The carpal tunnel syndrome is the most common entrapment neuropathy. With the report by Cannon and Love in 1946 of the favourable outcome in nine patients with median nerve entrapment at the wrist who underwent carpal tunnel release and neurolysis, the syndrome became increasingly recognized. During the next two decades, Carpal tunnel release is now the most common hand operation performed. Though the procedure is generally associated with low morbidity and relatively high success rates, failure of the surgeon to fully understand the anatomy, pathophysiology, and typical features of carpal tunnel syndrome, as well as the many pitfalls associated with its diagnosis and treatment, may lead to an unacceptable incidence of suboptimal results.
The history alone usually establishes the diagnosis of carpal tunnel syndrome. Abnormal findings on neurologic examination may support the diagnosis in patients with less typical symptoms, though objective abnormalities are generally sparse except in more advanced cases. Hypoesthesia in the median nerve distribution may be found, except over the thenar eminence and base of the palm. This is due to the palmar cutaneous branch of the median nerve arising proximal to and passing superficial to the transverse carpal ligament. This sensory branch is frequently spared the effects of entrapment within the carpal tunnel. Motor deficits are seen less frequently. The two most important muscles innervated by the distal median nerve are the opponens pollicis and the abductor pollicis brevis. The first is tested by having the patient oppose the thumb to the palm and draw it medially toward the base of the fifth digit against resistance. The latter is tested by resisting active abduction of the thumb away from the plane of the palm. Significant and long-standing denervation to these muscles leads to atrophy of the thenar eminence.
Two-thirds of patients will experience electrical sensations radiating into the palm and first three digits when the median nerve at the wrist crease is percussed. This is known as Tinel s sign, which classically has been associated with median nerve entrapment at the wrist; however, recent reports suggest this test to be of dubious value in establishing the diagnosis of carpal tunnel syndrome due to a high incidence of false-positive results in otherwise asymptomatic individuals. A more accurate predictor of carpal tunnel syndrome is Phalen‘s wrist-flexion test. The patient is asked to hold the forearms up in a vertical orientation with the wrists flexed for 60 seconds. Reproduction of the patient's painful dysesthesias provides a high degree of certainty of the diagnosis. Likewise, inflating a blood pressure cuff placed around the arm may reproduce the symptoms. Again, symptom exacerbation in this test is likely due to venous distention within the rigid confines of the carpal tunnel.
Electromyography and nerve conduction velocities should be obtained to confirm, not to establish, the diagnosis of carpal tunnel syndrome. Indications for surgery should not rest solely on the results of this test, but rather should be based on clinical judgment. Electrodiagnostic studies are helpful in differentiating the carpal tunnel syndrome from other disorders, such as cervical nerve root impingement or syndromes of the thoracic outlet.
The most sensitive and the earliest abnormality found is a prolonged sensory conduction latency across the wrist. Normally, the distal latency through the carpal tunnel to the abductor pollicis brevis is less than 4.5 milliseconds. A prolonged motor latency generally occurs later in the entrapment process. The amplitude of the action potential is frequently diminished. Denervation potentials in the opponens pollicis and abductor pollicis brevis indicate advanced and probably irreversible damage to the median nerve.
Nerve conduction velocities and latencies are subject to a number of physiologic variables, such as the age and metabolic status of the patient, as well as the temperature, vascular supply, and extent of edema in the arm. Numerous technical problems are associated with the determination itself. The treating physician or surgeon is responsible for being aware of the variability of the test and for assessing the results in light of the patient's clinical evaluation. Should the electrodiagnostic studies be equivocal, it might be prudent to wait up to 4-6 weeks to repeat the study before embarking on a course of surgical management. Though electrical abnormalities may not be evident in up to 10% of clinical cases of carpal tunnel syndrome, many surgeons will not consider decompression of the carpal tunnel without electrodiagnostic confirmation.
