UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
Bleeding in the first trimester of pregnancy
Wieringa-de Waard, M.
Publication date
2002
Link to publication
Citation for published version (APA):
Wieringa-de Waard, M. (2002). Bleeding in the first trimester of pregnancy.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.
Managementt of miscarriage
aa randomised controlled trial of expectant
managementmanagement versus surgical evacuation
Margreett Wieringa-de Waard, Jeroen Vos, Gouke J. Bonsel,Patrickk J. E. Bindels, Willem M. Ankum Submittedd for publication
Abstract t
OBJECTIVESS To determine safety and efficacy of either expectant management or surgicall evacuation for miscarriages.
DESIGNN Randomised controlled trial. SETTINGG Two Amsterdam hospitals.
PARTICIPANTSS Patients with a non-viable pregnancy or incomplete miscarriage at a gestationall age of less than 16 completed weeks.
INTERVENTIONN Allocation to either expectant management or surgical evacuation. Patientss who refused randomisation were managed according to their own preference. MAINN OUTCOME MEASURES Safety, complications, efficacy, success of intended treatment,, duration of clinical symptoms.
RESULTSS One hundred and twenty-two patients were randomised and 305 were managedd according to their choice. No differences were found in the number of emergencyy curettages and complications between expectant management and sur-gicall evacuation between the randomised groups. Intention-to-treat analysis showedd similar efficacy (92% versus 100%) for the randomised groups after six weeks.. The success rate of intended treatment at six weeks was 30/64 (47%) in womenn allocated to expectant management with a median time to evacuation after inclusionn of 19 days. After seven days, 37% of women in this group had a sponta-neouss complete miscarriage. Results in the preference groups were comparable withh the randomised groups.
CONCLUSIONSS In randomised patients the efficacy of expectant management and surgicall evacuation is similar. A waiting period of seven days after diagnosis may preventt about 40% of surgical procedures without loss of safety.
Introduction n
Inn many countries surgical uterine evacuation is the standard treatment for womenn with a miscarriage. Expectant management has been advocated as an alternativee in several observational studies in a primary care setting.1 2 Just onee randomised clinical trial compared both management options in a hospi-tall setting, suggesting the outcomes to be similar.3 However, the duration of expectantt management in this study was restricted to only three days. From aa clinical and health policy point of view we regard this information as impor-tantt but insufficient to base guidelines and individual decisions upon. A longer periodd of expectant management and more information on the preference of patientss are necessary. Patients' preferences might even play a decisive role iff no substantial differences exist in the effectiveness, costs and availability of bothh treatment modalities.
Wee conducted a randomised controlled trial comparing expectant management withh surgical evacuation. Eligible women who refused randomisation because off a strong preference for either surgery or expectant management were man-agedd according to their choice, and evaluated similarly to randomised patients. Byy using this particular study design, we aimed to enhance the generalisability off our findings.
Methods s
Protocol Protocol
Thee study was conducted between April 1998 and September 2000 in two Am-sterdamm hospitals: the Academic Medical Center and the Onze Lieve Vrouwe Gasthuis.. General practitioners working in the health district covered by these twoo hospitals were asked to refer women with first-trimester vaginal bleeding forr an ultrasound assessment. In addition, all women attending the emergen-cyy departments or the outpatient clinics of both hospitals because of first-trimesterr vaginal bleeding were also asked to participate.
Patientss with an established diagnosis of a non-viable pregnancy or incomplete miscarriagee at a gestational age of less than 16 completed weeks were includ-edd in the study. Transvaginal sonographic criteria for non-viability were: a meann gestational sac diameter > 15 mm without a measurable embryonic pole,, an embryo without cardiac activity, or a gestational sac diameter < 15 mm,, not showing any growth after a seven-days interval.4-5 An incomplete miscarriagee was diagnosed in case of ultrasound evidence of retained products off conception (RPOC) > 15 mm anteroposterior (AP) diameter. All transvaginal-scanss were performed by trained physicians using a transvaginal 6.5 MHz sonographicc probe (Hitachi Corporation, Tokyo, Japan).
