Transcript Gabapentin & Chronic Pain

Common Office Problems
G. Michael Allan
Associate Professor, Family Med, U of A
Objectives
1.
Neuro
A. Bell’s Palsy
B. Gabapentin in Chronic Pain.
C. X-ray Backs
2.
Women’s Health
A. Pap Testing (Cleaning the Cervix & Lubricant use)
B. Hot Flashes: Treatment Options
C. Incontinence
3.
4.
MSK:
Miscellaneous
1.
Anemia, Constipation,
Neuro
Bell’s Palsy: What to do and not to do?
• Clinical Question: Do corticosteroids or
anti-viral medications provide any benefit
to patients with Bell’s Palsy?
Bell’s Palsy: What to do and not to do?
• Scottish study1: 551 pts, within 72 hours,10 days of Acyclovir 400mg
5x/d & Prednisolone 25mg BID together or each separate or placebo.
– Prednisolone vs placebo for complete recovery:
• 3 months, 83% vs 64%, NNT =6
• 9 months, 94% vs 82%, NNT =8
– Acyclovir no advantage over placebo.
• Scandinavian study2: 839 pts, within 72 hours, Valcyclovir 1000mg TID x
7days & Prednisolone (60mg x5days then tapered by 10 mg/day) or
each separately or placebo.
– Prednisolone vs placebo for complete recovery
• at 12 months, NNT =7
– Valcyclovir no advantage over placebo.
• Acyclovir1 or Valycylovir2 added to Prednisolone no advantage.
Bell’s Palsy: What to do and not to do?
• Before Cochrane concluded insufficient evidence to recommend either
anti-virals3 or corticosteroids4 for Bell’s palsy.
• 2007 Japanese study:5 valcyclovir added to prednisolone
statistically significantly improved recovery over
prednisolone alone (NNT 15). However,
– Study smaller (296 pts), high drop-out, & unblinded.
– Another Japanese study same year = no benefit adding valcyclovir to
prednisolone.6
– 2 most recent studies1,2 larger, superior methodologically, and found
anti-viral therapies added to prednisolone provides no benefit over
prednisolone alone.
• Dosing of Prednisolone and Prednisone is equal (e.g. Prednisolone
25mg BID equals Prednisone 25mg BID).
Bell’s Palsy: What to do and not to do?
• Bottom-line: The best evidence indicates
that corticosteroids (in doses of Prednisolone
25mg BID or 60mg x 5days then tapered by
10mg/day) improve the odds of complete
recovery from Bell’s Palsy
• Anti-virals, used either alone or in addition to
prednisolone, offer no advantage.
Gabapentin & Chronic Pain:
Missing Evidence and Real Effect?
• Clinical Question: What is the evidence to
support gabapentin (or pregabalin) in
chronic peripheral neuropathic pain?
Gabapentin & Chronic Pain:
Missing Evidence and Real Effect?
• Review: 20 RCT Gabapentin for off-label use (17/20 pain)1
– 8 never published (40%)
– Outcomes in 12 published studies
• 4 used planned primary outcome.
• Of 180 predefined secondary, 122 (68%) not reported
• Meta-Analysis of all trials (including unpublished)2
– Moderate-marked improvement in pain for 13-17.5% more,
• NNT = 6-8 (2 weeks)
• Efficacy greatest post-herpetic neuralgia.
• No benefit for acute pain or in dose escalation >900mg (with
>adverse events)
– Adverse events: NNH 8 (dizziness, somnolence, confusion, etc)
Gabapentin & Chronic Pain:
Missing Evidence and Real Effect?
• Cochrane review3 (published only): relief in chronic pain,
NNT 3-4
– Another review:4 similar numbers but worse with unpublished trials.2
• Pregabalin best case: Chronic pain relief NNT 4-5 but only
published data.5
– Drug Review6 (+ unpublished trials), intermittently >placebo but not
consistent.
• No direct trial evidence of superiority over gabapentin.
• One trial active comparator: Pregabalin not >placebo but TCA
were.
• Other reviews: TCA similar7 or perhaps superior7,8 to
gabapentin or pregabalin. Possible time & trial quality bias7
• Publication bias & selective reporting likely more in industry
funded research, non-profit funded RCTs also selectively
report outcomes9
Gabapentin & Chronic Pain:
Missing Evidence and Real Effect?
• Bottom-line: The apparent benefit of
gabapentin in chronic pain was exaggerated
by publication and reporting biases.
• In carefully selected patients with peripheral
neuropathic pain, gabapentin may offer
moderate+ pain relief for 1 in every 6-8
patients but causes adverse events in a
similar number.
• No trial evidence pregabalin > gabapentin
What if they say, “What if,…”
• A 40 year old ♀ hospital cleaner is in concerned
about her back pain for almost 10 weeks.
–
–
–
–
Her range motion is 50%,
tender in paraspinal muscles around L3-5.
occasional radiation into buttock & upper legs.
Her neuro exam is normal
• She wants to know what is causing this!
• She isn’t satisfied with you explanation and would
like an x-ray.
Law of Unintended Consequences
• RCT, UK, 421 GP pts, low-back pain ≥ 6 weeks
– 60% female, mean age 39, back pain x 10 weeks
• At 3 months Stat diff in:
– Still in pain: 74% X-ray vs 65%, NNH 12
– Had f/u doctor visit: 53% X-ray vs 30%, NNH 5
– Self rated health status: 5% worse in X-ray group.
• After 6 more months
– Few borderline worse (disability & pain) but not Stat Sign
• However, ≥80% of both groups want X-rays.
– Those with x-rays more satisfied with visit.
– X-rays findings did not correlate to clinical
BMJ. 2001 Feb 17;322:400-5.
More evidence,…
• Meta-analysis1 of 6 trails, 1804 pts.
– Imaging vs usual: 4 X-ray, 2 CT or MRI
• Short term and Long-term outcomes of pain,
function, quality of life, mental health and
patient satisfaction did not differ sign
– Pain at 3 months was borderline worse with xray (SMD 0.19, CI -0.01 to 0.39)
• Same comparing MRI vs X-ray2
1. Lancet. 2009;373:463-72. 2. JAMA 2003;289:2810–8.
Women’s Health
Cooling the Hot Flashes
• A 51 year old woman is having a terrible time with
hot flashes. Her family history includes breast
cancer.
• You say: There’s lots to try but none work like HRT
• True: Well-designed Meta-analysis of 43 RCT’s
–
–
–
–
SSRI/SNRI (mid dose)= 1.13 ↓Hot Flashes/d (vs placebo)
Clonidine (≤0.075mg BID) = 0.95 - 1.63 ↓Hot flashes/d
Gabapentin (300mg TID) = 2.05 ↓Hot flashes/d
Soy Isoflavone Extract (50-70mg/d)= 0.97-1.22 ↓
• Endometrial safety with Isoflavone still unresolved.
• Estrogen best (2.5-3 ↓ Hot flashes/d)
JAMA 2006; 295: 2057-71.
Cleaning: Yes or No
• You are about to show the PAP test to a med
student and she asks about cleaningswabbing the cervix before the PAP test.
• You say: Cleaning of the Cervix will enhance
the Quality of the PAP test.
• False1
– RCT, 616 ♀, Family Practice,
– No difference
1. Can Fam Physician 2007;53:1328-1329.
PAP Case: I Forgot more than you Know
• You are brining you med student in to
observe her first PAP test. As you start, the
medical student asks “Is that lubricant? Our
gynecology professor says it wrecks the PAP
test.”
• You say: Lubricant on a Speculum will not
impact the PAP test?
PAP testing: Lubricant
• 4 RCTs address this question.
• The largest RCT1: 2906 patients
– Water soluble lubricant on speculum vs. tap water
– No difference in quality of cytology
• Two smaller RCTs:
– 182 patients2 and 70 patients3: No difference in PAP test
adequacy.2,3
• Quasi-randomized (randomized months), 3460
PAP tests4
– No difference in PAP test adequacy
• Conventional cervical cytology smears (glass slide)
were used in all studies1-4
PAP testing: Lubricant
• No RCT assessed lubricant on liquid-based PAP
– Retrospective review of 4068 liquid-based pap tests
• 15 (0.4%) obscuring material causing misinterpretation of pap
• ~ ½ may related to lubricant use (+ technologist inexperience)6
– Two studies applied lubricant directly into liquid-based
cervical cytology samples7,8
• One ↓ cell counts (after dilution) but adequacy not
assessed/reported7
• Other demonstrated no impact on liquid based PAP outcomes8
• One RCT also examined if lubricant affected testing
for Chlamydia and found no affect after 5535
samples.4 (Gonorrhea to uncommon to assess)
PAP cleaning & Bottom-line
• Bottom-line: A small amount of water soluble
lubricant on a speculum does not reduce the quality
of the PAP test and probably does not effect
microbiologic results either. The present evidence
suggests liquid-based PAP tests would be minimally
effected or not at all.
1) Can Fam Physician 2007;53:1328-1329.
Incontinence: Diagnosis
• You med student just saw a 50 year female with
incontinence and has planned a large work-up.
