reading

Contents

Cover

Title Page

Contributors

Preface

Abbreviations

Introduction

References and further reading

Part 1

Section 1: The legal structure of negligence

A few words about error

Medical negligence

Learning from system failures – the vincristine example

Reference

Section 2: Causes of diagnostic errors in general practice and how they can be avoided

How do general practitioners reach diagnoses?

Where do errors occur in diagnosis?

How can we minimize the risks of these errors?

References and further reading

Section 3: Bayesian reasoning and avoiding diagnostic errors

References and further reading

Section 4: A potpourri of advice on avoiding errors

History and examination

The telephone consultation

Communication problems

When lack of knowledge plays a part

The unexpectedly abnormal result

The standard of notes

Drug errors or prescribing errors

Consent

Confidentiality

Conditions that are ‘frequent flyers’ in negligence cases

Safety netting

References and further reading

Part 2: Clinical cases

Introduction

Case 1: A man with iron deficiency

Expert opinion

Legal comment

References and further reading

Case 2: When is a headache abrupt?

Expert opinion

Legal comment

References and further reading

Case 3: A woman with chest pain

Expert opinion

Legal comment

References and further reading

Case 4: A dizzy man

Expert opinion

Legal comment

References and further reading

Case 5: A dizzy man

Expert opinion

Legal comment

References and further reading

Case 6: A pulled calf muscle

Expert opinion

Legal comment

References and further reading

Case 7: A woman with hemiplegic migraine

Expert opinion

Legal comment

References and further reading

Online resources

Case 8: Irritable bowel syndrome after sickness in Goa

Expert opinion

Legal comment

References and further reading

Case 9: A young man with back pain

Expert opinion

Legal comment

References and further reading

Case 10: Irregular intermenstrual bleeding in a woman on the pill

Expert opinion

Legal comment

References and further reading

Case 11: A boy with a limp

Expert opinion

Legal comment

References and further reading

Case 12: A runner with a cough

Expert opinion

Legal comment

Further reading

Case 13: A woman with classical migraine

Expert comment

Legal comment

Further reading

Case 14: A young woman with diarrhoea and vomiting

Why was the diagnosis of appendicitis missed?

Expert comment

Legal comment

References and further reading

Case 15: Ill-fitting dentures in an elderly man

Expert comment

Legal comment

References and further reading

Case 16: Back pain in a middle-aged woman

Expert comment

Legal comment

References and further reading

Case 17: Cellulitis in a man's foot

Expert opinion

Legal comment

References and further reading

Case 18: A flare-up of ulcerative colitis

Expert comment

Mild

Moderate

Severe

Legal comment

References

Case 19: A woman with a skin lump on her leg

Do you think a claim against Dr Welch will succeed?

Expert comment

Legal comment

References and further reading

Case 20: A woman with microscopic haematuria

Expert comment

Legal comment

References

Case 21: A limping young girl

Expert comment

Legal comment

References and further reading

Case 22: A builder tripping over his feet

Expert comment

Legal comment

References and further reading

Case 23: An anxious young woman with hyperventilation

Expert comment

Legal comment

Further reading

Case 27: A febrile baby

Expert comment

Legal comment

Further reading

Part 3: Investigating and dealing with errors

1 Introduction

2 How errors and their recurrence are prevented in primary care

3 The role of the Primary Care Trusts

4 Other investigations

5 Legal advice – where to get it and how to pay

6 External inquiries

7 The role of the doctor

8 Emotional repercussions

9 Conclusion

Reference

Index

Eula

Case 1 A man with iron deficiency

Jeff is a 53-year-old man who returned from a two-week holiday visiting family in Kenya with symptoms of a febrile illness. He had taken antimalarials but was concerned about malaria. He consults Dr Wallace. She finds that he is apyrexial and looks well but she requests a full blood count and films for malarial parasites. The films are negative for parasites. The FBC shows a haemoglobin of 12.3 g/dL and a MCV of 67 fL (normal ranges > 13.5 g/dL and 70 to 100 fL). The results are filed.

What would you have done with the results?

Four months later Jeff consults another GP in the practice, Dr Rennie, with a two-week history of a dry cough and fatigue. Examination is unremarkable, urinalysis and TFTs are normal but the Hb is 10.7 g/dL with a MCV of 66 fL. Dr Rennie diagnoses an iron deficiency anaemia, requests haematinics, starts Jeff on ferrous sulphate and requests a repeat blood test in two months. The serum ferritin is 10 mcg/l. Dr Rennie suggests doing three faecal occult bloods. Two are done and are negative. Two months later the repeat FBC result comes back to the ST2 trainee, Dr Bordley. The man's haemoglobin is now 13.6 g/dL with a MCV of 70 fL.

