Abdominal Imaging

ª Springer Science+Business Media New York 2015 Published online: 24 March 2015

Abdom Imaging (2015) 40:2630–2644 DOI: 10.1007/s00261-015-0404-1

MR imaging of the retrorectal–presacral tumors: an algorithmic approach Hooman Hosseini-Nik,1 Keyanoosh Hosseinzadeh,2 Rajesh Bhayana,3 Kartik S. Jhaveri1,3 1

Joint Department of Medical Imaging (JDMI), University Health Network, Toronto, ON, Canada Department of Radiology, Wake Forest University School of Medicine, One Medical Center Boulevard, Winston-Salem, NC 27157, USA 3 Department of Medical Imaging, University of Toronto, Toronto, ON, Canada 2

Abstract The retrorectal–presacral space is located posterior to the mesorectum and anterior to the sacrum, and can harbor a heterogeneous group of uncommon masses. Retrorectal–presacral tumors may be classified as congenital, neurogenic, osseous, and miscellaneous. Magnetic resonance imaging (MRI) plays a crucial role in directing appropriate management through accurate diagnosis, detection of complications and anatomic extent. MRI aids in the selection of optimal surgical approach such as anterior, posterior, or combined—based on the lesion extent and relationship to adjacent structures. This article reviews the anatomy of the retrorectal–presacral space and the related tumors, optimal MRI protocol, MRI-based approach to differential diagnosis, and finally pertinent reporting pointers and implications of MR imaging findings for surgical management. Key words: Retrorectal space—Presacral space—Magnetic resonance imaging—Malignant tumor—Benign tumor—Neoplasm

The retrorectal–presacral space is located posterior to the mesorectal fascia and anterior to the sacrum resulting from the confluence of the embryonic hindgut, neuroectoderm, and notochord; hence it is a site of diverse group of tumors [1]. About 20–35% of these tumors are malignant at the time of diagnosis; although most of the

Correspondence to: Keyanoosh Hosseinzadeh; email: khossein@ wakehealth.edu

adult patients are 40–60-year-old women, malignant tumors are reported more frequently in men [2–4]. Since retrorectal–presacral tumors are generally silent or usually present with non-specific symptoms, their true prevalence is unknown [5]. It is estimated that one in 40,000 hospital admissions is due to a retrorectal–presacral tumor [6]. Retrorectal–presacral tumors are generally manifested by mass effect or less commonly by neurological symptoms. Pain is the most common symptom in adults, occurring especially in malignant or infected lesions [7, 8]. Altered bowel habit, obstructed defecation and tenesmus are the next most common symptoms [9]. Injury to the sacral roots (S2–S4) may cause fecal or urinary incontinence, sexual dysfunction, and/or paresthesia in lower extremities [4]. Magnetic resonance imaging (MRI) is the modality of choice for diagnosis and preoperative evaluation of retrorectal–presacral tumors, because of its ability to delineate peritumoral planes and to determine local invasion (sacral and/or rectal) and nerve involvement, with higher contrast resolution compared to computed tomography (CT) or endorectal ultrasound (ERUS) [2, 3]. The imaging features, tumor size, and location relative to the pelvic structures and sacral segments, and the presence or absence of local invasion are all essential factors in selecting the most appropriate surgical approach and for planning the biopsy and/or preoperative chemoradiation therapy, if required. In this review, we discuss the anatomy of the retrorectal–presacral space and its related tumors, optimal MRI protocol, MRI-based approach to differential diagnosis, and finally pertinent reporting pointers and implications of MR imaging findings for surgical management.

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

Anatomy The retroperitoneal portion of the rectum is surrounded by mesorectum, a connective tissue envelope containing fat, vessels, and lymphatics. The outermost layer of mesorectum forms the (meso)rectal fascia [10]. The space between the mesorectal fascia and sacrum (perirectal space) is divided by presacral fascia, into the anterior retrorectal and posterior presacral space [11] (Fig. 1). The retrorectal space is a potential space between the two low-intensity linear structures of mesorectal and pre-

2631

sacral fascia, which can be visualized on MRI only when it contains fluid, fat, or a mass (Fig. 2). At the level of the last sacral vertebrae, mesorectal and presacral fascias fuse to form the rectosacral (or Waldeyer’s) fascia, extending to the posterior wall of the rectum [1, 10]. On either side of the lower part of the rectum, two neurovascular bundles, called lateral ligaments, run between rectal visceral fascia and pelvic parietal fascia above the levator ani [12]. As a result, the retrorectal space extends laterally along the lateral ligaments of rectum, and is bounded anteriorly by the mesorectal fascia, posteriorly by the presacral fascia, superiorly by the peritoneal reflection, and inferiorly by the rectosacral fascia [13, 14] (Fig. 1). Retrorectal and presacral spaces are sometimes used interchangeably. Herein we apply the term ‘‘retrorectal–presacral space’’ to prevent confusion.

Classification

Fig. 1. Mid-sagittal and axial section schematics of pelvic cavity. a Peritoneum, b mesorectal fascia, c presacral fascia, d presacral space, e retrorectal space, f mesorectal space, g rectum, h rectosacral fascia, i levator ani, and j lateral ligaments.

Based on the predominant histopathology, retrorectal– presacral tumors are classified into four groups of congenital, neurogenic, osseous, and miscellaneous lesions; tumors in each group are further subcategorized as benign or malignant (Table 1) [5]. Most of these masses, especially the neurogenic and osseous ones, do not originate in retrorectal or presacral space; however, they may extend into or displace these spaces. Congenital lesions originate from embryologic remnants, accounting for almost two-thirds of retrorectal–presacral masses, and are more common in women [7]. Two-thirds of congenital lesions are benign [6], including developmental cysts, anterior sacral meningocele, and adrenal rest

Fig. 2. Normal anatomy on sagittal: A Sagittal T2-W MRI, B axial T2-W MRI. Rectorectal space is visible (due to fluid accumulation) between the anterior mesorectal fascia (arrow

heads) and posterior presacral fascia (straight arrows). The presacral space (curved arrows) is the area of intermediate signal intensity posterior to the presacral fascia.

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

2632

Table 1. Classification of retrorectal–presacral tumors Benign

lesions include giant-cell tumor, which is the second most frequent primary sacral neoplasm after chordoma [17].

