Intravesical Hemostatic Clip Migration After Robotic Prostatectomy: Case Series and Review of the Literature

Introduction

The Hem-o-lok™ Ligating clip (Weck Surgical Instruments; Teleflex Medical, Durham, NC) is a routinely used hemostatic tool in open, robotic, and laparoscopic surgeries. It is estimated that more than 100 million ligation clips have been used worldwide. ¹ It has been routine practice in our institution to use these clips for hemostatic control of the vascular pedicles during robotic-assisted laparoscopic radical prostatectomy (RALRP). The clips are made of sterile plastic and are therefore nonabsorbable. Migration of these clips has been reported intermittently throughout the literature dating back to ∼2008, primarily in single or dual patient case reports.^2–6 Blumenthal and Sutherland reported on what appears to be the largest case series thus far in the literature identifying two of their patients with bladder neck contractures secondary to migrated clips found in the urinary tract and 1 patient who was found to have migrated clips after being diagnosed with an anastomotic leak. ⁷ Other instances of migrated Weck clips include migration into the bladder with resultant stone formation around the clip and migration into the rectum.^8,9

In this review, we present our own case series of patients who have experienced intravesical Weck clip migration following robotic-assisted laparoscopic radical prostatectomy. Through an open discussion of our experience with this complication, we hope to stimulate further study into patient and operative characteristics, which may predict clip movement and suggest modified usage of this product in RALRP.

Materials and Methods

Implementation of our robotic-assisted laparoscopic urologic surgical program began in November 2006. An Institutional Review Board approved retrospective chart review was conducted on all patients who underwent RALRP between 2006 and 2011. Data were drawn from a comprehensive robotic prostatectomy database, in which data had been collected by an independent third party in a blinded prospective manner. The study population was then defined as those men who were confirmed to have undergone RALRP and, subsequently, presented to their surgeon's office with signs or symptoms of bladder pathology (including dysuria, urgency, frequency, difficulty voiding, or hematuria) and were then identified as having a migrated Weck clip in the urinary tract. Identification occurred following an office-based workup, including urine analysis, urine culture, and flexible cystoscopy. Patient age, preoperative Gleason Grade, operative time, intraoperative blood loss, length of initial postoperative hospital stay, and final pathologic tumor stage were collected from the comprehensive robotic-assisted laparoscopic radical prostatectomy database described above. Patient follow-up data in this database have been collected through calendar year 2013 providing follow-up times ranging from 2 years (for men who underwent RALRP in 2011) to 7 years (for men who underwent RALRP at the start of our experience in late 2006).

If a patient was found to have a clip migrated into their bladder, urethra, or bladder neck, he was then counseled regarding this finding and consented for surgical removal of the clip in the operating room. Intraoperative techniques required to remove the often encrusted or impacted Weck clips were recorded. Primary data points regarding the operative extraction included the type of intervention (laser lithotripsy, litholapaxy, or simple extraction) and the time to presentation from the patient's original RALRP was recorded.

Preoperative and intraoperative information and secondary procedure characteristics were compared across the two cohorts of patients undergoing RALRP using chi-square analysis for categorical variables.

Results

Five-hundred seventy robotic-assisted laparoscopic radical prostatectomies were performed at our institution between 2006 and 2011. Eight patients (1.4%) subsequently presented to their surgeon's office with complaint of lower urinary tract symptoms, were identified to have a migrated Weck clip in their urinary tract, and required an endoscopic extraction in the operating room. Perioperative characteristics between men in the two cohorts regarding their initial RALRP may be found in Table 1. There were no significant differences in the age (60.2 years vs. 62.4 years), preoperative Gleason Grade (6.6 vs. 6.8), or console time (170.6 minutes vs. 173.5 minutes) between the two groups. Estimated blood loss among those who did not experience clip migration was 260 mL compared to 240 mL among those men in whom a Weck clip had migrated, a difference which did not reach statistical significance. With regard to postoperative factors, men in whom a clip had migrated were hospitalized for a significantly longer period of time post prostatectomy (7.6 days vs. 2.1 days, P < .01) and, among the 8 men who demonstrated clip migration, three (37.5%) required transfusion compared to only 11 of 562 (1.9%) men who did not develop this complication (P < .01). Among men requiring transfusion, there was also a statistically significant difference in the volume of blood products transfused between the two cohorts, with men in the migrated clip group requiring a greater volume (1.4 units vs. 0.05 units, P < .01). The most common final pathologic tumor stage was pT2c and did not significantly differ between the two groups. The average time to presentation with lower urinary tract symptoms after initial RALRP among those found to have a migrated clip was 1.75 years.

