HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Goh, V. Articles by Bartram, C. I. Search for Related Content PubMed PubMed Citation Articles by Goh, V. Articles by Bartram, C. I.

Differentiation between Diverticulitis and Colorectal Cancer: Quantitative CT Perfusion Measurements versus Morphologic Criteria—Initial Experience¹

¹ From the Paul Strickland Scanner Centre, Mount Vernon Hospital, Northwood, England (V.G.); Department of Academic Radiology, University College London, Level 2 Podium, University College Hospital, 235 Euston Rd, London NW1 2BU, England (S.H., S.A.T.); and Intestinal Imaging Centre, St Mark's Hospital, Harrow, England (D.B., P.B., C.I.B.). From the 2005 RSNA Annual Meeting. Received October 11, 2005; revision requested November 30; revision received January 26, 2006; accepted February 9; final version accepted May 5. Supported in part by a grant from the Royal College of Radiologists, London, England. Address correspondence to S.H. (e-mail: s.halligan{at}ucl.ac.uk).

	ABSTRACT

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

 
Purpose: To determine whether computed tomographic (CT) perfusion measurements in prospectively recruited patients can be used to differentiate between diverticulitis and colorectal cancer and to compare this discrimination with that of standard morphologic criteria.

Materials and Methods: After institutional review board approval and written informed consent were obtained, 60 patients (24 men, 36 women; mean age, 69 years; range, 33.5–90.4 years; 20 patients with cancer, 20 with diverticulitis, and 20 with inactive diverticular disease) underwent CT perfusion imaging at the level of the colonic abnormality, and perfusion parameters were calculated. Analysis of variance was used to investigate any differences in perfusion between the patient groups. Two independent observers also analyzed an abdominopelvic CT study obtained immediately after the CT perfusion study and noted standard morphologic criteria for differential diagnosis. The sensitivity and specificity of CT perfusion measurements for determining the diagnostic category were compared with morphologic criteria by means of multivariate analysis to identify the most discriminatory criteria.

Results: Mean blood volume, blood flow, transit time, and permeability were significantly different between patients with cancer and those with diverticulitis (P < .0001); patients with cancer had the highest blood volume, blood flow, and permeability and the shortest transit time. The most discriminatory criteria for determining diagnostic category were blood volume, transit time, permeability, and presence of pericolonic nodes (P = .05, .02, .04, and .02, respectively). Blood volume and blood flow each had a sensitivity of 80% and had specificity of 70% and 75%, respectively, for cancer in comparison with standard morphologic criteria: less than 5 cm of bowel involvement (45% sensitivity, 95% specificity), presence of a mass (85% sensitivity, 90% specificity), pericolonic inflammation (75% sensitivity, 5% specificity), and pericolonic nodes (90% sensitivity, 45% specificity).

Conclusion: CT perfusion measurements enable differentiation and better discrimination, in comparison with morphologic criteria, between cancer and diverticulitis.

© RSNA, 2007

	INTRODUCTION

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

 
Computed tomography (CT) is the modality of choice for assessment of suspected diverticulitis. It can be used to confirm the diagnosis, detect complications, and help exclude cancer, which is an important differential diagnosis (1). Differentiation between diverticulitis and cancer can be challenging because of considerable overlap in CT features (2–5). For example, in a prospective study (5) of 72 patients in which the authors used morphologic criteria such as length of involvement, presence of a mass, pericolonic inflammation, and pericolonic nodes, an unequivocal diagnosis of diverticulitis or cancer could be made in only 51% of cases; thus, approximately half of the patients studied needed to undergo further investigation to establish a firm diagnosis.

Quantitative tissue perfusion measurements can be obtained by using dynamic contrast material–enhanced CT, and these measurements provide a measure of vascularity (6). In the abdomen, these measurements have been used to assess colorectal cancer, notably to predict response to therapy, but little attention has been paid to their possible application to differential diagnosis. Estimates of blood flow, volume, and permeability may potentially be used to differentiate tumor from inflammation. For example, findings in pulmonary nodules have indicated that perfusion measurements are considerably higher in malignant nodules than in benign nodules (7). It is possible that a similar distinction is possible in the abdomen and that perfusion measurements may enable better discrimination than do the current established morphologic criteria. Thus, the purpose of our study was to determine whether CT perfusion measurements in prospectively recruited patients can be used to differentiate between diverticulitis and colorectal cancer and to compare this discrimination with that of standard morphologic criteria.

