J reconstr Microsurg
DOI: 10.1055/s-0039-3400236
Letter to the Editor
Thieme Medical Publishers 333 Seventh Avenue, New York, NY 10001, USA.

Head and Neck Reconstruction: Does Surgical Specialty Affect Complication Rates?

Andrés M. Bur
1  Department of Otolaryngology—Head and Neck Surgery, University of Kansas, Kansas City, Kansas
,
Steven B. Cannady
2  Department of Otolaryngology—Head and Neck Surgery, University of Pennsylvania, Philadelphia, Pennsylvania
,
Urjeet A. Patel
3  Department of Otolaryngology—Head and Neck Surgery, Northwestern University, Chicago, Illinois
,
Eben L. Rosenthal
4  Department of Otolaryngology—Head and Neck Surgery, Stanford University, Palo Alto, California
,
Mark K. Wax
5  Department of Otolaryngology—Head and Neck Surgery, Oregon Health and Science University, Portland, Oregon
› Author Affiliations
Funding None.
Further Information

Publication History

09 August 2019

29 September 2019

Publication Date:
13 November 2019 (online)

Comments on “Head and Neck Reconstruction: Does Surgical Specialty Affect Complication Rates?”

Head and Neck Reconstruction: Does Surgical Specialty Affect Complication Rates?

We read with great interest the article by Drinane et al comparing outcomes after microvascular head and neck reconstruction between plastic and reconstructive surgery (PRS) and otolaryngology using the American College of Surgeons—National Surgical Quality Improvement Program (ACS-NSQIP) dataset.[1] Based on their analysis, the authors concluded that “plastic surgery patients have superior outcomes with regards to free-tissue transfers of the head and neck when compared to otolaryngology patients.” We have numerous concerns regarding the authors' methodology and the validity of their conclusions.

ACS-NSQIP provides information regarding the surgical specialty of the primary surgeon performing the procedure. However, for cases involving multiple surgeons from different specialties, the specialty of additional surgeons is not captured, nor is it possible to determine which specialty performed which procedure(s).[2] The authors erroneously assumed that the primary surgeon performed microvascular reconstruction in every case, which is likely to result in cases being misclassified with respect to which specialty performed the reconstruction.

The authors have not provided sufficient detail of their methods to allow their results to be reproduced. There is no description of how head and neck reconstructive cases were identified. Instead, the authors queried the ACS-NSQIP dataset by current procedural terminology (CPT) code using an incomplete list of free-tissue transfer codes (“15757—free skin flap with microvascular anastomosis” is notably missing). Further, the authors did not describe how cases involving free-tissue transfer outside of the head and neck were excluded.

The authors did not consider whether defects involved the upper aerodigestive tract. Other studies have found clean-contaminated wound class to be independently associated with length of stay, reoperation, and wound complications after head and neck microvascular reconstruction.[3] [4] Site of reconstruction has also been associated with risk of postoperative complications.[5] The authors comment that readmission was significantly higher for patients treated by otolaryngology, yet their results do not support this.

The power of large databases is that they provide large numbers of cases. However, to avoid drawing incorrect conclusions, proper statistical analysis is essential. The authors performed multiple, univariate comparisons of demographic and complication variables and set the level of significance to 0.05. In this scenario, adjusting for multiple comparisons using Bonferroni's correction or other statistical techniques is obligatory.

As the authors acknowledge, referral patterns and institutional practices influence the types of cases that are reconstructed by otolaryngology versus PRS. The authors report significant differences between these groups in several baseline characteristics as follows: chronic obstructive pulmonary disease (COPD), metabolic syndrome, weight loss, smoking, etc. However, they made no effort to account for these differences when analyzing the effect of surgical specialty on complications. Without controlling for confounding factors using multivariable logistic regression, a methodologically sound association between treating specialty and complications cannot be established to support the authors' conclusion.

We wholeheartedly support the authors' call to increase collaboration between PRS and otolaryngology-trained microvascular surgeons who perform head and neck reconstruction. However, publication of comparisons regarding who has “superior outcomes” using scientifically unsound methods does little to foster collaboration between our specialties and is to the detriment of our patients.