Intervention Review

You have free access to this content

Surgery for cervical intraepithelial neoplasia

  1. Pierre PL Martin-Hirsch1,*,
  2. Evangelos Paraskevaidis2,
  3. Andrew Bryant3,
  4. Heather O Dickinson3

Editorial Group: Cochrane Gynaecological, Neuro-oncology and Orphan Cancer Group

Published Online: 4 DEC 2013

Assessed as up-to-date: 23 NOV 2012

DOI: 10.1002/14651858.CD001318.pub3


How to Cite

Martin-Hirsch PPL, Paraskevaidis E, Bryant A, Dickinson HO. Surgery for cervical intraepithelial neoplasia. Cochrane Database of Systematic Reviews 2013, Issue 12. Art. No.: CD001318. DOI: 10.1002/14651858.CD001318.pub3.

Author Information

  1. 1

    Royal Preston Hospital, Lancashire Teaching Hospital NHS Trust, Gynaecological Oncology Unit, Preston, Lancashire, UK

  2. 2

    Ioannina University Hospital, Department of Obstetrics and Gynaecology, Ioannina, Greece

  3. 3

    Newcastle University, Institute of Health & Society, Newcastle upon Tyne, UK

*Pierre PL Martin-Hirsch, Gynaecological Oncology Unit, Royal Preston Hospital, Lancashire Teaching Hospital NHS Trust, Sharoe Green Lane, Fullwood, Preston, Lancashire, PR2 9HT, UK. martin.hirsch@me.com.

Publication History

  1. Publication Status: New search for studies and content updated (no change to conclusions)
  2. Published Online: 4 DEC 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]
Alvarez 1994

MethodsRCT


Participants375 women with cervical smears suggesting CIN 2 or 3, or 2 smears equivalent to CIN1
Women with adequate colposcopy included with entire lesion visible, not pregnant
Women with vaginitis, lesion extending to vagina, evidence of invasion excluded


InterventionsPrimary LLETZ
Colposcopic directed biopsy and endocervical curettage, Only if positive laser ablation of transformation zone


OutcomesHistological status of LLETZ or colposcopic specimens
Operators impression of significant peri-operative bleeding
Women's subjective opinion of peri-operative pain
Women's subjective opinion of post-operative severe discomfort, heavy discharge, severe bleeding
Residual disease (cytology) at 3 and 6 months


Notes195 randomised to LLETZ, 180 to Laser
All women had paracervical 1% lidocaine with 1:100,000 ephedrine
LLETZ group: 6 treated by laser ablation due to technical problems, 4 failed to attend for treatment
Laser group: 66 women did not require treatment, 114 required treatment
4 women were treated by LLETZ, 2 by cryosurgery due to technical problems


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskComputer generation was used to assign women to either LLETZ or laser, "they (patients) were assigned a treatment strategy by computer-randomised forms".

Allocation concealment (selection bias)Low risk"Computer-randomised forms contained in sealed opaque envelopes", were used as a method of concealment.

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
High risk% analysed: 190/375 (51%) and 107/375 (29%) for residual disease at 3 and 6 months respectively, "of the 190 who were compliant with follow up 3 months after treatment ... 107 returned for a second evaluation at 6 months".

All other outcomes assessed more than 51% of women.

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Berget 1987

MethodsRCT


Participants204 women with entire squamo-columnar junction visible
CIN 1 on 2 biopsies 3-6 months apart, CIN 2 or 3 not extending 3 mm into crypts
No extension onto vagina or lesion or 12.5 mm into canal


InterventionsCryotherapy
Laser ablation


OutcomesOperators impression of significant peri-operative bleeding >25cc
Women's subjective opinion of peri-operative pain (mild, moderate severe, Severe being that the woman would not consider the treatment again)
Women's subjective opinion of post-operative discomfort, heavy discharge, bleeding (none, mild, moderate, severe)
Post operative cervical stenosis
Satisfactory follow-up colposcopy at 3 months
Berget 1991 reports longer follow up for residual disease outcome: residual disease (histological) at 3, 9, 15, 21, 33, 45, 80 months


Notes103 randomised to laser, 101 randomised to cryotherapy
Laser performed ablated 2 mm lateral to transformation zone to a depth of 5-7mm
Cryo coagulation (double freeze thaw freeze technique) or more if the ice ball did not exceed the probe (25mm) by 4 mm.
Local analgesia was not routinely administered
6 laser and 2 cryotherapy women refused to be followed up
Women were offered repeat treatment with the same method of treatment as part of protocol. 3 laser and 6 cryotherapy women refused repeat treatment


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot reported, "patients fulfilling the criteria were randomized to either laser or cryo treatment".

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low riskFor residual disease:

% analysed:187/204 (92%)

Laser; 94/103 (91%)

Cryotherapy; 93/101 (92%)

All other outcomes had less loss to follow up

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Bostofte 1986

MethodsRCT


Participants123 women with CIN1,2,3


InterventionsLaser conisation
Knife conisation


OutcomesDuration
Peri-operative bleeding (quantity mls)
Post-operative bleeding (primary requiring treatment and secondary)
Post-operative pain (use of analgesics)
Adequate colposcopy
Cervical stenosis (failure to pass cotton swab)
Women complaining of dysmenorrhoea
Residual disease (3-36 months)


NotesAll procedures performed under general anaesthesia
Knife cone biopsy women had vaginal packing for 24 hours and 3 gms Tranexamic acid for 10 days. Sturmdorf sutures were not used, lateral cervical arteries used
Laser conisation women did not have vaginal packing or tranexamic acid
59 women randomised to laser conisation, 64 to knife conisation


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot reported

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low riskFor Inadequate colposcopy and cervical stenosis at follow up outcomes:

% analysed:113/123 (92%)

Laser: 56/59 (95%)

Knife: 57/64 (89%)

All other outcomes had less loss to follow up

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Cherchi 2002

MethodsRCT


Participants40 women with severe dysplasia/in situ carcinoma of the uterine cervix who underwent cervical conisation

Mean age in the trial was 34.8 years (SD=5.7 years)
There were 31 (77.5%) women with CIN II and 9 (22.5%) with CIN 3


InterventionsInterventions:

Unipolar energy scalpel (Medizin-Elektronik Elektroton 300, MARTIN, Tuttlingen, Germany)

Biopolar electrocautery scissors (Power Star; Ethicon, Inc, Somerville, NJ)

Biopolar electrocautery scissors are easy to handle; they have the same shape as surgical scissors, with an isolated nylon handle, and the two blades are separated by a thin ceramic layer, thus producing two active bipolar electrodes.


Outcomes
  • Duration of procedure
  • Peri-operative blood loss
  • Duration of recovery
  • Number of haemorrhages
  • Adequacy of margins of the lesion


NotesPrimary haemorrhage was deduced by fact that, haemorrhages was for number of women, therefore it had to be a woman's first haemorrhage

Adequacy of margins of the lesion: bipolar scissors: 11/20, monopolar scalpel: 9/20

Healing of cervix: bipolar scissors: 28.3 days (SD=4.4 days), monoploar scalpel: 35.2 days (SD=6.3 days)

Duration of recovery: bipolar scissors: 3.5 days (SD=1.5 days), monoploar scalpel: 6.4 days (SD=3.2 days)

There were no infections in either group.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Monopolar or bipolar assignment was obtained by means of a table of random digits".

Allocation concealment (selection bias)Low risk"Surgical methods were assigned randomly by drawing a sealed envelope ... An independent party filled and sealed the envelopes which were placed in a sealed box".

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% analysed: 40/40 (100%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Chirenje 2001

MethodsRCT


Participants400 women with histologically confirmed high grade squamous intraepithelial lesions

Mean age in the trial was 32.4 years (SD=6.2 years)


InterventionsLEEP: For each loop procedure the cervix was injected with 4 ml of 1% lignocaine with 1:100 000 epinephrine 1–2 mm beneath the cervical surface epithelium at 12, 3, 6 and 9 o’clock positions. We used a large speculum adapted for smoke evacuation and a 2×2 cm electrode was used for large lesions and 1×1 cm electrode for the smaller lesions. The electrosurgical generator (Surgitron Ellman International, New York, USA) was operated using the cutting mode recommended by the manufacturer.

Cryotherapy: The PCG-R Portable Cryosurgical Gun (Spembly Medical Ltd, UK) was used for cryotherapy. A large speculum was placed into the vagina and after the lesion was identified by colposcopy an appropriate-sized
probe to cover lesion and transformation zone was selected. A lubricant (KY jelly, Johnson and Johnson, South Africa) was applied to the probe before treatment of the cervix. The cervix was treated for 2 minutes, thawed and treated again for 2 minutes to allow an ice ball to form across the lesion and transformation zone.


Outcomes
  • Residual disease at 6 and 12 months follow up
  • Pain
  • Haemorrhage
  • Secondary haemorrhage
  • Discharge (watery/offensive)


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Treatment allocation was performed by a research nurse in a separate setting in accordance with computer-generated randomisation sequences stratified per treatment".

Allocation concealment (selection bias)Low risk"Treatment allocation was performed ... using consecutively numbered opaque sealed envelopes".

Blinding (performance bias and detection bias)
All outcomes
Unclear risk"The colposcopist was blinded with regard to treatment allocation". However, it was unclear as to whether the outcome assessor was blinded.

Incomplete outcome data (attrition bias)
All outcomes
Low riskFor residual disease at 6 months:

% of women analysed: 327/400 (82%)

By treatment arm:

LEEP: 159/200 (80%)

Cryotherapy: 168/200 (84%)

All other outcomes assessed more than 327 women.

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Crompton 1994

MethodsRCT


Participants80 women recruited with CIN 3
Women with a history of previous cervical surgery, peri- or post-menopausal or whose lesion extends to vagina


InterventionsLaser conisation
LLETZ


OutcomesSubjective scoring of pain by attendant nurse
Subjective scoring of pain by women by linear analogue scale
Peri-operative bleeding (none, spotting, requiring coagulation)
Operative time


NotesAll women had intracervical 4mls 2% lignocaine with 0.3 IU /mls octapressin prior to treatment
(1 spoiled data sheet)

43 women randomised to laser conisation
36 women randomised to LLETZ


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Women were then randomised to ... laser or the loop diathermy".

"Computer-generated simple randomisation", was used to assign women to either laser or LLETZ.

Allocation concealment (selection bias)Low risk"randomisation code was held in sealed opaque envelopes which also contained the data sheets".

Blinding (performance bias and detection bias)
All outcomes
Unclear risk"It was not possible to conceal the form of intervention used from the surgeon or attendant but the patient was not told whether she was having laser or loop treatment". However, it was not reported whether or not the outcome assessor was blinded.

