Eur Radiol DOI 10.1007/s00330-015-4047-y
MUSCULOSKELETAL
Accuracy and role of contrast-enhanced CT in diagnosis and surgical planning in 88 soft tissue tumours of extremities Lucia Verga 1 & Elena Maria Brach del Prever 2,1 & Alessandra Linari 3 & Sara Robiati 1 & Armanda De Marchi 1 & Domenico Martorano 1 & Michele Boffano 4 & Raimondo Piana 4 & Carlo Faletti 1
Received: 24 March 2015 / Revised: 8 September 2015 / Accepted: 23 September 2015 # European Society of Radiology 2015
Abstract Objectives Soft tissue tumours (STT) require accurate diagnosis in order to identify potential malignancies. Preoperative planning is fundamental to avoid inadequate treatments. The role of contrast-enhanced computed tomography (CT) for local staging remains incompletely assessed. Aims of the study were to evaluate CT accuracy in discriminating active from aggressive tumours compared to histology and evaluate the role of CT angiography (CTA) in surgical planning. Materials and methods This retrospective cohort series of 88 cases from 1200 patients (7 %) was locally studied by contrast-enhanced CT and CTA in a referral centre: 74 malignant tumours, 14 benign lesions. Contrast-enhancement patterns and relationship of the mass with major vessels and bone were compared with histology on surgically excised samples. Sensitivity, specificity, positive and negative predictive values (PPV, NPV) were evaluated in discriminating active from aggressive tumours.
Results Sensitivity in differentiating aggressive tumours from active lesions was 89 %, specificity 84 %, PPV 90 %, NPV 82 %. The relationship between mass and major vessels/bone was fundamental for surgical strategy respectively in 40 % and in 58 % of malignant tumours. Conclusion Contrast-enhanced CT and CTA are effective in differentiating aggressive masses from active lesions in soft tissue and in depicting the relationship between tumour and adjacent bones and major vessels. Key Points • Accurate delineation of vascular and bony involvement preoperatively is fundamental for a correct resection. • CT plays a critical role in differential diagnosis of soft tissue masses. • Contrast-enhanced CT and CT angiography are helpful in depicting tumoral vascular involvement. • CT is optimal for characterization of bone involvement in soft tissue malignancies.
* Lucia Verga [email protected]
Raimondo Piana [email protected]
Elena Maria Brach del Prever [email protected]
Carlo Faletti [email protected]
Alessandra Linari [email protected]
1
Department of Imaging, Azienda Ospedaliera Città della Salute e della Scienza, CTO Hospital, Via Zuretti 29, 10126 Torino, Italy
Sara Robiati [email protected]
2
Department of Public Health and Paediatrics, University of Turin, Via Santena 5-bis, 10126 Torino, Italy
Armanda De Marchi [email protected]
3
Department of Pathology, Azienda Ospedaliera Città della Salute e della Scienza, Regina Margherita Hospital, Piazza Polonia, 10126 Torino, Italy
4
Department of Orthopaedic Oncology and Reconstructive Surgery, Azienda Ospedaliera Città della Salute e della Scienza, CTO Hospital, Via Zuretti 29, 10126 Torino, Italy
Domenico Martorano [email protected] Michele Boffano [email protected]
Eur Radiol
Keywords Contrast-enhanced CT . CT angiography . Soft tissue sarcoma . Surgical planning . Tumour vascularization
Introduction Soft tissue sarcomas (STS) are a rare heterogeneous group representing 1 % of all tumours, often characterised by a large, and for a long time asymptomatic, mass in the limbs, where benign tumours and benign inflammatory lesions are very frequent [1, 2]. Depending on histotype and grading, STS may have a prolonged locally active behaviour with slow development (e.g. liposarcoma-lipomalike and some low grade liposarcomas), or they may have an aggressive development with rapid local invasion and, eventually, metastases at distance (e.g. high-grade pleiomorphic sarcoma) [1, 2]. The first-choice treatment of STS is surgical en bloc resection, to achieve local tumour control. The responsiveness to adjuvant therapies (chemo- and radiotherapy) is limited and low [3–6]. Surgical margins can be wide (the en bloc excision is intracompartmental with a cuff of normal tissue) or radical (the excision is extracompartmental); if the excision is marginal or contaminated, the resection is inadequate and it negatively influences the disease-free survival [1–6]. In the extremities and girdles, limb salvage versus amputation should be evaluated on the basis of the possibility of obtaining adequate surgical margins with acceptable residual function. For limb salvage, when the tumour involves the major vessel, two main surgical solutions are possible: 1) the wide en bloc tumour excision including the partially or totally encased vessel (and the vascular reconstruction with autologous transplant or vascular prosthesis has to be performed); 2) the wide en bloc excision including only the vessel adventitia (adventitia peeling), considered a sufficiently adequate margin, when the vessel is strictly adjacent to the tumour. The choice of the treatment strategy is often difficult, mainly based on imaging. The main challenge is to use a sensitive and specific imaging technique, able to discriminate a potentially malignant from a benign tumour and to depict the relationship of the tumour with major vessels and bone. In local staging, ultrasound (US) is useful to discriminate cysts from solid masses, to measure the diameter and to identify the location of a mass in relation to the deep fascia: a mass larger than 5 cm and deep with respect to the deep fascia should be considered potentially malignant [3–6]. Magnetic resonance imaging (MRI) is the gold standard to discriminate between potentially malignant aggressive masses and benign lesions [7–14] and to assess the compartment and the size for surgical planning [3, 5, 6]. Contrast-enhanced computed tomography (CT) is used for patients who, for various reasons, cannot be submitted to MRI [10] or when local logistic supplies are limited, such as in new
developing countries with low financial support [15, 16]. CT is accurate in identifying calcifications and assessing the degree of bone involvement [11, 15], but its accuracy in discriminating benign from malignant tumours in local staging is still undetermined. CT angiography (CTA) is well documented in showing the relationship of tumour with major vessels [16, 17]. Aims of this study were: 1) to evaluate the accuracy of contrast-enhanced CT in discriminating active from aggressive tumours in comparison to histology of surgically excised samples; 2) to evaluate the accuracy of contrast-enhanced CT and CTA in comparison to histology as an aid in deciding preoperatively the surgical margin and technique with reference to major vessels and bone. In this study, contrast-enhanced CT was part of a diagnostic strategy conducted by the interdisciplinary team in a regional referral centre for the study and treatment of bone and soft tissue sarcomas.
