by Elena Moro
For August 2018, we have selected: Corvol JC, Artaud F, Cornier-Dequaire F, et al., for the DIGPD Study Group. Longitudinal analysis of impulse control disorders in Parkinson disease. Neurology 2018. doi:10.1212/WNL.0000000000005816.
Impulse control disorders (ICDs) are frequent in Parkinson’s disease (PD) patients, and can have relevant consequences on social life. Whereas the association between treatment with dopamine agonists (DA) and the onset of ICDs is well known, there is controversy about the relationship with dose and duration of DA therapy.
In this multicenter, longitudinal cohort study, the authors analyzed PD patients consecutively included from May 2009 to July 2013, as part of the Drug Interaction With Genes in Parkinson’s Disease (DIGPD) Study. PD patients with ≤5 years’ disease duration at baseline were included in 8 French clinical centers and followed yearly over 5 years. The presence of ICDs was evaluated at each visit during a face-to-face semistructured interview. Compulsive gambling, shopping, eating, sexual behavior, hobbyism, and hypercreativity were defined as ICDs. Moreover, patients were evaluated with the full Movement Disorders Society Unified Parkinson Disease Rating Scale and the Mini-Mental State Examination. The association between DA and the ICDs prevalence at each visit was analyzed using generalized estimating equation (GEE) models. The Poisson regression with robust variance was used to estimate prevalence ratios (PRs) and 95% confidence intervals (Cis). Average daily DA dose and cumulative duration were analyzed as continuous variables. Moreover, differences between specific DAs and standard and extended-release (ER) presentations were also analyzed.
Data from 411 patients (244 men, 59.4%) were included in the analysis. At baseline, age at PD onset and PD duration were 59.7± 9.9 and 2.6 ± 1.5 years, respectively. At baseline, 93.7% of patients took levodopa (20.2%) or DA (29.4%) or both together (44.0%). There were 81 patients (19.7%) with ICDs, mainly compulsive binge/eating and compulsive sexual behavior. These patients were younger and had longer disease duration compared to the patients without ICDs, more likely obese, single, regular coffee drinkers, and using DAs more frequently and at higher doses. The average follow-up was 3.3 ± 1.7 years.
During the follow-up, 175 patients (42.6%) had ICDs at ≥1 visit, with increased prevalence from 19.7% (baseline) to 32.8% at 5 years. After the first visit with ICDs, 90 patients (51.4%) remained with ICDs until the end of the follow-up. Of the remaining 85, 58 patients endured without ICDs until the end whereas 27 relapsed.
Of the 306 patients without ICDs (260 ever DA users) at baseline, 94 developed ICDs (90 ever DA users), with to 46.1% cumulative incidence in 5 years (95% CI 37.4-55.7; never users 12.4%, 95% CI 4.8-30.0; ever users 51.5%, 95% CI 41.8-62.1). Men and younger patients developed ICDs more frequently over time. DA use in the past 12 months was associated with a 2.23-fold higher ICD prevalence (p=0.001). The prevalence increased with average daily dose. Cumulative DA duration and average daily dose showed linear association with the two variables, stronger for duration than dose, both overall and in ever users. LD was not strongly associated with ICDs. Fifteen out of 30 patients resolved ICDs progressively within one year after stopping DAs. Pramipexole and ropinirole did not differ in inducing ICDs, but both were related to higher prevalence of ICDs compared to other DAs. No difference was found between ER and standard presentation of DA in relation with ICDs.
“This study supports the strong association between DA treatment duration and dose with ICDs incidence in PD patients. The annual ICD incidence was of 103.3 per 1,000 (DA users 118.6 of 1,000), like other previous studies. However, the increase in ICD prevalence over time was more pronounced in incident patients that those with longer PD duration says Dr. Anna Castrioto. Division of Neurology, CHU Grenoble, Grenoble, France. “This finding emphasizes that ICDs tend to occur within the first years of the disease, with a trend to a stable frequency afterwards.”
“In this study, unlike others, levodopa was not associated with higher risk of developing ICDs. This finding might be related to the prospective nature of this longitudinal observational cohort study,” says Prof. Susan Fox, Division of Neurology, UHN, University of Toronto, Canada. “On the other hand, the relatively young age at baseline of this study population might be an important bias. Interestingly, ICDs disappeared in only 50% of the patients after stopping DA. This is intriguing and needs further studies.”
The other nominees for the August 2018 paper of the month are:
- Desjardins A, Gromeier M, Herndon J, et al. Recurrent glioblastoma treated with recombinant poliovirus. NEJM 2018; doi: 10.1056/NEJMoa16435. In this study, 61 patients with recurrent supratentorial glioblastoma received escalating doses of the recombinant nonpathogenic polio-rhinovirus chimera (PVSRIPO). The PVSRIPO binds the receptor CD155, widely present in the neoplastic cells. Infusion was delivered with a catheter inside the tumor through a pump. About 20% of the patients developed adverse events during the dose-expansion phase. At 24 and 36 months, 21% of the patients were still alive.
- Elkj LS, Bech BH, Sun Y, Laursen MT, Christensen J. Association between prenatal valproate exposure and performance on standardized language and mathematics test in school-aged children. JAMA Neurol 2018;75:663-671. In this prospective, population-based cohort study, all children born in Denmark between 1997 and 2006 were identified and assessed regarding to performance in national tests taken in primary and lower secondary state schools. Performances in Danish and mathematics was evaluated among valproate-exposed children, unexposed children and children exposed to other antiepilepsy drugs. Maternal use of valproate (and clonazepam) was associated with significant decrease in school performance.
- Mazzucco S, Li L, Binney L, Rothwell PM, on behalf of the Oxford Vascular Study Phenotyped Cohort. Prevalence of patent foramen ovale in cryptogenic transient attack and non-disabling stroke at older ages: a population-based study, systematic review, and meta-analysis. Lancet Neurol 2018;17:609-617. To assess the relationship between patent foramen ovale (PFO) and cryptogenic TIA or stroke in patients older than 60 years, the authors evaluated 572 patients using contrast-enhanced transcranial doppler (bubble-TCO) to detect probable PFO. The bubble-TCD was feasible in 90% of patients. PFO was significantly associated with cryptogenic vascular events in older patients.