ASAIO Journal 2014
Improvement in Blood Glucose Control in Patients With Diabetes After Implantation of Left Ventricular Assist Devices Maya Guglin,* Kim Maguire,† Taylor Missimer,* Cristiano Faber,‡ and Christiano Caldeira‡
Some reports suggest that the course of diabetes mellitus (DM) in heart failure (HF) may improve after implantation of left ventricular assist devices (LVADs). The objective of our study was to explore longitudinal changes in glycosylated hemoglobin (HbA1C) in patients with diabetes mellitus with advanced HF post-LVAD implantation. We retrospectively reviewed the records of all patients who received LVADs at our institution between 2002 and 2012 and selected those who 1) survived posthospital discharge and 2) had DM. We collected data on HbA1C before and after implantation of LVADs, daily doses of insulin, and antidiabetic drugs. Comparisons were done using Student’s t-test. A total of 50 patients met the inclusion criteria. HbA1C was 7.6 ± 1.6 before LVAD, 5.7 ± 0.9 within 3 months after the LVAD implant (p = 0.0001), 6.1 ± 1.0 (p = 0.004 in comparison with pre-LVAD level) in 3–6 months after the implant, 6.3 ± 1.0 (p = 0.01) in 6–9 months, and 5.3 ± 0.1 in 9–12 months (p = 0.002). There were no significant changes in body mass index. Favorable changes in clinical course of diabetes in patients with HF occur after the implantation of LVADs, persist for at least 1 year after the implant, and are likely associated with improved hemodynamics and metabolism after normalization of cardiac output. ASAIO Journal 2014; 60:290–293.
myocardial inflammatory mediators such as interleukin-2 or tumor necrosis factor.6 Not surprisingly, function of end-organ systems usually suffering in chronic HF, such as kidneys and liver, also improves.7,8 Surprisingly, function of pancreatic gland, also exposed to both chronic congestion and low output in HF, remains poorly studied. Meanwhile, status of increased insulin resistance is well known in HF, and the combination of HF and diabetes mellitus (DM) is extremely common and linked to poor outcomes.9,10 We hypothesized that function of the pancreatic gland should improve after LVAD implantation and retrospectively studied blood glucose control in patients with DM before and after LVAD implantation. Methods We retrospectively reviewed the records of all patients who underwent LVAD implant at the Tampa General Hospital program from 2002 till the end of 2012. Patients were included in our study if they had the diagnosis of DM, confirmed by being on diabetic diet, oral hypoglycemic agents, insulin, or combination of the above or abnormal fasting glucose or increased glycosylated hemoglobin (HbA1C). Patients were excluded if they had LVAD explantation, death, or heart transplantation within 3 months of LVAD implant. We also excluded those who did not have documented values of HbA1C. We collected baseline demographic, echocardiographic, laboratory parameters, and hemodynamic information from right heart catheterization before LVAD implantation. Similar data, except echocardiographic measurements which were not consistently available, were then collected from the time intervals up to 3 months after LVAD implant and 3–6 months after LVAD implant. Information about oral diabetic agents and insulin requirements (units per day) at various points in time after LVAD implant was analyzed. All laboratory data from the follow-up period were taken from outpatient’s office visits when patients were in stable conditions. Data on right heart catheterization after LVAD were also used. We collected all HbA1C level data up to 12 months after LVAD implant. The values before and after LVAD implant were compared using Student’s t-test. p values less than 0.05 were considered significant. The analysis was performed on SPSS software (SPSS version 21; SPSS Inc., Chicago, IL).
