1 Introduction

Antipsychotics are essential part of pharmacological treatment for schizophrenia and other psychotic disorders, such as bipolar disorders, dementia, anxiety, and depression (Solmi et al. 2017). Early synthesized antipsychotic drugs, including chlorpromazine, perphenazine, fluphenazine, trifluoperazine, and thioridazine, are known as first-generation antipsychotics. However, a large number of patients treated with first-generation antipsychotics developed severe side effects, including extrapyramidal adverse effects, tardive dyskinesia, and parkinsonism (Solmi et al. 2017; Lally and MacCabe 2015). Such adverse effects prompted the development of second-generation antipsychotics (SGAs), which have fewer side effects while maintaining the efficacy (Grajales et al. 2019).

Ileus is a serious gastrointestinal disease with potentially fatal consequences, such as shock, water-electrolyte disturbances, and intra-abdominal infections (Nielsen and Meyer 2012). Medication-induced ileus has been commonly seen in the use of anticholinergics, opioids, antihistamines, calcium antagonists, and antipsychotics (Isbister et al. 2001). Although SGAs have been widely used clinically, SGAs-associated ileus has only been reported occasionally from case reports and case-based reviews (McKinnon et al. 2009; Tyras et al. 2021). The safety assessment of antipsychotics in ileus has been reported in several studies previously (Nielsen and Meyer 2012; Hasegawa et al. 2024; Ingimarsson et al. 2018). However, these findings were derived from national medical information databases which might limit the generalizability of the results globally. Hence, a larger-scale real-world dataset is needed to reflect a more comprehensive situation of this rare adverse drug reaction (ADR). The United States Food and Drug Administration (FDA) defines “real-world data” as “data regarding the usage, or the potential benefits or risks, of a drug derived from sources other than traditional clinical trials” (Liu and Panagiotakos 2022). Since previous clinical trials were conducted in a limited number of participants with specific enrollment criteria, relevant results regarding SGAs-associated ileus might not provide an exhaustive presentation of real-world situations.

The FDA adverse event (AE) reporting system (FAERS) is a publicly available database of spontaneous AE reports, serving as an essential resource for real-world pharmacovigilance analysis (Khaleel et al. 2022). The method of pharmacovigilance analysis offers the advantage of large samples needed to assess rare ADR signals (Sakaeda et al. 2013). Thus, the aim of this study is to characterize and evaluate the risk signals of SGAs-related ileus through pharmacovigilance investigation using the FAERS database. We believe our study would provide referable evidence for the clinical monitoring and risk identification for SGAs-related ileus.

2 Materials and methods

2.1 Data acquisition

In this study, AE reports received from January 1 st, 2014 to December 31 st, 2023 were retrieved from the FAERS database (https://www.fda.gov/drugs/drug-approvals-and-databases/fda-adverse-event-reporting-system-faers). The AE term of “ileus” was identified with code 1021328 from the Medical Dictionary for Regulatory Activities (Version 24.0). According to the list of World Health Organization (WHO) essential medicines, the following 14 medicines were considered as SGAs: amisulpride, aripiprazole, asenapine, brexiprazole, cariprazine, clozapine, iloperidone, lurasidone, olanzapine, paliperidone, quetiapine, risperidone, sertindole, and zotepine (Todesco et al. 2023). Considering the very low number of reports for 9 of the SGAs (fewer than 15 reports each), we modified the SGAs list in our final analysis as follows: clozapine, risperidone, olanzapine, aripiprazole, and quetiapine (Fig. S1). In addition to monotherapy of the selected SGAs, we also analyzed combined therapy that contained more than one SGA. The detailed information of combined therapy was listed in Table S1.

The data mining tool OpenVigil FDA was applied to calculate the onset time of ADR, which was defined as the interval from the first time of SGAs treatment until the onset of ileus (Böhm et al. 2016). Additionally, each AE was assigned with a unique case ID, and cases owning the same ID were excluded from the analysis (duplicate cases). The detailed process of data collection was depicted in Fig. 1. Since all data of our pharmacovigilance study was originated from public databases with non-identifiable datasets, no ethical approval is required.

