Effect of baclofen on oesophageal motility and transient lower oesophageal sphincter relaxations in GORD patients: a 48-h manometric study

Introduction

The pathogenesis of gastro-oesophageal reflux disease (GORD) is multifactorial but the main factor is represented by an incompetence of the lower oesophageal sphincter (LOS). Transient LOS relaxations (TLOSRs) in particular are responsible for the majority of reflux episodes¹ and in the last years they have become an intriguing potential target for therapeutic approaches to GORD.^2–4 Baclofen, a gamma-aminobutyric acid B (GABA_B) receptor agonist which is a potent inhibitory neurotransmitter in the central nervous system, has emerged as one of the most promising drugs in this regard. Recent studies have shown that baclofen reduces the rate of TLOSRs and the rate of gastro-oesophageal reflux episodes and increases the basal LOS pressure (LOSP) both in normal subjects and in GORD patients.^5–7 Despite the huge amount of information about baclofen, the majority of the studies are limited either to a single dose or to relatively short periods of pH or manometric monitoring.^8,9 One study analysed the 24-h acid exposure after baclofen administration¹⁰ but, to date, there is still no information on the effect of this drug on the motility of oesophagus and LOS over a 24-h period. This information could be useful to optimize the therapeutic schedule, considering that the drug tends to disappear from the blood within 5–6 h after the consumption¹¹ and that multiple doses could cover the 24-h period better than a single administration.

Therefore, the aim of the study was to assess the effect of the GABA_B agonist baclofen on 24-h oesophageal and LOS motility in a group of GORD patients after multiple oral doses of the drug.

Materials and methods

Results

Twenty-one patients out of the 23 initially enrolled completed the study according to the protocol. Two patients (1M, 1F) were excluded because of LOSP < 10 mmHg not evidenced at the previous manometry. Fourteen patients received baclofen during the second 24-h period (Group A), whereas seven patients of Group B received placebo during the second day.

Deglutitive activity

The number of swallows during the 24 h resulted significantly lower after baclofen as compared to the baseline, 1254(850–1620) vs 1583(980–1752), respectively, P = 0.02. Placebo administration did not elicit any changes in the number of swallows throughout the 24-h period, that resulted 1528(770–1877), ns vs baseline (Fig. 1). The effect was also similar considering the daytime and the nighttime: in fact, the number of swallows was reduced after baclofen administration in both the periods with no change after placebo [Day: basal 1120(568–1340), baclofen 945(470–1225), P = 0.045; placebo 1096(630–1230), ns; Night: basal 440(235–558), baclofen 338(220–457), P = 0.045, placebo 468(260–533), ns].

Oesophageal peristalsis

The number of postdeglutitive oesophageal body waves during the 24 h resulted significantly lower after baclofen as compared to the baseline, 980(580–1450) vs 1245(730–1750), respectively, P = 0.04. Placebo administration did not elicit any changes in the number of oesophageal body waves throughout the 24-h period, resulted 1330(808–1596), ns vs baseline) (Fig. 2). As it happened for swallowings, the daytime and the nighttime periods showed a similar pattern: in fact, the number of primary peristaltic waves was reduced after baclofen in both periods with no change after placebo [Day: basal 920(450–1220), baclofen 735(410–1177), P = 0.04, placebo 950(560–1230), ns; Night: basal 398(267–540), baclofen 260(198–420), P = 0.045, placebo 368(255–580), ns].

No changes were detected in the amplitude of the primary oesophageal body waves neither after baclofen nor after the placebo in comparison with the baseline [basal 49.3(38–61) mmHg, baclofen 45.7(35–68) mmHg, ns; placebo 45.8(40–59), ns].

The percentages of simultaneous [basal 9.8%(6.3–14.1), baclofen 9.4%(5.6–16.2), ns; placebo 10.1%(7.2–14.6), ns] and of non-transmitted oesophageal contractions [basal 19.5%(11.3–24.5), baclofen 20.4%(12.8–26.2), ns; placebo 20.1%(13.2–25.6), ns] were similar in the three groups. Furthermore, the secondary peristalsis resulted unaffected by baclofen, showing similar characteristics in terms of percentage of propagation [basal 30.8%(16.3–44.8), baclofen 33.4%(15.6–46.2), ns; placebo 32.5.%(17.1–44.6), ns] and amplitude [basal 39.1(29–51) mmHg, baclofen 37.2(25–50) mmHg, ns; placebo 38.8(27–55), ns].