Malignant

Congenital Developmental cysts (Tailgut cyst, rectal duplication cyst, epidermoid, dermoid, and teratoma) Anterior sacral meningocele Adrenal rest tumors Neurogenic Schwannoma Neurofibroma Ganglioneuroma Osseous Giant-cell tumor Aneurysmal bone cyst Osteoblastoma

MRI technique

Chordoma Teratocarcinoma

Ependymoma Malignant nerve sheath tumors Neuroblastoma Ganglioneuroblastoma Osteogenic sarcoma Ewing’s sarcoma Myeloma Chondrosarcoma

Miscellaneous Lipoma Myelolipoma Myxoma Extramedullary hematopoiesis Fibroma Leiomyoma Desmoid tumor Abscess/hematoma

Liposarcoma Mucinous cystadenocarcinoma Fibrosarcoma Malignant fibrous histiocytoma Hemangiopericytoma Metastatic carcinoma Lymphoma

Modified from Ref. [5]

MR imaging of the pelvis utilizing 1.5 T or 3 T systems with multi-channel phased array torso coils provides optimal image quality. Routine MRI protocol must include multiplanar 2D or high spatial-resolution 3D T2weighted (T2-W) pulse sequences, together with obliquely oriented 2D T2-W sequences along the long axis of the sacrum to assess the relationship of the mass to the rectum, sacrum, sacral foramina, and nerve roots. Frequency-selective or inversion recovery fat-suppressed T2-weighted pulse sequences will improve the dynamic range for T2-weighting and tissue contrast and also confirm the presence of macroscopic fat. Routine T1weighted (T1-W) sequences with and without fat saturation will examine for macroscopic fat and multiphasic contrast-enhanced acquisitions must be acquired for appropriate characterization. In-phase (IP) and outof-phase (OP) T1 gradient-echo imaging is helpful in identification of intravoxel/intracellular lipid (Table 2).

MR imaging approach

tumors. Developmental cysts comprise up to 60% of all congenital retrorectal–presacral masses, and are reported more commonly in females (ratio 3:1) [7, 15]. The most common malignant congenital lesions are chordoma and teratocarcinoma. About 85% of neurogenic tumors are benign [16], and after congenital lesions, they are the second most common retrorectal–presacral tumors (up to 15%) [6]. Osseous tumors comprise up to 10% of retrorectal–presacral lesions, and originate from bone, cartilage, fibrous tissue, or marrow [8, 13]. The osseous

Retrorectal–presacral masses can be initially categorized into two groups with respect to the presence or absence of intralesional fat. Fat-containing masses are subdivided into cystic, solid, or complex lesions; margination provides further discrimination in the solid category. Nonfat-containing masses can be classified as cystic or solid. The cystic lesions are subclassified as unilocular or multilocular; and the solid lesions are differentiated based on the presence or absence of sacral destruction. With respect to these features, we propose an MR-based algorithmic approach to retrorectal–presacral masses,

Table 2. Sample MRI parameters for retrorectal–presacral tumors (1.5 T)

Repetition time (ms) Echo time (ms) Number of slices Bandwidth (Hz/Px) FOV (mm) Slice thickness (mm) Distance factor (%) Phase FOV (%) Number of acquisitions Matrix Phase encode direction Saturation band Acquisition time (min) a

Sagittal T2 FSEa

Axial T2 FSEa

Coronal T2 FSEa

Axial T1 IP/ OP GRE

3500 91 28 391 220 3 25 100 3 350 9 263 AP Ant 4

3320 91 40 391 220 4 25 100 2 350 9 263 Trans (R > L) N/A 5.5

3500 91 25 391 220 4 25 100 2 350 9 263 Trans (R > L) N/A 4

174 2.2, 4.5 54 390, 400 300 5 0 300 1 140 9 225 Trans (R > L) N/A 1

Coronal-oblique T2 FSE 4000 80 15 391 200 3 0 100 3 350 9 263 AP Sup & Inf 5

DWIb

3D T1 GREc

5800 96 30 1132 250 4 20 100 6 250 9 250 AP N/A 4.5

4.44 1.59 32 400 240 4 20 100 1 240 9 240 AP N/A 1

One of these sequences can be replaced with 3D T2 FSE, and one can be performed with fat-suppression b-value: 0, 50, 400 and 800 s/mm2 c Pre-contrast and 3 post-contrast phases Ant, anterior; AP, anteroposterior; FSE, fast spin echo; GRE, gradient refocused echo; Hi-res, high resolution; IP/OP, in-phase/out-of-phase; Inf, inferior; L, left; N/A, not applicable; Px, pixel; R, right; Sup, superior; Trans, transverse

b

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

Fig. 3.

2633

Differential diagnosis flowchart for retrorectal–presacral masses containing fat.

which is represented as two flowcharts in Figures 3 and 4. Care must be taken to differentiate sacral erosion, which can be caused by any longstanding mass in the vicinity of sacrum, from sacral destruction, which occurs with tumors with sacral or intrasacral origin or those invading it.

Fat-containing masses Teratomas are neoplasms derived from at least two germ cell layers [18]. When teratomas occur in retrorectal–presacral space, they usually have a sacrococcygeal component; although rarely they may show no connections to the sacrum or coccyx [19] (Fig. 5). Sacrococcygeal teratomas (SCT) are often found in females (ratio 3:1) and especially in the pediatric population [20]. These tumors are typically complex with solid and cystic components, but occasionally may be entirely cystic (uni- or multilocular) [21–23]. Macroscopic fat and intravoxel lipid can be detected in more than 80% of teratomas on imaging, either within the solid component or more typically as sebum (fat-fluid levels) in cystic parts [24]. Calcifications, presenting in more than 50% of teratomas [21], are readily visualized on CT, and appear as low-intensity foci on MRI. Up to 10% of

SCT diagnosed in infants are malignant, and the tendency for malignant transformation rises with increasing age [25]. Malignancy is more likely in predominantly solid tumors [26], and may be manifested by enhancement, necrosis, or hemorrhage in the solid component of the tumor [27], or more reliably by local invasion and/or regional lymphadenopathy. Dermoid cysts are ectodermal-derived lesions which may occur in retrorectal–presacral space, and are more common in young women [28]. These are round uni- or multilocular cysts with variable fat content [29, 30]. Cystic (mature) teratomas may mimic dermoid cysts on imaging; but they differ histologically, as teratomas originate from more than one germ cell layer. Liposarcomas are the most common malignant retroperitoneal soft-tissue tumors in patients over 50 and have no gender predilection [31, 32]. They are divided into five subtypes of well-differentiated (WDL), dedifferentiated (DDL), myxoid, round cell, and pleomorphic. WDL and DDL are more commonly found in the retroperitoneum, while other subtypes usually occur in the extremities. WDLs are typically ill-defined, round or lobulated, and usually septated tumors which are mostly composed of mature fat and sometimes contain small enhancing non-adipocytic solid components, while

2634

Fig. 4.

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

Differential diagnosis flowchart for retrorectal–presacral masses without fat.

