Litigation in breast surgery

Published on 10/04/2015 by admin

Filed under Surgery

Last modified 22/04/2025

Print this page

This article have been viewed 1584 times

Litigation in breast surgery

Tim Davidson and Tom Bates

Introduction

Litigation for clinical negligence has accelerated at an alarming rate. In 2009/10 the NHS paid out £787 m (including £164 m for defence and claimant legal costs).¹ In the USA, the value of malpractice claims for delay in the diagnosis of breast cancer 15 years ago was second only to that for neurological damage to neonates²^,³ and today this situation remains unchanged.⁴ Poor cosmetic outcome is also a frequent cause for litigation and heightened public awareness has increased patients’ expectations of recompense for real or perceived injury following both cosmetic and reconstructive breast surgery.

Basic principles

The legal process differs between countries and although this chapter is based on civil law in England and Wales, the general principles in use elsewhere are similar.⁵ For a claimant to succeed in law, she must satisfy the court (in the UK a judge, or in some countries a jury) that there was a failure or breach of duty of care (liability) and that as a foreseeable result she suffered an injury (causation). For the case to succeed, the court must find in favour of the claimant with regard both to liability and to causation. Clinical negligence cases are heard in civil court and the judge determines on the balance of probabilities whether the defendant is liable. This is entirely different from a criminal court determining guilt or innocence where the level of proof is beyond all reasonable doubt (which many equate to a degree of confidence > 95%). If the court finds in favour of the claimant, the court awards financial recompense to redress, as far as money is able, the injury that she has suffered.

Breach of duty

Duty of care

Any doctor – GP, radiologist, surgeon or pathologist – owes each individual patient a duty of care. This is rarely an issue. In the NHS the doctor acts as an employee of the hospital or health board, which is covered by the NHS Litigation Authority, which administers a scheme that acts as a mutual insurer for participating trusts (Clinical Negligence Scheme for Trusts).⁶ When acting in a private capacity, the doctor is covered by a professional defence organisation of his or her choice.

The Bolam test

A doctor is not negligent if he or she acts in accordance with a practice accepted at the time as proper by a responsible body of medical opinion. The Bolam test arises from the case of a patient who received electroconvulsive therapy and sustained fractures.⁷ Negligence was alleged because the patient was not given muscle relaxants and was inadequately restrained. Some doctors would have used muscle relaxants and restraints, others not. The doctor was not found negligent because he acted in accordance with a practice accepted at the time, even though other doctors may have advocated a different practice.

The Bolitho modification of Bolam adds the requirement that for the practice or opinion formed to be acceptable, it must be based on logical argument; an irrational practice cannot be argued as reasonable in court simply because a body of medical opinion agrees with its use.⁸

The Bolam test requires a higher degree of skill from a specialist in his or her own field than from a GP. If a patient is referred to a breast surgeon and the standard of care falls below that which the patient could reasonably have expected from a breast specialist, there has been a breach of the duty of care. The standard of care needs to be determined by an appropriate peer group, so a GP expert would be asked by the court to advise on breach of duty on the part of a GP whereas a specialist expert would be asked to give an opinion on the standard of care delivered by a breast surgeon.

Guidelines

National and local guidelines of good clinical practice are now in use throughout the NHS with guidelines covering patient referral, diagnosis, treatment and organisational arrangements within breast units.⁹^,¹⁰ Breaches of guidelines are not indicative of, or equivalent to, negligent practice and guidelines are constantly being amended in the light of scientific knowledge, healthcare resources, government targets, etc. Consideration must always be given to the time at which the alleged breach of duty took place and for a guideline to be relevant it must have been in the public domain at the time.

Clinical practice that complies with guidelines is inevitably much easier to defend against allegations of negligence. A diagnostic excision biopsy exceeding 20 g, as set out in NHS Breast Screening Programme guidelines, does not equate to negligent practice; however, a patient claiming excessive deformity after such a procedure is unlikely to succeed in litigation if her biopsy specimen weighed under 20 g. There is ongoing debate regarding the medico-legal implications of surgical guidelines.¹¹ Carrick et al.¹² reported that whereas 41% of surgeons surveyed believed that guidelines would protect them against litigation, 37% believed that they would increase their exposure to claims.

The Woolf report

In a review of the UK civil justice system, Lord Woolf singled out clinical negligence cases ¹⁵ because the difficulty in proving both liability and causation accounts for much of the excessive cost and the high proportion of cases that fail. The root of the problem, however, lies less in the complexity of the law than in the climate of defensiveness. The patient’s disappointment when treatment goes wrong is heightened by what she perceives to be a refusal to acknowledge fault and an attempt to cover up.

