Dengue is the most prevalent mosquito- borne viral disease in South East Asia [1]. It is caused by four dengue virus strains from the genus Flavivirus and transmitted by the Aedes aegypti mosquito. The risk of severe disease and death underscores the importance of early detection of dengue fever (DF) and monitoring for signs of progression to severe disease. Currently, there are no simple clinical and laboratory markers that can predict whether a patient with DF will develop life threatening dengue hemorrhagic fever (DHF) or dengue shock syndrome (DSS) [2]. There is a need to develop a simple effective predictor that can identify patients at risk of severe disease. Ideally, the test should be cheap, fast, highly sensitive and specific.

Previous studies have documented proteinuria during the course dengue fever. In 1995, Garcia et al observed proteinuria in 22% of dengue fever patients, 38% of whom had it within the first 4 days of the onset of constitutional symptoms [3]. But the authors did not compare the occurrence of proteinuria by severity of disease. In 2004, Wills et al demonstrated among children with DSS markedly reduced plasma concentrations of different-sized proteins with a corresponding increase in fractional urinary clearances of the same proteins. They suggested that a simple test of urine protein excretion may become a useful predictor for the subsequent development of DHF and DSS [4]. In 2010, Lumpaopong et al found proteinuria in 15% of children with DF compared to 27% of those with DHF (p = 0.072) [5].

We previously reported two patients with dengue hemorrhagic fever, who developed self-limiting gross nephrotic range proteinuria without evidence of renal damage such as an increase in serum creatinine, hematuria and urinary casts [6]. The objective of this study is to assess whether proteinuria could be used to indicate which patients with dengue fever would progress to DHF or DSS.

In our series, patients with uncomplicated DF did not develop significant proteinuria, while 96% of patients with impending dengue hemorrhagic fever and dengue shock syndrome developed that. In addition, they developed maximum proteinuria near the day of defervescence. This corresponded to the lowest platelet count as well that developed around the same period as the constitutional symptom i.e. fever disappeared [13]. We found a statistically significant association of higher degree of thrombocytopenia with significant proteinuria (Table 2). The degree of thrombocytopenia also corresponds to the severity of other types of hemorrhagic viral fever [14]. However we noted that proteinuria is self limiting and resolves on resolution of the illness.

Epidemiologically dengue fever is a frequently occurring illness in children and adult of the developing nations in Asia, Africa and South America with an increasing morbidity and mortality because of hemorrhagic complication or dengue shock syndrome [15–17]. Complicated dengue carries a higher mortality and morbidity rate in both pediatric and adult population. Early prediction and detection of dengue complication could enable physicians manage the complicated dengue cases with a better strategy for preventing mortality and morbidity. In a recent retrospective study, Lima et al, showed that detection of NS1 antigen of dengue virus (DENV) in tissues of dengue fatal cases had a strong positive association with confirmation of previously fatal cases of dengue fever. Even though, this DENV NS1 capture assay was studied as a test for a rapid and valuable postmortem dengue diagnostic test, they postulated that, with an increasing number of DHF and associated fatality, the availability of this new approach would be useful for the disease surveillance [18]. Although this is an approach for early diagnosis of DHF by antigen assay, it may not be practicable in clinical practice as it is a tissue diagnosis based on postmortem tissue analysis of fatal cases.

Association of proteinuria in dengue has been known to investigators for sometime. But, they could not find any difference in proteinuria between DF and DHF [5]. Unlike our study, these studies did not differentiate between significant and non-significant proteinuria. We could not identify in literature any study measuring daily UPCR for detecting peak proteinuria, as one sample of spot UPCR may not be representative of peak proteinuria for the representative case. These could explain the observation of lack of difference in proteinuria between DF and DHF in those studies.

The importance of proteinuria resulting from dengue induced secondary glomerulonephritis is elucidated in literature by various authors [19]. Renal biopsy in DHF, even showed the presence of IgA nephropathy that was transient in nature, showing the possibility of immune-complex disease in dengue fever because of DENV associated antigen-antibody-complex deposition in renal glomerular tissue [20]. Identification of soluble immune complex in DHF specific for dengue virus in serum reiterates this issue [21]. Antibodies to DENV were known to be associated with DHF in earlier studies as back as 1969 [22]. In -vitro studies with normal human B-lymphocytes demonstrated the detection of immune complex with preformed dengue antigen. This indicates that dengue immune complex might deposit in vascular and glomerular tissue leading to vasculitis of DHF and glomerulnephritis leading to proteinuria [23]. The degree of proteinuria (peak UPCR as studied in our series) might indicate the severity of dengue infection. The significant peak proteinuria could be a manifestation of an autoimmune pathogenetic mechanism that the virus triggers on the lympho-reticular system, resulting in glomerular leakage of protein due to glomerulonephritis associated with DHF [6]. Thus, peak UPCR could have an impact on predicting evolution of DHF. However, the glomerulonephritis itself resolves when the patient recovers from the illness.

Limitation of our study was the absence of renal biopsy to document and confirm glomerulonephritis in cases of significant UPCR as it was felt not necessary because of the self limiting nature of proteinuria. The other limitation was the lag period between admission of the case and initiation of collection of spot urine for UPCR, as urine was collected only after including the case in the study protocol after meeting the inclusion criteria. The study protocol would be initiated after the case had been transferred to study team from the admitting team. Since we measured daily UPCR for detecting the peak, this limitation might not have an impact on the final outcome. The onset of proteinuria could have been earlier in fact. In spite of these limitations we strongly feel that the peak UPCR in dengue should be practiced by primary care physicians when they encounter suspected cases of dengue as the onset of proteinuria and subsequent demonstration of peak UPCR could have a major implication in identification of incipient DHF and stratification of management strategy for such cases. The primary physician in the community has an advantage of seeing the patients earlier. They can administer the relevant tests including the UPCR more quickly. This may better assist the physician to stratify the severity of the illness earlier on. A similar analogy would be the diagnosis of preeclampsia/eclampsia, where the onset of new onset proteinuria from a patient without preexisting renal disease is highly significant of impending disease [24].

Further studies to define a commercially available urine dipstick equitable to the UPCR for detecting urinary protein in a patient with dengue with good sensitivity and specificity would reduce the difficulty of UPCR estimation. This test with a threshold value of 0.20 gm/mg could be used as an early warning system for predicting dengue complications. Further studies with large series of patient at multi-center set-up and improved protocol of collection of urine at primary care levels, are required to validate the important positive finding of our small series. Protocol could include use of use of urine dip-stick and UPCR together to validate more convenient urine dip-stick over the complicated and laboratory dependent UPCR.

We conclude that the onset and the peak of proteinuria using the UPCR have a significant association with subsequent development of DHF. Therefore, the daily UPCR could be a useful simple adjunct for physicians in the community hospital for early determination of dengue complications.