• You say: I bet I could get the Dx with 2 questions?
• True (probably): Mid size diagnostic cohort
• Accuracy Overall: correct 62%, partially correct
(Mixed) 23%, and 15% wrong
– Kappa : 3IQ (?s) = 0.65 – 0.69 & Specialists (gold
standard) Dx = 0.65
• In the last 3 months did you leak urine during
– Physical activity: coughing, sneezing, lifting, or exercise,
– You had the urge to go but couldn’t make it fast enough.
Ann Intern Med. 2006;144:715-723
Incontinence: Treatments
• Our 50 year old has mixed incontinence. What
might work?
• You say: Nothing works!
• False: Sys rev & Meta: 96 studies (14 K pts)
• Findings
– Works: Oxybutynin (5-10mg), Tolterodine (4mg), Pelvic
floor exercise
– Probable: Duloxetine (20-80mg)
– Possibly help (mixed): medical device (pessary), injected
bulking agents; topical HRT.
– No help: electric/magnetic stimulation,
– Harm: Oral HRT
Ann Intern Med 2008; 148; 459-73.
MSK
OA: Topical NSAIDs
• A 50 year old male is with knee OA. He finds
NSAIDs irritating to the stomach but doesn’t
get much help from Acetaminophen. He has
heard that creams may work.
• You say: Topical NSAIDs are less effective
than oral NSAIDs in knee OA.
• False
– RCT 622pts, oral vs topical, No diff except AE
– 2 Meta vs placebo, ES=0.40, NNT 5
J Rheumatolo 2004;31(10): 2002-12. BMJ 2004;329(7461):324.
March 05
www.Bandolier.com
Tennis Elbow: First do,…
• J. McEnroe (42) has lateral epicondylitis &
has tried NSAIDs (Topical & oral), splint &
physio. He wonders about steroid injections.
• He wonders what works.
Tennis Elbow: First do,…
• Steroid Injection1
–
–
–
–
–
RCT, 198 pts, 3 arms (injection/physio/usual)
Results: physio < pain meds
6 weeks Injection = NNT 8
52 wks Injection = NNH 4
Injections= ↑ recurrences & poor long term outcome
Improve Steroid Physio Usual
6 wks
78%
65%
27%
52 wks
68%
94%
90%
Tennis Elbow: First do,…
• Transdermal Nitro2
– Placebo vs ¼ Nitro 5mg/24hr patch (95 elbows)
– Nitro group: ↑ strength &
↓ pain/tenderness
– Asymptomatic ADL (6 mon): AB =21% or NNT 5
– Side-effects frequent (12% quit, NNH=8)
1. BMJ 2006;333:939
2. Am J Sports Med 2003;31:915–20
Rotator Cuff Injections:
Horseshoes & Hand-grenades
• You are teaching a medical student to inject a
sub-acromial bursa for rotator cuff but she is
nervous
• You say: Don’t worry, close is good enough.
• True (likely). RCT, 106 pts, mean 51 yrs, 61%
♀, Rot cuff >3 months
– RCT virtually no difference between radiologically
guided bursa injection and buttock injection
BMJ 2009;338:a3112
Grab-bag
All Plugged Up & No Way to Go
• A 42 y.o. female has chronic constipation.
BM’s are q4 days & she is uncomfortable
50% of the time. She wants to start Docusate
• You say: Docusate doesn’t work
• True: Sys Review of >50 trials & 13
therapies1
• Best Evidence for
–
–
–
Polyethylene Glycol (8 T - 743pts): 2-3bm/wk, NNT=2
Lactulose (10 T – 700 pts): 1-3 extra BM/wk, NNT=4
Psyllium (9 T – 997 pts): 1-3 extra BM/wk
Am J Gastroenterol 2005; 100: 936-71. 2) Am J Gastroenterol. 2007;102(7):1436-41.
Constipation Continued
• PEG dose is 17g daily & has been used X6
months2
• Poor Evidence for/against (Trial #): Bran*, Colchicine*,
Stimulant (Senna)*, Sorbital, Methylcellulose,
Misoprostal, Docusate
– (* mixed evidence, perhaps may be beneficial)
Am J Gastroenterol 2005; 100: 936-71. 2) Am J Gastroenterol. 2007;102(7):1436-41.
Iron: A Plug for Therapy
• 80 y.o. ♀ with fatigue has ongoing anemia (Hgb
101 + Ferritin 8). Unchanged x 4 yrs & scope then ve. Iron causes her constipation.
• You say: A low dose of iron works as well and
doesn’t cause constipation.
• True: RCT, 90 elderly anemic x2m (150, 50, 15mg)
– Hgb ↑ significantly (14 Hgb points), no diff between doses
– AE ↑ significantly as dose ↑, Dropout 15mg vs 150mg
NNH = 5
• 15mg elemental iron = 2.5 ml’s of Fer-In-Sol
30mg of elemental iron = Ferrous sulphate 150mg
Am J Med 2005; 118: 1142-7
“Butt-Out!”
• A 35 y.o. ♂ is considering quitting smoking.
She would like a pill but is nervous about
Varencline.
• You say: The only other option is Buproprion
• False: Meta-analysis; Bupropion (16 studies,
5K pts), Nortriptyline (5 studies, 1K pts)
– Nortrip 75-100mg = Bup 150/300mg (NNT = 11)
– Drop-out: Nortrip (4-9%), Bup (7-12%).
– 10 weeks: Nortrip $22 & Bup $175
Hughes JR, et al The Cochrane Library, 2004, Issue 3, Art. NO CD 000031.
“Sticks and Stones,…?”
• A 34 y.o. ♂ was Dx with a stone 2 nights
ago in ER. He’s referred to urology but it will
be 18 days. He is hoping for something to
help.
• You say: Sorry, but we only have pain relief.
• False: Sys Rev, 9 Trials (693 pts),
– Distal 1/3 of ureter & Absolute benefit for stones
≥5mm twice that for <5mm (31% versus 15%)
– Stone pass: 47% vs 78%, NNT = 4 at 4wks,
– nifedipine 30mg, terazosin 5mg, doxazosin 4mg
or tamusolin 0.4mg for 4 weeks.
1) Lancet 2006; 368: 1171-9. 2) Ann Emerg Med 2007; 50:552-63
Care of Elderly
Drug withdrawal
• You see a 81 year old, new to you, in the a nursing
home (he is on 10 medications). He is demented &
incontinent.
• You say: (To the family) He’d likely feel & be better
with less meds.
• True: Cohort,1 190 pts, mean age 81, 31% ♂, 93%
dementia/incontinent, 45%+ CVD,
– Mean 7 drugs. Stopped mean 2.8 drugs, 10% “failure”
– Death 21% (vs 45%) & hospitalization 12% (vs 30%)
• Other research shows polypharmacy increasing
(200% x 10 yrs) in elderly2 & associated with
increased mortality3
1) IMAJ 2007;9:430–434. 2) Dr Ross Upshur (pc). 3) Drugs Aging. 2009;26:1039-48.
First line to withdraw
• Okay, you are going to start to withdraw medicines.
Which do you start with?
• DART-AD: 165 pts, mean 85, 76% female, longterm care (high quality)
– Withdraw antipsychotic (placebo) or continue
• Outcomes
– Behavior: NPI behavior score worsened by 1.7% (2.4
/144). Not stat sign.
– Mortality: at 2 years, 71% continued anti-psychotic vs
46% placebo, (Diff = 25%, NNT 4)
• Patients on anti-psychotics deserve a break
Lancet Neurol 2009; 8:151–57. PLoS Med 5(4): e76.doi:10.1371/journal.pmed.0050076
First line to withdraw
• Okay, you are going to start to withdraw
medicines. Which do you start with?
• DART-AD: 165 pts, mean 85, 76% female, longterm care (high quality)
– Withdraw antipsychotic (placebo) or continue
• Outcomes
– Behavior: NPI behavior score worsened by 1.7% (2.4
/144). Not stat sign.
– Mortality: at 2 years, 71% continued anti-psychotic vs
46% placebo, (Diff = 25%, NNT 4)
• Patients on anti-psychotics deserve a break
Lancet Neurol 2009; 8:151–57. PLoS Med 5(4): e76.doi:10.1371/journal.pmed.0050076
Why is Santa both fat and jolly?
• An elderly overweight male toy distributor is in. H
won’t talk about weight loss (“I’ve lived forever”).
• Question: What is the mortality risk of being
overweight for older patients?
• Cohort age 70-74: 4,931 ♂ (x 8.1 yrs) & 5,042 ♀ (x9.6yrs).1
• Best BMI is 25-30 (overweight)
– At the low end of normal (BMI ~20) similar to BMI 35
– Inactivity generally more important: increase risk 28%
men and double in women
– Other studies support.2
J Am Geriatr Soc 2010; 58:234–241. 2. NEJM 2006; 3553: 779-87. JAMA 2006; 296:7986. J Am Geriatr Soc 2005; 53: 2112-8. J Am Geriatr Soc 57:2232–2238, 2009.
Why is Santa both fat and jolly?
The End