- Email: [email protected]
Does a uterine manipulator affect cervical cancer pathology or identification of lymphovascular space involvement?
Gynecologic Oncology 127 (2012) 98–101
Contents lists available at SciVerse ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Does a uterine manipulator affect cervical cancer pathology or identification of lymphovascular space involvement?☆ Joseph A. Rakowski a,⁎, Tien Anh N. Tran b, Sarfraz Ahmad a, Jeffrey A. James a, Lorna A. Brudie a, Peter J. Pernicone b, Michael J. Radi b, Robert W. Holloway a a b
Florida Hospital Gynecologic Oncology, Florida Hospital Cancer Institute, Orlando, FL 32804, USA Florida Hospital Department of Pathology, Orlando, FL 32804, USA
H I G H L I G H T S ► Pathology examination of cases utilizing uterine manipulators revealed increased surface disruption and parametrial carryover, but no change in LVSI. ► Uterine manipulators can be used in robotic-assisted laparoscopic radical hysterectomies without fear of altering pathology or creating false LVSI. ► Risk recurrence scores and treatment recommendations will not be falsely increased in cases where a uterine manipulator is utilized.
a r t i c l e
i n f o
Article history: Received 4 June 2012 Accepted 4 July 2012 Available online 16 July 2012 Keywords: Cervical cancer Uterine manipulator Robotic radical hysterectomy Laparotomy Lymphovascular space involvement Pathology
a b s t r a c t Objective. Uterine manipulators are a useful adjunct for robotic-assisted radical hysterectomy (RARH), but some surgeons avoid their use for fear of altering pathology or interpretation of lymphovascular space involvement (LVSI). We retrospectively compared clinico-pathological data and tumor pathology from patients with cervical cancer operated by laparotomy vs. RARH. Methods. Charts from cervical cancer patients who underwent radical hysterectomy from January-1997 to June-2010 were reviewed for tumor histology, grade, FIGO stage, lymph node status, LVSI, depth of invasion, and tumor size. A ConMed V-Care® uterine manipulator was used in all robotic cases. H&E stained slides from 20 robotic and 24 open stage IB1 cases with LVSI reported in the original pathology were re-reviewed by a blinded pathologist for analysis of tissue artifacts and LVSI. Results. Two-hundred-thirty-six cases (185 open, 51 robotic) with stages IA2, IB1 and IB2 cervical cancer were reviewed. No significant differences in histology (squamous cell carcinoma, 65% vs. 51%; p=0.1), IB1 lesion size (≤2 cm, 62% vs. 61%, p>0.1), LVSI (34% vs. 39%, p>0.1), and depth of stromal invasion (p>0.1) was found between open and robotic groups. Histologic examination of all IB1 cervical carcinomas revealed a higher degree of surface disruption [45% (9/20) vs. 12.6% (3/24), p=0.038] and artifactual “parametrial carryover” [65% (13/20) vs. 29% (7/24), p=0.037] in robotic vs. open groups, respectively, but no significant differences in the rate of LVSI. Conclusion. RARH cases that utilized a uterine manipulator did not show any clinico-pathological differences in depth of invasion, LVSI, or parametrial involvement compared to open cases. © 2012 Elsevier Inc. All rights reserved.
Introduction Minimally invasive gynecologic surgery, whether robotic-assisted or traditional laparoscopic, is considered a safe and effective approach, resulting in less perioperative morbidity and decreased hospitalization compared to laparotomy procedures [1–4]. These minimally invasive procedures produce similar pathologic specimen and allow the surgeon ☆ Presented at the 42nd Annual Meeting on Women's Cancer® of the Society of Gynecologic Oncology (SGO), in March 2011, at Orlando, FL. ⁎ Corresponding author at: Florida Hospital Gynecologic Oncology, Florida Hospital Cancer Institute, 2501 N. Orange Ave., Suite 800, Orlando, FL 32804, USA. Fax: +1 407 303 2435. E-mail addresses: [email protected] (J.A. Rakowski), sarfraz.ahmad@flhosp.org (S. Ahmad). 0090-8258/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2012.07.094
to gain better visualization of the pelvic anatomy. Uterine manipulators (i.e., Humi®, Zumi™, Hulka, or ConMed VCare®) allow the surgical assistant to easily maneuver the uterus, which improves exposure, aids dissection of the parametria with counter-traction, displaces the ureters from the thermal “injury zone” of bipolar devices, and helps clearly delineate the cervical/vaginal junction during colpotomy. However, some surgeons avoid the use of a uterine manipulator due to the theoretical risk of altering the tumor morphology or creating false lymphovascular space invasion (LVSI) [5–7]. This artifact has been described in cases of endometrial carcinoma as “pseudo-vascular invasion”, although some question the legitimacy of this concept and its etiology [8,9]. Nevertheless, pseudo-vascular invasion has not been investigated or reported in cervical cancer specimens removed with the use of a uterine manipulator.
