Changes in Anesthesia Can Reduce Periprocedural Urinary Retention After EVAR

Abstract

OBJECTIVE:

Enhanced recovery after surgery (ERAS) programs provide a streamlined approach for expedient post-operative care of high-volume procedures. Endovascular aortic repair (EVAR) has become standard treatment for abdominal aortic aneurysms and implementation of an early recovery program is warranted. Post-operative urinary retention (POUR) remains a problem lending to longer hospital stays and patient discomfort. We aim to demonstrate the utility of monitored anesthetic care (MAC) plus local anesthesia as a modality to minimize urinary retention following EVAR.

METHODS:

Single-center retrospective review from January 2017 to March 2020 of all patients undergoing standard elective EVAR under general anesthesia or MAC anesthesia. Local anesthetic at vessel access sites was used in all patients under MAC. Ruptured pathology and female sex were excluded from analysis. Patient characteristics, operative details, prostate measurements, and outcomes were abstracted from the electronic medical record. Urinary retention was defined as any requirement of straight catheterization, urinary catheter replacement, or discharge with urinary catheter. Chi square tests and logistic regression were used to determine predictors associated with POUR and increased hospital length of stay.

RESULTS:

Among 138 patients who underwent EVAR, eight (5.8%) were excluded due to ruptured pathology. Of the cohort, 113 (86.9%) were male with mean age of 73 years. Excluding female patients, 63 (55.8%) male patients underwent general anesthesia and 50 (44.3%) underwent MAC. Male patients under general anesthesia were more likely to have intraoperative urinary catheter placement when compared to MAC (82.5% vs 36%, respectively; p<0.001). POUR was identified in 17 patients (13.1%) of the entire study population with 15 events (88.2%) occurring in males. Excluding patients who were admitted to the ICU, twenty-two (19.5%) male patients stayed past post-operative day (POD) one, of which those who developed POUR were more likely to experience compared to those without POUR (45.6% versus 9.7%, respectively; p=0.001). On multivariable analysis, male patients who received MAC had a lower risk of developing POUR (OR 0.09, 95% CI 0.02 – 0.50). POUR was not associated with elective urinary catheter placement nor with pre-existing conditions such as diabetes, urinary retention, benign prostatic hypertrophy (BPH), or use of BPH medications. Additionally, neither prostate size nor volume was associated with developing POUR among male patients.

CONCLUSION:

MAC plus local anesthesia is associated with decreased rates of POUR after elective EVAR in male patients. ERAS pathways during elective EVAR interventions should implement MAC plus local anesthesia as an acceptable anesthetic option, where appropriate, in order to reduce urinary retention rates and subsequently decrease hospital length of stay in this patient cohort.

1.0. INTRODUCTION

Enhanced recovery after surgery (ERAS) programs provide a streamlined approach for expedient post-operative care of high-volume procedures. Since the implementation of endovascular aneurysm repair (EVAR) of abdominal aortic aneurysms (AAA), post-operative length of stay (LOS) has continued to decrease to the point where many patients now are discharged the day after the procedure.^1–3 Patients that usually meet eligibility criteria for early discharge include those with low preoperative risk and amenable aortic anatomy.^4–7

Despite careful patient selection, post-operative urinary retention (POUR) remains a problem in operations that implement ERAS pathways lending to longer hospital stays, patient discomfort, and risk for urinary tract infection (UTI).^8–10 Several risk factors have been demonstrated to be associated with POUR including history of urinary retention and certain comorbidities such as diabetes mellitus.¹¹ It is recognized that certain periprocedural measures can be implemented to ameliorate POUR such as preoperative voiding or administration of alpha-blockers however, use of intraoperative indwelling urinary catheters has varying results.^{8, 12, 13}

The incidence of POUR in patients undergoing elective EVAR has been documented between 4.5% and 13.0%.^{3, 14} These studies however, did not specifically look at predisposing or perioperative risk factors associated with POUR. Therefore, the objective of this study was to demonstrate the utility of a change in anesthetic using monitored anesthetic care (MAC) plus local anesthesia as a modality to minimize urinary retention following EVAR along with selective procedural usage of an indwelling urinary catheter as compared to general anesthesia (GA) with and without a catheter. We subsequently aimed to determine whether or not POUR was associated with increased hospital length of stay.

