Abstract 4515

Introduction:

Acute kidney injury (AKI) is one of the most common complications after allogeneic stem cell transplantation (SCT). Although various post-transplant risk factors for AKI have been reported, there have been few studies which comprehensively assessed pre-transplant comorbidities and investigated their impact on the occurrence of AKI.

Methods:

We performed a retrospective analysis of 207 consecutive adult patients undergoing myeloablative or non-myeloablative SCT between 2001 and 2009. Cumulative incidence of AKI during the first 100 days was analyzed. AKI was defined and classified according to Risk, Injury, Failure, Loss, and End-stage kidney disease (RIFLE) criteria. Renal dysfunction considered to be associated with multiple organ dysfunction syndromes before death was excluded. The time to the development of severe AKI (defined as RIFLE class I or class F) was calculated and the multivariate analysis of pre- and post-transplant variables was performed by a Cox proportional hazards model. Post-transplant factors were assessed as time-dependent variables. As a representative of pre-transplant comorbidities, we used the hematopoietic cell transplantation-specific comorbidity index (HCT-CI). To analyze the outcomes of patients who developed AKI, the multivariate analysis of OS and NRM was carried out by a landmark approach, in which the follow-up was started at the onset of AKI class R or severe AKI in patients who developed AKI class R or severe AKI, and at the median time of the incidence of AKI in patients who did not develop AKI.

Results:

Among 207 patients, myeloablative SCT was performed in 149 patients and non-myeloablative SCT in 58 patients. Overall, 158 patients (76.3%) developed AKI, and 92 patients (44.4%) developed severe AKI. The median interval to the occurrence of severe AKI was 30 days after SCT. The cumulative incidence of severe AKI within 100 days in patients with a HCT-CI score 0, 1–2, and ≥3 was 21.3 %, 48.8%, and 73.9%, respectively. In a multivariate analysis, the HCT-CI was independently and most strongly associated with severe AKI (HCT-CI 1–2: adjusted hazard ratio [HR] 2.42, P<0.01; HCT-CI ≥3: adjusted HR 4.69, P<0.01), although no patients had renal dysfunction defined by the HCT-CI scoring system (pre-transplant creatinine > 2.0 mg/dl) before SCT in our cohort. Among post-transplant factors, sepsis (HR 2.66, P<0.01) and use of vancomycin (HR 1.79, P=0.04) were significantly associated with the development of severe AKI. In total, 101 of 207 patients (48.8%) died, of which 52 patients (51.5%) died of non-relapse causes. In a landmark analysis, the 3-year OS was 61.4% in patients without AKI, 53.9% in patients with AKI class R, and 39.3% in patients with severe AKI (P<0.01). The 3-year NRM was 5.6% in patients without AKI, 16.6% in patients with AKI class R, and 40.8% in patients with severe AKI (P<0.01). Multivariate analysis showed that severe AKI was a significant risk factor for worse OS (HR: 2.10, P=0.01) and NRM (HR: 6.15, P<0.01), while the occurrence of AKI class R did not have a strong impact on OS (HR: 1.14, P=0.69) or NRM (HR: 2.31, P=0.20). According to the cause of death, 24 patients died of severe AKI within 100 days after SCT, consisting of 2 on-disease patients (8.3% of the 24 patients who relapsed within 100 days after SCT) and 22 patients in remission (12.0% of the 183 patients who did not relapse within 100 days after SCT). There was no significant difference in the proportions of severe AKI as a mortality cause between patients with or without relapse (P=1.0).

Conclusions:

We found that high HCT-CI scores, particularly ≥3, were the most important pre-transplant predictor of severe AKI. The development of severe AKI within the first 100 days after SCT was independently associated with worse prognosis, irrespective of the status of the primary disease. Therefore, the strategy to prevent the occurrence of life-threatening severe AKI in patients with a high HCT-CI score would be imperative to improve their survival after allogeneic SCT.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution