- Email: [email protected]
Ultrasound scanning in the radiotherapy department
Clin. Radiol. (1977) 28, 287-293 U L T R A S O U N D S C A N N I N G IN T H E R A D I O T H E R A P Y D E P A R T M E N T P. C. BADCOCK From the Department o f Radiotherapy, Hammersmith Hospital, London W12 OHS Apart from its use by diagnostic radiologists in the detection and staging of tumours, ultrasound scanning appears to offer particular advantages to the radiotherapist in treatment field planning. The following merits are almost self-evident: tumour, organ and body outlines are visualised, with the patient usually in the treatment position, sufficiently accurately for measurement, field planning and for checking of the treatment fields. Moreover, the safety and high patient-acceptability of the scanning technique allows reassessment of tumour morphology with early modification of the treatment regime. In certain cases, heavy reliance is now made on the information from ultrasound scanning since more conventional investigation would not allow imaging of the tumour-affected area in the treatment planes. On occasions carefully planned treatments based on conventional planning procedures have had to be modified because subsequent ultrasound scans have revealed inadequate cover of the tumour or non-optimal treatment field positioning. It would seem advisable that radiotherapy departments should acquire the expertise and personnel to ensure access to ultrasound scanning systems, both to increase the accuracy of defining the tumour-bearing area for treatment planning and to avoid methods of investigation involving invasive techniques.
Successful radiotherapy practice depends on irradiating tumours to a sufficiently high dose to cause their destruction without producing unacceptable damage to the surrounding normal tissues. Especially since the advent of supervoltage equipment, the planning of such treatment has become increasingly sophisticated allowing a better differential between the dose administered to the tumour and that received by the normal tissues. The potential for curing localised malignant disease has thus increased. Unfortunately the techniques for the acquisition of data relating to the patients' anatomy and tumour location have lagged behind progress made in methods of calculating dose distribution. Rapid strides have been made in the proper utilisation of ultrasound pulse-echo systems and B-scanning is now an accepted technique in diagnostic radiology and obstetrics (Wells, 1972). These noninvasive methods are particularly successful in imaging the outline of internal organs, tumours and cysts in cross-sectional planes without recourse to contrast media. Using this technique we have tried to improve the localisation of tumours and radiosensitive organs in the radiotherapy department at the Hammersmith Hospital which treats 1300 new patients each year. We have based our study on the experience of others mainly in departments in the USA (Smith and Holm, 1970; Cohen and Hass, 1971; Eule et al., 1973; Brascho, 1973). So far over 500 patients have been scanned and, after a slow start, we
are studying at a rate of 250 patients per year in a 1 day per week session. Apart from its use by diagnostic radiologists in the detection and staging of tumours, ultrasound scanning appears to offer particular advantages to the radiotherapist in treatment field planning although this application has so far been developed only in a few centres. The technique has many advantages. It is safe at the energy levels used, as on diagnostic machines no harmful effects have been demonstrated in reproducible studies. It is painless and non-invasive; no contrast medium is required, nor is any special preparation of the patient. It is rapid and reduces the requirement for special radiological techniques and isotope scans. Tumour, organ and body outlines are visualised, with the patient usually in the treatment position. Planes including the therapy plane, unable to be visualised by other techniques, may be seen. Although an EMI scanner has superior resolution, scanning in many planes is difficult and the process would be more time consuming and expensive. Because resolution depends on differing acoustic impedances rather than radiodensities, boundaries may be visualised that can not be seen radiologically thus making the technique superior even to transverse axial tomography in some situations. Because of the absence of distortion, accuracy is adequate for measurement and field planning and for verification of precision of the siting of treatment fields. In
288
CLINICAL RADIOLOGY
particular, the study of scans in a number of parallel planes can provide data for planning optimum dose distribution in three dimensions. MoreOver the safety and high patient acceptability of the scanning technique allows repeated reassessment of tumour morphology with consequent early modification of treatment regime. However, there are some disadvantages; being impedance maps of body cross-sections the scans are unfamiliar. A knowledge of physics is required not only for the production of the best scans but also for their interpretation. As in diagnostic radiology and nuclear medicine a pathological diagnosis is usually difficult though solid and cystic lesions are easily distinguished. However, tumours, perhaps because of their basic composition by a single cell type, tend to be relatively echo-free and are also