- Email: [email protected]
Insulin Treatment of Diabetic Rats Reduces Cardiac Function in a Lipopolysaccharide-Induced Systemic Inflammation Model
Journal of Surgical Research 171, 251–258 (2011) doi:10.1016/j.jss.2010.03.032
Insulin Treatment of Diabetic Rats Reduces Cardiac Function in a Lipopolysaccharide-Induced Systemic Inflammation Model Satoshi Hagiwara, M.D., Ph.D.,*,1 Hideo Iwasaka, M.D., Ph.D.,* Kyosuke Kudo, M.D.,* Akira Hasegawa, M.D.,* Jyunya Kusaka, M.D.,* Tomohisa Uchida, M.D., Ph.D.,† and Takayuki Noguchi, M.D., Ph.D.* *Department of Anesthesiology and Intensive Care Medicine; and †Department of Pathology, Oita University Faculty of Medicine, Yufu City, Oita, Japan Submitted for publication January 8, 2010
INTRODUCTION Background. Diabetes is a common comorbidity in patients with various medical conditions. Tight glucose control is known to improve systemic inflammation; however, whether it is effective in diabetic patients is unknown. The purpose of this study was to examine how strict glucose control affects systemic inflammation in diabetic patients. Materials and Methods. Male Wistar rats. We determined the effect of insulin therapy on cardiac function in a rat model of systemic inflammation. We administered lipopolysaccharide intravenously, with or without insulin, to streptozotocin-induced diabetic rats. After induction of systemic inflammation, we determined serum cytokine (IL-6 and TNFa) and nitrate/ nitrite levels and measured cardiac function. Results. Cytokine levels and cardiac function were significantly reduced in diabetic rats compared to non-diabetic rats. Moreover, insulin treatment was associated with higher cytokine levels and decreased cardiac function. Conclusion. In systemic inflammatory conditions, diabetes increases various proinflammatory mediators and inhibits cardiac function; insulin treatment exacerbates these effects. Thus, strict glucose control may not be desirable in diabetic patients with systemic inflammatory conditions. Ó 2011 Elsevier Inc. All rights reserved. Key Words: inflammation; sepsis; diabetes; insulin; lipopolysaccharide; cardiac dysfunction.
1 To whom correspondence and reprint requests should be addressed at Department of Anesthesiology and Intensive Care Medicine, Oita University Faculty of Medicine, 1-1 Idaigaoka Hasamamachi, Yufu City, Oita 879-5593, Japan. E-mail: saku@med. oita-u.ac.jp.
The World Health Organization estimates that approximately 180 million individuals are affected by diabetes worldwide, and this number is expected to double by the year 2030 [1]. Infection rates are also growing, and sepsis and systemic inflammation are significant problems in intensive care units [2]. In particular, untreated diabetic patients with severe infections are often seen in intensive care units. Hyperglycemia can often exacerbate inflammation. In a previous study, we found that acute hyperglycemia worsened lung injury and cardiac function after systemic inflammation was induced by LPS [3, 4]. However, the difference between acute hyperglycemia and chronic hyperglycemia due to diabetes is not well understood. Van den Berghe et al. reported that intensive insulin therapy to maintain blood glucose at or below 110 mg per dL reduces morbidity and mortality among critically ill patients [5]. The anti-inflammatory action of insulin is one of the reasons for this effect. Tight glucose control also reduces cytokine levels in people without diabetes [3, 4]. However, the mechanism by which serum glucose affects systemic inflammation is unknown. Cardiac dysfunction is common in intensive care units [6] and is clearly associated with many cases of septic shock and systemic inflammation [7]. Cardiac dysfunction is a life-threatening condition with a high rate of mortality. Nitric oxide (NO), which exerts numerous biological effects in the cardiovascular system, is an important factor in cardiac dysfunction [8]. The inhibition of inducible nitric oxide synthase (iNOS) reduces NO production in cardiac myocytes, thereby restoring contractile function and improving outcomes in patients with septic shock [9].
251
0022-4804/$36.00 Ó 2011 Elsevier Inc. All rights reserved.
252
JOURNAL OF SURGICAL RESEARCH: VOL. 171, NO. 1, NOVEMBER 2011
Cytokines are also important mediators in the progression of acute cardiac dysfunction. Tumor necrosis factor alpha (TNFa) plays a key role in the induction of NO synthase [10]. In addition, acute exposure of isolated muscles to TNFa significantly reduces specific muscle force without a loss in muscle mass or protein levels [11]. Interleukin (IL)-6, a multifunctional cytokine, is also associated with the progression of cardiac dysfunction [12]. In this study, we hypothesized that strict glucose control would affect the response to systemic inflammation in diabetic patients. To this end, we investigated whether strict glucose control in diabetic rats affected levels of systemic inflammation markers, such as serum cytokines and NO, and cardiac function.
Histological Analysis Hearts (n ¼ 4 in each group) were obtained from animals under 4% sevoflurane anesthesia. Heart tissue specimens were fixed with 10% formalin and embedded in paraffin. Sections were stained with hematoxylin and eosin. A pathologist blinded to group assignment analyzed the samples.
Nitrate/Nitrite Determination In biological fluids, nitric oxide (NO) is rapidly converted to nitrite and nitrate. Therefore, we determined NO levels after serum collection by cardiac puncture by measuring nitrite and nitrate (NOx) levels with commercial kits according to the manufacturer’s instructions (R and D Systems Inc., Minneapolis, MN). NOx levels were determined by a modified Griess assay and absorbance was measured at 540 nm. Nitrite and nitrate concentrations were calculated using a standard curve and expressed as mmol/L.
Cytokine Measurements MATERIALS AND METHODS Insulin (Ins) was obtained from Novo Nordisk A/S (Copenhagen, Denmark). Lipopolysaccharide (LPS, O127:B8) was obtained from Sigma (St. Louis, MO). All other reagents were of the highest available analytical grade.
Animals All protocols conformed to the National Institute of Health (NIH) guidelines and all animals received humane care in compliance with the Principles of Laboratory Animal Care. The study was approved by the Ethics Committee of Animal Research at the College of Medicine, Oita University, Oita, Japan. Male Wistar rats weighing 250 to 300 g (Kyudou, Saga, Japan) were used in all experiments. Rats were given free access to food and water. Diabetes was induced with streptozotocin and a model for systemic inflammation was generated using LPS. Animals were randomly assigned to one of six groups (n ¼ 20 in each group) and received treatments intravenously: (1) saline group: rats received 0.9 % NaCl solution on d 1 of the experiment and 4 wk later; (2) diabetes mellitus (DM) group: rats received 60 mg/kg streptozotocin, and 4 wk later received 0.9 % NaCl solution; (3) DMþIns group: rats received 60 mg/kg streptozotocin on d 1, and 4 wk later received a bolus injection of 0.9 % NaCl solution 3 h after starting continuous insulin infusion (20 IU/kg/ h); (4) LPS group: rats received 0.9 % NaCl solution, and 4 wk later received LPS (7.5 mg/kg); (5) DMþLPS group: rats received 60 mg/ kg streptozotocin, and 4 wk later received LPS (7.5 mg/kg); (6) DMþLPSþIns group: rats received 60 mg/kg streptozotocin, and 4 wk later received a bolus injection of LPS (7.5 mg/kg) 3 h after starting continuous insulin infusion (20 IU/kg/h). Samples of venous blood (0.3 mL) were obtained from the external jugular vein at various time points (0, 3, 6, and 9 h). Anesthesia was maintained by 4% sevoflurane (Maruishi Pharmaceutical Co., Osaka, Japan). All animals were sacrificed at 9 h while under deep anesthesia using pentobarbital sodium. Heart tissue specimens were quickly removed and processed as indicated below.
Blood Glucose Measurement Blood samples (n ¼ 6 in each group) were collected from the external jugular vein at each time point. Whole blood glucose levels were determined using the glucose oxidase method. The Glu-test-sensor blood glucose monitoring system was purchased from Sanwa Kagaku Kenkyusho Co., Ltd. (Nagoya, Japan).
Serum samples (n ¼ 6 in each group) were assayed using the sandwich enzyme-linked immunosorbent assay (ELISA) method. Ninetysix-well plates were precoated with monoclonal antibodies specific to rat IL-6 and TNFa (Invitrogen Corporation, Carlsbad, CA). Samples, negative controls, and diluted IL-6 and TNFa standard markers were added to their respective wells. Detection of IL-6 and TNFa was performed according to the manufacturer’s instructions. Absorbance at 450 nm was determined using an ELISA microplate reader (BioRad Laboratories, Hercules, CA).
Isolated Heart Perfusion and Assessment of Cardiac Function Animal hearts were isolated 9 h after LPS administration. Cardiac function was determined using a modified isovolumetric Langendorff technique as previously described [13–15]. Cardiac function was represented in terms of left ventricular developed pressure (LVDP), left ventricular end-diastolic pressure (LVEDP), and increased left ventricular pressure development during isovolumetric contraction (LV dP/dtmax) and relaxation (LV dP/dtmin). The LVDP is a physiologic parameter indicating the ability of the heart to contract and develop pressure. It is defined as the difference between LV systolic pressure and LVEDP (the ability of the heart to relax) at the end of the relaxation phase of the heart. A positive inotropic effect indicates a higher force of the myocardial contraction. The 6 dp/dt values characterize the rates of change in heart contraction and relaxation. Hearts were rapidly excised while still beating into oxygenated Krebs-Henseleit perfusion solution containing 11 mM glucose, 2.0 mM CaCl2, 4.3 mM KCl, 25 mM NaHCO3, 118 mM NaCl, 1.2 mM MgSO4, and 1.2 mM KH2PO4. Normothermic retrograde perfusion was performed with the same solution in an isovolumetric and non-recirculating mode. The perfusion buffer was saturated with a gas mixture of 95% O2 and 5% CO2 at pH 7.4. Perfusion pressure was maintained at 75 mmHg. A latex balloon was inserted through the left atrium into the left ventricle, and the balloon was filled with water (0.18– 0.28 mL). LVDP, LVEDP, LV dP/dtmax, and LV dP/dtmin were continuously recorded using a computerized pressure amplifier-digitizer. A three-way stopcock was mounted above the aortic cannula to induce global ischemia. After 20 min of perfusion (equilibration), hearts were subjected to 30 min of normothermic global ischemia followed by 30 min of reperfusion in a perfusate-filled organ bath chamber. Myocardial temperature was maintained at 37 C by circulating warm water.
Statistical Analysis All data are presented as mean 6 SD. Blood pressure and heart rate data were analyzed using one-way ANOVA. Mann-Whitney’s U-test
253
HAGIWARA ET AL.: INSULIN TREATMENT ATTENUATES CARDIAC FUNCTION IN DIABETES
FIG. 1. Effects of hyperglycemia and insulin on heart histology in lipopolysaccharide-administered rats. Heart tissue obtained from the saline group (A) magnification, 3100; DM group (B) magnification, 3100; DMþIns group (C) magnification, 3100; LPS group (D) magnification, 3100; DMþLPS group (E) magnification, 3100; or DMþLPSþIns group (F) magnification, 3100, were stained with hematoxylin and eosin. DM ¼ diabetes mellitus; LPS ¼ lipopolysaccharide; Ins ¼ insulin. (Color version of figure is available online.)
was used for comparisons between two independent groups. A P value < 0.05 was considered statistically significant.
RESULTS
In the present study, all rats in all treatment groups survived.
groups (Fig. 1B and C). In contrast, heart tissue obtained 6 h after LPS administration (LPS, DMþLPS, and DMþLPSþIns groups) demonstrated reduced myocytes compared to controls. Histologic changes were more pronounced in the DMþLPS and DMþLPSþIns groups (Fig. 1D–F). Effect of Insulin Administration on Diabetes
Effect of Diabetes and Insulin Treatment on Cardiac Histology in the LPS-Induced Systemic Inflammation Model
No histologic changes were observed in heart tissue of the saline group (Figure 1A) or the DM and DMþIns
Table 1 shows changes in blood glucose concentrations in the six groups following LPS and insulin treatments. At the 3-h time point, blood glucose levels were significantly higher in the DM group compared to the
TABLE 1 Changes in Blood Glucose Levels Following Endotoxin Challenge and Insulin Infusion
DM DMþIns DMþLPS DMþLPSþIns Saline LPS
0h
3h
6h
9h
412.1 6 74.2* 405.9 6 71.5* 409.4 6 64.3* 413.4 6 66.9* 171.5 6 17.1 176.3 6 14.5
417.0 6 82.6y 175.9 6 52.5 402.9 6 44.4y 209.4 6 80.3 165.5 6 25.1 166.0 6 22.7
433.3 6 68.1z 176.1 6 30.4 474.7 6 95.3z 182.8 6 87.9 162.7 6 29.0 89.0 6 40.9z
437.3 6 62.2x 155.4 6 23.8 537.6 6 78.2x 171.9 6 90.8 162.5 6 26.9 66.3 6 29.4x
Blood glucose levels at the indicated time points are shown for each group (n ¼ 10 for each group). Saline group, 0.9 % NaCl solution treatment; DM group, streptozotocin-induced diabetes; DMþIns group, streptozotocin-induced diabetes and continuous insulin infusion (20 IU/kg/ h); LPS group, LPS administration (7.5 mg/kg); DMþLPS group, streptozotocin-induced diabetes and LPS administration (7.5 mg/kg); DMþLPSþIns group, streptozotocin-induced diabetes and LPS administration following the initiation of continuous insulin infusion (20 IU/ kg/h). All data are expressed as mean 6 SD. * P < 0.05 compared with the Saline group at 0 h. y P < 0.05 compared with the Saline group at 3 h. z P < 0.05 compared with the Saline group at 6 h. x P < 0.05 compared with the Saline group at 9 h.
254
JOURNAL OF SURGICAL RESEARCH: VOL. 171, NO. 1, NOVEMBER 2011
FIG. 2. Changes in serum tumor necrosis factor a levels after lipopolysaccharide administration. Sera were obtained from saline (black), DM (white), DMþIns (gray), LPS (hatched), DMþLPS (border), and DMþLPSþIns (dot) groups (each group, n ¼ 10) at 3 h (A) or 6 h (B) after LPS administration. TNFa levels were determined by ELISA. All data are expressed as mean 6 SD. *P < 0.05 compared with the LPS group. DM ¼ diabetes mellitus; STZ ¼ streptozotocin; LPS ¼ lipopolysaccharide; Ins ¼ insulin.
saline group. Insulin treatment decreased serum glucose levels in the DMþIns group to normal levels. In contrast, glucose levels were significantly lower than normal following LPS administration (LPS group). In the DMþLPS group, blood glucose levels were significantly increased 9 h after LPS administration. In the DMþLPSþIns group, glucose levels were approximately the same as the saline group 9 h after LPS administration. Effect of Diabetes and Insulin Treatment on Serum Levels of Nitrate/Nitrite, TNFa, and IL-6
TNFa levels were markedly increased after LPS administration (Fig. 2). The DMþLPS group exhibited significantly increased TNFa levels compared with the
LPS group, and insulin treatment was also associated with a significant increase in TNFa (DMþLPSþIns). Similarly, the cytokine IL-6 was considerably elevated after LPS administration in LPS, DMþLPS, and DMþLPSþIns groups (Fig. 3). Diabetes was also associated with increased serum IL-6 levels and diabetic rats treated with insulin, in particular, exhibited increased serum IL-6 levels compared to the other groups. Serum nitrate/nitrite levels also increased after LPS administration (Fig. 4). Nitrate/nitrite levels were in higher in the DMþLPS group compared with the LPS group. Moreover, this peak was significantly higher in rats co-treated with insulin (DMþLPSþIns). Serum levels of IL-6, TNFa, and nitrate/nitrite remained unchanged at all time points in saline, DM, and DMþIns groups.
HAGIWARA ET AL.: INSULIN TREATMENT ATTENUATES CARDIAC FUNCTION IN DIABETES
255
FIG. 3. Changes in serum interleukin-6 levels after lipopolysaccharide administration. Sera were obtained from saline (black), DM (white), DMþIns (gray), LPS (hatched), DMþLPS (border), and DMþLPSþIns (dot) groups (each group, n ¼ 10) at 3 h (A) or 6 h (B) after LPS administration; IL-6 levels were determined by ELISA. All data are expressed as mean 6 SD. *P < 0.05 compared with the LPS group. DM ¼ diabetes mellitus; STZ ¼ streptozotocin; LPS ¼ lipopolysaccharide; Ins ¼ insulin.
Effect of Diabetes and Insulin Treatment on Myocardial Function in the LPS-Induced Systemic Inflammation Model
LPS administration (LPS, DMþLPS, and DMþ LPSþIns groups) resulted in markedly increased LVEDP compared with the saline group (Fig. 5A; left panel). LVEDP was especially elevated in the DMþLPS and DMþLPSþIns groups. In contrast, LPS administration resulted in markedly decreased LVDP (Fig. 5B; left panel), LV dP/dtmax (Fig. 5C; left panel), and LV dP/dtmin (Fig. 5D; left panel) compared with the saline group, especially in diabetic rats. After the 30 min ischemic reperfusion injury, all cardiac function parameters were significantly worsened. Cardiac function in DM and DMþIns groups was similar to that of the saline group; however, after ischemic reperfusion,
these parameters deteriorated in diabetic rats compared with nondiabetic rats (Fig. 5A–D; right panel). All parameters (LVEDP, LVDP, LV dP/dtmax, and LV dP/dtmin) deteriorated in the insulin treatment (DMþLPSþIns) group compared with the DMþLPS group (Fig. 5A–D; right panel). DISCUSSION
In this study, we demonstrated that diabetes induced cardiac dysfunction and increased serum cytokine and NOx levels in a rat model of systemic inflammation. Furthermore, normalizing serum glucose levels with insulin further deteriorated cardiac dysfunction via increasing serum cytokine and NOx levels.
256
JOURNAL OF SURGICAL RESEARCH: VOL. 171, NO. 1, NOVEMBER 2011
FIG. 4. Temporal changes in serum nitrate/nitrite levels after lipopolysaccharide administration. Sera were obtained from saline (black), DM (white), DMþIns (gray), LPS (hatched), DMþLPS (border), and DMþLPSþIns (dot) groups (each group, n ¼ 10) at 3 h (A) or 6 h (B) after LPS administration; nitrate/nitrite levels were determined by Griess assay. All data are expressed as mean 6 SD. *P < 0.05 compared with the LPS group. DM ¼ diabetes mellitus; STZ ¼ streptozotocin; LPS ¼ lipopolysaccharide; Ins ¼ insulin.
Acute cardiac dysfunction associated with systemic inflammation is commonly seen in clinical practice; however, its pathology is poorly understood. Vascular inflammation, with increased production and release of inflammatory cytokines, has been shown to play a role in the pathogenesis of cardiac dysfunction. TNFa has a direct effect on cardiomyocytes, downregulates sarcoplasmic reticulum proteins, inhibits contractility, and induces apoptosis [16], and other inflammatory cytokines (e.g., IL-6) induce heart failure in acute cardiac dysfunction patients [12]. In this study, we demonstrated that diabetic rats exhibited increased serum cytokine levels compared to nondiabetic rats after LPS-induced systemic inflammation. Moreover, acute normalization of serum glucose levels by insulin treatment further increased serum cytokine levels
and reduced cardiac function after LPS administration in our diabetic rat model. Thus, cardiac dysfunction in diabetic rats may be related to elevated cytokine levels. These results suggest that acute normalization of serum glucose levels may be avoided in diabetic patients. NO is produced by endothelial cells and cardiomyocytes, and regulates cardiac function in several ways including the modulation of cardiac contractile function [17]. Increased NOx is observed in the myocardium and vasculature of animals and patients with heart failure [18]. NOx is a major mediator of myocardial injury in various pathophysiologic conditions, and its effective neutralization may have significant therapeutic benefits. In the present study, serum NOx levels did not differ between diabetic and nondiabetic rats; however, LPS-induced systemic inflammation increased serum
HAGIWARA ET AL.: INSULIN TREATMENT ATTENUATES CARDIAC FUNCTION IN DIABETES
257
FIG. 5. Effects of hyperglycemia and insulin treatment on heart function in the rat systemic inflammation model. Each panel shows pre-IR (left) and post-I/R (right, 30 min later) indices. Myocardial function in rats from the saline (black), DM (white), DMþIns (gray), LPS (hatched), DMþLPS (border), and DMþLPSþIns (dot) groups (n ¼ 10/group) was determined at the indicated time points. (A) LVEDP, (B) LVDP, (C) LV dP/dtmax, and (D) LV dP/dtmin. Data are presented as mean 6 SD. *P < 0.05 compared with the LPS group. DM ¼ diabetes mellitus; STZ ¼ streptozotocin; LPS ¼ lipopolysaccharide; Ins ¼ insulin; I/R ¼ ischemic reperfusion.
