Utility of apparent diffusion coefficient in characterizationof different sinonasal pathologies

Mohamed S. Taha, M.D.,1 Lobna M. El Fiky, M.D.,1 Togan M. Taha, M.D.,2 Reda M. Sabra, M.D.,1

Tamer A. Youssef, M.D.,1 and Ihab M. Nada, M.D.3

ABSTRACTBackground: Sinonasal lesions are a heterogeneous group of lesions that span from a tumor to tumor-like nature. Characterization of such cases

preoperatively can improve the surgical control and the overall outcome of these patients.Objective: In this prospective study, we aimed at evaluation of the role of apparent diffusion coefficient (ADC) in the differentiation between benign and

malignant sinonasal lesions.Subjects and methods: All patients scheduled to have sinonasal surgical intervention at Ain Shams University Hospitals, Cairo, Egypt, were enrolled.

Diffusion-weighted (DW) magnetic resonance imaging (MRI) with calculation of ADC were done for all cases. Radiologic findings were then compared withhistologic findings, and the sensitivity, specificity, negative and positive predictive values (PPVs) of the conventional MRI, DW-MRI, and ADC value indifferentiation of benign from malignant sinonasal lesions were then calculated.

Results: There were 59 patients with median age of 43 years old. There were 20 cases of inflammatory lesions, 16 cases of benign tumors, and 23 cases ofmalignant lesions. The ADC values of all cases ranged from 0.4 � 10�3 to 2 � 10�3 (median � 1.5 � 10�3). The median ADC value for the malignant lesionswas 0.6 � 10�3, whereas that for the inflammatory conditions was 1.6 � 10�3 and that for the benign tumors was 1.5 � 10�3 with a highly significantdifference (p � .001). Analysis of the conventional MRI and DW-MRI to differentiate between malignant and benign lesions showed that the sensitivity,specificity, PPV, and negative predictive value (NPV) were 100%, 97%, 96%, and 100% and 91%, 97%, 95%, and 95%, respectively.

Conclusion: DW-MRI did not add significantly to the information gained from conventional MRI. It should be considered complimentary only to standardMRI in uncertain cases when malignancy is still a concern.

(Am J Rhinol Allergy 28, e181–e186, 2014; doi: 10.2500/ajra.2014.28.4098)

The sinonasal region has a very wide platform of pathologies thatstretch from various inflammatory lesions to benign and ma-

lignant diseases.1

Gross appearance of the sinonasal tract masses has limited value indifferentiating between different pathologies. Image-based differen-tiation is not always reliable, due to the wide variability of invasive-ness in different pathologies. The final diagnosis usually depends onthe pathological tissue result obtained by biopsy when feasible pre-operatively or intraoperatively during excisional biopsy.2–4

Computer tomography (CT) scan highlights bony structures, withbony destruction usually being indicative of an aggressive lesion.Magnetic resonance imaging (MRI) is superior to CT scan in its softtissue delineation. It has the ability to delineate sinus tumors frominflammatory disease, and it can better delineate tumor from theadjacent soft tissues.1 However, the imaging differentiation betweenbenign and malignant sinonasal lesions is often nonspecific especiallyin early cases.5

Recent studies reported that apparent diffusion coefficient (ADC)measurement in diffusion weighted (DW) MRI was useful in thedifferentiation of benign and malignant lesions of lymph nodes andsalivary glands.6 Low ADCs indicate limited diffusion of water mol-ecules in the tissue. Theoretically, therefore, a tumor or area within atumor with low ADC contains a greater number of cells than one withhigh ADC. Consistent with this notion, malignant masses have lowerADCs than benign lesions.7 These results suggest that DW-MRI maybe applicable for the differentiation of benign and malignant sinona-sal lesions. In this prospective study, we aimed at evaluation of therole of ADC in the differentiation between benign and malignantsinonasal lesions.

