Pulmonary Hypertension in Chronic Renal Failure Patients

Cardiovascular disease is the leading cause of morbidity and mortality in dialysis patients, accounting for 50% of deaths [1]. Pulmonary hypertension (PHT) can be the result of heart, lung or systemic disorders [2]. Regardless of the etiology, morbidity and mortality from long-standing PHT exceed that expected from the causative condition [3]. PHT is frequently found in patients with chronic renal failure (CRF). In one study, PHT was found in 40% of patients with end-stage renal disease (ESRD) on chronic haemodialysis therapy via arteriovenous access [4, 5]. Adekunle et al. [6] stated that PHT is an independent predictor of mortality in ESRD patients.The pathogenesis of PHT in patients with CRF has not been explained satisfactorily [7]. PHT involves vasoconstriction and obliteration of the lumen of small vessels in the lungs by plexiform lesions, resulting in increased resistance to flow [8]. Hormonal and metabolic derangement associated with ESRD might lead to pulmonary arterial vasoconstriction and an increase of the pulmonary vascular resistance [9].Pulmonary artery pressure (PAP) may be further increased by high cardiac output resulting from the arteriovenous access itself, worsened by commonly occurring anemia and fluid overload [10]. Subclinical left ventricular dysfunction also occurs in patients with ESRD, and is evidenced as abnormal myocardial diastolic rather than systolic dysfunction [11].Local vascular tone and function of pulmonary vessels are regulated by the balance between vasodilators, such as nitric oxide, and vasoconstrictors, such as thromboxane [12]. Patients with CRF show an endothelial dysfunction related to defective nitric oxide activity, which is not corrected by hemodialysis (HD) [13].Increased brain natriuretic hormone is associated with age, left ventricular hypertrophy, renal failure, and PHT. In the near future, brain natriuretic hormone will become an interesting new therapeutic modality [14]. N-terminal pro-brain natriuretic peptide (NT-proBNP) is a byproduct of brain natriuretic peptide (BNP) that has been shown to be of prognostic value in PHT [15].The aim of this study is to evaluate the prevalence of unexplained PHT in ESRD patients, and to explore some possible etiologic factors for this complication. The study was carried out on 45 patients with ESRD on regular HD (group 1) through arteriovenous fistula (AVF) 3 times per week in 4-hour sessions, and 31 patients with ESRD on conservative treatment (group 2). Exclusion criteria were chronic obstructive lung disease, chest wall disease or parenchymal lung disease, previous history of PHT, previous pulmonary embolism, collagen vascular disease, left ventricular ejection fraction ≤50%, and significant mitral or aortic valve disease.HD patients were being treated with standard bicarbonate dialysis (Na 138 mmol/l, HCO3 35 mmol/l, K 1.5 mmol/l, Ca 1.25 mmol/l, Mg 0.75 mmol/l) using low-flux HD with either cuprophan or semisynthetic membranes. Subjects gave their informed consent and the study protocol was approved by the institute’s Committee on Human Research.The following was undertaken in all patients:(1) Thorough medical history and clinical examination: A complete history was recorded with special emphasis to detect any clinical condition that predisposes toPHT, comorbidities, and history of renal disease (etiology of renal failure, age at onset, duration of HD therapy, and access location, brachial or radial).(2) Chest radiography and a standard 12-lead ECG: ECG findings suggestive of PHT were as follows: (1) right axis deviation; (2) a tall R-wave and small S-wave with a R/S ratio >1 in lead V1; (3) qR complex in lead V1; (4) rSR′ pattern in lead V1; (5) a large S-wave and small R-wave with a R/S ratio <1 in lead V5 or V6; or (6) S1, S2, S3 pattern. ST-T segment wave depression and inversion are often present in the right precordial leads. Right atrial enlargement is manifested as a tall P-wave (2.5 mm) in leads II, III, and AVF and frontal P-axis of 75° [16].(3) Laboratory investigations: Blood tests for BUN, creatinine, hemoglobin, hematocrit, calcium, phosphorus, alkaline phosphatase, and parathyroid hormone (PTH) level were sampled at the end of the HD therapy within 1 week after the echocardiography study. For regular tests (BUN, creatinine, hemoglobin, hematocrit, calcium and phosphorus) the mean values from the 6 months preceding the echo study were presented. Intact molecule PTH was measured by radioimmunoassay using commercially available kits (Biosource Europe SA, Nivelles, Belgium) [17].