- Email: [email protected]
Peritoneal Carcinomatosis in Primary Ovarian Cancer: Ultrasound Detection and Comparison with Computed Tomography
Ultrasound in Med. & Biol., Vol. -, No. -, pp. 1–9, 2017 Ó 2017 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/$ - see front matter
http://dx.doi.org/10.1016/j.ultrasmedbio.2017.02.016
d
Original Contribution PERITONEAL CARCINOMATOSIS IN PRIMARY OVARIAN CANCER: ULTRASOUND DETECTION AND COMPARISON WITH COMPUTED TOMOGRAPHY ZHENHONG QI, YIXIU ZHANG, QING DAI, YU XIA, and YUXIN JIANG Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China (Received 2 July 2016; revised 14 February 2017; in final form 21 February 2017)
Abstract—We retrospectively compared detection rates and consistency for diagnosis of peritoneal carcinomatosis (PC) of primary ovarian cancer (OC) between ultrasound (US) and computed tomography (CT) scans in 41 patients whose PC of OC (stages IIC–IV) had been diagnosed by histopathology findings. Compared with CT detection rates, those for US were significantly higher for metastases to the pelvic area (92.3% vs. 43.6%, p , 0.001) and bowel surface (64.0% vs. 16.0%, p 5 0.002); however, they did not significantly differ for other sites: omentum, diaphragm, lateral peritoneum, mesenteric, hepatic and splenic surfaces. Diagnostic consistency between US and CT scans were fair to moderate for splenic (k 5 0.806), hepatic (k 5 0.485), lateral peritoneum (k 5 0.450) and diaphragm (k 5 0.338) surfaces, but poorly consistent for other parts (k 5 0.144–0.229). In summary, US can complement CT scans, especially for detecting PC of primary OC metastases in pelvic and bowel surfaces. (E-mail: [email protected]) Ó 2017 World Federation for Ultrasound in Medicine & Biology. Key Words: Ovarian cancer, Peritoneum, Peritoneal carcinomatosis, Ultrasound, Computed tomography.
even at the bedside, and is reportedly .90% accurate in detecting benign and malignant intra-abdominal ovarian masses (Valentin 1999). However, very few data are available in existing literature regarding the involvement of PC in primary OC. This study compared detection rates by US and computed tomography (CT) scans for PC and evaluated their diagnostic consistency.
INTRODUCTION Ovarian cancer (OC) is the sixth most commonly diagnosed cancer among women worldwide and causes more deaths per year than any other cancer of the female reproductive system (Permuth-Wey and Sellers 2009). It is one of the most occult cancers, as 70% of tumors are diagnosed in the late stages of disease (FIGO III and IV). In addition, 70% of OC patients present with involvement of the peritoneum (Nougaret et al. 2012; Sehouli et al. 2009; Woodward et al. 2004). Aggressive surgery and chemotherapy are often combined to extend the survival of patients with OC; however, prognosis is still poor because of late-stage disease at diagnosis. The presence of a residual lesion after initial surgery is an important prognostic factor in patients with advanced OC. Therefore, detecting and detailing involvement of peritoneal carcinomatosis (PC) in patients suspected of having primary OC is critical. Ultrasound (US) is the least expensive technique; it is quick and simple to perform,
METHODS Study population We retrospectively enrolled into this study 41 women who had been diagnosed with primary OC with PC by histopathology findings from January 2008 to December 2015. All 41 women had been hospitalized in the Department of Obstetrics and Gynecology in Peking Union Medical College Hospital. They were evaluated with US examinations and CT scans independently, and then underwent optimal or suboptimal primary debulking surgery within 10 d. All ultrasound examinations were performed and evaluated by the same senior ultrasound physician; CT scan results were reported by seven radiologists. Surgical findings and pathologic diagnoses were regarded as gold standards for evaluating US and CT scan results. Of the 41 patients, 14 were pre-menopausal and 27 were
Address correspondence to: Yuxin Jiang, Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, No 1 Shuaifuyuan, Wangfujing, Beijing, China. E-mail: [email protected] 1
2
Ultrasound in Medicine and Biology
post-menopausal. None of them had taken estrogen during their lifetime. Only two patients had reported family histories of OC. This retrospective study was approved by the ethics committee of Peking Union Medical Hospital, which waived the need for written informed consent from the patients. All records data were de-identified and analyzed anonymously.
US examination The US examinations were performed using a LOGIQ 9 (GE Healthcare, Milwaukee, WI, USA) or an IU22 or HD11 (Philips, Eindhoven, Netherlands) with 9- to 3- or 8- to 4-MHz vaginal probes for transvaginal ultrasound imaging and 5- to 2-MHz probes for transabdominal ultrasound imaging. Transvaginal scans of the pelvic peritoneum in longitudinal and transverse views were performed first, followed by trans-abdominal scans. Scans focused on omentum, hepatic and splenic surfaces, diaphragm, lateral peritoneum and bowel surface. When ascites was observed, mesenteric and intestinal surfaces were carefully examined. Size, echo and color Doppler flow imaging (CDFI) of pelvic and abdominal peritoneal implants were recorded, along with the boundaries and relationships with lesions in surrounding tissues. Imaging was saved simultaneously. Detected PC in the abdominal cavity was carefully recorded with respect to the area involved, which included bowel surface, hepatic and splenic surfaces, diaphragm, lateral peritoneum, omentum, mesenteric surface and pelvis. Our method of examining patients for PC using ultrasound was as follows: 1. Pelvic implants: Transvaginal US was used to examine the pouch of Douglas, pelvic parietal peritoneum and surfaces of visceral peritoneum; lesions could be detected on of bowel, bladder, uterine and pelvic parietal peritoneum surfaces. 2. Omentum: Abdominal US of the upper and lower abdomen in the longitudinal or transverse direction was used. Lesions were usually located between the front abdominal wall and the bowel. 3. Hepatic surface: The patient was placed in the supine position. First, a transverse scan of the left liver at the level of the subxiphoid was performed. Then, an oblique scan was performed along the right costal margin of the liver, from left to right, followed by an intercostal oblique and coronal scan of the liver surface in the left lateral position. 4. Diaphragm surface: The right costal margin was scanned at an oblique angle, as was an intercostal oblique and coronal section in the left lateral position. 5. Splenic surface: Patient was moved to the right lateral position. The spleen can be examined via intercostal
Volume -, Number -, 2017
Table 1. CA-125 levels and ascites volumes CA-125 level (IU/mL)
n
Ascites volume (mL)
n
,100 100–200 201–500 501–1000 1001–2000 2001–5000 .5000 Total
2 4 10 7 6 7 5 41
Non-ascites #500 .500, #1000 .1000, #1500 .1500, #2000 .2000
4 12 8 9 5 3 41
or left coronal scanning, moving the probe backward and forward. 6. Mesenteric and bowel surfaces: Patient was placed in the supine position. Hyper-echoic bowel could be observed in all four quadrants. Operator looked for implants on mesenteric and bowel surfaces while scanning. 7. Lateral peritoneum: Patient was placed in the right or left lateral position. The left or right lateral abdomen, adjacent to the side of the abdominal wall, was scanned from top to bottom. CT examination Computed tomography scans were performed using GE 64-slice helical scanners (GE Medical Systems, Milwaukee, WI, USA). Images were reconstructed at 7-mm intervals. All patients received the intravenous contrast medium Ultravist (Bayer, Berlin, Germany) through a bolus injection with a high-pressure syringe through the forearm vessel at 3 mL/s. After injection of the contrast agent, arterial-phase, portal-phase and delayed-phase images were obtained after 30, 60, and 200 s, respectively. The scan range was from the top of the diaphragm to the pubic symphysis. Each physician recorded a diagnosis based on each CT scan. Pre-operative descriptions of OC by abdominal pelvic CT included size, morphology and unilateral or bilateral character of ovarian masses; any features of malignancy; uterus, bladder, bowel, pelvic side-wall invasions or implants; ascites in the pelvis or upper abdomen and its Table 2. Histologic findings in patients with persistently abnormal screens (n 5 41) Final pathologic type
N
Borderline serous papillary adenofibroma Endometrioid carcinoma Clear cell carcinoma Sarcoma cancer (malignant mixed tumor m€ullerian) Poorly differentiated neuroendocrine carcinoma Serous papillary adenocarcinoma 1 endometrial carcinoma Mucinous carcinoma Serous papillary adenocarcinoma or cystadenocarcinoma Total
1 3 2 2 1 2 3 27 41
US detection of peritoneal carcinomatosis d Z. QI et al.
