Variability in Sputum Sol Phase Proteins in Chronic Obstructive Bronchitis

Abstract

The variability of sputum sol phase proteins was studied in 27 patients with stable chronic obstructive bronchitis. Within individual patients (n 1/2 11), variability over 5 consecutive days was similar for each of 4 proteins studied: average coefficient of variation (CV) for albumin was 39.8% for alpha 1-antitrypsin (alpha 1AT), 32.1%, for alpha 1-antichymotrypsin (alpha 1ACh), 26.6%, and for immunoglobulin A (IgA), 35.1%. The calculation of sputum-to-serum ratios did not affect this variability. Between patients (all patients studied on a single day) the variability of sputum sol phase proteins was greater (CV: albumin 1/2 84.5%, alpha 1AT 1/2 90.1%, alpha 1ACh 1/2 58.3%, and IgA 1/2 71.6%) and was also unaffected by the use of sputum-to-serum concentration ratios. "Standardization" for albumin reduced the average within-patient variability for sputum alpha 1AT (CV from 32.1 to 19.7%) but it had no effect on alpha 1ACh and IgA. However, although between-patient CV for sputum alpha 1AT was also reduced (from 90.1 to 53.4%, 2 p less than 0.01) it significantly increased for alpha 1ACh (from 58.3 to 86.0%, 2 p less than 0.05) and IgA (71.6 to 170.1%, 2 p less than 0.001), suggesting that such standardization may be inappropriate for these proteins.

Effect of cigarette smoking, pulmonary inflammation, and lung disease on concentrations of carcinoembryonic antigen in serum and secretions.

“…13 The conventional techniques of standardising results of locally concentrated proteins for the albumin in the secretion has been shown to increase variability. 25 The present study confirmed this for carcinoembryonic antigen, and standardising for albumin increased both within and between patient variability (results not shown). Nevertheless, some variability of concentration remains even in a single subject.…”

“…The day to day variability of an individual subject and the variability between subjects was similar to that for other sputum proteins. 24 The higher concentrations of carcinoembryonic antigen in secretions than in serum suggest that it is concentrated in the secretions, either by local production or by preferential transport. 13 The conventional techniques of standardising results of locally concentrated proteins for the albumin in the secretion has been shown to increase variability.…”

Measurement of soluble proteins in lung secretions.

“…The use of a marker substance which has diffused from serum, such as albumin, may be inappropriate as a standard for this locally produced protein. 22 Measurement of several factors unique to the assembly of the secretory IgA molecule might, however, provide appropriate standardisation for its study. 35 4 It has been suggested that potassium may be a useful standard.36 Potassium ion does not, however, show a constant relationship to other components of bronchoalveolar lavage fluid when increasing volumes of lavage fluid are used.37 For this reason different results would be obtained even in the same individual when different initial volumes of lavage fluid were instilled.…”

“…In practice, although some variability persists as a consequence of the inherent variability of the immunological techniques (between batch variability 4%), the use of albumin as an internal standard appears to be sound. 22 Dilution of the secretion by bronchoalveolar lavage fluid might be expected to increase variability of the individual protein measurements (lower panel, table 1). This can, however, be eliminated by standardisation for albumin concentration and comparison with serum concentration can eliminate the disturbance caused by dilution and give results similar to those for the undiluted secretions (top panel, table 1).…”

“…Indeed, recent data suggest that the variability of IgA in sputum samples both within and between patients is similar to that of albumin despite the likelihood of salivary contamination. 22 The dilutional effect of saliva is still, however, the major problem and this is conventionally overcome by the estimation of protein ratios within the secretion,6 although recent work suggests that the use of such ratios may not significantly alter the variability of the results obtained.22 This suggests that at least some of the variability between protein concentration ratios in sputum is not simply due to the dilutional effect of saliva but may represent inherent variability of protein movements in the secretion itself. This point is further discussed below.…”