False-positive dengue non-structural protein 1 in pregnancy

Abstract

Dengue is a vector-borne infection, which contributes to significant morbidity and mortality in endemic areas. It manifests rapidly within 2 wk from febrile, critical to recovery phase. The point-of-care test (POCT) comprises the non-structural protein 1 (NS1) antigen, IgM and IgG, which aids rapid diagnosis, leading to timely treatment. Despite the high specificity of various POCTs, a few false-positive NS1 cases have been reported. We report the first case of false-positive NS1 antigen in pregnancy. A 24-y-old female in her second trimester of pregnancy presented with fever and chills for 9 d and a non-productive cough for 3 d. The dengue POCT showed NS1 antigen positive, IgM and IgG negative. Her symptoms were resolved with treatment of urinary tract infection with cefuroxime. Urine culture subsequently yielded Escherichia coli. Her recovery and delivery were uneventful. Her NS1 antigen was persistently positive for 18 d. The confirmatory real time-PCR for dengue and pan-Flaviviruses was negative.

Case presentation

A 24-y-old female at 24 wk and 6 d of gestation presented with fever and chills for 9 d. It was associated with a non-productive cough, reduced oral intake and postprandial vomiting for 3 d. There were no coryza symptoms, headaches, arthralgia, myalgia, rashes, retro-orbital pain or abdominal pain. She had no known medical illness and had an uncomplicated first pregnancy 5 y ago.

Upon presentation, her blood pressure was 92/56 mmHg, heart rate 107 beats per min, temperature 37.5°C and saturation 99% on ambient air. The physical examination was unremarkable. Her blood investigations showed a raised white cell count and elevated C-reactive protein (Table 1). A dengue point-of-care test (POCT) was carried out because of recent mosquito fogging in her neighborhood. The dengue POCT showed positive for NS1, and negative for IgM and IgG.

Table 1.

Investigations and results

Day of illness	9	10	13	14	16	18
Hemoglobin (g/dL)	11.7	9.1			9.9
White cells (103/µl)	12.2	13.8			10.1
Platelets (103/µl)	271	237			419
Sodium (mmol/L)	131				134
Potassium (mmol/L)	3.1				4.5
Urea (mmol/L)	2.1				2.5
Creatinine (µmol/L)	57				43
Total bilirubin (µmol/L)					5
Direct bilirubin (µmol/L)					4
Aspartate transferase (U/L)
Alanine transaminase (U/L)					17
Total protein (g/L)					56
Albumin (g/L)					30
Globulin (g/L)					26
Alkaline phosphatase (U/L)	147	195			210
Calcium (mmol/L)	2.22
Magnesium (mmol/L)	0.70
Phosphate (mmol/L)	0.68
International normalized ratio		1.16
Dengue POCT
● NS1	Pos	Pos	Pos	Pos	Pos	Pos
● IgM	Neg	Neg	Neg	Neg	Neg	Neg
● IgG	Neg	Neg	Neg	Neg	Neg	Neg
Culture
a. Blood	No growth
b. Urine	E. coli*
c. Sputum	No growth
Others
i. SARS CoV-2 antigen	Neg
ii. Certest rapid antigens#	Neg
iii. Leptospirosis IgM	Neg
iv. Blood film for malaria parasite	Neg	Neg	Neg
v. Mycoplasma pneumonia antibody			Neg
vi. QIAstat-Dx respiratory panel **			Neg
vii. Dengue virus RT-PCR			Neg
viii. Chikungunya virus RT-PCR			Neg
ix. Zika virus RT-PCR			Neg
x. Japanese encephalitis IgM			Neg
xi. Pan-Flavivirus RT-PCR			Neg

Neg: negative; Pos: positive;

RT-PCR: real time-PCR; SARS-CoV-2: severe acute respiratory syndrome coronavirus 2.

*Susceptible to ampicillin-sulbactam, cefuroxime, gentamicin and nitrofurantoin.

# Certest rapid antigens include Flu A, Flu B, RSV, Adenovirus.

**QIAstat-Dx respiratory panel includes Influenza A, Influenza A subtype H1N1/2009, Influenza A subtype H1, Influenza A subtype H3, Influenza B, Coronavirus 229E, Coronavirus HKU1, Coronavirus NL63, Coronavirus OC43, Parainfluenza virus 1, Parainfluenza virus 2, Parainfluenza virus 3, Parainfluenza virus 4, Respiratory Syncytial virus A/B, Human Metapneumovirus A/B, Adenovirus, Rhinovirus/Enterovirus, SARS-CoV-2, Mycoplasma pneumoniae, Legionella pneumophila, Bordetella pertussis.

She was initiated with intravenous cefuroxime because of the urine nitrite being positive. Her symptoms were resolved after 2 d of inpatient care. Urine culture subsequently yielded Escherichia coli. NS1 antigen remained positive up to day 18 of symptoms. Otherwise, clinical progress and other blood investigations were not suggestive of dengue infection. She was discharged well with urinary tract infection (UTI) treatment. She delivered a healthy child uneventfully at 37 wk of gestation. Dengue and pan-Flavivirus real-time PCR (RT-PCR) were sent to the reference laboratory, which showed negative results later.

Discussion

This was a case of false-positive NS1 antigen on dengue POCT. A dengue POCT was offered based on acute febrile illness with an epidemiological link to endemic areas. She did not have symptoms or blood investigations that suggestive of dengue infection. An Abbott Bioline Dengue Duo kit was used. It is a rapid lateral flow immunochromatographic test with a sensitivity and specificity of >98%. There was no concomitant testing using other POCTs because of the unavailability of other POCTs locally. The local guideline requires all NS1 antigen positive samples to be confirmed with dengue RT-PCR at the reference laboratory. For this case, the sample was also sent to the reference laboratory for additional pan-Flavivirus RT-PCT, which also yielded negative results.