	MATERIALS AND METHODS

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

 
GE Healthcare Technologies (Waukesha, Wis) provided the perfusion software for analysis. The authors retained control of all data collected and information submitted for publication.

Patients
Institutional review board approval was obtained, and all patients gave written informed consent. During a 22-month period, consecutive consenting patients assigned to undergo CT who had a clinical diagnosis of sigmoid colon carcinoma or sigmoid diverticulitis were prospectively recruited, as were those in whom the differential diagnosis between carcinoma and diverticulitis was uncertain. We intended to recruit 20 patients to each group (ie, colon cancer, diverticulitis, and inactive diverticular disease groups), and recruitment was stopped when the numbers in the groups were balanced. All patients were followed up in order to determine a firm diagnostic category.

Colonic adenocarcinoma was confirmed by means of histologic findings obtained at endoscopy or laparotomy. Patients with evidence of sigmoid diverticular disease at CT scanning were further subdivided into groups of those with and those without active inflammation (ie, diverticulitis) on the basis of imaging. However, the final categorization of active or inactive diverticular disease, as well as the exclusion of colon cancer, was confirmed by means of the following: clinical course and ultimate clinical outcome in the patient at medical chart review performed at 8 weeks and then at 1 year after initial presentation, colonoscopy and biopsy, barium enema examination, and laparotomy. Eight patients who did not undergo further confirmatory tests were excluded. Colonoscopy was performed in 27 patients, and barium enema examination was performed in five patients. Colonoscopy and barium enema examination were performed within a median of 62 days after CT (range, 4–325 days). Laparotomy was performed in eight patients, all of whom had diverticulitis. We recruited 60 patients in total (24 men and 36 women; mean age, 69 years; range, 33.5–90.4 years). There were 20 patients with colon cancer, 20 patients with diverticulitis, and 20 patients with inactive diverticular disease (to act as control subjects). The histopathologic T stage of colon cancer was pT2 in two patients, pT3 in 11 patients, and pT4 in seven patients.

CT Scanning
After fasting, patients ingested 1000 mL of water-soluble contrast material (2%–4% meglumine and sodium diatrizoate, Gastrografin; Bracco, Milan, Italy) to opacify the small bowel, per usual clinical practice, 30 minutes prior to CT scanning. An 18-gauge venous cannula was placed in the right antecubital fossa, and 20 mg of the spasmolytic hyoscine-N-butylbromide (Buscopan; Boehringer Ingelheim, Ingelheim am Rhein, Germany) was administered intravenously to abolish bowel peristalsis. Patient movement was minimized by placing a restraining band around the abdomen.

Patients were scanned by using a four–detector row CT scanner (LightSpeed Plus; GE Healthcare Technologies). An abdominopelvic study (section thickness, 10 mm; interval, 5 mm; high-speed mode; speed, 30 mm/sec; 120kV; 180 mA; 0.6-second rotation speed; field of view, 50 cm; and matrix, 512 x 512 mm) was obtained initially, without intravenous contrast material, in order to identify the segment of abnormality within the sigmoid colon. The images were then inspected by the supervising radiologist, and the scan coordinates were noted and were used to plan the acquisition of the subsequent dynamic study. A single radiologist (V.G.) with 5 years of experience in perfusion CT supervised the acquisition of all the perfusion studies.

A pump injector (Percupump Touchscreen; E-Z-Em, Westbury, NY) was used to inject 100 mL iopamidol 340 (Niopam 340; Bracco) intravenously at a rate of 5 mL/sec. Four contiguous sections, collimated to 5 mm, were obtained at 1-second intervals through the midpoint of the segment of interest by using a cine-mode acquisition (120 kV, 60 mA, field of view of 50 cm, and matrix of 512 x 512 mm). Acquisition was commenced 5 seconds after the start of intravenous injection to allow the acquisition of baseline nonenhanced images and was continued for a total duration of 65 seconds. This CT perfusion examination conferred an additional effective dose of 10 mSv.

The dynamic examination was followed with a diagnostic portal venous phase abdominopelvic CT examination that was started 75 seconds after the commencement of intravenous injection by using the following parameters: section thickness, 5 mm; section interval, 2.5 mm; high-speed mode; speed, 22.5 mm/sec; 120 kV; 280 mA; 0.6-second rotation speed; field of view, 50 cm; and matrix, 512 x 512 mm. This study was reviewed by a staff radiologist, and a report was issued as per usual day-to-day clinical practice. A total of six radiologists (V.G., S.H., and four nonauthor radiologists) with a mean of 13.5 years of experience in abdominal CT (range, 6–26 years) reported the findings for these diagnostic scans.

Image Analysis
Two radiologists (S.A.T., D.B., with 6 and 4 years of experience reporting abdominal CT findings, respectively) read all the portal venous studies independently by using a stand-alone workstation (Advantage 4.1; GE Healthcare Technologies). The following criteria were evaluated and recorded for each patient: bowel wall thickening of more than 5 mm, length of bowel wall involvement (<5 cm, 5–10 cm, or >10 cm), presence of a localized mass or abscess, presence of pericolonic stranding and/or edema, presence of pericolonic lymph nodes with a diameter exceeding 5 mm, and presence of large-bowel obstruction.

The radiologists were unaware of the underlying diagnosis and attempted to arrive at an overall diagnosis (cancer, diverticulitis, or inactive diverticular disease) by using morphologic criteria, which were derived from previous studies. Criteria used to establish a diagnosis of cancer included presence of a localized soft-tissue mass, presence of enlarged pericolonic nodes, length of bowel involvement of less than 5 cm, and presence of bowel obstruction (4,5,8). Criteria used to establish a diagnosis of diverticulitis included presence of pericolonic stranding and/or edema, presence of an abscess, and longitudinal extent of more than 10 cm (3,5,9,10). Criteria used to establish a diagnosis of inactive diverticular disease were presence of diverticula, with or without bowel wall thickening, with no evidence of active inflammation.

All 60 perfusion studies were then analyzed by a third observer (V.G., with 5 years of experience interpreting perfusion CT images) by using commercially available software (Body Tumor Perfusion 3.0; GE Healthcare Technologies). The initial 65-second dynamic study was downloaded onto the software, and the single 5-mm transverse image that best depicted the abnormal colonic segment was chosen from the four image levels available. A processing threshold of 0–120 HU was selected so that the subsequent analysis appropriately included both nonenhanced and contrast-enhanced soft tissue. The arterial input was determined by placing a circular region of interest with a mean size of 10 mm² within the iliac artery that was best visualized on the image. An arterial time-enhancement curve for the 65-second acquisition was generated automatically along with parametric maps of blood volume, blood flow, mean transit time, and permeability that showed all tissues within the selected processing threshold. A region of interest was then drawn freehand around the entire colonic segment of interest by using an electronic cursor and computer mouse. Care was taken to exclude pericolonic fat and intraluminal gas, a process that was facilitated by viewing a cine loop of the acquisition in order to gauge the degree and margins of patient movement. A tissue time-enhancement curve and the four perfusion parameters (blood volume, blood flow, mean transit time, and permeability) were derived automatically for the selected region of interest. Mean values for the four perfusion parameters were recorded for each patient.

Statistical Analysis
The reference category for each patient was determined as described previously: cancer, diverticulitis, or inactive diverticular disease. The mean ± standard deviation for blood volume, blood flow, mean transit time, and permeability were determined for each group, and values were compared by means of one-way analysis of variance by using a software package (Stata 7.0; Stata, College Station, Tex). Posttesting was performed by using the Bonferonni method. To determine which criteria (both morphologic and functional criteria) could best be used to discriminate between the groups, multivariate analysis was performed by using multinomial logistic regression or, if this was not possible, by using the Fisher exact test. A backward selection procedure was then used to identify which of the criteria were most discriminatory by removing nonsignificant variables one at a time until all remaining variables were statistically significant. Observer agreement for morphologic criteria was assessed by using analysis. Statistical significance was assigned at the 5% level. Sensitivity, specificity, and predictive values of all morphologic and functional criteria for distinguishing cancer from diverticulitis were also determined.

