69 / 242
69
A MULTI-CENTRE RANDOMISED DOUBLE-BLIND PLACEBO-CONTROLLED TRIAL
6
Power calculation
We powered to detect a difference in maximum pain score between the placebo and
alfentanil group, using nQuery Advisor 7.0 (Statistical Solutions Ltd., Cork, Ireland). Based
on our anaesthesiologists (S.E. and M.W.H.) experience a maximum pain reduction of 1.5
points was expected. A previously performed population screening trial served as pilot
data to assess distribution and standard deviation.
15
We calculated for a one-sided t-test, 1.5
point difference, 80% power, 0.05 α error and a standard deviation of 2.6. We assumed 5%
withdrawal and therefore our calculation resulted in groups of 45 patients.
Intervention
Two research physicians (J.H.R. and M.C.H.) generated a randomisation list using nQuery.
The list was kept by three research physicians (J.H.R., M.C.H. and J.A.W.T.) not involved in the
patient recruitment, CT colonography procedure or data collection. Included patients were
randomised into two groups of 45 patients (1:1 ratio and blocks of six). Group 1 received a
single bolus of 7.5 μg/kg actual body weight alfentanil (Rapifen, Janssen-Cilag, Tilburg, the
Netherlands). Group 2 received 0.9% saline solution as placebo. Both groups received 0.075
ml/kg actual body weight fluid. The physicians who kept the randomisation list prepared
study medication. The medication ampule was placed in a signed sealed envelope near the
CT scanner room to allow deblinding in case of a medical emergency. The study medication
was administered (double-blind) through a 20 Gauge intravenous cannula, 1½ minute
following administration of a spasmolytic agent. After procedure, one of the physicians who
kept the randomisation list collected the envelope.
CT colonography
We used a 24-hour preparation with low-fibre diet and three or four bottles 50 mL iodinated
contrast, meglumine ioxithalamate (Telebrix, Guerbet, Aulnay sous Bois, France).
31,32
For
bowel relaxation 1ml (20mg) butylscopalamine bromide (Buscopan, Boehringer-Ingelheim,
Ingelheim, Germany) or, if contraindicated 1 ml (1 mg) glucagon (GlucaGen, Novo
Nordisk A’S, Bagsvaerd, Denmark) was used.
33,34
An automated carbon dioxide insufflator
(PROTOCO2L, Bracco, EZEM, Lake Success, USA) and a flexible 20 French rectal catheter were
used with insufflation in three positions: right decubitus, supine and left decubitus position.
We aimed for three litres insufflation (1.3, 0.8 and 0.9 litres per position, respectively). The
insufflation pressure was gently increased during insufflation (maximum pressure of the
insufflator is 25 mmHg) and set on 20 mmHg when the target volume of three litres was
met or after five minutes of insufflation irrespective of the target volume. Subsequently,
scan acquisitions were performed in prone and supine position and intravenous contrast
medium iopromide (Ultravist 300, Bayer B.V., Mijdrecht, the Netherlands) was given in case