DDLs have less fatty component (showing as a discrete area or scattered foci throughout the lesion) and may be heterogeneous because of necrotic changes and/or calcification [33, 34]. Myxoid liposarcoma is an enhancing solid tumor which may demonstrate fluid intensity signal on T1W and T2-W MRI due to high myxomatous content, while round cell and aggressive pleomorphic liposarcoma have non-specific MR imaging features [34–36]. Lipomas are rare in the retroperitoneum with most of the pathologically proven cases felt to represent WDL liposarcomas misdiagnosed due to undersampling [31]. These masses characteristically appear on MRI as encapsulated homogenous fat-containing lesions; however, non-adipose elements such as fibrous septa and blood vessels have been described [37]. Myelolipomas are benign well-defined round or oval masses, composed of various proportions of fat and myeloid elements [38]; and when extra-adrenal, they are most frequently seen in the retrorectal–presacral space [31]. They are more common in middle-to-older aged women (ratio 2:1) [39, 40]. Macroscopic fat in myelolipoma is typically sufficient to be visible with standard fatsuppression sequences, while less commonly, intracellular

lipid is only apparent on chemical shift imaging [41]. Hematopoietic elements yield low signal on T1-W and intermediate signal intensity on T2-W images, and can show contrast enhancement [39, 42]. Unlike liposarcomas, myelolipoma have a ‘‘lumpy-bumpy’’ contour, and can show intratumoral hemorrhage with variable signal intensity depending on its stage (Fig. 6). Old hematomas can also calcify [43]. The posterior margin of myelolipomas may adhere to the presacral fascia and appear ill-defined on imaging, which makes resection difficult especially if performed laparoscopically. Extramedullary hematopoiesis (EMH) may rarely occur in the retrorectal–presacral space, sometimes secondary to previous sacral fractures [44]. EMH usually presents as multiple well-defined fat-containing masses adjacent to the anterior aspect of sacrum; however, it can occasionally manifest as presacral thickening as well [45]. Depending on the proportion of red marrow to adipose tissue, lesions may show variable signal intensity on MRI. The myeloid non-adipose tissues are iso- or hypointense on T1-W and T2-W images, and may show moderate enhancement [46, 47]. In the absence of hematological disorders, multifocality is the main finding

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

2635

Fig. 5. Cystic teratoma in a 33-year-old woman. A Sagittal T2-W, B sagittal post-contrast fat sat T1-W MRI, C axial in-phase (IP) T1-W MRI, D axial out-of-phase (OP) T1-W MRI. Smooth wall enhancement is evident on enhanced

image, without any enhancing septum or mural nodule; there is no evidence of sacral involvement. Signal loss on OP image is in favor of a fat-containing mass.

to suggest EMH over other lipomatous tumors, especially myelolipoma.

T1-W and are often heterogeneous on T2-W images mainly due to hemorrhage and/or cystic degeneration [51, 52] (Fig. 7). The typical, but not specific, appearance of neurofibromas is described as target-like, with central low signal intensity (fibrocollagenous tissue) and peripheral high signal intensity (myxoid stroma) on T2-W images [53]. In neurofibromatosis type 1, neurofibromas are usually plexiform presenting as lobular masses involving multiple nerves of a plexus [54]. Rarely, schwannoma and plexiform neurofibroma may undergo malignant transformation [55, 56]. The main markers of malignancy include large size (>5 cm) at presentation, irregular borders, and rapid growth; however, peripheral enhancement, perilesional edema-like zone, intratumoral

Non-fat-containing solid masses with intact sacrum Nerve sheath tumors are the most common neurogenic tumors in presacral space and include schwannomas and neurofibromas [48]. They are slightly more prevalent in men, and present most often in the third to fifth decades of life [49]. Nerve sheath tumors are well-defined enhancing masses, sometimes in relation to thickened nerve roots or widened neural foramina [50]. Schwannomas show low-to-intermediate signal intensity on

2636

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

Fig. 6. Extra-adrenal myelolipoma in a 69-year-old woman. A Axial in-phase (IP) T1-W MRI, B axial out-of-phase (OP) T1-W MRI, C Axial T2-W MRI, and D sagittal T2-W MRI. An encapsulated tumor with ‘‘lumpy-bumpy’’ contour (filled arrowheads) adhering to presacral fascia (hollow arrowheads). Intravoxel lipid shows signal intensity loss on

OP image. The mass shows mixed signal intensity, with macroscopic lipid, and hematopoietic tissue which is hypointense on T1-W and hypo to intermediate signal on T2W images. Note the intralesional rim-like hypointensity on sagittal T2-W image (arrow) suggesting hemosiderin from prior hemorrhage.

cystic change, heterogeneity on T1-W images, and intratumoral lobulation have also been suggested [57–60]. Myxomas are benign tumors, more frequently seen in 50–60-year-old women, which arise from abnormally differentiated fibroblasts usually in cardiac or limb muscles, or rarely in pelvis and in the retroperitoneum

[61]. Myxomas are well-circumscribed lobulated tumors, mimicking cysts by their homogeneous hypointensity on T1-W and hyperintensity on T2-W images [62]; however, their heterogeneous contrast enhancement differentiates them from cysts and schwannomas with cystic degeneration. Peritumoral high signal intensity on T2-W im-

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

2637

Fig. 7. Presacral schwannoma in a 71-year-old female. A Axial post-contrast T1-W fat-suppressed MRI, B sagittal fatsuppressed T2-W MRI. There is an enhancing solid mass

(with cystic components) just anterior to sacrum displacing rectum and uterus to left side. Tumor contiguity to sacral nerve roots is clearly visible (arrows).

Fig. 8. 56-year-old male with history of prostate cancer with biopsy-proven myxoma. A Coronal T1-W MRI, B axial fatsuppressed T2-W MRI. A well-defined lobulated mass with

uniformly low signal intensity on T1-W and uniformly high signal intensity on T2-W images. The perilesional high signal intensity is due to leak of myxomatous tissue (arrow).

ages may represent leak of myxomatous tissue [63] (Fig. 8). Lymphoma or metastases (mostly from colorectal carcinoma or carcinoid) also may develop in retrorectal– presacral space [4, 64, 65].

