Cumulative radiation exposure from repeated CT imaging in hospitalised COVID-19 patients: Does it matter? Authors: Marieta Mateva-Mladenova, Stefan Stanev, Dora Zlatareva, Desislava Kostova-Lefterova Published: 2026 ABSTRACT AbstractPurposeTo assess cumulative radiation exposure associated with repeated CT imaging in hospitalized COVID-19 patients and to identify factors contributing to high cumulative dose.Materials and MethodsThis retrospective observational study included 69 adult patients with confirmed COVID-19 who underwent three or more CT examinations during a single hospitalization. A total of 296 CT examinations comprising 350 acquisition phases were analyzed. Dose parameters (CTDIvol, DLP) were recorded, and cumulative effective dose (CED) was calculated using region-specific conversion factors. Continuous variables were expressed as median with interquartile range (IQR).ResultsThe median cumulative effective dose was 47.7 mSv (IQR 34.5–87.2; range 18.0–342.6). A total of 32 patients (46.4%) accumulated ≥ 50 mSv, while 16 patients (23.2%) exceeded 100 mSv during a single hospitalization, including one patient (1.4%) with exposure ≥ 200 mSv. CED showed a moderate correlation with the number of CT examinations (ρ=0.37, p = 0.0016) and a stronger correlation with the total number of acquisition phases (ρ=0.54, p < 0.00001). Patients with CED ≥ 100 mSv had a higher median number of acquisition phases (5.5 vs 3), whereas the difference in the number of CT examinations was less pronounced (4.5 vs 4). Non-contrast chest CT accounted for 70.95% of all examinations.ConclusionRepeated and multiphase CT imaging resulted in substantial cumulative radiation exposure in a selected subgroup of hospitalised COVID-19 patients. Nearly one quarter of patients exceeded 100 mSv during a single hospitalisation. These findings support careful justification of repeated CT examinations, limitation of multiphasic acquisitions when clinically feasible, and systematic dose monitoring in patients requiring intensive imaging follow-up.Keywords: Computed tomography, Cumulative radiation dose, COVID-19, Dose optimization, Multiphase imaging Introduction Patient exposure to ionizing radiation from medical imaging, particularly computed tomography (CT), has long been a critical consideration in the planning of diagnostic procedures [1], [2]. Reported patient doses for similar CT examinations vary significantly across countries, institutions, and even within departments [3]. Globally, a growing body of evidence has established a link between ionizing radiation and an increased risk of carcinogenesis [4], [5], [6], [7], [8]. The global impact of the coronavirus disease 2019 (COVID−19) pandemic has raised additional concerns regarding the cumulative radiation dose from repeated imaging, especially in young and middle-aged patients with longer life expectancy [9]. The pandemic, which began its rapid global spread in December 2019, has affected over 200 countries [10]. In Bulgaria, the first confirmed COVID−19 case was reported in early March 2020, followed by a rapid increase in infections. To date, over 760 million cases and nearly 7 million deaths have been reported worldwide, emphasizing the critical need for effective diagnostic tools [11]. Due to limited sensitivity and delayed turnaround time of RT-PCR testing, chest CT imaging emerged early in the pandemic as a valuable modality for diagnosing, staging, and monitoring COVID−19 pneumonia [12]. CT was rapidly adopted as a frontline diagnostic tool in many regions, particularly in cases with atypical clinical presentations, rapid disease progression, or uncertain therapeutic response [13], [14], [15], [16], [17], [18], [19]. In some countries, CT remained the primary imaging modality during the early pandemic phase due to the urgency of clinical decision-making and challenges with RT-PCR access [18], [19], [20], [21]. The limited ability of RT-PCR to assess disease severity, pulmonary involvement, and temporal progression further contributed to the increased reliance on CT imaging in hospitalized and follow-up patients [22], [23], [24]. Typical radiological findings in COVID−19 include bilateral, peripheral ground-glass opacities and posterior consolidations, predominantly affecting the basal lung zones [25]. In response to the overwhelming number of cases, national health authorities implemented variable diagnostic and treatment protocols, often incorporating CT as a core imaging tool, particularly in moderate to severe disease [26], [27]. Multiple national and institutional guidelines recommended serial CT examinations to monitor therapeutic response and