- Email: [email protected]
Surgical morbidity associated with radical trachelectomy and radical hysterectomy
Gynecologic Oncology 101 (2006) 450 – 454 www.elsevier.com/locate/ygyno
Surgical morbidity associated with radical trachelectomy and radical hysterectomy F. Alexander-Sefre *, N. Chee, C. Spencer, U. Menon, J.H. Shepherd Department of Gynaecological Oncology, St. Bartholomew’s Hospital, North East London Gynaecological Cancer Centre, London, UK Received 8 June 2005 Available online 15 December 2005
Abstract Objectives. To examine intra-operative, short and longer term morbidity associated with Radical Trachelectomy (RT) within our cohort of patients, compared with conventional Radical Hysterectomy (RH). Methods. A retrospective case note review comprising 29 RT and 50 RH patients. Patients who required adjuvant treatment were excluded. Operative data, short-term and long-term complications were recorded. Long-term problems were included only if they had been of persistent nature. Results. The median age and follow up period for the RH group were 40 years and 60 months and for RT group 30.5 years and 25 months. There were significant differences between RH and RT groups in median operative time (260 versus 187 min), blood loss (1000 versus 400 ml), transfusion requirement (75% versus 12%), analgesic requirement (8 versus 3.5 days) and hospital stay (11 versus 6 days). Bladder hypotonia requiring prolonged catheterisation was more frequent in RH group (P = 0.004). There was no apparent difference in psychosexual problems between the two groups (17 versus 16%). Complications of deep dyspareunia (P = 0.009), excessive vaginal discharge (P = 0.01), and upper thigh parasthesia (P = 0.05) were noted to be significantly higher in the RT group. Specific problems encountered in the RT group included; dysmenorrhea (24%), irregular menstruation (17%), recurrent candidiasis (14%), cervical suture problems (14%), isthmic stenosis (10%) and prolonged amenorrhea (7%). Conclusion. RT is associated with less operative and short-term morbidity compared with conventional RH. In addition to specific complications associated with RT, in our experience, there are long-term morbidities that are not as commonly observed in RH. D 2005 Elsevier Inc. All rights reserved. Keywords: Radical trachelectomy; Radical hysterectomy; Morbidity; Cervical cancer
Introduction Cervical cancer is the second commonest gynaecological malignancy world wide. Since the introduction of cervical screening programmes, there is an increasing number of young patients in reproductive years with early stage, low volume disease. Nearly 15% of all cervical cancers and 45% of surgically treated stage IB cervical cancers occur in childbearing women under age 40 [1,2]. Whilst it is generally accepted that superficially invasive disease (stage 1A1) can be treated * Corresponding author. Department of Gynaecological Oncology, Ward 24Glasgow Royal Infirmary, Castle Street, Glasgow G4 0SF, UK. Fax: +44 141 232 0860. E-mail address: [email protected] (F. Alexander-Sefre). 0090-8258/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2005.11.007
with conisation and hence preservation of fertility, radical hysterectomy or radiotherapy remains the mainstay of treatment of tumour beyond this stage in most centres worldwide. Such treatments produce good survival results, but may be regarded as an over treatment with subsequent loss of fertility and significant morbidity in certain cases. In 1994, Dargent first described radical vaginal trachelectomy (RT) after laparoscopic pelvic lymphadenectomy before the Society of Gynaecological Oncologists as a novel conservative surgical fertility preservation approach to management of cervical cancer patients with early stage disease [3]. Since then, several other presentations and publications have followed which has made RT an acceptable surgical treatment for a selected group of young patients with early stage disease wishing to preserve their fertility [4– 11]. However, despite adequate long-term results on survival and fertility rates for RT,
F. Alexander-Sefre et al. / Gynecologic Oncology 101 (2006) 450 – 454
there are no data on short- and long-term morbidity of RT when compared with conventional radical hysterectomy (RH). The objective of this study was to examine intra-operative, short-term (first 6 weeks) and longer term morbidity associated with RT within our cohort of patients as compared with conventional RH. Patients and methods In total, 29 and 50 consecutive patients were identified that had undergone RT and RH, respectively, without the need for adjuvant treatment or evidence of recurrent disease. All the patients were assessed with magnetic resonance imaging scans of pelvis and abdomen, and examination under anaesthesia to ensure that the disease did not extend to the endocervix or parametrium. The pathology of the diagnostic biopsy was reviewed to confirm that there was invasive disease and to check the depth of invasion and width of the lesion. The patients had been counselled concerning the standard accepted management of RH in this situation, and the uncertain nature of the cure rates and fertility rates for RT at the time of study. The women and their partners also received counselling and support from the clinical nurse specialist. All the operations were carried out at St Bartholomew’s Hospital, a tertiary referral centre for gynaecological oncology, during a 5-year period from 1996 to 2001. The operations were carried out by Fellows in Gynaecological Oncology or Senior Registrars in Obstetrics and Gynaecology under the direct supervision of experienced gynaecological oncologists and pelvic surgeons. Case notes were reviewed. The operative data and complications during the first 6 postoperative weeks were recorded based on a proforma. This included information on duration of surgery from knife to skin to closure, record of blood loss as estimated by the surgeon, need for blood transfusion, intraoperative complications, febrile episodes (defined as body temperature over 38-C after day one) and length of hospital stay. Gastrointestinal complications including: protracted nausea and vomiting lasting over 2 days, prolonged ileus as defined by absence of bowel sounds over 3 days and bowel obstruction based on both clinical and radiological