It has been suggested that lipoprotein abnormalities may contribute to the pulmonary arteriolar dysfunction observed in pulmonary arterial hypertension (PAH). High-density lipoprotein cholesterol (HDL) has vasodilatory, anti-inflammatory, and endothelial protective properties. We hypothesized that a higher serum HDL level may be advantageous for survival in PAH and that the serum HDL level at diagnosis would be an independent predictor of survival in PAH and be additive to previously validated predictors of survival. This study included all patients with PAH seen at the Mayo Clinic Pulmonary Hypertension Clinic from January 1, 1995, to December 31, 2009, who had a baseline HDL measurement. Mortality was analyzed over 5 years using the Kaplan–Meier method. Univariate and multivariable Cox proportional hazards ratios were calculated to evaluate the relation between baseline HDL level and survival. HDL levels were available for 227 patients. Higher HDL levels were associated with significantly lower mortality. Patients with an HDL >54 mg/dl at diagnosis had a 5-year survival of 59%. By comparison those with an HDL <34 mg/dl had a 5-year survival of 30%. On multivariate analysis, higher HDL was associated with an age-adjusted risk ratio for death of 0.78 (CI 0.67 to 0.91; p <0.01) per 10 mg/dl increase. In conclusion, HDL was an independent predictor of survival in PAH.
The pathophysiology of pulmonary arterial hypertension (PAH) involves derangements in the nitric oxide, prostacyclin, and endothelin cellular signaling pathways which are associated with pulmonary vascular remodeling, elevated pressures in the pulmonary arteries, and ultimately right ventricular failure. The estimated 5-year survival rate of patients with PAH is 50% to 60%. Although there are a large number of clinical and hemodynamic predictors of survival in PAH, studies of specific lipoprotein biochemical predictors of survival in PAH are few and poorly validated. High-density lipoprotein cholesterol (HDL) has been extensively studied for its role in the pathogenesis of systemic arterial disease. A higher HDL level has been found to be protective against atherosclerotic vascular disease by multiple mechanisms. Many of these mechanisms could also be protective in PAH. HDL promotes activation of endothelial nitric oxide synthase, which produces the potent vasodilator nitric oxide. In addition, HDL promotes the release of prostacyclin from endothelial cells and stabilizes prostacyclin, thus prolonging its half-life. Finally, HDL has been demonstrated to promote normalization of endothelial-dependent vascular reactivity. Previous smaller studies evaluating the relation between HDL levels and outcomes in PAH have had mixed results, so it remains unclear whether baseline HDL levels are predictive of long-term survival in PAH. We hypothesized that HDL level at diagnosis would be an independent predictor of survival in PAH and be additive to previously validated predictors of survival.
Methods
All patients diagnosed with group 1 pulmonary hypertension at the Pulmonary Hypertension Clinic at Mayo Clinic Rochester from January 1, 1995, to December 31, 2009, were identified from the Pulmonary Hypertension Clinic registry after institutional review board approval was obtained (IRB# 13-009774). Patients were included if they fulfilled contemporary diagnostic criteria for idiopathic, connective tissue disease associated, or portopulmonary PAH, and they had an HDL level checked within 3 months before or 1 month after diagnosis. All included patients had previously consented to the use of their medical record for research purposes as per Minnesota state law. Patients with PAH secondary to congenital heart disease were excluded because their survival on average is better than other patients with group 1 PAH.
A retrospective review of the medical record was performed to collect baseline clinical data obtained during the initial diagnostic workup. The availability of baseline data for study participants was as follows: HDL (n = 227), low-density lipoprotein cholesterol (n = 216), triglycerides (n = 227), fasting glucose (n = 201), echocardiogram (n = 227), 6-minute walk (n = 161), and pulmonary function tests (n = 161). All patients underwent a right-sided cardiac catheterization before diagnosis. The multicenter, observational US-based Registry to Evaluate Early and Long-term Pulmonary Arterial Hypertension Disease Management (REVEAL) Registry risk score was calculated as previously described. Patients were classified as having coronary artery disease if they had a previous angiogram demonstrating any degree of coronary artery disease, had previously undergone coronary revascularization, or had a previous myocardial infarction. Information regarding disease specific treatments, transplantation, and mortality was also collected. Lipid-lowering therapy reflects medication use at the time of baseline laboratory studies. Targeted therapy for PAH reflects treatment prescribed subsequent to diagnosis and was defined as one or more of the following agents: endothelin receptor antagonists, phosphodiesterase-5 inhibitors, or parenteral prostaglandins. Mortality data were supplemented by review of the Social Security Death Index.
