Is the Presence, Increased Number, or Certain Category of Patient-Reported Allergies Associated with a Poor Outcome...

The impact of patient-reported allergies (PRAs) on the outcome following shoulder arthroplasty remains unclear. Therefore, the purpose of this study was to address three questions in patients undergoing anatomic total shoulder arthroplasty (aTSA): (1) Is the presence of a PRA associated with poorer PROs or revision rate? (2) Is there a certain number of PRAs that are associated with poorer PROs? (3) Are there specific categories of PRAs that are associated with poorer PROs or revision rate?

A retrospective review was performed that included patients undergoing aTSA with a minimum follow-up of two years. Outcome measures included 2-year Simple Shoulder Test (SST) scores, the percentage of maximal possible improvement (%MPI) and revision rate. PRAs were categorized as allergies to (1) antibiotics, (2) pain medications, (3) "other" medications (medications other than antibiotics or pain medications) (4) food, and (5) materials (i.e. adhesives or latex). Univariable and multivariable regression analyses were used to determine if the presence of a PRA and certain allergy category were associated with a poor outcome, defined by an improvement of less than the minimal clinically important difference (MCID) for SST score and %MPI. Chi-squared testing was performed to determine if there was a threshold at which the number of PRAs was associated with a poor outcome.

302 of the 547 (55%) included patients reported one or more PRAs. On univariable analysis, the presence of a PRA was not associated with a poor outcome in terms of SST score or %MPI, nor an increased rate of revision surgery. However, patients who reported two or more PRAs were more likely to have a poor outcome based on %MPI compared to those without a PRA (p=0.002) and those with only one PRA (p=0.030). On univariable analysis of PRA categories, patients with PRAs categorized as pain medication and "other" medication allergies were more likely to have a poor outcome based on %MPI (p=0.003 for both). However, only the "other" medication category remained significant on subsequent multivariate analysis that included other significant covariates (p=0.021).

In this study, the simple presence of a PRA was not associated with a poor outcome or increase in revision surgery. However, patients who reported two or more allergies had poorer outcomes compared to those with an absence of or one PRA. When evaluating PRA types, allergies to medications other than antibiotics or pain medications were independently associated with a poor outcome.

III - Retrospective Cohort Comparison, Prognosis Study.