Drperlmutter.com2

Effects of Initiating Insulin and Metformin
on Glycemic Control and Inflammatory
Biomarkers Among Patients
With Type 2 Diabetes
The LANCET Randomized Trial
Context As diabetes is in part an inflammatory condition, the initiation of insulin and/or
metformin may beneficially reduce levels of inflammatory biomarkers such as high- sensitivity C-reactive protein (hsCRP).
Objective To determine whether insulin alone or combined with metformin lowers
levels of hsCRP, IL-6, and soluble tumor necrosis factor receptor 2 (sTNFr2) in pa-
tients with recent-onset type 2 diabetes mellitus.
Design, Setting, and Participants Randomized 2ϫ2 factorial trial of open-label
insulin glargine and placebo-controlled metformin in 500 adults with type 2 diabetes (median time from diagnosis, 2.0 years), suboptimal glycemic control, and elevated hsCRP levels. Patients were recruited from US office-based practices between Octo- Intervention Random allocation to 1 of 4 treatments (placebo metformin only, pla-
cebo metformin and insulin glargine, active metformin only, or active metformin and in- sulin glargine) with dose titration targeting fasting blood glucose less than 110 mg/dL.
vented with intensive glucose control are Main Outcome Measures Change in hsCRP level (primary end point) and change
conflicting. Recent findings from 3 stud- in IL-6 and sTNFr2 levels (secondary end points) from baseline to 14 weeks.
Results Levels of glucose and glycated hemoglobin (HbA
with active treatment vs placebo (all P values Ͻ.001). Levels of hsCRP were reduced in all 4 groups. There was no significant difference in hsCRP reduction among those allo- cated to insulin (−11.8%; 95% CI, −18.7% to −4.4%) or to no insulin (−17.5%; 95% CI, −23.9% to −10.5%) (P for difference=.25), or among those allocated to active met- formin (−18.1%; 95% CI, −24.4% to −11.1%) or placebo metformin (−11.2%; 95% CI, −18.1% to −3.7%) (P for difference=.17). In the individual treatment groups, de- spite a differential impact on glucose control, reductions in hsCRP in the metformin (−16.1%; analysis4 of these trials and 2 others5-7 95% CI, −25.1% to −6.1%) and metformin plus insulin (−20.1%; 95% CI, −28.8% to −10.4%) groups were no different than reductions with placebo alone (−19.0%; 95%CI, −27.8% to −9.1%; P=.67 and .87 vs placebo, respectively). By contrast, hsCRP re- duction was attenuated with insulin alone (−2.9%, 95% CI, −13.2% to 8.6%; P=.03 vs placebo). Similar findings were noted for levels of IL-6 and sTNFr2.
Conclusion In patients with recent-onset type 2 diabetes, treatment with insulin or
metformin compared with placebo did not reduce inflammatory biomarker levels de- Trial Registration clinicaltrials.gov Identifier: NCT00366301
gested that risk reduction might be at-tained in younger patients with short du- ration of disease, lower levels of glycatedhemoglobin (HbA Author Affiliations are listed at the end of this article.
tensified glucose control, and absence of Corresponding Author: Aruna D. Pradhan, MD, MPH,
Center for Cardiovascular Disease Prevention, Brigham
and Women’s Hospital, 900 Commonwealth Ave E, Boston, MA 02215-1204 (apradhan@partners.org).
1186 JAMA, September 16, 2009—Vol 302, No. 11 (Reprinted)
2009 American Medical Association. All rights reserved.
INSULIN AND METFORMIN FOR PATIENTS WITH DIABETES yet be beneficial in a subset of diabetic pa- tients, they also invite consideration of alternate therapeutic targets. Subclinical levels of 2 alternative inflammatory bio- defects of beta-cell insufficiency and in- sulin resistance,11-13 and elevations in lev- Interventions and Clinical
Follow-up
creatinine). Eligible participants were en- sis factor receptor 2 (sTNFr2), predict in- a sulfonylurea, or use of a thiazolidine- tion and stroke,19-21 and more recent clini- lished type 1 diabetes or positive anti– control, insulin resistance, or both with participants in the form of a patient- and ment, and clinical data on potential anti- Trial Design
lis, Indiana) and instructed on their ap- tigator-initiated 2ϫ2 factorial trial of (Ն2.0 mg/L). All patients provided writ- effect of insulin) and to evaluate whether associated with a difference in hsCRP re- combination groups, the initial pill dose that exercised oversight for the clinical sistent with a 2ϫ2 factorial design, sec- 2009 American Medical Association. All rights reserved.
