All articles by Yaniv Cosacov, Clinical Fellow, Minimally Invasive and  Bariatric Surgery, Cleveland Clinic Florida

Dear Readers,

Welcome to the May issue of The Surgeons’ Lounge. Here we feature the first of a two-part series that covers highlights from the 24th Annual Jagelman/34th Annual Turnbull International Colorectal Disease Symposium, hosted by Steven Wexner, MD, from Cleveland Clinic Florida in Weston and Feza Remzi, MD, from Cleveland Clinic. It is always our pleasure to bring you annual updates in the diagnosis and treatment of colorectal diseases from the leaders in this field. Stay tuned for the second half of this coverage that will appear in the June issue.



Samuel Szomstein, MD, FACS
Editor, The Surgeons’ Lounge

Dr. Szomstein is associate director, Bariatric Institute, Section of Minimally Invasive Surgery, Department of General and Vascular Surgery, Cleveland Clinic Florida, Weston.

A Message From the Symposium Director

Steven D. Wexner, MD, PhD (Hon), FACS, FRCS, FRCS(Ed)
Steven D. Wexner, MD, PhD (Hon), FACS, FRCS, FRCS(Ed)
Professor and Chair, Department of Colorectal Surgery
Cleveland Clinic Florida, Weston

It is with great pleasure that we present to you a recap of some of the highlights of our recent 24th Annual David G. Jagelman/34th Annual Rupert B Turnbull International Colorectal Disease Symposium that was held at the Marriott Harbor Beach Hotel in Fort Lauderdale, on February 12-17, 2013. Almost 100 lectures were delivered by more than 60 internationally acclaimed faculty. The presentations we have selected represent some of the highlights of the most innovative, controversial and current challenges in colorectal surgery.

We wish to thank all of the faculty who participated in our recent course, and in particular to thank the speakers who have taken additional time to contribute to The Surgeons’ Lounge column. As always, we wish to thank Samuel Szomstein, MD, FACS, the column’s editor and Frederick L. Greene, MD, FACS, senior medical adviser of General Surgery News, for allowing us to highlight some of the features of our course in this column.

We are in the final stages of completing the program for the silver anniversary (25 years) of the Annual David G. Jagelman /35th Annual Rupert B Turnbull International Colorectal Disease Symposium. This landmark event will be held on February 11-15, 2014, again at the Marriott Harbor Beach Hotel in Fort Lauderdale, Fla. We look forward to seeing all of the readers of General Surgery News at this event. In exchange, we can promise you an absolutely unforgettable educational experience, courtesy of the high-quality faculty who will be with us at that time.

Assessment of Technical Competency In Surgery

Eric J. Dozois, MD, professor of surgery and program director, Department of Colon and Rectal Surgery at Mayo Clinic in Rochester, Minn., discussed the importance of assessing surgical competency.

In the original charter of the Royal College of Surgeons of 1582, it was declared that “no matter of person shall employ our said craft of surgery unless he be worthy and expert in all the subjects belonging to these said crafts and be diligently and advisedly examined.” Thus, the question is: Are we properly assessing, among all areas of qualification and certification, the technical ability of the surgeon?

Currently in the United States, once a resident is signed off by the program director, he or she can either go directly into practice or complete a fellowship to become eligible for board certification in a specific surgical specialty. This traditional assessment of technical competency is not objective, and is conducted only through observation. It is not considered valid or reliable. Given that technical expertise is a cornerstone to successful care of patients in the surgical profession, are surgeons comfortable with the current process? Is a better training model needed? Furthermore, should this be in the certification or recertification pathway?

How can we do it better? Several approaches to the assessment of technical competency have been evaluated. The most recognized approach is the Objective Structured Assessment of Technical Skills (OSATS), which was developed at the University of Toronto, in Canada, in the mid-1990s. It consists of a “task-specific checklist tool” and a “global rating scale” for assessing performance. This method was found to be valid and reliable to assess technical skills in trainees. The OSATS includes several operative competency scores: respect for tissues, time and motion, instrument handling, suture handling, flow of operation and final outcome.

Other tools that have been developed include the Global Rating Index for Technical Skills (GRITS), the Operative Performance Rating Scale (OPRS) developed by a group at Southern Illinois University, the Hopkins Assessment of Surgical Competency (HASC), and specific Procedure Based Assessments (PBA), which is popular in Europe. All of these tools have focused on objective assessment of technical skills. These assessments take place in either the operating room or a simulated surgical environment.