Materials and methods A total of 1200 patients with soft tissue masses at limbs, trunk, and girdles were studied from January 2007 to December 2011 in a regional referral centre for the diagnosis and treatment of bone and soft tissue tumours, in accordance with the Guidelines on Soft Tissue Sarcomas established by the local regional health service [18]. All demographic details, clinical data and imaging were entered into a database as patients were investigated and treated, resulting in a prospective archive of soft tissue masses. All patients (for children and minors, their parents) gave their informed consent to perform the exams and to use their clinical data for study purposes. The local institutional board approved the study, which was carried out according to the Code of Ethics of the World Medical Association. The following exams were performed for local staging: X-ray when indicated, US without and with contrast medium enhancement (CEUS) [19], MRI without and with contrast medium when malignancy was suspected or size >5 cm, US-guided core needle biopsy when indicated [20]. Contrast-enhanced CT scan with angiographic study of the mass and relative body segment was performed according to the following criteria: 1) malignancy strongly suspected because of clinical and imaging characteristics; 2) adjacency to or suspected involvement of major vessels and/or bone; 3) intra- and perilesional calcifications/ ossifications shown by other radiological exams. According to these indications, the preoperative diagnostic strategy required contrast-enhanced CT investigation with angiographic study of the pathological segment in 88 patients with soft tissue masses (7 %), 43 males, 45
Eur Radiol
females, mean age 59 years (range 9–89); all patients had been previously studied by US, CEUS, and MRI. All 88 patients were included in this study. The lesions were located in the lower limb in 53 cases, in the upper limb in 16 cases, in the pelvic girdle in 11 cases, and in the trunk in eight cases. In 48 out of 88 cases the mass was intracompartmental, while in the remaining 40 the lesion was extracompartmental, according to Enneking classification [2]. The histological diagnosis was soft tissue sarcoma (STS) in 61 cases (20 low grade, LG; 41 high grade, HG), other malignant ST tumours in four cases, benign and inflammatory lesions in 14 cases, malignant and aggressive bone tumours with large soft tissue invasion in nine cases. Detailed histological diagnoses are reported in Table 1. In malignant tumours, local staging was completed with thorax and abdomen CT study to detect metastases. Seven out of 88 patients were submitted to
Table 1 Histological diagnosis and grading, defined on definitive surgical samples, of the 88 soft tissue lesions included in this study
preoperative chemotherapy, followed by new staging with CT and MRI in order to identify any changes in size, tissue characteristics and relationship with major vessel or bone. All 88 patients were operated on: in 82 cases, an en bloc excision with wide surgical margins and limb salvage was possible; amputation or disarticulation was performed in six cases. Diagnostic and therapeutic strategies were decided at the weekly Interdisciplinary Sarcoma Group meeting with dedicated radiologists, orthopaedic surgeons, pathologists, oncologists, and radiotherapists. Contrast-enhanced CT and CT angiography (CTA) technique A Multidetector Lightspeed Plus Advantage CT equipment, kV 80–120, mA 160–250 (General Electric Medical Systems, Milwaukee, WI, USA) was used. A
Histological diagnosis
Number of cases
Soft tissue malignant tumors
65
Liposarcoma Leiomyosarcoma
High Grade Soft Tissue Sarcoma: 41 Low Grade Soft Tissue Sarcoma: 20 Other ST Tumours: 4 15 Low Grade (mixoid variant 6), 5 High Grade 17 High Grade
Myxofibrosarcoma Pleomorphic Sarcoma
4 Low Grade, 4 High Grade 8 High-Grade
Malignant Peripheral Nerve Sheath Tumor (MPNST) Spindle Cell Sarcoma Soft Tissue Chondrosarcoma
4 High Grade
Fibrous Sarcoma
1 High Grade
2 High Grade 1 Low Grade
Soft Tissue Carcinoma Metastasis Merkel’s Cell Carcinoma Soft tissue benign tumors and inflammatory lesions Lipoma Giant Cell Tumour Elastofibroma Neurofibroma Leyomioma Desmoid Tumour
2 2
Myositis Ossificans Bone malignant and aggressive tumors Metastasis of Carcinoma Malignant Fibrous Histiocytoma (MFH) Osteosarcoma Ewing Sarcoma Giant Cell Tumour of Bone Total
1
14 7 2 1 1 1 1 9 2 2 2 1 2 88
Eur Radiol Table 2 Subgrouping of contrast-enhanced CT images according to the relationship between the pathologic mass and the major vessels
Type 1
Distance of mass from major vessels >1 cm
Type 2
Mass adjacent to major vessel with evidence of thin adipose film interposition
Type 3 Type 4a
Mass adjacent to major vessels without adipose film interposition Major vessel partially encased into the mass
Type 4b
Major vessel totally encased into the mass
dose of iodinated contrast medium (Iomeron™ 300 mg/ mL, 400 mg/mL, range 70–150 mL; in children and minor patients, 2 mL/kg) varying according to patient’s weight, general conditions, kidney function, and age, was administered at an infusion rate, chosen also according to heart functionality. To limit irradiation, online modulation of tube current and reduction of imaging length to the tumour region were used. Acquisitions were made during the arteriovenous phase at a varying time from infusion start depending on patient’s age and lesion site with a range of 60–70 s. Also, the thickness of the acquisition layer was modified in every case according to lesion site and size and patient’s age (scans of 2.5 or 1.25 mm, table (tbl) 7.5 mm/rotation, High Quality (HQ) pitch 3) with possibility of subsequent retro-reconstructions with a thinner layer at 1.25 mm with reconstruction interval at 0.75. The images were reconstructed with standard algorithm (for soft tissues) and bone plus algorithm (for bones).
The axial images were transmitted to an Advantage 4.0 workstation for the multiplane reconstructions (MPR), maximum intensity projection (MIP) to analyse vessel displacement and stenosis or occlusion; 3-dimensional (3D) reconstruction was performed to detect the longitudinal extent in case of suspected neurovascular involvement. Preoperative contrast-enhanced CT images were evaluated in a masked fashion by five radiologists engaged in the Sarcoma Group having at least 10 years of experience in musculoskeletal imaging. All data were compared with surgical findings, gross pathology and microscopic histology of the excised samples (hereinafter referred to as Bhistology^). The characteristics of contrast enhancement were assessed by ROI (region of interest), evaluated in six points along the major axis, before and after administration of contrast medium, and were scored as follows: absence of contrast enhancement; mild homogeneous regular, when contrast enhancement ranged from 10 to 20 HU
Table 3 Correlation between contrast-enhanced CT images and 1) the quality and distribution of enhancement; 2) the relationship of the mass with major vessel; 3) the bone involvement Contrast enhancement Relationship of the tumours with the major vessels
Relationship of the tumours with the bone
Lesions
None or Moderate or Type 1 Type 2 Type 3 Type 4a Type 4b None Adjacent Osteolysis mild severe
ST benign Lesions (14 cases) LG STS (20 cases) HG STS (41 cases)
11
3
4
4
6
0
0
6
8
0
17 1
3 40
8 16
4 8
7 7
1 3
0 7
8 27
11 11
1 3
3 1
1 8
2 2
0 2
0 2
0 0
2 3
1 0
1 0
2 9
Other malignant STT (4 cases) Malignant and aggressive bone tumours (9 cases) Legend: Lesions:
Soft Tissue (ST) benign tumours and aggressive lesions Low Grade Soft Tissue Sarcoma (LG STS) High Grade Soft Tissue Sarcoma (HG STS) Other malignant Soft Tissue Tumours (STT) Quality and distribution of enhancement: Absent or Mild: no contrast enhancement after contrast medium injection or mild (10–20 HU) homogeneous regular enhancement Moderate or severe: moderate (20–30 HU) or severe (>30 HU) dishomogeneous anarchic enhancement Relationship of the masses with the major vessels: see Table 2 Bone involvement: absent (none); pathological mass adjacent/in contiguity to bone; partial or severe and extended osteolysis.
Eur Radiol Table 4 Accuracy of contrast-enhanced CT in the diagnosis of active and aggressive soft tissue tumours in comparison to histology of surgically resected tumours Aggressive lesion at contrastenhanced CT Malignant at 49 final histological diagnosis Benign or 6 LG sarcoma at final diagnosis Total 55
Active lesions at contrastenhanced CT
Sensitivity Predictive Total (number Specificity Value of cases)
5
54
Sensitivity PPV 89 % 90 %
28
34
Specificity NPV 84 % 82 %
33
88
Legend: Active lesion at contrast-enhanced CT: absent or mild (10–20 HU) homogeneous regular contrast enhancement Aggressive lesion: moderate (20–30 HU) or severe (>30 HU) inhomogeneous irregular anarchic enhancement From the total of 88 cases studied by contrast-enhanced CT, the sensitivity in differentiating active lesions (such as benign and low grade malignant tumours) from aggressive tumours (such as high grade malignant tumours) was 89 % and specificity 84 %. The positive predictive value (PPV) for correct diagnosis was 90 % and negative predictive value (NPV) 82 %
(Hounsfield Units) and distribution was homogeneous in all areas; moderate irregular, when contrast enhancement ranged from 20 to 30 HU and distribution was irregular with avascular (necrosis or myxoid areas) and/or
Fig. 1 Contrast-enhanced CT of an intramuscular mass in the left thigh of a woman, 55 years old. Histological diagnosis: liposarcoma-lipomalike (low grade sarcoma). a basal examination: the mass has density of fat tissue, maximum diameter 6 cm. b contrast enhancement is mild homogeneous and regular, major vessels are dislocated by the mass, but
hypervascular zones; severe irregular inhomogeneous, when contrast enhancement was >30 HU and distribution was irregular and anarchic. Evaluation of accuracy of contrast-enhanced CT in discriminating active lesions from aggressive tumours in comparison to histology of surgically excided samples To evaluate accuracy, we grouped together as Bactive lesions^ all masses characterised by absent or mild regular enhancement, in accordance with regular limited vascular supply at histology and in vivo local active behaviour. We grouped together as Baggressive tumours^ all masses characterised by moderately or severely inhomogeneous irregular anarchic contrast enhancement, with avascular areas, in accordance with abundant neoangiogenesis and necrotic and/or hemorrhagic areas at histology; in vivo these tumours were characterised by rapid local and at distance invasion. Imaging was compared with the surgically excised samples and histological results. Sensibility, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. Evaluation of the role of CTA in deciding preoperatively the surgical margins and technique with reference to major vessels and bone Cross sectional images were used to classify the relationship between the mass and the major vessels (Table 2): 1) type 1: distance of the mass from the major vessel >1 cm; 2) type 2: mass adjacent to the major vessel with evidence of a thin adipose film interposition; 3) type 3: mass adjacent to vessel
a thin adipose film is interposed (type 2). According to diagnosis and relationship with vessels, the sarcoma was excised en bloc sparing the superficial femoral artery, but removing the adventitia considered adequate surgical margin
Eur Radiol
in relation to total malignant cases, and the number of soft tissue malignant tumours adjacent to bone or with osteolysis in relation to the total number of ST malignant tumours.
Results Accuracy of contrast-enhanced CT in discriminating active lesions from aggressive tumours in comparison to histology of surgically excided samples
Fig. 2 Contrast enhanced CT of an intramuscular mass in the right thigh of a woman, 52 years old. Histological diagnosis: high grade myxofibrosarcoma. a CT images with bone window, maximum diameter 6.5 cm. b contrast-enhanced CT image with soft tissue window shows severe irregular and inhomogeneous contrast enhancement; major vessels are dislocated without any detectable adipose film interposition (type 3); the femur has superficial erosion. According to diagnosis and relationship with vessel and bone, the sarcoma was excised en bloc with the superficial femoral artery adventia and periostium as adequate surgical margins
without adipose film interposition; 4) type 4a: vessel partially encased; and type 4b: totally encased in the lesion. The surgical margins were adequate when in type 2 and type 3 the dissection enclosed the vessel adventitia (adventitia peeling), and in type 4, the en bloc resection included the vessel with consequent vascular anastomosis/transplant or amputation. The relationship between the mass and the bone was classified as follows: mass distant from bone without any involvement, mass adjacent to bone with interposition of a thin adipose film, mass invading bone with osteolysis (even minimal cortical irregularities and/or thinning). When the tumour was in contact with the bone, the surgical margins were adequate when the en bloc resection enclosed the periosteum. When the tumour invaded the bone, the surgical margins were adequate when the en bloc resection enclosed the bone, with consequent bone transplant/ prosthesis or amputation. The percentage of malignant tumours in which the vessels and/or bone involvement was crucial for the surgical strategy to be adopted was calculated considering only pertinent cases, respectively the number of malignant tumours of types 3 and 4
Results concerning the contrast enhancement quality and distribution are schematically reported in Table 3. Contrast enhancement was absent or mild and regularly homogeneous in 11 benign lesions, 17 low grade STS, one high grade STS, three other ST malignant tumours, and one malignant bone tumour. Contrast enhancement was moderately or severely inhomogeneous with avascular areas in three benign lesions, three low grade STS, 40 high grade STS, one other ST tumour, and eight malignant or aggressive bone tumours. Considering the two groups of active lesions (benign tumours and low grade STS) and aggressive tumours, contrast enhancement was absent or mild and regular in 28 out of 34 cases of benign lesions or low grade STS (82 %), and moderately or severely inhomogeneous in 49 out of 54 high grade malignant tumours (90 %). Sensitivity of contrast-enhanced CT in discriminating the active from the aggressive tumours was 89 %, specificity was 84 %, with PPV of 90 %, and NPV of 82 % (Table 4). CT showed areas of necrosis in 26 out of 88 cases and calcifications in eight cases, confirmed by histology.
The role of CTA in deciding preoperatively surgical margins and technique with reference to major vessels and bone Results concerning the relationship of tumours with the major vessels and the bone are schematically reported in Table 3. As regards the involvement of major vessels, in 54 out of 88 cases the mass was distant or well separated by an adipose film (types 1 and 2) (Fig. 1). Absence of adipose film interposed between the tumour and major vessels was demonstrated in six benign lesions, seven low grade STS, and seven high grade STS (Fig. 2). The vessels were partially (type 4a) encased by the mass in one low grade and three high grade STS, and totally enclosed (type 4b) in seven high grade STS (Fig. 3), two other ST malignant tumours, and three malignant bone tumours. The correspondence between CT images and histology on surgically excised samples was complete in all cases (100 %). The strict relationship between the mass and major vessels detected by CTA was fundamental to obtain adequate surgical
Eur Radiol Fig. 3 Contrast enhanced CT of a large mass of the left thigh of a man, 68 years old. Histological diagnosis: high grade leiomyosarcoma. a basal examination: voluminous mass subverting the fibres of the quadriceps femoris, maximum diameter 10 cm. b Arterious phase: the superficial femoral artery is totally enclosed into the mass (type 4b). c Venous phase and d with bone window: contrast enhancement is abundant, irregular, and heterogeneous, with avascular and hypervascularised areas; partial erosion of internal cortex and hyperdensity of bone marrow are evident. e optimization of mass display and longitudinal extent with 3D reconstruction. f the surgically en bloc resected sarcoma with wide margins: the superficial femoral artery is enclosed in the resected mass (it will be reconstructed with great controlateral saphena)
margins (as described in Methods) in 32 out of 74 malignant tumours (43 %) characterised by types 3 and 4. As regards the involvement of bone, the mass was adjacent to bone in eight benign lesions, 11 low grade STS, 11 high grade STS, and one other malignant ST tumour; osteolysis was present in one low grade STS, three high grade STS, and two other malignant tumours. The involvement of bone detected by CT was essential for keeping adequate surgical margins (as described in Methods) in 17 malignant tumours out of 29 (58 %).
Discussion The contrast-enhanced CT is widely used in the investigation of many diseases such as cardiovascular, gastrointestinal, genitourinary, hepatobiliary, and cerebral diseases, but only few authors have presented it as a reference method in the staging of musculoskeletal tumours [9–11, 15]. Contrast-enhanced CT is used for its accuracy in angiographic studies [8, 17] to detect mineralization of soft tissue masses, depict bone details [10, 21–24], and as an excellent alternative to MR, especially when MRI is contraindicated or difficult to apply [11, 24]. The introduction of multilayer CT has permitted obtaining
multiphase vascular studies extended to several body districts in a short time. These technical implements have also allowed minimizing the artefacts induced by the patient’s movements, which enables obtaining a growing number of images, susceptible to additional multiplane and volumetric processes [22]. In the present study there is evidence that, in selected cases, contrast-enhanced CT can be also an investigation tool of chief importance in soft tissue tumours of the extremities and girdles both for diagnosis hypothesis and surgical planning. We observed that benign lesions and many low grade sarcomas, characterised in vivo by slow local development and absence of distal metastases and, therefore, defined as Bactive^ tumours, showed a regular and homogeneous CT contrast enhancement. On the contrary, high grade ST sarcomas and other malignant tumours, characterised by highly Baggressive^ invasion both locally and at distance, showed an irregular inhomogeneous and anarchic distribution of the contrast medium, with multiple avascular necrotic and/or hemorrhagic areas. These results are consistent with the pathophysiology of STS and histology and can be related to the new microvessels stimulated by angiogenic factors [25]. In our study, the accuracy of CT contrast enhancement distribution in differentiating active lesions tumours from aggressive sarcomas was high: sensitivity 89 %, specificity 84 %, PPV
Eur Radiol
90 %, and NPV 82 %. Whereas the CT ability to demonstrate hypo- and avascular areas is well documented [24], its accuracy in evaluating the different vascular network characteristics of the tumoral tissue seems to be a new observation. Our results seem to be comparable to MRI sensitivity (range 73– 91 %, in accordance with different parameters) and specificity (range 72–97 %) in discriminating malignant from benign lesions [15, 26, 27]. Furthermore, our results seem to be also in accordance with a preliminary report on CEUS high accuracy in identifying malignant tumours (sensitivity 89 %, specificity 85 %, PPV 86 %, NPV 88 %) [28]. The efficacy and safety of the surgical procedure as well as the incidence of local recurrences are strongly conditioned by preoperative planning [3–6]. Surgical technique varies according to site (compartment) and size of the mass, histotype and grading, presence of distant metastases, possibility of reconstruction, residual disability, response to radio-chemiotherapy, and patient’s characteristics and informed will [1–6]. Thanks to CTA, in our study, the strict relationship between the mass and major vessels was instrumental in selecting surgical strategy and surgical margins in 43 % of the malignant cases. These data are consistent with previous studies [17–29]; they demonstrated that axial CTA is very accurate and specific in identifying arterial encasement and releasable plane. In particular, the identification of contiguity of more than half of the artery circumference had specificity 93 %, negative predictive value 98 %, sensitivity 90 %. Our results are complementary to the study of Feydy et al. [30] concerning the accuracy of MRI in detecting vascular invasion by musculoskeletal tumours. Our study confirms the conclusions of Argin et al. [31] that CTA is extremely accurate in evaluating vascular invasion, even if we did not perform a comparison between MRI and CT. The local staging with contrast-enhanced CT in our series was completed at same time by general staging optimizing timing and resources. This is in accordance with the literature that considers CT to be the gold standard in general staging for thorax and abdomen study [9, 10, 32]. This paper has the following limits: first, it is a retrospective study, but the progressive entering of all data into a database as patients were investigated and treated resulted in a prospective archive of soft tissue masses. Eighty-eight cases is a small number, but substantial if viewed in relation to other studies [12, 15]. Furthermore, in our series, the 88 cases represented only 7 % of the patients studied in that period, in compliance with the radioprotection regulation which imposes reserving CT only to selected cases. US [3–6], possibly CEUS [28], and MRI continue to be the gold standard [7–15, 27, 33]. Concerning the different contrast enhancements in active and aggressive tumours, our assessment was done in a qualitative manner. In any case, qualitative vascular assessment is common practice in radiology, for exam ple, the
haemodynamics of blood supply to liver masses is assessed with the eye. Promising results have recently been observed with FDG-PET [34]. In the near future it will be interesting to compare the accuracy of contrast-enhanced CT to that of MRI and possibly of CEUS in a multicentre study, in order to obtain a sufficient number of cases for an accurate statistical evaluation.