J.A. Rakowski et al. / Gynecologic Oncology 127 (2012) 98–101
99
were performed from January 1997 to July 2007 prior to the first robotic radical hysterectomy at our institution. All RARH cases were performed between July 2007 and June 2010 with the assistance of a ConMed VCare® uterine manipulator. Clinical data extrapolated from the database included: tumor histology, tumor grade, tumor size, depth of invasion, International Federation of Gynecology and Obstetrics (FIGO) stage, lymph node status, and LVSI status. Depth of stromal invasion was reported as the percentage of the cervical stroma involved by the tumor and the cervical tumor size was a direct measurement of the greatest dimension in centimeters. All 83 cases of stage IB1 cervical cancer with reported LVSI were selected for histologic re-review by one pathologist (TNT) who was blinded to the surgical technique. Hematoxylin and eosin (H&E) stained slides were available for all 20 RARH cases that were positive for LVSI and 25 of 63 (40%) laparotomy cases (only cases b 10 years old available) with LVSI. Second review of the stage IB1 cervical cancer cases examined for: tumor size, histology, tumor surface disruption, size and location of the vessels containing tumor cells, number of LVSI (greater than or less than 5 LVSI), intravascular debris (inflammatory cells admixed with amorphous material in vessels), and the presence of parametrial carryover (defined as clusters and nests of tumor cells in the parametrium without any associated stromal reaction). Tumor surface disruption was identified by 2 criteria: artifactual elongated clefts formation that extends from the surface into the deeper portion of the tumor, or fragmentation of the superficial tumor surface with formation of non-cohesive clusters of tumor cells separated from the stroma by empty spaces/clefts and crushed artifacts in the tumor cells. The vessels were divided into small (open vascular space without any muscular wall) [Fig. 1A], medium (relatively dilated vessels with a thin muscular wall) [Fig. 1B], and large (dilated blood vessels with thick muscular wall) [Fig. 1C]. The primary objective was to determine whether any tumor surface disruption or LVSI patterns existed in the robotic group that was not seen in the laparotomy cases, which might have affected the final pathology diagnosis. Data was expressed as percent and/or mean with standard deviation (SD) for continuous variables and rates in the case of discrete variables. A Student's t-test or Chi-squared test was used for continuous and discrete variables, respectively. A z-test was performed to determine the statistical significance where data was analyzed for percent differences between the groups. In all instances, p-valueb 0.05 was used to determine the statistical significance. Fig. 1. (A) Lymphovascular space invasion (LVSI) in small caliber vessels (H&E ×100). Note: no muscular wall in the vessel. (B) LVSI in medium sized vessels (H&E ×100). Note: thin muscular wall of the vessel. (C) LVSI in large vessels (H&E ×100). Note: thick muscular wall of the vessel.
Considering the surgical advantages of proper uterine manipulation during robotic-assisted radical hysterectomy (RARH) as well as the potential for altering the cervical tumor, we investigated the potential effects of a uterine manipulator on cervical pathology. The purpose of this study was to compare clinico-pathologic data from RARH and open (laparotomy) radical hysterectomy cases for differences in LVSI, depth of invasion (DOI), or parametrial involvement. Additionally, all available slides from stage IB1 cervical cancer cases with LVSI were accessioned for re-review by a blinded pathologist to determine if surface disruption or any unusual LVSI pattern could be attributed to the uterine manipulator. Materials and methods A retrospective database review of all radical hysterectomy cases (open and robotic) performed at a single institution from January 1997 to June 2010 was examined once institutional review board (IRB) approval was granted. The majority of laparotomy (open) cases
Results A total of 236 patients underwent radical hysterectomy procedures for cervical carcinoma [185 laparotomy and 51 robotic cases]. Patient demographics were not statistically different between the two groups including mean age (43.7 years robotic vs. 48.7 years open) and mean Table 1 Patient demographics and tumor characteristics.
Number of cases Body mass index (kg/m2) Age (years) FIGO stage IA2 IB1 IB2 Histology Squamous cell Adenocarcinoma Neuroendocrine
Abdominal radical hysterectomy
Robotic radical hysterectomy
P-value
185 27.26 48.7 (range 29–89)
48 26.86 44.1 (range 26–80)
N/A 0.899 >0.05
42 (22.7%) 130 (70.3%) 13 (7%)
5 (10.4%) 42 (87.5%) 1 (2.1%)
0.091 0.026 0.351
120 (64.9%) 63 (34%) 2 (1%)
25 (52%) 23 (48%) 0 (0%)
0.140 0.104 0.828
Abbreviations: N/A — not applicable; FIGO = International Federation of Gynecology and Obstetrics.
100
J.A. Rakowski et al. / Gynecologic Oncology 127 (2012) 98–101
Table 2 Tumor pathology for stage IB1 cervical cancer cases.
Number of patients Depth of invasion Inner 1/3rd Middle 1/3rd Outer 1/3rd FIGO stage IB1 lesion size ≤2 cm >2 cm LVSI Positive Positive lymph nodes Adjuvant therapy (Cisplatin+ RT) GOG intermediate-risk group
Abdominal radical hysterectomy
Robotic radical hysterectomy
P-value
130
42
N/A
50 (38.5%) 38 (29.2%) 42 (32.3%)
22 (52.3%) 12 (28.6%) 8 (19.0%)
0.163 0.904 0.146
81 (62%) 49 (38%)
25 (59.5%) 17 (40.4%)
0.915 0.924
63 21 30 21
20 (39%) 7 (16.6%) 11 (26.2%) 8 (19%)
0.686 0.858 0.838 0.842
(34%) (16.1%) (23.1%) (16.1%)
Abbreviations: N/A = not applicable; FIGO = International Federation of Gynecology and Obstetrics; LVSI = lymphovascular space invasion; RT = radiation therapy; GOG = Gynecologic Oncology Group.
