ORIGINAL ARTICLE
A
quantitative analysis
of
the effect
glucose-insulin-potassium in acute myocardial infarction
of
S. Rasoul, T. Svilaas, J-P. Ottervanger, J.R. Timmer, A.W.J. van 't Hof, F. Zijlstra
Objective. To review the currently available data to investigate the clinical benefit of high- and lowdose glucose-insulin-potassium (GIK) in patients with ST-segment elevation acute myocardial infarction (STEMI). Design. Quantitative analysis of all randomised trials on GIK in patients with STEMI. Electronic and manual searches for randomised controlled trials of GIK in STEMI were performed with regard to inclusion criteria, dose of GIK and additional use of reperfusion therapy, and a metaanalysis with the primary endpoint 30-day mortality was performed. Patients. Data from 16 randomised trials, involving 26,273 patients, were included. Results. Studies were conducted between 1962 and 2005. Overall, hospital mortality was 9.6% after GIK compared with 10.2% in controls (p=0.088). GIK infusion was not associated with an increase in major adverse events. Conclusion. This quantitative analysis of GIK in patients with STEMI did not show a beneficial or detrimental effect of GIK infusion on 30-day mortality. GIK infusion should not be part of the standard therapy for patients with STEMI. (Neth HeartJ2006;14:19-23.) S. Rasoul J-P. Ottervanger J.R. Tlmmer A.W.J. van 't Hof Department of Cardiology, Isala Clinics, Zwolle, the Netherlands T. Svilaas F.
ZIljstra
University Medical Centre, University of Groningen, Groningen, the Netherlands
Correspondence to: J-P Ottervanger Isala Clinics, Department of Cardiology, Groot Wezenland 20, 8011 JW Zwolle, the Netherlands E-mail: [email protected]
qqp)
Netherlands Heart Journal, Volume 14, Number 1, January 2006
Keywords: glucose-insulin-potassium, GIPS, myocardial infarction, reperfusion
Since the early 1960s, the potential benefits of glucose-insulin-potassium (GIK) for ST-segment elevation acute myocardial infarction (STEMI) have been studied. 3 During the acute phase of myocardial infarction, insulin levels are low, while cortisol and glucagon levels are increased.4 This is associated with decreased insulin sensitivity, resulting in impaired glucose utilisation and an increased turnover of free fatty acids (FFA).s Although FFA are the primary energy source for nonischaemic myocardium, myocardium can use various forms of energy substrates, including glucose.6 Increased circulating FFA levels and reduced insulin sensitivity may limit cellular uptake of glucose and promotes the use of FFA.7 FFA may have a detrimental effect on ischaemic myocardium through varying pathways. Energy supply by FFA is associated with relatively high oxygen consumption in comparison with the utilisation of glucose.8 Moreover, in contrast to glucose, FFA can not be metabolised anaerobically. Furthermore, excess FFA metabolism increases susceptibility to ventricular arrhythmnias and reperfusion injury due to disturbances in calcium homeostasis and accumulation of free radicals.9"10 Administration of GIK lowers circulating levels of FFA by inhibitory effects of insulin on lipolysis." This decrease in FFA levels, in combination with an increase in glucose and insulin availability, promotes the myocardial use of glucose over FFA.9 Glucose is less oxygen consuming and may have a beneficial effect on the preservation of mechanical function and membrane stability.'2 GIK may also reduce arrhythmias after successful reperfusion.'3 As insulin itself induces coronary vasodilatation, myocardial metabolism could further be improved through enhanced myocardial perfusion.'4-16 Finally, intensive insulin therapy may normalise glucose levels, 19
A quantitative analysis of the effect of glucose-insulin-potassium in acute myocardial infarction
Table 1. GIK trials and GIK doses used in acute myocardial infarction.
Study
High-doe GIK DIGAMI" GIPS 1119 ECLA-1120 Heng2' Stanley22 Rogers23 Satler24 ECLA-125 GIPS_126 REVIVAL31 Low-dose GIK Pentecost2 MRC*3 ECLA-125 Mittra*27 Pilcher*28 Hjermann*29 Pol-GlK30
Publication year
Patients (n)
Glucose (%)
Insulin (IU/I)
(ml/kg/h)
1995 2005 2005 1977 1978 1979 1987 1998 2003 2004
620 889 20,201 27 110 134 17 135 940 312
5 20 25 50 30 30 30 25 20 20
80 Variable 50 21 50 50 50 50 50 40
1.5 1/24h 2 1.5 1.5 1.5 1.5 1.5 1.5 3 1.8
24 hours 12 hours 24 hours 6-12 hours 48 hours 48 hours 48 hours 24 hours 12 hours 24 hours
1968 1968 1998 1965 1967 1971 1999
200 968 133 170 102 204 954
10 16 10 24 24 20 10
30 10x2 20 10x2 10x2 16 20
1.5 1/24h
48 hours 14 days 24 hours 14 days 14 days 10 days 24 hours
Infusion rate
-
1.0 -
-
0.6
Infusion period
*Glucose, oral potassium and subcutaneous insulin.
which has been shown to be beneficial in critically ill patients.'7
A meta-analysis of the older randomised trials of GIK in patients with STEMI suggested beneficial effects of GIK, but studies in the reperfusion era were lacking.'8 Since then, several additional clinical trials have been performed.5'9-30 The more recent trials did not show benefits of GIK'19,20
We designed this quantitative analysis to review the currently available data to investigate the clinical benefit of high- and low-dose GIK in patients with STEMI. Methods We attempted to obtain results from all completed, published, randomised trials of GIK in STEMI. The literature was scanned by formal searches of electronic databases (MEDLINE, PubMed) and informal searches for studies concerning the potential mechanism of action of GIK Since it has been suggested that highdose GIKis superior to low-dose GIK,2' we performed stratified analyses of the effects of a high and low dose ofGIK High-dose GIKwas defined as an intravenous infusion ofGIK in a dose equal to or higher than used by Rogers et al., 30% glucose (300 mg/l), 50 IU/l regular insulin and 80 mmol/l potassium chloride at 1.5 ml/kg per hour infusion rate (approximately 5 units insulin/hour).23 Our primary efficacy outcome of interest was 30-day mortality. We calculated the odds ratio (OR) for 30-day/hospital mortality of patients
20
treated with GIK as compared with those treated with placebo or control. The OR and its 95% confidence interval (CI) were calculated for each trial and the (grand) totals. Results
A total of 16 randomised clinical trialswere identified and included in this meta-analysis.i3' ""19' Studies were conducted between 1962 and 2005. The total number ofpatients was 26,273 and the overall 30-day mortality of the whole population was 2603. Baseline characteristics ofall randomised trials are summarised in table 1. Numbers of enrolled patients and hospital/30-day mortality rates are summarised in table 2.
The studies used GIK in different doses, and the time from onset of symptoms of myocardial infarction to treatment varied, with a large proportion of patients being treated more than six hours after onset of symptoms. The first randomised prospective trial in the era of reperfusion, the DIGAMI study, including only patients with hyperglycaemia on admission, found that the combination of insulin-glucose infusion with an intensive insulin treatment for at least three months after discharge reduced one-year mortality (19 vs. 26%, p16.8 mmol/l as criterion to stop the infusion, only 1% of infusions were interrupted. Hjermann used a reduced dose of insulin, which prevented the occurrence of hypoglycaemia.29
Discussion This large meta-analysis of GIK involving 26,273 patients with STEMI did not show beneficial effects of GIK infusion on 30-day mortality.
Comparison with previous meta-analysis The results of our meta-analysis differ from the previous meta-analysis as performed by Fath-Ordoubadi and Beatt in which a beneficial effect of GIK was suggested.'8 In that meta-analysis a total of 1932 patients were included. Hospitalmortality was reduced from 21% in the placebo group to 16.1% in the GIK group (p=O. 004, OR0.72, 95% CI 0.57 to 0.90). Our meta-analysis included all of the more recent trials which failed to show significant beneficial effect of GIK.19,20 This may be due to methodological differences between the older clinical trials and the recent trials. Furthermore, treatment of STEMI has changed and clinical outcome of these patients has been improved. Compared with our meta-analysis, the meta-analysis performed by Fath-Ordoubadi and Beat showed a higher mortality in both control (21 vs. 10.2%) and GIK patients (16.1 vs. 9.7%).18 This may be due to the more frequent use of reperfusion therapy as well as 1-blockers, ACE inhibitors and statins in the more recent trials. In addition, the beneficial effects of GIK may have been overshadowed by improvement in the general management of STEMI. GIK and time to reperfusion It has been proposed thatglucose-insulin therapy may delay ischaemic necrosis to some degree before reperfusion, thus lengtheningthe period during which effective myocardial salvage is possiblewith reperfusion strategies.32 Time between admission and initiation of GIK varied among the studies. Mittra mentioned ten hours,27 whereas in the Medical Research Council 22
(MRC) study, 70% of patients were treated within 30 minutes after inclusion.3 In the ECIA study, time from onset ofsymptoms until the initiation of GIKtreatment was 10 to 11 hours.25 Whether the clinical effects of GIK on outcome are influenced by the time of initiation of treatment is not known. Reperfusion is mandatory as prolonged severe ischaemia, even in the presence of GIK, will eventually result in necrosis. It has been postulated that GIK infusion would be most effective when initiated before reperfusion, with continuation of the infusion for several hours after reperfusion.33 Future of GIK in myocardial infarction There may be several reasons for the controversial results of the trials with GIK The older trials had wellknown limitations with low patient numbers, poor design and methodological discrepancies. The more recent trials were larger, had a better design and in most trials GIK infusion was