Background/Purpose:

Elevated hemoglobin A1c (HbA1c) has been associated with a 9-fold increase in wound complications after total joint replacement (TJR). Data suggest that including HbA1c in perioperative screening lowers complications. The CDC recommends perioperative glycemic control as a prevention strategy for infections in patients with and without DM. We aimed to assess how frequently HbA1c was tested prior to TJR in a large claims database.

Methods:

Within Medicare Parts A/B/D (2010-2014), we identified patients who were ≥ 65 years old and underwent TJR (hip or knee). The index date was date of TJR; all patients were free of TJR and continuously enrolled in Medicare for ≥1 year prior to TJR. We created 4 mutually exclusive groups during the baseline period ranging from 365 to 90 days prior to TJR; 1) non-DM based on no ICD-9 code for DM or DM complications, no claim for insulin or anti-DM medication; 2) DM without medication, based on a diagnosis code for DM but no claim for insulin or non-insulin antidiabetic medication; 3) DM on non-insulin antidiabetic medication; and 4) DM on insulin with or without other antidiabetic medication. The outcome was HbA1c test ordered 90 days prior to TJR. Covariates including age, sex, race, comorbidities, medications, outpatient visits, and number of HbA1c tests, were collected during the baseline period. We calculated the proportion of patients receiving HbA1c or serum glucose test in the 90 days prior to TJR. We used logistic regression with adjustment for covariates to assess characteristics associated with HbA1c testing.

Results:

Seventy-two percent were non-DM, 11% DM without medication, 13% DM on non-insulin medications, and 4% DM on insulin. Mean age was 73 -75. Patients with DM had more outpatient visits, and co-morbid diseases compared to those without DM. During the baseline period mean number of HbA1c tests was 0.1 in the non-DM group, 0.8 in DM without medication, 1.1 in DM on non-insulin medications, and 1.3 in DM on insulin. Only 5% of patients without DM had an HbA1c testing within 90 days of TJR, compared to 26% in those with DM not on medication, 39% in DM on non-insulin medications and 43% with DM on insulin. Serum glucose testing was obtained more frequently- 37% of non-DM, and 46-50% of patients with DM. In patients without DM, non-white race and obesity lead to increased HbA1c testing. In patients with DM on no medication, those with DM complications had increased HbA1c testing. For all groups having HbA1c tested in the baseline period was associated with having HbA1c tested prior to TJR (Table).

Conclusion:

In this large population based cohort of patients undergoing TJR, HbA1c testing perioperatively occurred in less than half of patients with DM. The presence of diseases co-morbid to DM was not strongly associated with increased screening in the non-DM population. Further study on the utility of perioperative HbA1c monitoring and postoperative outcomes is warranted.

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/peri-operative-glycemic-evaluation-prior-to-total-joint-replacement-in-patients-with-and-without-diabetes/