Kanako Kunishima Kumamaru MD, PhD, Presenter: Nothing to Disclose

Hiraku Kumamaru, Abstract Co-Author: Nothing to Disclose

Hideo Yasunaga, Abstract Co-Author: Nothing to Disclose

Shigeki Aoki MD, PhD, Abstract Co-Author: Nothing to Disclose

Kuni Ohtomo MD, Abstract Co-Author: Research Grant, Bayer AG Research Grant, DAIICHI SANKYO Group

Frank John Rybicki MD, PhD, Abstract Co-Author: Research Grant, Toshiba Corporation

Ruth M. Dunne MBBCh, Abstract Co-Author: Nothing to Disclose

Amir Imanzadeh MD, Abstract Co-Author: Nothing to Disclose

In Japan, CT scanners are readily available and individual imaging costs are low. Thus, CT is an “easy to perform” test that can be overutilized. The purpose of this study is to evaluate the hospital level factors associated with higher CT utilization using a nationwide Japanese administrative database.

This retrospective (04/2012-03/2013) study used data from the Japanese nationwide administrative database of 1108 hospitals. The target population was patients who received total knee or hip replacement surgeries at hospitals meeting an inclusion criteria based on 30 or more procedures annually. We fitted logistic regression models with generalized estimating equations to identify hospital level factors associated with patient’s probability of undergoing CTs, while adjusting for patient level factors including age, sex, BMI, smoking status, pre-existent comorbidities, history of deep vein thrombosis, operative type (knee or hip), surgical time, transfusion during surgery, postoperative events, and length of stay.

A total of 448 hospitals met the inclusion criteria, and in these institutions, 37,110 patients underwent the procedure. During the median length of stay of 25 days (IQR:20-33), 7307 (19.7%) patients underwent post procedure CT. CT utilization was not homogeneous among hospitals. At majority (57.2%) of hospitals, less than 10% of patients had post procedure CT. However, more than 50% of patients had a CT at the 46 (10.3%) high utilization hospitals. Multivariable analysis identified academic status of the hospitals (99/448 hospitals) to be strongly associated with an increased probability of undergoing post procedure in-hospital CTs (adjusted odds ratio:4.22 (95%CI:1.91-9.32), p=0.003), while other hospital characteristics (number of beds, availability of radiologists, number of surgeons and total number of procedures during the year) were not.

We observed great variability in the utilization of post procedure CT in this Japanese cohort. Patients treated at academic facilities were more likely to have in-hospital CT after total knee or hip replacement compared to those treated at non-academic hospitals, even after adjusting for patient and hospital level factors.

This Japanese database study showed an association between higher CT utilization after total knee or hip replacement surgery and academic status of the hospitals, which may need further investigation.

Kumamaru, K, Kumamaru, H, Yasunaga, H, Aoki, S, Ohtomo, K, Rybicki, F, Dunne, R, Imanzadeh, A, Hospital Level Factors Associated with High Utilization of In-hospital CT in Japan.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14006802.html