Evaluation of the neonate is often a challenge. Neonates (children less than one month of age) have immature immune systems and are at higher risk for serious complications of bacterial and viral infections, including enteroviral diseases. Based on the high morbidity and mortality in this age group, the current standard of care for a febrile or irritable neonate includes a general sepsis workup including blood, urine, cerebral spinal fluid (CSF) cultures, laboratory studies, hospital admission, and intravenous (IV) antibiotics pending culture results.
More difficult is the evaluation of a neonate who presents with mild and nonspecific symptoms that are consistent with upper respiratory or viral processes. Infants with enteroviral infections, which include coxsackie, poliovirus, and echoviruses, vary in symptomology from self-limited disease to generalized multi-system organ failure and sepsis. Because of the variation in presentation as well as the difficulty in diagnosis, controversy exists concerning both the diagnosis and management of Enterovirus. Given the possible severe sequelae of enteroviral infection in neonates, this issue of Pediatric Emergency Medicine Practice explores the controversy of taking an evidence-based approach. The clinical presentation of enteroviral infection and the potential complications arising from the infection will be described. Furthermore, the diagnostic evaluation of the newborn baby including the utility of viral studies and treatment availability will be discussed.
Click here to download a PDF of the Evidence-Based Practice Recommendations for this issue.
A baby girl with perioral cyanosis and respiratory distress is rushed in from triage. As the baby is hooked up to the monitors and given supplemental oxygen, your nurses try to calm the family while you assess the newborn. As your thoughts run through the different causes of respiratory distress in an infant, you pick up on some of the story that the father is trying to tell. "This has been going on for a few days," he cries. "No one took us seriously, but I knew that this was more than just a cold!" The mother is sobbing in the background, claiming that now her baby will have to stay in the hospital again. You note that this baby is indeed in distress. The baby's RR is 70 and her HR is 180 beats per minute. While your nurses obtain IV access on the baby, you order a portable chest x-ray and ask the father, "What has been going on? When was the baby in the hospital?" He takes a deep breath and begins to tell you that his wife had a perfectly normal pregnancy but delivered at 36 weeks. The baby was doing well and was discharged home from the NICU 3 days later. At 4 days of life, the baby began to show signs of congestion and was not feeding well. The mother interjects that her older son was a "better eater" in comparison to her little girl. They took the baby to their local pediatrician, who advised them that the baby had some congestion and taught them how to use a nasal aspirator. They were assured that every baby is different and that their daughter simply may not nurse as well as their son. The following day, they returned to the pediatrician, who suggested that perhaps the mother did not have enough breast milk and that the baby was latching for prolonged periods because she was not satisfied. Again the baby was sent home, and an attempt was made to give some formula. Throughout the night, the baby's condition worsened. They called the pediatrician's service and were asked several questions by the nurse on call. The baby was afebrile, had been making urine, and the calculated respiratory rate according to the nurse's instruction was within the normal range. They were told to follow up with their doctor. By morning, the baby started to develop "weird" breathing while taking the bottle, and her color just did not look right. Frustrated and scared, her parents thought they would bring the baby to the hospital where she was born.
As the father finishes his story, you look at the patient's chest x-ray and ask the resident to call the pediatric cardiologist on call. You note the enlarged heart and ask for an electrocardiogram (ECG) and echocardiogram (ECHO). You explain to the baby's father that the baby has cardiomegaly. You know his next question is going to be "from what?" You do not have the answer. You begin to think of all of the possible causes of heart failure, and as you examine the neonate you begin to give some orders. You know that the baby may soon need to be intubated. You begin to discuss the differential with the parents, and as you mention viral myocarditis, the mother begins to shake uncontrollably. The father sees her response and asks "What is that? What causes it? Who could she have gotten that from? When would she have gotten that?" Finally, the mother looks up and says, "I had a really bad cold during the delivery, could she have gotten this from me?" How do you answer? Though you are unsure of the cause, you are also unsure of a favorable prognosis…
A literature review using PubMed and OVID was performed with keywords including enteroviral disease, neonatal infection, coxsackie in neonates, and viral infections in neonates. One hundred thirty-seven references were reviewed for this report and 85 are included. Although there is much literature about the septic neonate (with much of the discussion regarding bacterial causes and some discussion regarding herpesvirus) there is relatively less information that explicitly discusses the neonate infected with enteroviral disease.1 Despite extensive review, it was difficult to ascertain the incidence of enteroviral disease. The limitation in determining the true incidence of enteroviral disease stems from the ability of neonates to be asymptomatic with self-resolving infections. Without testing every neonate, it is impossible to predict the true number of neonates that are harboring infection and, of those infected, how many would eventually progress to worse disease. Clinicians often test only the most ill-appearing neonates, which distorts the true incidence and range of clinical presentations that can be studied.2 While some reports were available through the CDC, case reports provided much information on the severity of enteroviral disease, potential methods of transmission, and complications associated with specific serotypes.
Clinical guidelines, prediction rules, and standards of care have been created for the septic neonate, with the goal of diagnosing potentially treatable diseases; specific attention has not been given to enteroviral diseases.3-8 Due to the overlap of clinical presentation, the neonate who may have an enteroviral infection is subject to a workup that would rule out other treatable causes of infection without actually focusing on the diagnosis of enteroviral disease. The guidelines do not include established rules to diagnose patients with enteroviral disease. Some discussion of viral infections is used to consider placing those patients with viral infections at lower risk for serious bacterial infections, yet they are limited in their application to neonates.9 The lack of evidence or inclusion of Enterovirus in the guidelines may stem from the inability to treat enteroviral disease at this time. Clinicians often state that viral testing in their workup is futile and wasteful of resources, since it does not impact their treatment and disposition. While there is not much discussion on when to test for enteroviral infection, there is ample literature on the diagnostic tests that can be used to diagnose enteroviral disease. The continued advancements of these diagnostic tests offer better tools that may be adopted into clinical practice. Articles defining diagnostic methods for enteroviral disease were reviewed, and continued efforts to enhance the methods of detection were revealed. Clinical trials on treatment for enteroviral disease, including Cochrane reviews, were included with a search for trials that are currently underway. As the diagnostic modalities and treatment options become more available, the inclusion of enteroviral disease in the discussion of neonatal infections should become more impressive.
Enteroviruses are small (approximately 30 nm), single-strand ribonucleic acid (RNA) viruses that have the potential to cause infection. The genus Enterovirus belongs to the family Picornaviridae, a large family of morphologically identical, single-strand RNA viruses that share a common genomic and structural organization. Variations in the capsid proteins made of 60 subunits with 4 structural proteins (VP1-VP4) are responsible for the antigenic diversity among the 64 recognized human Enterovirus serotypes.10 The original classification of enteroviruses was based on their effects on tissue cultures and in animal studies. The genus Enterovirus was divided into 4 species: poliovirus, coxsackie A, coxsackie B, and echovirus.
Eventually, many strains of Enterovirus were discovered that did not match these specifications and could not be classified into one of the original subgenera. Therefore, since 1974, newly identified enteroviruses have been assigned Enterovirus type numbers. Each serotype is assigned to 1 of 5 specific subgenera: poliovirus, coxsackievirus A, coxsackievirus B, echoviruses, and enteroviruses. Additionally, several of the echovirus species had a different viral genome than the others and were therefore classified as a new genus, Parechovirus.
Genetic sequencing data has revealed that in many instances the enteroviruses do not have the genetic correspondence correlating with the original classification. A new species designation that is based on genetic analysis now divides human enteroviruses into 4 species (Human Enterovirus [HEV] A, B, C, and D) but keeps traditional names for individual serotypes.11,12 (See Figure 1.)
Various Enterovirus serotypes are associated with a diversity of clinical findings.12-14 These range from mild clinical illness to serious infection or death. (See Table 1.) The scope of this article is limited to the neonate. Infants present a unique challenge in that they do not always present with classic symptoms of enteroviral infections such as rash, mouth sores, or conjunctivitis; rather, they may present with nonspecific febrile illness.15 It is unclear at exactly what age these characteristics appear; however, many neonates can be asymptomatic carriers.15,16 In one study, 79% (59/75) of infants who had positive enteroviral cultures from rectal or throat swabs within 1 month of delivery had asymptomatic infections.16 Other asymptomatic neonates can develop rapidly progressive fatal disease.17 Neonates that present with signs and symptoms often have more serious illness and multi-organ system involvement.18 This is especially true of coxsackie A, coxsackie B, and echovirus.12
In the neonatal period, the majority of patients with a generalized enteroviral infection will present with symptoms at 3 to 5 days of life, with some presenting in the neonatal nursery and others after discharge home.17,19 For this latter group, the emergency department will become the first portal of entry for the clinically ill patient. Early symptoms may be mild and nonspecific and may include lethargy, decreased feeding, and transient respiratory complaints. One-third of the cases may have a biphasic illness characterized by a period of 1 to 7 days of recovery between the initial presentation and more severe disease progression.20
The more serious manifestations of enteroviral disease in neonates are myocarditis and/or hepatitis, which are often accompanied by encephalitis.14,18,21,22,23 While not unique or specific, many members of the Enterovirus species show a preference for certain tissues.18 (See Table 2.) Because there is often an overlap of clinical presentations, it is difficult to assign a serotype or to predict an outcome based on clinical suspicion. The National Enterovirus Surveillance System (NESS) reports cases of neonatal infections. The serotype found to be associated with myocarditis was often coxsackievirus B, and those with hepatitis and coagulopathy were shown to have echovirus 11 infections. These reports are similar to other evidence showing an association between a particular serotype and clinical presentations.24 For example, coxsackievirus B4 has been shown to be associated with a higher incidence of fatality in newborns compared to other enteroviruses. Coxsackieviruses B2-4 typically present with encephalomyocarditis, and echovirus 11 presents with hepatitis-hemorrhage syndrome and has been associated with overwhelming infection.19 Recombination of serotypes has caused an overlap of clinical presentation in the past 40 years which has blended some of these classic associations. Yet it is still evident that particular serotypes have more of a likelihood of presenting with a specific disease such as myocarditis, hepatitis, or encephalitis.
The cause of severity for disease in the neonate is unknown, though it may relate to their immune mechanisms.15 Though levels are comparable to those in the adult, the function of the T and B cell is immature. Protection to the neonate is offered through maternal immunoglobulin G (IgG), which is transferred to the neonate transplacentally, and immunoglobulin A (IgA), which is transported through breast milk.17,25 Other antimicrobial substances, such as lactoferrin, which is found in breast milk, may also play a role.26 Type-specific, transplacentally-acquired antibodies may prevent or modify severe neonatal enteroviral disease.25 There has been evidence showing some protective effect of breastfeeding on acquisition of enteroviral infections, with one study showing an association between lack of breastfeeding and frequent enteroviral infections in neonates.16,27 One study examined the throat and rectal swab cultures of neonates for Enterovirus within 24 hours of birth and then weekly for 1 month.
Acquisition of non-polio enteroviral infection was 12.8%.16 Lack of breastfeeding and lower socioeconomic status were shown to be factors relating to infection.15 In addition, other risk factors that appear to have an impact on severity of neonatal disease acquisition are prematurity, male sex, and onset of maternal enteroviral infection 2 weeks prior to delivery.18,28 Furthermore, it has been shown that severe infections tend to occur in the immediate neonatal period.22 One NICU outbreak demonstrated that the more severe cases resulting in neonatal death or morbidity occurred within the first week of life, which supports other published reports.29
In the case of the neonate, the dilemma becomes: when should the diagnosis of enteroviral infection be considered and when, if ever, should it impact the present guidelines of caring for the potentially septic neonate.
The Picornaviridae family plays a pivotal role in causing infections. Although studies point to the prevalence of disease associated with specific enteroviral infections, the prevalence of these infections and their impact on the population has been difficult to establish. A survey of the literature suggests that many cases of enteroviral infections are not reported. A large-scale attempt to gather patient data, known as NESS, is a "voluntary passive" surveillance system that summarizes the data and gathers information based on monthly reports of all enteroviral detections in state public health laboratories and other diagnostic laboratories. Because non-polio enteroviral infections are not nationally reportable, these numbers are almost certainly less than the true prevalence. Reported cases represent a subset of patients with more severe presentations who warranted advanced diagnostic testing. While clinical information has been reportable in the past, since 2002, access to patient information has become more difficult to collect; therefore, some of the clinical predictors may be lacking in more recent data collection.30
Despite these hurdles, the data makes it clear that enteroviral infection is a public health issue that has widespread prevalence spanning a large age range. In neonatal enteroviral disease, single cases or small clusters appear during periods of peak virus transmission.17 In temperate climates, enteroviral infections most commonly occur in the summer and fall, but in the tropics they may occur year round.18 Of importance, nonspecific symptoms far outnumber those with distinctive clinical syndromes. While multiple sub-types of the virus may be active at a particular time, 1 usually predominates in each geographic area. For example, using data from NESS for 2007, the CDC reported that coxsackievirus B1 was the predominant virus in the U.S. that year and accounted for 25% of enteroviral infections with known serotypes. Fifty-three percent of those patients were neonates. Five coxsackievirus B1-related deaths were reported with multisystem disease onset within the first 4 days of life.24
Table 3 shows the correspondence of various strains of enteroviral infection with age, season, and frequency. The severity of the outcome expressed as a percentage of fatality is attempted; this information is limited on a large scale due to difficulty in clinical data collection after 1998.
Recent use of pan-Enterovirus polymerase chain reaction assays (PCR) testing has aided in the detection and enhanced data collection of enteroviral disease. Reports in the literature have described the prevalence of disease in neonates and their common presenting symptoms. Case reports have detailed cases with complicated courses, severe infection, and death. In many cases, these reports specify species that are causative organisms. For example, Bendig et al describe a fatal case of myocarditis in a neonate that had Enterovirus-RNA detected by PCR on the day of birth and 10 days later.31 Kaplan et al examined fatal cases of coxsackievirus B disease reported in the literature. Of the 41 patients identified, 12 died at 2 to 12 days of age, 11 had a biphasic illness leading to death at 8 to 24 days of age, and 18 had progressive illness leading eventually to death. Myocarditis was present in all cases. Pulmonary hemorrhage was found in 30 infants and liver necrosis in 18 patients.20
For the purpose of this discussion we will separate the transmission of enteroviral disease into 3 categories: transplacental, intrapartum, and postnatal acquisition. Neonates may be secondarily affected by modes of transmission of enteroviral infection through the community. Enteroviral infection can be spread by the fecal-oral route and/or by oral-oral contamination (respiratory routes).19 This may include swimming pools, wading pools, and contaminated hands. Fecal viral shedding may continue for several weeks after infection while respiratory tract shedding may last up to 1 week. The usual incubation period is 3 to 6 days.32 Enteroviral infections can survive on surfaces for long periods of time allowing for transmission by fomites. This might explain the higher incidence of infection in lower socioeconomic populations with poor sanitary facilities.
Neonatal acquisition of enteroviral disease has additional modes of transmission that include prenatal or intrapartum exposure to maternal blood or secretions. Postnatal infection can occur by transmission through family members and health care workers.21 Nursery outbreaks with index cases have been described.17
The method of transplacental transmission of enteroviral infection is poorly understood. Maternal viremia during enteroviral infection is common. Of neonates with enteroviral infection, as many as 65% of their mothers had symptomatic disease during the perinatal period.33 Most intrauterine infection of Enterovirus has been based on indirect evidence of transplacental transmission and the possibility that active infection of the placenta has occurred. Although transplacental transmission of Enterovirus is probably an infrequent occurrence, Enterovirus has been recovered from placental tissues on several occasions.34,35 One recent case report documents the fetoplacental pathology of a third trimester fetus that was infected with coxsackievirus B3 and confirmed by molecular techniques.33 Coxsackievirus in animal studies demonstrates transplacental transmission to the fetus with coxsackievirus B3 and B4 successfully passing the placental barrier.36,37 Another study reported that infection with coxsackievirus B1 in late gestation mice was more severe and that transplacental infection of the fetus occurred transiently during maternal infection.38 Echoviruses 6, 7, 9, 11, 19, 27, and 33 have been identified in cases of transplacental acquired infection.12,39,40,41 In 1 case, echovirus was found in cord blood.17 Not only can gestational infection result in viral passage to the fetus but it may also cause placental compromise with indirect fetal damage.42,43 The transmission is assumed to be hematogenous dissemination that is initiated in the infected placenta.38 The virus has also been identified in amniotic fluid of pregnant women,44,45 possibly causing infection through primary infection of the pharynx and gastrointestinal system.12 Maternal factors such as compromised uteroplacental blood flow, coexisting infection, and advanced age increase the risk of transplacental fetal infection.42
In addition to transplacental acquisition on infection, intrapartum transmission of virus seems probable. According to Modlin, 65% of women who had infants with proven enteroviral disease were symptomatic within 1 week prior to delivery.22 The mothers complained of fever, respiratory symptoms, and/or abdominal pain.14 The virus has been recovered from feces, blood, oropharyngeal secretions, and vaginal secretions, offering many opportunities for transmission.19,46 Fecal carriage of Enterovirus can infect the newborn during vaginal delivery. One report found that 4% of mothers (2 of 55) had Enterovirus in their feces after delivery.47 A Boston study from 1979 found that 4 out of 7 women excreting Enterovirus 11 in the labor room had infants who were subsequently infected by 3 days of life.19
Epidemics have been reported in newborn nurseries due to spread between neonates from a primary infected infant. Outbreaks have been commonly associated with echovirus 11 and coxsackieviruses B1-5, often coinciding with seasonal peaks in the community.46 Most often the source of original infection is prenatally acquired from the mother. In a review, Modlin describes that in 4 of 16 newborn nursery outbreaks the primary case was an infant that was infected from its mother.22 This is often through the vertical transmission to the fetus as previously described, yet other methods have been described. A case report of echovirus 18 discovered in the banked breast milk of the mother of an infant with neonatal sepsis suggested a possible transmission of virus through breast milk.48 Enteroviral infection has also been shown to be spread to infants from health care workers. This may be caused by inadequate hand washing or respiratory droplets.17 Infants at highest risk of secondary infection are those requiring the most intensive nursing care, especially those with mouth care and gavage feeds.17
Recognizing the mode of patient transmission is important in the prevention of disease, prediction of outcome, and understanding of disease severity. While there is insufficient data to make a definitive statement, it is interesting to note that an infected newborn with presumed vertically transmitted prenatal infection seems to convey a worse outcome as compared to postnatally-infected neonates secondarily infected in nursery outbreaks.22
Enteroviral disease should be included in the differential of patients presenting with a broad range of clinical presentations. In considering such a wide constellation of symptoms, one must also consider metabolic derangements, viral infections [herpes simplex virus (HSV), cytomegalovirus (CMV), adenovirus], bacterial infections [Escherichia coli (E. Coli), Group B Streptococcus (GBS), Staphylococcus aureus], or genetic abnormalities. This includes prenatally, antenatally, and postnatally acquired conditions. Diseases that cause more severe presentations such as neonatal sepsis, myocarditis, hepatitis, hepatic necrosis, respiratory compromise, and neurologic changes must be considered. In addition, the differential may include causes of disease processes that present with signs and symptoms that are less specific and severe such as non-discreet rash, upper respiratory infection (URI), feeding intolerance, and conjunctivitis.
A toxic-appearing neonate presenting in the prehospital setting must be assessed, stabilized, and treated according to the Pediatric Advanced Life Support (PALS) guidelines. According to protocol, the ABCs must be addressed, and a secure airway should be established as needed. Oxygen supplementation may be warranted, and the appropriate size equipment must be used to ensure proper delivery. If IV access is established, it may be appropriate to administer an IV fluid bolus. Most importantly, it is important to recognize signs of possible heart failure; in these cases, fluids must be administered judiciously.
The evaluation of a neonate is quite different from that of an older child or adult, so it is important to have a staff that is familiar with caring for neonates. This is an important consideration when transporting a neonate to an emergency department. In the case of an unstable patient, it is vital to transfer the patient to an emergency facility immediately. When considering 2 equidistant hospitals or if the patient is stable, prehospital providers and community physicians should consider which emergency department could provide the best care for the patient. When allowed, neonates should be brought to emergency departments with physicians that are familiar and accustomed to treating neonates and supportive staff that are capable of dealing with the complications of enteroviral disease. This may include children's hospitals with dedicated pediatric emergency departments or emergency departments that have a high volume of pediatric patients and are accustomed to evaluating and treating the younger population.
The initial evaluation of a neonate presenting to the ED involves a quick assessment of whether the baby is critically ill, with the primary focus being to stabilize and treat an infant who shows signs of distress by first treating the presenting symptoms. After reassessing the ABCs and addressing complications such as possible respiratory compromise, dehydration, and lethargy, it is important to gather a complete history from the patient's family. Again, it is important to consider heart failure when considering the amount of fluid to be given to a neonate who may have enteroviral myocarditis.
Maternal and birth history should be thorough and complete. Pregnancy-related questions may include abnormal ultrasounds, screening tests, or pregnancy risks. This should include a detailed history of maternal infections, including HSV and HIV. Issues that should be addressed include maternal GBS colonization, intrapartum chemoprophylaxis with antibiotics, complications during delivery, or suggestions of chorioamnionitis such as prolonged rupture of membranes or maternal fever. Furthermore, maternal illness prior to or during delivery such as URI, fever, cough, or diarrhea should be explored.
A detailed neonatal history should be obtained that includes:
Signs and symptoms in the neonate should be thoroughly evaluated. Although parents of a newborn are often tired and stressed, their concerns and interpretation of changes from their baby's normal activity should be highly considered. Neglecting to address particular characteristics in the neonate's history and current state may lead the physician to miss important clues that would aid in diagnosis and management of the neonate. These include:
The general appearance of a neonate may vary from a comfortable, sleeping baby to a cranky, crying baby. Vital signs are important indicators for neonates. They are often difficult to obtain, especially with a crying baby, but attempts to get a true value should be made. Table 4 shows the normal range of vital signs for a neonate. While the presence of a fever may affect these numbers in an older infant, 1 study did not find a similar relationship between tachycardia and fever in a neonate.49 The immature nervous system may be an explanation for this finding; further studies need to be done. The physician must be careful to consider other causes of tachycardia. Furthermore, both the presence of a fever or the state of hypothermia may be an indicator of an infectious process.
The baby's physical examination must be thorough. Table 5 points out specific features that should be assessed on physical examination. Attention should be given to those factors that may affect further workup and management of the neonate such as heart and liver failure.
Clear recommendations exist for the evaluation of a febrile/toxic-appearing neonate, with much consideration given to bacterial infections.3-8 Initial guidelines were established by Baraff et al in 1993. The criteria for assessment of the febrile neonate are beyond the scope of this discussion, but they are crucial in the management of these patients. These criteria have been established to predict low-risk patients less than 3 months of age with serious bacterial infections and include the Rochester, Philadelphia, and Boston criteria.3-5 The purpose was to screen young infants at risk for serious bacterial infections that included bacteremia, urinary tract infection, or bacterial meningitis. Of these, only the Rochester criteria specifically included the neonate.50 Since these criteria have been proposed, some have tried to use these as a guideline to lower the age at which a full sepsis workup, including lumbar puncture, should be performed.51,52 However, despite the low probability of serious bacterial infection in neonates, under most circumstances all febrile/toxic-appearing neonates should have a sepsis evaluation and be hospitalized for parenteral antibiotics.3,53,54 Most physicians have similar practices with regard to neonatal patients,55 yet there is some variability; this is due to other factors that are involved such as age and fever.50 As many of the signs and symptoms of enteroviral disease are similar and indistinguishable from those of bacterial infections, neonates get evaluated based on the criteria developed for serious bacterial infections.56 According to standard practice, the evaluation for a febrile/toxic neonate includes a CBC with differential; blood culture; urinalysis and urine culture via sterile catheter; as well as lumbar puncture for culture, glucose, protein, and cell count. A chest x-ray and stool culture may also be considered. The challenge is in deciding what, if any, tests are needed in the afebrile non-toxic-appearing neonate who has a normal physical examination but questionable history. While there are no guidelines established to aid the clinician in these circumstances, a thorough neonatal history as previously described may reveal some pertinent clues that would encourage further workup.
The peripheral white blood count cannot be used alone to guide the physician in determining the presence and/or type of infection.57-59 When interpreting the white blood (cell) count (WBC) values, it is important to remember the normal values of the neonate which differ from that of an older age group.60 (See Tables 6 and 7.) Many studies of bacterial infections have examined the utility of WBC in evaluating a neonate. Although it is recommended as part of the workup of a neonate with a presumed infection, there is constant debate over the usefulness and predictability of WBC. One study of WBC as an indicator of bacteremia analyzed infants 0 to 89 days old, 25% of whom were neonates with a fever greater than 38°C (100.4°F). This study showed that the total peripheral WBC is limited in its ability to screen for bacteremia. Although it may predict higher risk of bacteremia at extreme values, there was no threshold that was determined to help stratify according to risk of bacteremia.61 Another study attempting to use the peripheral WBC to identify febrile neonatal patients with bacterial infections could not find a threshold that would discriminate patients with bacterial infections. Because none of the febrile infants in this study with a WBC less than 10 x 109 cells/L had a bacterial infection, it was suggested that future studies may support using WBC as a negative predictive value.62 In their study, Bonsu and Harper suggested that WBC can be used to estimate the risk of bacterial meningitis relative to isolated bacteremia. One study that examined the difference in risk between high and low WBCs found that values less than 5000 cells/mm3 were 7 times as likely to have bacterial meningitis compared to 3 times as likely for WBCs above 15,000 cells/mm3.63 Although most of the studies on WBC do not specifically include enteroviral infections, there appears to be a similar range of findings with regard to WBC. In enteroviral disease, WBC may be normal or elevated. Between 10% and 20% of neonates infected with Enterovirus have elevated WBCs.14
While CSF pleocytosis suggestive of viral disease is more consistently seen with viral illness in older patients, it may not be helpful in neonates. CSF pleocytosis may be absent or it may reach high levels mimicking bacterial meningitis.56,64-66 For example, 1 study that included patients up to 90 days of age in sepsis evaluation showed that CSF pleocytosis occurred in 69 of 138 (50%) patients with enteroviral disease. This study defined pleocytosis as a CSF WBC of > 22 mm3 in neonates, >15 mm3 in infants 4 to 8 weeks old, and > 7mm3 in infants older than 8 weeks. Those that had enteroviral meningitis with positive CSF PCR were more likely to have CSF pleocytosis than enteroviral-positive infants with negative CSF PCR (54% compared to 29%). Similar to reports on patients in varied age groups, this study also found CSF protein elevation in 62% of patients identified.65,67
PCR have been utilized in the testing of specimens from the CSF, pharynx, serum, and urine.69 One study of 502 children, two-thirds of whom were less than 1 year of age, reported 100% sensitivity and 97% specificity when using CSF PCR compared to CSF viral culture,72 which has a sensitivity of 65% of 75%. Virus titers are often too low to be detected in CSF with standard procedures.74 On the other hand, a 77% sensitivity of PCR for urine sample was found.72 Another study evaluating babies younger than 12 weeks found the PCR urine test to have a sensitivity of 47% compared to the sensitivity of 39% for urine cultures.75 These findings are consistent with other reports showing a higher sensitivity for PCR in comparison to culture.65,76
Pharyngeal swabs are noted to yield positive PCR results on occasion when the cultures are negative.72 This is thought to be secondary to the persistence of pharyngeal shedding of viral particles that are detected by PCR and not by Enterovirus culture.72,76 In one study by Rittichier et al in 2005, blood was the only positive result for Enterovirus in 22% of the Enterovirus-positive infants. According to this study, testing both blood and CSF specimens may increase diagnostic yield by approximately 20%. Therefore, it is recommended to test both blood and CSF.65 Although testing both blood and CSF may improve results, this is not a widely accepted practice. Testing for Enterovirus PCR in the blood is not a standard laboratory procedure and may be difficult to obtain. Furthermore, the results may not be helpful in determining enteroviral meningitis.
Because there is no universal standard PCR kit, individual laboratories may use different methods for obtaining and interpreting the results. This leaves room for testing inaccuracies and variability on the amount of time in which the results may be available. A consideration is to send the specimen of a neonate with suspected enteroviral infection to a laboratory that is adept at performing this test. When considering sending a specimen for Enterovirus PCR, it is important to know whether the hospital's laboratory is capable of running the specimen or whether it must be sent to another laboratory. This drastically changes the turnaround time for obtaining results.18, 65, 76,77
Because there is no universal standard PCR kit, individual laboratories may use different methods for obtaining and interpreting the results. This leaves room for testing inaccuracies and variability on the amount of time in which the results may be available. A consideration is to send the specimen of a neonate with suspected enteroviral infection to a laboratory that is adept at performing this test. When considering sending a specimen for Enterovirus PCR, it is important to know whether the hospital's laboratory is capable of running the specimen or whether it must be sent to another laboratory. This drastically changes the turnaround time for obtaining results.18, 65, 76,77
In certain cases, faster culture results may reduce the hospital stay.76 From a monetary perspective, this enhanced turnaround time may reduce healthcare spending because of shortened administration of medications and/or hospitalizations.71 Several studies have examined this effect and have shown that even considering the expenditure of PCR costs, clinical decision-making would be affected and healthcare spending would be reduced.70-73, 78,79 While this would have great implications in the care of children with infections and possible enteroviral disease, the inability to predict serious bacterial infections in neonates makes this not applicable to the neonate.
Based on the presentation of the neonate, other tests may be warranted. For example, a neonate who presents with signs and symptoms of enteroviral infection affecting the liver should have an abdominal panel performed with transaminases, ammonia, coagulation panel, and bilirubin. A neonate with respiratory distress should have a chest x-ray performed to rule out pneumonia or cardiomegaly. In the presence of an abnormal cardiac examination, an electrocardiogram (ECG), echocardiogram (ECHO), and cardiology consultation should be ordered with a high suspicion of viral myocarditis. This workup should also be considered on a baby with tachycardia or tachypnea out of the normal range after accounting for changes secondary to fever.
Recent studies are evaluating the use of serum markers in the evaluation of neonatal sepsis such as acute phase reactants and prolacitonin.60,80,81 The possibility of using these markers in identifying septic infants may help in the management of these patients. While many of these tests have good sensitivity, their specificity and positive predictive values are not as impressive.60 The role of cytokines in sepsis caused by viral organisms has not been fully investigated. Furthermore, the comparison between values in viral and bacterial infection has not been clarified. Future research may be able to guide clinicians by utilizing these tools to distinguish various causes of sepsis and to impact future management of these patients.
The treatment of enteroviral infection in the neonate depends in part on the type of infection and the presenting signs and symptoms. Unfortunately there is currently no FDA approved treatment of Enterovirus-related disease. Much of the care of the neonate with suspected or confirmed enteroviral disease is supportive.
At this time, clinicians who recognize an enteroviral infection in a newborn are mostly limited to treating the complications of enteroviral disease. This includes treating the signs and symptoms of heart failure, liver failure, and encephalopathy. While a discussion of each of these entities is beyond the scope of this narrative, it is imperative to recognize that although recognition of enteroviral disease may be important, treating the symptoms of the disease may be even more beneficial and life saving.
Though there is some suggestion that a neonate with enteroviral disease has a lower risk of serious bacterial infection, it is currently accepted practice to treat the septic-appearing infant until the underlying etiology is determined. Even though it will not treat enteroviral infection, the initiation of antibiotics such as ampicillin, gentamycin, or cefotaxime along with the consideration of vancomycin in the very ill-appearing infant is recommended.50,60 Similarly, infants suspected of HSV should be started on acyclovir to protect and treat herpes encephalitis.50
While there is much discussion on the use of PCR to shorten the length of hospital stay and stop preventative antibiotic treatment in a subset of patients, there is not much discussion on its applicability in the treatment of enteroviral infections. This is perhaps due to the lack of definitive treatment guidelines for enteroviral disease, since knowing the cause may not necessarily help in the treatment and management of the patient. If current trials show favorable results, perhaps the future will emphasize obtaining the results of enteroviral infection in order to help treat the disease itself.
In surveying the literature, no studies were found that addressed the impact of positive PCR for hospitalized neonates according to the most recent guidelines for suspected sepsis workup in a neonate. While some of the studies included neonates in their sample size, they were grouped together with older patients and the impact cannot be extrapolated.78 For example, studies on infections that included patients up to 90 days of age do not stratify the neonates. As previously discussed, these patients have different standards of care due to their higher risk of infection, sepsis, and complicated disease. Relevant questions would have included whether we can discharge non-septic-appearing neonates who have positive PCR without giving them a 48-hour course of antibiotics or without observing them for that same time period. Furthermore, it would include whether these numbers change according to the day of life that the baby was presenting. Finally, it would be important to know whether this logic could be applied to a baby less than 14 days old who has a higher risk of complicated enteroviral disease. The monetary impact on this group of patients cannot be considered until these questions are addressed with a significant sample of patients that could potentially alter the current guidelines for workup of a septic neonate. Therefore, at this point, the neonate cannot be discharged home before the culture results are confirmed negative even if their enteroviral PCR are negative.
Much work is underway to determine a way to protect neonates from overwhelming infections. This includes emphasis on both bacterial and viral causes of disease. The background of this work involves an investigation of the tendency towards more severe disease in neonates. Since neonatal disease may reflect the immune status of the newborn and studies have shown that babies who lack immunoglobulins transferred from their mothers are at higher risk, immunoglobulin administration may serve as a treatment for a select number of patients.
A Cochrane database review in 2004 evaluated the effect of treatment of infection in neonates with IVIG. The results demonstrated a borderline statistical significance in mortality in neonates with suspected infection and a statistically significant reduction in mortality with IVIG in cases of subsequently proven infection [typical RR 0.55 (95% CI, 0.31, 0.98)].83 These results have not yet been validated and therefore are not accepted practice. A large study being conducted by the International Neonatal Immunotherapy Study (INIS) has completed its recruitment phase and has randomized 3493 babies in a trial evaluating whether IVIG reduces death and disability in patients with suspected or proven infection. The patients were recruited from October 2001 to September 2007 in 9 different countries. Infants that were eligible for this study included those in the NICU that were receiving antibiotics and had a proven or suspected serious infection (not necessarily life-threatening) and had either a birth-weight of less than 1500 g, respiratory support via endotracheal tube, or evidence of infection in blood culture, CSF, or a sterile body fluid. Uncertainty of the need for IVIG in these patients was a precondition. The follow up phase will be completed at the end of 2009.84 Although the study does not look at Enterovirus specifically, the protocol includes patients with suspected or proven infection that does not limit to bacterial infections. It is unclear at this time how many infants with viral infection were included in this study, if at all, though perhaps a subgroup analysis of viral patients will be studied from this larger sample, serving to validate prior studies.
While IVIG is being studied for its global use in the treatment of neonatal sepsis including viral and bacterial causes of disease, trials for specific treatment modalities for enteroviral disease are also underway. Just as IVIG has been suggested in treating enteroviral disease, pleconaril (an antiviral medication) has been used in a variety of case studies.21,85
Pleconaril is a compound that integrates into the capsid of the Picornaviruses, including the enteroviruses, preventing the virus from attaching to cellular receptors and uncoating to release RNA into the cell. In 1996, pleconaril was made available on a compassionate care basis to patients with potentially life-threatening enteroviral infections, including those with neonatal enteroviral sepsis and myocarditis. While 5 out of 6 infants with neonatal enteroviral sepsis survived, the sample size is too small to determine the impact of the pleconaril.23 Further studies have similarly shown symptomatic improvement in patients being treated with pleconaril,21 yet no double-blinded placebo-controlled trial has been performed on neonates to date.
Pleconaril failed to secure FDA approval for use in adults because of the finding of induction of CYP3A enzyme activity and the potential for drug interactions, specifically with oral contraceptives.86 It was taken off the market for a period of time and was then licensed as an intranasal drug in consideration of treating the "common cold."86 Oral pleconaril is now being manufactured and is undergoing clinical trials in the U.S. and Canada. A phase II study sponsored by the National Institute of Allergy and Infectious Disease is in the process of recruiting patients for a double-blinded placebo-controlled virologic efficacy trial of pleconaril in the treatment of neonates with enteroviral sepsis syndrome. This study involves babies who are younger than 15 days of age at the onset of symptoms and who present with a very poor clinical presentation.87 If the outcome of this trial is favorable, pleconaril may be available in the near future and may have a role in clinical practice. While this would be applicable for the very sick neonate, using the drug may be the stepping stone for further research into its use and safety in the prevention of severe disease in a mild to moderately affected neonate. The questions arise of whether the drug will impact the treatment of less severe illnesses and whether it can be given to patients with mild enteroviral disease in order to prevent progression of disease. Before pleconaril or any other medication can be used, modalities to differentiate which patients are more likely to have a worsening of their disease progression must be determined. As it is assumed that many newborns may be harboring asymptomatic infection and only a few of these may progress to complicated enteroviral disease, the key to the treatment and prevention of severe disease is to distinguish these patients from the larger group. Once this is possible, more specific treatment modalities may be able to enhance the treatment of these patients.
Another area that could be investigated in the future is whether PCR blood analysis would be helpful in diagnosing an enteroviral infection. Although there are few labs that are currently running this test, the ability to obtain blood PCR may alter diagnostic workup. If a blood PCR sample is quick enough to give results before an LP is performed or antibiotics are given, it could have a tremendous impact on diagnostic workup. Enteroviruses have been found to be the cause of 60% to 90% of cases in neonates presenting with fever and CSF pleocytosis who are diagnosed with aseptic meningitis based on blood cultures remaining negative.82 A study published in the Journal of Pediatrics discusses the incidence of serious bacterial illness (SBI) in patients with viral diseases. The authors suggested that including viral diagnostic studies in combination with the Rochester criteria can aid in the identification of infants at risk for serious bacterial infections. The Rochester criteria is a tool that was created to help characterize low-risk and high-risk patients. (See Table 8). According to the study by Byington et al, using viral studies can help further stratify patients. For example, patients who are defined by the Rochester criteria as being high-risk and who have a proven viral infection have a lower risk of serious bacterial infection, UTI, and soft tissue infection compared to those in the same category without a proven viral infection. Those high-risk neonates without viral infections are 3.5 times as likely to have a serious bacterial infection.9 Similarly, a study of a cohort of 1061 infants less than 90 days of age who underwent enteroviral testing showed that none of the patients with enteroviral disease had evidence of bacterial meningitis. Of the 214 infants who were Enterovirus-positive in this study, 12 also had a UTI (5.6%), and 3 had concurrent bacteremia (1%). The 3 patients with bacteremia were less than 3 days of age, 1 of whom had life-threatening disease. It is unclear if this was secondary to Campylobacter sp. bacteremia or Enterovirus.65 While this study supports the idea that high-risk patients with viral infections have a lower risk of serious bacterial infection, larger studies comparing these populations would need to have significant power to provide strong evidence. They would need to show virtually no evidence of bacterial infection in patients with viral disease in order to change the current workup of a septic neonate, including a recommendation to support diagnostic viral studies in the workup of a neonate. This could provide more information on the actual prevalence of Enterovirus in neonates and allow us to further study the disease, its presentation, and its management. If this idea is confirmed, Enterovirus PCR may prove to be helpful in further categorizing patients at risk for viral and bacterial infections. It may lead to incorporating this test into some of the already recognized standards that some physicians apply in their workup of a febrile neonate.
A septic, ill-appearing infant must be assessed and admitted for further observation, care, and possible treatment. Any signs of complicated disease should be addressed with the appropriate support services rendered. An afebrile neonate with confirmed or suggested enteroviral infection may be discharged home with appropriate follow up in place.
Though confirmation of enteroviral disease may not affect treatment at this time, it may assist the physician in guiding parents and predicting prognosis. As previously discussed, the presentation and course of enteroviral disease is vast, making the determination of a complicated course difficult. Therefore, both PCR and viral culture might be recommended. Determining the serotype of the infection may be helpful because certain serotypes are associated with specific disease entities; those with serotypes that are more closely associated with myocarditis, encephalitis, and hepatitis should be watched for signs of organ involvement. While there may be a higher association with these serotypes, it must be understood that there is significant overlap and recombination of the Enterovirus species and therefore all neonates should be followed.
It is important to give parents of neonates with enteroviral disease discharge instructions, to ensure follow up with a physician, to discuss signs and symptoms of worsening disease or neonatal distress, and to impress upon them the importance of seeking care should the baby's status begin to change. This includes patients that have been hospitalized for a 48-hour course of antibiotics and have negative bacterial cultures. (See the Sample Discharge Instructions.) Immediate follow up with the pediatrician may give the primary medical doctor (PMD) a chance to establish a baseline assessment of the neonate that would allow the doctor to recognize even subtle changes in the neonate's behavior.
With its wide range of presentations, enteroviral disease may be a difficult diagnosis to make. This is especially true in the neonate who is at higher risk of developing complicated diseases such as myocarditis, encephalitis, and hepatitis. Although there is no available treatment for enteroviral disease, it is important for physicians to recognize that this viral infection can be life-threatening and not to dismiss a viral infection in a neonate. Avoidance of transmission should be attempted. Further research must address some of these issues in order to help further guide the physician in the diagnosis and treatment of this auspicious disease.
As the cardiologist arrives, the neonate's status has deteriorated and you are preparing to intubate her. At first attempt you are successful and the cardiologist notifies you that the PICU is ready to accept the patient. You are called by the cardiologist a few hours later who explains to you that despite all attempts, the baby rapidly deteriorated and expired. At the parent's request, an autopsy was performed and coxsackievirus B4 was determined to be the causative agent of the infant's myocarditis.
Evidence-based medicine requires a critical appraisal of the literature based upon study methodology and number of subjects. Not all references are equally robust. The findings of a large, prospective, randomized, and blinded trial should carry more weight than a case report.
To help the reader judge the strength of each reference, pertinent information about the study, such as the type of study and the number of patients in the study, will be included in bold type following the reference, where available.
Karen Goodman; Sylvia Garcia; Audrey Paul
March 1, 2009