INTRODUCTION

The lack of clarity of the signs and symptoms in the diagnosis of heart failure (HF) has led to the search for new biomarkers. In recent years, B-type natriuretic peptide (BNP) has become accepted as a marker of cardiac function, and its diagnostic and prognostic value has been demonstrated in patients with chronic HF.1-4 In these individuals, dyspnea produces a functional limitation and is usually the major symptom. The mechanisms of dyspnea on exertion in chronic HF are not fully understood, and they involve cardiac, pulmonary and peripheral vascular factors.5 The deterioration of pulmonary function is progressive throughout the course of the disease, and is associated with a poorer prognosis.6-9 However, to date, there is no known peripheral marker that, like BNP, aids us in monitoring lung function.

Surfactant protein B (SP-B) is found in the epithelium of the pulmonary alveolar-capillary membrane and is necessary for the maintenance of its function.10 The alveolar levels of SP-B remain constant, and it flows from the alveolus to the capillary blood as a consequence of a positive gradient, with rapid clearance from the capillary. If the alveolar-capillary barrier is damaged, SP-B leaks from the air space into the circulation as a result of the increased permeability.11-13 This increased flow enables its detection in peripheral blood, and its measurement could be used as a sensitive, noninvasive method for monitoring pulmonary function in HF patients. In this respect, SP-B has been detected in the peripheral circulation of patients with acute pulmonary edema, as a consequence of the sudden increase in the pulmonary capillary pressure and the damage to the alveolar-capillary unit.14-16 In patients with chronic HF, the sustained increase in capillary pressure is associated with structural changes17-19 and, to date, only one study has suggested that the SP-B levels could correlate with the functional class and the prognosis in these patients.20

The hypothesis of this study was that SP-B could be a marker of pulmonary function that would aid in the monitoring of functional deterioration in patients with chronic HF. The objectives of the present study are: a) to determine whether SP-B is detectable in patients with chronic HF, comparing them to a control group of healthy volunteers; and b) to assess whether its presence correlates with functional parameters in a cardiopulmonary exercise test.

METHODS

Study Population

The study included 43 consecutive outpatients from a specialized HF unit, with an established diagnosis of chronic HF, who met the following selection criteria: a) exertional dyspnea as the major limiting symptom (New York Heart Association [NYHA] class II-III); b) an optimized medical treatment (including a beta-blocker and an angiotensin-converting enzyme inhibitor or an angiotensin II receptor antagonist); c) a stable clinical condition, with no hospital admissions in the preceding 6 months; and d) not be diagnosed as having primary pulmonary disease and/or abnormal spirometry, defined as a forced expiratory volume in one second of less than 80% of the predicted value or a ratio of forced expiratory volume in one second to forced vital capacity of less than 70%. These criteria were established to achieve a homogeneous sample, without biases derived from conditions that could potentially affect the SP-B values.

All the procedures included in the study were carried out on the same morning: a) the patients had fasted since midnight the night before and, after a period of rest of 30 minutes, blood samples were collected from an antecubital vein; b) an echocardiographic study was performed (Sonos 5500, Philips, Andover, MA, USA); c) a symptom-limited cardiopulmonary exercise test was carried out; and d) within the first minute after the exercise test, a second blood sample was collected. We included an age and sex-matched control group (n=26) of asymptomatic healthy volunteers, who had no known history of disease or cardiovascular risk factors, and in whom the results of echocardiography and spirometry were normal. Under the same conditions with respect to rest and fasting as those of the group with chronic HF, a blood sample was obtained from each of the controls. This study was approved by the local ethics committee, and all the participants gave their written informed consent.

SP-B Determinations

The presence and amount of SP-B in serum was analyzed by means of Western blot. For each sample, 100 µg of total protein were incubated for 5 minutes at 95ºC in Laemmli sample buffer and were separated by means of electrophoresis under reductive conditions in 12% polyacrylamide gel. The gel was then transferred to a PVDF membrane (Immobilon-PSQ Membranes, Millipore, Bedford, MA, USA). A rabbit anti-PSB polyclonal antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) diluted 1:750 and a horseradish peroxidase-conjugated goat anti-rabbit antibody (Santa Cruz Biotechnology, Inc.) diluted 1:5000 were used as primary and secondary antibodies, respectively. For development, we employed the ECL Plus chemiluminescence kit (Amersham Pharmacia Biotech, Little Chalfont, UK) according to the manufacturer's instructions. Quantitative analysis was performed with the Gel-Pro Analyzer 3.1 software package (Sigma, St. Louis, MO, USA). The SP-B values are expressed as optical density units. A single specific band with an apparent molecular weight of 17 kDa was obtained, which corresponds to the active dimeric form of SP-B released from the bronchoalveolar barrier.21 The serum total protein concentration was determined by means of a bicinchoninic acid assay. The homogeneity of the total protein load in the different lanes of the gel was assessed by cutting the upper fourth of the membrane prior to blocking and staining it with 0.1% amido black dissolved in 45% methanol and 10% acetic acid. To verify the specificity of the immunoreactive bands, a pulmonary surfactant extract obtained from bronchoalveolar lavage performed in patients with primary lung tumors was utilized. The determinations were performed in triplicate in each sample. The plasma BNP concentration was measured in duplicate using a specific sandwich solid phase radioimmunoassay (ShionoRia BNP kit, CIS Bio International, Gif sur Yvette, France), as reported elsewhere.22

Cardiopulmonary Exercise Test

All patients underwent a dyspnea-limited cardiopulmonary exercise test on a treadmill according to a modified Bruce protocol. Gas exchange and ventilation were assessed by means of a pneumotachograph equipped with a gas analyzer (CPX System, Medical Graphics Corp., St. Paul, MN, USA), which measured minute ventilation (VE), oxygen consumption (VO2), and carbon dioxide consumption (VCO2) every 10 seconds by means of a mass spectrometer (Amis 2000, Innovision, MedGraphics CardiO2 System, Odense, Denmark). The patients were asked to exercise to exhaustion and all of them reached the anaerobic threshold and a respiratory quotient >1.05. Continuous monitoring with 12-lead electrocardiography was performed. Blood pressure was measured once every minute by means of a cuff sphygmomanometer. All the patients interrupted the test due to dyspnea; none of them experienced chest pain or ischemic ST segment changes. The maximum VO2 (VO2max) was established as the value of greatest importance in the final phase of exercise, and was expressed in mL/kg/min and as a percentage (%) of the predicted normal level (%VO2max or functional capacity). The predicted VO2max was calculated on the basis of sex, age, and weight.23 The slope of the ratio of minute ventilation to carbon dioxide production (VE/VCO2) was established as the coefficient of the ventilatory response to exercise or ventilatory inefficiency.24

Statistical Analysis

The Kolmogorov-Smirnov test was used to assess the normal distribution of quantitative variables. The SP-B and BNP values showed a deviation from normality and, thus, are expressed as medians (interquartile range [IQR]). The remaining quantitative variables were expressed as means (standard deviation [SD]) and the qualitative variables were expressed as number (%). For the comparison of the categorical variables, the c2 test and Fisher's exact test were used. The levels of SP-B in the group with chronic HF and in the control group were compared by means of the Mann-Whitney U test for independent samples. The SP-B levels at rest and during maximum effort were compared using the Wilcoxon test for paired samples. The association between SP-B levels and the diagnosis of chronic HF was studied by analyzing the receiver operating characteristic curve. For the study of the correlations, we used the logarithmic transformation of the levels of SP-B values (SP-Blog) and BNP (BNPlog) and simple linear regression analysis. The association between SP-B levels and the exercise parameters was evaluated by means of multiple linear regression analysis, adjusted for age, sex, left ventricular ejection fraction (LVEF), BNP and those variables with a P<.1 in the univariate analysis, and the β values and their 95% confidence intervals (95% CI) are provided. A P value less than .05 was considered significant. The statistical analysis was performed using a software package for social sciences (SPSS v. 15.0 for Windows, SPSS, Inc., Chicago, IL, USA).

RESULTS

Surfactant Protein B and Heart Failure

We studied a total of 43 patients with chronic HF whose clinical characteristics are shown in Table 1. The control group was made up of 26 healthy volunteers (age, 51.4 [9.6] years; 77% men). The presence of SP-B (Figure 1) was detected in the peripheral circulation of 35 HF patients (82%) versus 6 healthy volunteers (23%) (P<.001). In those with detectable SP-B, the concentration was higher in patients with HF: median [interquartile range, IQR] (interval), 174 [70-283] (13-1486) than in the healthy volunteers, 77 [41-152] (23-187) (P<.001) (Figure 2). The analysis of the receiver operating characteristic curve showed an area under the curve of 0.83 (95% CI, 0.72-0.91), and the detection of SP-B exhibited a sensitivity of 81%, a specificity of 77%, a positive predictive value of 86%, and a negative predictive value of 72% for the presence of HF. In the HF patients, the presence of circulating SP-B (vs its absence) was associated with a lower LVEF (31.4% [9.6%] vs 41.8% [15%]; P=.010). The SP-B levels did not correlate with any of the other clinical characteristics shown in Table 1 (P>.05 for all the analyses).

Figure. 1. Detection of surfactant protein B (SP-B) using Western blot analysis: control of bronchoalveolar lavage of patients with primary lung tumors (+); patients with heart failure (HF); healthy volunteers (S1, S2); and blank lane (X).

Figure 2 Quantification of circulating surfactant protein B (SP-B) levels in patients with heart failure and healthy controls.

Surfactant Protein B and Functional Capacity

The SP-Blog value at the time of peak exercise exhibited a nearly perfect correlation with the resting SP-Blog value (r=0.980; P<.001), but there were no significant differences between the levels at these 2 times (P=.164). The exercise/resting SP-B ratio was 0.99 (0.23) [0.85-1.06] (0.44-1.65). In multiple linear regression analysis (Table 2), a higher SP-Blog level was the principal determinant of a lower functional capacity (lower %VO2max), whereas age and SP-Blog at rest were positively associated with a greater ventilatory inefficiency (steeper VE/VCO2 slope). There was no association between SP-Blog and the duration of exercise or the maximum arterial blood pressure × heart rate product, and the response of SP-B to exercise (exercise/rest ratio) exhibited no association with the VO2max or the VE/VCO2 slope (P>.3 for all the analyses).