Key Words: heart failure, diabetes mellitus, left ventricular assist device, mechanical assisted circulation
In advanced systolic heart failure (HF), implantation of the
left ventricular assist device (LVAD) normalizes hemodynamics and results in profound favorable changes in multiple organs and systems. In the heart itself, there is a decrease in left ventricular dimensions, increase in left ventricular ejection fraction, and regression of cardiomyocyte hypertrophy.1 Hemodynamically, there is an increase in cardiac output and decrease in pulmonary capillary wedge pressure2 and pulmonary vascular resistance,3,4 resulting in improved right ventricular structure and function. Biochemically, there is a decrease in plasma epinephrine, norepinephrine, arginine vasopressin, renin, and angiotensin II,5 as well as in circulating and From the *Department of Cardiology, University of South Florida, Tampa, Florida; †Department of Medicine, University of South Florida, Tampa, Florida; and ‡Advanced Cardiothoracic Surgery and Tampa General Hospital, Tampa, Florida. Submitted for consideration December 24, 2013; accepted for publication in revised form February 5, 2014. Disclosures: The authors have no conflicts of interest to report. Reprint Requests: Maya Guglin, MD, PhD, FACC, Department of Cardiology, University of South Florida, 2 Tampa General Circle, Tampa, FL 33606. Email: [email protected]. Copyright © 2014 by the American Society for Artificial Internal Organs
Results Between 2005 and the end of 2012, the Tampa General Hospital transplant program implanted 317 LVADs. After we excluded patients who received short-term support only including extracorporeal membranous oxygenation (49) or
DOI: 10.1097/MAT.0000000000000064
290
291
DIABETES IMPROVES AFTER LVAD
pump exchanges (14), we had 254 patients who underwent a long-term pump placement and patients who received heart transplantation during same admission or died within 3 months of LVAD placement. Out of them, 81 patients had a concomitant diagnosis of DM. All diabetic patients had Type 2 diabetes. After we excluded patients whose diabetes status could not confirm because of the lack of information on their diabetic medications or insulin and normal laboratory values of glucose or HbA1C (11), and patients who had heart transplant within same hospital admission as LVAD placement or died within 3 months after the LVAD implant (13), we had 57 patients. In seven of them, there were no recorded values of HbA1C, and they were excluded as well. The remaining 50 patients comprised our study population. During the year after the implant, two of them died and 12 more underwent successful heart transplantation. The data on their diabetes management and laboratory values after the transplantation were not collected. Remaining 36 patients were still on LVAD support 12 months later. The baseline characteristics are summarized in Table 1. Majority were males (35 [70%]) with ischemic cardiomyopathy (31 [62%]). There were six patients who received HeartMate I, six who received HeartWare, and the remaining 38 patients received HeartMate II. Before the LVAD, 22 patients (43.0%) were on insulin, with the mean daily dose of 66.2 ± 66.0 units, and 24 (48%) were taking oral diabetic medications. In 3 months, 16 (32.0%) remained on insulin, requiring 39.7 ± 44 units daily, and eight (16.0%) were on oral hypoglycemic agents. In 6 months, there were 18 patients (36.0%) on insulin with the requirement of 38.7 ± 46.0 units a day, and 11 (22.0%) were on oral agents. The difference was significant only for the proportion of patients on oral diabetic agents before and after LVAD, p = 0.012 and p = 0.020 for 3 and 6 months after the LVAD, respectively, comparing with the pre-LVAD level. In terms of HbA1C, it was 7.6 ± 1.6 before LVAD, 5.7 ± 0.9 within 3 months after the LVAD implant (p = 0.0001), 6.1 ± 1.0 Table 1. Baseline Characteristics Before LVAD Age, years (SD) Men, % Ischemic cardiomyopathy, % HeartMate I, % HeartMate II, % HeartWare,% Body mass index, kg/m2 On insulin,% Insulin, units/day HbA1C, % Left ventricular ejection fraction, % Peak velocity of tricuspid regurgitation, m/sec Left ventricular end-diastolic dimension, mm Sodium, mmol/l Creatinine, mg/dl Total bilirubin, mg/dl Alanine aminotransferase, units/l Aspartate aminotransferase, units/l
56.7 (12.1) 35 (70.0) 31 (62.0) 6 (12.0) 38 (76.0) 6 (12.0) 29.7 (5.5) 22 (43.1) 66.2 (66.0) 7.6 (1.6) 15.2 (6.5) 2.7 (0.7) 69.5 (13.9) 135 (4.0) 1.6 (1.0) 1.4 (0.9) 61.8 (164.5) 44.3 (56.2)
HbA1C, glycosylated hemoglobin; LVAD, left ventricular assist device; SD, standard deviation.
