SECTION EDITOR: DENISE J. WEDEL.
Hemidiaphragmatic paresis after supraclavicular block of the brachial plexus is known to occur, although neither the clinical effect on respiration nor the incidence of this side effect is well defined. Conversely, hemidiaphragmatic paresis occurs after 100% of interscalene blocks and diminishes certain pulmonary function variables [1-3]. This has led to the contraindication of interscalene block in patients who cannot withstand unilateral diaphragmatic paresis and/or a 25% reduction in forced vital capacity . The risk of C3-5 (phrenic nerve) blockade is theoretically less with supraclavicular block because it is performed more caudad in the brachial plexus. However, the occurrence of clinically significant hemidiaphragmatic dysfunction with this block would likewise eliminate it as an alternative regional anesthetic technique in patients with preexisting pulmonary impairment.
Previous studies report a 1%-75% incidence of hemidiaphragmatic paresis after supraclavicular block [5-9]. Use of relatively crude techniques (plain chest radiography and fluoroscopy) may explain this widely disparate range of incidence. Ultrasonography and respiratory inductive plethysmography (RIP) accurately identify and quantify diaphragmatic paresis in a variety of clinical settings [10-13], including interscalene block [1,2,14].
Two clinical questions must be addressed concerning hemidiaphragmatic dysfunction as a side effect of supraclavicular block. First, does it occur often enough to warrant concern in patients with compromised pulmonary function? Second, does it necessarily correlate with significant impairment of respiratory function? We designed this volunteer, observational study to answer these questions by determining by ultrasonography and RIP the likelihood of hemidiaphragmatic paresis after supraclavicular block; by assessing the relationship, if any, of hemidiaphragmatic paresis to pulmonary function impairment in normal subjects; and by assessing the onset and duration of hemidiaphragmatic paresis relative to upper extremity motor and sensory block.
After institutional review board approval and informed consent, eight healthy, unsedated volunteers underwent right supraclavicular brachial plexus block using a modification of the "plumb-bob" technique . The technique used differed from the plumb-bob approach in that the needle entry point was in the interscalene groove, 1 cm above the clavicle, with the needle initially directed 15[degree sign] caudad in the sagittal plane. All blocks were performed with subjects in the supine position by one investigator (JMN). Subjects were monitored with electrocardiogram and pulse oximetry throughout the study. Thirty milliliters of plain lidocaine 1.5% (Abbott Laboratories, North Chicago, IL) was injected after eliciting a single paresthesia below the shoulder with a 22-gauge regional block needle (Becton-Dickinson, Franklin Lakes, NJ).
Baseline measurements were obtained for all test variables. The time interval between the block procedure and the first measurements was 5 min for ultrasonography, 10 min for pulmonary function testing (PFT), 15 min for plethysmography, and 20 min for sensory examination and motor strength testing. Thereafter, each test was performed at 20-min intervals until return to baseline of diaphragm or sensory function, and 90% baseline of motor function.
Ultrasonographic assessment of diaphragmatic motion was performed (3.5-MHz sector transducer; Acuson, Mountain View, CA) with subjects in the semisitting position and with the head of the bed 75[degree sign] upright . The diaphragm was imaged in the coronal plane with the transducer in the mid-axillary line . Diaphragmatic motion was initially measured during quiet respiration. The direction of hemidiaphragmatic motion was then assessed as either normal or paradoxical during a forced sniff maneuver. Normal hemidiaphragmatic excursion is caudad during this maneuver. Paradoxical motion, which establishes hemidiaphragmatic paralysis, is cephalad .
Inductive plethysmography was performed using a respiratory inductive plethysmograph (Respitrace Plus[trade mark sign]; Sensormedics Co., Yorba Linda, CA), with impedance straps placed circumferentially at the levels of the nipple and the umbilicus . The average expansion of the trunk during inspiration was thus measured at these two levels and corresponded to the thoracic and abdominal components, respectively. Measurements were first performed during tidal volume breathing (average of two breaths) and then during vital capacity breathing (average of two breaths). From these measurements, the relative contributions of the thoracic and abdominal components to the total respiratory expansion were calculated . Plethysmographic results are reported as %Thoracic (the thoracic component of total respiratory excursion). Negative abdominal excursion during inspiration establishes paradoxical diaphragmatic motion.
