Letters to the Editor: Letter to the Editor
To the Editor
I enjoyed the recent article by Choi et al1 regarding postoperative anemia and acute kidney injury in patients undergoing total hip arthroplasty. In addition to limitations highlighted by Shander and Roy2 in their excellent accompanying editorial, there are a few areas that warrant further clarification.
First, multiple exclusion criteria were used in the study including, among others, the use of vasoactive agents (eg, ephedrine, phenylephrine) perioperatively, the presence of unstable vital signs postoperatively (ie, mean arterial pressure <60 or those receiving colloid or red blood cells [RBCs]), and hemoglobin values that crossed the 10 g/dL threshold postoperatively. Although all 3 scenarios are common in patients undergoing orthopedic surgery, the authors give no indication that any of the approximately 2500 patients fell into these categories (lease see Figure 1 in their study). For example, it is highly improbable that no patient required intraoperative phenylephrine or had a postoperative hemoglobin value that crossed the 10 g/dL threshold (eg, falling from 10.1 g/dL on postoperative day 1 to 9.9 g/dL on postoperative day 2). In addition, the authors list end-stage renal disease with hemodialysis as an exclusion criterion. In Figure 1, however, they show that >600 patients were excluded because of “decreased glomerular filtration rate.” Assuming that this should be relabeled as “end-stage renal disease with hemodialysis” (which is a very distinct clinical entity), then the incidence would be remarkably high at approximately 16%. Hence, further clarification is needed regarding the actual included study population and relevant exclusion criteria.
Second, if patients had similar preoperative hemoglobin values, similar estimated blood loss, similar colloid administration, and similar RBC transfusion rates (surpassing 50% in each group) and volumes, why would one group develop postoperative anemia (arbitrarily defined as hemoglobin <10 g/dL) and the other not? Assuming the aforementioned factors to be equal, patients with postoperative anemia likely received more crystalloid resulting in hemodilution, suffered more blood loss intraoperatively or postoperatively regardless of estimated blood loss, or had decreased response to RBC transfusion. It should be noted that crystalloids have been associated with kidney injury,3 and the reporting of perioperative crystalloid volume and composition would provide readers with a clearer picture of the significance of postoperative anemia and renal outcomes.
Finally, a restricted analysis of the data based on the presence or absence of intraoperative RBC transfusion would greatly add to the study results, because patients receiving RBCs and achieving postoperative hemoglobin values ≥10 g/dL are likely to be distinct from (1) those receiving RBCs and not achieving this threshold and (2) those not requiring RBC transfusion and maintaining hemoglobin ≥10 g/dL.
I congratulate the group for their efforts to better understand the downstream effects of perioperative anemia, a critically important topic for all anesthesia providers.
Matthew A. Warner, MDDivision of Critical Care MedicineDepartment of AnesthesiologyMayo ClinicRochester, Minnesotawarner.firstname.lastname@example.org
1. Choi YJ, Kim SO, Sim JH, Hahm KD. Postoperative anemia is associated with acute kidney injury in patients undergoing total hip replacement arthroplasty: a retrospective study. Anesth Analg. 2016;122:19231928.
2. Shander A, Roy RC. Postoperative anemia: a sign of treatment failure. Anesth Analg. 2016;122:17551759.
3. Prowle JR, Bellomo R. Fluid administration and the kidney. Curr Opin Crit Care. 2013;19:308314.