{"paper_id":"02c9f607-257e-4746-903a-4e6e7132bb00","body_text":"EVIDENCED-BASED CLINICAL MEDICINE\nAbnormal Uterine Bleeding: A Management\nAlgorithm\nJohn W. Ely, MD, MSPH, Colleen M. Kennedy, MD, MS, Elizabeth C. Clark, MD, MPH,\nand Noelle C. Bowdler, MD\nAbnormal uterine bleeding is a common problem, and its management can be complex. Because of this\ncomplexity, concise guidelines have been difficult to develop. We constructed a concise but comprehen-\nsive algorithm for the management of abnormal uterine bleeding between menarche and menopause\nthat was based on a systematic review of the literature as well as the actual management of patients seen\nin a gynecology clinic. We started by drafting an algorithm that was based on a MEDLINE search for rel-\nevant reviews and original research. We compared this algorithm to the actual care provided to a ran-\ndom sample of 100 women with abnormal bleeding who were seen in a university gynecology clinic.\nDiscrepancies between the algorithm and actual care were discussed during audiotaped meetings\namong the 4 investigators (2 family physicians and 2 gynecologists). The audiotapes were used to revise\nthe algorithm. After 3 iterations of this process (total of 300 patients), we agreed on a final algorithm\nthat generally followed the practices we observed, while maintaining consistency with the evidence. In\nclinic, the gynecologists categorized the patient’s bleeding pattern into 1 of 4 types: irregular bleeding,\nheavy but regular bleeding (menorrhagia), severe acute bleeding, and abnormal bleeding associated\nwith a contraceptive method. Subsequent management involved both diagnostic and treatment interven-\ntions, which often occurred simultaneously. The algorithm in this article is designed to help primary\ncare physicians manage abnormal uterine bleeding using strategies that are consistent with the evidence\nas well as the actual practice of gynecologists. (J Am Board Fam Med 2006;19:590 – 602.)\nAbnormal uterine bleeding is a common problem, 1\nand its management can be complex. 2,3 Physicians\nare often unable to identify the cause of abnormal\nbleeding after a thorough history and physical ex-\namination.\n4,5 The management of abnormal bleed-\ning can involve many decisions about diagnosis and\ntreatment,\n3,6,7 which often occur simultaneously\nand without the benefit of comprehensive, evi-\ndence-based guidelines. The available evidence\ntends to focus on narrow treatment questions\nrather than the broad clinical approach to manage-\nment.\n8,17 It is not difficult to find long lists of\npotential causes of abnormal bleeding, but primary\ncare physicians need practical advice about how to\napproach this common problem.\nAbnormal uterine bleeding includes both dys-\nfunctional uterine bleeding and bleeding from\nstructural causes. Dysfunctional bleeding can be\nanovulatory, which is characterized by irregular un-\npredictable bleeding, or ovulatory, which is char-\nacterized by heavy but regular periods (ie, menor-\nrhagia).\n2 Structural causes include fibroids, polyps,\nendometrial carcinoma, and pregnancy complica-\ntions. Abnormal bleeding can also result from con-\ntraceptive methods.\nMany articles have reviewed the management of\nabnormal uterine bleeding,\n3,6,7,15,16,18,21 and they\noften include management algorithms. Although\nclinical algorithms have potential shortcom-\nings,\n22,25 there are data to support their benefit to\nboth physicians and patients. 26,29 Rather than sim-\nply listing causes of abnormal bleeding, manage-\nment algorithms force authors to face the same\ndecisions clinicians face. Most algorithms simply\nThis article was externally peer-reviewed.\nSubmitted 24 April 2006; revised 10 August 2006; ac-\ncepted 24 August 2006.\nFrom the Department of Family Medicine (JWE, ECC),\nand Department of Obstetrics and Gynecology (CMK,\nNCB), University of Iowa Carver College of Medicine, Iowa\nCity, IA.\nConflict of interest: none declared.\nCorresponding author: John W. Ely, MD, MSPH, Univer-\nsity of Iowa College of Medicine, Department of Family\nMedicine, 200 Hawkins Drive, 01291-D PFP, Iowa City, IA\n52242 (E-mail: john-ely@uiowa.edu).\n590 JABFM November–December 2006 Vol. 19 No. 6 http://www.jabfm.org\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from \n\nstate the author’s opinion about what to do. A\nMEDLINE search (1985 to present) found 76 re-\nview articles on abnormal uterine bleeding that\nappeared to address the topic comprehensively, and\n24 of these included an algorithm. Of these 24\nalgorithms, 23 were based on the opinions of the\nauthors and one was based on the available evi-\ndence.\n15 This single evidence-based algorithm ad -\ndressed only one aspect of abnormal bleeding\n(menorrhagia), and most of the diagnostic recom-\nmendations were based on grade C evidence (ex-\npert opinion). Authors who study clinical algo-\nrithms recommend validating them to assure their\nfeasibility in practice,\n30,32 but this is rarely done. 27\nNone of the 24 identified algorithms were system-\natically compared with actual practice. Our goal\nwas to produce a comprehensive algorithm for\nthe management of abnormal uterine bleeding that\nwas consistent with the evidence and feasible in\npractice.\nBleeding Patterns\nWe addressed abnormal uterine bleeding between\nmenarche and menopause. We excluded premen-\narchal bleeding because of its rarity. We excluded\namenorrhea and postmenopausal bleeding because\ntheir generally straightforward evaluation has been\nwell described elsewhere.\n3,4,33,34 Postoperative,\npostpartum, and pregnancy-related bleeding were\nalso excluded.