The surgeon should be aware of the similarity between carpal tunnel syndrome and other pathologic processes that cause pain and neurologic dysfunction of the radial aspect of the forearm and hand. These include brachial plexopathy from tumor, trauma, or inflammation, and, less often, thoracic outlet syndrome. The most common of these, however, is C6 and C7 radiculopathy caused by cervical disk herniation or spondylosis. Typically, this pain seems to originate in the neck and shoulder and radiates down the arm in a lancinating fashion into the radial aspect of the hand. Exacerbation of the pain with movement of the cervical spine is a significant diagnostic factor in this pain syndrome. In general, C6 and C7 nerve root compression produces motor deficits in the upper arm, such as biceps or triceps weakness, and diminished deep tendon reflexes. The intrinsic hand muscles are relatively unaffected.
The median nerve may become entrapped at locations other than the carpal tunnel. The pronator syndrome is produced by median nerve compression at one of a number of locations around the distal humerus, elbow, and proximal forearm. Entrapment at these sites produces pain on the volar surface of the forearm and hypoesthesia of the radial half of the hand. Weakness of the thenar muscles is observed less often. Symptoms are usually aggravated by exertion, especially with forceful flexion of the elbow or pronation of the forearm. Phalen's wrist-flexion test is typically negative. Likewise, the anterior interosseous syndrome causes pain in the proximal forearm. This pain is exacerbated by exercise and relieved by rest. Because it affects the anterior interosseous branch distal to where it leaves the median nerve in the cubital fossa, the motor and sensory innervation of the hand is not directly affected, though pain may be referred into the hand.
Carpal tunnel syndrome may coexist with other lesions of the nerve roots, brachial plexus, or median nerve. Concurrent cervical radiculopathy has been found in over 10% of electrically proven carpal tunnel syndrome. This is referred to as the double-crush syndrome. This putative syndrome is based on the concept that proximal compression of a nerve may weaken the nerve's ability to withstand a more distal compression.
Many cases of carpal tunnel syndrome, especially mild cases or those that present early in the evolution of the disease process, are self-limiting and resolve spontaneously without need for surgical intervention. In cases related to systemic illnesses, such as hypothyroidism or acromegaly, treatment of the underlying illness may result in improvement or even resolution of entrapment symptoms.
The occurrence of the carpal tunnel syndrome in pregnancy is thought to be related to fluid retention in connective tissues about the wrist. Nocturnal and exertional dysesthesias in the radial half of the palm occur in 10% to 25% of pregnant women. When the carpal tunnel syndrome occurs, the symptoms are more often bilateral. Onset of symptoms is typically during the third trimester. Relief occurs spontaneously, within a few weeks of delivery, in the majority of cases. Because of its transient nature, carpal tunnel syndrome during pregnancy is best treated by using conservative measures, such as splinting and analgesics. In the rare case of severe pain refractory to nonoperative therapy, it is reasonable to proceed with carpal tunnel decompression using a local anesthetic. That subsequent pregnancies are frequently associated with repeated episodes of carpal tunnel syndrome confirms the association of carpal tunnel syndrome and pregnancy.
Short-term immobilization of the wrist by splinting is among the most commonly used of nonoperative therapies. In general, the splint is worn only at night, though patients with diurnal dysesthetic pain should wear the splint at all times. It is important that the splint be constructed such that no pressure overlies the median nerve at the wrist. In the author's experience, if symptoms are not alleviated or improved after 6-8 weeks of splinting, further conservative management is unproductive. The patient is then offered the option of surgical decompression.
Diuretics may prove helpful in patients with carpal tunnel syndrome related to excessive fluid retention, such as is observed in congestive heart failure. Control of hyperglycemia in diabetics and weight loss in obese patients also may be of benefit in alleviating symptoms. Adequate analgesia may be obtained from nonsteroidal anti-inflammatory drugs, though their long-term efficacy in the treatment of carpal tunnel .syndrome has not been determined. Pyridoxine administration has not gained widespread acceptance as a useful therapy. If development of the syndrome is occupation related, an alteration of work activities or even a change of occupation may be necessary.
