Open vs. Closed haemorrhoidectomy

Study reference

Study characteristics

Patient characteristics

Intervention (I)

Comparison/control (C)

Follow-up

Outcome measures and effect size

Comments

Ho, 2007

SR and meta-analysis of RCT’s

Literature search up to November 2006

A: Ho et a.l., 1997

B: Carapeti et al., 1998

C: Gencosmanoglu et al., 2002

D: Arroyo et al., 2004

E: Uba et al., 2004

F: Johannsson et al.,2006

Study design: RCT

Country:

A: Singapore

B: UK

C: Turkey

D: Spain

E: Nigeria

F: Sweden

Source of funding:

Not stated

Inclusion criteria SR: RCT, open vs. closed wound management after haemorrhoidectomy, adult patients, instruments: scalpel, scissors, diathermy

Exclusion:

Tissue seasling technology, harmonic scalpel, stapling devices, submucosal technique, quasi randomized studies, trials using anal dilatation and sphincterotomy in surgical techniques.

6 studies included

Important patient characteristics at baseline:

N, mean age

A: 67, 45±1.7

B: 35, 45 (range 27-75)

C: 80, I: 40.5 (27–68), C: 39 (26–63)

D: 200, 43.5

E: 79, NR

F: 225,I: 52 (22–82), c: 53 (27–82)

Sex: n, % Male

A: 35 (52)

B: I: 7 (41) C: 8(44)

C: I: 14 (35) C: 18 (45)

D: 123 (62)

E: I: 26 (67) C: 33 (83)

F: I: 61 (53) C: 55 (50)

Groups comparable at baseline? Yes

Describe intervention (n)

A: Open haemorrhoidectomy (34)

B: Open haemorrhoidectomy (17)

C: Open haemorrhoidectomy (40)

D: Open haemorrhoidectomy (100)

E: Open haemorrhoidectomy (39)

F: Open haemorrhoidectomy (115)

Describe control:

A: Closed haemorrhoidectomy (33)

B: Closed haemorrhoidectomy (18)

C: Closed haemorrhoidectomy (40)

D: Closed haemorrhoidectomy (100)

E: Closed haemorrhoidectomy (40)

F: Closed haemorrhoidectomy (110)

End-point of follow-up:

Months:

A: mean 8.7 (SE 0.2)

B: median 1.5

C: median (range): 19.5 (4–40)

D: 12

E: 3

F: 12

For how many participants were no complete outcome data available?

Not described

Cure:

Defined as symptom free or minimal residual symptoms not requiring further treatment at the end of the study period (not necessarily stated in studies), n

A: 34 vs. 33

B: 16 vs. 18

C: 40 vs. 40

D: 100 vs. 100

E: 39 vs. 40

F: 84 vs. 88

Pooled effect (random effects model model):

1.4 [0.86, 2.19]; p=0.191

Heterogeneity (I²): Q=0.3; p=0.998

Maximum pain score, (postoperative, from 0=no pain to 10=maximum pain)

A: median (range) 5 (5-9) vs. 6 (2-8)

B: -

C: -

D: mean 8.5 vs. 5.5

E: -

F: median (range) 3.25 (0-9.25) vs. 3.2 (0-10)

Pooled effect (random effects model model):

0.27 [-0.65, 1.18], p=0.570

Heterogeneity (I²): Q=45.2; p<0.001

Incontinence

A: -

B: -

C: -

D: transient flatus 2 vs. transient fluid& flatus 1, transient flatus 1

E: -

F: flatus 13%, liquids 7%, solid 1% vs. flatus 13%, liquid 1%, solid 0%

Patient satisfaction

F: excellent: 59% vs. 62%

Good 29% vs. 34%

Acceptable 9% vs. 1%

Poor 3% vs. 3%

Complications, n

Overall

A: 5 vs. 3

B: 0 vs. 0

C: 0 vs. 5

D: -

E: 18 vs. 12

F: 5 vs. 12

Pooled effect (random effects model):

0.83 [0.33, 2.01]; p=0.681

Heterogeneity (I²): Q=8.4; p=0.079

Bleeding

A: 2 vs. 3

B: -

C: 0 vs. 0

D: -

E: 6 vs. 2

F: 2 vs. 4

Wound discharge or infection

A: 0 vs. 0

B: -

C: -

D: -

E: 0 vs. 2

F: 0 vs. 3

Anal stenosis

A: 0 vs. 0

B: -

C: 0 vs. 1

D: -

E: -

F: 0 vs. 0

Notes:

Authors report assessing study quality; however study quality of studies is not reported.

Overall description of data and methods for analysis is limited (for instance: means and medians are reported, unclear how the pooled estimates are calculated).

No GRADE reported.

Some sensitivity analyses were performed excluding studies of low quality, however not transparent why exactly, so therefore not reported on here.

Heterogeneity:

For maximum pain score there was significant heterogeneity (p<0.001). This was not explained. The authors hypothesise that postoperative care and analgesia play an important role (clinical heterogeneity).

SH vs. CH

Lumb, 2010

Cochrane review, see full Cochrane review for data.

