Table 2. Evidence Summary GRADE (1993–2018).
| Outcome | Absolute effects Anticipated | Relative Effect (95%CI) | Nº of participants (Studies) | Quality of Evidence (GRADE) | |
|---|---|---|---|---|---|
| LEEP risk | Cold cone risk | ||||
| Question: Should Cold Cone vs. LEEP be used for treatment of Cervical Intraepithelial Neoplasia (CIN)? Bibliography: Girardi et al. 1994; Mathevet et al., 1994; Duggan et al., 1999; Takac et al. 1999; Giacalone et al., 1999; Mathevet et al., 2003 | |||||
| Disease Recurrence (monitoring: range 6 months to 118 months) | 71 per 1,000 | 23 per 1,000 (6 to 81) | RR 0.32 (0.09 to 1.14) | 287 (3 ECAs) | ⨁◯◯◯ VERY LOW a,b |
| Residual Disease: (monitoring: up to 6 months) | 112 per 1,000 | 61 per 1,000 (34 to 108) | RR 0.54 (0.30 to 0.96) | 529 (4 ECAs) | ⨁◯◯◯ VERY LOW c,d |
| Positive Margins | 212 per 1,000 | 164 per 1,000 (115 to 232) | RR 0.77 (0.54 to 1.09) | 553 (4 ECAs) | ⨁◯◯◯ VERY LOW c,d,e |
| Minor bleeding during the first 24 hours after treatment | 54 per 1,000 | 57 per 1,000 (27 to 119) | RR 1.05 (0.50 to 2.21) | 469 (4 ECAs) | ⨁◯◯◯ VERY LOW c,f |
| Minor bleeding after the first 24 hours after treatment | 88 per 1,000 | 83 per 1,000 (36 to 187) | RR 0.94 (0.41 to 0.13) | 247 (2 ECAs) | ⨁◯◯◯ VERY LOW c,f |
| Cervical Stenosis (monitoring: range 3 months to 24 months) | 66 per 1,000 | 70 per 1,000 (29 to 169) | RR 1.06 (0.44 to 2.58) | 529 (4 ECAs) | ⨁◯◯◯ VERY LOW c,f |
|
| |||||
| Outcome | Absolute effects Anticipated | Relative Effect (95%CI) | Nº of participants (Studies) | Quality of Evidence (GRADE) | |
| LEEP risk | Cryotherapy Risk | ||||
| Question: Should Cryotherapy vs. LEEP be used for treatment of Cervical Intraepithelial Neoplasia (CIN)? Bibliography: Mitchell et al., 1998; Chirenje et al., 2001 | |||||
| Disease Recurrence (monitoring: range 6 to 24 months) | 77 per 1,000 | 144 per 1,000 (90 to 229) | RR 1.86 (1.16 to 2.97) | 598 (2 ECAs) | ⨁◯◯◯ VERY LOW a,b |
| Residual Disease: (monitoring: up to 6 months) | 37 per 1,000 | 65 per 1,000 (31 to 133) | RR 1.75 (0.85 to 3.60) | 596 (2 ECAs) | ⨁◯◯◯ VERY LOW a,b |
| Minor bleeding during the first 24 hours after treatment | 15 per 1,000 | 4 per 1,000 (1 to 25) | RR 0.27 (0.04 to 1.62) | 669 (2 ECAs) | ⨁◯◯◯ VERY LOW c,d |
| Minor bleeding after the first 24 hours after treatment | 484 per 1,000 | 237 per 1,000 (194 to 286) | RR 0.49 (0.40 to 0.59) | 625 (2 ECAs) | ⨁⨁⨁◯ MODERATE c |
| Cervical Stenosis: (monitoring: up to 24 months) | 3 per 1,000 | 6 per 1,000 (1 to 68) | RR 1.87 (0.17 to 20.38) | 596 (2 ECAs) | ⨁◯◯◯ VERY LOW a,d |
| Pain after 24h post treatment | 275 per 1,000 | 256 per 1,000 (204 to 322) | RR 0.93 (0.74 to 1.17) | 625 (2 ECAs) | ⨁⨁◯◯ LOW e,f |
| Infecciones después de 24h post tratamiento | 465 per 1,000 | 544 per 1,000 (502 to 595) | RR 1.17 (1.08 to 1.28) | 625 (2 ECAs) | ⨁◯◯◯ VERY LOW a,b |
GRADE (Degrees of Certainty of Evidence); LEEP: Loop Electrosurgical Excision Procedure
High certainty: we are very sure that the real effect is similar to the estimation of the effect
Moderate certainty: we are moderately confident in the estimate of the effect: the actual effect is likely to be close to the estimate of the effect, but there is a possibility that it will be substantially different
Low certainty: our confidence in the estimate of the effect is limited: the actual effect may be substantially different from the estimate of the effect
Very low certainty: we have very little confidence in the estimate of the effect: the actual effect is likely to be substantially different from the estimate of the effect
a. It was decided to decrease two levels due to High Risk of Wearing Bias (both studies had more than 10% losses)
b. It was decided to decrease one level in imprecision because of the wide CI crossing the 1.25 limit.
c. It was decided to decrease one level for Uncertainty Risk of detection bias (uncertainty in blinding of assessors)
d. It was decided to decrease two levels in inaccuracy because of the wide CI crossing the limits of 0.75 and 1.25 and because of the small number of events.
e. It was decided to decrease one level for Uncertainty Risk of detection (blinding of assessors).
f. It was decided to decrease one level in imprecision because of the wide CI that crosses the 0.75 limit.