effectiveness bulletin - qualitysafety.bmj.com · madsen-iversen symptom scores" had fallen by...

Quality in Health Care 1996;5:111-119

Effectiveness Bulletin

Benign prostatic hyperplasia

Arabella Melville, Jenny Donovan, Trevor Sheldon, Tim Peters

This paper reviews the evidence of effective-ness of treatment options for benign prostatichyperplasia, taken to mean the clinical problemof "men with lower urinary tract symptomssuggestive of bladder outlet obstruction."' It isbased on Effective Health Care, volume 2,number 2, which is a synthesis of informationderived from a systematic review of theliterature.2 Medline and Embase (1970-95)were searched, as well as bibliographies ofreview papers, and a range of experts wereconsulted for any references that might havebeen overlooked.

NHS Centre forReviews andDissemination,University ofYorkArabella MelvilleTrevor SheldonDepartment of SocialMedicine, Universityof BristolJenny DonovanTim PetersCorrespondence to:Arabella Melville,NHS Centre for Reviews andDissemination, University ofYork, York YO 1 5DD

Accepted for publication6 March 1996

BackgroundThe prostate is a wedge shaped gland sur-

rounding the urethra as it emerges from thebladder. Strictly speaking, benign prostatichyperplasia is a precise histological term whichdescribes age related growth of the prostate. Itis, however, more typically used as an all-encompassing term covering a wide range ofurinary symptoms, sometimes also referred toas "prostatism." Benign prostatic hyperplasiacan cause benign enlargement of the prostateor obstruction of bladder outlet.3The benign growth of the prostate gland can

result in acute or chronic retention of urine anda wide range of urinary symptoms. Patientswith acute retention require immediate reliefby means of a catheter; this is often followedby surgical treatment. Chronic urinary reten-tion, if left to develop, may cause damage to

the urinary tract and kidneys.4Changes in the prostate suggestive of benign

prostatic hyperplasia have been found in up to

70% of 60 to 70 year old men5 but estimatesof the prevalence of clinical benign prostatichyperplasia in this age group range from 5% to

43%, depending largely on the way the prob-lem is defined.6`9 Symptoms include hesitancy,dribbling, slow, erratic, and frequent urination,inability to empty the bladder completely,frequent night time urination, urgency, andurge incontinence. Benign prostatic hyper-plasia may be associated with recurrent urinaryinfection, bleeding, and bladder stones,although it has not been clearly established thatit causes these.'0

Untreated benign prostatic hyperplasiashows a variable and unpredictable pattern ofexacerbation and remission.'0 In one studyof 107 men followed up after five years, 15%thought that their symptoms had worsened,90/o underwent surgery, and 29% improvedwithout treatment. The proportion of un-

treated patients who became free from urinary

symptoms was similar for those with andwithout acute retention when first seen.'2 Inanother study, 123 patients were followed upfor four to seven years; 10% developed acuteretention but 48% had no urinary symptoms attheir final follow up. 13

Indications for treatmentThe most important indication for treatment isseverity of symptoms, which may be measuredby questionnaire.'4-l' Many urologists considerevidence of bladder outlet obstruction to be acrucial indication for surgery.'8 Most studieshave failed to show any relation between uro-dynamic measures and severity of symptoms oroutcome of treatment,'9-22 although one recentreport, based on a small study, suggests thatpatients with obstruction, as indicated bypressure-flow studies, show greater improve-ment after laser prostatectomy than those withno evidence of obstruction.23No specific level of symptoms can be taken

to be an absolute indication for any particulartreatment for benign prostatic hyperplasia onclinical grounds.24 Because of the complexnature of the problem, the balance of potentialrisks and benefits of possible treatments, andthe variable individual experience of benignprostatic hyperplasia, it is increasingly ac-cepted that patients should participate indecisions about treatment.25-27 However, theinformation received from urologists oftendoes not correspond either with publishedevidence or with patients' experience.28A video based shared decision making pro-

gramme has been developed in the UnitedStates29 and piloted in Britain.30 Use of thevideo has reduced surgery rates in some UnitedStates health maintenance organisations by upto 500/o.31 However, because the symptoms ofpatients who undergo surgery in Britain tendto be more severe,32 it is unlikely that surgeryrates would fall as much if this programmewere introduced in the United Kingdom.

Effectiveness oftreatment optionsWATCHFUL WAITING AND LIFESTYLE ADVICE

Many men are willing to tolerate symptomswhen fully informed about the natural historyof benign prostatic hyperplasia and the risksand benefits of treatment.30 33 A randomisedcontrolled trial in the United States comparedwatchful waiting with prostatectomy for menwith moderate symptoms.'9 Of the 276 patients17% assigned to watchful waiting were classi-fied as treatment failures (a failure rate of 6/100man-years) and 24%, including a third of the

ill

on 8 April 2019 by guest. P

rotected by copyright.http://qualitysafety.bm

j.com/

Qual H

ealth Care: first published as 10.1136/qshc.5.2.111 on 1 June 1996. D

ownloaded from

http://qualitysafety.bmj.com/

11lelville, DIlonozvan, Sheldon, Peters

"failures," underwent surgery during the studyperiod. After three years, men who had notundergone surgery reported more bother fromurinary difficulties and more interference withactivities of daily living. However, their meanMadsen-Iversen symptom scores" had fallenby 5 5 points, from a baseline value of 1406 to9-1. There were no significant differences be-tween the groups in mortality, general well-being, sexual performance, or social activities.

Drug treatmentCU-BL OC KERSa-Blocking drugs relax muscles in the bladderneck and prostate.34 Alfuzosin, indoramin,prazosin, and terazosin are licensed for treat-ment of benign prostatic hyperplasia in theUnited Kingdom.3' Table 1 shows the resultsof randomised controlled trials comparingthese drugs with placebo.3" " Studies with lessthan a total of 60 patients have been excludedbecause their results are less likely to bereliable. Most of these trials are meth-odologically flawed, but some, notably thestudies of terazosin by Brawer et al 4 and Leporet al,4' showed convincing, if modest, benefitsin a proportion of men.

All these drugs seem to produce rapid butsmall improvements in symptom scores (2 to3 points on the Boyarsky scale)1' '' andobjective measures, which may be maintainedfor at least three years in men who continue totake the drug.4' However, this open label trialmay have overestimated long term effects,because those who benefit from the drug aremore likely to continue taking it.

FINAS ITRIDEFinasteride, a 5-cx reductase inhibitor, shrinksthe prostate by reducing the level of dihydro-testosterone, a hormone implicated in prostaticenlargement.46 There have been three largescale multicentre trials of finasteride4 "' and aplacebo controlled dose-response study(table 2).5" In the two year Scandinavian studyof 707 patients with moderate symptoms ofbenign prostatic hyperplasia, 5 mg of finas-teride daily led to significant reductions insymptoms. This effect took some months todevelop, but it was sustained over the periodof the study." At baseline, both drug andplacebo groups scored around 13 on theMadsen-Iversen scale; after two years, themean score for the finasteride treated group

Table I Randomised placebo controlled trials of (x blockers used to treat benign prostatic lixperplasia (zvit/ more than 60 patients.)

