Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print August 23, 2006, 10.1148/radiol.2411050942
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2411050942v1
241/1/132    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stollfuss, J. C.
Right arrow Articles by Woertler, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stollfuss, J. C.
Right arrow Articles by Woertler, K.
(Radiology 2006;241:132-141.)
© RSNA, 2006

Experimental Studies

Rectal Carcinoma: High-Spatial-Resolution MR Imaging and T2 Quantification in Rectal Cancer Specimens1

Jens C. Stollfuss, MD, Karen Becker, MD, Andreas Sendler, MD, Stefan Seidl, MD, Marcus Settles, PhD, Florian Auer, MD, Ambros Beer, MD, Ernst J. Rummeny, MD and Klaus Woertler, MD

1 From the Departments of Radiology (J.C.S., M.S., F.A., A.B., E.J.R., K.W.), Pathology (K.B., S.S.), and Surgery (A.S.), Technische Universität München, Klinikum rechts der Isar, Ismaningerstrasse 22, 81675 Munich, Germany. Received June 5, 2005; revision requested August 1; revision received October 5; accepted November 4; final version accepted January 3, 2006. Address correspondence to J.C.S. e-mail: sto{at}roe.med.tum.de).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Purpose: To prospectively compare high-spatial-resolution T1-weighted, T2-weighted, and intermediate-weighted spectral fat-saturated magnetic resonance (MR) imaging for the differentiation of tumor from fibrosis and for delineation of rectal wall layers in rectal cancer specimens.

Materials and Methods: The local ethics committee approved the protocol, and written informed consent was obtained from each patient. Thin-section high-spatial-resolution MR imaging was performed in specimens obtained from 23 patients (16 men, seven women; median age, 64 years; age range, 39–84 years) immediately after resection. Seven patients underwent neoadjuvant treatment. T1-weighted spin-echo, T2-weighted fast spin-echo, and intermediate-weighted spectral fat-saturated MR images were obtained in the transverse plane. Differences in signal intensity between tumor and fibrosis and between tumor and rectal wall layers were evaluated by using visual scoring and measurements of T2 relaxation time. Statistical differences were evaluated by using the Wilcoxon signed rank test and a mixed-model regression analysis. All images were compared with whole-mount histopathologic slices (n = 86).

Results: T2-weighted MR images provided the best differentiation between tumor and fibrosis (P < .001). Mean visual signal intensity scores were –1.8 for T2-weighted MR images, –1.4 for intermediate-weighted spectral fat-saturated MR images, and –0.2 for T1-weighted MR images. T2 relaxation times were 97 msec ± 4.6 for tumor and 70 msec ± 3.8 for fibrosis (P < .001). Substantial overlap was noted between the tumor and the circular layer of the muscularis propria (97 msec ± 2.1), and less overlap was noted between the tumor and the longitudinal layer of the muscularis propria (88 msec ± 1.6).

Conclusion: T2-weighted MR imaging provides superior delineation of rectal wall layers and better differentiation of tumor from fibrosis in rectal cancer specimens compared with T1-weighted MR imaging and intermediate-weighted spectral fat-saturated MR imaging by using thin-section high-spatial-resolution sequences.

© RSNA, 2006


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
The prognosis of rectal cancer is influenced by multiple factors, such as local tumor extent, lymph node involvement, and the presence of distant metastasis (1–5). Of these factors, the presence and extent of extramural tumor spread influence both long-term survival and the risk of local recurrence (5–7). Most of the factors that influence the prognosis of rectal cancer have been assessed traditionally by using histopathologic examination. With the more widespread acceptance of neoadjuvant concepts, there is an increasing need for preoperative imaging methods to select patients for more aggressive multimodality treatment approaches on the basis of individual risk factors (8,9).

Magnetic resonance (MR) imaging has been shown to be a promising technique for preoperative local staging and may also provide measurements of the clinically relevant distance to the mesorectal fascia (10). Staging failures, however, have been known to occur with MR imaging in the differentiation of T2 tumors (ie, those confined to the rectal wall) and borderline T3 tumors (ie, those that infiltrate the mesorectum). There is also a tendency for overstaging that is mainly attributed to desmoplastic reaction, which can cause spiculations in the perirectal fat that may or may not contain viable tumor cells (11–13). To date, no consensus has been reached regarding which MR imaging sequence is optimal for the staging of rectal cancer, and multiple pulse sequences are often used in routine protocols. Thus, the purpose of our study was to prospectively compare high-spatial-resolution T1-weighted, T2-weighted, and intermediate-weighted spectral fat-saturated magnetic resonance (MR) imaging for the differentiation of tumor from fibrosis and for delineation of rectal wall layers in rectal cancer specimens.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Specimens
Rectum specimens from 23 consecutive patients (16 men, seven women; age range, 39–84 years; median age, 64 years) were examined in a fresh state immediately (15–30 minutes) after resection. All tumors were single nonobstructing lesions that were located within 15 cm of the anal verge. Fourteen of 23 tumors were located in the middle third of the rectum, five were located in the upper third of the rectum, and four were located in the lower third of the rectum. Seven patients received 45 Gy (1.8-Gy fractions) of preoperative pelvic radiation combined with 5-fluorouracil–based chemotherapy. Surgery was performed 5–6 weeks after neoadjuvant therapy. In 16 patients, no irradiation or chemotherapy was performed prior to surgery. All patients underwent total mesorectal excision (14) by means of either low anterior resection or abdominoperineal excision (15). The ethics committee of our institution approved the protocol, and written informed consent was obtained from each patient.

Specimen Handling
After total mesorectal excision, each specimen was opened along the anterior border proximal to the tumor-containing segment that was left untouched in 18 of 23 specimens. The specimens were opened by a pathologist (K.B. or S.S.). Five specimens were completely opened to ensure adequate distance to the aboral resection margin.

