Cir379 321.325

Evaluation of a Newly Developed Lateral FlowImmunoassay for the Diagnosis of Cryptococcosis Mark D. Lindsley,1 Nanthawan Mekha,2 Henry C. Baggett,3 Yupha Surinthong,2 Rinrapas Autthateinchai,2Pongpun Sawatwong,3 Julie R. Harris,1 Benjamin J. Park,1 Tom Chiller,1 S. Arunmozhi Balajee,1 andNatteewan Poonwan2 1Mycotic Diseases Branch, Centers for Disease Control and Prevention, Atlanta, Georgia; 2National Institute of Health, Department of Medical Sciences,Ministry of Public Health, Nonthaburi, Thailand; and 3International Emerging Infections Program, Thailand Ministry of Public Health–Centers forDisease Control and Prevention Collaboration, Nonthaburi, Thailand Cryptococcosis is a common opportunistic infection of human immunodeficiency virus (HIV)– infected individuals mostly occurring in resource-limited countries. This study compares the performance ofa recently developed lateral flow immunoassay (LFA) to blood culture and enzyme immunoassay (EIA) for thediagnosis of cryptococcosis.
Archived sera from 704 HIV-infected patients hospitalized for acute respiratory illness in Thailand were tested for cryptococcal antigenemia using EIA. All EIA-positive and a subset of EIA-negative sera were tested byLFA, with results recorded after 5 and 15 minutes incubation. Urine from patients with LFA- and EIA-positive sera was tested by LFA. Antigen results from patients with positive cryptococcal blood cultures were compared.
Of 704 sera, 92 (13%) were positive by EIA; among the 91 EIA-positive sera tested by LFA, 82 (90%) and 87 (96%) were LFA positive when read after 5 and 15 minutes, respectively. Kappa agreement of EIA and LFA for sera was 0.923 after 5 minutes and 0.959 after 15 minutes, respectively. Two of 373 EIA-negative sera were LFApositive at both time points. Of 74 urine specimens from EIA-positive patients, 52 (70.3%) were LFA positive. EIAwas positive in 16 of 17 sera from blood culture–positive patients (94% sensitivity), and all sera were positive by LFA(100% sensitivity).
A high level of agreement was shown between LFA and EIA testing of serum. The LFA is a rapid, easy-to-perform assay that does not require refrigeration, demonstrating its potential usefulness as a point-of-careassay for diagnosis of cryptococcosis in resource-limited countries.
Cryptococcosis, caused by Cryptococcus spp., is one of the 90% of deaths from cryptococcal meningitis occur in most common opportunistic infections among human sub-Saharan Africa and Southeast Asia [2]. Primary re- immunodeficiency virus (HIV)–infected individuals [1].
spiratory illness due to Cryptococcus, while uncommon Globally, an estimated 1 million new cases of crypto- in the United States and Europe, is more common in coccal meningitis occur per year, with more than 600 000 other regions, such as Southeast Asia [2].
deaths. An estimated 88% of global cases and more than Currently, cryptococcal diagnostics include micros- copy and/or culture-based methods, or detection ofcryptococcal antigen (CrAg) in body fluids using eitherlatex agglutination (LA) or enzyme immunoassay (EIA) Received 4 March 2011; accepted 29 April 2011.
Correspondence: S. Arunmozhi Balajee, MD, PhD, Mycotic Diseases Branch, methods. Although culturing of the organism is con- Centers for Disease Control and Prevention, Mailstop G11, 1600 Clifton Rd, sidered the gold standard diagnostic method, it has poor sensitivity, requires a large quantity of specimen, and Published by Oxford University Press on behalf of the Infectious Diseases Society requires laboratory infrastructure including electricity of America 2011. This is an Open Access article distributed under the terms of the (for centrifugation). Antigen tests such as LA or EIA Creative Commons Attribution Non-Commercial License (http://creativecommons.
org/licenses/by-nc/2.5/), which permits unrestricted non-commercial use, distribution, performed on cerebral spinal fluid or serum are highly and reproduction in any medium, provided the original work is properly cited.
sensitive and specific diagnostic options that are less 1058-4838/2011/534-0001$14.00DOI: 10.1093/cid/cir379 labor- and time-intensive than culture [3, 4]. However, LFA for Cryptococcal Diagnosis d CID 2011:53 (15 August) d 321 these methods require refrigeration, a cold chain for specimen Control and Prevention and the Ethical Review Committee of transport, and technical expertise; therefore, they are often performed only in reference/diagnostic labs far removed from All testing was performed at the National Institute of Health, patients, potentially limiting their clinical utility. In addition, the Nonthaburi, Thailand. The results of the LFA were compared costs of these tests are not affordable for many clinics. As a re- with those obtained from 2 commonly used diagnostic tests for sult, cryptococcosis often goes undiagnosed in resource-limited detecting Cryptococcus infection, blood culture and EIA. Blood cultures were obtained from a subset of study participants as Point-of-care tests (POCTs) show promise for enabling di- a part of routine clinical care. Cultures were processed by au- agnosis of infectious diseases in remote care centers in low- tomated BacTAlert blood culture system (bioMe´rieux), and resource countries. POCTs are assays that can produce results pathogen identification of isolates from positive blood cultures quickly, are simple to perform and interpret by personnel with was performed using standard microbiology methods (mor- no or minimal laboratory training, and ideally can be used phological and physiological tests).
without cold chain or advanced laboratory equipment [5]. To- Stored sera from all HIV-infected patients enrolled in the day, reliable and affordable POCTs are available for the de- study from 2004 to 2009 were tested by EIA, using the Premier tection of hepatitis B infection [6], HIV infection [7], malaria Cryptococcal Antigen enzyme immunoassay (Meridian Bio- [8, 9], syphilis [10], cholera [11], and some neglected tropical sciences). LFA testing was performed on serum specimens from diseases [12–14]. Additional advantages of POCTs include re- all patients with a positive serum EIA test and a random subset duction of patient anxiety and waiting time and decreases in of patients with a negative serum EIA test. Additionally, urine patient loss to follow-up [15]. To serve the communities most in from patients whose sera were positive by LFA and EIA were need, the World Health Organization (WHO) recommended tested by LFA. No LFA testing was performed on urine of pa- that POCTs should be Affordable, Sensitive, Specific, User- tients who had serum that tested EIA-negative.
friendly, Rapid, Equipment-free, and Delivered to those who The EIA was performed according to the manufacturer’s protocol using 50 lL of specimen. For the EIA, wells with re- In 2009, a lateral flow immunoassay (LFA) for the detection actions of optical densities ,0.07 units were considered negative; of cryptococcal antigen was developed by IMMY (Immuno- $0.07 to ,0.100 was considered indeterminate; and $0.100 was Mycologics) as a potential POCT for diagnosis of cryptococcal considered positive. The LFA is a semiquantitative test system for infection. The LFA is stable at room temperature, has a rapid the detection of capsular polysaccharide antigens of Cryptococcus turnaround time, requires very little technical skill, and can be species complex (Cryptococcus neoformans and Cryptococcus performed with minimal laboratory infrastructure. In this study, gattii). The LFA kit consists of immunochromatic test strips the performance of the LFA for CrAg detection was evaluated by impregnated with monoclonal antibodies optimized to detect all testing archived specimens from HIV-infected patients hospi- 4 cryptococcal serotypes and a diluent. The LFA kit can be stored talized with acute respiratory illness in Thailand, and comparing at room temperature for up to a year. To perform the assay, 20 lL of patient specimen was mixed with 2 drops of diluentin a 2-mL screw cap microtube (Sarstedt AG). The LFA strip was placed in the specimen and diluent cocktail and in-cubated at room temperature. Results were read after two The serum and urine specimens used in this study were collected incubation times: 5 minutes according to the manufacturer’s as part of a pneumonia etiology study that included HIV- instructions and a prolonged incubation time of 15 minutes.
infected patients in Thailand, described elsewhere [18]. In brief, The presence of 2 bands (control band and test band) in the test patients admitted to 1 of 8 hospitals in Sa Kaeo or 12 hospitals in zone of the LFA strip was interpreted as a positive result and Nakhon Phanom provinces in Thailand between 2004 and 2009 a single band in the test zone (control band) was interpreted were enrolled if they displayed any sign or symptom of active infection (temperature .38.2°C or ,35.5°C within 24 hours of A positive blood culture result for C. neoformans was used as admission; chills; abnormal total white blood cell count or dif- the gold standard for confirmed Cryptococcus infection. Sensi- ferential), and had evidence of lower respiratory illness (ab- tivity was calculated among the group of patients with a positive normal breath sounds, tachypnea, cough, sputum production, blood culture result as: (number of patients with positive serum or dyspnea). All patients provided informed consent for sample LFA) / (number of patients with positive blood culture) * 100.
collection and use of samples for research purposes. Serum and Specificity was not calculated in this group, as blood culture was urine specimens were collected soon after hospital admission not considered to be a method with sufficiently low false neg- and immediately after study consent. This study was approved ative rates. Agreement between serum or urine LFA results and by the Institutional Review Board of the Centers for Disease serum EIA results was calculated using the j statistic.
322 d CID 2011:53 (15 August) d Lindsley et al LFA using a 5-minute incubation, 52 (70.3%) were positive,and 22 (29.7%) were negative. Among urine specimens from A total of 704 HIV-infected study participants with available 63 patients with LFA-positive sera, 51 (81.0%) were LFA positive, specimens were identified and were tested as outlined in Figure 1.
Blood culture was performed on 325 patient specimens, with 18 All 91 EIA-positive sera were also tested by LFA using an of 325 yielding C. neoformans. Serum was not available from 1 of extended incubation time. When the results of the serum LFA the 18 culture-positive patients for analyses in the study; thus the were read after 15 minutes, 5 additional sera became positive EIA testing was performed on 17 culture-positive sera. Of the 17 (n 5 87; 95.6%). Of these 5 sera, only 1 came from a patient with serum specimens tested, 16 were positive by EIA and 17 were blood culture performed, which was negative for Cryptococcus positive by LFA with a 5-minute incubation time yielding a sen- spp. All EIA-negative specimens remained LFA negative after sitivity of 94% and 100%, respectively. Thirteen of the 18 patients the extended incubation period, increasing the j statistic to whose blood culture was positive for Cryptococcus also had a corresponding urine specimen, 12 of which were positive byLFA (92% sensitivity).
Sera from all 704 patients were tested using EIA; 92 (13.1%) were positive. No sera tested in this study gave in- Resource-limited countries, especially those in sub-Saharan determinate results by EIA. Of 91 EIA-positive sera that were Africa and Southeast Asia, continue to experience a high in- available for further analysis, 82 (90.1%) were positive by cidence of cryptococcosis; managing this disease is a persistent LFA with a 5-minute incubation time. Discordant results public health challenge [2]. Although several CrAg tests are between EIA and LFA were most often observed in serum currently available for cryptoccoccal diagnosis, these tests are exhibiting lower EIA values (data not shown). A random not readily accessible in resource-limited settings, resulting in no subset of 373 serum samples was selected from the 612 EIA- diagnosis or underdiagnosis of these often fatal infections. In the negative sera and tested using LFA: 371 of 373 (99.5%) were present study, a recently developed cryptococcal assay, the LFA, LFA negative, and 2 (0.5%) were LFA positive. The resultant was evaluated for test sensitivity and agreement with other j statistic describing agreement between LFA and EIA was available diagnostic methods using archived serum and urine specimens from HIV-infected patients in Thailand. Results Urine specimens were available from 74 patients whose sera showed that, with serum specimens, the LFA was 100% sen- were EIA positive. When these urine specimens were tested by sitive when compared with the gold standard blood culture.
Flow chart of the specimen testing algorithm. A total of 704 sera were tested by cryptococcal antigen enzyme immunoassay (CrAg-EIA); 91 EIA-positive sera were tested by lateral flow immunoassay (LFA); a subset of CrAg-EIA–negative sera (n 5 373) was further tested by LFA. Specimensfrom Cryptococcus blood culture–positive patients (n 5 17) were tested by LFA and EIA. A total of 74 urine samples from serum CrAg-EIA–positivepatients were tested by LFA. HIV, human immunodeficiency virus.
LFA for Cryptococcal Diagnosis d CID 2011:53 (15 August) d 323 Additionally, the LFA had a high level of agreement with the submitted to the United States Food and Drug Administration cryptococcal EIA. When urine was evaluated, the LFA was for approval (personal communication, Sean Bauman, IMMY).
found to be very sensitive (92%) when compared with blood For POCTs to have maximum value in remote settings, the use culture, and moderately sensitive (70.7%) when compared with of minimally invasive, easily obtained, processing-free specimens is optimal. This study was performed with sera and urine as test One of the limitations of this study was that the specimens specimens in a controlled laboratory setting (reference laboratory); were collected from patients hospitalized with acute respiratory however, in remote areas with insufficient technical expertise illness for whom complete clinical details were not available, or specimen processing capabilities, sera may not be the ideal including whether they had meningitis. Patients with more in- specimen type. The lower sensitivity of the LFA for urine vasive infections (ie, meningitis) may have a higher burden of compared with serum could be due to reduced excretion of circulating organisms and therefore antigen, and this may im- CrAg into the urine, compared with the blood. HIV infection pact the performance of tests that measure CrAg. Accordingly, [20] and treatment of HIV with the antiretroviral drugs te- discrepant results between serum EIA and LFA were more nofovir and indinavir [21, 22] have been previously demon- often observed in serum with lower EIA optical density values, strated to reduce the glomerular filtration rate and, possibly reflecting the presence of low levels of circulating potentially, reduce the amount of CrAg excreted in the urine.
Urine has been used for antigen detection in other fungal This study was performed in a reference laboratory and [23–25] and nonfungal diseases [26, 27]. The kinetics of ex- therefore the performance of the LFA in a field or hospital set- cretion of CrAg antigen in urine is unclear, and additional ting is unclear at this time. However, the LFA was simple to studies testing CrAg in urine need to be performed, including perform and did not require any additional laboratory equip- studies where urine is collected under controlled conditions.
ment. Incubation could be performed at room temperature and Another minimally invasive specimen type is whole blood the assay itself could be accomplished in 3 easy steps since this from a finger stick. Point-of-care assays for whole blood have method does not require any pretreatment of specimen (to re- been developed for viral and parasitic diseases [8, 9, 12, 13], all of move rheumatoid factor). Finally, the LFA yielded results that which have been useful in resource-limited countries. Further were unambiguous. Thus, the LFA has many characteristics that studies evaluating the LFA using finger-stick blood would en- may make it a valuable POCT in resource-limited areas. In ad- hance the accessibility of this assay. Additionally the test’s cost, dition, this study demonstrates that the CrAg LFA satisfies most ranging from $1.25 to $2.50 per test (depending on the country of the WHO ASSURED criteria [16, 17]: specifically, the assay is and volume of purchase) is based on the World Bank’s list of sensitive, user-friendly (small specimen volume, simple to use), economies, thereby ensuring affordability to the countries most rapid (10–15 minutes to perform), and equipment-free (including no requirement for refrigeration). The rapid turnaround time In summary, this study demonstrates that the LFA is a sensi- will allow diagnoses to be potentially provided during patient tive test for Cryptococcus spp compared with the gold standard visits, allowing treatment to begin immediately if warranted.
culture, and has a high level of agreement with EIA. Given the Recently, Jarvis et al [19] strongly recommended the integration ease of use, temperature stability, minimal requirements for of CrAg screening into national antiretroviral treatment pro- laboratory infrastructure, and potential low cost of this test, the grams in sub-Saharan Africa to reduce the human and economic LFA shows great promise as a POCT for diagnosis of crypto- costs due to the disease [19]. Although not tested in this study, coccosis. The availability of this assay as a POCT for use in the LFA may also have utility as a screening tool for early remote locations could have a meaningful impact on crypto- Extending the incubation time for sera from 5 to 15 minutes improved the sensitivity of the LFA when compared with EIA, thereby increasing agreement between the EIA and the LFA.
Since the time of this study, the manufacturer of the LFA has The authors express their gratitude to Apiwat Lapamnouysup for the technical assistance in this study, and acknowledge Immuno-Mycologics, modified the testing protocol, now recommending testing twice Inc. (Norman, Oklahoma), for the donation of the lateral flow immuno- the volume of patient specimen, reducing the recommended specimen:diluent ratio from 1:5 to 1:2, increasing the amount The findings and conclusions in this article are those of the author(s) and do not necessarily represent the views of the CDC. The use of product of conjugate in the chromatographic strip, and increasing the names in this manuscript does not imply their endorsement by the US recommended incubation time of the LFA to 10 minutes. Cur- Department of Health and Human Services.
rently the LFA has CE marking (a mandatory conformance Potential conflicts of interest. All authors: No reported conflicts.
All authors have submitted the ICMJE Form for Disclosure of Potential marking for the European Economic Area) for use in the Euro- Conflicts of Interest. Conflicts that the editors consider relevant to the content pean Union and in countries that use CE approval and has been of the manuscript have been disclosed in the Acknowledgments section.
324 d CID 2011:53 (15 August) d Lindsley et al 14. Peeling RW, Mabey D. Point-of-care tests for diagnosing infections in the developing world. Clin Microbiol Infect 2010; 16:1062–9.
1. Ong EL. Common AIDS-associated opportunistic infections. Clin Med 15. Loubiere S, Moatti JP. Economic evaluation of point-of-care diagnostic technologies for infectious diseases. Clin Microbiol Infect 2010; 16: 2. Park BJ, Wannemuehler KA, Marston BJ, Govender N, Pappas PG, Chiller TM. Estimation of the current global burden of cryptococcal 16. Mabey D, Peeling RW, Ustianowski A, Perkins MD. Diagnostics for meningitis among persons living with HIV/AIDS. AIDS 2009; 23:525–30.
the developing world. Nat Rev Microbiol 2004; 2:231–40.
3. Gade W, Hinnefeld SW, Babcock LS, et al. Comparison of the PRE- 17. WHO. Mapping the landscape of diagnostics for sexually transmitted MIER cryptococcal antigen enzyme immunoassay and the latex ag- infections. Key findings and recommendations.: UNICEF/UNDP/ glutination assay for detection of cryptococcal antigens. J Clin 18. Olsen SJ, Thamthitiwat S, Chantra S, et al. Incidence of respiratory 4. Sekhon AS, Garg AK, Kaufman L, et al. Evaluation of a commercial pathogens in persons hospitalized with pneumonia in two provinces enzyme immunoassay for the detection of cryptococcal antigen. My- in Thailand. Epidemiol Infect 2010; 138:1811–22.
19. Jarvis JN, Wainwright H, Harrison TS, Rebe K, Meintjes G. Pulmonary 5. Anderson D, Crowe S, Garcia M. Point-of-care testing. Curr HIV/AIDS cryptococcosis misdiagnosed as smear-negative pulmonary tubercu- losis with fatal consequences. Int J Infect Dis 2010; 14(Suppl 3): 6. Lin YH, Wang Y, Loua A, et al. Evaluation of a new hepatitis B virus surface antigen rapid test with improved sensitivity. J Clin Microbiol 20. Krawczyk CS, Holmberg SD, Moorman AC, Gardner LI, McGwin G Jr., Group HIVOS. Factors associated with chronic renal failure in HIV- 7. Pavie J, Rachline A, Loze B, et al. Sensitivity of five rapid HIV tests on infected ambulatory patients. AIDS 2004; 18:2171–8.
oral fluid or finger-stick whole blood: a real-time comparison in 21. Campbell LJ, Ibrahim F, Fisher M, Holt SG, Hendry BM, Post FA.
a healthcare setting. PLoS One 2010; 5:e11581.
Spectrum of chronic kidney disease in HIV-infected patients. HIV Med 8. Ghanchi NK, Beg MA, Hussain R. Estimation of parasite load using rapid diagnostic test ICT Now Malaria P.f/P.v in Plasmodium falcipa- 22. Horberg M, Tang B, Towner W, et al. Impact of tenofovir on renal rum malaria. Scand J Infect Dis 2009; 41:597–601.
function in HIV-infected, antiretroviral-naive patients. J Acquir Im- 9. Quintana M, Piper R, Boling HL, et al. Malaria diagnosis by dipstick assay in a Honduran population with coendemic Plasmodium falci- 23. Durkin M, Witt J, Lemonte A, Wheat B, Connolly P. Antigen assay with parum and Plasmodium vivax. Am J Trop Med Hyg 1998; 59:868–71.
the potential to aid in diagnosis of blastomycosis. J Clin Microbiol 10. Peeling RW, Ye H. Diagnostic tools for preventing and managing maternal and congenital syphilis: an overview. Bull World Health 24. Swartzentruber S, Rhodes L, Kurkjian K, et al. Diagnosis of acute pulmonary histoplasmosis by antigen detection. Clin Infect Dis 2009; 11. Mukherjee P, Ghosh S, Ramamurthy T, et al. Evaluation of a rapid immunochromatographic dipstick kit for diagnosis of cholera em- 25. Wheat LJ. Current diagnosis of histoplasmosis. Trends Microbiol phasizes its outbreak utility. Jpn J Infect Dis 2010; 63:234–8.
12. Rocha A, Braga C, Belem M, et al. Comparison of tests for the detection 26. Dirven K, Ieven M, Peeters MF, van der Zee A, De Schrijver K, of circulating filarial antigen (Og4C3-ELISA and AD12-ICT) and ul- Goossens H. Comparison of three Legionella urinary antigen assays trasound in diagnosis of lymphatic filariasis in individuals with mi- during an outbreak of legionellosis in Belgium. J Med Microbiol crofilariae. Mem Inst Oswaldo Cruz 2009; 104:621–5.
13. Weil GJ, Lammie PJ, Weiss N. The ICT Filariasis Test: a rapid-format 27. Neuman MI, Harper MB. Evaluation of a rapid urine antigen assay for antigen test for diagnosis of bancroftian filariasis. Parasitol Today 1997; the detection of invasive pneumococcal disease in children. Pediatrics LFA for Cryptococcal Diagnosis d CID 2011:53 (15 August) d 325

Source: http://check-poct.de/app/download/5780432500/Evaluation+of+a+Newly+Developed+Lateral+Flow.pdf