Special Feature

Arrhythmia and Sudden Death in Hemodialysis Patients:Protocol and Baseline Characteristics of the Monitoringin Dialysis Study

David M. Charytan,* Robert Foley,† Peter A. McCullough,‡§ John D. Rogers,| Peter Zimetbaum,¶

Charles A. Herzog,† and James A. Tumlin,** on behalf of the MiD Investigators and Committees

AbstractBackgroundDialysis patients have high rates of cardiovascular morbidity andmortality, but data on arrhythmiaburden, arrhythmia type, arrhythmia triggers, and the identity of terminal arrhythmias have historically beenlimited by an inability to monitor heart rhythm for prolonged periods.

Objectives To investigate arrhythmia and its association with sudden death in dialysis-dependent ESRD,describe the potential for implantable devices to advance study of dialysis physiology, review the ethicalimplications of using implantable devices in clinical studies, and report on the protocol and baseline results of theMonitoring in Dialysis Study (MiD).

Design, setting, participants, & measurements In this multicenter, interventional-observational, prospectivecohort study, we placed implantable loop recorders in patients undergoing long-term hemodialysis. Theproportion of patients experiencing clinically significant arrhythmias was the primary endpoint. For 6 months,we captured detailed data on the primary endpoint, symptomatic arrhythmias, other electrocardiographicvariables, dialysis prescription, electrolytes, dialysis-related variables, and vital signs. We collected additionalelectrocardiographic data for up to 1 year.

ResultsOverall, 66 patients underwent implantation in sites in the United States and India. Diabetes was presentin 63.6% of patients, 12.1% were age$70 years, 69.7% were men, and 53.0% were black. Primary and secondaryendpoint data are expected in 2016.

Conclusions Cardiac arrhythmia is an important contributor to cardiovascular morbidity and mortality indialysis patients, but available technology has previously limited the ability to estimate its true burden andtriggers and to define terminal rhythms in sudden death. Use of implantable technology in observational studiesraises complex issues but may greatly expand understanding of dialysis physiology. The use of implantable looprecorders inMiD is among the first examples of such a trial, and the results are expected to provide novel insightsinto the nature of arrhythmia in hemodialysis patients.

Clin J Am Soc Nephrol ▪: ccc–ccc, 2016. doi: 10.2215/CJN.09350915

IntroductionDialysis patients experience high cardiovascular andall-cause mortality. The death rate for all United Statesdialysis patients in 2011 was 198 per 1000 patient-years, with cardiac disease accounting for roughly40% (1). In the US Renal Data System database, twothirds of cardiac deaths are attributed to arrhythmia,making up 26% of mortality. Similarly, in the Hemo-dialysis Study and Die Deutsche Diabetes DialyseStudie trials, 22% and 26% of deaths, respectively,were sudden (2,3). The most convincing validationof registry data is provided by the EValuation OfCinacalcet HCl Therapy to Lower CardioVascularEvents study, the largest randomized dialysis trial,which enrolled 3883 hyperparathyroid hemodialysis(HD) patients. Overall, 25% of deaths were adjudicated

as sudden deaths (4), resolving speculation on therelative importance of (presumed) sudden cardiacdeath (SCD) in HD patients (Figures 1 and 2) (1,4).The underlying mechanisms of SCD, particularly

the specific type of terminal arrhythmia (which hasprofound implications for prevention), remain con-troversial. Nearly 5 years ago, the Kidney DiseaseImproving Global Outcomes Clinical Update Confer-ence on cardiovascular disease in CKD prescientlyconcluded, “Implantable loop recorders (ILRs) usedto identify terminal arrhythmias could prove useful,but a coordinated effort would be necessary givenlow enrollment rates anticipated in such studies” (5).Furthermore, the potential value of ILRs is not restrictedto lethal arrhythmias—the detection of clinically unsus-pected bradyarrhythmia and atrial fibrillation (AF)

*Department ofMedicine, Brigham &Women’s Hospital,Boston,Massachusetts;†Department ofMedicine, HennepinCounty MedicalCenter, University ofMinnesota,Minneapolis,Minnesota;‡Department ofMedicine, BaylorUniversity MedicalCenter, Baylor Heartand Vascular Institute,Baylor Jack and JaneHamilton Heart andVascular Hospital,Dallas, Texas;§Department ofMedicine, Division ofCardiology, The HeartHospital, Plano,Texas; |Department ofCardiology, ScrippsClinic, Torrey Pines,California;¶Department ofMedicine, Beth IsraelDeaconess MedicalCenter and HarvardMedical School,Boston,Massachusetts; and**Department ofMedicine, Universityof Tennessee,Chattanooga Collegeof Medicine,Chattanooga,Tennessee

Correspondence:Dr. David M.Charytan, RenalDivision, Brigham &Women’s Hospital,1620 Tremont Street,3rd Floor, Boston, MA02115. Email:Dcharytan@partners.org

www.cjasn.org Vol ▪ ▪▪▪, 2016 Copyright © 2016 by the American Society of Nephrology 1

. Published on January 13, 2016 as doi: 10.2215/CJN.09350915CJASN ePress

markedly widens their potential utility. In this manuscript,we review data on arrhythmia in HD patients and presentthe protocol and baseline data from the Monitoring in Di-alysis Trial (MiD), which used ILR to detect arrhythmia inthe setting of HD.

SCDHemodialysis initiation is a time of markedly increased

risk. Healthy People (HP) is a federally mandated programestablished to improve the health of Americans; one majorgoal of HP2020 is to “reduce new cases of CKD and its com-plications, disability, death, and economic costs” (1). Com-mensurate with the singularly high mortality rate in incidentdialysis patients, the HP2020 CKD-14.2 goal is to reduce thedeath rate within the first 3 months of dialysis initiation to 319.9deaths per 1,000 patient-years (from the 2011 rate of 335.4 per1000 patient-years). Figure 3 shows the early hazard of SCD(and other cause-specific mortality) at dialysis initiation (6).One major knowledge gap in understanding of SCD in

dialysis patients is ambiguity regarding the “terminal event,”particularly in distinguishing between sudden death due toarrhythmia and sudden death not preventable with cardiacdevices. Kidney Disease Improving Global Outcomes (5) iden-tified the paucity of autopsy data in dialysis-related SCD as amajor knowledge gap and highlighted the problematic natureof conventional definitions: “sudden, unexpected death

within an hour of symptom onset, or unwitnessed, unex-pected death without obvious non-cardiac cause in patientsknown to be well within the past 24 hours.” What exactly is“unexpected death” in a population with a high burden ofcomorbid illness who spend a disproportionate amount oftime in health care facilities? Furthermore, following with-drawal from dialysis, patients ultimately die of terminal ar-rhythmias, but this is death due to withdrawal. Withoutpatient-centered context, one could easily (and wrongly) inferSCD as the “primary” event from an ILR. Similarly, suchdiseases as subarachnoid hemorrhage or aortic dissectionmay mimic SCD in the absence of rhythm tracings.Many publications have highlighted strategies for risk

stratification to predict SCD in dialysis patients. For example,biomarkers indicating inflammation and malnutrition (e.g.,albumin) are associated with SCD risk (7,8). Additionally(but perhaps not surprisingly), there may be heritability toSCD propensity in dialysis patients. Chan et al. recently re-ported that genetically related family members on dialysishad a 1.7-fold increase in the odds of cardiac arrest comparedwith matched, unrelated controls (9). In the general popula-tion, rare mutations in ionic channels cause several distinctlong-QT syndromes, which are associated with a markedlyhigher risk of SCD (10). However, polymorphisms in long-QTsyndrome genes are relatively common (approximately 1% ofgeneral population) (11). These genetic polymorphisms maybe highly relevant and more dangerous in the context of con-ventional HD given the rapid declines in serum potassium,magnesium, and calcium, coupled with frequent exposure todrugs that cause QT prolongation (such as fluoroquinolones).Conventional 12-lead electrocardiography is a reason-

able method for assessment of the QT interval and the QTinterval corrected for heart rate (a value .460 msec for thelatter is considered prolonged). Although measurement ofQT dispersion as a predictor of risk has been studied indialysis patients, its role in clinical practice is still uncertain.Although ventricular arrhythmias (primarily prematureventricular contractions) are common in dialysis patients(based on 48-hour Holter monitoring), a prospective studyof 127 Italian HD patients followed for 4 years showed theywere not predictive of overall mortality (12,13).Few data are available on the types of lethal arrhythmias

in dialysis patients. One paper detailed 84 sudden cardiac ar-rest events in dialysis patients with wearable cardioverter-defibrillators (14). In this selected cohort, 78% of initialrhythms were ventricular tachycardia (VT; 64.3%) or

Figure 1. | Causes of death in prevalent dialysis patients, 2009–2011.AMI, acute myocardial infarction; CHF, congestive heart failure; CVA,cerebrovascular accident. Data obtained from reference 1. The datareported here have been supplied by the United States Renal DataSystem (USRDS). The interpretation and reporting of these data are theresponsibility of the author(s) and in noway should be seen as an officialpolicy or interpretation of the U.S. government.

Figure 2. | Adjudicated cause of death in the EValuation Of Cinacalcet HCl Therapy to Lower CardioVascular Events study population.CV, cardiovascular. Modified with permission from reference 4.

2 Clinical Journal of the American Society of Nephrology

ventricular fibrillation (VF; 14.3%), and 21.4% were asystole(i.e., not shockable). Unpublished data (kindly supplied byLinda Becker) covering emergency medical services dataon 47 cardiac arrests in nine outpatient dialysis centers inSeattle/King County between 1990 and 1996 showed thatin 29 (62%) arrests the rhythm was VT or VF. A recentAustralian study using ILRs, however, implicates brady-cardia and asystole as the major contributors to SCD inHD patients; this finding suggests pump failure or non-cardiac causes for the final sequence of clinical eventsleading to death (15,16). An important caveat was the ex-ceptionally long dialysis vintage of the study patients(mean6SD, 664 years); the type of arrhythmia (and termi-nal event) might be different in patients of newer versusolder dialysis vintage.

