August 21st, 2012
Shining a Light on Standards at Medical Journals
Druin Burch, BM BCh MA MSc, John Stephen Yudkin, MB BChir MD FRCP and Marion Marjorie Mafham, MBChB MD
The two BEGIN Basal-Bolus trials, published this past April in the Lancet and funded by Novo Nordisk, do not shed light on how best to treat patients with type 1 and type 2 diabetes, but rather on the consequences of poor standards in medical journals and some medical specialties. These open-label randomized studies share notable characteristics. The trials demonstrate that insulin degludec is noninferior to insulin glargine for glycemic control, as measured by HbA1c after 1 year. Each deals with a common disease yet recruited fewer than 10 patients per center. That may be an inefficient way to gather scientific data, but it’s very effective at getting a large number of units to start prescribing a new drug.
Strikingly, neither BEGIN article focuses its main conclusion on the primary outcome but instead on secondary measurements: nocturnal hypoglycemia in the first trial and overall hypoglycemia in the second. In both studies, the differences in outcomes were of marginal statistical significance, and there is no mention of adjustment for multiple testing. The focus on secondary endpoints encourages readers to believe that selecting insulin degludec over alternative treatments is warranted. Such lower rates of hypoglycemia rates in unblinded studies, however, should be considered hypothesis-generating at best. At worst, conclusions about them are completely spurious.
Another recent diabetes study in the Lancet, funded by Boerhinger Ingelheim, compares linagliptin with glimepiride for glycemic control. It follows the same pattern of giving undue emphasis to a hypothesis-generating outcome: The abstract highlights that linagliptin is associated with fewer cardiovascular events, even though this secondary outcome was not prespecified. Incredibly, Figure 3 in the paper further subdivides the cardiovascular events into individual outcomes, showing 3 strokes in the linagliptin group compared to 11 with glimepiride. Buried in the discussion is an assertion that “the likelihood that this finding was due to chance (type 1 error) cannot be discounted, because the study was neither planned nor powered for cardiovascular outcomes.” The mention of type 1 error brilliantly implies that the qualification is merely a statistical complexity, leaving the focus on the supposed advantages of linagliptin. But the only reasonable conclusion from the data is, at most, that different oral hypoglycemic agents might differ in their effects on hard outcomes.
Why does the Lancet repeatedly allow authors to get away with this sort of stuff? Is it because distribution of reprints is a major part of drug-company marketing and, therefore, a major source of revenue for the journal? The two insulin degludec trials and the linagliptin trial were written with the help of drug company–funded professional writers. Their skill in doing their job is clear. Equally clear is the lack of Lancet editorial effort to balance it. Yet these low editorial standards cannot be attributed entirely to commercial self-interest: In June the Lancet published a non–industry-funded trial of stroke thrombolysis (with recombinant t-PA) that had a negative primary outcome, yet again the article had an inappropriate emphasis on a positive secondary result.
The Lancet was founded in 1823 to support clarity and attack error. Its name is a pun: A lancet is both a window that lets in light and a surgical instrument used to evacuate pus. These articles, metaphorically, do neither. But the Lancet is not guilty of unusually bad behavior. All major medical journals frequently allow papers to improperly emphasize secondary endpoints; stopping them from doing so is somehow not seen as an editorial duty. Nor is the Lancet unique in making a large amount of money from drug-company purchases of article reprints that are then used by the drug makers as part of a marketing strategy. In a recent paper where the British Medical Journal and the Lancet revealed their income from drug company–purchased reprints, the New England Journal of Medicine, JAMA, and the Annals of Internal Medicine all declined to reveal their own figures. The culture across medical journals appears consistent both in allowing drug companies (often using professional writers) to inappropriately spin data, and in the journals then profiting when those same companies purchase the relevant reprints.