body mass index (BMI) (26.9 kg/m 2 robotic vs. 27.3 kg/m2 open) [Table 1]. The number of patients in FIGO stages IA2 and IB2 were similar in both groups, but a significantly higher number of robotic cases were stage IB1 (70.3% open vs. 87.5% robotic; p = 0.026). Squamous cell carcinoma was the most common histology (64.9% open vs. 52% robotic), and more adenocarcinoma was found in the robotic series (49% robotic vs. 34% open) but the difference was not statistically significant (p= 0.104). One-hundred-thirty open and 42 robotic radical hysterectomy cases with stage IB1 cervical carcinoma were then compared based on risk factors for recurrence of disease including tumor size, depth of invasion, and LVSI status [Table 2]. Depth of stromal invasion was not statistically different between the open and robotic cases for inner, middle, and outer-third comparisons. Likewise, there were no differences in tumor size, parametrial tumor spread, or lymphovascular space invasion. The IB1 cases were examined for GOG intermediate-risk factors (LVSI, tumor size and cervical stromal invasion) and there was no difference between the open and robotic cases (p= 0.84). Finally, we were unable to show any significant difference between positive lymph node spread (p= 0.86) or adjuvant therapy (p= 0.84) in the open vs. robotic radical hysterectomy cases. Patients with stage IB1 cervical carcinoma were reviewed and LVSI was identified in 20 robotic and 63 open cases. H&E slides were available for all robotic (20) cases and 25 (40%) open radical hysterectomy cases. Histology re-review of stage IB1 cervical carcinoma patients with LVSI demonstrated a higher rate of surface disruption (45% vs. 12.6%, p = 0.038) and parametrial carryover (65% vs. 29%, p = 0.037) for robotic compared to open cases [Table 3]. There were however, no significant differences in all other main categories such as vascular invasion in the small, medium, and large-size vessels, or presence of intravascular inflammatory debris [Figs. 1–3].
Fig. 2. Parametrial carryover (H&E ×100). Note: cluster of malignant cells in fibro-connective tissue without an associated desmoplastic reaction.
visualization of key anatomic structures throughout the dissection. However, there have been concerns that uterine cancer pathology may be altered (increased LVSI) from uterine manipulators [5–7] and some investigators choose not to use a manipulator in cervical cancer cases [10–12]. In this study, we compared cases of cervical cancer from RARH and open radical hysterectomy to determine if there were any significant differences in the pathology reports between the two groups, and any differences in specific key pathologic findings when secondarily examined by a blinded pathologist. Stage IB1 cervical cancer specimen were selected for blinded review in order to evaluate lesions with a higher rate of LVSI and to examine for possible alterations in risk scoring of GOG intermediate-risk lesions. Our secondary review did show increased surface tumor disruption and parametrial carry-over, but did not indicate significant alterations in the tumor pathology of the robotic cases which utilized a uterine manipulator. Specifically, there were no identifiable alterations in the frequency or character of LVSI or parametrial invasion when comparing patients operated by the two methods. There was also no significant difference in the mean depth of cervical stromal invasion between the RARH and open radical hysterectomy cases in this study. Although surface disruption and parametrial contamination were recognized more frequently in robotic specimens, this did not affect the ability to determine depth of invasion, LVSI, grade, or lesion size. Consequently, use of a uterine manipulators did not appear to alter the GOG risk scores for disease recurrence [13,14], and therefore were also very unlikely to influence treatment recommendations. Ultimately, long-term follow-up analyses of recurrence patterns and adjuvant therapies offered will be
Discussion Most gynecologic surgeons appreciate the advantages of uterine manipulation during hysterectomy procedures because of the improved
Table 3 Robotic vs. open pathology review.
Robotic Open p-value
Parametrial carryover
Small vessel LVSI
Medium vessel LVSI
Large vessel LVSI
Debris
Surface
65% (13/20) 29% (7/24) 0.037
75% 75% 0.727
40% 29% 0.655
25% 16.7% 0.761
25% 25% 0.727
45% 12.5% 0.038
Abbreviation: LVSI = lymphovascular space invasion.
Fig. 3. Tumor surface disruption and cleft formation (H&E ×40). Note: non-cohesive clusters of tumor cells in the surface disruption.
J.A. Rakowski et al. / Gynecologic Oncology 127 (2012) 98–101
necessary to confirm our interpretation of the pathology, or equivalency of the procedures. Retrospective studies investigating the possible effects of uterine manipulators on endometrial tumor pathology have offered differing opinions for the suspected causes of increased LVSI in manipulated cases. Some investigators have theorized that the uterine manipulator was either solely or partly the cause of increased LVSI [5–7]. Two studies proposed that the increased LVSI was possibly due to disruption and direct dissemination of tumor into lymph-capillary channels by the uterine manipulator [5], or from increased intrauterine pressure created by the manipulator mechanically transporting tumor into the lymphovascular spaces [6]. Conversely, Kitahara et al. [7] speculated that uterine manipulators probably disrupt surface tumor leading to a situation where the grossing process could artificially disseminate tumor cells or fragments into the lymphovascular spaces. In contrast to these three studies, Folkins et al. [8] and Frimer et al. [9] were unable to identify any differences in LVSI between the manipulated and non-manipulated endometrial cancer specimen. Frimer et al. [9] did report a correlation between LVSI and lymph node metastasis, but could not find a correlation between uterine manipulators and LVSI or micrometastasis. There have been no publications to date analyzing the effects of uterine manipulators on cervical cancer pathology. The blinded histologic comparison of