The Fate and Relevance of the Patella in Two-Stage Revision Total Knee Arthroplasty for Periprosthetic Joint Infection

Frequent complications of deficient patellae after component removal in 1.5-stage exchange arthroplasty for periprosthetic joint infection

PURPOSE

To compare the complication rates between deficient patellae (remnant patellae after component removal) and non-resurfaced patellae after 1.5-stage exchange arthroplasty for periprosthetic joint infection and identify risk factors for complications of deficient patellae.

METHODS

Complications in 76 deficient patellae (group D) and 41 non-resurfaced patellae (group N), with a minimum follow-up of 2 years after 1.5-stage exchange arthroplasty were retrospectively analyzed. Demographic characteristics and mean follow-up period did not differ between the groups. The incidences of patellar maltracking, erosion, and fractures were investigated. Factors affecting each patellar complication in group D were investigated in terms of age, sex, body mass index, postoperative hip-knee-ankle angle (HKA), remnant patellar thickness, and change in joint line height.

RESULTS

The incidence of patellar maltracking and erosion was significantly higher in group D than in group N [maltracking: 16 (21.1%) vs. 0 cases (0%), p = 0.001; erosion: 30 (39.5%) vs. 2 cases (4.9%), p < 0.001]. The incidence of patellar fractures tended to be higher in group D [6 (7.8%) vs. 0 cases, p = 0.090]. In group D, postoperative valgus limb alignment and younger age were identified as significant factors that affected patellar maltracking (odds ratio of postoperative HKA = 1.628, p = 0.005) and erosion (odds ratio of age = 0.909, p = 0.009), respectively.

CONCLUSION

Close observation of the complications of deficient patellae is required after 1.5-stage exchange arthroplasty. Valgus limb alignment should be avoided to reduce the risk of complications. Early second-stage revision with patellar reconstruction can be considered in younger patients with deficient patellae.

LEVEL OF EVIDENCE

III.

Decreased Patellar Fractures and Subluxation With Patellar Component Replacement at Stage-One Spacer

BACKGROUND

Periprosthetic joint infection is a devastating complication of total knee arthroplasty and is often treated with 2-stage revision. We retrospectively assessed whether replacing the patellar component with articulating stage-one spacers was associated with improved outcomes compared to spacers without patellar component replacement.

METHODS

A total of 139 patients from a single academic institution were identified who underwent an articulating stage-one revision total knee arthroplasty and had at least 1-year follow-up. Of the 139 patients, 91 underwent patellar component removal without replacement, while 48 had a patellar component replaced at stage-one revision. Patellar fracture and reinfection at any point after stage-one were recorded. Knee range of motion (ROM), patellar thickness, lateral tilt, and lateral displacement were measured at 6-weeks post stage-one. Chi-square, Fisher's exact, and t-tests were utilized for comparisons. There were no significant demographic differences between groups.

RESULTS

Patellar component replacement at stage-one revision was associated with fewer patellar fractures (2.1 versus 12.1%, P = .046), less lateral patellar displacement (1.7 versus 16.0 mm, P < .01), and improved pre to postoperative knee ROM 6 weeks after stage-one (+5.9 versus -11.4°, P = .03). There was no difference in reinfections after stage-2 revision for the replaced or unreplaced patellar groups (15.4 versus 15%, P = 1.000). While the mean time between stage-one and stage-2 was not different (5.2 versus 4.5 months, P = .50), at one-year follow-up, significantly more patients in the patellar component replacement group were satisfied and refused stage-2 revision (45.8 versus 3.3%, P < .001).

CONCLUSIONS

Replacing the patellar component at stage-one revision is associated with a decreased rate of patellar fracture and lateral patellar subluxation, improved ROM, and possible increased patient satisfaction, as reflected by nearly half of these patients electing to keep their spacer. There was no difference in reinfection rates between the cohorts.