Failure of total knee arthroplasty with or without patella resurfacing

Figures & data

Figure 1. Description of the selection procedure.

Table 1. Characteristics of primary total knee arthroplasties reported to the Norwegian Arthroplasty Register from 1994 to 2009

		Numbers of hospitals			Men	< 60 years	Median	OA^a	PO ^b	MB ^c
Prosthesis	n	n	n > 50	n > 100	%	%	age	%	%	%
Patella resurfacing ^d	11,887	70	46	37	28	13	72	81	26	34
Patella resurfaced	2,547	51	12	6	28	16	72	71	27	21
Patella non resurfaced	9,340	69	41	32	28	13	73	83	25	38
Prosthesis brand	25,590	79	61	53	31	15	72	84	27	34
NR AGC Universal	2,123	31	11	7	31	12	73	86	22	0
NR Tricon ^e	633	23	5	1	22	7	73	75	24	0
NR Genesis I	2,304	28	13	7	26	12	73	75	26	0
NR LCS ^f, g	3,526	36	15	10	28	13	72	88	26	100
NR Duracon ^h	1,283	18	10	5	33	14	71	87	30	0
NR NexGen	754	18	4	2	35	17	71	87	26	0
NR Profix	6,304	40	25	20	31	16	71	85	25	0.2
NR LCS Complete	4,090	36	15	13	33	16	70	90	30	100
NR e.motion	434	4	2	1	33	18	70	94	40	100
NR AGC Anatomic	1,298	18	4	4	37	15	70	88	26	0
PR AGC Universal	425	21	3	1	24	13	73	66	25	0
PR Tricon ^e	392	21	3	0	26	21	71	54	29	0
PR Genesis I	704	23	5	1	25	14	71	72	28	0
PR LCS ^i	532	12	2	1	27	13	73	92	23	100
PR Kinemax	294	12	2	0	19	9	74	87	22	0
PR NexGen	494	11	1	1	36	21	69	64	28	0

^a Primary osteoarthritis of the knee (%).

^b Previously operated in the knee (%).

^c Mobile-bearing prostheses (%).

^d Study population restricted to brands represented by both patella resurfaced and patella non resurfaced prostheses.

^e Tricon C or Tricon M femoral component used on the femoral side and Tricon II used on the tibial side.

^f 375 of the primary patella non resurfaced LCS prostheses were used in combination with an LCS Universal tibial component.

^g 3032 with mobile-bearing rotating platform system and 494 with mobile-bearing meniscal system.

^h One Duracon prosthesis had an all-polyethylene tibial component.

ⁱ 323 with mobile-bearing rotating platform system and 209 with mobile-bearing meniscal system.

Figure 2. Time trends in the use of cemented patella resurfaced and patella non resurfaced TKAs in Norway, 1994–2008. 2009 was not included since follow-up was only until December 10, 2009.

Table 2. Follow-up and numbers at risk by years after operation

Prosthesis	1994–2009					1994–2000					2001–2009
	Median follow-up (years)	Number at risk by years after operation				Median follow-up (years)	Number at risk by years after operation				Median follow-up (years)	Number at risk by years after op.
		0	5	10	15		0	5	10	15		0	5
Patella resurfacing ^a
Patella resurfaced	9.1	2,547	1,930	986	76	11	1,831	1,519	986	76	6.0	716	411
Patella non resurfaced	7.2	9,340	6,522	1957	116	10	4,255	3,506	1,957	116	5.8	5,085	3,016
Prosthesis brand
NR AGC Universal	6.9	2,123	1,445	376	14	9.8	841	708	376	14	5.7	1,282	737
NR Tricon	12	633	494	341	66	12	633	494	341	66		–	–
NR Genesis I	9.6	2,304	1,854	913	35	10	1,946	1,581	913	35	7.9	358	273
NR LCS	6.6	3,526	2,573	323	1	9.7	827	716	323	1	6.0	2,699	1,857
NR Duracon	1.7	1,283	248	118	4	11	204	173	118	4	1.5	1,079	75
NR NexGen	3.2	754	156	4	–	10	8	7	4	–	3.2	746	149
NR Profix	4.4	6,304	2,576	51	–	9.2	320	271	51	–	4.2	5,984	2,305
NR LCS Complete	1.9	4,090	59	–	–	–	–	–	–	–	1.9	4,090	59
NR e.motion	4.1	434	76	–	–	–	–	–	–	–	4.1	434	76
NR AGC Anatomic	2.7	1,298	152	15	–	10	31	25	15	–	2.7	1,267	127
PR AGC Universal	9.0	425	353	160	4	10	309	257	160	4	7.2	116	96
PR Tricon	13	392	313	226	67	13	392	313	226	67	–	–	–
PR Genesis I	11	704	581	396	5	11	672	557	396	5	7.2	32	24
PR LCS	8.9	532	446	150	–	9.9	346	293	150	–	7.5	186	153
PR Kinemax	10	294	240	138	30	11	265	214	138	30	7.2	29	26
PR NexGen	4.9	494	237	54	–	10	112	99	54	–	4.0	382	138

^a Study population restricted to brands represented with both patella resurfaced and patella non resurfaced prostheses.

Table 3a. Cox relative revision risk (RR) and survival percentages, estimated with all causes of revision as endpoint

Prosthesis	1994–2009
	n	Revised	RR	95% CI	p	5 yr	10 yr	15 yr	95% CI^a
Patella resurfacing^b,c
Patella non resurfaced	9,340	614	1			94.8	93.0	91.4	90.4–92.4
Patella resurfaced	2,547	172	0.84	0.71–1.00	0.052	95.9	94.3	92.1	90.7–93.6
Prosthesis brands ^d
NR AGC Universal	2,123	120	1			94.9	93.2	–	91.9–94.5
NR Tricon	633	72	1.67	1.24–2.24	0.001	93.4	88.6	85.0	81.7–88.3
NR Genesis I	2,304	222	1.43	1.14–1.79	0.002	92.4	90.4	–	89.1–91.6
NR LCS	3,526	177	0.84	0.67–1.06	0.14	95.8	94.1	–	93.2–95.1
NR Duracon	1,283	56	1.45	1.05–1.99	0.02	94.5	90.6	–	87.4–93.8
NR NexGen	754	23	0.79	0.51–1.24	0.3	95.7	–	–	93.9–97.5
NR Profix	6,304	195	0.66	0.52–0.82	< 0.001	96.7	95.7	–	94.9–96.5
NR LCS Complete	4,090	111	0.94	0.72–1.22	0.6	95.4	–	–	94.2–96.6
NR e.motion	434	2	0.09	0.02–0.37	0.001	99.5	–	–	98.9–100
NR AGC Anatomic	1,298	31	0.66	0.45–0.99	0.04	97.1	–	–	95.9–98.2
PR AGC Universal	425	14	0.48	0.27–0.83	0.009	97.2	96.6	–	94.9–98.4
PR Tricon	392	44	1.28	0.90–1.82	0.2	95.9	92.7	87.5	83.9–91.2
PR Genesis I	704	62	1.17	0.86–1.59	0.3	94.0	92.0	–	90.0–94.0
PR LCS	532	41	1.23	0.86–1.75	0.3	93.1	91.7	–	89.1–94.2
PR Kinemax	294	23	1.14	0.73–1.78	0.6	95.6	92.6	–	89.3–95.8
PR NexGen	494	11	0.40	0.22–0.74	0.004	98.6	96.7	–	94.6–98.8

^a Confidence interval for last reported survival percentage.

^b Cox relative risk estimates and survival percentages with adjustment for age group, sex, diagnosis, previous operation of the knee, and prosthesis brand.

^c Study population restricted to brands represented by both patella resurfaced and patella non resurfaced prostheses.

^d Cox relative risk estimates and survival percentages with adjustment for age group, sex, diagnosis, and previous operation of the knee.

Table 3b. Cox relative revision risk (RR) and survival percentages, estimated with all causes of revision as endpoint

Prosthesis	1994–2000 ^a				2001–2009
	n	Revised	RR	p	n	Revised	RR	p
Patella resurfacing ^b,c
Patella non resurfaced	4,255	349	1		5,085	265	1
Patella resurfaced	1,831	138	0.84	0.09	716	34	0.99	1.0
Prosthesis brands ^d
NR AGC Universal	841	51	1	–	1,282	69	1	–
NR Tricon	633	72	1.84	0.001	0	–	–	–
NR Genesis I	1,946	175	1.46	0.02	358	47	2.01	< 0.001
NR LCS	827	50	0.90	0.6	2,699	127	0.81	0.1
NR Duracon	204	18	1.41	0.2	1,079	38	1.39	0.1
NR NexGen	8	1	–	–	746	22	0.72	0.2
NR Profix	320	12	0.59	0.1	5,984	183	0.62	0.001
NR LCS Complete	0	–	–	–	4,090	111	0.85	0.3
NR e.motion	0	–	–	–	434	2	0.09	0.001
NR AGC Anatomic	31	1	–	–	1,267	30	0.63	0.03
PR AGC Universal	309	13	0.69	0.2	116	1	–	–
PR Tricon	392	44	1.46	0.07	0	–	–	–
PR Genesis I	672	57	1.25	0.2	32	5	–	–
PR LCS	346	20	0.97	0.9	186	21	–	–
PR Kinemax	265	22	1.32	0.3	29	1	–	–
PR NexGen	112	4	–	–	382	7	0.33	0.005

^a Follow-up until December 10, 2009

^{b – d} See Table 3a.

Figure 3. Survival rates (%) for cemented patella resurfaced and patella non-resurfaced TKAs in Norway, 1994–2009; all revisions. Cox regression results with adjustment for age group, sex, diagnosis, previous operation of the knee, and brand of prosthesis.

Table 4. Cox relative revision risk (RR), patella non resurfaced TKAs vs. patella resurfaced primary TKAs by reason for revision, 1994–2009

Reason for revision ^a	Revised patella resurfaced	Revised patella non-resurfaced	RR ^b	95% CI	p
Loose femur	21	65	0.82	0.49–1.39	0.5
Loose tibia	57	149	1.42	1.03–1.95	0.03
Loose patella	13
Dislocation, patella	7	22	0.98	0.42–2.32	1.0
Dislocation, other	4	12	2.22	0.71–6.95	0.2
Instability	30	73	0.96	0.61–1.50	0.9
Malalignment	17	43	1.18	0.65–2.12	0.6
Deep infection	41	93	1.32	0.90–1.95	0.2
Periprosthetic fracture	7	20	1.53	0.64–3.70	0.3
Defect polyethylene ins.	30	16	3.23	1.71–6.11	< 0.001
Pain alone	8	202	0.12	0.06–0.23	<0.001

^a More than one reason for revision may have been reported.

^b Cox relative risk of revision estimates (RR) adjusted for age group, sex, diagnosis, previous operation of the knee, and brand (AGC Universal, Tricon, Genesis I, LCS, and NexGen) are reported for patella resurfaced vs. patella non resurfaced prostheses.

Figure 4. Survival rates (%) for cemented patella resurfaced and patella non resurfaced TKAs in Norway, 1994–2009; specific reasons for revision. Cox regression results with adjustment for age group, sex, diagnosis, previous operation of the knee, and prosthesis brand [(c), (g), (i) without adjustment for brand].

Figure 5. Survival rates (%) for cemented patella resurfaced prosthesis brands in Norway, 1994–2009; all revisions. Cox regression results with adjustment for age group, sex, diagnosis, and previous operation of the knee.

Figure 6. Survival rates (%) for cemented patella non resurfaced prosthesis brands in Norway, 1994–2009; all revisions. Cox regression results with adjustment for age group, sex, diagnosis, and previous operation of the knee.