(Reprinted) JAMA, September 16, 2009—Vol 302, No. 11 1187
INSULIN AND METFORMIN FOR PATIENTS WITH DIABETES diately prior to routine study visits were Laboratory Analysis
Outcomes
hsCRP level from baseline to 14 weeks.
to-day variability of the assay at concen- study nurse every 3 days during the first study nurse or physician thereafter.
Study visits occurred in the clinic at 2-, (lnCRP) for statistical considerations.
red blood cells. The day-to-day variabil- points. During clinic visits, glucose dia- ity at % HbA1c values of 5.5 and 9.1 were chi 917 analyzer (Roche Diagnostics).
using 3 levels of control materials rangedfrom 3.5% to 9.6%. All blood glucosemeasurements were performed on cap- Figure 1. Study Flow Diagram
illary blood using glucometers from asingle manufacturer (Accu-Chek Ad- 610 Participants assessed for eligibility Statistical Analysis
1 Anti-GAD antibody positivity2 Age out of range 12 Ineligible antidiabetic therapy11 Eligibility lab value out of range 22 Fasting glucose ≥250 mg/dL during run-in tioning on baseline hsCRP and adjust-ing for treatment stratum were con- structed with the dependent variablebeing change in lnCRP. The means at sponding to study drug assignment. Ad-justed models included terms for base- an interaction term in the regressionmodel.Tofacilitatecomparisonofchange Numbers included in analysis had high-sensitivity C-reactive protein (hsCRP) measured at the 6-week and 14- week follow-up. GAD indicates glutamic acid decarboxylase.
1188 JAMA, September 16, 2009—Vol 302, No. 11 (Reprinted)
2009 American Medical Association. All rights reserved.
INSULIN AND METFORMIN FOR PATIENTS WITH DIABETES sidered for natural log transformation. All comes. All P values were unadjusted for had as much as a 30% independent effect.
2-tailed at the .05 significance level. All SAS version 9.2 (SAS Institute Inc, Cary, ence rate among completers of the trial.
linear regression models that adjusted for Sample Size and Interim Analyses
the baseline measure. All regression mod- events used Fisher exact testing for main Table 1. Baseline Characteristics of the Study Population
Placebo and
Metformin
Metformin
and Insulin
Placebo Alone
Glargine
Glargine
P
Total/Overall
Characteristics
Time since diabetes diagnosis, median (IQR), y 2-h postprandial SMBG, median (IQR), mg/dL Abbreviations: ACE-I, angiotensin-converting enzyme inhibitor; ARB, angiotensin-receptor blocker; CI, confidence interval; HbA1c, glycated hemoglobin; IQR, interquartile range; NSAID, nonsteroidal anti-inflammatory drug; SMBG, self-monitored blood glucose.
SI conversion factor: To convert glucose to mmol/L, multiply by 0.0555.
a P values for any difference in groups derived from the ␹2 test for categorical variables, analysis of variance for means, or Kruskal-Wallis for median values.
b Body mass index (BMI) calculated as weight in kilograms divided by height in meters squared.
2009 American Medical Association. All rights reserved.
(Reprinted) JAMA, September 16, 2009—Vol 302, No. 11 1189
INSULIN AND METFORMIN FOR PATIENTS WITH DIABETES line was −3.2 lbs (−1.4%) for placebo alone, and −1.6 lb (−0.7%) for metfor- min and insulin (all P Ͻ .04 vs base- line, and both P values nonsignificant formin vs placebo alone, P = .29; met- probable effectiveness or futility, a pre- P=.17]) while allocation to insulin and .jama.com), there were no qualitative dif- (ϩ0.2%; P=.47 vs baseline; P=.001 vs had altered baseline antidiabetic therapy was in fact 15%. The P value for the fi- (FIGURE 3).
those allocated to metformin placebo.