added to reperfusion therapy. However, even in the more recent trials, the metabolic effects of administration of GIK may not have been optimal. GIK infusion was started just a short time before reperfusion therapy and significant hyperglycaemia was induced in a large population of patients. Additional studies may be needed to find out whether GIK is effective when given early after symptom onset, for instance during transportation to hospital.34 Intensive insulin therapy to maintain blood glucose with a more optimal range reduces morbidity and mortality among critically ill patients in the surgical intensive care unit.'7 However, although the results of DIGAMI-1 were promising, the results of DIGAMI2 did not confirm the benefits of intensive insulin treatment in type 2 diabetic patients with STEMI.35
Conclusion This large meta-analysis of GIK therapy in patients with STEMI, involving 26,082 patients, did not show a beneficial or detrimental effect of GIK infusion on 30-day mortality. Therefore, GIK therapy in the current form should not be used in patients with STEMI. m
Acknowledgements This study was supported by a grant from the Netherlands Heart Foundation (2003 B277). References 1
2
3 4
Sodi-Pallares D, Testelli MR, Fishleder BL, Bisteni A, Medrano GA, Friedland C, et al. Effects of an intravenous infusion of a potassium-glucose-insulin solution on the electrocardiographic signs of myocardial infarction. A preliminary clinical report. AmJ Card 1962;9:166-81. Pentecost BL, Mayne NM, Lamb P. Controlled trial of intravenous glucose, potassium, and insulin in acute myocardial infarction. Lancet 1968;1:946-8. Medical Research Council Working Party on the Treatment of Myocardial Infarction. Potassium, glucose, and insulin treatment for acute myocardial infarction. Lancet 1968;2:1355-60. Vetter NJ, Strange RC, Adams W, Oliver MF. Initial metabolic and hormonal response to acute myocardial infarction. Lancet 1974;1:284-8.
Netherlands Heart Journal, Volume 14, Number 1, January 2006
Q
A quantitative analysis of the effect of glucose-insulin-potassium in acute myocardial infarction
5
6
7 8 9
10 11
12 13
14
15
16
17 18
19
20
pc
Malmberg K, Ryden L, Efendic S, Herlitz J, Nicol P, Waldenstrom A, et al. Randomised trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute myocardial infarction (DIGAMI study): effects on mortality at 1 year. JAm Coll Cardiol 1995;26:57-65. Neely JR, Rovetto MJ, Whitmer JT, Morgan HE. Effects of ischemia on function and metabolism of the isolated working rat heart. AmJPhysiol 1973;225:651-8. Rutenberg HL, Pamintuan JC, SoloffLA. Serum-free-fatty-acids and their relation to complications after acute myocardial infarction. Lancet 1969;2:559-64. Vik-Mo H, Mjos OD. Influence of free fatty acids on myocardial oxygen consumption and ischemic injury. Am J Cardiol 1981; 48:361-5. Oliver MF, Opie LH. Effects of glucose and fatty acids on myocardial ischaemia and arrhythmias. Lancet 1994;343:155-8. Tansey MJ, Opie LH. Relation between plasma free fatty acids and arrhythmias within the first twelve hours of acute myocardial infarction. Lancet 1983;2:419-22. McDaniel HG, Papapietro SE, Rogers WJ, Mantle JA, Smith LR, Russell RO Jr, et al. Glucose-insulin-potassium induced alterations in individual plasma free fatty acids in patients with acute myocardial infarction. Am HeartJ1981;102:10-5. Heng MK, Norris RM, Peter T, Nisbet HD, Singh BN. The effect of glucose-insulin-potassium on experimental myocardial infarction in the dog. Cardiovasc Res 1978;12:429-35. Eberli FR, Weinberg EO, Grice WN, Horowitz GL, Apstein CS. Protective effect of increased glycolytic substrate against systolic and diastolic dysfunction and increased coronary resistance from prolonged global underperfusion and reperfusion in isolated rabbit hearts perfused with erythrocyte suspensions. Circ Res 1991;68: 466-81. Marano L, Bestetti A, Lomuscio A, Tagliabue L, Castini D, Tarricone D, et al. Effects ofinfusion of glucose-insulin-potassium on myocardial function after a recent myocardial infarction. Acta Cardiol 2000;55:9-15. Laine H, Nuutila P, Luotolahti M, Meyer C, Elomaa T, Koskinen P, et al. Insulin-induced increment of coronary flow reserve is not abolished by dexamethasone in healthy young men. J Clin Endocrinol Metab 2000;85:1868-73. Sundell J, Knuuti J. Insulin and myocardial blood flow. Cardiovasc Res 2003;57:312-9. Van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, et al. Intensive insulin therapy in the critically ill patients. NEnglJMed 2001;345:1359-67. Fath-Ordoubadi F, Beatt KJ. Glucose-insulin-potassium therapy for treatment of acute myocardial infarction: an overview of randomised placebo-controlled trials. Circulation 1997;96:1152-6. Timmer JR, Svilaas T, Ottervanger JP, Henriques JPS, Dambrink J-HE, van den Broek SAJ, van der Horst ICC, Zijlstra F. Glucoseinsulin-potassium infusion in patients with acute myocardial infarction vithout signs of heart failure: the glucose-insulinpotassium-study II. Accepted for publication in JACC. Mehta SR, Yusuf S, Diaz R, Zhu J, Pais P, Xavier D, et al. CREATE-ECLA Trial Group Investigators. Effect of glucose-insulin-potassium infusion on mortality in patients writh acute STsegment elevation myocardial infarction: the CREATE-ECLA randomised controlled trial. JAMA 2005;293:437-6.