SUBJECTS AND METHODSAll patients scheduled to have sinonasal surgical intervention, for

newly diagnosed sinonasal lesions who presented to the ear, nose andthroat department of Ain Shams University Hospitals, Cairo, Egypt, atertiary care center, in the period from November 2011 to December2013, were enrolled in this prospective study. An informed consentwas signed by all participants after approval of the InstitutionalReview Board of Ain Shams University Hospitals. Before surgery, allparticipants had routine preoperative investigations, contrast-en-hanced CT scan, conventional MRI, and DW-MRI.

The consultant radiologist, who reviewed the scans, was asked tocomment on the different radiologic findings routinely, in addition tocalculate the ADC value of every patient. All pathology reportsavailable beforehand were blinded to the radiologist.

Radiologic findings were then compared with histologic findingsand the sensitivity, specificity, negative, and positive predictive val-ues (PPVs) of the conventional MRI, DW-MRI, and ADC value indifferentiation of benign from malignant sinonasal lesions were thencalculated.

MRI ProtocolThe study was done on a 1.5T MRI unit (Philips Healthcare) using

surface coil coverage from skull base to supraclavicular fossa, se-quences.

For conventional MRI, we obtained axial T1-weighted (time ofrepetion [TR]/echo time [TE]/number of signal intensity acquisi-tions � 500/15 ms/2) and fat suppression techniques (short tauinversion recovery [SPIR] or spectral selected attenuation inversionrecovery [SPAIR]) T2-weighted (TR/TE/number of signal intensityacquisitions 4784/80 ms/2 for SPIR; 6385/80 ms/2 for SPAIR) MRimages of the sinonasal lesions by using a multishot turbo spin echosequence with a turbo spin echo factor of 3 (T1-weighted) or 15(SPAIR). We used a 200-mm field of view, 256 � 224 scan and 512 �512 reconstruction matrix sizes, 4-mm section thickness, and 0.4-mmsection gap. Axial DW images (TR/TE/number of signal-intensity ac-quisitions � 4285/87 ms/4) were obtained by using single-shot spin-echo echo-planar imaging with an echo-planar imaging factor of 47.8,9

Departments of 1Otorhinolaryngology and 2Radiodiagnosis, Faculty of Medicine, AinShams University, and 3Department of Otorhinolaryngology, Faculty of Medicine, MisrUniversity for Science and Technology University, Cairo, EgyptThe authors have no conflicts of interest to declare pertaining to this articleAddress correspondence to Lobna M. El Fiky, M.D., 48 Ibn El Nafees Street, 6thDistrict, Madinet Nasr, 11371, Cairo, EgyptE-mail address: elfikylobna@hotmail.comCopyright © 2014, OceanSide Publications, Inc., U.S.A.

American Journal of Rhinology & Allergy e181

Image Analysis for DW-MRISequential gray-scale ADC map images that spanned the whole

lesion, except for the upper- and lower-most sections, and freehandregions of interest (ROIs) along the margins of the lesions, weremanually placed onto the ADC maps by using the corresponding fatsuppression techniques T2-weighted MR images as references forplacing the ROI. Then, average ADCs of the whole lesions weredetermined (� overall ADCs).9

Statistical MethodsThis was done on a personal computer using MedCalc for Win-

dows, version 12.5 (MedCalc Software, Ostend, Belgium) and DAG-Stat software. The D’Agostino-Pearson test was performed to test thenormality of numerical data distribution, presented as median. TheMann-Whitney U test was used for comparison of differences be-tween two groups, whereas the Kruskal-Wallis test was used tocompare differences among multiple groups. Qualitative data werepresented as number and percentage, and the �2 test was used forintergroup comparisons.10,11

The performance of MRI or DW-MRI for diagnosis of malignantconditions was examined using two-by-two contingency tables with

histopathologic diagnosis considered as the gold-standard for diag-nosis. The following quality indices were calculated: sensitivity, spec-ificity, PPV, and negative predictive value (NPV).

Receiver-operating characteristic (ROC) curve analysis was used toidentify the best cutoff criterion for identification of malignant le-sions. All p values were two-tailed; p � .05 was considered as statis-tically significant.

RESULTSA total of 59 patients were included in this study. Their age ranged

from 5 to 65 years (median � 43). A total of 43 (72.9%) of them weremales, whereas 16 (27.1%) were females. The final histopathology ofthe sinonasal lesions was inflammatory lesions in 20 cases (33.9%),benign tumors in 16 cases (27.1%), and malignant tumors in 23 cases(39%). The inflammatory lesions included 13 cases of nasal polypi, sixcases of fungal sinusitis, and one case of mucocele. Benign tumorsincluded seven cases of inverted papilloma, five cases of juvenileangiofibroma, two cases of ossifying fibroma, one case of extracranialmeningioma, and one case of schwannoma. For the malignant tu-mors, there were eight cases of squamous cell carcinoma, five cases oflymphoma, three cases of adenocarcinoma, three cases of sinonasal

Figure 1. Chronic invasive fungal sinusitis. (A and B) Coronal and axial postcontrast T1 fat saturated images showing mass involving right ethmoid andfrontal sinuses and right orbit (dotted arrow), mimicking tumor (C) DW imaging and (D) ADC value of 1.5 �10�3. ADC map displaying facilitated diffusion,mostly not malignancy.

Table 1. Descriptive statistics for the whole study population: categorical and numerical variables

Pathology Number (%) Male Female Median Age (years) Median ADC

Inflammatory lesions 20 (33.9) 13 (35%) 7 (65%) 33 1.6 � 10�3

Nasal polypi 13 (65%)Fungal sinusitis 6 (30%)Mucocele 1 (5%)Benign tumors 16 (27.1) 13 (81.3%) 3 (18.7%) 46 1.5 � 10�3

Inverted papilloma 7 (43.75%)Juvenile angiofibroma 5 (31.25%)Ossifying fibroma 2 (12.5%)Meningioma 1 (6.25%)Schwannoma 1 (6.25%)Malignant tumors 23 (39%) 17 (73.9%) 6 (26.1%) 53 0.6 � 10�3

SCC 8 (34.7%)Lymphoma 5 (21.7%)Adenocarcinoma 3 (13%)SNUC 3 (13%)Chondrosarcoma 2 (8.6%)Osteosarcoma 2 (8.6%)

ADC � apparent diffusion coefficient; SCC � squamous cell carcinoma; SNUC � sinonasal undifferentiated carcinoma.

e182 September–October 2014, Vol. 28, No. 5

undifferentiated carcinoma, two cases of chondrosarcoma, and twocases of osteosarcoma (Table 1).

The ADC values of all cases range from 0.4 � 10�3 to 2 � 10�3

(median � 1.5 � 10�3). There was an insignificant difference in ADCvalue across the various benign lesions (Fig. 1). The highest ADC valuewas in patients with allergic polyposis (2 � 10�3), whereas the lowestADC value was seen in a patient with mucocele (0.4 � 10�3) (Fig. 2).

There was also an insignificant difference in ADC value across thevarious malignant masses (Fig. 3). The lowest ADC value was seen inpatients with undifferentiated carcinoma (0.5 � 10�3) (Fig. 4),whereas the highest ADC value was seen in patient with squamouscell carcinoma developing in a recurrent inverted papilloma case(1.4 � 10�3).

The median ADC value for the malignant lesions was 0.6 � 10�3,whereas that for the inflammatory conditions was 1.6 � 10�3 and that forthe benign tumors was 1.5 � 10�3 with a highly significant difference(p � .001). When we combined inflammatory and benign tumors, themedian ADC of all benign conditions was 1.6 � 10�3 (p � .001).

Using ROC curve analysis, a cutoff ADC value of 1.3 � 10�3 wasused to differentiate benign from malignant lesions (Fig. 5). Ac-cordingly, all cases were properly diagnosed, except for the mu-cocele case misdiagnosed as malignant lesion and the case of SCCdeveloping in a recurrent inverted papilloma, misdiagnosed as abenign lesion.

Analysis of the conventional MRI to differentiate between malig-nant and benign lesions showed that the sensitivity, specificity, PPV,

Figure 2. Mucocele. Expansion of the right maxillary and ethmoid sinuses and nasal cavity by a nondestructive lesion (arrows), with significant low signalin (A) axial and (B) coronal T2 weighted images. (C) T1 weighted images coronal high signal likely due to viscid, protein-rich secretions. (D) DW imagingshows diffusion restriction with low ADC value of 0. 4 � 10�3. False �ve result for malignancy.

Figure 3. Undifferentiated carcinoma.(A) Axial and (B) coronal T2 weightedimages demonstrating soft tissue mass im-plicating maxillary and ethmoid sinusesbilaterally, left nasal cavity, with exten-sion to both orbits (arrows). (C) DW im-aging and (D) ADC map restricted diffu-sion with ADC value of 0.9 � 10�3

consistent with malignancy.

American Journal of Rhinology & Allergy e183

and NPV were 100%, 97%, 96%, and 100%, respectively. The diagnos-tic performance of DW-MRI for identification of malignant lesionsshowed that the sensitivity, specificity, PPV, and NPV were 91%, 97%,95%, and 95%, respectively (Table 2).

DISCUSSIONFor diagnosis, staging, and monitoring of sinonasal tumors and

tumor-like lesions, conventional MRI is the preferred imaging modal-ity due to its superior soft tissue contrast resolution compared with

CT scan.12 Both conventional MRI and CT are limited in terms ofdifferentiation between malignant and benign sinonasal lesions dueto the overlapping signal characteristics and enhancement of theselesions after contrast administration.13 DW imaging (DWI) is a non-invasive, functional imaging technique that is used to measure dif-ferences in tissue microstructural changes that are based on randomdisplacement of water molecules. These differences in water mobilityare quantified by using ADC measurement.14–19 Many recent studiesreported that DWI has been used to differentiate benign and malig-nant head and neck masses, to characterize cervical lymph nodes,parotid tumors, thyroid nodules, and sinonasal masses.20–35 The ADCobtained by DW-MRI is a marker of cell density and potentially maydistinguish malignant from benign lesions with significant lowerADC for malignant lesions than for benign lesions.13–18,35,36 In thepresent study, the median ADC value for malignant sinonasal tumorswas significantly lower (p � � .001) than that for benign lesions.

We calculated an optimal ADC cutoff value of 1.3 �10�3, whichwas best used for differentiating benign lesions from malignant le-sions of the sinonasal tract. Comparable threshold ADC values havebeen reported in the literature for other head and neck lesions.6,16,19–21

However, we should note that the correlation of the histopathologyof a lesion with ADCs is quite complex and affected by numeroustumor components beside cellularity, including the presence of ker-atin, collagen, myxoid stroma, and necrosis.36 Overlapping of ADCvalues between benign and malignant lesions can thus be found.

The lowest ADC value that we obtained in the benign group wasseen in a patient with mucocele (0.4 � 10�3) and thus misdiagnosedas a malignant tumor by ADC map. This is likely due to the presenceof viscid protein-rich secretions in such a lesion.

Vascular lesions can also have higher ADC values than other be-nign solid tumors, due to the excess extracellular spaces, the freediffusion within vascular lesions, the rate of perfusion, and hemosid-erin deposition.34,37 In the same time, excess fibrous tissue can lead tosubsequent restriction of diffusion.38

There was an insignificant difference in ADC value across thevarious malignant masses, even between carcinoma and lymphoma,in conjunction with the study by Ginat et al.21 However, a statisticallysignificant difference in the ADC values of carcinoma versus lym-phoma in the head and neck lesions was found in several otherstudies.7,20,39,40 Because lymphoma cells have relatively high nuclear-to-cytoplasm ratios and are densely packed, ADC value might belower than that of carcinoma, which has more areas of breakdownand necrosis.20,33 In our series, the patients of carcinoma did not haveexcessive areas of necrosis, explaining the inability to differentiatebetween carcinoma and lymphoma by ADC values. This situation canbe avoided using a morphologic images with the same geometry asDW images, to compare morphologic and functional images, allow-ing exact localization of the lesion and avoiding necrotic areas for ROI

Figure 4. Mandibular osteosarcoma. (A and B) Axial T2 weighted images and (C) coronal T2 weighted images show aggressive malignant mass arising fromthe ramus of right mandible, invading the hard palate, alveolar ridge of maxilla, right nasal cavity, and right orbit. (D) DW imaging malignant mass showsrestricted diffusion with low ADC value of 0.6 � 10�3 consistent with malignancy.

Figure 5. ROC curve for diagnosis of malignant lesions using ADC in thestudied sample. Area under the ROC curve (AUC) � 0.967 (p � .0001).Best cutoff criterion is ADC 1.3 � 10�3 (sensitivity � 95.65%, specificity �94.44%).

Table 2. Diagnostic performance of conventional MRI, DW-MRI,and AUC for diagnosis of malignant lesions using ADC foridentification of malignant lesions

Estimate MRI DW-MRI AUC using ADC

Sensitivity (%) 100 91 95.65Specificity (%) 97 97 94.44PPV (%) 96 95 91.75NPV (%) 100 95 91.1

ADC � apparent diffusion coefficient; MRI � magnetic resonance imaging;DW � diffusion weighted; AUC � area under the receiver-operating char-acteristic curve; PPV � positive predictive value, NPV � negative predic-tive value.

e184 September–October 2014, Vol. 28, No. 5

delineation.41 Addition of DW-MRI and ADC mapping during rou-tine MRI examination can be done without the need of extra time orcontrast injection.42

In our study, the lowest ADC value in the malignant group wasseen with undifferentiated carcinoma. This is in accordance withSumi et al.7 In undifferentiated carcinomas, the tumor cells are oftenoverlapping, with greater cellularity and, subsequently, restricteddiffusion resulting in ADCs significantly lower than those of moder-ately differentiated and well-differentiated carcinomas.20,33,39,40

The use of conventional MRI for differentiation between benignand malignant lesions in our study was very accurate with a speci-ficity reaching 97%. However, this usually depends on surroundingextrasinus extension without definite tissue characterization, makingspecific diagnosis only possible in advanced stages.43 The addition ofDW-MRI and ADC mapping can thus add to the specific diagnosis ofdifferent pathologies of the sinonasal area, especially in early caseslimited to the sinuses.44

In conclusion, the specific ability of DW-MRI to probe tissue mi-crostructure is an interesting complement to the currently used im-aging procedures in the characterization and even grading of malig-nancies. In our series, DW-MRI did not add clinical information in thework-up of most patients with sinonasal neoplasms. DW-MRI mighthave a narrow indication in patients when conventional MRI findingsare not consistent with the clinical picture, malignancy remains aconcern, and the patient is a poor surgical candidate for a biopsy orresection. Adding DW-MRI to conventional MRI will increase thecertainty in the tissue characterization of the sinonasal lesions. ADCmapping is an easy, promising tool in helping characterization ofsinonasal lesions.

REFERENCES1. Madani G, Beale TJ, and Lund VJ. Imaging of sinonasal tumors.

Semin Ultrasound CT MR 30:25–38, 2009.2. Pilch BZ, Bouquot J, and Thompson LDR. Squamous cell carcinoma.

In Pathology and genetics of head and neck tumors. Barnes EL,Eveson JW, Reichart P, et al. (Eds). Kleihues P, and Sobin LH (SeriesEds). World Health Organization Classification of Tumors. Lyon,France: IARC Press, 15–17, 2005.

3. McNicoll W, Hopkin N, Dalley VM, et al. Cancer of the paranasalsinuses and nasal cavities. Part II. Results of treatment. J LaryngolOtol 98:707–718, 1984.

4. Jackson RT, Fitz-Hugh GS, and Constable WC. Malignant neoplasmsof the nasal cavities and paranasal sinuses: (a retrospective study).Laryngoscope 87:726–736, 1977.