DISCUSSION

The findings of this study have implications for the understanding of the pathophysiology of the functional impairment in patients with chronic HF. Our work shows that SP-B is measurable and is present in a higher percentage and with higher levels in the peripheral circulation of HF patients than in that of a healthy population; likewise, the resting levels are correlated with ventilatory inefficiency during exercise (VE/VCO2 slope). Moreover, the description for the first time of its measurement in human serum by Western blot increases the relevance of this study.

Previous reports have shown that SP-B is present in the peripheral circulation of healthy controls and, at a higher concentration, in different lung diseases, as an index of pulmonary status.10,25,26 In situations of acute lung damage, such as acute pulmonary edema or adult respiratory distress syndrome, SP-B has been found to be increased in peripheral blood as a consequence of the damage to the alveolar-capillary unit.14,27 Our study demonstrates that SP-B is present in the peripheral circulation of patients with stable chronic HF, at significantly higher levels than those detected in healthy individuals. Until now, only one study, by de Pasquale et al, has demonstrated the presence of SP-B in 53 HF patients at significantly higher levels than in 17 healthy controls.20 The use in our study of Western blot for the detection of SP-B in human serum, rather than the enzyme-linked immunosorbent assay employed by de Pasquale et al, makes our work an important confirmatory study, as Western blotting is a more specific technique for the detection and comparison of protein levels. On the other hand, both studies show a wide range of levels, a circumstance that suggests a high interindividual variability, although the good correlation between the levels at rest and during exercise suggest a lower intraindividual variability.

In patients with HF, exercise is associated with an increase in pulmonary capillary pressures as well as pulmonary edema, which could play a role in an increase in the circulating SP-B.28 However, in our study, we did not observe a significant increase in SP-B during exercise. In 20 patients without HF who underwent a stress test for the diagnosis of heart disease, De Pasquale et al did not find an increase in SP-B in the overall study population, but did observe this effect in 10 patients in whom ischemia and changes in contractility were induced during exercise.29 In our study, the detection of SP-B was associated with a lower resting LVEF. However, given the absence of echocardiographic or hemodynamic monitoring during exertion, we are unable to determine whether the induction of ventricular systolic dysfunction during exercise would help to identify a subgroup of patients with increased SP-B. On the other hand, the SP-B levels were measured immediately after exercise, and we have no data as to whether serial measurements would have produced significantly different results.

As a measure of the status of the alveolar-capillary unit, SP-B might explain the functional impairment due to dyspnea during exercise. In our study, SP-B was not predictive of the VO2max, in contrast to BNP, which has been found to be a significant predictor of functional capacity.20,30 The VO2max reflects the increase in cardiac output upon exercise, a fact that would explain its closer association with BNP, since the latter is a biomarker of cardiac function. Previously, in 53 patients with HF, De Pasquale et al found a correlation among SP-B values, NYHA functional class and the 6-minute walk test.20 In contrast to that study, we found no correlation with more objective parameters such as the VO2max and the %VO2max. The inclusion in our study of stable patients with less functional impairment, as demonstrated by the low range of the baseline BNP and the relatively high VO2max values, may have had an influence on the results obtained. However, we observed a modest, but significant, association with ventilatory inefficiency, reflected in a steeper VE/ VCO2 slope. In patients with HF, there is a progressive loss of capacity for the diffusion of gases across the alveolar-capillary membrane, which, together with the increased congestion during exercise, implies a greater need for ventilation (VE) and a steeper VE/VCO2 slope or ventilatory inefficiency.7,31 Guazzi et al32 demonstrated that the status of the alveolar-capillary membrane is the pulmonary parameter that correlates best with the VE/VCO2 slope during exercise. This finding is in agreement with the results of our study, in which SP-B was associated with a steeper VE/VCO2 slope.

Thus, our study suggests that SP-B, as a marker of alveolar-capillary damage, is measurable in the peripheral circulation of HF patients and is associated with the impairment of pulmonary function in terms of ventilatory inefficiency during exercise. Consequently, its measurement could be used as a pulmonary marker in HF and could be of prognostic value in these patients. Additional studies involving larger populations are necessary in order to better define its role in patients with HF.

ABBREVIATIONS

BNP: B-type natriuretic peptide
HF: heart failure
IQR: interquartile range
SP-B: surfactant protein B

See editorial on pages 117-9

This study was partially funded by a 2006 grant from the Spanish Society of Cardiology Working Group on Heart Failure, Heart Transplantation, and Associated Therapies, Spain.

Correspondence: Dr. D.A. Pascual Figal.
Servicio de Cardiología. Hospital Universitario Virgen de la Arrixaca. Ctra. Madrid-Cartagena, s/n. 30120 Murcia. España.
E-mail: dapascual@servicam.com

Received April 28, 2008.
Accepted for publication October 17, 2008.