A Case Suspected for Yellow Fever Vaccine-Associated Viscerotropic Disease in the Netherlands

Authors


Abstract

Yellow fever (YF) 17D vaccine is one of the most successful vaccines ever developed. Since 2001, 56 cases of yellow fever vaccine-associated viscerotropic disease (YEL-AVD) have been published in the peer-reviewed literature. Here, we report a new case suspected for YEL-AVD in the Netherlands. Further research is needed to determine the true incidence of YEL-AVD and to clarify host and vaccine-associated factors in the pathogenesis of YEL-AVD. Because of the potential adverse events, healthcare providers should carefully consider vaccination only in people who are truly at risk for YF infection, especially in primary vaccine recipients.

Yellow fever (YF) virus, a mosquito-borne flavivirus, is endemic to sub-Saharan Africa and tropical areas of South America. The majority of YF virus infections in humans are asymptomatic but severe disease can present with jaundice and hemorrhagic fever, and a mortality rate of 20% to 50% can occur.[1] Because of the potential lethality of YF virus infections, vaccination with the live attenuated 17D YF vaccine is advised for individuals ≥9 months old who are traveling to endemic areas.[2] Over 600 million doses of 17D vaccine have been administered to humans since its development in 1937.[3] In general, the vaccine is well tolerated and has an acceptable safety profile.[4] Adverse reactions are typically mild and occur 2 to 11 days post vaccination. About 25% of vaccine recipients develop discomfort at the injection site, myalgia, headache, and low-grade fever.[4, 5] However starting in 2001, case descriptions of severe yellow fever vaccine-associated viscerotropic disease (YEL-AVD) after administration of the 17D vaccine were published. YEL-AVD was described as a condition that mimicked natural acquired YF.[6-8] As of July 2013, a total of 56 cases have been identified by the Brighton Collaboration Viscerotropic Disease Working Group.[9, 10]

Case Report

A 38-year-old man received his first YF vaccine (ARILVAX®, Chiron Vaccines Ltd, Liverpool, UK), a booster of revaxis diphtheria/tetanus/polio (DTP), and third TWINRIX® vaccine (GlaxoSmithKline BV, Brentford, Middlesex, UK) on October 22, 2013, in preparation for travel to Gambia. He had a medical history of only nasal congestion caused by hyper-reactivity. He did not take any medications and reported no allergies. Six hours after vaccination, he developed malaise, headache, arthromyalgia, nausea, and abdominal pain. His symptoms became progressively worse and 2 days later, he sought medical care because jaundice had occurred. He noticed pale stools and dark-colored urine.

On admission, he had a blood pressure of 116/77 mmHg, heart rate of 55 per minute, and temperature of 36.9°C. Other than scleral icterus, no abnormalities were found on physical examination. Initial laboratory tests revealed a white blood cell count (WBC) of 3.8 × 10E9 /L with lymphocytopenia (0.31 × 10E9 /L), thrombocytopenia (platelet count 59 × 10E9 /L), and normal hemoglobin count (Hb 10.4 mmol/L). Furthermore, C-reactive protein of 97 mg/L, elevated liver enzymes [aspartate aminotransferase (AST) 56 U/L, alanine aminotransferase (ALT) 88 U/L, total bilirubin 119 µg/L, and direct bilirubin 90 µmol/L], mild hyponatremia (Na 134 mmol/L), normal kidney function (creatinine 80 µmol/L), elevated D-dimer 19,860 µg/L, and normal fibrinogen of 3.5 g/L were detected. Abdominal ultrasonography showed splenomegaly and minimal hepatomegaly. The patient received intravenous fluids and was treated with analgesics. During his hospital stay, he was afebrile but developed slight pain in the right upper abdomen.

The patient's bacterial blood cultures were negative. Serological studies were negative for human immunodeficiency virus, hepatitis A–B–C–E-virus, Epstein–Barr virus, cytomegalovirus, parvovirus, toxoplasmosis, leptospirosis, and brucellosis. During hospitalization, his WBC, platelets, and liver function tests showed an increase, while his bilirubin levels decreased (Table 1). Six days after admission he was discharged with only abdominal pain as the residual symptom. Two weeks later, laboratory test results were normalized and the abdominal pain had resolved.

Table 1. Evolution of selected hematological and biochemical parameters in a patient with suspicion of yellow fever vaccine-associated viscerotropic disease
 Days from vaccination
Day 2Day 3Day 4Day 5Day 6Day 7Day 8Day 20
  1. WBC = white blood cell; Platelets: 150–400 × 10E9 /L; C-reactive protein: <10 mg/L; Total bilirubin: <17 µmol/L; Direct bilirubin: <15 µmol/L; AST = aspartate aminotransferase: <35 U/L; ALT = alanine aminotransferase: <45 U/L; AF = alkalic phosphatase: <120 U/L; GGT = gamma glutamyl transferase: <55 U/L; LD = lactate dehydrogenase: <250 U/L; CK = creatin kinase: <170 U/L; Fibrinogen: 2.0–4.0 g/L; D-dimer: <500 µg/L.

Temperature (°C)36.937.436.436.936.236.8
WBC (×10E9 /L)3.84.75.98.29.09.85.8
Neutrophils (%)7875585773
Lymphocytes (%)816313115
Platelets (×10E9 /L)59687993129245317
C-reactive protein (mg/L)975724128
Total bilirubin (µmol/L)11975554032252017
Direct bilirubin (µmol/L)90674635261917
AST (U/L)5662759216412611120
ALT (U/L)889412716628229428830
AF (U/L)7610112914517217717580
GGT (U/L)9610012013213112312352
LD (U/L)205175200174185
CK (U/L)58
Fibrinogen (g/L)3.5
D-dimer (µg/L)19,860

YF virus genome detection and viral load quantification in serum, obtained at days 3 and 10 post vaccination, were determined by real-time polymerase chain reaction (RT-PCR). No YF virus could be detected in blood.

Discussion

YEL-AVD was first reported in the literature in 2001.[6-8] It was considered to be a recent phenomenon until Engel and colleagues published a case that had occurred in a Brazilian woman vaccinated in 1975.[11] YEL-AVD is believed to result from widespread dissemination and replication of the live attenuated 17D YF vaccine virus and is described only in primary vaccine recipients.[2, 3] The estimated frequency of YEL-AVD among US and European travelers is 0.3 to 0.4/100,000 YF vaccine doses distributed with a higher rate for people aged ≥60 years (1.0/100,000 doses) and ≥70 years (2.0/100,000 doses).[4, 12, 13] The median time from vaccination to symptom onset is 3 days (range 1–8 days). The case-fatality ratio for all reported cases is 63%.[2] Advanced age[12, 14, 15] together with a history of thymus disease[16] have been identified as risk factors for developing YEL-AVD after primo vaccination.

A variety of clinical pictures of YEL-AVD have been described.[7, 17] The initial symptoms are nonspecific, including fever, headache, arthromyalgia, nausea, vomiting, and diarrhea. As the illness progresses jaundice can appear, together with laboratory abnormalities (thrombocytopenia, and elevation of total bilirubin, hepatic transaminases, and creatinine). Patients with severe YEL-AVD may develop hypotension and hemorrhage, as well as acute renal and/or respiratory failure. Less frequent manifestations of severe YEL-AVD are coagulopathy (elevated prothrombin time or activated partial thromboplastin time, elevated fibrin degradation products) and rhabdomyolysis.[2] The only therapy for YEL-AVD is symptomatic, assuming that other potential diseases have been excluded.

Although our patient was young and had no history of thymus disease, the case described strongly suggests YEL-AVD. The clinical picture of nonspecific symptoms within 2 days of vaccination (malaise, headache, arthromyalgia, nausea, and abdominal pain), jaundice, and abnormal laboratory findings (thrombocytopenia, elevated bilirubin and liver enzymes, elevated fibrin degradation) are consistent with other cases of YEL-AVD.[9] These symptoms have not been reported after revaxis DTP vaccination. GlaxoSmithKline reported that headache can occur in >10% of recipients, nausea and abdominal pain in 1% to 10%, and arthromyalgia in <0.01% after TWINRIX® vaccination. It is imaginable that the TWINRIX® vaccination may have contributed to these symptoms, but this cannot be confirmed. Physical examination, laboratory testing, serological studies, and abdominal ultrasound excluded other potential causes such as cholelithiasis or viral illness. Cholelithiasis is unlikely because our patient did not experience colic pain, and hepatitis developed with decreasing bilirubin levels. Furthermore, C-reactive protein level was elevated, but there were no signs of cholecystitis. Coincidental viral illness was excluded by negative findings on serological studies.

Finally, no bacteria were identified in blood specimens. This strongly suggests that the YF virus was the causal agent of the clinical events.

The Brighton Collaboration Viscerotropic Disease Working Group developed criteria for case definition in viscerotropic disease in 2012. Three levels of diagnostic certainty for viscerotropic disease have been established (Table 2),[9] with level 1 having the highest degree of specificity and the lowest degree of sensitivity for viscerotropic disease. These criteria can be applied when there is temporal association between symptoms and vaccination, and no clear alternative diagnosis can account for the symptoms. Our patient meets three major criteria: total bilirubin ≥1.5× upper limit of normal (ULN), platelet count <100,000/μL, and coagulopathy (elevation of D-dimer). On the basis of these criteria, level 1 evidence is established in case definition of viscerotropic disease.

Table 2. Major and minor criteria used in the case definition of viscerotropic disease
 Major criteriaMinor criteria
  1. Level 1 of diagnostic certainty—≥3 major criteria.

  2. Level 2 of diagnostic certainty—2 major criteria or 1 major criterion and ≥2 minor criteria.

  3. Level 3 of diagnostic certainty—≥3 minor criteria or 1 major criterion and 1 minor criterion.

  4. ALT = alanine aminotransferase; AST = aspartate aminotransferase; CPK = creatine phosphokinase; INR = international normalized ratio; ULN = upper limit of normal

HepaticTotal bilirubin ≥1.5× ULN (≥1.5× patient's baseline value if known, or ALT or AST ≥3× ULN (≥3× patient's baseline value if known)Jaundice
RenalCreatinine ≥1.5× ULN (≥1.5× patient's baseline value if known)

Urine output <500 mL/24 h for adults

Urine output <0.5 mL/kg for children

Positive urine dipstick test for blood with a negative urine microscopy exam for red blood cells

MusculoskeletalCPK ≥5× ULN 
RespiratoryOxygen saturation ≤88% on room air or requirements for mechanical ventilation

Increased respiratory rate for age:

6–11 months: >50 per minute

1–5 years: >40 per minute

≥6 years: >20 per minute

Platelet disorderPlatelets <100,000/µLPetechiae or purpura
HypotensionRequirements for vasopressor drugs to maintain systolic blood pressure

Systolic blood pressure <90 mmHg for adults

Systolic blood pressure <5th percentile for age in children <16 years

CoagulopathyINR ≥ 1.5 or prothrombin time ≥1.5× ULN or activated partial thromboplastin time ≥1.5× ULN or elevated fibrin degradation products or hemorrhage from more than one siteClinically evident hemorrhage: epistaxis, hematemesis, melena, hematochezia, hemoptysis, metrorrhagia or menorrhagia, gingival hemorrhage, persistent bleeding from needle puncture sites

Primary vaccine recipients often develop a low-level viremia following YF vaccination. The viremia usually occurs within 3 to 7 days post vaccination and persists for 1 to 3 days. However, RT-PCR could not detect the YF virus in the serum of our patient. This method was used with success by Reinhardt and colleagues.[18] By contrast, dos Santos and colleagues[19] were not able to detect the YF virus by RT-PCR. Although a longer duration of viremia is often reported in YEL-AVD patients, it might be that the viremia was present between days 4 and 9, whereas the serum was obtained at day 3 and 10. Belsher and colleagues[20] developed causality criteria for YEL-AVD. These criteria can be used when the case is an adverse event following YF vaccination, when symptom onset is within 10 days of vaccination, and when there is no laboratory evidence for an alternative diagnosis. In our case, the data to determine YF vaccine-associated causality were insufficient as the YF virus test results did not meet any of the criteria for definite, probable, or suspected causality. Hence, the correlation between symptoms and the YF virus could not be determined. Nevertheless, YEL-AVD is, in our patient, the most likely diagnosis because of the time course of the disease and the exclusion of other potential conditions by comprehensive laboratory testing, serological studies, blood cultures, and abdominal ultrasound.

Any illness starting with nonspecific symptoms within 10 days of receiving 17D vaccine should be intensively investigated for possible YEL-AVD. The increased attention to safety of the 17D vaccine since 2001 might be responsible for the higher incidence of YEL-AVD reports in the last 12 years. More research is needed to determine the true incidence of viscerotropic disease according to the Brighton Collaboration criteria and to clarify host- and vaccine-associated factors in the pathogenesis of YEL-AVD. Healthcare providers should carefully consider, in the case of each traveler, whether the 17D vaccine should be administered. They must take multiple factors into account: not only the risk of travel-associated YF virus infection with high morbidity, mortality, and country-entry requirements, but also the risk of serious adverse events following vaccination.

Declaration of Interests

The authors state they have no conflicts of interest to declare.

Ancillary