Real-World Experience of Holmium Laser Enucleation of the Prostate with Patients on Anticoagulation Therapy

Introduction

The prevalence of cardiovascular comorbidities necessitating anticoagulation (AC) and antiplatelet (AP) therapy increases with age. ¹ Benign prostatic hyperplasia (BPH) is a disease of the aged with 80% of men affected by 70 years of age. ² Therefore, managing BPH in patients taking oral AC or AP agents is a common challenge for urologists. The management becomes even more challenging when medical treatment fails or is not an option. ³

Traditional therapies such as transurethral resection of the prostate (TURP) and open/laparoscopic/robotic simple prostatectomy are not preferred or safe options for men necessitating anticoagulation. ⁴ Elzayat and colleagues first described holmium laser enucleation of the prostate (HoLEP) as a safe procedure with AP/AC therapy in 2006. ⁵ Since that time, more supporting evidence of smaller cohorts have emerged to support safety of HoLEP with regard to patients necessitating anticoagulation therapy. ^{6 –10} A large meta-analysis of BPH procedures, ¹¹ in addition to the American Urological Association (AUA) guidelines, supports HoLEP in patients necessitating anticoagulation therapy. ¹² However, in these series transfusion rates, ⁶ enucleation time, ¹⁰ and length of stay ^6,7 have been shown to be higher in AP and/or AC groups.

HoLEP technique and laser technology have improved since HoLEP was first introduced over 20 years ago, allowing for benefits in procedural hemostasis. ^13,14 Many of the studies mentioned include historical data that may underestimate the benefits of these new techniques and technology. In a parallel manner, improvements in AC/AP have led to the introduction and widespread use of direct oral anticoagulant (DOAC) drugs. ¹⁵ The use of DOACs in patients has not been extensively studied periprocedurally with HoLEP. We sought to present our recent experience in patients undergoing HoLEP who are taking an AP or AC therapy, including those on DOAC.

Patients and Methods

After obtaining institutional review board approval, we carried out a retrospective review of HoLEP procedures performed by three surgeons (A.E.K., M.E.R., T.L.) from July 2018 to December 2019. This timeframe was chosen to represent a modern cohort of technique and technology. All surgeons were endourology fellowship trained, had performed >100 cases, and were well past the learning curve at the time this study was conducted. HoLEP procedures were performed using a high-powered holmium laser with enucleation settings of 2 J and 40–60 Hz. All three surgeons performed HoLEP with a similar technique. ¹⁶ From these patients, three cohorts were created: no AP/AC (control), AP, and AC. Clopidogrel was the only AP agent included in the AP group. In our experience aspirin 81–325 mg has little impact on surgical outcomes; thus, these patients were included in the control cohort. Warfarin and DOACs, including factor Xa inhibitors and direct thrombin inhibitors, were included in the AC group.

All patients required preoperative clearance for operation, including recommendations from the AP/AC prescribing provider. The AP/AC prescribing physician was asked to provide surgical risk assessment and recommendations on date of cessation and restarting of AP/AC therapy. The plan for AP/AC continuation was modified if the patient called reporting significant hematuria postoperatively, but patients were advised to confirm this plan with their prescribing provider.

Demographics are presented as mean with standard deviation or number and percentage. A Kruskal–Wallis test was performed to detect differences between the three groups. When comparing the control group with either the AP or AC group, continuous variables were analyzed with a Mann–Whitney U test and categorical variables were assessed with Fisher's exact test. All statistics were performed with SAS (version 9.4; SAS Institute, Cary, NC).

Results

From July 2018 to December 2019, there were 472 patients with AP/AC data who underwent HoLEP. There were 377 (79.9%) in the control group, 30 (6.3%) on AP, and 65 (13.8%) on AC preoperatively. The mean age was 71.0 ± 9.3 years, body mass index (BMI) was 29.1 ± 17.6, and rate of catheter-dependent retention was 33.1% (n = 156). Mean prostate-specific antigen at operation was 8.3 ± 14.6 ng/mL and prostate size (based on preoperative transrectal ultrasound, computed tomography, or magnetic resonance imaging) was 103.2 ± 68.6 mL. The rate of alpha-1 antagonist use was 74.1% (n = 350) and 5-alpha reductase inhibitor was 28.7% (n = 135).

When comparing preoperative demographics across the groups (Table 1), there was only a significant difference in age (p < 0.001). There were no differences in perioperative outcomes, including enucleation time, morcellation time, total energy used, or specimen weight (Table 2). A similar number of patients underwent concomitant cystolitholapaxy (also performed with the holmium laser) between groups. In the control group, one patient required return to the operating room 7 days postoperatively for cystolitholapaxy because of a retained bladder and prostate stones causing obstruction. This patient had a large bladder diverticulum that obscured removal of all the stone fragments at the time of the initial procedure.

