An earlier incorrect version of this article appeared online. This article was corrected on July 30, 2024.

Anatomic Basis of Rectal Cancer Staging:

Clarifying Controversies and Misconceptions
Harmeet Kaur, MD* • Helena Gabriel, MD* • Muhammad O. Awiwi, MD • Ekta Maheshwari, MD • Camila Lopes Vendrami, MD
Tsuyoshi Konishi, MD, PhD • Melissa W. Taggart, MD • Michael Magnetta, MD, MS • Linda C. Kelahan, MD • Sonia Lee, MD
*H.K. and H.G. contributed equally to this work.
Author affiliations, funding, and conflicts of interest are listed at the end of this article.

Rectal MRI provides a detailed depiction of pelvic anatomy; specifically, the relationship of the tumor to key anatomic struc-
tures, including the mesorectal fascia, anterior peritoneal reflection, and sphincter complex. However, anatomic inconsisten-
cies, pitfalls, and confusion exist, which can have a strong impact on interpretation and treatment. These areas of confusion
include the definition of the rectum itself, specifically differentiation of the rectum from the anal canal and the sigmoid colon,
and delineation of the high versus low rectum. Other areas of confusion include the relative locations of the mesorectal fascia
and peritoneum and their significance in staging and treatment, the difference between the mesorectal fascia and circumfer-
ential resection margin, involvement of the sphincter complex, and evaluation of lateral pelvic lymph nodes. The impact of
these anatomic inconsistencies and sources of confusion is significant, given the importance of MRI in depicting the anatomic
relationship of the tumor to critical pelvic structures, to triage surgical resection and neoadjuvant chemoradiotherapy with the
goal of minimizing local recurrence. Evolving treatment paradigms also place MRI central in management of rectal cancer.
©
RSNA, 2024 • [Link]

Introduction perspectives in treatment of rectal cancer: prevention of local

As use of high-resolution rectal MRI in treatment of rectal recurrence (LR) and improvement in disease-free and over-
cancer continues to grow in popularity, a detailed under- all survival. High-resolution MRI, with its excellent depiction
standing of interpretive pitfalls, controversies related to rectal of the relationship of the primary tumor to pelvic structures,
and pelvic anatomy, staging, and new treatment paradigms is has had a particularly significant impact—through appropri-
essential for radiologists. ate deployment of nCRT and accurate surgical planning—in
The evolving treatment paradigms appear to be moving reducing LR.
management of rectal cancer in divergent directions: upfront The distinction between treatment of the upper and lower
total neoadjuvant therapy (TNT)—where chemotherapy is rectum is related to a growing appreciation of the fact that ana-
administered in the neoadjuvant setting along with neoadju- tomic factors—primarily involvement of the potential circum-
vant chemoradiotherapy (nCRT)—for a subset of lower rectal ferential resection margin (CRM)—rather than T and N catego-
tumors, while stage II and III upper rectal tumors above the ries are the key determinants of LR and surgical management.
anterior peritoneal reflection (APR) are increasingly being The current treatment paradigms are based on T and N status,
managed with only surgery. These approaches can be appreci- which while important for predicting prognostic outcomes—
ated by understanding two distinct but overlapping oncologic including overall and disease-free survival—have less impact

GASTROINTESTINAL IMAGING This copy is for personal

use only. To order copies,
contact reprints@[Link]
July 2024 Kaur and Gabriel et al

ble to those of tumors treated with both nCRT and surgery (3–
TestYour 5). Most of the tumors in which nCRT could be avoided were
Knowledge located above or at the APR. In contrast, most tumors below
the APR are near the MRF and sphincter and consequently
RadioGraphics 2024; 44(7):e230203 require nCRT or total neoadjuvant therapy. In summary, an-
[Link] atomic factors such as the location of the tumor in the upper
Content Codes: GI, OI or lower rectum, the relationship of the tumor to the potential
Abbreviations: AJCC = American Joint Committee on Cancer, APR = ante- surgical margin and anal sphincter, and high-risk prognostic
rior peritoneal reflection, CRM = circumferential resection margin, EMVI = factors (ie, T4 tumors, extramural vascular invasion, tumor
extramural vascular invasion, LPLN = lateral pelvic lymph node, LR = local deposits, and LPLN metastasis) are the prime determinants
recurrence, MRF = mesorectal fascia, nCRT = neoadjuvant chemoradiother-
apy, STO = sigmoid takeoff
of LR and the need for nCRT or total neoadjuvant therapy.
This is a radical departure from previous treatment guide-
TEACHING POINTS lines, where nCRT was suggested for all stage II and III rectal
„ The definition of the rectum varies between anatomists and clinicians cancers.
and among clinical specialties and societies. The surgical anal canal ex- In terms of mesorectal lymph nodes, the Optimierte Chi­
tends from the anal verge to the anorectal ring and is terminology used
rurgie und MRT (OCUM), QuickSilver, and MERCURY studies
by the AJCC and National Comprehensive Cancer Network. Describing
this region as the lower rectum (0–5 cm from the anal verge) can be con- have shown that lymph nodes assessed at MRI have no impact
fusing. Defining the rectum as extending from the anorectal ring to the on LR and can in large part be excluded in determination of
STO aligns most closely with clinically relevant and anatomic landmarks. the need for nCRT (5,6). Since it is well established that lymph
„ The APR divides the rectum into extraperitoneal (lower rectum) and node involvement in rectal cancer causes a significant de-
intraperitoneal (upper rectum) segments, with the section of the rec- cline in overall and disease-free survival rates, this approach
tum bridging the APR representing the midrectum. It also separates two
lymphatic drainage zones of the rectum, with only the extraperitoneal seemingly contradicts well-established oncologic principles.
rectum draining to lateral pelvic nodes, while the rectum above the APR Lymph node involvement in rectal cancer unequivocally
drains superiorly to the superior rectal and inferior mesenteric nodes. causes a decline in overall survival, but with rare exceptions
„ The MRF is a thin anatomic structure that envelops the rectum, me- has no impact on MRF involvement and hence on LR rates.
sorectal fat, nodes, and vascular-lymphatic supply to the rectum. It is These evolving treatment paradigms require specific ana-
circumferential below the APR and runs inferiorly over the surface of the
levator ani muscles, terminating at the top of the puborectalis muscle.
tomic delineation of the tumor relative to adjacent anatomic
Superior to the APR, the MRF is located posterolateral to the rectum till landmarks, thereby placing MRI—with its exquisite depiction
the level of the STO. of detailed anatomy—at the forefront of rectal cancer eval-
„ The point of breach of the muscularis propria by tumor typically oc- uation. Yet, anatomic nuances and inconsistencies exist in
curs at the base of the tumor and is ideally confirmed on multiplanar the literature and in practice, which can confound accurate
high-resolution T2-weighted images. This is the site where MRF or organ
involvement should be assessed.
interpretation. This article attempts to address these anatom-
„ All lymph nodes abutting the internal iliac artery and its branches up to
ic areas of confusion in staging rectal MRI using a consistent
the level of the infrapiriformis foramen are internal iliac nodes. Lymph format, which defines the issue and its impact on interpreta-
nodes lateral to the internal iliac vessels at the pelvic brim, posterior to tion and staging followed by a discussion clarifying the issue.
the external iliac vessels at the midpelvis, and inferior to the infrapiri- In areas where the inconsistencies have yet to be resolved, the
formis foramen are obturator nodes.
authors also provide reasonable approaches to various dilem-
mas based on the literature and experiences at tertiary-care
cancer centers.
on LR rates. The rectum above the APR is distinct from the rec-
tum below the APR in terms of proximity to critical anatomic Issue 1: Defining the Anal Canal
structures, pathways of lymphatic drainage, and propensity to There are confusing and overlapping definitions of the ana-
develop lateral pelvic lymph node (LPLN) metastasis; conse- tomic anal canal, surgical anal canal, and lower rectum.
quently, the APR is becoming an important landmark separat-
ing the rectum into regions with differing prognostic features, Impact
resulting in distinct treatment approaches. Clarity and consistency in terminology are important for
The excellent depiction with MRI of the relationship of the communication, clinical management, and comparability of
primary tumor to key pelvic anatomic structures—enabling trial populations.
accurate prediction of clear surgical margins—in conjunc-
tion with improvement of surgical techniques and nCRT has Discussion
produced phenomenal results, with a decline in the LR rate Inconsistencies in defining the anatomy in this region often
to less than 5% (1). Although currently neoadjuvant radiation stem from the lens through which various clinicians and spe-
therapy is deployed based on T and N category, its primary cialties view and use the anatomic information. The surgeon’s
impact is reduction of LR by 50%, with little or no effect on assessment is geared toward the type of surgical resection,
overall survival or disease-free survival (2). while radiation oncologists and medical oncologists focus on
A number of recent studies where upper or mid T3N+ tu- the propensity for development of peritoneal carcinomatosis
mors with a widely clear mesorectal fascia (MRF) were treated and LPLN metastasis. All these factors depend on the location
with upfront surgery showed LR rates of 2.3%–3%, compara- of the tumor within the rectum, as different portions of the

Volume 44 Number 7 2 [Link]

July 2024 Kaur and Gabriel et al

Figure 1. Anatomic and surgical anal canal in two different patients. (A) Sagittal T2-weighted MR
image shows the anatomic anal canal (white bracket), which extends from the dentate line (dotted
white line) to the anatomic anal verge. T2-hyperintense submucosal edema related to prior radi-
ation therapy facilitates definition of columnar rectal mucosa extending to the dentate line. The
anatomic anal canal lined by squamous epithelium appears relatively T2 hypointense. (B) Sagittal
T2-weighted image shows the surgical anal canal (yellow bracket), which is a functional structure
defined by the external anal sphincter extending from the anorectal ring to the anal verge, as
defined by the inferior edge of the intersphincteric groove (blue arrow). It has two components:
the distal portion of the anatomic rectum (blue bracket), which extends from the anorectal ring
to the dentate line (dotted white line), and the proximal portion of the anatomic anal canal (white
bracket), extending from the dentate line to the inferior edge of the intersphincteric groove.

