Table 2.
GRADE evidence profile; quality assessment per outcome
| Certainty assessment | № of patients | Effect | Certainty | Importance | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| № of studies | Study design | Risk of bias | Inconsistency | Indirectness | Imprecision | Other considerations | Rubber band ligation | Hemorrhoidectomy | Relative (95% CI) | Absolute (95% CI) | ||
| Urinary retention | ||||||||||||
| 6 | Randomised trials | Seriousa,b | Not serious | Seriousc | Not serious | None | 15/527 (2.8%) | 107/527 (20.3%) |
RR 0.15 (0.09 to 0.25) |
173 fewer per 1.000 (from 185 to 152 fewer) |
⨁⨁◯◯ LOW |
Critical |
| Postoperative pain | ||||||||||||
| 7 | Randomised trials | Seriousa,b,d,e | Not serious | Seriousc | Not serious | None | 68/557 (12.2%) | 438/553 (79.2%) |
RR 0.17 (0.11 to 0.28) |
657 fewer per 1.000 (from 705 to 570 fewer) |
⨁⨁◯◯ LOW |
Critical |
| Postoperative bleeding | ||||||||||||
| 7 | Randomised trials | Seriousa,b,d | Not serious | Seriousc | Seriousf | None | 22/557 (3.9%) | 84/553 (15.2%) |
RR 0.31 (0.15 to 0.66) |
105 fewer per 1.000 (from 129 to 52 fewer) |
⨁◯◯◯ VERY LOW |
Critical |
| Anal stenosis | ||||||||||||
| 5 | Randomised trials | Seriousa,b,d | Not serious | Seriousc | Not serious | None | 1/470 (0.2%) | 26/472 (5.5%) |
RR 0.11 (0.03 to 0.38) |
49 fewer per 1.000 (from 53 to 34 fewer) |
⨁⨁◯◯ LOW |
Critical |
| Recurrence | ||||||||||||
| 4 | Randomised trials | Seriousa,b,d,g | Serioush | Seriousc | Seriousf | None | 50/161 (31.1%) | 10/161 (6.2%) |
RR 4.77 (2.60 to 8.76) |
234 more per 1.000 (from 99 to 482 more) |
⨁◯◯◯ VERY LOW |
Critical |
| Anal incontinence | ||||||||||||
| 3 | Randomised trials | Serious a,b,d | Not serious | Seriousc | Seriousf | None | 0/120 (0.0%) | 5/116 (4.3%) |
RR 0.16 (0.02 to 1.28) |
36 fewer per 1.000 (from 42 fewer to 12 more) |
⨁◯◯◯ VERY LOW |
Critical |
CI Confidence interval; MD Mean difference; RR Risk ratio
aLack of allocation concealment and lack of blinding in all studies. This, however, is unavoidable in most surgical RCTs
bIncomplete accounting of patients and outcome events. Of all studies, only three mentioned the loss to follow-up
cSaeed, Murie, Bakhtawar, Khan, Ali not mentioned prespecified primary and secondary outcomes
dNo information on how the randomisation sequence was generated
eUnclear how postoperative pain was scored
fSeveral outcomes were reported in few studies and few patients and few events
gUnclear how patient satisfaction was defined
hUnclear how recurrence was defined