Introduction

Giant cell tumor (GCT) is a locally aggressive tumor of bone with distal radius as the third most common site [1]. Management of GCT of distal radius has largely remained controversial as rates of recurrence and metastasis have been reported to be higher in distal radius [2, 3]. Intralesional surgery as compared with resection provides better function although the recurrence rates are reported to be higher [4]. Given that radio-carpal articulation is significant for wrist joint function, reconstruction of the joint remains a challenge after resection. Given the paucity of data from India, we analyzed our patients with GCT of distal radius for functional and oncological outcomes with the question- “Is curettage justified for GCT of distal radius?”.

We compared the oncological and functional outcomes of the patients with GCT involving distal radius managed with extended intralesional curettage with cementing and resection-arthrodesis of the wrist.

Material and Method

We enrolled all patients operated for giant cell tumor of distal radius by a single surgeon(AT) between July 2011 and June 2021. We recorded demographic parameters, stage and grade of disease, and surgical details. The oncological and functional outcomes (Musculoskeletal Tumor Society score and Modified Mayo wrist score) were analyzed for all patients.

A total of 24 patients (14 females) with mean age of 31.6 years (range 17–58 years) were included in the study. Mean follow-up was 6.3 years (range 2 to 15.9 years). Eighteen patients presented with primary disease and six patients with recurrent disease following management elsewhere. None of the patients presented with metastasis but one patient developed multifocal disease 2 years after surgery, though the primary remained controlled.

All patients underwent complete work up including radiology and pre-operative biopsy. Out of the 18 primary presentations, six were grade II (all of which underwent EIC). The remaining 12 patients presented with grade III disease, of which three patients underwent extended intralesional curettage and 9 underwent resection and arthrodesis. Out of the six patients managed elsewhere, five underwent RRU while one case was managed with repeat curettage. Hence, 6 grade II and 3 grade III patients underwent EIC.

For extended intralesional curettage (EIC) a large window was created preferably at the cortical defect. The approach to curettage depended on the soft tissue component and cortical breach to preserve the cortex. In the absence of a soft tissue component, the dorsal approach was preferred to minimize the risk of tendon attrition, avoid contamination of multiple compartments/neurovascular bundle, and ease of applicability of plate for fixation. High-speed burring, 80% phenol, hydrogen peroxide, and pulse lavage were used as adjuvants. All patients underwent cementing reinforced with intra-cavity K wire wherever deemed necessary. Three out of these nine patients underwent reconstruction with longitudinal cement bone graft composite where cortical destruction was more than 50% (Fig. 1). For patients in resection group, wide resection and reconstruction with arthrodesis with radialisation of ulna was done which was supplemented by internal fixation with a contoured plate (Fig. 2).

Fig. 1
figure 1

From left to right showing pre-operative radiographs depicting an intra-osseous lesion in distal radius (A); intra-operative images showing the procedure of extended intralesional curettage (B); immediate postoperative radiographs of patient managed with extended curettage and cementing augmented with K wire to maintain radial height (C)

Full size image
Fig. 2
figure 2

From left to right showing pre-operative radiographs depicting extracompartmental lesion in distal radius with soft tissue extension (A); intra-operative images depicting resection of lesion with radialisation of ulna (B); immediate postoperative radiograph of patient managed with resection and reconstruction with single bone forearm with implant in situ (C)

Full size image

Wilcoxon–Mann–Whitney test was used to compare the Musculoskeletal Tumor Society score (MSTS) [5] and Modified Mayo Wrist Score [6]. Fischer exact test was used to compare the categorical variables of EIC compared to resection, recurrence rate, and rehabilitation time.

Results

There were a total of seven recurrences in our series with an overall rate of 29.1%. Table 1 depicts the distribution of the cases as new/old and according to the procedure done. Four of the recurrences were in cases initially managed elsewhere. All recurrences were managed with repeat surgery (wide excision of soft tissue recurrence in six and conversion to RRU in one). Thus, all patients were free of disease at the final follow-up. Out of the three recurrences in new cases, two were in the EIC group, of which only one was converted to resection. Hence, wrist joint could be salvaged in eight patients overall. The rate of recurrence in EIC group was higher than that in resection but was not statistically significant (p > 0.05). Patients managed elsewhere and then presenting to us for recurrence had higher rate of local recurrence (re-recurrence for the patient) (66.6%, p = 0.01). Two of the patients in this group had more than 2 recurrences (3 and 4, respectively). Average time to recurrence was 14 months (range 2–24 months).

Table 1 Recurrence rates in the various study groups
Full size table

EIC was the surgery of choice in Grade II disease. Three out of twelve Campanacci Grade III cases underwent EIC while the other nine patients were treated with RRU. One patient in EIC group had cement extrusion and fracture 2 months postoperatively and was managed with strut bone grafting and fixation. All patients in resection-arthrodesis group had a solid union. One patient had a delayed union and one patient had a surgical site infection, both of which were managed conservatively.

Mean rehabilitation time (time taken to reach optimal function) in the curettage group was 3.5 months as compared to 8.2 months in the resection-arthrodesis group (p > 0.05). The total range of motion (including wrist flexion + extension + supination + pronation) was 283 degrees in curettage group and 110° in resection group. Mean Modified Mayo Wrist score and MSTS score for both groups are tabulated (Table 2). Function on MSTS score for both groups was comparable but patients with EIC fared better when evaluated with the Mayo wrist score (Table 2).

Table 2 Wrist range of motion (ROM) and functional scores in both groups
Full size table

Discussion

Giant cell tumor is an aggressive tumor of bone usually seen in adults [7]. It usually affects the ends of long bones and may be rarely multifocal. Distal radius is the third most common site accounting for around 10–12% of all GCTs [8, 9]. GCT involving the distal end of radius has been treated as a distinct entity for various reasons such as higher incidence of local recurrences, complex osseous anatomy of the carpus, and scant soft tissue cover being some of these.

