Abdominal radical trachelectomy in West London
Dr S Saso, Clinical Research Fellow, Department of Surgery and Cancer, Institute of Reproductive & Developmental Biology, Imperial College London, Hammersmith Hospital Campus, Du Cane Road, London W12 0NN, UK. Email firstname.lastname@example.org
Please cite this paper as: Saso S, Ghaem-Maghami S, Chatterjee J, Naji O, Farthing A, Mason P, McIndoe A, Hird V, Ungar L, Del Priore G, Smith J. Abdominal radical trachelectomy in West London. BJOG 2012;119:187–193.
Objective Traditionally, the surgical management of invasive cervical carcinoma that has progressed beyond microinvasion has been a radical abdominal hysterectomy. However, this results in the loss of fertility, with significant consequences for the young patient. This report describes abdominal radical trachelectomy (ART) as a potential replacement for radical hysterectomy in patients with stage IA2–IIA cervical cancer who desire a fertility-sparing procedure without decreasing the curative rates.
Design Observational, retrospective study.
Setting Teaching hospital and regional cancer centre in London, UK.
Population Patients undergoing ART.
Methods Patients presenting during the period 2000–2009 with cervical cancer stage IA2–IIA were offered a trachelectomy, if they expressed a desire to preserve fertility. The type of trachelectomy (vaginal/abdominal) was chosen based on patient anatomy and neoplastic and magnetic resonance imaging characteristics. Each patient was counselled as to the experimental nature of the procedure.
Main outcome measures Survival, recurrence and fertility issues among ART patients.
Results A total of 30 patients underwent ART (open and laparoscopic) between 2001 and 2009. Three patients presented with a recurrence, two of which have died (median follow-up: 24 months). Only three patients required further surgical re-intervention because of operative complications. Ten patients attempted to conceive, resulting in three conceptions (30%) and two live children.
Conclusions Abdominal radical trachelectomy provides a feasible, cost-effective and safe treatment option for young women who have been diagnosed with early-stage cervical cancer and wish to preserve their fertility.
Traditionally, the surgical management of invasive cervical carcinoma that has progressed beyond microinvasion has been a radical abdominal hysterectomy and pelvic lymphadenectomy. This treatment method has shown excellent results in terms of survival, but results in the loss of fertility, which can cause added stress and reduced quality of life to a young female patient.
In current gynaecological oncology practice, fertility preservation has become a significant and meaningful issue when deciding on how to treat stage IA2–IB cervical cancer.1 This can be explained by the rising prevalence of cervical cancer in young women and the decision to delay childbearing made by women in the developed world.2,3 A fertility-preserving operation, known as ‘radical trachelectomy’, was coined in 1994 in a seminal paper by the late Daniel Dargent, who described a new technique suitable for exophytic tumours of stages IA–IIA.4 This procedure preserves the uterine body while removing the cervix, parametrium and upper one-third of the vagina.5,6 Two types of radical trachelectomy are now recognised: vaginal (VRT) and abdominal (ART).
The benefits of VRT in treating early-stage cervical cancer and successful fertility outcomes thereafter have been detailed previously.7,8 However, certain problems have arisen. Recurrence rates for example range from 12.5 to 29% for tumours that are >2 cm in diameter.5,9
The extent of the ART procedure is identical to that of a traditional type-III Wertheim hysterectomy. The similarity in the operative techniques between the two operations have allowed ART to become an acceptable alternative. This report proposes ART to be a potential solution in overcoming some of the problems presented by VRT.
Approval to perform ART was granted by the Institute Review Board of the Chelsea and Westminster Hospital Ethics Committee in 1997, which is the where J.R. Smith first performed an ART.10 Since the publication of that initial study, the selection criteria employed in order to choose suitable patients has remained unaltered. In our cancer centre, all patients presenting from 2000 to 2009 with cervical cancer at stages IA2–IIA who expressed a desire to preserve fertility were given the option of undergoing an ART. Each patient was counselled as to the novel nature of the procedure and appropriate consent was obtained. The surgical technique has been described previously.11,12
All patients were discharged from hospital once they demonstrated an adequate level of comfort and intake of food and fluid, only complained of minimal abdominal pain and had voided a sufficient volume of urine. They were also instructed, prior to discharge, to seek medical help if a significant volume of blood was passed per vaginam, if lower abdominal pain became intolerable or if problems with micturition became apparent.
