Don t use antipsychotics routinely to treat agitation and aggression in people with dementia

Transcription

1 CHANGE PAGE Don t use antipsychotics routinely to treat agitation and aggression in people with dementia Anne Corbett, 1 Alistair Burns, 2 Clive Ballard 1 1 Wolfson Centre for Age-Related Diseases, King s College London, London, UK 2 University of Manchester, Manchester M13 9NT, UK Correspondence to: C Ballard clive.ballard@kcl.ac.uk Cite this as: BMJ 2014;349:g6420 doi: /bmj.g6420 Change Page aims to alert clinicians to the immediate need for a change in practice to make it consistent with current evidence. The series advisers are Sera Tort, clinical editor, and David Tovey, editor in chief, the Cochrane Library. We welcome any suggestions for future articles ( us at practice@bmj.com). thebmj.com Previous articles in this series ЖЖAvoid prescribing antibiotics in acute rhinosinusitis (BMJ 2014;349:g5703) ЖЖAvoid surgery as first line treatment for nonspecific low back pain (BMJ 2014;349:g4214) ЖЖAdminister tranexamic acid early to injured patients at risk of substantial bleeding (BMJ 2012;345:e7133) ЖЖDon t use aspirin for primary prevention of cardiovascular disease (BMJ 2010;340:c1805) ЖЖPrescribe systemic corticosteroids in acute asthma (BMJ 2009;338:b1234) Behavioural and psychological symptoms of dementia such as agitation and aggression are commonly treated with atypical antipsychotic drugs, which are associated with severe side effects. 1 However, there is increasing evidence of potential harms associated with use of these drugs in people with dementia, and guidelines increasingly recommend restricting their use. In Europe, only risperidone is currently licensed for use in dementia and only for up to six weeks in patients with severe aggression, defined as causing risk or severe distress, which has not responded to other treatments. No antipsychotics are approved for this patient group in the United States. Best practice guidelines, including National Institute for Health and Care Excellence guidelines in the United Kingdom and American Psychiatric Association guidelines in the US, are similar. However, they do not distinguish between individual atypical antipsychotics and recommend a maximum treatment period of 12 weeks, except in exceptional circumstances. 2 3 The use of these drugs in people with dementia has decreased worldwide in the past decade. For example, a 2012 audit in the UK indicated a 50% reduction in prescriptions, although 16% of people with dementia continued to receive antipsychotic treatment. 1 Use is mostly off-licence, either prescriptions of antipsychotics such as quetiapine, olanzapine, aripiprazole, and haloperidol without a licence indication or the use of risperidone outside the strict licence indication. In addition, prescriptions often exceed the six week period specified in the licence indication for risperidone or the 12 week prescription period recommended in all of the major best practice guidelines, with 62% of people receiving atypical antipsychotics for six months or more. 1 Although current guidance promotes more judicious use of these drugs, with regular review of prescriptions, the THE BOTTOM LINE Despite continued use of antipsychotics to treat agitation and aggression in people with dementia, there is limited evidence of clinically meaningful benefit The potential harms of antipsychotic use (including increased cerebrovascular events and mortality) outweigh the benefits Risperidone is the only recommended antipsychotic, and should be used only in people with dementia who have pre-existing psychotic disorders or severe aggression. It should be prescribed for no more than 12 weeks Careful monitoring is the best practice alternative, with evidence that alternative treatments including analgesia and non-drug based approaches provide effective options SOURCES AND SELECTION CRITERIA This article is based on an updated review of the literature published in systematic reviews. We searched the electronic databases PubMed, Embase, and the Cochrane Library. We evaluated individual randomised controlled trials from systematic reviews to calculate the specific attributable risk of key adverse events associated with risperidone in people with Alzheimer s disease. Studies were selected according to the quality and robustness of the study design, with a focus on systematic reviews. guidelines are difficult to interpret, particularly as thresholds for severity are unclear. The evidence for change Modest short term clinical benefit Systematic reviews have analysed the clinical effectiveness of antipsychotics in people with dementia. These analyses are based on 18 placebo controlled randomised trials, most of which were conducted over a week period. However, many of these trials have not been published in full. The best evidence base exists for risperidone, with five fully published good quality randomised controlled trials (RCTs) and a total of 1761 participants. 