This Article Abstract Figures Only Full Text (PDF) Services 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 Vandecaveye, V Articles by Hermans, R Search for Related Content PubMed PubMed Citation Articles by Vandecaveye, V Articles by Hermans, R

Evaluation of the larynx for tumour recurrence by diffusion-weighted MRI after radiotherapy: initial experience in four cases

Departments of ¹Radiology ²Otorhinolaryngology, Head and Neck Surgery ³Pathology ⁵Radiation Oncology, University Hospitals Leuven, Leuven and ⁴Department of Experimental Radiobiology/LEO, Katholieke Universiteit Leuven, Belgium

Correspondence: Dr Robert Hermans, Department of Radiology, Herestraat 49, 3000 Leuven, Belgium.

	Abstract

Top Abstract Introduction Imaging technique Case presentations Discussion Conclusion References

 
Radiotherapy-induced changes in the soft tissues of the neck hamper the early detection of persistent or recurrent tumour by clinical examination and imaging procedures. Diffusion-weighted (DW) MRI is a non-invasive technique capable of probing tissue properties by measuring the movement of water. The purpose of the ongoing study is to examine the usefulness of DW-MRI for differentiation of persistent or recurrent tumour from post-radiotherapeutic sequelae or complications. Four patients, suspected of tumour recurrence after radiotherapy for laryngeal squamous cell carcinoma, were examined using a DW-MRI sequence on a clinical 1.5 T MR system prior to surgery. In two patients, the DW-MRI images showed an asymmetric hyperintense lesion on b1000 images with low apparent diffusion coefficient (ADC)-value, compatible with tumour on histopathology. All surrounding tissue presented high ADC values and absent signal on the b1000 images, histopathologically correlating to post-radiotherapeutic changes. The images of the third and fourth patient showed absent or minimal symmetric hyperintensity of the laryngeal soft tissues on the b1000 images and high ADC-values. In these cases, the histopathological diagnosis of radionecrosis was made and no tumour was found. In all four cases, differentiation of tumoral tissue from radiotherapy-induced tissue alterations was possible with DW-MRI.

	Introduction

Top Abstract Introduction Imaging technique Case presentations Discussion Conclusion References

 
The clinical and radiological detection of persistent or recurrent head and neck cancer is difficult in the early phase after radiotherapy (RT) [1]. Also, the differentiation between tumour recurrence and laryngeal necrosis can be challenging after radiotherapy.

Diffusion-weighted MRI (DW-MRI) is a non-invasive technique capable of probing the micro-environment of tissue by measuring water movement, and has not yet been reported for evaluation of head and neck lesions in a post-RT setting. Four patients are presented in whom DW-MRI was used to evaluate possible tumour recurrence in the larynx after RT for squamous cell cancer (SCC).

	Imaging technique

Top Abstract Introduction Imaging technique Case presentations Discussion Conclusion References

 
All examinations were approved by the local ethics committee.

The MRI study was performed on a 1.5 T SONATA scanner (Siemens, Erlangen, Germany). A T₂ weighted turbo spin-echo (TSE) sequence was performed in the transverse plane, with parameters: 48 slices, 4 mm slice thickness, 0.4 mm intersection gap, field of view (FOV) of 20 cmx25 cm, matrix of 291x512, repetition time (TR)/echo time (TE) = 3080 ms/106 ms, 2 averages, an echo train length of 9 and a resulting pixel resolution of 0.7 mmx0.5 mmx4.0 mm. The total acquisition time was 5 min 42 s. Then, a T₁ weighted TSE sequence in the transverse plane was acquired with the following parameters: 48 slices, 4 mm slice thickness, 0.4 mm intersection gap, FOV of 20 cmx25 cm, matrix of 250x512, TR/TE = 775 ms/8.3 ms, 3 averages, an echo train length of 19 and a resulting pixel resolution of 0.8 mmx0.5 mmx4.0 mm. The total acquisition time was 5 min 35 s. This sequence was performed before and after administration of 15 cm³ of gadolinium-BOPTA (Multihance; BRACCO, Milan, Italy). In one patient, a fat suppression pulse was added to the T₁ weighted TSE after contrast-administration, increasing the sequence time to 6 min 59 s. Additional coronal or sagittal T₁ sequences after contrast administration were used depending on tumour localization.

