Density (Hounsfield units)
Air
−1000
Fat
−30 to −100
Simple fluid
−10 to 20
Water
0
Blood
30–50
Muscle
45
Cortical bone
200–600
A limitation often encountered with CT imaging is that it is difficult to distinguish many biologic tissues since they have similar densities. The administration of different contrast agents introduces other densities that better define and distinguish biologic tissues. For example, when evaluating the GI tract, one of two different types of enteric contrast can be used depending on the type of pathology that is suspected. When evaluating the bowel wall for enhancement, luminal narrowing, and inflammatory changes, a neutral density or “negative” enteric contrast is used. This type of contrast distends the bowel lumen and, because it is less dense than the adjacent bowel wall, allows evaluation of their mural enhancement patterns [5]. When evaluating for obstruction, leak, or differentiation of the bowel from adjacent structures, a high density or “positive” enteric contrast is used. This type of contrast makes the bowel stand out against less dense abdominal structures and helps to assess transit through the bowel [6]. It is important to remember that positive contrast can linger in the bowel for several days and follow-up studies may be affected by residual contrast [7].
Another type of contrast that can be used to differentiate tissues on CT scans is intravenous contrast. Although many brands are available, they all contain iodine, a relatively dense atom with properties that result in increased absorption of X-rays. IV contrast is typically administered through a peripheral vein and increases the density wherever it accumulates. Initially it is located in the intravascular space, allowing studies such as arteriograms to be acquired if the scan is performed at the appropriate time after contrast is given. As contrast passes through capillary beds, some of it leaks out of the vessels into most of the interstitial tissues of the body, increasing their density to different degrees, helping to differentiate them on the scan. All living tissue enhances, a fact that helps in the characterization of nonliving tissue such as cysts, abscesses, and necrosis. Intravenous contrast is eventually eliminated through the kidneys.
CT is commonly used in imaging anorectal disease because of its widespread availability and relatively lower cost when compared to MRI. However, CT uses ionizing radiation like any other X-ray technique; a fact that needs to be considered when deciding among the different imaging modalities. CT is often used as a screening exam for a multitude of abdominal and pelvic pathologies, both in the hospital and in the outpatient setting. As a result of this, anorectal disease is often first identified on a CT scan as an incidental finding [8].
Scanning methods have been developed to screen the colon for polyps and cancer. Colonographic screening can identify previously unknown rectal masses. CT colonography is a low-dose screening technique which provides anatomic detail of the colon and rectum without the inherent invasiveness of a standard colonoscopy. Although no colonoscope is used in the procedure, adequate preparation and colonic distention is crucial to achieve a diagnostic result [9]. A full discussion of this technique is beyond the scope of this chapter.
3.2 Magnetic Resonance Imaging
MRI is an imaging technique that generates images based on the behavior of the hydrogen atoms in water molecules. The majority of the human body is composed of water molecules. When placed in the strong magnetic field of the MR machine, the hydrogen nuclei tend to align either with the field, in a lower-energy state, or against the field, in a higher-energy state. At equilibrium, there are more nuclei in the lower-energy state. When radio-frequency (RF) energy is applied at specific frequencies, that energy is absorbed by the low-energy nuclei, raising them into the higher-energy state. As they relax back to the lower-energy state, they release RF energy which is recorded by the system as the MRI signal. This signal is detected and reconstructed into images through a process called spatial localization. The result is an image that shows the amount of signal returning from each region of the tissue represented on the image with brightness that is proportional to the amount of signal returning [10].
MRI images can be acquired in multiple planes. A typical MRI examination consists of a combination of multiple views of the area of interest, each obtained to emphasize different components of the tissue and/or obtained in different orientations. These views are referred to as imaging sequences. They can be tailored to emphasize or eliminate fluid or fat. In addition, both intraluminal and intravascular contrast can be utilized to further enhance tissue differences in similar ways to those previously described for CT [11, 12]. It is important to use proper terminology when describing MRI findings. In general, things that appear brighter or more white are said to have “increased signal intensity” or be “hyperintense,” and things that appear darker or more black are said to have “decreased signal intensity” or be “hypointense.” Two common MRI sequences are referred to as “T1” and “T2” weighted. Only a few things appear hyperintense on T1-weighted sequences including fat, some hemorrhage, melanin, protein, and gadolinium-based contrast agents. Simple fluid is typically hypointense on T1-weighted sequences. T2-weighed images are “fluid sensitive,” and many water-containing substances typically appear hyperintense including fluid, fat, edema, and tumor (Table 3.2). Air has no signal and is black on both sequences. If desired, the signal from fat can be suppressed or eliminated from T2-weighted sequences in order to differentiate between fluid and fat. When trying to decide if a sequence is T1 or T2 weighted, look for structures that usually contain fluid such as the bladder or spinal canal; if they are hyperintense, the sequence is likely T2 weighted. Intravenous contrast, which is gadolinium based, can be added to T1-weighted sequences to image vessels or look for abnormal enhancement. Like CT, tissues with leakier capillary beds, such as those in areas of inflammation and tumor, will tend to accumulate more contrast and enhance more avidly than normal tissue [13].
