Share this article on:

A Comparison of Psoas Compartment Block and Spinal and General Anesthesia for Outpatient Knee Arthroscopy

Jankowski, Christopher J. MD*; Hebl, James R. MD*; Stuart, Michael J. MD†; Rock, Michael G. MD†; Pagnano, Mark W. MD†; Beighley, Christopher M. MS‡; Schroeder, Darrell R. MS‡; Horlocker, Terese T. MD*

doi: 10.1213/01.ANE.0000081798.89853.E7

The optimal anesthetic technique for outpatient knee arthroscopy remains controversial. In this study, we evaluated surgical operating conditions, patient satisfaction, recovery times, and postoperative analgesic requirements associated with psoas compartment block, general anesthetic, or spinal anesthetic techniques. Sixty patients were randomized to receive a propofol/nitrous oxide/fentanyl general anesthetic, spinal anesthesia with 6 mg of bupivacaine and 15 μg of fentanyl, or psoas compartment block with 40 mL of 1.5% mepivacaine. All patients received IV ketorolac and intraarticular bupivacaine. The frequency of postanesthesia recovery room admission was 13 (65%) of 20 for patients receiving general anesthesia, compared with 0 of 21 for patients receiving spinal anesthesia and 1 (5%) of 19 for patients receiving psoas block (P < 0.001). The median time from the end of surgery to meeting hospital discharge criteria did not differ across groups (131, 129, and 110 min for general, spinal, and psoas groups, respectively). In the hospital, 45% of general anesthesia patients received opioid analgesics, compared with 14% of spinal anesthesia and 21% of psoas block patients (P = 0.087). There was no difference among groups with respect to the time of first analgesic use or the number of patients requiring opioid analgesia. Pain scores were highest in patients receiving general anesthesia at 30 min (P = 0.032) and at 60, 90, and 120 min (P < 0.001). Patient satisfaction with anesthetic technique (P = 0.025) and pain management (P = 0.009) differed significantly across groups; patients receiving general anesthesia reported lower satisfaction ratings. We conclude that spinal anesthesia or psoas block is superior to general anesthesia for knee arthroscopy when considering resource utilization, patient satisfaction, and postoperative analgesic management.

IMPLICATIONS: Outpatient knee arthroscopy may be performed using a variety of anesthetic techniques. We report that spinal anesthesia and psoas compartment block are superior to general anesthesia when considering resource utilization, patient satisfaction, and postoperative analgesic management.

Departments of *Anesthesiology,

†Orthopedic Surgery, and

‡Biostatistics, Mayo Clinic, Rochester, Minnesota

Accepted for publication May 28, 2003.

Address correspondence and reprint requests to Terese T. Horlocker, MD, Department of Anesthesiology, Mayo Clinic, 200 First St. Southwest, Rochester, MN 55905. Address e-mail to

Knee arthroscopy is one of the most commonly performed orthopedic procedures in the United States. Various types of anesthesia, including local, neuraxial blockade, and general, have been successfully used (1–6). Peripheral techniques, including femoral and lateral femoral cutaneous block (with or without sciatic block), have been evaluated more recently as possible alternatives (7–10). Potential advantages of peripheral nerve blockade include less nausea/emesis and urinary retention, earlier ambulation, and improved postoperative analgesia, resulting in earlier hospital dismissal. No studies have compared peripheral blocks with current general or small-dose spinal anesthetic techniques.

Femoral nerve block is the most common peripheral regional technique used to provide anesthesia for knee arthroscopy. Although femoral block is easy to perform, the obturator and lateral femoral cutaneous nerves are not consistently blocked with this technique (11). Block failure, requiring a general anesthetic, may occur in 12%–15% of patients (7,8). Psoas compartment block, first described in 1974, represents a posterior approach to the nerves supplied by the lumbar plexus (12). Although initially described by using a loss-of-resistance technique, recent advancements in nerve stimulator and insulated needle technology have facilitated the identification of the lumbar plexus and increased the popularity of the posterior approach to the lumbar plexus. In contrast to the femoral approach, the psoas compartment block reliably produces blockade of the entire lumbar plexus, including the lateral femoral cutaneous and obturator nerves (11,13). Thus, psoas block may be suitable for outpatient knee arthroscopy. This study compared surgical operative conditions, analgesic requirements, postoperative recovery and discharge times, patient satisfaction, and side effects associated with three standardized anesthetic methods (psoas compartment block, spinal anesthesia, and general anesthesia), each using “optimal” medications and techniques.

Back to Top | Article Outline


After IRB approval, 60 patients undergoing elective outpatient knee arthroscopy were prospectively randomized to 1 of 3 standardized anesthetic techniques. All patients were between 18 and 60 yr old with an ASA physical status of I–II. Exclusion criteria included surgical procedures with ligamentous repair or reconstruction, morbid obesity (calculated body mass index >35 kg/m2), or a medical contraindication to any of the anesthetic techniques (allergy, neurologic condition, localized infection, bleeding disorder, gastroesophageal reflux, and impaired hepatic or renal function). Patients with a history of substance abuse or protracted nausea/vomiting after general anesthesia were also excluded. After obtaining written, informed consent, patients were randomized to receive psoas compartment block, spinal anesthesia, or general anesthesia. The surgeon was blinded to the regional anesthetic technique.

Patients did not receive premedication before arrival in the operating room. Standard monitors included continuous electrocardiogram and pulse oximetry. Noninvasive blood pressure measurements were performed at 5-min intervals. An 18- or 20-gauge IV cannula was placed, and an infusion of lactated Ringer’s solution was established. General anesthesia was induced with propofol 2 mg/kg IV and fentanyl 2–3 μg/kg IV and was maintained with a propofol infusion (75–200 μg · kg−1 · min−1) and 60% nitrous oxide by laryngeal mask airway. The propofol infusion was adjusted, and additional 1 μg/kg IV fentanyl doses were administered to maintain anesthetic depth and/or heart rate and blood pressure within 20% of baseline. The propofol infusion was discontinued 10 min before surgical closure. Ondansetron 4 mg IV was administered 30 min before surgical completion.

Patients undergoing psoas compartment block or spinal anesthesia received 0.03 mg/kg of IV midazolam during positioning for the regional technique. Psoas compartment block was performed with the patient in the lateral decubitus position, with the operative side dependent. The lumbar plexus was identified with a 10-cm, 21-gauge stimulating needle by using the approach described by Winnie et al. (12). On elicitation of a quadriceps response at <0.75 mA, 40 mL of 1.5% mepivacaine with 1:200,000 epinephrine was injected. Successful block was confirmed by documentation of decreased sensation in one or more terminal nerves of the lumbar plexus (femoral, lateral femoral cutaneous, or obturator) and quadriceps weakness. A single dose of 2–3 μg/kg of IV fentanyl was administered before surgical incision.

Spinal anesthesia was performed with the patient in the lateral decubitus position, with the operative side nondependent. A 27-gauge Whitacre needle was inserted at the L2-3, L3-4, or L4-5 interspace until free flow of cerebrospinal fluid was obtained. Six milligrams (0.8 mL) of 0.75% isobaric bupivacaine with 15 μg of fentanyl was injected. For both spinal and psoas blockade, intraoperative sedation was maintained with a propofol infusion (10–50 μg · kg−1 · min−1). If more than 50 μg · kg−1 · min−1 of propofol was required to maintain patient comfort, the regional technique was considered a failure, and general anesthesia was induced. All regional techniques were performed in the operating suite. Block time was measured for each patient and was defined as the time required to place the patient into the lateral decubitus position, identify landmarks, perform the technique, and inject the local anesthetic.

All patients received 15–30 mg of IV ketorolac before incision and again at the time of surgical closure. In addition, the surgeon infiltrated each arthroscopic portal site with 2–3 mL of 0.25% bupivacaine before surgery and on completion of the procedure injected 20 mL of 0.25% bupivacaine intraarticularly. The surgeon was asked to assess surgical operating conditions as either “satisfactory” or “unsatisfactory.”

