Advertisement

Patient-Reported Outcomes and Satisfaction 1 to 3 Years After Revisions of Total Knee Arthroplasties for Unexplained Pain Versus Aseptic Loosening

Open AccessPublished:October 15, 2022DOI:https://doi.org/10.1016/j.arth.2022.10.019

      Abstract

      Background

      It is unknown if patients are relieved of pain after knee arthroplasty revision for unexplained pain. The aim of this cross-sectional case-control study was to compare patient-reported outcome measures (PROMs) and satisfaction 1 to 3 years after revision of total knee arthroplasties (TKAs) for the indications of unexplained pain versus aseptic loosening.

      Methods

      We included 384 patients undergoing TKA revision for the indications of unexplained pain and aseptic loosening from January 1, 2018 to December 31, 2020 from the Danish Knee Arthroplasty Register. A total of 81 patients were revised for unexplained pain and 303 for aseptic loosening. Questionnaires including PROMs (Oxford Knee Score, EQ-5D-5L, and Forgotten Joint Score) and satisfaction with the surgery on a 0-100 scale (100 = not satisfied; 0 = very satisfied) were sent to digitally secured mailboxes. Time from revision to data collection was a median 3.1 years (range, 1.4-4.4 years).

      Results

      Median Oxford Knee Score was 25 (interquartile range [IQR] 15) versus 31 (IQR 18) 1-3 years after revisions for unexplained pain versus aseptic loosening, P = .009. Median EQ-5D-5L was 0.6 (IQR 0.4) versus 0.8 (IQR 0.3) for unexplained pain versus aseptic loosening, P = .009. Median Forgotten Joint Score was 50 (IQR 7) versus 50 (IQR 16) for unexplained pain versus aseptic loosening, P = .905. Satisfaction was 75 (IQR 38) for unexplained pain and 50 (IQR 73) for aseptic loosening, P < .001.

      Conclusion

      Patients undergoing TKA revision for the indication of unexplained pain had worse results on PROMs than those revised for aseptic loosening. Likewise, patients revised for unexplained pain were less satisfied compared to patients revised for aseptic loosening. This information is valuable to both surgeons and patients when candidates for revision surgery are selected, to obtain the best possible outcomes.

      Keywords

      The number of knee arthroplasty revisions performed annually is increasing []. It is well known that about 20% of patients undergoing primary knee arthroplasty experience persistent pain afterward but the proportion after revision might be even higher [
      • Lewis G.N.
      • Rice D.A.
      • McNair P.J.
      • Kluger M.
      Predictors of persistent pain after total knee arthroplasty: a systematic review and meta-analysis.
      ,
      • Petersen K.K.
      • Simonsen O.
      • Laursen M.B.
      • Nielsen T.A.
      • Rasmussen S.
      • Arendt-Nielsen L.
      Chronic postoperative pain after primary and revision total knee arthroplasty.
      ]. Some patients are revised because of unexplained pain without any other obvious knee pathology present, but it is unknown if these patients are relieved of pain after surgery. The indication of unexplained pain is controversial and generally not recommended but still widely used [
      • Vince K.G.
      The problem total knee replacement: systematic, comprehensive and efficient evaluation.
      ,
      • Reichel F.
      • Innmann M.
      • Gotterbarm T.
      • Schiltenwolf M.
      • Merle C.
      Predictors for persistent pain and dissatisfaction after total knee arthroplasty.
      ]. Although revision knee surgery for aseptic reasons may be as safe as primary surgery, patients are less satisfied after revision [
      • Lindberg-Larsen M.
      • Jørgensen C.C.
      • Bæk Hansen T.
      • Solgaard S.
      • Odgaard A.
      • Kehlet H.
      Re-admissions, re-operations and length of stay in hospital after aseptic revision knee replacement in Denmark: a two-year nationwide study.
      ,
      • Roman M.D.
      • Russu O.
      • Mohor C.
      • Necula R.
      • Boicean A.
      • Todor A.
      • et al.
      Outcomes in revision total knee arthroplasty (Review).
      ,
      • Lindberg-Larsen M.
      • Petersen P.B.
      • Corap Y.
      • Gromov K.
      • Jørgensen C.C.
      • Kehlet H.
      Fast-track revision knee arthroplasty.
      ]. The use of opioids and other analgesics does not seem to decrease considerably after revision because of pain, suggesting a lack of effect on pain relief [
      • Arndt K.B.
      • Schrøder H.M.
      • Troelsen A.
      • Lindberg-Larsen M.
      Opioid and analgesic use before and after revision knee arthroplasty for the indications "pain without loosening" versus "aseptic loosening" - a Danish nationwide study.
      ]. This also applied to patients revised for aseptic loosening. However, they might have had a mechanical problem, which was solved by revision.
      Investigations of patient’s perspectives are essential when it comes to estimations of pain and life quality. Data on patient-reported outcomes (PROs) after revision knee arthroplasty are limited [
      • Sabah S.A.
      • Alvand A.
      • Knight R.
      • Beard D.J.
      • Price A.J.
      Patient-reported function and quality of life after revision total knee arthroplasty: an analysis of 10,727 patients from the NHS PROMs program.
      ,
      • Baker P.
      • Cowling P.
      • Kurtz S.
      • Jameson S.
      • Gregg P.
      • Deehan D.
      Reason for revision influences early patient outcomes after aseptic knee revision.
      ]. It is unknown if the patients revised because of unexplained pain are as satisfied with the results as patients revised for the more well-established indication of aseptic loosening. Further knowledge of revisions performed because of unexplained pain is warranted to improve the selection of candidates for revision surgery.
      Therefore, the aim of this study was to compare patient-reported outcome measures (PROMs) and satisfaction 1 to 3 years after revision of total knee arthroplasties (TKAs) for the indications of unexplained pain versus aseptic loosening.

      Methods

      Study Design

      This cross-sectional nationwide case-control study was conducted in accordance with the consensus-based standards for the selection of health measurement instruments reporting guidelines for PROM studies [
      • Gagnier J.J.
      • Lai J.
      • Mokkink L.B.
      • Terwee C.B.
      COSMIN reporting guideline for studies on measurement properties of patient-reported outcome measures.
      ].