LVSI, surface disruption, intravascular debris and parametrial carryover in open and RARH specimens further supports our hypothesis that uterine manipulators do not affect cervical cancer pathology. There was a significant increase in surface disruption and parametrial carryover in robotic cases compared to open hysterectomies, which was probably due to a combination of surgical techniques and grossing artifacts. The uterine manipulator was probably responsible for the increased surface disruption which occurred during placement from cervical dilatation and the trauma with removal. Surface fragmentation of the tumor might result in a higher rate of artifactual displacement of tumor tissue into the parametrium by the cutting blade during the grossing process as suggested by Kitahara et al. [7]. The absence of an increased rate of lymph vascular invasion observed in robotic hysterectomies for cervical cancer in this study compared to reports of endometrial cancer cases deserves comment. First, the endocervical canal is small and confined compared to large and dilated endometrial cavities and cervical cancers are usually smaller and less polypoid than their endometrial counterparts. Kitahara et al. [7] only observed vascular pseudo-invasion in cases of polypoid endometrial carcinoma and hypothesized that the uterine manipulator may facilitate artifactual VSI (“vascular pseudoinvasion”). Second, the difference in the tissue composition of the cervical wall and myometrium could also account for the discrepancies between our study and those on endometrial cancer. The myometrium is composed of separate bundles of smooth muscle that allows more flexibility of the vascular lumen; however, the uterine cervix is comprised of compact dense connective tissue that renders the vessel wall more rigid with less room to collapse and dilate. The collapse of the lymphatics and vasculature in the uterus has been hypothesized to be a factor in the increased vascular pseudoinvasion in robotic hysterectomies for endometrial carcinoma [6]. Lastly, a closed pressure system created by the uterine manipulator balloon inside the uterine corpus which theoretically
101
could mechanically transport endometrial tissue into the vascular spaces through the effects of pressure is not possible in the endocervical canal in RARH for cervical carcinoma. This was thought to be the sole reason for increased LVSI in endometrial cancer cases reported by Logani et al. [6]. Thus, in contrast to the studies reporting possible altered pathology in endometrial cancer cases which utilized a uterine manipulator, we were unable to find an increased risk or vascular invasion of pseudoinvasion when using a uterine manipulator in cases of RARH for cervical carcinoma. Surgeons can use uterine manipulators without fear of significantly altering the pathologic interpretation of the cervical cancer specimens, falsely increasing the risk of recurrence pathology scores, or spuriously affecting treatment recommendations. Conflict of interest statement All co-authors declare that there are no conflicts of interest associated with this manuscript.
References [1] Gaia G, Holloway RW, Santoro L, Ahmad S, Di Silverio E, Spinillo A. Robotic-assisted hysterectomy for endometrial cancer compared with traditional laparoscopic and laparotomy approaches: a systematic review. Obstet Gynecol 2010;116:1422-31. [2] Reza M, Maeso S, Blasco JA, Andradas E. Meta-analysis of observational studies on the safety and effectiveness of robotic gynaecological surgery. Br J Surg 2010;97: 1772-83. [3] Lu D, Liu Z, Shi G, Zhou X. Robotic assisted surgery for gynaecological cancer. Cochrane Database Syst Rev 2012(1), doi:10.1002/14651858.CD008640.pub2 [Art. No.: CD008640]. [4] Ramirez PT, Adams S, Boggess JF, Burke WM, Frumovitz MM, Gardner GJ, et al. Robotic-assisted surgery in gynecologic oncology: a Society of Gynecologic Oncology Consensus Statement Developed by the Society of Gynecologic Oncology's Clinical Practice Robotics Task Force. Gynecol Oncol 2012;124:180-4. [5] Krizova A, Clarke BA, Bernardini MQ, James S, Kalloger SE, Boerner SL, et al. Histologic artifacts in abdominal, vaginal, laparoscopic, and robotic hysterectomy specimens: a blinded, retrospective review. Am J Surg Pathol 2011;35: 115-26. [6] Logani S, Herdman AV, Little JV, Moller KA. Vascular “Pseudo Invasion” in laparoscopic hysterectomy specimens: a diagnostic pitfall. Am J Surg Pathol 2008;32:560-5. [7] Kitahara S, Walsh C, Frumovitz M, Malpica A, Silva EG. Vascular pseudo-invasion in laparoscopic hysterectomy specimens for endometrial carcinoma: a grossing artifact? Am J Surg Pathol 2009;33:298-303. [8] Folkins AK, Nevadunsky NS, Saleemuddin A, Jarboe EA, Muto MG, Feltmate CM, et al. Evaluation of vascular space involvement in endometrial adenocarcinomas: laparoscopic vs. abdominal hysterectomies. Mod Pathol 2010;23:1073-9. [9] Frimer M, Khoury-Collado F, Murray MP, Barakat RR, Abu-Rustum NR. Micrometastasis of endometrial cancer to sentinel lymph nodes: is it an artifact of uterine manipulation? Gynecol Oncol 2010;119:496-9. [10] Shafer A, Boggess JF. Robotic-assisted endometrial cancer staging and radical hysterectomy with the da Vinci surgical system. Gynecol Oncol 2008;111(Suppl. 2): S18-23. [11] Estape R, Lambrou N, Diaz R, Estape E, Dunkin N, Rivera A. A case matched analysis of robotic radical hysterectomy with lymphadenectomy compared with laparoscopy and laparotomy. Gynecol Oncol 2009;113:357-61. [12] Magrina J, Kho R, Magtibay PM. Robotic radical hysterectomy: technical aspects. Gynecol Oncol 2009;113:28-31. [13] Delgado G, Bundy B, Zaino R, Sevin BU, Creasman WT, Major F. Prospective surgical-pathological study of disease-free interval in patients with stage IB squamous cell carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol 1990;38:352-7. [14] Sedlis A, Bundy BN, Rotman MZ, Lentz SS, Muderspach LI, Zaino RJ. A randomized trial of pelvic radiation therapy versus no further therapy in selected patients with stage IB carcinoma of the cervix after radical hysterectomy and pelvic lymphadenectomy: a Gynecologic Oncology Group study. Gynecol Oncol 1999;73:177-83.