(p = 0.004 in comparison with pre-LVAD level) in 3–6 months after the implant, 6.3 ± 1.0 (p = 0.01) in 6–9 months, and 5.3 ± 0.1 in 9–12 months (p = 0.002) (Figure 1). From other parameters, pulmonary arterial systolic pressure, pulmonary capillary wedge pressure, cardiac output, and liver function tests, particularly total bilirubin, as well as HbA1C, improved within 3 months after LVAD implant and remained at a better level than before the surgery in 6 months (Table 2 and Figure 2). The body mass index was 29.7 ± 5.5 before the LVAD implant, 28.7 ± 7.7 in 3 months, and 30.1 ± 4.2 in 6 months, and the difference was not significant. Discussion In patients with DM and HF, implantation of LVAD results not only in hemodynamic improvement but also in improvement or recovery of renal and hepatic function. In this retrospective study, we demonstrate that blood glucose control improves in parallel with other favorable changes in the body, within the same time frame. HbA1C, which reflects average blood glucose levels, decreased dramatically within first 3 months after LVAD implantation and remained at this level for up to 1 year. Because some of our patients received LVAD as bridge to transplant and many were transplanted within a year, further follow-up was not feasible. The dual nature of relation between HF and DM is becoming more and more evident. It is well known from Framingham study that DM multiplies the risk of HF up to eightfold,11 and this risk is dependent on the level of blood sugar control. A 1% increase in HbA1C is associated with an 8% increase in risk of HF and improvement in diabetes control decreased the risk of HF.12 However, several cohort studies established the opposite link between HF and DM. Thus, in patients after myocardial infarction, who develop HF, increased diuretic requirement was associated with an increased risk of diabetes.13 In other cohort studies, presence of HF doubled the risk of new-onset DM.14 The mechanism of increased risk of DM in HF is not well studied. Heart failure is a state of high insulin resistance. In Type 2 DM, patients usually have a combination of insulin resistance and insulin deficiency. Patients with HF have insulin sensitivity which is 58% lower than that in healthy subjects.15 High levels of catecholamines which are typically present in HF may contribute to this condition. Increased sympathetic activity inhibits pancreatic insulin secretion and stimulates hepatic gluconeogenesis and glycogenolysis, leading to hyperglycemia and therefore increasing insulin requirements.16 Another consideration is that pancreas, similar to liver and kidneys, is exposed to two major hemodynamic abnormalities characteristic for chronic HF such as decreased forward blood flow (low output) and increased central venous pressure (congestion). It is unknown how these two major forces in advanced HF may reflect on both insulin production and insulin sensitivity. Normalizing cardiac output, LVAD mostly resolves both low output and congestion. In 2011, small series of 15 patients from Columbia University for the first time demonstrated improved diabetic control after LVAD implantation.17
292 GUGLIN et al.
Figure 1. Glycosylated hemoglobin (HbA1C) before and after left ventricular assist device (LVAD) implantation.
One can hypothesize that improved physical activity accompanied by weight loss is a leading force behind better diabetes control after LVAD. However, our study shows that body mass index remained unchanged after LVAD implant (Table 2). It means that some other mechanisms, either directly or indirectly affected by improved hemodynamics, are responsible for lighter course of DM after LVAD. In addition, the follow-up of patients after LVAD implantation is much more thorough than the follow-up of regular diabetic patients with HF. But according to our data, there was a difference toward decrease, not increase, in use of antidiabetic drugs and a trend toward lower insulin requirements. This indicates that more aggressive management of diabetes cannot adequately explain the findings. Another study demonstrated that mechanical unloading after LVAD implantation corrects systemic and local metabolic derangements in advanced HF, leading to reduced myocardial levels of toxic lipid intermediates and improved cardiac insulin signaling.18
Limitations This is a retrospective study covering long period of time during which the hospital moved from paper to electronic medical records. Many records from earlier periods, before implementation of current electronic records in 2011, have incomplete data. Most patients, especially before 2011, received LVAD implantation as a bridge to transplant, and follow-up data longer than 6 months after the implantation date are scarce because they underwent heart transplantation. Also, because of the retrospective nature of the study, no detailed information on nutritional management or neurohormonal status was available. Conclusions Implantation of the LVAD results in improvement in glycemic control in patients with advanced systolic HF. These favorable changes are evident within 3 months from the implantation and sustain till at least 1 year. The timing of improved diabetic
Table 2. Hemodynamic and Laboratory Parameters at Baseline, and at 3 Months and 6 Months After LVAD Implantation
Body mass index, kg/m2 Cardiac index, L/min/m2 Pulmonary capillary wedge pressure, mm Hg Cardiac output, L/min HbA1C, % Insulin, units/day Pulmonary artery systolic pressure, mm Hg Alkaline phosphatase, units/l Sodium, mmol/l Creatinine, mg/dl Total bilirubin, mg/dl Alanine aminotransferase, units/l Aspartate aminotransferase, units/l
Before LVAD
3 Months After LVAD
6 Months After LVAD
p
29.7 (5.5) 1.8 (0.7) 24.4 (8.0) 3.4 (1.3) 7.6 (1.6) 66.2 (66.0) 52.4 (14.1) 87.7 (32.7) 135 (4.0) 1.6 (1.0) 1.4 (0.9) 61.8 (164.5) 44.3 (56.2)
28.7(7.7) 2.2 (0.5) 14.7 (8.2) 4.4 (1.1) 5.7 (0.9) 39.7 (44.0) 32.6 (12.7) 103.0 (43.0) 137.3 (2.5) 1.4 (0.7) 1.1 (1.3) 26.1 (14.1) 36.2 (23.4)
30.1 (4.2) 2.2 (0.5) 14.6 (6.0) 4.3 (1.2) 6.1 (1.0) 38.7 (46.0) 35.5 (12.6) 95.2 (25.4) 137.7 (2.2) 1.5 (1.2) 0.9 (0.3) 21.3 (7.2) 27.9 (7.0)
NS
HbA1C, glycosylated hemoglobin; LVAD, left ventricular assist device; NS, not significant; SD, standard deviation.