Discussion

Weck clip migration is a recognized, but rare potential complication following robotic-assisted laparoscopic radical prostatectomy.^2–6 Banks and Ramani initially identified this complication in 1 patient following laparoscopic prostatectomy in 2008 followed by Blumenthal and Sutherland, who identified migrated clips in 2 of 4 patients being treated for post-RALRP bladder neck contracture and 1 patient with an anastomotic leak.^2,7 Shin later described a fairly unique finding in which four Weck clips found floating in the bladder at the time of a urethral dilation had minimal encrustations compared to the prior literature, which had all noted moderate encrustation on the clips and required initial manipulation before extraction. ⁵ Overall, as summarized by Cormio and supported by Long and Palou's descriptions of metal clip migration, the vesicourethral anastomosis is felt to be a vulnerable location for this complication.^4,10,11

While only occurring in just over one percent of the total number of robotic-assisted laparoscopic radical prostatectomies at our institution between 2006 and 2011, we believe our experience has demonstrated that clip migration should be considered during the workup of post-RALRP patients with recurrent urinary tract infection (UTI), persistent lower urinary tract symptoms, or hematuria. Furthermore, this study indicates that even greater concern for this complication is indicated among men who experienced either prolonged hospitalizations or required transfusion with one or more units of packed red blood cells following their prostatectomy. Operative time, intraoperative blood loss, and preoperative Gleason Grade did not serve as predictive factors in clip migration. Office cystoscopy is an effective method for identifying patients with migrated clips. In one instance, a clip located at the bladder neck was able to be removed in the office; however, the overwhelming majority of migrated clips required return to the operating room where interventions such as laser lithotripsy, litholapaxy, and fulguration of bleeding extraction sites were necessary.

Patients who had clip migration into the bladder had a longer average hospital stay after their RALRP and required transfusion not only more frequently but also of a greater volume of blood products, following their procedure. This finding is interesting particularly in light of our earlier statement that intraoperative blood loss was not a significant predictor of our outcome. We hypothesize that the discordance between intraoperative estimated blood loss and need for postoperative transfusion may indicate the occurrence of delayed pelvic bleeding, which may result in a pelvic hematoma and serve to distract the posterior anastomosis allowing for free clips to migrate into the anastomosis or bladder.

This study does have some important limitations that need to be considered as well. Based on the significant discrepancy in the number of men in each of our two cohorts, we recognize the limitations of the statistical analysis that can be performed between the two groups. Ideally, a larger volume of men who experienced clip migration would strengthen the validity of our results, but achieving a more even distribution was not possible due to the limited nature of this complication. Selection bias may also have influenced our results given that the true denominator of all men in whom a clip migrated is not known. Data are only available for those men who were symptomatic enough to seek medical evaluation and, subsequently, were deemed appropriate for workup, which included office cystoscopy. Based on routine close follow-up of postprostatectomy patients and the significance of a migrated clip, however, it seems unlikely that there exists a large number of men in our cohort who have not been evaluated after surgery and who are symptomatic from undiagnosed migrated Weck clips. Finally, only 2 years of follow-up data are available for men who underwent RALRP in 2011 in comparison to almost 7 years of follow-up data being available for men who had this surgery early in our series. Having identified 1.75 years to be the average time between initial surgery and symptomatic identification of clip migration, it is possible that we have underestimated the true rate of migration among men toward the latter part of our study period.

Based on our experience, we have altered the utilization of Hem-o-lok Weck clips during RALRP. We now limit the number of clips used per case to six to eight clips. Every attempt is made to only place clips at the proximal end of the prostatic pedicles. Suture control of the distal aspect of the pedicle or to control bleeding along the neurovascular bundle is utilized as opposed to clips. We additionally use thrombin-based products along the neurovascular pedicles for additional hemostasis. All free floating clips are removed from the surgical field. The goal is to limit clip placement in the vicinity of the bladder neck or anastomosis and assure there are no free clips within the surgical field. Given the delay in presentation from the initial surgery to clip migration of ∼1.75 years, more longitudinal research is needed to determine if this change in surgical technique will be effective in mitigating clip migration. We plan to further validate the effectiveness of our change in operative protocol in reducing the number of migrated Weck clip complications as more data are collected.

Conclusions

Eight patients who underwent RALRP at our institution between 2006 and 2011 developed postoperative lower urinary tract symptoms that were determined to be secondary to a migrated Weck clip and, subsequently, required surgical extraction. Men with recurrent UTI, bothersome voiding symptoms, or hematuria following RALRP should be considered for cystoscopic evaluation, especially men who experienced a longer hospitalization or required large volume blood transfusions following their initial surgery. We postulate that clip dislodgement may lead to formation of a pelvic hematoma resulting in a distraction of the vesicourethral anastomosis and migration of the clip into the tissue defect. Therefore, we have limited the number of clips placed and their position to the proximal portion of the prostatic pedicles in an effort to minimize clip-related complications.