International Journal of Cardiology 174 (2014) 306–312

Contents lists available at ScienceDirect

International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard

Low fasting plasma glucose level predicts morbidity and mortality in symptomatic adults with congenital heart disease Hideo Ohuchi ⁎, Kenji Yasuda, Shin Ono, Yosuke Hayama, Jun Negishi, Kanae Noritake, Masanori Mizuno, Toru Iwasa, Aya Miyazaki, Osamu Yamada Department of Pediatric Cardiology and Adult Congenital Heart Disease, National Cerebral and Cardiovascular Center, Osaka, Japan

a r t i c l e

i n f o

Article history: Received 14 December 2013 Received in revised form 9 February 2014 Accepted 4 April 2014 Available online 15 April 2014 Keywords: Adult congenital heart disease Glucose Heart failure Mortality

a b s t r a c t Background: Adults with complex congenital heart disease (ACHD) have a high prevalence of abnormal glucose regulation (AGR: impaired glucose tolerance and diabetes mellitus). However, the impact of AGR on the prognosis remains unclear. Purpose: Our purpose was to clarify the prognostic value of AGR in ACHD. Methods and results: We performed a 75 g oral glucose tolerance test in 438 consecutive patients with ACHD (age 26 ± 8 years), including 38 unrepaired, 148 Fontan, 252 biventricular, and 27 healthy subjects and investigated associations between AGR and clinical events that required hospitalization or caused deaths from all-causes. When compared with the healthy group, fasting blood glucose level (FPG, mg/dl) was lower in the unrepaired and Fontan subjects (p b 0.05–0.01) and the prevalence of low FPG (≤ 80 mg/dl) was also higher in the unrepaired (58%), Fontan (47%), and biventricular group (33%) than in the healthy control (11%) (p b 0.0001). Postprandial hyperglycemia (area under the curve of glucose: PG-AUC) was higher in all ACHD groups (p b 0.0001 for all). New York Heart Association class and lower FPG independently predicted the hospitalization (FPG ≤ 84 mg/dl) and mortality (FPG ≤ 80 mg/dl) (p b 0.05–0.0001), while the PG-AUC was not an independent predictor. When compared with the asymptomatic ACHD, symptomatic ACHD with lower FPG had high hazard ratios of 2.2 (95% confidence interval [CI]: 1.3–3.8, p b 0.002) and 3.3 (95% CI: 1.2–11.9, p b 0.03) for the hospitalizations and all-cause mortality, respectively. Conclusions: Low FPG is not uncommon in ACHD and the low FPG predicts the morbidity and all-cause mortality in symptomatic ACHD. © 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Abnormal glucose regulation (AGR), diabetes mellitus and impaired glucose tolerance (IGT), is one of the major risk factors for poor prognosis in patients with cardiovascular disease [1–3] and AGR is characteristic of the pathophysiology of those patients with heart failure (HF) [4]. Therefore, AGR-conscious therapeutic strategy leads to a significant reduction of cardiovascular events and early interventions of this nature have been emphasized for the better outcome [5,6]. Furthermore, a routine 75 g oral glucose tolerance test (75 g-OGTT) has been recommended for cardiovascular patients because undiagnosed AGR has a significant detrimental impact on the prognosis [7]. On the other hand, a significant number of children with congenital

heart disease (CHD), especially those with complex malformations, reached their adulthood because of recent medical advances [8,9]. In those adults with complex CHD (ACHD), a high prevalence of AGR has been recognized despite the their young age [10]. Those ACHD show activated neurohormonal activities, abnormal cardiac autonomic nervous activity, and inflammatory cytokines as demonstrated in non-ACHD patients with HF [11–13] and some of those prognostic values are emerging [14]. However, a prognostic value of AGR still remains unclear, especially the prognostic power for the mortality in ACHD. Accordingly, the aim of the present study was to investigate an association between AGR and prognosis in our large cohort of ACHD. 2. Methods 2.1. Subjects

⁎ Corresponding author at: Department of Pediatrics, National Cardiovascular Center 5-7-1, Fujishiro-dai, Suita, Osaka 565-8565, Japan. Tel.: + 81 6 6833 5012; fax: + 81 6 6872 7486. E-mail address: [email protected] (H. Ohuchi).

http://dx.doi.org/10.1016/j.ijcard.2014.04.070 0167-5273/© 2014 Elsevier Ireland Ltd. All rights reserved.

We prospectively studied 471clinically stable adult subjects, including 444 ACHD (16 to 62 years) and 27 healthy volunteers (21 to 39 years) (Table 1). ACHD included 38 unrepaired cyanotic ACHD, including 7 with Eisenmenger syndrome, 149 Fontan

H. Ohuchi et al. / International Journal of Cardiology 174 (2014) 306–312

307

Table 1 Clinical characteristics of the study groups.

Cases Age (years) Male gender (%) Body mass index (kg/m2) Follow-up (years) Disease (repair)

NYHA class (I/II/III) Peak VO2 (ml/kg/min) Hemodynamics (n) CVP (mmHg) PAP (mmHg) EF (normal/reduced/poor) Hemoglobin (g/dl) CI (l/min/m2) SpO2 (mmHg) Neurohumoral factors Norepinephrine (pg/ml) BNP (pg/ml) Renin Activity (ng/ml/h) Medications (%) Diuretics Anticoagulant ACEI/ARB Beta blocker

Unrepaired

Fontan

Biventricular

Volunteers

p Value

38 32 ± 10⁎! 50 19 ± 4! – TF (10), UVH (8) VSD (4), DORV (4) TGA (3), Others (9) 2.5 ± 0.6 (1/17/19)⁎! 15 ± 4⁎!

149 22 ± 6!# 58 20 ± 3! 16 ± 5 TA (43), UVH (42) DORV (22), MA (14) PA (12), Others (16) 1.5 ± 0.6 (92/45/11)! 23 ± 7!

257 27 ± 8 50 21 ± 4 21 ± 7 TF (92), TGA (54) AVD (26), CoA/IAA (18) DORV (11), Others (56) 1.3 ± 0.5 (190/61/6) 26 ± 7

27 27 ± 5 44 22 ± 2 – –

– b0.0001 ns 0.0006 b0.0001 –

– –

b0.0001 b0.0001

6 ± 3⁎ 42 ± 31⁎! 18/16/4 18 ± 4⁎! 2.5 ± 0.9 83 ± 7⁎!

10 ± 2! 10 ± 3! 61/78/10 15 ± 2! 2.6 ± 0.6 94 ± 3!

6±4 16 ± 8 128/109/20 14 ± 2 2.6 ± 0.6 97 ± 2

– – – – – –

b0.0001 b0.0001 ns b0.0001 ns b0.0001

471 ± 303⁎!# 205 ± 336⁎!# 7.6 ± 7.7

422 ± 206!# 40 ± 62 12.6 ± 16.4!#

362 ± 214 56 ± 72 8.2 ± 10.2#

283 ± 97 9±3 1.0 ± 0.7

0.0003 b0.0001 b0.0001

47 58 39 24

48 72 38 28

37 33 25 22

– – – –

ns b0.0001 0.0085 ns

ACEI = angiotensin converting enzyme inhibitor, ARB = angiotensin receptor blocker, AVD = atrioventricular discordance, BNP = brain natriuretic peptide, CI = cardiac index, CoA/IAA = coarctation or interruption of the aorta, CVP = central venous pressure, DORV = double outlet right ventricle, EDVI = systemic ventricular end-diastolic volume index, EF = systemic ventricular ejection fraction, Ma = mitral atresia, NYHA = New York Heart Association, PA = pulmonary atresia, PAP = pulmonary artery pressure, SpO2 = arterial oxygen saturation, TA = tricuspid valve atresia, TOF = tetralogy of Fallot, TGA = transposition of the great arteries, UVH = univentricular heart, VSD = ventricular septal defect, VO2 = oxygen uptake. Values are mean ± SD. *,!, and # indicate p b 0.0 vs. Fontan, biventricular, and volunteer group, respectively. patients and 257 patients after biventricular repair (BVR). Of the unrepaired patients, 31 patients underwent cardiac catheterization to evaluate hemodynamics for possible surgical/catheter intervention for complex CHD anomalies and medical therapy for pulmonary hypertension in the other 7 Eisenmenger patients. Of the BVR patients, hemodynamics were assessed in 239 patients with cardiac catheterization with cardiovascular imaging. In the Fontan patients, follow up clinical evaluations, including cardiac catheterization with cardiovascular imaging, were performed as previously described [10,13], except for one. Of the BVR patients, 18 with atrioventricular discordance (functional repair) and 14 with transposition of the great arteries (atrial switch operation) had a morphologic right ventricle (RV) as a systemic ventricle (SV). Of the Fontan patients, a total cavopulmonary connection was created in 138 and an atriopulmonary connection in 11. The follow-up period from the last operation to the time of study was at least 6 months in post-surgical patients. All patients were free from intravenous medications. Medications, including diuretics, anticoagulant agents, and angiotensin converting enzyme inhibitors (ACEI) or angiotensin receptor blockers (ARB), β blockers and anti-arrhythmic agents, are shown in Table 1. Anti-coagulants and ACEI/ARB were more frequently given in the unrepaired and Fontan patients. No patients were taking inotropic agents, except for digoxin. The healthy volunteers who served as controls showed no significant lung or heart problems by routine physical check-up. 2.2. Hemodynamics As described above, cardiac catheterization was performed in 31 cyanotic, 148 Fontan and 239 BVR patients and hemodynamic variables were calculated. We estimated oxygen consumption from the age, sex, and heart rate and measured cardiac index (l/min/m2) using the Fick principle with the assumption that right and left pulmonary arterial saturations were equal in patients with either a Glenn or a total cavopulmonary connection because it is clinically difficult to measure accurate flow distribution in the bilateral pulmonary arteries. In patients who underwent cineventriculography, we used Simpson's rule to estimate morphological right and left ventricular volumes. End-diastolic ventricular volume was divided by body surface area to obtain end-diastolic volume index and SV ejection fraction (EF) was calculated [10,13]. When cineventriculography was not available, the SVEF was estimated by echocardiography using Pombo's method in patients with morphological left ventricle (LV) and a nuclear imaging technique was applied to estimate the SVEF in patients with non-LV SV. Because of the inaccuracy of these estimations for the SVEF, our patients were further divided into SVEF-based 3 groups, i.e., groups with normal (SVEF ≥ 60%), reduced (30% ≤ SVEF b 60%) and severely reduced (SVEF b 30%) and SVEF values were graded as 0, 1, and 2, respectively. 2.3. Exercise protocol Thirty-six unrepaired, 147 Fontan, and 247 BVR subjects underwent symptom-limited treadmill exercise within 1 week of cardiac catheterization [15] and cardiorespiratory