Data analysis was performed using JMP, version 10 (SAS Institute Inc., Cary, North Carolina). p Values <0.05 were considered statistically significant. Continuous variables that followed a normal distribution are presented as mean ± SD, continuous variables that did not follow a normal distribution are presented as median (25th to 75th percentile), and categorical variables are presented as number (percentage). Between-group differences were assessed by analysis of variance, Pearson chi-square, t tests, or Wilcoxon tests as appropriate. Survival was assessed using the Kaplan–Meier method and compared between groups using the log-rank test. Patients were censored at 5 years or date of transplant. The correlation between HDL and mortality was assessed by univariate and multivariable Cox proportional hazards modeling to adjust for known predictors of disease severity in PAH and for laboratory markers associated with HDL cholesterol. All baseline variables significantly associated with mortality by univariate analysis were initially included in a multivariable analysis. The least significant variables were then removed one at a time in a stepwise fashion until all variables remaining in the model had a p value <0.1. To handle incomplete test data, a categorical variable specifying whether a given test was performed for each patient was added to the model. Results are expressed as hazard ratios (95% CIs). Variables reflecting whether a patient received a phosphodiesterase-5 inhibitor, endothelin receptor antagonist, or parenteral prostaglandin were then added to the model to adjust for the influence of targeted therapy on survival. Two additional multivariate Cox proportional hazards analyses were performed. One of which included age, the REVEAL risk score, HDL, and the 3 variables for targeted therapy described previously. The second analysis included age, gender, body mass index, coronary artery disease, diabetes, low-density lipoprotein cholesterol, triglycerides, fasting glucose, and statin, fibrate, and niacin use. Baseline natriuretic peptide levels were not included in the multivariate analyses as they were not available in sufficient patient numbers from the early years of this study.
Results
HDL levels were available for 227 of 632 patients diagnosed with PAH from January 1, 1995, to December 31, 2009. Patients with HDL levels were more likely to be men (30% vs 22%; p <0.05) but were similar to those without HDL levels with respect to age, World Health Organization functional class, cardiac index, 6-minute walk distance, diffusing capacity of the lungs for carbon monoxide, and 5-year survival (p >0.05). The mean follow-up was 175 ± 99 weeks, and the median follow-up was 229 weeks (81 to 266 weeks). Patients were divided into 3 groups by HDL level: lowest quartile (HDL <34 mg/dl), middle 50% (HDL 34 to 54 mg/dl), and upper quartile (HDL >54 mg/dl). Baseline characteristics of the study population are presented in Table 1 . Higher HDL levels were associated with significantly greater 5-year survival overall ( Figure 1 ). The association between higher HDL and improved survival was also present for the subgroups of patients with idiopathic and connective tissue disease associated PAH ( Figure 2 ).
| Overall (n=227) | High Density Lipoprotein Cholesterol (mg/dL) | P value | |||
|---|---|---|---|---|---|
| < 34 (n=51) | 34-54 (n=119) | > 54 (n=57) | |||
| Age (years) | 55 ± 14 | 52 ± 12 | 53 ± 15 | 61 ± 11 | <0.01 |
| Female | 162 (71%) | 37 (73%) | 83 (69%) | 42 (75%) | 0.7 |
| Etiology of pulmonary arterial hypertension | |||||
| Idiopathic | 138 (61%) | 29(57%) | 76 (64%) | 33 (58%) | 0.3 |
| Connective tissue disease | 59 (26%) | 13 (25%) | 33 (28%) | 13 (23%) | |
| Portopulmonary | 30 (13%) | 9 (18%) | 10 (8%) | 11 (19%) | |
| World Health Organization functional class III/IV | 158 (71%) | 43 (86%) | 82 (69%) | 33 (61%) | 0.01 |
| 6 minute walk distance (m) | 330 (229 – 416) | 277 (178 – 373) | 331 (238 – 423) | 339 (273 – 427) | 0.1 |
| Diffusing capacity of the lungs for carbon monoxide (%) | 57 (44 – 78) | 59 (45 – 90) | 57 (41 – 77) | 54 (45-79) | 0.8 |
| REVEAL score | 8.1 ± 1.9 | 8.5 ± 2.0 | 8.1 ± 1.8 | 8.0 ± 2.0 | 0.2 |
| Severe right ventricular enlargement | 117 (52%) | 36 (71%) | 63 (53%) | 18 (32%) | <0.01 |
| Mean pulmonary arterial pressure (mmHg) | 51 ± 13 | 54 ± 13 | 52 ± 13 | 47 ± 13 | 0.03 |
