Radiation Exposure From Diagnostic Imaging in PICU
Radiation Exposure From Diagnostic Imaging in PICU
We enrolled 85 patients from October 2009 to April 2010. Parents of ten patients declined to participate for reasons including concern that the dosimeter might cause skin irritation, disinterest in clinical research, or because their children were chronically ventilated. Sixteen dosimeters were not available for analysis; two were discarded in the operating room during procedures and 14 were lost before discharge from the PICU. There were 69 dosimeters available for analysis. There were no complications during the study. Clinical characteristics of the subjects are described in Table 1. Of our subjects, 26% were obese (body mass index greater than the 95th percentile) and 13% were overweight (body mass index greater than the 85th percentile)
A total of 1195 radiologic studies were performed on these subjects during their PICU stay with most subjects undergoing α 20 studies. Diagnostic studies included CXRs (n = 992), other plain radiographs (n = 136), chest CT scans (n = 9 in six patients), and other CT scans (n = 35 in 19 patients). Twenty patients (29%) had one or more CT scans. These patients experienced a range of tube currents (65–356 mA) and driving forces (100–120 kVp). Twenty-three fluoroscopic studies were obtained in 11 patients with fluoroscopy time of 7 ± 6 (range, 0.1–20) minutes. Twelve of the fluoroscopic studies were for central venous access. Patients had on average (mean ± SD) 14 ± 16 CXRs. Forty-seven subjects had only plain radiographic studies (CXR group), and the remaining 22 patients underwent plain radiographs as well as other imaging modalities (CXR+ group).
Subjects in the CXR group had a median thoracic exposure of 1.02 (range, 0.13–28.26) mGy, during their PICU stay, and a median daily exposure of 0.16 (range, 0.02–1.99) mGy/day (Fig. 1A–B). The median exposure per study for patients who had only plain radiographs was 0.12 (range, 0.03–1.32) mGy/study. For this group, there was no significant difference in median daily exposures according to admission diagnosis (p = .133). Using multiple linear regression, daily exposure (mGy/day) could not be predicted from a linear combination of variables, including age, body mass index, gender, or length of stay (p = .498).
(Enlarge Image)
Figure 1.
Total (A) in mGy and average daily (B) in mGy/day thoracic radiation exposure in subjects with only plain chest radiographs (CXR group, open circles) or subjects with additional imaging studies (CXR+ group, solid circles). Subjects in the CXR+ group had significantly higher total and daily exposures compared with subjects in the CXR group. Median and interquartile ranges are shown.
Subjects in the CXR+ group had a median total thoracic exposure of 3.71 (range, 0.77–33.41) mGy and median daily exposure of 0.37 (range, 0.04–3.71) mGy/day (Fig. 1A–B), both of which were significantly higher than for subjects in the CXR group (p < .001 for both comparisons). Again, there was no significant difference in average daily exposures according to admission diagnoses (p = .137). Similarly, using multiple linear regression, daily exposure (mGy/day) could not be predicted from a linear combination of variables, including age, body mass index, gender, or length of stay (p = .638).
For all subjects (both CXR and CXR+ groups combined), average daily exposure was weakly correlated with weight (r = 0.470, p < .001) and age (r = 0.382, p = .001) but was not correlated with total number of studies (r = −0.006, p = .956). Total number of radiologic studies was correlated, as expected, with duration of ventilation (r = 0.941, p < .0001).
Results
Subjects
We enrolled 85 patients from October 2009 to April 2010. Parents of ten patients declined to participate for reasons including concern that the dosimeter might cause skin irritation, disinterest in clinical research, or because their children were chronically ventilated. Sixteen dosimeters were not available for analysis; two were discarded in the operating room during procedures and 14 were lost before discharge from the PICU. There were 69 dosimeters available for analysis. There were no complications during the study. Clinical characteristics of the subjects are described in Table 1. Of our subjects, 26% were obese (body mass index greater than the 95th percentile) and 13% were overweight (body mass index greater than the 85th percentile)
Diagnostic Imaging Studies
A total of 1195 radiologic studies were performed on these subjects during their PICU stay with most subjects undergoing α 20 studies. Diagnostic studies included CXRs (n = 992), other plain radiographs (n = 136), chest CT scans (n = 9 in six patients), and other CT scans (n = 35 in 19 patients). Twenty patients (29%) had one or more CT scans. These patients experienced a range of tube currents (65–356 mA) and driving forces (100–120 kVp). Twenty-three fluoroscopic studies were obtained in 11 patients with fluoroscopy time of 7 ± 6 (range, 0.1–20) minutes. Twelve of the fluoroscopic studies were for central venous access. Patients had on average (mean ± SD) 14 ± 16 CXRs. Forty-seven subjects had only plain radiographic studies (CXR group), and the remaining 22 patients underwent plain radiographs as well as other imaging modalities (CXR+ group).
Exposure
Subjects in the CXR group had a median thoracic exposure of 1.02 (range, 0.13–28.26) mGy, during their PICU stay, and a median daily exposure of 0.16 (range, 0.02–1.99) mGy/day (Fig. 1A–B). The median exposure per study for patients who had only plain radiographs was 0.12 (range, 0.03–1.32) mGy/study. For this group, there was no significant difference in median daily exposures according to admission diagnosis (p = .133). Using multiple linear regression, daily exposure (mGy/day) could not be predicted from a linear combination of variables, including age, body mass index, gender, or length of stay (p = .498).
(Enlarge Image)
Figure 1.
Total (A) in mGy and average daily (B) in mGy/day thoracic radiation exposure in subjects with only plain chest radiographs (CXR group, open circles) or subjects with additional imaging studies (CXR+ group, solid circles). Subjects in the CXR+ group had significantly higher total and daily exposures compared with subjects in the CXR group. Median and interquartile ranges are shown.
Subjects in the CXR+ group had a median total thoracic exposure of 3.71 (range, 0.77–33.41) mGy and median daily exposure of 0.37 (range, 0.04–3.71) mGy/day (Fig. 1A–B), both of which were significantly higher than for subjects in the CXR group (p < .001 for both comparisons). Again, there was no significant difference in average daily exposures according to admission diagnoses (p = .137). Similarly, using multiple linear regression, daily exposure (mGy/day) could not be predicted from a linear combination of variables, including age, body mass index, gender, or length of stay (p = .638).
For all subjects (both CXR and CXR+ groups combined), average daily exposure was weakly correlated with weight (r = 0.470, p < .001) and age (r = 0.382, p = .001) but was not correlated with total number of studies (r = −0.006, p = .956). Total number of radiologic studies was correlated, as expected, with duration of ventilation (r = 0.941, p < .0001).
Source...