Uterine artery embolization vs hysterectomy in the treatment ......GENERAL GYNECOLOGY Uterine artery embolization vs hysterectomy in the treatment of symptomatic uterine fibroids:
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
G
UofSE
Oabc
Seoa
Re
Cfi
SspcstlrWmitzm
FaAMh
R
RM
Sa
0
Research www.AJOG.org
ENERAL GYNECOLOGY
terine artery embolization vs hysterectomy in the treatmentf symptomatic uterine fibroids: 5-year outcomerom the randomized EMMY trialanne M. van der Kooij, MD; Wouter J. K. Hehenkamp, MD, PhD; Nicole A. Volkers, MD, PhD;rwin Birnie, PhD; Willem M. Ankum, MD, PhD; Jim A. Reekers, MD, PhD
(pslp
Ca
K
BJECTIVE: The purpose of this study was to compare clinical outcomend health related quality of life (HRQOL) 5 years after uterine artery em-olization (UAE) or hysterectomy in the treatment of menorrhagiaaused by uterine fibroids.
TUDY DESIGN: Patients with symptomatic uterine fibroids who wereligible for hysterectomy were assigned randomly 1:1 to hysterectomyr UAE. Endpoints after 5 years were reintervention rates, menorrhagia,nd HRQOL measures that were assessed by validated questionnaires.
ESULTS: Patients were assigned randomly to UAE (n � 88) or hyster-
broids: 5-year outcome from the randomized EMMY trial. Am J Obstet Gynecol 201
tsidUmsmrt(r(ph
See Journal Club, page 186
28.4%) had undergone a hysterectomy because of insufficient im-rovement of complaints (24.7% after successful UAE). HRQOL mea-ures improved significantly and remained stable until the 5-year fol-ow-up evaluation, with no differences between the groups. UAE had aositive effect both on urinary and defecation function.
ONCLUSION: UAE is a well-established alternative to hysterectomybout which patients should be counseled.
ey words: fibroid tumor, hysterectomy, menorrhagia, uterine artery
ctomy (n � 89). Five years after treatment 23 of 81 UAE patients embolization
ite this article as: van der Kooij SM, Hehenkamp WJK, Volkers NA, et al. Uterine artery embolization vs hysterectomy in the treatment of symptomatic uterine
0;203:105.e1-13.
tgasasowtgotostHtbt
MSTddvww
ymptomatic uterine fibroids aredisabling and are associated with
ignificant morbidity that affects ap-roximately 20-40% of women ofhildbearing age.1 The most commonymptom of uterine fibroids for whichreatment is sought is heavy or pro-onged menstrual bleeding, which mayesult in iron deficiency anemia.2
hen symptoms progress and phar-acotherapeutic options fail, surgical
ntervention may be necessary. Duringhe last decade, uterine artery emboli-ation (UAE) has been greeted as ainimally invasive treatment alterna-
rom the Departments of Radiology (Drs vannd Gynecology (Drs van der Kooij, Hehenkamsterdam, The Netherlands; and the Instituedical Center, Rotterdam (Dr Birnie), The N
ospitals are listed with the full-length article
eceived July 8, 2009; revised Oct. 22, 2009; a
eprints: Sanne M. van der Kooij, MD, Academieibergdreef 9, 1105 AZ Amsterdam, The Neth
upported by ZonMw “Netherlands Organizatiopplication no: 945-01-017) and by Boston Scie
ive for surgery in the reduction ofymptoms of heavy menstrual bleed-ngs caused by fibroids. Several ran-omized controlled trials comparedAE with hysterectomy and/or myo-ectomy and found similarly good re-
ults for both interventions up to 24onths of follow up.3-6 Earlier, we
eported on the 2-year results fromhe embolization vs hysterectomyEMMY) trial and compared clinicalesults,7 health-related quality of lifeHRQOL) outcomes,8 and meno-ausal symptoms9 between UAE andysterectomy. After 2 years the chance
Kooij and Volkers and Prof Dr Reekers), and Ankum), Academic Medical Center,f Health Policy and Management, Erasmus
herlands. The EMMY trial participants andww.AJOG.org.
r Health Research and Development” (Grantc Corporation, The Netherlands.
, Inc. • doi: 10.1016/j.ajog.2010.01.049
n
AUGUST 2010 Americ
o avoid a hysterectomy in the UAEroup was 76.5% while menorrhagiand HRQOL improved significantly,imilarly in both groups. Both UAEnd hysterectomy affected ovarian re-erve in women �45 years old. Basedn these 2-year follow-up results, UAEas considered to be a good alternative
o hysterectomy. Because fibroids mayrow back, menorrhagia can recur, orther symptoms that warrant hysterec-omy may emerge beyond the 2-yearsf follow-up period. Herefore, we ob-erved our cohort until 5 years afterreatment and investigated clinical andRQOL results between UAE and hys-
erectomy as well as outcomes betweenaseline and 5-year follow-up in pa-ients from the EMMY trial.
ATERIALS AND METHODStudy designhe EMMY study is a multicenter, ran-omized controlled trial, that was con-ucted in The Netherlands. Patients whoisited the gynecologic outpatient clinicsere invited to participate if they (1)ere premenopausal, (2) were diag-
enorrhagia, (4) had no other treatmentptions than a hysterectomy, and (5) hado desire for future pregnancy. Afterritten informed consent was obtained,atients were allocated randomly (1:1)
LAVH, laparoscopically assisted vaginal hysterectomy; LH, lapDerived, with permission, from Hehenkamp et al.8
a Data are given as mean � SD; b Data are given as median (
van der Kooij. UAE vs hysterectomy for uterine fibroids. A
atisfaction with the received treatment. c
tudy measuresatients received questionnaires at base-
ine and at fixed intervals for 2 years afterreatment.8 In addition, for logistic rea-ons, all patients received 1 question-aire in the autumn of 2007, at a median
ollow up period of approximately 5ears after primary treatment, which re-ulted in a variable follow-up evaluationn both groups. In tables and figures, the
edian of 5 years is depicted as a fixedoint in time for both treatment arms,espite this variation within the group.ll questionnaires were identical, except
ictorial Chart, Euro-Quality of Life-5,ealth Utilities Index Mark 3, sexual ac-
ivity, and body image questionnaires toptimize the response rate.10-15 The fol-
owing subjects were evaluated in the-year questionnaire: additional inter-entions between 2- and 5-year followp evaluation (in case of nonrespon-ents, the patients’ general practitionersere contacted by telephone to check for
ny additional procedures), menstrualharacteristics (intensity and regularityince UAE or no complaints because of
enopause; only in the UAE group), andeveral HRQOL measures that were as-
enopausal symptoms were queried byhe Kupperman score as modified by
iklund et al.16 Scores may range from
FIGURE 1Trial profile
rofile represents the flow of patients through thRI, magnetic resonance imaging; UAE, uterine artery embolizati
an der Kooij. UAE vs hysterectomy for uterine fibroids. Am J
–51; higher scores represent more seri- p
05.e4 American Journal of Obstetrics & Gynecolo
us menopausal symptoms. In additione inquired whether patients believed
hemselves to be in or beyond meno-
ial.
tet Gynecol 2010.
ause. Generic HRQOL was assessed by t
gy AUGUST 2010
he Medical Outcome Study Short Form6 (SF-36).17,18 The SF-36 can be dividednto the physical component summarycore (PCS) and the mental componentummary score (MCS).19 Scores mayange from 0 –100 (100 indicates the op-imal score) and were validated for theutch population. The Urogenital Dis-
ress Inventory (UDI)20,21 was used tonvestigate urinary symptoms. UDIcores range from 0 –100; higher scoresndicate worse functioning. For defeca-ion complaints, the Defecation Distressnventory (DDI) was used,22 with scoresanging from 0 –100; higher scores areess favorable. Patients were asked to ratehe overall quality of their urinary andtool function as “very good,” “good,”fairly good,” “not good or bad,” “fairlyad,” “bad,” or “very bad.” Further-ore, the use of pads for urinary incon-
inence or of laxatives was registered.atients were asked to indicate how sat-
sfied they were with the received treat-ent: “very satisfied,” “satisfied,” “fairly
atisfied,” “not satisfied or unsatisfied,”fairly unsatisfied,” “unsatisfied,” orvery unsatisfied.” We also inquiredhether patients would recommend therimary treatment to a friend. Finally,e asked women whether they would in-eed have chosen the assigned treatmentgain if they had had the opportunity too so.
tatistical analysisnalyses were done with SPSS statistical
oftware (version 16.0; SPSS Inc, Chi-ago, IL). Study outcomes were analyzedccording to original treatment assign-ent (intention to treat). Reinterven-
ions were also analyzed according toer-protocol analysis. A Kaplan-Meierurve was constructed to show the distri-ution of the secondary hysterectomiesver time. Differences in categoric dataere compared with �2 test or Fisher’s
xact tests, if appropriate. The Student test (or Mann-Whitney test, when appli-able) assessed differences in numericata. A probability value of � .05 wasonsidered statistically significant. Pre-ictors for failure (secondary hysterec-omy) were tested by logistic regressionnalysis. In this analysis, baseline charac-
e tron.
Obs
eristics (Appendix) were included for
www.AJOG.org General Gynecology Research
TABLE 2Reinterventions in UAE and hysterectomy group until 2 and 5 years after initial treatment
Primaryintervention Secondary intervention Reason for intervention
13-1 Diagnostic hysteroscopy with curettage Postmenstrual blood loss 12.......................................................................................................................................................................................................................................................................................................................................................................
