Published: November 1999

Selegiline in Parkinson's Disease

Prescriber Update 19: 14-18
November 1999

Dr Barry Snow, MBChB FRACP FRCPC, Neurologist, Auckland Hospital, Auckland, New Zealand

Selegiline (Eldepryl™, Selgene™, Apo-Selegiline™) was once thought to slow the progression of Parkinson’s disease and was, therefore, widely prescribed. New evidence indicates that selegiline does not have a neuroprotective effect, though some clinicians remain unconvinced. Selegiline does delay the need for levodopa therapy, however, the onset of disability or levodopa-induced motor fluctuations is not postponed by early selegiline treatment.
Recent studies also suggest that selegiline may increase mortality when co-prescribed with levodopa to patients with mild Parkinson's disease.
Selegiline is not without benefit as it smooths motor fluctuations in more advanced cases. Its best advantage is seen in patients with motor fluctuations that do not respond to alternative therapy. For those taking selegiline, switching therapy may not be advantageous unless side effects, such as insomnia, are bothersome.

Selegiline (Eldepryl™, Selgene™, Apo-Selegiline™) was introduced as adjunctive therapy for Parkinson’s disease. In the 1980s it was thought that selegiline was neuroprotective and slowed the progression of Parkinson’s disease, hence it was widely prescribed. It is now apparent that selegiline does not alter the long-term outlook of Parkinson’s disease. In addition, data have recently been published that suggest that selegiline may increase mortality in some patients. It is timely, therefore, to review the place of selegiline in the treatment of Parkinson’s disease.

Selegiline initially thought to retard the disease process

Selegiline has been used since 1975 as adjunctive therapy for Parkinson’s disease.1 It blocks the enzyme monoamine oxidase type B, which forms one of the metabolic pathways for dopamine in the brain. It was predicted that selegiline would prolong the effect of native or endogenous dopamine, and also prolong the effect of dopamine formed from therapeutically administered levodopa.

In the 1980s, evidence suggested that selegiline may retard the pathological process that leads to the development of Parkinson’s disease.2 Two possible mechanisms were postulated. The most popular was the blockade of oxidation and, therefore, free radical formation. A less popular putative mechanism was that selegiline may prevent the action of environmental toxins.

Latest evidence - delays the need for levodopa but no neuroprotective effect

The possibility that selegiline may be neuroprotective lead to the DATATOP study3 where patients with untreated Parkinson’s disease were randomized to selegiline or placebo. Patients on selegiline took a significantly longer time to need levodopa therapy than did patients on placebo. However, the study did not take adequate account of the mild symptomatic benefit of selegiline. This symptomatic benefit can be difficult to measure because selegiline is a "suicide" inhibitor of monoamine oxidase-B. This means that the MAO enzyme is irreversibly blocked and must be renewed by the formation of new enzyme. Therefore, the effect of selegiline lasts many days after withdrawal. A further review of patients in the DATATOP study concluded that the observed effect was most likely due to the symptomatic effect of selegiline and not any neuroprotective effect.4

While most observers now accept that selegiline does not have a neuroprotective effect, there is still a body of animal studies supporting this possibility, and some clinicians have continued to argue that selegiline may retard the progression of Parkinson’s disease.5,6

No long-term benefit from early selegiline treatment

While selegiline may not have a neuroprotective effect, it undoubtedly delays the need for levodopa therapy. The possibility that levodopa may be toxic to dopaminergic neurones has been intensively debated.7 In addition, prolonged levodopa therapy is often associated with disabling motor fluctuations. If selegiline delays the need for levodopa, then perhaps it will also delay the onset of motor fluctuations. This possibility was also addressed in follow-up studies of the DATATOP cohort.8 Those follow-up studies demonstrated that the "time gained" by selegiline therapy did not postpone the onset of disability or levodopa-induced motor fluctuations. Early treatment with selegiline, therefore, has not been shown to produce a long-term benefit.

Suggestion that selegiline increases mortality when co-prescribed with levodopa

Until 1995, most investigators focused on the possibility the selegiline might prolong the survival of patients with Parkinson’s disease; thus it came as a surprise when the Parkinson’s Disease Research Group (PDRG) of the United Kingdom reported increased mortality in patients on combined selegiline and levodopa therapy.9 That study compared patients on levodopa alone with patients on levodopa and selegiline. They found a 60% increase in relative mortality, which translated to 1 excess death for every 54 patients treated for 1 year. They also found that selegiline had no beneficial effect on parkinsonian disability.

