Reverse-Transcription Polymerase Chain Reaction Detection of the Enteroviruses Overview and Clinical Utility in Pediatric Enteroviral Infections Jose´ R. Romero, MD

● Objective.—This review focuses on commercial and inhouse–developed reverse-transcription polymerase chain reaction (RT-PCR) assays used for the detection of enteroviral infections. In addition to providing details on the performance of RT-PCR, its specificity, and sensitivity, the clinical utility of this diagnostic method with specific reference to its impact on hospitalization and cost savings is addressed. Data Sources.—MEDLINE was searched for reports relating to RT-PCR detection of the enteroviruses in adults and children. The search was restricted to studies reported in English language journals. Study Selection.—Reports documenting detailed information regarding the RT-PCR conditions, primers, sensitivity, specificity and, if relevant, clinical impact were selected for analysis. Data Extraction.—Details regarding method of extraction of the enteroviral genome, the primers used, RT-PCR conditions, and sensitivity and specificity of the assay were extracted from the literature. For reports detailing the use

of RT-PCR in the clinical management of enteroviral infections in children, the reduction in duration of hospitalization and health care cost savings were recorded. Data Synthesis.—Reverse-transcription PCR can increase the yield of detection of enteroviruses from cerebrospinal fluid by a mean of approximately 20% over tissue culture. Reverse-transcription PCR of cerebrospinal fluid has been shown to exhibit sensitivity and specificity values of 86% to 100% and 92% to 100%, respectively. Reductions of 1 to 3 days of hospitalization per patient are predicted if RTPCR is used to diagnose enteroviral meningitis in children. Conclusions.—Reverse-transcription PCR detection of enteroviral infections is an extremely rapid, sensitive, and specific diagnostic modality. Both commercial assays and assays developed in-house appear to be equivalent with regard to sensitivity and specificity. Reverse-transcription PCR diagnosis of enteroviral infections in children could reduce the length of hospitalization and result in significant health care cost savings. (Arch Pathol Lab Med. 1999;123:1161–1169)

T

An overview of the morphology, genomic characteristics, and biology of the enteroviruses is in order because of their relevance to the newest modality available for the detection of the enteroviruses, reverse-transcription polymerase chain reaction (RT-PCR). The enteroviruses are nonenveloped viruses that possess a single-stranded, positive (messenger)-sense RNA genome that is approximately 7500 nucleotides in length. The lack of a lipid envelope contributes to the structural stability of the enteroviruses, permitting the virus to survive the gastric pH and environmental stresses. The enteroviruses remain viable for prolonged periods in sewage, water, fomites, and on hands, thereby enhancing their transmissibility. When frozen, these viruses are stable for years and even decades.1,5,6 The enteroviral genome is organized into a 59 nontranslated region (59NTR), a polyprotein coding region, a shorter 39NTR, and a terminal polyadenylic tail (Figure 1).1 The viral coding region can be subdivided into P1, P2, and P3 regions.7 The P1 region encodes 4 structural proteins (VP1–4) that comprise the enteroviral capsid, while the P2 and P3 regions encode 7 nonstructural proteins essential for the enterovirus life cycle.1 The enteroviral 59NTR plays a critical role in the enteroviral life cycle and in RT-PCR detection. The 59NTR contains elements essential for replication of the viral genome

he enteroviruses form a genus within the family Picornaviridae (Pico meaning small; rna, RNA genome; viridae, viruses). Like all picornaviruses, the enteroviruses are small, icosahedral RNA viruses. The enteroviruses are a large and highly clinically important group of human pathogens. The genus is divided into 5 groups: the polioviruses, the group A coxsackieviruses, the group B coxsackieviruses, echoviruses, and the newer numbered enteroviruses, totaling 66 different serotypes (Table 1).1 Because of their biologic and molecular dissimilarity with the other members of this genus, it has been proposed that 2 of the enteroviruses, echoviruses 22 and 23, be removed and given their own independent genus (Paraenterovirus).1–4 Accepted for publication July 15,1999. From the Combined Division of Pediatric Infectious Diseases, University of Nebraska Medical Center and Creighton University; Department of Pathology and Microbiology, University of Nebraska Medical Center, Omaha, Neb. Presented at the Eighth Annual William Beaumont Hospital DNA Technology Symposium, DNA Technology in the Clinical Laboratory, Royal Oak, Mich, March 25–27, 1999. Reprints: Jose´ R. Romero, MD, Combined Division of Pediatric Infectious Diseases, 2500 California Plaza, Criss II, Room 409, Omaha, NE 68132. Arch Pathol Lab Med—Vol 123, December 1999

RT-PCR Detection of Enteroviruses—Romero 1161

Table 1.

The Enteroviruses

Group

Polioviruses Coxsackieviruses, group A Coxsackieviruses, group B Echoviruses Human enteroviruses * Coxsackieviruses A23; echoviruses 72 have been reclassified.

Serotypes*

1–3 1–22, 24 1–6 1–7, 9, 11–27, 29–33 68–71 8, 10, and 28; and enterovirus

and translation of its protein coding region.8,9 Given the crucial nature of these functions, it is not surprising that nucleotide sequences that have absolute (or near absolute) conservation among the enteroviruses have been identified within the 59NTR using nucleic acid hybridization and sequencing (reviewed in references 10 and 11). These conserved regions of high nucleotide identity (homology) have been exploited for the development of primers and probes used in RT-PCR for the detection of the enteroviruses. The enteroviruses are transmitted predominantly by the fecal-oral route.6,12 Once infected, an individual may shed virus from the oropharynx for 1 to 4 weeks and for up to 16 weeks in the stool. This issue is of particular importance to the clinician attempting to establish a causal relationship between a patient’s symptoms and an enteroviral infection. Because of the prolonged shedding from the upper respiratory and gastrointestinal tracts, the isolation of enteroviruses from these ‘‘permissive’’ sites does not conclusively establish the causality of a patient’s illness. Their presence may simply reflect a previous infection. Sites such as the central nervous system, vascular system, and urinary tract are normally ‘‘nonpermissive’’ for the enteroviruses (ie, enteroviruses are normally not found in these sites). The finding of enteroviruses in these sites (and in other nonpermissive sites) establishes causality. In temperate climates, the highest rates of enteroviral infection occur during the summer and fall.13–15 Although enteroviral infections occur in all age groups, children are the most common victims. Infection rates for infants are several-fold greater than those for adults.16–19 Thus, children account for the overwhelming majority of cases of enteroviral disease. The enteroviruses are responsible for a myriad of clinical syndromes and involve almost every organ system; manifestations range from nonfocal febrile illness to potentially life-threatening diseases such as meningitis, encephalitis, myocarditis, and fulminant neonatal sepsis.12,20– 23 With the exception of poliomyelitis, enteroviral infections are not required to be reported to the Centers for Disease Control and Prevention. Therefore, the true incidence of infections caused by specific enteroviruses is not known. However, in the United States, the nonpolio en-

teroviruses are estimated to cause 30 to 50 million infections per year.24 The enteroviruses account for at least 53% to 63% of hospital admissions in children with acute, nonfocal, febrile illness during the summer and fall.20,25,26 Approximately half of these infants and children will have enteroviral meningitis. Enteroviruses are responsible for greater than 90% of cases of aseptic meningitis for which an etiologic agent can be identified.26,27 It is believed that each year enteroviral infections result in 30 000 to 50 000 hospitalizations for meningitis alone.24 Owing to underreporting, this figure may be an underestimation of the actual number of cases. Recent reviews have discussed the limitations of tissue culture and non–amplification-based methods of nucleic acid detection for the identification of enteroviruses from clinical specimens.10,28,29 In brief, these limitations center around (1) the low sensitivity of tissue culture and nucleic acid hybridization (35% to 75% and 33%, respectively) and (2) the mean time required for the isolation of enteroviruses from cerebrospinal fluid (CSF) using tissue culture (3.7–8.2 days). Both of these issues are of particular importance in the clinical management of enteroviral infections, inasmuch as patients are frequently discharged (based on the results of the bacterial cultures) long before the results of the viral culture are available. The development of PCR technology30,31 and its subsequent adaptation for the detection of the enterovirus RNA genome via RT-PCR32–34 have provided clinicians with a sensitive, specific, and extremely rapid (ie, hours) method for the detection of the enteroviruses. Reverse-transcription PCR requires only a small sample size, generally 100 to 200 mL of fluid or 1 mg of tissue. The process is highly versatile and can be used to detect the enteroviral genome from multiple body fluids (cerebrospinal fluid, serum, urine, and pericardial fluid), tissue sources (heart, liver, and muscle), and sample sources (stool, rectal, and pharyngeal swabs). Handling and Processing of Clinical Specimens While reports in the literature have described procedures for the amplification of the enteroviral genome from multiple tissues, body fluids, and excreta (reviewed in reference 11), the most commonly used (and most clinically useful) specimens are CSF, serum and, possibly, urine. If obtained sterilely, CSF, serum, and urine require no additional processing prior to the extraction of the enterovirus RNA genome. As noted, the enteroviruses lack a lipid envelope; therefore, they exhibit significant environmental stability. Because of this stability, even ‘‘residual’’ specimen samples found in clinical laboratory refrigerators may be suitable for RT-PCR. While no appreciable loss in PCR signal is seen with short-term storage of specimens at 48C to 2208C,11 longer storage at these temperatures can result in a gradual decline of viral titer that could in turn affect the results of RT-PCR.35–37 Specimens

Figure 1. Organization of the enterovirus RNA genome. NTR indicates nontranslated region. See text for discussion. 1162 Arch Pathol Lab Med—Vol 123, December 1999

RT-PCR Detection of Enteroviruses—Romero

intended for long-term storage and, if possible, specimens that will not be processed immediately should be stored at 2708C. Care should be taken not to disrupt the viral capsid. The enteroviral RNA genome is protected from environmental ribonucleases as long it remains within the protective protein coat (capsid). Freeze-thawing of the specimen disrupts the viral capsid; therefore, storage of samples in frost-free refrigerators should be avoided. Large samples should be divided into smaller aliquots prior to freezing to avoid repetitive episodes of freezing and thawing. Cerebrospinal fluid that has been boiled (as is required when performing latex agglutination assays) is not suitable for RT-PCR, since the high temperature disrupts the viral capsid. Extraction of the Enteroviral RNA Genome Prior to reverse transcription and amplification of the enteroviral RNA genome, the genome must first be extracted from the protein capsid. In addition, proteins and lipids in the sample must be removed. Multiple methods have been devised to accomplish this.11 Traditionally, a mixture of phenol-chloroform-detergent-proteinase has been used.38 While effective, this method does not inactivate the ubiquitous ribonucleases present in body fluids, tissue, excreta, and the environment. Ribonucleases present in the CSF, for example, have been shown to lead to extremely rapid degradation of the enteroviral RNA genome.39,40 Guanidinium isothiocyanate (IsoQuick, ORCA Research Inc, Bothel, Wash),41 alone or in combination with phenol-chloroform (RNAzol-B, Tel-Test, Inc, Freindsward, Tex; Trizol, Life Technologies, Inc, Rockville, Md),42 inactivates ribonucleases and therefore improves the stability of the extracted RNA genome. Viral RNA extraction using guanidinium isothiocyanate alone or in combination with phenol chloroform requires that the extracted RNA be precipitated prior to use in subsequent phases of the amplification. Complete precipitation of extracted RNA is seldom achieved, so that some sample loss occurs. Furthermore, because the sample pellet is frequently difficult to see, sample loss can occur at this phase of the procedure. To reduce this possibility, extraction coupled with RNA purification via the capture of the extracted RNA by magnetic beads has been developed.43 In this procedure, the extracted viral RNA is hybridized to a biotinylated enterovirus-specific oligonucleotide that has been coupled onto a streptavidin-linked magnetic bead. After hybridization, the bead and the hybridized RNA are collected using a magnetic bead concentrator, washed, and used for RT-PCR. Uncoupling of the enteroviral genome from the magnetic bead is not necessary prior to reverse transcription. Most recently, a microspin-formatted procedure that exploits the selective nucleic acid binding properties of a proprietary silica-gel–based membrane has been introduced (QIAamp viral RNA mini kit, Qiagen, Valencia, Calif). This kit permits the purification of viral RNA from CSF, plasma, serum, and other cell-free body fluids. The extracted RNA is eluted from the membrane in an extremely small volume and can be used directly for reverse transcription. Reverse Transcription and Amplification of the Enteroviral Genome A detailed discussion of the reported techniques for reverse transcription and subsequent amplification of the Arch Pathol Lab Med—Vol 123, December 1999

enteroviral genome is beyond the scope of this review. Instead, an overview with emphasis on salient developments is provided. More in-depth discussion of these techniques, conditions, and enzymes can be found elsewhere.11,38,44 Because PCR requires a DNA template, the enteroviral RNA genome must first be ‘‘copied’’ or reverse-transcribed to produce a single-stranded complementary DNA (cDNA) copy of the viral RNA. This can be accomplished using a target-specific antisense or downstream oligonucleotide primer (see ‘‘Primers and Probes for the Detection of the Enteroviruses’’) or a commercially available mixture of random hexameric oligonucleotide primers and one of several commercially available reverse transcriptases derived from avian or murine retroviruses, avian myeloblastosis virus, or Moloney murine leukemia virus, respectively.38 The Moloney murine leukemia virus reverse transcriptase is available in its native form or as a genetically engineered version (SuperScript II, Life Technologies, Inc, Rockville, Md) that exhibits greater thermostability.45 Theoretical and practical advantages exist in favor of the use of each of the reverse transcriptases (reviewed in reference 11). Once generated, the cDNA can be amplified using a set of target-specific oligonucleotide primers and one of multiple commercially available, thermostable DNA polymerases.44 The most commonly used thermostable DNA polymerase for RT-PCR of the enteroviruses is that of Thermus aquaticus, Taq polymerase. A limited number of reports have employed other thermostable DNA polymerases, such as Vent polymerase (Thermococcus litoralis) (New England Biolabs, Beverly, Mass)46,47 or Tfl polymerase (Thermus flavus) (Promega, Madison, Wis).37 A major development in RT-PCR detection of the enteroviruses has been the creation of an assay in which both the reverse transcription and amplification are carried out by a single thermostable enzyme using one set of buffer conditions (Amplicor EV PCR, Roche Diagnostic Systems, Inc, Branchburg, NJ).48,49 This coupled assay exploits the inherent reverse-transcriptase activity present in the DNA polymerase of the thermostable bacterium Thermus thermophilus (Tth). The reverse transcriptase activity of the Tth DNA polymerase is enhanced in the presence of manganese.50 The major advantage of a single-enzyme–buffer assay is that it eliminates the need to open the reaction tube between the RNA reverse transcription and DNA amplification steps. By so doing, it decreases the possibility of contamination. An additional advantage of the use of the thermostable Tth polymerase is that the elevated reaction temperature required for this enzyme will destabilize secondary structures present in the RNA template and thus increase the efficiency of the reverse-transcription phase.45,51 Finally, the higher reaction temperature increases the specificity of the primer-template interaction. A commercially produced, single-enzyme, colorimetric, kitformatted assay for the detection of the enteroviruses (Amplicor EV PCR) has been tested (see ‘‘Sensitivity, Specificity, and Reproducibility of RT-PCR Assays’’),49 but it has not yet been licensed for use in the United States. Primers and Probes for the Detection of the Enteroviruses As noted previously, the existence of conserved regions of nucleotide identity within the enterovirus 59NTR has been exploited for the design of oligonucleotide primers and probes capable of near-universal amplification of the RT-PCR Detection of Enteroviruses—Romero 1163

Figure 2. Relative locations of the originally described primers and probes for the detection of the enteroviruses. Linear representation of the enterovirus 59 nontranslated and the viral protein coding region. Primers and probes are designated as hatched boxes below the genome. Numeration corresponds to the nucleotides (nts) relative to the poliovirus type 1 (Mahoney strain) genome.92

enteroviruses. Five initial reports of primers and probes used for enterovirus RT-PCR identified 7 regions of highly conserved nucleotide sequence within the 59NTR and a single region located in the VP2 coding region (Figure 2).32–35,52 When compared with the sequenced enteroviruses (polioviruses type 1 [Mahoney and Sabin strains], type 2 [Lansing and Sabin strains], and type 3 [Leon and Sabin]; coxsackieviruses A9, A16, A21, A24, B1, B3, B4, B5, and B6; echoviruses 2, 4, 6, 9 [Hill and Barty strains], 11, 12, 25, and 30; and enteroviruses 70 and 71 [Br and MS strains]), the reported primers and probes exhibit homologies that range from 81% to 100% with their target regions.11,47,53 Homology with echoviruses 22 and 23 is substantially lower.3,4 Subsequent reports have used multiple combinations and permutations of the originally described primers and probes directed to these regions in order to detect enteroviruses from clinical samples. A review of published reports15,32–35,37,43,49,54–90 indicates that the 2 primer and probe sets most frequently used for detection of the enteroviruses from clinical specimens are those reported by Rotbart and coworkers33,49 and Chapman et al.34 The primers designed by Rotbart have been incorporated into a commercially produced, single-enzyme, colorimetric format (Amplicor EV PCR). Three reports have extensively evaluated primers and probes identical to or with minor modifications to those first described by Rotbart for their ability to detect the 66 prototype enterovirus strains.60,71,83 The most inclusive of these reports tested all 66 prototype enterovirus strains.60 The authors were able to amplify 60 of the 66 prototype strains tested. The 6 serotypes not amplified were echoviruses 16, 22, and 23, and coxsackieviruses A11, A17, and A24. A subsequent study using 65 prototypic enterovirus strains (coxsackievirus A18 was not tested) detected all strains except echoviruses 22 and 23.71 In the most recent study, echoviruses 22 and 23 were again the only strains not amplified among the 62 enterovirus serotypes tested (coxsackieviruses A1, A2, A4, and A8 were not tested).83 Taken together, these reports indicate that primers and probes targeted to the highly conserved nucleotide regions 444–468, 542–560, and 577–596 of the enteroviral 59NTR (numeration relative to the poliovirus type 1 Mahoney strain genome91) are truly panreactive with the human enteroviruses. Given the molecular (and biologic) dissimilarity of echoviruses 22 and 23 with other members of the genus Enterovirus, it is not surprising that these primers failed to amplify these 2 serotypes.2–4 The improved sensitivity observed in the studies by Halonen et al71 and Kessler et al83 discussed may have been due to the use of guanidinium isothiocyanate to extract the viral genome (see ‘‘Extraction of the Enteroviral RNA Genome’’).71,92 1164 Arch Pathol Lab Med—Vol 123, December 1999

Postamplification Detection of the RT-PCR Product Postamplification detection of RT-PCR products can be accomplished through a number of methods. While agarose gel electrophoresis detection of the amplified product can be used, it does not verify the authenticity of the visualized band. Use of seminested or nested primers (ie, a primer or a set of primers internal to the original set used for amplification, respectively) to verify the authenticity of the amplification has been reported. The internal PCR product can be visualized using agarose gel electrophoresis.37,65,69,79,81 Nucleic acid hybridization of the amplified product to a probe targeted to conserved sequences internal to those of the amplification primers and labeled with biotin, digoxin, or a radioactive isotope is a more sensitive method for detection and verification of the amplified product. Hybridization may be accomplished through slot- or dotblot,32–35,52,55,59,60,62,79 Southern blot,34,36,43,75,79 or liquid-phase techniques.68,71 A colorimetric assay has been developed in which biotinylated primers are used for reverse transcription and amplification.49,93 The biotinylated amplification product is captured by an internal probe bound to the bottom of a microwell.93 The presence of the amplicon is detected by determining the optical density of a colorimetric reaction. Sensitivity, Specificity, and Reproducibility of RT-PCR Assays In the laboratory setting, RT-PCR has been shown to be sufficiently sensitive to detect as little as 1 to 100 molecules of the enterovirus genome.11,34,60,71 Using intact virions and various primer pairs, multiple investigators have demonstrated lower levels of detection to be in the range of 0.001 to 1 plaque-forming units35,65 or 0.003 to 0.1 50% tissue culture infective dose (TCID50).37,43,61,69,74,76,79,83 The possibility of cross-reactivity of primers used in RTPCR for the detection of enteroviruses with other infectious agents has been evaluated using both in-house and commercially developed (Amplicor EV PCR) kits.15,32,33,43,59,60,83 No cross-reactivity has been found with more than 30 different viruses, bacteria, fungi, and yeasts tested (Table 2). While the reported primers and probes appear to crossreact broadly among the enteroviruses, they also are highly specific for the enteroviruses. Reverse-transcription PCR has been shown to be extremely sensitive for the detection of the enteroviruses in clinical specimens. With RT-PCR it has been possible to detect enteroviral infection in agammaglobulinemic patients with tissue culture–negative meningoencephalitis or encephalopathy.66,94 Studies using paired analysis (viral culture and RT-PCR testing) of CSF specimens have demRT-PCR Detection of Enteroviruses—Romero

Table 2.

Viruses, Bacteria, Yeasts, and Fungi Tested for Cross-Reactivity With Reverse-Transcription Polymerase Chain Reaction Primers and Probes Used for Enteroviral Detection Viruses

Bacteria

Escherichia coli Haemophilus influenzae Listeria monocytogenes Mycobacterium tuberculosis Mycoplasma pneumoniae Neisseria meningitidis A, B, C, W135, Y Proprionibacterium acnes Pseudomonas aeruginosa Staphylococcus aureus Staphylococcus epidermidis Streptococcus agalactiae Streptococcus mitis Streptococcus mutans Streptococcus pneumoniae Streptococcus pyogenes Streptococcus sanguis

Adenovirus Cytomegalovirus Epstein-Barr virus Hepatitis A virus Hepatitis C virus Herpes simplex virus types 1 and 2 Human herpesvirus 6 Human immunodeficiency virus Influenza A virus Mumps virus Orf virus Parainfluenza virus types 1–3 Reovirus type 3 Respiratory syncytial virus Rubella virus Varicella-zoster

Table 3.

Yeast/Fungi

Candida albicans Cryptococcus neoformans

Paired Analysis of Tissue Culture Versus Reverse-Transcriptase Polymerase Chain Reaction (RT-PCR) for the Detection of Enterovirus From Cerebrospinal Fluid Source, y

RT-PCR Assay

Rotbart, 199033 Sawyer et al, 199468 Schlesinger et al, 199470 Riding et al, 199678 Hosoya et al, 199781 Yerly et al, 199675 Rotbart et al, 199782 Ahmed et al, 199785 Kessler et al, 199783 Andreoletti et al, 199888 Pozo et al, 199837

In-house In-house In-house In-house In-house Amplicor† Amplicor Amplicor Amplicor Amplicor In-house Amplicor In-house Amplicor Gorgievski-Hrisoho et al, 199889 Amplicor * Values presented as number positive/number tested. † Amplicor EV PCR, Roche Diagnostic Systems, Inc, Branchburg, NJ.

onstrated that RT-PCR is consistently and substantially more sensitive than tissue culture for the detection of enteroviruses (Table 3).* These studies have indicated that RT-PCR can increase the yield of detection of enteroviruses from the CSF by a mean of approximately 20% (range, 7% to 61%) over that seen with tissue culture. In addition, the time required to perform RT-PCR was significantly shorter than the time to positivity for tissue culture. Finally, the improved yield in enterovirus detection was observed regardless of whether an in-house or a commercial (Amplicor EV PCR) assay was used. Reverse-transcription PCR also appears to be more sensitive than tissue culture for the detection of enteroviruses from serum and urine,72,77,82 although these specimens have not been evaluated to the same extent as CSF. The sensitivity and specificity of RT-PCR, compared to the traditional ‘‘gold standard’’ of tissue culture, for the detection of the enteroviruses have been evaluated using different clinical specimens. Compared to tissue culture, RT-PCR of CSF has been shown to exhibit sensitivities that

* References 33, 37, 68, 70, 75, 78, 81, 82, 85, 88, 89. Arch Pathol Lab Med—Vol 123, December 1999

Tissue culture* (%)

9/13 112/217 6/17 6/140 2/23 13/38 36/209 5/61 27/103 1/44 26/50 26/50 1/29 1/29 16/68

(69) (52) (35) (4) (9) (34) (17) (8) (26) (2) (52) (52) (3) (3) (24)

RT-PCR* (%)

13/13 135/217 11/17 35/140 9/23 25/38 51/209 18/61 34/103 10/44 46/50 43/50 4/29 3/29 58/68

(100) (62) (65) (25) (39) (66) (24) (30) (33) (23) (92) (86) (14) (10) (85)

range from 86% to 100% and specificities ranging from 92% to 100%.† A direct comparison of the sensitivity and specificity of the Amplicor EV PCR and in-house RT-PCR assays has recently been reported.79 The performance of the Amplicor EV PCR assay was assessed in 13 different laboratories using a panel of 20 CSF samples containing known quantities of virus (10, 1, 0.1, and 0 TCID50). Additionally, in 5 laboratories the Amplicor EV PCR assay was compared with the in-house enteroviral RT-PCR assay normally used. The in-house assay was different in each of the 5 laboratories and varied with regard to method of RNA extraction, reverse transcription, number of cycles employed during amplification, and method of detection of the RT-PCR product. While it was found that the mean sensitivity values of the 5 in-house and the Amplicor RT-PCR assays were similar to each other for all viral quantities used, it was clearly evident that for all assays tested the mean sensitivity depended significantly on the quantity of virus present in

† References 35, 54, 68, 75, 82, 83, 85, 86, 89, 90.

RT-PCR Detection of Enteroviruses—Romero 1165

the CSF test sample (100% at 10 TCID50 and 75% at 1 TCID50). At the lowest viral concentration (TCID50 of 0.1), both assays exhibited low sensitivities. The mean sensitivity of the in-house assays was 26.6%, while that of the Amplicor EV PCR assay was 8.3%. The authors reported that the difference was not statistically significant. The specificity was found to be 100% for all of the RT-PCR assays. This study indicates that regardless of the format used, RT-PCR assays for enteroviral detection appear to be equally effective in the detection of the enteroviruses. However, the sensitivity of all assays evaluated appears to be highly dependent on the quantity of virus present in the sample. In the same study, the authors demonstrated that there were only small differences in the sensitivities of the Amplicor EV PCR tests performed in different laboratories, attesting to its reproducibility. These findings support an earlier study that addressed the reproducibility of the Amplicor EV PCR.86 In that study, 608 samples had duplicate aliquots run simultaneously. Only 7 pairs gave discordant results. An additional 104 samples had duplicate aliquots run in separate assays. For that set, no discrepancies in results were seen. The overall reproducibility of the assay was calculated to be 99%. Only a limited number of studies have evaluated the sensitivity and specificity of RT-PCR compared to tissue culture for the detection of enteroviruses using other samples. Based on these reports, serum appears to be another excellent source for enterovirus detection. The sensitivity and specificity values for RT-PCR using serum range from 81% to 92% and 98% to 100%, respectively.72,82 Urine appears to be the poorest nonpermissive sample source for detection of the enteroviruses using RT-PCR. Sensitivities of only 62% to 77% and specificities that range from 70% to 95% have been reported.72,77,82 The presence of an unidentified inhibitor(s) in the urine may account for the poor results observed. Nielsen et al77 found that nearly half of the urine samples tested had evidence of an inhibitory substance that reduced the expected optical density reading of urine samples spiked with a known quantity of poliovirus. Single Versus Multiple Nonpermissive Site Samples for RT-PCR Testing A growing body of evidence suggests that testing of specimens from multiple nonpermissive sites could improve the possibility of establishing enteroviral causality in infants and children with aseptic meningitis or nonfocal febrile illness. Abzug et al72 demonstrated that in neonates younger than 14 days of age with aseptic meningitis, 13 of 16 patients (81%) were positive for enterovirus by RTPCR using serum alone. Using aseptically collected urine alone, 10 of 16 patients (62.5%) had documented RT-PCR evidence of enteroviral infection. By combining the results of serum and urine RT-PCR tests, the authors were able to document enteroviral infection as the cause of meningitis in 14 of 16 patients (87.5%). A study by Andreoletti and coworkers investigated children with suspected acute neurological infection using RT-PCR.88 The authors established enteroviral causality in 10 of 44 patients (22.7%) by testing CSF alone, and in 9 of 44 patients (20.5%) using serum alone. By combining the results of RT-PCR testing of CSF and serum, they were able to document enteroviral causality in 14 (31.8%) of the 44 patients. This finding was found to be statistically sig1166 Arch Pathol Lab Med—Vol 123, December 1999

nificant (CSF plus serum vs CSF alone, P 5 .0l4; CSF plus serum vs serum alone, P 5 .007). In adults, the combination of the RT-PCR results of multiple specimens did not improve on the 43.7% of patients identified by testing of CSF alone. As part of a study designed to determine the epidemiology of enteroviral infection in febrile and afebrile infants younger than 90 days, 80 infants were found to be positive by RT-PCR for enterovirus using specimens from a nonpermissive site (CSF or blood).15 Fifty-five (69%) infants were identified by RT-PCR of the blood, and 62 (77.5%) were identified by RT-PCR of CSF. A combination of either positive blood or positive CSF RT-PCR identified 77 (96%) of the enterovirus-positive infants. These studies appear to indicate that by combining the RT-PCR results from specimens collected from multiple nonpermissive sites, it may be possible to improve the ability to establish an enteroviral causality in infants and children with aseptic meningitis. However, the limited numbers of patients in these studies precludes a formal recommendation for testing of multiple sites. It should also be noted that the usefulness of urine as a diagnostic substrate for enteroviral RT-PCR may be useful only in the first 2 weeks of life.29 Finally, although these studies are promising, Ahmed et al85 found that combining the results of RT-PCR testing of CSF with those from serum or urine did not enhance the ability to detect enteroviral infection. Clinical Utility of RT-PCR In the United States, enteroviral infections of infants and children result in a significant economic impact each year.15,87,94 Two recent studies have documented that the average cost of hospitalization for infants with enteroviral infection ranges from $4476 to $4921.15,87 In these 2 studies, the average length of hospitalization was found to be 3 to 4 days. Because of the length of time required for tissue culture to become positive (4–8 days), the majority of clinicians rely on bacterial cultures of blood, CSF, and urine as the major determinant for discharge of a febrile infant or child admitted for evaluation of possible sepsis or meningitis. In general, if patients are doing well clinically and the cultures obtained at the time of admission are negative at 48 to 72 hours, they are discharged. Because of the speed with which RT-PCR can be performed, it holds the promise of shortening the duration of hospitalization in enterovirus-infected children. However, RTPCR results will need to be available in less time than the currently relied upon bacterial cultures if they are to have an impact on the current clinical management of febrile infants and children. Beginning with the earliest reports documenting the efficiency of RT-PCR for detecting enteroviral meningitis in children, it was predicted that this assay could potentially lead to a reduction in hospitalization time and therapy.15,54,68,72,82 In their retrospective study of infants younger than 3 months of age infected with enteroviral meningitis, Schlesinger et al documented a mean hospitalization time of 4 days.70 The mean time to positive tissue culture for patients in whom CSF cultures were positive was 6.5 days. Using an in-house enteroviral RT-PCR assay whose results could be reported in 24 to 48 hours, the authors predicted a reduction of 1.2 days of hospitalization per patient had it been used for diagnosis. In their retrospective study, Ahmed et al85 identified 13 RT-PCR Detection of Enteroviruses—Romero

hospitalized infants in whom enteroviruses were isolated from CSF. All 13 patients had enteroviral genome detected from the CSF using the Amplicor EV PCR assay. Tissue culture required a mean of 4.2 days before a cytopathic effect was seen. The mean duration of hospitalization was 3.8 days. Based on the assumptions that RT-PCR results could be obtained after 24 hours and that a positive result would have prompted discharge, 2.8 days of hospitalization per patient would have been saved had RT-PCR been available. A retrospective study designed to evaluate the potential impact of RT-PCR diagnosis of enteroviral meningitis on hospitalization and cost savings identified enteroviral genome in the CSF of 21 patients aged 4 days to 18 years using the Amplicor EV PCR assay.95 Seventeen patients (81%) were hospitalized for a mean of 2.9 days per patient. Antibiotics were administered to 16 (95%) of 17 patients for an identical number of days. Cerebrospinal fluid was cultured in only 8 of 21 patients and was found to be positive in 6. The time to positive tissue culture ranged from 2 to 4 days. Based on the assumptions that discharge from the hospital would have occurred if the patient was afebrile and clinically stable, and that the RT-PCR results available by the end of the first fully staffed microbiology shift after the CSF was obtained were positive for enterovirus, the mean predicted length of hospitalization would have been 1.4 days per patient. A mean of 1.8 days of antibiotic therapy per patient would have been eliminated using RT-PCR. Based on the reduced number of days of hospitalization and antibiotic use, the authors estimated a mean cost savings of a minimum of $1000 per patient. The authors cautioned that any delay in reporting of RT-PCR results would adversely impact the length of stay and cost savings. Marshall et al87 conducted the most detailed analysis of potential cost savings through RT-PCR diagnosis of enteroviral meningitis to date. Their analysis included 18 patients younger than 6 months of age with CSF pleocytosis. It was assumed that because all patients exhibited pleocytosis, they would be most likely to remain hospitalized until bacterial meningitis was excluded. In 9 of these patients, the RT-PCR assay (Amplicor EV PCR) was positive. Total hospital charges for the RT-PCR–positive group was $44 289, and for RT-PCR–negative patients, $47 400. The authors devised 2 scenarios. In the first, it was assumed that clinicians would discharge afebrile patients in whom a viral diagnosis was established using RT-PCR. In this scenario, a reduction of total charges to $34 073 (23% reduction) for the RT-PCR–positive cohort would have resulted. In the second, more liberal scenario, it was assumed that the RT-PCR results would be available at 24 hours and that physicians would discharge enteroviruspositive patients regardless of fever at the time of diagnosis. In this scenario, total costs in the RT-PCR–positive cohort were reduced by 59% to $18 145. In both scenarios, it was assumed that charges for viral cultures would be eliminated and that the cost of RT-PCR would be $150 per patient. These studies clearly indicate that substantial cost reductions in the management of infants with enteroviral meningitis are possible if widespread use of RT-PCR becomes available. While promising, these studies do not predict how clinicians actually treating infants and children with enteroviral infections will use RT-PCR for the management of their patients. Questions remain as to Arch Pathol Lab Med—Vol 123, December 1999