Role of Dialysis BathsLow-potassium dialysate of 0 or 1mEq/L (17) or,2mEq/L

(18) and lower calcium dialysate (,2.5 mEq/L) (19) havebeen implicated as risk factors for SCD in large databasestudies. Similarly, low serum magnesium levels were asso-ciated with higher all-cause mortality in Japanese dialysispatients (20), but few data exist on SCD risk and magnesiumdialysate levels.

Dialytic CycleAlthough great enthusiasm exists among nephrology

professionals for more frequent maintenance HD, mostpatients dialyze three times weekly, necessitating two gapsof 1 day and one gap of 2 days. This has long been viewed asphysiologically challenging in patients with limited capacity

Figure 3. | Cause-specific mortality in 2011 incident dialysis patients following a first dialysis session in a free-standing facility according tothe death notification form. Modified from reference 6.

Clin J Am Soc Nephrol ▪: ccc–ccc, ▪▪▪, 2016 Design of the Monitoring in Dialysis Study, Charytan et al. 3

to maintain homeostasis in the presence of metabolic andvolume-related excursions from normality, where back-ground cardiovascular disease is the norm.In a nationally representative United States sample

between 2004 and 2007, one study compared rates of deathand cardiovascular admissions on the day after the 2-dayinterdialytic interval with rates on other days (21). As shownin Figure 4, although overall mortality, cardiovascular mor-tality, and cardiovascular admission rates were higher onthe day after the long interval, relative event disparitieswere especially marked for congestive heart failure (by afactor of 1.8) and dysrhythmia (by a factor of 1.9).Similar associations have been seen in several large studies

(22–24). For example, in a recent study of newly incident HDpatients in the United Kingdom between 2002 and 2006,hospital admission rates after the 2-day gap were 1.7-foldhigher, whereas all-cause mortality rates were 1.22 timeshigher. As with almost all of the literature to date, this studywas not specifically configured to compare rates of SCD.Nevertheless, out-of-hospital death rates were 1.59 timeshigher after the long interval than after the shorter intervaland only 1.06 times higher for in-hospital death (23).

Pathophysiology of Arrhythmia in CKDMultiple cardiovascular risk factors, including athero-

sclerotic heart disease, left ventricular hypertrophy (LVH),

and accelerated cardiac fibrosis, appear to contribute toarrhythmia pathogenesis in ESRD (25). Although arrhythmiaand ultimately sudden cardiac death can occur in patients(in the general population) with apparently structurally nor-mal hearts, most patients (particularly those with CKD)have underlying structural heart disease, and some type ofacute event interacts with the underlying substrate to pro-duce the fatal arrhythmia (26,27).Although many triggers have been identified, acute myo-

cardial ischemia is felt to be the most common initiatingevent in the general population (28). In patients withadvanced CKD (particularly those undergoing dialysis),myocardial ischemia is likely to be a contributor, but itis also plausible that myocardial ischemia (of the typemediated by epicardial coronary artery disease) may play aless predominant role and that other factors, such as inflam-mation and autonomic imbalance or increased sympatheticactivity (including sleep apnea), may be important contrib-utors to sudden cardiac death (29,30). Hypertrophic myocar-dium is predisposed to both atrial and ventriculararrhythmia through the induction of prolonged actionpotentials and increased repolarization defects in areasof ischemia, with underling fibrosis serving as a favorablesubstrate for propagation of arrhythmia. While hyperten-sion, diabetes, and other typical risk factors undoubtedlycontribute, dialysis and uremia may directly contribute toLVH and fibrosis, which appears to accelerate after

Figure 4. | Association of dialytic intervals cycle with mortality and cardiovascular hospitalizations. Annualized mortality (A) and cardio-vascular admission rates (B) per 100 patient-years, on different days of the dialysis week. Error bars represent 95% confidence intervals. CHF,congestive heart failure; CVD, cardiovascular disease; HD, hemodialysis (session); MI, myocardial infarction. Adapted from reference 21.

4 Clinical Journal of the American Society of Nephrology

dialysis initiation. In a prospective study of 596 patientswith ESRD who had no history of cardiovascular disease,for example, serial echocardiography demonstrated thatleft ventricular mass index increased significantly duringthe first 2 years of dialysis (31).Processes unique to CKD and ESRD may largely account

for increasing left ventricular mass index. Mall et al. analyzedmyocardial histology in patients with advanced renal dis-ease and noted that .90% of patients exhibited diffuse anduniform cardiac fibrosis with significantly greater levels ofmyocardial collagen than observed in hypertensive and val-vular cardiomyopathies and diffuse rather than the “patchy”fibrosis characterizing other conditions (32). Amann et al.subsequently demonstrated that myocardial matrix ex-pansion was associated with a reduction of capillarydensity (33). Low capillary density may partly explainsudden reductions in myocardial perfusion inducedduring dialysis. For example, McIntyre et al. studied patientswith ESRD who did not have significant coronary diseaseusing intradialytic position emission tomography and notedmarked, regional reductions in myocardial perfusion duringdialysis. Changes in perfusion were accompanied by re-gional wall-motion abnormalities, suggesting that functionalischemia during dialysis is not caused by flow-limitingatherosclerosis (34).The full pathogenesis of myocardial fibrosis and capil-

lary rarefaction in ESRD is beyond the scope of this review.Further study is needed to fully elucidate the pathwaysinvolved. However, several circulating factors, includingasymmetric dimethyl arginine (35,36), parathyroid hor-mone (37), aldosterone (38), fibroblast growth factor-23(39), angiotensin-2 (40), endogenous cardiac glycosides(41), vitamin D (42), and circulating angiogenesis inhibi-tors, have been implicated (35).These observations suggest that in addition to the effect

of LVH and coronary disease, transient myocardial ischemiainduced by a mismatch between capillary supply and thehypertrophied myocardium may be critical to the gener-ation of arrhythmias (43). Conversely, myocardial fibro-sis may slow or disrupt normal conduction, leading tobradycardia, asystole, and reentrant arrhythmias (43).Peridialytic changes in BP, volume status, sympathetictone, and electrolytes are likely to further contribute toarrhythmogenesis, but their exact contributions remainlargely unstudied.

AFIncidence and Prevalence of AFAlthough asystole or ventricular arrhythmias are the

most likely types of arrhythmia to result in sudden death,atrial arrhythmias, particularly AF, may result in significantmorbidity in patients with ESRD. AF is increasingly commonin HD patients. In a study of 258,605 older participants, theAF incidence in incident dialysis patients was 14.8/100person-years, and adjusted probabilities of developingAF during the first year of dialysis increased from 11.3%in 1995 to 14.3% in 2007 (44). Similarly, the prevalence ofAF (diagnosed from administrative claims) was 10.7% in2006—a three-fold increase from 1992 (45). Finally, theoverall AF prevalence in a meta-analysis of 25 dialysisstudies was 11.6% (46).

Dialysis as a TriggerMost evidence that HD triggers AF is indirect. Echocar-

diographic intra-atrial and interatrial activation times, forexample, correlate with ultrafiltration volume and shortensignificantly toward the end of HD (47). Similarly, thereare peridialytic changes in P-wave dispersion (standardelectrocardiography), P-wave duration, and the rootmean square voltage of the final 20 ms of the filtered Pwave (signal-averaged electrocardiography)—parametersthat measure atrial conduction delay and are associatedwith AF (48,49). In several studies, intradialytic changeswere prominent and were attenuated at the conclusion ofdialysis (50,51). In others, changes were not detected orwere seen only at the conclusion of dialysis and then quicklyattenuated (52). However, in one study, P-wave duration de-creased and root mean square voltage of the final 20 ms of thefiltered P wave decreased from the beginning to the end ofdialysis, changes suggesting a reduced risk of AF after HD(53). Changes in electrolytes were associated with P-waveparameters in several studies (51,52), but associations withultrafiltration appear more robust (50,51). On balance, thisliterature suggests that rapid ultrafiltration or other factorsduring HD acutely affect conduction in a way that couldtrigger AF but that effects rapidly attenuate after HD.More recently, implanted defibrillators were used to

monitor patients enrolled in the Implantable CardioverterDefibrillator-2 Trial. AF was detected in 14 of 40 dialysispatients—11 receiving HD, 3 receiving peritoneal dialysis,and 9 without known AF (54). AF frequency was three-foldhigher on dialysis days (P50.001) but did not differ withduration of the interdialytic interval. Intradialytic AF was13-fold more frequent than in the 7 hours before dialysis(P50.03) and two-fold higher than in the 7 hours after di-alysis (P50.001), although results were largely driven bytwo patients. Lower dialysate potassium concentration andhigher ultrafiltration volumes were associated with AF oc-currence. More recently, a trial using an ILR in 50 HD pa-tients found that the long interdialytic interval was the mostfrequent period of arrhythmia. AF was detected in 42% ofpatients, and new-onset AF was asymptomatic in 86% (16).

Consequences of AFAssociations with stroke are similar in the ESRD and

general populations (44–46,55,56). In a recent study, forexample, stroke incidence was 5.2/100 patient-years inpatients with AF compared with 1.9/100 patient-years indialysis patients without AF (46). AF is also strongly asso-ciated with mortality due to fatal stroke and other causes,the rate of which is roughly doubled (26.9 versus 13.4/100patient-years) when AF is present (44–46). In addition, AFis an independent risk factor for mesenteric ischemia, afrequently fatal event (57), and amputation (58); althoughthis has not been studied in the setting of ESRD, AF maylead to tachyarrhythmia-induced cardiomyopathy whensustained (59). Frequent AF could thus be an importantcause of an extended array of morbidity and mortality,including fatal stroke, heart failure, and deterioration inLV ejection fraction in HD patients.

