REASSESSING THE FLUID REQUIREMENT IN PATIENTS WITH SEPSIS

Background: It has long been recognized that a comprehensive bundled and coordinated care affects sepsis mortality.   The Center for Medicare and Medicaid Service’s (CMS) has mandated reporting of severe sepsis/septic shock management with emphasis on adherence to the bundles.  However, bundles may not be appropriate for all patients.  The current 3-hr bundle includes blood cultures, broad-spectrum antibiotics, lactate measurement and 30ml/kg rapid infusion in case of SBP<90 mmHg and/or lactate>4. Rapid fluid infusion is the most difficult measure to meet as it may be contra-indicated for certain groups of patients, including those with chronic renal failure (CRF) or chronic heart failure (CHF).  We aimed to compare the characteristics of septic patients in whom the fluid requirement was met and those who did not.

Methods: We used retrospective data from two hospitals, an 1171-bed academic center and a 235-bed community hospital, with a hospital-wide EMR-embedded sepsis screening program and a rigorous clinical adjudication process for cases that screened positive. Clinically adjudicated severe sepsis (or sepsis with SBP<90 and/or lactate >4) or septic shock cases from July 2015 to June 2016 were compiled and analyzed. Cases that met the fluid requirement were compared to those who did not based on CRF and CHF, time from admission to sepsis diagnosis (T0), length of hospital stay and mortality.

Results: Among 1177 cases of severe sepsis or septic shock, 47.2% were female and 52.8% male, 35% white and 21.4% black. Overall, 23.9% of patients met the 3-hr 30ml/kg fluid requirement.  Patients who met the fluid requirement (versus those who did not meet the requirement) were less likely to have CRF (20% vs 36%), CHF (13.9% vs 31.7%) with their T0 closer to admission (65hrs vs 127hrs). In addition, they were more likely to be admitted to the ICU after identification (44.5% vs 33.8%) but had shorter length of stay (15days vs 19.3days, p-value=0.0018) and lower in-hospital mortality rate (21.4% vs 31.1%). Table1.

Among patients with either CRF or CHF who received fluid (16%) (versus those who did not receive fluid), they had similar LOS (19.5 days vs 21.6 days, p-value 0.42) and a lower in-hospital mortality rate (27.2% vs 37.4%).

Conclusions: These preliminary data suggest that among sepsis patients who did not meet fluid criteria, there was a higher prevalence of CRF and CHF, conditions which could explain the lower adherence.  Moreover, we found that the T0 in the non-adherent group was twice as much as in the adherent group, further questioning the characteristics of these patients and the reasons for non-compliance which may be justified.  Nonetheless, among those with CRF or CHF who received the fluid requirement, they had lower in-hospital mortality, which may suggest a potential benefit of fluid even in this group, specifically in the subgroup with shorter T0.

In the current climate of hospital sepsis reporting and potential negative consequences for hospitals, it is crucial to understand statistics that may suggest suboptimal quality of sepsis management.

Table 1