Solid Pelvic Masses Caused by Endometriosis: MR Imaging Features

Size: px

Start display at page:

Download "Solid Pelvic Masses Caused by Endometriosis: MR Imaging Features"

Augustine Smith
6 years ago
Views:

1 X/94/ American Roentgen Ray Society Evan S. Siegelman1 Eric Outwater1 Tsailing Wang2 Donald G. Mitchell1 Received December 17, 1993; accepted after revision March 21, Department of Radiology, Thomas Jefferson University Hospital, Rm. 1096, Main Bldg., 132 S. 10th St., Philadelphia, PA Address correspondence to E. Outwater. 2 Department of Pathology, Thomas Jefferson University Hospital, Philadelphia, PA Solid Pelvic Masses Caused by Endometriosis: MR Imaging eatures OBJECTIVE. Solid fibrotic nodules of endometniosis can simulate penitoneal metastases on imaging studies. To assist with this distinction, we analyzed the MR appearance of solid masses of pelvic endometniosis. MATERIALS AND METHODS. A search of pathologic and surgical records of patients who had pelvic MR imaging with a phased-array multicoil disclosed 13 patients who had endometriosis proved at laparotomy. MR images were reviewed retrospectively by two unblinded radiologists for the signal intensity, size, and enhancement pattern of solid penitoneal masses. Eight solid masses in six patients were detected on MR images: four lesions were in the cul-de-sac, two in the bladder wall, and two In the rectal wall. our of the masses were excised at surgery and one was sampled by surgical biopsy; microscopy showed abundant fibrosis with small clustens of endometniotic glandular tissue. Three masses were inspected at surgery and found to represent dense fibrosis caused by endometniosis. RESULTS. Solid masses of endometniosis ranged in size from 1 to 5 cm. Seven of the eight masses had similar features on MR images: intermediate signal Intensity on Ti-weighted spin-echo images with punctate foci of high signal intensity, low signal intensity on T2-weighted images, and enhancement after administration of contrast material. CONCLUSION. MR imaging shows enhancing solid masses in some patients with endometniosis. These masses have MR features that might be useful in the differentiation between solid foci of endometniosis and penitoneal metastases. AJR 1 994;163: Solid penitoneal masses in the pelvis can suggest metastases from ovarian cancinoma. Solid masses of endometniosis implanted on the penitoneum have been described in several publications [1-4]. We undertook a retrospective review to analyze the MR findings of such lesions to identify findings that might be useful to distinguish solid deposits of endometniosis from metastases from ovarian carcinoma. Materials and Methods A search of pathologic and surgical records of patients who had pelvic MR imaging with a phased-array multicoil disclosed 1 3 patients who had endometriosis proved at laparotomy. The patients were years old. The MR studies of these patients were retrospectively reviewed by two radiologists in conference. The reviewers were not blinded to clinical data. Peritoneal masses that showed enhancement were evaluated for size, margins, and signal intensity on both Ti- and T2-weighted images. Enhancement was further quantified by region-of-interest analysis of the masses compared with pelvic muscle on the same image, which was an equivalent distance from the surface coil. The signal-intensity (SI) measurements were made from spin-echo (SE) Ti -weighted images before and after IV contrast enhancement. A mass-to-muscle enhancement ratio was calculated: (SI of mass on contrast-enhanced images/si of muscle on contrast-enhanced images)/(si of mass on unenhanced images/si of muscle on unenhanced images). The presence of endometriomas was determined according to the criteria of Togashi et al. [5].

358 SIEGELMAN T AL. AJR:163, August 1994 A total of eight solid masses in six patients with surgically proved endometniosis were detected on MR images; two patients each had two solid masses.

our of the masses were excised at surgery, one was sampled by surgical biopsy, and three were inspected at surgery and found to represent dense fibrosis due to endometriosis.

Pathologic examination of the five masses showed fibrosis, variable amounts of endometriotic glandular tissue, little hemorrhage, and scant hemosiderin deposition.

All patients were imaged with a four-coil phased-array (multicoil) system of receiver coils, which allowed a signal-to-noise ratio adequate for 4- to 5-mm-thick sections and a 20- to 24-cm field of

