Association Between Irritable Bowel Syndrome and Asthma a Meta-analysis and Systematic Review

Introduction

Parkinson'southward illness (PD) is the second most common neurodegenerative disease in the world after Alzheimer's disease, which is characterized by static tremor, bradykinesia, and rigidity (1). 1 to two per 1000 of the population are affected past PD, and the incidence increases with age (two). Typical neuropathological features include the presence of Lewy bodies or Lewy neurites- containing aggregation of α-synuclein and the loss of dopaminergic neurons in the substantia nigra. The resultant depletion of dopamine in the basal ganglia leads to move disorders (iii). The exact mechanism of PD remains unclear, and studies point that it is the issue of a complicated interplay between genetics and the surround. Some risk factors such as exposure to pesticides, traumatic brain injury, melanoma history have been confirmed (4–half dozen). Due to the slowly progressive nature, PD oft begins several years before diagnosis (vii, 8). Some non-motor features such every bit dementia, gastrointestinal changes, and olfactory dysfunction are common in early PD and may be the precursor symptoms of PD, with more and more attention (9, 10). A growing number of studies accept found significant associations betwixt Parkinson'due south disease and multiple gastrointestinal symptoms that originate from changes in intestinal part and structure or in associated neural structures (xi, 12). Furthermore, the gut microbiota and its relevant metabolites in PD have changed (thirteen, 14). These findings support the gut-to-encephalon transmission hypothesis that the pathological process may spread from the gut to the brain (fifteen, 16).

Irritable bowel syndrome (IBS) is one of the almost common functional bowel disorders, afflicting more 9% of adults worldwide according to Rome 3 criteria, and is characterized past changes in bowel addiction associated with chronic or recurrent intestinal pain that is either relieved or worsened by defecation (17, 18). The underlying mechanisms of IBS include genetic factors, mail infectious changes, altered gut microbiota, immunological alterations, low-course mucosal inflammation, and changes in abdominal permeability (19). Its diagnosis requires the exclusion of known organic diseases with similar symptoms, such as inflammatory bowel disease, tumors, and coeliac disease (xx). In recent years, the human relationship betwixt gut microbiota and irritable bowel syndrome has aroused significant interest in research and clinical fields. Some studies take constitute that gut microbiota may exist a potential source and driving gene of clinical symptoms (21, 22). Furthermore, certain neurological symptoms unrelated to the alimentary canal, such as anxiety, depression and mood disorders, are mutual in patients with IBS, which cannot exist explained by health-intendance-seeking beliefs lonely (23–25). These associations back up that defines IBS as a gut-brain disorder and that the pathophysiology of IBS may contribute to brain alterations.

PD and IBS may exist related because the pathophysiology of both potentially involves the gut microbiota and the gut-encephalon centrality. Previous studies have demonstrated that constipation is a prodromal symptom of PD (9). Considering the pregnant differences in definition and epidemiological characteristics betwixt IBS and constipation, exploring the association between IBS and PD tin help further identify populations at loftier run a risk for PD, provide new approaches for prevention, diagnosis, and treatment, and optimize policy-making. A previous meta-analysis by Fu et al. showed no statistical association between IBS and PD, but newly published studies based on a large population plant that the clan seems to exist (11, 26). Therefore, we conducted this meta-assay and systematic review to make up one's mind whether IBS patients have a college take a chance of PD.

Materials and Methods

The present systematic review and meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) guidelines (27). The protocol for this written report is not registered.

Search Strategies

We systematically searched the PubMed, Embase, Web of Scientific discipline, and Cochrane Libraries on Apr 25, 2021 using subject words combined with free words to identify all studies that investigate the correlation between IBS and PD, without whatsoever search filters or language restrictions. A manual search for additional relevant reports using the reference lists of included manufactures and related reviews was also performed.

(1) Subject area words: irritable bowel syndrome; irritable colon. Free words: irritable bowel syndromes; syndrome, irritable bowel; mucous colitis; colon spasm; colonic diseases, functional; colonospasm; functional colonic diseases; irritable colon syndrome; mucomembranous colitis; spastic colitis; spastic colon; unstable colon.