Conclusions Contrast-enhanced CT and CTA are accurate, sensitive, and specific in preoperative local staging of soft tissue masses. They can identify aggressive high grade tumours and indicate the exact extension and margins of the lesion allowing correct surgical planning. Contrast-enhanced CT is indicated in selected cases with strong suspicion for malignancy, large mass size, proximity to bone and major vessels. When MRI cannot be performed due to patient-related contraindications and local logistics (e.g. in less developed countries with low financial means), contrast-enhanced CT can help also for local staging. In agreement with Enneking, who used to say, BIn the study of tissue masses, there are no hip-shooters and gutreactors^, this work suggests to adopt not a unique imaging approach for all kinds of soft tissue tumours, but a multimodality tailored diagnostic strategy, where CT can play, in selected cases, a fundamental role.
Acknowledgments The scientific guarantor of this publication is Faletti Carlo. The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article. The authors state that this work has not received any funding. One of the authors has significant statistical expertise. Institutional Review Board approval was obtained. Written informed consent was obtained from all subjects (patients) in this study. No study subjects or cohorts have been previously reported. Methodology: retrospective, observational, performed at one institution.
References 1. 2. 3. 4.
5. 6. 7.
Campanacci M (1999) Bone and soft tissue tumours. SpringerVerlag, New York Enneking WF (1990) Clinical musculoskeletal pathology. 3 Rev Sub. University of Florida Press NICE Improving Outcomes Guidelines for soft tissue sarcoma. Cancer service guidance CSG Sarcoma. Issued: March 2006 Casali PG, Jost L, Sleijfer S, Verweij J, Blay JY (2008) Soft tissue sarcomas: ESMO clinical recommendations for diagnosis, treatment and follow up. Ann Oncol 19:ii89–ii93 Grimer R, Judson I, Peake D, Seddon B (2010) Guidelines for the management of soft tissue sarcomas. Sarcoma 2010:506182 Cutts S, Ferrero A, Piana R, Haywood R (2012) The management of soft tissue sarcomas. Review. Surgeon 10:25–32 Singer S (1999) New diagnostic modalities in soft tissue sarcoma. Semin Surg Oncol 17:11–22
Eur Radiol 8. 9.
10. 11.
12.
13.
14.
15.
16. 17.
18.
19.
20.
21.
22.
Balach T, Stacy GS, Haydon RC (2011) The clinical evaluation of soft tissue tumors. Radiol Clin N Am 49:1185–1196, vi Colleran G, Madewell J, Foran P, Shelly M, O’Sullivan PJ (2011) Imaging of soft tissue and osseous sarcomas of the extremities. Semin Ultrasound CT MR 32:442–455 Knapp EL, Kransdorf MJ, Letson GD (2005) Diagnostic imaging update: soft tissue sarcomas. Cancer Control 12:22–26 Tzeng C-WD, Smith JK, Heslin MJ (2007) Soft tissue sarcoma: preoperative and postoperative imaging for staging. Surg Oncol Clin N Am 16:389–402 Tuncbilek N, Karakas HM, Okten OO (2005) Dynamic contrast enhanced MRI in the differential diagnosis of soft tissue tumors. Eur J Radiol 53:500–505 Vilanova JC, Woertler K, Narvaez JA et al (2007) Soft tissue tumors update: MR imaging features according to the WHO classification. Eur Radiol 17:125–138 Daniel A Jr, Ullah E, Wahab S, Kumar V Jr (2009) Relevance of MRI in prediction of malignancy of musculoskeletal system—a prospective evaluation. BMC Musculoskelet Disord 10:125 Yamamoto T, Kurosaka M, Soejima T, Fujii M (2001) Contrastenhanced three-dimensional helical CT for soft tissue tumors in the extremities. Skelet Radiol 30:384–387 Cheong D, Letson GD (2011) Computer-assisted navigation and musculoskeletal sarcoma surgery. Cancer Control 18:171–176 Thévenin FS, Drapé JL, Biau D et al (2010) Assessment of vascular invasion by bone and soft tissue tumours of the limbs: usefulness of MDCT angiography. Eur Radiol 20:1524–1531 Comandone A, Brach del Prever EM, Aglietta M et al (2004, revised 2008) Guide Lines soft tissue sarcoma. Piedmont Oncological Network Press, Turin. www.reteoncologica.it Gay F, Pierucci F, Zimmerman V et al (2012) Contrast-enhanced ultrasonography of peripheral soft-tissue tumors: feasibility study and preliminary results. Diagn Interv Imaging 93:37–46 De Marchi A, Brach del Prever EM, Linari A et al (2010) Accuracy of core-needle biopsy after contrast-enhanced ultrasound in softtissue tumours. Eur Radiol 20:2740–2748 Masciocchi C, Lanni G, Conti L et al (2012) Soft-tissue inflammatory myofibroblastic tumors (IMTs) of the limbs: potential and limits of diagnostic imaging. Skelet Radiol 41:643–649 Ilaslan H, Sundaram M (2006) Advances in musculoskeletal tumor imaging. Orthop Clin N Am 37:375–391, vii
23.
24.
25. 26. 27.
28.
29.
30.
31.
32.
33.
34.
Errani C, Kreshak J, Ruggieri P et al (2013) Imaging of bone tumors for the musculoskeletal oncologic surgeon. Eur J Radiol 82:2083– 2091 Morris CD, Parsons TW 3rd, Schwab JH, Panicek DM (2012) Imaging interpretation of oncologic musculoskeletal conditions. Instr Course Lect 61:541–551 Huwart L, Michoux N, Van Beers BE (2007) Magnetic resonance imaging of angiogenesis in tumors. J Radiol 88:331–338 Fenstermacher MJ (2003) Imaging evaluation of patients with soft tissue sarcoma. Surg Oncol Clin N Am 12:305–332 Pang KK, Hughes T (2003) MR imaging of the musculoskeletal soft tissue mass: is heterogeneity a sign of malignancy? J Chin Med Assoc 66:655–661 Loizides A, Peer S, Plaikner M, Djurdjevic T, Gruber H (2012) Perfusion pattern of musculoskeletal masses using contrastenhanced ultrasound: a helpful tool for characterization? Eur Radiol 22:1803–1811 Li Y, Zheng Y, Lin J, Cai A, Zhou X, Wei X et al (2013) Evaluation of the relationship between extremity soft tissue sarcomas and adjacent major vessels using contrast-enhanced multidetector CT and three-dimensional volume-rendered CT angiography: a preliminary study. Acta Radiol 54:966–972 Feydy A, Anract P, Tomeno B, Chevrot A, Drapé JL (2006) Assessment of vascular invasion by musculoskeletal tumors of the limbs: use of contrast enhanced MR angiography. Radiology 238:611–621 Argin M, Isayev H, Kececi B, Arkun R, Sabah D (2009) Multidetector-row computed tomographic angiography findings of musculoskeletal tumors: retrospective analysis and correlation with surgical findings. Acta Radiol 50:1150–1159 King DM, Hackbarth DA, Kilian CM, Carrera GF (2009) Softtissue sarcoma metastases identified on abdomen and pelvis CT imaging. Clin Orthop Relat Res 467:2838–2844 Lee SY, Jee WH, Jung JY, Park MY, Kim SK, Jung CK et al (2015) Differentiation of malignant from benign soft tissue tumours: use of additive qualitative and quantitative diffusion weighted MR imaging to standard MR imaging at 3.0 T. Eur Radiol. doi:10.1007/ s00330-015-3878-x Fendler WP, Chalkidis RP, Ilhan H, Knösel T, Herrmann K, Issels RD et al (2015) Evaluation of several FDG PET parameters for prediction of soft tissue tumour grade at primary diagnosis and recurrence. Eur Radiol 25:2214–2221
MUSCULOSKELETAL
Accuracy and role of contrast-enhanced CT in diagnosis and surgical planning in 88 soft tissue tumours of extremities Lucia Verga 1 & Elena Maria Brach del Prever 2,1 & Alessandra Linari 3 & Sara Robiati 1 & Armanda De Marchi 1 & Domenico Martorano 1 & Michele Boffano 4 & Raimondo Piana 4 & Carlo Faletti 1
Received: 24 March 2015 / Revised: 8 September 2015 / Accepted: 23 September 2015 # European Society of Radiology 2015
Abstract Objectives Soft tissue tumours (STT) require accurate diagnosis in order to identify potential malignancies. Preoperative planning is fundamental to avoid inadequate treatments. The role of contrast-enhanced computed tomography (CT) for local staging remains incompletely assessed. Aims of the study were to evaluate CT accuracy in discriminating active from aggressive tumours compared to histology and evaluate the role of CT angiography (CTA) in surgical planning. Materials and methods This retrospective cohort series of 88 cases from 1200 patients (7 %) was locally studied by contrast-enhanced CT and CTA in a referral centre: 74 malignant tumours, 14 benign lesions. Contrast-enhancement patterns and relationship of the mass with major vessels and bone were compared with histology on surgically excised samples. Sensitivity, specificity, positive and negative predictive values (PPV, NPV) were evaluated in discriminating active from aggressive tumours.
