Appendectomy and risk of Parkinson’s disease in two large prospective cohorts of men and women.

Abstract

Background:

Prior work on appendectomy and PD has produced mixed results.

Objectives:

We examined whether history of self-reported appendectomy was related to risk of incident Parkinson’s Disease in the Nurses’ Health Study and the Health Professionals Follow-up Study.

Methods:

We used the Cox Proportional Hazards Model to estimate the Hazard Ratio of Parkinson’s Disease associated with self-report of appendectomy in men and women. Among women, we estimated the Hazard Ratio of Parkinson’s Disease associated with appendectomy for appendicitis and incidental appendectomy.

Results:

In pooled analyses, self-report of any appendectomy was not related to Parkinson’s Disease risk: the Hazard Ratio of Parkinson’s Disease comparing participants who reported any appendectomy to those who did not, was 1.08 (95% Confidence Interval: 0.94, 1.23). In women, appendectomy for appendicitis, but not incidental appendectomy, was associated with a modestly elevated risk of Parkinson’s Disease (Hazard Ratio: 1.23 (95% Confidence Interval: 1.00, 1.50).

Conclusions:

Overall, this study suggests limited to no association between appendectomy and Parkinson’s Disease risk.

Introduction

PD is increasingly recognized as a systemic disease, with well-known effects on the peripheral nervous system, and particularly the gut. Recently, it has proposed that the initial misfolding of alpha-synuclein, a protein key to PD pathology, may occur in the gut and then spread to the brain via retrograde axonal transport^1–4. A reduced risk of PD was observed after truncal vagotomy in the Danish⁵, and to a lesser extent, in the Swedish⁶ national registries. Furthermore, there has been growing interest in the role of the gut-brain axis in PD^3,7, including the role of the gut microbiome^8–16.

In a recent report,¹⁷ the appendix was found to be particularly rich in alpha-synuclein, relative to other areas of the gastrointestinal system, suggesting this organ as potential point of initiation of PD pathology. Several studies to date have examined the association between appendectomy and PD with mixed results. In an analyses of the Danish National Registry, with 34 years of follow-up, appendectomy was associated with a modest 15% (95% CI: 3%, 27%) increase in PD risk¹⁸. In contrast, no association between appendectomy and PD risk was found in a registry-based study in Ontario, Canada, comparing 42,999 participants with appendectomy to those with cholecystectomy, or no procedures¹⁹.

In this study, we sought to examine the association between self-reported appendectomy and incidence of PD in two large prospective cohort studies, the Nurses’ Health Study and the Health Professionals Follow-up Study.

Methods

Study Population:

We investigated the association between self-reported appendectomy and risk of PD in two large prospective epidemiological cohorts: the Nurses’ Health Study (NHS) and the Health Professionals Follow-up Study (HPFS). The NHS cohort was established in 1976 when 121,700 female registered nurses enrolled by filling out a mailed health-related baseline questionnaire. The HPFS cohort was initiated in 1986 when 51, 529 male health professionals, age 40–75 filled out a similar mailed questionnaire. Both cohorts have been followed every two years with mailed questionnaires inquiring about health habits and disease outcomes. Follow-up rates have consistently been very high in both cohorts, averaging over 90% at each follow-up cycle.

Exposure Ascertainment:

We used information on self-reported appendectomy to define our exposure variable. In the NHS, participants were asked on the 1992 questionnaire whether they ever had physician-diagnosed a) appendectomy for appendicitis or b) incidental appendectomy. For both appendectomy types, participants were asked to mark the year during which the procedure took place (before 1976, 1976–1980, 1981–1984, 1985–1989, 1990+). In the HPFS, history of appendectomy was asked in 1986; participants who reported appendectomy were asked to indicate the date range when their appendectomy took place (before 1955, 1955–1964, 1965–1974, 1975–1979 or 1980 to present). Type of appendectomy (for appendicitis vs. incidental) was not asked in HPFS.

Ascertainment of PD Cases:

Procedures for PD identification and diagnostic confirmation have been previous described^20,21. Briefly, we identified incident self-reported cases of PD through biennial questionnaires in both cohorts. For each self-reported PD case, after obtaining permission from the study participant, we contacted the treating neurologist or internist and asked for confirmation of the diagnosis and a copy of the patient’s medical record. Cases were adjudicated by a neurologist specialized in movement disorders who reviewed the medical records; if a medical record could not be obtained, individuals were considered to have PD if the treating neurologist or internist indicated that the diagnosis of PD was definite or probable. Only definite or probable cases were included in our analyses.

Statistical Analysis:

We used the Cox proportional hazards regression model to calculate age- and multivariable-adjusted hazard ratios (HR) of PD comparing participants who self-reported any appendectomy to those who did not report appendectomy. Analyses were conducted separately in each cohort and the cohort-specific HR were then pooled. Person-time based on age in months was accumulated from baseline (1992 in NHS and 1986 in HPFS) until the date of first PD symptoms, death, last completed questionnaire, or end of follow-up (June 2012 in both cohorts), whichever came first. Multivariate analyses were adjusted for age in months, smoking (never, past, current) pack years smoking, with additional adjustment for PMH use (never, past, current) at baseline in the NHS. In the NHS, additional analyses were performed, estimating the HR for appendectomy for appendicitis compared to no appendectomy as well as the HR for incidental appendectomy compared to no appendectomy. We also conducted sensitivity analyses excluding the first four years of follow-up, in both cohorts.

The cohort-specific regression analyses, including data manipulation and Cox regression were conducted using SAS software (SAS Institute, Cary, NC) and the pooling of the NHS and HPFS cohorts was conducted in STATA. The IRB at the Harvard School of Public Health and Brigham and Women’s Hospital approved the study.

Results

We identified 1,152 incident confirmed cases of PD, 583 in the NHS and 569 in HPFS. Baseline characteristics of the cohort are presented in Table 1. Overall, any appendectomy was not related to future PD risk, in cohort-specific or pooled analyses. In pooled analyses, the HR of PD comparing participants who reported any appendectomy to those who did not, was 1.08 (95% CI: 0.94, 1.23). Among men, this HR was 1.04 (95% CI: 0.84, 1.30) and for women it was 1.08 (95% CI: 0.91, 1.30). We did not observe significant heterogeneity by gender (p-heterogeneity: 0.72). The results of analysis for any appendectomy are presented in Table 2.

Table 1.

Any appendectomy and risk of Parkinson’s Disease in women (NHS) and men (HPFS).

	WOMEN					MEN				POOLED		p-heterogeneity (gender)
		PY*	Cases	HR	95% CI	PY	Cases	HR	95% CI	HR	95% CI
No Appendectomy		1249191	406	Ref		842764	460	Ref
Any Appendectomy	Model 1	442700	177	1.09	(0.92, 1.31)	167509	109	1.02	(0.83, 1.26)	1.06	(0.93, 1.22)	0.62
	Model 2			1.08	(0.91, 1.30)			1.04	(0.84, 1.29)	1.08	(0.94, 1.23)	0.72

Model 1 adjusted for age in months at baseline.

Model 2 adjusted for age in months, smoking (never, past, current) pack years smoking, PMH use (in NHS only: never, past, current) at baseline.

^*Person-Years of follow-up

Table 2.

Appendectomy and risk of Parkinson’s Disease in women (NHS) by type of appendectomy.

		PY	Cases	HR	95% CI
No Appendectomy		1249191	406	Ref
Appendectomy for Appendicitis	Model 1	273470	125	1.23	(1.0006, 1.50)
	Model 2			1.23	(1.002, 1.50)
No Appendectomy		1249191	406	Ref
Incidental Appendectomy	Model 1	165512	49	0.86	(0.64, 1.16)
	Model 2			0.83	(0.62, 1.12)

Model 1 adjusted for age in months at baseline in 1992.

Model 2 adjusted for age in months, smoking (never, past, current), pack years smoking, PMH use (never, past, current) at baseline in 1992.

^*Type of appendicitis was only asked in NHS. This question was not asked in HPFS.

^**Participants who reported incidental appendectomy were excluded from analyses of appendectomy for appendicitis and participants who reported appendectomy for appendicitis were excluded from analyses of incidental appendectomy

In the NHS, among women, history of appendectomy for appendicitis was associated with a modestly elevated risk of PD (HR comparing women who reported an appendectomy for appendicitis compared to those who did not report any appendectomy was 1.23 (95% CI: 1.00, 1.50). Also, in the NHS, self-reported incidental appendectomy was not associated with risk of PD (the HR comparing women who reported incidental appendectomy compared to those who did not report any appendectomy was 0.83 (95% CI: 0.62, 1.12)). The results of analyses by type of appendectomy are presented in Table 2.