2.0. METHODS

A single-center retrospective review of all patients who underwent elective standard EVAR for infrarenal abdominal aortic aneurysm from January 2017 to March 2020 was conducted. This study was approved by the university’s Health Science Institutional Review Board with waiver of informed consent. The hospital’s electronic medical records were queried using current procedure terminology (CPT) codes. Patients were included if the vascular anatomy met information for use (IFU) criteria for the specific endovascular aortic device used. EVAR for treatment of endoleaks or graft degeneration of previously placed endografts were also included. Patients requiring conversion to open repair and those with ruptured aortic pathology were subsequently excluded.

Our primary outcome variable of interest was development of POUR after elective EVAR with our secondary outcome variable including post-operative length of stay. Our primary predictor variable of interest was use of MAC anesthesia versus general anesthesia. All patients under MAC anesthesia received local anesthetic composed of plain 1% lidocaine (10 mg/ml) in areas where vessel access was to be obtained. Use of this local anesthetic at case completion was based on surgeon preference which we were unable to capture in the data. Two independent data extractors collected information from the patient electronic record, Epic (Epic Systems Corporation, Verona, WI), which included patient demographics, past medical history including comorbidities and urological history, surgical history including prostatectomy, outpatient medications and prostate measurements from the most immediate pre-operative computed tomography angiography (CTA) if applicable. History of urinary retention was documented if listed as a problem in the electronic medical record (EMR) or if the patient was on medications including alpha1-adrenergic blockers (tamsulosin, doxazosin, alfuzosin) or finasteride and had a history of benign prostatic hypertrophy (BPH). History of chronic kidney disease (CKD) and BPH were included if documented as an “active problem” within the EMR or documented in the preoperative assessment note performed by the anesthesia team. Patients were asked to discontinue any loop diuretics the day of surgery but continue other diuretics, including thiazides and spironolactone, in addition to BPH medications. Males with a history of prostatectomy were categorized as not having BPH given absence of a prostate preoperatively. Prostate size was measured as the largest transverse diameter on CTA axial slices and prostate volume was estimated by using the ellipsoid method of multiplying transverse, antero-posterior, and cranio-caudal measurements of the prostate subsequently multiplied by 0.52.^{15, 16} Prostate measurements excluded males with history of prostatectomy.

Pertinent intraoperative data collected included indication for EVAR, American Society of Anesthesiologists (ASA) surgical risk classification, type of anesthesia including doses of anesthetic components and relevant adjunctive procedures that may have impact on the pelvic circulations such as iliac branch extenders (IBE) and hypogastric artery coiling. Operative time and radiation measurements based on air kerma were considered surrogates for case complexity in addition to the adjunctive procedures. All procedures were conducted using standard “preclose” techniques.¹ It was standard protocol in all cases for patients to urinate prior to the operation however placement of an indwelling urinary catheter for the procedure was at the discretion of the primary surgeon most often based on expected operative time. In addition, postoperative length of stay, intensive care unit (ICU) admission, and complications were documented in addition to EVAR-related 30-day readmission and 30-day mortality. Postoperative LOS was categorized into discharge on or prior to post-operative day (POD) one versus after POD one.

Postoperative removal of the urinary catheter was documented as either within 24 hours of surgery or after. Patients were encouraged to ambulate after their “flat time” was completed, which ranged between 4 to 6 hours post-operatively, however, this was not specifically documented in the patient’s EMR. Our primary endpoint was postoperative urinary retention, which was defined as any patient requiring straight catheterization post-operatively, replacement of an indwelling urinary catheter after initial removal, or discharge with an indwelling urinary catheter after initial removal. Patients were bladder scanned six hours after initial catheter removal, if unable to void and bladder volume was greater than 600 ml, the patient was offered straight catheterization. Urinary tract infection was defined as either having a clinical diagnosis on urinalysis or symptoms of dysuria.

2.1. Statistical analysis

Our primary and secondary endpoints were analyzed using only men given evidence of their increased risk of POUR and our limited sample size.^{12, 17} Patient characteristics between those that developed POUR and those that did not were analyzed using bivariate chi square tests for categorical variables and either t test for normally distributed continuous variables or Mann-Whitney U test for non-normally distributed continuous variables. Continuous variables are presented as mean and standard deviation (SD) if normally distributed or median with interquartile range (IQR) if not. Covariates included in the multivariate logistic model were identified a priori based on literature review with manual selection of covariates from bivariate analysis at an alpha level of 0.05.¹⁸ Collinear variables were eliminated until a final model was estimated.¹⁹ All statistical analysis was performed using STATA SE version 16.1 (StataCorp LP, College Station, TX).

