Introduction

Despite recent advances in rapid diagnostics, smear microscopy remains the most widely used test in low-income high tuberculosis (TB) burden countries [1]. The sputum smear-positive case detection rates globally and in India have been stagnant at a level below the 70% target set by the World Health Organization (WHO) [2]. Hence, efforts to improve the performance of sputum smear microscopy are a high priority for global TB control [3]. One reason for poor case detection could be the suboptimal use of sputum smear microscopy which currently involves the examination of two specimens collected over two days. A substantial proportion of patients fail to complete the two-day evaluation, especially in medical college hospitals and district level health facilities, owing to costs and inconvenience involved in multiple visits. If this process could be front loaded with both samples collected on the same day, it may reduce the number of patient drop-outs during diagnostic evaluation and increase the number of patients diagnosed and initiated on treatment [4,5]

The WHO recommends that countries that have successfully implemented the policy for a two-specimen case-finding strategy should consider switching to same-day diagnosis, especially in settings where patients are likely to be lost during the diagnostic process [6]. While sputum smear microscopy has been criticized for its low sensitivity relative to culture, the recent WHO recommendation of shifting to a two-smear strategy and same-day microscopy are likely to reduce sensitivity even further. Hence, countries have to test this strategy in their settings before adopting it. Most of the studies conducted hitherto have been under trial conditions and there is a lack of evidence from routine programme settings [7–10] and limited, conflicting information from India. A study from South India showed that same day microscopy was as effective as the conventional strategy [11]; whereas another study from North India showed that same day microscopy missed a considerable number of smear positive TB patients [12]. Both the studies were conducted in tertiary care hospitals with a limited sample size and were not representative of routine programme conditions. Hence, we aimed to assess if same-day (spot-spot) microscopy is as effective as the conventional (spot-morning) strategy in detecting sputum smear positive TB under programmatic settings in Chhattisgarh state, India. The specific objectives were to compare between same-day (spot-spot) microscopy and the conventional (spot-morning) strategy, (i) the number (proportion) of presumptive TB patients (erstwhile referred to as ‘TB suspects’) who provided two sputum specimens, (ii) the quality of specimens, (iii) the number (proportion) of sputum smear positive TB cases diagnosed, and (iv) the incremental yield of the second specimen.

Methods

Data collection and data validation

The process is shown in Figure 1. Consecutive presumptive TB patients attending DMCs were provided information on the study and requested to provide two ‘spot’ sputum samples one hour apart. The patient was requested to return the next day with an early morning sample. Specimens were assessed macroscopically for quality and smears were stained using the hot Ziehl-Neelsen (ZN) technique. Examination of the second spot sample was performed in a blinded manner by the LT. The senior tuberculosis laboratory supervisor (STLS), who was responsible for the blinding, labeled the second spot specimen with a different number before examination by the LT. The original number, new number and results were recorded in a study register. Smears were considered positive if it had >1 acid fast bacilli per 100 oil immersion fields examined. All smears were graded according to the WHO/International Union Against Tuberculosis and Lung Disease scale [13].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Data collection procedure during the study, Chhattisgarh, India, October 2012-March 2013.

https://doi.org/10.1371/journal.pone.0074964.g001

Results

Of 2551 presumptive TB patients enrolled, 1749 (69%) were male and the mean (SD) age was 45 (21) years.

Table 1 provides the comparison between the two approaches. All participants provided the first spot specimen, 2361 (93%) provided the second spot specimen, and 2435 (95%) provided an early morning specimen. 1833 (72%) of the participants provided mucopurulent specimens in conventional microscopy as compared to 1540 (60%) in same day microscopy. The proportion of smear positive patients diagnosed by same day microscopy was 360 (14%), as compared to 431 (17%) by the conventional method (p<0.001). The sputum positivity did not vary by centre.

The incremental yield of the second sputum specimen (instances when patients were negative on first specimen and positive only on second) was found to be 15 (4%) in same day microscopy as compared to 86 (20%) in conventional microscopy.

Considering the results of all three specimens, there were a total of 433 smear positive TB cases. Of this, 73 (17%) were missed by same day microscopy compared to 2 (1%) by the conventional method (Table 2).

Discussion

This is the first study from India comparing same day microscopy with the conventional technique in routine program settings. Same day microscopy missed 17% of smear positive cases as compared to the conventional technique, thus the WHO recommendation of a same day sputum microscopy strategy may not be applicable in our setting.

The basis of the WHO recommendation comes from a meta-analysis and several studies which showed same day sputum microscopy has similar diagnostic accuracy as compared to the conventional technique [7–10]. While we are not sure of the reason for the difference in our findings, it might be related to the differences in patient characteristics and the spectrum of severity of TB in populations, which could be assessed indirectly using the distribution of smear quantification among smear positive TB patients [14]. In settings with a high TB burden, advanced disease and delayed access to care, most of the smear positive TB patients are likely to be ‘high positives’ (graded 2+ or 3+), and will be detected in a spot specimen. While early morning sputum specimens are known to more sensitive than spot specimens, the advantage is mitigated if most of the patients have advanced disease and are high positives [15].

Not all studies that have demonstrated equivalence between the two approaches have reported on the distribution of smear quantification [10]. In one study from Ethiopia which showed equivalence, none of the patients had scanty smears [16]. In contrast, nearly 50% of all smear positives in our study were low positives (either scanty or 1+) and hence the advantage of early morning specimen over second spot specimen is obvious. This may further indicate good access to services and patients reaching health facilities at an early stage in the state of Chhattisgarh. Introduction of the same day sputum microscopy may not be beneficial in such settings. Second, the higher yield of the early morning specimen may also be related to better quality of sputum specimens as compared to spot specimens. Third, most of the studies showing equivalent performance of same-day approach were done in well-controlled research settings with smear and culture examinations performed by well-trained personnel in reference laboratories [7–10]. In research settings, technicians are likely to spend more time reading the smears than in routine laboratories and hence are more likely to compensate for the difference between spot and overnight specimens; whereas, we conducted the study in routine settings and thus reflect operational realities.

The incremental yield of the second specimen was 4% in same day microscopy compared to 20% in the conventional method. Therefore, in situations where an early morning specimen cannot be obtained for any reason, a second spot will give at least some additional yield in sputum positivity and should be performed.

A surprising and counter-intuitive finding of our study was that the patient drop-out during diagnostic process was marginally higher with same-day approach as compared to conventional approach. The reason for this is not known and may require qualitative studies to assess why patients preferred to come on the morning of the second day to give their second sputum specimen instead of waiting for an hour on the same day. Inconvenience involved in waiting in over-crowded patient unfriendly health facilities and possible nosocomial exposure may have deterred the patients from waiting.

Since 2012, light emitting diode fluorescence microscopy (LED-FM) has been introduced in about 200 medical college microscopy centres in India. Given the better sensitivity (10% more) of LED-FM over ZN microscopy, it may be possible that LED-FM in spot specimens performs better in detecting acid-fast bacilli and the difference in performance between spot and early morning specimen may be lower [17]. Hence, it is possible that the same day approach could be as effective as conventional microscopy when LED-FM is used. This is a topic for future research.

There were a few limitations to our study. First, we conducted the study in seven purposively selected district hospitals and hence there may be an issue of generalizability of the findings. We deliberately chose busy district hospitals since we expected a higher chance of patient drop-out during the diagnostic process in these settings. If the same day approach did not confer any advantage in these settings, it is unlikely that it would be beneficial in more peripheral health institutions where most people are likely to provide two specimens. Second, all participants were provided detailed information including the need to provide second spot and early morning specimens during the process of informed consent. This might have influenced them to adhere to the diagnostic process and may not reflect the drop-outs that would have happened in routine conditions. Third, given the design of the study, we could not assess the effect of the different approaches on treatment uptake. Fourth, we did not have any information on HIV status of the participants or the proportion of TB patients with cavitary lesions and if that impacted the results. However, we feel that this might not have impacted greatly as the study was conducted in a low HIV prevalence setting (with an estimated HIV prevalence of 0.22% among adults aged 15-49 years in the year 2011) [18] and most of smear positive TB patients had their smears graded ‘low positive’ providing an indirect indication of likelihood of cavitary lesions.

Conclusion

Same day microscopy missed 17% of smear-positive cases when compared to the conventional strategy. In contrast to prior studies, this strategy did not increase the proportion of suspects providing the second sample. The study findings call for a need to revisit the WHO recommendation of switching to same-day diagnosis. Countries should be cautious in their approach, pilot test the same day microscopy strategy and ascertain its usefulness before making decisions on a wider scale-up.

Acknowledgments

This research was supported through an operational research course which was jointly developed and run by the International Union Against Tuberculosis and Lung Disease, South-East Asia Regional Office, New Delhi, India, National Institute for Research in Tuberculosis (NIRT), Chennai, India and Desmond Tutu TB Centre, Stellenbosch University, Cape Town, South Africa. The authors are thankful to the State TB Cell, District TB Officers and their staff of the seven districts of Chhattisgarh for enthusiastically conducting the study. Necessary administrative support to attend the course was extended by the World Health Organization (India Country Office). Special acknowledgment to STDC staff, Chhattisgarh for providing their valuable support in double data entry and validation.