7 Emergency gynaecology S.R. Ramphal * MBChB, FCOG (SA) Consultant Obstetrician and Gynaecologist/Lecturer Department of Obstetrics and Gynaecology, Nelson R Mandela School of Medicine, Private Bag 7, Congella, 4013, South Africa J. Moodley MBChB, FCOG (SA), FRCOG (UK), MD Consultant Obstetrician and Gynaecologist Women’s Health Group, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Durban, South Africa The management of gynaecological emergencies is directed at the preservation of life, health, sexual function and the perpetuation of fertility. Ectopic pregnancy (EP), pelvic inflammatory dis- ease (PID) and miscarriages are common gynaecological emergencies and early recognition and appropriate treatment is essential to avoid unwanted sequelae. Controversy will always exist in clinical medicine because management is mainly based on uncontrolled studies, expert opinion and personal experiences. It is estimated that only 10% of clinical treatments have been validated by prospective, randomised trials. Recent advances have led to earlier diagnosis and more con- servative treatment on an outpatient or day care basis in EP and miscarriages. Key words: ectopic pregnancies; pelvic inflammatory disease; miscarriages; randomised con- trolled trial. Box. Question and Literature Sources Question components Population: women with gynaecological emergencies (ectopic pregnancy/ miscarriage/acute pelvic sepsis Interventions: treatments for the above Outcomes: cure rates, failure rates, etc. * Corresponding author. Fax: 27 031 2604241. E-mail address: [email protected](S.R. Ramphal). 1521-6934/$ - see front matter ª 2006 Elsevier Ltd. All rights reserved. Best Practice & Research Clinical Obstetrics and Gynaecology Vol. 20, No. 5, pp. 729e750, 2006 doi:10.1016/j.bpobgyn.2006.04.001 available online at http://www.sciencedirect.com
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
7
Best Practice & Research Clinical Obstetrics and GynaecologyVol. 20, No. 5, pp. 729e750, 2006
doi:10.1016/j.bpobgyn.2006.04.001available online at http://www.sciencedirect.com
Emergency gynaecology
S.R. Ramphal* MBChB, FCOG (SA)
Consultant Obstetrician and Gynaecologist/Lecturer
Department of Obstetrics and Gynaecology, Nelson R Mandela School of Medicine, Private Bag 7,
Congella, 4013, South Africa
J. Moodley MBChB, FCOG (SA), FRCOG (UK), MD
Consultant Obstetrician and Gynaecologist
Women’s Health Group, Nelson R Mandela School of Medicine, University of KwaZulu-Natal,
Durban, South Africa
The management of gynaecological emergencies is directed at the preservation of life, health,sexual function and the perpetuation of fertility. Ectopic pregnancy (EP), pelvic inflammatory dis-ease (PID) and miscarriages are common gynaecological emergencies and early recognition andappropriate treatment is essential to avoid unwanted sequelae. Controversy will always exist inclinical medicine because management is mainly based on uncontrolled studies, expert opinionand personal experiences. It is estimated that only 10% of clinical treatments have been validatedby prospective, randomised trials. Recent advances have led to earlier diagnosis and more con-servative treatment on an outpatient or day care basis in EP and miscarriages.
Question componentsPopulation: women with gynaecological emergencies (ectopic pregnancy/miscarriage/acute pelvic sepsisInterventions: treatments for the aboveOutcomes: cure rates, failure rates, etc.
Similar to medical emergencies, the management of gynaecological emergencies isdirected towards preservation of life, health and function, in particular the conserva-tion of sexual function and the perpetuation of fertility if possible. Common gynaeco-logical emergencies present as an acute abdomen, abnormal vaginal bleeding, ora combination of both and are related to early pregnancy complications, pelvic inflam-matory disease (PID) and contraceptive issues. Recent advances in sonography,biochemical pregnancy testing, minimal access surgery and newer antibiotics haveled to early diagnosis and the adoption of more conservative approaches to treatment.
Evidence-based medicine involves integrating the current best evidence with clinicalexpertise and patient preferences to make optimal decisions.1 Under ideal conditions,all medical practice should be evidence-based. However, it has been estimated that only4% of decisions are based on robust evidence from clinical studies, 45% are basedsimply on moderate to strong consensus among physicians and 51% are based on littleevidence and consensus.2 A review of the evidence for ectopic pregnancy (EP), PID andmiscarriages and emergency contraception presenting to the emergency room will beoutlined.
Ectopic pregnancy
The incidence of EP has increased dramatically worldwide with a reported fourfold in-crease in the USA.3 The reasons for this include both the aetiological influence of sex-ually transmitted diseases, which has recently been compounded by the HIV epidemic4
and the fact that earlier diagnosis using sensitive b-human chorionic gonadotrophin(b-HCG) assays and the improvement in ultrasound techniques, particularly the trans-vaginal route, results in patients whose condition might previously have resolved spon-taneously, now requiring treatment.5
Early diagnosis and appropriate management is critical in that it allows the optionsof medical treatment or conservative surgical procedures such as salpingostomy orsegmental resection of the affected part of the tube.
Features of an EP include an empty uterus in the face of a positive pregnancy test,complex adnexal masses, or a ring structure in the fallopian tube.6 In women present-ing with less than 6 weeks amenorrhoea, it can be difficult to confirm an intrauterinepregnancy even with a transvaginal scan and the diagnosis of an EP is only about 40%correct, even in the best hands.7 In this scenario, a quantitative estimation of the b-HCG level is helpful in the diagnosis and may also facilitate conservative management
Literature sourcesElectronic databases: PubMed, MEDLINE, EMBASE, Cochrane Library, BestEvidence, etc.Manual search: Personal files of articles available with authors, reference listsof all known primary and traditional review articles.Contact with experts.