van der Kooij. UAE vs hysterectomy for uterine fibroids. Am J Obstet Gynecol 2010. (continued )
AUGUST 2010 American Journal of Obstetrics & Gynecology 105.e5
muvacptMstqfbsrtM(eTaPwaa�sa
RPP2t
lgtwcftmtom[mmwny3y
CRImUhap2h5ps
owwwpydr1vibucdotg
BTtm5epih(gwb
m J O
Research General Gynecology www.AJOG.org
1
ultiple regression analysis whenevernivariate analysis revealed probabilityalues of � .1. In the multiple regressionnalysis, a probability value of � .05 wasonsidered statistically significant. Re-eated measurements analysis was usedo evaluate longitudinal differences in
CS, PCS, DDI, UDI, and Wiklundcores between treatment strategies withime as repeated factor. Self-reporteduality of urinary and stool function atollow-up evaluation was compared withaseline and yielded 1 of 3 possible an-wers: worse, the same, or better. Logisticegression analysis was performed to testhe impact of improvement in SF-36
CS and PCS on satisfaction at 5 years“very satisfied” and “satisfied” vs “mod-rately satisfied” and “very unsatisfied”).o evaluate the impact of baseline vari-bles (Appendix) on the change in MCS,CS, UDI and DDI at 5 years comparedith baseline, multiple linear regression
nalysis was performed for those vari-bles that yielded probability values of
.1 in the univariate analysis. Nonre-ponders were not included in thenalyses.
ESULTSatientsatients were enrolled between March002 and February 2004. In the hysterec-
UAE, uterine artery embolization.Derived, with permission, from Volkers et al.7
van der Kooij. UAE vs hysterectomy for uterine fibroids. A
omy group, 75 women received the al- U
05.e6 American Journal of Obstetrics & Gynecolo
ocated treatment vs 81 in the UAEroup. Table 1 lists the baseline charac-eristics of the participating patients,hich include myoma characteristics; all
haracteristics were not significantly dif-erent. Figure 1 shows the flow of pa-ients through the trial: 93% of the
ailed 5-year questionnaires were re-urned, with a median follow-up periodf 59 months, ranging from 47-73onths (UAE: median, 60 months
range, 49 –73 months]; hysterectomy:edian, 58 months [range, 47–71onths]). The median age of all patientshen responding to the 5-year question-aire was 50 years, ranging from 39-63ears (UAE: median, 49 years [range,9 – 63 years]; hysterectomy: median, 49ears [range, 40 –59 years]).
linical outcomeeinterventions
n addition to the 19 secondary hysterecto-ies (23.5%) that were performed in theAE group in the first 2 years,7 another 4ysterectomies were required between 2nd 5 years, all because of insufficient im-rovement of bleeding complaints (Table). This adds up to a total of 23 secondaryysterectomies after a median follow up ofyears (28.4%). Distribution over time isresented in Figure 2. Per protocol analysishowed that, after a technically successful
ndary hysterectomy (24.7%). Of the 10omen who underwent unilateral UAE, 3omen underwent a hysterectomy, allithin the first 2 years of follow up. Multi-le regression analysis of failures within 5ears revealed only a higher body mass in-ex to be associated with failed UAE (oddsatio, 1.12; 95% confidence interval [CI],.02–1.24; P � .02). All additional inter-entions that were performed after UAE,ncluding hysterectomies, are listed in Ta-le 2. Twelve women in the UAE groupsed medication (tranexamic acid/oralontraception/levonorgestrel intrauterineevice) to remedy still symptomatic men-rrhagia (Table 3). After 5 years 8 of 75 ofhe women (10.7%) in the hysterectomyroup needed reintervention (Table 2).
leeding characteristicsable 3 shows various bleeding charac-
eristics of the UAE group. On averageenorrhagia decreased over time. Afteryears, 67 of 81 women (82.7%) were
ither symptom-free (n � 58) or re-orted great (n � 4) or moderate (n � 5)
mprovement. Of the 58 women who stillad their uterus after 5 years, 44 women75.9%) were symptom free or reportedreat or moderate improvement; 8omen (13.8%) reported their menstrualleeding to be unchanged, compared with
ot experiencing menorrhagia anymoreecause of menopause.
enopauseatients were asked the question: do you
eel that you are in or beyond menopause?n the UAE group 34.6% of the womennd in the hysterectomy group 47.1% ofhe women answered “yes,” which is sig-ificantly different (P � .03). The meaniklund score for menopausal symptoms
f both treatments is plotted over time inigure 3. Within group analysis revealed aignificant increase in the hysterectomyroup from baseline to 5 years (P � .04).he UAE group did not show a significant
urve represents preservation of the uterus afterhe last follow-up moment.AE, uterine artery embolization.
an der Kooij. UAE vs hysterectomy for uterine fibroids. Am J
ween the groups after 5 years. g
uality of life outcomeseneric HRQOLigure 4, A and B, display mental healthnd physical health scores throughouthe study period for both groups. Resultso the 2-year follow-up evaluation wereescribed earlier.8 Repeated measurementnalysis shows no differences between theroups during the 5-year follow-up periodor both the MCS and PCS scores. Table 4hows the differences in PCS and MCS be-ween and within groups over time. With-n-group analysis in the hysterectomyroup revealed significantly worse physicalealth after 5 years compared with 2 yearsP � .01), although mental health re-ained stable (P � .34). Within the UAE
f the uterus after UAE
. Censored means that a patient was lost to follo
tet Gynecol 2010.
