Complications and Influential Perioperative Factors Associated with SpaceOAR Hydrogel Placement

Objective To examine one academic institution's experiences with SpaceOAR placement, its associated complications, and periprocedural characteristics that affect outcomes for the purpose of quality improvement. Materials and Methods We conducted a retrospective review of 233 patients who received SpaceOAR from four surgeons and one radiation oncologist between 2018 and 2021. Variables such as demographics, oncologic parameters, radiation plan, and radiographic assessment of hydrogel placement were recorded. The Charlson Comorbidity Index (CCI) was used to assess comorbidity risk. Mann–Whitney and Fisher's exact tests were performed to compare patients with and without complications. Results Of the 233 patients who received SpaceOAR, 24 (10.3%) experienced toxicity. All complications were Clavien I or II, such as pelvic pain postplacement, pelvic fullness, bleeding, and lower urinary tract symptoms. 16 patients (6.9%) had some portion of the hydrogel injected into the rectal wall, but it was never clinically significant. The average CCI was 3.2 ± 0.95 for patients who experienced complications; the average CCI was 3.6 ± 1.6 (p=0.48) in the group without complications. Of the physicians with higher procedure volumes, Physician #1 had the highest rate of patient-reported complications at 11 out of 68 (16.2%) and Physician #2 had the lowest rate of complications at 4 out of 96 placements (4.2%). Multivariate analysis found that patients who had received hormone therapy previously had less odds of reporting complications after SpaceOAR placement. Conclusions The listed attending on the procedure had a significant correlation to complications with SpaceOAR placement on univariate analysis, and hormone therapy had some benefits to the tolerance for the procedure on multivariate analysis. Overall, the hydrogel placement was well tolerated with low incidence of mild and transient procedure-related toxicity.


Introduction
Radiation therapy (RT) is a mainstay for localized or locally advanced prostate cancer.Rectal injury, exhibiting as acute and chronic proctitis, is a common concern in radiotherapeutic treatment due to the proximity of the anatomical structures, and the concern for side efects limits dose escalation.
Te SpaceOAR System (Boston Scientifc, Marlborough, MA USA) is an FDA-cleared implant with ultrasound-guided perineum needle insertion, and it which adds about 1.3 centimeters between the prostate and rectum decreases radiation dose to the anterior rectal wall and decreases likelihood of late rectal side efects including rectal bleeding and proctitis [1].Te substance is 90% water mixed with polyethylene glycol (PEG), which forms into a gel-like substance in vitro, and it is absorbed and excreted in the urine naturally in about 6 months.Potential complications that could arise from the placement of the hydrogel include transient symptoms, such as pain associated with injection, pain, or discomfort from the hydrogel, site infammation, urinary urgency, constipation, rectal urgency or spasms, fainting, bleeding, infection, urinary retention, or rectal injury.More signifcant complications are rare but include infection, urinary retention, rectal injury, or ulcers.Severe complications are very unlikely but include the fstula, perforation, necrosis, allergic reaction (local reaction or more severe reaction, such as anaphylaxis), or embolism [1].Studies have reported a range of complication rates with hydrogel insertion.Tose fnding complications as low as 0% argue report SpaceOAR as a low-risk, potential-beneft prophylactic treatment, while those fnding more frequent complications as high as 10% argue that prophylaxis for RTrelated rectal toxicity is no longer necessary [2][3][4].One study analyzed the Manufacturer and User Facility Device Experience (MAUDE) database for all SpaceOAR-related adverse events.It argued that although complications with hydrogel injection are reportedly rare, the incidence of medical device reports increased each year, and the severity and debilitations, such as infections requiring surgical intervention, perirectal fstulae, signifcant bleeding, anaphylaxis requiring intensive care, and two deaths, should be considered [5].However, Babayan et al. responded that the yearly increase in the number of medical device reports was proportional to device usage [6].
In addition, perioperative variables and the Charlson Comorbidity Index (CCI) specifcally have not yet been assessed as prognostic factors.CCI is a validated method of assigning comorbidities a score from 1 to 6 based on their association with mortality and resource use.Te total points are then correlated to an estimated 10-year survival and inhospital mortality rate [7,8].Ghanem et al. found that CCI is a predictor of overall survival and late RT side efects including proctitis as the most prevalent gastrointestinal toxicity in prostate cancer patients [9].
We aim to elucidate the complication rate of SpaceOAR hydrogel placement and periprocedural characteristics that may factor into the outcomes so that we can improve the process at our academic hospital.In turn, we can provide patients with evidence-based information on procedural risks and benefts and potentially use perioperative factors to predict complications.

Materials and Methods
Tis research was not conducted under Institutional Review Board oversight as it was a quality improvement project designed not for generalizable knowledge and solely to identify and correct any processes with SpaceOAR insertion that are deviant from standard.We conducted a retrospective review of 233 patients who underwent SpaceOAR placement at our academic hospital by four surgeons (Physicians #1-4) and one radiation oncologist (Physician #5) between 2018 and 2021.SpaceOAR placements were done by four urological surgeons and one radiation oncologist.Our primary outcome was the percentage of patients experiencing complications from the procedure, and our secondary outcome was to detect some variables that could have led to the primary outcome.Variables such as demographics, medical history, oncologic parameters, and radiation plan were gathered.In addition, variables required to calculate the CCI were collected, and CCI was computed to assess preoperative comorbidity risk for each patient.
Physicians at this institution performing SpaceOAR insertion were a mix of junior and midcareer physicians.Tough they had diferent years of experience as practicing physicians, they all started performing this procedure around the same time period.All fve physicians performing the procedures had identical training with the industry representatives, including computer simulation training as well as a minimum of ten proctored procedures by an industry representative.Terefore, they all used the same instrumentation and technique disseminated by Boston Scientifc [10].
Mann-Whitney tests and Fisher's exact tests were performed to compare demographics, oncological history, CCI, and other variables between those who had complications and those who did not.All statistical tests were two tailed, and p < 0.05 was considered statistically signifcant.Statistical analyses were performed using GraphPad Prism 9.0.2 (San Diego, CA).
Multiple logistic regression was performed with clinically relevant factors, which were determined based on results on univariate analyses and relevance to patient outcome before or during the SpaceOAR insertion procedure: age, BMI, physician performing the procedure, if the patient received hormone therapy prior, CCI, and if the patient was on anticoagulation therapy.Race was not in the calculations due to a lack of diversity in this patient population to include it as a predicting factor in multivariate analysis.For the categorical variables, the physician with the least complications percentage-wise, patients who did not receive hormone therapy, and patients not on anticoagulation medications were used as the baseline comparators.
In comparisons between the cohort with complications and the cohort without complications, there were no

Discussion
Our study showed that complications from SpaceOAR placement at our academic institution was in line with previous studies.Te procedures were relatively safe with 10.7% of 233 patients experiencing transient complications similar to previous studies fnding no complications with placement up to 10% in the randomized controlled Pivotal Trial [2,3,11].
From the quality improvement standpoint, we found that there was a signifcant diference in the rate of patients with any complications or no complications between each physician performing the procedure.Pinkawa et al. demonstrated that there is a learning curve to hydrogel spacer placement fnding signifcantly increased symmetry of the gel after the frst 15 patients, more exclusion of the rectum from the planned target volume, and decreased dosage to the rectum [12].Tey had a previous study discovering that the total procedure time decreased from 25 minutes to 14 minutes and the needle insertion time decreased from 10 minutes to 5 minutes between the frst 10 and next 10 patients [13].In our study, 4 out of 5 physicians performed greater than 15 SpaceOAR placements.Te variability in procedure numbers refected our internal practice and referral patterns.However, a learning curve was not visible based on complication rates (Figure 1).Since complications as defned in our study, such as pain and lower urinary tract symptoms, are not the most directly correlated with skill, we also looked at rectal wall injection of hydrogel as identifed on MRI, and the incidences were spread throughout our study period [14].
Multivariate analysis revealed a signifcant diference in complications with those who received hormone therapy having lesser odds of reporting complications than those who did not receive hormone therapy.Te mechanism of this fnding is not clear and may not be causative, but presumably, androgen deprivation therapy (ADT) decreases the size of the prostate gland and induces tumor regression; the reduced tumor size is associated with decreased vascularity, which could improve complications such as bleeding, rectal fullness, and pain.A smaller prostate and tumor after ADT could also allow more precision during placement of the SpaceOAR gel with a more standard anatomy and more "space" in the pelvic cavity.
As with retrospective analysis of databases, there are limitations to this study.Additionally, the quality improvement goal of this project meant that the study was not systematically designed, nor did it include randomization.Our goal of comparing our institution's outcomes with the standard for this procedure can only be based on outcomes analysis from the consistency of data reported as part of care and information available in the notes.As with all procedures, benefts and risks must be weighed.Te distance between the prostate and rectum has been cited one of the factors most correlated to preventing rectal toxicity [15,16].Tis would also be the measure to determine a practitioner's "successful" placement and a marker to use for monitoring and comparing between physicians.In this study we were able to assess the risks, but we did not have the measurements of the thickness of the hydrogel to quantify the beneft of the procedure.Other limitations include not having the 4 Advances in Urology data on the amount of time since injection until radiation, the time since injection when measurement of the Space-OAR thickness is done, the total volume of each patient's injection by radiograph, and longer follow-up for more signifcant or later complications.Another limitation is the categorical nature of some of the variables collected, such as the hormone therapy where knowing the type, dosing, or length of treatment would stratify the patients and elucidate any associations, especially with the discovery that receiving ADT before SpaceOAR insertion has less odds of postprocedural complications.
With the quick advances in RT, our understanding of RT-related rectal toxicity would beneft from a randomized control trial between the diferent radiation modalities.Ten, assessing the complications of the hydrogel placement compared to adverse events with radiation without a spacer within modalities would also guide patient care.Future studies can look at confounding factors such as the distance between the prostate and rectum pre-hydrogel placement to account for individual anatomy, comparison with institutions in which residents perform these procedures under attending supervision, their year in training, and their case logs instead of the listed attending experiences.

Conclusion
In the population of men who had SpaceOAR hydrogel placement at our academic hospital, we found that the procedure had transient and mild complications at a rate in line with previous studies.Te factor that correlated most with diferences between having complications or no complications was the listed attending on the procedure.However, patients who received hormone therapy had less odds of reporting complications after procedures.Tis will lead to more detailed investigation into the complication rates and steps for improving hydrogel placement at this academic institution.

Figure 1 :
Figure 1: Quarterly timeline of the total number of cases and number of which one or more complications were reported per physician.

Table 1 :
Comparison of perioperative characteristics between patients with or without complications from SpaceOAR hydrogel placement.
* Each patient can receive more than one radiation treatment; percentages do not add up to 100.

Table 2 :
Multiple logistic regression including clinically relevant factors.
Bolded values are p < 0.05.Advances in Urologycompared to Physician #2 who was used as the reference.However, patients that received hormone therapy had less odds than patients who did not have hormone therapy to have complications (0.38 [0.15-0.99],p � 0.048).Age, BMI, CCI, and whether the patient was receiving anticoagulation were not signifcant on multivariate analyses.