Abnormal Glucose Metabolism and High-Energy Expenditure in Idiopathic Pulmonary Arterial Hypertension

Abstract

Rationale: Insulin resistance has emerged as a potential mechanism related to the pathogenesis of idiopathic pulmonary arterial hypertension (IPAH). However, direct measurements of insulin and glucose metabolism have not been performed in patients with IPAH to date.

Objectives: To perform comprehensive metabolic phenotyping of humans with IPAH.

Methods: We assessed plasma insulin and glucose, using an oral glucose tolerance test and estimated insulin resistance, and β-cell function in 14 patients with IPAH and 14 control subjects matched for age, sex, blood pressure, and body mass index. Body composition (dual-energy X-ray absorptiometry), inflammation (CXC chemokine ligand 10, endothelin-1), physical fitness (6-min walk test), and energy expenditure (indirect calorimetry) were also assessed.

Measurements and main results: Patients with IPAH had a higher rate of impaired glucose tolerance (57 vs. 14%; P < 0.05) and reduced glucose-stimulated insulin secretion compared with matched control subjects (IPAH: 1.31 ± 0.76 μU/ml⋅mg/dl vs. control subjects: 2.21 ± 1.27 μU/ml⋅mg/dl; P < 0.05). Pancreatic β-cell function was associated with circulating endothelin-1 (r = -0.71, P < 0.01) and CXC chemokine ligand 10 (r = -0.56, P < 0.05). Resting energy expenditure was elevated in IPAH (IPAH: 32 ± 3.4 vs. control subjects: 28.8 ± 2.9 kcal/d/kg fat-free mass; P < 0.05) and correlated with the plasma glucose response (r = 0.51, P < 0.01). Greater insulin resistance was associated with reduced 6-minute walk distance (r = 0.55, P < 0.05).

Conclusions: Independent of age, sex, blood pressure, and body mass index, patients with IPAH have glucose intolerance, decreased insulin secretion in response to glucose, and elevated resting energy expenditure. These abnormalities are associated with circulating markers of inflammation and vascular dysfunction.

Keywords: energy expenditure; glucose metabolism disorder; pulmonary hypertension.

Figure 1.

Basal glucose, insulin, and glycated hemoglobin levels. Fasting plasma glucose (idiopathic pulmonary arterial hypertension [IPAH]: 87 ± 10 mg/dl vs. control subjects: 94 ± 12 mg/dl) and glycated hemoglobin (HbA_1c) (IPAH: 5.58 ± 0.2% vs. control subjects: 5.71 ± 0.4%) were similar in patients with IPAH and matched control subjects (both, P > 0.05), whereas fasting insulin concentration was lower in patients with IPAH (IPAH: 9.1 ± 4.8 μU/ml vs. control subjects: 16.4 ± 10.9 μU/ml; P < 0.05). The bottom and top of each box represent the 25th and 75th percentiles, respectively; the band inside each box indicates the median value.

Figure 2.

Impaired fasting glucose (IFG) and impaired glucose tolerance (IGT) in patients with idiopathic pulmonary arterial hypertension (IPAH) and matched control subjects. One of 14 patients with IPAH (7%) had IFG, compared with 5 of 14 control subjects (36%) (P = 0.06). Eight of 14 patients with IPAH (57%) had IGT compared with only 2 of 14 matched control subjects (14%) (P = 0.02).

Figure 3.

Glucose and insulin response during the oral glucose tolerance test. (A) Glucose and insulin response curves in response to a 75-g oral glucose load in young lean healthy volunteers, patients with idiopathic pulmonary arterial hypertension (IPAH), matched control subjects, and patients with type 2 diabetes mellitus (DM2). Error bars represent the SEM. Patients with IPAH showed a trend to higher glucose levels compared with matched control subjects at time points 90 and 120 minutes (P values of 0.12 and 0.13, respectively). Patients with IPAH had a lower insulin response than did matched control subjects (P < 0.05 for time points 30 and 90 min, P = 0.06 for time point 60 min). (B) Boxplots of the incremental area under the curve (iAUC) for glucose and insulin during the oral glucose tolerance test. Mean glucose response in patients with IPAH was 10,408 ± 3,693 versus 8,177 ± 2,931 mg/dl⋅2 h in matched control subjects. Mean insulin response in IPAH was 11,674 ± 5,438 versus 16,803 ± 8,079 μU/ml⋅2 h in matched control subjects. The bottom and top of each box represent the 25th and 75th percentiles, respectively; the band inside each box represents the median value.

Figure 4.

Pancreatic β-cell function indices. Glucose-stimulated insulin secretion (GSIS) was calculated as the insulin incremental area under the curve divided by the glucose incremental area under the curve during the oral glucose tolerance test. GSIS was lower in patients with idiopathic pulmonary arterial hypertension (IPAH) compared with control subjects (IPAH: 1.31 ± 0.76 μU/ml⋅mg/dl vs. control subjects: 2.21 ± 1.27 μU/ml⋅mg/dl). The disposition index was calculated by multiplying GSIS by the Stumvoll index to account for varying degrees of insulin sensitivity, and it was lower in patients with IPAH (0.96 ± 0.48 a.u.) compared with control subjects (1.48 ± 0.74 a.u.). The bottom and top of each box represent the 25th and 75th percentiles, respectively; the band inside each box represents the median value. a.u. = arbitrary units.

Figure 5.

Substrate use and resting energy expenditure. (A) Respiratory quotient, a measure of substrate use, was similar in patients with idiopathic pulmonary arterial hypertension (IPAH) and control subjects. (B) Resting energy expenditure (REE) per kilogram of fat-free mass was higher in patients with IPAH than in control subjects. See Table 3 for values. The bottom and top of each box represent the 25th and 75th percentiles, respectively; the band inside each box represents the median value. (C) REE was associated with the glucose response. iAUC = incremental area under the curve.

Figure 6.

Correlations with inflammatory plasma biomarkers and insulin secretion. (A) Plasma IL-6 (4.47 ± 3.61 vs. 2.35 ± 1.73 pg/ml), CXC chemokine ligand 10 (CXCL10) (183 ± 102 vs. 109 ± 26 pg/ml), and endothelin-1 (3.2 ± 2.05 vs. 1.41 ± 0.47 pg/ml) were elevated in patients with idiopathic pulmonary arterial hypertension (IPAH) compared with control subjects. (B) Inverse correlation between CXCL10 and the insulin response during the oral glucose tolerance test. (C) Direct correlation between CXCL10 and resting energy expenditure (REE). (D) Inverse correlation between endothelin-1 (ET-1) and CXCL10 and indices of pancreatic β-cell function: glucose-stimulated insulin secretions (GSIS) and the disposition index (DI). The bottom and top of each box represent the 25th and 75th percentiles, respectively; the band inside each box represents the median value. FFM = fat-free mass; iAUC = incremental area under the curve.

Figure 7.

Correlation between insulin resistance and the 6-minute walk distance (6MWD). The percent predicted 6MWD had a moderate positive correlation with skeletal muscle insulin sensitivity. A higher Stumvoll index suggests less insulin resistance (i.e., more insulin sensitivity) and is associated with increased walk distance.