• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

PFMT with biofeedback

StudyStudy type and ELNo. of patientsPatient characteristicsInterventionComparisonLength of follow-upOutcome measuresEffect sizeAdditional comments
Berghmans 1996254RCT
EL = 1+
40Women, aged 18–70 years, urodynamic stress UI (mild-moderate [grade 1–2])
Exclusions: pronounced lesions of pudendal nerve, neurogenic bladder, urological or gynaecological surgery, pacemaker
PFMT + biofeedback (n = 20)PFMT (n = 20)4 weeks tx48 h pad test (mean change, g)−57 vs −54%, P = NSFunding: none declared.
PFMT: anatomy and function of pelvic floor, vaginal palpation every week, 12 ‘treatment sessions’ 3×/week in standing, crawling, side position, 25–35 min; home practice 3×/day. Duration of contractions 3–30 s, repeated 10–30 times. PFMT combined with coughing, climbing stairs, lifting, jumping, completed the program.
Biofeedback: vaginal probe attached to portable EMG. Contract-relax period, number of cycles, and tx time varied for treatments 1–12; sensitivity and threshold individualised to pts.
4 pts in biofeedback grp vs 0 had previous tx with medication and /or physical therapies
Cured/improved/ worse (n)3/14/3 vs 5/14/1
Leakage episodes/weekNo numerical data
Castleden 1984255RCT, cross-over
EL = 1−
19Women, mean age 55 years (23–85), stress UIPFMT + biofeedback (n = 19)PFMT (n = 19)4 weeks tx (2×2 week periods)VAS score at 4 weeks (scale not described)+23.9 (0, +79) vs +6.7 (−32, +26)Funding: Kingsdown medical supplied perineometer.
PFMT: 4–5 contractions every hour and interrupted micturition. Perineometer with vaginal catheter used at least 1×/day.
[EL = 1−] Limited information, especially re units for results.
No baseline data therefore unable to report % changes for outcomes.
Perineometer readings at 4 weeks (units not given)+2.0 (−3, +6) vs +1.5 (−1, +6)
Change in symptoms (n)Cure 14/19; worse 1/19 at 4 weeks
Glavind 1996256RCT
EL = 1+
40 randomised, 34 completedWomen, mean age, genuine stress UI
Exclusions: DO, previous UI surgery
PFMT + biofeedback (n = 19)PFMT (n = 15)4 weeks tx, follow-up at 3 months1 h pad test−88.4% (95% CI −78%, −94%) vs −54% (−2.1%, −78%)Funding: none declared.
Pts ability to contract PFM assessed by digital palpation. PFMT: 2–3 individual instruction; held contractions for 5–10 s, in supine, sitting, standing positions. Home exercise daily at least 3×/day and as often as possible.
Biofeedback: Dantec 21L20, 4 sessions 1×week, vaginal electrode and rectal catheter.
Objective cure (unclear whether measured by pad test)58% vs 20%, P = NS
Subjective assessmentCure: 26% vs 0%
Improvement 42% vs 29%
Acceptability of tx75% vs 52%
Sherman 1997257RCT
EL = 1+
39Women (active female duty soldiers), mean age 33 years, urodynamic UI (77% stress; 23% mixed)PFMT + biofeedback (n = 23)PFMT (n = 16)8 weeks txSubjective assessment of time between voids, degree of urgency, ability to stop urine stream, activity level, volume per void, severityNo sig. differences in changes between grpsFunding: none declared.
PFMT: initial educational session, exercises 10–10 contract-rest cycle ×5, home practice 20 min 2×/day. Pts put on bladder training schedules and taught urge control to use ‘when applicable’.
Pelvic examination at baseline to assess pelvic support and muscle tone.
Biofeedback: J&J Biofeedback system (EMG), vaginal and abdominal electrodes. Biofeedback home trainers given to people during the first week of tx.
Withdrawals: 1 BF grp, 5 PFMT grp: ‘no difference’ between those who withdrew vs continued, or in pad test results at 1st assessment.
