Breakthrough of chemiluminescence-based LAM urine test beyond HIV-positive individuals: Clinical diagnostic value of pulmonary tuberculosis in the general population

Abstract

To investigate the diagnostic value of a novel high-sensitivity urine lipoarabinomannan (LAM) test (chemiluminescence-based) for active tuberculosis in the general population. A retrospective study was conducted on 250 clinical suspected tuberculosis patients who were HIV-negative and visited the Fourth People’s Hospital of Foshan from January 2022 to December 2022. Among them, there were 135 cases of pulmonary tuberculosis, 34 cases of extrapulmonary tuberculosis, and 81 cases of non-tuberculosis. Urine samples were collected for LAM antigen detection before treatment, and laboratory data of sputum smear acid-fast staining (smear method), sputum culture, and GeneXpert method were collected. Using clinical diagnosis as the reference standard, the diagnostic efficacy of 4 methods for detecting active tuberculosis was evaluated. For the 135 cases of pulmonary tuberculosis, the sensitivity of sputum smears, sputm culture, sputm GeneXpert method, and urine LAM were 29.6% (40/135), 45.9% (62/135), 59.3% (80/135), and 51.9% (70/135), respectively. The combination of LAM + GeneXpert and LAM + culture had the highest sensitivity for detecting active pulmonary tuberculosis, which were 71.0% and 78.2%, respectively. For the detection of sputum culture-negative pulmonary tuberculosis, the positive rates of smear, GeneXpert, and LAM were 0.0% (0/73), 53.4% (39/73), and 52.1% (38/73), respectively. LAM + smear and LAM + Genexpert could detect 52.1% and 68.5% of sputum culture-negative patients, respectively. The high-sensitivity urine LAM test holds promise for tuberculosis diagnosis in the general population. It demonstrates high-sensitivity, enabling the detection of sputum culture-negative pulmonary tuberculosis patients. Furthermore, when combined with existing methods, it can enhance the overall detection rate.

1. Introduction

Tuberculosis (TB) is a chronic infectious disease caused by Mycobacterium tuberculosis (MTB) and represents a significant global health crisis.^[1] According to the World Health Organization (WHO)’s 2022 Global TB report, approximately 10.6 million individuals were affected by TB in 2021, reflecting a 4.5% increase from the previous year, with 1.6 million deaths reported (including 187,000 among HIV-positive individuals). China ranks third in the world in terms of TB burden, with a high number of TB patients.^[2] Early detection, diagnosis, and treatment are essential for effective TB management, and laboratory testing plays a critical role in TB diagnosis.^[3] While sputum smear microscopy is a simple and rapid method, its sensitivity is relatively low.^[4] Although culture remains the gold standard for diagnosis, the extended duration (6–8 weeks or more) is not conducive to rapid diagnosis.^[5] Molecular biology methods offer faster results and improved sensitivity and specificity; however, their implementation in primary healthcare settings is challenging due to the need for specialized equipment and facilities.^[6] The limitations of sputum-based TB diagnosis are particularly pronounced for individuals with extrapulmonary tuberculosis (EPTB) or those co-infected with HIV.^[7,8]

Therefore, the development and utilization of sensitive diagnostic tools and easily accessible biomarkers are crucial for early detection and diagnosis of TB.^[9] In recent years, significant advancements have been made in urinary lipoarabinomannan (LAM)-based diagnostics for TB. LAM is a glycolipid that is present in the cell wall of MTB and plays a role in MTB growth, cell wall integrity, and stimulation of host immune responses.^[10] During MTB metabolism, LAM can be released and subsequently detected.^[11] Notably, LAM possesses a small molecular weight (37 kDa), enabling it to pass through the glomerular basement membrane and be excreted in urine.^[12] Since 1997, urinary LAM has been extensively studied as a diagnostic biomarker for TB detection.^[13] LAM testing offers the advantages of convenience, requiring only urine samples, and not relying on sputum, making it a cost-effective and rapid diagnostic approach. As a result, it has been recommended by WHO.^[14] In contrast to traditional diagnostic methods, urine LAM testing offers distinct advantages in the diagnosis of HIV/TB patients. The commercially-available Alere determine TB LAM (AlereLAM) assay is the sole urine-based diagnostic product endorsed by the WHO for TB diagnosis. However, its sensitivity is limited to a moderate range (40–60%) and is primarily applicable to HIV-infected patients with low CD4 + T-cell counts.^[15] Consequently, there is an urgent need to enhance the sensitivity of LAM testing specifically in HIV-negative TB patients.^[16,17]

Research has demonstrated that leveraging the latest generation of anti-LAM monoclonal antibodies and high-throughput laboratory immunoassay platforms can significantly improve the sensitivity and specificity of urine LAM testing. This advancement enhances the detection performance and applicability of LAM testing, even within HIV-negative populations.^[18] Presently, widely used commercial and research LAM testing products encompass AlereLAM and the Fujifilm SILVAMP TB LAM. Although both methods employ colloidal gold as the detection platform, their sensitivity remains suboptimal. Moreover, they are exclusively employed for testing HIV-positive individuals, and research on chemiluminescence platforms offering heightened sensitivity has yet to be reported.

Moreover, the majority of research on urine LAM testing for TB has primarily concentrated on the context of HIV/TB coinfections, resulting in limited studies involving HIV-negative populations. Currently, there are no existing reports on the combined testing of urine LAM and other microbiological methods specifically tailored to HIV-negative individuals.

Therefore, the primary aim of this study is to conduct a comprehensive comparative analysis of diverse laboratory diagnostic techniques employed for the detection of active pulmonary tuberculosis (PTB). Furthermore, we intend to evaluate the clinical effectiveness of domestically produced and commercially available LAM antigen detection products. Additionally, our investigation aims to identify specific combinations of these products that can significantly augment the sensitivity of laboratory testing and enhance the accuracy of TB diagnosis specifically in HIV-negative individuals.

This manuscript is written following STARD 2015 checklist.

2. Methods

2.1. Study design and settings

This prospective observational study aimed to evaluate the diagnostic performance of urinary LAM in diagnosing PTB compared to sputum smear, culture, and GeneXpert. A total of 250 suspected TB patients were enrolled at the Department of Tuberculosis of the Fourth People’s Hospital of Foshan between October 2022 and March 2023 (Fig. 1). Eligibility criteria included suspected TB, age ≥ 18, availability of sputum smears, sputum culture, GeneXpert and urine LAM results, and non-HIV infection. Using the diagnostic criteria outlined in the “Diagnosis for pulmonary tuberculosis (WS 288–2017)”.^[19] and “Classification of Tuberculosis (WS 196–2017)”,^[20] 135 suspected patients were diagnosed with active PTB, 34 were diagnosed with EPTB and excluded, and 81 were diagnosed as non-tuberculosis patients (patients with pulmonary diseases but not MTB infection). A total of 216 patients were included in this study. The study adhered to the principles outlined in the Declaration of Helsinki and received approval from the Ethics Committee of the Fourth People’s Hospital of Foshan (No. 2022008). Written informed consent was obtained from all participants, allowing the utilization of their data for research purposes.

Figure 1.

Flowchart of included patients. EPTB = extrapulmonary tuberculosis, LAM = lipoarabinomannan, Non-TB = non-tuberculosis, PTB = pulmonary tuberculosis, TB = tuberculosis.

2.2. Sample collection

Sputum samples were collected from the patients using spontaneous expectoration. Fresh sputum was obtained from the patients first-morning expectoration after rinsing their mouths with water. The patients were instructed to forcefully expectorate from deep within their lungs into a glass or plastic cup or onto wax-coated paper. In cases where patients had no or minimal sputum, inhalation of a mist of saline solution (100 g/L) heated to 45°C was performed to facilitate sputum expectoration.

Three sputum samples were collected from each patient, including one in the evening, one in the early morning, and one at the time of collection. These 3 samples were combined and thoroughly mixed to form a single specimen. Subsequently, the sputum specimen was mixed with 5 mL of 0.9% saline solution to ensure homogeneity and facilitate laboratory handling. The prepared specimen was then sent to the laboratory for further analysis.

2.3. Sputum smears

The acid-fast staining of sputum smears was conducted in strict adherence to the instructions provided by the acid-fast staining reagents and the “Laboratory Testing Guidelines for Tuberculosis Diagnosis”. Microscopic examination was performed to detect the presence of MTB in the sputum samples. The acid-fast staining reagents used in this study were prepared within the laboratory and comprised of carbol fuchsin.

2.4. Sputum culture

Sputum specimens underwent a decontamination process using N-acetyl-L-cysteine-sodium hydroxide solution. Subsequently, the specimens were washed with phosphate buffer at pH 6.8. A volume of 0.5 mL from each specimen was then inoculated into mycobacteria growth indicator tube (MGIT) culture tubes, which contained nutrient supplements and antimicrobial agents. To facilitate culture and detection, the culture tubes were placed in a BACTEC MGIT 960 Mycobacterial Detection System. The process of mycobacterial isolation and culture examination followed the guidelines outlined in the “Laboratory Testing Guidelines for Tuberculosis Diagnosis”. The MGIT 960 detection system and associated reagents were provided by BD, USA.

2.5. GeneXpert

According to the instructions, the processing solution was mixed with the sputum sample at a ratio of 1:2. The mixture was then placed on a vortex mixer and vortexed for 0.5 minutes. After incubating at a specific temperature for 15 minutes, it was vortexed again for 0.5 minutes and incubated for an additional 5 minutes until the specimen was fully liquefied. The mixed solution was then added to the detection cartridge and placed in the GeneXpert instrument for testing. The GeneXpert detection system and accompanying reagents were provided by Cepheid, USA.

2.6. Urinary LAM detection (chemiluminescence)

For urinary LAM detection, 5 mL of midstream urine was collected from the patient and the test was performed according to the instructions. One-point five mL of the test sample was transferred to a 2 mL centrifuge tube. Then, 50 µL of the magnetic bead reagent (containing LAM-capturing antibodies) was added to the centrifuge tube and mixed. The tube was labeled for identification. The labeled centrifuge tube was placed in a rotating mixer and incubated at room temperature with a rotation speed of 30 to 50 rpm for 2 hours. After incubation, the centrifuge tube was placed on a magnetic rack for adsorption. Once the components were fully separated, the supernatant was discarded. To each tube, 200 µL of sample dilution solution was added. The mixture was thoroughly mixed using a vortex mixer and within 5 minutes, the sample was processed according to the operation manual of the LAM detection chemiluminescence analyzer. If the time exceeds 5 minutes, the sample needs to be mixed again before testing. Positive was defined as S/CO > 1, S means the optical density of sample, CO represents the Cut Off. The LAM detection system and accompanying reagents were provided by Leide Biosciences Co., Ltd, China.

2.7. Statistical analysis

The data were analyzed using SPSS 20.0 statistical software (SPSS Inc., Chicago, IL). Qualitative data, such as positive detection rate and sensitivity, were presented as frequencies and percentages (%). The differences between groups were compared using the chi-square test, and a P < .05 was considered statistically significant. According to the composite reference standard, the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and Kappa value were used to evaluate different methods and consistency.

3. Results

3.1. Demographic characteristics

We recruited 216 subjects, including 135 TB patients, and 81 non-tuberculosis subjects. Gender, age, BMI and smoking status of 2 groups did not show statistical difference and were comparable (Table 1).

Table 1.

Demographic characteristics of patients.

Variables	All (N, %)	PTB (N, %)	Non-TB (N, %)	χ2	P value
Number	216	135	81
Gender				1.309	.253
Male	150 (69.4%)	90 (66.7%)	60 (74.1%)
Female	66 (30.6%)	45 (33.3%)	21 (25.9%)
Age				3.172	.366
≤25	58 (26.9%)	35 (25.9%)	23 (28.4%)
26–39	60 (27.8%)	37 (27.4%)	23 (28.4%)
40–55	45 (20.8%)	33 (24.4%)	12 (14.8%)
>55	53 (24.5%)	30 (22.2%)	23 (28.4%)
BMI				0.95	.99
<18.5	3 (1.4%)	2 (1.6%)	1 (1.1%)
18.5–23.9	205 (94.9%)	118 (94.4%)	87 (95.6%)
24–27.9	7 (3.2%)	4 (3.2%)	3 (3.3%)
≥28	1 (0.5%)	1 (0.8%)	0 (0)
Smoking status				2.404	.121
Never smoked	197 (91.2%)	120 (88.9%)	77 (95.1%)
Ever somked	19 (8.8%)	15 (11.1%)	4 (4.9%)

BMI = body mass index, Non-TB = non-tuberculosis, PTB = pulmonary tuberculosis.

