|
||||||
|
|
|
|
|||||||||||
|
|||||||||||||
| ||||||||||||||||||||||||||||||||||||||
In December 2003, the American Association of Clinical Endocrinologists held a consensus conference for inpatient diabetes therapy, cosponsored by several organizations, including the American Diabetes Association (ADA). The conclusions of that conference have been well publicized, with strict glycemic targets suggested for hospitalized patients, whether critically ill or not.1 At the same time, the ADA published a long-awaited technical review on the topic,2 and due to the intense interest, it has initiated a grand rounds program that will be presented to more than 50 community hospitals in the fall of 2004.
|
In September, at the annual meeting of the European Association for the Study of Diabetes (EASD) in Munich, Lars Rydén, MD, and Klas Malmberg, MD, presented the results of the Diabetes Mellitus Insulin-Glucose Infusion in Acute Myocardial Infarction 2 (DIGAMI 2) study.
DIGAMI 2 was a follow-up to the first DIGAMI trial, which reported the effects of a glucose-insulin-potassium (GIK) infusion when patients with type 2 diabetes or an admission blood glucose above 198 mg/dl were admitted for suspected myocardial infarction.3
The results of the first study were dramatic. After 1 year, mortality was reduced in those who received the GIK infusion from 28% to 19%.3 The survival benefit was maintained after long-term follow-up: After an average of 3.4 years, mortality was reduced from 44% to 33%.4
Given the results of the first DIGAMI study and all of the recent clinical trials of critically ill patients with type 2 diabetes, DIGAMI 2 is perhaps the most surprising study of all (for news story, see page 1).
DIGAMI 2 STUDY DESIGN
Patients with suspected acute myocardial infarction were randomized to one of three groups: (1) acute insulin-glucose infusion followed by insulin-based long-term glucose control; (2) insulin-glucose infusion followed by standard glucose control (i.e., no insulin); or (3) routine metabolic management according to local practice.
The primary endpoint was all-cause mortality between groups 1 and 2 according to intention-to-treat analysis, while mortality differences between groups 2 and 3 and morbidity differences served as secondary and tertiary endpoints, respectively.
What are the major post-hospitalization treatment differences between the two DIGAMI studies? Although the definition of "intensive insulin therapy" varies, many have defined it as those individuals receiving three or four injections daily (i.e., basal-bolus therapy where frequent premeal capillary glucose measurements can assist in decisions for prandial and correction insulin replacement).5
In the first DIGAMI study, all subjects in the GIK group were asked to administer four daily injections, but specific data for numbers of people taking three or more injections are not provided. In DIGAMI 2, only 42% of subjects in group 1 received three or more daily injections, compared to 15% and 13% in groups 2 and 3, respectively. Perhaps then, it should not be surprising that in DIGAMI 2, endpoint glycated hemoglobin (A1C) was similar among all three groups at 6.8%. Furthermore, the primary treatment target of fasting blood glucose of 90–126 mg/dl for those in group 1 was never achieved, as fasting glucose levels were similar among groups at approximately 149 mg/dl.
Therefore, it should not be of any surprise, given both the inpatient and outpatient glycemic data, that mortality between groups 1 (23.4%) and 2 (22.6%) did not significantly differ. Similarly, there were no differences between groups 2 (22.6%) and 3 (19.3%). Importantly, the total study mortality was 18.4% compared to a predicted mortality of 22.2%. Furthermore, compared to a similar but nondiabetic population, mortality in the group with diabetes did not differ.
INTERPRETATION CONSIDERATIONS
How does one interpret these data, especially in the context of all of the recent trials reporting that both insulin infusions and improvements in glycemic control improve outcomes, including mortality?
While I have several thoughts on the answer, I want to first emphasize that this was an extremely difficult study to perform. The investigators noted that the study was underfunded, and it is not clear how that may have affected the requisite number of study coordinators, home blood-glucose monitoring equipment, or even expert diabetologists most familiar with outpatient insulin techniques. These likely were among the reasons that there were no differences in glycemic control by the end of the study period.
Importantly, the investigators had an even more difficult time showing differences in glycemia, given how well-controlled these individuals were at the time of study entry. Even though the numbers are not large, subsequent analysis to see if those with lower A1C levels in group 1 have different outcomes than those with higher levels in group 2 will be interesting.
"Protocol violations" are problematic with all studies, but deserve special comment with DIGAMI 2. The fact that 14% of patients in group 3 received an insulin infusion is quite concerning. That 41% in this group received any insulin during the inpatient period needs further clarification, as we were not told what if any glycemic thresholds were used to trigger insulin therapy. Even more surprising was that only 42% of group 1 patients received the admittedly arbitrary definition of "intensive insulin therapy." Coupling that with the fact that 15% and 13% of patients in groups 2 and 3 received this same therapy, I am not at all surprised by the final study results.
New clinical trials continue to support the fact that postprandial hyperglycemia is a major determinant of atherosclerosis progression, perhaps by reducing proinflammatory cytokines.6 Furthermore, a growing literature suggests that increases in glycemic variation cause an increase in oxidative stress, further promoting diabetes complications.7 The only way DIGAMI 2 investigators could target postprandial hyperglycemia and glycemic variability would be to inject prandial insulin.