Leakage episodes /day (mean change)−61 vs−67%, P = NS
Voids/night−60 vs−79%, P = NS
Urodynamic parametersNo sig. differences between grps in changes in bladder capacity, MUCP, detrusor contraction pressure
PFM maximal muscle contractionNo sig. difference between grps
Shepherd 1983258RCT
EL = 1−
22Women, mean age 48 years, urodynamic stress UIPFMT + biofeedback (n = 11)PFMT (n = 11)6 weeks treatment, follow-up at 18 weeksSelf-reported cure8/11 vs 3/11 (73% vs 27%)Funding: none declared.
PFMT: weekly clinic visits, home exercise program. No further information.
Biofeedback: intravaginal exercises connected to visual biofeedback.
[EL = 1−] Insufficient information regarding the interventions.
Self-reported cure or improvement10/11 vs 6/11 (91% vs 55%)
Aukee 2002259 and 2004260RCT
EL = 1+
35 randomised*
(31 attended 1 year follow-up) at 1 year
Women, mean ~51 years (21–70), with urodynamic stress UI; no previous surgery for UI; abdominal leak point pressure > 90
Exclusions: genital prolapse beyond hymen, pregnancy, severe concomitant diseases
PFMT (n = 15)PFMT + EMG biofeedback (n = 15)3 months tx, 1 year follow-upPFM activity, μV (supine and standing) at 3 monthsIncreases in both grps from baseline, P < 0.001, PFMT + EMG vs PFMT P = 0.024Funding: none declared.
PFMT; 5 sessions (weeks 0, 1, 4, 8–12), and practised at home (20 min/day 5×/week). Mean no. training days at home was 56.2 (range 21–87).
Home biofeedback grp given FemiScan device, containing a training programme and sound processor for verbal instructions. Mean no. training days at home was 68 (range 9–130); mean 47.5 days (range 6–93) without the device.
After 3 month tx period, pts advised to continue PFMT on own initiative.
#leakage index contains 13 types of physical exertions that trigger UI in women with SUI (1 never, 5 always), measured on ordinal scale of 1–5.
*Discrepancy in pt numbers between Aukee 2002 (first 30 pts) and 2004 (final number, 35).
24 h pad test at 3 months (adjusted for differences in baseline values)−18.1 g vs −17.3 g from baseline, P = NS
Leakage index# at 3 months (adjusted)−2.1 vs −8.8 from baseline, P = NS
Adverse effects at 3 months3 vs 2 pts reported pain while training.
2 pts from PFMT + EMG grp did not use EMG (found vaginal probe uncomfortable).
% had surgery for UI at 1 yearPFMT vs PFMT + EMG:
9/19 vs 5/16, (47% vs 31%) P = NS
Pages 2001261RCT
EL = 1−
51 randomised
40 analysed
Women, mean age 51 years (27–80), mild-moderate SUI
Exclusions: significant medical illness; drugs influencing bladder control and functioning
PFMT (n = 27)PFMT + biofeedback (n = 13)3 months (1 month tx, 2 further months follow-up)FrequencyFrom baseline (no between grp comparisons):
Daytime reduced in PFMT grp; night reduced in both grps, P < 0.01
Funding: none declared.
[EL = 1−] 11 (46%) from biofeedback grp excluded after randomisation owing to concurrent illness or decision to withdraw.
Physical therapy consisted of group therapy 5×/week for 60 min, plus initial educational session about PFM and incontinence. Also encouraged to do ~100 contractions at home/day.
BF grp had introductory session then individual therapy for 15 min 5×/week (Gemini 2000 TM apparatus).
*PFM contraction assessed by investigator, and by speculum and manometric measurements.
Cure (no incontinence)69% vs 62%
PFM contraction*Both grps increased strength of voluntary contraction and cough-induced contraction, and improvement in closure of introitus, P < 0.05 from baseline
Adverse effectsNone
Sung 2000262,263RCT
EL = 1−
90Women ≥ 18 years, stress UIPFMT (n = 30)PFMT + electrical stimulation biofeedback (n = 30
Control (no tx, n = 30)
6 weeksLeakage episodes (mean [SD] change, unclear whether in no per day/week)−0.2 (0.5) vs −1.0 (1.2) vs 0 (0.7), [−37 vs −9% vs 0%] P < 0.001 ES grp vs othersFunding: Hallym Academy of Science (Korea).
[EL = 1−];unclear whether randomisation refers to patient selection into the study, or allocation of treatment.
PFMT: Physio instruction, home training following instructions on videotape.
Electrical stimulation given alternately with biofeedback (Elite compact device): 2×20 min/week; ES 24 s, 35 and 50 Hz.
Quantity of leakage (units and method of measurement not stated)−0.2 (0.4) vs −0.7 (0.8) vs 0 (0.4), [−9 vs −28% vs 0%] P < 0.001 ES grp vs others
BFLUTSNo between-grp comparisons. Improvements in ES grp in difficulties in daily lives, fluid restriction, physical activities, personal relations, P < 0.001 vs baseline
Peak PFM pressure (perineometer, mean [SD] change, mmHg)+1.0 (2.4) vs +6.2 (4.9) vs −4.5 (6.2), [+3 vs +18 vs −12%] P < 0.001 active grps vs control
PFM duration of contraction (s), mean change (SD)+0.6 (0.6) vs +0.7 (1.2) vs −0.1 (0.6), [40 vs 44 vs −6%] P < 0.001 ES grp vs others
Morkved 2002264RCT
EL = 1+
103 randomised, 94 analysedWomen (mean age ~47), urodynamic stress UI (24 also had urge UI)
Exclusions: involuntary detrusor contractions, residual urine > 50 ml, previous surgery for stress UI, UTI, use of concomitant txs, other diseases, neurological or psychiatric disease
PFMT + Biofeedback (n = 53)PFMT (n = 50)6 monthsObjective cure*58% vs 46%, P = NSFunding: Norwegian industrial and regional development fund, Norwegian national insurance administration, Trondheim regional hospital.
PFMT individualised (1/week 4 months, every 2 weeks 2 months). Home training; 10 high intensity contractions/day.
BF using BF-106 device, vaginal probe, individualised for each pt.
*≤ 2 g leakage on provocative pad test with standardised bladder volume 300 ml saline.
#9 social settings, scale 0–10 impossible-possible to participate.
Subjective cure40% vs 30%, P = NS
Cure (48 h pad test)65% vs 57%, P = NS
Mean PFM strength, cmH2O (vaginal balloon catheter)At 6 months 12.3 (95% CI 9.5 to 15.1) vs 11.1 (8.1, 14.1), P = NS
Mean Social activity index score#At 6 months 0.4 (95% CI −0.1 to 0.6) vs 0.3 (0.0, 0.5), P = NS
Wang 2004265RCT
EL = 1+
120 randomised, 103 completedWomen 16–75 years, OAB > 6 months, frequency ≥ 8×/day, urge UI ≥ 1×/day, no other conservative tx
Exclusions: pregnancy, concurrent medical conditions, genital prolapse > stage II, residual urine > 100 ml
PFMT (n = 40)Biofeedback+PFMT (n = 38)
Electrical stimulation (n = 42)
12 weeksSelf-reported improvement or cure of urge UI38 vs 50% vs 51%, P = NSFunding: National Science Council.
PFM strength measured by 1 finger palpation (Oxford grading method), and vaginal pressured measured using balloon probe.
PFMT; at home, PERFECT scheme, 3×/day.
ES: Intravaginal electrode, biphasic pulsed current, freq 10 Hz, pulse width 400 μs, 10 s on, 5 s off, intensity 20–63 mA or 40–72 mA, 20 mins/session, twice/week.
Duration of compliance with home programme; median 14.5 (0–44) days PFMT, 8.5 (0–44) days BF+PFMT.
Baseline differences in gravidity, parity and menopausal status (BF+PFMT vs ES); leakage episodes/day 0.86 PFMT vs 0.92 BF+PFMT vs 2.09 ES.
*Power using Oxford grading system (0–5), vaginal pressure using balloon probe.
Self-reported cure of urge UI30 vs 38% vs 40%
PFM parameters (mean change)*Power: −2 vs −2.5 vs 0
Time of fast contraction: −5.8 vs −6.2 vs −3.0
Vaginal pressure: −36.0 vs −38.4 vs −8.9
P ≤ 0.012 for all comparisons of PFMT or BF +PFMT vs ES
QOL (King’s Health Q), mean changes (SD) in total score50.3 ± 171.4 vs 185.9 ± 176.6 vs 180.1 ± 176.0, P ≤ 0.004 for PFMT vs other groups
Aksac 2003266RCT
EL = 1+
50Postmenopausal women taking HRT, urodynamic stress UIPFMT + biofeedback (n = 20)PFMT + palpation (n = 20)
Control (HRT only n = 10)
8 weeks1 h pad tests−94% vs −89% vs +3%Funding: none declared.
BF: Myomed-932 device vaginal probe in EMG pressure mode; 20 min, 40 cycles of 10 s activity and 20 s relaxation, 3×/week.
Digital palpation: 5 s contraction, 10 s relaxation, 10×, 3×/day.
HRT: estradiol 2 mg + norethisterone 1 mg/day.
scale 0 to 5 where 0 = no contraction, 1 = minimal lasting < 1 s, 2 = weak contractions lasting 1–3 s, 3 = fingers of therapist elevated, contraction lasts 4–6 s, and can be repeated 3×, 4 = as 3 but contraction lasts 7–9 s, 5 = as 4 but lasts ≥ 9 s and pt able to repeat 4×.
*Scale of 1–4, 1 = once/day, 2 > 1/week, 3 < 1/week, 4 = 1/month.
(All parameters improved from baseline in both PFMT grps, and vs control grp P < 0.001. No sig. changes in control grp).
Perineometry (mean change, cmH2O)+30.9 vs +17.2 vs +1.3 (162 vs 85% vs 7%)
(actual values 50 ± 11.5 PFMT+BF vs 37.5 ± 8.7 PFMT + palpation, P < 0.001)
PFM strength digital palpation+1.6 vs +1.3 vs 0
Incontinence frequency* (change in mean score)+1.3 vs +1.2 vs +0.3
Wong 2001267RCT
EL = 1+
38Women, 30–62 years, urodynamic stress UI
Exclusions: 2nd/3rd degree uterine prolapse, previous failure of PFMT, previous surgery for UI, neurological pathology
PFMT + biofeedback (PF and abdominal muscles)
(n = 19)
PFMT +biofeedback (PF muscles)
(n = 19)
4 weeksLeakage episodes/week−2.0 vs −5.0, (57% vs 55%) P = NSFunding: none declared.
PFMT: 4× 30 min sessions 2×/week, fast and slow contractions.
Biofeedback: PRS9300 device. For PFMT + biofeedback (PF and abdominal muscles) group, 1 vaginal probe and 1 attached to abdominal wall; PF muscles group had vaginal probe only).
Pad test (mean change, g)−8.6 vs +13.9, (69% vs 153%) P = NS
PFM strength (mean change, cmH2O)+5.4 vs +8.8, (47% vs 68%) P = NS
PFM endurance (mean change, s)+1.3 vs +0.9, (26% vs 16%) P = NS
IIQ-7−4.8 vs −14.3, (25% vs 50%) P ≤ 0.04
UDI-6−8.0 vs −33.3, (22% vs 67%) P ≤ 0.04

From: Evidence tables for included studies

Cover of Urinary Incontinence
Urinary Incontinence: The Management of Urinary Incontinence in Women.
NICE Clinical Guidelines, No. 40.
National Collaborating Centre for Women's and Children's Health (UK).
London: RCOG Press; 2006 Oct.
Copyright © 2006, National Collaborating Centre for Women’s and Children’s Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK [www.cla.co.uk]. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.