Relevance of Subclinical AFAlthough associations of subclinical AF with clinical

outcomes have not been studied in ESRD, multiple studies

Clin J Am Soc Nephrol ▪: ccc–ccc, ▪▪▪, 2016 Design of the Monitoring in Dialysis Study, Charytan et al. 5

suggest that clinically silent AF is an important stroke riskfactor in the general population. In one randomizedpacemaker trial, the pacemaker monitoring function wasused to record rapid atrial events (rate $220 beats/min[BPM] lasting $5 minutes) presumed to represent AF(60). Overall, 51.3% of patients had events at a medianonset of 100 days. Atrial events were independently associ-ated with stroke (hazard ratio, 2.8; P50.001). Other studieshave demonstrated similar associations between subclinicalAF and subsequent overt AF or stroke (61–63). More recently,the Cryptogenic Stroke and Underlying AF (CRYSTAL AF)(64) study randomly assigned 441 stroke patients without AFwho were undergoing 24-hour monitoring to an ILR or con-ventional follow-up. AF lasting .30 seconds was detected in30% of intervention versus 3% of control patients at 3 years.These data suggest that subclinical AF is an important

contributor to stroke while illustrating the limits of clinicalobservation and standard diagnostics for advancing scientificdiscovery. In CRYSTAL AF, newly developed implantabletechnologies enabled long-term, continuous monitoring,thereby facilitating discovery of a high frequency of “silent”AF and a paradigm shift in the understanding of crypto-genic stroke. Improving understanding of SCD and its asso-ciation with dialysis is an obvious extension. More broadly,CRYSTAL AF highlights a new paradigm in which the

expanding capabilities and decreasing size of wearableor implantable technologies will facilitate capture of pre-viously unmeasurable parameters and collection of moregranular physiologic data, thereby improving under-standing of dialysis physiology.

LINQ and Reveal XT SystemSubcutaneous ILRs have been clinically available since

1998 (Medtronic Reveal; Medtronic, Minneapolis, MN), andILR use for detecting heart rhythm abnormalities in patientswith syncope, palpitations, AF, and other conditions inwhich arrhythmia is suspected has been well described inthe literature. These small loop recorders are placed in thesubcutaneous space and are leadless. ILRs continuouslyrecord the heart rhythm, and when programmed alertcriteria are met for diagnosis of arrhythmia, the event iscaptured and stored until retrieved in person or via remotehome monitoring. Most patients in the MiD study had aMedtronic Reveal XT ILR implanted, but recently enrolledpatients received an updated device, the Medtronic RevealLINQ (Figure 5). These monitors are implanted subcuta-neously with a simple outpatient procedure using localanesthesia. In patients without ESRD, device-related com-plications associated with ILR implantation include

Figure 5. | Reveal andMyCareLink systems. (A) Reveal (bottom) and Reveal (top) implantablemonitors. (B) MyCareLinkmonitor. The pictureddevice is used with the LINQ system and transmits stored information wirelessly from the implanted monitor to central servers when it islocated within approximately #30 feet of the implanted monitor. MyCareLink systems used with the Reveal XT device required that thetransmitter be connected to a landline and held up to the chest wall. (C) Information flow from implantable cardiac monitor to web-hostedphysician interface.

6 Clinical Journal of the American Society of Nephrology

infection (1.2%), device migration, and pain at the implantsite (,1%) (65). Battery life is approximately 3 years, andboth devices are conditionally approved for magnetic res-onance imaging (1.5 and 3.0 T) without reprogramming,although waiting for 6 weeks after insertion with the XTdevice is recommended (65,66). Medicare reimbursementfor an ILR is generally in the range of $6000–$7000, whichincludes the device and insertion-related costs.Both Reveal XT and Reveal LINQ can detect and recordAF

aswell as homemonitoring capabilities, but the LINQ deviceis 87% smaller (about a one third the size of a triple-Abattery), offers more memory (20% more), and providesautomatic nightly downloads of detected arrhythmias.While Reveal XT required holding a landline-connectedtransmitter over the chest wall, automatic nightly down-loads from the LINQmonitor to a bedside cellular phone–basedmonitor (MyCareLink monitor [Medtronic, Minneapolis,MN]), allow daily assessment of patient’s heart rhythms(Figure 5). This capability allows physicians to follow pa-tients daily instead of receiving the data only when thepatient has symptoms or at the time of scheduled interro-gation, as was the practice with the XT device.ILRs offer the ability to monitor patients for an extended

period (up to 3 years) (65). For patients with infrequentsymptoms or who require long-term rhythm evaluation,these monitors provide more information than shorter-term Holter monitoring, wearable patch technology, exter-nal loop recorders, or intermittent monitoring devices (67).Patient compliance may also be issue with external monitors,which limits their effectiveness (68).

MiD ProtocolOverviewThe primary objective of the MiD study (NCT01779856)

was to estimate the proportion of HD patients experiencingclinically significant arrhythmias during a 6-month primaryobservation period using the Reveal ILR system. Secondarygoals were to broadly characterize the occurrence of arrhyth-mia in HD-dependent ESRD and quantify associations withelectrolytes, HD procedure, and volume parameters.

Study PopulationEligibility criteria were designed to maximize general-

izability and patient safety (Table 1). All research compliedwith the Declaration of Helsinki and was approved byapplicable institutional review boards.

Study ProceduresEchocardiography (if not performed within the preceding

6 months), 12-lead electrocardiography, medical history, andmedications were obtained at baseline, and ILRs wereimplanted within 14 days. For the first 6 months afterimplantation, patients transmitted Reveal data immediatelybefore all dialysis sessions and after each session associatedwith a study blood draw. Postmortem transmission was en-couraged but was not specifically required by the protocol.Symptoms, dialysis prescription, ultrafiltration volumes, andperi- and postdialysis vital signs were collected at everydialysis session during the first 6 months. Blood was drawnboth before and after dialysis: twice weekly for 4 weeks andonce weekly thereafter through 6 months (Table 2 and 3).After 6 months, transmissions occurred at least weekly butdata collection was otherwise limited to adverse events.Study coordinators reviewed transmissions after the dialysis

session to identify prespecified, potentially dangerousarrhythmias, which mandated investigator review:VT $180 BPM for .15 seconds, asystole .5 seconds,waking (6 a.m.–10 p.m.) heart rate#40 BPM for$6 seconds,AF for.24 hours or for.12 hours over consecutive days, orsymptomatic arrhythmia.During months 0–6, the study sponsor reviewed

patient-marked events and transmissions with potentialarrhythmias. A core laboratory adjudicated those possiblyconsistent with the primary endpoint. Long-term follow-upcontinued for a maximum of 12 months. The study con-cluded when the last study participant had completed 6months of follow-up. ILR removal at study terminationwas optional.

EndpointsThe primary study objective was to estimate the pro-

portion of HD patients experiencing clinically significant

Table 1. Inclusion and exclusion criteria

Inclusion Criteria Exclusion Criteria

Age$ 21 yr Not suitable for implantation (e.g., cachexia, severedermatologic conditions)

In-center HD$3 times/wk oreGFR,15 ml/min per 1.73 m2 with expected HDinitiation within 2 mo

Expected survival ,6 moLeft-sided HD catheter in position expected to interferewith implantation

Ability to comply with protocol Thoracic surgery within 6 moInfection with 14 dBacteremia within 60 dHemoglobin ,10 g/dl on consecutive measurementswithin prior 2 mo

Transplantation expected within 6 moModality transfer expected within 6 moExisting pacemaker or ICD

HD, hemodialysis; ICD, implantable cardioverter-defibrillator.

Clin J Am Soc Nephrol ▪: ccc–ccc, ▪▪▪, 2016 Design of the Monitoring in Dialysis Study, Charytan et al. 7

cardiac arrhythmias (CSAs) over 6 months. CSAs arearrhythmias considered most likely to be associated withsyncope and cardiac arrest or to cause symptoms ofhypoperfusion. They were based on standard definitions(69–71), recommendations of an advisory panel, and thedetection capabilities of the Reveal device and includedthe following:

c VT$115 BPM lasting $30 seconds (the rate limit wassubsequently changed to $130 BPM with a protocolamendment).

c Bradycardia with heart rate #40 BPM for $6 seconds.c Asystole for $3 seconds.c Patient-marked (symptomatic) events where electrocardio-

graphic review showed an arrhythmia considered clinicallyrelevant in the judgment of the site cardiologist.

Secondary objectives were designed to assess devicesafety, characterize cardiac rhythm and associations withdialysis or clinical events, and assess the ability of theReveal device to detect short-term changes in electrocar-diographic morphology and their association with treat-ment parameters.The following secondary objectives were specified: (1)

quantifying device and procedure-related adverse events;

(2) recording health-related events, specifically death, car-diovascular events, and health care utilization; (3) analyzingthe association of arrhythmic events with health-relatedevents and HD treatment parameters (this objective includedCSA, other arrhythmias [e.g., AF], and parameters such asheart rate variability and heart rate trend); (4) quantifyingatrial arrhythmia burden and analyzing associations withHD treatment parameters; and (5) assessing association ofcaptured electrocardiographic morphology and pre- andpostdialysis serum electrolyte levels.

Statistical AnalysesThe planned sample size was 125 implanted patients.