| Cardiac index (L/min/m 2 ) | 2.4 (1.9-3.0) | 2.2 (1.8-3.1) | 2.4 (2.0-2.9) | 2.6 (2.1-3.3) | 0.3 |
| Right atrial pressure (mmHg) | 11 (7-16) | 15 (11-19) | 10 (7-14) | 10 (6-14) | <0.01 |
| Diabetes mellitus | 39 (17%) | 12 (24%) | 17 (14%) | 10 (18%) | 0.3 |
| Coronary artery disease | 50 (22%) | 10 (20%) | 24 (20%) | 16 (28%) | 0.44 |
| Low density lipoprotein cholesterol (mg/dL) | 90 (70 – 109) | 83 (65 – 104) | 93 (74 – 110) | 91 (60 – 117) | 0.1 |
| Triglycerides (mg/dL) | 126 (81 – 170) | 146 (107 -201) | 123 (79 – 172) | 115 (77 – 148) | <0.01 |
| Fasting glucose (mg/dL) | 99 (89 – 111) | 100 (87 – 114) | 98 (89 – 111) | 100 (91 – 111) | 0.9 |
| Baseline lipid therapy | |||||
| Statin | 37 (16%) | 5 (10%) | 21 (18%) | 11 (19%) | 0.4 |
| Fibrate | 7 (3%) | 2 (4%) | 2 (2%) | 3 (5%) | 0.4 |
| Niacin | 1 (<1%) | 0 | 1 (<1%) | 0 | 0.6 |
| Subsequent targeted therapy | |||||
| Endothelin receptor antagonist | 154 (68%) | 38 (75%) | 79 (66%) | 37 (66%) | 0.5 |
| Phosphodiesterase 5 inhibitor | 157 (70%) | 37 (73%) | 85 (71%) | 35 (63%) | 0.5 |
| Parenteral prostaglandin | 95 (42%) | 23 (45%) | 50 (42%) | 22 (39%) | 0.8 |
On univariate Cox proportional hazards analysis, higher HDL levels were significantly associated with mortality ( Table 2 ). In women, HDL levels were associated with an age-adjusted risk ratio for death of 0.76 (CI 0.64 to 0.90; p <0.01) per 10 mg/dl increase. A similar trend was observed in men but did not reach statistical significance. In the stepwise exclusion multivariate model, HDL was an independent predictor of mortality, as were etiology of PAH, diffusing capacity of the lungs for carbon monoxide, 6-minute walk distance, estimated glomerular filtration rate, and cardiac index ( Table 3 ). When variables for targeted therapy use were added to this model, HDL level remained a significant predictor of mortality (0.80 [0.68 to 0.93; p <0.01] per 10 mg/dl increase). On a separate multivariate analysis adjusting for age, gender, body mass index, coronary artery disease, diabetes, low-density lipoprotein cholesterol, triglycerides, fasting glucose, and statin, fibrate, and niacin use, HDL remained an independent predictor of mortality (hazard ratio 0.73 [0.62 to 0.86; p <0.01] per 10 mg/dl increase). Finally, HDL was an independent predictor of mortality (hazard ratio 0.81 [0.71 to 0.92]; p <0.01 per 10 mg/dl increase) in a multivariate model that included age, the REVEAL risk score, and variables reflecting use of targeted therapy with endothelin receptor antagonists, phosphodiesterase-5 inhibitors, and/or parenteral prostaglandins.
| Age-adjusted univariate hazard ratios | P value | |
|---|---|---|
| Sex (Female versus Male) | 0.88 (0.60-1.31) | 0.52 |
| World Health Organization functional class (I, II, III, IV) | 1.46 (1.11-1.93) | <0.01 |
| Etiology (connective tissue disease or portopulmonary vs. idiopathic) | 2.10 (1.45-3.04) | <0.01 |
| 6 minute walk distance (per 50 m increase) | 0.82 (0.78-0.87) | <0.01 |
| Diffusing capacity of the lungs for carbon monoxide (by 10% increase) | 0.94 (0.89-0.99) | <0.01 |
| Estimated glomerular filtration rate <50 (vs. >50 mL/min per 1.73 m 2 ) | 2.08 (1.37-3.12) | <0.01 |
| Echocardiographic parameters | ||
| Severe right ventricular enlargement | 1.81 (1.24-2.66) | <0.01 |
| Estimated right atrial pressure (per 5 mmHg increase) | 1.32 (1.07-1.63) | <0.01 |
| Right Heart Catheterization parameters | ||
| Mean pulmonary arterial pressure (per 10 mmHg increase) | 1.06 (0.91-1.22) | 0.46 |
| Cardiac index (per 1 L/min/m 2 increase) | 0.77 (0.61-0.98) | 0.03 |
| Right atrial pressure (per 5 mmHg increase) | 1.39 (1.18-1.62) | <0.01 |
| REVEAL Score | 1.34 (1.22-1.47) | <0.01 |
| High density lipoprotein cholesterol (per 10 mg/dL increase) | 0.74 (0.64-0.85) | <0.01 |
| Low density lipoprotein cholesterol (per 20 mg/dL increase) | 0.90 (0.80-1.00) | 0.06 |
| Triglycerides (per 50 mg/dL increase) | 0.98 (0.86-1.10) | 0.70 |
| Fasting glucose (per 40 mg/dL increase) | 1.10 (0.92-1.28) | 0.28 |
| Coronary artery disease | 0.92 (0.58-1.41) | 0.72 |
| Diabetes mellitus | 0.75 (0.44-1.28) | 0.2 |
| Statin use | 0.96 (0.57-1.54) | 0.9 |
| Fibrate use | 1.19 (0.36-2.84) | 0.7 |
| Niacin use | 1.39 (0.08-6.22) | 0.8 |
| Phosphodiesterase 5 inhibitor | 0.32 (0.19-0.52) | <0.01 |
| Endothelin receptor antagonist | 0.48 (0.31-0.73) | <0.01 |
| Parenteral prostaglandin | 1.66 (1.13-2.45) | 0.01 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