Exclusionn criteria for enrolment in the study were: under 18 years of age, in-abilityy to understand the Dutch or English informed consent form, and/or severee bleeding, pain or fever necessitating immediate surgical evacuation. Thee study was approved by the medical ethics committees of both hospitals.
Assignment Assignment
Afterr written informed consent had been obtained, patients were randomly al-locatedd by the attending physician to expectant management or surgical eva-cuationn (randomised groups) using central electronic randomisation. Randomisationn was stratified for referral setting (directly by GPs versus out-patientt clinics) and for gestational age (4-8, 8-12, and 12-16 weeks of amen-orrhea).. Eligible women who expressed a strong preference for one of the treatmentt options and refused informed consent for randomisation, were invitedd to participate in the observational study and received the treatment off their choice (preference groups). These women were asked to consent to thee same follow-up procedures as applied in randomised patients.
InterventionsInterventions and follow-up
Surgicall uterine evacuation using suction curettage was performed within a weekk after inclusion in the study under local or general anaesthesia in day-timee surgery. Planning of surgery depended on the availability of theatre facilitiess only and was independent of group assignment (randomised or pref-erence).. Local anaesthesia was attained by paracervical injection of scandi-caine,, after premedication with intravenous atropine. Occasionally, this was combinedd with midazolam for sedation. General anaesthesia was used when-everr cardio-pulmonary monitoring was required or when requested by the patient.. General anaesthesia was attained by the administration of propofol intravenously.. The cervical canal was dilated to a maximum of Hegar 12. Vacuumm aspiration was done using 8, 10, or 12 mm curettes (Rocket of Londonn Ltd, UK). Rhesus negative patients received 625 IU anti-D immuno-globulin.. Patients left the hospital after two to four hours of postoperative observation.. Expectant management involved bi-weekly scheduled visits to thee outpatient clinic. Further management in this group depended on clinical developments.. Women who became impatient while being managed expec-tantly,, and requested surgical evacuation as yet, were scheduled to undergo curettagee within a week.
Alll women (randomised and preference groups) were assessed clinically and sonographicallyy during the bi-weekly appointments until a complete evacua-tionn of the uterus was established either by surgical evacuation or through spontaneouss loss. Women had access to a telephone consultation at all times, andd emergency admission could be arranged, if necessary.
Too identify long term complications, active follow-up was continued for a periodd of up to three months.
DataData collection
Duringg the first visit the attending physician collected base line data on clinical signss and symptoms, obstetric history and gestational age. Additional informa-tionn on symptoms and sociodemographic data was collected by means of a struc-turedd questionnaire (two languages, Dutch and English, available on request). Alll patients were asked to report the amount of bleeding, the degree of abdominall pain and the ability to work in a standardised diary. Bleeding was registeredd daily on a validated pictorial blood loss assessment chart; pain was scoredd on a visual analogue scale.6
Duringg the bi-weekly visits, diaries were taken in, and patients received instructionss about the diary for the next interval.
OutcomeOutcome measures
Wee applied the following hierarchy in comparing the outcomes of both strate-gies:: safety, complications, efficacy, success rate and duration of vaginal bleedingg and pain.
Safety:Safety: this was defined as the absence of excessive bleeding requiring a
bloodd transfusion, or, in case of curettage, the absence of uterine perforation necessitatingg laparotomy.
Complications:Complications: we considered excessive bleeding (> 500 cc, not requiring a
bloodd transfusion), ascending genital tract infection, cervical tear and uterine perforationn (not necessitating laparotomy) as short-term complications. Intra-uterinee synechiae (Asherman's syndrome) demonstrated during hys-teroscopyy was considered a long-term complication. Whenever this complica-tionn was suspected on clinical grounds, i.e. in case of hypomenorrhea, dys-menorrheaa or amenorrhea, a hysteroscopy was performed.7
Efficacy:Efficacy: expectant management and surgical evacuation were considered to
havee been effective in case of sonographic evidence of RPOC < 15 mm (AP diameter)) at six weeks after inclusion in the study. The analysis followed the intention-to-treatt principle.
