All enrolled patients received r-ATG (Lymphoglobulin, Genzyme; Cambridge, USA) intravenously in a dose of 3.75 mg/kg body weight on Days 1–5 and cyclosporine-A (Sandimmun Neoral Novartis; Pharma GmbH, Nurnberg; Germany) orally in a dose of 5 mg/kg body weight over Days 1–180. The dose of cyclosporine was modified according to its serum concentration to keep it in a range of 100–200 ng/ml. Granulocyte colony-stimulating factor (Fligrastim-Neupogen, Amgen; Thousand Oaks; USA) was given subcutaneously or intravenously only to patients with severe infections non-responding to antibiotics and antifungal medications. The presence of remission was evaluated at Days 112, 180, and 360 from the onset of treatment. Complete remission was defined as the absolute neutrophil count >1.5 × 10⁹/L, platelets >100 × 10⁹/L, and hemoglobin concentration >11.0 g/L. Partial remission was diagnosed when the neutrophil count was >0.5 × 10⁹/L, platelets >20 × 10⁹/L, and hemoglobin concentration >8.0/L.

The Kaplan-Meier estimator was used to draw survival curves and curves showing the time of disease progression and relapse. Overall survival, expressed in years, was defined as the interval between the diagnosis and death of a patient. The event free survival, expressed in years, was defined as the time from the diagnosis to death, disease relapse, or the last follow-up. The time to disease relapse was considered as the number of years between the diagnosis of AA and its relapse. If no demise or relapse occurred, the last follow-up with the evaluation of treatment response, was considered as right-censoring. The longest follow-up period equaled 10 years. All statistical tests carried out during the analysis considered a p-value of 0.05 and confidence intervals that did not include 1 as defining significance. Statistical analysis was carried out with a commercial package Stata ver. 11.0.