Business & Finance

Antidiabetic Action of Bezafibrate in a Large Observational Database

Antidiabetic Action of Bezafibrate in a Large Observational Database
of 5
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
   Antidiabetic Action of Bezafibrate in aLarge Observational Database  J AMES  H. F LORY 1 S USAN  E LLENBERG ,  PHD 1 P HILIPPE  O. S ZAPARY ,  MD, MSCE 1,2 B RIAN  L. S TROM ,  MD, MPH 1 S EAN  H ENNESSY ,  PHARMD, PHD 1 OBJECTIVE — The purpose of this study was to test the hypothesis that bezafibrate, anapproved fibrate, can prevent or delay type 2 diabetes. RESEARCHDESIGNANDMETHODS — This was a retrospective cohort study usingdata from routine medical practice in the U.K., as captured by the General Practice ResearchDatabase (GPRD). Individuals chronically exposed to bezafibrate were compared with individ-uals chronically exposed to other fibrates. Hazard ratios (HRs) for incident type 2 diabetes werecalculatedusingaCoxproportionalhazardsmodel.Aposthocanalysiswasusedtoexaminetheeffect of bezafibrate on progression to use of oral antidiabetic medications or insulin in individ-uals with diabetes at baseline. RESULTS — Bezafibrate users had a lower hazard for incident diabetes than users of otherfibrates (HR 0.66 [95% CI 0.53–0.81]). This effect became stronger with increasing duration of therapy. Post hoc analysis of the effect of bezafibrate on progression of preexisting diabetes alsoshowed a lower hazard for progression to use of antidiabetic medication (0.54 [0.38–0.76]) orprogression to use of insulin (0.78 [0.55–1.10]). CONCLUSIONS — Bezafibrate appears to have clinically important antidiabetic properties.Randomized controlled trials should be considered to assess the utility of bezafibrate in treatingpatients with diabetes or in preventing diabetes in high-risk patients. Diabetes Care  32:547–551, 2009 T ype 2 diabetes is a major publichealth threat, expected to affectmore than 221 million peopleworldwide by 2010 (1). One key targetfor diabetes drug development is the per-oxisome proliferator–activated receptor(PPAR) (2,3). There are three isotypesthat are of specific interest in metabolicdiseases: PPAR-  , PPAR-  , and PPAR-  .The thiazolidinediones (e.g., pioglita-zone) are PPAR-   agonists used to treatdiabetes through improvement of insulinresponse. The fibrates are PPAR-   ago-nists used to treat dyslipidemia by raisingHDL and lowering triglycerides. PPAR-  remains an investigational drug targetwith potential uses in diabetes, dyslipide-mia, and obesity (4). Because dyslipidemiaand diabetes are commonly comorbid, at-tempts have been made to create dualPPAR-   /    agonists or pan-PPAR agonists,althoughnoneofthesehasreachedthemar-ket (3).Inresponsetoeffortstodeveloptheseagents, at least one observer has pointedoutthatthefibratebezafibrateactuallyisapan-PPAR agonist and affects insulin re-sistance(5).Posthocanalysesofaplacebo-controlled randomized trial showed thatbezafibrate may postpone or prevent type2diabetes(5,6).Duringamean6yearsof follow-up,hazardratios(HRs)versuspla-cebo for incident diabetes were 0.59(95%CI0.39–0.91)inobesepatients(5)and 0.70 (95% CI 0.49–0.99) in pre-diabetic patients (6). These clinical endpoint data were supported by biochemi-cal evidence showing that bezafibrateslowed progression of insulin resistance(7).Studies of the other fibrates (gemfi-brozil,fenofibrate,ciprofibrate,andclofi-brate) have not shown such effects(8–14), and these drugs are far more se-lective for PPAR-   than bezafibrate (15).Hence,itisreasonabletohypothesizethatthe status of bezafibrate as a pan-PPAR agonistmaygiveitantidiabeticpropertiesunique among fibrates. Although not approved in the U.S.,bezafibratehasbeenwidelyprescribedfordyslipidemia in the U.K. We used obser-vational data to