Contrast that with cultural inconsistency, across medical specialties, in the extent to which their trials rely on unjustified surrogate outcomes. There is a strong tradition within diabetology of conducting these sorts of poor-quality studies. Recent reviews have demolished the supposed benefits of intensive glucose lowering on macrovascular outcomes in patients with type 2 diabetes and the cardiovascular-mortality benefits of metformin. It is plain that HbA1c is not a reliable surrogate for weighing the harms and benefits of particular hypoglycemic agents in patients for whom the risks of macrovascular complications far outweigh those of microvascular eye and kidney disease. Nonetheless, the next study of a hypoglycemic agent will undoubtedly be along the lines of these two insulin degludec trials. The agent will be judged chiefly via nothing better than HbA1c reduction—and the trial will promptly be published and the drug licensed. In cardiology, imagine a company trying to license a new HDL-raising agent solely on the basis of its effects on HDL levels, without examining mortality and morbidity endpoints. It would simply never happen. What explains the difference in epistemological quality between the two specialties? Why are drugs for diabetes evaluated in ways that don’t show what they do to hard outcomes, when drugs for the heart now almost always are?
Cultural attitudes toward evidence exert a huge influence and deserve our attention. The tradition within cardiology of demanding high-quality evidence is not a function of any pre-existing moral or mental superiority among cardiologists. It comes simply from cardiologists’ having witnessed (over decades) the impact of a large number of high-quality trials that changed practice and saved lives. Experience has largely converted cardiologists to the value of seeking reliable data on hard outcomes and of subjecting even their favorite therapies to experimental validation. It is unfortunate that diabetologists have ignored the lessons learned by cardiologists. It is inexcusable and bizarre that journal editors have done the same.
Note: In June the authors published a correspondence piece on this topic in the Lancet.
What’s your take on editorial standards in medical journals and on the BEGIN trials in particular?
August 20th, 2012
TNF Inhibitors Linked to Reduction in MI for Psoriasis Patients
Larry Husten, PHD
Psoriasis patients who take TNF inhibitors have a significant reduction in the risk for myocardial infarction (MI), according to a retrospective cohort study published in Archives of Dermatology. Although previous research suggested that the anti-inflammatory effects of methotrexate, an older therapy, may be beneficial in this population, the cardiovascular effects of TNF inhibitors had not been well studied.
Researchers identified 8845 patients who were diagnosed with psoriasis or psoriatric arthritis within the Kaiser Permanente Southern California health plan. After a median of 4.3 years of observation, the overall rate of MI was 5.21 per 1000 patient-years.
- In the 1673 patients who received a TNF inhibitor (etanercept, infliximab, or adalimumab), the MI rate was 3.05 per 1000 patient-years.
- In the 2097 patients who received oral therapy or phototherapy, the MI rate was 3.85 per 1000 patient-years.
- In the 5075 patients who received topical therapy, the MI rate was 6.73 per 1000 patient-years.
TNF inhibitors and oral therapy/phototherapy were each superior to topical therapy, but the difference between TNF inhibitors and oral therapy/phototherapy was not significant. Compared with topical agents, TNF inhibitors and oral agents/phototherapy had hazard ratios (adjusted for other risk factors) of 0.50 and 0.54, respectively.
The authors write, “This is the first large scale retrospective cohort study to show that the use of TNF inhibitors for psoriasis is associated with a clinically and statistically significant reduction in MI risk and incident rate compared with psoriatic patients treated with topical agents.” However, they note that “prospective studies are needed and warranted to determine whether the use of TNF inhibitors may reduce the risk of major adverse cardiovascular events in patients with systemic inflammatory conditions.”
August 20th, 2012
Why Is the National Library of Medicine Still Indexing Reviews in Cardiovascular Medicine?
Kevin Lomangino, BA
The following guest post by Kevin Lomangino is reprinted with permission from HealthNewsReviews.org. Mr. Lomangino is an independent medical journalist and editor who is currently Editor-in-Chief of Clinical Nutrition Insight, a monthly evidence-based newsletter which reviews the scientific literature on nutrition for physicians and dietitians.
Last week, a study looking at off-label drug promotion reported a finding that was considered “unsurprising and disturbing” by some observers: Only 15% of physicians and scientists touting such off-label uses disclosed that they had relevant financial relationships with the manufacturers who produced the drugs.