SuccessSuccess of intended treatment: expectant management was considered to be
successfull if a spontaneous loss had occurred within six weeks. Surgical evac-uationn was successful if the curettage was performed without the need of a repeatt curettage within six weeks. Women with a spontaneous loss before the scheduledd curettage were considered as failures.
DurationDuration of clinical symptoms: this was based on patients' self- reported
symp-tomss as recorded in their diary. Emergency curettage was defined as the need to performm an unscheduled curettage for severe vaginal bleeding or pain.
SampleSample size: assuming no substantial differences between the two treatments
demonstratee a 20% difference in success rate of intended treatment (65% for expectantt management and 85% for curettage). To reach a power of 0.80 we neededd 162 patients to be randomised.
StatisticalStatistical analysis
Thee randomised groups were primarily analysed according to the intention-to-treatt principle.
Safetyy and efficacy were analysed using descriptive statistics, with the appli-cationn of the t-test, Chi-square and Wilcoxon-Mann-Whitney tests as appro-priate.. For the analysis of time until evacuation and time until stop bleeding orr pain, we applied conventional survival analysis methods and appropriate comparativee tests (log-rank test).
Medianss are 50% cumulative probabilities as estimated with Kaplan-Meier analysiss unless stated otherwise.
Alll statistical analyses were repeated for the preference groups with applica-tionn of comparative statistical tests for descriptive purposes only.
Thee Statistical Package of the Social Sciences (SPSS, version 9.0) was used forr all analyses.
R e s u l t s s
ParticipantParticipant flow and follow-up
Outt of 1101 women referred for an early pregnancy assessment, 449 were eli-giblee for the study (Figure 4.1). Twenty-two women were excluded because off severe bleeding or pain necessitating immediate curettage. Of the 427 remainingg women, 122 accepted randomisation while 305 expressed their ownn treatment preference and gave consent for data collection and follow-up.
Transvaginal l sonography y (n=1101) )
r+ +
Non-viablee pregnancy (n=419) ) > > Incompletee miscarriage (n=30) ) Excludedd (n=652) viable pregnancy -- complete miscarriage -- other diagnosis Randomisation n (nn = 122) Inclusion n (n=427) )a a
Preference e (n=305) ) Excludedd (n=22) -- immediate curettage necessary y —ww Expectant (n=64) —mm Curettage (n=58) - kk Expectant (n=126) - kk Curettage (n=179)Tablee 4 . 1 . Patient characteristics at inclusion according to treatment allocation and preference.3 3 Randomisedd group Expectant t (nn = 64) Curettage e (nn = 58) Preferencee group Expectantt Curettage <nn = 126) (n = 179) Age-yr,Age-yr, mean Parity Parity 0 0 1 1 > 1 1 PreviousPrevious experience Noo previous miscarriage
orr curettage Priorr curettage
Priorr spontaneous miscarriage Priorr curettage and spontaneous miscarriage e Unknown n GestationalGestational age 32.1 1 322 (50.0) 211 (32.8) 111 (17.3) 377 (57.8) 177 (26.6) 77 (10.9) 33 (4.7) 33.4 4 222 (37.9) 222 (37.9) 144 (24.1) 333 (56.9) 188 (31.0) 55 (8.6) 22 (3.4) << 8 wk 8-122 wk 12-166 wk Uncertain n
IntactIntact gestational sac
IncompleteIncomplete miscarriage
VaginalVaginal bleeding present BleedingBleeding until
inclusion-medianmedian daysb