sound attenuating. Some regions are inaccessible as they are shielded by bone or air-filled lung or gut. Poor penetration through thick layers of fat makes the examinations of obese patients more difficult. Resolution in deeper regions may be inadequate to show structures smaller than 3 cm. TECHNIQUE At the Hammersmith Hospital, it has been possible for radiotherapists to obtain direct access to NE4102 Diasonograph, supplied by the DHSS for an instrumental development programme in the Department of Medical Physics, without interference in the routine scanning for radiology and obstetrics. The transducer is either hand-held for contact scanning or submerged in a hand-held water bath. Liquid paraffin is used as the coupling agent as it is less likely to affect the treatment field marks on the skin. Scanning is usually performed in multiple planes 1 - 3 cm apart, in two directions at right angles in order to produce a three-dimensional picture of tumour and normal tissue morphology. Transverse scanning is usually performed first as the various anatomical features are more recognisable. More information can be gained from watching the development of the scan on the screen than by just looking at the final picture as a better appreciation is obtained of these areas of high acoustic attenuation. Surface markers are also included over anatomical landmarks or the position and boundaries of the radiation fields. It is important to produce an accurate body outline, without deforming the shape by undue pressure, and adequate visualisation of the deep structures. Although this can often be achieved in one study sometimes it is necessary to repeat the scans with different settings. It has been shown that it is more accurate to obtain body outlines in this manner than using lead strips or
mechanical jigs and easier than using plaster or plastic templates (Brown et al., 1972). A permanent record of the B mode scan is then made with polaroid photography. Most of our scans have been taken at approximately 2/5 scale - an accurate calibration having been made allowing direct measurement from the photographs. The scanning plane is then noted for easy reproduction. Developments are in hand for providing these data on line from the scanner to a dose calculating computer allowing multiplaner planning of dose distribution. At the moment however plans are drawn using direct measurement from the photographs or using slides of photograph negatives in a desk projection system for life-size display. Furthermore as it is an individual record of delineating the position of the tumour and tissues to be spared from irradiation for that particular patient, this technique is potentially much more accurate than the use of atlases of cross-sectional anatomy. Unlike many X-ray studies these scans are usually performed in the treatment position. Chest Wall. - Chest wall scanning to confirm that the radiotherapy adequately covers the full thickness of the involved region but avoids as much as possible the underlying lung, is useful for planning the sterilisation of post-mastectomy chest walls and the treatment of pleural tumours. In our experience water bath scanning techniques are necessary for accurate results to eliminate the 'opening burst' and a 2.5 MHz frequency transducer has been used. Although with objectives similar to other workers (Jackson et al., 1970) we have used B-scanning in this area marking anatomical features and the treatment fields on the scan by using a special cup fitting over the water bath. The thickness of the chest wall depends on the obesity of the patient, the site on the chest wall, previous surgery and the thickness of the tumour present. Minimum thicknesses, about 2 cm, were found in thin patients near the mid-clavicular line after radical mastectomy. Maximum thickness, about 7 cm, occurred in fat patients towards the axilla when no surgery had been performed. There is thus a wide variation of results for which allowance has to be made when planning radiotherapy with either tangential 3,-ray or direct electron fields in order to exclude as much lung as possible, so reducing the incidence of pulmonary fibrosis whilst adequately irradiating the tumour-bearing area. There is a good interface between the soft tissues and the acoustically impenetrable lung making thickness measurements far easier and quicker than by radiological methods (Fig. 1). The features of carcinoma in the unresected breast recorded by other workers have been noted (Kobayashi, 1973; Ekstrand et al., 1974). The tumours tend to be relatively echo free with loss of
ULTRASOUND
SCANNING
IN T H E R A D I O T H E R A P Y
Carcinoma o.f the breast Fig. 1 - Transverse scan o f the chest wall after m a s t e c t o m y . (Labelling as for Fig. 2 legend.)
Carcinoma of the bladder Fig. 3 - Transverse scan across the pelvis. (See Fig. 6 legend.)
Carcinoma of the prostate Fig. 5 - Longitudinal scan through the pelvis.(See Fig. 6 legend.)
DEPARTMENT
289
Fig. 2 - Transverse scan o f the chest wall through breast containing carcinoma, f, edge o f radiation fields (tangential and parasternal); i, m a s t e c t o m y scar; lu, lung; o, rib; s, skin surface; t, turnout. Fig. 4 - Longitudinal scan through the pelvis. (See Fig. 6 legend./ Fig. 6 - Oblique transverse scan across the pelvis, b, bladder; c, colon; o, pelvic bones (pubis a n d ischium); os, sacrum; p, prostate; r, rectum, s, skin surface; t, tumour.
