#103 Duloxetine (Cymbalta®): Jack of All Trades, Master of None?
CLINICAL QUESTION
How safe and effective is duloxetine for the treatment of chronic painful conditions?
BOTTOM LINE
Compared to placebo, duloxetine appears efficacious in neuropathic pain, improving pain by 50% or more for one in seven people. One in 20 people (over placebo) will have to quit due to adverse events.
Reading Tools for Practice Article can earn you MainPro+ Credits
Join NowAlready a CFPCLearn Member? Log in
EVIDENCE
Compared to placebo: Meta-analysis1 of five randomized controlled trials (RCTs) with 2,589 diabetic peripheral neuropathy patients over ≤12 weeks.
- ≥50% improvement in pain: duloxetine 46% vs. placebo 30% (Number Needed to Treat (NNT)=7).
- Mean pain scores improved by ~0.95 more with duloxetine than placebo (on 0–10 scale).
- Adverse events leading to discontinuation: Duloxetine 60 mg/day 12.6% vs. placebo 5.8% (Number Needed to Harm (NNH)=20).
- Adverse events included nausea (NNH=7), somnolence (NNH=14), dry mouth (NNH=14), and dizziness (NNH=22).
- Increasing dose no advantage: no difference in response but more adverse events.
- Similar results in one other RCT2 and several meta-analyses.3-5
- RCT6 of 804 diabetic peripheral neuropathy patients treated with either duloxetine 60 mg/day or pregabalin 300 mg/day for eight weeks.
- ≥50% improvement in pain: Duloxetine 40% vs. pregabalin 28% (NNT=9).
- ~12% discontinued treatment due to adverse effects in both groups.
- Trial sponsored by manufacturer of Cymbalta®.
- Previous small trials showed no conclusive difference between duloxetine and amitriptyline7,8 or pregabalin8,9 in neuropathic pain.
CONTEXT
- Duloxetine is also efficacious in other chronic painful conditions, including fibromyalgia1 (NNT=8), chemotherapy-induced neuropathic pain (NNT ~9),10 and osteoarthritis of the knee (NNT=6–8).11,12
- For depression, duloxetine has similar efficacy and overall safety to other second generation antidepressants, with no clear advantages compared to other agents.13
- Duloxetine trials are at moderate-to-high risk of bias: industry funding, short duration, high risk of selective outcome reporting, high drop-out rates, multiple outcomes without adjustment and possible selective publication.
Latest Tools for Practice
#370 Antibiotics or no antibiotics for acute diverticulitis, that is the question!
Do antibiotics change clinical outcomes for patients with acute uncomplicated diverticulitis?
#369 Remind me, do medications that target brain amyloid improve my dementia?
Are amyloid-targeting monoclonal antibodies safe and effective for mild cognitive impairment or Alzheimer’s dementia?
#368 Sodium Restriction in Heart Failure: Beneficial or pouring salt in the wound?
Does sodium restriction improve outcomes in patients with chronic heart failure?