4 5 In these trials, full data were presented for aggression and psychosis, but the information on non-aggressive symptoms of agitation such as restlessness, wandering, and shouting was incomplete, which may have led to bias. Adverse events were comprehensively reported in all five studies. 5 A meta-analysis reported a significant advantage for risperidone in the treatment of aggression ( 0.84 points on the BEHAV-AD rating scale (95% confidence interval 1.28 to 0.40) at a dose of 1 mg and 1.5 points ( 2.05 to 0.95) at 2 mg). 4 This threshold of change indicates a statistically significant difference but only borderline clinically meaningful benefit at the 2 mg dose. Evidence of clinically meaningful benefit is restricted to aggression none was seen for non-aggressive symptoms of agitation. 4 6 Although outside the focus of this paper, effectiveness is even more limited for the treatment of psychosis. Statistically significant benefit but no clinically meaningful benefit was seen only at 1 mg risperidone (BEHAV-AD mean difference 0.14, 0.25 to 0.03). The evidence for benefit is not equal for all anti-psychotics. For example, a meta-analysis of published trials found no evidence that quetiapine confers benefit in the treatment of behavioural and psychological symptoms of dementia. 4 The high placebo response rates (45% v 55% for risperidone) across all published trials indicate that benefit is often related to general benefits of good clinical practice, clinical review, treatment of comorbidity, and improved social interaction as a result of trial involvement November 2014 the bmj

2 Adverse effects The evidence of modest clinical effectiveness must be b alanced against the considerable risk of adverse events. A systematic review of 15 RCTs of antipsychotics in people with Alzheimer s disease reported a 1% attributable risk of mortality over 12 weeks of treatment (risk difference 0.01, to 0.02; P=0.01). A 12 month double blind RCT examining antipsychotic discontinuation in people with Alzheimer s disease, with follow-up of participants for up to five years, found a significant reduction in mortality associated with discontinuation (hazard ratio 0.58, 0.36 to 0.92), with a risk difference in mortality of 29% after 36 months. 4 A subsequent meta-analysis of all 18 RCTs of atypical antipsychotics for the treatment of behavioural and psychological symptoms of dementia also found a clinically significant acceleration of cognitive decline in people taking antipsychotics (mini-mental state examination mean difference 0.73 points, 0.38 to 1.09; P<0.0001) over 12 weeks. Reporting of adverse events is most complete for r is-peridone full data are not available for other antipsychotics because of the large number of unpublished studies. Of particular concern, meta-analyses have established a threefold increased risk of cerebrovascular events in p eople with Alzheimer s disease who take risperidone compared with those taking placebo (odds ratio 3.43, 1.60 to 7.32; Z=3.18; P=0.001), with a risk difference of 3.1% versus 1.0% in a pooled analysis Another important risk is an increased frequency of extrapyramidal symptoms (risk difference by meta-analysis 0.06, 0.03 to 0.09 over 12 weeks). Peripheral oedema, sedation, prolonged Q-Tc interval, infections, and abnormal gait have also been highlighted as potential problems. 6 Longer term use and discontinuation Evidence on the long term use of antipsychotics is emerging. A handful of recent trials have evaluated a range of antipsychotics over periods of six months and longer. Only one RCT directly evaluated the impact of an antipsychotic (quetiapine) on agitation, reporting no benefit compared with placebo over six months. 13 Overall, these studies have reported no benefit or very modest benefit in the treatment of the behavioural and psychological symptoms of dementia over six to 12 months. The exception is one recent trial that compared the impact of withdrawal versus continuation of haloperidol, which indicated ongoing benefit of continuation in people who had initially responded to this treatment. 14 A Cochrane review of nine randomised placebo controlled trials concludes that long term prescriptions of antipsychotics can be discontinued without a de trimental effect on neuropsychiatric symptoms. 15 Adverse events are also more marked with longer term use, with one RCT reporting 59% mortality compared with 30% in a placebo group after 36 months In another nine month RCT of 421 people with Alzheimer s disease, 18% of patients receiving risperidone withdrew from the trial owing to adverse events compared with 5% of those receiving placebo. 12 Alternatives to antipsychotics No direct alternatives to antipsychotics are available for the treatment of aggression or psychosis. Results of initial studies with several compounds including carbamazepine, mirtazapine, and prazosin are encouraging, but there is insufficient evidence to recommend their use in clinical practice. 18 Other compounds with similar chemical actions, such as benzodiazepines (other than mirtazapine) are possible candidates, but no evidence from clinical trials is currently available. A recent RCT of citalopram that built on preliminary studies indicated modest but significant benefit in the treatment of agitation at 30 mg per day, but the impact on prolongation of the Q-Tc interval was considered too great to recommend this as a treatment approach. 19 A trial is currently examining the efficacy and safety of a lower dose. Pain is a common underlying cause of agitation in dementia, and a recent RCT in 352 patients reported a 17% improvement in agitation after stepped treatment with analge sics, similar to the benefit seen with anti-psychotics. 20 Such treatment is not a direct alternative to antipsychotics but plays an important part in managing and preventing agitation, and may reduce the need for antipsychotics. Although there is a substantial evidence base for nonpharmacological approaches to the treatment of behavioural and psychological symptoms of dementia, a more modest number of studies focus on specific interventions for individual symptoms. A recent systematic review reported significant benefit in four of six RCTs that evaluated personalised activities, such as personalised social interaction and reminiscence therapy (box 1) for the treatment of agitation, with a median overall standardised effect size of The standardised effect size is calculated by dividing the mean difference between treatments for an outcome by the standard deviation of that measure across the sample. A standardised effect size of is usually considered to represent a moderate and clinically meaningful benefit. For context, the standardised effect size for risperidone for the treatment of aggression is about 0.2. However, such studies are usually smaller than trials on drug based interventions, and the comparison is usually with treatment as usual rather than a more specific control intervention. Such a design is likely to lead to a higher apparent effect size. The evidence supports the use of these approaches as an effective set of first line interventions, although their impact on severe intractable symptoms may be more limited. They are not an alternative to drug based treatment, but they are the most appropriate initial course of action and may reduce the need for antipsychotic use. Barriers to change The behavioural and psychological symptoms of dementia present a substantial treatment challenge for physicians, particularly with the current lack of a licensed drug based alternative and continuing ambiguity in thresholds for prescribing. Doctors often experience a considerable pressure to prescribe as a result of the distress that these symptoms cause to individuals, their families, and care staff in residential care settings. This complex challenge requires discussion with staff and family members so that they can air their concerns and to ensure that they fully understand the minimal clinical benefit and possible harms associated with antipsychotic treatment. the bmj 15 November

3 Box 1 Non-drug treatments for use as first line approaches for behavioural and psychological symptoms of dementia 21 The Seattle protocols A personalised approach to care planning that focuses on increasing physical and cognitive activities. This involves meeting with the patient and family to discuss the patient s interests and hobbies. A care plan can then be developed to ensure that the patient performs at least 60 minutes of enjoyable activities a week that also promote physical activity, such as gardening, gentle movement and balance exercises, or walking. Manuals for the Seattle protocols are available online. Personalised social interaction This approach focuses on ensuring that the patient regularly spends time interacting with other people. This involves a detailed assessment of the patient s medical and personal needs as well as ability to communicate, move around, and participate in activities. This is then used to create a tailored social interaction programme involving care staff, family members, friends, and others who are available to spend time with the patient. Brief psychosocial treatment This is a simplified version of the personalised social interaction approach above, which can be delivered by a family member. The goal is to provide 10 minutes of social interaction each day, which can involve talking about topics that interest the person (such as asking about past work, family, or hobbies), looking at photographs, or doing a puzzle together. Support is available through an initial 30 minute planning session with a trainer and brief weekly follow-up telephone calls. Simulated presence therapy This simple approach uses audio or video recordings from family members, such as recorded conversations, singing, or home videos, which are played to the patient in the care setting. The goal is to simulate social contact and to prompt memories and conversation. The