See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/320629012

Dengue fever in a south Asian metropolis: a report on 219 cases

Article  in  Iranian Journal of Microbiology · June 2017

CITATIONS READS

4 6,586

7 authors, including:

Shiv Sekhar Chatterjee

Diamond Harbour Government Medical College & Hospital. Diamond Harbour
45 PUBLICATIONS   1,253 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Shiv Sekhar Chatterjee on 11 July 2020.

The user has requested enhancement of the downloaded file.

 Volume 9 Number 3 (June 2017) 174-185

Dengue fever in a south Asian metropolis: a report on 219 cases

ORIGINAL ARTICLE

Shiv Sekhar Chatterjee1*, Ankush Sharma1, Shilpee Choudhury2, Sushil Kumar Chumber1, Ras Bage2,
Nittin Parkhe3, Uma Khanduri1

1
Department of Laboratory Diagnostic Services, St Stephen Hospital, Delhi, India
2
Department of Medicine, St Stephen Hospital, Delhi, India
3
Department of Radiology, St Stephen Hospital, Delhi, India

Received: November 2015, Accepted: November 2016

ABSTRACT

Background and Objectives: Yearly epidemics of Dengue fever occur post-monsoon in India’s capital, Delhi. A prospective
observational study was conducted during the outbreak months to understand the epidemiology and outcome of this infection
and its economic impact.
Materials and Methods: Febrile hospitalized (n=219) patients with dengue fever diagnosed by a combination of MAC-ELI-
SA, GAC-ELISA and NS1Antigen-ELISA were enrolled. Epidemiologic (including economic) parameters, clinical, radio-
logical and laboratory manifestations were noted and patients followed up over the period of hospital stay. Patient manage-
ment means and outcome were recorded and analysed.
Results: As per WHO-2009, 153 (69.9%) and 27 (12.3%) patients were classified as dengue with warning signs and Severe
Dengue respectively while according to WHO-1997 guidelines 39 (17.8%) and 18 (8.2%) patients were classified as DHF
and DSS respectively. 216 patients were from the city while three were travellers; hospitalization was more frequent among
the young and male gender. Fever, vomiting, aches and abdominal pain were the most common troublesome manifestations;
classical dengue triad was present in 55 (25.1%) patients; hemorrhagic, neurologic and mucocutaneous manifestations were
present in 44 (20.1%), 8 (3.7%) and 70 (32%) patients. Ascitis, pleural effusion, and Gall bladder wall oedema was found
in 53 (24.2%), 31 (14.1%) and 45 (20.5%) patients respectively. Mortality was 1.4% (3 deaths); in addition there was an
intra-uterine fetal death; mean expenditure per patient during the illness was US$ 377.25.
Conclusion: Dengue virus infection results in immense morbidity and substantial mortality.

Keywords: Dengue fever, Flavivirus, Dengue Hemorrhagic fever, Dengue shock syndrome

INTRODUCTION causing immense morbidity and occasional mortality

(1). Though world-wide in distribution, dengue poses
Dengue, a mosquito borne Flavivirus infection, a bigger challenge in resource poor countries, like In-
leads to recurrent epidemics in urban agglomerates dia, especially for the urban and rural poor. Recent-
ly, World Health Organization (WHO) has modified
*
Corresponding author: Dr. Shiv Sekhar Chatterjee, As- guidelines for management of dengue infection (2).
sistant Professor, Department of Laboratory Diagnostic Previous classification of Dengue Hemorrhagic fe-
Services, St Stephen Hospital, Delhi, India. ver (DHF) and Dengue shock syndrome (DSS) have
Tel: +9103325644070, +919748732366 been replaced by the new triage of Dengue without
Email: [email protected] warning signs, Dengue with warning signs and Se-