Diffusion-weighted echo planar images (EPI) were acquired with 48 slices in the transverse plane, bandwidth of 1502 Hz/pixel, 4 mm slice thickness, 0.4 mm intersection gap, FOV 20 cmx25 cm, matrix of 104x128, TR/TE = 7100 ms/84 ms, 3 averages and a resulting pixel resolution of 2.0 mmx2.0 mmx4.0 mm. The images were acquired using six different b-values (b = 0, 50, 100, 500, 750 and 1000 s mm^–2). The total acquisition time for this sequence was 5 min 48 s. All diffusion-sensitizing gradients were applied in three orthogonal directions and combined to create a 3-scan trace. An apparent diffusion coefficient (ADC) map was calculated automatically using the built-in manufacturer's software.

All sequences were acquired with identical geometry to allow correlation of the DW images with the TSE sequences.

Image analysis was performed on an off-line workstation using dedicated software (BioMAP; Novartis Pharma AG, Basel, Switzerland).

DW-MRI was analysed in a first step by visual inspection of DW images with b-value 1000 s mm^–2 (b1000 images) and ADC maps in correlation to the co-registered anatomical images. Afterwards, multiple regions of interest (ROIs) were placed by two authors in consensus (VV, 2 years of experience, and RH, 15 years of experience in head and neck radiology) over the larynx at the supraglottic, glottic and infraglottic levels and averaged, excluding the site of any suspect lesions. Separately, the suspect sites were delineated. For all ROIs the ADC values were calculated using all b-values.

More specifically, ADC-calculation was acquired from the native DW-MR images. The ROIs were drawn on the DW-MR images for b = 0 s mm^–2 and were then copied to correct positions on all other images (b = 50, 100, 500, 750 and 1000 s mm^–2) automatically by the software allowing for correct determination of signal intensity per b-value and thus allowing for accurate ADC-calculation.

Both the qualitative and quantitative analyses were correlated to the histological findings (localization of tumour on DW-MRI and histological specimen, and signal intensity versus histological tissue type).

Routine imaging procedures in three patients included a CT study of the head and neck during intravenous injection of a contrast agent, using a multidectector scanner (Siemens Sensation 16). Collimation was 16 mmx0.75 mm, feed/rotation 9.9 mm s^–1, tube voltage 120 kV and mAs_eff 250. The effective slice thickness was 1.5 mm, the reconstruction interval 0.75 mm; axial and coronal slices were reformatted with a thickness of 2 mm parallel to and perpendicular on the true vocal cords.

Whole-body FDG-PET studies were performed in two patients on a CTI Siemens ECAT 931 (Siemens, Knoxville, TN) with an in-plane spatial resolution of 8 mm and a transverse FOV of 10.1 cm for each bed position. The emission scan was initiated 60 min after the intravenous injection of 6.5 MBq kg^–1 FDG (to a maximum of 555 MBq). The raw imaging data were reconstructed in a 128x128 matrix with the use of an iterative reconstruction algorithm.

	Case presentations

Top Abstract Introduction Imaging technique Case presentations Discussion Conclusion References

 
Patient 1
3 months after completion of RT for a T1N0 SCC of the right true vocal cord (25 daily fractions of 2.2 Gy resulting in a total dose of 55 Gy), a 66-year old patient presented with progressive hoarseness.

CT showed a contrast-enhancing nodular lesion in the right true and false vocal cord, suggestive of tumour recurrence (Figure 1).

View larger version (92K): [in this window] [in a new window]   Figure 1. (a) Transverse CT image and (b) transverse contrast-enhanced T1 weighted MR image show an ulcerated lesion in the right false vocal cord (arrows), corresponding to (c) a hyperintense rim on the b1000 image (arrows) and (d) a hypointense rim on the apparent diffusion coefficient (ADC) map (arrows). (e) This rim corresponds histologically to hypercellular cancer tissue (T) in the lateral part of the false vocal cord; medially, a necrotic ulcer (NU) is present. (f) A more detailed view of the tumour shows multiple nests of densely packed squamous cell carcinoma (SCC) cells.

A direct laryngoscopy under anaesthesia showed a thickened right true vocal cord with adjacent granulation tissue. Biopsies were taken and histological examination confirmed the presence of tumour.

The patient underwent a total laryngectomy with unilateral neck dissection. Histopathology confirmed presence of SCC in the right vocal cord, characterized by densely packed cells with large cytoplasmatic content and irregular nuclei, multiple mitoses and intercellular bridging, anatomically corresponding to the suspect lesion on DW-MRI. The surrounding tissue showed post-radiotherapeutic changes, including variable amount of inflammation, fibrosis and limited necrosis.

Patient 2
6 months after completion of radiotherapy for a T2N0 SCC of the left true vocal cord (35 fractions of 2 Gy resulting in a total dose of 70 Gy), a 54-year-old patient showed on indirect laryngoscopy during routine follow-up an irregular appearance of the left vocal cord. CT showed an irregularly thickened, slightly enhancing left true vocal cord, compatible with tumour recurrence; the soft tissue infiltration extended into the false vocal cord, subglottic region, anterior and posterior commissure, and involvement of the cricoid cartilage was suspected (Figure 2). The PET scan showed moderate FDG-uptake at the level of the larynx; this finding was interpreted as tumour recurrence.

View larger version (77K): [in this window] [in a new window]   Figure 2. (a) Transverse CT image and (b) transverse contrast-enhanced T1 weighted MR image show thickening and increased contrast enhancement of the left vocal cord (arrows), corresponding to a hypermetabolic spot on (c) coronal FDG-PET image (arrow). The lesion is detected on (d) the b1000 DW-MR image (arrows) as a hyperintensity rim showing low intensity on (e) the apparent diffusion coefficient (ADC) map (arrows). (f) Histological section shows corresponding tumoral infiltration (arrows).

Direct laryngoscopy was performed at the time of the laryngectomy and showed the presence of a left sided laryngeal mass. No biopsies were taken and surgery was performed immediately. Histopathology confirmed SCC in the left vocal cord corresponding to the suspect lesion on DW-MRI (Figure 2). Typical neoplastic features were present, showing densely packed cells with large cytoplasmatic content and irregular nuclei, multiple mitoses and intercellular bridging. The surrounding tissue showed post-radiotherapeutic changes, including variable amounts of inflammation, fibrosis and limited necrosis.

Patient 3
A 45-year old patient presented with progressive pain, dysphagia and dyspnoea, 4 months after completion of RT for a T2N0 SCC of the right true vocal cord (35 fractions of 2 Gy resulting in a total dose of 70 Gy). Clinical examination showed diffuse laryngeal oedema. Progressive dyspnoea required placement of a tracheostomy. CT showed diffuse and pronounced thickening of the laryngeal soft tissues, with obliteration of the laryngeal lumen. No focal soft tissue mass was discerned. The presence of some small gas bubbles in the crico-arytenoid joints was interpreted as indicating laryngeal necrosis (Figure 3). However, PET showed a moderately hypermetabolic focus in the larynx; this was reported as being suspect for tumour recurrence. Similar to the CT findings, the TSE sequences on MRI (Figure 3) showed diffuse soft tissue thickening in the larynx, without a focal lesion. The b1000 images showed no asymmetric hyperintense signal and the ADC map showed diffuse hyperintensity with an ADC value (in mm² s^–1) of 1.84x10^–3. A direct laryngoscopy was not performed. Because of a worsening situation, the patient underwent total laryngectomy. Histological examination of the entire larynx showed severe radiotherapy-induced changes with necrosis and inflammation, including purulent infiltration; no neoplastic tissue was found.

View larger version (77K): [in this window] [in a new window]   Figure 3. (a) Transverse CT image and (b) transverse contrast-enhanced T1 weighted MR image show diffuse thickening and contrast-enhancement of the false vocal cords, without focal nodular mass. Small air bubble adjacent to the right arytenoid can be appreciated (a, arrow). (c) Coronal FDG-PET image shows tracer uptake at the laryngeal level. No asymmetric hyperintensity is revealed by (d) b1000 and (e) the ADC map shows diffuse hyperintensity of the soft tissues at the same level. Histological examination reveals necrosis, inflammation and purulent infiltration, without evidence for tumour recurrence. (f) Detailed histological image shows granulation tissue and inflammatory infiltrate.

The TSE-sequences on MRI showed diffuse laryngeal soft tissue thickening, more pronounced in the left true vocal cord, and a soft tissue defect in the posterior part of the right true vocal cord, suggesting laryngeal necrosis. However, based on conventional MRI findings a tumoral lesion could not be excluded on the left glottic level.

Histological examination on multiple laryngeal biopsies taken during panendoscopy suggested the presence of tumoral recurrence at the level of the true and false vocal cords. DW-MRI showed only slight symmetric hyperintensity on the b1000 images in the laryngeal soft tissues, no focal lesion with low ADC-value could be detected. The laryngeal soft tissues appeared hyperintense on the ADC map, with ADC value (in mm² s^–1) of 1.87x10^–3.

The patient underwent total laryngectomy. Histological examination of the entire larynx showed severe radiotherapy-induced changes with ulceration, necrosis and inflammation but no neoplastic tissue was found.

	Discussion

Top Abstract Introduction Imaging technique Case presentations Discussion Conclusion References

 
DW-MRI is a non-invasive technique able to depict the extent of random movement of water protons in biological tissues; the addition of two opposed magnetic field gradients makes the signal intensity dependent on the mobility of water molecules [2]. The amount of signal loss over the range of b-values correlates with the mobility of protons and is quantified by means of the ADC. The ADC value is mainly influenced by the size of the extracellular extravascular space (EES) containing free moving protons, with additive contributions from bulk water movement (e.g. intravascular flow) in the low b-value images and minor contributions from intracellular diffusion and transmembraneous transport [3]. Thus, any tissue architectural change causing structural barriers or influencing the proportion of the extracellular versus the intracellular compartment is expected to alter the ADC value.

The anatomic heterogeneity of the head and neck region with numerous air–soft tissue interfaces makes DW-MRI in this region prone to susceptibility artefacts, ghosting and image distortion. However, recent technical developments make EPI-based (including DW-MRI) imaging feasible in this anatomically highly demanding region. Application of parallel imaging decreases the echo-train length, which in turn reduces off-resonance and blurring artefacts [4], while application of high bandwidth [5], thin slices and meticulous shimming reduces image distortion and chemical shift artefacts. The acquisition of images with a large range of b-values allows a more accurate calculation of ADC and improves the image quality of ADC-maps by reducing movement artefacts and noise propagation.

Experimental and clinical data support the potential use of DW-MRI for in vivo characterization of tissue. Wang et al [5] showed a significantly smaller ADC for malignant lesions, including SCC, than for benign lesions in the head and neck. Furthermore, ADC measurements provide reliable information on remaining viable tumour tissue in the follow-up of human high-grade gliomas after RT [6]. For SCC in an animal model, Herneth et al show that DW-MRI differentiates viable from necrotic tumour tissue [7].

Recent progress in non-surgical treatment of head and neck cancer, by combining multifractioned high-dose radiotherapy (RT) with radiosensitizing measures [8] or chemotherapy [9], allows us to obtain tumour control even in advanced laryngeal SCC. The diagnostic accuracy of currently used clinical and imaging follow-up procedures may be compromised when organ preservation is attempted in such advanced disease. Treatment-induced tissue changes are anticipated to be more pronounced in such circumstances, menacing the early detection of persistent or recurrent tumour [10–12]. DW-MRI may allow differentiation between neoplastic tissue and post-radiotherapy inflammatory or necrotic tissue as the differences in tissue microstructure are expected to create differences in proton mobility.

In patients one and two, the appearance of recurrent tumoral tissue on DW-MRI is illustrated. The lesions were hyperintense on b1000 and hypointense on the ADC map, contrasting with the surrounding tissue. Histologically, the recurrent SCC showed densely grouped cells with large cytoplasmatic content and occasionally intercellular bridging. These tumoral characteristics are expected to restrict the movement of protons as the high cellular index and large cytoplasmatic content increase the cellular tissue fraction and reduce the EES [13].

The diffuse high ADC value and the absence of any focal restrictive signal on the b1000 images in the laryngeal soft tissues in the third and fourth patient correlated with diffuse laryngeal necrosis and absence of tumoral recurrence. The findings on DW-MRI were contradictory to the FDG-PET findings in the third patient. The increased laryngeal uptake of FDG in this patient was presumably caused by granulation tissue and metabolically active leukocytes [14]. Indeed, the results reported on the value of FDG-PET in the post-radiotherapy evaluation of laryngeal cancer are variable [11, 12, 15]. The specificity of this technique is diminished by inflammatory tissue alterations present early after radiotherapy.

In all four patients, the diffuse hyperintensity on the ADC maps in the normal soft tissues of the larynx and hypopharynx correlated with expected post-radiotherapeutic alterations, such as inflammation and interstitial oedema, promoting free movement of protons.

In all three patients examined with CT, a correct diagnosis could be made. This technique has a high accuracy for diagnosing recurrent laryngeal cancer after radiotherapy, but false positive and false negative results may occur [16]. Differentiation of tumour recurrence from therapy-induced laryngeal necrosis based on anatomical findings may be problematic, although some CT-findings allow the correct diagnosis to be made [17]. The accuracy of conventional MRI-techniques has shown to be similar to CT in the post-radiotherapeutic neck [11]. As illustrated in the fourth patient, the asymmetric soft tissue thickening, visible on conventional MRI, did not allow exclusion of the presence of tumour recurrence. However, DW-MRI showed no restrictive lesion in the laryngeal soft tissues, supporting the diagnosis of post-radiotherapeutic alterations or complications rather than tumour recurrence. This specific ability of DW-MRI to probe tissue microstructure is an interesting complement to the currently used imaging procedures in the evaluation of the post-radiotherapeutic neck.

	Conclusion

Top Abstract Introduction Imaging technique Case presentations Discussion Conclusion References

 
This report illustrates the ability of DW-MRI to characterize the tissue changes observed in the post-radiotherapeutic larynx. The use of DW-MRI may have a complementary role in the non-invasive evaluation of the larynx after RT. Early, and preferentially non-invasive, differentiation of tumour recurrence from a treatment-induced complication is desirable, as in a number of patients with the latter condition the laryngeal function may be saved by conservative measures. Further studies in a large patient population are ongoing to validate the reproducibility and diagnostic accuracy of the technique in the post-radiotherapeutic neck.

This work was partly financially supported by the research grant "Prof. em. A. L. Baert, Siemens Medical Solutions".

View larger version (85K): [in this window] [in a new window]   Figure 4. (a) Transverse non-fat saturated T1 weighted MR image before and (b) fat-saturated image after administration of contrast agent at the level of the vocal cords showing right-sided soft tissue ulceration (b, arrowhead) and tissue swelling on the left side (b, arrows), not allowing exclusion of tumour recurrence. Both (c) b1000 image and (d) apparent diffusion coefficient (ADC) map do not show a restrictive focal lesion, supporting the diagnosis of laryngeal necrosis without tumour recurrence. (e) Five and (f) 10 times magnified histopathological sections show necrotic tissue (N), and stromal tissue with reactive changes (S); multiple neutrophils and giant cells are visible, suggesting profound inflammatory reaction. No tumoral tissue was found.

	References

Top Abstract Introduction Imaging technique Case presentations Discussion Conclusion References

 