Emergency gynaecology 731
in selected cases.7 Serial quantitative b-HCG values over 48 h are also very helpful. Innormal pregnancy, the b-HCG level should at least double every 2e3 days, whereas inabnormal pregnancies, including EP, the doubling rate is slower with a <66% increaseover 48 h.8 If a quantitative b-HCG level exceeds 6500 IU/L with no evidence of anintrauterine pregnancy using abdominal ultrasound, a diagnostic laparoscopy shouldbe performed. Since vaginal ultrasound can detect an intrauterine pregnancy abouta week earlier than that of an abdominal scan, b-HCG levels above 1000 IU/L in the ab-sence of an intrauterine sac indicates the need for laparoscopy.9 The absence of an in-trauterine gestational sac and an HCG titre between 1000e1500 IU/L has been shownto be highly predictive of an EP (sensitivity 95%, specificity 95%, odds ratio (OR)¼ 24.8),thus making intervention rather than expectant management appropriate.9
Serum progesterone levels have been suggested as useful markers for the diagnosisof EP, however a systematic review of the accuracy of a single progesterone measure-ment concluded that although the progesterone concentration could identify womenat risk for EP, its clinical usefulness is limited because of its discriminative capacity todiagnose EP with certainty.10
Expectant management
Although surgery is the mainstay of management for EP, options of medical or expec-tant management are available for a selected group of patients. Successful clinical res-olution with expectant management has been reported in several small series but thisoption is not routinely practiced because of the unpredictability of the natural courseof EP and the prolonged period of hospitalisation and follow up that is required. Ananalysis of 10 prospective studies with a total of 347 patients managed expectantly,demonstrated that 69.2% of the EP resolved spontaneously. Favourable factors in-cluded a low initial b-HCG level (1e2000 IU/L), a haemoperitoneum of <50 mls, tubalmass <2 cm, the absence of recognisable fetal parts on ultrasound and the absence ofclinical symptoms. However, rupture of EP still occurs under these circumstances,making it uncommon for this form of management to be instituted.11
Laparoscopy versus laparotomy
Laparoscopy is considered by many to be the ‘gold standard’ for surgical treatment if thepatient is haemodynamically stable. Three randomised controlled trials (RCTs) com-pared traditional laparotomy with laparoscopy and in all three studies, the laparoscopicapproach was associated with significantly less blood loss, lower analgesic requirement,shorter hospital stay, quicker post operative recovery time and lower costs.12e14 Thesubsequent intrauterine pregnancy rate was 61% after laparoscopic surgery comparedto 52% after laparotomy, while the recurrent EP rate was lower after laparoscopy (8%)than after laparotomy (14.4%). However, the laparoscopic salpingotomy route was lesssuccessful than the open approach with regard to elimination of tubal pregnancy. Thepersistent EP rate was 12.2% in the laparoscopy group compared to 1.7% in the laparot-omy group. It is important to note that these three trials included only 231 patients,making interpretations of the small differences between the two interventions difficult.Further studies are needed to establish whether the persistent EP rate is still high in thelaparoscopic group and these studies should only include experienced surgeons withoptimal instrumentation. In a follow-up study, it was found that intra-operative adhe-sions, which are a risk factor for pelvic pain and infertility, developed significantly
732 S. R. Ramphal and J. Moodley
more often after laparotomy than after laparoscopic surgery ( p< 0.0001).15 At present,with the balance of evidence, the laparoscopic route should be favoured.
Salpingectomy or salpingostomy
The decision to perform a salpingectomy or a salpingostomy will depend on the size ofthe EP, the damage to the tube and the health of the contralateral tube. There is noRCT comparing outcome and assessing future fertility with either form of treatment.Non-randomised retrospective studies have shown no significant differences betweenthe two forms of treatment with regard to intrauterine pregnancy rates, but the rateof EP is increased after salpingostomy. Yao & Tulandi,16 reviewing a meta-analysis ofnine comparative studies, reported no significant difference in the intrauterine preg-nancy rate following salpingotomy (53%) and salpingectomy (49.3%). However, the re-current EP rate was higher after salpingotomy (15%) than after salpingectomy (10%).16
A recent study17 reviewed a cohort of 276 women undergoing salpingotomy or salpin-gectomy and showed a cumulative intrauterine pregnancy rate at 7 years of 89% fol-lowing salpingotomy compared with 66% following salpingectomy ( p< 0.05). Thehazard ratio for intrauterine pregnancy following salpingectomy was 0.63 (95%CI¼ 0.42e0.94) when compared to salpingectomy. In patients with contralateral tubalpathology, the chance of pregnancy was poor (hazard ratio¼ 0.463) and the risk ofrecurrent EP was high (hazard ratio¼ 2.25). Failure to totally remove all of the tro-phoblastic tissue is a definite problem with salpingotomy, leading to a 4.8e11% occur-rence of persistent EP18,19 compared with no cases following salpingectomy.20 In thepresence of a normal contralateral tube, salpingectomy is to be preferred to salpingos-tomy as it is associated with a lower rate of persistent trophoblast and subsequent re-peat EP, while future intrauterine pregnancy rates are similar. However, salpingostomyshould be considered as the primary treatment option in the presence of disease orabsent contralateral tube and the desire for future fertility, but the patient must bemade aware that there is an approximately 20% risk of a recurrent EP. If conceptionhas not occurred after 18 months following salpingostomy, it is unlikely to occurand in vitro fertilisation should be recommended.
It is prudent that b-HCG levels be measured after salpingostomy until they are un-detectable. Different regimens for monitoring b-HCG levels following salpingotomyhave been suggested. Yao & Tulandi recommended that the serum b-HCG level shouldbe measured before surgical treatment and once 7 days after treatment and if the levelis higher than expected, then the patient should be given a single intramuscular dose ofmethotrexate (MTX).16 Factors that increase the risk of persistent EP include earlygestation (<42 days from last menstrual period), small EP less than 2 cm, a high levelof b-HCG (>3000 IU/L) before treatment and milking the EP from the tube.21
Medical management
Early diagnosis has made medical therapy of EP an option. A review of controlled anduncontrolled studies has revealed that, in stable patients, a variety of medical treat-ments are as effective as surgery. Agents that have been used include hyperosmolarglucose, urea, cytotoxic agents, e.g. MTX, prostaglandins and mifepristone. Two recentRCTs have compared systemic MTX therapy with laparoscopic surgery. One22 com-pared MTX (at a dose of 1 mg/kg) with conservative laparoscopic surgery and founda 78% success rate for one dose of MTX compared with 92% for laparoscopic surgery.
Emergency gynaecology 733
In the MTX group, 16% required an additional dose and 5% required surgery duringthe follow-up period, while in the laparoscopic group, 8% had evidence of persistenttrophoblastic tissue. In a second randomised trial comparing MTX (50 mg/m2) withconservative laparoscopic surgery,23 success rates of 65% were reported for a singledose of MTX and 93% for laparoscopic surgery (95% CI¼ 10e47%; p< 0.05); 26% inthe MTX group required a further dose and 15% underwent laparoscopy during thefollow-up while 2% in the laparoscopic group had persistent trophoblast. An economicevaluation conducted alongside this trial showed that medical treatment was associatedwith reduced direct and indirect costs.24 These financial benefits however, were lost atb-HCG levels >1500 IU/L due to the need for prolonged follow-up and surgicalinterventions.
The criteria for medical management include haemodynamic stability, confirmationof the EP by ultrasound, significant risks associated with general anaesthetic, patientcompliance, lack of contraindication to medical therapy, small size of EP mass(<3.5 cm) and lack of fetal cardiac motion. The most studied agent for medical man-agement is MTX. Since the main advantage of medical therapy is its non-invasiveness,intramuscular MTX is more practical and less operator dependent than local injectionvia laparoscopy or ultrasound into the sac. A review of 24 studies found a surgical in-tervention rate of 9% when MTX was injected locally into the gestational sac, eitherthrough the laparoscope or under ultrasound guidance, and a rate of 3.2% in the sys-temically treated group.25 In that series, MTX was given in multiple doses in the sys-temically treated group. In a recent study addressing the route of MTX administrationin 137 women treated by either intramuscular (15 mg/m2) or local ultrasound-guidedadministration (1 mg/kg), the success rate was 67.1% and 92.5%, respectively.26 Thiswas not a randomised trial and further research is required to determine the mosteffective route of administration. Local injection of MTX under ultrasound guidancehad a success rate of 70e95% compared to the laparoscopically-injected group, whichvaried from 43e100%.16 The ultrasound route does not require operative inter-vention unlike laparoscopy. Another prospective randomised trial demonstrated thatwhen an ectopic sac is easily visualised by ultrasound, local injection of MTX is as ef-fective as systemic treatment and has minimal side effects.27 When direct injection andultrasound is technically difficult, then systemic MTX injection was recommended.
