CARDIOVASCULAR JOURNAL OF AFRICA • Vol 21, No 3, May/June 2010
146
AFRICA
Tc99m MIBI. However, a recent study failed to demonstrate any
effect of metoclopramide on interfering extra-cardiac activity.
A study done recently by Cherng
et al
.
9
concluded that diluted
lemon juice accelerated the transit of tetrofosmin through the
liver parenchyma, leading to improved image quality on myocar-
dial perfusion SPECT images. Prior to this, various studies had
been done in order to evaluate the effects of water, milk, milk-
shakes, and fatty meals on imaging time.
11
The rationale behind
these protocols is the following:
Clearance of activity located in the gall bladder and hepatic
•
•
duct into the duodenum and bowel may be stimulated by fatty
meals, milkshakes and intravenous cholecystokinin.
Water and sandwiches have been used to fill the stomach and
•
•
push the bowel away from the myocardium, thus exerting a
volume effect. In addition, peristaltic gastric motility may
be stimulated by oral consumption following pre-injection
fasting, and the upper gastro-intestinal tract activity may
therefore be diluted.
6
A study by Peace and Lloyd
11
evaluated the effect of imaging
time, the radiopharmaceutical used, and full-fat milk or water on
interfering extra-cardiac activity in myocardial SPECT in 260
patient acquisitions. The following conclusions were made from
this study:
The quantitative results indicated that oral administration of
•
•
full-fat milk or water made no significant difference to extra-
cardiac activity.
There was no significant difference between tetrofosmin and
•
•
sestamibi at any point.
Delayed imaging significantly reduced extra-cardiac activity
•
•
interfering with observer interpretation.
More importantly, however, was the development of a method
•
•
to quantify the myocardial-to-extra-cardiac activity ratio,
which correlated significantly with observer interpretation of
myocardial perfusion studies.
11
Other studies performed
Boz
et al
.
4
investigated the volume effect of the stomach in
patients undergoing a same-day exercise–rest protocol. After
the rest study, patients were given either 200 ml of water and
a sandwich or were fasted and imaged again within 30 mins.
Quantitative and observer analysis showed significantly reduced
activity inferior to the left ventricle in the meal group due to an
expanded stomach.
4
Hurwitz
et al
.
6
studied the effect of a milkshake taken either
immediately after injection or just before imaging. An early
drink decreased gallbladder activity, but had no effect on the liver
parenchyma at 15 and 110 min. Patients drinking a milkshake
just prior to imaging showed a prompt 26% reduction of activity
inferior to the myocardium.
6
Van Dongen and van Rijk qualitatively investigated the effects
of administering 450 ml of water 10 mins before imaging and/
or 150 ml of whole milk 10 mins after injection. Visual analysis
showed significantly fewer patients with interfering activity
when a milk-only protocol was used, compared with water only
or no drink, for images reconstructed by FBP. A combined milk-
and-water protocol showed the least number of studies with
interfering activity, compared with water-only and milk-only
protocols.
1
Most recently, Hara
et al
.
3
reported on the use of soda water
(given in the form of a carbonated soft drink) in conjunction
with adenosine stress myocardial perfusion imaging and in
the Monzen position. (This is a body-positioning manoeuvre
to increase the distance between the heart and the intestine by
raising the left arm and curving the trunk slightly to the right.)
According to their report, a carbonated beverage distended the
stomach with gas and lessened artifacts from intestinal radio-
tracer activity. Additional benefits included the faster clearance
of gas and a smaller amount of fluid needed to distend the stom-
ach, which is especially desirable in patients with left ventricular
dysfunction.
3
Interfering extra-cardiac activity therefore remains a signifi-
cant problem and a limitation to accurate interpretation of
myocardial perfusion imaging. In our study, interfering extra-
cardiac activity was present overall in 54% of the stress images
and 48% of the rest images, when the results were combined
from all three observers for both groups. Almost all the stress
studies were performed with Persantin, confirming once again
the higher incidence of interfering activity on pharmacological
stress studies.
In our Nuclear Medicine Department, the use of diluted lemon
juice has replaced the former protocol of bread and milk. Lemon
juice was perceived to reduce the amount of interfering activity
significantly, although the effects of the two protocols were never
formally compared. When evaluating the qualitative results of
this study, lemon juice appeared to be superior in reducing the
amount of interfering activity, with all three observers reporting
less interference in the lemon juice group. However, this was not
found to be statistically significant. The superiority of lemon
juice in this setting might be explained solely on the basis of a
much larger volume of fluid consumed, which might have led to
stomach distension.
Limitations of this study
The experimental design was sub-optimal, with individuals
being exposed to the same intervention during both parts of the
study (stress and rest). A better design might have been to expose
individuals to both interventions on different days. However, the
inherent differences in interfering activity on stress and rest stud-
ies (even in the absence of any intervention) present a diagnostic
dilemma.
A third and even a fourth comparison group would have been
helpful in validating the results with one group serving as a
control (receiving no intervention) and the other receiving milk
and bread (according to our previous protocol). However, consid-
ering the decrease in the amount of interfering activity perceived
with the use of lemon juice, the addition of the latter two groups
were considered to be unethical.
Upon final statistical analysis, the study appeared to be slight-
ly underpowered, which could have resulted in a type II error.
There were suspected errors during the randomisation process,
which might have resulted in the higher incidence of dyslipidae-
mia in the erythromycin group. The calculation method of the
MYO:EXT ratio as described by Peace and Lloyd
11
could not be
replicated exactly due to differences in acquisition parameters as
well as the software used for image processing.
Despite the above limitations, the use of diluted lemon juice
appeared to be beneficial in reducing interfering extra-cardiac
activity in MPI studies where sestamibi was used. Furthermore,
