Advice

Clinical and technical evidence

Clinical and technical evidence

A literature search was carried out for this briefing in accordance with the interim process and methods statement. This briefing includes the most relevant or best available published evidence relating to the clinical effectiveness of the technology. Further information about how the evidence for this briefing was selected is available on request by contacting mibs@nice.org.uk.

Published evidence

There are 5 studies summarised in this briefing, including 3,217 patients.

Table 1 summarises the clinical evidence as well as its strengths and limitations.

Overall assessment of the evidence

The studies show that the MR-proADM test is effective at predicting outcomes for infection, when used with clinical scoring systems. The studies also show that the test could triage patients and predict the type of treatment they will need.

There are no studies that show the MR‑proADM test used alongside the National Early Warning Score 2 (NEWS2). Although the studies are recent, they were done before NEWS2 was published. Three studies in the table used the test alongside the earlier NEWS. NEWS differs from NEWS2 because it does not consider oxygen saturation or confusion, and has less advice on serious sepsis and infection.

Further evidence showing how MR-proADM can improve clinical outcomes when used with NEWS/NEWS2 would provide evidence to support clinical adoption. This could be used to evaluate the resource impact of using MR‑proADM.

One study was done in the NHS and another included an NHS centre. The studies done outside the UK using different clinical scoring systems may be less relevant to the NHS.

Table 1 Summary of selected studies

Graziadio et al. 2019

Study size, design and location

300 people (with NEWS 2 to 5) in a prospective observational study.

Location: UK.

Intervention and comparator(s)

MR-proADM.

NEWS.

Key outcomes

NEWS and MR-proADM together predicted an increase of at least 2 in the NEWS (acuity increase) more accurately than NEWS alone. This increased the AUC to 0.61 (95% CI 0.54 to 0.69) from 0.55 (95% CI 0.48 to 0.62). When the confounding effects of chronic obstructive pulmonary disease or heart failure and interaction with MR‑proADM were included, the prognostic accuracy further increased the area under the curve to 0.69 (95% CI 0.63 to 0.76).

Strengths and limitations

This study was done in an NHS setting. MR-proADM was used with NEWS (the standard scoring system in the NHS). This study was done in 2015 before the development of NEWS2.

The authors state that the prognostic accuracy of MR-proADM might have been higher if more severely ill patients had been included. However, the aim of the study was to evaluate the usefulness of the test in less ill patients.

The study was funded by the company.

Eckart et al. 2019

Study size, design and location

1,303 adults seeking emergency department care in a secondary analysis of a multinational, observational study.

Location: France, Switzerland and USA.

Intervention and comparator(s)

MR‑proADM, white cell count and procalcitonin.

NEWS.

Key outcomes

The NEWS alone showed prognostic accuracy for 30-day mortality (AUC 0.73), with a multivariate adjusted odds ratio of 1.26 (95% CI 1.13 to 1.40, p<0.001). The AUCs for the prediction of mortality using MR-proADM, white cell count and procalcitonin were 0.78, 0.64 and 0.71, respectively.

Combining NEWS with all 3 blood markers or only with MR-proADM improved discrimination with an AUC of 0.82 (p=0.002). Combining the 3 inflammatory markers with NEWS improved prediction of intensive care admission compared with NEWS alone (AUC 0.70 versus 0.65, p=0.006).

Strengths and limitations

This study is a secondary analysis of an observational study, so the results of MR-proADM and other biomarkers were not used in clinical decision making. NEWS was calculated retrospectively and so the mortality rates may not reflect clinical practice when NEWS informs clinical decisions.

The authors excluded patients who did not have complete data, which may have led to a selection bias.

The study was funded by the company.

Saeed et al. (2019)

Study size, design and location

1,175 people presenting to the emergency department with suspected infection in an observational cohort study.

Location: UK, France, Italy, Sweden, Spain.

Intervention and comparator(s)

MR-proADM.

Also used: biomarkers (procalcitonin, lactate and C‑reactive protein) and clinical scores (SOFA, qSOFA, NEWS hazard ratio).

