Abstract
The commonest paralytic deformity in leprosy is the claw deformity of the fingers. Many surgical procedures have been described and are practiced to correct this deformity, but none is free from complications. A modification of the Zancolli lasso operation in which only half of the flexor digitorum superficialis tendon of the middle finger was used as the motor is described and a comparative study of this and the conventional procedure was carried out in 70 hands in 70 patients. The results suggest that the modification is technically simpler, with comparable results to those of the conventional procedure and fewer complications.
Claw deformity of the fingers, either alone or in combination with a Simeon thumb deformity, is by far the commonest deformity caused by leprosy. Most deformities due to this disease can be reversed by physiotherapy and steroids if treated early but, once established, they can only be corrected by surgery. Many procedures have been described to correct claw deformity of the fingers and each has advantages and disadvantages but none is without problems.
Procedures in which wrist extensors and flexors are used as motors (Brand, 1961; Riordan, 1953) require lengthening of their tendons to reach the points of insertion, for which harvest of another tendon, e.g. the palmaris or plantaris tendon, or use of fascia lata as the extension, is required. They also require that the lengthened tendon is routed to the palm, which can cause problems of crowding and pressure, particularly if the transfer passes through the carpal tunnel. Procedures that use finger flexors as motors are liable to swan neck deformity and contracture of the proximal inter-phalangeal joint in the donor finger, and these may require surgical treatment.
This study describes a modification of the Zancolli lasso operation for clawing of the fingers (Zancolli, 1957) and compares the results of use of this modification in 37 hands in 37 patients with the use of the conventional Zancolli lasso operation in 33 hands in 33 patients.
PATIENTS AND METHODS
This retrospective comparative study was conducted at Hoina Leprosy Research Trust, Muniguda, Orissa, India, during 2001–2005. Seventy of the 93 hands (75%) operated upon during this time had been followed-up for 1 year, or more, and were included in this study.
Thirty-three hands in 25 men and eight women of mean age 39 (range 16–55) years underwent correction of claw deformities of the fingers caused by leprosy using the standard modification of the Zancolli lasso procedure in which the whole flexor digitorum superficialis tendon of the middle finger was used as the motor. In the modified Zancolli lasso procedure, all four fingers are corrected by a single flexor digitorum superficialis tendon, which is divided into four strips, as against the original procedure in which each finger is corrected by one whole flexor digitorum superficialis tendon (Zancolli, 1979). All four fingers were corrected in all 33 hands in the 33 patients, with the whole flexor digitorum superficialis tendon of the middle finger split and used for four fingers. These patients form the first group in the study. Twenty had surgery to the right hand and 13 to the left hand. Seven also had a Simeon thumb deformity. Fifteen patients had borderline tuberculoid type of leprosy and 18 had the lepromatous type. The duration of the paralysis ranged from 1 to 10 years. All of them had completed multi-drug therapy. Skin smears were negative for lepra bacilli in all 33 patients. The average period of follow-up for this group was 2.6 (1–4.5) years.
Thirty-seven hands in 30 men and seven women with leprosy and a mean age of 32 (range 15–54) years underwent correction of claw deformities of their fingers using a new modification of the Zancolli lasso procedure in which only half of the flexor digitorum superficialis tendon of the middle finger was used as the motor. This modification is described as the “New Procedure” throughout this paper. All four fingers were corrected in all 37 hands in the 37 patients, with one half of the flexor digitorum superficialis tendon of the middle finger split into four longitudinal strips, each of which was used for one finger: the other half of the superficialis tendon was left in place. These patients form the second group in this study. Fifteen had surgery to the right hand and 22 to the left hand. Two also had a Simeon thumb deformity. Seventeen had borderline tuberculoid leprosy and 20 had the lepromatous type of leprosy. The duration of the paralysis ranged from 1 to 15 years. All of them had completed multi-drug therapy. Skin smears were negative for lepra bacilli in all 37 patients. The average period of follow-up for this group was 1.6 (1–3) years.
The criteria for selection for surgery were established claw deformities of two, or more, fingers, with all patients having had multi-drug anti-leprosy therapy for a minimum period of 6 months and having had no steroid therapy for at least 3 months.