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 79/80 (99%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Dey 2002

MethodsRCT


Participants289 women with CIN I,II,III


InterventionsLaser ablation
LLETZ


OutcomesResidual/recurent disease
Primary haemorrhage
Duration of pregnancy


Notes134 allocated to laser ablation. 120 received allocated treatment
155 allocated to LLETZ
151 received allocated treatment


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Consecutively numbered, sealed, opaque envelopes were used to allocate women to treatment with loop diathermy excision of the transformation zone or laser vaporisation according to a computer generated randomisation schedule stratified by centre".

Allocation concealment (selection bias)Low risk"Consecutively numbered, sealed, opaque envelopes were used".

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 285/289 (99%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Duggan 1999

MethodsRCT


Participants180 women recruited with all grades of CIN with the following inclusion criteria:
1) unsatisfactory colposcopy with positive biopsy
2) endocervical curettage with positive biopsy
3) possible microinvasion on biopsy


InterventionsLLETZ
Knife conisation


OutcomesAdequate colposcopy
Cervical stenosis
Incomplete resection margins
Residual disease at 3 months


Notes91 women randomised to LLETZ
89 women randomised to knife conisation


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Randomization to receive treatment by cold-knife conization (n = 89) or conization by the loop electrosurgical excision procedure (n = 91) was accomplished with a computer program using a permuted block design".

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
High risk% of women analysed: 140/180 (78%) were assessed for cervical stenosis. All other outcomes assessed more than 140 patients.

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Giacalone 1999

MethodsRCT


Participants78 women with CIN 2,3

Mean age in the trial was 35.6 years (SD=8.2 years)
There were 18 (27%) women with CIN II and 48 (73%) with CIN 3


InterventionsKnife conisation
LLETZ


OutcomesResidual disease

Post-operative bleeding
Cervical stenosis
Adequate colposcopy


Notes78 women randomised
Only 66 available for follow up:
38 knife cone
28 laser excision


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Women were assigned to the cold knife or loop excision group using a random-number table".

Allocation concealment (selection bias)Low risk"Group allocation predetermined and placed in consecutively numbered sealed envelopes".

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% analysed: 66/78 (85%)

No breakdown given in terms of groups

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Gilbert 1989

MethodsRCT


Participants200 women with CIN undergoing knife cone biopsy


InterventionsLateral haemostatic sutures and interrupted sutures if indicated
Vaginal pack with Monsels solution


OutcomesDuration of surgical procedure
Operative blood loss
Primary haemorrhage
Secondary haemorrhage


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskDetails about the sequence generation was not given, "Patients were randomly allocated to one or the other haemostatic method by opening one of a batch of sealed envelopes containing the appropriate instruction".

Allocation concealment (selection bias)Low risk"We performed the method allocation after the cone excision to ensure that previous knowledge of the haemostatic method could not influence the operator as to the size or shape of the cone".

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low riskFor all outcomes:

% analysed:200/205 (98%)

5 women did not wish to participate

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Healey 1996

MethodsRCT


Participants55 women with CIN


InterventionsRadical diathermy
LLETZ


OutcomesDuration of blood loss
Duration of watery/ blood stained discharge
Duration of yellow discharge
Duration of upper abdominal pain
Duration of lower abdominal pain
Duration of deep pelvic pain
Duration of vaginal pain


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskPatients were, "assigned a treatment using ... random numbers table".

Allocation concealment (selection bias)Low risk"They were then assigned a treatment using sealed envelopes".

Blinding (performance bias and detection bias)
All outcomes
Low risk"The treatments were colour coded so the patients and the investigators collecting and analysing the data were blinded to the treatment mode".

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 48/55 (87%)

By treatment arm

Radical diathermy: 24/26 (92%)
LLETZ: 24/29 (83%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Jobson 1984

MethodsRCT


Participants125 women with CIN 1,2,3
Women with satisfactory colposcopy, negative endocervical curettage, reproductive years


InterventionsLaser ablation
Cryotherapy


OutcomesVasovagal reaction
Patient acceptance (would patient have repeat treatment)
Satisfactory colposcopy at 4 months
Residual disease at 4 and 12 months


Notes42 women were randomised to laser ablation, 39 to cryotherapy and completed protocol
Laser performed ablated 2 mm lateral to transformation zone to a depth of 5-7mm. Women had pre-operative oral ibuprofen
Cryocoagulation (double freeze thaw freeze technique) or more if the ice ball did not exceed the probe (28mm) by 4-5mm. With or without analgesia


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot reported

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
High risk% of women analysed: 81/125 (65%)

"Characteristics of those patients lost from the study were similar in both ... arms".

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Kirwan 1985

MethodsRCT


Participants106 women with CIN 3
Adequate colposcopy and no extension to vagina


InterventionsLaser ablation
Cryotherapy


OutcomesResidual disease at 4 and 10 months


Notes71 women were randomised to laser ablation, 35 to cryotherapy
Laser performed ablated transformation zone to a depth of 7mm
Cryocoagulation (double freeze thaw freeze technique)


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot reported

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 106/106 (100%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Kristensen 1990

MethodsRCT


Participants183 women with CIN 2 or 3


InterventionsKnife cone with anterior+posterior Sturmdorf sutures
Knife cone without haemostatic sutures but with vaginal packing for 6-8 hours
Laser cone


OutcomesResection margins free of disease
Primary haemorrhage
Secondary haemorrhage
Cervical stenosis
Dysmenorrhoea


Notes62 women randomised to knife cone with sutures, 60 women to knife cone with packing, 61 to laser cone
All procedures performed under general anaesthesia
All procedures performed with lateral sutures and intracervical vasopressin