Netherlands Heart Joumal, Volume 14, Number 1, January 2006
21 Heng MK, Norris RM, Singh BN, Barratt-Boyes C. Effects of glucose-insulin-potassium on haemodynamics and enzyme release after acute myocardial infarction. Br HeartJ 1977; 39:748-57. 22 Stanley AW, Prather JW. Glucose-insulin-potassium, patient mortality and acute myocardial infarction: results from a prospective randomised study [abstract]. Circulation 1978;57 (Suppl II):II-62. 23 Rogers WJ, Segall PH, McDaniel HG, Mantle JA, Russell RO Jr, Rackley CE. Prospective randomised trial of glucose- insulin-potassium in myocardial infarction. Effects on myocardial hemodynamics, substrates and rhythm. Am J Cardiol 1979;43:801-9. 24 Satler LF, Green CE, Kent KM, Pallas RS, Pearle DL, Rackley CE. Metabolic support during coronary reperfusion. Am Heart J 1987;114:54-8. 25 Diaz R, Paolasso EA, Piegas LS, Tajer CD, Moreno MG, Corvalan R, et al. Metabolic modulation ofacute myocardial infarction. The ECLA (Estudios Cardiologicos Latinoamerica) Collaborative Group. Circulation 1998;98:2227-34. 26 Van der Horst IC, Zijlstra F, van 't HofAWJ, Doggen CJM, de Boer MJ, Suryapranata H, et al., for the Zwolle Infarction Study Group. Glucose-insulin-potassium infusion in patients treated with primary angioplasty for acute myocardial infarction: the glucoseinsulin-potassium study: a randomised trial. JAm Coil Cardiol 2003;42:784-91. 27 Mittra B. Potassium, glucose, and insulin in treatment of myocardial infarction. Lancet 1965;2:607-9. 28 Pilcher J, Etishamudin M, Exon P, Moore J. Potassium, glucose and insulin in myocardial infarction. Lancet 1967;1:1109. 29 Hjermann I. A controlled study of per oral glucose, insulin and potassium treatment in myocardial infarction. Acta Med Scand 1971;190:213-8. 30 Ceremuzynski L, Budaj A, Czepiel A, Burzykowski T, Achremczyk P, Smielak-Korombel W, et al. Low-dose glucose- insulin-potassium is ineffective in acute myocardial infarction: results of a randomised multicenter Pol-GIK trial. Cardiovasc Drugs Ther 1999;13:191-200. 31 Pache J, Kastrati A, Mehilli J, Bollwein H, Ndrepepa G, Schuhlen H, et al. A randomized evaluation of the effects of glucose-insulinpotassium infusion on myocardial salvage in patients with acute myocardial infarction treated with reperfusion therapy. Am Heart J2004:148:e3. 32 Van Overschelde JL, Janier MF, Bakke JE, Marshall DR, Bergmann SR. Rate of glycolysis during ischemia determines extent of ischemic injury and functional recovery after reperfusion.
AmjPhys 1994;267:H1785-H94. 33 Sack MN, Yellon DM. Insulin therapy as an adjunct to reperfusion after acute coronary ischemia: a proposed direct myocardial cell survival effect independent of metabolic modulation. JAm Coll
Cardiol2003;41:1404-7. 34 Apstein CS, Opie LH. A challenge to the metabolic approach to myocardial ischaemia. Eur HeartJ2005;26:956-9. 35 Malmberg K, Ryden L, Wedel H, Birkeland K, Bootsma A, Dickstein K, et al. Intense metabolic control by means of insulin in patients with diabetes mellitus and acute myocardial infarction (DIGAMI 2): effects on mortality and morbidity. Eur HeartJ 2005;26:650-61.
