Application technique: placement of a prostate–rectum spacer in men undergoing prostate radiation therapy


  • B.H. and M.H. contributed equally to the work.

Gencay Hatiboglu, Department of Urology, University of Heidelberg, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany. e-mail:


Study Type – Therapy (case series)

Level of Evidence 4

What's known on the subject? and What does the study add?

Different spacing agents have been tested to reduce incidential radiation exposure of the rectum during radiotherapy to the prostate. These agents all had certain drawbacks; either the created space was too small or the agents used did not stay in place during radiotherapy treatment.

The study describes the transperineal injection technique of a spacing agent in detail. Furthermore it shows the safety and efficacy of the spacing hydrogel used and shows that it overcomes some of the drawbacks of the previously examined spacing agents.


  • • To describe the technique used to apply a hydrogel spacer between the prostate and rectum so as to decrease the radiation dose to the rectum in patients with prostate cancer who are undergoing radiotherapy.


  • • A prospective study evaluating the safety and efficacy of prostate–rectum spacer injection was conducted in 29 male patients with prostate cancer scheduled for radiotherapy.
  • • Spacing hydrogel was injected into the perirectal space using a transperineal approach under real-time transrectal ultrasonography guidance.
  • • With the needle tip positioned beyond the rectourethralis muscle, saline injection opened the space between Denonvilliers' fascia and the anterior rectal wall, allowing needle advancement to the mid-prostate without rectal wall injury. Injection of hydrogel precursors further opened this space, which was then maintained as a result of hydrogel polymerization.
  • • Procedure duration and adverse events were monitored. Computed tomography and/or magnetic resonance imaging simulation scans were performed before and after injection. The hydrogel-created space was measured and the reduction in percent volume of the rectum receiving at least 70 Gy (rectal V70) was determined.


  • • Hydrogel injection resulted in mean (sd) additional prostate–rectum space relative to baseline of 9.87 (5.92) mm.
  • • The mean (sd) procedure time, as measured by needle insertion and removal, was 6.3 (3.2) min.
  • • The relative reduction in rectal V70 was 60.6%.
  • • There were no unanticipated adverse events associated with the hydrogel procedure or the hydrogel.


  • • Hydrogel spacer injection using hydrodissection is a fast and effective procedure to separate the rectal wall from the prostate in order to avoid rectal toxicity.
  • • Hydrogel spacer injection resulted in the addition of ∼1 cm of space
  • • Computed incidental radiation exposure, the rectal V70, was substantially reduced.



polyethylene glycol

rectal V70

volume of the rectum receiving at least 70 Gy


intensity-modulated radiation therapy.


Worldwide, prostate cancer is one of the most common cancers in men. For 2011, it was estimated that there were 903 500 new cases of prostate cancer worldwide [1].

For the definitive treatment of localized prostate cancer, radical prostatectomy or radiotherapy (RT) are often recommended [2,3]. Although both treatment options are considered almost equally effective, the side effects are varying. For radical prostatectomy these include erectile dysfunction and incontinence and for RT one of the main side effects is rectal toxicity from radiation exposure, owing to proximity to the prostate [4,5]. In addition, radiation margins of up to 10 mm around the prostate are required, owing to prostatic motion during treatment, to ensure that the entire gland receives the full dose of radiation [6,7]. Because of the anatomical proximity to the sensitive rectal mucosa, the rectum is commonly defined as the dose-limiting structure for prostate radiotherapy [8]. Storey et al. [9] reported that there were significantly more late rectal complications, such as bleeding, when 25% of the rectum received ≥70 Gy. To overcome this issue, several investigators have evaluated different materials, injected transperineally, to create space between the prostate and the rectum, including a blood patch [10], hyaluronic acid [11,12] and collagen injections [13]. Although these materials have been shown to reduce the amount of radiation to the rectum, each has certain drawbacks such as short persistence, degradation during RT and a non-uniform distribution. To address some of these drawbacks, a polyethylene glycol (PEG)-based hydrogel system (SpaceOAR System, Augmenix, Waltham, MA, USA) has been developed to temporarily create space between the rectal wall and the prostate during RT, thereby reducing the amount of radiation to the anterior rectum. The hydrogel spacer is a synthetic PEG-based material that is absorbable, water-soluble, non-toxic and non-immunogenic. It remains in place for 3 months, and then undergoes hydrolysis, liquefies, and is absorbed and cleared from the body in ∼6 months via renal filtration (Fig. 1).

Figure 1.

Transversal MRI view of prostate–rectum separation and spacer hydrogel distribution. After 6 months, the spacing hydrogel was completely absorbed.