3.0. RESULTS

3.1. Study population and demographic characteristics

There were 138 patients who met inclusion criteria during the study period. Eight participants (5.8%) were excluded due to ruptured pathology. Mean age of the cohort was 73 years with 17 (13.1%) being female. Only one patient (0.77%) had a documented congenital history of vascular aneurysmal degeneration in first degree relatives. Of the cohort, 122 (88%) were asymptomatic aortic aneurysms. Median aortic size prior to intervention was 5.6 cm [IQR 5.2, 6.0] with 36 patients (27.69%) having associated iliac aneurysms. When evaluating men (n=113), there were no statistically significant differences in demographics or medical history between those undergoing intervention under general anesthesia versus MAC. Twenty-seven (23.9%) had a history of urinary retention with 22 (19.5%) having been on BPH medications preoperatively (Table 1). POUR events were more common in men with history of CKD compared to those without (46.7% versus 21.4%, respectively; p=0.04) and preoperative diuretic use compared to those not taking diuretics (53.3% versus 26.5%, respectively; p=0.04). Excluding male patients with history of prostatectomy (n=11), median prostate size was 4.9 cm [IQR 4.3, 5.5] while median prostate volume was 38.9 cm³ [IQR 29.0, 57.3].

Table 1:

Preoperative Characteristics in Male Patients with and without POUR

Patient Characteristics	Total (n=113)	No POUR (n=98)	POUR (n=15)	P-value
Age	73.35 (7.93)	72.98 (8.02)	75.73 (7.13)	0.21
Non-Hispanic White	93 (83.8)	81 (83.5)	12 (85.7)	0.83
BMI	28.14 [24.95, 30.65]	28.45 [25.24, 30.66]	25.51 [23.32, 29.02]	0.12
Current Smoker	20 (17.7)	16 (16.3)	4 (26.7)	0.33
CAD	50 (44.3)	41 (41.8)	9 (60.0)	0.19
CHF	15 (13.3)	11 (11.2)	4 (26.7)	0.1
COPD	28 (24.8)	23 (23.5)	5 (33.3)	0.41
CVA/TIA	13 (11.5)	10 (10.2)	3 (20.0)	0.27
DM	20 (17.7)	19 (19.4)	1 (6.7)	0.23
CKD	28 (24.8)	21 (21.4)	7 (46.7)	0.04
PAD	4 (3.5)	3 (3.1)	1 (6.7)	0.48
Urinary Retention History	27 (23.9)	24 (24.5)	3 (20.0)	0.7
BPH	21 (18.6)	18 (18.4)	3 (20.0)	0.88
Prostatectomy	10 (8.9)	10 (10.2)	0 (0.0)	0.2
Preoperative medications
Beta Blocker	64 (56.6)	54 (55.1)	10 (66.7)	0.40
ARB/ACEi	56 (49.6)	52 (53.1)	4 (26.7)	0.06
Diuretic	34 (30.1)	26 (26.5)	8 (53.3)	0.04
BPH Rx	22 (19.5)	20 (20.4)	2 (13.3)	0.52
Prostate Measurements
Prostate Size (cm)	4.9 [4.3, 5.5]	4.9 [4.3, 5.6]	4.8 [4.7, 5.5]	0.79
Prostate Volume (cm3)	38.9 [29.0, 57.3]	38.1 [29.4, 57.6]	42.0 [28.4, 55.8]	0.88

BMI, body mass index; CAD, coronary artery disease; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; CVA, cerebral vascular accident; TIA, transient ischemic attack; DM, diabetes mellitus; PAD, peripheral arterial disease; ARB, angiotensin receptor blockers; ACEi, angiotensin-converting enzyme inhibitors; Rx, prescription Categorical variables are presented as absolute value (%). Age is presented as mean (SD); BMI, Prostate Size and Prostate Volume presented as median [IQR].

