ACTA REVIEW

Radical trachelectomy versus radical hysterectomy for thetreatment of early cervical cancer: a systematic reviewLI XU1, FU-QING SUN2 & ZAN-HONG WANG2

1Department of Respiratory Medicine, The First Affiliated Hospital of Chongqing Medical University, Chongqing, and2Department of Gynecology, The First Hospital, Shanxi Medical University, Taiyuan, PR China

Key wordsCervical cancer, radical trachelectomy, radicalhysterectomy, systematic review, controlledtrail

CorrespondenceZan-Hong Wang, Department of Gynecology,The First Hospital, Shanxi Medical University,Taiyuan, 030001, PR China.E-mail: wangzanhong@126.com

Conflict of interestThe authors have stated explicitly that thereare no conflicts of interest in connection withthis article.

Received: 21 January 2011Accepted: 21 June 2011

DOI: 10.1111/j.1600-0412.2011.01231.x

Abstract

Objective. To assess the efficacy and safety of radical trachelectomy (RT) and radicalhysterectomy (RH) for patients with early cervical cancer. Design. Systematic reviewwith meta-analysis. Population. Women who had early cervical cancer. Methods.Prospective controlled clinical trials comparing RT with RH were identified using apredefined search strategy. Recurrence, five-year recurrence-free survival rate, five-year overall survival rate, postoperative mortality, intraoperative and postoperativecomplications between the two operations were compared by using the methodsprovided by the Cochrane Handbook for Systematic Reviews of Interventions. Re-sults. Three controlled clinical trials involving 587 participants were included. Meta-analysis showed that there was no significant difference between the two groups inrecurrence rate [1.38; 95% confidence interval (CI) 0.58–3.28, p=0.47], five-yearrecurrence-free survival rate (1.17; 95% CI 0.54–2.53, p=0.69), five-year overallsurvival rate (0.86; 95% CI 0.30–2.43, p=0.78), postoperative mortality (1.14; 95%CI 0.42–3.11, p=0.80), intraoperative complications (1.66; 95% CI 0.11–25.28,p=0.72), postoperative complications (0.52; 95% CI 0.11–2.48, p=0.41), bloodtransfusion (0.29; 95% CI 0.06–1.36, p=0.12) and number of harvested lymphnodes. However, RT, compared with RH, reduced blood loss and shortened dura-tion to normal urine residual volume and postoperative hospital stay. Moreover,RT may achieve to normal conception rates, while RH makes patients sterile. Con-clusions. Radical trachelectomy has similar efficacy and safety to RH as the surgicaltreatment for early cervical cancer. Moreover, it reduced blood loss and shortenedthe duration to normal urine residual volumes and postoperative hospital stay. Rad-ical trachelectomy can be used to treat early stage cervical cancer as an alternativeoperation for patients who wish to preserve fertility.

Abbreviations: CCT, controlled clinical trial; CI, confidence interval; CLS, capillarylymphatic space; OR, odds ratio; RCT, randomized controlled trial; RH, radicalhysterectomy; RT, radical trachelectomy

Introduction

Worldwide, cancer of the cervix is the second most com-mon cancer in women and the leading cause of death amonggynecological cancers. Wertheim (1) first reported using ab-dominal hysterectomy to treat early cervical cancer in 1900.Nowadays, radical hysterectomy in combination with pelviclymphadenectomy has been extensively used to treat earlycervical cancer (2–6). With the use of cervical carcinomascreening, a decreased incidence and mortality from cervical

cancer and a noticeable shift from more advanced to earlierstage disease, together with late marriage, there will be anincreasing number of women who may wish to give birtheven if they are at risk of cervical cancer and thus possiblyrequiring hysterectomy. Approximately 15% of all cervicalcancers and 45% of surgically treated stage IB cancers occurin women under the age of 40years (7). Small invasive cancerhas become a more frequently encountered clinical problemand is often diagnosed in women who wish to preserve theirchildbearing prospects. It is a dilemma.

1200C© 2011 The Authors

Acta Obstetricia et Gynecologica Scandinavica C© 2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 1200–1209

A C TA Obstetricia et Gynecologica

L. Xu et al. Efficacy and safety of radical trachelectomy

Daniel Dargent performed the first vaginal radical trach-electomy associated with laparoscopic pelvic lymphadenec-tomy (LARVT or Dargent’s operation) in 1986 and describedthe approach in 1994 (8). This new surgical technique wasdesigned to effect a radical excision of the cervix (includ-ing the proximal part of parametrium and the upper vaginalcuff) while preserving the uterus corpus. In spite of initialdisbelief, this new approach was picked up by centers inEurope and North America (8–11). The treatment of morethan 900 women with early stage cervical cancer who subse-quently gave birth to around 200 children has been reviewed(12). Subsequently, radical trachelectomy (RT), a fertility-preserving surgical technique, has gained ever-increasingrecognition as a safe oncological alternative to radical hys-terectomy (RH) for early stage cervical cancer patients ofchildbearing age, and demand for this operation is likely togrow in the future (13,14). However, it has not yet been fullyevaluated as an alternative to RH. A new surgical approachfor cancer can only be fully accepted if it matches the stan-dard procedure, with regard to the extent of tissue removed,the number of harvested lymph nodes, the safety profile andthe survival outcomes. We assessed the existing literature andconducted a meta-analysis to compare the clinical effective-ness of these two operations.