A: Basdanis, 2004

B: Bikhchandani, 2004

C: Boccasanta

D: Cheetham, 2003

E: Chung, 2005

F:Correa, Rovelo, 2002

G:Ganio, 2001

H: Gravie, 2005

I: Hetzer, 2002

J: Huang, 2006

K: Kairaluoma, 2003

L: Mehigan, 2000

M: Nystrom, 2008

N:Ortiz, 2002

O: Ortiz, 2005

P: Palimento, 2003

Q: Racalbuto, 2004

R: Rowsell, 2000

S: Senagore, 2004

T: Shalaby, 2001

U: Thaha, 2008

V: Van de Stadt, 2005

Important patient characteristics at baseline:

Grade of haemorrhoids:

A: 3, 4

B: 3, 4

C: 4

D: symptomatic prolapsing hemorrhoids

E: 3

F: 3, 4

G: 3, 4

H: symptomatic hemorrhoidal disease

I: 2, 3

J: 3

K: 3, 4

L: symptomatic prolapsing hemorrhoids

M: 3, 4

N: 3, 4

O: 4

P: symptomatic hemorrhoidal disease

Q: 3, 4

R: 3

S: 3

T: 2, 3, 4

U: 2, 3, 4

V: 2, 3, 4

THD vs. CH

Elmer, 2013

Type of study:

RCT

Setting:

Hospital

Country:

Sweden

Source of funding: supported financially by Stockholm Council Public Health and Medical Services Comittee R&D department

Inclusion criteria:

Symptomatic grade 2-3 hemorrhoids

Exclusion criteria:

Acutely thrombosed haemorrhoids, anal fissure, anal abces, anal fistula age>80, continuous consumption of analgesics, IBD, fecal incontinence, anal stenosis, bleeding disorder, rectal prolapse, undergone RBL or sclerotherapy previous 3 months, OH within 3 monhts or any previous operation with HAL, THD or SH.

N total at baseline: 40

Intervention: 20

Control:20

Important prognostic factors²:

Not described

Transanal hemorrhoidal dearteralization with anopexy (n=20)

No antibiotics were given pre- or postoperatively. Surgery performed under general anaesthesia with the addition of a preoperative perianal block.

Postoperative care: stool softener was advised, and scheduled analgesia (1g paracetamol, 100mg dextroprppoxyphene 4x daily).

Open hemorrhoidectomy (n=20)

Performed without a retractor.

Length of follow-up:

1 year

Loss-to-follow-up:

Intervention:

N (%)

Reasons (describe)

Control:

N (%)

Reasons (describe)

Incomplete outcome data:

Intervention:

N (%)

Reasons (describe)

Control:

N (%)

Reasons (describe)

Postoperative pain (patient diary numeric rating scale, first 2 weeks):

Peak pain:

“significantly lower in THD group for 5 days during the first week (<0.05).

Overall pain: “not differ between groups”

*data between I and C not compared

Bleeding

After 2-4 months: 2 vs. 0

After 1 year: 3 vs. 0

Pain

After 2-4 months: 2 vs. 1

After 1 year 2 vs. 0

Soiling

After 2-4 months: 3 vs. 1

After 1 year: 6 vs. 3

Pruritus

After 2-4 months: 3 vs. 5

After 1 year: 1 vs. 2

Complications

Urinary retention: 4 vs. 3

Bleeding (hospital visit, however bleeding had stopped): 0 vs. 2

Reintervention on short term needed because of severe pain: 1 vs. 0

Partial reprolapse within first week (reintervention needed): 3 vs. 0

Reinterventions (>1 year): 2 vs. 2

The difference in postoperative pain between THD and CH may be less than expected based on previous literature.

The sample size is too small to conclude whether THD is comparable to OH when it comes to the reduction of symptoms and restoration of anatomy.

Note: baseline characteristics unclear

Denoya, 2013

Type of study:

RCT

Setting:

hospital

Country: USA

Source of funding: supported in part by unrestricted grant by THD America (funding source had no role in trial design, conduct, data analysis, data interpretation or writing the paper).

Inclusion criteria:

Symptomatic grade 3 -4 hemorrhoids (minimum 3 quadrants, with no external component)

Exclusion criteria: pt with external component, acute incarcerated grade 4 or recurrent haemorrhoids after surgery, HIV, IBD, inability to informed consent, age<18, colorectal anal cancer, thrombosed haemorrhoids, pregnancy.

N total at baseline: 40

Intervention: 20

Control:20

Important prognostic factors²:

age ± SD:

I: 48.5±11.0

C:53.9±9.1NS

Sex:

I: 55% M

C: 25% M

Groups comparable at baseline? Exept sex distribution

Dearterizlization (with Doppled guidance and mucopexy)

Postoperative care: oxycodone or hydrocodone for pain and instruction for use of water baths, stool softeners and laxatives on an as-needed basis. Patients were seen 2 weeks after surgery.

Hemorrhoidectomy (Ferguson)

Postoperative care: oxycodone or hydrocodone for pain and instruction for use of water baths, stool softeners and laxatives on an as-needed basis. Patients were seen 2 weeks after surgery.

Length of follow-up:

3 months

Loss-to-follow-up:

-

Incomplete outcome data:

Intervention: 1

Reasons: protocol violations

Intervention: 1

Reasons: protocol violations

Pain (brief pain inventory):

“did not differ significantly on postoperative days 1, 3 and 5; by day 7 dearteralization patients scoed better in terms of general activity, mood, ability to sleep and ability to return to work”

Pain intensity (scale 0-10) (mean±sd): 2.9± 3.5 vs. 7.9±2.9

QOL (SF-12): stated in methods, not reported on

Bleeding per rectum defined as requiring intervention:stated in methods, not reported on

Urinary retention (requiring reinsertion of catheter): 0 vs. 4 (23.5%, p=0.012).

Constipation requiring laxative use or emercency room visit: 2 (8.3%) vs. 4 (23.5%), P=0.175

Incontinence to flatus

1 vs. 1

Incontinence to stool

1 vs. 1

 Study underpowered, outcomes mentioned in methods section are not clearly reported on.