diagnosis were recorded. Urological complications such as; bladder hypotonia (high residual urine after 10th postoperative day requiring prolonged catheterisation), positive urine culture, presence of fistula were noted. Problems with wound dehiscence (partial or complete), wound sepsis (confirmed on microbiological swabs) and wound haematoma were recorded. Postoperative surgical complications such as pelvic haematoma and presence of lymphocyst as confirmed on imaging, deep venous thrombosis based on positive Doppler flow studies were noted. These were conducted only if patients had complained of symptoms suggestive of such complications. Cardiac problems that required input from cardiology team were also documented. Any respiratory complications including pneumonia when confirmed by positive sputum culture and chest X-ray, atelectasis and pulmonary embolism confirmed by ventilation perfusion scans were documented. Longer term morbidity beyond the first 6 weeks of surgery were identified by examining records of readmission or out patient follow up notes indicating persistent complaints. The diagnosis of psychological and sexual disorders were purely subjective and based on repeated records of patients complaints. The technique of RT has been described thoroughly elsewhere [12,13]. In brief; the procedure begins with a transperitoneal pelvic lymphadenectomy to rule out presence of lymph node metastasis. All the lymphatic tissue along the external and internal iliac vessels, up to the lower common iliac vessels, and around the obturator nerve are removed. Any suspicious nodes are submitted for frozen section examination. After the laparoscopic part of the procedure, the patient is repositioned in the lithotomy position as for a vaginal hysterectomy. At first, a Schuhardt incision was used for additional vaginal access. This is not necessary in the majority of women with deeper vaginal retractors being used. Infiltration of the vaginal epithelium using bupivicane facilitates the formation of the vaginal cuff. A sleeve is formed by incising circumferentially, approximately 2 cm distal to the cervix. Sharp dissection is used to separate the vagina from adjacent structures and this cylinder of tissue is grasped over the cervix for retraction. The anterior part of the specimen is further prepared by identifying the vesicovaginal space and hence the isthmus of the uterus. The bladder pillars were defined by formation of the uterovesical space and the lateral paravesical spaces. The ureters are then palpated in the pillar and are
451
mobilised cranially by blunt dissection. The descending vaginal and cervical branches of uterine arteries are divided and ligated. The peritoneal covering over Pouch of Douglas is swept away to define the uterosacral ligaments. The ligaments are divided and transfixed approximately 1 – 2 cm from the cervix. The cardinal ligaments are defined between paravesical spaces anteriorly and pararectal spaces posteriorly. Again, 1 – 2 cm of tissue is obtained lateral to the cervix to ensure an adequate margin of resection. The endocervical canal is identified using a Hegar dilator (Size 6) and the cervical isthmus is transected using diathermy. Frozen sections from the isthmic margin and proximal margin are examined to ensure absence of any residual disease. A single monofilament non-absorbable suture (1 nylon) is inserted in the remains of the isthmus as a cerclage similar to traditional Shirodkar suture. This is tightened around a Hagar 6 dilator. Using polyglycolic sutures the vaginal epithelium is re-attached to the isthmic (endocervical) epithelium at the uterine margin ensuring that the canal remains open. The remaining vaginal epithelium is closed transversely. Postoperatively patients remained in hospital until able to tolerate a regular diet, were fully mobile and bladder and bowel function returned to normal. Trial without catheter and measurement of residual urine following micturition was performed in all patients prior to discharge. Patients with a large postvoidal residual urine (>100 ml) as verified by catheterisation or with a bladder scan were advised to continue with indwelling catheter for a longer period. Women were asked to avoid pregnancy in the first 6 months postoperatively.
Results Intraoperative and short-term complications The median age and follow up periods for the RH group were 40 (21 – 65) years and 60 (2 –184) months, respectively, and 30.5 (23 – 36) years and 25 (7– 66) months for RT group. The median operative time for RT was 260 (125 – 395) min as opposed to 187 (110– 280) min for RH group (P = 0.0001). The median intraoperative blood loss was 400 (100 – 1000) ml for RT group and 1000 (350 –4000) ml for RH group (P = 0.0001). Seventy-five percent (34/45) of RH patients required 1 to 8 units of blood transfusion intraoperatively or soon after their operations. In contrast, only 12% (3/25) of RT patients required 2 to 4 units of blood transfusion (P = 0.0001). There were 3 documented intra operative complications within 25 RT cases (12%) and 5 in 47 RH cases (10%) (P = 0.61). The operative complications within RT group were 2 vascular injuries, one requiring laparotomy and one cystotomy. In the RH group, there were 4 vascular injuries and one cystotomy. All the vascular injuries were dealt with by gynaecological oncologists and none required assistance from vascular surgeons (Table 1). The need for regular postoperative analgesia was on average 3.5 days for RT group and 8 days for RH group (P = 0.004). Table 1 Comparative short-term morbidity (with in first 6 postoperative weeks) between RT and RH groups Morbidity
RT
RH
P value
Operative time (min) Blood loss (ml) Transfusion rate Hospital stay (days) Operative complication Regular analgesia (days) Febrile condition Bladder hypotonia
260 400 12% 6 12% 3.5 14% 4%
187 1000 75% 11 10% 8 18% 31%
0.0001 0.0001 0.0001 0.0001 0.61 0.004 0.43 0.004
(3/25) (3/25) (4/28) (1/27)
(34/45) (5/47) (9/43) (15/48)
452
F. Alexander-Sefre et al. / Gynecologic Oncology 101 (2006) 450 – 454