I’ll admit to some surprise at the extent to which these authors appear to have systematically hoodwinked their readers. But it’s true that the basic problem identified by the study – that scientific authors sometimes fail to disclose financial relationships that might influence their writing – is nothing new. In fact, the study brought to mind a troubling example of inadequate disclosure I wrote about in HealthNewsReviews.org in February. And it inspired me to complete this follow-up post that I’ve been putting off for some time.
First, some background: The piece I wrote in February focused on a cardiology journal, Reviews in Cardiovascular Medicine (RICM), that engages in questionable editorial practices that promote commercial interests. The journal was first flagged by blogger Marilyn Mann for publishing an article, sponsored and reviewed prior to publication by pharmaceutical manufacturer Abbott, extolling the benefits of Abbott drugs for the prevention of cardiovascular disease. After some more poking around, I found that the RICM editors also repeatedly failed to disclose — and in some cases, appeared to be actively concealing through misleading statements — relationships with other manufacturers whose products they were writing about in the journal. (I continue to welcome any attempt by these editors to explain the discrepancies identified, in case I have the wrong idea.)
As far as I know, there haven’t been any negative repercussions for RICM or its editors stemming from my post or other critical coverage. (In fact, the editors are so unconcerned that they didn’t even bother to update their disclosure statements on the RICM website, which as I pointed out months ago are woefully out of date and inaccurate. ) This isn’t all that unusual, really, considering that even authors who participate in fraudulent research don’t face much in the way of consequences for their actions. It seems that the dollar value these authors bring to their institutions trumps any ethical qualms about their publishing activities.
If RICM were a subscription-based business, readers might have an opportunity to lodge a protest by refusing to renew their subscriptions. RICM, however, is sent out free of charge through support from industry sponsors. And it seems that companies are all too happy to have RICM’s editors serving as their pitchmen.
There is one organization, though, that I had hoped would have more than a passing interest in meting out some discipline to RICM: the National Library of Medicine (NLM). As the curator of PubMed, the world’s foremost index of the scientific literature, the NLM has an obvious stake in assuring that the journals it indexes adhere to high editorial and ethical standards. And as it turns out, NLM even has a specific policy that seems designed to prevent the kinds of shenanigans that RICM engages in.
When it comes to sponsored journal supplements, NLM requires that disclosure statements be “specific and address any financial relationship the guest editors and authors have with the sponsoring organization and any interests that organization represents.” This policy would appear to preclude the NLM from indexing content such as the RICM supplement on Proteomic Approaches to Acute Coronary Syndromes, which received corporate sponsorship from diagnostic test manufacturer Alere.
As I noted in my earlier post on HealthNewsReviews.org, there is no acknowledgment in that supplement of Alere’s relationship with RICM editor Peter McCullough, MD, who is also coauthor of several articles in the supplement. These articles state that “no funding was provided to authors,” and one of the articles states that Dr. McCullough has “no real or apparent conflicts of interest to report.” This is despite the fact that Dr. McCullough has acknowledged being a consultant and speaker for Alere for about a decade, and was giving talks sponsored by the company within a year of the supplement’s publication.
And yet today RICM and its supplements continue to be indexed in the PubMed database, and comments from NLM officials suggest that this is unlikely to change anytime soon. When presented with the evidence of RICM‘s breach of NLM policy, their response to me was basically to say: Sorry, not our problem.
As Deborah Ozga, head of the Index Section wrote to me, “Staff members of the NLM Index Section check that supplements include disclosure statements and that the statements meet the criteria described in the conflict of interest Fact Sheet. Based on these requirements, [McCullough’s disclosure statement] is in compliance with NLM policy.”
But there’s an important qualifier: According to Ozga, “Staff members do not investigate the accuracy of disclosure statements.” (Emphasis mine.)
In other words, the NLM will go as far as checking that authors and editors have made a disclosure statement as required by their policy. But if you try to subvert that policy by making a false or misleading disclosure, they are not going to go out of their way to try and stop you. When I asked if there was any mechanism for NLM to evaluate possible improprieties in RICM’s disclosures, Ozga suggested that I raise my concerns with International Committee of Medical Journal Editors or a cardiovascular medicine organization.