Fig. 1
figure 1

Flowchart depicting the data collection process

Full size image

2.2 Statistical analysis

In this study, descriptive analysis and disproportionality analysis were conducted to investigate the risk signals of ileus relevant to SGAs administration. In descriptive analysis, we analyzed the relationships between SGAs-related ileus and various variables, such as age (pediatric and adolescent (< 18 years), adult (18–44 years), middle-aged (45–64 years), and elderly (≥ 65 years)), gender (female and male), reporter type (consumer and healthcare professional), and reporting country/region.

Disproportionality analysis is a fundamental approach in pharmacovigilance research for detecting ADR signals of medications. In the current study, we employed two widely accepted algorithms to mine ADR signals, namely reporting odds ratio (ROR) and Bayesian confidence propagation neural network (BCPNN) (Bate et al. 1998; van Puijenbroek et al. 2002). ROR and information component (IC) were the two major measures in disproportionality analysis, and confidence interval (CI) is a range of upper and lower limits which describes the possible mean of a sample (Almenoff et al. 2007; Szumilas 2010; O'Brien and Yi 2016). ROR and IC were calculated according to the two-by-two contingency table (Table 1), and the detailed information of formulas and significant criteria were listed in Table 2. A statistically significant signal was considered if both the lower limit 95% CI of ROR (ROR025) > 1 and the lower limit 95% CI of IC (IC025) > 0 (Sakaeda et al. 2013). All analysis and visualization were performed using R studio (Version 4.1.0) and GraphPad Prism (Version 9.3.1).

Table 1 Two-by-two contingency table for disproportionality analysis
Full size table
Table 2 Algorithms, formulas, and positive signal criteria for disproportionality analysis
Full size table

3 Results

3.1 Descriptive analysis

After data processing, a total of 13,269,320 AE reports during 2014 ~ 2023 were obtained, with 492,802 reports related to SGAs medication. After removing the duplicate records, 419 cases of SGAs-related ileus were included in the final pharmacovigilance analysis. Figure 2 and Table 3 displayed the clinical characteristics of the 419 cases. The annual number of reported cases represented a fluctuating trend from 2014 to 2023. Specifically, the numbers of reported cases in 2014 and 2016 were the smallest (n = 27 each), while the number of reported cases in 2018 was the greatest (n = 68) (Fig. 2a). Clozapine constituted the vast majority of the suspected drugs (n = 131, 31.26%), followed by quetiapine (n = 99, 23.63%), combined therapy (n = 55, 13.13%), aripiprazole (n = 54, 12.89%), olanzapine (n = 51, 12.17%), and risperidone (n = 29, 6.92%) (Fig. 2b). Regarding reporting country/region, the United States reported the most AEs (n = 125, 29.83%), followed by Japan (n = 107, 25.54%), Germany (n = 39, 9.31%), Australia (n = 32, 7.64%), and the United Kingdom (n = 19, 4.53%) (Fig. 2c). The clinical characteristics of patients were displayed in Fig. 2d, the proportion of men was greater than that of women (46.78% vs 44.15%), and most patients were distributed in the age range of 45–64 years (n = 145, 34.61%). Furthermore, these cases were mainly reported by the healthcare professionals (n = 404, 96.42%), thus providing reliable data on SGAs-related ileus to some extent.

Fig. 2
figure 2

Descriptive analysis of cases with SGAs-related ileus. a The annual number of cases during 2014 ~ 2023; b The distribution of cases among SGAs; c The global distribution of reports; d The distribution of age, gender, and reporter type

Full size image
Table 3 Clinical characteristics of cases with SGAs-related ileus
Full size table