examine the a priori hy-pothesisthatbezafibrateisuniqueamongfibrates in reducing diabetes risk. RESEARCH DESIGN ANDMETHODS — WeconductedacohortstudyusingtheGeneralPracticeResearchDatabase (GPRD). Personal informationwas removed before inclusion in the da-tabase. The requirement for informedconsent was waived by the University of Pennsylvania institutional review boardandtheGPRDIndependentScientificAd-visory Committee. Data source The GPRD contains data abstracted froma computerized medical record systemusedbyasubsetofgeneralpracticesintheU.K. Ninety-eight percent of the U.K.population receives all forms of healthcare through their general practitioners.The database is broadly representative of the U.K. population in terms of sex, age,and geography (16). We used data from1988 through 2002.The information prospectively col-lected in the database includes demo-graphic information, all prescriptionswritten by the general practitioner, clini-cal diagnoses, specialty consultationnotes, and hospital discharge diagnoses.Medical diagnoses are classified usingRead Clinical Classification and the Ox-ford Medical Information System codes.Participating general practices followprospectively designed protocols for re-cording computerized clinical informa- ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Fromthe 1 CenterforClinicalEpidemiologyandBiostatistics,UniversityofPennsylvaniaSchoolofMedicine,Philadelphia, Pennsylvania; and  2 Centocor Research and Development, Malvern, Pennsylvania, and Pre-ventive Cardiovascular Medicine and Lipid Clinic, Department of Medicine, University of PennsylvaniaHealth System, Philadelphia, Pennsylvania.Corresponding author: James Flory, 2 October and accepted 4 January 2009.Published ahead of print at on 8 January 2009. DOI: 10.2337/dc08-1809.No sponsor had any role in the design and conduct of the study; collection, management, analysis, andinterpretation of the data; and preparation, review, or approval of the manuscript.© 2009 by the American Diabetes Association. Readers may use this article as long as the work is properlycited, the use is educational and not for profit, and the work is not altered. See for details. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be herebymarked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. C l i n i c a l C a r e / E d u c a t i o n / N u t r i t i o n / P s y c h o s o c i a l R e s e a r c h O R I G I N A L A R T I C L E D IABETES  C ARE ,  VOLUME  32,  NUMBER   4, A PRIL  2009  547  tion and uploading it to the researchdatabase. Data reaching predefined qual-itystandardsaresodesignated.Morethan400 published epidemiological studieshave been performed using the GPRD(16,17). Study cohort From all patients being followed in theGPRD, we included only person-timefrom individuals who were exposed to afibrate. Individuals were only includedwho had been registered and up-to-standard with the GPRD for at least 12months before initiation of the exposuredrug, making this an inception cohort.Because the primary outcome of in-terest was incident diabetes, any diagnos-tic code for diabetes or any use of homeglucose-monitoring equipment or of drugs that are only used to treat diabetes(insulin, biguanides, sulfonylureas, thia-zolidinediones, or acarbose) before thefirstfibrateprescriptionorwithinthefirst90 days of fibrate therapy excluded thatindividual from participation in the pri-mary analysis. The rationale for this ex-clusion was to avoid including prevalentdiabetic subjects in the study cohort. Exposure definition The study group included those withmore than one prescription for bezafi-brate, as a way to identify those receivingchronic treatment. Because we excludedindividuals developing diabetes withinthe first 90 days of therapy, we began fol-low-up