Routine follow-up was organised in the gynaecology oncology clinic every 3 months in the first year, every 4 months in the second year, every 6 months in the next 3 years and annually thereafter. Each appointment consisted of a focused history and examination, followed by a ‘cervical’ smear and pelvic ultrasonography if deemed necessary. It is also worth mentioning that our patients had direct access to our Cancer Nurse Specialists if they required specific support. With regards to attempts at pregnancy, our recommendations were for the patients to use contraception during the first year following ART to allow the tissue to heal.
The project proposal was reviewed and approved as a clinical audit project by the Quality and Safety Co-ordinator for the Women’s and Children’s Clinical Programme Group at Queen Charlotte’s Hospital. Case notes were retrieved and studied and the following data were extracted: mean age, mean length of operating time, pre- and post-ART tumour histology (tumour type, stage and grade), size and margin status of cervical tumour, number and disease status of lymph nodes removed, estimated blood loss and type of cervical suture used to anastamose the uterine stump with the upper vagina. We also recorded postoperative complications, follow-up time and outcome, recurrence of disease and fertility outcomes.
Prior to surgery, all patients had preoperative magnetic resonance imaging (MRI) of the pelvis to assess tumour location, tumour size and involvement of the parametria and upper endocervix. Computerised tomography (CT) of the chest, abdomen and pelvis was performed to investigate possible distant spread. Examination under anaesthetic and cystoscopy with or without sigmoidoscopy was performed in order to complete the clinical staging process. Pathologic parameters were gathered from the pre- and postoperative reports. At our institution, all cervical specimens were evaluated by specialist gynaecologic pathologists. Case notes were reviewed by two of the investigators to confirm the accuracy of the data collected.
The patient characteristics and surgical outcomes are listed in Table 1. Following in-depth consultation between the patient and members of the gynaecology oncology team, thirty-one women chose to undergo an elective fertility-sparing radical abdominal trachelectomy. This operation was selected over a radical hysterectomy primarily in order to protect the patient’s fertility, while at the same time managing the chief diagnosis of cervical carcinoma.
Table 1. Pre-, intra- and postoperative characteristics
|Age||32.5 years (mean)|
|Follow-up||24 months (median)|
| ||Number of patients (%)|
|Stage IA2|| 2 (6.7)|
|Stage IB1|| 25 (83)|
|Stage IB2|| 2 (6.7)|
|Stage IIA|| 1 (3.3)|
|Squamous|| 15 (48)|
|Adenosquamous|| 4 (13)|
|Adenocarcinoma|| 10 (32)|
|Glassy cell carcinoma|| 1 (3.2)|
|Intra-operative and postoperative characteristics|
|Mean operative time (minutes)||170 (110–300)|
|Median number of lymph nodes retrieved|| 24 (7–52)|
| ||Number of patients (%)|
|ART – open approach|| 27 (90)|
|ART – laparoscopic approach|| 3 (10)|
|Postoperative antibiotics|| 8 (27)|
|Estimated blood loss (ml)||813 (50–4500)|
|Transfusion|| 6 (20)|
|Lymphatic invasion|| 3 (10)|
|Postoperative complications|| 14 (47)|
|Re-operation|| 3 (10)|
|Recurrence|| 3 (10)|
|Survival rate|| 28 (93)|
|Pregnancy rate|| 3 (10)|
At the time of diagnosis the mean age of the 31 patients was 32.5 years (range 23–41 years). The stages of cervical tumour that patients presented with were: stage IA2 (n = 2; 6.7%); stage IB1 (n = 25; 83.3%); stage IB2 (n = 2; 6.7%); and stage IIA (n = 1; 3.3%). Five tumours (16.7%) were >2 cm in diameter. Notably, patients with non-squamous tumours were advised of the risk of ovarian and distant metastases.
