LUMiC Endoprosthetic Reconstruction of Periacetabular Tumor Defects: A Multicenter Follow-up Study

Abstract

Update: This article was updated on July 17, 2024 because of a previous error, which was discovered after the preliminary version of the article was posted online. The byline that had read "Richard E. Evenhuis, MD 1 , Michiel A.J. van de Sande, MD, PhD 1,2 , Marta Fiocco, PhD 2,3,4 , Demien Broekhuis, MD 1 , Michaël P.A. Bus, MD, PhD 1 , and the LUMiC® Study Group*" now reads "Richard E. Evenhuis, MD 1 , Michiel A.J. van de Sande, MD, PhD 1,2 , Marta Fiocco, PhD 2,3,4 , Edwin F. Dierselhuis, MD, PhD 5 , Demien Broekhuis, MD 1 , Michaël P.A. Bus, MD, PhD 1 , and the LUMiC® Study Group*". The Department of Orthopaedic Surgery, Radboudumc, Nijmegen, The Netherlands, has been added as the affiliation for Edwin F. Dierselhuis, MD, PhD.

Background: We previously reported promising early results for periacetabular tumor reconstructions using the LUMiC prosthesis. The current study evaluates mid-term complications, revision rates, cumulative incidence of implant revision, and risk factors for complications in a multicenter cohort.

Methods: We assessed patients in whom a tumor defect after type P1b+2, P2, P2+3, or P1b+2+3 internal hemipelvectomy was reconstructed with a LUMiC prosthesis during the period of 2008 to 2022. Complications were reported according to the Henderson classification. Competing risks models were used to estimate the cumulative incidence of implant revision for mechanical and nonmechanical reasons, and reoperations for any complication. Cox models were used to study the effect of risk factors on dislocation and infection.

Results: One hundred and sixty-six patients (median follow-up, 4.2 years [interquartile range, 2.6 to 7.6 years]) were included. A total of 114 (69%) were treated for a primary malignant tumor, 46 (28%) for metastatic carcinoma, 5 (3%) for a benign aggressive lesion, and 1 (1%) for another reason. One hundred and sixty-five reoperations were performed in 82 (49%) of the patients; 104 (63%) of the reoperations were within 6 months. Thirty-two (19%) of 166 implants were revised: 13 (8%) for mechanical reasons, mainly dislocation (n = 5, 3%), and 19 (11%) for nonmechanical reasons, mainly periprosthetic joint infection (PJI) (n = 15, 9%). The cumulative incidences of revision for mechanical reasons and PJI (Henderson 1 to 4) at 2, 5, and 10 years were 11% (95% confidence interval [CI], 7% to 17%), 18% (12% to 25%), and 24% (16% to 33%), respectively. Previous surgery at the same site was associated with an increased dislocation risk (cause-specific hazard ratio [HR CS ], 3.0 [95% CI, 1.5 to 6.4]; p < 0.01), and resections involving the P3 region were associated with an increased infection risk (HR CS , 2.5 [95% CI, 1.4 to 4.7]; p < 0.01).

Conclusions: Despite a substantial reoperation risk, the LUMiC prosthesis demonstrated its durability in the mid-term, with a low mechanical revision rate and most patients retaining their primary implant. Most complications occur in the first postoperative months. Patients with previous surgery at the same site had an increased dislocation risk and might benefit from more conservative rehabilitation and aftercare. Measures should be aimed at reducing the PJI risk, especially in resections involving the P3 region.

Level of evidence: Therapeutic Level IV . See Instructions for Authors for a complete description of levels of evidence.

Fig. 1

Cumulative incidence of LUMiC revision for mechanical complications (Henderson 1 to 3) and PJI (Henderson 4), using a competing risks model.

Fig. 2

Cumulative incidence of reoperations for any complication, using a competing risks model.