C-reactive protein and NT-proBNP as surrogate markers for pulmonary hypertension in Gaucher disease

https://doi.org/10.1016/j.bcmd.2005.01.002Get rights and content

Abstract

Background

N-terminal pro-Brain Natriuretic Peptide (NT-proBNP) values correlate with mild-moderate pulmonary hypertension assessed by tricuspid insufficiency (TI) gradient ≥30 mm Hg in Gaucher disease. The purpose of this study is to ascertain improved risk stratification based on correlation with NT-proBNP and C-reactive protein (CRP), a standard marker of inflammation.

Methods

Patients with type I Gaucher disease were selected to reflect differing degrees of echocardiographically determined TI gradient values. NT-proBNP was performed by immunoassay and CRP by standard methods.

Findings

There were 45 patients (18 males; 40%); mean age = 42.5 (range: 4–80) years. Median NT-proBNP value = 153 (range: 46–6703) pg/ml; median CRP value = 0.145 (range: 0.02–2.69) mg/dl. There was a statistically significant correlation between these values (r = 0.445, P < 0.01). Elevations of CRP and NT-proBNP were risk factors for pulmonary hypertension with odds ratios of 8.47 and 4.9, respectively. The area under the Receiving Operator Characteristic (ROC) curve for diagnosis of pulmonary hypertension was 0.93 ± 0.04 for CRP, and 0.7 ± 0.1 for NT-proBNP. All patients with pulmonary hypertension had elevation of either CRP or NT-proBNP (100% sensitivity).

Conclusions

Elevated CRP was a better predictor of pulmonary hypertension in Gaucher disease than elevated NT-proBNP values. Elevated CRP (>0.5 mg/dl) or elevated NT-proBNP (>150 pg/ml) reduces the need to perform echocardiography by more than half, even in this group with over-representation of pulmonary hypertension. The role of inflammatory features in pulmonary hypertension in Gaucher disease is discussed. Further studies are required to ascertain if this approach is useful for prognosis of pulmonary hypertension.

Introduction

Primary pulmonary hypertension is a rare life-threatening disorder characterized by elevated pulmonary artery pressure whose etiology is incompletely understood; current therapies mitigate signs and symptoms, but progression to right heart failure is only slowed rather than reversed. Echocardiography is currently the preferred test for indirect evaluation of pulmonary hypertension, whether secondary or primary [1]. Recently, the use of the N-terminal fragment of the pro-hormone of brain natriuretic peptide (NT-proBNP) has been shown to be effective for diagnosis and prognosis in several cardiovascular disorders [2] and in symptomatic primary pulmonary hypertension [3].

Gaucher disease is caused by an enzymatic defect with accumulation of glucocerebroside in the cells of the monocyte–macrophage system. The presenting signs of systemic involvement generally include hepatosplenomegaly and hypersplenism. Infiltrative lung involvement and/or pulmonary hypertension secondary to severe systemic disease are increasingly recognized in both the neuronopathic and non-neuronopathic types of Gaucher disease [4].

Enzyme replacement therapy ameliorates most symptoms and signs including secondary lung involvement and pulmonary hypertension, but has also been implicated in inducing a primary-like form of pulmonary hypertension in a few patients [5]. The practice to routinely monitor tricuspid insufficiency (TI) gradient values by echocardiography has resulted in identifying patients with “incipient” pulmonary hypertension when TI gradient values are elevated beyond 30 mm Hg [6]. Often, withdrawal from enzyme therapy will halt the progression to overt symptoms and signs of severe pulmonary hypertension.

Recently, in a pilot study by our group, NT-proBNP values were seen to correlate with TI gradient values in a cohort of patients with Gaucher disease who were selected to represent a spectrum of TI gradient values [7]. Nonetheless, some patients with elevated TI gradients (≥30 mm Hg) and elevated NT-proBNP values (>200 pg/ml) did not have overt signs or symptoms of pulmonary hypertension. In another context, when attempting to correlate laboratory markers with specific organ severity/involvement in Gaucher disease such as bone and lung disease, we have shown a correlation between d-dimer values and TI gradient values [8] in an unselected cohort of adult patients. Our posited hypothesis was that elevated d-dimer values are suggestive of microthrombi. Yet, one may also entertain another tenable hypothesis: that d-dimer elevations are indicative of an inflammatory pathology.

Therefore, in attempting to more adequately assess patients with Gaucher disease who may be at risk for developing clinically significant pulmonary hypertension, a multi-marker approach was considered. In the current pilot study, high-sensitivity C-reactive protein (CRP) values were used as a more sensitive measure of inflammation along with d-dimer values, and NT-proBNP values as prognostic in pulmonary hypertension, and these were serially compared with TI gradient to ascertain whether laboratory measures may be sufficiently predictive of patients who may develop symptomatic pulmonary hypertension in Gaucher disease.

Section snippets

Methods

There were 48 patients from a cohort of 199 patients for whom there was long-term follow-up by echocardiography and from whom sera had been taken. The sample cohort was selected in order that there be a spectrum of echocardiographic findings. All 12 patients with TI gradients ≥30 mm Hg at the time of blood sampling with simultaneous evaluation of d-dimers and CRP during the 6-month period between October 2003 and March 2004 were included. Additional 36 patients who had been seen during this

Statistical analysis

Data were expressed as medians and ranges or as means. As most of the quantitative variables were not normally distributed, comparison between groups was performed using the non-parametric Mann–Whitney rank sum test. The Spearman non-parametric correlation coefficient was used to estimate the association between pairs of variables. The Pearson correlation coefficient was calculated only between the log transformation of the NT-proNP and CRP levels. NT-proBNP and CRP levels were dichotomized

Results

Of the original 48 patients, samples from 3 patients with TI gradients < 30 mm Hg were missing more than one lab value and therefore eliminated. The demographic characteristics of the remaining patients and the results of the assays are presented in Table 1. There was a statistically significant correlation (Spearman non-parametric correlation coefficient) between NT-proBNP and CRP (r = 0.445, P < 0.01).

Of the 16 patients with elevated NT-proBNP (>200 pg/ml), eight patients (50.0%) had normal

Discussion

A multi-marker approach would be attractive to enhance identification of patients at risk for pulmonary hypertension, particularly for the purpose of early intervention. Whereas routine echocardiography has resulted in more accurate documentation of trends towards elevated TI gradients and hence has allowed for withdrawal of enzyme therapy in those patients in whom this may be therapeutic [5], TI gradients alone are not sufficiently accurate for risk stratification. Since enzyme replacement

Acknowledgments

The authors would like to express their gratitude to Ms. Tali Bdolah-Abram for statistical analyses.

This study was partially supported by an unrestricted educational grant from Oxford GlycoSciences Ltd., UK.

References (10)

There are more references available in the full text version of this article.

Cited by (0)

View full text