MR Imaging
MR imaging was performed by using a 1.5-T MR imager (Intera upgrade; Philips Medical Systems, Best, the Netherlands) that was equipped with a radiofrequency coil (model 445PH-64; IGC Medical Advances, Milwaukee, Wis). This coil has a maximum transverse field coverage of 11 cm. The nondissected area of the specimen that contained the tumor was positioned in the coil after a small amount of ultrasonographic (US) gel was administered into the lumen for distention. The resection borders of the specimen were inked to enable orientation analogous to that of a patient in the supine position in the imager.

After localization, imaging was performed perpendicular to the long axis of the specimen (transverse plane) and covered the entire tumor area. Twenty sections were obtained by using a 2-mm section thickness and 0.2-mm intersection gap. The MR technique consisted of the following thin-section high-spatial-resolution MR imaging sequences: T2-weighted fast spin-echo MR imaging, intermediate-weighted fast spin-echo MR imaging with spectral fat saturation, T1-weighted spin-echo MR imaging, and multisection multi-echo spin-echo MR imaging, which was used to calculate T2 relaxation profiles (Table 1). All sequences were acquired in random order to prevent systematic errors resulting from potential specimen decay.


View this table:
[in this window]
[in a new window]

  		Table 1. MR Imaging Parameters

 
The intermediate-weighted sequence with spectral fat saturation was chosen because of the high signal-to-noise ratio of the resulting images, which are known to provide excellent anatomic detail and depiction of fine structures, although limited tissue contrast can be expected. Fluid and edema, however, can degrade image quality with fat saturation in terms of the differentiation of anatomic structures and are less emphasized on intermediate-weighted MR images than on long-echo-time MR images.

T1-weighted and intermediate-weighted MR images were obtained in all 23 specimens. T2-weighted and multisection multi-echo MR images were obtained in 11 of 23 specimens. Because of the limited time resources in clinical practice, only T2-weighted MR imaging (no multisection multi-echo MR imaging) was performed in six specimens. In another six specimens, only multisection multi-echo MR imaging (no T2-weighted MR imaging) was performed. The echo images that were obtained with an echo time of 105 msec were used for visual analysis in the latter six specimens, which were imaged with multisection multi-echo MR imaging only. The decision to perform multisection multi-echo MR imaging or T2-weighted MR imaging in these 12 of 23 specimens depended solely on the MR time schedule for that particular day of acquisition.

Histopathologic Analysis
After 24 hours, the formalin-fixed specimens were cut into 1-cm-thick slices. All tumor-containing slices were embedded in paraffin, and the slice with the maximum amount of tumor infiltration was used for further preparation. Between three and five whole-mount slices were prepared for microscopy from the latter slice, including the slice with the deepest tumor infiltration. The slices were examined by means of macroscopic and microscopic analysis by one pathologist who specialized in gastrointestinal pathologic analysis (K.B., 15 years of experience) and who was initially blinded to the MR imaging results.

Staining was performed with hematoxylin-eosin stain and periodic acid–Schiff stain, as well as with elastica–van Giesson stain for specimens obtained in patients who had undergone preoperative irradiation. Spatial correlation was achieved by identifying anatomic landmarks (eg, bowel contour, wall layers, blood vessels, and tumor deposits) that were visible both in the specimen and on the MR images. The extent of local tumor spread was assessed in each slice by using the TNM classification (16). An overall tumor stage was also assigned according to the maximum local tumor spread on any given slice.

Normal Architecture
For comparison, a total of 86 quartets of corresponding tissue slices and MR images were available in 23 specimens for the evaluation of normal architecture. MR images were reviewed in consensus by two radiologists (K.W. and J.C.S., with 10 and 7 years of experience, respectively, in gastrointestinal MR imaging) with respect to the effect of different pulse sequences on the anatomic and pathologic detail that could be discerned. The specimens and MR images were evaluated in a random order, with time intervals between the evaluations. Both readers knew where the tumors were located before starting visual analysis, and the pathologist provided assistance in the interpretation of histomorphology during reading sessions. The continuity and signal intensity (SI) of each layer of the rectal wall was evaluated, and the degree of infiltration by the primary tumor into the rectal wall was categorized according to the layer invaded.

Visual SI Scoring
Visual MR interpretation was performed by comparing the SI of the mucosa, submucosa, and circular and longitudinal layers of the muscularis propria with that of the tumor, as well as by comparing the SI of macroscopic fibrosis adjacent to the tumor or in the perirectal fat with that of the tumor. In one specimen (three slices) that was obtained after irradiation, areas with a mixture of residual tumor and fibrosis (in approximately 50% fractions) were present and were not considered for evaluation. Also, comparisons could not be made in three specimens (12 slices) that contained no viable tumor after irradiation, so 71 of 86 quartets in 19 specimens were left for visual SI scoring.

A single visual score per MR imaging sequence (wall layer vs tumor) was established with the consensus of two readers for each slice (n = 71). SI values were classified by using the following five-point scale: –2, markedly lower SI in wall layer than in tumor; –1, slightly lower SI in wall layer than in tumor; 0, SI of wall layer equal to that of tumor; 1, slightly higher SI in wall layer than in tumor; and 2, markedly higher SI in wall layer than in tumor.

The evaluation was guided by histopathologic findings so that analysis could be performed even when a given wall layer could not be clearly differentiated from the other layers by using MR images alone (eg, on T1-weighted MR images). All histopathologic slices contained at least part of the normal rectal wall layers. Thus, a total of 71 slices were available for each of the three sequences (T2-weighted, T1-weighted, and intermediate-weighted MR imaging) for all layers of the rectal wall (mucosa, submucosa, and circular and longitudinal layers of the muscularis propria), as well as for the perirectal fat. Fibrosis, however, was not present in all slices at histopathologic analysis; consequently, only 44 slices in 14 patients were available to compare with tumor tissue for each MR imaging sequence in the analysis of fibrosis. Desmoplastic stroma of the tumor and therapy-induced fibrosis are difficult to differentiate in terms of histopathologic features and are summarized for MR evaluation and referred to as fibrosis in the following text sections. SI scores were averaged for each MR imaging sequence over all sections and all patients. The results are presented as the mean ± standard deviation.

