- Email: [email protected]
The role of preoperative ultrasound evaluation of inguinal lymph nodes in patients with vulvar malignancy
Gynecologic Oncology 131 (2013) 113–117
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
The role of preoperative ultrasound evaluation of inguinal lymph nodes in patients with vulvar malignancy Nikolaus de Gregorio ⁎,1, Florian Ebner 1, Lukas Schwentner, Thomas W.P. Friedl, Miriam Deniz, Krisztián Látó, Rolf Kreienberg, Wolfgang Janni, Dominic Varga Universitätsfrauenklinik Ulm, Prittwitzstrasse 43, 89075fl, Germany
H I G H L I G H T S • We looked how well ultrasonography could identify positive nodes in vulvar cancer patients. • Despite good results, this method is not good enough to spare the patient a surgical staging.
a r t i c l e
i n f o
Article history: Received 13 June 2013 Accepted 24 July 2013 Available online 7 August 2013 Keywords: Vulvar cancer Inguinal lymph nodes Sentinel lymph node biopsy Inguinal ultrasound
a b s t r a c t Objectives. Inguinal lymphadenectomy in vulvar malignancies is associated with significant morbidity, especially in patients over 70 years old. Under certain conditions, surgical guidelines recommend biopsy and evaluation of the sentinel node in early vulvar cancer. The purpose of our study is to evaluate ultrasonography as a predictor of inguinal lymph node involvement. Methods. A retrospective study was performed with 60 patients who had vulvar malignancies (92% of which were squamous cell carcinomas) and who were treated at our hospital between 2002 and 2012. The patients ranged in age from 35 to 89 years, with a median age of 76 years. In total, 118 groin scans were retrospectively evaluated for sonographic evidence of lymph node involvement (i.e., absence of fatty hilum, irregular shape, cortical region diameter and vascularization pattern). The results were then compared with histopathologically confirmed lymph node status. Results. Histopathologically confirmed lymph node status was available for 107 of the inguinal nodes examined by ultrasound, and lymph node metastases were found in 38 (35.5%) cases. The presence or absence of inguinal lymph node metastases was correctly identified by sonography in 92 (86.0%) of the scanned areas. Sensitivity was 76.3%, specificity was 91.3%, and positive and negative predictive values were 82.9% and 87.5%, respectively. Conclusions. Ultrasonography of the inguinal lymph nodes showed a relatively high sensitivity and specificity for predicting inguinal tumor metastases. However, our results indicate that surgical lymph node staging is still needed to precisely determine inguinal lymph node status in vulvar cancer, especially because a missed lymph node-metastasis is often fatal. © 2013 Elsevier Inc. All rights reserved.
Introduction In treating vulvar cancer, it is currently unclear whether sentinel node biopsy is a sufficient substitute to prevent the morbidity linked with complete removal of the inguinal nodes. Standard treatment involves wide radical excision of the primary tumor and inguinal lymphadenectomy. Even with contemporary surgical techniques, half of treated patients suffer from post-surgical complications, such as ⁎ Corresponding author at: Universitätsfrauenklinik Ulm, Prittwitzstrasse 43, 89075 Ulm, Germany. Fax: +49 73150058502. E-mail address: [email protected] (N. de Gregorio). 1 Both authors contributed equally. 0090-8258/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ygyno.2013.07.103
impaired wound healing or acute infections [1]. These complications are predominantly short term; however, lymphedema caused by lymphadenectomy can be a lifelong issue. Nevertheless, because nodal involvement is negatively correlated with survival, lymph node status remains the most important predictor of overall prognosis [2]. Fiftythree percent of all recurrences develop in the groin, with a 5-yearsurvival rate of only 27%; in the case of a locoregional relapse, the 5-year-survival rate is 60% [3]. While little progress has been made in the treatment of lymphedema, the modern sentinel lymph node biopsy (SLNB) procedure is intended to prevent lymphedema by reducing the number of patients subjected to a full lymphadenectomy. The SLNB procedure was proposed by Morton et al. [4] as an alternative to observation or elective regional lymphadenectomy in women
114
N. de Gregorio et al. / Gynecologic Oncology 131 (2013) 113–117
with early cutaneous melanoma. This technique provides the opportunity for early intervention in women with SLN metastases and spares women without SLN metastases the morbidity of unnecessary regional lymphadenectomy. As with melanoma and breast cancer, vulvar cancer is in an excellent anatomic position for the SLN procedure. The tumor is easily reached for tracer injection, and the lymph drainage is limited to either groin. The Gynecologic Oncology Group (GOG) has studied various surgical and radiotherapy techniques for women with vulvar cancer. Consequently, the SLN procedure is only recommended for women with clinically negative lymph nodes [5]. As in breast cancer, palpation is one method of determining clinical lymph node status. Another method is ultrasonography. The aim of this study is to evaluate whether inguinal ultrasonography is a useful diagnostic tool to predict lymph node involvement in vulvar cancer patients. Accordingly, we retrospectively assessed ultrasound images obtained before surgery and compared the results to the corresponding histopathologically confirmed lymph node status.
distributed data were conducted using the Mann–Whitney-U test, while categorical data were analyzed using the Chi-square test or Fisher's exact test. When investigating the predictive factors for lymph node involvement, patients were excluded from the analysis if they had histopathologically confirmed negative lymph node status for one inguinal area but no histopathologically confirmed lymph node status for the other inguinal area, because we could not be certain about the lymph node status of the other side. To evaluate ultrasonography as a predictor of inguinal lymph node involvement, the scan of each inguinal area was treated as an independent event. All statistical tests reported here were two-sided, and a test result was regarded as statistically significant if p b 0.05. All statistical analyses were performed using IBM Statistical Package for the Social Sciences (SPSS) version 19.
Methods
Patient characteristics for the 60 patients are shown in Table 1. The majority of patients had squamous cell carcinoma, the most common tumor size was pT1b, and histological grade was mostly grade 2. Invasion depth (data available for 46 patients) ranged from 1 to 22 mm with a median of 4 mm. Sentinel node biopsy revealed positive nodes in 22 (25.6%) of the 86 cases. Histopathologically confirmed lymph node metastases were found in 38 (35.5%) of the 107 inguinal areas studied. A median of 3 nodes were dissected if the nodes were negative (range 1–19, n = 69), and a median of 7 nodes were dissected if the nodes were histopathologically positive (range 1–15, n = 38); the difference was statistically significant (Mann– Whitney-U-test, z = 4.48, p b 0.001). No significant difference was found in the number of nodes that were dissected between the left and right groins (Mann–Whitney-U-test, z = 1.21, p = 0.23).
In total, 60 patients with early vulvar malignancy treated from January 2002 to December 2012 at the Department of Gynecology and Obstetrics in Ulm were included in this study. Patients with tumor stage FIGO 1a, which is a negative predictor for nodal involvement, were excluded [6,7]. However, three patients with histologically proven pT1a were included, as in preoperative clinical examination a higher stage was assumed. Patients who presented with suspected vulvar malignancy were referred to the ultrasound department for a presurgical inguinal scan. Neither the patients nor the health care providers had pay for the additional ultrasound, which was offered on a voluntary basis to all patients studied. Ultrasonography was performed using GE 730 Expert and Voluson E8 ultrasound devices (GE Healthcare, Solingen, Germany). Both devices are equipped with linear probes at 12 MHz. Greyscale and color Doppler imaging were performed with the linear probes. Images were stored in the ViewPoint Software System (GE Healthcare, Solingen, Germany) and reviewed by three investigators according to the recommendations for axillary lymph nodes (i.e., the absence of fatty hilum, irregular shape, cortical region diameter ≥4 mm and peripheral vascularization) (Fig. 1 and Fig. 2) [8,9]. In the few cases in which the initial assessment of the three investigators differed, a consensus was reached after visually cross checking the images. Overall, ultrasound scans were available for 118 inguinal areas; for two of the 60 patients, ultrasonography was only performed on the right groin. Surgery was performed by 4 experienced surgeons. Following national guidelines, technetium-99 m was applied one day prior to the procedure, and blue dye was given intraoperatively to tag the sentinel node. If the tumor was unilaterally localized, the procedure was only performed on the ipsilateral groin. All stained and/or radionuclidemarked nodes were identified visually and with gamma probe, surgically removed and then subjected to a frozen section procedure. If a neoplasia was detected, a complete inguinofemoral lymphadenectomy was performed. In 39 patients, a successful sentinel node biopsy was performed for both inguinal areas, and in 8 patients, sentinel node biopsy was performed for one side only, resulting in a total of 86 sentinel node biopsies. Histopathologically confirmed lymph node status was obtained for both inguinal areas in 47 patients, while for the remaining 13 patients, histopathology was available only for one of the two inguinal areas (due to the tumor being unilaterally localized as described above). Thus, histopathologically confirmed lymph node status was available for 107 of the 118 scanned groins. In 3 cases (4.8%), lymph node metastases that were later confirmed by histopathological analysis had been missed on assessment of the fresh frozen sections. Statistical analyses were performed using the appropriate nonparametric tests. Comparisons regarding continuous non-normally
Results Patient characteristics
Predictive factors for lymph node involvement No significant difference in age was found between patients with and without positive nodes (Mann–Whitney U-test, z = 0.5, p = 0.62). However, the tumors of patients with lymph node metastases had a significantly higher invasion depth (median 10 mm, range 2–22 mm) than the tumors of patients without lymph node metastases (median 3 mm,
Table 1 Patient characteristics (n = 60). Age in years
Body Mass Index (n = 59)
Invasion depth in mm (n = 46)
Histological type
pT
pN
Histological grading
Mean Median Range Mean Median Range Mean Median Range Squamous cell carcinomas Melanomas Bowen-carcinoma Basalioma pT1a pT1b pT2 pT3 n/a pN0 pN1 pN2 pN3 G1 G2 G3 n/a
70.1 76 35–89 26.7 25.4 19.0–48.9 6.1 4.0 1–22 55 (91.7%) 3 (5.0%) 1 (1.7%) 1 (1.7%) 3 (5.0%) 33 (55.0%) 17 (28.3%) 5 (8.3%) 2 (3.3%) 31 (51.7%) 16 (26.7%) 8 (13.3%) 5 (8.3%) 10 (16.7%) 42 (70.0%) 4 (6.7%) 4 (6.7%)
N. de Gregorio et al. / Gynecologic Oncology 131 (2013) 113–117
115
Fig. 1. Physiological appearance of a groin node: Ovular with a fatty hilum, marked cortexmedullary border and narrow cortex. Fig. 3. Tumor grade is significantly related to the probability of positive nodes.
range 1–20 mm; Mann–Whitney U-test, z = 2.85, p = 0.004). In addition, patients with lymph node metastases had a significantly higher BMI compared to patients without lymph node metastases (Mann– Whitney U-test, z = 1.99, p = 0.046). Patients with higher tumor stages did not have a higher probability of positive nodes compared to patients with lower tumor stages (Chi-square test, χ2 = 3.2, p = 0.36). In contrast, tumor grade was significantly related to the probability of positive nodes (Chi-square test, χ2 = 11.19, p = 0.004). While no positive nodes were found in patients with grade I tumor, 50% and 56% of patients with grade II and III tumors, respectively, had lymph node metastases (Fig. 3).
Predictive value of ultrasound Ultrasonography correctly predicted the presence or absence of inguinal lymph node metastases in 92 (86%) of 107 cases (Fig. 4). In 6 cases, ultrasonography yielded a false positive diagnosis; in 9 cases, false negative results were found. The sensitivity and specificity were 76.3% and 91.3%, respectively; the positive predictive value was 82.9%, and the negative predictive value was 87.5% (see Table 2). In seven of the 9 false negative results, only one histopathologically confirmed
Fig. 2. Suspect groin node. Irregular shape without a fatty hilum, an indistinct cortexmedullary border and broadened cortex.
positive node was found, indicating that the sensitivity of ultrasound might be related to the number of positive nodes.
Subgroup analysis We further investigated whether the diagnostic potential of ultrasonography of inguinal areas differed among various patient subgroups. For patients with a tumor invasion depth of less than 5 mm, there were no false positive diagnoses and specificity (100%) was significantly higher than for patients with a tumor invasion depth of 5 mm or more (specificity 82.4%; Fisher's exact test, p = 0.026). However, the clinically important rate of false negative results was almost the same for patients with tumor invasion depth below or above 5 mm (22.2% and 23.8%, respectively). Consequently, no significant difference was found regarding the sensitivity of ultrasonography between these two groups (Fisher's exact test, p = 1.0). For patients younger than 76 years, specificity (100%) was significantly higher than for patients 76 years or older (83.3%; Fisher's exact test, p = 0.025). By contrast, sensitivity was higher but not significantly
Fig. 4. Sonography correctly predicts positive nodes.