Sensitivity Analyses
Thirteen participants (2.6%) were not in- tility. All participants enrolled at this (2 pills twice daily) at the final visit.
visits. In sensitivity analyses using im- Treatment Effects on hsCRP Level
TABLE 2 shows the main effects of insu-
sue of potential bias related to early ter- nearly identical. There was no significant level as well as effects within each of the treatment effect of either insulin (P=.25 vs no insulin) or metformin (P=.11 vs no metformin), with a persistent interaction between treatments (P=.03). As to the possibility of bias related to early termi- nation of the trial, estimates of the main tionally adjusted for baseline HbA1c level, effects and effects in individual treatment groups were similar in the interim analy- sis population as in the total study popu- baseline in all 4 study groups. In analy- randomization (FIGURE 1). Baseline
ses of main effects, there was no signifi- trial after the interim analysis decision to curtail enrollment, there was no signifi- vided in TABLE 1.
no insulin (P=.25) or among those allo- allocated to insulin vs no insulin (P=.29) Change in Glucose and HbA1c
ortometforminvsnometformin(P=.48).
Levels and Weight
no metformin (P=.17). An interaction Adherence and Safety
(P=.048) between interventions was ob- compared with placebo (all P values sons of subgroup effects. The addition of at clinic visits. Pill adherence, defined as Ͻ.001) with the greatest reductions in insulin to placebo metformin was asso- taking at least 80% of pills, was 98.1%all glycemic parameters among those tion; −2.9% for insulin and placebo met- (FIGURE 2). HbA1c level was near nor-
forminvs−19.0%forplaceboalone(P=.03 malized in the active treatment groups.
for difference in effects). This effect was total population (PϾ.99) or within the active metformin: −20.1% for insulin and open-label insulin glargine (P=.62). Ad- P=.55).Despiteagradientinachievedgly- 1190 JAMA, September 16, 2009—Vol 302, No. 11 (Reprinted)
2009 American Medical Association. All rights reserved.
INSULIN AND METFORMIN FOR PATIENTS WITH DIABETES Figure 2. Impact of Individual Treatments on Levels of Glucose, HbA1c, and Weight
nation therapy (n = 6) and confirmedhyperglycemia more common among and treatment with nonstudy insulin.
11, P=.01) and similar among those al- located to metformin or no metformin(8 vs 5, P = .40). Weight gain of 5% or more of baseline was more commonamong those allocated to insulin com- pared with no insulin (22 vs 7, P = .01) pared with no metformin (12 vs 17,P = .35). Gastrointestinal adverse ef- End-of-study geometric mean levels of fasting self-monitored blood glucose (SMBG), 2-hour postprandial SMBG, with no metformin (21 vs 7, P = .01) and glycated hemoglobin (HbA1c) are plotted for each of the 4 individual treatment groups. Models were adjustedfor treatment stratum and conditioned on baseline values. All P values Ͻ.001 for comparisons vs placebo. Error (TABLE 3).
bars indicate 95% confidence intervals (CIs).
Table 2. Effects of Study Medications on hsCRP Level: Main Effects and Individual Treatment Groups
End of Study
Baseline
Achieved
Participants,
Geometric
Geometric
% Change From
P
Adjusted % Change
P
Intervention
Mean (95% CI)
Mean (95% CI)
Baseline (95% CI)a
From Baseline (95% CI)c
Abbreviations: CI, confidence interval; hsCRP, high-sensitivity C-reactive protein.
a Adjusted for baseline treatment stratum, conditioned on baseline hsCRP.
b P value for comparison with no intervention or placebo as appropriate.
c Additionally adjusted for baseline glycated hemoglobin level, weight, and change in weight at each time point.
d P value for the prespecified comparison of those allocated to metformin and insulin with metformin alone=.55.
