Dig Dis Sci (2006) 51:1904–1909

DOI 10.1007/s10620-006-9284-0

ORIGINAL PAPER

A Comparison Between Sodium Alginate and Magaldrate

Anhydrous in the Treatment of Patients with Gastroesophageal
Reflux Symptoms
Edoardo G. Giannini · Patrizia Zentilin ·
Pietro Dulbecco · Elena Iiritano · Claudio Bilardi ·
Edoardo Savarino · Carlo Mansi · Vincenzo Savarino

Received: 20 December 2005 / Accepted: 1 March 2006 / Published online: 15 September 2006


C Springer Science+Business Media, Inc. 2006

Abstract The aims of the present study were to compare esophagus, which can result in acid damage to the esophageal
effects of sodium alginate and the antacid magaldrate mucosa (GORD). GORD causes a significant morbidity, an
anhydrous in adults with gastroesophageal reflux (GOR) increased loss of productivity, and a relevant burden on health
symptoms. economics [1, 2]. The prevalence in the general population
Patients with heartburn and/or acid regurgitation for at is estimated to vary from 20% to 40% [3], with increas-
least 3 days in the week before the study started (n = 203) ing rates reported in Western countries [4, 5]. In patients
were randomized to receive a single dose of sodium alginate with GORD, heartburn and acid regurgitation are the most
or magaldrate anhydrous at the onset of symptoms during a common symptoms, and other discomforts include epigas-
3-day run-in period. Patients with symptoms during the run- tric and/or retrosternal pain, and dysphagia, while extrae-
in (n = 191) were rerandomized to receive a 14-day treatment sophageal symptoms (e.g., noncardiac chest pain, laryngitis,
with either drug given as four daily doses. A speed of action coughing, and wheezing) may occur as well [5].
≤ 30 min was significantly more frequent among patients in The current medical management of GORD foresees the
the alginate group (49.4% vs. 40.4%; P = 0.0074). A trend administration of antacids, sodium alginate, and inhibitors
toward a more prolonged duration of action (median: 16.5 of acid secretion, such as H2 -receptor antagonists or pro-
vs. 12.7 hr) and a greater sum of the symptom intensity dif- ton pump inhibitors, coupled with lifestyle advice. In pa-
ference (median: 40.0 vs. 31.0) was observed in the sodium tients with GORD, it has been shown that alginate-containing
alginate group. Total disappearance of symptoms was re- preparations reduce the frequency and severity of symptoms
ported in 81.6% and 73.9% of patients in the sodium alginate [6, 7], decrease the number of both food and acid reflux
group and magaldrate group, respectively. We conclude that events [8], and limit the proximal migration of refluxed ma-
sodium alginate was faster than magaldrate in relieving GRO terial [7]. Antireflux properties of alginates are based on the
symptoms and showed a tendency toward a more prolonged interaction with gastric acid to form a strong viscous gel, or
duration of action and a higher level of efficacy. alginate raft, which floats on the top of the gastric contents
and effectively reduces by physical mechanism the reflux of
Keywords Sodium alginate . Magaldrate . Heartburn .
the gastric contents into the oesophagus [9]. The gel forms
Acid regurgitation
rapidly on exposure to gastric acid, occurring in vivo within
Introduction a few minutes of administration [10]. In raft-forming formu-
lations that contain potassium bicarbonate, the bicarbonate
Gastroesophageal reflux (GOR) is a common digestive dis- is converted to carbon dioxide, which is trapped within the
order that refers to the passage of gastric contents into the gel precipitate, thus converting it into a foam that floats on
the surface of the gastric contents and that can preferentially
move into the esophagus during episodes of GOR in place
E. G. Giannini · P. Zentilin · P. Dulbecco · E. Iiritano · C. Bilardi ·
of the acidic gastric contents [10–12]. Pharmacodynamic
E. Savarino · C. Mansi · V. Savarino (
)
Cattedra di Gastroenterologia, Dipartimento di Medicina Interna, studies showed that the alginate raft is characterized by a
Università degli Studi di Genova, gastric retention significantly higher than a standard meal
Viale Benedetto XV, no. 6, 16132 Genoa, Italy [9–12], and the time of raft permanence in the stomach was
e-mail: [email protected]
quantified as up to 4 hr [13].
Springer
Dig Dis Sci (2006) 51:1904–1909 1905

Taking into account that the acidic component of GORD severity, action taken, outcome, and relationship to the study
is mainly due to inappropriately located acidity (rather than treatment.
hyperacidity), and the frequency of reflux episodes and A blood sample was taken at study entry and at the end
duration of contact between gastric acid and esophageal of the treatment phase to assess the standard hematology
mucosa, sodium alginate has a mechanism of action which and blood chemistry parameters. Heart rate and blood pres-
is more respectful of gastric physiology than antacids, sure were measured at entry and end point using standard
whose role in GORD is a simple nonspecific neutralization procedures.
of gastric acidity.
Based on the above considerations, this study was aimed Evaluation of Treatment Outcomes
at comparing the efficacy and safety of sodium alginate,
which acts with a physical antireflux mechanism, and maga- At study entry, intensity of heartburn (defined as a retroster-
ldrate anhydrous, chosen as a representative antacid because nal burning sensation occurring in waves and tending to rise
of its wide use over decades in the treatment of GOR in upward toward the neck) and acid regurgitation in the pre-
symptomatic patients. vious week (average in the morning at awakening and in
the evening at bedtime) were recorded by interviewing the
patient, using a 4-point rating scale as follows: 0 = none (ab-
Patients and Methods sence of symptom); 1 = mild (minimal awareness of symp-
tom, which is easily tolerated); 2 = moderate (awareness of
Study Patients symptom, which is bothersome but tolerable, without im-
pairment of sleep or daily living); 3 = severe (symptom is
This open-label, prospective, randomized, parallel group hard to tolerate, with interference of daily activities and/or
clinical trial was conducted at 24 centers, all located in Italy sleeping). Patients had to record the time of drug administra-
(members of the study group are listed in the Appendix). tion and the time of symptom disappearance and recurrence
The study was carried out in accordance with Good Clini- during run-in, while heartburn and acid regurgitation were
cal Practice, International Conference on Harmonisation of recorded on a daily basis during the 14-day treatment phase
Technical Requirements for Registration of Pharmaceuticals according to the above rating scale.
for Human Use, and the Declaration of Helsinki. The primary efficacy variable was the total sum of scores
Patients of either gender aged ≥ 18 years were required for heartburn and acid regurgitation over the 14-day treat-
to have symptoms of GOR (heartburn and/or acid regurgita- ment period expressed as (a) the symptom intensity differ-
tion) for at least 3 days in the week prior to study entry and ence (SID), i.e., the difference obtained by subtracting the
during the 3-day run-in period. Eligible patients attended the patient’s sum of symptom score intensity (SSSI) at each time
screening/first randomization visit and entered a 3-day run- point from the patient’s baseline sum of symptom score; (b)
in to compare the speed and duration of action of sodium the sum of the symptom intensity difference (SSID), i.e., the
alginate oral suspension (Gaviscon Advance) and magal- baseline-adjusted sum of the SID scores over time, expressed
drate anhydrous oral suspension (Riopan gel). During the as the area under the time-effect curve for the treatment pe-
3-day run-in period, patients had to take a single oral 10-ml riod (14 days).
dose of the randomized study drug at the onset of symp- Other efficacy variables included speed of action (defined
toms of GOR (heartburn and/or acid regurgitation). Patients as the time elapsed from drug intake to complete symptom
with symptoms during the run-in were rerandomized to re- disappearance) and duration of action (defined as the time
ceive a 14-day treatment with sodium alginate suspension elapsed from complete symptom disappearance to symp-
or magaldrate anhydrous given in four 10-ml daily doses tom reappearance) during the run-in period and the time
(to be taken 1 hr after the main meals and at bedtime) and to complete disappearance of symptoms during the 14-day
follow-up visits took place after 7 and 14 (final visit) days of treatment period.
treatment.
Patients recorded the number of administered doses Statistical Analysis
on a diary card to allow assessment of compliance to
the assigned study drug based on the length of exposure. Statistical analysis was performed with the SAS System
The following treatments were not permitted at any time for Windows, release 8.2. All randomized patients who had
from 7 days before and throughout the study period: taken at least one dose of study drug and with at least one
H2 -receptor antagonists, motility stimulants, proton pump postbaseline follow-up were included in the intention-to-
inhibitors, theophylline or xantine derivatives (including treat (ITT) population. Patients included in the ITT analysis
coffee and tea), and other antacids. Patients were asked who did not have major protocol violations were included in
to report adverse events at each visit, together with their the per-protocol (PP) analysis. The primary efficacy variable