The ART was performed simultaneously with bilateral pelvic lymphadenectomy. The mean length of operating time was 2 hours 50 minutes (ranging from 1 hour 50 minutes to 5 hours). The tumour was excised completely in 27 patients, with the median number of lymph nodes removed from the pelvis equalling 24 (range 7–52). A total of 27 patients (90%) had no lymph node involvement. In the three patients (10%) who did demonstrate such spread, the lymph nodes involved were left internal iliac (1/23), right and left internal and external iliac (3/22), and right obturator (2/28). Lymph nodes removed were from the following groups: external, internal and common iliac, sacral and obturator. Microscopic evaluation demonstrated the resection margins of more than 3 mm to be tumour-free in all cases, with endo- and ectocervical and circumferential margins commented on most frequently. None of the tumours displayed perineural spread. During surgery, 11 patients (36.7%) had a frozen section sample sent for histology. All revealed tumour-free resection margins.
Magnetic resonance imaging revealed the largest tumour to be 5 cm × 3.7 cm × 6 cm, with a volume of 48 cc (stage IIA) and in contact with the posterior wall of the bladder. Following examination under anaesthesia (EUA) and imaging, as well as a team discussion, it was felt that ART could safely remove the tumour. The patient was also keen to preserve her fertility. She did not suffer from a recurrence during follow-up and is still alive.
Finally, with regards to the cervical suture, insertion depended on ease of application during surgery and surgeon’s choice. A total of 24 patients (80%) had a cervical cerclage to aid future pregnancies.
No intraoperative complications occurred in this case series, and 53.3% of patients reported no immediate, early or late postoperative complications. Three patients required major surgical re-intervention. One patient was found to have a haematocolpos 10 months post-ART after she presented with a history of amenorrhoea during the same 10-month period. Another patient who originally underwent a laparoscopic ART discharged significant volumes of lymphatic fluid vaginally following the operation, and on day 9 (postoperatively), presented with omental prolapse between sutures, which was repaired vaginally. A year later following a normal menstruating pattern and no symptoms during sexual intercourse, she presented with lower abdominal pain. Investigations revealed that the uterus had avulsed from the top of the vagina. A primary repair was performed and the patient has made a full recovery. A third patient underwent a laparoscopy, followed by a low transverse laparotomy, ovarian cystectomy, insertion of cervical suture and adhesiolysis where possible to treat a combination of abdominal pain and right ovarian cysts.
Recurrence and mortality
The 28 patients (93.3%) who underwent ART are alive at the time of writing this article, with a median follow up of 24 months (range 7–113 months). Two patients have died following a recurrence of the disease. One patient (30 years of age) presented 10 months after her ART for stage-IB1 cervical adenocarcinoma (diagnosed in June 2006) with non-specific respiratory symptoms, breathlessness and intermittent pleuritic chest pain, as well as several lumps situated in the anterior abdominal wall. Imaging revealed a highly suspicious right lung and abdominal wall lesions with paratracheal lymphadenopathy. CT-guided biopsies demonstrated neoplastic lesions of identical histology as the original cervical cancer, mucinous adenocarcinoma, and therefore confirmed a metastatic picture. The patient underwent surgical removal of the abdominal masses and several cycles of chemoradiotherapy, but died in 2008, 22 months after her initial diagnosis.
The second patient also underwent ART for stage-IB1 cervical adenocarcinoma. A follow-up MRI 3 months post-ART revealed extensive disease in the anterior portion of the uterus, invading towards the bladder and posteriorly towards the rectum. She was treated with radical chemoradiotherapy and 2 months post-treatment, a new MRI scan showed residual disease still present and new lung nodules, thus confirming further metastatic disease. As her condition worsened, the patient underwent a defunctioning colostomy formation following admission for large bowel obstruction. She died 11 months after diagnosis.
