The authors describe a dopamine agonist withdrawal syndrome (DAWS)...

The authors describe a dopamine agonist withdrawal syndrome (DAWS) in the context of a retrospective cohort study.  The withdrawal syndrome occurred in 5 subjects among 26 who underwent taper of the medication.  They suggest that this withdrawal symptom is specific for agonists (i.e., not relieved by other therapies including levodopa) and that there appears to be a specific vulnerability in patients who develop agonist-induced impulse control disorders.  They defined DAWS as ^"a severe, stereotyped cluster of physical and psychological symptoms that correlate with DA withdrawal in a dose-dependent manner, cause clinically significant distress or social/occupational dysfunction, are refractory to levodopa and other Parkinson disease medications, and cannot be accounted for by other clinical factors".  

I do not recall seeing this before and have not thought of the dopamine agonists as "addicting" in the classic sense of the term.  In fact, there are numerous accounts of successful discontinuation of dopamine agonists among patients who experienced typical side effects (e.g., sedation, lower extremity edema) but also among patients who developed impulse control disorders apparently induced by these agents.  Having said that, it seems quite plausible that a subset of patients who were vulnerable to agonist-induced impulse control disorders might not tolerated their discontinuation due to "withdrawal" symptoms such as those described in this paper given the likely involvement of the dopaminergic reward pathway in these patients. The fact that it is not alleviated by levodopa is interesting and suggests that it is specific for agonists.

It has been proposed that antiparkinsonian medications taken by patients with PD leads to over stimulation of dopamine receptors within the mesocorticolimbic pathways results in development of addictive behaviors, such as compulsive medication intake and impulse control disorders.  There is the phenomenon of dopa dysregulation syndrome in which patients use increasing dosages of levodopa (despite good motor function and sometimes severe dyskinesias) for its effect on their emotional state.  They can become hypomanic and engage in risky and atypical behaviors.  They often feel very restless, dysphoric and uncomfortable with any attempts to reduce their levodopa.  This appears consistent with a classical addiction syndrome.  It has been referred to in the past as hedonistic homeostatic dysregulation and tended to be reported more often in males of slightly younger age.  It generally occurs in individuals with advancing disease and motor fluctuations.  

On the other hand, there are the relatively more recently described impulse control disorders which are characterized by compulsive behaviors (other than compulsive medication intake) including gambling, shopping, eating, sexual behaviors, etc.  These occur in both early and later disease. These were initially thought to be associated more specifically with the dopamine agonists but it appears as though it can occur with levodopa as well.  These generally resolve with lowering the agonist dosage or discontinuing the medication. The authors of this paper are describing a withdrawal type syndrome specific to dopamine agonists occurring in a subset of patients, all of whom apparently demonstrated impulse control disorders while taking the agonist.  They note that symptoms appeared consistent with those seen in withdrawal from other drugs such as cocaine.  The patients did not improve with increasing or in other ways adjusting their levodopa dosage, suggesting that it is not just a dopaminergic "wearing off" phenomenon.  It was not clear whether or not the patients who experienced this syndrome needed/wanted ever increasing dosages of their dopamine agonist (as one might see in some addictions) or whether they just experienced withdrawal symptoms upon lower the dosage of discontinuing the medication.   

There are certainly instances of withdrawal syndromes that are associated with discontinuation of other medications in which the patients are not "addicted".  The most obvious example of this is the withdrawal syndrome from SSRI antidepressants.  This typically occurs 24-72 hours after the patient stops taking the SSRI.  It can cause gastrointestinal (e.g., diarrhea), neurological (e.g., dizziness), somatic (e.g., chills) and psychological (e.g., anxiety, irritability, agitation) symptoms.  It is generally self-limited but can be quite unpleasant for patients and can be mistaken for a medical or psychiatric illness.  As in the case of the dopamine agonist withdrawal described by the authors of the current paper, the withdrawal symptom occurrence (and severity) may be associated with the duration of therapy and the overall dosage used.   

In summary, this paper suggests that a subset of patients taking dopamine agonists may develop withdrawal symptoms when the dosage is lowered or the medication is discontinued.  It seems as though patients who experience impulse control disorders are more vulnerable to these symptoms.  There is no doubt that antiparkinsonian medications may affect the mesocorticolimbic reward pathways resulting in abnormal behaviors and that certain patients appear to be more vulnerable.  Whether or not we use the term "addiction" in patients who experience this withdrawal syndrome may be more a semantic concern.  It does, however,  highlight an important phenomenon and suggest that attempts to identify patients at risk and considering alternative therapies may be worthwhile.