(4) Doppler echocardiography: Two-dimensional and M-mode echocardiography were performed in all patients within 4 h following dialysis when the patients were at optimal dry weight according to hydration status, blood pressure and weight. Imaging was performed with Hewlett-Packard (Bathell, Wash., USA) Sonos 1000 equipped with 2.5- and 3.2-MHz phased pulsed-array transducers. Cardiac dimensions were measured according to the guidelines of the American Society of Echocardiography [18].Systolic right ventricular (or pulmonary artery) pressure was calculated using the modified Bernoulli equation: PAP = 4 × (tricuspid systolic jet)2 + 10 mm Hg (estimated right atrial pressure) [19]. Stroke volume estimated from the left ventricular outflow tract velocity time integral × diameter cardiac output was calculated by multiplying the stroke volume by the heart rate. PHT was defined as a systolic PAP >35 mm Hg [20].Diastolic dysfunction was assessed by using E/E′, where E is early mitral valve flow velocity and E′ early diastolic lengthening velocities. When the ratio E/E′ exceeds 15, left ventricular filling pressures are elevated and when the ratio is <8, left ventricular filling pressures are low [21].(5) AVF flow: In HD patients using color Doppler ultrasound, examinations were performed off HD by the same radiologist using a 5- to 10-MHz linear transducer. The arterial and venous blood flow rates were calculated using the equation: flow (ml/min) = time average velocity (mean of three cardiac cycles) (m/s) × cross-sectional area (mm2) × 60 [22].(6) NT-proBNP: NT-proBNP was detected in serum and plasma using commercially available assays, the half-life of BNP is much shorter (20 min) compared with NT-proBNP (120 min), and therefore NT-proBNP were selected from a clinical diagnostic perspective. After the patient had rested in the supine position for at least 20 min, blood samples for determination of NT-proBNP were collected in EDTA tubes, centrifuged and plasma stored at –80°C until analysis. All samples only underwent one freeze-thaw cycle for out-portioning and were otherwise treated and stored under the same conditions. Plasma concentrations of NT-proBNP were measured by a sandwich immunoassay on an Elecsys 2010 (Roche Diagnostics, Basel, Switzerland). The intra-assay variation was <3.0% and the total coefficient of variation ranged from 2.2 to 5.8% for low and high ranges of NT-proBNP, respectively.(7) TXB2: A 10-ml specimen of blood was drawn from the antecubital vein through a 19-gauge needle into the cooled syringe containing 0.1 ml of acetylsalicylate solution. After gentle mixing by inversion, the specimen was transferred into an iced siliconized Vacutainer tube containing 200 USP units of heparin. The cold tubes were centrifuged immediately (2,600 g, 10 min at room temperature). The plasma was then apportioned into polypropylene tubes and analyzed, either on the same day or stored refrigerated or frozen until assay.To 12 × 75 mm glass tubes, we added 200 µl phosphate buffer, 100 µl plasma, and 100 µl antibody (working dilution). The mixture was incubated overnight at 4°C. 50 µl (25 pg) of [3H]-TXB2 was added and incubated for 4 h at 4°C. 500 µl of solid-phase second antibody was added and incubated for 45 min at 4°C, centrifuged (5,140 g, 10 min at 4°C), the supernate decanted, the precipitate resuspended in 5 ml of Biofluor scintillation fluid (New England Nuclear), and the radioactivity of the suspension counted.SPSS for Windows Version 15 was used for analysis of data as follows: description of quantitative variables in the form of mean, SD and range; description of qualitative variables in the form of frequency and percentage. An independent sample t test was used to compare quantitative variables and a χ2 test was used to compare qualitative variable correlation coefficient tests (R-test) used to rank different variables against each other either directly or indirectly. A multiple regression test was used to determine the importance of different variables causing PHT.The two groups studied were comparable with regard to age (t = 0.07, p = 0.94) (group 1: 26 males/19 females; group 2: 17 males/14 females). The demographic and dialysis data of the groups studied are presented in table 1. The etiology of renal disease is presented in table 2.PHT (>35 mm Hg) was found in 20 patients (44.4%) in group 1 and in 10 patients (32.3%) in group 2. The biochemical data of groups 1 and 2 are presented in table 3. PAP in group 1 and 2 is shown in figure 1. The patients were classified into two groups according to the presence or absence of PHT. Both groups are compared