Evaluation of the MycoAKT Latex Agglutination Test for Rapid Diagnosis of Mycobacterium avium Complex Infections

Evaluation of the MycoAKT Latex Agglutination Test for Rapid Diagnosis of Mycobacterium avium Complex Infections

Evaluation of the MycoAKT Latex Agglutination Test for Rapid Diagnosis of Mycobacterium avium Complex Infections Juan P. Olano, Heather Holmes, and Ga...

142KB Sizes 0 Downloads 65 Views

Evaluation of the MycoAKT Latex Agglutination Test for Rapid Diagnosis of Mycobacterium avium Complex Infections Juan P. Olano, Heather Holmes, and Gail L. Woods

Rapid diagnosis of Mycobacterium avium complex (MAC) bacteremia is important for management of patients with the acquired immunodeficiency syndrome who have disseminated MAC. The purpose of this study was to determine the reliability of the MycoAKT latex agglutination test for direct detection of MAC in positive mycobacterial blood cultures. First, colonies of isolates of previously identified mycobacteria, including 35 MAC, were tested. Of the 55 isolates evaluated, 33 were identified as MAC by the latex test, including 31 of the known MAC and 2 M. chelonae (sensitivity, 88.6%; specificity, 90.0%). Second, broth from 20 ESP II and 20 BACTEC 12B bottles seeded with isolates of MAC were tested. Aliquots from 19 (95%) ESP II cultures and 16 (80%) 12B cultures were positive by the latex test. In phase 3, broth from 115

signal-positive ESP II blood cultures were tested by latex agglutination. Forty-three subcultures from these bottles grew mycobacteria (41 MAC and 2 Mycobacterium tuberculosis complex); the remainder grew no organisms. Broth from 40 of the blood cultures (39 that grew MAC and 1 from which no organisms were recovered) were latex positive; thus, the sensitivity, specificity, and positive and negative predictive values of the latex test for direct identification of MAC in ESP II blood cultures were 95.1, 98.6, 97.5, and 97.3%, respectively. The mean time to detection of MAC was 14.6 days (range, 6–34 days) with the direct latex test, compared with 18.3 days (range, 9–36 days) with subculture and probe (p , 0.05). © 1998 Elsevier Science Inc.

INTRODUCTION

identification of MAC was nucleic acid probes (GenProbe, Inc., San Diego, CA). However, recently a latex agglutination test (MycoAKT™, Remel, Lenexa, KS) that utilizes monoclonal antibodies against a phosphoglycolipid in the MAC cell wall became available (other latex tests can be used for identification of Mycobacterium tuberculosis complex and Mycobacterium kansasii) for testing colonies of mycobacteria on solid media. The purpose of this study was to determine the reliability of the latex test for identification of MAC directly in aliquots of broth removed from ESP II bottles (AccuMed International [formerly Difco Laboratories], Westlake, OH) that have been signaled as positive by the ESP II instrument. In the first phase, colonies of mycobacteria (both MAC and other mycobacteria) were tested. Second, aliquots of broth from ESP II bottles and BACTEC 12B bottles seeded with MAC (Becton Dick-

Disseminated infection with Mycobacterium avium complex (MAC) causes substantial morbidity and mortality in patients with the acquired immunodeficiency syndrome (Benson 1994). Given that relatively effective antimicrobial therapy is available for disseminated MAC, rapid and accurate diagnosis is important (Shafran et al. 1996). In most cases, diagnosis of disseminated MAC is based on a positive mycobacterial blood culture (Woods 1994). For many years, the only commercial kit method for rapid From the Department of Pathology, University of Texas Medical Branch, Galveston, Texas 77555-0740, USA. Address reprint requests to Juan P. Olano, Department of Pathology, University of Texas Medical Branch, Galveston, TX 77555-0740, USA. Received 11 July 1997; accepted 18 September 1997.

DIAGN MICROBIOL INFECT DIS 1998;30:71–74 © 1998 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010

0732-8893/98/$19.00 PII S0732-8893(97)00191-0

J.P. Olano et al.