2009 American Medical Association. All rights reserved.
(Reprinted) JAMA, September 16, 2009—Vol 302, No. 11 1191
INSULIN AND METFORMIN FOR PATIENTS WITH DIABETES for the main effects evaluated or in com- failed to lower the risk of incident CVD.
cemia and future vascular events. In this planation for a lack of clinical benefit.
other, as the choice of antidiabetic agent Figure 3. Impact of Individual Treatments
tant areas for further investigation.
tes are limited. The current findings are lispro insulin infusion, insulin glargine loss of 3.2 lb (−1.4%) as a result of ad- Models are adjusted for treatment stratum and baseline biomarkerlevels.hsCRPindicateshigh-sensitivityC-reactive protein; sTNFr2, soluble tumor necrosis factor recep-tor 2. Error bars indicate 95% confidence intervals.
betes, insulin appeared to have a greaterimpact than metformin on hsCRP andIL-6 reduction after 8 weeks of treat- Table 3. Occurrence of Adverse Events
Metformin
and Insulin
Metformin
and Insulin
Glargine
Glargine
Variable
hsCRP and TNF levels was observedwith intensive vs conventional insulin Abbreviation: SMBG, self-monitored blood glucose.
SI conversion factor: To convert glucose to mmol/L, multiply by 0.0555.
a Related serious adverse events were, for metformin alone, 1 case of severe diarrhea and 1 case of supraventricular tachycardia; and for placebo alone, 1 case of hyperglycemia requiring hospitalization.
1192 JAMA, September 16, 2009—Vol 302, No. 11 (Reprinted)
2009 American Medical Association. All rights reserved.
INSULIN AND METFORMIN FOR PATIENTS WITH DIABETES term effects. Unfortunately, inclusion of BS; Gail Borkowski, BA; Ellie Danielson, MIA; Geor-gina Friedenberg, MPH; Vincent Guzman; Nitsan Ha- vide insights into the complex interrela- levy, MD; Catherine Lapsley, BA; Tony Laurinaitis; Jean MacFadyen, BA; Annie Murray; Harriet Samuelson, atherosclerosis in diabetic patients. From MA; Daphne Suzin, BA, BS; Elaine Zaharris, BA; as wellas the dedicated laboratory personnel at the Clinical and Epidemiologic Research Laboratory, Boston Chil- (−7% for men, −14% for women) but sig- point trial data become available,39 these dren’s Hospital, including Gary Bradwin, BS; Shake´ Aya-nian, BS; Nathalie Bitar, BS; Susan Bradwin, CLA nificant changes in hsCRP level at 1 year (ASCP); Eileen Brennan, BS; Genevieve Briones, MPH; of follow-up in nondiabetic patients with Terry Law, BS; Anthony Ricupero, BS; and Emma Tzi- oumis, BS. Data and safety monitoring board mem-bers for the LANCET trial were as follows: Paul R. Con- cated to intensive lifestyle intervention lin, MD (VA Boston Healthcare System, Brigham and achieved the greatest benefit (−33% for Women’s Hospital, Harvard Medical School); Jose C.
Florez, MD, PhD (Massachusetts General Hospital, sure control; and, in appropriate patients, Broad Institute, Harvard Medical School); and Allison B. Goldfine, MD ( Joslin Clinic, Harvard Medical School).
Members of the data and safety monitoring board were strate no significant treatment effect, but compensated for time and effort related to data and Author Affiliations: Center for Cardiovascular Dis-
ease Prevention (Drs Pradhan, Everett, Cook, and Rid- LANCET Investigators: Listed are the site Principal In-
ker), Donald W. Reynolds Center for Cardiovascular vestigator, site Study Coordinator, and location, with Research (Drs Pradhan, Everett, Cook, and Ridker), number of participants randomized indicated in pa- Leducq Center for Molecular and Genetic Epidemiol- rentheses: C. Kent, S. Baker (50), Quincy, FL; C. Pin- ogy of Cardiovascular Disorders (Dr Ridker), Divi- ner, P. Bowers (24), Peak, SC; R. Chandra, M. Johns- sions of Cardiovascular Medicine (Drs Everett and Rid- ton (23), Lexington, MO; S. Moussa, H. Beltz (23), ker) and Preventive Medicine (Drs Pradhan, Everett, Whitehall, PA; N. Doyle, M. Dail (22), Wilson, NC; R.
