Correlation of Timika scoring with microbiological burden in multi-drug resistant tuberculosis patients initiated on bedaquiline treatment

Abstract

Keywords

Bacterial infection tuberculosis diagnosis public health Asia

Introduction

Tuberculosis (TB) is a public health concern of escalating importance. Mycobacterium tuberculosis (MTB) is an obligate pathogen of the mammalian respiratory system, and is the primary causative organism and most recognised member of the M. tuberculosis complex, infecting more than one-third of the world's human population. Globally, TB accounts for more deaths than any single infectious agent despite effective treatment and promising new developments easing diagnosis.¹ According to the global TB report 2024, an estimated total of 10.8 million people suffered from TB in 2023.² The worldwide spread of multidrug-resistant tuberculosis (MDR-TB) has added new dimensions to already existing problems. So, timely detection and optimal treatment at the earliest moment remain of utmost importance. The initial approach for the diagnosis of pulmonary TB (PTB) includes sputum smear examination, chest radiography and identification of typical clinical features such as low-grade fever, cough, haemoptysis, weight loss, anorexia and night sweats, all typically suggestive of TB.³

Imaging plays a crucial role in the diagnosis and follow-up of PTB patients. Chest radiography is accessible, cost-effective, and valuable for the presence and severity of the condition, and to monitor treatment response during therapy.⁴ The Timika chest radiography score is a simple system to grade the radiographic severity of the disease extent on CXR. It helps to stratify patients into high-risk groups in terms of the treatment outcome.⁵ It consists of two components: first, the percentage of lungs affected on both sides due to any TB-related lesions; and second, the presence of cavitation.^4,5 Understanding the correlation of the Timika score with the microbiological burden and disease severity provides a clearer picture of the course of the disease among newly diagnosed cases of MDR PTB patients initiated on bedaquiline treatment. The Timika score makes it easier to use a baseline chest radiograph to predict early infectiousness and exhibits performance comparable to that of more comprehensive and standardised information of the radiologist. We propose that this score could help physicians worldwide to identify high-risk TB patients in terms of treatment prognosis, especially in settings that lack resources.⁶

Material and methods

Ours was a one-year prospective observational study conducted in the University College of Medical Sciences (UCMS) and Guru Teg Bahadur (GTB) Hospital, Delhi. Thirty-four newly diagnosed adults (≥18 years) with MDR PTB by universal drug susceptibility testing, followed at DOTS Centre, were recruited into the study. Patients excluded were those transferred out, relocated, died or who had failed to collect specimens during follow-up. Informed written consent was obtained from all the participants before conducting the study. The study commenced after clearance from the Institutional Ethics Committee-Human Research (IEC-HR).

At the time of enrolment, clinical signs and symptoms were evaluated through a pre-designed questionnaire. The presence of risk factors for sputum conversion (smoking, alcohol intake, co-existent diabetes, tobacco consumption, substance abuse) was also noted. Patients were initiated on bedaquiline treatment, before which two early morning baseline sputum samples were collected for microscopic grading and culture.

Sputum specimens were homogenised and decontaminated by adding an equal volume of N-acetyl L-cysteine (NALC)–4% sodium hydroxide (NaOH) solution. After 20 min, the mixture was neutralised with phosphate-buffered saline and centrifuged. The sediment obtained was re-suspended in 1 mL sterile phosphate buffer and used for smear preparation, then subjected to Ziehl-Neelsen (ZN) staining and culture on the commercially available solid Lowenstein-Jensen media.⁷ Subsequent sputum samples were collected at an interval of four weeks for up to six months. Sputum smear grading on ZN stain was reported according to NTEP guidelines. Time to culture positivity was recorded. Sputum conversion was noted in the smear and culture during this period.