(2) Subject word: Parkinson affliction. Free words: idiopathic Parkinson disease; Lewy torso Parkinson affliction; Parkinson'south disease; main Parkinsonism; paralysis agitans; idiopathic Parkinsonism; Parkinson dementia circuitous; primary Parkinsonism.

The detailed search strategy in each database is available in Supplementary Material.

Study Selection

Reports that met the following specific criteria would exist considered for inclusion: (i) The report population was patients with PD or IBS. (ii) In the instance-control study, cases with PD were diagnosed according to standard clinical criteria or identified by the diagnostic lawmaking in patient medical records. Controls were healthy subjects or people without previous PD diagnosis from the same population. (iii) The IBS accomplice in the cohort study was diagnosed according to Rome criteria or questionnaire data, or identified past diagnostic code in patient medical records. The other cohort was from the same population without a history of IBS. (iv) Study outcome reported the following measures of clan: odds ratio (OR), relative risk (RR), or gamble ratio (HR) with corresponding 95% confidence interval (CI), or sufficient raw data were provided to summate these ratios. (v) The study type was cantankerous-sectional, instance-control or cohort design.

Reports that run across any of the following criteria would be excluded: (i) studies not derived from new data, including meta-analyses, systematic reviews, reviews, messages, and comments. (ii) Case reports or case series with fewer than ten patients. (iii) Briefing abstracts that total-text were already published. (4) Reports with fewer cases in studies with overlapping data. (five) Single-arm studies lacking a control group. (6) Non-human studies, such as animal experiments.

Based on the inclusion and exclusion criteria described above, two reviewers independently reviewed the records initially retrieved. The exclusion reasons were recorded and cross-checked.

Data Extraction and Quality Assessment

For eligible studies, 2 reviewers independently extracted the post-obit data: first author, year of publication, region, age, sex, report period, study design, population source, number of cases and controls or unlike cohorts, diagnostic criteria of IBS and PD, adapted confounding factors, follow-up fourth dimension, and take chances estimates with corresponding 95% CI.

The cohort design version and case-control design version of the Newcastle-Ottawa scale (NOS) were used to evaluate the included cohort studies and instance-control studies, respectively, which evaluated the selection, comparability, and exposure (example-control study)/outcome (cohort study) (28). Report with a score of more than 6 was classified every bit high quality, otherwise every bit depression quality.

Whatever discrepancy from study selection to quality cess was resolved past consensus through discussion amidst all authors.

Statistical Analysis

In this study, Stata/MP sixteen.0 was used for all data analyses. Adjusted take a chance estimates with corresponding 95% CI were used equally the chief endpoints of involvement for pooling. Only when a study did non report an adjusted gamble estimate, the rough risk gauge would be extracted and annotated. Given that the incidence of PD is low, we assumed that OR and HR could be considered equally. So, all the run a risk estimates were treated as HR in our meta-analysis, which represents the relative risk of developing PD in IBS patients compared to controls. We used the changed variance method of DerSimonian and Laird to pool the adjusted HRs, and P-value ≤ 0.05 was considered statistically significant (29). Cochran's Q-test and Higgins' I ² statistics were performed to appraise heterogeneity between included studies. Sensitivity analysis was performed by excluding i study at a fourth dimension and and then repeating the analysis. Furthermore, we tested the robustness of the pooled results past comparing the results from random-effects and stock-still-effect models. Low heterogeneity was considered when I ² <l% and P > 0.1, and the fixed-furnishings model was used. Inversely, the random-effects model was used, and the source of heterogeneity was analyzed by reviewing characteristics of included studies. Subgroup analysis was performed according to study design, gender, age, follow-up time, and report quality. We applied Begg'southward and Egger'southward tests to assess publication bias. When the P-values of both Begg'south and Egger's tests ≥ 0.05, it was considered that there was no publication bias.