The general rule is that ‘costs follow the event’ – so the unsuccessful party is responsible for the costs of both sides. Privately financed claimants or lawyers acting on a no-win, no-fee basis are therefore reluctant to pursue actions where the chances of success are small. However, if the claimant is supported by legal aid and loses the case, costs are not recoverable by the defendant. Most claims in the UK are legally aided because of the high costs of litigation; often the defendant is an NHS Trust, hence expenses incurred by both the claimant and defendant are funded from the public purse. In cases valued at less than £12 500, the median figure for the costs of litigation was 137% of the value of the claim. This explains the pressure to settle low-value claims even if defensible and a reluctance to appeal a doubtful judgment.

The recommendations of the Woolf report are intended to improve the resolution of disputes between patients and doctors, and reduce delay and costs while treating both parties fairly:

• Pre-litigation protocol where claimants should notify defendants with a written intention to sue 3 months before action. If liability is disputed, defendants should provide a reasoned answer.

• Lists on the Queen’s Bench to include a list of judges familiar with clinical negligence cases and training of trial judges in medical issues.

• Fast-track options for claims under £10 000 so that these can be litigated on a modest budget with a single expert acceptable to both parties appointed by the court.

• The medical expert is now required to address his report to the court and not to the instructing party,¹⁶ with an overriding duty ‘to provide objective unbiased opinion to the court on matters within his expertise, never assuming the role of an advocate’.

The burden (level) of proof

In civil litigation the court determines the facts, which means that the judge makes a decision on the balance of probabilities. This means that the successful claimant will normally recover damages in full, although in one case where the claimant was held to have lost an 80% chance of cure, a deputy high court judge directed that damages should be calculated accordingly.¹⁷

The all-or-none nature in awarding damages is arguably the most troubling aspect for experts involved in clinical negligence.¹⁸ For example, if the court finds that as a result of negligence a woman has suffered a reduction in her chance of survival from 60% to 40%, she will be awarded the full amount to compensate her (or her family) as though the loss had already occurred, on the basis that on balance she is now more likely to die. If the court finds, however, that her chance of survival has reduced from 90% to 60%, it may award her nothing on the basis that on balance her chance of survival remains unchanged.

In the case of Hotson v. East Berkshire Area Health Authority (1987), the House of Lords formulated the current UK position on causation. The defendant was in breach of duty in failing to diagnose a fracture of a femoral epiphysis following a fall. The child developed avascular necrosis with significant disability. The evidence was that the child had a 75% risk of developing this complication due to the accident and the trial judge and the Court of Appeal held that he was entitled to 25% of his damages for the 25% loss of a chance that prompt treatment might have prevented the complication. However, the House of Lords overturned the decision and held that on balance he was going to develop it even in the absence of negligence or, put another way, he had failed to establish that he was within the 25% who would not develop it.

The position on causation was again recognised in the appeal case of Gregg v. Scott (2002), a claim for failure to diagnose an axillary lump as non-Hodgkin’s lymphoma. The Court of Appeal held that the delay had reduced his chance of survival, although this had always been less than 50%, from 42% to 25% survival at 10 years. The case was appealed to the House of Lords who dismissed the claimant’s appeal for loss of a chance and upheld the traditional approach. So the burden of proof in deciding causation, for the time being at least, remains unchanged.

Delay in the diagnosis of breast cancer

Delay in diagnosis may occur as a result of failure to refer the patient from primary care or, once referral has taken place, for example with false-negative mammography, failure to perform triple assessment, misinterpretation of equivocal results or the misfiling of a positive test result.¹⁹

Litigation arising from delay in diagnosis concerns two main areas:

• Did the delay necessitate more extensive treatment?

Where the patient has had a mastectomy on the basis of tumour size, it is often claimed that earlier diagnosis would have made breast conservative surgery an option. For multifocal tumours, it can be argued that mastectomy would have been needed from the outset. With the advent of sentinel node biopsy, a claimant who has undergone axillary clearance for involved nodes may contest that earlier diagnosis at a time when her nodes were clear would have allowed less radical axillary surgery. When the patient has received chemotherapy it might be argued that this would not have been necessary if the diagnosis had been made sooner when, for example, the axillary nodes would on balance have been negative.

• Did the delay in diagnosis reduce the chance of cure?