290
CLINICAL RADIOLOGY
the distal boundary, despite high gain settings, which is believed to be due to high acoustic attenuation of the malignant tumour tissue and helps delineate the extent of the tumour within the normal breast echo pattern (Fig. 2). This technique has helped us determine the volume for irradiation not only in cases of breast carcinoma but also those patients with pleural tumours and sarcomata of the chest wall. Chest wall thickness measurement around the trunk enables more accurate assessment of lung corrections for neutron irradiated fields. By scanning longitudinally 25 mm lateral to the mid-line we feel we are visualising enlargement of the internal mammary nodes in the upper parasternal regions between the ribs. Carefully eliminating artifacts due to reverberation we sometimes see a pyramidal shaped dot pattern. These occur more frequently in those cases where the primary turnout is in the upper inner quadrant of the breast. Although ipsilateral occurrence is often seen they are never seen on the contralateral side unless present also in the ipsilateral side. We hope that some time these nodes seen pre-operatively may be measured and biopsied at the time of mastectomy. Urinary Bladder. - The bladder is scanned after the patient has urinated to assess the situation under the radiotherapy treatment conditions although we realise that this makes it more difficult to obtain scans with good definition. Contact scanning with a 1.5 MHz transducer is used. By measuring three diameters at right angles and using the formula for sphere, an approximate estimation of residual urine volume may be calculated. The aim of the grid of scans is to find the extent of the tumour within the bladder, its penetration through the bladder wall and extension out into the pelvis. The pelvic bones and rectum may also be visualised. With this information we can best check that our irradiated volume is adequate to include all of the tumour-bearing tissue whilst sparing the normal tissues (Figs. 3, 4). Recently in this department, at the end of a conventional course of radiotherapy an extra 1000 rads have been given to the primary tumour alone, on the basis of information from the scan, hoping that tumour control rates may be improved without extra normal tissue damage. Prostate. - With the renewed interest in radiotherapy of primary prostatic carcinoma we have scanned several patients to locate the position and size of the gland for small field, high-dose irradiation. The technique is similar to that for bladder tumours though the planes are closer and concentrated around the symphysis. Transverse planes are also taken with a tilt of about 20 ° into the pelvis (Figs. 5, 6). Although
the diameters are measured accurately for determination of field sizes a rough calculation only on the gland's volume has been made assuming a spherical shape. Pathologically some differentiation can be made. Benign hyperplasia is revealed by an enlarged gland with a persistence of an echo pattern and a regular margin. Neoplasia either primary or by invasion from a bladder tumour is seen as an echo-free gland with an irregular often rather poorly demarcated margin. Retroperitoneal Space. - Again with a 1.5 MHz transducer, contact scanning is performed with the patient lying prone to avoid the relatively impenetrable air-filled gut. Renal turnouts may be measured and located for pre-operative irradiation (Figs. 7, 8). The kidneys may also be scanned to delineate the areas for shielding in the treatment of other malignancies, i.e. carcinoma of the ovary, seminoma and the lymphomata. As they tend to be echo free and displace other organs, tumours in the retroperitoneal space such as lymphomata and pancreatic carcinoma may be localised and measured (Fig. 9). Life size display improves the accuracy of delineating the volume to be irradiated. Furthermore, serial measurements of the lymphomata allow assessment of the effectiveness of treatment either by irradiation or chemotherapy and modification if necessary. Tumours Elsewhere. - A few examples of several other tumours have been scanned (Figs. 10, 11, 12, 13). The thickness and lateral extent of skin turnouts can be measured for adequate radiotherapy (Fig. 14 ). We have scanned the neck for location of tumours in relation to the normal tissues such as the major vessels, thyroid, trachea and spinal cord. Large superficial soft tissue sarcomata have been scanned to assess extension both laterally and in depth allowing more sophisticated treatment techniques to be used (Fig. 15). With longer survivals of breast carcinoma patients on chemotherapy, malignant pericardial effusions are being seen, perhaps as the pericardium is relatively avascular and so inaccessible to cytotoxic agents. Often local radiotherapy has obliterated the anterior space leaving large posterior effusions to account for the clinical findings. We have just started trying to assess the size of spleens and para-aortic nodes which may be relevant in patients with lymphomata and the myeloproliferative disorders, both for their initial assessment and judging their response to treatment (Fig. 16). FUTURE PLANS As our proficiency has increased I hope we can extend our work into tumour kinetics. Easy reassessment of tumour size will allow early and accurate
ULTRASOUND
SCANNING
IN T H E R A D I O T H E R A P Y
Carcinoma o f the kidney Fig. 7 - Transverse scan across the a b d o m e n (patient prone). (See Fig. 9 legend.) Carcinoma o f the pancreas Fig. 9 Transverse scan across the a b d o m e n (patient prone). a, aorta; c, colon; k, kidney; ov, vertebra; s, skin surface, t, turnout; v, inferior vena cava. Fig. 11 Longitudinal scan through the pelvis. (See Fig. 13 legend.) 16
DEPARTMENT
291
Fig. 8 - Longtitudinal scan t h r o u g h the kidney bed (patient prone). (See Fig. 9 legend.) Carcinoma o f the colon - pelvic metastasis Fig. 10 - Transverse scan across the pelvis. (See Fig. 13 legend.) Carcinoma o f the colon - peritoneal metastasis Fig. 12 - Transverse scan across the abdomen. (See Fig. 13 legend.)
