Background: Growing evidence shows that acute kidney injury (AKI) leads to chronic kidney disease in pediatric and adult populations. Patients with sickle cell disease (SCD) that develop acute chest syndrome (ACS) are at risk for developing AKI due to the hypoxic, acidic environment in the renal medulla, the use of nephrotoxic agents (vancomycin, NSAIDs), hyposthenuria and potential dehydration experienced by patients prior to admission for ACS. As AKI and acute lung injury are linked by systemic organ cross-talk, AKI with or without oliguria, may exacerbate the pathologic state of ACS. For these reasons, improving our understanding of the epidemiology and pathophysiology of AKI during sickle cell crisis is paramount.

Methods: We conducted a six year retrospective observational evaluation of 236 episodes of ICD-9 defined ACS among 166 admitted pediatric patients. We excluded patients admitted for one day or had normal CXR throughout hospitalization despite being coded as ACS. AKI was defined by KDIGO definition of either a ≥ 0.3 mg/dL or 50% increase in serum creatinine (SCr) from baseline. Forty eight episodes were excluded as a baseline SCr was not identified. Variables analyzed included age, gender, SCD genotype, length of stay, CBC (baseline and admission), current SCD therapy, transfusion intervention and length of stay (LOS). Per institutional standard of care, patients with ACS are treated with Vancomycin, Azithromycin, and Ceftriaxone and receive NSAIDS (ibuprofen or toradol) for pain/fevers but the type/timing of transfusion is determined by SCD provider preferences. Logistic regression analysis was performed to assess associations between clinical variables and AKI for all patients and severe phenotypes. Multiple linear regression analysis was performed to determine if AKI was independently associated with hospital length of stay.

Results: The mean age of patients admitted with ACS was 9 years (range 0.8yrs-19) and 61% were male. Genotypes included SS (79%) SB0 thalassemia (4%), SB+ thalassemia (5%), SC (12%) and SD (1%). AKI was identified in 13 of 188 (7%) episodes of ACS. The majority of cases developed AKI within two days of admission and resolved 2.5 days later. Compared to children without AKI, those with AKI had a lower admission hemoglobin (6.3 vs 7.5 g/dL, p=0.02), lower decrease in hemoglobin from baseline (3.2 vs. 1.3 g/dL, p<0.001), and higher admission WBC (32.3 vs. 20.3 x103/uL, p<0.001). Finally, need for exchange transfusion was associated with AKI (χ2 = 9.8, p=0.007). Seven of 39 patients (18%) that underwent exchange transfusion, 5 of 137 (4%) simple transfusion patients, and 1 of 16 no transfusion patients (the one patient had severe RBC antibodies so was not transfused) developed AKI. The patients with and without AKI were similar for age, gender, SCD genotype, platelet count, and current SCD therapy. When evaluating whether AKI was associated with longer LOS, we found that those with AKI had longer LOS than those without (12.4 vs 5.0 days, p<0.001). Using nominal logistic regression, drop in hemoglobin from baseline (p<0.001), admission WBC (p=0.01), and length of stay (p<0.001) remained significantly associated with AKI (Χ2 =42, p<0.001). When excluding patients with milder SCD phenotypes (SC and SB+ thalassemia), similar results were observed. Among 164 ACS events with severe SCD genotypes, 11 (14%) patients developed AKI. Using nominal logistic regression to assess severe SCD genotypes, drop in hemoglobin, admission WBC, and LOS were significantly associated with the development of AKI (Χ2=36, p<0.001).

Conclusion: This study demonstrates that AKI is prevalent during admissions for ACS and is highly associated with a drop in Hb and increase in WBC on admission and the need for exchange transfusion. While AKI presented and resolved early during hospitalization, it is strongly associated with a prolonged hospital course. The impact of AKI on the pathophysiology of ACS in the acute setting as well as its role in progression to SCD nephropathy should be studied prospectively.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution