Randomized Cluster Crossover Trials for Reliable, Efficient, Comparative Effectiveness Testing: Design of the Prevention of Arrhythmia Device Infection Trial (PADIT)

doi:10.1016/j.cjca.2013.01.020

Canadian Journal of Cardiology

Volume 29, Issue 6, June 2013, Pages 652-658

https://doi.org/10.1016/j.cjca.2013.01.020 Get rights and content

Abstract

Randomized clinical trials are a major advance in clinical research methodology. However, there are myriad important questions about the effectiveness of treatments used in daily practice that are not informed by the results of randomized trials. This is in part because of important limitations inherent in the methodology of randomized efficacy trials which are performed with tight control of inclusion, exclusion, treatment, and follow-up. This approach enhances evaluation of clinical efficacy (performance in controlled situations) but increases complexity and is not well suited to test clinical effectiveness (performance under conditions of actual use). The cluster crossover trial is a new concept for efficient comparative effectiveness testing. Deep tissue infection occurs in 2% of patients after arrhythmia device implantation, usually requires system extraction, and increases mortality. There is variation in antibiotic prophylaxis used to reduce implanted device infections. To efficiently evaluate the comparative effectiveness of antibiotic strategies now in use, we designed a cluster crossover clinical trial, which randomized implanting centres to 1 of 2 prophylactic antibiotic strategies, which became the standard care at the centre for 6 months, followed by crossover to the other strategy, rerandomization, and second crossover. This method greatly reduces trial complexity because it aligns study procedures with usual clinical care and increases generalizability. Pilot studies have tested the feasibility and an 10,800-patient trial, funded by the Canadian Institutes of Health Research, is now under way. The cluster crossover randomized trial design is well suited to efficiently test comparative effectiveness of existing treatments where there is variability of practice, clinical equipoise, and minimal risk.

Résumé

Les essais cliniques aléatoires constituent une avancée majeure en méthodologie de la recherche clinique. Cependant, de nombreuses questions importantes portant sur l'efficacité des traitements utilisés dans la pratique quotidienne ne s'appuient pas sur les résultats des essais aléatoires. Cela tient en partie en raison des limites importantes inhérentes à la méthodologie des essais aléatoires sur l'efficacité qui sont réalisés en maîtrisant rigoureusement l'inclusion, l'exclusion, le traitement et le suivi. Cette approche améliore l'évaluation de l'efficacité clinique (performance dans des situations maîtrisées), mais augmente la complexité et ne se prête pas bien à l'essai d'efficacité clinique (performance dans des conditions d'utilisation actuelle). L'essai croisé par grappes est un nouveau concept d'essai comparatif de l'efficacité. L'infection de tissus profonds apparaît chez 2 % des patients après l'implantation d'un dispositif contre l'arythmie, nécessite habituellement l'extraction de l'appareil et augmente la mortalité. La prophylaxie antibiotique utilisée pour réduire les infections liées à l'implantation du dispositif diffère. Pour évaluer efficacement l'efficacité comparative des stratégies antibiotiques actuellement utilisées, nous avons réalisé un essai clinique croisé par grappes qui a réparti au hasard les centres d'implantation à 1 des 2 stratégies de prophylaxie antibiotique, qui est devenue la norme de soins du centre pour 6 mois, suivi par un essai croisé pour l'autre stratégie, d'une nouvelle répartition aléatoire et d'un deuxième essai croisé. Cette méthode réduit grandement la complexité de l'essai puisqu'elle harmonise les procédures de l'étude aux soins cliniques habituels et augmente la généralisabilité. Les études pilotes ont testé la faisabilité, et un essai de 10 800 patients financé par les Instituts de recherche en santé du Canada est en cours. La conception d'essai aléatoire croisé par grappes se prête bien pour tester efficacement l'efficacité comparative de traitements existants où il existe une variabilité de la pratique, une incertitude clinique absolue et des risques minimaux.