	RESULTS

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

 
Perfusion Measurements
The mean values for blood volume, blood flow, mean transit time, and permeability for the three patient groups (ie, the colon cancer, diverticulitis, and inactive diverticular disease groups) showed a significant difference (Table 1) for all four perfusion measurements when the three clinical categories of patients were compared (P < .0001) (Figure). Overall, patients with established cancer had the highest blood volume, blood flow, and permeability and the fastest transit time. Conversely, patients with inactive diverticular disease had the lowest blood volume, blood flow, and permeability and the slowest transit time. Values in patients with diverticulitis were intermediate between the two.

View larger version (16K): [in this window] [in a new window] [Download PPT slide]   Distribution of CT-derived perfusion parameters for individual patients with cancer, diverticulitis, and inactive diverticular disease (mean value is indicated for each group). A, Blood volume, B, blood flow, and, C, permeability were highest, and, D, transit time was shortest, in patients with cancer. Blood volume, blood flow, and permeability were lowest, and transit time was longest, in those with diverticular disease. P  values for group-to-group comparisons are also shown. Mean blood volume, flow, and permeability measurements were significantly different between patients with cancer and those with diverticulitis.

Multivariate Analysis
Multivariate analysis of morphologic features demonstrated that although the majority of criteria were found to be discriminatory initially, including length of bowel involved, presence of a localized mass, presence of pericolonic stranding and edema, and presence of an abscess (P < .001), when combinations of criteria were evaluated and adjusted for effect, four criteria were found to be the strongest discriminators between cancer and diverticulitis: blood volume (P = .05), mean transit time (P = .02), permeability (P = .04), and presence of pericolonic nodes (P = .02).

	DISCUSSION

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

 
CT has emerged as the technique of choice for assessment of acute diverticulitis and is also increasingly used as the initial investigation in patients suspected of having cancer. However, difficulties in distinguishing between these two diseases are well recognized (2–5). For example, bowel wall thickening and pericolonic stranding occur in both (10,11). Authors have therefore attempted to determine which features have the highest predictive value by performing retrospective studies. For example, Padidar et al (3) found that fluid in the sigmoid mesentery and engorgement of mesenteric vessels had positive predictive values of 89% and 100%, respectively, for diverticulitis. Chintapalli et al (4) found that pericolonic lymph nodes had a 71% sensitivity and an 85% specificity for cancer. However, for the individual patient, highly specific morphologic criteria remain elusive. For example, when morphologic features known to be discriminatory were applied in a prospective cohort of 72 patients (5), an unequivocal diagnosis of diverticulitis or cancer could be made in only 51% of patients.

We determined whether functional CT measurements differ between patients with cancer and those with diverticulitis. We also investigated the diagnostic potential of these measurements in comparison with established morphologic assessments. In keeping with findings of previous studies of inflammatory and malignant lung nodules (7,12–14), our findings demonstrated significant differences in vascular perfusion among all three groups of patients studied. Patients with cancer had significantly increased blood volume, blood flow, and permeability and accelerated transit time in comparison with control subjects with inactive diverticular disease. These findings were also observed in patients with diverticulitis but to a lesser extent. In our study, the sensitivity and specificity achieved by using threshold values based on our data were better than the sensitivities and specificities of the majority of morphologic criteria, with the exception of the presence of a localized mass. Although permeability had lower sensitivity than did blood volume and flow, it was the most specific perfusion measurement, at 90%. Despite significant differences in mean group values, there was considerable overlap that will inevitably limit the clinical utility of CT perfusion imaging in individual patients until such time that this prognostic model is validated further by means of larger prospective studies. Until then, our findings should be viewed cautiously.

Perfusion changes in cancer reflect its proangiogenic nature. It is now widely accepted that angiogenesis is central to tumor growth (15), although other mechanisms, such as co-option of existing vessels, are contributory (16). CT estimates of perfusion reflect the functional status of the tumor vascular bed, including vessel proliferation, leakiness of immature vessels, and intratumoral shunting. The inflammation associated with diverticulitis also increased local blood flow values above those of the control group with inactive diverticular disease, which reflects attempted repair: Cytokines provoke vasodilation, increase vascular permeability, recruit immune cells, and trigger healing. On CT scans, vasodilation is reflected as engorged mesenteric vessels in up to 30% of patients (3). However, the fact that angiogenesis is lacking in acute inflammation may explain why perfusion was less on average in these patients than in those with cancer.