292
CLINICAL RADIOLOGY
modification of the planned treatment whether it is by radiotherapy or chemotherapy. Repetition o f the scans in the same planes with the same machine settings allows an easy accurate comparison to be made. Studies have already been made on carcinoma o f the breast, soft tissue sarcomata, plasmocytoma and the l y m p h o m a t a where changes in tumour morphology have been appreciated earlier and more accurately than by palpation or symptomatic response - this is especially true of deep-seated, rather inaccessible tumours. By these means radiation fields have been modified but the capability for rapid replanning has to be increased for this to become a routine procedure. It is hoped that links being
developed with a computerised planning system will make this more practicable. However a difficulty we have experienced already is that, after irradiation, normally echo-free tumours become increasingly full of echoes, so reducing the distinctive characteristics between tumour and normal tissues. Assessment of the response of measurable tumours has allowed earlier modification o f cytotoxic chemotherapy programmes either to permit the use o f a less noxious combination o f drugs or to expose the necessity to find more effective agents. Thus it would seem advisable that the use of ultrasound scanning with its high patient acceptability, should be developed in the planning of
Fig. 13 - Longitudinal scan through the abdomen, b, bladder; o, pelvic bones (pubis and ischium); os, sacrum; s, skin surface; t, tumour; u, umbilicus.
Cutaneous lymphoma Fig. 14 - Oblique scan across the forehead, earn, external auditory meatus; oc, skull bones; t, tumour; wp, widow's peak.
Liposareoma o f the shoulder Fig. 15 - Transverse scan around the shoulder, j, shoulder joint; s, skin surface over the back; t, turnout.
Myeloproliferative disease Fig. 16 - Oblique scan along the long axis of the spleen, s, skin surface; sp, spleen.
ULTRASOUND SCANNING IN THE RADIOTHERAPY DEPARTMENT radiotherapy treatments. Radiotherapy departments should acquire the necessary expertise and personnel to ensure that access to ultrasound scanning systems is easily available both to increase the accuracy of defining the tumour-bearing area for treatment planning and to avoid other methods of investigation involving especially invasive techniques, as these are often time consuming, costly, unpleasant for the patient and occasionally hazardous. Furthermore we would stress the personal involvement of both the radiotherapist in the scanning, as an aid to him in treatment planning, and the physicist, to exploit the capabilities of the scanning machinery to the full. REFERENCES Brascho, D. (1973). Diagnostic ultrasound in radiation treatment planning. Journal of Clinical Ultrasound, 1, 320. Brown, R., Sartin, M. & Bogardns,C. (1972). Patient contours
293
in radiation therapy planning. 7th Annual Meeting of American Institute of Ultrasound in Medicine. Cohen, W. & Hass, C. (1971). Application of B-scan ultrasound in the planning of radiation therapy threatment ports. American Journal of Roentgenology, 111, 184. Ekstrand, IC, Blake, D. & Dixon,'R. (1974). Ultrasonography of the chest wall. Journal o f Clinical Ultrasound, 2, 119. Etde, J., Brochenstedt, F. & Salzman, E. (1973). Diagnostic ultrasound scanning - a valuable aid in radiation therapy planning. American Journal of Radiology, 117, 139. Jackson, S., Naylor, G. & Kerby, I, (1970). Ultrasonic measurement of post-mastectomy chest wall thickness. British Journal of Radiology, 43, 458. Kobayashi, T. (1973). Clinical evaluation of ultrasound techniques in breast tumours and malignant abdominal tumours. 2nd Ultrasonics in Medicine Congress, Rotterdam. Smith, E. & Holm, H. (1970). Ultrasonic scanning in radiotherapy treatment planning. Radiology, 96,433. Wells, P. (Ed.) (1972). Ultrasonics in Clinical Diagnosis. Churchill Livingstone, London.