Non-fat-containing solid masses with sacral destruction Sacrococcygeal chordomas are the most common solid tumors in the presacral space and account for 50% of all

2638

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

Fig. 9. Sacral chordoma in a 69-year-old female. A Sagittal unenhanced T1-W MRI, B sagittal T2-W MRI. A large heterogeneous mass with low-to-intermediate signal on T1-W image and intermediate signal on T2-W image, showing distal

sacral destruction and extension to retrorectal space and posterior paraspinal soft tissue. Mass had involved rectum at the time of resection.

cases of chordomas [66]. These malignant tumors are diagnosed more commonly in men (ratio 2:1), in the sixth and seventh decades [67]. Their main cause of mortality and morbidity is local aggressive invasion, causing pain or less frequently neurologic complications such as incontinence or impotence [68]. Sacrococcygeal chordomas are typically large and lobulated masses, usually with discrete margins, which originate in the midline of the sacrum, destroy the bone and extend into adjacent softtissue and/or sacral epidural space [69]. Chordomas are usually heterogeneous, generally iso- to hypointense on T1-W and hyperintense on T2-W images (due to high water content of mucin), may contain areas of hemorrhage (foci of high signal on T1-W images) and/or calcification (hypointense on T2-W images) (Fig. 9) and show variable enhancement [70]. Osseous tumors may be differentiated from chordoma by their different signal intensities or off-midline location. Giant-cell tumors, unlike chordoma, yield low-tointermediate T2 signal intensity, probably because of their high collagen content; and in the presence of aneurysmal bone cyst components, fluid–fluid levels may also be seen [71]. Chondrosarcomas arise from sacroiliac joint cartilage, and they almost always display calcification [72]. Intrasacral neurogenic tumors are centered in the sacrum, and therefore can easily be distinguished from

retrorectal to presacral based masses. These include nerve sheath tumors which can expand and destroy the sacrum. Intradural myxopapillary ependymomas may rarely expand and destroy the sacrum, extending into the presacral area, with imaging similar to chordomas but with more intense enhancement [73, 74].

Non-fat-containing cystic masses Tailgut cysts, also referred to as mucus-secreting cysts or cystic hamartomas, originate from the terminal portion of the hindgut. They occur in retrorectal–presacral space, and are more commonly seen in middle-aged women [75]. The proteinaceous content of the mucinous fluid may exhibit a range of signal intensities from low to high on T1-W and conversely from high to low on T2-W images (Fig. 10) [76, 77]. Tailgut cysts are most commonly multilocular, may be surrounded with a cluster of smaller cysts making a honeycomb pattern, and may contain T2 hypointense septa which enhance [77–79]. Tailgut cysts may be complicated by infection (forming abscess and/or perianal abscess) and/or malignant degeneration (in up to 13% of cases) [80, 81]. Both an acutely infected tailgut cyst and an ordinary abscess may present on MRI by marked cyst wall and perilesional enhancement. However, multilocularity and the presence of satellite lesions can help differentiate an infected tail-

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

2639

Fig. 10. Benign tailgut cyst in a 62-year-old woman. A Sagittal T2-W MRI. B axial post-contrast T1-W MRI. There is a multiloculated cystic lesion (asterisk) with internal septa

(arrows) in the left side of the retrorectal space. Smooth wall enhancement (arrowheads) is evident without any enhancing solid component or mural nodularity.

gut cyst (that should be surgically resected) from an ordinary abscess (that is usually managed by drainage) [77]. In a woman, a multilocular cystic lesion with T1 hyperintense content but without fat, favors the diagnosis of tailgut cysts over other cystic lesions such as duplication cyst, meningocele, cystic teratoma, cystic schwannoma, or mucinous adenocarcinoma. Nodular thickening of cyst wall and the presence of an enhancing intracystic polypoid mass may help differentiate an infected tailgut cyst from one with malignant transformation (Fig. 11) [79, 80]. Rectal duplication cysts are the result of hindgut sequestration during embryogenesis, constituting 5% of all developmental cysts in the retrorectal space [15], and sometimes associated with bladder, urethral, and anorectal abnormalities [82]. Rectal duplication cysts are unilocular cysts in continuity with the rectum. They have with a well-developed muscularis mucosa, and a deeper muscularis propria which is shared with the rectum, best identified on endorectal MRI as low signal bands separated by a hyperintense layer of submucosa [83]. Although the presence of air inside the cyst may suggest communication with the rectal lumen, it can also be seen in abscesses or even other cystic lesions [3]. Malignant degeneration has been reported in about 20% of rectal duplication cysts [84]. Epidermoid cysts are unilocular thin-walled benign cysts, more common in middle-aged women. The cysts arise from ectoderm and may communicate with the

skin, making a postanal dimple or sinus [3]. The cysts fluid is hypo- or hyperintense on T1-W and hyperintense on T2-W images (Fig. 12), and contains keratin debris, as T2 hypointense foci, in the dependent or nondependent portion [85, 86]. Anterior sacral meningocele is a well-defined unilocular thin-walled cyst containing CSF intensity fluid, sometimes with a visible stalk communicating with the thecal sac [15]. The lesion is more common in females (ratio 4:1) [87], and in 50% of cases may be accompanied with other congenital anomalies such as spina bifida, tethered spinal cord, imperforate anus, uterine/vaginal duplications, or presacral lipoma [5, 13, 15]. Extramucosal mucinous adenocarcinomas refer to abundant pools of mucin in retrorectal or perianal space formed by mucinous-type tumors. Mucinous adenocarcinomas have a worse prognosis than non-mucinous subtypes, and either originate from fistula-in-ano typically in middle-aged patients with longstanding perianal Crohn’s disease (incidence 0.7%) or arise from the anorectum in older patients [88–91]. These tumors are indistinguishable from each other. Their cystic component is often multi-septated and hyperintense on T2-W images, showing post-contrast nodular enhancement; and the fistulous communication between the cystic component and the rectum is frequently visualized [91, 92] (Fig. 13). History and the presence of a fistula aid in differentiating mucinous adenocarcinoma from malignant degeneration of developmental cysts [93].

2640

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

Fig. 11. Tailgut cyst with malignant transformation in a 72year-old woman. A Sagittal post-contrast T1-W MRI. B sagittal T2-W MRI. A complex large retrorectal cystic mass

occupying the entire pelvis, with multiple enhancing mural nodules (arrows) and septations (arrow head).

Fig. 12. Epidermoid cyst in a 47-year-old woman. A Axial T1-W MRI, B sagittal T2-W MRI. This is a unilocular cyst (arrow) filled with simple fluid (without visible keratin debris)

showing hypointensity on T1-W and hyperintensity on T2-W images (asterisk rectum, arrow head coccyx).

MRI reporting

as retrorectal–presacral, tumors should be located between sacrum and rectum. Anterior or anterolateral displacement and compression of rectum may help differentiate these masses from intramural rectal tumors.

The assessment of tumors on MRI should address the following four aspects: (1) location, (2) size, (3) morphology, and (4) margins and interface. To be regarded

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

2641

Fig. 13. 72-year-old female with biopsy-proven mucinous extramucosal adenocarcinoma of anorectum. A Axial T2-W MRI, B axial post-contrast T1-W MRI at a lower level. There is a large retrorectal multilocular cystic mass (asterisk) displacing

rectum (r) to the right, and showing nodular wall enhancement. Cyst content is hyperintense on T2-W image and hypointense on T1-W images; and the wide-mouth fistulous communication between the cyst and rectum is evident (arrows).