At the time of surgical closure, patients were evaluated with the Mayo Modified Discharge Scoring System (Appendix 1), an accepted modification of the Aldrete postoperative discharge criteria to assess readiness for postanesthesia care unit (PACU) discharge (14). Discharge criteria are based on motor activity, respiratory status (including oxygen saturation), hemodynamic stability, and level of consciousness. A Mayo Modified Discharge Scoring System score ≥8 signified readiness for PACU discharge. Patients were assessed every 10 min until discharge criteria were satisfied. If patients met criteria for PACU discharge at the time of surgical completion, they bypassed the PACU and were fast-tracked to the ambulatory surgical unit. Pain was assessed with a verbal analog scale (VAS) every 10 min until hospital discharge. In the PACU, pain was treated incrementally with 25 μg of IV fentanyl until the patient was comfortable. Postoperative nausea and vomiting were treated with ondansetron 4 mg IV.

Patients were admitted to the ambulatory surgical unit when PACU discharge criteria were met. Post-Anesthesia Discharge Scoring System scores were determined every 20 min until hospital discharge criteria were met (15). Assessment included mental status, stable vital signs, adequate analgesia, oral intake, absence of nausea, ability to ambulate (with crutches), and voiding (Appendix 2). Mild pain (VAS <5) was managed with acetaminophen 650 mg and propoxyphene 100 mg; moderate pain (VAS ≥5) was treated with acetaminophen 650 mg and oxycodone 10 mg. A Post-Anesthesia Discharge Scoring System score ≥9 was required for discharge.

Discharge time (defined as the time from completion of the surgical procedure until the patient met all discharge criteria), opioid requirements, and side effects such as nausea or pruritus requiring treatment were recorded. Patients were contacted by telephone at 1 and 7 days after surgery to evaluate patient satisfaction (1–10 scale), anesthesia-related complications, and postdismissal analgesic requirements.

Baseline patient and procedural characteristics, procedural outcomes, and postoperative pain outcomes were compared across general anesthesia, spinal anesthesia, and psoas compartment block groups simultaneously by using the Kruskal-Wallis test for continuous variables and Fisher’s exact test for categorical variables. Postoperative pain was compared at 30, 60, 90, and 120 min after surgery and was evaluated with a carry-forward method. Thus, for patients who were discharged before a given assessment period, the final pain score before discharge was used. P values ≤0.05 were considered statistically significant.

Back to Top | Article Outline


Sixty patients were enrolled in the study; 20 patients received general anesthesia, 21 patients received spinal anesthesia, and 19 patients received a psoas compartment block. There were no differences in age, height, weight, ASA physical status, operating room time, or intraoperative fentanyl administration among groups (Table 1). Block time did not differ between spinal and psoas groups. Operative conditions were satisfactory in all cases. One patient in the spinal group required general anesthesia because of failure of the regional technique. In addition, there was one protocol violation within the psoas group. The nonoperative extremity was initially blocked. When this was discovered, a second block was performed on the operative side, resulting in a total volume of 60 mL of local anesthetic. The patient subsequently reported a bilateral rapidly increasing sensory/motor block. General anesthesia was induced and the airway secured. The patient remained hemodynamically stable and was tracheally extubated after surgery. There were no long-term sequelae. However, persistent neuraxial blockade prolonged both PACU (233 min) and ambulatory surgical unit (172 min) times. In both cases, analyses were performed with an intention-to-treat approach whereby patients were included in their randomized groups.

Procedural outcomes are summarized in Table 2. The frequency of PACU admission was 13 (65%) of 20 for patients receiving general anesthesia, compared with 0 of 21 for patients receiving spinal anesthesia and 1 (5%) of 19 for patients receiving psoas block (P < 0.001). Time in the ambulatory surgical unit differed significantly (P = 0.011) across groups and was less in the general anesthesia group compared with the spinal (P = 0.006) and psoas (P = 0.018) block groups. However, the median time from the end of surgery to meeting hospital discharge criteria did not differ significantly across groups (131, 129, and 110 min for general, spinal, and psoas groups, respectively; P = 0.673).

In the hospital, 45% of general anesthesia patients received analgesics, compared with 14% of spinal anesthesia patients and 21% of psoas block patients (P = 0.087). There was no significant difference among groups with respect to the time of first analgesic use (P = 0.121) or the number of patients requiring opioid analgesia use in the first 24 h. No patient required antiemetic therapy in the PACU; one patient in the general and one patient in the spinal anesthesia group required an antiemetic in the ambulatory surgery unit (P = not significant).

Postoperative pain was compared at 30, 60, 90, and 120 min after surgery (Table 3). Pain scores differed significantly among groups at 30 min (P = 0.032) and at 60, 90, and 120 min (P < 0.001); the general anesthesia group reported higher pain scores. Median pain scores for spinal anesthesia and psoas block patients were 0.0 for all time intervals. There were no differences in the postoperative VAS pain scores between the spinal and psoas block groups.

All anesthetic techniques were associated with frequent patient satisfaction; more than 90% of patients reported satisfaction scores of ≥8 (1–10 scale) within each group. However, patient satisfaction with anesthetic technique differed significantly across groups (P = 0.025): patients receiving general anesthesia reported less satisfaction than those receiving spinal anesthesia (P = 0.010) or psoas block (P = 0.132). In addition, patient satisfaction with pain management differed across groups (P = 0.009) and was significantly higher with either regional technique compared with general anesthesia (spinal versus general, P = 0.012; psoas versus general, P = 0.026) (Table 2).

Back to Top | Article Outline


Outpatient knee arthroscopy may be performed under local, general, or regional anesthesia. Innovations in anesthetic equipment and medications continue to refine general and neuraxial anesthetic management. For example, the addition of intrathecal fentanyl to small-dose bupivacaine or lidocaine improves spinal anesthesia without prolonging recovery (2,5), whereas a propofol/nitrous oxide general anesthetic provides rapid recovery with small risk of postoperative nausea and vomiting (3). Peripheral regional techniques may also be suitable anesthetic alternatives for outpatient knee arthroscopy, given the infrequency of nausea/vomiting and urinary retention and the prolonged postoperative analgesia. However, there have been no previous investigations comparing the efficacy and perioperative outcomes of current general and neuraxial techniques with those of peripheral nerve blockade.

Several studies have evaluated the psoas approach in patients undergoing major orthopedic hip surgery (16–19) or knee procedures in combination with sciatic nerve blockade (20–22). However, no investigation has described its utility as the sole anesthetic for outpatient knee arthroscopy. We included the psoas compartment block, rather than femoral block, as our peripheral technique because of the more reliable blockade of the complete lumbar plexus associated with the posterior psoas approach (11,13). Previous studies have noted a 12%–15% failure rate of femoral nerve blockade in patients undergoing knee arthroscopy (7,8). In contrast, the psoas compartment approach in this investigation was associated with a 100% success rate and high patient satisfaction. Because complete anesthesia of the knee joint involves blockade of both the lumbar and sacral plexuses, the improved efficacy of the psoas technique may be due to more reliable lumbar plexus blockade, as well as partial epidural or sacral block, which occurs in 10%–25% of patients (17,22). Epidural spread may have been responsible for the high sensory and motor block in the patient who underwent bilateral psoas block with 60 mL of local anesthetic. Alternative etiologies include subdural, dural sleeve, or subarachnoid injection. No other patient had clinical evidence of sciatic or bilateral block at the time of surgical completion. In our experience, performance of the psoas compartment block with the operative side dependent and injection of 30–40 mL of local anesthetic solution minimizes the risk of significant central spread (20).

Adequate pain management is essential to facilitate rehabilitation after knee arthroscopy. Regional techniques provided superior analgesia at 30, 60, 90, and 120 minutes after surgery (the approximate time of hospital dismissal) when compared with general anesthesia. However, there were no significant differences in opioid use during the study period. Although the psoas compartment block theoretically would provide extended postoperative analgesia and decrease opioid requirements, some investigators have postulated that a multimodal analgesic approach such as that used within this study (short-acting opioids, local anesthetics, and nonsteroidal antiinflammatory drugs) maintains VAS pain scores <3 (1–10 scale) regardless of intraoperative anesthetic technique (4,6,23). This suggests there may be minimal discomfort associated with standard outpatient knee arthroscopy. Our results support these findings, with median VAS pain scores <2 in all groups after surgery. Therefore, although many clinicians advocate that peripheral nerve blockade is essential to provide prolonged postoperative analgesia, it may not be necessary in this patient population.

Anesthetic technique may influence resource utilization and institutional costs in the immediate postoperative period. We report that a significantly larger percentage of patients undergoing general anesthesia required PACU admission (65%) when compared with both the spinal (0%) and psoas (5%) techniques. The larger PACU admission rates resulted in an increased utilization of nursing resources and associated costs for general anesthesia. However, despite the increased frequency of PACU admission among patients undergoing general anesthesia, overall times from surgical completion to hospital discharge were not significantly different among groups. The similarity of discharge times may have resulted from our strict requirement for all patients to void before discharge. Many previous studies have allowed general and local anesthesia (but not spinal/epidural) patients to qualify for hospital dismissal without voiding (3). Mulroy et al. (24) recently concluded that ambulatory surgery patients may be discharged before voiding after short-acting spinal or epidural anesthesia. In their series of 131 patients fast-tracked to hospital dismissal, none had postoperative urinary problems, despite being discharged without voiding requirements. Additional studies are needed to further define postoperative requirements (voiding, ambulation, and oral intake) for safe patient dismissal.

All three anesthetic techniques were associated with extremely high patient satisfaction, with ≥90% of patients reporting VAS satisfaction scores of ≥8 (1–10 scale) for both anesthetic approach and postoperative pain management in all groups. However, patients receiving spinal anesthesia were significantly more satisfied with their care compared with general anesthesia patients. Contributing factors to this satisfaction may have been the ability of all spinal anesthesia patients to bypass the PACU and be admitted directly to the ambulatory surgery unit, where they were able to begin oral intake and mobilize. Furthermore, patients within both regional anesthesia groups were significantly more satisfied with their postoperative pain management when compared with general anesthesia patients. Therefore, overall patient satisfaction was highest among patients receiving regional techniques.

A potential difficulty in the results analysis of this prospectively randomized study must be discussed. Patients undergoing knee arthroscopy may have significant bias regarding anesthetic technique. In a previous arthroscopy study, Mulroy et al. (3) reported small recruitment (only 51 of 192 potential patients were included). Patients often refused randomization because they preferred to be awake for the procedure and observe it on a monitor, or they requested general anesthesia because of fear of awareness during surgery. Thus, a true optimization of anesthetic technique must account for patient preference.

In conclusion, we compared the surgical operative conditions, postoperative recovery and discharge times, analgesic requirements, patient satisfaction, and side effects of three anesthetic techniques for outpatient knee arthroscopy. Although there was no significant difference in hospital dismissal times among groups, small-dose spinal and psoas compartment blockade decreased overall resource utilization by avoiding PACU admission compared with general anesthesia. Furthermore, these regional techniques achieved superior patient satisfaction and immediate postoperative VAS pain scores compared with general anesthesia patients. Although these results are significant, individualization remains necessary. Surgical technique and duration, patient preferences and expectations, postoperative nursing management, and institutional practice models (such as fast-tracking) must all be taken into consideration when determining anesthetic management.

TABLE Cited Here...

TABLE Cited Here...

