Comparison of the efficacy and safety of transurethral laser versus open prostatectomy for patients with large-sized benign prostatic hyperplasia: A meta-analysis of comparative trials.

PURPOSE: The selection of open prostatectomy (OP) over transurethral laser surgery is controversial for large volume prostates. Thus, we aim to compare the efficacy and safety of transurethral laser versus OP, and provide the latest evidence of clinical practice for large-sized benign prostatic hyperplasia (BPH).
MATERIALS AND METHODS: This meta-analysis used Review Manager V5.3 software and the systematic literature search of Cochrane Library, Embase, PubMed, and Web of Science datasets was performed for citations published from 2000 to 2020 that compared transurethral laser with OP for the treatment of large BPH. Variables of interest assessing the two techniques included clinical characteristics, and the perioperation-, effectiveness-, and complication-related outcomes.
RESULTS: The meta-analysis included twelve studies containing 1,514 patients, with 792 laser and 722 OP. The transurethral laser group was associated with shorter hospital stay and catheterization duration, and less hemoglobin decreased in the perioperative variables. There was no significant difference in the international prostate symptom score, post-void residual urine volume, maximum flow rate, and quality of life score. Transurethral laser group had a significantly lower incidence of blood transfusion than OP group (odds ratio, 0.10; 95% confidence interval, 0.05 to 0.19; p<0.001; I²=8%), and no statistical differences were found with respect to the other complications.
CONCLUSIONS: Both OP and transurethral laser prostatectomy are effective and safe treatments for large prostate adenomas. With these advantages of less blood loss and transfusion, and shorter catheterization time and hospital stay, laser may be a better choice for large BPH. © The Korean Urological Association.

INTRODUCTION

Benign prostatic hyperplasia (BPH) is a chronic, progressive disease for elderly men. For patients over 70 years old, up to 4% have a prostatic volume more than 100 mL [1]. For some of patients with large BPH, surgical intervention is required when drug treatment (alpha-blockers, anticholinergics, etc. alone or combinations) is not effective. In addition, surgical intervention should also be considered for large-prostate patients who have moderate-to-severe lower urinary tract symptoms (LUTS) or who have complications from bladder outlet obstruction (BOO) due to BPH (renal insufficiency due to BOO, recurrent urinary tract infection [UTI], repeated gross hematuria, bladder stones, one or more episodes of urinary retention, etc.). Until recently, open prostatectomy (OP) has been considered for decades as the gold standard procedures for prostates larger than 80 mL [23]. However, OP brings considerable surgery-related complications, including more intraoperative bleeding, longer catheterization time and longer hospitalization [4]. Thus, more methods emerge endlessly, such as prostatic artery embolization, robot-assisted simple prostatectomy and transurethral laser surgery in managing large BPH [25].

Since holmium laser was initially reported for endoscopic resection of prostate in 1996 by Gilling et al. [6], transurethral laser surgery has become increasingly a popular option for the management of bothersome LUTS. The types of lasers have been a growing tendency to diversification and versatility for the treatment of BPH, and surgical approach has also been diversified (vaporization, resection, and enucleation, alone or combinations) recently. With the accumulation of surgical experience and proficiency in surgical techniques, the indications for transurethral laser prostatectomy have also expanded from conventional-volume to large-volume prostates. The efficacy and safety of transurethral laser prostatectomy compared with OP for patients with large prostates has been proven by several studies [278910].

With the development of laser technology and the improvement of surgical methods, the proportion of laser surgery for BPH is also increasing. Nevertheless, it remains debatable whether transurethral laser prostatectomy can replace OP as the gold standard of treatment for large prostates. Published studies on comparing these two surgical methods are still limited, especially for large BPH [1112131415]. So we made this meta-analysis comparing the efficacy and safety of transurethral laser surgery and OP to provide the latest evidence of clinical practice for surgical treatment of large-volume prostates.

MATERIALS AND METHODS

The protocol was registered in the ‘International prospective register of systematic reviews’ (https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021267469).

1. Search strategy

A bibliographic online search was from January 1, 2000 to December 31, 2020 on Cochrane Library, Embase, PubMed, and Web of Science databases, which compared two surgical methods (transurethral laser prostatectomy vs. OP) with the MESH search terms: “benign prostatic hyperplasia,” “laser therapy,” and “prostatectomy”. No language restriction was applied. We modified search strategy according to the requirements of each online database, and used “related articles” function to broaden online search.

