Abstract
Introduction and Objectives:
To evaluate the safety of holmium laser enucleation of the prostate (HoLEP) in patients on oral anticoagulation (OA) with respect to intra- and postoperative bleeding complications.
Methods:
Between January 2013 and October 2016, 2178 patients were included in this study, of whom 94 received direct oral anticoagulants (DOACs) and 151 received vitamin K antagonists (VKAs) before HoLEP. All patients either ceased OA (DOACs) or were bridged subtherapeutically (VKAs, international normalized ratio <2) during surgery. These patients were compared to a sample size of 1933 nonanticoagulated patients.
Results:
A significant longer postoperative stay was noted for the patients on DOACs (5.2 [4–6] days) and VKAs (5.3 [4–5] days) compared to the control group (4.5 [4–4] days). The mean drop in hemoglobin was significantly higher in the VKA group compared to the DOAC and control group. There was a significantly higher rate of postoperative bladder tamponades/secondary coagulation in patients on OA with 6 (7.9%)/3 (3.9%) patients in the DOAC group, 10 (7.4%)/6 (4.4%) patients in the VKA group compared to 37 (2.2%)/21 (2.1%) patients in the control group, respectively (p < 0.001). Eight patients required blood transfusions with a distribution of 1 (1.3%), 3 (2.2%), and 4 (0.2%) patients in the DOAC, VKA, and control group, respectively (p < 0.001).
Conclusions:
Our findings indicate that bridged patients who's DOACs and VKAs were ceased before HoLEP are at higher risk of intra- and postoperative bleeding complications. Nonetheless, HoLEP appears to be a safe and effective procedure in those patients.
Introduction
Symptomatic benign prostatic enlargement (BPE) marked by lower urinary tract symptoms (LUTSs) affect nearly 75% of the aging male population by the age of 70. 1 At the same time, the incidence of thromboembolic diseases increases steadily with more patients requiring oral anticoagulation (OA) at the time of prostate surgery. Depending on the cardiovascular comorbidity, the patients are anticoagulated with antiplatelet (AP) therapy, vitamin K antagonists (VKAs), or direct oral anticoagulants (DOACs).
For many decades, transurethral resection of the prostate (TURP) and open prostatectomy (OP) have been the standard treatment for surgical management of BPE. However, both procedures are accompanied with severe bleeding complications with transfusion rates as much as 6.4% during TURP 2 and 14% during OP 3 in the general population and as much as 33% in patients on OA. 4 Since its introduction in 1998, 5 holmium laser enucleation of the prostate (HoLEP) has been proven to be superior to TURP and OP in terms of functional outcome with less bleeding complications and shorter hospitalization rates, 1 regardless of prostate size. 6 Several studies have reported on the safety and efficacy of HoLEP in patients on OA with regards to AP and VKA in small study cohorts. 7 –9 However, until now there is no literature addressing the intake of DOACs during HoLEP, which are increasingly prescribed to prevent cardiac events. 10
Therefore, the aim of this study was to evaluate the safety and efficacy of HoLEP in patients on OA with DOACs and VKAs in comparison to a nonanticoagulated control group. To the best of our knowledge, this is the largest study, including 94 patients, on DOACs and 151 on VKAs before BPE treatment addressing a matter of high interest due to a steadily increasing intake of OA in the aging male population presenting with more cardiac comorbidities.
Materials and Methods
After obtaining institutional review board approval, we performed a retrospective review of our prospectively collected database of patients who underwent HoLEP between January 2013 and October 2016. Patients who required OA with DOACs and VKAs were identified and compared with the overall database cohort.
The purpose of this study was to assess the immediate and short-term outcomes with regards to bleeding complications in patients on DOACs and VKAs. Therefore, we excluded all patients receiving AP therapy (e.g., aspirin/clopidogrel) and patients with incomplete data for evaluation. From 2950 patients, 2178 were included in further analysis (94 patients on DOACs, 151 on VKA, and 1933 as controls). Demographic data, physical examination with digital rectal examination, and transrectal ultrasound were preoperatively performed in every patient.