2 358 SIEGELMAN T AL. AJR:163, August 1994 A total of eight solid masses in six patients with surgically proved endometniosis were detected on MR images; two patients each had two solid masses. our lesions were present in the cul-de-sac at or above the vaginal fornix, two involved the bladder wall, and two involved the rectal wall and pararectal fat. our of the masses were excised at surgery, one was sampled by surgical biopsy, and three were inspected at surgery and found to represent dense fibrosis due to endometriosis. Three of the masses were removed or sampled because they were thought to be malignant. Pathologic examination of the five masses showed fibrosis, variable amounts of endometriotic glandular tissue, little hemorrhage, and scant hemosiderin deposition. All MR imaging was performed with a 1.5-T magnet. One milligram of glucagon was injected intramuscularly before imaging to suppress bowel motion. All patients were imaged with a four-coil phased-array (multicoil) system of receiver coils, which allowed a signal-to-noise ratio adequate for 4- to 5-mm-thick sections and a 20- to 24-cm field of view for Ti- and T2-weighted SE imaging. Axial Ti-weighted SE images were obtained with these parameters: /11-18 (TRITE) interleaved sections without an interslice gap, two signal averages, and a 256 x 192 acquisition matrix. T2- weighted images were obtained in axial, coronal, and sagittal planes by using fast spin-echo (SE) sequences with two to four signal averages and a 256 x 256 acquisition matrix. Sixteen refocusing pulses per alpha pulse were spaced at 18- to 20-msec intervals to obtain an effective TE of msec and a TR of msec. All patients received an IV injection of gadopentetate dimeglumine (Magnevist, 0.i mmol/kg; Berlex Laboratories, Wayne, NJ). requency-selective fat-saturated Ti -weighted images were obtained with parameters similar to those used for the Ti-weighted SE images both before (n = 12) and after (n = 13) IV injection of gadopentetate dimeglumine. Results Solid masses from endometniosis were found in six (46%) of 13 patients who had multicoil MR imaging (igs. 1-4). The short-axis dimension of the masses varied from 1 to 5 cm. Enhancement ratios varied from 1.46 to p.?!,w :A - - Seven solid masses had similar MR imaging findings: predominant signal intensity approximating that of muscle on both Ti- and T2-weighted SE images and enhancement by both qualitative and quantitative criteria (igs. 1, 2, and 4). All lesions showed small internal foci of high signal intensity on the Ti -weighted images. These foci showed low signal intensity similar to that of the rest of the lesion on the T2- weighted SE images. All but one (ig. 2) of the lesions had spiculated outer margins with infiltration into the surrounding pelvic fat. On pathologic examination, these masses showed fibrosis and small clusters of endometniotic glandular tissue. Hemosidenin deposition was scant. An eighth mass (ig. 3) had low signal intensity on Ti -weighted SE images and high signal intensity (approaching that of fluid) on T2-weighted SE images. This tissue with high signal intensity nonetheless showed intense contrast enhancement when compared with muscle (ig. 3). Histologic examination of the excised specimen showed a solid mass of glandular endometniosis and surrounding fibrosis. In addition to the solid masses, three of the six patients had MR findings typical of endometniotic Cysts involving the ovaries: lesions with high signal intensity on Ti -weighted SE images and relatively low signal intensity on T2-weighted SE images. Discussion Endometniosis is the presence of endometnial glands and stroma outside the uterus [6]. The MR findings of pelvic endometniosis have been described [5, 7-16]. These reports emphasize the MR findings of endometnial cysts: multiplicity, high signal intensity on Ti-weighted images, and low signal intensity shading [12] on T2-weighted images. Although not emphasized in the MR reports, solid vasculanized masses of endometniosis are not rare [1]. Prominent fibrosis in these C ig. 1.-Enhancing solid pararectal mass of endometriosis in a 38-year-old woman with pelvic pain, painful defecation, and prior supracervical hysterectomy. Histologic examination confirmed abundant fibrous tissue with small foci of endometriosis. A, Sagittal multicoil T2-weighted SE (4600/1 36) MR image shows a low-signal mass (arrows) between cervix and rectal lumen. Cervix has thick walls and contains endocervical cysts. B, Sagittal multicoil Ti-weighted fat-suppressed SE (550116) MR images obtained after injection of gadopentetate dimeglumine shows enhancement within mass (arrows). Mass enhanced to a degree similar to cervix. C, Lateral view from double-contrast barium enema shows constricting lesion with no mucosal lesion.

A, Unenhanced Ti-weighted SE (616/16) MR image shows thickening of posterior vaginal fornix with foci of high signal intensity (arrowheads).

