This study was a meta-analysis to comprehensively evaluate the efficacy of GLP2 agonists in reducing PS requirements in patients with SBS. Our results revealed that GLP2 agonists can significantly reduce the burden of PS needs in patients with SBS by diminishing daily PN calorie intake, reducing weekly PS volume, and shortening the duration of PN infusion. In addition, this improvement in bowel function with GLP2 therapy was reflected as substantial increases in plasma citrulline levels and body weight, along with a decrease in fecal wet weight excretion.
Although PS is found to be lifesaving for patients with SBS, its long-term use is associated with a plethora of metabolic, septic, and mechanical complications [44, 45]. Reduced PN caloric requirements can be beneficial for a multitude of reasons. To start with, there is an increased risk of catheter-line associated bloodstream infections (CLABSI) in patients receiving PN [46]. Hypercaloric PN has been identified as an independent risk factor in the development of CLABSI [47, 48], and the recommended approach for reducing the risk of infectious complications is to lower the calorie demand. Moreover, a reduction in the daily caloric requirement leads to reduced PN infusion times, which maximizes patient mobility and greatly improves quality of life in patients with SBS[34]. According to our meta-analysis, daily PN calorie intake was significantly reduced from baseline after treatment with GLP2 agonists, which can be attributed to the pro-absorptive effects of the drug. Our analysis revealed a statistically significant reduction of 350 kcal/day from baseline, with some studies reporting a greater reduction than others. As demonstrated by a post-hoc analysis of the STEPS and STEPS2 extension trial, GLP2 therapy takes a longer duration to show maximal therapeutic effects, with a median response time of 4.3 months[15]. This is consistent with our findings which showed that the greatest reductions in daily PN caloric intake were reported by Marc et al. [29] and Joly et al. [27], in which the duration of treatment was 12 and 6 months respectively. Interestingly, our analysis also revealed that administering a higher dose of teduglutide than the recommended dosage of 0.05 mg/kg/day led to a much more diminished response to treatment. Patients receiving 0.1 mg/kg/day of teduglutide in Jeppesen et al [49] only exhibited a mean reduction of 106 kcal/day from baseline. It is still likely that the two active doses might have a similar efficacy, and that the discrepancy in response to treatment might just be due to the increased baseline energy and volume needs in patients receiving the higher dose. Further controlled studies exploring the effects of different dosages on GLP2 therapy in patients with similar baseline PS requirements are warranted to conclusively determine the optimal dosage required for maximum therapeutic benefit.
As a primary outcome measure, we also evaluated the decrease in weekly PN volume with GLP2 therapy, and our analysis revealed a reduction of approximately five liters per week. Studies focusing on pediatric populations or patients on long-term therapy demonstrated the most substantial reductions in PN volume. Lambe et al.'s study, which included pediatric subjects over a 48-week duration, reported an impressive reduction of over 27 liters per week, while Mark et al.'s study, conducted in adults over one year, demonstrated a modest reduction of approximately 10 liters per week (50,51). In contrast, Jeppsen et al.'s study, which targeted exclusively adult participants for a shorter duration of 12 weeks, reported the least reduction in PN volume of about 2.5 liters per week (52).
In our analysis, we also evaluated the proportion of patients achieving a ≥ 20% reduction in PS volume during GLP2 therapy known as “response rate” —a criterion widely recognized in literature to distinguish responders from non-responders (61).Consistent with the findings of a previous study by Fabio et al, which showed a significant increase in response rate between 6 months and 1 year, our analysis demonstrated similar findings: studies by Solar et al (62) and Joly et al (63), which exhibit a larger participant pool, consistently report that over 85% of participants achieved a > 20% reduction. These studies had an intervention duration ranging from 6 months to 1 year. Conversely, studies with fewer participants, namely Guz Mark et al (64) and Nakamura et al (57), demonstrated a lower proportion (nearly < 65%) undergoing ≥ 20% reduction, with these studies having a longer intervention duration, exceeding 18 months.
Previous literature suggests that patients with SBS who are completely weaned off PS have higher 10-year survival rates compared to those remaining dependent on PS (55). Therefore, gradual weaning off PN, achieved by reducing the frequency of PN days and substituting with an effective alternative, should be considered for individuals with SBS. Studies by Beth et al. and Lambe et al. demonstrate that substantial reductions in PN days correlate with a shorter duration of PN dependency (50,56). Conversely, studies showing less reduction in PN days indicate a more prolonged dependency as exemplified by the findings of Nakamura et al (57). Factors influencing the success of weaning off include the length of the remnant bowel and the presence of a colon (58). Without segmental reversal, weaning is rare in SBS patients with a remnant length < 70 cm (59). Studies with the least success in weaning, such as Jeppesen et al. and Beth et al., both report remnant bowel lengths falling below the 70 cm threshold (52,56). Conversely, studies reporting the highest rates of weaning, like Harpain et al., have longer remnant bowel lengths. David et al. reported a weaning success rate, but unfortunately did not provide information on short bowel length (60). Simlarly with reduction in weekly days off PN, a direct correlation is observed between greater remnant small bowel length and the effectiveness of teduglutide therapy. The greatest reduction was observed in Beth et al[50], which reported the highest reduction of 4 days per week and had a remnant small bowel length of 75 ± 23 cm. Conversely, studies with the least reduction, such as Guz Mark et al[51], reported just 1 day per week reduction and had a remnant bowel length of 23 ± 7 cm.
Our meta-analysis also demonstrated that GLP2 therapy significantly reduced fecal output by half a kg in patients with SBS. This improvement in intestinal wet weight absorption is particularly beneficial for patients with colon in-continuity, fecal or high output stomas, thus enhancing their social interaction and functionality. The greatest reduction in fecal wet weight excretion was reported by Jeppesen et al. (66), however, 9 out of 11 included participants were on antidiarrheal medication during the course of treatment introducing biasness. Furthermore, treatment duration was significantly longer in these participants, and during the washout period, the fecal wet weight output reverted back to baseline. This may suggest that the therapeutic effect of GLP2 agonists might not be permanent, and that long-term, possibly life-long therapy might be indicated. We encourage future trials with larger sample sizes to explore whether the beneficial effects of GLP2 therapy persist in the long term, particularly after treatment.
In the presence of stable or reduced PS requirement, an increase in the body weight of patients points towards an improvement in their nutritional status. Our analysis revealed a significant increase in body weight of SBS patients by approximately one kg. Whether this increase in body weight is due to an improvement in hydration status or an increase in fat mass remains to be determined. Thus, further studies including a concomitant measure of total body water are warranted to afford more clarity on the matter.
Previous studies indicate a strong link between plasma citrulline levels and residual small bowel length, particularly in patients with SBS (67). This can be attributed to the fact that circulating plasma citrulline is predominantly produced by the enterocytes of the small intestine (68). Our findings reveal a statistically significant increase in plasma citrulline levels with GLP2 therapy, likely due to the intestinotrophic effects of the drug.
5.1 Limitations
The results of our meta-analysis should be interpreted in light of certain limitations. Variations in patient’s inclusion criteria, nutritional status, spontaneous oral energy intake, and baseline PS requirements existed among studies, which likely contributed to the heterogeneity in our analysis. In addition, background or adjunctive therapies were not accounted for in our analysis which could be a possible cause for heterogeneity. The dosage and type of GLP2 agonists, as well as the follow up duration, were not standardized across studies, which may have introduced variability in the intervention. Subgroup analyses to account for these factors could not be performed owing to the small number of studies. Furthermore, we exclusively included articles published in the English language, potentially introducing a language bias in our study.