Table 3.2
General intensity on MR images
T1 appearance | T2 appearance | |
|---|---|---|
Air | Black | Black |
Fata | Hyperintense | Hyperintense |
Simple fluid | Hypointense | Hyperintense |
Hemorrhagic fluid | Hyperintense | Hypointense |
Muscle | Hypointense | Hypointense |
Cortical bone | Black | Black |
Standard rectal protocol MRI images are oriented in three planes to the rectum (axial, sagittal, and coronal). In the setting of cancer, the most important images for staging are high-resolution T2-weighted images that are oriented perpendicular to the long axis of the rectum. Air within the rectal lumen appears black on MRI. Certain lesions, especially small or polypoid masses, can be difficult to identify in the collapsed rectum. In order to accentuate the rectal wall or rectal lesions extending into the lumen, an aqueous gel can be used as a luminal contrast agent to fill the cavity. This appears very hyperintense on T2-weighted images and is hypointense on T1-weighted images [13]. In other cases, especially those with sessile lesions, overdistention of the rectum by the gel can cause underestimation of tumor size and involvement. Advanced techniques such as diffusion-weighted imaging (DWI) may be used for identification of subtle lesions and pelvic lymphadenopathy [13].
Some of the advantages of MR imaging for evaluating anorectal disease include superior resolution, increased anatomic detail, and imaging without ionizing radiation. Often, MRI can identify disease characteristics that suggest a specific pathology or pathologic subtype. Although there are many advantages to MRI, disadvantages include higher cost, longer exam times, and technical limitations that can result in imaging artifacts. MR is not considered safe in patients with certain implanted mechanical devices. Fortunately, developing technical innovations and improvements in methodologies continue to reduce these disadvantages.
3.3 Imaging Anatomy
The anal canal is the channel extending from the perineum, at the anal verge, cranially to the anorectal junction, where the rectal ampulla narrows at the puborectalis sling. The dentate line, an important morphologic landmark, can’t be seen on MRI but is located in the upper anus and is the level of transition between the rectal mucosa of the upper anus and the squamous epithelium of the lower anus [14]. The internal anal sphincter (IAS) which is formed by the circular muscle layer of the rectal wall appears hypointense on T2-weighted MRI. The muscular external anal sphincter (EAS), the inferiormost extension of the levator ani, also appears hypointense on T2-weighted MRI (Fig. 3.1a).
Fig. 3.1
Normal rectal anatomy on T2-weighted MR images. Coronal (a) and axial (b) images through the lower rectum, axial image through the mid-rectum (c) and sagittal near-midline image (d) including the upper rectum. The sphincter complex is seen in (a) including the levator ani (white arrow) which is contiguous with the external anal sphincter (white star) that surrounds the internal anal sphincter (white circle). The layers of the rectal wall are seen in coronal (a) and axial (b) including the outer T2 hyperintense mesorectal fat (black star), T2 hypointense muscularis propria (dashed black arrow), T2 intermediate to hyperintense submucosa (dashed white arrow), and thin T2 hypointense mucosa (black arrow). The mesorectal fascia (black arrow heads) is shown in the mid-rectum (c). The thin hypointense band representing the peritoneal reflection (white arrowheads) is seen on the sagittal image (d)
The rectum extends from the anorectal ring approximately 15 cm cephalad to about the level of the sacral promontory and is divided into upper, middle, and lower thirds. The five layers of the rectum, the mucosa, submucosa, muscularis propria, mesorectal fat, and mesorectal fascia, can be readily distinguished on a good-quality scan. On T2-weighted MR images, the rectal mucosa is represented by a thin hypointense line adjacent to the rectal lumen. The underlying submucosa is higher in signal intensity. The muscularis propria appears as a thin hypointense line surrounding the submucosa. The mesorectal fat, as most fat, appears hyperintense on both T1- and T2-weighted imaging. Finally, the mesorectal fascia is a thin hypointense line surrounding the mesorectal fat [13] (Fig. 3.1). Although all five layers are present through most of the rectum, the anatomy varies at both ends. The mesorectal fat and fascia are not present around the uppermost portion of the rectum, above the level of the peritoneal reflection, or around the lowest portion of the rectum as it transitions into the anus.
3.4 Anorectal Neoplasms
Imaging evaluation of primary anal and rectal carcinoma is currently performed with high-field pelvic MRI using dedicated rectal protocols that provide high-resolution, small field-of-view images. In everyday practice, this technique is supplanting previously standard endorectal ultrasound in the initial staging of disease.
3.4.1 Rectal Adenocarcinoma
As stated previously, the high inherent tissue contrast of MRI in combination with the high resolution possible with newer techniques allows visualization of the stratified rectal wall (Fig. 3.1). This level of detail makes it possible to stage nonlocally advanced rectal carcinoma (T1–T3) based on the depth of invasion of the tumor into the rectal wall. In many clinical cases, the distinction between T1 and T2 disease cannot be reliably made by MRI. However, the more clinically relevant distinction between those lower-grade tumors and T3 disease can usually be made. MRI is also excellent for the identification and evaluation of the local invasion found in T4 disease.
Originating in the mucosal layer, rectal adenocarcinomas are seen as intermediate-low signal lesions on T2-weighted images. Typically, they are hyperintense relative to muscle and hypointense relative to fat. The depth of invasion, best visualized on T2-weighted images obtained perpendicular to the rectal wall, corresponds to the tumor stage on MRI evaluation. T1-stage tumors may extend into the submucosal layer of the bowel wall, whereas T2-stage tumors extend into but not beyond the muscularis propria. Tumors that extend beyond this point, into the mesorectal fat, are designated T3 stage and are sometimes subtyped depending on their depth of invasion into the fat. When tumors invade into adjacent pelvic organs or extend into the peritoneal fat, they are T4 stage. Some modification of this system is required for tumors that extend very low or very high in the rectum [15] (Fig. 3.2).
Fig. 3.2
Rectal cancer T-staging. Axial T2-weighted images through the rectum (a) showing a T1 tumor (black arrow) without invasion into the rectal wall. (b) A T2 tumor (thick black arrow) disrupting (white arrow) the rectal mucosa (white arrow head). (c) A T3 tumor with nodular extension into the mesorectal fat (white dashed arrows). (d) A T4 tumor (white stars) extending through the mesorectal fascia into the prostate (short white arrows)
3.4.2 Circumferential Resection Margin (CRM)
Among the more important measurements provided by the radiologist, the radiographic CRM is the shortest radial distance from the tumor or a tumor deposit in the mesorectal fat to the mesorectal fascia. A CRM of 1 mm or less is considered involving the mesorectal fascia, increasing the likelihood that the mass will not be completely resected with a typical mesorectal excision [16–18].
3.4.3 Low Rectal Cancer
Because of the anatomic changes present in the bowel wall as the rectum transitions into the anus, the classification system described previously does not necessarily apply here. At this level, the mesorectal fat thins and ends at the puborectalis sling. Tumors that extend below this level, to the internal anal sphincter, do not fit this staging system. Taylor et al. propose a staging system for these tumors that is based on the degree of extension through the muscle layers [15]. In this system, stage 1 tumors involve the rectal wall but do not extend into the sphincter complex. Stage 2 tumors extend into but not beyond the internal anal sphincter. Stage 3 tumors extend into the intersphincteric space and stage 4 tumors invade the external anal sphincter and beyond (Fig. 3.3).
Fig. 3.3
Coronal (a) and axial (b) T2-weighted MR images showing a stage 4 low rectal tumor (white star) with extension into the IAS (white arrow heads), intersphincteric space (white arrows), and the EAS (black arrow heads)
3.4.4 High Rectal Cancer
Similarly, tumors that extend very high in the rectum require different evaluations when staging. If a high rectal tumor is above the level of the peritoneal reflection, any extension beyond the muscularis propria involves the peritoneal space, similar to typical colon carcinomas. It has been suggested that upper rectal cancers can be treated like colon cancer with limited benefit of neoadjuvant radiation [19]. It has also been suggested that these high intraperitoneal tumors are less aggressive, stressing the importance of localizing them before treatment [20]. On MR imaging, the peritoneal reflection can sometimes be seen directly. When it is not seen, its location can be inferred at the superior margin of the mesorectal fat. Since there are implications on surgical planning for resection, localization of the tumor in relation to the peritoneal reflection is important in the radiologic evaluation of these masses. Relationship to the sacral promontory can be easily determined and is important to report since it can be a helpful landmark during surgery as well.
3.4.5 Lymph Nodes
MRI excels at providing the anatomic detail needed to identify lymph nodes in the perirectal/mesorectal tissue and more distant locations in the pelvis and retroperitoneum. The difficulty is in the distinction between normal and abnormal nodes. Abdominal and pelvic lymph nodes are usually considered abnormal on cross-sectional imaging studies if their short axis measures 1 cm or more since statistically, these nodes are significantly more likely to be abnormal. However, in rectal cancer, the size distribution of normal and abnormal nodes shifts; there is a much higher percentage of abnormal nodes harboring metastatic disease with short axes less than 1 cm [21].
