Elsevier

Clinical Radiology

Volume 65, Issue 12, December 2010, Pages 989-996
Clinical Radiology

Original Paper
Comparative study of whole-body MRI and bone scintigraphy for the detection of bone metastases

https://doi.org/10.1016/j.crad.2010.07.002Get rights and content

Aim

To assess and compare the diagnostic accuracy of whole-body magnetic resonance imaging (MRI) and bone scintigraphy in the detection of metastases to bone.

Material and methods

Forty randomly selected patients with known malignant tumours were prospectively studied using bone scintigraphy and whole-body MRI. Two patients were excluded. Symptoms of bone metastasis were present in 29 (76%) patients and absent in nine (24%). Findings were classified into four categories according to the probability of bone metastasis: (1) negative, (2) probably negative, (3) probably positive, and (4) positive. Diagnostic accuracy was determined according to the area under the receiver operating characteristic (ROC) curve. The definitive diagnosis was reached using other imaging techniques, biopsy, or 12 months clinical follow-up.

Results

Metastases were present in 18 patients. The sensitivity, specificity, and diagnostic accuracy were 94, 90, and 92%, respectively, for whole-body MRI and 72, 75, and 74%, respectively, for bone scintigraphy. Diagnostic accuracy measured by the area under the ROC curve was significantly higher for whole-body MRI (96%) than for bone scintigraphy (77%; p<0.05). Interobserver agreement measured by the kappa index was significantly higher for whole-body MRI (0.895) than for bone scintigraphy (0.524; p<0.05). Whole-body MRI detected lesions in tissues other than bone in 17 (45%) patients.

Conclusions

Whole-body MRI is more accurate and more objective than bone scintigraphy for the detection of bone metastases. Whole-body MRI can also detect lesions in tissues other than bone.

Introduction

Malignant tumours are currently the third most common cause of death worldwide.1 Mortality due to malignant tumours and the probability of successful treatment depend primarily on the characteristics of the tumour, but they also depend on whether metastases are present. Knowing the stage of the patient’s disease is important to enable prognosis and individualized treatment.2, 3 Metastases very often affect the skeleton. The percentage of tumours with a tendency to metastasize to bone is 32%.4 The tumours that most often metastasize to bone are lung, breast, prostate, kidney, and thyroid carcinomas.4 About 75% of patients with these tumours will develop at least one bone metastasis.5

Plain-film radiography is the initial imaging study in the diagnostic work-up of bone metastases and the most frequently employed, together with bone scintigraphy. However, radiography can only show significant bone destruction with a mineral loss of greater than 50%;6 thus, there is a long latency period between the onset of bone destruction and its first signs on radiography. For this reason, other techniques are necessary for earlier diagnosis.

CT (computed tomography) is more sensitive than plain-film radiography; this technique is often used to study the extent of bone destruction,7 as well as to determine extension to soft tissues and to evaluate possible complications.8

Whole-body magnetic resonance imaging (MRI) is a method that does not use ionizing radiation. The protocol of combined T1-weighted and short-tau inversion recovery (STIR) sequences has proven very sensitive in discriminating benign from malignant bone marrow processes.9, 10 Recently the use of diffusion-weighted imaging (DWI) has been proposed.11, 12

Many centres consider bone scintigraphy the best imaging test for the early detection and staging of bone metastases. However, bone scintigraphy is prone to false positives from benign lesions that have increased bone formation7 and to false-negatives from metastases that do not have a sufficiently strong osteoblastic reaction.13 Nevertheless, bone scintigraphy is relatively inexpensive and fast, so it is still considered the reference standard for early detection and evaluation of the extent of bone metastases.2, 14, 15

Several studies have compared whole-body MRI with bone scintigraphy for screening for bone metastases.7, 13, 16, 17, 18, 19, 20 However, these few series were small and used a wide variety of protocols. Moreover, whole-body MRI is more expensive and less widely available than bone scintigraphy; thus, whole-body MRI has yet to be incorporated into diagnostic algorithms for screening for bone metastases.

The aim of the present study was to compare the diagnostic accuracy and reproducibility of whole-body MRI and bone scintigraphy in the diagnosis of bone metastases.

Section snippets

Patients

From November 2006 to May 2007, a prospective, cross-sectional study of 40 randomly selected patients diagnosed with primary malignant tumours was undertaken. All patients underwent bone scintigraphy and whole-body MRI to screen for bone metastases. Inclusion criteria were patients older than 18 years of age with (1) a histologically confirmed primary malignant solid tumour that required screening for bone metastases (breast and lung cancers), or (2) other malignant solid tumours and clinical

Results

Bone metastases were present in 18 patients and absent in 20. The results of the binary classification of patients by bone scintigraphy were: sensitivity 72.2% (13/18), specificity 75% (15/20), and diagnostic accuracy 73.7% (28/38). The four-point scale classification of patients by bone scintigraphy yielded a diagnostic accuracy (AUC) of 76.9% (Fig. 1).

The results of the binary classification of patients by whole-body MRI were: sensitivity 94.4% (17/18), specificity 90% (18/20), and diagnostic

Discussion

In the present study, the diagnostic accuracy for the detection of bone metastases was higher for whole-body MRI than for bone scintigraphy. These results are in line with those of other studies,13, 16, 18, 19, 20 although it is difficult to compare the results of the different studies due to differences in the selection criteria, techniques, and approach to data analysis used.

The high number of false-negative results for bone scintigraphy in the present study is probably due to this

References (25)

  • G.P. Schmidt et al.

    Whole-body imaging of the musculoskeletal system: the value of MR imaging

    Skeletal Radiol

    (2007)
  • G. Layer et al.

    Magnetic resonance imaging to detect bone marrow metastases in the initial staging of small cell lung carcinoma and breast carcinoma

    Cancer

    (1999)
  • Cited by (31)

    • Bone Metastases

      2022, Oncologic Imaging: A Multidisciplinary Approach
    • Bone Metastases: Mechanisms of the Metastatic Process, Imaging and Therapy

      2021, Seminars in Ultrasound, CT and MRI
      Citation Excerpt :

      As with any cut-off values in medicine, there is significant overlap in the values and important pitfalls that are beyond the scope of the current manuscript. The above sequences can be used in different combinations54-62 to provide whole-body MRI coverage, which has been shown to detect more bone metastases than skeletal scintigraphy.54,57,58,60,63-67 Technical advances, such as a rolling table platform56,60,61 and parallel imaging techniques, based on phased-array coil technology59,68,69 have made whole body MRI feasible in the typical 1-hour patient scheduling allotment.48

    • Diagnostic accuracy of imaging methods for the diagnosis of skeletal malignancies: A retrospective analysis against a pathology-proven reference

      2016, European Journal of Radiology
      Citation Excerpt :

      Compared with other imaging techniques, radiography was relatively insensitive in detecting bone metastases. These observations are in line with studies indicating that bone lesions become apparent on radiographs only after the loss of more than 50% of the bone mineral content [2]. We are aware that the low number of nuclear medicine examinations in general, as well as radiological studies in subgroup analyses, may have impaired the statistical power of the statistical analyses.

    • Comparison of whole-body MRI and skeletal scintigraphy for detection of bone metastatic tumors: A meta-analysis

      2013, Surgical Oncology
      Citation Excerpt :

      Representative spectrum (item 1) was not present in 57.1% [9–11,16] of the 7 articles. The execution of the reference standard in detail (item 9) was not present in 14.3% [18] of the 7 articles. Fig. 1(A–B) shows the forest plot of sensitivity and specificity for whole-body MRI and skeletal scintigraphy in detecting bone metastases.

    View all citing articles on Scopus
    View full text