0.001 0.02 NS 0.0002 NS 0.0006 NS 0.006 NS NS
DIABETES IMPROVES AFTER LVAD
Figure 2. Longitudinal changes of glycosylated hemoglobin (HbA1C), insulin requirements, and other parameters after left ventricular assist device (LVAD) implantation. PASP, pulmonary artery systolic pressure.
control is similar to the timing of recovery of other hemodynamic and metabolic parameters. Although improved blood glucose levels are likely related to an increase in cardiac output and decrease in congestion after LVAD implant, the exact mechanism needs further investigation. References 1. Nakatani S, McCarthy PM, Kottke-Marchant K, et al: Left ventricular echocardiographic and histologic changes: Impact of chronic unloading by an implantable ventricular assist device. J Am Coll Cardiol 27: 894–901, 1996. 2. Frazier OH, Benedict CR, Radovancevic B, et al: Improved left ventricular function after chronic left ventricular unloading. Ann Thorac Surg 62: 675–681; discussion 681, 1996. 3. Etz CD, Welp HA, Tjan TD, et al: Medically refractory pulmonary hypertension: Treatment with nonpulsatile left ventricular assist devices. Ann Thorac Surg 83: 1697–1705, 2007. 4. John R, Liao K, Kamdar F, Eckman P, Boyle A, Colvin-Adams M: Effects on pre- and posttransplant pulmonary hemodynamics in patients with continuous-flow left ventricular assist devices. J Thorac Cardiovasc Surg 140: 447–452, 2010.
293
5. James KB, McCarthy PM, Thomas JD, et al: Effect of the implantable left ventricular assist device on neuroendocrine activation in heart failure. Circulation 92(9 suppl): II191–II195, 1995. 6. Torre-Amione G, Stetson SJ, Youker KA, et al: Decreased expression of tumor necrosis factor-alpha in failing human myocardium after mechanical circulatory support: A potential mechanism for cardiac recovery. Circulation 100: 1189–1193, 1999. 7. Butler J, Geisberg C, Howser R, et al: Relationship between renal function and left ventricular assist device use. Ann Thorac Surg 81: 1745–1751, 2006. 8. Friedel N, Viazis P, Schiessler A, et al: Recovery of end-organ failure during mechanical circulatory support. Eur J Cardiothorac Surg 6: 519–522; discussion 523, 1992. 9. Smooke S, Horwich TB, Fonarow GC: Insulin-treated diabetes is associated with a marked increase in mortality in patients with advanced heart failure. Am Heart J 149: 168–174, 2005. 10. Gustafsson I, Brendorp B, Seibaek M, et al; Danish Investigatord of Arrhythmia and Mortality on Dofetilde Study Group: Influence of diabetes and diabetes-gender interaction on the risk of death in patients hospitalized with congestive heart failure. J Am Coll Cardiol 43: 771–777, 2004. 11. Kannel WB, Hjortland M, Castelli WP: Role of diabetes in congestive heart failure: The Framingham study. Am J Cardiol 34: 29–34, 1974. 12. Stratton IM, Adler AI, Neil HA, et al: Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): Prospective observational study. BMJ 321: 405–412, 2000. 13. Andersson C, Norgaard ML, Hansen PR, et al: Heart failure severity, as determined by loop diuretic dosages, predicts the risk of developing diabetes after myocardial infarction: A nationwide cohort study. Eur J Heart Fail 12: 1333–1338, 2010. 14. Amato L, Paolisso G, Cacciatore F, et al: Congestive heart failure predicts the development of non-insulin-dependent diabetes mellitus in the elderly. The Osservatorio Geriatrico Regione Campania Group. Diabetes Metab 23: 213–218, 1997. 15. Wong AK, AlZadjali MA, Choy AM, Lang CC: Insulin resistance: A potential new target for therapy in patients with heart failure. Cardiovasc Ther 26: 203–213, 2008. 16. Heck PM, Dutka DP: Insulin resistance and heart failure. Curr Heart Fail Rep 6: 89–94, 2009. 17. Uriel N, Naka Y, Colombo PC, et al: Improved diabetic control in advanced heart failure patients treated with left ventricular assist devices. Eur J Heart Fail 13: 195–199, 2011. 18. Chokshi A, Drosatos K, Cheema FH, et al: Ventricular assist device implantation corrects myocardial lipotoxicity, reverses insulin resistance, and normalizes cardiac metabolism in patients with advanced heart failure. Circulation 125: 2844–2853, 2012.
