Arboviruses associated with mosquitoes from nine Florida counties during 1993

Mosquitoes were collected for virus isolation tests from 36 sites in Bradford, Lake, Leon, Manatee, Orange, Osceola, Pasco, Putnam, and Sarasota counties, FL, from April 6 through October 11, 1993. A total of 158,129 adult specimens were collected in 726 trap nights using CDC light traps, usually baited with dry ice. At least 35 species were represented, although 60% of the collections was made up of 3 species (Aedes infirmatus, 6.5%; Anopheles crucians, 14.4%; and Culex nigripalpus, 39.4%). Four of the 36 collecting sites were located at waste-tire sites, where 254 trap nights yielded 27,455 specimens (17.4% of 9-county total). Forty-three virus strains were isolated from 2,812 mosquito pools consisting of 158,129 specimens. The viruses isolated include eastern equine encephalitis (EEE), 5 strains; Everglades (EVE), 2 strains; Keystone (KEY), 6 strains; Tensaw (TEN), 21 strains; trivittatus (TVT), one strain; Shark River (SR), one strain; and Flanders (FLA), one strain. In addition, 2 strains that are either KEY or Jamestown Canyon (JC) virus, and 4 ungrouped viruses remain to be identified. Twenty-one (48.8%) of the 43 virus strains were isolated from mosquitoes collected at waste-tire sites as follows: EEE (1), KEY (5), KEY/JC (1), TEN (13), and one ungrouped virus. The vector relations of the viruses are discussed and the potential importance of waste-tire sites as breeding habitats and harborages for vector and nuisance species is emphasized.

Detecting bunyaviruses of the Bunyamwera and California serogroups by a PCR technique

Many bunyaviruses of the Bunyamwera and California serogroups are medically important human pathogens. The development of an effective technique to detect the viruses by using molecular biologic tools, such as PCR, improves not only clinical diagnosis but also virologic surveillance of mosquito vectors in the field. In this study, we evaluated eight pairs of primers for reactivity with 44 viruses of the genus Bunyavirus, using a reverse transcriptase PCR technique. With a pair of serogroup-specific primers we designed, all viruses of the serogroups tested could be detected. Further, virus-specific primer pairs were identified for California encephalitis virus, Jamestown Canyon virus, La Crosse virus, and snowshoe hare virus for use in North America. Using this technique, we could detect one La Crosse virus-infected mosquito in a pool of 100 mosquitoes with undetectable plaque titers.

Isolations of Potosi virus from mosquitoes collected in the United States, 1989-94

Potosi (POT) virus, a recently characterized Bunyamwera serogroup virus, was discovered when it was isolated from Aedes albopictus collected at a waste-tire site in Potosi, Washington County, Missouri, during 1989. During the following year, POT virus was not isolated from 39,048 mosquitoes, including 17,519 Ae. albopictus, collected in Washington County. In 1991, mosquito collections from South Carolina, Ohio, and Michigan yielded 8 strains of POT virus: 6 from Coquillettidia perturbans and one each from Culex restuans and Psorophora columbiae. Additional collections of Ae. albopictus from several states during 1990-93 failed to yield further isolates of POT virus. In 1994, POT virus was isolated from Ae. albopictus and Anopheles punctipennis in North Carolina and from Ae. albopictus in Illinois. These represent the first virus isolations of any type for Ae. albopictus in those states. Thus far, POT virus has been isolated from 5 mosquito species in different genera in 6 states. The known geographic range of POT virus, based on virus isolations, has been extended from Missouri to the upper Midwest and the Atlantic seaboard. Potential vector relationships and possible transmission cycles of POT virus are discussed.

Assessment of arthropod vectors of infectious diseases in areas of U.S. troop deployment in the Persian Gulf

Beginning in August 1990, approximately 800,000 coalition troops were deployed to the Persian Gulf during Operations Desert Shield and Desert Storm. There was substantial concern about arthropod-borne diseases, particularly sand fly fever and cutaneous leishmaniasis, because of high morbidity rates in the Persian Gulf during World War II (WWII). In sharp contrast to WWII, there was no report of sand fly fever among coalition forces and only 31 cases of leishmaniasis among 697,000 U.S. troops. To further evaluate the risk of arthropod-borne diseases, an entomologic survey was conducted in 12 areas of Kuwait and Saudi Arabia. A total of 1,556 arthropods was collected during four survey periods in 1992. The suspected vectors of cutaneous Leishmania major infection, sand fly fever, West Nile fever, Rift Valley fever, and Crimean-Congo hemorrhagic fever were identified; however, there was no evidence of arboviruses or Leishmania among collected specimens nor from 51 trapped rodents. There are several possible reasons for the low risk of arthropod-borne infectious diseases among Desert Shield/Storm troops in an area where suspected vectors frequently were found: the use of insecticides and repellents, and the deployment of most ground troops to the open desert during the cooler, winter period--conditions least favorable for the transmission of arthropod-borne diseases.

The role of Aedes albopictus as an arbovirus vector

The recent rapid spread of Aedes albopictus to many areas previously free of such infestations, including Albania and Italy, has heightened concern among public health and vector control officials. Ae. albopictus is a proven or potential vector of several arboviruses, including some that are, or have been, actively transmitted in southern Europe. These virus/vector relationships are reviewed with a view toward assessing the risk posed by the current presence of Ae. albopictus in the region.

Preoperative total parenteral nutrition is not associated with mucosal atrophy or bacterial translocation in humans

Concerns have recently been expressed at suggestions that postoperative sepsis may be more common in patients who have received preoperative total parenteral nutrition (TPN). The mechanism suggested for this is that TPN causes intestinal mucosal atrophy leading to increased bacterial translocation from the gut as a source of systemic sepsis. This hypothesis was examined in 203 patients who had an elective laparotomy, 28 of whom required at least 10 days of preoperative TPN. Neither mucosal atrophy nor bacterial translocation was more common in parenterally fed patients than in enterally fed controls. In humans theoretical concerns about the adverse effects of TPN on intestinal integrity are unfounded.

Experimental infection of wading birds with eastern equine encephalitis virus

To study the susceptibility of wading birds to eastern equine encephalitis (EEE) virus and to determine their potential as reservoir or amplifying hosts, fledgling glossy ibises (Plegadis falcinellus) and snowy egrets (Egretta thula) were captured in New Jersey (USA) and shipped to Colorado (USA) where they were experimentally inoculated with EEE virus. All 16 snowy egrets and 14 (93%) of 15 of the glossy ibises inoculated became viremic with moderate titers, and all survivors developed neutralizing antibody. Six ibises and two egrets died during the first week after inoculation, and EEE virus was isolated from the tissues of three birds. Our experimental results support field evidence about the relative involvement of glossy ibises and snowy egrets in the epizootiology of EEE virus in New Jersey.

Isolation of potosi virus from Aedes albopictus in North Carolina

A total of 4,169 adult mosquitoes were aspirated from 3 tire disposal sites in North Carolina for virus assays. Aedes albopictus was the dominant species, with a relative abundance of approximately 99%. Potosi virus was isolated from one pool of 68 female Ae. albopictus. Priorities for future Potosi virus research and the implications of the North Carolina isolate are discussed.

Sensitive and specific colorimetric dot assay to detect eastern equine encephalomyelitis viral RNA in mosquitoes (Diptera: Culicidae) after polymerase chain reaction amplification

A sensitive and specific colorimetric dot assay following polymerase chain reaction (PCR) method has been developed to detect 0.1 pg of eastern equine encephalomyelitis viral (EEEV) RNA. The assay is 250-fold more sensitive than analysis by electrophoresis and is based on converting a 291-nucleotide sequence of the viral coat protein amino terminus into a double-stranded DNA (dsDNA) and amplifying the DNA using a specific primer pair and PCR. The amplified complementary DNA (cDNA) is denatured adsorbed onto a nylon strip, baked, and detected with a digoxigenin-labeled probe. Dots with viral cDNA are stained dark red, whereas controls do not stain or stain lightly. The assay is very specific and sensitive and detects only EEEV. RNA of Venezuelan equine encephalitis, St. Louis encephalitis, Keystone, Flanders, Tensaw, and western equine encephalitis viruses were not detected. EEEV (Ten Broeck) RNA was detected at the 10-ng level, indicating that the prototype we used may have different nucleotides in the region where the primer pair binds. The PCR amplified EEEV cDNA that was 92% homologous to the consensus sequence of EEEV. The detection of EEEV in the liver of an infected Emu bird and in field-collected mosquitoes from Florida and Massachusetts that were analyzed concurrently as blind samples by tissue culture plaque assay and by PCR dot analysis proved that the assay is sensitive and can be used to detect infected mosquitoes. The assay can detect at least 1 infected mosquito in a pool of 1,000 uninfected mosquitoes.