PFT were performed using a hand-held spirometer (Renaissance PB100; Puritan-Bennett Co., Wilmington, MA). Variables measured included forced vital capacity, forced expiratory volume in 1 s, forced expiratory flow from 25% to 75% of the vital capacity, peak expiratory flow, peak inspiratory flow, ratio of forced expiratory flow to forced inspiratory flow at 50% lung volume, and maximal voluntary ventilation.
Sensation to pinprick (18-gauge needle) was assessed until return of normal sensation at three locations of the right upper extremity: the base of the radial thumb, the ulnar border of the hand, and the lateral border of the shoulder. These three locations correspond to distributions of the median nerve, ulnar nerve, and axillary nerve, respectively.
Muscle strength was measured using an isometric force dynamometer (MicroFET; Hoggan Health Industries, Draper, UT). Strength was tested for shoulder abduction (axillary and suprascapular nerves), elbow flexion (musculocutaneous nerve), elbow extension (radial nerve), and metacarpophalangeal flexion (median and ulnar nerves).
Data are expressed as mean +/- SEM. Repeated-measures analysis of variance was used to determine change over time for motor and pulmonary function. Subjects with paradoxical hemidiaphragmatic motion by ultrasound (paralyzed) were distinguished from those without (nonparalyzed) paradoxical hemidiaphragmatic motion. Plethysmographic data for each group were compared using repeated-measures analysis of variance and Fisher's protected least significant difference. Correlation between ultrasound and RIP was examined using the z-test. A P value <0.05 was considered significant.
The age of the subjects (7 male and 1 female) ranged from 28 to 37 yr. Their height ranged from 163 to 188 cm, and weight from 51 to 100 kg. No subject exhibited decreased oxygen saturation or respiratory symptoms during the study period.
After supraclavicular block, ultrasonography showed paradoxical motion of the diaphragm with the forced sniff maneuver in 50% (95% confidence interval 14-86) of the subjects (four of eight). The onset time of paradoxical motion was 10 +/- 5 min, and the time until recovery from paradoxical diaphragmatic motion on ultrasound was 75 +/- 17 min.
Respiratory inductive plethysmography showed no overall change in %Thoracic during tidal volume breathing (Figure 1). During vital capacity breathing, %Thoracic increased significantly (P = 0.005) in those subjects with hemidiaphragmatic paralysis, beginning at 15 +/- 0 min and ending at 70 +/- 22 min (Figure 1). In the paralyzed group, %Thoracic increased to >100% at some time points because of negative abdominal excursion. Figure 2 shows representative plethysmograms before and after hemidiaphragmatic block. The duration of paradoxical motion demonstrated by RIP correlated with the duration of paradoxical motion demonstrated by ultrasound (r = 0.973, P < 0.04).
(Table 1) shows PFT data at baseline and 30 min after supraclavicular block. PFT values did not decrease over time in either paralyzed or nonparalyzed subjects.
Sensory and motor block of the upper extremity was produced in all subjects (Table 2). Onset of sensory block at the median, ulnar, and axillary distributions occurred at 23 +/- 5, 21 +/- 4, and 24 +/- 6 min, respectively. Duration of sensory block at these distributions was 85 +/- 19, 106 +/- 15, and 86 +/- 27 min, respectively. Motor block occurred at all four motor distributions examined (P < 0.0001). A representative time course of motor blockade is displayed in Figure 3.
There was no correlation between diaphragmatic paralysis (by ultrasound or RIP) and sensory or motor blockade of any individual nerve (Table 2). Diaphragmatic block was shorter in duration than sensory and motor block (Figure 4).
This observational study demonstrates three points. 1) Hemidiaphragmatic paresis followed supraclavicular block less frequently (50%; 95% confidence interval 14-86) than the reported 100% incidence after interscalene block . 2) No subject experienced a decline in PFT values or respiratory symptoms. 3) Sensory and motor blockade duration outlasted hemidiaphragmatic paresis.
Our study used sophisticated ultrasound and inductive plethysmography techniques to identify hemidiaphragmatic paresis in four of eight subjects. Our methodology should provide a better measure of diaphragmatic paresis than previous studies using plain chest radiography or fluoroscopy. These studies reported an incidence ranging from 1% (diagnostic method not specified)  to 28% using fluoroscopy  to 67% using fluoroscopy during sniffing  to 75% using plain chest radiography . Although the incidence of hemidiaphragmatic block after supraclavicular block is variable, it does not seem to approach the 100% involvement of interscalene block [1,3]. Moreover, the onset of hemidiaphragmatic paresis is consistently within 15 minutes of block placement; therefore, respiratory complications should be easily recognized before patient discharge.
It is tempting to speculate, as others have , that hemidiaphragmatic paresis is less common after supraclavicular block because local anesthetic is deposited at a more caudad level of the brachial plexus and is less likely to involve C3-5 nerve roots. How hemidiaphragmatic paresis actually occurs, and why supraclavicular block seems different than interscalene block, cannot be answered by our study. Two paralyzed subjects (Subjects 1 and 6) did not manifest sensory and/or motor block in the C5-6 distribution. Conversely, all nonparalyzed subjects had C5-6 motor and sensory changes. Thus, peripheral cervical nerve function is not predictive of hemidiaphragmatic paresis. Indeed, deep and superficial cervical plexus block (C2-4) results in only a 61% incidence of phrenic nerve involvement . We cannot determine from our study whether hemidiaphragmatic paresis results from direct blockade of the phrenic nerve, individual blockade of C3-5 nerve roots, or from some other mechanism.
Pulmonary function tests were not affected in healthy volunteers who experienced hemidiaphragmatic paresis, and, consistent with other reports , none of our subjects had symptoms of respiratory difficulty. These respiratory effects of supraclavicular block seem to be fundamentally different from those observed with interscalene block. Hemidiaphragmatic paresis after interscalene block is associated with a >25% reduction in forced vital capacity and forced expiratory volume in 1 s [2,3], and up to 39% of healthy patients report mild dyspnea or altered respiratory sensation . Furthermore, maintenance of tidal volume, minute volume, and PaCO (2) after interscalene block requires a compensatory increase in respiratory rate and %Thoracic during quiet tidal breathing, which implies an alteration of chest wall mechanics not observed with supraclavicular block . Hemidiaphragmatic paresis in our paralyzed subjects resolved more rapidly than their motor and sensory block. Thus, despite the possibility of hemidiaphragmatic paresis after supraclavicular block, its duration is less than sensory and motor block and is not accompanied by significant respiratory impairment in healthy subjects.
This study has several limitations. Although half of our subjects experienced impairment of diaphragmatic motion, this small study lacks the power to determine true incidence. Furthermore, experience with young, healthy volunteers cannot be extrapolated to patients with underlying pulmonary disease. Indeed, patients with chronic obstructive pulmonary disease have developed large reductions in PFT values after hemidiaphragmatic paresis [2,4]. In contrast to studies of interscalene blocks that used 34-52 mL of 1.5% mepivacaine with epinephrine and bicarbonate , we used only 30 mL of 1.5% lidocaine without epinephrine or bicarbonate. Whereas the addition of epinephrine would be expected to result in a more dense block, increasing the local anesthetic volume from 20 to 45 mL does not influence the incidence of hemidiaphragmatic paresis or pulmonary compromise after interscalene block . We believe that volume would be even less of a factor with the more caudad supraclavicular approach. Finally, our study design does not allow us to comment on hemidiaphragmatic paresis after continuous techniques  or long-acting local anesthetics.
Brachial plexus blockade can be a valuable technique for upper extremity surgery. For instance, interscalene block reduces length of stay and postoperative complications after shoulder arthroscopy , but it may be contraindicated in patients with preexisting lung disease. Is supraclavicular block a reasonable alternative for these patients? Some would argue that the often quoted 0.5%-6% incidence of pneumothorax with the classical Kulenkampff supraclavicular block technique  would contraindicate its use in any patient at risk of pulmonary complications. In our practice, we consider the risk of pneumothorax to be an acceptable <1% using either the subclavian perivascular technique  or a modification of the plumb-bob technique . However, although it is not associated with significant pulmonary impairment in healthy volunteers, hemidiaphragmatic paresis can occur unpredictably after supraclavicular block and may result in significant pulmonary dysfunction in patients with underlying lung disease. Therefore, the risk of pneumothorax notwithstanding, our study does not support supraclavicular block as an acceptable alternative regional anesthetic technique for these patients. This is consistent with previous reports of acute respiratory failure after supraclavicular block  and recommendations to avoid bilateral supraclavicular block in patients with underlying pulmonary disease .