PainPain until inclusion-median dayst NativeNative country
Western-Europee and USA Africann country
Surinamm and Antilles Otherr and unknown
9 9 36 6 15 5 4 4 60 0 4 4 48 8 3 3 0 0 39 9 3 3 11 1 11 1 (14.1) ) (56.3) ) (23.4) ) (6.3) ) (93.8) ) (6.2) ) (75.0) ) (0-9) ) (0-2) ) (60.9) ) (4.7) ) (17.2) ) (17.2) ) 88 (13.8) 299 (50.0) 188 (31.0) 33 (5.2) 544 (93.1) 44 (6.9) 455 (77.6) 2.5(1-6) ) 0.5(0-3) ) 299 (50.0) 88 (13.8) 122 (20.7) 99 (15.5) 32.5 5 577 (45.2 377 (29.4 322 (25.5 32.2 2 877 (48.6 533 (29.7 399 (21.8 722 (57.1) 344 (27.0) 111 (8.7) 55 (4.0) 44 (3.2) 99 (7.1) 655 (51.6) 400 (31.7) 122 (9.5) 1133 (89.7) 133 (10.3) 9 33 (73.8) 22 (0-6) 00 (0-1) 6 33 (50.0) 99 (7.1) 2 33 (18.3) 311 (24.6) 966 (53.6 566 (31.3 99 (5.0 144 (7.8 44 (2.2 188 (10.1 9 33 (52.0 500 (27.9 188 (10.1 1700 (95.0 99 (5.0 1311 (73.2 22 (0-7 00 (0-2 933 (52.0 100 (5.6 311 (17.3 455 (25.1 aa
Values are numbers with percentages in parentheses.
bb
Analysis Analysis
RandomisedRandomised groups
Noo significant differences in p a t i e n t c h a r a c t e r i s t i c s were p r e s e n t b e t w e e n t h e twoo r a n d o m i s e d groups ( T a b l e 4 . 1 ) . Also, no difference in prior e x p e r i e n c e withh one of t h e m a n a g e m e n t options was o b s e r v e d .
Safetyy of both t r e a t m e n t arms was excellent: only one case of excessive bleed-ingg r e q u i r i n g a blood transfusion occurred in t h e e x p e c t a n t l y m a n a g e d group. T h ee complication rate in randomised patients was low and did not differ signifi-cantlyy between the two m a n a g e m e n t strategies ( 4 . 7 % versus 3.4%) (Table 4.2). Accordingg to i n t e n t i o n - t o - t r e a t analysis t h e efficacy r a t e s at 6 weeks did not differr significantly and r e a c h e d 9 2 % ( 5 9 / 6 4 ) in t h e e x p e c t a n t group versus 100%% in t h e c u r e t t a g e group (p = 0.06); m e d i a n time until c o m p l e t e e v a c u a -tionn was 7 days v e r s u s 5 days ( p < 0 . 0 0 1 ) , r e s p e c t i v e l y . T h e m e d i a n d u r a t i o n off b l e e d i n g was 17 days for e x p e c t a n t m a n a g e m e n t and 13 days for c u r e t t a g e (pp = 0.04) while pain lasted for a m e d i a n time of 14 v e r s u s 11 days ( p > 0 . 1 0 ) . Successs of i n t e n d e d t r e a t m e n t : in t h e group allocated to e x p e c t a n t m a n a g e -m e n tt 30 out of 64 wo-men ( 4 6 . 9 % ) actually u n d e r w e n t a s p o n t a n e o u s loss withinn six w e e k s , while a n o t h e r two women e x p e r i e n c e d a c o m p l e t e loss even l a t e r .. T h e other 32 (50%) u n d e r w e n t surgical e v a c u a t i o n , 25 ( 3 9 % ) on t h e i r ownn r e q u e s t and 7 ( 1 0 . 9 % ) as an e m e r g e n c y p r o c e d u r e b e c a u s e of i n t o l e r a b l e bleedingg or pain. >» » o o c c TO TO C C O) ) in in (fl l _o o (fl l 3 3 O O V V c c re re c c O O Q. . C/3 3 20 0 oo 4 0 -6 0 --80 0 "T" " 7 7
No.. of patients at Risk
Randomisedd 64 Preferencee 126 30 0 66 6 —— R a n d o m i s e d —— Preference T T T T 144 2 1 2 8 Dayss after inclusion
19 9 36 6 11 1 16 6 7 7 11 1 35 5 4 2 2 1 1
Figuree 4.2. Kaplan-Meier estimates of the time until evacuation for expectant
