CURRENT STATUS OF REGIONAL ANESTHESIA FOR ADULT OUTPATIENTS n6m14mr
Dermot Fitzgibbon MB, BCh, FFARCSI

From the Department of Anesthesiology, University of Washington School of Medicine, Seattle, Washington

Address reprint requests to
Dermot Fitzgibbon, MB, BCh, FFARCSI
Department of Anesthesiology
University of Washington
School of Medicine
1959 N.E. Pacific Street
Seattle, WA 98195

BENEFITS OF REGIONAL ANESTHESIA
The challenge of anesthesia for ambulatory patients is to provide for rapid return to street readiness with the most effective postoperative analgesia and minimal undesirable side effects. Regional anesthesia, with its selective local action and relatively simple equipment, offers an excellent anesthetic choice in an outpatient facility. In addition to limiting the anesthetized area to the surgical site, the common side effects of general anesthesia (e.g., nausea, vomiting, lethargy) are reduced, the risks and side effects of endotracheal intubation are minimized, patient recovery time may be decreased, and improved analgesia is provided in the postoperative period. a7i a46i
A number of studies a51i a82i have evaluated the efficacy of ambulatory regional anesthesia. Urmey et al a82i prospectively recorded data on ambulatory surgery patients at an orthopedic speciality hospital where regional anesthesia was the first-line standard care; the various types of anesthesia administered are listed in Table 1 (Table Not Available) . Only 4.4% of patients who had regional anesthesia required admission compared with 12% of general anesthetics. Discharge times were similar for general, spinal, or epidural anesthesia (average of 3 hours); patients who had peripheral nerve blocks were discharged in approximately 2 hours. Failure of regional anesthesia, necessitating general anesthesia, occurred in only 1% of cases. The authors concluded that regional anesthesia in an ambulatory center is effective in all but a small percentage of patients. Osborne a51i evaluated outcome for 6000 consecutive procedures in a major public teaching hospital day surgery unit. Anesthesia-related complications were more frequent with general anesthesia (1:114) than with regional anesthesia (1:180) or local anesthesia plus sedation

712
TABLE 1 -- ANESTHETIC TECHNIQUES FOR AMBULATORY ORTHOPEDIC PROCEDURES
From Urmey WF, Stanton J, Sharrock NE: Initial one-year experience of a 97.3% regional anesthesia ambulatory surgery center. Reg Anesth 18:69, 1993; © Churchill Livingstone, with permission.
(Not Available)

(1:780). Recovery with regional or local anesthesia was significantly shorter than after general anesthesia.
Despite the potential advantages cited regional anesthesia should not be considered universally appropriate. Factors that contribute to a successful regional anesthetic include the appropriate selection of patients, anesthetic technique, and local anesthetic, use of sedative and hypnotic agents, and the skill of the anesthesiologist. Prior screening of patients through preanesthesia testing (PAT) clinics is very useful in determining the acceptability of patients for a regional anesthetic. Very young or excessively anxious patients may be poor candidates. Similarly, obese patients may present technical problems, especially for central neuraxial blocks. Patients of American Society of Anesthesiologists (ASA) physical status III or IV may be particularly good candidates for ambulatory regional anesthesia compared to general anesthesia, especially if their systemic diseases are medically stable.

SELECTION OF TECHNIQUE AND LOCAL ANESTHETIC
Outpatient regional techniques require some modification from standard inpatient procedures. Ideally, an outpatient regional technique should be rapid in onset and result in few if any acute or delayed complications (e.g., pneumothorax). The additional time needed to perform many regional blocks, as well as the time needed for the anesthetic to take effect, is a potential drawback when procedures are short and turnover between cases is rapid. Use of blocks that require more time than the procedure itself to perform should be limited to those situations where specifically indicated for medical reasons or the patient expresses a strong preference for a specific technique. Blocks that significantly impair the ability to ambulate and void should be tailored to the anticipated usual duration of surgery by appropriate selection of both local anesthetic agent and technique to minimize both recovery and discharge time. Prolonged analgesia from a block (e.g., foot, arm, or hand blocks) may be beneficial in some instances, particularly if the ability of the patient to perform various activities is not significantly impaired; however, prolonged anesthesia may provoke anxiety or be considered unpleasant or irritating by many patients when

