Incidence, Timing, and Implications of Postoperative Manipulation Under Anesthesia (MUA) Following Cemented Versus...

Recent increasing usage of cementless implants has raised the question of whether cemented versus cementless implants are more associated with the need for manipulation under anesthesia (MUA) following total knee arthroplasty (TKA). The current study utilized a large, national, administrative database to characterize the incidence of MUA after cemented versus cementless TKA and considered the potential association of MUA with revision rates.

A large national database was queried for adult patients who underwent elective TKA. Those who had a contralateral TKA within six months of index surgery were excluded. Those who received cemented versus cementless TKA were determined and exactly matched 4:1 on age, sex, and Elixhauser Comorbidity Index. The 6-month postoperative incidences of MUA were compared between subgroups. The 5-year prosthetic revision rates were compared between those who underwent MUA versus those who did not within each subgroup and between subgroups among those who underwent MUA, using log-rank tests. Of 567,715 TKA patients, cemented implants were used for 539,070 (95.0%) and cementless for 28,645 (5.0%).

Matched cemented and cementless subcohorts included 114,496 and 28,624 patients and had similar postoperative MUA rates of 3.4% for both groups (P = 0.5). Compared to patients who did not undergo MUA, those who did had higher 5-year revision rates within both cemented (7.5 versus 2.7%, P < 0.01) and cementless groups (7.1 versus 3.0%, P < 0.01). Among patients who underwent MUA, revision rates were similar between cemented and cementless groups (7.5 versus 7.1%, P = 1).

The 6-month postoperative incidences of MUA after TKA were equivalent between large, matched cemented and cementless populations (3.4% for both). While MUA was associated with decreased 5-year prosthetic survivorship for both cemented and cementless prostheses (underscoring its importance), it did not disproportionately impact rates for one versus the other.