3 AJR:163, August 1994 SOLID PELVIC MASSES CAUSED BY ENDOMETRIOSIS 359 ig. 2.-Enhancing solid mass of endometriosis in cul-de-sac of a 40- year-old woman with a prior laparoscopic diagnosis of endometriosis. A, Unenhanced Ti-weighted SE (616/16) MR image shows thickening of posterior vaginal fornix with foci of high signal intensity (arrowheads). B, SE (4200/119) MR image shows discrete nodule (arrow)to be wholly low signal intensity, similar to muscle. C, Ti-weighted fat-suppressed SE (700/18) MR image after injection of gadopentetate dimeglumine shows that nodule (arrow) enhances to same degree as cervix, indicating that It is vascularized tissue. D, Photomicrograph of excised nodule shows abundant fibrosis () with a paucity of glandular and stromal elements (arrowheads). (H and E, oniginal magnification x20) lesions is a well-recognized pathologic feature, particularly in those affecting the bladder and bowel walls [3, 4]. All eight lesions seen in patients in our study had punctate foci of high signal intensity on Ti-weighted images compatible with hemorrhage. In all cases, enhancement of the lesions with contrast material established that these lesions were composed of solid vasculanized tissue. Therefore, these lesions do not have low signal intensity on T2- weighted images caused by hemoglobin degradation products, as many endometniomas do. Pathologic correlation in these lesions showed abundant dense fibrosis intermixed with small endometnial glands and stroma. Hemosiderin deposition was scanty. One patient (ig. 3) had a pararectal mass showing signal intensity on T2-weighted images that was higher than that of the other masses and similar to that of normal endometnium, corresponding to the pathologic findings of abundant glandular and stromal tissue surrounded by a rim of fibrosis. Zawin et al. [9] reported 24 pelvic lesions in patients with pathologically proved endometniosis that showed low signal on both Ti - and T2-weighted images. They suggested that these lesions were acute hematomas within endometnial cysts, although no contrast material was given. Arriv#{233} et al. [8] illustrated a case of an endometnial mass behind the bladder in a postmenopausal woman that had low signal D.. V Jr! / i/i. intensity on T2-weighted images and appeared similar to our seven lesions; they suggest that solid endometnial implants should have imaging findings similar to those of the endometnium. Our results in a premenopausal population do not support this assertion. Togashi et al. [5] found that masses with low signal intensity on Ti -weighted images or with solid components were unlikely to represent endometniosis and were considered as criteria that excluded the diagnosis of endometnioma. Our findings show that the spectrum of MR imaging findings of pelvic endometniosis is broader than previously reported. The identification of solid penitoneal masses has been reported to suggest ovarian carcinoma [5, 17]. In patients with endometniosis, an erroneous diagnosis of malignant tumor can be fostered by the typical appearance of endometrial cysts: thick-walled with infiltrating outer margins, multiplicity involving both ovaries, evidence of stranding adhesions, and frequently resistive and pulsatility Doppler indexes in the range of malignant lesions [18, 19]. Nonetheless, Togashi et al. [5] have shown that endometnial cysts can be reliably distinguished from other adnexal masses, including a broad spectrum of ovarian malignant tumors. Our results show that solid penitoneal pelvic masses that have signal intensity consistent with fibrosis and internal punctate foci of high signal on Ti-weighted images compatible with

360 SIEGELMAN ET AL. AJR:163, August 1994 ;.,...,..1#{149},,,.., r #{149}.. *, cl. D,..,..* r ig. 3.

assodated pain and bleeding. Prior biopsies performed from endoluminal approach were nondiagnostic.

T2-weighted imag- Cs. Mass has infiltrative margins, low-signal-intensity fibrotic rim, and extramural location.

D, Photomicrograph of resected specimen shows rectal muscularis propria (arrows), fibrosis (), and abundant endometrial glands (asterisks). (H and E, original magnification x20) ig. 4.

4 360 SIEGELMAN ET AL. AJR:163, August 1994 ;.,...,..1#{149},,,.., r #{149}.. *, cl. D,..,..* r ig. 3.-Enhancing solid pararectal mass of endometriosis with abundant glandular and stromal tissue that was resected to exclude malignant tumor in a 41-year-old woman who had a history of rectal mass and assodated pain and bleeding. Prior biopsies performed from endoluminal approach were nondiagnostic. A and B, Unenhanced Ti-weighted SE (650/il, A) and SE (4700/ 136, B) MR images show a 4-cm pararectal mass (M) with intermediate signal intensity on Ti-weighted images and high signal intensity on T2-weighted imag- Cs. Mass has infiltrative margins, low-signal-intensity fibrotic rim, and extramural location. C, Contrast-enhanced Ti-weighted fat-suppressed SE (700/14) MR image shows intense enhancement of entire lesion. D, Photomicrograph of resected specimen shows rectal muscularis propria (arrows), fibrosis (), and abundant endometrial glands (asterisks). (H and E, original magnification x20) ig. 4.-Enhancing solid bladder-wall mass of endometriosis for which a biopsy was done to exclude malignant tumor in a 45-year-old asymptomatic wornan examined because bladder mass was detected sonographically. Histologic examination confirmed abundant fibrous tissue with small foci of endometriosis. A, Sagittal SE ( ) MR image shows a low-signal-intensity mass (arrows) in posterior bladder wall. Mass involves bladder muscularis. Lowsignal mucosa (arrowheads) is partially visible over mass. Uterine fibroids and a hydrosalpinx are present. B and C, Unenhanced (B) and contrast-enhanced (C) axial fat-suppressed Ti -weighted SE (550/i 8) MR images show enhancement throughout mass (arrows) after injection of gadopentetate dimeglumine. Punctate foci of high signal intensity are seen on unenhanced image (8).