713 it persists for many hours after hospital discharge and should be discussed with patients before instituting such a block.
Local anesthetic agents are commonly classified according to their relative potency and duration of action as follows: low potency and short duration (e.g., chloroprocaine), moderate potency and duration (e.g., lidocaine and mepivacaine), high potency and long duration (e.g., tetracaine, bupivacaine, and etidocaine). Selection of specific blocks are discussed later. Selecting the appropriate local anesthetic for a given regional anesthetic requires consideration of a number of factors, including potency, speed of onset, duration of action of local anesthetics, site and duration of surgery, the degree of muscle relaxation required, and the duration of analgesia desired. Duration of anesthesia with a given agent varies with the site of injection and frequently with the total mass of drug injected. a70i Thus, bupivacaine injected into the epidural space lasts approximately 2 to 3 hours whereas the same dose injected into the brachial plexus may last 10 to 11 hours. Vasoconstrictors, such as epinephrine, are added to increase the duration of action, provide an indication of intravascular injection, and reduce peak serum levels of local anesthetic. The extent to which epinephrine prolongs the duration of anesthesia depends on the specific local anesthetic used and the site of injection. Vasoconstrictors do not prolong the duration of action of all local anesthetics in all situations (Table 2) (Table Not Available) . Epinephrine prolongs the duration of action of all agents for peripheral nerve blocks except ropivacaine. a33i It also prolongs the duration of action of epidural chloroprocaine, lidocaine, and mepivacaine. The local anesthetic properties of the intrinsically more potent and longer acting agents (bupivacaine, etidocaine, tetracaine) are influenced less by the addition of epinephrine, particularly when such agents are used epidurally. Epinephrine does not markedly prolong the duration of motor block by epidural bupivacaine or etidocaine; however, it does extend the sensory block by these epidural agents. a71i The effects of epinephrine added to agents used for spinal anesthesia are discussed later.
The optimal dose of epinephrine is one that would produce maximal increase in the duration of a local anesthetic agent and minimal hemodynamic effects. Kennedy et al a36i showed that a supraclavicular brachial plexus block with 30 mL of 1.6% lidocaine has virtually no hemodynamic effects whereas the same agent with epinephrine 1:200,000 produced a dose-related increase in cardiac rate, cardiac output, and stroke volume that persisted for 90 minutes and decreases in peripheral resistance and concomitant changes in mean arterial pressure that persisted for 120 minutes. Absorbed epinephrine produces predominantly beta-adrenergic effects with little evidence of alpha-adrenergic effects at
TABLE 2 -- DURATION OF ACTION OF LOCAL ANESTHETICS AND EFFECT OF EPINEPHRINE
Adapted from Ellis JS: Local anesthetics. In Kirby, Gravenstein: Clinical Anesthesia Practice. Philadelphia, WB Saunders, 1994.
(Not Available)

714 doses up to 400 mug. Furthermore, epinephrine produced a dose-related increase in mean duration of anesthesia, but only up to a concentration of 1:200,000, above which the cardiocirculatory changes continued to increase without any further increase in the duration of anesthesia. The optimal dose of epinephrine in the above study was a 1:200,000 solution or 5 mug/ml concentration. If the use of epinephrine is desirable it should be added to the local anesthetic just before the local anesthetic is used. The reason for this is that commercial solutions of epinephrine containing epinephrine are buffered to a lower pH than the standard solution of that agent in an effort to oxidation of the epinephrine. Such acidification moves the pH farther from the pKa of that solution, reducing the availability of the free base and the rate of diffusion of the local anesthetic.
The speed of onset of local anesthetics is primarily related to the agent selected and the site of injection. Thus, agents such as chloroprocaine have a more rapid onset in the epidural space than agents such as lidocaine and bupivacaine, and lidocaine and mepivacaine have a more rapid onset than bupivacaine when used for peripheral nerve blocks. Efforts to increase the speed of onset of local anesthetics by the addition of bicarbonate have yielded contradictory results and appear to be minimally effective.

SEDATION/ANALGESIA
Many patients undergoing surgery with local or regional anesthesia prefer to be sedated. Small doses of short-acting drugs should be carefully titrated. Midazolam is an excellent agent for the moderately-anxious patient. The amnesia it produces does not correlate with the apparent level of sedation; fully conscious patients may have no recall of perianesthetic events. Low-dose propofol infusions also appear to be excellent agents for intraoperative sedation. a84i Adequate intraoperative sedation can usually be achieved with infusion rates of 25 to 100 mug/kg-1 /min-1 . When intraoperative amnesia is desired in addition to a rapid recovery, administration of small titrated doses of midazolam (0.5-3 mg intravenously) prior to the propofol infusion may offer advantages over either drug alone. Many outpatients find the use of local anesthetic techniques acceptable alternatives to both general and regional anesthesia when adequate sedation and anxiolysis are provided. The addition of a short-acting opioid (e.g., fentanyl, alfentanil) is especially useful if paresthesia are sought or nerve stimulation performed during a regional technique and obtundation is undesirable.

UPPER EXTREMITY BLOCKS
Brachial plexus anesthesia is suitable for many upper extremity procedures, most notably for orthopedic surgery. The axillary approach is suitable for forearm and hand surgery whereas the interscalene approach is useful for shoulder and more proximal upper extremity surgeries. Intravenous regional anesthesia (Bier block) provides adequate anesthesia of the hand and forearm for procedures of limited duration (less than 1 hour).

Axillary Brachial Plexus Block
The axillary approach is safe and effective for outpatients. a19i Localization of the plexus may be based on elicitation of paresthesias, palpation of a click as

715 the surrounding fascial sheath is pierced, piercing the axillary artery, or observation of motor responses to direct electrical stimulation of nerves. Although there is some controversy regarding the most reliable method of performing a successful axillary block, block of at least two nerves appears to be important to improving the success rate. a42i Urban et al a79i prospectively evaluated 508 patients who received either interscalene or axillary brachial plexus blocks for upper extremity surgery. They noted that major immediate complications were infrequent, with only one mild seizure in the axillary block group and evidence of intravascular injection in only two of the patients in the interscalene group; however, 23% of patients in the axillary group complained of pain, tenderness, or bruising in the axilla on the first day of surgery, and in 2 patients the pain persisted for 2 weeks. Paresthesia occurred in 19% of patients in the axillary group on the first postoperative day, and 7% of patients continued to have paresthesias 2 weeks after surgery. Similar problems of bruising and persistent numbness associated with axillary blocks were reported by Cooper et al. a13i

Interscalene Brachial Plexus Block

Despite its common use for open shoulder surgery, interscalene block is less widely accepted as a viable anesthetic technique in the fast turnover setting of ambulatory surgery. a66i Its major advantage is that it provides very effective postoperative analgesia as well as providing satisfactory anesthesia for surgery or arthroscopy of the shoulder. The block can also be used for surgery on the upper arm; however, the lower part of the brachial plexus, and in particular, the medial cutaneous nerve of the arm, intercostobrachial nerve, and the ulnar nerve are frequently missed by this approach. In situations where anesthesia of the medial aspect of the arm or forearm are required, additional block of the intercostobrachial nerve or supplementation by the axillary approach is required. The interscalene approach to the plexus is generally unsuitable for outpatient hand surgery. D'Alessio et al a16i reported on the use of interscalene block for ambulatory surgery. Compared with general anesthesia, the block required significantly less total nonsurgical intraoperative time use and resulted in fewer unplanned admissions for therapy of severe pain, sedation, or nausea and vomiting. A failure rate of 8.7% was observed. No airway problems other than hoarseness due to recurrent laryngeal nerve block were noted. Postoperatively, patients who had the block proceeded more rapidly through PACU and phase 2 recovery than comparable patients who received general anesthesia (72 minutes ± 24 versus 102 minutes ± 40, for regional versus general; P = 0.0001).
A number of different nerves (phrenic, recurrent laryngeal, and cervical sympathetic) may be blocked in addition to the roots of the brachial plexus when using the interscalene approach. Side effects commonly observed from block of these nerves are relatively minor and well tolerated. Involvement of the recurrent laryngeal and cervical sympathetic nerves is rarely significant, but patients may experience hoarseness, dysphagia, and blurred vision, and should be cautioned against drinking or eating while hoarseness and dysphagia are present. Reversible diaphragmatic paralysis has been reported to occur in up to 100% of cases. a83i In addition, greater than 25% mean reductions in functional residual capacity and FEV1 have also been associated with the block. a81i Complete or incomplete paralysis of a hemidiaphragm is usually well-tolerated, but spirometric studies have documented altered respiratory capacity for several hours. a83i a53i Urmey et al a80i have listed respiratory considerations for interscalene brachial plexus block.

716

LOCAL ANESTHETIC CHOICE FOR UPPER EXTREMITY BLOCK
Agents commonly used for peripheral nerve blocks, including brachial plexus blocks, are listed in Table 3 . In general, agents of intermediate potency exhibit a more rapid onset than the more potent agents. Etidocaine may be an exception because it produces a block of relatively rapid onset. a71i Mepivacaine provides a greater degree of motor block with a longer duration of sensory anethesia than lidocaine when used for brachial plexus block. a86i The variation in duration of anesthesia after brachial plexus block is also considerably greater than that observed after other types of regional anesthesia. As such, it is prudent to forewarn patients about to receive a brachial plexus block of the possibility of prolonged sensory and motor block, particularly when agents such as bupivacaine and etidocaine are used. When numbness is persistent after an upper extremity block, the patient's discharge need not be delayed until the block has resolved. Instructions can be given in the care of the extremity to prevent injury while sensation is lacking. Patients should also be instructed on the use of short-acting opioids and nonsteroidal anti-inflammatories prior to complete resolution of the block, and on regular use thereafter if significant postoperative pain is anticipated. Patients should be reassured that sensation will return after discharge and be given a contact telephone number or person if persistent beyond the expected duration.

