Evaluation of the phadebact gonococcus test, a coagglutination procedure for confirmation of Neisseria gonorrhoeae

Evaluation of eight methods for identification of pathogenic Neisseria species: Neisseria-Kwik, RIM-N, Gonobio-Test, Minitek, Gonochek II, GonoGen, Phadebact Monoclonal GC OMNI Test, and Syva MicroTrak Test

The performance of eight methods in identifying Neisseria species, particularly N. gonorrhoeae, was evaluated. These methods included four rapid carbohydrate utilization tests (Gonobio-Test, Neisseria-Kwik, RIM-N, and Minitek); the Gonochek II, a test which is based on the utilization of chromogenic substrates; and three monoclonal antibody tests (Syva MicroTrak, GonoGen, and Phadebact Monoclonal GC OMNI Test). In all, 182 isolates comprised in six species of Neisseria as well as Branhamella catarrhalis and Moraxella sp. were tested. Cystine-tryptic digest agar supplemented with sugars was included for reference purposes. In the carbohydrate utilization tests, the sensitivity and specificity of the Neisseria-Kwik and Minitek tests for the identification of N. gonorrhoeae were 100%. This compared with sensitivities and specificities, respectively, of 100 and 99.1% for the Gonobio-Test and 99.1 and 100% for cystine-tryptic digest agar sugars and the RIM-N test. The sensitivity and specificity of the Gonochek II test were 99.0 and 86.7%, respectively. Although most test kits did not claim to identify all Neisseria species, in several cases isolates of N. subflava were misidentified or could be misinterpreted as N. gonorrhoeae or N. meningitidis. With the monoclonal reagents, the Syva MicroTrak system was 100% sensitive and 100% specific. The GonoGen test was both 99.1% sensitive and specific, while the Phadebact Monoclonal GC OMNI Test was 99.1% sensitive but 91.2% specific. With this latter test, cross-reactions were observed with strains of B. catarrhalis, N. cinerea, and N. lactamica.

Difficulties in differentiating Neisseria cinerea from Neisseria gonorrhoeae in rapid systems used for identifying pathogenic Neisseria species

Neisseria cinerea and Neisseria gonorrhoeae may occur at the same body sites and may have similar colony morphologies. Ideally, systems used for rapid identification of N. gonorrhoeae should be able to differentiate N. cinerea from gonococci. We tested seven N. cinerea strains using the Gonochek II (Du Pont Diagnostics), Minitek (BBL Microbiology Systems), RapID-NH (Innovative Diagnostics, Inc.), RIM-N (American Microscan), and Phadebact (Pharmacia Diagnostics) systems. We found that the reactions produced by N. cinerea in Gonochek II, Minitek, and RapID-NH kits could be confused with the results produced by some strains of N. gonorrhoeae. The susceptibility of N. cinerea to colistin, its ability to grow on tryptic soy or Mueller-Hinton agar, and its inability to grow on modified Thayer-Martin medium help differentiate it from gonococci.

Gonococcal peritonitis in a patient treated with continuous ambulatory peritoneal dialysis (CAPD)

We report a case of gonococcal peritonitis in a sexually active female on continuous ambulatory peritoneal dialysis. The presumed route of entry into the peritoneal cavity was via the Fallopian tube. The episode responded rapidly to a standard antibiotic treatment protocol. Some episodes of culture-negative peritonitis may be caused by genital tract organisms which are not cultured by routine methods.

Lectin characterization of gonococci from an outbreak caused by penicillin-resistant Neisseria gonorrhoeae

A total of 40 Neisseria gonorrhoeae isolates, representing 19 penicillin-resistant isolates (from 8 heterosexual patients and 11 homosexual patients) and 21 penicillin-susceptible isolates (from 15 heterosexual patients and 6 homosexual patients) and obtained from the same geographic area, were examined. Lectin agglutination patterns were based on the reactivity of the isolates with the following 14 lectins: concanavalin A, Lens culinaris, Trichosanthes kinlowii, Griffonia simplicifolia I, Arachis hypogeae (peanut agglutinin), Glycine max (soybean agglutinin), Dolichos bifloris, Griffonia simplicifolia II, Solanum tuberosum (potato starch agglutinin), Triticum vulgaris (wheat germ agglutinin), Limax flavus, Phaseolus vulgaris, Ulex europaeus I, and Lotus tetragonolobus. All isolates were serotyped with monoclonal antibodies specific for gonococcal outer membrane protein I and auxotyped, and the plasmid content was determined. Resistant patient isolates were selected for their decreased penicillin susceptibility, and control isolates were selected for their penicillin susceptibility. Even though the patient isolates demonstrated resistance to penicillin, no phenotypic differences in lectin-grouping patterns were demonstrated between the two study groups; i.e., two predominant lectin groups were observed. No resistance-associated plasmids were detected. All patient isolates were serogroup IB (serovars IB-1, IB-2, and IB-4), whereas 12 of 21 control isolates were serogroup IA (P less than 0.05). Isolates obtained from different anatomical sites in the same patient (cervical and rectal) agreed with regard to lectin patterns and serovars but not auxotypes.

Epidemiological characterization of Neisseria gonorrhoeae by lectins

A total of 101 isolates of penicillinase-producing and non-penicillinase-producing Neisseria gonorrhoeae with known nutritional requirements, plasmid content, and serovars, were examined for lectin agglutination patterns. These isolates were from outbreaks in Georgia, California, Hawaii, and Pennsylvania. Cell suspensions made from 16- to 18-h cultures were mixed with 14 different lectins, and the resultant agglutination patterns were classified as agglutination groups. Among the 101 isolates tested, 24 different agglutination groups were demonstrated. Of the organisms tested, 55% were located in 3 of the 24 groups, and 86% of the isolates reacted with the lectins Trichosanthes kinlowii, Griffonia simplicifolia I, peanut agglutinin, soybean agglutinin, potato agglutinin, and wheat germ agglutinin. One isolate did not react with peanut or potato agglutinin, five isolates lacked reactivity with potato agglutinin, and six isolates did not react with wheat germ agglutinin. Of the wheat germ-negative isolates, four were from Pennsylvania and were identical with regard to auxotype, plasmid content, serovar, and lectin group. The other two wheat germ-negative isolates were from California and were unrelated by the same criteria to the four Pennsylvania isolates and to each other. Among the isolates tested, there were no differences in lectin groups with regard to the sex of the patient. In the Georgia collection, agglutination with one lectin group was confined to isolates of serogroup IA. This association was not observed for the other geographic areas. Some isolates showing identical auxotype, plasmid content, and serovars could be differentiated based on lectin agglutination patterns, whereas other isolates were identical by all testing criteria.

Evaluation of the RIM system and Gono Gen test for identification of Neisseria gonorrhoeae from clinical specimens

RIM-N (Austin Biological Laboratories) and Gono Gen (Micro Media Systems) were evaluated for accuracy and compared with conventional carbohydrate degradation tests for identification of Neisseria gonorrhoeae. A total of 127 fresh clinical isolates of N. gonorrhoeae were tested; 118 (93%) were identified by RIM-N, and 100 (79%) yielded positive reactions with Gono Gen. Seventy nongonococcal isolates including other Neisseria species, Branhamella catarrhalis, Kingella kingae, and Moraxella species were evaluated to determine the specificity of the test methods. Both rapid tests were 100% specific in the identification of N. gonorrhoeae isolates. RIM-N was the most sensitive test for the identification of N. gonorrhoeae and offers a useful, more rapid alternative to conventional carbohydrate degradation methods.

Validity of an enzyme immunoassay for detection of Neisseria gonorrhoeae antigens

An enzyme immunoassay (EIA; Gonozyme, Abbott Laboratories) for the antigenic detection of Neisseria gonorrhoeae in endocervical or urethral specimens was evaluated. EIA results were compared with results of conventional culture tests for N. gonorrhoeae. Specimens from 208 males (113 culture positive) and 252 females (72 culture positive) were tested. The sensitivity and specificity of EIA for specimens from males were 97.3 and 95.8%, respectively. The sensitivity and specificity of EIA for specimens from females were 79.2 and 87.2%, respectively.

Diagnostic value of interactions between members of the family Neisseriaceae and lectins

The lectin slide agglutination test for Neisseria gonorrhoeae has been modified and improved. Results show that wheat germ agglutinin and soybean lectin agglutinate 100% (193 of 193 tested) of clinical isolates of N. gonorrhoeae. Lectin-reactive meningococci can be readily identified by the hydrolysis of gamma-glutamyl-beta-naphthylamide. Branhamella catarrhalis, Neisseria lactamica, Neisseria sicca, Neisseria subflava, Neisseria perflava, and meningococcal serogroups A, B, C, X, Y, and Z do not interfere with the positive identification of N. gonorrhoeae. The frequently encountered problem of autoagglutination of members of the family Neisseriaceae may be circumvented by a short treatment of cellular suspensions with DNase. Based on agglutination assays, the enzyme treatment did not result in a loss of wheat germ agglutinin receptors from the bacteria. The lectin agglutination test, coupled with the gamma-glutamyl aminopeptidase assay, is proposed as a rapid and accurate means of identifying clinical isolates of gonococci.

Evaluation of Bactogen and Phadebact for detection of Haemophilus influenzae type b antigen in cerebrospinal fluid

Latex agglutination and coagglutination tests are commercially available as Bactogen and the Phadebact Haemophilus Test, respectively. We evaluated both for the detection of Haemophilus influenzae type b in cerebrospinal fluids. Both tests were positive in all of 51 culture-positive cases of meningitis caused by H. influenzae. Both were more sensitive than counterimmunoelectrophoresis. Antigen was also detected by Bactogen in seven of seven additional cerebrospinal fluid specimens (compared with four of seven by Phadebact) after 1 to 15 days of antimicrobial therapy. The cerebrospinal fluid of infants with meningitis owing to other common causative agents did not react with Bactogen or Phadebact. However, the cerebrospinal fluid of one patient with overwhelming infection owing to Proteus mirabilis reacted positively with Bactogen. Cost analysis revealed that Phadebact was less expensive to perform than Bactogen.

Rapidity and reliability of gonococcal identification by coagglutination after culture on modified New York City medium

The combination of culture on modified New York City (MNYC) medium and identification of neisserial isolates by the Phadebact gonococcus test was evaluated in routine laboratory practice. The sensitivity of coagglutination was 96.7% (318/329 isolates) and that of immunofluorescence (IF) 97.9% (322/329 isolates); the specificity of both methods was 96.8% (120/124 isolates). Of the 329 gonococcal isolates, 286 (86.9%) could be tested by coagglutination and 309 (93.9%) by IF after only 24 hours' incubation. Identification by coagglutination from primary cultures on MNYC medium is considered to be very rapid, simple, and efficient for the cultural diagnosis of anogenital gonorrhoea in women and urethral gonorrhoea in men. Because of the high prevalence of meningococci in the pharynx and their not infrequent occurrence in the anorectum of homosexual men the identity of isolates from these sites is best confirmed by sugar utilisation tests.

Identification of Streptococcus pneumoniae by the Phadebact coagglutination test

The Phadebact Pneumococcus Test is a coagglutination slide test for the serological identification of Streptococcus pneumoniae. Of 200 alpha-hemolytic streptococcal isolates, coagglutination test results agreed with those of optochin susceptibility and bile solubility in 189 cases, 105 of which were identified as S. pneumoniae by all three methods. The Phadebact test was 100% (113 of 113) sensitive and 98% (85 of 87) specific and was more sensitive than counterimmunoelectrophoresis in detecting the presence of pneumococcal antigen in cerebrospinal fluid. In fluids seeded with known amounts of pneumococcal antigen, it consistently detected lower levels than did counterimmunoelectrophoresis. The test provides a rapid and simple method for the definitive identification of S. pneumoniae.

Evaluation of the Phadebact Test for identification of Neisseria gonorrhoeae

The Phadebact Gonococcus Test (Pharmacia Diagnostics, Piscataway, N.J.) is used for the identification of Neisseria gonorrhoeae. In this test, boiled organisms are examined by using a 3-min coagglutination technique. A total of 313 isolates of Neisseria confirmed by the fluorescent-antibody technique or identified to the species level by the cysteine-tryptose agar utilization technique, were tested by the coagglutination technique. Of 229 isolates of N. gonorrhoeae, 13 (5.7%) were negative by the coagglutination technique on the first testing, 4 (1.7%) of which remained negative on subsequent testing after subculture. Of the 84 isolates of Neisseria other than N. gonorrhoeae, none showed a positive coagglutination reading. The Phadebact Gonococcus Test provides rapid, accurate identification for most isolates of N. gonorrhoeae. As with any new technique, standard procedures (fluorescent-antibody or cysteine-tryptose agar analysis or both) should be performed along with the new technique until laboratory workers are comfortable with the performance and interpretation of the test. We must, however, emphasize that it is very important to follow carefully the procedures outlined on the package insert.

