Abstract
Objective:
In radiotherapy of nasopharyngeal carcinoma (NPC) patients, the brachial plexus (BP) situated at both sides of the neck is often irradiated to high dose. This study was to evaluate different BP delineation methods and analyse the dosimetric consequences when applying BP dose constraints in radiotherapy planning of NPC.
Methods:
15 NPC cases radically treated with helical tomotherapy were recruited. Apart from the original treatment plan (Plan A), two new plans (Plans B and C) with additional BP dose constraints were computed using the same planning CT images, structures and planning parameters. Plan B consisted of BP contours based on Radiation Therapy Oncology Group (RTOG)-endorsed atlas; while those in Plan C were based on MR images registered with the planning CT images.
Results:
The mean BP volume by RTOG method was 19.04 ± 3.50 cm3 vs 10.44 ± 2.00 cm3 by CT/MRI method. The mean BP overlapping volume between the two contouring methods was 1.9 cm3 (0.38–4.03 cm3). There was significant difference between two methods (p < 0.001). The average Dmax, Dmean, D5%, D10% and D15% of both sides of BP in Plan A were significantly higher than those in both Plan B and Plan C. There were no significant dose differences in the targets and organs at risk (OARs) after applying dose constraints in Plan B and Plan C.
Conclusion:
RTOG method was recommended since larger BP volume provided better protection. Applying BP dose constraints during tomotherapy plan optimisation for NPC patients could significantly reduce the BP dose (p < 0.05) without compromising the doses to the targets and other OARs.
Advances in knowledge:
This is the first study comparing the delineation method based on RTOG-endorsed atlas with the conventional CT/MRI delineation method for BP in tomotherapy of NPC patients. Our results showed that BP dose could be significantly reduced after applying the dose constraints without compromising the doses to the target volumes and other OARs. The RTOG method was more favoured as it gave a relatively larger BP volume and therefore offered better organ sparing.
Introduction
Radiotherapy of nasopharyngeal carcinoma (NPC) patients involves the delivery of high dose to the cervical lymphatics up to 60–66 Gy. The brachial plexus (BP), which is situated at both sides of the neck, is often irradiated to high dose. The BP originates from the cervical and upper thoracic spinal cord and is responsible for the cutaneous and muscular innervation of the chest, shoulder and upper extremities.1 High radiation dose to BP can lead to late toxicity such as radiation-induced brachial plexopathy (RIBP), which is a peripheral nerve disorder2 leading to symptoms like incapacitation of shoulder, arm and hand pain, numbness, paraesthesia, muscle weakness, atrophy and paralysis.1–4 The development of radiation-induced nerve injury is a slow process with a latency period of 1–4 years and often irreversible. The main mechanism is thought to be demyelination leading to axon loss.5 Treatment for RIBP is symptomatic and often ineffective.2 Radiation dose and irradiated volume of BP are the crucial factors for RIBP.5 Therefore, high radiation dose to BP should be avoided. The recommended dose limits of BP by the Radiation Therapy Oncology Group (RTOG; RTOG 0522, 0412, 0615, 0617) range from 60 to 66 Gy using conventional dose scheme.1–3,6
Accurate delineation of BP is important for organ sparing in radiotherapy treatment planning, especially for intensity-modulated radiotherapy where the dose gradient at the target–organ at risk (OAR) interface is steep; a slight spatial deviation may lead to overdose of the organ.7 Currently, there are several recommendations for the delineation of BP,2,7,8 including those based on the RTOG-endorsed BP atlas and CT/MRI image registration. The aims of this study were to evaluate these two methods of BP delineation and to analyse the dosimetric consequences when applying dose constraints to the structure in radiotherapy planning of NPC treated by helical tomotherapy.
Methods and materials
15 adult NPC patients treated with helical tomotherapy in 2012 and 2013 were randomly recruited. The patient characteristics are shown in Table 1. They received radical radiotherapy to the nasopharynx (NP) and the cervical lymph nodes (LNs). All tomotherapy plans were computed by the tomotherapy planning system (TomoHD 1.2.1; Accuray Inc., Sunnyvale, CA). The contours of all targets and OARs including BP in both CT and MR images were delineated using Eclipse External Beam Planning System (v. 10.0; Varian Medical System Inc., Palo Alto, CA). PACS Station (GE Healthcare, Giles, UK) was used to retrieve both CT and MR images. MIM software (v. 5.6.5; MIM Software Inc., Cleveland, OH) was used for image registration and dose evaluation. SPSS Statistics (v. 21; IBM, Armonk, NY) was used for statistical analysis.