Variable dose, single dose and low dose intramuscular MTX regimens have beenstudied. The variable dose regimen involves alternating MTX with leucovorin. Duringtreatment, the b-HCG levels are evaluated daily and this regimen continues until theb-HCG level declines by more than 15% in 48 h. It has been reported that this regimenappears to be 82e85% effective in resolving EP and rates of subsequent fertility andtubal patencies are comparable to those for conservative surgical management.28 Asingle dose regimen (50 mg/m2 or 1 mg/kg body weight) has been administered in dif-ferent studies and success rates of 85e94%29 have been reported, which are compa-rable to rates using multiple dose regimens.29,30 If the HCG level fails to decrease by atleast 15% between day 4 and day 7 after a single intramuscular dose, then a furtherdose is given.22,23,31 Large uncontrolled studies have reported that about 14% ofwomen will require more than one dose of MTX and less than 10% of women treatedwith this regimen will require surgical intervention.32,33 The success rate increaseswith the addition of a second dose.30,34 The efficacy of single and multiple dose reg-imens has recently been compared in a meta-analysis of available studies.35 The successrate was defined as not requiring surgery and was 88.1% (940/1067) for single dosetherapy and 92.7% (241/260) for multiple dose therapy. Of significance, this differencewas much more marked when the results were adjusted for serum b-HCG values and
734 S. R. Ramphal and J. Moodley
the presence of fetal cardiac activity (OR¼ 4.74; 95% CI¼ 1.77e12.62). However,side-effects were lower with single dose therapy. Amongst women who were dueto receive a single dose, 13.6% required two or more doses. It appears that the serumb-HCG level, and not the size of the EP in clinically stable women, is the most impor-tant factor in the failure of medical treatment. In a prospective cohort study treatedwith single dose MTX, a success rate of 97% was achieved at a HCG level of<2000 IU/L, compared with 74% at levels above 2000 IU/L.36 In a retrospective reviewof 81 consecutive patients treated with MTX, success rates of 98% were found inwomen with HCG levels <1000 IU/L, falling to 80% at levels between 1000e4999 IU/Land to 38% at levels above 5000 IU/L.37 A finding unique to this study was that visual-isation of a yolk sac was a risk factor for failure. Similarly, fetal cardiac activity is a riskfactor for failure of MTX treatment. Overall, it would appear that if the strict criteriaare fulfilled, medical management is a reasonable alternative but costs, longer hospital-isation, prolonged follow-up and patient choice need to be considered.
ACUTE ABDOMEN
Pelvic inflammatory disease
Since the early 1960s, there has been a worldwide increase in the incidence of pelvicinflammatory disease (PID). It is the most common infectious disease that affectsyoung women (15e25 years) and accounts for 94% of the morbidity that is associatedwith sexually transmitted disease (STD) in well-resourced countries.
PID may be devastating in its effect on the afflicted individual and her family. Al-though it does not usually constitute an emergency in the sense that immediate treat-ment is life-saving, urgent therapy is required to minimise the effect of the disease onsubsequent fertility and reduce the risk of sequelae such as EP and chronic pain. Thisapplies to both mild and severe disease.
The diagnosis may be difficult because the signs and symptoms vary and often over-lap with other disease entities and may also vary according to the causative pathogen.The accuracy with which signs and symptoms predict the presence of PID has beenevaluated using laparoscopy as the ‘gold standard’. A comparison using clinical and lap-aroscopic diagnosis and a wide range of clinical features (Table 1) and strict laparo-scopic criteria (Table 2) reported that only 66% of woman with clinically diagnosedPID have the condition and that standard clinical criteria also fail to identify 10% of
Table 1. Minimal criteria for the provisional diagnosis of pelvic inflammatory disease.
Acute lower abdominal/pelvic painþ 2 of the following features:
A Abnormal vaginal discharge
A Fever
A Vomiting
A Menstrual abnormalities
A Urinary symptoms
A Proctitis symptoms
A Marked pelvic tenderness
A Palpable mass or swelling
A Erythrocyte sedimentation rate R15 mm/h
Emergency gynaecology 735
cases subsequently diagnosed by laparoscopy or laparotomy.38 A meta-analysis of 19studies evaluating the clinical diagnosis of PID in a variety of settings (grade 2 evi-dence), showed that between 30 and 40% of women were incorrectly diagnosed ashaving PID in the absence of laparascopic evidence.39 It also concluded that wherethere are strict clinical criteria, the accuracy of the clinical diagnosis is greater andwhere extended entry criteria are used, the specificity of the clinical diagnosis ispoor. Most studies in this meta-analysis did not address the issues of sensitivity, sincewomen with clinically mild disease were unlikely to be entered in to a study that re-quired hospitalisation for laparoscopy. A critical evaluation of the available evidenceconcerning the diagnosis of PID, based on clinical presentation, also examined therelationship between signs and symptoms and the presence of laparoscopic investiga-tion.40 All of the variables investigated (abnormal vaginal discharge, fever >38 �C,vomiting, menstrual irregularity, ongoing bleeding, symptoms of urethritis, rectal tem-perature >38 �C, marked tenderness of pelvic organs on bimanual examination, ad-nexal mass and erythrocyte sedimentation rate �15 mm in the first hour) had bothlow sensitivity and specificity.40 The discriminant analysis indicated that three variablessignificantly influenced the prediction of PID e erythrocyte sedimentation rate( p< 0.0001), fever ( p< 0.0001) and adnexal tenderness ( p< 0.0001) e and correctlyclassified 65% of patients with laparoscopically diagnosed PID (95% CI¼ 61e69%).40
Costs, limited access and surgical risks prevent the universal use of laparoscopy forthe diagnosis of PID.
Endometrial biopsy
A review of four studies, in which laparoscopy was compared with endometrial biopsy,concluded that if laparoscopy is the gold standard, then the sensitivity and specificity ofhistology obtained by endometrial biopsy is 50e87% and 57e92%, respectively.39 Thedrawback with endometrial biopsy studies is that firstly, in early cases, there may beevidence of endometritis before the infection reaches the fallopian tubes. Secondly, in-fection may reach the tubes via the parametria rather than via the tubes themselves,41
therefore it would be reasonable to argue that all cases with PID diagnosed histolog-ically or laparascopically, should be considered ‘true’ cases and therefore a new goldstandard should incorporate both factors.
Culdocentesis
Culdocentesis, which involves aspirating fluid from the pouch of Douglas through theposterior fornix, has largely been used to identify the microbiological cause of diseasein women with signs strongly suggestive of moderate to severe PID or to exclude anEP. Studies have demonstrated a poor correlation between the microbiological findings
Table 2. Laparoscopic criteria for the diagnosis of acute pelvic inflammatory disease.
