Reoperation after surgical treatment for benign prostatic hyperplasia: a systematic review

Context Surgical treatment is important for male lower urinary tract symptom (LUTS) management, but there are few reviews of the risks of reoperation. Objective To systematically evaluate the current evidence regarding the reoperation rates of surgical treatment for LUTS in accordance with current recommendations and guidelines. Evidence acquisition Eligible studies published up to July 2023, were searched for in the PubMed® (National Library of Medicine, Bethesda, MD, USA), Embase® (Elsevier, Amsterdam, the Netherlands), and Web of Science™ (Clarivate™, Philadelphia, PA, USA) databases. STATA® (StataCorp LP, College Station, TX, USA) software was used to conduct the meta-analysis. Random-effects models were used to calculate the pooled incidences (PIs) of reoperation and the 95% confidence intervals (CIs). Evidence synthesis A total of 119 studies with 130,106 patients were included. The reoperation rate of transurethral resection of the prostate (TURP) at 1, 2, 3, and 5 years was 4.0%, 5.0%, 6.0%, and 7.7%, respectively. The reoperation rate of plasma kinetic loop resection of the prostate (PKRP) at 1, 2, 3, and 5 years was 3.5%, 3.6%, 5.7%, and 6.6%, respectively. The reoperation rate of holmium laser enucleation of the prostate (HoLEP) at 1, 2, 3, and 5 years was 2.4%, 3.3%, 5.4%, and 6.6%, respectively. The reoperation rate of photoselective vaporization of the prostate (PVP) at 1, 2, 3, and 5 years was 3.3%, 4.1%, 6.7%, and 7.1%, respectively. The reoperation rate of surgery with AquaBeam® at 1, 2, 3, and 5 years was 2.6%, 3.1%, 3.0%, and 4.1%, respectively. The reoperation rate of prostatic artery embolization (PAE) at 1, 2, 3, and 5 years was 12.2%, 20.0%, 26.4%, and 23.8%, respectively. The reoperation rate of transurethral microwave thermotherapy (TUMT) at 1, 2, 3, and 5 years was 9.9%, 19.9%, 23.3%, and 31.2%, respectively. The reoperation rate of transurethral incision of the prostate (TUIP) at 5 years was 13.4%. The reoperation rate of open prostatectomy (OP) at 1 and 5 years was 1.3% and 4.4%, respectively. The reoperation rate of thulium laser enucleation of the prostate (ThuLEP) at 1, 2, and 5 years was 3.7%, 7.7%, and 8.4%, respectively. Conclusion Our results summarized the reoperation rates of 10 surgical procedures over follow-up durations of 1, 2, 3, and 5 years, which could provide reference for urologists and LUTS patients. Systematic review registration https://www.crd.york.ac.uk/PROSPERO, identifier CRD42023445780.