NOx levels in diabetic rats compared to nondiabetic rats, with particularly higher NOx levels in the insulin-treated diabetic group. Thus, diabetic patients may be especially susceptible to an increased inflammatory response. In 2002, van den Berghe et al. reported that intensive insulin therapy to maintain blood glucose at or below 110 mg per deciliter can reduce morbidity and mortality among critically ill patients [5]. In another study, insulin exerted an anti-inflammatory effect whereas glucose exerted a proinflammatory effect [19]. We previously demonstrated that acute hyperglycemia increased the inflammatory response and induced cardiac dysfunction. Moreover, normalizing serum glucose levels with insulin blocked these effects of glucose [3, 4]. In a recent study, systemic inflammation responses in diabetic rats were enhanced compared with normal rats in terms of acute hyperglycemia [20]. Some recent studies found that the use of intensive insulin therapy places critically ill patients with sepsis at increased risk of hypoglycemia, induces a shortage in the provision of energy, and is not associated with decreased hos-
pital mortality [21, 22]. In particular, Egi et al. found no clear association between hyperglycemia during the ICU stay and mortality in diabetic patients, and that hyperglycemia may have different biological and/or clinical implications in critical care patients with diabetes mellitus [23]. In other words, hyperglycemic responses in diabetic patients are very different from those in nondiabetic patients. In the present study, we demonstrated that diabetes induced serum cytokine and NOx levels, thereby inducing cardiac dysfunction via systemic inflammation and acute hyperglycemia. In contrast to the previous study, insulin increased serum cytokine and NOx levels in diabetic rats and worsened cardiac function compared to non-insulin-treated diabetic rats. These results suggest that diabetic rats respond to systemic inflammation differently than nondiabetic rats, and imply that strict glucose control should be avoided in uncontrolled diabetic patients. The present study has several limitations worth noting, including the short bleeding time of diabetic rats. In addition, the relatively small number of animals in each treatment group and the high levels of glucose
258
JOURNAL OF SURGICAL RESEARCH: VOL. 171, NO. 1, NOVEMBER 2011
induced by streptozotocin treatment may not correlate directly with the clinical conditions of ICU patients. Furthermore, animal model results should be interpreted cautiously due to potential differences between rodent and human diabetes and subsequent response to insulin treatment. The rat model of LPS-induced systemic inflammation is not completely analogous to clinical inflammation experienced by patients; however, LPS-induced inflammation is generally considered to adequately reproduce the full spectrum of inflammatory changes observed in patients with sepsis [24]. In conclusion, our results suggest that diabetes may deteriorate cardiac function in patients with systemic inflammatory conditions by increasing the inflammatory response. Normalization of serum glucose levels with insulin further increased inflammation and worsened cardiac function. We conclude that intensive glucose control may exacerbate the inflammatory response. ACKNOWLEDGMENTS The authors thank Hiroaki Kawazato and Aiko Yasuda for their helpful advice on hematoxylin and eosin staining.
REFERENCES 1. [WHO] World Health Organization: Diabetes facts. Available at: http://www.who.int/mediacentre/factsheets/fs312/en/index.html. Accessed February 23, 2008. 2. Gullo A, Iscra F, Di Capua G, et al. Sepsis and organ dysfunction: An ongoing challenge. Minerva Anestesiol 2005;71:671. 3. Hagiwara S, Iwasaka H, Hasegawa A, et al. Hyperglycemia contributes to cardiac dysfunction in a lipopolysaccharide-induced systemic inflammation model. Crit Care Med 2009;37:2223. 4. Hagiwara S, Iwasaka H, Hasegawa A, et al. Effects of hyperglycemia and insulin therapy on high mobility group box 1 in endotoxin-induced acute lung injury in a rat model. Crit Care Med 2008;36:2407. 5. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in the critically ill patients. N Engl J Med 2001; 345:1359. 6. Costa EL, Schettino IA, Schettino GP. The lung in sepsis: Guilty or innocent? Endocr Metab Immune Disord Drug Targets 2006; 6:213.
7. Rudiger A, Singer M. Mechanisms of sepsis-induced cardiac dysfunction. Crit Care Med 2007;35:1599. 8. Schulz R, Rassaf T, Massion PB, et al. Recent advances in the understanding of the role of nitric oxide in cardiovascular homeostasis. Pharmacol Ther 2005;108:225. 9. Lo´pez A, Lorente JA, Steingrub J, et al. Multiple-center, randomized, placebo-controlled, double-blind study of the nitric oxide synthase inhibitor 546C88: Effect on survival in patients with septic shock. Crit Care Med 2004;32:21. 10. Adamopoulos S, Parissis JT, Kremastinos DT. A glossary of circulating cytokines in chronic heart failure. Eur J Heart Fail 2001;3:517. 11. Torre-Amione G, Kapadia S, Lee J, et al. Tumor necrosis factora and tumor necrosis factor receptors in the failing human heart. Circulation 1996;93:704. 12. Tsutamoto T, Hisanaga T, Wada A, et al. Interleukin-6 spillover in the peripheral circulation increases with the severity of heart failure. J Am Coll Cardiol 1998;3:391. 13. Meng X, Brown JM, Ao L, et al. Endotoxin induces cardiac HSP70 and resistance to endotoxemic myocardial depression in rats. Am J Physiol 1996;271:C1316. 14. Brown MA, Jones WK. NF-kB action in sepsis: The innate immune system and the heart. Front Biosci 2004;9:1201. 15. Meng X, Brown JM, Ao L, et al. Norepinephrine induces cardiac heat shock protein 70 and delayed cardioprotection in the rat through 1 adrenoceptors. Cardiovasc Res 1996;32:374. 16. Mann DL. Recent insights into the role of tumor necrosis factor in the failing heart. Heart Fail Rev 2001;6:71. 17. Massion PB, Feron O, Dessy C, et al. Nitric oxide and cardiac function: Ten years after, and continuing. Circ Res 2003;93:388. 18. Chen Y, Traverse JH, Du R, et al. Nitric oxide modulates myocardial oxygen consumption in the failing heart. Circulation 2002;106:273. 19. Dandona P, Chaudhuri A, Ghanim H, et al. Insulin as an antiinflammatory and antiatherogenic modulator. J Am Coll Cardiol 2009;53:S14. 20. Hagiwara S, Iwasaka H, Shingu C, et al. The effect of experimental diabetes on high mobility group box 1 protein expression in endotoxin-induced acute lung injury. J Surg Res 2011;168:111. 21. Brunkhorst FM, Engel C, Bloos F, et al. Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med 2008;358:125. 22. Griesdale DE, de Souza RJ, van Dam RM, et al. Intensive insulin therapy and mortality among critically ill patients: A metaanalysis including NICE-SUGAR study data. CMAJ 2009; 180:821. 23. Egi M, Bellomo R, Stachowski E, et al. Blood glucose concentration and outcome of critical illness: The impact of diabetes. Crit Care Med 2008;36:2249. 24. Remick DG, Ward PA. Evaluation of endotoxin models for the study of sepsis. Shock 2005;24(Suppl 1):7.