      Participants and Data Sources

      Data on all knee arthroplasty revisions registered for the indication of unexplained pain or aseptic loosening exclusively in the period January 1, 2018 to December 31, 2020 were collected from the Danish Knee Arthroplasty Register (DKR). The DKR is a nationwide clinical database collecting data on primary and revision knee arthroplasties in Denmark since 1997 [
      • Pedersen A.B.
      • Mehnert F.
      • Odgaard A.
      • Schrøder H.M.
      Existing data sources for clinical epidemiology: the Danish Knee Arthroplasty Register.
      ]. All orthopaedic departments, including private hospitals, report preoperative and intraoperative data to the database. The completeness of the register was 97% for primary knee arthroplasties and 92% for revision knee arthroplasties in 2020 []. We retrieved demographic data on age, gender, and body mass index from the DKR. A total of 384 patients were included in the study (Fig. 1). There were 81 patients revised for the indication of unexplained pain and 303 for the indication of aseptic loosening. The overall response rate was 68%. The demographic characteristics were overall alike for responders revised for unexplained pain versus aseptic loosening (Table 1).
      Figure thumbnail gr1
      Fig. 1Flowchart of patients included/excluded in this study. DKR, Danish knee arthroplasty register; UKA, unicompartmental knee arthroplasty; TKA, total knee arthroplasty; PROM, patient-reported outcome measure.
      Table 1Demographic Characteristics of Included Patients for Responders and Nonresponders of Patient-Reported Outcomes.
      CharacteristicRespondersP ValueNonrespondersP Value
      PainAseptic LooseningPainAseptic Loosening
      n = 56 (69%)n = 206 (68%)n = 25 (31%)n = 97 (32%)
      Mean age in years (range)65 (29-82)69 (47-91).00365 (44-80)69 (43-92).039
      Women (%)31 (55%)123 (60%).55717 (68%)56 (58%).350
      BMI (Median [IQR]) (3 missing values)24 (IQR 13)22 (IQR 12).03221 (IQR 10)22 (IQR 12).649
      Surgical subgroup.572.137
       1. TKA-TKA49 (88%)174 (84%)16 (64%)76 (78%)
       2. Partial revision7 (12%)32 (16%)9 (36%)21 (22%)
      Time from primary surgery to revision (Mean [SD])4.7 (SD 3.9)6.4 (SD 5.5).3133.9 (SD 3.5)7.3 (SD 5.8).530
      Follow-up (Years from revision to data collection) (Median [range])3.4 (1.5-4.4)3 (1.4-4.4).0093 (1.6-4.4)2.9 (1.4-4.4).222
      PROM, Patient-Reported Outcome Measure; SD, standard deviation; BMI, body mass index; IQR, interquartile range; TKA, total knee arthroplasty; UKA, unicompartmental knee arthroplasty.
      We selected the study period 2018-2020, so that time from revision would not exceed 4 years. PROs from patients revised before this time period might be memory-biased and thus not relevant for this study. We included all revisions registered in the DKR for the two indications of investigation and therefore no sample size calculation was performed. We divided the revisions into surgical subgroups defined by the types of prostheses removed and inserted at surgery. We included surgical subgroup 1 (total revision of both femoral and tibial component in a TKA to new TKA) and surgical subgroup 2 (partial revision of either femoral or tibial component in a TKA). The excluded surgical subgroups were not relevant for the study and the number of revisions in most of the excluded groups was too low to perform a meaningful analysis (Fig. 1).

      Outcomes

      Data Collection

      All included patients received an e-mail with a link to an electronic questionnaire in a secured digital mailbox, which linked to the patient’s Danish personal registration number. If the questionnaires were not answered within 2 weeks, two reminder e-mails were sent with a 2-week interval. Patients who were not registered to the digital mailbox received a paper version of the questionnaire by postal mail. Paper versions were also sent on request. Study data were collected and managed using REDCap electronic data capture tools hosted at Odense Explorative Patient data Network (OPEN), Odense, Denmark [
      • Harris P.A.
      • Taylor R.
      • Thielke R.
      • Payne J.
      • Gonzalez N.
      • Conde J.G.
      Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support.
      ,
      • Harris P.A.
      • Taylor R.
      • Minor B.L.
      • Elliott V.
      • Fernandez M.
      • O'Neal L.
      • et al.
      The REDCap consortium: building an international community of software platform partners.
      ].

      Questionnaires

      We included the standardized questionnaires Oxford Knee Score (OKS), EQ-5D-5L, Forgotten Joint Score (FJS), and Copenhagen Knee range of motion (ROM) Scale, and we further asked questions about pain, satisfaction, and reason for revision.

      Oxford Knee Score

      OKS was calculated from the validated joint specific 12-item questionnaire developed in 1998 to measure outcomes after TKA and it was translated into Danish in 2009 [
      • Dawson J.
      • Fitzpatrick R.
      • Murray D.
      • Carr A.
      Questionnaire on the perceptions of patients about total knee replacement.
      ,
      • Anne Mørup-Petersen M.K.
      • Nielsen R.
      • Paulsen A.
      • Anders O.
      Translation and classical test theory validation of the Danish version of the Oxford Knee Score.
      ]. A score of 0 to 48 was calculated, with 48 being the best possible score. Calculation of the OKS followed recommendations from the developers [
      • Murray D.W.
      • Fitzpatrick R.
      • Rogers K.
      • Pandit H.
      • Beard D.J.
      • Carr A.J.
      • et al.
      The use of the Oxford hip and knee scores.
      ].

