Rectus sheath block with sedation versus general anesthesia for periumbilical hernia repair: A retrospective propensity score-matched study

Abstract

BACKGROUND

Umbilical hernia repair is a surgical procedure commonly performed under general anesthesia (GA). While safe, GA is associated with prolonged post-anesthesia care unit (PACU) recovery due to side effects including drowsiness and postoperative nausea and vomiting. Prolonged PACU stays can contribute to surgical delays, decreased patient satisfaction, and increased operating room costs. Ultrasound-guided rectus sheath block (RSB) with sedation is an alternative that may provide targeted analgesia and minimize opioid use. However, its impact on PACU recovery time compared to GA remains unclear. This study hypothesizes a shorter PACU recovery time for the regional anesthesia group compared to GA.

AIM

To compare PACU recovery time between RSB with sedation and GA for umbilical hernia repair.

METHODS

This single-center retrospective cohort study included adult patients undergoing umbilical hernia repair with either RSB and sedation or GA at Victoria Hospital, London Health Science Centre between January 1, 2017, and July 31, 2024. Multivariable logistic regression was used to model anesthetic modality as a function of age, sex, body mass index, and American Society of Anesthesiologists to generate a propensity score for each patient. The primary outcome was post-anesthetic recovery time. Secondary outcomes included perioperative times, postoperative pain, opioid use, nausea/vomiting, and desaturation events.

RESULTS

After screening 182 patients, 62 patients in each group were propensity matched. RSB with sedation was associated with 17 minutes (95%CI: -25.0 to -9.0) shorter post-anesthetic recovery time compared to GA. Likewise, there was a reduction in operation time by 5 minutes (95%CI: -9.0 to -0.10), and total length of stay by 30 minutes (95%CI: -47.0 to -11.0). While no differences were noted in the incidence of nausea/vomiting, severe pain episodes, and desaturation, the RSB group was correlated with a lower first pain score in PACU and opioid use.

CONCLUSION

RSB with sedation was associated with shorter recovery time, intraoperative time, and hospital length of stay. Future studies should focus on demonstrating a causal relationship between RSB and recovery time.

Key Words: Umbilical hernia repair; Regional anesthesia; Rectus sheath block; Post-anesthetic recovery period; Ambulatory surgery

Core Tip: While rectus sheath block (RSB) is a promising alternative to general anesthesia (GA) for umbilical hernia repair, there is a lack of evidence on its impact on hospital length of stay. This single-center retrospective cohort study included adult patients between 2017 to 2024, and measured the post-anesthetic recovery time between RSB and GA. After screening 182 patients, 62 patients were propensity matched. Compared to GA, RSB with sedation was associated with a reduction in post-anesthetic recovery time, operation time, and total length of hospital stay.

INTRODUCTION

It is estimated that more than 20 million hernia repairs are performed worldwide annually, making it one of the most common general surgery procedures[1]. In Canada, the aging population, increasing medical complexity of patients, and shortages of healthcare workers combined with the lasting ripple effects of the coronavirus disease 2019 pandemic, continues to contribute to surgical backlogs and prolonged wait times[2]. In turn, these challenges warrant the development of innovations and strategies that improve perioperative efficiency and patient recovery.

A potential area of focus is improving care in the post-anesthesia care unit (PACU), which is defined as the time period between the patient leaving the operating room and their clearance for hospital discharge. Umbilical hernias repairs are most often done under general anesthesia (GA). While safe, GA is associated with potential postoperative nausea/vomiting and drowsiness, which can contribute to prolonged stays in PACU[3-6]. Prolonged PACU stays can contribute to surgical delays, decreased patient satisfaction, and increased OR cost[3,6,7].

An alternative anesthetic for umbilical hernia repair could include local anesthesia infiltration by the surgeon or nerve blocks, like the ultrasound guided rectus sheath block (RSB). RSBs have been used in the pediatric population with GA for post-operative analgesia[8-10]. The targeted and improved analgesia provided by RSB can minimize opioid consumption[8,10,11], which is a known contributor to postoperative nausea and vomiting, and improve PACU recovery[5,6,12,13].

While RSB has been shown to be a feasible option for the primary anesthetic for umbilical hernia repairs in patients who need to avoid GA for medical comorbidities[11,12,14-16], the impact on hospital length of stay for RSB compared to GA for umbilical hernia repair has not been reported in the literature. Therefore, the aim of this retrospective review is to compare the PACU recovery time between patients receiving RSB with sedation as a primary anesthetic technique to those undergoing GA for umbilical hernia repair. Secondarily, this study seeks to investigate the impact of the anesthetic choice on postoperative pain and overall hospital flow including pre-operative time, operative (OR) time, and hospital length of stay. It is hypothesized that there will be a reduction in PACU recovery time in the regional anesthesia group.

MATERIALS AND METHODS

Study design and setting

This is a single-center, retrospective, propensity-score matched cohort study, approved by the Western University Health Sciences Research Ethics Board (No. 125656). Data was extracted from the institution’s electronic medical record at Victoria Hospital, London Health Sciences Centre by the co-investigators.

Study population

Patients who underwent umbilical hernia repair with either RSB and sedation or GA at Victoria Hospital, London Health Sciences Centre between January 1, 2017, and July 31, 2024, were screened for eligibility. Given the retrospective nature of the study, all the patients undergoing umbilical hernia repair during this period were included for screening and those who met the eligibility criteria were included in the analysis.

The inclusion criteria involved: (1) Patients aged 18 years or older with an American Society of Anesthesiologists (ASA) physical status of I-III; and (2) Those undergoing an elective one-day repair of midline periumbilical hernia (ventral and umbilical hernia) measuring up to 4 cm in size with or without mesh (BD Ventralex^TM).

Patients younger than 18 years of age, with a family history of malignant hyperthermia, and opioid dependence defined as a daily opioid requirement greater than 30 mg of oral morphine equivalents were excluded. Exclusion criteria also included patients who required admission post-operatively, underwent combined surgery (e.g., repair of two hernias or hernias plus lesion excision) or had procedures performed exclusively under local anesthesia. Patients with missing data were excluded from the study.

Recorded comorbidities included cardiac disease (coronary artery disease, congestive heart failure, and moderate to severe valvular disease), chronic obstructive pulmonary disease, diabetes, and obstructive sleep apnea.

Anesthetic techniques

All patients were offered the option of RSB with sedation, when the available anesthesiologist on the day of surgery was able to perform an RSB. If patients declined the technique or there was no capability of offering RSB, the patient would receive GA.

In the RSB group, patients would receive their block in the block room 30 minutes preoperatively. A pre-procedural pause was completed before all blocks. Patients were offered midazolam at the start of the procedure. If given, the dose of midazolam was between 1 mg and 2 mg. The RSB was performed under ultrasound guidance, at the level of hernia, using 15 mL of 0.5% ropivacaine on each side[16]. During the surgery, patients would receive propofol sedation. Propofol was administered at a rate between 50 μg/kg/minute and 150 μg/kg/minute and titrated at the discretion of the attending anesthesiologist to achieve the desired effect of a Ramsay sedation scale level of 4 or 5. Sedation level was assessed every 15 minutes.

For the GA group, patients would be induced with fentanyl 1-2 μg/kg and propofol 2-3 mg/kg. Anesthesia was then maintained with sevoflurane in air and oxygen mixture. The airway was managed with a laryngeal mask airway (LMA) without paralysis, unless there was a contraindicated to LMA such as reflux disease. At the end of the case, the surgeon would provide local infiltration with 20 mL of 0.5% bupivacaine.

Both groups received, on induction, intravenous dexamethasone and on skin closure, ketorolac and ondansetron. Most patients also received intravenous morphine equivalents of 2-5 mg, given at the end of the case at the discretion of the attending anesthesiologists.

LMA was removed prior to exiting the OR. For both groups, patients bypassed phase I recovery if, once emerging from anesthesia, they scored at least a 9 out of 10 on the modified Aldrete score. In phase II recovery, patients were eligible for discharge from hospital once they scored at least 9 out of 10 on the Post Anesthetic Discharge Scoring System.

Outcomes

Our primary outcome was post-anesthetic recovery time. This was defined as the time of arrival from the OR to either phase I or phase II recovery until the time the patient was ready for discharge from hospital.

Secondarily, the entire perioperative time was evaluated. The overall length of stay in hospital started preoperatively with arrival in surgical day care, included time spent in the OR, and ended once the patient was discharged from hospital. Additional secondary outcomes included PACU events such as postoperative pain, nausea and vomiting, and desaturation episodes. Postoperative pain was determined through several measures including the patient’s first recorded pain score in recovery, any documented episodes of significant pain through their stay which was defined as a pain score numeric rating above 6 out of 10, and through overall opioid consumption in oral morphine equivalents from time of arrival to PACU until hospital discharge. This pain score was the threshold to give postoperative opioids by our local practice. Postoperative nausea and vomiting episodes were considered to have occurred if nausea or vomiting was documented in the record, or if antiemetics were provided in PACU. A postoperative desaturation episode was defined as a charted oxygen saturation below 90% with or without supplemental oxygen.

Statistical analysis

Statistical analyses were performed using R (version 4.5.1) and were defined a priori. Patient characteristics, including age, sex, body mass index (BMI), ASA status, and comorbidities, were summarized using means and standard deviations or counts and percentages, as appropriate. An intention-to-treat approach was used. Propensity score matching was applied to address potential confounding. A multivariable logistic regression model was used to estimate the probability of receiving general vs regional anesthesia based on age, sex, BMI, and ASA status. Age and BMI were treated as continuous variables, while sex (male vs female) and ASA class (I, II, or III) were treated as categorical variables. Matching was performed using a nearest-neighbor approach with 1:1 matching, no replacement, and a caliper of 0.2 of the logit. Patients who could not be matched were excluded from post-matching analyses. Covariate balance before and after matching was assessed using standardized differences in means, with values below 0.10 considered acceptable. The Shapiro-Wilk test was used to assess normality of continuous outcomes. Normally distributed continuous variables were analyzed with Student’s t-test and non-parametric continuous variables were analyzed with Mann-Whitney U test. Fischer’s exact tests were done for categorical outcomes. Haldane-Anscombe corrections were applied for zero event outcomes. E-values were calculated to estimate the effect of unmeasured confounders on statistically significant post-matched outcomes. The statistical methods of this study were performed and reviewed by Lin C, the principal investigator, from the Department of Anesthesiology, Western University.

RESULTS

During the study period, 120 patients and 62 patients underwent GA and RSB for umbilical or ventral hernia repair. A total of 62 patients in each group were ultimately matched. No patients in the RSB group required conversion to GA. A total of 16 patients and 20 patients had mesh repair in the GA and RSB group respectively.

Demographic information pre-match and post-match were presented in Table 1. There were higher proportion of female patients, which is expected with this type of hernia. The GA group had higher incidence of comorbidities. The standardized mean differences of all co-variates were below 10%.

Table 1 Pre-match and post-match demographics, n (%)/mean ± SD.

	Pre-match		Post-match
	GA	SD1	GA	Regional anesthesia	SD1
Sample size	120		62	62
Mean age (years)	49.3 ± 14.1	0.15	52.6 ± 14.2	51.3 ± 13.7	0.09
Sex2
Male	73 (60.8)	0.04	40 (64.5)	40 (64.5)	0
Mean body mass index (kg/m²)	28.9 ± 4.8	0.03	28.9 ± 4.1	28.7 ± 5.1	0.03
American Society of Anesthesiologists
I	31 (25.8)	0.02	12 (19.4)	17 (27.4)	0.08
II	77 (64.2)	0.06	41 (66.1)	36 (58.1)	0.08
III	12 (10.0)	0.04	9 (14.5)	9 (14.5)	0
Comorbidities
Cardiac disease3	5 (4.2)		4 (6.5)	1 (1.6)
Chronic obstructive pulmonary disease	3 (2.5)		2 (3.2)	1 (1.6)
Sleep apnea4	2 (1.7)		1 (1.6)	1 (1.6)
Diabetes5	7 (5.8)		7 (11.3)	5 (8.1)

¹SD presented in proportion.

