Dengue provides a diagnostic challenge to health care providers, especially in ensuring rapid diagnosis to reduce the morbidity. One of the most important parameters depicting dengue severity is evidence of plasma leakage; hence prediction of plasma leakage in dengue is vital. Bedside ultrasonography is an attractive tool that can be used to detect evidence of plasma leakage, detecting free fluid in potential space and evidence of gall bladder wall thickening, hence providing another armamentarium towards establishing dengue diagnosis and severity.
The objective of the study was to address the accuracy of diagnosis and dengue severity categorization using ultrasonography without depending on laboratory index.
A prospective study carried out from September 2015 to March 2016. All adult patients between the age of 18- 65years old with history of fever and positive NS1 or serology IG G or IGM was recruited in the study. All patients underwent protocolized bedside ultrasonography conducted by certified providers which include sonographic assessment of gall bladder wall, pleural cavity, peritoneal cavity and pericardial space. Bedside ultrasound was done upon patient arrival, and the assessment is repeated after 60 minutes if the first assessment was negative. The sensitivity (Sn), specificity (Sp), positive predictive value (PPV) and negative predictive value (NPV) of ultrasonography in detecting plasma leakage was analyzed. Result was tabulated in SPSS VERSION 24
A total of 364 patients recruited and 228 had positive ultrasonography findings. 172 out of 228 (75.4%) was subsequently diagnosed as severe dengue with findings such as ascites/ pleural effusion and/or gallbladder (GB) wall thickening. 56 out of 228 (24.6%) who was initially diagnosed as non severe dengue were found to have positive ultrasonography findings, 28 (50%) them progressed to severe dengue. Patient whom came with dengue fever with warning sign 122 out of 228 (53%) has positive ultrasonography findings, 94 (77.0%) patients progressed into severe dengue. Fifty severe dengue patients whom came with no clinical sign or symptom suggestive of plasma leakage found to have positive ultrasound findings. Sensitivity and specificity gallbladder wall edema was more pronounced in severe than in nonsevere dengue patients and often preceded ascites/pleural effusion as the p value showed >0.001. The negative predictive value (NPV) of plasma leakage at Morrison’s pouch his 85.5%, NPV at splenorenal is 84.6% and NPV at rectovesicle is 82.4%. As for sensitivity (Sn) and specificity (Sp), fluid collection at retrovesicle pouch has a Sp of 89% and Sn 4.9%. This showed high specificity but very poor sensitivity. Meanwhile GB thickening showed specificity of 74.9% and NPV 86.3% which is highest among all other plasma leakage site. Pleural effusion also showed high specificity 79.8% and NPV 84%, but pericardial effusion showed high sensitivity 96.6% and low specificity with 83.7% negative predictive value. Gall bladder thickening noted more pronounce on the day 3-6 of illness in 93 patients, the diameter varied from 4.00mm-4.95mm. The GB wall diameter is >5.0mm in patient whom presented with combination (multiple sites) of plasma leakage such as gall bladder thickening and/or ascites and/or pleural effusion). When combining bedside ultrasonography with lactate showed to have p value of 0.00. Gall bladder thickening with raised haematocrit have p = 0.001.
Ultrasound is reliable prognostic and diagnostic tools in identifying plasma leakage, hence predicting the severity of diseases and progression of the diseases. Its also shown that when bedside ultrasound findings is combined with POCT result, the sensitivity and predictive values of diagnosing and classifying dengue severity is further enhanced. This is in contrast to using existing laboratory markers in isolation. Henceforth, such patients with evidence of plasma leakage should be managed as severe dengue and merits for intensive care monitoring to minimize the complication.