      EQ-5D-5L

      EQ-5D-5L consists of a 5-item questionnaire and the EQ visual analogue scale (EQ VAS) designed to measure health state. The Danish edition was validated in 2021 [
      • Jensen M.B.
      • Jensen C.E.
      • Gudex C.
      • Pedersen K.M.
      • Sørensen S.S.
      • Ehlers L.H.
      Danish population health measured by the EQ-5D-5L.
      ]. The EQ Index was calculated from the United Kingdom value set, which was developed from a population sample from 6 countries including Denmark [
      • van Hout B.
      • Janssen M.F.
      • Feng Y.S.
      • Kohlmann T.
      • Busschbach J.
      • Golicki D.
      • et al.
      Interim scoring for the EQ-5D-5L: mapping the EQ-5D-5L to EQ-5D-3L value sets.
      ].

      Forgotten Joint Score

      FJS was calculated from the 12-item questionnaire developed in 2012 and translated and validated in Danish in 2016 [
      • Behrend H.
      • Giesinger K.
      • Giesinger J.M.
      • Kuster M.S.
      The “forgotten joint” as the ultimate goal in joint arthroplasty: validation of a new patient-reported outcome measure.
      ,
      • Thomsen M.G.
      • Latifi R.
      • Kallemose T.
      • Barfod K.W.
      • Husted H.
      • Troelsen A.
      Good validity and reliability of the forgotten joint score in evaluating the outcome of total knee arthroplasty.
      ]. A total score of 0-100 was obtained. A high score indicated a high degree of “forgetting” the artificial joint. The FJS is an efficient tool for evaluation of small differences in knee performance after surgery. The FJS score was calculated following the instructions by the developers [
      • Behrend H.
      • Giesinger K.
      • Giesinger J.M.
      • Kuster M.S.
      The “forgotten joint” as the ultimate goal in joint arthroplasty: validation of a new patient-reported outcome measure.
      ].

      Copenhagen Knee Range of Motion

      We used the Copenhagen Knee ROM Scale to estimate the ROM of the revised knees [
      • Mørup-Petersen A.
      • Holm P.M.
      • Holm C.E.
      • Klausen T.W.
      • Skou S.T.
      • Krogsgaard M.R.
      • et al.
      Knee osteoarthritis patients can provide useful estimates of passive knee range of motion: development and validation of the copenhagen knee ROM scale.
      ]. The patients reported ROM from the 2-item scale with 11 illustrations of knee motion.

      Pain

      We asked the patients about their level of pain. The answers were reported on a VAS.“What was your average pain level the last month on a 0 to 100 scale” (0 = no pain; 100 = worst pain imaginable)

      Satisfaction

      We asked questions about the satisfaction after surgery.“How satisfied are you with the result of the surgery on a 0 to 100 scale” (0 = very satisfied; 100 = not satisfied)Improvement. “How are your knee problems now compared to prior to the operation?”

      Data Analyses

      Data were presented with means and standard deviations (SDs) for normally distributed continuous variables and median and interquartile range (IQR) for non-normally distributed continuous variables. Distributions were inspected for normality via quantile-quantile plots. Frequency counts and percentages were provided for categorical variables. Pearson’s Chi-squared tests were used to test for statistical differences between categorical measures. Wilcoxon Rank-Sum tests were used to test non-normally distributed continuous variables for statistical differences.
      Missing data of the respective PROMs were handled as recommended by the developers [
      • Murray D.W.
      • Fitzpatrick R.
      • Rogers K.
      • Pandit H.
      • Beard D.J.
      • Carr A.J.
      • et al.
      The use of the Oxford hip and knee scores.
      ,
      • Behrend H.
      • Giesinger K.
      • Giesinger J.M.
      • Kuster M.S.
      The “forgotten joint” as the ultimate goal in joint arthroplasty: validation of a new patient-reported outcome measure.
      ,].
      Statistical significance was set at the 5% level. For all analyses, we used Stata Statistical Software: Release 17. College Station, Texas: StataCorp LLC.