In situ polymerizing hydrogels were evaluated as prostate–rectum spacers in cadavers, showing satisfactory hydrogel distribution in all cases [14]. The mean (sd) prostate–rectum separation was 12.8 (1.2) mm. Overall, a prostate–rectum separation of 10 mm was found to be sufficient to reduce the mean volume of the rectum receiving at least 70 Gy (rectal V70) by 83.1% (P < 0.05), with no further reduction in rectal V70 at 15 mm of separation. In addition, hydrogel placement allowed increased planning target volume margins without exceeding the rectal dose tolerance. That preclinical study provided a basis for further evaluation of the device in human clinical trials.

In the present study, we describe the spacer hydrogel device, the transperineal/hydrodissection technique for its application and the results from a 29-patient pilot evaluation conducted using the device.


Twenty-nine men, indicated for a course of external beam RT as a result of pathologically confirmed prostate cancer, were enrolled in the present study between 19 July 2010 and 21 January 2011. All patients gave written informed consent, according to the Helsinki Declaration. The study was approved by the local ethics review boards. Patient characteristics, serum PSA levels and Gleason score grading were determined.


The present prospective, non-randomized, multicentre, single-arm, open-label study was carried out at four institutions of the European Union. Inclusion criteria were pathologically confirmed low- and intermediate-risk prostate cancer and a prostate volume <80 mL. Patients with metastatic disease, history of prostatectomy or other prostate surgery (e.g. TURP), previous radiotherapy to the prostate or pelvis, history of rectal or gastro-intestinal surgery or active inflammatory bowel disease were excluded.

After registering for the study, a baseline CT or MRI planning scan was done. All patients then received a transperineal injection of the hydrogel and the simulation scans were repeated within 3–5 days. Within 19 days of the hydrogel injection, the patients began intensity-modulated radiation therapy (IMRT; 78 Gy delivered over an 8-week period, 200 cGy per fraction, using five fractions per week).

The main objectives of the present study were to evaluate the safety and performance of the hydrogel system (SpaceOAR) when used to maintain space between the rectum and prostate. Functional success was defined as creation of at least 7.5 mm space between the posterior prostatic capsule (mid gland) and anterior rectal wall and was assessed by CT or MRI scans. Clinical success was defined as a minimum 25% reduction in rectal V70. Safety events that could be attributed to the device and procedure were continuously monitored.


The SpaceOAR System consists of components for the preparation of the absorbable hydrogel spacer with its delivery mechanism provided in a single-use kit (Fig. 2). The spacer hydrogel is formed by simultaneously injecting two solutions, the precursor and the accelerator, into the perirectal space. The solutions mix during injection, initiating a cross-linking reaction that results in the formation of a soft PEG-based gel within 10 s, without a measurable temperature rise. The hydrogel remains in the body for 12 weeks, after which hydrolysis causes the implant to liquefy, resulting in complete absorption (Fig. 1).

Figure 2.

Hydrogel system delivery assembly.


The hydrogel is injected transperineally using hydrodissection to facilitate placement. The potential perirectal space for hydrogel placement is between Denonvilliers' fascia and the rectal wall (Fig. 3). Denonvilliers' fascia consists of a single fibromuscular structure covering the posterior aspect of the prostate. This important anatomical boundary separates the prostate from the anterior rectal wall. It is important that the hydrogel is injected on the posterior side of Denonvilliers' fascia and anterior to the anterior rectal wall to minimize the risk of pushing cancer cells away from the high dose radiation field. Villers et al. [15] found that although prostate cancer invaded Denonvilliers' fascia in 19% of cases, there were no cases where the tumour invasion had progressed completely through the full thickness of the structure in their series of 243 prostatectomy specimens. In addition, care must be taken to avoid inadvertent injection into the rectal wall which can potentially lead to ischaemia or increased rectal wall stresses.

Figure 3.

Prostate anatomy and sagittal view of potential space.

The spacing hydrogel is injected using a transperineal approach under TRUS guidance. A transrectal approach cannot be used because of potential bacterial contamination.

Hydrodissection is a technique used to separate tissue planes through the use of fluid. It has been used for many years in various procedures including cataract surgeries and, more recently, in carpal tunnel procedures [16,17]. In cataract surgery, fluid is used to separate the lens nucleus from the cortex and in carpal tunnel procedures fluid dissection has been proven to be helpful in alleviating nerve compression. In these delicate procedures a small-gauge needle and small amounts of fluid are used to gently separate the planes. In the technique described in the present paper, the insertion of the hydrogel spacer is facilitated with the use of hydrodissection. Small volumes of fluid are carefully injected through an 18-G needle between Denonvilliers' fascia and the anterior rectal wall to open the potential space before injecting the hydrogel. By this means, enough space for the spacer hydrogel is created and steady dispersion of the spacing agent is ensured.


The insertion procedure itself is a short procedure that can be performed under local, spinal or light general anaesthesia in an outpatient setting. In the present study, use of prophylactic antibiotics was at the discretion of the treating physician at the time of the insertion procedure. Transperineal fiducial placement (if used) was performed before the hydrogel injection.

