R e v i e w s / C o m m e n t a r i e s / A D A S t a t e m e n t s
A Systematic Review of Drug Therapy to
Delay or Prevent Type 2 Diabetes

potentially limiting factors to widespread UMIT R. MAJUMDAR, MD
type 2 diabetes may be an important ther-apeutic modality in those patients in OBJECTIVE — To systematically review the evidence for the prevention of type 2 diabetes by
whom lifestyle interventions fail, are not sufficiently potent, or are not feasible. Anumber of different drug classes have RESEARCH DESIGN AND METHODS — Randomized controlled trials and cohort
studies examining the effect of oral hypoglycemic agents, antiobesity agents, antihypertensive portant distinction is whether such agents agents, statins, fibrates, and estrogen on the incidence of type 2 diabetes were identified from MEDLINE, EMBASE, the Cochrane Controlled Trials Registry, and searches of reference lists.
Two reviewers independently assessed studies for inclusion and performed data extraction.
term delay in the biochemical diagnosis of — Ten studies of oral hypoglycemic agents and 15 studies of nonoral hypoglycemic agents were found. Oral hypoglycemic agents and orlistat are the only drugs that have been studied in randomized controlled trials with diabetes incidence as the primary end point. In the are sustained, cost-effective, and free of largest studies of 2.5– 4.0 years’ duration, metformin (relative risk [RR] 0.69, 95% CI 0.57– 0.83), acarbose (0.75, 0.63– 0.90), troglitazone (0.45, 0.25– 0.83), and orlistat (hazard ratio this systematic review to evaluate the cur- [HR] 0.63, 95% CI 0.46 – 0.86) have all been shown to decrease diabetes incidence compared rent evidence for the prevention of type 2 with placebo; however, follow-up rates varied from 43 to 96%. Current evidence for statins, diabetes by pharmacological therapies.
fibrates, antihypertensive agents, and estrogen is inconclusive. In addition, the critical questionof whether drugs are preventing, or simply delaying, onset of diabetes remains unresolved.

— Currently, no single agent can be definitively recommended for diabetes prevention. Future studies should be designed with diabetes incidence as the primary outcome and should be of sufficient duration to differentiate between genuine diabetes prevention as controlled trials (RCTs) and cohort stud- opposed to simple delay or masking of this condition.
ies examining the effects of drug therapyon the subsequent incidence of type 2 di- Diabetes Care 28:736 –744, 2005
abetes. We searched the Cochrane Con-trolled Trials Registry (first quarter,2004), MEDLINE (1966 to June, week 3, Diabetes currently affects an esti- sive lifestyle interventions, primarily in 2004),andEMBASE(1980toweek26,
2004). Reference lists of original studies (IGT), may decrease the incidence of type diabetes is the fifth leading cause of death $132 billion in direct and indirect costs in beneficial effects on the entire cardiovas- the number of global cases of diabetes by cular risk profile as well as noncardiovas- ported, or provided sufficient data to cal- 2030 (1), the development of effective di- cular benefits related to weight loss and culate, type 2 diabetes incidence using an intention-to-treat analysis. In studies with multiple interventions, only the results of and feasibility in a nontrial setting remain drug intervention arms compared with aplacebo or control group were included.
● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● From the 1Department of Medicine, University of Alberta, Edmonton, Canada; the 2Institute of Health Economics, Edmonton, Canada; and the 3Department of Public Health Sciences, University of Alberta, Address correspondence and reprint requests to Raj Padwal, Department of Medicine, 2E3.22 Walter C.
Mackenzie HSC, University of Alberta Hospital, 8440-112th St., Edmonton, AB, Canada, T6G 2B7. E-mail:[email protected]
Received for publication 13 October 2004 and accepted in revised form 12 December 2004.
search, a specific search for the following Abbreviations: DPP, Diabetes Prevention Program; IGT, impaired glucose tolerance; RCT, randomized
controlled trial; STOP-NIDDM, Study To Prevent Noninsulin-Dependent Diabetes Mellitus.
Additional information for this article can be found in an online appendix at http://care.diabetesjournals.
2005 by the American Diabetes Association.
DIABETES CARE, VOLUME 28, NUMBER 3, MARCH 2005 Padwal and Associates
(hereafter, referred to as “statins”), estro- 0.69, 95% CI 0.57– 0.83); metformin was Thiazolidinediones. While the Trogli-
blockers, diuretics, and ␣-blockers. Arti- yses, the benefits of metformin were pri- (RR 0.45, 95% CI 0.25– 0.83) with trogli- cles were excluded if the intervention was marily observed in patients Ͻ60 years of tested in patients with preexisting diabe- age (RR 0.66, 95% CI 0.40 – 0.79 for pa- tes, the sample size was Ͻ50 patients, the tients 25– 44 years old) and patients with kg [95% CI 0.8 –1.4]), the nearly 33% at- citation was a review or duplicate article, a BMI Ն35 kg/m2 (0.47, 0.35– 0.63).
view of antihypertensive drugs and type 2 half of eligible patients, with one patient abetes was ascertained (12). In the 79% of studies published after August 2003 (11).
dependently examined abstracts of the re- found a significant reduction in diabetes in the placebo group. When results of the and performed data extraction. Cohen’s ␬ coefficients were calculated to assess in- overall analysis, metformin still signifi- terobserver agreement for study inclusion found no significant reduction in the in- 95% CI 0.46 – 0.86) and weight by 2.8 kg (95% CI 1.1– 4.5) compared with placebo in the Xenical in the Prevention of Diabe- differences in end point definitions, pa- (13–17). All of these studies had very low tient populations, and interventions.
the attrition rate was 57% (Table 2) (23).
Ninety-one percent of orlistat-treated pa- RESULTS — Of the 5,511 initial cita-
Acarbose. Acarbose was studied in one
tients experienced gastrointestinal side ef- fects in the first year of therapy compared [available at http://care.diabetesjournals .org]). Of these, 4,247 citations involved rolling 642 obese patients reported a non- significant reduction in the incidence of were excluded. After screening the titles group (RR 0.75, 95% CI 0.63– 0.90) dur- ing 39 months of observation (17). Nearly istat therapy (RR 0.25, 95% CI 0.05–1.2) tions, 36 full-text articles were reviewed and 10 articles met inclusion criteria. An low absolute incidence of diabetes within induced gastrointestinal toxicity. At study these trials, and attrition rates averaged the reference lists of all included reports.
A recently published systematic review of with 10.5% of placebo-treated patients.
In a secondary analysis, acarbose reduced drugs found that diabetes incidence is un- changed or increased by thiazide diuretics agents on diabetes incidence (Table 1).
Sulfonylureas. Two studies examined
Biguanides. The largest and most meth-
the effect of tolbutamide therapy on dia- blockers (11). Six placebo-controlled tri- odologically rigorous trial was the Diabe- als were included in this review. Thiazide high-normal/elevated fasting glucose lev- diuretic– based treatment regimens were els (19,20). Neither study reported a sta- associated with non–statistically signifi- to metformin or placebo (4). After a mean tistically significant decrease in the type 2 cant increases in the incidence of type 2 diabetes from 7.5 to 8.6% in the Systolic treated with metformin (relative risk [RR] DIABETES CARE, VOLUME 28, NUMBER 3, MARCH 2005 Drug therapy and type 2 diabetes
Table 1—Studies of oral hypoglycemic agents to reduce type 2 diabetes incidence
2%). Only five cases oftype 2 diabetes in theplacebo group.
a 2-h post-OGTT levelof Ͼ11.1 mmol/l at year5 or symptoms/signs mmol/l or a 2-h OGTTlevel of Ն11.1 mmol/l be 3 SDs above the meanto diagnose diabetes.
a 2-h post-OGTT levelof Ͼ11.1 mmol/l plussymptoms/signs 11.6%) Acarbose 150mg (83; 2%) vs. control(83; 11.6%) *Excluding patients with type 2 diabetes at baseline. †Refers to the percentage of patients with complete follow-up. ‡RR and CI, calculated from the data presentedusing intention-to-treat analysis. FPG, fasting plasma glucose; OGTT, oral glucose tolerance test.
Angiotensin receptor blocker therapy sig- the Elderly (EWHPE) trial (1.5, 0.85–1.6) 5.4 to 3.6% (0.66, 0.51– 0.85) and from nificantly decreased diabetes incidence in 22 to 6% in a small group of patients with heart failure (0.26, 0.13– 0.53) (27,28).
DIABETES CARE, VOLUME 28, NUMBER 3, MARCH 2005 Padwal and Associates
Table 2—Studies of other agents and type 2 diabetes incidence
XENDOS (Sweden) (23) 3,305 obese patients Repeated or confirmed bya whole-blood FPG Ն6.7mmol/l cholorthiazide weresecond-line agents.