There was a statistically higher rate of postoperative complications across AP/AC groups (p = 0.035), but not for unplanned emergency department (ED) visits (p = 0.557) or Clavien ≥3 complications (p = 0.16). Overall, there were five patients (1%) who required bedside catheter irrigation of clots and one patient a formal clot evacuation under anesthesia (0.2%), of which five were in the control group and one was in the AC group. The one patient in the AC group who required a bedside catheter irrigation was on a DOAC that was held perioperatively. There was no difference in the need for bedside clot evacuation between the groups (p = 0.667). In patients with a postoperative hemoglobin available (n = 247), there was no difference in change of hemoglobin from preoperative to postoperative between the groups. Two patients required blood transfusion and both were in the control group. One transfusion was performed at hemoglobin level of 8.6 in the context of a perioperative stroke. There was a statistically significant higher rate of complications in the AP group compared with the control group (31.3% vs 11.1%, p = 0.0343), but complications were not higher in the AC group compared with the control (22.2% vs 11.1%, p = 0.099). Table 3 provides a list of all complications by Clavien grade and group. Postoperatively (Table 4), there were no differences in postoperative voiding measures.

Table 5 provides the differences between the control group compared with the AP and AC groups individually. Between the AC group and the control group, only age was significantly different (75.5 ± 8.3 years vs 70.1 ± 9.34, respectively, p < 0.001). There was a statistically significant, but not clinically significant difference in BMI between the control group and the AP group (29.1 ± 19.6 vs 30.2 ± 5.28, respectively, p = 0.029). There was a lower rate of preoperative alpha antagonist use in AC compared with the control group (64.6% vs 75.7%, p = 0.049). There was an increase in 90-day complications in the AC group compared with the control group (p = 0.033), but not in Clavien ≥3 complications (p = 0.0542). There was no decrease in 90-day complications compared with those taking or not taking finasteride in the AP group (p = 0.38) or the AC group (p = 0.21). There were no other significant differences between control and the AP or AC groups.

There were no differences in preoperative subjective or objective voiding parameters between groups (Table 4). In 262 patients (56.3%) with postoperative follow-up, there were no differences among groups with the exception of a higher AUA symptom score in the AC group compared with the control group (11.1 ± 7.55 vs 8.34 ± 6.68, p = 0.04). Comparing pre- vs. postoperative voiding parameters within each group, there were significant improvements in all patients for all variables.

Of the 65 patients on AC therapy, 23 (35.3%) were on warfarin and the remaining 42 (64.7%) were on DOACs. At the time of the procedure, 6 (9.2%) AC patients had their anticoagulation therapy discontinued but bridged with heparin or low-weight molecular heparin, and one patient continued AC through operation. For patients on AP therapy, one patient (3.3%) continued clopidogrel through the time of procedure. The remainder of the patients stopped AP/AC perioperatively.

A comparison within the AC group of those on warfarin versus DOAC is given in Table 6. There was no demonstrated difference in preoperative demographics. The mean INR on the day of operation for patients on warfarin was 1.14 ± 0.13. Of note, patients on a DOAC were significantly more likely to have undergone a prior BPH procedure than those on warfarin (23.8% vs 0%, p = 0.011). There was no difference in operative parameters or postoperative hemoglobin drop. Patients on warfarin had a lower rate of catheter removal within 24 hours than those on a DOAC (64% vs 90%, p = 0.009). There was also a higher 90-day complication rate in those on warfarin compared with DOAC (52% vs 14%, p = 0.003), but no difference in Clavien 3 or higher complications (13% vs 4.8%, p = 0.337). Postoperative emergency room (ER) visit rates were also higher in those on warfarin than those on DOAC (35% vs 4.8%, p = 0.003).

Discussion

Herein we demonstrate that in the population of patients undergoing HoLEP, 20% are on clopidogrel, warfarin, or DOAC. We did find an increase in postoperative complications in the AP group (p = 0.034), but not in the AC group (p = 0.099). However, we did not find an increase in clot evacuations or transfusions as a result of the use of AP/AC drugs. These results are supported by prior studies of HoLEP and AP/AC drugs. ^{3,6 –8,10} The literature reports conflicting data on the impact of AP/AC on operative and perioperative outcomes. Our enucleation time was no different between the AP and AC compared with control groups, which differs from prior publications with mixed results showing both slower ^6,9 enucleation ¹⁰ and faster operative times. ⁷ Although there are differing results in the literature regarding transfusion rate, ⁶ our AP and AC cohorts did not experience a higher rate of transfusion compared with the control. However, the overall rate of transfusion for our entire cohort was very low (0.4%), which differs from the current literature rates of 2%–5% ^17,18 and is most likely a reflection of improved laser technology and evolution of HoLEP technique.