The upper and lower extents of the tumor should be identified with respect to the adjacent sacral vertebral bodies on sagittal images, and its largest two dimensions should be measured. Morphological characteristics of the tumor including, signal intensity, uniformity (heterogeneity vs. homogeneity), pattern of post-contrast enhancement, and nodules and septa in cystic lesions should be noted. The tumor margins need to be described as being sharply-defined (smooth, lobular, or irregular) or poorly-defined; and its interface with surrounding structures (incl. sacrum, rectum, cervix, pelvic side walls, and sciatic notches) should be assessed as being clear, abutting (narrow-based contact), contiguous (broadbased contact), eroding, or invasive. When tumor is adherent to the rectal wall or sacrum, MRI excels CT in making the distinction between an abutting mass and actual invasion; change in rectal wall signal intensity and thickening are two findings in favor of rectal wall invasion [69]. Care should be taken in defining the margins of WDLs or DDLs on imaging; otherwise parts of the tumor resembling the adjacent normal fatty tissue can be left after the surgical resection. Some tumors, like chordomas, may extend into gluteal muscles, spinal canal, and also along the posterior spinal muscles; identification of such extensions on preoperative MR images guides surgeons to a radical resection with negative margins which markedly decrease the local recurrence rate [68, 69, 94]. The MRI report will be completed by providing the list of differential diagnoses based on

imaging features (discussed earlier); and findings with possible impact on management (e.g., tumor’s boundaries and invasions) should be highlighted.

Management In the retrorectal–presacral location, 60% of solid and 10% of cystic tumors are malignant or bear areas with malignant transformation. In addition, some benign lesions are prone to malignant change; cystic lesions may become infected, and large masses of any type may contribute to dystocia in young women [2, 7, 8]; therefore most of these masses should be treated once diagnosed. Surgical resection is the therapeutic option of choice [8], which in some patients with malignant lesions may be followed by postoperative chemoradiation. On the other hand, preoperative neoadjuvant therapy may be curative in some tumors or may facilitate the resection of some large pelvic masses by decreasing their size [5]. Accurate MR imaging has decreased the need for routine preoperative biopsy, decreasing the complications such as tumor seeding or infection. Currently only inoperable tumors or those with imaging features strongly suggestive for metastasis or lymphoma may undergo preoperative biopsy [7, 9]. The most appropriate surgical approach for a given lesion depends on its boundaries, and relationship to and involvement of adjacent structures. There are three major surgical approaches, including anterior, posterior,

2642

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

and combined. The anterior (abdominal) approach is generally performed when the tumor is completely above the mid-body of S3 vertebral body [9, 95]. The posterior (transsacral, parasacral, or transperineal) approach is generally used for tumors that are completely below the mid-body of S3 without visceral, pelvic, or side wall involvement [4, 95]. The combined approach is indicated if the lesion has a portion above the mid-body of S3, or if invasion to sacrum, pelvic sidewall, or viscera is suspected [4]. In case of sacral invasion or when the tumor is firmly attached to the sacrum or coccyx (e.g., teratomas [7]), total or subtotal sacrococcygectomy will be indicated. This procedure may be complicated by pelvic instability when more than 50% of the S1 vertebral body is removed [96], or may lead to bladder or anorectal dysfunction, especially when S2–S4 nerves are sacrificed bilaterally [97]. Invasion to the pelvic side walls threatens the internal iliac vessels and ureters, and sciatic notch involvement may increase the risk of sciatic nerve injury or intractable hemorrhage during the surgery [9]. Follow-up is essential after surgical resection of both benign and malignant retrorectal–presacral lesions. Dozois et al. [5] suggest that patients with benign lesions have a baseline MRI 1-year following surgery and every 5 years thereafter. Those patients with malignant tumors should have annual pelvic MRI and chest CT scan in the first five postoperative years. Recurrent tumors exhibit the same signal intensity as the primary tumors; and post-contrast enhancement distinguishes them from seromas (a common postoperative finding), which do not enhance or present only rim enhancement [98].

2. 3. 4.

5. 6. 7. 8. 9.

10. 11.

12. 13. 14. 15.

Summary Retrorectal–presacral tumors are uncommon, but include a very broad differential diagnosis. Patients are often asymptomatic or present with vague symptoms, delayed diagnosis is common and tumors can be large or might have undergone malignant transformation by the time they are discovered. The role of radiologist in the multidisciplinary management of patients with retrorectal– presacral tumors is paramount. Optimal MR imaging and precise image interpretation and reporting of findings can obviate the need for routine preoperative biopsy, help in selecting the appropriate surgical approach with reduced complications, aid in planning preoperative chemoradiation therapy if required, and facilitate early detection of recurrence on surveillance examinations. Conflict of interest The authors declare that they have no conflict of interest.

References 1. Garcia-Armengol J, Garcia-Botello S, Martinez-Soriano F, Roig JV, Lledo S (2008) Review of the anatomic concepts in relation to the retrorectal space and endopelvic fascia: Waldeyer’s fascia and

16. 17.

18. 19. 20.

21.

22.

23. 24.

