Computer-Based Natural Language Search Applied to the Electronic Medical Record for Tonsil Triage

OBJECTIVES

To determine significant histologic findings in tonsils and categorize clinical settings in which they occur to identify cases benefiting from histopathologic examination using a computer-based natural language search (NLS) applied to the electronic medical record.

METHODS

The pathology database was queried for tonsillectomy cases accessioned between 2002 and 2018. Tonsils with microscopic examination were reviewed, and indication for examination and diagnoses were tallied. Clinical risk of malignancy was correlated with findings. A NLS was used to interrogate preoperative clinical records of the same group of patients. The search identified cases at risk of significant histologic findings and was implemented as part of standard practice.

RESULTS

Of the 18,733 bilateral tonsillectomies identified in the pathology database, 494 were palatine tonsils that underwent microscopic examination, 134 had indications concerning for malignancy, and 14 had significant findings on histologic examination. When the NLS was applied to the medical record of the same group, 223 cases were identified as having risk of malignancy, including all flagged by surgeons and pathologists and 89 additional cases. Clinical implementation resulted in identification of all cases benefiting from examination.

CONCLUSIONS

A NLS applied to the electronic medical record to select tonsils for examination was superior to relying on surgeons and pathologists.

Surgeon Preference on Sending Routine Tonsillectomy Specimens for Pathological Evaluation

INTRODUCTION

Despite the presence of a growing body of literature suggesting cost-ineffectiveness of routine pathologic analysis of tonsillectomy specimens, little is known about common institutional policies and practice patterns of pediatric otolaryngologists. The objectives of this study were to determine the prevalence of routine pathological evaluation of tonsillectomy specimens for uncomplicated pediatric adenotonsillectomy procedures and to evaluate opinions regarding this controversy among board-certified pediatric otolaryngologists.

METHODS

This was a cross-sectional survey study sent to board-certified pediatric otolaryngologists currently practicing and registered with the American Society of Pediatric Otolaryngology (ASPO) assessing their institutions' or practices' current policies on sending routine tonsillectomy specimens for pathology, their experience with this practice, and their opinions on whether routine pathologic analysis should be employed. Basic statistical analysis was then conducted.

RESULTS

Respondents mostly practiced in an academic setting (68.4%), with the next most common being academically affiliated private practice (21.8%), and private practice was the least common (9.8%). Most respondents (85.1%) did not agree with routine pathologic analysis of otherwise uncomplicated pediatric tonsillectomy specimens.

CONCLUSION

Most pediatric otolaryngologists who responded to this survey do not support routine pathological analysis of otherwise uncomplicated pediatric tonsillectomy specimens. However, the results are likely biased and should be interpreted carefully, since only a small percentage of pediatric otolaryngologists responded to the survey. Potential cost savings could be seen by patients, payers, and hospital systems with judicious use of surgical pathology, specifically in cases with concurrent signs or symptoms suspicious for malignancy.

Tonsillectomy and hematologic malignancy: Should routine pediatric tonsillectomy specimens be sent to pathology?

OBJECTIVES

Routine pediatric tonsillectomy ± adenoidectomy (T ± A) is one of the most common procedures for children worldwide, accounting for approximately 2000 procedures per year at our institution. To determine the utility of pathologic analysis of routine, nonsuspicious pediatric tonsil specimens, we investigated the incidence of hematologic and lymphoid malignancy diagnosed at the time of or following T ± A.

METHODS

Retrospective review of patients 0-18 years undergoing T ± A between 2012 and 2020 with or without pathologic analysis.

RESULTS

Included were 14,141 patients who underwent routine T ± A (mean age 11 ± 4.6 years, 48% female). Of these, tonsils of 2464 patients were sent to pathology, where zero were found to harbor malignancy. Seven patients (0.050%) developed malignancy after T ± A. Of these, 4 had unremarkable tonsils per pathology, and 3 did not have tonsils analyzed. There were 5 cases of Acute Lymphocytic Leukemia (ALL, 0.035%), 1 case of Acute Myeloid Leukemia (0.007%), and 1 case of Lymphoma (0.007%). The average length of time from T ± A to diagnosis was 2.4 ± 1.8 years.

CONCLUSION

No cases of occult malignancy were identified in specimens from routine T ± A with pathologic analysis, even among patients who later developed malignancy. This study suggests that sending routine pediatric T ± A specimens for formal pathologic analysis is an inefficient use of resources without appreciably improving the quality and safety of patient care.

Occult Malignancy in Adult Tonsillectomy for Benign Indication

OBJECTIVE

National pathology guidelines recommend full pathologic analysis for all adult tonsillectomy specimens. We evaluated the available data on occult malignancy in adult tonsillectomy for benign indication, and created a screening system to reduce the risk of missed malignancies if routine histopathologic examination were to be discontinued.

STUDY DESIGN

Retrospective chart review and systematic review of the literature.

SETTING

Tertiary care academic hospital and multi-hospital private healthcare system.

SUBJECTS AND METHODS

A systematic literature review identified case series of adult tonsillectomy. Retrospective chart review at our institutions from 2000 to 2016 produced an additional case series. The pooled rate of occult malignancy was determined, and re-analyzed using criteria based on preoperative risk factors designed to identify patients requiring full pathologic analysis. The predicted effects of prospective application of the proposed criteria were calculated. Pooled occult malignancy prevalence was estimated.

RESULTS

Literature review and our own case series yielded 12,094 total cases. Occult malignancy prevalence in the combined data was 0.033%, representing four occult malignancies. Three out of the four would have been selected for full pathology preoperatively with use of the proposed criteria. Statistical analysis indicates that the predicted frequency of occult malignancy incidence in cases negative for the criteria is 0.01%, or 1/10,000.

CONCLUSION

Application of the proposed criteria to adults undergoing tonsillectomy for benign indication identifies a subset of patients with an estimated incidence of occult malignancy similar to that reported for pediatric tonsillectomy, and potentially may permit safe elimination of pathologic analysis of their tonsil specimens.

LEVEL OF EVIDENCE

Pooled analysis of case series from the literature and a single institution, level 4.