| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||
Editorial |
Diagnostic radiology
Some papers create more interest and discussion than others and one example of such was the Commentary on the subject of nephrogenic systemic fibrosis [1]. This dealt with the recently recognized association between gadolinium-based contrast agents and this condition, and reported the methods being developed to eliminate this problem with newer contrast agents. The complicated correspondence on this subject highlighted how seriously the pharmaceutical companies are considering this issue.
Positron emission tomography (PET) is being increasingly used, especially for oncology patients, and it is vital to recognize variants that do not have an oncological basis. Fukuchi et al [2] described variations in myocardial uptake that could cause confusion, especially in fasting patients. The combination of PET with CT is also advancing rapidly. Pfannenberg et al [3] discussed the value of contrast-enhanced multiphase CT combined with PET, and showed that this technique changed the interpretation in 42% of patients and affected clinical management in 21%.
Tissue characterization has always been an ultimate goal of imaging and, although we are far from achieving it in most situations, Liu et al [4] elegantly demonstrated that, when using contrast-enhanced ultrasound for hepatocellular carcinoma, the timing of the lesion becoming hypoenhancing correlates well with tumour cell differentiation. A retrospective analysis of digital stereotaxis in breast screening demonstrated significant improvements in the assessment process compared with conventional stereotaxis but several potential biases in the study may also have contributed to improvements in the process [5]. Texture analysis of tissue-surrounding microcalcifications may play a role in the computer-aided diagnosis of digitized mammographical images and could help to reduce unnecessary biopsies [6].
With the increasing use and development of picture archiving and communication systems (PACS) in diagnostic radiology departments, there have been a number of papers reviewing image perception. Buls et al [7] evaluated the influence of display quality on radiologists' performance and showed how important a quality assurance (QA) programme is if performance is to be maximized. This was supported by the study of Thompson et al [8] who, rather worryingly, showed that a significant number of monitors did not perform within tolerances and had to be adjusted or replaced soon after delivery. These concerns over display screen QA and performance requirements were discussed further in a Commentary by Brettle [9], including topics such as selection, installation and maintenance; this article also highlighted the lack of clarity regarding the legislation relating to the use of display screens in radiology.
Donovan and Manning in their Commentary [10] explained how radiologists search an image for pathology and the importance of understanding this when new methods of presenting images are devised. They also describe how vital it is that new software is matched to the abilities and limitations of those using the equipment. The importance of ability was underlined by Brettle et al [11] who confirmed that the detectability of abnormalities increased with observer experience, but that this correlation depended on the composition of the local area anatomical noise.
In interventional radiology, Wildgruber et al [12] investigated the early endothelial and haematological responses to cryoplasty in comparison with angioplasty of the superficial femoral artery. Angioplasty is complicated by a restenosis rate of 30–40% within 6 months, and it has been suggested that cryoplasty in combination with balloon angioplasty will reduce this restenosis rate by inhibiting neointimal hyperplasia. The investigators studied the levels of various adhesion molecules, growth factors and cytokines that are released after endothelial injury in patients undergoing angiography, angioplasty and cryoplasty. They did not find any statistically significant difference between the group undergoing angioplasty and that undergoing cryoplasty. This study, along with other recently published trials, shows that, to date, no significant benefit of cryoplasty over angioplasty alone has been proven.
Two papers concerning haemoptysis were published in the BJR this year. Khalil et al [13] reported on the use of high-resolution CT (HRCT) in Respiratory Intensive Care Unit patients, concluding that HRCT and fibre optic bronchoscopy are complimentary techniques for bleeding site localization, with HRCT being more accurate in determining the aetiology of haemoptysis, especially in patients with bronchiectasis and tuberculosis. In May 2007, Poyanli et al [14] reported on their series of 140 patients with severe haemoptysis, all treated by bronchial artery embolisation. The cause of haemoptysis was tuberculosis in 136 cases and malignancy in the other 4. Embolisation successfully controlled the bleeding in 98.5% of cases, with 90% remaining free of bleeding at 1 month. Rebleeding occurred in 14 patients within 1–7 months, 12 of whom underwent repeat embolisation. Two patients with malignant tumour died from haemorrhage. There were no procedural complications. The authors conclude that safe and effective palliation is achieved by embolisation in the majority of cases after a single intervention, and that repeat embolisation is possible if necessary.
Two novel techniques were reported for dealing with ureteric strictures of various kinds. Thiruchelvam et al [15] described their method for dealing with difficult ureteroileal anastomotic strictures in patients with ileal conduits using a "double wire technique", which allows retrograde balloon dilatation and stent placement. Olsburgh et al [16] described the placement of extra-anatomic stents in two cases of transplant ureteric stenosis that had failed to respond to standard endourological techniques. Both procedures were performed under general anaesthesia and involved inserting an extra anatomic silicone–polytetrafluoroethylene bonded stent using fluoroscopic guidance (via an existing nephrostomy track) into an open cystostomy (via a subcutaneous track). Both patients had freely draining transplant pelvicalyceal systems at 12 months, with renal functions remaining at the patients' baseline levels.
