Evidence-based answers from the clinical inquiries Family Physicians Inquiries Network

Liza Jain, MD; Rebecca Benko, MD Tacoma Family Medicine Which drugs work best Residency, Tacoma, Wash for early Parkinson’s disease? Sarah Safranek, MLIS University of Washington Health Sciences Evidence-based answer Libraries, Seattle levodopa/carbidopa is the most ized controlled trials [RCTs]). Treating EDITOR A effective medical therapy for Par- early Parkinson’s disease with dopamine Janelle Guirguis- kinson’s disease, but it’s associated with agonists such as bromocriptine can im- Blake, MD Tacoma Family Medicine dyskinesia (strength of recommendation prove symptoms (SOR: B, Cochrane re- Residency, Tacoma, Wash [SOR]: A, Cochrane reviews and random- views, RCTs with heterogeneity).

Evidence summary 95% CI, 1.13-2.52), and nausea (OR=1.32; 95% Levodopa/carbidopa is the most commonly CI, 1.05-1.66). Patients treated with dopamine prescribed medication for Parkinson’s dis- agonists were also significantly more likely ease. Although its efficacy is established, to discontinue treatment because of adverse it can cause dyskinesia and dystonia.1 Re- events (OR=2.49; 95% CI, 2.08-2.98; P<.00001). cent studies (table) have evaluated the use of other medications early in the course of Bromocriptine studies hampered Parkinson’s disease in hopes of delaying the by poor quality waning effectiveness of levodopa over time. Two Cochrane reviews specifically evalu- ated the dopamine agonist bromocriptine.3,4 Dopamine agonists: The first focused on 6 head-to-head trials Dyskinesia reduction, but at a price with levodopa that enrolled 850 patients.3 A Cochrane review of 29 trials with 5247 pa- The studies were of poor quality, marred by tients compared dopamine agonists with methodological flaws and clinical heteroge- levodopa.2 Levodopa controlled symptoms neity. Problems included inadequate power, better than dopamine agonists, but inconsis- high variability in study duration (23 weeks tent data reporting prevented quantifying this to 5 years), differences in reporting, and lack result. of description of the randomization method Compared with the group taking levodo- in 3 of the 6 trials. Although bromocriptine pa, patients taking dopamine agonists dem- showed a trend toward lower incidence of onstrated a significant reduction in dyskinesia motor complications, many patients dropped (odds ratio [OR]=0.45; 95% CI, 0.37-0.54), dys- out of the studies because of increased non- tonia (OR=0.64; 95% CI, 0.51-0.81), and motor motor adverse effects and inadequate re- fluctuations (OR= 0.71; 95% CI, 0.58-0.87). sponse to treatment. However, patients taking dopamine The second review, of 7 trials with a to- agonists with or without levodopa expe- tal of 1100 patients, compared bromocrip- rienced significantly more adverse effects tine plus levodopa with levodopa alone.4 than patients taking levodopa alone. Side The studies were of poor quality for reasons effects included increased edema (OR=3.68; similar to the studies in the first review. Re- 95% CI, 2.62-5.18), somnolence (OR=1.49; searchers found no statistically significant or 95% CI, 1.12-2.00), constipation (OR=1.59; consistent evidence to determine whether 95% CI, 1.11-2.28), dizziness (OR=1.45; bromocriptine plus levodopa prevents or de- 95% CI, 1.09-1.92), hallucinations (OR=1.69; lays motor complications.

106 The Journal of Family Practice | February 2012 | Vol 61, No 2 TABLE Medications commonly used to treat Parkinson’s disease

Approximate monthly cost at usual dosage (in US $) for generic (brand name Medication class prices cited if no brand (generic)9 Advantages Disadvantages generic available)10 Carbidopa/levodopa First-line therapy; most Dyskinesia, dystonia, Sinemet (carbidopa/ effective at improving motor hallucinations $34.99-$101.98 levodopa) disability1

Sinemet CR (carbidopa/ No documented benefit $80.99-$295.97 levodopa controlled- of long-acting form1,8 (Highly variable due to release) dose range) COMT inhibitor Augments levodopa, may Same side effects as above Comtan (entacapone) improve activities of daily plus possible increased nausea, $310.97-$414.62 living6 vomiting, diarrhea6

Stalevo (carbidopa/ Possible increased $318.00-$1043.97 levodopa/entacapone) cardiovascular risk and prostate cancer

Dopamine agonist Reduced dyskinesias, dystonia, Nausea, dizziness, constipation, Mirapex (pramipexole) and motor complications2 somnolence, hallucinations, $239.99 edema2 Requip (ropinirole) $71.99-$143.98

Parlodel (bromocriptine) $385.97-$1133.92 MAO-B inhibitor Mild improved motor Limited efficacy and multiple Eldepryl (selegiline) symptoms of disease, adverse effects leading $101.99 decreased motor fluctuations to high dropout rate; not of treatment, possible recommended by Cochrane “levosparing effect”5 review5 Anticholinergic Improved symptoms, Confusion, memory loss, Cogentin (benztropine mostly tremor7 hallucinations, restlessness; $13.99-$22.99 mesylate) contraindicated in dementia7 Other No good updated studies, Symmetrel (amantadine) unproven long-term benefit, $43.17 nausea, dizziness, insomnia, can cause psychosis9 COMT, catechol-O-methyltransferase; MAO-B, monoamine oxidase type B.

