Integration of baroreflex and autoregulation control of bronchial blood flow in awake dogs.

AIM: Baroreflex control of the bronchial circulation is unresolved. Early studies suggested that baroreflexes dilate or have no effect, but recent studies in awake dogs suggested baroreflexes did not normally engage tonic vasoconstrictor efferents but during excitement systemic pressure rises may also trigger local sensory-motor dilator reflexes. We examined the postulate that bronchial flow is normally regulated at rest during controlled changes in pressure gradient (Pg) by integration of tonic autonomic activity with autoregulation. METHODS: Twelve greyhounds were instrumented under general anaesthesia by surgical implantation of pulsed Doppler flow transducers on the right bronchial artery (BA). After recovery baroreflex effects were evoked by raising and lowering aortic pressure using a lower thoracic aortic balloon in 11 animals, and in six of these after cholinoceptor plus adrenoceptor blockade. RESULTS: The right BA bed showed pressure-passive responses and the time-dependent bronchial bed effects in the autonomically intact state (INT) were largely similar to those in the blocked state (TAB). When results were replotted as pressure-flow relationships and analysed using covariance, the regression line over the pressure range 70-135 mmHg for TAB demonstrated a significant slope (P < 0.05), a linear regression elevated 120% (P = 0.006) above and parallel to INT (Y(Q) = 0.034 + 0.00033(X(Pg) - 104.6). The regression fell on the line of equal proportional change. CONCLUSION: Baroreflexes do not functionally engage the autonomic outflow to the bronchial circulation. Under controlled conditions of systemic Pg change, the bronchial circulation is normally controlled by the integration of resting autonomic tone, myogenic autoregulation and pressure-passive effects.

Autonomic control of bronchial blood flow and airway dimensions during strenuous exercise in sheep.

BACKGROUND: During exercise and recovery the transient and steady-state changes in autonomic activity regulating lower airway blood flow and dimensions are unknown. The aim of this study was to define changes in bronchial blood flow (Q(br)) and dimensions during moderate and strenuous exercise, and to analyse the role of vagal and sympathetic nerves. METHODS: Nine ewes (34-44kg) underwent left thoracotomy during general anaesthesia (thiopentone/isoflurane) and either (5 sheep=Group 1) a pulsed Doppler transducer was placed on the bronchial artery, or (4 sheep=Group 2) a pulsed Doppler transducer was placed on the bronchial artery, and transit-time and single crystal sonomicrometers were mounted on the left main bronchus. These measured continuously Q(br), bronchial circumference (Circ(br)) and wall thickness (Th(br)). Aortic pressure (P(a)) and central venous pressure catheters were placed in the superficial cervical artery and vein. Trained sheep exercised on a horizontal treadmill, i.e. Group 1, moderate exercise 2.2mph over 1.6, 6min recovery, for analysis of changes in Q(br) before and after cholinoceptor blockade; Group 2, strenuous exercise 4.4mph over 2, 10min recovery for analysis of changes in Q(br) and airway dimensions, before and after cholinoceptor plus alpha(1)-, alpha(2)-adrenoceptor blockade. beta-adrenoceptor systems were intact. RESULTS: In Group 1 during moderate exercise P(a) and heart rate (HR) rose. Q(br) and blood flow conductance (C(br)) fell immediately to 83% (P<0.001) before returning toward resting levels, but fell when exercise ceased to 89% (P<0.01) before recovering. Prior cholinoceptor blockade abolished the immediate fall in Q(br) and C(br), but not the recovery vasoconstriction. Later in recovery the bronchial bed dilated progressively over 6min (P<0.05). In Group 2 during strenuous exercise P(a) and HR rose substantially. Q(br) and C(br) fell to 68% and 54% (P<0.001), respectively, and there was early vasoconstriction in recovery. Circ(br) fell immediately and remained at 93% (P<0.01), and did not recover fully when exercise ceased. Th(br) did not change during or after exercise. Prior cholinoceptor plus alpha-adrenoceptor block caused P(a) and Q(br) to fall slightly during exercise, but the bronchovascular constriction during and after exercise was abolished, as was circumferential shortening in the airway. CONCLUSIONS: At exercise onset and steady-state, resetting the arterial baroreflex upward in sheep increases parasympathetic cholinergic vasoconstrictor activity and causes bronchial wall and bronchovascular smooth muscle contraction in concert with sympathetic adrenergic constriction of systemic vascular beds. Whether the known sigmoid baroreflex control of tracheal smooth muscle tension at rest is extended to tracheobronchial smooth muscle and its circulation during exercise is yet to be determined.

Effects of graded exercise on bronchial blood flow and airway dimensions in sheep.

Exercise stimulus-response relationships for airway blood supply and dimensions have not been described in mammalian species. These relationships are vital for postulates concerning integrated reflex factors normally controlling the airways and which may underlie the asthma syndromes of exercise. This study defines airways stimulus-response relationships in exercising sheep. Ewes between 35 and 40kg were instrumented at left thoracotomy under thiopentone/isoflurane general anaesthesia. Pulsed Doppler ultrasonic transducers were mounted on the bronchial artery, and transit-time plus single-crystal sonomicrometers on the left main bronchus. These recorded simultaneously and continuously bronchial blood flow (Q(br)) and conductance (C(br)), bronchial circumference (Circ(br)) and wall thickness (Th(br)). In Protocol 1 (P1), four sheep ran duplicate 5min protocols on a horizontal treadmill at continuous step-up-and-down speeds of 1min duration, namely, 0.8, 1.6, 2.2, 1.6 and 0.8mph (moderate exercise), followed by 10min recovery. In P2, four sheep ran duplicate 2min protocols at constant 4mph (strenuous exercise), and in P3, one sheep ran duplicate protocols each of 3min at 2.2, 4.4 and 6mph (severe exercise). Regression analysis and repeated measures ANOVA were used to assess differences between times, runs and exercise intensity. In P1, airway effects were directly related to graded exercise effort sustained over 5min. Peak effects occurred at 2.2mph, except for Th(br). Heart rate and P(a) rose (to 156% and 111% of resting, respectively), and Q(br) and C(br) fell (to 83% and 75%; both P<0.001). Circ(br) fell to 96% (P=0.02), and Th(br) rose at low speeds early and late, and thinned at the highest speed. In P2 and P3 for all variables the steady-state effects were systematically greater than for P1 (4.4mph: C(br) to 43%, Circ(br) to 93%; 6.6mph: C(br) to 25%, Circ(br) to 82%). There was no significant recovery hyperaemia, but there was residual post-exercise bronchoconstriction. The exercise stimulus-response relationships from rest to a maximal 6mph for sheep airway circumference and its bronchial circulation are inverse and functionally constrictor.

Analysis of bronchovascular downstream blood pressure changes in exercising sheep.

The relative roles of neural and pressure gradient factors, causing a fall or maintenance of bronchial blood flow in exercising sheep, are unknown. These were examined in sheep prepared under thiopentone/isoflurane general anaesthesia with a pulsed Doppler probe mounted on the bronchial artery, and aortic pressure (Pa) catheter in superficial cervical artery. After recovery, Swan-Ganz catheters were inserted under local anaesthesia into the pulmonary artery. Bronchial flow and conductance (Qbr, Cbr), and pressure gradients (Pg; i.e. aortic minus right atrial, Pg_RAP; pulmonary artery, Pg_Ppa; and, left atrial (wedge) Pg_LAP) were derived from continuous records, after switching between downstream sites during and after moderately severe treadmill exercise (3.8 km.h(-1), for 1.7 min, 6 min recovery). The protocol was repeated after combined alpha1,alpha2-adrenoceptor/cholinoceptor blockade using phentolamine methanesulfonate and methscopolamine bromide. Bronchial flow fell in both receptor intact (INT) and (BL) blocked state. Pa rose in INT, but downstream pressures rose only 3.7 (RAP), 2.8 (Ppa) and 2.0 (LAP) mmHg (P for each < 0.05) in both INT and BL. Pg_RAP and Pg_Ppa did not rise, but Pg_LAP rose 4.0 mmHg (P < 0.05). In BL, Pa fell, as did Pg_RAP (7.0 mmHg, P < 0.05), Pg_Ppa (8.9 mmHg, P < 0.001), but Pg_LAP did not change. Thus, downstream pressures change by small amounts, and pressure gradients to RAP and Ppa sites do not change during moderately severe exercise in normal sheep. The fall in Qbr in INT is due to neural factors, but in BL is due to a fall in Pg. The relative rise in Pg_LAP in both INT and BL favours redistribution within total Qbr to the pulmonary capillary/vein/left atrium site.

Both central command and exercise pressor reflex reset carotid sinus baroreflex.

