This is an ordinal scale with scores from 0 (no pain) to 10 (worst possible pain). Data presented represent the weekly mean of the scores of the average pain severity over the last 24 hours. Scores are based on daily assessments recorded by patients in their diaries. Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

This is an ordinal scale with scores from 0 (no pain) to 10 (worst possible pain). Data presented represent the weekly mean of the scores of the average pain severity over the last 24 hours. Scores are based on daily assessments recorded by patients in their diaries. Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

This is an ordinal scale with scores from 0 (no pain) to 10 (worst possible pain). Data presented represent the weekly mean of the scores of the daily nighttime pain severity over the last 24 hours. Scores are based on daily assessments recorded by patients in their diaries. Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

This is an ordinal scale with scores from 0 (no pain) to 10 (worst possible pain). Data presented represent the weekly mean of the scores of the daily worst pain severity over the last 24 hours. Scores are based on daily assessments recorded by patients in their diaries. Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

Measures severity of illness at the time of assessment compared with start of treatment. Scores range from 1 (normal, not at all ill) to 7 (among the most extremely ill patients). Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A scale that measures the patient's perception of improvement at the time of assessment compared with the start of treatment. The score ranges from 1 (very much better) to 7 (very much worse).

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A self-reported scale that measures the severity of pain based on the average pain over the past 24-hours. The severity scores range from 0 (no pain) to 10 (pain as severe as you can imagine). Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A self-reported scale that measures the severity of pain based on the worst pain experienced over the past 24-hours. The severity scores range from 0 (no pain) to 10 (pain as severe as you can imagine). Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A self-reported scale that measures the severity of pain based on the least pain experienced over the past 24-hours. The severity scores range from 0 (no pain) to 10 (pain as severe as you can imagine). Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A self-reported scale that measures the severity of pain based on the pain right now. The severity scores range from 0 (no pain) to 10 (pain as severe as you can imagine). Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A self-reported scale that measures the interference of pain in the past 24 hours on general activity. The Interference scores range from 0 (does not interfere) to 10 (completely interferes). Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A self-reported scale that measures the interference of pain in the past 24 hours on mood. The Interference scores range from 0 (does not interfere) to 10 (completely interferes). Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A self-reported scale that measures the interference of pain in the past 24 hours on walking ability. The Interference scores range from 0 (does not interfere) to 10 (completely interferes). Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A self-reported scale that measures the interference of pain in the past 24 hours on normal work. The Interference scores range from 0 (does not interfere) to 10 (completely interferes). Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A self-reported scale that measures the interference of pain in the past 24 hours on relations with other people. The Interference scores range from 0 (does not interfere) to 10 (completely interferes). Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A self-reported scale that measures the interference of pain in the past 24 hours on sleep. The Interference scores range from 0 (does not interfere) to 10 (completely interferes). Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A self-reported scale that measures the interference of pain in the past 24 hours on enjoyment of life. The Interference scores range from 0 (does not interfere) to 10 (completely interferes). Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

The Interference scores range from 0 (does not interfere) to 10 (completely interferes). There are 7 questions assessing the interference of pain in the past 24 hours for general activity, mood, walking ability, normal work, relations with other people, sleep, and enjoyment of life. Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

This is a nominal outcome reflecting whether or not a clinically-important efficacy outcome was achieved at endpoint. It is based on a comparison between baseline and endpoint scores on an ordinal scale with scores from 0 (no pain) to 10 (worst possible pain). Used were the weekly mean of the scores of the average pain severity over the last 24 hours. The weekly averages were based on daily assessments recorded by patients in their diaries.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

This is a nominal outcome reflecting whether or not a clinically-important efficacy outcome was achieved at endpoint. It is based on a comparison between baseline and endpoint scores on an ordinal scale with scores from 0 (no pain) to 10 (worst possible pain). Used were the weekly mean of the scores of the average pain severity over the last 24 hours. The weekly averages were based on daily assessments recorded by patients in their diaries.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

This is a nominal outcome reflecting whether or not a clinically-important efficacy outcome was achieved at endpoint. It is based on a comparison between baseline and endpoint scores on an ordinal scale with scores from 0 (no pain) to 10 (worst possible pain). Used were the weekly mean of the scores of the average pain severity over the last 24 hours. The weekly averages were based on daily assessments recorded by patients in their diaries.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

The LSEQ assesses the effects of psychoactive compounds on sleep and early morning behavior. Participants mark a series of 100 mm line analogue scales, indicating the direction and magnitude of any changes in behavioral state they experience following administration of the drug. Scores are represented in millimeters, higher scores indicate better sleep and better early morning behavior. Subscale score ranges: GTS=0-300, QOS=0-200, AFS=0-200, BFW=0-300. Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

The SDS is completed by the patient and is used to assess the effect of the patient's symptoms on their work/social/family life. Total scores range from 0 to 30, higher values indicate greater disruption in the patient's life. Item 1 assesses the effect of the patient's symptoms on their work/school schedule, Item 2 on their social life/leisure activities, and Item 3 on their family life/home responsibilities. Subscales scores range: 0-10, higher values indicate greater disruption in the patient's life. Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

All participants who were enrolled in the study.

