Abnormal Uterine Bleeding Associated with Hormonal Contraception -- American Family Physician

MAY 15, 2002 / VOLUME 65, NUMBER 10 www.aafp.org/afp AMERICAN FAMILY PHYSICIAN 2073

in the United States each year are related to

the misuse or discontinuation of OCPs.

Mechanisms of Action

COMBINATION ORAL CONTRACEPTIVE PILLS

Most combination OCPs contain ethinyl

estradiol (20 to 50 mcg) and a synthetic prog-

estin (e.g., norgestrel, norethindrone, levo-

norgestrel, desogestrel). These pills inhibit

ovulation in most women. They also induce

thickening of the cervical mucus, which im-

pedes transport of sperm to the uterus. With

perfect use, only 0.1 percent of women

become pregnant within the first year of

using a combination OCP.

PROGESTIN-ONLY CONTRACEPTIVE METHODS

Progestin-Only Pills. Birth control pills that

contain only progestin, often called “mini-

pills,” inhibit ovulation in about 50 percent of

women.

Their primary mechanism of action

is thickening of the cervical mucus. This effect

occurs within hours of taking a progestin-only

pill and peaks about four hours after the pill is

taken. However, the cervical mucus remains

bnormal uterine bleeding is a

common side effect of all forms

of hormonal contraception.

Although this bleeding is rarely

dangerous, many women find

it worrisome. In fact, women frequently dis-

continue hormonal contraception because of

irregular bleeding and other side effects.

1-3

One study

found that 32 percent of 1,657

women who started taking oral contraceptive

pills (OCPs) discontinued them within six

months; 46 percent of the discontinuations

were due to side effects.

Most women who discontinue hormonal

contraception do not use another contracep-

tive method and are therefore at high risk for

unintended pregnancy. An estimated one

third of the 3 million unintended pregnancies

Millions of women in the United States use some type of hormonal contraception: combina-

tion oral contraceptive pills (OCPs), progestin-only pills, medroxyprogesterone acetate injec-

tions, or subdermal levonorgestrel implants. Abnormal uterine bleeding is a common but

rarely dangerous side effect of hormonal contraception. It is, however, a major cause for the

discontinuation of hormonal contraception and the resultant occurrence of unplanned preg-

nancy. The evaluation of abnormal uterine bleeding in women who are using hormonal con-

traception includes an assessment of compliance, a thorough history and complete physical

examination to exclude organic causes of bleeding, and a targeted laboratory evaluation.

Pregnancy and the misuse of OCPs are frequent causes of abnormal uterine bleeding. Bleed-

ing is common during the first three months of OCP use; counseling and reassurance are ade-

quate during this time period. If bleeding persists beyond three months, it can be treated

with supplemental estrogen and/or a nonsteroidal anti-inflammatory drug (NSAID). Other

options are to change to an OCP with a higher estrogen content or to a different formulation

(i.e., a low-dose OCP containing a different progestin). Management strategies for women

with abnormal uterine bleeding who are using progestin-only contraceptive methods include

counseling and reassurance, as well as the administration of supplemental estrogen and/or

2002 American Academy of Family Physicians.)

Most women who discontinue hormonal contraception do

not use another contraceptive method and are therefore at

high risk for unintended pregnancy.

Abnormal Uterine Bleeding Associated

with Hormonal Contraception

SARINA SCHRAGER, M.D., University of Wisconsin Medical School, Madison, Wisconsin

O A patient infor-

mation handout on

birth control pills

and bleeding, writ-

ten by the author of

this article, is pro-

vided on page 2083.

thickened for only about 20 hours, which

makes the progestin-only pill less effective dur-

ing the last few hours before the next dose.

Progestin-only pills are useful in women

who cannot use OCPs that contain estrogen

or who do not want long-term contraception.

Breastfeeding women often use this form of

contraception.

With perfect use, only 0.5 per-

cent of women become pregnant within the

first year of using progestin-only OCPs.