\nWe found that gynecologists usually start the\nevaluation by determining the general pattern of\nabnormal bleeding (Figure 1). Thus, the algorithm\nstarts by asking the physician to categorize patients\naccording to the bleeding patterns defined in Table\n1. Subsequent figures present algorithms for each\npattern. The physician may have difficulty distin-\nguishing prolonged periods from irregular bleed-\ning, and we set an arbitrary bleeding duration of 12\ndays as a limit for menorrhagia. The distinction is\nFigure 1. Abnormal Uterine Bleeding between Menarche and Menopause.\nTable 1. Bleeding Patterns\nNormal: The normal interval is 21 to 35 days. The normal\nduration of bleeding is 1 to 7 days. The amount should be\nless than 1 pad or tampon per 3-hour period\nSevere acute bleeding: Bleeding that requires more than one\npad/tampon per hour or vital signs indicating hypovolemia.\nIrregular bleeding: Includes metrorrhagia,\nmenometrorrhagia, oligomenorrhea, prolonged bleeding,\nintermenstrual bleeding, or other irregular pattern.\nMenorrhagia: Heavy but regular cyclic bleeding plus /H110227 days\nof bleeding or clots or iron deficiency anemia. Prolonged\nbleeding /H1102212 days should be considered irregular regardless\nof cyclic pattern.\nhttp://www.jabfm.org Abnormal Uterine Bleeding 591\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from \n\nimportant because endometrial sampling can often\nbe avoided in patients with menorrhagia. However,\nthe conservative approach would be to follow the\nirregular bleeding algorithm (Figure 3) in border-\nline cases because it calls for endometrial sampling\nin women at high risk for endometrial cancer.\nSevere Acute Bleeding\nSevere acute uterine bleeding in the nonpregnant\npatient usually occurs in one of three settings: the\nadolescent with a coagulopathy (most commonly\nvon Willebrand disease\n35,36), the adult with submu -\ncous fibroids, or the adult taking anticoagulants.\nInitial management is based on hemodynamic sta-\nbility as outlined in Figure 2. The patient is given\nhigh-dose estrogen (orally or intravenously de-\npending on bleeding severity) and then a tapering\nschedule of oral contraceptives. One common oral\ncontraceptive regimen is ethinyl estradiol 30 /H9262g/\nnorgestrel 0.3 mg (eg, LoOvral) 1 active pill 4 times\ndaily for 4 days, followed by 3 times daily for 3\ndays, followed by 2 times daily for 2 days, followed\nby once daily for 3 weeks. The patient then stops\nthe pill for 1 week and then cycles in the usual\nmanner, 3 weeks on and 1 week off, for at least 3\nmonths. Once the patient is clinically stable, an\ninvestigation into the cause of bleeding includes\nscreening coagulation studies and possibly trans-\nvaginal ultrasound (TVUS). The ultrasound may\ninclude a saline-infused sonohysterogram, espe-\ncially when the endometrial stripe is thick, because\nof the increased sensitivity for endometrial polyps\nand submucous fibroids.\n37,38 In general, ultrasound\nis less likely to be helpful at menarche, and instead\nthe evaluation for coagulopathy, especially von\nWillebrand disease, becomes more relevant.\nIrregular Bleeding\nIrregular bleeding is a heterogenous category that\nincludes metrorrhagia, menometrorrhagia, oligo-\nmenorrhea, prolonged bleeding that can last weeks\nor months, and other irregular patterns. These pat-\nterns were lumped together in the algorithm be-\ncause their initial management is similar.\nPatients with minor variations of normal bleeding\nmay not require the evaluation outlined in Figure\n3. For example, irregular bleeding within 2 years of\nmenarche is usually due to anovulation, secondary\nto an immature hypothalamic-pituitary-ovarian ax-\nis.\n21,39,40 However, adolescents may request more\nthan simple reassurance and can be offered oral\ncontraceptives or a progestin as described in the\nalgorithm (Figure 3). Missed periods and pro-\nlonged intervals are expected in perimeno-\npause.\n41,42 Intervals may also decrease in the peri -\nmenopause, but repeated intervals less than 21 days\nor other irregular patterns require endometrial\nsampling. In any reproductive-aged woman, a few\ndays of premenstrual spotting, if it is contiguous\nwith the period, can be a normal variant, but the\ntotal duration should be less than 8 days.\n43 A few\ndays of postmenstrual spotting, if it is contiguous\nFigure 2. Severe Acute Bleeding in the Nonpregnant Patient.\n592 JABFM November–December 2006 Vol. 19 No. 6 http://www.jabfm.org\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from \n\nwith the period, can also be considered a normal\nvariant.43 Postmenstrual spotting is sometimes\ncaused by endometritis, which can be treated with\n100 mg of doxycycline twice daily for 10 days. Brief\nmidcycle spotting can occur at the time of ovula-\ntion due to the normal dip in serum estrogen lev-\nels.\n43 However, this is not common and should\nprompt an endometrial biopsy in women /H1102235 years\nold.2 A single early period ( /H1102121 days) may not\nrequire an endometrial biopsy even in a woman\nover age 35 if subsequent periods are regular and\nno other abnormal bleeding occurs. Early periods\nand occasional missed periods are common in\nyounger women and may result from mental stress\nor illness.\n44,45\nBefore beginning hormonal therapy, systemic\ncauses of abnormal uterine bleeding should be con-\nsidered:\n● If the uterus is tender, indicating possible chronic\nendometritis, the patient should be tested for\ngonorrhea and chlamydia and initially treated\nwith 100 mg of doxycycline twice daily for 10\ndays, pending culture results.\n46,47\n● Medications that can cause abnormal uterine\nbleeding include phenytoin, antipsychotics (eg,\nolanzapine, risperidone), tricyclic antidepressants\n(eg, amitriptyline, nortriptylene), and corticoste-\nroids (eg, prednisone, dexamethasone).\n48\n● Abnormal uterine bleeding can result from ad-\nvanced systemic disease such as liver failure or\nkidney failure.\n48 However, laboratory screening\nfor these diseases in the absence of obvious clin-\nical findings is not necessary because abnormal\nbleeding is a late manifestation. The exception is\nthyroid disease (hypothyroidism or hyperthy-\nroidism), which should be screened for early in\nthe evaluation with a thyroid-stimulating hor-\nmone (TSH).\n● Polycystic ovary syndrome (PCOS) is a common\ncause of abnormal uterine bleeding. 49 The diag -\nnostic criteria for PCOS include at least two of\nthe following\n50,51:\n1. Menstrual irregularity due to oligo- or anovu-\nlation.\n2. Signs of androgen excess, either on physical\nexamination (eg, hirsutism, acne) or laboratory\ntesting (eg, elevated testosterone).\n3. Evidence of polycystic ovaries by ultrasound.\nFigure 3. Irregular Bleeding in the Nonpregnant Patient.\nhttp://www.jabfm.org Abnormal Uterine Bleeding 593\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from \n\nIn addition to these criteria, other causes of hy-\nperandrogenism or abnormal bleeding must be ex-\ncluded before making the diagnosis of PCOS. Con-\nditions that should be ruled out include congenital\nadrenal hyperplasia (manifested by an elevated\nearly morning 17-hydroxyprogesterone), andro-\ngen-secreting tumors (manifested by a serum tes-\ntosterone /H11022200 ng/dL or dehydroepiandrosterone\nsulfate /H11022800/H9262g/dL), and hyperprolactinemia.\nIn women more than age 35 and those at risk for\nendometrial carcinoma (Figure 3), TVUS with or\nwithout a saline-infused sonohysterogram may be\nindicated before, after, or instead of endometrial\nbiopsy. TVUS can detect endometrial polyps, uter-\nine myomas, and endometrial hyperplasia.\n52,53 En-\ndometrial biopsy can detect hyperplasia, atypia, and\ncarcinoma. The conservative approach is to do the\nendometrial biopsy whether or not a TVUS is\nobtained. However, other factors may enter this\ndecision:\n● TVUS may be indicated if the patient will likely\nrequire operative management (eg, office biopsy\nwould be a technical challenge or fibroids sus-\npected on physical examination or probable need\nfor hysteroscopy or endometrial ablation).\n● High-quality TVUS is not available in many lo-\ncations. Also TVUS is costly and insurance status\nmay influence the order of testing.\n● One option is to first rule out neoplasia with the\nendometrial biopsy, then start hormonal therapy,\nand then obtain a TVUS only if abnormal bleed-\ning persists despite hormonal therapy.\n● TVUS is less invasive and less painful than en-\ndometrial biopsy. One study reported experience\nwith initial TVUS and no further evaluation if\nthe double-thickness endometrial stripe was /H110215\nmm.\n54 However, the conservative approach re -\nmains endometrial biopsy in women at risk for\nendometrial carcinoma.\nMenorrhagia\nMenorrhagia is defined as blood loss greater than\n80 mL per cycle. A more pragmatic but less precise\ndefinition is simply the patient’s perception of ex-\ncessive blood loss. Unfortunately, these judgments\ndo not correlate well with actual blood loss.\n55 Men-\norrhagia can often be managed without endome-\ntrial sampling because regular bleeding, even if\nheavy, is less concerning for endometrial cancer.\nHowever, if the bleeding is prolonged ( /H110227 days) or\ndoes not respond to hormonal therapy as outlined\nFigure 4. Menorrhagia in the Nonpregnant Patient.\n594 JABFM November–December 2006 Vol. 19 No. 6 http://www.jabfm.org\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from \n\nin Figure 4, further evaluation with TVUS or en-\ndometrial sampling is indicated. Platelet function\nanalysis to screen for von Willebrand disease\nshould be ordered in women with severe menor-\nrhagia or other signs of coagulopathy.\n36,56,58 For\ntreatment, women can be offered oral contracep-\ntives if not contraindicated (Table 2), progestins\n(Table 3), nonsteroidal anti-inflammatory drugs, or\nobservation. The decision between oral contracep-\ntives and progestins is often based on contraindi-\ncations to estrogen, most commonly smoking. A\nrecent clinical trial found that the levonorgestrel\nintrauterine device (IUD) (Mirena) resulted in\ncomparable quality of life scores and lower costs\ncompared with hysterectomy in women with men-\norrhagia.\n59 Women who prefer no hormones can\nbe started on nonsteroidal anti-inflammatory\ndrugs, which decrease blood loss.\n60,61\nHormonal Contraception\nBreakthrough bleeding occurs commonly with low-\ndose oral contraceptive pills (Figure 5). If the ab-\nnormal bleeding persists after the first 3 months, a\nhigher dose pill can be used, as indicated in Figure\n5. Gonorrhea and chlamydia in association with\noral contraceptives commonly leads to abnormal\nbleeding, and cervical cultures should be obtained.\nPatients on depo-medroxyprogesterone with\npersistent irregular bleeding can be treated with a\n7-day course of estrogen (eg, 1.25 mg of Premarin\ndaily, 1 mg of estradiol daily, or an estrogen patch\nsuch as 0.1 mg Climara). This can be repeated if the\nabnormal bleeding recurs.