Methods
All patients referred to the joint medical/surgical movement disorder clinic at our institution during the 10-year period from January 2002 to 2012 who were eligible for and offered STN DBS were included in this study. Patients either accepted the offer of surgery or elected to continue with medical treatment.
We compared the survival of patients undergoing STN-DBS as part of their PD management with that of patients who were offered the procedure but declined, and therefore continued with maximal medical therapy. Survival data and that of admission to residential (nursing) home care were obtained from follow-up at the movement disorder clinic, from general medical practitioners and from enquiries to the Registrar of Deaths (National Statistics). In order to compare possible confounding factors in both groups, we analysed: age, gender, ethnicity, duration of disease, past medical history, including depression and Parkinson's medication taken at the time surgery was offered. Levodopa equivalent doses were used to compare the medication taken by patients. Patients considered for surgery have severe PD demonstrated by Hoehn & Yahr severity scores of 3–4 in their worst off states, Dopa-responsive disease and display motor complications of therapy including dyskinesia and on/off fluctuations.
Patients who agree to have surgery undergo psychometric testing which includes: a Dementia Rating Scale II, a Wechsler Memory Scale 3rd edition (WMS III), an IQ test, and a Hospital Anxiety and Depression Scale assessment. Those patients showing satisfactory psychometric testing proceed to surgery. Patients showing evidence of neuropsychological impairment are deemed unsuitable for surgery. Patients with significant apathy or 'off period' hallucinations were excluded from both groups. The presence or absence of depressive symptoms was corroborated by a neurologist in a neuropsychiatry movement disorder clinic aided by a consultant psychiatrist.
Those patients declining surgery are referred back to their treating physician for ongoing medical therapy. Nearly half these patients were subsequently looked after by the DBS team neurologist. The opportunity exists for patients to be rereferred if they change their minds.
The majority of patients considered for surgery are aged between 40 years and 70 years, though age is not considered an absolute criterion for surgical eligibility.
Statistical Analysis
A range of variables was compared between those who underwent STN-DBS, and those who were deemed suitable but declined. Kaplan–Meier curves were then produced for these two groups, with a Log-Rank test used for comparing survival.
In order to adjust for any differences between the two groups at baseline, multivariable Cox regression models were produced. To account for delay between being offered surgery, and the surgery being performed, a time-dependent covariate was used. This treated all patients as being medically managed from the offer of surgery until surgery was performed, at which point they moved to the surgical group.
A logistic regression model was then produced from the same factors, to compare the rates of admission to residential care between the two treatment groups. All analyses were performed using IBM SPSS V.19 (IBM SPSS, Chicago, Illinois, USA), with p<0.05 deemed to be indicative of statistical significance.