rectum and anal canal have distinct anatomic relationships preferred and most reproducible landmark for defining the
and arterial-venous and lymphatic pathways. MRI with the anal verge at MRI and by clinical palpation. It should be noted
ability to demonstrate the rectal tumor and its relationship to that this is different from the anatomic definition of the anal
the sphincter, peritoneum, anorectal ring, bony landmarks, verge (Fig 1) (7).
and so on provides portable images accessible to all special- The surgical anal canal contains the anatomic anal canal
ties. This allows a single definition of the rectum, ideally the (lower one-third to one-half ) and a portion of the anatomic
anatomic definition, with MR images used to address special- rectum, which lies above the dentate line (upper one-third
ty-specific clinical questions. to one-half of surgical anal canal). This explains how rectal
The anatomic anal canal lined by stratified squamous epi- ade­nocarcinomas may occur within the surgical anal canal
thelium arises from the embryologic ectoderm, with vascular (8).
supply from the middle and inferior rectal vessels and lym- The dentate line is occasionally seen at MRI as the junction
phatic drainage to the superficial inguinal nodes. Its superior of vertical T2-hyperintense columnar epithelium (columns of
border is the dentate line, which is the junction between the Morgagni) with T2-hypointense squamous epithelium (Fig 1)
columnar epithelium of the lower rectum and the squamous (9). The anorectal ring and its relationship to the lower edge
epithelium of the anal canal. The anal canal extends to the of a rectal tumor is critical in determining the possibility of
anatomic anal verge, defined as the junction of the anal and sphincter-saving surgery. On coronal T2-weighted MR imag-
perianal skin (Fig 1) (7,8). es, the puborectalis arises from the inferior aspect of the pu-
In contrast, the surgical anal canal is a functional—not bic symphysis, extends on either side of the rectum, and fuses
anatomic—unit defined by surgeons, which extends from the posteriorly, creating the anorectal angle (Fig 2). On sagittal
top of the puborectalis sling (palpable as the anorectal ring T2-weighted images, the plane of the anorectal ring is defined
at rectal examination) to the anal verge and is essential from by a line drawn from the anorectal angle to the inferior aspect
the perspective of maintaining continence. It is defined by the of the pubic symphysis (Fig 2).
sphincter complex, composed of (a) the internal sphincter, The current norm at MRI and endoscopy and in some clin-
which is the continuation of the circular muscle layer of the ical societies (Table 1) is to define the lower 0–5 cm from the
muscularis propria separated by the intersphincteric plane, anal verge as the lower rectum. However, this overlaps with
which is a continuation of the longitudinal muscle layer of the the anatomic and surgical anal canals, and these variable ter-
muscularis propria, and (b) the external sphincter, composed minologies are the source of considerable confusion. The use
of distal fibers of the levator ani muscle, puborectalis mus- of the surgical anal canal for this region may be the most clin-
cle, and small external sphincter muscles. The surgical anal ically relevant definition, as suggested by the American Joint
canal extends to the anal verge, clinically defined as the in- Committee on Cancer (AJCC) and National Comprehensive
ferior-most aspect of the intersphincteric groove. This is the Cancer Network (Table 1).

Volume 44 Number 7 3 [Link]

July 2024 Kaur and Gabriel et al

Figure 2. Demonstration of the anorectal ring or top of the puborectalis muscle in two
patients with rectal adenocarcinoma. (A) Coronal T2-weighted image shows the thickened
puborectalis muscle (white arrows) on either side of the rectum. The top of the puborectalis
(straight solid line) defines the top of the surgical anal canal and sphincter. A mucinous low
rectal cancer (arrowhead) extends to abut the top of the internal sphincter. Note the longitu-
dinal muscle fibers of the muscularis propria continuing as the intersphincteric plane (blue
arrow). The left MRF (dotted white line) ending at the top of the puborectalis is near the rectal
wall due to tapering mesorectal fat. (B) Sagittal T2-weighted image shows a line (dotted white
line) drawn from the inferior symphysis pubis to the anorectal angle (arrow), which defines
the plane of the anorectal ring—the top of the surgical anal canal and sphincter—on the
sagittal view.

Issue 2: Defining the Rectum ual surgeons for defining rectal boundaries: for the inferior
Definitions of the rectum vary, related to different approach- boundary, 43% used the anal verge, 19% the dentate line, and
es in defining the rectum, using either anatomic, measure- 15% the anorectal ring; for the superior boundary, 35% used
ment-based, or landmark-based criteria. the peritoneal reflection, 21% the rectosigmoid, and 30% the
distance from the verge (Fig 4) (18).
Impact The rectum can be defined relative to adjacent pelvic land-
Inconsistent definitions of the rectum can lead to miscatego- marks (eg, the sacral promontory or anorectal ring) or by us-
rization of anal, rectal, and sigmoid cancers, impacting indi- ing measurements from the anal verge. However, there is no
vidual treatment and comparison of clinical trial outcomes. agreement on which landmarks should be used to define the
borders of the rectum; furthermore, the landmark-based and
Discussion measurement-based definitions do not align. In an editorial
The definition of the rectum varies between anatomists and referencing unpublished data, Mathis and Nelson (19) com-
clinicians and among clinical specialties and societies (Table pared endoscopic measurement-based criteria with different
2) (10–17). landmarks and found discrepancies. For example, measure-
The anatomic rectum extends superiorly from the point ment-based criteria place the upper rectal limit 12–15 cm from
of fusion of the taenia coli—which are present in the sigmoid the anal verge. Landmark-based criteria for the upper rectal
colon and coalesce to form the circumferential longitudinal border, like the sacral promontory, were found to be located
muscle of the rectum—to the dentate line inferiorly. 21–23 cm from the anal verge, and coalescence of the taeniae
The sigmoid colon and rectum have different vascular sup- coli was located 21 cm from the verge. This would result in tu-
plies and pathways of lymphatic drainage and lymph node mors classified as sigmoid based on measurement criteria to
spread. The sigmoid receives its vascular supply from the sig- be localized to the upper rectum based on landmarks.
moid arteries, with lymphatic drainage and metastatic lymph Another limitation of measurement-based criteria in de-
nodes following these vessels, located in the sigmoid meso­ fining the rectum is related to significant variation in the
colon. Conversely, the entire rectum gets its vascular supply length of the rectum between individuals. A study during
from the superior rectal arteries, with the lower rectum re- open proctectomy found significant variation based on gen-
ceiving additional vascular supply from the middle rectal ves- der and body habitus in the distance from the anal verge to
sels; lymphatic drainage and metastatic lymph nodes from the dentate line (P = .003), puborectalis muscle (P = .03), or
the rectum follow these vessels (Fig 3). peritoneal reflection (P = .02). Consequently, categorization
While the anatomic definition of the rectum is unchang- of rectal tumor location based on fixed measurements would
ing, there are numerous surgical and clinical definitions result in variable localization of tumors relative to these clini-
(Table 2). There is even significant variability among individ- cally and anatomically relevant landmarks (Fig 5) (20).

Volume 44 Number 7 4 [Link]

July 2024 Kaur and Gabriel et al

Table 1: Summary of Key Points

Issue Summary Conclusion

Definition of anal The surgical anal canal is the most widely accepted defini- Distal border: anal verge (inferior aspect of the
canal tion of the anal canal intersphincteric groove)
Proximal border: anorectal ring
Definition of rectum Measurement-based definitions of the rectum have limita- Distal border: anorectal ring or top of puborectalis
tions related to individual variability in the length of the muscle
rectum and limited clinical relevance Proximal border: sigmoid takeoff (STO)
Definition of the rectum should be based on clinically rele-
vant and anatomic landmarks
Subdivision of rectum Subdivisions of 0–5 cm, 5–10 cm, and 10–15 cm have signifi- The APR subdivides the rectum into the upper
into lower, mid, and cant limitations (intraperitoneal), mid (at the level of the APR),
upper rectum The APR should be used to subdivide the rectum and lower (extraperitoneal) rectum
Location of the MRF The anterior peritoneum gradually encircles the upper rec- Above the APR, the MRF is located posterolateral-
relative to the tum as it ascends from the level of the APR to the STO ly and decreases as the rectum ascends and the
peritoneum above The peritoneum and MRF merge at the lateral angle of the peritoneum encircles the rectum
the APR rectum with no clear demarcation at MRI or CT Peritoneal involvement is T4a
MRF involvement impacts only the need for nCRT
and surgical planning, not the T category
Does the CRM = the The MRF is an anatomic plane The CRM should not be used on preoperative MRI
MRF? The CRM is the potential surgical plane along the nonperi- reports as a surrogate for the MRF
tonealized surface of the rectum
Criteria for MRF Tumor beyond the muscularis propria, EMVI, spiculated T3 tumors, EMVI, spiculated heterogeneous
involvement heterogeneous lymph nodes, or tumor deposits < 1 mm lymph nodes, and tumor deposits are assessed
from the MRF = MRF involvement for MRF involvement
Smooth heterogeneous or benign lymph nodes < 1 mm from Smooth encapsulated malignant or benign lymph
the MRF do not equate to MRF involvement nodes are not assessed for MRF involvement
T staging of anal The T category is a prognostic guide based on survival data, ESGAR and SAR suggest detailing components of
sphincter involve- not only on anatomic considerations sphincter involvement pending guidance from
ment The eighth edition of the AJCC manual currently does not the AJCC
provide specific guidelines
The CAP and some surgical societies upstage levator and
external sphincter involvement to T4b
Assessment of LPLN LPLN size criteria are applicable to T3 or T4 rectal tumors Pretreatment MRI size criteria: 5–10 mm (short
involvement below the APR axis) for LPLNs
Almost all metastatic LPLNs localize to the internal iliac and Posttreatment MRI size criteria: >4-mm internal
obturator node compartments iliac lymph nodes have a high risk of residual
Measurement of small LPLNs is subject to significant in- disease
terobserver variability
Changing treatment Upper rectum: Upper rectum: stage II or III tumors above the APR
paradigms Intraperitoneal are increasingly treated like sigmoid tumors
Lower incidence of MRF or organ involvement with upfront surgery, unless the posterior later-
Potential for T4a and peritoneal spread al MRF is involved
No LPLN metastasis Lower rectum: stage II or III tumors generally re-
Lower rectum: quire total neoadjuvant therapy (TNT) or nCRT
Extraperitoneal
Higher incidence of MRF or organ involvement
LPLN metastasis
Higher incidence of pulmonary metastasis
Note.—AJCC = American Joint Committee on Cancer, CAP = College of American Pathologists, EMVI = extramural vascular invasion, ESGAR =
European Society for Gastrointestinal and Abdominal Radiology, SAR = Society of Abdominal Radiology.