The patient was placed in the lithotomy position. A linear side-fire TRUS probe, covered by a stand-off balloon was used for best ultrasonography imaging and proper needle and anatomical visibility. The device was mounted to an adjustable stepper unit, freeing both the physician's hands.

Under real-time TRUS guidance in the sagittal plane, the injection needle (18G × 15 cm), with an attached syringe containing saline or lidocaine 0.5–1.0% diluted in 15 mL saline, was inserted ∼1 cm above the TRUS probe through the perineum and carefully advanced to the pelvic floor muscles at an angle parallel to the TRUS probe. After penetration of the rectourethralis muscle, the tip was positioned inferior to the prostatic apex, just between Denonvilliers' fascia and the anterior rectal wall (Fig. 4A). Hydrodissection was started using small volumes of fluid to slowly open up the potential space between Denonvilliers' fascia and the anterior rectal wall. Once the correct space started to open, the needle was advanced into the space, and saline injection/needle advancement continued until the needle tip reached mid-gland. By this means, space for the spacer hydrogel was created (Fig. 4B). After confirming the correct position of the needle with both sagittal and axial ultrasound views, and aspirating to ensure the needle tip was not intravascular, the saline syringe was removed. The syringe assembly was then attached, maintaining needle position, and 10 mL of hydrogel was administered in one continuous motion in 8–10 s without moving the needle during injection (Fig. 4C). In the case of insufficient hydrodissection or unintentional rectal wall perforation, spacer hydrogel injection was not performed and the procedure was aborted.

Figure 4.

A, Needle alignment and placement with ultrasonography sagittal view. B, Hydrodissection with ultrasonography sagittal view. C, Hydrogel injection with ultrasonography sagittal view.

The distance from the rectal wall to the prostate/Denonvilliers' fascia was measured before and after the injection procedure. The duration of the overall procedure and injection, difficulty of procedure, and amount of hydrogel used were determined. Adverse events during and after the spacer injection procedure were assessed.


Radiotherapy treatment planning was performed at each centre on simulation scans acquired before and after hydrogel injection and the rectal V70 was calculated. CT/MRI images taken at baseline, after hydrogel application (∼1 week), end of IMRT (∼3 months) and 6 months after insertion were reviewed by an independent radiation oncologist who measured the mid-gland space between the prostate and anterior rectal wall for every patient. The space created resulting from hydrogel application was calculated by subtracting the baseline space from the space measured at each subsequent time point.


The mean patient age was ∼67 years. The baseline characteristics were reflective of a low-to-intermediate-risk prostate cancer population. Patients were classified in the T1c-T2c stage with an approximately even distribution of Gleason scores 6 (48%) and 7 (52%). Most patients (59%) had a PSA level between 4 and 10 ng/mL. The median prostate size was 58 mL. Hydrogel placement was performed successfully in all patients. None of the procedures needed to be aborted because of insufficient hydrodissection or other complications.

Prostate–rectum space was determined by an independent data reviewer from the CT or MR images taken before and after hydrogel placement. Before hydrogel insertion, the mean (sd) mid-gland space between the prostate and rectum was 4.8 (2.4) mm. After insertion it was 14.7 (5.2) mm, resulting in a mean (sd) space creation of 9.87 (5.92) mm 1 week after insertion (Fig. 5) and 10.47 (4.33) mm and 1.68 (2.94) mm 3 and 6 months after, respectively. The amount of space created was not affected by prostate size (Fig. 6).

Figure 5.

Transveral ultrasonography view before and after hydrogel insertion.

Figure 6.

Mid-gland space after hydrogel insertion (mm) vs prostate volume (mL).

The calculated dose to the anterior rectum was obtained from treatment plans created from the pre- and post-hydrogel insertion simulation scans. The mean (sd; range) rectal V70 before vs after hydrogel placement was substantially reduced from 14.6 (7.11; 3.1–32.7)% to 5.8 (4.6; 0.1–19.5)% resulting in a mean (sd) relative reduction of 60.6 (22.6)%.

Functional (7.5 mm space after hydrogel placement) and clinical success (≥25% reduction in rectal V70) was achieved in 96.6% (28/29) and 96.3% (26/27) of patients, respectively. Two patients were excluded from clinical success analysis as their planning simulations could not be uploaded for the review process because of technical problems. The one patient that did not meet the criteria for functional success had a post mid-gland distance measurement of 6.3 mm, but met the criteria for clinical success with a reduction in rectal V70 of 49%. The one patient who did not meet the criteria for clinical success had a rectal V70 of 7.3% before hydrogel placement and a rectal V70 of 7.6% after placement, despite an increase in mid-gland distance (2.7 mm before and 15.2 mm after hydrogel placement). A review of the planning images showed a larger rectal volume before (289 mL) than after placement (189 mL), which could explain why no change in rectal V70 was observed.

