Background:
There is an escalation of frequency and magnitude of dengue epidemics in Malaysia, with a concomitant increase in patient hospitalization. Prolonged hospitalization (PH) due to dengue virus (DENV) infections causes considerable socioeconomic burden. Early identification of patients needing PH could optimize resource consumption and reduce health care costs. This study aims to identify clinicopathological factors present on admission that are associated with PH among patients with DENV infections.
Methods:
This study was conducted in a tertiary referral hospital in Southern Malaysia. Relevant clinical and laboratory data upon admission were retrieved from medical records of 253 consecutive DENV nonstructural protein 1 (NS1) antigen and PCR-positive hospitalized patients. The DENV serotype present in each patient was determined. Patients were stratified based on duration of hospital stay (<4 vs. ≥4 days). Data were analyzed using IBM® SPSS® 25.0. Multivariate logistic regression was performed to examine the association between PH and admission parameters.
Results:
Of 253 DENV hospitalized patients, 95 (37.5%) had PH (≥4 days). The mean duration of hospital stay was 3.43 ± 2.085 days (median = 3 days, interquartile range = 7 days). Diabetes mellitus (adjusted odds ratio [AOR] = 6.261, 95% confidence interval [CI] = 2.130-18.406, p = 0.001), DENV-2 serotype (AOR = 2.581, 95% CI = 1.179-5.650, p = 0.018), duration of fever ≤4 days (AOR = 2.423, 95% CI = 0.872-6.734, p = 0.09), and a shorter preadmission fever duration (AOR = 0.679, 95% CI = 0.481-0.957, p = 0.027) were independently associated with PH. However, PH was not found to be associated with symptoms on admission, secondary DENV infections or platelet count, hematocrit, or liver enzyme levels on admission.
Conclusions:
Early identification of these factors at presentation may alert clinicians to anticipate and recognize challenges in treating such patients, leading to more focused management plans that may shorten the duration of hospitalization.
* Title and MeSH Headings from MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine.