Which factors affect the ability to kneel following total knee arthroplasty? An outpatient study of 100 postoperative knee replacements

Introduction

Total knee replacement (TKR) has consistently been shown to be a highly effective surgical procedure in terms of pain relief and improved functional outcomes based upon patient questionnaire scoring systems. Of the functional outcomes assessed, kneeling has been shown to be of great importance to patients both prior to TKR and in terms of functional disability postoperatively. ^{1 –3} Studies also suggest that osteoarthritic knees do not kneel well but that this improves following TKR. ^4,5

Kneeling is an important function for many activities of daily living as well as having religious and occupational value. Kneeling plays an important cultural role for patients from Middle-Eastern backgrounds being of significance in religious and social activities. ^4,5 Importantly kneeling has also been shown to be an intermediate position in rising from the floor in the older population. ⁶ Due to the distribution of load placed through the native knee during kneeling those who kneel frequently in their activities of daily living are predisposed to the development of osteoarthritis. ^{7 –11}

Consistently a patient’s perception of their ability to kneel following TKR has been shown to be not concordant with their actual ability, with studies showing between 56% and 68% of patients reporting being unable to kneel when in reality between 64% and 82% were actually able to kneel. ^{12 –14} Of those who reported they were unable to kneel, 49% reported it was fear of harming the prosthesis or lack of information that prevented them from kneeling. ¹² Despite kneeling being considered on the Oxford Knee Score patient outcome questionnaire, these studies have shown that a patient’s perception of their kneeling ability cannot be relied upon.

Despite the value of kneeling to our patients and the value of kneeling functionally postoperatively, relatively little consideration has been given to the implant-based factors which may improve kneeling ability. The aim of this study was to assess whether patella resurfacing has an impact on a patient’s ability to kneeling following TKR. Secondary aims were to review whether posterior cruciate ligament (PCL) retention affects kneeling ability, whether patient’s perception of their kneeling ability matched their actual kneeling ability and also to assess the degree of flexion achievable and its relationship with the patient’s ability to kneel.

Materials and methods

We observed 100 knees in 79 consecutive patients attending outpatient follow-up clinics of four different consultants at a specialist orthopedic hospital at least 5 months following TKR. The implants varied from PCL retaining or PCL sacrificing depending on the surgeon’s preference with the option of a patella resurfacing button.

The patients were asked to fill out a questionnaire asking whether they thought they could kneel, the reasons for not being able to kneel, activities they struggled with if they were unable to kneel, and overall satisfaction score. Overall satisfaction score was obtained using a visual analogue scale from 0 to 10. The notes were the reviewed to determine the date of the operation, whether the patella was resurfaced, whether it was a revision or primary replacement, and whether the PCL was retained or sacrificed. Finally, the degree of passive flexion was determined, and the patient’s ability to kneel was assessed.

The patient was first asked to demonstrate the ability to kneel on a low soft examination couch. If they were able to kneel without difficulty or pain they were then asked to demonstrate their kneeling ability onto a cushion on the floor. They were asked to hold this position for 15 s. If the patient was unable to kneel either on the couch or on the floor due to discomfort or pain they were deemed unable to kneel.

The implants used in our cohort were the AGC and AGC PS (Biomet Inc., Warsaw, Indiana, USA), Vanguard and Vanguard PS (Biomet Inc.,), PFC PS (DePuy International, Leeds, UK), and RHK (Biomet Inc.).

The data were then analyzed to show whether kneeling ability was affected, the patella was native or resurfaced, the PCL was retained or sacrificed, and the knee replacement was a primary or revision prosthesis.

Results

The mean age at clinic appointment of the patients was 71.6 years (range 52.2–96 years) with 56 patients being female and 44 male patients. The mean time from knee replacement to clinic review was 3.3 years (range 0.43–17.9 years). The mean degree of flexion achieved in clinic was 105° (80°–130°). The mean satisfaction score as determined by a visual analogue scale was 7.6 (range 0–10). In those patients seen in clinic less than 12 months postoperatively, the mean range of motion was 104.7° compared to 105.2° in those seen at more than a year postoperatively, a nonsignificant difference on two-tailed t-test analysis (p = 0.879).

Of the knees observed in this study, 84 had undergone primary knee replacement with 16 having undergone revision replacement. Fifty had a PCL-retaining prosthesis implanted, with 50 having a PCL-sacrificing prosthesis. Forty-two underwent patella resurfacing at the time of operation, with 57 keeping the native patella and 1 patient having undergone a patellectomy prior to their knee replacement due to previous trauma to that knee.