Methods

Search strategy

The methodology followed the Quality of Reporting ofMeta-analyses (QUORUM) statement (15). The followingdatabases were searched systematically: MEDLINE, EM-BASE, PUBMED, SCIENCE-DIRECT, BIOSIS Previews, andthe China Biological Medicine Database (CBM) CNKI(China National Knowledge Infrastructure Whole ArticleDatabase) from January 1994 to November 2010, as wellas the Cochrane Central Register of Controlled Trials (CEN-TRAL) on the Cochrane Library issue 4, 2010. All controlledtrials comparing radical trachelectomy and radical hysterec-tomy in the surgical treatment of cervical cancer were iden-tified. In this review, the following search terms were used:cervical cancer, cervical tumor, cervical carcinoma, uterinecervical neoplasm, radical hysterectomy and radical trach-electomy. Both free text and a MeSH search for keywordswere used. The references for retrieved articles together withthe proceedings of relevant conferences were hand-searchedin order to identify other potentially eligible studies for in-clusion in the analysis and missed by the initial search orany unpublished data. Additional cross-searches were per-formed in MEDLINE and PUBMED using the names of in-vestigators who were the lead authors of at least one eligiblestudy.

Inclusion and exclusion criteria

The types of studies were comparative trials with recordedoncological outcomes (recurrence, five-year recurrence-freesurvival rate, five-year overall survival rate and postoperativemortality), operative results (intraoperative and postopera-tive complications) and obstetrical outcomes were included.Noncomparative studies, case series, case reports and studiesusing historical controls were excluded. Patients with Inter-national Federation of Gynecology and Obstetrics (FIGO)stage I–IIA cervical cancer were included from trials com-paring radical trachelectomy and radical hysterectomy. Trialsin which patients had distant metastasis, synchronous malig-nancy in other organs, serious cardiovascular or respiratorydisorders, hepatic or renal failure were excluded. In the caseof overlap or duplicate studies, we retained only the mostcomprehensive one. Trials were included. Trials comparingthe results of these two operations but where the patientshad also received preoperative or postoperative chemother-apy were excluded.

All outcomes were defined prior to the literature search.The primary outcomes were oncological results (recurrence,five-year recurrence-free survival rate, five-year overall sur-vival rate and postoperative mortality). The secondary out-comes were operative outcomes (operation time, blood lossduring operation, transfusion, intra-/postoperative compli-cations and so on) and obstetrical outcomes.

Data extraction and statistical analyses

All data were extracted independently by two reviewers (FQSand ZHW), and differences of opinion were resolved by con-sensus. Titles and abstracts were scanned first to make a list ofpossibly related literature, and then full texts were obtainedfor those articles identified as either relevant or not clear, butonly trials coincident with our predetermined criteria wereincluded.

As there was no consensus about quality assessment ofnonrandomized studies, the Cochrane Handbook for Sys-tematic Reviews of Interventions, in which criteria for non-randomized studies were the same as randomized controlledtrials (RCTs), was used to assess the methodological qual-ity of included studies (16). The criteria included six items,as follows: (1) adequate sequence generation; (2) allocationconcealment; (3) blinding, defined as such if results of mea-surements were masked and if regarded as of low risk withregard to both performance and detection bias (only wounddegree of surgery and postoperative morbidity included, notconsidering five-year survival rates and postoperative mor-tality); (4) incompletely addressed outcome data; (5) data freeof selective reporting; and (6) free of other bias. We consid-ered missing outcome data in the categories <10% as low risk

C© 2011 The AuthorsActa Obstetricia et Gynecologica Scandinavica C© 2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 1200–1209 1201

Efficacy and safety of radical trachelectomy L. Xu et al.

of bias, 10–15% as moderate risk of bias and >15% as highrisk of bias. The results of quality assessment were classifiedas follows: yes (low risk of bias); probably yes (moderate riskof bias); and no (high risk of bias). No rating of the studieswas performed; each was accepted or rejected based on thesix items noted above.

Using a predefined data extraction form, two reviewers(FQS and ZHW) extracted data about characteristics of in-cluded studies and baseline characteristics of patients inde-pendently, which included following items: the first author,publication date, country of the investigators, number ofpatients, interventions, age, recurrence, five-year recurrence-free survival rate, five-year overall survival rate, postopera-tive mortality, operation time, blood loss during operation,transfusion, number of harvested lymph nodes, postoper-ative hospital stay, duration to normal urine residual vol-umes and obstetrical outcomes. If necessary, the authors ofthe original articles were contacted for additional data. Finalagreement was achieved through discussion.