There was a significant difference between the median length of hospital stay between the 2 groups; 11 days (5 – 31) for RH and 6 days (3– 12) for RT group (P = 0.0001). Febrile conditions were noted in 14% (4/28) of RT group and 18% (9/48) of RH cases whilst in hospital (P = 0.43). Bladder hypotonia requiring prolong catheterisation was noted in 4% (1/27) of RT patients as compared with 31% (15/48) of RH patients (P = 0.004). Swelling of the vulva lasting for 3 days was noted in 1 RT case. There were no significant reported gastrointestinal or cardiorespiratory complications in the two groups in their 6 week postoperative period. Long-term complications Within RH group 5 patients (10%) had recorded evidence of persistent anxiety or depression requiring treatment. There were similar number of patients in RT group (17%), but this finding did not reach statistical significance (P = 0.27). Three RH patients (6%) reported loss of sexual desire or sensation. Such complaint was not recorded in RT group (Table 2). Trends towards higher incidence of lymphoedema, lymphocyst formation and urinary incontinence were noted within the RH group; 7 patients presented with symptomatic lymphocysts in RH group as compared with 1 in RT group (P = 0.13). Peripheral lymphodema was recorded in 8 cases post-RH and 1 case post-RT (P = 0.08). Urinary stress incontinence was noted in 5 RH patients and none of RT (P = 0.09). Complaints of deep dyspareunia were noted in 6 RT and 1 RH patient (P = 0.009). Excessive vaginal discharge was reported by 4 RT patients and none of RH patients (P = 0.01). Leg parasthesia over upper part of thigh and weakness in adduction was reported more frequently in RT (4 patients) group than RH (1 patient) (P = 0.05). Granulation tissue formation requiring cauterisation was noted in 4 RH and 3 RT patients (P = 0.52). Chronic pelvic pain was reported by 4 RH patients and 3 RT patients (P = 0.50). Two RH patients required further surgery for problems associated with their abdominal wound scars. Such problem was unreported in RT group. Specific problems encountered in the RT group included; dysmenorrhoea (24%, 7/29), irregular menstruation or intermenstrual bleeding (17%, 5/29), recurrent candidiasis (14%, 4/ Table 2 Comparative long-term morbidity (beyond the first 6 postoperative weeks) between RT and RH groups Morbidity
RT group (%)
RH group (%)
P value
Stress incontinence Lymphodema Lymphocyst Wound problem Vaginal granulation tissue Chronic pelvic pain Dysparaunia Vaginal discharge Neuropathy Psychological Loss of sexual desire/sensation
0 3 3 0 10 10 20 14 14 17 0
10 16 14 4 8 8 2 0 2 10 6
0.09 0.08 0.13 0.15 0.52 0.50 0.009 0.01 0.05 0.27 0.24
(1/29) (1/29) (3/29) (3/29) (6/29) (4/29) (4/29) (5/29)
(5/50) (8/50) (7/50) (2/50) (4/50) (4/50) (1/50) (1/50) (5/50) (3/50)
Table 3 Specific problems associated with RT Morbidity
Prevalence (%)
Dysmenorrhoea Dysplastic smears Irregular or intermenstrual bleeding Problems with cercalage suture Excessive vaginal discharge Isthmic stenosis Amenorrhea
24 24 17 14 14 10 7
(7/29) (7/29) (5/29) (4/29) (4/29) (3/29) (2/29)
29), isthmic stenosis requiring dilatation (10%, 3/29) and prolonged amenorrhea (7%, 2/29) (Table 3). Dysplastic isthmic smears were reported in 24% (7/29) of cases over the study period. In addition there were problems associated with cervical cerclage suture in 4 cases (14%); in 1 case the suture had been rejected, and another was cut to allow spontaneous abortion to be completed . They both underwent abdominal reinsertion. In the 2 remaining cases the suture had to be trimmed due to interference with intercourse. Discussion The primary outcome measure of any cancer treatment is undoubtedly survival but other factors such as associated morbidity and quality of life cannot be ignored. Patients therefore need to be aware of potential complications and longterm morbidities associated with any proposed procedure and in particular in gynaecological oncology the impact on fertility, sexual function and body image. In addition, before making a decision about the choice of surgery patients need to make a comparison with alternative surgical approaches and their associated benefits and morbidities. Based on the current literature, RT has comparable survival and recurrence rates to conventional RH and as a result it is becoming an established surgical approach for patients with an early stage disease who wish to preserve their fertility [4– 11]. However, despite the reported intraoperative or postoperative morbidities by various groups, the more long-term associated problems with RT and its comparison with conventional RH has not as yet been fully investigated. In our study we have shown that RT is associated with significantly less morbidity when compared with RH and the patients are discharged home much earlier. RT patients are less likely to have major blood loss and their requirement for regular analgesia is much less. Furthermore, the need for prolonged catheterisation due to bladder hypotonia happens rarely in RT. The main drawback for RT, however remains the longer mean operative time, which is partly due to the slow learning curve in early cases, as has been confirmed with further experience. These findings are in concordance with previous reported series from major North American and European centres (Table 4). The rate of accidental intraoperative injuries for RT in our study does not differ significantly from RH and are mainly vascular injuries or cystotomies that were all dealt by gynaecological oncologists. Such injuries again happened in earlier cases in RT group whilst establishing
F. Alexander-Sefre et al. / Gynecologic Oncology 101 (2006) 450 – 454 Table 4 Comparative intraoperative findings from 5 major series Centre
Toronto
Lyon
LA
Quebec
London
Author No OR time (min) Blood Loss (ml) Transfusion rate % Hospital stay (days)
Covens 81 180 300 9.4 1
Dargent 82 130 NR 6.5 7
Morrow 21 318 293 NR 3
Roy 44 240 200 1.4 3
Shepherd 29 260 400 12 6
