AJR 2005; 184:1709
© American Roentgen Ray Society
Utilization Patterns for CT Pulmonary Angiography
Georg Zettinig and
Thomas Leitha
Medical University of Vienna Vienna University Hospital Vienna,
Austria
Prologo et al. [1] reported
a decreasing pretest probability of pulmonary embolism in conjunction with an
increasing number of patients referred to CT pulmonary angiography between
1998 and 2003. Although it was not the primary goal of their study, the
authors also provided data in Table 1 that showed only a minor decrease of
ventilation–perfusion imaging (488 vs 330) during the respective time
periods. In their conclusion, Prologo et al. addressed the necessity of
analyzing the clinical setting of referrals.
With due respect, we would like to draw the authors' attention to data in
our study [2] published in 2002
that substantiate their recent findings. Based on an analysis of 2,676
inpatients and outpatients at the Vienna University Hospital between March
1992 and December 1998 and between April 1997 and December 1998, we were able
to show that the introduction of CT pulmonary angiography significantly
affected the selection of patients referred for ventilation–perfusion
lung scanning. The percentage of patients with high and intermediate
probability for pulmonary embolism decreased significantly, from 15.2% to 9.4%
(p < 0.01) and from 10.2% to 7.3% (p < 0.05),
respectively. Low-probability scans increased significantly, from 37.8% to
42.7% (p < 0.05), whereas the total number of
ventilation–perfusion scans obtained annually (mean, 446 ± 135
[SD] scans) showed no significant decrease. Further analysis of the clinical
practice at our institution showed that, especially for outpatients, the
introduction of helical CT had affected patient referral because
ventilation–perfusion scanning was used primarily to exclude rather than
to confirm pulmonary embolism.
In light of our data, we came to the same conclusions that Prologo et al.
[1] discussed 2 years later.
Severe changes in the pretest probability of a referred patient population
might decrease the clinical utility of a diagnostic pathway because its
initial validation was based on different parameters. The compatibility of the
results of the two articles, although investigating this diagnostic problem in
different areas and in different time periods and focusing on two different
and competing diagnostic tools (ventilation–perfusion scanning and CT
pulmonary angiography), indicates a universal clinical problem in the
diagnostic pathways of pulmonary embolism, and we concur with the authors that
further investigation is crucial.
References
- Prologo JD, Gilkeson RC, Diaz M, Asaad J. CT pulmonary angiography:
a comparative analysis of the utilization patterns in emergency department and
hospitalized patients between 1998 and 2003. AJR2004; 183:1093
–1096[Abstract/Free Full Text]
- Zettinig G, Baudrexel S, Leitha T. Introduction of helical computed
tomography affects patient selection for V/Q lung scan. Nucl
Med 2002;41:91
–94
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
References
