Background: The criteria by Camitta for diagnosis in severe aplastic anemia (SAA) has been used since 1976. However, there has been no attempt to verify the Camitta`s criteria, that the survival in patients with SAA may differ by absolute neutrophil count (ANC), platelet count, and corrected reticulocyte count (CRC), which are components of the Camitta`s criteria.. Patients and Method: 117 SAA patients diagnosed by the Camitta`s criteria were analyzed, retrospectively. Response by immunosuppressive therapy (IST) or stem cell transplantation (SCT) significantly affected overall survival (OS) by multivariate analysis (p=0.008). Therefore, we excluded treatment responders for analysis. Finally, 92 SAA patients were analyzed, including non- responders by IST or SCT and who were treated with supportive care by using univariate and multivariate analyses. Results: The median age of analyzed patients was 54.5 years. Male to female ratio was 1:1. The median follow-up duration was 23.48 months (range 0.27-1345). The median ANC, platelet count, and CRC were 394/uL, 12,000/uL, and 0.39%, respectively. In the univariate and multivariate analyses, ANC by more than or less than 500/uL was the only significant factor for OS (p=0.015, HR 2.694, 95% C.I. 1.209-6.002). There were no significant differences in OS between groups that were diagnosed SAA by ANC+platelet count, ANC+CRC. and platelet count+CRC. Conclusion: The criterion of ANC <500/uL is the only significant factor affecting OS. Thus, ANC should be an essential, not an optional criterion for diagnosing SAA. This study suggests the possibility that the Camitta`s criteria be modified. Studies in large cohorts are needed to transform the Camitta`s criteria.