Section snippets

The Need for Efficient Testing of Clinical Effectiveness

Randomized clinical trials are a major advance in clinical research methodology which have generated considerable knowledge about the efficacy of therapies and have facilitated introduction of many advances in treatment. However, there are myriad important questions about the effectiveness of treatments used in daily practice that are not informed by the results of randomized trials. Thus, for many clinical decisions, physicians do not know “what works best,” and they must rely on intuition,

Problem of Arrhythmia Device Infection

Cardiac arrhythmia devices (pacemakers, implantable cardioverter defibrillators [ICDs], and cardiac resynchronization therapy) are life-saving treatments that are widely used. In 2006 there were 27,286 device implants or replacements in Canada (21,054 pacemakers, 4812 ICDs, and 1420 cardiac resynchronization therapy). Device infection occurs in approximately 2% of all cases,2, 3, 4, 5, 6, 7 and it is a serious and potentially catastrophic consequence of these procedures.² Device infection rates

Prevention of Device Infection

It is currently recommended that a single preoperative dose of cephalosporin be used at the time of surgery to reduce the risk of device infection. This recommendation is based on several small randomized trials in patients who received devices, and on a meta-analysis.4, 13 Present professional guidelines (American Heart Association/American College of Cardiology) advocate preoperative cefazolin (or vancomycin in patients who are penicillin-allergic), a practice that is widely practiced in

PADIT Pilot Studies and Development of the Cluster Crossover Design

There is a need to establish if the additional antibiotic measures are effective, but the challenges of a randomized efficacy trial to test an additional antibiotic strategy are great. Based on an estimated infection rate of 2% in higher risk patients with conventional single-antibiotic prophylaxis, we calculated that enrolment of almost 11,000 patients would be required to have 80% power to detect a 35% reduction in infection rates. Although technically feasible with sufficient funding, grants

Randomized Cluster Crossover Trial

Cardiac rhythm device implantation in Canada is largely performed at highly specialized centres that implant from 200 to 1000 devices per year. To achieve these high volumes with low complication rates, these centres use highly standardized operating procedures, which include quality assurance, careful monitoring of cases, and highly trained teams of specialized health care professionals. The clinical effectiveness of interventions delivered at the time of implant to prevent device infection

PADIT Cluster Crossover Trial Design

The primary hypothesis of PADIT is that a strategy of incremental antimicrobial prophylaxis will reduce the risk of hospitalization for device infection, compared with single dose preoperative cefazolin, in patients undergoing higher risk arrhythmia device procedures. All patients at each centre will receive the antibiotic strategy to which the centre is randomized, as standard procedure during each study period, but only higher-risk patients (protocol-defined), will be included in the

Minimal Risk of the Antibiotic Strategies

The antibiotic strategies tested in this protocol are already widely used in Canada to prevent surgical site infections, with favourable safety profiles, especially in the very short periods of treatment planned. The fact that these antibiotics are already routinely used at many device implant centres, alone or in combination, despite any evidence from randomized trials of their effectiveness, clearly shows that physicians consider the risks of these treatments to be very low. Cephazolin and

Statistical Considerations

There are unique statistical considerations for a cluster-randomized trial which have been well described.25, 26 Because the unit of randomization is the centre, sample size and statistical power depend greatly on the intraclass correlation (ICC), which is a measure of the variation in outcomes between centres. If there is a high ICC, then there is a need for more centres, and the number of patients per centre is less relevant. The sample size for present cluster randomized crossover design for

Ethics Committee Approval

It is often not possible for a patient to choose to avoid the intervention or to give consent to it in a cluster randomized trial because the intervention is applied at the level of the health system and not the patient. Several observers have recognized that cluster randomized trials present a challenge to the usual approach to clinical research.21, 22 However, this type of trial is becoming much more common in response to the need for more information about knowledge translation and clinical

Study Organization

This study was developed from the Arrhythmia Working group of CANNeCTIN in collaboration with the Canadian Heart Rhythm Society Device Committee. CANNeCTIN is a clinical trials network funded by the Canadian Institutes of Health Research (CIHR) to increase clinical trials research capacity throughout Canada. The Population Health Research Institute in Hamilton serves as the hub of the network and has provided coordination for the working groups and pilot studies. In the case of PADIT, the 2

Discussion

There is great variation in everyday clinical practice, often because of lack of knowledge about the clinical effectiveness of different treatment options. This variation reveals the deficiency of knowledge but highlights the opportunity to improve care if one can mobilize these differences to test the comparative effectiveness of practices in use. The cluster crossover design is a way to do that. Randomized clinical trials are the gold standard for valid assessment of treatments but are often

Funding Sources

Dr Krahn is a Career Investigator of the Heart and Stroke Foundation of Ontario (CI6498). The study was supported by the CANNeCTIN network (CIHR grant 88370), and the impending trial (CIHR grant 119442).

Disclosures

The authors have no conflicts of interest to disclose.

References (33)

T. Mela et al.
Long-term infection rates associated with the pectoral versus abdominal approach to cardioverter-defibrillator implants
Am J Cardiol
(2001)
M.R. Sohail et al.
Management and outcome of permanent pacemaker and implantable cardioverter-defibrillator infections
J Am Coll Cardiol
(2007)
A.J. Greenspon et al.
16-year trends in the infection burden for pacemakers and implantable cardioverter-defibrillators in the United States
J Am Coll Cardiol
(2011)
L.J. Gula et al.
Arrhythmia device lead extraction: factors that necessitate laser assistance
Can J Cardiol
(2008)
C.L. Byrd
Managing device-related complications and transvenous lead extraction
M.R. Sohail et al.
Infective endocarditis complicating permanent pacemaker and implantable cardioverter-defibrillator infection
Mayo Clin Proc
(2008)
F.H. Edwards et al.
The society of thoracic surgeons practice guideline series: antibiotic prophylaxis in cardiac surgery, part I: duration
Ann Thorac Surg
(2006)
J.A. Cairns et al.
Canadian network and centre for trials internationally (CANNeCTIN): a national network for Canadian-led trials in cardiovascular diseases and diabetes mellitus
Can J Cardiol
(2010)
R. Platt et al.
Cluster randomized trials in comparative effectiveness research: randomizing hospitals to test methods for prevention of healthcare-associated infections
Med Care
(2010)
P.A. Gould et al.
Complications associated with implantable cardioverter-defibrillator replacement in response to device advisories
JAMA
(2006)

J.C. de Oliveira et al.

Efficacy of antibiotic prophylaxis before the implantation of pacemakers and cardioverter-defibrillators: results of a large, prospective, randomized, double-blinded, placebo-controlled trial

Circ Arrhythm Electrophysiol

(2009)

D. Klug et al.

Risk factors related to infections of implanted pacemakers and cardioverter-defibrillators: results of a large prospective study

Circulation

(2007)

M.R. Sohail et al.

Risk factor analysis of permanent pacemaker infection

Clin Infect Dis

(2007)

F.L. Paulin et al.

Management of infections involving implanted cardiac electrophysiologic devices

Curr Treat Options Cardiovasc Med

(2008)

S. Kapa et al.

Complication risk with pulse generator change: implications when reacting to a device advisory or recall

Pacing Clin Electrophysiol

(2007)

A. Da Costa et al.

Antibiotic prophylaxis for permanent pacemaker implantation: a meta-analysis

Circulation

(1998)

Cited by (50)

Adherent skin barrier drape use is associated with a reduced risk of cardiac implantable device infection: Results from a prospective study of 14,225 procedures
2024, Heart Rhythm O2
Cardiac implantable electronic device (CIED) infection is a costly and highly morbid complication. Perioperative interventions, including the use of antibiotic pouches and intensified perioperative antibiotic regimens, have demonstrated marginal efficacy at reducing CIED infection. Additional research is needed to identify additional interventions to reduce infection risk.
We sought to evaluate whether adherent skin barrier drape use is associated with a reduction in CIED infection.
A prospective registry of all CIED implantation procedures was established at our institution in January 2007. The registry was established in collaboration with our hospital infection prevention team with a specific focus on prospectively identifying all potential CIED infections. All potential CIED infections were independently adjudicated by 2 physicians blinded to the use of an adherent skin barrier drape.
Over a 13-year period, 14,225 procedures were completed (mean age 72 ± 14 years; female 4,918 (35%); new implants 10,005 (70%); pulse generator changes 2585 (18%); upgrades 1635 (11%). Of those, 2469 procedures (17.4%) were performed using an adherent skin barrier drape. There were 103 adjudicated device infections (0.73%). The infection rate in patients in the barrier use groups was 8 of 2469 (0.32%) as compared with 95 of 11,756 (0.8%) in the nonuse group (P = .0084). In multivariable analysis, the use of an adherent skin barrier drape was independently associated with a reduction in infection (odds ratio 0.32; 95% confidence interval 0.154–0.665; P = .002).
The use of an adherent skin barrier drape at the time of cardiac device surgery is associated with a lower risk of subsequent infection.
Benzodiazepine-Free Cardiac Anesthesia for Reduction of Postoperative Delirium (B-Free): A Protocol for a Multi-centre Randomized Cluster Crossover Trial
2023, CJC Open
Delirium is common after cardiac surgery and is associated with adverse outcomes. Administration of benzodiazepines before and after cardiac surgery is associated with delirium; guidelines recommend minimizing their use. Benzodiazepine administration during cardiac surgery remains common because of its recognized benefits. The Benzodiazepine-Free Cardiac Anesthesia for Reduction of Postoperative Delirium (B-Free) trial is a randomized cluster crossover trial evaluating whether an institutional policy of restricting intraoperative benzodiazepine administration (ie, ≥ 90% of patients do not receive benzodiazepines during cardiac surgery), as compared with a policy of liberal intraoperative benzodiazepine administration (ie, ≥ 90% of patients receive ≥ 0.03 mg/kg midazolam equivalent), reduces delirium. Hospitals performing ≥ 250 cardiac surgeries a year are included if their cardiac anesthesia group agrees to apply both benzodiazepine policies per their randomization, and patients are assessed for postoperative delirium every 12 hours in routine clinical care. Hospitals apply the restricted or liberal benzodiazepine policy during 12 to 18 crossover periods of 4 weeks each. Randomization for all periods takes place in advance of site startup; sites are notified of their allocated policy during the last week of each crossover period. Policies are applied to all patients undergoing cardiac surgery during the trial period. The primary outcome is the incidence of delirium at up to 72 hours after surgery. The B-Free trial will enroll ≥ 18,000 patients undergoing cardiac surgery at 20 hospitals across North America. Delirium is common after cardiac surgery, and benzodiazepines are associated with the occurrence of delirium. The B-Free trial will determine whether an institutional policy restricting the administration of benzodiazepines during cardiac surgery reduces the incidence of delirium after cardiac surgery.
Clinicaltrials.gov registration number: NCT03928236 (First registered April 26, 2019).
L’état confusionnel est fréquent après une chirurgie cardiaque et il est associé à des complications. L’administration de benzodiazépines avant et après une chirurgie cardiaque est associée à l’état confusionnel; dans les lignes directrices, on recommande de réduire leur utilisation au minimum. L’administration de benzodiazépines pendant une chirurgie cardiaque demeure fréquente, en raison des leurs bienfaits reconnus. L’essai B-Free (Benzodiazepine-Free Cardiac Anesthesia for Reduction of Postoperative Delirium ou l’anesthésie sans benzodiazépine en contexte de chirurgie cardiaque pour la réduction de l’état confusionnel postopératoire) est un essai à répartition aléatoire par grappes et avec permutation, visant à évaluer si une politique institutionnelle de restriction de l’administration peropératoire de benzodiazépines (c.-à-d. que ≥ 90 % des patients ne reçoivent pas de benzodiazépines durant une chirurgie cardiaque) réduit l’état confusionnel, comparativement à une politique d’administration peropératoire libérale de benzodiazépines (c.-à-d. que ≥ 90 % des patients reçoivent ≥ 0,03 mg/kg d’équivalent du midazolam). Des hôpitaux effectuant au moins 250 chirurgies cardiaques par année sont inclus dans l’essai si leurs équipes d’anesthésie cardiaque acceptent d’appliquer les deux politiques relatives aux benzodiazépines en vertu de la répartition aléatoire et si les patients sont évalués toutes les 12 heures, en ce qui a trait à l’état confusionnel postopératoire, dans le cadre des soins cliniques habituels. Les hôpitaux mettent en œuvre la politique d’administration restreinte ou libérale de benzodiazépines durant 12 à 18 périodes de permutation de 4 semaines chacune. La répartition aléatoire de l’ensemble des périodes a lieu avant le début de l’essai à l’hôpital; les établissements sont avisés de la politique qui leur est attribuée au cours de la dernière semaine de chaque période de permutation. Les politiques sont appliquées à tous les patients qui subissent une chirurgie cardiaque durant la période de l’essai. Le critère d’évaluation principal est l’incidence de l’état confusionnel dans les 72 heures suivant l’intervention chirurgicale. L’étude B-Free inclura au moins 18 000 patients qui subiront une chirurgie cardiaque dans 20 hôpitaux en l’Amérique du Nord. L’état confusionnel est fréquent après une chirurgie cardiaque, et les benzodiazépines sont associées à la survenue de l’état confusionnel. L’essai B-Free permettra de déterminer si une politique institutionnelle de restriction de l’administration de benzodiazépines durant une chirurgie cardiaque réduit l’incidence de l’état confusionnel après une telle chirurgie.
