Share this article on:

Chikungunya Virus Infections Among InfantsWho Classification Not Applicable

van Keulen, Veronique MD*; Huibers, Minke MD*†; Manshande, Meindert MD*; van Hensbroek, Michaël Boele MD, PhD; van Rooij, Linda MD, PhD*

The Pediatric Infectious Disease Journal: March 2018 - Volume 37 - Issue 3 - p e83–e86
doi: 10.1097/INF.0000000000001826
Brief Reports

Chikungunya virus caused an epidemic on Curaçao in 2014–2015. Infants are highly at risk for clinical syndromes as sepsis-like illness and central nervous system disease. Clinical recognition is important if laboratory test, polymerase chain reaction and enzyme-linked immunosorbent assay are not available. The World Health Organization created criteria for identification of probable Chikungunya virus. None of the studied infants met these criteria. We believe the criteria suggested by World Health Organization should be modified for infants.

From the *Department of Pediatrics, St. Elisabeth Hospital, Willemstad, Curaçao; and Global Child Health Group, Emma Children’s Hospital, Academic Medical Centre (AMC), Amsterdam, The Netherlands.

Accepted for publication November 7, 2017.

The authors have no funding or conflicts of interest to disclose.

V.K. and M.H. contributed equally.

Address for correspondence: Veronique van Keulen, MD, Johannes Vermeerplantsoen 31, 2251 GR, Voorschoten, The Netherlands. E-mail:

The Chikungunya virus (ChikV) is an arthropod-borne virus transmitted by the Aedes mosquito. In 2014 more than 1 million probable ChikV cases were documented in the Caribbean.1 Presenting symptoms of ChikV in children are sudden onset of fever, severe acute polyarthralgia, myalgia and erythema.1 Clinical syndromes in infants are poorly described, but case series and case reports document sepsis-like illness, central nervous system disease and bullous dermatitis.1 ChikV can be diagnosed by means of polymerase chain reaction, virus-specific RNA detection and enzyme-linked immunosorbent assay.1–3 If these tests are not available, the World Health Organization (WHO) suggests using their clinical algorithm, which is designed to identify patients with a probable ChikV infection.4

In this case series, we describe the symptoms of ChikV-infected young infants not infected as a result of mother-to-child transmission during the ChikV epidemic of 2014–2015 on the island of Curaçao. Secondly, we evaluate the applicability of the WHO criteria for ChikV infections in this specific patient population.

Back to Top | Article Outline


Infants from 1 week to 6 months of age, who were admitted with fever to the pediatric ward of the St. Elisabeth Hospital, Willemstad, Curaçao, between September 1, 2014 and January 31, 2015, with positive antibody tests for ChikV, were included. Infants younger than 1 week of age were excluded due to the possibility of mother-to-child transmission.

The included infants were evaluated according to the WHO criteria for ChikV infections.4 Probable cases are defined as acute onset of fever (>38.5°C) with severe arthralgia or arthritis, not explained by other medical conditions and residing in an area where ChikV was reported. Confirmed cases have positive ChikV tests.4

To estimate the risk of severe bacterial infection (SBI) and identify the severity of the symptoms, all children were evaluated according to the Rochester criteria5 (Table 1). These criteria are evolved from studies designed to test the hypothesis that well-appearing febrile infants who meet defined history, physical examination and laboratory criteria are unlikely to have SBI.5 Infants were observed for at least 3 days on the pediatric ward. Medical information and clinical syndromes were reported. Irritability was defined as restless activity with altered mental status (ie, inconsolable crying) and/or opisthotonus. Compromised circulation was defined as cold peripheral extremities, capillary refill time >3 seconds, tachycardia and/or anuria for more than 6 hours, and respiratory distress was defined as blood saturations below 93%, tachypnea and/or signs of dyspnea.



In accordance with Dutch guidelines for fever of unknown origin, infants <2 months of age with fever underwent a full sepsis workup (including blood cultures, full blood count, C-reactive protein, lumbar puncture for cerebrospinal fluid [CSF] cultures and cell count for signs of infections [white blood cell count, red blood cell count, protein levels, glucose levels and checking for pleocytosis] and urine sampling for cultures and screening). All infants >2 months of age underwent blood sampling to screen for SBI. Depending on the clinical presentation, a lumbar puncture and urine collection were considered. Suspected bacterial infections were treated with antibiotics, regardless of ChikV probability.