Insulin Treatment of Diabetic Rats Reduces Cardiac Function in a Lipopolysaccharide-Induced Systemic Inflammation Model Satoshi Hagiwara, M.D., Ph.D.,*,1 Hideo Iwasaka, M.D., Ph.D.,* Kyosuke Kudo, M.D.,* Akira Hasegawa, M.D.,* Jyunya Kusaka, M.D.,* Tomohisa Uchida, M.D., Ph.D.,† and Takayuki Noguchi, M.D., Ph.D.* *Department of Anesthesiology and Intensive Care Medicine; and †Department of Pathology, Oita University Faculty of Medicine, Yufu City, Oita, Japan Submitted for publication January 8, 2010
INTRODUCTION Background. Diabetes is a common comorbidity in patients with various medical conditions. Tight glucose control is known to improve systemic inflammation; however, whether it is effective in diabetic patients is unknown. The purpose of this study was to examine how strict glucose control affects systemic inflammation in diabetic patients. Materials and Methods. Male Wistar rats. We determined the effect of insulin therapy on cardiac function in a rat model of systemic inflammation. We administered lipopolysaccharide intravenously, with or without insulin, to streptozotocin-induced diabetic rats. After induction of systemic inflammation, we determined serum cytokine (IL-6 and TNFa) and nitrate/ nitrite levels and measured cardiac function. Results. Cytokine levels and cardiac function were significantly reduced in diabetic rats compared to non-diabetic rats. Moreover, insulin treatment was associated with higher cytokine levels and decreased cardiac function. Conclusion. In systemic inflammatory conditions, diabetes increases various proinflammatory mediators and inhibits cardiac function; insulin treatment exacerbates these effects. Thus, strict glucose control may not be desirable in diabetic patients with systemic inflammatory conditions. Ó 2011 Elsevier Inc. All rights reserved. Key Words: inflammation; sepsis; diabetes; insulin; lipopolysaccharide; cardiac dysfunction.
1 To whom correspondence and reprint requests should be addressed at Department of Anesthesiology and Intensive Care Medicine, Oita University Faculty of Medicine, 1-1 Idaigaoka Hasamamachi, Yufu City, Oita 879-5593, Japan. E-mail: saku@med. oita-u.ac.jp.
The World Health Organization estimates that approximately 180 million individuals are affected by diabetes worldwide, and this number is expected to double by the year 2030 [1]. Infection rates are also growing, and sepsis and systemic inflammation are significant problems in intensive care units [2]. In particular, untreated diabetic patients with severe infections are often seen in intensive care units. Hyperglycemia can often exacerbate inflammation. In a previous study, we found that acute hyperglycemia worsened lung injury and cardiac function after systemic inflammation was induced by LPS [3, 4]. However, the difference between acute hyperglycemia and chronic hyperglycemia due to diabetes is not well understood. Van den Berghe et al. reported that intensive insulin therapy to maintain blood glucose at or below 110 mg per dL reduces morbidity and mortality among critically ill patients [5]. The anti-inflammatory action of insulin is one of the reasons for this effect. Tight glucose control also reduces cytokine levels in people without diabetes [3, 4]. However, the mechanism by which serum glucose affects systemic inflammation is unknown. Cardiac dysfunction is common in intensive care units [6] and is clearly associated with many cases of septic shock and systemic inflammation [7]. Cardiac dysfunction is a life-threatening condition with a high rate of mortality. Nitric oxide (NO), which exerts numerous biological effects in the cardiovascular system, is an important factor in cardiac dysfunction [8]. The inhibition of inducible nitric oxide synthase (iNOS) reduces NO production in cardiac myocytes, thereby restoring contractile function and improving outcomes in patients with septic shock [9].
251
0022-4804/$36.00 Ó 2011 Elsevier Inc. All rights reserved.
252
JOURNAL OF SURGICAL RESEARCH: VOL. 171, NO. 1, NOVEMBER 2011
Cytokines are also important mediators in the progression of acute cardiac dysfunction. Tumor necrosis factor alpha (TNFa) plays a key role in the induction of NO synthase [10]. In addition, acute exposure of isolated muscles to TNFa significantly reduces specific muscle force without a loss in muscle mass or protein levels [11]. Interleukin (IL)-6, a multifunctional cytokine, is also associated with the progression of cardiac dysfunction [12]. In this study, we hypothesized that strict glucose control would affect the response to systemic inflammation in diabetic patients. To this end, we investigated whether strict glucose control in diabetic rats affected levels of systemic inflammation markers, such as serum cytokines and NO, and cardiac function.
Histological Analysis Hearts (n ¼ 4 in each group) were obtained from animals under 4% sevoflurane anesthesia. Heart tissue specimens were fixed with 10% formalin and embedded in paraffin. Sections were stained with hematoxylin and eosin. A pathologist blinded to group assignment analyzed the samples.
Nitrate/Nitrite Determination In biological fluids, nitric oxide (NO) is rapidly converted to nitrite and nitrate. Therefore, we determined NO levels after serum collection by cardiac puncture by measuring nitrite and nitrate (NOx) levels with commercial kits according to the manufacturer’s instructions (R and D Systems Inc., Minneapolis, MN). NOx levels were determined by a modified Griess assay and absorbance was measured at 540 nm. Nitrite and nitrate concentrations were calculated using a standard curve and expressed as mmol/L.
Cytokine Measurements MATERIALS AND METHODS Insulin (Ins) was obtained from Novo Nordisk A/S (Copenhagen, Denmark). Lipopolysaccharide (LPS, O127:B8) was obtained from Sigma (St. Louis, MO). All other reagents were of the highest available analytical grade.
Animals All protocols conformed to the National Institute of Health (NIH) guidelines and all animals received humane care in compliance with the Principles of Laboratory Animal Care. The study was approved by the Ethics Committee of Animal Research at the College of Medicine, Oita University, Oita, Japan. Male Wistar rats weighing 250 to 300 g (Kyudou, Saga, Japan) were used in all experiments. Rats were given free access to food and water. Diabetes was induced with streptozotocin and a model for systemic inflammation was generated using LPS. Animals were randomly assigned to one of six groups (n ¼ 20 in each group) and received treatments intravenously: (1) saline group: rats received 0.9 % NaCl solution on d 1 of the experiment and 4 wk later; (2) diabetes mellitus (DM) group: rats received 60 mg/kg streptozotocin, and 4 wk later received 0.9 % NaCl solution; (3) DMþIns group: rats received 60 mg/kg streptozotocin on d 1, and 4 wk later received a bolus injection of 0.9 % NaCl solution 3 h after starting continuous insulin infusion (20 IU/kg/ h); (4) LPS group: rats received 0.9 % NaCl solution, and 4 wk later received LPS (7.5 mg/kg); (5) DMþLPS group: rats received 60 mg/ kg streptozotocin, and 4 wk later received LPS (7.5 mg/kg); (6) DMþLPSþIns group: rats received 60 mg/kg streptozotocin, and 4 wk later received a bolus injection of LPS (7.5 mg/kg) 3 h after starting continuous insulin infusion (20 IU/kg/h). Samples of venous blood (0.3 mL) were obtained from the external jugular vein at various time points (0, 3, 6, and 9 h). Anesthesia was maintained by 4% sevoflurane (Maruishi Pharmaceutical Co., Osaka, Japan). All animals were sacrificed at 9 h while under deep anesthesia using pentobarbital sodium. Heart tissue specimens were quickly removed and processed as indicated below.
