Discussion
In this study, our objective was to quantify the absolute PCT parameters of intracerebral tumours. We report one of the largest series concerning the application and efficacy of PCT in the preoperative histopathological grouping of cerebral intra-axial tumours. By applying a CT perfusion method and based on the absolute perfusion parameters, we differentiated cerebral lymphomas from high-grade gliomas (Figs. 4, 5).
Fig. 4 A 58-year-old man with histopathological diagnosis of glioblastoma multiforme WHO IV. Axial contrast-enhanced MIP reconstruction image (a) shows a contrast-enhanced mass, which demonstrates intensely elevated blood flow (b) and volume (c) as well as strongly increased regional K trans (d) in comparison with the normal cortical and subcortical cerebral parenchyma
Fig. 5 A 72-year-old woman with histopathological diagnosis of primary cerebral lymphoma. In comparison with the contralateral normal cerebral parenchyma, the lesion depicted in the right lentiform nucleus demonstrates the typical perfusion characteristics of lymphoma: enhancement (a), no significant increase in CBF (b) or CBV values (c), though intensely increased regional permeability (K Trans, d), indicating a massive disturbance of the blood-brain barrier
Dynamic perfusion imaging is based on the assessment of tissue-related distribution of the contrast medium, which acts as a tracer. Microvascularisation and diffusion across the endothelial membrane into the interstitial space determine the distribution of contrast agent after infusion. The absolute perfusion parameters correspond to the microvascular density, in histopathological examination, which has been considered to be the ‘gold standard’ for such an evaluation, because of its direct association with angiogenic growth factor expression, tumour growth, and metastatic occurrence [10, 11]. In accordance with previously published studies, significantly increased CBV was noted in high-grade gliomas. The increased vascular proliferation of the neoplastic tissue and the hypothesis that feeding arterioles are more vasodilated than normal in neoplasms support these findings [8, 12, 13].
Increased vascular permeability has also been correlated with malignancy and has been evolving as a surrogate marker of tumour angiogenesis and, thus, tumour grade [14]. Higher permeability has been associated with higher tumour grade and has also been shown to decrease, responding to antiangiogenic therapy [3, 15–17]. Our results demonstrated significantly higher permeability values for both high-grade gliomas and lymphomas in comparison with healthy tissue. Both entities are characterised by a histopathologically proven blood-brain barrier disturbance [18, 19].
It has already been shown that PWI provides valuable information concerning tumour perfusion, facilitating the preoperative classification and grading of gliomas [3–6]. Our data stress the role of PCT in the preoperative differential diagnosis of primary central nervous system lymphomas (PCNSL) from high-grade gliomas. PCNSL, a discrete histopathological entity, constitute up to 6% of malignant central nervous system (CNS) tumours [20]. Although PCNSL present certain characteristic magnetic resonance imaging (MRI) findings, it can be difficult or even impossible to differentiate them, on the basis of imaging features, from high-grade gliomas on standard CT or MRI, because of their diffuse infiltrative growth [21, 22]. Histopathologically, contrary to high-grade gliomas, PCNSL are characterised by the absence of neovascularisation. Our results show that lymphomas can be differentiated from high-grade gliomas by comparing CBV and CBF parameters using PCT. Both histopathological entities presented significantly higher permeability values compared with normal brain parenchyma, but only high-grade gliomas presented with significantly higher values of regional CBV and CBF parameters than those of normal cerebral parenchyma. Our results are comparable with those previously reported for a series of patients who underwent PWI [19]. Because PCNSL and high-grade gliomas require different therapeutical management and differ in prognosis, precise diagnosis is crucial [20, 23].
Gliomas, the most frequent cerebral tumours in adults, exhibit varying degrees of cellular and nuclear pleomorphism, mitotic activity, vascular proliferation and necrosis [24, 25]. This histopathological heterogeneity explains the difficulties concerning the preoperative assessment and biopsy and thus histopathological grading. Histopathological assessment of tissue, the current standard for tumour grading, presents inherent limitations, including sampling errors, intra-observer variation and the changing nature of central nervous system tumours [26, 27]. Accurate histopathological diagnosis is crucial to define the appropriate management and prognosis according to tumour grade. For low-grade gliomas, conservative treatment and monitoring for detecting transformation or active proliferation are of great importance. Because the degree of vascular proliferation is one of the most critical elements in the determination of tumour grade and prognosis, the preoperative non-invasive assessment and quantification of glioma vascularity can be helpful to determine the malignant potential of the tumour, to select an appropriate biopsy site, to evaluate transition from low-grade to a high-grade glioma, and also to monitor treatment response [28]. Despite the small number of patients with low-grade gliomas, our data implicate that high-grade gliomas could be differentiated from low-grade gliomas on the basis of all three PCT parameters studied. Low-grade gliomas exhibited no different perfusion parameters compared with normal parenchyma (Fig. 6).
Fig. 6 A 42-year-old man with histopathological diagnosis of low-grade glioma WHO II. The contrast-enhanced MIP image (a) shows no enhancement within the tumour in the right frontal lobe (1). The perfusion maps reveal no significant difference in the perfusion parameters CBF (b), CBV (c) and K Trans (d) between the lesion (1) and the unaffected side (2)
In spite of the feasible advantages of PCT over PWI, the relative limitations of PCT and consequently of our study are the radiation dose involved with the procedure and also the limited coverage area of the cerebral parenchyma compared with PWI. The latter limitation could be overcome in the future with the new volume PCT technique, which enables the measurement of the whole brain and three-dimensional qualitative and quantitative imaging. Thus, whole-brain PCT could lead to even more precise grading of intra-axial brain tumours. Potential limitations also include the different contrast agent protocols applied. However, we statistically demonstrated no methodological heterogeneity regarding the perfusion parameters generated.
In conclusion, with the role of imaging beginning to shift towards providing complementary information concerning tumour dynamics and physiology, our results revealed the promising and determinant role of the perfusion technique in the preoperative histopathological assessment of cerebral intra-axial tumours. By detecting and quantifying microvascular density and capillary permeability PCT helps to distinguish cerebral lymphomas from high-grade gliomas and facilitates the preoperative histopathological grouping of brain gliomas. Information obtained by the in vivo monitoring of tumour proliferation and angiogenesis may support the role of PCT in the clinical routine not only preoperatively but also in the post-treatment assessment and follow-up of patients with intra-axial cerebral tumours. However, additional studies are required to differentiate between patients with high- and low-grade gliomas as well as between those with radionecrosis and recurrence.