1. Parsons JT. The effect of radiation on normal tissues of the head and neck. In: Million RR, Cassisi NJ, editors. Management of head and neck cancer: a multidisciplinary approach. Philadelphia, PA: Lippincott, 1994:245–89
2. Stejskal EO, Tanner J. Spin diffusion measurements: spin-echoes in the presence of a time-dependent field gradient. J Chem Phys 1965;42:288–92.[CrossRef]
3. Le Bihan D, Breton E, Lallemand D, Aubin ML, Vignaud J, Laval-Jeantet M. Separation of diffusion and perfusion in intravoxel incoherent motion MR imaging. Radiology 1988;168:497–505.[Abstract/Free Full Text]
4. Taouli B, Martin AJ, Qayyum A, Merriman RB, Vigneron D, Yeh BM, et al. Parallel imaging and diffusion tensor imaging for diffusion-weighted MRI of the liver: preliminary experience in healthy volunteers. AJR Am J Roentgenol 2004;183:677–80.[Abstract/Free Full Text]
5. Wang J, Takashima S, Takayama F, Kawakami S, Saito A, Matsushita T, et al. Head and neck lesions: characterization with diffusion-weighted echo-planar MR imaging. Radiology 2001;220:621–30.[Abstract/Free Full Text]
6. Hein PA, Eskey CJ, Dunn JF, Hug EB. Diffusion-weighted imaging in the follow-up of treated high-grade gliomas: tumor recurrence versus radiation injury. AJNR Am J Neuroradiol 2004;25:201–9.[Abstract/Free Full Text]
7. Herneth AM, Guccione S, Bednarski M. Apparent diffusion coefficient: a quantitative parameter for in vivo tumor characterization. Eur J Radiol 2003;45:208–13.[CrossRef][Medline]
8. Kaanders JH, Pop LA, Marres HA, Bruaset I, van den Hoogen FJ, Merkx MA, et al. ARCON: experience in 215 patients with advanced head-and-neck cancer. Int J Radiat Oncol Biol Phys 2002;52:769–78.[CrossRef][Medline]
9. Browman GP, Hodson DI, Mackenzie RJ, Bestic N, Zuraw L, Cancer Care Ontario Practice Guideline Initiative Head and Neck Cancer Disease Site Group. Choosing a concomitant chemotherapy and radiotherapy regimen for squamous cell head and neck cancer: a systematic review of the published literature with subgroup analysis. Head Neck 2001;23:579–89.[CrossRef][Medline]
10. Viani L, Stell PM, Dalby JE. Recurrence after radiotherapy for glottic carcinoma. Cancer 1991;67:577–84.[CrossRef][Medline]
11. Briggs RJ, Gallimore AP, Phelps PD, Howard DJ. Laryngeal imaging by computerized tomography and magnetic resonance following radiation therapy: a need for caution. J Laryngol Otol 1993;107:565–8.[Medline]
12. Terhaard CH, Bongers V, van Rijk PP, Hordijk GJ. F-18-fluoro-deoxy-glucose positron-emission tomography scanning in detection of local recurrence after radiotherapy for laryngeal/pharyngeal cancer. Head Neck 2001;23:933–41.[CrossRef][Medline]
13. Lyng H, Haraldseth O, Rofstad EK. Measurement of cell density and necrotic fraction in human melanoma xenografts by diffusion-weighted magnetic resonance imaging. Magn Reson Med 2000;43:828–36.[CrossRef][Medline]
14. Kubota R, Yamada S, Kubota K, Ishiwata K, Tamahashi N, Ido T. Intratumoral distribution of fluorine-18-fluorodeoxyglucose in vivo: high accumulation in macrophages and granulation tissues studied by microautoradiography. J Nucl Med 1992;33:1972–80.[Abstract/Free Full Text]
15. McGuirt WF, Greven KM, Keyes JW Jr, Williams DW 3rd, Watson NE Jr, Geisinger KR, et al. Positron emission tomography in the evaluation of laryngeal carcinoma. Ann Otol Rhinol Laryngol 1995;104:274–8.[Medline]
16. Hermans R, Pameijer FA, Mancuso AA, Parsons JT, Mendenhall WM. Laryngeal or hypopharyngeal squamous cell carcinoma: can follow-up CT after definitive radiotherapy be used to detect local failure earlier than clinical examination alone? Radiology 2000;214:683–7.[Abstract/Free Full Text]
17. Hermans R, Pameijer FA, Mancuso AA, Parsons JT, Mendenhall WM. CT findings in chondroradionecrosis of the larynx. AJNR Am J Neuroradiol 1998;19:711–8.[Abstract]

This article has been cited by other articles:

				A. Srinivasan, R. Dvorak, K. Perni, S. Rohrer, and S.K. Mukherji Differentiation of Benign and Malignant Pathology in the Head and Neck Using 3T Apparent Diffusion Coefficient Values: Early Experience AJNR Am. J. Neuroradiol., January 1, 2008; 29(1): 40 - 44. [Abstract] [Full Text] [PDF]

				D.-M. Koh and D. J. Collins Diffusion-Weighted MRI in the Body: Applications and Challenges in Oncology Am. J. Roentgenol., June 1, 2007; 188(6): 1622 - 1635. [Abstract] [Full Text] [PDF]

				BJR review of the year - 2006 Br. J. Radiol., March 1, 2007; 80(951): 147 - 151. [Full Text] [PDF]

				D-M Koh and A R Padhani Diffusion-weighted MRI: a new functional clinical technique for tumour imaging Br. J. Radiol., August 1, 2006; 79(944): 633 - 635. [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Services 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 Vandecaveye, V Articles by Hermans, R Search for Related Content PubMed PubMed Citation Articles by Vandecaveye, V Articles by Hermans, R