Bmuh Mtdical BulUtix (1991) Vol. 47, No 2, pp. 295-3OJ © The Brituh Council 1991
Diagnosis by team work: An approach to conservatism M Galea R W Blarney Professorial Department of Surgery, Nottingham City Hospital, Nottingham, UK
In the Nottingham Breast Clinic 5000 new cases are seen annually; 3000 are sent because the general practitioner believes that he/she has found a lump. It is the diagnosis and management of this common problem that is considered in this chapter. The traditional management of the palpable breast lump included excision to establish accurate diagnosis; indeed up until 20 years ago even simple cysts were excised. Now most surgeons accept that a woman with no residual palpable abnormality after aspiration of blood-free fluid from a cyst requires no further treatment. Most solid breast lumps are benign on histological examination; their routine removal a mistake in judgement. The challenge in the management of a palpable lesion is to correctly diagnose all the cancers without having to remove those which are benign, unless the patient wishes.
Until 1987 our management of the symptomatic patient consisted firstly of careful palpation of the breast to decide whether or not a lump was present. Once the examiner had decided that a lump was palpable, he carried out aspiration to determine whether it was cystic, and if so he aspirated it completely; the area was then re-palpated and as long as the lump had completely resolved no further investigation was carried out. If the lump was solid, 'Trucut' needle biopsy was performed followed always by operative excision of the lump. The findings on 'Trucut' biopsy were regarded as valid only if positive, allowing a preoperative diagnosis of cancer. If biopsy did not reveal a cancer the lump was still
296
BREAST DISEASE
removed and sent for histology, this based on our finding that 'Trucut' biopsy was only 76% sensitive in the diagnosis of palpable breast cancers (Table 1). A summary of this management scheme is given in Figure 1. Several observations may be made with respect to this scheme. We still regard palpation as the first and single most important investigative step in the symptomatic woman. In our clinic it is
Fig. 1 The Nottingham Breast Clinic management scheme for the palpable breast lump prior to 1987.
THE BRITISH COUNCIL 1991
297
Table 1 Results of Trucut biopsy in 932 palpable breast cancers Trucut diagnosis Carcinoma Suspicious Benign
n = 932
Percentage
708 44 180
76 5 19
Thu gives • sensitivity for an absolute diagnosis of cancer of 76%
the initial palpation which determines whether further investigation should be carried out. 'Trucut' needle biopsy still has the advantage of providing a core of breast tissue for conventional histological techniques, which are widely available. But the sampling of benign lumps by 'Trucut' biopsy often proves inadequate, because of deflection of the needle by the inherent rubbery-type toughness of benign tissue. 'Trucut' biopsy cannot safely diagnose the benign nature of the small breast lump. CYTOLOGY The increasing interest in the use offineneedle aspiration cytology (FNAC) for the diagnosis of breast lesions led us to introduce this diagnostic method alongside 'Trucut' biopsy over the period 1986-87. FNAC is obtained without the use of local anaesthetic; it is rapidly complete and although sometimes painful is no more so than the injection of local anaesthetic. It gives a higher true positive rate for the diagnosis of cancer than does 'Trucut' biopsy—87% sensitivity vs. 76% (Table 2). FNAC does suffer the disadvantage of very occasionally giving a false positive result (the best readers of breast cytology quote a rate of approximately 1 per 1000 cases), a problem usually associated with interpretation of samples from fibroadenomas (Fig. 2). Cytology therefore requires great expertise but even then the cytopathologist can be confident in his assessment only if the preparation contains an adequate number of epithelial cells. In some Scandinavian and American Table 2 Results of FNAC in 213 palpable breast cancers FNAC diagnosis Carcinoma Suspicious Equivocal Benign
n = 213
Percentage
185 17 3 8
87 8 1 4
A 95% sensitivity is obtmmed for 'carcinoma' or 'suspicious of carcinoma'.
