Sonographic plasma leakage marks the dengue patient who will get worse

Design

systematic review with random-effects meta-analysis where data allowed, QUADAS-2 assessed

Population

6,979 patients across 56 publications; 6,189 with dengue, remainder Crimean-Congo, hantavirus, yellow fever and Lassa

Primary outcome

ultrasound findings and their ability to predict severe disease

Effect

pericardial effusion and several other plasma-leakage findings associated with severe disease; high heterogeneity, no pooled threshold

This systematic review screened 457 publications and included 56 comprising 6,979 patients with viral infections associated with haemorrhagic fevers — 44 publications and 6,189 patients on dengue, with smaller sets on Crimean-Congo haemorrhagic fever, hantavirus, yellow fever and Lassa fever. Patients were mostly recruited while acutely unwell at tertiary centres during periods of high transmission.

The common pathophysiology is plasma leakage, and it produces a recognisable set of sonographic findings: gallbladder wall thickening, ascites, pleural effusions and abnormal lung ultrasound, pericardial effusions, abdominal lymphadenopathy, splenomegaly, abnormal pancreatic appearances, and hyperechoic kidneys or liver. Pericardial effusion and several other findings were associated with severe disease. The authors propose incorporating ultrasound into severity definitions and set out a framework they call Ultrasound-based assessment of Plasma leakage Severity, explicitly as hypothesis-generating.

For an emergency department in an Indian dengue season this is the most directly usable item on the desk, because the equipment is already at the bedside and the assessment is being made anyway — usually on haematocrit trend, platelet count and a warning-sign checklist. What the review supports is scanning deliberately for leakage rather than waiting for it to declare itself haemodynamically, in a disease where the deterioration is fast and the window for fluid management is narrow. What it does not yet support is a scored threshold: heterogeneity was high, the recruitment was hospital-based and skewed to severe disease, and no prospective validation exists.

• Scan deliberately for ascites, pleural and pericardial effusion and gallbladder wall thickening in suspected severe dengue
• Record the findings and repeat the scan rather than reading one image in isolation
• Keep using established warning signs and haematocrit trend — ultrasound supplements them
• Do not use a sonographic score to discharge a patient; none has been validated

Why it matters

The machine is already at the bedside and the deterioration it might anticipate is the one that kills.

Don't overread it

This is hypothesis-generating: no sonographic severity score has been prospectively validated, and the cohorts were hospital-based.

The statistics, in plain English

The review reports crude proportions and pooled estimates where data allowed, with high heterogeneity throughout — meaning the studies disagree enough that a single sensitivity or specificity figure would be misleading. Recruitment at tertiary centres during peak transmission inflates the proportion with severe disease, so predictive values from these data will not transfer to a general emergency department. Bias in the timing domain matters especially in dengue, where findings depend heavily on which day of illness the scan is done.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.