3.2. Comparison the diagnostic performance of 4 methods

PPV, NPV, and Kappa value of smear for detecting PTB were 29.6%, 92.6%, 87.0%, 44.1%, and 0.182, respectively. The sensitivity, specificity, PPV, NPV, and Kappa value of sputum culture for detecting PTB were 45.9%, 97.5%, 96.9%, 52.0%, and 0.370, respectively. The sensitivity, specificity, PPV, NPV, and Kappa value of GeneXpert for detecting PTB were 59.3%, 98.8%, 98.8%, 59.3%, and 0.512, respectively. The sensitivity, specificity, PPV, NPV, and Kappa value of LAM detection for detecting PTB were 51.9%, 97.5%, 97.2%, 54.9%, and 0.427, respectively. The sensitivity of LAM was significantly higher than sputum smears (51.9% vs 29.6%, P = .000), slight higher than Sputum culture (51.9% vs 45.9%, P = .330), and slight lower than GeneXpert (51.9% vs 59.3%, P = .221) (Table 2).

Table 2.

The diagnostic performance of four methods.

Methods	Results	CRS (n = 216)		Sensitivity	Specificity	PPV	NPV	Kappa
		PTB	Non-TB
Sputum smears	Positive	40	6	29.6%	92.6%	87.0%	44.1%	0.182
	Negative	95	75
Sputum culture	Positive	62	2	45.9%	97.5%	96.9%	52.0%	0.370
	Negative	73	79
GeneXpert	Positive	80	1	59.3%	98.8%	98.8%	59.3%	0.512
	Negative	55	80
LAM	Positive	70	2	51.9%	97.5%	97.2%	54.9%	0.427
	Negative	65	79

Sensitivity = True Positive Cases/ (True Positive Cases + False Negative Cases) × 100%; Specificity = True Negative Cases/ (True Negative Cases + False Positive Cases) × 100%; PPV = True Positive Cases/ (True Positive Cases + False Positive Cases) × 100%; NPV = True Negative Cases/ (True Negative Cases + False Negative Cases) × 100%.

CRS = composite reference standard, LAM = lipoarabinomannan, Non-TB = non-tuberculosis, NPV = negative predictive value, PPV = positive predictive value, PTB = pulmonary tuberculosis.

3.3. Comparison the diagnostic performance of combined testing

The diagnostic performance of LAM detection, in combination with sputum smear, sputum culture, and GeneXpert, was higher compared to individual testing methods. The sensitivity, specificity, PPV, NPV, and Kappa value of LAM combine with sputum smears was 54.9%, 95.1%, 94.8%, 55.7%, 0.437. The sensitivity of LAM combined with sputum smear, was significantly higher than sputum smear alone (54.9% vs 29.5%, P = .000); The sensitivity, specificity, PPV, NPV, and Kappa value of LAM combine with sputum culture was 78.2%, 97.5%, 98.1%, 72.9%, 0.713. The sensitivity of LAM combined with sputum culture, was significantly higher than sputum culture alone (78.2% vs 45.9%, P = .000). The sensitivity, specificity, PPV, NPV, and Kappa value of LAM combine with GeneXpert was 71.0%, 97.5%, 98.0%, 66.9%, 0.629. The sensitivity of LAM combined with GeneXpert, was significantly higher than GeneXpert alone (71.0% vs 59.3%, P = .001). The best combine testing is LAM combine with sputum culture, the sensitivity is significantly higher than LAM combined with sputum smear (78.2% vs 54.5%, P = .000), and slightly higher than LAM combine with GeneXpert (78.2% vs 71.0%, P = .161) (Table 3).

Table 3.

The diagnostic performance of combine testing.

Combine	Results	CRS (n = 216)		Sensitivity	Specificity	PPV	NPV	Kappa
		PTB	Non-TB
LAM + Sputum smears	Positive	74	4	54.5%	95.1%	94.8%	55.6%	0.437
	Negative	61	77
LAM + Sputum culture	Positive	106	2	78.2%	97.5%	98.1%	72.9%	0.713
	Negative	29	79
LAM + GeneXpert	Positive	96	2	71.0%	97.5%	98.0%	66.9%	0.629
	Negative	39	79

Sensitivity = True Positive Cases/ (True Positive Cases + False Negative Cases) × 100%; Specificity = True Negative Cases/ (True Negative Cases + False Positive Cases) × 100%; PPV = True Positive Cases/ (True Positive Cases + False Positive Cases) × 100%; NPV = True Negative Cases/ (True Negative Cases + False Negative Cases) × 100%.

CRS = composite reference standard, LAM = lipoarabinomannan, Non-TB = non-tuberculosis, NPV = negative predictive value, PPV = positive predictive value, PTB = pulmonary tuberculosis.

3.4. The detection performance of different methods for culture-negative samples

A total of 73 TB samples with sputum culture-negative result were selected as the gold standard (Table 4). When the sputum culture results were negative, the positive rate of smear, GeneXpert and LAM was 0%, 53.4% and 52.1%, respectively. The positive rate of LAM is significantly higher than sputum smears (52.1% vs 0%, P = .000), is similar to GeneXpert (52.1% vs 53.4%, P = .931). The positive rate of LAM combined with GeneXpert, was significantly higher than GeneXpert alone (53.4% vs 68.5%, P = .042).

Table 4.

The positive rate of other methods in sputum culture-negative.

	Sputum culture-negative (n = 73)
	Sputum smears	GeneXpert	LAM	LAM + Sputum smears	LAM + GeneXpert
n	0	39	38	38	50
Positive rate	0.0%	53.4%	52.1%	52.1%	68.5%

LAM = lipoarabinomannan.

4. Discussion

Throughout the past decade, TB remains a major public health challenge around the world. China accounts for more than 10% of the global TB burden.^[21] WHO estimates that approximately 10 million individuals develop active TB annually, but 30% of these (3 million) are never diagnosed or reported.^[22] Therefore, there is a critical need for a clear and rapid diagnosis of TB, guiding drug use, and controlling the development of diseases. With low infection control risks, noninvasive, and convenient sample collection, the urine-based LAM assay was identified as a focus.^[23] The urine-based LAM assay, a simple point-of-care test, has been used in the commercial development of rapid detection of MTB.^[24] In addition, LAM assay reduces costs over culture and diagnoses patients with EPTB with slightly higher accuracy than smear.^[25]

In this study, the positive detection rates of sputum smears, sputum culture, and GeneXpert were found to be 29.6%, 45.9%, and 59.3%, respectively, aligning with previously reported literature values of 24.0%,^[26] 40.0%,^[26] and 63.9%.^[27] Notably, the sensitivity of the urinary LAM assay (chemiluminescence) was determined to be 51.9%, which is significantly higher than that of sputum smears and sputum culture but lower than that of the GeneXpert method. Importantly, our LAM assay demonstrated marked clinical efficacy, surpassing the performance of both the Fujifilm SILVAMP TB LAM and AlereLAM assays, with an average 30% improvement in sensitivity for detecting TB in HIV patients.^[28]

Moreover, the integration of various laboratory diagnostic methods has opened up new avenues for increasing the positive detection rate in clinical practice.^[29] In light of this, our study aimed to explore the combined detection of the urine LAM assay with sputum smears, culture, and GeneXpert. The results revealed that the positive detection rates achieved through combined detection using the urine LAM assay in conjunction with sputum smears, culture, and GeneXpert were 54.5%, 78.2%, and 71.0%, respectively, among patients with PTB. These findings highlight the substantial improvement in diagnostic sensitivity when employing the combined detection approach, as compared to individual pathogen-specific methods. Peng et al^[30] also found that LAM combined with other methods can improve positive detection rate in pleural effusion. The combination of LAM and Xpert yielded a sensitivity of 71.0%, comparable to the best LAM combination reported in another study,^[31] but the reagents used in this research have not been commercially produced. Notably, the combination of LAM and culture showed the highest sensitivity of 78.2%. This combination holds greater value in regions with limited medical resources or impoverished settings.

In clinical practice, approximately 50% of patients do not have pathogen-specific results or exhibit negative test results.^[32] This study aimed to investigate the detection rates of sputum culture-negative patients using sputum smears, urine LAM assay, and GeneXpert. Our findings revealed that LAM had a detection rate of 52.1%, which was significantly higher than that of sputum smears. Furthermore, combining urine LAM assay with GeneXpert led to a significant increase in the detection rate relative to GeneXpert used alone. For culture-negative patients, we recommend a combined detection approach using urine LAM assay with GeneXpert to enhance the overall detection rate.