Vector competence of Aedes albopictus from Pine Bluff, Arkansas, for a St. Louis encephalitis virus strain isolated during the 1991 epidemic

The vector competence of Aedes albopictus from Pine Bluff, AR, was assessed for a St. Louis encephalitis (SLE) virus strain isolated during the 1991 epidemic. Aedes albopictus were fed on hamsters with viremia levels of 10(4.6)-10(4.9) Vero cell plaque-forming units (PFU)/ml. At 7 and 15 days postbloodfeeding, transmission trials were conducted using individual suckling mice. Three of 313 Ae. albopictus were determined to be infected with SLE virus with titers of 10(6.3)-10(7.0) PFU/mosquito. At 15 days postbloodfeeding, one of 209 Ae. albopictus that refed transmitted virus resulting in a 15-day population transmission rate of 0.5%. The infection threshold (i.e., the amount of virus required to infect from 1 to 5% of mosquitoes) was determined to be approximately 10(2.3) PFU/mosquito. Virus inoculated intracoelomically into Ae. albopictus replicated and reached mean titers above 10(6.0) PFU/mosquito on day 6. The combination of low susceptibility to infection and a mammalophilic bloodfeeding pattern suggests that Ae. albopictus is unlikely to play a significant role in SLE transmission.

The prevalence of gut translocation in humans

BACKGROUND/AIMS

Gut translocation of enteric organisms across the intact intestinal mucosa has been postulated as a potential source of sepsis in susceptible patients. However, little is known of its occurrence or significance in humans. The aim of this study was to determine the prevalence of gut translocation of bacteria in humans and attempt to identify any predisposing factors to its occurrence.

METHODS

A consecutive series of 267 general surgical patients were examined for evidence of bacterial translocation by bacterial analysis of intestinal serosa and mesenteric lymph nodes taken at the time of surgery.

RESULTS

Translocation occurred in 10.3% of patients overall. Both aerobic and anaerobic bacteria translocated. Excluding patients with distal intestinal obstruction and those with inflammatory bowel disease in whom translocation was more common, the prevalence was 5%. Neither jaundice, nutritional status, nor total parenteral nutrition predisposed to translocation. Similarly, mucosal atrophy did not predispose to this phenomenon. The development of postoperative septic complications was twice as common in patients with translocation as in those without, but mortality was unaffected.

CONCLUSIONS

Translocation occurs as a spontaneous event in humans, but its clinical significance remains to be defined.

PCR-based detection of arboviral RNA from mosquitoes homogenized in detergent

An improved method for the extraction of viral RNAs was developed to facilitate the reverse transcription (RT)-PCR detection of mosquitoes infected with Western equine encephalitis virus or La Crosse virus. The solubilization method, which uses only EDTA and sodium dodecyl sulfate (SDS) followed by dilution of sample, allows accurate viral detection through the use of random hexamers for the RT followed by specific primers for the PCR. Identities of the reaction products were confirmed either by sequencing or restriction endonuclease digestion. Previous methods for the extraction of RNA for the coupled RT-PCR depended on combinations of guanidinium isothiocyanate, acid phenol, detergents and multiple centrifugations. Ideally, routine detection of viral RNAs for diagnostic purposes should bypass many of the above steps, while still providing a sensitive assay. Our level of detection is 1 infected mosquito in a group of 100.

Analysis of a yellow fever virus isolated from a fatal case of vaccine-associated human encephalitis

The virulence of a yellow fever (YF) virus (P-16065) isolated from a fatal case of vaccine-associated viral encephalitis was investigated. P-16065 appeared identical to its parent vaccine virus (17D-204 USA, lot 6145) when examined with monoclonal antibodies except that YF wild type-specific MAb S24 recognized P-16065 but not 17D-204 USA 6145. Thus, a mutation of at least one epitope on the envelope (E) protein had occurred. Unlike 17D-204 USA 6145 and other 17D vaccine viruses, P-16065 was neuroinvasive and virulent for mice after intranasal inoculation, and neurovirulent for monkeys after intracerebral inoculation. The E protein of P-16065 differed from 17D-204 USA by two amino acids at positions 155 and 303. Changes at amino acid position 155 are found in other YF vaccine viruses that are not neurovirulent, and it is therefore postulated that the change at position 303 is involved in the alteration of the phenotype of P-16065 and may be important for virulence of YF virus.

An integrated target sequence and signal amplification assay, reverse transcriptase-PCR-enzyme-linked immunosorbent assay, to detect and characterize flaviviruses

We previously described a reverse transcriptase-PCR using flavivirus genus-conserved and virus species-specific amplimers (D. W. Trent and G. J. Chang, p. 355-371, in Y. Becker and C. Darai; ed., Frontiers of Virology, vol. 1, 1992). Target amplification was improved by redesigning the amplimers, and a sensitive enzyme-linked immunosorbent assay (ELISA) technique has been developed to detect amplified digoxigenin (DIG)-modified DNA. A single biotin motif and multiple DIG motifs were incorporated into each amplicon, which permitted amplicon capture by a biotin-streptavidin interaction and detection with DIG-specific antiserum in a colorimetric ELISA. We evaluated the utility of this assay for detecting St. Louis encephalitis (SLE) viral RNA in infected mosquitoes and dengue viral RNA in human serum specimens. The reverse transcriptase-PCR-ELISA was as sensitive as isolation of SLE virus by cell culture in detecting SLE viral RNA in infected mosquitoes. The test was 89% specific and 95 to 100% sensitive for identification of dengue viral RNA in serum specimens compared with isolation of virus by Aedes albopictus C6/36 cell culture and identification by the indirect immunofluorescence assay.

Larval diet, adult size, and susceptibility of Aedes aegypti (Diptera: Culicidae) to infection with Ross River virus

The relationship of larval nutrition and adult body size to the susceptibility of Aedes aegypti (L.) to Ross River virus infection was examined. Large adult mosquitoes produced by feeding larvae a high-level diet consumed significantly more virus particles than did smaller mosquitoes. However, when a correction for body size was made, smaller mosquitoes were found to consume significantly more virus per unit of body weight. A host viremia of 2.4 log10 PFU/ml failed to infect mosquitoes of any size. Large Ae. aegypti mosquitoes were significantly more susceptible than small mosquitoes when fed on hosts with viremias of 4.8, 5.7, 6.4, and 7.5 log10 PFU/ml, but differences in susceptibility were less apparent at higher viremias.

Evaluation of Questor urine screening system for bacteriuria and pyuria

AIMS

To evaluate the Questor automated bacteriuria and pyuria screening system; to compare its performance with that of a reference method; and to assess its usefulness in a routine clinical laboratory.

METHODS

The Questor urine screening system was compared with a comprehensive regimen to detect urinary tract infection, using pour-plate viable counts to determine the numbers of bacteria present in urine samples, a wide range of other cultural methods, microscopic findings and clinical information.

RESULTS

The optimal performance in detecting significant growths was a sensitivity of 93%, a specificity of 74%, a positive predictive value of 43% and a negative predictive value of 98%. The list price per test is 0.17 pounds and the capital cost of the system is 39,950 pounds. Questor can test 50 samples an hour and can be operated by one member of the laboratory staff, who is not required to make interpretative judgments--for example, a medical laboratory assistant.

CONCLUSIONS

The sensitivity and specificity of the Questor was better than that obtained from other screening systems using the same protocol. The system was easy to use and is a useful addition to the methods available for screening for bacteriuria.

Total parenteral nutrition by peripheral vein — Substitute or supplement to the central venous route? A prospective trial

51 consecutive gastroenterological patients who required total parenteral nutrition (TPN) were entered into this study. Two patients were withdrawn because of specific nutritional requirements, leaving 49 patients for randomisation. 23 patients were allocated to receive peripheral parenteral nutrition (PPN) and 26 to receive feeding through a central venous line (CPN). There was no significant difference between the groups with respect to the median duration of feeding (9.4 +/- 3.6 days; 12.0 +/- 7.8 days) but significant morbidity occurred more frequently in the CPN group (11%) compared to the PPN group (0%). TPN by the designated route was not possible in 4 patients in the PPN group and in 3 of the CPN group. Of the 19 patients commenced on PPN, 13 continued without complication until resumption of oral feeding (median 10.7 +/- 3.2 days); 6 of these patients had to be converted to central venous feeding for completion of their nutritional requirements. Of the 23 patients commenced on CPN, 21 completed their nutritional course (median 11.8 +/- 5.3 days), 2 patients in the CPN group required conversion to PPN to complete their nutritional course. This study shows that PPN is a feasible, safe alternative to CPN in many patients. It is not necessary to subject all patients who require TPN to the risks and expense of central venous cannulation.