Prostate measurements exclude patients with history of prostatectomy (n=11)

3.2. Operative characteristics

Excluding women, during EVAR, 63 (55.8%) men underwent GA and 50 (44.3%) underwent MAC. Urinary catheters tended to be placed more frequently in men under GA versus MAC (82.5% vs 36%, respectively; p<0.001). Median operative times were longer in patients under GA compared to MAC (159 minutes, IQR 120–213, versus 135, IQR 103–164, respectively; p<0.01) in addition to median air kerma levels (1883 mGy, IQR 1,216–2,790, versus 883 mGy, IQR 608–1,326, respectively; p<0.001). Compared to GA, patients under MAC had higher doses of propofol (median 172.7 mg, IQR 150 – 200, versus 447 mg, IQR 254 – 705, respectively; p<0.001) and lower doses of fentanyl (median 150 mcg, IQR 125 – 200, versus 75 mcg, IQR 50 – 100, respectively; p<0.001).

The majority of endovascular devices included 68 (60.18%) Excluder grafts (Gore, Flagstaff, AZ), 26 (23.01%) Endurant grafts (Medtronic, Minneapolis, MN), and 8 (7.08%) Ovation grafts (Endologix, Irvine, CA). Fourteen patients (13.39%) required IBE devices and five (4.42%) had either unilateral or bilateral hypogastric coiling performed. Patients requiring IBE use tended to be under general anesthesia more frequently than those with hypogastric coiling or no adjunctive intervention (92.86% versus 60% and 50%, respectively; p=0.01). Interventions requiring IBE or hypogastric coiling involved longer operative times (median 211 minutes, IQR 169 – 313, and median 149 minutes, IQR 139 – 193, respectively) compared to no adjunctive intervention (median 136 minutes, IQR 109 – 172; p<0.001). Patients requiring IBEs were more likely to have intraoperative urinary catheters placed (92.86%) compared to either hypogastric coil embolization or no adjunctive intervention (60% and 57.45%, respectively; p=0.04). However, there was no statistically significant association between IBE use or hypogastric embolization with the development of POUR (7.14% and 20.0%, respectively) when compared to no adjunctive intervention (13.83%). Femoral artery access was obtained either percutaneously with Perclose ProGlide devices (n=106, 93.81%; Abbott, Plymouth, MN) or surgical cut-down (n=7, 6.19%); all cut-downs were performed under general anesthesia. Of the seven cut-downs, one was due to perclose device failure and two were performed with direct access to the external iliac arteries via lower quadrant oblique incisions given either presence of groin infection or occluded femoral vessels. The remainder were performed for severely calcified femoral vessels obviating use of a perclose device.

3.3. Postoperative outcomes

Among men, there were 3 (2.83%) access site complications, consisting of small hematomas only requiring observation, in those who had percutaneous access. Nine (7.96%) required ICU admission which was associated with higher ASA class (6.1% with ASA 2 versus 28.57% with ASA 3; p<0.01). Indications for ICU admission included severe obstructive sleep apnea (n=4), surgeon preference given extent of intervention performed and patient comorbidities (n=3), and intraoperative atrial fibrillation with rapid ventricular response (n=2). In addition, patients who underwent access site cut-down compared to percutaneous access were prone to ICU admission (42.86% versus 5.66%, respectively; p<0.001) however, there was no association with access site complications. Excluding ruptures, there was only one 30-day mortality in the elective repair group (0.93%) in an octogenarian who had an aspiration event eventually leading to hypercapneic respiratory failure.

3.4. Factors associated with POUR and LOS

Of the entire cohort, the total POUR event rate was 14% which included two female patients versus 15 male patients. In male patients, POUR was associated with history of CKD and pre-operative diuretic use (Table 1). Among male patients with POUR, median prostate volume was 42.0 cm³ versus 38.2 cm³ in those without POUR (p=0.79). No pre-existing conditions such as diabetes, urinary retention history, BPH, or use of BPH medications was associated with POUR. Of note, 10 male patients had a history of prostatectomy and none developed POUR versus 15 (14.56%; p=0.20) who did not have a history of prostatectomy. Neither placement of a preoperative urinary catheter nor time to removal of the urinary catheter was associated with POUR (Table 2). Male patients who underwent endovascular repair under MAC experienced less POUR events compared to those under general anesthesia (4.0% vs 20.6%, respectively: p=0.01). There was no statistically significant association between the amount of crystalloid administered intraoperatively with POUR. On multivariable analysis, after adjusting for covariates, MAC anesthesia remained protective against POUR (OR 0.09, 95% CI 0.02 – 0.50, p=<0.01) while other known risk factors were not significantly associated with POUR in this cohort (Table 3). Three male patients with pre-operative urinary catheters (4.29%; p=0.17) developed dysuria after urinary catheter removal and were subsequently treated with antibiotics. Only one of these patients had developed POUR (Table 4).