Tablee 4.2. Outcome measures according to treatment allocation and preference; intention to treatt analysis. Randomisedd group Expectantt Curettage (n=64)) ( n = 5 8 ) Preferencee group Paa Expectant Curettage (nn = 126) (n = 179) CurettageCurettage performed Secondd curettage Emergencyy curettage 322 (50.0) 48 (82.8) <0.001 22 (3.1) 3 (5.2) >0.10 77 (10.9) 6 (10.3) >0.10 611 (48.4) 154(86.0) 22 (1.6) 12 (6.7) 199 (15.1) 18 (10.1) Safety Safety
Bleeding,, transfusion needed
Complications Complications Haemorrhagee (>500cc) Cervicall tear Uterinee perforation Infection n Asherman'ss syndrome Total l
EfficacyEfficacy at six weeks 1 1 2 2 1 1 3 3 59 9 (1.6) ) (3.1) ) (1.6) ) --(4.7) ) (92.2) ) 1 1 1 1 2 2 58 8 --(1.7) ) --(1.7) ) (3.4) ) (100) ) >0.10 0 >0.10 0 >0.10 0 >0.10 0 >0.10 0 0.06 6 3 3 5 5 5 5 106 6 (2.4) ) (4.0) ) --(4.0) ) (84.1)c c --77 (3.9) --11 (0.6) 22 (1.1) 11 (0.6) 111 (6.2) 171(95.5)c c
MedianMedian time until evacuation^ 7 (3-16) 5 (2-7)
MedianMedian time to stop bleeding*3 17(10-26) 13(9-17)
<0.0011 10 (4-18) 5 (2-8) 0.044 22(15-35) 15(9-25)
MedianMedian time to stop pain1
SuccessSuccess of Intended treatmenttreatment at 6 weeks 144 (7-24) 11 (6-26) >0.10 300 (46.9) 45 (77.6) 0.001 177 (7-24) 11 (6-21) 533 (42.1) 141 (78.8) Numberss in parentheses are the percentages.
aa
Randomised groups.
bb
Median time after inclusion defined as the 50% probability of evacuation, cessation of bleeding or pain follow-ingg Kaplan-Meier estimation. In parentheses the 25% probability and 75% probability respectively.
cc
Women who were lost to follow-upp were treated as failures (12 in the expectant group, 7 in the curettage group). .
Thee median time to reach complete evacuation of uterine contents (including thee waiting time of censored patients undergoing curettage) analysed accord-ingg to the intended treatment was 19 days, while 37% of the women had a spontaneouss loss within seven days (Figure 4.2).
AA successful evacuation of the uterus was reached in 45 of the 58 patients al-locatedd to surgical evacuation (77.6%). In ten women (17%) a spontaneous loss occurredd before the scheduled curettage. If these 10 women were not consid-eredd failures of the intended treatment, the success rate of surgical evacuation wass 93.8% (45/48). A second curettage was needed in three cases (5.2%) be-causee of incompleteness of the first procedure. The rate of emergency curet-tagess and second curettages did not differ between the randomised groups.
PreferencePreference groups
Base-linee characteristics between randomised patients and those managed ac-cordingg to their preference did not differ (Table 4.1). With regard to safety, bleedingg requiring a blood transfusion occurred three times. The complica-tionn rate was 4% for expectant management versus 6.2% for surgical evacua-tionn (Table 4.2).
RandomisedRandomised versus preference groups
Thee efficacy (intention-to-treat analysis) was lower in the preference groups thann in the trial (expectant: 84.1%, curettage: 95.5%), because 12 (9.5%) womenn in the expectant group and 7 (3.9%) in the curettage group were lost too follow-up and were considered as failures in the present analysis. The rate off repeat curettages, emergency curettages and complications was similar in alll groups. The success rates of intended treatment of expectant versus surgi-call management in the preference groups (42.1% versus 78.8% respectively) weree comparable to those in randomised patients.
