This website is intended for healthcare professionals

Phenol Block for Upper Limb Spasticity

Posted in Rehabilitation Articles on 10th Dec 2012


Christopher Roy

Christopher W Roy recently retired as Consultant in Rehabilitation Medicine at the Southern General Hospital, Glasgow, where he had been appointed in 1993. He had previously (since 1988) worked as Senior Lecturer in Rehabilitation Medicine at Wellington School of Medicine, New Zealand. Clinical interests include spasticity, and the locomotor consequences of neurological disease. Research interests include shoulder pain in paralysed people, and measuring the effectiveness of rehabilitation.

Correspondence to:
Email. Christopher Roy

This article follows that of Moheb Gaid, which appeared in a recent edition of ACNR,1 where phenol nerve block of lower limb spasticity was discussed. Here, attention is given to upper limb procedures. Before any focal technique is performed, intercurrent disease as a cause of increased spasticity must be excluded, and intervention at a local rather than systemic level deemed appropriate. This process is similar to preparation for botulinum toxin injection, which has been well described elsewhere.2 This article assumes that these preparatory considerations have been completed.

Although a treatment that has been used since the 1960s,3 the widespread use of botulinum toxin for focal spasticity has reduced the frequency of phenol procedures, particularly in the upper limb; many situations call for the injection of smaller muscles, which are eminently susceptible to botulinum toxin techniques.  In a series conducted in the late 1990’s, Ahmed and colleagues4 found that of 254 consecutive Phenol procedures for spasticity 35% applied to the upper limb (10% biceps, and 25% wrist and/or finger flexor procedures).  Greater availability of botulinum toxin has altered this picture; in a recent period of approximately 18 months in our own department, 283 procedures for treating spasticity were performed using botulinum toxin. Of this, 65% were for upper limb indications, and the remainder lower limb; whereas in the same period, of 196 peripheral phenol procedures, the vast majority applied to the lower limb, with a few procedures only employed in the upper limb, or trunk. Nevertheless, there remain a number of situations where phenol remains a useful option, in view of the (generally) longer duration of effect of phenol, the large dose range that can be used, and the relatively low cost of the medication: typically, sufficient phenol to treat elbow flexors might cost 5% of the botulinum dose required. These reasons are summarised in Table I.

Phenol blockade may be performed either to a nerve trunk, or to motor points of target muscles. The technique of nerve block is generally similar to that of those lower limb procedures described in Gaid’s paper,1 but this paper will explain the technique of motor point blockade, used in many upper limb procedures. This technique also has applicability in some lower limb procedures, and thus will add to Gaid’s discussion of that. These techniques are generally low risk, and have been performed both in hospital and community settings. Although phenol concentrations of 4-8% have been used for neurolysis, and stimulation frequencies of 0.5-4Hz, for simplicity we report here doses needed for 6% aqueous phenol, and 2HZstimulation, as our usual practice.

In principle motor point blockade can be applied to virtually any superficial muscle, but the more common sites of intervention in the upper limb are listed in Table 2, together with those areas where nerve block is preferred.

Upper limb procedures employing nerve blocks as part of an open surgical procedure have been described,5 but this is less commonly used now, and will not be further discussed here. Figure 1 shows a typical patient where upper limb phenol blockade was found helpful.

Figure 1: Severe multifocal upper limb spasticity

Actions of Phenol, and adverse reactions

A comprehensive description of these is detailed in a helpful review by Gracies et al.6 In brief, phenol has transient local anaesthetic actions, but later neurolytic, and in some cases myolytic activity. Gracies cites laboratory work identifying microvascular changes also, but the clinically significance of this is uncertain. As a phenolic, there must be a theoretical risk of mutagenic changes, but the author is not aware of clinical reports of this.

The principal adverse effects that should be communicated to the patient before a procedure are bruising and pain. Bruising is of course a possibility in any procedure at or near muscle, but may be more common in spasticity procedures as the muscle may contract due to stimulation or spasm whilst the needle is in situ. On occasion, appreciable haematoma formation can occur, and in one case in the author’s experience, this organised, resulting in contracture. Caution is needed therefore in anticoagulated patients. In our department, we generally employ an empirical limit of an INR of 2.0, especially in areas, such as the forearm flexor compartment, where space is constrained. Similar considerations apply to those with bleeding diatheses.