Much has been written about the use of corticosteroid injections into the carpal tunnel, but few objective reports of positive results are available. Gelberman et al reported their results of a prospective trial of steroid injection and splinting. Patients with mild and early symptoms noted a beneficial effect after 3 weeks; however, those patients with more profound symptoms failed to respond to this therapy. Others feel steroid injections may alleviate symptoms though only temporarily. In the last decades, enthusiasm for this treatment modality has waned. Its primary indication, however, may remain as a means of controlling symptoms during temporary or reversible causes of carpal tunnel syndrome, such as is observed during pregnancy, or following failed carpal tunnel syndrome surgery. Administration of corticosteroids into the carpal tunnel must be precise in order to avoid injury to the median nerve. A 25-gauge needle is inserted into the wrist at a point 1 cm proximal to the flexion, crease between the tendons of the palmaris longus and the flexor carpi radialis and at an angle of 45 degrees to the long axis of the forearm. The needle is advanced approximately 1 cm until the flexor retinaculum is pierced. If painful dysesthesias in the distribution of the median nerve are produced, the needle is withdrawn and reinserted at a slightly different location. One to two millilitres of a mixture of triamcinolone or a similar corticosteroid and 1 % lidocaine are slowly injected. If the patient does not obtain relief following the first injection, further injections, in general, should not be pursued.
The indications for carpal tunnel release are (1) rapidly progressive thenar wasting and hand dysfunction, or (2) substantial symptoms that are unrelieved by conservative measures. The surgeon should be confident of the diagnosis and have thoroughly excluded other causes of hand dysfunction and pain. In cases of bilateral carpal tunnel syndrome, it is rarely necessary to operate on both hands simultaneously. The more severely affected hand (or in the case of symmetric disease, the nondominant hand) should be decompressed first. Surgery on the contralateral hand may be performed 6 weeks later, after the first hand has recovered and has regained full function. Most would agree that the patient undergoing simultaneous bilateral procedures would be rendered functionally impaired and quite dependent, albeit for a short period of time. Often the symptoms in the less severely affected hand will spontaneously resolve in the interim and will not require operative intervention. In Gainer and Nugent's series of 430 cases, only one-third of patients with bilateral carpal tunnel syndrome required bilateral operations. When bilateral surgery was needed, an interval between operations of at least 3 weeks was allowed.
Rarely does a patient require general anaesthesia to undergo carpal tunnel release. The author prefers a regional block or, if the anaesthesiologist is not well versed in this technique, a locally administered anesthetic agent. The use of a tourniquet is unnecessary. There are as many methods of incision as there are surgeons performing carpal tunnel releases. The type of incision is of little importance as long as it follows three basic guidelines: (1) The incision should be designed to avoid potential division of the palmar cutaneous branch of the median nerve; (2) It should be carried far enough into the palm to confidently divide the most distal aspect of the ligament in its entirety; and (3) If it is to cross the flexion crease of the wrist, it should not do so in a perpendicular fashion. A simple transverse wrist incision, though widely used in the past, is discouraged. This incision does not offer adequate exposure of the deep palm in order to ensure that the thickest and most distal portion of the transverse carpal ligament has been completely divided. In addition, the palmar cutaneous branch of the median nerve may be inadvertently divided, producing numbness over the thenar eminence or pain related to neuroma formation. As has been previously and eloquently stated. Once the incision is made, sharp dissection is carried through the subcutaneous fatty tissue to the underlying anterior antebrachial fascia in the distal forearm and wrist and the palmar aponeurosis in the hand. Loupe magnification and headlamp illumination improve the visualization of anatomic structures. Hemostasis is maintained throughout the procedure by coagulating and dividing the small subcutaneous vessels using bipolar electrocautery. Beneath the superficial fascia, the tendon of the palmaris longus is encountered. The division of the transverse carpal ligament begins just radial to this easily identifiable tendon. Rarely, the palmaris longus tendon may overlie and appear to compress the median nerve when the wrist is extended. The transverse carpal ligament is sharply divided using a No. 15 blade or fine Metzenbaum scissors. The division is performed under direct vision. The incision is made on the ulnar side of the median nerve where the palmar cutaneous and the recurrent motor branches are less likely to be encountered. As the dissection crosses the wrist into the palm, the transverse carpal ligament becomes noticeably tougher and thicker. The muscles of the thenar and hypothenar eminences originate from the transverse ligament at this site.