The individual and pooled statistics were calculated us-ing the fixed effect model (Mantel–Haenszel method) (17),but if a p-value of the heterogeneity test was less than0.1, a random-effect model was used. The results were ex-pressed with incidence odds ratio (OR) for dichotomousdata and weighted mean differences for continuous data (16)(http://www.cochrane-handbook.org/), and 95% confidenceintervals (CI) were also calculated. Heterogeneity between in-cluded studies was tested using I2 statistics, which describesthe percentage of total variation across studies that are dueto heterogeneity rather than chance (18). The interpretationof I2 depends on the magnitude and direction of effects, aswell as the strength of evidence for heterogeneity (such asp-value from the χ2 test, or a confidence interval for I2). Ifheterogeneity was present, we checked the potential reason

with reference to aspects of study design and quality, differ-ences in intervention and baseline characteristics of includedpatients, by using methods of subgroup and sensitivity anal-ysis. Statistical analysis was performed using RevMan 5.0.18,which was provided by the Cochrane Collaboration. Funnelplots were drawn to assess publication bias.

Results

Selection of included studies

Two trial assessors agreed on the selection of three CCTs.The QUORUM flow diagram illustrates the main reasons fortrial exclusions (Figure 1). The overall sample included 587patients in these three CCTs (19–21) comparing RT with RH.

Description and risk of bias of included studies

Characteristics of included studies, baseline characteristics ofpatients and potential bias in included studies are listed inTables 1–3, respectively. There were 248 patients in the RTand 339 patients in the RH groups, and the sample size ofstudies ranged from 150 to 257 women. No randomizationand concealment of allocation were used. According to ourdefinition of blinding, this was unclear in all three includedstudies. Completeness of data was good; only Marchiole et al.(21) recorded four patients lost to follow up (two in eachgroup). The description of selective reporting and other biaspossibilities was not present in detail. Apart from a significantdifference in age in the studies of Beiner et al. (20) and Diazet al. (19), baseline characteristics were similar between RTand RH groups.

622 potentially eligible abstracts from computer searches

16 trials retrieved for detail

3 prospective CCTs finally included

606 abstracts excluded Reasons: no valid survival data, reviews, meta-analysis, retrospective studies

13 trials excluded Reasons: low quality, duplicates, retrospective design

Figure 1. Quality of Reporting ofMeta-analyses (QUORUM) flow chart forstudies on radical trachelectomy and radicalhysterectomy. CCT, controlled clinical trial.

1202C© 2011 The Authors

Acta Obstetricia et Gynecologica Scandinavica C© 2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 1200–1209

L. Xu et al. Efficacy and safety of radical trachelectomy

Table 1. Characteristics of included studies on radical trachelectomy and radical hysterectomy.

Study (author, year) Country Date Design Methods Interventions

Diaz et al. (2008;19) USA November 2001 toJune 2007

CCT Single-center, parallel group,prospective study withoutrandomization

RT vs. RH

Beiner et al. (2008;20) Canada March 1994 to July2007

CCT Single-center, parallel group,prospective study withoutrandomization

RVT vs. RH

Marchiole et al. (2007;21) France December 1986 toDecember 2003

CCT Single-center, parallel group,prospective study withoutrandomization

LARVT vs. LARVH

(continue) Patients Outcomes

RT: patients with stage IB1 cervical carcinomaRH: all women with stage IB1 cervical carcinoma and who would have been eligible

for a fertility-sparing alternative

Oncological outcomes, operative results, obstetricaloutcomes, OR, blood loss, blood transfusion, lymphnode count

RVT: all patients who sought preservation of fertility with cervical cancer, tumor size≤2cm, and did not meet Society of Gynecologic Oncologists’ definition ofmicroinvasive cancer

RH: patients with stage IA and IB cervical cancer

Oncological outcomes, operative results, OR, blood loss,blood transfusion, hospital stay, time to normal urineresidual volume

LARVT: women affected by an early cervical cancer (stage I–IIA)LARVH: patients with stage I–IIA cervical cancer

Oncological outcomes, operative results, OR, bloodtransfusion, lymph node count

Abbreviations: CCT, controlled clinical trial; LARVH, laparoscopic assisted radical vaginal hysterectomy;, LARVT, laparoscopic assisted radical vaginaltrachelectomy; OR, operative time; RH radical hysterectomy; RT, radical trachelectomy; and RVT, radical vaginal trachelectomy.

Table 2. Baseline characteristics of patients in included studies.