What would you do now? What is your differential diagnosis?

Dr Bordley asks the receptionists to contact Jeff for a routine appointment. Unfortunately Jeff fails to make an appointment as requested.

Six months later Jeff consults Dr Rennie again, this time with fatigue. He is clinically anaemic. He stopped taking the ferrous sulphate some months earlier. Abdominal examination suggests a right lower quadrant abdominal mass. A FBC shows a Hb of 9.1 g/dL with an MCV of 66 fL. Dr Rennie refers Jeff urgently to a consultant gastroenterologist and a colonoscopy shows a circumferential, stenosing caecal carcinoma. He undergoes a right hemicolectomy and the cancer is staged as Dukes C1.

Jeff sues the three general practitioners on the premise that his iron deficiency anaemia should have been investigated one year earlier and that this would have led to an earlier diagnosis of the cancer with a better prognosis.

Expert opinion

As with many medico-legal cases there are often a number of errors which compound the delay in diagnosis. Iron deficiency in a man in the developed world is rarely due to dietary deficiency (Goddard et al., 2005). In contrast, menstruating women can easily tip into a negative iron balance because of a significant monthly loss with menstruation. Anaemia indicates significant iron deficiency following exhaustion of the marrow stores. Iron deficiency anaemia in a man is rare and always needs investigating. It must never be assumed to be dietary because such an assumption is unsafe and rarely true.

Since July 2000 the UK ‘Two Week Rule’ referral guidelines have recommended urgent referral for men with an iron deficiency anaemia of less than 11 g/dL in a man and 10 g/dL in a non menstruating woman (NHS Executive, 2000). However, the 2005 British Society of Gastroenterology (BSG) guidance is to refer for investigation all men and women with an unexplained iron deficiency anaemia of any degree (Goddard et al., 2005). The likely causes are occult blood loss from a carcinoma or peptic ulceration (particularly with NSAIDs) or malabsorption due to, for example, Coeliac's disease. Most right-sided colorectal cancers present with an isolated iron deficiency. Right-sided colorectal cancers often do not cause the characteristic alteration in bowel habit (to loose, frequent stools) that occurs with left-sided cancers. Cancers proximal to the splenic flexure will also not produce rectal bleeding.

One study of 695 patients with iron deficiency anaemia referred under the BSG guidance found colorectal cancer in 6.4% (James et al., 2005). A second study of 431 patients with a haemoglobin below 12 g/dL in men and 11 g/dL in women found 7.4% had colorectal cancer (Logan et al., 2002). Another case-controlled study using a retrospective analysis of primary care computerized records identified 6442 patients with colorectal cancer and 45 066 controls. For men with an iron deficiency anaemia the positive predictive value of iron deficiency anaemia for colorectal cancer was 13.3% (Hamilton et al., 2008).

The request for faecal occult bloods is an example of a clinician not understanding the performance characteristics (sensitivity and specificity) and therefore significance of a requested test. The test was incorrectly being requested as a ‘rule out test’ – if the FOBs were negative the GP would consider that occult bleeding had been ‘ruled out’. However, the sensitivity of the test was too low to be a ‘rule out test’. The sensitivity of three FOBs is only 50% to 90% for colorectal cancer (Hewitson et al., 2007). Many people with colorectal cancer will still have three negative tests. To rule out a diagnosis, the test has to have a high sensitivity (like a barium enema) and be negative. The acronym is SnOUT – high sensitivity negative test rules it out.

The initial assessing GP, Dr Wallace, should have recognized that iron deficiency anaemia in a man, even if mild, requires confirming with a serum ferritin level and then investigating if it is confirmed. Jeff would not have met the criteria for an urgent ‘Two Week Rule’ referral (the Hb needs to be below 11 g/dL) but should have been referred non-urgently for specialist gastroenterological opinion.

The second GP Dr Rennie correctly requested a repeat full blood count and a serum ferritin but should then have referred the patient urgently to see why he had an iron deficiency anaemia. The faecal occult blood tests were inappropriate. They are colorectal cancer screening tests for an asymptomatic population.

The ST2 doctor Dr Bordley should have recognized that there was a high probability that Jeff had serious gastro intestinal disease (cancer or an ulcer) and ensured that he was seen rather than have merely delegated contacting the man to a receptionist.

Legal comment

Expert comment confirms that Dr Wallace ought to have also carried out a serum ferritin test to rule out iron deficiency anaemia, an indicator of colorectal cancer. Although Jeff did not meet the criteria for an urgent two-week referral he ought to have been referred non-urgently for specialist gastroenterological opinion because of the low haemoglobin results.