Introduction
Lower urinary tract symptoms (LUTSs) related to benign prostatic hyperplasia (BPH) are very common in older men and seriously affect their quality of life (1).Although a1-adrenoceptor antagonists and 5a-reductase inhibitors are first-line drugs with good efficacy, many adverse events such as dizziness, asthenia, postural hypotension, and low libido may occur as a result of treatment with them (1).In addition, there are some patients who have poor drug responsiveness or for whom these drugs are eventually unable to delay disease progress.Therefore, many patients ultimately require surgical intervention (1).According to the current guidelines, indications of the need for surgery include renal insufficiency, refractory urinary retention, recurrent urinary tract infections (UTIs) or gross hematuria, bladder stones, or the patient being refractory to or unwilling to use other therapies (2,3).Transurethral resection of prostate (TURP) has long been considered the "gold standard" for the surgical management of LUTSs/BPH (4).In recent decades, many new technologies and procedures have been widely used and recommended by clinical guidelines, such as plasma kinetic resection of prostate (PKRP), holmium laser enucleation of the prostate (HoLEP), and photoselective vaporization of the prostate (PVP) (2,3).Based on the current guidelines, the most suitable type of surgery for a patient depends on their prostate volume (PV), physical condition, and economic situation, and can also even be dependent on the preference of the surgeon and the machines owned by the hospital (2,3).
Since the physical characteristics of the surgical technique and the anatomy of the prostate vary across patients, some may suffer bladder neck contracture (BNC), urethral stricture, or other complex complications postoperatively, and these may need surgical retreatment (1).In addition, some surgical procedures do not provide patients with satisfactory relief from their symptoms, or do not prevent the reappearance of bladder outlet obstruction over time after surgery, which may also require surgical retreatment.Moreover, some recommended surgical procedures are still under investigation such as surgery with AquaBeam ® and prostatic artery embolization (PAE), of which the efficacy, safety, and tolerability still need to be confirmed (3).When selecting an appropriate surgical approach, knowledge of the reoperation rates could be used to predict the cost and management of disease in the years following the operation.
In the past, many studies have reported on the reoperation rate after various kinds of surgery.For patients who had undergone TURP, an Austrian nationwide study reported that the retreatment rate at the 1-year follow-up was 3.7%, and that this increased by approximately 1%-2% with each subsequent year (5,6).A recent study reported that the rate of secondary surgery for TURP, transurethral incision of prostate (TUIP), and PVP at the 5-year follow-up was 10.3%, 13.6%, and 11.6%, respectively (7).Other procedures such as PAE and transurethral microwave thermotherapy (TUMT) were reported to have a higher risk of retreatment (8,9).Recently, a systematic review also summarized the pharmacologic and surgical retreatment rates after newer officebased treatments, including water vapor thermal therapy (WVTT), prostatic urethral lift (PUL), and that using a temporarily implanted nitinol device (iTIND) (10).However, there is still a limited number of reviews on the reoperation rate of common surgeries recommended by the guidelines.We therefore conducted an updated systematic review and meta-analysis to summarize the reoperation rates of common surgical treatment for LUTSs/BPH.This review could be important to both BPH/LUTS patients and urologists when they are selecting an appropriate surgical procedure.

Literature search
This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (11).The protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) database (registration number CRD42023445780).
Studies were searched for in the PubMed ® (National Library of Medicine, Bethesda, MD, USA), Embase ® (Elsevier, Amsterdam, the Netherlands), and Web of Science ™ (Clarivate ™ , Philadelphia, PA, USA) databases up to July 2023.The primary outcomes were the rates of surgical retreatment during follow-up.The search strategy is provided in the Supplementary Files.The initial screening, which included reading the title and abstract, was performed by the two authors independently (WXH and TD).
Subsequently, the full text of potentially relevant studies was acquired for further confirmation and the data extraction process.Any conflicts that arose between the two authors during article selection and data extraction were resolved through discussion with an arbitrator (ZPN).

Inclusion and exclusion criteria
Articles that met the following criteria were included: (1) those that reported on the surgical retreatment rate of BPH/LUTS patients who had undergone operations in hospitals during the follow-up period; (2) those that were focused only on procedures recommended in the recent guidelines of the Association of University Administrators (AUA) and the European Association of Urology (EAU), including TURP, PKRP, TUIP, open prostatectomy (OP), thulium:yttrium aluminum garnet laser (Tm : YAG), enucleation of the prostate (ThuLEP), HoLEP, PVP, surgery with AquaBeam, PAE, and TUMT; (3) those that reported on a randomized controlled trial (RCT), non-randomized prospective study, or retrospective study; (4) those that were original peer-reviewed human participant research studies; (5) those that were published in English; and (6) those with a follow-up duration of 1, 2, 3, or 5 years.Studies such as reviews, editorials, commentaries, meeting abstracts of unpublished studies, and case reports were excluded.For duplicate publications, the higher-quality study, or the study that had been most recently published was selected.

Data extraction
Data were extracted from eligible studies by the two authors independently (WXH and TD).The extracted data included the first author's surname, publication year, country of research, study design, patient information, follow-up time, and rates of surgical retreatment.The patient information collected included the patient's number, age, prostate volume (PV), International Prostate Symptom Score (IPSS), postvoid residual volume (PVR), and maximum urinary flow rate (Q max ).It should be noted that surgical retreatment included both the management of the prostatic obstruction and of postoperative complications such as bladder neck contracture or urethral stricture.For some studies, we calculated the rate for further investigation if authors reported only the number of retreatment patients.

Quality assessment
The risk of bias (RoB) and quality of each eligible study were assessed by two authors independently (WXH and TD).For RCTs, the RoB was assessed, summarized, and then visualized using the Cochrane Collaboration RoB tool embedded in the RevMan (The Cochrane Collaboration, The Nordic Cochrane Centre, Copenhagen, Denmark) software (version 5.4).For single-arm studies, the RoB was assessed in accordance with the EAU guidelines for systematic reviews (12).

Data synthesis
For each surgical type, the baseline characteristics of patients were summarized and then pooled using Microsoft Excel ® (Microsoft Corporation, Redmond, WA, USA) software (2016).In addition, the pooled incidences (PIs) and corresponding 95% confidence intervals (CIs) of the surgical retreatment rates were evaluated and stratified by the surgical type and follow-up duration (i.e., 1 year, 2 years, 3 years, and 5 years) using STATA (version 17.0; StataCorp LP, College Station, TX, USA).A random-effects model was used to estimate the pooled incidences.

Study selection and characteristics
The study selection process is presented in the PRISMA flow chart shown in Figure 1.A total of 119 studies met our inclusion criteria.The baseline characteristics of the included studies are presented in Table 1.A total of 130,106 patients were included, of whom 100,295 had undergone TURP, 1,530 had undergone PKRP, 90 had undergone TUIP, 4,621 had undergone OP, 3,956 had undergone HoLEP, 1,584 had undergone ThuLEP, 14,058 had undergone PVP, 217 had undergone surgery with AquaBeam, 1,796 had undergone PAE, and 1,959 had undergone TUMT procedures.Forty-two studies were RCTs, 29 were nonrandomized prospective studies, and 48 were retrospective singlearm case series.Forty-nine studies were conducted in Europe, 34 were conducted in Asia, 22 studies were conducted in North America, six studies were conducted in Africa, and two studies were conducted in Oceania.In addition, another six multiinstitutional studies were conducted in Europe and North America.

Risk of bias
The quality and RoB assessments are summarized in the Supplementary Files.For the 42 RCT studies, the RoBs of the 32 studies were considered unclear, whereas 47 of the 77 single-arm studies were assessed as having a high RoB.

Baseline patient characteristics
As shown in Table 2, the preoperative characteristics of the patients were summarized and pooled in accordance with the type of procedure.It appeared that the PV, IPSS, and PVR values of patients who had undergone TUIP, OP, or TUMT were different from those of other groups.For patients who had undergone TURP, the average age was 70 years, the average PV was 55 cm 3 , the average IPSS was 22, the average PVR was 184 mL, and the average Q max was 8 mL per s.For patients who had undergone PKRP, the average age was 69 years, the average PV was 67 cm 3 , the average IPSS was 21, the average PVR was 112 mL, and the average Q max was 7 mL per s.For patients who had undergone TUIP, the average age was 71 years, the average PV was 26 cm 3 , the average IPSS was 16, the average PVR was 139 mL, and the average Q max was 9 mL per s.For patients who had undergone OP, the average age was 71 years, the average PV was 106 cm 3 , the average IPSS was 24, the average PVR was 147 mL, and the average Q max was 6 mL per s.For patients who had undergone HoLEP, the average age was 70 years, the average PV was 79 cm 3 , the average IPSS was 21, the average PVR was 186 mL, and the average Q max was 8 mL per s.For patients who had undergone ThuLEP, the average age was 70 years, the average PV was 65 cm 3 , the average IPSS was 24, the average PVR was 138 mL, and the average Q max was 7 mL per s.For patients who had undergone PVP, the average age was 72 years, the average PV was 63 cm 3 , the average IPSS was 22, the average PVR was 166 mL, and the average Q max was 8 mL per s.For patients who had undergone surgery with AquaBeam, the average age was 67 years, the average PV was 79 cm 3 , the average IPSS was 23, the average PVR was 117 mL, and the average Q max was 7 mL per s.For patients who had undergone PAE, the average age was 66 years, the average PV was 86 cm 3 , the average IPSS was 22, the average PVR was 124 mL, and the average Q max was 10 mL per s.For patients who had undergone TUMT, the average age was 67 years, the average PV was 48 cm 3 , the average IPSS was 21, the average PVR was 76 mL, and the average Q max was 9 mL per s.