Contents lists available at SciVerse ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Does a uterine manipulator affect cervical cancer pathology or identification of lymphovascular space involvement?☆ Joseph A. Rakowski a,⁎, Tien Anh N. Tran b, Sarfraz Ahmad a, Jeffrey A. James a, Lorna A. Brudie a, Peter J. Pernicone b, Michael J. Radi b, Robert W. Holloway a a b
Florida Hospital Gynecologic Oncology, Florida Hospital Cancer Institute, Orlando, FL 32804, USA Florida Hospital Department of Pathology, Orlando, FL 32804, USA
H I G H L I G H T S ► Pathology examination of cases utilizing uterine manipulators revealed increased surface disruption and parametrial carryover, but no change in LVSI. ► Uterine manipulators can be used in robotic-assisted laparoscopic radical hysterectomies without fear of altering pathology or creating false LVSI. ► Risk recurrence scores and treatment recommendations will not be falsely increased in cases where a uterine manipulator is utilized.
a r t i c l e
i n f o
Article history: Received 4 June 2012 Accepted 4 July 2012 Available online 16 July 2012 Keywords: Cervical cancer Uterine manipulator Robotic radical hysterectomy Laparotomy Lymphovascular space involvement Pathology
a b s t r a c t Objective. Uterine manipulators are a useful adjunct for robotic-assisted radical hysterectomy (RARH), but some surgeons avoid their use for fear of altering pathology or interpretation of lymphovascular space involvement (LVSI). We retrospectively compared clinico-pathological data and tumor pathology from patients with cervical cancer operated by laparotomy vs. RARH. Methods. Charts from cervical cancer patients who underwent radical hysterectomy from January-1997 to June-2010 were reviewed for tumor histology, grade, FIGO stage, lymph node status, LVSI, depth of invasion, and tumor size. A ConMed V-Care® uterine manipulator was used in all robotic cases. H&E stained slides from 20 robotic and 24 open stage IB1 cases with LVSI reported in the original pathology were re-reviewed by a blinded pathologist for analysis of tissue artifacts and LVSI. Results. Two-hundred-thirty-six cases (185 open, 51 robotic) with stages IA2, IB1 and IB2 cervical cancer were reviewed. No significant differences in histology (squamous cell carcinoma, 65% vs. 51%; p=0.1), IB1 lesion size (≤2 cm, 62% vs. 61%, p>0.1), LVSI (34% vs. 39%, p>0.1), and depth of stromal invasion (p>0.1) was found between open and robotic groups. Histologic examination of all IB1 cervical carcinomas revealed a higher degree of surface disruption [45% (9/20) vs. 12.6% (3/24), p=0.038] and artifactual “parametrial carryover” [65% (13/20) vs. 29% (7/24), p=0.037] in robotic vs. open groups, respectively, but no significant differences in the rate of LVSI. Conclusion. RARH cases that utilized a uterine manipulator did not show any clinico-pathological differences in depth of invasion, LVSI, or parametrial involvement compared to open cases. © 2012 Elsevier Inc. All rights reserved.
Introduction Minimally invasive gynecologic surgery, whether robotic-assisted or traditional laparoscopic, is considered a safe and effective approach, resulting in less perioperative morbidity and decreased hospitalization compared to laparotomy procedures [1–4]. These minimally invasive procedures produce similar pathologic specimen and allow the surgeon ☆ Presented at the 42nd Annual Meeting on Women's Cancer® of the Society of Gynecologic Oncology (SGO), in March 2011, at Orlando, FL. ⁎ Corresponding author at: Florida Hospital Gynecologic Oncology, Florida Hospital Cancer Institute, 2501 N. Orange Ave., Suite 800, Orlando, FL 32804, USA. Fax: +1 407 303 2435. E-mail addresses: [email protected] (J.A. Rakowski), sarfraz.ahmad@flhosp.org (S. Ahmad). 0090-8258/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2012.07.094
to gain better visualization of the pelvic anatomy. Uterine manipulators (i.e., Humi®, Zumi™, Hulka, or ConMed VCare®) allow the surgical assistant to easily maneuver the uterus, which improves exposure, aids dissection of the parametria with counter-traction, displaces the ureters from the thermal “injury zone” of bipolar devices, and helps clearly delineate the cervical/vaginal junction during colpotomy. However, some surgeons avoid the use of a uterine manipulator due to the theoretical risk of altering the tumor morphology or creating false lymphovascular space invasion (LVSI) [5–7]. This artifact has been described in cases of endometrial carcinoma as “pseudo-vascular invasion”, although some question the legitimacy of this concept and its etiology [8,9]. Nevertheless, pseudo-vascular invasion has not been investigated or reported in cervical cancer specimens removed with the use of a uterine manipulator.