Springer
1906 Dig Dis Sci (2006) 51:1904–1909

Fig. 1 Patients’ flow through

the various phases of the study

was analyzed in both populations, while the other efficacy Results

parameters were analyzed only in the ITT population. The
last observation carried forward (LOCF) method was used A total of 203 patients were randomized to run-in phase,
to deal with missing data. 101 to sodium alginate and 102 to magaldrate: 7 patients (4
The primary variables (SID, SSID), speed, and duration of and 3 in the two groups, respectively) were lost to follow-up
action were analyzed using the Mann-Whitney U test; daily and 5 (3 and 2 in the two groups, respectively) did not have
symptoms were analyzed using Cox’s regression model (also symptoms during run-in. Therefore, a total of 191 patients
used to analyze speed of action during run-in). The frequency were randomized to the treatment phase: 93 were assigned
of adverse events (coded by MedDRA system organ class) to sodium alginate and 98 to magaldrate. Overall, 180 pa-
in the two groups was analyzed using χ 2 test. Analysis of tients (88 and 92 patients in the two groups, respectively)
laboratory parameters was done within and between groups completed the treatment phase and 11 patients quit the study
using paired and unpaired t-test, respectively. Heart rate and prematurely. One complete patient in the sodium alginate
blood pressure over time were analyzed using repeated- group did not have any evidence of drug intake and was ex-
measures ANOVA with time points as the factor within cluded from the ITT population (179 total; 87 and 92 patients
subjects. in the two groups, respectively), while 2 major violators in

Springer
Dig Dis Sci (2006) 51:1904–1909 1907

Table 1 Speed of action: patients’ distribution by treatment

Speed of action Sodium alginate Magaldrate P

<15 min 11.2% 6.7%

15–30 min 38.2% 33.7%
Total 49.4% 40.4% 0.0074

Note. Data were analyzed by a logistic regression model including

age, gender, and clinical symptoms at baseline as covariates.

the magaldrate group did not enter the PP population (177

total; 87 and 90 patients in the two groups, respectively).
Figure 1 illustrates the patients’ flow through the various
phases of the study.
Groups assigned to sodium alginate and magaldrate were
well matched for demographics, baseline characteristics, and Fig. 3 Symptom intensity difference (SID) over baseline in the ITT
population (SE in bars)
clinical symptoms, both at the start of the run-in phase and
at the start of the treatment phase. Symptom severity at the
start of the treatment phase also was similar in the two groups Symptom severity (SSSI) in the treatment phase for the
(mean SSSI, 4.75 ± 2.47 in the sodium alginate group and ITT population is presented in Fig. 2. From Day 1 on-
4.82 ± 2.79 in the magaldrate group). ward, the mean SSSI score in the sodium alginate group
Both study drugs exhibited a good speed of action dur- was always lower than that for the magaldrate group. Con-
ing run-in. Although the comparison between groups did sequently change from baseline (SID) tended to be greater
not show statistically significant differences, the time to dis- in the sodium alginate group than in the magaldrate group
appearance of symptoms in the sodium alginate group was (Fig. 3): differences between groups were evident as early
shorter (median, 35.0 min) than in the magaldrate group (me- as Day 2 (net difference, 0.35), were maximal on Day 8
dian, 44.0 min). In a further analysis, carried out according (0.41) and Day 9 (0.48), and were maintained on the last day
to the criteria adopted by Chevrel (speed of action grouped (Day 14: 0.25). Mean values of the SSID were 42.8 (median,
as <15, 15–30, and >30 min [14]), a statistically signifi- 40.0) in the sodium alginate group and 38.7 (median, 31.0)
cant difference in favor of sodium alginate in the frequency in the magaldrate group. Due to the minimal difference in the
of patients with a speed of action ≤ 30 min was observed populations, results obtained in the PP population were con-
(49.4% vs. 40.4%, respectively; P = 0.0074). The distribu- sistent with those found in primary ITT analysis. As shown
tion of speed of action by treatment is reported in Table 1. in Fig. 4, disappearance of symptoms was reported in 71 of
87 patients (81.6%; 95% CI, 73.4%–87.9%) in the sodium
The duration of action during run-in tended to be more alginate group and in 68 of 92 patients (73.9%; CI, 65.3%–
prolonged, but not significantly so, in the sodium alginate 81.2%) in the magaldrate group.
group (median, 16.5) than in the magaldrate group (median, All patients in both groups exhibited a satisfactory level
12.7). of compliance to study drug (i.e., ≥ 75% of scheduled). A
total of 21 adverse events, mainly consisting of gastrointesti-
nal complaints or painful conditions, were reported by 13

Fig. 2 Results of the sum of symptom score intensity (SSSI) in the Fig. 4 Percentage of patients (95% CI in bars) with symptom disap-
ITT population (SE in bars) pearance during the treatment phase

Springer
1908 Dig Dis Sci (2006) 51:1904–1909

patients. The number of adverse events during the treatment more prolonged in the sodium alginate group than in controls
phase was 6 in the sodium alginate group (in 5 patients; 5.5% ( +30.0%) and, in an analysis defined poststudy only, that the
of the safety population) and 14 in the magaldrate group (in 7 relief of symptoms was achieved within 30 min by more pa-
patients; 7.2%). One patient in the sodium alginate group (di- tients in the sodium alginate group than in the magaldrate
arrhea) and two in the magaldrate group (one with diarrhea group (P<0.01). Considering the results of previous studies
and one with several occurrences of gastrointestinal discom- with those of the present one, we can assume that sodium al-
fort) quit the study prematurely due to adverse events. Two ginate mainly starts to induce symptom relief within 10 min
events in the sodium alginate group (diarrhea and nausea) after its intake, with complete disappearance of symptoms
and one in the magaldrate group (feces abnormality) were within 30 min [6–9, 13, 14]. The raft lasts about 4 hr, while
considered to be potentially drug-related. One event in the the GOR symptom-free period observed in the study lasted
magaldrate group (non-drug-related biliary colic) caused a for a median of 16.5 hr [13, 14].
patient’s hospitalization and, therefore, was considered to be Noteworthy, all patients treated with alginate responded
a serious adverse event. No drug-induced clinically signifi- to single-dose therapy during run-in, thus suggesting a po-
cant abnormalities were observed at the final visit in either tential high sensitivity of sodium alginate in detecting GOR.
group for any laboratory variable. Results for heart rate and On the basis of this observation, the use of a single sodium
blood pressure did not point into evidence any change from alginate dose as a diagnostic tool in patients with symptoms
baseline in either treatment group. suggestive of GOR seems promising, although this requires
further investigation before being formally proposed. Indeed,
since the currently used standard short-term treatment with
Discussion proton pump inhibitors as a test for GOR (PPI test) requires
a 1- to 2-week period of high-dose treatment and has a sen-
Sodium alginate products act by a physical antireflux mech- sitivity of approximately 80%, with a poor specificity [15,
anism and are widely used for the relief of heartburn and 16], use of a single dose of sodium alginate might exhibit
symptoms related to GOR. Symptom relief occurs rapidly; a higher sensitivity, better compliance derived from single
therefore, alginate-containing preparations may provide an dosing, and rapid interpretation of results.
onset of relief that is comparable to or even faster than that Although this study was aimed at assessing the superior-
of antacids which act by a buffer effect aimed at neutralizing ity of sodium alginate over magaldrate in terms of symptom
acidic gastric contents and thus produce a nonacid reflux, severity (sum of the symptom intensity difference; SSID), the
which, however, does not correspond to a innocuous one difference in SSID between groups was not statistically sig-
(e.g., bile, pepsin). nificant. However, the SID from baseline was always greater
The main mechanism of action of alginate-containing in the sodium alginate group than in the magaldrate group
preparations is the interaction of the drug with gastric acid, from Day 2 onward, with a maximal difference at Days 8
thus forming a viscous gel (alginate raft). In vivo, the algi- and 9. The difference in effects was still evident at Day 14,
nate raft forms within a few minutes of drug administration considering both the total score and the individual symptoms
and, by floating on top of the gastric contents, preferen- heartburn and acid regurgitation, either in the daytime or in
tially moves into the esophagus during episodes of GOR in the nighttime. Notably, sodium alginate seemed to be partic-
place of the acidic gastric contents, thus effectively reduc- ularly effective in controlling nighttime heartburn, which is
ing the likelihood of mucosal damage [9–12]. Since the raft recognized not only as a bothersome common GOR symp-
can be retained on top of the stomach contents for several tom, but also as an important factor in the pathogenesis of
hours, alginate-based raft-forming formulations can provide complications of sleep-related GOR [17], such as occurrence
longer-lasting relief compared to antacids [7–14]. Therefore, of asthma-like symptoms [18] or impairment of quality of
the present study was aimed at comparing sodium alginate life [19], particularly when nighttime heartburn persists on
and magaldrate, a reference antacid, with respect to symp- a long-term basis. Furthermore, complete disappearance of
tom relief efficacy, not only in terms of the extent of overall both heartburn and acid regurgitation during the 14-day treat-
symptom relief but also in terms of the speed and duration of ment period was achieved by 81.6% of patients treated with
action and safety. The use of a run-in and a 14-day treatment sodium alginate, compared with 73.9% of patients receiving
phase (for patients presenting symptoms during the run-in magaldrate. Nevertheless, the differences observed in treat-
phase) by using double randomization permitted speed and ment efficacy between the two tratment arms are relatively
duration of action to be measured following a single dose. small, and further studies on larger populations are needed
The groups of patients assigned to either study drug were to confirm these preliminary findings.
well matched for baseline characteristics and symptom in- Due to the mode of action exerted via the formation of
tensity both at the start of run-in and at the start of the treat- a physical barrier, and considering that it is excreted as an
ment phase. Results showed that the duration of action was unmodified agent, no adverse effects are expected during

Springer
Dig Dis Sci (2006) 51:1904–1909 1909

treatment with sodium alginate preparations. The results of 4. Niemcryk SJ, Joshua-Gotlib S, Levine DS (2005) Outpatient expe-
the present study confirm the excellent safety profile of this rience of patients with GERD in the United States: analysis of the
1998–2001 National Ambulatory Medical Care Survey. Dig Dis
drug. Sci 50:1904–1908
In summary, the results of the present study show that 5. Okamoto K, Iwakiri R, Mori M, Hara M, Oda K, Danjo A, Ootani
treatment with sodium alginate produced fast relief of GOR A, Sakata H, Fujimoto K (2003) Clinical symptoms in endoscopic
symptoms. Furthermore, secondary analysis showed that sig- reflux esophagitis: evaluation in 8031 adult subjects. Dig Dis Sci
48:2237–2241
nificantly more patients receiving sodium alginate achieved 6. Chatfield S (1999) A comparison of the efficacy of the alginate
rapid symptom disappearance (<30 min) than with magal- preparation, Gaviscon Advance, with placebo in the treatment of
drate. This is of particular importance, since prompt symp- gastroesophageal reflux disease. Curr Med Res Opin 15:152–159
tom remission is especially needed by patients with GOR. 7. Zentilin P, Dulbecco P, Savarino E, Parodi A, Iiritano E, Bilardi
C, Reglioni S, Vigneri S, Savarino V (2001) An evaluation of
Sodium alginate has demonstrated a high level of efficacy,
the antireflux properties of sodium alginate by means of combined
fast speed of action, and prolonged symptom-free period. multichannel intraluminal impedance and pH-metry. Aliment Phar-
These characteristics together with its specific antireflux macol Ther 21:29–34
mechanism of action make sodium alginate a valid tool for 8. Washington N, Steele RJ, Jackson SJ, Washington C, Bush D
(1998) Patterns of food and acid reflux in patients with low-grade
treatment of GOR symptoms. oesophagitis—the role of an anti-reflux agent. Aliment Pharmacol
Ther 12:53–58
9. Mandel KG, Daggy BP, Brodie DA, Jacoby HI (2000) Review
Appendix article: alginate-raft formulations in the treatment of heartburn and
acid reflux. Aliment Pharmacol Ther 14:669–690
10. Korman MG, Nicholson L, Chan JG (1990) In-vivo anti-reflux and
Study coordinator V. Savarino (Genova). raft properties of alginates. Aliment Pharmacol Ther 4:615–622
Investigators G. Bernasconi (Gallarate, VA), E. Corazz- 11. Marciani L, Little SL, Snee J, Coleman NS, Tyler DJ, Sykes J,
iari E (Roma), R. Cuomo (Napoli), R. De Franchis (Milano), Jolliffe IG, Dettmar PW, Spiller RC, Gowland PA (2002) Echo-
planar magnetic resonance imaging of Gaviscon alginate rafts in-
M. Del Piano (Novara), A. D’Odorico (Padova), O. Ferraù vivo. J Pharm Pharmacol 54:1351–1356
(Messina), A. Franzè (Parma), G. Gasbarrini (Roma), S. 12. Hampson FC, Farndale A, Strugala V, Sykes J, Jolliffe IG, Dettmar
Marchi (Pisa), A. Pera (Torino), M. Perego (Pavia), P. Ravelli PW (2005) Alginate rafts and their characterisation. Int J Pharm
(Seriate, BG), A. Russo (Catania), V. Stanghellini (Bologna), 294:137–147
13. Taylor G, Warren SJ, Kelleway IW, Patel B, Little SL (1997) Gastric
V. Stoppino (Foggia), C. Surrenti (Firenze), F. Tosato residence of Gaviscon Advance and Liquid Gaviscon in healthy
(Roma), P. Usai (Cagliari), F. Valente (Taormina, ME), S. volunteers. J Pharm Pharmacol 49(Suppl 4):73 (131)
Vigneri (Palermo), M. Zilli (Udine), and G. Zoli (Cento, FE). 14. Chevrel B (1980) A comparative crossover study on the treat-
ment of heartburn and epigastric pain: Liquid Gaviscon and a
magnesium-aluminium antacid gel. J Int Med Res 8:300–302
Acknowledgments This study was supported by Novartis Consumer
15. Juul-Hansen P, Rydning A (2003) Endoscopy-negative reflux dis-
Health SpA, Italy.
ease: What is the value of a proton-pump inhibitor test in everyday
clinical practice? Scand J Gastroenterol 38:1200–1203
16. Numans ME, Lau J, de Wit NJ, Bonis PA (2004) Short-term treat-
References ment with proton-pump inhibitors as a test for gastroesophageal
reflux disease: a meta-analysis of diagnostic test characteristics.
Ann Intern Med 140:518–527
1. Dean BB, Crawley JA, Schmitt CM, Wong J, Ofman JJ (2003) 17. Orr WC (2004) Sleep and gastroesophageal reflux disease: a wake-
The burden of illness of gastroesophageal reflux disease: impact up call. Rev Gastroenterol Disord 4(Suppl 4):S25–S32
on work productivity. Aliment Pharmacol Ther 17:1309–1317 18. Gislason T, Janson C, Vermeire P, Plaschke P, Bjornsson E, Gis-
2. Joish VN, Donaldson G, Stockdale W, Oderda GM, Crawley J, lason D, Boman G (2002) Respiratory symptoms and nocturnal
Sasane R, Joshua-Gotlib S, Brixner DI (2005) The economic im- gastroesophageal reflux: a population-based study of young adults
pact of GERD and PUD: examination of direct and indirect costs in three European countries. Chest 121:158–163
using a large integrated employer claims database. Curr Med Res 19. Farup C, Kleinman L, Sloan S, Ganoczy D, Chee E, Lee C, Re-
Opin 21:535–544 vicki D (2001) The impact of nocturnal symptoms associated with
3. Valle C, Broglia F, Pistorio A, Tinelli C, Perego M (1999) Preva- gastroesophageal reflux disease on health-related quality of life.
lence and impact of symptoms suggestive of gastroesophageal re- Arch Intern Med 161:45–52
flux disease. Dig Dis Sci 44:1848–1852

Springer