recordings could also be used as part of an enjoyable activity, such as looking through old photographs. This approach works best when working closely with families to agree on the most suitable recordings on the basis of the patient s interests. NEST approach This more in-depth approach is suitable for use by an occupational therapist. It involves evaluating patients needs, and the environment, stimulation, and techniques (NEST) that they require; this is then used to select different activities they can engage in. Identification of pain is included in best practice guidelines for the management of agitation, but it is rarely applied systematically, probably because of the lack of practical guidance. Furthermore, there is currently limited provision of simple first line evidence based alternatives, such as those outlined in box 1, both by primary care and multidisciplinary specialist teams. Guidelines and training programmes exist, but lack of standardised or specific training may result in interventions that provide monitoring or advice for staff but do not deliver simple evidence based treatments. This may in turn lead to the m isperception that non-drug based approaches are in effective. Training for some of these approaches, such as p romoting enjoyable activities, is simple and can be delivered in a half day session. Box 2 Current guidelines for use of risperidone in people with behavioural and psychological symptoms of dementia 22 Risperidone is the only recommended treatment Maximum length of treatment is 12 weeks It should be used only in patients with severe aggression that is causing risk or severe distress and where alternative approaches have failed The decision to prescribe should be made only after a careful risk assessment, particularly cerebrovascular risk (taking into account hypertension, diabetes, smoking, atrial fibrillation, and previous stroke) Report all suspected side effects to the appropriate regulatory body How should we change our practice? We propose that in line with current European licence indications in patients with severe aggression that have not responded to alternative treatment approaches, risperidone should be used for a maximum of six weeks (box 2). The maximum treatment period could be extended to 12 weeks to be consistent with international best practice guidelines. 2 3 Other antipsychotics should not be prescribed for people with dementia. Exceptions should be made only when the individual has a pre-existing psychotic disorder independent of a diagnosis of dementia. For most people with dementia, the risk of harm of antipsychotic treatment outweighs the likelihood of benefit. Careful ongoing monitoring and support, without the prescription of antipsychotics, is important because clinical trials show that 40-45% of people will experience clinically meaningful benefit from the generic components of good clinical practice. 4 6 For people who have already been taking antipsychotics for more than 12 weeks, these drugs should be reviewed and discontinued unless at least two previous attempts at discontinuation have led to severe exacerbation of symptoms. In 70% of people, no worsening of symptoms is seen after withdrawal of antipsychotics. 16 Carers may require support over the first four weeks because of anxiety about worsening of symptoms, and monitoring is recommended for up to three months. When possible it can also be helpful to recommend simple non-drug based interventions such as social interaction (box 1). Non-drug based treatments and improved pain management provide an appropriate first line treatment for agitation and aggression in many people, although these are not a direct alternative to antipsychotics. Routinely assess for pain in all people experiencing behavioural and psychological symptoms of dementia. If there is any suggestion that pain may be contributing, start a trial of treatment for pain through stepped analgesia, starting with paracetamol in the absence of ongoing analgesia. A first line approach should also include prompt access to simple effective evidence based non-drug treatments, such as personalised activities with social interaction (box 1). If these approaches do not help, refer the patient to a clinical psychologist to assess whether a short term prescription of an antipsychotic is warranted or whether the risk of prescribing outweighs any benefit. In this case, ongoing psychosocial interventions in combination with close monitoring would be more beneficial. CB and AC would like to thank the National Institute for Health Research (NIHR) Mental Health Biomedical Research Centre and Dementia Unit at South London and Maudsley NHS Foundation Trust and King s College London for supporting this work. Contributors: All authors contributed equally to the content of this article. CB is guarantor. Competing interests: We have read and understood BMJ policy on declaration of interests and declare the following interests: AC has received consultancy and speaking honorariums from Acadia, Lundbeck, and Novartis. AB has received speaking honorariums from Healthcare Education Services, Hammond Care, and Betrinac; writes court reports; is chair of the DMEC for an RCT funded by Acadia Pharmaceuticals; receives royalties as editor in chief of the International Journal of Geriatric Psychiatry; and is employed as clinical director for dementia, NHS England. CB has received consultancy and speakers fees from Novartis, Bristol Myer Squibb, Acadia, Otsuka, Orion, Napp, Roche, and Heptares; has received grant funding from Acadia and research contract funding from Lundbeck and Takeda, and will be a member of an advisory board for Otsuka November 2014 the bmj

4 Provenance and peer review: Commissioned; externally peer reviewed. 1 Barnes TR, Banerjee S, Collins N, Treloar A, McIntyre SM, Paton C. Antipsychotics in dementia: prevalence and quality of antipsychotic drug prescribing in UK mental health services. Br J Psychiatry 2012;201: Rabins PV, Blacker D, Rovner BW, Rummans T, Schneider LS, Tariot PN, et al. American Psychiatric Association practice guideline for the treatment of patients with Alzheimer s disease and other dementias. Second edition. Am J Psychiatry 2007;164(12 suppl): National Institute for Health and Care Excellence. Dementia: supporting people with dementia and their carers in health and social care Schneider LS, Dagerman K, Insel PS. Efficacy and adverse effects of atypical antipsychotics for dementia: meta-analysis of randomized, placebo-controlled trials. Am J Geriatr Psychiatry 2006;14: Ballard C, Waite J. The effectiveness of atypical antipsychotics for the treatment of aggression and psychosis in Alzheimer s disease. Cochrane Database Syst Rev 2006;1:CD Ballard C, Howard R. Neuroleptic drugs in dementia: benefits and harm. Nat Rev Neurosci 2006;7: Schneider LS, Dagerman KS, Insel P. Risk of death with atypical antipsychotic drug treatment for dementia: meta-analysis of randomized placebo-controlled trials. JAMA 2005;294: US Federal Drugs Agency. Public health advisory: deaths with antipsychotics in elderly patients with behavioural disturbances postmarketdrugsafetyinformationforpatientsandproviders/ucm De Deyn PP, Rabheru K, Rasmussen A, Bocksberger JP, Dautzenberg PL, Eriksson S, et al. A randomized trial of risperidone, placebo, and haloperidol for behavioral symptoms of dementia. Neurology 1999;53: Brodaty H, Ames D, Snowdon J, Woodward M, Kirwan J, Clarnette R, et al. A randomized placebo-controlled trial of risperidone for the treatment of aggression, agitation, and psychosis of dementia. J Clin Psychiatry 2003;64: Katz IR, Jeste DV, Mintzer JE, Clyde C, Napolitano J, Brecher M. Comparison of risperidone and placebo for psychosis and behavioral disturbances associated with dementia: a randomized, double-blind trial. Risperidone Study Group. J Clin Psychiatry 1999;60: Schneider LS, Tariot PN, Dagerman KS, Davis SM, Hsiao JK, Ismail MS, et al. Effectiveness of atypical antipsychotic drugs in patients with Alzheimer s disease. N Engl J Med 2006;355: Ballard C, Margallo-Lana M, Juszczak E, Douglas S, Swann A, Thomas A, et al. Quetiapine and rivastigmine and cognitive decline in Alzheimer s disease: randomised double blind placebo controlled trial. BMJ 2005;330: Devanand DP, Pelton GH, Cunqueiro K, Sackeim HA, Marder K. A 6-month, randomized, double-blind, placebo-controlled pilot discontinuation trial following response to haloperidol treatment of psychosis and agitation in Alzheimer s disease. Int J Geriatr Psychiatry 2011;26: Declercq T, Petrovic M, Azermai M, Vander Stichele R, De Sutter AI, van Driel ML, et al. Withdrawal versus continuation of chronic antipsychotic drugs for behavioural and psychological symptoms in older people with dementia. Cochrane Database Syst Rev 2013;3:CD Ballard C, Hanney ML, Theodoulou M, Douglas S, McShane R, Kossakowski K, et al. The dementia antipsychotic withdrawal trial (DART-AD): longterm follow-up of a randomised placebo-controlled trial. Lancet Neurol 2009;8: Ballard C, Lana MM, Theodoulou M, Douglas S, McShane R, Jacoby R, et al. A randomised, blinded, placebo-controlled trial in dementia patients continuing or stopping neuroleptics (the DART-AD trial). PLoS Med 2008;5:e Ballard C, Corbett A. Agitation and aggression in people with Alzheimer s disease. Curr Opin Psychiatry 2013;26: Porsteinsson AP, Drye LT, Pollock BG, Devanand DP, Frangakis C, Ismail Z, et al. Effect of citalopram on agitation in Alzheimer disease: the CitAD randomized clinical trial. JAMA 2014;311: Corbett A, Husebo B, Malcangio M, Staniland A, Cohen-Mansfield J, Aarsland D, et al. Assessment and treatment of pain in people with dementia. Nat Rev Neurol 2012;8: Testad I, Corbett A, Aarsland D, Lexow KO, Fossey J, Woods B, et al. The value of personalized psychosocial interventions to address behavioral and psychological symptoms in people with dementia living in care home settings: a systematic review. Int Psychogeriatr 2014;26: Medicines and