Contraceptive Injections. Depot medroxy-

progesterone acetate (Depo-Provera) is an

intramuscular progestin injection (150 mg) that

provides approximately 14 weeks of adequate

contraceptive levels. Because of the high dose of

progestin, ovulation is inhibited in most

women.

With perfect use, only 0.3 percent of

women become pregnant within the first year of

using medroxyprogesterone injections.

Contraceptive Implants. Levonorgestrel

(Norplant) consists of six subdermal im-

plants that release a constant low level of the

progestin levonorgestrel over a five-year

period: 0.05 to 0.08 mg per day for the first

year and 0.03 mg per day for the remaining

four years. Ovulation is inhibited in most

women.

The implants also induce a thick-

ened cervical mucus and cause endometrial

changes that impede implantation. With per-

fect use, only 0.09 percent of women become

pregnant within the first year of using levo-

norgestrel implants.

All forms of hormonal contraception are

listed as pregnancy category X.

Terminology

The term “breakthrough bleeding” refers to

bleeding at an unexpected time during the

menstrual cycle, with the bleeding sufficient to

require use of a tampon or sanitary napkin.

“Spotting” refers to unexpected bleeding that

does not require any protection.

The term

“intermenstrual bleeding” simply relates to

the timing of abnormal bleeding, not its

amount. In many studies, interchangeable use

of these terms makes interpretation of

research findings difficult. In this article, the

term “abnormal uterine bleeding” is defined

as any bleeding that occurs at an unpre-

dictable time during the menstrual cycle.

Evaluation of Abnormal Uterine Bleeding

Although hormonal contraception is a com-

mon cause of abnormal uterine bleeding, other

causes also need to be considered (Table 1).The

evaluation of women who have abnormal uter-

ine bleeding and are using hormonal contra-

ception is summarized in Ta b le 2.

Compliance with hormonal contraception

should be assessed, and a menstrual calendar

should be reviewed to determine the pattern

of bleeding. Often, women are unaware of the

impact missed contraceptive pills can have on

their menstrual cycle. Even one missed pill

can cause breakthrough bleeding.

Clinical clues from the history and physical

examination can guide laboratory testing

(Table 3).Ifthe cervix is inflamed, samples

should be obtained for Chlamydia trachomatis

and Neisseria gonorrhoeae testing. If the uterus

is enlarged, a pregnancy test is indicated. If

2074

AMERICAN FAMILY PHYSICIAN www.aafp.org/afp VOLUME 65, NUMBER 10 / MAY 15, 2002

Progestin-only pills are especially useful in women who are

breastfeeding.

TABLE 1

Differential Diagnosis

of Abnormal Uterine Bleeding

Hormonal contraception

Intrauterine or ectopic pregnancy

Endometrial or cervical polyp

Endocrine abnormalities (hypothyroidism,

hyperthyroidism, hyperprolactinemia)

Cervicitis

Cervical dysplasia or carcinoma

Bleeding disorders

Liver or renal failure

Endometrial hyperplasia or carcinoma

Uterine leiomyomas

heavy bleeding is present, testing for anemia is

appropriate. Pelvic ultrasonography or

endometrial biopsy can exclude endometrial

abnormalities and uterine leiomyomas.

Combination Oral Contraceptive Pills

FACTORS THAT INFLUENCE ABNORMAL

UTERINE BLEEDING

In the past 20 years, the estrogen dose in

OCPs has decreased from more than 150 mcg

of ethinyl estradiol to 35 mcg or less. The

most common low-dose OCPs now contain

no more than 35 mcg of ethinyl estradiol.

Although several OCPs contain 50 mcg of

ethinyl estradiol, these pills are not used rou-

tinely because they are associated with an

increased risk of thromboembolic disease.

Because the lower doses of estrogen in

OCPs are insufficient to sustain endometrial

integrity, abnormal uterine bleeding has

become more common.

The most frequent

cause of breakthrough bleeding with OCPs is

progestin-induced decidualization and

endometrial atrophy, which result in men-

strual breakdown and irregular bleeding.

