Elsevier

Clinical Radiology

Volume 73, Issue 4, April 2018, Pages 410.e9-410.e15
Clinical Radiology

Detection of clinically occult primary tumours in patients with cervical metastases of unknown primary tumours: comparison of three-dimensional THRIVE MRI, two-dimensional spin-echo MRI, and contrast-enhanced CT

https://doi.org/10.1016/j.crad.2017.10.020Get rights and content

Highlights

  • Preop localization of the primary tumour is critical in patients with neck LN metastases.

  • 3D THRIVE is more sensitive at detecting primary tumours than is SE T1-weighted MR.

  • Using the 3D THRIVE sequence may improve biopsy targeting and therapeutic planning.

Aim

To evaluate and compare the utility of contrast-enhanced three-dimensional (3D) T1-weighted high-resolution isotropic volume examination (THRIVE), spin-echo (SE) T1-weighted magnetic resonance imaging (MRI), and computed tomography (CT) for detecting clinically occult primary tumours in patients with cervical lymph node metastases.

Materials and methods

Seventy-three consecutive patients with tumours that went undetected during endoscopic or physical examinations underwent preoperative contrast-enhanced CT and MRI (SE and 3D THRIVE) after gadolinium injection. Guided biopsy results served as reference standards. The diagnostic performances of the imaging techniques were compared with McNemar's tests.

Results

Primary tumours were identified in 59 (80.8%) of the 73 patients after surgery. Of these, 36 were found in the palatine tonsil, 11 in the base of the tongue, seven in the nasopharynx, and five in the pyriform sinus. The sensitivity (72.9%) and accuracy (71.2%) of 3D THRIVE for detecting primary tumours were higher than were those of SE T1-weighted MRI (49.2% and 53.4%, p≤0.002) or CT (36.4% and 46.4%, p≤0.001). The specificities of these techniques did not differ. The diagnostic performance of 3D THRIVE (area under the curve [AUC]=0.681) for detecting tumours did not differ from that of SE T1-weighted MRI or CT (AUC=0.671 and 0.608, p>0.05).

Conclusion

3D THRIVE was more sensitive at detecting primary tumours than was SE T1-weighted MRI or CT in patients with cervical metastases of unknown primary tumours. This sequence may improve biopsy and therapeutic planning in these patients.

Introduction

When a primary tumour cannot be identified in a patient with biopsy-proven cervical lymph node metastasis despite thorough clinical examination, it is defined as cervical lymph node metastasis from an unknown primary tumour.1 Such patients should undergo endoscopic examinations of the upper aerodigestive tract during the initial clinical evaluation because accurate preoperative localisation of the primary tumour in head and neck squamous cell carcinoma is a prerequisite for selecting appropriate treatment plans and predicting patients' prognoses.1, 2, 3 In addition, preoperative cross-sectional imaging has become mandatory to ensure accurate localisation of the primary tumour.4, 5

Magnetic resonance imaging (MRI) has been widely used in pre- and postoperative evaluations of patients with head and neck cancer because it provides excellent tissue contrast between the tumour and the surrounding structures.6, 7, 8 It was previously reported that the T1-weighted high-resolution isotropic volume examination (THRIVE) sequence, a three-dimensional (3D) ultrafast spoiled gradient MRI sequence that incorporates a frequency-selective fat-saturation pulse, may serve as an efficient alternative to the two-dimensional (2D) spin-echo (SE) T1-weighted MRI sequence, as it provides more detailed anatomical information and good spatial resolution with reduced artefacts for preoperative head and neck cancer staging.9 Based on this information, the aim of the present study was to investigate and compare the clinical usefulness of the contrast-enhanced THRIVE sequence, contrast-enhanced SE T1-weighted sequence, and contrast-enhanced CT for the preoperative localisation of clinically occult primary tumours in patients with cervical lymph node metastasis.

Section snippets

Patients

This retrospective study was approved by the institutional review board, which waived the requirement for informed consent. Between January 2011 and December 2016, 86 consecutive patients who initially presented with palpable lumps in the neck that were confirmed to be metastatic squamous cell carcinoma by fine-needle aspiration or biopsy of the cervical lymph nodes, but in whom primary tumours were not detected by full physical and endoscopic evaluations of the upper aerodigestive tract, were

Results

The clinical characteristics of the 73 patients are listed in Table 1. Forty (54.8%) of the 73 patients showed single-level involvement of the metastatic lymph nodes, while 33 patients (45.2%) had multiple-level involvement. Except for the two patients who presented with metastatic lymph node involvement in levels I and III, respectively, all of the other patients showed level II involvement.

The final histopathological examinations that were performed after guided biopsy according to the CT and

Discussion

In the present study, the utility of contrast-enhanced 3D THRIVE, SE T1-weighted MRI, and CT to depict clinically occult primary tumours in 73 patients were compared against the biopsy results as the reference standard. The present data showed that the maximum detection rate of clinically occult primary tumours was 73%. The diagnostic performance of the contrast-enhanced 3D THRIVE sequence was superior to those of contrast-enhanced SE T1-weighted MRI and contrast-enhanced CT, as 3D THRIVE had

References (26)

  • M. Keberle et al.

    Current concepts in imaging of laryngeal and hypopharyngeal cancer

    Eur Radiol

    (2002)
  • R. Sigal et al.

    CT and MRI of squamous cell carcinoma of the tongue and floor of the mouth

    RadioGraphics

    (1996)
  • A.D. Singhi et al.

    Comparison of human papillomavirus in situ hybridization and p16 immunohistochemistry in the detection of human papillomavirus-associated head and neck cancer based on a prospective clinical experience

    Cancer

    (2010)
  • Cited by (0)

    View full text