Results Sensitivity in differentiating aggressive tumours from active lesions was 89 %, specificity 84 %, PPV 90 %, NPV 82 %. The relationship between mass and major vessels/bone was fundamental for surgical strategy respectively in 40 % and in 58 % of malignant tumours. Conclusion Contrast-enhanced CT and CTA are effective in differentiating aggressive masses from active lesions in soft tissue and in depicting the relationship between tumour and adjacent bones and major vessels. Key Points • Accurate delineation of vascular and bony involvement preoperatively is fundamental for a correct resection. • CT plays a critical role in differential diagnosis of soft tissue masses. • Contrast-enhanced CT and CT angiography are helpful in depicting tumoral vascular involvement. • CT is optimal for characterization of bone involvement in soft tissue malignancies.
* Lucia Verga [email protected]
Raimondo Piana [email protected]
Elena Maria Brach del Prever [email protected]
Carlo Faletti [email protected]
Alessandra Linari [email protected]
1
Department of Imaging, Azienda Ospedaliera Città della Salute e della Scienza, CTO Hospital, Via Zuretti 29, 10126 Torino, Italy
Sara Robiati [email protected]
2
Department of Public Health and Paediatrics, University of Turin, Via Santena 5-bis, 10126 Torino, Italy
Armanda De Marchi [email protected]
3
Department of Pathology, Azienda Ospedaliera Città della Salute e della Scienza, Regina Margherita Hospital, Piazza Polonia, 10126 Torino, Italy
4
Department of Orthopaedic Oncology and Reconstructive Surgery, Azienda Ospedaliera Città della Salute e della Scienza, CTO Hospital, Via Zuretti 29, 10126 Torino, Italy
Domenico Martorano [email protected] Michele Boffano [email protected]
Eur Radiol
Keywords Contrast-enhanced CT . CT angiography . Soft tissue sarcoma . Surgical planning . Tumour vascularization
Introduction Soft tissue sarcomas (STS) are a rare heterogeneous group representing 1 % of all tumours, often characterised by a large, and for a long time asymptomatic, mass in the limbs, where benign tumours and benign inflammatory lesions are very frequent [1, 2]. Depending on histotype and grading, STS may have a prolonged locally active behaviour with slow development (e.g. liposarcoma-lipomalike and some low grade liposarcomas), or they may have an aggressive development with rapid local invasion and, eventually, metastases at distance (e.g. high-grade pleiomorphic sarcoma) [1, 2]. The first-choice treatment of STS is surgical en bloc resection, to achieve local tumour control. The responsiveness to adjuvant therapies (chemo- and radiotherapy) is limited and low [3–6]. Surgical margins can be wide (the en bloc excision is intracompartmental with a cuff of normal tissue) or radical (the excision is extracompartmental); if the excision is marginal or contaminated, the resection is inadequate and it negatively influences the disease-free survival [1–6]. In the extremities and girdles, limb salvage versus amputation should be evaluated on the basis of the possibility of obtaining adequate surgical margins with acceptable residual function. For limb salvage, when the tumour involves the major vessel, two main surgical solutions are possible: 1) the wide en bloc tumour excision including the partially or totally encased vessel (and the vascular reconstruction with autologous transplant or vascular prosthesis has to be performed); 2) the wide en bloc excision including only the vessel adventitia (adventitia peeling), considered a sufficiently adequate margin, when the vessel is strictly adjacent to the tumour. The choice of the treatment strategy is often difficult, mainly based on imaging. The main challenge is to use a sensitive and specific imaging technique, able to discriminate a potentially malignant from a benign tumour and to depict the relationship of the tumour with major vessels and bone. In local staging, ultrasound (US) is useful to discriminate cysts from solid masses, to measure the diameter and to identify the location of a mass in relation to the deep fascia: a mass larger than 5 cm and deep with respect to the deep fascia should be considered potentially malignant [3–6]. Magnetic resonance imaging (MRI) is the gold standard to discriminate between potentially malignant aggressive masses and benign lesions [7–14] and to assess the compartment and the size for surgical planning [3, 5, 6]. Contrast-enhanced computed tomography (CT) is used for patients who, for various reasons, cannot be submitted to MRI [10] or when local logistic supplies are limited, such as in new
developing countries with low financial support [15, 16]. CT is accurate in identifying calcifications and assessing the degree of bone involvement [11, 15], but its accuracy in discriminating benign from malignant tumours in local staging is still undetermined. CT angiography (CTA) is well documented in showing the relationship of tumour with major vessels [16, 17]. Aims of this study were: 1) to evaluate the accuracy of contrast-enhanced CT in discriminating active from aggressive tumours in comparison to histology of surgically excised samples; 2) to evaluate the accuracy of contrast-enhanced CT and CTA in comparison to histology as an aid in deciding preoperatively the surgical margin and technique with reference to major vessels and bone. In this study, contrast-enhanced CT was part of a diagnostic strategy conducted by the interdisciplinary team in a regional referral centre for the study and treatment of bone and soft tissue sarcomas.