Study (author, year)

Diaz et al. (2008; 19) Beiner et al. (2008; 20) Marchiole et al. (2007; 21)

Age (years) [mean (range)] 32(20–43) vs. 37(22–45),p<0.001

31(20–44) vs. 34(19–44),p<0.01

–

Sample size 40 vs. 110 90 vs. 90 118 vs. 139FIGO IB1 IA/IB I–IIA

Histology

SCC 20 vs. 57 NS 39 vs. 39 NS 90 vs. 102 NSAdenocarcinoma/adenosquamous 20 vs. 52 50 vs. 50 25 vs. 33Other – 1 vs. 1 3 vs. 4CLS invasion 9 vs. 30 NS 61 vs. 61 NS 43 vs. 35 NSLymph node metastasis 6 vs. 6 NS 0 vs. 0 NS 5 vs. 6 NSFollow up (months) [mean(range)] 44(3–201) 51 vs. 58 95(31–234) vs. 113(36–249)∗

Note: Results are presented as No. RT vs. No. RH.Abbreviations: CLS, capillary lymphatic space; FIGO, International Federation of Gynecology and Obstetrics; NS, not significantly different; SCC,squamous cell carcinoma.∗There were two patients lost to follow up in each group.

Effects of interventions

Two studies (437 patients) (20,21) reported postoperativerecurrence rates, and the pooled OR was 1.38 (95% CI0.58–3.28, p=0.47), suggesting that there was no significantdifference between the groups; the p-value of the heterogene-

ity test was 0.15 (Figure 2). The recurrence rate in the radicaltrachelectomy group was not higher than in the radical hys-terectomy group. For postoperative mortality, the pooled ORwas 1.14 (95% CI 0.42–3.11, p=0.80), suggesting that therewas no significant difference between the groups (p-value forheterogeneity test 0.32; Figure 2).

C© 2011 The AuthorsActa Obstetricia et Gynecologica Scandinavica C© 2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 1200–1209 1203

Efficacy and safety of radical trachelectomy L. Xu et al.

Table 3. Potential bias of included studies.

Study (author, year)

Adequatesequence

generationAllocation

concealment Blinding

Incompleteoutcome data

addressedFree of selective

reporting Free of other bias

Diaz et al. (2008; 19) No No Probably yes Yes Probably yes Probably yesBeiner et al. (2008; 20) No No Probably yes Yes Probably yes Probably yesMarchiole et al. (2007; 21) No No Probably yes Yes Probably yes Probably yes

Study or Subgroup1.1.1 Rate of recurrenceBeiner et al. (2008; 20)Marchiole et al. (2007; 21)Subtotal (95% CI)Tota l events

Test for overall effect: Z = 0.73 (P = 0.47)

1.1.3 Post-operative mortalityBeiner et al. (2008; 20)Marchiole et al. (2007; 21)Subtotal (95% CI)Tota l events

Test for overall effect: Z = 0.25 (P = 0.80)

Events

57

12

35

8

Total

90118208

90118208

Events

19

10

17

8

Total

90139229

90139229

Weight

10.8%89.2%

100.0%

13.6%86.4%

100.0%

M-H, Fixed, 95% CI

5.24 [0.60, 45.74]0.91 [0.33, 2.53]1.38 [0.58, 3.28]

3.07 [0.31, 30.08]0.83 [0.26, 2.70]1.14 [0.42, 3.11]

RT RH Odds Ratio Odds RatioM-H, Fixed, 95% CI

0.01 0.1 1 10 100Favours RT Favours RH

Figure 2. Comparisons of recurrence and postoperative mortality between radical trachelectomy (RT) and radical hysterectomy (RH).

The pooled OR for five-year recurrence-free survival in allthree studies was 1.17 (95% CI 0.54–2.53, p=0.69; p-value forheterogeneity 0.13; Figure 3). Five-year overall survival ratecould be estimated from two studies (437 patients) (20,21).The pooled OR was 0.86 (95% CI 0.30–2.43, p=0.78). Therewas no difference between RT and RH (p-value for hetero-geneity 0.53; Figure 3).

Two studies (20,21) could be used to estimate intraoper-ative complications and postoperative overall complicationswith pooled ORs of 1.66 (95% CI 0.11–25.28, p=0.72) and0.52 (95% CI 0.11–2.48, p=0.41), respectively (p-value forheterogeneity 0.008 and 0.01; Figure 4 ), suggesting no differ-ence between the groups. Postoperative infectious and non-infectious complications showed pooled ORs of 0.46 (95%CI 0.16–1.33, p=0.15) and 0.53 (95% CI 0.09–3.03, p=0.48),respectively (p-value for heterogeneity test 0.13 and 0.10).Compared with RH, RT neither increased nor decreased the

risk of intraoperative and postoperative complications (Fig-ures 4 and 5).