298
BREAST DISEASE
Fig. 2 Cytology from a clinically and radiologically benign lump: although showing some features consistent with a fibroadenoma many cells are large with nuclear atypia.
centres the diagnostic cytologist takes the FNAC specimens. This is not the practice in this country where most of the specimens are taken and smears made on slides by the attending surgeon. This requires training and the experience of regular use. Good technique and close communication between surgeon and cytopathologist are essential at all times. For the technique used in Nottingham a 10 ml disposable syringe and 23 gauge needle are used. The lump to be aspirated is fixed between two fingers and the needle advanced through the skin into it. Suction is then applied. The needle is passed to and fro through the lump working round it in a clockface manner so that 10-12 passes are made; after each pass the needle is drawn out of the lump but not through the skin. This ensures that the tumour is not sheered which might cause the sample to be obscured by blood. Suction is then released and the needle withdrawn through the skin. The needle contents are then expressed
THE BRITISH COUNCIL 1991
299
onto two slides and smears made. These are air dried but some pathologists prefer to 'quickfix' one slide. Further material is obtained by rinsing the needle and syringe with suitable fixative and sending the specimen for centrifugation. The exact staining procedure depends upon the pathologist's preference; in Nottingham the May-Grunwald-Giemsa stain is used. The cytological classification used in Nottingham is given in Table 3. The findings are interpreted along with the surgeon's criteria on which FNAC has been performed (Table 4). The two categories are matched to determine how the patient is managed.
Table 3 Nottingham Cytology Classification Inadequate:
a sparsely cellular smear, or a poorly preserved or prepared smear
Benign:
an adequate sample containing benign epithelial cells
Equivocal:
an aspirate with an overall benign pattern in which mild cellular atypia is seen. This atypia is usually nuclear
Suspicious:
cellular features not absolutely diagnostic of malignancy. Applies to: (i) a scanty, poorly preserved or poorly prepared sample with some features of malignancy in some cells; (ii) a satisfactory smear with some, but no overt, features of malignancy; and (iii) a sample with an overall benign pattern with some cells showing distinct malignant features
Carcinoma:
cells present which are unequivocally malignant
Table 4 Clinical categories for correlation with FNA findings Category A
Prcoperative diagnosis of cancer. The lump has clinical and/or imaging features which are not entirely benign. A classification of 'malignant' (Table 2) allows definitive operation for cancer. Even if cytology is 'benign' the lump will be removed for diagnostic histology.
Category B Preoperative diagnosis of breast lesion uncertain. Cytological classification of 'benign' epithelial cells is required. If 'inadequate' the lump will be removed. Category C Preoperative diagnosis of benign disease. FNA to ensure that no suspicious cells are found from a lump that would not be removed even if cytology 'inadequate'. Therefore, further investigation or treatment only if cytology is 'malignant' or 'suspicious': for example, apparent fibroadenoma below the age of 35 years. Category D Preoperative diagnosis difficult to ascertain because of an unusual situation. The cytologist should be alerted to this—e.g. previous cancer treated with irradiation, pregnancy.
300
BREAST DISEASE
IMAGING Diagnostic imaging has improved as a result of better equipment and film quality but it is the emergence of dedicated mammographic radiologists and radiographers, whose expertise encompasses diagnostic ultrasound and localization techniques, that has been the most important step forward in this modality. The imaging classification of mammograms and ultrasound used in Nottingham is: Benign Probably benign Indeterminate Probably malignant Malignant Imaging now has a clear place in our decision-making, influencing the management of breast cancer. For example in determining suitability for treatment with breast conservation, a pre-operative mammogram is essential to ensure that the cancer is neither larger than clinically appreciated nor diffuse or multicentric. However, we believe that the place of imaging in the diagnosis of the symptomatic patient is often over emphasized. In a review of mammography in symptomatic breast disease, we found it to be rarely essential to the diagnosis of cancer. Thus, in 5080 symptomatic patients the diagnosis of 562 cancers would have been revealed, on clinical grounds and 'Trucut' biopsy, in all but 11. In 3 of these 11 patients the cancer was in the opposite breast to the one of complaint.1
THE TEAM APPROACH In the breast clinic in Nottingham City Hospital 100 new patients with breast symptoms attend each week. Under these circumstances the passing of typed reports between staff of different disciplines is both ineffective and inefficient especially where uncertainty in interpretation exists. We now hold an 'interdisciplinary clinic' alongside the main breast clinic in which surgeon, radiologist and pathologist assess all women presenting with a palpable lump at the time of attendance. This clinic allows women to be assessed fully at one visit. Those without cancer can be reassured at this first attendance, saving weeks of anxiety; needle aspiration is carried out only after imaging investigations; the radiologist involved in the clinic becomes
THE BRITISH COUNCIL 1991
301
specialised in breast disease and technical expertise is gained by non-medical staff in mammography including specialised views, ultrasound and preparation for cytology. The management pathway for the assessment of women with a symptomatic breast lump as now practised in this multidisciplinary clinic is shown in Figure 3. A non-operative policy has been
Symptomatic clinic
palpable
IoterdiadplioarT AMcaanient CHmc : MTTir day
(Category C)
(Category B)
(Category A)
Fig. 3 Present day management of the palpable breast lump in the Nottingham Multidisciplinary Assessment clinic.
302
BREAST DISEASE
Table S Criteria to be satisfied for leaving a palpable lump in sttu 1. Clinically the lump feels entirely benign 2. Dependant on age: < 35 1 x FNAC cytology with no suspicious cells 35-50 Ultrasound benign 2 x FNAC cytology (6 weeks apart): benign epithelial cells > 50 Ultrasound benign Mammogram benign 2 x FNAC cytology (6 weeks apart): benign epithelial cells
extended to women of all ages who fulfil the stria criteria shown in Table 5. If any of these criteria remains unsatisfied the lump is excised for diagnosis. The rationale for adopting these criteria are: (1) Breast cancer is rare under the age of 35 years. In this age group a breast lump believed to be benign on clinical and cytological grounds can be watched safely.2 (2) Implementation of the National Breast Cancer Screening Programme in the UK has shown many older women to have asymptomatic breast abnormalities. Excision of all of these would not only throw the screening programme into disrepute but cause considerable unnecessary morbidity. An anomalous situation has arisen in which women with a palpable lump which clinically, mammographically and cytologically is benign may still be advised to have it removed, whereas a woman with a benign-appearing lesion that cannot be felt or checked by cytology is advised to keep it in situ. The adoption of this policy has dramatically reduced the number of operations for lumps proving to be benign on cytology. This can be illustrated by considering the group required to fulfil the most stringent criteria—i.e. women over 35 years of age. In the multidisciplinary clinic over a 1 year period (August 1989-90), we have assessed 200 such women, each with a discrete benign-feeling lump. Of these: 92 had simple cysts that resolved on aspiration; 80 women fulfilled our criteria for non-operative management, of whom only 9 chose to have an operation (histology: all benign); and 28 women failed to fulfil the criteria and were advised operative excision. The indication for operation in the 28 women failing to fulfil the criteria and the specimen histology are given in Table 6. By adhering to this protocol only 13 women could be considered to have had an unnecessary operation. The pivotal role of FNAC in
THE BRITISH COUNCIL 1991
303
Table 6 Reasons for failure to fulfill criteria for non-operative management of a clinically benign lump in 28 women > 35 years and histology of the resected specimen Histology Total Imaging: US FNAC: FNAC and Imaging: Size alone: >3cm