1. Pronounced hyperaemia of tubal surface
2. Oedema of the tubal wall
3. Sticky exudate on tubal surface and from fimbriated ends
All three required for diagnosis
736 S. R. Ramphal and J. Moodley
of peritoneal fluid obtained by culdocentesis and that obtained directly by laparos-copy.42 Currently, there are no studies that compare culdoncentesis with laparoscopyor endometrial biopsy and therefore it is not possible to evaluate its usage as a diag-nostic technique.
Imaging techniques
Pelvic imaging in the form of ultrasound, computerised tomography (CT) and magneticresonance imaging (MRI) have been used to visualise and evaluate the pelvis in suspectedPID. Traditional transabdominal ultrasound detects tubo-ovarian abscesses and may beuseful in excluding other diagnostic possibilities such as torsion of an ovarian cyst. Acomparison of abdominal ultrasound with laparoscopy found that abnormalities com-patible with PID were recorded in 78% of patients.43 However, inflamed fallopian tubeswithout thickening will not be detected on ultrasound.44 Transvaginal ultrasound is use-ful and has a sensitivity of 80% in diagnosing PID and in ruling out tubo-ovarian abscesses,ovarian cyst and ovarian torsion.45 A report on 30 patients with severe pelvic pain, andthe clinical suspicion of PID, showed that ultrasound had an 81% sensitivity and 78%specificity when compared with laparoscopy.46 Despite numerous publications describ-ing ultrasound findings, there are no studies in unselected populations of woman withacute pelvic pain in which ultrasound studies and laparoscopy have been performed in-dependently.47 Colour and power Doppler should show an increased flow (hyperaemia)in the walls and incomplete septi of the inflamed tube but there have been no reportedcomparative studies of its value.
MRI may be useful and has a sensitivity and specificity of 95% and 93%, respectively,when compared with laparoscopy. However, this technique is expensive and unlikely tofind widespread acceptance. CT scanning has also been evaluated and is thought toplay a role when ultrasound findings are equivocal, but there are no prospective stud-ies comparing this technique with other diagnostic techniques.
Microbiological specimens
The true significance of microbiological tests that could be used to diagnose PID is ques-tionable. Microbiological tests could identify 80% of cases, but not all patients withpelvic pain and pathogens in the lower genital tract have PID. However, the identifica-tion of pathological organisms in the lower tract may add weight to the diagnosis of PID.
Serological markers
Investigators have evaluated the role of laboratory markers, namely white blood cellcount, erythrocyte sediment rate and C-reactive protein, as diagnostic markers ofPID. The results of these studies lack consistency in that some investigators haveshown statistical significance in these markers when woman with PID are comparedwith those without the condition,38,48 while others have not.49,50 To date, there isno single test that has adequate sensitivity and specificity for routine use and increasingthe number of tests to aid the diagnosis increases the specificity but decreases thesensitivity.
Emergency gynaecology 737
Clinical data
The Center for Disease Control (CDC) has formulated criteria (Table 3) for thediagnosis of PID that are divided to facilitate sensitivity and specificity.51 The minimumcriteria listed are well suited for improving the sensitivity of diagnosis. In patients whoare at high risk (where the prevalence is high) all three minimum criteria are requiredin making the diagnosis. The additional criteria may help to improve the diagnosticspecificity, which is useful in patients who are at low risk (in whom the prevalence
is low).51 According to one study,52 the CDC criteria have a sensitivity of 83%. Ofnote, the single finding of adnexal tenderness has a sensitivity of 95% in ‘at risk’women,52 resulting in some investigators recommending treatment for PID basedon this sign alone in a high risk population.53
TREATMENT
Since there are no reliable clinical diagnostic criteria for PID and the positive pre-dictive value of a clinical diagnosis is 65e90%, empirical treatment is recommended.However, to date, there is no systematic review or RCT to support or refute em-pirical treatment for suspected PID, nor is there reliable evidence of harm whentreated. Mild disease may be treated on an out-patient basis and it is essentialthat all possible measures to ensure patient compliance are applied as early cessationof treatment may promote long-term sequelae. The goals of treatment are twofold:(1) relief of the acute symptoms and inflammation and (2) prevention of the long-term sequelae associated with PID. Current CDC treatment regimens are shownin Table 4. Optimal management of patients with PID should be individualised onthe basis of clinical setting and patient characteristics. Systematic reviews of RCTsand case series have found that several regimens of parenteral followed by oral an-tibiotic treatment are effective in resolving acute symptoms and signs associated withPID. However, there is no good evidence on the optimal duration of treatment orcomparing oral versus parenteral treatment. Two systematic reviews, which identi-fied 26 studies of 16 antimicrobial regimens in 1925 woman with PID, found antibi-otics to be effective in relieving the symptoms associated with PID, with clinical andmicrobiological cure rates of 90e100%.54,55 In these reviews, duration of treatmentwas not addressed, although the most common treatment period was 14 days.
Table 3. Center for disease control criteria for the diagnosis of pelvic inflammatory disease.
Minimum criteria:
A Uterine/adnexal tenderness
A Cervical motion tenderness
A No other cause of above signs notedAdditional criteria:
A Oral temperature of >101 �F (>38.3 �C)
A Abnormal cervical or vaginal mucopurulent discharge
A Presence of white blood cells on saline microscopy of vaginal secretions
A Elevated erythrocyte sedimentation rate
A Elevated C-reactive protein
A Laboratory documentation of cervical infection with Neisseria gonorrhoea or Chlamydia trachomatis
738 S. R. Ramphal and J. Moodley
Furthermore, the reviews did not analyse outcome by oral or parental route, al-though most regimens started with parental treatment and continued with oraltreatment at different points. At present, there is no data that clearly establisheswhich patients with mild to moderate PID would best benefit from in-patient versusout-patient management or oral versus parenteral therapy.
In woman who have an intra-uterine contraceptive device (IUCD), common practiceis to remove the IUCD when making the diagnoses of PID. An IUCD increases the risk ofdeveloping PID in the first few weeks after insertion56 and in a RCT, it was evident thatremoving an IUCD does not alter the response to treatment.57 Unfortunately, this studyhad suboptimal outcome measures. An observational study showed no benefit in re-moving an IUCD.58 Hence, one can conclude that an IUCD may be left in situ in womanwith clinically mild PID but should be removed in cases of severe disease.
ADNEXAL MASSES
Adnexal masses (ovarian cysts, ovarian torsion and tubo-ovarian abscesses) are notuncommon presentations in the emergency department with a typical complaint ofacute pelvic pain. Ultrasound is the best diagnostic tool to evaluate these masses. Itcan detect the presence of free pelvic fluid, differentiate solid from cystic structuresand identify masses with septations and internal structures. Blood flow to the ovarycan also be assessed: it is diminished in the case of ovarian torsion and increased inmalignancy.
Tubo-ovarian abscess
Tubo-ovarian abscess (TOA) is not a disease entity, but rather a finding on the spec-trum of PID. It can occur in up to one third of patients admitted with PID.59
Table 4. Center for disease control treatment regimens.