2. Inclusion criteria and exclusion criteria

The certain criteria were defined ahead of search. The comparative trials that met the following inclusion criteria were included: (1) the literatures compared transurethral laser prostatectomy and OP; (2) enrolled patients with LUTS secondary to BPH preoperatively; (3) included OP performed by a traditional open approach (not robotic or laparoscopic); and (4) provided at least one indicator for analysis, such as hemoglobin decline, operation time, catheterization duration, hospital stay, micturition parameters, and the complications.

There were three exclusion criteria were used: (1) included prostate volumes less than 70 mL and did not mention a large prostate; (2) no parameters for comparative analysis in the results; (3) enrolled patients with vesical calculus, urethral stricture, or previous prostate surgery.

3. Data extraction

Two reviewers (H.B.W. and B.Y.G.) independently screened and evaluated the following parameters: (1) general data: journal, first author, year of publication, study design, study quality, number of patients, age of the patients, and prostate size; (2) perioperative variables: hemoglobin decline, operative time, catheterization time, and hospital stay; (3) effectiveness-related outcomes during the follow-up period: International Prostate Symptom Score (IPSS), maximum flow rate (Qmax), postvoid residual (PVR), quality of life score (QoL), and length of follow-ups; (4) surgical complications: blood transfusion rate, transitory urge incontinence, stress incontinence, bladder neck/urethral stenosis, UTI, surgical intervention for bleeding, and acute urinary retention (AUR).

4. Assessment of study quality

Jadad composite scale and Newcastle–Ottawa scale were used to score the methodological quality of the included randomized controlled trials (RCTs) and non-RCTs, respectively. The Newcastle–Ottawa scale was used to assess the quality of the initial design, content and ease of use. The score of each individual study quality was from 0 to 9 according to the star system of Newcastle–Ottawa scale. This procedure was independently performed by Y.F.T. and W.Z., and disagreements were resolved by D.H.Z.

5. Statistical analysis and meta-analysis

The data was calculated by using Review Manager V5.3 software (Cochrane Collaboration, Oxford, UK). Transurethral laser surgery was regarded as the experimental intervention, and the other group was regarded as the control intervention. Dichotomous variables and continuous variables were represented as odds ratio (OR) and weighted mean difference (WMD), with 95% confidence interval (CI). Heterogeneity was evaluated by the chi-squared test and p-value [16]. A fixed-effects model was applied to pool the results, and there was no evident heterogeneity if p>0.1 or I2<50%, otherwise, a random effects model was applied. Considering the limitations of the Cochrane software, we used corresponding statistical method to convert median and interquartile range into mean and standard deviation [1718]. A subgroup analysis of RCTs and non-randomised studies of the effects of interventions (NRSIs), was applied to probe possible discrepancies between groups. We also conducted subgroup meta-analyses of perioperative variables on the type of surgical technology: enucleation and non-enucleation. Two-sided tests with p<0.05 were considered to be statistically significant.

RESULTS

1. Study characteristics

As shown in Fig. 1, a total of 608 articles were preliminarily retrieved through electronic search, and 12 studies were enrolled finally [2819202122232425262728]. The characteristics of the enrolled studies are shown in Table 1, which includes 5 RCTs and 7 non-RCT studies (792 patients in the transurethral laser group and 722 patients in the OP group). One study was excluded due to lack of data available for analysis [29]. Through full-text analysis, the other 3 articles were considered to be the same study with different follow-up points, and consolidated data was analyzed with enrolled corresponding articles [303132].

Fig. 1

Flow diagram of studies identified, included, and excluded from the analysis.

Table 1

Preoperative characteristics of clinical trials included in the meta-analysis

Study (reference)	Year	Follow-up months	Publication type	Surgical type	No. of patients	Age (y)	Qmax (mL/s)	PSA (ng/mL)	QoL	IPSS	Prostate size (g)
Moody and Lingeman [23]	2001	3, 20	Non-RCT	Holep	10	74.8±3.6	NA	NA	NA	19±2.6	>100
				OP	10	71.1±1.3	NA	NA	NA	22.4±3.3	>100
Naspro et al. [28]	2006	1, 3, 12, 24	RCT	Holep	41	66.26±6.55	7.83±3.42	6.33±3.43	4.07±0.93	20.11±5.84	113.27±35.33
				OP	39	67.27±6.72	8.32±2.37	6.99±4.28	4.44±0.96	21.60±3.24	124.21±38.52
Salonia et al. [27]	2006	NA	RCT	Holep	34	67.4±6.7	8.9±4.2	8.0±8.4	4.6±1.0	19.6±7.0	113.8±37.0
				OP	29	68.0±6.4	8.4±2.4	7.2±4.7	4.4±1.0	21.6±3.5	121.0±34.9
Zhang et al. [26]	2007	3	RCT	Holep	32	72.1±6.3	6.1±2.9	NA	4.7±1.2	27.4±5.5	139.6±26.4
				OP	28	75.7±5.1	6.7±2.8	NA	4.6±1.0	25.1±6.4	157.2±35.1
Kuntz et al. [24]	2008	1, 3, 6, 12, 18, 24, 36, 48, 60	RCT	Holep	60	69.2±8.4	3.8±3.6	NA	NA	22.1±3.3	114.6±21.6
				OP	60	71.2±8.3	3.6±3.8	NA	NA	21.0±3.6	113.0±19.2
Alivizatos et al. [25]	2008	1, 3, 6, 12, 18	RCT	PVP	65	74 (67–80)	8.6 (6.7–10.5)	6.2 (3.1–8.44)	3 (2–4)	20 (15–22)	93 (85–100)
				OP	60	67.5 (65–74)	8 (5.8–10.2)	6.3 (2.9–8.6)	3 (2.25–4)	21 (16.25–23.75)	96 (86.2–100)
Elshal et al. [22]	2016	1, 3, 12, 24	Non-RCT	Holep	92	67.7±6.9	NA	9.4 (0.5–43)	NA	NA	166.7±49.7
				OP	91	68.3±7.1	NA	9.8 (0.2–57)	NA	NA	161.4±35.7
Lanchon et al. [21]	2018	28 (PVP)	Non-RCT	PVP	57	77±9	NA	7.3±5.4	4±1	18±6	103±25
		35 (OP)		OP	54	74±9	NA	10.1±7.9	4±1	20±7	142±58
Misraï et al. [8]	2018	1, 2, 6	Non-RCT	GreenLEP	204	68 (65–73)	NA	5 (3.2–8.5)	NA	NA	100 (80–120)
				OP	204	69 (64–76)	NA	5.9 (3.7–9)	NA	NA	90 (80–120)
Rosenhammer et al. [20]	2018	NA	Non-RCT	Holep	72	69.9±7.9	NA	7.32±4.37	NA	NA	112.9±25.8
				OP	72	70.8±6.6	NA	7.36±4.36	NA	NA	112.6±21.2
Enikeev et al. [19]	2019	3, 6	Non-RCT	ThuFlep	90	67.4±7.7	7.5±1.9	NA	3.9±0.8	21.8±1.1	127.8±40.2
				OP	40	67.0±1.0	7.9±2.8	NA	4.5±1.5	24.6±3.3	114.1±38.9
Nestler et al. [2]	2019	NA	Non-RCT	ThuVEP	35	71.2 (65.1–74.8)	NA	NA	4.2 (3–5)	20 (18–23)	90.8 (73–121)
				OP	35	70.6 (66.1–74.3)	NA	NA	5 (5–6)	23 (21–25)	95 (84–132)

Values are presented as number only, mean±standard deviation, or median (interquartile range).

OP, open prostatectomy; Qmax, maximum flow rate; PSA, prostate-specific antigen; QoL, quality of life score; IPSS, International Prostate Symptom Score; RCT, randomized controlled trial; Holep, holmium laser enucleation of prostate; NA, not available; PVP, photoselective vaporization of the prostate; GreenLEP, green laser enucleation of the prostate; ThuFLEP, thulium fiber laser enucleation of the prostate; ThuVEP, thulium vapoenucleation of prostate.

2. Outcomes of perioperative variables

No significant difference was observed in operative time (WMD, 5.71; 95% CI, -8.36 to 19.77; p=0.43; I2=97%), and resected prostatic weight (WMD, -2.37; 95% CI, -11.82 to 7.07; p=0.62; I2=86%). The experimental intervention group was associated with shorter hospital stay (WMD, -4.68; 95% CI, -5.51 to -3.86; p<0.001; I2=96%), shorter catheterization duration (WMD,-4.01; 95% CI, -4.91 to -3.10; p<0.001; I2=99%), and less hemoglobin decreased (WMD, -1.15; 95% CI, -1.43 to -0.87; p<0.001; I2=0%). Regarding the subgroup analysis of enucleation and non-enucleation, the perioperative variables had no obvious change compared with the overall analysis. These outcomes of perioperative variables are shown in Table 2.