complications, contributing to a significant rise in cumulative patient dose [28], [29], [30], [31], [32]. Several studies have since explored CT utilization patterns in COVID−19 patients, addressing scan frequency, protocol optimization, dose metrics, image quality, and risk-benefit balance [9], [33], [34]. The increased use of CT during the pandemic has justifiably raised concern regarding potential long-term health consequences, particularly the radiation-induced carcinogenic risk in younger patients who underwent repeated examinations [9]. This underscores the importance of applying dose-reduction strategies, including careful anatomical targeting, lower tube current and voltage settings, higher pitch values, and the use of iterative reconstruction algorithms [25], [35]. At present, there is limited knowledge about the long-term consequences of the pandemic, including those related to radiation exposure from diagnostic imaging. Considering the young age of many patients and their higher lifetime risk from stochastic radiation effects, the question of increased neoplastic risk becomes particularly relevant. Adhering to the ALARA (As Low As Reasonably Achievable) principle, CT imaging should be performed judiciously, with strict attention to clinical justification, protocol optimization, and dose monitoring [36]. Importantly, the experience from the COVID−19 pandemic offers valuable lessons for future public health crises. The implementation of structured imaging protocols, real-time dose tracking systems, and evidence-based guidance for imaging utilization during pandemics should become an integral part of emergency preparedness planning. This will help ensure that diagnostic effectiveness is balanced against long-term safety, particularly for vulnerable populations. Although several studies have evaluated CT utilisation patterns, protocol optimisation and radiation dose metrics during the COVID−19 pandemic, most have focused primarily on individual CT examinations or cumulative exposure over longer observational periods. Data regarding short-term cumulative radiation exposure accumulated during a single hospitalisation in patients requiring repeated CT imaging remain limited. In particular, the potential for substantial cumulative dose accumulation within acute clinical management pathways has not been sufficiently characterised. Purpose The purpose of this study was to assess short-term cumulative radiation exposure associated with repeated CT imaging during a single hospitalisation in COVID−19 patients, to evaluate the frequency and characteristics of repeated imaging, and to identify factors associated with higher cumulative dose. Methods and materials This study employed a retrospective, observational design to evaluate the cumulative radiation exposure in patients diagnosed with COVID−19 who underwent multiple CT examinations during their hospitalization. The analysis was performed using clinical and imaging data from 866 patients treated at one of Bulgaria’s largest COVID−19 referral centers over a two-year period (April 2020 – October 2022). All CT examinations were performed using a Toshiba/Canon Aquilion 64-slice MDCT scanner (installed in 2009; software version V3.35ER004). Study cohort and inclusion criteria From the initial cohort of 866 patients, a subgroup of 69 patients was identified based on the following inclusion criteria: 1) age ≥ 18 years; 2) laboratory-confirmed COVID−19 diagnosis (positive RT-PCR); 3) performed three or more chest CT scans per patient during hospitalization period and 4) CT imaging performed for purposes of diagnosis, assessment of disease progression, therapeutic response, or complications. Patients with fewer than three CT examinations or with CT findings unrelated to COVID−19 (e.g., malignancy, tuberculosis) were excluded. Data collection For each patient, the following parameters were collected: demographic data (age, sex); number and type of CT examinations; CT phases (native or contrast-enhanced, single or multiphase); the displayed information for the CT dose indices, volume CT dose index (CTDIvol, mGy) and dose-length product (DLP, mGy.cm), and scan indication and justification. The effective dose (E, mSv) was estimated using published region-specific conversion coefficients according to the formula E = DLP × k, where k represents the anatomical region-specific conversion coefficient expressed in mSv/mGy.cm [38], [39], [40]. The following conversion coefficients in mSv/mGy.cm were applied in the analysis: 0.027 for chest and CTPA; 0.024 for abdomen; 0.020 for abdomen and pelvis; 0.0255 for chest and abdomen; 0.021 for chest-abdomen-pelvis; 0.020 for pelvis; 0.002 for head and 0.024 for three-phase liver CT. The applied coefficients were based on ICRP 103-related effective dose estimates [41]. Each acquisition phase, including native and contrast-enhanced phases, was analysed individually, and cumulative effective dose (CED) was calculated as the sum of all phase-specific effective doses for each patient. All CTDIvol values were obtained using a 32 cm PMMA phantom according to standardized protocol reporting. DLP values were recorded for each individual phase, including both native and contrast-enhanced acquisitions. Because automatic tube current modulation was used, a single fixed tube current value was not available or representative for each acquisition; therefore, the analysis relied on recorded CTDIvol and DLP values from the scanner dose reports. CT protocol exposure parameters, including tube voltage, pitch factor, scanned anatomical region, use of tube current modulation, and reconstruction approach, were documented where available to assess inter-examination variability and to identify potential areas for protocol optimization. Chest CT was performed according to clinical indications. CT pulmonary angiography (CTPA) was performed in cases of suspected pulmonary embolism, while additional abdominal–pelvic or head CT examinations were performed when extra-thoracic complications were suspected. Three non-contrast chest CT protocol variants were used (routine, optimized low-dose, and high-dose for obese patients), with protocol selection based on clinical and patient-related factors. CT dose indices (CTDIvol and DLP) were summarized, defined by CT examination type and patient age group. For Toshiba CT scanners running software version 4.63 or earlier (typically systems installed before 2013), the dose report displays the maximum CTDIvol rather than the average CTDIvol reported by other vendors [37]. In protocols using tube current modulation (TCM), this can overestimate CTDIvol and may bias comparisons with published reference values. Importantly, this type of reporting does not affect the DLP value, which is based on the average CTDIvol (together with scan length). When SUREExposure tube current modulation is used, CTDIvol is reported based on the maximum tube current used in the exam, DLP is reported based on the average tube current. Therefore, CTDIvol values from older Toshiba software versions, which applies to the CT scanner used in the present study (software version V3.35ER004), should be interpreted with caution, and comparisons across scanners or DRLs should preferably rely on DLP when this issue is present. Because retrospective access to raw tube current modulation data was not available, the quantitative magnitude of CTDIvol overestimation could not be determined in the present study. Statistical analysis was performed using IBM SPSS Statistics version 17. Continuous variables were tested for normality using the Kolmogorov–Smirnov test and demonstrated non-normal distribution; therefore, non-parametric statistical methods were applied. Continuous variables were summarized as median values with interquartile range (IQR) and range. Spearman’s rank correlation coefficient (Spearman’s rho) was used to assess associations between cumulative effective dose, number of CT examinations, and number of acquisition phases, as these variables were non-normally distributed and the study cohort was relatively small. A two-sided p-value < 0.05 was considered statistically significant. Comparisons between patient subgroups were performed using the Mann–Whitney U test for continuous non-normally distributed variables. Associations between categorical variables were evaluated using the Chi-square test or Fisher’s exact test, as appropriate. Results A total of 69 patients met the inclusion criteria and were included in the final analysis. Of these, 40 were male and 29 were female, with a mean age of 63 years (range: 36–88 years). All patients had a laboratory-confirmed diagnosis of COVID−19 and underwent three or more chest CT scans during their hospitalization. The hospital length of stay among this subgroup ranged from 6 to 75 days, with an average of 25 days. From the initial cohort of 866 COVID−19-positive patients, this subgroup was identified based on the need for repeated CT imaging, including additional examinations of the abdomen, pelvis, head, CT pulmonary angiography (CTPA), or CT coronary angiography (CTCA). CT examination distribution In total, 296 CT examinations comprising 350 acquisition phases were analyzed. The most frequently performed imaging modality was non-contrast chest CT, accounting for 210 scans (70.95%). Table 1 