Selection of cases and safety

The criteria for carrying out RVT have been almost unchanged for more than 10 years ( Table 1 ). Cervical cancer spreads predominantly in a lateral direction to parametrium, cardinal ligaments and upper vagina, but small stage IB1 carcinoma infrequently extends to the corpus uteri. No randomized–controlled trials comparing RVT and radical hysterectomy in the treatment of early cervical cancer have been published because it is considered unethical to enlist young participants who want to preserve fertility to an arm that involves infertility. In addition, such a study would need a rather large sample size to demonstrate equivalence between the methods. However, matched case–control studies from the USA and Canada have confirmed RVT to be an oncological safe procedure with a recurrence rate of less than 5%.

Imaging

The use of diagnostic imaging by computed tomography, magnetic resonance imaging, positron emission tomography (or both positron emission tomography and computed tomography) to assess the size and eventual spread of the primary tumour is encouraged, but not mandatory. Its use to rule out affection of the urinary system, lymphatic or systemic metastasis is applied differently among centres. Magnetic resonance imaging is by some centres considered crucial in characterizing the tumour before selecting patients for RVT. In my institution, positron emission tomography and computed tomography are used routinely to rule out spread of the disease and conisation is carried out before the trachelectomy in order to characterize the growth optimally. Frequently, the upper margin in the cone specimen will be close, but clear so the purpose of the following trachelectomy is to confirm this and if necessary extend the resection margins.

Pathology

With fertility-preserving procedures, FIGO stage should be characterized in a well-conducted cone biopsy assessed by a dedicated pathologist. I believe such conisation is necessary to plan the amount of cervix to be removed for radical surgery. Cone biopsy is carried out even when the tumour is visible to find out the extent along the cervical canal and depth of infiltration. If the cone specimen is ‘radical’, the following trachelectomy will represent the required extra margin. If the crucial upper margin is clear, further resection of the cervix is saved. The vaginal margin is easier as additional resection of the vaginal wall does not compromise fertility. To evaluate trachelectomy surgical margins intraoperatively, a special protocol is suitable.

Selection of cases and safety

The criteria for carrying out RVT have been almost unchanged for more than 10 years ( Table 1 ). Cervical cancer spreads predominantly in a lateral direction to parametrium, cardinal ligaments and upper vagina, but small stage IB1 carcinoma infrequently extends to the corpus uteri. No randomized–controlled trials comparing RVT and radical hysterectomy in the treatment of early cervical cancer have been published because it is considered unethical to enlist young participants who want to preserve fertility to an arm that involves infertility. In addition, such a study would need a rather large sample size to demonstrate equivalence between the methods. However, matched case–control studies from the USA and Canada have confirmed RVT to be an oncological safe procedure with a recurrence rate of less than 5%.

Imaging

The use of diagnostic imaging by computed tomography, magnetic resonance imaging, positron emission tomography (or both positron emission tomography and computed tomography) to assess the size and eventual spread of the primary tumour is encouraged, but not mandatory. Its use to rule out affection of the urinary system, lymphatic or systemic metastasis is applied differently among centres. Magnetic resonance imaging is by some centres considered crucial in characterizing the tumour before selecting patients for RVT. In my institution, positron emission tomography and computed tomography are used routinely to rule out spread of the disease and conisation is carried out before the trachelectomy in order to characterize the growth optimally. Frequently, the upper margin in the cone specimen will be close, but clear so the purpose of the following trachelectomy is to confirm this and if necessary extend the resection margins.

Pathology

With fertility-preserving procedures, FIGO stage should be characterized in a well-conducted cone biopsy assessed by a dedicated pathologist. I believe such conisation is necessary to plan the amount of cervix to be removed for radical surgery. Cone biopsy is carried out even when the tumour is visible to find out the extent along the cervical canal and depth of infiltration. If the cone specimen is ‘radical’, the following trachelectomy will represent the required extra margin. If the crucial upper margin is clear, further resection of the cervix is saved. The vaginal margin is easier as additional resection of the vaginal wall does not compromise fertility. To evaluate trachelectomy surgical margins intraoperatively, a special protocol is suitable.

Tumour size (volume)

The size of the lesion is considered to be the most important risk factor, as the recurrence rate significantly increases when the tumour is above 2 cm. Roughly 90% of recurrent cases are within that range.

Lymphovascular invasion

Studies on radical hysterectomy show that lymphovascular space involvement increases the risk for recurrence. In some studies, more than half of the women considered for trachelectomy had this unfavourable characteristic. It has therefore been discussed whether it should be considered an exclusion criteria instead. However, most groups have included these patients. Most women who had recurrence nevertheless had tumours with lymphovascular invasion.