As we could not obtain the standard deviation (SD) of op-erative time, blood loss, lymph node count, hospital stay andtime to normal urine residual volumes from most includedtrials, pooled meta-analysis could not be done on these out-comes, but they are listed in Table 4. In the study by Diazet al. (19), the operating time of radical trachelectomy wassignificantly longer, but this was not the case in the two otherstudies. Beiner et al. found no significant difference in oper-ating time between RT and RH (20). In two studies (19,20),blood loss during the operation was compared, with a sig-nificant difference between the groups (Table 4). All threestudies could be used to estimate the need for blood transfu-sions, with apooled OR of 0.29 (95% CI 0.06–1.36, p=0.12;p-value for heterogeneity 0.02; Figure 6), suggesting no dif-ference. In two of the studies (19,21), the count of lymph

1204C© 2011 The Authors

Acta Obstetricia et Gynecologica Scandinavica C© 2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 1200–1209

L. Xu et al. Efficacy and safety of radical trachelectomy

Study or Subgroup1.2.1 Five-year recurrence-free survival rateDiaz et al. (2008; 19)Beiner et al. (2008; 20)Marchiole et al. (2007; 21)Subtotal (95% CI)Tota l events

Test for overall effect: Z = 0.40 (P = 0.69)

1.2.2 Five-year overall survival rateBeiner et al. (2008; 20)Marchiole et al. (2007; 21)Subtotal (95% CI)Tota l events

Test for overall effect: Z = 0.28 (P = 0.78)

Events

3886

113

237

89112

201

Total

4090

118248

90118208

Events

9590

132

317

90132

222

Total

11090

139339

90139229

Weight

20.9%36.8%42.3%

100.0%

19.5%80.5%

100.0%

M-H, Fixed, 95% CI

3.00 [0.65, 13.75]0.11 [0.01, 2.00]1.20 [0.37, 3.88]1.17 [0.54, 2.53]

0.33 [0.01, 8.20]0.99 [0.32, 3.03]0.86 [0.30, 2.43]

RT RH Odds Ratio Odds RatioM-H, Fixed, 95% CI

0.01 0.1 1 10 100Favours RH Favours RT

Figure 3. Comparisons of rates of five-year recurrence-free survival and five-year overall survival between radical trachelectomy (RT) and radicalhysterectomy (RH).

nodes removed was compared. Marchiole et al. (21), removedmore lymph nodes at RT than at RH (p<0.05), while in theother study there was no difference (19). Beiner et al. (22) re-ported the hospital stay and duration to normal urine residualvolumes, both of which were significantly shorter for RT. Onlyone study reported on pregnancy outcomes, i.e. a total of ninepregnancies and four deliveries, one second-trimester mis-carriage and two terminations. Because there were too fewincluded studies, we did not draw funnel plots to test forpublication bias.

Discussion

The standard surgical management for early stage cervicalcarcinoma is a radical abdominal hysterectomy and pelvicwith or without para-aortic lymph node dissection, obvi-ously eliminating the possibility of future conception. Radi-cal trachelectomy is a fertility-preserving procedure that hasrecently gained worldwide acceptance as a method of surgi-cally treating small invasive cancers of the cervix. We havecompared oncological outcomes and operative results usingthe method provided by the Cochrane Collaboration. Themeta-analysis showed that RT has similar efficacy and safetyto RH for early cervical cancer surgical treatment.

Undoubtably, the most accurate method to answer thequestions whether RT is more efficient and much safer than

RH is a randomized control trial; however, such a study is notfeasible for several reasons, as follows: (1) the issues involvedin recruiting young patients with small tumors who are anx-ious to preserve fertility is ethically difficult; (2) the samplesize required to detect a difference in the recurrence-free sur-vival rate as small as 5% is approximately 1 000; and (3) fewsurgeons are sufficiently skilled to perform this procedure.Therefore, the next best methodology to evaluate the efficacyof this procedure is a case–control design. Three CCTs withalmost 600 patients were included in our review. In the studyof John et al. (19), radical vaginal trachelectomy and radi-cal abdominal trachelectomy were done in 40 patients withstage IB1 cervical carcinoma, and included 28 radical vaginaltrachelectomies and 12 radical abdominal trachelectomies.According to our inclusion criteria, the two processes bothpreserve fertility. Recurrences, five-year recurrence-free sur-vival rates, five-year overall survival rates and postoperativemortality were similar between the two operations. Com-pared with RH, blood loss, postoperative hospital stay andtime to normal urine residual volume were significantly re-duced in RT; however, the operation time and lymph nodecount in RT were not analyzed in this review.

Heterogeneity was found between intra-/postoperativecomplications and blood transfusion in all three studies. Sev-eral factors contributed to heterogeneity. Firstly, there was a

C© 2011 The AuthorsActa Obstetricia et Gynecologica Scandinavica C© 2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 1200–1209 1205

Efficacy and safety of radical trachelectomy L. Xu et al.