Peripheral procedures using Phenol have been reported as commonly producing severe pain during injection, and the chance of late deafferentation pain. In our experience (which extends to over 2000 cases), such side effects are rare. In Ahmed’s series,4 the overall late complication rate (combining haematoma and neuralgic pain) was 2%, and this is in accord with our longer experience. This rate, lower than some quote, may be because of our preference for avoiding injection of large mixed nerves where possible, preferring injection of more-or-less purely motor nerves, or motor-point procedures.  Immediate pain lasts only whilst phenol is being injected. In our series of 196 cases mentioned above, pain during injection was recorded in four cases; but there may have been some under-recording, as in general we estimate a 5% chance of immediate pain. Deafferentation (neuralgic) pain is more troublesome when it does occur. Generally it appears one to two weeks after the injection, and lasts for some weeks. It most cases, it is self-limiting, but some require management with agents such as Gabapentin. We have not seen cases of post-phenol neuralgias lasting more than four to five months, but patients are warned that permanent cases have been reported.

Systemic side effects are unlikely at the doses commonly employed in spasticity procedures. The lethal dose of phenol has been reported as at least 8.5 G,6 which is over 140 mls of 6% phenol. Arrythmias were encountered in a case where 40mls of 6% phenol was injected to the coeliac plexus. A ‘house rule’ in our department was established as a maximum 50% of that, ie 20mls of 6% in any one day. However, other body areas could be injected a day or two later.

The effect of Phenol may be felt immediately, presumably due to its local anaesthetic action, but the full effect may take two to three weeks to develop. The duration of action is very variable, from a few weeks only to some years. In practice, patients are advised to hope for six to nine months benefit. We rarely re-inject a site less than three months after a first injection. Benefit is not universal in our experience, but we counsel that useful benefit may be expected in 60-80% of patients. Following injection, appropriate exercise and orthotic intervention, as outlined by Gaid, forms an integral part of the treatment.

Motor point blockade

Motor points lie on the surface of the muscle. Each muscle may have a number of motor points. The locations vary greatly from individual to individual, and their location must therefore be identified as the first stage of blockade procedures. It is said that they tend to cluster around the midpoint of a muscle’s length; in the author’s experience this is largely, but by no means wholly, correct. The advantages of injecting motor points rather than nerve trunks are that individual muscles can be targeted, and that the absence of sensory fibres reduces the chance of dysaesthesia following injection; the disadvantages are the difficulty in injecting deep muscles, and the larger number of injections required.

The technique is as follows:

1)      Using a nerve stimulator, an exploring electrode is systematically swept over the skin overlying the target muscle, and muscle response identified by palpation and/or vision. Initially, a low stimulus amplitude is used, to aid patient tolerance. Maximum response for minimum amplitude is sought, aiming to locate the point to within 1cm. At 2Hz stimulation, it follows that the maximum sweep speed to achieve this is 2cm/second.

2)      If no suitable point is found, the stimulus amplitude is increased according to tolerance until a motor point is identified. The point is marked with a suitable skin pen, and further points sought. Typically in a larger muscle such as biceps, three or four points may be identified; frequently a smaller muscle yields only one point.

3)      After suitable skin antisepsis, an insulated needle is introduced through the skin at a marked point, and advanced more deeply to reach the surface of the muscle. A resistance is usually felt as the needle penetrates the fascia; the next resistance beyond that is taken to indicate the muscle surface. It is preferable not to enter the muscle, to limit bruising, and as the blockade is accomplished more easily at the muscle surface.

4)      Using the muscle stimulator, the point of the needle is moved to identify the strongest response. Whilst stimulation continues, 1ml of phenol is injected. If all stimulated contraction is not abolished, further smaller aliquots of phenol are given, titrated to response. It is rare for more than 2mls to be required at one point; that requirement suggests imprecise localisation, and prompt repositioning of the needle.

5)      The procedure is repeated for each point found at Stage 1 above.

Specific areas of intervention

Internal rotation and adduction of the arm

Pectoralis major is in most cases readily amenable to motor point block, which is illustrated in Figure 2.  Typically three to four points are found, most commonly just medial to the axilla.

