Introduction

Tuberculosis (TB) is an infectious disease caused by the bacillus Mycobacterium tuberculosis. TB is not only typically affecting the lungs known as pulmonary TB, but also affects any part of the body as single or multiple sites known as extra-pulmonary TB [1,2,3]. Tuberculous bacillus causes a focal infection in the site where it is deposited after inhalation [4]. TB remained one of the top 10 causes of death worldwide. It still remains the world’s endemic infection although between 2000 and 2015 the number of deaths fell by 22%. The TB epidemic is larger than previously estimated. There were an estimated 10.4 million incident TB cases worldwide in 2015, while the rate of decline incidence from 2014 to 2015 remained at only 1.5% although the treatment averted a million deaths. This condition could be due to persisting gaps between treatment and diagnostic modalities. According to WHO report, one-third of pulmonary TB was undiagnosed or delayed diagnosed caused continued transmission in communities [1].

The diagnosis of TB remained unchanged for many decades and it probably would have no progress. The ultimate diagnosis of TB depends on the recognition of Mycobacterium tuberculosis on histological examination and/or bacteriological culture. Several diagnostic tests for TB have been developed including sputum smear microscopy more than 100 years ago [1]. Unfortunately, the most of undiagnosed pulmonary TB was cases with negative smear sputum microscopic. A systemic review showed that the sensitivity of this test is only 31–69% [5, 6]. Molecular diagnostic tools have been developed, but it cannot be routinely used for diagnosis of pulmonary TB [5, 7,8,9]. Currently, the Xpert® MTB/RIF assay is recommended by WHO for the diagnosis of TB. The test has much better accuracy than microscopy and culture methods. The limitation of this test is requiring more developed laboratory capacity and sometimes difficult to collect adequate specimen. Other limitation is it takes 2 weeks to provide results [1, 10]. The diagnosis of extra-pulmonary TB is more difficult compared with that of pulmonary TB. Extra-pulmonary TB involves relatively inaccessible sites for bacteriological confirmation as well [11, 12]. In these situations, rapid non-invasive imaging modality is necessary to detect and localize the site of TB.

Nuclear medicine technique is a non-invasive diagnostic modality which is highly sensitive and specific to detect and localize the lesion at early stage and less time consuming. A wide variety of radiopharmaceuticals have been used for infection/inflammation imaging. 67Ga-citrate is a high-sensitive agent for infection/inflammation imaging, but non-specific, since malignancy diseases will provide similar result [13, 14]. 99mTc-albumin, 99mTc-nanocolloid, 99mTc-tetrofosmin, FDG-PET, and 99mTc-MIBI are other sensitive and non-specific radiopharmaceutical agents used for diagnosis of infections including TB [15,16,17,18]. 111In-oxine-WBC, 99mTc-HMPAO-WBC, 99mTc-HIg and peptide labeled are specific process agents for infection/inflammation imaging. However, these radiopharmaceuticals are basically not specific to separate bacterial infection from sterile inflammation. 99mTc-ciprofloxacin can be a more specific radiopharmaceutical for bacterial infection imaging, because it is taken up by living bacteria and it inactivates DNA gyrase [5, 19,20,21]. Unfortunately 99mTc-ciprofloxacin cannot differentiate TB from other bacterial infection, since ciprofloxacin acts as a broad-spectrum antibiotic that can be taken up by any living bacteria [19,20,21,22,23,24]. 99mTc-INH has been developed to image TB, but clinically is not widely use [25].

Ethambutol is an active specific antibiotic against mycobacterium. It inhibits mycolic acid in bacterial cell membrane [11, 26,27,28,29]. Verma et al. showed that 99mTc-labeled ethambutol is specifically taken up by Mycobacterium tuberculosis, and can be image using gamma camera [30]. In vivo studies showed that the whole body bio-distribution of 99mTc-ethambutol was consistent with the pharmacokinetic characteristics of ethambutol as anti-tuberculosis drug [30,31,32,33,34,35]. 99mTc-ethambutol remains in tubercular lesion as it is bound to mycolic acid in the cell wall of bacteria, but will be cleared out from non-tubercular lesions [34, 36]. 99mTc-ethambutol was not seen in dormant TB cases. The advantages of 99mTc-ethambutol scintigraphy are it is a non-invasive procedure and provides result faster than the cytological test with minimal or no side effects [35,36,37].

The aim of this study was to evaluate the diagnostic value of 99mTc-ethambutol scintigraphy in detecting and localizing of TB infection.

