Portal de Periódicos da CAPES

Acute kidney injury during parvovirus B19‐induced transient aplastic crisis in sickle cell disease

2018; Wiley; Volume: 93; Issue: 8 Linguagem: Inglês

10.1002/ajh.25140

1096-8652

Jamie Oakley, Rima S. Zahr, Inmaculada Aban, Varsha Kulkarni, Rakesh P. Patel, Julia L. Hurwitz, David J. Askenazi, Jane S. Hankins, Jeffrey D. Lebensburger,

Bone and Joint Diseases

To the Editor: Acute kidney injury (AKI) is associated with progression to chronic kidney disease.1 In Sickle Cell disease (SCD), AKI occurs during pain crisis and acute chest syndrome (ACS) and is associated with an acute decline in hemoglobin (Hb).2, 3 Two pathophysiologic mechanisms may account for the association between acute anemic events and the development of AKI. First, acute anemia can reduce oxygen delivery to the kidneys promoting ischemic tubular injury. Second, hemolysis, leading to an increase in cell free hemoglobin and heme, can reduce blood flow to the kidneys causing AKI.4 In our prior studies of AKI during pain or ACS, it was unclear which of these pathologic factors play a stronger role in AKI development: the Hb decline itself or an increase in free heme/hemoglobin secondary to hemolysis. To evaluate the role of anemia independently of hemolysis on AKI, we examined renal function of SCD patients who developed acute parvovirus B19-induced transient aplastic crisis (TAC). TAC results in a severe anemia episode caused by prolonged reticulocytopenia rather than acute hemolysis as seen in ACS or pain crisis.5 We hypothesized that the severe Hb decline induced by TAC would lead to AKI. Second, we compared the odds of developing AKI during TAC based on the decline in hemoglobin to the odds of developing AKI during admission for pain crisis or ACS. The University of Alabama at Birmingham (UAB) institutional review board approved the study at UAB and St. Jude Children's Research Hospital (St. Jude). UAB identified cases through a retrospective study and St. Jude identified cases through a 24 month prospective study of TAC (iSCREEN, NCT02261480). TAC cases were confirmed by parvovirus B19 serology or PCR. The primary outcome measure was the development of AKI defined as an increase in serum creatinine (SCr) by ≥0.3 mg/dL or 50% increase from baseline. Therefore, peak SCr during TAC and baseline steady state (most recent measurement within 12 months) were recorded. Cases were excluded if patient developed pain crisis or ACS occurred during TAC. Additional variables included: Hb, white blood cell count (WBC), absolute neutrophil count, platelet count, and absolute reticulocyte count (ARC) from steady state and during TAC. Continuous variables were analyzed using Wilcoxon test. To evaluate differences in change in Hb on the odds of developing AKI, changes in Hb during TAC were compared to those during pain crisis or ACS admissions by fitting a generalized linear mixed model for binary outcome with random intercept to incorporate the correlation among repeated hospitalizations for the same subject. We utilized 2 previously described patient cohorts to analyze the impact of changes in Hb on the development of AKI during pain crisis and ACS.2, 3 Analyses were done in SAS version 9.4 (Cary, North Carolina). We identified 34 cases of confirmed acute parvovirus B19-induced TAC with baseline and hospital serum creatinine measurements. The mean age of participants was 6.8 (s.d. 4.4) years and 20 participants (59%) were female. Twenty-two participants had genotypes HbSS or HbSB0-thalassemia, 9 HbSC or 3 HbSB+ thalassemia. Only 3 participants (9%) developed AKI during their TAC event, all with HbSS or SB0-thalassemia. Patients with AKI had a non-significant lower admission Hb during TAC event than patients without AKI (3.4 vs 5.7 g/dL, P = .052) and a non-significant larger decrease in Hb from baseline to TAC event (4.7 vs 3.4 g/dL, P = 0.08). For the 3 AKI cases, the Hb levels were 2.1, 3.0, and 5.0 g/dL on admission and the decline in hemoglobin from baseline to admission was 5.0, 4.6, and 4.5g/dL. When restricting outcomes to 22 patients with HbSS or SB0-thalassemia, we identified no significant differences in admission Hb (3.4 vs 4.8 g/dL, P = .2) or change in Hb from baseline (4.7 vs 3.5 g/dL, P = .1) among patients with or