Positive predictive value of a sepsis-screening protocol for patients with tetraplegia.

Objective: To determine the positive predictive value (PPV) of a sepsis-screening protocol in patients with cervical spinal cord injury (SCI).

Design/method: Retrospective review of all patients with cervical SCI who screened positive for two or more systemic inflammatory response syndrome (SIRS) criteria while hospitalized in acute care or inpatient rehabilitation units over a 3.5-year period. Sepsis was defined by the occurrence of (1) any culture order followed by an intravenous (IV) antibiotic within 72 hours or (2) an IV antimicrobial followed by a culture order within 24 hours.

Results: A total of 134 patients screened positive for two or more SIRS criteria. Of these, 36 patients (26.9%) were diagnosed with sepsis. Factors associated with a true-positive SIRS screen on multivariable analysis included American Spinal Injury Association Impairment Scale (AIS) grade A-C (vs. D; p < .001). The PPV of the screen was 38% in patients with AIS A-C and 9% in patients with AIS D. Altered mental status (AMS) was strongly associated with a diagnosis of sepsis; 16 of 18 (88.9%) of those with AMS had sepsis (p < .001). Age, sex, and neurologic level of injury were not associated with true-positive screening. For patients with new SCI, the first true-positive screen occurred a median of 31 days post-injury. The most common SIRS criteria combinations in patients with true-positive screens were elevated heart rate and either abnormal white blood cell count (43% of true positives) or abnormal temperature (26% of true positives). Abnormally low body temperature (<36°C) contributed to false-positive screening for 10 of 38 (26%) AIS D patients who screened positive.

Conclusion: Sepsis screening using SIRS criteria in hospitalized patients with tetraplegia has a PPV of 26.9%; it is significantly higher in patients with AIS A-C versus D injuries. AMS, when combined with a positive SIRS screening, is strongly associated with sepsis.