Key outcomes

MR-proADM was associated with predicting 28-day mortality with significantly more accuracy compared with other biomarkers and scores, in a cohort of patients in the emergency unit (n=1,175, sensitivity 92%, 95% CI 80% to 97%; specificity 75%, 95% CI 72% to 78%). Patient subgroups with high MR‑proADM concentrations (1.54 nmol/litre or higher) and low biomarker (procalcitonin below 0.25 ng/mlitre, lactate below 2.0 mmol/litre or C‑reactive protein below 67 mg/litre) or clinical score (NEWS below 4 or CRB-65 score below 2) values were characterised by a significantly longer length of hospitalisation (p<0.001), rate of intensive care unit admission (p<0.001), elevated mortality risk (for example, NEWS hazard ratio, 32.6, 95%CI 9.4 to 113.6) and a greater number of disease progression events (p<0.001), compared with similar subgroups with low MR-proADM concentrations (below 1.54 nmol/litre).

Strengths and limitations

There were 2 patient cohorts considered, 1 consisted of prospective patients enrolled after presenting to the emergency department in 5 countries. This was compared with a subgroup of a cohort from a previous study done in the Netherlands. This second cohort was used to validate the results measured prospectively. This method of comparison is subject to bias because the 2 cohorts of data were collected under different conditions.

The authors published further analysis in a letter to the journal indicating that delayed antibiotics for patients with low MR-proADM concentrations might result in fewer adverse effects, potentially allowing for a more detailed clinical assessment before any further treatment. Further studies in larger patient populations are needed to confirm these findings.

The study was funded by the company.

Bellia et al. (2018)

Study size, design and location

126 people in a consecutive case series.

Location: Italy.

Intervention and comparator(s)

MR-proADM (measured on admission, T0 and after 24 hours, T24).

Also used: APACHE II and SAPS II.

Key outcomes

Multivariate analysis showed that T0 MR-proADM was a significant predictor of mortality (odds ratio: 1.27; 95% CI 1.03 to 1.55; p=0.022). Receiver operating characteristic curves analysis showed that MR-proADM on admission identified non-survivors with high accuracy and was less accurate than APACHE II and SAPS II scores (AUC 0.71; 95% CI 0.62 to 0.78; p=0.0002 for MR-proADM; AUC 0.71; 95% CI: 0.62 to 0.79; p<0.0001 for APACHE II; AUC 0.8; 95% CI 0.71 to 0.87; p<0.0001 for SAPS II).

Strengths and limitations

This study uses the technology alongside APACHE II and SAPS II clinical scoring systems, which are not standard of care in the NHS.

Stalenhoef et al. (2018)

Study size, design and location

313 people with suspected febrile urinary tract infections in a consecutive case series.

Location: Netherlands.

Intervention and comparator(s)

MR-proADM.

Also used: procalcitonin, C-reactive protein, and a clinical score (PRACTICE).

Key outcomes

MR-proADM had the highest diagnostic accuracy for predicting a complicated febrile urinary tract infection (AUROC [95% CI] 0.86 [0.79 to 0.92]), followed by procalcitonin (AUROC [95% CI] 0.69 [0.58 to 0.80]). MR-proADM concentrations were unique in being significantly elevated in patients directly admitted and in outpatients who then needed hospitalisation, compared with those completing treatment at home. A virtual triage algorithm with an MR-proADM cut-off of 0.80 nmol/litre resulted in a hospitalisation rate of 66%, with only 2% secondary admissions.

Strengths and limitations

This study only considered MR-proADM in people with suspected febrile urinary tract infections. This study uses the PRACTICE clinical scoring system and other biomarkers, which may not be standard of care in the NHS.

One of the authors is employed by the company.

Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation; AUC, area under the curve; AUROC, area under receiver operating characteristic curve; CI, confidence interval; qSOFA, quick sequential organ failure assessment; NEWS, National Early Warning Score; SAPS, Simplified Acute Physiology Score; SOFA, sequential organ failure assessment.

Recent and ongoing studies