J.A. Rakowski et al. / Gynecologic Oncology 127 (2012) 98–101
99
were performed from January 1997 to July 2007 prior to the first robotic radical hysterectomy at our institution. All RARH cases were performed between July 2007 and June 2010 with the assistance of a ConMed VCare® uterine manipulator. Clinical data extrapolated from the database included: tumor histology, tumor grade, tumor size, depth of invasion, International Federation of Gynecology and Obstetrics (FIGO) stage, lymph node status, and LVSI status. Depth of stromal invasion was reported as the percentage of the cervical stroma involved by the tumor and the cervical tumor size was a direct measurement of the greatest dimension in centimeters. All 83 cases of stage IB1 cervical cancer with reported LVSI were selected for histologic re-review by one pathologist (TNT) who was blinded to the surgical technique. Hematoxylin and eosin (H&E) stained slides were available for all 20 RARH cases that were positive for LVSI and 25 of 63 (40%) laparotomy cases (only cases b 10 years old available) with LVSI. Second review of the stage IB1 cervical cancer cases examined for: tumor size, histology, tumor surface disruption, size and location of the vessels containing tumor cells, number of LVSI (greater than or less than 5 LVSI), intravascular debris (inflammatory cells admixed with amorphous material in vessels), and the presence of parametrial carryover (defined as clusters and nests of tumor cells in the parametrium without any associated stromal reaction). Tumor surface disruption was identified by 2 criteria: artifactual elongated clefts formation that extends from the surface into the deeper portion of the tumor, or fragmentation of the superficial tumor surface with formation of non-cohesive clusters of tumor cells separated from the stroma by empty spaces/clefts and crushed artifacts in the tumor cells. The vessels were divided into small (open vascular space without any muscular wall) [Fig. 1A], medium (relatively dilated vessels with a thin muscular wall) [Fig. 1B], and large (dilated blood vessels with thick muscular wall) [Fig. 1C]. The primary objective was to determine whether any tumor surface disruption or LVSI patterns existed in the robotic group that was not seen in the laparotomy cases, which might have affected the final pathology diagnosis. Data was expressed as percent and/or mean with standard deviation (SD) for continuous variables and rates in the case of discrete variables. A Student's t-test or Chi-squared test was used for continuous and discrete variables, respectively. A z-test was performed to determine the statistical significance where data was analyzed for percent differences between the groups. In all instances, p-valueb 0.05 was used to determine the statistical significance. Fig. 1. (A) Lymphovascular space invasion (LVSI) in small caliber vessels (H&E ×100). Note: no muscular wall in the vessel. (B) LVSI in medium sized vessels (H&E ×100). Note: thin muscular wall of the vessel. (C) LVSI in large vessels (H&E ×100). Note: thick muscular wall of the vessel.
Considering the surgical advantages of proper uterine manipulation during robotic-assisted radical hysterectomy (RARH) as well as the potential for altering the cervical tumor, we investigated the potential effects of a uterine manipulator on cervical pathology. The purpose of this study was to compare clinico-pathologic data from RARH and open (laparotomy) radical hysterectomy cases for differences in LVSI, depth of invasion (DOI), or parametrial involvement. Additionally, all available slides from stage IB1 cervical cancer cases with LVSI were accessioned for re-review by a blinded pathologist to determine if surface disruption or any unusual LVSI pattern could be attributed to the uterine manipulator. Materials and methods A retrospective database review of all radical hysterectomy cases (open and robotic) performed at a single institution from January 1997 to June 2010 was examined once institutional review board (IRB) approval was granted. The majority of laparotomy (open) cases
Results A total of 236 patients underwent radical hysterectomy procedures for cervical carcinoma [185 laparotomy and 51 robotic cases]. Patient demographics were not statistically different between the two groups including mean age (43.7 years robotic vs. 48.7 years open) and mean Table 1 Patient demographics and tumor characteristics.
Number of cases Body mass index (kg/m2) Age (years) FIGO stage IA2 IB1 IB2 Histology Squamous cell Adenocarcinoma Neuroendocrine
Abdominal radical hysterectomy
Robotic radical hysterectomy
P-value
185 27.26 48.7 (range 29–89)
48 26.86 44.1 (range 26–80)
N/A 0.899 >0.05
42 (22.7%) 130 (70.3%) 13 (7%)
5 (10.4%) 42 (87.5%) 1 (2.1%)
0.091 0.026 0.351
120 (64.9%) 63 (34%) 2 (1%)
25 (52%) 23 (48%) 0 (0%)
0.140 0.104 0.828
Abbreviations: N/A — not applicable; FIGO = International Federation of Gynecology and Obstetrics.
100
J.A. Rakowski et al. / Gynecologic Oncology 127 (2012) 98–101
Table 2 Tumor pathology for stage IB1 cervical cancer cases.
Number of patients Depth of invasion Inner 1/3rd Middle 1/3rd Outer 1/3rd FIGO stage IB1 lesion size ≤2 cm >2 cm LVSI Positive Positive lymph nodes Adjuvant therapy (Cisplatin+ RT) GOG intermediate-risk group
Abdominal radical hysterectomy
Robotic radical hysterectomy
P-value
130
42
N/A
50 (38.5%) 38 (29.2%) 42 (32.3%)
22 (52.3%) 12 (28.6%) 8 (19.0%)
0.163 0.904 0.146
81 (62%) 49 (38%)
25 (59.5%) 17 (40.4%)
0.915 0.924
63 21 30 21
20 (39%) 7 (16.6%) 11 (26.2%) 8 (19%)
0.686 0.858 0.838 0.842
(34%) (16.1%) (23.1%) (16.1%)
Abbreviations: N/A = not applicable; FIGO = International Federation of Gynecology and Obstetrics; LVSI = lymphovascular space invasion; RT = radiation therapy; GOG = Gynecologic Oncology Group.