14,573 patients at high Physician reported or new Ն11.1 mmol/l with clinicalevidence of diabetes DIABETES CARE, VOLUME 28, NUMBER 3, MARCH 2005 Drug therapy and type 2 diabetes
Table 2—Continued
Current ERT users (132) vs. OR 1.1 (0.6–2.0) Hammond et al. (U.S.) 582 estrogen-deficient with at least 5years follow-upwere included) 1,006 postmenopausal FPG Ն7.8 mmol/l or 2-h *Excluding patients with DM2 at baseline; †refers to the percentage of patients with complete follow-up; ‡RR and CIs calculated from the data presented usingintention-to-treat analysis. OGTT, oral glucose tolerance test; FPG, fasting plasma glucose; CAD, coronary artery disease; SES, socioecononmic status; FH, familyhistory; RPG, random plasma glucose; ERT, estrogen replacement therapy; HTN, hypertension.
(0.78, 0.64 – 0.96) and resulted in a non- In a post hoc analysis of 303 patients with significant decrease in diabetes incidence tion and Prognosis in the elderly (SCOPE) was associated with a reduction in diabe- study, and there was insufficient evidence controlled statin trials reported conflict- ing results regarding the effect of statin risk of developing type 2 diabetes (11).
therapy on diabetes incidence (Table 2).
gen use and diabetes incidence (Table 2).
disease enrolled in the International Vera- 0.50 – 0.99) (33). In the Heart Protection with a significant reduction in the inci- the placebo arm (1.15, 0.99 –1.34) and in men (RR 0.85, 95% CI 0.77– 0.95) (Table statin in Ischemic Disease (LIPID) study, 2) (31). Diabetes incidence was not a pre- therapies, which could potentially affect never users (RR 0.82, 95% CI 0.70 – 0.96) diabetes incidence. In the Valsartan Anti- (39). Diabetes incidence in former estro- gen users was not significantly different from never users (1.07, 0.95–1.2). Of the tients at high cardiovascular risk, a valsar- remaining four cohort studies (40 – 43), only one reported a significant covariate- DIABETES CARE, VOLUME 28, NUMBER 3, MARCH 2005 Padwal and Associates
type 2 diabetes was 3% lower in the life- pared with nonusers. Several trials failed tality in order to accept that any of these to adjust for potentially important covari- drugs are beneficial in patients who have ates such as family history, weight, or base- lifestyle modification was efficacious re- gardless of age, sex, BMI, or ethnic back- CONCLUSIONS — In summary, a
lifestyle interventions can be successfully pact of different drugs on diabetes inci- equally effective form outside of a clinical trial setting, recidivism remains a major verse effects associated with drug-related agents, antiobesity drugs, statins, fibrates, diabetes prevention strategies. For exam- estrogen, and antihypertensive drugs.
ber of participants achieving weight loss ple, hypoglycemia is a potentially limiting side effect of sulfonylurea therapy, occur- ring at a frequency of 3% in patients with mia Study (47). Gastrointestinal toxicity contributed to high discontinuation rates decreases in diabetes incidence with met- ies, and troglitazone is no longer available istat; however, high attrition rates were because of the risk of serious hepatotox- found in trials of the latter two agents.
icity (48). Given the likelihood of long- Evidence for statins, estrogen, and antihy- pertensive agents is conflicting and is lim- tive effect. Drugs that acutely lower serum agents, additional data regarding adverse ited to cohort studies and secondary post events and adherence will be required.
concentrations to a lower cutoff level than that required for the formal diagnosis of atic review (including ours) is the possi- diabetes. In the posttrial washout periods studies reporting diabetes incidence as a higher incidence of diabetes in the treat- secondary or post hoc end point were dif- effectiveness analysis from a societal per- spective, the metformin intervention cost trial follow-up periods were longer. In the adjusted life year gained over the 3-year TRIPOD study, diabetes incidence, ␤-cell phies yielded more valid studies for inclu- function, and insulin sensitivity remained sion than our original search strategy.
stable in the troglitazone arm for at least 8 Regardless, while the possibility of miss- months after drug discontinuation (21).
$35,000, respectively. In all analyses, the ing trials reporting secondary or post hoc However, the type 2 diabetes incidence in analyses exists, we feel that it is unlikely the posttrial period was very low, and rea- cally attractive than metformin. A second that any definitive studies were missed.
also factored in estimates of cost savings els achieved by oral hypoglycemic agents, for each case of diabetes presumably pre- it is likely that drug-induced weight loss is contributing to the observed reduction in onstrate arrest of the disease process.
