New findings regarding the influence of assistants on surgical outcomes in penile prosthesis implantation

Penile prosthesis implantation is the definitive treatment for refractory erectile dysfunction, yet exposure to this procedure during training of urology residents is often limited. To assess the effects of resident participation in penile prosthesis surgery, we compared surgical outcomes in a retrospective case series of 253 penile prosthesis surgeries by a single surgeon at the same institution between 2017 and 2020 with the assistance of either a registered nurse first assistant (RNFA) or a resident. Pertinent patient characteristics and surgical complications including device complications, surgical site infection, postoperative bleeding, iatrogenic injury, cardiovascular events, pulmonary events, and urinary retention were documented. Measured outcomes included operative time, Emergency Room (ER) visits, unplanned postoperative visits, pain medication refills, and surgical complications. Compared to RFNAs, resident-assisted penile prosthesis surgery was associated with significant increase in mean operative time (71.4 min vs. 87.9 min, p < 0.01) and postoperative ER visits (3.0% vs. 10.6%, p = 0.03) but not surgical complications (19.7% vs. 20.8%, OR 1.03, 95% CI [0.46 −2.30]) or other measured outcomes. Compared to a dedicated RFNA, Resident assistance increased operative time by approximately 17 min, but did not increase post-operative surgical complications, supporting the notion that resident assistance in these procedures may be appropriate as an integral part of training.


INTRODUCTION
Erectile dysfunction (ED) is one of the most prevalent diseases in the United States, affecting as many as 52% of men between ages 40 to 70 [1].Approximately 3% of patients with ED refractory to medical management undergo penile prosthesis surgeries, which significantly improves patients' quality of life [2,3].Given the prevalence of ED and high satisfaction rate of penile prosthesis placement [4,5], exposure to this surgery during residency training is vital.
The goals of an academic training program are to balance efficiency, beneficial results, and patient safety with resident education.However, obtaining in-depth exposure to penile prosthesis surgeries during residency can be challenging.In a 2015 survey, only 15% of urology training programs have a dedicated prosthetic surgeon, and 75% of penile prostheses are placed by surgeons who do so less than 5 times per year [6,7].Moreover, operative volume for surgery residents has also decreased since implementation of work hour restrictions [8], possibly further limiting the acquisition of penile prosthesis implantation surgical skills for trainees.
Prior studies have reported that resident participation with appropriate supervision does not lead to worse outcomes in a wide array of urological procedures, but only limited data are available in prosthetic urology.McAbee et al. reported a single cohort case series of penile prosthesis implantation surgeries with resident involvement which did not show increased postoperative complication compared to the national average [9].However, prior study only investigated the influence of a resident assistant on patient outcomes.
The limited prosthetic exposure of residency raises the question if utilizing dedicated registered nurse first assistants (RNFA) would influence the surgical outcomes differently.Reports by nonurological specialties find mortality rates in a variety of surgical procedures to be independent of types of first assistant, but certain adverse events are associated with resident involvement in specialized procedures [10][11][12].Given the limited reports on how assistants may influence the outcomes of penile prosthesis surgeries, we report a single-surgeon case series comparing the surgical outcomes of penile prosthesis surgery performed with the assistance of RNFA versus residents.

Study design
Medical records of patients who underwent primary and revision penile prosthesis surgery by a single surgeon at the same institution between 2017 and 2020 were retrospectively reviewed with IRB approval (Dartmouth CPHS 0031589).Patient characteristics including age, prior penile prosthesis, race, BMI, smoking status, diabetes, preoperative laboratory results, and follow-up duration were recorded.Assistants for each surgery were recorded.Assistants were assigned based on availability and without preference regarding junior versus senior resident training level.RNFA were always available, while residents were either rotating through as part of an optional elective rotation and/or assigned if available.In all cases the attending surgeon was the primary operator and had only one assistant, whose assignment was arbitrary and based on available personnel with no option for attending surgeon preference.All implants are prepared by the scrub nurse or technician.All patients were seen at 2 and 6 weeks postoperatively, with subsequent follow-up on as needed basis.

Measured outcomes
Surgical outcomes were defined as operative time, Emergency Room (ER) visits, and documented postoperative complications during the follow-up period.We defined our primary outcome as postoperative complications, and secondary outcomes as Clavien-Dindo classifications (CD), operative time, ER visits, unplanned post-op visits, and pain medication refill during follow-up period.Complications include device complications, surgical site infection, postoperative bleeding, iatrogenic injury, cardiovascular event, pulmonary event, and urinary retention.Clavien-Dindo classifications of each reported complication was also recorded [13].

Inclusion and exclusion criteria
All patients who underwent primary and revision penile prosthesis surgery by a single surgeon at the same institution between 2017 and 2020 were screened.Patients without adequate documentation of follow-up, type of assistant, or complications with CD classifications were excluded from the analysis.There were no patient-related exclusion criteria.

Statistics
Baseline characteristics were stratified and outcomes compared with assistant type (resident vs. RFNA) as the independent variable.Categorical variables were compared using Pearson's Chi-square test or Fisher's exact test.Continuous variables with were reported in median with interquartile range (IQR) and compared using Wilcoxon's rank sum test.Association of covariates to complications is reported in odds ratios estimated by univariate and multivariate analysis with logistic regression model.All statistical tests were performed using JMP Pro 15 (SAS Institute, Cary NC, USA) with a two-sided significance level set at P < 0.05.

RESULTS
A total of 253 consecutive independent penile prosthesis surgeries with an assistant were identified.We excluded 2 patients because no follow-up visits were documented, leaving 251 patients for analysis.All cases only had one assistant.There were no patients with both a RNFA and resident as assistants.Patients were followed for a median of 11.1 (IQR 1.5-14) months.
Device-related complications were erosions, perforations, malposition/migration of components, fibrosis requiring switch from IPP to malleable prosthesis, patient-requested revision due to dissatisfaction, or device malfunction (Supplementary Table 2).Among the device complications, 21 of the 29 events were CD grade 3b or above and underwent surgical revisions.
Increased ER visits were associated with resident-assisted surgeries (3.0% vs. 10.6%, p = 0.03).However, there were no statistically significant differences in the rate of unplanned post-op visits (35.6% vs. 38.3%,p = 0.74), pain medication refill (13.9% vs. 12.8%, p = 0.79) or penile prosthesis surgery-associated complications (19.7% vs. 20.8%,p = 0.83) between the RNFA or residents cohort.Moreover, there was no statistically significant difference in the distribution of the CD grade of the complications between the two groups (Table 2).
Patient variables including prior penile prosthesis, BMI ≥ 30, diabetes, smoking status and preoperative laboratory values were included in univariate and multivariate analysis, they were not found to be associated with increased complication rate.Resident- assisted penile prosthesis surgeries were also not associated with increased complication rate compared those assisted by RNFA (Table 3).

DISCUSSION
One goal of urology residency is to expose trainees to a wide variety of surgical techniques in order to provide a foundation for future practice.Given the efficacy of penile prosthesis surgeries in treating refractory ED [2,3], residents should have adequate exposure to these cases.This exposure must also be measured against surgical goals of efficiency, satisfactory outcomes, and low rates complications.Dedicated RNFAs are often utilized in many surgical procedures when residents or MD-surgeon support is not readily accessible [10][11][12], however there are no prior reports on how different assistants might affect the outcomes of penile prosthesis surgeries.
Our study evaluates these metrics in a novel fashion by comparing penile prosthesis cases performed with residents and RNFA.
Prior findings in urological surgical education demonstrated that resident involvement universally increases operative time in open, laparoscopic, and robotic surgeries [14][15][16][17][18]. Studies from general surgery demonstrated similar findings when comparing resident directly to non-trainee first assist (PA/RN) [10,11].Our study showed that operative time was noted to be slightly increased (71.4 min vs. 87.9min, p < 0.01) with resident compared with RNFA in penile prosthesis surgeries.One possible explanation is that the trainees are not as familiar with surgical techniques compared to the dedicated RNFA, leading to increased OR time.However, McMillan et al. previously demonstrated with that robotic-assisted laparoscopic prostatectomy assisted by senior residents have longer operative time compared to those assisted by junior residents [16].Another study from the National Surgery Quality Improvement Program (NSQIP) database looking at male one-stage urethroplasties showed that operative time were only significantly increased when chief residents were involved, but not the junior residents [19].Our results also showed that longer operative time was associated with higher PGY level.These results suggest that the increased operative time is likely attributed by increased surgical education and delegation of case responsibilities commensurate with PGY level, whereas RNFA increases efficiency typically by remaining consistent across procedures.Stratified operative time analysis also showed that only new penile prosthesis surgeries assisted by resident are associated with increased operative time compared to RNFA, but not the revisions.This observation further supports that increased OR time is likely attributed by selective bias for cases appropriate for increased resident involvement.Prior study has also linked prosthesis surgeon with higher volume shorter operative time [7], which is also consistent with our findings, since the median cases assisted by a single RNFA is higher than that of resident in this series (35 vs. 6).A possible area of future study is to see if the senior resident's operative time would start to decrease after certain number of cases, in order to further establish the learning curve of penile prosthesis surgery.
Prior studies have shown that resident involvement is not associated with increased complication rate in a variety of urological procedures [14][15][16][17][18]. Specific to penile prosthesis implantation surgery, McAbee et al. reported a single cohort case series with resident involvement, which did not show increased postoperative complication compared to the national average [9].However, there are reports that associate certain adverse events with trainee involvement.A study from NSQIP database suggested that senior trainees involvement with open radical prostatectomy is associated with increased overall morbidity than with junior trainees, whereas morbidity of laparoscopic radical prostatectomy is independent of trainee involvement [20].Another study reported that resident involvement in transurethral resection of bladder tumor is associated with increased odds of 30-day readmission regardless of PGY level, but not with the overall complication and reoperation rate [21].Studies in general surgery education that directly compared outcomes with trainee involvement to that of nontrainee first assist (PA/RN) have also shown increased readmission or ICU admission rates in specialized procedures [10,11].
In our 251-patient series, there was increased rate of ER visits (3.0% vs. 10.6%)associated with resident involvement, but no difference in unplanned post-op visits, pain medication refill, or overall complication rate (19.7% vs. 20.8%).On multivariate analysis, resident involvement is also not associated with increased overall complication rate (OR 1.07, 95% CI 0.46-2.30).There was also no association of more severe complication with resident involvement, as the CD grade distribution is comparable between the 2 groups.Moreover, the penile prosthesis revision rates, as suggested by CD 3b complications in the RNFA (9.9%) and resident (6.3%) cohorts, are both lower than the previously reported 18% among US veterans who underwent IPP placement [22].These results suggest that complication rate of penile prosthesis surgeries is independent of the type of first assist for high volume surgeons with low complications.Unfortunately, our result does not provide insight into the increased ER visits associated with resident involvement.One possible explanation is that the primary surgeon may be more involved in the discharge process and instructions when operating without a resident, thereby preventing unnecessary ER visits.Surgeon's clinic availability for same visits may also fluctuate at different times, which is a potential confounder for increased ER visits not captured by our database.
Prior studies have suggested patient risk factors such as smoking or DM increase overall complications [23] and are associated with penile prosthesis revision surgeries [22].In this case series, history of smoking or DM were not independent predictors of complications on multivariate analysis.Although not statistically significant, there is increased complication rate (OR 2.34, CI 0.84-6.52)associated with current smoking status.A possible area of future investigations would be if the type or experience level of the assistant would improve complication rate in certain high risk patient populations.
Lastly, there is no observable difference of the infection rates between the 2 cohorts (0.5% vs. 2.1%), although the low number of postoperative infections (n = 2) may not have enough power to detect the differences.Infection is one of the most feared complications for penile prosthesis surgeries and often requires surgical explants of the penile prosthesis.The overall infection rate of our series is 0.8%, lower than previously reported 1-4% associated with IPP surgeries [24,25].Our results suggest that with appropriate supervision by a surgeon with penile prosthesis implantation expertise, the infection rate is independent of first assistant, and it would be safe for either residents or RNFA to assist in penile prosthesis without compromising outcome.This is the first report on how different types of first assist may affect the outcomes of penile prosthesis surgeries.Our study has several limitations due inherent bias of a retrospective study.Study power is limited due to small sample size in the resident cohort.Resident participation in penile prosthesis surgeries were relatively recent at our institution, leading different length of follow-up duration, disproportionate number of cases between the two groups, and possibly skewing the number of events.Generalizability of the study may also be limited due to the singlesurgeon experience in a racially homogenous patient population.Different avenues for future study may be using national or multiinstitutional database to validate these observations.CONCLUSION Resident-assisted penile prosthesis surgeries are associated with increased OR time but is not associated with increased complications compared to those assisted by dedicated RNFA.Under the supervision of experienced implant surgeon, these results suggest residents can participate in specialized prosthetic surgeries without compromising patient outcomes.

Table 1 .
Baseline patient characteristics of IPP surgeries assisted by RNFA or resident.Categorical variables (Prior IPP, Race, DM, Smoking status, urine culture) were reported as the number of cases (and percentage of total) and compared using Pearson's Chi-square test.Continuous variables (Age, follow-up duration) were reported in median with IQR and compared using Wilcoxon's rank sum test.IQR interquartile range, IPP Inflatable penile prosthesis, BMI body mass index, DM diabetes mellitus, RNFA registered nurse first assist.
aIndicates statistical significance at the 0.05 level.S.-c.Wu et al.

Table 2 .
Operative outcomes and complications of IPP surgeries assisted by RNFA or resident.
Categorical variables (ER visit, unplanned post-op visit, pain med refill, complications and Clavien-Dindo classification) were reported as the number of cases (and percentage of total) and compared using Pearson's Chi-square test or Fisher's exact test.Continuous variables (operative time) were reported in median with IQR and compared using Wilcoxon's rank sum test.IQR interquartile range, IPP Inflatable penile prosthesis, RNFA registered nurse first assist, ER Emergency room.a Indicates statistical significance at the 0.05 level.

Table 3 .
Predictors of IPP surgery complications on univariate and multivariate analysis.All variables reported in median with IQR.Association of covariates to complications is reported in odds ratios estimated by univariate and multivariate analysis with logistic regression model.UVA univariate analysis, MVA multivariate analysis, IQR interquartile range, IPP Inflatable penile prosthesis, BMI body mass index, DM diabetes mellitus, RNFA registered nurse first assist.