23
A
quantitative analysis
of
the effect
glucose-insulin-potassium in acute myocardial infarction
of
S. Rasoul, T. Svilaas, J-P. Ottervanger, J.R. Timmer, A.W.J. van 't Hof, F. Zijlstra
Objective. To review the currently available data to investigate the clinical benefit of high- and lowdose glucose-insulin-potassium (GIK) in patients with ST-segment elevation acute myocardial infarction (STEMI). Design. Quantitative analysis of all randomised trials on GIK in patients with STEMI. Electronic and manual searches for randomised controlled trials of GIK in STEMI were performed with regard to inclusion criteria, dose of GIK and additional use of reperfusion therapy, and a metaanalysis with the primary endpoint 30-day mortality was performed. Patients. Data from 16 randomised trials, involving 26,273 patients, were included. Results. Studies were conducted between 1962 and 2005. Overall, hospital mortality was 9.6% after GIK compared with 10.2% in controls (p=0.088). GIK infusion was not associated with an increase in major adverse events. Conclusion. This quantitative analysis of GIK in patients with STEMI did not show a beneficial or detrimental effect of GIK infusion on 30-day mortality. GIK infusion should not be part of the standard therapy for patients with STEMI. (Neth HeartJ2006;14:19-23.) S. Rasoul J-P. Ottervanger J.R. Tlmmer A.W.J. van 't Hof Department of Cardiology, Isala Clinics, Zwolle, the Netherlands T. Svilaas F.
ZIljstra
University Medical Centre, University of Groningen, Groningen, the Netherlands
Correspondence to: J-P Ottervanger Isala Clinics, Department of Cardiology, Groot Wezenland 20, 8011 JW Zwolle, the Netherlands E-mail: [email protected]
qqp)
Netherlands Heart Journal, Volume 14, Number 1, January 2006
Keywords: glucose-insulin-potassium, GIPS, myocardial infarction, reperfusion
Since the early 1960s, the potential benefits of glucose-insulin-potassium (GIK) for ST-segment elevation acute myocardial infarction (STEMI) have been studied. 3 During the acute phase of myocardial infarction, insulin levels are low, while cortisol and glucagon levels are increased.4 This is associated with decreased insulin sensitivity, resulting in impaired glucose utilisation and an increased turnover of free fatty acids (FFA).s Although FFA are the primary energy source for nonischaemic myocardium, myocardium can use various forms of energy substrates, including glucose.6 Increased circulating FFA levels and reduced insulin sensitivity may limit cellular uptake of glucose and promotes the use of FFA.7 FFA may have a detrimental effect on ischaemic myocardium through varying pathways. Energy supply by FFA is associated with relatively high oxygen consumption in comparison with the utilisation of glucose.8 Moreover, in contrast to glucose, FFA can not be metabolised anaerobically. Furthermore, excess FFA metabolism increases susceptibility to ventricular arrhythmnias and reperfusion injury due to disturbances in calcium homeostasis and accumulation of free radicals.9"10 Administration of GIK lowers circulating levels of FFA by inhibitory effects of insulin on lipolysis." This decrease in FFA levels, in combination with an increase in glucose and insulin availability, promotes the myocardial use of glucose over FFA.9 Glucose is less oxygen consuming and may have a beneficial effect on the preservation of mechanical function and membrane stability.'2 GIK may also reduce arrhythmias after successful reperfusion.'3 As insulin itself induces coronary vasodilatation, myocardial metabolism could further be improved through enhanced myocardial perfusion.'4-16 Finally, intensive insulin therapy may normalise glucose levels, 19
A quantitative analysis of the effect of glucose-insulin-potassium in acute myocardial infarction
Table 1. GIK trials and GIK doses used in acute myocardial infarction.
Study
High-doe GIK DIGAMI" GIPS 1119 ECLA-1120 Heng2' Stanley22 Rogers23 Satler24 ECLA-125 GIPS_126 REVIVAL31 Low-dose GIK Pentecost2 MRC*3 ECLA-125 Mittra*27 Pilcher*28 Hjermann*29 Pol-GlK30
Publication year
Patients (n)
Glucose (%)
Insulin (IU/I)
(ml/kg/h)
1995 2005 2005 1977 1978 1979 1987 1998 2003 2004
620 889 20,201 27 110 134 17 135 940 312
5 20 25 50 30 30 30 25 20 20
80 Variable 50 21 50 50 50 50 50 40
1.5 1/24h 2 1.5 1.5 1.5 1.5 1.5 1.5 3 1.8
24 hours 12 hours 24 hours 6-12 hours 48 hours 48 hours 48 hours 24 hours 12 hours 24 hours
1968 1968 1998 1965 1967 1971 1999
200 968 133 170 102 204 954
10 16 10 24 24 20 10
30 10x2 20 10x2 10x2 16 20
1.5 1/24h
48 hours 14 days 24 hours 14 days 14 days 10 days 24 hours
Infusion rate
-
1.0 -
-
0.6
Infusion period
*Glucose, oral potassium and subcutaneous insulin.