\nIn patients with an IUD, abnormal bleeding may\nbe associated with endometritis. After culturing the\ncervix, patients with a tender uterus can be treated\nwith 100 mg of doxycycline twice daily for 10 days\nand possible removal of the IUD. In the absence of\nendometritis, patients with a copper IUD (Para-\ngard) can be treated with one cycle of the oral con-\ntraceptive pill or 10 mg of medroxyprogesterone daily\nfor 7 days. Patients with a progestin-releasing IUD\n(Mirena, Progestasert) can be treated with one cycle\nof the oral contraceptive pill. If the abnormal bleeding\npersists, the IUD can be removed and alternative\ncontraceptive methods discussed.\nTable 2. Oral Contraceptive Pill\nCombination Oral Contraceptive Pill\nIf the goal is to achieve amenorrhea, the OCP can be given\ncontinuously, but is usually withdrawn every 3 to 4 months\nto allow endometrial shedding and avoid irregular bleeding.\nIrregular bleeding\nIn most women, suspect a thin endometrium and cycle on\nOCP (eg, Necon 1/35) for at least 3 months. If PCOS is\nsuspected (ie, thick endometrium), consider cyclic progestin\n(Table 3), and then continue cyclic progestin or switch to\nOCP.\nIf there is heavy bleeding at the time of the visit, start a\nmoderate-estrogen OCP (eg, LoOvral) one active pill QID\n/H110034 days, then one TID /H110033 days, then one BID /H110032 days,\nthen daily /H110033 weeks, then skip 1 week, then cycle on OCP\nfor at least 3 months.\nMenorrhagia\nCan start OCP any time but typically on Sunday following\nfirst day of menses.\nContraindications to OCP\nPrevious thromboembolic event or stroke\nHistory of estrogen-dependent tumor\nActive liver disease\nPregnancy\nHypertriglyceridemia\nOlder than 35 years and smokes /H1102215 cigarettes per day\nOlder than 40 years is not a contraindication but many\nphysicians favor progestin for this age group.\nTable 3. Progestin Therapy\nProgestin therapy\nIn most cases, use a cyclic progestin, usually\nmedroxyprogesterone (Provera) because of its low cost. If\nPMS-like side effects are unacceptable, consider micronized\nprogesterone (Prometrium), norethindrone (Aygestin), or\nmegestrol (Megace).\nCyclic progestins\nStart medroxyprogesterone 10 mg daily for 14 days, then off\n14 days, then on 14 days, and so on without regard to\nbleeding pattern. If bleeding occurs before completing the\n14-day course, the patient can double the dose (20 mg) and\n/H11032reset the clock/H11032(count the first day of bleeding as day 1 and\nstart medroxyprogesterone on day 14) or not reset the clock\nand continue the schedule without regard to bleeding pattern.\nIf the patient is bleeding at the time of the visit, start\nmedroxyprogesterone 10 mg daily and increase every 2 days\nas needed to stop the bleeding (20 mg, 30 mg, 40 mg, 60\nmg, 80 mg) until bleeding stops. However, the patient\nshould be warned that intolerable PMS-like side effects may\ndevelop with high doses. Continue for 14 days and then\ncycle 14 days on, 14 days off, and so on.\nContinuous progestins\nContinuous progestins may be indicated if the goal is to\nachieve amenorrhea (eg, busy professional or athlete,\nintractable menstrual migraine, catamenial seizures, severe\nmental retardation). Maintaining amenorrhea is often more\ndifficult than cycling a progestin (ie, there may be\nunpredictable spotting). Options include:\n— Oral progestin: medroxyprogesterone Provera 10 to 20 mg\ndaily or /H11032Minipill/H11032(eg, 0.35 mg of norethindrone daily)\n— Depo-medroxyprogesterone (Depo-Provera) 150 mg IM\nevery 13 weeks. Often used in adolescents to improve\ncompliance. Less often used in ages /H1102240 years due to risk\nof osteoporosis.\n— Levonorgestrel IUD (Mirena).\nhttp://www.jabfm.org Abnormal Uterine Bleeding 595\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from \n\nComment\nIn this review, we developed an algorithm for the\nmanagement of abnormal uterine bleeding and\ncompared it to actual practice. The algorithm is\ngenerally consistent with previous comprehensive\nalgorithms. For example, Albers and colleagues\npresented an algorithm that covered several pages\nin a recent review.\n3 Space limitations forced the\nauthors to use general recommendations such as\n“medical management” rather than specific drugs.\nOther algorithms have solved the space problem by\nlimiting their algorithms to single aspects of abnor-\nmal bleeding, such as only menorrhagia\n15,62,63 or\nonly amenorrhea. 64,67 Some reviews start from the\npathophysiologic perspective, addressing topics\nsuch as “anovulatory bleeding”\n2 or “dysfunctional\nuterine bleeding,” 68 but this approach may be less\nhelpful to clinicians because patients do not present\nwith these labels.\nLittle is known about how to develop clinical\nalgorithms. Authors recognize the importance of\nvalidating clinical algorithms, but they have little\nadvice about how to do it or even what is meant by\n“validation.”\n30,31 Validation could involve building\nalgorithms that optimize patient preferences, phy-\nsician preferences, compliance with the evidence,\nconformity with physicians’ diagnostic reasoning\nprocesses, or, as in this study, conformity with\nactual practice. Algorithms could be tested by de-\ntermining whether physicians follow the “correct”\npath (validity) and whether they follow the same\npath (reliability).\nAlthough the algorithm presented in this article\nis based on the practice of gynecologists in a ter-\ntiary setting, the recommendations should be gen-\nerally applicable to primary care settings in the\nUnited States because they consist of routine tests,\nsuch as pregnancy tests and endometrial biopsies,\nand simple treatments, such as oral contraceptives\nand progestins.\nThe algorithm is lengthy, and busy clinicians\nmight find it unwieldy. However, a clinician with\nan individual patient could focus on only the first\nfigure (Figure 1) plus the one other figure that\naddresses the specific bleeding pattern. Although\nwe could have shortened the algorithm by using\ngeneral recommendations, such as “medical ther-\napy,” or “appropriate laboratory evaluation,” we\nwanted a practical tool that could stand alone at the\npoint of care.