One other patient has presented with disease recurrence (November 2010). She was diagnosed with a grade-2 squamous cervical carcinoma (stage IB) in November 2007 and underwent ART with bilateral pelvic lymphadenectomy (24 lymph nodes, all clear). Resection margins were tumour free. Her postoperative recovery period had been uneventful until the time of recurrence. It consisted of a pelvic mass, extending to pelvic side walls and pressing against rectum and bladder, with right ovarian involvement. She underwent chemoradiotherapy with no subsequent evidence of persistent disease.
Prior to ART, four patients had a history of pregnancy, with only one patient out of the four reporting the birth of a healthy baby. The other three patients had suffered one or more miscarriages. Ten patients reported a desire to conceive during the follow-up period. Fertility treatment or oocyte cryopreservation was not sought by any of the patients prior to ART. The mean age of this particular group was 30 years (range 26–36 years), with a median follow up of 53 months (range 11–113 months). The pregnancy rate was 30%, with two of the patients delivering successfully.
Two patients conceived spontaneously 5 and 5 years after ART, respectively, and both experienced an uneventful pregnancy. Healthy term female infants were delivered at 37 weeks of gestation via a caesarean section. Noting the sole reliance on ovarian vessels for uterine perfusion during pregnancy, we were reassured by the weight of the ‘term’ babies, which were within normal range. The third patient conceived 6 years post-ART via in vitro fertilisation (IVF), but had to undergo a termination of pregnancy at 17 weeks of gestation. This was a consequence of spontaneous rupture of membranes and subsequent fulminating chorioamnioitis. Notably all three patients had a cervical cerclage suture inserted at the time of ART, with the suture removed in the chorioamnioitis case.
Carcinoma of the cervix is the second most common cancer in women in developing countries, and among all gynaecological malignancies remains the leading cause of death worldwide. In developed countries, however, screening of cervical cancer has dramatically reduced the incidence of this disease, as well as the mortality caused by it.13 Despite this, around 45% of surgically treated stage-IB cancers occur in women under the age of 40 years, with the age of nulliparous women increasing in developed countries.13,14
Cervical cancer survival has improved in almost all categories, and more patients of reproductive age survive their cancers. Hence, a greater number of patients will request the possible preservation of fertility in the future. Fertility preservation in cervical cancer patients has become a meaningful quality-of-life objective, both for the patients and the gynaecologic oncology surgeons treating them, especially as the mean age of childbearing is slowly increasing in the developed world.15
Although worldwide most cervical cancer patients are treated with hysterectomy, fertility-preserving methods are becoming more accepted and available.15 This important treatment alteration followed the realisation that the uterine body can be preserved without an apparent compromise of therapeutic efficacy.15 The current accepted fertility-preserving procedure in Europe is VRT with laparoscopic pelvic lymphadenectomy, and with over 600 cases reported in the literature to date, provides the greatest experience of a method of fertility-sparing surgery in early cervical cancer. ART is becoming a favoured fertility-sparing procedure in the UK and USA, with extremely promising and improved oncological results (1.9% recurrence rate in tumours <2 cm in diameter following ART), but with worse fertility outcomes in comparison with VRT (15.5 versus 30% of all pregnant women from the international series) for tumours <2 cm in diameter.13 The two most recent patient series on ART reported promising results. A retrospective study in Japan reported ART outcomes for 61 women were followed-up for a median of 27 months. Six recurrences (9.8%) were diagnosed, with none occurring in patients with a tumour diameter of <2 cm, except in one case with adenocarcinoma. The cumulative pregnancy rate among the women who attempted to conceive was 13.8% (4/29).16
A retrospective review of a prospectively maintained database of 64 ART patients revealed no recurrences at a median follow-up of 22.8 months. The pregnancy rate was 20%.17
In our opinion, and when looking at the literature, ART seems to be a reasonable option for selected patients whose tumours are <4 cm in diameter. We are aware that the fertility rates currently favour VRT, and this issue requires further investigation. Despite this, we believe that in the future gynaecological oncologists should consider both ART and VRT with every patient diagnosed with early-stage cervical cancer (tumour size <2 cm in diameter) that wishes to preserve their fertility. The choice should depend on the eligibility criteria (Table 2), treatment centre, availability of trained personnel and patient choice. In particular the patient should be made aware of recurrence, mortality and fertility outcomes of ART and VRT, as well as the experience of surgeons at that institution in both procedures. The techniques for an open ART and routine radical hysterectomy are extremely similar. Therefore, in theory, a learning curve should not exist as more trainee surgeons should potentially be able to master the operative techniques of open ART relatively easily because radical hysterectomy is readily taught and reproducible in any gynaecologic oncology unit. In contrast, VRT may be a more difficult technique to learn because it is limited to a few centres with small patient numbers (i.e. for VRT), and where specialised operative techniques are available. We know that there is a high rate of operation complications when surgeons are on the learning curve of VRT. These include cystotomy, enterotomy and pelvic haematoma.18,19 The overall incidence of intra- and postoperative complications is likely to be lower with ART than with VRT when performed in average-sized gynaecologic oncology centres. Our cohort of patients supports this theory with no intra-operative injury demonstrated. Only three patients required further surgical re-intervention because of complications related to the operation: surgical treatment of haematocolpos; laparoscopic repair of an omental prolapse; and low transverse laparotomy, ovarian cystectomy and adhesiolysis. With regards to urinary symptoms, one patient only experienced a single episode of urinary retention in the first month postoperatively.