1Trial Drug Durationl Patients Aauin oiutcomiit A. 1lain us (Commenwsimeiasuost

Jardin Alfuzosin 6 Months Alfuzosin1991 t) 7 5 mg;

after 14days, someincreasedto 10 mg"dependingonresponse

Chose, Indoramin1990(o 20 mg at

night(n =43) or

20mgtwice dailv(n = 38)

8 Weeks

Sertcelik Prazosin 4 Weeks1990"8 2 mg twice

daily after 2

weeks onlosser dose

Steven Prazosinmetal 2mg twice1993 "' daily; after

8 days ofgraduallyincreasingdose

12 Weekstreatment, 12weeksplacebo, in

randomlyallocatedorder

Chapple Prazosin 12 Weekset al 2 mg twice1992'" dailv after 8

weeks davsof graduallyincreasingdose

n = 251, placebon = 267; men age41-86, Boyarskiscore > 6, nosevereconcomitantillness or drugtreatment likelyto affect results

Indoramin 20 mgn = 39, 40 mgn = 37, placebon = 34. Men age46-84, with nomajor concurrentdisease

Drug n = 40placebo n = 40;men, 44-76, whohad not hadprostatic surgeryand did not takedrugs likely toaffect results

82 GP Patients,age > 50, withmoderatesymptoms; eachacted at his ownplacebo controlin crossoverdesign

93 Men at 2hospital centres,with severesymptomsawaiting surgery.At end of study,prazosin n 34,placebo n 41

Peak and meanflow! rates,residual urine(n= 189),Boyarski score,effects on specificsymptoms,overalltherapeutic effectas judged bfiinvestigator andpatient

Peak urinaryflows, voidingtime, voidingvolume, reportedsymptomaticimprovement

Prrostatic weight,peak flow rate,residual urine,diaries of diurnaland nocturnalfrequency,urgency, and,obstructivesymptoms'

Symptom scoresat baseline, 4, 8and 12 weeks.Analogue ratingscale for patients'overallassessment ofsymptoms, from"no problems"(0) to "totalblockage" (100)

Diary card recordof urination,symptomevaluation byclinicians, flowrates, voidingpressure, voidedand residualvolumes

Increase in peak flow not significant comparedwith placebo; difference in mean floss ratesignificant (treatment group 0)9 ml, placebo0 6 ml). Residual urine in treatment groupdecreased from 80 ml to 49 ml, placebo 88 ml to80 ml (P = 0-0 17). Overall therapeutic effectjudged good by investigators for 61"'., of patients.by patients for 50°/'; 42'0X of investigators and40'/o of patients judged placebo good. Significanttreatment benefit (P < 00(5) for nocturia, daytime frequency, hesitancy, urgency, "qualitv ofstream"

Symptom improvement reported in 78°/, ofpatients in 40 mg group, 64'/( in 20 mg group,530// in placebo group. Peak flow rates increased4 90 ml (CI 2 56 to 7 22) in 40 mg group,2 79 ml. (0 53 to 5 05) in 20 mg group, I-75 nml(CI 0-67 to 4 17) in placebo group

Flow rate increase 5 ml in treatment group, 2 mlin placebo group (P < 0001 ). 25'/, Reduction indaily frequency in treatment group, 7 5'/o inplacebo group. Treatment judged effective for95'/,, of patients in drug group, 40'o in placebogroup. Residual urine decreased 30 ml in placebogroup, 1 5 ml in treatment group

Symptom scores: reduction in treatment groupcompared with placebo at 12 seeks -0 8 points(CI -22 to 0-6, NS). Benefits of treatment,compared with placebo on analogue scale at 12weeks: -5 4 units (CI -0 7 to 101). Adverseeffects: 205 reported by 75 patients in prazosinphase, 152 by 64 patients in placebo phase; noneserious or long lasting

No significant difference between groups infrequency of urination, nocturia, urgency,hesitancy, voided volume, residual volume.Investigators' overall evaluation of efficacy: 56'..of prazosin group improved . 21% of placebo;P = 0 02. Drug rated tolerable (risks notsignificant) by 68% in prazosin group, 55'/o inplacebo group; P = 0 05. Voiding pressurechange: significant difference (P = 0 02). Flosrate change: prazosin 1 ml/s, placebo - 0 ml/s(P= 0(01)

Investigators not blind and31' sswithdrawal rate, sojudgements of therapeuticbenefit probably exaggerated,although objective measuresless likely to be affected.Overall incidence of adverseeffects similar in both groups

( omparisons svere betweenbaseline and 8 week figuressNithin treatment groups, notbetsveen treatment andplacebo. Groups differed atoutset, so benefit not reliablyattributable to treatment

Investigators not blind, sobenefits probably exaggerated.Baseline voiding frequency intreatment group higher thanplacebo, so impros ement masnot be due to treatment.Effectiveness, frequency, andflows rate benefits inconsistentvith data on residual urine2 1"'/, Dropout rate, nointention to treat analysis

Ihis ssras a double blindcrossover trial, therefore likelyto produce relatively reliableresults. Benefits not dramatic,more notable for obstructivethan irritative symptoms.Placebo effects much smallerthan in most benign prostatichyperplasia studies

This double blind trial showssome benefit linked svithtreatment, but 19'svithdrasval rate combinedssith lack of intention to treatanalysis reduces reliability ofresults

112



j.com/

Qual H


ownloaded from



Table 1 - continued

Tial Drug Duration Patients Main outcome Main results Commentsmeasures

Brawer Terazosin, 24 Weeks Terazosin n = 81, Boyarskietal 1-10mg,19934' titrated to

patient'sresponseafter 4weekplacebolead in

Lepor Terazosinet al 2, 5, or199242 10 mg daily

Di TerazosinSilverio 2, 5, or199243 10mg

dosageincreasedover 4weeks

Lloyd Terazosin,etal 2,5,or199244 10mg,

dosageincreasedover 4weeks

12 Weeks

4 Weeksplacebo leadin, 4 weeksincreasingdoses forhigher dosegroups, 4weeks with alldose levels

4 Weeksplacebo leadin, 4 weekswithincreasingdoses forhigher dosegroups, 4weeks with alldose levels

placebo n = 79.Men age > 45,scoring point ormore on at least2 items ofBoyarskisymptom scale,no "absoluteindications forprostatectomy",detrusorinstability, orsignificantcardiopulmonarydisease

Terazosin 2 mg;n = 74; 5 mgn=72; 10mgn = 70; placebon = 69. Men age44-77, not takingdrugs likely tointerfere withstudy, no seriousheart or kidneydisease ordiabetes, norecent urinarytract infection

Terazosin 2 mgn = 34, 5 mgn=36, 10mgn = 32, placebon = 35.Normotensivemen, not takingantihypertensivedrugs

Terazosin 2 mgn= 19, 5mgn= 19, 10mgn = 22, placebon = 20.Normotensivemen, mean age65-7, free fromurinary and renalproblems otherthan benignprostatichyperplasia, nottaking drugslikely to affectaction ofterazosin

symptom scores(scale max 27),"globalassessmentscores" assignedby interviewers,urine flow

Boyarskisymptom score,urinary flowrates, symptomdiary, patientself-assessment

Decrease in symptom score for drug group(baseline mean 11) significantly greater thanplacebo from first assessment at week 2.Beneficial effect maintained, increasing to week20. Treatment led to fall in score 3-4 pointsgreater than placebo (CI -4-6 to -2-2). Peak flowrate increase 1-4 ml greater than placebo(P < 0-05). Investigator's global assessmentbetter for drug group (P < 0-05). 12 Men in druggroup and 7 in placebo group dropped outbecause of adverse events (mainly dizziness).Blood pressure fell only in patients withuntreated hypertension in drug group. Lowerrate of urinary tract infection in drug group

Improvement in total symptom score (baselinemean, 10), compared with placebo; 2 mg: -1-0(P = 0 097); 5 mg, -1-3 (P = 0 042), 10 mg, -2-3(P < 0 001). Effects apparent and maintainedfrom week 4. Peak flow rates, change comparedwith placebo: 2 mg 1 1 ml/s, 5 mg 0-6 ml/s,10 mg 1 9 ml/s: significant for 10 mg group only(P = 0-009). Changes in voided volume notsignificant

Flow rates, Peak flow rates increased 34% in terazosinresidual urine, groups, 17% in placebo group, but difference notsymptom scores significant. Mean flow rate increase for 10 mg

group significantly greater than placebo. Nosignificant difference between groups in volumeof residual urine or symptom scores

Symptom scores,flow rates, voidedvolume andresidual urine,symptom diaries

No significant changes in any group whencompared with placebo. Peak flow rate change interazosin 2 mg group 1-3 ml/s, 5 mg 2 1 mg/s,10 mg 2-8 ml/s, placebo 2-5 ml/s. Other changessimilarly unimpressive. Greatest apparent drugeffect on obstructive symptoms: -3-8 (CI -5-6 to-2-0) in 2 mg group, -3 4 (CI -5 0 to 1-8) in5 mg group, -3-9 (CI -5 3 to -2 5) in 10 mggroup, -1-7 (CI -3 3 to -0 1) in placebo group