This is a controversial area since there is public expectation, promoted over the years by health campaigners, that earlier diagnosis offers better chance of a cure. Where expert opinion is divided, the court often prefers the evidence in favour of the delay having caused a reduced survival time. A review of claims for diagnostic delay in breast cancer in Sweden over a 10-year period concluded that delays had an impact on treatment in 23% of cases and adversely affected prognosis in 11% of those patients for whom the delay was longer than 12 months.²⁰

Delay in primary care

The GP who sees many cases with benign breast symptoms each year but only one or two breast cancers is in a difficult position and GPs are increasingly facing litigation for delays in referral. The main area of contention is where long-standing or recurrent breast nodularity coexists with an (initially) undetected lump. Relying on a negative mammogram without an expert clinical examination, ultrasound or needle biopsy may increase the risk of false reassurance.¹⁹

GPs today are therefore faced with referring most women with breast symptoms for a specialist opinion and referral guidelines have been in use in the UK since 1995.21–23

Delay after specialist referral

Breast units throughout the UK submit audit data on compliance with 2-week target referrals for suspected cancers.²⁴ The prioritisation of referral letters can be counter-productive if non-urgent cases have to wait longer, but current NHS guidelines stipulate all symptomatic breast referrals are now seen within 2 weeks. NHS ‘31/62’ cancer targets (allowing 31 days from urgent referral to diagnosis and 62 days to commencing treatment) will identify units in breach of national guidelines.

The role of triple assessment, and the circumstances in which it fails, are critical. The specialist centre is also faced with the problem that women under 35 form the majority of the diagnostic workload (66%) but the fewest number of breast cancers (3%).²¹

The North American experience

Two studies commissioned by the Physician Insurers Association of America (PIAA)²^,³ showed delay in the diagnosis of breast cancer to be the commonest cause of clinical litigation, and a striking feature of both studies was young age: women under 50 accounted for 69% of claimants and received 84% of the damages paid, whereas only 25% of cancers occur in women under 50.

The most common reasons cited for the delay were (in descending order):

• physical findings failed to impress;

• failure to follow up the patient;

• negative mammogram report or misreading of the mammogram;

• failure to perform a biopsy.

In 487 cases where liability was admitted, the mean delay was 14 months. The mean payout was $301 000, with higher damages for longer delays and to younger patients.³

Diagnosis of breast cancer

Triple assessment is the foundation upon which clinicians diagnose breast lumps. However, the extent to which the accuracy of these tests is reduced in younger women is not well appreciated (Fig. 18.1). It has been suggested that perfection of diagnosis will require removal of every solid mass,²⁵ but this would represent a retrograde step. The practice of defensive medicine, in place of conventional wisdom, may well be encouraged by a litigious public and diagnostic tests where the sensitivity falls below 95%.

Figure 18.1 Sensitivity of clinical and mammographic examination by age. Reproduced from Dixon JM, Mansel RE. Symptoms, assessment and guidelines for referral. In: Dixon JM (ed.) ABC of breast diseases, 2nd edn. London: BMJ Books, 2000; p. 6. With permission from Wiley Publishing Ltd.

Physical examination

About 70% of all breast cancers are palpable, but with tumours of 0.6–1 cm diameter this figure falls to 50%.²⁶ The larger the breast and the greater the density of breast tissue, the more difficult physical examination becomes. Cyclical changes in breast parenchyma may require repeat examination at different phases of the menstrual cycle. Coexisting benign lumps, scars and distortion from previous surgery, the ridge of tissue above the inframammary fold, changes during pregnancy and lactation, and the underlying ribs all add to the uncertainty of clinical examination. Other difficulties include: inflammatory cancers masquerading as infection; the presence of implants with an associated fibrous capsule; and the effect of hormone replacement therapy, which increases the density of breast parenchyma both clinically and radiologically.

The sensitivity of clinical examination in women aged 30–39 can be as low as 25%.²⁶ A sensitivity over 90% can only be expected in older women, when the atrophic nature of the breast parenchyma and the low incidence of benign disease combine to make clinical diagnosis a relatively simple task. The low sensitivity of clinical examination, coupled with the low incidence of breast cancer and the considerable numbers of young women attending breast clinics, must largely explain why failure of physical findings to impress the clinician was one of the most common reasons for delay in diagnosis in the PIAA study.³

Mammography

False-negative mammography is one of the principal reasons for delay in diagnosis,³^,²⁷ since it gives the clinician and patient false reassurance. Age is an important factor in false-negative reporting (Fig. 18.1), with the number of cancers missed inversely proportional to age: 36% of cancers in women aged 40 compared with just 9% in those aged 75. The net result of medicolegal pressure on breast radiologists has been an increase in recall and biopsy rates.²⁸^,²⁹

Ultrasound

The use of ultrasound is now an integral part of breast imaging in a patient of any age with a lump (see Chapter 1), especially when an abnormality is not detectable on clinical or mammographic examination. Ultrasound-guided core biopsy has replaced fine-needle aspiration cytology (FNAC) in screen-detected tumours, although both techniques are currently acceptable in the symptomatic clinic. The expertise required for ultrasound examination and guided core biopsy has placed breast imaging outside the competency of the general radiologist. The breast surgeon using ultrasound in the clinic may be similarly compromised unless training in the technique can be verified.