In order to increase the likelihood of correctly identifying abnormal lymph nodes, the evaluation of these nodes takes into account not only size but morphologic characteristics as well. Imaging features that make a node suspicious for neoplastic involvement include irregular borders and heterogeneous signal intensity [22–24] (Fig. 3.4). Nodal staging in rectal cancer is based on the number of abnormal nodes present. N1 disease is characterized by the presence of 1–3 abnormal nodes; four or more abnormal nodes indicate N2 disease [15].
Fig. 3.4
Axial T2-weighted MR image through the rectum (a) showing a small smoothly marginated, oval mesorectal lymph node (white arrow) which is likely benign, inferior to an enlarged, irregularly marginated rounded lymph node (black arrow) which is likely malignant. A more caudal axial image (b) shows a similar-appearing malignant lymph node (black arrow head) and an opposite tumor nodule (white star). The corresponding slice from the apparent diffusion coefficient map (c) derived from the diffusion-weighted images (not shown) shows low signal within both lesions (dashed white arrows) indicating restricted water diffusion, compatible with tumor
3.4.6 Vascular Invasion
Tumor growth that extends beyond the muscularis propria into small vessels associated with the rectum in the region of the tumor is associated with poor prognosis, independent of other factors [25, 26]. High-resolution MR imaging can identify these small vessels and can evaluate them for the absence of flow and changes in morphology that suggest vascular invasion. It is also possible on post-contrast scans to distinguish tumoral enhancement seen in invasion from normal vascular opacification.
3.4.7 Mucinous Tumors
Although staged similarly, mucinous tumors of the rectum typically have a markedly different appearance on T2-weighted imaging because of their mucin content. Mucin, having high water content, is usually hyperintense and these mucinous tumors usually appear as multiseptated cystic masses (Fig. 3.5). Recent data suggests that patients with these tumors may not benefit from neoadjuvant therapy [27]. The ability of MRI to characterize these lesions allows this to be taken into consideration at the time of treatment planning.
Fig. 3.5
High-resolution axial (a) and sagittal (b) T2-weighted images through the lower rectum showing a large multiloculated high-signal mucinous neoplasm extending into the anal canal. Multiple thin septations within the lesion are hypointense
3.4.8 Surgical Planning
In addition to traditional small field-of-view T2-weighted imaging performed to evaluate the stage of the tumor, at our institution, we utilize additional large-field T2-weighted imaging and post-contrast T1-weighted imaging in multiple planes. These images help identify tumor involvement with adjacent structures. This can aid the surgeon in planning complex resections near the spine and the pelvic sidewall and in and around the sciatic foramen (Fig. 3.6).
Fig. 3.6
Post-contrast T1-weighted axial MRI images showing (a) direct tumor extension of tumor into the mesorectal fat and inflammation extending to the pelvic sidewall. Recurrent adenocarcinoma (b) involving the pelvic sidewall to the hypointense cortical bone (short arrow) without extension into the greater sciatic notch (long arrow)
3.4.9 Posttreatment
Patients who have undergone neoadjuvant therapy, typically those with T3 tumors and/or those with N1 or N2 disease, are usually restaged with pelvic MRI. The extent of disease is reassessed using the same criteria used in the primary evaluation. Downstaging after treatment can result in modification of the patient’s treatment plan [28].
3.4.10 Anal Carcinoma
As opposed to rectal carcinoma staging, where depth of invasion determines the stage, anal cancer is primarily staged based on its size, with the length of the long axis of the tumor determining the T1–T3 stage. Less than 2 cm, 2–5 cm, and greater than 5 cm correspond to stages 1, 2, and 3, respectively [29]. The stage is unaffected by involvement of local structures such as the sphincter complex, rectum, or anal verge. Involvement of distal structures, however, such as the bladder and vagina, does upstage the mass to stage 4 (Fig. 3.7).
Fig. 3.7
Anal carcinoma . Coronal (a) and axial (b) T2-weighted images through the anus showing an intermediate intensity T4N0 anal carcinoma involving the IAS (white arrow heads) on the right and extending into the expanded intersphincteric space (white star) and through the posterior, inferior aspect of the EAS (black arrowheads) into the gluteal fat and finally through the skin (not shown)
3.4.11 Lymph Node Staging
Also contrary to rectal carcinoma staging, nodal staging of anal cancer is not based on the number of involved nodes but rather their distribution. So, involvement of perirectal nodes alone is considered N1; unilateral involvement of the inguinal or internal iliac nodes is N2 and involvement of both groups or bilateral involvement is considered N3 [30, 31].
3.4.12 Posttreatment Imaging
Posttreatment imaging can be quite challenging. After treatment findings can range from interval progression of disease to a complete radiologic response. Newer techniques, such as diffusion-weighted imaging, that assign image contrast based on the freedom of water movement within tissues can be used to help identify residual disease [14].
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