This randomised double blindtrial shows definite benefit ofterazosin treatment. Adversereactions to terazosin make itunsuitable for some patients,but response (beneficial oradverse) is likely to becomeapparent within 2 weeks ofstarting treatment

This multicentre double blindtrial with intention to treatanalysis shows dosedependent beneficial effects ofterazosin on symptoms andflow rates. Adverse effectswere not significantly differentfrom placebo

Significant difference betweenresults obtained by differentinvestigators (P = 0-01) butlittle evidence of drug effect:few other comparisonssignificant

Small numbers mean effectsdo not reach significance

Table 2 Randomised placebo controlled trials offinasterideTria Dose Duration Patients Main outcome Main results Comments

measures

Andersen 5 mg Daily 2 Years Finasteride Symptom score Improvement in symptom scores in finasteride This is the first double blindet al n = 347, placebo (Madsen- group significantly greater than placebo from 8 placebo controlled trial199547 n = 346 Men Iversen) flow months. 2 Point improvement at 24 months evaluating effects of any

< 80 y, moderate rates, prostate (baseline 13), with divergence between groups medical treatment for benignsymptoms, not volume tending to increase. Placebo group symptoms prostatic hyperplasia over 2more than 2 returned to baseline at 20 months after initial years. The results suggest thatsevere symptoms, improvement. Change from month 12 to 24 finasteride treatment has aenlarged significant. Maximum flow rates improved in small but sustained beneficialprostate, flow finasteride group to month 8, improvements effect on symptoms and israte 5-15 ml/s, sustained thereafter; placebo returned to acceptable to most patientsno significant baseline. Prostate volume in finasteride groupabnormalities or decreased by 20% and decrease maintained; inserious pathology placebo group, prostate volume increased 1%

by end of year 2. Similar numbers (18%)withdrew from each group. 9% Higher incidenceof sexual dysfunction in finasteride group

Gormley 1 mg and 12 Months Finasteride 5 mg Flow rates, After 12 months, total symptom scores had fallen Large numbers in studyet al 5 mg Daily after 2 week n = 297, 1 mg symptom scores 1 point (2%) in placebo group, 2 points (9%) in allowed relatively small199248 placebo n = 298, placebo on 45 point scale 1 mg finasteride group, 3 points (21%) in 5 mg changes to be significant.

lead-in n = 300. Men (modified group. Significant difference (P < 0 05) between Drug effects developed40-83 y, Boyarski), 5 mg and placebo. Urinary flow increased gradually over 12 months.symptoms of prostatic volume 0-2 ml/s in placebo group, 1-4 ml/s in 1 mg 12% withdrawal rate, similarobstruction, finasteride group, 1-6 ml/s in 5 mg group. in all 3 groupsenlarged Prostatic volume fell 1-2 ml in placebo group,prostate, peak 11 8 ml in 1 mg group, 11 1 ml in 5 mg groupflow < 15 ml/s (P < 0 01 for comparisons between drug groups

and placebo). Frequency of adverse effectssimilar in all groups, but higher incidence ofdepressed libido, impotence, and ejaculatoryproblems with finasteride treatment

113



j.com/

Qual H


ownloaded from


1Ielv/ille, Donovan, Sheldon, Peters

lab/c 2 - cootinuccl

7)I,11al Ds.D DItrationl PaitiIts A ill AlitC(ili1 in[t cilltS(Lou(e/ 17 tsIl IS1ilC(Us

Finasteride mg and 12 MonthsStudy 5 mg DailyGroup1993"'

Stoner Studs 1: 5,19929 10, 20, 40,

and 80 mg;Studs 2:0 2, 0 5, 1,5, and40 mg

Studs 1: 12weeks, plus 12drug freeweeks; studs2: 24 sleeks

Finasteride 5 mgn = 246, 1 mgn = 249, placebon =255. Men40-80 v, goodphysical andmental health,ss mptoms ofobstruction,prostate> 30 cm', peakfloss < 15 ml/S

Studs 1: placebon= l4,5mgn =10, 10mgn = 15, 20 mgn= 16,40mgn =14, 80mgn .1 Studv 2:placebo n = 25,0-2mgn= 18,0) mgn= 15l mgn = 17,5mgn= 15,4l0mgn= 13.Men 40--80 ,ss mptoms ofobstruction,prostate> 30 cm' noclinicalabnormality

Flow rates,symptom scoreson 45 point scale(modifiedBovarski),troublesomenessratings (22 pointscale), prostaticvolume

Study 1: prostatevolume. Studs 2:symptom scores,peak floss rates,hormone profile.Both studies:adverse events

Symptom scores in finasteride 5 mg group fellfrom 18 6 to 15 3; in placebo group from 18 2to 16-2; difference significant after 8 months.Finasteride 1 mg group symptom scores notsignificantly different from placebo. Urologicalstatus, as assessed bs investigators, xx assignificantly better in both finasteride groupsthan placebo at 6-12 months (P < 00 15).4Troublesomeness of symptoms significantlyreduced in 5 mg group only from 8 months.Floss rates: both finasteride groups significantlyimproved, compared With placebo, at 12months (P < 0)025). Prostate volumedecreased 23-6" a, in 1 mg group, 22-4" O in -) mggroup, 5" placebo at 12 months (P < (3 1).Incidence of impotence: finasteride 5 mg,4 ()`,, 1 mg, 4 0"/,, placebo 04",; P < 0 005).No other adverse experiences variedsignificantly betsseen groups

Studs 1 result: significant decrease in meanprostate volume in 1, 5, and 40 mg groups z

placebo. Study 2: Significant increase in floassrates in treatment groups v placebo, from xs eek7. Mean difference between pooled higher dosegroups and placebo 3-7 ml/s (P = 0 03).Symptom scores: no significant differencebetsveen 0(2 and 0 5 mg treatment groups andplacebo, significant improvement in 1 ands mg groups (P < 005). Significant decrease indihNdrotestosterone in all treatment groups;hormone concentrations and prostate volumereturned to baseline at end of drug free period.Adsverse events: 1 3 "serious' on finasteride, 1on placebo, possibly drug related

Study design and results verysimilar to Gormlev et al. Fallin symptom score infinasteride 5 mg group of 0()points more than placebo on45 point scale after 1 X(P < 0 005) although sertsmall compared with scoredifferences achieved bssurger-

These studies suggest thatfinasteride is moderatelyeffective in reliesingsymptoms of benign prostatichyperplasia at doses of mgand above. The drug seems toreduce prostate volume whbiletreatment continues, throughreversible reduction ofdihs drotestosterone

had fallen to 11, whereas the placebo groupmean score had returned to baseline after aninitial improvement. Other measures suggestedgradual deterioration in the placebo group overthe course of the study, in contrast to a smallbut consistent improvement in the finasteridegroup. The only adverse effect attributable tofinasteride was sexual dysfunction, whichaffected 19'/o of patients in the treatment groupand 100/, in the placebo group.Both the other major trials of finasteride45 -"

lasted one year, with three year open label (non-blinded) extensions. These studies showedimprovements of 1 to 2 points on a modifiedBoyarsky scale compared with placebo, whichis consistent with the Scandinavian data. Abouthalf the men in these studies continued to usefinasteride after the first, double blind year, andthe benefits, although slight, were maintained.)1However, this non-blinded comparison mayoverestimate long term benefit.

NA I URAI REMEDIES AND) OTHER AGFNTISAlthough several natural remedies are used totreat benign prostatic hyperplasia, there is noconvincing evidence of efficacy for them.'However, in double blind randomised controlledtrials, both Cernilton, a pollen extract, andCubicin, derived from pumpkin seeds, producedsignificant reductions in subjective symptomsand residual urine.)3 ' Neither product wasreported to cause any side effects. These trialsinvolved small numbers of patients and shouldbe repeated on a larger scale.