the technique and only 1 case required conversion to laparotomy. Rarity of severe intraoperative complications has been also reported by other groups [6,10]. Examining the long-term reported morbidities, there appear to be problems that never occurred within the RT group and were specifically related to RH. These included, loss of sexual desire or sensation, stress incontinence and abdominal wound problems. Such complaints are most likely a direct result of radical pelvic surgery or an indirect psychological impact of this operation. The higher incidence of lymphodema and lymphocyst formation in RH group may also be due to more radical surgery and hence greater disruption of lymphatics. Similarly, there are problems specific to RT group, for instance, deep dysparaunia, excessive vaginal discharge and recurrent candidiasis. The excessive vaginal discharge and recurrent candidiasis are thought to be most likely due to presence of cervical cerclage suture acting as a niche for abnormal vaginal organisms, despite usage of monofilament nylon suture material and coverage by vaginal epithelium. Deep dyspareunia, in two patients was noted to be due to the cervical suture, requiring trimming of the protruding ends. One must bear in mind however, that the higher level of reported anxiety or depression may have a confounding effect in the complaint of deep dyspareunia. However, this is only an observation which requires verification in prospective studies with objective means of assessment and with a control group from general population. In two cases the cervical suture had fallen out necessitating laparotomy and abdominal reinsertion. The RT procedure as carried out in our unit involves complete excision of cervix and therefore no cervical tissue in which one could safely reinsert the suture vaginally. It was therefore deemed safer and more practical to insert this second suture around the isthmus from above using an abdominal approach under direct vision. In our centre, RH is limited to patients with early disease not exceeding 4 cm in maximum diameter. A standard surgical approach was followed which involved type I RH and systematic dissection of lymphatic tissue along common iliac, internal, external iliac vessels and obturator fossa. Similarly patients only qualified for RT if their longest tumour diameter did not exceed 2 cm and was 1 cm away from internal cervical os. They similarly underwent systematic node dissection like RH group. Because of this standard surgical approach and limited number of patients in each study group, further stratification based on variations in tumour size and the number of previous cone biopsies were avoided and only morbidity data was collected. However it would be of interest
453
to examine such minor variations between the patients in future large prospective studies. In addition future comparative studies with laparoscopic Radical Hysterectomy which is not yet an established technique in any centre in UK would be informative. Cervical stenosis was not an uncommon problem in our series and the 3 patients who presented with this had to undergo examination under anaesthesia and isthmic dilatation. In two of the three patients the cervical os was opened directly with use of dilators and in one case ultrasound guidance was needed. Surprisingly, such complaint has not been reported in a larger study by Plante et al. [10]. This may be due to slight variation in surgical technique which needs to be investigated further. Parasthesia over upper medial aspect of thigh and some degree of weakness in adduction of leg were noted in 14% of our patients. This is most likely due to manipulation and direct thermal or electrical injury by monopolar diathermy to cutaneous branches of obturator, ilioinguinal and genitofemoral nerves. Since the study, our practice has been modified by using harmonic scalpel and bipolar diathermy in order to reduce such complications. Most countries in which RT is practiced have formed a central register for their patients pooling follow up data and fertility outcomes. To our best knowledge, however none have so far included record of short- and long-term physical and psychological morbidities. Introduction of this additional data would be of great importance for patients in order to reach a well-informed decision and to decide whether the indications for RT should be expanded. In conclusion, RT is associated with less operative and short-term morbidity compared with conventional RH. There are specific complications associated with RT and there are also long-term morbidities that are more commonly observed in RH patients. References [1] Covens A, Rosen B, Murphy J, Laframboise S, DePetrillo AD, Lickrish G, et al. Changes in the demographics and perioperative care of stage IA(2)/IB(1) cervical cancer over the past 16 years. Gynecol Oncol 2001;81(2):133 – 7. [2] Sonoda Y, Abu-Rustum NR, Gemignani ML, Chi DS, Brown CL, Poynor EA, et al. A fertility-sparing alternative to radical hysterectomy: how many patients may be eligible? Gynecol Oncol 2004; 95(3):534 – 8. [3] Dargent D, Burn JL, Roy M, Remi I. Pregnancies following radical trachelectomy for invasive cervical cancer. Gynaecol Oncol 1994;52:105 [Abstract 14]. [4] Shepherd JH, Mould T, Oram DH. Radical trachelectomy in early stage carcinoma of the cervix: outcome as judged by recurrence and fertility rates. Bri J Obstet Gynaecol 2001;108(8):882 – 5. [5] 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(11): 2273 – 9. [6] 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(8):1877 – 82. [7] Schlaerth JB, Spirtos NM, Schlaerth AC. Radical trachelectomy and pelvic lymphadenectomy with uterine preservation in the treatment of cervical cancer. Am J Obstet Gynecol 2003;188(1):29 – 34.
454
F. Alexander-Sefre et al. / Gynecologic Oncology 101 (2006) 450 – 454
[8] Lee CL, Huang KG, Wang CJ, Yen CF, Lai CH. Laparoscopic radical trachelectomy for stage Ib1 cervical cancer. J Am Assoc Gynecol Laparosc 2003;10(1):111 – 5. [9] Burnett AF, Roman LD, O’Meara AT, Morrow CP. Radical vaginal trachelectomy and pelvic lymphadenectomy for preservation of fertility in early cervical carcinoma. Gynecol Oncol 2003;88(3):419 – 23. [10] Plante M, Renaud MC, Francois H, Roy M. Vaginal radical trachelectomy: an oncologically safe fertility-preserving surgery. An updated series of 72 cases and review of the literature. Gynecol Oncol 2004;94(3):614 – 23.
[11] Plante M, Roy M. New approaches in the surgical management of early stage cervical cancer. Curr Opin Obstet Gynecol 2001; 13(1):41 – 6. [12] Plante M, Renaud MC, Francois H, Roy M. Vaginal radical trachelectomy. In: Levine d, Barakat RR, editors. Atlas of Procedures in Gynaecological Oncology. London’ Martin Dunitz; 2003. p. 207 – 21. [13] Shepherd JH, Crawford RA, Oram DH. Radical trachelectomy: a way to preserve fertility in the treatment of early cervical cancer. Br J Obstet Gynaecol 1998;105(8):912 – 6.