Vector and host relationships of California serogroup viruses in western Siberia

During 1990 and 1991, adult mosquitoes were collected along the Ob River and its tributaries in western Siberia from approximately 51 degrees 18'N to 66 degrees 4'N. Fifteen virus strains were isolated from 74,196 mosquitoes tested in 1,874 pools. These included Tahyna virus from Aedes cataphylla-punctor subgroup (one) and Ae. excrucians (one), and Inkoo (INK) virus from Ae. communis (one), Ae. communis subgroup (one), Ae. hexodontus (two), Ae. punctor subgroup (two), Ae. punctor complex (one), and unidentified Aedes species (three). In addition, a single Ae. euedes yielded a strain of snowshoe hare (SSH) virus and a strain of Getah, an alphavirus. A Bunyamwera serogroup virus was isolated from Ae. excrucians. With the exception of the two isolates from a single mosquito, minimum infection rates among mosquito taxa ranged from 0.4 to 16.7 per 1,000. The INK virus isolates were widely distributed geographically; however, seven of the 10 isolates were from two sites north of the Arctic Circle. During 1991, sera from two mouse species, five vole species, and four shrew species were collected along the upper Ob River for serologic tests. The prevalence of neutralizing antibody to SSH virus in these sera was 80%. Prevalence rates in the four most abundant species were Apodemus agrarius, 73%; Clethrionomys rutilus, 71%; Microtus arvalis, 80%; and Sorex araneus, 91%. This is the first attempt to clarify the vector and vertebrate host relationships of California serogroup viruses in western Siberia.

Entomologic investigations of an epidemic of St. Louis encephalitis in Pine Bluff, Arkansas, 1991

An epidemic of St. Louis encephalitis (SLE) occurred in Jefferson County, Arkansas during July-August 1991. At least 26 human cases were involved, with 25 cases in the town of Pine Bluff. Twelve isolates of SLE virus were obtained from mosquitoes collected in Pine Bluff between August 13 and 24: 11 from pools of Culex pipiens quinquefasciatus, resulting in a minimum infection rate of 1.6 per 1,000 (n = 6,768) for this subspecies, and one isolate from a pool of 22 mosquitoes identified as Cx. (Culex) spp. Three of the SLE-positive pools, two from Cx. p. quinquefasciatus and one from Cx. (Cux.) spp., also yielded isolates of Flanders virus. Larval surveys resulted in the collection of seven species in four genera from 28 larva-positive habitats and the identification of one significant site of Cx. p. quinquefasciatus production. Ecologic assessments conducted at 12 randomly selected residences resulted in the identification of 17 larva-positive habitats, for an average mosquito-positive habitat rate of 1.4 per residence, and a Cx. p. quinquefasciatus larva-positive habitat rate of 0.6 per residence. Aedes albopictus and Cx. p. quinquefasciatus were the species most frequently encountered in larval surveys in residential neighborhoods.

Comparison of Signal and Bactec NR-660 blood culture systems

The Signal blood culture system was compared with the Bactec NR-660. A total of 1617 blood culture sets yielded 143 (8.8%) significant isolates; 113 (79.0%) were from positive bottles in both the Bactec and Signal systems. Twelve organisms (8.4%) were detected and isolated from the Signal system only and another 18 (12.6%) from the Bactec system only. Of these 18, five were Signal-positive but the organism was not recovered and four organisms were isolated from negative Signal bottles on terminal subculture. The time taken to detection for each system was similar; the Signal system detected 68% and the Bactec 63% of significant positives within 24 h. At 48 h Bactec detected 91% and the Signal 85%. A significantly-reduced number of bottles which gave a positive signal but were negative by microscopical and cultural methods was found, compared with previous reports. The 1 h incubation period prior to the insertion of the Signal growth indicator device was considered to be the cause of this reduction in the proportion of false positives. Fifty-five percent (42/77) of the Bactec false positives were due to delta growth value. This is when there is an increase in the growth index of > or = 15 without the positive threshold level of 30 being attained. This occurred in the anaerobic bottle on day 2 with 42 bottles. Another 40% (31/77) of the false positives had a growth value between the positive threshold of 30 and a value of 35. Eighty (4.9%) of Bactec and 65 (4.0%) of Signal sets yielded clinically non-significant isolates.(ABSTRACT TRUNCATED AT 250 WORDS)