T2 Relaxation Time Mapping
T2 relaxation time maps were calculated on a pixel-by-pixel basis for all images by using a linear curve-fitting algorithm. SI as a function of time was fitted by using a monoexponential function for each pixel. The T2 map was generated from the slope of the best fit. A total of 5100 relaxation profiles were calculated (15 echo images from multisection multi-echo MR imaging per slice x 20 slices per specimen x 17 specimens). The numeric values for T2 relaxation time were represented on a gray-scale image.

The data for T2 relaxation time were assessed by a single reader (J.C.S.) on the basis of a region of interest (ROI) analysis by using parametric T2 relaxation time images in 17 of 23 specimens; in the remaining six specimens, multisection multi-echo MR imaging was not performed. Again, evaluation was guided by histopathologic findings. A total of 69 slices were available in 17 specimens (between three and five slices per specimen) for analysis of rectal wall layers, including the mucosa, submucosa, and circular and longitudinal layers of the muscularis propria. No measurements were taken from the muscularis mucosae because they were too thin to draw adequate ROIs. At histopathologic analysis, 60 of 69 slices showed tumor tissue that could be evaluated. Measurable fibrosis was present in 42 of 69 slices at histopathologic examination.

Irregular ROIs were drawn by one reader (J.C.S.) who used visual correlation with histomorphology to ensure correct positioning of the ROI. ROIs encompassed only a portion of the lesion or rectal wall layers. The average T2 relaxation time for each ROI was used for further calculations. The average T2 relaxation time for each ROI was averaged for all slices and all patients (n = 69 for normal wall layers, n = 60 for tumor tissue, and n = 42 for fibrosis). Results are presented as the mean T2 relaxation time in milliseconds ± standard deviation for each group of tissues. Relaxation rates (R2 = 1/T2) are given in inverse seconds ± standard deviation.

Statistical Analysis
A one-sample sign test was used to determine whether the mean score of the relative visual SI differences (rectal wall layers vs tumor and fibrosis vs tumor) that were obtained by using one of the MR imaging sequences was different from zero (zero would mean equal SI for the tested structure and tumor tissue). A P value of less than .05 was considered to indicate a statistically significant difference. Analysis was performed by using StatView (version 5.0; SAS Institute, Cary, NC).

The matched-pairs Wilcoxon signed rank test was used to evaluate statistical significance by comparing the different MR imaging sequences with respect to visual SI scores. As described above, a single score was established by the consensus of two readers. There were 71 scores for the analysis of rectal wall layers relative to tumor tissue and 44 scores for the analysis of fibrosis relative to tumor tissue for each MR imaging sequence. A P value of less than .05 was considered to indicate a statistically significant difference when comparing scores as a measure of the ability to differentiate between rectal wall layers and tumor tissue or between fibrosis and tumor tissue by using different MR imaging sequences. Analysis was performed by using StatView (version 5.0; SAS Institute).

One reader (J.C.S.) established the T2 relaxation time for tumor tissue, fibrotic tissue, and all normal layers of the rectal wall (except for the muscularis mucosae). Because there was an unequal number of samples for each tissue type and more than one sample per specimen, we performed a mixed-model regression analysis to test for statistical differences between the T2 relaxation times for each of the different types of tissue. A P value of less than .05 was considered to indicate a statistically significant difference. Analysis was performed by using SPSS (version 13.0; SPSS, Chicago, Ill).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Histopathologic Results
At gross examination, eight lesions were classified as fungating and/or ulcerative (size range, 2.5–5.5 cm); these lesions were found to be stage T3 tumors at histopathologic analysis. In addition, there were 10 sessile elevated tumors (five stage T2 tumors and five stage T3 tumors; size range, 0.8–5.0 cm). There was one semipedunculated lesion adjacent to a polyp (1.5 cm) that was classified as a stage T2 tumor. One circular lesion that infiltrated the peritoneum was classified as a stage T4 tumor. Three specimens did not show evidence of viable tumor after preoperative irradiation and chemotherapy.

Normal Architecture
On T2-weighted MR images, the rectal wall was apparent as five distinct layers in 19 cases; these layers included the mucosa, muscularis mucosae, submucosa, and circular and longitudinal layers of the muscularis propria (Fig 1). In four cases, the mucosa and the muscularis mucosae did not appear as distinct layers on T2-weighted MR images. The SI of the mucosa relative to that of the circular layer of the muscularis propria was similar in all cases. The SI of the submucosa varied from intermediate to high compared with that of the circular layer of the muscularis propria. The circular and longitudinal muscle layers were distinct in terms of morphology and SI (Fig 2a, 2d). The longitudinal layer showed lower SI than the circular layer in 19 of 23 cases.


Figure 1
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a: Small tumor growing adjacent to polyp in 62-year-old man. (a) Transverse thin-section (2-mm) T2-weighted spin-echo MR image (4958/105 [repetition time msec/echo time msec]) demonstrates five rectal wall layers, including the mucosa (muc), muscularis mucosae (black arrow in a and b), submucosa (sub), and circular (m) and longitudinal (M) layers of the muscularis propria. Perirectal fat (F) is also seen. Transition zone (white arrows in a and b) between tumor tissue (T), mucosa, and hyperplastic polyp (P) is demonstrated. No fat plane is visible between circular muscle and tumor because of similar SI. Polyp and mucosal layer show slightly lower SI compared with tumor. Lumen is filled with US gel. (b) Histologic slice confirms T2 tumor. Transition zone between tumor and hyperplastic polyp on one end and normal mucosal layer on other end can be clearly identified. (Hematoxylin-eosin stain; original magnification, x1.)

 

Figure 1
View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b: Small tumor growing adjacent to polyp in 62-year-old man. (a) Transverse thin-section (2-mm) T2-weighted spin-echo MR image (4958/105 [repetition time msec/echo time msec]) demonstrates five rectal wall layers, including the mucosa (muc), muscularis mucosae (black arrow in a and b), submucosa (sub), and circular (m) and longitudinal (M) layers of the muscularis propria. Perirectal fat (F) is also seen. Transition zone (white arrows in a and b) between tumor tissue (T), mucosa, and hyperplastic polyp (P) is demonstrated. No fat plane is visible between circular muscle and tumor because of similar SI. Polyp and mucosal layer show slightly lower SI compared with tumor. Lumen is filled with US gel. (b) Histologic slice confirms T2 tumor. Transition zone between tumor and hyperplastic polyp on one end and normal mucosal layer on other end can be clearly identified. (Hematoxylin-eosin stain; original magnification, x1.)

 

Figure 2
View larger version (94K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a: Sessile T3 tumor with small portion infiltrating perirectal fat in 48-year-old man. (a) Transverse T2-weighted fast spin-echo MR image (5958/105) demonstrates relatively small (1.8-cm) sessile tumor (T) in dorsal aspect of rectum, with small portion (white arrow in a–c) growing into perirectal fat (F). Longitudinal layer (M) of the muscularis propria shows lower SI than circular layer (m) of the muscularis propria. Circular layer of the muscularis propria and tumor have similar SI. Muscularis mucosae are depicted as a separate layer of low SI (black arrow in a and b). (b) Corresponding intermediate-weighted MR image (1400/42) shows five distinct rectal wall layers. SI difference between tumor and muscularis propria is small compared with that in a. (c) Corresponding T1-weighted MR image (500/22) does not demonstrate separation of muscularis propria into distinct layers; SI of tumor and that of muscularis propria are very similar. (d) Corresponding histopathologic slice shows tumor growing along perivascular spaces (large black arrow); mucularis mucosae is also seen (small black arrow). (Periodic acid–Schiff stain; original magnification, x1.)

 

Figure 2
View larger version (98K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b: Sessile T3 tumor with small portion infiltrating perirectal fat in 48-year-old man. (a) Transverse T2-weighted fast spin-echo MR image (5958/105) demonstrates relatively small (1.8-cm) sessile tumor (T) in dorsal aspect of rectum, with small portion (white arrow in a–c) growing into perirectal fat (F). Longitudinal layer (M) of the muscularis propria shows lower SI than circular layer (m) of the muscularis propria. Circular layer of the muscularis propria and tumor have similar SI. Muscularis mucosae are depicted as a separate layer of low SI (black arrow in a and b). (b) Corresponding intermediate-weighted MR image (1400/42) shows five distinct rectal wall layers. SI difference between tumor and muscularis propria is small compared with that in a. (c) Corresponding T1-weighted MR image (500/22) does not demonstrate separation of muscularis propria into distinct layers; SI of tumor and that of muscularis propria are very similar. (d) Corresponding histopathologic slice shows tumor growing along perivascular spaces (large black arrow); mucularis mucosae is also seen (small black arrow). (Periodic acid–Schiff stain; original magnification, x1.)

 

Figure 2
View larger version (107K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c: Sessile T3 tumor with small portion infiltrating perirectal fat in 48-year-old man. (a) Transverse T2-weighted fast spin-echo MR image (5958/105) demonstrates relatively small (1.8-cm) sessile tumor (T) in dorsal aspect of rectum, with small portion (white arrow in a–c) growing into perirectal fat (F). Longitudinal layer (M) of the muscularis propria shows lower SI than circular layer (m) of the muscularis propria. Circular layer of the muscularis propria and tumor have similar SI. Muscularis mucosae are depicted as a separate layer of low SI (black arrow in a and b). (b) Corresponding intermediate-weighted MR image (1400/42) shows five distinct rectal wall layers. SI difference between tumor and muscularis propria is small compared with that in a. (c) Corresponding T1-weighted MR image (500/22) does not demonstrate separation of muscularis propria into distinct layers; SI of tumor and that of muscularis propria are very similar. (d) Corresponding histopathologic slice shows tumor growing along perivascular spaces (large black arrow); mucularis mucosae is also seen (small black arrow). (Periodic acid–Schiff stain; original magnification, x1.)

 

Figure 2
View larger version (136K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2d: Sessile T3 tumor with small portion infiltrating perirectal fat in 48-year-old man. (a) Transverse T2-weighted fast spin-echo MR image (5958/105) demonstrates relatively small (1.8-cm) sessile tumor (T) in dorsal aspect of rectum, with small portion (white arrow in a–c) growing into perirectal fat (F). Longitudinal layer (M) of the muscularis propria shows lower SI than circular layer (m) of the muscularis propria. Circular layer of the muscularis propria and tumor have similar SI. Muscularis mucosae are depicted as a separate layer of low SI (black arrow in a and b). (b) Corresponding intermediate-weighted MR image (1400/42) shows five distinct rectal wall layers. SI difference between tumor and muscularis propria is small compared with that in a. (c) Corresponding T1-weighted MR image (500/22) does not demonstrate separation of muscularis propria into distinct layers; SI of tumor and that of muscularis propria are very similar. (d) Corresponding histopathologic slice shows tumor growing along perivascular spaces (large black arrow); mucularis mucosae is also seen (small black arrow). (Periodic acid–Schiff stain; original magnification, x1.)

 
On intermediate-weighted MR images, the rectal wall was apparent as four distinct layers in 16 cases; these layers included the mucosa, submucosa, and circular and longitudinal layers of muscularis propria (Fig 2b). In seven cases, the mucosa and the muscularis mucosae appeared as distinct layers on intermediate-weighted MR images. The SI of the submucosa was high relative to that of the circular layer of the muscularis propria. The layers of the muscularis propria constantly showed intermediate SI, with only a minor difference between the longitudinal and circular layers.

On T1-weighted MR images, the whole intestinal wall was apparent as three distinct layers in 19 cases; these layers included the mucosa, submucosa, and muscularis propria (Fig 2c). The circular and longitudinal layers of the muscularis propria were apparent as distinct layers on T1-weighted MR images in only four cases. The SI of the submucosa varied from low to intermediate relative to that of the circular layer of the muscularis propria. The muscularis propria constantly showed low SI.

Visual SI Scoring
The SI of the mucosa was slightly higher than that of the tumor for all sequences, with mean relative SI scores (mucosa vs tumor) ranging from 0.4 to 0.6 (Table 2). The one-sample sign test showed that the mean SI scores (mucosa vs tumor) were significantly different (P < .001) from zero (equal SI) for all three individual sequences. In other words, the SI of the mucosa was significantly higher than that of the tumor. There were no significant differences, however, when comparing SI scores for the three MR imaging sequences, as established by using the matched-pairs Wilcoxon signed rank test (z value: T2-weighted MR imaging vs intermediate-weighted MR imaging, –1.8 [P > .05]; intermediate-weighted MR imaging vs T1-weighted MR imaging, –1.4 [P > .05]; and T2-weighted MR imaging vs T1-weighted MR imaging, –0.6 [P > .05]).


View this table:
[in this window]
[in a new window]

  		Table 2. SI Scores for Rectal Wall Layers and Perirectal Fat Compared with Tumor Tissue

 
The SI of the submucosa was moderately higher than that of the tumor for all sequences, with mean SI scores ranging from 0.7 to 0.9. Again, the mean relative SI score was significantly different (P < .001) from zero (equal SI) for all three sequences, as established by using the one-sample sign test. The relative SI scores did not differ significantly between sequences, as established by using the matched-pairs Wilcoxon signed rank test (z value: T2-weighted MR imaging vs intermediate-weighted MR imaging, –1.6 [P > .05]; intermediate-weighted MR imaging vs T1-weighted MR imaging, –1.8 [P > .05]; and T2-weighted MR imaging vs T1-weighted MR imaging, –0.6 [P > .05]).

The circular layer of the muscularis propria and the tumor showed almost identical SI for all sequences. Mean SI scores ranged from –0.1 to 0.1. Neither the results of the one-sample sign test with respect to an individual MR imaging sequence (T2-weighted MR imaging, P = .17; intermediate-weighted MR imaging, P = .28; and T1-weighted MR imaging, P = .45) nor the differences in relative SI between MR imaging sequences (z value: T2-weighted MR imaging vs intermediate-weighted MR imaging, 0.03 [P > .05]; intermediate-weighted MR imaging vs T1-weighted MR imaging, 1.2 [P > .05]; and T2-weighted MR imaging vs T1-weighted MR imaging, –1.9 [P > .05]) were statistically significant.

Relative visual SI differences for the longitudinal layer of the muscularis propria were not homogeneous, with mean SI scores ranging from –0.9 to –0.2. The mean relative SI score was significantly different (P < .001) from zero (equal SI) for all three sequences, as established by using the one-sample sign test. However, when comparing the different sequences with respect to the longitudinal muscle layer, the relative SI scores for T2-weighted MR imaging (Fig 2a) were significantly lower than those for both intermediate-weighted (Fig 2b) and T1-weighted (Fig 2c) MR imaging, as established by using the matched-pairs Wilcoxon signed rank test (z value: T2-weighted MR imaging vs intermediate-weighted MR imaging, –6.6 [P < .001]; intermediate-weighted MR imaging vs T1-weighted MR imaging, 0.6 [P > .05]; and T2-weighted MR imaging vs T1-weighted MR imaging, –5.2 [P < .001]).

The SI of fibrotic tissue (spiculations) was generally lower than that of the tumor on T2-weighted (Fig 3a, 3d) and intermediate-weighted (Fig 3b) MR images, with mean SI scores of –1.8 for T2-weighted MR imaging and –1.4 for intermediate-weighted MR imaging (P < .001). The relative difference in SI between fibrosis and tumor tissue on T1-weighted MR images was smaller (–0.2) but still significantly different from zero (P = .002), as established by using the one-sample sign test (Fig 3c). By comparing the different sequences, we determined that significantly lower scores were achieved by using T2-weighted and intermediate-weighted MR imaging compared with T1-weighted MR imaging (z value: T2-weighted MR imaging vs intermediate-weighted MR imaging, –4.3 [P < .001]; intermediate-weighted MR imaging vs T1-weighted MR imaging, –6.3 [P < .001]; and T2-weighted MR imaging vs T1-weighted MR imaging, –5.9 [P < .001]).


Figure 3
View larger version (77K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a: Dissected specimen from 65-year-old man 6 weeks after combined irradiation and chemotherapy. (a) Transverse T2-weighted fast spin-echo MR image (5958/105) demonstrates tumor clusters (T) embedded in fibrotic bands (white arrow in a–d) that show substantially lower SI than tumor tissue. Fibrosis (black arrow in a–d) is visualized adjacent to rectal wall in perirectal fat (F). (b) On intermediate-weighted MR image (1400/42), fibrotic changes in perirectal fat are obscured because of fat saturation. Tumor clusters are embedded in fibrotic bands. High SI adjacent to rectal wall is most likely the result of increased fluid content in perirectal tissue. (c) On T1-weighted MR image (500/22), differentiation of tumor from fibrosis is difficult because both show low SI relative to perirectal fat. Fibrosis adjacent to wall in perirectal fat is well visualized. (d) Corresponding histopathologic slice. (Hematoxylin-eosin stain; original magnification, x1.)