72 inson, Sparks, MD) were evaluated. Lastly, patient blood cultures inoculated to ESP II bottles were tested when signaled as positive by the ESP instrument.

Phase One (Colonies on a Solid Medium) Fifty-five isolates, including 35 MAC and 20 other mycobacteria (1 M. gordonae, 1 M. terrae, 1 M. scrofulaceum, 5 M. chelonae, 1 M. phlei, 1 M. fortuitum, 3 M. kansasii, 2 M. xenopi, and 5 M. tuberculosis) that had been identified previously by nucleic acid probes (AccuProbe, Gen-Probe, Inc., San Diego, CA), conventional biochemical tests, high performance liquid chromatography (performed at the Texas Department of Health), or a combination of the latter two techniques, were subcultured to Lowenstein-Jensen slants. The latex agglutination test was then performed according to the manufacturer’s directions. Briefly, 200 mL of methanol were added to an extraction tube containing glass beads (provided with the latex kit). Two to three colonies were scraped from the surface of the slant and mixed in the methanol. Tubes were vigorously agitated on a vortex mixer for 30 s, placed in a dry heat block for 30 min at 50°C, and centrifuged at 17,000 3 g for 5 min. The supernatant was transferred to another tube, which was heated in a dry heat block at 85°C until the methanol completely evaporated (at least 15 min). Extraction reagent (100 mL) was added, and tubes were vigorously agitated on a vortex mixer for 20 s and then centrifuged at 17,000 3 g for 10 min. A 50-mL aliquot was transferred to a latex card, which was rocked on a three-dimensional rotator for up to 13 min. Positive and negative controls were included with each batch of isolates tested, as specified by the manufacturer. If a discrepancy between the latex agglutination test result and the previous identification (i.e., MAC or not MAC) occurred, the latex test was repeated; and if the isolate had been identified as MAC, the probe also was repeated.

Phase 2 (Seeded Broth Media) Suspensions of 20 MAC isolates (previously identified by AccuProbe) equal to the density of a 1.0 McFarland standard were prepared in sterile phosphate buffered saline and serially diluted to 1:10,000. A 0.1-mL volume of the 1:10,000 dilution was inoculated to a BACTEC 12B bottle and to an ESP II bottle. BACTEC 12B bottles were incubated at 37°C and monitored by the BACTEC 460 instrument until the growth index reached $999. ESP II bottles were placed in the ESP instrument until they signaled positive. Bottles meeting these criteria were vigorously agitated on a vortex mixer for 20 s, and 1.5 mL of broth were transferred to a latex kit extraction

tube, which was centrifuged at 17,000 3 g for 5 min. The supernatant was removed, and as recommended by the manufacturer, the tube was frozen at 220°C for 20 min. The latex test was then performed according to the manufacturer’s directions as described for phase 1.

Phase 3 (Patient Blood Cultures) One hundred fifteen blood cultures signaled as positive by the ESP II instrument between August and November 1996 were included in the study. Blood specimens for mycobacterial culture were collected in Isolator tubes (Wampole Laboratories, Cranbury, NJ), processed according to manufacturer’s directions, and inoculated (0.5 mL) into an ESP II bottle. Bottles were placed in the ESP II instrument, where they were continuously monitored for bacterial growth for 6 weeks or until signaled as positive. When a bottle signaled positive, it was vigorously agitated on a vortex mixer for 20 s, aliquots were removed for preparation of a smear that was stained for acid-fast bacilli (AFB) and for subculture to a Middlebrook 7H11 agar biplate (incubated for 6 weeks or until colonies appeared), and 1.5 mL were transferred to an extraction tube for latex testing. The tube was then processed for latex agglutination as described for the seeded bottles in phase 2. Latex test results were compared with results of subculture to a solid medium and identification of colonies by AccuProbe (for MAC and M. tuberculosis complex).

Statistical Analysis Times to identification by direct testing of aliquots of broth from signal-positive ESP II bottles and by the usual laboratory protocol described for phase 3 were compared by using the Student’s t-test. For phase 1 of this study, colonies of 33 of the 55 mycobacteria were identified as MAC by the latex test. Thirty-one of 33 were isolates previously identified as MAC, and two were previously identified as M. chelonae. Results of repeat latex testing of the four latex-negative MAC, which were identified as MAC when retested by AccuProbe, and the two latexpositive M. chelonae were identical to initial results. The sensitivity and specificity of the latex test when testing colonies on a solid medium were 88.6 and 90.0%, respectively. In phase 2, 19 of the 20 (95%) seeded ESP II bottles were positive by the MAC latex test, compared with only 16 (80%) of the 12B bottles. In the last phase of the study, smears from 42 of the 115 signal-positive ESP II blood cultures were positive for AFB, and 43 subcultures (37.4%) grew a mycobacterium (summarized in Table 1). Isolates from 41 of the positive cultures were identified as MAC (smears from all except one were AFB posi-

Rapid Diagnosis of MAC Infections TABLE 1

73

Latex Agglutination for Direct Detection of Mycobacterium avium Complex (MAC) in Positive ESP II Blood Cultures MAC Latex Result

Final Blood Culture Result

Positive

Negative

Total

Positive for MAC Negative for MAC Total

39 1a 40

2 73b 75

41 74 115

a b

No organisms were recovered. Includes two M. tuberculosis isolates.

tive), and two were M. tuberculosis complex. No organisms were recovered from the remaining 72 (63.5%) cultures. Forty ESP II aliquots were positive for MAC by the latex test, including 39 from cultures from 29 patients in which MAC was identified and 1 from a culture that was negative. Thus the sensitivity, specificity, and positive and negative predictive values of the latex test for direct identification of MAC in signal-positive ESP II cultures of blood were 95.1, 98.6, 97.5, and 97.3%, respectively. The mean time to identification of MAC with the usual laboratory protocol was 18.3 days (range, 9 –36 days) compared to a mean of 14.6 days (range, 6 –34 days) for direct identification by the latex test (p , 0.05). When testing colonies on a solid medium, the performance of the MAC latex test in our hands was comparable to the manufacturer’s claims. The latex test did not identify all MAC isolates; this also is true of AccuProbe, which has a sensitivity of 95 to 97% for identification of MAC when testing colonies on a solid medium (Goto et al. 1991). Included among the 35 MAC tested were 4 isolates from 4 separate patients that had been negative by AccuProbe on 2 separate occasions. These isolates were subsequently referred to the Texas Department of Health, where they were identified as MAC by high performance liquid chromatography and conventional biochemical tests. Three of these four isolates were positive with the MAC latex test. In contrast to AccuProbe, however, the MAC latex test is positive with some isolates of M. chelonae (a limitation acknowledged by the manufacturer). In practice this should not be an issue because the colonies of MAC and M. chelonae grow at different rates on solid media and generally have a different morphologic appearance; however, it rarely could create a problem when the latex test is used directly to test a positive liquid blood culture. Results of our experiments in which we tested liquid cultures seeded with MAC were unexpected. They also differ somewhat from the results of a smaller study in which broth from eight of eight BACTEC 12B bottles seeded with MAC were latex positive (Greene and Gleason-Beavis 1997). At this