Four months later Jeff returned to his GP practice with a dry cough and fatigue and saw Dr Rennie. Although Dr Rennie diagnosed iron deficiency anaemia she failed to refer Jeff for investigation of the unexplained iron deficiency.

The mistakes continued when the ST2 Trainee Dr Bordley failed to recognize the high probability that Jeff had serious gastro intestinal disease from the results of the repeat FBC.

Six months later Jeff consulted Dr Rennie again when she found a right-sided abdominal mass on examination and appropriately referred Jeff to a consultant gastroenterologist.

All three doctors failed to consider any differential diagnoses. They appeared not to be aware of BSG guidance or the Two Week Rule guidance. Communication between all three doctors about Jeff's unexplained anaemia appears to have been poor. A breach of duty by the GPs seems clear. The next question is what the results of that breach were. It is fortunate that the cancer seems to be confined to the right side of the colon and has not spread beyond it. Had Jeff been properly diagnosed with iron deficiency anaemia and appropriately referred to a specialist gastroenterologist at the outset, his prognosis may not have been any different. (He may still have required surgery.) If expert opinion confirms that it is unlikely that the prognosis would have been any different had the cancer been identified earlier, then Jeff could expect to be awarded damages for approximately one year's pain and suffering with contributions being made by all three doctors.

Key learning points

Specific to the case

Iron deficiency anaemia in men should never be assumed to be dietary as this is unlikely and an unsafe assumption.
Approximately 10% of men with unexplained iron deficiency anaemia will have colorectal cancer, often of the caecum.

General points

It is important to know the performance characteristics of a test before relying on it to ‘rule in’ or ‘rule out’ a diagnosis.
It is only sensible to ‘rule out’ a condition if the test has a high sensitivity and is negative – SnOUT.

References and further reading

Goddard AF, James MW, McIntyre AS, Scott BB (2005) Guidelines for the management of iron deficiency anaemia. British Society of Gastroenterology.

Hamilton W, Lancashire R, Sharp D, Peters TJ, Cheng KK, Marshall T (2008) The importance of anaemia in diagnosing colorectal cancer: a case-control study using electronic primary care records. Br J Cancer 98: 323–7.

Hewitson P, Glasziou P, Irwig L, Towler B, Watson E (2007) Screening for colorectal cancer using the faecal occult blood test, Hemoccult. Wiley Online Library. DOI: 10.1002/14651858.CD001216.pub2

James MW, Chen CM, Goddard WP, Scott BB (2005) Risk factors for gastrointestinal malignancy in patients with iron deficiency anaemia. European Journal of Gastroneterology and Hepatology 17: 1197–1203.

Logan E, Yates J, Steward R, Fielding K, Kendrick D (2002) Investigation and management of iron deficiency anaemia in general practice: a cluster randomised controlled trail of a simple management prompt. Postgraduate Medical Journal 78: 533–7.

NHS Executive (2000) Referral Guidelines for Suspected Cancer, 19–20. London, Department of Health.

Case 2 When is a headache abrupt?

Hannah was 36 when she presented to an Out of Hours general practitioner, Dr Walmesley, situated in an emergency department (ED). She had a headache. She was a known migraine sufferer but had not suffered from migraine for a couple of years and felt that this was the worst attack she had suffered for many years and that her routine analgaesia was inadequate. She stated that four years earlier she had been referred to a specialist with episodes of vertigo associated with headache and the diagnosis had been migraine. Her current headache had come on 6 hours earlier. It was initially in the neck and back of the head and had progressively worsened over 10–15 minutes to being very severe. She had vomited once. On examination she was afebrile with no neck stiffness and no abnormalities other than mild photophobia on retinal examination. Dr Walmesley diagnosed migraine and prescribed strong opiates and an anti-emetic.

What would you have done?

Two days later the woman consulted her own general practitioner Dr Palmer. She told the doctor that she had been to hospital two days earlier and that the hospital had diagnosed migraine but that the painkillers were only relieving the headache for a couple of hours. She had a minor degree of neck ache and stiffness. There was no abnormality on examination and Dr Palmer suggested the use of a tryptan.

What would you do now? What is your differential diagnosis?

Ten days later Hannah was found dead in bed. A post-mortem revealed a large recent subarachnoid haemorrhage (SAH) originating from a 25 mm basilar artery berry aneurysm. Hannah's family sued the general practitioners for negligently making a diagnosis of migraine. It was argued that a competent general practitioner would have admitted her for urgent investigation, that an MRI would have demonstrated the berry aneurysm and that it would have been treated with an endovascular coil.

Do you think their claim will succeed?

Expert opinion

The suspicion and referral of patients suspected of having a SAH or ‘warning bleeds’ present a very significant problem for general practitioners and ED staff. Few general practitioners would fail to suspect a SAH in a patient with the typical presentation of an abrupt onset of a very severe occipital headache (thunderclap headache).