 

Figure 3
View larger version (82K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b: Dissected specimen from 65-year-old man 6 weeks after combined irradiation and chemotherapy. (a) Transverse T2-weighted fast spin-echo MR image (5958/105) demonstrates tumor clusters (T) embedded in fibrotic bands (white arrow in a–d) that show substantially lower SI than tumor tissue. Fibrosis (black arrow in a–d) is visualized adjacent to rectal wall in perirectal fat (F). (b) On intermediate-weighted MR image (1400/42), fibrotic changes in perirectal fat are obscured because of fat saturation. Tumor clusters are embedded in fibrotic bands. High SI adjacent to rectal wall is most likely the result of increased fluid content in perirectal tissue. (c) On T1-weighted MR image (500/22), differentiation of tumor from fibrosis is difficult because both show low SI relative to perirectal fat. Fibrosis adjacent to wall in perirectal fat is well visualized. (d) Corresponding histopathologic slice. (Hematoxylin-eosin stain; original magnification, x1.)

 

Figure 3
View larger version (85K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3c: Dissected specimen from 65-year-old man 6 weeks after combined irradiation and chemotherapy. (a) Transverse T2-weighted fast spin-echo MR image (5958/105) demonstrates tumor clusters (T) embedded in fibrotic bands (white arrow in a–d) that show substantially lower SI than tumor tissue. Fibrosis (black arrow in a–d) is visualized adjacent to rectal wall in perirectal fat (F). (b) On intermediate-weighted MR image (1400/42), fibrotic changes in perirectal fat are obscured because of fat saturation. Tumor clusters are embedded in fibrotic bands. High SI adjacent to rectal wall is most likely the result of increased fluid content in perirectal tissue. (c) On T1-weighted MR image (500/22), differentiation of tumor from fibrosis is difficult because both show low SI relative to perirectal fat. Fibrosis adjacent to wall in perirectal fat is well visualized. (d) Corresponding histopathologic slice. (Hematoxylin-eosin stain; original magnification, x1.)

 

Figure 3
View larger version (109K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3d: Dissected specimen from 65-year-old man 6 weeks after combined irradiation and chemotherapy. (a) Transverse T2-weighted fast spin-echo MR image (5958/105) demonstrates tumor clusters (T) embedded in fibrotic bands (white arrow in a–d) that show substantially lower SI than tumor tissue. Fibrosis (black arrow in a–d) is visualized adjacent to rectal wall in perirectal fat (F). (b) On intermediate-weighted MR image (1400/42), fibrotic changes in perirectal fat are obscured because of fat saturation. Tumor clusters are embedded in fibrotic bands. High SI adjacent to rectal wall is most likely the result of increased fluid content in perirectal tissue. (c) On T1-weighted MR image (500/22), differentiation of tumor from fibrosis is difficult because both show low SI relative to perirectal fat. Fibrosis adjacent to wall in perirectal fat is well visualized. (d) Corresponding histopathologic slice. (Hematoxylin-eosin stain; original magnification, x1.)

 
T2 Relaxation Time Mapping
A typical analysis can be illustrated by using monoexponential fits to the multi-echo T2 data and corresponding parametric T2 image (Fig 4). Differences in average T2 relaxation times (Table 3), which were calculated on the basis of a mixed-model regression analysis, were significant for comparing tumor tissue with all other types of tissue, except the circular layer of the muscularis propria (tumor vs mucosa, P < .001; tumor vs submucosa, P < .001; tumor vs circular layer of the muscularis propria, P = .61; tumor vs longitudinal layer of the muscularis propria, P < .001; and tumor vs fibrosis, P < .001). Although the average difference was significant, some overlap in T2 relaxation time was noted between the tumor and the longitudinal layer of the muscularis propria. The similar SI of the muscle and the tumor may be a source of false-positive results after irradiation (Fig 5).


Figure 4
View larger version (27K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a: Typical monoexponential curve fits T2 relaxation data for tumor tissue and different rectal wall layers in 62-year-old man. (a) All curves were normalized at an echo time of zero for clarity. Monoexponential function is given in legend. The slope of the best fit line represents R2. Values are given in milliseconds; R2 values in Table 2 are calculated in seconds. (b) Corresponding parametric T2-weighted MR image (3000/15–225) of specimen includes ROIs for tumor (T), mucosa (muc), submucosa (sub), and circular (m) and longitudinal (M) layers of the muscularis propria. Tumor in left lateral aspect of rectal wall was classified as a stage T2 tumor.

 

Figure 4
View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b: Typical monoexponential curve fits T2 relaxation data for tumor tissue and different rectal wall layers in 62-year-old man. (a) All curves were normalized at an echo time of zero for clarity. Monoexponential function is given in legend. The slope of the best fit line represents R2. Values are given in milliseconds; R2 values in Table 2 are calculated in seconds. (b) Corresponding parametric T2-weighted MR image (3000/15–225) of specimen includes ROIs for tumor (T), mucosa (muc), submucosa (sub), and circular (m) and longitudinal (M) layers of the muscularis propria. Tumor in left lateral aspect of rectal wall was classified as a stage T2 tumor.

 

View this table:
[in this window]
[in a new window]

  		Table 3. Average Relaxation Time

 

Figure 5
View larger version (114K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a: Complete tumor remission after combined irradiation and chemotherapy in 44-year-old woman. (a) Transverse in vitro T2-weighted turbo spin-echo MR image (4958/105) shows substantial mass effect in left lateral aspect, with intermediate SI that may be interpreted as residual tumor (white arrow). Fibrotic changes are present and are located adjacent to wall and in perirectal fat (black arrow). (b) Corresponding microscopy slice shows smooth muscle hyperplasia (white arrow), without evidence of tumor (stage T0). Spiculations consisted of desmoplastic reaction (black arrow) without tumor cells. m = circular layer of muscularis propria, M = longitudinal layer of muscularis propria. (Elastica–van Giesson stain; original magnification, x1.)

 

Figure 5
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b: Complete tumor remission after combined irradiation and chemotherapy in 44-year-old woman. (a) Transverse in vitro T2-weighted turbo spin-echo MR image (4958/105) shows substantial mass effect in left lateral aspect, with intermediate SI that may be interpreted as residual tumor (white arrow). Fibrotic changes are present and are located adjacent to wall and in perirectal fat (black arrow). (b) Corresponding microscopy slice shows smooth muscle hyperplasia (white arrow), without evidence of tumor (stage T0). Spiculations consisted of desmoplastic reaction (black arrow) without tumor cells. m = circular layer of muscularis propria, M = longitudinal layer of muscularis propria. (Elastica–van Giesson stain; original magnification, x1.)

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 References
 
Most staging errors that occur with MR imaging involve the overestimation of tumor extent and are mainly attributed to desmoplastic reactions, which can cause spiculations in the mesorectal fat that may or may not contain viable tumor cells at histopathologic examination (10). The presence of fibrotic changes, however, can also influence the accuracy of preoperative MR imaging when MR imaging is used to determine the involvement of the mesorectal fascia (17). This is problematic because both the T stage and the distance to the lateral resection margin may serve as inclusion criteria for preoperative (neoadjuvant) treatment strategies.

In more recent studies, T2-weighted MR images were used for the evaluation of tumor extent in relation to the rectal wall layers and the mesorectal fascia (12,13,18,19). Sequences that implement fat-suppression techniques are successfully used in many fields of MR imaging to increase the contrast between normal tissue and pathologic processes. The role of fat-suppressed images for the prediction of tumor penetration through the rectal wall is unclear, and we tested the performance of intermediate-weighted MR imaging with spectral fat saturation, which is known to provide excellent anatomic detail. Theoretically, one might expect that, with increasing spatial resolution, SI differences will become less relevant in relation to morphologic detail.

Normal Architecture and Visual SI Scoring
Our results indicate that T2-weighted MR imaging provides superior contrast for tumor tissue, fibrotic tissue, and anatomic rectal wall layers compared with T1-weighted and intermediate-weighted MR imaging. Intermediate-weighted MR imaging provided excellent morphologic detail of the rectal anatomy but was not superior to T2-weighted MR imaging in the differentiation of rectal wall layers. As one might expect, relatively little overall tissue contrast could be obtained by using intermediate-weighted MR imaging. The SI of the circular and longitudinal muscle layers was virtually identical on intermediate-weighted MR images owing to the intermediate weighting of the sequence. SI differences between tumor and fibrosis were also lower on intermediate-weighted MR images compared with T2-weighted MR images, as established by using visual SI scoring. Furthermore, the presence of fibrosis in the mesorectal compartment and the mesorectal fascia may be obscured with fat saturation.

Substantially less anatomic detail and contrast could be obtained by using T1-weighted MR imaging. The limited value of unenhanced T1-weighted MR images for the evaluation of intramural tumor extent is concordant with the results of other in vitro studies (20,21). Imai et al (21) compared T1-weighted MR images with T2-weighted MR images obtained in fresh and fixed specimens of 18 colorectal tumors and found that the best delineation of bowel wall layers occurred on long-repetition-time long-echo-time MR images. In addition to overall tissue shrinkage, specimens that were prepared with formalin fixation had lower SI in the muscularis propria than did specimens in the fresh state. Thus, we performed MR imaging without formalin fixation in all specimens to avoid potential influences on SI.

For visual SI scoring, T2-weighted, T1-weighted, and intermediate-weighted spectral fat-saturated MR images did not showed differences in SI between tumor tissue and the circular layer of the muscularis propria. This is reflected in the difficulty to detect smaller rectal tumors at the T1 or T2 stage in vivo. The longitudinal muscle layer, however, showed lower SI compared with tumor tissue in most specimens on T2-weighted MR images. These findings are partly in contrast to the results of specimen and in vivo studies conducted by researchers who found no visible differences between the circular and longitudinal layers (12,20,21). Imai et al (21), however, described lower SI in the longitudinal layer in three of 18 cases for which differentiation of the circular and longitudinal layers was possible. The higher resolution used in our study may account for the differences.

T2 Relaxation Time Mapping
T2 mapping was performed on the basis of a multisection multi-echo MR imaging sequence and was used to confirm statistically significant differences in T2 relaxation time between tumor tissue and all other layers of the rectal wall, except for the circular layer of the muscularis propria. Some overlap in T2 relaxation time was noted for the longitudinal layer of the muscularis propria. Thus, smooth muscle hyperplasia associated with radiation therapy may be difficult to distinguish from tumor tissue because of the resulting mass effect.

Limitations
There are limitations to our study that should be noted. A limited number of specimens were investigated, and not all sequences were applied in all specimens. In addition, some sections that demonstrated no tumor or a mixture of tumor and fibrosis were excluded from analysis, and no contrast material–enhanced T1-weighted MR imaging sequences could be performed in our in vitro model. However, although different opinions exist regarding the value of contrast-enhanced MR imaging sequences, there is growing evidence that contrast-enhanced MR imaging does not add substantial information to T2-weighted MR images (22).

The thin-section high-spatial-resolution sequences that were used in our ex vivo study are not suitable for routine in vivo protocols because of the long imaging durations, and the measurements of T2 relaxation time are probably not relevant for in vivo imaging. The ex vivo T2 relaxation times presented in our study, as well as the visual SI differences, may not completely apply to an in vivo situation. Differentiation between the longitudinal and circular layers of the muscularis propria is not relevant for T staging. In addition, no entities other than adenocarcinomas were included in our study (eg, mucinous tumors were excluded because they exhibit different morphologic features at MR imaging depending on mucinous content).

In conclusion, T2-weighted MR imaging provides superior delineation of rectal wall layers and differentiation of tumor from fibrosis compared with T1-weighted and intermediate-weighted spectral fat-saturated MR imaging in our ex vivo model. Despite overlap between the SI of the tumor and that of the circular layer of the muscularis propria, even small lesions could be visualized by using thin-section high-spatial-resolution T2-weighted MR imaging.

Practical application: Further studies, including in vitro imaging with very high-field-strength systems, may improve the delineation of rectal tumors. In this aspect, quantitative measurements of T2 relaxation time may provide an objective basis for comparison of results obtained with 1.5- and 3.0-T systems.


