Enchondroma is one of the most common benign bone tumours of the hand. It originates from cartilage and is commonly located in the proximal metaphysis of the proximal phalanx.1 The tumour usually presents as an incidental finding or pathological fracture.

Despite being the most common bone tumour in the hand, standardised treatment protocols are lacking.1 Options vary from observation alone, curettage alone, and curettage with bone grafting (recently with artificial bone substitute). At the Prince of Wales Hospital (PWH), we used to treat enchondroma of the hand by complete curettage and filling the defect with autologous bone graft. Although autologous bone graft provides both biological and mechanical advantages in managing the bone void, this procedure is not without risk. Patients must undergo general anaesthesia to obtain the bone graft from the iliac crest, and most patients have considerable postoperative pain, which limits their walking ability for a variable period.

Recently, studies evaluating the clinical application of artificial bone substitute have shown promising results.2 3 However, this is a relatively new technique in local practice. In a cohort study, we retrospectively analysed the treatment outcomes of patients with hand enchondroma and compared the results for autologous bone graft and artificial bone substitute.

From January 2001 to December 2013, all patients with symptomatic monostotic enchondroma of the phalanges or metacarpals treated at the PWH or the Alice Ho Miu Ling Nethersole Hospital (AHNH) underwent thorough curettage according to the standard protocol. The bone defects were filled by either autologous bone graft or artificial bone substitute, depending on the surgeon’s and patient’s preferences. All operations were done by the same team of orthopaedic specialists. For patients presenting with pathological fracture, the fracture was first managed conservatively until healed. The surgery for the tumour was performed 3 months after initial presentation.

In the artificial bone substitute group, an incision was centred on the lesion, and the extensor tendon was retracted, with no subperiosteal dissection. A small oval cortical window was made by connecting multiple drill hole perforations prepared by a 0.9-mm Kirschner wire. The tumour was removed by small-angle curettage and clearance was checked under fluoroscopic control (Figs a and b). The cavity was then filled with artificial bone substitute (Fig c).

A custom-made paper funnel was used for precise insertion of bone substitute to avoid spillage to the surrounding soft tissue, which could be difficult to remove (Figs d and e). Bone substitute granules were impacted tightly by using a punch (Fig f). The piece of oval cortical bone was placed back in position, and the periosteum was repaired where possible; alternatively, the window was sealed with fibrin glue (Tisseel; Baxter Healthcare Corp, Deerfield [IL], US) to contain the bone substitute (Fig g). The wound was closed with fine nylon suture. A radiograph was taken to confirm filling of the defect and absence of fracture (Fig h). Free mobilisation was allowed postoperatively.

In the autologous bone graft group, the operation was done under general anaesthesia. The surgical approach and procedures to the affected bone were the same as for the artificial bone substitute group, except that autologous cancellous bone grafts harvested from iliac crest were used instead of artificial bone substitute. We do not usually obtain the bone graft from the ipsilateral distal radius as the quantity is insufficient for packing the wound. Free mobilisation was allowed postoperatively.

The operative details and postoperative clinical and radiological outcomes were reviewed by an independent reviewer. Fisher’s exact test was used for sex, tumour site, and pain score. Mann-Whitney U test was used for the Disabilities of the Arm, Shoulder and Hand questionnaire (DASH) score, and t test was used for the other parameters. Clinically, the active range of motion, symptoms, and function measured by the Chinese and shortened version of the DASH (QuickDASH)4 were evaluated. Plain radiographs were taken at standard intervals (1 week, 4 weeks, 3 months, 6 months, and annually) postoperatively to determine bone incorporation. Bone incorporation was defined as a seamless appearance with no gap between the cancellous bone and the bone substitute. For any suspicious symptoms or radiographic appearance, computed tomography or magnetic resonance imaging (MRI) was performed to look for any recurrence.