Intravenous Regional Anesthesia
Surgical anesthesia during intravenous regional anesthesia (IVRA) is produced by multiple and complementary mechanisms, including block of peripheral small nerves and nerve endings (initial effect), block of nerve trunks at a proximal site (main anesthetic component), ischemia, and compression of nerve trunks. a63i The block can be used for various upper extremity operations, including both soft-tissue and orthopedic procedures, primarily in the hand and forearm. a5i It has also been used for foot procedures with a calf tourniquet. a38i The agent most commonly used for an upper extremity block is 0.5% lidocaine, approximately 40 mL for upper extremity procedures, and 0.25% lidocaine 50 to 60 mL for lower extremity procedures. Because of the risk of toxic reactions, bupivacaine is not a suitable agent. Although chloroprocaine, because of its extremely short serum half-life, might appear to be the ideal choice for IVRA, it is contraindicated because it can cause phlebitis on intravenous injection. a31i IVRA is a safe and effective way to provide anesthesia for surgery distal to the elbow of less than 1 hour's duration. A limiting feature of the technique is the onset of tourniquet pain. Clinical investigations involving unmedicated unanesthetized volunteers
TABLE 3 -- LOCAL ANESTHETICS USED FOR PERIPHERAL NERVE BLOCKS
Agent Concentration Duration (minutes)
2-Chloroprocaine 2%-3% 30-75
Lidocaine 1%-2% 50-120
Mepivacaine 1%-2% 120-300
Bupivacaine 0.25%-0.5% 300-720
Etidocaine 0.5%-0.75% 300-720

717 have shown that upper extremity tourniquet inflation can be tolerated for 29 to 34 minutes in motivated, healthy subjects. a3i a15i The use of a double tourniquet reduces but does not eliminate this problem. Onset of analgesia and anesthesia after injection is rapid, so surgery or manipulation may begin within 5 to 10 minutes. Lidocaine typically produces sensory loss earliest on the radial forearm and in the first dorsal web space. The onset of fingertip anesthesia is variable and unpredictable, as is decrease in motor function. a78i Supplementation of IVRA with a digital nerve block may be used if digital anesthesia is inadequate. Normal sensation and motor power return rapidly after tourniquet release. In some situations rapidity of recovery with loss of analgesia may be considered a disadvantage. Infiltration of the wound with long acting local anesthetic or peripheral nerve block by the surgeon prior to cuff release and application of dressings may overcome this problem in the early postoperative period. Recently, Reuben et al a61i studied the effects of IVRA using ketolorac 60 mg and 38 mL of 0.5% lidocaine, and noted that patients who had received ketoralac experienced less postoperative pain (both in PACU and in the first 24 hours), and concluded that ketorolac improves IVRA with 0.5% lidocaine, both in terms of controlling intraoperative tourniquet pain and by diminishing postoperative pain.

ABDOMINAL, PERINEAL, AND LOWER EXTREMITY BLOCKS
Lumbar epidural anesthesia is suitable for pelvic, lower abdominal, and lower extremity (excluding foot) surgery. The onset and quality of sensory and motor block of the fifth lumbar and first sacral roots by epidural anesthesia are often delayed or incomplete, with a failure of the S1 segment sensory block in up to 46% of patients when using lidocaine hydrochloride and epinephrine. a28i Spinal anesthesia is superior to epidural anesthesia for lower extremity and perineal surgery, and is useful for lower extremity, urologic, and herniorrhaphy procedures in the ambulatory setting. Anesthesia and analgesia for mid- and forefoot surgery can be satisfactorily achieved with peripheral nerve blocks of the ankle and foot. Complete analgesia lasting from 10 to 25 hours after surgery has been described with ankle blocks using bupivacaine 0.5%. a67i Details on how to perform this block are provided by Schurman. a69i

Spinal Anesthesia
The advantages of spinal anesthesia for ambulatory surgery include ease of administration, rapid onset, and high reliability. Potential disadvantages include the possibility of postdural puncture headache, urinary retention, and transient radicular irritation with lidocaine. In North America three local anesthetics are commonly used to produce spinal anesthesia: lidocaine, tetracaine, and bupivacaine (Table 4) . Lidocaine produces short-to-intermediate-acting anesthesia and is ideally suited for ambulatory regional anesthesia. The only mixture of lidocaine currently approved by the Food and Drug Administration (FDA) for subarachnoid use is a 5% solution made hyperbaric in 7.5% dextrose. Recently, observations have been made that 5% hyperbaric lidocaine may cause back and bilateral leg pain. Pinczower et al a55i described the problem in nine patients who received hyperbaric lidocaine for spinal anesthesia in an ambulatory setting. The dose of lidocaine ranged from 40 to 100 mg. The pain was described as either sharp or cramping with or without associated back pain. None of these patients

