IJCS | Volume 31, Nº2, March / April 2018

167 Table 3 – Variables associated with the occurrence of mediastinitis Variables OR (n = 65) 95% CI OR (n = 52) p-value* Intercept 0.008 (4.3e -5 -0582) 0.042 Age 0.997 (0.918-1.086) 0.952 Post-surgical hospitalization time, days 1.032 (1.008-1.067) 0.019 Fever No - - - Yes 10.316 (2.124-61.738) 0.005 Number of risk factors 1.573 (0.978-2.640) 0.065 The p-values refer to the significance tests of the regression coefficients. Hosmer-Lemeshow test: p = 0.620 (non-rejection of model adequacy). OR = odds ratio; 95% CI: 95% confidence interval. Similar results were observed in studies conducted in Brazil: Guaragna et al. 8 evaluated patients submitted to cardiac surgery from 1997 to 2000. Most of them were males (also in the mediastinitis group, 71.1%). The incidence of mediastinitis was 2.9% and themortality rate was 15.8%. Moreover, a study of patients submitted to CABG from 1996 to 2007 showed an incidence of mediastinitis of 3.3% and a 6% mortality rate. 6 Another study carried out in patients submitted to cardiac surgery, from 2007 to 2009, had 68.5% of the sample submitted to CABG and 31.5% to valve replacement. The mean age was 59.9 years and most of them were males. The incidence of post-sternotomy mediastinitis was 1.3%, and mortality was 42.8%. 16 A study carried out in the United Kingdom evaluated patients submitted to CABG from 1999 to 2009. The mean age was 67 years and 77.3% were males. The incidence of patientswho developedpost-sternotomy mediastinitis was 0.59%. Mortality was higher in patients who developed mediastinitis. 17 Regarding the profile of the patients described in these studies, 1,3,4,6,8,15-17 most of them were males, mean age was around 52 to 67 years, and CABG was the most often performed surgery. Similar results were observed in both studies performed at the institutionmentioned here, that is, the present study and the one by Coelho et al. 15 , regarding gender, mean age and type of surgery. Male patients are more likely to developmediastinitis and gender is even considered an independent risk factor for its development. 18,19 One of the probable mechanisms is related with the anatomical aspects of the male chest (hair follicles in the region of the sternotomy), favoring bacterial growth and infection. Guaragna et al. 8 verified this association: most patients who developed mediastinitis were males. In the present study, it was not possible to evaluate this association, as gender was used for pairing of the controls. Regarding the type of surgery, the correlation between CABG and the development of mediastinitis is well documented in the literature. More than 80% of the cases of infection were mentioned as a surgery-associated complication. 3 The studies by Souza et al. 4 , Sá et al. 1 and Magalhães et al. 3 verified this association, as observed in this study in which CABG, besides being the most often performed surgery, showed a higher proportion of patients with mediastinitis, although it did not show statistical difference. The incidence of mediastinitis in the evaluated studies, as described in the guideline of the American College of Cardiology/AmericanHeart Association (ACC/AHA) in 2012, 11 is between 0.4 and 4.0%. In this study, it was not possible to evaluate the incidence of mediastinitis, due to the study design. However, in the cohort studies, the lowest incidence ofmediastinitiswas observed in the study of Ariyaratnamet al. 17 , carried out in the UnitedKingdom. Studies performed in Brazil showed an incidence varying from 1.3% to 3.3%, 3,4,6,8,16 except for the study by Sá et al. 1 , which showed an incidence of 5.6%. The authors justified the high incidence for having considered all types of cardiac surgeries including CABG, which is, according to them, associated with a greater risk of mediastinitis. Other studies 1,3,4 also found an association betweenCABG and increased risk of mediastinitis, but they showed an Pinto et al. Factors Associated with Post-Sternotomy Mediastinitis Int J Cardiovasc Sci. 2018;31(2)163-172 Original Article