There were 24 patients (eight men and 16 women), with a mean age of 40 years. Overall, 17 cases involved the phalangeal bones and seven involved the metacarpal bones; 13 patients underwent autologous bone graft and 11 had artificial bone substitute. Five patients in each group presented with pathological fracture, among whom nine were managed conservatively until the fracture was healed. Patients’ demographics, including site and size of the tumour, presentation, and time to operation are shown in Tables 15 and 2. In all patients, the histology confirmed the diagnosis of enchondroma. The operative details and postoperative outcomes are shown in Tables 3, 4, and 5.

For the artificial bone substitute group, 10 of 11 patients were operated on using intravenous local anaesthesia or regional plexus block. One patient was operated on under general anaesthesia as the MRI showed suspicion for malignancy, and frozen section was performed during the operation. The mean surgical time was 81 minutes (range, 60-135 minutes). All surgeons used Bio-1 granules (SBM France, Lourdes, France) except for one patient for whom injectable bone substitute (Norian SRS; Synthes USA, Paoli [PA], US) was used because of the surgeon’s preference.

For the autologous bone graft group, all 13 patients were operated on under general anaesthesia. The mean surgical time was 106 minutes (range, 60-150 minutes), which was 25 minutes longer than for the artificial bone substitute group (P=0.008).

The mean follow-up period was 59 months (mean 59 months, range 4-102 months in the autologous bone graft group and mean 59 months, range 12-153 months in the artificial bone substitute group). All patients demonstrated satisfactory bone incorporation. There were no significant observable radiological differences between the groups 1 year after operation. Functional recovery was similar in both groups. There were no significant differences in QuickDASH scores (mean 2.3 for artificial bone substitute group and 5.1 for autologous bone graft group; P=0.128).

One patient in each group developed soft tissue complications 3 weeks after the operation. The patient in autologous bone graft group presented with erythema over the surgical site. The condition improved after intravenous antibiotic treatment was given and the patient was diagnosed to have a low-grade superficial infection. The patient in the artificial bone substitute group presented with discharge from the wound and radiograph showed a trace amount of tiny calcifications in the soft tissue adjacent to the affected digit. The culture swab of the discharge fluid showed negative growth. The patient was treated with empirical antibiotics and surgical debridement showed a small amount of bone substitute in the subcutaneous plane of the wound. The diagnosis was probable inflammation secondary to foreign body reaction, rather than a genuine infection. Both patients had satisfactory wound healing and bone healing.

Recurrence of enchondroma was suspected in one patient in the artificial bone substitute group. The patient had a radiolucent lesion at the proximal part of the affected metacarpal on plain radiograph during routine follow-up 8 years after the index operation. The patient subsequently underwent a second operation to remove the lesion.

Joosten et al2 first reported treatment of enchondroma with artificial bone substitute in 2000. Eight patients were treated with hydroxyapatite cement to fill the bone cavity. All of the patients gained full function of the hand and no complications were observed during 1-year follow-up. Subsequently, studies from Japan3 and South Korea6 have also shown satisfactory outcomes using calcium bone cement and calcium-based pellets, respectively.

At the PWH and AHNH, we treat all hand enchondromas surgically because the tumour will usually grow, weaken the bone, and result in pathological fracture. Since the bone will be further weakened by curettage alone, we believe that replacement with an osteogenic or osteoconductive substance will facilitate bone healing and remodelling so that this fracture-prone period can be shortened. We traditionally treated hand enchondroma with curettage and filled the defect with autologous bone graft. However, we started treating with artificial bone substitute in 2001. In 2010, we changed to routine use of autologous bone graft because there are several advantages of reduced donor site morbidity, use of local anaesthesia, reduced operating time (mean, 25 minutes less), and the surgery can be performed on a day-case basis.

