Introduction

GreenLight photoselective vaporization of the prostate (PVP) is a safe and effective treatment option for benign prostatic hyperplasia (BPH) in small-to-moderate-sized prostates (30–80 cc) [1,2,3,4]. In the context of larger prostates (> 100 cc), it has been shown to be an acceptable alternative to available modalities [5,6,7], but has been associated with longer operative time, more bleeding, and rising retreatment rates after 3 years, amongst other limitations [7,8,9].

It is challenging to disentangle the effect of larger prostate volumes from the effect of the median lobe on these outcomes, as both are closely related. Many men presenting with greater prostate volumes often present with a median lobe, also known as an intravesical prostatic protrusion (IPP) [10, 11]. The median lobe may complicate the surgical procedure, making it more difficult to safely achieve adequate tissue removal [12,13,14]. Conditions such as a high bladder neck, bladder neck bleeding, and/or IPP obstructing the visualization of the ureteral orifices can lead to incomplete vaporization and less favourable functional outcomes [15]. These difficulties have led some to investigate whether a combination of vaporization-resection would be beneficial over vaporization alone in patients with median lobes [16].

However, previous studies investigating the impact of the median lobe on BPH surgery outcomes have been limited in their ability to adjust for prostate size due to a small sample sizes [12, 16,17,18,19]. Despite it being commonly accepted that prostate configuration has little effect on the efficiency of GreenLight laser PVP, there are no studies specifically studying the impact of a median lobe on outcomes of interest while adjusting for potential sources of confounding such as prostate volume. As such, we sought to evaluate the impact of median lobe presence on outcomes of GreenLight PVP while adjusting for prostate size, amongst other factors, using a large, international GreenLight PVP database.

Methods

Data source

Data were obtained from the Global GreenLight Group (GGG) database [20]. At the time of analysis, the database was composed of 3809 men with established BPH who underwent treatment using solely GreenLight PVP using the XPS-180 W system between 2011 and 2019, as performed by high-volume, experienced surgeons, at a total of seven centers in Canada, Mexico, Brazil, Argentina, Italy, France, and Germany. Patients requiring a hybrid procedure involving enucleation were not included in the database. Indications for surgery in each country were based on similar criteria established by the Canadian Urological Association, American Urological Association, and European Association of Urology guidelines on BPH surgical management [2, 21, 22].

Cohort exclusion criteria

We excluded patients with a previous history of transurethral resection of the prostate (TURP), a prostate size under 30 cc, missing median lobe information, known prostate cancer, previous pelvic radiation, or known neurological disorders. All other patients were included.

Definition of the exposure

A median lobe was identified during preoperative cystoscopy as a distinct intravesical protrusion and each patient was coded as either having a median lobe or not having a median lobe. No other median lobe measurement was available. All patients underwent a preoperative cystoscopy. The presence of a median lobe was also further confirmed on transrectal ultrasound (TRUS) (Fig. 1). Prostate size was measured with TRUS using the ellipsoid formula (height × width × length × π/6) [23].

Fig. 1
figure 1

Cystoscopy and ultrasound findings of a prostatic median lobe. a Cystoscopy findings with absence of a median lobe. b Cystoscopy findings demonstrating distinct intravesical median lobe protrusion. c Corresponding ultrasound image confirming the anatomy/configuration of intravesical prostatic protrusion

Full size image

Study parameters

Patients completed the International Prostate Symptom Score (IPSS) questionnaire for lower urinary tract symptoms, including the 6-point quality of life (QoL) score questionnaire. Uroflowmetry was performed, as well as measurements of the post-void residual (PVR). Prostate-specific antigen (PSA) levels were determined in a standard laboratory blood assessment. Prostate size was measured using the TRUS approach. American Society of Anesthesiologists (ASA) score was used as an indicator for patients’ general health/fitness status. Operative time, postoperative hematuria rate, blood transfusion rate, hospital length of stay (LOS) and readmission rates were also documented. Questionnaires, uroflowmetry, PVR and laboratory tests were required at the postoperative visits at 6 and 12 months (Fig. 2).

Fig. 2
figure 2

One-year functional outcomes of Greenlight PVP based on presence of median lobe. a IPSS score, b QoL score, c PVR, d PSA, e Qmax

Full size image

Covariates

Covariates included age, prostate size as measured via transrectal ultrasound (categorized as 30–80 cc, 80–150 cc, ≥ 150 cc in adjusted analyses), body mass index (BMI), and ASA score (categorized as 0, 1, 2, ≥ 3).

Statistical analysis

Means and standard deviations were reported for continuous variables. Categorical variables were presented as frequencies and proportions.

In unadjusted analyses comparing the outcomes of patients with median lobes to those of patients without median lobes, the two-sided t-test was used for continuous outcomes, the Pearson chi-square test for dichotomous outcomes, and the Wilcoxon rank-sum for the non-parametric outcome of hospital LOS. Unadjusted analyses were performed to provide a reference to compare our cohort to that of previous unadjusted analyses of smaller cohorts.

