Abstract
Keywords
INTRODUCTION
The pathological hallmark of idiopathic Parkinson’s disease (PD) is the formation of Lewy bodies or Lewy neurites comprised of anomalous accumulation of the hyperphosphorylated alfa-synuclein (α-syn) protein [1, 2]. According to the model of Braak et al. [3], the propagation of the pathology in the brain starts from the dorsal motor nucleus of vagus and anterior olfactory nucleus, which has further raised the discussion that the primary cause of vagal accumulation could be the enteric nervous system by means of retrograde transportation of α-syn by preganglionic fibers of the vagal nerve [4]. Detection of the α-syn protein in the entire enteric nervous system in PD patients [5] together with the observation of autonomic dysfunction in prodromal PD [6] makes this assumption highly plausible.
It has been reported that the appendix vermiformis is enriched with α-syn containing mucosal nerve fibers more than any other structure in the gastrointestinal (GI) tract [7]. Moreover, lack of a blood-tissue barrier in this mucosal region may also increase the predisposition towards α-syn aggregation in the central nervous system (CNS). Thus, it has been suggested that appendectomy might decrease the exposure of brainstem to α-syn and reduce the risk or delay the onset of PD [7]. The aim of this study was to test this assumption by including patients with PD (as an α-synucleinopathy), patients with non-α-syn parkinsonism and control subjects. Group comparisons were made and the disease specific features in PD patients with and without appendectomy were assessed in order to test the following hypotheses: Patients with PD have a lower prevalence of appendectomy than the other groups. In a given time, the probability of PD diagnosis in individuals with appendectomy is lower than the probability in those without (hazard ratio (HR) <1). PD patients with previous appendectomy have delayed symptom onset or milder disease severity compared to PD patients without. The interval between appendectomy and the first symptom as well as the age of appendectomy reflects the magnitude of exposure to α-syn, and therefore predicts disease severity in the PD patients with appendectomy. Individuals with early appendectomy (<20) have a lower risk of PD diagnosis compared with other diagnoses, and age of appendectomy is related with disease severity in PD patients.
METHODS
In this study, a retrospective evaluation was made of the electronic database records of patients referred to the Movement Disorders Unit of the Department of Neurology, University of Ankara between January 2007 and May 2016. Approval of the study was granted by the Local Ethics Committee and all procedures were in accordance with the most recent version of the Declaration of Helsinki.
Study cases
The subjects were divided into three groups of PD, (non-α-syn) parkinsonism and controls (neurological or non-neurological patients without movement disorders). All patients with PD were diagnosed according to the UK Brain Bank Criteria [8] by an experienced movement disorders specialist (MCA). Apart from the PD group, another group of parkinsonism but without α-synucleinopathy was included into the analyses. The goal in including the non-α-synucleinopathy parkinsonism group was to test the proposed pathophysiological mechanism independent from the clinical symptoms. This group was comprised of atypical parkinsonism and secondary parkinsonism such as vascular or drug-induced parkinsonism, progressive supranuclear palsy, corticobasal syndromes, normal pressure hydrocephalus and Wilson’s disease. Patients with multisystem atrophy (MSA) and essential tremor (ET) were excluded in order to avoid misclassification bias and undiagnosed PD respectively. The control was group composed of patients referred to the movement disorders unit but who lacked parkinsonism and had stroke, dementia, depression, polyneuropathy or ataxia.
The history of appendectomy was obtained from the history given by the patient (or a family member) using a standard form, on which the patient provided information about past surgical operations including the date (written as year). Information with regard to age, gender and age at symptom onset was also obtained. The interval between appendectomy and the first symptom of the disease was also calculated. In patients with PD, disease severity was assessed by Hoehn and Yahr (H&Y) Scale. The daily levodopa equivalent dose (LEDD) in the last examination was also recorded.
