Summary of findings 4. Retreatment ‐ long term.

Minimally invasive treatments versus transurethral resection of the prostate
Patient or population: men with moderate to severe lower urinary symptoms due to benign prostatic hyperplasia Interventions: minimally invasive treatments. Comparator (reference): transurethral resection of the prostate Setting: hospital procedure – outpatient follow‐up
Outcome: retreatment Defined as: number of participants requiring a follow‐up procedure for lower urinary tract symptoms including another minimally invasive treatment or TURP (this does not include procedures to treat complications ‐ these are included under major adverse events) Follow‐up: 12 ‐ 60 months
10 studies 799 participants	Anticipated absolute effect (95% CI) *		Relative effect (95% CI)	Certainty of the evidence	Ranking (SUCRA) **
	With TURP	With a minimally invasive procedure
PUL (UroLift) (mixed estimate)	Median rate of retreatment: 12 per 1000a	17 more per 1000 (6 fewer to 121 more)	RR 2.39 (0.51 to 11.10)	⊕⊝⊝⊝ VERY LOW b c d	2.2 (68.8%)
PAE (mixed estimate)		41 more per 1000 (3 more to 173 more)	RR 4.39 (1.25 to 15.44)	⊕⊝⊝⊝ VERY LOW b d e	3.0 (50.8%)
TUMT (mixed estimate)		104 more per 1000 (16 more to 470 more)	RR 9.71 (2.35 to 40.13)	⊕⊕⊕⊝ LOW b d	3.7 (32.1%)
CRFWVT (Rezūm) (pairwise)	Based on one study with 197 participants, we are very uncertain about the effects of CRFWVT on retreatment compared to sham at three months follow‐up (RR 1.36, 95% CI 0.06 to 32.86).			⊕⊝⊝⊝ VERY LOW f	Data could not be included in NMA to preserve the transitivity of each network
TIND (pairwise)	Based on one study with 185 participants, we are very uncertain about the effects of TIND on retreatment compared to sham at three‐month follow‐up (RR 0.67, 95% CI 0.11 to 3.89).			⊕⊝⊝⊝ VERY LOW f	Data could not be included in NMA to preserve the transitivity of each network
CI: confidence interval; CRFWVT: convective radiofrequency water vapor therapy; IPSS: International Prostate Symptom Score; NMA: network meta‐analysis; QoL: quality of life; PAE: prostatic arterial embolization; PUL: prostatic urethral lift; RR: risk ratio; SUCRA: surface under the cumulative ranking curve; TIND: temporary implantable nitinol device; TUMT: transurethral microwave thermotherapy; TURP: transurethral resection of prostate. Network meta‐analysis summary of findings table definitions. * Estimates are reported as risk difference and confidence interval (CI). ** Rank statistics is defined as the probability that a treatment out of n treatments in a network meta‐analysis is the best, the second, the third, and so on until the least effective treatment. Between brackets are the surface under the curve (SUCRA) estimates.
GRADE Working Group grades of evidence (or certainty of the evidence). High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

^aAverage rate of retreatment in the control group (1.15%) or 12 per 1000. TURP was the highest rank intervention for this outcome with a mean rank of 1.1 (SUCRA 96.4%)

^bDowngraded by one level due to major concerns on within‐study bias: nearly all studies contributing to this estimate had an overall high risk of bias.

^cDowngraded by one level due to major concerns on imprecision: wide confidence interval.

^dDowngraded by one level due to major concerns on incoherence: the network does not present close loops to assess incoherence.

^eDowngraded by one level due to some concerns on imprecision and inconsistency (heterogeneity): wide confidence interval and prediction interval.

^fDowngraded by three levels due to concerns on within‐study bias (single study at high risk of bias) and severe imprecision (wide confidence interval).