116
N. de Gregorio et al. / Gynecologic Oncology 131 (2013) 113–117
Table 2 Binary classification table. Lymph node metastases
Ultrasound Overall
Nonsuspect Suspect
Overall
No
Yes
63 6 69
9 29 38
72 35 107
different for older patients (86.4%) compared to younger ones (62.5%; Fisher's exact test, p = 0.13). For patients with a BMI above 30, sensitivity was numerically lower as compared to patients with a BMI below 30 (66.7% vs 78.6%); however, the difference was not statistically significant (n = 22, Fisher's exact test, p = 0.66). Specificity was very similar for patients with a BMI above or below 30 (92.3% and 91.1%, respectively; n = 84, Fisher's exact test, p = 1.0). Finally, neither specificity nor sensitivity was significantly different between patients treated from 2002 to 2008 and patients treated from 2009 to 2012 (Fisher's exact test, both p N 0.4). Discussion This study demonstrates that groin ultrasound scans, although more reliable than other non-invasive diagnostic tools in predicting nodal status in vulvar malignancies, lack the high sensitivity and specificity that are necessary to avoid surgery, especially because missed positive nodes have a strong impact on prognosis and treatment. The relatively large number of patients examined in this rare tumor entity (to our knowledge, the largest sample size published to date) and the high percentage of nodal involvement observed in our patient sample strengthen our findings. The rate of positive lymph nodes found in our study is higher than in most other series reported [10]. This potential selection bias might originate from the fact that in contrast to other trials, patients with a clinically rated tumor stage T1a were not included in this study, as our national treatment guidelines do not require a surgical groin staging for these patients. This patient selection might have led to a higher rate of node positives as compared to other published data. Another potential selection bias might be caused by the referral practice as our hospital is a regional cancer referral center, with a high number of severe cases transferred from other hospitals, resulting in an increased rate of node positive patients. It is difficult to exactly predict how a higher proportion of nodenegative patients would have influenced our results. Generally speaking, sensitivity and specificity are not expected to change due to an increased proportion of node-negative patient, while the positive predictive value would be expected to decrease and the negative predictive value would be expected to increase. That is, given that both sensitivity and specificity should not be affected by changes in the proportion of node-negative patients in the study population, our main results and conclusions would remain the same even if the proportion of nodenegative patients in our sample would be higher. In our study, we found no significant association between tumor stage and the probability of having positive lymph nodes; however, there was a non-significant tendency for higher tumor stages to be associated with an increased probability of having positive lymph nodes. One possible reason for the lack of a significant relationship might be that our sample included only three patients with a pT1a tumor, which is known to be a negative predictor for lymph node involvement. In addition, our sample included only five pT3 and no pT4 tumors, and this low number of large tumors might be another reason why the test for an association between tumor stage and nodal status did not reach statistical significance. Because up to 70% of patients have groin or lower-limb problems after radical groin surgery, the preoperative identification of nodal
positive disease is a major issue. While many studies focus on proper identification of the sentinel node [11,12], less effort has been exerted in trying to avoid the procedure completely. The difference in the number of dissected nodes between radical groin dissection and the sentinel node technique tends to be lower than in other cancers, for example, breast cancer [2,13,14], although current data suggest that axillary dissection performed after a positive sentinel node does not have a positive impact on prognosis. At the same time, locoregional recurrence of breast cancer in the axilla is still considered to be curable, while a groin relapse of a vulvar malignancy is often considered to be incurable due to limited therapeutic options. Mäkelä et al. [15] found that ultrasonography is more sensitive than preoperative or intraoperative palpation (82% vs 9% vs 55%). High sensitivity (92%) for groin ultrasound alone has been described previously; the addition of fine-needle aspiration cytology (FNAC) improves the results slightly [16]. Our study showed similar sensitivity and specificity when compared to earlier publications of ultrasound in combination with FNC [17]. This outcome might be due to the improvement of ultrasound technology. In addition, our sonographers routinely perform axillary lymph node ultrasounds and are experts in lymph node evaluation. Abang et al. [18] published sonographic criteria showing a sensitivity of 89% for determining lymph node status. These criteria use the short axis diameter and long axis/short axis ratio to identify positive nodes. However, the results of this study should be interpreted carefully because of the small sample size of 20 patients and the involvement of only a small number of lymph nodes, which resulted in a specificity of only 38%. Several other imaging techniques of the groin have been published during the last several years. In comparison to ultrasound, slightly lower rates of positive node detection have been reported with contrast enhanced magnetic resonance imaging (CE-MRI) [19]. Even lower sensitivity (52%) and positive predictive values (46%) have been reported for conventional MRI, leading the authors to conclude that there is no role for MRI in the diagnosis of groin nodes [20]. Similar data were published for preoperative CT-scans [21], where the sensitivity and positive predictive value were both 58%. Few reports are available regarding the use of positron emission tomography (PET) to detect groin metastases. Cohn et al. [22] published a study with 15 patients, but concluded PET to be relatively insensitive because a negative PET did not reliably indicate a pathologically negative groin. Inevitably, every retrospective study has limitations. A selection bias due to unknown factors cannot be ruled out. One limitation is the relatively small number of patients due to the low incidence compared to other cancers, such as breast cancer. The low incidence makes it difficult to design a prospective or randomized trial. Another limitation is that the final histological diagnosis is often not known at the time of the ultrasound; therefore, not only the squamous cell carcinomas but also a small amount of other pathologies are included. A possible confounding factor could be the improvement in ultrasound and surgical techniques during the 10-year data collection period, which might have resulted in inadequate imaging and thus worse diagnostic potential of ultrasonography in patients who were treated earlier in the study. However, the key diagnostic values did not differ significantly between patients treated from 2002 to 2008 and patients treated from 2009 to 2012 (see Results). Regarding surgical techniques, no major changes were identified when reviewing the patients' operative reports. As far as we know, no data on how BMI influences groin ultrasound have been previously published. However, there is data concerning axillary ultrasound suggesting that BMI is not associated with false negative results [23]. Another potential bias might be, that the patients' body weight was self reported, and therefore most likely rather underestimated. Overall, ultrasound seems to be the most efficacious of these imaging techniques because of its high sensitivity, low overall costs and the absence of exposure to radiation and/or contrast agents. Preoperative groin ultrasound is now a routine procedure in patients with vulvar
N. de Gregorio et al. / Gynecologic Oncology 131 (2013) 113–117
malignancy in our hospital. A sonographically suspicious node would result in the patient either having a preoperative needle biopsy or the suspect node would be identified with sonographically guided wire tag and excised in addition to the sentinel node. Promising future developments might include the contrast enhanced ultrasound (CEUS) technique, where microbubble contrast is used to visualize lymphatic channels and sentinel nodes. This technique has already been successfully demonstrated in breast cancer patients [24,25]. Conclusions Ultrasonography of the inguinal lymph nodes showed a relatively high sensitivity and specificity for predicting inguinal tumor metastases and is also a cost effective, easy-to-learn and easy-to-perform method that does not expose the patient to radiation or contrast agents. However, our results indicate that surgical lymph node staging is still needed to precisely determine inguinal lymph node status in vulvar cancer, especially because a missed lymph node-metastasis (i.e., a false negative diagnosis) is often fatal. Conflict of interest statement All authors disclose no conflicts of interest.