With 10% attrition this would have enabled estimation ofthe proportion of patients experiencing CSAs with a 95%confidence interval (95% CI) half-width of less than 0.1 forany proportion. Preliminary data from the first 50 enrolledpatients revealed CSA rates of $70%, which allowed esti-mation of 95% CI half-widths #0.15 without additionalenrollment. Thus, recruitment of the full sample size wasnot necessary to reliably estimate CSA event rates or todetermine whether there was a clinically relevant inci-dence of CSA in the HD population. Given the financialand logistic challenges of recruitment and limited potentialfor qualitative effect from additional enrollment, the study

Table 2. Schedule for blood draws

WeeklySession

First SessionPredialysis

First SessionPostdialysis

Second or Third SessionPredialysis

Second or Third SessionPostdialysis

Wk 1 A1B1LTS D1Ea C DWk 2 C D C DWk 3 C D C DWk 4 C D C DWk 5–26 Bb or C D

For laboratory tests performed in panels A–E, see Table 3. LTS, sample long-term storage.aOptional blood draw.bHemoglobin, hematocrit, and iron collected monthly according to standard of care.

Table 3. Blood collection panels

A B C D E (Optionala)

Brain natriuretic peptide Albumin Albumin Albumin AlbuminCK/CK-MB Bicarbonate Bicarbonate Bicarbonate BicarbonateHemoglobin A1c BUN BUN BUN CalciumHigh-sensitivity C-reactive protein Calcium Calcium Calcium CreatinineParathyroid hormone Creatinine Creatinine Creatinine MagnesiumTroponin-T Hematocrit Magnesium Magnesium Phosphorous

Hemoglobin Phosphorous Phosphorous PotassiumIron Potassium Potassium SodiumMagnesium Sodium SodiumPhosphorousPotassiumSodium

CK, creatine kinase; CK-MB, creatinine kinase muscle brain; BUN, blood urea nitrogen.aOptional blood collections 15 and 30 minutes after dialysis on first session after Reveal implantation.

8 Clinical Journal of the American Society of Nephrology

was capped at 66 implanted patients. This final samplesize of 66 enables the estimation of the proportion ofpatients experiencing CSA with a 95% CI half-width of,0.13 for any proportion.The primary CSA proportion will be calculated from the

cohort completing 6-month follow-up with 95% CI boundsestimated by the Clopper–Pearson “exact” method usingthe MiD-p modification (72). The analysis will be repeatedin two populations: (1) patients with complete follow-up,plus those with incomplete data who experienced CSAs,and (2) all implanted patients, assuming that those withincomplete follow-up and no CSAs while under observa-tion had no events during unobserved follow-up. This willbe supplemented with Kaplan–Meier survivor plots ofCSA-free survival. A negative binomial model, appropriatefor recurrent events and allowing for variable follow-up (73),will be used to estimate mean and 95% CI of CSA perpatient-year.

Special ConsiderationsUse of an invasive monitoring device in a clinical study is

an unusual feature of MiD and raises unique issues byexposing patients to risk in an otherwise observationalstudy. The ethical issues are not unique to this design.Radiographic endpoints are frequently used in studies, forexample, and also require radiation exposure without clearbenefit; phase 1 drug studies or “first-in-human” devicestudies also expose people to risk without clear benefit.Given the need to understand SCD and arrhythmia cause

in the dialysis population, the minimally invasive nature ofILR devices, and the low rate of infections when used inother populations, the use of Reveal XT and LINQ in MiDwas felt to be ethically acceptable. In general, we advocateassessing “interventional-observational” research designs,with an ethical framework that balances the importance ofthe knowledge gained against potential risks. High-risk,non–minimally invasive devices would rarely be acceptablewithout a potential for individual benefit. Conversely, asdevices become smaller and less invasive, they may enableobservational studies of nonserious conditions. Use of non-invasive tools should generally be considered as an alterna-tive, and investigators are obligated to carefully reviewindividual risk before enrolling patients in an observationalstudy requiring use of an implantable monitor.A particularly thorny issue is that monitoring data can

alter observed outcomes and bias estimated event rates.Detected arrhythmias in MiD, for example, could mandatechanges in clinical care (e.g., pacemaker implantation,change in dialysis prescription) that modify the risk ofsubsequent arrhythmia, thereby reducing CSAs and result-ing in underestimation of the true population CSA rate.Full disclosure to participants and clinicians maximizes

potential benefits to participants but also maximizes bias.Sequestering the monitoring data eliminates outcomecontamination, but nondisclosure of potentially harmful andtreatable conditions discovered during research is ethicallyunacceptable.With end-of-study batch analysis of raw monitoring data,

results and interpretation of the monitoring data do notbecome available until after the study has concluded. This issimilar to storing blood samples for batch analysis at the endof a study. Because the samples (in this case monitoring

information) are not analyzed or interpreted until after thestudy, this approach eliminates the ethical problems inherentto withholding analyzed data; however, it imposes sub-stantial data storage requirements and may inhibit the abilityto optimally understand clinical features of events or todetermine the best clinical response. Batch analysis at regularintervals offers a useful intermediate approach. Finally, itmay be possible to select endpoints, monitoring time frames,or data disclosures in order to limit biasing of importantendpoints while allowing clinical use of research data. InMiD, arrhythmias felt to be dangerous, such as asystole andsustained VT, were reviewed in an ongoing manner andshared with study patients and their clinicians. Conversely,disclosure of arrhythmias of uncertain importance, such asnonsustained VT, was not required.

Baseline Characteristics of the Study PopulationPatients were enrolled in India (23 of 66) and the United

States (43 of 66). Several characteristics of the studypopulation differed from those of the United States dialysispopulation (1). Mean age was lower; only 12.1% of patientswere$70 years of age. A higher percentage of patients wereblack (53.0%), Asian (34.8%), and male (69.7%). Diabeteswas present in 63.6% of patients, while nearly half of thepatients (48.5%) had a history of ischemic heart disease. Aminority of patients had a history of arrhythmia (31.8%),and 10.6% had a history of AF (Table 4). Use of cardiovas-cular medications was common, with 55%, 33%, and 48% ofpatients using b-blockers, angiotensin-converting enzymeinhibitors, or angiotensin-receptor blockers or statins, respec-tively. Laboratory parameters (Table 5) suggested adequatedelivery of dialysis with a mean single pool Kt/V of 1.560.4.Potassium (5.061.0 mEq/L), hemoglobin (10.6 61.2 g/dl),and phosphorous (5.562.0 mg/dl) were each mildly ele-vated but within the expected range for patients undergoinglong-term dialysis.Several characteristics differed significantly between

United States and Indian patients. In particular, Indianpatients weighed less (66.5 versus 97.6 kg), were morelikely to have diabetic kidney disease (56.5% versus34.9%), were more likely to dialyze via an arteriovenousfistula (91.3% versus 59.5%), had a shorter dialysis vin-tage (2.2 versus 3.5 years), and were less likely to havehypertension (60.9% versus 97.7%) than participants inthe United States. Laboratory characteristics were gener-ally similar, but bicarbonate (18.8 versus 23.8 mEq/L)and sodium (134.1 versus 138.4 mEq/L) concentrationswere lower in India.This regional variation and increased variability in labo-

ratory and clinical characteristics present challenges andopportunities. Analysis of country-specific arrhythmia ratesis important, but the wide range of characteristics, such asbody mass index, hypertension, and bicarbonate concen-trations, may provide a greater ability to analyze associa-tions of these factors with arrhythmia than would have beenpossible had the study recruited only within the UnitedStates. Global enrollment should thus limit the effect ofcountry-specific dialysis practices on our findings and mayenhance relevance to the global HD population whilerequiring more cautious extrapolation to the United Statespopulation.

Clin J Am Soc Nephrol ▪: ccc–ccc, ▪▪▪, 2016 Design of the Monitoring in Dialysis Study, Charytan et al. 9

MiD as a ParadigmThe MiD study illustrates the potential of invasive

monitoring to improve understanding of the pathophysi-ology of ESRD and elucidate the true nature or cause of

clinical events in dialysis patients. From our experience, itseems clear that when investigators are engaged, risks arereasonably low, research questions are important, andclinical benefit is possible, dialysis patients can be engaged

Table 4. Baseline characteristics of enrolled patients

Characteristics All Patients (n566) Patients in UnitedStates (n543)

Patients in India(n523) P Value

Meanage at implant6SD,yr (n/N) 56.3612.2 (66/66) 55.8611.6 (43/43) 57.2613.5 (23/23) 0.66Male sex 69.7 (46/66) 62.8 (27/43) 82.6 (19/23) 0.16Race ,0.001Asian 34.8 (23/66) 0.0 (0/43) 100.0 (23/23)Black 53.0 (35/66) 81.4 (35/43) 0.0 (0/23)Other 1.5 (1/66) 2.3 (1/43) 0.0 (0/23)White 10.6 (7/66) 16.3 (7/43) 0.0 (0/23)

Hispanic ethnicity 0.0 (0/66) 0.0 (0/43) 0.0 (0/23)Cause of ESRDDiabetes 42.4 (28/66) 34.9 (15/43) 56.5 (13/23) 0.01GN 7.6 (5/66) 7.0 (3/43) 8.7 (2/23)Hypertension 37.9 (25/66) 51.2 (22/43) 13.0 (3/23)Other 12.1 (8/66) 7.0 (3/43) 21.7 (5/23)

Median ESRD vintage(IQR), yr (n/N)

2.4 (1.2–5.3) (65/66) 3.5 (1.2–5.7) (42/43) 2.2 (1.1–3.0) (23/23) ,0.001

Prior kidney transplant 13.6 (9/66) 11.6 (5/43) 17.4 (4/23) 0.71Previous peritoneal dialysis 10.6 (7/66) 14.0 (6/43) 4.3 (1/23) 0.41Current vascular accessAV fistula 70.8 (46/65) 59.5 (25/42) 91.3 (21/23) 0.01AV graft 24.6 (16/65) 35.7 (15/42) 4.3 (1/23)Catheter 4.6 (3/65) 4.8 (2/42) 4.3 (1/23)

Diabetes 0.61None 36.4 (24/66) 37.2 (16/43) 34.8 (8/23)Type 1 4.5 (3/66) 7.0 (3/43) 0.0 (0/23)Type 2 59.1 (39/66) 55.8 (24/43) 65.2 (15/23)