Deeper in the palm, the fibers of the palmar aponeurosis begin to blend with those of the transverse ligament, thus increasing its thickness and tenacity. Here it is of utmost importance to proceed cautiously and with vigilance for an aberrant course of the recurrent motor branch. A grooved dissector may be placed beneath the ligament in order to guide the knife blade through the tissues. The ligament may be incised likewise with a scissors throughout its' entire length. The decompression is incomplete until the transverse ligament is divided in its entirety. The median nerve itself is then examined for pseudoneuroma formation or compression from adjacent masses, such as neurofibromas or ganglion cysts. The underlying flexor synovialis is also examined; presence of severe tenosynovitis is of prognostic importance. Internal neurolysis or epineurolysis is not routinely performed unless the procedure is a re-exploration with the only significant finding being scarring in and about the nerve. The wound is then closed with a subcutaneous or subcuticular absorbable suture, taking care not to reapproximate the transverse carpal ligament over the median nerve. Nylon suture is used to reapproximate the thick palmar skin. Using absorbable Vicryl subcuticular suture give the feeling of inferior cosmetic results. A bulky dressing is placed and the hand wrapped lightly with an elastic bandage. It is important that the patient be reminded to keep the hand elevated above the level of the heart for at least 24 hours to limit postoperative edema and venous congestion and to move the fingers.
The patient's hand should be bandaged with bulky dressing material in the palm for 1 day after surgery. Active flexion and extension of the digits as well as thumb abduction and opposition are encouraged to prevent the effects of prolonged immobilization. Sutures are removed 10 days after surgery. The patient is allowed to gradually increase the use of the hand but is discouraged from strong gripping or other exertional uses until 3 weeks postoperatively, at which time the patient is allowed to resume full activities. In patients with significant motor deficits, physiotherapy rarely needed after surgery.
Generally, relief of painful dysesthesias occurs almost immediately following carpal tunnel release. Should the pain and tingling persist, the possibility of incomplete division of the transverse carpal ligament or of erroneous diagnosis should be entertained. A deep, aching sensation exacerbated by activity may develop in the thenar and hypothenar eminences and distal forearm related to swelling at the base of the palm. Referred to as "pillar pain," the condition is self-limiting and generally resolves within a few months. Sensory deficits should show definite improvement by 6-12 weeks, though motor deficits are much slower to resolve. In cases of severe denervation and atrophy, complete recovery of motor function should not be expected.
Results of Nonoperative Therapy
One-half obtained relief of their symptoms with nonoperative therapy. In those patients unrelieved by splinting and nonsteroidal analgesics, additional benefit may be obtained from steroidal injections into the carpal ligament; however, the effect is generally only temporary and 65% to 90% of patients undergoing this treatment modality will experience recurrence of symptoms.
Results of Surgery
Relief of pain and improvement in motor and sensory deficits occur in 90 % of patients undergoing carpal tunnel release. Complete or near complete resolution of signs and symptoms was achieved in 82 % of patients with an additional 10% obtaining moderate relief. The remaining 8% achieved little or no relief or were worse. Poor surgical outcomes fall into two main categories: (1) failure to relieve symptoms and (2) adverse effects and complications. The former is more common, probably related to failure to completely divide the transverse carpal ligament or to adequately manage other compressive lesions within the carpal tunnel. A significant number of surgical failures, however, may be due to misdiagnosis, such as mistaking cervical radiculopathy, brachial plexopathy, or diabetic neuropathy for carpal tunnel compression. This may be especially true if surgical indications are based on electrodiagnostic tests alone without appropriate regard for the clinical presentation and physical findings.
Mackinnon classified postoperative failures into four groups: (1) neurologic complications, such as injury to the palmar cutaneous branch and other neural elements, (2) vascular injury, including palmar hematomas, (3) tendon problems, such as bowstringing, tendon adhesions, and trigger finger, and (4) wound complications, including infection, wound dehiscence, and hypertrophic or painful scarring across the wrist crease. Except in cases of incomplete sectioning of the ligament and wound problems, reexploration following failed carpal tunnel release may prove disappointing and of limited benefit.