Surgical morbidity associated with radical trachelectomy and radical hysterectomy F. Alexander-Sefre *, N. Chee, C. Spencer, U. Menon, J.H. Shepherd Department of Gynaecological Oncology, St. Bartholomew’s Hospital, North East London Gynaecological Cancer Centre, London, UK Received 8 June 2005 Available online 15 December 2005
Abstract Objectives. To examine intra-operative, short and longer term morbidity associated with Radical Trachelectomy (RT) within our cohort of patients, compared with conventional Radical Hysterectomy (RH). Methods. A retrospective case note review comprising 29 RT and 50 RH patients. Patients who required adjuvant treatment were excluded. Operative data, short-term and long-term complications were recorded. Long-term problems were included only if they had been of persistent nature. Results. The median age and follow up period for the RH group were 40 years and 60 months and for RT group 30.5 years and 25 months. There were significant differences between RH and RT groups in median operative time (260 versus 187 min), blood loss (1000 versus 400 ml), transfusion requirement (75% versus 12%), analgesic requirement (8 versus 3.5 days) and hospital stay (11 versus 6 days). Bladder hypotonia requiring prolonged catheterisation was more frequent in RH group (P = 0.004). There was no apparent difference in psychosexual problems between the two groups (17 versus 16%). Complications of deep dyspareunia (P = 0.009), excessive vaginal discharge (P = 0.01), and upper thigh parasthesia (P = 0.05) were noted to be significantly higher in the RT group. Specific problems encountered in the RT group included; dysmenorrhea (24%), irregular menstruation (17%), recurrent candidiasis (14%), cervical suture problems (14%), isthmic stenosis (10%) and prolonged amenorrhea (7%). Conclusion. RT is associated with less operative and short-term morbidity compared with conventional RH. In addition to specific complications associated with RT, in our experience, there are long-term morbidities that are not as commonly observed in RH. D 2005 Elsevier Inc. All rights reserved. Keywords: Radical trachelectomy; Radical hysterectomy; Morbidity; Cervical cancer
Introduction Cervical cancer is the second commonest gynaecological malignancy world wide. Since the introduction of cervical screening programmes, there is an increasing number of young patients in reproductive years with early stage, low volume disease. Nearly 15% of all cervical cancers and 45% of surgically treated stage IB cervical cancers occur in childbearing women under age 40 [1,2]. Whilst it is generally accepted that superficially invasive disease (stage 1A1) can be treated * Corresponding author. Department of Gynaecological Oncology, Ward 24Glasgow Royal Infirmary, Castle Street, Glasgow G4 0SF, UK. Fax: +44 141 232 0860. E-mail address: [email protected] (F. Alexander-Sefre). 0090-8258/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2005.11.007
with conisation and hence preservation of fertility, radical hysterectomy or radiotherapy remains the mainstay of treatment of tumour beyond this stage in most centres worldwide. Such treatments produce good survival results, but may be regarded as an over treatment with subsequent loss of fertility and significant morbidity in certain cases. In 1994, Dargent first described radical vaginal trachelectomy (RT) after laparoscopic pelvic lymphadenectomy before the Society of Gynaecological Oncologists as a novel conservative surgical fertility preservation approach to management of cervical cancer patients with early stage disease [3]. Since then, several other presentations and publications have followed which has made RT an acceptable surgical treatment for a selected group of young patients with early stage disease wishing to preserve their fertility [4– 11]. However, despite adequate long-term results on survival and fertility rates for RT,
F. Alexander-Sefre et al. / Gynecologic Oncology 101 (2006) 450 – 454
there are no data on short- and long-term morbidity of RT when compared with conventional radical hysterectomy (RH). The objective of this study was to examine intra-operative, short-term (first 6 weeks) and longer term morbidity associated with RT within our cohort of patients as compared with conventional RH. Patients and methods In total, 29 and 50 consecutive patients were identified that had undergone RT and RH, respectively, without the need for adjuvant treatment or evidence of recurrent disease. All the patients were assessed with magnetic resonance imaging scans of pelvis and abdomen, and examination under anaesthesia to ensure that the disease did not extend to the endocervix or parametrium. The pathology of the diagnostic biopsy was reviewed to confirm that there was invasive disease and to check the depth of invasion and width of the lesion. The patients had been counselled concerning the standard accepted management of RH in this situation, and the uncertain nature of the cure rates and fertility rates for RT at the time of study. The women and their partners also received counselling and support from the clinical nurse specialist. All the operations were carried out at St Bartholomew’s Hospital, a tertiary referral centre for gynaecological oncology, during a 5-year period from 1996 to 2001. The operations were carried out by Fellows in Gynaecological Oncology or Senior Registrars in Obstetrics and Gynaecology under the direct supervision of experienced gynaecological oncologists and pelvic surgeons. Case notes were reviewed. The operative data and complications during the first 6 postoperative weeks were recorded based on a proforma. This included information on duration of surgery from knife to skin to closure, record of blood loss as estimated by the surgeon, need for blood transfusion, intraoperative complications, febrile episodes (defined as body temperature over 38-C after day one) and length of hospital stay. Gastrointestinal complications including: protracted nausea and vomiting lasting over 2 days, prolonged ileus as defined by absence of bowel sounds over 3 days and bowel obstruction based on both clinical and radiological diagnosis were recorded. Urological complications such as; bladder hypotonia (high residual urine after 10th postoperative day