²Biologic sex.

³Include coronary disease, moderate to severe valvulopathy, heart failure.

⁴Obstructive sleep apnea.

⁵Include type 1 and 2 diabetes.

GA: General anesthesia.

All continuous outcomes were not normally distributed and were analyzed with Mann-Whitney U test. The outcomes were presented in Table 2. The average total post anesthetic length of stay for all patients was 113 minutes. For post-anesthetic recovery time, there was a modest reduction of 17 minutes (95%CI: -25.0 to -9.0; P < 0.001) associated with the RSB group. E value for this result was 3.2, which was moderately robust against unaccounted biases. Likewise, there was reduction in the operating time and total length of stay by 5 minutes (95%CI: -9.0 to -0.10; P = 0.010) and 30 minutes (95%CI: -47.0 to -11.0; P = 0.002) respectively, but not preoperative time. Odds ratio for bypassing phase I recovery in the RSB group was 18.6 (95%CI: 6.83-58.02; P < 0.001) This was correlated with lower first pain scores in PACU and lower opioid use in the RSB group. However, no difference in incidence of nausea/vomiting, severe pain episodes and desaturation were detected.

Table 2 Pre-match and post-match measured outcomes, n (%).

Outcome	Pre-match1	Post-match2		Mean difference (95%CI)	P value	E value
	GA (n = 120)	GA (n = 62)	Regional anesthesia (n = 62)
Total recovery time in minutes, mean (IQR)	81 (27.2)	82 (30)	65 (28)	-17.0 (-25.0 to -9.0)	< 0.001	3.2
Preoperative time in minutes, mean (IQR)	89.5 (53.2)	91.5 (54.5)	89 (67)	-2.0 (-15.0 to 13.0)	0.81
Operating time in minutes, mean (IQR)	45 (17)	45 (13.8)	41 (11.8)	-5.0 (-9.0 to -1.0)	0.010	2.4
Total length of stay in minutes, mean (IQR)	224 (82.8)	229 (79)	198 (63.8)	-30.0 (-47.0 to -11.0)	0.002	2.7
First postoperative pain score, mean (IQR)	3 (3.25)	3 (3.75)	0 (3)	-2.0 (-2.0 to -1.0)	0.001	2.9
Postoperative opioid use in oral morphine equivalents, mean (IQR)	10 (10)	10 (10)	0 (10)	0.00 (-10.0 to 0.00)	0.001	2.8
Adverse events
Nausea or vomiting	8 (6.7)	5 (8.1)	5 (8.1)	1 (0.22-4.60)	1
Severe pain	14 (11.7)	7 (11.3)	5 (8.1)	0.69 (0.16-2.70)	0.76
Desaturation	10 (8.3)	6 (9.7)	0 (0)	0.07 (0.004-1.26)	0.07
Phase I recovery bypass	13 (10.8)	7 (11.3)	44 (71.0)	18.6 (6.83-58.02)	0.001	36.7

¹Outcomes under pre-matched samples were unadjusted; presented as median and interquartile range, or number and percentage.

²Outcomes from matched samples were adjusted from a multivariable regression model with age, body mass index, sex, American Society of Anesthesiologists, cardiac disease, chronic obstructive pulmonary disease, diabetes, and obstructive sleep apnea as covariates, presented as median and interquartile range, or number and percentage.

