Clinical application of the urinary lipoarabinomannan (AIMLAM) test in PLHIV with TB

Abstract

Objective: To evaluate the diagnostic performance and clinical utility of a new antigen test for the detection of urinary lipoarabinomannan (AIMLAM) in people living with HIV (PLHIV) with comorbid tuberculosis (TB).

Methods: This study included 82 PLHIV who were presumed to have TB and were admitted to Yunnan Infectious Disease Hospital from December 1, 2023, to July 1, 2024. General clinical data were collected. Urine samples were collected from all patients and subjected to AIMLAM antigen detection via chemiluminescence. Appropriate samples (such as sputum or other samples) were collected from patients in good condition for MGIT 960 mycobacterial culture (MGIT 960), GeneXpert-MTB/RIF testing (Xpert), and acid‒fast bacillus staining (AFB). According to the gold standard for outcome prediction, the patient’s history, clinical symptoms, confirmed microbiological features, and radiological/imaging findings, etc., were integrated to predict the overall outcome.

Results: Urinary AIMLAM antigen detection demonstrated greater sensitivity (83.3%) and specificity (70.7%) than MGIT 960 culture, GeneXpert-MTB/RIF, and AFB staining for diagnosing advanced HIV disease (AHD)-related TB. The diagnostic efficacy of urinary AIMLAM antigen detection was superior to that of the other tests, with an area under the curve (AUC) of 0.774, indicating significant clinical value in diagnosing TB among patients with AHD. There were 33 AIMLAM antigen-positive cases with CD4 cell counts ≤ 200 cells/µL, but the results for 14 cases were identified as false positives because of coinfections with nontuberculous mycobacteria (NTM), invasive fungal infections, or severe bacterial infections. Statistically significant differences in signal-to-cutoff ratio (S/CO) values from AIMLAM testing were observed among patients with mycobacterial, fungal, and bacterial infections (P < 0.05), but no significant difference was detected between patients with TB and those with NTM infections (P > 0.05).

Conclusion: Urine AIMLAM antigen detection is an effective tool for initial screening of TB in patients with AHD. It has higher sensitivity than MGIT 960, GeneXpert-MTB/RIF, and AFB smear, but the method needs to be optimized to increase its specificity. Positive AIMLAM results with high S/CO values suggest mycobacterial infection, although it cannot be used to differentiate TB from NTM infections. Low S/CO values in HIV patients warrant the consideration of fungal or other opportunistic infections. Species-specific confirmation is needed before treatment initiation.

Keywords: MTB; AIMLAM antigen; Lipomannan; Opportunistic infection; PLHIV; Sensitivity.

Fig. 1

ROC curves for AIMLAM antigen detection and other diagnostic methods. AIMLAM antigen detection had a greater AUC than the other methods did, indicating its better overall diagnostic performance

Fig. 2

Comparison of AIMLAM antigen S/CO values. The S/CO values in patients with MTB infection is significantly higher compared to those with fungal or other pathogenic infections. Note: *: Statistically significant, P < 0.05. **: Statistically significant, P < 0.01