time, the specific reason for the difference we found in performance between ESP II and 12B cultures is unknown, but it is possible that the nutrient composition of the media plays a role. Based on our findings, we chose not to proceed with testing patient specimens in 12B bottles; and until additional data are available, we do not recommend using the latex test directly on broth from positive 12B bottles. With regard to direct identification of MAC in signal-positive ESP II bottles, the latex test performed well and provided results in a significantly shorter period of time than was possible by our usual protocol. The latex test failed to identify only one positive culture and yielded two false-positive results. The sensitivity of the latex test was higher than has been reported previously for AccuProbe when used for direct identification of MAC in liquid cultures (Evans et al. 1992; Chapin-Robertson et al. 1993; Reisner et al. 1994; LaBombardi et al. 1997). Moreover, with the latex test, the additional wash steps required when using AccuProbe for direct detection of MAC in pellets from positive blood cultures are not necessary (LaBombardi et al. 1997). Although we limited our study only to signal-positive ESP II blood cultures, based on the importance of detecting MAC bacteremia, we have no reason to believe that the performance of the latex test would be different with other specimen types. However, our results suggest that when testing aliquots from a positive broth culture for MAC by latex agglutination, it probably would be wise to report positive results as preliminary, followed by a final report based on examination of growth of the subculture on a solid medium. In summary, we believe that the latex test is a reliable alternative to the AccuProbe test for identification of MAC, and is preferable when performing direct testing of a positive liquid blood culture. The costs associated with the two tests are similar. The list price of the AccuProbe is $330 for a 20-test kit, and included among the 20 tests are the controls. Additional materials needed for the AccuProbe are the detection reagent ($25 for 100 tests) and the culture identification reagent ($75 for 120 tests). Equipment requirements for the AccuProbe include a waterbath and a luminometer to read the results. The list price of the latex test is $420 for a 25-test kit, which also contains additional reagents for 15 control procedures (i.e., controls are not included among the 25 tests). Equipment needed for the latex test are a heat block and a three-dimensional rocker.

This study was supported in part by an educational grant from Remel. We thank Ms. Shirley Wright for her secretarial assistance.

74

J.P. Olano et al.

REFERENCES Benson CA (1994) Disease due to the Mycobacterium avium complex in patients with AIDS: epidemiology and clinical syndrome. Clin Infect Dis 18(suppl):S218–S222. Chapin-Robertson K, Dahlberg S, Waycott S, Corrales C, Kontnick C, Edberg SC (1993) Detection and identification of mycobacterium directly from BACTEC bottles by using DNA-rRNA probe. Diag Microbiol Infect Dis 17:203–207. Evans KD, Nakasone AS, Sutherland PA, de la Maza LM, Peterson EM (1992) Identification of Mycobacterium tuberculosis and Mycobacterium avium-intracellulare directly from primary BACTEC cultures by using acridinium-ester-labeled DNA probes. J Clin Microbiol 30:2427–2431. Goto M, Oka S, Okuzumi K, Kimura S, Shimada K (1991) Evaluation of acridinium-ester-labeled DNA probes for identification of Mycobacterium tuberculosis and Mycobacterium avium-Mycobacterium intracellulare complex in culture. J Clin Microbiol 29:2473–2476. Greene JA, Gleason-Beavis K (1997) Rapid identification of mycobacteria from broth culture using the Myco AKT. Abstract C-443. Abstracts of the 97th General Meeting of the

American Society for Microbiology. Washington, DC: American Society of Microbiology. Labombardi VJ, Carter L, Massarella S (1997) Use of nucleic acid probes to identify mycobacteria directly from Difco ESP-Myco bottles. J Clin Microbiol 35(4):1002–1004. Reisner BS, Gatson AM, Woods GL (1994) Use of GenProbe AccuProbes to identify Mycobacterium avium complex, Mycobacterium tuberculosis complex, Mycobacterium kansasii and Mycobacterium gordonae directly from BACTEC TB broth cultures. J Clin Microbiol 32(12):2995– 2998. Shafran SD, Singer J, Zarowny DP, Phillips P, Salit I, Walmsley SL, Fong IW, Gill J, Rachlis AR, Lalonde RG, Fanning MM, Tsoukas CM (1996) A comparison of two regimens for the treatment of Mycobacterium avium complex bacteremia in AIDS: rifabutin, ethambutol, and clarithromycin versus rifampin, ethambutol, clofazimine, and ciprofloxacin. N Engl J Med 335:377–383. Woods GL (1994) Disease due to the Mycobacterium avium complex in patients infected with human immunodeficiency virus: diagnosis and susceptibility testing. Clin Infect Dis 18(suppl):S227–S232.