Neurobiological mechanisms invoking the release of endogenous opioids and depression of stress hormone release are believed to be the basis of acupuncture analgesia. This study compared plasma β-endorphin and cortisol levels with self assessment scores of intensity of pain, before and after 10 days of electro-acupuncture treatment in patients suffering from chronic pain as a result of osteoarthritis knees. Forty patients of either sex over 40 years with primary osteoarthritis knee were recruited into a single-blinded, sham-controlled study. For electro-acupuncture group the points were selected according to the Traditional Chinese Medicine Meridian Theory. In the sham group needles were inserted at random points away from true acupoints and no current was passed. Both groups were treated for 10 days with one session every day lasting for 20–25 min. Pre- and post-treatment Western Ontario and McMaster Universities (WOMAC) index of osteoarthritis knee and Visual Analogue Scale (VAS) for pain were recorded and blood samples were taken for the measurement of plasma cortisol and β-endorphin levels. Following electro-acupuncture treatment there was a significant improvement in WOMAC index and VAS (p = 0.001), a significant rise in plasma β-endorphin (p = 0.001), and a significant fall in plasma cortisol (p = 0.016). In conclusion electro-acupuncture resulted in an improvement in pain, stiffness and disability. Of clinical importance is that an improvement in objective measures of pain and stress/pain associated biomarkers was shown above that of a sham treatment; hence demonstrating acupuncture associated physiological changes beyond that of the placebo effects.