Successful radiotherapy practice depends on irradiating tumours to a sufficiently high dose to cause their destruction without producing unacceptable damage to the surrounding normal tissues. Especially since the advent of supervoltage equipment, the planning of such treatment has become increasingly sophisticated allowing a better differential between the dose administered to the tumour and that received by the normal tissues. The potential for curing localised malignant disease has thus increased. Unfortunately the techniques for the acquisition of data relating to the patients' anatomy and tumour location have lagged behind progress made in methods of calculating dose distribution. Rapid strides have been made in the proper utilisation of ultrasound pulse-echo systems and B-scanning is now an accepted technique in diagnostic radiology and obstetrics (Wells, 1972). These noninvasive methods are particularly successful in imaging the outline of internal organs, tumours and cysts in cross-sectional planes without recourse to contrast media. Using this technique we have tried to improve the localisation of tumours and radiosensitive organs in the radiotherapy department at the Hammersmith Hospital which treats 1300 new patients each year. We have based our study on the experience of others mainly in departments in the USA (Smith and Holm, 1970; Cohen and Hass, 1971; Eule et al., 1973; Brascho, 1973). So far over 500 patients have been scanned and, after a slow start, we
are studying at a rate of 250 patients per year in a 1 day per week session. Apart from its use by diagnostic radiologists in the detection and staging of tumours, ultrasound scanning appears to offer particular advantages to the radiotherapist in treatment field planning although this application has so far been developed only in a few centres. The technique has many advantages. It is safe at the energy levels used, as on diagnostic machines no harmful effects have been demonstrated in reproducible studies. It is painless and non-invasive; no contrast medium is required, nor is any special preparation of the patient. It is rapid and reduces the requirement for special radiological techniques and isotope scans. Tumour, organ and body outlines are visualised, with the patient usually in the treatment position. Planes including the therapy plane, unable to be visualised by other techniques, may be seen. Although an EMI scanner has superior resolution, scanning in many planes is difficult and the process would be more time consuming and expensive. Because resolution depends on differing acoustic impedances rather than radiodensities, boundaries may be visualised that can not be seen radiologically thus making the technique superior even to transverse axial tomography in some situations. Because of the absence of distortion, accuracy is adequate for measurement and field planning and for verification of precision of the siting of treatment fields. In
288
CLINICAL RADIOLOGY
particular, the study of scans in a number of parallel planes can provide data for planning optimum dose distribution in three dimensions. MoreOver the safety and high patient acceptability of the scanning technique allows repeated reassessment of tumour morphology with consequent early modification of treatment regime. However, there are some disadvantages; being impedance maps of body cross-sections the scans are unfamiliar. A knowledge of physics is required not only for the production of the best scans but also for their interpretation. As in diagnostic radiology and nuclear medicine a pathological diagnosis is usually difficult though solid and cystic lesions are easily distinguished. However, tumours, perhaps because of their basic composition by a single cell type, tend to be relatively echo-free and are also sound attenuating. Some regions are inaccessible as they are shielded by bone or air-filled lung or gut. Poor penetration through thick layers of fat makes the examinations of obese patients more difficult. Resolution in deeper regions may be inadequate to show structures smaller than 3 cm. TECHNIQUE At the Hammersmith Hospital, it has been possible for radiotherapists to obtain direct access to NE4102 Diasonograph, supplied by the DHSS for an instrumental development programme in the Department of Medical Physics, without interference in the routine scanning for radiology and obstetrics. The transducer is either hand-held for contact scanning or submerged in a hand-held water bath. Liquid paraffin is used as the coupling agent as it is less likely to affect the treatment field marks on the skin. Scanning