September 2019 HSC Section 1 Congenital and Pediatric Problems

Research Original Investigation

Association of Respiratory, Allergic, and Infectious Diseases With Removal of Adenoids and Tonsils

ing about why tonsillectomy or adenoidectomy was performed (eg, indications, disease burden, national guide- lines), and must assume it was a vague interplay of clini- cian recommendations and patient preference. The surgi- cal group is unlikely tohave the same known (andunknown) risk factors for future illness as the control group, making it impossible to separate out treatment effects from other influences. Not knowing how the patients were chosen for surgery also makes it impossible to generalize results be- yond the study sample. 4. Measurement bias results from errors in how diseases or outcomes are measured and can be problematic when relying on diagnostic codes in a database. Codes are often not used consistently or accurately by clinicians, and do not reflect disease frequency, duration, or severity, which can be important in understanding the clinical signifi- cance of results. The Bottom Line for Patients and Clinicians Clinicians may wish to counsel families of children 9 years or younger that tonsillectomy may have a slight impact on fu- ture risk for some chronic diseases, which include asthma, COPD, pneumonia, infectious diseases, skin conditions, and upper and lower respiratory disorders. The evidence, how- ever, is limited to 1 study, says nothing about causation, and is subject to many biases inherent in observational research that greatly limit our confidence in the results. Nonetheless, both young surgeons and old barbers would be wise to be familiar with this study and remain alert to new research that brings further clarity to what is presently a rather tenuous association.

1. Confounding occurs when a variable that is related to both the predictor (eg, surgery) and outcome variables (eg, disease incidence) causes an apparent, but false, associa- tion. For example, the authors did not control for smoke exposure (primary or secondhand), which could increase the risk for surgery and independently increase the risk for many diseases (asthma, infections, sinusitis, COPD, etc). Similarly, antibiotic consumption in childhood (also not measured) is associated with tonsil and adenoid surgery and with certain acute and chronic disease related to an altered microbiome, bacterial resistance, and adverse events. 2. Reverse causation is present when B occurs after A, is as- sumed to be caused by A, but the reverse is true. Finding more otitismedia after adenoidectomy is best explained by realizing that children who are otitis-prone at baseline are most likely to have surgery, and most likely to have more otitis media after surgery than healthy controls. Children with asthma, allergies, and frequent respiratory infections consumemore health care than controls, aremore likely to see specialists, and more likely to become surgical candi- dates. The surgery did not cause these conditions, it is more likely that the conditions themselves led to the sur- gery. Last, we do not know the indications for surgery (si- nusitis, tonsillitis, otitis media, nasal obstruction, sleep disturbance, etc) in this cohort of 60000 children, which could greatly impact the subsequent presence of acute and chronic disease. 3. Selection bias distorts outcomes when patients who re- ceived treatment differ from control patients in past expo- sures and other important characteristics. We know noth-

ARTICLE INFORMATION Author Affiliation: Otolaryngology, SUNY

Disclosure of Potential Conflicts of Interest and none were reported.

guideline: tonsillectomy in children. Otolaryngol Head Neck Surg . 2011;144(1)(suppl):S1-S30 . 4 . Byars SG, Stearns SC, Boomsma JJ. Association of long-term risk of respiratory, allergic, and infectious diseases with removal of adenoids and tonsils in childhood [published online June 7, 2018]. JAMA Otolaryngol Head Neck Surg . doi: 10.1001 /jamaoto.2018.0614 5 . Sessler DI, Imrey PB. Clinical research methodology 2: observational clinical research. Anesth Analg . 2015;121(4):1043-1051 .

Downstate Medical Center, Brooklyn, New York. Corresponding Author: Richard M. Rosenfeld, MD, MPH, Distinguished Professor of Otolaryngology, SUNY Downstate Medical Center, 450 Clarkson Ave, MSC 126, Brooklyn, NY 11203 ( richrosenfeld @msn.com ).

REFERENCES 1 . McCoul ED. Legacies of the Tonsil Hospital. Otolaryngol Head Neck Surg . 2010;143(1):4-7, 7.e1-7.e2 . 2 . Fowler RH. Tonsil Surgery . Philadelphia: F.A. Davis Co; 1931. 3 . Baugh RF, Archer SM, Mitchell RB, et al; American Academy of Otolaryngology-Head and Neck Surgery Foundation. Clinical practice

Published Online: June 7, 2018. doi: 10.1001/jamaoto.2018.0622

Conflict of Interest Disclosures: The author has completed and submitted the ICMJE Form for

JAMA Otolaryngology–Head & Neck Surgery July 2018 Volume 144, Number 7 (Reprinted)

jamaotolaryngology.com

© 2018 American Medical Association. All rights reserved.

112