variables included resting and peak heart rates and systolic blood pressures, exercise time and peak oxygen uptake (VO2: ml/kg/min). 2.4. Plasma neurohormonal activities After at least 15 minutes supine rest, the plasma norepinephrine concentration (NE, by high-performance liquid chromatography) [16], brain natriuretic peptides, and renin activity (PRA) were determined in all ACHD and volunteers [17,18]. 2.5. Glucose tolerance Glucose metabolic variables included fasting plasma glucose (FPG) (mg/dl) (glucose oxidase electrode method, Auto & Stat GA-1160, Kyoto Dai-ich Kagaku, Japan), insulin (μU/ml) (electro-chemi-luminescence immunoassay), and hemoglobin A1c (HbA1c) concentrations (high-performance liquid chromatography). A standard 75 g-OGTT was administered with measurements of plasma glucose (PG) and plasma insulin levels before, 30, 60 and 120 minutes post-challenge in all ACHD, except for those with known diabetes mellitus (one in the Fontan and 5 BV patients) who were excluded from the 75 g-OGTT-associated analysis. Abnormalities of glucose tolerance were defined by the following criteria: impaired FPG as a FPG between 110 and 125 (mg/dl), IGT as a PG between 140 and 199 (mg/dl) 120 minutes after the oral glucose challenge, and provisional or newly diagnosed diabetes [19]. We also calculated areas under the PG response curve using fasting, 30, 60, and 120-PG concentrations with the trapezoid rule as an index of postprandial hyperglycemia. To focus on the response, we used values of the area under the PG response curve — FPG × 120 (minutes) (AUC-PG). Homeostasis model assessment (HOMA) was used to assess hepatic insulin resistance (IR), respectively [20]. Informed consent was obtained from all patients and healthy controls and/or their parents. This study protocol was approved by the Ethics Committee of the National Cerebral and Cardiovascular Center. 2.6. Clinical events that required hospitalization Clinical events included cardiovascular and non-cardiovascular events that required hospitalization. Cardiovascular events included arrhythmia, HF, catheter or surgical intervention. Pacemaker implantation for severe bradycardia or atrioventricular block, and medical and catheter ablation for treatment of clinically significant arrhythmias were considered arrhythmia-related cardiovascular events. Noncardiovascular events included hemoptysis, thromboembolism, stroke and infectious endocarditis. Patients were categorized as HF if it complicated worsening HF as defined by evidence of at least one of the following: orthopnea, nocturnal dyspnea, pulmonary edema, increasing peripheral edema with or without inappropriate body weight gain, or radiological signs of congestive HF [21]. Protein losing

308

H. Ohuchi et al. / International Journal of Cardiology 174 (2014) 306–312

enteropathy, its relapse (confirmed by 99Technetium labeled albumin bowel scintigraphy) and renal failure were categorized as HF in the present study. Indications for catheter or surgical intervention for hemodynamic abnormalities, such as stenosis of the right ventricular outflow or atrioventricular regurgitation, were decided at our institutional clinical conference.

2.7. Statistical analysis Data are expressed as the mean ± SD. Differences in demographics, functional capacity, hemodynamics, glucose metabolic and neurohormonal variables were evaluated using one-way ANOVA with Turkey's post hoc test among ≥ three groups. Simple regression analysis was used to evaluate relationships between continuous variables and multivariate linear regression analysis was used to detect the main correlates. Comparisons of prevalence of medications and glucose metabolic abnormalities were evaluated with chi-square or Fisher's exact test. We used univariate Cox's proportional hazards model to predict the associations of clinical factors, i.e., age, gender, body mass index (BMI), functional capacity, SV systolic function, arterial oxygen saturation, neurohormonal levels and glucose metabolic variables. To ease the interpretability of the results, hazard ratios have been computed for the 100 (pg/ml) of NE and 10 (pg/ml) of brain natriuretic peptide. When the variables were statistically significant, the receiver-operator characteristic (ROC) curve analysis was applied to determine these cutoff values to identify efficient prognostic prediction, i.e., the value with maximal area under the ROC curve. Variables that proved significant variables in univariate analysis (p b 0.05) were included in the multivariate analysis of Cox's regression model to determine independent predictors. Clinical event free status was estimated using the Kaplan-Meier method, and differences in the event free status between groups were assessed using log rank tests. Analyses were performed with the software JMP 10 pro (SAS Institute, Cary, NC, USA). A p value of b 0.05 was considered statistically significant.

3. Results Clinical characteristics, including demographic, hemodynamic, and medications are shown in Table 1. The unrepaired patients were older and Fontan patients were younger than the other two groups. BMI was lower in the ACHD, especially in the unrepaired patients. As for the hemodynamics, the unrepaired patients showed the highest pulmonary pressure and lowest arterial oxygen saturation, while the Fontan patients had the highest CVP. However, there were no differences in SVEF or cardiac index among the ACHD groups. Neurohormonal activities were higher in the ACHD groups than in the volunteers, except for the brain natriuretic peptide and PRA because of the skewed distribution. As for the BVR patients, relatively high prevalence of medication use indicated that our BVR patients were somewhat sicker patients than other types of simple postoperative BVR patients, such as patients after closure of atrial or ventricular septal defect.

a)

3.1. Glucose metabolic variables FPG just before 75 g-OGTT (baseline FPG) was lower in the unrepaired and Fontan patients than in the volunteers, whereas the HbA1c was higher, indicating postprandial hyperglycemia in these patients. Baseline insulin levels and HOMA-IR were higher in the Fontan and BVR patients, indicating insulin resistance, although the values of HOMA-IR did not reach statistical difference. When we divided our patients into 3 subgroups according to baseline FPG, i.e., those with FPG b 80 (mg/dl), those with FPG between 80 and 90 (mg/dl), and those with FPG N 90 (mg/dl), the prevalence of low FPG was higher in all ACHD groups, especially in the Fontan and unrepaired groups (p b 0.0001, Fig. 1-a). AUC-PG was higher in all ACHD groups with no group differences and the prevalence of AGR was markedly higher in all ACHD groups (Table 2, Fig. 1-b).