with regard to different variables in table 4. Patients with PHT had a significantly higher AVF blood flow (t = 5.9, p < 0.001) (fig. 2), proBNP (t = 7.9, p < 0.001), TXB2 (t = 5.9, p < 0.001) (fig. 2), age (t = 2.1, p = 0.03), BUN (t = 1.7, p = 0.09) and serum creatinine (t = 1.8, p = 0.06) than those without PHT. No significant differences were found between the two groups with regard to serum calcium (t = –0.86, p = 0.38), serum phosphorus (t = 0.37, p = 0.7), intact PTH (t = 1.2, p = 0.21), and hemoglobin (t = –0.7, p = 0.48).In patients with PHT, 35% (7 patients) had radial arteriovenous access and 65% (13 patients) had brachial access. Also, 76.7% (23 patients) had diastolic dysfunction and 23.3% (7 patients) did not. Multiple regression analysis for PHT with different variables shows that AVF flow is the most significant variable (t = 4.0, p < 0.001), followed by proBNP (t = 3.5, p = 0.005). A correlation study showed that PHT correlates with AVF flow (r = 0.7, p < 0.001), proBNP (r = 0.8, p < 0.001) (fig. 3), and TXB2 (r = 0.6, p < 0.001) (fig. 4). Risk calculation for PHT shows that patients with brachial arteriovenous access have an odds ratio of 2.3 (95% CI 0.7–7.9), and patients with diastolic dysfunction have an odds ratio of 21.9 (95% CI 6.5–73.0).We studied PHT in HD patients and in patients with ESRD in the predialysis stage. Elevated pulmonary artery systolic pressure (>35 mm Hg) was found in 44.4% in group 1 and 32.3% in group 2.Our results show that mean PAP is significantly higher in HD patients than predialysis patients. Ishimura et al. [23] stated that both ESRD and long-term HD via AVF might be involved in the pathogenesis of PHT. The cause of a higher PAP in group 1 may be related to the process of HD or the hemodynamic changes caused by AVF. Creation of an AVF produces significant elevations in left ventricular end-diastolic diameter, fractional shortening, and cardiac output [24]. Thus, increasing right ventricular cardiac output [4]and LVDD produced by AVF contribute to elevating PAP. Pulmonary arterial remodeling occurs following creation of a systemic AVF, leading to flow-induced PHT [25].The process of HD itself may be a contributing factor for elevated PAP, but the exact cause is not known and vasoconstrictors such as endothelin may be involved [4]. Microbubble emboli are another cause. The recurrent microbubble lung injury may explain the increased PAP, of long-term dialysis patients [26]. In addition, HD causes recurrent episodes of hypoxemia due to partial blockage of the pulmonary capillary bed by white cells or silicone microemboli [27]. Recurrent hypoxemia is associated with elevation of PAP.We found a significant difference between groups 1 and 2 with regard to TXB2 and proBNP. The increase of TXB2 in the HD group may be related to the process of HD which induces a detectable extracorporeal increase of TXB2 through blood membrane interaction causing degranulation of neutrophils [28]. NT-proBNP is possibly largely eliminated by glomerular filtration only [29], so its level is elevated in patients with renal failure. NT-proBNP concentrations can be increased in the setting of HD [30].Our results show that patients with PHT have a significantly higher age than those without PHT; they also have higher levels of BUN and serum creatinine. Nakhoul et al. [31 ]and Amin et al. [32 ]did not find a significant difference with regard to age between HD patients with and without PHT. The effect of age on PHT is not clear however; in PHT, intravascular thrombosis in pulmonary arteries occurs with increasing age [33].The presence of a significantly higher BUN and serum creatinine levels in a group with PHT was reported by Amin et al. [32], but Nakhoul et al. [31] did not find a significant difference with regard to BUN and serum creatinine. The effect of uremic toxin on PAP had been postulated as an etiological factor in HD-associated PHT through endothelial dysfunction that has been described in PHT [34] and uremia [35], which diminishes the vasodilatation response to the arteriovenous access-induced elevated cardiac output.When comparing patients with elevated PAP to those with normal PAP, patients with a high PAP have a higher AVF blood flow rate, which indicates that blood shunt through AVF is an important cause of elevated PAP. The effect of an increase in blood flow on microvascular pressures depends on the state of recruitment of the capillary bed. When the lung is already fully recruited, increases in blood flow result in large increases in pulmonary artery and venous pressures [36].Previous studies have emphasized the