Statistical analyses
Groups of patients were compared with regard to demographics and disease characteristics. Distribution of the data was assessed by histograms and the Kolmogorov-Smirnov test. In order to test the first hypothesis and group demographics, the data of the three groups were compared using one-way ANOVA, with further between-group comparisons by post-hoc tests of Tukey or Games-Howell according to the violation of the test assumptions. The second hypothesis was assessed by calculating the HR for appendectomy using Cox regression. Since the standard Cox regression model does not take the timing of the exposure (in this case appendectomy) into account, a Cox regression model with time-dependent covariates was applied. For the third hypothesis, PD patients with and without appendectomy were compared in respect of age at symptom onset, disease severity and LEDD using the independent samples t-test. Prediction of disease severity by the interval between appendectomy and first symptom within the PD group with appendectomy (fourth hypothesis), was assessed using proportional odds ordinal regression model with the H&Y scale as dependent variable, and age and appendectomy-first symptom interval as covariates. Finally, disease severity between PD patients with early (≤20) vs. late (>20) appendectomy was compared using logistic regression with age and gender as possible confounders. Relative risk of PD diagnosis in early appendectomy group was also calculated. With respect to previous literature [9], additional explorative comparisons were also performed after dividing the PD group as early (≤55) and late (>55)-onset. The threshold of significance was set to p < 0.05. All analyses were made using SPSS Statistics 22.0.0 software (SPSS Ltd., Chicago, IL).
RESULTS
Between January 2007 and May 2016, the data of 1760 patients from the data bank were evaluated. Following the exclusion of 7 participants who could not be diagnosed and 128 patients from the parkinsonism group with diagnosis of ET or MSA, a total of 1625 patients remained for the analysis, of whom 839 had PD, 633 had parkinsonism and 153 were controls. The parkinsonism group was significantly older than the other two groups and the percentage of males was significantly lower in the control group (Table 1).
Demographic data and disease characteristics of the groups
Data are shown as mean (± SD) unless stated otherwise. n, number. cChi square. TPost hoc test-Tukey. gPost hoc test- Games-Howell. tIndependent samples t-test. *Including patients without appendectomy. AParkinson’s disease group, BParkinsonism group, CControl group.
A total of 134 (8.2%) patients had a history of appendectomy, comprising 55.5% males (p = 0.49). The prevalence of appendectomy was not significantly different between the groups (PD: 8.2%, parkinsonism: 8.4%, controls: 7.9%, p = 0.98). Regarding the history of appendectomy, the odds of having PD against other diagnoses (OR) were not significantly different (OR = 0.99, 95% CI = 0.70–1.4, p = 0.96). When the timing of the appendectomy was taken into account using the Cox regression with the time varying exposure method, appendectomy was not determined to have any effect on the outcome of PD diagnosis (HR = 1.12, 95% CI = 0.87–1.44, p = 0.39). Further details are shown in Table 1.
The comparison of the PD patients with (n = 69) and without (n = 770) appendectomy revealed no significant differences in respect of symptom onset, disease duration, disease severity or daily LEDD (Table 2). The ordinal regression analysis within the PD group, which was used to test if the appendectomy-first symptom interval significantly predicted the severity of the disease showed that disease severity could be significantly predicted by age with an OR of 1.05 (95% CI = 1.01–1.1, Wald X2(1) = 4012, p = 0.045), but not by the appendectomy-first symptom interval (OR = 0.99, 95% CI = 0.95–1.03, p = 0.54).
Comparison of Parkinson’s disease patients with and without appendectomy
Data are shown as mean (± SD) unless stated otherwise. n, number. cChi square. t Independent samples t-test.
The probability of the PD diagnosis in the late appendectomy group (>20) was not significantly different than the early appendectomy group (≤20). (Relative Risk = 0.9, p = 0.54, CI = 0.64–1.27). To assess the effect of early appendectomy on LEDD and H&Y in PD patients, separate logistic regressions were performed accounting for age and gender as covariates. Appendectomy age (lower or higher than 20) was not significantly associated with LEDD (B = –0.001, p = 0.35 Exp(B) = 0.999, CI = 0.9–1.01) or H&Y stage (B = 0.69, p = 0.22, Exp(B) = 2, CI = 0.7–6.1). Further comparisons in early (≤55) or late disease-onset (>55) PD patients did not yield significant results (Data not shown).
DISCUSSION
The aim of this study was to investigate the relationship between appendectomy and PD. None of the tested hypotheses was accepted in our study, i.e. the comparison of appendectomy between the groups, and the comparison of disease specific features within the PD patients with and without a history of appendectomy did not show any effect of appendectomy on PD. Furthermore, when the age of the appendectomy and the period between appendectomy and the first symptom was taken into account, there was neither a decreased risk for PD diagnosis nor milder symptoms detected in PD patients with a history of appendectomy.