Several societies including the AJCC (Table 2) place the point where the rectum flexes anteriorly from the sacrum at
distal boundary of the rectum at the anorectal ring. The CT or MRI (Fig 6). The flexion away from the sacrum is relat-
proximal rectal boundary remains controversial (Table 2). ed to the mobility of the intraperitoneal sigmoid mesocolon
Recently, the sigmoid takeoff (STO) has been proposed as relative to the retroperitoneal mesorectum and serves as a
an imaging-based landmark for the junction of the sigmoid surrogate marker for the true ana­tomic junction of the rec-
mesocolon and mesorectum (21). The STO is defined as the tum and sigmoid. This flexion point can also be defined by

Volume 44 Number 7 5 [Link]

July 2024 Kaur and Gabriel et al

Table 2: Definitions of the Rectum

Proximal Rectal Boundary Distal Rectal Boundary Subdivisions of Rectum Organization or Other Source (Year)
<15 cm from anal verge Anorectal ring or top of … American Society of Colon and Rectal
pubo­rectalis ring Surgeons (ASCRS) (2020)*
<15 cm from anal verge Anal verge Lower (0–5 cm), mid (5–10 cm), and European Society for Medical Oncolo-
upper (10–15 cm) rectum gy (ESMO) (2017)
Line from sacral promontory Anorectal ring or top of APR defines lower (below), mid (at re- National Comprehensive Cancer
to top of pubic symphysis pubo­rectalis ring flection), and upper (above) rectum Network (NCCN) (2020)
Coalescence of taeniae and Anorectal ring or top of APR defines lower (below), mid (at re- American Joint Committee on Cancer
sacral promontory pubo­rectalis ring flection), and upper (above) rectum (AJCC) (2017)
Lower edge of S2 Anorectal ring or top of APR defines lower rectum (below) and Japanese Society for Cancer of the
pubo­rectalis ring upper rectum (above) Colon and Rectum (JSCCR)
<15 cm from anal verge Anal verge Lower (0–5 cm), mid (5–10 cm), and Union for International Cancer Con-
upper (10–15 cm) rectum trol (UICC) (2017)
<12 cm from anal verge Anal verge … National Cancer Institute (NCI)
(2000)
<15 cm from anal verge Anal verge Lower (0–5 cm), mid (5–10 cm), and Endoscopic and MRI-based measure-
upper (10–15 cm) rectum ments
Sources.—References 10–17.
* Reference 58.

vasculature—that is, the point where the sigmoid arteries in Discussion

the sigmoid mesocolon and the superior rectal arteries di- The APR simply defines a point of contact or reflection of the
verge (22,23). visceral peritoneum from the bladder and seminal vesicles or
A study to validate the STO found correlation between uterus onto the anterior rectal wall and is not interchange-
radiologic and pathologic findings (23). In a study of 100 pa- able with the visceral peritoneum.
tients, the STO had a wide range of 9.4–19 cm from the anal The APR divides the rectum into extraperitoneal (lower
verge, reflecting expected variability between individuals. If rectum) and intraperitoneal (upper rectum) segments, with
the standard cutoff of less than 15 cm for defining the rectum the section of the rectum bridging the APR representing the
is used, tumors classified as sigmoid using the STO would midrectum. It also separates two distinct lymphatic drainage
be classified as upper rectal in 84% of cases (Fig 7) (24). The zones of the rectum, with only the extraperitoneal rectum
Dutch 2019 guidelines have instituted use of the STO as the draining to lateral pelvic nodes, while the rectum above the
separation between the rectum and sigmoid (25). APR drains superiorly to the superior rectal and inferior mes-
However, the STO is not without problems: a recent audit of enteric nodes (Fig 3).
stakeholders treating rectal cancer found only 53% agreement The APR can be defined in 81.6%–88.5% of rectal MRI stud-
with expert reference localization of the STO, which improved ies (27). On axial T2-weighted images, the APR is seen as a
to only 70% after training, thus showing that there is a learn- V configuration—likened to a seagull configuration—at the
ing curve in implementing this landmark (Fig 6) (26). point where the peritoneum reflects onto the anterior rectum
In summary, the basic definition of the rectum is anatom- (Fig 8). The seminal vesicles and uterocervical angle serve as
ic; however, clinicians have modified the definition based on reliable landmarks for defining the APR (27). The distance
their clinical perspectives. Since measurement-based criteria from the APR to the anorectal ring and the relationship of
have limitations, MRI reports should include—in addition to rectal tumors to the APR has excellent interreader agreement
tumor measurement from the anal verge—the relationship of (27).
the lower edge of the tumor to (a) the top of the puborectalis Use of the APR to subdivide the rectum into the upper and
sling, which defines the lower edge of the rectum, or to (b) the lower rectum reflects the current National Comprehensive
STO, which separates the upper rectum and sigmoid. Cancer Network (NCCN), AJCC, and Japanese Society for Can-
cer of the Colon and Rectum (JSCCR) guidelines (Table 2) (Fig
Issue 3: Defining the APR and Its Significance 9). This approach is distinct from measurement-based subdi-
1. The APR subdivides the rectum into two clinically and ana- vision of the rectum into the low rectum at 0–5 cm, the mid-
tomically distinct regions. rectum at 5–10 cm, and the upper rectum at 10–15 cm, which
2. The distinction between the APR and the visceral perito- does not account for individual variability in length of the rec-
neum requires clarification. tum and anal canal (Fig 10).
The upper and lower rectum, separated by the APR, are
Impact distinct in terms of investment by peritoneum, lymphatic
The subdivision into the high and low rectum should be based drainage zones, and proximity to key anatomic structures
on the relationship of the APR to the rectum. like the MRF, inferior hypogastric plexus, levator ani, and
Volume 44 Number 7 6 [Link]
July 2024 Kaur and Gabriel et al

Figure 3. Vascular and lymphatic supply of the

anatomic sigmoid colon, rectum, and anal canal. The
sigmoid receives its vascular supply from sigmoid
arteries arising from the inferior mesenteric artery,
with lymphatic drainage and metastatic lymph nodes
following these vessels, located in the sigmoid meso­
colon. The entire rectum (blue shaded area with purple
dashed outline) gets its vascular supply from the
inferior mesenteric and superior rectal arteries, with
lymphatic drainage and adenopathy (green dotted lines
and lymph nodes) developing along these vessels. The
lower rectum below the APR (blue dashed line) receives
additional vascular supply from the middle rectal
vessels. The lymphatic drainage from the lower rectum
drains to the internal iliac and obturator lymph node
compartments (yellow shaded area). The anatomic anal
canal receives vascular supply from the middle and
inferior rectal vessels, and lymphatic drainage is to the
superficial inguinal nodes (not shown).

likelihood of MRF or levator and sphincter involvement (Figs

2A, 3). These tumors are predisposed to LPLN metastasis and
have a higher incidence of pulmonary metastasis because of
systemic venous drainage along the middle rectal veins (28).
For these reasons, most of these tumors require nCRT or total
neoadjuvant therapy.
In contrast, rectal tumors above the APR are unlikely to de-
velop LR due to the anatomic widening of the pelvis, leading
to distance between the tumor and vital pelvic structures (3–
5). Indeed, several trials have shown that the majority of stage
II or III rectal cancers above the APR can proceed directly to
surgery (29). In the Optimierte Chi­rurgie und MRT (OCUM)
trial, 89% of upper rectal tumors went directly to surgery with
no LR at 3 years (3,5,30). Tumors in the upper rectum, recto-
sigmoid, and sigmoid colon are increasingly being treated in
a similar manner with upfront surgery.

Figure 4. Sagittal T2-weighted image Issue 4: Relationship of MRF and Peritoneum

shows variable boundaries of the rectum. In- There are significant anatomic misconceptions about the re-
ferior border of the rectum: anal verge (black
lationship of the MRF relative to the peritoneum around the
dashed line) or anorectal ring (white dashed
line). Superior border of the rectum: sacral upper rectum.
promontory (red dashed line), 15 cm from
the anal verge (orange dashed line), sigmoid Impact
takeoff (STO) (arrow), or bottom of S2 (blue This can impact T staging, and if the MRF is involved, surgical
dashed line). planning and the need for nCRT.

Discussion
anal sphincter. Consequently, the relationship of the tumor to In a global online survey of more than 300 radiologists, there
the APR can affect management. The rectum below the APR is was insufficient agreement (73%–79%) among respondents
surrounded by tapering mesorectal fat, which increases the on T staging of tumors involving the peritoneum versus the
Volume 44 Number 7 7 [Link]
July 2024 Kaur and Gabriel et al

Figure 5. Variability in the length of the

anal canal and rectum between individuals.
(A) Sagittal T2-weighted image shows a
5.5-cm-long surgical anal canal (blue double
arrow) in a male patient. The APR (white
arrow) is 12 cm above the anal verge. The
lower rectum from the anorectal ring to the
APR is 6–7 cm. A small anterior rectal tumor
is incidentally noted (white dotted outline).
(B) Sagittal T2-weighted image shows a 2.5-
cm surgical anal canal (blue double arrow)
in a female patient. The APR (white arrow)
is located 11 cm above the anal verge. The
lower rectum from the anorectal ring to the
APR is 8–9 cm.