summarizes the distribution of CT types. Multiphase acquisitions were predominantly observed in contrast-enhanced abdominal–pelvic CT and CTPA protocols. Table 1CT examination types and frequencies.CT TypeNumber of CT examsNumber of CT phasesPercentage (%)Non-contrast chest CT21021070.95CT pulmonary angiography (CTPA)10203.38Non-contrast head CT13134.39Contrast-enhanced chest CT7142.36Contrast-enhanced chest& abdomen CT260.68Non-contrast abdominal CT13134.39Non-contrast abdomen&pelvis CT24248.11Contrast-enhanced abdomen&pelvis14424.73CT coronary angiography120.34Contrast-enhanced head CT260.68Total296350100.00 CT examination types and frequencies. CT protocols and dose indices Among the chest CT examinations, three protocol types were identified: a routine protocol, an optimized low-dose protocol and a high-dose protocol for obese patients. The choice of protocol was selectively applied depending on clinical indications, patient weight and radiologist discretion. All non-contrast chest CT examinations were performed using single-phase inspiratory acquisitions. Multiphase acquisitions were predominantly observed in contrast-enhanced abdominal–pelvic CT and CT pulmonary angiography protocols. The routine chest was used for the majority of non-contrast chest CTs (Table 2). Of the 210 non-contrast chest CT examinations, 25 (12%) were performed using the optimized low-dose protocol, 171 (82%) using the routine protocol, and 14 (6%) using the high-dose (135 kV) protocol applied in obese patients. Across all analyzed acquisition phases (n = 350), the median DLP value was 525 mGy.cm (IQR: 365–758), and the median CTDIvol was 21.1 mGy (IQR: 13.2–30.6). The optimized low-dose protocol, although less frequently applied, resulted in lower dose values with 10.7 mGy (IQR: 9.4–12.1) CTDIvol and 367 mGy.cm (IQR: 316–418) DLP. The high-dose protocol for obese patients was leading to higher doses with 41.6 mGy (IQR: 39.3–42.6) CTDIvol and 1220 mGy.cm (IQR: 1012–1316) DLP. Table 2Exposure parameters and dose indices by Chest CT protocol for a single phase(median values with IQR).Protocol Type (% of non contrast chest)Median Age (y)Tube Voltage (kV)TCMPitchCTDIvol (IQR), mGyDLP (IQR), mGy.cmRoutine [82]60120Yes0.82821.1 (13.2–30.6)525 (365−758)Optimized [12]61100Yes1.10910.7 (9.4–12.1)367 (316–418)High-dose [6]46135Yes0.82841.6 (39.3–42.6)1220 (1012–1316)Note: The CT Software version is earlier than v4.63 (V3.35ER004) and the displayed CTDIvol values are the maximum CTDIvol (not average), potentially overestimating CTDIvol in TCM protocols; DLP is not affected. Exposure parameters and dose indices by Chest CT protocol for a single phase(median values with IQR). Note: The CT Software version is earlier than v4.63 (V3.35ER004) and the displayed CTDIvol values are the maximum CTDIvol (not average), potentially overestimating CTDIvol in TCM protocols; DLP is not affected. The optimized low-dose protocol demonstrated substantially lower CTDIvol and DLP values compared with the routine protocol, whereas the high-dose protocol used in obese patients resulted in markedly increased dose indices. These findings illustrate the considerable influence of protocol selection and patient-related factors on radiation exposure during repeated CT imaging. Cumulative effective dose The median cumulative effective dose (CED) was 47.7 mSv (IQR 34.5–87.2 mSv), ranging from 18.0 to 342.6 mSv. Nearly half of the patients (32/69; 46.4%) accumulated ≥ 50 mSv, while 16 patients (23.2%) exceeded 100 mSv during a single hospitalization (Table 3). One patient (1.4%) reached a cumulative exposure ≥ 200 mSv. Table 3Distribution of cumulative effective dose (CED).Interval (mSv)n%< 2557.2%25–49.93246.4%50–74.968.7%75–99.91014.5%100–199.91521.7%≥ 20011.4% Distribution of cumulative effective dose (CED). CED demonstrated a moderate correlation with the number of CT examinations (ρ=0.37, p = 0.0016) and a stronger correlation with the total number of acquisition phases (ρ=0.54, p < 0.00001). Patients who accumulated ≥ 100 mSv had a higher median number of acquisition phases (5.5 vs 3), whereas the difference in the number of CT examinations was less pronounced (4.5 vs 4), suggesting an important contribution of multiphasic imaging to cumulative exposure in these patients. Patients with clinical complications demonstrated significantly longer hospitalisation, higher number of CT examinations, higher cumulative effective dose and greater number of acquisition phases (p < 0.05 for all comparisons). No statistically significant association was observed between cumulative effective dose and assisted ventilation requirement. The observed cumulative exposure