Mahmoud Q. Moammar

Cardiac troponin I levels and alveolar-arterial oxygen gradient in patients with community-acquired pneumonia.

BACKGROUND Patients with community-acquired pneumonia (CAP) appear to have cardiac stress as demonstrated by elevated B-type natriuretic peptide (BNP). We hypothesised that myocardial stress and decrease in oxygenation might also lead to elevations of cardiac troponin I (cTnI) levels in serum. OBJECTIVE The aim of this study was to see if cTnI was associated with the alveolar-arterial oxygen gradient (DeltaA-a), a marker of severity in CAP. METHODS Retrospective cohort study of 901 CAP patients with no evidence of acute coronary syndrome presenting to a large, tertiary-care, urban teaching hospital over a 3-year period. RESULTS A strong linear trend between log(10)cTnI and DeltaA-a was observed (r(2)=0.76) with a statistically significant Spearman correlation coefficient (r(s)=0.75; p<0.0001) between cTnI and DeltaA-a. A cTnI value of 0.5 ng/ml discriminated mild CAP from moderate-severe CAP with an OR=208 (95% CI: 50.5-408; p<0.0001). CONCLUSIONS These data suggest that decreased blood O(2) levels as suggested by elevated DeltaA-a may lead to acute myocardial damage and that cTnI may be useful as a biomarker to stratify risk in subjects with CAP.

ALVEOLAR–ARTERIAL OXYGEN GRADIENT, PNEUMONIA SEVERITY INDEX AND OUTCOMES IN PATIENTS HOSPITALIZED WITH COMMUNITY ACQUIRED PNEUMONIA

1 The alveolar–arterial oxygen gradient (ΔA–a) provides a useful assessment of ventilation/perfusion (V/Q) abnormalities. The objectives of the present study were to: (i) examine the correlation between the ΔA–a and the pneumonia severity index (PSI); and (ii) determine whether these measures were comparable in predicting clinical outcomes. 2 The present study was conducted at a 750‐bed teaching hospital. It examined a retrospective cohort of 255 patients with community acquired pneumonia (CAP) over a 2 year period. Association between the CAP and ΔA–a was investigated by regression models and correlation, as well as two logistic models for subjects bifurcated by low‐risk/moderate‐to‐high risk. The decision levels (DL) for both PSI and ΔA–a were then compared as predictors of both length of stay (LOS) and survival. 3 The correlation between PSI and ΔA–a was strong (ρ = 0.76; P < 0.0001) and was best modelled by a curvilinear relationship. Both logistic models indicated a strong association (P < 0.001) between ΔA–a and PSI and yielded an optimal DL for the ΔA–a of < 89 mmHg. Inter‐test agreement of ΔA–a with PSI was 76.9% (κ = 0.60; 95% confidence interval 0.47–0.72; P < 0.0001). At < 89 mmHg, the odds ratios for LOS were similar to those at PSI = 90 in predicting LOS in the range 3–7 days, inclusive. There was no significant difference in the ability of ΔA–a and PSI to predict survival for either the low‐ or high‐risk group (P = 0.363 and P = 0.951, respectively). 4 The ΔA–a correlates well with PSI and performs comparably in predicting two major outcomes in subjects hospitalized with CAP.