Now, I’m not suggesting that the NLM should routinely verify the disclosures made in journal articles. But when presented with credible evidence of a breach of their policy, shouldn’t they do more than turn a blind eye to the infraction? If the intent of the policy is to prevent dissemination of biased information to PubMed users, clearly there needs to be a more robust enforcement mechanism attached to it.
It’s worth reiterating that this issue has important implications for patients. In the world of medical publishing, inclusion in the PubMed database confers considerable prestige and credibility. More importantly, it makes content from the journal visible to millions of researchers and clinicians online who otherwise would probably never be aware of it. (More than a billion searches were logged on PubMed in 2009.)
So when RICM publishes industry-sponsored patient care recommendations that contradict guidelines from the American Heart Association (as happened with the Abbott-sponsored review of fibrate drugs), PubMed’s imprimatur helps give credence and additional distribution to the industry-supported view. That may be part of the reason why sales of fibrate drugs continued to climb for most of the past decade, even in the face of evidence showing that these drugs were no more effective than placebo for improving heart disease outcomes.
August 16th, 2012
FDA Calls for Imaging of Implantable Cardiac Devices with Riata Leads
The FDA is recommending imaging studies for patients whose implantable cardioverter/defibrillator or cardiac resynchronization therapy defibrillator has Riata or Riata ST leads.
The studies are to check for abnormalities in the leads’ insulation, which can cause the devices to deliver inappropriate or no shock therapy. The abnormalities may be detected with two-view chest X-ray or fluoroscopy. The leads are manufactured by St. Jude Medical, which reported that there are some 79,000 Riata leads implanted in the U.S.
The agency recommends close monitoring of affected patients, who should be notified of the risk. Patients who’ve not had a recent evaluation should get one, and their devices should be checked for electrical abnormalities. The FDA also advises that clinicians consider remote monitoring of patients.
August 16th, 2012
Children Should Have Their Cholesterol Checked
Samuel S. Gidding, MD
Dr. James H. Stein, in his recent post on CardioExchange, presents several arguments against cholesterol screening in childhood. I would like to present the view of those who drafted the document and suggest that some of Dr. Stein’s arguments are incorrect.
First, given that conflicts of interest were part of his argument, I must be transparent about mine. I have never participated in a drug-industry-sponsored trial of a cholesterol-lowering medication. I had no conflicts with industry until late 2009, after the main work on the guideline was complete. Both of these are related to evidence gaps identified in the literature review for the guideline. The first is as a member of the Data and Safety Monitoring Board for a trial of losartan to lower blood pressure in severely hypertensive children younger than 6 years of age. The reimbursement for this work is deposited into a research fund for medical students and residents who lack another source of support for their research. The second is to conduct a clinical trial of fish oil for elevated triglycerides in adolescents. No clinical trial data on medication use for elevated triglycerides in children exist, and the NHLBI-sponsored statement made no recommendations about use of medications for this purpose. This money is provided as a grant, the trial is completely of my design, and my research program manages the trial without input from the company. The company has no oversight role for the trial, other than that I have to recruit a specified number of patients and complete a manuscript to receive payment. My reimbursement from the grant is for managing the coordinating center of the trial (there are 3 sites), analyzing the data, and performing various medical duties in conducting the trial.
The NHLBI guideline was strictly conducted according to the recommendations of the Institute of Medicine (IOM) and other bodies that are interested in guideline integrity. Several thousand research papers meeting preselected criteria as evidence were considered in the evaluation of recommendations for 14 different risk factors. A paper describing the panel’s process has recently been published. The Lipid section of the guideline was drafted by the three members of the panel who have the most knowledge about lipids. However, this group’s recommendation was debated and voted on by the entire committee, which had a diverse composition, consistent with IOM recommendations. NHLBI officials, cognizant of conflict of interest issues, were also present and contributed to the discussion. With one exception, none of the other 11 committee members had participated in an industry-sponsored cholesterol-drug trial.