2 22 1 3 28
Benign
Carcinoma
2 8 0 3 13
0 14 1 0 15
assessment is clear, emphasizing the necessity for a competent aspirator and experienced cytopathologist. Although these data do not provide information on possible false negative assessments, to date no women with a breast lump left in situ has returned with breast cancer. We believe that this policy of non-operative management of a discrete breast lump after multidisciplinary assessment is well tolerated, avoids hospital admission and appears safe. In our experience less than 10% request to be rid of their lump. At present, follow-up at 1 year is recommended to reassess the few lumps that continue to enlarge. Excision for histological diagnosis of lumps > 3 cm is then advised. The majority of lumps managed in this way are fibroadenomas. A further benefit from triple assessment is that many benign 'lumpy' areas, where the clinician was initially undecided as to whether a lump was present or not, prove to be areas of benign fibrocystic change rather than a true lump. Many of these women would have had a biopsy in the past. Now that the problem of diagnosis can be resolved by combined assessment, to operate on such 'pseudo-lumps' is unusual. To summarize, our diagnostic policy accurately and safely aims to define benign from malignant so that the excision biopsy of benign lumps becomes infrequent. In the setting of a multidisciplinary clinic with breast specialists, our protocol for non-operative management of benign-feeling lumps appears safe and is well tolerated. CONCLUSION There is no doubt in our minds that the team assessment of breast symptoms has great advantages:
304
BREAST DISEASE
Many women without cancer are diagnosed 'benign' and reassured at their first visit, keeping anxiety to a minimum; Discussion of difficult cases in a multidisciplinary fashion avoids protracted deliberation; The opportunity exists for excellent training in the management of breast disease for medical and non-medical staff; Prompt diagnosis and treatment for cancers. The non-operative management of 'benign' lumps has been safe following our strict criteria. There are still a number of patients that require or wish excision of lumps which prove to be histologically benign—perhaps some of our criteria will be relaxed in the future. Our ultimate goal is to leave all benign lesions in situ whilst achieving 100% specificity in selecting the breast cancers from amongst them. Experience, combined expertise and the techniques we already have, make this a feasible proposition. REFERENCES 1 Locker AP, Manhire AR, Strickland V, Caseldine J, Blarney RW. Mammography in Symptomatic Breast Disease. Lancet 1989; i: 887-89 2 Wilkinson S, Anderson TJ, Rifkind E, Chetty U, Forrest APM, Fibroadenoma of the breast: a follow-up of conservative management. Br J Surg 1989; 76: 390-1
Diagnosis by team work: An approach to conservatism M Galea R W Blarney Professorial Department of Surgery, Nottingham City Hospital, Nottingham, UK
In the Nottingham Breast Clinic 5000 new cases are seen annually; 3000 are sent because the general practitioner believes that he/she has found a lump. It is the diagnosis and management of this common problem that is considered in this chapter. The traditional management of the palpable breast lump included excision to establish accurate diagnosis; indeed up until 20 years ago even simple cysts were excised. Now most surgeons accept that a woman with no residual palpable abnormality after aspiration of blood-free fluid from a cyst requires no further treatment. Most solid breast lumps are benign on histological examination; their routine removal a mistake in judgement. The challenge in the management of a palpable lesion is to correctly diagnose all the cancers without having to remove those which are benign, unless the patient wishes.
Until 1987 our management of the symptomatic patient consisted firstly of careful palpation of the breast to decide whether or not a lump was present. Once the examiner had decided that a lump was palpable, he carried out aspiration to determine whether it was cystic, and if so he aspirated it completely; the area was then re-palpated and as long as the lump had completely resolved no further investigation was carried out. If the lump was solid, 'Trucut' needle biopsy was performed followed always by operative excision of the lump. The findings on 'Trucut' biopsy were regarded as valid only if positive, allowing a preoperative diagnosis of cancer. If biopsy did not reveal a cancer the lump was still
296
BREAST DISEASE
removed and sent for histology, this based on our finding that 'Trucut' biopsy was only 76% sensitive in the diagnosis of palpable breast cancers (Table 1). A summary of this management scheme is given in Figure 1. Several observations may be made with respect to this scheme. We still regard palpation as the first and single most important investigative step in the symptomatic woman. In our clinic it is
Fig. 1 The Nottingham Breast Clinic management scheme for the palpable breast lump prior to 1987.