Maternal fever is an emergency that demands immediate investigation and treatment to avoid complications or death. UTI and pyelonephritis are common causes of maternal fever.¹  Escherichia coli, Klebsiella spp., Proteus mirabilis and Pseudomonas aeruginosa are the common isolates from urine culture.¹ In dengue-endemic regions, pregnant mothers are affected equally. Dengue is a vector-borne infection transmitted by Aedes species mosquitoes. Dengue fever in pregnancy is associated with a 3–4 times increased incidence risk of developing severe dengue, which can result in increased maternal mortality and preterm delivery.² A single-center study in the dengue-endemic area reported that the two main causes of maternal fever were dengue (24.3%) and UTI (12.2%).¹ Among these, 31.8% had severe dengue and 50% had warning signs, which resulted in 15.9% (n=7) maternal death and 81.8% fetal or neonatal deaths.¹

Rapid diagnosis is crucial as dengue manifests quickly through febrile, critical to recovery phases. Prompt supportive therapy is needed to prevent complications. The dengue POCT consists of NS1 antigen, IgM and IgG. NS1 antigen presents from day 1 of infection up to day 9. NS1 antigen sensitivity ranged from 42.9 to 64.3%, specificity as 99.2–100%, positive predictive value as 85.7–100.0% and negative predictive value as 93.9–96.1%.³ In combination with IgM or IgG, the sensitivity improved.³ Dengue confirmatory tests via dengue RT-PCR and pan-Flavivirus RT-PCR. RT-PCR is usually carried out in reference laboratories and not at the service laboratory of the hospital. Therefore, delayed return of results is expected for RT-PCR tests. Few cases of false-positive NS1 have been reported. The causes include communicable and non-communicable diseases.

Cross-reactivities of NS1 antigen in POCTs have been reported for Japanese encephalitis virus (JEV) and zika virus (ZIKV).^4,5 Dengue virus, ZIKV and JEV are from the same genus Flavivirus. Besides clinical features, the three viruses also share similarities of the NS1 antigen. Furthermore, ZIKV shares the same vectors as the dengue virus. The Zika outbreak in Brazil raised awareness that infection during the first or second trimester can cause birth defects such as microcephaly and congenital blindness.⁶ The diagnostic test for ZIKV is ELISA-based ZIKV serology.⁶ ZIKV-NSI-IgM declines rapidly after 29 d of infection; therefore, the sensitivity is 79% during the first month and 68% up to 2 mo postinfection.⁶ Although Zika is not endemic in Malaysia, its impact on pregnant patients must be considered due to its serious implications. Besides, the same vector potentially transmits the virus in tropical and subtropical regions. The Malaysian service laboratory only offered RT-PCR, which might impede the local diagnostic capability of Zika.

JEV is transmitted via Culex species mosquitoes. Sarawak in Malaysian Borneo had a JEV outbreak in 1998 that, in 2000, brought about an important public health intervention, the incorporation of the JE vaccine into the childhood vaccination schedule.⁷ Most infections are either asymptomatic or cause only a transient fever, leading to underdiagnosis. However, encephalitis cases can result in up to 30% mortality and 50% disability among survivors.⁷ JEV IgM is the diagnostic test of choice. Negative JEV IgM in pairs at least 7 d apart ruled out JE infection. In this case, the second JEV IgM result was available due to miscommunication with the laboratory.

Two case reports of hematological malignancies, namely acute myeloid leukemia and natural killer cell lymphoma.⁸ One patient presented with a prolonged fever for 1 mo, which was associated with weight loss, and the second patient had white cells of 43×10⁹/liter, which were not typical for dengue.

UTI and/or pyelonephritis coinfection with dengue should be considered in dengue-endemic areas. Both conditions present with maternal fever and early symptoms are often non-specific.⁹ In this case, the patient presented with non-specific febrile illness with leukocytosis and hyponatremia. Serial monitoring of basic blood investigations did not support the diagnosis of dengue. As our center does not offer in house RT-PCR for dengue and pan-Flavivirus, the clinician made the diagnosis of treating for UTI and not for dengue based on basic blood investigations and clinical progress. The urine culture yielded E. coli on the third day of culture and clinical improvement confirmed the diagnosis of UTI. Persistently positive NS1 antigen up to 18 d and persistently negative IgM and IgG ruled out dengue infection. RT-PCR of dengue and pan-Flaviviruses at reference laboratories returned 2 mo later supported the diagnosis.

Conclusions

UTI and dengue fever are important and treatable causes of maternal fever. UTI and dengue coinfection should be considered in endemic areas. As the treatment of UTI and dengue varies, a definitive diagnosis is dependent on the laboratory tests. Close monitoring and treatment based on clinical progress are needed if the confirmatory investigations are delayed. An alternative diagnosis of NS1 antigen should be considered if the patient's clinical course does not correspond with the clinical course of dengue infection. False-positive NS1 antigen can occur in other Flavivirus infections or other conditions, such as hematological malignancies.

Authors’ contributions

TS and YO carried out clinical management. YO and TW drafted the manuscript. TS, YO and TW critically revised the manuscript for intellectual content. All the authors read and approved the final version of the manuscript.

Funding

None.

Competing interests

None declared.

Ethical approval

This study was approved by the Ministry of Health Medical Research Ethics Committee under the National Medical Research Registry of Malaysia (NMRR-20-1699-55994). Written consent was obtained from the patient for writing this manuscript.

Data availability

The data underlying this article are available in the article and in its online supplementary material.