Background. Our understanding of constipation, a common symptom in persons with Parkinson’s disease (PwPD), is based on small cohorts. The relationship between constipation and other non-motor and motor symptoms of PD at the outset and over time is unclear.
Aims. To evaluate the relationship between constipation and other symptoms at baseline and over time in persons with (PwPD-C) vs without constipation (PwPD-noC).
Methods. In the Michael J. Fox Foundation online survey of PwPD, 38,261 persons had a physician-confirmed diagnosis of PD of which 30,123 noted if they “experienced constipation (< 3 bowel movements a week) or strained to pass a stool in the last month”. We analyzed the relationship between constipation and (1) self-evaluated daily activities related to motor (Part II of UPDRS), other nonmotor symptoms (NMS), quality of life (QoL), comorbidity score, depression, and daily activities at baseline and the (2) development of new NMS over time. This analysis used uni- and multivariable logistic and Cox proportional hazards models.
Results. Of 30,123 PwPD, 22,186 (74%) had constipation. Compared to PwPD-noC, PwPD-C were older, less likely to have PD for <5 years, had a higher comorbidity score, worse depression, and a greater impairment in daily activities (Table). Constipation increased with each increase in severity of motor symptoms at the index survey (Figure, left panel). For example, the risk of constipation was greater in persons with slight vs no, mild vs slight, moderate vs mild, and severe vs moderate freezing of gait. At baseline, only selected NMS (eg, talking during sleep) were associated with constipation (Figure, center panel). However, at a median duration of 2.9 years after the index survey, the risk of developing all other new non-motor symptoms was greater in PwPD-C than in PwPD-noC even after adjusting for age at PD diagnosis and at index survey, sex, and other covariates at index date (i.e., cognitive impairment, UPDRS, NMS and comorbidity scores) (Figure, right panel). For selected symptoms that are recognized to be associated with PD, the HR (95% CI) of developing a new symptom in PwPD-C vs PwPD-noC was as follows: anosmia (HR 1.58 [1.49 – 1.67]), dysphagia (HR 1.67 [1.58 – 1.76]), urinary problems (HR 1.42 [1.34 – 1.50]), falls (HR 1.72 [1.62 – 1.81]), and talking in sleep (HR 1.84 [1.74 – 1.94]). For most NMS, this risk of progression was greater in PwPD who were older, had cognitive impairment, or more severe motor or NMS scores or comorbid conditions at the index survey.
Interpretation. Among PwPD, constipation is associated with more severe motor and non-motor symptoms and poor QoL at baseline and a greater risk of developing other non-motor symptoms over time. Prospective studies are necessary to confirm these findings and to understand the mechanisms for these differences between PwPD-C and PwPD-noC.

<b>Table. </b>Comparison of Persons with vs without Constipation in Persons with PD.

Table. Comparison of Persons with vs without Constipation in Persons with PD.

<b>Figure. </b>Relationship between Constipation and Motor Symptoms at Baseline (left panel), Nonmotor Symptoms at Baseline (center panel) and Nonmotor Symptoms in Future (right panel). <b>Left panel:</b> logistic regression analysis between constipation (dependent variable) and different severity of motor symptoms, age at index survey and sex (predictor variable). i.e., the association of severe vs moderate and moderate vs mild motor symptoms and constipation. <b>Center panel:</b> logistic regression analysis with constipation as the outcome variable and individual NMS score as the predictor variable. The reference group is an absence of symptom. <b>Right panel:</b> multivariate analyses adjusting additionally for age at diagnosis, age at first survey, sex, cognitive impairment, and index UPDRS score.

Figure. Relationship between Constipation and Motor Symptoms at Baseline (left panel), Nonmotor Symptoms at Baseline (center panel) and Nonmotor Symptoms in Future (right panel). Left panel: logistic regression analysis between constipation (dependent variable) and different severity of motor symptoms, age at index survey and sex (predictor variable). i.e., the association of severe vs moderate and moderate vs mild motor symptoms and constipation. Center panel: logistic regression analysis with constipation as the outcome variable and individual NMS score as the predictor variable. The reference group is an absence of symptom. Right panel: multivariate analyses adjusting additionally for age at diagnosis, age at first survey, sex, cognitive impairment, and index UPDRS score.

Rodolfo Savica

Appearances

CONSTIPATION IN A LARGE ONLINE COHORT OF PERSONS WITH PARKINSON DISEASE

Presenter

Speakers