References [1] McComiskey MH, Robinson A, Hoo C, Khan K, Price JH. Radical excision of vulval disease and multidisciplinary tissue flap reconstruction: the belfast experience. Int J Gynecol Cancer 2012;22:323–7. [2] Hampl M, Hantschmann P, Michels W, Hillemanns P. Validation of the accuracy of the sentinel lymph node procedure in patients with vulvar cancer: results of a multicenter study in Germany. Gynecol Oncol 2008;111:282–8. [3] Maggino T, Landoni F, Sartori E, Zola P, Gadducci A, Alessi C, et al. Patterns of recurrence in patients with squamous cell carcinoma of the vulva. A multicenter CTF Study. Cancer 2000;89:116–22. [4] Morton DL, Thompson JF, Essner R, Elashoff R, Stern SL, Nieweg OE, et al. Validation of the accuracy of intraoperative lymphatic mapping and sentinel lymphadenectomy for early-stage melanoma: a multicenter trial. Multicenter Selective Lymphadenectomy Trial Group. Ann Surg 1999;230:453–63 [discussion 463–5]. [5] De Hullu JA, Hollema H, Piers DA, Verheijen RH, van Diest PJ, Mourits MJ, et al. Sentinel lymph node procedure is highly accurate in squamous cell carcinoma of the vulva. J Clin Oncol 2000;18:2811–6. [6] Oonk MHM, Hollema H, de Hullu JA, van der Zee AGJ. Prediction of lymph node metastases in vulvar cancer: a review. Int J Gynecol Cancer 2006;16:963–71. [7] De Hullu JA, Hollema H, Lolkema S, Boezen M, Boonstra H, Burger MPM, et al. Vulvar carcinoma. The price of less radical surgery. Cancer 2002;95:2331–8. [8] Gruber I, Hahn M, Fehm T, Hann von Weyhern C, Stäbler A, Winckelmann A, et al. Relevance and methods of interventional breast sonography in preoperative axillary lymph node staging. Ultraschall Med 2012;33:337–43 [Stuttgart, Germany:1980].
117
[9] Kebudi A, Calişkan C, Yetkin G, Celebi S, Işgör A, Mesrur Halefoğlu A, et al. The role of pre-operative B mode ultrasound in the evaluation of the axillary lymph node metastases in the initial staging of breast carcinoma. Acta Chir Belg 2005;105:511–4. [10] Sohaib SAA, Moskovic EC. Imaging in vulval cancer. Best Pract Res Clin Obstet Gynaecol 2003;17:543–56. [11] Akrivos N, Rodolakis A, Vlachos G, Sotiropoulou M, Papantoniou V, Biliatis I, et al. Detection and credibility of sentinel node in vulvar cancer: a single institutional study and short review of literature. Arch Gynecol Obstet 2011;284:1551–6. [12] Klar M, Bossart M, Stickeler E, Brink I, Orlowska-Volk M, Denschlag D. Sentinel lymph node detection in patients with vulvar carcinoma; feasibility of intraoperative mapping with technetium-99 m-labeled nanocolloid. Eur J Surg Oncol 2011;37:818–23. [13] Bilimoria KY, Bentrem DJ, Hansen NM, Bethke KP, Rademaker AW, Ko CY, et al. Comparison of sentinel lymph node biopsy alone and completion axillary lymph node dissection for node-positive breast cancer. J Clin Oncol 2009;27:2946–53. [14] Martínez-Palones JM, Pérez-Benavente MA, Gil-Moreno A, Díaz-Feijoo B, Roca I, García-Jiménez A, et al. Comparison of recurrence after vulvectomy and lymphadenectomy with and without sentinel node biopsy in early stage vulvar cancer. Gynecol Oncol 2006;103:865–70. [15] Mäkelä PJ, Leminen A, Kääriäinen M, Lehtovirta P. Pretreatment sonographic evaluation of inguinal lymph nodes in patients with vulvar malignancy. J Ultrasound Med 1993;12:255–8. [16] Hall TB, Barton DPJ, Trott PA, Nasiri N, Shepherd JH, Thomas JM, et al. The role of ultrasound-guided cytology of groin lymph nodes in the management of squamous cell carcinoma of the vulva: 5-year experience in 44 patients. Clin Radiol 2003;58: 367–71. [17] Moskovic EC, Shepherd JH, Barton DP, Trott PA, Nasiri N, Thomas JM. The role of high resolution ultrasound with guided cytology of groin lymph nodes in the management of squamous cell carcinoma of the vulva: a pilot study. Br J Obstet Gynaecol 1999;106:863–7. [18] Abang Mohammed DK, Uberoi R, de B Lopes A, Monaghan JM. Inguinal node status by ultrasound in vulva cancer. Gynecol Oncol 2000;77:93–6. [19] Kataoka MY, Sala E, Baldwin P, Reinhold C, Farhadi A, Hudolin T, et al. The accuracy of magnetic resonance imaging in staging of vulvar cancer: a retrospective multi-centre study. Gynecol Oncol 2010;117:82–7. [20] Bipat S, Fransen GA, Spijkerboer AM, van der Velden J, Bossuyt PMM, Zwinderman AH, et al. Is there a role for magnetic resonance imaging in the evaluation of inguinal lymph node metastases in patients with vulva carcinoma? Gynecol Oncol 2006;103: 1001–6. [21] Land R, Herod J, Moskovic E, King M, Sohaib SA, Trott P, et al. Routine computerized tomography scanning, groin ultrasound with or without fine needle aspiration cytology in the surgical management of primary squamous cell carcinoma of the vulva. Int J Gynecol Cancer 2006;16:312–7. [22] Cohn DE, Dehdashti F, Gibb RK, Mutch DG, Rader JS, Siegel BA, et al. Prospective evaluation of positron emission tomography for the detection of groin node metastases from vulvar cancer. Gynecol Oncol 2002;85:179–84. [23] Choi JS, Kim MJ, Moon HJ, Kim E-K, Yoon JH. False negative results of preoperative axillary ultrasound in patients with invasive breast cancer: correlations with clinicopathologic findings. Ultrasound Med Biol 2012;38:1881–6. [24] Sever A, Broillet A, Schneider M, Cox K, Jones S, Weeks J, et al. Dynamic visualization of lymphatic channels and sentinel lymph nodes using intradermal microbubbles and contrast-enhanced ultrasound in a swine model and patients with breast cancer. J Ultrasound Med 2010;29:1699–704. [25] Ouyang Q, Chen L, Zhao H, Xu R, Lin Q. Detecting metastasis of lymph nodes and predicting aggressiveness in patients with breast carcinomas. J Ultrasound Med 2010;29:343–52.