Of the 100 knees belonging to 79 patients, 46 reported that prior to their knee replacement they were unable to kneel. Following TKR only 17 perceived they were able to kneel. Table 1 presents the reasons as to why patients reported they didn’t think they could kneel. Table 2 presents activities patients struggled with as a result of being unable to kneel. Of the 100 knees observed in clinic, 60 demonstrated the ability to actually kneel without difficulty.

OPEN IN VIEWER

Table 1.

The results from the free text question as to why those patients whom reported they were unable to kneel reported such.

Why patients are unable to kneel	Number of responses
Pain/uncomfortable	43
Anxiety	16
Can’t get up	14
Stiff	12
Told not to	11
Other joint pain	4
Others	3

OPEN IN VIEWER

Table 2.

The activities as listed by patients in the free text area of which activities they struggle with as a result of being unable to kneel.

Activities patients struggle with	Number of responses
Picking things up	27
Cleaning	21
Gardening	20
Bathing	6
Do it yourself home renovation	5
Occupation	4
Sports	4
Sitting on floor	3
Praying	2

Table 3 presents the results of our study separated into degrees of flexion achievable in clinic at the time of recruitment. The knees that were able to flex to 100° or further were more likely to be able to kneel, while knees with flexion less than 100° were more likely to be unable to kneel. The χ ² statistical analysis data confirm that flexion angle significantly improves a patient’s kneeling ability (p = 0.0148). Assessing the impact of patient age on kneeling ability and degree of flexion, we observed no significant correlation between age and kneeling ability (r = 0.001) and by χ ² analysis after dividing the cohort into youngest and oldest halves, there was no significant impact of age on kneeling ability (p = 0.540)

OPEN IN VIEWER

Table 3.

Results comparing the kneeling ability of patients based upon the maximum degree of flexion achievable in clinic.

Degrees of flexion	Kneeling ability
	Unable	Able
80–89	6	0
90–99	13	11
100–109	8	16
110–119	3	9
120–129	9	20
130–139	0	1

Table 4 presents the results of the study broken down into implant type +/− patella resurfacing, indicating whether the patient was actually able to kneel in clinic on the day of recruitment.

OPEN IN VIEWER

Table 4.

Results comparing implant design with actual kneeling ability.

Implant design	Kneeling ability
	Unable	Able
Sacrificed + Resurfaced	8	17
Retained + Resurfaced	1	16
Sacrificed + Native	14	10
Retained + Native	17	16
Sacrificed + Patellectomy	1	0

Of the 42 knees which had undergone patella resurfacing, 33 demonstrated an ability to kneel with 9 unable to kneel. In comparison, of the 57 knees in which the native patella was retained, 26 were able to kneel with 31 unable to kneel without difficulty. The patient with the previous patellectomy was excluded from analysis. Two-tailed Fisher’s exact test demonstrates that resurfacing a patient’s patella significantly increases the likelihood of a patient being able to kneel following TKR (p = 0.001). Two-sided power analysis for a dichotomous outcome measure with an acceptable significance level of 0.05 and the proportions as above indicates that for a sample size of 42 subjects per group, the resulting power is satisfactory at 0.89. When looking at the native patella and resurfaced patella groups separately, the state of the PCL was shown to have no significant within group impact on the ability to kneel (p = 0.359 and p = 0.06 respectively).

Fifty knees had a PCL retaining prosthesis implanted, of which 32 could kneel and 18 were unable to kneel. A PCL sacrificing prosthesis was used in 50 patients, with 28 of these being able to kneel and the remaining 22 unable to kneel. Two-tailed Fisher’s exact test demonstrated that there was no significant improvement in kneeling ability when comparing PCL retaining versus sacrificing prostheses (p = 0.541). Two-sided power calculation of these results indicates that for a significance level of 0.05 and the proportions as indicated the 50 subject sample size gives an unsatisfactory power of 0.13.

Of the 16 knees which attended clinic for follow-up following revision knee replacement, 8 were able to kneel with the remaining 8 unable. Of the 84 patients with primary knee replacements, 52 were able to kneel and 32 were unable. The difference between revision and primary knee replacements was not statistically significant (p = 0.413). Of the 21 patients presenting with bilateral knee replacements, only 1 (4.7%) was unable to kneel on one side due to difficulties associated with the contralateral knee replacement. A further 6 patients were able to kneel on one side but not on the other, with inability to kneel on the contralateral side not impacting the patient’s ability to kneel on the ipsilateral replacement.