Materials and methods A total of 1200 patients with soft tissue masses at limbs, trunk, and girdles were studied from January 2007 to December 2011 in a regional referral centre for the diagnosis and treatment of bone and soft tissue tumours, in accordance with the Guidelines on Soft Tissue Sarcomas established by the local regional health service [18]. All demographic details, clinical data and imaging were entered into a database as patients were investigated and treated, resulting in a prospective archive of soft tissue masses. All patients (for children and minors, their parents) gave their informed consent to perform the exams and to use their clinical data for study purposes. The local institutional board approved the study, which was carried out according to the Code of Ethics of the World Medical Association. The following exams were performed for local staging: X-ray when indicated, US without and with contrast medium enhancement (CEUS) [19], MRI without and with contrast medium when malignancy was suspected or size >5 cm, US-guided core needle biopsy when indicated [20]. Contrast-enhanced CT scan with angiographic study of the mass and relative body segment was performed according to the following criteria: 1) malignancy strongly suspected because of clinical and imaging characteristics; 2) adjacency to or suspected involvement of major vessels and/or bone; 3) intra- and perilesional calcifications/ ossifications shown by other radiological exams. According to these indications, the preoperative diagnostic strategy required contrast-enhanced CT investigation with angiographic study of the pathological segment in 88 patients with soft tissue masses (7 %), 43 males, 45
Eur Radiol
females, mean age 59 years (range 9–89); all patients had been previously studied by US, CEUS, and MRI. All 88 patients were included in this study. The lesions were located in the lower limb in 53 cases, in the upper limb in 16 cases, in the pelvic girdle in 11 cases, and in the trunk in eight cases. In 48 out of 88 cases the mass was intracompartmental, while in the remaining 40 the lesion was extracompartmental, according to Enneking classification [2]. The histological diagnosis was soft tissue sarcoma (STS) in 61 cases (20 low grade, LG; 41 high grade, HG), other malignant ST tumours in four cases, benign and inflammatory lesions in 14 cases, malignant and aggressive bone tumours with large soft tissue invasion in nine cases. Detailed histological diagnoses are reported in Table 1. In malignant tumours, local staging was completed with thorax and abdomen CT study to detect metastases. Seven out of 88 patients were submitted to
Table 1 Histological diagnosis and grading, defined on definitive surgical samples, of the 88 soft tissue lesions included in this study
preoperative chemotherapy, followed by new staging with CT and MRI in order to identify any changes in size, tissue characteristics and relationship with major vessel or bone. All 88 patients were operated on: in 82 cases, an en bloc excision with wide surgical margins and limb salvage was possible; amputation or disarticulation was performed in six cases. Diagnostic and therapeutic strategies were decided at the weekly Interdisciplinary Sarcoma Group meeting with dedicated radiologists, orthopaedic surgeons, pathologists, oncologists, and radiotherapists. Contrast-enhanced CT and CT angiography (CTA) technique A Multidetector Lightspeed Plus Advantage CT equipment, kV 80–120, mA 160–250 (General Electric Medical Systems, Milwaukee, WI, USA) was used. A
Histological diagnosis
Number of cases
Soft tissue malignant tumors
65
Liposarcoma Leiomyosarcoma
High Grade Soft Tissue Sarcoma: 41 Low Grade Soft Tissue Sarcoma: 20 Other ST Tumours: 4 15 Low Grade (mixoid variant 6), 5 High Grade 17 High Grade
Myxofibrosarcoma Pleomorphic Sarcoma
4 Low Grade, 4 High Grade 8 High-Grade
Malignant Peripheral Nerve Sheath Tumor (MPNST) Spindle Cell Sarcoma Soft Tissue Chondrosarcoma
4 High Grade
Fibrous Sarcoma
1 High Grade
2 High Grade 1 Low Grade
Soft Tissue Carcinoma Metastasis Merkel’s Cell Carcinoma Soft tissue benign tumors and inflammatory lesions Lipoma Giant Cell Tumour Elastofibroma Neurofibroma Leyomioma Desmoid Tumour
2 2
Myositis Ossificans Bone malignant and aggressive tumors Metastasis of Carcinoma Malignant Fibrous Histiocytoma (MFH) Osteosarcoma Ewing Sarcoma Giant Cell Tumour of Bone Total
1
14 7 2 1 1 1 1 9 2 2 2 1 2 88
Eur Radiol Table 2 Subgrouping of contrast-enhanced CT images according to the relationship between the pathologic mass and the major vessels
Type 1
Distance of mass from major vessels >1 cm
Type 2
Mass adjacent to major vessel with evidence of thin adipose film interposition
Type 3 Type 4a
Mass adjacent to major vessels without adipose film interposition Major vessel partially encased into the mass
Type 4b
Major vessel totally encased into the mass
dose of iodinated contrast medium (Iomeron™ 300 mg/ mL, 400 mg/mL, range 70–150 mL; in children and minor patients, 2 mL/kg) varying according to patient’s weight, general conditions, kidney function, and age, was administered at an infusion rate, chosen also according to heart functionality. To limit irradiation, online modulation of tube current and reduction of imaging length to the tumour region were used. Acquisitions were made during the arteriovenous phase at a varying time from infusion start depending on patient’s age and lesion site with a range of 60–70 s. Also, the thickness of the acquisition layer was modified in every case according to lesion site and size and patient’s age (scans of 2.5 or 1.25 mm, table (tbl) 7.5 mm/rotation, High Quality (HQ) pitch 3) with possibility of subsequent retro-reconstructions with a thinner layer at 1.25 mm with reconstruction interval at 0.75. The images were reconstructed with standard algorithm (for soft tissues) and bone plus algorithm (for bones).
The axial images were transmitted to an Advantage 4.0 workstation for the multiplane reconstructions (MPR), maximum intensity projection (MIP) to analyse vessel displacement and stenosis or occlusion; 3-dimensional (3D) reconstruction was performed to detect the longitudinal extent in case of suspected neurovascular involvement. Preoperative contrast-enhanced CT images were evaluated in a masked fashion by five radiologists engaged in the Sarcoma Group having at least 10 years of experience in musculoskeletal imaging. All data were compared with surgical findings, gross pathology and microscopic histology of the excised samples (hereinafter referred to as Bhistology^). The characteristics of contrast enhancement were assessed by ROI (region of interest), evaluated in six points along the major axis, before and after administration of contrast medium, and were scored as follows: absence of contrast enhancement; mild homogeneous regular, when contrast enhancement ranged from 10 to 20 HU
Table 3 Correlation between contrast-enhanced CT images and 1) the quality and distribution of enhancement; 2) the relationship of the mass with major vessel; 3) the bone involvement Contrast enhancement Relationship of the tumours with the major vessels
Relationship of the tumours with the bone
Lesions
None or Moderate or Type 1 Type 2 Type 3 Type 4a Type 4b None Adjacent Osteolysis mild severe
ST benign Lesions (14 cases) LG STS (20 cases) HG STS (41 cases)
11
3
4
4
6
0
0
6
8
0
17 1
3 40
8 16
4 8
7 7
1 3
0 7
8 27
11 11
1 3
3 1
1 8
2 2
0 2
0 2
0 0
2 3
1 0
1 0
2 9
Other malignant STT (4 cases) Malignant and aggressive bone tumours (9 cases) Legend: Lesions:
Soft Tissue (ST) benign tumours and aggressive lesions Low Grade Soft Tissue Sarcoma (LG STS) High Grade Soft Tissue Sarcoma (HG STS) Other malignant Soft Tissue Tumours (STT) Quality and distribution of enhancement: Absent or Mild: no contrast enhancement after contrast medium injection or mild (10–20 HU) homogeneous regular enhancement Moderate or severe: moderate (20–30 HU) or severe (>30 HU) dishomogeneous anarchic enhancement Relationship of the masses with the major vessels: see Table 2 Bone involvement: absent (none); pathological mass adjacent/in contiguity to bone; partial or severe and extended osteolysis.
Eur Radiol Table 4 Accuracy of contrast-enhanced CT in the diagnosis of active and aggressive soft tissue tumours in comparison to histology of surgically resected tumours Aggressive lesion at contrastenhanced CT Malignant at 49 final histological diagnosis Benign or 6 LG sarcoma at final diagnosis Total 55
Active lesions at contrastenhanced CT
Sensitivity Predictive Total (number Specificity Value of cases)
5
54
Sensitivity PPV 89 % 90 %
28
34
Specificity NPV 84 % 82 %
33
88
Legend: Active lesion at contrast-enhanced CT: absent or mild (10–20 HU) homogeneous regular contrast enhancement Aggressive lesion: moderate (20–30 HU) or severe (>30 HU) inhomogeneous irregular anarchic enhancement From the total of 88 cases studied by contrast-enhanced CT, the sensitivity in differentiating active lesions (such as benign and low grade malignant tumours) from aggressive tumours (such as high grade malignant tumours) was 89 % and specificity 84 %. The positive predictive value (PPV) for correct diagnosis was 90 % and negative predictive value (NPV) 82 %
(Hounsfield Units) and distribution was homogeneous in all areas; moderate irregular, when contrast enhancement ranged from 20 to 30 HU and distribution was irregular with avascular (necrosis or myxoid areas) and/or
Fig. 1 Contrast-enhanced CT of an intramuscular mass in the left thigh of a woman, 55 years old. Histological diagnosis: liposarcoma-lipomalike (low grade sarcoma). a basal examination: the mass has density of fat tissue, maximum diameter 6 cm. b contrast enhancement is mild homogeneous and regular, major vessels are dislocated by the mass, but
hypervascular zones; severe irregular inhomogeneous, when contrast enhancement was >30 HU and distribution was irregular and anarchic. Evaluation of accuracy of contrast-enhanced CT in discriminating active lesions from aggressive tumours in comparison to histology of surgically excided samples To evaluate accuracy, we grouped together as Bactive lesions^ all masses characterised by absent or mild regular enhancement, in accordance with regular limited vascular supply at histology and in vivo local active behaviour. We grouped together as Baggressive tumours^ all masses characterised by moderately or severely inhomogeneous irregular anarchic contrast enhancement, with avascular areas, in accordance with abundant neoangiogenesis and necrotic and/or hemorrhagic areas at histology; in vivo these tumours were characterised by rapid local and at distance invasion. Imaging was compared with the surgically excised samples and histological results. Sensibility, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. Evaluation of the role of CTA in deciding preoperatively the surgical margins and technique with reference to major vessels and bone Cross sectional images were used to classify the relationship between the mass and the major vessels (Table 2): 1) type 1: distance of the mass from the major vessel >1 cm; 2) type 2: mass adjacent to the major vessel with evidence of a thin adipose film interposition; 3) type 3: mass adjacent to vessel
a thin adipose film is interposed (type 2). According to diagnosis and relationship with vessels, the sarcoma was excised en bloc sparing the superficial femoral artery, but removing the adventitia considered adequate surgical margin
Eur Radiol
in relation to total malignant cases, and the number of soft tissue malignant tumours adjacent to bone or with osteolysis in relation to the total number of ST malignant tumours.