Improvement in Blood Glucose Control in Patients With Diabetes After Implantation of Left Ventricular Assist Devices Maya Guglin,* Kim Maguire,† Taylor Missimer,* Cristiano Faber,‡ and Christiano Caldeira‡
Some reports suggest that the course of diabetes mellitus (DM) in heart failure (HF) may improve after implantation of left ventricular assist devices (LVADs). The objective of our study was to explore longitudinal changes in glycosylated hemoglobin (HbA1C) in patients with diabetes mellitus with advanced HF post-LVAD implantation. We retrospectively reviewed the records of all patients who received LVADs at our institution between 2002 and 2012 and selected those who 1) survived posthospital discharge and 2) had DM. We collected data on HbA1C before and after implantation of LVADs, daily doses of insulin, and antidiabetic drugs. Comparisons were done using Student’s t-test. A total of 50 patients met the inclusion criteria. HbA1C was 7.6 ± 1.6 before LVAD, 5.7 ± 0.9 within 3 months after the LVAD implant (p = 0.0001), 6.1 ± 1.0 (p = 0.004 in comparison with pre-LVAD level) in 3–6 months after the implant, 6.3 ± 1.0 (p = 0.01) in 6–9 months, and 5.3 ± 0.1 in 9–12 months (p = 0.002). There were no significant changes in body mass index. Favorable changes in clinical course of diabetes in patients with HF occur after the implantation of LVADs, persist for at least 1 year after the implant, and are likely associated with improved hemodynamics and metabolism after normalization of cardiac output. ASAIO Journal 2014; 60:290–293.
myocardial inflammatory mediators such as interleukin-2 or tumor necrosis factor.6 Not surprisingly, function of end-organ systems usually suffering in chronic HF, such as kidneys and liver, also improves.7,8 Surprisingly, function of pancreatic gland, also exposed to both chronic congestion and low output in HF, remains poorly studied. Meanwhile, status of increased insulin resistance is well known in HF, and the combination of HF and diabetes mellitus (DM) is extremely common and linked to poor outcomes.9,10 We hypothesized that function of the pancreatic gland should improve after LVAD implantation and retrospectively studied blood glucose control in patients with DM before and after LVAD implantation. Methods We retrospectively reviewed the records of all patients who underwent LVAD implant at the Tampa General Hospital program from 2002 till the end of 2012. Patients were included in our study if they had the diagnosis of DM, confirmed by being on diabetic diet, oral hypoglycemic agents, insulin, or combination of the above or abnormal fasting glucose or increased glycosylated hemoglobin (HbA1C). Patients were excluded if they had LVAD explantation, death, or heart transplantation within 3 months of LVAD implant. We also excluded those who did not have documented values of HbA1C. We collected baseline demographic, echocardiographic, laboratory parameters, and hemodynamic information from right heart catheterization before LVAD implantation. Similar data, except echocardiographic measurements which were not consistently available, were then collected from the time intervals up to 3 months after LVAD implant and 3–6 months after LVAD implant. Information about oral diabetic agents and insulin requirements (units per day) at various points in time after LVAD implant was analyzed. All laboratory data from the follow-up period were taken from outpatient’s office visits when patients were in stable conditions. Data on right heart catheterization after LVAD were also used. We collected all HbA1C level data up to 12 months after LVAD implant. The values before and after LVAD implant were compared using Student’s t-test. p values less than 0.05 were considered significant. The analysis was performed on SPSS software (SPSS version 21; SPSS Inc., Chicago, IL).