Parenteral regimen A
A Cefotetan, 2 g i.v. every 12 h; OR cefoxitin 2 g i.v. every 6 h AND doxycycline, 100 mg, p.o. or i.v.
every 12 h.Parenteral regimen B
A Clindamycin 900 mg i.v. every 8 h AND gentamicin, loading dose i.v. or i.m. (2 mg/kg) followed by
maintenance dose (1.5 mg/kg) every 8 h (single daily dose may be substituted)Alternative parenteral regimens
A Ofloxacin 400 mg i.v. every 12 h; OR levofloxacin 500 mg i.v. every 24 h with OR without metro-
nidazole 500 mg i.v. every 8 h; OR ampicillin/salbactam 3 g i.v. every 6 h AND doxycycline 100 mg
p.o. or i.v. every 12 hOral regimen A
A Oflaxacin 400 mg p.o. twice daily for 14 days; OR levofloxacin 500 mg p.o. daily for 14 days with OR
without metronidazole 500 mg p.o. twice daily for 14 daysOral regimen B
A Ceftriaxone 250 mg i.m. in a single dose; OR cefoxitin 2 g i.m. in a single dose AND probenecid, i5
p.o. in a single dose; OR other 3rd generation cephalosporin PLUS doxycycline, 100 mg p.o. twice
daily for 14 days with OR without metronidazole 500 mg p.o. twice daily for 14 days
i.m., intramuscularly; i.v., intravenously; p.o., by mouth.
Emergency gynaecology 739
Pelvic ultrasound plays an important role in providing answers with regard to man-agement. While the ovaries and fallopian tubes are easily discernable anatomically,their borders are less well defined by ultrasound and they are therefore commonlyreferred to as tubo-ovarian complexes. Pelvic ultrasound has a sensitivity of 93%and a specificity of 98% in the diagnosis of TOA.60 Treatment includes analgesia, fluidsand intravenous antibiotics and although 60e80% will resolve without surgical inter-vention, it is prudent to identify those individuals who require immediate surgical in-tervention, viz. those with ruptured TOA with generalised peritonitis and septic shock,because delay can be fatal.61
Adequate drainage of purulent fluid and intravenous antibiotics are key to success-ful treatment. Ultrasound-guided drainage, drainage via colpotomy,62 percutaneousdrainage63 and vaginal drainage64 have been performed, but the results are not consis-tent. In one study,65 ultrasound-guided transvaginal aspiration of pyosalpinges andtubo-ovarian abscesses in 15 patients was found to be effective, simple and safe andsymptoms and signs subsided in all women within 3 days. Single step ultrasound-guidedaspiration in conjunction with the installation of intracavity antibiotics successfullyavoided additional surgical intervention.66
EARLY PREGNANCY LOSSES
Vaginal bleeding with associated abdominal pains is a common complaint in the first 20weeks of pregnancy. Miscarriage is the most common complication of pregnancy, withup to 50% of all conceptions failing,67 most occurring before the women is aware thatshe is pregnant. Miscarriages are known to occur in 15% of confirmed pregnancies,with most occurring prior to the 12th week.68 If cardiac activity is detected by ultra-sound early in the pregnancy, the chance of miscarriage is reduced to 5.5%.69 Althoughthe introduction of ultrasound has revolutionised the appreciation of pregnancy,there has been very little change in clinical management over the last 50 years andmanagement is still based on traditional guidelines rather than evidence-basedmedicine.
The process of miscarriage ranges from threatened miscarriage (TM), progressingto inevitable miscarriage and culminating in complete miscarriage. The patient witha TM usually presents with vaginal bleeding with no clots and may have associatedlower abdominal pains. Ultrasound plays a critical role in prognosticating the outcomeof the pregnancy, namely:
1. An empty gestational sac with a diameter of at least 15 mm at 7 weeks and 21 mmat 8 weeks has a diagnostic accuracy of 90.8% in predicting miscarriage in womanwho have symptoms.70 A mean sac diameter of 17 mm without an embryo or13 mm without a yolk sac can predict non-viable gestation with a specificity of100% and a positive predictive value of 100%.71
2. Fetal heart activity and the lack of adverse prognostic factors conveys a variableprognosis. It is expected that once the fetal pole is 5 mm long, fetal heart activityshould be visible with transvaginal ultrasound.72 Most prospective series reporta fetal loss rate of 3.4e5.5% if bleeding occurs after fetal heart rate activitystarts.73,74 In one study, the identification of fetal activity by ultrasound carrieda 97% likelihood for the pregnancy continuing beyond 20 weeks.75 However, mis-carriage rates of 20e31% has been reported in other studies where fetal heartactivity was identified.76,77
740 S. R. Ramphal and J. Moodley
3. Discrepancy between gestational age and crown rump length (CRL) as well as fetalbradycardia are also adverse prognostic factors.78,79 A prospective study was per-formed using transvaginal ultrasound on 270 pregnant patients for bleedingbetween 5 and 12 weeks. Using logistic regression analysis on risk factors(fetal bradycardia, discrepancy between gestational sac and crown lump length,and discrepancy between menstrual and sonographic age by more than oneweek), the risk of abortion was increased from 6% when none of the factorswere present to a maximum of 84% when all were present.70
4. The presence of subchorionic haematoma on ultrasound has contentious prognos-tic value with regard to the outcome of the pregnancy. Although a large separationhas been associated with a threefold increased risk of miscarriage (19% versus71%) in women with bleeding,80 the presence or the size of the haematoma hasno bearing on miscarriage rates (10% versus 11%) and the rate of prematuredelivery was the same in patients with and without haematoma (11%) in anotherprospective series.81
Maternal serum biochemistry has also been proposed as a predictor of pooroutcome in women with TM. Women with TM in their first trimester who eventuallymiscarry have lower serum HCG values compared with women continuing the preg-nancy and with asymptomatic pregnant women.76 In a prospective controlled study,a single serum free b-HCG measurement taken in early pregnancy with a cut-off valueof 20 ng/ml was found to differentiate between the ’normal’ (controlled and threat-ened continuing) and the ‘abnormal’ (non-continuing TM and tubal) pregnancieswith 88.3% sensitivity and a positive predictive value of 82.6%.82 Similarly, an evaluationof 358 women presenting with vaginal bleeding in the first 18 weeks of pregnancyfound that a single progesterone value of <45 nmol/l was able to differentiate betweenabnormal and normal (ongoing) pregnancy with a sensitivity of 87.6% and a specificityof 87.5%.83
Other serum markers such as serum inhibin A, activin A, Ca125 and pregnancy-associated placental protein have also been evaluated as markers in the outcome ofTM. Studies have reported low and poor predictive values limiting the practical useof these tests.84
Once the patient is diagnosed with a TM the attending medical staff should explainthe miscarriage process to the patient. Exercise need not be discouraged and bed restshould not be totally advocated.85 Although there is no definite evidence that bedrest can affect the course of pregnancy, abstinence from the working environmentfor a few days will help women feel safer, thus providing emotional relief.86 Ina RCT evaluating the impact of bed rest on the course of TM, 61 women with vaginalbleeding and viable pregnancies at <8 weeks were randomly allocated to intramuscu-lar HCG, intramuscular placebo or bed rest. The miscarriage rates in the threegroups were 30%, 40% and 75%; significant differences between the HCG and bedrest groups, but not between the placebo and bed rest groups.87 Although theHCG group performed significantly better than the bed rest group, the lack of pro-found benefit over placebo, the concern about the potential development of ovarianhyperstimulation syndrome and the fact that TM may be the result of various condi-tions irrelevant to luteal function, prevented further testing and application of HCGtreatment in general gynaecological practice. An observational cohort study of 230women with TM in whom bed rest was recommended, showed that women whoadhered to this suggestion had a miscarriage rate of 9.9%, compared with 23.3% inwomen who continued their usual activities ( p¼ 0.006). However the lack of
Emergency gynaecology 741
randomisation and retrospective design of the outcome data collection preclude adefinite conclusion.88
According to published series,89,90 progesterone is prescribed in 13e40% ofwomen with TM in an attempt to support the deficient corpus luteum or induce re-laxation of a cramping uterus. A recent meta-analysis addressed the impact of proges-terone supplementation on miscarriage rates in various clinical settings but did notprovide a separate analysis for progesterone in TM.91 Two published papers92,93
have attempted to assess this relationship. One included three different regimens ofprogestogens. Having miscarriage as the outcome measure, the random effect risk ra-tio (RR) was 1.10 (95% CI¼ 0.92e1.31) for the progestogen group.92 In the otherstudy, which provided ultrasonographic evidence of fetal heart activity, the RR of mis-carriage was 1.09 (90% CI¼ 0.90e1.33) for the progestogen group.93 Hence given theavailable data and poor quality of evidence, progestogens do not appear to improvethe outcome in TM.