\nWe sought to develop a good algorithm, but it\nwas not clear how to define “good.” A good algo-\nFigure 5. Oral Contraceptive Pill-associated Bleeding.\n596 JABFM November–December 2006 Vol. 19 No. 6 http://www.jabfm.org\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from \n\nrithm might be cost-effective, evidence-based, in-\ntuitive, efficient (arrives at a treatment plan quickly\nwithout unnecessary steps), comprehensive (no\nneed to consult other information resources), non-\ninvasive (avoids endometrial biopsy when possible),\npractice-based (works in practice), filled with ac-\nFigure 6. Depo-medroxyprogesterone or Progesterone Only Pill-associated Bleeding.\nFigure 7. Intrauterine Device-associated Bleeding.\nhttp://www.jabfm.org Abnormal Uterine Bleeding 597\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from \n\ntion-oriented advice (“don’t just talk about the\nproblem, tell me what to do”), and able to account\nfor patient preferences. A good algorithm should\nlead to favorable patient outcomes in a randomized\nclinical trial, but trials involving comprehensive al-\ngorithms for complex problems, such as abnormal\nuterine bleeding, are generally not feasible.\nThe algorithm in this study was initially based\non the evidence but modified to match the actual\ncare of patients. The strength of the evidence for\nthe major recommendations in the algorithm are\nsummarized Table 4. The validation procedures we\nfollowed were time consuming and may not be\npractical for algorithms that address other topics.\nHowever, even limited attempts to test an algorithm\nor compare it with actual patient care might reassure\nauthors and readers of its usability in practice.\nAppendix\nHandheld Computer Version of Algorithm for the\nManagement of Abnormal Uterine Bleeding\nA. Initial approach\n1. History and physical examination\n2. Rule out pregnancy\n3. Determine bleeding pattern\na. Severe acute bleeding\nb. Irregular bleeding\nc. Menorrhagia\nd. Contraceptive method (oral contraceptive\npill (OCP), depo-medroxyprogesterone,\nIUD)\nB. Severe acute bleeding (not pregnant)\n1. Orthostatic hypotension or hemoglobin /H1102110\ng/dL or profuse bleeding. Admit to the hos-\npital. Premarin 25 mg IV q4 hours /H1100324\nhours /H1100125 mg of promethazine PO or IM or\nper rectum every 4 to 6 hours as needed for\nnausea. Dilation and curettage (D&C) if no\nresponse after 1 to 2 doses of Premarin.\nTransfuse if hemoglobin /H110217.5 g/dL. Simul-\ntaneous with IV Premarin, start LoOvral, 1\nactive pill QID /H110034d, TID /H110033d, BID /H110032d,\nQD /H110033 weeks, then one week off, then cycle\nfor at least 3 months. If OCP contraindi-\ncated, cycle 10 mg of Provera for 14 days, off\n14 days, on 14 days, and so on for at least 3\nmonths. Obtain TVUS, TSH, complete\nblood cell count (CBC), platelet count, pro-\nthrombin time, activated partial thrombo-\nplastin time, and platelet function analysis.\nStart oral iron.\n2. No orthostatic hypotension, hemoglobin\n/H1135010 g/dL, bleeding not profuse. Outpatient\nmanagement: 2.5 mg of Premarin PO QID\nplus 25 mg of promethazine PO or IM or per\nrectum every 4 to 6 hours as needed for\nTable 4. Strength of Evidence for Major Management Recommendations\nRecommendation\nStrength of\nRecommendation*\nTSH. Obtain a thyroid-stimulating hormone (TSH) serum level in women with irregular bleeding or\nmenorrhagia.69–73\nB\nAge 35. Obtain an endometrial biopsy in women over age 35 with irregular bleeding. 2,74 B\nUnopposed estrogen. Obtain an endometrial biopsy in women with prolonged unopposed estrogen\nregardless of age (most commonly, a woman with polycystic ovary syndrome (PCOS) with few or\nno periods for more than 2 years). 2,75\nC\nTransvaginal ultrasound. Consider transvaginal ultrasound or saline-infused sonohysterogram for\nperimenopausal women with irregular bleeding. 4,54,76\nC\nHormonal therapy for irregular bleeding. Offer oral contraceptives or a progestin for cycle\nregulation in women with irregular bleeding, after ruling out structural causes, systemic causes, and\ncontraindications to the oral contraceptive.\n2,8,77\nB\nHormonal therapy for menorrhagia. Offer oral contraceptives or a progestin to decrease bleeding\nin women with menorrhagia after ruling out structural causes, systemic causes, and\ncontraindications to the oral contraceptive.\n8,10,13,77–79\nB\nNonsteroidal anti-inflammatory drugs for menorrhagia. Offer nonsteroidal anti-inflammatory\ndrugs for women with menorrhagia, after ruling out structural causes and systemic causes. 60,61\nB\n* Strength of recommendation classified according to the 3-component SORT system 80: A, recommendation based on consistent and\ngood-quality patient-oriented evidence80; B, recommendation based on inconsistent or limited quality patient-oriented evidence 80;C ,\nrecommendation based on consensus, usual practice, opinion, disease-oriented evidence, or case series for studies of diagnosis,\ntreatment, prevention, or screening. 80\n598 JABFM November–December 2006 Vol. 19 No. 6 http://www.jabfm.org\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from \n\nnausea. D&C if no response after 2 to 4 doses\nof Premarin or sooner if bleeding /H110221 pad/\nhour. After acute bleeding start LoOvral, 1\nactive pill QID /H110034d, TID /H110033d, BID /H110032d,\nQD /H110033 weeks, then 1 week off, then cycle\nfor at least 3 months. If OCP contraindi-\ncated, cycle 10 mg of Provera for 14 days, off\n14 days, on 14 days, and so on for at least 3\nmonths. Obtain TVUS, TSH, CBC, platelet\ncount, prothrombin time, activated partial\nthromboplastin time, and platelet function\nanalysis. Start oral iron.