Article Text

DTB Select: 3 | March 2017


Potential for abuse and misuse of pregabalin and gabapentin | Risk of depression and suicidal thoughts with ▼apremilast | Are patients overly optimistic about outcomes? | Pioglitazone and cardiovascular outcomes | Antipsychotics and risk of acute respiratory failure in patients with COPD | Improving adherence to lipid-lowering drugs | ▼Canagliflozin associated with increased risk of amputation | Medical research news stories: how independent and qualified are commenters?

Statistics from

Potential for abuse and misuse of pregabalin and gabapentin

The Advisory Council on the Misuse of Drugs (ACMD) has highlighted concerns over the misuse of pregabalin and gabapentin.1 This follows concerns raised by the Health and Social Care Board of Northern Ireland and Public Health England.2,3 The chair of the ACMD has written to the UK government to request that pregabalin and gabapentin be classed as controlled drugs.1

The ACMD conducted a review of the potential harms associated with the misuse of these drugs.1 It noted that, in the UK, prescribing of pregabalin and gabapentin has increased by 350% and 150%, respectively, over 5 years. UK survey data suggests that abuse of pregabalin and gabapentin is common among opioid-using and prison populations.

Pregabalin is reported to produce a high or elevated mood in some users, and is thought to have greater abuse potential (due to rapid absorption and onset of action, and higher potency) than gabapentin, which can give rise to feelings of relaxation, calmness and euphoria.1 In 2013, pregabalin and gabapentin were implicated in 19 and 17 deaths, respectively, In 2014, the Office for National Statistics reported 38 deaths where pregabalin was mentioned on the death certificate. The corresponding figure for gabapentin was 26.

The ACMD also cites reports of abuse, dependence and involvement of pregabalin and gabapentin in opioid overdose deaths in Finland and Germany, where pregabalin abuse with high doses is reported as increasingly common and there is significant concern over the possibility of fatalities when either drug is combined with opioids.

The ACMD asserts that pregabalin and gabapentin present a risk of addiction (with the potential for illegal diversion and medicinal misuse) and that the harms associated with pregabalin and gabapentin are comparable to those of tramadol. The ACMD has recommended that both agents are controlled as Class C substances under the Misuse of Drugs Act 1971 (and scheduled under the Misuse of Drugs Regulations 2001 [as amended] as Schedule 3).

Comment: Concerns over the potential for abuse with pregabalin and gabapentin have been known for some time.4,5 In England, 4.8 million prescriptions for pregabalin and 5.7 million prescriptions for gabapentin were dispensed in 2015.6 The letter from the ACMD is a welcome acknowledgement of the need to address the problems associated with such widely prescribed drugs.


Risk of depression and suicidal thoughts with ▼apremilast

Healthcare professionals are being alerted to an increased risk of psychiatric symptoms, including depression and suicidal thoughts, among patients taking apremilast (Otezla), a phosphodiesterase-4 inhibitor licensed for the treatment of plaque psoriasis and psoriatic arthritis in adults.1–3

The safety update was issued by the Medicines and Healthcare products Regulatory Agency (MHRA) following a regulatory review of evidence from clinical trials and postmarketing cases.1 The MHRA alert acknowledges that depression, suicidal thoughts and suicidal behaviours are more common in patients with psoriasis or psoriatic arthritis than in the general population. Nevertheless, the review suggested that there was a causal association between the use of apremilast and suicidal thoughts and behaviour (including completed suicide) in patients with or without a history of depression. These events were reported with an estimated frequency of 1 in 1,000 to 10 in 1,000 patients taking apremilast.

In placebo-controlled clinical trials, depression occurred more commonly among patients receiving apremilast.2 There have also been postmarketing reports of depression associated with apremilast use, including some serious cases.

The MHRA has recommended the following actions for healthcare professionals:2

  • carefully assess the balance of benefits and risks of starting or continuing treatment with apremilast in patients with a history of psychiatric symptoms or in those taking other medicines likely to cause psychiatric symptoms;

  • stop treatment with apremilast if patients experience new or worsening psychiatric symptoms; and

  • tell patients to contact a healthcare professional if they notice changes in their mood.