Blood Glucose Measurement Blood samples (n ¼ 6 in each group) were collected from the external jugular vein at each time point. Whole blood glucose levels were determined using the glucose oxidase method. The Glu-test-sensor blood glucose monitoring system was purchased from Sanwa Kagaku Kenkyusho Co., Ltd. (Nagoya, Japan).
Serum samples (n ¼ 6 in each group) were assayed using the sandwich enzyme-linked immunosorbent assay (ELISA) method. Ninetysix-well plates were precoated with monoclonal antibodies specific to rat IL-6 and TNFa (Invitrogen Corporation, Carlsbad, CA). Samples, negative controls, and diluted IL-6 and TNFa standard markers were added to their respective wells. Detection of IL-6 and TNFa was performed according to the manufacturer’s instructions. Absorbance at 450 nm was determined using an ELISA microplate reader (BioRad Laboratories, Hercules, CA).
Isolated Heart Perfusion and Assessment of Cardiac Function Animal hearts were isolated 9 h after LPS administration. Cardiac function was determined using a modified isovolumetric Langendorff technique as previously described [13–15]. Cardiac function was represented in terms of left ventricular developed pressure (LVDP), left ventricular end-diastolic pressure (LVEDP), and increased left ventricular pressure development during isovolumetric contraction (LV dP/dtmax) and relaxation (LV dP/dtmin). The LVDP is a physiologic parameter indicating the ability of the heart to contract and develop pressure. It is defined as the difference between LV systolic pressure and LVEDP (the ability of the heart to relax) at the end of the relaxation phase of the heart. A positive inotropic effect indicates a higher force of the myocardial contraction. The 6 dp/dt values characterize the rates of change in heart contraction and relaxation. Hearts were rapidly excised while still beating into oxygenated Krebs-Henseleit perfusion solution containing 11 mM glucose, 2.0 mM CaCl2, 4.3 mM KCl, 25 mM NaHCO3, 118 mM NaCl, 1.2 mM MgSO4, and 1.2 mM KH2PO4. Normothermic retrograde perfusion was performed with the same solution in an isovolumetric and non-recirculating mode. The perfusion buffer was saturated with a gas mixture of 95% O2 and 5% CO2 at pH 7.4. Perfusion pressure was maintained at 75 mmHg. A latex balloon was inserted through the left atrium into the left ventricle, and the balloon was filled with water (0.18– 0.28 mL). LVDP, LVEDP, LV dP/dtmax, and LV dP/dtmin were continuously recorded using a computerized pressure amplifier-digitizer. A three-way stopcock was mounted above the aortic cannula to induce global ischemia. After 20 min of perfusion (equilibration), hearts were subjected to 30 min of normothermic global ischemia followed by 30 min of reperfusion in a perfusate-filled organ bath chamber. Myocardial temperature was maintained at 37 C by circulating warm water.
Statistical Analysis All data are presented as mean 6 SD. Blood pressure and heart rate data were analyzed using one-way ANOVA. Mann-Whitney’s U-test
253
HAGIWARA ET AL.: INSULIN TREATMENT ATTENUATES CARDIAC FUNCTION IN DIABETES
FIG. 1. Effects of hyperglycemia and insulin on heart histology in lipopolysaccharide-administered rats. Heart tissue obtained from the saline group (A) magnification, 3100; DM group (B) magnification, 3100; DMþIns group (C) magnification, 3100; LPS group (D) magnification, 3100; DMþLPS group (E) magnification, 3100; or DMþLPSþIns group (F) magnification, 3100, were stained with hematoxylin and eosin. DM ¼ diabetes mellitus; LPS ¼ lipopolysaccharide; Ins ¼ insulin. (Color version of figure is available online.)
was used for comparisons between two independent groups. A P value < 0.05 was considered statistically significant.
RESULTS
In the present study, all rats in all treatment groups survived.
groups (Fig. 1B and C). In contrast, heart tissue obtained 6 h after LPS administration (LPS, DMþLPS, and DMþLPSþIns groups) demonstrated reduced myocytes compared to controls. Histologic changes were more pronounced in the DMþLPS and DMþLPSþIns groups (Fig. 1D–F). Effect of Insulin Administration on Diabetes
Effect of Diabetes and Insulin Treatment on Cardiac Histology in the LPS-Induced Systemic Inflammation Model
No histologic changes were observed in heart tissue of the saline group (Figure 1A) or the DM and DMþIns
Table 1 shows changes in blood glucose concentrations in the six groups following LPS and insulin treatments. At the 3-h time point, blood glucose levels were significantly higher in the DM group compared to the
TABLE 1 Changes in Blood Glucose Levels Following Endotoxin Challenge and Insulin Infusion
DM DMþIns DMþLPS DMþLPSþIns Saline LPS
0h
3h
6h
9h
412.1 6 74.2* 405.9 6 71.5* 409.4 6 64.3* 413.4 6 66.9* 171.5 6 17.1 176.3 6 14.5
417.0 6 82.6y 175.9 6 52.5 402.9 6 44.4y 209.4 6 80.3 165.5 6 25.1 166.0 6 22.7
433.3 6 68.1z 176.1 6 30.4 474.7 6 95.3z 182.8 6 87.9 162.7 6 29.0 89.0 6 40.9z
437.3 6 62.2x 155.4 6 23.8 537.6 6 78.2x 171.9 6 90.8 162.5 6 26.9 66.3 6 29.4x
Blood glucose levels at the indicated time points are shown for each group (n ¼ 10 for each group). Saline group, 0.9 % NaCl solution treatment; DM group, streptozotocin-induced diabetes; DMþIns group, streptozotocin-induced diabetes and continuous insulin infusion (20 IU/kg/ h); LPS group, LPS administration (7.5 mg/kg); DMþLPS group, streptozotocin-induced diabetes and LPS administration (7.5 mg/kg); DMþLPSþIns group, streptozotocin-induced diabetes and LPS administration following the initiation of continuous insulin infusion (20 IU/ kg/h). All data are expressed as mean 6 SD. * P < 0.05 compared with the Saline group at 0 h. y P < 0.05 compared with the Saline group at 3 h. z P < 0.05 compared with the Saline group at 6 h. x P < 0.05 compared with the Saline group at 9 h.
