A systematic review of ultrasound to predict severe disease in patients with dengue and other viral infections associated with hemorrhagic fevers

Abstract

Background: Viral infections associated with hemorrhagic fevers (VaHFs) cause millions of cases each year across a broad geography with severity ranging from asymptomatic cases to clinical disease with a high case fatality rate. Commonalities in pathophysiology across VaHFs include vascular damage and plasma leakage with the possibility of hemorrhage, organ failure, shock, and death. These changes are seen on autopsy, and evidence of plasma leakage can be readily detected on ultrasound. Methods: We conducted a systematic review to assess ultrasound findings in VaHFs and their ability to predict severe disease (PROSPERO CRD42024620862). PubMed was searched through May 2025. Any articles presenting original, patient-level findings on ultrasound in VaHFs with serologic and/or molecular confirmation were included. Case series and duplicative datasets were excluded. A QUADAS-2 tool was used to assess bias. Study level characteristics and patient-level data on the ultrasound findings were extracted. The crude proportions of ultrasound findings were reported, and weighted meta-analysis using a random-effects model was performed and displayed in forest plots when the data allowed. Sensitivity, specificity, and positive/negative likelihood ratios for severe disease were calculated and visualized using a random effects model in SROC plots when the data allowed. Results: In total, 457 unique publications were screened for inclusion, and 56 publications comprising 6979 participants were included in the review, including 44 publications (6189 participants) on dengue, six (494 participants) on Crimean-Congo hemorrhagic fever, three (134 participants) on hantavirus disease, two (116 participants) on yellow fever, and one (46 participants) on Lassa fever. Participants were most often recruited while acutely sick during periods of high transmission at tertiary care centers. Ultrasound findings included gallbladder wall thickening, ascites, pleural effusions and abnormal lung ultrasound, pericardial effusions, abdominal lymphadenopathy, splenomegaly, abnormal pancreatic ultrasound, and hyperechoic kidneys or liver. Pericardial effusions and several other findings were associated with severe disease. Conclusions: Common pathophysiology leading to plasma leakage results in sonographic signs across viral infections associated with hemorrhagic fevers, and some findings can help predict severe disease. Important limitations included reliance on a single database and possibility of missing manuscripts from local journals, the small number of publications on non-dengue VaHFs, patient recruitment overwhelmingly from the in-hospital setting at tertiary referral centers during high-transmission periods, high degree of heterogeneity, and bias in the timing domain among the studies on dengue. We first propose ultrasound findings be incorporated into new definitions of severity for VaHFs and then introduce an Ultrasound-based assessment of Plasma leakage Severity (UPS) as a hypothesis generating framework to help guide future, prospective research. We discuss some virus and organ-specific considerations and conclude with detailed considerations for future research. Author summary: Viral infections associated with hemorrhagic fevers (VaHFs) may lead to vascular damage and plasma leakage with the possibility of hemorrhage, organ failure, shock, and death. Plasma leakage can be readily detected on ultrasound, and in this review, the authors assess the frequency of gallbladder wall thickening, ascites, pleural effusions and abnormal lung ultrasound, pericardial effusions, abdominal lymphadenopathy, splenomegaly, abnormal pancreatic ultrasound, and hyperechoic kidneys or liver in VaHFs, as well as the ability of these findings to predict severe disease. The review was dominated by studies on dengue, and there was great heterogeneity in the frequency of ultrasound findings in part related to variability in the era the study was completed, the age of participants, and the location of the original study. Several findings were associated with severe disease, and the authors raise many important considerations for future research. The authors recommend ultrasound findings be incorporated into definitions of severity for VaHFs and then propose an ultrasound-based grading scale for plasma leakage severity (UPS) as a hypothesis generating framework to help spark future research.

Original publication: PLOS Neglected Tropical Diseases (2026-09-17). Source. Source DOI: 10.1371/journal.pntd.0014609.