5. Eida S, Sumi M, Sakihama N, et al. Apparent diffusion coefficientmapping of salivary gland tumors: prediction of the benignancy andmalignancy. AJNR Am J Neuroradiol 28:116–121, 2007.

6. Sumi M, and Nakamura T. Diagnostic importance of focal defects inthe apparent diffusion coefficient-based differentiation between lym-phoma and squamous cell carcinoma nodes in the neck. Eur Radiol19:975–981, 2009.

7. Sumi M, Ichikawa Y, and Nakamura T. Diagnostic ability of apparentdiffusion coefficients for lymphomas and carcinomas in the pharynx.Eur Radiol 17:2631–2637, 2007.

8. Biomarkers Definitions Working Group. Biomarkers and surrogateendpoints: preferred definitions and conceptual framework. ClinPharmacol Ther 69:89–95, 2001.

9. European Society of Radiology. White paper on imaging biomarkers.Insights Imaging 1:42–45, 2010.

10. Mackinnon A. A spreadsheet for the calculation of comprehensivestatistics for the assessment of diagnostic tests and inter-rater agree-ment. Comput Biol Med 30:127–234, 2000.

11. Altman DG. Practical statistics for medical research. London: Chap-man and Hall, 1991.

12. Maroldi R, Ravanelli M, Borghesi A, et al. Paranasal sinus imaging.European J Radiol 66:372–386, 2008.

13. Vogl T, Dresel S, Schedel H, et al. [MRI of the nasopharynx withGd-DTPA: its value and differential diagnostic criteria]. Rofo 150:516–522, 1989.

14. Schmidt GP, Kramer H, Reiser MF, et al. Whole-body magneticresonance imaging and positron emission tomography-computed

tomography in oncology. Top Magn Reson Imaging 18:193–202,2007.

15. Razek AA, Megahed AS, Denewer A, et al. Role of diffusion-weighted magnetic resonance imaging in differentiation between theviable and necrotic parts of head and neck tumors. Acta Radiol49:364–370, 2008.

16. Wang J, Takashima S, and Takayama F. Head and neck lesions:characterization with diffusion-weighted echo-planar MR imaging.Radiology 220:621–630, 2001.

17. Hermans R, and Vandecaveye V. Diffusion-weighted MRI in headand neck cancer. Cancer Imaging 7:126–127, 2007.

18. Vandecaveye V, De Keyzer F, Nuyts S, et al. Detection of head andneck squamous cell carcinoma with diffusion weighted MRI after(chemo)radiotherapy: correlation between radiologic and histo-pathologic findings. Int J Radiat Oncol Biol Phys 67:960–971, 2007.

19. Abdel Razek A, Mossad A, and Ghonim M. Role of diffusion-weighted MR imaging in assessing malignant versus benign skull-base lesions. Radiol Med 116:125–132, 2011.

20. Razek AA, Sieza S, and Maha B. Assessment of nasal and paranasalsinus masses by diffusion-weighted MRI. J Neuroradiol 36:206–211,2009.

21. Ginat DT, Mangla R, Yeaney G, et al. Diffusion weighted imaging fordifferentiating benign from malignant skull lesions and correlationwith cell density. AJR Am J Roentgenol 198:597–601, 2012.

22. Filippi M, Cercignani M, Inglese M, et al. Diffusion tensor mag-netic resonance imaging in multiple sclerosis. Neurology 56:304–311, 2001.

23. Bammer R, and Fazekas F. Diffusion imaging in multiple sclerosis.Neuroimaging Clin N Am 12:71–106, 2002.

24. Maeda M, and Maier S. Usefulness of diffusion-weighted imagingand the apparent diffusion coefficient in the assessment of head andneck tumors. J Neuroradiology 35:71–78, 2008.

25. Thoeny HC, and De Keyzer F. Extracranial applications of diffusion-weighted magnetic resonance imaging. Eur Radiol 17:1385–1393,2007.

26. Wang J, Takashima S, Takayama F, et al. Head and neck lesions:characterization with diffusion-weighted echo-planar MR imaging.Radiology 220:621–630, 2001.