3.2. Clinical correlates and morbidity and mortality During a mean follow-up of 51 ± 26 (1 to 95) months from 75 g-OGTT, we encountered 18 deaths and 136 clinical events that required hospitalization. The causes of deaths were surgery-associated and HF in 5 each, sudden death in 3, hepatocellular carcinoma in 2, and gastrointestinal bleeding, protein losing enteropathy, stroke, and other in one each. The causes of hospitalizations were arrhythmia in 50, HF in 33, surgery in 25, relapse of protein losing enteropathy in 6, and others in 22. Associations of clinical variables with the hospitalization and all-cause mortality are summarized in Tables 3 and 4, respectively. As for the hospitalizations, according to univariate Cox model, many clinical variables were associated with the hospitalizations, including FPG during 75 g-OGTT, AUC-PG and a presence of AGR. Logistic regression model with ROC analysis revealed that values of 84 (mg/dl), 6.16 (g/dl) and 5.4 (ng/ml/h) were the cut-off values for the FPG, AUC-PG, and PRA, respectively. Free rates from hospitalizations based on the 3 FPG-subgroups are shown in Fig. 2-a, showing a higher clinical event rate in ACHD with lower FPG. Of the clinical variables, older age, NYHA class, use of β blocker, high PRA, and low FPG (≤ 84 mg/dl) were the independent predictors of hospitalizations. If peak VO2, instead of NYHA class, was included in the multivariate, the predictors were the same, including peak VO2 instead of NYHA class (Table 3). We divided our patients into 4 subgroups according to presence of symptoms (asymptomatic: NYHA class I or symptomatic: NYHA class ≥ II) and

b) (p < 0.0001)

Rate (%)

(p < 0.0001)

Rate (%)

100%

100%

80%

80%

60%

60%

40%

40%

20%

20%

0%

0% Control

BVR

Fontan

Unrepaired

Control

BVR

Fontan

Unrepaired

Fig. 1. Prevalence of low fasting plasma glucose (FPG b 80 mg/dl) in red and that of FPG N 90 (mg/dl) in blue (a) and that of abnormal glucose regulation (b), i.e., impaired glucose tolerance (IGT, green) and diabetes mellitus (red)

H. Ohuchi et al. / International Journal of Cardiology 174 (2014) 306–312

309

Table 2 Glucose metabolic variables in study groups.

75 g-OGTT Baseline glucose (mg/dl) Baseline insulin (^U/ml) AUC PG (g/dl) AUC insulin (mU/ml) IGT/DM (%) Baseline HOMA-IR HbA1c (%)

Unrepaired

Fontan

Biventricular

Volunteers

p

38 79 ± 7!!# 3.7 ± 2.4⁎⁎⁎!!! 6.3 ± 2.9### 5.1 ± 2.7⁎⁎ 16/3 (50) 0.7 ± 0.5⁎⁎⁎!!!

148 82 ± 7!# 7.5 ± 4.8# 6.7 ± 3.2### 8.5 ± 6.1### 52/14 (46) 1.6 ± 1.1 5.7 ± 0.4!!###

252 84 ± 8 7.0 ± 4.4# 6.6 ± 2.9### 7.4 ± 5.5### 99/11(44)⁎⁎⁎ 1.5 ± 1.0 5.5 ± 0.5

27 87 ± 7 4.7 ± 2.1 3.3 ± 2.5 3.3 ± 1.5 1/0 (4) 1.0 ± 0.6 4.9 ± 0.3

b0.0001 b0.0001 b0.0001 b0.0001 0.0004 b0.0001 b0.0001

5.8 ± 0.5!###

AUC PG = area under the curve of plasma glucose during OGTT, DM = diabetes mellitus, HOMA-IR = insulin resistance by homeostasis model assessment, IGT = impaired glucose tolerance. Values are mean ± SD. *, **, and *** indicate p b 0.05, 0.01, and 0.001 vs. Fontan, respectively. !, !!, and !!! indicate p b 0.05, 0.01, and 0.001 vs. biventricular, respectively. #, ##, and ### indicate p b 0.05, 0.01, and 0.001 vs. volunteers, respectively.

low FPG (≤ 84 mg) and those hospitalization free rates are shown in Fig. 2-b. When compared with the asymptomatic ACHD with a high FPG, hazard ratios of asymptomatic ACHD with a high FPG, symptomatic ACHD with a high FPG, and those with a low FPG were 1.55 (95% confidence interval [CI]: 0.86–2.94, p = 0.1448), 3.13 (95% CI: 1.59–6.26, p = 0.001), and 6.85 (95% CI: 4.01–12.5, p b 0.0001), respectively. Thus, symptomatic ACHD with a low FPG had a high hazard ratio of 2.19 (95% CI: 1.34–3.76, p = 0.0015) when compared with symptomatic ACHD with a high FPG. If the 5 known diabetic ACHD were included in the analysis of FPG, there were again few statistical differences. As for the all-cause mortality, age, NYHA class, peak VO2, arterial oxygen saturation, uses of diuretics and anti-coagulants, and low FPG were associated with the mortality (see Fig. 3-a for the mortality). The cut-off value of the FPG was 80 (mg/dl). On the multivariate analysis using dichotomous FPG groups, i.e., low (≤ 80 mg/dl) and high FPG (N80 mg/dl), NYHA class and low FPG were independent predictors of the mortality. If peak VO2, instead of NYHA class, was used in the analysis, a low FPG was the only independent predictor (Table 4). We divided our ACHD into 4 subgroups according to the symptoms and FPG and those survival curves are shown in Fig. 3-b. There were no deaths in the asymptomatic ACHD with a high FPG. When compared