3.2 Reaction outcomes and actions taken with the drug

After removing the incomplete reports, a total of 219 cases were identified for the analysis of reaction outcomes. Figure 3a and Table S2 exhibited that most patients were observed in recovered state (n = 119, 54.34%), other outcomes included recovering (n = 56, 25.57%), fatal (n = 24, 10.96%), not recovered (n = 19, 8.68%), and recovered with sequelae (n = 1, 0.46%). Within each SGA treatment condition, quetiapine ranked first in the rate of fatalities (n = 10, 23.26%), followed by olanzapine (n = 3, 13.26%), clozapine (n = 6, 9.09%), combined therapy (n = 3, 8.33%), and aripiprazole (n = 2, 6.90%). As for recovered state, clozapine ranked first in proportion (n = 43, 65.15%), ahead of combined therapy (n = 22, 61.11%), quetiapine (n = 24, 55.81%), olanzapine (n = 11, 50.00%), risperidone (n = 9, 39.13%), and aripiprazole (n = 10, 34.48%). Risperidone showed the highest rate of recovering (n = 13, 56.52%), trailed by aripiprazole (n = 13, 44.83%), clozapine (n = 13, 19.70%), combined therapy (n = 7, 19.44%), olanzapine (n = 4, 18.18%), and quetiapine (n = 6, 13.95%). Moreover, olanzapine led in the rate of not recovered, (n = 4, 18.18%), followed by aripiprazole (n = 4, 13.79%), combined therapy (n = 4, 11.11%), quetiapine (n = 3, 6.98%), clozapine (n = 3, 4.55%), and risperidone (n = 1, 4.35%). There was only one case from clozapine treatment group containing outcome of recovered with sequelae.

Fig. 3
figure 3

The (a) reaction outcomes and (b) actions taken with the drug of cases with SGAs-related ileus

Full size image

After discarding reports with missing information, a total of 179 cases were included in the analysis of actions taken with the drug. As illustrated in Fig. 3b and Table S2, drug withdrawn (n = 120, 67.04%) was recorded in most cases, and the same trend was found within each SGA treatment group: aripiprazole (n = 16, 66.67%), clozapine (n = 30, 57.69%), olanzapine (n = 17, 68.00%), quetiapine (n = 26, 68.42%), risperidone (n = 12, 92.31%), and combined therapy (n = 19, 70.37%).

3.3 Time to onset of SGAs-related ileus

After excluding the reports with missing or inaccurate data, a total of 82 cases were analyzed for the time of onset. As demonstrated in Fig. 4 and Table S2, ileus was mostly observed within 30 days after the first-time SGAs administration (n = 35, 42.68%), and the same trend was found in treatment group of clozapine (n = 16, 42.11%), quetiapine (n = 3, 42.86%), risperidone (n = 2, 66.67%), and combined therapy (n = 11, 57.90%). Whereas in aripiprazole group, ileus was more frequently observed in 31 ~ 60 days after the first-time SGAs administration (n = 6, 75.00%). In olanzapine group, ileus was mostly found in 31 ~ 60 or 151 ~ 180 days after the first-time treatment of SGAs (n = 2 each, 28.57%). The median onset of SGAs-related ileus was 40 days (range 0 ~ 6070) for all cases. Quetiapine has the greatest median of onset (114 days, range 0 ~ 3838), while the median onset of risperidone was the smallest (0 day, range 0 ~ 43).

Fig. 4
figure 4

The time to onset of SGAs-related ileus

Full size image

3.4 Disproportionality analysis

As shown in Fig. 5 and Table 4, SGAs including clozapine (ROR = 3.30, 95%Cl: 2.77 ~ 3.92), quetiapine (ROR = 1.93, 95%Cl:1.58 ~ 2.35), olanzapine (ROR = 1.66, 95%Cl: 1.26 ~ 2.18), aripiprazole (ROR = 1.36, 95%Cl: 1.04 ~ 1.78), and combined therapy (ROR = 1.84, 95%Cl: 1.41 ~ 2.40) showed significant signals of ileus. The ROR025 of these SGAs were greater than 1, indicating an increased risk of ileus after SGAs administration. However, we did not observe a significant ADR signal in risperidone (ROR = 0.68, 95%Cl: 0.47 ~ 0.98) (Fig. 5a). Likewise, IC exhibited similar results that clozapine (IC = 1.69, 95%Cl: 1.42 ~ 2.00), quetiapine (IC = 0.93, 95%Cl: 0.76 ~ 1.13), olanzapine (IC = 0.71, 95%Cl: 0.54 ~ 0.94), aripiprazole (IC = 0.44, 95%Cl: 0.33 ~ 0.57), and combined therapy (IC = 0.86, 95%Cl: 0.66 ~ 1.13) were significantly associated with an increased risk of ileus, whereas the ADR signal of risperidone was not significant (IC = −0.55, 95%Cl: −0.38 ~ −0.79) (Fig. 5b).

