Perioperative Mischief: The Price of Academic Misconduct
Vineet Chopra, MD, MSc,a Kim A. Eagle, MDb
aDivision of General Internal Medicine and bDivision of Cardiovascular Medicine, University of Michigan Health System, Ann Arbor.
Recent allegations of fraud committed by one of the most prolific researchers in perioperative medicine,Don Poldermans, have left many clinicians in a state of disbelief. With over 500 peer-reviewed publica-tions, Poldermans heavily influenced the clinical practice of perioperative beta-blockers and statins innoncardiac surgery, shaping guidelines and national policies on the use of these treatments. The effects offraud in perioperative medicine are particularly caustic owing to a profound domino effect. Manyinvestigators devoted their academic careers to following the footsteps of investigators such as Poldermans.
Similarly, funding agencies supported this line of enquiry, incurring significant cost and expense. Mostimportantly, hundreds of patients were exposed to treatments that may have been harmful in an effort toadvance this research agenda. How should perioperative clinicians utilize beta blockade now that aconsiderable portion of the literature is enshrouded in uncertainty? In this brief review, we reiterate andemphasize basic principles about the indications and administration of perioperative beta blockade.
Because research misconduct in perioperative medicine can be so damaging, we present strategies toprevent such events in the future. Without such reform, fraud in research may very well continue. The pricefor such misconduct is simply too great to pay.
2012 Elsevier Inc. All rights reserved. The American Journal of Medicine (2012) 125, 953-955 KEYWORDS: Beta-blocker; Fraud; Perioperative
The perioperative community is in a state of turmoil after fictitious data and submitting two reports to conferences one of its preeminent investigators, Don Poldermans, was which included knowingly unreliable data” were cited as fired from Erasmus Medical Center in the Netherlands.
reasons for his Sadly, warning signs did exist Poldermans and his research team heavily influenced the before this avalanche of allegations. In a 2005 systematic literature on the utility of beta-blocker and statin treatment review of randomized trials stopped early for benefit, Mon- in noncardiac surgery. With well over 500 peer-reviewed tori et al isolated a prematurely halted Poldermans study, publications in high-impact journals, their work has in- finding that “the very large treatment effect is likely too formed national policies, evidence-based guidelines, scien- good to be true. It is inconsistent with the researchers’ tific discourse, and the management of millions of patients.
expectations; with the magnitude of effect (ie, RRs [relative Thus, when allegations of scientific misconduct emerged, risks] of 0.65 to 0.85) of ␤-blockers in tens of thousands of perioperative practitioners across the world were left in a patients with acute myocardial infarction or chronic man- agement of congestive heart failure; with results in day-to- The precise nature of the accusations surrounding Pol- day clinical practice; and with results of other dermans is unclear. In an official press release, concerns Indeed, never before (nor since) this initial study has such a including “carelessness in collecting the data for his re- profound effect of perioperative beta blockade on cardiac search, using patient data without written permission, using Unfortunately, this is not the first time that controversy Funding: None.
has surfaced in the perioperative literature. Indeed, the first Conflict of Interest: None.
Authorship: Both authors had access to the data and a role in writing
ever perioperative beta-blocker trial experienced a healthy dose of skepticism when benefit was reported not immedi- Requests for reprints should be addressed to Vineet Chopra, MD, MSc, ately, but months after surgery in those randomized to this Division of General Internal Medicine, University of Michigan Health For an intervention designed to reduce cardiac System, 300 North Ingalls Building, Room 7C-17, Ann Arbor, MI 48109.
events around the time of surgery, this seemed an odd result.
0002-9343/$ -see front matter 2012 Elsevier Inc. All rights reserved.
The American Journal of Medicine, Vol 125, No 10, October 2012 Further criticism came from scrutiny of the study’s meth- clinicians became more willing to risk the development ods, which excluded immediate deaths that (if included) of hypotension, bradycardia, and stroke in their quest to would have statistically nullified the observed benefit from achieve heart-rate control. Guideline-writing committees beta-blocker While the integrity of the authors of the American College of Cardiology Foundation/ and that of the study has never been questioned, the trial is American Heart Association issued focused updates to CLINICAL SIGNIFICANCE
● The evidence regarding the cardiac benefit of perioperative beta blockade is in doubt owing to allegations of research fraud.
● Perioperative beta blockade must be ad- ● Greater oversight and structural reform is necessary to prevent perioperative re- inglorious origins of this literature, it is telling that conver- 1. First, we must exercise clinical restraint and recognize sations of integrity sparsely occurred. For example, a sys- the double-edged sword that is perioperative beta- tematic review 6 years ago opined that “design flaws in the blocker treatment. While beta-blockers blunt surgical first clinical trials advocating the use of ␤-blockade in hemodynamic stress to preserve myocardial oxygen sup- non-cardiac surgery made it very likely that these results ply, the price paid is failure to augment cardiac output in were Rather than investigate these concerns further, situations where blood loss, systemic vasodilation, and more daring studies to demonstrate a positive effect fromperioperative beta blockade were launched, effectively intravascular volume shifts are the norm. This scenario is overselling the benefits while underplaying the risks asso- more compelling in the elderly, who, owing to preexist- ciated with this treatment. The result was a research agenda ing vascular stiffness and reductions in peak heart rates, dedicated to larger, more expensive, and purportedly more have narrower volume “set-points” and are more at risk definitive studies. For example, a manuscript summarizing of systemic collapse. We must therefore limit beta block- the design of the Perioperative Ischemic Evaluation ade to patients with established coronary artery disease (POISE) study reported how Ͼ6300 patients in 182 cen- already on this therapy, or in those with documented ters and 21 countries had been recruited for the largest ischemia and concerning cardiac symptoms facing high- randomized perioperative beta-blocker trial to risk surgery. Any generalizations beyond these subsets Many design elements of this seminal trial (including the are not supported by current evidence and may be asso- controversial high dose of metoprolol succinate admin- istered to beta-blocker naive patients), were likely in- 2. Second, in a post-POISE era, we can no longer afford to formed by the evidence available at the time. One might overlook the risks associated with this treatment. When- say that the ripple arising from the early perioperative ever considered, beta-blockade should be started weeks, beta-blockade studies had morphed into a veritable not days, before surgery with slow and careful titration to individual heart rate and blood pressure. Correspond- Undoubtedly, the greatest tragedy of the perioperative ingly, unless concerns about hemodynamic instability beta-blocker saga lies in the untold harm it has caused to surface, beta-blockers must not be discontinued abruptly patients undergoing noncardiac surgery. Trials that in patients undergoing surgery in order to avoid precip- linked perioperative heart-rate control to improved car- itating a withdrawal state leading to sympathetic activa- diac outcomes indubitably steered clinical practice tion and cardiac events. However, even when adminis- towards more aggressive beta blockade and tighter tered for appropriate indications, careful monitoring for heart-rate control. Almost paradoxically, perioperative hypotension and bradycardia coupled with judicious withholding of medication must be planned. Without erative research community must recalibrate its ethical, such supervision, perioperative beta blockade remains a moral, and scientific compass to ensure that this never dangerous and potentially lethal practice.
happens again. The price of academic misconduct is simply 3. Third, we must strive to improve oversight of perioper- ative clinical trials. With several sources of bias avail-able, perioperative researchers, journals, and funding References
agencies must develop and nurture equipoise by ensur- 1. Erasmus Medical Center. Official press release. Available at: ing the integrity of study Falsifying trial results to fit an investigator’s agenda rather than report- ing the truth epitomizes the most extreme outcome from 2. Montori VM, Devereaux PJ, Adhikari NK, et al. Randomized trials failing to address this challenge. How may we address stopped early for benefit: a systematic review. JAMA. 2005;294(17):2203-2209.
this problem? First, as good housekeeping begins at 3. Mangano DT, Layug EL, Wallace A, Tateo I. Effect of atenolol on home, better oversight from organizations where re- mortality and cardiovascular morbidity after noncardiac surgery. Mul- search is conducted is needed. Parent institutions should ticenter Study of Perioperative Ischemia Research Group. N Engl develop and maintain auditing programs to ensure that investigators are following appropriate research ethics, 4. Chopra V, Plaisance B, Cavusoglu E, Flanders SA, Eagle KA. Peri- policies, and procedures. Second, the submission of raw operative beta-blockers for major noncardiac surgery: Primum NonNocere. Am J Med. 2009;122:222-229.
data and databases along with manuscripts should be 5. Wetterslev J, Juul AB. Benefits and harms of perioperative beta- encouraged. This is especially true when extraordinary blockade. Best Pract Res Clin Anaesthesiol. 2006;20:285-302.
treatment size(s) or practice-changing recommendations 6. Devereaux PJ, Yang H, Guyatt GH, et al. Rationale, design, and are made. These data should go through careful, unbi- organization of the PeriOperative ISchemic Evaluation (POISE) trial: ased scrutiny by methodological experts including peer- a randomized controlled trial of metoprolol versus placebo in patients reviewers and in-house statistical editors. While this ap- undergoing noncardiac surgery. Am Heart J. 2006;152:223-230.
7. Fleischmann KE, Beckman JA, Buller CE, et al. 2009 ACCF/AHA proach may not prevent falsification of data, it raises the focused update on perioperative beta blockade: a report of the Amer- bar for such practice and creates another “check-point” ican college of cardiology foundation/American heart association task that researchers must pass. Third, for this approach to be force on practice guidelines. Circulation. 2009;120:2123-2151.
successful, journals must develop infrastructure and pol- 8. Devereaux PJ, Yang H, Yusuf S, et al. Effects of extended-release icies to secure these data from inadvertent leakage or metoprolol succinate in patients undergoing non-cardiac surgery dissemination. As the cost and effort required to review (POISE trial): a randomised controlled trial. Lancet. 2008;371:1839-1847.
this information can be substantial, funding agencies 9. Bangalore S, Wetterslev J, Pranesh S, Sawhney S, Gluud C, Messerli should develop mechanisms to support this FH. Perioperative beta blockers in patients having non-cardiac sur-gery: a meta-analysis. Lancet. 2008;372:1962-1976.
The consequences of research fraud are legion. As viv- 10. Chopra V, Davis M. In search of equipoise. JAMA. 2011;305:1234- idly illustrated by the perioperative beta-blocker studies, inestimable expense and hundreds of investigators and pa- 11. White C. Suspected research fraud: difficulties of getting at the truth.
tients have become victims of misinformation. The periop-


Eugenio anessi pessina


Microsoft word - 01 prologo.doc

Nel verano del pasado año 2004 la ciudad de Córdoba tenía el privilegio de presenciar, envuelta en el mágico entorno de los jardines del ' Alcázar y dentro de los conciertos programados por el Festival de la Guitarra, una actuación en directo de Paco de Lucía con su actual grupo. Presentaban su nuevo disco Cositas buenas, y nos deleitaban con versiones renovadas de temas pertenecientes

© 2010-2018 Modern Medicine