|Year : 2019 | Volume
| Issue : 4 | Page : 258-263
Early diagnosis of smear-negative childhood pulmonary tuberculosis and its substantial yield in gastric lavage/aspirates through cartridge-based nucleic acid amplification test (Xpert Mycobacterium tuberculosis/rifampicin assay)
Prasanta Kumar Das, Somtirtha B Ganguly, Bodhisatya Mandal
State TB Demonstration Cum Training Centre, Intermediate Reference Laboratory, Kolkata, West Bengal, India
|Date of Submission||04-Jun-2019|
|Date of Acceptance||17-Sep-2019|
|Date of Web Publication||03-Dec-2019|
Dr. Prasanta Kumar Das
State TB Demonstration Cum Training Centre, Intermediate Reference Laboratory, Kolkata, West Bengal
Source of Support: None, Conflict of Interest: None
Introduction: Tuberculosis (TB) still remains the major public health threat in India. Early diagnosis is a priority as any delay may complicate the prognosis further leading to the failure of an effective control. India accounts for 6% incidence of pediatric TB cases in a population that has 40% as estimated latent TB cases. Gastric Lavage /aspirates were taken under study for pediatric patients as key population. Collection of sputum samples from the pediatric patients has been very difficult. The most common implementation of cartridge-based nucleic acid amplification tests (CBNAATs) in diagnosis of pediatric TB has augmented the detection rates. As a mandate for universal drug sensitivity, testing of smear-negative pediatric samples is set as a priority. The program included the smear-negative pulmonary samples contributing to high yield after sputum includes induced sputum, bronchoalveolar lavage, gastric aspirate, and gastric lavage. The study involves an analysis probing into the percentage contribution of these samples and the substantial yields of gastric lavage among these samples. Methods: Pulmonary pediatric samples were tested on CBNAAT by Xpert Mycobacterium tuberculosis/rifampicin (RIF) (Cepheid) based on the manufacturer's instruction. All necessary aseptic measures were taken. The data were captured in the Xpert software automatically during the tests and exported to the Microsoft Excel sheets for further analysis. A defined study design against each and every objective was set up. Results: From earlier studies, it was found that a point prevalence of 6%–7% of pediatric TB exists among the tested specimens every quarter. The periodic prevalence was found to be 5%, and the incidence rates ranged from 4.5% to 5%. RIF resistance detection showing a seasonal variation ranged from 13% to 15% among the CBNAAT-positive cases. Conclusions: The study concluded that more skilled collection centers for biological specimens are required to address the undetected pulmonary TB cases among the pediatric age group, especially below 6 months of age. About 5% prevalence and around 4.9% incidence is an alarming situation in the TB control scenario of West Bengal. Gastric lavage showed a major detection in children aged below 6 months of whom collecting sputum samples were difficult. This prompted suboptimal detection levels due to the dearth of sample collection modalities at the peripheries.
Keywords: Cartridge-based nucleic acid amplification test, childhood pulmonary tuberculosis, gastric lavage
|How to cite this article:|
Das PK, Ganguly SB, Mandal B. Early diagnosis of smear-negative childhood pulmonary tuberculosis and its substantial yield in gastric lavage/aspirates through cartridge-based nucleic acid amplification test (Xpert Mycobacterium tuberculosis/rifampicin assay). Biomed Biotechnol Res J 2019;3:258-63
|How to cite this URL:|
Das PK, Ganguly SB, Mandal B. Early diagnosis of smear-negative childhood pulmonary tuberculosis and its substantial yield in gastric lavage/aspirates through cartridge-based nucleic acid amplification test (Xpert Mycobacterium tuberculosis/rifampicin assay). Biomed Biotechnol Res J [serial online] 2019 [cited 2020 Jan 18];3:258-63. Available from: http://www.bmbtrj.org/text.asp?2019/3/4/258/272190
| Introduction|| |
It is estimated that India accounts for 25% of global tuberculosis (TB) burden of the world. The pediatric population accounted for 6% of the total burden. In West Bengal, India, before the availability of cartridge-based nucleic acid amplification test (CBNAAT), the pediatric TB was seen in 3.8% cases among the total population tested.
As the initial diagnosis was solely dependent on smear microscopy, paucibacillary samples were left undetected. It is estimated that India accounts for around 40% of the latent TB infection. The pediatric case detections were less owing to a dearth of a sensitive technology not being able to detect the underlying cause.,, The drug resistance cannot be detected by smear microscopy. Nucleic acid amplification methods are able to detect Mycobacterium tuberculosis complex (MTBC) categorically along with the point mutations at the rifampicin (RIF) resistance determining region. Moreover, the analytical sensitivity of smear microscopy is not sufficient enough as it accounts for a poor-positive predictive value owing to its nonspecific detections.
The following study explored the increased case detection for the pediatric group among a variety of samples after the availability of CBNAAT in the remotest districts of West Bengal. The alarming data of >4% pediatric TB prompted in further collection of samples. It has been seen that collecting sputum samples from pediatric age group has been very difficult. Various samples other than sputum were collected and their percentage positivity was studied.
The role of gastric lavage in the diagnosis of pediatric TB population is assessed.
| Methods|| |
The study explores a multidimensional analysis of the CBNAAT performed on the samples among the pediatric population covering the entire state of West Bengal.
The objectives of the study included figuring out percentage positivity among the Gastric aspirates / lavage samples sent at Intermediate Reference Laboratory, Kolkata.
The retrospective study was undertaken covering a period from May 2014 to May 2018.
All pediatric samples, pulmonary, irrespective of their geographical location that came to the Intermediate Reference Laboratory with a request for CBNAAT have been included in the study.
The methods involve a retrospective multidimensional analysis of the CBNAAT's data of the pediatric samples coming to Intermediate Reference Laboratory, Kolkata, from May 2014 to May 2017. The proposal was reviewed and approved by the Research and Ethical committee Intermediate Reference Laboratory, Kolkata, India. The Ethics Committee Approval number STDC/IRL–CBNAAT Peadiatrc project No. 012/2014 dated January 5, 2014, to 2017.
Data were captured for the entire period from the recording and registration register; the study designed was worked out based on the available data using computational software.
Selective screening helped in tracking the contribution of the referral sites. The age- and sample-related propensity in the case detection was also studied. The probe comprised not only the detection of MTBC by CBNAAT but also the incidence and concordance studies of the RIF-resistant cases.
Before the processing of the sample for Gene Xpert, around 20 μl was taken for making smears on grease-free slides.
These slides were stained and examined under a fluorescent microscope.
The samples were processed as per the standard operating procedures led down by the GeneXpert MTB/RIF guidelines. For Xpert, around 2–5 ml of sample was taken. In case of cerebrospinal fluid (CSF), the test was run with at least of 1 ml of sample. The samples were mixed with double amount of buffered solution. The sample along with the buffer solution was made to stand for 15 min with intermittent gentle shaking in between for effective mixing. The tests were then run on CBNAAT machines as per the manufacturer's user manual.
A 100 μl of the buffer sample deposits (after centrifugation at 3000 g for 20 min) was sent for LPA (Genotype MTBDRplus, V2 of Hains Life Sciences to see the concordance).
Gastric lavage showed a major detection in children after sputum as a biological specimen contributing around 8.35% of the total positive detection. The contribution to RIF resistance was 11.1% as seen in other studies. Data were captured on Microsoft Excel worksheets, and age- and sample-wise propensity results were expressed in terms of percentage. A Student's t-test was done between the paired data of the quarterly point; the prevalence showing P < 0.05 was obtained. Odds ratio between the smear results and CBNAAT positivity was calculated out using MedCalc Software (https://www.medcalc.org/calc/odds_ratio.php).
| Results|| |
It was observed that after sputum, gastric lavage contributed mostly in the pediatric case detection.
Their contribution toward RIF resistance varied from 12% to 16% among the positives; CSF contributed to 3% of the pediatric samples.
Resistance percentage ranged from 8% to 9% among the positives [Table 1].
|Table 1: Percentage contribution of the individual samples to the entire load of samples received (n=14,088)|
Click here to view
Prevalence and its significance
A Student's t-test was done between the quarterly point prevalence showing significance at 95% confidence interval (CI) (P< 0.05), and the point prevalence of MTB detection in the pediatric age group varied from 6% to 7% quarter-to-quarter period, whereas the prevalence of MTB detection was 5%.
Point and periodic prevalence of RIF-resistant MTB among the population tested varied from 0.2% to 0.5%. Point prevalence of RIF resistance MTB among the detected found to range between 5% and 15% with a seasonal predominance during winter months.
With the sample size of 14,088, the MTB detected positive by CBNAAT was 979 (6.9%). Proportion of positive results was found to be 0.069 with the lower bound of 0.065 and the upper bound of 0.071 at 95% CI.
The proportion of RIF resistance among the positives ranged from 0.110 to 0.153 at 95% CI.
Sample- and age-wise detection
Among the sample types, the gastric lavage contributed the most after sputum. The proportion positivity varied from one sample to other. As different samples came in different numbers, it was not equitable to comment on the percentage positivity. Similar number of different samples would have helped in achieving significant data [Table 2].
Detection was poor in all most all the biological specimens in the age group below 6 months; however, the gastric aspirate and gastric lavage helped in significant case detection levels after sputum.
The percentage detection was maximum in the age group between 1 and 5 years and >1 year (29.51% and 57.27%, respectively), followed by the age group <1 year (10.42%) [Table 3].
|Table 3: The age-wise contribution toward positive case detection by cartridge-based nucleic acid amplification tests|
Click here to view
The overall percentage detection was minimum in the age group below 6 months (2.78%).
Gastric lavage contributed to 8.35% of the total positive detection.
CSF contributed to almost 2.37% of the entire positives detected, with around 13.79% RIF resistance among the CBNAAT-CSF positives.
The MTB cases from gastric aspirate/ Lavage were detected mainly in the overcrowded area. Some sporadic cases of MDR TB were also found which is epidemiologically a threat to the community.
Smear microscopy and cartridge-based nucleic acid amplification test positivity
On calculating odds ratio between the smear results and CBNAAT positivity, it was found that an odds ratio of >1 with CI 95% with P < 0.0001 shows higher detection of MTB through CBNAAT in smear-positive cases than negative cases; however, the analytical sensitivity of CBNAAT and smear microscopy is 100 colony-forming unit (CFU)/ml and 5000–10,000 CFU/ml, respectively.
Based on the higher sensitivity and high-positive predictive value of CBNAAT, the positivity among the smear negative showed a substantial percentage. Among the smear Negative pulmonary samples it has been observed that 48 percent cases found to be positive by CBNAAT(15) and our study for gastric lavage/aspirate reflects about 6% through CBNAAT which were initially Smear Negative.
| Discussion|| |
Among the top ten causes of death worldwide, TB accounts for one of the most important factors; approximately, 1.2 million children became ill and 0.17 million died with it. Five hundred children deaths due to TB and over three quarters of a million children fall ill with TB each year.
Clinical diagnosis being nonspecific and considerable interreader variation in radiological findings prompts for a diagnostic technology of a high-positive predictive value.
Microbiological confirmations would help in giving a clearer picture for diagnosis and prognosis of disease in the course of treatment.
The Intermediate Reference Laboratory at Kolkata under State TB Demonstration and Training Center (Government of West Bengal, India) has a molecular biology setup that runs LPA for antitubercular drugs as well as CBNAATs. The laboratory is well managed with internal quality control (QC) and external QC done by the National Reference Laboratories on behalf of Central TB Division, Government of India.
A total of 14,088 numbers of pediatric specimens were tested in a span of 4 years (2014–2018) coming from all the districts of West Bengal, India.
The specimens comprised several samples that included pulmonary as well as extrapulmonary samples. Pulmonary samples comprised sputum, induced sputum, and bronchoalveolar lavage, whereas extrapulmonary samples comprised gastric lavage, ascitic fluid, plural fluid, pericardial fluid, lymph nodes (axillary and cervical), CSF, synovial fluids, and tissues.
The study undertook multidimensional cohort analysis of the increased case detection rates with regard to the referrals, nature of the samples, and age group contribution to the positivity (whether indicative of any age-wise pattern). The retrieval rates in paucibacillary samples and the occurrence of RIF-resistant MTBC strains among these samples were looked for. The strains which were found resistant by CBNAAT were subjected to LPA by MTBDRplus and V2 Hain for concordance analysis. Clinical manifestation correlating to an affirmative detection was also a component of the study smear-related correlation and their outcomes were analyzed.
Sample- and age-wise positivity
Gastric lavage showed a maximum detection after sputum. It is suggested that for pediatric population for <1 year of age, collection of other biological specimens such as gastric aspirate, gastric lavage, and induced sputum increases the yield where sputum collection is difficult.
Among the age group between 6 months and 1 year, the gastric lavage showed a detection rate of almost 16% after sputum which showed an MTB detection rate of 46.6%. Detection was mostly found in the age group between 6 months and 1 year.
Although most of the samples that frequented were sputum, CSF from inpatient formed a major portion (8.6%) of the total samples tested. Detection was high in CSF among the age group above 6 months. The percentage of RIF-resistant strain detected within this group varied from 8% to 7%.
The study also tried to find out the possibility of predominance of pediatric TB in a particular specific geographical location.
The detection of Mycobacterium tuberculosis from gastric aspirate/lavage were mainly from the overcrowded areas. Sporadic cases of MDR TB were also detected reflecting an epidemiologically a threat to the community.
Most of the pediatric TB cases were referred from the pediatric hospitals. TB units, peripheral health institutions, and medical colleges together contributed only 10%–12% of the total intake. The majority of the samples came from the government pediatric hospital. A substantial number of CSF (8.68%) came from inpatient department of these hospitals.
Private practice involvement was found to be significant. Gastric lavage and gastric aspirate from the pediatric hospital augmented the detection rates. The scope of widening the pediatric TB care support by increasing the number of skilled collection center across the state will help in further increase in the detection rates.
However, the MDR detection cases during the study duration were dispersed throughout and not confined to the urban slum area.
Smear and cartridge-based nucleic acid amplification test positivity
The odds ratio corroborates with the higher analytical sensitivity of CBNAAT. The study showed that CBNAAT was able to capture 48% of smear-negative samples.
Prevalence and its significance
The correlation between the quarterly prevalence rates was found to be significant using Student's t-test.
At 95% CI (P< 0.05), the point prevalence of MTB detection in the pediatric age group varied from 3% to 5% quarter-to-quarter period, whereas the prevalence of MTB detection was found to be 3.88%.
Point and periodic prevalence of RIF-resistant MTB among the population tested varied from 0.2% to 0.5%. Point prevalence of RIF resistance MTB among the detected found to vary between 5% and 15% with a seasonal predominance during summer months. This corroborated with the earlier findings.
Five percent prevalence and around 4.9% incidence is an alarming situation in the TB control scenario of West Bengal. This statistics is an alarming threat to the community and a rationality to be kept for implementation of universal drug susceptibility testing.
With the sample size of 14,088, the MTB detected positive by CBNAAT was 537 (3.8%) [Table 4]. Proportion of positive results was found to be 0.038 with the lower bound of 0.035 and the upper bound of 0.041 at 95% CI.
|Table 4: Sample type contribution toward Mycobacterium tuberculosis detection and rifampicin resistance detection|
Click here to view
The proportion of RIF resistance among the positives ranged from 0.116 to 0.146 at 95% CI.
Clinical correlation showed cough and fever as the most correlated manifestations.
Cough predominated mainly on the higher age group.
Cartridge-based nucleic acid amplification test and line probe assay concordance
The confirmed RIF-resistant MTB through CBNAAT was subjected twice in different batches for LPA (MTBDRplus and V2 Hains) to further validate and establish the concordance with CBNAAT.
Ninety percent result showed concordance with LPA. However, 5% of cases were sensitive to RIF through LPA and 5% were found negative for MTB on repeated test on LPA. The established discordance of two molecular techniques showed varied specificity and sensitivity.
It has been observed that sputum collection has been an ordeal for pediatric population showing symptoms of pulmonary TB. The study showed substantial percentage detection of pediatric TB in pulmonary samples other than sputum. The group consisted of bronchoalveolar lavage, bronchial brushings, tracheal aspirates, and induced sputum. It was found that gastric lavage formed the majority of the cases from which TB was detected after sputum. This was observed from the percentage positivity of the different samples.
The odds ratio shows that chances of getting CBNAAT detection are not dependent of smear positivity, around 45.5% of smear negative turned out to be CBNAAT positive. However, most of the samples carried either low or very low-bacillary load. This corroborates with the analytical sensitivity studies on CBNAAT as compared to the smear microscopy. However, it was found that in earlier studies, the percentage of samples were nontuberculous mycobacterium (NTM) which became CBNAAT negative even after smear positivity. The extensive study did not show a single case being smear positive while CBNAAT negative.
| Conclusions|| |
The study showed the paramount importance of samples other than sputum, in detecting pulmonary TB in the age group between 6 months and 1 year. Gastric lavage and gastric aspirate contributed to 15% and 16%, respectively, in the pediatric case detection rates. Substantial numbers when translated to absolute figures remain undiagnosed due to dearth of appropriate sample collection.
Point prevalence of 6%–7% pediatric TB is seen in West Bengal, India. More units for sample collection would be of great help augmenting the case detection further. Detection from the sputum was more in moderate age groups (between 6 months and 1 year); however, the gastric lavage contributed significantly in the age group below 6 months. Around 40% cases which were missed by fluorescent microscopy using standard national guidelines were detected by nucleic acid amplification tests.
The study show MTB detection 3% among the CSF in West Bengal which stands as a public health challenge. CBNAAT was able to capture several of smear-negative paucibacillary cases. Odds ratio >1 depicted that the chances of getting a CBNAAT positive among the smear-positive cases were >90%; however, the percentage of NTM among the missed 10% could be done with additional resources.
Gastric lavage is an important sample and accounts for substantial yield in the pediatric age group of below 6 years.
We are thankful to all staff at Intermediate Reference Laboratory, West Bengal, India, for providing technical inputs and all patients suffering and the staffs contributing for their dedication.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Mukherjee A, Chowdhury R, Singla R, Saha I, Dutta R, Das T, et al.
Comparison between childhood and adult tuberculosis in a rural tuberculosis unit of West Bengal: A retrospective study. Lung India 2014;31:116-20.
] [Full text]
Chakravorty S, Sen MK, Tyagi JS. Diagnosis of extrapulmonary tuberculosis by smear, culture, and PCR using universal sample processing technology. J Clin Microbiol 2005;43:4357-62.
Raizada N, Sachdeva KS, Swaminathan S, Kulsange S, Khaparde SD, Nair SA, et al.
Piloting upfront Xpert MTB/RIF testing on various specimens under programmatic conditions for diagnosis of TB and DR-TB in paediatric population. PLoS One 2015;10:e0140375.
Rufai SB, Kumar P, Singh A, Prajapati S, Balooni V, Singh S, et al.
Comparison of Xpert MTB/RIF with line probe assay for detection of rifampin-monoresistant Mycobacterium tuberculosis.
JClin Microbiol 2014;52:1846-52.
Desikan P. Sputum smear microscopy in tuberculosis: Is it still relevant? Indian J Med Res 2013;137:442-4.
] [Full text]
Iravane JA, fatema SA, Damle Ajit S. GeneXpert for gastric lavage to diagnose pulmonary tuberculosis in children. Int J Microbiol Res 2017;9:14.
Helb D, Jones M, Story E, Boehme C, Wallace E, Ho K, et al.
Rapid detection of mycobacterium tuberculosis and rifampin resistance by use of on-demand, near-patient technology. J Clin Microbiol 2010;48:229-37.
Bodmer T, Ströhle A. Diagnosing pulmonary tuberculosis with the Xpert MTB/RIF test. J Vis Exp 2012; p. 3547.
Lombardi G, Di Gregori V, Girometti N, Tadolini M, Bisognin F, Dal Monte P, et al.
Diagnosis of smear-negative tuberculosis is greatly improved by Xpert MTB/RIF. PLoS One 2017;12:e0176186.
World Health Organization. Global Tuberculosis Report. World Health Organization; 2016.
World Health Organization. Roadmap for Childhood Tuberculosis: Towards Zero Deaths. Geneva: World Health Organization; 2013.
Hesseling AC, Schaaf HS, Gie RP, Starke JR, Beyers N. A critical review of diagnostic approaches used in the diagnosis of childhood tuberculosis. Int J Tuberc Lung Dis 2002;6:1038-45.
Zar HJ, Hanslo D, Apolles P, Swingler G, Hussey G. Induced sputum versus gastric lavage for microbiological confirmation of pulmonary tuberculosis in infants and young children: A prospective study. Lancet 2005;365:130-4.
Das PK, Ganguly SB, Mandal B, Khan A. Prevalence of rifampicin-resistant pediatric tuberculosis by cartridge-based nucleic acid amplification test at the intermediate reference laboratory under revised national tuberculosis control program India: A multidimensional approach. Biomed Biotechnol Res J 2018;2:300-5. [Full text]
[Table 1], [Table 2], [Table 3], [Table 4]