whether clinicians will be willing to discontinue antibiotics in patients documented to have enteroviral infections by RT-PCR. Finally, it has yet to be proven that a rapid diagnosis of enteroviral infection in infants and children results in earlier discharge from the hospital. Only the results of future prospective outcomes-based studies designed to evaluate the use of RT-PCR in the management of enterovirus in infants, children, and adults will be able to address these issues. In summary, RT-PCR has been conclusively shown to be superior to tissue culture with regard to speed and sensitivity for the detection of enteroviruses in infants and children. The primers and probes used for enterovirus amplification have been documented to be extremely specific for the enteroviruses. Multiple studies have documented versatility of RT-PCR for the detection of enteroviruses from multiple body fluids, tissues, and excreta. Finally, recent reports indicate that RT-PCR could have an impact on the clinical management of patients infected with an enterovirus. References 1. Rueckert RR. Picornaviridae: the viruses and their replication. In: Fields BN, Knipe DM, Howley PM, et al, eds. Virology. 3rd ed. Philadelphia, Pa: LippincottRaven Publishers; 1996:609–654. 2. Coller BA, Chapman NM, Beck MA, Pallansch MA, Gauntt CJ, Tracy SM. Echovirus 22 is an atypical enterovirus. J Virol. 1990;64:2692–2701. 3. Oberste MS, Maher K, Pallansch MA. Complete sequence of echovirus 23 and its relationship to echovirus 22 and other human enteroviruses. Virus Res. 1998;56:217–223. 4. Ghazi F, Hughes PJ, Hyypia T, Stanway G. Molecular analysis of human parechovirus type 2 (formerly echovirus 23). Gen Virol. 1998;79(pt 11):2641– 2650. 5. Morens DM, Pallansch MA. Epidemiology. In: Rotbart H, ed. Human Enterovirus Infections. Washington, DC: American Society for Microbiology; 1995:3. 6. Melnick JL. Enteroviruses: polioviruses, coxsackieviruses, echoviruses, and newer enteroviruses. In: Fields BN, Knipe DM, Howley PM, et al, eds. Fields Virology. 3rd ed. Philadelphia, Pa: Raven Publishers; 1996:655. 7. Rueckert RR, Wimmer E. Systematic nomenclature of picornavirus proteins. J Virol. 1984;50:957–959. 8. Andino R, Rieckhof GE, Achaso PL, Baltimore D. Poliovirus RNA synthesis utilizes a RNP complex formed around the 5’-end of viral RNA. EMBO J. 1993; 12:3587–3589. 9. Ehrenfeld E. Initiation of translation by picornavirus RNAs. In: Hershey JWB, Mathews MB, Sonenberg N, eds. Translational Control. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 1996:549–573. 10. Rotbart HA. Nucleic acid detection systems for enteroviruses. Clin Microbiol Rev. 1991;4:156–168. 11. Romero JR, Rotbart HA. PCR-based strategies for the detection of human enteroviruses. In: Ehrlich GD, Greenberg SJ, eds. PCR-Based Diagnostics in Infectious Disease. Cambridge, Mass: Blackwell Scientific Publications; 1994:341. 12. Cherry JD. Enteroviruses: coxsackieviruses, enteroviruses and polioviruses. In: Feigin RD, Cherry JD, eds. Textbook of Pediatric Infectious Diseases. 4th ed. Philadelphia, Pa: WB Saunders; 1998:1787–1839. 13. Moore M. Enteroviral disease in the United States, 1970–1979. J Infect Dis. 1982;146:103. 14. Strikas RA, Anderson LJ, Parker RA. Temporal and geographic patterns of isolates of nonpolio enterovirus in the United States, 1970–1983. J Infect Dis. 1986;153:346–351. 15. Byington CL, Taggart EW, Carroll KC, Hillyard DR. A polymerase chain reaction–based epidemiologic investigation of the incidence of nonpolio enteroviral infections in febrile and afebrile infants 90 days and younger. Pediatrics. 1999;103:E27. 16. Gelfand HA, Holguin AH, Marchetti GE, et al. A continuing surveillance of enterovirus infections in healthy children in six United States cities, I: viruses isolated during 1960 and 1961. Am J Hyg. 1963;78:358–375. 17. Marier R, Rodriguez W, Chloupek RJ, et al. Coxsackievirus B5 infection and aseptic meningitis in neonates and children. Am J Dis Child. 1975;129:321– 325. 18. Wilfert CM, Lauer BA, Cohen M, Costenbader ML, Myers E. An epidemic of echovirus 18 meningitis. J Infect Dis. 1975;131:75–78. 19. Rorabaugh ML, Berlin LE, Heldrich F, et al. Aseptic meningitis in infants younger than 2 years of age: acute illness and neurologic complications. Pediatrics. 1993;92:206–211. 20. Modlin JF. Enterovirus infection in infants and children. Adv Pediatr Infect Dis. 1997;12:155–180. 21. Modlin JF. Perinatal echovirus and group B coxsackievirus infections. Clin Perinatol. 1988;15:233–246.

RT-PCR Detection of Enteroviruses—Romero 1167

22. Modlin JF, Kinney JS. Perinatal enterovirus infections. Adv Pediatr Infect Dis. 1989;2:57–78. 23. Modlin JF, Rotbart HA. Group B Coxsackie disease in children. Curr Top Microbiol Immunol. 1997;223:53–80. 24. Oberste MS, Maher K, Kilpatrick DR, Pallansch MA. Molecular evolution of the human enteroviruses: correlation of serotype with VP1 sequence and application to picornavirus classification. Virology. 1999;73:1941–1948. 25. Dagan R. Nonpolio enteroviruses and the febrile young infant: epidemiologic, clinical and diagnostic aspects. Pediatr Infect Dis J. 1996;15:67–71. 26. Berlin LE, Rorabaugh ML, Heldrich F, Roberts K, Doran T, Modlin JF. Aseptic meningitis in infants younger than 2 years of age: acute illness and neurologic complications. Pediatrics. 1993;92:206–211. 27. Rotbart HA. Meningitis and encephalitis. In: Rotbart H, ed. Human Enterovirus Infections. Washington, DC: American Society for Microbiology; 1995: 271. 28. Rotbart HA, Romero JR. Laboratory diagnosis of enteroviral infections. In: Rotbart H, ed. Human Enterovirus Infections. Washington, DC: American Society for Microbiology; 1995:401. 29. Romero JR. Enteroviruses: diagnosis, screening and surveillance. Infect Med. 1999. In press. 30. Mullis KB, Faloona FA. Specific synthesis of DNA in vitro via a polymerasecatalyzed chain reaction. Methods Enzymol. 1987;155:335–350. 31. Saiki R, Gelfand DH, Stoffel S, et al. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science. 1988;239:487– 491. 32. Hyypia T, Auvinen P, Maaronen M. Polymerase chain reaction for human picornaviruses. J Gen Virol. 1989;70:3261–3268. 33. Rotbart HA. Enzymatic RNA amplification of the enteroviruses. J Clin Microbiol. 1990;28:438–442. 34. Chapman NM, Tracy S, Gauntt CJ, Fortmueller U. Molecular detection and identification of enteroviruses using enzymatic amplification and nucleic acid hybridization. J Clin Microbiol. 1990;28:843–850. 35. Olive DM, Al-Mufti S, Al-Mulla W, et al. Detection and differentiation of picornaviruses in clinical samples following genomic amplification. J Gen Virol. 1990;71:2141–2147. 36. Yang CF, De L, Holloway BP, Pallansch MA, Kew OM. Detection and identification of vaccine-related polioviruses by the polymerase chain reaction. Virus. 1991;20:159–179. 37. Pozo F, Casas I, Tenorio A, Trallero G, Echevarria JM. Evaluation of a commercially available reverse transcription-PCR assay for diagnosis of enteroviral infection in archival and prospectively collected cerebrospinal fluid specimens. J Clin Microbiol. 1998;36:1741–1745. 38. Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 1989. 39. Rotbart HA, Levin MJ, Villarreal LP, Tracy SM, Semler BL, Wimmer E. Factors affecting the detection of enteroviruses in cerebrospinal fluid with coxsackievirus B3 and poliovirus 1 cDNA probes. J Clin Microbiol. 1985;22:220–224. 40. Rotbart HA, Levin MJ, Murphy NL, Abzug MJ. RNA target loss during solid phase hybridization of body fluids: a quantitative study. Mol Cell Probes. 1987; 1:347–358. 41. Chirgwin JM, Przbyla AE, MacDonald RJ, et al. Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry. 1979;18:5294–5299. 42. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem. 1987; 162:156–159. 43. Muir P, Nicholson F, Jhetam M, Neogi S, Banatvala JE. Rapid diagnosis of enterovirus infection by magnetic bead extraction and polymerase chain reaction detection of enterovirus RNA in clinical specimens. J Clin Microbiol. 1993;31: 31–38. 44. Bej AK, Mahbubani MH. Thermostable DNA polymerases for in vitro DNA amplifications. In: Griffin HG, Griffin AM, eds. PCR Technology Current Innovation. Boca Raton, Fla: CRC Press Inc; 1994:219–237. 45. Kotewicz ML, Sampson CM, D’Alessio JM, Gerard GF. Isolation of cloned Moloney murine leukemia virus reverse transcriptase lacking ribonuclease H activity. Nucleic Acids Res. 1988;16:265–277. 46. Romero JR, Rotbart HA. Sequence analysis of the downstream 59 nontranslated region of seven echoviruses with different neurovirulence phenotypes. J Virol. 1995;69:1370–1375. 47. Romero JR, Chapman NM, Rotbart HA, Tracy SM. Polymerase chain reaction detection of the human enteroviruses. In: Becker Y, Darai G, eds. PCR: Protocol for Diagnosis of Human and Animal Virus Diseases. Berlin, Germany: Springer-Verlag; 1996:259–266. 48. Young KKY, Resnick RM, Myers TW. Detection of hepatitis C virus RNA by a combined reverse transcription-polymerase chain reaction assay. J Clin Microbiol. 1993;31:882–886. 49. Rotbart HA, Sawyer MH, Fast S, et al. Diagnosis of enteroviral meningitis by using PCR with a colorimetric microwell detection assay. J Clin Microbiol. 1994;32:2590–2592. 50. Myers TW, Gelfand DH. Reverse transcription and DNA amplification by a Thermus thermophilus DNA polymerase. Biochemistry. 1991;30:7661–7666. 51. Buell GN, Wickens MP, Payvar F, Schimke RT. Synthesis of full length cDNAs from four partially purified oviduct mRNAs. J Biol Chem. 1978;253:2471– 2482. 52. Gow JW, Behan WM, Clements GB, Woodall C, Riding M, Behan PO.