Study or Subgroup1.3.1 Rate of intra-operative complicationsBeiner et al. (2008; 20)Marchiole et al. (2007; 21)Subtotal (95% CI)Total events

Test for overall effect: Z = 0.37 (P = 0.72)

1.3.3 Rate of post-operative overall complicationsBeiner et al. (2008; 20)Marchiole et al. (2007; 21)Subtotal (95% CI)Total events

Test for overall effect: Z = 0.83 (P = 0.41)

1.3.4 Rate of post-operative noninfectious complicationsBeiner et al. (2008; 20)Marchiole et al. (2007; 21)Subtotal (95% CI)Total events

Test for overall effect: Z = 0.71 (P = 0.48)

Events

123

15

425

29

123

24

Total

90118208

90118208

90118208

Events

28

10

1628

44

627

33

Total

90139229

90139229

90139229

Weight

49.1%50.9%

100.0%

45.3%54.7%

100.0%

34.5%65.5%

100.0%

M-H, Random, 95% CI

6.77 [1.47, 31.19]0.43 [0.11, 1.65]

1.66 [0.11, 25.28]

0.22 [0.07, 0.67]1.07 [0.58, 1.95]0.52 [0.11, 2.48]

0.16 [0.02, 1.33]1.00 [0.54, 1.87]0.53 [0.09, 3.03]

RT RH Odds Ratio Odds RatioM-H, Random, 95% CI

0.01 0.1 1 10 100Favours RT Favours RH

Figure 4. Comparisons of rates of intraoperative and postoperative complications between radical trachelectomy (RT) and radical hysterectomy (RH).

Study or Subgroup1.4.1 Rate of post-operative infectious complicationsBeiner et al. (2008; 20)Marchiole et al. (2007; 21)Subtotal (95% CI)Tota l events

Test for overall effect: Z = 1.44 (P = 0.15)

Events

32

5

Total

90118208

Events

101

11

Total

90139229

Weight

91.5%8.5%

100.0%

M-H, Fixed, 95% CI

0.28 [0.07, 1.04]2.38 [0.21, 26.57]0.46 [0.16, 1.33]

RT RH Odds Ratio Odds RatioM-H, Fixed, 95% CI

0.01 0.1 1 10 100Favours RT Favours RH

Figure 5. Comparison of rate of postoperative infectious complications between radical trachelectomy (RT) and radical hysterectomy (RH).

significant difference in patients’ age (19,20), though otherbaseline characteristics were similar, which suggested a poten-tial selection risk. Secondly, surgical outcomes were relatedto the experience of the surgeon (23). Considering the learn-

ing curves of surgeons, performance bias might have beenbeen introduced. Thirdly, as seen in Table 3, blinding wasunclear in all included studies, so measurement bias mightexist. Although common in surgical clinical trials, this may

1206C© 2011 The Authors

Acta Obstetricia et Gynecologica Scandinavica C© 2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 1200–1209

L. Xu et al. Efficacy and safety of radical trachelectomy

Table 4. Comparisons of operative time, blood loss, lymph node count, postoperative hospital stay and duration to normal urine residual volume

between radical trachelectomy and radical hysterectomy.

Study (author, year)

Diaz et al. (2008; 19) Beiner et al. (2008; 20) Marchiole et al. (2007; 21)

Operative time[mean(range)/mean±SD]

300(100–630) vs.240(134–450)min, p=0.023

2.9(1.5–5) vs. 2.8(1.6–5)h,p=0.75

179±54 vs. 187±53min, NS

Blood loss (ml) [mean(range)] 150(50–600) vs. 425(75–2700),p<0.001

300(100–1100) vs.600(100–2300), p<0.001

–

Lymph node count[mean(range)/mean±SD]

26(12–64) vs. 28(1–64), NS – 19±8(5–42) vs. 17±7(5–44),p<0.05

Postoperative hospital stay(days) [mean(range)]

– 1(0–7) vs. 6(1–20), p<0.001 –

Time to normal urine residualvolume (days [mean(range)]

– 1(1–49) vs. 6(2–63), p<0.001 –

Abbreviation: NS, not significantly different.

Study or Subgroup1.5.1 Rate of blood transfusionDiaz et al. (2008; 19)Beiner et al. (2008; 20)Marchiole et al. (2007; 21)Subtotal (95% CI)Total events

Test for overall effect: Z = 1.57 (P = 0.12)

Events

12

15

18

Total

4090

118248

Events

72124

52

Total

11090

139339

Weight

24.8%32.6%42.6%

100.0%

M-H, Random, 95% CI

0.38 [0.04, 3.17]0.07 [0.02, 0.33]0.70 [0.35, 1.40]0.29 [0.06, 1.36]

RT RH Odds Ratio Odds RatioM-H, Random, 95% CI

0.01 0.1 1 10 100Favours RT Favours RH

Figure 6. Comparison of rate of blood transfusion between radical trachelectomy (RT) and radical hysterectomy (RH).

weaken the reliability of our results to some extent. Fourthly,lost data could affect results, and there were two patients lostto follow up in each group in one of the studies (21). Finally,publication bias might exist with the small number of in-cluded studies.