\nC. Irregular bleeding in nonpregnant patient\n1. TSH. Prolactin if oligomenorrhea.\n2. If more than age 35 or prolonged unopposed\nestrogen, obtain endometrial biopsy and con-\nsider TVUS.\n3. Consider as a cause endometritis (tender\nuterus), medications (phenytoin, antipsy-\nchotics, tricyclic antidepressants, corticoste-\nroids), advanced systemic disease, or polycys-\ntic ovary syndrome.\n4. If the patient does not want to achieve preg-\nnancy, start oral contraceptive (eg, Necon\n1/35) and cycle at least 3 months. If the oral\ncontraceptive is contraindicated, start 10 mg\nof Provera QD for 14 days, off 14 days, on\n14 days, and so on for at least 3 months. If\nabnormal bleeding persists, offer higher\ndose oral contraceptive (eg, Demulen 1/50)\nor higher dose Provera (20 mg, 30 mg, 40\nmg, 60 mg, 80 mg). If abnormal bleeding\npersists, consider TVUS and endometrial\nbiopsy.\n5. Contraindications to oral contraceptives in-\nclude history of thromboembolic event or\nstroke, estrogen-dependent tumor, active\nliver disease, pregnancy, hypertriglyceride-\nmia, smoking more than 15 cigarettes per day\nwhen age is /H1135035.\nD. Menorrhagia in nonpregnant patient\n1. TSH. Hemoglobin. Consider platelet func-\ntion analysis. Consider TVUS if abnormal\nuterus on pelvic examination.\n2. Cycle oral contraceptive (eg, Necon 1/35). If\noral contraceptive contraindicated, 10 mg of\nProvera QD x 14 days, off 14 days, on 14\ndays, and so on for at least 3 months. Other\noptions include nonsteroidal antiinflamma-\ntory drugs (eg, 400 mg of ibuprofen TID for\n4 days, staring day 1 of menses) or no treat-\nment.\n3. If response inadequate, obtain TVUS to\nidentify polyps, myomas, endometrial hyper-\nplasia, adenomyosis.\nE. Oral contraceptive pill-associated bleeding\n1. Menorrhagia. Refer to menorrhagia algo-\nrithm above.\n2. Breakthrough bleeding. If breakthrough\nbleeding occurs during the first 3 months,\nencourage continued use. If breakthrough\nbleeding occurs after 3 months of use or\npatient requests intervention sooner, test for\nchlamydia and gonorrhea, ask about compli-\nance, consider changing to higher estrogen\npill (eg, Necon 1/35, Demulen 1/35,\nDemulen 1/50, LoOvral). If more than age\n35, obtain endometrial biopsy.\n3. Amenorrhea. Rule out pregnancy. Consider\nhigher estrogen pill (eg, Necon 1/35,\nDemulen 1/35, Demulen 1/50, LoOvral). Or\nmay continue same pill because endometrial\nhyperplasia should not develop on oral con-\ntraceptives.\nF. Depo-medroxyprogesterone or progesterone-\nonly pill-associated bleeding.\n1. Amenorrhea. Advise that amenorrhea or\nscant bleeding is expected.\n2. If unacceptable irregular bleeding and pa-\ntient more than age 35 or otherwise at risk\nfor endometrial carcinoma, do endometrial\nbiopsy.\n3. If less than age 35 and not otherwise at high\nrisk for endometrial carcinoma and first 4 to\n6 months of use, can encourage continued\nuse or substitute oral contraceptive, or tem-\nporarily increase frequency of injections (eg,\nevery 2 months).\n4. If less than age 35 and not otherwise at high\nrisk for endometrial carcinoma and after first\n4 to 6 months of use, offer 1.25 mg of Pre-\nmarin QD for 7 days. Can repeat Premarin\ncourse if abnormal bleeding recurs. Consider\nother methods of contraception if bleeding\npersists.\nG. IUD-associated bleeding\n1. Uterus tender; 100 mg of doxycycline BID\nfor 10 days. Consider removal.\n2. First 4 to 6 months of use. Encourage con-\ntinued use. Can offer NSAID (eg, 400 mg of\nhttp://www.jabfm.org Abnormal Uterine Bleeding 599\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from \n\nibuprofen TID for 4 days, starting day 1 of\nmenses).\n3. After 4 to 6 months of use, consider oral\ncontraceptive for one cycle or, if copper\nIUD, 10 mg of Provera QD for 7 days. If\nunacceptable bleeding persists, consider re-\nmoval.\nH. Endometrial biopsy results (eg, pipelle aspira-\ntion; Figure 8).\n1. Polyp. Consider hysteroscopic removal or\nD&C or observation.\n2. Disordered endometrium or stromal collapse\nor proliferative endometrium or secretory\nendometrium. Return to appropriate algo-\nrithm based on bleeding pattern.\n3. Endometritis; 100 mg of doxycycline BID for\n10 days.\n4. Hyperplasia without atypia. Cyclic or contin-\nuous progestin (eg, 10 mg of Provera QD for\n14 days, off 14 days, on 14 days, and so on).\nRepeat biopsy after 3 to 6 months. Refer if\nhyperplasia persists.\n5. Atypia or hyperplasia with atypia or carci-\nnoma. Refer for further counseling and treat-\nment.\nReferences\n1. Nicholson WK, Ellison SA, Grason H, Powe NR. Patterns\nof ambulatory care use for gynecologic conditions: a na-\ntional study. Am J Obstet Gynecol 2001;184:523–30.\n2. ACOG practice bulletin: management of anovulatory\nbleeding. Int J Gynaecol Obstet 2001;72:263–71.\n3. Albers JR, Hull SK, Wesley MA. Abnormal uterine\nbleeding. Am Fam Phys 2004;69:1915–26.\n4. Goldstein SR. Menorrhagia and abnormal bleeding\nbefore the menopause. Best Pract Res Clin Obstet\nGynaecol 2004;18:59 – 69.\n5. Farrell E. Dysfunctional uterine bleeding. Aust Fam\nPhysician 2004;33:906 – 8.\n6. Kilbourn CL, Richards CS. Abnormal uterine bleed-\ning. Diagnostic considerations, management op-\ntions. Postgrad Med 2001;109:137– 8, 141– 4, 147–\n50.\n7. Oriel KA, Schrager S. Abnormal uterine bleeding.\nAm Fam Physician 1999;60:1371– 80; Discussion\n1381–2.\n8. Kuppermann M, Varner RE, Summitt RL, Jr., et al.