High Ischemia-Modified Albumin Concentration Reflects Oxidative Stress But Not Myocardial Involvement in Systemic Sclerosis
Posted Sep 11 2009 4:56pm
By Didier Borderie and Colleague
Systemic sclerosis (SSc) is a connective tissue disease characterized by widespread vascular lesions and fibrosis of the skin and internal organs. In SSc, vasospasm causes frequent episodes of reperfusion injury and free-radical-mediated endothelial disruption. Primary myocardial involvement is far more common than initially suspected on clinical grounds (1)(2)(3)(4)(5) and affects survival rates because it is associated with a poor prognosis (6)(7). Myocardial fibrosis is thought to occur secondarily to repeated focal ischemia in the coronary microcirculation as a result of abnormal vasoreactivity, with or without associated structural vascular disease (4)(5). The early and accurate identification of cardiac involvement is therefore of paramount clinical importance.
The concentration of ischemia-modified albumin (IMA), as measured by the albumin cobalt binding test (Ischemia Technologies, Inc.), is a new marker to rule out transient myocardial ischemia (8)(9). This test measures the binding of exogenous cobalt to the NH2 terminus of human albumin. In the presence of myocardial ischemia, structural changes occur in the NH2 terminus of albumin, rapidly reducing its capacity to bind transition metal ions after an ischemic event (10).
We assessed the accuracy of the albumin cobalt binding test for detecting ischemia in SSc patients and investigated the roles of myocardial ischemia and peripheral oxidative stress in this condition. We also considered carbonyl residues and advanced oxidation protein products (AOPP) as factors indicative of protein oxidation.
We included consecutive patients hospitalized for systematic follow-up who fulfilled the American Rheumatism Association preliminary criteria for SSc. The exclusion criteria were pregnancy; symptoms of heart failure, including class III or IV dyspnea (New York Heart Association); venous distension and recent major lower limb edema; pulmonary arterial hypertension (systolic arterial pressure >40 mmHg and/or mean artery pressure >25 mmHg, determined by echocardiography); severe pulmonary involvement (forced vital capacity or carbon monoxide diffusing capacity
<50% of the predicted normal value); renal involvement (creatinine concentration > 106 µmol/L); or severe disease complications such as cancer or gangrene. At the time of the study, none of the patients was taking medication for cardiac or vascular disease. If previously treated with vasodilators, patients were asked to stop taking these drugs 3 days before admission. This interruption period corresponds to five times the half-life of calcium channel blockers and angiotensin-converting enzyme. All patients gave informed consent for all procedures, and the study was approved by the local ethics committee (Paris, Cochin).
We assessed the following in all patients: blood cell count, Westergren erythrocyte sedimentation rate, serum creatinine concentration, and anti-centromere and anti-topoisomerase I antibody concentrations. The concentration of high-sensitivity C-reactive protein was measured by immunoturbidimetry on a Roche modular PP instrument using the CRP latex Tina-quant® assay (Roche Diagnostics). Pulmonary involvement was assessed by computed tomography scan, forced vital capacity, and the ratio of carbon monoxide diffusion capacity to hemoglobin concentration. Pulmonary arterial systolic pressure was determined by Doppler echocardiography at rest. The thickness of the skin was quantified on a scale of 0–3, by use of the modified Rodman skin scoring technique, for each of 17 body surface areas (11)(12).
All patients underwent thallium-201 myocardial single-photon-emission computerized tomography at rest, using a gamma camera (Starport 400AT; General Electric) interfaced with an ADAC computer (DPS 3300). Myocardial perfusion was assessed semiquantitatively by two experienced practitioners, using a blind protocol and a 17-segment model as follows: score 4 for normal, 3 for mild reduction (not definitely abnormal), 2 for moderate reduction (definitely abnormal), 1 for severe reduction, and 0 for absence of uptake (13). We determined both a "global perfusion score" (sum of the mean perfusion scores for each of the 17 segments) and the "number of perfusion defects" (with a perfusion defect defined as a score 2). The controls were healthy laboratory staff who agreed to provide blood. Blood samples (10 mL) were collected in tubes without anticoagulant. The samples were centrifuged at 3000g for 10 min within 1 h of collection, and the resulting sera were stored at –80 °C until use. The interval between sample collection and analysis was
Cardiac troponin I (cTnI) was measured by an immunochemiluminescent assay on an ACS180 analyzer (Bayer Diagnostics). In accordance with the manufacturer’s data, the lower limit detection of this assay was 0.03 µg/L; the total imprecision (CV) was 10% and 20%, respectively, at 0.4 and 0.1 µg/L. Serum IMA was measured on a Roche Modular PP instrument. Because no cutoff value was validated for this analyzer, 85 kilounits/L, the upper limit of the range of IMA concentrations for a reference population (95th percentile of a population of 283 apparently healthy individuals), was used as a cutoff point for ischemia, in accordance with data reported by the manufacturer and determined with the Roche Modular P analyzer. The inter- and intraassay CVs were
<3.7% and 4.3%, respectively, at IMA concentrations of 106 and 60 kilounits/L (n = 20). Protein carbonyl groups were determined by ELISA (14), and the AOPP concentrations were determined as described by Witko-Sarsat et al. (15). The lowest concentrations determined in our laboratory with CVs of 10% and 20% were, respectively, 0.20 and 0.12 µmol/g for protein carbonyl groups and 25 and 17 µmol/L for AOPP. Concentrations of IMA and markers of oxidative stress are expressed as medians with ranges. Data were analyzed by the Mann–Whitney test for group comparisons and the Spearman rank correlation test for assessment of the relationship between quantitative variables. P values
<0.05 were considered significant.
