Role of Diffusion weighted MR Imaging in the Evaluation of Ovarian Tumors

Size: px

Start display at page:

Download "Role of Diffusion weighted MR Imaging in the Evaluation of Ovarian Tumors"

Damon Oliver
6 years ago
Views:

1 Role of Diffusion weighted MR Imaging in the Evaluation of Ovarian Tumors Thesis Submitted For Partial Fulfillment of the M.Sc. Degree In Radio diagnosis By Safaa Ibrahim Saif El Nasr Supervised by Dr. Soha Talaat Hamed Prof. of Radio diagnosis Dr. Fatma Mohamed Awad Assist. Prof. of Radio diagnosis Cairo University 2012

2 بسماالله الرحمن الرحيم س ب ح ان ك لا ع ل م ل ن ا إ لا م ا ع ل م ت ن ا إ ن ك أ نت ال ع ل يم ال ح ك يم " "ق ال وا صدق االله العظيم سورة البقرة الا ية (٣٢)

3 Acknowledgment First of all I would like to thank ALLAH for giving me the power to complete this work, may he be generous on me and give me the knowledge to help others. Words can never express my feelings, respect and gratitude to Prof. Dr. Soha Talaat Hamed Professor of Radiodiagnosis, Faculty of medicine, Cairo-University, for her continues care, support and for her invaluable guidance, constructive criticism in supervising this work. Also special thanks for Dr. Fatma Mohamed Awad Assist. Prof. of Radio diagnosis, Faculty of medicine, Cairo-University for her effort and support in this thesis. I wish to express my deep gratitude to Dr. Sahar Mansour Lecturer of Radiodiagnosis, for her help, concern and being always giving and caring. It is my pleasure to express my deep appreciation to my professors and colleagues in the Radiology department, Cairo University. My especial thanks and deep appreciation to Dr. Reham Osama Assist. Lecturer for her patience and unlimited kind support. Also special thanks for women s imaging unit nurses and all MRI unit operators. Last, but not least, I would like to express my respect, appreciation and thanks to my Parents and my Brothers for their continues praying, everlasting love and care.

4 Contents Contents Abstract List of abbreviations List of tables List of figures Introduction Aim of work Review of literature: Anatomy of the ovary. Pathology of ovarian tumors Technique of pelvic MRI Diffusion-Weighted Imaging (DWI) MRI manifestations of ovarian tumors Appearance of ovarian tumors in DWI. Management of ovarian tumors. Page I II III IV V VI Patients & methods 64 Results 70 Case presentation 80 Discussion 92 Summary and conclusion 98 References 102 Arabic summary

5 Abstract Ovarian cancer is a leading cause of death among women. Preoperative tissue characterization can help the surgeon to plan for adequate procedures. DWI is one of the evolving functional MR imaging. When added and interpreted together with the conventional MR imaging, the specificity and accuracy of conventional MR imaging findings have shown to be increased. Keywords Ovarian cancer - Difussion Weighted Imaging - functional imaging. I

6 List of Abbreviations ADC BOT DCE MRI DWI FOV FSE Gd HCG MRS NPV PET/CT PPV RF SE SPAIR : Apparent diffusion coefficient : Borderline ovarian tumors : Dynamic Contrast Enhanced MRI : Diffusion weighted imaging : Field of view : Fast spin echo : gadolinium : Human chorionic gonadotropin : Magnetic Resonance Spectroscopy : Negative predictive value : Positron emission tomography computed tomography : Positive predictive value : Radiofrequency : Spin echo. : Spectral Adiabatic Inversion Recovery II

7 List of tables Figure Title Epithelial tumors of the ovary Sex cord stromal tumors of the ovary Germ cell tumor of the ovary. TNM classification of ovarian tumors. Different MR sequences for evaluating the adnexa One of the protocols used for applying the DWI for the pelvis. Interpretation of DWI Findings. Criteria to differentiate benign versus malignant epithelial tumors. Criteria to differentiate between serous and mucinous tumors: Metabolites Detected with Proton MR Spectroscopy Different complaints and their percentage. the maximum diameters of the tumors in cm. Correlation between the numbers of cases diagnosed as benign or malignant by the conventional MR imaging, DWI and their pathological diagnosis. Difference between the MRI and the DWI in Sensitivity, Specificity, PPV, NPV and accuracy. Comparison between the ADC values of solid component between benign and malignant tumors. Comparison between the ADC values of cystic component between benign and malignant tumors. Page III

