Predicting the severity of Dengue based on a few clinical and laboratory indicators: A retrospective analysis

Dr. Purwa Doke; Dr. Bharat Purandare

Volume : IX, Issue : IX, September - 2019

Predicting the severity of Dengue based on a few clinical and laboratory indicators: A retrospective analysis

Dr. Purwa Doke, Dr. Bharat Purandare

Abstract :

Background: Dengue has acquired an important place amongst the feile illnesses, for three reasons. Firstly, for its increasing incidence in an epidemic form; secondly, it’s diverse severity. And lastly, for a near absence of an indicator system predicting its severity/mortality. Although there have been some attempts like identifying the molecular indicators to predict the severity of Dengue fever; yet a well acknowledged and a simple investigations based system is sought after. Aim:The aim of the study was to measure the association between some laboratory investigations and clinical findings with the outcomes especially admission to ICU, among the admitted Dengue patients. Materials and methods:The study was carried out in a teaching hospital located in Pune Municipal Corporation area. It is a 850 bedded hospital having basic and super–speciality services. Secondary data from the hospital records section was collected. Diagnosis confirmed by either NS1 antigen or IgM antibodies was accepted. Patients discharged from 1st July to 31st December 2015 were first enlisted. Their relevant details were coded and entered in an excel sheet and analysed. Pearson correlation coefficient between quantitative variables and ICU admission were calculated and χ2 test for qualitative variables was applied. A simple system having a maximum score of 10 was developed, based on five variables. Results:During the study duration, 370 patients were admitted. There was a seasonal variation showing a peak in the month of November.The mean platelet count was 61,165. Few patients having platelet count less than 50,000 had some bleeding tendency. Diabetes was the commonest (13%) comorbidity. The association between a higher total score and ICU admission was observed. Conclusions:The scoring system is easy and its predictability is satisfactory.