The quality of the tools discussed has been studied, and in a recent review in the British Journal of Surgery, the validity and the reliability of objective methods for technical skills assessment within surgery and gynecology were reported. A total of 104 assessments were included, and 27 consisted of intraoperative evaluation while the rest were simulation-based assessments. Virtual reality and OSATS were the most studied. The conclusion was that most of the tools were valuable for feedback and measuring progress of training, but few could be used for examinations or credentialing. OSATS was the most standardized and validated assessment.

The American Board of Colon and Rectal Surgery (ABCRS) has set a goal, to be accomplished over the next five years, to move away from “case numbers” as the marker of technical proficiency and instead use direct, “validated” skills assessment tools to define candidates’ technical competency. The goal is to determine if a single assessment tool could be used for certifying colorectal trainees. A pilot study has been designed by members of the ABCRS’ Operative Competency Evaluation Committee to answer whether a reliable and valid objective assessment of colorectal technical skills could be developed, and to evaluate whether an examination could have the ability to discriminate between candidates based on level of training.

The study entitled “A Novel Approach to Assessing Technical Competence of Colorectal Surgery Residents: The Development and Evaluation of the Colorectal Objective Structured Assessment of Technical Skills (COSATS)” will be published in Annals of Surgery later this year. The study included an eight-station examination of 20 trainees, 10 general surgery residents and 10 colorectal surgery residents. The trainees were examined using a task checklist from the global rating scale.

The analysis found that an overall checklist and global rating scores effectively discriminated resident groups. The colorectal surgery residents significantly outperformed general surgery residents in all eight stations. With a pass/fail assessment, nine out of 10 colorectal surgery residents passed compared with three out of 10 general surgery residents. The results were very encouraging, and as a result, a pilot study is in progress where an examination of skills is performed as part of the certification process.

Dr. Dozois concluded that the current method of evaluating technical skills only at the completion of training can serve no longer, and if the goal of certification is to ensure that a surgical candidate is competent in all aspects of the surgical profession, it is time to include technical skills assessment as part of surgery credentialing.

Laparoscopic Ventral Rectopexy for Rectal Prolapse Syndromes

Professor Andre D’Hoore discussed his experience at the University Hospital Gasthuisberg in Leuven, Belgium, for the treatment of rectal prolapse syndromes using the laparoscopic approach for ventral rectopexy.

External or internal rectal prolapse and rectocele are the most difficult clinical problems in colorectal surgery. Symptoms vary from obstructive defecation to fecal incontinence, resulting from chronic sphincter damage.

Several surgical procedures have been developed in an attempt to restore these conditions. To date, the surgical community has not yet accepted a standard method, although abdominal rectopexy is considered superior to perineal or transanal approaches because of lower recurrence rates and better functional outcomes. However, the induction or worsening of postoperative constipation was observed as the most common side effect of rectopexy. An inherent step in classic rectopexy is the full mobilization of the rectum. Autonomic nerve injury during extensive posterolateral rectosigmoid mobilization may lead to postoperative dysmotility and impaired evacuation. In contrast, transanal partial rectal resection or plication may induce or worsen incontinence. Laparoscopic ventral rectopexy (LVR) using a polypropylene mesh has been introduced to combine the good functional outcome of the abdominal procedure while avoiding both postoperative constipation and incontinence.

In a study between January 1999 and December 2008, 405 patients underwent LVR for rectal prolapse syndromes. The median age of participants was 55 years (range, 16-88) and 93% of patients were women. More than 41% (n=168) had undergone previous pelvic surgery, of which 39% were hysterectomies. In 27 patients (6.7%), LVR was performed for recurrent rectal prolapse. Data concerning operative difficulties and conversion, postoperative morbidity, and recurrence were gathered from a prospective database. The mean follow-up was 25 months (range, 6-143). An extensive, institutional questionnaire that assessed symptoms of anorectal and sexual dysfunction was used.

Nearly half of the patients suffered from internal rectal prolapse (45.9%; n=186). Other indications were total rectal prolapse (43%; n=174) and isolated rectocele and/or enterocele (11.1%; n=45). In 95 patients (23.5%), laparoscopic dissection of the rectovaginal septum was completed with a small perineotomy to treat a complex supra-anal rectocele.