8 List of figures Figure Title Normal anatomy of the female reproductive system. Normal zonal anatomy in a premenopausal woman Schematic drawing showing sites of origin of ovarian cancer. Diagram showing diffusion of water molecules Schematic illustrates the effect of a diffusion-weighted sequence on water molecules serous cystadenoma of left ovary Mucinous cystadenoma of the ovary Borderline papillary serous tumor of the ovary Endometriod carcinoma with concomitant endometrial carcinoma Clear cell carcinoma of the ovary Brenner tumor of the ovary Granulosa cell tumor of the ovary Fibrothecoma of the ovary Sertoli-Leydig cell tumor of the ovary. Sclerosing stromal tumor of the ovary. Steroid cell tumor of the ovary. Mature cystic teratoma Immature teratoma Dysgerminoma of the ovary. Primary ovarian choriocarcinoma. Bilateral metastasis to the ovaries (Krukenberg tumors) from gastric carcinoma. Collision tumor of the ovary. Algorithm for diagnosis of ovarian mass Signal intensity-time curves of benign, borderline, and invasive ovarian tumors. Proton MRS in cases of malignant and benign ovarian tumor DW appearance of malignant ovarian tumor DW appearance of benign ovarian tumor DW appearance of peritoneal deposit DW appearance of lymph node in cases of recurrent ovarian cancer DW appearance of post-operative changes Conservative or radical surgery for treatment of BOTs Algorithm for imaging of a suspected ovarian mass. Chart showing different pathological types of benign tumors. Chart showing different pathological types of malignant tumors. Chart showing the different compositions of the tumors and their percent. Algorithm showing Indications for different pelvic MRI sequences for characterization of ovarian masses (detected by US). Page IV

9 Introduction Introduction Functional imaging is becoming increasingly important in the evaluation of cancer patients because of the limitations of morphologic imaging, particularly in the assessment of response to therapy. Recent technical advances allow the use of diffusion MR imaging in abdominal and pelvic applications after it has been established as a useful functional imaging tool in neurologic applications for a number of years. (Whittaker et al, 2009) This unique noninvasive modality has demonstrated the capacity to help discriminate between benign and malignant lesions, increase the contrast between lesions and surrounding tissues, and improve the detection and delineation of peritoneal implants at both initial staging and follow-up. Moreover, diffusion-weighted imaging provides quantitative information about tissue cellularity that may be used to distinguish viable tumors from treatment-related changes (Kyriazi et al, 2010) When diffusion-weighted MR imaging is used in gynecologic applications, cancers have shown lower apparent diffusion coefficient (ADC) values. Increasing ADC values is noted in carcinomas responding to radiation therapy, so it can be used as a biomarker for treatment response, and in the evaluation of recurrence, discriminating localized from multifocal disease which is a critical factor in opting for secondary cytoreduction (Inada et al, 2008), (McVeigh et al, 2008). As for peritoneal implants from ovarian cancer, the diagnosis represents a privilege for diffusion weighted MR imaging, as the small seeds invaginated within peritoneal reflections, or coating the serosal surface of intestinal loops and solid viscera, are often masked by the similarity of their attenuation or signal intensity to that of adjacent structures using CT or conventional MRI. On diffusion-weighted imaging, malignant deposits on the visceral peritoneum are more conspicuous because of signal suppression from surrounding ascites, bowel contents, and fat (Low et al, 2009) V

10 Introduction Avoiding the potential pitfalls, of the technique, can be accomplished when diffusion weighted images are interpreted in association with anatomic MR images. Increasing familiarity with diffusion coefficient calculation and software manipulation, will allow radiologists to provide new information for the diagnosis of patients with known or suspected gynecologic malignancies (Fujii et al, 2008) Limitations of diffusion weighted MR imaging, in abdomen and pelvis, due to motion and susceptibility artifacts has been overcome by the development of new imaging techniques, particularly novel methods of data acquisition and parallel imaging, allowing much faster data acquisition with fewer artifacts, resulting in significant improvement in image quality in body applications (Qayyum, 2009) V