Fig. 5
figure 5

Disproportionality analysis of (a) reporting odds ratio and (b) information component for SGAs-related ileus

Full size image
Table 4 Disproportionality analysis of SGAs-related ileus
Full size table

4 Discussion

To our knowledge, this is the first pharmacovigilance analysis investigating the links between the risk of ileus and SGAs medication. SGAs are effective agents in the treatment of psychotic disorders, and the commonly observed ADRs of SGAs contain metabolic abnormalities, prolactin elevation, dizziness and sedation, sexual dysfunction, cardiotoxicity, cataract, ileus, and constipation (Solmi et al. 2017). Although our results suggested a relatively low incidence of SGAs-related ileus in all cases of SGAs-associated AEs (419 out of 492,802), it warrants significant clinical attention due to the potential serious consequences of patients (Sarangi et al. 2021). In addition to the 419 AE cases we collected, several life-threatening cases of SGA-induced ileus have been documented (Every-Palmer et al. 2020; Osterman et al. 2017; Hibbard et al. 2009; Palmer et al. 2008; Shirazi et al. 2016), reflecting the necessity for clinical surveillance of this ADR. In our current study, the AE reports covered the populations that are typically excluded from clinical trials, such as children, pregnant women, and the elderly (Jacobson et al. 2024). Therefore, we investigated SGAs-related ileus in this large-scale dataset, aiming to broaden the understanding of this rare but serious ADR and minimize the risk of it.

Previous findings have revealed three major neurotransmitter mechanisms that might be involved in SGAs-related ileus: muscarinic cholinergic receptor antagonism, serotonin (5-hydroxytryptamine, 5-HT) receptor antagonism, and histamine H1 receptor antagonism (Xu et al. 2021). Acetylcholine is an intestinal excitatory transmitter that acts at muscarinic receptors on intestinal smooth muscle cells and adjacent Cajal cells (Harrington et al. 2010; Dome et al. 2007). The antagonism of muscarinic M1 and M3 cholinergic receptors can reduce intestinal contractility and delay intestinal transit, thereby causing ileus (Singh et al. 2022). Interestingly, we found that SGA with relatively lower muscarinic receptor affinity, such as aripiprazole, also displayed an increased risk of ileus, which might be linked to the antagonistic effects on 5-HT receptors (Every-Palmer et al. 2019; Keepers et al. 2020). 5-HT is capable of regulating gastrointestinal motility by acting on the vagal afferent pathways and slowing the colonic transport rate (Gershon 2004). Another mechanism contributing to ileus might be the antagonistic effect of SGAs on H1 receptor. Histamine acts on H1 receptors in gastrointestinal smooth muscles to promote intestinal peristalsis, and multiple SGAs have been shown to have H1 antagonism effect, including clozapine, olanzapine, and quetiapine (Keepers et al. 2020; Michl et al. 2014; Mandola et al. 2019). Of note, we should be aware that since SGAs can act on multiple receptors simultaneously, the mechanism of action of this ADR has pharmacologically complex properties (Taylor et al. 2005). Sedation is another common ADR of SGAs that cannot be ignored. The ADR of sedation can lead to gastrointestinal motility disturbances via the alterations of autonomic nervous motor patterns (Keepers et al. 2020; Arana 2000; Citrome 2017). The underlying pharmacological mechanisms may involve multiple gastrointestinal tract receptors, including GABA receptors, histamine receptors, and opioid receptors (Michl et al. 2014; Chang et al. 2020).