with the asymptomatic ACHD, hazard ratios of the mortality in symptomatic ACHD with and without a low FPG were 6.19 (95% CI: 1.20–44.8, p = 0.0297) and 20.6 (95% CI: 5.60–132, p b 0.0001), respectively. Thus, symptomatic ACHD with a low FPG had a high hazard ratio of 3.32 (95% CI: 1.16–11.9, p = 0.0248) when compared with symptomatic ACHD with a high FPG. If the 5 known diabetic ACHD were included in the analysis of FPG, there were again few statistical differences. On the other hand, the conventional classification of AGR, i.e., normal, IGT, and diabetes mellitus, had little impact on the hospitalization and all-cause mortality in all our ACHD patients (Fig. 4-a and b).

4. Discussion In addition to reconfirming a high prevalence of AGR in ACHD, we made the following new observations: 1) a low FPG was strongly associated with a high morbidity and all-cause mortality in symptomatic ACHD despite a weak adverse impact of IGT/diabetes mellitus on the morbidity in our young ACHD. Therefore, a low FPG, including AGR, is one of the characteristic ACHD pathophysiologic features closely associated with the disease severity and prognosis.

Table 3 Univariate and multivariate predictors of unscheduled hospitalization in all ACHD. Unscheduled hospitalization Univariate

Patient characteristics Age (years) Male Body mass index NYHA class Peak VO2 (per 1.0 ml/kg/min) SV systolic function (normal = 1, reduced = 2, severe = 3) Arterial oxygen saturation (per 1%) Medications Diuretics Anti-coagulant ACEI/ARB Beta blocker Neurohumoral factors Norepinephrine (per 100 pg/ml) Renin activity (ng/ml/h) BNP (per 10 pg/ml) Glucose metabolism Low glucose (b84) Insulin (^U/ml) HbA1c (%) HOMA-IR High AUC PG (N6.165) IGT/DM vs. NGT

Multivariate (model 1)

Multivariate (model 2)

HR

95% CI

p Value

HR

95% CI

p Value

HR

95% CI

p Value

1.04 1.24 0.95 2.93 0.91 1.67 0.94

1.02–1.06 0.89–1.75 0.91–1.00 2.34–3.67 0.88–0.93 1.30–2.14 0.92–0.97

b0.0001 0.2075 0.0329 b0.0001 b0.0001 b0.0001 b0.0001

1.03

1.01–1.05

0.0052

1.03

1.01–1.05

0.0079

2.34 –

1.61–3.40 –

b0.0001 –

– 0.96

– 0.92–0.99

– 0.0285

2.71 1.92 1.82 2.56

1.92–3.87 1.35–2.75 1.29–2.56 1.81–3.61

b0.0001 0.0002 0.0008 b0.0001

1.60

1.06–2.39

0.0256

1.56

1.02–2.40

0.0418

1.16 1.03 1.01

1.10–1.22 1.02–1.04 1.01–1.02

b0.0001 b0.0001 0.0009

1.02

1.01–1.04

0.0028

1.03

1.01–1.04

0.0015

1.95 0.97 1.24 0.86 1.69 1.43

1.34–2.93 0.93–1.01 0.93–1.58 0.70–1.02 1.19–2.44 1.02–2.00

0.0004 0.1409 0.1369 0.0957 0.0034 0.0396

1.94

1.28–3.02

0.0016

1.95

1.28–3.06

0.0016

1.31 –

0.90–1.94 –

0.1576 –

1.45 –

0.99–2.13 –

0.0583 –

CI = confidence interval, HR = hazard ratio. Multivariate analysis was performed using NYHA class in model 1 or peak VO2 in model 2.

310

H. Ohuchi et al. / International Journal of Cardiology 174 (2014) 306–312

Table 4 Univariate and multivariate predictors of mortality in all ACHD. Mortality Univariate

Patient characteristics Age (years) Male Body mass index NYHA class Peak VO2 (per 1.0 ml/kg/min) SV systolic function (normal = 1, reduced = 2, severe = 3) Arterial oxygen saturation (per 1%) Medications Diuretics Anti-coagulant ACEI/ARB Beta blocker Neurohumoral factors Norepinephrine (per 100 pg/ml) Renin activity (ng/ml/h) BNP (per 10 pg/ml) Glucose metabolism Low glucose (b80 mg/dl) Insulin (^U/ml) HbA1c (%) HOMA-IR High AUC PG (N6.315 g/dl) AGM vs. NGT

Multivariate (model 1)

HR

95% CI

p Value

1.06 0.90 0.94 4.90 0.81 1.48 0.91

1.01–1.10 0.35–2.30 0.81–1.06 2.65–9.65 0.73–0.88 0.74–2.92 0.87–0.96

0.0182 0.8188 0.3399 b0.0001 b0.0001 0.2613 0.0011

4.24 3.07 1.55 0.76

1.52–15.0 1.10–10.9 0.59–3.95 0.21–2.13

0.0049 0.032 0.3637 0.6208

1.14 1.01 1.02

0.97–1.30 0.98–1.04 1.00–1.03

0.1 0.4501 0.0532

5.46 0.90 1.59 0.58 0.68 1.05

1.95–19.3 0.76–1.02 0.81–2.54 0.26–1.03 0.26–1.72 0.40–2.67

0.0008 0.0930 0.1528 0.0635 0.4105 0.9156

Multivariate (model 2)

HR

95% CI

p Value

HR

95% CI

p Value

2.69 –

1.10–6.93 –

0.0302 –

– 0.90

– 0.79–1.01

– 0.0735

4.01

1.40–14.4

0.0085

4.04

1.41–14.5

0.0084

Abbreviations are same as in the Tables 1–3. Multivariate analysis was performed using NYHA class in model 1 or peak VO2 in model 2.