Healthcare Regulatory Agency. Antipsychotic drugs. www. mhra.gov.uk/safetyinformation/generalsafetyinformationandadvice/ Product-specificinformationandadvice/Productspecificinformationandadvice-A-F/Antipsychoticdrugs/index.htm. EASILY MISSED? Pancreatic cancer Ajith K Siriwardena, 1 2 Alison M Siriwardena 3 1 Regional Hepato-Pancreato-Biliary Unit, Manchester Royal Infirmary, Manchester M13 9WL, UK 2 University of Manchester, Manchester, UK 3 Bramhall Park Medical Centre, Stockport, UK Correspondence to: A K Siriwardena ajith.siriwardena@cmft.nhs.uk Cite this as: BMJ 2014;349:g6385 doi: /bmj.g6385 THE BOTTOM LINE A 74 year old man who had been self medicating with antacids for recurrent indigestion consulted his general practitioner when symptoms persisted for three months. After assessment his GP prescribed a proton pump inhibitor and arranged review in four weeks. Two months later he developed back pain and was prescribed paracetamol with advice about mobilisation. Within four weeks he developed jaundice with dark urine and was admitted to his local hospital. Contrast enhanced computed tomography confirmed a diagnosis of locally advanced, Consider the diagnosis of pancreatic cancer in all patients presenting with painless obstructive jaundice and refer for further assessment Consider the diagnosis in patients with persistent non-specific gastrointestinal symptoms combined with recent onset or atypical back pain or recent onset diabetes Early referral, assessment, and surgical resection with adjuvant chemotherapy are the mainstays of care for patients with resectable disease, but most patients are not candidates for surgery In patients with unresectable disease, direct treatment towards the relief of jaundice, adequate analgesia, establishment of a tissue diagnosis, consideration of chemotherapy, pancreatic exocrine replacement, and appropriate palliative care unresectable pancreatic cancer (figure). He received palliative chemotherapy but died five months later. Pancreatic cancer More than eight in 10 cases of pancreatic cancer are ductal adenocarcinoma arising from the epithelium of the exocrine pancreas. 1 Obstruction of the lower common bile duct gives rise to the classic presentation of painless obstructive jaundice (dark urine, pale stools, and pruritus owing to hyperbilirubinaemia). 2 Tumours can also arise from the body or tail of the gland, and these are particularly difficult to diagnose because symptoms tend to be a late feature. 3 4 Surgical resection followed by adjuvant chemotherapy is the optimum treatment, but most patients are unsuitable for surgery because of comorbidity or the tumour being locally advanced or metastatic. 7 8 Why is it missed? Pancreatic cancer is easily missed at initial presentation because the early symptoms are non-specific and common. Non-specific symptoms can be categorised as systemic or local. Systemic features include weight loss and the onset of diabetes or impaired glucose tolerance. 9 Local symptoms include abdominal pain and back pain. the bmj 15 November

5 thebmj.com Previous articles in this series ЖЖNasal septal haematoma (BMJ 2014;349:g6075) ЖЖPerthes disease (BMJ 2014;349:g5584) Ж Ж Kawasaki disease (BMJ 2014;349:g5336) ЖЖPostnatal depression (BMJ 2014;349:g4500) ЖЖMotor neurone disease (BMJ 2014;349:g4052) Ж Ж Copper deficiency ( BMJ 2014;348:g3691) Abdominal pain, often epigastric and radiating to the back, is easily mistaken for dyspepsia by patients and doctors. Back pain is also common in pancreatic cancer, 10 possibly because extrapancreatic spread can involve peri-pancreatic neural plexuses. However, back pain is common in the age group most affected by pancreatic cancer, and in the absence of other symptoms or signs is unlikely to raise a suspicion of pancreatic cancer at first presentation. Jaundice is a relatively late symptom. Typically, the onset of jaundice precipitates consultation and rapid onward referral. Obstruction of the main pancreatic duct can lead to compromise in exocrine function, which can manifest as fat intolerance, steatorrhoea, and weight loss. 10 This clinical picture is a late stage phenomenon, but milder symptoms may be noticeable at an early stage. Why does this matter? About half of patients who present with pancreatic cancer are admitted as an emergency either through emergency departments or general practice; this is the highest proportion of GP emergency referrals for any cancer. 11 Only about 11% are referred under the two week wait system for suspected cancer referrals. 11 Patients presenting as emergencies have a lower 12 month survival rate (9%) than those referred under the two week wait rule (19%) or as routine or elective GP referrals (26%). 11 These differences in survival time suggest that the symptoms and signs that currently precipitate a suspected cancer referral are late features. The age standardised one year survival for pancreatic cancer in England ( ) was 17% for men and 19% for women. 