Figure 6. Demonstration of the STO in two

patients. (A) Sagittal T2-weighted image
shows the STO (white arrow) as the point
where the rectum flexes anteriorly from the
sacrum and becomes horizontal. This flexion
point often correlates with the point of diver-
gence of the sigmoid vessels (blue arrow) and
superior rectal arteries (red arrow). (B) Sagittal
T2-weighted image shows a 5-cm mucinous
tumor (arrowhead) bridging the STO (white
arrow). This prevents flexion of the sigmoid
and creates the impression that the STO is
located more proximally (black arrow). This is
a pitfall that can be avoided by using vascular
landmarks. The STO reflects the junction of
the sigmoid mesocolon and mesorectum and
is an accurate surrogate for the true anatomic
junction of the sigmoid colon and rectum, that
is, the confluence of the taeniae coli.

Figure 7. Differences in measurements versus anatomic landmarks for defining the rectum and
sigmoid colon. Sagittal T2-weighted image shows the distal margin of a rectal tumor (arrowhead)
located 13 cm above the anal verge. This is an upper rectal tumor based on measurement criteria
(10–15 cm). The tumor is located above the STO (arrow) and would be classified as a sigmoid
tumor based on using the STO as a landmark for the upper rectal border. Measurement guide-
lines place some sigmoid tumors into the high rectal group, leading to potential overtreatment.
Anatomic landmarks are a more reliable means of defining rectal segments and adjusting for
individual variability.

The lower rectum below the APR is surrounded by MRF. The

APR is where the rectum emerges from its extraperitoneal lo-
cation and acquires a peritoneal covering, which progressively
encircles the rectum. As the peritoneum gradually invests the
rectum anterolaterally, the MRF reduces and ends at the STO,
above which the entire sigmoid is encircled by peritoneum
except at the root of the sigmoid mesocolon (Fig 11). The peri-
toneum above the APR merges with the MRF laterally, but dis-
MRF. This seems to have been partially related to a lack of tinct demarcation between these two structures is often lack-
clarity related to the relative relationship of the MRF and peri- ing (Fig 11D). Radiologists should be cognizant of this fact and
toneum above the APR (31). that tumors may involve one or both structures in this region,

Volume 44 Number 7 8 [Link]

July 2024 Kaur and Gabriel et al

Figure 8. The APR in two patients. (A) Sag-

ittal T2-weighted image shows the APR (white
arrow). Peritoneum over the fundus of the
bladder covers the tips of the seminal vesicles
and reflects off the anterior rectal wall (arrow-
heads). The point of contact of the peritoneum
over the rectal wall is the APR. The seminal
vesicles (orange arrow) and uterocervical an-
gle serve as consistent landmarks for defining
the APR. (B) Axial T2-weighted image shows
the APR with its seagull appearance (arrows).

Figure 10. Differences in measurements

versus anatomic landmarks in defining the
upper, mid, and lower rectum in a 64-year-old
patient with rectal adenocarcinoma. Sagittal
T2-weighted image shows the distal margin
of a rectal adenocarcinoma 7 cm above the
anal verge. This is a midrectal tumor based
on measurement criteria (at 5–10 cm) and a
low rectal tumor based on landmark-based
criteria, as the tumor is located below the APR
(arrow), faintly seen just below the tips of the
seminal vesicles.

Figure 9. Sagittal T2-weighted image

shows the definition of the upper and lower
rectum. The inferior border is the green
dashed line extending from the inferior
pubic symphysis along the top of the anorec-
tal ring. The APR (arrowhead) separates the the rectum. It is circumferential below the APR (Fig 11B) and
upper rectum (white bracket) and lower inferiorly runs over the surface of the levator ani muscles, ter-
rectum (black bracket). The upper border is
minating at the top of the puborectalis muscle (Fig 2).
the STO (arrow).
The CRM is a potential surgical dissection plane along the
nonperitonealized surface of the rectum, which may or may
upstaging in cases of peritoneal involvement to a T4a tumor not follow the MRF, depending on the surgical procedure that
with potential for peritoneal spread (Fig 12). In contrast, MRF is performed.
involvement does not lead to upstaging of T3 tumors to T4b (32) Total mesorectal excision (TME) is the most common sur-
(Table 1). In assessing peritoneal involvement, it is important to gery performed for rectal tumors, characterized by dissec-
note that abutment does not equate to involvement. Defining tion along the MRF. Occasionally, the surgeon may deviate
the tumor base and points of tumor transgression beyond the from the “innermost dissectible plane” of the MRF if tumor
muscularis propria is a key step to avoid pitfalls (Fig 13). involves or extends beyond the MRF, necessitating a curative
resection margin that would need to extend beyond the MRF
Issue 5: Confusion regarding the CRM and MRF (33,34).
The CRM is often erroneously equated with the MRF. For low rectal tumors not eligible for a sphincter-saving
TME, the surgical approach may require an intersphincteric,
Impact extrasphincteric, or extralevator abdominoperineal resection
The CRM represents a variable potential surgical margin, de- (Fig 14). In each of these procedures, the CRM is different. The
pending on the type of surgery, while the MRF is a fixed ana- CRM does not apply to the portion of the rectum that is cov-
tomic structure. The two are not interchangeable. ered by peritoneum.

Discussion Issue 6: What Constitutes MRF Involvement?

The MRF is a thin anatomic structure that envelops the rec- What constitutes MRF involvement? Specifically, which struc-
tum, mesorectal fat, nodes, and vascular-lymphatic supply to tures—that is, the primary tumor, mesorectal lymph nodes,
Volume 44 Number 7 9 [Link]
July 2024 Kaur and Gabriel et al

extramural vascular invasion (EMVI), or tumor deposits— of imaging, a recent multidisciplinary expert consensus rec-
should be assessed relative to the MRF and how should the ommended that heterogeneous spiculated nodules in the
distance to the MRF be measured and defined? mesorectal fat could be either lymph nodes or tumor deposits
and should be treated differently from smooth well-encap-
Impact sulated mesorectal lymph nodes. The MRF should be consid-
Incorrectly diagnosed MRF involvement can result in over- or ered as involved if a heterogeneous spiculated lymph node or
undertreatment. tumor deposit is in proximity to the MRF, but not a smooth
encapsulated lymph node even if heterogeneous in appear-
Discussion ance and potentially metastatic (Fig 18) (31) (Table 1).
The MRF circumferentially surrounds the lower rectum, runs Accurately defining involvement of the MRF by the prima-
over the surface of the levators, terminates at the top of the ry tumor and EMVI or tumor deposits requires adherence to
puborectalis sling, fuses with the APR anteriorly, and ascends a few key steps:
in the lateral and presacral region above the APR. 1. MRF involvement should be assessed in T3 tumors at the
In terms of the primary tumor, MRF assessment is applica- base of the tumor, the most common site of transgression of
ble only in T3 tumors, not in T1 or T2 tumors. In low rectal tu- the muscularis propria (Fig 15). Assessment of the point of
mors extending into the anal canal, MRF involvement should breach of the muscularis propria by tumor signal intensity
be assessed only for the component of the tumor in the lower extending beyond the muscularis propria and approaching
rectum, as the MRF terminates at the top of the puborectalis the MRF, ideally confirmed on multiplanar high-resolution
sling. MRF involvement by the primary tumor is best assessed T2-weighted images, is essential. Occasionally, the intact rec-
at the base of the tumor, which is the most likely point of mus- tal wall adjacent to the tumor approximates the MRF and is
cularis propria transgression (Fig 15). misdiagnosed as MRF involvement. Tumor confined within
With respect to EMVI, a study by Birbeck et al (35) reported the muscularis propria at the point of abutment does not in-
that MRF involvement by EMVI resulted in a 30.4% LR rate, dicate MRF involvement (29).
while tumor deposit involvement of the MRF resulted in a 2. In terms of measurement from the MRF, the closest dis-
40.9% LR rate. On the basis of such data, a recent expert con- tance from the most penetrating component of the primary
sensus recommended that EMVI and tumor deposits within tumor, EMVI, or spiculated lymph node or tumor deposit
1 mm of the MRF should be reported as MRF involvement at should be obtained. MRF is described as involved when any
MRI (Figs 16, 17) (31). Additionally, EMVI or tumor deposits at of these structures are less than 1 mm from the MRF (Figs 15,
MRI have significant prognostic impact and are associated 16). Counterintuitively increasing the cutoff from 1 mm to 5
with a decline in overall survival and disease-free survival mm reduced accuracy due to higher false-positive results, re-
(36). sulting in patients receiving unnecessary neoadjuvant ther-
In terms of mesorectal lymph nodes, the Colorectal and apy (6).
Anal Disease-focused Panel of the Society of Abdominal Ra- 3. MRI has high interobserver agreement and negative pre-
diology has recommended that lymph nodes in proximity dictive value of 94%–98.1% in excluding MRF involvement.
to the MRF be described but not reported as MRF involve- The positive predictive value is only slightly lower at 86.5%.
ment (37). This recommendation recognizes that most visi- This has been shown consistently in multiple trials and stud-
ble lymph nodes are benign and that radiologic evaluation of ies over the decades and is a significant contributor to the suc-
lymph nodes is limited in its accuracy. Interestingly, a study cess of MRI in reducing LR (4,42).
by Nagtegaal et al (38) found that even tumor-bearing lymph
nodes involving the CRM did not confer any increased risk of Issue 7: Anal Sphincter Involvement and T Stage
LR. Shihab et al (39) found that it is uncommon for a mesorec- The eighth edition of the AJCC manual does not specify
tal lymph node to be the only factor responsible for a positive whether anal sphincter involvement should be considered
CRM at histopathologic analysis. T4b disease. However, the College of American Pathologists
The eighth edition of the AJCC manual defines tumor de- and some surgical societies have different recommendations.
posits as discontinuous nodules of tumor in the mesorectal
fat, with no identifiable nodal or vascular tissue. The definition Impact
of tumor deposits at TNM staging has changed between the There is a lack of clarity on T stage for rectal tumors involv-
fifth and eighth editions and continues to evolve, presenting ing the anal sphincter and conflicting recommendations from
challenges in their reporting. Tumor deposits were initially different expert groups.
described 90 years ago by Gabriel (40) to occur in conjunction
with EMVI as in-transit tumor thrombi forming a nodular mass Discussion
with a spiculated border contiguous with EMVI, an appearance The AJCC defines the T category by “the size and/or contigu-
described at MRI as the comet tail sign (41). ous extension of the primary tumor” and T4b when the tumor
Tumor deposits may also be completely replaced me- directly involves or adheres to adjacent organs or structures.
sorectal lymph nodes. However, the distinction between a Currently, the AJCC gives no guidance on what constitutes
metastatic partially replaced lymph node and a lymph node “other structures” (43).
entirely replaced by tumor is a pathologic diagnosis that can- The College of American Pathologists has taken a position
not be made with imaging. To accommodate this limitation that “invasion of the external sphincter and/or levator ani

Volume 44 Number 7 10 [Link]

July 2024 Kaur and Gabriel et al

Figure 11. Progressive envelopment of the upper rectum by

peritoneum. (A) Sagittal illustration shows the peritoneal cav-
ity (pink overlay) and peritoneum (curvilinear white line). (B–
F) Cross sections show the extraperitoneal or lower rectum,
the APR, and the relationship of the MRF to the peritoneum in
the upper rectum. The sigmoid mesocolon above the STO is
also depicted. There are corresponding MR or CT images for
each section. (B) Axial illustration and T2-weighted MR image
below the APR show the extraperitoneal or lower rectum sur-
rounded by MRF on the illustration (dark brown line) and MR
image (arrowheads). (C) Axial illustration and T2-weighted MR
image at the APR show the seagull appearance of the APR on
the illustration (white line) and MR image (arrow). An anterior
rectal tumor is near but not involving the APR. Arrowheads on
the MR image = MRF. (D) Axial illustration and T2-weighted MR
image above the APR. The illustration shows that peritoneum
coverage anteriorly (curved white line) has increased, with a
corresponding reduction in MRF (brown line). The MR image
shows smaller MRF posteriorly (arrowheads). The peritone-
um (blue arrows) is delineated by ascitic fluid, which allows
visualization of the lateral confluence of the peritoneum and
MRF (orange arrows). The uterus is seen anterior to the rec-
tum, and incidental note is made of a left ovarian mass (white
arrow). (E) Axial illustration and T2-weighted MR image of the
upper rectum just proximal to the STO. The illustration shows
peritoneum almost completely encircling the rectum, with a
small amount of presacral MRF (brown line) remaining. The
MR image shows the peritoneum outlined in a white line and
presacral MRF (arrowheads) posteriorly. (F) Axial illustration
and coronal contrast-enhanced CT image of the sigmoid colon
and mesocolon. The illustration shows the sigmoid mesoco-
lon with sigmoid arteries suspended in the peritoneal cavity
(pink shading) and entirely encircled by peritoneum, except
at the point of attachment of the mesocolon. The CT image
shows ascitic fluid outlining the sigmoid colon and fat and
arteries in the sigmoid mesocolon (arrow).

Volume 44 Number 7 11 [Link]

July 2024 Kaur and Gabriel et al

Figure 12. Combined peritoneal-MRF involvement in two patients with upper rectal tumors located
above the APR. (A) Axial T2-weighted image shows a large tumor deposit or spiculated lymph node
infiltrating the peritoneum (blue arrow), the junction of the MRF and peritoneum laterally (orange ar-
row), and right posterior MRF (arrowhead). (B) Coronal T2-weighted image shows a large upper rectal
tumor infiltrating through the muscularis propria along the left lateral wall. It involves the MRF posteri-
orly and laterally (arrowhead) and the peritoneum anteriorly (arrow) over the fundus of the bladder.

Figure 13. Pitfall in assessment of peritoneal involvement in a 64-year-old man with upper rectal
adenocarcinoma. (A) Sagittal T2-weighted image shows an upper rectal tumor (white arrow) above
the APR. The tumor appears to abut and possibly involve the peritoneum over the posterior bladder
wall (black arrow). (B) Axial T2-weighted image shows a posterior ulcerated tumor with rolled edg-
es (white outline). The anterior rectal wall is not involved by the tumor (blue arrow); consequently,
abutment of the peritoneum should not be misconstrued as involvement. The tumor infiltrates
through the muscularis propria at the base of the ulcer crater (white arrow). The erroneous impres-
sion on the sagittal image (A) is related to the plane of the image (blue line), which passes through
the rolled edge.

muscle(s) be classified as T4b,” although no reference is pro- for determining the surgical approach (14). If the levator or
vided to support this statement. It seems reasonable that oth- external sphincter is involved, an abdominoperineal resec-
er striated muscles not inherently related to the rectum would tion is needed, while in tumors confined to the mucosa or in-
constitute “additional structures” and that their involvement ternal sphincter, a sphincter-sparing surgery can potentially
would upstage rectal tumors to T4b (44). be performed.
A few surgical societies and international expert panels The current position of the European Society for Gastroin-
upstage levator or sphincter involvement to T4b, as tumor in- testinal and Abdominal Radiology (ESGAR) and the Colorectal
volvement of components of the anal sphincter is important and Anal Disease-focused Panel of the Society of Abdominal

Volume 44 Number 7 12 [Link]

July 2024 Kaur and Gabriel et al

Figure 14. The CRM does not equal the MRF. (A) Coronal oblique high-resolution T2-weighted image
show a low rectal adenocarcinoma (*) involving the external anal sphincter and puborectalis muscle.
The CRM (dashed line) will be expanded to include the levator ani, making extralevator resection most
appropriate. (B) Illustration shows a coronal plane cross section along the longitudinal axis of the anal
canal, with a low rectal adenocarcinoma (purple mass in the upper anal canal) involving the levator
ani and puborectalis muscle, in addition to the intersphincteric plane and external sphincter. The
curative resection margin (orange line) deviates from the MRF (blue line) in the lower pelvis to include
the levator ani and puborectalis muscle and the ischiorectal and ischioanal fat to ensure an adequate
resection margin.

Figure 16. MRF involvement by EMVI in a

59-year-old patient with rectal carcinoma.
Coronal T2-weighted image shows the prima-
ry tumor (*) and linear, tubular intermediate
tumor signal intensity expanding a vessel
(arrow), compatible with EMVI. The EMVI less
than 1 mm from the MRF (arrowheads) reflects
involvement.

Figure 15. Assessment of MRF involve-

ment by the primary rectal tumor in a
69-year-old man with rectal adenocarcino- Issue 8: LPLN Involvement
ma. Axial oblique T2-weighted MR image
1. All rectal tumors and LPLN compartments are not equally at
shows an anterior ulcerated low rectal tumor
(dashed line). The point of breach of the
risk for developing LPLN metastasis and lateral LR after nCRT.
muscularis propria is at the middle portion 2. The boundaries of LPLN compartments have not been
or base of the tumor. The point of maximum consistently defined at CT or MRI.
extension into the anterior mesorectum 3. There is no LPLN size that is definitive for establishing
(white arrow) is 2 mm from the uninvolved the presence or absence of LPLN metastasis.
MRF (orange arrow).
Impact
Application of LPLN size criteria requires knowledge of rectal
Radiology is that radiologists base their T category on the lymphatics, the anatomy of LPLN compartments, and the im-
portion of the tumor above the anorectal ring and detail the plications of size criteria.
extent of sphincter involvement for preoperative planning
(45,46). Discussion
Importantly, modifications to the TNM stage should reflect In the East, especially Japan, prophylactic LPLN resection in
prognostic implications and not be driven by purely anatomic all T3 or T4 rectal tumors below the APR has been the pre-
considerations. Conflating the two is problematic (Fig 19) (Ta- ferred treatment approach. In the West, nCRT has been used
ble 1). to sterilize metastatic LPLNs due to surgical morbidity and
Volume 44 Number 7 13 [Link]
July 2024 Kaur and Gabriel et al

Figure 17. Assessment of MRF involvement by tumor deposits and a tumor deposit or spiculated
lymph node in two patients with rectal adenocarcinoma. (A) Coronal high-resolution T2-weighted im-
age shows a tumor deposit (white arrow). Heterogeneous nodules with irregular borders contiguous
to tumor thrombus in a vessel is termed the comet tail appearance of a tumor deposit at MRI. This tu-
mor deposit less than 1 mm from the MRF (blue arrow) constitutes involvement. (B) Axial T2-weighted
image shows a near-circumferential tumor with a heterogeneous spiculated nodule (arrow) involving
the right MRF and seminal vesicle. In the absence of the comet tail appearance, round heterogeneous
spiculated nodules may represent tumor deposits or spiculated lymph nodes and cannot be distin-
guished with imaging. A heterogeneous spiculated lymph node or tumor deposit less than 1 mm from
the MRF constitutes MRF involvement.

Figure 18. Assessment of MRF involvement by well-encapsulated benign and malig-

nant mesorectal lymph nodes in two patients. (A) Axial T2-weighted image shows a likely
benign mesorectal lymph node with a smooth capsule (white arrow), which is less than
1 mm from the right MRF (blue arrow). The MRF should not be documented as involved.
Note is made of an anterior ulcerated rectal tumor (arrowhead). (B) Axial T2-weighted
image shows a midrectal tumor. A potentially metastatic, heterogeneous, smooth encap-
sulated node in the left mesorectum (white arrow) less than 1 mm from the MRF (blue
arrow) should not be documented as MRF involvement, as even tumor-bearing lymph
nodes do not confer increased risk of LR (38). The benign right mesorectal node also does
not constitute MRF involvement.

the nonsuperiority of LPLN dissection compared with nCRT Anatomically, only the rectum below the APR has lym-
(47). However, the Lateral Node Study Consortium (LNSC) phatic drainage to LPLNs and the potential to develop LPLN
study of T3 or T4 low rectal cancer treated with total mesorec- metastasis (Figs 3, 20) (50). Of LPLN metastases, 92.3% de-
tal excision (TME) and nCRT found that 19.5% of patients with velop in the internal iliac node and obturator node compart-
greater than or equal to 7-mm LPLNs at pretreatment MRI de- ments (51). Internal iliac nodes represent a sentinel lymph
veloped LR in the LPLN compartments. This suggests that, in node for rectal cancer, and enlarged internal iliac nodes
this subgroup of rectal tumors, nCRT is inadequate and LPLN are more likely to be metastatic than comparable obturator
dissection is required to reduce lateral LR (48,49). nodes (48). Accurate definition of these LPLN compartments

Volume 44 Number 7 14 [Link]

July 2024 Kaur and Gabriel et al

Figure 20. Low rectal cancer in a 63-year-

Figure 19. Sphincter involvement in old man. Coronal T2-weighted image shows
low rectal cancer in a 49-year-old woman. a low rectal tumor abutting the top of the
Coronal T2-weighted image shows a low internal sphincter (teal arrow) and clearly
rectal tumor (arrowheads) with invasion of below the APR (white arrow). This portion
the internal sphincter (blue arrows) and the of the rectum has lymphatic drainage to the
inner fibers of the external sphincter (orange internal iliac and obturator compartments
arrows). (curved arrow). A large metastatic right inter-
nal iliac node is identified (light blue arrow).

at CT or MRI is key to identification of metastatic LPLNs in

rectal cancer. 2. Midpelvis: A line is drawn from the internal iliac artery
Categorization of LPLNs in the surgery literature is well branches to the obliterated umbilical artery. All nodes lateral
established (14). The radiology literature also has an existing to this plane abutting and posterior to the external iliac ves-
classification of LPLNs at CT or MRI (52). However, there is sels are obturator nodes.
lack of alignment between the surgical and radiologic defini- A line along the pelvic brim where external iliac vessels
tions of LPLN compartments, particularly the obturator node exit the pelvis marks the anterior boundary of the obturator
and internal iliac node compartments. The Lateral Node compartment. An oblong lymph node adjacent to the external
Study Consortium (LNSC) has attempted to standardize the iliac vessels at the level transected by this plane is outside the
classification of LPLN compartments at imaging by creating obturator compartment (48).
a color-coded CT-MRI atlas. However, a recent study of 53 ra- 3. Low pelvis: All lymph nodes below the infrapiriformis
diologists found that only 46%–62% of radiologists could cor- foramen are obturator nodes. Obturator nodes in this location
rectly localize internal iliac nodes and obturator nodes. This were described for the first time, to our knowledge, by the Lat-
increased to 72%–77% after a 2-hour training session but did eral Node Study Consortium (Fig 22).
not reach 100% agreement with expert localization (53). We In terms of prognosis and staging, internal iliac nodes are
suggest the following simplified approach to define the inter- categorized by the AJCC as regional or N category for rectal
nal iliac and obturator lymph node compartments based on cancer (13). The overall survival for internal iliac node metas-
the LNSC color-coded CT-MRI atlas (Fig 21): tasis is comparable to that of N2 category mesorectal lymph
nodes. Obturator node involvement is more controversial, as
Internal Iliac Lymph Node Compartment.—All lymph nodes these lymph nodes have a worse overall survival than inter-
abutting the internal iliac artery and its branches from its nal iliac nodes but a superior overall survival to that of oth-
origin up to the point where the terminal arterial branches er LPLN compartments. However, resection of these lymph
exit the infrapiriformis foramen are internal iliac nodes. The nodes has been associated with excellent survival outcomes,
internal pudendal and inferior gluteal artery are the terminal supporting classification of both as regional lymph nodes
branches of the internal iliac artery that exit through the in- (54). Distinction of obturator nodes from M category external
frapiriformis foramen (Figs 21–23). iliac nodes is exceedingly important, as there is a significant
difference in overall survival between these two lymph node
Obturator Lymph Node Compartment.—Lymph nodes lateral groups and because external iliac lymph nodes—unlike obtu-
to the internal iliac vessels at the pelvic brim, posterior to the rator nodes—are not candidates for LPLN dissection (55).
external iliac vessels at the midpelvis, and inferior to the in- Since LPLN metastasis is rare in T1 or T2 low rectal tumors,
frapiriformis foramen are obturator nodes. We propose place- with an incidence of only 5.4%–8.2%, LPLN size criteria are
ment of the following landmark planes: generated only for T3 or T4 tumors, which have a 16.5%–37.2%
1. Pelvic brim: A line drawn along the lateral aspect of the incidence of metastatic LPLNs and which localize primarily to
internal iliac vessels separates the cephalic portion of the in- internal iliac nodes and obturator nodes (51). LPLN size crite-
ternal iliac compartment and the obturator compartment. ria are not applicable for upper rectal or T1 or T2 tumors and

Volume 44 Number 7 15 [Link]

July 2024 Kaur and Gabriel et al

Figure 21. Axial MR or CT images and illustrations at the pelvic brim, mid pelvis, and lower pelvis with pro-
posed pelvic planes to simplify categorization of LPLN compartments. (A) Axial T2-weighted image at the pelvic
brim. A red line marks the lateral aspect of the internal iliac vessels, separating the cephalic portion of the inter-
nal iliac compartment (blue outline) and the obturator compartment (orange outline). Illustration demonstrates
the internal iliac node compartment (blue outline) and obturator compartment (orange outline). (B) Axial CT
image at the midpelvis. A red line from the internal iliac artery branches along the plane of the obliterated
umbilical artery (black arrows) and separates the internal iliac compartment (blue outline) from the obturator
compartment (orange outline). Anteriorly, the obliterated umbilical artery is medial and should not be confused
with the laterally located vas deferens (white arrow). Black arrow in illustration = obliterated umbilical artery,
white arrow in illustration = internal iliac vessel branches. (C) Axial contrast-enhanced CT image at the lower
pelvis. A red line marks the anterior boundary of the obturator compartment as the external iliac vessels exit the
pelvis. The flat oblong LPLNs medial to the external iliac vessels at this level (arrows) do not receive visceral lym-
phatics from pelvic organs and are not involved in rectal cancer. These are not obturator compartment nodes
but external iliac nodes. Blue arrow in illustration = terminal internal pudendal artery branch of the internal iliac
artery, orange arrow in illustration = terminal inferior gluteal artery branch of the internal iliac artery.

apply primarily to the obturator node and internal iliac node intrinsic limitations of using lymph node size as a surrogate
compartments, as 92% of assessed LPLNs localized to these for metastatic disease. In the data from the largest recent
compartments (49). study of LPLNs, the Lateral Node Study Consortium found
It is important to recognize the limitations of lymph node that using a criterion of greater than or equal to 7 mm at
size criteria related to variability in measurement and the pretreatment MRI resulted in nearly 20% lateral LR after

Volume 44 Number 7 16 [Link]

July 2024 Kaur and Gabriel et al

Figure 22. Coronal MR image and illustra-

tion clarify the location of the infrapiriformis
foramen and the relative locations of the in-
ternal iliac and deep obturator node compart-
ments in the coronal plane. Axial T2-weighted
images show these nodes in the axial plane.
(A) Coronal T2-weighted image shows the
internal pudendal artery (blue arrow) and
inferior gluteal artery (orange arrow) exiting
through the infrapiriformis foramen (green
arrow), located below the piriformis muscle
(blue arrowhead). The obturator compart-
ment (orange outline) continues below the
caudal extent of the internal iliac compart-
ment (blue outline). (B) Coronal illustration
shows the same findings as the coronal MR
image (A). The continuation of the obturator
compartment below the infrapiriformis fora-
men described by the Lateral Node Study Con-
sortium has not been previously described
in the radiology literature, to our knowledge.
(C) Axial T2-weighted image above the infrapiriformis foramen (blue dashed line in B) shows an internal iliac node that abuts the inferior gluteal
artery (white arrow) and internal pudendal artery (orange arrow) within the pelvis. (D) Axial T2-weighted image inferior to the infrapiriformis
foramen (orange dashed line in B) shows an obturator compartment node. The right internal pudendal artery (orange arrow) and inferior gluteal
artery (white arrow) have exited through the infrapiriformis foramen, which marks the end of the internal iliac vessel compartment.

Figure 23. Axial T2-weighted images in two

patients demonstrate LPLN categorization.
(A) Axial T2-weighted image at the midpelvis
shows a lymph node (blue arrow) posterior to
the right external iliac vessels and adjacent
to the obturator nerve (white arrow). The red
line runs from the internal iliac branches to the
obliterated umbilical artery (not shown). All
lymph nodes lateral to the red line and poste-
rior to the external iliac vessels are obturator
compartment nodes. (B) Axial T2-weighted
image just below the pelvic brim shows a large
LPLN (arrow) abutting left internal iliac artery
branches and medial to the plane along the
lateral margin of the internal iliac vessels (red
line). This is an internal iliac node.

nCRT. However, this and other studies acknowledge a linear on expert measurement) and 8–11 mm for a 9.3-mm node
relationship between LPLN size, likelihood of lymph node (31).
metastasis, and consequent development of lateral LR. The In terms of assessment of LPLNs at posttreatment MRI,
different size cutoffs will yield variable lymph node positivi- there is no consensus agreement on the size criteria for
ty and lateral LR rates. The data from three studies are sum- LPLNs after nCRT (53). However, the Lateral Node Study Con-
marized in Table 3 (49,56,57). Furthermore, a recent study sortium reported that residual internal iliac lymph nodes
suggests that minor differences in lymph node size criteria greater than 4 mm (short axis) at post-nCRT MRI showed a
may be irrelevant, as there is significant interobserver vari- 52.3% 5-year rate of lateral LR compared with only 9.5% for ob-
ation in lymph node size measurements. This study found turator lymph nodes, reflecting the significance of even small
measurement ranges of 2–7 mm for a 5.7-mm node (based residual internal iliac nodes (Table 4) (48).