The algorithm of managing anticoagulation therapy is based on a standardized clinical protocol. After cardiologic consultation, patients on DOAC therapy ceased OA 48 hours before surgery without control of the international normalized ratio (INR). Whenever possible, coumadin was stopped 10 days before surgery. To avoid cardiac events, bridging with low molecular weight heparin (LMWH) once a day and adapted to weight was performed for patients on DOACs. Patients on coumadin received the same bridging therapy after the INR dropped to 1.5. Patients with artificial heart valves received LMWH twice a day. Postoperatively, all patients were bridged subtherapeutically for 14 days with LMWH before restarting treatment. Coumadin was as well routinely restarted 14 days after surgery depending on the INR and comorbidities.
HoLEP was performed using a 100 W pulsed Ho:YAG laser (Lumenis 100 W, Dreieich, Germany), a 550 μm bare-ended, reusable laser fiber (slimline 550; Lumenis Ltd.), and a 26F continuous-flow laser resectoscope in combination with a mechanical tissue morcellator (R. Wolf, Piranha™, Knittlingen, Germany).
The technique has been previously reported in detail. 11 In all cases, a three-lobe technique was performed. We started with a paracollicular incision toward the prostate capsule. After 5- and 7-o'clock incisions, the middle lobe was enucleated along the prostate capsule and was deposited into the bladder. Thereafter, the left lateral lobe was enucleated with an early apical release of the lateral lobe until 2 o'clock. Then, a diagonal incision of the apical mucosa was made from 2 to 12 o'clock of the prostate. Enucleation of the lateral lobe along the prostatic capsule toward the bladder neck and beyond 12 o'clock was performed. The anterior commissure was then resected, and the left lobe was completely enucleated and pushed into the bladder. The right lobe was enucleated accordingly to the left lobe. After enucleating all lobes, careful hemostasis with 60 W was performed. The prostate tissue was then morcellated and aspirated inside the bladder. After surgery, a 20F three-way foley catheter was inserted for continuous irrigation with normal saline and stopped after 24 hours based on the standard department protocol. Perioperatively, all patients received an antibiotic prophylaxis. After removal of the catheter and the ability to void adequately, the patients were discharged from hospital. All patients were reassessed at 3 months for LUTS evaluation using the International Prostate Symptom Score (I-PSS) and to evaluate further bleeding complications.
All statistical calculations were performed using SPSS 22 (IBM Corp, Armonk, NY) for Mac OS® X. A comparison between the groups (No anticoagulation vs DOACs vs VKAs) was performed using the Mann–Whitney U test for ordinal variables, a chi-square or Fisher's exact test for categorical variables, and the Student t-test for parametric continuous variables. Patient data were expressed as median (interquartile range) or mean (standard deviation). A p-value ≤0.05 was considered statistically significant.
Results
A total of 2178 patients were included in this study, of whom 245 patients received OA and 1933 served as controls. Of the OA study cohort, 94 patients received DOACs, and 151 patients received anticoagulative therapy with VKAs. The most common indications for OA were atrial fibrillation, venous thromboembolism, and presence of a mechanical heart valve. Except for patient's age (70 vs 75 years) at time of surgery, no significant differences were seen between the groups at baseline (Table 1). Preoperatively, 846 (38.8%) patients presented in urinary retention with an indwelling catheter in situ.
Baseline Characteristics
Data as n (%) or median (interquartile range).
I-PSS = International Prostate Symptom Score.
Table 2 lists the perioperative parameters. The mean operative time was 60.3 minutes with a mean morcellation efficiency of 4.6 g/min without significant differences between the groups. The overall hemoglobin decrease was low with 0.9 (0.2–1) g/dL. However, the mean drop in hemoglobin was significantly higher in the VKA group compared to the DOAC group with 1.1 (0.4–1.5) g/dL and 0.6 (0.1–1.2) g/dL, respectively. The mean catheterization time was significantly longer in the DOAC (2.6 [2–3] days) and VKA (2.9 [2–3] days) groups compared to the control group (2.2 days, p < 0.001). Simultaneously, a significantly longer mean postoperative stay was noted for patients on DOACs (5.2 [4–6] days) and VKAs (5.3 [4–5] days) compared to the control group (4.5 [4–4] days) (p < 0.001).
Perioperative Data
Data indicated as median (interquartile range).
Measured from insertion until removal of the resectoscope.
Resected weight/morcellation time.
The immediate and short-term complications are summarized in Table 3. The most common complication was the development of urinary tract infections (UTIs) that occurred in 372 (19.4%) patients without significant differences between the groups. Preoperatively, 38.8% of the patients showed up in urinary retention with an indwelling catheter in situ, and 41.9% were already treated with antibiotics according to an antibiogram.