Radiofrequency ablation (RFA) is well established in treating liver, lung and kidney tumours. Marcy et al [17] described their preliminary experience when using RFA to treat breast tumours in four breast cancer patients aged >70 years, who had requested minimally invasive treatment. Five tumours were treated in the four patients under local anaesthesia and using ultrasound guidance. They report successful RFA lump ablation in four out of the five tumours. There was one local recurrence at 4 months and one abscess occurred at 9 months. All patients were alive and without other signs of recurrence or metastases at 29.4 months. The authors conclude that RFA may be a reasonable, less invasive alternative to lumpectomy in elderly patients with selected early carcinoma of the breast, but note that large controlled trials are required to validate the efficacy of this therapy.
During 2007, several original papers were published in the BJR highlighting the advancement in physiological MRI techniques: namely MR perfusion, MR spectroscopy, functional MRI (fMRI) and diffusion tensor imaging. These techniques have of course been well documented for many years, but their clinical application is still relatively "small scale". These papers demonstrated the use of these techniques in the clinical environment, and were produced by major research establishments with a strong clinical involvement.
Narayana et al [18] described the use of MR spectroscopy and fMRI for the identification of high-grade structures and critical eloquent areas in patients undergoing intensity-modulated radiation therapy (IMRT). To maximize the benefit of IMRT, areas of high-grade tumour must be identified and targeted separately from low-grade tumour and normal eloquent brain in an attempt to reduce the incidence of recurrence within the radiation field. However, dose escalation carries the risk of damage to areas that are eloquent or of lower grade, and techniques to better identify the areas of high-grade tumour in non-critical areas would be of significant benefit. Narayana et al used MR spectroscopy to identify high-grade components within heterogeneous gliomas (defined by a choline:creatine ratio of >3, in accordance with previous correlative work), delineating areas within the gross tumour volume that were metabolically and histologically more aggressive. Furthermore, fMRI was performed using paradigms to define the motor areas, language areas or both, and the images were overlaid onto structural images to delineate anatomically critical eloquent loci. The combination of the spectroscopic and fMRI data helped to identify areas of tumour that were high grade and non-eloquent, and therefore suitable for a focal incremental IMRT dose.
Haroon et al [19] validated a "new" perfusion MR method for evaluation of the vascularity of gliomas. There is a clear relationship between microvascular density, tumour grade and prognosis, and perfusion MRI has been suggested as a method for its non-invasive evaluation. This technique has also been advocated for the targeting of stereotactic biopsy and for modulation of post-surgical radiotherapy. Traditionally used perfusion techniques have significant disadvantages owing to the spatial distortion that occurs as a result of paramagnetic variation in areas of extravascular contrast leakage. Haroon et al [19] have validated a more recently described perfusion method (based on T1 dynamic contrast enhancement, instead of T2* susceptibility weighted imaging). This has the advantage of reducing the amount of spatial distortion, as it is free from leakage effects, with theoretical advantages in terms of temporal resolution, tissue coverage and freedom from image distortion. The paper validated the quantitative effectiveness of the technique, and suggested that it should be used as the perfusion method of choice in the evaluation of cerebral tumours. This could have a significant effect on MR practice as a result.
Lai et al [20] reported the use of diffusion tensor imaging in the assessment of long tract involvement in acute lacunar cerebral infarction. Lacunar infarcts are a common cause of neurological disability, but it is difficult to assess the precise location of the infarcts in relation to the main sensory and motor pathways passing through the internal capsule. This is of importance for prognostic reasons and in order to evaluate the secondary effect of focal lacunar infarction. In their study, Lai et al [20] elegantly demonstrated the corticospinal tracts, and mapped their involvement by lacunar infarction in a series of patients. The extent of involvement of the corticospinal tract by lacunar infarcts was demonstrated by diffusion tensor imaging and correlated with the degree of disability. The authors point out that this helps to establish topographical accuracy of diffusion tensor imaging in the fibre tracking of the corticospinal tract, and shows the utility of this technique for prognostic reasons in the clinical arena.
Throughout 2007, the BJR has continued to publish important papers on the interface between imaging and the practice of oncology. In addition to the advances in RFA, MR spectroscopy and fMRI mentioned previously [17, 18], further information on how novel imaging techniques might influence practice has been published by Jackson et al [21] (distortion-corrected T2 weighted MRI for prostate radiotherapy planning) and by Veit-Haibach et al [22] (the use of fluorodeoxyglucose (FDG)-PET/CT in restaging of patients with recurrent breast cancer). Another important aspect of imaging, which should not be overlooked, is that of interpretation. This will become increasingly important, as oncologists must interpret complex sets of images in order to delineate tumours and critical organs accurately. It is salutary to be reminded that imaging is not about black and white; it is about shades of grey. It is also about what the interpreter brings to the process, and not just a simple matter of objective appraisal of data [10, 23–24]. We would do well to remember this when making clinical decisions, particularly in that artificial environment that is the multidisciplinary team meeting.
