Interdisciplinary perspectives on computed tomography in sepsis: survey among medical doctors at a large university medical center

Abstract

Objectives: This study aims to describe physicians' perspectives on the use of computed tomography (CT) in patients with sepsis.

Methods: In January 2022, physicians of a large European university medical center were surveyed using a web-based questionnaire asking about their views on the role of CT in sepsis. A total of 371 questionnaires met the inclusion criteria and were analyzed using work experience, workplace, and medical specialty of physicians as variables. Chi-square tests were performed.

Results: Physicians considered the ability to detect an unknown focus as the greatest benefit of CT scans in sepsis (70.9%, n = 263/371). Two clinical criteria - "signs of decreased vigilance" (89.2%, n = 331/371) and "increased catecholamine demand" (84.7%, n = 314/371) - were considered highly relevant for a CT request. Elevated procalcitonin (82.7%, n = 307/371) and lactate levels (83.6%, n = 310/371) were consistently found to be critical laboratory values to request a CT. As long as there is evidence of infection in one organ region, most physicians (42.6%, n = 158/371) would order a CT scan based on clinical assessment. Combined examination of the chest, abdomen, and pelvis was favored (34.8%, n = 129/371) in cases without clinical clues of an infection source. A time window of ≥ 1-6 h was preferred for both CT examinations (53.9%, n = 200/371) and CT-guided interventions (59.3%, n = 220/371) in patients with sepsis.

Conclusion: Despite much consensus, there are significant differences in attitudes towards the use of CT in septic patients among physicians from different workplaces and medical specialties. Knowledge of these perspectives may improve patient management and interprofessional communication.

Key points: Despite interdisciplinary consensus on the use of CT in sepsis, statistically significant differences in the responses are apparent among physicians from different workplaces and medical specialties. The detection of a previously unknown source of infection and the ability to plan interventions and/or surgery based on CT findings are considered key advantages of CT in septic patients. Timing of CT reflects the requirements of specific disciplines.

Keywords: Clinical reasoning; Physicians; Sepsis; Surveys and questionnaires; Tomography, X-ray computed.

Fig. 1

Flow chart of survey participation and questionnaire inclusion and exclusion. Of the 2502 physicians contacted, 575 participated in the survey. After excluding 204 questionnaires, 371 met the inclusion criteria and were included in the analysis

Fig. 2

Arguments in favor of a CT scan in sepsis — why should a patient be examined by CT? Overview of results for whether physicians believe the listed options to represent a major benefit of CT examinations in septic patients. With 97.0% (n = 360/371, “strongly agree” and “somewhat agree” counted together), the majority of participants agreed that the ability of CT to detect an unknown focus is a great benefit, followed by 96.8% (n = 359/371) of respondents seeing a great advantage of CT in planning interventions and/or surgeries. CT to confirm a suspected focus was considered a major advantage by 73.6% (n = 273/371). 55.0% (n = 204/371) of participants considered CT to be of great benefit as a diagnostic tool for exclusion. The modification of anti-infective therapy after a CT scan was the option most disagreed with: 51.7% (n = 192/371) of physicians ticked the options “strongly disagree” or “somewhat disagree.” CT, computed tomography

Fig. 3

Descriptive results for physicians’ responses regarding whether they considered the listed clinical criteria to support a CT request in patients with sepsis. Except for the “elderly patient” criterion, each clinical criterion listed was considered by more than 50% of the participants as a reasonable argument in favor of a CT scan in patients with sepsis. The criterion that most physicians selected to support a CT request was “signs of decreased vigilance” (89.2%, n = 331/371), followed by “increased catecholamine demand” (84.7%, n = 314/371). A critical qSOFA and a critical SOFA score were similarly classified. Whereas a qSOFA score with ≥ 2 applicable criteria was considered relevant by 76.9% (n = 285/371) of participants, a SOFA score increased by ≥ 2 points was considered relevant by 80.0% (n = 297/371). At 55.5% (n = 206/371), the SIRS criteria were most frequently classified as only somewhat relevant. Significant levels (p-values) for response patterns in relation to work experience, workplace, and medical specialty are compiled in Table 1. Detailed frequency data for the analyzed variables can be found in the supplementary material. CT, computed tomography; qSOFA, Quick Systemic Organ Failure Assessment; SOFA, Systemic Organ Failure Assessment; SBP, systolic blood pressure; MAP, mean arterial pressure; SIRS, Systemic Inflammatory Response Syndrome