718
TABLE 4 -- LOCAL ANESTHETICS USED FOR SPINAL ANESTHESIA
Agent Concentration Usual Dose Baricity Duration (minutes)

Lidocaine 5% in 7.5% glucose 50-100 mg Hyperbaric 45-80
2% 40-60 mg ? Hypobaric 60-100
Bupivacaine 0.75% in 8.25% glucose 9-15 mg Hyperbaric 90-240
0.5% 15 mg Isobaric 90-240
Tetracaine 0.5% in 5% glucose 10-20 mg Hyperbaric 150-300
0.5% 10-20 mg Isobaric 150-300
0.1% 7.5-10 mg Hypobaric 150-300

demonstrated objective neurologic deficits. In all cases, the symptoms resolved fully within 1 week. Tarkkila et al a76i estimate that approximately 10% of patients who receive hyperbaric 5% lidocaine may experience these problems. Because of concerns of neurotoxicity relating to either a high lidocaine or dextrose concentration, a41i a68i a76i the manufacturing company now recommends dilution of the mixture with equal volume of cerebrospinal fluid (CSF) prior to subarachnoid injection. A number of studies suggest equal if not greater efficacy with lower concentrations of lidocaine. a43i a77i Liew a43i reported on the use of 25 mg of 0.5% lidocaine for minor outpatient gynecologic procedures; 93% of patients developed a block to T10. The mean duration of sensory block was 32.5 minutes, and all patients had complete resolution of motor block within 1 hour. Bupivacaine 0.75% in 8.25% dextrose is useful for procedures lasting 2 to 2.5 hours, althought the time to spontaneous voiding may be considerably longer. a23i Although subarachnoid bupivacaine without a vasoconstrictor possesses an anesthetic profile similar to tetracaine without a vasoconstrictor, differences do exist between the drugs. The depth and duration of motor block is probably greater with tetracaine a11i making it less desirable for outpatient anesthesia. Bupivacaine, when compared with tetracaine, has been reported to cause less hypotension a4i and to have a lower incidence of lower extremity tourniquet pain. a12i
The effect of epinephrine on subarachnoid anesthesia is confusing. Chambers a9i noted that different doses of epinephrine (100-300 mug) added to hyperbaric lidocaine did not prolong the duration of anesthetic block to any clinically useful effect, but time to full recovery was delayed by 40 to 50 minutes; similar effects were observed with subarachnoid bupivacaine. a8i Chiu a10i showed that 200 mug of epinephrine added to 50 mg of hyperbaric lidocaine delayed the ability to void urine by approximately 1 hour, and may delay discharge. Moore et al a47i studied the effect of epinephrine added to lidocaine for lower extremity surgery using the occurrence of intraoperative pain as an end-point rather than thoracic dermatome sensory regression. The duration of anesthesia prior to the occurrence of pain was 87 minutes ± 16 minutes for the plain lidocaine group and 128 minutes ± 23 minutes for the lidocaine with epinephrine group. Epinephrine may have a differential vasoconstrictive effect at different levels. Kozody et al a40i showed that 200 mug of intrathecal epinephrine caused dural vasoconstriction, implying that epinephrine prolongs a lidocaine block primarily in the lumbosacral area. Thus, some prolongation of lower extremity anesthesia may be achieved by the addition of epinephrine but at a potential cost of delayed recovery and discharge. For routine lower extremity or perineal surgery the author recommends the use of 50 to 60 mg 5% lidocaine in 7.5% dextrose diluted in equal volumes with CSF and injected without epinephrine. If prolongation of lower extremity anesthesia is necessary (greater than 60 minutes) the author

719 recommends the addition of 200 mug of epinephrine to lidocaine or the use of bupivacaine 8 to 12 mg without epinephrine. Subarachnoid tetracaine with epinephrine is not recommended for ambulatory surgery.

