Keywords
dengue - dengue-like illness - concurrent infection - RT-PCR - NS1 antigen
Introduction
Dengue virus infections are a significant cause of morbidity and mortality and lead
to adverse economic effects in many developing tropical countries.[1] The incidence of dengue fever is on the rise worldwide, and in some areas of Asia,
complications of the disease are a leading cause of serious illness and death in children.[2]
[3] Over the past two decades in India, there has been a dramatic increase in dengue fever (DF), dengue hemorrhagic fever (DHF), and dengue shock syndrome (DSS), and their
epidemics.[4]
[5]
[6] The identification of dengue cases is by distinct clinical features, but they can
present with varied manifestations.[7]
[8]
[9] Dengue remains a puzzling disease in many aspects, such as the virus-vector and
host–virus relationship, and clinical expression variability.[10]
[11]
[12] The dengue epidemics in India are cyclical and are more frequent, expanding into
the rural areas and dengue viral strains are circulating in the community either by
local evolution or importations, leading to a heterogeneous distribution of serotypes/genotypes
in different geographical areas.[10]
[11] This in turn alters the infectivity of the virus and gets reflected as changes in
the clinical spectrum of the disease. The aim of the present study was to find the
incidence of concurrent infection of dengue virus and to correlate the difference
in clinical features, laboratory diagnoses, and outcomes between dengue and dengue-like
illness
Materials and Methods
Ethical Clearance
The protocol was approved by the Institutional Ethics Review Board. Consent was obtained
from all patients.
Type of the Study
Prospective study.
Study Duration: Six years (July 2014–June 2020).
Inclusion Criteria
According to specific inclusion criteria, 2,256 patients with fever (presenting within
5–7 days of onset with body temperature above 100°F at the time of blood sample collection)
and fulfilling the case definition criteria of DF and DHF of the World Health Organization
(WHO) were included in the study. Clinical and demographic data were collected by
interviewing the patients or their attendants and meticulous physical examination
of the patients conducted by their treating physician.
Specimen Collection
Two blood samples were collected from each patient and one blood sample was sent on
ice to the molecular laboratory for the detection of dengue viruses, the NS1 antigen,
and IgM and IgG antibodies. The second sample was used for a complete blood hemogram
and for other investigations such as Widal test, malarial test by card and peripheral
smear, chikungunya test by card method and for typhus fever test. The clinical basis
for diagnosing patients with dengue virus infection was based on the WHO definitions.[11]
Reports of hematological investigations, dengue serology, and data obtained from daily
follow-up were analyzed. Hospitalized patients were categorized into DF, DHF, and
DSS according to the WHO severity grading scale.[11] The blood indices were initially measured on a continuous scale and then categorized
on the basis of clinically meaningful cut offs. Thrombocytopenia was defined as a
platelet count less than 100,000 cells/mm3 blood. A hemotocrit more than 20% rise was considered raised. Similarly, leucopenia
was defined as a white cell count less than 4,000 cells/mm3.
Viral RNA was extracted from serum samples using the QIAamp Viral RNA mini kit (Qiagen,
Germany) according to the manufacturers' instructions. The extracted RNA was stored
at −70°C or immediately used for RT-PCR. RT and PCR were performed in one tube using
a universal primer and a one-step RT kit (QIAGEN, GmbH, Hilden, Germany); the reaction
was then placed in a thermal cycler (Eppendorf, Germany). The preliminary product
was further used for nested PCR in another reaction tube.[12] Nested PCR was performed with a thermal cycler. The secondary PCR product was subjected
to agarose gel electrophoresis using a 2% agarose gel (Bangalore Gene) in Tris-borate
buffer, followed by staining with ethidium bromide and visualization on a UV transilluminator
at 302 nm. The NS1 antigen and IgM and IgG antibodies were detected with immuno-chromatographic
test (ICT). The test kit used was the dengue NS1 antigen and antibody combi card supplied
by J. Mitra and Co. Pvt. Ltd. (New Delhi, India).[13]
Statistical Analysis
Data were analyzed using the Epi-info software. The categorical data were shown in
terms of numbers and percentages and analyzed using the Z-test for proportions and
chi-square test.
Results
Out of 2,256 patients enrolled for the study, 1,412 (62.6%) were male and 844 (37.4%)
were female. In addition, 1,356 (60.1%) were from rural area and 900 (39.9%) were
urban population. The age distribution is shown in [Table 1]. The study population age ranged from 2 to 44 years.