AdditionalAdditional findings in all treatment groups
Approximatelyy 25% of the women in each group did not have vaginal bleeding att inclusion. The outcomes (efficacy and complications), however, were iden-ticall to those in women with manifest vaginal bleeding at inclusion.
Discussion n
Wee studied the outcome of expectant management and surgical evacuation in ann unselected cohort of women with a diagnosis of a non-viable pregnancy or incompletee miscarriage. By conducting a randomised trial within this group, andd by concomitantly accounting for the data of the observational study of womenn who refused randomisation, we aimed to increase the generalisability off our findings to the normal practice situation. We compared within and betweenn the randomised and preference groups and did not find any
differ-encess in the rate of serious adverse events, emergency curettages, second curet-tagess or complications.
Intention-to-treatt analysis showed the efficacy in both randomised arms to be similarr (92% in the expectant and 100% in the curettage group). In patients treatedd according to their preference the efficacy was lower, i.e. 84% versus 95%,, mainly because women who were lost to follow-up were considered to bee failures.
Whenn analysed according the intended treatment, our study shows expectant managementt to be safe, while a waiting period of seven days after diagnosis resultedd in a spontaneous loss in 37% of women.
Ann unexpected finding was the high proportion of women expressing a strong preferencee for one of the management options. As a consequence, the 162 randomisedd women required according to our power calculation was not reachedd within the pre-set study period. As a result, formally our study shouldd be regarded underpowered. However, in women, managed according theirr own preference, our findings were similar to those in the randomised groupss with regard to efficacy, effect of intended treatment and safety. While awaree that combining of the results of randomised and non-randomised patientss might be challenged, in this case of high similarity of both patient characteristicss and outcome data, we don 't expect the results to change with largerr numbers of randomised patients. One might argue to combine patient dataa to enhance power and to generalise findings.8"10
AA short-term complication, haemorrhage, was only registered as a complication off surgery, where the amount of blood loss is easy to assess. In the expectant managementt group, haemoglobin concentrations were measured only whenever indicatedd on clinical grounds; this did not result in any blood transfusions. Thee only randomised trial comparable to the present study, earlier published byy Nielsen and Hahlin, reported spontaneous complete miscarriages within threee days in 79% of patients allocated to expectant management.3 In con-trast,, we found only 47% during a much longer observation period. This dif-ferencee might be explained by a different interval between the onset of symp-tomss and inclusion of patients in their and our study. Time lapse between the onsett of bleeding and inclusion in our study was three days on average. This timee lapse is explained by the Dutch insurance system, where women with earlyy pregnancy problems require a formal referral by their GP to a fetal assessmentt unit.
Off the 1101 women in our study, 20% had already miscarried completely at base-linee and this might explain the difference with Nielsen and Hahlin. Probably,, they included patients at an earlier stage and possibly included moree patients with an incomplete miscarriage, a group virtually not repre-sentedd in our study.
Thee average duration of bleeding and pain was also longer compared to Nielsenn and Hahlin's findings. This might be explained by the selection of
p a t i e n t ss (only a few incomplete m i s c a r r i a g e s ) , t h e much longer period of ex-p e c t a n tt m a n a g e m e n t in our s t u d y and t h e use of a s t a n d a r d i s e d diary, in whichh p a t i e n t s t h e m s e l v e s , not t h e i n v e s t i g a t o r s , r e g i s t e r e d vaginal b l e e d i n g r e q u i r i n gg s a n i t a r y p r o t e c t i o n .
C h i p c h a s ee also performed a r a n d o m i s e d study of 35 women, all with r e t a i n e d p r o d u c t ss of conception with small d i a m e t e r s , which in our s t u d y w e r e not evenn included b u t r e g a r d e d as c o m p l e t e m i s c a r r i a g e s .1 1
W ee performed our curettages after several days, not immediately. This is stan-dardd practice in the N e t h e r l a n d s . Obviously, if curettages would have been per-formedd earlier, i.e. immediately after e n t r a n c e in t h e study, the n u m b e r of emergencyy curettages (not-planned curettage) would have been less.