RECOVERY AFTER SPINAL ANESTHESIA
The generally accepted sequence of return of function after spinal block is motor, sensory, and sympathetic; however, several studies have found recovery of sympathetic activity may occur before complete regression of the motor or sensory spinal block, a18i a62i although Axelsson a2i demonstrated that motor strength in the lower extremities was restored 40 to 140 minutes on average before restoration of detrusor strength after subarachnoid injections of bupivacaine and tetracaine. Pflug a54i considered the ability to urinate a final indication of reversal of sympathetic paralysis because an intact, functioning sympathetic nerve supply to the bladder and urethra is necessary for this function. Suitable criteria for ambulation after spinal anesthesia include normal perianal (S4-5) pinprick sensation, plantar flexion of the foot, and proprioception of the big toe. Discharge criteria after spinal anesthesia include normal sensation, ability to walk, and ability to urinate. Patients should be instructed not to lift heavy objects or strain for 24 hours.

POSTDURAL PUNCTURE HEADACHE
The use of spinal anesthesia in younger patients for ambulatory procedures has been discouraged by reports of a high incidence of postdural puncture headaches (PDPH). a25i Outpatients appear to have a higher risk of PDPH than inpatients. a25i a49i In ambulatory surgery the possibility of PDPH assumes more prominence because it may impair the ability to return to normal activities shortly after the procedure. For patients planning to travel long distances soon after surgery, the occurrence of a severe headache in a geographical area remote from the hospital may be difficult to manage. Headache following dural puncture is typically delayed in onset and postural in nature. PDPH presumably occurs when a slow leak of CSF leads to contraction of the subarachnoid space and compensatory expansion of the pain-sensitive intracerebral veins. A variety of factors are involved in the incidence of PDPH, including age, sex, needle diameter, and needle tip design. a26i a48i Patients younger than 45 years old are at greater risk from a higher incidence and severity of headache requiring treatment. Efforts have been made to reduce the incidence of PDPH by changing the size and design of the needle. In a meta-analysis of PDPH and spinal needle design, Halpern a30i showed a reduction in the incidence of PDPH when noncutting needles rather than cutting needles were used, unless the discrepancy in needle diameter was very large. There was also a reduction in PDPH when a small-diameter spinal needle compared with a larger diameter needle of the same type was used. Quincke point needles may cause persistent dural tears whereas blunt or noncutting spinal needles (Whitacre, Sprotte) may spread dural fibers and decrease CSF loss after dural puncture, thereby reducing the incidence of PDPH. a65i The use of 22- and 25-gauge Quincke needles in young patients cannot be encouraged as few studies with large patient numbers have demonstrated a PDPH incidence under 10%. a75i Although the incidence of headache significantly decreases with the smaller diameter needles, such as 27-gauge and 29-gauge, a22i technical difficulties may be increased. a17i a44i The incidence of PDPH in young

720 patients with a 25-gauge Whitacre needle varies between 0.6% and 3.0%. a6i a7i The incidence of headache after a 24-gauge Sprotte is probably similar, a21i a64i although Wiesel et al a85i report an incidence of 15.2% with 24-gauge Sprotte needles in patients under 45 years old. Needle size is less of a factor in the incidence of PDPH when noncutting needles are used compared with Quincke-type needles. a14i a27i Smith et al a72i recently compared the use of 25- and 27-gauge Whitacre needles in obstetric patients and observed increased technical difficulties with the use of the smaller gauge needles. The possibility of a lower incidence of PDPH with the 27-gauge needle was unproven in this study. Campbell et al a7i noted that 25-gauge Whitacre and 24-gauge Sprotte needles were comparable with respect to ease of insertion and incidence of PDPH, but that the Whitacre needles were substantially cheaper than the Sprotte needles.
Based on these studies and on other currently available information, the author recommends the routine use of 25-gauge Whitacre needles for young patients in the ambulatory setting as the incidence of PDPH does not appear to be significantly different with either Sprotte needles or the smaller gauge Whitacre needles, and the cost-savings may be significant with 25-gauge Whitacre. Although advances in needle tip design may have reduced the incidence of PDPH, the problem still persists, and a higher incidence should be expected in the younger outpatient age groups. Alternative methods of regional anesthesia for outpatient anesthesia should be considered if either the risk of PDPH is unacceptable to the patient or access to appropriate medical care or advice is difficult should this problem arise. If PDPH occurs and does not respond to conservative approaches, an epidural blood patch on an outpatient basis can be highly effective. a59i Patients are instructed to rest quietly for 1 hour after injection of autologous blood. Patients can be discharged but should subsequently avoid straining and should maintain good oral fluid intake at home.