Histology

Adenocarcinoma or adenosquamous carcinomas constitutes more than one-third of tumours in women treated with RVT, and these types do not seem to increase the risk of recurrence. Some women with small-cell neuroendocrine tumours have been reported with rapid recurrence, and many clinicians will discourage RVT in such cases and rather recommend radical hysterectomy and chemotherapy.

Resection margins and nodal metastasis

About 10% of women will need further treatment owing to positive margins or metastatic lymph nodes. Also, margins in the final pathology will sometimes be closer than found during surgery by frozen section. For these women, individual decisions about definite treatment have to be settled.

Operative technique

Technically, the procedure is a combination of modern laparoscopic surgery and classic vaginal skills. The original method for lymphadenectomy in the pelvis by laparoscopy was published 1991 by Querleu et al., and used to commence the operation called Coelio Schauta – a minimally invasive alternative to the open Wertheim operation. The vaginal approach to removing the tumour with adequate margins in the cervix and parametrium is similar to the beginning of a radical vaginal hysterectomy, a method evolved by Schauta who was contemporary with Wertheim in Vienna in the early 1900. Until now, relatively few specialized centres have reported on vaginal trachelectomy, and probably this is due to the limited opportunity to learn radical vaginal surgery through apprenticeship. The procedures are seldom indicated, so there is a need to centralize cases in order to overcome the learning curve and keep the surgical team well trained. All RVTs in Denmark (10–15 each year; 5.2 million inhabitants) are carried out at the Copenhagen University Hospital, Rigshospitalet.

Lymphadenectomy in the pelvis

To evaluate lymph-node status, the external iliac, superficial obturator and interiliac nodes are removed by laparoscopy, and suspicious nodes are sent for frozen section. In case of metastasis, surgery is abandoned and the woman is prepared for chemo-radiation. A node harvest is between 20 and 30 lymph nodes.

Vaginal part

About 2 cm of vaginal cuff is defined, injected with a local anaesthetic containing vasoconstrictor, cut and dissected. The vesicouterine fold is developed sharply, but the peritoneum over the anterior cul-de-sac is not opened. The paravesical space is tunnelled and entered with a small retractor, and the ureter is identified so that the descending branch of the uterine artery can be located and divided. In my practice of more than 80 RVTs, clamping and suturing was initially carried out. More recently, bipolar diathermy has become the preferred tool of choice, as it is the quick, simple and economic to use. After opening the posterior cul-de-sac and dividing the sacrouterine ligaments, the parametrium can be cut and secured. The uterus is now mobile and the isthmic region can be examined and measured with a uterine sound to define the level of the internal orifice. The specimen is cut with monopolar diathermy leaving about 1 cm of the cervix. The upper margin is cut from the specimen with a cold knife, and together with the main specimen sent for immediate pathological examination aiming for a 10 mm free margin. This consistent use of cold knife and diathermy will enable the pathologist to distinguish the surgical margins reliably, which is of great importance in cases of residual disease.

The reconstruction starts with cerclage of the cervix to improve its competency. In our institution, we have used a 2:0 Gore-Tex stitched according to McDonald encircling the cervix two times and the knot is placed posterior, where it can be well covered by vaginal epithelium. This preferred material is easy to handle, strong though soft and seems to incorporate well into the neo-cervix. Finally, vaginoisthmic anastomosis is carried out by systematically suturing the vaginal wall to the cervix with a resorbable 2:0 Monocryl. Some will place some kind of stent in the remaining cervical canal in order to prevent future stenosis, but the efficacy of this action has not been documented. The vaginal technique is beautifully described and illustrated in Dargent’s original work. For a detailed and comprehensive description, Sonoda and Abu-Rustum is recommended. Although lymphadenectomy is mandatory, an open retroperitoneal approach can be used if equipment and laparoscopists are not available. The vaginal part of the operation is more demanding than routine vaginal surgery.

Postoperative regimen

The patient is encouraged to mobilize as much as possible after surgery. The catheter is removed the next day, voiding is monitored and intermittent catheterization used and taught to the patient when necessary.

Operative results

The operating time is around 3 h, including the time waiting for pathology. Blood loss is usually less than 400 ml.

Complications

Reported complication rates of RVT are in the range of 8–13%, comprising ureteric damage and bladder perforation, and bleeding reported in the early part of the learning curve. Many of these problems are cystotomies that are simple to deal with intraoperatively. The postoperative complication rate is about 12%. Isthmic stenosis pain, lymph cysts, lymph oedema and thigh numbness are the most frequent problems. The rate of 5% isthmic stenosis is probably related to the amount of cervix actually resected in order to obtain radicality, and some investigators argue that the whole cervical canal should be excised in cases of adenocarcinoma. The use of a clinician checklist as a clinical instrument has been proposed.