THD vs. SH

Sajid, 2012

Study characteristics and results are extracted from the SR

SR and meta-analysis of RCT’s

Literature search up to may 2011

A: Festen, 2009

B: Giordano, 2011

C: Ramirez, 2005

Study design: RCT

Setting and Country:

Not described

Source of funding:

Not described

Inclusion criteria SR: RCT, all languages comparing THD with SH for any grade of haemorrhoids.

Exclusion criteria SR: not reported

3 studies included

Important patient characteristics at baseline:

N

A: I: 23, C:18

B: I: 28 C: 24

C: I: 29 C: 28

Groups comparable at baseline? Not reported.

Describe intervention:

A: THD (n=23)

B: THD (n=28)

C: THD (n=29)

Describe control:

A: SH (n=18)

B: SH (n=24)

C: SH (n=28)

End-point of follow-up:

Not described

For how many participants were no complete outcome data available?

Not described

Treatment success rate, RR [95% CI]:

A: 0.94 [0.70, 1.27]

B: 0.96 [0.82, 1.13]

C: 0.85 [0.69, 1.06]

Pooled effect (fixed effects model): 0.92 [0.81, 1.04]

Heterogeneity (I²): 0%

Postoperative complications (all). RR [95%CI]

A: 0.52 [0.10-2.80)

B: 0.57 [0.18-1.79]

C: 0.19 [0.01-3.86]

Pooled effect (fixed effects model): 0.48 [0.20, 1.18]

Heterogeneity (I²): 0%

Recurrence of hemorrhoids, RR [95%CI]

A: 0.78 [0.27, 2.29]

B: 1.29 [0.23, 7.07]

C: 2.90 [0.64, 13.16]

Pooled effect (fixed effects model): 1.33 [0.62, 2.84]

Heterogeneity (I²): 0%

Postoperative pain, mean difference [95%CI]

A: -2.00 [-2.06, -1.94]

B: -1.50 [-2.86, -0.14]

C: -2.00 [-4.13, 0.13]

Pooled effect (fixed effects model): -2.00 [-2.06, -1.94]

Heterogeneity (I²): 0%

Facultative:

Brief description of author’s conclusion: based on this review of 3 RCT’s conducted on 150 patients, THD is as effective as SH in terms of treatment success rate, operation time, postoperative complications and incidence of recurrence. In addition, THD is also associated with significantly less postoperative pain compared to SH.

Remark: study by Ramirez is abstract only, potential source of bias. Quality of studies was not high due to methodological problems. Also, significant differences about inclusion and exclusion criteria, also different definitions of outcomes used in the studies

Study reference

Study characteristics

Patient characteristics ²

Intervention (I)

Comparison/control (C) ³

Follow-up

Outcome measures and effect size ⁴

Comments

SH vs. CH

Kim, 2013

Type of study:

RCT

Setting:

Hospital

Country:

Germany

Source of funding: not stated

Inclusion criteria:

Symptomatic circumferential third-degree haemorrhoidal disease

Exclusion criteria: single third-degree haemorrhoids, acute incarcerated haemorrhoids, intercurrent acute anal fissure and/or fistula, or prior haemorrhoidectomy.

N total at baseline: 130

Intervention: 65

Control: 65

Important prognostic factors²:

age (range):

I: 53 (32-78)

C: 56 (21-79)

Sex m/f:

I: 39/22

C: 47/14

Groups comparable at baseline? Yes

SH (n=61)

under general anaesthesia

As described by Longo, 2/0 polypropylene purse-string suture including the mucosa and submucosa applied 4cm above the dentate line.

Postoperatively: advice to clean perianal wounds with water on a daily basis and after bowel movements.

Pain medication: analgesics (paracetamol 1000mg every 6h or novalgin 1000mg every 8h), rectal xylocaine gel if required, and laxatives (lactulose 10ml every 12h).

MM (n=61)

under general anaesthesia

Removing haemorrhoidal tissue via an upside-down V-shaped incision on the anal dermis. A surgical wound with sharp edges remained open at the end of the procedure.

Postoperatively: advice to clean perianal wounds with water on a daily basis and after bowel movements.

Pain medication: analgesics (paracetamol 1000mg every 6h or novalgin 1000mg every 8h), rectal xylocaine gel if required, and laxatives (lactulose 10ml every 12h).

Length of follow-up:

Median 5.93 (range 2.83-10.68) years

Loss-to-follow-up:

Intervention:

N (%)

Reasons (describe)

Incomplete outcome data:

5 patients were excluded, because they had surgery elsewhere. 3 patients were excluded because of incomplete data collection.

I vs. C

Pain (VAS)

Week 1

3.1 vs. 6.2, p<0.001

Week 2

0.5 vs. 3, p<0.001

Week 4

0.05 vs. 0.6, p<0.001

Recurrence, n (%)

Year 2 (n=122):

2 (3.3) vs. 4 (6.6) p=0.402

Year 3 (n=120)

2 (3.3) vs 3 (4.9) p=0.648

Year 4 (n=101)

3 (4.9) vs. 3 (4.9) p=1.00

Year 5 (n=76)

4 (6.6) vs. 4 (6.6) p=1.00

Total

11 (18) vs. 14 (23) p=0.65

Prolonged wound healing,

n (%) at 4 wks:

1 (1.6%) vs. 6 (9.8%), p=0.052

Itching and burning, n (%)

Week 4

3 (4.9) vs. 12 (19.7), p<0.001

Incontinence (Cleveland clinic incontinence score), n (%):

Week 4

0 vs. 0

2 years (n=122)

1 (1.6) vs. 1 (1.6)

3 years (n=120)

1 (1.6) vs 0 p=0.315

4 years (n=101)

2 (3.3) vs 1 (1.6) p=0.315

5 years (n=76)

2 (3.3) vs. 1 (1.6) p=0.559

No incontinence for liquid or solid stools occurred during follow-up.