Evaluation of the Phadebact Gonococcus Test in the identification of Neisseria gonorrhoeae in a routine diagnostic laboratory

The Phadebact Gonococcus Test, a coagglutination procedure for the confirmation of identity of presumptive N gonorrhoeae isolates, was evaluated under normal working conditions in a routine diagnostic laboratory and compared with an immunofluorescence technique. Of 166 isolates of N gonorrhoeae from urogenital, rectal, and pharyngeal sites, 164 gave a positive coagglutination reaction, and one of the two negative isolates gave a positive reaction on retesting after subculture. There were no cross-reactions with other organisms tested. This was in contrast with the immunofluorescence technique; with this method, three of 46 isolates of N meningitidis fluoresced brightly, and a further 17 isolates of N meningitidis gave reactions that were difficult to interpret. The coagglutination test is rapid, simple and a more specific alternative to immunofluorescence.

Evaluation of the coagglutination test for the identification of Neisseria gonorrhoeae in primary cultures

The coagglutination (COA) test for the identification of Neisseria gonorrhoeae was compared with immunofluorescence and sugar degradation tests on 1710 gonococcal isolates, 72 of which produce beta-lactamase. The COA test gave a positive result for 98.6% of the strains. Treatment of suspensions with Streptomyces enzyme reduced the incidence of inconclusive results due to autoagglutination to 1.2%. Cross-reactivity of the gonococcal antiserum was minimised by absorption with meningococci and Moraxella species. The COA provide a simple, quick, and reliable method for identifying N gonorrhoeae in culture.

Enhancement of coagglutination reactions of the Phadebact gonococcus test by ethylenediaminetetraacetate and ethylene glycol-bis(beta-aminoethyl ether)-N,N-tetraacetate

Incubation of gonococci under conditions optimal for autolysis resulted in increased sensitivity and enhancement of the coagglutination reaction of the Phadebact gonococcus test. These conditions included an alkaline pH (pH 8.3) and the presence of divalent cation chelators such as ethylenediaminetetraacetic acid or ethylene glycol-bis(beta-aminoethyl ether)-N,N-tetraacetic acid. Heating cell suspensions at 90 degrees C for 15 min before assay by coagglutination produced a further increase in sensitivity and enhancement of the reaction. Gonococcal lipopolysaccharide was found to be an important antigen in these coagglutination reactions. The detection of lipopolysaccharide was markedly enhanced by the addition of chelating agents.

Comparison of the Phadebact Gonococcus Test with the rapid fermentation method

The Phadebact Gonococcus Test (Pharmacia Diagnostics, Piscataway, N.J.), a coagglutination technique, was compared with the rapid fermentation method of Kellogg and Turner (D. S. Kellogg, Jr., and E. M. Turner, Appl. Microbiol. 25: 550--552, 1973). A total of 93 organisms isolated on Martin-Lewis media were determined to be Neisseria gonorrhoeae based on the following criteria: presence of gram-negative diplococci, oxidase positivity, and appropriate reaction in the rapid fermentation method. These 93 isolates were then serologically tested with the Phadebact test. The direct method was attempted on the first 46 N. gonorrhoeae isolates. Difficulty in interpreting results was encountered in 39%. Thereafter, the alternate method of boiling was instituted on an additional 47 N. gonorrhoeae isolates, with 2 isolates producing noninterpretable results. All 93 isolates were frozen for a maximum of 2 months in skim milk at -25 degrees C. These isolates were thawed and retyped with the alternate boiling procedure, with 97% being confirmed as N. gonorrhoeae. In addition, 33 Neisseria meningitidis isolates, 14 Neisseria species, and 7 Moraxella species were tested with similar techniques. No positive reactions were observed. A cost effectiveness study using 5, 10, and 20 microliters of the gonococcal reagent was undertaken to reduce the cost of the test. When 10 and 20 microliters of reagent were used, no difficulty was encountered in interpreting the reaction. The coagglutination technique was difficult to read when 5 microliters of reagent was used.

The systems approach to diagnostic microbiology

"Classical" and "conventional" methods for microbial identification are still utilized in clinical microbiology laboratories; however, significant advances in methodology have taken place in the last two decades. In the transition from classical to contemporary methodologies, the reference point has changed from multistep procedures to unitary procedures with marked emphasis on standardization, speed, reproducibility, and most recently, mechanization and automation. The most evident expression of this transition is the adaptation or streamlining of classical methods in the form of "miniaturized identification systems" and their commercial availability. This review analyzes the development of the underlining principles that make the systems approach to diagnostic microbiology possible.