with the 91st day of fibrate ther-apy. The duration of each prescriptionwas either provided in the database or,when this information was missing, esti-mated from the number of pills dis-pensed. Exposure was assumed tocontinue 30 days after the end of the ex-pected duration of the last prescription.Gaps over 60 days longer than expectedbetweenprescriptionswereconsideredtomark a last prescription, although a pa-tient could reenter the cohort with thenext prescription. Clustering methodswere used to account for single patientscontributing multiple blocks of time tothe cohort, and sensitivity analysis wasdone in which patients were censored atthefirstgapandnotallowedtoreenterthecohort.Control groups were defined usingparallelcriteria,withperson-timeforcon-trol subjects defined by exposure to otherfibrateswithoutanyhistoryofbezafibrateuse. The prespecified plan for this studywastomaximizepowerbyconsideringallnonbezafibrate fibrates as a single expo-sure and only distinguishing between in-dividual fibrates in a secondary analysis. Any patient who switched from onestudy group to another was censored atthe time of the switch. In a secondaryanalysis, each specific fibrate constitutedits own exposure group, compared withbezafibrate. Outcome definition The outcome of interest was clinical diag-nosisofortreatmentfordiabetes,definedby at least two codes indicative of diabe-tes. Such codes included any diagnosticcode for diabetes, any prescription forhome glucose-monitoring equipment, orany prescription for insulin or an oral an-tidiabetic drug. Post hoc analysis In a post hoc analysis, two additional co-horts were created. These consisted of in-dividuals who would have been eligiblefor the primary study but were excludedbecause of diabetes occurring before the91st day of fibrate treatment. These indi-viduals with baseline diabetes were di-vided into two groups. The new cohortconsisted of individuals who had un-treated diabetes at baseline (as identifiedby medical codes for diabetes or use of home glucose-monitoring equipment butno use of any antidiabetic medication).For this cohort, the outcome of interestwasprogressiontouseofanyantidiabeticmedication. In addition, individuals whowereusingoralantidiabetictherapy(abi-guanide, sulfonylurea, thiazolidinedione,or acarbose) at baseline were treated as aseparate cohort, with progression to useof insulin as the study outcome. Statistical analysis  All exposure groups were first comparedon baseline variables. For each exposuregroup, event rates were calculated. Next,Cox proportional hazard models wereused to estimate unadjusted and adjustedHRs.Fullyadjustedmodelswerereportedwith all variables included in the model.Covariates included the year that theexposure therapy was initiated. They alsoincluded a preidentified list of factorsknown to be associated either positivelyor negatively with diabetes. These in-cluded sex, age, history of stroke, historyof myocardial infarction, and use of thefollowing drugs: ACE inhibitors, calciumchannelblockers,  -blockers,thiazidedi-uretics, loop diuretics, and corticoste-roids (18). These drugs were analyzed asbaselinecovariatesandinsensitivityanal-ysis as time-varying covariates. BMI andsmoking status were available only for aportion of the population and were in-cluded only in secondary analyses. Thepresence of comorbidities was deter-mined on the basis of identification of GPRD medical diagnostic codes in theyear before the first fibrate prescription.Five secondary analyses were per-formed:  1 ) comparison of bezafibrate us-ersversususersofeachindividualfibrate;  2 ) stratification by duration of therapy;  3 )stratification of bezafibrate users into ap-proximate quartiles of average dosage(  200, 200–400, 400–600, and 600  mg/day) with use of the