requiring prolonged catheterisation), positive urine culture, presence of fistula were noted. Problems with wound dehiscence (partial or complete), wound sepsis (confirmed on microbiological swabs) and wound haematoma were recorded. Postoperative surgical complications such as pelvic haematoma and presence of lymphocyst as confirmed on imaging, deep venous thrombosis based on positive Doppler flow studies were noted. These were conducted only if patients had complained of symptoms suggestive of such complications. Cardiac problems that required input from cardiology team were also documented. Any respiratory complications including pneumonia when confirmed by positive sputum culture and chest X-ray, atelectasis and pulmonary embolism confirmed by ventilation perfusion scans were documented. Longer term morbidity beyond the first 6 weeks of surgery were identified by examining records of readmission or out patient follow up notes indicating persistent complaints. The diagnosis of psychological and sexual disorders were purely subjective and based on repeated records of patients complaints. The technique of RT has been described thoroughly elsewhere [12,13]. In brief; the procedure begins with a transperitoneal pelvic lymphadenectomy to rule out presence of lymph node metastasis. All the lymphatic tissue along the external and internal iliac vessels, up to the lower common iliac vessels, and around the obturator nerve are removed. Any suspicious nodes are submitted for frozen section examination. After the laparoscopic part of the procedure, the patient is repositioned in the lithotomy position as for a vaginal hysterectomy. At first, a Schuhardt incision was used for additional vaginal access. This is not necessary in the majority of women with deeper vaginal retractors being used. Infiltration of the vaginal epithelium using bupivicane facilitates the formation of the vaginal cuff. A sleeve is formed by incising circumferentially, approximately 2 cm distal to the cervix. Sharp dissection is used to separate the vagina from adjacent structures and this cylinder of tissue is grasped over the cervix for retraction. The anterior part of the specimen is further prepared by identifying the vesicovaginal space and hence the isthmus of the uterus. The bladder pillars were defined by formation of the uterovesical space and the lateral paravesical spaces. The ureters are then palpated in the pillar and are
451
mobilised cranially by blunt dissection. The descending vaginal and cervical branches of uterine arteries are divided and ligated. The peritoneal covering over Pouch of Douglas is swept away to define the uterosacral ligaments. The ligaments are divided and transfixed approximately 1 – 2 cm from the cervix. The cardinal ligaments are defined between paravesical spaces anteriorly and pararectal spaces posteriorly. Again, 1 – 2 cm of tissue is obtained lateral to the cervix to ensure an adequate margin of resection. The endocervical canal is identified using a Hegar dilator (Size 6) and the cervical isthmus is transected using diathermy. Frozen sections from the isthmic margin and proximal margin are examined to ensure absence of any residual disease. A single monofilament non-absorbable suture (1 nylon) is inserted in the remains of the isthmus as a cerclage similar to traditional Shirodkar suture. This is tightened around a Hagar 6 dilator. Using polyglycolic sutures the vaginal epithelium is re-attached to the isthmic (endocervical) epithelium at the uterine margin ensuring that the canal remains open. The remaining vaginal epithelium is closed transversely. Postoperatively patients remained in hospital until able to tolerate a regular diet, were fully mobile and bladder and bowel function returned to normal. Trial without catheter and measurement of residual urine following micturition was performed in all patients prior to discharge. Patients with a large postvoidal residual urine (>100 ml) as verified by catheterisation or with a bladder scan were advised to continue with indwelling catheter for a longer period. Women were asked to avoid pregnancy in the first 6 months postoperatively.
Results Intraoperative and short-term complications The median age and follow up periods for the RH group were 40 (21 – 65) years and 60 (2 –184) months, respectively, and 30.5 (23 – 36) years and 25 (7– 66) months for RT group. The median operative time for RT was 260 (125 – 395) min as opposed to 187 (110– 280) min for RH group (P = 0.0001). The median intraoperative blood loss was 400 (100 – 1000) ml for RT group and 1000 (350 –4000) ml for RH group (P = 0.0001). Seventy-five percent (34/45) of RH patients required 1 to 8 units of blood transfusion intraoperatively or soon after their operations. In contrast, only 12% (3/25) of RT patients required 2 to 4 units of blood transfusion (P = 0.0001). There were 3 documented intra operative complications within 25 RT cases (12%) and 5 in 47 RH cases (10%) (P = 0.61). The operative complications within RT group were 2 vascular injuries, one requiring laparotomy and one cystotomy. In the RH group, there were 4 vascular injuries and one cystotomy. All the vascular injuries were dealt with by gynaecological oncologists and none required assistance from vascular surgeons (Table 1). The need for regular postoperative analgesia was on average 3.5 days for RT group and 8 days for RH group (P = 0.004). Table 1 Comparative short-term morbidity (with in first 6 postoperative weeks) between RT and RH groups Morbidity
RT
RH
P value
Operative time (min) Blood loss (ml) Transfusion rate Hospital stay (days) Operative complication Regular analgesia (days) Febrile condition Bladder hypotonia
260 400 12% 6 12% 3.5 14% 4%
187 1000 75% 11 10% 8 18% 31%
0.0001 0.0001 0.0001 0.0001 0.61 0.004 0.43 0.004
(3/25) (3/25) (4/28) (1/27)
(34/45) (5/47) (9/43) (15/48)
452
F. Alexander-Sefre et al. / Gynecologic Oncology 101 (2006) 450 – 454