3
Fig. 1. Transvaginal ultrasound findings of pelvic implant lesions by ultrasound. (a) Bilateral stage IIIC ovarian clear-cell carcinoma in a 42-y-old woman (MASS 5 ovarian tumor). Arrows point to peritoneal implants ,5 mm. Presence of pelvic fluid facilitates detection of these small metastases. (b) Scan of a 69-y-old woman with bilateral stage IIIC ovarian sarcoma reveals a peritoneal metastasis on the bowel surface. Although ascitic fluid is absent, there is no gas in the intestines; the small implant (0.8 3 0.6 cm) is visible (1). (c) Right stage IIIC ovarian endometrioid carcinoma in a 69-y-old woman. Anechoic ascitic fluid clearly outlines the small implant (0.7 3 0.8 cm) on the bowel surface (arrow); color Doppler flow imaging reveals abundant blood flow in the lesion. (d) Right stage IIIC ovarian serous papillary adenocarcinoma in a 59-y-old woman. Implant (2.2 3 2.5 cm) on rectal surface was observed on ultrasound (1) and computed tomography (arrow). (e) Bilateral stage IIIC ovarian serous papillary adenocarcinoma in an 82-y-old woman. Implant (0.9 cm) appears sheet-like and irregular on bladder surface (1). (f) Transvaginal scan of pelvic bowel in a 69-y-old woman with bilateral stage IIIC ovarian serous papillary adenocarcinoma reveals a thickening hypo-echoic implant (1.4 cm) attached to both sides of the bowel (1). BL 5 bladder; BO 5 bowel; CX 5 cervix; UT 5 uterus.
volume; omentum metastases; involvement of small bowel mesentery and surface of bowel; surface or parenchymal hepatic and splenic metastases; implants in para-colic gutters, sites of other peritoneal/serosal implants outside the pelvis; sites of lymph nodes with short-axis diameters .1 cm or suspicious clusters of smaller lymph nodes. Lymph node and hepatic and splenic parenchyma metastases were not included in this study. Statistical analysis Categorical variables were described using frequencies and percentages. We calculated detection rates of US and CT scans at each site. The McNemar test was applied to compare the statistical differences in detection rates between US and CT scans, with surgical findings and histopathology diagnoses serving as gold standards. p , 0.05 was considered to indicate significance. We used k tests to evaluate the consistency of US and CT in diagnosing peritoneal metastasis of ovarian cancer (k . 0.81:
very good consistency, 0.61–0.80: good, 0.41–0.60: moderate, 0.21–0.40: fair, and ,0.20: poor). Statistical analyses were performed using the Statistical Package for the Social Sciences (Version 23.0, IBM, Armonk, NY, USA). RESULTS Demographic and clinical information The median age of the 41 patients was 55 y; their average age was 55.8 6 12.6 y (range: 27–82 y). CA125 levels were elevated for all 41 patients (#500 IU/ mL: n 5 16, .500 IU/mL: n 5 25). Ascites was documented in 37 patients (.500 mL: n 5 25, #500 mL: n 5 12). There were no correlations between CA-125 values and ascites volumes (Table 1). Operative and pathologic results On the basis of the surgical findings, bilateral ovarian lesions were present in 23 patients and unilateral
4
Ultrasound in Medicine and Biology
Volume -, Number -, 2017
Fig. 1. (continued).
Fig. 2. Omental cake and nodular omental metastases. (a) Left stage IIIC ovarian serous papillary adenocarcinoma in a 61-y-old woman. Longitudinal scan of umbilical upper abdomen reveals abundant blood flow in thickened mediumechoic omentum (arrows, OMEN). (b) Bilateral stage IIIC ovarian serous papillary adenocarcinoma in a 50-y-old woman. Thickened, rigid omentum in ascites fluid; color Doppler flow imaging confirmed vessel in the omentum (1: thickened omentum, 1.8 cm). (c,d) Left stage IIIC ovarian serous adenocarcinoma in a 45-y-old woman. Computed tomography (c) and ultrasound (d) images reveal nodular omental metastases (arrows).
US detection of peritoneal carcinomatosis d Z. QI et al.
ovarian lesions in 18 patients, for a total of 64 lesions, of which 12 were solid and 52 were cystic-solid; diameters measured ,5 cm (n 5 28), 5–10 cm (n 5 300) and .10 cm (n 5 6). The FIGO stages of the ovarian tumors ranged from IIC to IV (2 stage IIC, 1 stage IIIB, 35 stage IIIC [the majority stage, at 85.4%] and 3 stage IV). In the 2 patients with stage IIC tumors, one had a sheetlike implant on her peritoneal bladder fold with 300 mL ascites, which was detected during surgery, and the other had a 0.8 3 0.4 3 0.4-cm implant in the pouch of Douglas without ascites; both implants were missed by US and CT. Final pathologic diagnoses are listed in Table 2. US and CT findings In US detection, metastatic lesions appeared as hypoechoic or medium-echoic and nodular or sheet-like, or the omentum was diffusely thickened. Blood flow was detected in the peritoneal metastases through CDFI. US images tastatic lesions in different parts are provided in Figures 1–7. Peritoneal lesions ,5 mm were easily
5
detected in the presence of pelvic fluid by transvaginal ultrasound (Fig. 1a). Lesions were clearly visible on the surface of liver, with the liver serving as the acoustic window (Fig. 3a); especially between the liver and right kidney, lesions ,1 cm were detectable (Fig. 3b). Because the greater omentum is adjacent to the frontal abdominal walls, metastatic lesions between the abdominal wall and bowel can easily be detected as they are not affected by bowel gas (Fig. 2a–d). Although lesions on the mesenteric and bowel surfaces may not be easily detected by transabdominal US because of bowel gas or large amounts of ascites, we noticed that a moderate volume of ascites was helpful in lesion detection (Fig. 6). However, metastases to the diaphragm are difficult to find. In CT scans, metastatic lesions appeared nodular (Figs. 1d, 2c,d and 5a) or local (Fig. 3c,d), or as diffuse thickening in the greater omentum, or linear or nodular enhancement in the omentum. As for US, the lesions were easily detected with the presence of ascites. Sagittal and coronal sections can help detect metastasis in the diaphragm.
