Statistics from Altmetric.com
Shannon and colleagues1 conclude that conversion to core biopsy increases sensitivity and specificity and reduces the inadequate rate, and advocate the abandonment of fine needle aspiration (FNA) in the preoperative diagnosis of breast disease.
In their experience, the inadequate rate for FNA in the symptomatic breast disease context of 35% is certainly untenable, but it would be appropriate to evaluate the application of the technique rather than the technique itself. The authors provide no data on the structure of their FNA service: were the aspirates performed by a limited number of trained and experienced aspirators? What indications were considered appropriate for FNA? Was there rapid assessment of FNA with repeat of inadequate samples?
An additional factor may be the technique used: 21 gauge needles are very large bore for symptomatic breast FNA, and better results can be obtained with smaller gauge needles (23 gauge),2 whereby discomfort is reduced, blood contamination is reduced, and cell yield is satisfactory.
The structure of a service where the inadequate rate of FNA is as high as 35% must be seriously questioned. In our experience in symptomatic breast disease practice, the use of FNA within a triple assessment context complemented by core biopsy in situations of significant discordance between modalities is a robust and accurate diagnostic modality.
By limiting the technique to a small number of skilled aspirators (in our case two cytopathologists) and incorporating rapid microscopic evaluation of adequacy at a triple assessment clinic, the inadequate rate is low (< 15%) and other performance criteria are acceptable (complete sensitivity, 88%; full specificity, 80%; false negative rate, 5%; false positive rate, 0%: parameters calculated using methodology detailed in National Health Service breast screening programme guidelines for cytology). The addition of core biopsy in situations where triple assessment is discordant allows accurate preoperative diagnosis in virtually all cases.
Absolute and complete sensitivity with FNA will never match that of core biopsy, but that is not always the point: most patients attending symptomatic breast disease clinics do not have malignant disease, and a benign diagnosis on triple assessment including FNA allows a definitive diagnosis to be made at a single outpatient visit, alleviates patient anxiety, and offers considerable cost effectiveness in terms of outpatient clinic time and resources. The requirement for core biopsy is reduced by the use of FNA, particularly in the context of benign disease. Clear protocols for further investigation of situations of discordant triple assessment (including core biopsy) safeguard against false negative diagnosis and missed cancer.
Core biopsy is certainly a robust and reliable diagnostic modality, but it has some important disadvantages in terms of delay in diagnosis and patient discomfort (I would dispute that core biopsy is less traumatic than FNA1 when FNA is performed using a 23 gauge needle).
When used in the appropriate setting FNA has an important role in preoperative diagnosis of symptomatic breast disease and should not be abandoned without due consideration to its appropriate use and the structure of the service within which it is applied.
We are grateful to Dr Jeffers for his response. Our paper1 and his letter raise important issues regarding the future viability of a cytology led service for the preoperative assessment of breast disease. In our centre, fine needle aspiration cytology (FNAC) and core biopsies were performed in the main by either consultant radiologists with a specialist interest in breast disease (where radiological assistance is required) or specialist breast surgeons. The investigations were not performed within the setting of a single attendance (“one stop”) clinic.
We accept that FNAC supplemented by core biopsy in a rapid diagnosis setting may be considered by clinicians and patients alike to be a satisfactory service, but we would argue that this is the only potential benefit FNAC has to offer. To be a viable proposition this assumes the regular availability of an experienced cytopathologist seeing many aspirates on a regular basis, together with appropriate cross cover for annual leave/sickness. Dr Jeffers indicates that this is so in his department, but we suggest that this may not be the case in many centres and would be costly to set up.
The single attendance (one stop) clinic can be advertised as an attractive and a popular innovation, but in this climate of evidence based medicine the demonstration of advantages in patient management and cost benefit analysis are required.2 A recently published analysis of one stop breast clinics directly dealing with these issues has shown a reduction in patient anxiety (which was only detectable over the first 24 hours but not at three weeks or three months).3 For the first time, this study analysed costs and showed that savings from the reduction of frequency of outpatient visits were more than offset by the increased costs associated with same day diagnosis (£32 for each patient). The authors conclude that this additional cost to the National Health Service may not be justified by the short term reduction in anxiety.
Dr Jeffers argues for a role for FNAC in benign breast disease to provide immediate reassurance. Perhaps it is time to consider whether simply calling a lesion benign or malignant (neither of which constitutes a diagnosis) is enough. We would argue that core biopsies are more likely to provide a definite benign diagnosis. We show reduced suspicious rates rather than having to accept C1 as an absence of malignant cells and to struggle with the cases given a C3 diagnosis, which would probably proceed to core biopsy any way. Surgical colleagues have shown that automated core biopsy has a superior diagnostic power compared with FNAC in breast cancer diagnosis.4
Other arguments for favouring core biopsy over FNAC for palpable lesions are well rehearsed (assessment of invasion, tumour type, grade, and oestrogen receptor status are more amenable by the former). Mammographically detected lesions in the context of screening require a different approach because there are radiological abnormalities that require pathological explanation (such as microcalcification, distortion, asymmetric density, etc) and that often cannot be investigated satisfactorily by means of FNAC. In addition, we suggest that core biopsies are more readily and accurately interpreted by pathologists who are considered “non-specialist” compared with needle aspirates, resulting in less reliance on the presence of a small group of individuals for a clinic to run smoothly, and fewer difficulties in the provision of cover for staff absence.