Patients with diabetic neuropathy and total symptoms score(TSS) >7 points were invited to this open multicenter study. Patients were free of pain medications and severe diabetic complications .Patients started alpha lipoic acid (ALA)1800 mg for 4 weeks. Patients with a decrease >3 points in the TSS were randomly allocated to 600 mg of ALA (ALA group) or no medications (ALA withdrawal) for 16 weeks. In each visit investigators evaluated any change in the TSS and the necessity of rescue medication to control symptoms (mainly pain). At the end of the study investigators compared between ALA and ALA withdrawal groups TSS levels and the frequency of use of rescue medications. Physicians were free to manage glucose to maintain Hba1c close to the ADA target (HbA1c <7%).

This trial was conducted in accordance with the Declaration of Helsinki and was approved by the ethics committee of Universidad Popular Autónoma del Estado de Puebla, Mexico. All participants provided a written informed consent. Type 2 diabetic patients (according to American Diabetes Association (ADA) criteria) with symptomatic diabetic sensorimotor polyneuropathy (DSPN) defined as the presence of neuropathic symptoms (pain, paresthesias, or numbness) were invited to participate in this open-label multicenter trial. Inclusion criteria were: total symptom score (TSS) >7 points, HbA1c<10%, and serum creatinine <2 mg/dl. Exclusion criteria were evidence of active cardiovascular disease, malignancy, or any other conditions causing neuropathic pain, use of analgesic, antidepressant, or antiepileptic drugs, or any other medication aimed to relief neuropathic pain. In addition, child-bearing female patients not using any effective birth control method and under surveillance of a board-certified gynecologist were excluded.

Phase 1. All patients meeting inclusion criteria received 600 mg of alpha lipoic acid (ALA) (Meda Pharma, Germany) orally tid, 30 min after each main meal for 4 weeks. During phase 1, no medication for relief of neuropathic pain was allowed. Each participating site was in charge to maintain glycemic control based on the investigator's judgment attempting that all patients were treated according to the american diabetes association (ADA) guidelines. All patients were seen once a week, and at each site visit, TSS was assessed along with a pill count to ensure drug adherence, presence of adverse events and, if needed, treatment adjustments to maintain glucose levels within the ADA targets. Patients with a TSS reduction >3 points by the end of phase 1 were selected to proceed with phase 2 of the study. Patients with a decrease <3 points in TSS or that used other neuropathic pain drugs were excluded from study phase 2.

Phase 2. Patients with a decrease of ≥3 TSS points after phase 1 were randomized to receive 600 mg of ALA orally qd for 16 weeks or ALA withdrawal. Patients were scheduled to visit the clinic every 2-3 weeks for TSS, monofilament and assessment. If needed, the patient was prescribed analgesic rescue medication which was monitored at each visit. Primary endpoint was the change in TSS in the two groups studied in phase 2 and the frequency of use of rescue medications Neurological examination was performed at baseline and after phase 1 and 2 including the monofilament test, vibration perception threshold (VPT), and ankle reflexes. A 10g nylon monofilament (Thio-Feel ® Meda Pharma, Germany) was applied to four anatomical sites in each foot (1st, 3rd and 5th metatarsal heads and plantar surface of distal hallux) as previously described (correct answer = 1 point, with a maximum of 4 points in each foot). Eight correct answers were considered normal, 1-7 correct answers indicated reduced monofilament sensation, while absent sensation was assumed if no answer was correct. VPT was evaluated using a 128-Hz tuning fork (Thio-Vib ®, Meda Pharma,Germany) applied bilaterally at the tip of the great toe. Responses were categorized as abnormal (no perception of vibration), present (examiner perceives vibration <10 seconds after patient reported disappearance of vibration perception) and reduced (examiner perceives vibration >10 sec after patient reported disappearance of vibration perception). Ankle reflexes were graded as normal, decreased, and absent

Inclusion Criteria:

• Type 2 diabetic patients (according to American Diabetes Association (ADA) criteria)
• Symptomatic diabetic polyneuropathy
• Total Symptom Score (TSS) >7 points,
• HbA1c<10%,
• Serum creatinine <2 mg/dl.

Exclusion Criteria:

• Active cardiovascular disease
• Malignancy
• Any other conditions causing neuropathic pain
• Use of analgesic, antidepressant, or antiepileptic drugs, or any other medication aimed to relief neuropathic pain.
• Child-bearing female patients not using any effective birth control method and under surveillance of a board-certified gynecologist

5 months