Key Words: heart failure, diabetes mellitus, left ventricular assist device, mechanical assisted circulation
In advanced systolic heart failure (HF), implantation of the
left ventricular assist device (LVAD) normalizes hemodynamics and results in profound favorable changes in multiple organs and systems. In the heart itself, there is a decrease in left ventricular dimensions, increase in left ventricular ejection fraction, and regression of cardiomyocyte hypertrophy.1 Hemodynamically, there is an increase in cardiac output and decrease in pulmonary capillary wedge pressure2 and pulmonary vascular resistance,3,4 resulting in improved right ventricular structure and function. Biochemically, there is a decrease in plasma epinephrine, norepinephrine, arginine vasopressin, renin, and angiotensin II,5 as well as in circulating and From the *Department of Cardiology, University of South Florida, Tampa, Florida; †Department of Medicine, University of South Florida, Tampa, Florida; and ‡Advanced Cardiothoracic Surgery and Tampa General Hospital, Tampa, Florida. Submitted for consideration December 24, 2013; accepted for publication in revised form February 5, 2014. Disclosures: The authors have no conflicts of interest to report. Reprint Requests: Maya Guglin, MD, PhD, FACC, Department of Cardiology, University of South Florida, 2 Tampa General Circle, Tampa, FL 33606. Email: [email protected]. Copyright © 2014 by the American Society for Artificial Internal Organs
Results Between 2005 and the end of 2012, the Tampa General Hospital transplant program implanted 317 LVADs. After we excluded patients who received short-term support only including extracorporeal membranous oxygenation (49) or
DOI: 10.1097/MAT.0000000000000064
290
291
DIABETES IMPROVES AFTER LVAD
pump exchanges (14), we had 254 patients who underwent a long-term pump placement and patients who received heart transplantation during same admission or died within 3 months of LVAD placement. Out of them, 81 patients had a concomitant diagnosis of DM. All diabetic patients had Type 2 diabetes. After we excluded patients whose diabetes status could not confirm because of the lack of information on their diabetic medications or insulin and normal laboratory values of glucose or HbA1C (11), and patients who had heart transplant within same hospital admission as LVAD placement or died within 3 months after the LVAD implant (13), we had 57 patients. In seven of them, there were no recorded values of HbA1C, and they were excluded as well. The remaining 50 patients comprised our study population. During the year after the implant, two of them died and 12 more underwent successful heart transplantation. The data on their diabetes management and laboratory values after the transplantation were not collected. Remaining 36 patients were still on LVAD support 12 months later. The baseline characteristics are summarized in Table 1. Majority were males (35 [70%]) with ischemic cardiomyopathy (31 [62%]). There were six patients who received HeartMate I, six who received HeartWare, and the remaining 38 patients received HeartMate II. Before the LVAD, 22 patients (43.0%) were on insulin, with the mean daily dose of 66.2 ± 66.0 units, and 24 (48%) were taking oral diabetic medications. In 3 months, 16 (32.0%) remained on insulin, requiring 39.7 ± 44 units daily, and eight (16.0%) were on oral hypoglycemic agents. In 6 months, there were 18 patients (36.0%) on insulin with the requirement of 38.7 ± 46.0 units a day, and 11 (22.0%) were on oral agents. The difference was significant only for the proportion of patients on oral diabetic agents before and after LVAD, p = 0.012 and p = 0.020 for 3 and 6 months after the LVAD, respectively, comparing with the pre-LVAD level. In terms of HbA1C, it was 7.6 ± 1.6 before LVAD, 5.7 ± 0.9 within 3 months after the LVAD implant (p = 0.0001), 6.1 ± 1.0 Table 1. Baseline Characteristics Before LVAD Age, years (SD) Men, % Ischemic cardiomyopathy, % HeartMate I, % HeartMate II, % HeartWare,% Body mass index, kg/m2 On insulin,% Insulin, units/day HbA1C, % Left ventricular ejection fraction, % Peak velocity of tricuspid regurgitation, m/sec Left ventricular end-diastolic dimension, mm Sodium, mmol/l Creatinine, mg/dl Total bilirubin, mg/dl Alanine aminotransferase, units/l Aspartate aminotransferase, units/l
56.7 (12.1) 35 (70.0) 31 (62.0) 6 (12.0) 38 (76.0) 6 (12.0) 29.7 (5.5) 22 (43.1) 66.2 (66.0) 7.6 (1.6) 15.2 (6.5) 2.7 (0.7) 69.5 (13.9) 135 (4.0) 1.6 (1.0) 1.4 (0.9) 61.8 (164.5) 44.3 (56.2)
HbA1C, glycosylated hemoglobin; LVAD, left ventricular assist device; SD, standard deviation.