role of AVF in the development of PHT in renal patients [31, 37, 38]. The Vascular Access Society guidelines [39 ]define an AVF with a high flow as that having a flow of 1.0–1.5 l/min and a cardiopulmonary recirculation >20%.The characteristic vascular changes due to increased pulmonary blood flow include intimal hyperplasia/fibrosis, medial hypertrophy, and formation of plexiform lesions. These changes lead to a decreased compliance of the pulmonary vasculature [40]. Patients with PHT show no significant difference from those without regarding hemoglobin, serum calcium, serum phosphorus and intact PTH. Similar findings have also been reported by Nakhoul et al. [31] and Amin et al. [32].Although anemia is associated with a compensatory increased cardiac output and is expected to be one of the causes of increased cardiac output, it seems that its contribution is minimal in patients with PHT. In the group with PHT, more patients have brachial AVF than radial ones, the reverse is the case in the group without PHT, yet the difference between the two groups is not significant. It therefore appears that arteriovenous access blood flow rather than the site is the factor related to the development of PHT.Diastolic dysfunction of the heart is significantly higher in the group with PHT [41]. Faraz et al. [42] stated that diastolic dysfunction contributes to PHT in combination with other factors which matches our results. Diastolic dysfunction may contribute to the development of PHT by causing an elevated left atrial pressure. Calculating the risk for development of PHT shows that patients with brachial AVF have a more than double risk for development of PHT, while patients with LVDD have a very high risk compared to those without LVDD.In the group with PHT, TXB2 is significantly higher than those without PHT. Pulmonary veins are also the primary sites of action of thromboxane. The increased synthesis of these vasoactive agents that may lead to pulmonary venous constriction, increased microvascular pressures [43]. Pulmonary veins show a greater sensitivity than arteries to a number of vasoconstrictor stimuli, including thromboxane [44]. The process of HD is associated with increased thromboxane production [28]. However, this does not happen in sham dialysis in healthy subjects in spite of complement activation and marked leukopenia, indicating the presence of other factors in ESRD patients [45] which may increase the sensitivity of pulmonary vessels to thromboxane.Our results show that BNP is higher in the PHT group in patients with renal failure. Leuchte et al. [15] stated that BNP is a simple, non-invasive and observer-independent parameter for assessing disease severity in patients with primary PHT [46]. In HD patients, BNP changes are small and without any correlation with either atrial or ventricular volume and continue to be higher than normal controls [47]. Creation of AVF is associated with elevated BNP which also correlates with LVDD [24]. Natriuretic peptides are elevated in patients with ESRD as a result of left ventricular hypertrophy and ventricular dysfunction and markedly diminished renal clearance. Although BNP is a vasodilator, BNP infusion did not significantly improve pulmonary hemodynamics in most patients with pulmonary arterial hypertension [48].The role of BNP in PHT in patients with CRF is not clear and may be because BNP is an important predictor of cardiovascular congestion [49] and LVDD which increases the intravascular pressure in pulmonary veins [42]. Multiple regression tests show that AVF flow is an important independent factor contributing to PHT followed by proBNP level.Our results show that PHT is common in CRF patients and is more so in patients receiving HD than those on conservative treatment. Patients with PHT have a higher AVF blood flow, proBNP, TXB2, age and more frequent LVDD.Since pulmonary arterial hypertension usually presents at a very late stage, early diagnosis is a key to proper management and a better outcome. Estimation and follow-up of PAP using Doppler echocardiography may be indicated in all patients with ESRD undergoing HD via an arteriovenous access. Surgical reduction of the arteriovenous access may also be indicated. Standardization of arteriovenous access size with a shunt flow estimation is necessary too. Improvement of PHT following kidney transplantation can be another indication for kidney transplantation in patients with severe PHT [4]. Endothelin receptor antagonists can reduce pulmonary blood pressure in CRF patients, the pharmacokinetics and safety of which have already been demonstrated [41].