SurgeryThere are two types of surgical treatment forbenign prostatic hyperplasia. The first, prosta-

tectomy, involves removal (resection) or de-struction of the inner tissue of the gland, usuallythrough the urethra. When this operation iscarried out by electrocautery, it is known astransurethral resection of the prostate, but it canalso be carried out with heat generated by laseror microwaves. When access through the urethrais difficult or inappropriate - for example whenthe gland is too large to resect- the operationis carried out through an incision in theabdomen; this is open prostatectomy.The second type of operation - transurethral

incision of the prostate or bladder neck incision- involves one or two incisions in the prostateto relieve constriction. This is considered un-suitable for men with particularly large glands.

SURGE RN' RATIESAbout 450/1 of patients with diagnosed benignprostatic hyperplasia in the United Kingdomare treated surgically (David Neal, Universityof Newcastle, personal communication). Ofthese, in 1992, 90 90 underwent resection,1.6% open prostatectomy, 5.71)0 incision, and0-2% laser or microwave prostatectomy.)About 55 000 prostatectomies are carried

out by the National Health Service (NHS) inEngland each year,' and 9400 in privatehospitals (Jon Nicol, University of Sheffield,personal communication). The age standard-ised prostatectomy rate in England and Wales,at 9 per 10 000 per year, is similar to Norwayand Australia, but is only 300/0 of the rate inthe United States. ' There are large variationsin the age adjusted rates for resection betweenEnglish districts, ranging from 1 5 to 26 7 per10 000 per year,"' but it is not clear what ratewould be appropriate.

1 14



j.com/

Qual H


ownloaded from



TRANSURETHRAL RESECTION OF THE PROSTATE

The effectiveness of resection was comparedwith watchful waiting in a United Statesmulticentre randomised controlled trial.'9 Of280 men assigned to surgery, 249 underwentresection and were followed up for a mean of2-8 years. At baseline, the mean Madsen-Iversen symptom score in this group was 14-6;three years after surgery, this score had fallenby 9-6 points.Both the randomised controlled trial'9 and

the United Kingdom national prostatectomyaudit55 found that men with severe symptomsare most likely to benefit. The UnitedKingdom audit showed that overall, perhaps a

quarter ofmen fail to improve symptomaticallyafter resection.58 These results are less positivethan the 88% improvement rate derived froma synthesis of United States case series by theagency for health care policy and research.22These findings may be related to the selectionof studies or regression to the mean caused byrandom variation in symptoms over time.The mean reported probability of surgical

complications calculated by the agency forhealth care policy and research is 15% (90%CI 2.15% to 33.30/o).22 Long term adverseeffects of resection in some men include impo-tence, incontinence, and urethral strictures,although the multicentre randomised con-

trolled trial already described did not findincreases in either impotence or inconti-nence.'9 The national prostatectomy audit59 ofroutine care in the United Kingdom found that6-6% of men free from incontinence beforeprostatectomy developed incontinence severe

enough to cause a problem, and a further 6-3%with slight non-problematic incontinenceinitially had severe incontinence afterwards.Twenty one per cent of men had increaseddifficulties with erection after surgery, but for8% erection problems were reduced. In 73.4%of patients resection caused dry, or retrograde,ejaculation.22The mortality for patients Undergoing

resection for benign prostatic hyperplasia inhospitals in the Northern Region of Englandwas 0.3% at 30 days and 1-7% at 90 days.60Ninety day mortality after resection in theUnited States is similar to the British rate, at1-5% (90% CI 0 5 to 3 3%).22 There are

notable variations between sites, withsignificantly fewer deaths and complications inhospitals where more than 100 patients were

treated over the eight month period of theaudit. However, because the risk of death ishighly dependent on age and severity of comor-bidity,6' these variations are difficult tointerpret.

Reoperation rates increase progressively withtime.62-64 The United States trial'9 reportedthat 100% required further surgery within threeyears, and a retrospective study of outcomesfor over 50 000 patients found that between89% and 9-7% undergo a second resectionwithin five years.62

OPEN PROSTATECTOMY

Open prostatectomy is used for particularlylarge glands (80 g or over) and for patients with

hip problems which prevent correct positioningfor resection.65 67 Observational studies reportthat it leads to slightly higher rates of improve-ment than resection and that reoperation ratesare consistently lower.62 64 However, compli-cations are more common. One randomisedcontrolled trial reported considerably higherlevels of infections and blood transfusions afteropen prostatectomy than after resection.68 TheUnited States estimate22 for the median prob-ability of surgical complications is 21% (90%CI 700% to 42-7%); 0 5% of men are reportedto become totally incontinent after surgery,and reported rates of impotence vary from32 3% for perineal prostatectomy to 16-2% forretropubic prostatectomy. Retrograde ejacu-lation affects around 70% of patients.69The 90 day mortality in the United States

was 2-4% (90% CI 1-0% to 4-6%).22 A largeretrospective study found that open prostat-ectomy was associated with lower long termmortality than resection,62 but subsequentobservational studies suggested that differ-ences in the severity of comorbidity mayaccount for the apparent superiority of openprostatectomy.6'

TRANSURETHRAL INCISION OF THE PROSTATE

Transurethral incision is only consideredsuitable for men with glands of 30 g, or less,22but this includes at least half of those whocurrently undergo resection.55 Randomised con-trolled trials comparing incision with resec-tion'112 suggest that the two procedures can beequally effective, showing a 12 point reductionin the Madsen-Iversen symptom score (table 3).The United Kingdom national prostatectomyaudit figures also show no significant differencesbetween incision and resection in postoperativeseverity of symptoms.59 Retreatment rates afterincision seem to be similar to those forresection.94 The incision is a quicker operationwhich causes less tissue damage than otherforms of prostate surgery, so adverse effects areless common and less serious.70 The 90 daymortality has been estimated at 0.7% (90% CI0 2% to 1.50/o).22

LASER TREATMENT

Laser treatment is growing rapidly in popu-larity among urologists. There are two maintechniques: transurethral ultrasound guidedlaser prostatectomy and visual laser pros-tatectomy. Tissue sloughs off gradually aftertreatment, leading to improvement after aboutsix weeks.73Randomised controlled trials comparing

laser prostatectomy with resection suggest thatlaser treatment reduces symptoms by about 10units on the 35 point American UrologicalAssociation scale,'4 compared with 13 units forresection (table 4).7476 Laser treatment seemsto be slightly less effective than resection, butthe difference in clinical outcome may bebalanced by its safety advantages. Laserprostatectomy causes much less bleeding andfewer perioperative problems; in one study, forexample, complications affected patients whohad a resection three times as often as patientstreated by laser.75 Reported adverse effects of

115



j.com/

Qual H


ownloaded from


1A6elville, Donovca, Sheldon, Peters

Table 3 Raodomoised controlled trials comiparinig transiorethral resectiOn (T7URP) zith traosurethral incision (TUIP)

I')i7ll 1'atielits Jo/h1zr lip AaoiM OIitoiLa i)'xM1alr.d C oiiliillisMwasllues

Riehmann TURP Mean, 34 Sxmptom scores Symptom scores: baseline 16, both groups; t'alling to lTIP is as effective as TURP.ot/l 1995 n 56, TUIP months; (Madsen- minimum at 1 year, TURP 4, TUIP 45. Thereafter, and its equivalence in terms of

n 61; diminishing Iversen); peak gradual increase to 10, years 4-6. No differences between effectiveness is maintained oserprostates numbers w ith floss rates; blood treatment groups significant. Peak flow rate (Q ) the 6 year period of this studs.< 20 gm, no increasing time, loss; duration of baseline, TURP 9u7, TUIP 7 8; thereafter, to 6 years, Although peak floss rates aresuspicion of but 17 patients surgery, TURP about 19, TUIP 15. Subjectise assessment of slightly higher after TURP, thismalignancy, followed up for catheterisation outcome: satisfaction rate declined to 60°/,", both groups, at is not clinically significant, asno previous 6 s and 3 years, then rose to 800/o in TUIP group at 6 s. Additional floss after TUIP remainsprostatic hospitalisation; treatment for obstruction: 16"',, in TURP group, 233" in relatively high for this agesurgery sexual 'IUIP group; NS. Operating time: TURP 50 min, TUIP group. Overall, TUI1' is clearly

functioning 20 min. Blood loss: TURP 150 ml, TUIP 25 ml. better than TURP for thisHospitalisation: TURP 4 days, TUIP 3 das s; catheter group of patients, with equaltime: TURP 2 days, TUIP I das. 1 Death due to surgery benefit but loser levels ofin TURP group, none in TUIP group. Retrograde adverse effects and costejaculation: 68"/O after 'URP, 35" after TUIP (P = 01)2)

Soonaxvalla I'URP All patients 3 Patient Satisfaction (excellent or fair): 3 months and 2 sears: 2 Year data for satisfaction andetf / 1992 n = 1 10, months; 1 t; satisfaction; floss TURP, 9Vt0, TUIP, %95"50O. Peak flows rates: 'IURP, residual urine unreliable: no

TUIP n = 137; 2 s, rates; residual baseline 8 0, 3 months 20 1, 2 years 19 9; TUIP, baseline change in figures xxith fallingn = I 10; n = 47 urine; blood 7 9, 3 months 19 4, 2 sears 18-9. "High residue": TURP, patient numbers. Nevertheless,men, transfusion; baseline 73" o 3 months 692"); TUIP, baseline 74"/,o, 3 data suggest that TURP and45-87 s duration of months 7 3(0V. Transfusion: 38(35" ,,) in TURP group, TUIP are equally effective, andprostates surgery and none in TUIP group. Mean duration of surgery: TURP 59 that their equivalence is< 30 gm, no hospitalisation min, TUIP 20 min. Mean hospitalisation: TURP 7 1 dass, maintained. No late recurrencesuspicion of TUIP 6 0 days. More perioperative complications after of obstruction, but 3 in TUIPmalignancy TURP, including 6 4 o,'ICUR syndrome group later undersvent TURP,

and 2 in TURP group hadrepeat resection

Nielsen 1 URP s Successful or At 1 sear, 18 of 23 TURPs judged successful, 18 of'22 88' $O Of patients in lUIP1988' n -5, TUI IP unsuccessful TUIPs. 3 Patients in TUIP group had postoperatis e group had estimated prostate

n 24; men (incontinence or retention and underwent TURP; all had prostates > 30 g. sveight > 30 g, usually> 60 , increased Another, also w ith prostate > 30 g, xw as judged considered too large for thisconsecutive e frequency of unsatisfactory at 1 sear. Flos rates after TURI' operation. Poor measures of'patients svith urination), significantly higher than after TUIP at 2 months, patient satisfaction makeobstruction urinars culture, difference not significant at I sear (TURP, 12 ml/s; TUIP results imprecise. TUIP seemsdue to benign flox rates 9 ml/s). No difference between groups in positive urinars safer and quicker than TURP,prostatic culture at 1 year. 4 In TURP group developed strictures, 1 possible slightly less effectis e inhyperplasia incontinent. Significantly less bleeding and shorter terms of floss rates49"/, in acute operation times in TUIP groupretention

Table 4 Raodonn'sed controlled trials eomoparoiJg trao1siret/iral resectioui (TURP) zvith laser prostatectoov

Prial (oulipirl'sou) Patiiltx Fo/l/o,- Muilllutcol)1u Mlcil esult/ (uoulul.ieultlip MCCI'tS111etS

AnsonandWatson1995''

TURP v

endoscopiclaser ablationof the prostate(ELAP)

Coxnvles TURP vvisualet al/ laser ablation1995%5 of the prostate

(\VLAP)

Schulze TURP v

et a/ transurethral1994" ultrasound

guided laserinducedprostatectomy( 1 UI IP)

TURP n = 75;laser n = 76. Age> 50s, nosuspicion ofprostate cancer,no anticoagulantmedication

TUR1' n = (,9,VLAP n = 56,men > 50 s, notin retention, nomedicalcondition thatshould affectsuitability foreither procedure

40 Men, < 75 a,prostate Xolume2 7r ml,symptom scores(Boyarski) > 15,peak floss rate(Q_.j < 15 ml/s.13 of each groupin acute retentionbefore surgery

I v AmericanUrologicalAssociationSx mptomscores (35point scale),peak floss rate(Q ,111)) residualurine, adverseeffects

I ! AmericanUrologicalAssociationsymptomscores, peakfloss rate,residual urine,quality of life

6 BovarskiMonths symptom score

peak flosv rate,residual urine,blood loss,hospital stash

Symptom score fell from 182 to 5 1 (CI 3.8 to 6 4) inTURP group, 18 1 to 7 7 (CI 6 3 to 9Q 1) in lasergroup; (Q .,,) rose from 10 to 21 8 ml/s (CI 18s5 to25 1) in TURP group, 9D5 to 154 (CI 136 to 17 2)in laser group; residual urine fell from 62 to 46 ml inTURP group, 70 to 69 ml in laser group. 5 Treatmentfailures in laser group. Dysuria: 15 {, in TURP groupat 4 xs weeks, 41" ,, in laser group; at 3 months, 1" O inTURP group, 1 5'S., in laser group. Catheterization:297 days after TURP, 1292 days after laser.Significantly more infections after laser treatment.16" 0 Transfusions after TURI, none after laser. 1Death in each group

Fall in symptom scores from baseline at 1 sear: I URP13 3, laser 9 0 (P < 0 04). Increase in peak flosw:TURP 9-5, laser 6 9 ml/s (P = 0127). Fall in residualurine: TURP 139 ml, laser 55 ml (P < 0(01). Qualityof life: 93",, patients improved after TURP, 78",, afterlaser. Serious complications: TURP 2 1, laser 6(P < 0 01). Non-serious complications: TURP 17,laser29 (P < !0( 1)

At six months, symptom score fell to 1 8 in TURPgroup, 3S0 in laser group; Q,1,, rose to 24 4 in TURPgroup, 18X5 in laser group; residual urine fell to121 ml in TURP group, 42 5 in laser group. Meanestimated blood loss in TURP group 515 ml, . 51 mlin laser group; postoperative hospital stay 6 7 das s forTURP group, 296 days for laser group. After lasertreatment, patients catheterised for 34 days onaverage (range 9-66). Improvement after lasertreatment developed gradually over 3 months

1 URP is significantly moreeffective than lasertreatment, and laser causesfesver perioperatis eproblems. High les cls ofdysuria and longercatheterisation in lasergroup raise concerns aboutpatient acceptability

TURP and VLAP are notequivalent; laser treatmentis safer, but TURP is moreeffective. Complication ratefor TURP (36"//, of patients)is high; authors comment"One suspects that the ratesquoted in the literature max,in fact, be too losseSimilar benefits for the twsoprocedures after 6 months.Although laser treatmentinvolves shorterhospitalization and lessblood loss, this is balancedby longer catheterisationand long delay beforepatients experience benefits

laser prostatectomy include impotence (4%),urethral stricture (5%), incontinence (3.8%),and retrograde ejaculation (5*4%) .There have been no reports on patient

acceptability, but the need for catheterizationfor up to 34 days,7" high prevalence of persis-tent local irritation and dysuria,>' and delay ofseveral months before full benefits develop, aredisadvantages. No randomised controlled trialpublished so far has compared different laser

systems. The durability of the procedure andits long term effects are not known.

HYPERTHERMIA, IHERMOTHERAFPY, AND)THERMAL ABLATION

In these types of treatment, localised heat isgenerated in the prostate by microwaves,producing temperatures ranging from 42-44 Cfor hyperthermia to 60-75 C for thermalablation.

116



j.com/

Qual H


ownloaded from



There have been several published and un-published trials comparing microwave thermo-therapy with sham treatment, most of whichsuggest that microwave thermotherapy pro-duces benefits in excess of placebo.79 However,the results seem to vary widely betweenstudies. One recent double blind randomisedtrial of 145 men compared the effects ofhyperthermia at 450C delivered by one of tworoutes (transurethral or transrectal) with a370C control.8" After one year, there was noevidence that the treatment had any effect onobjective measures, and although transurethralmicrowave treatment seemed more effectivethan its corresponding sham, the data suggestthat this sham group may have been excep-tional. The authors concluded that hyper-thermia was not an effective treatment forbenign prostatic hyperplasia.

Transurethral needle ablation uses radio-frequency energy to destroy prostate tissue.8' Itis reported to be well tolerated withoutanaesthesia but its effectiveness is unknown.

HIGH INTENSITY ULTRASOUND

This uses a transrectal probe to produceprostatic lesions. Early results from small trialssuggest that it is well tolerated and reducessymptoms,82 but most patients have transienturinary retention and long term effects areunknown. This treatment is still experimental.

BALLOON DILATATION

Balloon dilatation involves insertion of aballoon into the prostate through the urethra.Two small randomised controlled trials havebeen published which showed no benefitcompared with cytoscopy83 or resection.84 Flowrates returned to pretreatment levels within ayear.84 It is not clear that balloon dilatationoffers any lasting benefit.

STENTS

Stents are flexible prostheses inserted into theurethra to hold it open. There have been norandomised controlled trials to assess their rolein the treatment of benign prostatic hyper-plasia. Observational studies report a fairlyhigh success rate, but problems include dis-placement, calcification, infection, inconti-nence, and discomfort.22 85

Costs and cost effectiveness oftreatmentWatchful waiting is likely to be the lowest costoption. Most men with mild or moderatesymptoms are likely to require no more thanperiodic checks by their general practitioners(GPs) and may benefit from lifestyle adviceand bladder training. Up to half may find thatthe problem resolves spontaneously.Drug treatment is likely to cost (1995 prices)

£325 per man-year for finasteride, £347 forterazosin, and £60 for prazosin (which is outof patent).3

Costs of surgical procedures dependcrucially on the duration of hospital stay,theatre time, complications, and retreatmentrates. Open prostatectomy has been estimatedto cost 85% more than resection.86 Against thismust be balanced lower retreatment rates after

open prostatectomy. The most cost-effectivesurgical operation seems to be incision whichis as effective as resection with shorter times intheatre and in hospital and a lower compli-cation rate.

Analyses of cost use and cost effectivenesscomparing resection, finasteride, and watchfulwaiting have recently been published by theCanadian Coordinating Office for HealthTechnology Assessment.87 This considereddirect costs to the healthcare system, andassessed benefits in terms of quality adjustedlife years (QALYs) with figures from theagency for health care policy and research forthe effectiveness of each type of treatment.This suggests that for moderate symptoms,watchful waiting followed by resection if symp-toms become less tolerable is the most costeffective strategy. However, surgery offers clearbenefits for patients with severe symptoms. Italso suggests that for patients with a relativelyshort life expectancy - < 3 years - drugtreatment may be more cost effective. Theseresults are sensitive to assumed estimates ofeffectiveness, and do not include considerationof other options such as incision, laser treat-ment, and ao blocking drugs.A comparison of the cost effectiveness of

visual laser treatment and resection carried outin Australia88 suggested that visual laser pros-tatectomy was more expensive than resection,although the difference was not great. Theresults of this study, which involved 71 patientsrandomised to resection or visual laserprostatectomy, suggested that the outcomes ofthe two procedures were equivalent, and there-fore cost effectiveness depended on the relativecosts of treatment and care after discharge.The figures were particularly sensitive to thecost of the laser fibre, which might change withtime and changing techniques.A United States review of hospital charges

for 115 patients with benign prostatic hyper-plasia at a single institution found that resec-tion cost over $2000 more than laser pros-tatectomy. The single greatest differencebetween the treatments was due to the fact thatlaser treatment was carried out on an out-patient basis, whereas resectioned patientsremained in hospital for an average of threedays.89

ConclusionsEach type of treatment involves a differentbalance of risks and benefits, and patientsshould be encouraged to participate in makinginformed decisions about their management.Watchful waiting is likely to be most cost

effective and appropriate for men whose symp-toms are not unacceptably severe, as many arewilling to tolerate their symptoms wheninformed about the natural history of thecondition. Drug treatment generally has asmall effect on symptoms.

Surgery, whether resection or incision of theprostate, leads to considerable improvementsfor most patients with severe symptoms and isprobably the most cost effective treatment forsuch men. Transurethral incision may be thebest choice for half of these men, as it is

117



j.com/

Qual H


ownloaded from


1l1eiville, Donocan, S/eldon, Peters

associated with lower levels of morbidity andmortality than resection, but less than 5% are

given this operation." New technologies needto be compared with transurethral resection or

incision in randomised controlled trials ofsufficient power and duration to evaluate theirlong term effectiveness and cost effectiveness.

We thank the ftollossing people for their advice and assistance:Paul Abrams, Michael Barrs', Nick Black, Joanna Coast, Adaml)arkins, Mark Emberton, John Fitzpatrick, Andres Long.MartinMcKee, David Neal.

I C consensus Committee of N'HO.BPHC'oiisclu/i ini/L/LiiMonaco: WHO, 1995.

2 Donos'an JI COastJ, Peters '1T. I'ic7ian ciii fi)r bciiig pi) stLitiL/Ivpe'plasici. Bristol: Unisersity of Bristol, 1996

3 Donos'an JL, Frankel SJ, Nanchahal 1K, Coast J,Williams MH. Prostatectomsr for benign prostatichyperplasia. In: Stes ens A, Rafters J, eds. H/alt/ ca/i rIi i'

LSS'SSilic//i, 2 Oxford: Radcliffe Medical Press, 1I994:140-201.

4 Feest TG, Round A, Hamad S. Incidence of severe acuterenal failure in adults: results of a community based studs.BRM7 1993;306:481-3.

5 Lsnch TH, Waymount B, Beacock CJ, Dunn JA,Hughes MIA, W~allace MA. Follows up after transurethralresection oft the prostate: sho needs it. 1A.1I7199 1,302:2.

Watanabe H. Natural history of benign prostatic hyper-troph'. Ut/oiiisolildMcd Bioli 1986;12:567 7 1.

7 Garrassav WM, Collins GN, Lee RJ. High prevalence ofbenign prostatic hypertrophs in the community. ILai/iL,1991;338:469-71.

8 Guess HA. The press alence of benign prostatic hyperplasiain population surveys. In: Garrasay M, ed. EpidlLniio/l(og(iifpr'istatci c/iscasc. Berlin: Springer-Verlag, 1995:121 31.

9 Hunter DJW', McKee CM, Black N, Sanderson CFB.Urinary symptoms: prevalence and sesen'ts' in British men

aged 55 and over. Epiclki/i//)l/Coiniiiiiun HeLithi/ 1994;48:569-73.

11) Barry MJ. Natural history of untreated benign prostatichyperplasia. In: Garrassay M, ed. Epic/cinu)1o'vio'/ip/)itcltdisease. Berlin: Springer-Verlag, 1995.

11 BirkhoffJD, Wiederhorn AR, Hamilton ML, Zinsser HH.Natural histor' of benign prostatic hvpertroph! and acuteurinars retention. lUro/ogvs 1976;7:48--52.

12 C raigen AA, Hickling JB, Saunders CR, C carpenter RG.Natural history of prostatic obstruction. (ioll Gmn PrLi,[Occais Papl 1969;18:226- 32.

13 Ball AJ, Feneley RC, Abrams PH. The natural history of

untreated 'prostatism". BR 7 TIi! 1981;53:61 3-6.11 Barrs MJ, Fosler FJ, O'LIeary MP, Bruskesitz RC

Holtgresse HL, Mebust WK, ct Li!. TIhe AmericanUrological Association symptom index for benignprostatic hyperplasia. 7 Ciii)l 1992;148:1549 7.

15 Mebust WK, Bosch R, Donovan JL, C)kada K,

O'LearsvMA. Symptom evaluation, quality of life andsexuality. In: Cockett Al K, Khours S, Aso Y, ChatelainC eds. Pi// Lieel/ -tiIL Seco micl U H() IiitL iOlatii)il l(,'()/Iis1ttini //iRBPH. Jersey: Scientific C(ommunicaticilInternational, 1993:129 -5i.

16 Bovarskv S, Jones G, Paulson DF, Prout GR Jr. A ness lookat bladder neck obstruction bh the food and drugadministration regulators: guide lines for investigation ofbenign prostatic hypertrophsy. Traoisactioiis ()f thl A Hi.iiLa/Association Si/Ciit/rlii's' ,Stiogcoiis 19977;68:29 32.

1 7Madsen PO, Iversen 1. A point system for selectingoperative candidates. In: Hinman F, ed. R'bo1gni pristitliihis'pcrt'ophov. Nes York: Springer-Verlag 1983:763-5.

18 Abrams P, Blaisas J, Nordling J, Griffiths DJ, Jensen KM.

The objectis e evaluation of bladder outflos obstruction.In: Cockett ATK, Khoury S, Ato Y, Chatclain C, eds.Pri/cedc iogi ii?/i Seco/idO B 'I'IHO) Ioit matoia ooal C uisiIItaii)II

inl BPH. Jersey: Scientific C(ommunication International,1993:1 1-226.

19 Wasson JH, Reda DJ, Bruskessitz RC, Elinson J, Keller AM,Henderson WG. A comparison of transurethral surgeryssith watchful ssaiting for moderate symptoms of benignprostatic hyperplasia. NAEiig/7 Med 1995;332:75-9.

21) Lepor H. Rigaud G. The efficacy of transurethral resection

of the prostate in men ssith moderate symptoms ofprostatism. Urol 1990;148:533-7.

21 Frimodt-Moller PC, Jensen KiM-E, Iversen P, Madsen PO,Bruskessitz RC. Analsysis of presenting symptoms in

prostatism. Urol 1984;132:272-6.22 McConnell JD, Barrs MJ, Bruskesitz RC, Lt al, eds. BeRig,,

pos/tatic /spe' plasica: diagnosis anid trcatmnlct. Cl/iicLLipractice guidve/nc. Rockville, MD: US Department ofHealth and Human Sers'ices, Agency for Health CarePolicy and Research, 1994.

23 san Venrooij GEPM, Boon I-A, de Dier RPE. Internationalprostate symptom score and quality of life assessment

s'ersus urodvnamic parameters in men s ith benignprostatic hyperplasia symptoms. 7 L'i/1995;153:1516-9.

24 Hunter DJW, McKee CM, Sanderson CEB, Black NA.Appropriate indications for prostatectomy in the UK

results of a consensus panel. 7 I-pidei(,n'! (,)iiiiiitiu`tVHcaulth 1994;48:58-64.

25 Isaacs J'. Importance of the natural history of benignprostatic hyperplasia in the evaluation of pharmacologicintervention. PrlstitiSiLtppl 1990;3:1 7.

26 \X'ennberg J. Outcomes research, cost containment andfear of' health care rationing. Iiil! 7 ALcL 1Q9Q0;323:1202 4.

2' Werennberg JE, Mulles AG, Hanles D, !tal. An assessmentcf' prostatectom ftor benign uirinais tract obstruction.7AIA 1988;259:3021-7 30.

28 Meredith P, Emberton M, WX'ood C, Smith J. Comparisonof patients' needs for information on prostate surgery ssithprinted materials provided by surgeons. alui/itv ibi H.Llt/iC(LoL- 1995;4:18 23.

2") Kasper JF, Mulley AG, Wennberg J'. Developing shareddecision-making programs to improve the quality ofhealth care. Qal/itvRK.t ei' BRlu/tiis 1992:183 90.

30 Sheppcrd S, Coulter A, Farmer A. Using interactive videosin general practice to inform patients about trcatmecntchoices: a pilot studs. h'ili/s'1V PILCtiLL 1995;(in press).

31 WagnerEH, Barrett 1', Barrs MJ, Barloss W, Foss ler IJ. Iheeffect of a shared dcecisionmaking program on rates ifsurgery for benign prostatic hyperplasia: pilot results. .lL!CL/iL 1995;33:765 7).

32 Black N, Glickman ME, D)ing J, Flood AB. Internationalvariation in intersention rates: sshat are the implicationsfor patient selection? hit _I 7IlLi/iil Assess HLe/ltli ('Lr.1995;11:719-32.

33 Hunter DJW', McKee C.M, Black NA, Sanderson C(AB.Health care sought and receis ed bs men ssith urinaryssrmptoms, and their siesss on prostatectoms. BR 7(JmP/ciLt 1995;44:27 3(.

34 Hedlund H. Andersson KE, Fk A. Effects of praz/osin inpatients with benign prostatic obstruction. 7 Uiil 1(833130:27 8.

35 British Medical Association, Rmoyal Pharmaceutical Socict.Rritish ilioiLi/ /ooiu/ars London: BMA, RPS.

36 Jardin A, Bensadoun H, Delauche-C aallier MC, Attal P,BPH-AIF Group. Alfuzosin for treatment of benignprostatic hvpertrophs. ILioot 1)99 1;337:1 457 61.

37 Cbhoss W, Hahn D, Sandhu D, Slanes P. Henshass R,Das G, Lt l!. Multicentre controlled trial of indoramin inthe ss mptomatic relief of benign prostatic hs pertroplhs B17 Uro! 1990;65:36 -8.

38 Sertcelik 'NIN, Unal S, Ozgur S, Imriamoglu A. Prazosin, aselective ii receptor blocker, in the treatment of benignprostatic hspertrophsy. ('itoore I'/vpeitL/ti/ ResLc li 1 99);48:1(610 74.

39 Steven II), Coffes GA, Graham N.MH, Wi'lodarczsk J,C urtis P. The effect of prazosin on patients sithsvmpti/ms of benign prostatic hvpertrophs. Autst Io'LiiiP/li.i'iii 1993;22:1 260 4.

40 Chapple CR, Stott 'A, Abrams PH, Chnistmas TJ, Milros EJG.A 2-ss eek placebo controlled double-blind studs ofprazosin in the treatment of prostatic obstruction due tobenign prostatic hyperplasia. Br 7 Uii)i 1992;70:285- 94.

41 Brasser MK, Adams G, Epstein H, Terazosin BPH StudsGroup. Terazosin in the treatment of benign prostatichyperplasia. Ac//liLz /)I'Lsii/ .LioL 1993;2:929 3'.

42 Ilepor H, Auerbacl S, Iuras-Baez A, Naras an ', Solos\asNtLowse F, Lt ol. A randomized, placebo-controlled miulti-center studs of the efficacy and safety of terazosin in thetreatment of benign prostatic hyperplasia. 7 L'oi/ 1 i))2;148:1467 74.

43 Di Silscrio F. Use of terazosin in the medical treatment ofbenign prostatic hyperplasia: experience in Itals. Ri o7 l(l1992;70(suppl 1P:22 6.

44 I losd SN, Buckles JF, C hilton CP, Ibrahim I, Kaisars AV,Kirk D. Terazosin in the treatment of benign prostatichs'perplasia: a multicentre, placebo-controlled trial. Bi JL'iii/ 1992;70(suppl 1):17 21.

45 Lepor H. Long-term efficacy and safety of terazosin inpatients sith benign prostatic hyperplasia. Uiili/ii 199;45:406 13.

46 Stoner IF, Finasteride Studv Group. Maintenance ot clinicalefficaes sith finasteride therapy for 24 months in patientssith benign prostatic hyperplasia. Ai1li Iitieii ALed 1)94;154:83 8.

47 Andersen JT, Ekman j), Wolf H, Beisland HO.Johansson JE, Kontturi MN, eIt il. Scandinas ian BPH StudyGroup. Can finasteridc reverse the progress of benignprostatic hyperplasia? A 2 sear placebo-controlled studs.L'ro/(ii 1995;46:631 7.

48 Gormles GJ, Stoner E, Bruskewitz RC, Lt Li!. The effect offinasteride in men with benign prostatic hyperplasia. TheFinasteride Studs' Group. N' F/g! 7 Aled 1992;327:118i 91.

49 The Finasteride Study Group. Finasteride in the treatmentof benign prostatic hvperplasia. Posutate 1993;22:291 9.

50 Stoner E, Finasteride Study Group. The clinical effects ofa 5-reductase inhibitor, finasteride, cn benign prostatichyperplasia. Y Crol 1992;147:1298-31)2.

-1 Stoner E, Finasteridc Study Group. Three- ear safety andefficacy data on the use of finasteride in the treatmentof benign prostatic hyperplasia. ULioogs' 1994;43:284 93.

52 Fitzpatrick JM, Lynch 'H. Phytotherapeutic agents in themangement of symptomatic benign prostatic hyperplasia.U'i'ii/ ('lin North Am11995;22:407 12.

53 Buck AC, Cox R, Rees RN'M, Ebeling L, John A. reat-ment of outflow tract obstruction due to benign prostatichyperplasia sith the pollen extract, Cernilton. A double-blind, placebo-controlled studs. Rr 7 C'ro! l 991);66:398 41)4

118



j.com/

Qual H


ownloaded from



54 Carbin BE, Larsson B, Lindahl 0. Treatment of benignprostatic hyperplasia with phytosterols. Br J Urol 1990;66:639-41.

55 Emberton M, Neal DE, Black N, Harrison M, Fordham M,McBrien MP, et al. The national prostatectomy audit: theclinical management of patients during hospitaladmission. BrJ7 Urol 1995;75:301-16.

56 Department of Health. Hospital episode statistics. London:DH, 1993/4.

57 McPherson K. Why do variations occur? In: Andersen TF,Mooney G, eds. The challenge of medical practice variations.London: Macmillan, 1990.

58 Neal DE. Prostatectomy: an open or closed case. BrJ Urol1990;66:449-54.

59 Emberton M, Neal DE, Black N, Fordham M, Harrison M,McBrien MP, et al. The effect of prostatectomy onsymptom severity and quality of life. Br J Urol 1996;77:233-47.

60 Thorpe AC, Clearly R, Coles J, Vernon S, Reynolds J,Neale DE. Death and complications following pros-tatectomy in 1400 men in the Northern Region ofEngland. BrJ7 Urol 1994;74:559-65.

61 Concato J, Horwitz RI, Feinstein AR, Elmore JG, Schiff SF.Problems of comorbidity in mortality after prostatectomy.JAMA 1992;267:1077-82.

62 Roos NP, Wennberg JE, Malenka DJ, Fisher ES,McPherson K, Andersen TF, et al. Mortality andreoperation after open and transurethral resection of theprostate for benign prostatic hyperplasia. N Engl J Med1989;320:1120-4.

63 Sidney S, Quesenberry CP Jnr, Sadler MC. Reoperationand mortality after surgical treatment of benign prostatichypertrophy in a large prepaid medical care program. MedCare 1992;30:117-25.

64 de Wildt MJAM, de la Rosette JJMCH, Debruyne FMJ.Retreatment rate after surgical and non-surgical treatments.Benign prostatic hyperplasia. Wiley-Liss, 1994:597-613.

65 Smith JM. Open prostatectomy. In: Fitzpatrick JM,Krane RJ, eds. The prostate. London: ChurchillLivingstone, 1989:207-14.

66 Milroy E, Chapple C. The aetiology and treatment of benignprostatic obstruction. Practitioner 1988;232:1141-5.

67 Walsh A. Indications for prostatic surgery and selection ofoperation. In: Fitzpatrick JM, Krane RJ, eds, The prostate.London: Churchill Livingstone, 1989:137-42.

68 Meyhoff HH, Nordling J, Hald T. Clinical evaluation oftransurethral versus transvesical prostatectomy. Arandomized study. Scand3 Urol Nephrol 1984;18:201-9.

69 Meyhoff HH, Nordling J. Long term results of transurethraland transvesical prostatectomy. A randomized study.ScandJ7 Urol Nephrol 1986;20:27-33.

70 Reihmann M, Knes JM, Heisey D, Madsen PO,Bruskewitz RC. Transurethral resection versus incision ofthe prostate: a randomized, prospective study. Urology1995;45:768-75.

71 Soonawalla PF, Pardanani DS. Transurethral incisionversus transurethral resection of the prostate. A subjectiveand objective analysis. BrJr Urol 1992;70:174-77.

72 Neilsen HO. Transurethral prostatectomy versus trans-urethral prostatectomy in benign prostatic hypertrophy: aprospective randomized study. BrJr Urol 1988;61:435-8.

73 Gerber GS. Lasers in the treatment of benign prostatichyperplasia. Urology 1995;45: 193-9.

74 Anson K, Watson G. The current status of the use of lasersin the treatment of benign prostatic hyperplasia. BrJr Urol1995;75(suppl 1):34-41.

75 Cowles RS, Kabalin JN, Childs S, Lepor H, Dixon C,Stein B, et al. A prospective randomized comparison oftransurethral resection to visual laser ablation of theprostate for the treatment of benign prostatic hyperplasia.Urology 1995;46:155-60.

76 Schulze H, Martin W, Hoch P, Senge T. TULIP v TURP:a prospective randomized study. J Urol 1994;151:228A.

77 Babayan RK, Roth RA, McCullough DL, Gordon JO,Reese JH, Fuselier HA, et al. TULIP - two year results.J Urol 1994;151:228A.

78 Oesterling JE. Benign prostatic hyperplasia: medical andminimally invastive treatment options. N Engl J Med1995;332:99-109.

79 de Wildt MJAM, de la Rosette JJMCH. Transurethralmicrowave thermotherapy: an evolving technology in thetreatment of benign prostatic enlargement. Br J7 Urol1995;76:531-8.

80 Abbou CC, Payan C, Viens-Bitker C, Richard F, Boccon-Gibod L, Jardin A, et al. French BPH HyperthermiaGroup. Transrectal and transurethral hyperthermiaversus sham treatment in benign prostatic hyperplasia: adouble-blind randomized multi-centre clinical trial. Br JUrol 1995;76:619-24.

81 Rasor JS, Zlotta AR, Edwards SD, Schulman CC.Transurethral needle ablation (TUNA): thermal gradientmapping and comparison of lesion size in a tissue modeland in patients with benign prostatic hyperplasia. Eur Urol1993;24:411-4.

82 Madersbacher S, Kratzik C, Szabo N, Susani M, Vingers L,Marberger M. Tissue ablation in benign prostatichyperplasia with high-intensity focused ultrasound. EurUrol 1993;23(suppl 1):39-43.

83 Lepor H, Sypherd D, Machi G, Derus J. Randomizeddouble-blind study comparing the effectiveness ofballoondilation of the prostate and cystoscopy for the treatmentof symptomatic benign prostatic hyperplasia. J Urol1992;147:639-42.

84 Donatucci C, Donohue R, Berger N, Raife M, Kreder K,Crawford E. Randomized clinical trial comparing balloondilatation to transurethral resection of prostate for benignprostatic hyperplasia. Urology 1993;42:42-9.

85 Smith PH, Marberger M, Conort P, Denonec M, Foster I.Other non-medical therapies (excluding lasers) in thetreatment of BPH. In: Cockett ATK, Khoury S, Aso Y,Chatelain C, eds. Proceedings of the Second WHOInternational Consultation on BPH. Jersey: ScientificCommunication International, 1993:451-506.

86 Meyhoff MH, Nordling J, Hald T. Economy in transurethralprostatectomy. ScandJ7 Urol Nephrol 1985;19: 17-20.

87 Canadian Coordinating Office for Health TechnologyAssessment. Cost-effectiveness and cost-utility analyses offinasteride therapy for the treatment of benign prostatichyperplasia. CCOHTA, Ottawa, 1995.

88 Jackson T, Street A, Costello A, Crowe H. Cost-effectiveness of laser ablation of the prostate: prematureevaluation. Int 7 Technol Assess Health Care 1995;11:3595-610.

89 Kabalin JN, Butler ED. Costs of minimally invasive surgerycompared with transurethral electrocautery resection ofthe prostate. Westem J7Med 1995;162:5426-9.

119



j.com/

Qual H


ownloaded from


effectiveness bulletin - qualitysafety.bmj.com · madsen-iversen symptom scores" had fallen by...

Documents