Results

Study Characteristics

The search strategy described above yielded 476 potential records, of which 53 came from PubMed, 236 from Embase, 162 from Web of Scientific discipline, 19 from Cochrane library, and six from reference lists of relevant articles. After excluding duplicated records and screening titles and/or abstracts, 452 articles were excluded, every bit they were reviews, systematic reviews, written report protocols, animal trials, letters, or comments. The remaining 24 reports were further assessed by a full-article review. Of these, eight did non report IBS, 6 did non study PD, three were conference abstracts that were already published in full, two did non have a command grouping, and cases in one (30) were included in 1 eligible study with a larger sample size. The list of articles excluded after a full-text reading is available in Supplementary Material. Finally, 4 studies involving 2,044,110 subjects were included (26, 31–33). Three of them were based on the nationwide or regional registers (26, 31, 32), and one was a case-control study based on patients recruited (33) (Effigy i).

www.frontiersin.org

Figure one. Flow diagram of the study select process.

Three big-scale studies were conducted in Taiwan region, Sweden, and Finland (26, 31, 32). Both the Taiwanese and Finnish study designs were cohort studies (26, 31). Sweden conducted nested instance-command and accomplice studies based on the Swedish Patient Register and Swedish Twin Registry, respectively (32). Just the PD cases of both were identified from the Patient Annals, and the number of PD patients included in the case-control study was more than that in the cohort report, that is, the cases in the accomplice report were included in the example-command report (32). Therefore, we included only the Swedish case-control study in the meta-analysis, except for the subgroup analysis by the study blueprint. Iii studies based on national or regional databases each included more than 100,000 subjects, and all PD patients were diagnosed after the onset of IBS (26, 31, 32), while the example-control study conducted past Lee et al. recruited a relatively modest group of 87 subjects and the sequence of PD and IBS diagnosis was unknown (33). Except for the cohort study in Sweden that used an algorithm involving self-reported abdominal symptoms based on the ROME Ii criteria to identify IBS cases (32), studies based on the nationwide or regional register all used the International Classification of Diseases codes to place PD and IBS cases (26, 31, 32). Study conducted by Lee et al. used Korean Rome III and U.k. PD Society Encephalon Bank to diagnose patients with IBS and PD, respectively (33). The median or mean age of patients in all studies was over 50 years (26, 31–33). The Taiwanese and Finnish studies had a median follow-upwardly of ~vi years (26, 31), the Swedish cohort had a median follow-up of fifteen.1 years (32), and the study by Lee et al. had a median follow-upward of 46 months (33). The detailed information is shown in Table 1.

www.frontiersin.org

Table 1. Main characteristics of the studies included in this study.

All accomplice studies and case-command study based on nationwide or regional annals had loftier quality with scores ranging from 8 to 9 (26, 31, 32). These studies were consistently of reasonable quality, with advisable participants option, clearly defined exposures, and adequate follow-up for outcomes. Nonetheless, the Taiwanese and Finnish studies did non control detection bias (26, 31). The case-control study conducted past Lee et al. had an NOS score of six, mainly due to the potential pick bias (33) (Table 2).

www.frontiersin.org

Tabular array two. The quality assessment of included studies.

The Association Between IBS and Take a chance of PD

Considering the relatively small sample size and high risk of bias in the Korean study, we removed this study from the overall quantitative meta-assay and included it only for subsequent qualitative and subgroup analyses. The pooled effect of three studies involving one,999,798 subjects showed a significant association between IBS and PD (26, 31, 32). Compared with those without IBS, the 60 minutes of PD onset was 1.48 (95% CI: 1.35–1.62, P < 0.001) amidst patients with IBS. The heterogeneity in this analysis was insignificant (I ² = 0.0%, P _{heterogeneity} = 0.585) (Effigy ii). All patients in this assay were diagnosed with PD after IBS onset.

www.frontiersin.org

Figure 2. Forest plot of the associations betwixt irritable bowel syndrome and risks of Parkinson's disease.

Subgroup Analyses

We performed meta-analyses of subgroup based on the study design, gender, and study quality (Table 3). Since the available information of age and follow-up time did not support quantitative analysis, we conducted a qualitative evaluation for them as comprehensive equally possible.

www.frontiersin.org

Tabular array 3. Subgroup analysis of the association between IBS and PD run a risk.

According to written report design, pooled issue of the cohort studies showed that the IBS accomplice had a significantly higher risk of PD than the non-IBS accomplice (60 minutes = 1.49; 95% CI: 1.31–1.69), with subtle heterogeneity (I ² = 0.0%, P _{heterogeneity} = 0.428) (26, 31, 32). The pooled result of case-control studies showed that there was also a significant association between IBS and PD (HR = 1.43; 95% CI: 1.26–1.62), with adequate heterogeneity (I ² = 47.8%, P _{heterogeneity} = 0.166) (32, 33).

In subgroup analysis according to gender, IBS was significantly associated with take chances of PD in both male and female subjects, with pooled HRs of 1.45 (95% CI: 1.25–ane.67, I ^two = 0.three%, P _{heterogeneity} = 0.367) and i.58 (95% CI: 1.27–ane.96, I ^two = 62.3%, P _{heterogeneity} = 0.071), respectively (26, 31, 32). The high heterogeneity may be due to the small-scale sample bias in the Finnish study. After excluding the Finnish study, the heterogeneity of female subgroup disappeared (I ^two = 0.0%, P _{heterogeneity} = 0.878), and the statistical significance all the same existed (Hr = one.43, 95% CI: 1.26–i.62).

Co-ordinate to the NOS score, excluding the study of Lee et al. to combine the results of low bias take a chance studies, the pooled Hr was 1.48 (95% CI: 1.35–1.62), with subtle heterogeneity (I ² = 0.0%, P _{heterogeneity} = 0.585), indicating that the results from high-quality studies were consistent (26, 31, 32).

In terms of impacts of age, in the study based on the Taiwanese population, information technology was constitute that the HR of PD increased by 10% with age (per i yr) (31). In the Swedish population, the risk of PD afterwards IBS diagnosis tended to be higher in the elderly (≥l years at the time of IBS diagnosis) than in young people (<50 years), although the interaction test was non statistically meaning (P = 0.11) (32). In the Finnish population, the authors reported only that the results were like to the main assay in subjects aged at least 50 years at the time of IBS diagnosis, merely data were not presented (26). The bear on of historic period was not assessed in the Korean study (33).

Studies in Taiwan and Finland performed stratified analysis for the clan between IBS and gamble of PD according to the follow-up time that started from the IBS diagnosis until the occurrence of PD (26, 31). The Taiwanese written report showed a significant association between IBS and PD in both groups with less than and over 2 years of follow-upward, but the adventure of developing PD was college during the early 2 years of follow-upward (Hour = i.77; 95% CI: 1.33–2.36 vs. 1.38; 95% CI: i.16–i.66, respectively) (31). In the Finnish population, the highest 60 minutes was observed during the showtime ii years of follow-up (HR = 2.96; 95% CI: 1.78–4.92), the relative departure betwixt the two cohorts diminished during the two to 5 years of follow-up (HR = 1.08; 95% CI: 0.61–1.93), and the Hour demonstrated an increase at follow-upwardly over 5 years (Hour = 1.53; 95% CI: 0.98–2.38) although non statistically significant (26). Furthermore, in the Swedish population, stratified analysis was likewise conducted according to the interval between IBS diagnosis and PD diagnosis (32). A meaning association between IBS and run a risk of PD was observed when follow-upwards was less than five years (Hour = ane.31; 95% CI: 1.06–1.61), more than 5 years (HR = one.53; 95% CI: 1.30–1.79), less than 10 years (HR = one.47; 95% CI: 1.26–1.72), and more than 10 years (HR = i.38; 95% CI: 1.xi–1.71) (32). Similar to age, the impact of follow-up was not assessed in the Korean study (33).

Evaluation for Publication Bias

The P-values of Begg's and Egger'southward tests were 1.00 and 0.59, respectively, indicating that there was no publication bias. Due to the pocket-sized number of included studies, the funnel plot was not carried out.