Table 2

Overall analysis of perioperative variables comparing transurethral laser with OP

Perioperative outcome		Study	No. of patients (laser/OP)	WMD (95% CI)	p-value	Study heterogeneity				Favors
						χ2	df	I2 (%)	p-value
Operative time
	Total	[2,8,19,21,23-28]	628/559	5.71 (-8.36 to 19.77)	0.43	343.16	9	97	<0.001	None
	RCT	[24-28]	232/216	27.49 (16.54 to 38.44)	<0.001*	34.52	4	88	<0.001	OP
	NRSI	[2,8,19,21,23]	396/343	-14.71 (-25.78 to -3.65)	<0.001*	36.88	4	89	<0.001	Laser
	Enucleation	[2,8,19,23,24,26-28]	506/445	2.92 (-13.12 to 18.96)	0.72	213.77	7	97	<0.001	None
	Non-enucleation	[21,25]	122/114	16.34 (-7.41 to 40.08)	0.18	26.63	1	96	<0.001	None
Hemoglobin decrease
	Total	[2,22,24,25,28]	293/285	-1.15 (-1.43 to -0.87)	<0.001*	3.91	4	0	0.42	Laser
	RCT	[24,25,28]	166/159	-0.97 (-1.31 to -0.64)	<0.001*	0.13	2	0	0.94	Laser
	NRSI	[2,22]	127/126	-1.58 (-2.10 to -1.07)	<0.001*	0.00	1	0	0.96	Laser
	Enucleation	[2,22,24,28]	228/225	-1.19 (-1.50 to -0.87)	<0.001*	3.67	3	18	0.30	Laser
	Non-enucleation	[25]	65/60	-1.02 (-1.63 to -0.41)	0.001*	NA	NA	NA	NA	Laser
Resected prostate weight
	Total (Enucleation)	[2,8,19,20,22-24,26-28]	670/608	-2.37 (-11.82 to 7.07)	0.62	65.60	9	86	<0.001	None
	RCT	[24,26-28]	167/156	-10.24 (-16.54 to -3.93)	0.001*	7.29	3	59	0.06	Laser
	NRSI	[2,8,19,20,22,23]	503/452	-1.68 (-5.54 to 2.19)	0.40	53.16	5	91	<0.001	None
Catheterization day
	Total	[2,8,19,21,22,24-28]	710/640	-4.01 (-4.91 to -3.10)	<0.001*	664.82	9	99	<0.001	Laser
	RCT	[24-28]	232/216	-3.67 (-5.60 to -1.75)	0.0002*	458.40	4	99	<0.001	Laser
	NRSI	[2,8,19,21,22]	478/424	-4.30 (-5.21 to -3.38)	<0.001*	126.87	4	97	<0.001	Laser
	Enucleation	[2,8,19,22,24,26-28]	588/526	-4.14 (-5.21 to -3.07)	<0.001*	662.86	7	99	<0.001	Laser
	Non-enucleation	[21,25]	122/114	-3.62 (-4.41 to -2.83)	<0.001*	1.96	1	49	0.16	Laser
Hospital stay
	Total	[2,8,19,21-28]	720/650	-4.68 (-5.51 to -3.86)	<0.001*	274.35	10	96	<0.001	Laser
	RCT	[24-28]	232/216	-3.97 (-4.22 to -3.72)	<0.001*	200.50	4	98	<0.001	Laser
	NRSI	[2,8,19,22,23]	488/434	-4.43 (-4.56 to -4.29)	<0.001*	63.59	5	92	<0.001	Laser
	Enucleation	[2,8,19,22,24,26-28]	598/536	-4.98 (-5.97 to -3.98)	<0.001*	260.16	8	97	<0.001	Laser
	Non-enucleation	[21,25]	122/114	-3.59 (-4.00 to -3.18)	<0.001*	0.75	1	0	0.39	Laser

OP, open prostatectomy; WMD, weighted mean difference; CI, confidence interval; RCT, randomized controlled trial; NRSI, non-randomised studies of the effects of intervention; NA, not available.

*Statistically significant results (p<0.05).

3. Effectiveness-related outcomes during the follow-up period

The micturition parameters at 1 month, 3 months, 6 months, and 12 months postoperatively were collected and combined (Table 3). There were no significant differences between the transurethral laser and OP groups in postoperative IPSS, QoL, Qmax, and PVR in the follow-up periods. At 6 months postoperatively, the pooled Qmax of RCT group was statistically different, while the other parameters of subgroups showed no obvious change compared with the overall analysis.

Table 3

Overall analysis of postoperative efficacy comparing transurethral laser with OP

Postoperative outcome			Study	No. of patients (laser/OP)	WMD (95% CI)	p-value	Study heterogeneity				Favors
							χ2	df	I2 (%)	p-value
IPSS
	1 month (RCT)		[24,25,28]	163/152	0.50 (-0.94 to 1.93)	0.50	8.72	2	77	0.01	None
	3 months (RCT)		[24-26,28]	192/177	0.33 (-0.25 to 0.91)	0.27	2.67	3	0	0.44	None
	6 months
		Total	[8,19,24,25]	413/240	-0.16 (-0.47 to 0.15)	0.31	0.40	3	0	0.94	None
		RCT	[24,25]	119/110	-0.22 (-1.11 to 0.67)	0.62	0.19	1	0	0.66	None
		NRSI	[8,19]	294/130	-0.15 (-0.49 to 0.18)	0.37	0.19	1	0	0.66	None
	12 months (RCT)		[24,25,28]	154/140	0.29 (-0.34 to 0.91)	0.37	0.03	2	0	0.99	None
Qmax (mL/s)
	1 month (RCT)		[24,25,28]	163/152	0.45 (-0.56 to 1.45)	0.38	1.24	2	0	0.54	None
	3 months (RCT)		[24-26,28]	192/182	0.76 (-0.19 to 1.71)	0.12	4.09	3	27	0.25	None
	6 months
		Total	[8,19,24,25]	413/240	0.63 (-0.63 to 1.89)	0.32	11.75	3	74	0.008	None
		RCT	[24,25]	119/110	1.71 (0.05 to 3.37)	0.04*	1.76	1	43	0.18	OP
		NRSI	[8,19]	294/130	-0.20 (-1.62 to 1.21)	0.78	3.89	1	74	0.05	None
	12 months (RCT)		[24,25,28]	154/140	-0.44 (-2.58 to 1.69)	0.69	5.17	2	61	0.08	None
QoL
	3 months (RCT)		[25,26,28]	138/127	0.07 (-0.28 to 0.43)	0.69	13.76	2	85	0.001	None
PVR (mL)
	3 months (RCT)		[24-26]	151/138	-0.79 (-7.91 to 6.32)	0.83	5.53	2	64	0.06	None

OP, open prostatectomy; WMD, weighted mean difference; CI, confidence interval; IPSS, International Prostate Symptom Score; RCT, randomized controlled trial; NRSI, non-randomised studies of the effects of intervention; Qmax, maximum flow rate; QoL, quality of life score; PVR, postvoid residual volume.

*Statistically significant results (p<0.05).

4. Complications

The transurethral laser group had a significantly lower incidence of blood transfusion than the OP group (OR, 0.10; 95% CI, 0.05 to 0.19; p<0.001; I2=8%). No statistical differences were found with respect to UTI, stress incontinence, AUR, bladder neck/urethral stenosis, and surgical intervention for bleeding (Table 4). As for subgroup analysis, the complication of AUR in NRSI group was statistically different (OR, 1.93; 95% CI, 0.61 to 6.10; p=0.26; I2=53% [RCT: OR, 1.06; 95% CI, 0.26 to 4.22; p=0.94; studies=2/4; I2=51%, NRSI: OR, 4.53; 95% CI, 1.24 to 16.52; p=0.02; studies=2/4; I2=0%]), while the other indicators of subgroups had no obvious change compared with the overall analysis.

Table 4

Overall analysis of complications comparing transurethral laser with OP

Complication		Study	No. of patients (laser/OP)	OR (95% CI)	p-value	Study heterogeneity				Favors
						χ2	df	I2 (%)	p-value
Urinary tract infection
	Total	[19,21,22,25]	304/245	0.94 (0.51 to 1.73)	0.85	5.15	3	42	0.16	None
	RCT	[25]	65/60	0.75 (0.33 to 1.72)	0.50	NA	NA	NA	NA	None
	NRSI	[19,21,22]	239/185	1.23 (0.49 to 3.08)	0.65	4.96	2	60	0.08	None
Transitory urge incontinence
	Total	[19,23,28]	141/89	0.53 (0.25 to 1.10)	0.09	0.86	2	0	0.65	None
	RCT	[28]	41/39	0.67 (0.27 to 1.66)	0.39	NA	NA	NA	NA	None
	NRSI	[19,23]	100/50	0.32 (0.09 to 1.14)	0.08	0.00	1	0	0.97	None
Stress incontinence
	Total	[8,19,21,23,25,26,28]	499/358	1.61 (0.91 to 2.85)	0.10	4.28	6	0	0.64	None
	RCT	[25,26,28]	138/127	1.56 (0.46 to 5.30)	0.48	0.25	2	0	0.88	None
	NRSI	[8,19,21,23]	361/231	-0.15 (-0.49 to 0.18)	0.14	4.03	3	26	0.26	None
Acute urinary retention
	Total	[19,21,25,28]	220/207	1.93 (0.61 to 6.10)	0.26	6.39	2	53	0.09	None
	RCT	[25,28]	106/99	1.06 (0.26 to 4.22)	0.94	2.03	1	51	0.15	None
	NRSI	[19,21]	114/108	4.53 (1.24 to 16.52)	0.02*	0.20	1	0	0.66	OP
Blood transfusion
	Total	[2,8,19,21,22,24,25,28]	644/583	0.10 (0.05 to 0.19)	<0.001*	7.57	7	8	0.37	Laser
	RCT	[24,25,28]	166/159	0.10 (0.03 to 0.35)	0.0003*	1.49	2	0	0.47	Laser
	NRSI	[2,8,19,21,22]	294/130	-0.15 (-0.49 to 0.18)	<0.001*	6.08	4	34	0.19	Laser
Surgical intervention for bleeding
	Total	[19,21,22,24]	289/229	0.86 (0.30 to 2.51)	0.79	1.01	3	0	0.80	None
	RCT	[24]	50/44	0.87 (0.17 to 4.56)	0.87	NA	NA	NA	NA	None
	NRSI	[19,21,22]	239/185	0.86 (0.21 to 3.46)	0.83	1.01	2	0	0.60	None
Bladder neck/urethral stenosis
	Total	[8,21,23,24,25,26,28]	462/367	0.61 (0.31 to 1.21)	0.16	3.89	6	0	0.69	None
	RCT	[24,25,26,28]	191/176	0.92 (0.37 to 2.26)	0.85	0.00	3	0	>0.999	None
	NRSI	[8,21,23]	271/191	0.35 (0.12 to 1.04)	0.06	2.08	2	4	0.35	None

OP, open prostatectomy; OR, odds ratio; CI, confidence interval; RCT, randomized controlled trial; NRSI, non-randomised studies of the effects of intervention; NA, not available.

*Statistically significant results (p<0.05).

5. Sensitivity analysis, bias analyses, and quality assessment

According to our previous results, there were 2 low-quality RCTs and 3 high-quality RCTs according to the Jadad scale [33]. Sensitivity analysis was performed using 3 high-quality RCTs, and the corresponding results were shown in Table 5. There was no obvious change except for the operative time (WMD, 28.85; 95% CI, 14.01 to 43.69; p=0.0001; I2=93%), and Qmax (WMD, 1.49; 95% CI, 0.40 to 2.57; p=0.007; I2=43%) at 6 months postoperatively. Most indicators were stable, which was consistent with the overall analysis. We used funnel plot to evaluate publication bias of enrolled articles in this meta-analysis. Taking prostate volume as an example, no obvious asymmetry was shown in the funnel plot (Fig. 2). Table 6 summarized the quality assessment of enrolled non-RCTs. The Newcastle–Ottawa scale scores from zero to nine stars. For included non-RCTs, 6 studies were of a high quality with a score of eight to nine stars.

Table 5

Sensitivity analysis of 3 high-quality RCTs comparing transurethral laser with OP

Outcome of interest			Study	No. of patients (laser/OP)	OR/WMD (95 % CI)	p-value	Study heterogeneity				Favors
							χ2	df	I2 (%)	p-value
Perioperative outcomes
	Operative time (min)		[24,25,28]	166/159	28.85 (14.01 to 43.69)	0.0001*	26.99	2	93	<0.00001	OP
	Hemoglobin decrease (g/dL)		[24,25,28]	166/159	-0.97 (-1.31 to -0.64)	<0.001*	0.13	2	0	0.94	Laser
	Resected prostate weight (g)		[24,28]	101/99	-14.83 (-40.13 to 10.47)	0.25	6.44	1	84	0.01	None
	Catheterization (d)		[24,25,28]	166/159	-4.44 (-7.11 to -1.76)	0.001*	359.11	2	99	<0.00001	Laser
	Hospital stay (d)		[24,25,28]	166/159	-4.64 (-7.19 to -2.08)	0.0004*	158.09	2	99	<0.00001	Laser
Effectiveness-related outcomes
	IPSS
		1 month	[24,25,28]	163/152	0.50 (-0.94 to 1.93)	0.50	8.72	2	77	0.01	None
		3 months	[24,25,28]	160/149	0.28 (-0.38 to 0.94)	0.40	2.58	2	23	0.28	None
		6 months	[24,25]	119/110	-0.22 (-1.11 to 0.67)	0.62	0.19	1	0	0.66	None
		12 months	[24,25,28]	154/140	0.29 (-0.34 to 0.91)	0.37	0.03	2	0	0.99	None
	Qmax (mL/s)
		1 month	[24,25,28]	163/152	0.45 (-0.56 to 1.45)	0.38	1.24	2	0	0.54	None
		3 months	[24,25,28]	160/149	0.16 (-1.79 to 2.10)	0.88	4.06	2	51	0.13	None
		6 months	[24,25]	119/110	1.49 (0.40 to 2.57)	0.007*	1.76	1	43	0.18	OP
		12 months	[24,25,28]	154/140	-0.44 (-2.58 to 1.69)	0.69	5.17	2	61	0.08	None
	QoL (3 mo)		[25,28]	106/99	0.05 (-0.63 to 0.73)	0.88	13.72	1	93	0.0002	None
	PVR (3 mo, mL)		[24,25]	119/110	3.15 (-1.74 to 8.04)	0.21	0.73	1	0	0.39	None
Complications
	Stress incontinence		[25,28]	106/99	1.58a (0.20 to 12.20)	0.66	0.25	1	0	0.62	None
	Acute urinary retention		[25,28]	106/99	1.06a (0.26 to 4.22)	0.94	2.03	1	51	0.15	None
	Blood transfusion		[24,25,28]	166/159	0.10a (0.03 to 0.35)	0.0003*	1.49	2	0	0.47	Laser

RCT, randomized controlled trial; OP, open prostatectomy; OR, odds ratio; WMD, weighted mean difference; CI, confidence interval; IPSS, International Prostate Symptom Score; Qmax, maximum flow rate; QoL, quality of life score; PVR, postvoid residual volume.

aValues are presented as OR.

*Statistically significant results (p<0.05).

Fig. 2

Funnel plot for prostate volume. Funnel plot with pseudo 95% confidence limits. SE, standard error; MD, mean difference; RCT, randomized controlled trial; NRSI, non-randomised studies of the effects of intervention.

Table 6

The Newcastle–Ottawa scale scores for quality assessment of non-RCTs

Study	Selection				Comparability based on the design or analysis	Outcome			Total score
	Representativeness of the exposed cohort	Selection of the non-exposed cohort	Ascertainment of exposure	Outcome of interest was not present at start of study		Assessment of outcome	Follow-up long enough for outcomes to occur	Adequacy of follow-up of cohorts
Moody and Lingeman [23]	1	1	1	1	2	1	1	0	8
Elshal et al. [22]	1	1	1	1	2	1	1	1	9
Lanchon et al. [21]	1	1	1	1	2	1	1	1	9
Misraï et al. [8]	1	1	1	1	2	1	1	1	9
Rosenhammer et al. [20]	1	1	1	1	2	1	0	0	7
Enikeev et al. [19]	1	1	1	1	2	1	1	0	8
Nestler et al. [2]	1	1	1	1	2	1	1	0	8

RCT, randomized controlled trial.

DISCUSSION

Since OP was initially reported by Eugene Fuller in 1894, this open surgical approach had been used for more than a hundred years [34]. While OP retrieves great prostatic volume and has the advantage of a short operative time, it is always considered as the gold standard for surgical treatment with prostates larger than 80 g [13]. However, OP is still considered to associate with significant postoperative complications and mortality [242835]. Therefore, in the past three decades, new minimally invasive surgical methods were developed by engineers and surgeons, and expected to replace traditional OP for large BPH patients.