Moreover, GCT in distal radius is reported to be more aggressive with large proportion presenting with Campanacci grade 3 lesions. The frequency of grade 3 lesions in distal radius is reported to be 70–80% in previous studies (66.7% in our series) which is higher than that seen in the knee [6, 10,11,12]. Although the association of metastasis in relation with site is controversial, many studies have positively associated primary at distal radius with higher risk of metastasis [13, 14]. Another observation in corollary is that of higher recurrence rate attributed to the location of the disease. Again, a recurrence rate of 25–85% has been reported in the literature, varying according to the procedure done [10, 15, 16].

The threshold of choosing resection over curettage for GCT of distal radius is generally low, and the reasons are many- high rates of recurrence [2, 3], high percentage of those presenting with Campanacci Grade 3 disease [4, 11], and acceptable functional outcomes of wrist arthrodesis/fibula auto-transplant [15, 17]. This has been the reason behind more work being reported on resection and much less on curettage [18,19,20,21]. Studies focussing on outcomes of both EIC and resection are even rarer [16, 22, 23]. This study presents the outcomes of both these techniques in distal radius GCTs, performed by the same surgeon and with a long-term follow-up for median 6.3 years; a significant addition to existing literature.

Given the obvious benefits of EIC in distal radius, our approach was to consider EIC in all patients where we could maintain the mechanical integrity of the bone after EIC (with cement with or without bone grafts/K wire/splintage as deemed necessary). We then discussed the merits and demerits of both approaches with the patient, and then made an informed collective decision about the procedure. In this retrospective review, we can conclude that the choice of EIC in the selected patients resulted in a better function, with acceptable oncological outcomes.

In general, the debate between curettage and resection for GCT of bone is never-ending [16, 22]. Curettage has been popularly advocated in GCT radius where there is no wrist involvement, less than 50% cortex involved, or destruction in less than two planes [24]. Even in larger Campancci grade 3 tumors with extensive destruction, curettage is recommended as adequate treatment if augmented with additional stabilization [25]. This becomes more relevant for GCT of bone involving the distal radius, for reasons discussed in the previous paragraph. A recent study reported that extraosseous soft tissue component with more than 50% destruction could also be managed with a longitudinal sandwich technique where bone graft is placed over the deficient cortex protected with gel foam before cementing the cavity [26]. In this series, three of the patients in the EIC group underwent reconstruction with this longitudinal graft-cement composite technique. None of the above patients had recurrence or any other significant surgical complication on their last follow-up (Fig. 3).

Fig. 3
figure 3

Pre-operative radiograph showing large lesion in distal radius with cortical destruction (A); postoperative X-ray of wrist at 3 months (B) and 6 months (C) showing reconstruction with modified longitudinal sandwich technique

Full size image

In our study, there were two recurrences in the EIC group (11.1%) with primary presentation as compared to a single recurrence in the resection group (22.2%). One out of the two recurrences was managed with repeat EIC and is disease free till date. Hence, a major proportion (n = 8, 89%) of the patients in the curettage group retained their natural joint, prompting us to conclude that curettage is a viable option for patients affected by GCT of distal radius, even some of those with Campanacci Grade 3 disease. A recent study found no increased risk of recurrence between grade 2 or 3 diseases; furthermore, it was observed that the recurrences in the extended curettage group could be managed with repeat curettage with preservation of the joint [27].

Another observation made was of higher recurrence rates in cases presenting after initial intervention elsewhere. This finding is in contrast with previous studies where no increased risk is seen in previously intervened cases [2, 22]. Higher recurrence rates could be attributed to the heterogeneity in the standard operating procedure, soft tissue contamination with curetted disease in previous surgery and the dubious use of adjuvants [28, 29]. Hence, emphasis should be placed on proper selection of procedure for the patients and thoroughness at the initial encounter itself.

Resection followed by reconstruction in distal radius is recommended for GCTs with large soft tissue component. Reconstruction can be done by translocation of ulna, iliac crest autograft, fibular arthroplasty or prosthesis [17,18,19,20,21, 30]. Various factors such as feasibility, ease of procedure, cost, and durability of reconstructed segment have to be regarded while choosing the method of reconstruction. In this series, all patients underwent resection arthrodesis with radialisation of ulna as it gives a stable and pain-free wrist joint.

A number of reviews have reported curettage to be functionally superior to resection, where EIC was seen to have a higher recurrence rate but better functional outcomes [16, 22, 23]. In our study, the resection group had comparable functional scores with the EIC group (Table 2). This could be explained by the upper limb being nonweight bearing and functional compensation by the contralateral limb. Due to subjective nature of MSTS score, a more objective Modified Mayo Wrist score was used to compare function among the patients. Patients in curettage group fared better in assessment of Mayo Wrist score because of the inclusion of objective parameters like range of motion and grip strength. EIC group also had significantly shorter rehabilitation time, another factor not taken into account in the MSTS score.

This study had numerous limitations including its retrospective design and a small sample size. Curettage was more likely to be performed in Grade II lesions, so a selection bias in the outcomes cannot be ruled out. However, this study reports the outcome of curettage and resection for GCT of distal radius by a single surgeon, with a single type of reconstruction after resection, ensuring uniformity of the operative procedures, with a good follow-up.

Conclusion

In this retrospective study of 24 patients with GCT of distal radius, extended intra-lesional curettage resulted in an earlier rehabilitation with a mobile wrist and an acceptable disease control when compared with resection and arthrodesis with radialisation of ulna.