EARLY PREGNANCY UNITS
Early pregnancy units (EPUs) evolved as an efficient way of organising the services re-quired to manage complications in early pregnancy.94 They were designed initially as‘fast tracking’ diagnostic and evacuation services for those with miscarriages, but be-cause of the increasing numbers of women who present with early pregnancy bleedingand who seek reassurance, these units have developed into full early pregnancy carefacilities. They have become useful since they provide an efficient use of resources95
and reduce the demands made on the acute gynaecological team and the need for in-patient beds. They are also well accepted and attractive to patients, providing informa-tion and reassurance to those using these clinics. Many publications confirm thephysiological impact of early pregnancy loss on women, their partners and families.For some individuals, the distress is severe and protracted, even when the miscarriageoccurs early in the first trimester96 and these units provide expert psychologicalsupport.
Surgical management of miscarriage
Surgical uterine evacuation for the management of miscarriage usually involves vacuumaspiration or sharp metal curettage. Since the 1800s, surgical uterine evacuation hasbeen the standard treatment offered to women who miscarry, based on an assumptionthat retained tissue runs the risk of infection and haemorrhage. Sharp metal curettageis a procedure not without complications and, aside from the small anaesthetic risk, ithas also been associated with a 4e10% rate of early complications including infection,bleeding, cervical trauma and uterine perforation.97,98 In addition, it has been associ-ated with long term complications including decreased fertility and abnormal menstru-ation.99 A RCT comparing suction curettage with sharp curettage concluded thatsuction curettage was safer and easier.100 In a recent Cochrane review, vacuum aspi-ration was found to be safer, quicker to perform and less painful than sharp curettage,as shown by statistically significant findings of decreased blood loss (RR¼ 0.28, 95%CI¼ 0.10e0.73), decreased perception of pain (RR¼ 0.74, 95% CI¼ 0.33e0.90)and a shorter duration of the vacuum aspiration procedure (�1.2 mins weightedmean difference, 95% CI¼�1.5e0.87 mins).101 However, the conclusions from thisreview might be limited by the small sample sizes in the trials and the large loss to
742 S. R. Ramphal and J. Moodley
follow up in one of the trials analysed. It has also been suggested that vacuum aspira-tion is more cost effective than sharp curettage.98
Surgical evacuation under local anaesthesia is well described and commonly prac-ticed in the USA97 as well as Asian and African countries. This technique may beappropriate for selected women and its wide use needs further evaluation.
The value of routine antibiotics before surgical evacuation of the uterus in womanwith incomplete miscarriage is controversial, however most clinicians use antibiotics ifthere are signs of infection. A Cochrane review found little research on the routineuse of prophylactic antibiotics in incomplete miscarriages.102 The few RCTs thathave been conducted do not provide evidence that prophylactic antibiotics decreasethe rate of post abortal sepsis.
Non-surgical management of miscarriage
Recently some studies have reported the effectiveness of expectant treatmentwhereas others have explored the role of medical treatment with various prostaglan-dins (PG) or progesterone agents in evacuating the uterus without surgical curettage inearly miscarriages. In a randomised study comparing medical and surgical evacuation,20% of women expressed a strong preference for medical management.103 The rea-sons given for these choices were avoidance of general anaesthesia and a feeling of be-ing more in control. In a relatively small open study involving 43 women and addressingthe non-surgical management of incomplete miscarriages, a 95.3% success rate wasachieved using a single 0.5 mg dose of intramuscular Sulprostone or a 400 mg doseof oral misoprostol.104 The authors concluded that it might be a cost saving alterna-tive. More recently, a RCT105 evaluated medical versus surgical therapy in first trimes-ter miscarriages; a total of 94 women were randomised to either 600 mg misoprostolor surgical curettage in an under-resourced setting. Overall success rate of medicaltherapy was 91.5% compared to 100% in the surgical arm at evaluation with ultrasoundon day 7. However, another randomised study106 showed only a 13% successful evac-uation in the medical arm after a single 400 mg dose of misoprostol compared to 97%in the surgical group. A drawback of this latter study was that the protocol called fora surgical evacuation 12 h after the misoprostol dose if the products of conceptionwere not fully expelled, thus underestimating the potential of this form of treatmentin that successful outcomes could have occurred given more time. One randomisedtrial showed no statistical difference in efficacy between surgical and medical evacua-tion for miscarriage and early fetal demise at gestation <71 days or sac diameter<24 mm.103 Although there was a reduction in clinical pelvic infection after medicalevacuation (7.1 versus 13.2; p< 0.001), patient acceptability for both methods wasequal. Further research needs to be done in order to support the role of medical ther-apy in spontaneous miscarriages.