Postoperative bleeding, n (%) at 4wks

3 (4.9%) vs. 3 (4.9%) p=0.309

Bleeding episodes

Minimal bleeding occurrence after defectation if laboratory findings found regular haemoglobin level and recurrence was excluded at anoscopy, n(%)

2 years

vs. 2 (3.3) vs 4 (6.6) p=0.402

3 years

3 (4.9) vs. 3 (4.9) p=1.00

4 years

3 (4.9) vs, 3 (4.9) p=1.00

5 years

3 (4.9) vs. 3 (4.9) p=1.00

Total

11 vs. 13

Author conclusion:

This study shows comparable long-term results for both procedures but suggests increased patient comfort with significantly less pain, burning and itching in the postoperative course after SH. We consider SH being as effective as the MM procedure and recommend its use for symptomatic circumferential third-degree haemorrhoidal disease on the basis of its short- and long-term benefits.

Ammaturo, 2012

Type of study:

RCT

Setting:

Hospital

Country:

Italy

Source of funding: not stated

Inclusion criteria:

Symptomatic grade 3 haemorrhoids

Exclusion criteria:

Acute thrombosed, piles, other concomitant anal disease, previous anal surgery, haematological disorders, ongoing treatment with oral anticoagulants, ASA 3 or 4

N total at baseline: 80

Intervention: 40

Control:40

Important prognostic factors²:

median age (range):

I: 45 (20-71)

C:48 (24-74)

Sex m/f:

I: 23/17

C: 25/15

Groups comparable at baseline? yes

SH under spinal anaesthesia

PPH stapling gun according to technique as described by longo. Actively spurting vessels were suture0ligated. Anal tampon was left in the rectum at the end of procedure for 4-5hrs to detect haemorrhage.

Without bowel preparation (only one enema 3hrs preoperatively). Short term prophylactic antibiotics given 3hrs preoperatively.

MM under spinal anaesthesia

Internal and external piles dissected entirely up to anorectal ring with scissors and pedicle was interrupted after suture-ligation. Haemostatic dressing was left in the anal canal et the end of the procedure for 4-5hrs.

Without bowel preparation (only one enema 3hrs preoperatively). Short term prophylactic antibiotics given 3hrs preoperatively.

Length of follow-up:

Median 35 months (range 24-51)

Loss-to-follow-up:

Intervention: 1 (2.5%)

Reasons: not described

Control: 0

Incomplete outcome data:

-

I vs. C

Bleeding during defecation, n(%)

6 months: 3 (7.5%) vs. 2 (5%)

2 years: 7 (18%) vs. 1 (2.5%)

Prolapse during defecation, n(%)

6 months: 3 (7.5) vs. 0

2 years: 5 (13%) vs. 0

Bleeding was severe and continuous during every defecation in 5/7 of SH group, wile light and sporadic in only 1 patient of MM group

Pain

VAS during first 10 days, mean (range)

2.5 (2-5) vs. 6.8 (3-9)

Reoperation, n(%) 2years:

5 (13%) vs. 0

Hospital stay, mean (range):

1.3 (1-2) vs. 2.5 (1-4)

Complications, n(%)

After 10 days:

Haemorrhage: 0 vs. 3 (7.5%)

Urinary retention: 7 (17.5%) vs. 16 (40%)

External haemorrhoidal thrombosis: 4 (10%) vs. 0

Anal sepsis: 0 vs. 0

Reoperation: 0 vs. 1 (2.5%)

Patient satisfaction (based on symptoms after surgery: 0=worsened, 1=unchanged, 2= ameliorated, 3=relieved), mean

6 months: 2.6 vs. 2.5

2 years: 2.06 vs. 2.8

Author conclusion: SH offers greater short term benefits with reduced pain, shorter length of hospital stay, earlier return to work and high patients satisfaction but it is less effective than MMH as a definitive cure of grade 3 haemorrhoids.

Arslani, 2012

Type of study:

RCT

Setting:

Hospital

Country:

Croatia

Source of funding: not stated

Inclusion criteria:

Newly diagnosed grade 3 haemorrhoidal disease.

Exclusion criteria: combined procedure for other anorectal disease, incarcerated or gangrenous haemorrhoids requiring emergency surgery, faecal incontinence, urinary retention, other anorectal disease, anal stenosis.

N total at baseline: 98

Intervention: 46

Control: 52

Important prognostic factors²:

Age median (range):

I: 52 (17-72)

C: 50 (18-78)

Sex %F:

I:25 (54.3)

C: 29 (55.8)

Groups comparable at baseline? yes

SH with PPH 0.

Preoperative: oral manitol day before, enema on day of surgery, IV fluids. Anaesthesia: general endotracheal anaesthesia.

LH with LigaSure

Preoperative: oral manitol day before, enema on day of surgery, IV fluids. Anaesthesia: general endotracheal anaesthesia.

Length of follow-up:

24 months

Loss-to-follow-up:

Not described

Incomplete outcome data:

Not described

I vs. C

Recurrence (defined as recurring symptoms or new prolapse, after 24 months).