The putative mechanism of retrograde transportation of α-syn from the GI tract to the brain stem via the vagal nerve which was suggested by Braak et al. [4] has recently been strengthened by several studies. In one vagotomy study, Svensson et al. demonstrated that the decrease in risk for PD was associated only with truncal vagotomy, but not with superselective vagotomy [10]. Furthermore, they showed a duplication in the annual incidence of PD diagnosis in the general population compared to patients with truncal vagotomy after 20 years, (1.28 vs. 0.65 per 1000) indicating a stronger effect over time. In another study, Mendes et al. collected the data of 295 PD patients and found that a history of appendectomy may delay the onset of PD but only in patients with late disease onset (≥55 years) [9]. The current study did not confirm this latter finding. A population-based study performed by Marras et al. also found no risk but a slightly increased risk of future PD within 5 years, in patients who had undergone appendectomy compared to cholecystectomy in the past 10 years [11].It was argued that this finding may have been due to the increased contact with the health care system within the first 5 years of appendectomy, which could have increased the possibility of detection of parkinsonian signs. Interestingly another recently published study from Svensson et al. also found a small increase in the risk of PD after 10 years, in patients who had undergone appendectomy [12]. These surprising results from two population-based studies are difficult to interpret. Svensson et al. speculated about the possible role of an infectious agent causing appendicitis or an inflammation somehow leading to retrograde transportation of misfolded α-syn protein.
The significance of the appendix with regard to α-syn content of the GI system is based on the report that the appendix is enriched with α-syn particularly in the mucosal nerves rather than the submucosal or myenteric plexuses which then facilitates the exposure of the enteric nervous system by eliminating the blood-nerve barrier [7]. However, to what extent this mechanism contributes to the exposure of the brain stem is unclear, considering the fact that Lewy body pathology can be found in all sites of the GI system in PD patients [13, 14]. Moreover, the phosphorylated α-syn pathology seems to have a diminishing rostro-caudal distribution, i.e. it is abundant in the higher structures such as the submandibular gland and lower esophagus, but less frequent in the colon and rectum [15]. This evidence indicates that although the appendix is enriched with the α-syn protein content, access to the CNS via retrograde vagal transport could be achieved in any anatomical location of the GI tract. For example, in a rat model, intestinal injection of human α-syn was seen to result in the detection of the pathology in the brainstem after 48 hours [16]. Thus, it could be reasonable to argue that appendectomy or any other partial resection of the GI tract may not prevent the CNS from exposure to abnormal α-syn protein, although it may, to some extent, reduce the amount of aggregate in a given time. It is possible that only truncal vagotomy could entirely prevent the retrograde transportation of α-syn.
Given the fact that neurodegeneration precedes the clinical diagnosis of PD by decades, the surgery-first symptom interval should also be taken into account while evaluating the effect of appendectomy. It has been shown that Lewy body pathology in the GI tract is detected up to 20 years prior to the diagnosis of PD [13]. Therefore, one may argue about the lack of benefit from the appendectomy in patients with a relatively short appendectomy-first symptom interval since the α-syn accumulation in the brain stem could have already started before the removal of the appendix. In the current study, the mean interval between the appendectomy and the first symptom was approximately 30 years, which was assumed to be sufficient to include participants who underwent appendectomy before the beginning of the neurodegenerative process. Moreover, patients with early appendectomy (before the age of 20) did not differ from those with later appendectomy.
Several limitations of the present study need to be discussed. The major limitation is that there was no systematic prospective assessment in which appendectomy status was determined. The history of appendectomy relied on chart review methodology, which was predominantly dependent on the history given by the patient (or a family member), and therefore, was not immune to bias, since some patients might not have mentioned a previous appendectomy. Although the lifetime risk of appendicitis has been reported to be 7–8% [17], which is similar to our results, the possibility of missing surgical history can be considered an important limitation of our study. However, fewer than 15% had an H&Y score > 2.5 in the PD group and thus the possibility of missing history of appendectomy due to the accompanying cognitive impairment can be considered low. Second, the diagnosis of PD was based on the UK Brain Bank criteria and achieved by expert opinion, but was not approved by functional imaging. The majority of the PD patients had an H&Y score of 2, and in some the diagnosis could still be confounded. Third, there was a relatively smaller number (n = 69) of PD patients with a history appendectomy, which may have had insufficient statistical power to reveal an effect. Finally, smoking was not taken into account. Given that smoking is associated with increased risk of appendectomy [18, 19], any positive result could have included some amount of variance explained by smoking and therefore, would be biased. However, the large sample size, the inclusion of other parkinsonian syndromes and the detailed statistical evaluation are the strengths of this study. In conclusion, these results suggest that removal of the appendix does not alter the occurrence, time of onset or severity of PD.
CONFLICT OF INTEREST
The authors have no conflict of interest to report.