Apremilast has been licensed in the UK since January 2015 for the treatment of moderate to severe chronic plaque psoriasis or active psoriatic arthritis in adults who have not responded to, or who are unable to take, other systemic treatments.3

Comment: When DTB reviewed the safety and efficacy of apremilast,4 we noted that the European Medicines Agency's risk management plan recommended that patients taking apremilast should be monitored for early signs and symptoms of depression. This recommendation was informed by earlier reports of an increased risk of suicidal behaviour with another PDE-4 inhibitor,roflumilast (licensed for maintenance treatment of severe chronic obstructive pulmonary disease associated with chronic bronchitis).5 Patients prescribed apremilast for plaque psoriasis or psoriatic arthritis should be carefully monitored for the emergence or exacerbation of psychiatric symptoms. Suspected adverse drug reactions to apremilast should be reported via the Yellow Card scheme. (


Are patients overly optimistic about outcomes?

Patients often have inaccurate expectations of the benefits and harms of medical interventions, reports a systematic review, which found that patients usually overestimate the benefits of treatments, tests and screening, while underestimating the harms.1

The review's authors identified 35 studies (27,323 participants) that quantitatively measured patient or public expectation of the benefit and/or harm of a medical intervention. The majority of studies (22) assessed participant expectation of benefit; ten studies evaluated benefit and harm expectations; three studies assessed expectations of harm only. Of the studies, 15 evaluated treatments only, 14 explored expectations relating to screening, three focused on tests and three on screening and treatment. Of the 18 treatments considered, only four were examined in more than one study. Fifteen of the studies on tests or screening focused on cancer. The range of study designs and outcomes used precluded meta-analysis.

Study participants had received, were about to receive or were considering the intervention of interest; were attending or had recently attended a healthcare facility; or were members of the general population. They were asked to quantitatively estimate the likely benefit and/or harm of an intervention, and this estimate was compared with the study authors' ‘correct’ answer (contemporaneous estimate).

The review found that:

  • Among the 32 studies that assessed expectations of benefit, 21 provided data. Most participants overestimated benefits for 22 of 34 outcomes that provided overestimation data. A majority of participants correctly estimated the benefit of two outcomes: the proportion of people with improved vision after cataract surgery and the accuracy of the cervical smear test. There was only one outcome where at least half of participants underestimated the benefit (reduction in lower back pain after back surgery).

  • Among studies that assessed expectations of harm, data were available for 15 outcomes. The majority of participants underestimated harm for 10 outcomes and correctly estimated harm associated with two outcomes (proportion of people who still need glasses after cataract surgery and risk of miscarriage following amniocentesis).

(Funding provided by the Australian Department of Health and Ageing.)

Comment: Discussions about benefits and harms are an essential part of the process of helping patients decide on treatment options.2 The findings of this systematic review underscore the need to improve the communication and understanding of the benefits and harms of interventions. Shared decision making, the participation of clinician and patient in reaching an informed healthcare decision, can help to identify uncertainties relating to the outcomes of treatment.3,4 This may address unrealistic expectations and help to reduce the use of unnecessary interventions in clinical practice.


Pioglitazone and cardiovascular outcomes

The cardiovascular benefits and harms of drugs used to manage diabetes is a subject of ongoing debate. Although many drugs are licensed to manage glycaemic control in people with type 2 diabetes, evidence of improved cardiovascular outcomes for these treatments is limited.1 The European Medicines Agency (EMA) requires evidence that any new drug for glycaemic control does not increase the risk of cardiovascular complications. In addition, the cardiovascular effects of many licensed drugs are being assessed through clinical trials and systematic reviews. The authors of a systematic review of trials of pioglitazone have concluded that it reduces the risk of major cardiovascular events in people with insulin resistance, prediabetes or diabetes mellitus, but is associated with an increased risk of heart failure and weight gain.2

The systematic review and meta-analysis included nine randomised controlled trials with 12,026 participants (mean age 62 years; mean study duration 2.7 years). Three of the trials were non-blinded. Participants, with or without cardiovascular disease, had a history of insulin resistance, pre-diabetes (impaired fasting glucose and/or impaired glucose tolerance) or diabetes mellitus. The primary outcome was the risk of major adverse cardiovascular events (composite of non-fatal myocardial infarction, non-fatal stroke and cardiovascular death). Safety outcomes included heart failure, fracture, all-cause mortality, cancer, bladder cancer, oedema, weight gain and hypoglycaemia.

Pioglitazone reduced the risk of major adverse cardiovascular events in patients with insulin resistance or pre-diabetes compared with placebo (2 trials; relative risk [RR] 0.77, 95% CI 0.64 to 0.93). Pioglitazone also reduced the risk of myocardial infarction in these patients (2 trials; RR 0.68, 95% CI 0.49 to 0.96) but not that of recurrent stroke (2 trials; RR 0.81, 95% CI 0.65 to 1.01).

In patients with diabetes, pioglitazone was associated with a lower risk of major adverse cardiovascular events (5 trials; RR 0.83, 95% CI 0.72 to 0.97). The risk of myocardial infarction (5 trials; RR 0.80, 95% CI 0.62 to 1.03) or stroke (5 trials; RR 0.78, 95% CI 0.60 to 1.02) was not reduced in this population.

Meta-analysis of trial data that included patients with insulin resistance, pre-diabetes or diabetes mellitus found that, compared with placebo, pioglitazone increased the risk of heart failure (5 trials; RR 1.32, 95% CI 1.14 to 1.54), bone fracture (4 trials; RR 1.52, 95% CI 1.17 to 1.99), oedema (7 trials; RR 1.63, CI 1.52 to 1.75), weight gain (4 trials; RR 1.60, CI 1.50 to 1.72) and hypoglycaemia (5 trials; RR 1.24, CI 1.13 to 1.35). Risk differences for all-cause mortality (7 trials; RR 0.93, 95% CI 0.80 to 1.09), any cancer (4 trials; RR 0.91, 95% CI 0.77 to 1.08) and bladder cancer (2 trials; RR 1.87, 95% CI 0.98 to 3.57) were not statistically significant.

(The research was supported by the Ministry of Science and Technology, Taiwan.)

Comment: In the UK, pioglitazone is licensed as second- or third-line treatment of type 2 diabetes mellitus.3 Although pioglitazone, in combination with standard therapy, reduced the composite endpoint in patients with insulin resistance, pre-diabetes or diabetes mellitus, it is not clear what specific cardiovascular benefit an individual may reasonably expect from treatment with pioglitazone. Furthermore, the increased risk of serious adverse events, alongside the inclusion of primary and secondary prevention populations in the review, complicated the analysis of the overall harms and benefits.


Antipsychotics and risk of acute respiratory failure in patients with COPD

A possible link between the use of antipsychotic drugs and acute respiratory failure (ARF) in people with chronic obstructive pulmonary disease (COPD) has been reported.1

The authors of the study note that acute respiratory distress requiring intubation or mechanical ventilation has been described in individuals with COPD subsequent to the administration of an antipsychotic agent.1 Cases have involved typical and atypical antipsychotics administered orally or by injection. However, the safety of antipsychotics has not been examined in a population-based study of patients with COPD.

The study authors used a case-crossover study design to compare the use of antipsychotics immediately prior to an ARF event (case period) and during a preceding control period for each patient with COPD and ARF. The case period was defined as 1 to 14 days before the ARF event (based on the risk period noted in previous case reports). The control period was 75 to 88 days before the ARF event (to mitigate potential effects of drugs with extended half-lives). Using the Taiwan National Health Insurance Research Database, the authors identified eligible patients (COPD with newly diagnosed ARF in hospital or emergency care settings requiring intubation or mechanical ventilation) from the 1st January 2000 to the 31st December 2011. Analysis was restricted to idiopathic ARF by excluding patients with a history of ARF, lung cancer and cardiogenic, traumatic or septic ARF.

The study identified 5,032 patients (mean age 74 years) with ARF among 61,620 patients with COPD, corresponding to an incidence rate of 3.5 per 100 person-years. In total, 590 patients with ARF (11.7%) filled at least one antipsychotic prescription during the case period compared with 443 (8.8%) during the control period, resulting in an increased risk of ARF regardless of antipsychotic class and administration route (adjusted odds ratio 1.66, 95% CI 1.34 to 2.05; p<0.001).

