March 28 (HealthDay News) -- An impaired sense of smell could be an early indicator of Parkinson's disease, occurring up to four years before motor skill problems appear, recent research shows.

The study, published in the February issue of the Annals of Neurology, followed 2,267 men who received olfactory testing at the Kuakini Medical Center in Honolulu at least once during two periods in the 1990s. They were followed for up to eight years and, during that time, 35 of the men developed the disease.

An odor identification problem preceded the development of Parkinson's by at least four years in these men. Decreased odor identification was associated with older age, smoking, more coffee consumption, less frequent bowel movements, lower cognitive function and excessive daytime sleepiness, but even after adjusting for these factors, those with the lowest odor identification scores had a five times greater risk of developing Parkinson's than those with the highest scores.

The results strengthen findings from earlier studies that suggest olfactory impairment begins between two and seven years before diagnosis.

Why problems with smelling accompany Parkinson's is not completely understood; however, nerve loss and the formation of Lewy bodies -- abnormal clumps of proteins inside nerve cells that are thought to be a marker of the disease -- are known to take place in the olfactory structures of patients with the disease.

The study authors noted that one previous study involving brain dissection of deceased patients with neurological disease found that olfactory structures are the earliest brain regions affected by Lewy degeneration, which supports the idea that an impaired sense of smell could be one of the earliest signs of Parkinson's disease.

An impaired sense of smell could also be caused by impaired sniffing, which may be another motor symptom of Parkinson's.

Olfactory testing, along with screening for other potential early indicators of Parkinson's disease such as constipation or sleep disturbances, could provide a simple and relatively economic means of identifying individuals at high risk who could participate in trials of medications designed to prevent or slow disease progression, the authors concluded in a prepared statement.

SOURCE: American Neurological Association, news release, March 20, 2008