The diagnosis of ChikV was confirmed via the detection of serum-specific IgM antibodies by applying the enzyme-linked immunosorbent assay technique (Euroimmune AG, Lubeck, Germany) to an acute blood sample. A second sample, collected at least 2 weeks later, with a quantitative 4-fold or higher rise in immunoglobulin G (IgG) titer, confirmed the diagnosis. Cutoff values were IgM/IgG values above 1.1 g/L (“accepted positive”), values from 0.8 to 1.0 g/L (“dubious positive”) and below 0.8 g/L (“negative”).2

Infants were followed up to 6 months after admission at the outpatient clinic by their own pediatrician for a clinical (neurologic) examination. Data were entered and analyzed using SPSS (IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp). The study was approved by the Research Ethics Committee of the St. Elisabeth Hospital, Curaçao.

Back to Top | Article Outline


Twenty-two infants were included, all with fever, which persisted for 2–7 days (mean, 3 days). Table 2 shows all baseline characteristics. Irritability was seen in 77% of cases (17/22). Skin lesions as maculopapular rash was seen in 64% (14/22) and bullae/vesicles in 14% (3/22). No acrocyanosis or limb edema was seen in our case series. Compromised circulation was seen in 36% (8/22) and respiratory distress in 14% (3/22). Feeding difficulties, needing nasal tube feeding, in 12/22 (55%). In 80% of the infants, the Rochester criteria showed increased risk of SBI.



Leukopenia (<6 × 109/L) was seen in 23% (5/22) and leukocytosis in 5% (1/22), while none of the infants were found to have thrombocytopenia (<150 × 109/L). C-reactive protein was raised in 41% (9/22) (>0.5 g/L; median 1.5, interquartile range 2.83). CSF showed no signs of pleiocytosis and normal protein and glucose levels. All cultures of CSF (performed in 64% of cases) and blood (performed in 95% of cases) were negative. Two infants were diagnosed with urinary tract infections based on a positive urine culture.

Intravenous antibiotics were started in 82% (18/22) of the infants because of clinically suspected SBI. Respiratory support (oxygen flow using nasal cannula) and volume resuscitation (fluid boluses of NaCl 0.9% 10–20 mL/kg) were needed in, respectively, 9% (2/22) and 36% (8/22) of the infants.

At 6-month follow-up, a clinical examination was performed. This included physical neurologic examination and screening for development disorders by interviewing the parents. The results were reassuring.

Back to Top | Article Outline


In this case series, we describe severe symptoms of ChikV-infected young infants, which are difficult to distinguish from bacterial infections or encephalitis. Secondly, this is the first clinical report on the application of the WHO criteria among hospitalized ChikV-infected infants during an epidemic. We used the WHO criteria to identify infants with a probable ChikV infections in all cases, but none of the 22 included infants fulfilled these criteria.4 While arthralgia and/or arthritis among ChikV-infected adults are common and serve as one of the leading symptoms of the WHO classification, the prevalence among infants is only 30%–50%.1 , 4 In our cohort, these symptoms were not reported, no limb edema or signs of synovitis were seen. However, identification of specific pain as sign of arthritis is difficult in infants. Irritability or a proxy of altered mental status could be a sign of pain, but could also be an indication of a sepsis-like illness or neurologic disease like encephalitis.6 The incidence of irritability in our group was 70%; also in the absence of pain grading and limb edema, we cannot state, with certainly, that arthralgia or arthritis was absent in our group. However, due to this uncertainty, these criteria are not usable in young infants.

Skin abnormalities are another leading symptoms of ChikV infections.7 , 8 Maculopapular erythema is reported in approximately half of the adult population and to a lesser extent in children.7 Symptoms occur up to 24–48 hours after the start of fever.1 , 7 We found skin lesions in 64% of the infants.