LAM is a specific antigen on the cell wall of MTB that can be filtered by the kidneys and detected in urine.^[13] Urine is a sterile and easily accessible body fluid that can be tested even in patients with sputum deficiency, and LAM is a thermally stable antigen that remains detectable in boiled urine through the use of highly sensitive immunological techniques.^[33] Numerous studies have indicated that urine LAM testing may improve the accuracy of the diagnosis of active tuberculosis, especially in critically ill patients with complicated H1V infection and low CD4 + T lymphocyte counts.^[34,35] In this study, the high-sensitivity urine LAM test (chemiluminescence-based) was used for the first time in the diagnosis of tuberculosis in HIV-negative population, and it was found that its sensitivity has surpassed that of sputum smear and sputum culture. In addition, it has been reported that the specificity of LAM testing can be affected by certain nontuberculous mycobacterial infections, leading to false positive results.^[36] Therefore, in clinical practice, LAM testing may not be able to distinguish between nontuberculous mycobacterial and MTB. In addition, other factors such as the stage of TB infection, bacterial load and immune response will also affect the detection rate of LAM,^[34,35] and further research is needed to explore the factors that affect LAM concentration in the urine and improve its diagnostic performance of TB as much as possible.

This study has several limitations. Firstly, it is a single-center study and did not cover data from multiple centers or a large sample size. Therefore, further research is needed to expand to broader regions and include a larger number of samples. Secondly, it is important to combine other diagnostic methods, such as TB-DNA, to enhance the diagnostic value for PTB and better assist in clinical diagnosis and treatment. Lastly, this study lacks an evaluation of the efficacy of other microbiological methods for the diagnosis of EPTB. Therefore, further research is required to select the optimal laboratory diagnostic methods. However, considering the ease of operation and patient burden, LAM detection still holds significant clinical value and can compensate for the limitations of existing bacteriological techniques.

5. Conclusion

The study evaluated the diagnostic value of urinary LAM antigen detection (chemiluminescence) for TB in HIV-negative patients. The results showed that the sensitivity of the LAM detection method was higher than sputum smears, slightly higher than sputum culture, and slightly lower than the GeneXpert method. When combined with other diagnostic methods, LAM detection showed improved sensitivity compared to individual methods. The combination of LAM detection with sputum culture demonstrated the highest sensitivity. Additionally, in cases where sputum culture results were negative, LAM detection had a higher positive rate compared to sputum smears and a similar positive rate to the GeneXpert method. Overall, these findings suggest that urinary LAM antigen detection can be a valuable adjunctive diagnostic tool for HIV-negative TB cases, particularly when used in combination with sputum culture. Further research and validation studies are warranted to confirm these results and explore the potential clinical utility of this detection method.

Acknowledgements

This study was supported by the Leading Talents in Entrepreneurship and Innovation of Guangzhou Development Zone (Phase II) (No. 2021-L033) and Yang Cheng Innovation and Entrepreneurship Talent Support Program (No.2016011). We thank the Leide Biosciences Co., Ltd for providing kits and testing services.

Author contributions

Conceptualization: Yingxuan Zhang, Shihao Chen, Hongxia Wei, Qianhong Zhong, Yiwu Yuan, Jianrong Lou, Xilin Zhang.

Data curation: Yingxuan Zhang, Shihao Chen, Yiwu Yuan, Xilin Zhang.

Formal analysis: Yingxuan Zhang, Shihao Chen.

Funding acquisition: Jianrong Lou.

Investigation: Yingxuan Zhang, Shihao Chen, Yiwu Yuan, Yongping Wang.

Methodology: Yingxuan Zhang, Hongxia Wei, Qianhong Zhong, Yongping Wang, Xilin Zhang.

Resources: Shihao Chen, Xilin Zhang.

Software: Yingxuan Zhang, Shihao Chen, Jianrong Lou.

Supervision: Jianrong Lou, Xilin Zhang.

Validation: Hongxia Wei, Shihao Chen.

Writing – original draft: Yingxuan Zhang, Shihao Chen.

Writing – review & editing: Hongxia Wei, Jianrong Lou, Xilin Zhang.