MAO-B inhibitors: Minimally effective progression and treatment response. with troubling side effects Compared with the control groups (either A Cochrane review of monoamine oxidase placebo or levodopa at study onset), the MAO- type B (MAO-B) inhibitors included 10 tri- B group (either alone or with levodopa) showed als with 2422 participants.5 The review found significant improvement on the motor section statistically, but not clinically, significant (weighted mean difference [WMD]=−3.81 on a improvements in scores on 2 sections of 108-point scale; 95% CI, −5.36 to −2.27) and ac- the United Parkinson Disease Rating Scale tivities of daily living section (WMD=−1.50 on (UPDRS), a standardized assessment tool that a 52-point scale; 95% CI, −2.53 to −0.48). Fewer facilitates accurate documentation of disease motor complications occurred in the MAO-B

jfponline.com Vol 61, No 2 | February 2012 | The Journal of Family Practice 107 clinical inquiries

group (OR=0.75; 95% CI, 0.59-0.94) than the P=.025) but not mentation or motor symptoms. control group. Lower doses and shorter treat- Dyskinesia and wearing-off symptoms ment with levodopa were necessary to control (motor fluctuations) didn’t differ significant- symptoms in the MAO-B group. ly between the 2 groups. LCE was associated The clinical impact of MAO-B inhibitors with a higher incidence of adverse effects on Parkinson’s symptoms was small, and than LC, and involved mostly nausea (26.6% almost all patients required the addition of vs 13.5%) and diarrhea (8.7% vs 2.8%). levodopa to the treatment regimen after 3 or 4 years. Withdrawals because of medica- Anticholinergics may help, tion side effects were significantly higher in but cause adverse mental effects the MAO-B inhibitor group than controls Another Cochrane review compared anti- (OR=2.36; 95% CI, 1.32-4.20). Side effects in- cholinergic agents with placebo or no treat- cluded nausea, confusion, hallucinations, ment in 9 studies that included 221 patients.7 and postural hypotension. Concerns about Meta-analysis wasn’t possible because of het- cardiovascular adverse effects raised in pre- erogeneity in patient populations, outcomes, vious studies, especially with selegiline, and measurements and incomplete report- weren’t found to be significant (OR=1.15; ing. Compared with placebo, anticholiner- 95% CI, 0.92-1.44). Because of their minimal gic agents may improve Parkinson’s-related effectiveness and worrisome adverse effects, motor symptoms but have significant mental Dopamine MAO-B inhibitors aren’t recommended for adverse effects, including confusion, memory agonists routine use in early Parkinson’s disease. problems, restlessness, and hallucinations. can reduce dyskinesia, but COMT inhibitors may boost have significant levodopa/carbidopa’s effects Recommendations adverse effects. A randomized double-blinded trial followed The most recent guidelines (2002) from the 423 patients for 39 weeks to compare the American Academy of Neurology recom- combination of the catechol-O-methyltrans- mend levodopa and dopamine agonists as ferase (COMT) inhibitor entacapone and first-line therapies.8 Levodopa is more ef- levodopa/carbidopa (LCE) with levodopa/ fective at improving the motor symptoms of carbidopa alone (LC).6 The researchers Parkinson’s disease but is associated with found statistically significant improvements a higher risk of dyskinesia than dopamine with LCE in UPDRS scores for activities of agonists. No compelling evidence suggests daily living (mean change from baseline=3.0 a difference in efficacy between long- and for LCE vs 2.3 for LC on a 52-point scale; short-acting levodopa. JFP

References trial of levodopa/carbidopa/entacapone versus levodopa/ 1. Hauser RA. Levodopa: past, present, and future. Eur Neurol. carbidopa in early Parkinson’s disease. Mov Disord. 2009;62:1-8. 2009;24:541-550. 2. Stowe RL, Ives NJ, Clarke C, et al. Dopamine agonist therapy 7. Katzenschlager R, Sampaio C, Costa J, et al. Anticholiner- in early Parkinson’s disease. Cochrane Database Syst Rev. gics for symptomatic management of Parkinson’s disease. 2008;(2):CD006564. Cochrane Database Syst Rev. 2003;(2):CD003735. 3. van Hilten JJ, Ramaker CC, Stowe R, et al. Bromocriptine ver- 8. Miyasaki JM, Martin W, Suchowersky O, et al. Practice pa- sus levodopa in early Parkinson’s disease. Cochrane Database rameter: initiation of treatment for Parkinson’s disease: Syst Rev. 2007;(4):CD002258. an evidence-based review: report of the Quality Standards 4. van Hilten JJ, Ramaker CC, Stowe R, et al. Bromocriptine/le- Subcommittee of the American Academy of Neurology. vodopa combined versus levodopa alone for early Parkinson’s Neurology. 2002;58:11-17. disease. Cochrane Database Syst Rev. 2007;(4):CD003634. 9. Drugs for Parkinson’s disease. Treat Guidl Med Lett. 2011; 5. Macleod AD, Counsell CE, Ives N, et al. Monoamine oxidase 9:1-6. B inhibitors for early Parkinson’s disease. Cochrane Database 10. Drugstore.com Online Pharmacy. Pharmacy drug costs. Syst Rev. 2005;(3):CD004898. Available at http://www.drugstore.com. Accessed August 30, 6. Hauser RA, Panisset M, Abbruzzese G, et al. Double-blind 2011.

Visit us @ jfponline.com

108 The Journal of Family Practice | February 2012 | Vol 61, No 2