Table 2. Eligibility of patient for ART: what are the criteria?
|Cervical carcinoma: FIGO stage IA1 with lymph-vascular space involvement (LVSI) or stage IA2/IB1|
|Histological subtypes: squamous, adenosquamous, adenocarcinoma|
|No evidence (clinical or radiological) to suspect extra-pelvic spread or lymph node metastases|
|Lesion diameter >2 cm|
|Desire to preserve fertility|
|No diagnosis or indication to suspect infertility|
|Favourable age (relating to fertility outcomes)|
Abdominal radical trachelectomy (ART) is a safer option with regards to bulky, exophytic or >2-cm diameter tumours, as the extent and location of tumours this size and shape requires an increased thoroughness of parametrial, sacrouterine, vesico-cervical and pelvic lymphatic tissue resection, which is not possible with VRT, where the approach limits the parametrial resection to tissue in the medial half of the broad ligament. With regards to the two deaths from our cohort, one of the patients was advised that ART would not give her the greatest chance of cure because of her bulky tumour mass (>2 cm). However, she still opted for ART for fertility reasons (the patient was only 29 years of age at the time of diagnosis), which unfortunately did not lead to her recovery.
We also note other less common but still important cases in which ART should be the primary treatment option: paediatric cervical cancer patients; diagnosis of cancer in the cervical stump after subtotal hysterectomy; and cervical cancer patients with distorted vaginal anatomy.20 However, as mentioned before, VRT has certain advantages over ART, mainly a shorter hospital stay and postoperative recovery, and possibly less bladder and bowel dysfunction.
Only ten of our cohort attempted to conceive, in part because of our earlier recommendation that each patient waits for 2 years following the procedure while undergoing close surveillance. This has now changed to 1 year. Our pregnancy rate was 30%, with a delivery rate of 20% (term pregnancies). This is comparable with the rates reported in the literature: worldwide, 194 patients have undergone ART, 15% became pregnant and 10% delivered.13
We have previously reported the birth of two healthy term infants and the loss of three fetuses at 7, 8 and 15 weeks of gestation of five pregnant patients who underwent ART (Hungarian patient cohort).21 The patients were followed-up for 40 months with no record of relapse. However, on closer analysis of the ‘lost’ fetuses, we note that two of these three were in fact the first two ARTs performed. We believe this to be of significance because since then, the surgical technique has constantly improved as both knowledge and experience of the procedure have advanced.
We acknowledge, however, that our conclusions have their limitations, mainly as a result of the size of our data set and duration of follow-up. Our data set includes only 30 patients, and the median follow-up length of time is only 24 months. Our aim is not to influence other gynaecologic oncology surgeons, but to describe our thoughts of how ART may be used in the future. At our unit we offer both ART and VRT, with the final decision resting with the principal surgeon.