As the dose and potency of both estrogen

and progestin increase, the incidence of break-

through bleeding decreases.

11,12

How the dif-

ferent pill formulations containing low doses

of estrogen (less than 50 mcg of ethinyl estra-

diol) differ in their propensity to cause abnor-

mal uterine bleeding remains unclear.

In addition to problems with terminology,

variations in formulations have make direct

comparisons of OCPs difficult. Some pills are

monophasic and have consistent doses of both

ethinyl estradiol and progestin throughout the

21-day cycle. Some are triphasic and have

three different doses of estrogen and prog-

estin. The type of progestin also varies.

Three

recent studies

14-16

have shown similar inci-

dences of abnormal uterine bleeding with

monophasic and triphasic low-dose pills. A

fourth study

found that women taking a

triphasic pill had significantly less abnormal

bleeding than those taking a monophasic pill.

Abnormal uterine bleeding patterns can be

related to the ratio of estrogen to progestin.

In addition, every woman metabolizes hor-

mones differently.

These factors further

complicate the interpretation of study results.

FACTORS THAT INCREASE BLEEDING

None of the studies comparing bleeding

with different OCPs controlled for cigarette

smoking or C. trachomatis infection. Yet both

Abnormal Uterine Bleeding

MAY 15, 2002 / VOLUME 65, NUMBER 10 www.aafp.org/afp AMERICAN FAMILY PHYSICIAN 2075

TABLE 2

Evaluation of Abnormal Uterine Bleeding

in Women Using Hormonal Contraception

History and physical examination (including pelvic

examination)

Assessment of compliance with hormonal

contraception

Review of menstrual calendar

Pregnancy test

Papanicolaou’s test (to evaluate for cervical

pathology)

Appropriate laboratory tests (e.g., hemoglobin level,

thyroid-stimulating hormone level, prolactin level)

Tests for Chlamydia trachomatis and Neisseria

gonorrhoeae, if indicated

Pelvic ultrasonography, if indicated

Endometrial biopsy, if indicated

TABLE 3

Clinical Clues and Appropriate Laboratory Tests

for Abnormal Uterine Bleeding

Clinical clue Appropriate tests

Fatigue or weight gain Thyroid-stimulating hormone level

Galactorrhea Prolactin level

Cervicitis Papanicolaou’s test, Chlamydia trachomatis

and Neisseria gonorrhoeae tests

Enlarged uterus Pregnancy test, pelvic ultrasonography

Edema Evaluation of kidney function

Nausea, fatigue, missed Pregnancy test

pills

Heavy bleeding Coagulation profile, evaluation of endometrium

for hyperplasia or carcinoma

of these factors have been associated with

increased abnormal uterine bleeding in

women taking combination OCPs.

10,19,20

One study

showed that smokers were

47 percent more likely to experience abnor-

mal uterine bleeding than nonsmokers. Ciga-

rette smoking is associated with antiestro-

genic effects and may lower estrogen levels.

Another study

found that 29 percent of

2076

AMERICAN FAMILY PHYSICIAN www.aafp.org/afp VOLUME 65, NUMBER 10 / MAY 15, 2002

Abnormal Uterine Bleeding with Combination OCPs

After 3 months

of OCP use

Provide counseling

and reassurance.

If abnormalities are found,

provide appropriate

management.

Provide pregnancy

test and counseling.

First 3 months

of OCP use

History and physical examination; selected

laboratory tests (see Table 2)

Treat with ibuprofen (e.g., Advil,

Motrin), 800 mg 3 times

daily for 1 to 2 weeks or

until bleeding stops.

Change pill formulation (i.e., pill with higher

estrogen dose or different progestin).

Treat with supplemental

estrogen for 1 to 2

weeks or until bleeding

stops (see Table 4).

No improvement

Repeat treatment or discontinue combination OCPs

and suggest another form of contraception.

No improvement

Assess compliance

with OCP use.

No missed pillsMissed pills

FIGURE 1. Algorithm for the suggested management of abnormal uterine bleeding in women

using combination oral contraceptive pills (OCPs).

Information from references 18 and 20.

The Author

SARINA SCHRAGER, M.D., is assistant professor in the Department of Family Medicine

at the University of Wisconsin Medical School, Madison. She received her medical

degree from the University of Illinois at Chicago College of Medicine. Dr. Schrager

completed a family practice residency and a primary care women’s health fellowship

at MacNeal Memorial Hospital, Berwyn, Ill.

Address correspondence to Sarina Schrager, M.D., Department of Family Medicine,

University of Wisconsin Medical School, 777 S. Mills St., Madison, WI 53715 (e-mail:

[email protected]). Reprints are not available from the author.

women taking OCPs who experienced new

abnormal uterine bleeding had asymp-

tomatic chlamydial cervicitis or chronic

endometritis.

Management of Abnormal

Uterine Bleeding

COMBINATION ORAL CONTRACEPTIVE PILLS

As many as 30 percent of women experi-

ence abnormal uterine bleeding in the first

month that they use combination OCPs.

The incidence of bleeding decreases signifi-

cantly by the third month of use.

The management of abnormal uterine

bleeding in women who are taking combina-

tion OCPs begins with counseling about

compliance (Figure 1).

18,21

If the bleeding oc-

curs within the first three months of pill use

and compliance is good, supportive counsel-

ing is all that is needed.

After three months and if other causes of

bleeding (including pregnancy) are excluded,

treatment with supplemental estrogen

and/or a nonsteroidal anti-inflammatory

drug (NSAID) often alleviates the bleeding

(Table 4).

18,21

Adding extra estrogen while

maintaining the same dose of progestin in-

creases endometrial thickness, thereby stabi-

lizing the endometrium and blood vessels.

If the bleeding persists despite the use of

supplemental estrogen and/or an NSAID and

compliance is good, another low-dose OCP

containing a different progestin could be

tried. However, only minimal evidence sug-

gests that switching OCPs further reduces

bleeding.

18,22

Changing to a 50-mcg OCP increases the

dose of both estrogen and progestin. Side

effects, including nausea and breast tender-

ness, may increase. Adding a second OCP on

a daily basis is not a good option, because

this doubles the estrogen and progestin

doses, further increasing side effects. When

OCPs are doubled, the progestin component

tends to dominate; therefore, endometrial

atrophy and subsequent irregular bleeding

increase.

PROGESTIN-

ONLY CONTRACEPTIVE METHODS

Abnormal uterine bleeding in women who

are using progestin-only contraceptive meth-

ods is treated with supplemental estrogen to

stabilize the endometrium and/or an NSAID

to decrease endogenous prostaglandins while

bleeding is present (Figure 2).

7,8,23-25

Progestin-Only Pills. Menstrual patterns are

affected in more than one half of women who

use progestin-only pills for hormonal contra-

ception. Menstrual changes include irregular

bleeding, short cycles (caused by an inade-

quate luteal phase), and amenorrhea.

7,8

Because progestin-only pills are short act-

ing, timely ingestion is important. Women

should be counseled to take their progestin-

only pill at the same time every day. Variances

of as little as two to three hours can cause

abnormal uterine bleeding.

Use of an NSAID and/or supplemental estro-

Abnormal Uterine Bleeding

MAY 15, 2002 / VOLUME 65, NUMBER 10 www.aafp.org/afp AMERICAN FAMILY PHYSICIAN 2077

TABLE 4

Treatment Options for Abnormal Uterine Bleeding

in Women Using Hormonal Contraception

Treatment Dosage

Nonsteroidal anti-inflammatory 800 mg three times daily for 1 to 2 weeks

drug such as ibuprofen (e.g., or until bleeding stops

Advil, Motrin)

Supplemental estrogen

Conjugated equine estrogens 0.625 to 1.25 mg per day for 1 to 2 weeks

(Premarin)

Ethinyl estradiol (Estinyl) 20 mcg per day for 1 to 2 weeks

Estradiol (Estrase) 0.5 to 1 mg per day for 1 to 2 weeks

Information from references 18 and 21.