Radiotherapy and oncology
Imaging is an essential part of delivering effective treatment with radiation. We need to be aware that the imaging itself is not without risk, but that these risks need to be put into perspective [25], particularly for concomitant imaging [26, 27]. This difficulty is further compounded when we attempt to conduct clinical research in radiation oncology where we may find a degree of tension between the various bodies involved in the governance of such projects [28, 29]. Legislative and administrative minefields are not the only obstacle to expanding the role of clinical research in UK radiotherapy. There is also the question of capacity, as shown by the important survey carried out under the auspices of ACORRN (Academic Clinical Oncology and Radiobiology Research Network) [30]. The demands of service in a target-driven environment prevent us from asking (and perhaps even answering) the questions we need to ask. Doing is encouraged at the expense of reflection: we simply keep on keeping on.
The management of errors and incidents in radiotherapy received wide coverage in the BJR during 2007. Jones et al [31] explored the radiobiological aspects of deriving compensation for the interruption of palliative treatment. One of the conclusions of a survey carried out by Dale et al [32] on methods for compensating treatment interruptions and errors was a need for improved training in practical aspects of radiobiology for all radiotherapy staff. In his 2006 Report on the Health of the Nation [33], the Chief Medical Officer (CMO) for England and Wales made some interesting observations about safety in clinical radiotherapy. The BJR had already published a Commentary earlier in the year on radiotherapy incidents and on the need for a system to disseminate the lessons to be learnt; it also described the remit of a multidisciplinary working party established by the Royal College of Radiologists to examine these issues [34]. The use of in vivo dosimetry (IVD) as an important safety measure was advocated in the CMO's report. However, a survey showed that there had been little change in IVD practice in the UK over the previous 10 years [35]. The BJR was able to contribute usefully to the debate on this important topic [36], including a detailed analysis of the wider issues raised by the CMO's report [37].
Radiobiology
Radiobiology featured prominently in 2007, with the publication of a special issue highlighting the programme of the joint meeting in Leicester of the Association for Radiation Research and the European Society for Radiation Biology. The dominant theme of the contributions to this issue was radiation injury to normal tissues, with particular emphasis on the signalling pathways involved in mediating and responding to damage. The importance of oxidative and nitrosative stress in regulating the severity of late normal tissue reactions was reviewed by Zhao et al [38], who identified anti-inflammatory interventions as a promising therapeutic approach. Oxidative stress was also implicated in the mechanism whereby tumour necrosis factor-
, released form irradiated cells, acts as a bystander signalling molecule capable of inducing genetic damage in endothelial cells [39]. The concept of reactive oxygen and nitrogen species as mediators of bystander signalling was supported by a study in normal human fibroblasts [40], which showed that the free radical scavenger dimethyl sulfoxide could eliminate the bystander signalling between cells that had received a low dose of particle irradiation and those that were unirradiated but still showed DNA double-strand breaks. Lindsay et al [41] reported markedly different response kinetics for key genes, including p53, p21 and Bax, involved in the cellular and tissue response to irradiation, and emphasised that these responses were tissue specific. This could contribute to differences in the kinetics and severity of radiation damage in various tissues. The importance of a key pathway involving signalling by the Rho family of proteins in response to radiation was reviewed by Haydont et al [42], with reference to fibrogenic processes in normal tissues. The roles of other stromal elements including enteric neurones [43] and mesenchymal stem cells [44] were also reviewed. Other papers focused on the genetic risk in leukaemia [45] and the use of chromosome translocations as a reliable indicator of past radiation exposure [46]. Another potentially important factor involved in the pathogenesis of radiation induced damage to normal tissue — hypoxia — was identified by Westbury et al [47] in a study using the hypoxia markers carbonic anhydrase IX and pimonidazole in human breast tissue. An increase in our understanding of the factors that determine whether any given irradiated tissue will exhibit late radiation reactions will offer the prospect of a meaningful intervention to reduce their severity, either before or after irradiation.
Radiation protection
Two areas in which the balance of radiation risk and diagnostic benefit becomes more crucial are (i) radiological examinations of the very young, because of the increase in tissue sensitivity with decreasing age, and (ii) mammography screening of asymptomatic individuals.
10 years after the issue of the European Directive requiring member states to establish diagnostic reference levels (DRLs), efforts are still being made to establish DRLs for children to reflect their change in stature with age. Onnasch et al [48] concluded that dose–area product (DAP) divided by body weight should be used for DRLs in paediatric cardiac catheterization instead of mean DAP values for different age groups. Kiljunen et al [49] recommended the specification of DRLs for paediatric chest examinations as a curve of the entrance surface dose, and DAP values as a function of the projected patient thickness.
The risk/benefit balance also assumes a greater importance in mammography, where asymptomatic individuals are screened with low kVp X-rays; this topic continues to generate scientific investigation alongside technological and clinical developments in the technique. Svahn et al [50] found that the dose to the breast from digital mammography could be reduced to one half of the level currently used in Sweden without affecting diagnostic accuracy. Law et al [51] have compiled a set of breast cancer detection/induction ratios for women at different screening ages, and for younger women at different ages with and without a family history of breast cancer. The current debate over the sensitivity to radiation-induced cancer of genetically susceptible women who are carriers of the BRCA1 and BRCA2 genes was summarized succinctly in an editorial [52].
References
| ||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