roup, no differences over time were noted g
AUGUST 2010 Americ
MCS, P � .36; PCS, P � .18). In the mul-ivariate analysis, none of the baseline vari-bles was associated with improvement ofF-36 MCS scores at 5 years. The increasen SF-36 PCS score after 5 years was influ-nced positively by a hemoglobin level thatas �12.0 g/dL at baseline (� � 8.50; 95%I, 3.28–13.6; P � .002) and age (� �0.51 per year; 95% CI, –0.87 to –0.15;� .006): having a low hemoglobin level
t baseline resulted in more improvementn PCS; older women had less improve-
ent in PCS scores at 5 years.
rinary and defecation functionigure 4, C, depicts urinary functionUDI). Repeated measurements analysishowed no differences between the
p and had not undergone a hysterectomy before
n o
UAE w u
Obs
roups after 5 years. After 6 months, the
an Journal of Obstetrics & Gynecology 105.e7
UcctfeFtmtwlm
htsat–(p(PtsDpt(nt(bpf7aerc(yc
piygt(
SAs(8gsalh(h7Uwa[Uht
CTlcteowbTaitltbivwItedtsnfi
TmSrU
vA
m J O
Research General Gynecology www.AJOG.org
1
DI score stabilized in both groups at aontinuously significant higher levelompared with baseline until 5 years af-er treatment without any significant dif-erences between groups (Table 4). Def-cation function (DDI) is shown inigure 4, D. In the UAE group, a persis-ent significant improvement from 6
onths onward was found. In the hys-erectomy group, no significant changesere demonstrated compared with base-
ine. After 5 years, repeated measure-ents analysis showed the UAE group to
FIGURE 3Wiklund score
he graph represents the Wiklund score forenopausal symptoms until 5 years of follow up.cores range from 0–51; the higher scores rep-esent more serious menopausal symptoms.AE, uterine artery embolization.
an der Kooij. UAE vs hysterectomy for uterine fibroids.m J Obstet Gynecol 2010.
TABLE 3Menstruation changes to 5 years aVariable, n (%)
Because of hysterectomy..........................................................................................................
Because of self-reported menopause...................................................................................................................
UAE, uterine artery embolization.a Includes 23 hysterectomies.
van der Kooij. UAE vs hysterectomy for uterine fibroids. A
(
05.e8 American Journal of Obstetrics & Gynecolo
ave significantly better defecation func-ion than the hysterectomy group. Amaller number of fibroid tumors wasssociated with improved UDI scores af-er 5 years (� � –3.87; 95% CI, – 6.29 to1.44; P � .02); the intended treatmentembolization) was associated with im-rovement of the DDI score after 5 years� � –12.65; 95% CI, –22.08 to –3.22;
� .01). For the variables UAE andime, a significant interaction effect washown in the repeated measurements onDI scores. Urinary incontinence wasresent at baseline in 18.5% of UAE pa-ients vs 14.7% of hysterectomy patientsP � .52). After 5 years, urinary inconti-ence was reported by 27.2% of UAE pa-
ients vs 22.7% of hysterectomy patientsP � .31). After 5 years, most patients inoth groups reported a similar or im-roved quality of urinary or defecation
unction compared with baseline (UAE,0.4% and 67.9%; hysterectomy, 77.3%nd 61.3%), without significant differ-nces between groups (P � .20 and .61,espectively). The use of laxatives de-reased over time in the UAE group onlyfrom 9.7% at baseline to 1.3% at 2ears).8 However, after 5 years, an in-reased use of laxatives was found
roportion was stable at 5.8%, with anncrease after 5 years to 8.7%. After 5ears, 17.3% of women in the UAEroup and 10.0% of women in the hys-erectomy group used incontinence padsP � .23).
atisfactionfter 5 years, most patients were (very)
atisfied about the received treatmentie, 85.3% of women in the UAE vs8.6% of women in the hysterectomyroup (P � .37; Table 5). Logistic regres-ion analysis showed none of the vari-bles to be associated with satisfactionevels at 5 years after treatment. In theysterectomy group, 62 of 70 women88.6%) would advise a friend to have aysterectomy. In the UAE group, 61 of9 women (77.2%) would recommendAE to their friends (P � .07). Mostomen expressed a preference for the
ctual received treatment (56/79 women70.8%] from the UAE group preferredAE; 44/70 women [62.9%] from theysterectomy group preferred hysterec-omy; P � .10).
OMMENThis article describes the results of a
arge, long-term, randomized trial thatompared UAE with hysterectomy in thereatment of menorrhagia in the pres-nce of uterine fibroids. After a medianf 5 years, 23 hysterectomies (28.4%)ere performed in the UAE group, allecause of uncontrolled menorrhagia.he success rate of 71.6% (or 76.5% aftertechnically successful UAE procedure)
s comparable to prospective uncon-rolled single arm UAE studies23,24 butower than those reported in retrospec-ive studies.25-27 This may be explainedy the fact that our patients could partic-
pate only in the EMMY study when se-ere bleeding complaints were present,hile other treatment options had failed.
n contrast to other studies, all our pa-ients had a classic indication for hyster-ctomy and were indeed willing to un-ergo surgery. Because 82.6% (19/23) ofhe secondary hysterectomies in thistudy occurred within 2 years after UAE,ot undergoing a hysterectomy in therst 2 years after UAE might be a predic-
fte
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........