However, case studies indicate that abrupt onset of a headache may not occur in up to 50% of cases (Linn et al., 1994). The time from onset to peak severity maybe several minutes (Linn et al., 1998). This may explain why case studies of SAH find initial rates of misdiagnosis of between 23% and 51%.3 The patients who are misdiagnosed tend to be less unwell, usually with no transient loss of consciousness and have no neurological signs. The commonest incorrect diagnosis is migraine (21% of misdiagnoses).

Even if an explosive headache is the only presenting symptom only 10% will have SAH (Edlow & Caplan, 2000). Nevertheless, it is obviously worth referring and investigating 10 patients (or even 100) to detect one SAH at an early stage. The more difficult question is whether to refer all patients with the ‘worst headache ever’ even if it comes on over several minutes. One US study found that 20 out of 107 patients presenting to an ED department with the ‘worst ever’ headache had SAH (Morgenstern et al., 1998). However, one community study found that 9.1% of the population reported at least one ‘almost unbearable’ headache in the previous year (Newland et al., 1978). Potentially, the whole population of headache sufferers will have their ‘worst ever headache’ at some stage. While the presence of neck stiffness makes the diagnosis more likely one unpublished study from Durham found neck stiffness was absent in 64% of cases of SAH. Nuchal rigidity may take 3 to 12 hours to develop, and maybe absent in small bleeds.

Duration of headache can be helpful in excluding SAH. It is generally accepted that the headache from a SAH usually lasts days and certainly does not last less than an hour (Davenport, 2002).

The conclusion of the studies is that clinical suspicion of a SAH needs to be set at a low level because up to half the cases may not present with a typical thunderclap headache. Nevertheless, close questioning about the onset of a headache is critical to the assessment of any new headache and details about onset should be recorded. Brief duration headaches that resolve are not due to SAH but general practitioners should have a low threshold for urgently referring sudden onset headaches or ‘first and worst’ headaches even if the onset is over a few minutes and there are no neurological signs such as nuchal rigidity.

Dr Walmesley should have recognized that SAH was a significant possibility (the chance of SAH was probably about 1 in 10) and admitted Hannah urgently to hospital for a CT head scan and a lumbar puncture. The second general practitioner Dr Palmer should have retaken the crucial history about the onset of the symptoms and realized that a SAH may have been missed and again admitted her urgently. Dr Palmer should not have been reassured by the earlier negative assessment ‘in hospital’.

Legal comment

The duty of a GP is to act in accordance with a responsible body of medical opinion skilled in that particular field. The expert in this case believes that both Dr Walmesley and Dr Palmer have failed to do so. He says they should both have recognized the possibility of SAH and arranged for further investigations.

However, a lawyer acting for either of those two doctors may wish to explore whether there is another responsible body of medical opinion which takes a different view. Maybe it could be argued for Dr Walmesley that the lack of neurological symptoms and the history of a gradual progression of the headache made it reasonable for him to diagnose migraine at the first consultation? (No doubt it will be more difficult for an expert to argue this for Dr Palmer who saw Hannah two days later when she had a persistent headache and some neck ache and stiffness.)

The lawyers for the doctors will also want to explore the consequences of the failure to refer Hannah for investigation. Was it more likely than not that Hannah could have been saved if referred for a CT scan? If so, then Hannah's family will have a valid claim for bereavement damages of £11 800. The administrators of her estate could sue for the pain and suffering she endured before her death, due to the GP's negligence. If Hannah has children, they will have a potentially substantial claim as her dependents for the loss of care and support from their mother during their childhood. The family will also be entitled to compensation representing her financial contribution to the household.

Key learning points

Specific to the case

The headache in a SAH may take several minutes to reach peak severity.
The history of the onset is critical to making the diagnosis.
There may be no physical signs.
Consider SAH in all cases of sudden onset severe headache and the ‘first and worst’ headache that lasts more than an hour.

General points

General practitioners should not rely on a diagnosis of a headache made by the clinician who saw the patient first. Even in hospital the assessing doctor may have been inexperienced. Always retake the history, particularly of the onset.

References and further reading

Davenport R (2002) Acute headaches in the emergency department. J Neurol Neurosurg Psychiatry 72: 33ii–37.

Edlow JA, Caplan LR (2000) Avoiding pitfalls in the diagnosis of subarachnoid hemorrhage. N Engl J Med 342: 29–36.

Hankey GJ, Nelson MR (2009) Easily missed: Subarachnoid haemorrhage. BMJ 339: 569–70.