A1C, while the major glycemic outcome in most diabetes studies, is a very crude marker for hyperglycemia, as it does not in and of itself differentiate glycemic variability or the contribution of fasting compared with postprandial hyperglycemia. However, a recent report noted that in 290 patients with type 2 diabetes, the lower the A1C, the greater the contribution to postprandial hyperglycemia. For those with A1C values below 7.3% (similar to group 1 in the DIGAMI 2 trial), approximately 70% of the A1C value is due to postprandial excursions. If this is the case, it is interesting to speculate how the A1C differences, let alone the cardiovascular outcomes, would have been different if at least some emphasis had been placed on postprandial hyperglycemia. As noted by Esposito and colleagues, attention to postprandial hyperglycemia may trump A1C, especially in well-controlled diabetes.7
TAKE-HOME MESSAGES
Perhaps the most important message from DIGAMI 2 is that in well-controlled diabetes, cardiovascular outcomes are similar to those in a nondiabetic population.
A surprising point (for me anyway) is that metformin provided no cardiovascular benefit. As expected, there were significant improvements in mortality with both statins and ACE inhibitors. While we should all commend the investigators for performing this important study, it does not change how I feel about the way we need to be caring for patients with acute myocardial infarction. It certainly does not negate the results of the first DIGAMI study, nor does it change the previous reports of the beneficial effects of insulin infusion in this situation, even without a previous diagnosis of diabetes.8
While mortality was similar between those with and a similar population without diabetes, I believe that should not be the primary goal. Rather, I would prefer to see reduced cardiovascular outcomes in both groups. What DIGAMI 2 really tells us is we still have a long way to go before we reduce mortality to zero, or close to it. It will require further studies and will also require changes in our current protocols. For example, Furnary and colleagues believe glucose levels after cardiothoracic surgery should be near normal while receiving an insulin infusion for 3 days after surgery.9 DIGAMI 2 was not able to replicate this.
Or perhaps, as Malmberg stated, what we really need is a "statin for diabetes." Let's face it: Achieving glycemic control, both for the physician and the patient (and the patient's family), both in the hospital and at home, is much more difficult than prescribing a statin, ACE inhibitor, or beta-blocker. For the immediate future, diabetes will require more effort, with frequent glucose testing, dietary observance, and ideally, significant exercise.
Finally, I predict passionate debates about these results, perhaps similar
to a past study in diabetes with controversial conclusions. The University
Group Diabetes Program, announced in 1970, was so controversial that some
practitioners went decades without even speaking to their colleagues. I hope
that rather than igniting such a degree of passion, this study inspires us
with more material for healthy debate. 
|
References
2. Clement S, Braithwaite SS, Magee MF, Ahmann A, Smith EP, Schafer R,
Hirsch IB: (Technical Review) Management of diabetes and hyperglycemia in
hospitals. Diabetes Care 27:553–591, 2004.
3. DIGAMI Study Group, Malmberg K, Rydén L, Efendic S, Herlitz J, Nicol P, Waldenstrom A, Wedel H, Welin L: A randomized trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute myocardial infarction: Effects on mortality at 1 year. J Am Coll Cardiol 26:57–65, 1995.[Abstract]
4. Malmberg K: Prospective randomized study of intensive insulin
treatment on long-term survival after acute myocardial infarction in patients
with diabetes mellitus. BMJ 314: 1512–1515, 1997.
5. Dewitt DE, Hirsch IB: Outpatient therapy for type 1 and type 2
diabetes: Scientific review. JAMA 289: 2254–2264, 2003.
6. Esposito K, Giugliano D, Nappo F, Marfella K, for the Campanian Postprandial Hyperglycemia Study Group: Regression of atherosclerosis by control of postprandial hyperglycemia in type 2 diabetes mellitus. Circulation 110:214–219, 2004.[Medline]
7. Quagliaro L, Piconi L, Assalone R, Martinelli L, Motz E, Ceriello
A: Intermittent high glucose enhances apoptosis related to oxidative stress in
human umbilical vein endothelial cells: The role of protein kinase C and
NAD(P)H-oxidase activation. Diabetes 52: 2795–2804, 2003.
8. Fath-Ordoubadi F, Beatt KJ: Glucose-insulin-potassium therapy for treatment of acute myocardial infarction: An overview of randomized place-bo-controlled trials. Circulation 96: 1152–1156, 1997.[Medline]
9. Furnary AP, Gao G, Grunkemeier GL, Wu Y, Zerr KJ, Bookin SQ, Floten
HS, Starr A: Continuous insulin infusion reduces mortality in patients with
diabetes undergoing coronary artery bypass grafting. J Thorac
Cardiovasc Surg 125:1007–1021, 2003.
CiteULike 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||
|
||||||
|
| DOC News | Diabetes | Diabetes Care | Clinical Diabetes | Diabetes Spectrum |