Preparation for surgery by the physiotherapists included mobilisation of the joints of the involved hand, isolated exercises of the muscle to be transferred to create an awareness of its individual action and strengthen it, as well as general exercises and education on sensory orientation.
Surgical technique
Group 1 (“modified Zancolli lasso procedure”)
The modified Zancolli lasso procedure (Zancolli, 1979) using the whole flexor digitorum superficialis tendon of the middle finger was carried out in 33 hands. The flexor digitorum superficialis of the middle finger was preferred to that of the ring finger as the flexor digitorum profundus of the ring finger is weak in most hands affected by leprosy as a result of high ulnar paralysis (Malviya, 1997).
A transverse incision was made across the palm in the distal palmar crease and skin flaps were raised to expose the flexor tendons and their A1 and A2 pulleys. The lumbrical muscles were identified on the radial side of the flexor tendons of each finger. Through a transverse incision at the crease of the proximal interphalangeal joint of the middle finger, the flexor sheath was exposed. A transverse slit was made in it, the profundus tendon was retracted and both halves of the superficialis tendon were severed close to their insertion. The chiasma was divided and the two halves of the tendon were separated. The severed superficialis tendon was withdrawn into an incision in the middle of the palm, and divided into four slips. This division was done through a separate proximal palmar incision to avoid acute angles at the point of division. The slips were then tunnelled alongside the lumbrical muscles to the distal palmar crease incision. Each slip was then passed under the A1 and the proximal half of the A2 pulley of each finger, brought out through a slit made in the A2 pulley, folded back towards the palm and sutured to itself in such a way that its loop enclosed the A1 and the proximal half of the A2 pulley. The slit made in the A2 pulley was closed around the slip. The flexor digitorum superficialis slips were sutured to set the metacarpophalangeal joints in flexion; 30° for the middle finger, 40° for the index and ring fingers and 50° for the little finger. These angles were chosen to enable better cupping of the palm.
Group 2 (“new procedure”)
The new procedure, using half of the flexor digitorum superficialis tendon of the middle finger as the motor, was performed in 37 hands. The flexor tendons and the pulleys were exposed through an incision in the distal palmar crease, as described above. A curvilinear incision was made on the ulnar side of the palmar crease of the proximal interphalangeal joint of the middle finger (Figs 1A and 1B) and the tendon sheath was opened longitudinally. After retracting the profundus tendon, the ulnar half of the superficialis tendon was detached from its insertion into the ulnar side of the base of the middle phalanx and separated from the radial half of the tendon by dividing the chiasma (Figs 2A and 2B). The detached half of the superficialis tendon was brought into the incision in the midpalm and further separated from its intact radial half (Figs 3A and 3B). Another incision was made just proximal to the distal wrist crease and the two halves of the superficialis tendon were split right up to the musculotendinous junction (Figs 4A and 4B). Through the incision in the midpalm, the detached half of the superficialis tendon was divided into four slips (Figs 5A and 5B), and each slip was then tunnelled to the incision in the distal palmar crease and attached to the A1 and the proximal half of the A2 pulley, as already described. The tendon slips were sutured to set the metacarpophalangeal joints in the positions described for the modified Zancolli lasso procedure.
A below-elbow plaster of Paris cast was applied with the wrist in the neutral position and the metacarpophalangeal joints at 70° flexion, leaving the interphalangeal joints free. Mobilisation of the interphalangeal joints was started after 48 hours. The isolated exercises to contract the transferred muscle independently, which the patient had learnt pre-operatively, were commenced after 3 weeks, along with integration exercises to train the transferred muscle to work in unison with the other muscles to bring about movement patterns, for example, flexing the metacarpophalangeal joint using the transfer and then flexing the interphalangeal joints to make a fist. Two weeks later, training to carry out purposeful movements such as transferring beads from one tray to another was started. Patients were seen for follow-up every month for 3 months, every 3 months up to 1 year and then annually for 5 years.
Assessment
Evaluation included measurements of (1) the appearance of each finger, using the angles at the metacarpophalangeal and proximal interphalangeal joints during active extension, (2) the ability to hold the metacarpophalangeal joint of each finger in 90° flexion with the proximal interphalangeal joint extended, (3) the flexion of each finger and (4) the angle of the distal transverse metacarpal arch measured on the dorsal aspect of the hand, with the fingers in extension.
The appearance of the finger was graded as good, fair or poor, using a graph (Fig 6) and plotting the angle of the metacarpophalangeal joint on the X-axis and the angle of the proximal interphalangeal joint on the Y axis, measured with the finger in active extension (open hand position) (Palande, 1976).
The ability to hold the metacarpophalangeal joint in 90° flexion with the proximal interphalangeal joint extended was graded as good when the angle at the proximal interphalangeal joint was 30° or less, fair when this angle was between 31° and 59° and poor when it was 60° or more.
Flexion of the finger was graded as good when the tip of the finger touched the distal palmar crease, fair when it touched the proximal palmar crease and poor when it could not touch the palm.
The distal transverse metacarpal arch was assessed with all the fingers in extension. It was graded as good when the arch was convex dorsally and its angle (measured on the dorsal aspect of the hand at the level of the neck of the third metacarpal) was 15° or more; fair when the arch was convex dorsally and its angle was between 10° and 14° and poor when the arch was convex dorsally and its angle was less than 10°, or if the arch was concave dorsally (reversal of the arch).
The results of assessing the appearance of the finger, the ability to hold the metacarpophalangeal joint in 90° flexion with the proximal interphalangeal joint extended and the flexion of the finger were analysed on an individual finger basis and also as a composite for the hand by summating the results of all four fingers. The summated results were graded as good, when the individual results were good for all the four fingers, fair when the individual results were good in some fingers and fair in others or fair in all the four fingers, and poor when any of the individual results were poor, even in one finger.
Grip strength was measured using a dynamometer (North Coast Medical Inc., Morganhill, California).
Complications assessed during follow-up were swan neck deformity, contracture of the proximal interphalangeal joint, overcorrection resulting in the intrinsic plus deformity, undercorrection resulting in recurrent clawing, joint stiffness and restriction of interphalangeal joint mobility.
Statistical analysis of the results and complications of the two groups of hands were performed. The chi-square test was used to compare categorical variables. For two-by-two tables, Fisher’s exact test was used. A P-value of 0.05 or less was regarded as statistically significant.
RESULTS
The average follow-up for Group 1 (modified Zancolli lasso procedure) was 2.6 (1–4.5) years and for Group 2 (New procedure) it was 1.6 (1–3) years. The results of the analysis of the appearance of the fingers are shown in Table 1, those for the ability to hold the metacarpophalangeal joint in 90° flexion with the proximal interphalangeal joint extended in Table 2, those for flexion of the fingers in Table 3 and those for the distal transverse metacarpal arch in Table 4. Grip strength measurements were available for 22 hands in Group 1 and for 23 hands in Group 2, and the mean strengths were 14 (range 6–25) kg and 16 (range 8–25) kg, respectively. Complications encountered are shown in Table 5.
Statistical analysis
Comparison of the results of the appearance of the fingers, the ability to hold the metacarpophalangeal joint in 90° flexion with the proximal interphalangeal joint extended, flexion of the fingers, the distal transverse metacarpal arch and grip strength showed no significant differences between the two treatment groups, except for the appearance of the composite hand (P<0.02).
Statistical comparison revealed significantly less swan neck deformities and contractures of the proximal interphalangeal joint after the new procedure, as compared to after the modified Zancolli lasso procedure.
DISCUSSION
Although numerous procedures have been described for the correction of clawing of the fingers, only a few are used in practice, including Brand’s operation with the extensor carpi radialis longus as the motor (Brand, 1961), Bunnell’s operation (Bunnell, 1942) and the modified Zancolli lasso operation (Zancolli, 1979) with flexor digitorum superficialis as the motor.
Extensor carpi radialis longus is an excellent motor for claw-finger correction as its normal function of wrist extension and its intended function of metacarpophalangeal joint flexion are phasic movements. Also, it has sufficient power to flex the metacarpophalangeal joints and is expendable if extensor carpi radialis brevis and extensor carpi ulnaris are normal. However, it has to be lengthened with free graft and its isolation and integration are difficult. Palmaris longus, when present, is also a good motor for claw-finger correction, but is not phasic. Unless it is rudimentary, it also has enough power to provide flexion at the metacarpophalangeal joints and is expendable. Palmaris longus also has to be lengthened with a free graft.
