QScience Highlights
Diagnosis

Simplifying the diagnosis of acute fever

Platelet levels in patients with acute fevers in tropical countries could help guide diagnosis and treatment. 

Published online 30 July 2016

Evaluating platelet levels in patients with acute fevers in tropical countries could help resource-poor clinics get a better picture of how to diagnose and treat them, when used together with other clinical signs and symptoms.

Evaluating platelet levels in patients with acute fevers in tropical countries could help resource-poor clinics get a better picture of how to diagnose and treat them, when used together with other clinical signs and symptoms.

© BSIP SA / Alamy Stock Photo

Researchers reviewed clinical and laboratory data of 402 patients with acute fever lasting 24 hours to three weeks upon admission to a hospital in South India. By analysing various platelet indices in these patients, alone and together with other data, and comparing them with actual diagnoses, the team found that a high platelet count in feverish patients could suggest tuberculosis1.

Levels of other platelet indices, when used together with total white blood cell count, could also be used to help differentiate between common infections.

“Platelet count is the first thing we notice on a complete blood count (CBC) report in a patient with acute febrile illness and is commonly low in the ‘big five’ tropical illnesses: dengue, malaria, scrub typhus, leptospirosis and enteric fever,” says Stalin Viswanathan of Indira Gandhi Medical College & Research Institute.

Several different platelet indices have been researched as possible biomarkers in various diseases, such as atherosclerosis and rheumatoid arthritis. “Since [platelet indices] are available at no extra cost in CBC reports, we thought they would be helpful in resource-poor settings such as ours,” explains Viswanathan.

He and his colleague Vijayakumar Saravanakumari developed a flowchart based on the patient data they gathered that used various blood indices, including platelets, to see if it was possible to use this tool to narrow down the diagnosis of febrile illnesses. When this ‘classification tree’ used platelet indices alone, it was found to have a nearly 50% margin of error higher. When it was combined with clinical symptoms, complete blood count, liver function tests and echocardiogram heart rate, this dropped down to 37%.

Decision trees based on these indices and other commonly available tests in hospitals could help doctors in poorer areas decide what types of treatment to initiate for patients with acute febrile illness, the researchers say.

The team categorized patients into seven major etiological groups: bacterial, viral, rickettsial, parasitic, mycobacterial, undifferentiated and non-infective. Among the groups, tuberculosis, a mycobacterial infection, had the highest platelet counts and plateletcrit, consistent with other studies on the disease. This suggests that these indices have good sensitivity and specificity for the diagnosis of tuberculosis.

“These results may not be applicable to other, especially non-tropical, locations,” says Viswanathan. “Platelet indices plus total [white blood cell] count should preferably be the first test along with blood culture and peripheral smear [for patients with acute fever],” he says.

The researchers next hope to study and thus standardize normal platelet ranges in a large population size in their locality to better determine abnormal levels. They also plan to study how platelet indices change during the duration of a hospital stay up until the time of discharge.

Reference

  1. Viswanathan, S. & Saravanakumari, V. Are platelet indices useful in diagnosis of tropical acute febrile illnesses? J. Local Glob. Health Sci. 2016, 3 (2016). | article

DOI: 10.1038/qsh.2016.123

Creative Commons License
The text of this article is licensed under a Creative Commons Attribution 4.0 International License. For rights on images refer to their source.