low-dose cate-goryasareferencegroup; 4 )restrictiontosubjects with baseline BMI data and in-corporation of BMI into multivariablemodeling; and  5 ) restriction to subjectwith baseline smoking data and incorpo-ration of smoking into multivariablemodeling. RESULTS — Bezafibrate was used farmore commonly (12,161 users) than anyother fibrate (4,191 users). Of the otherfibrateusers,1,465usedciprofibrate,502used clofibrate, 824 used fenofibrate, and1,400 used gemfibrozil. Baseline charac-teristics of bezafibrate users and otherfibrate users were consistent with previ-ously published research (Table 1) (18).Because of the large sample size, mostbaseline differences were statisticallysignificant. However, few clinically sig-nificant differences were observed. Of interest, however, the prevalence of re-corded obesity was very similar (5% vs.6%) between the two groups. However,bezafibrate users were more likely to befemale than other fibrate users (48% vs.40%).Users of all other fibrates were lesslikelytohavebaselinediabetesthanusersof bezafibrate (relative risk 0.90 [95% CI0.82–0.98] adjusted for year of the firstfibrate prescription). However, bezafi-brate users were less likely to have diabe-tes before treatment initiation comparedwith only one subgroup, the fenofibrateusers (1.25 [1.08–1.42] adjusted for yearof first fibrate prescription). Among bezafibrate users, 272 newcases of diabetes occurred, for an inci-dence rate of 8.5 cases per 1,000 patient-years (95% CI 7.5–9.5). Among users of the other fibrates, 131 new cases of dia-betes occurred, for an incidence rate of 14.4 cases per 1,000 patient-years (12.1–17.1). Bezafibrate for prevention of diabetes 548  D IABETES  C ARE ,  VOLUME  32,  NUMBER   4, A PRIL  2009  Cox proportional hazard regressionresults are shown in Table 2. The unad- justed HR for the comparison betweenbezafibrate and all other fibrates was 0.58(95%CI0.47–0.72).Adjustingforyearof treatmentinitiationattenuatedtheassoci-ation slightly, yielding a HR of 0.64(0.52–0.79).Noothervariablesmodifiedthe point estimate by as much as 10%.The fully adjusted HR for incident type 2diabetes was 0.66 (0.53–0.81,  P   0.0001). Analyses were repeated withstratification by year of treatment initia-tion, with no substantial change in theresults or evidence of heterogeneity of re-sults by year (data not shown).Table 2 also shows each individual fi-bratetreatedasadistinctreferencegroup.Bezafibrate had similar adjusted HRscompared with those for ciprofibrate,clofibrate, and gemfibrozil (Table 2).However, compared with fenofibrate,bezafibrate was associated with a particu-larly low hazard for diabetes (HR 0.41,95% CI 0.29–0.58).Table 3 shows the results of the dura-tion-response analysis. The HR declinedmonotonically as duration of therapyincreased.No significant relationship with aver-age daily dose was seen (data not shown).Neither restriction to individuals withBMIdatanoradjustmentforBMIsubstan-tially altered the results, although BMI of 25–30 kg/m 2 was associated with an HR of 3.72 (95% CI 1.89–7.29) and BMI  30 kg/m 2 was associated with an HR of 6.98 (95% CI 3.51–13.88), with BMI  25 kg/m 2 as a reference group. Thesame was true for restriction to individ-uals with baseline smoking data and in-clusion of that information in themultivariable model (not shown).In post hoc analysis, individuals withbaseline diabetes were classified as eitherunmedicated (no record of use of antidi-abetic medication) or receiving oral an-tidiabetic medication. The distribution of baselinecharacteristicsbetweenexposuregroupswasgenerallysimilartothatinthesrcinal cohort (data not shown).Table4showsHRscalculatedforpro-gression from unmedicated diabetes atthe time of fibrate initiation to use of anyantidiabetic medication (including