www.buffalo.edu

• For 116 enrolled patients, final treatment outcomes were judged a success by WHO guidelines

[9] in 110 patients, but were judged to fail in the other 6 patients.

• At 2 month follow up, sputum exams were negative in 72 patients in the success group and 2

patients in the fail group (P = 0.04, Chi test).

• DT showed improvement in 22 patients in the success group and in no patients in the fail group

(P = 0.18).

• Chest radiography is widely used not only in the screening and diagnosis of pulmonary

tuberculosis (TB), but also plays a significant role in the evaluation of treatment response [1, 2].

• However, even in patients with active pulmonary TB disease, chest radiographs may show only

subtle findings or even appear normal [2]. In patients with normal or inconclusive chest

radiographs, computed tomography (CT) plays a major role in providing more precise

information of TB, yet, the potential radiation hazards and relatively high costs remain

problematic [1, 3] in particular in low and medium resource healthcare environments.

• Digital tomosynthesis (DT) has a lower cost and lower effective radiation dose of about 0.1

mSv (30-70 times lower than that for multidetector CT, comparable to the dose of a

posteroanterior and lateral chest radiographs) [4, 5, 6].

• DT is an imaging technique based on reconstruction of multiple radiographic image projections

acquired over a limited angle and provides improved conspicuity by reducing the visual

distraction from overlying structures [7].

• Several previous studies have demonstrated that DT is superior to conventional radiography for

the detection of pulmonary nodules and lung lesions in patients with pulmonary mycobacterial

disease [4, 5, 8].

Background

Digital Tomosynthesis of Chest : Serial Radiographic Response in Patients with Pulmonary Tuberculosis

John M. Sabol, PhD3, Myung Jin Chung, MD1, Geewon Lee, MD1,Won-Jung Koh, MD2, and

Kyung Soo Lee, MD1

1. Department of Radiology, Sungkyunkwan University School of Medicine. Seoul, KOREA

2. Department of Pulmonology, Sungkyunkwan University School of Medicine. Seoul, KOREA

3. GE Healthcare, WI, USA

Purpose

• The purpose of this study was to predict outcomes of TB treatment using digital

tomosynthesis (DT) at presentation and at follow-up studies after antibiotic therapy.

Projection Images

Lesion A Lesion B

1 2 3 4 5

Tube Sweeping

+5 +4 +3 +2 +1 focused on plane of lesion A

+1

+2

+3

+4

+5

focused on plane of lesion B

Sectional Image reconstruction

Figure 2. Diagram of Tomosynthesis

Materials and Methods

• Serial DT scans of 110 patients with DSTB (n=82) and MDRTB (n=28).

• DT scans at presentation were examined for the presence of lung abnormalities including

bronchiolitis, airspace consolidation, nodules, bronchiectasis, atelectasis or volume loss, and

cavities.

• Serial DTs were examined for change in the size of cavities and overall assessment of

improvement or aggravation at intervals of 2 and 6 months after initiation of treatment.

• Four board certified chest radiologists reviewed with consensus retrospectively

• Overall extent of disease was scored on a semi-quantitative %-age scale. “Improvement” was

defined if the extent of disease at follow-up decreased 5% or more from initial extent.

Figure 1. Wall stand example

• AFB (-) conversion was acquired

during 2nd line CTx.

• Until 3 months, cavity in left apex

decreased on DT.

• At 4 months, cavity regrew on DT.

However sputum AFB was still (-)

• At 10 months, sputum AFB was still

(-), but cavity continuously grew.

• Therefore, wedge resection was

done, and AFB was (+) in specimen

culture.

AFB 1+ AFB - AFB - AFB -

Initial 3 month 4 month 10 month

Figure 7. M/35, s/p Rt pneumonectomy d/t known Multi-Drug Resistant tuberculosis

Init

ial

6 m

on

th F

/U

Tuberculosis was proven as drug sensitive by sputum AFB culture.

1. Initial DR does not depict any lesion

2. Initial LD DT clearly depicts fine micro nodules in right lung apex

3. Initial CT confirms this lesion as fine clustered bronchiolitis called

‘Galaxy sign’

4. F/U DT shows almost complete disappearance of the lesion in

right apex

1. 2.

3. 4.

DR DT

DR DT

Figure 6. M/ 46, cough

Init

ial

2 m

on

th F

/U

• Tuberculosis was proven as drug

sensitive by sputum AFB culture.

1.Small nodular opacity is noted in left

middle lung zone on initial DR.

2.There is thin walled cavity on initial

DT (arrow). Thin walled cavity can

not be demarcated enough on DR.

3.F/U DR shows no significant change

of nodule.

4.F/U DT shows near complete

obliteration of cavity and residual

nodular lesion.

1. 2.

3. 4.

DR DT

DR DT

Figure 5. M/ 18, abnormal finding on mass screening

Conventional radiography Serial sectional images of DT reconstructed from projection images

1 2 3 4 5 6

Figure 3. Reconstruction of Multiple planar image in Digital Tomosynthesis

Results

Final Treatment Response

Success Fail Sum

Enrolled patients 110 6 116

2 month Exam 110 6 116

6 month Exam 86 3 89

2 month Sputum

Success Fail Sum

Negative 72 2 74

Positive 26 4 30

Total 98 6 104

2 month CR

Success Fail Sum

Improve 54 2 58

Non-Improve 56 4 58

Total 110 6 116

2 month DT

Success Fail Sum

Improve 27 0 27

Non-Improve 83 6 89

Total 110 6 116

6 month DT

Success Fail Sum

Improve 43 3 46

Non-Improve 43 0 43

Total 86 3 89

• In 6 month follow up, DT showed

improvement in 40 patients in

success group and three patients in

fail group (P = 0.11).

• Although not statistically significant, there were no false positive cases that showed

radiographic improvement on DT 2 months after initiation of antibiotic therapy. CR and sputum

exam showed some false positives. Thus, we conclude that early radiographical

improvement visualized on DT can help predict TB treatment success.

References 1. Jeong, Y.J. and K.S. Lee, Pulmonary tuberculosis: up-to-date imaging and management. AJR Am J Roentgenol, 2008. 191(3): pp. 834-844.

2. Woodring, J.H., et al., Update: the radiographic features of pulmonary tuberculosis. AJR Am J Roentgenol, 1986. 146(3): pp. 497-506.

3. Lee, K.S., et al., Utility of CT in the evaluation of pulmonary tuberculosis in patients without AIDS. Chest, 1996. 110(4): pp. 977-984.

4. Kim, E.Y., et al., Pulmonary mycobacterial disease: diagnostic performance of low-dose digital tomosynthesis as compared with chest radiography. Radiology, 2010. 257(1): pp.

269-277.

5. Vikgren, J., et al., Comparison of chest tomosynthesis and chest radiography for detection of pulmonary nodules: human observer study of clinical cases. Radiology, 2008. 249(3):

pp. 1034-1041.

6. Sabol, J.M., A Monte Carlo Estimation of Effective Dose in Chest Tomosynthesis Medical Physics, 2009 36(12):pp. 5480-5487.

7. Dobbins, J.T., 3rd and H.P. McAdams, Chest tomosynthesis: technical principles and clinical update. Eur J Radiol, 2009. 72(2): pp. 244-251.

8. James, T.D., et al., Digital tomosynthesis of the chest for lung nodule detection: interim sensitivity results from an ongoing NIH-sponsored trial. Med Phys, 2008. 35(6): pp. 2554-

2557.

9. WHO, Treatment of tuberculosis: guideline 4th ed. 2010.

Figure 4: Comparison of semi-quantitative

assessment extent of disease in

DT between initial and 2 or 6 month follow-

up . At 6 months after treatment, extent of

disease was judged as decreased

(improvement) in most patients even in

treatment failures. This phenomenon

causes lower specificity and lower utility of

conventional radiographic examination in

late follow up period.