Back to Top | Article Outline


1. Parnass SM, McCarthy RJ, Bach BR Jr, et al. Beneficial impact of epidural anesthesia on recovery after outpatient arthroscopy. Arthroscopy 1993; 9: 91–5.
2. Ben-David B, Solomon E, Levin H, et al. Intrathecal fentanyl with small-dose dilute bupivacaine: better anesthesia without prolonging recovery. Anesth Analg 1997; 85: 560–3.
3. Mulroy MF, Larkin KL, Hodgson PS, et al. A comparison of spinal, epidural, and general anesthesia for outpatient knee arthroscopy. Anesth Analg 2000; 91: 860–4.
4. Jacobson E, Forssblad M, Rosenberg J, et al. Can local anesthesia be recommended for routine use in elective knee arthroscopy? A comparison between local, spinal, and general anesthesia. Arthroscopy 2000; 16: 183–90.
5. Ben-David B, DeMeo PJ, Lucyk C, Solosko D. A comparison of minidose lidocaine-fentanyl spinal anesthesia and local anesthesia/propofol infusion for outpatient knee arthroscopy. Anesth Analg 2001; 93: 319–25.
6. Wong J, Marshall S, Chung F, et al. Spinal anesthesia improves the early recovery profile of patients undergoing ambulatory knee arthroscopy. Can J Anaesth 2001; 48: 369–74.
7. Patel NJ, Flashburg MH, Paskin S, Grossman R. A regional anesthetic technique compared to general anesthesia for outpatient knee arthroscopy. Anesth Analg 1986; 65: 185–7.
8. Cappelleri G, Casati A, Fanelli G, et al. Unilateral spinal anesthesia or combined sciatic-femoral nerve block for day-case knee arthroscopy: a prospective, randomized comparison. Minerva Anestesiol 2000; 66: 131–6.
9. Bonicalzi V, Gallino M. Comparison of two regional anesthetic techniques for knee arthroscopy. Arthroscopy 1995; 11: 207–12.
10. Sansone V, De Ponti A, Fanelli G, Agostoni M. Combined sciatic and femoral nerve block for knee arthroscopy: 4 years’ experience. Arch Orthop Trauma Surg 1999; 119: 163–7.
11. Parkinson SK, Mueller JB, Little WL, Bailey SL. Extent of blockade with various approaches to the lumbar plexus. Anesth Analg 1989; 68: 243–8.
12. Winnie AP, Ramamurthy S, Durrani Z, Radonjic R. Plexus blocks for lower extremity surgery: new answers to old problems. Anesthesiol Rev 1974; 1: 11–6.
13. Biboulet P, Morau D, Ryckwaert Y, et al. Extent of blockade along the thigh after the anterior and the posterior approach to the lumbar plexus [abstract]. Anesthesiology 2000; 93: A816.
14. Aldrete JA, Kroulik D. A postanesthetic recovery score. Anesth Analg 1970; 49: 924–34.
15. Chung F. Are discharge criteria changing? J Clin Anesth 1993; 5: 64S–8.
16. Brands E, Callanan VI. Continuous lumbar plexus block: analgesia for femoral neck fractures. Anaesthesia 1978; 6: 256–8.
17. Chudinov A, Berkenstadt H, Salai M, et al. Continuous psoas compartment block for anesthesia and perioperative analgesia in patients with hip fractures. Reg Anesth Pain Med 1999; 24: 563–8.
18. Ryckwaert Y, Macaire P, Choquet O, et al. Postoperative analgesia by continuous psoas compartment block after total hip arthroplasty [abstract]. Anesthesiology 2000; 93: A893.
19. Stevens RD, Van Gessel E, Flory N, Fournier R. Lumbar plexus block reduces pain and blood loss associated with total hip arthroplasty. Anesthesiology 2000; 93: 115–21.
20. Horlocker TT, Hebl JR, Kinney MAO, Cabanela ME. Opioid-free analgesia following total knee arthroplasty: a multimodal approach using continuous lumbar plexus (psoas compartment) block, acetaminophen and ketorolac. Reg Anesth Pain Med 2002; 27: 105–8.
21. Greengrass RA, Klein SM, D’Ercole FJ, et al. Lumbar plexus and sciatic nerve block for knee arthroplasty: comparison of ropivacaine and bupivacaine. Can J Anaesth 1998; 45: 1094–6.
22. Farny J, Girard M, Drolet P. Posterior approach to the lumbar plexus combined with a sciatic block using lidocaine. Can J Anaesth 1994; 41: 486–91.
23. Reuben SS, Sklar J. Pain management in patients who undergo outpatient arthroscopic surgery of the knee [review]. J Bone Joint Surg Am 2000; 82: 1754–66.
24. Mulroy MF, Salinas FV, Larkin KL, Polissar NL. Ambulatory surgery patients may be discharged before voiding after short-acting spinal and epidural anesthesia. Anesthesiology 2002; 97: 315–9.
© 2003 International Anesthesia Research Society