the rectosacral fascia. Colorectal Dis 10(3):298–302. doi:10.1111/j. 1463-1318.2007.01472.x Glasgow SC, Birnbaum EH, Lowney JK, et al. (2005) Retrorectal tumors: a diagnostic and therapeutic challenge. Dis Colon Rectum 48(8):1581–1587. doi:10.1007/s10350-005-0048-2 Yang BL, Gu YF, Shao WJ, et al. (2010) Retrorectal tumors in adults: magnetic resonance imaging findings. World J Gastroenterol 16(46):5822–5829 Bosca A, Pous S, Artes MJ, et al. (2012) Tumours of the retrorectal space: management and outcome of a heterogeneous group of diseases. Colorectal Dis 14(11):1418–1423. doi:10.1111/j.1463-1318. 2012.03016.x Dozois E, Marcos M (2011) Presacral tumors. In: Beck D, Roberts P, Saclarides T (eds) The ASCRS textbook of colon and rectal surgery, vol. 2. New York: Springer, pp 359–374 Jao SW, Beart RW Jr, Spencer RJ, Reiman HM, Ilstrup DM (1985) Retrorectal tumors. Mayo Clinic experience, 1960–1979. Dis Colon Rectum 28(9):644–652 Hobson KG, Ghaemmaghami V, Roe JP, Goodnight JE, Khatri VP (2005) Tumors of the retrorectal space. Dis Colon Rectum 48(10):1964–1974. doi:10.1007/s10350-005-0122-9 Bullard Dunn K (2010) Retrorectal tumors. Surg Clin North Am 90:8 Macafee DA, Sagar PM, El-Khoury T, Hyland R (2012) Retrorectal tumours: optimization of surgical approach and outcome. Colorectal Dis 14(11):1411–1417. doi:10.1111/j.1463-1318. 2012.02994.x Scha¨fer AO (2010) Anorectal anatomy: clinical implications for the MR radiologist. In: Scha¨fer AO, Langer M (eds) MRI of rectal cancer. Berlin: Springer Zhang C, Ding ZH, Li GX, et al. (2010) Perirectal fascia and spaces: annular distribution pattern around the mesorectum. Dis Colon Rectum 53(9):1315–1322. doi:10.1007/DCR.0b013e3181e74 525 Wang GJ, Gao CF, Wei D, Wang C, Meng WJ (2010) Anatomy of the lateral ligaments of the rectum: a controversial point of view. World J Gastroenterol 16(43):5411–5415 Neale JA (2011) Retrorectal tumors. Clin Colon Rectal Surg 24(3):149–160. doi:10.1055/s-0031-1285999 Wolpert A, Beer-Gabel M, Lifschitz O, Zbar AP (2002) The management of presacral masses in the adult. Tech Coloproctol 6(1):43–49. doi:10.1007/s101510200008 Dahan H, Arrive L, Wendum D, et al. (2001) Retrorectal developmental cysts in adults: clinical and radiologic-histopathologic review, differential diagnosis, and treatment. Radiographics 21(3): 575–584. doi:10.1148/radiographics.21.3.g01ma13575 Hassan I, Wietfeldt ED (2009) Presacral tumors: diagnosis and management. Clin Colon Rectal Surg 22(2):84–93. doi: 10.1055/s-0029-1223839 Diel J, Ortiz O, Losada RA, et al. (2001) The sacrum: pathologic spectrum, multimodality imaging, and subspecialty approach. Radiographics 21(1):83–104. doi:10.1148/radiographics.21.1.g01ja 0883 Azizkhan RG, Caty MG (1996) Teratomas in childhood. Curr Opin Pediatr 8(3):287–292 Wishnia SC, Rosen JE, Hamid MA, Haas S, Moreno-Ruiz N (2008) Management of a presacral teratoma in an adult. J Clin Oncol 26(15):2586–2589. doi:10.1200/JCO.2007.15.6034 Bartels SA, van Koperen PJ, van der Steeg AF, et al. (2011) Presacral masses in children: presentation, aetiology and risk of malignancy. Colorectal Dis 13(8):930–934. doi:10.1111/j.1463-1318. 2010.02312.x Keslar PJ, Buck JL, Suarez ES (1994) Germ cell tumors of the sacrococcygeal region: radiologic-pathologic correlation. Radiographics 14(3):607–620 (quiz 621–602). doi:10.1148/radiographics. 14.3.8066275 Simpson PJ, Wise KB, Merchea A, et al. (2014) Surgical outcomes in adults with benign and malignant sacrococcygeal teratoma: a single-institution experience of 26 cases. Dis Colon Rectum 57(7): 851–857. doi:10.1097/DCR.0000000000000117 Herman TE, Siegel MJ (2002) Cystic type IV sacrococcygeal teratoma. J Perinatol 22(4):331–332. doi:10.1038/sj.jp.7210706 Davidson AJ, Hartman DS, Goldman SM (1989) Mature teratoma of the retroperitoneum: radiologic, pathologic, and clinical corre-

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

25. 26. 27. 28. 29. 30. 31.

32.

33. 34. 35.

36.

37.

38. 39. 40. 41. 42. 43. 44.

45.

46.

lation. Radiology 172(2):421–425. doi:10.1148/radiology.172.2. 2664866 Altman RP, Randolph JG, Lilly JR (1974) Sacrococcygeal teratoma: American Academy of Pediatrics Surgical Section Survey1973. J Pediatr Surg 9(3):389–398 Monteiro M, Cunha TM, Catarino A, Tome V (2002) Case report: sacrococcygeal teratoma with malignant transformation in an adult female: CT and MRI findings. Br J Radiol 75(895):620–623 Renato F, Paolo V, Girolamo M, et al. (1996) Malignant retroperitoneal teratoma: case report and literature review. Acta Urol Belg 64(3):49–54 Munteanu I, Badulescu A, Mastalier B, et al. (2013) Retrorectal dermoid cyst: a rare clinical entity. Curr Health Sci J 39(3):179–183 Coco C, Manno A, Mattana C, et al. (2008) Congenital tumors of the retrorectal space in the adult: report of two cases and review of the literature. Tumori 94(4):602–607 Jones M, Khosa J (2013) Presacral tumours: a rare case of a dermoid cyst in a paediatric patient. BMJ Case Rep . doi:10.1136/ bcr-2013-008783 Craig WD, Fanburg-Smith JC, Henry LR, Guerrero R, Barton JH (2009) Fat-containing lesions of the retroperitoneum: radiologicpathologic correlation. Radiographics 29(1):261–290. doi:10.1148/ rg.291085203 Kransdorf MJ (1995) Benign soft-tissue tumors in a large referral population: distribution of specific diagnoses by age, sex, and location. AJR Am J Roentgenol 164(2):395–402. doi:10.2214/ajr.164. 2.7839977 Francis IR, Cohan RH, Varma DG, Sondak VK (2005) Retroperitoneal sarcomas. Cancer Imaging 5:89–94. doi:10.1102/ 1470-7330.2005.0019 Song T, Shen J, Liang BL, et al. (2007) Retroperitoneal liposarcoma: MR characteristics and pathological correlative analysis. Abdom Imaging 32(5):668–674. doi:10.1007/s00261-007-9220-6 Shin NY, Kim MJ, Chung JJ, et al. (2010) The differential imaging features of fat-containing tumors in the peritoneal cavity and retroperitoneum: the radiologic-pathologic correlation. Korean J Radiol 11(3):333–345. doi:10.3348/kjr.2010.11.3.333 Gebhard S, Coindre JM, Michels JJ, et al. (2002) Pleomorphic liposarcoma: clinicopathologic, immunohistochemical, and followup analysis of 63 cases: a study from the French Federation of Cancer Centers Sarcoma Group. Am J Surg Pathol 26(5):601–616 Gaskin CM, Helms CA (2004) Lipomas, lipoma variants, and welldifferentiated liposarcomas (atypical lipomas): results of MRI evaluations of 126 consecutive fatty masses. AJR Am J Roentgenol 182(3):733–739. doi:10.2214/ajr.182.3.1820733 Sutker B, Balthazar EJ, Fazzini E (1985) Presacral myelolipoma: CT findings. J Comput Assist Tomogr 9(6):1128–1130 Kammen BF, Elder DE, Fraker DL, Siegelman ES (1998) Extraadrenal myelolipoma: MR imaging findings. AJR Am J Roentgenol 171(3):721–723. doi:10.2214/ajr.171.3.9725304 Singla AK, Kechejian G, Lopez MJ (2003) Giant presacral myelolipoma. Am Surg 69(4):334–338 Cyran KM, Kenney PJ, Memel DS, Yacoub I (1996) Adrenal myelolipoma. AJR Am J Roentgenol 166(2):395–400. doi:10.2214/ ajr.166.2.8553954 Baker KS, Lee D, Huang M, Gould ES (2012) Presacral myelolipoma: a case report and review of imaging findings. J Radiol Case Rep 6(6):1–9. doi:10.3941/jrcr.v6i6.1095 Dann PH, Krinsky GA, Israel GM (2008) Case 135: presacral myelolipoma. Radiology 248(1):314–316. doi:10.1148/radiol. 2481050321 Macki M, Bydon M, Papademetriou K, Gokaslan Z, Bydon A (2013) Presacral extramedullary hematopoiesis: an alternative hypothesis. J Clin Neurosci 20(12):1664–1668. doi:10.1016/j.jocn. 2013.06.003 Denies E, Duwel V, Delvaux P (2012) Presacral extramedullary haematopoiesis: a diagnostic update and case report of a late diagnosis. Int J Surg Case Rep 3(9):474–476. doi:10.1016/j.ijscr. 2012.06.002 Ginzel AW, Kransdorf MJ, Peterson JJ, Garner HW, Murphey MD (2012) Mass-like extramedullary hematopoiesis: imaging features. Skeletal Radiol 41(8):911–916. doi:10.1007/s00256-0111323-z