The free graft most commonly used to lengthen these transfers is harvested from fascia lata. Although this necessitates another incision in the thigh and stripping of the graft, it does not involve sacrifice of another muscle, as when either the palmaris longus or the plantaris tendon is used. Whichever muscle and graft is used, after anastomosis at the distal forearm level, it may pass through the carpal tunnel, causing crowding, which may lead to compression of the median nerve, especially if it is already inflamed. Instrumentation during passage of the graft through the carpal tunnel can also cause nerve damage (Malviya, 2002).
Flexor digitorum superficialis of the middle or ring finger is also used commonly for claw-finger correction (Bunnell, 1942; Zancolli, 1979). These muscles are phasic but the muscle power is much greater than needed. The tendon is also quite bulky and, even after splitting it into four slips, each slip is thicker than those used in extensor carpi radialis longus or palmaris longus transfers. When the flexor digitorum superficialis is used as the motor, harvest of the graft and its related problems are avoided.
Flexor digitorum superficialis of the middle finger is the preferred motor because of various identified features of comparative anatomy (Brand, 1985; Malviya, 2005). It is independent and is the strongest flexor digitorum superficialis muscle with the longest muscle fibres. Its tension fraction is 3.4%, compared to 2%, and less for the other fingers and its mass fraction is 4.7%, compared to 3%, and less for the other fingers. Over and above these advantages, the flexor digitorum superficialis of the middle finger is preferable to that of the ring finger as the flexor digitorum profundus of the ring finger is quite often weak in leprosy as a result of a high ulnar paralysis (Malviya, 1997).
Swan neck deformity is one of the common complications encountered after removal of the flexor digitorum superficialis tendon from a finger. Reddy and Kolumban (1981) observed this deformity in 83% of fingers and Brandsma and Ottenhoff-de Jonge (1992) reported it in 15%, more often in those with hypermobile joints. Extensor dominance at the proximal interphalangeal joint as a result of removal of the superficialis tendon and the force of the newly transferred flexor at the metacarpophalangeal joint contribute to this deformity (Bourrel, 1997). Many surgeons leave one of the distal superficialis slips a little longer and suture it to the sheath to overcome this problem.
Flexion contracture of the proximal interphalangeal joint, another complication, occurs as a result of scarring due to adhesion of the distal stump of the flexor digitorum superficialis (Fritschi, 1984). Brandsma and Ottenhoff-de Jonge (1992) reported this complication in 26% of fingers. Early mobilisation of the interphalangeal joints obviates this complication.
When only half of a flexor digitorum superficialis is used, swan neck deformity is less likely to occur, as the remaining half prevents the proximal interphalangeal joint from assuming the position of hyperextension. Also, flexion contracture does not occur at the proximal interphalangeal joint as the ulnar half of the superficialis tendon is detached, without leaving a distal stump. As the contribution of the radial half of the flexor digitorum superficialis of the middle finger is more important for tripod pinch, it is retained and the ulnar half of the tendon is used for the transfer. As the width of the slips of the flexor digitorum superficialis used in the new procedure is only half of that of the slips with the conventional lasso procedure, tunnelling and suturing them is technically easier.
Comparison of the results of the two procedures failed to show any statistically significant difference between them with respect to the results, including the grip strengths of the hands, except for the appearance of the hands. The appearances of the hands following the new procedure were superior to those following the conventional modified Zancolli lasso procedure.
Swan neck deformity was encountered in four, and contracture of the proximal interphalangeal joint in ten, of the 33 hands treated with the conventional modified Zancolli lasso procedure. In contrast, these complications did not occur with the new procedure, which is technically simpler, and produces comparable results to those of the conventional modified Zancolli lasso procedure.
Footnotes
Acknowledgements
This study was conducted at the Hoina Leprosy Research Trust, Muniguda, Orissa, India, with the kind permission of the Director of the Trust and the Chief Executive of the Lepra Society. The author would like to thank Dr Dinker D Palande, Vice Chairman of the Lepra Society, for help with the study and Mr MS Rajagopalan for his valuable statistical advice. The author would also like to thank Mr A Kameswara Rao, Physiotherapy Coordinator of the Lepra Society, Mr P Nageswar Patra and the team of physiotherapists and Mrs Kanakalatha Tandi and the team of nurses for their help with the study.