5 vs. 1 NS

Pain (VAS), median, IQR

During first 5 days

3 (3-4) vs. 3 (3-5) p=0.358

Complications, n (%)

Urinary retention:

1 (2.2) vs. 2 (3.8) p=0.898

Early bleeding (48h):

3 (6.5) vs. 1 (1.9), p=0.504

Late bleeding (after48h)

3 (6.5) vs. 2 (3.8), p=0.883

Faecal incontinence

2 (4.3) vs. 1 (1.9) p=0.919

Anal stenosis

2 (4.3) vs. 1 (1.9) p=0.919

Overall complications

11 (23.9) vs. 7 (13.5) p=0.293

“It has been clearly shown that SH has a slightly higher recurrence than conventional excision techniques and offers more comfort to the patient. Also that it has a higher change of recurrence than conventional haemorrhoidectomy, and that it is the patients choice to decide whether to accept a higher change of recurrence for higher immediate postoperative comfort and faster return to normal activity. Today SH is by most authors considered unsuitable for grade 4 haemorrhoids because of the high recurrence.”

Author conclusion:

We conclude that SH and LH can be used to treat grade 3 haemorrhoidal disease with the same efficacy.

Sakr, 2010

Type of study:

RCT

Setting:

Hospital

Country:

Kuwait

Source of funding: not stated

Inclusion criteria:

Symptomatic grade 3 or 4 haemorrhoids

Exclusion criteria: complicated (thrombosed or strangulated) haemorrhoid, coexisting perianal disease, previous perianal surgery, compromised patient condition (unfit for general anaesthesia), pregnancy, history of bleeding tendency.

N total at baseline: 68

Intervention: 34

Control: 34

Important prognostic factors²:

age ± SD:

I: 43.7±4.66

C: 39.3±4.68

P=0.718

Sex: % M

I: 21 (61.8%)

C: 19 (55.9%)

Grade 3/4, n(%)

I: 27 (79.4)/ 7 (20.6)

C:24(70.6)/10(29.4)

Groups comparable at baseline? yes

SH

Preoperative: enema before surgery. General or spinal/epidural anaesthesia. No routine antibiotics.

Postoperative: oral analgesic (acet-aminophen) every 4hrs, intramuscular injection pethidine (1mg/kg) as requested. Bulk-forming agent (psyllium) twice daily. Patients were not discharged until pain was effectively managed (<4VAS) with oral analgesics.

LH

Preoperative: enema before surgery. General or spinal/epidural anaesthesia. No routine antibiotics.

Postoperative: oral analgesic (acet-aminophen) every 4hrs, intramuscular injection pethidine (1mg/kg) as requested. Bulk-forming agent (psyllium) twice daily. Patients were not discharged until pain was effectively managed (<4VAS) with oral analgesics.

Length of follow-up:

12 months

Loss-to-follow-up:

Not stated

Incomplete outcome data:

Not stated

I vs. C

Recurrence at 12 months:

4 (11.8) vs. 1 (2.9) p=0.163

Pain (VAS), mean±sd

Day 1:

5.29±0.914 vs. 5.53±1.02 p=0.260

Day 2:

3.9±0.06 vs. 3.7±0.07

P=0.215

Day 14:

1.8±0.1 vs. 1.6±0.08 p=0.237

Complications

Urine retention, haemorrhage, wound sepsis, gas incontinence

Within 2 weeks after operation

At 2 weeks:

Constipation: 5 (14.7) vs. 6 (17.6) p=0.742

Urine retention:

1 (2.9) vs. 2 (5.9) p=0.555

Haemorrhage:

2 (5.9) vs. 1 (2.9) p=0.555

Wound infection

0 vs. 1 (2.9) p=0.314

Gas incontinence:

4 (11.8) vs. 2 (5.9), p=0.393

Residual prolapse:

8 (23.5) vs. 2 (5.9) p=0.040

At 12 weeks:

Anal stenosis

2 (5.9) vs. 1 (2.9) p=0.555

Anal fissure

2 (5.9) vs. 1 (2.9) p=0.555

Author conclusion:

“both LigaSure haemorrhoidectomy and SH yield comparable goof results with a short operative time and minimal side effects in the treatment of grade 3 and 4 haemorrhoids, but with a lower rate of residual prolapse for the LigaSure procedure… influence of degree of severity of haemorrhoidal disease (grade 3 or 4) should be investigated.

Open versus closed haemorrhoidectomy

Shaik, 2013

Type of study:

RCT

Setting:

Hospital

Country:

Pakistan

Source of funding: not stated

Inclusion criteria:

Late 2th degree not responding to conservative treatment, or grade 3 or 4 haemorrhoids.

Exclusion criteria: complicated, secondary, external haemorrhoids or associated with anorectal disorders.

N total at baseline: 213

Intervention: 110

Control: 103

Important prognostic factors:

age ± SD:

I: 45.13±5.6

C: 46.20±4.5

Sex, (%) M

I: 81.8%

C: 77.6%

Degree of haemorrhoids

2/3-4, n(%)

I: 27 (24.54)/83 (75.45)

C: 24 (23.3)/79 (76.6)

Groups comparable at baseline? yes

Open (MM) under spinal or general anaesthesia, with diathermy. Anal canal was plugged.

Preoperative: enema at night and morning before surgery.

Closed (Ferguson) under spinal or general anaesthesia with 2/0 polyglactin; wound was closed with 3/0 polyglactin.

Preoperative: enema at night and morning before surgery.

Length of follow-up:

Until wounds had completely healed (no mean/median reported)

Loss-to-follow-up:

Not stated

Incomplete outcome data:

Not stated

I vs. C

Pain (linear analogue scale first 24h), mean±sd:

6.12±1.5 vs. 5.0±0.02 p<0.05

Recurrence

2 (1.8) vs. 1 (0.97), NS

Complications, n(%)

Urine retention:

13 (11.8) vs. 4 (3.88) p=0.01

Reactionary haemorrhage:

4 (3.63) vs, 0, NS

Anal stenosis:

0 vs. 3 (2.91%), NS

Anal incontinence:

1 (0.9%) vs. 0, NS

Wound infection:

1 (0.9%) vs. 2 (1.9%), NS

Author conclusion: “closed haemorrhoidectomy offered an advantage regarding postoperative pain, bleeding control, early mobility and rapid wound healing”.

Pokharel, 2009

Type of study:

RCT

Setting:

hospital

Country:

Kathmandu

Source of funding: not stated

Inclusion criteria:

Symptomatic grade 3 or 4 haemorrhoids for which conservative methods had failed.

Exclusion criteria: none stated

N total at baseline: 56

Intervention: 28

Control: 28

Important prognostic factors:

Groups nog compared

33 men, 23 women

Age, mean: 50 years

Groups comparable at baseline?

MM under spinal anaesthesia

Diathermy was used. Wounds were left open. Anal canal was plugged.

Ferguson under spinal anaesthesia. Ligation of the haemorrhoidal pedicle. Mucosa was closed with 3/0 vicryl once haemostasis was attained. Anal canal packing was done.

Length of follow-up:

6 months

Loss-to-follow-up:

Not stated

Incomplete outcome data:

Not stated

I vs. C

Pain (VAS), mean±sd:

Directly postoperative: 7.8±0.7 vs. 4.9±0.7, p<0.05

After 24h:

6.9±0.8 vs. 2.3±0.8, p<0.05

After 1 month:

3.1 vs. 2.1 NS

Complications

not clearly reported on

Very limited data representation, therefore limited use of the study. Only Pain is usable. Overall study quality and description is very limited.

Author conclusion: “we found important advantages in using the closed technique over de MM technique. These included less pain in the immediate postoperative period and 24h later, fewer pills needed to control pain, better and faster wound healing. Disadvantage is longer operation time. We think that the use of closed haemorrhoidectomy for the treatment of 3 and 4the degree haemorrhoids brings the advantage in our hospital set up and in our population”

Mik, 2008

Type of study:

RCT

Setting:

Hospital

Country:

Poland

Source of funding: not stated

Inclusion criteria:

Symptomatic grade 4 haemorrhoids.

Exclusion criteria:

Previous surgery for anal disease, inflammatory bowel diseases.

N total at baseline: 78

Intervention: 26

Control: 42

Important prognostic factors:

age ± SD:

I: 53±12

C:49±9

Sex (F/M):

I:16/16

C: 15/19

Groups comparable at baseline? yes

Milligan Morgan under subarachnoidal anaesthesia.

Piles were removed using electro cautery

External dressings postoperatively

Preoperative: admitted day before surgery. 15ml paraffin for 2 weeks.

Ferguson under subarachnoidal anaesthesia.

Piles were removed using electro cautery. After ligating mucosa and anoderm were stitched using a running 3-0 chromic catgut suture.

Preoperative: admitted day before surgery. 15ml paraffin for 2 weeks.

Length of follow-up:

18 months

Loss-to-follow-up:

Intervention: 7 (18%)

Reasons: not described

Control: 8 (20.5%)

Reasons: not described

Incomplete outcome data:

Intervention:

N (%)

Reasons (describe)

Control:

N (%)

Reasons (describe)

I vs. C

Recurrence

After 6 months

2 (5.9%) vs. 1 (2.9%) (treated by RBL), p=0.59

After 18 months

0 vs. 1 (2.9%), p=1

Overall:

2 (6.9%) vs. 2 (5.8%)

Pain

Severe (>7 VAS) postoperative pain:

18 (62.1%) vs. 20 (64.7%)

Bleeding

After 2 weeks:

7 (24.1%) vs. 10 (29.4%), p=0.77

After 1 month:

5 (17.2%) vs. 3 (8.8%)

After 6 months:

1 (3.4%) vs. 2 (5.9%), p=0.59

After 18 months:

1 (3.4%) vs. 1 (2.9%), p=0.3

Pruritis

After 2 weeks:

5 (17.2) vs. 9 (26.5%) p=0.54

After 18 months:

2 (6.9%) vs. 0, p=0.22

Wound infection

after 2wks:

0 vs. 4 (11.8%)

After 1 month

0 vs. 0

Complications

Urine retention (postoperative

): (17.2%) vs. 7 (20.6%)

Long term complications

Anal stricture

1 (3.4%) vs. 0

Incontinence (gas and faecal)

At 6 months:

0 vs. 0

At 18 months:

0 vs. 2 (5.9%) (mild to gas)

Soiling (18 months):

0 vs. 1

High loss-to-follow-up, which is not accounted for. Strange: patients are randomized

Hadi, 2011

Type of study:

RCT

Setting:

Hospital

Country: Pakistan

Source of funding: not stated

Inclusion criteria:

2nd, 3rd and 4th degree haemorrhoids and having no systemic illness

Exclusion criteria:

recurrent and thrombosed haemorrhoids or those having associated diseases like perianal abscess, fistulae, fissures, ulcerative colitis and rectal cancer

N total at baseline: 50

Intervention:

Control:

Important prognostic factors²:

Not described for groups seprately

Groups comparable at baseline? Not clear

Open haemorrhoidectomy under general anaesthesia

Postoperatively:

Observation in surgical ward for 24 hours. high fiber diet, sitz baths with warm water, antibiotic ointment with local anaesthetic, regular analgesia

Closed haemorrhoidectomy under general anaesthesia

Postoperatively:

Observation in surgical ward for 24 hours. high fiber diet, sitz baths with warm water, antibiotic ointment with local anaesthetic, regular analgesia

Length of follow-up:

2 months

Loss-to-follow-up:

Not described

Incomplete outcome data:

Not described

I vs. C

Pain:

Mild: 13 (52%) vs. 8 (32%)

Moderate: 11 (44%) vs. 10 (40%)

Severe: 6 (24%) vs. 2 (8%)

Bleeding:

Early:

15(60%) vs. 6 (24%)

Late:

2 (8% vs. 1 (4%)

Overall unclear description of outcomes, blinding procedure, baseline characteristics etc.