At present, transurethral laser prostatectomy is applied in clinical, including holmium laser, green laser, and thulium laser for pursuing a much better efficacy and safety [11]. For example, transurethral holmium laser enucleation has been proven to be a size-independent method for surgical treatment of LUTS/benign prostatic obstruction in prostates ≥30 cc currently, and postoperative functional outcomes and Clavien–Dindo grade ≥II complications show no difference between all sizes [36]. Therefore, transurethral laser prostatectomy has gradually occupied a certain proportion in the surgical treatment of different sizes prostates. However, it comes to clinical dispute when faced with treatment of large prostates with refractory LUTS. How to choose a more suitable surgical method for BPH with large prostates is difficult but be of great importance, and it is worthy of comparing transurethral laser prostatectomy and OP for large-sized prostates. In our previous meta-analysis, transurethral laser prostatectomy was proven to be an appropriate treatment for large prostates with the advantages of less intraoperative blood loss, and shorter postoperative catheterization time and hospital stay [33]. However, there were two limitations in the previous study. On one hand, the number of enrolled studies and the sample size were small (5 studies and 448 cases). On the other hand, all the studies of previous result were carried out twelve years ago, and there might be a certain bias due to lack of studies in the last 5 years. The published meta-analysis needs to be updated, and transurethral bipolar enucleation needs to be eliminated to avoid potential bias [12]. Built on previous work of us and Lin, new meta-analysis included recent studies with larger sample sizes is very necessary to resolve this clinical dispute of large prostates [1233].

Twelve studies were enrolled in the current analysis, and the relevant outcomes were appraised carefully. Transurethral laser surgery was considered to have various advantages over OP in terms of blood loss, hospital stay, and catheterization days, while there was no significant difference was observed in resected prostate weight (p=0.62) and operative time (p=0.43). In the previous meta-analysis, OP was proved to have the advantage of a shorter operative time and more resected weight compared with laser surgery, which was inconsistent with the current result [33]. All studies included in the previous meta-analysis were before 2008, and the current meta-analysis includes 2016 to 2019. The probable explanations for changes in operation duration include the following: First, with the optimization of the learning curve of transurethral laser prostatectomy, the accumulation of operations makes urologists more efficient and more skilled. Second, surgical duration is further reduced with the development of tissue morcellator and the increase of laser power. In addition, improved surgical techniques have a very good effect on permitting less operation time, such as the diversity of surgical methods. Transurethral laser surgery is no longer a single resection, enucleation or vaporization, but a fusion of two or three. Therefore, with the faster development and further promotion of laser technology, its efficiency is progressively improving, even more than open surgery. As for resected tissue weight, endoscopic laser is able to remove the adenoma close to the surgical capsule without depending on the volume of the prostate, similarly to what the index finger does during an OP procedure [15]. Furthermore, the wide application of tissue morcellator protects resected specimens and reduces the loss of specimen volume in transurethral laser prostatectomy. Although the pooled data revealed that the resected tissue weights between the two groups were not significantly different, 1 trial reported that transurethral laser yielded much more specimen weights compared with OP [23]. The subgroup analysis was not carried out due to limited sample size.

The surgical goal of BPH should be safe removal of adenoma to relieve BOO, while improving LUTS significantly. No statistically significant difference was observed in related parameters of Qmax, PVR, QoL, and IPSS at the different follow-ups. Although LUTS significantly lessened for both groups in the early follow-up of the meta-analysis, the effect can be sustained depends on more long-term follow-ups. However, similar studies about large prostates will be less, as OP is increasingly being replaced by laparoscopic or robotic prostatectomy.

The overall complication rates were not significantly different between transurethral laser surgery and OP, except for transfusion rate of the meta-analysis. Laser had excellent hemostatic property, which makes the transfusion rate of laser surgery less than that of OP. Unfortunately, the number of the enrolled studies comparing complications was small, and some studies just provided overall complication rate without furnishing the specific adverse events, which may cause some bias, but our meta-analysis indicates that transurethral laser prostatectomy is at least as safe as OP. The smooth urination continence after the surgery is the key for surgical treatment of large-sized BPH. Compared with OP, transurethral laser prostatectomy for large prostates had similar incidence of AUR, stress incontinence, and transitory urge incontinence. In previously published studies, transurethral laser prostatectomy and OP were similar in terms of urination continence [1113141516]. Our analysis also confirmed these results.

There are some certain constraints in this study. First of all, the sample size of most enrolled studies was not large enough. Although one large-scale study was included, this study was not an RCT research, which has selection bias [8]. Next, some papers did not provide micturition parameters (Qmax, PVR, QoL, and IPSS), which is the key to the surgery, although some papers reported these parameters. Last but not least, most enrolled studies in the meta-analysis reported an insufficient follow-up period, and follow-up period of most studies is not more than two years.

CONCLUSIONS

Both OP and transurethral laser prostatectomy are effective and safe treatments for large prostate adenomas. With these advantages of less blood loss and transfusion, and shorter catheterization time and hospital stay, laser may be a better choice for large BPH who need surgical treatment.