Total Intravenous Anesthesia Compared to Inhalational Anesthesia in Patients Undergoing Arthroscopic Rotator Cuff Repair

Background Inhalation anesthesia (IA) and total intravenous anesthesia (TIVA) are common general anesthesia techniques. During rotator cuff repair (RCR), an interscalene block is beneficial for intraoperative and early postoperative pain control. This study aimed to evaluate postoperative outcomes and opioid usage in patients undergoing arthroscopic RCR with an interscalene block and either IA or TIVA. Methodology A retrospective observational study was performed of 478 patients undergoing RCR at a single institution. Demographics, surgical details, intra and postoperative medications, and 90-day outcomes were collected. Univariate and multivariate analyses were performed to evaluate differences between groups. Results In total, 309 (64.6%) patients received IA and 169 (35.3%) received TIVA. Patients receiving IA were more likely to have comorbidities, such as diabetes (p = 0.002), sleep apnea (p = 0.006), gastroesophageal reflux disease (p < 0.001), and hypertension (p < 0.001). After adjusting for differences between groups in the multivariate analysis, patients who received TIVA had significantly shorter surgical time (β = -14.85, p < 0.001) and perioperative time (β = -21.01, p < 0.001) and significantly lower first post-anesthesia care unit Pasero opioid-induced sedation scores (β = -0.022, p = 0.040). Patients who received TIVA were less likely to receive intraoperative narcotics (odds ratio = 0.38; p = 0.031). Conclusions TIVA appears to be a safe and effective anesthetic for patients undergoing arthroscopic RCR. TIVA is a potentially beneficial alternative to IA for this patient population.


Introduction
Rotator cuff injuries are a common orthopedic condition, affecting over 25% of all people over the age of 60 and nearly 50% of those over the age of 70 [1,2].Rotator cuff repairs (RCRs) are one of the most commonly performed orthopedic procedures, with over 250,000 performed each year in the United States alone [1].RCR is a safe and effective surgical procedure that benefits individual patients and the wider society [3,4].However, RCR is often associated with significant postoperative pain and opioid use; the choice of anesthetic technique used can influence patient outcomes and is an important consideration [5,6].General anesthesia with an endotracheal tube is typically the standard technique used in RCR due to several factors, including patient positioning and facilitating hypotensive anesthesia to minimize intraoperative blood loss [6].In patients undergoing RCR, the use of peripheral nerve blocks is proven to be a beneficial adjunct to general anesthesia (GA) for both intraoperative and early postoperative pain control; interscalene blocks are one of the most effective peripheral nerve blocks in RCR [6][7][8].
The most commonly used forms of GA in these procedures are inhalational anesthesia (IA) and total intravenous anesthesia (TIVA) [9].Historically, IA has been used as the standard form of GA as it acts on GABA receptors to induce central nervous system (CNS) depression, although it may also induce myocardial and respiratory depression [9,10].The use of TIVA is becoming more common as a method of GA as it is known for its rapid anesthetic effects and comparatively quicker recovery period; TIVA has the potential for decreased opioid use, decreased nausea, and increased patient satisfaction [11,12].The most commonly used TIVA agents include propofol, ketamine, and etomidate; these agents are administered intravenously through their effect on various GABA and adrenergic receptors to quickly induce anesthesia [9,13].Within the fields of anesthesiology and orthopedic surgery, there is limited data regarding the potential differences in clinical outcomes for patients undergoing RCR based on the form of GA used [14,15].The primary objective of this study was to compare intra and early postoperative opioid use in patients undergoing RCR via IA or TIVA.Secondary objectives included comparing concomitant intra and postoperative medications used, short-term postoperative outcomes, and 90-day emergency department (ED) visits and admissions.

Study population and setting
This study was deemed exempt by the institutional review board.A retrospective observational study of patients undergoing primary arthroscopic RCR from January 2019 through February 2022 was conducted.Procedures were performed by one of nine board-certified orthopedic surgeons in a single hospital-based outpatient surgery center.Patients were excluded from the study if they underwent open RCR, revision RCR, or if they did not receive a preoperative interscalene peripheral nerve block.