Results of the sensitivity analyses excluding the first four years of follow-up, were not materially different from our main results.

Discussion

Overall, in this study within two large cohorts of men and women, we did not observe a significant association between self-reported appendectomy and risk of incident PD. In pre-planned sensitivity analyses, we did, however, observe an elevated risk of PD among women who reported appendectomy for appendicitis compared to no appendectomy.

Our results are largely in agreement with prior work on appendectomy and PD, indicating no or only modestly elevated risk of PD associated with appendectomy. In a nationwide cohort within the Danish National Registry, Svensson et al.¹⁸ observed a slightly (14%) increased risk of PD associated with an operation code of appendectomy 10 or more years after the procedure. In a population based study using linkage to Ontario, Canada health insurance databases, Marras et al.¹⁹ compared hazard of PD between individuals having undergone appendectomy, cholecystectomy, or neither procedure, with 42,999 individuals in each group. The authors of this study observed an increased incidence of PD within 5 years of appendectomy, but no significant increase in risk after 5 years of follow-up. Marras et al. acknowledge that cases with incident PD within the first 5 years of the procedure likely already had the disease at the time the appendectomy was performed. To address this issue in our study, we conducted sensitivity analyses excluding the first four years of follow-up, in both cohorts. The results of these sensitivity analyses were not materially different from our main results. It would have been interesting to examine whether age at appendectomy modifies these findings, but we did not have sufficient data to be able to do so, because only 17% of the participants answered the question regarding the date range when their appendectomy was performed.

In a study of 295 PD cases, Mendez et al.²², did not observe a significant overall association between appendectomy and age of PD onset, but reported a possible delay in PD onset among older patients with appendectomy. However, this study was restricted to PD patients, and did not include a healthy comparison group.

The finding of an elevated risk of PD among women who reported appendectomy for appendicitis, but not among those who reported incidental appendectomy is consistent with the postulated role of inflammation or invasion by a bacterial pathogen in triggering PD pathogenesis. A recent study reported evidence of elevated interleukin-1αIL-1α), among other inflammatory factors, in PD patients²³, and elevation of IL-1 has also been noted in patients with acute appendicitis²⁴. It is possible that appendicitis in these women could have been the result of invasion by a neurotrophic enteric pathogen that contributed to triggering of PD pathology¹. Alternatively, appendicitis may have promoted a systemic inflammatory response that may have contributed to PD onset, such as the triggering of propagation of alpha-synuclein form the gut/appendix to the brain. Finally, the surgical removal of an inflamed appendix, as opposed to a non-inflamed, may have released harmful inflammatory pathogens into the surrounding gut environment. In theory, the differences in the findings between Canada and Denmark, discussed above could be due to a difference in the proportion of incidental appendectomies vs. appendectomies for appendicitis between the two countries, or in a difference in antibiotic use pre-or post-surgery. Rates of appendicitis have varied across geographic location over the past century, with somewhat higher rates in Denmark compared to Canada²⁵, however, to our knowledge data on rates of appendectomy for appendicitis vs incidental appendectomy are not available.

Strengths of our study include a prospective design with a long follow-up and a large number of neurologist-confirmed PD cases. The use of self-report of appendectomy is a potential limitation of the study. However, the cohorts in these studies comprise nurses and health professionals, and their self-report of medical diagnoses and conditions have proven reliable. A limitation is that the self-report of appendectomy was recorded at only one point in the study (1992 in NHS and 1986 in HPFS); we thus cannot exclude the possibility that some participants underwent appendectomy at later times. If the under-reporting of appendectomy was not differentially related to PD incidence, such misclassification of exposure would likely bias the results to the null.

Furthermore, although we observed a somewhat elevated risk of PD among women who reported appendectomy for appendicitis, data on type of appendectomy was not available in men, limiting our ability to perform such stratified analyses in the HPFS.

In summary, in this study that included two large prospective cohorts of men and women, history of appendectomy overall was not related to risk of PD. We found a suggestion of an elevated risk of PD associated with appendectomy for appendicitis.