Table 1.
The patient characteristics
| Variables | |
| Mean age (range) | 47.7 years (31–78) |
| Gender | |
| Male | 8 (53.3%) |
| Female | 7 (46.7%) |
| AJCC stage | |
| I | 4 (26.7%) |
| II | 3 (20%) |
| III | 3 (20%) |
| IV | 2 (13.3%) |
| Unknown | 3 (20%) |
AJCC, American Joint Committee on Cancer.
Patients were immobilised in supine position on T-plates with T-vacloks and attached to head rest. Patients’ head, neck and upper thoracic region were immobilised by thermoplastic mask. Tongue depressor was used. Planning CT images were acquired by GE LightSpeed CT simulator (GE Healthcare, Giles, UK) with 2.5 mm slice thickness and without i.v. iodine-based contrast. The scan region was from the vertex down to the mid-thorax at the level of fifth thoracic vertebra.
The MRI scans, either T1 or T2 pulse sequence, covered the patients’ whole neck regions down to the level of second thoracic vertebra with slice thickness of at least 3 mm. Patients’ MR images were fused with the planning CT images by rigid registration using the MIM software. If the CT and MR images were acquired in very different positions, two sets of rigid image registrations would be used: one focused mainly on the NP region and the other focused on the neck region, in which the latter was relevant for this study. Both planning CT images and registered MR images were imported to the Eclipse Planning System (Figure 1). Target volumes including NP and LNs, and OARs were contoured in the corresponding slices (Figure 2). Pseudostructures included “PTV Margin”, a 1 cm avoidance structure added to planning target volume (PTV) in all directions, and “Low Dose”, a 3 cm avoidance structure in addition to the “PTV Margin” in all directions. The purpose of these pseudostructures was to minimise dose spillage to the adjacent normal tissues.
Figure 1.
Planning CT/MR image fusion based on rigid registration using MIM software.
Figure 2.
NP targets (left) Including NP-GTV and NP-PTV and LN targets (right) Including LN-GTV and LN-PTV contoured by clinical oncologist on planning CT images. GTV, gross tumour volume; NP, nasopharynx; PTV, planning target volume; LN, lymph node.
The left and right BP (BP-Lt and BP-Rt) were delineated on the CT images according to the RTOG-endorsed BP contouring atlas (RTOG method; Table 2). Then another set of contours (BP-Lt-MR and BP-Rt-MR) was delineated using the registered MR images (CT/MRI method). Illustrations of BP delineation by both methods are shown in Figure 3a,b. When there was overlapping between BP and PTV, an additional “nPTV BP” structure that excluded the overlapping part would be created for plan optimisation.
Table 2.
A summary of the RTOG-endorsed BP contouring atlas7
| Step | Descriptions |
| 1 | Identify and contour C5, T1 and T2 |
| 2 | Identify and contour the subclavian and axillary neurovascular bundle |
| 3 | Identify and contour anterior and middle scalene muscles from C5 to insertion onto the first rib |
| 4 | To contour the BP OAR using a 5 mm diameter paint tool |
| 5 | Start at the neural foramina from C5 to T1; this should extend from lateral aspect of the spinal canal to the small space between the anterior and middle scalene muscles |
| 6 | For CT slices, where no neural foramen is present, contour only the space between the anterior and middle scalene muscles |
| 7 | Continue to contour the space between the anterior and middle scalene muscles; eventually the middle scalene will end in the region of the subclavian neurovascular bundle |
| 8 | Contour the BP as the posterior aspect of the neurovascular bundle inferiorly and laterally to one to two CT slices below the clavicular head |
| 9 | The first and second rib serve as the medial limit of the OAR contour |
BP, brachial plexus; OAR, organ at risk; RTOG, Radiation Therapy Oncology Group.
Figure 3.
(a) Comparison of BP contours by RTOG-endorsed contouring atlas (RTOG method; left) and by registered MRI (CT/MRI method; right) from spine level of C5 down to T2. (b) Overlapping of BP contours by RTOG method (dark blue) and CT/MRI method (dark green) on planning CT. BP, brachial plexus; RTOG, Radiation Therapy Oncology Group.
After adding the BP contours in the original plan (Plan A), the dose–volume histogram of BP was generated using the MIM software, which gave the dose information when no dose constraint was applied to BP. Plan A1 was the original plan with BP delineated by RTOG method, whereas Plan A2 was the original plan with BP delineated by CT/MRI method. New Plan B and Plan C were then computed with BP delineated using the RTOG method and CT/MRI method respectively, and with application of dose constraints to BP. A summary of BP dose constraints is shown in Table 3. All the other plan settings (Table 4) and OAR dose constraints (Table 5) were the same as Plan A. The end point was that Plan B and Plan C achieved the same dose outcome for the targets and OARs as in Plan A.
Table 3.
Additional dose constraints for BP in IMRT optimisation of NPC
| Dose (Gy) | |
| Dmax | 60 |
| D5% | 58 |
| D10% | 56 |
| D15% | 54 |
BP, brachial plexus; IMRT, intensity-modulated radiotherapy; NPC, nasopharyngeal carcinoma.
Table 4.
The routine treatment settings for radiotherapy of NPC using helical tomotherapy
| Parameters | Settings |
| Field width | 2.5 cm |
| Dose calculation grid | Fine |
| Pitch | 0.287 |
| Modulation factor | 3.0–4.0 |
NPC, nasopharyngeal carcinoma.
Table 5.
Dose constraints for OARs for IMRT optimisation of NPC
| OARs | Dmax (Gy) | D10% (Gy) | Dmean (Gy) |
| Brain stem | <54 | – | – |
| Spinal cord | <45 | – | – |
| Optic chiasm | <50 | – | – |
| Pituitary | – | – | <50 |
| Lt optic nerve | <50 | – | – |
| Rt optic nerve | <50 | – | – |
| Lt parotid gland | – | – | <28 |
| Rt parotid gland | – | – | <28 |
| Lt lens | <10 | – | <6 |
| Rt lens | <10 | – | <6 |
| Lt eye | <40 | – | <25 |
| Rt eye | <40 | – | <25 |
| Larynx | <63 | – | <40 |
| Oral cavity | <66 | <60 | <45 |
| Lt temporal lobe | <66 | <45 | – |
| Rt temporal lobe | <66 | <45 | – |
| Lt middle ear | – | – | <55 |
| Rt middle ear | – | – | <55 |
| Mandible | – | – | <45 |
IMRT, intensity-modulated radiotherapy; Lt, left; OAR, organ at risk; NPC, nasopharyngeal carcinoma; Rt, right.
Apart from checking the BP dose against the RTOG-recommended dose limit, the BP volumes, maximum dose, mean dose and dose received by 5, 10 and 15% volume of BP were recorded and compared among the three plans. To analyse the volumetric difference between two BP contouring methods, paired t-test was performed using the SPSS Statistics. To analyse the dosimetric differences of targets and OARs among different plans, one-way analysis of variance for repeated measure with the post-hoc multiple comparison test of Tukey’s honestly significant difference was performed.
Results
BP contour
The mean volume of BP contoured by the RTOG and CT/MRI methods was 19.04 ± 3.50 and 10.44 ± 2.00 cm3 respectively. There was significant difference between the two methods of BP delineation (p < 0.001). The overlapping BP volume of two delineation methods ranged from 0.38 to 4.03 cm3.
BP dose
With the application of dose constraints to BP in the optimization process, its dose was in general significantly lower than the plan without dose constraint. For the BP delineation based on RTOG method, there was significant reduction of BP dose in Plan B over Plan A1 in Dmax, D5%, D10% and D15% (p < 0.05), with the range of 1.73–3.55 Gy. For the BP delineation based on CT/MRI method, there was significant difference of BP doses between Plan A2 and Plan C in terms of Dmax and D10% (p < 0.05). The range of dose reduction was 1.77–2.47 Gy. Details of the dose outcome are shown in Table 6.
Table 6.
Comparison of BP dose between plans with and without applying dose constraint in planning optimisation
| RTOG method | CT/MRI method | |||||
| Plan A1 | Plan B | Plan A2 | Plan C | |||
| (Mean ± SD) | (Mean ± SD) | p-value | (Mean ± SD) | (Mean ± SD) | p-value | |
| Dmax (Gy) | ||||||
| Left | 61.96 ± 3.70 | 59.95 ± 3.56 | 0.002 | 60.79 ± 9.62 | 58.59 ± 9.37 | 0.005 |
| Right | 61.63 ± 3.37 | 59.89 ± 3.34 | 0.003 | 60.47 ± 8.52 | 58.35 ± 8.22 | 0.016 |
| Dmean (Gy) | ||||||
| Left | 43.51 ± 11.29 | 40.91 ± 9.82 | 0.058 | 40.49 ± 11.16 | 38.29 ± 10.01 | 0.008 |
| Right | 42.17 ± 11.13 | 40.27 ± 10.03 | 0.097 | 38.13 ± 10.28 | 37.28 ± 9.84 | 0.199 |
| D5% (Gy) | ||||||
| Left | 57.21 ± 6.72 | 54.70 ± 1.50 | 0.012 | 54.48 ± 10.72 | 52.16 ± 10.32 | 0.010 |
| Right | 56.29 ± 7.53 | 54.37 ± 6.71 | 0.037 | 54.03 ± 10.11 | 52.60 ± 10.31 | 0.142 |
| D10% (Gy) | ||||||
| Left | 55.17 ± 8.66 | 51.97 ± 7.87 | 0.005 | 52.16 ± 11.33 | 50.01 ± 10.95 | 0.010 |
| Right | 54.27 ± 9.53 | 51.57 ± 8.66 | 0.010 | 51.31 ± 10.80 | 49.53 ± 10.69 | 0.043 |
| D15% (Gy) | ||||||
| Left | 53.65 ± 9.84 | 50.01 ± 8.88 | 0.004 | 50.47 ± 11.69 | 47.99 ± 11.11 | 0.010 |
| Right | 52.66 ± 10.68 | 49.66 ± 9.71 | 0.005 | 49.14 ± 11.22 | 47.43 ± 11.20 | 0.057 |
BP, brachial plexus; RTOG, Radiation Therapy Oncology Group.
Bold p-values indicate statistical significance.
Target and OAR doses
There was no significant difference in the dose coverage of NP targets (NP-GTV and NP-PTV) and nodal targets (LN-GTV Lt/Rt and LN-PTV Lt/Rt) among Plans A, B and C (p > 0.05; Table 7). No observable difference was found in their dose distributions (Figure 4). For the OARs, only those situated below the base of skull were included in the analysis as the more distant structures were unlikely to be influenced by the change of BP dose. Overall, there was no significant difference in all the OAR doses among Plans A, B and C (p > 0.05; Table 8).
Table 7.
Comparison of the target dose (D95%) among Plans A, B and C
| Plan A (mean ± SD) |
Plan B (mean ± SD) |
Plan C (mean ± SD) |
ANOVA p-value | HSD significance (p < 0.01) |
||
| NP-GTV | D95% | 72.67 ± 2.22 | 72.81 ± 2.29 | 72.79 ± 2.27 | 0.984 | – |
| NP-PTV | D95% | 65.59 ± 3.08 | 65.39 ± 3.17 | 65.39 ± 3.18 | 0.979 | – |
| LN-GTV Lt | D95% | 68.57 ± 3.84 | 68.44 ± 3.85 | 68.33 ± 3.80 | 0.985 | – |
| LN-PTV Lt | D95% | 61.82 ± 2.83 | 61.68 ± 3.14 | 61.65 ± 3.12 | 0.987 | – |
| LN-GTV Rt | D95% | 67.23 ± 4.27 | 67.09 ± 4.11 | 67.05 ± 4.12 | 0.993 | – |
| LN-PTV Rt | D95% | 61.82 ± 2.73 | 61.67 ± 2.94 | 61.63 ± 2.91 | 0.982 | – |
ANOVA, analysis of variance; GTV, gross tumour volume; HSD, honestly significant difference; NP, nasopharynx; PTV, planning target volume; LN, lymph node; Lt, left; Rt, right.
Figure 4.
Isodose distribution showing BP and LN targets of Plans A, B and C of one of the subjects (left to right). BP, brachial plexus; LN, lymph node.
Table 8.
Comparison of the OAR doses among Plans A, B and C
| Plan A (mean ± SD) |
Plan B (mean ± SD) |
Plan C (mean ± SD) |
ANOVA p-value |
HSD significance (p < 0.01) |
||
| Spinal cord | Dmax | 31.07 ± 5.39 | 29.64 ± 5.15 | 29.69 ± 4.74 | 0.688 | – |
| Lt parotid gland | Dmean | 25.65 ± 6.22 | 24.51 ± 5.95 | 24.12 ± 5.92 | 0.773 | – |
| Rt parotid gland | Dmean | 24.72 ± 5.53 | 23.84 ± 5.44 | 23.59 ± 5.27 | 0.836 | – |
| Larynx | Dmax | 59.96 ± 7.45 | 59.30 ± 7.74 | 59.37 ± 7.56 | 0.966 | – |
| Dmean | 27.77 ± 5.68 | 26.57 ± 5.79 | 26.77 ± 5.71 | 0.829 | – | |
| Oral cavity | Dmax | 68.18 ± 2.52 | 68.31 ± 2.39 | 68.35 ± 2.41 | 0.982 | – |
| Dmean | 38.97 ± 5.08 | 38.07 ± 5.63 | 38.38 ± 5.17 | 0.895 | – | |
| Lt middle ear | Dmean | 37.82 ± 9.66 | 36.67 ± 9.73 | 37.20 ± 9.75 | 0.949 | – |
| Rt middle ear | Dmean | 36.66 ± 9.89 | 35.45 ± 10.02 | 35.81 ± 9.86 | 0.943 | – |
| Mandible | Dmean | 37.37 ± 4.95 | 36.87 ± 4.83 | 36.66 ± 5.13 | 0.922 | – |
ANOVA, analysis of variance; HSD, honestly significant difference; Lt, left; OAR, organ at risk; Rt, right.
Discussion
Comparison between RTOG and CT/MRI methods
When delineating BP by the RTOG method, the mean total volume was 19.04 ± 3.50 cm3, which was close to the magnitude 20.8 cm3 reported by Min et al.9 However, it was significantly smaller than the volume reported from the article that proposed the RTOG-endorsed delineation atlas.10 For the BP contours based on CT/MRI method, the mean total volume was 10.44 ± 2.00 cm3. There was a significant difference between these two contouring methods. The mean overlapping volume was 1.91 cm3 (range: 0.38–4.03 cm3). From the visual inspection of the contours in axial CT images, the CT/MRI method mainly delineated the nerve roots of BP, whereas the RTOG method included both the roots and trunks. It was expected that the accuracy of BP delineated by CT/MRI method was influenced by the choice of MR image sequences. The anatomy and morphology of proximal nerve segments of BP could be demonstrated in T1 weighted (T1W) sequences, while the T2 short tau inversion–recovery (STIR) sequence could demonstrate the peripheral BP nerve disorders.11,12 Tagliafico et al13 concluded that T1W sequences offered higher signal-to-noise ratio in quantitative comparison and higher visualisation score in qualitative comparison than T2 weighted (T2W) sequences for the imaging of BP.
MR image acquisition
BP is located in between the anterior and middle scalene muscle. However, BP has the same signal intensity as muscle, i.e. low to intermediate on both T1W and T2W MR images. MR neurography, which uses a fat-suppressed heavily T2W sequence such as T2W short tau inversion–recovery imaging, could enhance the signal of BP to become slightly hyper-intense.14 Hyper-intense nerve fibres area could be clearly distinguished from the hypointense muscle and saturated fat.12,15 The image acquisition plane could also affect the MR image quality of BP. BP was well demonstrated in coronal and sagittal images. The sagittal oblique plane had the advantages of depicting the course of BP in cross-section, which was the most reliable imaging plane for visualisation of the trunks, divisions, cords and branches of BP. The coronal and axial images were useful in demonstrating the long axis of BP. The axial plane could further give additional information about the nerve roots as they exit their foramina.16 In addition, high magnetic fields could offer better signal-to-noise ratio and excellent spatial and contrast enhancement resolutions, and thus produce better image quality of BP.15
BP delineation
The RTOG-endorsed BP contouring atlas is a systematic guideline for BP delineation. Training is still required before the implementation of the contouring method. The atlas makes use of several prominent anatomy landmarks to estimate the position of BP. However, some landmarks were not easy to identify. For anterior and middle scalene muscles, the image contrast needed to be adjusted carefully to have better visualisation. Min et al9 studied the learning process of this delineation method. There was a very significant learning curve with a mean of 70 min (range: 35–120 min) for the first case, while the mean duration varied between 13 and 20 min after delineating five cases.This implied that there might be potential difficulties for this BP delineation method. However, this method was still justified since it could significantly reduce the BP dose and subsequent complications.
Targets and OAR doses
There was no significant difference between the dose coverage in terms of D95% of NP and LN targets among the three different plans. In NPC, the primary target (NP) usually extended from the brain stem level down to the level of first cervical vertebra, which was at a considerable distance away from BP. Therefore, the change of dose at BP was unlikely to influence the target dose coverage and there was no significant difference in NP target doses after the application of dose constraints to BP. For the nodal target volume (LN), although it was closer to BP and even with overlaps in some patients, the dose differences in nodal targets after the application of BP dose constraints were still not significant since higher priority was given to the dose coverage of the target volumes during optimisation. In other words, the application of BP dose constraints in the tomotherapy plan optimisation did not show any significant effect on the target doses. All OARs showed no significant dose difference after the application of dose constraints to BP, regardless of their relative distance from BP. It was because the volume of BP was small, and the optimisation process of the tomotherapy planning system could use alternative beam angles to deliver the target dose and keep similar doses to the OARs.
BP dose
Without the application of dose constraints to BP, regardless of the delineation methods, the mean Dmax of BP exceeded the RTOG-recommended dose limit 60 Gy. This implied that it carried a risk of RIBP, which was one of the common radiation-induced peripheral nervous system complications.17 Our results demonstrated that the application of dose constraints was effective in reducing the maximum BP dose below 60 Gy. In addition, the D5%, D10% and D15% of the BP were also significantly lower than that of the plans without constraints.
For the delineation of BP based on RTOG method, additional dose constraints could reduce the average Dmax of the left and right BPs by 2.01 and 1.7 Gy respectively, which were below 60 Gy. Similar result was found in the CT/MRI method. Application of BP dose constraints could reduce the Dmax by 2.20 and 2.13 Gy in the left and right sides respectively, and both could be kept below 60 Gy.
Conclusion
Delineation of BP using RTOG method produced a larger BP volume compared with CT/MRI method. By application of dose constrains to BP regardless of the delineation methods, its dose could be significantly reduced without affecting the target dose coverage and OAR doses. Balance between BP delineation accuracy and organ protection was important. RTOG method was recommended at the moment since it had a step-by-step guideline for BP delineation to enhance consistency and reproducibility. However, specific guidelines for CT/MRI method should be developed to increase delineation accuracy. For future, the potential of CT/MRI method with certain planning OAR volume margin in replacing the RTOG method will be worth to be investigated.
Contributor Information
Chi-Him Li, Email: chihim.li@gmail.com.
Vincent WC Wu, Email: vincent.wu@polyu.edu.hk.
George Chiu, Email: george.chiu@hksh.com.
References
- 1.Chen AM, Hall WH, Li J, Beckett L, Farwell DG, Lau DH, et al. Brachial plexus-associated neuropathy after high-dose radiation therapy for head-and-neck cancer. Int J Radiat Oncol Biol Phys 2012; 84: 165–9. doi: 10.1016/j.ijrobp.2011.11.019 [DOI] [PubMed] [Google Scholar]
- 2.Truong MT, Romesser PB, Qureshi MM, Kovalchuk N, Orlina L, Willins J. Radiation dose to the brachial plexus in head-and-neck intensity-modulated radiation therapy and its relationship to tumor and nodal stage. Int J Radiat Oncol Biol Phys 2012; 84: 158–64. doi: 10.1016/j.ijrobp.2011.10.079 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hall WH, Guiou M, Lee NY, Dublin A, Narayan S, Vijayakumar S, et al. Development and validation of a standardized method for contouring the brachial plexus: preliminary dosimetric analysis among patients treated with IMRT for head-and-neck cancer. Int J Radiat Oncol Biol Phys 2008; 72: 1362–7. doi: 10.1016/j.ijrobp.2008.03.004 [DOI] [PubMed] [Google Scholar]