174
 DENGUE FEVER IN DELHI

vere Dengue to expedite adequate therapy (2). Cri- active or iv) NS1Ag and GAC-ELISA were included
teria for presumptive diagnosis, hospital admission, in the study (n=219).
fluid and blood component administration, and dis-
charge have been forwarded. These may significantly Exclusion criteria for observational study. 1)
impact dengue mortality; however their usefulness is Patient tested with rapid Immunochromatograph-
not validated in India (2). Repeated dengue epidem- ic test for Dengue antibodies or Antigen, 2) Patient
ics occur in Delhi, National Capital Region, India (1, treated on outpatient basis (not admitted), 3) Pa-
3-5) where a large proportion of people live under tient MAC-ELISA and NS1Ag-ELISA negative,
the poverty line (6) or in unhygienic conditions (6). GAC-ELISA alone reactive, 4) Febrile patient with
We planned this study to highlight the clinical and blood culture positive bacterial sepsis or significant
laboratory features of acute dengue infection and its pyuria and bacteriuria or malaria parasite document-
financial impact on the affected patients. ed by thick, thin films and antigen detection test.
Informed consent was taken from each patient en-
rolled in the study. Institutional ethics committee ap-
MATERIALS AND METHODS proval was taken for conducting this study. Patient’s
occupation, daily income, loss of income if any, total
A prospective observational study was undertaken expenses due to the illness were recorded. Meticu-
at St Stephen Hospital, Delhi from August 2013 to lous clinical examination was carried out daily; all
January 2014. A total of 918 suspected dengue pa- patients underwent the following investigations as
tients were tested with for NS1 antigen, anti-dengue per the clinical requirement–complete blood count
IgM and IgG antibodies by ELISA. Antibody testing (CBC), blood culture, urine culture, thin and thick
was done with commercially available Panbio Dengue smears for malarial parasite, rapid malarial antigen
IgM capture ELISA (Inverness Medical Innovations, testing (Sure Test, pan-LDH-2 antigen detection,
Cat No. E-DEN01M/EDEN01M05) (MAC-ELISA) Microgene Diagnostics), Erythrocyte Sedimentation
and Dengue IgG capture ELISA (Inverness Medical Rate (Westergren’s method)), liver function tests
Innovations, Cat No. E-DEN02G) (GAC-ELISA). (LFT), renal function tests, chest X-ray (CXR) and
One Negative control, one Reactive Control and ultrasonography (USG). CBC was done daily for the
Callibrators in triplicate was used in each ELISA first 4 days of hospital stay and then as and when re-
test as per manufacturer instructions. Index value quired depending on the clinical situation. Clinically,
[ratio of sample absorbance and cut-off value (cal- presence of tachypnea, chest retractions, decreased
culated from calibrator ODs and calibration factor breath sounds and decreased vocal resonance were
given in each kit)] and Panbio units were calculated considered signs of pleural effusion. Presence of
for each sample and results are classified as positive abdominal distension with fullness of the flanks
[Index value >2.2], negative (Index Value<1.8) and and presence of shifting dullness or fluid thrill was
equivocal (Index Value~1.8-2.2) according to the taken as evidence of ascites. The extent of hemocon-
manufacturer’s instructions. Any initial equivocal centration was quantitated by taking a difference
result was retested to confirm the result. NS1 anti- between the maximum hematocrit at admission or
gen (NS1Ag) testing was done with commercially anytime during the hospital stay and the minimum
available PlateliaTM Dengue capture 96-well sand- hematocrit recording at convalescence or discharge
wich format ELISA (Bio-Rad, Cat. No.72830). One (7). Dengue with warning signs was defined as lab-
Negative control, two calibrators, and one Posi- oratory confirmed dengue with any of the following:
tive Control were used in each run, and the cut-off abdominal pain or tenderness, Persistent vomiting,
was calculated as mean of calibrators. All ELISAs clinical fluid accumulation, mucosal bleed, lethargy,
were performed on a fully automated EVOLIS restlessness, Liver enlargement >2 cm and increase
platform. in hematocrit (>20%) concurrent with rapid decrease
in platelet count (to below 40000/ul) (2). Severe den-
Inclusion criteria for observational study. Fe- gue was defined as patients with any of the following
brile hospitalized patients with either i) MAC-ELISA features: severe plasma leakage with shock and/or
and GAC-ELISA reactive, or ii) MAC-ELISA alone fluid accumulation with respiratory distress, severe
reactive, or iii) NS1Ag-ELISA and MAC-ELISA re- bleeding, or severe liver, renal, cardiac, and pulmo-

http://ijm.tums.ac.ir 175
IRAN. J. MICROBIOL. Volume 9 Number 3 (June 2017) 174-185http://ijm.tums.ac.ir
Shiv Sekhar Chatterjee ET AL .

nary or central nervous system impairment (2). DHF Epidemiological findings. Among the dengue pa-
was diagnosed as per the older WHO guidelines as tients, 149 were male and 70 females. 191 patients
a probable case of dengue fever with hemorrhag- were of north-indian descent, 11 of northeast-indian
ic tendencies and thrombocytopenia along with the descent, 11 south-indian, and 6 of east-indian descent.
presence of evidence of plasma leakage manifested 216 patients were from National Capital Region (72
by any one or more of the following i.e., a rise in the Trans-Yamuna River East Delhi, 3 from Dwarka, 2
average hematocrit for the age and sex by >20%; a from Gaziabad, 2 from Sonepat, Fig. 2) while 3 were
>20% drop in the hematocrit following volume re- travellers. Of those infected, 70 (32%) were students,
placement compared to the baseline; signs of plasma 42 (19.2%) home-makers, 44 (20.1%) employed in
leakage i.e., pleural effusion, ascites, hypoprotein- various offices, 8 (3.8%) factory workers or mechan-
emia (7). The area specific hematocrit cut off values ics, 2 (0.9%) retired, 2 (0.9%) unemployed, and one
for hemoconcentration was defined as >36.3% in less (0.5%) child below school going age. Mean monthly
than 12 years age group (4) and >37.5% in those more income of those gainfully employed was Rs 9872.00
than 12 years (8). Criteria for splenomegaly on USG (US$158.53); mean income loss in these patients due
was span more than 11 cm on greatest dimension to illness was Rs. 2100.60 (US$33.73). Four (1.8%)
or weight greater than 250gm (9) and that for hep- patients gave history of past episode of dengue fever.
atomegaly was span at mid-clavicular line greater
than 15.5 cm (10). Guidelines of the National Vector Clinical and radiological findings. The clinical
Borne Disease Control Program (NVBDCP) in India findings among 219 dengue patients are detailed in
were used as criteria for assessing appropriateness of Fig. 3. Fever duration ranged from 1 day to 28 days
platelet transfusion (11). Each patient’s total expendi- (mean 5.5 ± 0.4 days); maximum recorded tempera-
ture was determined from the final bill generated for ture ranged from 101°C to 105°C (mean 102.5 ±
each patient. Statistical analysis was carried out on 0.1°C) and saddle back fever was noted in 16 (7.3%)
the data in Microsoft Excel software. patients. The classical dengue triad was present in 55
patients (25.1%). A positive tourniquet test was noted
in 11 patients of 155 patients tested. Hemorrhagic,
RESULTS neurologic and mucocutaneous manifestations were
present in 44 (20.1%), 8 (3.7%) and 70 (32%) patients
During 2013, out of 918 suspected dengue patients respectively and are detailed in Table 3. The second-
(572 males, 346 females), 689 [422 (61.2%) males, ary rash was maculopapular in 39 (73.6%) patients,
267 (38.8%) females) were reactive in any of the three petechial in 9 (15.3%), and macular in 5 (8.5%). The
tests, MAC-ELISA, GAC-ELISA or NS1Ag-ELISA macular and maculopapular rashes were bilateral in
(Table 1). The various combinations of positive tests 43 patients, started mostly on days 4 (10, 22.7%), 5
noted in these patients are shown in Table 1. Maxi- (11, 25%) and 6 (12, 27.3%), lasted a mean duration
mum and minimum incidence of dengue was in the of 2.77 +/- 0.44 days, started distally at upper or low-
age group of 20-30 years (235, 34.1%) and 1-5 years er limbs or both in 37 (82.2%) patients, on the neck
(3, 0.4%) (Fig. 1). Temporally cases began in August in five, and centrally in two, was accompanied by
(1 patient), numbered 70, 357, 148, and 9 in Septem- itching on the rash in 8 (13.6%), and progressed to
ber, October, November, and December (Fig. 1). involve more central areas like upper arms, thighs,
Among these 689 patients, we prospectively neck, chest, back and abdomen in 21 patients but
observed 219 randomly selected patients (148 remained localised to the initial site in 23 patients.
male, 71 female, 45 IgM+veIgG-veNS1Ag-ve, 69 Features of shock recorded include cool, clammy
IgM+veIgG-veNS1Ag+ve, 55 IgM+veIgG-veN- extremities with warm trunk (18, 8.2%), weak pulse
S1Ag-ve, 41 IgM+veIgG+veNS1Ag+ve and 9 (14, 6.4%), and blueness around mouth (2, 0.9%).
IgM-veIgG+veNS1Ag+ve) (Table 1). Among these Mean duration of onset of shock from the start of fe-
219 patients, three (all male) patients died during ver was 3.1 ± 1.2 days. Clinical examination revealed
hospital stay (Mortality 1.4%). As per WHO 1999 hepatomegaly in 18 (8.2%) patients and splenomeg-
criteria, 4 patients (1.8%) were categorized as dengue aly in one patient. Ultrasonographic assessment re-
shock syndrome, 7 (3.2%) as Dengue hemorrhagic vealed enlarged liver (> 17 cm) in further 29 (13.2%)
fever, and 208 as dengue fever (95%) (Table 2). (total 47 (21.5%) patients with hepatomegaly), and an

176 IRAN. J. MICROBIOL. Volume 9 Number 3 (June 2017) 174-185 http://ijm.tums.ac.ir

 DENGUE FEVER IN DELHI

Table 1. MAC-ELISA, GAC-ELISA and NS1Ag-ELISA result combinations in dengue patients (n=918).

Test Number of patients reactive (%)

Anti-Dengue IgM capture ELISA 530 (76.9%)
Anti-Dengue IgG capture ELISA 433 (62.8%)
NS1 Antigen ELISA 415 (60.2%)
Test Combinations reactive
All three (IgM, IgG, NS1Ag) tests reactive 217 (31.5%)
Any two tests reactive
IgM+ve IgG+ve NS1Ag-ve 237 (34.4%)
IgM+ve IgG-ve NS1Ag+ve 63 (9.1%)
IgM-ve IgG+ve NS1Ag+ve 45 (6.5%)
Any one test reactive
NS1Ag+ve only, IgM-ve, IgG-ve 90 (13.1%)
IgM+ve only, IgG-ve, NS1Ag-ve 13 (1.9%)
IgG+ve only, IgM-ve, NS1Ag-ve 24 (3.5%)

MAC-ELISA:anti-Dengue IgM capture ELISA, GAC-ELISA:anti-Dengue IgG capture ELISA, NS1Ag:NS1 Antigen.

Fig. 1. Age, gender distribution and month of presentation of Dengue fever patients

http://ijm.tums.ac.ir 177
IRAN. J. MICROBIOL. Volume 9 Number 3 (June 2017) 174-185 http://ijm.tums.ac.ir
Shiv Sekhar Chatterjee ET AL .

Table 2. Characteristics of dengue patients (n=219) enrolled in observational study.

Indicator N Mean Age ± 3SE Mortality

Participants 219 (100) 29.33 ± 2.5 3 (1.37) (+ 1 IUFD)
Gender
Female 71 (32.4) 34.8 ± 4.8 0
Male 148 (67.6) 26.7 ± 2.6 3 (2.03)
Classification of patients as per WHO 1997
Dengue Fever 162 (74.0) 29.7 ± 2.8 0
Dengue Hemorrhagic Fever 39 (17.8) 29.7 ± 5.5 2 (28.6)
Dengue Shock Syndrome 18 (8.2) 31.5 ± 10.5 1 (25.0)
Classification of patients as per WHO 2009
Dengue fever without warning signs 39 (17.8) 27.9 ± 6.6 0
Dengue with warning signs 153 (69.9) 28.2 ± 2.8 0
Severe Dengue 27 (12.3) 32.1 ± 8.1 3 (11.1)
Co-morbidities
Pregnancy 4 (1.8) 30 ± 14.7 One IUFD
Anaemia (Mild to moderate grade) 27 (12.3) 37.2 ± 8.6 1 (3.7)
Hypertension 13 (5.9) 47.1± 7.1 1 (7.7)
Diabetes mellitus 13 (5.9) 48.7 ± 5.5 1 (7.7)
Bronchial Asthma 5 (2.3) 44.8 ± 16.5 0
Hypothyroidism on supplemental L-Thyroxine 4 (1.8) 42.3 ± 7.8 0
Smokers 11 (5) 36.6 ± 9.9 0
Chronic Alcoholics 12 (5.5) 34.0 ± 9.9 1 (8.3)
Obesity 3 (1.4) 49.3 ± 3.6 0
Malnourishment 1 (0.5) - 0

IUFD:Intra-uterine fetal death, WHO:World Health Organization.

enlarged but clinically non-detectable splenomegaly local cut-off in 212 (96.9%) patients, hemoconcen-
in 41 (19.2%) patients (total 42 (19.2%) patients with tration > 20% and in 15-20% range was found in 43
splenomegaly). Of 42 patients with splenomegaly, 37 (19.8%) and 28 (12.6%) patients respectively. A rise
demonstrated only marginal enlargement of spleen in eosinophils by 2.5% in differential was noted in 72
(12.1-14.0 cm on largest dimension) while 5 patients (32.8%) patients; a leukoerythroblastic blood film in
demonstrated a little larger spleen (14.0-18.0cm). As- 5 patients (2.3%) during some point in their hospital
citis, bilateral pleural effusion, and right sided pleu- stay while left shift and smear cells were document-
ral effusion was found in 53 (24.2%), 14 (6.4%), and ed in 65 (29.7%) and 4 (1.9%) patients respectively.
17 (7.7%) patients by a combination of clinical and Mean ESR at 1 hour was 14.35±2.90 mm.
radiological methods. Gall bladder wall oedema and
thickening was noted in 45 (20.5%) patients. B] Biochemical parameters: Biochemical pa-
rameters of the dengue patients are detailed in Table
Laboratory parameters. 4. Mean ALT, AST, and alkaline phosphatase were
140.7±60.1 IU/dl, 237.8±115.0 IU/dl and 86.2±17.9
A] Hematological parameters: Hematological IU/dl respectively. Mean albumin: globulin ratio and
parameters of the dengue patients are detailed in Ta- ALT: AST ratio was 1.30 ± 0.07 and 0.70 ± 0.06 re-
ble 4. Of note, 200 (91.1%), 111 (57.7%), 204 (93%), spectively.
93 (42.3%) patients respectively developed throm-
bocytopenia, leucopenia, reactive lymphocytes, Course of Illness. 167 (76.2%) patients’ required
and a rise in Absolute Lymphocyte count > 2000/ul intravenous crystalloid fluid therapy, 2 patients in
during their hospital stay. Hematocrit was above the addition required intravenous colloid therapy; 69

178 IRAN. J. MICROBIOL. Volume 9 Number 3 (June 2017) 174-185 http://ijm.tums.ac.ir

 DENGUE FEVER IN DELHI

Fig. 2. Residences of the 216 Dengue patients who were National Capital Region residents

(31.4%), 4 (1.5%), and 3 (1.1%) required platelet gesics and antiemetics was required for 46 (20.9%)
rich plasma transfusion, packed red blood cells, and patients. Renal replacement therapy in the form of
fresh frozen plasma respectively. Out of 69 patients hemodialysis was given to one (0.4%) patient. Twelve
receiving platelet transfusion, 20 had active bleed- patients (4.5%) had to be treated in intensive care
ing (29%), another 13 patients (18.8%) had platelet during hospital stay. Three patients (1.2%), (all male)
counts below 10 × 109 / µl during their course of ill- died during hospital stay and there was one case of
ness while 9 (13%) suffered shock and had an ab- intra-uterine fetal death. Average duration of hospi-
normal coagulogram. The rest 28 patients (40.6%) tal stay was 6.3 ± 1.8 days. Two patients experienced
who received platelets had no appropriate indication hospital acquired blood stream infection (Staphylo-
as per NBVDCP guidelines (13). All 69 patients re- coccus aureus and Klebsiella pneumoniae) and three
ceived 2 or more units (average 3.38 units received); hospital acquired UTI (one each due to Escherich-
no patient received single unit platelet transfusion. ia coli, Enterococcus faecalis, and Streptococcus
Close observation for development of danger signs bovis) during their hospital stay. Four of the above
and symptomatic treatment with anti-pyretics, anal- recovered after change of venous lines, urinary cath-

http://ijm.tums.ac.ir 179
IRAN. J. MICROBIOL. Volume 9 Number 3 (June 2017) 174-185 http://ijm.tums.ac.ir
Shiv Sekhar Chatterjee ET AL .

Fig. 3. Clinical manifestations of dengue fever patients (n=219).

w/wo : with / without, Ghabrahat:Hindi term meaning sense of unease/restlessness

eters and appropriate antibiotic therapy. In the fifth and encephalopathy in one patient and severe hepati-
patient, bilateral lower limb maculopapular rash got tis and encephalopathy in the other). All patients who
infected leading to lower limb cellulitis and pustu- recovered (216) were afebrile at the time of discharge
lar eruptions which yielded Staphylococcus aureus and the mean platelet count at discharge was 129057
and Streptococcus pyogenes on culture. He further ± 14073 / ul. Mean expenditure per patient during
developed features of septicaemia (blood culture the illness was calculated to be Rs.23492.00 (US$
yielding Staphylococcus aureus). On Day-15 of hos- 377.25).
pital stay the patient deteriorated with myocarditis,
arrhythmias and died on Day-18. The other two male
patients who died (both on Day-5 of hospitalization) DISCUSSION
had features of dengue shock syndrome, with mas-
sive mucosal bleeds. Multi-organ involvement oc- Outbreaks of dengue fever with consequent hospi-
curred in both of them (myocarditis, severe hepatitis talization of severe cases are becoming frequent since

180 IRAN. J. MICROBIOL. Volume 9 Number 3 (June 2017) 174-185 http://ijm.tums.ac.ir

 DENGUE FEVER IN DELHI

Table 3. Hemorrhagic, neurologic and mucocutaneous manifestions noted in dengue patients.

Hemorrhagic manifestation Number of Patients (%)

Easy bruising (especially at venipuncture site) 13 (5.9)
Petechiae 9 (4.1)
Fresh bleeding per Rectum 2 (0.9)
Epistaxis and Bleeding Gums 10 (4.6)
Bleeding Gums 5 (2.3)
Bleeding per-Vagina 2 (0.9)
Malena 8 (3.7)
Hematemesis 6 (2.7)
Hemoptysis 1 (0.5)
Hematuria 2 (0.9)
Neurological manifestations
Impaired consciousness 7 (3.2)
Focal Neurological deficit 1 (0.5)
Seizures 1 (0.5)
Mucocutaneous manifestations
Initial Exanthema (Rash) 20 (9.2)
Facial Flushing 2 (0.9)
Central flushing on chest & abdomen 6 (2.7)
Distal Flushing 13 (5.9)
Secondary Rash (Between Day1-10 of fever onset) 55 (25.1)
Conjunctival congestion 7 (3.2)
Oral blisters and ulcers 4 (1.8)
Itching 23 (10.5)
Itching over Rash 10 (4.6)

1980s (12) especially in the south East Asian region. three travellers were also among those hospitalised.
With no specific treatment available, prevention re- This finding highlights the serious situation of dengue
mains a cornerstone. Preventive measures are not ef- in the capital; complicated by the lack of vaccines and
fectively applied in developing countries due to lack chemoprophylaxis against the disease. Travellers and
of funding, trained manpower, community ignorance residents alike must strictly guard against mosqui-
and non-participation (13). The morbidity and finan- to bites through use of mosquito nets and repellents.
cial cost of this infection is tremendous on the afflicted Public health education regarding protection against
subjects (14). In addition fatalities due to dengue fe- mosquito bites is important but the growing resistance
ver are not uncommon as seen in our study (3 deaths, of mosquitoes to insecticides is a concerning factor
one IUFD). Majority of those hospitalised and two of (16).
those who died were young adults (20-40 years). The Fever with chill and rigor, and aches were the com-
male predominance among those hospitalised may be monest symptom. These apart vomiting and lack of
due to rampant gender discrimination in India (15). appetite were very commonly noted. Many patients
The seasonal distribution of dengue cases is well had troublesome persistent vomiting as their main
documented, with a peak incidence after the heaviest complaint; abdominal pain was another significant
monsoon rainfall (12). The exact month of highest in- trouble for some patients. Vomiting and abdominal
cidence may differ according to the timing of the rain- pain have been previously documented to be partic-
fall. Mudpools, open water reservoirs and stagnant ularly troublesome symptoms (17). Dengue triad was
water in houses and roofs are breeding sites for Aedes documented only in a quarter of the patients (25.1%).
(Stegomyia) aegypti mosquitoes (2, 12). Most affect- The tourniquet sign (11 out of 155 patients tested)
ed patients were residents of NCR, mainly school and was even more infrequently documented. The ab-
college going students and homemakers; however sence of the classical dengue triad or the tourniquet

http://ijm.tums.ac.ir 181
IRAN. J. MICROBIOL. Volume 9 Number 3 (June 2017) 174-185 http://ijm.tums.ac.ir
Shiv Sekhar Chatterjee ET AL .

Table 4. Hematological and biochemical parameters of 219 dengue patients.

Hematological Parameters Number of Patients (%)

At admission Later during hospital stay Overall
Leukopenia (TLC < 4000/ul) 87 (39.8) 24 (10.8) 111 (50.6)
Leukocytosis (>11,000/ul) 9 (4.2) 19 (8.9) 28 (12.8)
ALC > 3500/ul 19 (8.5) 52 (23.9) 71 (32.4)
Rise of ALC > 2000/ul 93 (42.5)
Reactive Lymphocytes 126 (57.6) 78 (35.5) 204 (93.2)
Hemoconcentration ~ 15-20% 28 (12.6)
Hemoconcentration > 20% 43 (19.8)
Hematocrit above cut-off 212 (96.9)
Thrombocytopenia (<100,000/ul) 166 (75.7) 34 (15.4) 200 (91.3)
Eosinophilia (AEC> 500/ul) 2 (0.8) 35 (15.8) 37 (16.9)
≥2.5% ↑ in Eo% during hospitalization 72 (32.8)
Leukoerythroblastic blood film 3 (1.4) 2 (0.9) 5 (2.3)
Left shift 39 (14.6) 26 (9.7) 65 (29.7)
Smear cells 1 (0.5) 3 (1.4) 4 (1.9)
Deranged aPTT 15 (6.9)
Deranged Prothrombin Time 7 (3.1)
Biochemical Parameters Number of patients
Raised Alanine Transaminase 161 (73.5)
Raised Aspartate Transaminase 184 (83.8)
Hypoproteinemia (<6g/dl) 66 (30.2)
Hypoproteinemia (<5g/dl) 12 (5.6)
Hypoalbuminemia (<3g/dl) 35 (15.9)
S. Creat.>1.2 mg/dl or s.BUN > 24 mg/dl 24 (11.1)
Raised Alkaline phosphatase 14 (6.3)
Hyperbilirubinemiaa 13 (6)

TLC: Total Leucocyte count, ALC:Absolute Lymphocyte count, AEC:Absolute Eosinophil count, S. Creat:Serum Creatinine,
S. BUN:serum Blood Urea Nitrogen, Eo:Eosinophil, aPTT:activated partial Thromboplastin Time.
a
:All 13 patients had predominant conjugated hyperbilirubinemia.

sign in many patients discourages the use of these for er that with initial flushing is very low compared to
diagnosis or triage. In another two studies, one from same study (20). The macular or maculopapular rash
Eastern India (17) and another from the north (4), pos- occurred commonly between day-4 and day-6 of fever
itive tourniquet sign were found in 31% and 25.5% of (74%), started distally (36, 72%) and progressed cen-
dengue patients. Though hemorrhagic manifestations trally or remained localized. Itching on the rash was
were noted in 44 patients, no individual site or lesion reported only in minority (13.6%) unlike Afzar NA
was responsible for majority of the cases; so it is im- et al. who reported pruritus on 62% of rashes (20).
portant to be aware and on the look out of the myriad In DSS patients’ onset of shock is usually around day
hemorrhagic presentations. Other studies have found 3 of fever (2, 12); in our study shock occurred most
similar wide distribution of hemorrhage patterns in commonly between day 2 and day 4 of fever (61.1%),
dengue patients (4, 17-19). Of cutaneous manifesta- mean duration between onset of fever and shock be-
tions, maculopapular or petiechial secondary skin rash ing 3.1 ± 1.2 days. Hepatomegaly (47 patients, 20%),
was the most common manifestation, however initial detected by combination of clinical and sonological
flushing on the face or limbs or trunk was seen in mi- examinations was a common feature. Though clini-
nority (9.2%) of patients. Proportion of peteichial rash cally imperceptible, splenomegaly was noticed in 41
is similar to that seen patients from Lahore; howev- (19.2%) patients on ultrasound examination. Major-

182 IRAN. J. MICROBIOL. Volume 9 Number 3 (June 2017) 174-185 http://ijm.tums.ac.ir

 DENGUE FEVER IN DELHI

ity of patients with splenomegaly had marginal en- 24 (11.1%) patients. A leukoerythroblastic blood film
largement while five demonstrated a moderately large and smear cells are found in even fewer patients of
spleen (14.0-18.0cm). Similarly, another study noted dengue hemorrhagic fever. Leukoerythroblastic blood
27/70 patients with clinical hepatomegaly and 9/70 film probably represents a bad prognostic factor in
with only sonological evidence while the same with dengue patients as 3 out of the 5 patients died of the
splenomegaly was 3/70 and 15/70 patients respective- disease.
ly (21). Other markers of plasma leakage (2, 12) were A combination of dengue antigen and antibody de-
also detected better by radiological methods including tection by ELISA tests was used in our study to confirm
ascitis and pleural effusion in 24.2% and 14.1% pa- the diagnosis. However, MAC-ELISA and GAC-ELI-
tients respectively, as noted in other studies (17, 21). SA can show serological cross-reactivity among fla-
Further we noted that right sided pleural effusion was vivirus infections like Murray Valley fever, Japanese
earlier to appear and pleural effusion was often uni- Encephalitis, St. Louis Encephalitis, Yellow fever and
lateral. Many such minimal effusions were detected West Nile (2, 25). Rheumatoid factor, malaria, Chi-
more frequently by USG than the conventional CXR. kungunya, Lyme disease, Scrub Typhus, Hanta virus
Thickened and oedematous gall bladder wall changes infection and leptospirosis are additional scenarios
are characteristic of the plasma leakage syndrome in producing false positive MAC-ELISAs (2, 25, 26). A
dengue (22) and may be marker of severe dengue (22). persisting IgM against Dengue virus originating from
Significant thrombocytopenia, marked leucopenia a previous infection will result in false diagnosis (27,
(50.6%) and lymphocytosis (32.8%) with presence of 28). IgM anti-Dengue usually persists for 2-6 months
many reactive lymphocytes in the peripheral smear (28, 29) with a median time period of 179 days for pri-
(93%) are commonly noted features in dengue fever. mary and 139 days for secondary infections (29). An-
In our laboratory we often utilize these parameters to ti-Dengue IgG antibodies persists for an even longer
suspect a dengue case during outbreak periods. To- time even up to years (2, 30). False positive anti-Den-
gether with significant transamnitis (83.4% patients) gue virus IgG detection has been documented in bacte-
and hemoconcentration (19.8%), these findings are a remia, leptospirosis, Q fever, and other viral infections
valuable clue to the laboratory regarding the diagno- like Chikungunya, Tick-borne encephalitis, Varicella,
sis. Interestingly, in addition a rise in eosinophil count Cytomegalovirus, and Epstein-Barr infections (28,
is often documented later in the course of dengue as 31). False negative GAC-ELISA can occur in primary
seen in 32.8% of our patients. The cut-off levels of dengue infections where these antibodies are slow to
Hb% as a marker for hemoconcentration devised pre- appear (32). Testing GAC-ELISA on paired samples
viously (4, 8) seems to need revision as were many is thus more worthwhile (2, 27). NS1Ag is often only
as 96.9% dengue patients were above cut-off at ad- transiently present in secondary dengue infections
mission. In contrast, however, hemoconcentration and may be masked by antigen-antibody complex-
in excess of 20% and 15% was documented in only es. False positive NS1Ag tests have been detected in
19.8% and 32.4% patients only. The cut-off levels acute Zika virus infection, Cytomegalovirus infection
need to indicate only the at-risk patients (those who and in patients with haematological malignancies
will go on to develop severe dengue or DHF) (8) and (33-35). In light of such discrepancies, clinical di-
not all cases of dengue. Leukopenia occurred in more agnosis in our hospital depends on a combination of
than half of dengue patients; in contrast Shukla V et al. antigen and antibody detection rather than a single as-
found leucopenia in only 10% of cases (21). As noted say.
in our study, transamnitis with ALT and AST greater Close observation and symptomatic management is
than double the upper normal range is documented in all that is required in most cases of simple dengue (2,
majority of patients (16, 21, 23). The AST: ALT ratio 12). Anti-pyretic were nearly universally prescribed
1.42 which is similar to that of 1.8 and 2.0 seen in oth- for these patients while anti-emetics were required in
er studies (21, 23). More severe liver dysfunction as considerable number of patients. Careful intravenous
indicated by hyperbilirubinemia, deranged coagula- fluid replacement remains the cornerstone for man-
tion profile, raised alkaline phosphatise and significant aging severe cases of dengue (2, 12, 17). 167 of our
hypoproteinemia (and hypoalbuminemia) are found in patients received intravenous crystalloids as therapy;
only a minority of cases and indicates severe disease fluid overload was monitored for in these cases. Fluid
(24). Reversible renal dysfunction was also found in overload can lead to serious consequences (17). Two

http://ijm.tums.ac.ir 183
IRAN. J. MICROBIOL. Volume 9 Number 3 (June 2017) 174-185 http://ijm.tums.ac.ir
Shiv Sekhar Chatterjee ET AL .

of our patients required intravenous colloid solutions detection and virus culture was not possible. Further,
while four patients were treated with fresh frozen serotype / genotype identification could not be done
plasma for refractory shock. Most authors report no during the study.
relation between platelet counts and clinical bleeds,
however the duration of shock has important implica-
tions regarding severe bleeding (11). Thus appropriate
monitoring of shock and sequential hematocrit mon- REFERENCES
itoring may have been more appropriate (2, 11, 12).
Active bleeding may be a more appropriate indication 1. Guo RN, Lin JY, Li LH, Ke CW, He JF, Zhong HJ, et
platelet transfusion rather than low platelet count. al. The prevalence and endemic nature of dengue in-
Only 29% of dengue related platelet transfusions in fections in Guangdong, South China: an epidemiologi-
cal, serological, and etiological study from 2005-2011.
our study population had active bleeding while 40.6%
PLoS One 2014; 9: e85596.
patients had no appropriate indications as per NBVD-
2. World Health Organization (WHO) and the Special
CP guidelines (11). Other hospitals in Delhi also re- Programme for Research and Training in Tropical Dis-
port high usage of platelet transfusions, often inappro- eases (TDR) (2009). Dengue guidelines for diagnosis,
priately (3, 5). WHO as well as the Indian NBVDCP treatment, prevention and control. New Edition, Gene-
authorities recommend packed cell transfusion in only va, Switzerland.
those cases who show 10% or more loss of blood vol- 3. Makroo RN, Raina V, Kumar P, Kanth RK. Role of
ume and in those with refractory shock with declining platelet transfusion in the management of dengue pa-
hematocrit (11, 12). In addition to platelets, four pa- tients in a tertiary care hospital. Asian J Transfus Sci
tients required such packed red cell transfusion while 2007; 1: 4-7.
three patients were given fresh frozen plasma in view 4. Gomber S, Ramachandran VG, Kumar S, Agarwal KN,
of coagulopathy. Gupta P, Gupta P, et al. Hematological observations as
diagnostic markers in dengue hemorrhagic fever–a re-
Mortality in our study was 1.2%; additionally there
appraisal. Indian Pediatr 2001; 38: 477-481.
was a case of intra-uterine fetal death. All three dead
5. Chaurasia R, Zaman S, Chatterjee K, Das B. Retro-
patients were male, bread-winners of their family and spective review of platelet transfusion practices during
belonged to 20-55 years working age group. One pa- 2013 Dengue Epidemic of Delhi, India. Transfus Med
tient died due to refractory shock along with myocar- Hemother 2015; 42: 227-231.
ditis and severe hepatic involvement; the second due 6. Daudé É, Mazumdar S, Solanki V. Widespread fear of
to refractory shock with severe blood loss and hepatic dengue transmission but poor practices of dengue pre-
involvement superadded on chronic renal failure and vention: A study in the slums of Delhi, India. PLoS One
diabetes mellitus; and the third due to myocarditis, 2017; 12: e0171543.
cellulitis and staphyloccal sepsis. Severe hepatic in- 7. World Health Organization (1997). Dengue Hemor-
volvement with transaminases greater than 1000 IU/dl rhagic fever: Diagnosis, Treatment, Prevention and
and myocarditis have previously also been associat- Control. 2nd edition, Geneva, Switzerland.
8. Balasubramanian S, Anandnathan K, Shivbalan S,
ed with poor prognosis. The morbidity due to dengue
Datta M, Amalraj E. Cut off Hematocrit for hemocon-
is no less, mean duration of hospital stay was more
centration in Dengue hemorrhagic fever. J Trop Pedi-
than 6 days in our study, mean expenditure per pa- atr 2004; 50: 123-124.
tient due to the illness was greater than US$377; and 9. Benter T, Klühs L, Teichgräber U. Sonography of the
further nosocomial infections complicated the course spleen. J Ultrasound Med 2011; 30: 1281-1293.
in as many as 5 patients. Shepard DS et al. in their 10. Gosink BB, Leymaster CE. Ultrasonic determination
pan-India study conducted from 2006-2012 who not- of hepatomegaly. J Clin Ultrasound 1981; 9: 37-44.
ed an average expenditure of $235.20 in a hospital- 11. Directorate of National Vector Borne Disease Con-
ised patient in private set-up (14). The higher expen- trol Programme, Directorate General of Health Ser-
diture in our study may reflect inflationary changes vices, Ministry of Health & Family Welfare, Gov-
and the fact that Delhi is one of the costliest cities ernment of India (2008). Guidelines for clinical
in India. case management of dengue fever/ dengue hemor-
rhagic fever/ dengue shock syndrome, Delhi, India.
A major limitation in our study is the dependence of
http://nvbdcp.gov.in/Doc/Clinical%20Guidelines.pdf
dengue diagnosis on serological and antigen detection
(last accessed August 13, 2017).
assays only. Due to resource limitations, viral RNA 12. World health Organisation: regional office for

184 IRAN. J. MICROBIOL. Volume 9 Number 3 (June 2017) 174-185 http://ijm.tums.ac.ir

 DENGUE FEVER IN DELHI

South-east Asia (2011). Comprehensive guide- haemorrhagic necrosis of liver is an unpredictable fatal
lines for prevention and control of dengue and den- complication in dengue infection: a postmortem study.
gue hemorrhagic fever. Revised and expanded BMC Infect Dis 2014, 14: 141-150.
edition, New Delhi, India. Available from: URL: 25. Hunsperger EA, Yoksan S, Buchy P, Nguyen VC, Se-
www.searo.who.int/entity/vector_borne_tropical_ karan SD, Enria DA, et al. Evaluation of commercially
diseases/ (last accessed August 13, 2017). available anti-dengue virus immunoglobulin M tests.
13. Anand T, Kumar R, Saini V, Meena G, Ingle G. Knowl- Emerg Infect Dis 2009; 15: 436-440.
edge and use of personal protective measures against 26. Blacksell SD, Jarman RG, Gibbons RV, Tanga-
mosquito borne diseases in a resettlement colony of nuchitcharnchai A, Mammen MP Jr, Nisalak A, et al.
delhi. Ann Med Health Sci Res 2014; 4: 227-232. Comparison of seven commercial antigen and antibody
14. Shepard DS, Halasa YA, Tyagi BK, Adhish SV, Nan- enzyme-linked immunosorbent assays for detection of
dan D, Karthiga KS, et al. Economic and disease bur- acute dengue infection. Clin Vaccine Immunol 2012;
den of dengue illness in India. Am J Trop Med Hyg 19: 804-810.
2014; 91: 1235-1242. 27. De Paula SO, Fonseca BA. Dengue: a review of the lab-
15. Himabindu BL, Arora R, Prashanth NS. Whose prob- oratory tests a clinician must know to achieve a correct
lem is it anyway? Crimes against women in India. Glob diagnosis. Braz J Infect Dis 2004; 8: 390-398.
Health Action 2014; 7: 23718. 28. Andries AC, Duong V, Ngan C, Ong S, Huy R, Sroin
16. Shetty V, Sanil D, Shetty NJ. Insecticide susceptibility KK, et al. Field evaluation and impact on clinical man-
status in three medically important species of mosqui- agement of a rapid diagnostic kit that detects Dengue
toes, Anopheles stephensi, Aedes aegypti and Culex NS1, IgM and IgG. PLoS Negl Trop Dis 2012; 6: e1993.
quinquefasciatus, from Bruhat Bengaluru Mahanaga- 29. Prince HE, Matud JL. Estimation of dengue virus IgM
ra Palike, Karnataka, India. Pest Manag Sci 2013; 69: persistence using regression analysis. Clin Vaccine Im-
257-267. munol 2011; 18: 2183-2185.
17. Chatterjee N, Mukhopadhyay M, Ghosh S, Mondol M, 30. Ngwe Tun MM, Muta Y, Inoue S, Morita K. Persistence
Das C, Patar K. An observational study of dengue fe- of neutralizing antibody against dengue virus 2 after 70
ver in a tertiary care hospital of eastern India. J Assoc years from infection in Nagasaki. Biores Open Access
Physicians India 2014; 62: 224-227. 2016; 5: 188-191.
18. Liu JW, Lee IK, Wang L, Chen RF, Yang KD. The use- 31. Groen J, Koraka P, Velzing J, Copra C, Osterhaus AD.
fulness of clinical-practice-based laboratory data in fa- Evaluation of six immunoassays for detection of den-
cilitating the diagnosis of dengue illness. Biomed Res gue virus-specific immunoglobulin M and G antibod-
Int 2013; 2013: 198797. ies. Clin Diagn Lab Immunol 2000; 7: 867-871.
19. Thomas EA, John M, Bhatia M. Cutaneous manifesta- 32. Hu D, Di B, Ding X, Wang Y, Chen Y, Pan Y, et al.
tions of dengue viral infection in Punjab (north India). Kinetics of non-structural protein 1, IgM and IgG an-
Int J Dermatol 2007; 46: 715-719. tibodies in dengue type 1 primary infection. Virol J
20. Thomas EA, John M, Kanish B. Mucocutaneous mani- 2011; 8: 47.
festations of Dengue fever. Indian J Dermatol 2010; 55: 33. Gyurech D, Schilling J, Schmidt-Chanasit J, Cassinotti
79-85. P, Kaeppeli F, Dobec M. False positive dengue NS1 an-
21. Shukla V, Chandr A. A study of hepatic dysfunction in tigen test in a traveller with an acute Zika virus infec-
dengue. J Assoc Physicians India 2013; 61: 460-461. tion imported into Switzerland. Swiss Med Wkly 2016;
22. Michels M, Sumardi U, de Mast Q, Jusuf H, Puspita 146: w14296.
M, Dewi IM, et al. The predictive diagnostic value of 34. Chung SJ, Krishnan PU, Leo YS. Two cases of
serial daily bedside ultrasonography for severe dengue false-positive dengue non-structural protein 1 (NS1)
in Indonesian adults. PLoS Negl Trop Dis 2013; 7: antigen in patients with hematological malignancies
e2277. and a review of the literature on the use of NS1 for
23. Kittitrakul C, Silachamroon U, Phumratanaprapin W, the detection of dengue infection. Am J Trop Med Hyg
Krudsood S, Wilairatana P, Treeprasertsuk S. Liver 2015; 92: 367-369.
function tests abnormality and clinical severity of den- 35. Fry SR, Meyer M, Semple MG, Simmons CP, Sekaran
gue infection in adult patients. J Med Assoc Thai 2015; SD, Huang JX, et al. The diagnostic sensitivity of den-
98 Suppl 1: S1-8. gue rapid test assays is significantly enhanced by us-
24. Kularatne SA, Imbulpitiya IV, Abeysekera RA, Wa- ing a combined antigen and antibody testing approach.
duge RN, Rajapakse RP, Weerakoon KG. Extensive PLoS Negl Trop Dis 2011; 5: e1199.

http://ijm.tums.ac.ir 185
IRAN. J. MICROBIOL. Volume 9 Number 3 (June 2017) 174-185 http://ijm.tums.ac.ir

View publication stats