Volume 44 Number 7 17 [Link]

July 2024 Kaur and Gabriel et al

Table 3: Rates of Lateral LR for Different-sized LPLNs at Pre-nCRT MRI or CT

LPLN Size (Short Axis) at Pretreat- Rate of Lateral Compartment LR

ment MRI (mm) Location of Tumor and T Category after nCRT at 4–5 Years (%) References
>10 T3 or T4 <8 cm from the anorectal ring or verge 33–36.7 39, 44, 45
5 to <10 T3 or T4 <8 cm from the anorectal ring or verge 10.1–20 44, 45
<5 T3 or T4 <8 cm from the anorectal ring or verge 6.4 45

Table 4: LPLN Size Criteria at Pre- and Posttreatment MRI

LPLN Size (Short Axis) at LPLN Size (Short Axis) at Rate of Lateral LR at 5
LPLN Compartment Pretreatment MRI (mm) Posttreatment MRI (mm) Years (%)
Internal iliac node >7 >4 52.3
compartment
Obturator node >7 >4 9.5
compartment

Conclusion References
Rectal MRI has transformed staging and management of 1. Kapiteijn E, Marijnen CA, Nagtegaal ID, et al. Preoperative radiotherapy
combined with total mesorectal excision for resectable rectal cancer. N
rectal cancer. Despite great strides in performance and im- Engl J Med 2001;345(9):638–646.
age interpretation, numerous anatomic nuances and areas 2. Peeters KC, Marijnen CA, Nagtegaal ID, et al; Dutch Colorectal Cancer
of confusion exist, including the definitions of the anal ca- Group. The TME trial after a median follow-up of 6 years: increased lo-
cal control but no survival benefit in irradiated patients with resectable
nal, rectum, APR, and upper and lower rectum; the loca- rectal carcinoma. Ann Surg 2007;246(5):693–701.
tion of the MRF in relation to the upper rectum; anatomic 3. Ruppert R, Kube R, Strassburg J, et al; other members of the OCUM
structures to be assessed for MRF involvement; distinction Group. Avoidance of Overtreatment of Rectal Cancer by Selective
Chemoradiotherapy: Results of the Optimized Surgery and MRI-based
between the CRM and MRF; and definition of boundaries Multimodal Therapy Trial. J Am Coll Surg 2020;231(4):413–425.e2.
of LPLN compartments most commonly involved in rectal 4. MERCURY Study Group. Diagnostic accuracy of preoperative magnet-
cancer at CT and MRI. This article attempts to identify and ic resonance imaging in predicting curative resection of rectal cancer:
prospective observational study. BMJ 2006;333(7572):779.
address these areas of confusion and provide clarifying ex- 5. Rectal Cancer Alliance of Canada (RCAC). QuickSilver: a phase II study
planations to promote more accurate rectal MRI interpreta- using magnetic resonance imaging criteria to identify “good progno-
tions. sis” rectal cancer patients eligible for primary surgery. JMIR Res Protoc
2015;4(2):e41.
6. Taylor FG, Quirke P, Heald RJ, et al; MERCURY study group. One mil-
Author affiliations.—From the Departments of Abdominal Imaging (H.K.), limetre is the safe cut-off for magnetic resonance imaging prediction
Colon and Rectal Surgery (T.K.), and Anatomical Pathology (M.W.T.), Universi- of surgical margin status in rectal cancer. Br J Surg 2011;98(6):872–879.
ty of Texas MD Anderson Cancer Center, PO Box 301402, Unit 1473, Houston, 7. Lee JM, Kim NK. Essential anatomy of the anorectum for colorectal sur-
TX 77230-1402; Department of Radiology, Northwestern University, Chicago, geons focused on the gross anatomy and histologic findings. Ann Colo-
Ill (H.G., C.L.V., L.C.K.); Department of Radiology, University of Texas Health proctol 2018;34(2):59–71.
Science Center, Houston, Tex (M.O.A.); Department of Radiology, University 8. Brierley JD, Gospodarowicz MK, Wittekind C, eds. TNM Classification
of Pittsburgh Medical Center, Pittsburgh, Pa (E.M.); Department of Radiolo- of Malignant Tumours, 8th ed. Union for International Cancer Control
gy, Endeavor Health Medical Group, Evanston, Ill (M.M.); and Department of (UICC), 2016.
Radiological Sciences, University of California, Irvine, School of Medicine, Or- 9. Liu X, Wang Z, Ren H, Wang Z, Li J. Accuracy of magnetic resonance
ange, Calif (S.L.). Recipient of a Cum Laude Award for an education exhibit at imaging in defining dentate line in anal fistula. BMC Med Imaging
the 2022 RSNA Annual Meeting. Received August 8, 2023; revision requested 2022;22(1):201.
September 11 and received December 20; accepted January 30, 2024. Address 10. Charmichael J, Mills S. Anatomy and Embryology of the Colon, Rectum
correspondence to H.K. (email: hkaur@[Link]). and Anus. In: Steele S, Hull T, Hyman N, Maykel J, Read T, Whitlow C,
eds. The ASCRS Textbook of Colon and Rectal Surgery. 4th ed. Switzer-
Acknowledgments.—We would like to thank Kelly Kage for her outstanding il- land: Springer Nature, 2022.
lustrations. The late Chusilp Charnsangavej, MD, was the inspiration behind 11. Glynne-Jones R, Wyrwicz L, Tiret E, et al; ESMO Guidelines Committee.
this work. Rectal cancer: ESMO Clinical Practice Guidelines for diagnosis, treat-
ment and follow-up. Ann Oncol 2017;28(Suppl 4):iv22–iv40.
Disclosures of conflicts of interest.—H.K. Travel and honorarium paid by 12. Benson AB, Venook AP, Al-Hawary MM, et al. Rectal cancer, version
Medtronic for presenting at meeting, cochair of Society of Abdominal Radiol- 2.2022, NCCN clinical practice guidelines in oncology. J Natl Compr
ogy Disease-focused Panel on Colorectal Cancer. M.W.T. Lodging and travel Canc Netw 2022;20(10):1139–1167.
expenses from American Society for Clinical Pathology (ASCP) and United 13. AJCC Cancer Staging Manual, 8th ed. Amin MB, Edge SB, Greene FL, et
States and Canadian Academy of Pathology (USCAP), lodging from Indian al (eds). Springer Cham; 2017.
Society of Gastrointestinal and Abdominal Radiology. L.C.K. Research pay- 14. Japanese Society for Cancer of the Colon and Rectum. Japanese Classi-
ments from Canon Medical for clinical study. All other authors, the editor, fication of Colorectal, Appendiceal, and Anal Carcinoma: the 3d English
and the reviewers have disclosed no relevant relationships. Edition [Secondary Publication]. J Anus Rectum Colon 2019;3(4):175–
195.
15. UICC Manual of Clinical Oncology. Geneva, Switzerland: Union for In-
ternational Cancer Control, 2015.

Volume 44 Number 7 18 [Link]

July 2024 Kaur and Gabriel et al

16. Nelson H, Petrelli N, Carlin A, et al; National Cancer Institute Expert Pan- cumferential margin involvement is still an important predictor of local
el. Guidelines 2000 for colon and rectal cancer surgery. J Natl Cancer Inst recurrence in rectal carcinoma: not one millimeter but two millimeters is
2001;93(8):583–596. the limit. Am J Surg Pathol 2002;26(3):350–357.
17. Nougaret S, Reinhold C, Mikhael HW, Rouanet P, Bibeau F, Brown G. The 39. Shihab OC, Quirke P, Heald RJ, Moran BJ, Brown G. Magnetic resonance
use of MR imaging in treatment planning for patients with rectal carcino- imaging-detected lymph nodes close to the mesorectal fascia are rarely
ma: have you checked the “DISTANCE”? Radiology 2013;268(2):330–344. a cause of margin involvement after total mesorectal excision. Br J Surg
18. McMullen TP, Easson AM, Cohen Z, Swallow CJ. The investigation of pri- 2010;97(9):1431–1436.
mary rectal cancer by surgeons: current pattern of practice. Can J Surg 40. Gabriel WB. Perineo-abdominal excision of the rectum in one stage. Proc
2005;48(1):19–26. R Soc Med 1935;28(3):212–213.
19. Mathis KL, Nelson H. Defining the Rectum. Ann Surg 2019;270(6):960– 41. Lord AC, Moran B, Abulafi M, et al. Can extranodal tumour deposits be
961. diagnosed on MRI? Protocol for a multicentre clinical trial (the COMET
20. Wasserman MA, McGee MF, Helenowski IB, Halverson AL, Boller AM, trial). BMJ Open 2020;10(10):e033395.
Stryker SJ. The anthropometric definition of the rectum is highly variable. 42. Stelzner S, Ruppert R, Kube R, et al. Selection of patients with rectal can-
Int J Colorectal Dis 2016;31(2):189–195. cer for neoadjuvant therapy using pre-therapeutic MRI: results from
21. Lee S, Kassam Z, Baheti AD, et al. Rectal cancer lexicon 2023 revised and OCUM trial. Eur J Radiol 2022;147:110113.
updated consensus statement from the Society of Abdominal Radiology 43. Gress DM, Edge SB, Greene FL, et al. Principles of Cancer Staging. 8th ed.
Colorectal and Anal Cancer Disease-Focused Panel. Abdom Radiol (NY) AJCC Cancer Staging Manual. Chicago, Ill: American Joint Committee on
2023;48(9):2792–2806. Cancer, 2017.
22. DʼSouza N, de Neree Tot Babberich MPM, d’Hoore A, et al. Definition of 44. Burgart LJ, Chopp WV, Jain D. Protocol for the Examination of Resection
the rectum: an international, expert-based Delphi consensus. Ann Surg Specimens from Patients with Primary Carcinoma of the Colon and Rec-
2019;270(6):955–959. tum. [Link]
23. D’Souza N, Lord A, Shaw A, et al. The sigmoid take-off: An anatomical im- [Link]. Published June 2022. Accessed August 8, 2023.
aging definition of the rectum validated on specimen analysis. Eur J Surg 45. Beets-Tan RGH, Lambregts DMJ, Maas M, et al. Magnetic resonance im-
Oncol 2020;46(9):1668–1672. aging for clinical management of rectal cancer: updated recommenda-
24. D’Souza N, Balyasnikova S, Tudyka V, et al. Variation in landmarks for the tions from the 2016 European Society of Gastrointestinal and Abdominal
rectum: an MRI study. Colorectal Dis 2018;20(10):O304–O309. Radiology (ESGAR) consensus meeting. Eur Radiol 2018;28(4):1465–1475.
25. Dutch guidelines for locoregional staging of rectal cancer. [Link] [Published correction appears in Eur Radiol 2018;28(6):2711.]
[Link]/richtlijn/colorectaal_carcinoom_crc/diagnostiek_bij_ 46. Kassam Z, Lang R, Arya S, et al. Update to the structured MRI report
crc/locoregionale_stadi_ring_rectumcarcinoom.html. Updated May 8, for primary staging of rectal cancer: perspective from the SAR Dis-
2022. Accessed January 12, 2020. ease Focused Panel on Rectal and Anal Cancer. Abdom Radiol (NY)
26. Bogveradze N, Lambregts DMJ, El Khababi N, et al; MRI Rectal Study 2022;47(10):3364–3374.
Group. The sigmoid take-off as a landmark to distinguish rectal from sig- 47. Kusters M, Beets GL, van de Velde CJ, et al. A comparison between the
moid tumours on MRI: reproducibility, pitfalls and potential impact on treatment of low rectal cancer in Japan and the Netherlands, focusing on
treatment stratification. Eur J Surg Oncol 2022;48(1):237–244. the patterns of local recurrence. Ann Surg 2009;249(2):229–235.
27. Gollub MJ, Maas M, Weiser M, et al. Recognition of the anterior peritoneal 48. Ogura A, Konishi T, Beets GL, et al; Lateral Node Study Consortium. Later-
reflection at rectal MRI. AJR Am J Roentgenol 2013;200(1):97–101. al Nodal Features on Restaging Magnetic Resonance Imaging Associated
28. Li W, Peng J, Li C, et al. Prognosis and risk factors for the development of with Lateral Local Recurrence in Low Rectal Cancer after Neoadjuvant
pulmonary metastases after preoperative chemoradiotherapy and radi- Chemoradiotherapy or Radiotherapy. JAMA Surg 2019;154(9):e192172.
cal resection in patients with locally advanced rectal cancer. Ann Transl 49. Ogura A, Konishi T, Cunningham C, et al; Lateral Node Study Consortium.
Med 2020;8(4):117. Neoadjuvant (Chemo)radiotherapy with Total Mesorectal Excision Only
29. Kaur H, Gabriel H, Taggart M, et al. MRI staging in an evolving manage- Is Not Sufficient to Prevent Lateral Local Recurrence in Enlarged Nodes:
ment paradigm for rectal cancer, from the AJR special series on cancer Results of the Multicenter Lateral Node Study of Patients with Low cT3/4
staging. AJR Am J Roentgenol 2021;217(6):1282–1293. Rectal Cancer. J Clin Oncol 2019;37(1):33–43.
30. Kennedy ED, Simunovic M, Jhaveri K, et al. Safety and feasibility of using 50. Heald RJ, Ryall RD. Recurrence and survival after total mesorectal exci-
magnetic resonance imaging criteria to identify patients with “good prog- sion for rectal cancer. Lancet 1986;1(8496):1479–1482.
nosis” rectal cancer eligible for primary surgery: the phase 2 nonrandom- 51. Kobayashi H, Mochizuki H, Kato T, et al. Outcomes of surgery alone for
ized QuickSilver clinical trial. JAMA Oncol 2019;5(7):961–966. lower rectal cancer with and without pelvic sidewall dissection. Dis Colon
31. Lambregts DMJ, Bogveradze N, Blomqvist LK, et al. Current controversies Rectum 2009;52(4):567–576.
in TNM for the radiological staging of rectal cancer and how to deal with 52. Park JM, Charnsangavej C, Yoshimitsu K, Herron DH, Robinson TJ, Wal-
them: results of a global online survey and multidisciplinary expert con- lace S. Pathways of nodal metastasis from pelvic tumors: CT demonstra-
sensus. Eur Radiol 2022;32(7):4991–5003. tion. RadioGraphics 1994;14(6):1309–1321.
32. Taylor FG, Quirke P, Heald RJ, et al; Magnetic Resonance Imaging in Rectal 53. Sluckin TC, Hazen SMJA, Horsthuis K, et al. Significant improvement
Cancer European Equivalence Study Group. Preoperative magnetic reso- after training in the assessment of lateral compartments and short-axis
nance imaging assessment of circumferential resection margin predicts measurements of lateral lymph nodes in rectal cancer. Eur Radiol. 2023;
disease-free survival and local recurrence: 5-year follow-up results of the 33(1): 483–492.
MERCURY study. J Clin Oncol 2014;32(1):34–43. 54. Matsuda T, Sumi Y, Yamashita K, et al. Outcomes and prognostic fac-
33. Disease-focused Panel in Colorectal and Anal Cancer. Society of Abdom- tors of selective lateral pelvic lymph node dissection with preoperative
inal Radiology. [Link] chemoradiotherapy for locally advanced rectal cancer. Int J Colorectal Dis
els/. Accessed August 8, 2023. 2018;33(4):367–374.
34. Glimelius B, Beets-Tan R, Blomqvist L, et al. Mesorectal fascia instead of 55. Chen Z, Sasaki K, Murono K, et al; Japanese Society for Cancer of the Co-
circumferential resection margin in preoperative staging of rectal cancer. lon and Rectum. Oncologic Status of Obturator Lymph Node Metastases
J Clin Oncol 2011;29(16):2142–2143. in Locally Advanced Low Rectal Cancer: A Japanese Multi-Institutional
35. Birbeck KF, Macklin CP, Tiffin NJ, et al. Rates of circumferential resection Study of 3487 Patients. Ann Surg Oncol 2022;29(7):4210–4219.
margin involvement vary between surgeons and predict outcomes in rec- 56. Kusters M, Slater A, Muirhead R, et al. What to Do with Lateral Nodal Dis-
tal cancer surgery. Ann Surg 2002;235(4):449–457. ease in Low Locally Advanced Rectal Cancer? A Call for Further Reflection
36. Lord AC, D’Souza N, Shaw A, et al. MRI-diagnosed tumor deposits and and Research. Dis Colon Rectum 2017;60(6):577–585.
EMVI status have superior prognostic accuracy to current clinical TNM 57. Kim MJ, Kim TH, Kim DY, et al. Can chemoradiation allow for omission
staging in rectal cancer. Ann Surg 2022;276(2):334–344. of lateral pelvic node dissection for locally advanced rectal cancer? J Surg
37. Gollub MJ, Lall C, Lalwani N, Rosenthal MH. Current controversy, confu- Oncol 2015;111(4):459–464.
sion, and imprecision in the use and interpretation of rectal MRI. Abdom 58. You YN, Hardiman KM, Bafford A, et al. The American Society of Colon
Radiol (NY) 2019;44(11):3549–3558. and Rectal Surgeons Clinical Practice Guidelines for the Management of
38. Nagtegaal ID, Marijnen CA, Kranenbarg EK, van de Velde CJ, van Krieken Rectal Cancer. Dis Colon Rectum 2020;63(9):1191–1222.
JH; Pathology Review Committee; Cooperative Clinical Investigators. Cir-

Volume 44 Number 7 19 [Link]

 Erratum for: Anatomic Basis of Rectal Cancer Staging:
Clarifying Controversies and Misconceptions
September 2024 • Volume 44 • Number 9

Originally published in:

[Link]
Anatomic Basis of Rectal Cancer Staging: Clarifying Controversies and Misconceptions
Harmeet Kaur, Helena Gabriel, Muhammad O. Awiwi, Ekta Maheshwari, Camila Lopes Vendrami, Tsuyoshi Konishi, Melissa W.
Taggart, Michael Magnetta, Linda C. Kelahan, Sonia Lee
Erratum in:
[Link]
On page 14, the following sentence was corrected: However, the Lateral Node Study Consortium (LNSC) study of T3 or T4 low rectal cancer
treated with total mesorectal excision (TME) and nCRT found that 19.5% of patients with greater than or equal to 7-mm LPLNs at pretreatment
MRI developed LR in the LPLN compartments.
Figure 1A and 1B were replaced to include new keys, and the figure legend was updated to define the keys.
In the reference list, new reference 53 was added, and the references were renumbered in the text and reference list:
53. Sluckin TC, Hazen SMJA, Horsthuis K, et al. Significant improvement after training in the assessment of lateral compartments and
short-axis measurements of lateral lymph nodes in rectal cancer. Eur Radiol. 2023; 33(1): 483–492.

©
ERRATUM This copy is for personal use
only. To order copies, contact
RSNA, 2024 reprints@[Link]