is usually performed in multiple planes 1 - 3 cm apart, in two directions at right angles in order to produce a three-dimensional picture of tumour and normal tissue morphology. Transverse scanning is usually performed first as the various anatomical features are more recognisable. More information can be gained from watching the development of the scan on the screen than by just looking at the final picture as a better appreciation is obtained of these areas of high acoustic attenuation. Surface markers are also included over anatomical landmarks or the position and boundaries of the radiation fields. It is important to produce an accurate body outline, without deforming the shape by undue pressure, and adequate visualisation of the deep structures. Although this can often be achieved in one study sometimes it is necessary to repeat the scans with different settings. It has been shown that it is more accurate to obtain body outlines in this manner than using lead strips or
mechanical jigs and easier than using plaster or plastic templates (Brown et al., 1972). A permanent record of the B mode scan is then made with polaroid photography. Most of our scans have been taken at approximately 2/5 scale - an accurate calibration having been made allowing direct measurement from the photographs. The scanning plane is then noted for easy reproduction. Developments are in hand for providing these data on line from the scanner to a dose calculating computer allowing multiplaner planning of dose distribution. At the moment however plans are drawn using direct measurement from the photographs or using slides of photograph negatives in a desk projection system for life-size display. Furthermore as it is an individual record of delineating the position of the tumour and tissues to be spared from irradiation for that particular patient, this technique is potentially much more accurate than the use of atlases of cross-sectional anatomy. Unlike many X-ray studies these scans are usually performed in the treatment position. Chest Wall. - Chest wall scanning to confirm that the radiotherapy adequately covers the full thickness of the involved region but avoids as much as possible the underlying lung, is useful for planning the sterilisation of post-mastectomy chest walls and the treatment of pleural tumours. In our experience water bath scanning techniques are necessary for accurate results to eliminate the 'opening burst' and a 2.5 MHz frequency transducer has been used. Although with objectives similar to other workers (Jackson et al., 1970) we have used B-scanning in this area marking anatomical features and the treatment fields on the scan by using a special cup fitting over the water bath. The thickness of the chest wall depends on the obesity of the patient, the site on the chest wall, previous surgery and the thickness of the tumour present. Minimum thicknesses, about 2 cm, were found in thin patients near the mid-clavicular line after radical mastectomy. Maximum thickness, about 7 cm, occurred in fat patients towards the axilla when no surgery had been performed. There is thus a wide variation of results for which allowance has to be made when planning radiotherapy with either tangential 3,-ray or direct electron fields in order to exclude as much lung as possible, so reducing the incidence of pulmonary fibrosis whilst adequately irradiating the tumour-bearing area. There is a good interface between the soft tissues and the acoustically impenetrable lung making thickness measurements far easier and quicker than by radiological methods (Fig. 1). The features of carcinoma in the unresected breast recorded by other workers have been noted (Kobayashi, 1973; Ekstrand et al., 1974). The tumours tend to be relatively echo free with loss of
ULTRASOUND
SCANNING
IN T H E R A D I O T H E R A P Y
Carcinoma o.f the breast Fig. 1 - Transverse scan o f the chest wall after m a s t e c t o m y . (Labelling as for Fig. 2 legend.)
Carcinoma of the bladder Fig. 3 - Transverse scan across the pelvis. (See Fig. 6 legend.)
Carcinoma of the prostate Fig. 5 - Longitudinal scan through the pelvis.(See Fig. 6 legend.)
DEPARTMENT
289
Fig. 2 - Transverse scan o f the chest wall through breast containing carcinoma, f, edge o f radiation fields (tangential and parasternal); i, m a s t e c t o m y scar; lu, lung; o, rib; s, skin surface; t, turnout. Fig. 4 - Longitudinal scan through the pelvis. (See Fig. 6 legend./ Fig. 6 - Oblique transverse scan across the pelvis, b, bladder; c, colon; o, pelvic bones (pubis a n d ischium); os, sacrum; p, prostate; r, rectum, s, skin surface; t, tumour.