Baseline full-size postero-anterior chest radiographs with proper position and penetration were taken from all patients for the assessment of the Timika score in consultation with a radiologist. Each image was independently evaluated by two radiologists, who were blinded to clinical information to avoid bias. They visually estimated the findings and recorded them using a standardised form. The extent of lung involvement was assessed using the radiological Timika score system.⁶ The cavitatory disease was defined radiographically as a gas-containing lucent space within the lung parenchyma, surrounded by a fibrotic wall with thickness >1 mm.⁸

The Timika score is calculated as depicted in Figure 1: ^6,8

For the first component of the scoring system, the chest radiograph is divided into six zones of approximately similar size, with two horizontal lines designating the upper, middle and lower zones for each lung.

For each zone, the percentage area that showed active disease (consolidation, nodules) involvement is estimated depending on the visual estimation of the extent of opacification (5 or 10–100% in 10% increments).

The percentage area of all six zones showing active disease is added and divided by 600 to get the total percentage of lung affected (0 to 100 points).

The second component is the presence of at least one cavity; a constant value of 40 is added to the above value to obtain the final score.

Assessment reliability, using the intraclass correlation coefficient was 0.75, indicating good agreement among assessors based on established thresholds for Timika scoring assessment.

Definitions used were: (a) Time to culture positivity: The duration between the inoculation of a clinical specimen onto a culture medium and the visible mycobacterial growth. (b) Sputum conversion: The interval between the date of MDR-TB treatment initiation and the date of the first of the two negative consecutive cultures taken at least 30 days apart. (c) Non-conversion: Patients with positive cultures until the sixth month as not having converted.⁹

Scores therefore ranged from 0 to 140: the higher the score, the more severe the condition of the patient.⁴

Data analysis was performed using SPSS v20 (IBM Corp.). Descriptive statistics were elaborated in the form of means/standard deviations and medians/IQRs for continuous variables. Normality was assessed using the Shapiro–Wilk test. Non-normally distributed data were analysed using the Wilcoxon and Kruskal–Wallis tests. Categorical variables were summarised as frequencies and percentages. Correlations between continuous variables were examined using Pearson's correlation for normally distributed data and Spearman's correlation otherwise. Statistical significance was set at p < 0.05. Diagnostic capability/ accuracy of Timika scoring for culture conversion and non-conversion was analysed by ROC (Receiver Operating Characteristic curve) analysis.

Results

The distribution of the study population in terms of the Timika score (Fig. 2) uses a histogram (Bar chart): The x-axis represents the Timika scores, divided into bins (e.g. 0–10, 10–20). The y-axis represents the count of individuals per bin. The Timika score ranged from 3.2 to 85, with a mean (SD) of 37.16 (25.28), a median (IQR) of 29.00 (18.22–59.15), and the variable Timika score was not normally distributed (Shapiro–Wilk test: p = 0.014). A majority of participants had Timika scores between 10 and 30, with the most frequent bin containing approximately nine participants.

Correlation between the mean time to culture positivity with various ranges of Timika score (Fig. 3) signifies that the specimens from patients who had higher Timika scores took a shorter time (in weeks) to grow on the culture media, with weekly excluded culture converted samples simultaneously.

Socio-demographic profile of the study population in terms of the Timika score is depicted in Table 1. It depicts the association of the Timika score with various socio-demographic patient parameters. It shows a significant association between the Timika score on one hand and sputum smear microscopy grading (p = 0.015) along with the presence of any risk factors (p = 0.036) on the other hand. The risk factors included were history of smoking, substance abuse, alcohol intake, and diabetes, among which the coexistance of diabetes was significantly associating with the Timika score (p = 0.021).

Association between sputum smear microscopy grades and cavitatory lesion on chest radiograph is shown in Table 2. The proportion of participants with cavitatory lesions appeared to be higher in the 3+ grade group (52.6%) compared to the 2+ group (15.4%), although a higher proportion of cavitatory lesions was observed in the 3 + group, the difference did not reach statistical significance (p = 0.064).

The association between the Timika score and sputum smear microscopy grades is shown in Table 3. There was a significant difference between the three groups in terms of the Timika score (χ² = 8.340, p = 0.015), with the median Timika score being the highest in the 3+ microscopy grade group. Strength of association (Kendall's Tau) = 0.41 (Medium effect size).

Table 1.

Socio-demographic correlation with Timika scoring in the study population (n = 34).

Parameters				Timika scoring	p-value
Sociodemographic finding	Age			Correlation coefficient (rho) = 0.21	0.241¹
	Gender		Male	39.32 ± 24.68	0.710²
	Gender		Female	35.83 ± 26.15	0.710²
	History of contact		Present	52.56 ± 32.35	0.224²
	History of contact		Absent	34.50 ± 23.54	0.224²
	Sputum smear microscopy grading***		1+	16.60 ± 18.95	0.015³
			2+	22.68 ± 13.17
			3+	49.23 ± 26.15
	Presence of any Risk factor***		Present	49.02 ± 22.72	0.036²
	Presence of any Risk factor***		Absent	29.82 ± 24.43	0.036²
	Risk factors	Diabetes***	Present	56.80 ± 21.59	0.021²
		Diabetes***	Absent	31.12 ± 23.49	0.021²
		History of substance abuse	Present	42.88 ± 21.74	0.481²
		History of substance abuse	Absent	36.17 ± 26.05	0.481²
		History of alcohol intake	Present	80.00 ± 0	0.126²
		History of alcohol intake	Absent	35.86 ± 24.49	0.126²
		History of smoking	Present	36.07 ± 27.37	0.903²
		History of smoking	Absent	37.27 ± 25.55	0.903²

Table 2.

Association between sputum smear microscopic grades and cavitary lesion on chest x-ray in the study population (n = 34).

Cavitary lesion on chest radiograph	Sputum smears microscopic grade				Fisher's exact test
Cavitary lesion on chest radiograph	1+	2+	3+	Total	χ2	p-value
Present	0 (0.0%)	2 (15.4%)	10 (52.6%)	12 (35.3%)	5.848	0.064
Absent	2 (100.0%)	11 (84.6%)	9 (47.4%)	22 (64.7%)
Total	2 (100.0%)	13 (100.0%)	19 (100.0%)	34 (100.0%)

Table 3.

Correlation between Timika score and sputum smear microscopy grades in the study population (n = 34).

Timika score	Sputum smear microscopy grading			Kruskal–Wallis Test
Timika score	1+	2+	3+	χ2	p-value
Mean (SD)	16.60 (18.95)	22.68 (13.17)	49.23 (26.15)	8.340	0.015
Median (IQR)	16.6 (9.9–23.3)	20 (18–23.3)	56.6 (27.3–72.45)
Range	3.2–30	5–46.6	5–85

A ROC curve analysis showing diagnostic performance of the Timika score in predicting culture conversion or non-conversion is shown in Figure 4. AUC for the Timika score was 0.781 (95% CI: 0.63–0.933), thus demonstrating fair diagnostic performance but no statistical significance (p = 0.200). At a cut-off of 56.6, it predicts culture non-conversion with a sensitivity of 100%, specificity of 75%, positive predictive value of 20.0% (3–56), negative predictive value of 100.0% (86–100) and a diagnostic accuracy of 76.5% (59–89). The cut-off and the diagnostic parameters found are not reliable, as the test is not statistically significant.

Discussion

TB diagnosis is primarily based on a patient's history, examination, chest radiography and smear microscopy. Sputum smear microscopy and culture are used as the basis for sputum conversion to assess the treatment response in PTB. According to some systematic reviews and meta-analyses, sputum examination has low sensitivity and moderate specificity to predict relapse or failure during treatment.¹⁰ So, it is important to know about infectiousness and the disease severity of patients at diagnosis, which ascertains the treatment result. Many surrogate markers help to estimate disease severity more precisely: high bacillary load on baseline sputum examination, extent of pulmonary involvement and cavitatory lesion on diagnostic radiograph. Imaging has a role in risk stratification, especially in patients not producing sputum or sputum-negative PTB at the time of diagnosis.¹¹

Among the risk factors identified among cases, such as coexistence of diabetes, history of substance abuse, alcohol intake, and smoking, only the coexistence of diabetes demonstrated a significant association with the Timika score.

A higher likelihood of elevated baseline SSG in patients with cavitatory lesions is logical. Timika score calculation with machine learning makes it easy for radiologists to implement the score.

Our findings indicate a clear association between higher imaging scores and higher baseline SSG, suggesting that more extensive radiographic involvement correlates with a greater mycobacterial burden. Additionally, the imaging score at diagnosis significantly correlated with the body mass index, percentage of predicted forced expiratory volume in 1 s (FEV1), haemoglobin level, exercise tolerance and quality of life in another study, further emphasising the prognostic value of radiographic scoring in assessing the disease burden and predicting clinical outcomes.⁵

The likely contributing factor for non-conversion could be the co-existence of diabetes and a history of substance abuse. To the best of our knowledge, no studies establish a correlation between the Timika score and the TTP. Culture conversion and non-conversion demonstrated fair diagnostic performance, but it was not statistically significant.

Our study concludes that the Timika CXR score significantly correlates with higher baseline sputum smear grading and also signifies its correlation with the mean time to culture positivity. Cavitatory lesions on chest radiography are significantly associated with higher baseline mycobacterial load, as determined by sputum smear and culture. The Timika score appears to be a valuable tool in predicting infectiousness and disease severity at diagnosis, allowing for potential modifications in treatment strategy and duration among PTB patients.

Limitations

This study is limited by its small sample size and single-centre design. More comprehensive studies with a larger sample size and a multicentre prospective design could yield more insightful findings.

Footnotes

Acknowledgements

The authors thank the HOD Microbiology for departmental support, Dr Puneeta Hyanki for help at DOTS Center and Dr Rajat Jhamb for clinical support.

ORCID iDs

Neelam Arya

Bineeta Kashyap

Anupama Tandon

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

The top 10 causes of death [Internet]. [cited 2024 Jan 1]. Available from: https://www.who.int/news-room/fact-sheets/detail/the-top-10-causes-of-death.

WHO global report 2024 [Internet]. Geneva (citated Feb 2025). Available from [https://www.who.int/teams/global-programme-on-tuberculosis-and-lung-health/tb-reports/global-tuberculosis-report-2024].

Campbell

Bah-Sow

. Pulmonary tuberculosis: diagnosis and treatment. Br Med J 2006; 332: 1194–1197.

Krishnamoorthy

Knudsen

Rajaa

, et al. Accuracy of the Timika X-ray scoring system to predict the treatment outcomes among tuberculosis patients in India. Indian J Tuberc 2022; 69: 476–481.

Ralph

Ardian

Wiguna

, et al. A simple, valid, numerical score for grading chest x-ray severity in adult smear-positive pulmonary tuberculosis. Thorax 2010; 65: 863–869.

Pal Singh

Singh

Kaur

, et al. The importance of chest X-ray (Timika score) to predict the clinico-bacteriological profile of pulmonary tuberculosis patients. I J Appl Res 2021; 11: 55–58.

Revised National TB Control Programme Training manual for M. tuberculosis Culture and Drug susceptibility Testing.pdf. [Internet]. [cited 2025 Feb 05]. Available from: https://tbcindia.gov.in/writereaddata/l892s/6995271860Training%20manual%20M%20tuberculosis%20C%20DST.pdf

Chakraborthy

Shivananjaiah

Ramaswamy

, et al. Chest X-ray score (Timika score): an useful adjunct to predict treatment outcome in Tuberculosis. Adv Respir Med 2018; 86: 205–210.

Maretbayeva

Rakisheva

Adenov

, et al. Culture conversion at six months in patients receiving bedaquiline- and delamanid-containing regimens for the treatment of multidrug-resistant tuberculosis. Int J Infect Dis 2021; 113: S91–S95.

10.

Horne

Royce

Gooze

, et al. Sputum monitoring during tuberculosis treatment for predicting outcome: systematic review and meta-analysis. Lancet Infect Dis 2010; 10: 387–394.

11.

Nakiyingi

Bwanika

Ssengooba

, et al. Chest X-ray interpretation does not complement Xpert MTB/RIF in diagnosis of smear-negative pulmonary tuberculosis among TB-HIV co-infected adults in a resource-limited setting. BMC Infect Dis 2021; 21: 63.