Table 1
Demographic, sex, and age of study subjects
|
Features
|
RT-PCR
|
Total
|
|
Positive
|
Negative
|
|
Demographic
|
Rural
|
840
|
516
|
1,356
|
|
Urban
|
466
|
434
|
900
|
|
Sex
|
Male
|
789
|
623
|
1,412
|
|
Female
|
517
|
327
|
844
|
|
Age
|
< 5 y
|
129
|
149
|
278
|
|
6–10 y
|
522
|
304
|
826
|
|
11–15 y
|
379
|
263
|
642
|
|
> 15 y
|
286
|
224
|
510
|
Out of 2,256 clinically suspected dengue cases, 1,306 were positive and 950 were negative
by RT-PCR test.
Various clinical features are summarized in [Table 2]. Among the RT-PCR-positive dengue cases, fever was the most common clinical presentation
occurring in all patients. There was no specific pattern of fever, and the fever ranged
from 38°C to 40°C. Other common clinical features were retroorbital pain (85.9%),
flushing in 77.5%, and rashes in 84.8% of patients. ARDS was seen in 9.7%, splenomegaly
in 27.5%, ascites in 20.3%, and encephalopathy in 3.4% of patients.
Table 2
Clinical manifestations of patients with RT-PCR-positive and -negative cases
|
Clinical features
|
RT-PCR
|
|
Positive
|
Negative
|
|
Number
|
Percentage
|
Number
|
Percentage
|
|
Fever
|
1,306
|
100.0
|
950
|
100.0
|
|
Retro orbital pain
|
1,122
|
85.9
|
870
|
91.6
|
|
Flushing
|
1,012
|
77.5
|
712
|
74.9
|
|
Rash
|
1,108
|
84.8
|
332
|
34.9
|
|
ARDS
|
127
|
9.7
|
188
|
19.8
|
|
Encephalopathy
|
44
|
3.4
|
92
|
9.7
|
|
Hepatitis
|
141
|
10.8
|
57
|
6.0
|
|
Malena
|
73
|
5.6
|
361
|
38.0
|
|
Epistaxis
|
12
|
0.9
|
18
|
1.9
|
|
Hematemesis
|
13
|
1.0
|
6
|
0.6
|
|
Hepatomegaly
|
99
|
7.6
|
327
|
34.4
|
|
Splenomegaly
|
38
|
2.9
|
104
|
10.9
|
|
Ascites
|
265
|
20.3
|
54
|
5.7
|
|
Pleural effusion
|
164
|
12.6
|
39
|
4.1
|
|
Cyanosis
|
14
|
1.1
|
8
|
0.8
|
|
Convulsion
|
55
|
4.2
|
113
|
11.9
|
|
Oliguria
|
14
|
1.1
|
10
|
1.1
|
|
Hypoglycemia
|
16
|
1.2
|
6
|
0.6
|
|
Abscess
|
14
|
1.1
|
21
|
2.2
|
|
Pneumonia
|
17
|
1.3
|
32
|
3.4
|
|
Hematuria
|
8
|
0.6
|
6
|
0.6
|
|
Gum bleeding
|
2
|
0.2
|
6
|
0.6
|
|
Sub conjunctival hemorrhage
|
2
|
0.2
|
22
|
2.3
|
|
Total
|
1,306
|
100.0
|
950
|
100.0
|
Among the RT-PCR-negative dengue cases, fever was the most common clinical presentation
occurring in all patients. Retroorbital pain was seen in 91.6% of patients, 74.9%
of patients exhibited flushing, rashes were seen in 34.9% of patients, ARDS was seen
in 19.8%, ascites in 5.7%, and encephalopathy in 9.7% of patients ([Table 2]).
A platelet count less than 100,000 was observed in 1,838 (81.5%) patients, a platelet
count of 50,000 to 100,000 was observed in 835 (37%) patients, a platelet count of
20,000 to 50,000 was observed in 856 (37.9%) patients, and a platelet count less than
20,000 was observed in 147 (6.5%) patients ([Table 3]).
Table 3
Platelet count during the admission among patients with suspected dengue fever
|
Age (Years)
|
Platelet count (cells/mm3)
|
Total
|
|
< 20,000
|
20,001 to 50,000
|
50,001 to 100,000
|
> 100,000
|
|
< 5
|
20
|
120
|
80
|
58
|
278
|
|
6–10
|
78
|
328
|
294
|
126
|
826
|
|
11–15
|
42
|
232
|
318
|
50
|
642
|
|
> 15
|
7
|
175
|
143
|
180
|
510
|
|
Total
|
147
|
856
|
835
|
414
|
2,256
|
Of 2,256 samples, 1,306 (57.9%) tested positive for dengue viral RNA by RT-PCR. Seven
hundred ninety-eight cases were infected with a single DENV serotype, and 608 had
a concurrent infection with two DENV serotypes. Of the 798 single-infection cases,
392 (54.2%) were typed as DENV-2, 218 (29.2%) as DENV-3, 114 (8.3%) as DENV-1, and
74 (8.3%) as DENV-4. DENV-2 dominated the outbreak, accounting for 49.1% of the positive
samples, followed by DENV-3 (27.3%).
The overall prevalence of concurrent infections was 46.6%. Coinfection with serotypes
DENV-2 and DENV-3 was found to account for 67.8% of all concurrent infections. Other
combinations included the following: DENV-1 and DENV-3 (81 of 608, 16.7%); DENV-2
and DENV-4 (48 of 608, 11.1%); DENV-2 and DENV-4 (48 of 608, 5.6%); and DENV-3 and
DENV-4 (44 of 608, 5.6%). Thus, DENV-2 and DENV-3 were the most commonly combined
serotypes observed over the period.
Out of 1,306 RT-PCR-positive cases, 1,148 were found to have positive serology (NS1,
NS1 and IgM, IgM,) and 86 cases were positive among 921 dengue RT-PCR-negative cases.
Among the serologically positive cases, 1,134 cases were NS1 positive alone, 122 cases
were NS1 and IgM positive, and 12 cases were only IgM positive among the 1,306 cases
of RT-PCR-positive cases ([Table 4]). Among RT-PCR-negative cases, 15 cases were NS1-positive, 22 were NS1 and IgM positive,
8 cases were IgM alone positive, 12 were IgM and IgG positive, and 29 cases were positive
for IgG in RT-PCR-negative cases ([Table 4]).
Table 4
Comparison of dengue serology with RT-PCR-positive cases
|
Dengue serology
|
Result
|
RT-PCT
|
|
Positive
|
Negative
|
|
Number
|
Percentage
|
Number
|
Percentage
|
|
NS1
|
Positive
|
1,134
|
86.8
|
15
|
1.6
|
|
Negative
|
172
|
13.2
|
935
|
98.4
|
|
NS1 + IgM
|
Positive
|
122
|
9.3
|
22
|
2.3
|
|
Negative
|
1,184
|
91.7
|
928
|
97.7
|
|
IgM
|
Positive
|
12
|
0.9
|
8
|
0.8
|
|
Negative
|
1,294
|
99.1
|
942
|
99.2
|
|
IgM + IgG
|
Positive
|
0
|
0
|
12
|
1.3
|
|
Negative
|
1,306
|
100
|
938
|
98.7
|
|
IgG
|
Positive
|
0
|
0
|
29
|
3.1
|
|
Negative
|
1,306
|
100
|
921
|
96.9
|
Among the 1,306 RT-PCR-positive dengue cases, 660 (50.5%), 376 (28.8%), and 270 (20.7%)
were classified as classical DF, as well as DHF, and DSS.
Out of 950 RT-PCR-negative cases, 44 samples tested positive for malaria by card test
and also by peripheral smear, 69 samples were positive for enteric fever by Widal
test, and 126 samples were positive for scrub typhus by ELISA test, and 10 were positive
for chikungunya.
Discussion
Dengue virus causes a broad spectrum of illnesses, ranging from mild undifferentiated
fever to classical DF, as well as DHF and DSS.[3]
[4]
[14] Dengue is a major public health problem in central Karnataka. Over the past 15 years,
we have been observing varied clinical manifestations of dengue,[4]
[7]
[12] which are rather different from the past reports from this region as well as from
other parts of the country.
The seasonal drift of dengue virus infection is revealed by the zenith of positive
cases observed during the month of August to December, which is in concordance with
previous outbreaks.[3]
[7]
[14]
[15]
[16] The majority of patients (62.6%) in our series were males and from rural area (60.1%).
The commonest age group involved were young children varying from 6 to 10 years of
age. In the present study, 2,256 patients were admitted with suspected DF. Among these,
1,306 patients (57.8%) were confirmed to have the dengue disease by RT-PCR, and 44
samples tested positive for malaria by card test and also by peripheral smear, 69
samples were positive for enteric fever by Widal test, and 126 samples were positive
for scrub typhus by ELISA test, and 10 were positive for chikungunya.
Most developing countries have epidemics of febrile illnesses, which can be confused
with DF.[14]
[16] A recent review of published studies was unable to make any conclusions on the signs
and symptoms that can clinically distinguish dengue from other febrile illnesses.[7]
[9]
[14]
Even in the present study, there was not much significant difference in the clinical
features of DF (RT-PCR-positive) and dengue-like illness (RT PCR-negative). Dengue
fever is generally described as a short febrile illness. The WHO criteria mention
an illness of 2 to 7 days' duration.[5] In the present study, we have observed a longer duration of fever than in previous
years. The mean duration of fever in survivors in this study was almost 7.4 days and
the longest duration of fever was more than 3 weeks 4 days in 32 patients. The predominant
clinical features were fever, retroorbital pain, flushing, and myalgia, and vomiting
was the most frequent symptom, which has also been observed in other studies.[5]
[6]
[7] The major difference from the previous report is the frequent occurrence of acute
respiratory distress syndrome, splenomegaly, prolonged fever, and encephalopathy.
In addition, 9.7% of DF cases showed ARDS, while 19.8% of dengue-like cases showed
ARDS.
Although hepatomegaly is among the WHO clinical criteria for DF, splenomegaly is not
generally held to be a feature of dengue infection. Earlier studies in India do not
describe a high frequency of splenomegaly.[7]
[8]
[9]
[10]
[11]
[12] But in our study, we observed 27.5% of the cases had splenomegaly among DF. Our
study corroborates with a recent study from Delhi that has reported a higher percentage
(32.4%) of splenomegaly in children with dengue.[7]
Even though a high incidence of encephalopathy has been reported in several studies,
it was observed in 3.4% of cases in DF and 9.7% in dengue-like fever in our study.
On an analysis of laboratory findings, it was observed that platelets were below 100,000/mm3 in 81.5% of dengue suspected cases and 6.5% of cases had platelets less than 20,000/mm.3 Thrombocytopenia is thought to be due to depression of the bone marrow observed in
acute stage of dengue virus infection. Other explanations are direct infection of
megakaryocytes by the virus, leading to increased destruction of platelets or the
presence of antibodies directed against platelets.[17] A leukocyte count less than 4,000/mm3 was seen in 77.6% of RT-PCR-positive dengue cases and hematocrite value more than
20% for age and sex was observed in 65% of cases. Coagulopathy is also frequent in
most patients with DSS. In our study, prolongation of PTT was quite common in DSS
cases of both groups. The other important abnormality was raised liver enzymes. sGOT
and sGPT were raised above the normal limit in 47.7% of cases and 46.2% of cases,
respectively, in the study cohort.
The most challenging problem associated with patient management in dengue infection
is rapid diagnosis. Although the commercially available MAC ELISAs or ICT offer improvements
over other conventional assays for the diagnosis, they do not offer serotype-specific
diagnosis. Diagnosis based on the detection of antigen (NS1) can be achieved on first
day, but in the present study, the NS1 was positive only in 96.2% of cases and detection
of IgM antibodies can only be achieved after 5 to 7 days of illness. The detection
of concurrent infections can be made by virus isolation in tissue culture, followed
by indirect immunofluorescence using serotype-specific monoclonal antibodies and/or
RT-PCR.[11] However, RT-PCR offers accuracy and speed along with serotype-specific diagnosis
of various circulating dengue viruses and information about co-circulation of different
subtypes. It is now notably clear that the epidemics caused by multiple dengue virus
serotypes have become more frequent on a global basis in the last 18 years.[9] General belief is that concurrent infections by different dengue serotypes occur
during epidemics only, where multiple virus serotypes are being transmitted. The co-circulation
of multiple dengue serotypes in the same region has been documented in several countries
for decades.[14]
[15]
[16] It has been hypothesized that concurrent infections by multiple dengue virus serotypes
may influence the clinical expression of the disease. This is considered as a single
major factor for the emergence of DHF.
The detection of dengue virus RNA by RT-PCR showed that DENV-2 and DENV-3 were the
most common etiologic agents, followed by DENV-1 in this period of study. Only three
dengue virus serotypes were found to be co-circulating. Previously from Delhi and
its surrounding areas, only two concurrent dengue cases were identified[18] but now the number has significantly increased. One fact that may be of considerable
importance in concurrent infections is the occurrence of recombination events. Their
recombination may lead to the emergence of more virulent strains.
