Initial Experience with Aspirin Use During Robotic Radical Prostatectomy

Introduction

Prostate cancer is the second leading cause of cancer death in American men. Approximately 192,000 new cases were diagnosed in 2009, and deaths from prostate cancer were estimated at 27,000. ¹ The current lifetime risk of prostate cancer for men living in the United States is estimated at approximately 1 in 6. Between 2001 and 2005, the median age at diagnosis for white American men was 68 years, and the median age at death was 80 years; the median age at diagnosis for black men was 65 years, and the median age of death was 77 years. ²

Unfortunately, this same population of older men is also at risk for coronary artery disease. The National Health and Nutrition Surveys suggest that 13.2 million persons in the United States have coronary artery disease, half of whom have had myocardial infarctions. ³ Stenting is now performed in the majority of percutaneous coronary interventions. A review of almost 150,000 procedures performed at 139 hospitals in the National Cardiovascular Data Registry between 1998 and 2000 found that stents were used in 77% of cases. ⁴ New cardiology guidelines recommend continuous antiplatelet therapy for some patients with cardiac stents. Aspirin (acetylsalicylic acid [ASA]) use is relatively contraindicated during urologic surgery because of the increased risk of bleeding. Robotic radical prostatectomy (RobRP) is increasing in use. It is associated with decreased intraoperative blood loss, improved recovery time, and equivalent oncologic outcomes compared with open surgery.^5–8 We have selectively continued ASA use during RobRP for men at high risk for coronary artery stent occlusion. The objective of this study is to compare the outcomes of men who underwent RobRP on ASA with those of a cohort of men who were not on ASA during RobRP.

Patients and Methods

We performed an Institutional Review Board–approved retrospective review of our urologic oncology database, in which registration, diagnostic, treatment, and follow-up details are prospectively entered for all men with prostate cancer cared for at our institution. We reviewed RobRP procedures performed by a single surgeon (CJK) in patients with prostate cancer between October 2007 and February 2010. We screened patient records for cardiac stents and aspirin status. We obtained preoperative cardiology consultation for patients with cardiac stents concerning continued aspirin use. We reviewed complete demographic, medical history, prostate cancer diagnostic details, D'Amico risk group, ⁹ 2002 TNM stage, ¹⁰ details of surgery (duration, blood loss, intraoperative blood transfusions), and outcomes (margin status, complications, length of hospital stay). We estimated operative blood loss in collaboration with our anesthesiologist. We also reviewed postoperative data, including transfusion requirements, length of hospital stay, hematocrit changes, and complications. ¹¹

RobRP was performed using the da Vinci^® S model (Intuitive Surgical, Sunnyvale, CA). Dissection was performed with combination of sharp dissection and selective electrocautery. Hemostasis was achieved with a combination of electrocautery, surgical clips and thrombin–gelatin matrix (Floseal^®, Baxter, Deerfield, IL). Electrocautery was not used at the neurovascular bundle during nerve sparing procedures. All patients on ASA had bilateral nerve-sparing surgery. No differences in technique were used in patients on perioperative ASA. Lymph node dissection was performed in patients with D'Amico intermediate- or high-risk prostate cancer. ¹² We removed our specimens through the umbilical trocar site with a specimen entrapment bag (Endo Catch™, Covidien, Mansfield, MA). At the end of the procedure, we systematically inspected the pelvis to assure hemostasis and placed a Jackson–Pratt drain at the end of the procedure. Patients were discharged when they were ambulatory and could care for their catheter, and their pain was adequately controlled. We followed up postoperatively at 1 week, 6 weeks, and every 3 months thereafter.

We performed statistical analysis with SPSS (SPSS Inc., Chicago, IL). Kruskal–Wallis and chi-squared tests were used to compare continuous and categorical variables between groups, with P<.05 considered to be significant.

Results

A total of 249 patients underwent robotic prostatectomy. Of these, 6 men with cardiac stents had continued ASA (Group 1), and 7 men with cardiac stents discontinued ASA (Group 2) for the RobRP. The remaining 236 patients served as our control group (Group 3). Patient age, body mass index, biopsy Gleason sum, clinical T stage, D'Amico risk, surgical details, and postoperative details were compared between groups (Tables 1 and 2). Follow-up ranged from 0 to 27 months.

