Contact isolation in trauma patients: an analysis of infectious complications
Author(s):
Mark Hamill, Virginia Tech Carilion School of Medicine; Christopher Reed, Virginia Tech Carilion School of Medicine; Sandy Fogel, Virginia Tech Carilion School of Medicine; Eric Bradburn, Virginia Tech Carilion School of Medicine; Kinga Powers, Virginia Tech Carilion School of Medicine; Katie Love, Virginia Tech Carilion School of Medicine; Christopher Baker, Virginia Tech Carilion School of Medicine; Bryan Collier, Virginia Tech Carilion School of Medicine
Background: Recent studies have demonstrated that contact isolation (CI) may have a significant impact on patient outcomes. We sought to characterize the relationship between CI, victims of trauma and infectious sequelae at our Level I trauma center.
Hypothesis: The use of CI in the trauma population is associated with increases in pneumonia (PNA) and urinary tract infection (UTI) rates.
Methods: An existing database of all trauma patients admitted to our Level I trauma center during 2011 and 2012 was queried. Data including demographics, injury severity score (ISS), comorbidities, CI status, and infections were collected. Diagnosis of PNA or UTI was based on clinical documentation in the medical record. A chart review revealed the reason for CI, including specific organisms. Infections were excluded if the CI was the direct result of a new infection. Due to differences in demographics between groups, logistic regression was then performed.
Results: A total of 4,423 patients were admitted over this period. Of these, 4,317 (97.6%) had complete records and were included in the subsequent analysis. CI was in place in 251 (5.8%) patients; 4,066 (94.2%) were not isolated. 176 (70.1%) had CI initiated for MRSA nasal colonization. 22 (8.8%) had no reason for CI documented. PNA occurred in 193 (4.5%); 58 (23.1%) in the CI group vs 135 (3.3%) in the non-CI group. UTI was diagnosed in 167 (3.9%); 48 (19.1%) in the CI group vs 119 (2.9%) in the non-CI group. Using logistic regression, CI, ISS and gender were significantly associated with PNA and UTI. COPD, alcohol abuse, and obesity were also associated with PNA. Age was significantly associated with UTI.
Demographics
|
|
CI |
Non-CI |
p-value |
Numbers |
251 (5.8%) |
4066 (94.2%) |
<0.0001 |
Age |
60 (32) |
45 (32) |
<0.0001 |
Male Gender |
62.2 |
65.5 |
0.2761 |
ISS |
17 (13) |
9 (13) |
<0.0001 |
PNA |
58 (23.1%) |
135 (3.3%) |
<0.0001 |
UTI |
48 (19.1%) |
119 (2,9%) |
<0.0001 |
Logistic Regression Data
|
|
Pneumonia
|
UTI
|
|
Odds Ratio |
p-value |
Odds Ratio |
p-value |
Age |
NS |
NS |
0.98 |
<0.0001 |
Male Gender |
2.10 |
0.0002 |
0.45 |
<0.0001 |
ISS |
0.92 |
<0.0001 |
0.95 |
<0.0001 |
Hx Alcohol Abuse |
1.78 |
0.0195 |
NS |
NS |
COPD |
2.28 |
0.0056 |
NS |
NS |
Obesity |
2.05 |
0.0056 |
NS |
NS |
CI |
3.70 |
<0.0001 |
4.20 |
<0.0001 |
Conclusions: PNA and UTI risks in trauma patients were significantly associated with the use of CI. As the majority of these patients had CI precautions for asymptomatic colonization, the CI provided them no direct benefit. As CI is associated with multiple negative outcomes, its use in the trauma population needs to be carefully reevaluated.