All the arguments discussed in critiques of the guideline emerged in this debate; in fact, the committee felt so strongly about the evidence limitations that a specific chapter of the guideline was written to include them, particularly the cost issue. Nonetheless, the committee, with one exception, voted for universal screening because the weight of the evidence review favored this decision despite evidence gaps. The weight came from these facts: atherosclerosis begins in the second decade of life; this atherosclerosis (including future atherosclerosis) is strongly related to non-HDL cholesterol levels; a small but significant number of children can be identified with high cardiovascular risk; treatments highly likely to be successful are available; and genetic diseases causing both high and low LDL cholesterol were highly suggestive of the risk of high LDL cholesterol and of the benefit of lifetime low LDL cholesterol levels.
It is inaccurate of Dr. Stein to say that the evidence evaluation for the cholesterol treatment recommendation does not reflect the types of studies included to support the recommendation. The grade given is B (not A): evidence from clinical trials with potential flaws, high-quality and consistent observational studies, and Mendelian randomization (genetic research). I hope that the guideline will ultimately lead to high-quality clinical trials of lipid-lowering treatment, initiated by sponsors other than industry.
Included in guideline development (and recommended by the IOM) was a period for public comment, review by several interested professional organizations, and internal review by both NHLBI and DHHS. Notably, despite receiving over a thousand critiques, only one challenged the universal screening for cholesterol recommendation, and that critique presented only opinion rather than evidence against the recommendation. Given the committee’s deliberations and the extensive external review prior to publication, it is my opinion that the argument against cholesterol screening in children is a well-considered — but a minority — position.
I would like to challenge several other arguments in Dr. Stein’s post — as the extensive evidence review, conducted using a process now considered “state of the art,” did not support these assertions. Most important, no published evidence of harm from cholesterol screening exists. Guidelines regarding cholesterol testing have been in place for about 20 years, with no research demonstrating harm from this practice. In West Virginia, where universal screening of cholesterol occurs in schools voluntarily, over 60% of children consent to testing, and no harm has been identified in this program. Our evidence review specifically sought publications with side effects or unintended consequences of screening — and none were found. The age for cholesterol screening was chosen partly because most states require a medical examination for admission to a public school, and blood is often drawn for other purposes. Girls, on average, do have slightly higher total-cholesterol levels than boys — but women have equal rates of heart disease as men, only at slightly older ages. The number of cases of familial hypercholesterolemia missed by current guidelines is significant. Indeed, one of the reasons for the universal screening recommendation was that numerous studies conducted since the 1992 guideline was published suggest that 25% to 30% of cases are missed by selective screening. Given the frequency of 1:300-500 in the population, this is not a trivial number (and there is no gender bias in these levels). The dietary guidelines presented are for all children, not just those with high cholesterol. One of the reasons for delay in publication was the need to synchronize the recommendations with the evidence-based 2010 Dietary Guidelines for Americans.
The cardiovascular health of children has deteriorated significantly, principally because of the obesity epidemic and poor nutrition habits. Many, indeed almost one third of U.S. children, have a different type of dyslipidemia (low HDL cholesterol and high triglycerides), and it has been recommended, prior to the publication of the universal screening recommendations, that these children have their lipids checked. I am concerned that the debate about the usefulness of checking cholesterol might send the wrong message to the general population — namely, that accurately knowing your cardiovascular risk is not useful for this vulnerable population.
A key tenet of the IOM recommendations for development of guidelines is that they undergo periodic review with reconsideration of the evidence. For the lipid section in particular, the evidence gaps identified by Dr. Stein and others cited in his post had insufficient weight as evidence compared with the many studies supporting the final recommendations. Nonetheless, the current guideline should not be the last word. Evidence gaps point the way for future research, and this new work should inform future deliberations about this important issue.
Note: In addition to James Stein’s post, Larry Husten has written a post on industry PR efforts to influence the debate on cholesterol screening for children.
August 16th, 2012
Cochrane Review: Benefits of Treating Mild Hypertension Not Clear
The benefits of treating patients with mildly elevated blood pressure who are free of cardiovascular disease are unclear, according to a Cochrane review. These findings conflict with current hypertension treatment guidelines in the U.S., Canada, and Europe.