(The funding arrangement for the study was not stated.)

Comment: The authors of the study suggest that use of an antipsychotic agent is associated with a modest increase in the risk of acute respiratory failure in patients with COPD. However, the study design, which serves to investigate transient effects of an intermittent exposure on acute outcomes, cannot establish causality. The clinical significance of this association remains unclear.


Improving adherence to lipid-lowering drugs

Automated reminders, counselling sessions and other interventions that provide intensified patient care can improve adherence to drug regimens for hyperlipidaemia, as well as lipid levels, according to an updated Cochrane review.1

Despite strong evidence that lipid-lowering medicines improve cardiovascular outcomes, adherence to these treatments is often poor. Earlier versions of this review (published in 2004 and 2010) found no clear benefit for any type of intervention aimed at improving adherence. However, the most recent update identified 24 additional randomised controlled trials that satisfied the review's eligibility criteria (randomised controlled trials of interventions of any type intended to increase adherence to self-administered lipid-lowering medication compared with usual care or no intervention). The review now includes 35 studies with 925,171 participants (mean age range 49–77 years).

The authors grouped the interventions into seven general categories: intensified patient care (e.g. automated reminders, pharmacist-led interventions and multidisciplinary educational or counselling sessions); simplification of drug regimens; patient education and information; complex behavioural approaches; computer-based decision-support systems; administrative improvements; and large-scale pharmacy-led automated telephone interventions.

Of these, only intensified patient care consistently showed an overall improvement in adherence rates, total cholesterol and low-density lipoprotein (LDL) cholesterol compared with usual care. The effects persisted during short- and long-term follow-up.

Over the short term (6 months or less), intensified patient care resulted in:

  • improved adherence compared with usual care (odds ratio [OR] 1.9, 95% confidence interval [CI] 1.3 to 2.9; 7 studies, 11,204 participants, moderate-quality evidence);

  • decreased total cholesterol, mean difference 17.2 mg/dL [0.4mmol/L] (95% CI 1.2 to 33.1; 4 studies, 430 participants, low-quality evidence); and

  • decreased LDL-cholesterol, mean difference 19.5 mg/dL [0.5mmol/L] (95% CI 8.5 to 30.5; 3 studies, 333 participants, moderate-quality evidence).

Over the long-term (more than 6 months), intensified patient care resulted in:

  • improved adherence compared with usual care (OR 2.9, 95% CI 1.9 to 4.3; 3 studies, 663 participants, high-quality evidence);

  • decreased total cholesterol, mean difference 17.6 mg/dL [0.5mmol/L] (95% CI 15.0 to 20.2; 2 studies, 127 participants, high-quality evidence).

In a sensitivity analysis, the effectiveness of these interventions was sustained when studies that contributed to heterogeneity were removed.

The authors note that the included studies did not report usable data for health outcomes, adverse effects or costs/resource use.

Comment: Non-adherence to medication is a well-recognised problem among people with long-term medical conditions.2 National guidance has highlighted that interventions to improve adherence should be considered on a case-by-case basis and should address the concerns and needs of individual patients.3 The results of the Cochrane review are encouraging in light of the documented high discontinuation rates and lack of adherence to lipid-lowering medications among patients at risk of cardiovascular morbidity and mortality. However, such interventions are likely to require a co-ordinated team-based approach to provide reminders, pharmacist-led interventions and educational/counselling sessions.


▼ Canagliflozin associated with increased risk of amputation

The European Medicines Agency's Pharmacovigilance Risk Assessment Committee (PRAC) has issued a warning that an increase in cases of lower-limb amputation (mostly affecting the toes) has been observed in patients taking canagliflozin compared with placebo in two clinical trials.1 The ongoing CANVAS and CANVAS-R studies involve patients at high risk of cardiovascular events.2,3 Data from both studies will be combined as part of a prespecified integrated analysis of cardiovascular safety outcomes to satisfy the USA Food and Drug Administration's post-marketing requirement for canagliflozin.2

Although an increased risk has not been seen with ▼dapagliflozin or ▼empagliflozin, data are limited and PRAC has advised that the risk of amputation may also apply to these drugs.1 Based on available data, PRAC has recommended that a warning on the risk of lower limb amputation should be included in the prescribing information for these drugs, highlighting the importance of routine preventative foot care. Lower-limb amputation should be listed in the prescribing information as an uncommon adverse effect (occurring in between 1 to 10 patients in 1,000).