The injection device was deemed to be very easy or easy to use in 98% of the cases. The mean (sd) overall procedure time was 16 (7.8) min from TRUS insertion to removal. The mean (sd) needle-in to needle-out procedure time was 6.31 (3.2) min. Most patients tolerated the injection of the hydrogel well. No patients experienced a device- or procedure-related adverse event.


Results from the present clinical study show that, with proper injection technique using hydrodissection, it was feasible to administer the hydrogel material within the potential space between Denonvilliers' fascia and the rectal wall without any device- or procedure-related adverse events. The injection device was considered easy to use. The high overall functional success rate of 96.6% (28/29 patients) showed that the hydrogel was effective at displacing the rectal wall away from the prostate. In addition, this space of ∼1 cm was maintained throughout the course of IMRT. The hydrogel then gradually liquefied through hydrolysis and was absorbed in ∼6 months. Furthermore, clinical success was achieved in 96.3% (26/27) of the patients. Administration of the hydrogel resulted in a substantial reduction in computed incidental radiation exposure to the anterior rectum (a mean rectal V70 reduction of 60.6%).

Several other studies have evaluated other materials (e.g. a blood patch [10], hyaluronic acid [11,12] and collagen injections [13]), injected transperineally for a rectum–prostate separation and found that these materials had limitations. The blood patch technique involves injecting a small sample of the patient's blood between the rectum and the prostate [10]. In a small pilot study of three patients the blood patch provided only a modest separation between the rectum and prostate. The patch was reported to stay in place for only 1 week. Collagen availability can be somewhat volatile and the material is commonly supplied in a lyophilized form, which must be reconstituted [13]. In addition, collagen tends to clump during preparation and obtaining the proper consistency can be difficult. Hyaluronic acid can undergo degradation upon radiation exposure [18]. The material is also somewhat viscous, which can affect its ability to distribute evenly upon injection.

The PEG-based hydrogel used in the present study, has several advantages over the aforementioned materials. It is injected as a liquid into the perirectal fat, where it expands the space and then polymerizes (solidifies) into a soft hydrogel within 10 s. The hydrodissection technique is used to open up the perirectal space and provides a ‘landing zone’ for the spacer, ensuring excellent distribution and durable separation of the prostate from the anterior rectal wall. The hydrogel remains in place for 3 months, during radiotherapy, and then gradually liquefies via hydrolysis and is absorbed in ∼6 months, as shown in the present study. Furthermore, the safety of PEG-based hydrogels has been shown previously; they have been used safely for a variety of neurosurgical applications, e.g. a watertight dural closure [19], and there were no complications associated with the use of the hydrogel sealant when applied to the nerve root and surrounding area after lumbar microdiscectomy for adhesion prevention [20]. A number of other applications using the PEG-based hydrogel have also been shown to have a favourable safety profile. No adverse events or toxicity were attributed to the material when it was used as an adhesion barrier in women undergoing laparoscopic or open uterine myomectomy [21] or patients undergoing loop ileostomy closure [22]. Lastly, the PEG-based hydrogel was shown to be safe when used as a temporary ocular surface bandage for patients undergoing clear corneal cataract surgery [23]. Taken together, these results demonstrate the versatility and safety of PEG-based hydrogel material in various applications.

The transperineal TRUS-guided injection approach used in the present study has been used previously for several other urological procedures including: placement of gold fiducial markers [24,25], prostatic biopsies [26–28], transperineal injection of botulin toxin [29,30], and placement of brachytherapy seeds [31], so the techniques used for the application of the prostate–rectum spacer hydrogel were not expected to raise new safety issues. Use of the side-fire TRUS probe and the stand-off balloon improved visibility, a stepper unit stabilized the probe and allowed both hands to be free to control the hydrogel applicator. Overall the safety results show that the injected hydrogel appears to be well tolerated. The injection procedure was also found to be fast; overall procedure time from TRUS insertion to removal was 16 min, while the needle insertion to needle removal only took 6 min.

In conclusion, the SpaceOARTM System provides an effective means to temporarily provide greater separation between the anterior rectal wall and the prostate, thus reducing the radiation dose to the rectum. The hydrogel appears to be safe and well tolerated when administered appropriately. As urologists are already familiar with the transperineal approach, the presented technique is safe and easy to learn. Moreover the urologist accompanies patients with prostate cancer before and after any kind of treatment and, by this means, the urologist can contribute to a safer radiotherapy procedure by injecting the spacing hydrogel transperineally, a procedure that can even be done easily in an outpatient setting. The urologist can ensure that there is less rectal toxicity and therefore fewer adverse events after radiotherapy for his/her patients.


The independent data reviewer: Danny Song, M.D. Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University, Baltimore, MD, USA.


None declared.