C. Pino1, L. Gerhardinger3, J. W. Scott3, N. F. Sangji3, M. R. Hemmila3, A. N. Miller4, B. W. Oliphant1,2 1Detroit Medical Center, Department Of Orthopaedic Surgery, Detroit, MI, USA 2University Of Michigan, Department Of Orthopaedic Surgery, Ann Arbor, MI, USA 3University Of Michigan, Department Of Surgery, Ann Arbor, MI, USA 4Washington University, Department Of Orthopaedic Surgery, St. Louis, MO, USA
Introduction: The adoption of ICD-10-CM (International Classification of Diseases, Tenth Revision, Clinical Modification) resulted in a substantial increase in the number of possible diagnoses along with an increase in the granularity for describing severity and location. From a trauma standpoint, the ability to define the laterality of the diagnosis is key for patients with multiple extremity injuries. However, in spite of this potential, it is unclear whether the appropriate extremity side (left or right) is actually being selected or if the unspecified option is chosen or defaulted to through cross-walking registry software. Also, it is unknown if certain fractures occur disproportionately on one side.
Methods: Data from the National Trauma Data Bank (NTDB) from 2017-2019 and the Michigan Trauma Quality Improvement Program (MTQIP) from 2017-2021 were utilized. Upper and lower extremity fractures (excluding spine, pelvis and hand) were classified using ICD-10-CM codes and the laterality (left, right or unspecified) was recorded. The percentage of the sidedness of the injuries for each injury group was calculated. A fractional probit model was used to assess for factors (e.g. trauma center level, Injury Severity Score and total number of fractures) that impacted the percentage of unspecified fractures.
Results: We identified 1,146,199 patients and 1,578,550 fractures in NTDB and 43,445 patients and 45,643 fractures in MTQIP. In NTDB 6.2% of all fractures were unspecified (range: 1.7% [middle cuneiform] to 25.6% [other tarsal bone]) and in MTQIP 4.1% of all fractures were unspecified (range: 0% % [middle cuneiform] to 32.6% [forearm]). In NTDB the biggest difference in laterality was found in the talus (R: 55.8% vs. L: 36.6%) and in MTQIP this occurred in ICD-10 codes for “other” fractures of the feet and toes (R: 67.6% vs. L: 29.7%). In MTQIP, ISS between 15-35 was associated with a higher risk of being coded as unspecified while level 2 and level 3 centers were less likely to code an unspecified fracture. A patient’s total number of fractures did not affect the likelihood of an unspecified code. (Table)
Conclusion: Overall, the percentage of unspecified fractures coded in large trauma registries does not appear to be substantial, however there are some specific injuries with a sizable proportion. Interestingly, the biggest differential in laterality occurred in the foot and raises the question of an association with driving automobiles. The factors associated with an increased likelihood of having an unspecified code should be examined for means to lessen this occurrence.