Linn FHH, Wijdicks EFM, van Gijn J, Weerdesteyn-van Vliet FAC, van der Graaf Y, Bartelds AIM (1994) Prospective study of sentinel headache in aneurysmal subarachnoid haemorrhage. The Lancet 344: 590–3.

Linn FHH, Rinkel GJE, Algra A, van Gijn J (1998) Headache characteristics in subarachnoid haemorrhage and benign thunderclap headache. J Neurol Neurosurg Psychiatry 65: 791–3.

Morgenstern LB, Luna-Gonzales H, Huber JC, Jr, et al. (1998) Worst headache and subarachnoid hemorrhage: prospective, modern computed tomography and spinal fluid analysis. Annals of Emergency Medicine 32: 297–304.

Newland C, Illis LS, Robinson PK, Batchelor BG, Waters WE (1978) A survey of headache in an English city. Res Clin Stud Headache 5: 1–20.

Case 3 A woman with chest pain

Brenda was 40 when she consulted Dr Marks with a cough and right submammary pain. Two weeks previously she had recently returned from a month's holiday in Brazil where she had been treated for a chest infection with two courses of antibiotics. Dr Marks noted that Brenda felt slightly short of breath. She was taking the combined oral contraceptive (COCP), was a nonsmoker, and had a body mass index of 24.1. Her blood pressure was 159/95 and there were some crepitations at the right base. Dr Marks made a diagnosis of ‘pleurisy’ without recording any further detail about the pain.

What other information would you obtain?

Brenda told Dr Marks that she had had chest radiographs in Brazil that showed a lung infection. There was no observable calf or thigh swelling.

What would be your differential diagnosis and how would you discriminate between them?

A few days later Brenda became increasingly troubled by her chest pain and shortness of breath and went to A&E. She was admitted to hospital. Both the junior doctors who initially saw her suspected a pulmonary embolus. Brenda told them her pain had first occurred shortly after her flight to Brazil, and that her leg had swelled up at the same time. The diagnosis was confirmed by CT pulmonary angiogram, which showed widespread multiple pulmonary emboli.

Brenda sued Dr Marks for failing to consider the possibility of a pulmonary embolus.

Do you think her claim will succeed?

Expert opinion

Chest pain and complaints of breathlessness are both common in general practice. In this case the pain was unilateral and pleuritic in nature (sharp, well localized in time and position and worse on inspiration). It was associated with breathlessness. The differential diagnosis in this case included:

pneumonia
pulmonary infarction (a consequence of pulmonary embolus)
lung abscess
bronchiectasis
malignancy, e.g. bronchogenic carcinoma
pneumothorax
asbestos pleural disease, including mesothelioma.

The reason Dr Marks made the diagnosis of ‘pleurisy’ (presumably infective) was probably Brenda's own account of having been diagnosed with a chest infection on holiday. Possibly because of this, Dr Marks did not record a thorough history of the timing and onset of the symptoms some 6 weeks earlier. He does not appear to have enquired about associated features such as haemoptysis. He prematurely fixed on the diagnosis of ‘pleurisy’ and did not consider a wide differential diagnosis to explain what the underlying cause of Brenda's chest pain was.

Pulmonary emboli are relatively rare in general practice. The quoted incidence of pulmonary emboli is 25 per 100 000 per year. Thus a general practitioner with a list of 2000 will see one case every two years. According to Dalen and Master (2002):

It is well-recognized that the signs and symptoms of pulmonary embolism are nonspecific and that, as a result, the clinical recognition of pulmonary embolism is notoriously inaccurate. The lack of sensitivity of the clinical diagnosis of pulmonary embolism is evident from postmortem studies demonstrating that the majority of cases of pulmonary embolism detected postmortem were not diagnosed (or treated) prior to death.

It is estimated in the USA that only 26% of pulmonary emboli are diagnosed and treated. 74% of cases are undiagnosed or diagnosed only after death (Dalen and Master, 2002).

In the Prospective Investigation of Pulmonary Embolism Diagnosis study (PIOPED) of the patients who survived long enough to have their pulmonary emboli diagnosed by pulmonary angiography (Stein & Henry, 1997):

65% presented with ‘classical’ symptoms of pulmonary infarction (death of lung tissue due to the clot). These probably represent the smallest volume emboli that fragment and impact in the distal bronchial tree (Ryu et al., 1998).
22% presented with isolated breathlessness and no other features. Angiographic studies suggest that these are due to intermediate sized emboli that lodge in the pulmonary tree without producing infarction but significantly affect gas exchange and cause some degree of, possibly transitory, pulmonary hypertension (Ryu et al., 1998).
8% presented with circulatory collapse. These are the large emboli that break off from the deep pelvic veins and impact at the principle trunks of the pulmonary artery. They present with the features described by Virchow: severe acute breathlessness, poorly localized chest pain and distress (probably from right ventricular distension and ischaemia), syncope or pre syncope, followed by recovery (if the patient survives) (Ryu et al., 1998).
5% had no symptoms at all and were diagnosed on the basis of clinical suspicion after an abnormal chest X-ray (Ryu et al., 1998).

Most clinicians (and all clinical prediction rules) rely significantly on the presence of risk factors to alert the doctor to the possibility of pulmonary embolism in clinically unobvious cases.

The British Thoracic Guideline for the management of suspected pulmonary embolism (June 2003) gives a list of risk factors and divides them into major risk factors (increasing the risk by a factor of 5 to 20) and minor risk factors (increasing risk by a factor of 2 to 4) (British Thoracic Society, 2003). This is reproduced as Case Table 3.1.

Table 3.1 Risk factors for venous thromboembolism.

Major risk factors (relative risk 5–20)
Surgerya	Major abdominal/pelvic surgery Hip/knee replacement Postoperative intensive care
Obstetrics	Late pregnancy Caesarian section Puerperium
Lower limb problems	Fracture Varicose veins
Malignancy	Abdominal/pelvic Advanced/metastatic
Reduced mobility	Hospitalization Institutional care
Miscellaneous	Previous VTE
Minor risk factors (relative risk 2–4)
Cardiovascular	Congenital heart disease Congestive cardiac failure Hypertension Superficial venous thrombosis Indwelling central vein catheter
Oestrogens	Oral contraceptive Hormone replacement therapy
Miscellaneous	COPD Neurological disability Thrombotic disorders Long-distance sedentary travel Obesity Otherb
aWhere appropriate prophylaxis is used, relative risk is much lower. bInflammatory bowel disease, nephrotic syndrome, chronic dialysis, myeloproliferative disorders, paroxysmal nocturnal haemoglobinuria, Behçet's disease.

A recent history of long-haul air travel is widely considered to increase the risk of venous thromboembolism. The evidence is effectively reviewed in the NHS Clinical knowledge summaries (NHS, 2011). In reality only a small proportion (about 1 in 4000–5000 flights) of air travellers have a deep vein thrombosis as a result. However, use of the combined oral contraceptive, obesity, Factor V Leiden mutation and extreme height increase this risk. The World Health Organization data estimated the odds ratio for the use of the COCP when flying to be 40. Therefore the prior probability of Brenda having a PE could have been as high as 1%.

The difficulty arises when patients do not present with the ‘classical symptoms’ of pulmonary embolism result from pulmonary infarction:

pleuritic chest pain (sharp, well localized chest pain occurring on breathing in)
cough
haemoptysis
tachycardia
tachypnoea.

In this case Brenda does not seem to have had haemoptysis, although it may not have been recorded. However the persistence of symptoms of chest pain and shortness of breath over a six-week period was rather atypical for a previous diagnosis of chest infection. While pulmonary embolus is an uncommon diagnosis in general practice it should have been considered because of this combination of symptoms.

Thromboembolic disease is one of the top three medical causes of litigation against GPs. Cases tend to involve important disputes of fact about what history was given by the patient, or ought to have been elicited by a competent GP. It is also common for there to be an allegation that the GP simply failed to think of the possibility of thromboembolic disease. Both of these criticisms potentially apply in this case.

Should scoring systems be used to define risk? The Wells Rule (Wells et al., 2000) is quite well known and may even be adopted for local care pathways between primary and secondary care. However it should be borne in mind that these scoring systems have been derived from populations presenting in secondary care in whom pulmonary embolus (or deep vein thrombosis as the case may be) have already been considered as a diagnosis. They have therefore not been validated for use in primary care. The AMUSE study has validated a Clinical Prediction Rule for use in primary care but relies heavily upon and requires a D Dimer result (Buller et al., 2009).

Legal comment

It is suggested by the legal expert that Dr Marks failed in his duty because he did not elicit an adequate history. If he had elicited information that Brenda's pain came on shortly after her flight to Brazil and that her leg had swelled up at the same time, then he would have surely been alerted to the potential diagnosis of pulmonary embolism.

The number of potential diagnoses for chest pain and breathlessness make it incumbent on a GP to explore the patient's history carefully. It would therefore be difficult for a lawyer to obtain a contrary expert opinion.

We are not told what the outcome was for Brenda. Assuming successful treatment, she will have a claim for any pain or suffering before her problem was correctly diagnosed and treated. Since this was only a few days after the consultation, it would seem that this is a relatively small claim against Dr Marks, perhaps £2000–£3000.

Key learning points

Specific to the case

Always include pulmonary embolus in the differential diagnosis of pleuritic chest pain.
In the absence of clear pointers to the diagnosis of pulmonary embolus, a key feature is the plausibility of the alternative diagnosis.
If, as in this case, the diagnosis that was made was not plausible, there may be difficulty in defending the medical management.

General points

It is good practice to always ‘second guess’ a presumed diagnosis.
Is the presumed diagnosis the most likely?
Is there another diagnosis that one ‘cannot afford to miss’?

References and further reading

British Thoracic Society Standards of Care Committee Pulmonary Embolism Guideline Development Group (2003) British Thoracic Society guidelines for the management of suspected acute pulmonary embolism. Thorax 58: 470–83.

Buller HR, Cate-Hoek AJ, Hoes AW, et al. (2009) Safely ruling out deep venous thrombosis in primary care Annals of Internal Medicine 150: 229–35.

Dalen M, Master M (2002) Pulmonary embolism: What have we learnt since Virchow? Natural history, pathophysiology and diagnosis. Chest 122: 1440–56.

Meyer G, Roy P-M, Gilberg S, Perrier A (2010) Easily missed? Pulmonary embolism. BMJ 340: c1421.

NHS (2011) Clinical knowledge summaries http://www.cks.nhs.uk/dvt_prevention_for_travellers/evidence/supporting_evidence/risk_of_travel_related_dvt#-390015

Ryu J, Jay H, Olson EJ, Pellikka PA (1998) Clinical recognition of pulmonary embolism: problem of unrecognized and asymptomatic cases. Mayo Clinic Proceedings 73: 873–9.

Stein PD, Henry J (1997) Clinical characteristics of patients with acute pulmonary embolism stratified according to their presenting symptoms. Chest 112: 974–80.

Wells PS, Anderson DR, Rodger M, Ginsberg JS, Kearon C, Gent M, et al. (2000) Derivation of a simple clinical model to categorize patients probability of pulmonary embolism: increasing the models utility with the SimpliRED D-dimer. Thromb Haemost 83: 416–20.

Case 4 A dizzy man

Bernard was 60 years old when he was visited by an Out of Hours general practitioner, Dr Carter, on a weekend evening. He stated that he woke that day, turned over in bed and felt that the room was spinning. He had vomited twice in the bed. During the morning he felt unsteady on his feet and had become very anxious because he lived on his own and was unsure how he would cope. He had felt nonspecifically unwell for a couple of days with a sore throat and malaise. He had made himself some hot lemon with whisky that evening but had been unable to drink it. He had had a coronary angioplasty and stent two years earlier and was on a statin, aspirin and a beta blocker.

What would you do now?

Dr Carter established that the vertigo was intermittently so severe that Bernard had difficulty standing. He had difficulty walking but managed if he kept his head still. He felt comfortable once he was sitting still or lying down. On direct questioning he had a slight headache, was not aware of any weakness, had a sore throat and a hoarse voice.

What would be your differential diagnosis and how would you discriminate between them?

Dr Carter noted that the man looked well but anxious, had a temperature of 37.4 °C, pulse 64/min regular and a BP 174/92 mmHg. Dr Carter suggested doing a Hallpike test but Bernard was anxious about provoking the vertigo. Dr Carter recorded that Bernard probably had a viral labyrinthitis and pharyngitis and prescribed stemetil and paracetamol.

Bernard's son visited him the next day and found his father was unwell with a high fever and a severe cough. He took him to the local ED department where a chest X-ray demonstrated a pneumonia. He was admitted onto a general medical ward. The following day one of the nurses noted that Bernard was choking when drinking fluids. Further assessment by the medical registrar demonstrated that the man had a hoarse voice, a weak cough and that palatal movements were asymmetric on saying ‘aah’ (the uvula moving to the left). Bernard regurgitated water on drinking. Light touch was subjectively different on the right cheek to the left, there was a right Horner's syndrome, sustained nystagmus and he could not walk unaided. A cranial MRI scan demonstrated patchy infarction of the right lateral medulla and inferior cerebellum. The diagnosis was made of an aspiration pneumonia secondary to a brainstem stroke. Bernard was unwell and dehydrated and his neurological deficit extended. A further MRI showed more extensive dorsolateral medullary infarction. Bernard made only a partial recovery.

Bernard sued Dr Carter for failing to consider the possibility of a stroke and for omitting to check for dysphagia, and sued the hospital for initially missing the stroke and treating him for pneumonia.

Do you think his claim will succeed?