Because the average time interval between onset of ␤-cell dysfunction and develop- ment of type 2 diabetes is 10 years (45), or to indirectly promote weight loss.
follow-up periods will have to be substan- dence (Table 3) (47,51–56). Short-acting tially longer. In a recent retrospective co- lifestyle modification interventions in the inhibitors, newer thiazolidinediones, and diabetes prevention trials (9,10). In con- trast to drug therapy, intensive lifestyle classes being investigated. The majority of tion in the latter half of this decade.
three largest studies to date, despite mod- est degrees of weight loss of ϳ5 kg or less have investigated the effects of several dif- their requirements for insulin therapy.
ferent drug classes on type 2 diabetes in- DIABETES CARE, VOLUME 28, NUMBER 3, MARCH 2005 Drug therapy and type 2 diabetes
Table 3—Ongoing and future RCTs of drug therapy to prevent type 2 diabetes
Patients at risk of type 2 diabetes.
*Six-year results for both studies have been published only as abstracts. Primary analyses showed no significant difference between groups.
3. Tuomilehto J, Lindstrom J, Eriksson JG, Khunti KK: Prevention of Type 2 diabetes mellitus: a review of the evidence and its application in a UK setting. Diabet Med cebo, while the evidence for orlistat, st- Uusitupa M: Prevention of type 2 diabetes 11. Padwal R, Laupacis A: Antihypertensive therapy and incidence of type 2 diabetes: with impaired glucose tolerance. N Engl a systematic review. Diabetes Care 27: are not definitive and no single agent can prevention. It is critical that future studies of type 2 diabetes with lifestyle interven- tion or metformin. N Engl J Med 346:393– tes in the Diabetes Prevention Program.
onset diabetes as the primary outcome, so as to differentiate between genuine diabe- 13. Yang W, Lin L, Qi J: The preventive effect tes prevention as opposed to simple delay study. Chin J Endocrinol Metab 17:131– Acknowledgments — S.R.M. and F.A.M. are
pertension. Can J Cardiol 20:55–59, 2004 Population Health Investigators of the Alberta 6. Goldstein DJ: Beneficial effects of modest 14. Li CL, Pan CY, Lu JM, Zhu Y, Wang JH, weight loss. Int J Obes Relat Metab Disord vestigators of the Canadian Institutes of Health Research (CIHR). F.A.M. holds the University paired glucose tolerance. Diabet Med 16: of Alberta Merck Frosst/Aventis Chair in Pa- tient Health Management. J.A.J. is a Health 15. Fontbonne A, Charles MA, Juhan-Vague I search Chair in Diabetes Health Outcomes.
Review). In The Cochrane Library. Issue 2.
Lee for his assistance in article translation.
upper-body fat distribution. Diabetes Care References
16. Jarrett RJ, Keen H, Fuller JH, McCartney 1. Wild S, Roglic G, Green A, Sicree R, King and control. Ann Intern Med 119:764 –770, impaired glucose tolerance (‘borderline diabetes’). Diabetologia 16:25–30, 1979 for 2030. Diabetes Care 27:1047–1053, tional Institute of Diabetes, Digestive and 17. Chiasson JL, Josse RG, Gomis R, Hanefeld M, Karasik A, Laakso M: Acarbose for pre- 2. American Diabetes Association: Economic of type 2 diabetes. Diabetes Care 25:742– costs of diabetes in the U.S. in 2002. Dia- STOP-NIDDM randomized trial. Lancet DIABETES CARE, VOLUME 28, NUMBER 3, MARCH 2005 Padwal and Associates
18. Chiasson JL, Josse RG, Gomis R, Hanefeld NIDDM trial. JAMA 290:486 – 494, 2003 19. Sartor G, Schersten B, Carlstrom S, Me- tients with chronic heart failure. Circula- cose tolerance: prevention of diabetes by tolbutamide and diet regulation. Diabetes Olofsson B, Ostergren J, Yusuf S: Effects randomised controlled trial. Lancet 361: 20. Keen H, Jarrett RJ, Ward JD, Fuller JH: ity in patients with chronic heart failure: Borderline diabetics and their response to the CHARM-Overall programme. Lancet tolbutamide. Adv Metab Disord 2 (Suppl.