THE BRITISH COUNCIL 1991
297
Table 1 Results of Trucut biopsy in 932 palpable breast cancers Trucut diagnosis Carcinoma Suspicious Benign
n = 932
Percentage
708 44 180
76 5 19
Thu gives • sensitivity for an absolute diagnosis of cancer of 76%
the initial palpation which determines whether further investigation should be carried out. 'Trucut' needle biopsy still has the advantage of providing a core of breast tissue for conventional histological techniques, which are widely available. But the sampling of benign lumps by 'Trucut' biopsy often proves inadequate, because of deflection of the needle by the inherent rubbery-type toughness of benign tissue. 'Trucut' biopsy cannot safely diagnose the benign nature of the small breast lump. CYTOLOGY The increasing interest in the use offineneedle aspiration cytology (FNAC) for the diagnosis of breast lesions led us to introduce this diagnostic method alongside 'Trucut' biopsy over the period 1986-87. FNAC is obtained without the use of local anaesthetic; it is rapidly complete and although sometimes painful is no more so than the injection of local anaesthetic. It gives a higher true positive rate for the diagnosis of cancer than does 'Trucut' biopsy—87% sensitivity vs. 76% (Table 2). FNAC does suffer the disadvantage of very occasionally giving a false positive result (the best readers of breast cytology quote a rate of approximately 1 per 1000 cases), a problem usually associated with interpretation of samples from fibroadenomas (Fig. 2). Cytology therefore requires great expertise but even then the cytopathologist can be confident in his assessment only if the preparation contains an adequate number of epithelial cells. In some Scandinavian and American Table 2 Results of FNAC in 213 palpable breast cancers FNAC diagnosis Carcinoma Suspicious Equivocal Benign
n = 213
Percentage
185 17 3 8
87 8 1 4
A 95% sensitivity is obtmmed for 'carcinoma' or 'suspicious of carcinoma'.
298
BREAST DISEASE
Fig. 2 Cytology from a clinically and radiologically benign lump: although showing some features consistent with a fibroadenoma many cells are large with nuclear atypia.
centres the diagnostic cytologist takes the FNAC specimens. This is not the practice in this country where most of the specimens are taken and smears made on slides by the attending surgeon. This requires training and the experience of regular use. Good technique and close communication between surgeon and cytopathologist are essential at all times. For the technique used in Nottingham a 10 ml disposable syringe and 23 gauge needle are used. The lump to be aspirated is fixed between two fingers and the needle advanced through the skin into it. Suction is then applied. The needle is passed to and fro through the lump working round it in a clockface manner so that 10-12 passes are made; after each pass the needle is drawn out of the lump but not through the skin. This ensures that the tumour is not sheered which might cause the sample to be obscured by blood. Suction is then released and the needle withdrawn through the skin. The needle contents are then expressed
THE BRITISH COUNCIL 1991
299
onto two slides and smears made. These are air dried but some pathologists prefer to 'quickfix' one slide. Further material is obtained by rinsing the needle and syringe with suitable fixative and sending the specimen for centrifugation. The exact staining procedure depends upon the pathologist's preference; in Nottingham the May-Grunwald-Giemsa stain is used. The cytological classification used in Nottingham is given in Table 3. The findings are interpreted along with the surgeon's criteria on which FNAC has been performed (Table 4). The two categories are matched to determine how the patient is managed.
Table 3 Nottingham Cytology Classification Inadequate:
a sparsely cellular smear, or a poorly preserved or prepared smear
Benign:
an adequate sample containing benign epithelial cells
Equivocal:
an aspirate with an overall benign pattern in which mild cellular atypia is seen. This atypia is usually nuclear
Suspicious:
cellular features not absolutely diagnostic of malignancy. Applies to: (i) a scanty, poorly preserved or poorly prepared sample with some features of malignancy in some cells; (ii) a satisfactory smear with some, but no overt, features of malignancy; and (iii) a sample with an overall benign pattern with some cells showing distinct malignant features
Carcinoma:
cells present which are unequivocally malignant
Table 4 Clinical categories for correlation with FNA findings Category A
Prcoperative diagnosis of cancer. The lump has clinical and/or imaging features which are not entirely benign. A classification of 'malignant' (Table 2) allows definitive operation for cancer. Even if cytology is 'benign' the lump will be removed for diagnostic histology.
