373 / 412
When asked about persisting complaints at the end of the
trial, 11 of 26 placebo recipients (42%) felt themselves to be
free of symptoms compared to 22 of 28 patients ( 78%)
from the esomeprazole group (two patients of each study
group could not clearly answer the question with “yes” or
“no”). This difference is statistically significant (
P
0.006).
All documented adverse events were considered mild
and not treatment-related. The most common ones were
bronchitis (n 7), pharyngitis (n 5), and headache (n
5). One patient receiving esomeprazole was lost to fol-
low-up after the initial visit. Two female patients com-
plained of abdominal pain two and three weeks after receiv-
ing the study medication and were withdrawn from
treatment. Their treatment was unblinded, and as they were
part of the placebo group their complaints could not be
caused by esomeprazole. Another female patient developed
a skin rash six weeks after the beginning of therapy. Al-
though this drop-out patient was not a placebo recipient,
allergologic testing excluded drug hypersensitivity to es-
omeprazole
( Fig 1 ).
DISCUSSION
Our study is the second largest to assess the efficacy of a
PPI therapy for symptoms and signs associated with LPR
compared to placebo and the second one choosing esome-
prazole for a controlled trial. The results of this study
demonstrate a significant placebo effect in the therapy of
LPR-related symptoms as reported by patients especially
within the first 6 weeks of PPI treatment. This phenomenon
was already observed in former placebo-controlled studies
investigating the efficacy of PPI therapy in patients with
Table 1
Baseline characteristics*
Variable
Control
(n 28)
PPI
(n 30)
P
value†
Sex: male
13 (46.4%) 17 (56.7%) 0.160
Smoker
5 (17.9 %)
6 (20.0%) 0.871
Total RSI before
therapy
21.79 6.69 23.1 7.45 0.503
Total RFS before
therapy
14.89 2.5 14.9 2.75 0.969
Age
47.6 16 49 13.9 0.663
n,
number of subjects in each group;
PPI,
proton pump
inhibitor (esomeprazole);
RFS,
reflux finding score;
RSI,
re-
flux symptom index.
*Comparison of baseline characteristics between groups, re-
ported as mean standard deviation (SD) or proportion of
subjects (%).
†
P
value Mann-Whitney
U
test or
2
test, all nonsignificant.
Table 2
Change from baseline within each group after six weeks
Visit I vs visit II
Esomeprazole
Placebo
Difference
RSI
Total
9.87 1.61‡
6.93 1.61‡
2.94 2.27
Hoarseness
1.03 0.31†
1.07 0.35†
0.04 0.47
Throat clear
1.20 0.31†
1.32 0.32†
0.12 0.45
Throat mucus
1.10 0.28†
0.75 0.22†
0.35 0.35
Difficulty swallowing
1.23 0.30†
0.57 0.27*
0.66 0.41
Coughing after meals
0.53 0.34
0.11 0.32
0.64 0.47
Breathing difficulties
0.50 0.29
0.57 0.25*
0.07 0.39
Annoying cough
0.87 0.39*
0.68 0.27*
0.19 0.48
Throat sensations
1.60 0.32‡
1.25 0.34†
0.35 0.46
Heartburn
1.53 0.32‡
0.75 0.27†
0.78 0.42*
RFS
Total
3.47 0.47‡
2.46 0.71†
1.00 0.85
Subglottic edema
0.07 0.07
0.21 0.19
0.28 0.12
Ventricular
0.60 0.17†
0.36 0.18
0.24 0.25
Erythema
0.77 0.21†
0.43 0.26
0.34 0.33
Vocal fold edema
0.33 0.09†
0.46 0.13†
0.13 0.16
Diffuse laryngeal edema
0.60 0.13‡
0.21 0.11
0.39 0.16*
Posterior commissure hypertrophy
0.80 0.15‡
0.54 0.13†
0.26 0.20
Granuloma/granulation tissue
0.07 0.07
0.00 0.18
0.07 0.19
Thick endolaryngeal mucus
0.40 0.23
0.29 0.25
0.11 0.33
Data are given as mean differences standard error of the mean (SEM). The first two data columns show differences from
baseline and the third demonstrates the difference between study groups.
P
values are from Wilcoxon signed rank test and
Mann-Whitney
U
test as appropriate.
*
P
0.05; †
P
0.01; ‡
P
0.001.
Reichel et al Double-blind, placebo-controlled trial with . . .
148