Background
Methods
Study design and setting
Selection of participants
Definition of outcomes
Sepsis
No sepsis
Sepsis screening models
NEWS2 (20) | RETTS (21) | Predict Sepsis screening tool 1 (27) | Predict Sepsis screening tool 2 (27) | Predict Sepsis screening tool 3 (27) | |
---|---|---|---|---|---|
Number of included variables | 6 | 9 variables in VS part described below + ESS* | 6 | 4 | 6 |
Score considered positive for suspected sepsis | ≥ 5 is recom-mended (20) but some sepsis alerts (35) suggest ≥ 7 | RETTS red is used in some sepsis alerts (36) | ≥ 2 | ≥ 2 | ≥ 2 |
Included variables with weights | |||||
Respiratory rate | ≤ 8 = 3 9–11 = 1 12–20 = 0 21–24 = 2 ≥ 25 = 3 | > 30 or < 8 = red 26–30 = orange 8–25 = green | X | X | > 24 = 1 |
Oxygen saturation | ≤ 91 = 3 92–93 = 2 94–95 = 1 ≥ 96 = 0 | < 90 with O2 = red < 90 without O2 = orange 90–95 without O2 = yellow > 95 without O2 = green | X | X | < 94 = 1 |
Systolic blood pressure | ≤ 90 = 3 91–100 = 2 101–110 = 1 111–219 = 0 ≥ 220 = 3 | < 90 = red | ≤ 100 = 2 | ≤ 100 = 2 | ≤ 100 = 2 |
Heart rate | ≤ 40 = 3 41–50 = 1 51–90 = 0 91–110 = 1 111–130 = 2 ≥ 131 = 3 | Regular > 130 or irregular > 150 = red > 120 or < 40 = orange > 110 or < 50 = yellow 50–110 = green | X | X | > 110 = 1 |
Consciousness | (AVPU in parenthesis) GCS ≤ 14 (C, V, P, or U) = 3 GCS ≥ 15 (A) = 0 | (RLS in parenthesis) GCS < 8 (unconscious) = red GCS 8–12 (2–3) = orange GCS 13–14 (acutely disoriented) = yellow GCS 15 (alert) = green | GCS < 15 = 2 History of acute altered mental status = 1 | History of acute altered mental status and/or GCS < 15 = 1 | GCS < 15 = 2 |
Temperature, degrees C | ≤ 35,0 = 3 35,1–36,0 = 1 36,1–38,0 = 0 38,1–39,0 = 1 ≥ 39,1 = 2 | > 41 or < 35 = orange 38.6–41 = yellow 35–38.5 = green | 38.1–38.5 = 1 > 38.5 = 2 | 38.1–38.5 = 1 > 38.5 = 2 | 38.1–38.5 = 1 > 38.5 = 2 |
Other variables | X | Obstructed airway = red Stridor = red Ongoing seizures = red | Gastrointestinal symptoms = 1 P-Lactate > 4.0 = 2 | Gastrointestinal symptoms = 1 | X |
Measurements; data collection and handling
Statistical analyses
Ethical approval
Results
Characteristics
Variable | All patients | Patients with outcome sepsis | ||
---|---|---|---|---|
Number* (%*) | Median (IQR) | Number* (%*) | Median (IQR) | |
Age (yr) | 78 (72–85) | 78 (70–84) | ||
< 65 years | 36/323 (11.1) | 18/144 (12.5) | ||
65–74 years | 78/323 (24.1) | 39/144 (27.1) | ||
≥ 75 years | 209/323 (64.7) | 87/144 (60.4) | ||
Gender | ||||
Male | 189/323 (58.5) | 90/144 (62.5) | ||
Ambulance priority | ||||
1 | 62/320 (19.4) | 48/143 (33.6) | ||
2 | 222/320 (69.4) | 88/143 (61.5) | ||
3 | 36/320 (11.3) | 7/143 (4.9) | ||
Admitted to in-hospital care | 273/323 (84.5) | 139/144 (96.5) | ||
Outcome | ||||
Sepsis | 144/323 (44.6) | |||
No sepsis | 179/323 (55.4) | |||
Septic shocka | 17/138 (12.3) | |||
ICD-code upon hospital discharge | ||||
ICD-code sepsis | 29/321 (9.0) | 24/142 (16.9) | ||
ICD-code infection | 211/321 (65.7) | 118/142 (83.1) | ||
In-hospital mortality | 21/323 (6.5) | 15/144 (10.4) |
Performance of the screening models
NEWS2 (20) ≥ 5 | NEWS2 ≥ 7 | RETTS (21) ≥ orange | RETTS red | Clinical judgment | Predict Sepsis screening tool 1 (27) | Predict Sepsis screening tool 2 (27) | Predict Sepsis screening tool 3 (27) | |
---|---|---|---|---|---|---|---|---|
Total score considered positive for predicted sepsis | ≥ 5 | ≥ 7 | RETTS ≥ orange | RETTS red | Clinical judgment sepsis by ambulance personnel | ≥ 2 | ≥ 2 | ≥ 2 |
Sensitivitya (95%CI) | 0.74 (0.65–0.80) | 0.58 (0.50–0.66) | 0.83 (0.75–0.88) | 0.23 (0.17–0.31) | 0.40 (0.32–0.49) | 0.90 (0.84–0.94) | 0.87 (0.80–0.92) | 0.91 (0.85–0.95) |
Specificitya (95%CI) | 0.61 (0.53–0.68) | 0.78 (0.71–0.84) | 0.45 (0.38–0.53) | 0.92 (0.86–0.95) | 0.74 (0.67–0.80) | 0.42 (0.35–0.50) | 0.49 (0.41–0.56) | 0.39 (0.32–0.47) |
PPVa (95%CI) | 0.60 (0.53–0.67) | 0.68 (0.59–0.76) | 0.55 (0.48–0.62) | 0.69 (0.54–0.81) | 0.55 (0.45–0.65) | 0.56 (0.49–0.62) | 0.58 (0.51–0.64) | 0.55 (0.48–0.61) |
NPVa (95%CI) | 0.74 (0.66–0.81) | 0.70 (0.63–0.76) | 0.76 (0.67–0.84) | 0.60 (0.54–0.65) | 0.61 (0.54–0.67) | 0.84 (0.75–0.91) | 0.82 (0.73–0.89) | 0.84 (0.74–0.91) |
Pos LRa (95%CI) | 1.88 (1.53–2.32) | 2.68 (1.96–3.65) | 1.51 (1.30–1.76) | 2.73 (1.55–4.83) | 1.53 (1.12–2.10) | 1.55 (1.36–1.78) | 1.69 (1.44–1.97) | 1.49 (1.31–1.70) |
Neg LRa (95%CI) | 0.43 (0.33–0.57) | 0.53 (0.44–0.65) | 0.38 (0.27–0.56) | 0.84 (0.77–0.92) | 0.81 (0.71–0.93) | 0.23 (0.14–0.39) | 0.27 (0.18–0.42) | 0.23 (0.14–0.39) |
AUCa for the model without cut-off-based on sum of scores (95%CI) | 0.73 (0.68–0.79) | 0.73 (0.68–0.79) | not possible to calculate | not possible to calculate | not possible to calculate | 0.77 (0.72–0.82) | 0.75 (0.70–0.80) | 0.77 (0.71–0.82) |
AUCa for the model with specific cut–off | 0.67 (0.61–0.73) | 0.68 (0.62–0.74) | 0.64 (0.58–0.70) | 0.57 (0.51–0.64) | 0.57 (0.51–0.63) | 0.66 (0.60–0.72) | 0.68 (0.62–0.74) | 0.65 (0.59–0.71) |