Epidural Anesthesia
Effective use of epidural local anesthesia requires an understanding of local anesthetic potency and duration and a realistic estimate of the length of the procedure. A variety of different agents are used for epidural anesthesia (Table 5) . 2-Chloroprocaine (2-CP), an amino ester local anesthetic, is a short-acting agent that allows efficient matching of surgical procedure length and duration of epidural analgesia. It is available in 2% and 3% concentrations, with the latter preferable for surgical anesthesia.
Kopacz a39i noted that the duration of sensory anesthesia after epidural injection of 20 mL of 3% 2-CP and 1.5% lidocaine was significantly shorter (133 minutes ± 28 minutes for 2-CP, 182 minutes ± 38 minutes for lidocaine) than

TABLE 5 -- LOCAL ANESTHETIC DRUGS USEFUL FOR OUTPATIENT EPIDURAL ANESTHESIA
Drug Concentration Typical Volume (mL) Duration (minutes) *
Chloroprocaine 2%-3% 15-24 30-90
Lidocaine 1.5%-2% 15-24 60-90
Mepivacaine 1.5%-2% 15-24 90-120
*Solution contains 1:200,000 concentration of epinephrine.

721 after 1.5% mepivacaine (247 minutes ± 42 minutes). In addition, discharge times were significantly shorter for 2-CP (269 minutes ± 62 minutes) and lidocaine (284 minutes ± 62 minutes) than for mepivacaine (357 minutes ± 71 minutes). Each of these solutions contained 5 mug/mL epinephrine. Deck et al a20i compared the effects of epidural 3% 2-CP and 1.5% lidocaine without epinephrine. Patients receiving 2-CP had significantly faster times to block resolution, ambulation, and discharge than those receiving lidocaine. Patients receiving 2-CP had resolution of block (120 minutes ± 15 minutes versus 190 minutes ± 44 minutes) and were discharged sooner (127 minutes ± 16.8 minutes versus 195 minutes ± 43.8 minutes) than patients in the lidocaine group. The implication of these studies are that moderate epidural doses of 3% 2-CP without epinephrine may be the epidural solution of choice for ambulatory epidural anesthesia.
Prolonged and sometimes permanent neurologic deficits have been reported after inadvertent subarachnoid injections of 2-CP during attempted epidural anesthesia. a59i The combination of a low pH and the presence of sodium bisulfite may have been responsible for the neurotoxic reactions. a29i 2-CP itself does not appear to be neurotoxic. Partly because of these reports, the preparation of the drug has undergone an evolution resulting in 2-CP preparations with different additives. Earlier preparations of the drug contained methylparaben. In 1987, a new preparation that was free of both methylparaben and sodium bisulfite was marketed under the brand name Nesacaine-MPF (Nesacaine methylparaben free, Astra Pharmaceutical, Westborough, MA); however, this preparation contains disodium ethylenetetraacetic acid (EDTA), which has been associated with a syndrome of prolonged backache following regression of epidural anesthesia with Nesacaine-MPF. a24i a50i

2-CHLOROPROCAINE BACK PAIN
In 1989, Fibuch a24i published the first of several reports describing a syndrome of prolonged backache following regression of epidural anesthetic with EDTA containing Nesacaine-MPF. Orkin a50i reported a 40% incidence of backache following 2-CP epidural anesthesia in ambulatory patients. Stevens a73i found a 100% incidence of back pain after large doses of 2-CP. The pain typically begins during regression of sensory anesthesia. It is described as being deep, aching, or burning in character. The pain is poorly localized to the lumbar region and severe in intensity. It is not easily treated with conventional analgesics and may last 24 hours or more. It has not been reported to cause either a chronic pain syndrome or any lasting neurologic sequelae. This syndrome of severe back pain was not reported prior to the addition of EDTA; therefore, EDTA has been the leading suspect in the cause of back pain. One possibility is that EDTA leaking into the lumbar musculature may result in localized hypocalcemia and tetanus. a24i Other possible contributing factors include the low pH (2.5-3.5) of 2-CP as 2-CP is more acidic than the other commonly used local anesthetics. Stevens a74i concluded that large doses (> 40 mL) of 2-CP that contains EDTA resulted in a high incidence of back pain. Use of a lower volume (< 25 mL) significantly reduced the incidence of back pain. Treatment with epidural fentanyl (100 mug) has provided prompt relief. a73i
If 2-CP is planned for epidural use, a block might prudently first be established with 1.5% lidocaine (to exclude inadvertent subarachnoid injection and avoid potential neurotoxic effects from 2-CP) and then maintained with 2-CP. For short procedures, epinephrine should be avoided. The minimal volume of 2-CP necessary should be used. The use of epidural bupivacaine is not recommended for ambulatory surgery due to its unnecessarily prolonged duration.