Researchers analyzed data from nearly 9000 participants in four trials and found no obvious benefit of drug treatment in patients with mild hypertension (systolic BP, 140 to 159 mm Hg and/or diastolic BP, 90 to 99 mm Hg) in terms of total mortality or cardiovascular events at 5 years’ follow-up. They did, however, see an increased likelihood of drug withdrawal due to adverse effects (relative risk, 4.8).
August 15th, 2012
Observational Study Fills Gaps in Understanding AF Patients with Kidney Disease
Larry Husten, PHD
Although people with atrial fibrillation (AF) and people with chronic kidney disease (CKD) are at elevated risk for stroke and other vascular events, people with both conditions have not been well studied, since those with CKD have been excluded from most clinical trials of stroke prevention for AF. The problem is further compounded because the presence of CKD increases the risk for bleeding associated with anticoagulation.
Now, a large observational study from Denmark published in the New England Journal of Medicine provides new evidence to better estimate risk and to help determine a treatment strategy for patients with both AF and CKD. Jonas Bjerring Olesen and colleagues analyzed data from 132,372 Danish patients with AF and found 3587 who also had chronic CKD and 901 who received dialysis or a transplant.
Compared to patients with no renal disease, patients with CKD were at significantly elevated risk for stroke or thromboembolism, bleeding, MI, and death:
Stroke or thromboembolism (event rate per 100 person-years):
- No renal disease: 3.61
- Chronic CKD: 6.44
- CKD requiring renal replacement therapy: 5.61
Bleeding (event rate per 100 person-years):
- No renal disease: 3.54
- Chronic CKD: 8.77
- CKD requiring renal replacement therapy: 8.89
Myocardial infarction (event rate per 100 person-years):
- No renal disease: 1.88
- Chronic CKD: 5.81
- CKD requiring renal replacement therapy: 5.98
Death (event rate per 100 person-years):
- No renal disease: 11.21
- Chronic CKD: 38.65
- CKD requiring renal replacement therapy: 29.35
The investigators found that warfarin but not aspirin reduced the risk for stroke or thromboembolism in the patients with CKD. Both warfarin and aspirin were associated with an increased risk for bleeding, however. “Thus,” the authors wrote, “the net clinical effect of warfarin treatment requires careful assessment in patients with chronic kidney disease, and the data do not provide clear guidance regarding indications for anticoagulant therapy in patients with both atrial fibrillation and chronic kidney disease. Certainly, close monitoring of the international normalized ratio is required when warfarin is administered.”
August 13th, 2012
Troponin Test May Allow Rapid MI Rule-Out in the Emergency Department
Larry Husten, PHD
More than three-quarters of people with chest pain can be triaged within an hour of arrival at the emergency department with a novel strategy utilizing high-sensitivity cardiac troponin (hs-cTnT), according to a study from Switzerland published in the Archives of Internal Medicine. The strategy is promising, according to an accompanying editorial, but much work remains before it can be implemented in clinical practice.
Tobias Reichlin and colleagues first studied 436 patients and developed a treatment algorithm utilizing hs-cTnT baseline changes and absolute changes over the initial hour. The algorithm was then tested in a second validation cohort of 436 patients, with the following results:
- 60% were classified as “rule-out”
- 17% were classified as “rule-in”
- 23% required further observation
- Overall sensitivity and negative predictive value: 100% for rule-out
- Specificity for rule-in: 97%
- Positive predictive value for rule-in: 84%
- Prevalence of MI in the observational group: 8%
- 30-day survival: 99.8% in the rule-out group, 98.6% in the observational group, and 95.3% in the rule-in group
The authors claim that their strategy “may obviate the need for prolonged monitoring and serial blood sampling in 3 of 4 patients.”