insu-lin), as well as HRs for progression fromuseofanoralantidiabeticdrugatbaselineto insulin use. Bezafibrate was associatedwith a lower hazard of progression to an-tidiabetic medication use compared withfibrates (HR 0.54, 95% CI 0.38–0.76).TheHRswerenotsubstantiallyalteredbymultivariable adjustment.The analysis was repeated with indi-viduals who used oral antidiabetic medi-cations at baseline, with progression toinsulin therapy as the outcome. Bezafi-bratewasassociatedwithanonsignificanttrend toward a lower hazard of progres-siontoinsulintherapy(adjustedHR0.78,95% CI 0.55–1.10).No significant differences in HRswere observed for sex. Data on ethnicitywere not available for this study. CONCLUSIONS — This study pro-vides strong evidence that bezafibrate hasantidiabetic properties, supporting both Table 1— Baseline characteristics and number of events in exposure groups Bezafibrate All other fibrates P  value fordifference n  12,161 4,191Person-years 32,091 9,067Mean duration of use (years) 2.6 2.2   0.0001Mode year of treatment initiation 1993 1994   0.0001 Age (years)50 20 22 0.011250–59 33 33 0.984060–69 37 33   0.0001  69 10 12 0.0008Male sex 52 60   0.0001History of myocardial infarction 1 1 0.7529History of stroke 0 0 0.9676History of ACE inhibitor/angiotensin receptorblocker use5 6 0.0018History of calcium channel blocker use 24 22 0.1756History of    -blocker use 16 17 0.2382History of loop diuretic use 5 5 0.2178History of thiazide diuretic use 9 8 0.2883History of corticosteroid use 3 3 0.6388Never smoker 19 21 0.5114Ever smoker 39 41 0.5114Not reported 42 38   0.0001BMI  25 kg/m 2 9 8 0.044925–29.9 kg/m 2 13 13 0.6363  29.9 kg/m 2 5 6 0.0035Not reported 73 72 0.0523Number of cases of incident diabetes 272 131Cases/1,000 person-years (95% CI) 8.5 (7.5–9.5) 14.4 (12.1–17.1)   0.001 Data are % unless indicated otherwise. Individuals with baseline diabetes were excluded. History of cardio-vascular events and drug use refer to history in the year before cohort entry.  P  values were generated using  2 and  t  tests. Table 2— Prespecified secondary analyses consisting of HRs for exposure to bezafibrate Reference groupHRs (95% CI) for incident type 2 diabetes inindividuals exposed to bezafibrateUnadjusted Fully adjusted All fibrate users 0.58 (0.47–0.72) 0.66 (0.53–0.81)Ciprofibrate users 0.53 (0.39–0.73) 0.72 (0.52–0.99)Clofibrate users 1.17 (0.63–2.14) 0.78 (0.54–1.14)Gemfibrozil users 0.30 (0.21–0.42) 0.84 (0.46–1.55)Fenofibrate users 0.81 (0.57–1.19) 0.41 (0.29–0.58) Fully adjusted HRs are adjusted for year of treatment initiation, age, sex, history of congestive heart failure,historyofmyocardialinfarction,andhistoryofuseofthiazidediuretics,loopdiuretics,  -blockers,calcium-channel blockers, ACE inhibitors, angiotensin receptor blockers, or steroids. Flory and Associates D IABETES  C ARE ,  VOLUME  32,  NUMBER   4, A PRIL  2009  549  in vitro data and earlier post hoc analysessuggestingthatbezafibratecanpreventordelaytheonsetoftype2diabetes(5–7).Itfurther indicates that this effect is uniqueto bezafibrate among the fibrates. Thesefindings have important implications forresearch. Our findings are bolstered bythe similarity of subjects in exposuregroups on clinically relevant character-istics, the fact that the finding of a pro-tective effect is of a clinically relevantmagnitude, statistically significant, androbust to sensitivity analyses includingadjustment for BMI, and the monotonicduration-response relationship.Theresultsoftheposthocanalysisarereassuring.Itwasreasonabletoworrythatfenofibrate was more likely to be pre-scribed to individuals with a high risk fordiabetes or unrecorded diabetes, creatinga falsely elevated hazard for developmentof diabetes during fenofibrate treatmentcompared with bezafibrate