30. Lithell H, Hansson L, Skoog I, Elmfeldt D, S: Peroxisome proliferators-activated re- 21. Buchanan TA, Xiang AH, Peters RK, Kjos of type 2 diabetes mellitus in patients with Prognosis in the Elderly (SCOPE): princi- coronary artery disease. Circulation 109: ervation of pancreatic B-cell function and intervention trial. J Hypertens 21:875– cological treatment of insulin resistance in high-risk Hispanic women. Diabetes 51: 31. Pepine CJ, Handberg EM, Cooper-DeHoff 22. Durbin RJ: Thiazolidinedione therapy in study. Ann Intern Med 138:1–9, 1992 39. Manson JE, Rimm EB, Colditz GA, Willett ance and insulin resistance. Diabetes Obes onist hypertension treatment strategy for 23. Torgerson JS, Hauptman J, Boldrin MN, study: a randomized study of orlistat as an mellitus. Ann Epidemiol 2:665– 673, 1992 adjunct to lifestyle changes for the pre- 32. Julius S, Kjeldsen SE, Weber M, Brunner 40. Rossi R, Origliani G, Modena MG: Trans- tients. Diabetes Care 27:155–161, 2004 Laragh J, McInnes GT, Mitchell L, Plat F, ing JP, Sjostrom L: Effects of weight loss women. Diabetes Care 27:645– 649, 2004 adults. Arch Intern Med 160:1321–1326, domized trial. Lancet 363:2022–2031, The effect of estrogen use on levels of glu- 25. Savage PH, Pressel SL, Curb D, Schron 33. Freeman DJ, Norrie J, Sattar N, Neely DG, pausal women. Diabetes Care 25:500 –504, field JL, Stamler J: Influence of long-term, dence for a protective treatment effect in term estrogen replacement therapy. I. Met- sive therapy on glucose, lipid, uric acid, abolic effects. Am J Obstet Gynecol 133: Study. Circulation 103:357–362, 2001 sion. Arch Intern Med 158:741–751, 1998 26. Fletcher A, Amery A, Birkenhager W, Bul- estrogen therapy on the risk of non-insu- lin-dependent diabetes mellitus. Am J ized placebo-controlled trial. Lancet 361: 35. Keech A, Colquhoun D, Best J, Kirby A, P, Nissinen A, O’Brien E, O’Malley K, Pelemans W, Petrie J, Staessen J, Terzoli L, Thijs L, Tuomilehto J, Webster J, Wil- BV: The Da Qing IGT, Diabetes Study.
liams B: Risks and benefits in the trial of Effects of diet and exercise in preventing Blood Pressure in the Elderly. J Hypertens glucose. Diabetes Care 26:2713–2721, tolerance. Diabetes Care 20:537–544, 1997 27. Yusuf S, Gerstein H, Hoogwerf B, Pogue J, 36. Sever PS, Dahlof B, Poulter NR, Wedel DIABETES CARE, VOLUME 28, NUMBER 3, MARCH 2005 Drug therapy and type 2 diabetes
view of 6 years’ therapy of type II diabetes.
54. Navigator press release: the NAVIGATOR 46. Yee A, Majumdar SR, Simpson SH, McAli- study [article online], 2001. Available from use in type 2 diabetes mellitus is associ- ated with a delay in starting insulin. Dia- Switzerland, and the Untied Kingdom.
47. Herlihy OM, Morris RJ, Karunakaran S, challenges in improving prognosis. Am J years does not delay progression to diabe- Cardiol 89 (Suppl.):18A–26A, 2002 tes (Abstract). Diabetologia 43 (Suppl. 1): 56. Aventis press release: Aventis announces vention Trial (Abstract). Diabet Med 20 five-year ORIGIN trial to investigate reduc- 48. Kohlroser J, Mathai J, Reichheld J, Banner troglitazone: report of two cases and re- insulin [article online], 2004. Available from
tion. Am J Gastroenterol 95:272–276, 2000 subjects with impaired fasting glucose.
Diabetes Obes Metab 3:443– 451, 2001 57. Canadian Diabetes Association Website: search Group: Within-trial cost-effective- 53. The Oxford Center for Diabetes Endocri- ness of lifestyle intervention or metformin Evaluation trial [article online], 2004.
for the primary prevention of type 2 dia- view [article online], 2002. Available from betes. Diabetes Care 26:2518 –2523, 2003 Section_Professionals/canoe_trial.asp.
50. Palmer AJ, Roze S, Valentine WJ, Spinas DIABETES CARE, VOLUME 28, NUMBER 3, MARCH 2005



A Prospective, Randomized Pilot Evaluation of Topical Triple Antibiotic Versus Mupirocin for the Prevention of Uncomplicated Soft Tissue Wound Infection ROBERT HOOD, MD,* KENNETH M. SHERMOCK, PHARMD,† Little data exists comparing the safety and efficacy of triple antibiotic itracin zinc, and polymixin B sulfate) or placebo petrolatum ointment (TAO) and mupirocin for prevention of

La Lettre d’Oriade I Septembre 2013 Pour les infections à gonocoque : Comme la plupart des bactéries, le gonocoque a acquis des résistances à de nombreux résistances des germes, Ainsi, les fluoroquinolones n’ont plus de place dans cette indication, le gonocoque et constatez notamment étant résistant à la ciprofloxacine dans plus d’1 cas sur 3 en 2009 (contre 1 cas sur 15

Copyright © 2018 Medical Abstracts