Clinical Recommendations to Manage Gastrointestinal Adverse Events in Patients Treated with Glp-1 Receptor Agonists: A Multidisciplinary Expert Consensus

1Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have represented a paradigm shift in the treatment of type 2 diabetes (T2D) and obesity. The incretin effect induced by GLP-1 RAs allows for glycaemic control in a glucose-dependent manner more efficiently than other therapeutic classes without increasing the risk of hypoglycaemia. Interestingly, some of them are able to cross the blood–brain barrier and act on the brain to stimulate satiety, which leads to food intake reduction and, consequently, body weight loss, which occurs at the expense of fat mass. As a result, the risk of progression to T2D decreases, and improvements in lipid profile, blood pressure or sleep apnoea have been observed. GLP-1 RAs also exert pleiotropic actions with metabolic, hepatic, renal and cardiovascular beneficial effects. Of note, a recent meta-analysis encompassing those clinical trials focused on the cardiovascular safety of GLP-1 RAs found significant reductions in MACE3 (a composite of cardiovascular death, non-fatal myocardial infarction, and non-fatal stroke), hospitalization for heart failure and progression of chronic kidney disease. The currently commercialized GLP-1 RAs with indications for T2D or obesity are summarized in Supplementary Table S1.

2Ample clinical evidence arising from trials and real-world scenarios highlights that the most frequent adverse events (AEs) associated with GLP-1 RA are those of a gastrointestinal (GI) nature, namely nausea, vomiting, diarrhoea and constipation. According to the literature addressing clinical trials, GI AEs usually develop in 40–70% of treated patients, although they have sometimes been reported in up to 85%.

3GI AEs arise irrespective of the half-life (long/short action) or route of administration (subcutaneous/oral) of the chosen GLP-1 RA. They are usually transient, typically starting during the dose-escalation period and generally resolving shortly after the maintenance dose is reached, and, in most cases, they are mild to moderate in severity. A recent report summarizing the results of several trials reported that the majority (99.5%) of documented GI AEs in people with obesity on GLP-1 RA treatment were non-serious. A meaningful body of evidence in real-world settings roughly replicates these observations. Nevertheless, it is essential that patients and healthcare professionals (HCPs) are aware of the right procedures to follow to prevent GI AEs from arising or, if they occur, to mitigate their effects and improve adherence and persistence to the treatment.

4GI AEs may lead to the temporary or permanent discontinuation of GLP-1 RA treatment. Although interruption has been reported to occur in up to 12% of GLP-1 RA-treated patients (vs. ~2% in those treated with placebo), permanent discontinuations range between 1.6–6% of treated patients (vs. <1% with placebo), according to clinical trial programs. In the real-world setting, persistence with GLP-1 RA therapy has been found to range between 40% and 60% and between 34% and 67% at 180 and 360 days, respectively. The results seem to be better with once-weekly administered GLP-1 RAs compared to those requiring once-daily injections. The adequate management of GI AEs would potentially improve a patient’s experience while undergoing treatment with GLP-1 RA, preventing discontinuations secondary to GI AEs.

5The literature addressing the management of GI AEs in clinical practice associated with the use of GLP-1 RA from a multidisciplinary perspective is scarce. For this reason, we formed a multidisciplinary team consisting of HCPs of several specialties and a diabetes nurse educator (DNE) with experience in the management of patients treated with GLP-1 RA in their daily practice. The main goal of this document is to reach a consensus regarding patient education and procedures to minimize the frequency and intensity of GLP-1 RA-associated GI AEs. Hopefully, this multidisciplinary consensus may contribute to expanding our knowledge of the practical use of GLP1-RAs, providing useful tools for physicians to use when managing people with T2D or overweight/obesity with this therapeutic class. More importantly, the better management of GI AEs could potentially have a positive impact on the adherence/persistence, effectiveness of treatment and quality of life (QoL) among GLP-1 RA-treated patients.

6A multidisciplinary expert panel comprising three endocrinologists, two nephrologists, two primary care physicians (PCPs), two cardiologists, two internists and one DNE, all of them with clinical expertise in the management of people with T2D or obesity with GLP-1 RA-based therapies, convened across virtual meetings to reach a consensus. The participants performed a comprehensive literature search using prespecified keys (Supplementary Table S2). The experts agreed that GI AEs should be the main topic to focus on when addressing the optimal medical care of patients treated with GLP-1 RA. The proper management of such occurrences is of paramount importance since it contributes significantly to treatment adherence and persistence and may improve QoL. Hepatobiliopancreatic AEs are also discussed in brief because, although they are extremely infrequent, it is important to exercise caution with those patients with a history of pancreatic and gallbladder disorders. Finally, widely accepted myths and erroneous considerations concerning other AEs and profiles of subjects wrongly considered unfit to receive this medication are also reviewed to ensure that as many patients as possible can benefit from treatment with GLP-1 RA.

7As anticipated, the AEs that were most frequently associated with GLP-1 RAs are those of a GI nature. Table 1 shows a summary of their frequency of occurrence in phase III clinical trials. Among the most common GI side effects, namely nausea, vomiting, diarrhoea and constipation, nausea appears systematically as the most frequent event in all clinical trials. The prevalence of the other GI AEs is lower. Overall, the onset of GI AEs is slightly higher in those trials designed to assess the efficacy and safety of GLP-1 RAs in people with obesity, which may be ascribed to the fact that doses are higher than those used in clinical trials in people with T2D. It is worth mentioning that long-acting agents have been associated with less nausea and vomiting but with more diarrhoea, which might be explained by a more sustained effect of these compounds on GLP-1 intestinal receptors. Finally, it is worth mentioning that flatulence may occasionally appear, although studies reporting its frequency are lacking.

8Patient education in terms of how to take and deal with satiety once GLP-1 RAs are started is crucial for ensuring treatment compliance. It is accepted that persistence improves when weight is adequately managed and safe, straightforward treatments are used. GLP-1 RA-based treatments comply with these characteristics. The proper education of patients by HCPs on their expectations when initiating treatment, how to prevent AEs, and how to treat them if they appear is particularly important. Patients have to learn that, although GI AEs may occur, these will be transient and of mild/moderate severity in the majority of cases and that following specific dietary recommendations will relieve symptoms (Table 2, Figure 1, Figure 2 and Figure 3).

9Compliance with the guidelines contained in Table 2 will be useful in avoiding the onset of GI AEs and mitigating their intensity in those cases where they develop. Indeed, it is important to start low and go slow: the lowest dose of medication should be used to start, and HCPs should ensure that patients follow their instructions closely. In those rare cases of particularly intense and/or persistent GI AEs, HCPs can apply further measures (Figure 4): If GI AEs appear during the dose-escalating phase, HCPs could modify the planned schedule by implementing one or several of the following points:Extend the dose escalation phase duration (2–4 more weeks with the previous dose or temporary suspension).Avoid dose escalation while GI AEs persist.If a GI AE is experienced when moving up to a higher dose, go back on the lower one and stay on it for a few days. Then, increase the dose gradually, taking advantage of the multidose pen-injector when available.In the case of persistent tolerability limitations set a dose lower than the maximum one recommended by the technical data sheet as a maintenance dose. Withhold treatment temporarily until the resolution of AEs and then resume treatment.

10If GI AEs appear during the dose-escalating phase, HCPs could modify the planned schedule by implementing one or several of the following points:Extend the dose escalation phase duration (2–4 more weeks with the previous dose or temporary suspension).Avoid dose escalation while GI AEs persist.If a GI AE is experienced when moving up to a higher dose, go back on the lower one and stay on it for a few days. Then, increase the dose gradually, taking advantage of the multidose pen-injector when available.In the case of persistent tolerability limitations set a dose lower than the maximum one recommended by the technical data sheet as a maintenance dose. Withhold treatment temporarily until the resolution of AEs and then resume treatment.

11Extend the dose escalation phase duration (2–4 more weeks with the previous dose or temporary suspension).

12Avoid dose escalation while GI AEs persist.

13If a GI AE is experienced when moving up to a higher dose, go back on the lower one and stay on it for a few days. Then, increase the dose gradually, taking advantage of the multidose pen-injector when available.

14In the case of persistent tolerability limitations set a dose lower than the maximum one recommended by the technical data sheet as a maintenance dose.

15Withhold treatment temporarily until the resolution of AEs and then resume treatment.

16Appropriate management during the dose-escalation phase could help patients to develop tachyphylaxis, i.e., to avoid excessive delay in gastric emptying upon treatment with GLP-1 RAs, especially with the long-acting ones.

17Start a differential diagnosis procedure to rule out underlying conditions causing the symptoms or their exacerbation.

18Make sure that the patient understands and complies with the guidelines regarding diet habits (Table 2, Figure 1 and Figure 2).

19Start symptomatic treatment focused on the specific GI AE (see below).

20Switching to another GLP-1 RA may be considered. Although reported data from trials and real-world series do not show definitive differences between GLP-1 RAs in terms of tolerability (Table 1), there are studies claiming that the tolerability profile may vary between different compounds. In fact, the switching strategy has already been proposed in the context of the treatment of people with T2D. It is advisable to start the new GLP-1 RA at its lowest escalation dose. If the patient is on treatment with semaglutide, switching the route of administration, i.e., from s.c. to oral semaglutide or vice versa, may be an option.

21Real-world studies confirm that nausea is the most frequent GI AE that arises upon initiation of GLP-1 RA. The frequency varies among clinical trials, although it usually ranges between 15 and 50% (Table 1). Prevalence is higher during the first 4–5 weeks of treatment, when gastric emptying is more significantly delayed, decreasing thereafter. Symptoms are usually moderate and disappear 8 days or less after their onset. To alleviate symptomatology, the following measures are advisable:Comply with recommendations regarding intake habits and food composition (Table 2, Figure 1, Figure 2 and Figure 3).If symptoms still persist, consider anti-emetic and/or prokinetic medications (Figure 4). Domperidone (10–20 mg three to four times daily, oral dosage, not in children < 12 years) should be used rather than metoclopramide, especially in older patients, to minimize the risk of extrapyramidal side effects. Among substituted benzamides, cinitapride may be an alternative to metoclopramide. In the event that oral semaglutide is being used, a period of 30 min must elapse between the administration of both medications. If drugs to mitigate nausea (or other GI AEs) are needed for over a month when the maintenance GLP-1 RA dose has been reached, a dose reduction should be considered for the patient to tolerate the drug with no need for pharmacological support.

22Comply with recommendations regarding intake habits and food composition (Table 2, Figure 1, Figure 2 and Figure 3).

23If symptoms still persist, consider anti-emetic and/or prokinetic medications (Figure 4). Domperidone (10–20 mg three to four times daily, oral dosage, not in children < 12 years) should be used rather than metoclopramide, especially in older patients, to minimize the risk of extrapyramidal side effects. Among substituted benzamides, cinitapride may be an alternative to metoclopramide. In the event that oral semaglutide is being used, a period of 30 min must elapse between the administration of both medications. If drugs to mitigate nausea (or other GI AEs) are needed for over a month when the maintenance GLP-1 RA dose has been reached, a dose reduction should be considered for the patient to tolerate the drug with no need for pharmacological support.

24Vomiting occurs at a frequency that usually ranges between 5 and 20%, i.e., lower than that reported for nausea (Table 1). Symptoms have been reported to disappear in 1–8 days. Initial measures do not substantially differ from those already reviewed (Figure 1). In case of abnormally long persistence or unexpected severity, additional actions may be undertaken (Table 2, Figure 2, Figure 3 and Figure 4): Maintaining hydration is particularly important. Small amounts of food should be taken in more frequent meals.Consider anti-emetic and/or prokinetic medications. Domperidone should be used rather than metoclopramide, as explained above. In case of persistence and/or remarkable severity, and where the patient presents with dizziness, confusion and fatigue, standard procedures to clinically manage severe vomiting can be initiated. Although rarely, intravenous rehydration may be necessary.

25Maintaining hydration is particularly important.

26Small amounts of food should be taken in more frequent meals.

27Consider anti-emetic and/or prokinetic medications. Domperidone should be used rather than metoclopramide, as explained above.

28In case of persistence and/or remarkable severity, and where the patient presents with dizziness, confusion and fatigue, standard procedures to clinically manage severe vomiting can be initiated. Although rarely, intravenous rehydration may be necessary.

29If the above measures are not as efficient as expected, either a decrease in GLP-1 RA dose or a temporary interruption of the treatment may be considered.

30The frequency of diarrhoea also varies among trials, usually ranging between 5 and 25% (Table 1). Diarrhoea initiates during the first four weeks of treatment, after which the incidence notably decreases. Symptoms have been reported to last for about three days in people with obesity treated with GLP-1 RA. The patients must be encouraged to follow the specific recommendations to avoid/mitigate diarrhoea (Table 2, Figure 1, Figure 2 and Figure 3), which should already have been taught to them as part of their education prior to GLP-1 RA initiation. In the event of persistence, in spite of compliance with these and the other guidelines, probiotic and/or antidiarrheal supplements such as loperamide could be considered. GLP-1 RA may exacerbate diarrhoea in patients receiving metformin treatment, especially if they are also taking omeprazole. In such cases, a reduction in the dose of metformin may be advisable (Figure 4).

31Constipation generally occurs at frequencies in the range of 4–12%, i.e., lower than those documented with other GI AEs. Nevertheless, some trials have reported a prevalence of up to 25 and 35% in people with obesity (Table 1). Accordingly, real-world series confirmed that constipation occurs more frequently in patients with overweight/obesity than with T2D. The onset can be in the first 16 weeks of treatment, particularly during the first 28 days. Constipation symptoms have been reported to last longer than those of the other GI AEs. The symptoms persisted for a median duration of 47 days in people with obesity on GLP-1 RA therapy, which may be ascribed to the more chronic nature of the condition. Secondary to the feeling of gastric fullness after receiving the drug, patients tend to reduce water intake, which may predispose them to this AE. Patients should be advised to increase their mobility, and intake of water and fibre, as well as to consider the use of stool softeners (Figure 1, Figure 2 and Figure 3). Clinicians should not discard a reduction in the GLP-1 RA dose upon a worsening of symptoms (Figure 4).

32Pancreatobiliary complications have been documented, albeit seldom. A comprehensive review encompassing 30 trials focusing on GLP-1 RA safety in people with T2D concluded that the risk of gallbladder complications or acute pancreatitis (AP) associated with this medication was generally low. In people with obesity, the incidence of gallbladder-related events was always <3%. Cholelithiasis was reported in 0–<1% patients in the majority of cohorts, regardless of having T2D or obesity. Its association with GLP-1 RA use, although unusual, has been linked to a combination of factors. Among these, the relevant weight loss often experienced by people with obesity may promote biliary lithogenicity, such as after bariatric surgery. Other explanations might be a direct action of the drug on biliary secretion and/or the modification of gallbladder motility. On the other hand, cholecystitis episodes were anecdotal. Finally, although higher circulating levels of lipase and amylase were reported in patients on GLP-1 RA therapy in many trials, the increases were rarely higher than three or five-fold the upper limit of normal (ULN), respectively, returned to normal levels after medication withdrawn, and were poor predictors of AP.

33A meta-analysis that included studies lasting for ≥24 months, with more than 9000 patients treated with GLP-1 RA during such period, did not find an association between GLP-1 RA therapy and AP either. Another one covering 55 randomized controlled trials and five observational studies with more than 300,000 participants, drew the same conclusion. Nevertheless, a recent meta-analysis encompassing up to 76 randomized controlled trials concluded that GLP-1 RA treatment was associated with a significant, albeit low, increased risk of gallbladder or biliary diseases (relative risk (RR), 1.37; 95% confidence interval (CI), 1.23–1.52). When the trials were stratified according to pathology, RR was 2.29 (95% CI, 1.64–3.18) in trials focused on people with obesity, near two-fold higher than that observed in trials concerning the treatment of T2D or other diseases (1.27; 95% CI, 1.14–1.43). Higher GLP-1 RA doses and, especially, the duration of treatment influenced the risk.

34The proportion of patients who developed pancreatobiliary complications such as AEs in the course of the more significant clinical trials is summarized in Table 3. AP was always experienced by less than 1% of patients treated with GLP-1 RAs. Furthermore, many of the cases were reported in subjects with a previous history of pancreatitis or gallbladder disease. Thus, caution must be exercised in patients with these antecedents. Ursodeoxycholic acid may be administered to patients with a history of cholelithiasis.

35HCPs must be aware of the possibility of these rare side effects to act early and avoid the complications related to volume depletion, such as severe renal failure. Patient education to recognize pancreatitis symptoms is also a recommended practice.

36If developed, pancreatobiliary complications should be managed according to the guidelines of each centre.

37When GLP-1 RAs started to be used, the first observations of certain occurrences led to misconceptions and erroneous conclusions regarding either the risks associated with the use of these drugs or the profiles of patients who would not benefit from these therapies.

38As previously stated, the most common GIs AEs included nausea, vomiting and diarrhoea. It has, therefore, been discussed whether the effects of GLP-1 RAs on weight loss are linked to these GI AEs. A study performed to assess whether or not there was a direct association between GI AEs and the extent of weight loss concluded that weight loss was largely independent of GI AEs.

39Many of these patients can benefit from GLP-1 RA therapy even though their dietary habits are poor. For example, patients with binge eating disorder or night-eating syndrome may benefit from GLP-1 RA therapy as long as they also receive proper psychological/psychiatric care. The exception are those patients with anorexia nervosa. GLP-1 RAs would also be contraindicated in patients with bulimia and self-induced vomiting until the psychiatric disorder was resolved using psychotherapy or pharmacologic means. Indeed, HCPs should take care to provide those patients with eating disorders who are going to start GLP-1 RA treatment with educational tips aimed at promoting a healthy lifestyle.

40The reduced cardiovascular and renal morbimortality associated with the use of GLP-1 RAs is beneficial for this particularly risky population. Caution when using GLP-1 RAs in older patients, especially those ≥75 years old, has been suggested to prevent the risk of sarcopenia. Although, as said above, weight loss induced by GLP-1 RAs comes mainly from reductions in fat mass rather than lean mass, nutritional advice and exercise programmes to achieve muscle strengthening are recommended to minimize sarcopenic risk.

41On the other hand, regarding the concern that GLP-1 RA doses indicated by the technical data sheet may be harmful to the elderly, it must be remarked that pivotal trials have repeatedly shown that the beneficial effects on weight loss and glycemic control can be achieved with no need to reach the recommended maintenance doses. An intermediate dose may be considered as the maintenance dose in fragile patients, thus minimizing risks associated with GI AEs. The Clinical Frailty Scale may sometimes be useful to guide dose adjustment. Finally, it must be remarked that GLP-1 RAs may be particularly beneficial for older patients receiving insulin treatment. The improved glycaemic control achieved with the new medication will probably make it possible to reduce insulin doses, which could decrease the risk of hypoglycaemic events and the consequences associated with these.

42A recently reported study focusing on patients with upper gastrointestinal disease, either chronic gastritis, gastroesophageal reflux disease, or both, undergoing GLP-1 RA treatment showed that this medication was well tolerated, with no need for dose adjustment. All of the AEs were mild-to-moderate, with no withdrawals because of them. Patients with chronic intestinal pathology are candidates for GLP-1 RA use. Furthermore, preliminary data suggest that these compounds might play a beneficial role in intestinal diseases. Nevertheless, the data sheets of several GLP-1 RAs do not recommend their administration to patients with severe gastrointestinal disease due to the limited experience in managing this condition.

43To address further practical considerations, four clinical scenarios are described that are aimed at providing clues for the better management of GI AEs in those patients at higher risk of these complications that are going to undergo GLP-1 RA-based therapy (Table 4).

44According to this multidisciplinary consensus, GLP-1 RAs can be considered easy-to-manage, highly effective medications with a good profile of tolerability to treat people with obesity or T2D. Patients should be educated to be aware that they may experience GI AEs, and that these will probably be mild to moderate in intensity and transient. Patients must also be educated on following a series of guidelines that will help them to prevent or, at least, mitigate GI AEs. HCPs must be aware that comprehensive dietary education, flexibility during the dose-escalation phase and appropriate symptomatic treatment of persistent GI AEs are factors of paramount importance to minimize the GI side effects associated with GLP-1 RA use.