In an accompanying comment, L. Kristin Newby writes that the Swiss study “is a major advance in understanding the application of hsTn testing that with continued development could substantially improve evaluation of ED patients with suspected MI.” However, she notes that the excellent results obtained in this initial study will probably not be equalled in the real world. In addition, she writes, “although touted as ‘simple’ by the authors, the need for multicomponent algorithms that are different for rule-in and rule-out and that vary by age group or other parameters will challenge application by busy clinicians unlikely to remember or accurately process the proposed algorithm. As such, it will be imperative that hsTn algorithms, if validated, are built into clinical decision support layered onto electronic health records so that testing results are provided electronically to physicians along with the algorithmic interpretation to allow systematic application in triage and treatment.”
August 13th, 2012
Selections from Richard Lehman’s Literature Review: August 13th
Richard Lehman, BM, BCh, MRCGP
CardioExchange is pleased to reprint selections from Dr. Richard Lehman’s weekly journal review blog at BMJ.com. Selected summaries are relevant to our audience, but we encourage members to engage with the entire blog.
JAMA 8 Aug 2012 Vol 308
More Research is…Readily Available (pg. 575): When Stephen Lock was editor of the BMJ, he banned the expression “further research is needed” on the grounds that further research is always needed. Here is a piece by John Oiannidis which argues that further research is very often badly needed – and that it is very often instantly achievable, simply by using data that has already been collected. But to do that, researchers need to know where the data are to be found, and to have automatic access to it. The rather clunky title of this essay, “The Importance of Potential Studies That Have Not Existed and Registration of Observational Data Sets”, does little justice to a powerful discussion of issues that lie at the very heart of the coming revolution in research methodology. We are – I hope – on the verge of an era where all data from every human trial ever carried out will be available to any researcher; and more than that, an era when every clinical encounter will in principle be available for analysis after the removal of patient identifiers. As Ioannidis points out, much of this is already possible and is not being done. The “potential studies that have not existed” which he refers to are investigations that could settle contentious issues – if only people would take the trouble to look at datasets that are already in existence.
(Note for pedants: I have deliberately used “data” both as a plural and as a singular in two sentences, because this is common practice. I bet that within ten years, data will have followed “media” and become uniformly singular. Personally, I don’t like it; but that’s language for you. It’s like a bacteria. Neuter plurals counted as singular in ancient Greek.)
Normal Weight, Diabetes, and Risk of non-Cardiovascular Death (pg. 581): Here’s another significant insight into the enigma that is labelled “type 2 diabetes”. We all know of people who have this condition despite being of normal weight (BMI less than 25), and in fact in the five large cohorts looked at here, the proportion of subjects with normal BMI at the time of onset of diabetes varied between 9 and 21%, with a mean of 12%. The striking finding of this analysis is that these individuals have a doubling of mortality risk at 15 years compared with overweight or obese individuals with T2DM. And oddly enough, this is mainly accounted for by non-cardiovascular causes of death.
Total Cholesterol Levels in U.S. Kids (pg. 591): But now for a really puzzling observation: between 1988 and 2006, the mean level of total cholesterol in American children has declined. During this period, US kids aged between 6 and 19 have become fatter and less active. I don’t know what can account for this, and nor does the writer of the editorial that accompanies it, except to call it a reason for optimism.
Lancet 11 Aug 2012 Vol 380
Statins and Diabetes (pg. 565): One way to push people over the arbitrary threshold of 7.0 mmol/L fasting glucose is to give them a thiazide diuretic: another is to give them a statin. They are then officially “diabetic”, but does that mean that they are condemned to progressive beta-cell failure? In the case of thiazides, the answer is simply no. In the case of statins, we don’t know: but we do know that the cardiovascular benefits of continuing the drug easily outweigh any negative effects from hyperglycaemia. This analysis of data from the JUPITER trial confirms that this applies across the glucose range, including people at high risk of diabetes.
HDL-C and MI Protection (pg. 572): I tried counting the authors of this celebrated mendelian randomisation study of plasma high density lipoprotein cholesterol and myocardial infarction, but I had to give up due to vertigo. There must be about 120. If you believe them, there is no likelihood of any causal link between HDL-C levels and protection from MI. If you don’t believe them, you can start all over again: perform two mendelian randomisation analyses. First, use as an instrument a single nucleotide polymorphism (SNP) in the endothelial lipase gene (LIPG Asn396Ser) and test this SNP in 20 studies (20 913 myocardial infarction cases, 95 407 controls). Second, use as an instrument a genetic score consisting of 14 common SNPs that exclusively associate with HDL cholesterol and tested this score in up to 12 482 cases of myocardial infarction and 41 331 controls. As a positive control, also test a genetic score of 13 common SNPs exclusively associated with LDL cholesterol.