P value compared post-matched samples between general anesthesia and regional anesthesia groups, using P < 0.05 as statistically significant. Mann-Whitney U test was used for continuous outcomes and Fisher’s exact test for categorical outcome. GA: General anesthesia; IQR: Interquartile range.

DISCUSSION

To date, there have been no comparative studies investigating the benefit of using RSB as a primary anesthetic technique vs GA for umbilical hernia. Manassero et al[12] provided RSB for umbilical hernia repair as the sole anesthetic technique in a 30-patient case series. The technique was deemed effective if additional local anesthetic infiltration was not required, which occurred in 75% of the patients, and no patients required conversion to GA. Several single-case reports[15,16] demonstrated successful use of RSB without GA in high-risk patients for umbilical hernia or colostomy revision. Wang et al[11] conducted a randomized trial comparing RSB vs local infiltration, both with and without GA, in patients with cirrhosis. No patients required conversion to GA in this study either, and a lower dose of rescue analgesia and incidence of nausea and vomiting were noted in the RSB group.

RSB has traditionally been utilized as a component of multimodal analgesia as it lacks visceral coverage[17]. Moderate sedation is therefore essential to ensure success when using RSB as the primary anesthesia. Further local infiltration, gentle retraction and manipulation of the peritoneum and bowel components by the surgeons may also help to reduce the depth of sedation. To provide coverage for T9 to T11 dermatomes, RSB has also been commonly placed at the level of umbilicus. In our practice, RSB was placed at the level of hernia. We believe this, in theory, provided higher dose of local anesthetic sooner to the corresponding level of the anterior cutaneous branch of the intercostal nerve.

The benefits of utilizing regional anesthesia over GA included shorter hospital length of stay, better pain control and a lower risk of nausea and vomiting. Our retrospective study also demonstrated a correlation between RSB and shorter recovery time, phase I bypass. The shorter recovery time and phase I bypass are likely a result of avoidance of volatile anesthetic, therefore reducing the risk of postoperative nausea and vomiting and drowsiness. The decrease in recovery time was 17 minutes in our study, about 20% reduction of the median recovery time of 82 minutes. The cumulative effect of decreasing recovery length of stay by a modest amount for each patient can make a difference over the course of several patients in a day and possibly allow for scheduling of additional patients. It is also helpful for overall perioperative flow to have patients bypass phase I recovery. In our center, one limiting factor to the arrival of patients to PACU is the number of nurses available to recover patients. If patients are able to move straight to phase II recovery this can open up staffing for PACU recovery of other patients, minimizing PACU holds. Similarly, RSB has been shown to provide good analgesia for umbilical hernia[10,14] with patients demonstrating a lower first PACU pain score and less opioid use in the postoperative period.

There are important limitations in our study. As a retrospective study, a causal relationship could not be established. Despite using propensity score to balance the confounders, it was not possible to account for all variables that may lead to differences in patient recovery. The E values for statistically significant data range from 2.4 to 3.2, suggesting moderate robustness against confounders. On the other hand, E value for phase I bypass was 36.7, which was strong against biases. Another important bias was the issue with preemptive analgesia. In the RSB group the patients receive an analgesic technique upfront whereas the GA group receive it in forms of opioids and local anesthetic infiltration at the end. However, the pain outcome was modestly better in the RSB group with 2-point reduction in VAS. Notably different local anesthetics were used in the RSB and GA groups with important difference the duration. There may be difference in unmeasured post-hospital-discharge outcomes. As a tertiary hospital and single center study, factors such as the presence of learners, a block room, anesthetic and surgical techniques may not be applicable for other locations.

CONCLUSION

RSB with sedation is an attractive anesthetic technique for umbilical hernia. RSB was associated with shorter recovery time in addition to shorter intraoperative time and hospital length of stay and correlated with a modest reduction of first pain score and opioid use. Our study demonstrated RSB to be a feasible anesthetic technique with modest improvement in outcomes. Future studies should focus on demonstrating a causal relationship between RSB and shortened recovery time.