Impact of preoperative frailty on complications, readmissions, and functional recovery following total knee arthroplasty in elderly patients

Table 1 Pre-matching and post-matching standardized mean differences and characteristics of elderly total knee arthroplasty patients with preoperative frailty and non-frail cohort

Characteristic	Frail pre-match (n = 612)	Non-frail pre-match (n = 2512)	SMD pre-match	Frail post-match (n = 512)	Non-frail post-match (n = 2048)	SMD post-match
Age (mean ± SD)	73.5 ± 5.5	71.2 ± 4.8	0.45	72.3 ± 5.1	72.1 ± 4.9	0.04
Gender: Female (%)	65.0	58.0	0.14	60.9	60.4	0.01
Body mass index (mean ± SD)	32.1 ± 4.2	29.8 ± 3.9	0.56	30.5 ± 4.0	30.4 ± 3.9	0.03
American Society of Anesthesiologists III and IV (%)	45.0	25.0	0.42	30.1	29.0	0.02
Diabetes mellitus (%)	55.0	35.0	0.41	40.6	40.1	0.01
Hypertension (%)	80.0	60.0	0.43	69.9	69.5	0.01
Chronic obstructive pulmonary disease (%)	30.0	15.0	0.35	19.9	20.0	0.00

It presents the baseline characteristics and standardized mean differences for the frail and non-frail groups before and after propensity score matching. The post-matching standardized mean differences, which are all < 0.1, confirm the excellent balance achieved between the two groups for the matching covariates: (1) Age; (2) Gender (female); (3) Body mass index; (4) American Society of Anesthesiologists status (III and IV); (5) Diabetes mellitus; (6) Hypertension; and (7) Chronic obstructive pulmonary disease. Pre-match n adjusted for illustration; actual initial frail n = 512 before exclusions. SMD: Standardized mean difference.

Table 2 Ninety-day postoperative complications between elderly total knee arthroplasty patients with preoperative frailty and a matched-control cohort

Complication	Frailty (%)	Control (%)	Odds ratio	95%CI	P value
Composite	21.1	9.2	2.61	2.05-3.32	< 0.001
Surgical site infection	5.3	1.8	3.12	2.18-4.46	< 0.001
Deep vein thrombosis	4.1	1.5	2.85	1.92-4.23	< 0.001
Pulmonary embolism	3.7	1.0	4.02	2.45-6.59	< 0.001
Periprosthetic fracture	2.0	0.8	2.56	1.45-4.52	0.001

It compares the percentage rate of 90-day complications and their respective adjusted odds ratios (OR) with 95%CI and P value between the frail and control groups. Frail patients had significantly increased odds of all complications, including composite complications (OR = 2.61, P < 0.001), surgical site infection (OR = 3.12, P < 0.001), deep vein thrombosis (OR = 2.85, P < 0.001), pulmonary embolism (OR = 4.02, P < 0.001), and periprosthetic fracture (OR = 2.56, P = 0.001).

Table 3 Healthcare utilization between elderly total knee arthroplasty patients with preoperative frailty and a matched-control cohort

Utilization	Frailty	Control	Odds ratio	95%CI	P value
Readmission (%)	17.0	4.2	4.74	3.56-6.31	< 0.001
Length of stay (days), mean ± SD)	5.6 ± 1.9	3.9 ± 1.5	-	-	< 0.001

It compares metrics of healthcare utilization. Frail patients had significantly prolonged hospital length of stay (5.6 ± 1.9 days vs 3.9 ± 1.5 days; P < 0.001) and a higher 90-day readmission rate (17.0% vs 4.2%; adjusted odds ratio = 4.74, P < 0.001) compared to non-frail controls.

Table 4 One-year functional recovery (Knee Society Score) between elderly total knee arthroplasty patients with preoperative frailty and a matched-control cohort

Outcome	Frailty (mean ± SD)	Control (mean ± SD)	P value
Baseline KSS	45.2 ± 9.8	46.1 ± 9.4	0.312
1-year KSS	75.5 ± 10.2	85.1 ± 8.1	< 0.001
Delta KSS	30.3 ± 11.4	39.0 ± 10.2	< 0.001

It presents the mean ± SD of the Knee Society Score (KSS) outcomes. Baseline KSS was similar between groups (P = 0.312). At one year, the frail group had significantly lower mean 1-year KSS scores (75.5 ± 10.2 vs 85.1 ± 8.1; P < 0.001) and a smaller mean delta KSS (change from baseline) (30.3 ± 11.4 vs 39.0 ± 10.2; P < 0.001) compared to the control group. KSS: Knee Society Score.