Page 41 - ANZCP Gazette APRIL 2022

6
Perfusion 00(0)
Table 3. Postoperative outcomes for patients accepting or refusing transfusion. Accept transfusion
Refuse transfusion
p
<0.001 0.25 0.77 0.37 1.00 1.00 0.49 0.46 0.68
N Postoperative
Any AKI
RIFLE risk class or greater AKI RIFLE injury class or greater AKI Blood loss in 4hours, mL Ventilation time, min
Ventilation time > 48 hours Length of postoperative stay, days Intra-aortic balloon pump
Return to operating theatre Myocardial infarction
Stroke
New dialysis requirement
New renal insufficiency Combined morbidity
In hospital mortality
118
118
22% 8% 14% 5%
0.003 0.018 0.14 0.016 0.80 0.53
8%
175 (120, 280)
10 (6, 17.5) 5%
7 (6, 10) 0%
5% 3% 2% 1% 6% 12% 2%
4% 150 9 4% 6 3% 4% 1% 2% 0% 4% 9% 3%
(97.5, 200) (7, 16)
(5, 8)
AKI: acute kidney injury; RIFLE: Risk, Injury, Failure, Loss of renal function and End-stage renal disease.
Categorical data are represented as % and continuous data as median (interquartile range). Combined morbidity; incidence of postoperative ventilation>48hours, new renal insufficiency, stroke or return to theatre.
lower in patients refusing transfusion than in the con- trol group (8% vs. 22%; Relative Risk (RR), 2.5; 95% confidence interval (CI), 1.26-4.97; p = 0.003). A signifi- cant reduction in any AKI in patients refusing transfu- sion was seen in the later period only (11% vs. 21% (p = 0.18) early, 6% vs. 23% (p = 0.006) late). Patients refusing transfusion had significantly lower postopera- tive blood loss and shorter postoperative length of stay. There was no difference in morbidity or mortality for patients who refused blood products in this study com- pared with patients accepting transfusion.
Discussion
This study found a significantly lower incidence in the primary endpoint of AKI in JW patients who refused blood product transfusion compared with a matched cohort of patients accepting transfusion. While the pre- vious meta-analysis of JWs undergoing cardiac surgery found no difference in renal outcome compared with matched controls, the definition of renal failure was typ- ically a creatinine >2 mg/dL.5 The exception was the study by Guin et al. that used the same definition of any AKI as reported in this analysis, and while they showed no difference between JWs and the controls, the inci- dence of any renal failure was high by comparison (77.8% vs. 77.3%, respectively).
We detail the management of CPB from a large registry integrating electronically acquired physiological data not previously reported in patients refusing blood products compared with patients consenting to transfusion. Unlike
other previous studies, we propensity matched patients preoperative Hb in each group so that CPB factors associ- ated with Hb such as DO2i were not confounded. This was performed a priori since previous studies have reported patients refusing transfusion having higher preoperative Hb. Interestingly, a post hoc analysis revealed no differ- ence in the preoperative Hb between patients refusing transfusion and all patients in the ANZCPR accepting transfusion (n = 30,054) in the ANZCPR (132 vs. 131 g/L; 2007-2012 and 137 vs. 134 g/L; 2013-2018).
Management of MAP during CPB remains contro- versial. In a recent randomised trial, Kandler et al.18 found no difference in AKI in coronary artery bypass patients randomised to a MAP greater than or less than 60mmHg. Our study showed a small but significant increase in MAP in JW patients refusing transfusion in the later period (64 vs. 59 mm Hg) but no difference in the management of MAP during CPB between groups over the total study period (mean median MAP 62mmHg). Postoperative blood loss was lower in patients refusing transfusion as was postoperative length of stay, consistent with other studies.5
Not surprisingly, use of cell salvage in patients refusing transfusion was more than triple that of patients accepting transfusion, highlighting a marked difference in clinical practice with respect to blood conservation strategies offered to patients accepting transfusion. We saw no change over time with the use of RAP; however, the sig- nificantly higher RAP volumes in the later time period likely contributed to the reduced net circuit prime volume and to the shift to a higher nadir Hb. The importance of
Continued on page 40
APRIL 2022 | www.anzcp.org 38