Based on the BMD values and the weight, the age-normalized percentiles (Z-scores) were determined to allow for comparison with historical control groups. "No difference" in comparison with the historical control groups was defined as a Z-score between '-0.5' and '0.5', a lower value was defined as a value below '-0.5', and a higher value above '0.5'.

Responders were defined as subjects with reduction in pain intensity from baseline of at least 30% in the Visual Analog Scale (VAS) at Week 24. VAS consisted of a 100 unit long straight line, with verbal anchors at either end, representing a continuum of pain intensity. One end of the line with 0 score as "absence of pain" while the other end of the line with 100 score as "unbearable pain". The assessment of pelvic pain on a VAS was done once every 4 weeks till the end of the treatment (Week 52).

The cardinal symptoms inlcuded in the modified Biberoglu and Behrman severity profile were pelvic pain, dysmenorrhea, and dyspareunia (the latter only in those subjects having sexual intercourse), analyzed at all visits with symptom severity scores from 0 (none) to 3 (severe). Negative value for change from baseline indicates an improvement. In the listed categories, 'N' signifies those subjects who were evaluable for this measure.

The cardinal symptoms inlcuded in the modified Biberoglu and Behrman severity profile were pelvic pain, dysmenorrhea, and dyspareunia (the latter only in those subjects having sexual intercourse), analyzed at all visits with symptom severity scores from 0 (none) to 3 (severe). Negative value for change from baseline indicates an improvement. In the listed categories, 'N' signifies those subjects who were evaluable for this measure.

In order to judge therapeutic effectiveness and to compare subjects' complaints, a severity profile score of pelvic pain was assessed using a rating scale: missing; 0 = none; 1 = mild (occasional pelvic discomfort); 2 = moderate (noticeable discomfort for most of the cycle); 3 = severe (requires strong analgesics and persists during cycle when not menstruating) based on the subject's self-assessment of symptoms. In the listed categories, 'N' signifies those subjects who were evaluable for this measure.

In order to judge therapeutic effectiveness and to compare subjects' complaints, a severity profile score of dysmenorrhea was assessed using a rating scale: missing; 0 = none; 1 = mild (some loss in work efficiency); 2 = moderate (in bed part of day, occasional loss of work efficiency); 3 = severe (in bed one or more days, incapacitation) based on the subject's self-assessment of symptoms. In the listed categories, 'N' signifies those subjects who were evaluable for this measure.

In order to judge therapeutic effectiveness and to compare subjects' complaints, a severity profile score of dyspareunia was assessed using a rating scale.

In order to judge therapeutic effectiveness and to compare subjects' complaints, a severity profile score of pelvic tenderness was assessed using a rating scale: missing; 0 = none (no pain during intercourse); 1 = mild (minimal tenderness on palpation); 2 = moderate (extensive tenderness on palpation); 3 = severe (unable to palpate because of tenderness) based on the gynecological palpation by the attending physician. In the listed categories, 'N' signifies those subjects who were evaluable for this measure.

In order to judge therapeutic effectiveness and to compare subjects' complaints, a severity profile score of induration was assessed using a rating scale: missing; 0 = none (no pain during intercourse); 1 = mild (uterus freely mobile, induration in the cul-de-sac); 2 = moderate (thickened and indurated adnexa and cul-de-sac, restricted uterine mobility); 3 = severe (nodular adnexa and cul-de-sac, uterus frequently frozen) based on the gynecological palpation by the attending physician. In the listed categories, 'N' signifies those subjects who were evaluable for this measure.

The investigator rating scale used in this study was based on the validated CGI scale, which is widely used as a simple tool to assess the overall effect of treatments. The investigator or a sub-investigator rated the total improvement according to the following scale: Score 1 = very much improved; Score 2 =much improved; Score 3 = minimally improved; Score 4 = no change; Score 5 = minimally worse; Score 6 = much worse; Score 7 = very much worse. None of the subjects reported Score 7. In the listed categories, 'N' signifies those subjects who were evaluable for this measure.

The subject rating scale used in this study was based on the validated CGI scale, which is widely used as a simple tool to assess the overall effect of treatments. The subject was asked to rate her satisfaction with the study treatment according to the following scale: Score 1 = very much satisfied; Score 2 = much satisfied; Score 3 = minimally satisfied; Score 4 = neither satisfied nor dissatisfied; Score 5 =minimally dissatisfied; Score 6 = much dissatisfied; Score 7 = very much dissatisfied. In the listed categories, 'N' signifies those subjects who were evaluable for this measure.