Perioperative protocol
All surgeries were performed on an outpatient basis under either inhaled or total intravenous GA.All patients received an interscalene peripheral nerve block performed preoperatively.The peripheral nerve blocks were performed with consistent technique among anesthesiologists with 20 mL of ropivacaine 0.5%.The IA protocol consisted of preoperative sedation using a combination of midazolam, fentanyl, and/or propofol, followed by induction and airway management via laryngeal mask airway (LMA) or endotracheal tube.Anesthesia was maintained using inhalational anesthetics, most commonly sevoflurane.The TIVA protocol consisted of preoperative sedation using a combination of midazolam, fentanyl, and/or propofol, followed by induction and maintenance of GA through a continuous intravenous infusion of propofol and oxygen delivered typically by nasal cannula with the occasional use of LMA as appropriate.Additional medications for pain and nausea were used at the discretion of the anesthesiologist.All medication dosages were determined by the anesthesiologist as appropriate for the individual patient.All surgical procedures were performed in the beach chair position.
The Pasero opioid-induced sedation scale (POSS) is a scoring system commonly used in post-anesthesia care units (PACUs) to evaluate sedation.Patients are scored from 1 to 4, with 1 being the most alert and 4 being somnolent and unresponsive; 1 and 2 are considered acceptable scores, and 3 and 4 require intervention [16].

Data collection and analysis
Patient demographics, comorbidities, intraoperative and postoperative medications, and surgical and PACU details were manually recorded from the electronic medical record.Ninety-day hospital returns, including ED visits and hospital admissions, were obtained using the Chesapeake Regional Information System for Our Patients (CRISP) and accounted for visits to both our institution and outside regional institutions.The primary endpoints were the need for narcotics and the total milligram morphine equivalent (MME) required intraoperatively and in the PACU.Secondary endpoints included differences in nausea, pain, sedation scores, operational throughput measures, and complications resulting in 90-day hospital returns.Univariate statistics including two-sided independent and paired-sample t-tests and chi-square tests were used to evaluate differences in demographics, comorbidities, intraoperative and postoperative medication usage, short-term postoperative outcomes, and hospital returns.Multivariate analysis was performed to evaluate the effect of TIVA on outcomes that demonstrated significant differences at α ≤ 0.10 on univariate analysis after controlling for demographic and comorbidities that varied between groups in the univariate analyses.These included first PACU POSS, intraoperative MME, intraoperative narcotics, minutes in OR, perioperative minutes, postoperative MME, and postoperative narcotics.All statistical analyses were performed using R Studio (version 4.2.2, 2009-2023 RStudio, PBC).Statistical significance was assessed at p-values <0.05.