Postoperative Complications During Three-Month Follow-Up
Data as n (%).
Data as fractions, since no data from all patients available at 3-month follow-up.
UTI = urinary tract infection; n.s. = not significant.
With regards to bleeding complications, there was a significantly higher rate of postoperative bladder tamponades in patients on OA with 6 (7.9%) in the DOAC group and 10 (7.4%) in the VKA group compared to 37 (2.2%) patients in the control group (p < 0.001). Out of these patients, 3 (3.9%), 6 (4.4%), and 21 (2.1%) patients each required a secondary coagulation of the prostate fossa after a failed trial to manage the bleeding conservatively (p = 0.003). Overall, eight patients required blood transfusions due to a significant hemoglobin decrease with a distribution of one (1.3%), three (2.2%), and four (0.2%) patients in the DOAC, VKA, and control group, respectively (p < 0.001).
Indications for reoperation during the first 3 months were bladder neck contractures (0.5% [9 patients]), urethral strictures (0.9% [17 patients]), and secondary apical resection (1.1% [22 patients]) in patients who were not able to void adequately after removal of the urinary catheter. Three months after surgery, I-PSS improved significantly in all groups (20 vs 6) without differences between the groups. Tables 4 and 5 list the age-stratified subgroup analysis of the patients on DOACs and VKAs. Interestingly, the age group from 71 to 80 had a significantly higher I-PSS at the 3 month follow-up mark compared to the other age groups in the DOAC group. In the VKA group, these patients also presented with a higher I-PSS; however, the difference between the subgroups did not differ significantly. Older patients (>80 years) had a significantly longer hospitalization rate in the VKA group compared to patients <80 years (6 vs 4 days), whereas this difference did not pass the level of significance in the DOAC group (5 vs 4 days).
Age-Stratified Outcomes for Holmium Laser Enucleation of the Prostate in Patients Under Chronic Phenprocoumon Therapy (n = 151 patients)
Data indicated as median (interquartile range).
Age-Stratified Outcomes for Holmium Laser Enucleation of the Prostate in Patients Under Chronic Rivaroxaban, Apixaban, or Dabigatran Therapy (n = 94 Patients)
Data indicated as median (interquartile range).
Discussion
OA is frequently used in patients with thromboembolic diseases all over the world. Due to the aging population, this issue becomes increasingly relevant. However, for the use of the “newer” oral anticoagulants such as dabigatran, rivaroxaban, and apixaban, only sparse data exist with respect to bleeding complications and still need to be investigated in large observational studies. 12 The American College of Chest Physicians developed evidence-based guidelines for clinical practice to minimize adverse outcomes and simplify patient management for patients on OA. 13 Although TURP is considered as an operation with a high risk of bleeding, acetylsalicylic acid should be continued during surgery. VKAs and DOACs are supposed to be discontinued due to a standard protocol that has been previously reported in detail. 13
In this study, we could prove that the use of DOACs and VKAs during HoLEP led to more bleeding complications compared to nonanticoagulated patients, however, still significantly lower compared to TURP. 4,14,15 We noted a significant improvement in symptom parameters assessed by I-PSS 3 months after HoLEP, which is comparable to I-PSS of the meta-analysis that was published by Cornu and colleagues. 6 Interestingly, the improvement of I-PSS was significantly less in patients on OA compared to the nonanticoagulated group, especially in the elderly patients >71 years.
At baseline, a significantly higher number of older patients on OA showed up with an indwelling catheter in situ with a higher incidence of UTIs that also influences the perioperative outcome of HoLEP. Furthermore, the analysis of this subgroup revealed significantly larger prostate glands and a higher intake of OA. Patients with perioperative UTIs usually have more irritative symptoms postoperatively and require a secondary treatment with antibiotics. In addition, the elderly patients usually wait much longer and undergo more psychologic strain until they have surgery or are forced to undergo surgery because of the transurethral catheter. The I-PSS in elderly patients is much more affected by other comorbidities such as heart failure, nycturia, neurologic disorders, and so on. At the age of 80, the I-PSS is only correlated to an obstruction in ∼50% of patients. 16 There is no comparable study of subgroup analysis regarding the I-PSS of a single center; however, this should be taken into account when analyzing subjectively measured parameters of large and heterogenous study cohorts.
The mean drop in hemoglobin was 0.9 g/dL, which is low and well comparable to the current literature. 6 Regarding the different groups, the hemoglobin decrease was significantly higher in the VKA group (1.1 g/dL) compared to the DOAC and control group with 0.6 and 0.9 g/dL, respectively. Concerning bleeding complications in patients on OA, only sparse data exist. 12 In total, 252 patients were analyzed in seven studies, 134 after HoLEP 7 –9,17 and 118 after thulium vapoenucleation of the prostate (ThuVEP), 18,19 respectively. The overall transfusion rate was 5.5% without differences between the groups, but significantly higher compared to our transfusion rate of 1.6% in patients on OA. The pooled results of clot retention and secondary surgery due to bleeding were shown to be almost identical in the literature with 6.9% and 3.7% compared to our data with 6.5% and 3.7%, respectively. However, it is important to note that the analysis of postoperative bleeding included as well patients on AP therapy in contrast to our data. 7,9,17
Photoselective vaporization of the prostate (PVP) has proven to be safe in patients on OA. However, only nine studies have been published with patients on VKAs, 20 –28 and only one study, including 13 patients, elucidated the complication rate during the use of DOACs. 20 In total, 806 patients were analyzed in all those studies together with OA during PVP. Nineteen (2.3%) patients developed postoperative bleeding with clot retention, six (0.7%) patients required secondary coagulation of the prostate fossa, and four (0.5%) patients received blood transfusions. Compared to our study cohort, these data are comparable to the nonanticoagulated group and less compared to the DOAC and VKA group. These data indicate why PVP is considered as the best minimally invasive technique for prostate surgery in patients on OA with respect to hemostatic properties. Furthermore, it could also correlate with the significantly shorter hospital stay of the investigated study population that secondary hemorrhage is either not seen or treated in different hospitals after restarting VKA therapy. The commonly performed TURP is still considered as the reference standard in BPE treatment, although the data clearly show a higher risk of bleeding and a higher need for blood transfusions in patients on OA. 4,14,15,29 The incidence of significant blood loss during TURP requiring blood transfusion differs from 1% to 33%. 4,14
Overall, the major problem of all studies concerning OA during prostate surgery is the proposed low level of evidence of three and four due to the study design and the small study cohorts. 30 Another issue by assessing the risk of bleeding during BPE therapy is the absence of standardized clinical trials that investigate clearly defined patient groups with respect to OA and the energy source used (e.g., HoLEP, 7 –9,17 ThuVEP, 18,19 GreenLight laser 20 –23,26,27,30 ). Because of these mixed up patient cohorts without differentiating between AP and VKA therapy, no clear conclusions can be drawn for practicing HoLEP in clinical routine. Not only mixed patient cohorts lead to unintelligibility but also the bridging therapy does not become clear in the study protocols of most studies. During surgery and for the postoperative course it is of great importance whether patients are bridged therapeutically (INR >2) or subtherapeutically (INR <2). 8,9,17 In our study, the OA was ceased in all patients. Patients on DOACs were postoperatively bridged with LMWH once a day, whereas patients on VKA were bridged after the INR dropped to 1.5 once a day and in case of artificial heart valves twice a day. This could explain the lower blood transfusion rate in our study compared to Bishop and colleagues of 7.7% in patients with a mean INR of 2.6. 17
The main limitation of this actual study is the retrospective study design of the prospectively collected data, but an ethical approval for a randomized trial comparing patients on continuous OA with ceased OA is hardly receivable, because of an increased risk of cardiac events in the latter group. Another limitation is the missing availability of the short-term complications according to the Clavien-Dindo classification and the standardized assessment of the patient comorbidities with the Charlson Comorbidity Index. By assessing these two parameters, this study would have been easier to compare to other studies assessing OA intake during BPE treatment. Nevertheless, this study adds value to the current literature in terms of OA during HoLEP and represents the largest series of patients with DOACs and VKAs with regards to perioperative safety and efficacy. Although patients on OA represent a challenging cohort for any transurethral technique, we consider HoLEP as superior compared to TURP due to the comparison of our data with the current literature. But, whenever possible we suggest to cease OA therapy with DOACs or VKAs during transurethral treatment of the prostate.
Conclusions
This study implies that patients on OA are at higher risk for bleeding complications intra- and postoperatively. However, it further proves that HoLEP is a safe and effective procedure for patients with bleeding disorders with little need for blood transfusions.
Footnotes
Author Disclosure Statement
No competing financial interests exist.