Fig. 3. Nodular and sheet-like implants on the liver surface. (a). Bilateral stage IV ovarian serous papillary adenocarcinoma in a 55-y-old woman. With the patient in the supine position, oblique left liver scan along the right costal margin reveals a hypo-echoic implant (1.7 3 0.9 cm, arrows) on the surface of the left liver lobe. (b) Bilateral stage IIIC ovarian serous papillary carcinoma in a 59-y-old woman. Small hypo-echoic metastasis (0.8 cm, arrow) is clearly visible on the surface of right liver lobe, with the liver and right kidney serving as the acoustic window. (c,d) Bilateral stage IIIC ovarian serous cystadenocarcinoma in a 55-y-old woman. Computed tomography axis (c) and ultrasonic coronal plane (d) images reveal sheet-like metastases on the liver surface (arrows). RK 5 right kidney.
6
Ultrasound in Medicine and Biology
Fig. 4. Right stage IV ovarian serous papillary cystadenocarcinoma in a 62-y-old woman. Intercostal oblique scan reveals a 0.6-cm hypo-echoic sheet-like lesion on the diaphragm surface and partial indistinct borders with the liver (1, metastasis).
Comparison of US and CT Peritoneal carcinomatosis was diagnosed in 39 patients and misdiagnosed in 2 patients by US and CT; the detection rates with US and CTare listed in Table 3. Respec-
Volume -, Number -, 2017
tive US and CT detection rates for PC were 92.3% versus 43.6% in the pelvis; 73.7% versus 89.5% in the omentum; 21.9% versus 31.2% on diaphragm surfaces; 38.5% versus 61.5% in the lateral peritoneum; 64.0% versus 16.0% in the bowel; mesenteric: 21.1% versus 57.9% in the mesentery; hepatic: 83.3% versus 58.3% in the liver; and 87.5% versus 62.5% in the spleen. There were significant differences in detecting pelvic (p , 0.001) and bowel surface (p 5 0.002) masses, but not masses in other areas. Diagnostic consistency between US and CT scans was fair to moderate in splenic (k 5 0.806), hepatic (k 5 0.485), lateral peritoneum (k 5 0.450) and diaphragm (k 5 0.338) surfaces. Diagnostic consistency was poor in omental (k 5 0.229), pelvic (k 5 0.179), bowel (k 5 0.171) and mesenteric (k 5 0.144) surfaces. Among omental lesions, 9 cases were detected only by CT and 3 only by US, whereas the detection rate for omental lesions increased to 97.4% (37/38) when US and CT scans were combined. Nineteen pelvic lesions were detected only by US, and the 17 pelvic implants detected by CT were all covered by US. In addition, among the 10 metastatic lesions ,1 cm on the hepatic surface, bowel surface and
Fig. 5. Nodular and sheet-like implants on the spleen surface. (a) Same patient as in Figure 3c and d. Intercostal ultrasound oblique scan reveals a medium-echoic lesion (2.8 3 1.5 cm, arrows) of the splenic hilum. Computed tomography axis image reveals a nodular lesion of the splenic hilum. (b) Right stage IIIC ovarian serous papillary adenocarcinoma in a 62-y-old woman. Coronal scan reveals a hypo-echoic sheet-like lesion (arrows) on the spleen surface. M 5 metastasis; SP 5 spleen.
US detection of peritoneal carcinomatosis d Z. QI et al.
Fig. 6. Bilateral ovarian serous cystadenocarcinoma in a 51-y-old woman. Right lower abdominal scan reveals a 1.5 3 1.2-cm hypoechoic mesenteric implant (M). AS 5 ascites.
pelvis (Figs 1a–c and 3b), 7 (63.6%) were detected by US and 3 (27.3%) were detected by CT.
DISCUSSION Reportedly, US is the most commonly used screening method, and was as sensitive as CA-125 level assessment in the largest multicenter study for OC screening (Menon et al. 2009). However, few reports and data are available describing how PC can be effectively detected via either trans-vaginal or transabdominal ultrasound (Savelli et al. 2005). In addition, no routine standard screening for PC in patients with suspected primary OC is available at our hospital. Our study supported the important role of US in PC diagnosis, including summarizing the US standard scanning guideline in case enhancing the detection rate of PC.
Fig. 7. Same patient as in Figure 6. Scan of lower right abdomen of patient in the left lateral position reveals a 3.8 3 1.7-cm hypo-echoic lesion on the lower right abdominal wall (1: lesion).
7
Diagnosis of peritoneal metastasis requires both a thorough and meticulous examination of all peritoneal surfaces (Hanbidge et al. 2003) and an operator who is familiar with the routes of OC metastasis, to detect specific sites of metastasis after a standard scan. The US detection rates in our study were significantly higher than those for CT for both pelvic and bowel surface implants. Our final results indicated that the US detection rate for pelvic metastases was 92.3% (36/39), similar to the 88% sensitivity reported by Savelli et al. (2005), but our CT detection rate was 43.6% (17/39), which was significantly lower than that in Savelli et al. In our study, 64% (16/25) of implants on bowel surfaces were detected, mainly through transvaginal ultrasound (Fig. 1b,c,f). The CT detection rate for bowel surfaces in our study was low at 16%, which was similar to the 14% reported by Forstner et al. (2010). Ascites was detected in 90.2% of patients (37/41) with PC, possibly because of improved visualization of the peritoneal layers, allowing better detection of pelvic and bowel implants (Henrich et al. 2007). In our study, results for US detection depended both on the patient’s condition and on her implant sites. Generally, it was easier to detect implants in the presence of moderate ascites or in thin patients. The presence of ascites was particularly beneficial in observing pelvic and parietal peritoneal lesions and the uterus–rectum fossa while avoiding ovarian tumors. In contrast, in obese patients or patients with intestinal gas, involvement of mesenteric and bowel surfaces made detection more difficult. We found US and CT scans to be moderately consistent in detecting hepatic and splenic surfaces; both had high detection rates. The liver and spleen form the acoustic window; particularly between the liver and right kidney, implants were easily detected, even those ,1.0 cm (Fig. 3b). The omentum is close to the anterior abdominal wall in a very superficial location, and without the influence of intestinal gas, masses located on the omentum and thickening of the omentum were more easily observed with US. Interestingly, although our detection rates for the omentum were high for both imaging modes (US: 73.7%, CT: 89.5%), their consistency was poor, which means that different cases of omentum metastasis were detected by US and CT, respectively. The detection rate increased to 97.4% when we combined the two methods. Tempany et al. (2000) reported an overall sensitivity of only 69% for US in detecting peritoneal metastases in patients with stage III–IV OC, significantly less than 92% for CT; therefore, routine use of US for OC staging is not recommended. Coakley et al. (2002) also reported an overall high sensitivity (85%–93%) for use of CT for ovarian cancer staging, but the detection rate for small peritoneal lesions was poor, with a sensitivity of only 25%–50% for lesions with a maximal diameter ,1 cm.
8
Ultrasound in Medicine and Biology
Volume -, Number -, 2017
Table 3. Peritoneal carcinomatosis in 41 patients who were diagnosed by ultrasound, CT and histopathology after surgery Site of implant
Pathology and surgery
No. of patients diagnosed by ultrasound (detection rate)
No. of patients diagnosed by CT scan (detection rate)
McNemar test (p value)
k index
p
Pelvic Omentum Diaphragm surface Lateral peritoneum Bowel surface Mesenterium Hepatic surface Splenic surface
39 38 32 26 25 19 12 8
36 (92.3%) 28 (73.7%) 7 (21.9%) 10 (38.5%) 16 (64.0%) 4 (21.1%) 10 (83.3%) 7 (87.5%)
17 (43.6%) 34 (89.5%) 10 (31.2%) 16 (61.5%) 4 (16.0%) 11 (57.9%) 7 (58.3%) 5 (62.5%)
0.000 0.146 0.508 0.109 0.002 0.065 0.453 0.500
0.179 0.229 0.338 0.450 0.171 0.144 0.485 0.806
0.045 0.112 0.027 0.002 0.120 0.271 0.001 0.000
CT 5 computed tomography.
In our study, for implants ,1 cm on hepatic surfaces and pelvis, the detection rate with US (63.6%) was higher than that with CT (27.3%), highlighting the importance of US as a complement to CT. Fischerova (2011) also reported that US was helpful in selecting patients with advanced OC for explorative laparotomies, with a high rate of successful optimal primary debulking. We also tried to summarize some guidelines for detection among sites examined in this study. Thickened and rigid omentum, known as ‘‘omental cake’’ (Fig. 2a,b), and nodular or sheet-like lesions attached to the peritoneum are all signs of abdominal and pelvic peritoneal involvement (Coakley et al. 2002; Conte et al. 1994; Savelli et al. 2005; Sehouli et al. 2009) (Figs. 1 and 3–6). We suggest completing the examination with both transabdominal and transvaginal scans and with the patient placed in different positions, such as the supine position and left and right lateral positions, as well as with various sectional views, such as abdominal longitudinal and transverse, intercostal oblique, subcostal oblique and coronal. These various scan positions and sections can reveal a thickened omentum and implants on the liver and spleen surfaces. Our study has some limitations. First, the number of patients involved was too small for subgroup analyses, especially for detection of hepatic and splenic surfaces. In addition, whereas the US exam was performed and evaluated by the same US physician, the CT scan results were reported by seven radiologists; although their reports were based on the standard scanning method and criteria, some bias might still exist. Furthermore, contrast CT scan was used to identify tumor vascularity, whereas contrast-enhanced ultrasound was not used, which may reduce the US detection rate of the metastasis. We hope to develop another study with a larger sample size, which can supply more reliable results on this issue, and also to decrease possible bias by involving more radiologists and sonographers to standardize the scanning method and reporting system. In summary, on the basis of our analysis, US detection rates for peritoneal carcinomatosis in pelvis and on bowel surfaces are higher than CT rates in primary OC.
We also observed that US and CT are moderately consistent in detecting lesions on hepatic and splenic surfaces, with both modes having high detection rates. Standard US scanning is a useful, feasible and economic complement to CT scans in detecting PC in patients suspected of having primary OC before surgery. Acknowledgments—The authors have no conflict of interest that could be perceived as prejudicing the impartiality of the research reported. We thank the Departments of Radiology, and Obstetrics and Gynecology at the Peking Union Medical College Hospital for their valuable help.
REFERENCES Coakley FV, Choi PH, Gougoutas CA, Pothuri B, Venkatraman E, Chi D, Bergman A, Hricak H. Peritoneal metastases: Detection with spiral CT in patients with ovarian cancer. Radiology 2002;223:495–499. Conte M, Guariglia L, Benedetti Panici P, Scambia G, Matonti G, Mancuso S. Preoperative ultrasound assessment of omental spread in ovarian cancer. Gynecol Obstet Invest 1994;38:213–216. Fischerova D. Ultrasound scanning of the pelvis and abdomen for staging of gynecological tumors: A review. Ultrasound Obstet Gynecol 2011;38:246–266. Forstner R, Sala E, Kinkel K, Spencer JA, European Society of Urogenital Radiology. ESUR guidelines: Ovarian cancer staging and follow-up. Eur Radiol 2010;20:2773–2780. Hanbidge AE, Lynch D, Wilson SR. US of the peritoneum. Radiographics 2003;23:663–685. Henrich W, Fotopoulou C, Fuchs I, Wolf C, Schmider A, Denkert C, Lichtenegger W, Sehouli J. Value of preoperative transvaginal sonography (TVS) in the description of tumor pattern in ovarian cancer patients: Results of a prospective study. Anticancer Res 2007;27:4289–4294. Menon U, Gentry-Maharaj A, Hallett R, Ryan A, Burnell M, Sharma A, Lewis S, Davies S, Philpott S, Lopes A, Godfrey K, Oram D, Herod J, Williamson K, Seif MW, Scott I, Mould T, Woolas R, Murdoch J, Dobbs S, Amso NN, Leeson S, Cruickshank D, McGuire A, Campbell S, Fallowfield L, Singh N, Dawnay A, Skates SJ, Parmar M, Jacobs I. Sensitivity and specificity of multimodal and ultrasound screening for ovarian cancer, and stage distribution of detected cancers: Results of the prevalence screen of the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS). Lancet Oncol 2009;10:327–340. Nougaret S, Addley HC, Colombo PE, Fujii S, Al Sharif SS, Tirumani SH, Jardon K, Sala E, Reinhold C. Ovarian carcinomatosis: How the radiologist can help plan the surgical approach. Radiographics 2012;32:1775–1800. Permuth-Wey J, Sellers TA. Epidemiology of ovarian cancer. Methods Mol Biol 2009;472:413–437. Savelli L, De Iaco P, Ceccaroni M, Ghi T, Ceccarini M, Seracchioli R, Cacciatore B. Transvaginal sonographic features of peritoneal carcinomatosis. Ultrasound Obstet Gynecol 2005;26:552–557. Sehouli J, Senyuva F, Fotopoulou C, Neumann U, Denkert C, Werner L, G€ulten OO. Intra-abdominal tumor dissemination pattern and
US detection of peritoneal carcinomatosis d Z. QI et al. surgical outcome in 214 patients with primary ovarian cancer. J Surg Oncol 2009;99:424–427. Tempany CM, Zou KH, Silverman SG, Brown DL, Kurtz AB, McNeil BJ. Staging of advanced ovarian cancer: Comparison of imaging modalities—Report from the Radiological Diagnostic Oncology Group. Radiology 2000;215:761–767.