4.1. AGR and cardiovascular disease Many studies have demonstrated a close association between AGR and the pathophysiology of cardiovascular disease in patients with non-ACHD [1–7]. In addition to central obesity, insulin resistance due to activated sympathetic nervous system and/or renin–angiotensin– aldosterone system, change in skeletal muscle properties, and diuretic use have been considered as possible mechanisms for AGR in patients with HF [22–26]. Along with these mechanisms, renal dysfunction may also be involved [27,28]. Oxidative stress due to postprandial hyperglycemia impairs endothelial function [29], leading to arteriosclerosis, a major cause of ischemic heart disease. On the other hand, HF patients have a high prevalence of insulin resistance [4,30] and the insulin resistance-associated AGR is associated with the poor prognosis [30]. Furthermore, recent studies emphasize an important prognostic value of pre-diabetic AGR, i.e., IGT, because this impairment

a)

has a similar prognostic impact on cardiovascular patients when compared with that of diabetes mellitus [5]. In the present study, however, there was no such significant difference among the three AGR categorygroups, i.e., normal glucose regulation, IGT, and diabetes mellitus in terms of the AGR-group related prognosis. Shorter periods of AGR in our ACHD and the rare AGR-related organ complications, such as diabetic nephropathy, might explain the weak impact of AUC-PG on the morbidity and mortality. However, long-lasting adverse influences of AGR on the cardiovascular system must be of future concern, and furthermore, earlier intervention(s) targeting AGR is now being recommended because of the limited efficacy of current management for established AGR-associated pathophysiology [6]. Regarding the diagnosis of AGR, the 75-reg OGTT is recommended because of a poor association between FPG, HbA1c and IGT and/or diabetes mellitus [7,31,32]. In the care of young ACHD, a much longerterm meticulous follow-up strategy is essential to preserve better

b)

Free Rate (%) 100

(p = 0.0063)

90 FPG

80

90

70 80

FPG < 90

60 50

FPG < 80

40 0

20

40

60

80

100

Follow Up from 75g-OGTT (months) Fig. 2. The Kaplan-Meier clinical event-related hospitalization free rate curves were stratified into 3 groups by the fasting plasma glucose level (FPG) (a) and were stratified into 4 groups based on the symptoms (symptomatic patients ≥ New York Heart Association: NYHA class II) (b). The Kaplan-Meier curves were divided by the cutoff values according to area under the receiver-operator characteristic curve.

H. Ohuchi et al. / International Journal of Cardiology 174 (2014) 306–312

a)

311

b)

Free Rate (%) 100

80

FPG < 90 FPG

90

90 FPG < 80

80

(p = 0.0206) 70 0

20

40

60

80

100

Follow Up from 75g-OGTT (months) Fig. 3. The Kaplan-Meier all-cause mortality curves were stratified into 3 groups by the fasting plasma glucose level (FPG) (a) and were stratified into 4 groups based on the symptoms (symptomatic patients ≥ NYHA class II) (b). The Kaplan-Meier curves were divided by the cutoff values according to area under the receiver-operator characteristic curve.

cardiovascular function over their life-span and we believe that the 75 g-OGTT will provide us useful information not only for the management but also insights into the unique ACHD pathophysiology. In addition, AGR is now of concern for female ACHD who are pregnant because they have a high prevalence of AGR as shown here and even mild AGR during pregnancy has an adverse impact on the neonatal outcome [33]. Thus, 75 g-OGTT may also be important in the management of young female ACHD.

Furthermore, hypoglycemia induces a hypercoagulant state via platelet aggregation, changes of plasma coagulant factors, and increases plasma levels of endothelin-1 [41]. Thromboembolic events are sometimes lethal in unrepaired patients with right to left shunting and Fontan patients [42]. In general, because AGR is one of the causes of a hypercoagulant state [43], so hypoglycemia may trigger thromboembolic events in some ACHD with AGR. 4.3. Study limitations

4.2. Low FPG and ACHD This is the first demonstration of a prognostic importance of low FPG in ACHD. Recently, several studies have suggested a prognostic impact of low FPG as well as hyperglycemia at admission on an early stable condition in adults with acute myocardial infarction and HF [34–36], especially for the prediction of all-cause mortality. Several mechanisms for the association have been proposed. First, acute hypoglycemia induces catecholamine surge due to sympathetic nervous activation that leads to Ca2 + overload of the cardiomyocyte and hypokalemia both of which cause QT prolongation, one of the major precursors of lethal arrhythmias, such as ventricular tachycardia/fibrillation [37]. In fact, arrhythmia is the leading cause of hospitalization [38] in ACHD and pressure and/or volume load to cardiomyocyte due to abnormal hemodynamics and/or residual lesions which may be an additional substrate for arrhythmias. Second, a detrimental effect of hypoglycemia on myocardial metabolism with insulin resistance may lead to reduced SV contraction [39]. In addition, hypoglycemia may decrease myocardial blood flow reserve in ACHD, especially those with AGR [40].

a)