Mean diabetes duration6SD,yr (n/N)

17.9612.8 (37/66) 21.0613.6 (24/43) 12.169.0 (13/23) 0.04

Hyperlipidemia 60.6 (40/66) 74.4 (32/43) 34.8 (8/23) 0.003Hypertension 84.8 (56/66) 97.7 (42/43) 60.9 (14/23) ,0.001Ischemic heart disease 48.5 (32/66) 51.2 (22/43) 43.5 (10/23) 0.61Congestive heart failure 25.8 (17/66) 39.5 (17/43) 0.0 (0/23) ,0.001Coronary artery bypass surgery 13.6 (9/66) 11.6 (5/43) 17.4 (4/23) 0.71History of arrhythmia 31.8 (21/66) 48.8 (21/43) 0.0 (0/23) ,0.001Atrial fibrillation 10.6 (7/66) 16.3 (7/43) 0.0 (0/23) 0.09Smoking 0.02Current 7.6 (5/66) 11.6 (5/43) 0.0 (0/23)Never 69.7 (46/66) 58.1 (25/43) 91.3 (21/23)Past 22.7 (15/66) 30.2 (13/43) 8.7 (2/23)

Mean weight6SD, kg (n/N) 86.7628.8 (66/66) 97.6628.7 (43/43) 66.5614.9 (23/23) ,0.001Body mass index $40 kg/m2 9.1 (6/66) 14.0 (6/43) 0.0 (0/23) 0.08Mean systolic BP6SD, mmHg(n/N)

140.8623.4 (66/66) 139.6625.8 (43/43) 143.0618.2 (23/23) 0.57

Mean diastolic BP6SD, mmHg(n/N)

76.8612.9 (66/66) 75.5615.1 (43/43) 79.166.7 (23/23) 0.18

MedicationsNonaspirin anticoagulants oranti platelet agents

15 (10/66) 16 (7/43) 13 (3/23) 1.00

b-Blockers 55 (36/66) 58 (25/43) 48 (11/23) 0.45Calcium channel blockers 59 (39/66) 51 (22/43) 74 (17/23) 0.11ACEI or ARB 33 (22/66) 47 (20/43) 9 (2/23) 0.002Statin 48 (32/66) 47 (20/43) 52 (12/23) 0.80Aspirin 45 (30/66) 49 (21/43) 39 (9/23) 0.60

Unless otherwise noted, values are the number/number of patients (percentage). Categorical P value is based on a two-sided Fisherexact test. ContinuousP value is based on an unpaired t test. IQR, interquartile range;AV, arteriovenous; ACEI, angiotensin-convertingenzyme inhibitor; ARB, angiotensin-receptor blocker.

10 Clinical Journal of the American Society of Nephrology

and their collaboration secured in observational studies usinginvasive monitoring procedures.Future research on ILR technology in ESRD should

include studies to determine the association between sub-clinical AF and the risk of stroke, long-term studies withsufficient sample size to capture terminal cardiac rhythmsat the time of SCD, and interventional studies testing theeffect of individualization of the dialysis prescription onarrhythmia. Devices with miniaturized impedance technologywill facilitate studying associations of volume status witharrhythmia and more sophisticated investigations into linksbetween ultrafiltration rate, BP, volume status, cardiovasculardeath, and intradialytic hypotension. Outside of the cardio-vascular arena, implantable glucose monitors stand out as atechnology that could be readily used to definitively determinethe true relationship between serum glucose levels andhemoglobin A1c or clinical outcomes in ESRD (74). Simi-larly, one can envision using pressure or flow monitoringdevices to better understand access physiology and trajec-tories of access maturation and failure.

SummaryUnderstanding the causal sequence of events in the pathway

to death is critical to identifying opportunities for prolong-ing life in ESRD. Despite advances in renal care, the periodfrom the last clinical assessment until out-of-hospital deathremains ripe for investigation (75–78). In persons under-going in-center HD, regular health care contact limits thetime during which no vital information is recorded; how-ever, out-of-center SCD remains common, with little infor-mation about the hours and days leading up to the finalminutes of life. Information on these periods in home-carepatients is more limited with respect to cardiac rhythm andvital status (79).Conventional cardiac pacemakers have reduced the mor-

tality associated with bradycardia and allow bradycardicpatients to have normal actuarial survival compared withage- and sex-matched cohorts (80). In those with acceptedindications for implantable cardioverter-defibrillators

(ICDs), the rate of appropriate therapy for lethal rhythmsis approximately 20% over the life of the implant (80), andprimary and secondary prevention of SCD with ICDs inpatients with left ventricular dysfunction improves survivalof these groups. In a subset of patients with left ventriculardysfunction who have ventricular dyssynchrony (e.g., pro-longed left bundle-branch block morphology QRS complexon electrocardiography), the addition of a left ventricularlead and biventricular pacing improves survival from bothsudden and pump failure death (82).Nevertheless, the specific translation of these therapies to

patients with advanced CKD and ESRD is limited, as notedabove, by a lack of data that allow inferences on theirpotential risks and benefits. ICDs, for example, face higherdefibrillation thresholds that may inhibit successful cardio-version of VT/fibrillation in patients with ESRD (83). Thismay reflect greater degrees of overall left ventricular fibro-sis and hypertrophy compared with those without years ofCKD. There is also limited information about the need forbackup bradycardia pacing in patients with ESRD whohave ICDs implanted yet may have a clinical event wherebradycardia is the complicating rhythm. Finally, andmost critically, the proportion of death that is trulyavoidable with device therapy versus how much is trulyattributable to pump failure or noncardiac primary cau-ses that cannot be influenced by any form of devicetherapy remains unknown.The MiD study is uniquely designed to gain critical

insights into patients both in the clinic and hospital and,more important, in the community and households in whichthey live. Recording the most recent baseline rhythm, criticaltriggers (premature depolarizations, pauses), pathologicintermediate rhythms (monomorphic, polymorphic VT), andterminal patterns (heart block, VT, VF, systole, and otherforms of pulseless electrical activity) will provide crucial dataallowing investigators to begin piecing together the sequenceof events and proximate causes of the last minutes of life inESRD. Similarly, identification of clinical, demographic, orelectrophysiologic variables associated with the final elec-trical sequence may prove useful for predicting lethal, yet

Table 5. Baseline laboratory data of study participants

Laboratory Value All Patients (n) Patients in United States (n) Patients in India (n) P Value

BUN, (mg/dL) 59.7617.8 (59) 58.6615.8 (40) 62.2621.8 (19) 0.47Creatinine, mg/dl 10.063.4 (59) 10.963.5 (40) 8.362.4 (19) 0.01Sodium, mEq/L 137.064.5 (59) 138.463.9 (40) 134.164.4 (19) ,0.001Potassium, mEq/L 5.061.0 (58) 4.860.7 (40) 5.461.3 (18) 0.07Calcium, mg/dl 8.760.8 (59) 8.860.8 (40) 8.460.8 (19) 0.08Bicarbonate, mEq/L 22.263.7 (59) 23.862.9 (40) 18.862.9 (19) ,0.001Magnesium, mg/dl 2.460.5 (59) 2.260.3 (40) 2.860.6 (19) 0.001Phosphorous, mg/dl 5.562.0 (59) 5.762.1 (40) 5.061.9 (19) 0.19Hemoglobin, g/dla 10.661.2 (56) 10.861.0 (38) 10.261.5 (18) 0.12Serum albumin, g/dl 3.9460.33 (57) 4.0360.26 (40) 3.7560.39 (19) 0.01Parathyroid hormone, pg/ml 473.76371.9 473.76371.9 NA NAspKT/V 1.560.4 (59) 1.560.3 (40) 1.460.6 (19) 0.44

Values expressed are mean6SD. Categorical P value is based on a two-sided Fisher exact test. Continuous P value is based on anunpaired t test. spKT/V, single pool Kt/V; NA, not available.aPatients in the United States with hemoglobin,10 g/dl on two consecutive local laboratory drawswere excluded. Data presented arefrom baseline hemoglobin measure at the core laboratory at first postimplant session.

Clin J Am Soc Nephrol ▪: ccc–ccc, ▪▪▪, 2016 Design of the Monitoring in Dialysis Study, Charytan et al. 11

treatable, rhythms, for the potential utility of and identi-fying those most likely to benefit from ICD insertion, fordetermining whether back-up pacing functions are needed,or for determining whether patients would benefit fromwithdrawal of nodal blockade (e.g., b-blockers).By capturing the full constellation of cardiac rhythm

during a prolonged period, the MiD study will provideinformation on the extent to which previously undetectedAF or other tachyarrhythmias and bradyarrhythmias con-tribute to morbidity in ESRD. Finally, as noted previously,the detailed capture of clinical data in MiD is critical forcorrelating captured rhythms with concurrent clinicalsymptoms to avoid confusing “agonal” rhythms that appearsecondary to some other life-ending event, such as sepsis orstroke (for which application of cardiac pacing devices is notappropriate), with primary arrhythmia as a cause of death.Analysis of blood chemistries and data on the dialysis

prescription from MiD may alternatively identify prescrip-tion characteristics or electrolyte levels strongly associatedwith the occurrence of arrhythmia. Thus, the final resultscould suggest ways to tailor the dialysis prescription in orderto limit arrhythmia. In this regard, associations betweenblood-dialysate potassium or calcium gradients and arrhyth-mia occurrence in MiD are eagerly anticipated.In the future, particularly as point-of-care information

becomes available, long-term monitoring may allow ne-phrologists to alter the dialysis prescription in response tominute-by-minute changes in electrocardiographic mor-phology or the peridialytic occurrence of arrhythmia.Long-term monitoring might allow the identification ofpatients with subclinical AF likely to benefit from anti-coagulation. ILR studies may also facilitate determination ofwhether certain dialysate-serum electrolyte gradients areassociated with a higher likelihood of arrhythmia andshould be avoided. Finally, clinical trials comparingoutcomes following randomization to an ILR with detectionof and response to arrhythmia versus routine standard ofcaremay be considered to determine whether the potential ofthese devices can reduce mortality in the dialysis population.In summary, we believe that leveraging modern cardiac

rhythm monitoring in persons with ESRD provides animportant opportunity to advance the science of cardiorenalmedicine. We await the results contributed by dedicated andaltruistic patients, families, and investigators who are carry-ing out the protocol at the time of this writing.