treatment. Itwashenceusefultodoaposthocanalysisconfined to individuals who already haddiabetes. In this post hoc analysis, bezafi-brate also appeared to have antidiabeticproperties.Taken together, these findings sup-port and complement previous observa-tions. Post hoc analyses of the BezafibrateInfarction Prevention (BIP) Study havesuggestedthatbezafibratemayreducethehazard for incident diabetes, with pointestimates for the HR of 0.59 and 0.70(5,6). The fully adjusted point estimatefrom our study (0.66, 95% CI 0.3–0.81)is very consistent with those earlier re-sults. These additional results are impor-tant because they confirm a post hocanalysis in a new study with this as a pre-specified hypothesis, they generalize theresults to a broader population than thesrcinal post hoc analysis did, and theyprovide considerably more precise pointestimates. Another publication from the BIPshowed that bezafibrate attenuated pro-gression of the homeostasis model assess-ment of insulin resistance marker forinsulin resistance in all patients, suggest-ing that bezafibrate might also slow pro-gression of diabetes (7). Our results wereconsistent with that hypothesis as well,both for progression from diagnosis toany use of antidiabetic drugs (HR 0.54,95% CI 0.38–0.76) and for progressionfrom use of oral antidiabetics to insulin(0.78, 0.55–1.10). The finding for pro-gressiontoinsulinwasatrendbutwasnotstatistically significant. In addition, thefindings on diabetes progression shouldbe noted to be post hoc and are not ad- justed for multiple comparisons.Themajorlimitationofthisstudywasthe potential for unadjusted confound-ing,aprobleminanyobservationalstudy.Despite the similar indications for thefibrates, the drugs are clearly not identi-cally prescribed. Most strikingly, bezafi-brate was by far the most commonly usedfibrate in the U.K. and had the highestproportion of female users. Of mostconcern, rates of baseline diabetes dif-fered by exposure group. Adjusting forthese baseline differences did notchange our results. Further, it is reas-suring that bezafibrate still appeared tobe protective even when compared withciprofibrate, clofibrate, or gemfibrozil(which were not preferentially pre-scribed to diabetic subjects comparedwith bezafibrate). It is especially reas-suring that a post hoc analysis of diabe-tesprogressionthatcouldnothavebeenconfounded by preferential prescribingto diabetic subjects still showed a pro-tective effect from bezafibrate. We inturn note that this post hoc analysis wasnot part of the srcinal study design,was subject to false-positive results dueto multiple comparisons, and usedrough proxies for diabetes progression(progression to oral or insulin therapy).No observational study can completelyexclude the possibility of confoundingby indication, and the results of thisstudy need to be confirmed in a subse-quent randomized study. Another po-tential limitation of this study is thelikelihood that some incident cases of diabetes were not captured by the data-base; however, such misclassificationwould most likely be nondifferentialand would bias any finding toward thenull.In summary, this study strongly sup-portstheideathatbezafibratecanpreventtype 2 diabetes, confirming a post hocanalysis in a prior study. The effect sizeestimates from this study are comparableto those reported for other studies assess-ingtheuseofthiazolidinedionesandmet-formin to prevent diabetes (19). Givenconcerns about cardiovascular risk withexisting oral antidiabetic agents (20,21),bezafibrate may offer a unique opportu-nity to treat or prevent diabetes whilemaintaining a favorable cardiovascularrisk-benefit profile. However, it wouldnot be appropriate to establish a new in-dication without randomized controlledtrial data to confirm these findings, simi-lar