11 Aim of the work This study aims at the reviewing and emphasizing the role of diffusionweighted MR imaging in the diagnosis of ovarian tumors. VI

12 Anatomy of Ovary 1 Anatomy of the ovary Gross Anatomy The ovaries are almond shaped but may vary in size, position, and appearance, depending on the age and the reproductive activities of the individual (DeLancey et al, 1997). Fig (1.1): showing normal anatomy of the female reproductive system (Tanakaet al, 2004) The normal adult woman ovaries range from cm long, cm thick, and cm wide, with a weight of 3 8 gm. (kleeman, 2007). The ovary is encapsulated by a thin whitishh fibrous Capsule the tunica albuginea. ( Tortora et al, 1998) ). called The ovary can be divided into: Outer cortex which consists of a cellular connective tissue stroma in which the ovarian follicles are embedded. Inner medulla which is composed of loose connective tissue which contains blood vessels and nerves (Kleeman, 2007). 1

13 Anatomy of Ovary 1 The ovary is attached by the mesovarium to the posterior surface of the broad ligament. Further support is given by the ovarian ligament proper and the suspensory ligament of the ovary that is continuous with the broad ligament attaching to the pelvic sidewall and in which the ovarian vessels and lymphatics run (Federle et al, 2006). Blood supply of the ovary The ovarian artery originates from the abdominal aorta, below the level of the renal arteries. The ovarian arteries supply the ovaries, uterine tubes, the upper portion of the body and fundus of the uterus, and anastomose with the uterine arteries (Winkler et al, 1986). The ovarian vein is typically single but may be multiple and will fuse forming single vein which accompanies the ovarian artery along its retroperitoneal course. The right vein drains into the inferior vena cava and the left one drains into the left renal vein (Tukeva et al, 1999). Lymphatic drainage of the ovary The ovarian lymphatics ascend with the ovarian vessels drain almost exclusively into to the para-aortic lymph nodes, close to the origin of the ovarian arteries (Reynolds et al, 2006). Other small branches drain via the broad ligament to the external, internal, and common iliac groups of nodes (Livengood et al, 2006). 2

14 Anatomy of Ovary 1 MR appearance of the ovary (Paul et al, 2004) On T1WIs: The adult ovary appears of intermediate signal intensity with low- signal follicles (unless hemorrhagic). On T2WIs: Multiple high-signal follicles of intensity central stroma Low-signal intensity capsule. varying sizes within low signal Fig (a) Fig (b) Fig..(1.2 ). Normal zonal anatomy in a premenopausal woman. Axial T1-WI (a) and axial T2-WI (b). Both ovaries display multiple small follicles in subcortical location which show intermediate signal on T1-weighted images and very bright signal on the T2- weighted images (Paul et al, 2004) 3

15 Pathology of Ovarian Tumors 2 Pathology of Ovarian Tumors Incidence Ovarian cancer is the fifth leading cause of cancer death among women after (lung, breast, colorectal, and pancreatic cancers) and has a high likelihood of recurrence despite aggressive treatment strategies (Hongju et al, 2011). It is considered the second most common gynecologic malignancy (after cervical Cancer) and most of women are diagnosed in late stages of the disease, (stage III or stage IV cancer), with five year survival reaching 20%. Less than 30% of women are diagnosed with stage I ovarian cancer, and, of these, 90% will survive to five years. (Hippisley-Cox et al, 2012). Epidemiology The diagnosis is primarily in women above the age of 50. Its diagnosis before the age of 30 is rare, even among women affected by hereditary syndromes. After the age of 30, the incidence of ovarian cancer starts to rise (Chu et al, 2008). Risk Factors Positive family history. Genetic syndromes (Folsom et al, 2004). Nulliparity. Childbirth after 35 years. Late menopause. Estrogen replacement therapy for more than five years. Early onset of menses (Sam et al, 2002). 4

Pathology of Ovarian Tumors 2 Histological

1): Serous Tumors: Benign, borderlinee and malignant.

Endometrioid Tumors: Benign, borderline and malignant.