AcknowledgmentsThe authorswould like to acknowledge the support and feedback

from the study team at Medtronic including John Burnes and AmyRoettger.

The MiD study was funded by Medtronic and designed byMedtronic in collaboration with an advisory committee thatincluded the authors.

The following are the MiD investigators and committees: Nephrol-ogy investigators: Don Williamson (Kidney Care Associates, Augusta,GA), Prabir Roy-Chaudhury (University of Cincinnati Medical Cen-ter, Cincinnati, OH), James Tumlin (South East Renal Research Institu-tion, Chattanooga, TN), Vijay Kher (Medanta, The Metacity, Kidney &Urology Institute, Gurgaon, India), Vikranth Reddy (CAREHospital,Hyderabad, India), Kowdle Chandrasekhar Prakash (Apollo Hospi-tals, Chennai, India), David Charytan (Brigham&Women’s Hospital,Boston, MA), Suresh Chandra Tiwari (Fortis Vasant Kunj Hospital,

Delhi, India), SaurabhPokhariyal (FortisMemorial Research Institute,Gurgaon, India), Amber Podoll (University of Texas, Houston,Houston, TX), Sanjeev Jasuja (Apollo Hospitals, Delhi, India).Cardiology investigators: G. Leslie Walters (Augusta Cardiology

Clinic, Augusta, GA), Kraig Wangsnes (Cardiovascular Associates,Augusta, GA), Alexandru Costea (University of Cincinnati MedicalCenter, Cincinnati, OH), Selcuk Tombul (Diagnostic CardiologyGroup, Chattanooga, TN), Balbir Singh (Medanta, The Metacity,Kidney & Urology Institute, Gurgaon, India), Brajesh Mishra(Medanta, The Metacity, Kidney & Urology Institute, Gurgaon,India), Sachin Yalagudri (CARE Hospital, Hyderabad, India),Abhijeet Shelke (CARE Hospital, Hyderabad, India), CalamburNarasimhan (CARE Hospital, Hyderabad, India), A.M. Karthigesan(Apollo Hospitals, Chennai, Chennai, India), Abraham Oomman(Apollo Hospitals, Chennai, India), K.P. Pramod Kumar (ApolloHospitals, Chennai, India), Bruce Koplan (Brigham & Women’sHospital, Boston, MA), Upendra Kaul (Fortis Vasant Kunj Hospital,Delhi, India), TapanGhose (FortisVasantKunjHospital,Delhi, India),RipenGupta (Fortis Vasant Kunj Hospital, Delhi, India), Arvind Sethi(Fortis EscortsHospital, Delhi, India), Nikhil Kumar (FortisMemorialResearch Institute, Gurgaon, India), RameshHariharan (University ofTexas, Houston, Houston, TX), Rajnish Sardana (Apollo Hospitals,Delhi, India), Arif Wahab (Apollo Hospitals, Delhi, India), N.N.Khanna (Apollo Hospitals, Delhi, India).Nephrology coinvestigators: Mark Smith (Kidney Care Asso-

ciates, Augusta, GA), Suresh Kamath (University of CincinnatiMedical Center, Cincinnati, OH), ClaudeGalphin (South East RenalResearch Institution, Chattanooga, TN), Puneet Sodhi (Medanta, TheMetacity, Kidney & Urology Institute, Gurgaon, India), RajsekaraChakravarthy (CARE Hospital, Hyderabad, India), Subba RaoBudithi (Apollo Hospitals, Chennai, India), Finnian McCausland(Brigham & Women’s Hospital, Boston, MA), Sanjeev Gulati(Fortis Vasant Kunj Hospital, Delhi, India), Munawer Dijoo (FortisVasant Kunj Hospital, Delhi, India), Upendra Singh (Fortis EscortsHospital, Delhi, India), Salil Jain (Fortis Memorial ResearchInstitute, Gurgaon, India), Vishal Saxena (Fortis Memorial ResearchInstitute, Gurgaon, India), Gaurav Sagar (Apollo Hospitals, Delhi,India).Advisory committee: David Charytan (Brigham & Women’s

Hospital, Boston, MA), Rachel Fissell (Vanderbilt University,Nashville, TN), Robert Foley (Hennepin County Medical Center,Minneapolis, MN), Charles A. Herzog (Hennepin County MedicalCenter, University of Minnesota, Minneapolis, MN), PeterMcCullough (Baylor University Medical Center, Baylor Heartand Vascular Institute, Baylor Jack and Jane Hamilton Heart andVascular Hospital, Dallas, TX), John D. Rogers (Scripps Clinic-Torrey Pines, La Jolla, CA), James A. Tumlin (South East RenalResearch Institution, Chattanooga, TN), Peter Zimetbaum (BethIsrael Deaconess Medical Center, Boston, MA).Adverse events committee: Manish Assar (Baylor University

Medical Center, Dallas, TX), Mark Kremers (Mid Carolina Cardi-ology, Charlotte, NC), Wolfgang C. Winkelmayer (Baylor Collegeof Medicine, Houston, TX).

DisclosuresAll of the authors received research support and/or consulting

fees fromMedtronic (Minneapolis,MN) in relationship to thedesignof the study. D.M.C. received consulting from Questcor Pharma-ceuticals (Anaheim, CA), expert witness fees related to dialysatecomposition from FreseniusMedical Care (Waltham,MA), and feesrelated to service on clinical events committee from PLC MedicalSystems (Milford,MA). J.D.R. received speakers bureau, consultant,

12 Clinical Journal of the American Society of Nephrology

andresearch support fromMedtronic; speakersbureauandresearchsupport from St Jude Medical (St. Paul, MN); and speakers bureaufromBiotronik (Berlin, Germany). C.A.H. received research supportfrom Zoll (Chelmsford, MA) and stock/equity interest from Cam-bridge Heart, Boston Scientific (Tewksbury, MA).

References1. U.S. Renal Data System:USRDS 2013 Annual Data Report: Atlas

of End-Stage Renal Disease in the United States, Bethesda, MD,National Institutes of Health, National Institute of Diabetes andDigestive and Kidney Diseases, 2013

2. Wanner C, Krane V,MarzW,Olschewski M,Mann JF, Ruf G, RitzE; German Diabetes and Dialysis Study Investigators: Atorvas-tatin in patients with type 2 diabetes mellitus undergoing he-modialysis. N Engl J Med 353: 238–248, 2005

3. Cheung AK, SarnakMJ, YanG, BerkobenM, Heyka R, Kaufman A,Lewis J, Rocco M, Toto R, Windus D, Ornt D, Levey AS; HEMOStudy Group: Cardiac diseases in maintenance hemodialysis pa-tients: Results of theHEMOStudy.Kidney Int 65: 2380–2389, 2004

4. WheelerDC, LondonGM, Parfrey PS, BlockGA, Correa-Rotter R,Dehmel B,Drueke TB, Floege J, Kubo Y,Mahaffey KW,GoodmanWG,Moe SM, TrotmanML, Abdalla S, ChertowGM, Herzog CA;EValuation Of Cinacalcet HCl Therapy to Lower CardioVascularEvents (EVOLVE) Trial Investigators: Effects of cinacalcet on ath-erosclerotic and nonatherosclerotic cardiovascular events inpatients receiving hemodialysis: The EValuation Of CinacalcetHCl Therapy to Lower CardioVascular Events (EVOLVE) trial. JAm Heart Assoc 3: e001363, 2014

5. Herzog CA, Asinger RW, Berger AK, Charytan DM, Dıez J, HartRG, Eckardt KU, Kasiske BL, McCullough PA, Passman RS,DeLoach SS, Pun PH, Ritz E: Cardiovascular disease in chronickidney disease. A clinical update from Kidney Disease: Improv-ing Global Outcomes (KDIGO). Kidney Int 80: 572–586, 2011

6. Weinhandl E, Constantini E, Everson S, Gilbertson D, Li S, Solid C,AngerM, Bhat JG, DeOreo P, KrishnanM,NissensonA, JohnsonD,Ikizler TA, Maddux F, Sadler J, Tyshler L, Parker T 3rd, Schiller B,Smith B, Lindenfeld S, Collins AJ: Peer kidney care initiative 2014report: dialysis care andoutcomes in theUnitedStates.AmJKidneyDis 65: S1–S140, 2015

7. DrechslerC,GrootendorstDC,PilzS, TomaschitzA,KraneV,DekkerF, Marz W, Ritz E, Wanner C: Wasting and sudden cardiac deathin hemodialysis patients: A post hoc analysis of 4D (Die DeutscheDiabetes Dialyse Studie). Am J Kidney Dis 58: 599–607, 2011

8. Parekh RS, Plantinga LC, Kao WH, Meoni LA, Jaar BG, Fink NE,Powe NR, Coresh J, Klag MJ: The association of sudden cardiacdeathwith inflammation and other traditional risk factors.KidneyInt 74: 1335–1342, 2008

9. Chan KE, Newton-Cheh C, Gusella JF, Maddux FW: Heritabilityof risk for sudden cardiac arrest in ESRD. J Am Soc Nephrol 26:2815–2820, 2015

10. NakanoY, ShimizuW:Genetics of long-QT syndrome [Publishedonline ahead of print June 25, 2015]. J HumGenet doi: 10.1038/jhg.2015.74.