722

KNEE SURGERY
Arthroscopic procedures on the knee joint form a large proportion of outpatient surgical procedures. The surgery may be solely diagnostic, and in this situation may last less than 30 minutes. Therapeutic procedures (e.g., meniscectomy) may last 2 to 3 hours. The knee is innervated by L3, L4, and L5 nerve roots anteriorly and the first two sacral roots posteriorly. A number of different regional anesthetic techniques have been used for knee surgery (Table 6) . Randel a57i considered epidural anesthesia superior to spinal or general anesthesia for outpatient knee arthroscopy. Patel et al a52i noted favorable operating conditions and good postoperative analgesia with a 3-in-1 femoral nerve block, a lateral cutaneous nerve block of thigh, and 10 mL 0.25% bupivacaine as supplementary intra-articular anesthesia, if required. A variety of analgesic techniques have been evaluated and advocated for managing postoperative pain after arthroscopic knee surgery. These techniques have largely focused on the use of intra-articular anesthetics, intra-articular opioids, and systemic nonsteroidal anti-inflammatory agents. Although studies have shown various degrees of success, intra-articular injections of local anesthetics and opioids are currently popular for postoperative analgesia. Intra-articular 0.25% bupivacaine, usually in doses of 20 mL, is associated with significantly improved early (1-6 hours) postoperative analgesia, and generally does not appear to be effectively maintained thereafter. a1i a35i Conflicting results exist regarding the efficacy of intra-articular morphine with some studies revealing effective analgesia for up to 48 hours a34i a37i
TABLE 6 -- REGIONAL ANESTHETIC TECHNIQUES FOR KNEE SURGERY
Technique Agents Advantages Disadvantages
Epidural Lidocaine 1.5% ± 2-CP Faster recovery time compared with spinal or general anesthesia. a57i Relatively slow onset time; wet tap; patchy block.
Spinal Lidocaine, bupivacaine Rapid onset; dense block. PDPH; unpredictable duration of surgery.
Nerve block (3-in-1, sciatic) Lidocaine + epinephrine Prolonged postoperative analgesia; no impairment of voiding. Slow onset; inability to ambulate.
Femoral nerve block + intra-articular injection Bupivacaine 0.25% + epinephrine Prolonged postoperative analgesia; no impairment of voiding; early discharge. Some difficulty with ambulation; ? patchy anesthesia.
Local anesthesia (skin + intra-articular) Lidocaine 0.5% + epinephrine Use of epinephrine should eliminate need for tourniquet; ? postoperative analgesia. Minor procedures only; potential for large doses of local anesthetic.

723
TABLE 7 -- LOCAL ANESTHETIC AGENTS USED FOR INFILTRATION TECHNIQUES
Agent Concentration Duration without Epinephrine (minutes) Duration With Epinephrine (minutes)
Lidocaine 0.5-1.0 30-60 120-360
Mepivacaine 0.5-1.0 45-90 120-360
Bupivacaine 0.25-0.5 120-240 180-420
Etidocaine 0.5-1.0 120-180 180-420

and others failing to demonstrate any clinical effect. a32i a56i Reuben a60i recently evaluated the effect of intravenous and intra-articular ketorolac (60 mg) with intra-articular bupivacaine (30 mL 0.25%) in patients undergoing elective arthroscopic meniscal surgery. Intra-articular ketorolac in combination with bupivacaine significantly decreased postoperative pain, decreased analgesic requirements, and increased analgesic duration after arthroscopic knee surgery. Patel et al a52i suggest that a 3-in-1 femoral nerve block may be extremely useful for outpatient knee arthroscopy, providing both excellent postoperative analgesia and a high degree of patient acceptance.

LOCAL ANESTHESIA
Local infiltration of the operative site with dilute solutions of local anesthetics is a simple and safe technique for outpatients. Because the injection of local anesthetics can be associated with significant discomfort, the use of intravenous sedative and analgesic drugs (so-called "conscious sedation") has become very popular among surgeons. a45i A variety of surgical procedures, including vasovasotomy, circumcision, hydrocele and spermatocele repairs, cystoscopy, inguinal herniorrhaphy, plastic and reconstructive surgery, and breast biopsy with needle localization, are suitable for local infiltration. In addition, local anesthetic supplementation (e.g., infiltration with 0.25% bupivacaine) decreases incisional pain in the recovery room and potentially hastens recovery time. a46i Agents commonly used for local infiltration are listed in Table 7 .

CONCLUSION
Regional anesthesia is ideally suited for many types of adult outpatient surgery. It requires appropriate choice of both anesthetic agent and technique. Central neuraxial anesthesia provides ideal anesthetic conditions for many procedures, and to avoid delays in time to discharge caused by either prolonged motor block or inability to void urine, the use of short-acting agents without epinephrine (lidocaine for subarachnoid, 2-CP for epidurals) is recommended. A variety of methods for improved postoperative analgesia is now available to minimize both patient discomfort and impairment of function. Increased use of regional anesthesia and analgesia in the ambulatory setting may result in greater satisfaction for both patient and physician alike.