In contrast to adults and older children, ChikV infection in infants causes severe symptoms.1 , 4 The Rochester criteria were developed and are widely used to subdivide febrile infants into groups with a high or a low risk for SBI, with a sensitivity of 92%.5 In our study, only 20% of the infants showed no increased risk for SBI based on the Rochester criteria. However, the urine culture was positive in only 2 patients; all other cultures of CSF and blood remained negative. Hence, because 80% of the infants met criteria for a SBI, no significant clinical differentiation between SBI and ChikV infections could be made. Starting antibiotics in this group is therefore validated because bacterial infections can be found among serology-positive ChikV infants.9

Previous studies describe neurologic complications in children in 14%–25% of cases.1 , 6 In our cohort, no clinical neurologic sequelae were found in 6 months of follow-up (data not shown). Gérardin et al6 describes a 3-year outcome of ChikV-associated encephalitis, with a better prognosis for infants than adults. Complications after 3 years exist of behavioral changes and neurocognitive impairments for neonates infected by mother-to-child transmission. Not much is known for ChikV-infected infants, so follow-up is imported. Our research group is currently conducting additional research concerning long-term complications.

Laboratory diagnostics are not helpful in early disease recognition of our ChikV-infected infants. Forty-one percent of our cases showed a slight increase of C-reactive protein (maximum 11 mg/dL). Thrombocytopenia is described as common in neonates and found in infants,9 but was not found in our cohort. Rapid diagnostic tests therefore remain vital. However, ChikV is often seen in countries where tests are not available due to a lack of resources, patient overloads or practical problems. Hence, a good clinical recognition of this disease is most important.

Our study results are based on a cohort of 22 hospitalized infants. Confirming ChikV infections through RNA detection with polymerase chain reaction is more sensitive in comparison with the antibody detection used in our study. Unfortunately, this was not possible at our hospital and is a limitation of our study. Because of the low sensitivity of the antibody test, the number of infected infants and the size of the ChikV problems in infants may have been underestimated. A second sample after the convalescent phase would have been helpful, but this was impossible due to the enormous test loads during the epidemic and running out of the serology test.

Sensitive clinical criteria are needed in settings where rapid diagnostic tests, such as polymerase chain reaction, are not available, for early recognizing of ChikV infections (and neurologic complications), to prevent overtreatment with antibiotics and overcrowded hospital wards. More research is needed to determine the sensitivity and specificity of the WHO criteria, which were not useful in our population.

Back to Top | Article Outline


Symptoms of ChikV infections in young infants are poorly described. Symptoms are difficult to distinguish from bacterial infections and clinical syndromes involving the central nervous system can occur and should be recognized. When fast laboratory tests are not available, clinical recognition of ChikV during an epidemic is important. The WHO criteria for probable ChikV infection may be inappropriate for this specific group of patients. We suggest that the criteria are reconsidered and modified.

Back to Top | Article Outline


1. Ritz N, Hufnagel M, Gérardin P. Chikungunya in children. Pediatr Infect Dis J. 2015;34:789–791.
2. Grivard P, Le Roux K, Laurent P, et al. Molecular and serological diagnosis of Chikungunya virus infection. Pathol Biol (Paris). 2007;55:490–494.
3. Prince HE, Seaton BL, Matud JL, et al. Chikungunya virus RNA and antibody testing at a National Reference Laboratory since the emergence of Chikungunya virus in the Americas. Clin Vaccine Immunol. 2015;22:291–297.
4. World Health Organisation. Epidemiology and case definition Chikungunya. 2012. Available at: Accessed September 2014.
5. Jaskiewicz JA, McCarthy CA, Richardson AC, et al. Febrile infants at low risk for serious bacterial infection–an appraisal of the Rochester criteria and implications for management. Febrile Infant Collaborative Study Group. Pediatrics. 1994;94:390–396.
6. Gérardin P, Couderc T, Bintner M, et al; Encephalchik Study Group. Chikungunya virus-associated encephalitis: a cohort study on La Réunion Island, 2005-2009. Neurology. 2016;86:94–102.
7. Valamparampil JJ, Chirakkarot S, Letha S, et al. Clinical profile of Chikungunya in infants. Indian J Pediatr. 2009;76:151–155.
8. Robin S, Ramful D, Zettor J, et al. Severe bullous skin lesions associated with Chikungunya virus infection in small infants. Eur J Pediatr. 2010;169:67–72.
9. Elenga N, Folin M, Vandamme YM, et al. Chikungunya infection in hospitalized febrile infants younger than 3 months of age. Pediatr Infect Dis J. 2017;36:736–740.

Chikungunya; arthropod-borne virus; infants; Americas; WHO

Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.