14-item subject-rated scale assessing medication related changes in sexual activity + functioning. Structured interview/questionnaire. It measures five dimensions of sexual behavior: pleasure; desire/frequency; desire/interest; arousal; and orgasm. The total score is obtained across all 5 dimensions, ranging from 14 to 70. Subscale score ranges: desire/frequency=2-10; desire/interest=3-15; pleasure=1-5; arousal=3-15; orgasm=3-15. Higher scores = better sexual functioning. Least-squares means: adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

14-item subject-rated scale assessing changes in sexual activity and functioning; structured interview/questionnaire, designed to measure medication related changes in sexual functioning. 5 dimensions of sexual behavior: pleasure; desire/frequency; desire/interest; arousal; orgasm. Total score: obtained across all 5 dimensions, ranges from 14 to 70. Subscale scores: desire/frequency=2-10; desire/interest=3-15; pleasure=1-5; arousal=3-15; orgasm=3-15. Higher scores = better sexual functioning. Categories: better=positive change in score; same=no change in score; worse=negative change in score.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) and with baseline values within the normal range were included in the analysis.

The Portland Neurotoxicity Scale is a 15-item, patient-completed questionnaire designed to assess the degree of impact of anti-epileptic drug therapy on a number of cognitive and somatomotor parameters. The total score ranges from 15-135 with higher scores indicating more toxicity. The cognitive toxicity score ranges from 10-90 and the somatomotor toxicity score ranges from 5-45, for both higher scores indicate more toxicity. Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

The Portland Neurotoxicity Scale is a 15-item, patient-completed questionnaire designed to assess the degree of impact of anti-epileptic drug therapy on a number of cognitive and somatomotor parameters. Categories: better=negative change in score; same=no change in score; worse=positive change in score.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

Contribution to reduction in pain directly by treatment and indirectly by treatment through the reduction of depressive symptoms using path analysis. The direct treatment effect estimates the mean drug difference in pain reduction directly through treatment; the indirect treatment effect estimates the contribution that treatment plays to the mean drug difference in pain reduction indirectly through the reduction in mood symptoms; the total effect estimates the drug difference in reducing pain in sum through the specified path of direct and indirect treatment effects.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

A 21-item, patient-completed questionnaire to assess characteristics of depression. Each of the 21 items corresponding to a symptom of depression is summed to give a single score. There is a four-point scale for each item ranging from 0 to 3. Total score of 0-13 is considered minimal range, 14-19 is mild, 20-28 is moderate, and 29-63 is severe. Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

The Resource Utilization Scale measures direct and indirect costs (collected only for US sites). Direct costs include inpatient and outpatient costs, while indirect costs include lost days of work and caregiver time spent with patients. Inpatient costs include costs associated with hospitalizations and time spent in emergency rooms and psychiatric rooms. Outpatient costs include costs associated with visits to various health care providers, home health care by health care providers, and partial care.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

Number of participants who discontinued. The reasons for discontinuation are presented in the participant flow.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

This is the number of days required to first achieve a nominal outcome reflecting whether or not a clinically-important efficacy outcome was achieved. It is based on a comparison between baseline and post-baseline scores on an ordinal scale with scores from 0 (no pain) to 10 (worst possible pain). Used were the weekly mean of the scores of the average pain severity over the last 24 hours. The weekly averages were based on daily assessments recorded by patients in their diaries.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

This is the number of days to first achieve ≥50% reduction, baseline to endpoint, in the weekly means of the 24-hour average pain severity via daily patient assessments using an ordinal scale (scores from 0 (no pain) to 10 (worst possible pain). The median time to first ≥50% reduction with some measure of dispersion could not be calculated for each treatment group. The number of patients who reached a ≥50% reduction in weekly mean 24 hour average pain score are presented in outcome measure 20.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