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
The role of preoperative ultrasound evaluation of inguinal lymph nodes in patients with vulvar malignancy Nikolaus de Gregorio ⁎,1, Florian Ebner 1, Lukas Schwentner, Thomas W.P. Friedl, Miriam Deniz, Krisztián Látó, Rolf Kreienberg, Wolfgang Janni, Dominic Varga Universitätsfrauenklinik Ulm, Prittwitzstrasse 43, 89075fl, Germany
H I G H L I G H T S • We looked how well ultrasonography could identify positive nodes in vulvar cancer patients. • Despite good results, this method is not good enough to spare the patient a surgical staging.
a r t i c l e
i n f o
Article history: Received 13 June 2013 Accepted 24 July 2013 Available online 7 August 2013 Keywords: Vulvar cancer Inguinal lymph nodes Sentinel lymph node biopsy Inguinal ultrasound
a b s t r a c t Objectives. Inguinal lymphadenectomy in vulvar malignancies is associated with significant morbidity, especially in patients over 70 years old. Under certain conditions, surgical guidelines recommend biopsy and evaluation of the sentinel node in early vulvar cancer. The purpose of our study is to evaluate ultrasonography as a predictor of inguinal lymph node involvement. Methods. A retrospective study was performed with 60 patients who had vulvar malignancies (92% of which were squamous cell carcinomas) and who were treated at our hospital between 2002 and 2012. The patients ranged in age from 35 to 89 years, with a median age of 76 years. In total, 118 groin scans were retrospectively evaluated for sonographic evidence of lymph node involvement (i.e., absence of fatty hilum, irregular shape, cortical region diameter and vascularization pattern). The results were then compared with histopathologically confirmed lymph node status. Results. Histopathologically confirmed lymph node status was available for 107 of the inguinal nodes examined by ultrasound, and lymph node metastases were found in 38 (35.5%) cases. The presence or absence of inguinal lymph node metastases was correctly identified by sonography in 92 (86.0%) of the scanned areas. Sensitivity was 76.3%, specificity was 91.3%, and positive and negative predictive values were 82.9% and 87.5%, respectively. Conclusions. Ultrasonography of the inguinal lymph nodes showed a relatively high sensitivity and specificity for predicting inguinal tumor metastases. However, our results indicate that surgical lymph node staging is still needed to precisely determine inguinal lymph node status in vulvar cancer, especially because a missed lymph node-metastasis is often fatal. © 2013 Elsevier Inc. All rights reserved.
Introduction In treating vulvar cancer, it is currently unclear whether sentinel node biopsy is a sufficient substitute to prevent the morbidity linked with complete removal of the inguinal nodes. Standard treatment involves wide radical excision of the primary tumor and inguinal lymphadenectomy. Even with contemporary surgical techniques, half of treated patients suffer from post-surgical complications, such as ⁎ Corresponding author at: Universitätsfrauenklinik Ulm, Prittwitzstrasse 43, 89075 Ulm, Germany. Fax: +49 73150058502. E-mail address: [email protected] (N. de Gregorio). 1 Both authors contributed equally. 0090-8258/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ygyno.2013.07.103
impaired wound healing or acute infections [1]. These complications are predominantly short term; however, lymphedema caused by lymphadenectomy can be a lifelong issue. Nevertheless, because nodal involvement is negatively correlated with survival, lymph node status remains the most important predictor of overall prognosis [2]. Fiftythree percent of all recurrences develop in the groin, with a 5-yearsurvival rate of only 27%; in the case of a locoregional relapse, the 5-year-survival rate is 60% [3]. While little progress has been made in the treatment of lymphedema, the modern sentinel lymph node biopsy (SLNB) procedure is intended to prevent lymphedema by reducing the number of patients subjected to a full lymphadenectomy. The SLNB procedure was proposed by Morton et al. [4] as an alternative to observation or elective regional lymphadenectomy in women
114
N. de Gregorio et al. / Gynecologic Oncology 131 (2013) 113–117
with early cutaneous melanoma. This technique provides the opportunity for early intervention in women with SLN metastases and spares women without SLN metastases the morbidity of unnecessary regional lymphadenectomy. As with melanoma and breast cancer, vulvar cancer is in an excellent anatomic position for the SLN procedure. The tumor is easily reached for tracer injection, and the lymph drainage is limited to either groin. The Gynecologic Oncology Group (GOG) has studied various surgical and radiotherapy techniques for women with vulvar cancer. Consequently, the SLN procedure is only recommended for women with clinically negative lymph nodes [5]. As in breast cancer, palpation is one method of determining clinical lymph node status. Another method is ultrasonography. The aim of this study is to evaluate whether inguinal ultrasonography is a useful diagnostic tool to predict lymph node involvement in vulvar cancer patients. Accordingly, we retrospectively assessed ultrasound images obtained before surgery and compared the results to the corresponding histopathologically confirmed lymph node status.
distributed data were conducted using the Mann–Whitney-U test, while categorical data were analyzed using the Chi-square test or Fisher's exact test. When investigating the predictive factors for lymph node involvement, patients were excluded from the analysis if they had histopathologically confirmed negative lymph node status for one inguinal area but no histopathologically confirmed lymph node status for the other inguinal area, because we could not be certain about the lymph node status of the other side. To evaluate ultrasonography as a predictor of inguinal lymph node involvement, the scan of each inguinal area was treated as an independent event. All statistical tests reported here were two-sided, and a test result was regarded as statistically significant if p b 0.05. All statistical analyses were performed using IBM Statistical Package for the Social Sciences (SPSS) version 19.
Methods
Patient characteristics for the 60 patients are shown in Table 1. The majority of patients had squamous cell carcinoma, the most common tumor size was pT1b, and histological grade was mostly grade 2. Invasion depth (data available for 46 patients) ranged from 1 to 22 mm with a median of 4 mm. Sentinel node biopsy revealed positive nodes in 22 (25.6%) of the 86 cases. Histopathologically confirmed lymph node metastases were found in 38 (35.5%) of the 107 inguinal areas studied. A median of 3 nodes were dissected if the nodes were negative (range 1–19, n = 69), and a median of 7 nodes were dissected if the nodes were histopathologically positive (range 1–15, n = 38); the difference was statistically significant (Mann– Whitney-U-test, z = 4.48, p b 0.001). No significant difference was found in the number of nodes that were dissected between the left and right groins (Mann–Whitney-U-test, z = 1.21, p = 0.23).
