril (Altace®), among others. The FDA acknowledged the findings of this study, but declined to revise product labeling for the class: ACE inhibitors remain category C in the first trimester of pregnancy, and category D in the second and third. Prescribing information for the class continues to recommend discontinuing ACE inhibitors as soon as possible if a patient becomes pregnant while taking them. .