Discussion

In previous studies, Palmer et al. ¹³ found that patella resurfacing had no significant impact on a patient’s ability to kneel, while Hassaballa et al. ¹⁴ showed that patellofemoral replacement in isolation resulted in the patient being less likely to kneel when compared to TKR and unicompartmental knee replacement.

For the first time, our study has shown there is a statistically significant benefit for resurfacing the patella with regard to a patient’s ability to kneel. We have also demonstrated that this benefit occurs independent of whether the PCL is retained or sacrificed. As shown in previous studies in the upright kneeling position, occurring with 90° of knee flexion, the patella and tibial tuberosity act as contact points with the ground, while at deeper degrees of flexion the patella is in contact only with the condyles and as such is not transmitting force. ^13,15 Although not previously demonstrated, the results achieved in this study are unsurprising as with a native patella there is direct contact and force transmitted between the internal patella surface and the prosthetic femoral groove, while in the resurfaced patella this contact is borne by the implanted patella button.

We have also demonstrated that patients with a greater degree of flexion are more likely to be able to kneel. This again is an unsurprising result as to achieve the position of upright kneeling there must be at least 90° of flexion. As such, those patients who have a less flexible knee are more likely to find the action of kneeling uncomfortable and be unable to achieve this position as stiffness of the knee becomes an important factor.

The results of this study have shown again that TKR is an effective operation with regard to patient satisfaction with a mean satisfaction score of 7.6 (range 0–10). According to previous studies, it has also been demonstrated that only 46% of patients, prior to knee replacement, are able to kneel due to the effects of arthritis but following TKR this improved to 60% demonstrating an ability to kneel. ^4,5,13 The discrepancy between perception of kneeling ability and actual kneeling ability has also been shown again with 17% reporting being able to kneel upon questioning compared to the 60% who actually proved their ability to kneel. These figures are in keeping with the previous numbers presented by other studies. ^{12 –14}

In vivo kinematic analysis by Incavo et al. has previously demonstrated that patients are able to safely kneel within the design constraints of both PCL-retaining and sacrificing implant designs. ¹⁶ Lee has demonstrated that in PCL-retaining prosthesis upon kneeling there is a significantly larger medial and lateral tibiofemoral contact area than in a PCL-sacrificing design with no significant difference in the contact pressures observed. ¹⁷ As such with both PCL-retaining and sacrificing prostheses being shown to, on average, achieve flexion greater than 90° it would be expected that there should be no difference in kneeling ability. Our results support this theory, showing no significant differences. Due to the relatively small observed difference between the two designs the risk of a type II error is increased and as such further studies with a larger sample size may be required to truly determine the impact the state of the PCL has on a patient’s ability to kneel following TKR.

The most frequent reason people feel they are unable to kneel is due to the perceived pain they would feel upon kneeling. ¹³ Although previous studies have shown as many as 80% of respondents list advice from professionals or anxiety of harming the prosthesis as factors preventing them from kneeling, only 32.5% of our cohort listed these as reasons. The list of activities in Table 2 listed by the patients who reported being unable to kneel show the impact that being unable to kneel has on quality of life with simple acts such as picking things up off the floor, cleaning, and gardening featuring frequently.

The study represents a pragmatic observational approach to assessing kneeling ability in consecutive patients attending the outpatient clinic, and this has resulted in a range of patients including those with bilateral knee replacements, a variable period of follow-up post-knee replacement and a range of prostheses. Previous reports have demonstrated that that range of motion tends to plateau at around 12 months postoperatively, however in our cohort we noted a nonsignificant difference between the ranges of motion between those seen at less than 1 year postoperatively and those seen at greater than a year postoperatively and as such they were included in analysis. ^18,19 We also found no significant impact of age on degree of flexion in the replaced knee, nor in ability to kneel.

Conclusion

Kneeling remains an important function in patients’ receiving TKR and plays a key role in many activities of daily life directly impacting on a patient’s quality of life. We have shown that patella resurfacing significantly improves a patient’s chance of being able to kneel following TKR although whether PCL retention or sacrifice has any impact remains unclear. Kneeling has been shown to be achievable and safe in the short term, but there remains a lack of evidence regarding the long-term effects of regularly kneeling, in particular in the rate of polyethylene wear and stress loading. As such caution should remain regarding our advice to patients; however, in those who would significantly benefit from being able to kneel we should consider resurfacing the patella.