Observational and controlled trials of expectant versus surgical management showsa wide variation in reported efficacy (25e100%).107e109 In a review of nine studies, thesensitivity analyses comparing the RCToutcome with that from observational studies ofexpectant treatment showed a lower average of success of 79.5% for the RCT sub-group of 122 pooled patients.110 This difference appears to be due to the successful out-come rate of 79% reported by Nielsen & El Hahlin from their 103 patients.107 Of theirexpectant treatment sub-group, 21% had a dilatation and curettage if a transvaginal scan(TVS) after 3 days showed products of of more than 15 mm in anterior posterior (AP)diameter. It could well be that the 79% success figure might have been higher had these
Emergency gynaecology 743
patients been followed expectantly for a longer period. Success in this study was definedas the absence of products of conception by TVS 3 days after the patients sought care.Chipchase & James108 included all patients with an AP diameter of<50 mm and showeda 94.7% success rate. Their study was small (19 patients in the expectant group) and pa-tients were clinically reviewed on three occasions up to 6 months. Furthermore, themean AP diameter of tissue was 11 mm, which would have been defined as ‘completemiscarriage’ by Nielsen & Hahlin and excluded from their study. The investigators inthese two studies admit that their sonographic inclusion criteria are arbitrary and, todate, it is still unknown what size limit for the products of conception, as viewed byTVS, is the appropriate upper cut-off point for expectant treatment. Rulin et al111 re-ported that ultrasound assessment of the uterine cavity showing heterogenous shadow-ing with an AP diameter of 15 mm or less is less likely to confirm genuine retainedproducts of conception histologically. These cases of ‘complete miscarriages’ are bestmanaged conservatively, since there is a trend towards fewer complications comparedto surgical management (3.0 versus 5.8%, p¼ 0.06).112 In a recent randomised, double-blind, placebo controlled trial comparing medical and expectant management of first tri-mester miscarriage, 104 women were randomised to either 600 mg of vaginal misopros-tol or intravaginal placebo.113 The primary outcome measure was complete miscarriageby day 7 as shown by an endometrial thickness of �15 mm on TVS. The success rate ofmedical management was 88.6% compared with 44.2% in the expectant group. In thesubgroup analysis, for women diagnosed with incomplete miscarriage there was no dif-ference in the success rate (100% versus 85.8%), while for women with early pregnancyfailure there was a success rate of 87% with misoprostol compared with 29% in theexpectant management group (OR¼ 15.96; 95% CI¼ 5.26e48.37). The complete mis-carriage rate was achieved more quickly in the medical group than in the expectantgroup by day 1 (32.7 versus 5.8%) and day 2 (73.1 versus 13.5%) of treatment,respectively.
It is prudent that medical and expectant management be offered only in units wherepatients have 24 h access. Although medical management results in resorption of re-tained tissue, bleeding may occur if tissue is passed. With medical evacuation, up to30% will miscarry or bleed in the priming phase. With expectant management, therole of serum b-HCG has not yet been defined.
In early fetal demise or delayed miscarriage, where there is the presence of a closedcervix, priming with anti progesterone or the administration of a higher dosage of PGswith longer duration of use may be a more effective regime. Recently a large study of in-duced abortion showed that combination therapy (prostaglandin and anti-progesteroneagent) was more effective than a single agent.114
In a Cochrane review115 addressing medical methods for first trimester miscar-riage, the authors concluded the following:-
1. The most common combined regimen (mifepristone/misoprostol) is an effectiveand safe method for termination of pregnancy in the first trimester. The effectof mifepristone did not seem to be affected by lowering the dose from the previ-ously recommended 600 mg to 200 mg when combined with misoprostol of atleast 400 mg. In earlier studies, it had been shown that the linear doseeresponseeffect of mifepristone did not occur in doses above 100 mg.116
2. With regard to the role of gestational age, there was insufficient data to confirmthe findings of a World Health Organisation (WHO) Task Force that there is adoubling of failure rates when comparing miscarriages at less than 7 weeks tothose at 9 weeks or more.117
744 S. R. Ramphal and J. Moodley
3. A combination regimen with PG is more effective than PG alone. When split intosub-groups of early and late first trimester, this effect was not apparent but thismay be due to the small number in each group. Similarly, mifepristone alone isless effective than a combination regimen with PG.
4. Different routes of application have been used and vaginal application of misopros-tol seems to be superior to oral administration, being more effective and havingfewer side effects.
SUMMARY
Ectopic pregnancy (EP), acute pelvic inflammatory disease (PID) and miscarriages arecommon conditions seen in the emergency room. The use of sonography, in particular,the vaginal probe, modern pregnancy testing, laparoscopic techniques and newer an-tibiotics have led to early diagnosis offering the best chances for avoiding risks tolife, health and fertility. Furthermore, there is now more robust evidence, based onrandomised controlled trials (RCTs) and meta-analysis, that allows clinicians and theirpatients to decide on expectant, medical and conservative surgical management tominimise morbidity while preserving physical, reproductive and emotional wellbeing.However, there is a need for prospective randomised trials to address managementissues relating to medical therapy for EP and PID, including treatment options, antibi-otic choices, duration of treatment and evaluation of different methods in the manage-ment of first trimester miscarriages.
Practice points
A If an ectopic pregnancy (EP) is suspected, then it is mandatory that it isexcluded.
A A combination of transvaginal ultrasound and serum human chorionic gonad-otrophin (HCG) measurements can reliably diagnosis EP pregnancies.
A Laparoscopic surgery, rather than open surgery, is now the main method ofaccess in haemodynamically stable EP.
A b-HCG levels should be followed up after salpingostomy until they areundetected.
A Medical therapy has an established place in the treatment of EP in selectedpatients.
A Early recognition and empirical antibiotic treatment of pelvic inflammatorydisease (PID) is essential to reduce sequelae of chronic pain and infertility.
A Although laparoscopy is the gold standard in the diagnosis of PID, its routineuse is not practical.
A In threatened miscarriage (TM), little evidence supports the use of prosta-glandin therapy and bed rest.
A Medical abortion is non-invasive, safe and an effective alternative to earlysurgical abortion in selected cases.
Emergency gynaecology 745
REFERENCES
1. Sackett DL, Rosenberg WMC, Gray JAM et al. Evidence based medicine: what it is and what it isn’t.
British Medical Journal 1996; 312: 71e72.
2. Field MJ & Lohr KN. Guidelines for Clinical Practice. Institute of Medicine. Washington DC: National
Academy Press, 1992, p. 34.
3. Goldner TE, Lawson HW, Xia Z & Atrash HLK. Surveillance of ectopic pregnancy: United States
4. Sperling RS, Friedman Jr. F, Joyner M et al. Seroprevalence of human immunodeficiency virus in women
admitted to the hospital with pelvic inflammatory disease. Journal of Reproductive Medicine 1991; 36:
122e124.
5. Goldstein SR, Snyder GR, Watson C & Danon M. Very early pregnancy detection with endovaginal
ultrasound. Obstetrics and Gynecology 1988; 72: 200e204.
6. Cartwright PS. Diagnosis of ectopic pregnancy. Obstetrics and Gynecology Clinics of North America 1991;
18: 19e37.
7. Read MD & Bigrigg MA. The management of gynaecological emergencies. Current Obstetrics and
Gynaecology 1994; 4: 98e103.
8. Lindblom B, Hahlin M & Sjoblom P. Serial human chorionic gonadotrophin determinations by fluoroim-
muno-assay for differentiation between intrauterine and ectopic gestation. American Journal of Obstetrics
and Gynecology 1989; 161: 397e400.
9. Cacciatore B, Stenman UH & Ylostalo P. Diagnosis of ectopic pregnancy by vaginal ultrasonography in
combination discriminatory serum HCG level of 1000IU/L(IRP). British Journal of Obstetrics and
Gynaecology 1990; 97: 904e908.
10. Mol B, Lijmer T, Ankum W et al. The accuracy of single serum progesterone measurement in the di-
agnosis of ectopic pregnancy: a meta-analysis. Human Reproduction 1998; 13: 3220e3227.
*11. Garry R. Towards evidence-based management of tubal ectopic pregnancies. Gynaecological Endoscopy
1999; 8: 65e70.