In conclusion, supraclavicular block of the brachial plexus is associated with an unpredictable, but not absolute, risk of brief hemidiaphragmatic paresis, which is not accompanied by clinical evidence of respiratory compromise in healthy subjects.
1. Urmey WF, Talts KH, Sharrock NE. One hundred percent incidence of hemidiaphragmatic paresis associated with interscalene brachial plexus anesthesia as diagnosed by ultrasonography. Anesth Analg 1991;72:498-503.
2. Urmey WF, McDonald M. Hemidiaphragmatic paresis during interscalene brachial plexus block: effects on pulmonary function and chest wall mechanics. Anesth Analg 1992;74:352-7.
3. Pere P, Pitkanen M, Rosenberg PH, et al. Effect of continuous interscalene brachial plexus block on diaphragm motion and ventilatory function. Acta Anaesthesiol Scand 1992;36:53-7.
4. Urmey WF, Gloeggler PJ. Pulmonary function changes during interscalene brachial plexus block: effects of decreasing local anesthetic injection volume. Reg Anesth 1993;18:244-9.
5. Dhuner KG, Mober E, Onne L. Paresis of the phrenic nerve during brachial plexus block analgesia and its importance. Acta Chir Scand 1955;109:53-7.
6. Knoblanche GE. The incidence and aetiology of phrenic nerve blockade associated with supraclavicular brachial plexus block. Anaesth Intensive Care 1979;7:346-9.
7. Winnie AP, Collins VJ. The subclavian perivascular approach of brachial plexus anesthesia. Anesthesiology 1964;25:353-63.
8. Shaw WM. Paralysis of the phrenic nerve during brachial plexus anesthesia. Anesthesiology 1949;10:627-8.
9. Moore DC. Regional block. 4th ed. Springfield, IL: Charles C Thomas, 1965.
10. Davies SC, Hill AL, Holmes RB, et al. Ultrasound quantitation of respiratory organ motion in the upper abdomen. Br J Radiol 1994;67:1096-102.
11. Houston JG, Morris AD, Howie CA, et al. Technical report: quantitative assessment of diaphragmatic movement-a reproducible method using ultrasound. Clin Radiol 1992;46:405-7.
12. Sackner JD, Nixon AJ, Davis B, et al. Non-invasive measurement of ventilation during exercise using a respiratory inductive plethysmograph. Am Rev Respir Dis 1980;122:867-71.
13. Gonzalez H, Haller B, Watson HL, Sackner MA. Accuracy of respiratory inductive plethysmograph over wide range of rib cage and abdominal compartmental contributions to tidal volume in normal subjects and in patients with chronic obstructive pulmonary disease. Am Rev Respir Dis 1984;130:171-4.
14. Fujimura N, Namba H, Tsunoda K, et al. Effect of hemidiaphragmatic paresis caused by interscalene brachial plexus block on breathing pattern, chest wall mechanics, and arterial blood gases. Anesth Analg 1995;81:962-6.
15. Brown DL, Cahill DR, Bridenbaugh LD. Supraclavicular nerve block: anatomic analysis of a method to prevent pneumothorax. Anesth Analg 1993;76:530-4.
16. Houston JG, Fleet M, Cowan MD, McMillan NC. Comparison of ultrasound with fluoroscopy in the assessment of suspected hemidiaphragmatic movement abnormality. Clin Radiol 1995;50:95-8.
17. Castresana M, Masters R, Castresana E, et al. Incidence and clinical significance of hemidiaphragmatic paresis in patients undergoing carotid endarterectomy during cervical plexus block anesthesia. J Neurosurg Anesth 1994;6:21-3.
18. Brown AR, Weiss R, Greenberg C, et al. Interscalene block for shoulder arthroscopy: comparison with general anesthesia. Arthroscopy 1993;9:295-300.
19. Concepcion M. Acute complications and side effects of regional anesthesia. In: Brown DL, ed. Regional anesthesia and analgesia. Philadelphia: WB Saunders, 1996:446-61.
20. Hood J, Knoblanche G. Respiratory failure following brachial plexus block [letter]. Anaesth Intensive Care 1979;7:285-6.