9
Valentin L. Prospective cross-validation of Doppler ultrasound examination and gray-scale ultrasound imaging for discrimination of benign and malignant pelvic masses. Ultrasound Obstet Gynecol 1999;14:273–283. Woodward PJ, Hosseinzadeh K, Saenger JS. From the archives of the AFIP: Radiologic staging of ovarian carcinoma with pathologic correlation. Radiographics 2004;4:225–246.
http://dx.doi.org/10.1016/j.ultrasmedbio.2017.02.016
d
Original Contribution PERITONEAL CARCINOMATOSIS IN PRIMARY OVARIAN CANCER: ULTRASOUND DETECTION AND COMPARISON WITH COMPUTED TOMOGRAPHY ZHENHONG QI, YIXIU ZHANG, QING DAI, YU XIA, and YUXIN JIANG Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China (Received 2 July 2016; revised 14 February 2017; in final form 21 February 2017)
Abstract—We retrospectively compared detection rates and consistency for diagnosis of peritoneal carcinomatosis (PC) of primary ovarian cancer (OC) between ultrasound (US) and computed tomography (CT) scans in 41 patients whose PC of OC (stages IIC–IV) had been diagnosed by histopathology findings. Compared with CT detection rates, those for US were significantly higher for metastases to the pelvic area (92.3% vs. 43.6%, p , 0.001) and bowel surface (64.0% vs. 16.0%, p 5 0.002); however, they did not significantly differ for other sites: omentum, diaphragm, lateral peritoneum, mesenteric, hepatic and splenic surfaces. Diagnostic consistency between US and CT scans were fair to moderate for splenic (k 5 0.806), hepatic (k 5 0.485), lateral peritoneum (k 5 0.450) and diaphragm (k 5 0.338) surfaces, but poorly consistent for other parts (k 5 0.144–0.229). In summary, US can complement CT scans, especially for detecting PC of primary OC metastases in pelvic and bowel surfaces. (E-mail: [email protected]) Ó 2017 World Federation for Ultrasound in Medicine & Biology. Key Words: Ovarian cancer, Peritoneum, Peritoneal carcinomatosis, Ultrasound, Computed tomography.
even at the bedside, and is reportedly .90% accurate in detecting benign and malignant intra-abdominal ovarian masses (Valentin 1999). However, very few data are available in existing literature regarding the involvement of PC in primary OC. This study compared detection rates by US and computed tomography (CT) scans for PC and evaluated their diagnostic consistency.
INTRODUCTION Ovarian cancer (OC) is the sixth most commonly diagnosed cancer among women worldwide and causes more deaths per year than any other cancer of the female reproductive system (Permuth-Wey and Sellers 2009). It is one of the most occult cancers, as 70% of tumors are diagnosed in the late stages of disease (FIGO III and IV). In addition, 70% of OC patients present with involvement of the peritoneum (Nougaret et al. 2012; Sehouli et al. 2009; Woodward et al. 2004). Aggressive surgery and chemotherapy are often combined to extend the survival of patients with OC; however, prognosis is still poor because of late-stage disease at diagnosis. The presence of a residual lesion after initial surgery is an important prognostic factor in patients with advanced OC. Therefore, detecting and detailing involvement of peritoneal carcinomatosis (PC) in patients suspected of having primary OC is critical. Ultrasound (US) is the least expensive technique; it is quick and simple to perform,
METHODS Study population We retrospectively enrolled into this study 41 women who had been diagnosed with primary OC with PC by histopathology findings from January 2008 to December 2015. All 41 women had been hospitalized in the Department of Obstetrics and Gynecology in Peking Union Medical College Hospital. They were evaluated with US examinations and CT scans independently, and then underwent optimal or suboptimal primary debulking surgery within 10 d. All ultrasound examinations were performed and evaluated by the same senior ultrasound physician; CT scan results were reported by seven radiologists. Surgical findings and pathologic diagnoses were regarded as gold standards for evaluating US and CT scan results. Of the 41 patients, 14 were pre-menopausal and 27 were
Address correspondence to: Yuxin Jiang, Department of Ultrasound, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, No 1 Shuaifuyuan, Wangfujing, Beijing, China. E-mail: [email protected] 1
2
Ultrasound in Medicine and Biology
post-menopausal. None of them had taken estrogen during their lifetime. Only two patients had reported family histories of OC. This retrospective study was approved by the ethics committee of Peking Union Medical Hospital, which waived the need for written informed consent from the patients. All records data were de-identified and analyzed anonymously.
US examination The US examinations were performed using a LOGIQ 9 (GE Healthcare, Milwaukee, WI, USA) or an IU22 or HD11 (Philips, Eindhoven, Netherlands) with 9- to 3- or 8- to 4-MHz vaginal probes for transvaginal ultrasound imaging and 5- to 2-MHz probes for transabdominal ultrasound imaging. Transvaginal scans of the pelvic peritoneum in longitudinal and transverse views were performed first, followed by trans-abdominal scans. Scans focused on omentum, hepatic and splenic surfaces, diaphragm, lateral peritoneum and bowel surface. When ascites was observed, mesenteric and intestinal surfaces were carefully examined. Size, echo and color Doppler flow imaging (CDFI) of pelvic and abdominal peritoneal implants were recorded, along with the boundaries and relationships with lesions in surrounding tissues. Imaging was saved simultaneously. Detected PC in the abdominal cavity was carefully recorded with respect to the area involved, which included bowel surface, hepatic and splenic surfaces, diaphragm, lateral peritoneum, omentum, mesenteric surface and pelvis. Our method of examining patients for PC using ultrasound was as follows: 1. Pelvic implants: Transvaginal US was used to examine the pouch of Douglas, pelvic parietal peritoneum and surfaces of visceral peritoneum; lesions could be detected on of bowel, bladder, uterine and pelvic parietal peritoneum surfaces. 2. Omentum: Abdominal US of the upper and lower abdomen in the longitudinal or transverse direction was used. Lesions were usually located between the front abdominal wall and the bowel. 3. Hepatic surface: The patient was placed in the supine position. First, a transverse scan of the left liver at the level of the subxiphoid was performed. Then, an oblique scan was performed along the right costal margin of the liver, from left to right, followed by an intercostal oblique and coronal scan of the liver surface in the left lateral position. 4. Diaphragm surface: The right costal margin was scanned at an oblique angle, as was an intercostal oblique and coronal section in the left lateral position. 5. Splenic surface: Patient was moved to the right lateral position. The spleen can be examined via intercostal
Volume -, Number -, 2017
Table 1. CA-125 levels and ascites volumes CA-125 level (IU/mL)
n
Ascites volume (mL)
n
,100 100–200 201–500 501–1000 1001–2000 2001–5000 .5000 Total
2 4 10 7 6 7 5 41
Non-ascites #500 .500, #1000 .1000, #1500 .1500, #2000 .2000
4 12 8 9 5 3 41
or left coronal scanning, moving the probe backward and forward. 6. Mesenteric and bowel surfaces: Patient was placed in the supine position. Hyper-echoic bowel could be observed in all four quadrants. Operator looked for implants on mesenteric and bowel surfaces while scanning. 7. Lateral peritoneum: Patient was placed in the right or left lateral position. The left or right lateral abdomen, adjacent to the side of the abdominal wall, was scanned from top to bottom. CT examination Computed tomography scans were performed using GE 64-slice helical scanners (GE Medical Systems, Milwaukee, WI, USA). Images were reconstructed at 7-mm intervals. All patients received the intravenous contrast medium Ultravist (Bayer, Berlin, Germany) through a bolus injection with a high-pressure syringe through the forearm vessel at 3 mL/s. After injection of the contrast agent, arterial-phase, portal-phase and delayed-phase images were obtained after 30, 60, and 200 s, respectively. The scan range was from the top of the diaphragm to the pubic symphysis. Each physician recorded a diagnosis based on each CT scan. Pre-operative descriptions of OC by abdominal pelvic CT included size, morphology and unilateral or bilateral character of ovarian masses; any features of malignancy; uterus, bladder, bowel, pelvic side-wall invasions or implants; ascites in the pelvis or upper abdomen and its Table 2. Histologic findings in patients with persistently abnormal screens (n 5 41) Final pathologic type
N
Borderline serous papillary adenofibroma Endometrioid carcinoma Clear cell carcinoma Sarcoma cancer (malignant mixed tumor m€ullerian) Poorly differentiated neuroendocrine carcinoma Serous papillary adenocarcinoma 1 endometrial carcinoma Mucinous carcinoma Serous papillary adenocarcinoma or cystadenocarcinoma Total
1 3 2 2 1 2 3 27 41
US detection of peritoneal carcinomatosis d Z. QI et al.