Japanese encephalitis on Saipan: a survey of suspected mosquito vectors

An outbreak of Japanese encephalitis (JE) occurred on Saipan, Commonwealth of Northern Mariana Islands, in October 1990. Adult and larval mosquitoes were collected during September-October 1991 to retrospectively determine the probable mosquito vector(s). Virus was not isolated from 119 mosquito pools composed of 7,250 adult specimens as follows: Aedes vexans nocturnis (14%), Culex tritaeniorhynchus (39%), Cx. sitiens group (11%), Culex (Culex) species (35%), and < 1% each of Ae. albopictus, Ae. oakleyi, Aedes saipanensis, Cx. annulirostris marianae, and Cx. fuscanus. Three additional species were collected only as larvae: Anopheles indefinitus, Ae. neopandani, and Cx. quinquefasciatus. Among the vectors of JE incriminated in other areas, Cx. tritaeniorhynchus was the predominant species in our collections and the principal species feeding on swine. This is the first published record of the occurrence of this species on Saipan. Culex tritaeniorhynchus is abundant and widely distributed on the southern half of Saipan where human JE cases occurred in 1990, and where swine seroconversions were detected. Although the identity of the mosquito vector(s) responsible for the 1990 outbreak cannot be established with certainty, our results suggest that Cx. tritaeniorhychus was probably involved.

Host-feeding patterns of Aedes albopictus (Diptera: Culicidae) at a temperate North American site

Precipitin tests and ELISA were used to investigate host-feeding patterns of 172 blood-fed Aedes albopictus (Skuse) collected at Potosi, MO, during the summers of 1989 and 1990. One hundred ten (64.0%) mosquitoes had fed on mammals, 29 (16.9%) on birds, and none on turtles or snakes. Thirty-three (19.2%) mosquitoes failed to react in all tests. Eighty-six (78.2%) of the 110 mammalian feeds were positive for lower taxa as follows: rabbit, 24.5%; deer, 14.5%; dog, 13.6%; human, 8.2%; squirrel, 7.3%; opossum, 4.5%; myomorph rodents other than Rattus, 3.6%; raccoon, 0.9%; and bovine, 0.9%. Positive feeds were not detected for the following mammals: cat (n = 99); horse (n = 95); Rattus (n = 84); and swine (n = 84). Fourteen (48.3%) of the 29 avian feeds were positive for lower taxa as follows: Passeriformes, 24.1%; Columbiformes, 17.2%; Ciconiiformes, 3.4%; and quail, 3.4%. These data, the first on host-feeding patterns for Ae. albopictus populations in the New World, indicate that Ae. albopictus is an opportunistic feeder that utilizes a wide variety of hosts and, therefore, has the potential to become involved in the transmission cycles of indigenous arboviruses.

Susceptibility parameters of Aedes albopictus to per oral infection with eastern equine encephalitis virus

Aedes albopictus (Skuse) mosquitoes were fed on snowy egrets, Egretta thula (Thayer and Bangs), that had been infected by subcutaneous inoculation of eastern equine encephalitis (EEE) virus. Freshly fed mosquitoes were frozen and tested to determine how much virus they had ingested. Other fed mosquitoes from the same lots were incubated for 7 d at 27 degrees C before testing. Seven lots of Ae. albopictus fed on viremic birds. Based on average amounts of virus ingested and day 7 virus infection rates in mosquitoes from the same lots, the amount of virus required to infect 50% of the mosquitoes was calculated to be 10(2.8) Vero cell plaque-forming units (PFU). The infection threshold (i.e., the amount of virus required to infect from 1 to 5% of mosquitoes) was determined to be < or = 10 PFU per blood meal. These parameters indicate that Ae. albopictus is sufficiently susceptible to infection with EEE virus to enable it to acquire infectious doses from a wide variety of viremic birds and possibly from equines.

Comparison of Sentinel and Bactec blood culture systems

AIMS

To evaluate the Sentinel automated blood culture system and to compare its performance with that of Bactec.

METHODS

The Sentinel blood culture system was evaluated in three centres. The performance of the system was assessed in comparison with the routine blood culture method used in these centres, the Bactec system.

RESULTS

Blood culture sets (n = 2180) consisting of Sentinel aerobic and anaerobic, and Bactec aerobic and anaerobic bottles yielded 218 (10%) clinically important isolates. One hundred and fifty five (71%) of the isolates were detected by both systems; 35 (16%) were detected by Sentinel only; and 28 (13%) by Bactec only. For the duration of the evaluation, the Sentinel system was deliberately configured so that it was impossible to detect positive results during the first 12 hours. The times to positivity after the first 12 hours were similar. Data gathered during and subsequent to the evaluation have been used by the manufacturer to refine the algorithm so that positive results can be detected at a minimum of 2.25 hours.

CONCLUSIONS

After a period of familiarization the Sentinel system was considered easy to use. Sentinel is a useful addition to the methods available for the detection of bacteria in blood cultures.