Abstract
Objectives: To determine the ideal number of sputum acid-fast bacilli (AFB) smears and polymerase chain reaction (PCR) required for discontinuing tuberculosis (TB) isolation among patients with suspected pulmonary TB.
Methods: This was a single-center, record-based retrospective study of all admitted patients diagnosed with culture-proven pulmonary TB between 2010 and 2018. The study was conducted at Prince Sultan Military Medical City (PSMMC) in Riyadh, Saudi Arabia, a large tertiary care center consisting of 1,200 beds. Data were obtained from our TB notification records. Patients with smear-positive TB were investigated. Only the first 3 sputum smears for AFB were included in the analysis. The PCR results for Mycobacterium tuberculosis (MTB) were also included in the study. The incremental yield of the second and third smears was assessed.
Results: Overall, 240 patients were MTB-culture positive. A total of 126 (52.5%) patients were smear and culture positive, whereas 114 were culture positive but smear negative. Of 126 patients who were AFB smear positive, 98 (77.8%) were detected in the first specimen, 13 (10.3%) in the second specimen, and only 9 (7.1%) in the third specimen. Polymerase chain reaction for MTB was positive in 122 (96.8%) smear-positive patients. Four patients did not undergo a PCR test.
Conclusion: A single Xpert MTB/resistance to rifampicin test detected all smear-positive patients, whereas the third smear did not significantly contribute to MTB isolation.
Tuberculosis (TB) infection remains a leading cause of morbidity and mortality worldwide, with an estimated 10.0 million people affected with TB disease in 2017, whereas an estimated 1.3 (range, 1.2-1.4) million deaths occurred among the human immunodeficiency virus (HIV)-negative population. An annual TB incidence rate of 10 (8.6-12)/100,000 population has been reported by the World Health Organization (WHO) in Saudi Arabia.1 The “WHO End TB” strategy stresses on the need to prevent TB infection in settings with high risk of transmission including healthcare facilities.2 Early suspicion and placing individuals with presumptive TB in an airborne infection isolation (AII), ideally using single-patient and negative-pressure ventilation rooms, are essential components of a hospital TB program. Furthermore, the current Center for Disease Control and Prevention (CDC) guidelines for the prevention of Mycobacterium tuberculosis (MTB) transmission in the healthcare setting recommended administrative controls that aim to reduce exposure to individuals with presumptive pulmonary TB and respiratory protection controls using disposable N95 respirators by healthcare workers.3 Three respiratory specimens collected 8 to 24 hours apart, with at least 1 early morning specimen, should be sent to the laboratory for acid-fast bacilli (AFB) microscopy and culture. Isolation can be discounted if 3 smears are negative for AFB. However, AII is costly, scarce, or unavailable in many hospitals. The Middle East respiratory coronavirus (MERS-CoV) outbreak in the Arabian Peninsula has increased the demand for such AII rooms in countries like Saudi Arabia. Moreover, a systematic review identified a negative impact of isolation on the patients’ mental health, whereas patients’ safety and satisfaction were also unfavorably affected. In addition, few studies also found that the contact time between healthcare workers and isolated patients decreased.4 Thus, measures for decreasing isolation time are needed. On the contrary, nosocomial outbreaks with sensitive, and more alarmingly, MDR TB have has been previously reported.5,6 Consequently, a balance should be achieved to prevent disease transmission in health facilities.
This study was conducted to determine the ideal number of sputum AFB smears and polymerase chain reaction (PCR) required for discontinuing TB isolation among patients with suspected pulmonary TB.
Methods
This is a single-center records based retrospective study of all patients diagnosed with culture-proven pulmonary TB between 2010 and 2018. The TB notification records were utilized to identify patients admitted with PTB. The patients’ charts were reviewed by the authors for demographic data, sputum AFB smears, and polymerase chain reaction (PCR) and culture results. Only the first 3 sputum smears were included in the analysis. Polymerase chain reaction results for MTB were also included in the study.
The study was conducted at Prince Sultan Military Medical City (PSMMC) in Riyadh, Saudi Arabia, a large tertiary care center consisting of 1,200 beds. All patients attending the emergency department with suspected clinical and/or radiological features of pulmonary TB are immediately triaged to a negative pressure isolation room. Three sputum specimens are collected every 8 hours with one specimen collected early morning. Among patients who fail to expectorate, sputum induction is performed with 3% hypertonic saline nebulized in a negative pressure room. Occasionally, patients may require bronchoscopy to obtain appropriate specimen. In addition to routine sputum microscopy and culture, GeneXpert can be requested on all first specimens. Patients with smear-positive TB are kept in isolation, whereas individuals with alternative diagnosis or 3 negative smears are discharged.
We included all adults with at least one sputum culture positive for MTB. Patients with negative culture or non-tuberculous mycobacteria (NTM) were excluded. All patients were assessed according to the first positive smears and PCR results.
In the microbiology laboratory, all samples are cultured on solid media (Lowenstein-Jensen) and liquid media Mycobacterium growth indicator tubes (MGITs) and a smear was prepared for AFB detection. Cultures were considered negative if no growth was observed after 8 weeks of incubation. Smears that were prepared from concentrated samples were heat fixed and stained with Auramine and read under the fluorescent microscope for the presence of AFB. Positive smears were confirmed by over staining with ZN stain. Rapid detection of MTB complex DNA in the sputum or concentrated sediments that were either AFB smear positive or negative was performed by PCR test using GeneXpert RT-PCR MTB/RI.
Statistical analysis
Statistical analyses were performed using SPSS 20.0 (SPSS Inc., Chicago, IL, USA). Continuous data were expressed as mean±standard deviation (minimum - maximum) while categorized data were expressed as number (%). Cross tabulation was made between first and second sample results and between third and combined first and second samples. Cross tabulation was also made between PCR and AFB smears results. Categorical variables were compared using likelihood ratio. All p-values were 2-tailed, and p≤0.05 was considered statistically significant. The study was conducted after approval from the Research Ethical Committee of the Prince Sultan Military Medical City, Riyadh, Saudi Arabia.
Results
A total of 240 patients were MTB culture positive. The mean+SD age of the patients was 51.2±22.5 years. A majority of the patients were men (n=155, 64.6%). Only 7 (2.9%) of patients were HIV positive; however, 42 (17.5%) patients were not tested for HIV (Table 1).
Demographic and clinical characteristics of the patients with pulmonary tuberculosis (N=240).
A confirmed positive culture of MTBC was used as a reference standard. A total of 126 patients were smear and culture positive, whereas 114 were culture positive but smear negative. Polymerase chain reaction was performed in 83% of the study group, which was positive in 152 of 199 (76.4%) tested patients (Table 1). The overall sensitivity of the AFB smear and PCR were 52.5% and 76.4% respectively (Table 2). Tables 3 & 4 shows cross tabulations between smears that showed positivity in the 1st or 2nd smear or both of them and results of the 3rd smear. Of 126 patients who were culture and AFB smear-positive, 98 (77.8%) were detected in the first specimen, 13 (10.3%) patients who were negative in the first specimen were positive in the second specimen. Only 9 (7.1%) were positive in the third specimen while negative in both first and second sample. The likelihood of positivity of the third specimen when both first and second smear were negative was 2.91 (LHR 2.9, p=0.23) (Tables 3 & 4). Six samples were negative for AFB beyond the third specimen (5 in fourth sample, one in the sixth sample); however, all of these patients were PCR positive based on the first sample.
Sensitivities of acid-fast bacilli (AFB) smear and polymerase chain reaction (PCR) using culture as gold standard.
Cross tabulation between 1st and 2nd acid-fast bacilli (AFB) smears.
Cross tabulation between 3rd acid-fast bacilli (AFB) smear and both 1st and 2nd smears.
Polymerase chain reaction was positive in 122 (96.8%) smear-positive patients (Table 5). Four patients did not undergo PCR; their smears were detected in the first sample, one from the third sample, and 2 from the fourth sample. The PCR was 100% sensitive in detecting smear positive patients when the test is performed. None of the smear positive group was PCR negative. In smear-negative patients, the PCR was positive in 30 (26.3%) of cases with significant difference between the groups (p≤0.001).
Cross tabulation between sputum AFB smear and PCR.
Discussion
The CDC guidelines recommend that all patients with suspected pulmonary TB have 3 negative sputum AFB smears before the discontinuation of isolation.3 Similarly, the European Centre for Disease Prevention and Control (ECDC) suggests a minimum of 2 sputum specimens subjected for AFB examination.7 This study was conducted to determine the number of smears and/or PCR to determine deisolation of suspected TB patients. Overall, the sensitivity of AFB smear in identification of TB in our study was 52.5% (95% confidence interval [CI] 39.8%-52.8%). The first 2 smears together detected 88% (111 out of 126) of patients, whereas only 7.1% of patients were detected using the third serial smear alone. The contribution of the third sample was not significant with a likelihood ratio for positivity of 2.9 (p=0.23) These findings are in agreement with previous reports. Percentages of patients identified as AFB smear positive in at least the third specimen alone widely varied (0%, 5.3%, 8.6%, 3.2%, and 11.1%).8 Studies performed in Colorado, USA, showed an incremental yield of the third specimen (2%), whereas in Tanzania, the increment was 2.5% in a cohort of 49,930 patients with suspected TB.9 Similarly, 42 laboratories from 4 high-income countries showed that an additional yield from a third serial smear was 0.7% to 7.2%. However, a large number of smears, 122.7 to 796.3, were needed to detect one extra case with the third serial smear, whereas 164.8 to 2133.4 slides had to be examined to detect an incremental failure with the second serial smear.10 In our cohort, 2 sputum smears would have missed 9 smear-positive patients over 9 years. However, the PCR would have prevented all these patients released to the general ward. Moreover, patients strongly suspected of PTB should be kept in isolation even with negative screening. Utilizing clinical decisions and/or PCR testing can help decrease immature deisolations.
Evidences that using the PCR can shorten the isolation time in PTB suspects have been growing.11-14 Nucleic acid amplification test (NAAT) can rapidly diagnose TB and detect resistance. Xpert®MTB/RIF assay endorsed by the WHO is a rapid, cartridge-based NAAT. The result can be delivered within 2 hour.15 In the case of respiratory specimens, the sensitivity and specificity of the Xpert® MTB/RIF assay were 100% (95% CI: 80-100%) and 100% (95% CI: 95%-100%).16 In our study the PCR detected all smear positive patients. Similar to previous studies a single PCR in our study was equivalent to 3 serial smears finding.14 The sensitivity and specificity of the PCR in the tested patients in this study was 100.0%. Interestingly when the PCR was performed, 30 patients (26.3%) of our smear negative patients turned to be sputum PCR positive. These patients could have transmitted the infection if released to the general ward. Of note, 10-20% of TB patients acquires their infection from smear and probably PCR negative patients; hence, deisolation should be guided by the clinical probability of the disease.17-19
All of our smear positive patients were detected with a single PCR examination. It is not yet clear whether a single expert or 2 examinations should guide discontinuation of AII. Single NAA test can decrease the demand for negative pressure rooms by approximately 75% when compared to serial sputum AFB smears.12 The sensitivity of 1 Xpert, 2 Xperts, 2 smears, or 3 smears compared to culture was 0.85 (95% CI: 0.61-0.96), 0.95 (95% CI: 0.73-1.0), 0.70 (95% CI: 0.46-0.88), and 0.80 (95% CI: 0.56-0.93), respectively. The NPV of the 4 strategies was also similar (0.99 for 3 smears, 0.98 for 2 smears, 0.99 for 1 Xpert, 1.00 for 2 Xperts. The FDA has recently approved testing of 1 or 2 sputum PCRs as an alternative to serial AFB smears for discontinuation of airborne precautions.20
Despite current CDC recommendations that nucleic acid amplification tests (NAATs) be performed on at least one respiratory specimen from all patients evaluated for TB, molecular testing for TB in the United States is underutilized, due to lack of awareness or due to costs and logistical considerations.3 Remarkably and following increasing awareness, our current PCR request rate in suspected PTB improved to 83%.
We believe our findings support studies endorsing a single PCR to shorten the isolation period in suspect pulmonary TB. The third serial smear specimen had a low incremental yield and would lead to delay in removing patients from AII rooms. This is associated with economic burden and can contribute to patients’ stigmatization and adversely affect their outcome. Polymerase chain reaction findings and critical clinical decisions will guard against immature deisolation in these patients. Our data did not allow for calculation of the days saved and the economic impact of NAA. However, a recent review showed molecular testing to be associated with shorter median isolation period (2.9 versus 2.5 days), time to hospital discharge (6.0 versus 4.9 days) and lower costing with savings of $13347 per isolated TB negative patients.21
Study limitations
It is a retrospective, single center study. In the cohort analyzed, the number of patients with HIV co-infection was very low that the outcome may not be generalized to areas where the prevalence of HIV/TB co-infection is high. The exact timing of the sample collection and release of patients from isolation could not be identified in all patients prohibiting calculation of time and economic saving. One strength of our study is the high PCR request rate (83%). Furthermore, NAAT performed on our smear negative patients detected 30 (26.3%) of culture positive, smear negative patients. This can contribute to the diagnosis and help in the isolation decision.
In conclusion, a single PCR examination helps discontinue negative pressure room isolation while a third sputum smear did not contribute significantly to the deisolation decision in patients suspected of pulmonary TB.
Acknowledgment
We extend our thanks to our TB co-coordinators, Moaidh Alzahrani and Saud Aldalbahi, for their help in the study and to Professor Enas Hamed for the statistical analysis. We would like to thank Editage (www.editage.com) for English language editing
Footnotes
Disclosure. Authors have no conflict of interests, and the work was not supported or funded by any drug company. Dr. Ali M. Albarrak is a member of the Editorial Team, and was therefore excluded from any final editorial decisions regarding this paper.
- Received May 21, 2019.
- Accepted September 15, 2019.
- Copyright: © Saudi Medical Journal
This is an open-access article distributed under the terms of the Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.