3
Fig. 1. Transvaginal ultrasound findings of pelvic implant lesions by ultrasound. (a) Bilateral stage IIIC ovarian clear-cell carcinoma in a 42-y-old woman (MASS 5 ovarian tumor). Arrows point to peritoneal implants ,5 mm. Presence of pelvic fluid facilitates detection of these small metastases. (b) Scan of a 69-y-old woman with bilateral stage IIIC ovarian sarcoma reveals a peritoneal metastasis on the bowel surface. Although ascitic fluid is absent, there is no gas in the intestines; the small implant (0.8 3 0.6 cm) is visible (1). (c) Right stage IIIC ovarian endometrioid carcinoma in a 69-y-old woman. Anechoic ascitic fluid clearly outlines the small implant (0.7 3 0.8 cm) on the bowel surface (arrow); color Doppler flow imaging reveals abundant blood flow in the lesion. (d) Right stage IIIC ovarian serous papillary adenocarcinoma in a 59-y-old woman. Implant (2.2 3 2.5 cm) on rectal surface was observed on ultrasound (1) and computed tomography (arrow). (e) Bilateral stage IIIC ovarian serous papillary adenocarcinoma in an 82-y-old woman. Implant (0.9 cm) appears sheet-like and irregular on bladder surface (1). (f) Transvaginal scan of pelvic bowel in a 69-y-old woman with bilateral stage IIIC ovarian serous papillary adenocarcinoma reveals a thickening hypo-echoic implant (1.4 cm) attached to both sides of the bowel (1). BL 5 bladder; BO 5 bowel; CX 5 cervix; UT 5 uterus.
volume; omentum metastases; involvement of small bowel mesentery and surface of bowel; surface or parenchymal hepatic and splenic metastases; implants in para-colic gutters, sites of other peritoneal/serosal implants outside the pelvis; sites of lymph nodes with short-axis diameters .1 cm or suspicious clusters of smaller lymph nodes. Lymph node and hepatic and splenic parenchyma metastases were not included in this study. Statistical analysis Categorical variables were described using frequencies and percentages. We calculated detection rates of US and CT scans at each site. The McNemar test was applied to compare the statistical differences in detection rates between US and CT scans, with surgical findings and histopathology diagnoses serving as gold standards. p , 0.05 was considered to indicate significance. We used k tests to evaluate the consistency of US and CT in diagnosing peritoneal metastasis of ovarian cancer (k . 0.81:
very good consistency, 0.61–0.80: good, 0.41–0.60: moderate, 0.21–0.40: fair, and ,0.20: poor). Statistical analyses were performed using the Statistical Package for the Social Sciences (Version 23.0, IBM, Armonk, NY, USA). RESULTS Demographic and clinical information The median age of the 41 patients was 55 y; their average age was 55.8 6 12.6 y (range: 27–82 y). CA125 levels were elevated for all 41 patients (#500 IU/ mL: n 5 16, .500 IU/mL: n 5 25). Ascites was documented in 37 patients (.500 mL: n 5 25, #500 mL: n 5 12). There were no correlations between CA-125 values and ascites volumes (Table 1). Operative and pathologic results On the basis of the surgical findings, bilateral ovarian lesions were present in 23 patients and unilateral
4
Ultrasound in Medicine and Biology
Volume -, Number -, 2017
Fig. 1. (continued).
Fig. 2. Omental cake and nodular omental metastases. (a) Left stage IIIC ovarian serous papillary adenocarcinoma in a 61-y-old woman. Longitudinal scan of umbilical upper abdomen reveals abundant blood flow in thickened mediumechoic omentum (arrows, OMEN). (b) Bilateral stage IIIC ovarian serous papillary adenocarcinoma in a 50-y-old woman. Thickened, rigid omentum in ascites fluid; color Doppler flow imaging confirmed vessel in the omentum (1: thickened omentum, 1.8 cm). (c,d) Left stage IIIC ovarian serous adenocarcinoma in a 45-y-old woman. Computed tomography (c) and ultrasound (d) images reveal nodular omental metastases (arrows).
US detection of peritoneal carcinomatosis d Z. QI et al.
ovarian lesions in 18 patients, for a total of 64 lesions, of which 12 were solid and 52 were cystic-solid; diameters measured ,5 cm (n 5 28), 5–10 cm (n 5 300) and .10 cm (n 5 6). The FIGO stages of the ovarian tumors ranged from IIC to IV (2 stage IIC, 1 stage IIIB, 35 stage IIIC [the majority stage, at 85.4%] and 3 stage IV). In the 2 patients with stage IIC tumors, one had a sheetlike implant on her peritoneal bladder fold with 300 mL ascites, which was detected during surgery, and the other had a 0.8 3 0.4 3 0.4-cm implant in the pouch of Douglas without ascites; both implants were missed by US and CT. Final pathologic diagnoses are listed in Table 2. US and CT findings In US detection, metastatic lesions appeared as hypoechoic or medium-echoic and nodular or sheet-like, or the omentum was diffusely thickened. Blood flow was detected in the peritoneal metastases through CDFI. US images tastatic lesions in different parts are provided in Figures 1–7. Peritoneal lesions ,5 mm were easily
5
detected in the presence of pelvic fluid by transvaginal ultrasound (Fig. 1a). Lesions were clearly visible on the surface of liver, with the liver serving as the acoustic window (Fig. 3a); especially between the liver and right kidney, lesions ,1 cm were detectable (Fig. 3b). Because the greater omentum is adjacent to the frontal abdominal walls, metastatic lesions between the abdominal wall and bowel can easily be detected as they are not affected by bowel gas (Fig. 2a–d). Although lesions on the mesenteric and bowel surfaces may not be easily detected by transabdominal US because of bowel gas or large amounts of ascites, we noticed that a moderate volume of ascites was helpful in lesion detection (Fig. 6). However, metastases to the diaphragm are difficult to find. In CT scans, metastatic lesions appeared nodular (Figs. 1d, 2c,d and 5a) or local (Fig. 3c,d), or as diffuse thickening in the greater omentum, or linear or nodular enhancement in the omentum. As for US, the lesions were easily detected with the presence of ascites. Sagittal and coronal sections can help detect metastasis in the diaphragm.
Fig. 3. Nodular and sheet-like implants on the liver surface. (a). Bilateral stage IV ovarian serous papillary adenocarcinoma in a 55-y-old woman. With the patient in the supine position, oblique left liver scan along the right costal margin reveals a hypo-echoic implant (1.7 3 0.9 cm, arrows) on the surface of the left liver lobe. (b) Bilateral stage IIIC ovarian serous papillary carcinoma in a 59-y-old woman. Small hypo-echoic metastasis (0.8 cm, arrow) is clearly visible on the surface of right liver lobe, with the liver and right kidney serving as the acoustic window. (c,d) Bilateral stage IIIC ovarian serous cystadenocarcinoma in a 55-y-old woman. Computed tomography axis (c) and ultrasonic coronal plane (d) images reveal sheet-like metastases on the liver surface (arrows). RK 5 right kidney.
6
Ultrasound in Medicine and Biology
Fig. 4. Right stage IV ovarian serous papillary cystadenocarcinoma in a 62-y-old woman. Intercostal oblique scan reveals a 0.6-cm hypo-echoic sheet-like lesion on the diaphragm surface and partial indistinct borders with the liver (1, metastasis).
Comparison of US and CT Peritoneal carcinomatosis was diagnosed in 39 patients and misdiagnosed in 2 patients by US and CT; the detection rates with US and CTare listed in Table 3. Respec-
Volume -, Number -, 2017
tive US and CT detection rates for PC were 92.3% versus 43.6% in the pelvis; 73.7% versus 89.5% in the omentum; 21.9% versus 31.2% on diaphragm surfaces; 38.5% versus 61.5% in the lateral peritoneum; 64.0% versus 16.0% in the bowel; mesenteric: 21.1% versus 57.9% in the mesentery; hepatic: 83.3% versus 58.3% in the liver; and 87.5% versus 62.5% in the spleen. There were significant differences in detecting pelvic (p , 0.001) and bowel surface (p 5 0.002) masses, but not masses in other areas. Diagnostic consistency between US and CT scans was fair to moderate in splenic (k 5 0.806), hepatic (k 5 0.485), lateral peritoneum (k 5 0.450) and diaphragm (k 5 0.338) surfaces. Diagnostic consistency was poor in omental (k 5 0.229), pelvic (k 5 0.179), bowel (k 5 0.171) and mesenteric (k 5 0.144) surfaces. Among omental lesions, 9 cases were detected only by CT and 3 only by US, whereas the detection rate for omental lesions increased to 97.4% (37/38) when US and CT scans were combined. Nineteen pelvic lesions were detected only by US, and the 17 pelvic implants detected by CT were all covered by US. In addition, among the 10 metastatic lesions ,1 cm on the hepatic surface, bowel surface and
Fig. 5. Nodular and sheet-like implants on the spleen surface. (a) Same patient as in Figure 3c and d. Intercostal ultrasound oblique scan reveals a medium-echoic lesion (2.8 3 1.5 cm, arrows) of the splenic hilum. Computed tomography axis image reveals a nodular lesion of the splenic hilum. (b) Right stage IIIC ovarian serous papillary adenocarcinoma in a 62-y-old woman. Coronal scan reveals a hypo-echoic sheet-like lesion (arrows) on the spleen surface. M 5 metastasis; SP 5 spleen.
US detection of peritoneal carcinomatosis d Z. QI et al.
Fig. 6. Bilateral ovarian serous cystadenocarcinoma in a 51-y-old woman. Right lower abdominal scan reveals a 1.5 3 1.2-cm hypoechoic mesenteric implant (M). AS 5 ascites.
pelvis (Figs 1a–c and 3b), 7 (63.6%) were detected by US and 3 (27.3%) were detected by CT.
DISCUSSION Reportedly, US is the most commonly used screening method, and was as sensitive as CA-125 level assessment in the largest multicenter study for OC screening (Menon et al. 2009). However, few reports and data are available describing how PC can be effectively detected via either trans-vaginal or transabdominal ultrasound (Savelli et al. 2005). In addition, no routine standard screening for PC in patients with suspected primary OC is available at our hospital. Our study supported the important role of US in PC diagnosis, including summarizing the US standard scanning guideline in case enhancing the detection rate of PC.
Fig. 7. Same patient as in Figure 6. Scan of lower right abdomen of patient in the left lateral position reveals a 3.8 3 1.7-cm hypo-echoic lesion on the lower right abdominal wall (1: lesion).
7
Diagnosis of peritoneal metastasis requires both a thorough and meticulous examination of all peritoneal surfaces (Hanbidge et al. 2003) and an operator who is familiar with the routes of OC metastasis, to detect specific sites of metastasis after a standard scan. The US detection rates in our study were significantly higher than those for CT for both pelvic and bowel surface implants. Our final results indicated that the US detection rate for pelvic metastases was 92.3% (36/39), similar to the 88% sensitivity reported by Savelli et al. (2005), but our CT detection rate was 43.6% (17/39), which was significantly lower than that in Savelli et al. In our study, 64% (16/25) of implants on bowel surfaces were detected, mainly through transvaginal ultrasound (Fig. 1b,c,f). The CT detection rate for bowel surfaces in our study was low at 16%, which was similar to the 14% reported by Forstner et al. (2010). Ascites was detected in 90.2% of patients (37/41) with PC, possibly because of improved visualization of the peritoneal layers, allowing better detection of pelvic and bowel implants (Henrich et al. 2007). In our study, results for US detection depended both on the patient’s condition and on her implant sites. Generally, it was easier to detect implants in the presence of moderate ascites or in thin patients. The presence of ascites was particularly beneficial in observing pelvic and parietal peritoneal lesions and the uterus–rectum fossa while avoiding ovarian tumors. In contrast, in obese patients or patients with intestinal gas, involvement of mesenteric and bowel surfaces made detection more difficult. We found US and CT scans to be moderately consistent in detecting hepatic and splenic surfaces; both had high detection rates. The liver and spleen form the acoustic window; particularly between the liver and right kidney, implants were easily detected, even those ,1.0 cm (Fig. 3b). The omentum is close to the anterior abdominal wall in a very superficial location, and without the influence of intestinal gas, masses located on the omentum and thickening of the omentum were more easily observed with US. Interestingly, although our detection rates for the omentum were high for both imaging modes (US: 73.7%, CT: 89.5%), their consistency was poor, which means that different cases of omentum metastasis were detected by US and CT, respectively. The detection rate increased to 97.4% when we combined the two methods. Tempany et al. (2000) reported an overall sensitivity of only 69% for US in detecting peritoneal metastases in patients with stage III–IV OC, significantly less than 92% for CT; therefore, routine use of US for OC staging is not recommended. Coakley et al. (2002) also reported an overall high sensitivity (85%–93%) for use of CT for ovarian cancer staging, but the detection rate for small peritoneal lesions was poor, with a sensitivity of only 25%–50% for lesions with a maximal diameter ,1 cm.