Materials and methods

Retrospective cross-sectional study was done in the Department of Nuclear Medicine and Molecular Imaging, Dr. Hasan Sadikin General Hospital/Faculty of Medicine, Universitas Padjadjaran Bandung, Indonesia. Secondary data were collected from medical record. Subjects were patients who were referred for 99mTc-ethambutol scintigraphy to confirm or exclude TB infection from 2009 to 2015. The duration between the start of ethambutol drug and the day of 99mTc-ethambutol scintigraphy test was recorded. Side effect from injection of radiopharmaceuticals was observed. Subjects without histopathological or microbiological data and under TB treatment for more than 2 weeks were excluded from this study. This study was approved by Health Research Ethic Committee Faculty of Medicine Universitas Padjadjaran no. 592/UN6.C1.3.2/KEPK/PN/2015.

Ethambutol cold kit was developed by Center for Radioisotope and Radiopharmaceuticals Technology, National Nuclear Energy Agency of Indonesian. Whole body and SPECT/CT imaging in the suspected areas were done at 1 and 4 h following 370–740 MBq intravenous injection of 99mTc-ethambutol. SPECT/CT was done in the suspected area of TB based on whole body imaging. Twenty-four hours imaging was done if necessary. The results of 99mTc-ethambutol scintigraphy were compared with that of histopathological or microbiological test.

99mTc-ethambutol labeling of procedure

Ethambutol cold kit comes in 2 vials. One vial (A) contains 3.5 mg ethambutol and 5 mg mannitol. The other vial (B) contains 400 µg SnCl2.2H2O and 2 mg sodium pyrophosphate. One mL aquades was added to vial A and mixed well. All the solution was taken from vial A and put into vial B using sterilized fine syringe, and mixed well. 370–740 MBq freshly eluted 99mTc was added to vial B. The mixture in vial B was swirled and allowed to react for 5–10 min at room temperature. Instant thin-layer chromatography (ITLC) was performed for quality control using acetone as a solvent to determine radiochemical purity of 99mTc-ethambutol. The radiochemical purity should be more than 90% to be used in this study. This based on the formulation for the pre-determined radiochemical purity that 88–94% radiochemical purity of 99mTc-ethambutol can be used [38].

Imaging protocol

There is no specific preparation for ethambutol scan. Anterior and posterior whole body and SPECT/CT images were done using a standard dual-head Gamma Camera SPECT/CT (GE-Infinia®) with Low-Energy High-Resolution (LEHR) Collimator at 1, 4 and if necessary 24 h following intravenous injection of 99mTc-ethambutol. Injection dose of 99mTc-ethambutol was 370–740 MBq. SPECT/CT images were done covering lungs and other suspected area to improve the sensitivity in detection of lesions. The subjects were observed for any signs and symptoms from side effect of radiopharmaceutical injection. Common side effects include problems with vision, headaches, nausea, joint pain, tiredness and allergic reaction.

Interpretation

The interpretation of image was based on the quality of tracer uptake. Images were analyzed visually for qualitative analysis. Normal distribution of 99mTc-ethambutol is seen as high uptake in kidney, urinary bladder, liver and spleen. There is no tracer uptake shown on bone, bone marrow,, epiphysis, stomach and thyroid, soft tissue and lung as well. Significant urinary activity was seen in 1 h, since the kidney is the main excretory organ [34]. Any increased pathological uptake of 99mTc-ethambutol seen in the area out of normal distribution compared with that of the opposite area was considered as positive results. The pathological uptake is increasing gradually at 4- or 24-h images compared with that of normal uptake. Negative result was considered if normal radioactivity distribution was seen without any pathological tracer uptake, or increased tracer uptake at 1 h, but decreasing at 4 h image [30, 32].

Statistic analysis

Statistic analysis was done to determine the sensitivity, specificity, negative and positive predictive value as well as accuracy by comparing the result of 99mTc-ethambutol scintigraphy with that of histopathological or microbiological using a 2 × 2 table.

Results

One hundred and sixty-eight subjects out of 221 fulfilled the inclusion and exclusion criteria. They were included in this study. There were 110 (65.5%) males and 58 (34.5%) females with mean age 34.52 ± 11.94 years and ranged 2–79 years old. Fifty-three subjects with spondylitis TB were excluded due to no data of histopathology or microbiological test. Final diagnosis of pulmonary TB was 52 subjects, lymphadenitis TB 33 subjects, spondylitis TB 40 subject, peritoneal TB 20 subject and other extra-pulmonary TB 5 subjects. Non-TB infection was found in 18 subjects. There were 23 subjects with history of taking ethambutol treatment less than 2 weeks prior to ethambutol scintigraphy. The average dose of 99mTc-ethambutol was 721.87 ± 95.09 MBq with average percentage of labeling efficiency as 95.82 ± 0.86% (Table 1). Normal whole body distribution of 99mTc-ethambutol at 1 and 4 h after injection of radiopharmaceutical is shown in Fig. 1. Positive finding was seen on 1- and 4-h whole body images (Fig. 2) and 4-h SPECT/CT images (Fig. 3). Increase pathological tracer uptake was seen in the upper lobe of the right lung. This finding was confirmed as pulmonary TB by microbiological examination. Figure 4 shows a positive result of 99mTc-ethambutol scintigraphy in subjects with final diagnosis of spondylitis TB with paravertebral abscess.

Table 1 The characteristic subject and the result of 99mTc-ethambutol scintigraphy and histopathology/microbiological
Fig. 1
figure 1

Whole body scan (voluntary healthy human) shows normal bio-distribution of 99mTc-ethambutol at 1 and 4 h after injection

Fig. 2
figure 2

One- and 4-h whole body 99mTc-ethambutol scintigraphy of patient with pulmonary TB. Anterior and posterior images showed increase pathological tracer uptake in the upper lobes of the right lung

Fig. 3
figure 3

The characteristic of radiotracer uptake based on diversity of TB pathology within a single individual patient at the same time. Microbiological test was positive

Fig. 4
figure 4

99mTc-ethambutol SPECT/CT of patient with approved spondylitis TB by histopathological examination. Specimen was taken during surgery. CT images (left side) showed deformity of thoracic spines with paravertebral abscess. 99mTc-ethambutol images (right side) showed increased pathological tracer uptake at several thoracic spines and surrounding paravertebral abscess

There was concordance between 99mTc-ethambutol scintigraphy results and mycobacterial or histopathological finding in 156 of 168 subjects (92.9%). There were 131 (78%) subjects who were positive on both 99mTc-ethambutol scintigraphy and microbiological or histopathological finding, while the other 25 (14.9%) subjects were negative on both examinations. The results were discordant in 12 (7.1%) subjects. Five subjects were positive on 99mTc-ethambutol scintigraphy but negative on microbiological or histopathological test. The other 7 subjects were negative on 99mTc-ethambutol scintigraphy, but positive on microbiological or histopathological test. All 23 subjects with the history of taking ethambutol treatment showed true-positive results. The duration time between the first day of taking ethambutol drug and the day of 99mTc-ethambutol scintigraphy test was 7–12 days.

Table 2 shows a 2 × 2 table between 99mTc-ethambutol scintigraphy and histopathological/microbiological test for pulmonary TB, extra-pulmonary TB and total subject. Table 3 shows the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy of 99mTc-ethambutol scintigraphy in the diagnosis of pulmonary TB as 93.9, 85.7, 93.9, 85.7 and 91.4%, respectively, for extra-pulmonary TB 95.5, 77.8, 97.9, 63.6, and 85.1%, respectively, and for total tuberculosis 94.9, 83.3, 96.3, 78.1 and 92.9%, respectively. There were no signs and symptoms of the side effects observed after the injection of radiopharmaceuticals.

Table 2 2 × 2 table between 99mTc-ethambutol scintigraphy and histopathologic/microbiological test
Table 3 Sensitivity, specificity, positive predictive value, negative predictive value and accuracy of 99mTc-ethambutol in the diagnosis of tuberculosis

Discussion

In this study, we found that two-third of the subjects were male with very wide range of age from 2 years to 79 years. We excluded 53 subjects with clinical diagnosis of having spondylitis TB. Histopathological or microbiological data were not available because no specimen is available due to the subject refusing to undergo surgery or tissue biopsy. The radiochemical purity of 99mTc-ethambutol used in this study was higher compared to that recommended by other study based on the formulation [31].

Increased pathological uptake of 99mTc-ethambutol was observed at focal lesion of active tuberculosis in1- and 4-h images. This pathological uptake was due to increase and retention of radiopharmaceuticals by live M. tuberculosis. The retention of 99mTc-ethambutol in the 24-h image increased the specificity and positive predictive value of 99mTc-ethambutol scintigraphy, particularly, in doubtful cases. Any radiotracer uptake by infected/inflammation lesions at the initial phase is nonspecific, since most of the tracers are distributed at the blood pool lesion and extracellular space [15]. This phenomenon is seen in 99mTc-ethambutol scintigraphy as well.

In this study, we found 92.9% subjects showed concordance and 7.1% discordance between 99mTc-ethambutol scintigraphy result and mycobacterial or histopathological finding. The sensitivity and positive predictive value of 99mTc-ethambutol scintigraphy for the diagnosis of pulmonary is more than 90% as well as for extra-pulmonary TB, but the specificity and negative predictive values were less than 90%. 99mTc-ciprofloxacin is more specific among other radiopharmaceuticals for the diagnosis of bacterial infection. Lee et al. showed that the sensitivity and specificity of 99mTc-ciprofloxacin SPECT for detecting active pulmonary tuberculosis before treatment were 80.0 and 90.9%, respectively. The positive predictive value was 88.9% and the negative predictive value was 83.3% [5]. We found that the sensitivity, negative and positive predictive values of 99mTc-ethambutol scintigraphy were higher compared to that in the study by Lee et al., but in the contrary, the specificity was slightly lower. Theoretically, the specificity of 99mTc-ethambutol should be higher compared with that of 99mTc-ciprofloxacin, since 99mTc-ethambutol is more specific for Mycobacterium tuberculosis. Low negative predictive value in the diagnosis of TB could be due to a limited number of subjects with non-TB. Negative TB was found in only 30 subjects, while positive TB was found in 138 subjects. Less number of subjects with negative TB were found in extra-pulmonary TB compared with that for pulmonary TB. It is recommended to do another study with comparable number of subjects between positive and negative TB.

Tuberculosis lesions in human are very complex with a wide range of pathological features. This variety of lesions could be observed within a single individual patient simultaneously [39,40,41]. Due to their characteristic; these lesions will provide either false-positive or false-negative results on 99mTc-ethambutol scintigraphy. False-positive result could be caused by: (1) hypervascularization of non-specific infection or inflammation [30], (2) difficulties to get the adequate specimen [41], (3) and at least 104 number of acid-fast bacilli/mL of specimen culture are required to provide positive result [42, 43]. In this study, we found 5 false-positive subjects: 3 subjects from the group of pulmonary TB and the other 2 subjects from the group of extra-pulmonary TB showed false-positive result. Hypervascularization could be found in TB granuloma. High uptake of radioactivity seen at solid lesions or nodules in lung parenchyma suggested granuloma cellular uptakes just around the lesions of largely filled cavities. Granuloma contains predominantly intracellular M. tuberculosis with good vascularization which showed high uptake of 99mTc-ethambutol. Negative smear sputum microscopic test in those subjects could be because granuloma did not contact with airway lung structure [39, 40, 44]. Subject with false positive on 99mTc-ethambutol scintigraphy, but abnormal clinical feature and chest X-ray were considered as suggestive for active pulmonary tuberculosis. Those subjects were treated with anti-tuberculosis drugs. On the follow-up, they showed better improve clinical features. This false-positive case can be considered as true positive based on good response following anti-tuberculosis treatment, although smear sputum microscopic test was negative. Performing serial late images could minimize false-positive result of 99mTc-ethambutol scintigraphy due to hypervascularization. Twenty-hour images following injection of 99mTc-ethambutol could be done if necessary.

In this study, we found false-negative results in 7 subjects, 3 subjects belong to pulmonary TB group and 4 subjects belong to extra-pulmonary TB group. They showed negative result on 99mTc-ethambutol scintigraphy, but positive on microbiological or histopathological test. False-negative result could be found in necrotic caseous lesions that spread and destroy vasculature which lead to lack of radiopharmaceutical supply from blood [16]. These necrotic lesions can remain solid with very few bacilli. Necrotic caseous lesions can be seen on X-ray without radiotracer uptake that caused false negative results on 99mTc-ethambutol scintigraphy with positive smear sputum microscopic test [39, 40].

During the period of 5 years performing 99mTc-ethambutol scintigraphy, we found 13 negative 99mTc-ethambutol scintigraphy results in patients suspected of having TB. All of them were under intensive phase treatment using ethambutol. Since the characteristic of 99mTc-ethambutol is similar to ethambutol used for treatment, the question arises whether negative result is true negative or false negative due to competition between 99mTc-ethambutol as radiopharmaceutical and ethambutol as anti-tuberculosis drug. In our study, 23 subjects with positive pulmonary TB based on smear sputum microscopic test were positive on 99mTc-ethambutol scintigraphy. All of these subjects showed positive result on the second 99mTc-ethambutol scintigraphy performed after taking ethambutol drug for 7–12 days. This finding showed that taking drug during intensive treatment less than 2 weeks would not affect the result of 99mTc-ethambutol scintigraphy.

In this study we did not find any sign and symptom related to injection of 99mTc-ethambutol. Problem in vision due to optic neuropathy or cardio-hepatotoxicity occurs very rarely after several months of therapeutics. Ethambutol is freely given to pediatric patients. Adverse effects of ethambutol are rare and dose dependent. The diagnostic dose administered is only less than 3.5 mg. In clinical context, it was considered as a safe radiopharmaceutical, even in children [31].

Conclusion

This study showed that 99mTc-ethambutol scintigraphy is a useful diagnostic imaging technique to detect and localize both intra- and extra-pulmonary tuberculosis. 99mTc-ethambutol scintigraphy is safe to be performed even in pediatric patient. Consuming ethambutol less than 2 weeks does not influence the result of 99mTc-ethambutol scintigraphy.