Transitional Cell Tumors (Brennerr tumors) : Benign,

16 Pathology of Ovarian Tumors 2 Histological Classification of Ovarian Tumors (Hulse et al, 2004) Figure (2.1) Schematic drawing showing sites of origin of ovarian cancer. (Hulse et al, 2004) Surface Epithelial Tumors (Table 2. 1): Serous Tumors: Benign, borderlinee and malignant. Mucinous Tumors: Benign, borderline and malignant. Endometrioid Tumors: Benign, borderline and malignant. Clear Cell Tumors: Benign, borderline and malignant. Transitional Cell Tumors (Brennerr tumors) : Benign, borderline and malignant. Mixed and unclassified tumors. Sex Cord Stromal Tumors (Table 2. 2): Granulosa Cell Tumors. Thecoma. Sertoli-Leydig Cell Tumors. Sex Cord Tumor with annular tubules. Steroid Cell Tumors. Germ cell tumors ( Table 2 3): Teratoma: o Benign polyphasic and monophasic such as struma ovarii. o Malignant. Dysgerminoma. Yolk Sac Tumor. Choriocarcinoma. Gonadoblastoma. 5

17 Pathology of Ovarian Tumors 2 Miscellaneous tumors: Wilms tumor. Lymphoma. Small cell tumor with hyper-calcaemia. Metastatic tumors, accounting for about 5% of ovarian malignancies, usually arise from breast, colon, endometrium, stomach and cervical cancers. (Scully, 1975). Serous Mucinous Endometrioid Clear cell Brenner Incidence 20 50% of malignant tumors 10% of malignant tumors 20% of malignant tumors 6% of malignant tumors 1 2% of malignant tumors Aggression Benign 60% Malignant 25% Bilateral 25% of benign 65% of malignant Benign 80% Malignant 10% 5% of benign 20% of malignant Almost always malignant Almost always malignant but 75% Stage 1 Rarely malignant 40% 40% 6% Post-menopausal Age Post-menopausal Postmenopausal Postmenopausal Any age, 50% over 50 years Typical features Predominantly cystic. Malignant lesions have more solid components. Psammoma bodies in 30% Multilocular cysts containing haemorrhage or cellular debris Variable cystic/solid components. Associated with endometrial hyperplasia and carcinoma Usually unilocular cyst with few mural nodules protruding into the lumen Solid and homogenous. Occasionally cystic Usually small ( cm). Extensive calcification Table (2.1) Epithelial tumors: make up to 60 90% of all ovarian tumors; and make up 90% of all malignant ovarian tumors. (Hulse et al, 2004) 6

18 Pathology of Ovarian Tumors 2 Fibroma Thecoma Granulosa cell Sertoli Leydig cell Incidence Rare commonest 5 10% of all ovarian malignancies Aggression Benign Benign Malignant potential increases with size. 5yr survival is 90% Less than 0.2% of all ovarian malignancies Aggressiveness depends on degree of differentiation. 5yr survival is 70% Bilateral Unilateral Unilateral Unilateral in >95% Unilateral Age 4th and 5 th Reproductive years Any age but commonest in Reproductive years decades postmenopausal years Typical features Solid. Associated with pleural effusion (Meigs syndrome) Solid. Produces oestrogen and associated with abnormal vaginal bleeding, endometrial hyperplasia and carcinoma Multi-cystic. May be haemorrhagic or necrotic. Can secrete oestrogen and is associated with abnormal vaginal bleeding, hyperplasia and carcinoma Can be solid or cystic. Synthesise androgens resulting in masculinisation. Table (2.2). Sex cord stromal tumors: 5% of all ovarian malignancies; 85 90%. It is characterized by steroid synthesis. (Hulse et al, 2004) Dysgerminoma mature Teratoma immature Yolk sac Choriocarcinoma Incidence Commonest malignant germ cell. 2 5% of primary ovarian malignancies Aggression Malignant.( 90% 5year Survival) Commonest ovarian neoplasm Benign Bilateral 10% Usually unilateral Less than 1% of ovarian malignancies Malignant (80 90% 5 year Survival) 1% of ovarian tumors. 2nd commonest malignant germ cell tumor Malignant (2 year survival 60 70%) Usually unilatera Unilateral Very rare Malignant. Poor prognosis Unilateral Age 2nd and 3rd decades Any age 10 20yrs Under 30 years Under 20 years Typical features Solid with cystic areas secondary to necrosis or haemorrhage Fat/fluid or Hair /fluid levels. Calcification, bone and teeth Predominantly solid. Contain fetal/ embryonal tissue Cystic or solid. Areas of haemorrhage Usually solid. Often a component of other malignant germ cell tumor Table (2.3) Germ cell tumors: 15 20% of all ovarian malignancies; most common in the 1 stage and 2nd decades; 60 70% Stage I at diagnosis. (Hulse et al, 2004) 7