Expert opinion

Vertigo is a difficult presentation for a general practitioner and requires careful assessment. It is relatively common. A full time general practitioner can expect to see 10–20 cases of acute vertigo per year (McCormick et al., 1995). The vast majority will be due to malfunctions of the labyrinth – vertigo caused by a peripheral lesion. Roughly 40% will be due to benign paroxysmal positional vertigo (BPPV), 40% will be due to acute vestibular neuritis and a significant proportion of the rest may have vestibular migraine (Barraclough & Bronstein, 2009). However, a critical aspect of the general practitioner's assessment must be to distinguish the tiny number of cases of vertigo due to brainstem lesions (mostly brainstem strokes) from the large number of peripheral causes of vertigo.

In this case there were a number of features that should have alerted Dr Carter to consider the possibility of a central (brain stem) cause rather than a peripheral cause. At 60 Bernard was in the age range where cerebrovascular events are not uncommon. It was a concern that he had difficulty walking and drinking and that his voice was hoarse. The latter were suggestive of dysphagia and dysphonia.

Dr Carter should have established a few additional features. It was not clear from the history if the vertigo was positional or sustained. He should have seen if the man could drink a glass of water. If he could drink then significant dysphagia was unlikely. He should have seen him walk and checked that there was no facial, hand or arm weakness. All these are quite easy tests and if normal, make a brainstem stroke unlikely. If the tests were abnormal the man should have been referred immediately into hospital with a possible stroke (NICE, July 2008, Clinical guidelines on stroke).

Because structures in the brainstem are closely packed together, vertigo in the absence of any other cranial nerve features (such as diplopia, facial weakness, facial numbness, dysphagia, dysphonia) or long tract symptoms (such as weakness or numbness of the limbs) is unlikely to be due to a central cause. Fewer than 1% of 407 patients with posterior circulation strokes in the New England Medical Centre Posterior Circulation Registry (NEMC-PCR) presented with a single isolated symptom (Caplan et al., 2004). Another study of 1666 patients aged over 44 presenting to a US emergency department with ‘dizziness’ found 53 (3.6%) were due to a stroke or a TIA. In patients with ‘dizziness’ without other symptoms or signs only 0.7% had had a stroke or TIA (Kerber et al., 2006).

Dr Carter attempted to make a positive diagnosis but was negligent in not checking that there were no other neurological signs – particularly dysphagia, dysphonia, palatal weakness, facial weakness or numbness or ataxia.

Legal comment

The expert says it was negligent of Dr Carter not to have checked Bernard for neurological signs. If he had done so, he would have probably found abnormalities which would have led him to admit Bernard to hospital.

As it was, in the hours before his son came to visit, it seems that Bernard developed an aspiration pneumonia because of his dysphagia. That pneumonia may have worsened the neurological deficit which will now affect Bernard for the rest of his life. He has sued both the GP and the hospital.

As well as looking at breach of duty the lawyers for each will take expert opinions on the cause of the additional neurological deficit. Maybe after all, despite any failings by either party, that deficit is entirely unrelated. It may have been simply a continuation of the brain stem infarction. In that case, Bernard's claim will fail.

But if Bernard's worse outcome is probably the result of the aspiration pneumonia, then Dr Carter's MDO will have to pay Bernard compensation to reflect his additional disabilities.

The MDO may seek a contribution from the hospital. If there is expert medical evidence to suggest that it too was negligent then there is likely to be negotiation between the lawyers for those two defendants.

Bernard's compensation will be assessed according to a comparison between the likely outcome if he had been properly managed, and the actual outcome. The costs of any additional care now required because of the additional deficit will be calculated by reference to Bernard's life expectancy.

Key learning points

Specific to the case

In cases of acute vertigo it is essential to consider the possibility of a brainstem stroke.
Check the onset and duration of the vertigo and whether it is positional or not. Check that there is no headache or hearing loss.
On examination check eye movements and nystagmus (should be horizontal only and ideally whether the pupils are equal. Check there is no facial weakness, dyaphagia or dysphonia. Check for gait or limb ataxia and look in the ears for a cholesteatoma.

General points

It is always necessary to be wary of subtle details that ‘do not fit’.

References and further reading

Barraclough K, Bronstein A (2009) Diagnosis in general practice: vertigo, BMJ 339: 749–52.

Caplan L, Chung C-S, Wityk R, et al. (2004) New England Medical Center Posterior Circulation Registry. Annals of Neurology 56: 389–98.

Kerber KA, Brown DL, Lisabeth LD, Smith MA, Morgenstern LB (2006) Stroke among patients with dizziness, vertigo, and imbalance in the emergency department: a population-based study. Stroke 37: 2484–7.

McCormick A, Fleming D, Charlton J (1995) Morbidity Statistics from General Practice: Fourth National Study, 1991–1992. London, Office of Population Censuses and Surveys.

NICE (2008) Diagnosis and initial management of acute stroke and transient ischaemic attack (TIA), Clinical guidelines CG68, July 2008.