The PDRG report of excess mortality associated with selegiline caused much controversy.10,11 Criticisms of the paper included:

The PDRG authors addressed these issues in discussion papers.12 They also returned to the study patients with a follow-up study.13 The passage of time since the original report9 permitted the gathering of more death certificate data. This led to a revision of the relative mortality to 35% or 1 excess death per 75 patients treated for 1 year. The increased risk was of borderline statistical significance. They also examined the cause of death in more detail, but this did not change the original conclusion that the excess mortality was due to Parkinson’s disease and not another identified cause. Patients who died on selegiline were more likely to have had possible dementia and a history of falls. The paper was not able to address the main outstanding criticism of the PDRG paper, namely, the inconsistency with the mortality data from other studies.

Selegiline may smooth muscle fluctuations

While the above synthesis may suggest that selegiline has no place in the therapy of Parkinson’s disease, the medication is not without use. The inhibition of levodopa metabolism by selegiline can effectively smooth out predictable motor fluctuations in some patients.1,14 Thus for patients with motor fluctuations and without dementia or falls, selegiline may be both beneficial and safe.

There is no longer any compelling reason to prescribe selegiline as monotherapy for new patients with Parkinson’s disease. Selegiline may be of use for patients with motor fluctuations although it is not necessarily better than alternative strategies such as slow-release levodopa, dopamine agonists or COMT inhibition.

For patients who are already taking selegiline, I personally do not routinely stop the medication. Some patients experience a feeling of wellbeing on selegiline, and they may experience withdrawal symptoms on stopping the drug. Many patients taking selegiline are troubled by insomnia; in these patients, I stop selegiline therapy.


Correspondence to Dr Barry Snow, Neurophysiology, 3rd Floor, Auckland Hospital, Private Bag 92094, Auckland, New Zealand. Telephone 025 314409, fax 09 307 8912, e-mail

  1. Birkmayer W, Riederer P, Youdim MBH, Linauer W. The potentiation of the anti akinetic effect after L-dopa treatment by an inhibitor of MAO-B, deprenyl. J Neural Transmission 1975;36:303-26.
  2. The Parkinson Study Group. DATATOP: a multicenter controlled clinical trial in early Parkinson's disease. Arch Neurol 1989;46:1052-1060.
  3. The Parkinson Study Group. Effect of deprenyl on the progression of disability in early Parkinson's disease. N Engl J Med 1989;321:1364-71.
  4. The Parkinson Study Group. Effects of tocopherol and deprenyl on the progression of disability in early Parkinson's disease. N Engl J Med 1993;328:176-83.
  5. Tatton WG, Greenwood CE. Rescue of dying neurons: a new action for deprenyl in MPTP parkinsonism. J Neurosci Res 1991;30:666-72.
  6. Olanow CW, Hauser RA, Gauger L, et al. The effect of deprenyl and levodopa on the progression of Parkinson's disease. Ann Neurol 1995;38:771-7.
  7. Agid Y, Chase T, Marsden D. Adverse reactions to levodopa: drug toxicity or progression of disease? Lancet 1998;351:851-2.
  8. Parkinson Study Group. I. Impact of deprenyl and tocopherol treatment on Parkinson’s disease in DATATOP subjects not requiring levodopa. II. Impact of deprenyl and tocopherol treatment on Parkinson’s disease in DATATOP subjects requiring levodopa. Ann Neurol 1996;39:29-36, 37-45.
  9. Lees AJ, on behalf of the Parkinson’s Disease Research Group of the United Kingdom. Comparison of therapeutic effects and mortality data of levodopa and levodopa combined with selegiline in patients with early, mild Parkinson’s disease. BMJ 1995;311:1602-7.
  10. See correspondence BMJ 1996;312:702-5.
  11. Olanow CW, Fahn S, Langston JW, Godbold J. Selegiline and mortality in Parkinson’s disease. Ann Neurol 1996;40:841-5.
  12. Lees AJ, Head J, Ben-Sholomo Y. Selegiline and mortality in Parkinson’s disease: another view. Ann Neurol 1997;41:282-3.
  13. Ben-Shlomo Y, Churchyardn A, Head J, Lees AJ, et al. Investigation by Parkinson's Disease Research Group of United Kingdom into excess mortality seen with combined levodopa and selegiline treatment in patients with early, mild Parkinson's disease: further results of randomised trial and confidential inquiry. BMJ 1998;316:1191-6.
  14. Golbe LI. Deprenyl as symptomatic therapy in Parkinson's disease. Clin.Neuropharmacol 1988;11:387-400.