In decerebrate unanesthetized cats, we determined whether either "central command," the exercise pressor reflex, or the muscle mechanoreceptor reflex reset the carotid baroreflex. Both carotid sinuses were vascularly isolated, and the carotid baroreceptors were stimulated with pulsatile pressure. Carotid baroreflex function curves were determined for aortic pressure, heart rate, and renal vascular conductance. Central command was evoked by electrical stimulation of the mesencephalic locomotor region (MLR) in cats that were paralyzed. The exercise pressor reflex was evoked by statically contracting the triceps surae muscles in cats that were not paralyzed. Likewise, the muscle mechanoreceptor reflex was evoked by stretching the calcaneal tendon in cats that were not paralyzed. We found that each of the three maneuvers shifted upward the linear relationship between carotid sinus pressure and aortic pressure and heart rate. Each of the maneuvers, however, had no effect on the slope of these baroreflex function curves. Our findings show that central command arising from the MLR as well as the exercise pressor reflex are capable of resetting the carotid baroreflex.

Heart-lung interactions: the sigh and autonomic control in the bronchial and coronary circulations.

1. The Darwin hypothesis that human and animal expressions of emotion are the product of evolution and are tied to patterns of autonomic activity specified to progress the emotion remains under challenge. 2. The sigh is a respiratory behaviour linked with emotional expression in animals and humans from birth to death. The aim of the present study was to explore Darwin's hypothesis with respect to tied autonomic activity underlying sigh-induced changes in the bronchial and coronary circulations. 3. Awake dogs were prepared using pulsed ultrasonic flow probes on the right bronchial artery, parent intercostal artery and brachial artery, or on the right, circumflex and anterior descending coronary arteries. Central venous (CVP) and arterial pressures (AP) were measured; heart rate and flow conductances were derived. Three spontaneous sighs were monitored before and during random blockade of individual and combinations of cholinoceptors, alpha-adrenoceptors and beta-adrenoceptors using methscopolamine, phentolamine and propranolol infusions. The data were subject to a 2(3) factorial analysis. 4. A spontaneous sigh is marked by a transient fall and return (< 3 s) in CVP of 18 mmHg (from 4 +/- 1 to -14 +/- 2 mmHg), usually followed by apnoea lasting 23 +/- 2 s. There is an immediate tachycardia and small rise in AP (phase 1) then, during apnoea, bradycardia and a fall in AP (phase 2). During phase 2, bronchial and coronary blood flow and conductance rise two- to three-fold over 30s (peak at 8s). The vascular changes are absent in parent intercostal and brachial beds.

Effect of fentanyl on baroreflex control of circumflex coronary conductance.

1. Fentanyl, a synthetic mciro-opioid receptor agonist, is the preferred induction and maintenance anaesthetic agent in cardiac surgery. 2. Its actions on myocardial blood flow are poorly understood. There are reports of intra-operative myocardial ischaemia. Its reported actions on cardiorespiratory control vary widely, but do involve hypertension, bradycardia and peripheral vasoconstriction. 3. Accordingly, the postulate that fentanyl would cause coronary vasoconstriction and myocardial disadvantage was examined in awake dogs with a continuous wave Doppler flow probe mounted on the circumflex coronary artery. 4. Continuous intravenous infusion of fentanyl citrate (550 ng/kg per min) raised plasma concentrations of fentanyl to 3.37 ng/mL in a linear fashion at 20 min. There was a fall in core temperature of 0.7 degrees C and, although no apparent depression of ventilation or fall in arterial or coronary sinus PO2, there was a rise in PCO2 and H+ concentration. Some dogs salivated and panted transiently. Thus, fentanyl may reset temperature regulation in low doses but, at higher doses, is associated with metabolic acidosis. 5. In sinus rhythm, the arterial pressure of the dogs fell slightly, then rose to 115% of resting control. Circumflex flow and conductance rose early, then conductance steadily declined to 83%. Heart rate fell, then rose before returning to pre-infusion levels. The early circumflex coronary vasodilator effects, but not the later vasoconstrictor effects, were reduced in dogs with paced hearts. 6. In dogs with paced hearts, a dose-effect study using 138, 275, 550 and 1100 ng/kg per min fentanyl suggested that, at low plasma concentrations of 1-2 ng/mL, vasodilatation does occur in both coronary and systemic circulations; however, at higher doses, intense coronary and systemic vasoconstriction supervenes. 7. The dose-response effect of fentanyl on arterial baroreflex control of circumflex conductance was examined during the immediate 8 s circumflex vasodilator response to a step rise in aortic pressure caused by inflation of an intra-aortic balloon. At low plasma concentrations of fentanyl, baroreflex control of circumflex conductance appears to be enhanced but, with increasing plasma concentrations of fentanyl, appears to be depressed. 8. Therefore, the effects of fentanyl are dose dependent. At low plasma concentrations, left ventricular blood flow and its baroreflex control is enhanced but, at higher concentrations, it is depressed.

Bronchovascular role in pulmonary congestion.

1. A postulated role for the bronchial circulation in the development of pulmonary congestion may be based on recent studies of bronchovascular control. 2. The bronchial circulation is the nutrient blood supply of the conducting airways and, therefore, plays an important role in the function of the bronchial mucosa. Mucosal swelling secondary to elevation of mucosal capillary hydrostatic pressure may decrease airway calibre, increase resistance to airflow and precipitate symptoms of pulmonary congestion. 3. Resting mucosal capillary hydrostatic pressure is relatively constant due to autoregulation of bronchial blood flow and is maintained low by nett bronchovascular constriction due to the dominance of autonomic vasoconstriction over nitric oxidedependent vasodilatation. 4. Bronchial blood flow is also regulated by cardiac afferent reflexes. Stimulation of cardiac vagal and spinal afferents produces vasodilatation and vasoconstriction, respectively. Tonic activity of cardiac spinal afferents probably contributes to the resting autonomic vasoconstriction. 5. Therefore, mild heart failure, which is associated with abnormal cardiovascular reflex function, may decrease cardiac spinal afferent-mediated bronchial vasoconstriction and produce active dilatation due to stimulation of cardiac vagal afferents by excessive myocardial stretch, leading to bronchial mucosal swelling and pulmonary congestion.

Autonomic control of bronchial circulation in awake sheep during rest and behaviour.

1. We tested the hypothesis that the pattern and the intensity of autonomic mechanisms causing vasoconstriction in the resting bronchial circulation of awake dogs also exists in awake sheep. It was also postulated that sighing behaviour and the associated bronchovascular dilatation induced by non-adrenergic, non-cholinergic (NANC) mechanisms observed in the dog exist in sheep. 2. Bronchial arterial blood flow to lower airways of both lungs of awake sheep was measured continuously using pulsed Doppler flow probes mounted on the bronchial artery at prior thoracotomy. 3. Cumulative and factorial analysis of responses to randomized combinations of autonomic alpha 1-, alpha 2-, beta 1- and beta 2-adrenoceptors and cholinoceptor autonomic blockade suggests that resting vasoconstrictor activity is less in sheep than in dogs. At normal aortic pressure, the autonomic activity of these receptor groups in the sheep lowers bronchial blood flow and conductance by 30%, whereas in the awake dog, the corresponding autonomic effect is 50%. 4. Tonic autonomic control of bronchial conductance can be partitioned in sheep to show significant and separate alpha- and beta-adrenoceptor vasoconstrictor activity at a ratio of 1.8:1, an effect normally offset by a weaker vasodilator alpha-/beta-adrenoceptor interaction. In contrast to the situation in awake dogs, cholinoceptors do not play a role in awake sheep. 5. Nitric oxide (NO) synthase inhibition in sheep using NG-nitro-L-arginine following blockade of alpha- and beta-adrenoceptors and cholinoceptors causes hypertension, but minor changes, if any, in pulmonary pressures or heart rate. Bronchial flow and conductance, however, fall from a higher resting conductance by approximately 50%, suggesting that, normally, resting bronchial flow conductance is dominated by strong tonic NO vasodilator effects that interact with weaker tonic autonomic vasoconstrictor effects. 6. Superimposed (respiratory) behaviours of sighing, sneezing and coughing, which involve negative swings in intrathoracic pressure and the movement of inspired air, evoke large active bronchovascular dilator effects. These appear to be largely NANC in origin and appear to be dependent, in part, on mechanisms associated with NO release. It is postulated that the C-fibre axon reflex using substance P, calcitonin gene-related peptide and neurokinin A may be involved. Vocalization and eructation do not evoke bronchovascular effects.

Shift of the posteroinferior aspect of the capsule for recurrent posterior glenohumeral instability.

Thirty-five shoulders in thirty-four patients were treated with a superior shift of the posteroinferior aspect of the capsule because of recurrent posterior glenohumeral subluxation and dislocation. The physical examination revealed three types of posterior instability in these patients preoperatively: unidirectional (six shoulders), bidirectional (posterior and inferior) (seven shoulders), and multidirectional (posterior and inferior dislocation with anterior subluxation) (twenty-two shoulders). Eleven shoulders had had previous operative procedures. At the time of the index operation, the most common abnormal findings in these shoulders were capsular redundancy and excessive volume of the glenohumeral joint. Complete detachment of the posterior aspect of the labrum was found in only four shoulders. There was no excessive glenoid retroversion in these patients. All thirty-four patients were available for follow-up at an average of five years (range, two to twelve and a half years) postoperatively. Over-all, the result for seventeen of the thirty-five shoulders was rated as excellent; eleven, as good; one, as fair; and six, as poor. Four shoulders became unstable again. Six of the seven unsatisfactory results were in shoulders that had had previous attempts at stabilization. A successful result was achieved in twenty-three of the twenty-four shoulders in which the superior shift of the posteroinferior aspect of the capsule was the initial repair.

Surgical treatment of complete acromioclavicular dislocations.

Forty-four patients, ranging in age from 17 to 57 years (average, 32), were evaluated an average of 4 years (range, 2 to 9) after surgical reconstruction for Allman-Tossy Grade III acromioclavicular dislocations. Twenty-seven patients underwent repair for acute injuries (< 3 weeks after injury) and 17 patients underwent reconstructions for chronic injuries (> 3 weeks). Coracoclavicular fixation with heavy nonabsorbable sutures was used to correct superior displacement in all cases. In addition, transfer of the coracoacromial ligament to the distal clavicle was performed in 15 of the 27 early repairs and 17 of the 17 late reconstructions. Overall, 26 of 27 (96%) early repairs and 13 of 17 (77%) late reconstructions achieved satisfactory results. There was a trend for better results and return to sports or heavy labor with early repairs; however, this was not statistically significant (P = 0.065). When the results of early repairs were compared with those of late reconstructions performed more than 3 months after injury, the results of the shoulders undergoing early repair were significantly better (P < 0.01). Overall, 39 of 44 (89%) patients achieved a satisfactory result. Surgical reconstruction for acromioclavicular dislocation provides reliable results including use of the arm for sports or repetitive work.

Isolated nerve injuries about the shoulder.

Infraclavicular nerve injuries are rare and potentially disabling problems. A retrospective study of 24 patients with 28 nerve injuries is presented, including 18 axillary, 7 suprascapular, and 3 musculocutaneous nerve injuries. Vague shoulder pain and weakness of the involved muscle groups were the main symptoms of nerve injury in these patients. All patients had atrophy of the specific muscles involved. Diagnosis of these nerve lesions can often be difficult because of this vague presentation. Followup from date of injury averaged 60 months and included evaluation by questionnaire, repeat physical examinations, and serial electromyograms. There were 21 complete or satisfactory nerve recoveries, while 7 patients had unsatisfactory results. The etiology of the injury appeared to be an important factor with respect to outcome. Eight of 10 nerve injuries secondary to blunt trauma went on to complete recovery, and 4 of 6 nerve injuries secondary to shoulder dislocation recovered completely. None of the 7 nerves injured during surgery recovered completely. No patient with spontaneous onset of nerve dysfunction had an unsatisfactory result. Poor results were noted in patients with initial total denervation as shown by electromyogram and in patients with intraoperative nerve damage.

Operative treatment of failed repairs of the rotator cuff.

Thirty-one patients had a repeat repair after a failure of a previously performed repair of the rotator cuff, and the causes of the original failure and the results of the repeat repair were analyzed. The clinical situation was complex, with multiple factors contributing to the failures. A large or massive tear of the cuff in thirty patients (97 per cent) and persistent subacromial impingement in twenty-eight patients (90 per cent) were the most common factors associated with failure. The over-all result of the repeat repair was satisfactory in sixteen patients (52 per cent) and unsatisfactory in fifteen (48 per cent). Twenty-five patients (81 per cent) had satisfactory relief of pain. However, fourteen patients (45 per cent) had persistent weakness that led to an unsatisfactory result. The factors that were associated with a better result were an intact acromion, an intact origin of the deltoid muscle, and the finding that the remaining tissue of the rotator cuff was of good quality. The factors that were associated with an inferior result were a previous lateral acromionectomy, a previously detached origin of the deltoid muscle, and the finding that the tissue of the rotator cuff that was available at the time of the repeat operation was of poor quality. Persistent pain is the primary indication for a repeat repair. The functional results are not as predictable, especially when the tissue of the cuff is poor and the deltoid origin has been detached previously.

Open reduction and internal fixation of two- and three-part displaced surgical neck fractures of the proximal humerus.

Twenty-two patients aged 20 to 82 years (average 56 years) were followed for 1.1 to 8.9 years (average 3.3 years) after open reduction and internal fixation of two- and three-part displaced surgical neck fractures of the proximal humerus. There were 14 two-part displaced surgical neck fractures, seven three-part displaced greater tuberosity and surgical neck fractures, and one three-part displaced lesser tuberosity and surgical neck fracture. Fixation was achieved with heavy nonabsorbable sutures or wire that incorporated the rotator cuff tendons, tuberosities, and shaft. In cases with significant surgical neck comminution, humeral Enders nails were incorporated in a tension-band construct to provide longitudinal stability. Eighteen (82%) of the 22 patients had good or excellent results. Three (14%) of the 22 had satisfactory results, and one (5%) had an unsatisfactory result. The use of a technique of limited internal fixation for these displaced fractures without the use of plates and screws achieved fracture stability and a high percentage of acceptable results.

Prosthetic replacement in rotator cuff-deficient shoulders.

We reviewed a series of 30 shoulders in 25 patients who had glenohumeral arthritis and rotator cuff deficiency and who underwent prosthetic replacement. Nineteen shoulders underwent humeral head replacement, and 11 shoulders had total shoulder arthroplasty. Meticulous mobilization and reconstruction of the deficiencies in the thin atrophic rotator cuff tissues were attempted in all shoulders. Emphasis was placed on anteroposterior stability, and this was achieved in all shoulders; superior coverage was fully achieved in 15 shoulders and was partially achieved in 11. All shoulders had less pain after surgery, and 93% achieved satisfactory pain relief. Total shoulder arthroplasty and humeral hemiarthroplasty were found to provide similar results with respect to pain relief, functional improvement, and patient satisfaction. Shoulders with hemiarthroplasty gained significantly more active elevation (+52° vs + 2°) after surgery. Cuff repair was easier when a humeral head prosthesis alone was used because less lateralization of the humerus occurred. Also, operative time, anesthesia time, and blood loss were decreased with hemiarthroplasty. Because the lack of glenoid resurfacing did not adversely affect pain relief or function and avoided the potential problem of glenoid loosening, we favor humeral hemiarthroplasty as a treatment for glenohumeral arthritis in the rotator cuff-deficient shoulder.

Open reduction and internal fixation of two-part displaced fractures of the greater tuberosity of the proximal part of the humerus.

Twelve patients, ranging in age from thirty-four to seventy-two years (average, fifty-three years), were evaluated an average of five years (range, two to eight years) after open reduction and internal fixation of a two-part displaced fracture of the greater tuberosity of the proximal part of the humerus. The indication for operative reduction was one centimeter or more of displacement of the fracture as seen on the diagnostic radiographs. The anterosuperior deltoid-splitting approach, combined with rotation of the humerus, allowed adequate exposure of the retracted tuberosity. Internal fixation of the greater tuberosity with heavy, non-absorbable sutures and careful repair of the rotator cuff permitted early passive motion. All fractures healed without postoperative displacement. Six patients had an excellent result and six had a good result; active elevation averaged 170 degrees. There was one partial, transient palsy of the axillary nerve.

Anterior acromioplasty for subacromial impingement in patients younger than 40 years of age.

The indications and results of anterior acromioplasty in patients older than 40 years of age are well documented; however, little is reported concerning this procedure in younger patients because the majority are treated conservatively. Over the last seven years, 26 patients younger than 40 years of age were treated by anterior acromioplasty for subacromial impingement resistant to conservative therapy. The indication for surgery was failed conservative therapy in 17 patients. Six patients had impingement despite previous surgery, such as failed acromioplasty, biceps tenolysis, and biceps tenodesis. Two patients had impingement secondary to greater tuberosity malunion. The average duration of symptoms was 27 months and of conservative therapy, 17 months. All patients had a bursectomy and anterior acromioplasty. In addition, eight had a complete acromioclavicular arthroplasty for concurrent acromioclavicular arthritis, four had a modified acromioclavicular arthroplasty for impingement under the acromioclavicular joint, three had calcium excisions, and one had a rotator cuff repair. The greater tuberosity malunions were advanced in one patient and excised in one. The results were 81% (21) excellent and good, 15% (four) fair, and 4% (one) poor. The average follow-up period was 33 months. The majority of younger patients with subacromial impingement can be successfully treated conservatively. In those patients with persistent symptoms, open acromioplasty can reliably give good results with regard to pain and function.