In:
The Laryngoscope, Wiley, Vol. 126, No. 4 ( 2016-04), p. 851-857
Abstract:
There is growing interest in reducing the exposure of patients to allogeneic blood transfusions by lowering preoperative cross‐matched blood ordering and adopting alternative practices, such as autologous blood donations. Our aim was to investigate the predictors for perioperative blood transfusion (PBT) in head and neck cancer patients undergoing neck dissection (ND). Study Design Retrospective cohort study. Methods Retrospective observational study. All patients who underwent ND between January 2011 and August 2014. The primary outcome measure was PBT. Predictors tested included: gender, age, American Society of Anesthesiologists comorbidity score, Charlson comorbidity index, preoperative hemoglobin level, head and neck primary tumor location, tumor and nodal staging, side and laterality of ND, central versus lateral ND, elective ND, preoperative chemotherapy/radiotherapy/I 131 therapy, history of previous ND, other surgical procedures in addition to the ND, bone resection, use and type of reconstruction, and the use of bony free flap reconstruction. Results Twenty‐one preoperative and operative variables were tested for an association with PBT using univariate and multivariate analyses. Multivariate analysis found only the following three predictors to be significantly associated with PBT in patients undergoing ND: low preoperative hemoglobin level, advanced N stage, and concurrent reconstructive surgery. Conclusion Evaluation of specific risk factors for predicting the need for PBT prior to neck dissection may be helpful in identifying the head and neck cancer patients in whom preoperative ordering of cross‐matched blood is required or who could benefit from alternative means, such as preoperative autologous blood donation. Level of Evidence 4. Laryngoscope , 126:851–857, 2016
Type of Medium:
Online Resource
ISSN:
0023-852X
,
1531-4995
Language:
English
Publisher:
Wiley
Publication Date:
2016
detail.hit.zdb_id:
2026089-1
Bookmarklink