.........
or for being long-term hysterectomy
fSubfdewptlwctioicotbt
Uttstctods5tigohvp(rehfltpfstOofos
aqywbveptmmjitcstdgwHtwgacl
itldrfpdatoriesofipttgdtastv
AmrIDU
v
www.AJOG.org General Gynecology Research
ree. This corresponds with other data.23
ome studies have pointed out that fail-re rates are likely to increase wheneverilateral embolization cannot be per-ormed.25,28 In our study, 10 women un-erwent unilateral UAE. Three of themventually underwent hysterectomy, allithin the 2-year follow-up period. Thisercentage (3/10) of secondary hysterec-omies is more or less the same as in bi-ateral embolization in this study group,hich does not underline the higher per-
entage of failure in unilateral emboliza-ion. Our finding that a high body massndex at baseline is a predictor for failuref UAE is a new finding that might be of
mportance when a patient is beingounseled for UAE. However, the risk forbese patients to undergo surgery is ex-ensive also. The gynecologist shouldalance the risks and benefits of both op-ions with the individual patient.
Control of bleeding of women in theAE group who still had their uterus af-
er 5 years in our study was 75.9%. Al-hough failure was defined strictly as aecondary hysterectomy, 12 patients inhe UAE group still needed to use medi-ation to remedy menorrhagia, and 8 ofhem showed no improvement of men-rrhagia; these women are potential can-idates for reintervention. For this rea-on, we will approach them after another
years to establish the definitive long-erm failure rate of UAE in this trial. Ev-dently even a hysterectomy does notuarantee an intervention-free life; inur study, 10.7% of patients in theysterectomy group needed a reinter-ention, most of them because of com-lications caused by the hysterectomyadhesiolysis, a vesicovaginal fistula, oreconstruction surgery). Some studiesven demonstrate that women after aysterectomy are at higher risk for pelvicoor repair,29 which underlines the rein-
ervention risk after a hysterectomy to beresent. However, this could not be
ound in our study; the UDI did nothow a difference in complaints betweenhe UAE and the hysterectomy group.varian failure after UAE is a well-rec-gnized complication that may resultrom inadvertent embolization of utero-varian collateral vessels.30,31 Earlier we
howed both UAE and hysterectomy to o
ffect ovarian reserve.9 As a conse-uence, especially those women �45ears old seem to be at higher risk thanomen without UAE/hysterectomy forecoming menopausal after both inter-entions. In our 5-year analysis, how-ver, in the hysterectomy group, a higherercentage of patients reported subjec-ively to believe that they were beyond
enopause than in the UAE group. Thisight be explained by the absence of ob-
ective symptoms (ie, menstrual periods)n women who had a hysterectomy andherefore might be biased, although itan provide an indication. The Wiklundcore (evaluating the menopausal symp-oms instead of bleeding) did not show aifference in menopause between bothroups after 5 years. Comparing HRQOL,e showed that the main increase inRQOL occurred in the first months after
reatment8 and remained stable for 5 yearsithout showing differences between theroups. Hemoglobin level �12.0 g/dL andge at baseline were predictive of PCS in-rease; worse physical condition at base-ine (low hemoglobin level and being
FIGURE 4Graphs show scores until 5 years o
, Short Form 36 (SF-36 ): Mental Component Sean a better mental quality of life); B, Short Form
ange from 0 –100 (higher scores mean a betnventory (UDI ): scores range from 0–100 (higheistress Inventory (DDI ): scores range from 0–1AE, uterine artery embolization.
an der Kooij. UAE vs hysterectomy for uterine fibroids. Am J
lder) predicted a higher increase in phys- e
AUGUST 2010 Americ
cal condition after 5 years. Apparently,hese patients had the most to gain in theong-term. Higher age at baseline that pre-icted a worse physical condition probablyeflects the normal decline in physicalunction that comes with age. For clinicalractice, this substantiates that the best in-ication for treatment is heavy symptomst baseline. Evident differences betweenhe 2 groups in urinary function were notbserved during the 5-year follow-up pe-iod; both groups showed a comparablemprovement. A positive effect of hyster-ctomy on urinary function has been de-cribed before;32 the positive effect of UAEn urinary function, however, is a newnding. Defecation function only im-roved in the UAE group. After 5 years,his group had a significantly better defeca-ion function than the hysterectomyroup, which did not show an increase or aecrease in defecation function. The posi-ive effect of UAE on defecation function isnother new finding that has not been de-cribed before. The increasing use of laxa-ives in the UAE group from 1.3-13.3% is aery contradictory finding. The possible
ollow-up evaluation
mary: scores range from 0–100 (higher scores(SF-36 ): Physical Component Summary; scoresphysical quality of life); C, Urogenital Distresscores indicate worse functioning); D, Defecation(higher scores indicate worse functioning).
dditionally, the subjective change inverall quality of defecation, comparedith baseline, was not reported to be dif-
erent between both groups, which makeshe huge difference in (the objective) DDIfter 5 years difficult to ground. From 2ears onward, in both groups, a mild butignificant deterioration in urinary andefecation function was found, whichrobably can be ascribed to the effect of
ncreasing age on the prevalence of pelvicoordysfunction.34 Thenumberoffibroid
umors appeared to be the only predictoror UDI improvement, which probably isxplained by the mechanical effect of en-arged fibroid uteri. Satisfaction, treatmentreference (for the received treatment),ndrecommendationof thereceivedtreat-ent to a friend were all similarly good in
oth groups without differences betweenhem, which confirms earlier findings.35
hese findings support that patients per-eive both treatment alternatives as accept-ble options. Despite the accumulatingeneficial evidence from randomized tri-ls, the implementation of UAE as an alter-ative to hysterectomy in gynecologicractice is relatively slow. In Europe, it isstimated that �5% (Cardiovascular andnterventional Radiological Society of Eu-ope survey 2008, unpublished data) po-ential UAE candidates are being offeredhis alternative, and most women who un-ergo UAE seem to be those who discov-red this alternative to surgery through thenternet. In conclusion, UAE is a provenaluable treatment alternative for surgeryn women with symptomatic uterine fi-roids. In view of the currently availablevidence, the time is ripe to counsel allomen who are candidates for hysterec-
omy for their symptomatic uterine fi-roids on the possibility of UAE. In 5 years,
n 71.6% of all women who underwentAE, a hysterectomy was avoided, and
here was no difference in HRQOL be-ween groups. Besides this, one has to keepn mind that there might be a chance that,nstead of a hysterectomy, a less invasiventervention or the use of medication maye needed, which is reflected in the num-er of patients who reported their menor-hagia complaints as unchanged com-ared with baseline. However, because
these women chose not to have a second-
05.e10 American Journal of Obstetrics & Gynecology AUGUST 2010
aotomnom
AWgmRBC(S(TniAMRLKMNGOAsH(SLRHmMG
(avHusdBUt(RA
R1eF2Mtap3Us24PtidA5Hmft26fi
st7ktsfG8eboc(9eet1st1p1h21ttC1Sad1rwO1mqr1
m J O
www.AJOG.org General Gynecology Research
ry hysterectomy so far, the chance wouldnly increase that they would continue noto request one in the future. The mean agef our patient group is 50 years now, andenopause is looming. Of course, this is
ot a certainty and can be certified only bybservation of this patient group untilenopause. f
CKNOWLEDGMENTSe thank all participating patients, EMMY-trial
roup members and other contributors whoade the trial possible: W. Hehenkamp, J.eekers, W. Ankum, E. Birnie, M. Burger, G.onsel, and N. Volkers (Academic Medicalenter, Amsterdam); S. de Blok and C. de Vries
Onze Lieve Vrouwe Gasthuis, Amsterdam); T.alemans and G. Veldhuyzen van Zanten
Atrium Medical Center, Heerlen); D. Tinga and. Prins (Groningen University Hospital, Gro-ingen); P. Sleijffers and M. Rutten (Bosch Med-
cal Center, Den Bosch); M. Smeets and N.arts (Bronovo Hospital, The Hague); P. van deroer and D. Vroegindeweij (Medical Centerijnmond-Zuid, Rotterdam); F. Boekkooi and L.ampmann (St. Elisabeth Hospital, Tilburg); G.leiverda (Flevo Hospital, Almere); R. Dik and J.arsman (Gooi-Noord Hospital, Laren); C. deooijer, I. Hendriks, and G. Guit (Kennemerasthuis, Haarlem); H. Ottervanger and H. vanverhagen (Leyenburg Hospital, The Hague);. Thurkow (St. Lucas/Andreas Hospital, Am-terdam); P. Donderwinkel and C. Holt (Martiniospital, Groningen); A. Adriaanse and J. Wallis
Medical Center Alkmaar, Alkmaar); J. Hirdes, J.chutte, and W. de Rhoter (Medical Centereeuwarden, Leeuwarden); P. Paaymans and. Schepers-Bok (Hospital Midden-Twente,engelo); G. van Doorn, J. Krabbe, and A. Huis-an (Medisch Spectrum Twente, Enschede);. Hermans and R. Dallinga (Reinier de Graaf
TABLE 5Satisfaction until 5 years after UAE
Variable
12 mo7
UAE(n � 81)
Hy(n
Very satisfied 29 48...................................................................................................................
Very unsatisfied 1 0...................................................................................................................
UAE, uterine artery embolization.Derived, with permission, from Volkers et al.7
van der Kooij. UAE vs hysterectomy for uterine fibroids. A
asthuis, Delft); F. Reijnders and J. Spithoven d
Slingeland Hospital, Doetichem); W. de Jagernd P. Veekmans (St. Jans Gasthuis, Weert); P.an der Heijden, M. Veereschild, and J. van denout (Twenteborg Hospital, Almelo); I. van Se-meren, A. Heintz, R. Lo, and W. Mali, (Univer-ity Hospital Utrecht, Utrecht); J. Lind and Th.e Rooy (Westeinde Hospital, The Hague); M.ulstra and F. Sanders (Diakonessenhuistrecht, Utrecht); J. Doornbos (De Heel Hospi-
al, Zaandam); P. Dijkhuizen and M. van KintsRijnstate Hospital, Arnhem); Ph. Engelen and. Heijboer (Slotervaart Hospital, Amsterdam);. Dijkman (BovenIJ Hospital, Amsterdam).
EFERENCES. Buttram VC, Reiter RC. Uterine leiomyomata:tiology, symptomatology and management.ertil Steril 1981;36:4433-45.. Fraser IS, Critchley HO, Munro MG, Broder. Can we achieve international agreement on
erminologies and definitions used to describebnormalities of menstrual bleeding? Hum Re-rod 2007;22:635-43.. Edwards RD, Moss JG, Lumsden MA, et al.terine artery embolization versus surgery forymptomatic uterine fibroids. N Engl J Med007;356:360-70.. Hehenkamp WJ, Volkers NA, DonderwinkelF, et al. Uterine artery embolization versus hys-
erectomy in the treatment of symptomatic uter-ne fibroids (EMMY trial): peri- and postproce-ural results from a randomized controlled trial.m J Obstet Gynecol 2005;193:1618-29.. Mara M, Fucikova Z, Maskova J, Kuzel D,aakova L. Uterine fibroid embolization versusyomectomy in women wishing to preserve
ertility: preliminary results of a randomized con-rolled trial. Eur J Obstet Gynecol Reprod Biol006;126:226-33.. Pinto I, Chimeno P, Romo A, et al. Uterinebroids: uterine artery embolization versus ab-
pective, randomized and controlled clinicalrial. Radiology 2003;226:425-31.. Volkers NA, Hehenkamp WJ, Birnie E, An-um WM, Reekers JA. Uterine artery emboliza-ion versus hysterectomy in the treatment ofymptomatic uterine fibroids: 2 years’ outcomerom the randomized EMMY trial. Am J Obstetynecol 2007;196:519.e1-11.. Hehenkamp WJ, Volkers NA, Birnie E, Reek-rs, JA, Ankum WM. Symptomatic uterine fi-roids: treatment with uterine artery embolizationr hysterectomy: results from the randomizedlinical embolisation versus hysterectomyEMMY) trial. Radiology 2008;246:823-32.. Hehenkamp WJ, Volkers NA, Broekmans FJ,t al. Loss of ovarian reserve after uterine arterymbolization: a randomized comparison with hys-erectomy. Hum Reprod 2007;22:1996-2005.0. Higham JM, O’Brien PM, Shaw RW. As-essment of menstrual blood loss using a pic-orial chart. BJOG 1990;97:734-9.1. Brooks R. EuroQol: the current state oflay. Health Policy 1996;37:53-72.2. Rabin R, de Charro F. EQ-5D: a measure ofealth status from the EuroQol group. Ann Med001;33:337-43.3. Feeny D, Furlong W, Torrance GW. Multiat-ribute and single-attribute utility functions forhe health utilities index mark 3 system. Medare 2002;40:113-28.4. Stead ML, Crocombe WD, Fallowfield LJ.exual activity questionnaires in clinical trials:cceptability to patients with gynaecologicalisorders. BJOG 1999;106:50-4.5. Stead ML, Fountain J, Napp V. Psychomet-ic properties of the body image scale in womenith benign gynaecological conditions. Eur Jbstet Gynecol Reprod Biol 2004;114:215-20.6. Wiklund I, Holst J, Karlberg J, et al. A newethodological approach to the evaluation ofuality of life in postmenopausal women. Matu-itas 1992;14:211-24.7. Kupperman M, Varner RE, Summit RL Jr,
ysterectomy versus medical treatment onealth-related quality of life and sexual
unctioning: the medicine or surgery (Ms)andomized trial. JAMA 2004;291:1447-55.8. Hurskainen R, Teperi J, Rissanen P, AaltoM, Grenman S, Kivela A. Quality of life andost-effectiveness of levonorgestrel-releasing
ntrauterine system versus hysterectomy forreatment of menorrhagia: a randomised trial.ancet 2001;357:273-7.9. Ware JE Jr. SF-36 health survey update.pine 2000;25:3130-9.0. Shumaker SA, Wyman JF, Uebersax JS,cClish D, Fantl JA. Health-related quality of lifeeasures for women with urinary incontinence:
he Incontinence Impact Questionnaire and therogenital Distress Inventory: Continence Pro-ram in Women (CPW) research group. Qualife Res 1994;3:291-306.1. Uebersax JS, Wyman JF, Shumaker SA,cClish DK, Fantl JA. Short forms to assess lifeuality and symptom distress for urinary incon-inence in women: the Incontinence Impactuestionnaire and the Urogenital Distress In-
entory: Continence Program in Women (CPW)esearch group. Neurourol Urodyn 1995;14:31-9.2. Roovers JP, van den Bom JG, Huub van
er Vaart C, Fousert DM, Heintz AP. Does G
05.e12 American Journal of Obstetrics & Gyneco
ode of hysterectomy influence micturition andefecation? Acta Obstet Gynecol Scand001;80:945-51.3. Spies JB, Bruno J, Czeyda-Pommersheim F,agee ST, Ascher SA, Jha RC. Long-term out-
ome of uterine artery embolization of leiomyo-ata. Obstet Gynecol 2005;106:933-9.4. Lohle PN, Voogt MJ, de Vries J, et al. Long-erm outcome of uterine artery embolization forymptomatic uterine leiomyomas. J Vasc Intervadiol 2008;19:319-26.5. Gabriel-Cox K, Jacobsen GF, ArmstrongA, Hung YY, Learman LA. Predictors of hys-
erectomy after uterine artery embolization foreiomyoma. Am J Obstet Gynecol 2007;196:88.e1-6.6. Muller-Hulsbeck S. Long-term results afterbroid embolization. Radiologe 2008;48:660-5.7. Hirst A, Dutton S, Wu O, et al. A multi-centreetrospective cohort study comparing the effi-acy, safety and cost-effectiveness of hysterec-omy and uterine artery embolisation for thereatment of symptomatic uterine fibroids: theOPEFUL study. Health Technol Assess 2008;2:1-248.8. Park AJ, Bohrer JC, Bradley LD, et al. Inci-ence and risk factors for surgical interventionfter uterine artery embolization. Am J Obstet