In the UK, canagliflozin is licensed for use in adults with type 2 diabetes mellitus to improve glycaemic control as monotherapy or as add-on therapy when other glucose-lowering medicinal products (including insulin), combined with diet and exercise, do not provide adequate glycaemic control.4,5

Comment: This new group of oral glucose-lowering drugs (sodium-glucose co-transporter-2 [SGLT2] inhibitors) inhibit renal glucose reabsorption and promote glycosuria. All three drugs have been shown to reduce HbA1c by a similar amount and to produce modest reductions in body weight and blood pressure.6 Adverse effects of SGLT2 inhibitors include an increased risk of genital infection, hypoglycaemia when used in combination with sulfonylureas and volume depletion when used in combination with diuretics. In addition, there have been safety warnings about the risk of diabetic ketoacidosis with SGLT2 inhibitors (this has occurred with relatively low levels of blood glucose). Healthcare professionals need to be aware of the risk of increased lower limb amputation. Suspected adverse drug reactions to SGLT2 inhibitors should be reported through the Yellow Card scheme. (


Medical research news stories: how independent and qualified are commenters?

Independent commenters and editorial writers who comment upon medical research may lack relevant clinical or academic expertise, have undisclosed financial or academic conflicts of interest, or both, an observational study has found.1

Using a data set of clinical research publications from seven medical journals with a high-impact factor (New England Journal of Medicine, the Journal of the American Medical Association, The Lancet, PLoS Medicine, JAMA Internal Medicine, BMJ and Annals of Internal Medicine), the authors identified 131 articles (published between 1 January to 31 March, 2013) with a clinical outcome that generated 591 news stories, of which 92 featured independent comment. Independent commenters were defined as those other than study authors, study sponsors or journal editorialists who were not involved in the development or marketing of the intervention assessed in the research.

The main study outcomes were prevalence of relevant academic expertise (co-authored five or more relevant papers during the preceding 5-year period) and clinical expertise, prevalence and reporting of academic conflicts of interest (as defined by the International Committee of Medical Journal Editors), financial conflicts of interest (financial connection with any entity with commercial interests in the medical research) and disposition of comments toward study findings.

The 92 news stories, generated by 40 research publications (21 of which were accompanied by an editorial), contained 104 comments attributed to 102 independent commenters. Twenty-three comments were from spokespersons for advocacy organisations. On six occasions (26%), the advocacy organisation received undisclosed sponsorship from a commercial entity with an interest relevant to the medical research.

Although editorial writers were more likely to have relevant academic expertise than independent commenters (86% vs. 53%, p=0.007), the proportion with pertinent clinical expertise did not differ significantly between the two groups (48% vs. 56%). None of the editorial writers lacked both academic and clinical expertise, while 25% of independent commenters had neither. Of the 26 independent commenters that lacked either academic or financial expertise, 18 were spokespersons for advocacy organisations.

Of the 21 editorial comments, an academic conflict of interest was present in nine instances (43%), and a financial conflict of interest in nine (38%). Academic and financial conflicts of interest were present in 56 (54%) and 33 (32%) of the 104 comments, respectively. Of the 56 academic conflicts of interest, 25 (45%) were reported in the news stories. Only 11 (33%) of the financial conflicts of interest were reported.

Independent commenters were favourably disposed to research when their academic and financial conflicts of interest were consistent with the results of the medical research (academic conflict: 34 [97%] of the 35 comments expressing positive views; financial conflict: 13 [95%] of the 14 comments expressing positive views).

(The authors provided details of their research funding arrangements.)

Comment: Recommendations on disclosure of potential conflicts of interest by authors are provided by ICMJE and Committee on Publication Ethics.2,3 Disclosure of of potential conflicts of interest (mandated by many medical research journals) provides the reader an opportunity to consider academic and pecuniary interests that might influence their understanding of the author's work. The results of this study suggest that the reporting of medical research in news stories would benefit from a similar degree of stringency.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.