12 In the large ESPAC-3 study, resection of the primary tumour followed by adjuvant chemotherapy was associated with a median survival of 43 (95% confidence interval 34 to 56) months, with 29% of patients being alive five years after surgery. 8 Thus, although not all patients with pancreatic cancer are candidates for surgery, evidence suggests that surgery improves survival and earlier diagnosis of non-metastatic tumours may allow more patients to have resection. Earlier diagnosis would also avoid the trauma associated with diagnostic delay. How is pancreatic cancer diagnosed? Clinical features Pancreatic cancer should be considered in any patient with recent onset painless obstructive jaundice. In this setting, management should consist of early referral for further assessment. 13 Diagnosis is more difficult in patients without ja undice. Here the coexistence of persistent non-specific gastrointestinal symptoms with recent onset or atypical back pain, particularly in the presence of recently diagnosed diabetes or weight loss, should raise an index of suspicion about pancreatic cancer. A large study based on a primary care database developed an algorithm for the diagnosis of pancreatic cancer (table). 14 Independent predictors in both men and women were age, smoking, type 2 diabetes, chronic pancreatitis, abdominal pain, anorexia, and weight loss (table). 14 The 10% of patients with the highest predicted risks made up 62% of all pancreatic cancers diagnosed over the next two years. 14 A case-control analysis that compared 3635 newly diagnosed patients with age and sex matched controls found that nine features were associated with pancreatic cancer (all P<0.01). 15 These were jaundice, abdominal pain, nausea or vomiting, back pain, constipation, diarrhoea, weight loss, malaise, and new onset diabetes. Investigations Biochemical liver function tests may help to distinguish between an obstructive and a hepatocellular picture in patients with painless jaundice and guide referral to medical or surgical teams. Upper abdominal ultrasonography may suggest the diagnosis, but a normal result does not exclude pancreatic cancer. 13 For the diagnosis of pancreatic cancer, ultrasonography has a sensitivity of 90% and specificity of 97%. 16 Ultrasound is operator dependent and false negative results can occur if the pancreas is obscured by gas in the overlying transverse colon. 17 UK guidelines currently recommend high resolution contrast computed tomography as the diagnostic test of choice. 13 It has a reported sensitivity of % and a specificity of % for the diagnosis of pancreatic cancer, 17 although sensitivity for lesions less than 2 cm is only 68-77%. 17 The test also provides information on prognosis by indicating the presence or absence of liver metastasis and invasion of local vascular structures. How is pancreatic cancer managed? The modern management of patients with pancreatic cancer requires multidisciplinary care in a high volume specialist Contrast enhanced computed tomogram of the upper abdomen showing a hypodense (dark) mass (M) in the head of the pancreas. Note that an endobiliary stent (E) is in situ HOW COMMON IS IT? In 2012, pancreatic cancer was the 10th most common cancer in the UK and the fifth most common cause of cancer related death 5 Incidence is strongly related to age, with a sharp rise after years In 2011, there were 8773 new cases in the UK: 4328 (49%) in men and 4445 (51%) in women, giving a male:female ratio of almost 1:1 1 Incidence and mortality figures are similar, with about 8000 people dying from the disease each year 1 Incidence seems to be higher in the Western world, possibly because of the increased prevalence of obesity November 2014 the bmj

6 Algorithm for the diagnosis of pancreatic cancer. Data expressed as adjusted hazard ratios (95% confidence interval) 14 Factor Women (95% CI) Men (95% CI) Smoking status Non-smoker 1 1 Ex-smoker 0.97 (0.77 to 1.23) 1.37 (1.12 to 1.67) Light smoker 1.53 (1.04 to 2.25) 1.44 (1.03 to 2.03) Moderate smoker 2.32 (1.74 to 3.10) 1.63 (1.20 to 2.20) Heavy smoker 2.39 (1.65 to 3.48) 1.88 (1.36 to 2.61) Medical history Type 2 diabetes 2.07 (1.66 to 2.58) 2.11 (1.76 to 2.52) Chronic pancreatitis 3.15 (1.17 to 8.46) 3.94 (1.93 to 8.01) Current symptoms and symptoms in preceding year Current appetite loss* 3.90 (2.61 to 5.82) 2.46 (1.43 to 4.23) Current weight loss* 3.27 (2.35 to 4.56) 12.5 (7.84 to 19.9) Current abdominal pain* 4.09 (3.46 to 4.84) 5.23 (4.48 to 6.11) Current abdominal distension* 3.04 (1.68 to 5.50) NS Current dysphagia* NS 2.56 (1.60 to 4.10) Constipation in past year* NS 1.91 (1.35 to 2.71) *Compared with person without this characteristic. Interaction term at mean age in men. The model for men also included an interaction between weight loss and the age term. Hazard ratios adjusted for all other terms in the table and for age. NS=not significant. centre. In patients with few (or no) comorbidities and a small non-metastatic tumour, current guidelines advocate surgical resection followed by adjuvant chemotherapy. 