There was a significant difference between the median length of hospital stay between the 2 groups; 11 days (5 – 31) for RH and 6 days (3– 12) for RT group (P = 0.0001). Febrile conditions were noted in 14% (4/28) of RT group and 18% (9/48) of RH cases whilst in hospital (P = 0.43). Bladder hypotonia requiring prolong catheterisation was noted in 4% (1/27) of RT patients as compared with 31% (15/48) of RH patients (P = 0.004). Swelling of the vulva lasting for 3 days was noted in 1 RT case. There were no significant reported gastrointestinal or cardiorespiratory complications in the two groups in their 6 week postoperative period. Long-term complications Within RH group 5 patients (10%) had recorded evidence of persistent anxiety or depression requiring treatment. There were similar number of patients in RT group (17%), but this finding did not reach statistical significance (P = 0.27). Three RH patients (6%) reported loss of sexual desire or sensation. Such complaint was not recorded in RT group (Table 2). Trends towards higher incidence of lymphoedema, lymphocyst formation and urinary incontinence were noted within the RH group; 7 patients presented with symptomatic lymphocysts in RH group as compared with 1 in RT group (P = 0.13). Peripheral lymphodema was recorded in 8 cases post-RH and 1 case post-RT (P = 0.08). Urinary stress incontinence was noted in 5 RH patients and none of RT (P = 0.09). Complaints of deep dyspareunia were noted in 6 RT and 1 RH patient (P = 0.009). Excessive vaginal discharge was reported by 4 RT patients and none of RH patients (P = 0.01). Leg parasthesia over upper part of thigh and weakness in adduction was reported more frequently in RT (4 patients) group than RH (1 patient) (P = 0.05). Granulation tissue formation requiring cauterisation was noted in 4 RH and 3 RT patients (P = 0.52). Chronic pelvic pain was reported by 4 RH patients and 3 RT patients (P = 0.50). Two RH patients required further surgery for problems associated with their abdominal wound scars. Such problem was unreported in RT group. Specific problems encountered in the RT group included; dysmenorrhoea (24%, 7/29), irregular menstruation or intermenstrual bleeding (17%, 5/29), recurrent candidiasis (14%, 4/ Table 2 Comparative long-term morbidity (beyond the first 6 postoperative weeks) between RT and RH groups Morbidity
RT group (%)
RH group (%)
P value
Stress incontinence Lymphodema Lymphocyst Wound problem Vaginal granulation tissue Chronic pelvic pain Dysparaunia Vaginal discharge Neuropathy Psychological Loss of sexual desire/sensation
0 3 3 0 10 10 20 14 14 17 0
10 16 14 4 8 8 2 0 2 10 6
0.09 0.08 0.13 0.15 0.52 0.50 0.009 0.01 0.05 0.27 0.24
(1/29) (1/29) (3/29) (3/29) (6/29) (4/29) (4/29) (5/29)
(5/50) (8/50) (7/50) (2/50) (4/50) (4/50) (1/50) (1/50) (5/50) (3/50)
Table 3 Specific problems associated with RT Morbidity
Prevalence (%)
Dysmenorrhoea Dysplastic smears Irregular or intermenstrual bleeding Problems with cercalage suture Excessive vaginal discharge Isthmic stenosis Amenorrhea
24 24 17 14 14 10 7
(7/29) (7/29) (5/29) (4/29) (4/29) (3/29) (2/29)
29), isthmic stenosis requiring dilatation (10%, 3/29) and prolonged amenorrhea (7%, 2/29) (Table 3). Dysplastic isthmic smears were reported in 24% (7/29) of cases over the study period. In addition there were problems associated with cervical cerclage suture in 4 cases (14%); in 1 case the suture had been rejected, and another was cut to allow spontaneous abortion to be completed . They both underwent abdominal reinsertion. In the 2 remaining cases the suture had to be trimmed due to interference with intercourse. Discussion The primary outcome measure of any cancer treatment is undoubtedly survival but other factors such as associated morbidity and quality of life cannot be ignored. Patients therefore need to be aware of potential complications and longterm morbidities associated with any proposed procedure and in particular in gynaecological oncology the impact on fertility, sexual function and body image. In addition, before making a decision about the choice of surgery patients need to make a comparison with alternative surgical approaches and their associated benefits and morbidities. Based on the current literature, RT has comparable survival and recurrence rates to conventional RH and as a result it is becoming an established surgical approach for patients with an early stage disease who wish to preserve their fertility [4– 11]. However, despite the reported intraoperative or postoperative morbidities by various groups, the more long-term associated problems with RT and its comparison with conventional RH has not as yet been fully investigated. In our study we have shown that RT is associated with significantly less morbidity when compared with RH and the patients are discharged home much earlier. RT patients are less likely to have major blood loss and their requirement for regular analgesia is much less. Furthermore, the need for prolonged catheterisation due to bladder hypotonia happens rarely in RT. The main drawback for RT, however remains the longer mean operative time, which is partly due to the slow learning curve in early cases, as has been confirmed with further experience. These findings are in concordance with previous reported series from major North American and European centres (Table 4). The rate of accidental intraoperative injuries for RT in our study does not differ significantly from RH and are mainly vascular injuries or cystotomies that were all dealt by gynaecological oncologists. Such injuries again happened in earlier cases in RT group whilst establishing
F. Alexander-Sefre et al. / Gynecologic Oncology 101 (2006) 450 – 454 Table 4 Comparative intraoperative findings from 5 major series Centre
Toronto
Lyon
LA
Quebec
London
Author No OR time (min) Blood Loss (ml) Transfusion rate % Hospital stay (days)
Covens 81 180 300 9.4 1
Dargent 82 130 NR 6.5 7
Morrow 21 318 293 NR 3
Roy 44 240 200 1.4 3
Shepherd 29 260 400 12 6