to those recently conducted to studythe antidiabetic properties of colesevelam(22). In light of the increasing populationriskofdiabetes,atrialthatcouldestablishthe effectiveness of an inexpensive and Table 3— HRs stratified by years of cumulative use Reference groupFully adjusted HRs (95% CI) stratified by years of cumulative exposure Year 1 Years 2–3 Years 4–5 All fibrate users 0.74 (0.52–1.05) 0.62 (0.44–0.89) 0.57 (0.35–0.93) Fully adjusted HRs are adjusted for year of treatment initiation, age, sex, history of congestive heart failure,historyofmyocardialinfarction,andhistoryofuseofthiazidediuretics,loopdiuretics,  -blockers,calcium-channel blockers, ACE inhibitors, angiotensin receptor blockers, or steroids. Table 4— Cox proportional hazard models for ad hoc analysis Reference groupFully adjusted HRs (95% CI)For progression from unmedicatedbaseline diabetes to use of oralantidiabetic therapyFor progression from baseline useof oral antidiabetic therapy to useof insulin All fibrate users 0.54 (0.38–0.76) 0.78 (0.55–1.10)Ciprofibrate users 0.44 (0.28–0.69) 0.78 (0.50–1.22)Fenofibrate users 0.57 (0.32–1.02) 0.86 (0.52–1.42)Gemfibrozil users 0.74 (0.38–1.43) 0.57 (0.31–1.05)  All models treat bezafibrate as the exposure; reference group varies by row. Clofibrate was not used alone asa reference group because of an insufficient number of observations in the clofibrate group to supportmultivariable modeling. Fully adjusted models are adjusted for year of treatment initiation, age, sex, historyof congestive heart failure, history of stroke, and history of drug use (ACE/angiotensin receptor blocker,calcium channel blocker, loop diuretic, thiazide diuretic,   -blocker, or steroid). Bezafibrate for prevention of diabetes 550  D IABETES  C ARE ,  VOLUME  32,  NUMBER   4, A PRIL  2009  safe agent for both prophylaxis and treat-ment should be strongly considered forprioritization by funding agencies. Acknowledgments — This research was sup-ported by a National Institutes of Health T-32and Clinical and Translation Science AWARDS (CTSA) grant, a CTSA-AutomatedClaims and Medical Record Databases intra-mural grant, and pooled pharmacoepidemiol-ogy training money that includes industrysupport. J.H.F. also is grateful for the supportof the Center for Clinical Epidemiology andBiostatistics and the University of Pennsylva-nia CERT (Center for Education and Researchon Therapeutics). J.H.F. received tuition support from phar-macoepidemiology training funds, which in-clude funding from multiple companies thatmanufacture fibrates. S.E. is involved in a fi-nancial, research, or advisory capacity with anumber of companies that manufacture fi-brates. P.O.S. is involved in a financial, re-search, or advisory capacity with a number of companies that manufacture fibrates. S.H. isacoinvestigatoronastudysponsoredbyAb-bott Laboratories and has served as a con-sultant to Wyeth and sanofi-aventis. Noother potential conflicts of interest relevantto this article were reported.Parts of this study were presented in ab-stract form at the 110th annual meeting of the American Society for Clinical Pharmacologyand Therapeutics, Washington, DC, 18–21March 2009. References 1. ZimmetP,AlbertiKG,ShawJ:Globalandsocietal implications of the diabetes epi-demic.  Nature  414:782–787, 20012. Hansen MK, Connolly TM: Nuclear re-ceptors as drug targets in obesity, dyslip-idemia and atherosclerosis.  Curr OpinInvestig Drugs  9:247–255, 20083. Brophy JM: Selling safety—lessons frommuraglitazar.  JAMA  294:2633–2635, 20054. Narkar VA, Downes M, Yu RT, Embler E, WangYX,BanayoE,MihaylovaMM,Nel-sonMC,ZouY,JuguilonH,KangH,ShawRJ, Evans RM: AMPK and PPAR   agonistsare exercise mimetics.  