5 AJR:163, August 1994 SOLID PELVIC MASSES CAUSED BY ENDOMETRIOSIS 361 hemorrhage should not lead to the diagnosis of ovarian cancinoma. Although the low signal intensity of the fibrotic masses on T2-weighted images is atypical for ovarian malignant tumors [1 7, 20], pelvic malignant tumors can also have these findings. In conclusion, solid enhancing masses are part of the spectrum of MR findings in endometniosis. In patients with known or suspected endometniosis, such lesions, especially when seen in the cul-de-sac or posterior bladder, are not necessarily indicative of peritoneal metastases. REERENCES 1. Sampson JA. Perforating hemorrhagic (chocolate) cysts of the ovary. Arch Surg 1921:3: Benagiano G, Brosens I. The history of andomatriosis: identifying the disease. Hum Reprod 1991:6: Clement PB. Pathology of endometriosis. In: Rosen PP, echner RE, ads. Pathology annual, vol. 25. Norwalk, CT: Appleton & Lange, 1990: Garbie AB, Merrill JA. Pathology of andomatriosis. Clin Obstet Gynecol 1988:31: Togashi K, Nishimura K, Kimura I, at al. Endomatrial cysts: diagnosis with MR imaging. Radiology 1991:180: Brosens IA. The endometrial implant. In: Thomas E, Rock J, ads. Modern approaches to endometriosis. Boston: Kluwar, 1991: Outwatar E, Schiebler ML, Owen RS, Schnall MD. Characterization of hemorrhagic adnexal lesions with MR imaging: blinded reader study. Radiology i993;186: Arriv#{233} L, Hricak H, Martin MC. Pelvic endometriosis: MR imaging. Radiology 1989:171: Zawin M, McCarthy 5, Scoutt L, Comite. Endometriosis: appearance and detection at MR imaging. Radiology 1989:171: Nishimura K, Togashi K, Itoh K, at al. Endometrial cysts of the ovary: MR imaging. Radiology 1987:162: Mitchell DG, Mintz MC, Spritzer CE. Adnexal masses: MR imaging observations at 1.5T, with US and CT correlation. Radiology 1987:162: Nyberg DA, Porter BA, Olds MO, Olson DO, Anderson R, Wesby GE. MR imaging of hemorrhagic adnexal masses. J ComputAssist Tomogr 1987: 11: Sugimara K, Okizuka H, Imaoka I, at al. Pelvic endometriosis: detection and diagnosis with chemical shift MR imaging. Radiology 1993:188: Kier R, Smith RC, McCarthy SM. Value of lipid and water suppression MR images in distinguishing between blood and lipid within ovarian masses. AJR 1992:158: Takahashi K, Okada 5, Kitao M, Sugimura K. Magnetic resonance imaging using lat saturation technique is useful for diagnosing small andomatrioma: a case report. ertil Stenl 1992:58: Stevens 5K, Hricak H, Campos Z. Taratomas versus cystic hemorrhagic adnaxal lesions: differentiation with proton-selective fat-saturation MR imaging. Radiology 1993:186: Stevens 5K, Hricak H, Stem JL. Ovarian lesions: detection and charactarization with gadolinium-enhanced MR imaging at 1ST. Radiology 1991 ;181 : Hamper UM, Sheth 5, Abbas M, Rosenshein NB, Aronson D, Kurman RJ. Transvaginal color Doppler sonography of adnexal masses: differencas in blood flow impedance in benign and malignant lesions. AJR i993;160: Hata K, Hata T, Manabe A, Sugimura K, Kitao M. A critical evaluation of transvaginal doppler studies, transvaginal sonography, magnetic resonance imaging, and CA 125 in detecting ovarian cancer. Obstet Gynecol 1992:80: Ghossain MA, Buy JN, Lignares C, at al. Epithelial tumors of the ovary: comparison of MR and CT findings. Radiology 1993:181:

Value of Lipid- and Water- Suppression MR Images in Distinguishing Between Blood and Lipid Within Ovarian Masses

321 Value of Lipid- and Water- Suppression MR Images in Distinguishing Between Blood and Lipid Within Ovarian Masses Ruben Kier1 The distinction between blood and lipid in ovarian masses on MR imaging