which has been shown to be beneficial in critically ill patients.'7
A meta-analysis of the older randomised trials of GIK in patients with STEMI suggested beneficial effects of GIK, but studies in the reperfusion era were lacking.'8 Since then, several additional clinical trials have been performed.5'9-30 The more recent trials did not show benefits of GIK'19,20
We designed this quantitative analysis to review the currently available data to investigate the clinical benefit of high- and low-dose GIK in patients with STEMI. Methods We attempted to obtain results from all completed, published, randomised trials of GIK in STEMI. The literature was scanned by formal searches of electronic databases (MEDLINE, PubMed) and informal searches for studies concerning the potential mechanism of action of GIK Since it has been suggested that highdose GIKis superior to low-dose GIK,2' we performed stratified analyses of the effects of a high and low dose ofGIK High-dose GIKwas defined as an intravenous infusion ofGIK in a dose equal to or higher than used by Rogers et al., 30% glucose (300 mg/l), 50 IU/l regular insulin and 80 mmol/l potassium chloride at 1.5 ml/kg per hour infusion rate (approximately 5 units insulin/hour).23 Our primary efficacy outcome of interest was 30-day mortality. We calculated the odds ratio (OR) for 30-day/hospital mortality of patients
20
treated with GIK as compared with those treated with placebo or control. The OR and its 95% confidence interval (CI) were calculated for each trial and the (grand) totals. Results
A total of 16 randomised clinical trialswere identified and included in this meta-analysis.i3' ""19' Studies were conducted between 1962 and 2005. The total number ofpatients was 26,273 and the overall 30-day mortality of the whole population was 2603. Baseline characteristics ofall randomised trials are summarised in table 1. Numbers of enrolled patients and hospital/30-day mortality rates are summarised in table 2.
The studies used GIK in different doses, and the time from onset of symptoms of myocardial infarction to treatment varied, with a large proportion of patients being treated more than six hours after onset of symptoms. The first randomised prospective trial in the era of reperfusion, the DIGAMI study, including only patients with hyperglycaemia on admission, found that the combination of insulin-glucose infusion with an intensive insulin treatment for at least three months after discharge reduced one-year mortality (19 vs. 26%, p16.8 mmol/l as criterion to stop the infusion, only 1% of infusions were interrupted. Hjermann used a reduced dose of insulin, which prevented the occurrence of hypoglycaemia.29
Discussion This large meta-analysis of GIK involving 26,273 patients with STEMI did not show beneficial effects of GIK infusion on 30-day mortality.
Comparison with previous meta-analysis The results of our meta-analysis differ from the previous meta-analysis as performed by Fath-Ordoubadi and Beatt in which a beneficial effect of GIK was suggested.'8 In that meta-analysis a total of 1932 patients were included. Hospitalmortality was reduced from 21% in the placebo group to 16.1% in the GIK group (p=O. 004, OR0.72, 95% CI 0.57 to 0.90). Our meta-analysis included all of the more recent trials which failed to show significant beneficial effect of GIK.19,20 This may be due to methodological differences between the older clinical trials and the recent trials. Furthermore, treatment of STEMI has changed and clinical outcome of these patients has been improved. Compared with our meta-analysis, the meta-analysis performed by Fath-Ordoubadi and Beat showed a higher mortality in both control (21 vs. 10.2%) and GIK patients (16.1 vs. 9.7%).18 This may be due to the more frequent use of reperfusion therapy as well as 1-blockers, ACE inhibitors and statins in the more recent trials. In addition, the beneficial effects of GIK may have been overshadowed by improvement in the general management of STEMI. GIK and time to reperfusion It has been proposed thatglucose-insulin therapy may delay ischaemic necrosis to some degree before reperfusion, thus lengtheningthe period during which effective myocardial salvage is possiblewith reperfusion strategies.32 Time between admission and initiation of GIK varied among the studies. Mittra mentioned ten hours,27 whereas in the Medical Research Council 22
(MRC) study, 70% of patients were treated within 30 minutes after inclusion.3 In the ECIA study, time from onset ofsymptoms until the initiation of GIKtreatment was 10 to 11 hours.25 Whether the clinical effects of GIK on outcome are influenced by the time of initiation of treatment is not known. Reperfusion is mandatory as prolonged severe ischaemia, even in the presence of GIK, will eventually result in necrosis. It has been postulated that GIK infusion would be most effective when initiated before reperfusion, with continuation of the infusion for several hours after reperfusion.33 Future of GIK in myocardial infarction There may be several reasons for the controversial results of the trials with GIK The older trials had wellknown limitations with low patient numbers, poor design and methodological discrepancies. The more recent trials were larger, had a better design and in most trials GIK infusion was added to reperfusion therapy. However, even in the more recent trials, the metabolic effects of administration of GIK may not have been optimal. GIK infusion was started just a short time before reperfusion therapy and significant hyperglycaemia was induced in a large population of patients. Additional studies may be needed to find out whether GIK is effective when given early after symptom onset, for instance during transportation to hospital.34 Intensive insulin therapy to maintain blood glucose with a more optimal range reduces morbidity and mortality among critically ill patients in the surgical intensive care unit.'7 However, although the results of DIGAMI-1 were promising, the results of DIGAMI2 did not confirm the benefits of intensive insulin treatment in type 2 diabetic patients with STEMI.35