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot reported

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 163/183 (89%) for cervical stenosis at follow up

By treatment arms

Laser: 56/61 (92%)

Knife: 107/122 (88%)

All other outcomes have follow up greater than 89%

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Kwikkel 1985

MethodsRCT


Participants105 women with CIN 1,2,3
Adquate colposcopy, no suspicion of invasion


InterventionsLaser ablation
Cryotherapy


OutcomesPeri-operative pain
Peri-operative bleeding
Residual disease at 3-18 months


NotesLaser performed ablating the transformation zone to a depth of 6-7mm
Cryocoagulation (double freeze thaw freeze technique) using a probe (18mm)
2 women in cryotherapy group, 2 women in laser group lost to follow up


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot reported

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 101/105 (96%)

"One patient in the cryotherapy group was lost to follow up ... an additional patient treated with cryotherapy and two treated with laser ... are excluded".

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Larsson 1982

MethodsRCT


Participants110 women with CIN 3


InterventionsLaser conisation
Knife conisation


OutcomesPeri-operative blood loss (insufficient data for analysis)
Primary haemorrhage (bleeding requiring intervention in first 4 days)
Secondary haemorrhage (bleeding after 4th day)


Notes55 women were randomised to laser conisation, 55 to knife conisation
All procedures performed under general anaesthesia
Blood loss estimated by alkaline haematin extraction from swabs etc


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot reported

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 110/110 (100%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Mathevet 1994

MethodsRCT


Participants110 women with CIN 1,2,3
Squamo-columnar junction not completely visible


InterventionsKnife cone
Laser cone
LLETZ


OutcomesEctocervical resection margin involved with disease
Endocervical resection margin involved with disease
Presence of thermal artifact not permitting evaluation of resection margins
Peri-operative bleeding requiring haemostatic sutures (loop+laser only)
Secondary haemorrhage
Cervical stenosis
Satisfactory colposcopy
Residual disease at 6 months

Residual disease at 36 months


Notes37 women were randomised to knife conisation, 37 to laser conisation, 36 to loop
All 3 treatments performed as an out-patient procedure with 10-20 mls 1% xylocaine with ephedrine.
At knife conisation haemostasis was achieved by Sturmdorf sutures, laser cone by laser coagulation and Monsels solution, loop excision by coagulation and Monsels solution

In 2004 update 86 patients were followed up for more than 3 years. Of these 28 had been treated with the cold knife, 29 with LEEP and 29 by laser.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Allocation was accomplished by drawing envelopes containing the names of the procedures (distribution of the different procedures was done according to a hazard table)".

Allocation concealment (selection bias)Unclear riskIt was not reported whether or not sealed, opaque envelopes were used

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 110/110 (100%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Mitchell 1998

MethodsRCT


Participants498 women with CIN 1,2,3
Women over 18 yrs, using contraception, biopsy proven CIN, satisfactory colposcopy with lesion entirely visible

Age details in the trial were as follows: <25 years: 171 (44%), 25-29: 105 (27%), >29: 114 (29%)
There were 123 (32%) women with CIN 1, 124 (32%) with CIN 2 and 143 (36%) with CIN 3


InterventionsCryotherapy
Laser ablation
Loop Excision


OutcomesResidual disease
Primary haemorrhage
Secondary haemorrhage


Notes139 women were randomised to cryotherapy, 121 to laser ablation, 130 to loop excision


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"A physician assistant who was not involved in treatment used the computer generated list to assign the random treatment and scheduled the patient".

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
High risk% of women analysed: 390/498 (78%)

No breakdown of numbers in treatment arms

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Nagar 2004

MethodsRCT


ParticipantsAll women over 20 years of age who required a LLETZ for suspected CIN on colposcopy

Mean age in the trial was 34.4 years (SD=9.2 years)


InterventionsPure cut: the standard Valleylab force 2 electrosurgical generator (Valleylab, CO) was set to 90W

Blend 1: both cut and coagulation were set to 60W. The diathermy setting blend 1 is a combination of 50% cutting waveform and 50% coagulating waveform

Prior to the procedure, the extent of the lesion was determined with 3% acetic acid and the cervix was injected with 4-6 ml of 3% prilocaine hydrochloride with felypressin. The size of the disposable loop was selected by the operator and performed in the standard way.


OutcomesResidual disease (follow-up smear 6 months after LLETZ procedure)

Grading and depth of thermal artefact


NotesResidual disease was assessed at 6 months follow up based on a smear result

Mean depth of thermal artefact at the epithelial margin was 0.292 mm in the blend group and 0.270 mm in the cut group (P=0.237). It was not possible to obtain a SD so this could not be displayed on a forest plot.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear risk"Randomization was performed using sequential sealed envelopes. No blocking or stratification was carried out".

Allocation concealment (selection bias)Unclear risk"Randomization was performed using sequential sealed envelopes". It was unclear whether these were opaque sealed envelopes.

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk"Three eligible women declined to take part in the trial and three were not randomized for unknown reasons. The remaining 49 women were randomized to either the cut or the blend setting for the LLETZ procedure.