7 However, most patients are not candidates for surgery, and supportive care aims to relieve jaundice; establish a tissue diagnosis of cancer; and offer palliative ch emotherapy, pancreatic exocrine replacement therapy, adequate analgesia, and palliative care. 7 Contributors: Both authors reviewed the primary data cited in this article. Both contributed equally to writing the paper. AKS is guarantor. Competing interests: We have read and understood BMJ policy on declaration of interests and have none to declare. Provenance and peer review: Commissioned; externally peer reviewed. Patient consent not required (patient anonymised, hypothetical, or dead). 1 Cancer Research UK. Types of pancreatic cancer. www. cancerresearchuk.org/cancer-help/type/pancreatic-cancer/about/ types-of-pancreatic-cancer. 2 Cancer Research UK. Pancreatic cancer symptoms. www. cancerresearchuk.org/cancer-help/type/pancreatic-cancer/about/ pancreatic-cancer-symptoms. 3 Watanabe I, Sasaki S, Konishi M, Nakagohri T, Inoue K, Oda T, et al. Onset symptoms and tumor locations as prognostic factors of pancreatic cancer. Pancreas 2004;28: Brennan MF, Moccia RD, Klimstra D. Management of adenocarcinoma of the body and tail of the pancreas. Ann Surg 1996;223: Cancer Research UK. Cancer incidence and mortality in the UK. publications.cancerresearchuk.org/downloads/product/cs_report_ TOP10INCMORT.pdf. 6 Simard EP, Ward EM, Siegel R, Jemal A. Cancers with increasing incidence trends in the United States: 1999 through CA Cancer J Clin 2012;62: National Comprehensive Cancer Network. NCCN clinical practice guidelines in oncology. Pancreatic adenocarcinoma. Version Neoptolemos JP, Stocken DD, Bassi C, Ghaneh P, Cunningham D, Goldstein D, et al. Adjuvant chemotherapy with fluouracil plus folinic acid versus gemcitabine following pancreatic cancer resection: a randomized controlled trial. JAMA 2010;304: Sah RP, Nagpal SJS, Mukhopadhyay D, Chari ST. New insights into pancreatic cancer-induced paraneoplastic diabetes. Nat Rev Gastroenterol Hepatol 2013;10: Bond-Smith G, Banga N, Hammond TM, Imber CJ. Pancreatic adenocarcinoma. BMJ 2012;344:e National Cancer Intelligence Network. Routes to diagnosis: exploring emergency presentations briefings/routes_to_diagnosis_exploring_emergency_presentations. 12 Cancer Research UK. Pancreatic cancer survival statistics. www. cancerresearchuk.org/cancer-info/cancerstats/types/pancreas/ survival/pancreatic-cancer-survival-statistics. 13 Pancreatic section of the British Society of Gastroenterology. Guidelines for the management of patients with pancreatic cancer periampullary and ampullary carcinomas. Gut 2005;54 (suppl V):V Hippisley-Cox J, Coupland C. Identifying patients with suspected pancreatic cancer in primary care: derivation and validation of an algorithm. Br J Gen Pract 2012;62:e Stapley S, Peters TJ, Neal RD, Rose PW, Walter FM, Hamilton W. The risk of pancreatic cancer in symptomatic patients in primary care: a large casecontrol study using electronic records. Br J Cancer 2012;106: Karlson B-M, Ekbom A, Lindgren PG, Källskog V, Rastad J. Abdominal US for diagnosis of pancreatic tumor: prospective cohort analysis. Radiology 1999;213: Shrikhande SV, Barreto SG, Goel M, Arya S. Multimodality imaging of pancreatic ductal adenocarcinoma: a review of the literature. HPB (Oxford) 2012;14: This is one of a series of occasional articles highlighting conditions that may be more common than many doctors realise or may be missed at first presentation. The series advisers are Anthony Harnden, professor of primary care, Department of Primary Care Health Sciences, University of Oxford, and Richard Lehman, general practitioner, Banbury. To suggest a topic for this series, please us at practice@bmj.com. ANSWERS TO ENDGAMES, p 34 For long answers go to the Education channel on thebmj.com STATISTICAL QUESTION Explanatory trials versus pragmatic trials Statements a and c are true, whereas b and d are false. ANATOMY QUIZ Magnetic resonance angiogram, with contrast, of the carotid arteries A: Brachiocephalic artery B: Right subclavian artery C: Right common carotid artery D: Right external carotid artery E: Right internal carotid artery PICTURE QUIZ White pupillary reflex in a 3 year old boy 1 Retinoblastoma. The most common diseases that can simulate retinoblastoma are Coat s disease, persistent fetal vasculature (a congenital ocular disorder in which fetal vasculature persists), vitreous haemorrhage, rhegmatogenous retinal detachment, coloboma, and astrocytic hamartoma. 2 Fundoscopy, ultrasound, and magnetic resonance imaging (MRI) are the examinations of choice in patients being investigated for retinoblastoma. 3 Several treatment options are available, including local ablative therapy, chemotherapy, and enucleation. 4 Retinoblastoma is a disease that can be inherited. 5 In certain cases retinoblastoma can affect the anterior chamber, while fundoscopy is normal. the bmj 15 November