RBL vs. CH

Asghar Khan, 2013

Type of study:

RCT

Setting:

Hospital

Country:

Pakistan

Source of funding: not stated

Inclusion criteria:

>14 years, 2^nd and 3^rd degree haemorrhoids.

Exclusion criteria: anal fissure, large skin tag, perianal abscess, thrombosed piles, pregnant and cirrhotic liver disease.

N total at baseline: 120

Intervention: 60

Control: 60

Important prognostic factors²:

age ± SD:

I: 39.80±16.25

C: 40.16±17.54

P=0.97

Sex:

I: 83.3 % M

C:80 % M

P=0.34

Grade 2/3

I: 56.7%/43%

C: 51.7%/48.3%

P=0.583

Groups comparable at baseline? yes

Milligan Morgan haemorrhoidectomy

Rubber band ligation

2 rubber bands at base of each pile. Maximum of 3 piles were banded in one session.

Length of follow-up:

6 months

Loss-to-follow-up:

Not stated

Incomplete outcome data:

Not stated

I vs. C

Recurrence (after 6 months)

2 (3.3%) vs. 8(13.3%), p=0.048

Pain (% severe pain postoperatively)

55 (91.7%) vs. 9 (15%), p=0.001

Bleeding

Immediate

11 (18.3%) vs. 4 (6.7%), p=0.053

Secondary (5-10 days postoperatively):

1.7% vs. 10.0% p=0.051*

*bleeding was mild and stopped spontaneously

Incontinence

3 (5%) vs. 0%, p=0.079*

*mild flatus incontinence which settled without surgical intervention. No faecal incontinence.

Anal stenosis

5 (8.3%) vs. 0 (0%), p=0.022

Satisfaction

56 (93.3%) vs. 52 (86.7%), p=0.224

Author conclusion: RBL has comparatively less post-operative complications. It is safe, effective but the satisfaction rate of patients in our society is not comparable to open haemorrhoidectomy in short term due to repeated procedure.

THD vs. SH (PPH)

Verre, 2013

Type of study:

RCT

Setting:

Hospital

Country:

Italy

Source of funding: not stated

Inclusion criteria:

Symptomatic essential haemorrhoids degree 3 or 4.

Exclusion criteria:

-

N total at baseline: 122

Intervention: 63

Control: 59

Important prognostic factors:

age:

I: 47.8

C: 48.9

Sex (m/f)

I: 24/39

C: 22/37

Degree (3/4):

I: 25 (39.6%)/ 38 (60.3%)

C: 27 (45.7%)/ 35(59.3%)

Groups comparable at baseline? yes

THD under epidural/caudal anaesthesia

Preoperative: Fleet; metronidazole 500mg IV.

Postoperative: acetaminophen 1000mg every 6 hours the first 3 days + ketorolac 30mg as rescue dose. Fiber rich diet with administration of psyllium fiber + at least 2lt of water/day.

SH under epidural/caudal anaesthesia

Preoperative: Fleet; metronidazole 500mg IV.

Postoperative: acetaminophen 1000mg every 6 hours the first 3 days + ketorolac 30mg as rescue dose. Fiber rich diet with administration of psyllium fiber + at least 2lt of water/day.

Length of follow-up:

3 months

Loss-to-follow-up:

Intervention:

4 (6.8%)

Reasons: failed to attend follow-up visit

Control:

6 (9.5%)

Reasons: failed to attend follow-up visit

Incomplete outcome data:

-

I vs. C

Postoperative bleeding, n (%)

0 vs. 5 (7.9%), p=NS

Pain, n (%)

Postoperative pain (n(%) of patients with >5 on VAS scale postoperatively):

4 (6.7%) vs. 10 (15.8%), p=NS

At rest after 1 week, n (%):

4 (6.7%) vs. 7 (11.1%) NS

At rest after 1 month, n (%)

1 (1.6%) vs. 6 (9.5%) NS

At rest after 3 months, n (%)

0 vs. 3 (5.2%) NS

Complications (at 3 months)

Tenesmus:

1 (1.8%) vs. 3 (5.2%)

Soiling:

2 (3.6%) vs. 3 (5.2%)

Itching:

0 vs. 3 (5.2%)

Follow-up to short, and important outcomes are not reported.

Infantino, 2013

Type of study:

RCT

Setting:

Multi-centre

Country:

Italy

Source of funding: not stated

Inclusion criteria:

Symptomatic 3^rd degree haemorrhoids, 18-70 years, mentally capable of understanding and signing informed consent, no previous proctologic operations or pelvic-perianal radiotherapy

Exclusion criteria:

Faecal incontince, outlet obstruction (Wexner score and ODS), contemporaneous presence of other anorectal disease, irritable bowel syndrome, inflammatory bowel disease, neurological disease affecting anal sensation, chronic anal pain, full-thickness rectal prolapse, pregnancy, immunosuppression, and/or coagulation defects.

N total at baseline:172

Intervention: 85

Control: 87

Important prognostic factors:

age ± SD:

I:47.6±11.9

C: 46.2±11.5

Sex (M/F):

I: 58/27

C: 58/26

Anaesthesia used

General: 18 vs. 15

Spinal: 42 vs. 63

Local+sedation: 12 vs.0

Local: 5 vs. 0

Pudendal: 8 vs. 56

Groups comparable at baseline? yes

THD under loco regional, spinal or general anaesthesia (surgeons and patients preference)

Preoperative: phosphate enema, no antibiotic prophylaxis.