Supplemental estrogen and/or a nonsteroidal anti-inflam-

matory drug (e.g., ibuprofen) can correct abnormal uterine

bleeding in women who are using progestin-only methods

of contraception.

2078 AMERICAN FAMILY PHYSICIAN www.aafp.org/afp VOLUME 65, NUMBER 10 / MAY 15, 2002

Abnormal Uterine Bleeding with Progestin-Only Contraception

Assess menstrual calendar

to evaluate bleeding pattern.

If abnormalities are found,

provide appropriate

management.

Provide counseling.

History and physical examination; selected

laboratory tests (see Table 2)

First-line therapy:

Treat with ibuprofen (e.g., Advil,

Motrin), 800 mg 3 times daily for

1 to 2 weeks or until bleeding stops.

Add estrogen for 1 to 2 weeks or

until bleeding stops (see Table 4).

Second-line therapy

Resolution of bleeding

Repeat treatments for further bleeding episodes.

Bleeding persists.

Add low-dose combination

OCP for 2 to 3 months.

Add low-dose combination OCP

for 2 to 3 months.

Remove implants if bleeding persists.

Levonorgestrel

implants (Norplant)

Medroxyprogesterone acetate injections

Change to

combination OCP.

Progestin-only pills

Assess compliance:

Are progestin-only pills being taken

at the same time every day?

Are medroxyprogesterone acetate

injections (Depo-Provera) being

received every 3 months?

CompliantNoncompliant

FIGURE 2. Algorithm for the suggested management of abnormal uterine bleeding in women

using progestin-only contraceptive methods. (OCP = oral contraceptive pill)

Information from references 7, 8, and 23 through 25.

gen can be helpful.

Occasionally, it may be

necessary to change to a combination OCP.

Contraceptive Injections and Implants. Ab-

normal bleeding is common in women using

long-acting progestin-only contraceptive

methods. The absolute dose of progestin is

higher in a medroxyprogesterone injection

than in levonorgestrel implants. Episodes of

unpredictable bleeding occur during the first

year in 70 percent of women who use contra-

ceptive injections and in up to 80 percent of

women who use contraceptive implants.

After one year of using medroxyproges-

terone injections, up to 50 percent of women

experience amenorrhea. With increasing

duration of use, the incidence of amenorrhea

may reach 80 percent. Women who use levo-

norgestrel implants less commonly have

amenorrhea, and bleeding abnormalities tend

to decrease after the first year of use.

Approx-

imately one third of women who use contra-

ceptive implants continue to ovulate and have

regular cycles.

A study

of almost 500 women who used

levonorgestrel implants noted bleeding side

effects as a major reason for discontinuation.

In this study, discontinuation was more closely

related to increased or decreased bleeding than

to irregular, unpredictable bleeding.

In one survey,

the duration of treatment

for bleeding related to the use of progestin-

only contraceptives ranged from seven to 10

days to “frequent.” Usually, treatment was ini-

tiated because patients were annoyed by the

abnormal bleeding or worried about becom-

ing pregnant.

One study

compared levonorgestrel (0.03

mg twice daily), ethinyl estradiol (0.05 mcg

per day), and ibuprofen (800 mg three times

daily) with placebo for the treatment of pro-

longed bleeding in women who were using

contraceptive implants. Women in all three

treatment groups had fewer bleeding days

than those in the placebo group. Ethinyl estra-

diol was the most effective treatment, followed

by ibuprofen and then levonorgestrel.

A more recent study

evaluated the use of

50 mcg of ethinyl estradiol, an OCP contain-

ing 50 mcg of ethinyl estradiol and 250 mcg of

levonorgestrel, and placebo in women who

had abnormal uterine bleeding while using

contraceptive implants. The combination

OCP was more effective than ethinyl estradiol

alone, which was more effective than placebo.