1168 Arch Pathol Lab Med—Vol 123, December 1999

Enteroviral RNA sequences detected by polymerase chain reaction in muscle of patients with postviral fatigue syndrome. BMJ. 1991;302:692–696. 53. Zell R. The Jena Picornavirus Database: 1997 Institute of Virology, Friedrich Schiller University, Jena, Germany. Available at: http://www.uni- jena.de/;i6zero/ database.html. Accessed February 10, 1999. 54. Rotbart HA. Diagnosis of enteroviral meningitis with the polymerase chain reaction. J Pediatr. 1990;117:85–89. 55. Jin O, Sole MJ, Butany JW, et al. Detection of enterovirus RNA in myocardial biopsies from patients with myocarditis and cardiomyopathy using gene amplification by polymerase chain reaction. Circulation. 1990;82:8–16. 56. Rotbart HA, Kinsella JP, Wasserman RL. Persistent enterovirus infection in culture-negative meningoencephalitis: demonstration by enzymatic RNA amplification. J Infect Dis. 1990;161:787–791. 57. Redline RW, Genest DR, Tycko B. Detection of enteroviral infection in paraffin-embedded tissue by the RNA polymerase chain reaction technique. Am J Clin Pathol. 1991;96:568–571. 58. Thoren A, Robinson AJ, Maguire T, Jenkins R. Two-step PCR in the retrospective diagnosis of enteroviral viraemia. Scand J Infect Dis. 1992;24:137–141. 59. Abebe A, Johansson B, Abens J, Strannegard O. Detection of enteroviruses in faeces by polymerase chain reaction. Scand J Infect Dis. 1992;24:265–273. 60. Zoll GJ, Melchers WJ, Kopecka H, Jambroes G, van der Poel HJ, Galama JM. General primer-mediated polymerase chain reaction for detection of enteroviruses: application for diagnostic routine and persistent infections. J Clin Microbiol. 1992;30:160–165. 61. Weiss LM, Movahed LA, Billingham ME, Cleary ML. Detection of coxsackievirus B3 RNA in myocardial tissues by the polymerase chain reaction. Am J Pathol. 1991;138:497–503. 62. Glimaker M, Abebe A, Johansson B, Ehrnst A, Olcen P, Strannegard O. Detection of enteroviral RNA by polymerase chain reaction in faecal samples from patients with aseptic meningitis. J Med Virol. 1992;38:54–61. 63. Glimaker M, Johansson B, Olcen P, Ehrnst A, Forsgren M. Detection of enteroviral RNA by polymerase chain reaction in cerebrospinal fluid from patients with aseptic meningitis. Scand J Infect Dis. 1993;25:547–557. 64. Martin AB, Webber S, Fricker FJ, et al. Acute myocarditis: rapid diagnosis by PCR in children. Circulation. 1994;90:330–339. 65. Severini GM, Mestroni L, Falaschi A, Camerini F, Giacca M. Nested polymerase chain reaction for high-sensitivity detection of enteroviral RNA in biological samples. J Clin Microbiol. 1993;31:1345–1349. 66. Webster AD, Rotbart HA, Warner T, Rudge P, Hyman N. Diagnosis of enterovirus brain disease in hypogammaglobulinemic patients by polymerase chain reaction. Clin Infect Dis. 1993;17:657–661. 67. Thoren A, Widell A. PCR for the diagnosis of enteroviral meningitis. Scand J Infect Dis. 1994;26:249–254. 68. Sawyer MH, Holland D, Aintablian N, Connor JD, Keyser EF, Waecker NJ Jr. Diagnosis of enteroviral central nervous system infection by polymerase chain reaction during a large community outbreak. Pediatr Infect Dis J. 1994;13:177– 182. 69. Nicholson F, Meetoo G, Aiyar S, Banatvala JE, Muir P. Detection of enterovirus RNA in clinical samples by nested polymerase chain reaction for rapid diagnosis of enterovirus infection. J Virol Methods. 1994;48:155–166. 70. Schlesinger Y, Sawyer MH, Storch GA. Enteroviral meningitis in infancy: potential role for polymerase chain reaction in patient management. Pediatrics. 1994;94:157–162. 71. Halonen P, Rocha E, Hierholzer J, et al. Detection of enteroviruses and rhinoviruses in clinical specimens by PCR and liquid-phase hybridization. J Clin Microbiol. 1995;33:648–653. 72. Abzug MJ, Loeffelholz M, Rotbart HA. Diagnosis of neonatal enterovirus infection by polymerase chain reaction. J Pediatr. 1995;126:447–450. 73. Aintablian N, Pratt RD, Sawyer MH. Rapidly recurrent enteroviral meningitis in non-immunocompromised infants caused by different viral strains. J Med Virol. 1995;47(2):126–129. 74. Casas I, Klapper PE, Cleator GM, Echevarria JE, Tenorio A, Echevarria JM. Two different PCR assays to detect enteroviral RNA in CSF samples from patients with acute aseptic meningitis. J Med Virol. 1995;47:378–385. 75. Yerly S, Gervaix A, Simonet V, Caflisch M, Perrin L, Wunderli W. Rapid and sensitive detection of enteroviruses in specimens from patients with aseptic meningitis. J Clin Microbiol. 1996;34:199–201. 76. Sharland M, Hodgson J, Davies EG, Booth J, Jeffery S. Enteroviral pharyngitis diagnosed by reverse transcriptase-polymerase chain reaction. Arch Dis Child. 1996;74:462–463. 77. Nielsen LP, Modlin JF, Rotbart HA. Detection of enteroviruses by polymerase chain reaction in urine samples of patients with aseptic meningitis. Pediatr Infect Dis J. 1996;15:625–627. 78. Riding MH, Stewart J, Clements GB, Galbraith DN. Enteroviral polymerase chain reaction in the investigation of aseptic meningitis. J Med Virol. 1996;50: 204–206. 79. Lina B, Pozzetto B, Andreoletti L, et al. Multicenter evaluating of a commercially available PCR assay for diagnosing enterovirus infection in a panel of cerebrospinal fluid specimens. J Clin Microbiol. 1996;34:3002–3006. 80. Chambon M, Delage C, Bailly JL, et al. Fatal hepatic necrosis in a neonate with echovirus 20 infection: use of the polymerase chain reaction to detect enterovirus in liver tissue. Clin Infect Dis. 1997;24:523–524. 81. Hosoya M, Honzumi K, Suzuki H. Detection of enterovirus by polymerase chain reaction and culture in cerebrospinal fluid of children with transient neu-

RT-PCR Detection of Enteroviruses—Romero

rologic complications associated with acute febrile illness. J Infect Dis. 1997;175: 700–703. 82. Rotbart HA, Ahmed A, Hickey S, et al. Diagnosis of enterovirus infection by polymerase chain reaction of multiple specimen types. Pediatr Infect Dis J. 1997;16:409–411. 83. Kessler HH, Santner B, Rabenau H, et al. Rapid diagnosis of enterovirus infection by a new one-step reverse transcription-PCR assay. J Clin Microbiol. 1997;35:976–977. 84. Kuan MM. Detection and rapid differentiation of human enteroviruses following genomic amplification. J Clin Microbiol. 1997;35:2598–2601. 85. Ahmed A, Brito F, Goto C, et al. Clinical utility of the polymerase chain reaction for diagnosis of enteroviral meningitis in infancy. J Pediatr. 1997;131: 393–397. 86. Rotbart HA. Reproducibility of AMPLICOR enterovirus PCR test results. J Clin Microbiol. 1997;35:3301–3302. 87. Marshall GS, Hauck Ma Buck G, Rabalais GP. Potential cost savings through rapid diagnosis of enteroviral meningitis. Pediatr Infect Dis J. 1997;16: 1086–1087. 88. Andreoletti L, Blassel-Damman N, Dewilde A, et al. Comparison of use of cerebrospinal fluid, serum, and throat swab specimens in diagnosis of enteroviral acute neurological infection by a rapid RNA detection PCR assay. J Clin Microbiol. 1998;36:589–591.

Arch Pathol Lab Med—Vol 123, December 1999

89. Gorgievski-Hrisoho M, Schumacher JD, Vilimonovic N, Germann D, Matter L. Detection by PCR of enteroviruses in cerebrospinal fluid during a summer outbreak of aseptic meningitis in Switzerland. J Clin Microbiol. 1998;36:2408– 2412. 90. van Vliet KE, Glimaker M, Lebon P, et al. Multicenter evaluation of the Amplicor enterovirus PCR test with cerebrospinal fluid from patients with aseptic meningitis: The European Union Concerted Action on Viral Meningitis and Encephalitis. J Clin Microbiol. 1998;36:2652–2657. 91. Koch F, Koch G. The Molecular Biology of Poliovirus. New York, NY: Springer-Verlag; 1985:150–161. 92. Arruda E, Hayden FG. Detection of human rhinovirus RNA in nasal washings by PCR. Mol Cell Probes. 1993;7:373–379. 93. Loeffelholz MJ, Lewinski CA, Silver SR, et al. Detection of Chlamydia trachomatis in endocervical specimens by polymerase chain reaction. J Clin Microbiol. 1992;30:2847–2851. 94. Pichichero ME, McLinn S, Rotbart HA, Menegus MA, Cascino M, Reidenberg BE. Clinical and economic impact of enterovirus illness in private pediatric practice. Pediatrics. 1998;102:1126–1134. 95. Romero JR, Hinrichs SH, Cavalieri SJ, Perry D, Leser JS, Onyeuku J. Potential health care cost savings from PCR-based rapid diagnosis of enteroviral meningitis. Pediatr Res. 1996;39:139A.

RT-PCR Detection of Enteroviruses—Romero 1169