8
Ultrasound in Medicine and Biology
Volume -, Number -, 2017
Table 3. Peritoneal carcinomatosis in 41 patients who were diagnosed by ultrasound, CT and histopathology after surgery Site of implant
Pathology and surgery
No. of patients diagnosed by ultrasound (detection rate)
No. of patients diagnosed by CT scan (detection rate)
McNemar test (p value)
k index
p
Pelvic Omentum Diaphragm surface Lateral peritoneum Bowel surface Mesenterium Hepatic surface Splenic surface
39 38 32 26 25 19 12 8
36 (92.3%) 28 (73.7%) 7 (21.9%) 10 (38.5%) 16 (64.0%) 4 (21.1%) 10 (83.3%) 7 (87.5%)
17 (43.6%) 34 (89.5%) 10 (31.2%) 16 (61.5%) 4 (16.0%) 11 (57.9%) 7 (58.3%) 5 (62.5%)
0.000 0.146 0.508 0.109 0.002 0.065 0.453 0.500
0.179 0.229 0.338 0.450 0.171 0.144 0.485 0.806
0.045 0.112 0.027 0.002 0.120 0.271 0.001 0.000
CT 5 computed tomography.
In our study, for implants ,1 cm on hepatic surfaces and pelvis, the detection rate with US (63.6%) was higher than that with CT (27.3%), highlighting the importance of US as a complement to CT. Fischerova (2011) also reported that US was helpful in selecting patients with advanced OC for explorative laparotomies, with a high rate of successful optimal primary debulking. We also tried to summarize some guidelines for detection among sites examined in this study. Thickened and rigid omentum, known as ‘‘omental cake’’ (Fig. 2a,b), and nodular or sheet-like lesions attached to the peritoneum are all signs of abdominal and pelvic peritoneal involvement (Coakley et al. 2002; Conte et al. 1994; Savelli et al. 2005; Sehouli et al. 2009) (Figs. 1 and 3–6). We suggest completing the examination with both transabdominal and transvaginal scans and with the patient placed in different positions, such as the supine position and left and right lateral positions, as well as with various sectional views, such as abdominal longitudinal and transverse, intercostal oblique, subcostal oblique and coronal. These various scan positions and sections can reveal a thickened omentum and implants on the liver and spleen surfaces. Our study has some limitations. First, the number of patients involved was too small for subgroup analyses, especially for detection of hepatic and splenic surfaces. In addition, whereas the US exam was performed and evaluated by the same US physician, the CT scan results were reported by seven radiologists; although their reports were based on the standard scanning method and criteria, some bias might still exist. Furthermore, contrast CT scan was used to identify tumor vascularity, whereas contrast-enhanced ultrasound was not used, which may reduce the US detection rate of the metastasis. We hope to develop another study with a larger sample size, which can supply more reliable results on this issue, and also to decrease possible bias by involving more radiologists and sonographers to standardize the scanning method and reporting system. In summary, on the basis of our analysis, US detection rates for peritoneal carcinomatosis in pelvis and on bowel surfaces are higher than CT rates in primary OC.
We also observed that US and CT are moderately consistent in detecting lesions on hepatic and splenic surfaces, with both modes having high detection rates. Standard US scanning is a useful, feasible and economic complement to CT scans in detecting PC in patients suspected of having primary OC before surgery. Acknowledgments—The authors have no conflict of interest that could be perceived as prejudicing the impartiality of the research reported. We thank the Departments of Radiology, and Obstetrics and Gynecology at the Peking Union Medical College Hospital for their valuable help.
REFERENCES Coakley FV, Choi PH, Gougoutas CA, Pothuri B, Venkatraman E, Chi D, Bergman A, Hricak H. Peritoneal metastases: Detection with spiral CT in patients with ovarian cancer. Radiology 2002;223:495–499. Conte M, Guariglia L, Benedetti Panici P, Scambia G, Matonti G, Mancuso S. Preoperative ultrasound assessment of omental spread in ovarian cancer. Gynecol Obstet Invest 1994;38:213–216. Fischerova D. Ultrasound scanning of the pelvis and abdomen for staging of gynecological tumors: A review. Ultrasound Obstet Gynecol 2011;38:246–266. Forstner R, Sala E, Kinkel K, Spencer JA, European Society of Urogenital Radiology. ESUR guidelines: Ovarian cancer staging and follow-up. Eur Radiol 2010;20:2773–2780. Hanbidge AE, Lynch D, Wilson SR. US of the peritoneum. Radiographics 2003;23:663–685. Henrich W, Fotopoulou C, Fuchs I, Wolf C, Schmider A, Denkert C, Lichtenegger W, Sehouli J. Value of preoperative transvaginal sonography (TVS) in the description of tumor pattern in ovarian cancer patients: Results of a prospective study. Anticancer Res 2007;27:4289–4294. Menon U, Gentry-Maharaj A, Hallett R, Ryan A, Burnell M, Sharma A, Lewis S, Davies S, Philpott S, Lopes A, Godfrey K, Oram D, Herod J, Williamson K, Seif MW, Scott I, Mould T, Woolas R, Murdoch J, Dobbs S, Amso NN, Leeson S, Cruickshank D, McGuire A, Campbell S, Fallowfield L, Singh N, Dawnay A, Skates SJ, Parmar M, Jacobs I. Sensitivity and specificity of multimodal and ultrasound screening for ovarian cancer, and stage distribution of detected cancers: Results of the prevalence screen of the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS). Lancet Oncol 2009;10:327–340. Nougaret S, Addley HC, Colombo PE, Fujii S, Al Sharif SS, Tirumani SH, Jardon K, Sala E, Reinhold C. Ovarian carcinomatosis: How the radiologist can help plan the surgical approach. Radiographics 2012;32:1775–1800. Permuth-Wey J, Sellers TA. Epidemiology of ovarian cancer. Methods Mol Biol 2009;472:413–437. Savelli L, De Iaco P, Ceccaroni M, Ghi T, Ceccarini M, Seracchioli R, Cacciatore B. Transvaginal sonographic features of peritoneal carcinomatosis. Ultrasound Obstet Gynecol 2005;26:552–557. Sehouli J, Senyuva F, Fotopoulou C, Neumann U, Denkert C, Werner L, G€ulten OO. Intra-abdominal tumor dissemination pattern and
US detection of peritoneal carcinomatosis d Z. QI et al. surgical outcome in 214 patients with primary ovarian cancer. J Surg Oncol 2009;99:424–427. Tempany CM, Zou KH, Silverman SG, Brown DL, Kurtz AB, McNeil BJ. Staging of advanced ovarian cancer: Comparison of imaging modalities—Report from the Radiological Diagnostic Oncology Group. Radiology 2000;215:761–767.
9
Valentin L. Prospective cross-validation of Doppler ultrasound examination and gray-scale ultrasound imaging for discrimination of benign and malignant pelvic masses. Ultrasound Obstet Gynecol 1999;14:273–283. Woodward PJ, Hosseinzadeh K, Saenger JS. From the archives of the AFIP: Radiologic staging of ovarian carcinoma with pathologic correlation. Radiographics 2004;4:225–246.