levels, particularly in patients exceeding 100 mSv during a single hospitalisation, highlight the importance of careful justification and optimisation of repeated CT imaging in patients requiring intensive follow-up. An illustrative example of a patient with very high cumulative exposure is presented in Fig. 1. A 58-year-old female patient underwent multiple CT examinations over a 49-day hospitalization period, including repeated non-contrast chest CT scans, contrast-enhanced CT of the abdomen and pelvis (arterial and venous phases), and two CT pulmonary angiography (CTPA) studies. This case illustrates how repeated and multiphasic imaging performed for clinical complications may lead to substantial cumulative radiation exposure within a short time interval. Example of a patient with high cumulative radiation exposure (>200 mSv): A) Axial chest CT image demonstrating typical subacute COVID−19 findings with bilateral, subpleural ground-glass opacities and crazy paving pattern. B) Axial chest CT image obtained 49 days after hospitalization, demonstrating partial resolution of the infectious lung changes. C) Contrast-enhanced CT images of the abdomen and pelvis showing multiple retroperitoneal hematomas detected during follow-up imaging. Discussion During the COVID−19 pandemic, variability in national and institutional guidelines led to differing approaches regarding the use of chest CT in patients with suspected or confirmed infection. In some settings, CT was used beyond its conventional indications, contributing to a substantial increase in imaging volume. Although a single CT examination is generally sufficient to assess pulmonary involvement, repeated chest CT scans may occur due to reinfection, clinical deterioration, complications, or prolonged therapy [42]. In our cohort, 69 out of 866 hospitalized COVID−19 patients underwent three or more chest CT examinations during a single hospitalization, reflecting the intensity of imaging required in selected cases with a complicated clinical course. International experience further illustrates this trend. A multi-center retrospective study in Turkey demonstrated a marked increase in chest CT utilization during the pandemic, with examinations for infectious indications increasing approximately five-fold in 2020 compared with the preceding year, and repeated imaging rising more than fourfold [34]. Similarly, an interrupted time series analysis from Fars province, Iran, showed that the average monthly number of chest CT scans increased from approximately 326 before the pandemic to over 3000 after its onset [43]. These findings demonstrate how public health pressures may rapidly escalate imaging demand. In addition, multinational data have documented considerable variability in CT utilization, protocols, and radiation doses across institutions during the early pandemic phase [8]. Such variability highlights the importance of maintaining optimized imaging pathways and consistent radiation protection strategies, particularly during healthcare crises. Persistent radiological abnormalities after COVID−19 infection provide a clinical rationale for follow-up imaging in selected patients. Zhao et al. demonstrated higher rates of residual CT findings in patients with severe disease compared with moderate cases [19]. Similarly, Han et al. reported fibrotic-like changes in more than one-third of patients at 6-month follow-up after severe COVID−19 pneumonia, suggesting sustained structural lung alterations [44]. Long-term evaluation has further shown that radiological abnormalities may persist at 1-year follow-up in a substantial proportion of survivors, particularly in those with severe initial disease [45]. These findings support the selective use of follow-up CT in patients with prolonged or severe disease course. Repeated CT examinations inevitably increase cumulative radiation exposure. Although individual chest CT doses may remain within accepted diagnostic ranges, multiple examinations performed within a limited time interval can lead to substantial cumulative effective dose. This consideration is particularly relevant in younger patients, who have a longer post-exposure lifetime during which radiation-related effects may potentially manifest, and for whom radiation protection guidance emphasises particular caution because lifetime attributable risk is age-dependent [49], [50], [51], [52]. In the present cohort, patients with clinical complications demonstrated significantly higher cumulative effective dose, greater number of acquisition phases and longer hospitalisation, supporting the interpretation that increased radiation exposure was associated with more