Although HoLEP has been shown to be feasible on uninterrupted AP/AC, ^{5,7 –10} it is desirable for a patient to have this therapy interrupted if deemed safe. The majority (97%) of patients in our cohort were able to hold their AP/AC drugs before HoLEP, at the discretion of the prescribing provider. Discontinuation of the AP/AC medication recommendation is in line with current data on perioperative anticoagulation. The BRIDGE trial demonstrated that in patients with atrial fibrillation on warfarin, interrupting AC without bridging was noninferior with respect to preventing arterial thromboembolic events and predictably had a lower rate of bleeding. ¹⁹ Aspirin is a common agent used in patients on AC (between 30% and 40%, ^19,20 ) and 60% of patients remained on aspirin in the BRIDGE trial. ¹⁹ Given the supporting data from the BRIDGE trial and our observation that aspirin had little influence on intra- or postoperative outcomes, we did not include it as an AP. With this in mind, our data support the BRIDGE trial with no increase in complications in patients holding or continuing anticoagulation at the time of operation.

One unique aspect of this study is the sub-characterization of those on warfarin and DOAC. Since the introduction of DOAC, the rate of prescribing has increased significantly compared with that of warfarin. ¹⁵ In our AC cohort, DOACs represented almost two-thirds of the cohort. We found that DOACs did not present a greater risk and that in fact complications and unplanned ER visits were higher in those with warfarin. Most of the complications in the AC group occurred after restarting warfarin. In addition, the passage of the initial voiding trial was significantly less in those patients on warfarin. There were no demographic differences to explain the varying outcomes between those patients using warfarin and those using DOAC, and further analysis with a larger patient cohort is required.

A large meta-analysis of DOAC vs warfarin phase 3 trials did not find any association with worse periprocedural outcomes/complications with use of either drug class when anticoagulation was interrupted. ²⁰ We speculate that the anticoagulation level on warfarin is variable for the patient. In the perioperative period, the patients are often on antibiotics that can affect the impact of warfarin, and many patients become supratherapeutic compared with their target resulting in unexpected bleeding. DOAC medications, on the contrary, are not affected by antibiotics typically used for genitourinary indications ²¹ and provide more predictable anticoagulation.

HoLEP has multiple advantages over other endoscopic BPH procedures, which allow for favorable outcomes in patients on AP/AC therapy. ³ The holmium laser is hemostatic, and newer technology has improved this characteristic by increasing not only energy, but method of energy delivery. ¹⁴ In addition, enucleation procedures allow for a single plane of operation, resulting in efficient initial coagulation of vessels as opposed to resection procedures that require repetitive vessel coagulation as resection is carried deeper. Using principles from simple prostatectomy, HoLEP allows for enucleation to be safely performed on AP/AC medications when open and minimally invasive techniques cannot. ⁴ The simplicity of working in one plane results in increased efficiency of HoLEP as gland size increases as opposed to TURP. ¹³ The differences in postoperative bleeding complications are significantly higher in TURP than HoLEP ⁴ with a relative risk of transfusion of HoLEP compared with TURP of 0.20 (confidence interval, 0.08–0.47). ¹² Additional risk of TURP may be owing to a possible increase in postoperative hypercoagulability in patients undergoing TURP than HoLEP because of increased thrombin generation. ²² Furthermore, our study demonstrates safety of HoLEP in AP/AC populations, with successful same-day discharge in AP and AC cohorts, which is not possible with TURP or simple prostatectomy.

Our study is not without weaknesses. This is a retrospective review of a single, tertiary care center. A prospective trial would be the ideal way to assess if AP/AC therapy should be continued, bridged, or held for the best outcome at the time of operation; however, we defer to the patient's cardiologist, hematologist, or primary care provider to assess risk/benefit of holding anticoagulation perioperatively. Although we do contact patients within 7 days to discuss early operative status and pathology results and referring providers will contact us if postoperative complications and issues arise, not all patients return for follow-up appointments given the tertiary nature of our center. Therefore, some outcome results may be missing, but missing variables would be distributed across all groups of the study. Despite these limitations, we provide a description of warfarin vs DOAC in patients undergoing HoLEP, which the literature generally treats in the same category.

Conclusions

HoLEP is a safe procedure in patients on clopidogrel, warfarin, or DOAC therapy, especially in the real-world experience of intermittent dosing periprocedurally. We found that complications rates, although low grade, were higher in AP patients compared with AC patients. Within the AC group, we found that warfarin decreased success of catheter removal, higher 90-day complication rate, and higher return to the ED rate compared with DOAC. Thus, in patients who require AC and are undergoing HoLEP, DOACs may be superior to warfarin regarding postoperative outcomes, but further study is needed to elucidate this relationship.