290
CLINICAL RADIOLOGY
the distal boundary, despite high gain settings, which is believed to be due to high acoustic attenuation of the malignant tumour tissue and helps delineate the extent of the tumour within the normal breast echo pattern (Fig. 2). This technique has helped us determine the volume for irradiation not only in cases of breast carcinoma but also those patients with pleural tumours and sarcomata of the chest wall. Chest wall thickness measurement around the trunk enables more accurate assessment of lung corrections for neutron irradiated fields. By scanning longitudinally 25 mm lateral to the mid-line we feel we are visualising enlargement of the internal mammary nodes in the upper parasternal regions between the ribs. Carefully eliminating artifacts due to reverberation we sometimes see a pyramidal shaped dot pattern. These occur more frequently in those cases where the primary turnout is in the upper inner quadrant of the breast. Although ipsilateral occurrence is often seen they are never seen on the contralateral side unless present also in the ipsilateral side. We hope that some time these nodes seen pre-operatively may be measured and biopsied at the time of mastectomy. Urinary Bladder. - The bladder is scanned after the patient has urinated to assess the situation under the radiotherapy treatment conditions although we realise that this makes it more difficult to obtain scans with good definition. Contact scanning with a 1.5 MHz transducer is used. By measuring three diameters at right angles and using the formula for sphere, an approximate estimation of residual urine volume may be calculated. The aim of the grid of scans is to find the extent of the tumour within the bladder, its penetration through the bladder wall and extension out into the pelvis. The pelvic bones and rectum may also be visualised. With this information we can best check that our irradiated volume is adequate to include all of the tumour-bearing tissue whilst sparing the normal tissues (Figs. 3, 4). Recently in this department, at the end of a conventional course of radiotherapy an extra 1000 rads have been given to the primary tumour alone, on the basis of information from the scan, hoping that tumour control rates may be improved without extra normal tissue damage. Prostate. - With the renewed interest in radiotherapy of primary prostatic carcinoma we have scanned several patients to locate the position and size of the gland for small field, high-dose irradiation. The technique is similar to that for bladder tumours though the planes are closer and concentrated around the symphysis. Transverse planes are also taken with a tilt of about 20 ° into the pelvis (Figs. 5, 6). Although
the diameters are measured accurately for determination of field sizes a rough calculation only on the gland's volume has been made assuming a spherical shape. Pathologically some differentiation can be made. Benign hyperplasia is revealed by an enlarged gland with a persistence of an echo pattern and a regular margin. Neoplasia either primary or by invasion from a bladder tumour is seen as an echo-free gland with an irregular often rather poorly demarcated margin. Retroperitoneal Space. - Again with a 1.5 MHz transducer, contact scanning is performed with the patient lying prone to avoid the relatively impenetrable air-filled gut. Renal turnouts may be measured and located for pre-operative irradiation (Figs. 7, 8). The kidneys may also be scanned to delineate the areas for shielding in the treatment of other malignancies, i.e. carcinoma of the ovary, seminoma and the lymphomata. As they tend to be echo free and displace other organs, tumours in the retroperitoneal space such as lymphomata and pancreatic carcinoma may be localised and measured (Fig. 9). Life size display improves the accuracy of delineating the volume to be irradiated. Furthermore, serial measurements of the lymphomata allow assessment of the effectiveness of treatment either by irradiation or chemotherapy and modification if necessary. Tumours Elsewhere. - A few examples of several other tumours have been scanned (Figs. 10, 11, 12, 13). The thickness and lateral extent of skin turnouts can be measured for adequate radiotherapy (Fig. 14 ). We have scanned the neck for location of tumours in relation to the normal tissues such as the major vessels, thyroid, trachea and spinal cord. Large superficial soft tissue sarcomata have been scanned to assess extension both laterally and in depth allowing more sophisticated treatment techniques to be used (Fig. 15). With longer survivals of breast carcinoma patients on chemotherapy, malignant pericardial effusions are being seen, perhaps as the pericardium is relatively avascular and so inaccessible to cytotoxic agents. Often local radiotherapy has obliterated the anterior space leaving large posterior effusions to account for the clinical findings. We have just started trying to assess the size of spleens and para-aortic nodes which may be relevant in patients with lymphomata and the myeloproliferative disorders, both for their initial assessment and judging their response to treatment (Fig. 16). FUTURE PLANS As our proficiency has increased I hope we can extend our work into tumour kinetics. Easy reassessment of tumour size will allow early and accurate
ULTRASOUND
SCANNING
IN T H E R A D I O T H E R A P Y
Carcinoma o f the kidney Fig. 7 - Transverse scan across the a b d o m e n (patient prone). (See Fig. 9 legend.) Carcinoma o f the pancreas Fig. 9 Transverse scan across the a b d o m e n (patient prone). a, aorta; c, colon; k, kidney; ov, vertebra; s, skin surface, t, turnout; v, inferior vena cava. Fig. 11 Longitudinal scan through the pelvis. (See Fig. 13 legend.) 16
DEPARTMENT
291
Fig. 8 - Longtitudinal scan t h r o u g h the kidney bed (patient prone). (See Fig. 9 legend.) Carcinoma o f the colon - pelvic metastasis Fig. 10 - Transverse scan across the pelvis. (See Fig. 13 legend.) Carcinoma o f the colon - peritoneal metastasis Fig. 12 - Transverse scan across the abdomen. (See Fig. 13 legend.)