Although the studies

3,23,24

were in women

using levonorgestrel implants, the results can

be extended to those using medroxyproges-

terone injections and progestin-only pills,

because the mechanism of abnormal bleed-

ing is the same. Some experts recommend

giving a second medroxyprogesterone injec-

tion less than three months after the first

injection to induce amenorrhea sooner.

However, another study

concluded that an

early second injection does not change men-

strual patterns.

ROLE OF COUNSELING

Counseling given before any method of

hormonal contraception is initiated has been

shown to improve compliance and continua-

tion of that method. Counseling should

address the possible side effects of each con-

traceptive method and include a discussion of

how abnormal uterine bleeding can affect the

woman’s life.

A 1988 study

of more than 5,000 women

using four different types of hormonal con-

traceptives (combination OCPs, progestin-

only pills, medroxyprogesterone injections,

and an estrogen-releasing vaginal ring not yet

available in the United States) found that the

subjective experience of bleeding was more

important than the actual bleeding pattern.

Thus, women were able to tolerate profound

changes in bleeding patterns if they were

counseled about possible bleeding first.

Another study

also showed increased con-

tinuation rates with medroxyprogesterone

injections in women who received pretreat-

ment counseling about side effects.

The author indicates that she does not have any con-

flicts of interest. Sources of funding: none reported.

Abnormal Uterine Bleeding

MAY 15, 2002 / VOLUME 65, NUMBER 10 www.aafp.org/afp AMERICAN FAMILY PHYSICIAN 2079

Abnormal Uterine Bleeding

REFERENCES

1. Sangi-Haghpeykar H, Poindexter AN 3d, Bateman

L, Ditmore JR. Experiences of injectable contracep-

tive users in an urban setting. Obstet Gynecol

1996;88:227-33.

2. Paul C, Skegg DC, Williams S. Depot medroxypro-

gesterone acetate. Patterns of use and reasons for

discontinuation. Contraception 1997;56:209-14.

3. Diaz J, Faundes A, Olmos P, Diaz M. Bleeding com-

plaints during the first year of Norplant implants

use and their impact on removal rate. Contracep-

tion 1996;53:91-5.

4. Rosenberg MJ, Waugh MS. Oral contraceptive dis-

continuation: a prospective evaluation of fre-

quency and reasons. Am J Obstet Gynecol 1998;

179(3 pt 1):577-82.

5. Rosenberg MJ, Waugh MS, Long S. Unintended

pregnancies and use, misuse and discontinuation of

oral contraceptives. J Reprod Med 1995;40:355-60.

6. Trussell J. Contraceptive efficacy. In: Hatcher RA.

Contraceptive technology. 17th ed. rev. New York:

Ardent Media, 1998:779-99.

7. Progestin-only oral contraceptives. In: Wallach M,

Grimes DA, Chaney EJ, et al., eds. Modern oral

contraception: updates from The Contraception

Report. Totowa, N.J.: Emron, 2000:242-50.

8. Hatcher RA. Depo-Provera, Norplant, and prog-

estin-only pills. In: Hatcher RA. Contraceptive tech-

nology. 17th ed. rev. New York: Ardent Media,

1998:467-509.

9. Archer DF. Management of bleeding in women

using subdermal implants. Contemp Ob Gyn 1995;

40(7):11-25.

10. Thorneycroft IH. Cycle control with oral contracep-

tives: a review of the literature. Am J Obstet Gyne-

col 1999;180(2 pt 2):280-7.

11. Kaunitz AM. Oral contraceptive estrogen dose con-

siderations. Contraception 1998;58(3 suppl):S15-21.

12. Sulak P, Lippman J, Siu C, Massaro J, Godwin A.

Clinical comparison of triphasic norgestimate/

35 micrograms ethinyl estradiol and monophasic

norethindrone acetate/20 micrograms ethinyl

estradiol. Contraception 1999;59:161-6.