Surgical retreatments after different procedures
In Table 3, the surgical retreatment rates of various procedures in different follow-up years are shown.Most of the evidence was derived from studies on TURP, PKRP HoLEP, and PVP, as there were fewer studies on TUIP, OP, ThuLEP, AquaBeam, PAE, and TUMT.For almost every procedure, the risk of surgical retreatment increased over time.

Discussion
This systematic review comprehensively summarized the reoperation rates after surgeries for male LUTS management.We found that the retreatment rates increased over time and differed among procedures.Our results can be used to counsel both the urologists and patients regarding the different therapeutic strategies.
As the gold standard of surgical treatment for BPH/LUTSs, it was reported after a nationwide analysis of 20,671 patients that the surgical retreatment rate of TURP was 3.7% for 1 year and 9.5% for 5 years (5), which was similar to our current result.As the most widely investigated alternative to TURP and PKRP (bipolar TURP) was found to have a comparable efficacy in regard to the long-term follow-up, but was safer during the perioperative period (3).Numerous studies have reported that PKRP exhibited similar rates of surgical retreatment as TURP (3), which was consistent with our results.Moreover, TUIP was recommended for patients with a PV of < 30 mL and those without a middle lobe (2,3).A meta-analysis of six trials published 13 years ago showed that reoperation was more common after TUIP (18.4%) than it was after TURP (7.2%) (128).The follow-up periods of the six trials included above were different, which may introduce some bias; however, our updated review showed a similar result in that the reoperation rate of TUIP was 13.4% in 5 years, which was higher among these surgical procedures.The higher risk of surgical retreatment associated with TUIP may be due to its method, which involves only incising the bladder outlet without removing prostatic tissue.However, TUIP has been underutilized in the urological community over the years, the reasons for this include concerns related to the limitations of PV as an indicator of the need for surgery and also its long-term efficacy (129).In contrast to TUIP, during OP, the whole prostate is removed, which is recommended for patients with a PV of > 80 mL (2,3).A nationwide analysis reported that the surgical retreatment rate of OP was 3.0% for 1 year and 6.0% for 5 years (130), which was similar to our result.Although its long-term reoperation rate seems lower than those of the other procedures, OP showed poorer perioperative safety than the other transurethral approaches, and was associated with higher rates of blood transfusions and even death (131,132).Therefore, OP was less popular than the other minimally invasive surgeries.However, in recent years, prostatectomy with laparoscopy or robot-assisted surgery showed better safety and were also recommended by guidelines (2).
As an alternative to open enucleation, some studies reported that HoLEP has a lower risk of reoperation than TURP or PKRP (59, 133,134), whereas another reported that there was no difference (135, 136).Indeed, our results suggest that the reoperation rates for HoLEP are similar to (and possibly slightly lower than) those for TURP or PKRP.Enucleation using another laser, ThuLEP has a rate of reoperation that is similar to (and possibly slightly higher than) that of HoLEP, which may be due to them being similar procedures.A recent interesting study from Italy reported that an improved ThuLEP technique successfully preserved the ejaculation function in most patients (137), which suggested its potential in decreasing the reoperation rates.
For vaporization, PVP has been used in clinical settings for many years and there are many related studies that have shown it has a similar efficacy to TURP (2,3).A previous meta-analysis published by Zhou and colleagues reported that the reoperation rate after PVP was higher than that after TURP (138).However, there were only three related trials included in Zhou's study, and the follow-up durations of these trials were different (138).Our current summarized results, which included 53 trials, reported that the reoperation rates are similar between PVP and TURP.The difference between the results of these two meta-analyses may be due to the number of articles included.AquaBeam has come under investigation in recent years and two related trails, WATER and WATER II, reported the reoperation rate associated with it (50, 117, 118).However, there are few studies on this technique and a lack of long-term follow-up data.Although the rate of surgical retreatment appeared to be better than other procedures in our current review, whether or not AquaBeam could be an alternative to traditional procedures still needs