[Utility of the medial region of pro-adrenomodulin for the detection of true bacteremia in elderly patients treated in the emergency department for suspected infection]

OBJECTIVE

The prediction of bacteremia in the emergency department (ER) is important for initial decision-making. The elderly population is a diagnosis challenge. The objective was to evaluate the accuracy of mid regional pro-adrenomedullin (MR-proADM) to identify true bacteremia (BV) in elderly patients attended in 3 hospital emergency departments.

METHODS

Observational study including patients ≥75 years of age or older attended in the ER for suspected infection in whom a blood culture (BC) was extracted. Sociodemographic, comorbidity, hemodynamic and analytical variables, biomarkers [MR-proADM, procalcitonin (PCT), C-reactive protein (CRP) and lactate] and final diagnosis were collected. The primary outcome was a true positive on a blood culture.

RESULTS

A total of 109 patients with a mean age of 83 (SD: 5.5) years were included. A final diagnosis of BV was obtained in 22 patients (20.2%). The independent variables to predict it were PCT (OR: 13.9; CI95%: 2.702-71.703; p=0.002), MR-proADM (OR: 4.081; CI95%: 1.026-16.225; p=0.046) and temperature (OR: 2.171; CI95%: 1.109-4.248; p=0.024). Considering the cut-off point for MR-proADM (2.13 mg/dl), a sensitivity (Se) of 73%, specificity (E) of 71%, a positive predictive value (PPV) of 39%, a negative predictive value (NPV) of 91%, a positive likelihood ratio (LHR+) of 2.53 and a negative likelihood ratio (LHR-) of 0.38; for PCT (0.76 mg/dl) a Se of 90%, E of 65%, PPV of 40%, NPV of 96%, LHR+ 2,64 and a LHR- of 0.14 were obtained. When combining both, a Se of 69%, E of 84%, PPV of 52%, NPV of 91%, LHR+ of 4.24 and LHR- of 0.38 were observed.

CONCLUSIONS

Elevated levels of PCT and MR-proADM were independently associated with an increased risk of BV and the combination of both improves the accuracy to identify these patients.

[Current situation of sepsis care in Spanish emergency departments]

OBJECTIVE

To describe the approach to the patients with suspected sepsis in the Spanish emergency department hospitals (ED) and analyze whether there are differences according to the size of the hospital and the number of visits to the emergency room.

METHODS

Structured survey of those responsible for the 282 public EDs that serve adults 24 hours a day, 365 days a year. It was asked about assistance and management in the emergency room in the care of patients with suspected sepsis. The results are compared according to hospital size (large ≥ 500 beds vs medium-small <500) and influx to the emergency room (discharge ≥ 200 visits / day vs medium-low <200).

RESULTS

A total of 250 Spanish EDs responded (89%). Sepsis protocols are available in 163 (65%) EDs median weekly sepsis treated ranged from 0-5 per week in 39 (71%) ED, 6-10 per week in 10 (18%), 11-15 per week in 4 (7%), and more than 15 activations per week in 3 centers (3.6%). The criteria used for sepsis diagnosis were the qSOFA/SOFA in 105 (63.6%) of the hospitals, SIRS in 6 (3.6%), while in 49 (29.7%) they used both criteria simultaneously. In 79 centers, the sepsis diagnosis was computerized, and in 56 there were tools to help decision-making. 48% (79 of 163) of the EDs had data on bundles compliance. In 61% (99 of 163) of EDs there was training in sepsis and in 56% (55 of 99) it was periodic. Considering the size of the hospital, large hospitals participated more frequently as recipients of patients with sepsis and had an infectious, sepsis and short-stay unit, a microbiologist and infectious disease specialist on duty.

CONCLUSIONS

Most EDs have sepsis protocols, but there is room for improvement. The computerization and development of alerts for diagnosis and treatment still have a long way to go in EDs.