Send letter to journal:
Re: CT radiation dose

Dear Editor,

We read the paper by Jimenez et al with interest regarding the potential role of limited slice high resolution CT (HRCT) in children with cystic fibrosis (1). Studies such as this utilising fewer CT sections at greater intervals in children with diffuse lung disease warrant further evaluation. We certainly agree with the authors’ assertion that CT should be used judiciously in children and that technical parameters must be modified in accordance with the size of the child.

We believe however that Jimenez et al have not fully addressed the issues of radiation burden reduction in children. They have examined their patients with different CT scanners but in all patients used ‘100-130mA, with exposure times between 1 and 3 seconds, resulting in 100-300mAs’. Exposure times can be machine dependent, but are now much faster with newer multidetector CT scanners which should be widely available. The mA they have used on their patient group is unnecessarily high in our opinion. [The mA (milliAmperage) used is the major determinant of CT dose, together with, albeit to a lesser degree, the kV (kilovoltage) setting of the machine. Of note, mA multiplied by time, in seconds, equals mAs].

Examples of truly low dose HRCT techniques have been in use for some time with authors stating that acceptable image quality can be obtained using only 10 or 20mA, which is a fraction of the dose used by Jimenez et al in their study (2-5). Whilst such extremely low dose techniques may not be widely utilised, it is fair to state that lower mA studies are possible in every day practice even with older single slice helical CT scanners. There is a balance between diagnostic image quality [with sufficient signal] and image noise when extremely low mA doses are used. If the dose is too low the images may become too ‘noisy’ thus impairing the overall acceptability of the images, and rendering them non-diagnostic. We currently perform HRCT sections at a maximum of 30mAs in children of less than 35kg, and use no more than 55mAs in children weighing 35-54kg in our routine practice. To keep the radiation burden to a minimum the mA, and thus mAs also, should be set as low as reasonably achievable (ALARA principle) whilst maintaining diagnostic image quality.

Kieran McHugh, FRCR
Catherine Owens, FRCR.

Consultant Paediatric Radiologists
Great Ormond Street Hospital for Children
London
WC1N 3JH

References:

1. Jimenez S, Jimenez JR, Crespo M, et al. Computed tomography in children with cystic fibrosis: a new way to reduce the radiation dose. Arch Dis Child 2006;91:388-390.

2. Zwirewich CV, Mayo JR, Muller NL. Low-dose high-resolution CT of lung parenchyma. Radiology 1991;180:413-417.

3. Naidich DP, Marshall CH, Gribbin C, et al. Low-dose CT of the lungs: preliminary observations. Radiology 1990;175:729-731.

4. Ambrosino MM, Genieser NB, Roche KJ, et al. Feasibility of high- resolution low-dose chest CT in evaluating the pediatric chest. Pediatr Radiol 1994;24:6-10.

Send letter to journal:
Re: CXR and HRCT in cystic fibrosis.

Dear Editor,

We read with interest the article by Jiménez et al reporting the use of low dose, high resolution computer tomography (HRCT) in children with cystic fibrosis (CF) and the excellent commentary from Langton-Hewer in the May issue. We agree with Langton-Hewer’s conclusion that whilst HRCT’s provide an earlier and detailed insight into progression of CF related lung disease, they offer limited advantages to patient management in the current therapeutic environment.

We have carried out a study that supports these conclusions. We undertook a prospective, blinded comparison of ultra low dose HRCT, chest xray (CXR) and spirometry in 30 children (aged 5.02-17.35 years, 13 male) with CF when free of acute exacerbation. Scans were performed at 100 kv and 30 mAS. One mm slice thicknesses at 10 mm intervals were taken. These were presented for scoring as hard copy images with either the whole image dataset or alternate images (equalling to 10 mm or 20 mm slice intervals). The chest x-rays and HRCT’s were scored independently by 2 consultant paediatric radiologists using Mafessanti and Bhalla (HRCT) and Crispin-Norman (CXR) scores. Assuming a 20 mm slice interval, our dose calculations demonstrated a total absorbed dose equivalent to two chest xray examinations (PA + lateral) for the ultra low dose HRCT protocol.

Our results confirmed the validity of a low dose limited slice technique (i.e. 20 mm) and we concur with Jiménez et al with respect to the technical feasibility of this method in the sequential assessment of CF in school aged children. However CXR scores and HRCT scores were in very close agreeance (r = 0.9) using either scoring system across the range of scores recorded. Whilst HRCT provided more detailed information about the characteristics of lung disease, CXR scores were as sensitive as the HRCT scores for recording its presence.

Whilst there was a good correlation between FEV1 and HRCT scores (r=0.6) we did not identify useful spirometry parameters for the early detection of lung disease shown on HRCT and CXR.

Given the limited therapeutic options for pre-symptomatic CF lung disease, we suggest that routine low dose HRCT is unlikely to significantly contribute to patient management in this age group. This might, of course, alter if novel therapeutic interventions become available for which changes in HRCT appearances are valid as a clinically relevant outcome parameter. At present CXRs provide a practical and adequately sensitive test for the routine assessments of respiratory status.

Competing interests: none declared

REFERENCES:

1. Jiménez S, Jiménez JR, Crespo M, et al. Computed tomography in children with cystic fibrosis: a new way to reduce radiation dose. Arch Dis Child 2006;91:388-90.