A 392-patient trial will test which sequence of three common painkillers—amitriptyline, pregabalin, and duloxetine—works best for the burning, stabbing nerve pain that afflicts up to a third of people with diabetes. Diabetic peripheral neuropathic pain (DPNP) is notoriously difficult to treat. Doctors have several drug options but no clear evidence on the best order to try them, or whether combining two drugs works better than switching to a different single drug. Patients often endure months of trial-and-error prescribing, with many never achieving adequate relief. This trial directly compares three treatment pathways—each starting with a different drug and adding a second if the first fails—to find the most effective and cost-effective approach. If one pathway clearly outperforms the others, the NHS could adopt it as a standard protocol, sparing thousands of patients from prolonged pain and reducing the guesswork in primary care. The health economics analysis will tell funders whether the best clinical option also delivers value for money. For a condition that disrupts sleep, mood, and daily function, a definitive treatment pathway would be a practical, everyday improvement for a large and growing patient group.
View original technical description
DESIGN: Multicentre, double-blind, crossover RCT of Treatment Pathways (TP) for diabetic peripheral neuropathic pain (DPNP). Each TP has 2 periods (6-week monotherapy followed by 10-week combination therapy: amitriptyline followed by pregabalin; pregabalin followed by amitriptyline and duloxetine followed by amitriptyline). Eligible patients will be randomised to 1 of 6 treatment sequences. Each sequence will examine all 3 TPs in stratified order. An internal pilot will be incorporated assessing recruitment and retention feasibility, with stop/go criteria at 6 and 12 months into recruitment. SETTING: 8 DPNP hospital centres aligned with 80 primary care practices in the UK. TARGET POPULATION: Primary/secondary care DPNP patients >/=18 yrs old. INCLUSION: Daily pain for > 3 months, neuropathy confirmed by Michigan Neuropathy Screening Instrument (MNSI) score >/= 3, Douleur Neuropathique 4 (DN4), numeric rating scale (NRS) 24-hour average pain of >/=4, stable diabetes control (A1c<108mmol/mol). EXCLUSION: Significant cardiovascular/liver disease, non-diabetic neuropathies, alcohol/substance abuse, major psychiatric disorders, contraindications to study medications, pregnancy/breast feeding. HEALTH TECHNOLOGIES TO BE ASSESSED: Patients will be randomised 1:1:1:1:1:1 to 6 parallel sequences stratified by site and titrated to maximum tolerated dose. Treatment response will be assessed by 7-day average 24-hr NRS score at the end of monotherapy. Patients with scores 3) will receive the second agent in combination. Patients intolerant of monotherapy (or with no change in pain score) will be switched to the 2nd agent. Before starting the next TP, the study drug will be tapered down (3 days) and washed out (4 days). Blinding will be maintained with over-encapsulated drugs, matching placebo and identical dosing regimen for all groups. PRIMARY OUTCOME MEASURE: 7-day average 24-hour pain on NRS (0=no pain; 10=worse pain imaginable) during the final follow-up week. SECONDARY OUTCOME MEASURES: Head- to-head comparison of monotherapy, BPI-MSF measure of pain interference with function, mood assessed by BDI and HADS, health status assessed by SF-36, sleep duration and quality examined using diaries, responder rates analysis (30% and 50% pain relief), EuroQoL-5D-5L, Patient Global Impression of Improvement (PGI-I) modified health resource questionnaire based on Client Service Receipt Inventory (CSRI) to capture health resources used. Adverse events and serious adverse events will be assessed. The relationship of pain phenotype to treatment response will be assessed using Neuropathic Pain Symptom Inventory (NPSI). SAMPLE SIZE: A 0.5 mean change between groups represents an 8% increase in the proportion of subjects improving by at least 1 point (the MCID in an individual). Using 0.5 change, within subject SD of 1.65, two sided significance level of 0.0167, and 90% power, assuming 25% attrition, we require 392 subjects. ANALYSIS: The primary outcome and other continuous outcomes will be analysed using a mixed model with patient, treatment, sequence and period entered into the model. Patient will be entered as a random term. Contrasts will be used to evaluate the difference in means. HEALTH ECONOMICS: A cost-utility analysis will be conducted. QALYs will be calculated using the EQ-5D-5L. An NHS and PSS perspective for costs will be taken, with personal and societal costs included within a sensitivity analysis. The analysis will be reported to the NICE reference case standard. The main outcome will be incremental cost per QALY gained.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know