No significant difference was seen regarding age (P=.233), biopsy grade sum (P=.233), clinical T stage (P=.246), or D'Amico risk (P=.146). Patients without cardiac stents were found to have a significantly lower mean prostate-specific antigen at diagnosis than patients with stents (6.6±4.2 ng/mL versus 9.4±4.6 and 9.2±6 ng/mL, P=.041). Operative details were comparable among the groups with respect to mean operative times (193±25, 190±28, and 187±34 minutes, P=.653), estimated blood loss (180±84, 111±35, and 156±91 mL, P=.329), hematocrit changes (–8.8±2.9%, –8.1±5.7%, and –9.3±3.4%, P=.867), and transfusion rates (P=0.820). Seven patients in Group 3 (control group) did receive postoperative blood transfusions; however, all these patients had underlying hematologic and hepatic disease. No patients in Group 1 or 2 required blood transfusions. Furthermore, there were no significant differences in length of hospital stay among these groups (mean, 1.3±0.5, 1.4±0.8, and 1.2±0.7 days, P=.160).

No significant difference was seen in either low-grade (Clavien Grade I or II) or high-grade (Clavien Grade III–V) complications (P=.834 and .773, respectively). No patients in Group 1 or 2 had significant complications, including perioperative myocardial infarctions. In Group 3, however, 8 patients required procedural interventions: 3 (1.3%) for symptomatic lymphoceles, 1 (0.4%) for psoas abscess, 2 (0.8%) for anastomotic leak, and 2 (0.8%) for incisional hernia. Patients had no sequelae after intervention. There was one death in our series. The patient was discharged on postoperative day 1 but returned to the emergency department 1 week later for abdominal pain. During his evaluation, he vomited and experienced an aspiration event resulting in acute respiratory distress, sepsis, and ultimately multiorgan failure despite medical therapy.

Table 3 illustrates individual demographic, surgical, and postoperative details of patients with continued perioperative ASA use. All patients who were continued on ASA therapy successfully underwent RobRP, without need for conversion.

Discussion

Our initial series suggests the feasibility and safety of performing a RobRP on patients who continue ASA. This is important because some patients with coronary stents may require lifelong antiplatelet therapy. Holding antiplatelet medications can increase the rate of acute coronary thrombosis. Ferrari et al. ¹³ interviewed 1236 patients hospitalized for acute coronary syndrome and possible ASA interruption. On review, 4.1% of coronary events and 13% of recurrences involved ASA withdrawal within 10 days of interruption. Other authors have had similar findings.^14,15 Patients who discontinue ASA after being on chronic ASA therapy may experience acute myocardial infarction within 10 days of withdrawal as a consequence of progressive recovery of platelet cyclooxygenase activity, which can initiate a rebound effect leading to acute coronary thrombosis. ¹⁵

In a meta-analysis, Burger et al. ¹⁶ estimated that patients undergoing most surgical procedures have a 50% increased risk of bleeding complications. Urologic and pelvic surgeries do have increased bleeding risks, especially with antiplatelet therapy. During transurethral prostatectomy, the blood transfusion rate was increased by a factor of 2.7 in patients on ASA compared with control groups, ¹⁷ and it is important that there were two fatalities in this series. Increased bleeding from transurethral surgery procedures may be due to the inability to easily tamponade the surgical site and the presence of urinary plasminogen activator. To our knowledge no studies of open or laparoscopic radical prostatectomy or RobRP and antiplatelet use have been published. In our study, we found no significant difference among the groups with respect to estimated blood loss, transfusion rate, or change in hematocrit.

The introduction of laparoscopic and robotic prostatectomy was accompanied by a decrease in perioperative blood loss and transfusion, possibly because of pneumoperitoneum and better visualization of vascular structures.^18–20 All patients who were maintained on ASA successfully underwent RobRP and, compared with our control group, had no differences in median hospital stay, operating room time, estimated blood loss, or overall complication rates. Subjectively we did notice a subtle difference in the ease of bleeding; however, this was easily controlled with selective electrocautery and clips. Whether patients on ASA would also have low blood loss during open surgery is not clear.

Our study does have several limitations. The treatment group in this initial study comprised only 6 patients, clearly limiting its statistical power and the conclusions we can infer. This retrospective cohort study is also limited because of potential selection bias and confounding bias. With such a small sample size, we could potentially underestimate bleeding risk. Moreover, many of our patients are outside referrals with sometimes limited postoperative follow-up. However, we did follow up all patients on ASA for at least 6 weeks.

In conclusion, our initial experience suggests that continuing ASA therapy for men undergoing RobRP is not associated with increased in operative time, bleeding, transfusions, hospital stay, or complications. Further study is required to validate these observations.