Postoperative: stool softeners and oral analgesic as required.

SH under loco regional, spinal or general anaesthesia (surgeons and patients preference)

Preoperative: phosphate enema, no antibiotic prophylaxis.

Postoperative: stool softeners and oral analgesic as required.

Length of follow-up:

Mean 17±0.4 months

Loss-to-follow-up:

Not described

Incomplete outcome data:

Intervention:

-

Control:

3

Reasons: refusing the operation after randomization

I vs. C

Pain

Postoperative pain (VAS):

20.2 ±4.9 vs. 24.1±5.0 NS

After 1 week:

Only shown in figure, no statistically significant diff.

Postoperative bleeding, n(%)

5 (5.9) vs. 3 (3.6) NS

Recurrence (or persistence) at final follow-up:

12 (14%) vs. 6 (7.1%) p=0.22

Further treatment needed

10 (11.3%) vs. 6 (7.1%) p=0.44

Incontinence (Wexner score):

Only figure is presented in article: no significant differences

Quality of life (measured by VAS-score): improved in equal extent

Complications

Early (<30 days postoperatively)

Urinary retention: 5(5.9) vs. 12 (14.3)

Thrombosis: 2 (2.4) vs. 3 (3.6)

Haematoma: 5 (5.9) vs. 4 (4.8)

Late

Abscess: 0 vs. 1

Faecal urgency: 0 vs. 1

Pain >30 days: 0 vs. 1

Obstructed defecation: 0 vs. 1

Total: 0 vs. 6 p=0.028

Author conclusion: both minimally invasive techniques are effective in treating 3^rd degree haemorrhoids. THD seems to be associated with a better cost-effective ratio, better functional outcome and no late complications, while differences in postoperative pain and recurrence between the two procedures did not reach significance.

Ho, 2007

Yes

Yes

No

No * characteristics of the intervention not clearly described

Not applicable

Unclear* stated that quality is assessed, however no reportation of quality

Unclear * outcomes are differently defined and reported (mean, median), but still pooled.

Unclear

No

Sajid, 2012

Yes

Yes

No

No

Not applicable

Unclear* stated that quality is assessed, however no reportation of quality

Unclear (not described how outcomes in studies were reported/ when/defined)

No

No

*Cochrane reviews are not assessed.

Study reference

(first author, publication year)

Describe method of randomisation¹

Bias due to inadequate concealment of allocation?²

(unlikely/likely/unclear)

Bias due to inadequate blinding of participants to treatment allocation?³

(unlikely/likely/unclear)

Bias due to inadequate blinding of care providers to treatment allocation?³

(unlikely/likely/unclear)

Bias due to inadequate blinding of outcome assessors to treatment allocation?³

(unlikely/likely/unclear)

Bias due to selective outcome reporting on basis of the results?⁴

(unlikely/likely/unclear)

Bias due to loss to follow-up?⁵

(unlikely/likely/unclear)

Bias due to violation of

intention to treat analysis?⁶

(unlikely/likely/unclear)

SH vs. CH

Kim, 2013

Computer randomized

Unlikely

Unlikely

Unlikely (no blinding)

Unlikely (no blinding)

Unlikely

Unlikely

Unclear (no ITT)

Ammaturo, 2012

Not described

Unclear

Unlikely (no blinding)

Unlikely (no blinding)

Unlikely (no blinding)

Unlikely

Unlikely

Unlikely

Arslani, 2012

Slips of papers from the box method

Unlikely

Unlikely (no blinding)

Unlikely “

Unlikely (no blinding)

Unlikely

Unclear

Unclear

Sakr, 2009

Computer generated schedule

Unlikely

Unlikely (blinding)

Unlikely

Unlikely (blinding)

Unlikely

Unclear

Unlikely

Open vs. Closed

Shaikh, 2013

Not described

Unclear

Unlikely (no blinding)

Unclear (no blinding)

Unclear (no blinding)

Unlikely

Unclear

Unclear

Pokharel, 2009

By alternate patients

Unclear

Unlikely

Unclear (no blinding)

Unclear (no blinding)

Likely (not all outcomes are well reported on)

Unclear (not described)

Unlikely

Mik, 2008

Computer generated codes

Unclear

Unclear

Unlikely

Unlikely

Unlikely

Likely

Likely (high loss to follow-up, no ITT)

Hadi, 2011

Odd and even numbers

Unclear

Unclear

Unclear

Unclear

Unclear

Unclear

Unclear

CH vs. RBL

Khan, 2013

Not described

Unclear

Unlikely (no blinding)

Unclear (no blinding)

Unclear (no blinding)

Unlikely

Unclear

Unclear

THD vs. CH

Elmer, 2013

Sealed envelopes

unlikely

Unclear (not blinded)

Unclear (not blinded)

Unclear (not blinded)

Unlikely

Unlikely

Unlikely

Denoya, 2013

Computed-generated randomization scheme

Unlikely

Unlikely

Unlikely

Unlikely

Likely (not clearly reported on!)

Unlikely

Unlikely

THD vs. SH

Verre, 2013

Not described

Unclear

Unlikely (no blinding)

Unclear (no blinding)

Unclear (no blinding)

Unlikely

Unclear

Unclear (no ITT)

Infantino, 2013

Computer generated

Unlikely

Unlikely (no blinding)

Unlikely (no blinding)

Unlikely (no blinding)

Unlikely

Unlikely

Unlikely (no ITT)