Discussion
Overall there were few, but potentially important, differences between groups, with TIVA patients experiencing similar or improved intraoperative and recovery outcomes than those receiving IA.The TIVA group demonstrated significantly shorter OR time, overall perioperative time, lower intraoperative dexamethasone administration, and lower first PACU POSS scores.The decreased OR time, perioperative time, and first PACU POSS score also remained significant in multivariate analysis after controlling for confounding variables.The percentage of patients receiving intraoperative narcotics was not significantly different in univariate analysis, but was significant upon multivariate analysis, with patients receiving TIVA being less likely to receive intraoperative narcotics.
GA with IA is the gold standard anesthesia technique for a wide variety of surgical procedures and is typically safe and effective, although it is accepted that some percentage of patients will experience side effects such as nausea, vomiting, poorly controlled postoperative pain, and postoperative cognitive impairment [6,17].Several studies from a variety of surgical specialties indicate that TIVA may improve the quality of recovery after surgery by mitigating some of the uncomfortable effects of IA [11,[17][18][19].In contrast to a study by Kim et al., there was no difference in postoperative nausea scores, pain scores, and opioids given in this study when comparing IA and TIVA [11].The one postoperative clinical outcome where TIVA patients performed better in this study was the first postoperative POSS score, although the mean POSS score in both groups was within an acceptable range; POSS scores less than 3 do not require any additional monitoring or intervention [16].The reasons for improved POSS scores are not clear, although TIVA using propofol has been shown to improve analgesia and anxiolysis [13,20,21].
Another notable finding of this study was the decreased operative time and overall perioperative time in the TIVA group.Increased operative time has been identified as an independent risk factor in postoperative complications following arthroscopic shoulder surgery, including RCR [22][23][24][25].Agarwalla et al. found that a 15-minute increase in operative time in arthroscopic RCR increased the risk of several postoperative complications, including surgical site infection, pulmonary embolism, and extended hospital stay [26].We found a decrease in operative time between groups, as patients receiving TIVA had a shorter OR time overall, although this did not result in a difference in postoperative complications in this study.Further, the decreased OR time led to a significantly shorter overall perioperative time in TIVA patients, even though the six-minute reduction in PACU time for TIVA patients did not reach statistical significance independently.
Intraoperative and postoperative analgesia are significant concerns in patients undergoing RCR, as this surgery is associated with significant postoperative pain that affects both short-term patient comfort and longer-term recovery [7].Regional anesthesia, often in the form of an interscalene nerve block, combined with GA has proven very effective at managing postoperative pain in these patients [6][7][8]15].It is not known whether the type of GA, when combined with regional anesthesia, influences postoperative pain and intra and postoperative opioid consumption.When comparing IA and TIVA in general surgery, patients receiving TIVA require fewer opioid medications in the perioperative period [27,28].Similar results have been found in orthopedic spine surgery, with patients receiving TIVA requiring fewer opioids in the postoperative period [11,29].In contrast, patients in this study received equivalent total MMEs intraoperatively and in the PACU, although patients receiving TIVA were less likely to get any opioid medication intraoperatively.We hypothesize that our results deviated from those of the general and spine surgery literature given the consistent use of regional anesthesia in this population, which resulted in relatively low use of opioid analgesia, as both groups received under 20 MMEs both intra and postoperatively, on average.
Collectively, the benefits of TIVA observed in this study, namely, decreased intraoperative narcotic utilization rates, postoperative sedation, OR time, and perioperative time, suggest this anesthetic approach may hold value in facilitating the rapid throughput of RCR patients.As arthroscopic procedures, including RCR, continue to shift to the ambulatory surgery center setting, an increased focus on factors influencing perioperative throughput has occurred [30][31][32][33][34].While many factors that have been shown to influence throughput time, such as technical aspects of the repair and concomitant procedures performed, sex, and comorbidities, are not modifiable, anesthetic selection is within the control of the care team [32,35,36].The overall perioperative time savings of approximately 21 minutes per patient observed in this study is not insignificant and could allow for the performance of additional cases during the operative day.Given this benefit and the similar complication rates observed in comparison to IA, TIVA appears to be an attractive anesthetic option, particularly in the ambulatory care setting.
Despite the potential benefits of TIVA, not all patients are appropriate candidates.Anesthetic underdosing resulting in awareness during anesthesia is more likely to occur using TIVA techniques, and patients with a history of anesthetic resistance are at increased risk for awareness [37][38][39].Further, caution should be taken when utilizing TIVA in morbidly obese patients.Although commonly practiced, appropriately dosing TIVA agents to avoid intraoperative awareness is more challenging in this population, as excess fat may result in increased diffusion of highly lipid-soluble propofol from the plasma [40].While the results of the current study highlight the potential benefits of TIVA, the ultimate decision for anesthetic technique must continue to be made on a patient-specific basis, based on the clinical judgment of the anesthesiologist.
This study does have several limitations.First, the study was performed at a single institution and therefore the findings may not be generalizable to the wider patient population.It is also a retrospective study and subject to an inherent selection bias, although we were able to utilize multivariate analysis to control for potentially confounding factors.However, it is possible that unmeasured confounders influenced our results, including surgeon and anesthesiologist variability.A strength of the study is the consistent use of interscalene peripheral nerve blocks, thus eliminating a potentially significant confounding factor.Finally, this study is limited in its evaluation of only intraoperative and immediate postoperative outcomes.Further study is required to evaluate whether TIVA administration is associated with differences in post-discharge pain and narcotic utilization and patient perception of the operative experience.Based on the positive results presented, we suggest a prospective randomized evaluation is warranted to generate high-quality evidence evaluating the effects of TIVA on RCR outcomes.

Conclusions
In comparison to IA, TIVA appears to be a safe and effective anesthetic for patients undergoing arthroscopic RCR with regional nerve blocks.Given the observed benefits, including reduced intraoperative narcotic utilization, postoperative sedation, and throughput time, we suggest TIVA be considered as a potentially beneficial alternative to IA for this patient population.There was no significant difference between anesthesia types in postoperative nausea, postoperative pain scores, or complications which suggests at minimum TIVA is equivalent to IA.