2643

47. Rajiah P, Hayashi R, Bauer TW, Sundaram M (2011) Extramedullary hematopoiesis in unusual locations in hematologically compromised and noncompromised patients. Skeletal Radiol 40(7):947–953. doi:10.1007/s00256-011-1129-z 48. Dozois EJ, Wall JC, Spinner RJ, et al. (2009) Neurogenic tumors of the pelvis: clinicopathologic features and surgical outcomes using a multidisciplinary team. Ann Surg Oncol 16(4):1010–1016. doi: 10.1245/s10434-009-0344-5 49. Hirano K, Imagama S, Sato K, et al. (2012) Primary spinal cord tumors: review of 678 surgically treated patients in Japan. A multicenter study. Eur Spine J 21(10):2019–2026. doi:10.1007/s00586012-2345-5 50. Hayasaka K, Tanaka Y, Soeda S, Huppert P, Claussen CD (1999) MR findings in primary retroperitoneal schwannoma. Acta Radiol 40(1):78–82 51. Hughes MJ, Thomas JM, Fisher C, Moskovic EC (2005) Imaging features of retroperitoneal and pelvic schwannomas. Clin Radiol 60(8):886–893. doi:10.1016/j.crad.2005.01.016 52. Cerofolini E, Landi A, DeSantis G, et al. (1991) MR of benign peripheral nerve sheath tumors. J Comput Assist Tomogr 15(4): 593–597 53. Murphey MD, Smith WS, Smith SE, Kransdorf MJ, Temple HT (1999) From the archives of the AFIP. Imaging of musculoskeletal neurogenic tumors: radiologic-pathologic correlation. Radiographics 19(5):1253–1280. doi:10.1148/radiographics.19.5.g99se 101253 54. Ros PR, Eshaghi N (1991) Plexiform neurofibroma of the pelvis: CT and MRI findings. Magn Reson Imaging 9(3):463–465 55. Korf BR (1999) Plexiform neurofibromas. Am J Med Genet 89(1):31–37 56. Woodruff JM, Selig AM, Crowley K, Allen PW (1994) Schwannoma (neurilemoma) with malignant transformation. A rare, distinctive peripheral nerve tumor. Am J Surg Pathol 18(9):882–895 57. Wasa J, Nishida Y, Tsukushi S, et al. (2010) MRI features in the differentiation of malignant peripheral nerve sheath tumors and neurofibromas. AJR Am J Roentgenol 194(6):1568–1574. doi: 10.2214/AJR.09.2724 58. Hrehorovich PA, Franke HR, Maximin S, Caracta P (2003) Malignant peripheral nerve sheath tumor. Radiographics 23(3): 790–794. doi:10.1148/rg.233025153 59. Tucker T, Friedman JM, Friedrich RE, et al. (2009) Longitudinal study of neurofibromatosis 1 associated plexiform neurofibromas. J Med Genet 46(2):81–85. doi:10.1136/jmg.2008.061051 60. Matsumine A, Kusuzaki K, Nakamura T, et al. (2009) Differentiation between neurofibromas and malignant peripheral nerve sheath tumors in neurofibromatosis 1 evaluated by MRI. J Cancer Res Clin Oncol 135(7):891–900. doi:10.1007/s00432-008-0523-y 61. Ireland DC, Soule EH, Ivins JC (1973) Myxoma of somatic soft tissues. A report of 58 patients, 3 with multiple tumors and fibrous dysplasia of bone. Mayo Clin Proc 48(6):401–410 62. Shanbhogue AK, Fasih N, Macdonald DB, et al. (2012) Uncommon primary pelvic retroperitoneal masses in adults: a patternbased imaging approach. Radiographics 32(3):795–817. doi: 10.1148/rg.323115020 63. Walker EA, Fenton ME, Salesky JS, Murphey MD (2011) Magnetic resonance imaging of benign soft tissue neoplasms in adults. Radiol Clin North Am 49(6):1197–1217. doi:10.1016/j.rcl.2011. 07.007 64. Luong TV, Salvagni S, Bordi C (2005) Presacral carcinoid tumour. Review of the literature and report of a clinically malignant case. Dig Liver Dis 37(4):278–281. doi:10.1016/j.dld.2004.10.014 65. Strupas K, Poskus E, Ambrazevicius M (2011) Retrorectal tumours: literature review and vilnius university hospital ‘‘santariskiu klinikos’’ experience of 14 cases. Eur J Med Res 16(5):231–236 66. Llauger J, Palmer J, Amores S, Bague S, Camins A (2000) Primary tumors of the sacrum: diagnostic imaging. AJR Am J Roentgenol 174(2):417–424. doi:10.2214/ajr.174.2.1740417 67. Gerber S, Ollivier L, Leclere J, et al. (2008) Imaging of sacral tumours. Skeletal Radiol 37(4):277–289. doi:10.1007/s00256-0070413-4 68. Bergh P, Kindblom LG, Gunterberg B, et al. (2000) Prognostic factors in chordoma of the sacrum and mobile spine: a study of 39 patients. Cancer 88(9):2122–2134