without AKI. WBC, ARC, and platelets during the TAC event were not significantly associated with the development of AKI. Next, we developed a model to evaluate the impact of change in Hb from baseline for patients with HbSS or HbSB0-thalassemia admitted with either TAC, ACS, or pain crisis on AKI. For this analysis we used 346 cases of HbSS/SB0-thalassemia who experienced ACS or pain, among which 45 developed AKI.2, 3 Adjusted for change in Hb, pain subjects had higher odds of developing AKI relative to acute parvovirus B19-induced TAC (OR = 11.23, P = .002). Adjusted for change in Hb, no evidence was found to conclude that the odds of developing AKI during ACS versus acute parvovirus B19-induced TAC was higher (OR = 2.94, P = .134). We illustrate this association further by plotting the estimated probability (based only on the fixed effects for ease of interpretation) of AKI as a function of change in Hb by sickle cell complication (Figure 1). In particular, for patients experiencing a 3 g/dL decrease in Hb from baseline, patients with acute parvovirus B19-induced TAC would have a 6% (95% CI 0.02, 0.17) probability of AKI, patients with ACS have a 15% (95% CI 0.08, 0.25) probability of AKI and patients with pain crisis have a 40% (95% CI 0.26, 0.55) probability of developing AKI. Estimated probability with 95% confidence intervals of AKI by change in hemoglobin from baseline for acute chest syndrome admissions, pain admissions, and parvovirus admissions Our data demonstrate that AKI occurs in children with SCD during acute parvovirus B19-induced TAC; however, this complication only occurred when the Hb declined by more than 4 g/dL from baseline or the Hb was less than 5 g/dL at admission. For each 1g/dL Hb decline during acute parvovirus B19-induced TAC, we identified an estimated 2% odds of AKI, which contrasts with our prior work in ACS and vaso-occlusive pain crisis that showed an estimated 80% and 50% odds of developing AKI with each 1 g/dL drop in Hb. Therefore, we conclude that the mild drop in Hb when hemolysis occurs may itself play a more significant role in the pathology of AKI during SCD crisis as compared to agenerative anemia, unless the decline in Hb leads to a very severe anemia. Our study provides plausibility to the hypothesis that acute hemolysis promotes AKI via free tetrameric Hb, which is filtered by the glomerulus, dissociates into dimers and then releases free heme. The latter causes AKI by either direct cytotoxic injury, oxidative stress, or tubular injury through endothelial dysfunction. Recent data suggests that free heme reduces renal blood flow or promotes endothelial dysfunction during ACS.6 Therefore, SCD patients presenting with an acute drop in Hb should be monitored for the development of sickle cell-related complications and end-organ toxicity, including AKI. A few limitations are worth noting. The urine output definition of AKI was not utilized. SCr has limitations in SCD and may underestimate AKI during the TAC event. Also, we did not obtain LDH or plasma free heme/hemoglobin levels as standard of care during admission for TAC. In conclusion, we show that an acute decline in Hb during TAC may lead to AKI, albeit with a lower risk in relation to an acute Hb decline during ACS or pain crisis. This difference may be due to the added contribution of the free heme or nephrotoxic medication use during a pain event, compounding the acute Hb decline. To explore the hypothesis that free heme contributes to AKI during vaso-occlusion, we are prospectively collecting blood for measurement of free heme/hemoglobin and urine (urine biomarkers of AKI) on all patients admitted for either ACS or pain (NCT 03105271). The authors would like to thank Jola Dowdy, CRA of the iSCREEN study, Alyssa Cotton, CRNP, Rhiannon Penkert PhD and all the providers at UAB and St. Jude for caring for the patients with sickle cell anemia. The authors also recognize Dr. Michael DeBaun for his critical role in developing the concept of this manuscript and pushing the authors to improve the care for patients living with sickle cell anemia. Additional Supporting Information may be found online in the supporting information tab for this article. Supporting information 1 Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.