Does patella resurfacing really matter? Pain and function in 972 patients after primary total knee arthroplasty

Figures & data

Figure 1.  Description of the selection procedure.

Table 2.  Patient characteristics by prosthesis type and prosthesis brand

	AGC	Genesis I	LCS	NexGen	Total
No. of prostheses
Resurfaced	99	132	184	89	504
Non-resurfaced	106	134	180	48	468
All	205	266	364	137	972
No. of hospitals
Resurfaced	16	18	9	3	40
Non-resurfaced	12	21	19	7	47
All					56
Men %
Resurfaced	32	25	32	29	30
Non-resurfaced	32	29	24	40	29
All					29
Mean (SD) age (years) ^a
Resurfaced	76 (7.2)	77 (7.1)	75 (8.1)	74 (8.1)	76 (7.7)
Non-resurfaced	76 (8.6)	78 (6.0)	76 (7.6)	74 (9.3)	76 (7.7)
All					76 (7.7)
Mean (SD) time since operation (years) ^b
Resurfaced	7.2 (2.5)	9.2 (1.7)	6.5 (1.8)	5.2 (1.9)	7.1 (2.4)
Non-resurfaced	7.1 (2.4)	9.2 (1.6)	6.5 (1.8)	3.6 (0.9)	7.1 (2.5)
All					7.1 (2.4)
Charnley Category C %
Resurfaced	57	69	61	66	63
Non-resurfaced	66	65	69	61	66
All					65
Mean (SD) preoperativ EQ-5D index score
Resurfaced	0.48 (0.23)	0.47 (0.22)	0.44 (0.23)	0.45 (0.23)	0.46 (0.23)
Non-resurfaced	0.47 (0.20)	0.46 (0.22)	0.48 (0.23)	0.43 (0.20)	0.47 (0.22)
All					0.46 (0.22)

^a Mean age when completing the questionnaire.

^b Mean time since operation when completing the questionnaire.

Figure 2.  Mean outcome scores for resurfaced and non-resurfaced prostheses. The first 5 outcomes from the left represent the KOOS subscales. Adjustments were made for age, sex, preoperative EQ-5D index score (except for the outcome ΔEQ-5D), Charnley category, and prosthesis brand. Outcomes were measured on a scale from 0 to 100 units (worst to best).

Figure 3.  Mean outcome scores for resurfaced and non-resurfaced prostheses, for each brand of prosthesis. The first 5 outcomes from the left represent the KOOS subscales. Adjustments were made for age, sex, preoperative EQ-5D index score (except for the outcome ΔEQ-5D), and Charnley category. Outcomes were measured on a scale from 0 to 100 units (worst to best).

Table 3.  Mean difference ^a in outcome between resurfaced and non-resurfaced prostheses

Prosthesis brands	Pain		Symptoms		ADL		Sport&Rec		QOL		Pain(VAS)		Satisfaction(VAS)		ΔEQ-5D
	Δ	p	Δ	p	Δ	p	Δ	p	Δ	p	Δ	p	Δ	p	Δ	p
Total ^b	0.8	0.6	0.2	0.9	1.4	0.4	1.4	0.5	–0.2	0.9	-0.1	0.9	–0.7	0.7	0.4	0.9
AGC ^c	0.8	0.8	0.0	1.0	2.1	0.5	0.2	1.0	–1.9	0.6	–0.4	0.9	–1.2	0.7	1.4	0.8
Genesis I ^c	4.0	0.2	0.7	0.8	3.2	0.3	1.8	0.6	1.0	0.8	2.5	0.4	1.1	0.7	6.3	0.1
LCS ^c	–1.3	0.6	–0.1	1.0	0.9	0.8	2.9	0.4	1.0	0.8	–1.3	0.6	–2.5	0.4	–4.1	0.2
NexGen ^c	–0.6	0.9	–1.3	0.7	–4.0	0.4	–4.3	0.4	–4.3	0.4	–1.7	0.6	0.0	1.0	–1.3	0.8

^a Differences (Δ) = mean scores among non-resurfaced prostheses minus mean scores among resurfaced prostheses.

^b Differences in mean are adjusted for age, sex, preoperative EQ-5D index score (except for ΔEQ-5D), Charnley category and prosthesis brand.

^c Differences in mean are adjusted for age, sex, preoperative EQ-5D index score (except for ΔEQ-5D) and Charnley category.

Table 4.  Effects on mean outcome of gender, age, preoperative EQ-5D index score, time since operation, Charnley category and prosthesis brand

Risk predictors	n	Pain		Symptoms		ADL		Sport&Rec		QOL		Pain(VAS)		Satisfaction(VAS)		ΔEQ-5D
		Δ ^a	p	Δ ^a	p	Δ ^a	p	Δ ^a	p	Δ ^a	p	Δ ^a	p	Δ ^a	p	Δ ^a	p
Sex
Male	284	ref		ref		ref		ref		ref		ref		ref		ref
Female	679	–8.6	e	–6.5	e	–10.1	e	–15.5	e	–8.1	e	–6.1	e	–6.7	e	–2.9	0.2
Age (years) ^b
<65	94	ref		ref		ref		ref		ref		ref		ref		ref
65–70	108	7.1	0.05	4.3	0.1	3.5	0.3	1.2	0.8	2.7	0.5	4.0	0.2	0.9	0.8	–0.2	1.0
70–80	396	7.9	0.01	6.3	0.01	1.9	0.5	–1.5	0.7	3.1	0.4	7.3	0.01	1.9	0.6	–4.6	0.2
>80	365	10.9	e	10.0	e	3.5	0.3	1.8	0.6	8.4	0.02	8.5	0.004	3.0	0.4	–11.5	0.002
Preoperative EQ-5D index score ^c
<0.30	286	ref		ref		ref		ref		ref		ref		ref
0.30–0.69	469	8.6	e	4.7	0.002	7.2	e	3.8	0.09	6.0	0.006	5.9	e	3.2	0.1
>0.69	148	11.3	e	9.2	e	12.1	e	9.1	0.004	13.3	e	9.3	e	7.5	0.006
Years since operation	–0.5	0.2	–0.3	0.4	–0.6	0.2	0.2	0.7	–0.1	0.8	–0.1	0.9	0.2	0.8	–0.7	0.2
Charnley Category ^c
A	195	ref		ref		ref		ref		ref		ref		ref		ref
B	128	–13.1	e	–9.1	e	–11.2	e	–17.8	e	–16.8	e	–13.2	e	–8.7	0.004	–11.2	0.001
C	585	–9.6	e	–6.6	e	–14.5	e	–15.3	e	–13.7	e	–10.6	e	–7.3	0.001	–5.9	0.02
Prosthesis brand
AGC ^d	196	ref		ref		ref		ref		ref		ref		ref		ref
Genesis I	266	–1.5	0.5	–2.9	0.2	–2.6	0.3	–3.1	0.3	–4.2	0.2	–3.2	0.2	–3.1	0.2	–0.6	0.8
LCS	364	1.2	0.6	1.2	0.5	0.9	0.7	3.4	0.2	–0.1	1.0	1.9	0.4	0.5	0.8	1.8	0.5
NexGen	137	5.4	0.08	4.9	0.05	4.0	0.2	3.9	0.3	6.5	0.07	8.3	0.004	7.9	0.02	5.9	0.1

^a Differences in mean scores are adjusted for all other variables in a linear regression model.

^b Age when completing the questionnaire.

^c Information on preoperative EQ–5D index score was not given by 60 patients and on Charnley category by 55 patients.

^d 9 patients not having the AGC Universal design were excluded (5 resurfaced and 4 non-resurfaced).

^e p-value < 0.001.

Table 5.  Sensitivity analysis of potential bias due to different revision criteria from resurfaced and non-resurfaced implants

Ratio ^a	Non-resurfaced			Resurfaced			Mann-Whitney
	Primary (n)	Revised (n)	Mean pain ^b	Primary (n)	Revised (n)	Mean pain ^b	p-value
– ^c	468	–	1.6	504	–	1.6	0.9
– ^d	–	145 ^e	2.7	–	23	2.6	0.7
1.0	468	23	1.7	504	23	1.7	1.0
1.3	468	30	1.7	504	23	1.7	0.7
1.5	468	35	1.7	504	23	1.7	0.6
2.0	468	46	1.7	504	23	1.7	0.5
2.5	468	58	1.8	504	23	1.7	0.1
3.0	468	69	1.8	504	23	1.7	0.07
3.5	468	81	1.8	504	23	1.7	0.02

^a Approximate ratio between number of revised non-resurfaced and revised resurfaced

implants.

^b Score from the EQ-5D question regarding pain and discomfort (1=best,3=worst)

^c Original data; patients with unrevised implants.

^d Additional data; patients with revised implants. Among the 23 revised resurfaced prosthe-

ses the most common reasons for revision were loose distal component (n=11) and pain

(n=7). Among the 145 revised non-resurfaced prostheses the most common reasons for

revision were loose distal component (n=38), instability (n=13), deep infection (n=14) and

pain (n=72). More than one reason for revision may have been given.

^e 3 of the 148 available patients with revised non-resurfaced implants failed to answer the

specific EQ-5D question.