19 Pathology of Ovarian Tumors 2 Borderline ovarian tumors (BOTs) Borderline ovarian tumors are also known as (ovarian tumors of low malignant potential). They are noninvasive tumors, with long term follow up studies describing overall survival rates up to 99.7% at 5 years and 80% to 90% at 15 years. (Silverberg et al, 2004). BOTs are subtype of epithelial tumors and as such may have serous, mucinous, clear cell, endometrioid, Brenner (transitional cell), or mixed histology. Serous types are the most commonly observed (65%), followed by mucinous BOTs (32%). (Jones et al, 2006). They commonly affect white women in their fourth decade of life, and are mainly localized to the ovary at time of presentation.(jones et al, 2006) The WHO established criteria which include tumors with budding, epithelial stratification, mitotic activity, nuclear atypia, but lacking stromal invasion. Ovarian tumors containing 2 or more of the above features involving 10% or more of the tumor may be classified as borderline tumor. (Seidman et al, 2004). A unique feature of borderline tumors is the noninvasive behavior of extra-ovarian tumor implants (contralateral ovary, omentum, and peritoneum) in the advanced stages of the disease. They behave in a benign fashion with no tendency for invasion, unlike invasive tumor implants from malignant ovarian carcinoma. In a minority of cases, invasive peritoneal implants were observed which can behave as low-grade carcinomas do. (Burkholz et al, 2005). 8

20 Pathology of Ovarian Tumors 2 Table (2.4) shows TNM classification (Grab et al, 2000) Primary tumor (T) TX T0 Primary tumor cannot be assessed. No evidence of primary tumor. T1 I Tumor limited to ovaries (one or both). T1a IA Tumor limited to one ovary; capsule intact, no tumor on ovarian surface. No malignant cells in ascites or peritoneal washing. T1b IB Tumor limited to both ovaries; capsules intact, no tumor on ovarian surface. No malignant cells in ascites or peritoneal washings. T1c IC Tumor limited to one or both ovaries with any of the following: capsule ruptured, tumor on ovarian surface, malignant cells in ascites or peritoneal washings. T2 II Tumor involves one or both ovaries with pelvic extension and/or implants. T2a IIA Extension and/or implants on uterus and/or tube(s). No malignant cells in ascites or peritoneal washings. T2b IIB Extension to and/or implants on other pelvic tissues. No malignant cells in ascites or peritoneal washings T2c IIC Pelvic extension and/or implants (T2a or T2b) with malignant cells in ascites or peritoneal washings. T3 Tumor involves one or both ovaries with microscopically III confirmed peritoneal metastasis outside the pelvis. T3a IIIA Microscopic peritoneal metastasis beyond pelvis (no macroscopic tumor). T3b IIIB Macroscopic peritoneal metastasis beyond pelvis < 2 cm in greatest dimension. T3c IIIC Peritoneal metastasis beyond pelvis >2 cm in greatest dimension and/or regional lymph node metastasis. Regional lymph nodes (N) NX N0 N1 Distant metastasis (M) MX M0 M1 Regional lymph nodes cannot be assessed. No regional lymph node metastasis. Regional lymph node metastasis. Distant metastasis cannot be assessed. No distant metastasis. Distant metastasis (excludes peritoneal metastasis). 9

بسم هللا الرحمن الرحيم. Prof soha Talaat

بسم هللا الرحمن الرحيم. Prof soha Talaat بسم هللا الرحمن الرحيم Ovarian tumors The leading indication for gynecologic surgery. Preoperative characterization of complex solid and cystic adnexal masses is crucial for informing patients about possible