There are different types of bone substitute available in the market. In this study, either Bio-1 granules or injectable Norian7 was used. These bone substitutes are synthetic materials made with resorbable calcium phosphate. The composition comprises calcium and phosphate ions, which are biocompatible with natural bone minerals.8 An in-vitro study shows that calcium phosphate allows osteoblast fixation and proliferation,8 followed by osteointegration and bone resorption mimicking normal bone healing. Calcium phosphate is available in granules or cubes and in an injectable form.

In this study, complete curettage of the tumour was achieved, with histological confirmation of the diagnosis. There were no significant differences in QuickDASH scores between the two groups.

Autologous bone graft takes around 4 to 6 months to incorporate while, for artificial bone substitute, the time to incorporation depends on the type of bone substitute used. Bio-1 takes approximately 9 to 12 months to incorporate. There were no significant radiological differences between the groups at 1 year postoperatively. Norian stays in the bone for longer than Bio-1 and is not completely resorbed up to 3 years postoperatively.

The mean follow-up period of this study was 59 months, which is longer than in most studies. The numbers of patients in each treatment group were comparable with other studies. We observed suspected recurrence in the affected metacarpal in one patient, who had undergone operation 8 years previously. A radiolucent lesion was noted beneath the bone substitute. We postulated that there might have been residual enchondroma cells seeding at the base of the lesion after curettage, which were displaced proximally during impaction of the bone substitute.

There were several limitations to this study. First, there was a difference in patient age between the two groups, which is a confounding variable. This might be accounted for by the relatively low incidence of enchondroma despite it being the most common upper limb tumour. Second, the choice of artificial bone substitute was not standardised, as two different substitutes were used. Norian SRS injectable bone substitute was used in one patient and Bio-1 was used in the other patients. Third, radiological assessment postoperatively might not be accurate. Despite all radiographs being reviewed by experienced orthopaedic specialists, the diagnosis of bone incorporation was subjective, with the chance for inter- and intra-observer bias. Finally, this study was retrospective and non-randomised.

Overall, most patients gained full range of motion and satisfactory function, with radiological evidence of bone incorporation and, later, bone growth. The application of artificial bone substitute gives comparable functional and radiological results in treating enchondroma of the hand. The procedure allows reduction in operating time, elimination of donor site morbidity, and day-case surgery under local or regional anaesthesia. Meticulous curettage and bone substitute impaction without spillage to the surrounding soft tissues are key to achieving good outcomes and avoiding complications.

1. Sassoon AA, Fitz-Gibbon PD, Harmsen WS, Moran SL. Enchondromas of hand: factors affecting recurrence, healing, motion, and malignant transformation. J Hand Surg Am 2012;37:1229-34. Crossref

2. Joosten U, Joist A, Frebel T, Walter M, Langer M. The use of in situ curing hydroxyapatite cement as an alternative of bone graft following removal of enchondroma of the hand. J Hand Surg Br 2000;25:288-91. Crossref

3. Yasuda M, Masada K, Takeuchi E. Treatment of enchondroma of the hand with injectable calcium phosphate bone cement. J Hand Surg Am 2006;31:98-102. Crossref

4. Lee EW. Chinese QuickDASH (PWH, HK version). Physiotherapy Department, Prince of Wales Hospital. Toronto: Institute for Work and Health; 2006.

5. Takigawa K. Chondroma of the bones of the hand. A review of 110 cases. J Bone Joint Surg Am 1971;53:1591-600.

6. Choy WS, Kim KJ, Lee SK, Yang DS, Park HJ. Treatment for hand enchondroma with curettage and calcium sulfate pellet (OsteoSet®) grafting. Eur J Orthop Surg Traumatol 2012;22:295-9. Crossref

7. Hak DJ. The use of osteoconductive bone graft substitutes in orthopaedic trauma. J Am Acad Orthop Surg 2007;15:525-36.

8. Le Huec JC, Clément D, Lesprit E, Faber J. The use of calcium phosphate, their biological properties. Eur J Orthop Surg Traumatol 2000;10:223-9. Crossref