(p = 0.004 in comparison with pre-LVAD level) in 3–6 months after the implant, 6.3 ± 1.0 (p = 0.01) in 6–9 months, and 5.3 ± 0.1 in 9–12 months (p = 0.002) (Figure 1). From other parameters, pulmonary arterial systolic pressure, pulmonary capillary wedge pressure, cardiac output, and liver function tests, particularly total bilirubin, as well as HbA1C, improved within 3 months after LVAD implant and remained at a better level than before the surgery in 6 months (Table 2 and Figure 2). The body mass index was 29.7 ± 5.5 before the LVAD implant, 28.7 ± 7.7 in 3 months, and 30.1 ± 4.2 in 6 months, and the difference was not significant. Discussion In patients with DM and HF, implantation of LVAD results not only in hemodynamic improvement but also in improvement or recovery of renal and hepatic function. In this retrospective study, we demonstrate that blood glucose control improves in parallel with other favorable changes in the body, within the same time frame. HbA1C, which reflects average blood glucose levels, decreased dramatically within first 3 months after LVAD implantation and remained at this level for up to 1 year. Because some of our patients received LVAD as bridge to transplant and many were transplanted within a year, further follow-up was not feasible. The dual nature of relation between HF and DM is becoming more and more evident. It is well known from Framingham study that DM multiplies the risk of HF up to eightfold,11 and this risk is dependent on the level of blood sugar control. A 1% increase in HbA1C is associated with an 8% increase in risk of HF and improvement in diabetes control decreased the risk of HF.12 However, several cohort studies established the opposite link between HF and DM. Thus, in patients after myocardial infarction, who develop HF, increased diuretic requirement was associated with an increased risk of diabetes.13 In other cohort studies, presence of HF doubled the risk of new-onset DM.14 The mechanism of increased risk of DM in HF is not well studied. Heart failure is a state of high insulin resistance. In Type 2 DM, patients usually have a combination of insulin resistance and insulin deficiency. Patients with HF have insulin sensitivity which is 58% lower than that in healthy subjects.15 High levels of catecholamines which are typically present in HF may contribute to this condition. Increased sympathetic activity inhibits pancreatic insulin secretion and stimulates hepatic gluconeogenesis and glycogenolysis, leading to hyperglycemia and therefore increasing insulin requirements.16 Another consideration is that pancreas, similar to liver and kidneys, is exposed to two major hemodynamic abnormalities characteristic for chronic HF such as decreased forward blood flow (low output) and increased central venous pressure (congestion). It is unknown how these two major forces in advanced HF may reflect on both insulin production and insulin sensitivity. Normalizing cardiac output, LVAD mostly resolves both low output and congestion. In 2011, small series of 15 patients from Columbia University for the first time demonstrated improved diabetic control after LVAD implantation.17
292 GUGLIN et al.
Figure 1. Glycosylated hemoglobin (HbA1C) before and after left ventricular assist device (LVAD) implantation.
One can hypothesize that improved physical activity accompanied by weight loss is a leading force behind better diabetes control after LVAD. However, our study shows that body mass index remained unchanged after LVAD implant (Table 2). It means that some other mechanisms, either directly or indirectly affected by improved hemodynamics, are responsible for lighter course of DM after LVAD. In addition, the follow-up of patients after LVAD implantation is much more thorough than the follow-up of regular diabetic patients with HF. But according to our data, there was a difference toward decrease, not increase, in use of antidiabetic drugs and a trend toward lower insulin requirements. This indicates that more aggressive management of diabetes cannot adequately explain the findings. Another study demonstrated that mechanical unloading after LVAD implantation corrects systemic and local metabolic derangements in advanced HF, leading to reduced myocardial levels of toxic lipid intermediates and improved cardiac insulin signaling.18
Limitations This is a retrospective study covering long period of time during which the hospital moved from paper to electronic medical records. Many records from earlier periods, before implementation of current electronic records in 2011, have incomplete data. Most patients, especially before 2011, received LVAD implantation as a bridge to transplant, and follow-up data longer than 6 months after the implantation date are scarce because they underwent heart transplantation. Also, because of the retrospective nature of the study, no detailed information on nutritional management or neurohormonal status was available. Conclusions Implantation of the LVAD results in improvement in glycemic control in patients with advanced systolic HF. These favorable changes are evident within 3 months from the implantation and sustain till at least 1 year. The timing of improved diabetic
Table 2. Hemodynamic and Laboratory Parameters at Baseline, and at 3 Months and 6 Months After LVAD Implantation
Body mass index, kg/m2 Cardiac index, L/min/m2 Pulmonary capillary wedge pressure, mm Hg Cardiac output, L/min HbA1C, % Insulin, units/day Pulmonary artery systolic pressure, mm Hg Alkaline phosphatase, units/l Sodium, mmol/l Creatinine, mg/dl Total bilirubin, mg/dl Alanine aminotransferase, units/l Aspartate aminotransferase, units/l
Before LVAD
3 Months After LVAD
6 Months After LVAD
p
29.7 (5.5) 1.8 (0.7) 24.4 (8.0) 3.4 (1.3) 7.6 (1.6) 66.2 (66.0) 52.4 (14.1) 87.7 (32.7) 135 (4.0) 1.6 (1.0) 1.4 (0.9) 61.8 (164.5) 44.3 (56.2)
28.7(7.7) 2.2 (0.5) 14.7 (8.2) 4.4 (1.1) 5.7 (0.9) 39.7 (44.0) 32.6 (12.7) 103.0 (43.0) 137.3 (2.5) 1.4 (0.7) 1.1 (1.3) 26.1 (14.1) 36.2 (23.4)
30.1 (4.2) 2.2 (0.5) 14.6 (6.0) 4.3 (1.2) 6.1 (1.0) 38.7 (46.0) 35.5 (12.6) 95.2 (25.4) 137.7 (2.2) 1.5 (1.2) 0.9 (0.3) 21.3 (7.2) 27.9 (7.0)
NS
HbA1C, glycosylated hemoglobin; LVAD, left ventricular assist device; NS, not significant; SD, standard deviation.