First, our subjects did not represent a “real world” ACHD population and they were sicker than those attending the usual ACHD outpatient clinic because invasive hemodynamic evaluation was required for some medical indications. Consequently, our present results may not apply to adults with simple CHD. However, symptomatic ACHD patients are those who require comprehensive managements and we believe that information respecting metabolic problems, including AGR, is vital for better management strategies in future ACHD practice. Second, the reproducibility of 75 g-OGTT is unclear and it should be reconfirmed in future studies. As for the reliability of FPG, we believe that overnight fasting before 75 g-OGTT guaranteed its validity. 5. Conclusion Low FPG was closely associated with both a high morbidity and mortality in symptomatic ACHD. In addition, we reconfirmed a high prevalence of AGR in ACHD. We believe that future comprehensive

b) Free Rate (%)

Free Rate (%) 100

(p = 0.1664)

DM

100

NGT

80

NGT

90 IGT

60

IGT

80

DM

40

(p = 0.9501)

20

70 0

20

40

60

80

Follow Up from 75g-OGTT (months)

100

0

20

40

60

80

100

Follow Up from 75g-OGTT (months)

Fig. 4. The Kaplan-Meier clinical event-related hospitalization free rate (a) and all-cause mortality free rate curves (b) were stratified into 3 groups by conventional criteria of abnormal glucose regulation, i.e., normal (NGT), impaired (IGT), and diabetes mellitus (DM).

312

H. Ohuchi et al. / International Journal of Cardiology 174 (2014) 306–312

management of symptomatic ACHD should include assessment of AGR including FPG if long-term outcomes are to be improved.

[21]

Acknowledgement We are grateful to Drs. Peter M. Olley, Adjunct Professor of Pediatrics, Sapporo Medical University, and Setsuko Olley for assistance in preparing the article. References [1] Kannel WB, Hjortland M, Castelli WP. Role of diabetes in congestive heart failure: the Framingham study. Am J Cardiol 1974;34:29–34. [2] Thrainsdottir IS, Aspelund T, Thorgeirsson G, et al. The association between glucose abnormalities and heart failure in the population-based Reykjavik study. Diabetes Care 2005;28:612–6. [3] MacDonald MR, Petrie MC, Varyani F, et al. CHARM Investigators. Impact of diabetes on outcomes in patients with low and preserved ejection fraction heart failure: an analysis of the Candesartan in Heart failure: assessment of reduction in mortality and morbidity (CHARM) programme. Eur Heart J 2008;29:1377–85. [4] Witteles RM, Tang WH, Jamali AH, Chu JW, Reaven GM, Fowler MB. Insulin resistance in idiopathic dilated cardiomyopathy: a possible etiologic link. J Am Coll Cardiol 2004;44:78–81. [5] Chiasson JL, Josse RG, Gomis R, Hanefeld M, Karasik A, Laakso M. STOP-NIDDM Trial Research Group. Acarbose treatment and the risk of cardiovascular disease and hypertension in patients with impaired glucose tolerance: the STOP-NIDDM trial. JAMA 2003;290:486–94. [6] ACCORD Study Group, Gerstein HC, Miller ME, et al. Long-term effects of intensive glucose lowering on cardiovascular outcomes. N Engl J Med 2011;364:818–28. [7] Bartnik M, Malmberg K, Norhammar A, Tenerz A, Ohrvik J, Rydén L. Newly detected abnormal glucose tolerance: an important predictor of long-term outcome after myocardial infarction. Eur Heart J 2004;25:1990–7. [8] Marelli AJ, Mackie AS, Ionescu-Ittu R, Rahme E, Pilote L. Congenital heart disease in the general population: changing prevalence and age distribution. Circulation 2007;115:163–72. [9] Shiina Y, Toyoda T, Kawasoe Y, et al. Prevalence of adult patients with congenital heart disease in Japan. Int J Cardiol 2011;146:13–6. [10] Ohuchi H, Miyamoto Y, Yamamoto M, et al. High prevalence of abnormal glucose metabolism in young adult patients with complex congenital heart disease. Am Heart J 2009;158:30–9. [11] Sharma R, Bolger AP, Li W, et al. Elevated circulating levels of inflammatory cytokines and bacterial endotoxin in adults with congenital heart disease. Am J Cardiol 2003;92:188–93. [12] Bolger AP, Sharma R, Li W, et al. Neurohormonal activation and the chronic heart failure syndrome in adults with congenital heart disease. Circulation 2002;106:92–9. [13] Ohuchi H, Takasugi H, Ohashi H, et al. Stratification of pediatric heart failure on the basis of neurohormonal and cardiac autonomic nervous activities in patients with congenital heart disease. Circulation 2003;108:2368–76. [14] Giannakoulas G, Dimopoulos K, Bolger AP, et al. Usefulness of natriuretic peptide levels to predict mortality in adults with congenital heart disease. Am J Cardiol 2010;105:869–73. [15] Ohuchi H, Nakajima T, Kawade M, Kamiya T. Measurement and validity of the ventilatory threshold in patients with congenital heart disease. Pediatr Cardiol 1996;17:7–14. [16] Mori K. Automated measurement of catecholamines in urine, plasma and hemogenates by high-performance liquid chromatography with fluorometric detection. J Chromatogr 1981;218:631–7. [17] Kono M, Yamaguchi A, Tsuji T, et al. An immunoradiometric assay for brain natriuretic peptide in human plasma. Kaku Igaku 1993;13:2–7. [18] Ikeda I, Iinuma K, Takai M, et al. Measurement of plasma renin activity by a simple solid phase radioimmunoassay. J Clin Endocrinol Metab 1982;54:423–8. [19] Puavilai G, Chanprasertyotin S, Sriphrapradaeng A. Diagnostic criteria for diabetes mellitus and other categories of glucose intolerance: 1997 criteria by the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus (ADA), 1998 WHO consultation criteria, and 1985 WHO criteria. World Health Organization. Diabetes Res Clin Pract 1999;44:21–6. [20] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function

