Increased risk of Parkinson's disease in individuals hospitalized with conditions related to the use of methamphetamine or other amphetamine-type drugs

Abstract

Background

Since methamphetamine and other amphetamine-type stimulants (meth/amphetamine) can damage dopaminergic neurons, researchers have long speculated that these drugs may predispose users to develop Parkinson's disease (PD), a dopamine deficiency neurological disorder.

Methods

We employed a retrospective population-based cohort study using all linked statewide California inpatient hospital episodes and death records from January 1, 1990 through December 31, 2005. Patients at least 30 years of age were followed for up to 16 years. Competing risks analysis was used to determine whether the meth/amphetamine cohort had elevated risk of developing PD (ICD-9 332.0; ICD-10 G20) in comparison to a matched population-proxy appendicitis group and a matched cocaine drug control group. Individuals admitted to hospital with meth/amphetamine-related conditions (n = 40,472; ICD-9 codes 304.4, 305.7, 969.7, E854.2) were matched on age, race, sex, date of index admission, and patterns of hospital admission with patients with appendicitis conditions (n = 207,831; ICD-9 codes 540–542) and also individuals with cocaine-use disorders (n = 35,335; ICD-9 codes 304.2, 305.6, 968.5).

Results

The meth/amphetamine cohort showed increased risk of PD compared to both that of the matched appendicitis group [hazard ratio (HR) = 1.76, 95% CI: 1.12–2.75, p = 0.017] and the matched cocaine group [HR = 2.44, 95% CI: 1.32–4.41, p = 0.004]. The cocaine group did not show elevated hazard of PD compared to the matched appendicitis group [HR = 1.04, 95% CI: 0.56–1.93, p = 0.80].

Conclusion

These data provide evidence that meth/amphetamine users have above-normal risk for developing PD.

Introduction

Methamphetamine and other amphetamine-type stimulants (meth/amphetamine) comprise the second most widely used class of illicit drugs in the world (United Nations Office on Drugs and Crime, 2008). Such consumption patterns, along with serious concerns specifically about methamphetamine toxicity (Thrash et al., 2009), have had a major influence on drug policy legislation (e.g., U.S. Combat Methamphetamine Epidemic Act of 2005) (Sununu, 2005) and health service utilization in the United States (e.g., one-third of all recent publicly funded substance-abuse treatment episodes in California were due primarily to methamphetamine) (Substance Abuse and Mental Health Services Administration, 2010a, Substance Abuse and Mental Health Services Administration, 2010b). In addition, humans are exposed to licit amphetamine to promote wakefulness in narcoleptic patients, maintain alertness in armed forces personnel, facilitate weight reduction in the obese, and treat the symptoms of attention deficit hyperactivity disorder (ADHD) in children (Kish, 2008). Currently, the absence of powerful longitudinal studies in this area is a major critical barrier to understanding and anticipating the full, long-term impact of meth/amphetamine consumption.

It has been more than 30 years since the discovery that methamphetamine and its metabolite amphetamine can harm brain dopamine neurons in experimental animals (Fibiger and McGeer, 1971, Seiden et al., 1976, Ricaurte et al., 1984, Ryan et al., 1990). Because of the animal findings, there is concern that use of meth/amphetamine might damage dopamine neurons in humans and thereby increase the risk of developing Parkinson's disease (PD), a dopamine deficiency brain disorder (Guilarte, 2001, Caligiuri and Buitenhuys, 2005, Thrash et al., 2009).

Biochemical brain studies of young methamphetamine users (who do not show the symptoms of PD) have disclosed changes in levels of some dopamine markers (Wilson et al., 1996a, McCann et al., 1998). However, the findings have yet to (and may never) confirm actual structural damage to or loss of dopamine neurons, because of the likelihood that such markers are not stable measures of dopamine neuron integrity (Boileau et al., 2008).

Significant and enduring dopamine toxicity caused by meth/amphetamine might only become clinically evident in susceptible users who have advanced to middle or older age—a time characterized by some age-related loss of dopamine neurons—and, as a result, longitudinal cohort designs offer a rigorous way to test this possibility. Given the high cost and common obstacles (e.g., participant loss to follow-up) associated with long-term longitudinal studies of illicit drug users, especially in regards to the estimation of low-incidence (but quite debilitating) conditions such as PD, a large-scale record-linkage approach may be one of the only feasible and effective designs available to assess the potential link between meth/amphetamine use and incidence of PD. In a previous epidemiological investigation of a small sample of older hospitalized meth/amphetamine users (≥50 years old) in California, we introduced a record-linkage approach which provided preliminary data suggesting that use of meth/amphetamine, sufficient to warrant a hospital diagnosis, might be associated with developing PD (Callaghan et al., 2010). Our present study adds significantly to this preliminary work by including: (1) a much larger and age-diversified group of meth/amphetamine users from California; (2) a sufficiently sized cocaine drug-control cohort; (3) a longer follow-up time (up to 16 years); and (4) the use of a more sophisticated statistical technique (i.e., competing risks analysis), along with the addition of death-record information, to account for potential differences in mortality across cohort groups. Here, we assess the risk of developing Parkinson's disease among meth/amphetamine users in comparison to that of population-proxy and stimulant-drug controls.