Table 2:

Operative Characteristics in Male Patients with and without POUR

Operative Variables	Total (n=113)	No POUR (n=98)	POUR (n=15)	P-Value
Symptomatic aneurysm	6 (5.3)	5 (5.1)	1 (6.7)	0.80
Length of Surgery (min)	143 [116, 176]	140 [109, 174]	159 [130, 199]	0.10
MAC Anesthesia	50 (44.3)	48 (49.0)	2 (13.3)	0.01
Propofol (mg)	200 [150, 447]	210 [150, 447]	155 [120, 379]	0.25
Fentanyl (mcg)	125 [75, 175]	125 [75, 175]	113 [75, 150]	0.94
Midazolam (mg)	1 [0, 2]	1 [0, 2]	0 [0,1]	0.03
Phenylephrine (mcg)	200 [0, 3390]	100 [0, 3060]	1545 [0,6245]	0.10
Crystalloid (L)	1.4 [0.9, 1.8]	1.4 [0.9, 1.8]	1.5 [0.8, 2.0]	0.77
Contrast (ml)	80 [68, 115]	80 [67, 121]	85 [68, 101]	0.79
Urinary Catheter	70 (62.0)	61 (62.2)	9 (60.0)	0.87
Urinary Catheter	6 (8.96)	5 (8.5)	1 (12.5)	0.19
Removal >24 Hours

Categorical variables are presented as absolute value (%). Age is presented as mean (SD), remaining continuous variables presented as median [IQR]

Time to urinary catheter removal excludes patients without intraoperative urinary catheters (n=43) and those with unknown time of removal (n=3)

Table 3:

Operative and Perioperative Predictors of POUR in Males

Covariates	OR	95% CI	P-value
MAC Anesthesia	0.09	(0.02 – 0.50)	<0.01
Urinary Catheter	0.29	(0.07 – 1.17)	0.08
Age >= 80 years	1.05	(0.25 – 4.44)	0.94
CKD	2.67	(0.69 – 10.36)	0.16
Urinary Retention
History	0.86	(0.19 – 3.84)	0.85
Diuretic	2.32	(0.64 – 8.43)	0.20

OR, odds ratio; CI, confidence interval; CKD, chronic kidney disease

Table 4:

Postoperative Outcomes Associated with POUR in Males

Outcomes	Total (n=113)	No POUR (n=98)	POUR (n=15)	P-value
New BPH Rx	12 (10.6)	3 (3.1)	9 (60.0)	<0.001
UTI	3 (2.7)	2 (2.0)	1 (6.7)	0.3
LOS > POD 1	14 (13.5)	9 (9.7)	5 (45.6)	<0.01
30-day readmission	3 (2.7)	3 (3.1)	0 (0.0)	0.51

UTI, urinary tract infection; POD 1, postoperative day 1

Categorical variables are presented as absolute value (%)

LOS>POD 1 excludes patients admitted to the ICU (n=9)

When excluding patients who were admitted to the ICU (n=9), twenty-two (19.5%) male patients stayed past POD one which was associated with development of POUR compared to those without (45.6% versus 9.7%, respectively; p=0.001). On multivariable analysis, male patients with POUR had increased likelihood of staying passed POD one (OR 4.56, 95% CI 1.11 – 18.73) however, length of stay was not associated with type of anesthesia used (Table 5).

Table 5:

Operative and Perioperative Predictors for Length of Stay Past Postoperative Day One in Males

Covariates	OR	95% CI	P-value
MAC Anesthesia	0.65	(0.1 – 3.17)	0.59
Procedure Time >= 3 Hours	3.01	(0.72 – 12.53)	0.13
CKD	4.60	(1.32 – 15.99)	0.02
CHF	1.33	(0.23 – 7.86)	0.75
POUR	4.56	(1.11 – 18.73)	0.04

OR, odds ratio; CI, confidence interval; CKD, chronic kidney disease; CHF, congestive heart failure

Categorical variables are presented as absolute value (%)

Analysis excludes those patients admitted to the ICU (n=9)

4.0. DISCUSSION

Enhanced recovery after surgery pathways have had profound impact in various surgical fields as technology continues to advance and protocols are improved. Despite a move toward these pathways, vascular surgery seems to lag behind. A recent systematic review and meta-analysis performed by Visioni et al on ERAS pathways in non-colorectal surgery actually excluded vascular surgery cases, although they were explicitly looking at major abdominopelvic interventions.²⁰ Nevertheless, several studies have been performed trying to determine the feasibility of same-day or early discharge after elective endovascular aortic repair.^{4, 14} In an attempt to determine suitable candidates for same-day discharge after elective EVAR, Montross et al demonstrated that 84% of patients had general anesthesia, 6% required ICU admission and 13% had developed POUR.³ Our cohort had similar findings in regard to ICU admission and POUR rates however had less interventions performed under general anesthesia (55.8%). An important difference with their analysis however, was the inclusion of females and only including asymptomatic abdominal aortic aneurysms.

Our observational study demonstrated that MAC anesthesia is associated with less post-operative urinary retention rates in male patients. Additionally those with post-operative urinary retention tended to stay in the hospital greater than one day post-operatively. Our findings add to the evidence supporting MAC and local anesthesia as a modifiable factor for POUR however, there remains a paucity of literature looking at predictors of POUR during endovascular aortic repair.¹¹ We routinely use local anesthetic composed of plain 1% lidocaine (10 mg/ml) in patients under MAC anesthesia to anesthetize the access sites however, it is variable as to whether or not this same local anesthetic is used at case completion given it is surgeon dependent. It has been our experience that most patients with percutaneous access do not complain of access site pain and if so it is usually controlled with 650 mg of oral acetaminophen given every six hours as needed. If patients require more pain control, they are offered 5 mg of oxycodone or 50 mg of tramadol every six hours as needed for a maximum of three days post-operatively to minimize narcotic use. A recent systematic review looking at different anesthetic approaches to EVAR noted 14 studies that included patients who had elective EVAR surgery with inpatient mortality, 30-day mortality, duration of surgery, and length of hospital stay being the most commonly reported outcome measures. These studies did not specifically look at anesthesia effect on POUR nor its association with length of stay.²¹ Studies that evaluate POUR within the realm of vascular surgery include patients undergoing carotid endarterectomy (CEA) interventions. Boitano et al noted a POUR rate of 28.2% in 294 male patients undergoing CEA that was associated with older age, CKD, and history of urinary retention.⁸ CKD as a risk factor is concordant with our findings however our study demonstrated no association with age or history of retention with POUR likely due to our small sample size. This group also demonstrated a protective effect of intraoperative urinary catheter use which is similar to our study as we demonstrated an approach toward significance with an OR of 0.29 (p=0.08). This is thought to be due to passive bladder emptying preventing overdistention and therefore avoiding detrusor muscle injury.¹¹ Boitano et al also noted no association between urinary catheter use and post-operative UTI which is concordant with our findings although other studies have demonstrated evidence to the contrary.¹⁰

Tam et al looked at enhanced recovery pathways in colorectal surgery and demonstrated that ERAS pathways had shorter indwelling urinary catheter times (median 1 day versus median 2 days in non-ERAS group) and that these pathways were associated with a statistically significant 96% increased odds of developing POUR.⁹ They conclude that early removal of urinary catheters portends an increased risk of POUR. Our study did not demonstrate such findings with the majority of our patients having urinary catheters removed less than 24 hours postoperatively. Their interventions however, were done under general or spinal anesthesia and their threshold for catheter reinsertion or straight catheterization was 300–400 ml on bladder ultrasound. In our cohort our threshold to reinsert a urinary catheter was 600 ml, with studies demonstrating that waiting until 800 ml to be safe.²² Thresholds for straight catheterization or reinsertion of urinary catheters is usually dependent on surgeon or trainee preferences which makes it difficult to generalize results without a standardized protocol and therefore makes it difficult to make meaningful comparisons with other studies that utilize different thresholds. We seldom use epidural or spinal anesthesia for our elective EVAR cases unless the patient has a comorbid condition such as severe sleep apnea or chronic obstructive pulmonary disease that prohibits the use of MAC given the risk of airway compromise. If an epidural or spinal anesthetic were to be used, intraoperative urinary catheters are always placed and it is our institution’s practice post-operatively to remove the urinary catheter after removal of the anesthetic catheter with encouragement of supervised early ambulation.

Though we did not find a direct association between length of hospital stay and type of anesthetic used, several studies have been performed demonstrating decreased length of stay favoring MAC and local anesthesia. A study using the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database from January 2005 to December 2008 evaluating 6,009 elective EVAR cases noted that only 9.3% of cases were performed under local anesthesia. This study demonstrated decreased length of stay for patients undergoing interventions under local/MAC versus GA (median 1 day versus 2 days, respectively) with an OR of 20.00 for patients under GA (95% CI 14.1–26.2).⁷ Similar findings are reported by Cheng et al who looked at elective EVARs performed in several Kaiser Permanente hospitals and noted that 14.3% were performed under MAC/local which was associated with a statistically significant decreased length of stay compared to general anesthesia (mean 35.1 hours versus 43.7 hours, respectively).²³ On the contrary, in using the VQI database from 2014 to 2017, Van Orden et al noted that only 8.77%% of elective EVARs were under local anesthesia demonstrating no difference in mean hospital length of stay compared to general anesthesia (2.3 mean days versus 2.6 days) however, they excluded open access cases and included emergent repairs.²⁴ It is worth mentioning that despite EVAR becoming the standard of care for elective abdominal aortic aneurysm repair, these studies demonstrate that the great majority continue to be performed under general anesthesia.

Specific factors with utilization of MAC anesthesia that are associated with decreased POUR in our cohort could not be specifically elucidated. Patients under MAC tended to have higher doses of propofol and lower doses of fentanyl though no significant associations were found with POUR events. A potential contributing factor could include muscular paralysis used during general anesthesia which our study did not specifically measure. Additionally we were unable to determine an association between pre-operative prostate size nor prostate volume with POUR nor was there an association with history of BPH or urinary retention. These results could likely be due to our limited sample size and potentially warrant further investigation.

Our findings suggest that the type of anesthetic used during elective EVAR can lead to increased rates of post-operative urinary retention and thus increased hospital length of stay. These findings may have implications for implementing and standardizing enhanced recovery protocols that include MAC anesthesia for patients undergoing elective EVAR. When appropriate, we have attempted to minimize the use of general anesthetic in patients undergoing elective EVAR at our institution, specifically in patients with extensive cardiac comorbidities. The use of certain anesthetic agents with MAC anesthesia is currently dependent on the patient’s comfort during the case in order to minimize patient movement while performing the intervention which subsequently decreases the operative time in addition to other factors such as radiation and contrast dosage. Our goal is to minimize these factors in addition to minimizing use of certain anesthetic agents, specifically opiates, if possible. Currently we recommend that all patients void prior to intervention and a urinary catheter be placed intraoperatively if the case is expected to take longer than 3 hours to avoid bladder overdistention. In patients undergoing general anesthesia, we recommend urinary catheter removal within 12 hours of the case and ambulation as early as possible. Patients under MAC should have urinary catheters removed at case completion with bladder scans performed routinely in the post-anesthetic care unit (PACU). Straight catheterization should be attempted if > 600 ml is noted on bladder scans to prevent bladder overdistention. Careful attention should be given to patients with CKD who still produce urine and to patients on loop diuretics as these patients may require longer times to void. We have also attempted to decrease the use of intraoperative urinary catheters in older patients, especially in octogenarians where we believe the risk of urethral injury due to larger prostate sizes could lead to retention and longer hospital stays. Possible next steps in research should involve prospective endeavors comparing different anesthetic modalities in this patient cohort as they relate to post-operative outcomes including POUR.

Limitations of our study include the inherent limitations of a retrospective review from a single university-affiliated tertiary care medical center where generalizability to all patients undergoing EVAR is limited. Given this study is not a randomized controlled trial, there is risk of selection bias, specifically regarding the selection of anesthesia in patients undergoing elective EVAR, which was based on either surgeon or anesthesiology preference, though patient demographics and medical history did not differ between the two groups. This study evaluated male patients only and therefore may not apply to female patients undergoing EVAR. There is also the risk of type II error as our study size was small which can explain why we found no significant association between historical factors associated with POUR such as age and history of urinary retention.

5.0. CONCLUSION

MAC plus local anesthesia is associated with decreased rates of POUR after elective EVAR in male patients. ERAS pathways for elective EVAR interventions should consider implementing MAC plus local anesthesia as an acceptable anesthetic option, where appropriate, in order to reduce urinary retention rates and subsequently decrease hospital length of stay in this patient cohort.

HIGHLIGHTS.

• Enhanced recovery pathways for endovascular aortic repair

• Monitored anesthetic care decreases urinary retention post endovascular aortic repair

• Urinary retention leads to higher odds of increased hospital length of stay