Evaluation of point of care serum cryptococcal antigen by lateral flow immunoassay for diagnosis of cryptococcosis and cryptococcal meningitis in HIV-positive patients.

BACKGROUND: Cryptococcal meningitis (CM) is the initial acquired immunodeficiency syndrome (AIDS) defining illness in 2% of patients with CD4 levels <100/μL and a leading cause of mortality in AIDS in the developing world. It is the most common opportunistic infection of the central nervous system in AIDS in various Indian studies. Detection of serum cryptococcal antigen (SCRAG) is the most widely used diagnostic method for cryptococcosis. The presence of cerebrospinal fluid cryptococcal antigen (CSF CRAG) is diagnostic of CM. CRAG can be determined by latex agglutination (LAT), enzyme-linked immunosorbent assay and now, by lateral flow (LFA)immunoassay. LFA is a point of care test that rapidly detects CRAG. AIMS AND
OBJECTIVES: This study compares LAT and LFA for the detection of serum CRAG and diagnosing CM.
MATERIALS AND METHODS: Two hundred and ten patients of HIV/AIDS were submitted to SCRAG LFA by dipstick. A sample was also sent to laboratory for SCRAG by LAT. CSF examination was done for those who were positive for SCRAG LFA and those who had symptoms suggestive of meningitis. SCRAG by LFA was compared with SCRAG by LAT, CSF CRAG by LAT and LFA, CSF cryptococcal culture and CSF India ink examination for Cryptococcus.
RESULTS: Fifteen patients were found positive for SCRAG by LFA dipstick. All of them were also positive for SCRAG by LAT. Twelve of them had C. D4 count below below 100 cells/mm3. CSF CRAG was positive in all 12 SCRAG positive who were submitted to CSF examination.
CONCLUSION: We found that serum detection of CRAG by LFA dipstick is as sensitive as CRAG detection in serum by LAT and CSF CRAG detection by LFA and LAT. It is thus a rapid test for diagnosing CM in HIV patients with low CD4 counts. Copyright:

INTRODUCTION

Cryptococcal meningitis (CM) is the initial acquired immunodeficiency syndrome (AIDS) defining illness in 2% of patients with CD4 levels <100/µL and a leading cause of mortality in AIDS in the developing world.[1] It is the most common opportunistic infection of the central nervous system in AIDS in various Indian studies.[234]

Early diagnosis and treatment decreases CM-related mortality. The detection of serum cryptococcal antigen (SCRAG) is the most widely used diagnostic method for cryptococcosis. The presence of cerebrospinal fluid cryptococcal antigen (CSF CRAG) is diagnostic of CM. CRAG can be determined by latex agglutination (LAT),[5] enzyme-linked immunosorbent assay[6] and now, by lateral flow (LFA) immunoassay.[78] LFA is a point of care test that rapidly detects CRAG. This study compares LAT and LFA for detection of serum CRAG and diagnosing CM.

MATERIALS AND METHODS

A cross-sectional observational study was done with 210 HIV/AIDS patients with or without signs of meningitis presenting to the department of medicine and/or anti-retroviral therapy (ART) clinic at a teaching hospital in New Delhi. The study included patients above the age of 18 years who had documented HIV infection, irrespective of CD4 counts and ART status. HIV infection in them was confirmed by series of 3 tests as per the National AIDS Control Organization guidelines. All patients with history of cryptococcosis and patients already on fluconazole therapy were excluded from the study.

Patients were then subjected to detailed history, examination, and laboratory investigations. At the time of examination serum CRAG-LFA with dipstick was done at bedside or in the clinic. Simultaneously, one or more samples were sent to laboratory for detection of CRAG by LAT.

In all CRAG LFA-positive patients, CSF testing was done for CRAG detection by LFA at bedside and same sample was sent to the laboratory for detection of CRAG by LAT. CSF culture and India ink examination was also done. CSF study and CRAG were also done in all patients with symptoms suggestive of meningitis.

We compared CRAG by LFA with CRAG by LAT and CSF CRAG by LFA with CSF CRAG by LAT. Titers of CRAG by both methods were also compared. We also compared SCRAG with CSF CRAG, CSF India ink and CSF cryptococcal culture for diagnosing CM. SCRAG-LAT was taken as the gold standard for serum cryptococcal antigenemia and CSF-CRAG LAT as the gold standard for diagnosing CM. The results obtained by LFA and LAT were also compared with clinical signs, symptoms, CD4 count, and ART status.

Lateral flow assay technique

The CRAG LFA provides both qualitative and semi quantitative results. The test is done as follows: First a drop of LFA specimen diluent is taken in a tube, then 40 µL of patient specimen is added to it. Next, the CRAG LFA test strip is inserted into the tube and incubated for 10 min to interpret the results. The presence of two lines (test and control lines), regardless of the intensity of the test line indicates a positive result. A single control line indicates a negative test result. If the control line does not appear, the results are invalid and the test should be repeated. For semi-quantitative results, the patient's titers are reported as the highest dilution that yields a positive test result.

Latex agglutination test

The LAT test was done using the commercial preparation PASTOREX TM CRYPTOPLUS (BIO-RAD, FRANCE), according to the instructions of producer. 120 µl of CSF (1:2 diluted in glycine buffer) is mixed with 20 µl of pronase and incubated at 56°C for 30 min. An aliquot of 25 µl is transferred on the slide card on the designated rings and mixed with applicator sticks. The test is carried out with a positive control. The card is then rocked by hand for 5 min and agglutination read immediately. Results are then read out visually based on degree of agglutination and rated on a scale ranging from 0 to 4+. A reading of 2+ was considered positive.

Positive samples are then titrated by serial dilutions. Antigen titer value are defined as dilution prior to which antigen is not detected.

RESULTS

Of 210 HIV patients who underwent screening for cryptococcal infection, CRAG was positive in 15 patients by LFA. Of these 4 patients were asymptomatic and 11 had symptoms of meningitis [Table 1]. Nine patients had CD4 counts <50 cells/mm3, 3 had CD4 counts between 50 cells/mm3 and 100 cells/mm3 and 3 had counts between 100 cells/mm3 to 200 cells/mm3. None of the patients with CD4 above 200 cells/mm3 was CRAG positive [Table 2]. All those with CRAG positive by LFA had CRAG positive by LAT as well.

Table 1

Serum cryptococcal antigen relation with symptom status*

Status	Total (number of patients)	SCRAG +ve	SCRAG −ve
Symptomatic	51	11	40
Asymptomatic	159	4	155

*The P value is <0.00001. Significant at P<0.01. SCRAG=Serum cryptococcal antigen

Table 2

Serum cryptococcal antigen relation with CD4 count*

CD4 count	Number of patients
	Total	SCRAG (LF) +ve
<50	39	9
50–100	95	3
>100	76	3

*The P value is 0.000763. The result is significant at P<0.01. SCRAG=Serum cryptococcal antigen; LF=Lateral flow

Out of 15 CRAG-positive patients, 3 asymptomatic patients did not give consent for lumbar puncture. Remaining 12 (11 symptomatic and one asymptomatic) were subjected to lumbar puncture and CSF study was done. All of them had CRAG positivity in CSF both by LAT and LFA. Four patients out of 12 patients with serum and CSF CRAG positivity had no growth on fungal culture. Five out of the 12 patients, positive for serum and CSF CRAG had CSF negative for India ink staining [Table 3].

Table 3

Cerebro spinal fluid results in 12 patients of serum cryptococcal antigen positive by lateral flow who were subjected to cerebro spinal fluid examination*

Result	CSF crag (LF)	CSF crag (LAT)	CSF India ink	CSF culture
Positive	12	12	8	9
Negative	0	0	4	3
Total	12	12	12	12

*3 asymptomatic patients who were SCRAG positive did not consent for CSF examination. SCRAG=Serum cryptococcal antigen; CSF=Cerebro spinal fluid; LF=Lateral flow; LAT=Latex agglutination test

None of the symptomatic patients who were SCRAG negative by Lateral flow assay (LFA) had CSF CRAG positive by LAT or LFA. CSF cryptococcal culture and India ink was also negative in all these patients [Table 4].

Table 4

Cerebrospinal fluid results in 43 symptomatic patients who were serum cryptococcal antigen negative by lateral flow

Result	CSF crag (LFA)	CSF crag (LAT)	CSF India ink	CSF culture
Positive	0	0	0	0
Negative	43	43	43	43
Total	43	43	43	43

CSF=Cerebrospinal fluid; LFA=Lateral flow assay; LAT=Latex agglutination test

CRAG titers by LAT and LF were compared in 7 serum and 4 CSF samples [Table 5]. It was seen that consistently CRAG was positive in higher dilutions by LF.

Table 5

Correlation of crag titers by latex agglutination test and lateral flow assay in serum and cerebrospinal fluid*

Serum crag titres (LAT)	Serum crag titres (LFA)
1:256	1:320
1:256	1:320
1:128	1:160
1:1024	1:1280
1:256	1:320
1:512	1:640
1:64	1:80
CSF crag titers (LAT)	CSF crag titers (LFA)
1:1024	1:1280
1:1024	1:1280
1:64	1:80
1:512	1:640

*R=1, the P value is <0.00001. The result is significant at P<0.01. CRAG=Cryptococcal antigen; CSF=Cerebrospinal fluid; LFA=Lateral flow assay; LAT=latex agglutination test

DISCUSSION

As CM has a high prevalence (>3.2%) and the fact that cryptococcal subclinical infection is high, a rapid diagnostic test needs to be introduced and evaluated. SCRAG by LFA is a point of care test to detect cryptococcemia. In our study, LFA was compared against LAT, taking CRAG by LAT as gold standard. This was done as previous studies have shown superiority of LAT over other methods of cryptococcal detection.[910]

The prevalence of cryptococcal antigenemia in our study was 7.1% among HIV-infected people, irrespective of CD4 counts and ART status and this was comparable to study by Park et al. (0.04%–12%)[1] and Casadevall from New York (6.1–8.5%).[11]

The mean age group of our study population was 38.8 years (19–78 years). Majority of them were males. Sexual transmission was the route of spread in almost all the cases. Most of the patients were diagnosed with HIV infection more than a month before being enrolled in our study. Hence, majority of them were on ART. These demographic characteristics of our study population were similar to studies conducted earlier by Chakraborty et al.,[12] Amelu et al.[13] and Ogba and Abia Bassey.[14]

About 34.6% of our patients had one or the other symptoms suggesting meningitis. Among them, 15 patients were found to be CRAG positive. Fever was observed in 13 patients (86.67%), headache in 11 (73.33%), vomiting in 6 (40%), and neck rigidity in 10 patients (66.67%). Most of the CRAG-positive patients had symptomatic meningitis (80%; 12 out of 15). This was in agreement with study by Baradkar et al.[15] and Clark et al.[16] Our study also found that there is an increased chance of CRAG to be positive in patients with lower Glasgow Coma Scale (GCS). Zuger et al.[17] had shown that patients with higher GCS at diagnosis had better prognosis in CM.

CRAG prevalence in our study was 7.1% (15/210). This was comparable to study by Park et al.[1] but lower to those done outside India mainly in sub-Saharan countries where the cryptococcal fungal burden is higher than the rest of the world.

The CD4 count of our study population ranged from 6 to 780 cells/mm3. No patient receiving ART was detected to be CRAG positive. Of the patients positive for CRAG, 9 out of 15 (60%) had CD4 <50 cells/mm3; three out of 15 (20%) had CD4 between 50 and 100 cells/mm3 and the remaining three (20%) had CD4 >100 cells/mm3. Previous studies by Micol et al.,[18] Pongsai et al.[19] also found that there is decreasing cryptococcal prevalence with increasing CD4 counts and this was in agreement with the present study.

In the present study, all patients (15/210) who were CRAG positive by LAT were also CRAG positive by LFA. Those with CRAG negative by LA were negative by LFA also. Thus, the sensitivity and specificity of CRAG LFA is 100% each and this was in agreement with several previously published studies by Huang et al.,[20] Klausner et al.,[21] McMullan et al.,[22] Rugemalila et al.[23] and Escandón et al.[24] A study results showed the CRAG LFA had excellent agreement with the CRAG by LAT.

All 12 CRAG-positive patients who were subjected to lumbar puncture, had CSF CRAG positive by both, LA and LFA. This observation was in agreement with multicentric study by Rolfes M,[25] Vijayan et al.[21] and Kabanda et al.[26]

In a study by Lourens et al.,[27] high concentrations of the CRAG resulted in decreased visual intensity of the test lines (hook effect). This accounts for 100% sensitivity of LFA only on diluted CSF samples for the diagnosis of CM, especially when high organism load is present. However, in our study patients with CRAG positive in high dilutions also had CRAG positive by LFA. We detected CRAG positive by LFA and CSF CRAG positive by LFA in patients with titers as low as 1:10 as well as in very high titers as 1:10,000. Although antigenemia is highly predictive of the development of CM and an independent predictor of mortality, our study did not find any statistical significance between the two. This may be due to small sample size (15 samples) and short duration of our study.

Of 54 symptomatic patients whose CSF was analyzed, 12 patients were positive by both CRAG LFA and CSF LFA. Thus, CRAG LFA is 100% sensitive and specific to diagnose CM. This agreed with study by Kozel and Bauman.[28]

CSF of the above patients was also subjected to India ink staining and fungal culture. India ink and fungal cultures were negative in all those in whom CRAG was negative by LA and LFA. Of the CSF CRAG-positive cases, 66.6% were positive for CSF India ink staining and 75% were positive for CSF fungal culture. When LFA and LA were taken as gold standard for diagnosis of CM, sensitivity, specificity, positive predictive value, and negative predictive value of CSF fungal culture were 66.7%, 100%, 100%, 97.9%, and 58.3%, respectively. The corresponding values in case of India ink staining were 97.5%, 100%, 100%, respectively.

This is because India ink is highly operator dependable and the sensitivity of culture depends on quantity of CSF sample used for culture.[25]

No specific importance was given to CSF volume inoculated in our study. Perhaps, larger volume of inoculation might yield better sensitivity for cryptococcal detection by fungal culture. Thus, conventional methods are inferior to CRAG LFA in detection of cryptococcal infection.

All cases of cryptococcal antigenemia in our study had more advanced disease with low CD4 counts. We had excluded previously diagnosed cases of cryptococcosis and our study had no cases of cryptococcal immune reconstitution inflammatory syndrome. None of our patients on ART had cryptococcal antigenemia and this is in concordance with previous studies that ART confers protection against cryptococcosis.

There were some limitations in our study. Only one asymptomatic patient who had CRAG positive was submitted to CSF examination. A larger group of asymptomatic CRAG positive patients need to be evaluated by CSF examination before treating all CRAG-positive patients as CM. Further studies on point of care CRAG by LFA need to focus on asymptomatic patients with low CD4 counts as they stand to gain maximum from this strategy.

CONCLUSION

Early diagnosis is the key to improve survivability in CM. CRAG is a rapid point of care test recommended by the WHO for diagnosing cryptococcal infection. The reports of LFA are obtained in <10 min.

Our study is the first study in India to evaluate CRAG detection by LFA method by comparing it against the LAT method. We found that serum detection of CRAG is as sensitive as CRAG detection in serum by LAT, CSF CRAG detection by LFA and LAT, thus making it possible to implement this method on a large scale even in resource poor, remote areas where access to electricity, well-equipped laboratory, cold chains is not possible. In future, LFA might replace LAT as diagnostic test of choice to diagnose CM.

Financial support and sponsorship

AC was supported by the research grant from ICMR, Delhi, vide grant no OMI/13/2014-ECD-I.

Conflicts of interest

There are no conflicts of interest.