Only one morphologic criterion, the presence of pericolonic lymph nodes, was identified as a strong predictor at multivariate analysis. In contrast, three functional criteria were identified as strong predictors, supporting the notion that functional measurements may be better discriminators. However, it is interesting that both of our observers found high sensitivity and specificity for both cancer and diverticulitis when asked to assign a diagnostic category on the basis of their overall impression of the CT scan: Only one cancer was misinterpreted as acute diverticulitis. In contrast, inactive diverticular disease was more frequently misclassified. Of importance, if observers had access to perfusion measurements in concert with the CT scans, the one case of cancer that was misinterpreted as acute diverticulitis would have been correctly classified, the two cases of acute diverticulitis misinterpreted as cancer would have been correctly classified, and the four cases of inactive diverticular disease that were misclassified as cancers would also have been correctly classified, which indicates the potential additional benefit of perfusion measurements.

While the prevalence of individual morphologic abnormalities was similar to that in previous studies (5,10), we found the interobserver agreement was generally poor, which is a surprising observation given the accuracy with which an overall diagnostic category of cancer or diverticulitis was reached. However, we did find good observer agreement for the presence of pericolonic nodes, which was one of the few individual discriminatory morphologic criteria we identified. While in our present study we did not investigate observer agreement for perfusion measurements, authors of previous studies (17,18) have found this to be acceptable, with intraclass correlation coefficients of greater than 0.8. This may be because the estimate is less subjective than morphologic assessments.

Our study had limitations. Although the study was prospective, discrimination among groups was defined by means of retrospective analysis of the data, which is inevitable when one is developing a prognostic model (19). Our thresholds will need prospective validation, and, until this validation is performed, the clinical utility remains unproved. Patient numbers were relatively small, which may mask or enhance the differences we observed because of wider confidence intervals. Also, data acquisition occurred at a single tissue level and comprised acquisition of four contiguous 5-mm transverse CT images with a z-axis tissue coverage of 2 cm. Inevitably, this acquisition did not encompass the entire bowel segment of interest. It has been shown that tumor perfusion is both temporally and spatially heterogeneous (20) and that whole-tumor assessment may provide more reliable assessment (21). However, whether this translates into an improvement in diagnostic assessment requires further investigation.

In summary, we found that functional perfusion CT measurements enable differentiation and better discrimination, in comparison with morphologic criteria, between cancer and diverticulitis. In clinical practice, a combination of CT perfusion measurement and established morphologic evaluation may facilitate the differential diagnosis.

	ADVANCES IN KNOWLEDGE

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

 

• We demonstrated that quantitative perfusion measurements help differentiate between diverticulitis and cancer and may be more discriminatory than morphologic criteria.
  
• We documented the range of perfusion measurements in diverticulitis, colon cancer, and inactive diverticular disease.
  

	FOOTNOTES

 
See Materials and Methods for pertinent disclosures.

Author contributions: Guarantors of integrity of entire study, V.G., S.H.; 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, V.G.; clinical studies, V.G., S.A.T., D.B.; statistical analysis, V.G., P.B.; and manuscript editing, V.G., S.H., C.I.B.

	References

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

 

