A quality assessment of genetic association studies supporting susceptibility and outcome in acute lung injury

INTRODUCTION: Clinical observations and animal models provide evidence that the development of acute lung injury (ALI), a phenomenon of acute diffuse lung inflammation in critically ill patients, is influenced by genetic factors. Association studies are the main tool for exploring common genetic variations underlying ALI susceptibility and/or outcome. We aimed to assess the quality of positive genetic association studies with ALI susceptibility and/or outcome in adults in order to highlight their consistency and major limitations. METHODS: We conducted a broad PubMed literature search from 1996 to June 2008 for original articles in English supporting a positive association (P ≤ 0.05) of genetic variants contributing to all-cause ALI susceptibility and/or outcome. Studies were evaluated based on current recommendations using a 10-point quality scoring system derived from 14 criteria, and the gene was considered as the unit of replication. Genes were also categorized according to biological processes using the Gene Ontology. RESULTS: Our search identified a total of 29 studies reporting positive findings for 16 genes involved mainly in the response to external stimulus and cell signal transduction. The genes encoding for interleukin-6, mannose-binding lectin, surfactant protein B, and angiotensin-converting enzyme were the most replicated across the studies. On average, the studies had an intermediate quality score (median of 4.62 and interquartile range of 3.33 to 6.15). CONCLUSIONS: Although the quality of association studies seems to have improved over the years, more and better designed studies, including the replication of previous findings, with larger sample sizes extended to population groups other than those of European descent, are needed for identifying firm genetic modifiers of ALI.