Sensitivity Analysis

Sensitivity assay was performed for pooled results by excluding one study at a time and using both random-effects and stock-still-effects models. Results were stable and the conclusion was non changed when excluding ane study at a time (Figure 3). Results from random and stock-still-furnishings models were inconsistent in the case-command design subgroup. By reviewing the included studies, we found that the command subjects included in the study past Lee et al. came from patients without a history of neurological disorders who visited the gastroenterology outpatient at their institution, which may be a greater probability of IBS (33). Therefore, the pooled result of example-control studies using random-effects model that gives greater weight to the subjects included by Lee et al. may exist unreliable due to the risk of patient pick bias. Except for this subgroup, the pooled results using dissimilar effects models were similar in other analyses (Table 4).

www.frontiersin.org

Effigy 3. The sensitivity analysis of the associations between irritable bowel syndrome and risks of Parkinson'south illness.

www.frontiersin.org

Tabular array four. Comparing of use of random-furnishings vs. fixed-furnishings models.

Discussion

This is the beginning meta-analysis and systematic review that focuses on the clan between IBS patients and risk of PD. Our study plant a 48% increased hazard of PD in subjects with IBS compared to those without IBS. At that place was little prove of heterogeneity. Based on high-quality studies involving more than 2 meg subjects, the results from this analysis can be considered a stringent and reliable estimate, despite the relatively small number of studies included.

Compared with the study of Fu et al. (11), we added two large nationwide sample studies from Sweden and Republic of finland (26, 32), and excluded the example-control study of Mertsalmi et al. with patients that might overlap with the Swedish nationwide written report (26, 30). In fact, the previous meta-analysis did not observe statistical clan betwixt IBS and PD mainly due to the study of Lee et al. (xi, 33). The study by Lee et al. did not focus on exploring the association betwixt IBS and PD and therefore may innovate a risk of bias in patient pick. In addition, due to the big heterogeneity, the random-furnishings model was used for their analysis, which resulted in a more conservative estimate (11). Due to the significantly increased data, the present meta-analysis obtained a more than authentic prediction interval, and the sample size included was sufficient to demonstrate the association between IBS and risk of PD.

Subgroup analysis showed that meaning increased chance of PD was observed in IBS subjects across genders. In the Finnish study, no college run a risk of PD was observed in male IBS subjects, possibly due to a smaller sample size and therefore lack of statistical power (26). Majority of included studies suggested that the 60 minutes for PD increased with the age at IBS diagnosis (26, 31, 32).

Unlike the association between inflammatory bowel disease and PD (34), we found that the report pattern did not seem to be the source of heterogeneity in each analysis, and the results of the subgroup analyses did not differ when divided again according to case-control and cohort studies, which farther demonstrated the stability of pooled results. In addition, the Swedish study conducted a sensitivity analysis adjusting for constipation, and the effect showed that the association betwixt IBS and PD remained and constipation could non entirely explicate such association (32).

It is important to determine the causal relationship between IBS and PD. Except for the written report conducted by Lee et al. (33), all PD patients were diagnosed later the IBS onset, which seems to endeavour to demonstrate that IBS is a gamble factor for PD (26, 31, 32). However, non-motor symptoms of PD often occur before the onset of motor symptoms, so IBS may be a manifestation of early on PD that has non been diagnosed. In the Finnish study, the HR for PD in subjects with IBS compared to those without IBS was highest during the get-go 2 years of follow-upward and the association decreased between 2 and 5 years of follow-up, which appeared to support the opposite causation (26). Notwithstanding, the 60 minutes increased subsequently more than 5 years of follow-up, although statistical significance was not observed, possibly due to reduced statistical power limited by a smaller sample size. When the definition for PD was more liberal to include more samples, statistical significance over again occurred (26). The association between IBS and elevated PD risk was nonetheless observed in patients followed for more than 2 years in the Taiwanese population (31). Furthermore, the Swedish report showed a 38% increased hazard of PD amongst subjects with IBS diagnosis for more than ten years (32). Therefore, the electric current evidence did not completely explicate IBS equally a premotor symptom of PD. In addition, the association between IBS and PD may exist the result of a mutual exposure.