the technique and only 1 case required conversion to laparotomy. Rarity of severe intraoperative complications has been also reported by other groups [6,10]. Examining the long-term reported morbidities, there appear to be problems that never occurred within the RT group and were specifically related to RH. These included, loss of sexual desire or sensation, stress incontinence and abdominal wound problems. Such complaints are most likely a direct result of radical pelvic surgery or an indirect psychological impact of this operation. The higher incidence of lymphodema and lymphocyst formation in RH group may also be due to more radical surgery and hence greater disruption of lymphatics. Similarly, there are problems specific to RT group, for instance, deep dysparaunia, excessive vaginal discharge and recurrent candidiasis. The excessive vaginal discharge and recurrent candidiasis are thought to be most likely due to presence of cervical cerclage suture acting as a niche for abnormal vaginal organisms, despite usage of monofilament nylon suture material and coverage by vaginal epithelium. Deep dyspareunia, in two patients was noted to be due to the cervical suture, requiring trimming of the protruding ends. One must bear in mind however, that the higher level of reported anxiety or depression may have a confounding effect in the complaint of deep dyspareunia. However, this is only an observation which requires verification in prospective studies with objective means of assessment and with a control group from general population. In two cases the cervical suture had fallen out necessitating laparotomy and abdominal reinsertion. The RT procedure as carried out in our unit involves complete excision of cervix and therefore no cervical tissue in which one could safely reinsert the suture vaginally. It was therefore deemed safer and more practical to insert this second suture around the isthmus from above using an abdominal approach under direct vision. In our centre, RH is limited to patients with early disease not exceeding 4 cm in maximum diameter. A standard surgical approach was followed which involved type I RH and systematic dissection of lymphatic tissue along common iliac, internal, external iliac vessels and obturator fossa. Similarly patients only qualified for RT if their longest tumour diameter did not exceed 2 cm and was 1 cm away from internal cervical os. They similarly underwent systematic node dissection like RH group. Because of this standard surgical approach and limited number of patients in each study group, further stratification based on variations in tumour size and the number of previous cone biopsies were avoided and only morbidity data was collected. However it would be of interest
453
to examine such minor variations between the patients in future large prospective studies. In addition future comparative studies with laparoscopic Radical Hysterectomy which is not yet an established technique in any centre in UK would be informative. Cervical stenosis was not an uncommon problem in our series and the 3 patients who presented with this had to undergo examination under anaesthesia and isthmic dilatation. In two of the three patients the cervical os was opened directly with use of dilators and in one case ultrasound guidance was needed. Surprisingly, such complaint has not been reported in a larger study by Plante et al. [10]. This may be due to slight variation in surgical technique which needs to be investigated further. Parasthesia over upper medial aspect of thigh and some degree of weakness in adduction of leg were noted in 14% of our patients. This is most likely due to manipulation and direct thermal or electrical injury by monopolar diathermy to cutaneous branches of obturator, ilioinguinal and genitofemoral nerves. Since the study, our practice has been modified by using harmonic scalpel and bipolar diathermy in order to reduce such complications. Most countries in which RT is practiced have formed a central register for their patients pooling follow up data and fertility outcomes. To our best knowledge, however none have so far included record of short- and long-term physical and psychological morbidities. Introduction of this additional data would be of great importance for patients in order to reach a well-informed decision and to decide whether the indications for RT should be expanded. In conclusion, RT is associated with less operative and short-term morbidity compared with conventional RH. There are specific complications associated with RT and there are also long-term morbidities that are more commonly observed in RH patients. References [1] Covens A, Rosen B, Murphy J, Laframboise S, DePetrillo AD, Lickrish G, et al. Changes in the demographics and perioperative care of stage IA(2)/IB(1) cervical cancer over the past 16 years. Gynecol Oncol 2001;81(2):133 – 7. [2] Sonoda Y, Abu-Rustum NR, Gemignani ML, Chi DS, Brown CL, Poynor EA, et al. A fertility-sparing alternative to radical hysterectomy: how many patients may be eligible? Gynecol Oncol 2004; 95(3):534 – 8. [3] Dargent D, Burn JL, Roy M, Remi I. Pregnancies following radical trachelectomy for invasive cervical cancer. Gynaecol Oncol 1994;52:105 [Abstract 14]. [4] Shepherd JH, Mould T, Oram DH. Radical trachelectomy in early stage carcinoma of the cervix: outcome as judged by recurrence and fertility rates. Bri J Obstet Gynaecol 2001;108(8):882 – 5. [5] 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(11): 2273 – 9. [6] 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(8):1877 – 82. [7] Schlaerth JB, Spirtos NM, Schlaerth AC. Radical trachelectomy and pelvic lymphadenectomy with uterine preservation in the treatment of cervical cancer. Am J Obstet Gynecol 2003;188(1):29 – 34.
454
F. Alexander-Sefre et al. / Gynecologic Oncology 101 (2006) 450 – 454
[8] Lee CL, Huang KG, Wang CJ, Yen CF, Lai CH. Laparoscopic radical trachelectomy for stage Ib1 cervical cancer. J Am Assoc Gynecol Laparosc 2003;10(1):111 – 5. [9] Burnett AF, Roman LD, O’Meara AT, Morrow CP. Radical vaginal trachelectomy and pelvic lymphadenectomy for preservation of fertility in early cervical carcinoma. Gynecol Oncol 2003;88(3):419 – 23. [10] Plante M, Renaud MC, Francois H, Roy M. Vaginal radical trachelectomy: an oncologically safe fertility-preserving surgery. An updated series of 72 cases and review of the literature. Gynecol Oncol 2004;94(3):614 – 23.
[11] Plante M, Roy M. New approaches in the surgical management of early stage cervical cancer. Curr Opin Obstet Gynecol 2001; 13(1):41 – 6. [12] Plante M, Renaud MC, Francois H, Roy M. Vaginal radical trachelectomy. In: Levine d, Barakat RR, editors. Atlas of Procedures in Gynaecological Oncology. London’ Martin Dunitz; 2003. p. 207 – 21. [13] Shepherd JH, Crawford RA, Oram DH. Radical trachelectomy: a way to preserve fertility in the treatment of early cervical cancer. Br J Obstet Gynaecol 1998;105(8):912 – 6.