Prognostic marker of nonfatal pulmonary thromboembolism: decreased glomerular filtration rate or increased age?

We read the article titled “Chronic kidney disease: Prognostic marker of nonfatal pulmonary thromboembolism” by Ouatu et al. (1) published in Anatol J Cardiol 2014 Dec 31 with great interest. In this article, the authors aimed to elucidate the relationship between venous thromboembolism-related mortality and renal dysfunction assessed by a regression-based MDRD formula. As a result of their investigation, the authors proposed that GFR is an independent predictor of 2-year mortality in pulmonary embolism besides troponin, dyslipidemia, acceleration time of pulmonary ejection, pericardial effusion, and BNP. Chronic kidney disease is a well-known prognostic factor, indicating increased morbidity and mortality in various cardiovascular diseases and acute pulmonary embolism. Impairment of renal functions may be related to preexisting chronic kidney disease or deteriorations secondary to hemodynamic failure (2). In clinical practice, renal functions are usually evaluated using creatinine-based formulae, which are based on age and gender. This situation may cause biases even after adjustment for age and gender in statistical analysis when evaluating the data for independence. It may not be cost-effective to evaluate renal functions with inulin or radioisotope-based quantitative determinants of GFR other than regression-based GFR formulae in a relatively large number of cases. According to the current guidelines (3), various prediction rules have been proposed for the prognostic assessment of patients with acute pulmonary embolism, and the pulmonary embolism severity index is one of the most widely used scores. This scoring system and its simplified form are composed of several variables including “age.” Male gender is also a poor prognosis predictor in the original form of the scoring system. In the current article by Ouatu et al. (1), gender difference was not significant between survivors and non-survivors, while age was significantly higher in non-survivors. We wonder if the authors adjusted their findings for age and possibly for gender or if they brought these variables into regression models. Otherwise, it is hard to propose GFR as an independent predictor of mortality owing to the highly possible collinearity between age and GFR. These concerns could be kept in mind while evaluating the results of this study.