Category B Preoperative diagnosis of breast lesion uncertain. Cytological classification of 'benign' epithelial cells is required. If 'inadequate' the lump will be removed. Category C Preoperative diagnosis of benign disease. FNA to ensure that no suspicious cells are found from a lump that would not be removed even if cytology 'inadequate'. Therefore, further investigation or treatment only if cytology is 'malignant' or 'suspicious': for example, apparent fibroadenoma below the age of 35 years. Category D Preoperative diagnosis difficult to ascertain because of an unusual situation. The cytologist should be alerted to this—e.g. previous cancer treated with irradiation, pregnancy.
300
BREAST DISEASE
IMAGING Diagnostic imaging has improved as a result of better equipment and film quality but it is the emergence of dedicated mammographic radiologists and radiographers, whose expertise encompasses diagnostic ultrasound and localization techniques, that has been the most important step forward in this modality. The imaging classification of mammograms and ultrasound used in Nottingham is: Benign Probably benign Indeterminate Probably malignant Malignant Imaging now has a clear place in our decision-making, influencing the management of breast cancer. For example in determining suitability for treatment with breast conservation, a pre-operative mammogram is essential to ensure that the cancer is neither larger than clinically appreciated nor diffuse or multicentric. However, we believe that the place of imaging in the diagnosis of the symptomatic patient is often over emphasized. In a review of mammography in symptomatic breast disease, we found it to be rarely essential to the diagnosis of cancer. Thus, in 5080 symptomatic patients the diagnosis of 562 cancers would have been revealed, on clinical grounds and 'Trucut' biopsy, in all but 11. In 3 of these 11 patients the cancer was in the opposite breast to the one of complaint.1
THE TEAM APPROACH In the breast clinic in Nottingham City Hospital 100 new patients with breast symptoms attend each week. Under these circumstances the passing of typed reports between staff of different disciplines is both ineffective and inefficient especially where uncertainty in interpretation exists. We now hold an 'interdisciplinary clinic' alongside the main breast clinic in which surgeon, radiologist and pathologist assess all women presenting with a palpable lump at the time of attendance. This clinic allows women to be assessed fully at one visit. Those without cancer can be reassured at this first attendance, saving weeks of anxiety; needle aspiration is carried out only after imaging investigations; the radiologist involved in the clinic becomes
THE BRITISH COUNCIL 1991
301
specialised in breast disease and technical expertise is gained by non-medical staff in mammography including specialised views, ultrasound and preparation for cytology. The management pathway for the assessment of women with a symptomatic breast lump as now practised in this multidisciplinary clinic is shown in Figure 3. A non-operative policy has been
Symptomatic clinic
palpable
IoterdiadplioarT AMcaanient CHmc : MTTir day
(Category C)
(Category B)
(Category A)
Fig. 3 Present day management of the palpable breast lump in the Nottingham Multidisciplinary Assessment clinic.