Figure 1. Dissection starts at the sacral promontory with preservation of the right hypogastric nerve. The caudal extension of the peritoneal incision follows the dotted line.
Figure 2. The deepest part of the fold of Douglas is retracted and incised. The rectovaginal septum is opened without any lateral dissection.

Surgical Technique

Limited bowel preparation and a single dose of a broad-spectrum antibiotic, as well as thrombophylaxis, were given preoperatively to the patient. The “bean bag” allows steep Trendelenburg in the operating room: A modified lithotomy position with both arms tucked was used. A bladder catheter was inserted, the vagina disinfected, and the camera port was introduced at the umbilicus. Three additional ports were inserted into the right flank (5 mm), the left iliac fossa (5 mm) and the right lower quadrant (12 mm). A 30-degree optic was used and temporary hysteropexy enhanced the pelvic view. The entire small bowel was retracted out of the pelvis. Next, the mesosigmoid was retracted to the left and a peritoneal incision was made over the sacral promontory and caudally extended along the rectum and over the deepest part of the pouch of Douglas (special care was taken to preserve the right hypogastric nerve). The sacral promontory needed to be sufficiently dissected to allow safe mesh fixation (too medial a dissection must be avoided to safeguard the left iliac vein) (figure 1). The rectovaginal septum was opened after a firm retraction of the deepest part of the fold of Douglas. The Denonvilliers fascia was incised, and the anterior aspect of the rectum was dissected, leaving all fibrous tissue against the posterior vaginal wall (figure 2). A Marlex mesh (approximately 3×17 cm) was sutured to the ventral aspect of the distal rectum using nonabsorbable sutures to inhibit further rectal intussusception. Extracorporeal suturing seemed the most appropriate in the deepest part. Care was taken to ensure that the mesh lay flat on the rectum to avoid any mechanical erosion due to mesh kinking. The mesh was then fixed to the sacral promontory using an endoscopic "tacker" device, and secured with one stitch of ethibond 2.0. No traction was exerted on the rectum, which remained in the sacrococcygeal hollow (figure 3). The posterior vaginal apex (vaginal vault) was then elevated and sutured to the same strip of mesh (figure 4). Two lateral sutures incorporated the remainder of the uterosacral ligament. More suturing may be needed depending on the degree of middle compartment prolapse. Ideally, the sutures should not perforate the vaginal wall. The mesh could be left broader at that site to allow adequate vault suspension. The lateral borders were closed over the mesh using the V-Loc 90 absorbable wound-closure device elevating the neo-Douglas over the colpopexy. This maneuver was important to avoid any later small bowel entrapment and/or erosion. It can be difficult to complete the rectovaginal septum dissection to the pelvic floor level. In treating a complex rectocele, the surgeon may decide to complete the laparoscopic dissection with a small perineotomy. The incision is made immediately dorsal to the vaginal orifice to open the perineal body and dissected to join the laparoscopic dissection plane, allowing mesh fixation in the deepest part of the rectovaginal septum and restoring the perineal body. The perineotomy, however, could be avoided in most patients with a total rectal prolapse. Postoperatively, no drain is needed in most cases, and the patient can be discharged from day 1 onward.

Figure 3. A strip of polypropylene is sutured to the anterior aspect of the rectum and fixed without traction on the sacral promontory.
Figure 4. The posterior vaginal wall is elevated and sutured to the same mesh.

Outcome After Laparoscopic Ventral Rectocolpopexy

Conversion to laparotomy was required in eight patients (2%). The mean hospital length of stay (LOS) was 5.1 days and the last 50 patients had an LOS of 3.2 days. Perioperative mortality did not occur. Minor morbidity was noted in 74 patients (18%); urinary tract infection in 23 (5.9%); superficial wound dehiscence in 18 (4.6%); prolonged ileus in12 (3.1%); and postoperative hematoma or bleeding in nine (2.3%). Six patients (1.5%) underwent a re-intervention under general anesthesia within 30 days of surgery. Ten patients (2.5%) developed dyspareunia. Prolonged neuralgia (six weeks) at the right lower quadrant port was documented in six patients (1.5%). five patients (1.3%) were seen with mesh erosion. All of the patients underwent a combined approach with perineotomy for a grade III supra-anal rectocele. five patients (1.3%) had a trocar-site hernia. No major septic complications were observed.