In the included studies, except for the small-scale sample study conducted by Lee et al. (33), the results of three nationwide or regional studies adjusted for a multifariousness of possible misreckoning factors (26, 31, 32). However, only the Swedish study assessed the touch of surveillance bias on results (32). Subjects with IBS oftentimes accept a greater number of medical visits than those without IBS, then they are more than probable to notice their body changes and take before and greater probability of getting the PD diagnosis. Therefore, it is possible to overestimate the increased risk of PD in subjects with IBS during the whole menstruum of analysis. In addition, this may partly explain that subjects with IBS had the highest 60 minutes for PD in the first 2 years of follow-up compared to those without IBS. In the Swedish study, the association was adulterate afterwards adjusting the total number of hospital visits, but information technology remained pregnant.

The exact mechanism of interaction betwixt IBS and PD remains unclear. Many factors related to IBS, such equally depression-grade mucosal inflammation, allowed activation disorders, altered intestinal permeability, metabolic abnormalities, and disturbances of the neuroendocrine system, may include some of the fundamental elements of PD (19). Furthermore, these factors could make the gastrointestinal tract more vulnerable to pathogens, thereby becoming the initiation site for α-synuclein assemblage (22, 35). The correlation between the bacterial gut disturbances and the development and severity of IBS has been observed in previous studies, and the gut-encephalon axis plays a key role in the perpetuation of IBS, which are like to the hypothesis that PD originates in the gut (36–38). When intestinal dysfunction and/or intestinal flora are disturbed, it may cause local and systemic inflammation and enteric neuroglial activation, which will somewhen trigger the evolution of α-synuclein pathology (xiii, 39). Withal, further enquiry is needed to confirm the correlation, and reverse causality is possible.

There are some limitations in this written report. First, the number of included studies is relatively pocket-size, so it is unclear whether this connectedness exists in other ethnicities and regions in the earth. Second, although the population from the national or regional databases provided a big sample for the study. However, the data in these databases are non collected for specific research questions, so some potential confounding factors that may affect the chance of PD, such equally Body Mass Index, dietary habits, exposure to pesticides, coffee consumption, and medication utilise were not controlled. Moreover, it is unclear whether there are differences in the clan between the unlike subtypes of IBS and PD. Next, the sensitivity and specificity of using diagnostic codes to determine patients are express, and underdiagnosis and misclassification of IBS and PD are inevitable. Fifth, in that location is a time error in the diagnosis of IBS and PD, so the temporal relationship assay may non exist accurate. Finally, although Egger' and Begg' tests showed no publication bias, there may withal be potential publication bias due to the small number of studies.

Conclusion

The present meta-analysis indicated that there is a significant association between IBS and PD, and subjects with IBS take a college adventure of developing PD compared to those without IBS. This increased risk was observed in both genders and increased with the historic period at IBS diagnosis. However, it is unclear whether IBS is a risk factor for PD, and there may be a reverse causation. Future studies with large samples and long follow-up in different geographic areas are required to further confirm this clan and clarify the causation and pathological mechanism between IBS and PD.

Information Availability Argument

The original contributions generated for the study are included in the commodity/Supplementary Textile, further inquiries can be directed to the corresponding author/s.

Author Contributions

QS and ZS contributed to the conception and design of the study. XZ and QH searched the database. XZ, YY, and ML extracted data. ZS and RZ conducted data assay. XZ, ZS, and ML wrote and revised the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version.

Funding

This study was supported past the International Cooperation Project on new Traditional Chinese medicine for prevention and treatment of Parkinson's disease (00023129) and the project of Sichuan Administration of traditional Chinese medicine, 2021MS107 [czybf (2021) No. thirteen-130].

Conflict of Interest

The authors declare that the research was conducted in the absenteeism of whatever commercial or fiscal relationships that could be construed every bit a potential conflict of interest.

Publisher's Note

All claims expressed in this article are solely those of the authors and do non necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may exist made by its manufacturer, is not guaranteed or endorsed past the publisher.

Supplementary Material

The Supplementary Material for this article tin be found online at: https://www.frontiersin.org/articles/10.3389/fneur.2021.720958/full#supplementary-material

References

3. Braak H, Del Tredici G, Rüb U, De Vos RAI, Jansen Steur ENH, Braak East. Staging of encephalon pathology related to sporadic Parkinson's disease. Neurobiol Aging. (2003) 24:197–211. doi: 10.1016/S0197-4580(02)00065-9