Cytology

For many years FNAC has allowed rapid and cost-effective tissue diagnosis of a palpable lump. How reliable is FNAC? Dixon et al.²⁶ reported that the sensitivity of FNAC could be increased from 66% to 99% by restricting the biopsy to one aspirator. In comparison, the sensitivity of FNAC in women under 36 was as low as 78%,²⁵ though Dixon et al.²⁶ found that accuracy was not related to age if inadequate samples were excluded from the calculations.

In a review of 112 reports of FNAC of breast masses by Layfield et al.³⁰ the overall accuracy was over 95% but delays in diagnosis of greater than 50 days still occurred, and 85% of such delays were in women under 55³¹ and with smaller tumours.³² In small tumours, a sampling error is probably more common than misinterpretation of the cytology and ultrasound-guided core biopsy is advisable.

Efficacy of triple assessment

If it is assumed that breast cancer detection by clinical examination, imaging and FNAC are independent of each other, it is possible to calculate the theoretical rate at which all three tests will give a false-negative result. The false-negative rates of the three investigations have been multiplied and expressed as a percentage in Table 18.1. It is probable that the sensitivities of these tests are not totally independent of each other, and therefore the predicted rate that all three tests will be false negative for an individual is a conservative estimate. For a woman under 35 with breast cancer, there may be a 12% chance that all three tests will give a false-negative result. Using optimal sensitivities, the chance that all three tests will produce a false-negative result falls to approximately 1 in 1000 patients with these data skewed towards the older age groups; the likely overall rate of false-negative triple assessment in the clinic is between 1.4% and 4%.³²^,³³

Table 18.1

False-negative rates of triple assessment for women under 35 compared with the generally quoted results ¹⁹^,³¹

Tumour doubling time

The usual threshold diameter for detection of a breast cancer by physical examination is 1 cm; such a tumour consists of 10⁹ cells and is the result of 30 doublings. It is possible for mammography to detect tumours as small as 2 mm, which equates to a tumour of 10⁷ cells and about 23 doublings.

Assuming a constant doubling time, early detection of breast cancer is a misnomer, since at least two-thirds of the biological life of the tumour will have been completed by the time of detection.³⁴ In medicolegal terms, if the alleged delay in diagnosis was 14 months for a cancer with a doubling time of 90 days, such a delay would equate to the number of cells increasing by one order of magnitude (i.e. from 10⁹ to 10¹⁰ cells). This represents a major increase in tumour load but is a very short period in the lifespan of the tumour and it is difficult to be sure that this period of delay would have a significant effect. In civil law, however, the court wants to know whether such an effect is more likely than not to alter the patient’s prognosis or treatment.

Lymph node status remains the most important prognostic indicator at the time of primary treatment. Tubiana and Koscielny ³⁵ suggested that breast cancer represents a continuum from slow-growing tumours with late axillary involvement and distant dissemination to the most aggressive, rapidly growing and early metastasising subtype. Assuming the growth rate of the nodal metastasis approximates to that of the primary tumour, it is possible to estimate the theoretical time at which the tumour must have metastasised, and not infrequently this would have occurred before the threshold size for detection of the primary tumour.³⁴

It is generally accepted that breast cancer begins as a single cell or a small group of cells that exhibit an exponential growth pattern. The time taken for a tumour to double in volume is known as the doubling time. Doubling times for breast cancers have been estimated by measuring the size of mastectomy scar recurrences ³⁶ and by serial mammographic evaluation.³⁷

Pearlman ³⁶ categorised patients as having fast- (< 25 days), intermediate- (26–75 days) and slow-growing (> 76 days) tumours based on measurement of tumour doubling time; 5-year survival rates were 5%, 62% and 100% respectively. Peer et al.³⁸ estimated mean doubling time in women under 50 to be 80 days (confidence limits 44–147 days) and in women over 50 to be 157 days (confidence limits 121–204 days). Although lymph node metastases are more commonly found in fast-growing tumours, it would be wrong to assume that the survival of patients was entirely a consequence of tumour doubling time. Galante et al.³⁷ emphasised the importance of the metastatic potential of the tumour, suggesting that within fast-, intermediate- and slow-growing tumours there may be subsets with high and low metastasising potential.

