Statins and Diabetes

Baby I'm afraid of a lot of things but I ain't scared of loving you.
- Karen O

Diabetes scares me probably more than any other disease state.  Diabetes puts patients at risk of slow healing wounds, peripheral neuropathy, increased infections, kidney failure, cardiovascular events, and is the number one cause of blindness.

My roommate has the best definition of diabetes that I've ever heard:

If you eat too much candy, your foot is going to fall off.

She has a point.

There is no way to prevent type 1 diabetes, but research into preventing type 2 diabetes is endless.  Most recently there has been evidence showing that statins increase the risk of developing diabetes.  An inquisitive NP at my rotation was asking me about a Huffington Post article in regards to the topic, so in the past few days I have attempted to dissect the evidence and assess the risk.

Attention was first called to incident diabetes with statin use when the JUPITER trial was published in NEJM in 2008.  Despite screening for cardiovascular risk factors, an estimated half of all myocardial infarctions occur in patients that are at their target LDL goal.  C-reactive protein, a marker of vascular inflammation, has been shown to be a risk factor for CV events independent of cholesterol levels.  Statins have been shown to reduce C-reactive protein levels in addition to reducing LDL.

The JUPITER trial looked at patients at a LOW risk of CV events but high C-reactive protein levels: men over 50 and women over 60 who had no history of cardiovascular disease, and LDL less than 130mg/dL, and a C-reactive protein level greater than 2.0mg/L.  Patients with diabetes, blood pressure above 190/100, low renal function, or use of anti-lipid therapy were excluded.  17,802 patients randomized to 20mg rosuvastatin or placebo were followed for a median of 1.9 years to the occurrence of the first event: MI, nonfatal stroke, hospitalization for unstable angina, arterial revascularization, or cardiovascular death.

Patients taking rosuvastatin 20mg had a decreased risk of cardiovascular events; one event prevented for every 95 patients treated over two years or 31 patients treated over four years.  However, the study also showed a statistically significant increase in the onset of diabetes in patients taking rosuvastatin.

The Buttery

The PROSPER study published in 2002 also showed a significant increase in diabetes in patients taking pravastatin.

In response, Sattar et al. conducted a meta-analysis of trials containing over 1000 patients treated with a statin for at least a year.  The analysis included 13 trials and 91,140 patients.  The statins used were atorvastatin, lovastatin, rosuvastatin, simvastatin, and pravastatin; where six of 13 trials utilized pravastatin.

Sattar et al. found a 9% increase in new onset diabetes in patients taking a statin, or one more new case of diabetes in a patient on a statin for every 225 patients treated for four years.  Increase in diabetes was seen even without the inclusion of the JUPITER trial.  The only other correlation to the development of diabetes was increasing age.

The analysis by Sattar et al. only included one trial, ASCOT-LLA, that utilized atorvastatin.  The stir in the medical community lately has been over an analysis by Waters et al. published in 2011 looking at the incidence of diabetes in IDEAL, TNT, and SPARCL which all utilized atorvastatin.

New York

I am disappointed that Waters et al. did not conduct a meta-analysis, but they found a trend to new-onset diabetes in the TNT and IDEAL trials and an increase in new-onset diabetes in SPARCL.  Patients that developed diabetes were more likely at baseline to have higher fasting glucose, BMI, white blood cell count, blood pressure total cholesterol/HDL ratio, and triglycerides.  80mg of atorvastatin was more likely to worsen glycemic control than 10 or 20mg atorvastatin.  Age differences, sex, and smoking were not associated with incident diabetes.

The real question is how this will change the way we treat patients.  Despite a slight increase in diabetes, statins have been shown to decrease the risk of cardiovascular events and mortality.  Waters tells heartwire, "Compared with their risk of a cardiovascular event, their risk of developing diabetes is paltry" and advises patients not to stop taking their statins.  Dr. Blumenthal of John Hopkins Medical Institute also comments to theheart.org that this analysis will not change his use of statins.  ALLHAT similarly showed an increase in diabetes in patients taking chlorthalidone, but these patients had improved mortality outcomes.

The available evidence shows that the benefit that patients receive from statins outweighs the risk of diabetes.  To me, these analyses show that no medication comes without risk.  Its a reminder to those that believe or believed that statins should be in our drinking water that this is a prescription medication.  There is a risk of liver dysfunction, rhabdomyolysis, and now, a risk of diabetes associated with their use.   Patients should be monitored accordingly.

I would attempt to make lame parallels between the JUPITER trial and the planet, but I know that I cannot do the planet justice.  I think I have failed my parents.  My mother builds satellites and my dad builds video servers by day, and charts the sky by night.  Check out his amazing nebula photos at:
http://astrospotter.zenfolio.com/

Hyannis 2008

References:
Ridker PM et al.  Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein.  The New England Journal of Medicine 2008;359(21):2195-207.

Shepherd J et al. Pravastatin in elderly individuals at risk of vascular disease: a randomised controlled trial.  Lancet 2002;360:1623-30.

Sattar N et al.  Statins and risk of incident diabetes: a collaborative meta-analysis of randomised statin trials.  Lancet 2010;375:735-42.

Waters DD et al.  Predictors of new-onset diabetes in patients treated with atorvastatin.  Journal of the American College of Cardiology 2011;57(14):1535-45.

Hughes S.  More data on diabetes risk with statins.  March 30, 2011; http://www.theheart.org/article/1203383.do 

Nitrate use in Heart Failure

Isosorbide dinitrate can be used with hydralazine in heart failure as an alternative to ACE-inhibitors or ARBs.  The medications act complementary to dilate the blood vessels.  Hydralazine reduces afterload by decreasing both pulmonary and systemic vascular resistance.  Its effects on vasodilation are not completely understood.  It also has some moderate inotropic effects.  It reduces renal vascular resistance, though not by as much as ACE-inhibitors.¹ 

Nitrates like isosorbide dinitrate relax blood vessels by releasing NO.  Isosorbide dinitrate improves exercise capacity in patients with heart failure^1,2.  Nitrates can inhibit vascular and myocardial remodeling³.  Nitrates must be dosed so that there is a six to eight hour period of negligible drug levels or tolerance may develop.  Hydralazine may decrease nitrate tolerance. ^1,3

The V-HeFT trial in 1986 showed decreased mortality with the use of hydralazine and isosorbide dinitrate compared to prazosin ².  The V-HeFT II trial compared isosorbide dinitrate to enalapril in patients with moderate heart failure and found decreased survival due to a higher incidence of sudden death in the isosorbide dinitrate-hydralazine arm.  There was no difference in mortality rate in African Americans taking isosorbide dinitrate-hydralazine compared to enalapril. ²   

The A-HeFT trail in 2004 compared the addition of isosorbide dinitrate with hydralazine or placebo to standard heart failure treatment in patients of African descent in NYHA stage III or IV heart failure.  The trial was terminated early due to increased survival in patients taking hydralazine with isosorbide dinitrate.  The target dosing used in this trial was 75mg hydralazine and 40mg isosorbide dinitrate three times daily for a total daily dose of 225mg hydralazine and 120mg isosorbide dinitrate.⁴

Compliance is more difficult with a regimen containing isosorbide dinitrate and hydralazine because these medications are dosed three to four times a day.  The new product, BiDil, contains both medications but is about twice as expensive and must also be taken multiple times per day.  There is also a high discontinuation rate due to incidence of headache and GI upset.³

There is a lack of trials evaluating the use of isosorbide dinitrate without hydralazine in heart failure; these medications should be used together³.   The ACC/AHA guidelines recommend to consider the addition of hydralazine with isosorbide dinitrate in African Americans.  However, hydralazine with isosorbide should not be considered before an ACE-inhibitor if the patient has no history or ACE-inhibitor intolerance or if they are tolerating ACE-inhibitor therapy.³  The HFSA guidelines recommend to consider the use of hydralazine and isosorbide dinitrate in African Americans in standard therapy, as they do not respond as well to ACE-inhibitors as white patients.  The HFSA also recommends to consider the addition of hydralazine with isosorbide dinitrate in African Americans with stage II or III heart failure and LV dysfunction even if their regimen includes an ACE-inhibitor or beta-blocker.⁵

*Note: images are not of heart failure but are of the beach

References:

1. Brunton LL, Lazo JS, Parker KL, editors.  Goodman & Gilman’s: the pharmacological basis of therapeutics.  11th ed.  New York: McGraw-Hill Companies, Inc.; 2006.
2. Elkayam U, Bitar F.  Effects of nitrates and hydralazine in heart failure: clinical evidence before the African American heart failure trial.  American Journal of Cardiology 2005;96(suppl):37i-43i.
3. Hunt SA, Abraham WT, Chin MH, Feldman AM, Francis GS, Ganiats TG et al.  ACC/AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult.  Circulation 2005;112:e154-e235.
4. Taylor AL, Ziesche S, Yancy C, Carson P,  D’Agostino R, Ferdinand K et al.  Combination of isosorbide dinitrate and hydralazine in blacks with heart failure.  The New England Journal of Medicine 2004;351(20):2049-2057.
5. Adams KF, Lindenfeld J, Arnold JMO, Baker DW, Barnard DH, Baughman KL et al.  Executive summary: HFSA 2006 comprehensive heart failure practice guideline.  Journal of Cardiac Failure 2006; 12(1):10-38.

Insulin u-500

Although insulin u-500 has been available since the 1950s, the use has sharply increased in only the past five years as a result of increasing rates of obesity and insulin resistance.

Insulin u-500 is a concentrated form of regular insulin, containing five times the units of insulin per mL as u-100.  Although it is a regular insulin, its pharmacokinetic profile may more resemble that of NPH;  a dose of u-500 may last up to 8 or even 24 hours. and there is a delay associated with the peak of action.¹  Although the total dose of u-500 may be more completely absorbed than large doses of u-100, there is a delay in absorption. ² 

There have been limited pharmacokinetic studies done.  Despite the availability of algorithms for the initiation of insulin u-500, much of the conversion may be trial and error. ¹  Pharmacokinetic reports have shown that a sharp rise in insulin concentration occurs 30 minutes after the dose is injected and peaks at 5 hours.  Insulin level will remain high even at 7 hours.  Insulin u-500 should be given 30 minutes before eating. ^2,3

Initiation of insulin u-500 is recommended for patients that are taking greater than 200 units of insulin daily and have not reached or have difficulty maintaining their goal A1c.  The total daily dose (TDD) of insulin should be added up.  If the patient’s A1c is less than 8%, 10-20% of the TDD should be subtracted. ^1,3

Insulin u-500 has both basal and prandial insulin properties.  On initiation, the TDD can be divided into three doses to be taken with breakfast, lunch and dinner although adequate glycemic control has also been achieved with the use of insulin u-500 in two divided doses.¹  If the TDD is 200-300 units, initiate a BID dosing regimen; 300-750, a TID regimen, and for doses 750-2000 consider dosing four times daily⁴.

Dividing the total daily dose²:  

BID dosing: Give 60% TDD qAM, 40% qPM

TID dosing:  Give 40-45% qAM, 30-40% at noon, and 20-30% qPM

no greater than 10% should be given at bedtime

Insulin u-500 acts more as prandial insulin in some patients, others more as basal; it is unclear if this is due to patient weight, insulin resistance, or the dose given^2,3.  The extent of peaking in insulin concentration has varied between studies.³

The patient should be followed every 2 weeks.  If over half of the blood sugar levels are 70-130, no dose change is needed.  If patient is experiencing hypoglycemic episodes, the dose can be decreased 25 units daily.  For fasting glucose levels of 130-200, the dose can be increased 25 units daily and for fasting glucose levels greater than 200, the daily dose may be increased 50 units. ¹

Insulin u-500 is associated with more weight gain than regular insulin.  For every 1% decrease in A1c, a 2kg increase in body weight can be expected³.  Despite the weight gain, patients will be a decreased risk of diabetic complications later on because of more tightly controlled blood sugar.  Weight loss should continue to be emphasized in patients taking insulin u-500.  Decreased weight will improve insulin sensitivity and may allow the patient to eventually step back down to using insulin u-100.  Weight gain with the switch to this insulin should be something assessed as part of the risks and benefits to starting insulin u-500. ^2,3

References:

1.  Ballani P et al.  Clinical experience with u-500 regular insulin in obese, markedly insulin-resistant type 2 diabetic patients.  Diabetes Care 2006; 29(11):2504-5.

2.  Daily AM et al.  Extreme insulin resistance: indications and approaches to the use of u-500 insulin in type 2 diabetes mellitus.  Curr Diabetes Rep 2011; 11:77-82.

3.  Segal AR et al.  Use of concentrated insulin human regular (u-500) for patients with diabetes.  American Journal of Health System Pharmacists 2010; 67:1526-35.

4.  Cochran E et al.  The use of u-500 in patients with extreme insulin resistance.  Diabetes Care 2005; 28(5): 1240-4.