Low-dose computed tomography in COVID-19: systematic review

Ivan A. Blokhin; Блохин Иван Андреевич; Ivan A. Blokhin; Denis А. Rumyantsev; Румянцев Денис Андреевич; Denis А. Rumyantsev; Maria M. Suchilova; Сучилова Мария Максимовна; Maria M. Suchilova; Anna P. Gonchar; Гончар Анна Павловна; Anna P. Gonchar; Olga V. Omelyanskaya; Омелянская Ольга Васильевна; Olga V. Omelyanskaya

doi:10.17816/DD119870

Low-dose computed tomography in COVID-19: systematic review

Authors: Blokhin I.A.¹, Rumyantsev D.А.¹, Suchilova M.M.¹, Gonchar A.P.¹, Omelyanskaya O.V.¹
Affiliations:
1. Moscow Center for Diagnostics and Telemedicine
Issue: Vol 4, No 1 (2023)
Pages: 25-37
Section: Systematic reviews
URL: https://jdigitaldiagnostics.com/DD/article/view/119870
DOI: https://doi.org/10.17816/DD119870
ID: 119870

Cite item

Full Text

Abstract
Full Text
About the authors
References
Supplementary files
Statistics

Abstract

BACKGROUND: The increased number of computed tomography scans during the COVID-19 pandemic has emphasized the task of decreasing radiation exposure of patients, since it is known to be associated with an elevated risk of cancer development. The ALARA (as low as reasonably achievable) principle, proposed by the International Commission on Radiation Protection, should be adhered to in the operation of radiation diagnostics departments, even during the pandemic.

AIM: To systematize data on the appropriateness and effectiveness of low-dose computed tomography in the diagnosis of lung lesions in COVID-19.

MATERIALS AND METHODS: Relevant national and foreign literature in scientific libraries PubMed and eLIBRARY, using English and Russian queries “low-dose computed tomography” and “COVID-19,” published between 2020 and 2022 were analyzed. Publications were evaluated after assessing the relevance to the review topic by title and abstract analysis. The references were further analyzed to identify articles omitted during the search that may meet the inclusion criteria.

RESULTS: Published studies summarized the current data on the imaging of COVID-19 lung lesions and the use of computed tomography scans and identified possible options for reducing the effective dose.

CONCLUSION: We present techniques to reduce radiation exposure during chest computed tomography and preserve high-quality diagnostic images potentially sufficient for reliable detection of COVID-19 signs. Reducing radiation dose is a valid approach to obtain relevant diagnostic information, preserving opportunities for the introduction of advanced computational analysis technologies in clinical practice.

Keywords

computed tomography, low-dose computed tomography, literature review, COVID-19, COVID-19 diagnosis

Full Text

BACKGROUND

At the time of writing this paper (December 22, 2022), the number of coronavirus disease 2019 (COVID-19) cases reaches 650 million.¹ Spread of the disease and consequent mortality can be prevented and reduced with a combination of measures, including early diagnosis.²

The main method of laboratory diagnostics is reverse-transcription polymerase chain reaction (RT PCR). At the first peak of the coronavirus pandemic, disadvantages of this technique were revealed, such as a high rate of false-positive results, limited test availability, and long waiting time for results. [1] Moreover, computed tomography (CT) of the chest can also provide false-negative results in patients with signs of COVID-19. [2]

According to Russian [3] and international³ guidelines, diagnostic radiology for COVID-19-associated pneumonia includes radiography and CT. Chest X-ray imaging has low sensitivity to viral pneumonia [4], so CT plays an important role in the diagnosis of COVID-19-associated pneumonia and its complications. [5]

The active use of CT during the pandemic leads to high radiation exposure to the population. [6, 7] To assess changes in patient condition during the hospital stay, 2–6 CTs are usually performed within a short period since a clear trend toward CT regression of abnormal changes is one of the criteria for patient discharge.[3] Patients with suspected COVID-19 may undergo 1–2 outpatient CT scans to detect signs of the disease. [8. 9]

Radiation exposure is significantly associated with the increased risk of developing cancer. [10] In radiology departments, even in a pandemic, the “as low as reasonably achievable” principle proposed by the International Commission of Radiological Protection [11] should be observed. In March 2020, Kang et al. proposed the use of low-dose CT (LDCT) as the first stage of radiation diagnosis in patients with COVID-19-associated pneumonia. [6] The important role of LDCT in COVID-19 was also highlighted by the webinar “COVID-19 and Chest CT: Protocol and Dose Optimization,” which was held in April 2020 and attended by 1633 people from 100 countries. During the video conference, it was found that 55%, 43%, and 2% of healthcare institutions use standard (CTDIvol 5–10 mGy), low-dose (CTDIvol <5 mGy), and high-dose protocols (CTDIvol >10 mGy). [12] However, even a superficial assessment of an X-ray workstation reveals a significant number of scan parameters that affect radiation exposure, [13], and the relationship between different protocol settings and radiation dose may not be obvious, especially considering the pathology examined.

This literature review aimed to systematize data on opportunities for reducing radiation exposure during lung CT in patients with COVID-19.

MATERIALS AND METHODS

Relevant Russian and foreign literature sources were reviewed in PubMed and eLIBRARY scientific libraries for search queries “low dose computed tomography COVID-19” and “низкодозная компьютерная томография COVID-19” for the period from 2020 to 2022.

Publications were included in the review after assessing their relevance according to their titles and abstracts. The review included original studies and meta-analyses, and literature reviews, case reports, and congress abstracts were excluded. References were also reviewed for relevant studies on general principles of CT dose reduction that might have been published before 2020. When such articles were found, the most recent study was included in the review.

RESULTS AND DISCUSSION

In total, 45 foreign papers and five Russian papers were reviewed. The latest search date was December 22, 2022.

Methods for Reducing Radiation Exposure

Methods for reducing radiation exposure can be divided into hardware and software ones. Hardware methods are related to tube potential, tube current, pitch factor, and X-ray beam filter. Software methods are related to the reconstruction filter, slice thickness, and iterative reconstructions.

Hardware Methods. Tube potential (kVp) is nonlinearly related to radiation exposure. [14] Zarb et al. [15] showed that a decrease in tube potential by 14%–17% provides a radiation dose decrease by 32%–38%. In this case, reducing the tube potential increases the noise level while performing non-contrast-enhanced examinations. A phantom study showed that these parameters are interrelated via an exponent of −1.3. [16] Moreover, reducing the tube potential in contrast-enhanced examinations improves the quality of images by significantly reducing radiation exposure. [17]

Tube current (mAs) is linearly related to radiation exposure. [18] For example, a 50% decrease in tube current leads to a 50% decrease in the effective dose [17], while the signal-to-noise ratio is inversely proportional to the square root of the current. [19]

A pitch factor for multislice CT has almost no effect on radiation exposure. [20] As the pitch factor increases, the signal-to-noise ratio decreases, and the tomograph automatically increases the tube current to prevent deterioration of image quality. [21]

An X-ray beam filter is used to absorb low-energy photons that cannot pass through the patient tissues and do not reach the detectors. Therefore, the use of an additional tin filter can significantly reduce radiation exposure during CT [22] but requires additional costs for scanner modification.

Software Methods. The choice of the reconstruction filter (convolution kernel) does not affect radiation exposure but affects the signal-to-noise ratio, amplifying or smoothing out the difference between pixels of different organs or structures. [23]

A low slice thickness is associated with lower image quality but decreases the risk of missing small abnormal changes. Thus, slice thickness can be optimized. For example, for the examination of pulmonary nodes, this parameter may be 2 mm. [24]

The main way to reduce “noise” is iterative reconstructions, which allow CTs with lower radiation doses and a similar signal-to-noise ratio to the standard data reconstruction technique. [25] The use of artificial neural networks for image reconstruction is one of the promising methods. [26, 27]

Based on the literature review, reducing the tube current reduces radiation exposure, and the signal-to-noise ratio can be optimized using a reconstruction filter that smoothens the difference between adjacent pixels (soft tissue) and iterative reconstruction.

Low-dose CT in the diagnosis of COVID-19

In the literature review, a single, well-defined low-dose protocol is warranted for COVID-19 (Table 1 [28–54]). The reduced radiation exposure dose is achieved mainly by changing the tube potential, tube current, use of iterative reconstructions, and a tin filter. Some studies reviewed had shortcomings related to data presentation: dosimetric parameters (CTDI, DLP, SSDE, and effective dose) were not mentioned, and small sample sizes were used.

Table 1. Parameters of low-dose computed tomography for diagnosis of COVID-19 based on the literature review

Автор, год, ссылка	Напряжение трубки, кВ	Сила тока трубки, мАс	Средняя доза облучения, мЗв	Толщина среза, мм	Фильтр реконструкции	Использование итеративной реконструкции
Author, year, link	Tube potential (kV)	Tube current, mA	Average radiation dose (mSv)	Slice thickness (mm)	Reconstruction filter	Use of Iterative Reconstruction
Blokhin et al. (2022) [28]	120	10–500, noise level 36 (SD)	3	1	FC51, FC07	No
Filatova et al. (2020) [29]	100/110	40–120	1,27	1	-	Yes
Afshar et al. (2022) [30]	110	20	1–1,5	2	D40s	-
Fukumoto et al. (2022) [31]	120	20–25	СTDI 1.3 mGy	5	Lung and soft tissue	-
Bieba et al. (2022) [32]	Depending on weight	Depending on weight	-	1 and 3	-	-
Barrio et al. (2022) [33]	100/150	Anthropomorphic current modulation system	-	1	Br32 Bl60	-
Thieß et al. (2022) [34]	100	10–100	0,53	0,5 b 0,625	Fc01 Fc85	Yes
Greffier et al. (2021) [35]	100/120	10	0,2	1	I30f, mediastinal, I50f, lung images	Yes
Karakaş et al. (2021), [36]	80	40	0,18	5	lung	Yes
Julie et al. (2021) [37]	120	45	-	1,2	-	-
Desmet et al. (2021) [38]	80–140	20–30	0,64	0,6	-	-
Aslan et al. (2021) [39]	80	35–50	0,2856	3	lung	Yes
Stoleriu et al. (2021) [40]	120	40–113	35–100 mGy×cm 0.78–2.91 mGy	1,25	Medium Soft	Yes
Bai et al. (2021) [41]	120	120–380	1,21±0,10	1,25	Standard	Yes
Agostini et al. (2021) [42]	100	95	0,39	1,5	Sharp	Yes
Zali et al. (2021) [43]	100–120	50–100	-	1-3	-	-
Argentieri et al. (2021) [44]	80	20	0,219	2	Sharp	-
Leger et al. (2020) [45]	120	45	0,49	1,2	-	-
Hamper et al. (2020) [46]	100	20–120	0,5	0,625–1	Lung	Yes
Li et al. (2020) [47]	120	30	1,22±0,14	1	-	Yes
Dangis et al. (2020) [48]	100	20	0,56	1	Lung (150f)	Yes
Radpour et al. (2020) [49]	100–120	50–100	-	1–3	-	-
Kang et al. (2020) [6]	80–100	10–25	0,203	0,6	-	Yes
Tofighi et al. (2020) [50]	100	40	2,03	-	-	No
Tabatabaei et al. (2020) [51]	120	30	1,8	3	-	-
Schulze-Hagen et al. (2020) [52]	80	35	1,7	1 and 3	170f 130f	-
Zhao Yue et al. (2020) [53]	100	50	1,5	1	-	-
Castelli et al. (2020) [54]	120	45	0,47	1,2	-	-

Interestingly, a parameter to be changed when optimizing the scanning protocol can be universal for various clinical tasks. Therefore, in LDCT for lung cancer screening, different groups of authors also changed the tube current [55, 56]. However, the development of a specialized LDCT protocol should be initiated with a study on a model object (phantom) to select the optimal method for reducing the exposure. For example, Gombolevsky et al. [57] developed the LDCT protocol for the diagnosis of COVID-19 using a phantom with thickening plates, while setting the automatic tube current control system (Sure Exposure 3D) to a sufficient level to detect ground-glass lesions with a maximum reduction in radiation exposure (SD = 36). A comparison of the protocol selected according to the results of the phantom study with standard CT and LDCT for lung cancer screening is shown in Fig. 1.

Figure 1. Comparison of a dedicated low-dose computed tomography protocol for COVID-19 (SD = 36) with standard and low-dose computed tomography for lung cancer screening. Data on radiation exposure and axial tomograms of the phantom at the level of the lower and middle zones of the lungs. Low-dose computed tomography for lung cancer screening was developed considering the need for radiation exposure limitation as preventive measures according to SanPin (disease control and prevention standards) and has the lowest signal-to-noise ratio. The proposed protocol for low-dose computed tomography for COVID-19 considers the densitometric characteristics of ground-glass lesions with a significant reduction in radiation exposure.

Any special low-dose protocols require clinical validation and comparison with the gold standard. Therefore, clinical trials of the developed LDCT protocol for COVID-19 used standard CT as a reference technique. [28] Some of the clinical images obtained using the developed protocol are shown in Figs. 2 and 3.

Figure 2. Radiation exposure is reduced by 5 times. Patient, 59 y. o., BMI 29 kg/m². Computed tomography with a soft tissue filter (effective dose: 9.7 mSv), low-dose computed tomography with a soft tissue filter (effective dose: 2.1 mSv). In the upper lobe of the left lung, there was a peripheral ground-glass lesion.

Figure 3. Radiation exposure is reduced by 1.5 times. Patient, 44 y. o., BMI 46 kg/m². Computed tomography with a soft tissue filter (effective dose: 15.3 mSv), low-dose computed tomography with a soft tissue filter (effective dose: 10.5 mSv). Bilateral peripheral ground-glass lesions.

Limitations of LDCT

Kim et al. [58] showed that obesity (body mass index > 25) appears to limit the use of chest LDCT in routine practice due to X-ray absorption by adipose tissue. However, studies on the inter-expert agreement of COVID-19 examinations indicate the opposite. [59]

In addition, LDCT empirically appears to be related to the negative effect of increased image noise on the operation of AI systems, including the calculation of an emphysema index in densitometric analysis, [60] and the radiomic analysis of subsolid pulmonary nodules. [61] Effect of a scanning protocol on the results of a quantitative analysis can be reduced using relative parameters, for example, the percentage of affected lung tissue in COVID-19, [62] or by normalizing the obtained data with special algorithms. [63]

CONCLUSION

Methods are presented to reduce radiation exposure during chest CT and maintain high-quality diagnostic images that are hypothetically sufficient to reliably detect signs of COVID-19. Although there is no single way to optimize scan protocols, dose reduction provides relevant diagnostic information and retains the ability to incorporate advanced computer-assisted technologies into clinical pathways.

ADDITIONAL INFORMATION

Funding source. This article was prepared by a team of authors as part of the research work (EGISU number: AAAA-A20-120071090058-7) in accordance with the Program of the Moscow Department of Health “Scientific support of metropolitan health” for 2020-2022.

Competing interests. The authors declare that they have no competing interests.

Authors’ contribution. All authors made a substantial contribution to the conception of the work, acquisition, analysis, interpretation of data for the work, drafting and revising the work, final approval of the version to be published and agree to be accountable for all aspects of the work. The largest contributions were distributed as follows: I.A. Blokhin — editing and approval of the final manuscript text, advisory support; D.A. Rumyantsev — data analysis, article text writing; M.M. Suchilova, A.P. Gonchar — editing and approval of the final manuscript text; O.V. Omelyanskaya — study concept and design.

¹World Health Organization. Novel Coronavirus (COVID-19) situation. Available at https://who.sprinklr.com.

² Centers for Disease Control and Prevention. Coronavirus 2019 disease (COVID-19). Available at https://www.cdc.gov/coronavirus/2019-ncov/index.html.

³ World Health Organization. Use of chest imaging in COVID-19: A rapid advice guide [11 June 2020]. Available at https://apps.who.int/iris/handle/10665/332336.