11. Sun Q, Song K, Shen X, Cai Y: The association between KCNQ1gene polymorphism and type 2 diabetes risk: A meta-analysis.PLoS One 7: e48578, 2012

12. Multicentre, cross-sectional study of ventricular arrhythmiasin chronically haemodialysed patients. Gruppo Emodialisi ePatologie Cardiovasculari. Lancet 2: 305–309, 1988

13. Sforzini S, Latini R, Mingardi G, Vincenti A, Redaelli B: Ven-tricular arrhythmias and four-year mortality in haemodialysispatients. Gruppo Emodialisi e Patologie Cardiovascolari. Lancet339: 212–213, 1992

14. Wan C, Herzog CA, Zareba W, Szymkiewicz SJ: Sudden cardiacarrest in hemodialysis patients with wearable cardioverter de-fibrillator. Ann Noninvasive Electrocardiol 19: 247–257, 2014

15. Wong MC, Kalman JM, Pedagogos E, Toussaint N, Vohra JK,Sparks PB, Sanders P, Kistler PM, Halloran K, Lee G, Joseph SA,Morton JB: Bradycardia and asystole is the predominant mech-anism of sudden cardiac death in patients with chronic kidneydisease. J Am Coll Cardiol 65: 1263–1265, 2015

16. Wong MC, Kalman JM, Pedagogos E, Toussaint N, Vohra JK,Sparks PB, Sanders P, Kistler PM, Halloran K, Lee G, Joseph SA,

Morton JB: Temporal distribution of arrhythmic events in chronickidney disease: Highest incidence in the long interdialytic pe-riod. Heart Rhythm 12: 2047–2055, 2015

17. Karnik JA, Young BS, LewNL, HergetM,Dubinsky C, Lazarus JM,Chertow GM: Cardiac arrest and sudden death in dialysis units.Kidney Int 60: 350–357, 2001

18. Pun PH, Lehrich RW, Honeycutt EF, Herzog CA, Middleton JP:Modifiable risk factors associated with sudden cardiac arrestwithin hemodialysis clinics. Kidney Int 79: 218–227, 2011

19. Pun PH, Horton JR, Middleton JP: Dialysate calcium concen-tration and the risk of sudden cardiac arrest in hemodialysispatients. Clin J Am Soc Nephrol 8: 797–803, 2013

20. Sakaguchi Y, Fujii N, Shoji T, Hayashi T, Rakugi H, Isaka Y:Hypomagnesemia is a significant predictor of cardiovascular andnon-cardiovascular mortality in patients undergoing hemodial-ysis. Kidney Int 85: 174–181, 2014

21. Foley RN, Gilbertson DT, Murray T, Collins AJ: Long interdialyticinterval and mortality among patients receiving hemodialysis.N Engl J Med 365: 1099–1107, 2011

22. Bleyer AJ, Russell GB, Satko SG: Sudden and cardiac death ratesin hemodialysis patients. Kidney Int 55: 1553–1559, 1999

23. Fotheringham J, Fogarty DG, El Nahas M, Campbell MJ,Farrington K: The mortality and hospitalization rates associatedwith the long interdialytic gap in thrice-weekly hemodialysispatients. Kidney Int 88: 569–575, 2015

24. Zhang H, Schaubel DE, Kalbfleisch JD, Bragg-Gresham JL,Robinson BM, Pisoni RL, Canaud B, Jadoul M, Akiba T, Saito A,Port FK, Saran R: Dialysis outcomes and analysis of practicepatterns suggests the dialysis schedule affects day-of-week mor-tality. Kidney Int 81: 1108–1115, 2012

25. Wolk R: Arrhythmogenic mechanisms in left ventricular hyper-trophy. Europace 2: 216–223, 2000

26. Huikuri HV, Castellanos A, Myerburg RJ: Sudden death due tocardiac arrhythmias. N Engl J Med 345: 1473–1482, 2001

27. Herzog CA, Mangrum JM, Passman R: Sudden cardiac death anddialysis patients. Semin Dial 21: 300–307, 2008

28. Spaulding CM, Joly LM, Rosenberg A, Monchi M, Weber SN,Dhainaut JF, Carli P: Immediate coronary angiography in survi-vors of out-of-hospital cardiac arrest. N Engl J Med 336: 1629–1633, 1997

29. Zoccali C, Mallamaci F, Parlongo S, Cutrupi S, Benedetto FA,Tripepi G, Bonanno G, Rapisarda F, Fatuzzo P, Seminara G,Cataliotti A, Stancanelli B, Malatino LS: Plasma norepinephrinepredicts survival and incident cardiovascular events in patientswith end-stage renal disease. Circulation 105: 1354–1359, 2002

30. Stenvinkel P: Inflammation in end-stage renal disease: the hiddenenemy. Nephrology (Carlton) 11: 36–41, 2006

31. Foley RN, Curtis BM, Randell EW, Parfrey PS: Left ventricularhypertrophy in new hemodialysis patients without symptomaticcardiac disease. Clin J Am Soc Nephrol 5: 805–813, 2010

32. Mall G, Huther W, Schneider J, Lundin P, Ritz E: Diffuse inter-myocardiocytic fibrosis in uraemic patients. Nephrol DialTransplant 5: 39–44, 1990

33. Amann K, Breitbach M, Ritz E, Mall G: Myocyte/capillary mis-match in the heart of uremic patients. J Am Soc Nephrol 9: 1018–1022, 1998

34. McIntyre CW, Burton JO, Selby NM, Leccisotti L, Korsheed S,Baker CS, Camici PG: Hemodialysis-induced cardiac dysfunc-tion is associated with an acute reduction in global and seg-mental myocardial blood flow. Clin J Am Soc Nephrol 3: 19–26,2008

35. Charytan DM, Padera R, Helfand AM, Zeisberg M, Xu X, Liu X,Himmelfarb J, Cinelli A, Kalluri R, Zeisberg EM: Increased con-centration of circulating angiogenesis and nitric oxide inhibitorsinduces endothelial to mesenchymal transition and myocardialfibrosis in patients with chronic kidney disease. Int J Cardiol 176:99–109, 2014

36. Matsumoto Y, Ueda S, Yamagishi S, Matsuguma K, Shibata R,Fukami K, Matsuoka H, Imaizumi T, Okuda S: Dimethylargininedimethylaminohydrolase prevents progression of renal dysfunc-tion by inhibiting loss of peritubular capillaries and tubulointer-stitial fibrosis in a rat model of chronic kidney disease. J Am SocNephrol 18: 1525–1533, 2007

37. Repo JM, Rantala IS, Honkanen TT, Mustonen JT, Koobi P,Tahvanainen AM, Niemela OJ, Tikkanen I, Rysa JM, Ruskoaho HJ,

Clin J Am Soc Nephrol ▪: ccc–ccc, ▪▪▪, 2016 Design of the Monitoring in Dialysis Study, Charytan et al. 13

Porsti IH: Paricalcitol aggravates perivascular fibrosis in rats withrenal insufficiency and low calcitriol. Kidney Int 72: 977–984,2007

38. Paulis L, Matuskova J, Adamcova M, Pelouch V, Simko J,Krajcirovicova K, Potacova A, Hulin I, Janega P, Pechanova O,Simko F: Regression of left ventricular hypertrophy and aorticremodelling in NO-deficient hypertensive rats: Effect ofL-arginine and spironolactone. Acta Physiol (Oxf) 194: 45–55,2008

39. Faul C, Amaral AP, Oskouei B, Hu MC, Sloan A, Isakova T,Gutierrez OM, Aguillon-Prada R, Lincoln J, Hare JM, Mundel P,Morales A, Scialla J, FischerM, Soliman EZ, Chen J, Go AS, RosasSE, Nessel L, Townsend RR, FeldmanHI, St John SuttonM,Ojo A,Gadegbeku C, Di Marco GS, Reuter S, Kentrup D, Tiemann K,Brand M, Hill JA, Moe OW, Kuro-O M, Kusek JW, Keane MG,Wolf M: FGF23 induces left ventricular hypertrophy. J Clin Invest121: 4393–4408, 2011

40. Brilla CG: Renin-angiotensin system mediated mechanisms:Cardioreparation and cardioprotection. Heart 84 Suppl 1: i18–19:discussion i50, 2000

41. Kennedy DJ, Vetteth S, Periyasamy SM, Kanj M, Fedorova L,Khouri S, Kahaleh MB, Xie Z, Malhotra D, Kolodkin NI, LakattaEG, Fedorova OV, Bagrov AY, Shapiro JI: Central role for thecardiotonic steroid marinobufagenin in the pathogenesis of experi-mental uremic cardiomyopathy. Hypertension 47: 488–495, 2006

42. Mizobuchi M, Nakamura H, Tokumoto M, Finch J, Morrissey J,Liapis H, Slatopolsky E: Myocardial effects of VDR activators inrenal failure. J Steroid Biochem Mol Biol 121: 188–192, 2010

43. Saffitz JE: The pathology of sudden cardiac death in patients withischemic heart disease–arrhythmology for anatomic pathologists.Cardiovasc Pathol 14: 195–203, 2005

44. Goldstein BA, Arce CM, Hlatky MA, Turakhia M, Setoguchi S,Winkelmayer WC: Trends in the incidence of atrial fibrillation inolder patients initiating dialysis in the United States. Circulation126: 2293–2301, 2012

45. Winkelmayer WC, Patrick AR, Liu J, Brookhart MA, Setoguchi S:The increasing prevalence of atrial fibrillation among hemodi-alysis patients. J Am Soc Nephrol 22: 349–357, 2011

46. Zimmerman D, Sood MM, Rigatto C, Holden RM, Hiremath S,Clase CM: Systematic review and meta-analysis of incidence,prevalence and outcomes of atrial fibrillation in patients on di-alysis. Nephrol Dial Transplant 27: 3816–3822, 2012