The premise that early detection of a tumour will lead to cure depends on the concept that at the time of earlier treatment the tumour has not metastasised. The period of time between the earliest possible detection of the cancer and the time at which the tumour metastasises has been described as the cancer control window.³⁹ If the tumour has already metastasised by the time it reaches the threshold size for detection, there is no window and only effective systemic therapy might cure the patient. It is often argued by the claimant and her legal team that delay in treatment is the cause of metastasis rather than the inherent biology of the tumour itself.

Vignettes on breach of duty

The following vignettes illustrate areas of breach of duty that have arisen in medicolegal breast cases and the comments following each vignette raise issues that might be discussed in conference with counsel.

Vignette 1: False-positive cytology, lymphoedema and alteration of records

A 60-year-old patient was referred to a breast clinic with a clinically suspicious breast lump. Mammography was suspicious (M4) and FNAC was malignant (C5). The patient underwent wide local excision and axillary clearance for what subsequently proved to be a histologically benign condition. She complained of a poor cosmetic outcome and a painful swollen arm, which she had not been warned about. The nursing records stated ‘patient was warned of the risk of lymphoedema’.

Comments

• A false-positive cytology result is a rare but potentially devastating event. From a medico-legal standpoint, an expert review of the cytology should be the first step. If experts are agreed that the cytologist acted reasonably in reporting the slides as malignant, attention will then focus on the surgeon.

• The surgeon has to be aware that a false-positive cytology can occur and for this reason most surgeons will not carry out mastectomy without histological confirmation from a core biopsy. Histology is mandatory in the absence of a fully concordant malignant triple assessment and should have been obtained in this case. Axillary clearance for a benign condition is a potentially significant injury and it is unfortunate that this patient developed lymphoedema.

• The tense of the warning suggested to the judge that this was a retrospective note. Judges place most reliance on contemporaneous records and alterations to the notes are often quite obvious. Any addition to the notes must be clearly identified as such by signing and dating the record and any temptation to alter the record after the event must be firmly resisted.

Vignette 2: Failure of preoperative assessment and postoperative management

A 69-year-old woman presented with a small but obvious carcinoma in the tail of the breast. The surgeon performed FNAC, which showed malignant cells (C5), and he carried out wide local excision of the tumour without preliminary breast imaging and without performing axillary staging. The tumour was grade 2 but the resection margins were not reported. The patient was started on tamoxifen but was not referred for consideration of radiotherapy.

Comments

• The management of this patient falls short of an acceptable standard in a number of respects (breach of duty), but until such time as she develops evidence of recurrence, which may never occur, any resultant harm (causation) remains potential.

• A preoperative mammogram might have shown widespread microcalcification or multifocal tumour and in this situation conservative surgery would not have been appropriate. There might also have been an undetected cancer in the contralateral breast.

• To fail to report the margin status and managing a patient with conservative surgery without consideration of radiotherapy is unacceptable.

• It is clear that this case was not discussed at a multidisciplinary meeting.

• Axillary node status should have been assessed and has been recommended in UK guidelines since 1995.⁹

Vignette 3: False-negative cytology, trainees in the clinic and lobular carcinoma

A 38-year-old patient was referred to a breast clinic with a breast lump. Mammography was negative and a specialist registrar found an indefinite lump of which he was not suspicious. He carried out FNAC but failed to achieve an adequate specimen (C1). At follow-up 6 weeks later a repeat FNAC showed an adequate benign sample (C2). She was discharged from the clinic but re-presented 6 months later with an invasive lobular carcinoma at the same site.

Comments

• Trainees must be adequately supervised and only permitted to see patients by themselves when their trainer is satisfied that they are competent and understand local protocols.¹⁰

• There is a learning curve in achieving adequate samples with FNAC. The rate of inadequate FNAC samples should be monitored for each clinician.

• Clinicians should be aware that lobular cancers are prone to false-negative mammography. There is also an increased difficulty with false-negative cytology interpretation.

• The diagnosis of breast cancer in young women is difficult and in the situation where a patient complains of a lump but on clinical examination or mammography nothing is obvious, ultrasound is advisable, with guided biopsy if a lesion is seen.

Vignette 4: Pneumothorax following FNAC

A dental nurse aged 32 presented with a lump in the tail of the breast. An experienced clinical assistant in the breast clinic carried out FNAC. Unfortunately, he pierced the pleura and caused a pneumothorax that required hospitalisation and pleural drainage. The patient was not warned of the risk and complained of persistent chest pain over a period of many months.