Cook, and Ridker), Brigham and Women’s Hospital L. Murphy, D. Murphy, P. Goodrich (22), Humboldt, and Harvard Medical School, Boston, Massachu- TN; J. Howard, J. Brosius (17), Charlotte, NC; A. Rudo, setts; Department of Pathology (Dr Rifai), Children’sHospital Medical Center and Harvard Medical School, S. Holmes, K. Kennedy (17), Westminster, MD; G. Platt, Boston; and Division of Cardiovascular Medicine (Dr J. Q. Love, C. Thomason (15), Green Cove Springs,FL; J. Rybicki, A. Rybicki, B. Marquinhos (14), Phila- Pradhan), Boston VA Medical Center, Boston.
Author Contributions: Dr Pradhan had full access to all
delphia, PA; S. Patel, J. Severs (13), Irvington, KY; E.
of the data in the study and takes responsibility for the Ranasinghe, T. Seneviratne, D. Williams (12), ShakerHeights, OH; R. Grimball, L. Woods, B. Williamson (11), integrity of the data and the accuracy of the data analysis.
Study concept and design: Pradhan, Cook, Ridker.
Sulphur, LA; P. McLaughlin, F. McLaughlin (11), Mount Acquisition of data: Pradhan, Everett, Rifai, Ridker.
Sterling, KY; L. C. Atwood, A. Bryant (10), Indepen- Analysis and interpretation of data: Pradhan, Everett, dence, KS; R. Cole, A. Cole (10), Stuart, VA; D. Whitt,M. Hammer (10), Pickerington, OH; P. McCaffrey, D.
Cook, Ridker.
Drafting of the manuscript: Pradhan, Ridker.
Duran (8), Pueblo, CO; S. Schwartz, C. Sheets (8), San Critical revision of the manuscript for important in- Antonio, TX; M. Souza, S. Lobo, A. Carvalho (8), EastProvidence, RI; J. Bonelli, M. Saldana (7), Silver Spring, tellectual content: Pradhan, Everett, Cook, Rifai, Ridker.
Statistical analysis: Pradhan, Cook.
MD; S. Chooljian, K. Chooljian (7), Fresno, CA; H. Mari- ano, M. Espiritu, L. Protacio (7), Fresno, CA; D. Ri- Administrative, technical, or material support: Pradhan, vas, J. Lopez (7), Tampa, FL; M. Waseem, L. Hudnet, M. Grammer (7), Baltimore, MD; J. Guerrero, A. Sala- Study supervision: Pradhan, Ridker.
zar (6), Houston, TX; S. Kulback, C. Honeycutt (6), Financial Disclosures: During the period of this project,
Birmingham, AL; A. Smith, C. Bencko (6), Adams- Dr Pradhan reported receiving research support from ville, AL; B. Collins, A. Dupree, L. Bunn, B. Capps, A.
Sanofi-Aventis. Dr Ridker reported having received in- DeFreese (5), Pell City, AL; L. Handke, J. Supencheck vestigator-initiated research grant support from the (5), Lincoln, NE; J. Herrod, L. Noll, M. Duron (5), Si-erra Vista, AZ, T.-S. Lee, T. Pacos (5), Dunkirk, NY; National Heart, Lung, and Blood Institute; the Na-tional Cancer Institute; the Donald W. Reynolds Foun- B. McCracken (5), Greenville, IL; W. McGarity, J. Waller dation; the Leducq Foundation; Astra-Zeneca; Novar- (5), Decatur, GA; J. Almand, S. Singh (4), Grand Prai- tis; Merck; Abbott; Roche; and Sanofi-Aventis; having rie, TX; C. Chappel, T. Koenig (4), Kissimmee, FL; R.