Results Accuracy of contrast-enhanced CT in discriminating active lesions from aggressive tumours in comparison to histology of surgically excided samples
Fig. 2 Contrast enhanced CT of an intramuscular mass in the right thigh of a woman, 52 years old. Histological diagnosis: high grade myxofibrosarcoma. a CT images with bone window, maximum diameter 6.5 cm. b contrast-enhanced CT image with soft tissue window shows severe irregular and inhomogeneous contrast enhancement; major vessels are dislocated without any detectable adipose film interposition (type 3); the femur has superficial erosion. According to diagnosis and relationship with vessel and bone, the sarcoma was excised en bloc with the superficial femoral artery adventia and periostium as adequate surgical margins
without adipose film interposition; 4) type 4a: vessel partially encased; and type 4b: totally encased in the lesion. The surgical margins were adequate when in type 2 and type 3 the dissection enclosed the vessel adventitia (adventitia peeling), and in type 4, the en bloc resection included the vessel with consequent vascular anastomosis/transplant or amputation. The relationship between the mass and the bone was classified as follows: mass distant from bone without any involvement, mass adjacent to bone with interposition of a thin adipose film, mass invading bone with osteolysis (even minimal cortical irregularities and/or thinning). When the tumour was in contact with the bone, the surgical margins were adequate when the en bloc resection enclosed the periosteum. When the tumour invaded the bone, the surgical margins were adequate when the en bloc resection enclosed the bone, with consequent bone transplant/ prosthesis or amputation. The percentage of malignant tumours in which the vessels and/or bone involvement was crucial for the surgical strategy to be adopted was calculated considering only pertinent cases, respectively the number of malignant tumours of types 3 and 4
Results concerning the contrast enhancement quality and distribution are schematically reported in Table 3. Contrast enhancement was absent or mild and regularly homogeneous in 11 benign lesions, 17 low grade STS, one high grade STS, three other ST malignant tumours, and one malignant bone tumour. Contrast enhancement was moderately or severely inhomogeneous with avascular areas in three benign lesions, three low grade STS, 40 high grade STS, one other ST tumour, and eight malignant or aggressive bone tumours. Considering the two groups of active lesions (benign tumours and low grade STS) and aggressive tumours, contrast enhancement was absent or mild and regular in 28 out of 34 cases of benign lesions or low grade STS (82 %), and moderately or severely inhomogeneous in 49 out of 54 high grade malignant tumours (90 %). Sensitivity of contrast-enhanced CT in discriminating the active from the aggressive tumours was 89 %, specificity was 84 %, with PPV of 90 %, and NPV of 82 % (Table 4). CT showed areas of necrosis in 26 out of 88 cases and calcifications in eight cases, confirmed by histology.
The role of CTA in deciding preoperatively surgical margins and technique with reference to major vessels and bone Results concerning the relationship of tumours with the major vessels and the bone are schematically reported in Table 3. As regards the involvement of major vessels, in 54 out of 88 cases the mass was distant or well separated by an adipose film (types 1 and 2) (Fig. 1). Absence of adipose film interposed between the tumour and major vessels was demonstrated in six benign lesions, seven low grade STS, and seven high grade STS (Fig. 2). The vessels were partially (type 4a) encased by the mass in one low grade and three high grade STS, and totally enclosed (type 4b) in seven high grade STS (Fig. 3), two other ST malignant tumours, and three malignant bone tumours. The correspondence between CT images and histology on surgically excised samples was complete in all cases (100 %). The strict relationship between the mass and major vessels detected by CTA was fundamental to obtain adequate surgical
Eur Radiol Fig. 3 Contrast enhanced CT of a large mass of the left thigh of a man, 68 years old. Histological diagnosis: high grade leiomyosarcoma. a basal examination: voluminous mass subverting the fibres of the quadriceps femoris, maximum diameter 10 cm. b Arterious phase: the superficial femoral artery is totally enclosed into the mass (type 4b). c Venous phase and d with bone window: contrast enhancement is abundant, irregular, and heterogeneous, with avascular and hypervascularised areas; partial erosion of internal cortex and hyperdensity of bone marrow are evident. e optimization of mass display and longitudinal extent with 3D reconstruction. f the surgically en bloc resected sarcoma with wide margins: the superficial femoral artery is enclosed in the resected mass (it will be reconstructed with great controlateral saphena)
margins (as described in Methods) in 32 out of 74 malignant tumours (43 %) characterised by types 3 and 4. As regards the involvement of bone, the mass was adjacent to bone in eight benign lesions, 11 low grade STS, 11 high grade STS, and one other malignant ST tumour; osteolysis was present in one low grade STS, three high grade STS, and two other malignant tumours. The involvement of bone detected by CT was essential for keeping adequate surgical margins (as described in Methods) in 17 malignant tumours out of 29 (58 %).
Discussion The contrast-enhanced CT is widely used in the investigation of many diseases such as cardiovascular, gastrointestinal, genitourinary, hepatobiliary, and cerebral diseases, but only few authors have presented it as a reference method in the staging of musculoskeletal tumours [9–11, 15]. Contrast-enhanced CT is used for its accuracy in angiographic studies [8, 17] to detect mineralization of soft tissue masses, depict bone details [10, 21–24], and as an excellent alternative to MR, especially when MRI is contraindicated or difficult to apply [11, 24]. The introduction of multilayer CT has permitted obtaining
multiphase vascular studies extended to several body districts in a short time. These technical implements have also allowed minimizing the artefacts induced by the patient’s movements, which enables obtaining a growing number of images, susceptible to additional multiplane and volumetric processes [22]. In the present study there is evidence that, in selected cases, contrast-enhanced CT can be also an investigation tool of chief importance in soft tissue tumours of the extremities and girdles both for diagnosis hypothesis and surgical planning. We observed that benign lesions and many low grade sarcomas, characterised in vivo by slow local development and absence of distal metastases and, therefore, defined as Bactive^ tumours, showed a regular and homogeneous CT contrast enhancement. On the contrary, high grade ST sarcomas and other malignant tumours, characterised by highly Baggressive^ invasion both locally and at distance, showed an irregular inhomogeneous and anarchic distribution of the contrast medium, with multiple avascular necrotic and/or hemorrhagic areas. These results are consistent with the pathophysiology of STS and histology and can be related to the new microvessels stimulated by angiogenic factors [25]. In our study, the accuracy of CT contrast enhancement distribution in differentiating active lesions tumours from aggressive sarcomas was high: sensitivity 89 %, specificity 84 %, PPV
Eur Radiol
90 %, and NPV 82 %. Whereas the CT ability to demonstrate hypo- and avascular areas is well documented [24], its accuracy in evaluating the different vascular network characteristics of the tumoral tissue seems to be a new observation. Our results seem to be comparable to MRI sensitivity (range 73– 91 %, in accordance with different parameters) and specificity (range 72–97 %) in discriminating malignant from benign lesions [15, 26, 27]. Furthermore, our results seem to be also in accordance with a preliminary report on CEUS high accuracy in identifying malignant tumours (sensitivity 89 %, specificity 85 %, PPV 86 %, NPV 88 %) [28]. The efficacy and safety of the surgical procedure as well as the incidence of local recurrences are strongly conditioned by preoperative planning [3–6]. Surgical technique varies according to site (compartment) and size of the mass, histotype and grading, presence of distant metastases, possibility of reconstruction, residual disability, response to radio-chemiotherapy, and patient’s characteristics and informed will [1–6]. Thanks to CTA, in our study, the strict relationship between the mass and major vessels was instrumental in selecting surgical strategy and surgical margins in 43 % of the malignant cases. These data are consistent with previous studies [17–29]; they demonstrated that axial CTA is very accurate and specific in identifying arterial encasement and releasable plane. In particular, the identification of contiguity of more than half of the artery circumference had specificity 93 %, negative predictive value 98 %, sensitivity 90 %. Our results are complementary to the study of Feydy et al. [30] concerning the accuracy of MRI in detecting vascular invasion by musculoskeletal tumours. Our study confirms the conclusions of Argin et al. [31] that CTA is extremely accurate in evaluating vascular invasion, even if we did not perform a comparison between MRI and CT. The local staging with contrast-enhanced CT in our series was completed at same time by general staging optimizing timing and resources. This is in accordance with the literature that considers CT to be the gold standard in general staging for thorax and abdomen study [9, 10, 32]. This paper has the following limits: first, it is a retrospective study, but the progressive entering of all data into a database as patients were investigated and treated resulted in a prospective archive of soft tissue masses. Eighty-eight cases is a small number, but substantial if viewed in relation to other studies [12, 15]. Furthermore, in our series, the 88 cases represented only 7 % of the patients studied in that period, in compliance with the radioprotection regulation which imposes reserving CT only to selected cases. US [3–6], possibly CEUS [28], and MRI continue to be the gold standard [7–15, 27, 33]. Concerning the different contrast enhancements in active and aggressive tumours, our assessment was done in a qualitative manner. In any case, qualitative vascular assessment is common practice in radiology, for exam ple, the
haemodynamics of blood supply to liver masses is assessed with the eye. Promising results have recently been observed with FDG-PET [34]. In the near future it will be interesting to compare the accuracy of contrast-enhanced CT to that of MRI and possibly of CEUS in a multicentre study, in order to obtain a sufficient number of cases for an accurate statistical evaluation.
Conclusions Contrast-enhanced CT and CTA are accurate, sensitive, and specific in preoperative local staging of soft tissue masses. They can identify aggressive high grade tumours and indicate the exact extension and margins of the lesion allowing correct surgical planning. Contrast-enhanced CT is indicated in selected cases with strong suspicion for malignancy, large mass size, proximity to bone and major vessels. When MRI cannot be performed due to patient-related contraindications and local logistics (e.g. in less developed countries with low financial means), contrast-enhanced CT can help also for local staging. In agreement with Enneking, who used to say, BIn the study of tissue masses, there are no hip-shooters and gutreactors^, this work suggests to adopt not a unique imaging approach for all kinds of soft tissue tumours, but a multimodality tailored diagnostic strategy, where CT can play, in selected cases, a fundamental role.
Acknowledgments The scientific guarantor of this publication is Faletti Carlo. The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article. The authors state that this work has not received any funding. One of the authors has significant statistical expertise. Institutional Review Board approval was obtained. Written informed consent was obtained from all subjects (patients) in this study. No study subjects or cohorts have been previously reported. Methodology: retrospective, observational, performed at one institution.
References 1. 2. 3. 4.
5. 6. 7.
Campanacci M (1999) Bone and soft tissue tumours. SpringerVerlag, New York Enneking WF (1990) Clinical musculoskeletal pathology. 3 Rev Sub. University of Florida Press NICE Improving Outcomes Guidelines for soft tissue sarcoma. Cancer service guidance CSG Sarcoma. Issued: March 2006 Casali PG, Jost L, Sleijfer S, Verweij J, Blay JY (2008) Soft tissue sarcomas: ESMO clinical recommendations for diagnosis, treatment and follow up. Ann Oncol 19:ii89–ii93 Grimer R, Judson I, Peake D, Seddon B (2010) Guidelines for the management of soft tissue sarcomas. Sarcoma 2010:506182 Cutts S, Ferrero A, Piana R, Haywood R (2012) The management of soft tissue sarcomas. Review. Surgeon 10:25–32 Singer S (1999) New diagnostic modalities in soft tissue sarcoma. Semin Surg Oncol 17:11–22
Eur Radiol 8. 9.
10. 11.
12.
13.
14.
15.
16. 17.
18.
19.
20.
21.
22.
Balach T, Stacy GS, Haydon RC (2011) The clinical evaluation of soft tissue tumors. Radiol Clin N Am 49:1185–1196, vi Colleran G, Madewell J, Foran P, Shelly M, O’Sullivan PJ (2011) Imaging of soft tissue and osseous sarcomas of the extremities. Semin Ultrasound CT MR 32:442–455 Knapp EL, Kransdorf MJ, Letson GD (2005) Diagnostic imaging update: soft tissue sarcomas. Cancer Control 12:22–26 Tzeng C-WD, Smith JK, Heslin MJ (2007) Soft tissue sarcoma: preoperative and postoperative imaging for staging. Surg Oncol Clin N Am 16:389–402 Tuncbilek N, Karakas HM, Okten OO (2005) Dynamic contrast enhanced MRI in the differential diagnosis of soft tissue tumors. Eur J Radiol 53:500–505 Vilanova JC, Woertler K, Narvaez JA et al (2007) Soft tissue tumors update: MR imaging features according to the WHO classification. Eur Radiol 17:125–138 Daniel A Jr, Ullah E, Wahab S, Kumar V Jr (2009) Relevance of MRI in prediction of malignancy of musculoskeletal system—a prospective evaluation. BMC Musculoskelet Disord 10:125 Yamamoto T, Kurosaka M, Soejima T, Fujii M (2001) Contrastenhanced three-dimensional helical CT for soft tissue tumors in the extremities. Skelet Radiol 30:384–387 Cheong D, Letson GD (2011) Computer-assisted navigation and musculoskeletal sarcoma surgery. Cancer Control 18:171–176 Thévenin FS, Drapé JL, Biau D et al (2010) Assessment of vascular invasion by bone and soft tissue tumours of the limbs: usefulness of MDCT angiography. Eur Radiol 20:1524–1531 Comandone A, Brach del Prever EM, Aglietta M et al (2004, revised 2008) Guide Lines soft tissue sarcoma. Piedmont Oncological Network Press, Turin. www.reteoncologica.it Gay F, Pierucci F, Zimmerman V et al (2012) Contrast-enhanced ultrasonography of peripheral soft-tissue tumors: feasibility study and preliminary results. Diagn Interv Imaging 93:37–46 De Marchi A, Brach del Prever EM, Linari A et al (2010) Accuracy of core-needle biopsy after contrast-enhanced ultrasound in softtissue tumours. Eur Radiol 20:2740–2748 Masciocchi C, Lanni G, Conti L et al (2012) Soft-tissue inflammatory myofibroblastic tumors (IMTs) of the limbs: potential and limits of diagnostic imaging. Skelet Radiol 41:643–649 Ilaslan H, Sundaram M (2006) Advances in musculoskeletal tumor imaging. Orthop Clin N Am 37:375–391, vii
23.
24.
25. 26. 27.
28.
29.
30.
31.
32.
33.
34.
Errani C, Kreshak J, Ruggieri P et al (2013) Imaging of bone tumors for the musculoskeletal oncologic surgeon. Eur J Radiol 82:2083– 2091 Morris CD, Parsons TW 3rd, Schwab JH, Panicek DM (2012) Imaging interpretation of oncologic musculoskeletal conditions. Instr Course Lect 61:541–551 Huwart L, Michoux N, Van Beers BE (2007) Magnetic resonance imaging of angiogenesis in tumors. J Radiol 88:331–338 Fenstermacher MJ (2003) Imaging evaluation of patients with soft tissue sarcoma. Surg Oncol Clin N Am 12:305–332 Pang KK, Hughes T (2003) MR imaging of the musculoskeletal soft tissue mass: is heterogeneity a sign of malignancy? J Chin Med Assoc 66:655–661 Loizides A, Peer S, Plaikner M, Djurdjevic T, Gruber H (2012) Perfusion pattern of musculoskeletal masses using contrastenhanced ultrasound: a helpful tool for characterization? Eur Radiol 22:1803–1811 Li Y, Zheng Y, Lin J, Cai A, Zhou X, Wei X et al (2013) Evaluation of the relationship between extremity soft tissue sarcomas and adjacent major vessels using contrast-enhanced multidetector CT and three-dimensional volume-rendered CT angiography: a preliminary study. Acta Radiol 54:966–972 Feydy A, Anract P, Tomeno B, Chevrot A, Drapé JL (2006) Assessment of vascular invasion by musculoskeletal tumors of the limbs: use of contrast enhanced MR angiography. Radiology 238:611–621 Argin M, Isayev H, Kececi B, Arkun R, Sabah D (2009) Multidetector-row computed tomographic angiography findings of musculoskeletal tumors: retrospective analysis and correlation with surgical findings. Acta Radiol 50:1150–1159 King DM, Hackbarth DA, Kilian CM, Carrera GF (2009) Softtissue sarcoma metastases identified on abdomen and pelvis CT imaging. Clin Orthop Relat Res 467:2838–2844 Lee SY, Jee WH, Jung JY, Park MY, Kim SK, Jung CK et al (2015) Differentiation of malignant from benign soft tissue tumours: use of additive qualitative and quantitative diffusion weighted MR imaging to standard MR imaging at 3.0 T. Eur Radiol. doi:10.1007/ s00330-015-3878-x Fendler WP, Chalkidis RP, Ilhan H, Knösel T, Herrmann K, Issels RD et al (2015) Evaluation of several FDG PET parameters for prediction of soft tissue tumour grade at primary diagnosis and recurrence. Eur Radiol 25:2214–2221