      Ethics and Funding

      Permission from the Danish Data Protection Agency was achieved (Journal no. 19/14,416). We achieved accept to contact the patients in our study from the Head of Departments of all included departments performing the revisions.
      The authors had no conflicts of interest to declare.

      Results

      Patient-Reported Outcome Measures

      Median OKS was 25 (IQR 15) for unexplained pain versus 31 (IQR 18) for aseptic loosening 1-3 years after revision, P = .009 (Table 2 and Fig. 2). Median EQ-5D-5L was 0.6 (IQR 0.4) for unexplained pain and 0.8 (IQR 0.3) for aseptic loosening 1-3 years after revision, P = .009 (Table 2 and Fig. 3). There were no differences in these scores within indication groups comparing revisions after 1-2 years versus > 2 to 3 years (Supplementary Table 1).
      Table 2Patient-Reported Outcomes of Patients Revised for the Indication of Unexplained Pain Versus Aseptic Loosening.
      PROMUnexplained PainAseptic LooseningP Value
      N = 56N = 206
      Oxford Knee Score
      Median (IQR).
      25 (IQR 15)31 (IQR 18).009
      Wilcoxon Rank-Sum test.
      EQ-5D-5L Index
      Median (IQR).
      0.6 (IQR 0.4)0.8 (IQR 0.3).009
      Wilcoxon Rank-Sum test.
      EQ VAS
      Median (IQR).
      61 (IQR 37)60 (IQR 40).276
      Wilcoxon Rank-Sum test.
      FJS
      Median (IQR).
      50 (IQR 7)50 (IQR 16).905
      Wilcoxon Rank-Sum test.
      Copenhagen Knee ROM
       Flexion
      Median (IQR).
      5 (IQR 1)5 (2).035
      Wilcoxon Rank-Sum test.
       Flexion deficit (0-4)23 (41%)54 (26%).035
       Extension
      Median (IQR).
      4 (IQR 1)4 (IQR 2).083
      Wilcoxon Rank-Sum test.
       Extension deficit (0-3)21 (42%)61 (29%).258
      If no statistical test is mentioned for P values, Chi-squared test was used.
      PROM, Patient-reported outcome measure; IQR, interquartile range; EQ-5D-5L Index, a value of 1 indicates the best quality of life and 0 indicate the worst; EQ VAS, EuroQol Visual Analogue Scale: 100 = best health imaginable and 0 = worst health imaginable; Copenhagen Knee ROM = Copenhagen Knee Range of Motion.
      a Median (IQR).
      b Wilcoxon Rank-Sum test.
      Figure thumbnail gr2
      Fig. 2OKS for the indications of revision unexplained pain and aseptic loosening presented as kernel curves. OKS, oxford knee score.
      Figure thumbnail gr3
      Fig. 3EQ-5D Index for the indications of revision unexplained pain and aseptic loosening presented as kernel curves.
      There were no differences in EQ VAS and FJS between indication groups. The flexion ability estimated by the Copenhagen Knee ROM was slightly better for revisions for aseptic loosening than for unexplained pain (Table 2). PROMs at a surgical subgroup level showed similar results (Supplementary Table 2).