[22]

[23]

[24]

[25]

[26] [27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37] [38] [39] [40]

[41] [42] [43]

from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28:412–9. Solomon SD, Wang D, Finn P, et al. Effect of candesartan on cause-specific mortality in heart failure patients: the Candesartan in Heart failure Assessment of Reduction in Mortality and morbidity (CHARM) program. Circulation 2004;110:2180–3. Desideri G, Ferri C, Bellini C, De Mattia G, Santucci A. Effects of ACE inhibition on spontaneous and insulin-stimulated endothelin-1 secretion: in vitro and in vivo studies. Diabetes 1997;46:81–6. Baron AD. The coupling of glucose metabolism and perfusion in human skeletal muscle. The potential role of endothelium-derived nitric oxide. Diabetes 1996;45(Suppl. 1):S105–9. Ran J, Hirano T, Fukui T, et al. Angiotensin II infusion decreases plasma adiponectin level via its type 1 receptor in rats: an implication for hypertension-related insulin resistance. Metabolism 2006;55:478–88. Kumashiro N, Tamura Y, Uchida T, et al. Impact of oxidative stress and peroxisome proliferator-activated receptor gamma coactivator-1alpha in hepatic insulin resistance. Diabetes 2008;57:2083–91. Zillich AJ, Garg J, Basu S, Bakris GL, Carter BL. Thiazide diuretics, potassium, and the development of diabetes: a quantitative review. Hypertension 2006;48:219–24. Meyer C, Woerle HJ, Dostou JM, Welle SL, Gerich JE. Abnormal renal, hepatic, and muscle glucose metabolism following glucose ingestion in type 2 diabetes. Am J Physiol Endocrinol Metab 2004;287:E1049–56. Dimopoulos K, Diller GP, Koltsida E, et al. Prevalence, predictors, and prognostic value of renal dysfunction in adults with congenital heart disease. Circulation 2008;117:2320–8. Kawano H, Motoyama T, Hirashima O, et al. Hyperglycemia rapidly suppresses flow-mediated endothelium-dependent vasodilation of brachial artery. J Am Coll Cardiol 1999;34:146–54. Doehner W, Rauchhaus M, Ponikowski P, et al. Impaired insulin sensitivity as an independent risk factor for mortality in patients with stable chronic heart failure. J Am Coll Cardiol 2005;46:1019–26. Egstrup M, Schou M, Gustafsson I, Kistorp CN, Hildebrandt PR, Tuxen CD. Oral glucose tolerance testing in an outpatient heart failure clinic reveals a high proportion of undiagnosed diabetic patients with an adverse prognosis. Eur J Heart Fail 2011;13:319–26. Rydén L, Standl E, Bartnik M, et al. Task Force on Diabetes and Cardiovascular Diseases of the European Society of Cardiology (ESC); European Association for the Study of Diabetes (EASD). Guidelines on diabetes, pre-diabetes, and cardiovascular diseases: executive summary. The Task Force on Diabetes and Cardiovascular Diseases of the European Society of Cardiology (ESC) and of the European Association for the Study of Diabetes (EASD). Eur Heart J 2007;28:88–136. Landon MB, Spong CY, Thom E, et al. Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network. A multicenter, randomized trial of treatment for mild gestational diabetes. N Engl J Med 2009;361:1339–48. Wei M, Gibbons LW, Mitchell TL, Kampert JB, Stern MP, Blair SN. Low fasting plasma glucose level as a predictor of cardiovascular disease and all-cause mortality. Circulation 2000;101:2047–52. Svensson AM, McGuire DK, Abrahamsson P, Dellborg M. Association between hyper- and hypoglycaemia and 2 year all-cause mortality risk in diabetic patients with acute coronary events. Eur Heart J 2005;26:1255–61. Ukena C, Dobre D, Mahfoud F, et al. Hypo- and hyperglycemia predict outcome in patients with left ventricular dysfunction after acute myocardial infarction: data from EPHESUS. J Card Fail 2012;18:439–45. Marques JL, George E, Peacey SR, et al. Altered ventricular repolarization during hypoglycaemia in patients with diabetes. Diabet Med 1997;14:648–54. Opotowsky AR, Siddiqi OK, Webb GD. Trends in hospitalizations for adults with congenital heart disease in the U.S. J Am Coll Cardiol 2009;54:460–7. Boudina S, Abel ED. Diabetic cardiomyopathy revisited. Circulation 2007;115:3213–23. Rana O, Byrne CD, Kerr D, et al. Acute hypoglycemia decreases myocardial blood flow reserve in patients with type 1 diabetes mellitus and in healthy humans. Circulation 2011;124:1548–56. Wright RJ, Frier BM. Vascular disease and diabetes: is hypoglycaemia an aggravating factor? Diabetes Metab Res Rev 2008;24:353–63. Tsang W, Johansson B, Salehian O, et al. Intracardiac thrombus in adults with the Fontan circulation. Cardiol Young 2007;17:646–51. Jax TW, Peters AJ, Plehn G, Schoebel FC. Hemostatic risk factors in patients with coronary artery disease and type 2 diabetes — a two year follow-up of 243 patients. Cardiovasc Diabetol 2009;8:48.

Low fasting plasma glucose level predicts morbidity and mortality in symptomatic adults with congenital heart disease.

Adults with complex congenital heart disease (ACHD) have a high prevalence of abnormal glucose regulation (AGR: impaired glucose tolerance and diabete...
503KB Sizes 0 Downloads 3 Views