complex clinical management and intensified imaging follow-up. These findings reinforce the importance of applying radiation protection principles such as ALARA in scenarios requiring repeated CT imaging [36]. Although diagnostic reference levels (DRLs) are primarily intended for optimisation of individual examinations rather than cumulative patient exposure assessment, repeated and multiphasic imaging may lead to substantial cumulative dose accumulation despite acceptable dose levels for individual CT examinations [53]. In our study, cumulative exposure reached high levels within a single hospitalization. Nearly one quarter of patients (23.2%) exceeded 100 mSv, and one patient reached ≥ 200 mSv. Although effective dose is not intended for individual risk prediction and should be interpreted as a radiological protection quantity rather than a patient-specific risk estimate [41], cumulative exposure levels exceeding 50 mSv and particularly 100 mSv within a relatively short period are generally considered relevant from a radiation protection perspective because they indicate substantial accumulated exposure from diagnostic imaging [41], [54]. Such exposure levels reinforce the importance of careful justification, optimisation and dose monitoring, particularly in patients requiring repeated follow-up CT examinations. While previous reports have highlighted increased CT utilization and per-examination dose variability during the pandemic [8], [12], [34], [42], most analyses focused on individual scan metrics rather than cumulative patient-level exposure. Our findings demonstrated a stronger correlation between cumulative dose and the total number of acquisition phases than with the absolute number of CT examinations. However, this observation should be interpreted cautiously, since cumulative effective dose is inherently derived from the sum of phase-specific radiation exposures. This interpretation is consistent with previous CT dose literature showing that multiphasic CT protocols increase total examination dose because each additional acquisition phase contributes additional DLP and effective dose, and that reducing unnecessary phases is an established dose-optimisation strategy [9], [22], [55]. Тhe relative contribution of each protocol type could not be isolated in the present analysis. Because multiple CT protocol types were frequently combined within the same hospitalisation, protocol-specific contribution to total cumulative dose could not be reliably quantified at patient level. These findings add patient-level short-term cumulative exposure data to the existing literature on CT utilisation and dose during the COVID−19 pandemic. In comparison with published cumulative dose data, the exposure pattern observed in our cohort appears substantially different in magnitude and temporal concentration. Differences in effective dose estimation methods should also be considered when comparing cumulative dose data across studies, as published reports may use different anatomical conversion coefficients, phantom assumptions or ICRP-based weighting factors. In a study evaluating multiple low-dose CT examinations in COVID−19 patients, Bai et al. reported a mean cumulative effective dose of 5.25 mSv, with a maximum of 7.48 mSv after up to six CT examinations [46]. In contrast, the median CED in our cohort was 47.7 mSv, and doses exceeded 100 mSv in 23.2% of patients, indicating a substantially higher cumulative dose. Similarly, Moghadam et al. reported that only 0.67% of patients reached a CED ≥ 100 mSv within one year, and 3.5% exceeded 50 mSv in a large regional hospital network [47]. In our cohort, 46.4% accumulated ≥ 50 mSv and 23.2% exceeded 100 mSv during a single hospitalization, suggesting a substantially higher short-term cumulative exposure compared with general CT populations. Longitudinal data from Ferretti et al., analyzing more than 400,000 patients over an 11-year period, demonstrated that 6.6% of patients accumulated ≥ 100 mSv [48]. In contrast, nearly one-quarter of patients in our study exceeded this threshold within weeks rather than years, reflecting an accelerated cumulative exposure model driven by intensive multiphasic imaging during acute disease management. The illustrative case presented in Fig. 1 further demonstrates how prolonged hospitalisation, repeated follow-up imaging, multiphasic abdominal CT examinations and repeated CTPA studies performed for clinical complications may collectively result in very high cumulative radiation exposure within a relatively short time interval. However, differences in patient selection, disease severity, and imaging protocols should be considered