Packed Red Blood Cell Transfusion Associates With Acute Kidney Injury After Transcatheter Aortic Valve Replacement

Akeel M. Merchant; Javier A. Neyra; Abu Minhajuddin; Lauren E. Wehrmann; Richard A. Mills; Sarah K. Gualano; Dharam J. Kumbhani; Lynn C. Huffman; Michael E. Jessen; Amanda A. Fox


BMC Anesthesiol. 2019;19(99) 

In This Article


Demographic, Peri-procedural and Clinical Characteristics

Table 1 describes clinical and procedural characteristics of the 116 subjects included in the study, with stratification according to whether the patient did or did not develop AKI after TAVR. Twenty subjects (17.2%) developed AKI after TAVR: 19 developed Stage 1 AKI and one developed Stage 2 AKI. No patients developed Stage 3 AKI or required dialysis. Subjects had a mean age of 81 years with a SD of 7.5 years, and 55% of subjects were male. Of the total cohort, 31 subjects (26.7%) were transfused at least 1 unit of pRBCs in the perioperative period. Two patients died within 7 days after TAVR, with one developing AKI prior to death and the other not developing AKI. Post-TAVR ICU stay was significantly longer in the post-TAVR AKI group (median 2, IQR 1, 4 days) versus the group that did not develop post-TAVR AKI (median 2, IQR 1, 2 days) (P = 0.02). Post-TAVR hospital stay was also significantly greater in the AKI group (median 5, IQR 3, 12 days) versus the no-AKI group (median 3, IQR 2, 5 days) (P = 0.01).

Univariate Associations With Development of Post-TAVR AKI

Table 1 shows univariate associations between patient demographic, procedural and clinical variables and the development of post-TAVR AKI with data stratified according to whether patients did or did not develop post-TAVR AKI. Patients who developed post-TAVR AKI received significantly more pRBC transfusions than patients who did not (mean units transfused in AKI group 1.7 versus 0.4 units in no AKI group; P = 0.03; Table 1). Of the patients who developed post-TAVR AKI, 55% received pRBC transfusion, while only 21% of the patients who did not develop AKI received pRBCs (P = 0.002).

More patients who developed AKI were anemic prior to TAVR (75% of the AKI group versus 51% of the no AKI group, P = 0.05). Furthermore, patients who developed AKI had significantly lower nadir Hgb measurements obtained during the TAVR procedure and the first 24 h following TAVR (mean Hgb 8.8 ± 1.5 g/dL in the AKI group versus 9.8 ± 1.7 g/dL in the no AKI group; P = 0.02). Figure 1 shows the patients who did and did not receive pRBC transfusions and the nadir Hgb recorded during TAVR or the 24 h following TAVR for each patient. Since patients may receive pRBCs in the setting of ongoing bleeding and need for volume resuscitation in the operating room, it is possible that some actual nadir Hgb was lower than recorded.

Figure 1.

Nadir measured hemoglobin (intra-procedure and first 24 h post-TAVR) and number of patients transfused and not transfused pRBCs at these hemoglobin values. pRBC = packed red blood cells

Continuous intra-procedure blood pressure monitoring was done via arterial line. Neither the occurrence of any episode of intraoperative hypotension with MAP < 60 mmHg for ≥5 consecutive minutes, the total number of these intraoperative hypotension episodes, nor the total number of intraoperative minutes included in these hypotensive episodes was significantly associated with development of AKI after TAVR. Having a period of ≥10 consecutive minutes of MAP < 60 mmHg was also not significantly associated with development of post-TAVR AKI. Intra-procedure rapid pacing was utilized in 90% of the patients who developed post-TAVR AKI and in 75% of the patients who did not. This difference was not statistically significant.

Maximum number of concurrent inotropes and vasopressor drugs utilized during the TAVR procedure did not differ significantly between the patients who did and did not develop post-TAVR AKI. However, maximum number of concurrent inotropes and vasopressor drugs utilized during the post-TAVR period (up to hospital discharge or through post-procedure day 5) was significantly greater in the patients who developed post-TAVR AKI than in those who did not develop AKI (mean number of concurrent vasoactive drugs was 1.1 ± 1.2 SD in the AKI group and 0.5 ± 0.7 SD in the no AKI group; P = 0.03).

That there is no significant difference in mean contrast volume administered to the patients who did and did not develop AKI (108 mL in the AKI group and 103 mL in the no-AKI group). There was also no significant difference between the AKI and no-AKI groups with regards to percentage of subjects who received greater than or equal to 150 mL of contrast during TAVR.

Multivariable Adjusted Associations Between pRBC Transfusion and Development of Post-TAVR AKI

In order to adjust for potential confounders of the association between pRBC transfusion and the development of AKI after TAVR, a multivariate analysis was performed using a logistic regression model with total number of pRBCs transfused, nadir Hgb, pre-procedural estimated glomerular filtration rate (eGFR), and post-TAVR maximum number of inotropes and vasopressors used concurrently (Table 2). Total units of pRBC transfused (OR = 1.67 per unit, 95% CI 1.13–2.47; P = 0.01) remained independently associated with post-TAVR AKI after adjustments for these other clinical risk factors. Nadir Hgb was no longer significantly associated with post-TAVR AKI after adjusting for these additional variables. Figure 2 further illustrates the finding that pRBC transfusion rather than peri-procedural nadir Hgb seems to drive the association with post-TAVR AKI; the TAVR study cohort is stratified into 4 groups based on if subjects' nadir Hgb measured during the TAVR procedure and the 24 h post-TAVR was recorded as < 8 g/dL versus ≥8 g/dL, and then within these nadir Hgb categories patients are stratified according to whether they were or were not transfused pRBCs. AKI development was not significantly different between the patients who were transfused pRBC who had a nadir Hgb of < 8 g/dL and those who were transfused pRBCs and had a nadir Hgb ≥ 8 g/dL (P = 0.45). Hgb < 8 g/dL was selected since anesthesiologists typically aim to maintain a Hgb > 7 g/dL and may consider transfusion in the setting of potential ongoing procedural blood loss once Hgb falls below 8 g/dL. Post-TAVR maximum number of concurrently administered inotropes and vasopressors (OR = 2.09 increase for each drug, 95% CI 1.19–3.67; P = 0.01) also independently associated with post-TAVR AKI in the multivariable clinical model.

Figure 2.

Number of patients with post TAVR acute kidney injury (AKI) stratified by packed red blood cell (pRBC) transfusion and periprocedural anemia (nadir hemoglobin < 8 g/dL versus ≥8 g/dL)