Conclusion This large meta-analysis of GIK therapy in patients with STEMI, involving 26,082 patients, did not show a beneficial or detrimental effect of GIK infusion on 30-day mortality. Therefore, GIK therapy in the current form should not be used in patients with STEMI. m
Acknowledgements This study was supported by a grant from the Netherlands Heart Foundation (2003 B277). References 1
2
3 4
Sodi-Pallares D, Testelli MR, Fishleder BL, Bisteni A, Medrano GA, Friedland C, et al. Effects of an intravenous infusion of a potassium-glucose-insulin solution on the electrocardiographic signs of myocardial infarction. A preliminary clinical report. AmJ Card 1962;9:166-81. Pentecost BL, Mayne NM, Lamb P. Controlled trial of intravenous glucose, potassium, and insulin in acute myocardial infarction. Lancet 1968;1:946-8. Medical Research Council Working Party on the Treatment of Myocardial Infarction. Potassium, glucose, and insulin treatment for acute myocardial infarction. Lancet 1968;2:1355-60. Vetter NJ, Strange RC, Adams W, Oliver MF. Initial metabolic and hormonal response to acute myocardial infarction. Lancet 1974;1:284-8.
Netherlands Heart Journal, Volume 14, Number 1, January 2006
Q
A quantitative analysis of the effect of glucose-insulin-potassium in acute myocardial infarction
5
6
7 8 9
10 11
12 13
14
15
16
17 18
19
20
pc
Malmberg K, Ryden L, Efendic S, Herlitz J, Nicol P, Waldenstrom A, et al. Randomised trial of insulin-glucose infusion followed by subcutaneous insulin treatment in diabetic patients with acute myocardial infarction (DIGAMI study): effects on mortality at 1 year. JAm Coll Cardiol 1995;26:57-65. Neely JR, Rovetto MJ, Whitmer JT, Morgan HE. Effects of ischemia on function and metabolism of the isolated working rat heart. AmJPhysiol 1973;225:651-8. Rutenberg HL, Pamintuan JC, SoloffLA. Serum-free-fatty-acids and their relation to complications after acute myocardial infarction. Lancet 1969;2:559-64. Vik-Mo H, Mjos OD. Influence of free fatty acids on myocardial oxygen consumption and ischemic injury. Am J Cardiol 1981; 48:361-5. Oliver MF, Opie LH. Effects of glucose and fatty acids on myocardial ischaemia and arrhythmias. Lancet 1994;343:155-8. Tansey MJ, Opie LH. Relation between plasma free fatty acids and arrhythmias within the first twelve hours of acute myocardial infarction. Lancet 1983;2:419-22. McDaniel HG, Papapietro SE, Rogers WJ, Mantle JA, Smith LR, Russell RO Jr, et al. Glucose-insulin-potassium induced alterations in individual plasma free fatty acids in patients with acute myocardial infarction. Am HeartJ1981;102:10-5. Heng MK, Norris RM, Peter T, Nisbet HD, Singh BN. The effect of glucose-insulin-potassium on experimental myocardial infarction in the dog. Cardiovasc Res 1978;12:429-35. Eberli FR, Weinberg EO, Grice WN, Horowitz GL, Apstein CS. Protective effect of increased glycolytic substrate against systolic and diastolic dysfunction and increased coronary resistance from prolonged global underperfusion and reperfusion in isolated rabbit hearts perfused with erythrocyte suspensions. Circ Res 1991;68: 466-81. Marano L, Bestetti A, Lomuscio A, Tagliabue L, Castini D, Tarricone D, et al. Effects ofinfusion of glucose-insulin-potassium on myocardial function after a recent myocardial infarction. Acta Cardiol 2000;55:9-15. Laine H, Nuutila P, Luotolahti M, Meyer C, Elomaa T, Koskinen P, et al. Insulin-induced increment of coronary flow reserve is not abolished by dexamethasone in healthy young men. J Clin Endocrinol Metab 2000;85:1868-73. Sundell J, Knuuti J. Insulin and myocardial blood flow. Cardiovasc Res 2003;57:312-9. Van den Berghe G, Wouters P, Weekers F, Verwaest C, Bruyninckx F, Schetz M, et al. Intensive insulin therapy in the critically ill patients. NEnglJMed 2001;345:1359-67. Fath-Ordoubadi F, Beatt KJ. Glucose-insulin-potassium therapy for treatment of acute myocardial infarction: an overview of randomised placebo-controlled trials. Circulation 1997;96:1152-6. Timmer JR, Svilaas T, Ottervanger JP, Henriques JPS, Dambrink J-HE, van den Broek SAJ, van der Horst ICC, Zijlstra F. Glucoseinsulin-potassium infusion in patients with acute myocardial infarction vithout signs of heart failure: the glucose-insulinpotassium-study II. Accepted for publication in JACC. Mehta SR, Yusuf S, Diaz R, Zhu J, Pais P, Xavier D, et al. CREATE-ECLA Trial Group Investigators. Effect of glucose-insulin-potassium infusion on mortality in patients writh acute STsegment elevation myocardial infarction: the CREATE-ECLA randomised controlled trial. JAMA 2005;293:437-6.