body mass index (BMI) (26.9 kg/m 2 robotic vs. 27.3 kg/m2 open) [Table 1]. The number of patients in FIGO stages IA2 and IB2 were similar in both groups, but a significantly higher number of robotic cases were stage IB1 (70.3% open vs. 87.5% robotic; p = 0.026). Squamous cell carcinoma was the most common histology (64.9% open vs. 52% robotic), and more adenocarcinoma was found in the robotic series (49% robotic vs. 34% open) but the difference was not statistically significant (p= 0.104). One-hundred-thirty open and 42 robotic radical hysterectomy cases with stage IB1 cervical carcinoma were then compared based on risk factors for recurrence of disease including tumor size, depth of invasion, and LVSI status [Table 2]. Depth of stromal invasion was not statistically different between the open and robotic cases for inner, middle, and outer-third comparisons. Likewise, there were no differences in tumor size, parametrial tumor spread, or lymphovascular space invasion. The IB1 cases were examined for GOG intermediate-risk factors (LVSI, tumor size and cervical stromal invasion) and there was no difference between the open and robotic cases (p= 0.84). Finally, we were unable to show any significant difference between positive lymph node spread (p= 0.86) or adjuvant therapy (p= 0.84) in the open vs. robotic radical hysterectomy cases. Patients with stage IB1 cervical carcinoma were reviewed and LVSI was identified in 20 robotic and 63 open cases. H&E slides were available for all robotic (20) cases and 25 (40%) open radical hysterectomy cases. Histology re-review of stage IB1 cervical carcinoma patients with LVSI demonstrated a higher rate of surface disruption (45% vs. 12.6%, p = 0.038) and parametrial carryover (65% vs. 29%, p = 0.037) for robotic compared to open cases [Table 3]. There were however, no significant differences in all other main categories such as vascular invasion in the small, medium, and large-size vessels, or presence of intravascular inflammatory debris [Figs. 1–3].
Fig. 2. Parametrial carryover (H&E ×100). Note: cluster of malignant cells in fibro-connective tissue without an associated desmoplastic reaction.
visualization of key anatomic structures throughout the dissection. However, there have been concerns that uterine cancer pathology may be altered (increased LVSI) from uterine manipulators [5–7] and some investigators choose not to use a manipulator in cervical cancer cases [10–12]. In this study, we compared cases of cervical cancer from RARH and open radical hysterectomy to determine if there were any significant differences in the pathology reports between the two groups, and any differences in specific key pathologic findings when secondarily examined by a blinded pathologist. Stage IB1 cervical cancer specimen were selected for blinded review in order to evaluate lesions with a higher rate of LVSI and to examine for possible alterations in risk scoring of GOG intermediate-risk lesions. Our secondary review did show increased surface tumor disruption and parametrial carry-over, but did not indicate significant alterations in the tumor pathology of the robotic cases which utilized a uterine manipulator. Specifically, there were no identifiable alterations in the frequency or character of LVSI or parametrial invasion when comparing patients operated by the two methods. There was also no significant difference in the mean depth of cervical stromal invasion between the RARH and open radical hysterectomy cases in this study. Although surface disruption and parametrial contamination were recognized more frequently in robotic specimens, this did not affect the ability to determine depth of invasion, LVSI, grade, or lesion size. Consequently, use of a uterine manipulators did not appear to alter the GOG risk scores for disease recurrence [13,14], and therefore were also very unlikely to influence treatment recommendations. Ultimately, long-term follow-up analyses of recurrence patterns and adjuvant therapies offered will be
Discussion Most gynecologic surgeons appreciate the advantages of uterine manipulation during hysterectomy procedures because of the improved
Table 3 Robotic vs. open pathology review.
Robotic Open p-value
Parametrial carryover
Small vessel LVSI
Medium vessel LVSI
Large vessel LVSI
Debris
Surface
65% (13/20) 29% (7/24) 0.037
75% 75% 0.727
40% 29% 0.655
25% 16.7% 0.761
25% 25% 0.727
45% 12.5% 0.038
Abbreviation: LVSI = lymphovascular space invasion.
Fig. 3. Tumor surface disruption and cleft formation (H&E ×40). Note: non-cohesive clusters of tumor cells in the surface disruption.
J.A. Rakowski et al. / Gynecologic Oncology 127 (2012) 98–101
necessary to confirm our interpretation of the pathology, or equivalency of the procedures. Retrospective studies investigating the possible effects of uterine manipulators on endometrial tumor pathology have offered differing opinions for the suspected causes of increased LVSI in manipulated cases. Some investigators have theorized that the uterine manipulator was either solely or partly the cause of increased LVSI [5–7]. Two studies proposed that the increased LVSI was possibly due to disruption and direct dissemination of tumor into lymph-capillary channels by the uterine manipulator [5], or from increased intrauterine pressure created by the manipulator mechanically transporting tumor into the lymphovascular spaces [6]. Conversely, Kitahara et al. [7] speculated that uterine manipulators probably disrupt surface tumor leading to a situation where the grossing process could artificially disseminate tumor cells or fragments into the lymphovascular spaces. In contrast to these three studies, Folkins et al. [8] and Frimer et al. [9] were unable to identify any differences in LVSI between the manipulated and non-manipulated endometrial cancer specimen. Frimer et al. [9] did report a correlation between LVSI and lymph node metastasis, but could not find a correlation between uterine manipulators and LVSI or micrometastasis. There have been no publications to date analyzing the effects of uterine manipulators on cervical cancer pathology. The blinded histologic comparison of LVSI, surface