received consulting fees or lecture fees from Astra- McDavid, K. Phillips, T. Robbins (4), Johnson City, TN; izable to all individuals with type 2 dia- Zeneca, Novartis, Merck, Merck-Schering Plough, Sa- S. Ross, A. Coleburn (4), Florence, SC; Z. Adefuin, A.
nofi-Aventis, ISIS, Dade-Behring, and Vascular Bio- Luciano, H. Ferrillo-Dilulio (3), Bridgeport, CT; S. Ad- genics; and being listed as a coinventor on patents held kins, M. Perry, R. Phillips (3), Weber City, VA; W.
by the Brigham and Women’s Hospital that relate to Brown, D. J. Bolesta (3), Thorndale, PA; K. Chris- the use of inflammatory biomarkers in cardiovascular tensen, T. Boomgaarden (3), Idaho Falls, ID; K. Ea- disease and diabetes. These patents have been li- ton, B. Cody (3), Dothan, AL; B. Frandsen, L. Bitz (3), censed to Siemens and Astra-Zeneca. No other dis- Port Orchard, WA; D. Gandhi, V. Gandhi (3), Hart- selle, AL; H. Mariano, G. Cruz, A. Francisco, S. Go (3), Funding/Support: LANCET was an investigator-
Anaheim, CA; T. Milko, C. Goodwin (3), Graysville, initiated trial funded by Sanofi-Aventis.
AL; M. Mudrick, H. Mudrick (3), Chester, PA; T. Rouse, Role of the Sponsor: The funder had no role in the
M. Palmer (3), Saranac, MI; J. Scheuer Jr, J. Baldwin design and conduct of the study; in the collection, (3), Camden, SC; L. Curry, H. Helman, L. Chamber- analysis, and interpretation of the data; or in the prepa- lin, K. Ford (2), South Bend, IN; J. Delgado, K. An- ration, review, or approval of the manuscript.
drus (2), Ashland, OR; M. Gandhi, M. Gandhi, R. Smith Additional Information: eTable is available at http:
(2), Woodruff, SC; M. Harless, F. C. Bell (2), Harts- ville, SC; S. Lewis, S. Lisovskiy (2), Concord, CA; J. Mo- Additional Contributions: We thank the 500 study par-
relli, M. Whitehouse (2), Stoneboro, PA; E. A. Osea, ticipants and also acknowledge the tireless efforts of A. Francisco, A. Aquino, E. Heredia (2), Harbor City, status among individuals with type 2 dia- the LANCET Coordinating Center (Brigham and Wom- CA; J. Weil, D. S. Patel, K. Davies (2), Las Vegas, NV; en’s Hospital) study staff without whom this study J. Perez, S. Revels, J. Bryant (2), Sarasota, FL; G. Port- would not have been possible, including David Bates, nay, L. Brazee (2), Billerica, MA; K. Pritchett, M. Reno, 2009 American Medical Association. All rights reserved.
(Reprinted) JAMA, September 16, 2009—Vol 302, No. 11 1193
INSULIN AND METFORMIN FOR PATIENTS WITH DIABETES M. Murray, L. Methvin (2), O’Fallon, IL; P. Punjabi, 13. Shoelson SE, Lee J, Goldfine AB. Inflammation and
27. Sinha A, Formica C, Tsalamandris C, et al. Ef-
J. Amill (2), Feasterville, PA; D. Shrestha, R. Ferguson insulin resistance. J Clin Invest. 2006;116(7):1793- fects of insulin on body composition in patients with (2), Bay City, MI; N. Tuanquin, T. Tuanquin (2), Lo- insulin-dependent and non–insulin-dependent gan, WV; P. Barrington, K. Turner (1), Snellville, GA; 14. Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker
diabetes. Diabet Med. 1996;13(1):40-46.
J. Cuellar, C. Sitton (1), Wentzville, MO; I. Fenton, P.
PM. C-reactive protein, interleukin 6, and risk of de- 28. Mathieu P, Poirier P, Pibarot P, Lemieux I, Despres
Finch (1), Vernon Hills, IL; K. Hershon, S. Goon, R.
veloping type 2 diabetes mellitus. JAMA. 2001; JP. Visceral obesity. Hypertension. 2009;53(4): Moskowitz, (1), New Hyde Park, NY; J. Katsetos, P.
