The safety and efficacy of tubal sterilization: an international overview

Quinacrine non-surgical female sterilization in Bangladesh

This study was undertaken to evaluate the efficacy, safety, and acceptability of transcervical applications of quinacrine along with other adjuvants, such as ampicillin and ibuprofen, for sterilization. The cohort consisted of 750 normal women who requested sterilization and volunteered for this method at the family planning clinic of a tertiary hospital and community clinics in Chittagong, Bangladesh. Several different protocols were followed from October 1989 to April 1999. Each woman received one or two insertions of 180 mg or 252 mg quinacrine with or without adjuvants including 55.5 mg ibuprofen or 125 mg ampicillin. Supplementary contraception was given in the form of combined oral contraceptive pills, barrier methods, or injection of depot medroxyprogesterone acetate for 3 months. Details of each protocol are described in the text. The gross pregnancy failure rate for insertion of 180 mg in 590 women was 3.9% compared to 1.9% for the 160 who received 252 mg. There were no serious complications, and side effects were transient. We conclude that quinacrine non-surgical sterilization is a safe, acceptably effective method when two insertions of 252 mg quinacrine with medroxyprogesterone injection as a supplement is used.

Tubal sterilization: focus on the U.S. experience

OBJECTIVE

To review the frequency, effectiveness, and clinical sequelae of tubal sterilization with a focus on the U.S. experience.

DESIGN

A review of U.S. health care statistics and English-language literature using a MEDLINE search, bibliographies of key references, and U.S. government publications.

PATIENT(S)

Women seeking tubal sterilization.

INTERVENTION

Tubal sterilization.

MAIN OUTCOME MEASURE(S)

Effectiveness and long-term risks and benefits.

RESULT(S)

Half of the 700,000 annual bilateral tubal sterilizations (TS) are performed postpartum and half as ambulatory interval procedures. Eleven million U.S. women 15-44 years of age rely on TS for contraception. Failure rates vary by method with one third or more resulting in ectopic pregnancy. Reversal is most successful after use of methods that destroy the least tube. Evidence of menstrual or hormonal disturbance after TS is weak, although some studies find higher rates of hysterectomy among previously sterilized women. Decreased risk of subsequent ovarian cancer has been observed among sterilized women.

CONCLUSION(S)

Tubal sterilization is highly effective and safe. Failures, although uncommon, occur at higher rates than previously appreciated. Evidence for hormonal or menstrual changes due to TS is weak. Tubal sterilization is associated with decreased risk of ovarian cancer.

31,781 cases of non-surgical female sterilisation with quinacrine pellets in Vietnam

The quinacrine method of non-surgical female sterilisation involves transcervical intrauterine insertion of 252 mg quinacrine as pellets during the proliferative phase of the menstrual cycle; the drug causes inflammation and fibrosis of the proximal fallopian tube. We have carried out a field trial of 31,781 cases in twenty-four provinces of Vietnam from Jan 2, 1989, until October, 1992. There were 818 pregnancies after the procedure, of which 80 were carried to term. Some women received only one dose of quinacrine; the majority received two doses with an interval of one month. Cumulative life-table pregnancy rates per 100 women at 1 year (for studies of at least 50 cases followed for 12 months) were 2.63 (SE 0.17) among 9461 women who received two doses and 5.15 (0.48) among 2225 who received only one dose. Failure rates (pregnancies) were strongly affected by the skill of the doctor or midwife. There were no deaths and only 8 serious complications were reported (0.03%); by contrast, in a similar series of women undergoing surgical sterilisation, 30 deaths and between 540 and 1812 serious complications would be expected. All reported side-effects were minor and of short duration. There were 19 ectopic pregnancies and the incidence was 0.89 per 1000 woman-years of use. There was one birth defect (anencephaly), in a fetus conceived 2.5 months after quinacrine insertions; however, we believe it is not related to the procedure. An estimated 242 maternal deaths will be averted by these 31,781 sterilizations. This method is safe and acceptably effective for female sterilisation.

Sterilization needs in the 1990s: the case for quinacrine nonsurgical female sterilization

Much evidence suggests that demand for sterilization is a function of supply of surgical sterilization services in less-developed countries. If such services were greatly expanded, the number of procedures performed would grow dramatically. While the prevalence of sterilization is estimated to increase from 23.5% to 28.8% of married women of reproductive age in the 1990s, there will actually be 106,432,000 more couples of reproductive age at the end of this decade than at its beginning who use either no method or a far less effective method with much lower continuation rates than sterilization--nearly a 20% increase. To achieve a mean sterilization prevalence of 47% of married women of reproductive age in the less-developed world, as now seen in the Republic of Korea and Puerto Rico, the number of sterilizations would need to be more than double the current projection for the 1990s: 328,429,000 rather than 159,000,000. The quinacrine pellet method for nonsurgical female sterilization offers hope that this enormous shortfall in sterilization services can be overcome in this decade.

Analysis of sterilization failure in Brazil

A study of 13,423 female sterilization procedures performed from 1981 to 1984 in Rio de Janeiro, Brazil, was conducted to determine the level of sterilization failure. The lifetable cumulative failure rate was 0.54 at 12 months per 100 initial sterilizations, increasing to 1.04 at 48 months. Results of a multivariate analysis indicated that the differential in the rate of sterilization failure by woman's age at sterilization at 12 and 24 months was statistically highly significant. However, the differential rate of failure by the other variables such as whether sterilized during a training period, number of sterilizations per surgeon per day, and year of sterilization were not statistically significant.

Four-year follow-up of electrocoagulation and tubal ring sterilizations in Costa Rica

An earlier analysis of 299 laparoscopic sterilizations comparing electrocoagulation and tubal ring occlusion techniques found no significant differences in rates of surgical complications. The risk of potentially serious complications, such as bowel/bladder burns, was considered higher with electrocoagulation be the preferred technique. Examination of the same women through 48 months poststerilization between the two techniques with respect to the incidence of gynecologic surgery performed subsequent to sterilization. Gynecologic abnormalities were similar for women in both groups. The rate of pregnancy was higher for tubal rings than for electrocoagulation (2.1 compared to 0.7 at 48 months) but this difference was not statistically significant.

Comparative risks and costs of male and female sterilization

Couples who are considering elective sterilization should compare the risks and costs of male and female sterilization procedures as part of the decision process. Morbidity, mortality, failure rates, and short-term costs associated with male and female sterilization procedures were estimated from data available in previous case series. Male sterilization procedures were found to have zero attributable deaths and significantly less major complications when compared to female sterilization procedures. No less than 14 deaths a year can be attributed to female sterilization procedures in the US. Male and female sterilization procedures have efficacy rates that are not significantly different from each other. The short-term costs of female sterilization are 3.0 to 4.1 times that of vasectomy.

Statistical comparison of Pearl rates

Statistical procedures for hypothesis testing and interval estimation of the difference in a pair of Pearl rates are presented. The p value of the test statistic is evaluated exactly from the binomial distribution or approximately from the standard normal distribution. Interval estimation is provided by the test-based method of Miettinen. The procedures are applied to two published data sets, and the conceptual link between the Pearl rate and life-table methods is discussed.