Clinical recurrence was noted in 4.6% of 174 patients after LVR for total rectal prolapse. Only four of these eight patients underwent further perineal surgery. Recurrence for internal rectal prolapse was lower (0.5%), but the need for perineal surgery during follow-up was higher (4.3%).

Failure of the mesh fixation onto the sacral promontory was noted in four patients during relaparoscopy. In one patient, dehiscence of the rectal fixation was seen, and in another, incomplete reduction of the prolapse at the time of mesh fixation evidently resulted in a persistent prolapse.

Significant improvement in symptoms was reported in 85.6% of patients at final follow-up. Obstructed defecation resolved completely in 71.1%, whereas new-onset constipation was documented in only 10 patients (2.3%). Fecal incontinence improved in 84.5% of patients.

Obstructed defecation, which was present in 120 patients with internal rectal prolapse before LVR, had resolved in 59.2%. Constipation was induced in 3%. Fecal incontinence improved in 88.9% of patients with internal rectal prolapse, and at final follow-up, 70.4% of patients reported improvement of functional outcome.

In summary, Professor D’Hoore noted that surgical treatment of rectal prolapse syndromes remains controversial in colorectal surgery. A large number of operations are described in the literature. LVR was developed in an attempt to fulfill the three main objectives of prolapse surgery: restoration of the anatomy in a reliable, safe and reproducible way; improvement of anorectal function; and avoidance of functional sequelae (constipation and incontinence). After LVR for total rectal prolapse, a significant improvement occurred in 85% of patients at final follow-up. Potential functional problems should be investigated before LVR in patients with internal rectal prolapse. Moreover, mechanical and functional obstruction may coexist. LVR, with or without perineotomy, was found to be safe with relatively low morbidity and functional outcomes supported its efficacy. The indication for LVR in patients with internal rectal prolapse should be optimized.

Expert Reviews Use of MRI in Rectal Cancer

A review highlighting how magnetic resonance imaging (MRI) could be used in the assessment of rectal cancer to improve patient outcomes was presented by Gina Brown, MD, consultant radiologist and reader in gastrointestinal cancer imaging at The Royal Marsden hospital in England.

The core aspects of rectal cancer surgery have changed tremendously over the past few decades. In the Dutch TME and CR07 trials that assessed rectal cancer treatments, one of the problems was the lack of homogeneity of the excised specimens and the challenge this presented in standardizing and auditing the specimen (on which statistics and prognostic criteria are based). These trials found that lymph nodes staying behind are strong predictors of recurrence. Today, entire nodes are removed en bloc with total mesorectal excision (TME). The question is can we still identify predictors for local recurrence and what are the predictors for distant failure?

These are important points to consider when optimizing patient selection and outcomes. Dr. Brown explained that best practices for primary surgery should soon be revealed. If a surgeon could efficiently predict that local recurrence would be low, then maybe he or she would not irradiate every patient they see. Or perhaps those at very high risk for distant metastasis or those who probably would not survive even after extensive surgery could be identified.

Dr. Brown said that the most important predictor of local recurrence is at the hand of the surgeon and depends on the quality of the TME specimen obtained, which is extremely important for histologic and pathologic assessment of the tumor. A good TME specimen with no positive lymph nodes shows that the patient is at very low risk for local recurrence. It is well known that it is not the T stage of the tumor, but more importantly the marginal involvement of the specimen, which has a fourfold risk for local recurrence, if positive. The latter also is true for MRI: Wherever there is margin involvement, there is increased risk for recurrence and a 3.5 hazard ratio for local recurrence. This is so, regardless of the state of the lymph node involvement. Therefore, the two most important factors with regard to the tumor are its height and margin status.

Figure. Survival prediction of positive margin (unfavorable) vs. no MRI margin involvement (favorable).

One end point from MERCURY (Magnetic Resonance Imaging in Rectal Cancer European Equivalence Study) was the assessment of the diagnostic accuracy of MRI in predicting circumferential margin involvement of the tumor and its equivalence to extramural depth of tumor invasion with histopathology. The figure shows the prediction for those who reported a positive margin and thus an unfavorable prognosis compared with those who had no MRI margin involvement, thus having a favorable prognosis. The latter two were correlated with survival over time. Using MRI allows the surgeon to predict local recurrence according to margin involvement.