In the present study, we found the highest frequency and strongest risk signal of clozapine-associated ileus in FAERS database. We also exploratively collected data from other two widely applied ADR reporting systems, the EudraVigilance which primarily collects and analyzes AE reports within the European Economic Area (https://www.adrreports.eu/en/index.html) and the World Health Organization AE reporting system, namely VigiBase (https://www.who-umc.org). Interestingly, we found similar distribution patterns of SGA-related ileus among these three databases (Fig. S2). Indeed, ileus related to clozapine has been reported more frequently than other SGAs (Tyras et al. 2021; Tyras et al. 2021; Shirazi et al. 2016; Castillo-García et al. 2016; Khokhar et al. 2018; De Las et al. 2024). Although the underlying reason might be that clozapine has the highest anticholinergic affinity compared to other SGAs (Dean et al. 2023), we believe that potential reporting biases cannot be ignored. Due to influencing factors such as regulatory changes, media attention, Weber effect, and notoriety effect, medical staff and consumers would unconsciously pay more attention to ADR of clozapine-associated ileus (Neha et al. 2021; Matsuda et al. 2015). Our disproportionality analysis did not detect a strong link between the risk of ileus and risperidone (ROR025 = 0.68, IC025 = −0.38), which echoed the findings from a very recent pharmacovigilance study that risperidone has the least effect on gastrointestinal motility compared to other antipsychotics (He and Li 2025). Although risperidone has negligible affinity to muscarinic receptors, there were several case reports of severe intestinal motility inhibition caused by risperidone (Dome et al. 2007; Keepers et al. 2020; Rattehalli et al. 2016; Citrome et al. 2001). One possible reason could be the sympathetic hyperactivity triggered by risperidone-driven dopamine receptor blockade (Ramamourthy et al. 2013). Another reason could be that risperidone affects gastrointestinal motility through its relatively high affinity to 5-HT2 A receptors (Keepers et al. 2020; Guzel and Mirowska-Guzel 2022). Further, extensive evidence indicated that olanzapine has stronger muscarinic receptor affinity than quetiapine (Keepers et al. 2020; Dean et al. 2023; Lavrador et al. 2021). However, we found a much higher number of reports with quetiapine than olanzapine, which might support the view that there are growing cases of quetiapine misuse and abuse worldwide (Gjerden et al. 2017). Based on the above-mentioned findings derived from clinical data and pharmacokinetic results, we exploratively summarized the receptor binding properties and sedation effect of SGAs in Table S3. Moreover, we found that combined therapy did not show elevated risk of ileus versus aripiprazole (ROR025 = 0.93), clozapine (ROR025 = 0.41), olanzapine (ROR025 = 0.76), or quetiapine (ROR025 = 0.69), but displayed higher risk compared to risperidone (ROR025 = 1.72) (Fig. S3). Since the statistical power was limited by the small sample size of combined therapy group (n = 55), we performed descriptive analysis in reaction outcomes. We found that combined therapy showed a high rate in “recovered” state (61.11%), outperforming most monotherapies (except clozapine, 65.15%). As for fatality rate, combined therapy demonstrated a relatively low rate at 8.33%, only slightly higher than aripiprazole (6.90%) (Table S2). These findings collectively revealed the comparably safe profiles of combined therapy in terms of ileus risk and clinical outcomes. Taken together, we should reinforce the fact that proper monitoring is still needed while using SGAs other than clozapine.

Findings from previous retrospective analysis have suggested that the mortality rate of ileus was significantly higher in middle-aged and older adults compared to younger people (Koşar and Görgülü 2021, Elgar et al. 2022, Carroll et al. 2016), which is consistent with our data that the median age of patients with SGAs-associated ileus was 58 years. On the other hand, the connection between gender and SGAs-related ileus remains unclear, some studies observed that gastrointestinal hypomotility was independent of gender in patients receiving clozapine (Every-Palmer et al. 2016, 2019), whereas other study suggested that female patients receiving antipsychotics were more likely to develop constipation (Barbui et al. 2005). Although we did not find significant gender difference, it would be of interest to explore this point for follow-up studies. Of note, we noticed that although the median onset time of SGA-associated ileus was 40 days, a considerable proportion of AEs occurred more than 1 year after the first-time SGAs administration (n = 24, 29.27%), including clozapine, olanzapine, quetiapine, and combined therapy. Moreover, there were 3 cases involving clozapine medication exceeded 5 years in onset time of ileus, which was in line with previous evidence indicating that clozapine-associated ileus is highly likely to occur in the maintenance phase (more than 4 years) rather than acute phase (Nielsen and Meyer 2012). Considering the late onset property of SGAs-related ileus, we therefore recommend that healthcare professionals should increase their vigilance on patients with long-term use of SGAs. Several other SGAs-related factors of ileus could not be ignored either. Patients receiving SGAs medication tend to have sedentary lifestyle and low fiber diets which are considered as risk factors of ileus (Rognoni et al. 2021). Further, patients with schizophrenia are relatively insensitive to pain, which might reduce their somatic awareness of ileus (Coronado et al. 2021). As such, high-fiber diet, adequate fluid intake, and physical exercise would be beneficial in reducing the risk of SGAs-related ileus (Camilleri et al. 2017).