292
CLINICAL RADIOLOGY
modification of the planned treatment whether it is by radiotherapy or chemotherapy. Repetition o f the scans in the same planes with the same machine settings allows an easy accurate comparison to be made. Studies have already been made on carcinoma o f the breast, soft tissue sarcomata, plasmocytoma and the l y m p h o m a t a where changes in tumour morphology have been appreciated earlier and more accurately than by palpation or symptomatic response - this is especially true of deep-seated, rather inaccessible tumours. By these means radiation fields have been modified but the capability for rapid replanning has to be increased for this to become a routine procedure. It is hoped that links being
developed with a computerised planning system will make this more practicable. However a difficulty we have experienced already is that, after irradiation, normally echo-free tumours become increasingly full of echoes, so reducing the distinctive characteristics between tumour and normal tissues. Assessment of the response of measurable tumours has allowed earlier modification o f cytotoxic chemotherapy programmes either to permit the use o f a less noxious combination o f drugs or to expose the necessity to find more effective agents. Thus it would seem advisable that the use of ultrasound scanning with its high patient acceptability, should be developed in the planning of
Fig. 13 - Longitudinal scan through the abdomen, b, bladder; o, pelvic bones (pubis and ischium); os, sacrum; s, skin surface; t, tumour; u, umbilicus.
Cutaneous lymphoma Fig. 14 - Oblique scan across the forehead, earn, external auditory meatus; oc, skull bones; t, tumour; wp, widow's peak.
Liposareoma o f the shoulder Fig. 15 - Transverse scan around the shoulder, j, shoulder joint; s, skin surface over the back; t, turnout.
Myeloproliferative disease Fig. 16 - Oblique scan along the long axis of the spleen, s, skin surface; sp, spleen.
ULTRASOUND SCANNING IN THE RADIOTHERAPY DEPARTMENT radiotherapy treatments. Radiotherapy departments should acquire the necessary expertise and personnel to ensure that access to ultrasound scanning systems is easily available both to increase the accuracy of defining the tumour-bearing area for treatment planning and to avoid other methods of investigation involving especially invasive techniques, as these are often time consuming, costly, unpleasant for the patient and occasionally hazardous. Furthermore we would stress the personal involvement of both the radiotherapist in the scanning, as an aid to him in treatment planning, and the physicist, to exploit the capabilities of the scanning machinery to the full. REFERENCES Brascho, D. (1973). Diagnostic ultrasound in radiation treatment planning. Journal of Clinical Ultrasound, 1, 320. Brown, R., Sartin, M. & Bogardns,C. (1972). Patient contours
293
in radiation therapy planning. 7th Annual Meeting of American Institute of Ultrasound in Medicine. Cohen, W. & Hass, C. (1971). Application of B-scan ultrasound in the planning of radiation therapy threatment ports. American Journal of Roentgenology, 111, 184. Ekstrand, IC, Blake, D. & Dixon,'R. (1974). Ultrasonography of the chest wall. Journal o f Clinical Ultrasound, 2, 119. Etde, J., Brochenstedt, F. & Salzman, E. (1973). Diagnostic ultrasound scanning - a valuable aid in radiation therapy planning. American Journal of Radiology, 117, 139. Jackson, S., Naylor, G. & Kerby, I, (1970). Ultrasonic measurement of post-mastectomy chest wall thickness. British Journal of Radiology, 43, 458. Kobayashi, T. (1973). Clinical evaluation of ultrasound techniques in breast tumours and malignant abdominal tumours. 2nd Ultrasonics in Medicine Congress, Rotterdam. Smith, E. & Holm, H. (1970). Ultrasonic scanning in radiotherapy treatment planning. Radiology, 96,433. Wells, P. (Ed.) (1972). Ultrasonics in Clinical Diagnosis. Churchill Livingstone, London.