27. Suzuki C, Maeda M, Hori K, et al. ADC of pituitary macroadenomaevaluated with line-scan diffusion-weighted imaging. J Neuroradiol4:228–235, 2007.

28. Matsushima N, Maeda M, Takamura M, et al. ADC of benign andmalignant salivary gland tumors. Comparison with histopathologicalfindings. J Neuroradiol 3:183–218, 2007.

29. Habermann CR, Arndt C, Graessner J, et al. Diffusion-weightedecho-planar MR imaging of primary parotid gland tumors: is aprediction of different histologic subtypes possible? AJNR Am JNeuroradiol 30:591–596, 2009.

30. Sumi M, Sakihama N, Sumi T, et al. Discrimination of metastaticcervical lymph nodes with diffusion-weighted MR imaging in pa-tients with head and neck cancer .AJNR Am J Neuroradiol 24:1627–1634, 2003.

31. Abdel Razek AA, Soliman NY, Elkhamary S, et al. Role of diffusion-weighted MR imaging in cervical lymphadenopathy. Eur Radiol16:1468–1477, 2006.

32. Razek AA, Sadek AG, Kombar OR, et al. Role of apparent diffu-sion coefficient values in differentiation between malignant andbenign solitary thyroid nodule. AJNR Am J Neuroradiol 29:563–568, 2008.

33. Abdel Razek AA, Gaballa G, Elhawarey G, et al. Characterization ofpediatric head and neck masses with diffusion-weighted MR imag-ing. Eur Radiol 19:201–208, 2009.

34. White ML, Zhang Y, and Robinson RA. Evaluating tumors andtumorlike lesions of the nasal cavity, the paranasal sinuses, and theadjacent skull base with diffusion-weighted MRI. J Comput AssistTomogr 30:490–495, 2006.

35. Anderson JR, and Tumors I. General features, types and examplesMuir’s textbook of pathology. London, England: Edward Arnold,127–156, 2008.

36. Srinivasan A, Dvorak R, Perni K, et al. Differentiation of benign andmalignant pathology in the head and neck using 3T apparent diffu-sion coefficient values: early experience. AJNR Am J Neuroradiol29:40–44, 2008.

American Journal of Rhinology & Allergy e185

37. Sakamoto J, Yoshino N, Okochi K, et al. Tissue characterization ofhead and neck lesions using diffusion-weighted MR imaging withSPLICE. Eur J Radiol 69:260–268, 2009.

38. Rijswijk CS, Kunz P, Hogendoorn PC, et al. Diffusion–weighted MRIin the characterization of soft tissue tumors. J Magn Reson Imag5:302–307, 2002.

39. Maeda M, Kato H, Sakuma H, et al. Usefulness of the apparent diffusioncoefficient in line scan diffusion-weighted imaging for distinguishingbetween squamous cell carcinomas and malignant lymphomas of thehead and neck. AJNR Am J Neuroradiol 26:1186–1192, 2005.

40. King AD, Ahuja AT, Yeung DK, et al. Malignant cervical lymphade-nopathy: diagnostic accuracy of diffusion-weighted MR imaging.Radiology 245:806–813, 2007.

41. Thoeny HC. Diffusion-weighted MRI in head and neck radiology:applications in oncology. Cancer Imaging 10:209–214, 2011.

42. Sahin C, and Arıbal E. The role of apparent diffusion coefficientvalues in the differential diagnosis of breast lesions in diffusion-weighted MRI. Diagn Interv Radiol 19:457–462, 2013.

43. Guo AC, Cummings TJ, Dash RC, et al. Lymphomas and high-gradeastrocytomas: comparison of water diffusibility and histologic char-acteristics. Radiology 224:177–183, 2002.

44. De Bondt RBJ, Hoeberigs MC, Nelemans PJ, et al. Diagnosticaccuracy and additional value of diffusion-weighted imaging fordiscrimination of malignant cervical lymph nodes in head andneck squamous cell carcinoma. Neuroradiology 51:183–192,2009. e

e186 September–October 2014, Vol. 28, No. 5