Statins: Treat Everyone, Treat to Risk, or Treat to Target? (pg. 581): You’ll note that the preceding study did confirm that LDL-C is causal in adverse cardiovascular events; and only LDL-lowering strategy that we know to prevent CV events is statin therapy. Now there are three camps where statins are concerned: the “put them in the water supply” camp at one extreme, and the “treat to target” camp at the other, and half way between the “treat to risk” camp. This meta-analysis of individual data from 27 randomised trials confirms that statins lower cardiovascular risk at all levels, with the most pronounced effect of course at the top of the risk scale. Some friends of mine would argue that statins taken from the age of say 16 would prevent atheroma altogether; others among my friends rail at the mass medication of society and seem to wish that statins had never been invented; I myself think they should be offered to everyone above some arbitrary age, say 50. In their editorial, Shah Ebrahim and Juan Casas appear to agree, though their last paragraph does offer the tempting alternative of moving to Mauritius.
Reviewing Hypertension Treatment (pg. 591): The management of “hypertension” in primary care is one of the most boring jobs we do, and I fear we don’t do it very well. Most of the individuals who trudge regularly into our surgeries every six months gain no benefit whatsoever from the drugs we give them – the number needed to prevent one stroke is typically 100-400 – and most of the time we don’t even measure their BP adequately anyway. We medicate the herd for small benefit, while often struggling with the patients at highest risk. Fortunately most of the drugs we use are cheap and safe, but the market is so huge that drug companies continue to develop new drugs which might work for some subgroups of patients with resistant hypertension. This review usefully lists them, but the real revolution is likely to come from a different direction – devices and procedures which permanently reset the sympathetic nervous system. In ten years’ time, I wouldn’t be surprised to see renal sympathetic denervation becoming a routine procedure: or will it be continuous carotid stimulation?
BMJ 11 Aug 2012 Vol 345
Clopidogrel and PPIs: In laboratory tests of platelet function, proton pump inhibitors cancel out the effect of clopidogrel because they inhibit the P450 2C19 enzyme. So if you look at a cohort of people who are taking clopidogrel (with aspirin) you might expect to find a higher rate of coronary events in those who are also taking a PPI. And in fact, as this study from the UK GP Research Database shows, there is a 30% or so higher risk in this group, if you look at the cohort as a whole. In fact there is an even higher risk difference in non-vascular death, which begins to make you wonder whether there is something else going on in these people who are taking PPIs. Maybe the groups are not comparable: and the cunning authors then go on to test this hypothesis by looking at within individual differences in CV outcomes during periods on and off PPI treatment. Here the effect direction is reversed. This is intriguing, and probably means that most people can take clopidogrel and PPIs together with impunity: but there are some loose ends here. Somebody needs to repeat this exercise using another database, looking more precisely at the individual clinical reasons for co-treatment with PPIs, aspirin and clopidogrel.
Pharma Innovation: The most important contribution to this week’s BMJ is undoubtedly an analysis by Donald Light and Joel Lexchin of the truth behind the “innovation crisis” in the pharmaceutical industry. They demonstrate that it is a widely touted myth, aimed at putting pressure on regulatory agencies to help the poor ailing industry by setting a lower bar for licensing new products. In fact only one in ten products has any added clinical value, whereas the FDA is currently granting “priority status” to 44% of new drugs. The remaining 90% simply drive up health costs without adding benefit, and most industry effort is put into developing such me-too drugs. And this is a highly successful business model: pharmaceutical R&D budgets rose by $34.2bn between 1995 and 2010, while profits rose by $200bn. The pharmaceutical industry could innovate if it had the incentives to, and indeed does do so one time in ten: but most of the time it is creaming off easy money from health systems that can ill afford it.