2644

H. Hosseini-Nik et al.: MR imaging of the retrorectal–presacral tumors

69. Sung MS, Lee GK, Kang HS, et al. (2005) Sacrococcygeal chordoma: MR imaging in 30 patients. Skeletal Radiol 34(2):87–94. doi: 10.1007/s00256-004-0840-4 70. Farsad K, Kattapuram SV, Sacknoff R, Ono J, Nielsen GP (2009) Sacral chordoma. Radiographics 29(5):1525–1530. doi:10.1148/rg. 295085215 71. Murphey MD, Nomikos GC, Flemming DJ, et al. (2001) From the archives of AFIP. Imaging of giant cell tumor and giant cell reparative granuloma of bone: radiologic-pathologic correlation. Radiographics 21(5):1283–1309. doi:10.1148/radiographics.21.5.g01se 251283 72. Liu G, Wu G, Ghimire P, Pang H, Zhang Z (2013) Primary spinal chondrosarcoma: radiological manifestations with histopathological correlation in eight patients and literature review. Clin Imaging 37(1):124–133. doi:10.1016/j.clinimag.2012.02.010 73. Wippold FJ 2nd, Smirniotopoulos JG, Moran CJ, Suojanen JN, Vollmer DG (1995) MR imaging of myxopapillary ependymoma: findings and value to determine extent of tumor and its relation to intraspinal structures. AJR Am J Roentgenol 165(5):1263–1267. doi:10.2214/ajr.165.5.7572515 74. Shors SM, Jones TA, Jhaveri MD, Huckman MS (2006) Best cases from the AFIP: myxopapillary ependymoma of the sacrum. Radiographics 26(Suppl 1):S111–S116. doi:10.1148/rg.26si065020 75. Johnson AR, Ros PR, Hjermstad BM (1986) Tailgut cyst: diagnosis with CT and sonography. AJR Am J Roentgenol 147(6):1309– 1311. doi:10.2214/ajr.147.6.1309 76. Yang DM, Park CH, Jin W, et al. (2005) Tailgut cyst: MRI evaluation. AJR Am J Roentgenol 184(5):1519–1523. doi:10.2214/ ajr.184.5.01841519 77. Aflalo-Hazan V, Rousset P, Mourra N, et al. (2008) Tailgut cysts: MRI findings. Eur Radiol 18(11):2586–2593. doi:10.1007/s00330008-1028-4 78. Kim MJ, Kim WH, Kim NK, et al. (1997) Tailgut cyst: multilocular cystic appearance on MRI. J Comput Assist Tomogr 21(5):731–732 79. Moulopoulos LA, Karvouni E, Kehagias D, et al. (1999) MR imaging of complex tail-gut cysts. Clin Radiol 54(2):118–122 80. Lim KE, Hsu WC, Wang CR (1998) Tailgut cyst with malignancy: MR imaging findings. AJR Am J Roentgenol 170(6):1488–1490. doi:10.2214/ajr.170.6.9609159 81. Mathis KL, Dozois EJ, Grewal MS, et al. (2010) Malignant risk and surgical outcomes of presacral tailgut cysts. Br J Surg 97(4): 575–579. doi:10.1002/bjs.6915 82. Yousefzadeh DK, Bickers GH, Jackson JH Jr, Benton C (1983) Tubular colonic duplication–review of 1876–1981 literature. Pediatr Radiol 13(2):65–71 83. Beattie CH, Garvey CJ, Hershman MJ (1999) Endorectal magnetic resonance imaging of a rectal duplication cyst. Br J Radiol 72(861):896–898

84. Gibson TC, Edwards JM, Shafiq S (1986) Carcinoma arising in a rectal duplication cyst. Br J Surg 73(5):377 85. Yang DM, Yoon MH, Kim HS, et al. (2001) Presacral epidermoid cyst: imaging findings with histopathologic correlation. Abdom Imaging 26(1):79–82 86. Fujimoto H, Murakami K, Kashimada A, et al. (1993) Large epidermal cyst involving the ischiorectal fossa: MR demonstration. Clin Imaging 17(2):146–148 87. Villarejo F, Scavone C, Blazquez MG, et al. (1983) Anterior sacral meningocele: review of the literature. Surg Neurol 19(1):57–71 88. Freeman HJ, Perry T, Webber DL, Chang SD, Loh MY (2010) Mucinous carcinoma in Crohn’s disease originating in a fistulous tract. World J Gastrointest Oncol 2(7):307–310. doi:10.4251/ wjgo.v2.i7.307 89. Iesalnieks I, Gaertner WB, Glass H, et al. (2010) Fistula-associated anal adenocarcinoma in Crohn’s disease. Inflamm Bowel Dis 16(10):1643–1648. doi:10.1002/ibd.21228 90. Smith R, Hicks D, Tomljanovich PI, et al. (2008) Adenocarcinoma arising from chronic perianal Crohn’s disease: case report and review of the literature. Am Surg 74(1):59–61 91. Kim MJ, Park JS, Park SI, et al. (2003) Accuracy in differentiation of mucinous and nonmucinous rectal carcinoma on MR imaging. J Comput Assist Tomogr 27(1):48–55 92. Hussain SM, Outwater EK, Siegelman ES (1999) Mucinous versus nonmucinous rectal carcinomas: differentiation with MR imaging. Radiology 213(1):79–85. doi:10.1148/radiology.213.1.r99se3879 93. Hama Y, Makita K, Yamana T, Dodanuki K (2006) Mucinous adenocarcinoma arising from fistula in ano: MRI findings. AJR Am J Roentgenol 187(2):517–521. doi:10.2214/AJR.05.0011 94. Fuchs B, Dickey ID, Yaszemski MJ, Inwards CY, Sim FH (2005) Operative management of sacral chordoma. J Bone Joint Surg Am 87(10):2211–2216. doi:10.2106/JBJS.D.02693 95. Woodfield JC, Chalmers AG, Phillips N, Sagar PM (2008) Algorithms for the surgical management of retrorectal tumours. Br J Surg 95(2):214–221. doi:10.1002/bjs.5931 96. Dozois EJ, Privitera A, Holubar SD, et al. (2011) High sacrectomy for locally recurrent rectal cancer: can long-term survival be achieved? J Surg Oncol 103(2):105–109. doi:10.1002/jso.21774 97. Nakai S, Yoshizawa H, Kobayashi S, Maeda K, Okumura Y (2000) Anorectal and bladder function after sacrifice of the sacral nerves. Spine 25(17):2234–2239 98. Davies AM, Hall AD, Strouhal PD, Evans N, Grimer RJ (2004) The MR imaging appearances and natural history of seromas following excision of soft tissue tumours. Eur Radiol 14(7):1196–1202. doi:10.1007/s00330-004-2255-y