Tuberculosis (TB) is a leading cause of morbidity and mortality in HIV-infected patients accessing antiretroviral treatment (ART) programs or requiring medical admission to hospitals in sub-Saharan Africa.1–5 However, diagnosis of TB in these patient groups is often missed or delayed as a result of the nonspecific clinical presentation, high rates of sputum smear negative, extrapulmonary and disseminated disease, and the limited utility of pulmonary radiology.6,7 There is a great need for new diagnostic tools to increase the sensitivity and speed of diagnosis in these patient groups, especially in view of their high mortality and the risk of nosocomial TB transmission.
A range of new diagnostics for TB is now emerging, employing various different technologies.8 One area of renewed interest has focused on the potential for TB diagnosis to be made from analysis of urine samples. Urine is an attractive sample for diagnosis as it is simple to obtain, even from very ill patients who may not be able to produce sputum. Urine sampling does not generate hazardous infectious aerosols, and urine is relatively clean and easy to handle in the laboratory. Urine may be cultured,9 tested by polymerase chain reaction for mycobacterial transrenal DNA10 or tested for specific mycobacterial antigens such as lipoarabinomannan (LAM).11–13
More recently, the Xpert MTB/RIF rapid molecular assay14,15 (hereafter referred to as ‘Xpert’) has been endorsed by the World Health Organization (WHO) as a replacement for sputum smear microscopy.16 However, there are also reports of Xpert being used to diagnose TB by testing a range of extrapulmonary samples, including urine.17,18 In this study, we therefore determined the diagnostic sensitivity of Xpert MTB/RIF when testing small volumes of urine obtained from HIV-infected patients being screened for TB before starting ART in South Africa.
The ART service in Gugulethu township, Cape Town, South Africa, and the burden of TB among these patients have previously been described.2,19 The patients in this study were enrolled into the parent study12 between March 12, 2010, and April 20, 2011, to assess the diagnostic accuracy of the urine LAM point-of-care assay (Determine TB-LAM). Consecutive eligible HIV-infected patients were recruited from among patients newly referred to the clinic for ART. Study eligibility criteria included age >18 years, being ART naive, and having no current TB diagnosis. All participants provided written informed consent, and the study was approved by the research ethics committees of the University of Cape Town, South Africa, and the London School of Hygiene and Tropical Medicine, United Kingdom.
After recruitment at their first visit to the clinic, patients' demographic details were recorded and they were clinically characterized. A standardized TB symptom-screening questionnaire (including the WHO symptom screen for HIV-associated TB20) was completed, and routine baseline investigations and screening tests for TB were done. Two sputum samples were requested from each patient; a spot specimen was first obtained followed by a second obtained by induction using nebulized 3% hypertonic saline. Urine samples were also collected and stored at −20°C within 3 hours. Blood CD4 cell counts and plasma viral load were measured on all patients via the routine laboratory services. Chest radiographs were obtained and reported by an experienced reader certified in the use of the chest radiograph reading and recording system.21,22
Patients were followed up within the routine ART service and patients diagnosed as having TB were referred to TB clinics within the township for treatment. ART service records of those with TB were reviewed to ascertain deaths within the first 90 days after enrollment.
Sputum specimens were processed using standardized protocols, and quality assurance procedures by a centralized accredited laboratory as described in detail previously.12,23 In brief, samples were decontaminated with N-acetyl-L-cysteine and sodium hydroxide and concentrated by centrifugation. Smears prepared from the sputum pellets underwent fluorescence microscopy. Equal volumes of the remaining pellet were tested by the Xpert assay and by liquid culture using Mycobacterial Growth Indicator Tubes (Becton Dickinson, Sparks, MA), which were incubated for up to 6 weeks. Cultures positive for acid-fast bacilli were identified as Mycobacterium tuberculosis complex using the MTBDRplus assay (Hain Lifesciences, Nehren, Germany), which also provided genotypic assessment of susceptibility to rifampicin and isoniazid. Phenotypic drug susceptibility testing was also done in liquid culture using the modified proportion method.
Frozen urine samples were defrosted and retrospectively analyzed for the presence of LAM using both the Clearview MTB-ELISA (Alere, Waltham, MA) the Determine TB-LAM point-of-care test strips (Alere, Lot #101102) with strict adherence to the manufacturer instructions.12 Defrosted urine samples (2.0 mL) were also concentrated by centrifugation, the pellet resuspended in 0.75 mL of phosphate buffer and then tested using the Xpert assay according to the manufacturer instructions.
Definitions and Analysis
Patients were defined as having TB if Mycobacterium tuberculosis was cultured from one or more sputum samples. Patients were then categorized as having urine Xpert-positive or urine Xpert-negative disease. Clinical, microbiological and radiological characteristics and 90-day outcomes of patients with Xpert-positive and Xpert-negative TB were compared using the Wilcoxon rank-sum test, t test, χ2, and Fisher exact tests as appropriate. Logistic regression analysis was used to identify factors independently associated with urine Xpert-positive test results. All statistical tests were 2-sided at alpha = 0.05.
Among 602 patients recruited, 535 produced at least 1 sputum sample and a specimen of urine. Sputum culture results were available from 516 patients and these yielded 85 diagnoses of culture-positive TB. The remainder (n = 431) were sputum culture negative. TB cases were young adults of whom more than a half were female (Table 1). The median plasma viral load was high and the median (interquartile range) CD4 cell count was 138 cells/μL (interquartile range: 63–205). A positive WHO TB symptom screen was recorded in 82.1% of TB patients but only 25.0% reported a cough lasting ≥2 weeks. Disease was sputum smear positive in 28.6% of cases and smear negative in 71.4% (Table 1). Xpert testing of sputum was positive in 58.3% from the first sample and 70.2% when testing both sputum samples.
Yield of TB Diagnoses From Testing Urine With Xpert MTB/RIF
Of the patients with sputum culture–positive TB and available urine Xpert results (n = 84), 16 tested positive, giving a sensitivity of 19.0% [95% confidence interval (CI): 11.3–29.1]. The sensitivity was strongly associated with blood CD4 cell count (Fig. 1A). It was higher in those with counts <50 cells per microliter (44.4%; 95% CI: 21.5 to 69.2) compared with those with counts of 50–150 cells per microliter (25.0%; 95% CI: 10.7 to 44.9) or >150 cells per microliter (2.7%; 95% CI: 0.1 to 14.2).
Xpert testing of both sputum and urine samples produced a small incremental diagnostic yield over Xpert testing of sputum only. When Xpert testing was done on one sputum sample, an additional 4 cases of TB were diagnosed by also testing urine samples with Xpert (sensitivity: 58.3% vs. 63.1%). When testing 2 sputum samples, an additional 2 cases of TB were diagnosed by also testing urine samples with Xpert (70.2% vs. 72.6%).
The sensitivity of urine Xpert testing was approximately two-thirds of that observed using the Determine TB-LAM point-of-care assay for LAM. The overall sensitivity of this assay for all TB patients was 28.6% (19.2–39.5) and for those with CD4 cell counts of <50 cells per microliter, 50–150 cells per microliter, and >150 cells per microliter, sensitivities were 66.7% (41.0–86.7), 32.1% (15.9–52.4), and 8.1% (1.7–21.9), respectively (Fig. 1B).
Comparison of Urine Xpert-Positive and Xpert-Negative Cases
We compared the characteristics and microbiological results of patients with urine Xpert-positive (n = 16) and urine Xpert-negative (n = 68) disease (Table 1). Those with urine Xpert-positive disease were younger, tended to have lower body mass indices and hemoglobin concentrations, and higher neutrophil counts. They also had more advanced HIV, with higher viral loads and a much lower median CD4 cell count compared with patients with urine Xpert-negative disease.
Multivariate logistic regression analysis was done and this included CD4 cell count, viral load, and those patient characteristics and blood tests results that were found to be significantly associated with Xpert status in univariate analysis in Table 1. A positive urine Xpert result was independently associated with just 2 patient characteristics—younger age and lower CD4 cell counts. Assigning patients with CD4 cell counts of >150 cells per microliter as the reference group, the adjusted odds of a positive urine Xpert result among patients with CD4 cell counts of either 50–150 cells per microliter or <50 cells per microliter were 13.6 (95% CI: 1.33 to 139) and 20.1 (95% CI: 1.72 to 233), respectively (P = 0.010).
Although there were no significant differences in the pulmonary radiographic features of disease comparing the 2 groups of patients (Table 1), patients with urine Xpert-positive TB had microbiological evidence of high mycobacterial burden. They had higher frequencies of positive sputum smear microscopy and positive sputum Xpert results, shorter time to culture positivity, and a higher frequency of detectable LAM in urine (Table 1).
Four cases of multidrug-resistant TB were confirmed by phenotypic and genotypic assessment of sputum culture isolates. Rifampicin resistance was detected in all of these cases by Xpert testing of sputum. Xpert detected M. tuberculosis in the urine of only one of these patients but correctly identified rifampicin resistance. In addition, results of Xpert done on urine samples also reported rifampicin resistance in 2 other patients whose sputum isolates were drug susceptible. These were either false-positive Xpert results or represented mixed infection with resistant and susceptible strains.
Urine Samples From Non-TB Cases
Of patients (n = 431) with negative sputum cultures, we tested urine samples from a sample of consecutive patients equal in number to the TB cases studied (n = 84). The characteristics of these patients (data not shown) were very similar to those of non-TB patients in the overall cohort included in the parent study,12 indicating that these were a representative sample. Urine specimens from these 84 patients were tested using Xpert, and all were negative.
Outcomes up to 90 days from screening were assessed. During follow-up, 4 of 16 (25%) of urine Xpert-positive patients died compared with 1 of 68 (1.5%) of urine Xpert-negative patients (P = 0.004). All deaths occurred within the first 30 days.
In this study, one-fifth of ambulatory HIV-infected patients with advanced immunodeficiency and sputum culture-positive pulmonary TB had M. tuberculosis that was detectable when testing small volumes of urine with the Xpert assay. The diagnostic yield was strongly associated with blood CD4 cell count with almost one half of those with CD4 cell counts <50 cells per microliter testing positive, whereas the yield at CD4 counts >150 cells per microliter was very low. There was a small incremental diagnostic yield from Xpert assays when spot urine samples were tested in addition to sputum samples. This represents an alternative rapid diagnostic modality in HIV-infected patients with advanced immunodeficiency.
Previous studies have reported that concentration of urine from HIV-infected patients by centrifugation and subsequent culture has an appreciable diagnostic yield.9,24,25 However, this entails collection and processing of comparatively large volumes of urine, for example pooling morning samples over 3 days.25 Urine culture is infrequently useful as a diagnostic modality in this clinical population because of the prolonged time to culture positivity. In this study, however, we tested very small (2.0 mL) samples of urine and still found an important diagnostic yield that was similar to that obtained by culture of concentrated large volumes of urine in a study of HIV-infected patients in Thailand.9 The sensitivity of Xpert in our study might have been even greater had a whole urine sample been collected and concentrated.
A major advantage of Xpert is rapidity, with results being generated within 2 hours compared with culture for which the time to positivity often exceeds 3 weeks for paucibacillary clinical samples.11,23 Rapid diagnosis is particularly important for patients with high mortality risk. Our analyses demonstrated that similar to sputum testing,26 urine Xpert testing is most likely to be positive in patients with the most advanced immunodeficiency and poor prognostic characteristics (low body mass indices and low hemoglobin concentrations). One quarter of those who retrospectively tested urine Xpert-positive TB died. Thus, this represents a particularly useful diagnostic modality in patients with low CD4 cell counts and high risk of death. We assessed ambulatory patients and yet the assay may have even greater utility among sick patients requiring hospital admission who may be unable to obtain good quality sputum samples and in whom risk of disseminated TB is high.5
In contrast to detection in urine of nonorganism-associated transrenal DNA fragments, the Xpert assay detects intact Mycobacterium tuberculosis bacilli. This is because the cartridge-based processing entails sputum lysis, washing and deposition of whole mycobacteria on a filter membrane before ultrasonic disruption, and real-time PCR amplification and detection.15 Thus, detection of M. tuberculosis in urine using Xpert indicates renal tract involvement with TB as the bacilli would otherwise be unable to enter the urine. The sensitivity of urine Xpert was approximately two-thirds that of Determine TB-LAM which, in contrast to Xpert, detects mycobacterial cell wall LAM antigen rather than the whole organism. This is consistent with findings from another study from Cape Town investigating the mechanisms of LAM antigenuria.27 We do not know whether the sensitivity of urine Xpert might have been similar to that of the LAM assay had much larger volumes of urine been concentrated by centrifugation before testing.
Xpert has been found to have very high sensitivity but sub-optimal specificity for rifampicin resistance.15 In this study, 2 patients had rifampicin resistant M. tuberculosis when testing urine with Xpert, but the strains isolated from sputum were susceptible. It is not possible to ascertain whether these are false-positives (which are usually stochastic events, related to poor binding of detection probe B to the amplicon during the assay15) or whether the patients could have had disease due to 2 different strains. It was not possible to investigate this with molecular typing without paired isolates.
It was notable that urine Xpert results were only positive in patients with positive sputum cultures and were negative in all those who tested sputum culture negative. Detection of M. tuberculosis in both urine and sputum samples indicates the presence of disseminated disease with both pulmonary and renal tract involvement and this also reflects the lack of disease anatomic compartmentalization in patients with advanced immunodeficiency. Assessment of both sputum and urine samples using Xpert provides a means of rapid assessment for disseminated TB.
Strengths of this study include the recruitment of carefully characterized consecutive patients in a well-documented clinical cohort. We do not know whether the diagnostic yield might have been greater had larger volumes of urine been tested, and the relative sensitivity of urine Xpert and urine culture was not assessed. Urine testing was done retrospectively and it is not known whether prospective testing would have accelerated diagnosis or improved patient outcomes. Further prospective evaluation is needed to determine whether this provides a useful diagnostic modality during clinical care in both high and low TB burden settings.
In conclusion, analysis of urine samples using the Xpert assay provides useful diagnostic yield for TB among patients with advanced HIV-associated immunodeficiency. High rates of detection among those with low CD4 cell counts provide evidence of frequent disease dissemination.
The authors are grateful to the Foundation for Innovative New Diagnostics, Geneva, Switzerland, for providing access to the Xpert MTB/RIF assay cartridges with preferential pricing. Alere provided the LAM assays free of charge. The authors thank sister Pearl Pahlana and the staff of the Hannan Crusaid ART clinic.
1. Lawn SD, Harries AD, Anglaret X, et al.. Early mortality among adults accessing antiretroviral treatment programmes in sub-Saharan Africa. AIDS. 2008;22:1897–1908.
2. Lawn SD, Myer L, Bekker LG, et al.. Burden of tuberculosis in an antiretroviral treatment programme in sub-Saharan Africa: impact on treatment outcomes and implications for tuberculosis control. AIDS. 2006;20:1605–1612.
3. Moore D, Liechty C, Ekwaru P, et al.. Prevalence, incidence and mortality associated with tuberculosis in HIV-infected patients initiating antiretroviral therapy in rural Uganda. AIDS. 2007;21:713–719.
4. Bassett IV, Wang B, Chetty S, et al.. Intensive tuberculosis screening for HIV-infected patients starting antiretroviral therapy in Durban, South Africa. Clin Infect Dis. 2010;51:823–829.
5. Cohen T, Murray M, Wallengren K, et al.. The prevalence and drug sensitivity of tuberculosis among patients dying in hospital in KwaZulu-Natal, South Africa: a postmortem study. PLoS Med. 2010;7:e1000296.
6. Lawn SD, Wood R. Tuberculosis in antiretroviral treatment services in resource-limited settings: addressing the challenges of screening and diagnosis. J Infect Dis. 2011;204(suppl 4):S1159–S1167.
7. Reid MJ, Shah NS. Approaches to tuberculosis screening and diagnosis in people with HIV in resource-limited settings. Lancet Infect Dis. 2009;9:173–184.
8. Pai M, Minion J, Sohn H, et al.. Novel and improved technologies for tuberculosis diagnosis: progress and challenges. Clin Chest Med. 2009;30:701–716. viii.
9. Monkongdee P, McCarthy KD, Cain KP, et al.. Yield of acid-fast smear and mycobacterial culture for tuberculosis diagnosis in people with human immunodeficiency virus. Am J Respir Crit Care Med. 2009;180:903–908.