In total, 60 patients with early vulvar malignancy treated from January 2002 to December 2012 at the Department of Gynecology and Obstetrics in Ulm were included in this study. Patients with tumor stage FIGO 1a, which is a negative predictor for nodal involvement, were excluded [6,7]. However, three patients with histologically proven pT1a were included, as in preoperative clinical examination a higher stage was assumed. Patients who presented with suspected vulvar malignancy were referred to the ultrasound department for a presurgical inguinal scan. Neither the patients nor the health care providers had pay for the additional ultrasound, which was offered on a voluntary basis to all patients studied. Ultrasonography was performed using GE 730 Expert and Voluson E8 ultrasound devices (GE Healthcare, Solingen, Germany). Both devices are equipped with linear probes at 12 MHz. Greyscale and color Doppler imaging were performed with the linear probes. Images were stored in the ViewPoint Software System (GE Healthcare, Solingen, Germany) and reviewed by three investigators according to the recommendations for axillary lymph nodes (i.e., the absence of fatty hilum, irregular shape, cortical region diameter ≥4 mm and peripheral vascularization) (Fig. 1 and Fig. 2) [8,9]. In the few cases in which the initial assessment of the three investigators differed, a consensus was reached after visually cross checking the images. Overall, ultrasound scans were available for 118 inguinal areas; for two of the 60 patients, ultrasonography was only performed on the right groin. Surgery was performed by 4 experienced surgeons. Following national guidelines, technetium-99 m was applied one day prior to the procedure, and blue dye was given intraoperatively to tag the sentinel node. If the tumor was unilaterally localized, the procedure was only performed on the ipsilateral groin. All stained and/or radionuclidemarked nodes were identified visually and with gamma probe, surgically removed and then subjected to a frozen section procedure. If a neoplasia was detected, a complete inguinofemoral lymphadenectomy was performed. In 39 patients, a successful sentinel node biopsy was performed for both inguinal areas, and in 8 patients, sentinel node biopsy was performed for one side only, resulting in a total of 86 sentinel node biopsies. Histopathologically confirmed lymph node status was obtained for both inguinal areas in 47 patients, while for the remaining 13 patients, histopathology was available only for one of the two inguinal areas (due to the tumor being unilaterally localized as described above). Thus, histopathologically confirmed lymph node status was available for 107 of the 118 scanned groins. In 3 cases (4.8%), lymph node metastases that were later confirmed by histopathological analysis had been missed on assessment of the fresh frozen sections. Statistical analyses were performed using the appropriate nonparametric tests. Comparisons regarding continuous non-normally
Results Patient characteristics
Predictive factors for lymph node involvement No significant difference in age was found between patients with and without positive nodes (Mann–Whitney U-test, z = 0.5, p = 0.62). However, the tumors of patients with lymph node metastases had a significantly higher invasion depth (median 10 mm, range 2–22 mm) than the tumors of patients without lymph node metastases (median 3 mm,
Table 1 Patient characteristics (n = 60). Age in years
Body Mass Index (n = 59)
Invasion depth in mm (n = 46)
Histological type
pT
pN
Histological grading
Mean Median Range Mean Median Range Mean Median Range Squamous cell carcinomas Melanomas Bowen-carcinoma Basalioma pT1a pT1b pT2 pT3 n/a pN0 pN1 pN2 pN3 G1 G2 G3 n/a
70.1 76 35–89 26.7 25.4 19.0–48.9 6.1 4.0 1–22 55 (91.7%) 3 (5.0%) 1 (1.7%) 1 (1.7%) 3 (5.0%) 33 (55.0%) 17 (28.3%) 5 (8.3%) 2 (3.3%) 31 (51.7%) 16 (26.7%) 8 (13.3%) 5 (8.3%) 10 (16.7%) 42 (70.0%) 4 (6.7%) 4 (6.7%)
N. de Gregorio et al. / Gynecologic Oncology 131 (2013) 113–117
115
Fig. 1. Physiological appearance of a groin node: Ovular with a fatty hilum, marked cortexmedullary border and narrow cortex. Fig. 3. Tumor grade is significantly related to the probability of positive nodes.
range 1–20 mm; Mann–Whitney U-test, z = 2.85, p = 0.004). In addition, patients with lymph node metastases had a significantly higher BMI compared to patients without lymph node metastases (Mann– Whitney U-test, z = 1.99, p = 0.046). Patients with higher tumor stages did not have a higher probability of positive nodes compared to patients with lower tumor stages (Chi-square test, χ2 = 3.2, p = 0.36). In contrast, tumor grade was significantly related to the probability of positive nodes (Chi-square test, χ2 = 11.19, p = 0.004). While no positive nodes were found in patients with grade I tumor, 50% and 56% of patients with grade II and III tumors, respectively, had lymph node metastases (Fig. 3).
Predictive value of ultrasound Ultrasonography correctly predicted the presence or absence of inguinal lymph node metastases in 92 (86%) of 107 cases (Fig. 4). In 6 cases, ultrasonography yielded a false positive diagnosis; in 9 cases, false negative results were found. The sensitivity and specificity were 76.3% and 91.3%, respectively; the positive predictive value was 82.9%, and the negative predictive value was 87.5% (see Table 2). In seven of the 9 false negative results, only one histopathologically confirmed
Fig. 2. Suspect groin node. Irregular shape without a fatty hilum, an indistinct cortexmedullary border and broadened cortex.
positive node was found, indicating that the sensitivity of ultrasound might be related to the number of positive nodes.
Subgroup analysis We further investigated whether the diagnostic potential of ultrasonography of inguinal areas differed among various patient subgroups. For patients with a tumor invasion depth of less than 5 mm, there were no false positive diagnoses and specificity (100%) was significantly higher than for patients with a tumor invasion depth of 5 mm or more (specificity 82.4%; Fisher's exact test, p = 0.026). However, the clinically important rate of false negative results was almost the same for patients with tumor invasion depth below or above 5 mm (22.2% and 23.8%, respectively). Consequently, no significant difference was found regarding the sensitivity of ultrasonography between these two groups (Fisher's exact test, p = 1.0). For patients younger than 76 years, specificity (100%) was significantly higher than for patients 76 years or older (83.3%; Fisher's exact test, p = 0.025). By contrast, sensitivity was higher but not significantly
Fig. 4. Sonography correctly predicts positive nodes.