0.001 0.02 NS 0.0002 NS 0.0006 NS 0.006 NS NS
DIABETES IMPROVES AFTER LVAD
Figure 2. Longitudinal changes of glycosylated hemoglobin (HbA1C), insulin requirements, and other parameters after left ventricular assist device (LVAD) implantation. PASP, pulmonary artery systolic pressure.
control is similar to the timing of recovery of other hemodynamic and metabolic parameters. Although improved blood glucose levels are likely related to an increase in cardiac output and decrease in congestion after LVAD implant, the exact mechanism needs further investigation. References 1. Nakatani S, McCarthy PM, Kottke-Marchant K, et al: Left ventricular echocardiographic and histologic changes: Impact of chronic unloading by an implantable ventricular assist device. J Am Coll Cardiol 27: 894–901, 1996. 2. Frazier OH, Benedict CR, Radovancevic B, et al: Improved left ventricular function after chronic left ventricular unloading. Ann Thorac Surg 62: 675–681; discussion 681, 1996. 3. Etz CD, Welp HA, Tjan TD, et al: Medically refractory pulmonary hypertension: Treatment with nonpulsatile left ventricular assist devices. Ann Thorac Surg 83: 1697–1705, 2007. 4. John R, Liao K, Kamdar F, Eckman P, Boyle A, Colvin-Adams M: Effects on pre- and posttransplant pulmonary hemodynamics in patients with continuous-flow left ventricular assist devices. J Thorac Cardiovasc Surg 140: 447–452, 2010.
293
5. James KB, McCarthy PM, Thomas JD, et al: Effect of the implantable left ventricular assist device on neuroendocrine activation in heart failure. Circulation 92(9 suppl): II191–II195, 1995. 6. Torre-Amione G, Stetson SJ, Youker KA, et al: Decreased expression of tumor necrosis factor-alpha in failing human myocardium after mechanical circulatory support: A potential mechanism for cardiac recovery. Circulation 100: 1189–1193, 1999. 7. Butler J, Geisberg C, Howser R, et al: Relationship between renal function and left ventricular assist device use. Ann Thorac Surg 81: 1745–1751, 2006. 8. Friedel N, Viazis P, Schiessler A, et al: Recovery of end-organ failure during mechanical circulatory support. Eur J Cardiothorac Surg 6: 519–522; discussion 523, 1992. 9. Smooke S, Horwich TB, Fonarow GC: Insulin-treated diabetes is associated with a marked increase in mortality in patients with advanced heart failure. Am Heart J 149: 168–174, 2005. 10. Gustafsson I, Brendorp B, Seibaek M, et al; Danish Investigatord of Arrhythmia and Mortality on Dofetilde Study Group: Influence of diabetes and diabetes-gender interaction on the risk of death in patients hospitalized with congestive heart failure. J Am Coll Cardiol 43: 771–777, 2004. 11. Kannel WB, Hjortland M, Castelli WP: Role of diabetes in congestive heart failure: The Framingham study. Am J Cardiol 34: 29–34, 1974. 12. Stratton IM, Adler AI, Neil HA, et al: Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): Prospective observational study. BMJ 321: 405–412, 2000. 13. Andersson C, Norgaard ML, Hansen PR, et al: Heart failure severity, as determined by loop diuretic dosages, predicts the risk of developing diabetes after myocardial infarction: A nationwide cohort study. Eur J Heart Fail 12: 1333–1338, 2010. 14. Amato L, Paolisso G, Cacciatore F, et al: Congestive heart failure predicts the development of non-insulin-dependent diabetes mellitus in the elderly. The Osservatorio Geriatrico Regione Campania Group. Diabetes Metab 23: 213–218, 1997. 15. Wong AK, AlZadjali MA, Choy AM, Lang CC: Insulin resistance: A potential new target for therapy in patients with heart failure. Cardiovasc Ther 26: 203–213, 2008. 16. Heck PM, Dutka DP: Insulin resistance and heart failure. Curr Heart Fail Rep 6: 89–94, 2009. 17. Uriel N, Naka Y, Colombo PC, et al: Improved diabetic control in advanced heart failure patients treated with left ventricular assist devices. Eur J Heart Fail 13: 195–199, 2011. 18. Chokshi A, Drosatos K, Cheema FH, et al: Ventricular assist device implantation corrects myocardial lipotoxicity, reverses insulin resistance, and normalizes cardiac metabolism in patients with advanced heart failure. Circulation 125: 2844–2853, 2012.