Avoidance of Complications
Adverse effects and complications of carpal tunnel release are under-reported. Only those authors who deal with a large volume of patients referred for failure of prior decompressions have adequate data for evaluation. Louis et al reported on such a group of patients, 26 of whom had undergone prior carpal tunnel release that resulted in postoperative complications. Fourteen patients had painful neuromas related to division of the palmar cutaneous branch, making this the most common complication. In two patients, failure to relieve symptoms was related to the use of transverse wrist incisions. Painful and hypertrophic scarring from perpendicular wrist incisions was observed in three patients. Two patients experienced stiffness of the interphalangeal joints following prolonged immobilization, two had neuromas of the superficial branch of the radial nerve, and three experienced dysesthesia, possibly secondary to reflex sympathetic dystrophy. In a series of 186 patients reported by MacDonald et al, there were 34 complications in 22 patients. The complications included 12 cases of incomplete division of the transverse carpal ligament (in 8 of these cases, a transverse incision was used), 11 injuries to the palmar cutaneous branch of the median nerve, 4 cases of reflex sympathetic dystrophy, 2 each of hypertrophic scarring (from perpendicular incisions), palmar hematoma, and "bowstringing" of the flexor tendons, and 1 case of adherence of flexor tendons. Superficial wound infections have been reported to occur in 0.5 % to 6% of cases. Fortunately, injury to the recurrent motor branch is a rarely reported complication. Litigation must be kept in mind.
This unique technique deserves additional comment. In 1983, Paine and Polyzoidis described a method of carpal tunnel release using a retinaculotome. In their technique, a small transverse wrist incision is made centred over the palmaris longus tendon. The flexor retinaculum is opened medial to the median nerve and the retinaculotome is inserted. This instrument has a blunt foot plate beneath a sharp vertically oriented blade and is designed to acutely incise the ligament while deflecting the underlying median nerve. The instrument is passed beneath the transverse carpal ligament, dividing the ligament as it is advanced more distally into the palm. The procedure is repeated in a proximal direction to divide the remaining ligament. Adequacy of ligament division is determined by feel and sound, which the authors describe as being quite characteristic. In their review of 516 "closed" procedures using the retinaculotome, 89% of their patients reported satisfactory results. Their few failures were thought to be due to incomplete division of the ligament. Wound hematom as occurred in less than 1 % of the cases. They reported no injuries to the recurrent motor branch.
Pagnanelli and Barrer reported a series of 577 hands decompressed using a slight modification of the Paine and Polyzoidis "closed" Retinaculotomy. Satisfactory results were obtained in 93 % of cases. Complications were minor and few; no injuries to the median nerve or its branches occurred. Length of postoperative recovery appeared to be shorter with this technique as compared to the more traditional "open" procedure, and incisional discomfort was perhaps lessened. Despite the comparable results achieved by closed Retinaculotomy, it has not gained widespread acceptance among surgeons performing carpal tunnel release. However, the limited experience with this innovative technique challenges surgical dogma that the transverse carpal ligament must be divided its entire length under direct visualization. Further experience is needed to evaluate the role of this procedure in the surgical treatment of carpal tunnel syndrome.
Carpal tunnel syndrome is a common affliction, especially among persons whose occupations require repetitive wrist motion. Numerous systemic disease processes can predispose to the development of this disorder. Symptoms are quite characteristic as are physical findings; however, because of occasional similarity between other nerve compression syndromes, electrodiagnostic tests may be necessary to confirm the diagnosis of carpal tunnel syndrome. Early and mild cases are frequently self-limiting and resolve with time or conservative measures. This is especially true with regard to pregnancy-induced carpal tunnel syndrome. Symptoms are adequately managed in one-half of the cases by using wrist splinting, analgesics, physiotherapy, and occasionally steroid injections into the carpal tunnel. In those cases unrelieved by nonoperative means, surgical decompression of the carpal tunnel can be undertaken with an 80% to 90% success rate. Surgical failures can be minimized if care is taken to avoid injury to the palmar cutaneous and recurrent motor branches of the median nerve, and if an incision is made that avoids hypertrophic scarring in the wrist yet allows adequate visualization for division of the distal extent of the transverse carpal ligament. With a proper understanding of the anatomy, pathophysiology, clinical presentation, and available treatment modalities for median nerve entrapment at the wrist, the surgeon can expect to offer substantial relief to the majority of patients afflicted with carpal tunnel syndrome.