Study patients

Between January 2008 and January 2011, patients diagnosed with definite IE (according to the modified Duke criteria) at Cumhuriyet University Faculty of Medicine and Yuksek Ihtisas Education and Research Hospitals were enrolled in this study . Patients with chronic liver disease and those with a history of haematological disease other than anaemia (such as leukaemia, myeloproliferative diseases or bone marrow infiltration) were excluded. The prospective data of 100 consecutive IE patients were analysed retrospectively. In addition to RDW levels on admission, clinical, echocardiographic and laboratory findings were recorded for each subject. Predisposing heart diseases, including prosthetic valve, pre-existing valvular disease (rheumatic heart disease and degenerative valves), congenital heart disease, implantable cardiac device, nosocomial infection, previous history of IE, malignancy and immunosuppression were assessed.

Complications during hospitalization, such as heart failure, renal failure, abscess formation, embolic events (excluding cerebral), cerebrovascular events and surgical treatment for IE, were recorded. Duration of hospital stay and in-hospital mortality were noted. Cerebrovascular events were defined in case of the following presentations: intracranial haemorrhage; ischaemic stroke; or transient ischaemic attacks. Renal failure was defined by serum creatinine concentration > 2 mg/dL during hospital stay. Anaemia was defined according to the World Health Organization criteria (haemoglobin < 13 g/dL in men and < 12 g/dL in women). The study endpoint was the incidence of all-cause death within 1 year after index hospitalization. Clinical event data were collected during the follow-up period for all patients by reviewing medical files and by telephone contact. The study was performed in accordance with the Declaration of Helsinki for human research and was approved by the local ethics committee.

Laboratory

Blood samples were obtained after admission, following overnight fasting. Baseline RDW, haemoglobin, haematocrit, platelet count and white blood cell count values were measured using an automated haematology analyser. C-reactive protein, glucose and creatinine concentrations were measured accordingly. At least three sets of blood samples for cultures were obtained from each patient immediately after hospital admission. Any other available fluid, tissue (valves, vegetations or intracardiac abscesses removed at surgery) or foreign body samples (pacemaker leads, catheters) were used to isolate microorganisms.

Echocardiography

All patients underwent two-dimensional transthoracic echocardiography within 24 hours of admission. Echocardiographic examinations were performed with the Vivid 7 system (GE Healthcare, Wauwatosa, WI, USA) in two participating centres. Transoesophageal echocardiography was performed when image quality with transthoracic echocardiography was insufficient for an accurate diagnosis or in cases of high clinical suspicion of IE, prosthetic valve involvement and suspicion of complications. Vegetation, abscess formation and valvular destruction, such as perforation of leaflet and chordal rupture, were evaluated. Vegetation size was measured by using different echocardiographic windows; the maximal length was obtained. Existence of rocking motion of the prosthetic valve with an excursion of > 15° in at least one direction gave the diagnosis of dehiscence. Left ventricular ejection fraction was calculated by the modified Simpson’s method. Severe valvular regurgitation was identified according to guideline recommendations . Pulmonary artery systolic pressure was estimated by continuous wave Doppler imaging of the tricuspid regurgitation using the Bernoulli equation .

Statistics

Continuous variables were expressed as mean ± standard deviation or median with interquartile range; categorical variables were expressed as number and percentage. A chi-square test or Fisher’s exact test was performed to compare categorical variables. The normality of distributions of variables was assessed using the Kolmogorov-Smirnov test. Student’s t -test was used for normally distributed continuous variables; the Mann-Whitney U test was used when the distribution was skewed. The discrimination of RDW for 1-year survival was evaluated using the area under the receiver operating characteristic (ROC) curve. The optimal cut-off point for ROC curves was determined for maximizing the sensitivity and specificity of the RDW values. Patients with IE were categorized into two groups on the basis of the cut-off value. Kaplan-Meier cumulative survival curves were used to display survival in two patient subgroups and log-rank values were calculated to assess the statistical significance. Univariate Cox proportional-hazards analyses were used to evaluate the relationship between variables and overall mortality. Variables that had a P value < 0.1 in the univariate analysis were used in a multivariable Cox proportional-hazards model to determine the independent prognostic factors of mortality. The results of the regression analysis are presented as hazard ratios and 95% confidence intervals. All statistical analyses were performed using SPSS software version 17.0 (SPSS Inc., Chicago, IL, USA). A P value of 0.05 was considered statistically significant.