disruption, intravascular debris and parametrial carryover in open and RARH specimens further supports our hypothesis that uterine manipulators do not affect cervical cancer pathology. There was a significant increase in surface disruption and parametrial carryover in robotic cases compared to open hysterectomies, which was probably due to a combination of surgical techniques and grossing artifacts. The uterine manipulator was probably responsible for the increased surface disruption which occurred during placement from cervical dilatation and the trauma with removal. Surface fragmentation of the tumor might result in a higher rate of artifactual displacement of tumor tissue into the parametrium by the cutting blade during the grossing process as suggested by Kitahara et al. [7]. The absence of an increased rate of lymph vascular invasion observed in robotic hysterectomies for cervical cancer in this study compared to reports of endometrial cancer cases deserves comment. First, the endocervical canal is small and confined compared to large and dilated endometrial cavities and cervical cancers are usually smaller and less polypoid than their endometrial counterparts. Kitahara et al. [7] only observed vascular pseudo-invasion in cases of polypoid endometrial carcinoma and hypothesized that the uterine manipulator may facilitate artifactual VSI (“vascular pseudoinvasion”). Second, the difference in the tissue composition of the cervical wall and myometrium could also account for the discrepancies between our study and those on endometrial cancer. The myometrium is composed of separate bundles of smooth muscle that allows more flexibility of the vascular lumen; however, the uterine cervix is comprised of compact dense connective tissue that renders the vessel wall more rigid with less room to collapse and dilate. The collapse of the lymphatics and vasculature in the uterus has been hypothesized to be a factor in the increased vascular pseudoinvasion in robotic hysterectomies for endometrial carcinoma [6]. Lastly, a closed pressure system created by the uterine manipulator balloon inside the uterine corpus which theoretically
101
could mechanically transport endometrial tissue into the vascular spaces through the effects of pressure is not possible in the endocervical canal in RARH for cervical carcinoma. This was thought to be the sole reason for increased LVSI in endometrial cancer cases reported by Logani et al. [6]. Thus, in contrast to the studies reporting possible altered pathology in endometrial cancer cases which utilized a uterine manipulator, we were unable to find an increased risk or vascular invasion of pseudoinvasion when using a uterine manipulator in cases of RARH for cervical carcinoma. Surgeons can use uterine manipulators without fear of significantly altering the pathologic interpretation of the cervical cancer specimens, falsely increasing the risk of recurrence pathology scores, or spuriously affecting treatment recommendations. Conflict of interest statement All co-authors declare that there are no conflicts of interest associated with this manuscript.
References [1] Gaia G, Holloway RW, Santoro L, Ahmad S, Di Silverio E, Spinillo A. Robotic-assisted hysterectomy for endometrial cancer compared with traditional laparoscopic and laparotomy approaches: a systematic review. Obstet Gynecol 2010;116:1422-31. [2] Reza M, Maeso S, Blasco JA, Andradas E. Meta-analysis of observational studies on the safety and effectiveness of robotic gynaecological surgery. Br J Surg 2010;97: 1772-83. [3] Lu D, Liu Z, Shi G, Zhou X. Robotic assisted surgery for gynaecological cancer. Cochrane Database Syst Rev 2012(1), doi:10.1002/14651858.CD008640.pub2 [Art. No.: CD008640]. [4] Ramirez PT, Adams S, Boggess JF, Burke WM, Frumovitz MM, Gardner GJ, et al. Robotic-assisted surgery in gynecologic oncology: a Society of Gynecologic Oncology Consensus Statement Developed by the Society of Gynecologic Oncology's Clinical Practice Robotics Task Force. Gynecol Oncol 2012;124:180-4. [5] Krizova A, Clarke BA, Bernardini MQ, James S, Kalloger SE, Boerner SL, et al. Histologic artifacts in abdominal, vaginal, laparoscopic, and robotic hysterectomy specimens: a blinded, retrospective review. Am J Surg Pathol 2011;35: 115-26. [6] Logani S, Herdman AV, Little JV, Moller KA. Vascular “Pseudo Invasion” in laparoscopic hysterectomy specimens: a diagnostic pitfall. Am J Surg Pathol 2008;32:560-5. [7] Kitahara S, Walsh C, Frumovitz M, Malpica A, Silva EG. Vascular pseudo-invasion in laparoscopic hysterectomy specimens for endometrial carcinoma: a grossing artifact? Am J Surg Pathol 2009;33:298-303. [8] Folkins AK, Nevadunsky NS, Saleemuddin A, Jarboe EA, Muto MG, Feltmate CM, et al. Evaluation of vascular space involvement in endometrial adenocarcinomas: laparoscopic vs. abdominal hysterectomies. Mod Pathol 2010;23:1073-9. [9] Frimer M, Khoury-Collado F, Murray MP, Barakat RR, Abu-Rustum NR. Micrometastasis of endometrial cancer to sentinel lymph nodes: is it an artifact of uterine manipulation? Gynecol Oncol 2010;119:496-9. [10] Shafer A, Boggess JF. Robotic-assisted endometrial cancer staging and radical hysterectomy with the da Vinci surgical system. Gynecol Oncol 2008;111(Suppl. 2): S18-23. [11] Estape R, Lambrou N, Diaz R, Estape E, Dunkin N, Rivera A. A case matched analysis of robotic radical hysterectomy with lymphadenectomy compared with laparoscopy and laparotomy. Gynecol Oncol 2009;113:357-61. [12] Magrina J, Kho R, Magtibay PM. Robotic radical hysterectomy: technical aspects. Gynecol Oncol 2009;113:28-31. [13] Delgado G, Bundy B, Zaino R, Sevin BU, Creasman WT, Major F. Prospective surgical-pathological study of disease-free interval in patients with stage IB squamous cell carcinoma of the cervix: a Gynecologic Oncology Group study. Gynecol Oncol 1990;38:352-7. [14] Sedlis A, Bundy BN, Rotman MZ, Lentz SS, Muderspach LI, Zaino RJ. A randomized trial of pelvic radiation therapy versus no further therapy in selected patients with stage IB carcinoma of the cervix after radical hysterectomy and pelvic lymphadenectomy: a Gynecologic Oncology Group study. Gynecol Oncol 1999;73:177-83.