Comments

• Did the clinician fail in his duty of care by advancing the needle too far or by failing to warn the patient? The literature suggests a risk of around 1 in 10 000.40–42 It is not rare to strike a rib with the needle at FNAC. Piercing the pleura may be more common because a trivial pneumothorax may go undetected.

• In several cases the court has found pneumothorax to be a rare but recognised complication that can occur without breach of duty, although one case ruled this injury to be negligent, which was an unexpected judgment.⁴² There is often a reluctance on the part of trusts to defend or appeal low-value cases under £10 000 (as would be the case for damages here).

• In defending a claim of negligence, it is important to be able to show that the clinician was experienced in the technique. Were this to be a trainee, it would be necessary to establish that he or she had been properly supervised.

• Pneumothorax has been reported following needle localisation by a radiologist and following aspiration of axillary seroma, performing FNAC and core biopsy. Using image guidance should theoretically reduce the risk.

Delay in diagnosis: causation issues

In a case of delay in diagnosis, it is often relatively straightforward to establish whether there has been a breach of duty of care. The next hurdle to be overcome by the claimant is to satisfy the court, on the balance of probabilities, that this delay caused her harm. Expert opinion is often divided and if the experts cannot agree then the court makes a judgment. The public and the judiciary have the expectation that early diagnosis carries a better outlook. It is not surprising therefore that counter-arguments of lead-time bias and innate tumour biology tend to fall on deaf ears.

Having established breach of duty, the issue of causation may include the allegation that less treatment would have been required. It may also be argued that psychological damage has ensued. However, the main issue centres around whether the claimant has suffered a reduced expectation of life or, if she has died by the time the matter comes to court, whether she would have lived longer. Richards et al.,⁴³ after a systematic review, concluded that there is an average reduction in 5-year survival of 7% with a delay of 3–6 months and 12% with a delay of more than 6 months. Dische et al.⁴⁴ extrapolated a 1.8% decrease in survival from Richards et al.’s data for each 1-month delay up to 6 months. In the same issue of the Lancet, Sainsbury et al.⁴⁵ reported no such effect but produced the apparently contradictory finding that patients with the shortest delay had the worst prognosis, a finding previously reported by Afzelius et al.⁴⁶

The survival in patients with the best-prognosis tumours (small, special type, grade 1, node negative) approaches that of the normal population, and clearly a formula attributing a reduction in survival for each month of delay would be inappropriate in such tumours. However, if one accepts this, then for some tumours with a worse prognosis there must be a greater loss of survival.⁴⁴ It would nevertheless seem intuitive that the most aggressive tumours with rapid metastatic potential are likely to be incurable from early in their natural history, suggesting diagnostic delays in such cancers are less likely to have an impact on survival.

Vignettes on causation

The following vignettes illustrate some of the causation issues that have arisen once liability has been established. Delay in diagnosis remains an area of considerable uncertainty and the comments reflect our experience of differing arguments presented to the court.

Vignette 5: 12-month delay in diagnosis of node-positive carcinoma

A 32-year-old woman was referred to a breast clinic with a lump in the breast. Ultrasound showed an indeterminate opacity 1 cm in diameter consistent with a fibroadenoma but no sample was taken by FNAC or core biopsy. She was discharged from the clinic but returned a year later with a clinical carcinoma at the same site. This measured 2.1 cm on both ultrasound and histology and was grade 3; one of four nodes was positive. Liability for delay in diagnosis was admitted in failing to carry out a biopsy at the first visit.

Comments

• The Nottingham Prognostic Index (NPI)47–49 and Adjuvant!Online⁵⁰ are often used to determine the difference in outcome over the period of delay.

• Both NPI and Adjuvant!Online rely on the following assumptions about the individual case:

Assumption 1: the tumour grade remains constant – this is usually agreed by both sides.

Assumption 2: earlier tumour size. In this particular case a record of tumour size at the first visit was available, but if no clinical measurement was recorded and there was no imaging, an approximate tumour size has to be derived by working back from the tumour volume at diagnosis using tumour doubling times. This calculation assumes (i) that tumour growth is exponential, (ii) that in calculating tumour diameter the tumour approximates a sphere or an ellipsoid, and (iii) that the doubling time chosen is appropriate. Tumour size itself is a weak determinant of prognosis but the derived earlier tumour size is further used to calculate the likely nodal status.