Lower oesophageal sphincter basal activity and TLOSRs

The basal LOS tone was significantly increased in the group of patients assuming baclofen in comparison with the baseline whilst no modifications were detected after the placebo [basal 20.5 mmHg (14–32); baclofen: 24.8 mmHg (15–35), P = 0.02; placebo 21.5 mmHg (14–30), ns] (Fig. 3). The daytime and the nighttime periods confirmed that the basal LOS tone was increased after baclofen with no change after the placebo [Day: basal 22.8 mmHg (15–31), baclofen 24.5 mmHg (15–35), P = 0.05, placebo 22.4 mmHg (14–29), ns; Night: basal 18.6 mmHg (14–25), baclofen 20.2 mmHg (15–27), P = 0.045, placebo 19.3 mmHg (14–29), ns].

The number of TLOSRs during the 24 h decreased after baclofen in comparison with the basal period and, even in this case, the placebo did not show any significant effects [basal 57.5(38–73); baclofen 41.5(29–62), P = 0.01; placebo 55(27–69), ns] (Fig. 4). The 24-h profile obtained by calculating the number of TLOSRs in each 2-h window confirm that the baclofen reducing effect was maintained almost homogeneously throughout the entire day (Fig. 5).

Finally, no differences were identified in the occurrence of phase III of migrating motor complexes (MMCs) recorded at the LOS level during the night [number of cycles detected: basal 3(1–4); baclofen 3(2–4), ns; placebo 3(2–4), ns]. Although no changes were detected as well in the amplitude of phase III contractions and in the latency between each front within the different groups of patients, these periods were not considered in the total analysis of LOS basal tone.

Discussion

Our study confirms that baclofen reduces the frequency of TLOSRs and elicits a greater basal LOS tone in GORD patients. The new finding we demonstrated is that it is possible to maintain both of these effects for longer periods than previously known with multiple oral administrations of the drug. The GABA_B agonist baclofen is already known to affect the LOS motor pattern, both the basal tone and the TLOSRs occurrence rate but, to date, all the information present in literature have been related to short-lasting recordings, mainly limited to the postprandial periods. The only 24-h analysis regards the pH-metric pattern 10, with the demonstration of a prolonged significant reduction of the acid presence inside the oesophageal lumen.

The therapy of GORD, for years based mainly on a strong acid suppression, could benefit from a more aetiological approach. Transient LOS relaxations seem to represent one of the main targets of new-developing strategies, in view of their major role in the genesis of gastro-oesophageal reflux episodes 2, 3, 4. Although the majority of reflux episodes occur in the postprandial period, it is clear that facilitating factors of GORD are present throughout the 24 h and not only after meals. Our study therefore aimed at demonstrating whether multiple daily doses of the drug could prolong the control of LOS motor pattern, thus offering a better barrier against reflux episodes.

The 24-h profile obtained in our study indicates that the number of TLOSRs is continuously reduced by baclofen. Our data represent the first report looking at the profile of transient relaxations throughout a 24-h period, in the postprandial and in the fasting state, during daytime and at night. The difference in the nighttime period is actually not statistically significant. This is probably due to the low number of TLOSRs recorded at that time, as a trend towards a reduction was detected after baclofen but not in the placebo group. Our results at the moment suggest that the administration of the night dose of baclofen could probably be avoided, although it seems to represent a first step in the attempt of reducing TLOSRs at night. What is not surprising is that the greater difference has been observed after meals, as this represents the period when TLOSRs more frequently occur.¹³ Nevertheless, we found that the effect of baclofen is also present in the fasting state; this suggests that the GABA_B pathway is involved in TLOSRs occurrence independently from the period of occurrence, both in the fasting and in the fed state. As far as the basal LOS tone the population enrolled in our study has been selected by ruling out patients with values <10 mmHg evidenced at a previous manometric exam. This decision, as explained above, tended to minimize the risk of missing LOS relaxations during the analysis. It could be that in this way we missed a group of patients potentially very representative of the disease. Despite this bias it seems equally important in our opinion that the basal LOS tone of the patients recruited is homogeneously affected by baclofen throughout the whole recording period. Lower oesophageal sphincter pressure is not constant during the 24 h; it is well known, in fact, that the LOSP tends to fall after a test meal,¹⁴ there is evidence as well of a relationship between the cyclic interdigestive motor activity, particularly the phase III of MMCs, and the basal activity of LOS.¹⁵ Furthermore, a reduced or absent LOSP at night is recognized as the major contributor to oesophageal acid exposure.¹⁶ In our recordings during the whole 24 h, the tone of LOS was greater in patients assuming baclofen than in the control group. This seems to indicate that baclofen is able to reinforce the barrier function of the LOS independently from all its physiological changes occurring during the entire day.

Although the LOS motility seems to be the major target of baclofen, we also detected a reduction in both deglutitive activity and oesophageal primary waves. This aspect had been partially analysed⁵ with the demonstration of a fall in swallows without reporting the amount of primary peristaltic waves; it is reasonable to assume that the lower deglutitive activity elicits a parallel reduction in the number of primary peristaltic activity. Subsequently, it has been clearly demonstrated that baclofen reduces the oesophageal acid exposure both after acute and chronic administrations.⁸ We therefore hypothesize that the reduction in swallows and in primary peristalsis represent a consequence of the lower amount of acid inside the oesophageal wall¹⁰ with subsequent reduced stimulus for clearing motor mechanisms¹⁷ rather than a primary effect of the drug itself. Nevertheless a possibility exists of a concomitant direct action by baclofen on spontaneous and evoked swallows, as demonstrated in animals,¹⁸ and this possibility is reinforced by the lack of significant alterations in secondary peristalsis in our patients.

One potential limit of our protocol could rely on the sphinctometer we used to record the LOS, as the activity of the sphincteric region is preferentially monitored using the sleeve device.^19,20 However, it is well accepted that the sphinctometer, introduced in the early 1990s²¹ can identify TLOSRs with the same criteria used for the sleeve.²² The only limitation seems due to a slight underestimation of LOS relaxation episodes, particularly in case of low LOSP²³ whilst our patients were within the normal basal LOSP values and patients with basal LOS tone < 10 mmHg were not recruited. Therefore this bias, if present, could have affected both the baclofen and the control group in the same manner, thus maintaining the general sense of our findings.

The statistical power of our results is not very strong, in that the difference between the two groups remained around 0.02 and 0.05 significance. This in some way limits the potential clinical implications of baclofen as a therapeutic agent. What we wanted to test, however, was the duration of the action after multiple daily doses and it remains that the drug is able to reduce the occurrence of a reflex in a way and for a period to date not demonstrated yet with other agents and this must be considered an important starting point. The occurrence of side effects reported in many other studies when administering baclofen is surely a further major concern. In our protocol, the amount of such a phenomenon is not very high, as only two patients of the 14 totally studied described headache and dizziness, respectively, however, not so serious so as to be removed from the study. It has to be considered that our protocol was entirely performed on in-patients, who although allowed to do any kind of mild motor activities, were continuously monitored clinically by medical and paramedical personnel. Moreover, the drug was taken only for 1 day and it is well known that the majority of undesired effects usually occur after longer periods. It is clear that to date the side effects still represent the major limitating factor against a widespread use of baclofen in GORD population, at least in the formulations and at the dosages currently available. Nevertheless, the primary target of this study was to prolong the analysis of the baclofen effect on motor phenomena during a 24-h temporal window where our data seem to reproduce the beneficial effects detected in shorter monitorings.

In conclusion, baclofen is able to modify the oesophageal and LOS motor pattern not only after meals, but also in fasting conditions. Although the oesophagus of GORD patients is more exposed to the acidic stimulus in the postprandial phase, it is known that other situations can affect the natural course of the disease irrespective of meal assumption, such as nocturnal acid breakthrough²⁴ or night-occurring reflux episodes.²⁵ As these conditions are often scarcely responsive to the antisecretory treatment, the use of multiple daily doses of baclofen or of similar drugs with potentially fewer side effects could represent a further therapeutic option to be used in such patients.