This is the number of days to first achieve an outcome using the weekly means of the 24-hour average pain severity via daily patient assessments using an ordinal scale (scores from 0 (no pain) to 10 (worst possible pain). Sustained response: ≥30% reduction, baseline to endpoint, with 30% reduction from baseline ≥2 weeks prior to endpoint, remaining at ≥20% reduction between. The median time to sustained response with some measure of dispersion could not be calculated for each treatment group.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

This is the number of days required to first achieve a nominal outcome reflecting whether or not a clinically-important efficacy outcome was achieved. It is based on a comparison between baseline and post-baseline scores on an ordinal scale with scores from 0 (no pain) to 10 (worst possible pain). Used were the weekly mean of the scores of the average pain severity over the last 24 hours. The weekly averages were based on daily assessments recorded by patients in their diaries.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

This is an ordinal scale with scores from 0 (no pain) to 10 (worst possible pain). Data presented represent the weekly mean of the scores of the average pain severity over the last 24 hours. Scores are based on daily assessments recorded by patients in their diaries.

It was planned to analyze participants stratified by the presence or absence of a co-morbidity with GAD. However, due to the low number of participants with GAD in the study this analysis was not possible.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

Ordinal scale: 0=no pain, 10=worst possible pain. Data=weekly mean of scores of average pain severity over last 24 hours (h). Scores: daily assessments recorded by patients in diaries. Only patients adhering to key protocol criteria included: baseline Weekly Mean 24h Average Pain Score ≥4; 80-120% compliant with study Drug, each visit; baseline Michigan Neuropathy Screening Instrument Physical Assessment Total Score ≥3; gabapentin taper ≤14 days, no HbA1c ≥12% post randomization; no contraindicated medications used. Least-squares means=adjustment due to baseline severity + investigative site.

The per-protocol sub-population of all randomized participants with a baseline value and >= 1 non-missing post-baseline value (modified intent-to-treat population) was included in the analysis.

This is an ordinal scale with scores from 0 (no pain) to 10 (worst possible pain). Data presented represent the weekly mean of the scores of the average pain severity over the last 24 hours. Scores are based on daily assessments recorded by patients in their diaries. De novo: use of gabapentin for <56 contiguous days prior to randomization. Prior use: use of gabapentin for >=56 contiguous days prior to randomization. Least-squares means represent adjustment due to baseline severity and investigative site.

Analyzed were all participants with a baseline and at least 1 non-missing post-baseline value. Last observation carried forward analysis.

Presented are numbers of participants who discontinued due to a change from baseline in laboratory analytes or vital signs.

All randomized patients.

Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

Least-squares means represent adjustment due to baseline severity and investigative site.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

Elevated systolic blood pressure: >=130 millimeter mercury (mm Hg) + an increase of >=10 mm Hg if baseline <130 mm Hg.

Elevated diastolic blood pressure: >=85 mm Hg + an increase of >=10 mm Hg if baseline <85 mm Hg.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) and with baseline values within the normal range were included in the analysis.

Elevated heart rate: >=100 beats per minute (bpm) + an increase of >=10 bpm if baseline <100 bpm.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) and with baseline values within the normal range were included in the analysis.

Treatment-emergent high body weight: weight at last visit >=107% of baseline weight.

Treatment-emergent low body weight: weight at last visit <=93% of baseline weight.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

Aspartate aminotransferase = AST Alanine aminotransferase = ALT Gamma glutamyl transferase = GGT Alkaline phosphatase = AlkPhos

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) were included in the analysis.

Treatment-emergent: within range at baseline, out of range after baseline. Ranges in Units/Liter (U/L). Aspartate Aminotransferase (AST): female (f): >34, male (m): >36. Alanine Aminotransferase (ALT): f:<69 years (yr) >34, ≥69yr >32; m: <69yr >43, ≥69yr >35. Total Bilirubin (TBili): >21. Gamma Glutamyl Transferase (GGT): f: <59yr >49, ≥59yr >50; m: <59yr >61, ≥59yr >50. Fasting Plasma Glucose (FPG): <59yr >6.4, ≥59yr >6.7. Hemoglobin A1C (HbA1C) >6%. Alkaline Phosphatase (AlkPhos): f: 18-50yr >106, 50-70yr >123, 70-80yr >164, ≥80yr >221; m: 18-50yr >129, 50-70yr >131, 70-80yr >156, ≥80yr >187

All randomized participants with a baseline value and at least 1 non-missing post-baseline value (modified intent-to-treat population) and with baseline values within the normal range were included in the analysis.