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
References
 


   FOOTNOTES
 

Abbreviations: ROI = region of interest • SI = signal intensity

Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, J.C.S., K.B.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, J.C.S., K.B., S.S., M.S., F.A., A.B.; experimental studies, all authors; statistical analysis, J.C.S., A.S., M.S.; and manuscript editing, all authors


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 

  1. Dukes CE, Bussey HJ. The spread of rectal cancer and its effect on prognosis. Br J Cancer 1958;12:309–320.[Medline]
  2. Harrison JC, Dean PJ, el-Zeky F, Vander Zwaag R. From Dukes through Jass: pathological prognostic indicators in rectal cancer. Hum Pathol 1994;25:498–505.[CrossRef][Medline]
  3. Jass JR, Love SB. Prognostic value of direct spread in Dukes' C cases of rectal cancer. Dis Colon Rectum 1989;32:477–480.[Medline]
  4. Wolmark N, Fisher B, Wieand HS. The prognostic value of the modifications of the Dukes' C class of colorectal cancer: an analysis of the NSABP clinical trials. Ann Surg 1986;203:115–122.[Medline]
  5. Shepherd NA, Baxter KJ, Love SB. Influence of local peritoneal involvement on pelvic recurrence and prognosis in rectal cancer. J Clin Pathol 1995;48:849–855.[Abstract/Free Full Text]
  6. Cawthorn SJ, Parums DV, Gibbs NM, et al. Extent of mesorectal spread and involvement of lateral resection margin as prognostic factors after surgery for rectal cancer. Lancet 1990;335:1055–1059.[CrossRef][Medline]
  7. Willett CG, Badizadegan K, Ancukiewicz M, Shellito PC. Prognostic factors in stage T3N0 rectal cancer: do all patients require postoperative pelvic irradiation and chemotherapy? Dis Colon Rectum 1999;42:167–173.[CrossRef][Medline]
  8. Barrett MW. Chemoradiation for rectal cancer: current methods. Semin Surg Oncol 1998;15:114–119.[CrossRef][Medline]
  9. Colorectal Cancer Collaborative Group. Adjuvant radiotherapy for rectal cancer: a systematic overview of 8,507 patients from 22 randomised trials. Lancet 2001;358:1291–1304.[CrossRef][Medline]
  10. Beets-Tan RG, Beets GL. Rectal cancer: review with emphasis on MR imaging. Radiology 2004;232:335–346.[Abstract/Free Full Text]
  11. Vogl TJ, Pegios W, Mack MG, et al. Accuracy of staging rectal tumors with contrast-enhanced transrectal MR imaging. AJR Am J Roentgenol 1997;168:1427–1434.[Abstract/Free Full Text]
  12. Brown G, Richards CJ, Newcombe RG, et al. Rectal carcinoma: thin-section MR imaging for staging in 28 patients. Radiology 1999;211:215–222.[Abstract/Free Full Text]
  13. Beets-Tan RG, Beets GL, Vliegen RF, et al. Accuracy of magnetic resonance imaging in prediction of tumour-free resection margin in rectal cancer surgery. Lancet 2001;357:497–504.[CrossRef][Medline]
  14. Heald RJ, Husband EM, Ryall RD. The mesorectum in rectal cancer surgery: the clue to pelvic recurrence? Br J Surg 1982;69:613–616.[Medline]
  15. Miles WE. A method of performing abdomino-perineal excision for carcinoma of the rectum and of the terminal portion of the pelvic colon (1908). CA Cancer J Clin 1971;21:361–364.[Free Full Text]
  16. Wood DA. Clinical staging and end results classification: TNM system of clinical classification as applicable to carcinoma of the colon and rectum. Cancer 1971;28:109–114.[CrossRef][Medline]
  17. Brown G, Radcliff AG, Newcombe RG, et al. Preoperative assessment of prognostic factors in rectal cancer using high-resolution magnetic resonance imaging. Br J Surg 2003;90:355–364.[CrossRef][Medline]
  18. Brown G, Kirkham A, Williams GT, et al. High-resolution MRI of the anatomy important in total mesorectal excision of the rectum. AJR Am J Roentgenol 2004;182:431–439.[Abstract/Free Full Text]
  19. Beets-Tan RG, Beets GL. Rectal cancer: how accurate can imaging predict the T stage and the circumferential resection margin? Int J Colorectal Dis 2003;18:385–391.[CrossRef][Medline]
  20. Blomqvist L, Rubio C, Holm T, Machado M, Hindmarsh T. Rectal adenocarcinoma: assessment of tumour involvement of the lateral resection margin by MRI of resected specimen. Br J Radiol 1999;72:18–23.[Abstract]
  21. Imai Y, Kressel HY, Saul SH, et al. Colorectal tumors: an in vitro study of high-resolution MR imaging. Radiology 1990;177:695–701.[Abstract/Free Full Text]
  22. Vliegen RF, Beets GL, von Meyenfeldt MF, et al. Rectal cancer: MR imaging in local staging—is gadolinium-based contrast material helpful? Radiology 2005;234:179–188.



This article has been cited by other articles:


Home page
 Br. J. Radiol.Home page
D M KOH, A DZIK-JURASZ, B O'NEILL, D TAIT, J E HUSBAND, and G BROWN
Pelvic phased-array MR imaging of anal carcinoma before and after chemoradiation
Br. J. Radiol., February 1, 2008; 81(962): 91 - 98.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
I. Yamada, S. Okabe, M. Enomoto, K. Sugihara, N. Yoshino, A. Tetsumura, J. Kumagai, and H. Shibuya
Colorectal Carcinoma: In Vitro Evaluation with High-Spatial-Resolution 3D Constructive Interference in Steady-State MR Imaging
Radiology, December 19, 2007; (2007) 2462070128.
[Abstract] [Full Text]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2411050942v1
241/1/132    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stollfuss, J. C.
Right arrow Articles by Woertler, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stollfuss, J. C.
Right arrow Articles by Woertler, K.

HOME HELP FEEDBACK SUBSCRIPTIONS