- 4.Wouter van Es H, Engelen AM, Witkamp TD, Ramos LM, Feldberg MA. Radiation-induced brachial plexopathy: MR imaging. Skeletal Radiol 1997; 26: 284–8. doi: 10.1007/s002560050236 [DOI] [PubMed] [Google Scholar]
- 5.Amini A, Yang J, Williamson R, McBurney ML, Erasmus J, Allen PK, et al. Dose constraints to prevent radiation-induced brachial plexopathy in patients treated for lung cancer. Int J Radiat Oncol Biol Phys 2012; 82: e391–e398. doi: 10.1016/j.ijrobp.2011.06.1961 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Truong MT, Nadgir RN, Hirsch AE, Subramaniam RM, Wang JW, Wu R, et al. Brachial plexus contouring with CT and MR imaging in radiation therapy planning for head and neck cancer. Radiographics 2010; 30: 1095–103. doi: 10.1148/rg.304095105 [DOI] [PubMed] [Google Scholar]
- 7.Yi SK, Hall WH, Mathai M, Dublin AB, Gupta V, Purdy JA, et al. Validating the RTOG-endorsed brachial plexus contouring atlas: an evaluation of reproducibility among patients treated by intensity-modulated radiotherapy for head-and-neck cancer. Int J Radiat Oncol Biol Phys 2012; 82: 1060–4. doi: 10.1016/j.ijrobp.2010.10.035 [DOI] [PubMed] [Google Scholar]
- 8.Kong FM, Ritter T, Quint DJ, Senan S, Gaspar LE, Komaki RU, et al. Consideration of dose limits for organs at risk of thoracic radiotherapy: atlas for lung, proximal bronchial tree, esophagus, spinal cord, ribs, and brachial plexus. Int J Radiat Oncol Biol Phys 2011; 81: 1442–57. doi: 10.1016/j.ijrobp.2010.07.1977 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Min M, Roos D, Keating E, Penniment M, Carruthers S, Zanchetta L, et al. External evaluation of the radiation therapy oncology group brachial plexus contouring protocol: several issues identified. J Med Imaging Radiat Oncol 2014; 58: 360–8. doi: 10.1111/1754-9485.12175 [DOI] [PubMed] [Google Scholar]
- 10.Guiou M, Hall W, Jennelle R, Mathai M, Purdy J, Chen A. Prospective evaluation of dosimetric variables associated with brachial plexopathy after radiation therapy for head and neck cancer. Int J Radiat Oncol Biol Phys 2008; 72: S385. doi: 10.1016/j.ijrobp.2008.06.1248 [DOI] [Google Scholar]
- 11.Torres C, Mailley K, Del Carpio O'Donovan R, O’donoval RDC. MRI of the brachial plexus: modified imaging technique leading to a better characterization of its anatomy and pathology. Neuroradiol J 2013; 26: 699–719. doi: 10.1177/197140091302600614 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mallouhi A, Marik W, Prayer D, Kainberger F, Bodner G, Kasprian G. 3T MR tomography of the brachial plexus: structural and microstructural evaluation. Eur J Radiol 2012; 81: 2231–45. doi: 10.1016/j.ejrad.2011.05.021 [DOI] [PubMed] [Google Scholar]
- 13.Tagliafico A, Succio G, Emanuele Neumaier C, Serafini G, Ghidara M, Calabrese M, et al. MR imaging of the brachial plexus: comparison between 1.5-T and 3-T MR imaging: preliminary experience. Skeletal Radiol 2011; 40: 717–24. doi: 10.1007/s00256-010-1050-x [DOI] [PubMed] [Google Scholar]
- 14.McGary JE, Grant WH, Teh BS, Paulino AC, Butler E. Dosimetric evaluation of the brachial plexus in the treatment of head and neck cancer. Int J Radiat Oncol Biol Phys 2007; 69: S464–S465. doi: 10.1016/j.ijrobp.2007.07.1650 [DOI] [Google Scholar]
- 15.van Es HW. MRI of the brachial plexus. Eur Radiol 2001; 11: 325–36. doi: 10.1007/s003300000644 [DOI] [PubMed] [Google Scholar]
- 16.van Es HW, Bollen TL, van Heesewijk HP. MRI of the brachial plexus: a pictorial review. Eur J Radiol 2010; 74: 391–402. doi: 10.1016/j.ejrad.2009.05.067 [DOI] [PubMed] [Google Scholar]
- 17.Mukherji SK, Castillo M, Wagle AG. The brachial plexus. Seminars in Ultrasound, CT and MRI 1996; 17: 519–38. doi: 10.1016/S0887-2171(96)90002-5 [DOI] [PubMed] [Google Scholar]