We investigated 32 consecutive SSc patients [mean (SD) age, 54.1 (11.6) years], including 26 women. The clinical and laboratory data for these patients are presented in Table 1. None of the following variables was correlated with IMA values: age, pulmonary fibrosis, carbon monoxide diffusion, autoantibody status, high-sensitivity C-reactive protein, and erythrocyte sedimentation rate.
In SSc patients, the median global myocardial perfusion score was 37 (range, 8–47; scores
<41 were considered abnormal), and the median of number of perfusion defects was 11 (range, 5–17). All patients had a cTnI concentration
<0.4 µg/L (10% CV); two patients had a cTnI value >0.1 µg/L (20% CV). The median (range) IMA concentration was 87 kilounits/L (55–115 kilounits/L; Fig. 1 ). Nineteen patients (59%) had IMA concentrations 85 kilounits/L, exceeding the 95th percentile for a population of 283 apparently healthy individuals in accordance with the manufacturer’s data determined with the Roche Modular P analyzer.
The IMA concentration was not correlated with the global myocardial perfusion score (r = 0.13; P = 0.48) or the number of perfusion defects (r = 0.23; P = 0.2). SSc patients diagnosed less than 5 years previously had higher median IMA concentrations [93 (74–115) kilounits/L] than did patients with longer disease durations [83 (55–106) kilounits/L; P
<0.05; Fig. 1 ]. IMA concentrations were inversely correlated with disease duration (r = –0.48; P
<0.01) and positively correlated with skin score (r = 0.54; P = 0.002).
Concentrations of serum markers of oxidative stress were significantly higher in SSc patients than in controls: carbonyl residues, 0.82 (0.37–1.09) µmol/g vs 0.34 (0.26–0.64) µmol/g (P
<0.001); AOPP, 95.1 (36.6–280) µmol/L vs 78.2 (43.2–129) µmol/L (P
<0.05). The IMA concentration was correlated with carbonyl residue concentration (r = 0.59; P = 0.002) but not with AOPP concentrations. However, neither carbonyl residues nor AOPP were correlated with disease duration.
SSc patients had high IMA concentrations, but the IMA concentration was not correlated with global myocardial perfusion score or the number of perfusion defects, despite functional impairment of the coronary microvasculature (16). The IMA concentration was associated with disease duration and skin score in SSc patients, reflecting the strong dependence of activity on the intensity of free radical reactions in the first few years of the disease, especially in patients with diffuse forms, who have high skin scores early in their disease (17)(18).
Free radicals generated by reperfusion injury and the inflammatory process may be of major importance in SSc patients (19). This study confirms that SSc is associated with excessive protein oxidative stress, as reflected by the high concentrations of carbonyl groups and AOPP. These results suggest that protein oxidation may occur early in the pathogenesis of SSc and may indicate underlying subclinical disease (oxidative stress or vascular dysfunction).
IMA has been studied primarily in selected populations thought to display myocardial involvement only in the absence of confounding clinical conditions. However, other organs seem to be responsible for the increase in IMA. Apple et al. (20) reported high concentrations of IMA 24–48 h after endurance exercise and suggested that this was attributable to delayed gastrointestinal or skeletal muscle ischemia. High IMA concentrations do not seem to depend purely on myocardial involvement. Thus, IMA may increase during ischemia-reperfusion, affecting any organ, and cannot be considered a specific cardiac marker in diseases associated with oxidative stress.
The IMA concentration was closely related to the concentration of carbonyl groups but not AOPP concentrations. This result is not surprising because these markers do not provide the same information concerning the extent of oxidative damage to proteins (15), the half-lives of these damaged proteins, and/or their clearance rate. Albumin is the most abundant serum protein, with a mean concentration of 0.63 mmol/L, and is a powerful extracellular antioxidant. The biochemical mechanism modifying the N-terminal region of albumin during ischemia is unclear, but reperfusion after an ischemic event may damage serum albumin as much as, if not more than, ischemia itself (21). These modifications to albumin may involve hypoxia, acidosis, or free radical damage, most of which occur within minutes. IMA seems to have a short half-life, returning to baseline values in 6–12 h, as shown recently in patients with stable angina pectoris after transient ischemia induced during elective percutaneous transluminal angioplasty (9). However, it is unclear whether the changes in the N-terminal region of albumin are reversible or lead to preferential degradation by proteolytic systems, as reported for other oxidized proteins (22). Nevertheless, the high values obtained in the albumin cobalt binding test for SSc patients reflect a succession of ischemic-reperfusion episodes, given the short half-life of IMA.