Regarding the dosage of SGAs, we found only a few cases marked with SGAs overdose (n = 19, 4.53%), thus indicating that our findings mostly reflected the profile of ADRs in the routine-dose SGAs scenario. Interestingly, we found 16 of these cases contained quetiapine extended-release dosage form. Since sustained/controlled-release form usually contains high dose of medications, patients are more likely to orally overdose themselves without a proper guidance from healthcare professionals (Ni et al. 2023). In addition, one literature review revealed that anticholinergic ADRs of SGAs appeared to be dose-dependent, including clozapine, olanzapine, and quetiapine (Yoshida and Takeuchi 2021). Unfortunately, due to our limited access to the specific dosage information or serum drug levels in databases, the dose-dependent profile between SGAs and ileus need to be further validated in the future.

We also identified several concomitant medications with potential risk of ileus in Table S4, including ondansetron, diphenhydramine, haloperidol, metoclopramide, benztropine, morphine, and fentanyl (de Alvarenga et al. 2017; Maheshwari and Sood 2023; Karunarathna 2024; de Boer et al. 2017). Besides, anticholinergics used to treat extrapyramidal side effects of schizophrenia patients might also increase the risk of ileus (Ali et al. 2021; Bolden et al. 1992). However, a firm influence of confounders (e.g., concomitant drugs and comorbidities) was difficult to determine as the detailed medical records were not reachable, such as administration timing, dosing sequence, blood concentration, allergy history, and drug interactions. Future investigations integrating comprehensive medical records, including patient baseline condition, medical history, and medication timeline, would enable deeper exploration of the confounding factors.

Despite the potential seriousness of ileus, knowledge about its clinical strategies or preventive measures remained poor, even among psychiatric clinical staff (De Hert et al. 2016). Based on our results of actions taken with the drug and the experience from other case reports (Tyras et al. 2021; Chiang and Lan 2018), we suggest that the focus of ileus management should be on discontinuing or switching SGAs to prevent potential fatalities. Early symptoms of ileus normally include abdominal pain, vomiting, abdominal distension, and nausea, which should be closely monitored during patient care (Vilz et al. 2017). Prophylactic laxatives, adjustment of electrolyte or acid–base disturbances, and early enteral feeding are considered effective for ileus (Vilz et al. 2017). Moreover, imaging of abdominal plain film, ultrasound, and computed tomography would be helpful for diagnosis, whereas the decision of surgical intervention should be made by experienced surgeons (Jang et al. 2010).

This study should be interpreted within its limitations. On the one hand, since that FAERS relies on voluntary reporting, underreporting, misreporting, and selection biases are thereby inevitable, which resulted a reduction in the number of cases collected. In addition, due to the lack of detailed clinical records and medication history of patients, the potential confounders of concomitant drugs and comorbidities were rarely analyzed. Therefore, education and training among healthcare staff and patients would be of great value to raise their awareness and improve the quality of AE reports. On the other hand, due to the intrinsic limitation of FAERS database, the data collected only allowed us to generate a statistical association of drug-AE correlation rather than a definite proof of causality. Further investigations integrating other data sources, such as electronic health records, medical insurance database, as well as causality assessment tools of WHO-Uppsala Monitoring Center scale and Naranjo scale, would provide more comprehensive safety assessment (Rezende de Menezes et al. 2021; Naranjo et al. 1981). Moreover, the newly developed machine learning model would be promising to optimally analyze FAERS data together with the tools and data sources described above (Ali and Aoun 2023).

5 Conclusion

In this study, we conducted a real-world pharmacovigilance analysis of SGAs-related ileus using the FAERS database. Our results suggested that SGAs administration increased the risk of ileus, and the characteristics of AEs would help clinicians understand the potential risk signals of SGAs-related ileus. Moreover, our results in the disproportionality analysis would assist clinicians in selecting SGAs for patients with relevant medication history, and provide valuable insight into the prognosis and safe use of SGAs.