Assumption 3: the nodal status at the time of the breach of duty is usually unknown and often disputed by the experts. Axillary node status is invariably presumed to have been negative by the claimant, and this claim is supported by tables for tumour grade and tumour size that show the probability of positive nodes.⁵¹ Only in grade 2 and 3 tumours greater than 2.5 cm does the probability of positive nodes rise above 50%. Therefore, on balance it may often be argued that the nodes would have been negative at the earlier time when the tumour was smaller.

Vignette 6: 2-year delay in diagnosis of node-negative grade 1 carcinoma

A 40-year-old woman presented with a lump in the breast and a triple assessment was carried out. The tumour measured 1.5 cm on ultrasound and mammography, and FNAC was reported as benign (C2). She was discharged but 2 years later returned with a carcinoma 3 cm in diameter on histology. The patient was treated by breast conservation and postoperative radiotherapy.

The tumour was grade 1 and four nodes sampled were clear. Review of the original cytology indicated that this had been under-reported and an expert opinion graded the slides unequivocally malignant (C5). Breach of duty was admitted.

Comments

• The standard of care must be judged against the standard reasonably to be expected of a cytologist working at the same level and not that of a world expert in the field.

• Although the tumour doubled in diameter over the 2-year period of delay, it remained within the good prognosis group. The treatment would have been the same with an earlier diagnosis and therefore the case did not succeed on causation.

Vignette 7: 3-year delay in diagnosis of a carcinoma missed on screening

A woman of 50 responded to an invitation for screening and was recalled for a localised area of microcalcification. There was a soft-tissue opacity and the appearance was judged benign on further magnification views and ultrasound. She was returned to routine screening but 3 years later the screening films showed an obvious carcinoma at the same site. This was a 2-cm grade 2 infiltrating carcinoma with an extensive in situ component; four of 10 nodes were positive. An expert opinion from a breast-screening radiologist rated the original films as suspicious for ductal carcinoma in situ (DCIS; M4) and stated that the microcalcification should have been biopsied by any competent breast radiologist.

Comments

• Delay due to radiological misinterpretation tends to be a matter of years rather than months. Screening by 3-yearly mammography probably reduces breast cancer deaths by up to 25%, and it is likely that a delay in diagnosis of 3 years will affect survival in a proportion of cases.

• The potential loss of survival in this case would be considerable since the original lesion would, on the balance of probabilities, have been an area of high-grade DCIS with a near-normal expectation of life if adequately treated at age 50. At age 53 the patient now has a relatively poor prognosis but on balance is still more likely to survive 10 years (Table 18.2).

Table 18.2

Nottingham Prognostic Index groups

Group	NPI value	10-year survival (%)
Excellent (EPG)	2.0–2.4	96
Good (GPG)	2.41–3.4	93
Moderate I (MPGI)	3.41–4.4	82
Moderate II (MPGII)	4.41–5.4	75
Poor (PPG)	5.41–6.4	53
Very poor (VPG)	≥ 6.41	39

From the Nottingham Primary Breast Cancer Series. Data relate to patients with primary operable breast cancer, treated from 1990 to 1996.⁴⁷

Vignette 8: 14-month delay in the breast clinic and failure to recommend chemotherapy

A 30-year-old woman was referred with a lump in the breast and was seen by a succession of specialist registrars. An initial ultrasound showed a 1-cm opacity consistent with a fibroadenoma but the FNAC was reported as mildly atypical (C3) and the pathologist advised ‘consider biopsy to confirm’ but no action was taken. A series of 6-month follow-up appointments were given. Finally the GP referred the patient again to the consultant, who immediately diagnosed breast cancer. The tumour was 4 cm, grade 3 and heavily node positive. The patient was treated by wide local excision, axillary clearance and radiotherapy. She was not given chemotherapy or hormone treatment until she developed bone secondaries 16 months later. Breach of duty was admitted for the delay in diagnosis and the failure to give chemotherapy at the time of diagnosis.

Comments

• Discussion of the discordant triple assessment at the breast multidisciplinary meeting (MDM) did not take place.

• Arranging a 6-month follow-up for a breast lump that is presumed benign is illogical.

• The judge took evidence from five expert witnesses. It was agreed that when first seen the tumour would have been grade 3, 1 cm in diameter and node negative. The tumour had increased in size to 4 cm in the 14 months of delay and opinion was divided as to whether this was ever potentially curable. The experts for the claimant were far more optimistic than those for the defence but in the event the judge preferred the latter.⁵²

• The judge considered that, on balance, the breach of duty ‘caused her to die 18 months before, sadly, she would have died anyway’.