Fig. 4

Descriptive results for physicians’ responses regarding whether they considered the listed ancillary criteria to support a CT request in patients with sepsis. Although a sonographically suspected focus of infection was considered an indication for a CT scan by more participants (83.3%, n = 309/371) than an abnormal chest x-ray (64.9%, n = 241/371), a general desire for confirmation of previous imaging findings by CT is apparent. More than half of physicians would also request CT in patients with evidence of SARS-CoV-2 (72.2%, n = 268/371) and a positive blood culture (66.6%, n = 247/371). Of the laboratory parameters listed, elevated PCT (82.7%, n=307/371) and elevated lactate (83.6%, n=310/371) were considered the strongest arguments for requesting a CT. The leukocyte count (76.2%, n = 282/371) and elevated CRP (70.9%, n = 263/371) were also selected by physicians as reasons for a CT request. Conversely, the criterion “elevated IL-6” was not considered a criterion supporting a CT indication by 58% (n = 215/371). No difference was found in the response pattern for “sonographically suspected focus of infection” and “SARS-CoV-2 detected.” Regardless of the variables examined, all physicians marked these parameters as supportive of a CT scan. Significant levels (p-values) for response patterns in relation to work experience, workplace, and medical specialty are stated in Table 1. Detailed frequency data for the analyzed variables can be found in the supplementary material. CT, computed tomography; PCT, procalcitonin; SARS-CoV-2, severe acute respiratory syndrome coronavirus type 2; CRP, C-reactive protein; IL-6, interleukin-6

Fig. 5

Time window for CT examinations and CT-guided interventions in patients with sepsis. The ≥ 1–6 h time window was most frequently selected by participants for both CT examinations for suspected sepsis (53.9%, n = 200/371) and CT-guided interventions (59.3%, n = 220/371). While the second most frequently chosen time window for a CT scan was <1h (36.7%, n = 136/371), it was the ≥ 6–12h time window for CT-guided interventions with 25.6% (n = 95/371). Conversely, only 9.7% (n = 36/371) of physicians indicated a preference for CT-guided interventions within 1 hour, and 5.9% (n = 22/371) for CT examinations within 6 to 12 hours. The time window ≥ 12–24 h for both a CT scan (3.5%, n = 13/371) and a CT-guided intervention (5.4%, n = 20/371) was the least frequently ticked option. CT, computed tomography; h, hour/hours

Fig. 6

a Time window for CT interventions and effect of medical specialty. Significant differences (p < 0.001, chi-square test) were found between radiologists and each of the other four medical specialties. While the other medical specialties preferred the time window of ≥ 1–6 h, radiologists mainly chose the time window of ≥ 6–12 h with 42.4% (n = 14/33). While the option of ≥ 12–24 h was chosen by 27.3% (n = 9/33) of radiologists, surgeons were the group most likely to choose this option among the other medical specialties, but at only 4.5% (n = 2/44). The time window of < 1h was not selected once by radiologists. Conversely, the option to have a CT intervention within 1 hour was chosen by 18.2% (n = 8/44) of surgeons. b Time window for CT-guided interventions and effect of the workplace. Significant differences (p < 0.001, chi-square test) were found between the radiology department and each of the other workplaces. The majority of physicians preferred a time window of ≥ 1–6 h for a CT-guided intervention. Radiologists, on the other hand, preferred the ≥ 6–12 h time window (43.8%, n = 14/32). While the option of ≥ 12–24 h was chosen by 28.1% (n = 9/32) of radiologists, only one physician each from the emergency department (2.0%), OR (2.4%), outpatient clinic (3.1%), and other (5.6%) chose this option. Only radiologists never selected the time window of < 1 hour even once. CT, computed tomography; ICU, intensive care unit; OR, operating room; h, hour/hours