Subscapularis7 motor point injection has also been described, via a medial approach at the level of the spine of the scapula. The patient should be positioned sitting forward, and the operator grasps the medial border of the scapula and pulls it posteriorly, to widen the access. Some patients’ bodily habitus will make this procedure difficult or undesirable. Here, skin stimulation is clearly valueless, but the motor point may be expected by advancing the needle close to the anterior surface of the scapula about 7-10cm. The patient should be warned of the possibility of pneumothorax.

Latissimus dorsi is approached by identifying motor points just medial to the posterior axillary fold, near the musculo-tendinous junction. This site is chosen as well away from the so-called ‘bare’ area of pleura near the inferior border of the muscle.

Elbow flexion

The three muscles to be considered are biceps, brachialis, and brachioradialis. Musculo-cutaneous nerve block will impact on biceps and brachialis, but not brachioradialis. The technique is described by Keenan,5 with medial insertion of the needle to pass between the short head of biceps and the brachialis. Some additionally use the brachial artery to aid localisation.

As an injection of a mixed nerve, there may be concern over a greater risk of post-injection neuralgia, and so we have preferred motor point injection of biceps8 and/or brachioradialis. Biceps is generally less important as an elbow flexor than the other muscles, but it is important when the forearm is pronated, which is commonly the situation in spasticity. In brachioradialis, it is rare to identify more than one point. Typically, this is found about 1-2cm distal to the elbow crease, on the lateral side of the supinated forearm.

Wrist and finger flexion

Motor points of the superficial flexors (flexor carpi radialis, ulnaris, and flexor digitorum superficialis)  are found by scanning the volar surface of the forearm with a surface electrode from approximately 4cm distal to the elbow crease to the musculo-tendinous junction. The motor points of these muscles are often found very close to each other, approximately halfway between the elbow and the wrist (or a little more proximally). Since wrist and finger spasticity often co-exist, it is not usually essential to differentiate which points are selected, but if needed individual muscles (and sometimes individual digits) can be selected by careful observation of muscle stimulation. Flexor digitorum profundus is more difficult to treat by this method, but can sometimes be identified.

Since this compartment has limited space, the total volume of injection is kept low – preferably less than 5mls – achieved by very careful localisation of the motor points.

Thumb-in – palm

Thenar eminence muscles may often respond to a single injection to the recurrent motor branch of the median nerve, as described by Keenan9. This is injected where the nerve enters the thenar eminence, at the junction of a line drawn proximally from the radial border of the long finger, and a line formed by projecting the medial edge of the thumb parallel to the first palmar crease. The innervation of the thenar muscles is variable, but in the majority of subjects, this nerve supplies them. 2-3mls of 6% phenol are injected, titrated to ablation of stimulation. Flexor Pollicis Longus may also need injection.


Despite the ready availability of alternative procedure, the long duration of action and relative economy of phenol blockade mean that these procedures remain a useful part of spasticity management.


1.      Gaid M. Phenol Nerve Block for the Management of Lower Limb Spasticity. ACNR, 2012;12(3):23-5.

2.      Royal College of Physicians, British Society of Rehabilitation Medicine, Chartered Society of Physiotherapy, Association of Chartered Physiotherapists Interested in Neurology. Spasticity in adults: management using Botulinum toxin. National guidelines. London: RCP, 2009

3.      Khalili AA, Harmel MH, Forster S, Benton JG. Management of spasticity by selective peripheral nerve block with dilute phenol solutions in clinical rehabilitation. Arch Phys Med Rehabil 1964;45:513-19.

4.      Ahmed S, Hagen P, Forrest S, Roy C. Unpublished data (personal communication), 1995

5.      Keenan MAE. Management of the spastic upper extremity in the neurologically impaired adult. Clinical Orthopaedics and Related Research,1988;233:116-25.

6.      Gracies J-M, Elovic E, McGuire JR, Simpson DM. Traditional Pharmacologic treatments for spasticity. Part I: Local treatments. In: Spasticity: Etiology, evaluation, management and the role of botulinum toxin. We Move; 2002;65-93.

7.      Chironna RL, Hecht JS. Subscapularis motor point block for the painful hemiplegic shoulder. Arch Phys Med Rehabil 1990;71(6):428-9.

8.      Batcheler R, Roy CW. The use of motor point blocks to relieve spasticity in biceps muscle. NZ J Physiother, 1991;19(3):6-7.

9.      Keenan MAE, Botte MJ. Technique of percutaneous phenol block of the recurrent motor branch of the median nerve. J Hand Surg 1987:12A;806.

Download this Article