Another factor that was predictive of local recurrence was the distance of the tumor from the mesorectal fascia. If the tumor was 1 mm or more away from the fascia, then the chances of local recurrence were found to be low. Using the 1-mm cutoff achieved the highest agreement, corresponding best with histopathologic reports. This shows that if a patient’s tumor can be seen on MRI with a distance larger than 1 mm, then local irradiation might not be in the patient’s best interest as chances for local recurrence are low.

Another implication is that neoadjuvant treatment could regress the tumor to negative margins, again making local recurrence less likely. The same is true for nodes involving the fascia where preoperative radiation can take away the node from the fascia, and thus provide better prognosis. This downstaging of the tumor should take into account the low prevalence for nodal involvement of the fascia.

Dr. Brown explained that tumors that are less than 5 cm from the anal verge are at a higher risk for margin involvement because of the difficulty in performing the distal TME dissection that goes through the intersphincteric plane, risking tumor perforation. This mostly occurs just above the puborectalis sling. Therefore, another set of data that could help predict prognosis is the extent of tumor involvement of the sphincter complex. The dangerous and high-risk tumors for marginal involvement are classified according to the depth of spread at the level of the sphincter complex. Those tumors extending into the muscularis are classified as stages III and IV, so it is possible to identify patients with low rectal cancer who are at high risk for local recurrence. To tackle the latter problem, a method was developed to allow surgeons to identify which plane to operate on, using MRI for guidance: the standard TME plane, the TME plane with sphincter preservation or the extralavator plane. The latter system would help the surgeon lower the risk for and prevent local recurrence.

Other preoperative factors that were seen to affect disease-free survival rates using univariate analysis include circumferential resection margin stage, advanced T stage and MRI extramural vascular invasion status, disease in the sidewall compartment, and venous mural involvement. More trials are currently under way in which chemotherapy is given upfront and positron emission tomography imaging is used to identity high-risk patients, the results of which are should be published soon.

Dr. Brown concluded that using MRI to stratify patients with low rectal cancer is crucial for better prognosis of patients and to prevent unnecessary procedures with the understanding that some patients will only need primary surgery with or without combined chemoradiation therapy, whereas others who have tumors that need additional therapy and or more radical surgery. MRI is expected to be an integral part for future preoperative assessment of tumors in patients with low rectal cancers.

Sacral Neuromodulation and Tibial Nerve Stimulation For Fecal Incontinence

David C.C. Bartolo, MD, professor of surgery, The University of Western Australia in Perth, explained that fecal incontinence (fi) is a highly distressing condition that affects more people than current data would suggest. For the most part, obstetric injuries, including the use of forceps during childbirth, are responsible. Injury is associated with two pathologies, which may coexist to varying degrees—tearing of the anal sphincter muscle and anal canal or damaging the nerves supplying that region.

Repairing a denervated muscle serves little purpose. In the past, the success rates of sphincter repair were greatly exaggerated: Early outcomes were shown to deteriorate with time and patients may have reported better outcomes than actually existed. Also, if the nerve supply to the sphincter is damaged, there is a high probability that there also will be interference with the normal functioning of the rectum and possibly the colon. This means that the coordination between anal muscles and the rectum is altered, which compromises effective continence and defecation.

In 1995, Matzel et al published pioneering work on sacral nerve stimulation (SNS). Over the past decade, SNS has become an established treatment for fi in Europe and recently also has been approved in the United States. Originally, it was believed that an intact external anal sphincter was a prerequisite for SNS since the proposed mechanism of action was an augmented function of a generalized weakened striated anal sphincter. But several observational studies have shown that SNS is successful in some patients with sphincter defects.

SNS has had successful results that appear to be multimodal. Local effects on rectal awareness and urgency (and in some studies on sphincter function), and central effects are seen on positron emission tomography (PET) scans and functional magnetic resonance imaging (MRI) studies of the brain. Some studies reported improvement in sphincter pressures whereas others did not, and some described altered rectal awareness, which possibly allowed an earlier perception of the need to defecate. A 2008 study by Michelsen showed that SNS in fi not only induced changes in the left colon (explained by presacral nerve stimulation of the long colonic nerves), but also resulted in changes to the right colon that must be vagally induced. Laurberg et al used PET scans to record cerebral blood flow before and after 30 minutes of continuous stimulation. After two weeks of continued stimulation, they repeated the scan before, and 30 minutes after, the arrest of the stimulation. The initial stimulation activated a region of the contralateral frontal cortex and after two weeks of stimulation, during the repeated measurement, activity was found in parts of the ipsilateral caudate nucleus, a region of the brain thought to be involved in learning and reward processing. They concluded that SNS induced changes using the afferent projections of the vagus nerve.