Netherlands Heart Joumal, Volume 14, Number 1, January 2006
21 Heng MK, Norris RM, Singh BN, Barratt-Boyes C. Effects of glucose-insulin-potassium on haemodynamics and enzyme release after acute myocardial infarction. Br HeartJ 1977; 39:748-57. 22 Stanley AW, Prather JW. Glucose-insulin-potassium, patient mortality and acute myocardial infarction: results from a prospective randomised study [abstract]. Circulation 1978;57 (Suppl II):II-62. 23 Rogers WJ, Segall PH, McDaniel HG, Mantle JA, Russell RO Jr, Rackley CE. Prospective randomised trial of glucose- insulin-potassium in myocardial infarction. Effects on myocardial hemodynamics, substrates and rhythm. Am J Cardiol 1979;43:801-9. 24 Satler LF, Green CE, Kent KM, Pallas RS, Pearle DL, Rackley CE. Metabolic support during coronary reperfusion. Am Heart J 1987;114:54-8. 25 Diaz R, Paolasso EA, Piegas LS, Tajer CD, Moreno MG, Corvalan R, et al. Metabolic modulation ofacute myocardial infarction. The ECLA (Estudios Cardiologicos Latinoamerica) Collaborative Group. Circulation 1998;98:2227-34. 26 Van der Horst IC, Zijlstra F, van 't HofAWJ, Doggen CJM, de Boer MJ, Suryapranata H, et al., for the Zwolle Infarction Study Group. Glucose-insulin-potassium infusion in patients treated with primary angioplasty for acute myocardial infarction: the glucoseinsulin-potassium study: a randomised trial. JAm Coil Cardiol 2003;42:784-91. 27 Mittra B. Potassium, glucose, and insulin in treatment of myocardial infarction. Lancet 1965;2:607-9. 28 Pilcher J, Etishamudin M, Exon P, Moore J. Potassium, glucose and insulin in myocardial infarction. Lancet 1967;1:1109. 29 Hjermann I. A controlled study of per oral glucose, insulin and potassium treatment in myocardial infarction. Acta Med Scand 1971;190:213-8. 30 Ceremuzynski L, Budaj A, Czepiel A, Burzykowski T, Achremczyk P, Smielak-Korombel W, et al. Low-dose glucose- insulin-potassium is ineffective in acute myocardial infarction: results of a randomised multicenter Pol-GIK trial. Cardiovasc Drugs Ther 1999;13:191-200. 31 Pache J, Kastrati A, Mehilli J, Bollwein H, Ndrepepa G, Schuhlen H, et al. A randomized evaluation of the effects of glucose-insulinpotassium infusion on myocardial salvage in patients with acute myocardial infarction treated with reperfusion therapy. Am Heart J2004:148:e3. 32 Van Overschelde JL, Janier MF, Bakke JE, Marshall DR, Bergmann SR. Rate of glycolysis during ischemia determines extent of ischemic injury and functional recovery after reperfusion.
AmjPhys 1994;267:H1785-H94. 33 Sack MN, Yellon DM. Insulin therapy as an adjunct to reperfusion after acute coronary ischemia: a proposed direct myocardial cell survival effect independent of metabolic modulation. JAm Coll
Cardiol2003;41:1404-7. 34 Apstein CS, Opie LH. A challenge to the metabolic approach to myocardial ischaemia. Eur HeartJ2005;26:956-9. 35 Malmberg K, Ryden L, Wedel H, Birkeland K, Bootsma A, Dickstein K, et al. Intense metabolic control by means of insulin in patients with diabetes mellitus and acute myocardial infarction (DIGAMI 2): effects on mortality and morbidity. Eur HeartJ 2005;26:650-61.
23