ynecol 2008;199:671.e1-6. B
logy AUGUST 2010
9. Blandon RE, Bharucha AE, Melton LJ 3rd,t al. Incidence of pelvic floor repair after hys-erectomy: a population-based cohort study.m J Obstet Gynecol 2007;197:664.e1-7.0. Istre O. Management of symptomatic fi-roids: conservative surgical treatment modali-ies other than abdominal or laparoscopic myo-ectomy. Best Pract Res Clin Obstet Gynaecol008;22:735-47.1. Tulandi T, Sammour A, Valenti D, Child TJ,eti L, Tan SL. Ovarian reserve after uterine ar-
ery embolization for leiomyomata. Fertil Steril002;78:197-8.2. Narushima M, Otani T, Itoh Y, Kai S, Kondo, Hayashi H. Clinical effect of transabdominalimple hysterectomy on micturition function.inyokika Kiyo 1993;39:797-800.3. Chen GD. Pelvic floor dysfunction in agingomen. Taiwan J Obstet Gynecol 2007;46:74-8.4. Agdi M, Valenti D, Tulandi T, et al. Intraab-ominal adhesions after uterine artery emboli-ation. Am J Obstet Gynecol 2008;199:82.e1-3.5. Walker WJ, Barton-Smith P. Long-term fol-
ow up of uterine artery embolisation: an effec-ive alternative in the treatment of fibroids.
JOG 2006;113:464-8.
m J Obstet Gynecol 2010.
www.AJOG.org General Gynecology Research
APPENDIX
Predictors for failure
Age (continuous)...................................................................................................................
Ethnicity (white as reference category)...................................................................................................................
Body mass index (continuous)...................................................................................................................
Duration of menorrhagia symptoms (� or �1...................................................................................................................
Location of dominant fibroid (submucosal, subclassified)...................................................................................................................
Flow in dominant fibroid using ultrasound (hypor hypovascular)...................................................................................................................
T2 signal intensity on magnetic resonance imisointens, hypointens, mixed)...................................................................................................................
HRQOL, health related quality of life; MCS, Mental Componen
van der Kooij. UAE vs hysterectomy for uterine fibroids. A
Effect of baseline variables on HRQOL
Age (continuous).............................................................................................................................................................................................................................................................
Ethnicity (white as reference category).............................................................................................................................................................................................................................................................
Body mass index (continuous).............................................................................................................................................................................................................................................................
Any previous treatment (yes/no).............................................................................................................................................................................................................................................................
y) Duration of menorrhagia symptoms (continuously).............................................................................................................................................................................................................................................................
Anemia before treatment (yes/no).............................................................................................................................................................................................................................................................
No. of fibroid tumors (continuous).............................................................................................................................................................................................................................................................
Baseline SF-36 MCS (continuously, not on MCS change outcome).............................................................................................................................................................................................................................................................
Baseline SF-36 PCS (continuously, not on PCS change outcome).............................................................................................................................................................................................................................................................
t Summary; PCS, Physical Component Summary; SF-36, Short Form 36; UAE, uterine artery embolization.
AUGUST 2010 American Journal of Obstetrics & Gynecology 105.e13