116
N. de Gregorio et al. / Gynecologic Oncology 131 (2013) 113–117
Table 2 Binary classification table. Lymph node metastases
Ultrasound Overall
Nonsuspect Suspect
Overall
No
Yes
63 6 69
9 29 38
72 35 107
different for older patients (86.4%) compared to younger ones (62.5%; Fisher's exact test, p = 0.13). For patients with a BMI above 30, sensitivity was numerically lower as compared to patients with a BMI below 30 (66.7% vs 78.6%); however, the difference was not statistically significant (n = 22, Fisher's exact test, p = 0.66). Specificity was very similar for patients with a BMI above or below 30 (92.3% and 91.1%, respectively; n = 84, Fisher's exact test, p = 1.0). Finally, neither specificity nor sensitivity was significantly different between patients treated from 2002 to 2008 and patients treated from 2009 to 2012 (Fisher's exact test, both p N 0.4). Discussion This study demonstrates that groin ultrasound scans, although more reliable than other non-invasive diagnostic tools in predicting nodal status in vulvar malignancies, lack the high sensitivity and specificity that are necessary to avoid surgery, especially because missed positive nodes have a strong impact on prognosis and treatment. The relatively large number of patients examined in this rare tumor entity (to our knowledge, the largest sample size published to date) and the high percentage of nodal involvement observed in our patient sample strengthen our findings. The rate of positive lymph nodes found in our study is higher than in most other series reported [10]. This potential selection bias might originate from the fact that in contrast to other trials, patients with a clinically rated tumor stage T1a were not included in this study, as our national treatment guidelines do not require a surgical groin staging for these patients. This patient selection might have led to a higher rate of node positives as compared to other published data. Another potential selection bias might be caused by the referral practice as our hospital is a regional cancer referral center, with a high number of severe cases transferred from other hospitals, resulting in an increased rate of node positive patients. It is difficult to exactly predict how a higher proportion of nodenegative patients would have influenced our results. Generally speaking, sensitivity and specificity are not expected to change due to an increased proportion of node-negative patient, while the positive predictive value would be expected to decrease and the negative predictive value would be expected to increase. That is, given that both sensitivity and specificity should not be affected by changes in the proportion of node-negative patients in the study population, our main results and conclusions would remain the same even if the proportion of nodenegative patients in our sample would be higher. In our study, we found no significant association between tumor stage and the probability of having positive lymph nodes; however, there was a non-significant tendency for higher tumor stages to be associated with an increased probability of having positive lymph nodes. One possible reason for the lack of a significant relationship might be that our sample included only three patients with a pT1a tumor, which is known to be a negative predictor for lymph node involvement. In addition, our sample included only five pT3 and no pT4 tumors, and this low number of large tumors might be another reason why the test for an association between tumor stage and nodal status did not reach statistical significance. Because up to 70% of patients have groin or lower-limb problems after radical groin surgery, the preoperative identification of nodal
positive disease is a major issue. While many studies focus on proper identification of the sentinel node [11,12], less effort has been exerted in trying to avoid the procedure completely. The difference in the number of dissected nodes between radical groin dissection and the sentinel node technique tends to be lower than in other cancers, for example, breast cancer [2,13,14], although current data suggest that axillary dissection performed after a positive sentinel node does not have a positive impact on prognosis. At the same time, locoregional recurrence of breast cancer in the axilla is still considered to be curable, while a groin relapse of a vulvar malignancy is often considered to be incurable due to limited therapeutic options. Mäkelä et al. [15] found that ultrasonography is more sensitive than preoperative or intraoperative palpation (82% vs 9% vs 55%). High sensitivity (92%) for groin ultrasound alone has been described previously; the addition of fine-needle aspiration cytology (FNAC) improves the results slightly [16]. Our study showed similar sensitivity and specificity when compared to earlier publications of ultrasound in combination with FNC [17]. This outcome might be due to the improvement of ultrasound technology. In addition, our sonographers routinely perform axillary lymph node ultrasounds and are experts in lymph node evaluation. Abang et al. [18] published sonographic criteria showing a sensitivity of 89% for determining lymph node status. These criteria use the short axis diameter and long axis/short axis ratio to identify positive nodes. However, the results of this study should be interpreted carefully because of the small sample size of 20 patients and the involvement of only a small number of lymph nodes, which resulted in a specificity of only 38%. Several other imaging techniques of the groin have been published during the last several years. In comparison to ultrasound, slightly lower rates of positive node detection have been reported with contrast enhanced magnetic resonance imaging (CE-MRI) [19]. Even lower sensitivity (52%) and positive predictive values (46%) have been reported for conventional MRI, leading the authors to conclude that there is no role for MRI in the diagnosis of groin nodes [20]. Similar data were published for preoperative CT-scans [21], where the sensitivity and positive predictive value were both 58%. Few reports are available regarding the use of positron emission tomography (PET) to detect groin metastases. Cohn et al. [22] published a study with 15 patients, but concluded PET to be relatively insensitive because a negative PET did not reliably indicate a pathologically negative groin. Inevitably, every retrospective study has limitations. A selection bias due to unknown factors cannot be ruled out. One limitation is the relatively small number of patients due to the low incidence compared to other cancers, such as breast cancer. The low incidence makes it difficult to design a prospective or randomized trial. Another limitation is that the final histological diagnosis is often not known at the time of the ultrasound; therefore, not only the squamous cell carcinomas but also a small amount of other pathologies are included. A possible confounding factor could be the improvement in ultrasound and surgical techniques during the 10-year data collection period, which might have resulted in inadequate imaging and thus worse diagnostic potential of ultrasonography in patients who were treated earlier in the study. However, the key diagnostic values did not differ significantly between patients treated from 2002 to 2008 and patients treated from 2009 to 2012 (see Results). Regarding surgical techniques, no major changes were identified when reviewing the patients' operative reports. As far as we know, no data on how BMI influences groin ultrasound have been previously published. However, there is data concerning axillary ultrasound suggesting that BMI is not associated with false negative results [23]. Another potential bias might be, that the patients' body weight was self reported, and therefore most likely rather underestimated. Overall, ultrasound seems to be the most efficacious of these imaging techniques because of its high sensitivity, low overall costs and the absence of exposure to radiation and/or contrast agents. Preoperative groin ultrasound is now a routine procedure in patients with vulvar
N. de Gregorio et al. / Gynecologic Oncology 131 (2013) 113–117
malignancy in our hospital. A sonographically suspicious node would result in the patient either having a preoperative needle biopsy or the suspect node would be identified with sonographically guided wire tag and excised in addition to the sentinel node. Promising future developments might include the contrast enhanced ultrasound (CEUS) technique, where microbubble contrast is used to visualize lymphatic channels and sentinel nodes. This technique has already been successfully demonstrated in breast cancer patients [24,25]. Conclusions Ultrasonography of the inguinal lymph nodes showed a relatively high sensitivity and specificity for predicting inguinal tumor metastases and is also a cost effective, easy-to-learn and easy-to-perform method that does not expose the patient to radiation or contrast agents. However, our results indicate that surgical lymph node staging is still needed to precisely determine inguinal lymph node status in vulvar cancer, especially because a missed lymph node-metastasis (i.e., a false negative diagnosis) is often fatal. Conflict of interest statement All authors disclose no conflicts of interest.
References [1] McComiskey MH, Robinson A, Hoo C, Khan K, Price JH. Radical excision of vulval disease and multidisciplinary tissue flap reconstruction: the belfast experience. Int J Gynecol Cancer 2012;22:323–7. [2] Hampl M, Hantschmann P, Michels W, Hillemanns P. Validation of the accuracy of the sentinel lymph node procedure in patients with vulvar cancer: results of a multicenter study in Germany. Gynecol Oncol 2008;111:282–8. [3] Maggino T, Landoni F, Sartori E, Zola P, Gadducci A, Alessi C, et al. Patterns of recurrence in patients with squamous cell carcinoma of the vulva. A multicenter CTF Study. Cancer 2000;89:116–22. [4] Morton DL, Thompson JF, Essner R, Elashoff R, Stern SL, Nieweg OE, et al. Validation of the accuracy of intraoperative lymphatic mapping and sentinel lymphadenectomy for early-stage melanoma: a multicenter trial. Multicenter Selective Lymphadenectomy Trial Group. Ann Surg 1999;230:453–63 [discussion 463–5]. [5] De Hullu JA, Hollema H, Piers DA, Verheijen RH, van Diest PJ, Mourits MJ, et al. Sentinel lymph node procedure is highly accurate in squamous cell carcinoma of the vulva. J Clin Oncol 2000;18:2811–6. [6] Oonk MHM, Hollema H, de Hullu JA, van der Zee AGJ. Prediction of lymph node metastases in vulvar cancer: a review. Int J Gynecol Cancer 2006;16:963–71. [7] De Hullu JA, Hollema H, Lolkema S, Boezen M, Boonstra H, Burger MPM, et al. Vulvar carcinoma. The price of less radical surgery. Cancer 2002;95:2331–8. [8] Gruber I, Hahn M, Fehm T, Hann von Weyhern C, Stäbler A, Winckelmann A, et al. Relevance and methods of interventional breast sonography in preoperative axillary lymph node staging. Ultraschall Med 2012;33:337–43 [Stuttgart, Germany:1980].
117
[9] Kebudi A, Calişkan C, Yetkin G, Celebi S, Işgör A, Mesrur Halefoğlu A, et al. The role of pre-operative B mode ultrasound in the evaluation of the axillary lymph node metastases in the initial staging of breast carcinoma. Acta Chir Belg 2005;105:511–4. [10] Sohaib SAA, Moskovic EC. Imaging in vulval cancer. Best Pract Res Clin Obstet Gynaecol 2003;17:543–56. [11] Akrivos N, Rodolakis A, Vlachos G, Sotiropoulou M, Papantoniou V, Biliatis I, et al. Detection and credibility of sentinel node in vulvar cancer: a single institutional study and short review of literature. Arch Gynecol Obstet 2011;284:1551–6. [12] Klar M, Bossart M, Stickeler E, Brink I, Orlowska-Volk M, Denschlag D. Sentinel lymph node detection in patients with vulvar carcinoma; feasibility of intraoperative mapping with technetium-99 m-labeled nanocolloid. Eur J Surg Oncol 2011;37:818–23. [13] Bilimoria KY, Bentrem DJ, Hansen NM, Bethke KP, Rademaker AW, Ko CY, et al. Comparison of sentinel lymph node biopsy alone and completion axillary lymph node dissection for node-positive breast cancer. J Clin Oncol 2009;27:2946–53. [14] Martínez-Palones JM, Pérez-Benavente MA, Gil-Moreno A, Díaz-Feijoo B, Roca I, García-Jiménez A, et al. Comparison of recurrence after vulvectomy and lymphadenectomy with and without sentinel node biopsy in early stage vulvar cancer. Gynecol Oncol 2006;103:865–70. [15] Mäkelä PJ, Leminen A, Kääriäinen M, Lehtovirta P. Pretreatment sonographic evaluation of inguinal lymph nodes in patients with vulvar malignancy. J Ultrasound Med 1993;12:255–8. [16] Hall TB, Barton DPJ, Trott PA, Nasiri N, Shepherd JH, Thomas JM, et al. The role of ultrasound-guided cytology of groin lymph nodes in the management of squamous cell carcinoma of the vulva: 5-year experience in 44 patients. Clin Radiol 2003;58: 367–71. [17] Moskovic EC, Shepherd JH, Barton DP, Trott PA, Nasiri N, Thomas JM. The role of high resolution ultrasound with guided cytology of groin lymph nodes in the management of squamous cell carcinoma of the vulva: a pilot study. Br J Obstet Gynaecol 1999;106:863–7. [18] Abang Mohammed DK, Uberoi R, de B Lopes A, Monaghan JM. Inguinal node status by ultrasound in vulva cancer. Gynecol Oncol 2000;77:93–6. [19] Kataoka MY, Sala E, Baldwin P, Reinhold C, Farhadi A, Hudolin T, et al. The accuracy of magnetic resonance imaging in staging of vulvar cancer: a retrospective multi-centre study. Gynecol Oncol 2010;117:82–7. [20] Bipat S, Fransen GA, Spijkerboer AM, van der Velden J, Bossuyt PMM, Zwinderman AH, et al. Is there a role for magnetic resonance imaging in the evaluation of inguinal lymph node metastases in patients with vulva carcinoma? Gynecol Oncol 2006;103: 1001–6. [21] Land R, Herod J, Moskovic E, King M, Sohaib SA, Trott P, et al. Routine computerized tomography scanning, groin ultrasound with or without fine needle aspiration cytology in the surgical management of primary squamous cell carcinoma of the vulva. Int J Gynecol Cancer 2006;16:312–7. [22] Cohn DE, Dehdashti F, Gibb RK, Mutch DG, Rader JS, Siegel BA, et al. Prospective evaluation of positron emission tomography for the detection of groin node metastases from vulvar cancer. Gynecol Oncol 2002;85:179–84. [23] Choi JS, Kim MJ, Moon HJ, Kim E-K, Yoon JH. False negative results of preoperative axillary ultrasound in patients with invasive breast cancer: correlations with clinicopathologic findings. Ultrasound Med Biol 2012;38:1881–6. [24] Sever A, Broillet A, Schneider M, Cox K, Jones S, Weeks J, et al. Dynamic visualization of lymphatic channels and sentinel lymph nodes using intradermal microbubbles and contrast-enhanced ultrasound in a swine model and patients with breast cancer. J Ultrasound Med 2010;29:1699–704. [25] Ouyang Q, Chen L, Zhao H, Xu R, Lin Q. Detecting metastasis of lymph nodes and predicting aggressiveness in patients with breast carcinomas. J Ultrasound Med 2010;29:343–52.