Silvia (1), Milford, CT; D. Kenton, R. Zambrana (1), 15. Barzilay JI, Abraham L, Heckbert SR, et al. The re-
29. Fonseca VA, Theuma P, Mudaliar S, Leissinger
Deerfield Beach, FL; L. Kuskin, S. Gifford (1), Harris- lation of markers of inflammation to the develop- CA, Clejan S, Henry RR. Diabetes treatments have burg, PA; J. Marshall (1), Hoover, AL; M. Samson, R.
ment of glucose disorders in the elderly. Diabetes. 2001; differential effects on nontraditional cardiovascular risk factors. J Diabetes Complications. 2006;20 16. Festa A, D’Agostino R Jr, Tracy RP, Haffner SM;
Insulin Resistance Atherosclerosis Study. Elevated lev- 30. Reynolds LR, Kingsley FJ, Karounos DG, Tannock
REFERENCES
els of acute-phase proteins and plasminogen activa- LR. Differential effects of rosiglitazone and insulin tor inhibitor-1 predict the development of type 2 glargine on inflammatory markers, glycemic control, 1. Gerstein HC, Miller ME, Byington RP, et al; Action
diabetes. Diabetes. 2002;51(4):1131-1137.
and lipids in type 2 diabetes. Diabetes Res Clin Pract.
to Control Cardiovascular Risk in Diabetes Study Group.
17. Freeman DJ, Norrie J, Caslake MJ, et al; West of Scot-
Effects of intensive glucose lowering in type 2 diabetes.
land Coronary Prevention Study. C-reactive protein is an 31. Aas AM, Seljeflot I, Torjesen PA, Diep LM, Thorsby
N Engl J Med. 2008;358(24):2545-2559.
independent predictor of risk for the development of dia- PM, Birkeland KI. Blood glucose lowering by means 2. Patel A, MacMahon S, Chalmers J, et al; ADVANCE
betes in the West of Scotland Coronary Prevention Study.
of lifestyle intervention has different effects on adi- Collaborative Group. Intensive blood glucose control Diabetes. 2002;51(5):1596-1600.
pokines as compared with insulin treatment in sub- and vascular outcomes in patients with type 2 diabetes.
18. Hu FB, Meigs JB, Li TY, Rifai N, Manson JE. In-
jects with type 2 diabetes. Diabetologia. 2006; N Engl J Med. 2008;358(24):2560-2572.
flammatory markers and risk of developing type 2 3. Duckworth W, Abraira C, Moritz T, et al; VADT
diabetes in women. Diabetes. 2004;53(3):693- 32. Mao XM, Liu H, Tao XJ, Yin GP, Li Q, Wang SK.
Investigators. Glucose control and vascular complica- Independent anti-inflammatory effect of insulin in tions in veterans with type 2 diabetes. N Engl J Med.
19. Ridker PM, Cushman M, Stampfer MJ, Tracy RP,
newly diagnosed type 2 diabetes. Diabetes Metab Res Hennekens CH. Inflammation, aspirin, and the risk of 4. Ray KK, Seshasai SR, Wijesuriya S, et al. Effect of
cardiovascular disease in apparently healthy men.
33. Schaumberg DA, Glynn RJ, Jenkins AJ, et al. Effect
intensive control of glucose on cardiovascular out- N Engl J Med. 1997;336(14):973-979.
of intensive glycemic control on levels of markers of comes and death in patients with diabetes mellitus.
20. Ridker PM, Hennekens CH, Buring JE, Rifai N. C-
inflammation in type 1 diabetes mellitus in the dia- Lancet. 2009;373(9677):1765-1772.
reactive protein and other markers of inflammation betes control and complications trial. Circulation. 2005; 5. UK Prospective Diabetes Study (UKPDS) Group. In-
in the prediction of cardiovascular disease in women.
tensive blood-glucose control with sulphonylureas or N Engl J Med. 2000;342(12):836-843.