\nEffect of hysterectomy vs medical treatment on\nhealth-related quality of life and sexual functioning:\nthe medicine or surgery (Ms) randomized trial.\nJAMA 2004;291:1447–55.\n9. Tantiwattanakul P, Taneepanichskul S. Effect of\nmefenamic acid on controlling irregular uterine\nbleeding in DMPA users. Contraception 2004;70:\n277–9.\n10. Rauramo I, Elo I, Istre O. Long-term treatment of\nmenorrhagia with levonorgestrel intrauterine system\nversus endometrial resection. Obstet Gynecol 2004;\n104(6):1314 –21.\n11. Searle J, Grover S, Santin A, Weideman P. Random-\nised trial of an integrated educational strategy to\nreduce investigation rates in young women with dys-\nFigure 8. Endometrial Biopsy (Pipelle aspiration).\n600 JABFM November–December 2006 Vol. 19 No. 6 http://www.jabfm.org\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from \n\nfunctional uterine bleeding. Aus tNZJObstet\nGynaecol 2002;42:395– 400.\n12. Kennedy AD, Sculpher MJ, Coulter A, et al. Effects\nof decision aids for menorrhagia on treatment\nchoices, health outcomes, and costs: a randomized\ncontrolled trial. JAMA 2002;288:2701– 8.\n13. Istre O, Trolle B. Treatment of menorrhagia with\nthe levonorgestrel intrauterine system versus endo-\nmetrial resection. Fertil Steril 2001;76:304 –9.\n14. Cooper KG, Jack SA, Parkin DE, Grant AM. Five-\nyear follow up of women randomised to medical\nmanagement or transcervical resection of the endo-\nmetrium for heavy menstrual loss: clinical and qual-\nity of life outcomes. BJOG 2001;108:1222– 8.\n15. Working Party for Guidelines for the Management\nof Heavy Menstrual Bleeding. An evidence-based\nguideline for the management of heavy menstrual\nbleeding. N Z Med J 1999;112:174 –7.\n16. Bongers MY, Mol BW, Brolmann HA. Current\ntreatment of dysfunctional uterine bleeding. Matu-\nritas 2004;47:159 –74.\n17. Tscherne G. Menstrual irregularities. Evidence-\nbased clinical practice. Endocr Dev 2004;7:129 –39.\n18. Buckingham K, Fawdry A, Fothergill D. Manage-\nment of vaginal bleeding presenting to the accident\nand emergency department. J Accid Emerg Med\n1999;16:130 –5.\n19. Iglesias EA, Coupey SM. Menstrual cycle abnormal-\nities: diagnosis and management. Adolesc Med 1999;\n10:255–73.\n20. Minjarez DA, Bradshaw KD. Abnormal uterine\nbleeding in adolescents. Obstet Gynecol Clin North\nAm 2000;27:63–78.\n21. Minjarez DA. Abnormal bleeding in adolescents. Se-\nmin Reprod Med 2003;21:363–73.\n22. Neale EJ, Chang AM. Clinical algorithms. Med\nTeach 1991;13:317–22.\n23. Sadler C. Pitfalls in the use of clinical algorithms.\nOrtho Clin North Am 1986;17:545–547.\n24. Fred HL. Algorithms: let’s give them back. Hosp\nPract (Off Ed) 2000;35:15– 6.\n25. Kassirer JP, Kopelman RI. Diagnosis and decisions\nby algorithms. Hosp Pract (Off Ed) 1990;25:23– 4,\n27, 31.\n26. McDonald CJ, Wilson GA, McCabe GP, Jr. Physi-\ncian response to computer reminders. JAMA 1980;\n244:1579 – 81.\n27. Shoemaker WC, Corley RD, Liu M, et al. Develop-\nment and testing of a decision tree for blunt trauma.\nCrit Care Med 1988;16:1199 –208.\n28. Wirtschafter DD, Sumners J, Jackson JR, Brooks\nCM, Turner M. Continuing medical education using\nclinical algorithms: a controlled trial assessment of\neffect on neonatal care. Am J Dis Child 1986;140:\n791–7.\n29. Margolis CZ, Cook CD, Barak N, Adler A,\nGeertsma A. Clinical algorithms teach pediatric de-\ncision making more effectively than prose. Med Care\n1989;27:576 –92.\n30. Margolis CZ. Pediatric algorithms. J Pediatr 1987;\n110:417– 8.\n31. Feinstein AR. An analysis of diagnostic reasoning:\nIII. The construction of clinical algorithms. Yale\nJ Biol Med 1974;1:5–32.\n32. Horabin I, Lewis BN. Algorithms. Englewood Cliffs\n(NJ): Educational Technology Publications; 1978.\n33. Speroff L, Fritz MA. Clinical gynecologic endocri-\nnology and infertility. 7th ed. Baltimore: Lippincott\nWilliams & Wilkins; 2004.\n34. Epstein E, Valentin L. Managing women with post-\nmenopausal bleeding. Best Pract Res Clin Obstet\nGynaecol 2004;18:125– 43.\n35. James A, Matchar DB, Myers ER. Testing for von\nWillebrand disease in women with menorrhagia: a\nsystematic review. Obstet Gynecol 2004;104:381– 8.\n36. Shankar M, Lee CA, Sabin CA, Economides DL,\nKadir RA. von Willebrand disease in women with\nmenorrhagia: a systematic review. BJOG 2004;111:\n734 – 40.\n37. Dijkhuizen FP, Mol BW, Bongers MY, Brolmann\nHA, Heintz AP. Cost-effectiveness of transvaginal\nsonography and saline infused sonography in the\nevaluation of menorrhagia. Int J Gynaecol Obstet.\n2003;83:45–52.\n38. Turner RT, Berman AM, Topel HC. Improved\ndemonstration of endometrial polyps and submucous\nmyomas using saline-enhanced vaginal sonohyster-\nography. J Am Assoc Gynecol Laparosc 1995;\n2:421–5.\n39. Mitan LA, Slap GB. Adolescent menstrual disorders.\nUpdate. Med Clin North Am 2000;84:851– 68.\n40. Strickland JL, Wall JW. Abnormal uterine bleeding\nin adolescents. Obstet Gynecol Clin North Am\n2003;30:321–35.\n41. Kaunitz AM. Gynecologic problems of the peri-\nmenopause: evaluation and treatment. Obstet Gy-\nnecol Clin North Am 2002;29:455–73.\n42. Weiss G. Menstrual irregularities and the perimeno-\npause. J Soc Gynecol Investig 2001;8(1 Suppl Pro-\nceedings):S65– 6.\n43. Field CS. Dysfunctional uterine bleeding. Prim Care\n1988;15:561–74.\n44. Crosignani PG, Vegetti W. A practical guide to the\ndiagnosis and management of amenorrhoea. Drugs\n1996;52:671– 81.\n45. Wathen PI, Henderson MC, Witz CA. Abnormal uterine\nbleeding. Med Clin North Am 1995;79:329 – 44.\n46. Chen KT. Acute and chronic endometritis. Available\nfrom: UpToDate.com. Accessed on March 25, 2006.\n47. Michels TC. Chronic endometritis. Am Fam Physi-\ncian 1995;52:217–22.