13. DelConte A, Loffer F, Grubb GS. Cycle control with

oral contraceptives containing 20 micrograms of

ethinyl estradiol. A multicenter, randomized com-

parison of levonorgestrel/ethinyl estradiol (100

micrograms/20 micrograms) and northethin-

drone/ethinyl estradiol (1000 micrograms/20

micrograms). Contraception 1999;59:187-93.

14. Rosenberg MJ, Meyers A, Roy V. Efficacy, cycle con-

trol, and side effects of low- and lower-dose oral

contraceptives: a randomized trial of 20 micro-

grams and 35 micrograms estrogen preparations.

Contraception 1999;60:321-9.

15. Chavez A, DelConte A. A comparison of cycle con-

trol with monophasic levonorgestrel/ethinylestra-

diol 100 micrograms/20 micrograms versus tripha-

sic northethindrone/ethinylestradiol 500-750-1000

micrograms/35 micrograms: a multicenter, ran-

domized, open-label study. Eur J Contracept Re-

prod Health Care 1999;4(2):75-83.

16. Reisman H, Martin D, Gast MJ. A multicenter ran-

domized comparison of cycle control and labora-

tory findings with oral contraceptive agents con-

taining 100 microg levonorgestrel with 20 microg

ethinyl estradiol or triphasic northethindrone with

ethinyl estradiol. Am J Obstet Gynecol 1999;181(5

pt 2):45-52.

17. Stubblefield PG. Menstrual impact of contraception.

Am J Obstet Gynecol 1994;170(5 pt 2):1513-22.

18. Approach to common side effects. In: Wallach M,

Grimes DA, Chaney EJ, et al., eds. Modern oral

contraception: updates from The Contraception

Report. Totowa, N.J.: Emron, 2000:70-6.

19. Rosenberg MJ, Waugh MS, Stevens CM. Smoking

and cycle control among oral contraceptive users.

Am J Obstet Gynecol 1996;174:628-32.

20. Krettek JE, Arkin SI, Chaisilwattana P, Monif GR.

Chlamydia trachomatis in patients who used oral

contraceptives and had intermenstrual spotting.

Obstet Gynecol 1993;81(5 pt 1):728-31.

21. Hatcher RA, Guillebaud J. The pill: combined oral

contraceptives. In: Hatcher RA. Contraceptive tech-

nology. 17th ed. rev. New York: Ardent Media,

1998:405-66.

22. Cerel-Suhl SL, Yeager BF. Update on oral contra-

ceptive pills. Am Fam Physician 1999;60:2073-84.

23. Diaz S, Croxatto HB, Pavez M, Belhadj H, Stern J,

Sivin I. Clinical assessment of treatments for pro-

longed bleeding in users of Norplant implants.

Contraception 1990;42:97-109.

24. Alvarez-Sanchez F, Brache V, Thevenin F, Cochon L,

Faundes A. Hormonal treatment for bleeding irreg-

ularities in Norplant implant users. Am J Obstet

Gynecol 1996;174:919-22.

25. Nelson AL. Counseling issues and management of

side effects for women using depot medroxypro-

gesterone acetate contraception. J Reprod Med

1996;41(5 suppl):391-400.

26. Nutley T, Dunson TR. Treatment of bleeding prob-

lems associated with progestin-only contraceptives:

survey results. Adv Contracept 1997;13:419-28.

27. Harel Z, Biro FM, Kollar LM. Depo-Provera in ado-

lescents: effects of early second injection or prior

oral contraception. J Adolesc Health 1995;16:379-

84.

28. Belsey EM. The association between vaginal bleed-

ing patterns and reasons for discontinuation of

contraceptive use. Contraception 1988;38:207-25.

29. Lei ZW, Wu SC, Garceau RJ, Jiang S, Yang QZ,

Wang WL, et al. Effect of pretreatment counseling

on discontinuation rates in Chinese women given

depo-medroxyprogesterone acetate for contracep-

tion. Contraception 1996;53:357-61.

2080 AMERICAN FAMILY PHYSICIAN www.aafp.org/afp VOLUME 65, NUMBER 10 / MAY 15, 2002