1. Balthazar EJ. Diverticular disease. In: Gore RM, Levine MS, Laufer I, eds. Textbook of GI radiology. Philadelphia, Pa: Saunders, 1994; 1072–1095.
2. Balthazar EJ, Megibow A, Schinella RA, Gordon R. Limitations in the CT diagnosis of acute diverticulitis: comparison of CT, contrast enema, and pathologic findings in 16 patients. AJR Am J Roentgenol 1990;154:281–285.[Abstract/Free Full Text]
3. Padidar AM, Jeffrey RB Jr, Mindelzun RE, Dolph JF. Differentiating sigmoid diverticulitis from carcinoma on CT scans: mesenteric inflammation suggests diverticulitis. AJR Am J Roentgenol 1994;163:81–83.[Abstract/Free Full Text]
4. Chintapalli KN, Esola CC, Chopra S, Ghiatas AA, Dodd GD. Pericolic mesenteric lymph nodes: an aid to distinguishing diverticulitis from cancer of the colon. AJR Am J Roentgenol 1997;169:1253–1255.[Abstract/Free Full Text]
5. Chintapalli KN, Chopra S, Ghiatas AA, Esola CC, Fields SF, Dodd GD. Diverticulitis versus colon cancer: differentiation with helical CT findings. Radiology 1999;210:429–435.[Abstract/Free Full Text]
6. Miles KA, Hayball M, Dixon AK. Colour perfusion imaging: a new application of computed tomography. Lancet 1991;337:643–645.[Medline]
7. Zhang M, Kono M. Solitary pulmonary nodules: evaluation of blood flow patterns with dynamic CT. Radiology 1997;205:471–478.[Abstract/Free Full Text]
8. Xiong L, Chintapalli KN, Dodd GD 3rd, et al. Frequency and CT pattern of bowel wall thickening proximal to cancer of the colon. AJR Am J Roentgenol 2004;182:905–909.[Abstract/Free Full Text]
9. Pradel JA, Adell JF, Taourel P, Djafari M, Monnin-Delhom E, Bruel JM. Acute colonic diverticulitis: prospective comparative evaluation with US and CT. Radiology 1997;205:503–512.[Abstract/Free Full Text]
10. Kircher MF, Rhea JT, Kihiczak D, Novelline RA. Frequency, sensitivity, and specificity of individual signs of diverticulitis on thin section helical CT with colonic contrast material: experience with 312 cases. AJR Am J Roentgenol 2002;178:1313–1318.[Abstract/Free Full Text]
11. Horton KM, Corl FM, Fishman EK. CT evaluation of the colon: inflammatory disease. RadioGraphics 2000;20:399–418.[Abstract/Free Full Text]
12. Swensen SJ, Brown LR, Colby TV, Weaver AL, Midthun DE. Lung nodule enhancement at CT: prospective evaluation. Radiology 1996;201:447–455.[Abstract/Free Full Text]
13. Swensen SJ, Viggiano RW, Midthun DE, et al. Lung nodule enhancement at CT: multicenter study. Radiology 2000;214:73–80.[Abstract/Free Full Text]
14. Yi CA, Lee KS, Kim EA, et al. Solitary pulmonary nodules: dynamic enhanced multi–detector row CT study and comparison with vascular endothelial growth factor and microvessel density. Radiology 2004;233:191–199.[Abstract/Free Full Text]
15. Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med 1971;285:1182–1186.[Medline]
16. Vermeulen PB, Gasparini G, Fox SB, et al. Quantification of angiogenesis in solid human tumours: an international consensus on the methodology and criteria of evaluation. Eur J Cancer 1996;32A:2474–2484.[CrossRef]
17. Fiorella D, Heiserman J, Prenger E, Partovi S. Assessment of the reproducibility of postprocessing dynamic CT perfusion data. AJNR Am J Neuroradiol 2004;25:97–107.[Abstract/Free Full Text]
18. Goh V, Halligan S, Hugill JA, Bassett P, Bartram CI. Quantitative assessment of colorectal cancer perfusion using MDCT: inter- and intra-observer agreement. AJR Am J Roentgenol 2005;185:225–231.[Abstract/Free Full Text]
19. Altman DG, Royston P. What do we mean by validating a prognostic model. Stat Med 2000;19:453–473.[CrossRef][Medline]
20. Endrich B, Vaupel P. The role of the microcirculation in the treatment of malignant tumors: facts and fiction. In: Molls M, Vaupel P, eds. Blood perfusion and microenvironment of human tumors: implications for clinical radiooncology. Berlin, Germany: Springer-Verlag, 2001; 19–39.
21. Ng QS, Goh V, Klotz E, et al. Quantitative measurement of lung cancer perfusion using MDCT: does whole tumour analysis improve measurement reproducibility. Presented at the 105th annual meeting of the American Roentgen Ray Society, New Orleans, La, May 15–20, 2005.

This article has been cited by other articles:

				R. G. Sheiman, A. Sitek, V. Goh, and S. Halligan CT Perfusion Imaging: Know Its Assumptions and Limitations Radiology, February 1, 2008; 246(2): 649 - 650. [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Goh, V. Articles by Bartram, C. I. Search for Related Content PubMed PubMed Citation Articles by Goh, V. Articles by Bartram, C. I.