In adjusted analyses, multivariable linear regression models adjusting for the previous listed covariates were fitted to compare continuous outcomes between patients with and without median lobes. For dichotomous outcomes, multivariable logistic regression models adjusting for the covariates were used to estimate the odds of the outcome between both groups. For rare outcomes (transfusions and hematuria), we used Firth’s multivariable logistic regression, again adjusting for all covariates [24]. Hospital LOS was non-normally distributed and as such was log-transformed for adjusted regression analyses. However, all results are presented in non-log-transformed measures to facilitate interpretability.

Statistical analyses were performed using Stata version 14.0 (StataCorp, Texas, USA). Statistical significance was defined as a two-sided p < 0.05. Institutional review board approval was obtained prior to the conduct of this study.

Results

Baseline characteristics

A total of 1650 men met our inclusion criteria. Patients in the median lobe group were on average 1 year younger than patients without a median lobe (p = 0.04). A median lobe was identified in 621 (37.6%) patients. The prostate volume, measured by TRUS, was significantly larger in the median lobe group compared to the absent median lobe group (86.44 ml vs 78.30 ml, respectively; p < 0.01). ASA scores varied considerably between the two groups with favorable scores found in patients in the median lobe group (p < 0.01). IPSS scores were comparable in both groups (p = 0.08). Patients in both groups demonstrated peak urinary flow rates around 7 ml/s (p = 0.13). Additional patient baseline characteristics can be found in Table 1.

Table 1 Baseline patient demographics
Full size table

Unadjusted perioperative and functional outcomes

Operative time was longer in the median lobe group compared to the control group (76.17 min and 65.35 min, respectively, p < 0.01). The median lobe group also had a significantly longer lasing time (41.70 min vs 38.78 min, respectively, p < 0.01). No difference was appreciated for lasing energy (p = 0.93) or hospital LOS (p = 0.07). At both 6- and 12-month follow-up, mean maximum urinary flow rate (Qmax) improvements were significantly greater in patients with median lobes (an improvement of 13.0 ml/s vs. 11.75 ml/s at 6-months and 13.4 ml/s vs. 11.79 ml/s at 12 months for median lobe vs no median lobe group, respectively; p = 0.01 and p < 0.01, respectively). Additional unadjusted perioperative and functional outcomes can be found in Table 2 and Fig. 1.

Table 2 Unadjusted functional and perioperative outcomes
Full size table

Adjusted perioperative and functional outcomes

In analyses adjusting for age, BMI, ASA score, and prostate volume category, the operative time in the median lobe group was 6.72 (95% CI 3.22–10.23; p < 0.01) minutes longer than the control group. Patients with median lobes had 2.90 min longer lasing time (95% CI 1.02–4.78; p < 0.01). Comparable postoperative LOS was seen in both groups. There were no statistically significant differences for both transfusion rates [OR 0.42 (95% CI 0.02–10.44; p = 0.60)] and hematuria rates [OR 1.20 (95% CI 0.67–2.15; p = 0.53)]. Similarly, patients with a median lobe had comparable odds [OR 1.07 (95% CI 0.63–1.82; p = 0.796)] of being readmitted within 30 days.

IPSS changes at 6-month follow-up were not statistically different in the median group compared to the control group [+ 0.91 (95% CI − 0.33 to 2.14; p = 0.15)]. Comparatively, the IPSS drop after 12 months demonstrated a 1.59-point improvement (95% CI 0.11–3.08; p = 0.04) in the median lobe group. Changes in QoL scores in the median lobe group were similar to the control group at 6- and 12-month follow-up [+ 0.21 (95% CI − 0.09 to 0.51; p = 0.165) and 0.32-point improvement (95% CI − 0.01 to 0.66; p = 0.06), respectively]. The decrease in PVR after 6 months was 46.51 ml (95% CI 4.65–88.36; p = 0.03) greater in patients with median lobes compared to men without median lobes. However, the difference was non-significant at 12 months of follow-up [45.34 ml improvement (95% CI − 3.7 to 94.4 l; p = 0.07)]. Patients with median lobes had a similar Qmax to patients without median lobes at both 6- and 12-month follow-up [difference: − 0.03 (95% CI − 1.23 to 1.17; p = 0.23) and − 0.66 (95% CI − 1.74 to 0.42; p = 0.96), respectively].

Discussion

Larger prostate sizes are generally associated with more complications and worse outcomes compared to smaller glands [7,8,9]. Increasing prostatic volume is also associated with more complex anatomy, notably the presence of median lobes [10, 11]. They are thought to complicate surgery, particularly due to concerns of inadequate prostatic adenoma resection. However, previous studies comparing Greenlight PVP outcomes of patients with or without median lobes were limited by small sample sizes and could not adjust for prostate volume [15, 18, 19]. In our analysis of the GGG database adjusting for prostate volume categories and other factors, we found no significant differences in outcomes between patients with or without median lobes undergoing GreenLight PVP.

Patients presenting with LUTS and median lobe enlargement have been found to have decreased peak urinary flow rates and higher incidence of urinary retention when compared to patients with other prostatic configurations [19, 25]. Previous literature suggested that the presence of a median lobe may result in worse functional outcomes. Using a fluid structural interaction analysis approach using 3D models, Zheng et al. found that variations in prostate anatomy such as large IPPs lead to reduced flow efficiency [26]. In the clinical setting, the presence of a large median lobe has been found to be a predictor of trial without catheter failure [27]. Furthermore, studies have shown that greater length of IPP is associated with lower response to alpha-blockers and 5-alpha-reductase-inhibitors [28, 29]. This is believed to be secondary to the IPP of the enlarged median lobe creating a ball-valve mechanism with the opening of the bladder neck and/or causing abnormal bladder contraction with loss of the funnel effect of the bladder neck during voiding [15].

Men with median lobes undergoing GreenLight PVP using the 180 W XPS system had similar postoperative outcomes to those without a median lobe. Like Gu et al.’s findings, the operation time in our study was significantly longer in cases with a median lobe both in non-adjusted and adjusted analysis, although the absolute difference was small. This can likely be explained by a more extensive prostatic floor treatment as more time is required to vaporize the additional prostatic tissue of the enlarged median lobe. Zorn et al. described that patients with a median lobe will receive a second grooved incision at 7 o’clock prior to prostatic floor treatment [13]. Additionally, Kim et al. mention that a large protruding median lobe increases the difficulty of ureteral orifice identification [16]. Gu et al. studied the impact of median lobe presence on GreenLight PVP using the 120 W HPS [18]. In their study, the bilobed prostate group was compared to the trilobed group and was found to have a shorter lasing time and operation time. However, significant differences were present at baseline between the groups. Therefore, due to a lack of an adjusted analysis, it is impossible to appreciate the effect of median lobe on both perioperative and postoperative outcomes in their study.

In our study, there were no differences in safety outcomes in both groups. Patients in both groups had similar postoperative hospital LOS, hematuria rates, and rehospitalization rates, even after adjusting for age, prostate volume, BMI and ASA scores.

Adjusted 1-year functional outcomes in both groups were comparable except for statistically significant favorable outcomes in the median lobe group with regards to 6-month change in PVR and 12-month change in IPSS. Using GreenLight PVP 120 W HPS, Kim et al. noted a statistically significant difference in total IPSS and voiding IPSS sub-score in favor of the median lobe group at 1-month and 3-month follow-up, but not at 6-month follow-up [19]. This greater initial improvement is thought to be due to the different underlying mechanisms of obstruction between both groups. In patients without an obstructing median lobe, the cause of obstruction is predominantly due to increased intraluminal pressure of the prostatic urethra. In comparison, in patients with median lobes, the sudden elimination of the protruding intravesical mass may lead to a more rapid initial improvement of outlet obstruction [19].

We found a significant difference in Qmax change from baseline at 6- and 12-month follow-up in unadjusted analysis. However, the mean Qmax was similar between both groups at each follow-up period. Kim et al. also noted a statistically favourable change in Qmax from baseline at 6-month follow-up in their median lobe group compared to the control group [19]. However, in contrast to our findings, patients with median lobes in their study had a significantly lower preoperative Qmax compared to those without median lobes. Furthermore, after adjusting for covariates in our study, we no longer appreciated a difference in Qmax change from baseline in between the groups. These findings therefore support the hypothesis that the presence of a median lobe is not an independent predictor of the postoperative outcomes of GreenLight PVP.

The limitations of this study include its retrospective design and short follow-up period. The GGG database includes data of multiple high-volume international urological centers with experienced surgeons, which may not be representative of the lower-volume or more novice surgeons. The extent of median lobe IPP was not quantified as median lobe volume and anatomy were not collected. Although, previous studies suggest that there is no statistically significant correlation between degree of IPP and postoperative outcomes such as IPSS score, storage symptom score, voiding symptom score, QoL, Qmax and PVR [15].

Nevertheless, this is the first study which looks at the outcomes of GreenLight PVP in patients with median lobes while adjusting for confounding factors. The large sample size of our study allowed us to adjust for confounders, further isolating the impact of the median lobe on operative outcomes. Future research should examine the impact of median lobes on other modalities, namely enucleation which should be considered for patients presenting with a median lobe.

Conclusion

In a retrospective analysis of a large, international database of GreenLight PVP using the XPS-180 W system, we found no clinically significant differences in procedural or postoperative outcomes between patients with and without median lobes. Any differences previously observed in the literature may be due to low numbers limiting the ability to adjust for prostate size. Our findings support the hypothesis that GreenLight PVP is an effective modality for prostates with median lobes. Future studies will be required to determine the impact of median lobe presence on outcomes based on the dimensions of the median lobe.