Clinicaltrials.gov registration number: NCT03928236 (First registered April 26, 2019).
Infections in interventional electrophysiology
2020, Annales de Cardiologie et d'Angeiologie
En fréquence et en gravité, l’implantation de stimulateurs cardiaques et de défibrillateurs est de loin l’activité cardiologique la plus risquée sur le plan infectieux. Le nombre d’implantations de dispositifs cardiaques implantables ne cesse d’augmenter avec près de 2 millions de poses par an au niveau mondial. Le profil des patients implantés se révèle parfois à haut risque infectieux. Plusieurs mesures permettent de réduire le risque d’infection de matériel. Le bénéfice de l’antibioprophylaxie systématique est bien démontré à condition de respecter ses modalités de prescription, en particulier concernant le choix de molécule et les délais d’administration avant l’intervention. Malgré les précautions prises lors de l’intervention chirurgicale (asepsie, antibiothérapie prophylactique…), le taux d’infections péri-opératoires est évalué à près de 2 %. Il s’agit d’une complication associée à une morbi-mortalité élevée et un coût important pour le système de santé. Les Staphylococcus aureus (SA) et epidermidis (SE) sont les agents pathogènes les plus souvent impliqués. La prévention est essentielle car le traitement d’une infection sur dispositif cardiaque endocavitaire est toujours difficile, du fait des résistances aux antibiotiques et surtout de la nécessité quasi systématique d’explanter tout le matériel. Le pace maker sans sonde et le défibrillateur sous cutané semblent être des alternatives à ces dispositifs cardiaques endocavitaires, avec certaines limites. Récemment, un consensus sur la prévention, le diagnostic et le traitement des infections de dispositifs cardiaques endocavitaires implantables a été proposé par un groupe d’experts.
The implantation of pacemakers and defibrillators carries the highest risk of infection in interventional electrophysiology. The use of implantable cardiac devices is continually increasing with almost 2 million devices implanted worldwide each year. The recipients’ profile may also be associated with an increased risk of infection. Several measures can be implemented to reduce the risk of device-related infection. Systematic antibiotic prophylaxis has proven to be beneficial provided that prescription modalities are respected, especially with respect to the selection of the appropriate molecule and timing of administration prior to the procedure. Despite all the precautions taken during surgery (asepsis, prophylactic antibiotic therapy….) the estimated rate of peri-procedural infection is around 2%. Device related infections are associated with a high rate of morbidity and mortality as well as substantial healthcare costs. Staphylococcus aureus (SA) and epidermidis (SE) are the pathogenic agents involved in most cases. Prevention is crucial given the difficulties in treating such infections because of the near-systematic need to remove the device and antibiotic resistance. Leadless pacemakers and subcutaneous defibrillators are potential alternatives to implantable endocardial devices, albeit with certain limitations. A group of experts has recently issued consensus paper on the prevention, diagnosis and treatment of infections associated with endocardial implantable cardiac devices.
The Future Directions of Research in Cardiac Anesthesiology
2020, Advances in Anesthesia
Citation Excerpt :
This is known as a randomized cluster crossover trial. Every time a cluster crosses over some of statistical power lost because of clustering is regained, provided there are no carry-over effects.55 In situations where they may be a carry-over effect (ie, where it may be difficult to resume the “control” intervention after the intervention of interest has been in place) an alternate design, the stepped wedge trial, may be used.
Selecting the Optimal Level of Clustering: An Approach to Trial Design Decision Making
2020, Journal of the American College of Surgeons
Risk Factors for Infections Involving Cardiac Implanted Electronic Devices
2019, Journal of the American College of Cardiology
Citation Excerpt :
The design of PADIT has previously been described (11), and the primary results have been published (1).
Cardiac implantable electronic device infection is a major complication that usually requires device removal. PADIT (Prevention of Arrhythmia Device Infection Trial) was a large cluster crossover trial of conventional versus incremental antibiotics.
This study sought to investigate independent predictors of device infection in PADIT and develop a novel infection risk score.
In brief, over 4 6-month periods, 28 centers used either conventional or incremental prophylactic antibiotic treatment in all patients. The primary outcome was hospitalization for device infection within 1 year (blinded endpoint adjudication). Multivariable logistic prediction modeling was used to identify the independent predictors and develop a risk score for device infection. The prediction models were internally validated with bootstrap methods.
Device procedures were performed in 19,603 patients, and hospitalization for infection occurred in 177 (0.90%) within 1 year of follow-up. The final prediction model identified 5 independent predictors of device infection (prior procedures [P], age [A], depressed renal function [D], immunocompromised [I], and procedure type [T]) with an optimism-corrected C-statistic of 0.704 (95% confidence interval: 0.660 to 0.744). A PADIT risk score ranging from 0 to 15 points classified patients into low (0 to 4), intermediate (5 to 6) and high (≥7) risk groups with rates of hospitalization for infection of 0.51%, 1.42%, and 3.41%, respectively.
This study identified 5 independent predictors of device infection and developed a novel infection risk score in the largest cardiac implantable electronic device trial to date, warranting validation in an independent cohort. The 5 independent predictors in the PADIT score are readily adopted into clinical practice. (Prevention of Arrhythmia Device Infection Trial [PADIT Pilot]; NCT01002911)

View all citing articles on Scopus

: See page 657 for disclosure information.

View full text

Special ArticleRandomized Cluster Crossover Trials for Reliable, Efficient, Comparative Effectiveness Testing: Design of the Prevention of Arrhythmia Device Infection Trial (PADIT)

Abstract

Résumé

Section snippets

The Need for Efficient Testing of Clinical Effectiveness

Problem of Arrhythmia Device Infection

Prevention of Device Infection

PADIT Pilot Studies and Development of the Cluster Crossover Design

Randomized Cluster Crossover Trial

PADIT Cluster Crossover Trial Design

Minimal Risk of the Antibiotic Strategies

Statistical Considerations

Ethics Committee Approval

Study Organization

Discussion

Funding Sources

Disclosures

Am J Cardiol

J Am Coll Cardiol

J Am Coll Cardiol

Can J Cardiol

Mayo Clin Proc

Ann Thorac Surg

Can J Cardiol

Cluster randomized trials in comparative effectiveness research: randomizing hospitals to test methods for prevention of healthcare-associated infections

Med Care

Complications associated with implantable cardioverter-defibrillator replacement in response to device advisories

JAMA

Efficacy of antibiotic prophylaxis before the implantation of pacemakers and cardioverter-defibrillators: results of a large, prospective, randomized, double-blinded, placebo-controlled trial

Circ Arrhythm Electrophysiol

Risk factors related to infections of implanted pacemakers and cardioverter-defibrillators: results of a large prospective study

Circulation

Risk factor analysis of permanent pacemaker infection

Clin Infect Dis

Management of infections involving implanted cardiac electrophysiologic devices

Curr Treat Options Cardiovasc Med

Complication risk with pulse generator change: implications when reacting to a device advisory or recall

Pacing Clin Electrophysiol

Antibiotic prophylaxis for permanent pacemaker implantation: a meta-analysis

Circulation

Special Article
Randomized Cluster Crossover Trials for Reliable, Efficient, Comparative Effectiveness Testing: Design of the Prevention of Arrhythmia Device Infection Trial (PADIT)