34. Caballero AE, Delgado A, Aguilar-Salinas CA, et al.
insulin compared with conventional treatment and risk 21. Ridker PM, Rifai N, Rose L, Buring JE, Cook NR.
The differential effects of metformin on markers of en- of complications in patients with type 2 diabetes Comparison of C-reactive protein and low-density li- dothelial activation and inflammation in subjects with (UKPDS 33). Lancet. 1998;352(9131):837-853.
poprotein cholesterol levels in the prediction of first impaired glucose tolerance. J Clin Endocrinol Metab.
6. UK Prospective Diabetes Study (UKPDS) Group.
cardiovascular events. N Engl J Med. 2002;347 Effect of intensive blood-glucose control with met- 35. De Jager J, Kooy A, Lehert P, et al. Effects of short-
formin on complications in overweight patients with 22. Ridker PM, Danielson E, Fonseca FA, et al; JUPITER
term treatment with metformin on markers of endo- type 2 diabetes (UKPDS 34). Lancet. 1998;352 Study Group. Rosuvastatin to prevent vascular events thelial function and inflammatory activity in type 2 in men and women with elevated C-reactive protein.
diabetes mellitus. J Intern Med. 2005;257(1): 7. Dormandy JA, Charbonnel B, Eckland DJ, et al;
N Engl J Med. 2008;359(21):2195-2207.
PROactive investigators. Secondary prevention of mac- 23. Ridker PM, Danielson E, Fonseca FA, et al; JUPI-
36. Haffner S, Temprosa M, Crandall J, et al; Diabe-
rovascular events in patients with type 2 diabetes in TER Trial Study Group. Reduction in C-reactive pro- tes Prevention Program Research Group. Intensive life- the PROactive Study. Lancet. 2005;366(9493): tein and LDL cholesterol and cardiovascular event rates style intervention or metformin on inflammation and after initiation of rosuvastatin. Lancet. 2009;373 coagulation in participants with impaired glucose 8. American Diabetes Association. Standards of medi-
tolerance. Diabetes. 2005;54(5):1566-1572.
cal care in diabetes: 2009. Diabetes Care. 2009; 24. Dandona P, Chaudhuri A, Ghanim H, Mohanty
37. Holman RR, Paul SK, Bethel MA, Matthews DR,
P. Insulin as an anti-inflammatory and antiathero- Neil HA. 10-year follow-up of intensive glucose con- 9. Kahn SE. Glucose control in type 2 diabetes. JAMA.
genic modulator. J Am Coll Cardiol. 2009;53(5) trol in type 2 diabetes. N Engl J Med. 2008;359 10. Hayden JM, Reaven PD. Cardiovascular disease
25. Chu NV, Kong AP, Kim DD, et al. Differential ef-
38. Nesto RW, Lago RM. Glucose: a biomarker in acute
in diabetes mellitus type 2. Curr Opin Lipidol. 2000; fects of metformin and troglitazone on cardiovascu- myocardial infarction ready for prime time? Circulation.
lar risk factors in patients with type 2 diabetes. Dia- 11. Pickup JC, Crook MA. Is type II diabetes mellitus
betes Care. 2002;25(3):542-549.
39. Origin Trial Investigators; Gerstein H, Yusuf S,
a disease of the innate immune system? Diabetologia.
26. Fujimoto WY, Jablonski KA, Bray GA, et al; Dia-
Riddle MC, Ryden L, Bosch J. Rationale, design, and betes Prevention Program Research Group. Body size baseline characteristics for a large international 12. Donath MY, Storling J, Maedler K, Mandrup-Poulsen
and shape changes and the risk of diabetes in the dia- trial of cardiovascular disease prevention in people T. Inflammatory mediators and islet beta-cell failure. J Mol betes prevention program. Diabetes. 2007;56(6): with dysglycemia. Am Heart J. 2008;155(1):26- 1194 JAMA, September 16, 2009—Vol 302, No. 11 (Reprinted)
2009 American Medical Association. All rights reserved.

Source: http://drperlmutter.com/wp-content/uploads/2013/07/6-Not-drug-insulin.pdf