• Unfortunately, the damages amounted to less than the defence had already paid into court. Civil litigation rules resulted in the family winning the case but receiving none of the settlement.

Vignette 9: 8-month delay in diagnosis in primary care

A 46-year-old woman attended her GP with a new lump that the GP assessed as a 1.5-cm smooth benign lump; in view of the past history of benign breast nodularity, the GP recommended a conservative approach and it was only when she saw a different GP 8 months later that referral to the clinic and a diagnosis of breast cancer was made. She underwent mastectomy and axillary clearance for a 42-mm grade 2 carcinoma with seven of 20 nodes involved, followed by chemotherapy, radiotherapy, trastuzumab and endocrine therapy. She had suffered persistent shoulder discomfort since the time of axillary surgery.

Comments

• Following an opinion from a GP expert, breach of duty on the part of the GP was admitted and the claim for causation was both for more extensive treatment and loss of chance of survival.

• The claimant’s breast expert maintained that she would had undergone wide excision and sentinel node biopsy at the earlier time and that her expected survival had been reduced from 93% to 53%.

• At a pre-trial meeting of experts, they agreed jointly that the nodes would still have been involved at the earlier time and that the axillary surgery would have been unchanged, and agreed that she was, on balance, still more likely to survive than die from the breast cancer.

• The experts agreed that she had lost the chance of breast conservation as a result of the delay and following their joint report the case was settled for £50 000 on the basis that she had undergone mastectomy and would require future breast reconstruction.

• Less than 5% of clinical negligence cases are contested in court with over 95% of cases either dropped or settled before the trial date.

• Since the discussion of discordant triple assessments in breast MDMs has become widespread good practice, claims for delay in diagnosis against breast specialists have reduced and are now mainly against GPs who fail to refer cases to the breast clinic.

Poor cosmetic outcome

The mastectomy rate in the UK remains close to 40%, and where mastectomy is required the increasing demand for reconstruction has been addressed in national guidelines ⁵³ and detailed audit.⁵⁴ Litigation has risen in line with patients’ expectations,⁵⁵ although the typical size of claim for an unsatisfactory cosmetic outcome is considerably less than for delay in diagnosis.

Clinical notes should record evidence of discussions with the patient regarding either immediate or later breast reconstruction and should document that alternative treatment options and potential complications of surgery have been discussed. The presence of the breast-care nurse to augment and further document the surgeon’s counselling is advisable and her written record, usually kept separately from the clinical notes, can make all the difference when allegations of inadequate preoperative information are part of a claim for breach of duty on the part of the reconstructive surgeon.

Preoperative markings should be made on the ward by the operating surgeon, explained to and agreed with the patient without pressure of time, and a record kept of this operative planning. Hasty skin marking in the anaesthetic room or on a patient already anaesthetised are not compatible with a reasonable standard of care.

A request for immediate reconstruction can present difficulties in offering the most appropriate reconstruction and yet not breach guidelines for commencing treatment. This tension has presented medicolegally with a patient offered a more basic form of reconstruction with a subpectoral implant in order to expedite mastectomy who, once the cancer had been dealt with, felt that the advice fell below a reasonable standard. The breast surgeon must make available to the patient the appropriate range of reconstructive options, including those that might require the involvement of a plastic surgical colleague.

Should a patient pursue litigation for her cosmetic outcome, the Bolam principle ⁷ would apply in determining breach of duty, i.e. the practice would be compared with that held to be reasonable by a similar body of professionals, in this case breast surgeons trained to undertake breast reconstruction. An audit of the surgeon’s operative results is not currently expected by the court, nor are there prescribed minimum numbers of procedures to comply with good practice. At present there is no certificate of competency in training in breast reconstruction within the UK, but current reconstructive guidelines⁵³ outline the requirements for good practice.

When a poor cosmetic outcome is due to a recognised complication rather than poor judgment and provided the patient has been warned of the risk, she does not have a case against the surgeon.

Reduction mammoplasty, breast augmentation and surgery for gynaecomastia remain high-risk areas for patient dissatisfaction and should never be undertaken on an occasional basis. Breast operations represent the biggest group of plastic surgical procedures giving rise to claims.⁵⁶ It is vital that evidence of appropriate discussions about potential problems is adequately recorded in the notes and the correspondence, together with the use of appropriate information leaflets and clinical photographs to explain the proposed procedure. Consent forms should record the signed informed consent as well as documenting that the literature has been received and understood. Operation notes, where relevant, should include details of implant type, serial and lot number, and manufacturer, for possible future reference.⁵⁷