47. Tekce H, Ozturk S, Aktas G, Tekce BK, Erdem A, Ozyasar M,Duman TT, Yazici M: The effects of a single dialysis session onatrial electromechanical conduction times and functions.KidneyBlood Press Res 37: 622–630, 2013

48. Steinberg JS, Zelenkofske S, Wong SC, Gelernt M, Sciacca R,Menchavez E: Value of the P-wave signal-averaged ECG forpredicting atrial fibrillation after cardiac surgery. Circulation 88:2618–2622, 1993

49. Fukunami M, Yamada T, Ohmori M, Kumagai K, Umemoto K,Sakai A, KondohN,Minamino T, Hoki N:Detection of patients atrisk for paroxysmal atrial fibrillation during sinus rhythm by Pwave-triggered signal-averaged electrocardiogram. Circulation83: 162–169, 1991

50. Ozben B, Toprak A, Koc M, Sumerkan M, Tanrikulu AM, Papila-Topal N, Kefeli US, Cincin AA, Baykan O, Fak AS: P wave dis-persion increases during hemodialysis sessions. Nephron ClinPract 112: c171–c176, 2009

51. Shimada K, Tomita T, Kamijo Y, Higuchi M, Ito K, Koizumi Y,Yamamoto K, AizawaK, KoshikawaM, KasaiH, IzawaA,MiyashitaY, Kumazaki S, Koyama J, Ikeda U: Hemodialysis-induced P-wavesignal-averaged electrocardiogram alterations are indicative ofvulnerability to atrial arrhythmias. Circ J 76: 612–617, 2012

52. Szabo Z, Kakuk G, Fulop T, Matyus J, Balla J, Karpati I, Juhasz A,Kun C, Karanyi Z, Lorincz I: Effects of haemodialysis on maxi-mum P wave duration and P wave dispersion. Nephrol DialTransplant 17: 1634–1638, 2002

53. Jaroszy�nski AJ, Głowniak A, Sodolski T, Załuska W, Widomska-Czekajska T, Ksiazek A: Effect of haemodialysis on signal-averagedelectrocardiogramP-waveparameters.NephrolDial Transplant21:425–430, 2006

54. Buiten MS, de Bie MK, Rotmans JI, Gabreels BA, van Dorp W,Wolterbeek R, Trines SA, Schalij MJ, Jukema JW, Rabelink TJ, vanErven L: The dialysis procedure as a trigger for atrial fibrillation:

New insights in the development of atrial fibrillation in dialysispatients. Heart 100: 685–690, 2014

55. Wolf PA, Abbott RD, Kannel WB: Atrial fibrillation as an in-dependent risk factor for stroke: the Framingham Study. Stroke22: 983–988, 1991

56. Wetmore JB, Ellerbeck EF, Mahnken JD, Phadnis M, Rigler SK,Mukhopadhyay P, Spertus JA, Zhou X, HouQ, Shireman TI: Atrialfibrillation and risk of stroke in dialysis patients. Ann Epidemiol23: 112–118, 2013

57. Li SY, Chen YT, Chen TJ, Tsai LW, YangWC, Chen TW:Mesentericischemia in patients with end-stage renal disease: A nationwidelongitudinal study. Am J Nephrol 35: 491–497, 2012

58. CombeC, Albert JM, Bragg-Gresham JL, Andreucci VE, DisneyA,Fukuhara S, GoodkinDA,Gillespie BW, Saito A, JadoulM, PisoniRL: The burden of amputation among hemodialysis patients inthe Dialysis Outcomes and Practice Patterns Study (DOPPS). AmJ Kidney Dis 54: 680–692, 2009

59. Zimmermann AJ, BossardM, Aeschbacher S, Schoen T, VoellminG, Suter Y, Lehmann A, Hochgruber T, Pumpol K, Sticherling C,Kuhne M, Conen D, Kaufmann BA: Effects of sinus rhythmmaintenance on left heart function after electrical cardioversionof atrial fibrillation: implications for tachycardia-induced car-diomyopathy. Can J Cardiol 31: 36–43, 2015

60. Glotzer TV, Hellkamp AS, Zimmerman J, Sweeney MO, Yee R,Marinchak R, Cook J, Paraschos A, Love J, RadoslovichG, Lee KL,Lamas GA; MOST Investigators: Atrial high rate episodes de-tected by pacemaker diagnostics predict death and stroke: reportof the Atrial Diagnostics Ancillary Study of the MOde SelectionTrial (MOST). Circulation 107: 1614–1619, 2003

61. Healey JS, Connolly SJ, Gold MR, Israel CW, Van Gelder IC,Capucci A, Lau CP, Fain E, Yang S, Bailleul C,Morillo CA, CarlsonM, Themeles E, Kaufman ES, Hohnloser SH; ASSERT Investiga-tors: Subclinical atrial fibrillation and the risk of stroke. N Engl JMed 366: 120–129, 2012

62. KarthikeyanG,Ananthakrishnan R,DevasenapathyN,Narang R,Yadav R, Seth S, Singh S, Goswami KC, Bahl VK: Transient,subclinical atrial fibrillation and risk of systemic embolism inpatients with rheumatic mitral stenosis in sinus rhythm. Am JCardiol 114: 869–874, 2014

63. Zimetbaum P,Waks JW, Ellis ER, Glotzer TV, Passman RS: Role ofatrial fibrillation burden in assessing thromboembolic risk. CircArrhythm Electrophysiol 7: 1223–1229, 2014

64. Sanna T, Diener H-C, Passman RS, Di Lazzaro V, Bernstein RA,Morillo CA, Rymer MM, Thijs V, Rogers T, Beckers F, Lindborg K,Brachmann J: Cryptogenic stroke and underlying atrial fibrilla-tion. N Engl J Med 370: 2478–2486, 2014

65. Medtronic Inc.Reveal LINQICMClinicianManualAvailableathttp://manuals.medtronic.com/wcm/groups/mdtcom_sg/@emanuals/@era/@crdm/documents/documents/contrib_185856.pdf. AccessedJuly 15, 2015

66. Medtronic Inc. Reveal LINQMRIConditions forUse. 2015.Availableathttp://www.medtronicdiagnostics.com/us/cardiac-monitors/reveal-linq/reveal-linq-cardiac-monitor/mri-conditions-for-use/index.htm.Accessed August 31, 2015

67. Zimetbaum P, Goldman A: Ambulatory arrhythmia monitoring:Choosing the right device. Circulation 122: 1629–1636, 2010

68. Gladstone DJ, Sharma M, Spence JD; EMBRACE Steering Com-mittee and Investigators: Cryptogenic stroke and atrial fibrilla-tion. N Engl J Med 371: 1260, 2014

69. Echt DS, Liebson PR, Mitchell LB, Peters RW, Obias-Manno D,Barker AH, Arensberg D, Baker A, Friedman L, Greene HL,Huther ML, Richardson DW; CAST Investigators: Mortality andmorbidity in patients receiving encainide, flecainide, or placebo.The Cardiac Arrhythmia Suppression Trial. N Engl J Med 324:781–788, 1991

70. Zimetbaus PJ, Josephson ME, Wylie JV: Management of non-sustained ventricular tachycardia. UpToDate. 2015. Available athttp://www.uptodate.com/contents/management-of-non-sustained-ventricular-tachycardia. Accessed June 1, 2015

71. Ganz LI: Sinus bradycardia. UpToDate. 2015. Available at http://www.uptodate.com/contents/sinus-bradycardia. AccessedJune 1, 2015

72. Newcombe RG: Two-sided confidence intervals for the singleproportion: Comparison of seven methods. Stat Med 17: 857–872, 1998

14 Clinical Journal of the American Society of Nephrology

73. Cook RJ, Lawless J: The Stastical Analysis of Recurrent Events,New York, Springer, 2007

74. Williams ME, Lacson E Jr, Teng M, Ofsthun N, Lazarus JM: He-modialyzed type I and type II diabetic patients in the US: Char-acteristics, glycemic control, and survival. Kidney Int 70: 1503–1509, 2006

75. Dumler F, McCullough PA: Optimal dialysis for the end-stagerenal disease patient with cardiovascular disease. Adv ChronicKidney Dis 11: 261–273, 2004

76. Keeley EC, McCullough PA: Coronary revascularization in pa-tients with end-stage renal disease: Risks, benefits, and optimalstrategies. Rev Cardiovasc Med 4: 125–130, 2003

77. McCullough PA: Opportunities for improvement in the cardio-vascular care of patients with end-stage renal disease. AdvChronic Kidney Dis 11: 294–303, 2004

78. Woo KT,Wong KS, Chan CM: Clinical trials of the past decade inthe management of chronic kidney disease. Rev Recent ClinTrials 4: 159–162, 2009

79. Culleton BF, Asola MR: The impact of short daily and nocturnalhemodialysis on quality of life, cardiovascular risk and survival.J Nephrol 24: 405–415, 2011

80. Kusumoto FM, Phillips R, Goldschlager N: Pacing therapy in theelderly. Am J Geriatr Cardiol 11: 305–316, 2002

81. Prystowsky EN: Prevention of sudden cardiac death.Clin Cardiol28[Suppl 1]: I12–I18, 2005

82. Yousuf O, Chrispin J, Tomaselli GF, Berger RD: Clinical man-agement and prevention of sudden cardiac death. Circ Res 116:2020–2040, 2015

83. Wase A, Basit A, Nazir R, Jamal A, Shah S, Khan T, Mohiuddin I,White C, Saklayen M, McCullough PA: Impact of chronickidney disease upon survival among implantable cardioverter-defibrillator recipients. J Interv Card Electrophysiol 11: 199–204, 2004

Published online ahead of print. Publication date available at www.cjasn.org.

This article contains supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.09350915/-/DCSupplemental.

Clin J Am Soc Nephrol ▪: ccc–ccc, ▪▪▪, 2016 Design of the Monitoring in Dialysis Study, Charytan et al. 15