The fact that the final decision rests with the principal surgeon was perfectly highlighted when you apply the ART eligibility criteria (Table 2) to our data. Our patient cohort ‘conflicted’ with the criteria on two accounts: size and stage. Imaging revealed only five patients with tumour size >2 cm in diameter, and three patients were diagnosed with a stage-IB2/IIA tumour. With respect to the latter three patients, all are still alive at the time of writing. Our conclusions below are partly based on these findings.
Finally, it is worth asking the question of whether ART is an adequate fertility-preserving procedure when only a third of the cohort attempted to become pregnant post-ART and the mortality rate was 6.7%. Scrutiny of the results refutes this proposition. First, one of the two patients that died was advised against ART. Second, all women in our cohort underwent ART as opposed to total abdominal hysterectomy in order to preserve their fertility for the purpose of future pregnancy. A third of the cohort has tried to get pregnant post-ART, but this is only to date (median follow-up: 2 years). From the remaining 20 patients who have not attempted pregnancy, some did not have a partner/were not married at the time of the operation, whereas others felt that they wanted to wait longer prior to starting a family. Hence, the low number of women attempting to conceive to date should not be confused with the fact that all of our ART patients were hoping to fall pregnant eventually.
The ART patient series described in this report represents the largest data set from a single institution in the UK. It is also the first such series published in the UK. We believe that ART provides a feasible, cost-effective and most importantly safe treatment option for women who have been diagnosed with early-stage cervical cancer and who wish to preserve their fertility. Careful selection of patients for trachelectomy is of paramount importance. Both ART and VRT should be considered prior to deciding which operation is more suitable for treating tumours <2 cm in diameter. For tumours >2 cm in diameter, ART seems to be the most appropriate choice. Decision making should involve a fully informed and consented patient in all cases. Discussion in a gynaecologic oncology multidisciplinary team meeting, pre-operative counselling and postoperative support is vital.
Fertility outcomes and the role of adjuvant therapy in node-positive post-ART patients deserve further investigation. We encourage other gynaecological oncology centres to focus on these two issues prior to reporting their experiences in future series of patients undergoing ART. We also believe that ART should qualify as a ‘required skill’, and should therefore feature on the gynaecologic oncology surgical curriculum.
Finally, we believe that the results of this study suggest the need for a treatment comparison of ART versus VRT for patients that are diagnosed with early-stage cervical cancer who wish to preserve their fertility. This should be a multicentre, prospective randomised trial of: ART versus VRT versus cone biopsy and lymphadenectomy for lesion size 0–2 cm, correlating recurrence rates and fertility outcomes, as primary end points.22 Secondary aims should try to define the incidence of specific postoperative complications within the three trial groups, arrive at a more suitable method of preoperative patient selection for a specific procedure and investigate how to apply adjuvant therapy in node-positive patients. A closer look at surgical techniques employed in the formation of a neocervix, application of a permanent cervical cerclage (and timing: during ART or pregnancy) and process of re-anastomosis are also worthy of study.22
Globally, gynaecologic oncology teams should urge each other to ensure that existing or planned trials include factors in their demographic data enabling the future collection of relevant data to test the above hypotheses further. In other words, perform individual patient and survival data meta-analyses.
Disclosure of interests
JRS, GDP and LU are the co-inventors of the ART technique.
Contribution to authorship
SS and SG-M were responsible for the design, drafting and revision for important intellectual content of the original manuscript. SS, JC and ON were responsible for data collection. SG-M, AF, PM, AM and VH are all consultant gynaecological oncologists, and have performed ART on the patients described in this article. In addition, together with LU and GDP, they were responsible for providing important intellectual content throughout the production, drafting and final approval of the version of the article to be published. JRS is the guarantor for this paper and accepts full responsibility for the work and/or the conduct of the study. His involvement was critical to every phase of this work and he controlled the decision to publish.
Details of ethics approval
Approval to perform ART was granted by the Institute Review Board of the Chelsea and Westminster Hospital Ethics Committee in 1997, which is where ART was first performed. Approval to collect data regarding ART patients was granted by the Quality and Safety Co-ordinator for the Women’s and Children’s Clinical Programme Group at Queen Charlotte’s Hospital.