302
BREAST DISEASE
Table S Criteria to be satisfied for leaving a palpable lump in sttu 1. Clinically the lump feels entirely benign 2. Dependant on age: < 35 1 x FNAC cytology with no suspicious cells 35-50 Ultrasound benign 2 x FNAC cytology (6 weeks apart): benign epithelial cells > 50 Ultrasound benign Mammogram benign 2 x FNAC cytology (6 weeks apart): benign epithelial cells
extended to women of all ages who fulfil the stria criteria shown in Table 5. If any of these criteria remains unsatisfied the lump is excised for diagnosis. The rationale for adopting these criteria are: (1) Breast cancer is rare under the age of 35 years. In this age group a breast lump believed to be benign on clinical and cytological grounds can be watched safely.2 (2) Implementation of the National Breast Cancer Screening Programme in the UK has shown many older women to have asymptomatic breast abnormalities. Excision of all of these would not only throw the screening programme into disrepute but cause considerable unnecessary morbidity. An anomalous situation has arisen in which women with a palpable lump which clinically, mammographically and cytologically is benign may still be advised to have it removed, whereas a woman with a benign-appearing lesion that cannot be felt or checked by cytology is advised to keep it in situ. The adoption of this policy has dramatically reduced the number of operations for lumps proving to be benign on cytology. This can be illustrated by considering the group required to fulfil the most stringent criteria—i.e. women over 35 years of age. In the multidisciplinary clinic over a 1 year period (August 1989-90), we have assessed 200 such women, each with a discrete benign-feeling lump. Of these: 92 had simple cysts that resolved on aspiration; 80 women fulfilled our criteria for non-operative management, of whom only 9 chose to have an operation (histology: all benign); and 28 women failed to fulfil the criteria and were advised operative excision. The indication for operation in the 28 women failing to fulfil the criteria and the specimen histology are given in Table 6. By adhering to this protocol only 13 women could be considered to have had an unnecessary operation. The pivotal role of FNAC in
THE BRITISH COUNCIL 1991
303
Table 6 Reasons for failure to fulfill criteria for non-operative management of a clinically benign lump in 28 women > 35 years and histology of the resected specimen Histology Total Imaging: US FNAC: FNAC and Imaging: Size alone: >3cm
2 22 1 3 28
Benign
Carcinoma
2 8 0 3 13
0 14 1 0 15
assessment is clear, emphasizing the necessity for a competent aspirator and experienced cytopathologist. Although these data do not provide information on possible false negative assessments, to date no women with a breast lump left in situ has returned with breast cancer. We believe that this policy of non-operative management of a discrete breast lump after multidisciplinary assessment is well tolerated, avoids hospital admission and appears safe. In our experience less than 10% request to be rid of their lump. At present, follow-up at 1 year is recommended to reassess the few lumps that continue to enlarge. Excision for histological diagnosis of lumps > 3 cm is then advised. The majority of lumps managed in this way are fibroadenomas. A further benefit from triple assessment is that many benign 'lumpy' areas, where the clinician was initially undecided as to whether a lump was present or not, prove to be areas of benign fibrocystic change rather than a true lump. Many of these women would have had a biopsy in the past. Now that the problem of diagnosis can be resolved by combined assessment, to operate on such 'pseudo-lumps' is unusual. To summarize, our diagnostic policy accurately and safely aims to define benign from malignant so that the excision biopsy of benign lumps becomes infrequent. In the setting of a multidisciplinary clinic with breast specialists, our protocol for non-operative management of benign-feeling lumps appears safe and is well tolerated. CONCLUSION There is no doubt in our minds that the team assessment of breast symptoms has great advantages:
304
BREAST DISEASE
Many women without cancer are diagnosed 'benign' and reassured at their first visit, keeping anxiety to a minimum; Discussion of difficult cases in a multidisciplinary fashion avoids protracted deliberation; The opportunity exists for excellent training in the management of breast disease for medical and non-medical staff; Prompt diagnosis and treatment for cancers. The non-operative management of 'benign' lumps has been safe following our strict criteria. There are still a number of patients that require or wish excision of lumps which prove to be histologically benign—perhaps some of our criteria will be relaxed in the future. Our ultimate goal is to leave all benign lesions in situ whilst achieving 100% specificity in selecting the breast cancers from amongst them. Experience, combined expertise and the techniques we already have, make this a feasible proposition. REFERENCES 1 Locker AP, Manhire AR, Strickland V, Caseldine J, Blarney RW. Mammography in Symptomatic Breast Disease. Lancet 1989; i: 887-89 2 Wilkinson S, Anderson TJ, Rifkind E, Chetty U, Forrest APM, Fibroadenoma of the breast: a follow-up of conservative management. Br J Surg 1989; 76: 390-1