There were some significant differences between the RTand RH groups. Blood loss, time to normal urination andpostoperative hospital stay were significantly shorter in theRT group, similar to reports by other centers performingRT (24), while there was no difference as regards intra-/postoperative complications, with the pooled ORs being 2.77(95% CI 0.35–22.19) and 0.55 (95% CI 0.18–1.69, p=0.30),respectively. This suggests that RT is superior to RH to someextent. Several studies (25–28) reported intraoperative com-plication rates in RT from 0 to 25%. In this review, there were23 patients (9%) who suffered such complications, and 16(6%) had accidental cystotomy in the RT group. Monitoringcomplication rates continuously is important for any newsurgical procedure.

The development of fertility-sparing surgery has providedpatients with early stage cervical carcinoma with the opportu-nity to pursue conception after the treatment of their disease.Beiner et al. (22) indicated that pregnancy rates for thosewomen who attempted conception were 41–79%. The rate offirst-trimester miscarriage was 18% for the pooled data, simi-lar to that of the general population. Plante et al. (24) showedthat 31 of the women became pregnant (43%) and had atotal of 50 pregnancies, including 10 miscarriages (eight firsttrimester and two second trimester), while Shepherd et al.(26) reported that the five-year accumulated conception ratewas 52.8%. Diaz et al. (19) reported nine pregnancies, includ-ing three term cesarean sections, one late preterm delivery at35weeks, one second-trimester spontaneous abortion andtwo second-trimester terminations. However, Covens et al.(11) noted five pregnancies among women, including threeterm cesarean sections and two spontaneous first-trimesterabortions, yielding an actuarial conception rate of 37% atone year, a lower rate of conception.

C© 2011 The AuthorsActa Obstetricia et Gynecologica Scandinavica C© 2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 1200–1209 1207

Efficacy and safety of radical trachelectomy L. Xu et al.

It is important to evaluate emotional and sexual aspects aswell as quality-of-life issues after surgery for malignant cer-vical tumors. Only one study has hitherto reported that theseaspects did not differ significantly by type of surgery (29). Weincluded prospective controlled clinical trials and retainedonly the most comprehensive studies, which is different froma previous systematic review (30). Radical trachelectomy hassimilar efficacy and safety to RH for early cervical cancer, butis associated with reduced blood loss, shorter time to resump-tion of urinary function and shorter postoperative hospitalstay. An important subset of young women with early stagecervical cancer can benefit from a fertility-sparing surgicalapproach by radical trachelectomy.

Funding

The research was supported by the National Natural ScienceFoundation of China (grant no. 30740016).

References

1. Wertheim E. The extended abdominal operations for

carcinoma uteri (based on 5000 operative cases). Am J Obstet

Gynecol. 1921;66:169.

2. Piver MS, Rutledge F, Smith JP. Five classes of extended

hysterectomy for women with cervical cancer. Obstet

Gynecol. 1974;44:265–72.

3. Kinney WK, Hodge DO, Egorshin EV, Ballard DJ, Podratz

KC. Identification of a low-risk subset of patients with

stage IB invasive squamous cancer of the cervix possibly

suited to less radical surgical treatment. Gynecol Oncol.

1995;57:3–6.

4. Morgan PR, Anderson MC, Buckley CH, Murdoch JB, Lopes

A, Duncan ID, et al. The Royal College of Obstetricians and

Gynaecologists micro-invasive carcinoma of the cervix study:

preliminary results. Br J Obstet Gynecol. 1993;100:664–8.

5. Jones WB, Mercer GO, Lewis JL Jr, Rubin SC, Hoskins WJ.

Early invasive carcinoma of the cervix. Gynecol Oncol.

1993;51:26–32.

6. Dursun P, Ayhan A, Kuscu E. New surgical approaches for

the management of cervical carcinoma. Eur J Surg Oncol.

2008;34:487–96.

7. Covens A, Rosen B, Murphy J, Laframboise S, DePetrillo AD,

Lickrish G, et al. Changes in the demographics and

perioperative care of stage IA2/IB1 cervical cancer over the

past 16 years. Gynecol Oncol. 2001;81:133–7.

8. Dargent D, Martin X, Sacchetoni A, Mathevet P.

Laparoscopic vaginal radical trachelectomy: a treatment to

preserve the fertility of cervical carcinoma patients. Cancer.

2000;88:1877–82.

9. Shepherd JH, Crawford RA, Oram DH. Radical

trachelectomy: a way to preserve fertility in the treatment of

early cervical cancer. Br J Obstet Gynecol. 1998;105:

912–16.

10. Roy M, Plante M. Pregnancies after radical vaginal

trachelectomy for early-stage cervical cancer. Am J Obstet

Gynecol. 1998;179:1491–6.

11. Covens A, Shaw P, Murphy J, DePetrillo D, Lickrish G,

Laframboise S, et al. Is radical trachelectomy a safe

alternative to radical hysterectomy for patients with

stage IA-B carcinoma of the cervix? Cancer.1999;86:

2273–9.