LeRoi et al published landmark studies of the crossover by which they turned off the neuromodulation and found persisting benefits in terms of elevated sphincter pressure compared with preimplantation. Thus, the benefits could represent some form of training or long-term modulation. However, loss of benefit over time has been increasingly reported. Although unclear, long-term failure may result from changes around the stimulation leads.

A major advantage of SNS is the peripheral nerve evaluation (PNE), which uses a temporary wire attached to a stimulator. Because success rates vary between 50% and 80%, progression to definitive implant is advised only after positive PNE. The most commonly used criterion for a positive PNE test is a 50% reduction in the number of incontinence episodes. Unfortunately, not all positive temporary outcomes translate into permanent ones, and the reason for this is unclear. Thus, some advocate multiple lead testing, which may result in better permanent site selection.

Tan et al conducted a meta-analysis of 34 studies on SNS for fecal incontinence published between 1995 and 2008. The researchers evaluated functional, physiologic and quality-of-life (QoL) outcomes. Overall, 944 patients undergoing PNE and 665 patients undergoing permanent SNS were included. Weekly incontinence episodes and incontinence scores were significantly reduced with SNS and the ability to defer defecation was increased. QoL indices improved following SNS, and mean anal pressures increased significantly (P<0.001). Patients under the age of 56 years showed smaller functional—but greater physiologic and QoL—improvements. Results were similar between sphincter-intact and impaired subgroups. The complication rate was 15% for permanent SNS, with 3% resulting in permanent explantation.

Another important development is the spread of indications. Initially, SNS was used in patients with an intact anal sphincter, and later in patients with sphincter defects, and comparable outcomes have been achieved with SNS and sphincter repair in appropriately selected patients. Patients who are functioning poorly after rectal reconstructive surgery, diabetes, various neurologic diseases, prolapse surgery and other etiologies, are now being studied and also could be potentially offered SNS implantation. SNS for irritable bowel syndrome shows improvement in symptoms and QoL, although the mechanism is still not understood.

Kamm et al reported on patients with constipation undergoing PNE, in which 73% proceeded to permanent SNS implantation. Of those, 63% achieved a 50% improvement with stool frequency, from 2.3 to 6.6 evacuations per week, which is in the normal reference range. On an intention-to-treat basis, just under two-thirds achieved an improvement of 50%, signifying the importance of further study of the use of SNS in severe constipation. Dinning, Lubowski and colleagues were able to study propagating waves in the colon using a long multiport manometer. They found that stimulation of S3 induced pancolonic propagating sequences. Conversely, S2 stimulation produced an increase in retrograde wave sequences.

When comparing SNS with an artificial bowel sphincter (ABS), both have high rates of explantation. Incontinencescores are better with ABS, but this method was associated with more incontinence events. Typically, if SNS fails, one may then consider ABS or muscle transfer.

After initial successful PNE, if there is loss of function with the definitive implant, a number of actions should be taken. The device and lead integrity should be checked. If there is loss of sensation, lead replacement may be required, especially if there is high impedance on testing.

Percutaneous and/or transcutaneous nerve stimulation (PCTNS) offers a second form of neuromodulation in which the tibial nerve is stimulated. The latter contains sensory fibers that may spread the stimulation through the sacral nerve plexi. PCTNS potentially could be applied at night while the patient is sleeping. Another advantage is the low cost. In a recent report, 144 patients with either tibial stimulation or a sham procedure were compared and no significant benefits for fi or urgency were observed on manometry. The role of PCTNS currently is questionable.

Dr. Bartolo said that PNE should be considered for patients with fi, regardless of the cause of the incontinence. The surgeon should know how to evaluate the motor contraction intraoperatively, offering the best site for implantation, perhaps by using multiple leads. The 50% benefit currently noted is still too low, and better results on PNE should be available before performing permanent implantation.