Abbott Proclaim 3660 Clinician Manual page 39

The secondary endpoints that were assessed included
Determining superiority of burst stimulation compared to tonic stimulation using VAS pain

diary scores for average daily overall pain (after non-inferiority was demonstrated)
Comparing the responder rate, which is defined by a 30% decrease in the VAS pain diary

scores for average daily overall pain
Comparing the percentage of paresthesia coverage

The following descriptive endpoints and additional data were assessed:
Demographics, including gender, age, height, weight, ethnicity, and marital status

Pain history, including primary diagnosis, pain duration, pain etiology, and prior treatments

Adverse events

Summary of adverse events related to tonic and burst stimulation

Surgery and device information

Programming and stimulation mode data

Recharging data

Comparison of the responder rate, which is defined by a 50% decrease in the VAS pain diary

scores for average daily overall pain
Average daily trunk pain and average daily limb pain as assessed by the VAS

Comparison of VAS pain diary scores for worst daily overall pain

Patient satisfaction with the device

Stimulation mode preference (tonic or burst)

Patient Global Impression of Change (PGIC)

Comparison of the quality of life in physical and mental components of Short Form 36

(SF-36)
Comparison of function in the Oswestry Disability Index (ODI), version 2.1a

Pain quality as assessed by the Short-Form McGill Pain Questionnaire (SF-MPQ-2)

Pain catastrophizing as assessed by the Pain Catastrophizing Scale (PCS)

Depression as assessed by the BDI‡-II clinical assessment

Medication usage

Primary Statistical Analysis Plan
To analyze the primary endpoint—evaluating non-inferiority of burst stimulation to tonic
stimulation in overall VAS scores—the non-inferiority margin was set to 7.5 mm for the overall VAS
score on a scale of 0 to 100. The standard deviation of the difference between the VAS scores for
tonic and burst stimulation was assumed to be 18.4 points. Setting the Type I error rate to 0.05, a
minimum sample size of 76 subjects was required to achieve 80% power to demonstrate non-
inferiority of burst stimulation to tonic stimulation.
The secondary endpoints were tested if the primary endpoint was met. Each secondary endpoint
was tested at a 5% significance level. Statistical tests were not performed for descriptive endpoints
and additional data.
35