10. Green C, Huggett JF, Talbot E, et al.. Rapid diagnosis of tuberculosis through the detection of mycobacterial DNA in urine by nucleic acid amplification methods. Lancet Infect Dis. 2009;9:505–511.
11. Lawn SD, Edwards DJ, Kranzer K, et al.. Urine lipoarabinomannan assay for tuberculosis screening before antiretroviral therapy diagnostic yield and association with immune reconstitution disease. AIDS. 2009;23:1875–1880.
12. Lawn SD, Kerkhoff AD, Vogt M, et al.. Diagnostic accuracy of a low-cost, urine antigen, point-of-care screening assay for HIV-associated pulmonary tuberculosis before antiretroviral therapy: a descriptive study. Lancet Infect Dis. 2012;12:201–209.
13. Shah M, Variava E, Holmes CB, et al.. Diagnostic accuracy of a urine lipoarabinomannan test for tuberculosis in hospitalized patients in a High HIV prevalence setting. J Acquir Immune Defic Syndr. 2009;52:145–151.
14. Boehme CC, Nabeta P, Hillemann D, et al.. Rapid molecular detection of tuberculosis and rifampin resistance. N Engl J Med. 2010;363:1005–1015.
15. Lawn SD, Nicol MP. Xpert MTB/RIF assay: development, evaluation and implementation of a new rapid molecular diagnostic for tuberculosis and rifampicin resistance. Future Microbiol. 2011;6:1067–1082.
16. WHO. Automated Real-Time Nucleic Acid Amplification Technology for Rapid and Simultaneous Detection of Tuberculosis and Rifampicin Resistance: Xpert MTB/RIF System Policy Statement. Geneva, Switzerland: WHO; 2011. Available at: http://whqlibdoc.who.int/publications/2011/9789241501545_eng.pdf
. Accessed June 28, 2011.
17. Lawn SD, Zumla AI. Diagnosis of extrapulmonary tuberculosis using the Xpert MTB/RIF assay. Exp Rev Anti-Infect Ther. 2012. In press.
18. Hillemann D, Rusch-Gerdes S, Boehme C, et al.. Rapid molecular detection of extrapulmonary tuberculosis by the automated GeneXpert MTB/RIF system. J Clin Microbiol. 2011;49:1202–1205.
19. Lawn SD, Myer L, Orrell C, et al.. Early mortality among adults accessing a community-based antiretroviral service in South Africa: implications for programme design. AIDS. 2005;19:2141–2148.
20. Getahun H, Kittikraisak W, Heilig CM, et al.. Development of a standardized screening rule for tuberculosis in people living with HIV in resource-constrained settings: individual participant data meta-analysis of observational studies. PLoS Med. 2011;8:e1000391.
21. Dawson R, Masuka P, Edwards DJ, et al.. Chest radiograph reading and recording system: evaluation for tuberculosis screening in patients with advanced HIV. Int J Tuberc Lung Dis. 2010;14:52–58.
22. Den Boon S, Bateman ED, Enarson DA, et al.. Development and evaluation of a new chest radiograph reading and recording system for epidemiological surveys of tuberculosis and lung disease. Int J Tuberc Lung Dis. 2005;9:1088–1096.
23. Lawn SD, Brooks SV, Kranzer K, et al.. Screening for HIV-associated tuberculosis and rifampicin resistance before antiretroviral therapy using the Xpert MTB/RIF assay: a prospective study. PLoS Med. 2011;8:e1001067.
24. Shafer RW, Kim DS, Weiss JP, et al.. Extrapulmonary tuberculosis in patients with human immunodeficiency virus infection. Medicine. 1991;7:384–397.
25. Gopinath K, Singh S. Urine as an adjunct specimen for the diagnosis of active pulmonary tuberculosis. Int J Infect Dis. 2009;13:374–379.
26. Lawn SD, Kerkhoff AD, Vogt M, et al.. Characteristics and early outcomes of patients with Xpert MTB/RIF-negative pulmonary tuberculosis diagnosed during screening before antiretroviral therapy. Clin Infect Dis. 2012;54:1071–1079.
27. Wood R, Racow K, Bekker LG, et al.. Host and pathogen factors impacting detection of lipoarabinomannan in urine during tuberculosis treatment and association with mycobacteriuria. BMC Infect Dis. 2012;12:47.