F. N. Williams1, P. Strassle2, S. Jones1, B. Cairns1 1UNC,Surgery/Burns,Chapel Hill, NC, USA 2UNC,Surgery,Chapel Hill, NC, USA
Introduction: Outcomes in burn patients, including mortality, are affected by age of the patient and the extent of the burn injury. This is reflected in the revised Baux score, which is calculated by adding age plus percent total body surface area burned plus 17 if there is an inhalational injury. The result corresponds to a predicted mortality. Tight glucose control following burn injury improves complication and mortality rates. It remains unclear whether a pre-existing diagnosis of diabetes in burn patients influences key outcomes such as infectious complications and mortality.
Methods: The Burn Center registry, Hospital Epidemiology database, electronic medical records, and billing data were linked. Adult patients (≥18 years old) admitted between January 1, 2004 and December 31, 2013 with a burn injury were included. Only the first hospitalization within this time frame was included. Diabetes mellitus was identified using both comorbidities listed on the burn registry and d diagnostic codes attached to the inpatient hospitalization (ICD-9-CM 250). Multivariable Cox proportional hazard models were used to estimate the increased risk of diabetes on 60-day mortality and hospital-acquired infections, after adjusting for patient and burn characteristics. Only patients hospitalized for 2 or more days (i.e. at risk for infection as per CDC definitions) were included in HAI analyses.
Results: 5,539 patients met the inclusion criteria. 665 (11.8%) had a diabetes mellitus (DM) diagnosis. Diabetic patients were significantly more likely to be female (34.1% vs. 26.6%, p<0.0001), African American (36.9% vs. 26.0%, p<0.0001), and older (median age 56.7 years old vs. 39.9 years old, p<0.0001). Diabetic patients were more likely to have contact burns (8.9% vs. 4.7%, p=<0.0001) and inhalational injury (11.0% vs. 8.1%, p=0.01). No differences were seen in median burn size (4.0% vs. 4.0%, p=0.44). The median revised Baux scores was also higher among diabetics (64.0 vs. 47.6), p<0.0001. Patients with DM were more likely to be admitted to the ICU (14.3% vs. 10.6%, p<0.0001) and were hospitalized for a median 11 days (interquartile range [IQR] 4 – 26), compared to a median 7 days (IQR 2 – 13) for patients without diabetes, p<0.0001. Only 242 patients (4.4%) were hospitalized longer than 60 days and administratively censored prior to discharge or death. Overall, 243 (4.4%) died during their inpatient hospitalization. After 30 days, diabetic patients had a higher mortality risk (RD 0.03, 95% CI 0.00, 0.05) compared to non-diabetic patients, and after 60 days the risk was higher (RD 0.07, 95% CI 0.01, 0.12). Patients with DM were significantly more likely to have an HAI after 60 days.
Conclusion: Comorbid conditions can lead to worse outcomes. Diagnosed diabetics fare worse after burn injury than matched non-diabetics.