12. Gien LT, Covens A. Fertility-sparing options for early stage

cervical cancer. Gynecol Oncol. 2010;117:350–7.

13. Plante M. Vaginal radical trachelectomy: an update. Gynecol

Oncol. 2008;111:S105–10.

14. Einstein MH, Park KJ, Sonoda Y, Carter J, Chi DS, Barakat

RR, et al. Radical vaginal versus abdominal trachelectomy for

stage IB1 cervical cancer: a comparison of surgical and

pathologic outcomes. Gynecol Oncol. 2009;112:73–7.

15. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup

DF. Improving the quality of reports of meta-analyses of

randomised controlled trials: the QUOROM statement.

Quality of Reporting of Meta-analyses. Lancet.

1999;354:1896–900.

16. Julian PH, Sally G. Cochrane Handbook for Systematic

Reviews of Interventions Version 5.0.2 [updated September

2009] The Cochrane Collaborations. 2009. Available online

at: http://www.cochrane-handbook.org/ (accessed July 17,

2011)

17. Mantel N, Haenszel W. Statistical aspects of the analysis of

data from retrospective studies of disease. J Nat Cancer Inst.

1959;22:719–48.

18. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring

inconsistency in meta-analyses. BMJ. 2003;327:557–60.

19. Diaz JP, Sonoda Y, Leitao MM, Zivanovic O, Brown CL, Chi

DS, et al. Oncologic outcome of fertility-sparing radical

trachelectomy versus radical hysterectomy for stage IB1

cervical carcinoma. Gynecol Oncol. 2008;111:255–60.

20. Beiner ME, Hauspy J, Rosen B, Murphy J, Laframboise S,

Nofech-Mozes S, et al. Radical vaginal trachelectomy vs.

radical hysterectomy for small early stage cervical cancer: a

matched case–control study. Gynecol Oncol.

2008;110:168–71.

21. Marchiole P, Benchaib M, Buenerd A, Lazlo E, Dargent D,

Mathevet P. Oncological safety of laparoscopic-assisted

vaginal radical trachelectomy (LARVT or Dargent’s

operation): a comparative study with laparoscopic-assisted

vaginal radical hysterectomy (LARVH). Gynecol Oncol.

2007;106:132–41.

22. Beiner ME, Covens A. Surgery insight: radical vaginal

trachelectomy as a method of fertility preservation for

cervical cancer. Nat Clin Pract Oncol. 2007;4:353–61.

23. Lee JH, Ryu KW, Lee JH, Park SR, Kim CG, Kook MC, et al.

Learning curve for total gastrectomy with D2 lymph node

dissection: cumulative sum analysis for qualified surgery.

Ann Surg Oncol. 2006;13:1175–81.

24. Plante M, Renaud MC, Hoskins IA, Roy M. Vaginal radical

trachelectomy: a valuable fertility-preserving option in the

1208C© 2011 The Authors

Acta Obstetricia et Gynecologica Scandinavica C© 2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 1200–1209

L. Xu et al. Efficacy and safety of radical trachelectomy

management of early-stage cervical cancer. A series of 50

pregnancies and review of the literature. Gynecol Oncol.

2005;98:3–10.

25. Shepherd JH, Milliken DA. Conservative surgery for

carcinoma of the cervix. Clin Oncol (R Col Radiol).

2008;20:395–400.

26. Shepherd JH, Spencer C, Herod J, Ind TE. Radical vaginal

trachelectomy as a fertility-sparing procedure in women with

early-stage cervical cancer cumulative pregnancy rate in a

series of 123 women. BJOG. 2006;113:719–24.

27. Carter J, Sonoda Y, Abu-Rustum NR. Reproductive concerns

of women treated with radical trachelectomy for cervical

cancer. Gynecol Oncol. 2007;105:13–16.

28. Dursun P, LeBlanc E, Nogueira MC. Radical vaginal

trachelectomy (Dargent’s operation): a critical review of the

literature. Eur J Surg Oncol. 2007;33:933–41.

29. Carter J, Sonoda Y, Baser RE, Raviv L, Chi DS, Barakat RR,

et al. A 2-year prospective study assessing the emotional,

sexual, and quality of life concerns of women undergoing

radical trachelectomy versus radical hysterectomy for

treatment of early-stage cervical cancer. Gynecol Oncol.

2010;119:358–65.

30. Han L, Yang XY, Zheng A, Wang L, Chu YX, Wang Y.

Systematic comparison of radical vaginal trachelectomy and

radical hysterectomy in the treatment of early-stage cervical

cancer. Int J Gynecol Obstet. 2011;112:149–53.

C© 2011 The AuthorsActa Obstetricia et Gynecologica Scandinavica C© 2011 Nordic Federation of Societies of Obstetrics and Gynecology 90 (2011) 1200–1209 1209