Research agenda
A Randomised controlled trials (RCTs) are needed to assess the benefits andharm of expectant management, medical management and surgery in ectopicpregnancies (EPs). All studies should address the long-term outcomes offertility, repeat EP, quality of life assessment and cost-effective analysis.
A RCTs are required to assess the role of salpingotomy and salpingectomy insurgically treated patients.
A Research on appropriate management (surgical or medical) in patients withpersistent EP after salpingostomy.
A Well-designed studies of diagnostic techniques in pelvic inflammatory disease(PID) are urgently needed (these will require a high level of organisational andlogistical skill).
A Better quality studies (RCTs) are required to evaluate bed rest and proges-terone support in threatened abortion after controlling for confoundingvariables.
A Further research is needed to clarify parameters and indications for expec-tant management in early miscarriage.
A Research on the side-effects, bleeding patterns, acceptability, financial impactand quality of life outcomes of the different methods of treatment for firsttrimester miscarriages.
746 S. R. Ramphal and J. Moodley
12. Murphy AA, Nager CW, Wujek JJ et al. Operative laparoscopy vs laparotomy for the management of
ectopic pregnancy: a prospective trial. Fertility and Sterility 1992; 57: 1180e1185.
13. Vermesh M, Silva PD, Rosen GF et al. Management of ruptured ectopic gestation by linear salpingos-
tomy: a prospective randomized clinical trial of laparoscopy vs laparotomy. Obstetrics and Gynecology
1989; 73: 400e404.
14. Lundorff P, Thorburn J, Hahlin M et al. Laparoscopic surgery in ectopic pregnancy; a randomized trial
vs laparotomy. Acta Obstetricia et Gynecologica Scandinavica 1991; 70: 343e348.
15. Lundorff P, Hahlin M, Kallfelt B et al. Adhesion formation after laparoscopic surgery in tubal preg-
nancy: a randomized trial versus laparotomy. Fertility and Sterility 1991; 55: 911e915.
16. Yao M & Tulandi T. Current status of surgical and nonsurgical management of ectopic pregnancy.
Fertility and Sterility 1997; 67: 421e433.
17. Bangsgaard N, Lund C, Ottesen B & Nilas L. Improved fertility following conservative sur-
gical treatment of ectopic pregnancy. British Journal of Obstetrics and Gynaecology 2003; 110:
765e770.
18. Pouly JL, Mahnes H, Mage G et al. Conservative laparoscopic treatment of 321 ectopic pregnancies.
Fertility and Sterility 1986; 46: 1093e1097.
19. Maymon R, Shulman A, Halperin R et al. Ectopic pregnancy and laparoscopy - Review of of 1197
patients treated by salpingectomy or salpingostomy. European Journal of Obstetrics, Gynecology, and
Reproductive Biology 1995; 62: 61e67.
20. Dubuisson JB, Morice B, Chapron C et al. Salpngectomy - the laparoscopic surgical choice for ectopic
pregnancy. Human Reproduction 1996; 11: 1199e1203.
21. Seifer DB. Persistent ectopic pregnancy: an argument for heightened vigilance and patient compliance.
Fertility and Sterility 1997; 68: 402e404.
22. Saraj A, Wilcox J, Najmabadi S et al. Resolution of hormonal markers of ectopic gestation: a random-
ized trial comparing single dose intramuscular MTX with salpingostomy. Obstetrics and Gynecology
1998; 92: 989e994.
23. Sowter M, Farquhar C, Petrie K & Gudex G. A randomized trial comparing single dose systemic meth-
otrexate and laparoscopic surgery for the treatment of unruptured tubal pregnancy. British Journal of
Obstetrics and Gynaecology 2001; 108: 192e203.
24. Sowter M, Farquhar C & Gudex G. An economic evaluation of single dose systemic methotrexate and
laparoscopic surgery for the treatment of unruptured ectopic pregnancy. British Journal of Obstetrics
and Gynaecology 2001; 108: 204e212.
25. Kooi S & Kock HCLV. A review of literature on nonsurgical treatment in tubal pregnancies. Obstetrical
and Gynecological Survey 1992; 47: 739e749.
26. Nazac A, Gervaise A, Bouyer J et al. Predictors of success in methotrexate treatment of women with
unruptured tubal pregnancies. Ultrasound in Obstetrics and Gynecology 2003; 21: 181e185.
27. Fernandez H, Bourger P, Ville Y et al. Treatment of unruptured tubal pregnancy with methotrexate:
pharmacokinetic analyses of local vs intramuscular administration. Fertility and Sterility 1994; 62:
943e947.
28. Hajenius PJ, Engelsbel S, Mol BW et al. Randomised trial of systemic methotrexate versus laparoscopic
salpingostomy in tubal pregnancy. Lancet 1997; 350: 774e779.
29. Henry MA & Gentry WL. Single injection of methotrexate for treatment of ectopic pregnancies.
American Journal of Obstetrics and Gynecology 1994; 171: 1584e1587.
30. Stovall TG & Ling FW. Single dose methotrexate: an expanded clinical trial. American Journal of Obstet-
rics and Gynecology 1993; 168: 1759e1765.
*31. Sowter M & Frappell J. The role of laparoscopy in the management of ectopic pregnancy. Reviews in
Gynaecological Practice 2002; 2: 273e282.
32. Lipscomb G, McCord M, Stovall T et al. Predictors of success of methotrexate treatment in women
with ectopic pregnancies. The New England Journal of Medicine 1999; 341: 1974e1978.
33. Lipscomb C, Bran D, McCord M & Portera J. Analyses of three hundred and fifteen ectopic pregnan-
cies treated with single dose methotrexate. American Journal of Obstetrics and Gynecology 1998; 178:
1354e1358.
34. Parker J, Bisits A & Proietto AM. A systematic review of single-dose intramuscular methotrexate for
the treatment of ectopic pregnancy. Australian and New Zealand Journal of Obstetrics and Gynaecology
1996; 88: 775e778.
Emergency gynaecology 747
*35. Barnhart K, Gosman G, Ashby R & Sammel M. The medical management of ectopic pregnancy: a meta-
analysis comparing single dose and multidose regimens. Obstetrics and Gynecology 2003; 101: 778e784.
36. Gamzu R, Almog B, Levin Y et al. Efficacy of methotrexate treatment in extrauterine pregnancies de-
fined by stable or increasing human chorionic gonadotrophin concentrations. Fertility and Sterility 2002;
77: 761e765.
37. Potter M, lopine L & Jamieson D. Predictors of success with methotrexate treatment of tubal ectopic
pregnancy at Grady Memorial hospital. American Journal of Obstetrics and Gynecology 2003; 188: 1192e
1194.
38. Jacobson L & Westrom L. Objectivized diagnosis of acute pelvic inflammatory disease. American Journal
of Obstetrics and Gynecology 1969; 105: 1088e1098.
*39. Munday PE. Pelvic inflammatory disease - an evidence-based approach to diagnosis. The Journal of
Infection 2000; 40: 31e41.
*40. Simms I, Warburton F & Westrom L. Diagnosis of pelvic inflammatory disease: time for a rethink.