Forr m a n y y e a r s , t h e a r g u m e n t of safety has b e e n used in justifying a surgical a p p r o a c hh in t h e m a n a g e m e n t of m i s c a r r i a g e s . H o w e v e r , several large obser-vationall studies have m a d e clear t h a t serious complications may result from surgicall e v a c u a t i o n , i.e. l a p a r o t o m y for u t e r i n e p e r f o r a t i o n .1 2 - 1 4
Inn conclusion, the success rate of expectant m a n a g e m e n t is less favourable t h a n earlierr r e p o r t e d by other studies. It fails in about half of cases, mainly b e c a u s e womenn are not p r e p a r e d to wait for more t h a n two weeks. E x p e c t a n t manage-m e n t ,, however, is safe and up to 4 0 % of surgical procedures can be p r e v e n t e d byy awaiting the n a t u r a l course of events within the first week. T h e generalis-abilityy of these findings is probably high, because we r e p o r t e d on all consecu-tivee p a t i e n t s with first-trimester bleeding from a well defined population. E x p e c t a n tt m a n a g e m e n t of m i s c a r r i a g e can be offered as a safe option to well-informedd women who wish to avoid s u r g e r y .
References s
1.. Wiebe E, Janssen P. Management of spontaneous abortion in family practices and hospitals.. Fam Med 1998;30:293-6.
2.. Ambulatory Sentinel Practice Network. Spontaneous abortion in primary care. J Am Boardd Fam Pract 1988;1:15-23.
3.. Nielsen S, Hahlin M. Expectant management of first-trimester spontaneous abortion. Lancett 1995;345:84-6.
4.. Coulam CB, Goodman C, Dorfmann A. Comparison of ultrasonographic findings in spontaneouss abortions with normal and abnormal karyotypes. Hum Reprod 1997;12:823-6. .
5.. Deaton JL, Honore GM, Huffman CS, Bauguess P. Early transvaginal ultrasound followingg an accurately dated pregnancy: the importance of finding a yolk sac or fetal heartt motion. Hum Reprod 1997;12:2820-3.
6.. Higham JM, O'Brien PM, Shaw RW. Assessment of menstrual blood loss using a pic-toriall chart. Br J Obstet Gynaecol 1990;97:734-9.
7.. Westendorp IC, Ankum WM, Mol BW, Vonk J. Prevalence of Asherman's syndrome afterr secondary removal of placental remnants or a repeat curettage for incomplete abortion.. Hum Reprod 1998;13:3347-50.
8.. Pocock SJ, Elbourne DR. Randomised trials or observational tribulations? N Engl J Medd 2000;342:1097-9.
9.. Mant D. Can randomised trials inform clinical decisions about individual patients? Lancett 1999;353:743-6.
10.. Torgerson D, Sibbald B. Understanding controlled trials: What is a patient preference trial?? BMJ 1998;316:360-2.
1 1 .. Chipchase J, James D. Randomised trial of expectant versus surgical management of spontaneouss miscarriage. Br J Obstet Gynaecol 1997;104:840-1.
12.. Johnson N, Priestnall M, Marsay T, Ballard P, Watters J. A randomised trial evaluat-ingg pain and bleeding after first trimester miscarriage treated surgically or medical-ly.. Eur J Obstet Gynecol Reprod Biol 1997;72:213-5.
13.. Hinshaw KKS. Medical management of miscarriage. In: Grudzinskas JG, O'Brien PMS.. Problems in early pregnancy. London: RCOG Press, 1997;284-95.
14.. Ankum WM, Wieringa-de Waard M. Bindels PJE. Management of spontaneous mis-carriagee in the first trimester: an example of putting informed shared decision making intoo practice. BMJ 2001;322:1343-6.