What to do in the case of stenosis

Women undergoing RVT, who develop stenosis, will experience cyclic pain with no or sparse and protracted menstruation in combination with dilatation of the cavum uteri visible on ultrasound. In my experience, this is primarily a result of the vaginal epithelium growing over the external orifice. Inspection of the vaginal vault shows wrinkles converging to the point where the opening last presented. Sometimes it is obvious that small canals have evolved under the epithelium, creating dark red brown nodules resembling endometriotic stains. Actually, this can be considered an iatrogenic variety of endometriosis as histologic examination of the resected epithelium regularly confirms. Finding the external orifice for dilatation is not straight forward, and this condition can be managed by using a small loop electrosurgical excision procedure electrode to excise the described epithelium overlying the canal. If the excision is confined only to include the epithelial thickness, it will not damage the cerclage or impair the cervical competence but relieve the pressure inside the uterus. Such measures can be necessary as preparation for fertility assistance.

Completion of treatment

A number of women (9%) have required further treatment because of adverse prognostic factors that arise after surgery, such as metastatic lymph nodes (chemo-radiation). Close or involved surgical margins usually request repeated radical surgery.

Follow up

Evidence does not support the value of follow-up visits, and traditionally we follow these women clinically every 4 months for 2 years, and every 6 months thereafter. Cytology is checked once a year and imaging only on special indication. Colposcopy with frequent cytological smears is recommended by several centres, but interpretation of these smears can sometimes be difficult as they might show glandular atypia of endometrial origin. In addition, up to 41% of smears contain squamous cells only. Re-epithelialization of the vaginal vault and remaining cervix is usually completed within 6 months, but the patient should be free to conceive whenever she wishes. Routine hysterectomy after achievement of child bearing is not recommended; however, a few women will need this when they present with symptoms such as pain caused by stenosis or adenomyosis.

Consensus is lacking on whether to use imaging routinely during follow up or to use it only when clinically indicated. Imaging appearances of the female pelvis after RVT include diffuse vaginal wall thickening, haematomas, lymphoceles, exaggerations of the pelvic venous plexuses, as well as adenomyosis and endometriosis.

Oncological outcome

Case-control studies have confirmed the safety of RVT, with a recurrence rate of about 5% and 5-year survival of 99%.

Reproductive outcome

The greater part of couples will conceive by themselves, although a fair number will need assisted reproductive advice or interventions as these women are already facing declining fertility due to age, and for that reason cannot afford to wait in vain.

Pregnancy rates vary between 41 and 79% in studies of women actually attempting pregnancy. Prematurity is a significant problem for 12% of live births. The rate of miscarriage in the first trimester is the same as in the general population (16–20%), but in the second trimester the rate is doubled to 8–10%. The reason for this is most likely ascending infection through the shortened cervix, void of mucus plug resulting in chorioamnionitis, and premature rupture of membranes. The same mechanism is considered responsible for the reported preterm delivery rates of between 20 and 30%.

What to do in case of miscarriage

In early cases, initial medical approach with misoprostol is recommended. It is, however, possible to evacuate the uterus through the cervix which, in spite of the cerclage, can be dilated to Hegar-size 7–8, making this an option up to weeks 10–12. The procedure needs to be carried out by an experienced gynaecologist familiar with the appearance of the vaginal vault in these women, since the external orifice can be difficult to find and grasp, increasing the risk of perforation and incomplete emptying. Ultrasonic and even laparoscopic guidance has been proposed in these situations. In case of second trimester miscarriage, evacuation by means of hysterotomy will sometimes be necessary as, in my experience, the cerclage is not easily accessible for removal.

Follow-up during pregnancy

It is generally agreed among centres that women who become pregnant after RVT should be followed in close collaboration with an associated maternal–fetal specialist. Various suggestions, such as routine genital tract infection screening, prophylactic antibiotics, reduced physical activity or bed rest, and routine steroid have been put forward but are all without firm evidence of benefit for this special cohort. In case of premature rupture of membranes without signs of infection, it is recommended to be expectant until 32–34 weeks of pregnancy. Ultrasonic monitoring of the cervical length and dilatation as well as the position of the cerclage is another possibility of surveillance.

Mode of delivery

Because of permanent cerclage, which cannot be removed by simple means at time of delivery, a caesarean section is planned around the end of week 37. Standard operative technique applies, and it is recommended to place the uterotomy in the upper isthmus to avoid pushing the bladder down and to be careful not to tear the uterine vessels. Several of these women have given birth more than once.

Concerns

Regardless of superior oncological and obstetrical outcomes, significant amounts of distress and reproductive concerns have been identified in women planning fertility-preserving surgery and those who have undergone it. These worries have lasted up to 6 months after surgery but seem to be reversible. Preoperative information is important in preparation before RVT as is the postoperative support for these women.