Of the 55 eligible patients

% of women analysed: 49/55 (89%)

Of the 49 patients who were randomised

% of women analysed: 49/49 (100%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Oyesanya 1993

MethodsRCT


Participants300 women with CIN 1,2,3
Women with adequate colposcopy, no evidence of invasion


InterventionsLaser conisation
LLETZ


OutcomesDuration of treatment
Patient subjective assessment of pain (none/minimal, moderate, severe)
Peri-operative blood loss (difference in weight of blood stained/dry swabs)
Secondary haemorrhage
Presence of thermal artifact not permitting evaluation of resection margins
Dysmenorrhoea
Residual disease at 3-12 months


Notes150 women randomised to laser conisation, 150 to loop excision
Intra-cervical 6mls Citanest (0.5% prilocaine with octapressin) used pre-operatively


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot reported, merely states that, "women were randomized to receive treatment either by loop diathermy excision or laser excisional conization".

Allocation concealment (selection bias)Low risk"The women were randomized to receive treatment either by loop diathermy excision or laser excisional conization by drawing from a box of sealed, opaque, mixed envelopes of the same color and size, each of which contained the name of one of the procedures".

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 295/300 (98%) or secondary haemorrhage, dysmenorrhoea and residual disease:

"All patients except two in group 1 and three in group 2 attended at least one follow-up clinic. Attempts to contact these patients by telephone, letter, or through their general practitioners failed."

100% of women were assessed for all other outcomes as they could be measured during or immediately after surgery.

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Panoskaltsis 2004a

MethodsRCT


ParticipantsWomen who fulfilled criteria for cervical treatment for CIN and had not been treated previously. This included a cytological or colposcopic suspicion of CIN I or worse and unsatisfactory colposcopic examination in the presence of moderate or severe dyskaryosis or persistent mild dyskaryosis. Women with adenocarcinoma in situ were not included in the study

Mean age in the trial was 32.6 years (SD=9.4 years)
There were 149 (37.25%) women with CIN 1, 124 (31%) with CIN 2, 66 (16.5%) with CIN 3, in 33 (8.25%) women there was no pretreatment histology and was another category in 28 (7%) women


InterventionsLLETZ: performed using a 1.5, 1.8 or 2.2 cm tungsten diathermy loop (Rocket, Watford, UK) according to the surgeon’s preference for a specific lesion. During a LLETZ procedure, surgeons used diathermy settings according to their usual practice

Needle excision: performed with a 2 cm long tungsten wire (Rocket) using a pure coagulation setting of 35W. The intention with both techniques was to remove the specimen in one piece if possible


Outcomes
  • Duration of procedure
  • Peri-operative pain
  • Peri-operative complications
  • Cervical stenosis


NotesThere was no difference in the cumulative risk of developing a recurrent or residual high grade intraepithelial lesion between the two groups at follow up (log-rank test= 0.13, P= 0.72) but the study would have needed over 1200 subjects to assess recurrence rates reliably

Duration of procedure (secs): median= 90 [range: 60–120] for LLETZ procedure and median= 210 (range: 180–300) for NETZ.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Women were randomised by telephone to the trial office where a computer generated randomisation list was kept securely. Block randomisation was used with varying sized
groups".

Allocation concealment (selection bias)Low risk"Women were randomised by telephone to the trial office where a computer generated randomisation list was kept securely".

Blinding (performance bias and detection bias)
All outcomes
Unclear risk"Histological analysis was performed by the routine pathological service in each hospital. Histopathologists were unaware of the type of treatment patients received". However, it was unclear whether or not those analysing the data were blinded to the treatment mode.

Incomplete outcome data (attrition bias)
All outcomes
Low riskFor cervical stenosis outcome

% of women analysed: 339/400 (85%)

By treatment arm

LLETZ: 167/200 (84%)

Needle excision: 172/200 (86%)

All other outcomes analysed at least 85% of women

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Paraskevaidis 1994

MethodsRCT


Participants40 women undergoing elective hysterectomy


InterventionsLaser conisation
LLETZ


OutcomesDuration of procedure
Depth of thermal injury


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot reported, "women were serially placed into one or two groups".

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 40/40 (100%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Partington 1989

MethodsRCT


Participants100 women with CIN 1,2,3
Women with adequate colposcopy , no evidence of invasion, lesion no more than 5mm into canal


InterventionsLaser conisation
Laser ablation


OutcomesDuration of treatment
Significant peri-operative bleeding
Women's subjective opinion of peri-operative pain (mild, moderate, severe)
Secondary haemorrhage (seen in out-patients)
Secondary haemorrhage (required admission)
Adequate colposcopy
Cervical stenosis
Dysmennorrhoea
Residual disease at 6, 12, 24 months


Notes50 women randomised to laser conisation, 50 women randomised to laser ablation
Haemostasis achieved by pressure with a cotton swab or Monsel solution
Laser Excision 2mm margin to lesion and to a depth of 2-3mm
Laser ablation to a depth of 10mm
Intra-cervical 3% prilocaine with octapressin used pre-operatively


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot reported

Allocation concealment (selection bias)Low risk"Patients were randomised to treatment ... by drawing sealed envelopes".

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 93/100 (93%) for residual disease

By treatment arm

Laser conisation: 45/50 (90%)
Laser ablation: 48/50 (96%)

All other outcomes analysed all 100 patients

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Sadek 2000

MethodsRCT


Participants105 women with histologically verified CIN.


Interventions
  • Needle excision of the transformation zone (NETZ), in which a new, specially designed diathermy needle is used
  • Large loop excision (LLETZ)
  • Cold knife


Outcomes
  • Mean operating time including anaesthesia
  • Free resection margins
  • Residual disease
  • Mean postoperative duration of vaginal discharge
  • Success rate after single treatment


NotesFollow-up time is three years; all patients were evaluated at 3, 6, 9, 12, 18, 24 and 36 months after surgery

Mean operating time including anaesthesia (NETZ 7.4 minutes, LLETZ 8.6 minutes, cold knife 17.7 minutes, P<0.05)

Conversion to general anaesthesia (NETZ 11%, LLETZ 22%, cold knife 37%, P=0.04)

Free resection margins (NETZ 85%, LLETZ 37%, cold knife 68%, P<0.05)

Residual disease (NETZ 2%, LLETZ 28%, cold knife 14%, P<0.05)

Mean post-operative duration of vaginal discharge (NETZ 9 days, LLETZ 12 days, cold knife,13 days, P<0.05)

Success rate after single treatment (NETZ, 97.1%, cold knife 85.7%, LLETZ 71.4%).


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot reported

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 105/105 (100%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Santos 1996

MethodsRCT


Participants447 women with CIN 1,2,3
Women with suspicion of invasion, extensive lesion, pregnant were excluded

153 patients (24%) were candidates for other type of treatment due to conditions contraindicating 1-day management


InterventionsLLETZ
Laser conisation


OutcomesResidual disease
Significant peri-operative bleeding
Secondary haemorrhage
Cervical stenosis at follow up
Satisfactory colposcopy at follow up


Notes145 women randomised to laser conisation, 147 to loop
Intracervical 6mls 2% lidocaine with 1:80,000 ephedrine used pre-operatively


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"A list was elaborated using an aleatory numbers’ table, and according to that order new patients were correlatively allocated at each therapeutic arm".

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 294/294 (100%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Schantz 1984

MethodsRCT


Participants142 women with ectocervical CIN 1 and 2


InterventionsSingle freeze
Double freeze
Cryotherapy


OutcomesResidual disease at 6 months


Notes61 underwent single freeze
81 underwent double freeze


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"By using random numbers the patients were assigned to either single-freeze or double freeze treatment".

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 142/142 (100%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Takac 1999

MethodsRCT


Participants240 women with CIN 1,2,3.
All procedures were done as in-patients

Mean age in the trial was 32.5 years (SD=8 years).
There were 9 (3.75%) women with CIN 1, 71 (30%) with CIN 2, 152 (63%) with CIN 3, 6 (2.5%) women had stage Ia cervical cancer and 2 (0.75%) women had negative histology.


InterventionsKnife conisation: wound was sutured with two semicircular sutures (Vicryl, Ethicon, Edinburgh, UK), thus reconstructing the cervix
LLETZ: performed using an Elektrotom 400 unit (Brechtold, Tuttlingen, Germany) with loop devices ranging in size from 10 to 20 mm. The procedure was performed using a blended current with the cut frequency set at 40 W and the coagulation frequency set between 20 and 40 W


OutcomesCompleteness of excision (endo/ectocervial disease involvement)
Adequate colposcopy rates after treatment
Primary haemorrhage


Notes120 randomised to Knife cone
120 randomised to LLETZ


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear riskNot reported

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 120/120 (100%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Townsend 1983

MethodsRCT


Participants200 women with CIN 1,2,3
Adequate colposcopy, no evidence of invasion


InterventionsLaser ablation
Cryotherapy


OutcomesSevere cramps
Vasomotor symptoms
Residual disease at 6 months.


Notes100 women randomised to laser ablation, 100 randomised to cryotherapy
Cryo coagulation (single freeze thaw technique) using a probe (18mm) with ice ball extending 5 mm beyond abnormal epithelium
Laser ablation of all transformation zone

No patient was lost to follow up


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Unclear risk"The study was designed ... in as objective a way as possible by alternating cases randomly insofar as possible on the bases of histologic grade and lesion size". This study seems to deploy the method of minimisation and aims to minimise the imbalance between the number of patients in each treatment group over two important prognostic factors.

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 200/200 (100%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

Vejerslev 1999

MethodsMulti-centre RCT at the Glostrup Hospital and the Hvidovre Hospital in Denmark


Participants222 women with a histological diagnosis of CIN II-III, or persistent CIN I

Median age (years) in the trial at the Glostrup Hospital was 33 (range: 18-74) in the loop group and 32 (range: 21-58) in the laser group

At the Hvidovre Hospital the median age (years) was 31 (range: 18-56) in the loop group and 29 (range: 19-75) in the laser group

The age range in the trial was 18-75 years and there was no significant difference in age between the 4 groups.
There were 9 (4%) women with CIN 1, 45 (20%) with CIN 2 and 168 (76%) with CIN 3


InterventionsLaser conisation: performed using a CO2 laser (A: Sharplan Model 733A, B: Sharplan model 1050, Laser Industries Ltd) attached to the colposcope. The laser energy was delivered in continuous (A) or pulsatile pulse mode (B) using power densities ranging from 3000–5500 W/cm². After removal of the tissue a defocused beam coagulated the cervical lesion

Diathermy loop conisation: this was done without simultaneous colposcopic guidance. The excision zone and depth was determined by the colposcopic description in the patients record, whether atypical cytology and/or histology was present in samples from the exocervix, the endocervix, or both, and guided by application of acetic acid solution and Schiller’s iodine. The size and shape of the diathermy loop (Niko-Med) were chosen among three (wideXdeep: 10X5 mm, 15X7 mm, or 20X9 mm) to excise the lesion in a single sweep. Larger lesions required excision in two or three sections. A Davol model 2000 BP II(A) or 2000(B) (Electro Medical Systems) electrosurgical generator supplied the diathermy power. A combination of cutting and coagulation was used for excision. The power setting was adjusted according to the size of the loop and the cone. Hemostasis was achieved by ball diathermy


Outcomes
  • Peri-operative severe bleeding
  • Bleeding (ordinal scale: none, >1day, 1-7 days, 8-14 days, 15-31 days)
  • Time for excision
  • Vaginal discharge
  • Cervical stenosis


NotesMedian duration of procedure (mins) at the Glostrup Hospital was 4 (range: 0.5-60) in the loop group and 20 (range: 3.5-60) in the laser group

Median duration of procedure (mins) at the Hvidovre Hospital was 3 (range: 1-14) in the loop group and 10 (range: 3-25) in the laser group

Residual disease based on abnormal cytology at 6 or 9 months.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low risk"Women admitted for conization were allocated by a computer-generated randomisation code to either loop or laser excision".

Allocation concealment (selection bias)Unclear riskNot reported

Blinding (performance bias and detection bias)
All outcomes
Unclear riskNot reported

Incomplete outcome data (attrition bias)
All outcomes
Low risk% of women analysed: 222/222 (100%)

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasUnclear riskInsufficient information to assess whether an important risk of bias exists

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Bar-AM 2000Not an RCT, "the findings of the study group were compared with those of the last 161 consecutive patients who underwent LEETZ alone (the control group)"

Boardman 2004Trial does not report outcome measures as specified in protocol

Ferenczy 1985Quasi-RCT

Gentile 2001Review article

Girardi 1994Quasi-RCT

Gunasekera 1990Quasi-RCT

Lisowski 1999Not an RCT, "the choice between LLETZ or laser CO2 was made based on a pre-treatment examination (cytology, colposcopy, microbiology test and punch biopsy)"

O'Shea 1986Quasi-RCT

Panoskaltsis 2004bCommentary on an earlier published RCT

Singh 1988Quasi-RCT

 
Comparison 1. Single freeze cryotherapy versus double freeze cryotherapy

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Residual Disease within 12 months1Risk Ratio (IV, Random, 95% CI)Subtotals only

 
Comparison 2. Laser ablation versus cryotherapy

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Residual Disease (All Grades of CIN)6935Risk Ratio (IV, Random, 95% CI)1.13 [0.73, 1.76]

 2 Residual Disease (CIN1, CIN2, CIN3)4567Risk Ratio (IV, Random, 95% CI)1.51 [0.91, 2.51]

    2.1 CIN1
473Risk Ratio (IV, Random, 95% CI)2.75 [0.68, 11.11]

    2.2 CIN2
4289Risk Ratio (IV, Random, 95% CI)1.37 [0.65, 2.88]

    2.3 CIN3
4205Risk Ratio (IV, Random, 95% CI)1.38 [0.62, 3.09]

 3 Peri-operative Severe Pain3493Risk Ratio (IV, Random, 95% CI)2.00 [0.64, 6.27]

 4 Peri-operative Severe Bleeding2305Risk Ratio (IV, Random, 95% CI)5.83 [0.71, 47.96]

 5 Vaso-motor Symptoms1Risk Ratio (IV, Random, 95% CI)Subtotals only

 6 Malodorous Discharge2400Risk Ratio (IV, Random, 95% CI)0.30 [0.12, 0.77]

 7 Inadequate Colposcopy at Follow-up2272Risk Ratio (IV, Random, 95% CI)0.38 [0.26, 0.56]

 8 Cervical Stenosis at Follow-up2464Risk Ratio (IV, Random, 95% CI)1.45 [0.45, 4.73]

 
Comparison 3. Laser conisation versus knife conisation

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Residual Disease (All Grades of CIN)2194Risk Ratio (IV, Random, 95% CI)0.64 [0.22, 1.90]

 2 Primary Haemorrhage2306Risk Ratio (IV, Random, 95% CI)0.53 [0.18, 1.54]

 3 Secondary Haemorrhage3359Risk Ratio (IV, Random, 95% CI)0.91 [0.34, 2.40]

 4 Inadequate Colposcopy at Follow-up2160Risk Ratio (IV, Random, 95% CI)0.57 [0.39, 0.81]

 5 Cervical Stenosis at Follow-up41007Risk Ratio (IV, Random, 95% CI)0.38 [0.19, 0.76]

 6 Significant Thermal Artifact Prohibiting Interpretation of Resection Margin1Risk Ratio (IV, Random, 95% CI)Subtotals only

 
Comparison 4. Laser conisation versus laser ablation

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Residual Disease (All Grades of Disease)1Risk Ratio (IV, Random, 95% CI)Subtotals only

 2 Peri-operative Severe Bleeding1Risk Ratio (IV, Random, 95% CI)Subtotals only

 3 Secondary Haemorrhage1Risk Ratio (IV, Random, 95% CI)Subtotals only

 4 Inadequate Colposcopy at Follow-up1Risk Ratio (IV, Random, 95% CI)Subtotals only

 
Comparison 5. Laser conisation versus loop excision

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Residual Disease4889Risk Ratio (IV, Random, 95% CI)1.24 [0.77, 1.99]

 2 Duration of Procedure3419Mean Difference (IV, Random, 95% CI)11.66 [1.37, 21.95]

 3 peri-operative severe bleeding1Risk Ratio (IV, Random, 95% CI)Subtotals only

 4 Peri-operative Severe Pain2594Risk Ratio (IV, Random, 95% CI)4.34 [0.25, 75.67]

 5 Secondary Haemorrhage4889Risk Ratio (IV, Random, 95% CI)1.41 [0.72, 2.76]

 6 Significant Thermal Artefact on Biopsy2373Risk Ratio (IV, Random, 95% CI)2.38 [0.61, 9.34]

 7 Depth of Thermal Artifact1Mean Difference (IV, Random, 95% CI)Subtotals only

 8 Inadequate Colposcopy2339Risk Ratio (IV, Random, 95% CI)1.38 [0.48, 3.97]

 9 Cervical Stenosis3560Risk Ratio (IV, Random, 95% CI)1.21 [0.57, 2.57]

 10 Vaginal discharge1Risk Ratio (IV, Random, 95% CI)Subtotals only

 
Comparison 6. Laser ablation versus loop excision

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Residual Disease3911Risk Ratio (IV, Random, 95% CI)1.15 [0.59, 2.25]

 2 Peri-operative Severe Pain1281Risk Ratio (IV, Random, 95% CI)0.38 [0.02, 7.91]

 3 Primary Haemorrhage2560Risk Ratio (IV, Random, 95% CI)0.35 [0.04, 3.14]

 4 Secondary Haemorrhage2560Risk Ratio (IV, Random, 95% CI)0.54 [0.14, 2.10]

 
Comparison 7. Knife conisation versus loop excision

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Residual Disease3279Risk Ratio (IV, Random, 95% CI)0.47 [0.20, 1.08]

 2 Primary Haemorrhage2306Risk Ratio (IV, Random, 95% CI)1.04 [0.45, 2.37]

 3 Inadequate Colposcopy at Follow-up3291Risk Ratio (IV, Random, 95% CI)1.63 [0.85, 3.15]

 4 Cervical Stenosis3251Risk Ratio (IV, Random, 95% CI)1.12 [0.44, 2.84]

 
Comparison 8. Radical diathermy versus LLETZ

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Duration of blood loss1Mean Difference (IV, Random, 95% CI)Subtotals only

 2 Blood stained / watery discharge1Mean Difference (IV, Random, 95% CI)Subtotals only

 3 Yellow discharge1Mean Difference (IV, Random, 95% CI)Subtotals only

 4 White discharge1Mean Difference (IV, Random, 95% CI)Subtotals only

 5 Upper Abdominal Pain1Mean Difference (IV, Random, 95% CI)Subtotals only

 6 Lower Abdominal Pain1Mean Difference (IV, Random, 95% CI)Subtotals only

 7 Deep Pelvic Pain1Mean Difference (IV, Random, 95% CI)Subtotals only

 8 Vaginal Pain1Mean Difference (IV, Random, 95% CI)Subtotals only

 
Comparison 9. Knife cone biopsy: haemostatic sutures versus none

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Primary Haemorrhage2522Risk Ratio (IV, Random, 95% CI)0.42 [0.06, 3.23]

 2 Secondary Haemorrhage2515Risk Ratio (IV, Random, 95% CI)2.68 [1.27, 5.66]

 3 Cervical Stenosis2307Risk Ratio (IV, Random, 95% CI)1.75 [0.65, 4.72]

 4 Inadequate Colposcopy at Follow-up1Risk Ratio (IV, Random, 95% CI)Subtotals only

 5 Dysmenorrhoea2277Risk Ratio (IV, Random, 95% CI)2.50 [1.41, 4.45]

 
Comparison 10. Bipolar electrocautery scissors versus monopolar energy scalpel

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Peri-operative bleeding1Mean Difference (IV, Random, 95% CI)Subtotals only

 2 Duration of procedure1Mean Difference (IV, Random, 95% CI)Subtotals only

 3 Primary haemorrhage1Risk Ratio (IV, Random, 95% CI)Subtotals only

 
Comparison 11. LEEP versus cryotherapy

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Residual disease at 6 months1Risk Ratio (IV, Random, 95% CI)Subtotals only

 2 Residual disease at 12 months1Risk Ratio (IV, Random, 95% CI)Subtotals only

 3 Primary haemorrhage1Risk Ratio (IV, Random, 95% CI)Subtotals only

 4 Secondary haemorrhage1Risk Ratio (IV, Random, 95% CI)Subtotals only

 5 Offensive discharge1Risk Ratio (IV, Random, 95% CI)Subtotals only

 6 Watery discharge1Risk Ratio (IV, Random, 95% CI)Subtotals only

 7 Peri-operative severe pain1Risk Ratio (IV, Random, 95% CI)Subtotals only

 
Comparison 12. Pure cut setting versus blend setting when performing LLETZ

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Residual disease at 6 months1Risk Ratio (IV, Random, 95% CI)Subtotals only

 2 Depth of thermal artefact at deep stromal margin1Mean Difference (IV, Random, 95% CI)Subtotals only

 
Comparison 13. LLETZ versus NETZ

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Residual disease at 36 months1Risk Ratio (IV, Random, 95% CI)Subtotals only

 2 Peri-operative pain1Risk Ratio (IV, Random, 95% CI)Subtotals only

 3 Peri-operative blood loss interfering with treatment1Risk Ratio (IV, Random, 95% CI)Subtotals only

 4 Bleeding requiring vaginal pack1Risk Ratio (IV, Random, 95% CI)Subtotals only

 5 Cervical stenosis at follow-up1Risk Ratio (IV, Random, 95% CI)Subtotals only

 
Comparison 14. Knife conisation versus NETZ

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Residual disease at 36 months1Risk Ratio (IV, Random, 95% CI)Subtotals only

 
Comparison 15. LLETZ versus Knife conisation

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Residual disease at 36 months1Risk Ratio (IV, Random, 95% CI)Subtotals only