a lot of studies and long-term follow-up to be carried out.Previous studies indicate that PAE, another surgical procedure that remains under investigation, has a higher risk of surgical retreatment than that shown in our results (9).Due to the variability of blood supply to the human prostate, non-target embolization may occur, and secondary surgical retreatment is required (139).In addition, it takes time for the prostate to shrink after vessel embolization, and PV will also stop decreasing and begin to increase after a period of time (140).Therefore, both complications and insufficient treatment response may result in a higher risk of reoperation.Overall, the efficacy and reliability of PAE remain undetermined, and further investigations and improvements are still needed.TUMT, one of the earliest technologies used for the treatment of BPH/LUTSs, has been used and studied less in recent years, due to its higher risk of retreatment and the emergence of newer, minimally invasive technologies (2,8).Our current results confirmed that it has a higher rate of surgical retreatment.In fact, TUMT was not recommended by the latest version of the EAU guidelines, whereas the AUA guidelines still suggest that this is a reasonable approach.However, considering its higher reoperation rate and the newer, minimally invasive technologies, TUMT will likely be displaced within the next several years (2,141).
There are some limitations or shortcomings in our current analysis and review which must be acknowledged.First, RoB was in some of the studies included through assessment.Second, our review focused only on the reoperation rates at follow-up periods of 1, 2, 3, and 5 years.However, the follow-up duration was different among studies; examples of follow-up periods were 6 months, 4 years, long term (> 5 years), and some did not last for a 'regular' (i.e., a multiple of a year, half a year, or 1 year) length of time.Therefore, our results are limited by the lack of data obtained during these follow-up durations.Third, 10 surgical procedures were included in our current review, of which the indication that recommended by guidelines are different.The baseline characteristics and therapeutic outcomes of patients may also have varied.Meanwhile, the great difference between data retrieved across techniques may also have led to bias.Fourth, the risk of misestimating the reoperation rate must be noted since patients lost to follow-up are common in studies.Finally, the reoperation rates of other surgical approaches excluded in our review while recommended by guidelines were also obtained during the literature search.However, as their surgical methods were outdated, less commonly used, or they were associated with a smaller number of studies, we excluded them from our current review.
In future, studies of higher quality and longer follow-up durations should be included.With the development of surgical approaches and techniques, the reoperation rate data should also be updated every few years.Meanwhile, the reoperation rate should be further refined based on its cause, and studies exploring the reason for reoperation are needed.In addition to the reoperation rate, the cost of surgical management across procedures varies, and sometimes there are even huge differences, which also affects what procedures are available for patients to choose from (142, 143).For example, a recent study reported that robotic-assisted simple prostatectomy (RASP) showed comparable efficacy and safety with a shorter hospitalization than laparoscopic simple prostatectomy (LSP) (144).However, considering the cost and unavailability of robot-assisted surgery, LSP is also a better alternative (144).Therefore, studies that evaluate the cost-effectiveness of these surgical approaches are also needed.Overall, these further investigations may lead to a reduction in the reoperation rate or prevent some common reoperation cases, which may give more information for clinical practitioners, better improve patient quality of life, and reduce medical expenses for patients.

Conclusions
Our results summarized the reoperation rates of 10 surgical procedures over follow-up durations of 1, 2, 3, and 5 years.There was a great difference in the reoperation rate among these procedures.The OP, AquaBeam, PKRP, and HoLEP procedures exhibited a lower reoperation rate, whereas the PAE and TUMT procedures exhibited a higher rate.These data could provide reference for urologists and BPH/LUTS patients.was supported by grants from the National Natural Science Foundation of China (No. 81772734).

FIGURE 1 Preferred
FIGURE 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram.

TABLE 1
Study characteristics.

TABLE 2
Pooled estimates for baseline confounders.

TABLE 3
Surgical retreatment after different operation procedures.