254
JOURNAL OF SURGICAL RESEARCH: VOL. 171, NO. 1, NOVEMBER 2011
FIG. 2. Changes in serum tumor necrosis factor a levels after lipopolysaccharide administration. Sera were obtained from saline (black), DM (white), DMþIns (gray), LPS (hatched), DMþLPS (border), and DMþLPSþIns (dot) groups (each group, n ¼ 10) at 3 h (A) or 6 h (B) after LPS administration. TNFa levels were determined by ELISA. All data are expressed as mean 6 SD. *P < 0.05 compared with the LPS group. DM ¼ diabetes mellitus; STZ ¼ streptozotocin; LPS ¼ lipopolysaccharide; Ins ¼ insulin.
saline group. Insulin treatment decreased serum glucose levels in the DMþIns group to normal levels. In contrast, glucose levels were significantly lower than normal following LPS administration (LPS group). In the DMþLPS group, blood glucose levels were significantly increased 9 h after LPS administration. In the DMþLPSþIns group, glucose levels were approximately the same as the saline group 9 h after LPS administration. Effect of Diabetes and Insulin Treatment on Serum Levels of Nitrate/Nitrite, TNFa, and IL-6
TNFa levels were markedly increased after LPS administration (Fig. 2). The DMþLPS group exhibited significantly increased TNFa levels compared with the
LPS group, and insulin treatment was also associated with a significant increase in TNFa (DMþLPSþIns). Similarly, the cytokine IL-6 was considerably elevated after LPS administration in LPS, DMþLPS, and DMþLPSþIns groups (Fig. 3). Diabetes was also associated with increased serum IL-6 levels and diabetic rats treated with insulin, in particular, exhibited increased serum IL-6 levels compared to the other groups. Serum nitrate/nitrite levels also increased after LPS administration (Fig. 4). Nitrate/nitrite levels were in higher in the DMþLPS group compared with the LPS group. Moreover, this peak was significantly higher in rats co-treated with insulin (DMþLPSþIns). Serum levels of IL-6, TNFa, and nitrate/nitrite remained unchanged at all time points in saline, DM, and DMþIns groups.
HAGIWARA ET AL.: INSULIN TREATMENT ATTENUATES CARDIAC FUNCTION IN DIABETES
255
FIG. 3. Changes in serum interleukin-6 levels after lipopolysaccharide administration. Sera were obtained from saline (black), DM (white), DMþIns (gray), LPS (hatched), DMþLPS (border), and DMþLPSþIns (dot) groups (each group, n ¼ 10) at 3 h (A) or 6 h (B) after LPS administration; IL-6 levels were determined by ELISA. All data are expressed as mean 6 SD. *P < 0.05 compared with the LPS group. DM ¼ diabetes mellitus; STZ ¼ streptozotocin; LPS ¼ lipopolysaccharide; Ins ¼ insulin.
Effect of Diabetes and Insulin Treatment on Myocardial Function in the LPS-Induced Systemic Inflammation Model
LPS administration (LPS, DMþLPS, and DMþ LPSþIns groups) resulted in markedly increased LVEDP compared with the saline group (Fig. 5A; left panel). LVEDP was especially elevated in the DMþLPS and DMþLPSþIns groups. In contrast, LPS administration resulted in markedly decreased LVDP (Fig. 5B; left panel), LV dP/dtmax (Fig. 5C; left panel), and LV dP/dtmin (Fig. 5D; left panel) compared with the saline group, especially in diabetic rats. After the 30 min ischemic reperfusion injury, all cardiac function parameters were significantly worsened. Cardiac function in DM and DMþIns groups was similar to that of the saline group; however, after ischemic reperfusion,
these parameters deteriorated in diabetic rats compared with nondiabetic rats (Fig. 5A–D; right panel). All parameters (LVEDP, LVDP, LV dP/dtmax, and LV dP/dtmin) deteriorated in the insulin treatment (DMþLPSþIns) group compared with the DMþLPS group (Fig. 5A–D; right panel). DISCUSSION
In this study, we demonstrated that diabetes induced cardiac dysfunction and increased serum cytokine and NOx levels in a rat model of systemic inflammation. Furthermore, normalizing serum glucose levels with insulin further deteriorated cardiac dysfunction via increasing serum cytokine and NOx levels.
256
JOURNAL OF SURGICAL RESEARCH: VOL. 171, NO. 1, NOVEMBER 2011
FIG. 4. Temporal changes in serum nitrate/nitrite levels after lipopolysaccharide administration. Sera were obtained from saline (black), DM (white), DMþIns (gray), LPS (hatched), DMþLPS (border), and DMþLPSþIns (dot) groups (each group, n ¼ 10) at 3 h (A) or 6 h (B) after LPS administration; nitrate/nitrite levels were determined by Griess assay. All data are expressed as mean 6 SD. *P < 0.05 compared with the LPS group. DM ¼ diabetes mellitus; STZ ¼ streptozotocin; LPS ¼ lipopolysaccharide; Ins ¼ insulin.
Acute cardiac dysfunction associated with systemic inflammation is commonly seen in clinical practice; however, its pathology is poorly understood. Vascular inflammation, with increased production and release of inflammatory cytokines, has been shown to play a role in the pathogenesis of cardiac dysfunction. TNFa has a direct effect on cardiomyocytes, downregulates sarcoplasmic reticulum proteins, inhibits contractility, and induces apoptosis [16], and other inflammatory cytokines (e.g., IL-6) induce heart failure in acute cardiac dysfunction patients [12]. In this study, we demonstrated that diabetic rats exhibited increased serum cytokine levels compared to nondiabetic rats after LPS-induced systemic inflammation. Moreover, acute normalization of serum glucose levels by insulin treatment further increased serum cytokine levels
and reduced cardiac function after LPS administration in our diabetic rat model. Thus, cardiac dysfunction in diabetic rats may be related to elevated cytokine levels. These results suggest that acute normalization of serum glucose levels may be avoided in diabetic patients. NO is produced by endothelial cells and cardiomyocytes, and regulates cardiac function in several ways including the modulation of cardiac contractile function [17]. Increased NOx is observed in the myocardium and vasculature of animals and patients with heart failure [18]. NOx is a major mediator of myocardial injury in various pathophysiologic conditions, and its effective neutralization may have significant therapeutic benefits. In the present study, serum NOx levels did not differ between diabetic and nondiabetic rats; however, LPS-induced systemic inflammation increased serum
HAGIWARA ET AL.: INSULIN TREATMENT ATTENUATES CARDIAC FUNCTION IN DIABETES
257
FIG. 5. Effects of hyperglycemia and insulin treatment on heart function in the rat systemic inflammation model. Each panel shows pre-IR (left) and post-I/R (right, 30 min later) indices. Myocardial function in rats from the saline (black), DM (white), DMþIns (gray), LPS (hatched), DMþLPS (border), and DMþLPSþIns (dot) groups (n ¼ 10/group) was determined at the indicated time points. (A) LVEDP, (B) LVDP, (C) LV dP/dtmax, and (D) LV dP/dtmin. Data are presented as mean 6 SD. *P < 0.05 compared with the LPS group. DM ¼ diabetes mellitus; STZ ¼ streptozotocin; LPS ¼ lipopolysaccharide; Ins ¼ insulin; I/R ¼ ischemic reperfusion.