Cell  134:405–415,20085. Tenenbaum A, Motro M, Fisman EZ, Adler Y, Shemesh J, Tanne D, Leor J,Boyko V, Schwammenthal E, Behar S: Ef-fect of bezafibrate on incidence of type 2diabetes mellitus in obese patients.  Eur Heart J   26:2032–2038, 20056. Tenenbaum A, Motro M, Fisman EZ,Schwammenthal E, Adler Y, GoldenbergI, Leor J, Boyko V, Mandelzweig L, BeharS: Peroxisome proliferator-activated re-ceptorligandbezafibrateforpreventionof type 2 diabetes mellitus in patients withcoronary artery disease.  Circulation  109:2197–2202, 20047. Tenenbaum A, Fisman EZ, Boyko V,Benderly M, Tanne D, Haim M, Matas Z,MotroM,BeharS:Attenuationofprogres-sion of insulin resistance in patients withcoronary artery disease by bezafibrate.  Arch Intern Med  166:737–741, 20068. Keech AC, Mitchell P, Summanen PA,O’Day J, Davis TM, Moffitt MS, TaskinenMR, Simes RJ, Tse D, Williamson E, Mer-rifield A, Laatikainen LT, d’Emden MC,Crimet DC, O’Connell RL, Colman PG,FIELD Study Investigators: Effect of feno-fibrate on the need for laser treatment fordiabeticretinopathy(FIELDstudy):aran-domised controlled trial.  Lancet  370:1687–1697, 20079. Whitelaw DC, Smith JM, Nattrass M: Ef-fects of gemfibrozil on insulin resistanceto fat metabolism in subjects with type 2diabetes and hypertriglyceridaemia.  Dia-betes Obes Metab  4:187–194, 200410. Ohrvall M, Lithell H, Johansson J, VessbyB: A comparison between the effects of gemfibrozil and simvastatin on insulinsensitivity in patients with non-insulin-dependent diabetes mellitus and hyperli-poproteinemia.  Metabolism  44:212–217,199511. Vega GL, Ma PT, Cater NB, Filipchuk N,Meguro S, Garcia-Garcia AB, Grundy SM:Effects of adding fenofibrate (200 mg/ day) to simvastatin (10 mg/day) in pa-tients with combined hyperlipidemia andmetabolic syndrome.  Am J Cardiol  91:956–960, 200312. The FIELD Study Investigators: Effects of long-term fenofibrate therapy on cardio-vascular events in 9795 people with type2 diabetes mellitus (the FIELD study):randomised controlled trial.  Lancet  366:1849–1861, 200513. Raslova´ K, Nagyova´ A, Dobia´sova´ M, Pta´ckova´ K, Dusinska´ M: Effect of cipro-fibrateonlipoproteinmetabolismandox-idative stress parameters in patients withtype 2 diabetes mellitus and atherogeniclipoprotein phenotype.  Acta Diabetol  37:131–134, 200014. Herna´ndez-MijaresA,LluchI,VizcarraE,Martínez-Triguero ML, Ascaso JF, Car-mena R: Ciprofibrate effects on carbohy-drate and lipid metabolism in type 2diabetes mellitus subjects.  Nutr MetabCardiovasc Dis  10:1–6, 200015. Willson TM, Brown PJ, Sternbach DD,Henke BR: The PPARs: from orphan re-ceptorstodrugdiscovery.  J Med Chem 43:527–550, 200016. LewisJD,BrensingerC,BilkerWB,StromBL: Validity and completeness of theGeneral Practice Research Database forstudies of inflammatory bowel disease. Pharmacoepidemiol Drug Saf   11:211–218,200217. General Practice Research Database [on-line], 2002. Available from Accessed 7 July200818. YangCC,JickSS,TestaMA:Whoreceiveslipid-lowering drugs: the effects of co-morbidities and patient characteristics ontreatment initiation.  Br J Clin Pharmacol 55:288–298, 200319. Chiasson JL: Prevention of type 2 diabe-tes: fact or fiction?  Expert Opin Pharmaco-ther   8:3147–3158, 200720. Nissen SE, Wolski K: Effect of rosiglita-zone on the risk of myocardial infarctionand death from cardiovascular causes. N Engl J Med  356:2457–2471, 200721. GoldfineAB:Assessingthecardiovascularsafety of diabetes therapies.  N Engl J Med 359:1092–1095, 200822. Fonseca VA, Rosenstock J, Wang AC,Truitt KE, Jones MR: Colesevelam HClimproves glycemic control and reducesLDL cholesterol in patients with inade-quately controlled type 2 diabetes on sul-fonylurea-based therapy.  Diabetes Care 31:1479–1484, 2008 Flory and Associates D IABETES  C ARE ,  VOLUME  32,  NUMBER   4, A PRIL  2009  551
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!