\n48. Brenner PF. Differential diagnosis of abnormal uter-\nine bleeding. Am J Obstet Gynecol 1996;175(3 Pt\n2):766 –9.\nhttp://www.jabfm.org Abnormal Uterine Bleeding 601\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from \n\n49. Gordon CM. Menstrual disorders in adolescents.\nExcess androgens and the polycystic ovary syn-\ndrome. Pediatr Clin North Am 1999;46:519 – 43.\n50. Revised 2003 consensus on diagnostic criteria and\nlong-term health risks related to polycystic ovary\nsyndrome (PCOS). Hum Reprod 2004;19:41–7.\n51. Barbieri RL, Ehrmann DA. Diagnosis and treatment\nof polycystic ovary syndrome in adults. Available\nfrom: UpToDate.com. Accessed on March 25, 2006.\n52. Kelekci S, Kaya E, Alan M, Alan Y, Bilge U, Mol-\nlamahmutoglu L. Comparison of transvaginal\nsonography, saline infusion sonography, and office\nhysteroscopy in reproductive-aged women with or\nwithout abnormal uterine bleeding. Fertil Steril\n2005;84:682– 6.\n53. Valenzano MM, Lijoi D, Mistrangelo E, Fortunato\nT, Costantini S, Ragni N. The value of sonohyster-\nography in detecting intracavitary benign abnormal-\nities. Arch Gynecol Obstet 2005;272:265– 8. Epub\n2005 Oct 13.\n54. Goldstein SR, Zeltser I, Horan CK, Snyder JR,\nSchwartz LB. Ultrasonography-based triage for\nperimenopausal patients with abnormal uterine\nbleeding. Am J Obstet Gynecol 1997;177:102– 8.\n55. Warner PE, Critchley HO, Lumsden MA, Camp-\nbell-Brown M, Douglas A, Murray GD. Menorrha-\ngia I: measured blood loss, clinical features, and\noutcome in women with heavy periods: a survey with\nfollow-up data. Am J Obstet Gynecol 2004;\n190:1216 –23.\n56. Kouides PA, Conard J, Peyvandi F, Lukes A, Kadir\nR. Hemostasis and menstruation: appropriate inves-\ntigation for underlying disorders of hemostasis in\nwomen with excessive menstrual bleeding. Fertil\nSteril 2005;84:1345–51.\n57. Kujovich JL. von Willebrand’s disease and menor-\nrhagia: prevalence, diagnosis, and management.\nAm J Hematol 2005;79:220 – 8.\n58. Philipp CS, Miller CH, Faiz A, et al. Screening\nwomen with menorrhagia for underlying bleeding\ndisorders: the utility of the platelet function analyser\nand bleeding time. Haemophilia 2005;11:497–503.\n59. Hurskainen R, Teperi J, Rissanen P, et al. Clinical\noutcomes and costs with the levonorgestrel-releasing\nintrauterine system or hysterectomy for treatment of\nmenorrhagia: randomized trial 5-year follow-up.\nJAMA 2004;291:1456 – 63.\n60. Grover V, Usha R, Gupta U, Kalra S. Management\nof cyclical menorrhagia with prostaglandin syn-\nthetase inhibitor. Asia Oceania J Obstet Gynaecol\n1990;16:255–9.\n61. Fraser IS, McCarron G. Randomized trial of 2 hor-\nmonal and 2 prostaglandin-inhibiting agents in\nwomen with a complaint of menorrhagia. Aust N Z\nJ Obstet Gynaecol 1991;31:66 –70.\n62. Rosenfeld JA. Treatment of menorrhagia due to dys-\nfunctional uterine bleeding. Am Fam Physician\n1996;53:165–72.\n63. Rees M. Menorrhagia. Br Med J (Clin Res Ed) 1987;\n294:759 – 62.\n64. Bulusu S. Secondary amenorrhoea J R Soc Med\n1996;89:220P–1P.\n65. Aloi JA. Evaluation of amenorrhea. Compr Ther\n1995;21:575– 8.\n66. Burnett RG. Diagnostic strategies for amenorrhea.\nPostgrad Med 1990;87:241–7, 250.\n67. Franks S. Primary and secondary amenorrhoea. Br\nMed J (Clin Res Ed) 1987;294:815–9.\n68. Bayer SR, DeCherney AH. Clinical manifestations\nand treatment of dysfunctional uterine bleeding.\nJAMA 1993;269:1823– 8.\n69. Rowland AS, Baird DD, Long S, et al. Influence of\nmedical conditions and lifestyle factors on the men-\nstrual cycle. Epidemiology 2002;13:668 –74.\n70. Weeks AD. Menorrhagia and hypothyroidism. Evi-\ndence supports association between hypothyroidism\nand menorrhagia BMJ 2000;320:649.\n71. Krassas GE, Pontikides N, Kaltsas T, et al. Distur-\nbances of menstruation in hypothyroidism. Clin En-\ndocrinol (Oxf) 1999;50:655–9.\n72. Koutras DA. Disturbances of menstruation in thy-\nroid disease. Ann N Y Acad Sci 1997;816:280 – 4.\n73. Krassas GE, Pontikides N, Kaltsas T, Papadopoulou\nP, Batrinos M. Menstrual disturbances in thyrotox-\nicosis. Clin Endocrinol (Oxf) 1994;40:641– 4.\n74. SEER data Available from: http://seer.cancer.gov/\nfaststats/sites.php?stat /H11005Incidence&site /H11005Corpus/H11001\nand/H11001Uterus%2C/H11001NOS/H11001Cancer&x/H1100513&y/H1100516.\nAccessed on April 15, 2006.\n75. Farhi DC, Nosanchuk J, Silverberg SG. Endometrial\nadenocarcinoma in women under 25 years of age.\nObstet Gynecol 1986;68:741–5.\n76. Dubinsky TJ. Value of sonography in the diagnosis\nof abnormal vaginal bleeding. J Clin Ultrasound\n2004;32:348 –53.\n77. Learman LA, Summitt RL Jr., Varner RE, et al.\nHysterectomy versus expanded medical treatment\nfor abnormal uterine bleeding: clinical outcomes in\nthe medicine or surgery trial. Obstet Gynecol 2004;\n103(5 Pt 1):824 –33.\n78. Hurskainen R, Teperi J, Rissanen P, et al. Clinical\noutcomes and costs with the levonorgestrel-releasing\nintrauterine system or hysterectomy for treatment of\nmenorrhagia: randomized trial 5-year follow-up.\nJAMA 2004;291:1456 – 63.\n79. Lahteenmaki P, Haukkamaa M, Puolakka J, et al.\nOpen randomised study of use of levonorgestrel re-\nleasing intrauterine system as alternative to hyster-\nectomy. BMJ 1998;316:1122– 6.\n80. Ebell MH, Siwek J, Weiss BD. Strength of Recom-\nmendation Taxonomy (SORT): a patient-centered\napproach to grading evidence in the medical litera-\nture. J Fam Pract 2004;53:111–20.\n602 JABFM November–December 2006 Vol. 19 No. 6 http://www.jabfm.org\n on 13 June 2026 by guest. Protected by copyright.http://www.jabfm.org/ J Am Board Fam Med: first published as 10.3122/jabfm.19.6.590 on 7 November 2006. Downloaded from","source_license":"CC0","license_restricted":false}