NOx levels in diabetic rats compared to nondiabetic rats, with particularly higher NOx levels in the insulin-treated diabetic group. Thus, diabetic patients may be especially susceptible to an increased inflammatory response. In 2002, van den Berghe et al. reported that intensive insulin therapy to maintain blood glucose at or below 110 mg per deciliter can reduce morbidity and mortality among critically ill patients [5]. In another study, insulin exerted an anti-inflammatory effect whereas glucose exerted a proinflammatory effect [19]. We previously demonstrated that acute hyperglycemia increased the inflammatory response and induced cardiac dysfunction. Moreover, normalizing serum glucose levels with insulin blocked these effects of glucose [3, 4]. In a recent study, systemic inflammation responses in diabetic rats were enhanced compared with normal rats in terms of acute hyperglycemia [20]. Some recent studies found that the use of intensive insulin therapy places critically ill patients with sepsis at increased risk of hypoglycemia, induces a shortage in the provision of energy, and is not associated with decreased hos-
pital mortality [21, 22]. In particular, Egi et al. found no clear association between hyperglycemia during the ICU stay and mortality in diabetic patients, and that hyperglycemia may have different biological and/or clinical implications in critical care patients with diabetes mellitus [23]. In other words, hyperglycemic responses in diabetic patients are very different from those in nondiabetic patients. In the present study, we demonstrated that diabetes induced serum cytokine and NOx levels, thereby inducing cardiac dysfunction via systemic inflammation and acute hyperglycemia. In contrast to the previous study, insulin increased serum cytokine and NOx levels in diabetic rats and worsened cardiac function compared to non-insulin-treated diabetic rats. These results suggest that diabetic rats respond to systemic inflammation differently than nondiabetic rats, and imply that strict glucose control should be avoided in uncontrolled diabetic patients. The present study has several limitations worth noting, including the short bleeding time of diabetic rats. In addition, the relatively small number of animals in each treatment group and the high levels of glucose
258
JOURNAL OF SURGICAL RESEARCH: VOL. 171, NO. 1, NOVEMBER 2011
induced by streptozotocin treatment may not correlate directly with the clinical conditions of ICU patients. Furthermore, animal model results should be interpreted cautiously due to potential differences between rodent and human diabetes and subsequent response to insulin treatment. The rat model of LPS-induced systemic inflammation is not completely analogous to clinical inflammation experienced by patients; however, LPS-induced inflammation is generally considered to adequately reproduce the full spectrum of inflammatory changes observed in patients with sepsis [24]. In conclusion, our results suggest that diabetes may deteriorate cardiac function in patients with systemic inflammatory conditions by increasing the inflammatory response. Normalization of serum glucose levels with insulin further increased inflammation and worsened cardiac function. We conclude that intensive glucose control may exacerbate the inflammatory response. ACKNOWLEDGMENTS The authors thank Hiroaki Kawazato and Aiko Yasuda for their helpful advice on hematoxylin and eosin staining.
REFERENCES 1. [WHO] World Health Organization: Diabetes facts. Available at: http://www.who.int/mediacentre/factsheets/fs312/en/index.html. Accessed February 23, 2008. 2. Gullo A, Iscra F, Di Capua G, et al. Sepsis and organ dysfunction: An ongoing challenge. Minerva Anestesiol 2005;71:671. 3. Hagiwara S, Iwasaka H, Hasegawa A, et al. Hyperglycemia contributes to cardiac dysfunction in a lipopolysaccharide-induced systemic inflammation model. Crit Care Med 2009;37:2223. 4. Hagiwara S, Iwasaka H, Hasegawa A, et al. Effects of hyperglycemia and insulin therapy on high mobility group box 1 in endotoxin-induced acute lung injury in a rat model. Crit Care Med 2008;36:2407. 5. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in the critically ill patients. N Engl J Med 2001; 345:1359. 6. Costa EL, Schettino IA, Schettino GP. The lung in sepsis: Guilty or innocent? Endocr Metab Immune Disord Drug Targets 2006; 6:213.
7. Rudiger A, Singer M. Mechanisms of sepsis-induced cardiac dysfunction. Crit Care Med 2007;35:1599. 8. Schulz R, Rassaf T, Massion PB, et al. Recent advances in the understanding of the role of nitric oxide in cardiovascular homeostasis. Pharmacol Ther 2005;108:225. 9. Lo´pez A, Lorente JA, Steingrub J, et al. Multiple-center, randomized, placebo-controlled, double-blind study of the nitric oxide synthase inhibitor 546C88: Effect on survival in patients with septic shock. Crit Care Med 2004;32:21. 10. Adamopoulos S, Parissis JT, Kremastinos DT. A glossary of circulating cytokines in chronic heart failure. Eur J Heart Fail 2001;3:517. 11. Torre-Amione G, Kapadia S, Lee J, et al. Tumor necrosis factora and tumor necrosis factor receptors in the failing human heart. Circulation 1996;93:704. 12. Tsutamoto T, Hisanaga T, Wada A, et al. Interleukin-6 spillover in the peripheral circulation increases with the severity of heart failure. J Am Coll Cardiol 1998;3:391. 13. Meng X, Brown JM, Ao L, et al. Endotoxin induces cardiac HSP70 and resistance to endotoxemic myocardial depression in rats. Am J Physiol 1996;271:C1316. 14. Brown MA, Jones WK. NF-kB action in sepsis: The innate immune system and the heart. Front Biosci 2004;9:1201. 15. Meng X, Brown JM, Ao L, et al. Norepinephrine induces cardiac heat shock protein 70 and delayed cardioprotection in the rat through 1 adrenoceptors. Cardiovasc Res 1996;32:374. 16. Mann DL. Recent insights into the role of tumor necrosis factor in the failing heart. Heart Fail Rev 2001;6:71. 17. Massion PB, Feron O, Dessy C, et al. Nitric oxide and cardiac function: Ten years after, and continuing. Circ Res 2003;93:388. 18. Chen Y, Traverse JH, Du R, et al. Nitric oxide modulates myocardial oxygen consumption in the failing heart. Circulation 2002;106:273. 19. Dandona P, Chaudhuri A, Ghanim H, et al. Insulin as an antiinflammatory and antiatherogenic modulator. J Am Coll Cardiol 2009;53:S14. 20. Hagiwara S, Iwasaka H, Shingu C, et al. The effect of experimental diabetes on high mobility group box 1 protein expression in endotoxin-induced acute lung injury. J Surg Res 2011;168:111. 21. Brunkhorst FM, Engel C, Bloos F, et al. Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med 2008;358:125. 22. Griesdale DE, de Souza RJ, van Dam RM, et al. Intensive insulin therapy and mortality among critically ill patients: A metaanalysis including NICE-SUGAR study data. CMAJ 2009; 180:821. 23. Egi M, Bellomo R, Stachowski E, et al. Blood glucose concentration and outcome of critical illness: The impact of diabetes. Crit Care Med 2008;36:2249. 24. Remick DG, Ward PA. Evaluation of endotoxin models for the study of sepsis. Shock 2005;24(Suppl 1):7.