when interpreting cross-study comparisons. In particular, the present cohort consisted exclusively of hospitalised patients undergoing repeated CT imaging during acute disease management, including multiphasic contrast-enhanced examinations performed for suspected complications. This differs substantially from broader population-based CT utilisation studies and likely contributed to the higher cumulative exposure levels observed in the present analysis. Limitations First, the analysed cohort included only 69 patients who underwent three or more CT examinations during a single hospitalisation. Therefore, the results should be interpreted as reflecting a selected subgroup of hospitalised COVID−19 patients requiring repeated imaging, rather than the overall COVID−19 inpatient population. Consequently, the reported cumulative exposure values should not be interpreted as representative of the average cumulative radiation exposure among all hospitalised COVID−19 patients, but rather as characterising patients with more intensive imaging needs. The limited sample size also reduces statistical power, particularly in subgroup analyses such as patients exceeding 100 mSv. For this reason, subgroup-related findings should be considered descriptive and exploratory. In addition, the observed correlations between cumulative effective dose and imaging frequency or acquisition phases should be interpreted in the context of the mathematical structure of cumulative dose calculation and do not necessarily represent independent associations. Second, the analysed CT examinations were heterogeneous with respect to anatomical region, contrast materia use and number of acquisition phases. Although this heterogeneity reflects real clinical practice in patients with complicated disease course, pooling all examinations into a single cumulative effective dose metric may obscure the relative contribution of individual CT protocols or anatomical regions. Therefore, CED was interpreted as a patient-level cumulative exposure metric rather than a protocol-specific dose indicator. Third, the study had a retrospective single-centre design and included a relatively limited number of patients, which may restrict the generalisability of the findings. Organ-specific radiation doses were not estimated, as dose assessment was based on effective dose calculations derived from DLP conversion coefficients. In addition, long-term clinical outcomes and potential radiation-related effects were not evaluated, since the study was focused on cumulative exposure during acute hospitalisation. Detailed phase-specific tube current modulation data were not available retrospectively for all examinations; therefore, the direct contribution of tube current variation to CTDIvol, DLP and effective dose could not be quantified. Conclusion Repeated CT imaging in this selected subgroup of hospitalised COVID−19 patients resulted in substantial short-term cumulative radiation exposure, with nearly one quarter of patients exceeding 100 mSv during a single hospitalisation. Higher cumulative exposure was observed in patients undergoing repeated and multiphasic CT imaging, particularly in the setting of clinical complications and prolonged hospitalisation. Although these findings should be interpreted cautiously because of the cumulative nature of CED calculation and the exploratory study design, they highlight the importance of careful justification of repeated CT examinations, limitation of multiphasic acquisitions when clinically feasible, protocol optimisation and systematic dose monitoring in patients requiring intensive imaging follow-up. Statement for studies in humansanimals This study used anonymised data collected as part of routine clinical practice. Written informed consent was obtained from all participants. According to institutional policies, formal ethical approval was not required. CRediT authorship contribution statement Dora Zlatareva: Writing – review & editing, Visualization, Supervision. Stefan Stanev: Writing – review & editing, Validation, Software, Formal analysis, Data curation. Desislava Kostova-Lefterova: Writing – review & editing, Writing – original draft, Validation, Supervision, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Marieta Mateva-Mladenova: Writing – review & editing, Writing – original draft, Validation, Resources, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Declaration of Generative AI and AI-assisted technologies in the writing process During the preparation of this work, the authors used ChatGPT (OpenAI) to assist with language editing. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. References