      Pain

      The average pain score was significantly worse for unexplained pain than aseptic loosening, P = .008 (Table 3 and Fig. 4).
      Table 3Questions on Pain and Satisfaction.
      QuestionUnexplained PainAseptic LooseningP Value
      n = 56n = 206
      Pain
       What was your average pain level the last month on a 0-100 scale; 0 = no pain; 100 = worst pain imaginable
      Median (IQR).
      62 (IQR 41)45 (IQR 56).008
      Wilcoxon Rank-Sum test.
      Satisfaction
       How satisfied are you with the result of the surgery on a 0-100 scale; 0 = very satisfied; 100 = not satisfied
      Median (IQR).
      75 (IQR 38)50 (IQR 73)< .001
      Wilcoxon Rank-Sum test.
       Improvement. How are your knee problems now compared to prior to the operation?.356
      Importantly improved25 (69%)116 (77%)
      Not importantly improved11 (31%)35 (23%)
       Do you find your present situation acceptable considering your daily level of function?.005
      Yes19 (35%)109 (57%)
      No35 (65%)83 (43%)
       The question was only asked to patients replying no to the above: Do you think the treatment has failed?.790
      Yes21 (64%)53 (66%)
      No12 (36%)27 (34%)
       Would you go through the surgery again?.263
      Yes19 (35%)91 (47%)
      Maybe21 (39%)67 (35%)
      No14 (26%)36 (18%)
      If no statistical test is mentioned for P values, Chi-squared test was used.
      IQR, interquartile range.
      a Median (IQR).
      b Wilcoxon Rank-Sum test.
      Figure thumbnail gr4
      Fig. 4Average level of pain (0 = no pain; 100 = worst pain imaginable) presented as kernel curves for the indications of revision unexplained pain and aseptic loosening.

      Satisfaction

      The average satisfaction score was significantly worse for unexplained pain than aseptic loosening, P < .001 (Table 3 and Fig. 5). Patients revised for unexplained pain were also less likely to find their knee problem importantly improved or their daily level of function acceptable (Table 3). Scores for pain and satisfaction at a surgical subgroup level showed similar results (Supplementary Table 3).
      Figure thumbnail gr5
      Fig. 5Average level of satisfaction with the surgery (0 = very satisfied; 100 = not satisfied) presented as kernel curves for the indications of revision unexplained pain and aseptic loosening.

      Discussion

      This was a nationwide study of PROM and satisfaction data from 384 patients collected 1 to 3 years after knee arthroplasty revision for the indications of unexplained pain versus aseptic loosening. We found significantly lower OKS and EQ-5D-5L Index scores for patients revised for unexplained pain. Patients revised for unexplained pain were less satisfied with the result of the surgery. There were 69% revised for pain versus 77% revised for aseptic loosening who considered the result of the surgery an important improvement. The author group has previously conducted a survival study including the same indications as in this study: unexplained pain and aseptic loosening. We found similar rerevision rates between groups [
      • Leta T.H.
      • Lygre S.H.
      • Skredderstuen A.
      • Hallan G.
      • Gjertsen J.E.
      • Rokne B.
      • et al.
      Outcomes of unicompartmental knee arthroplasty after aseptic revision to total knee arthroplasty: a comparative study of 768 TKAs and 578 UKAs revised to TKAs from the Norwegian arthroplasty register (1994 to 2011).
      ].

      Patient-Reported Outcome Measures

      Baker et al investigated PROMs of a cohort of 996 revision patients recorded by the National Joint Registry for England and Wales from 2008 to 2010 [
      • Baker P.
      • Cowling P.
      • Kurtz S.
      • Jameson S.
      • Gregg P.
      • Deehan D.
      Reason for revision influences early patient outcomes after aseptic knee revision.
      ]. Mean postrevision OKS was 26.4 (95% CI 23.5 to 29.3) for unexplained pain and 27.8 (95% CI 26.6 to 28.9) for aseptic loosening/lysis. Sabah et al investigated a cohort of 10,727 revision patients recruited from the UK National Health Service PROMs dataset from 2013 to 2019. They reported postrevision OKS of 29.0 without a specification of indications [
      • Sabah S.A.
      • Alvand A.
      • Knight R.
      • Beard D.J.
      • Price A.J.
      Patient-reported function and quality of life after revision total knee arthroplasty: an analysis of 10,727 patients from the NHS PROMs program.
      ]. These scores are concordant with those of our study for both indication groups, although we did find a statistical significant difference between groups.
      EQ-5D Index values of 0.5-0.7 after revision have been reported [
      • Sabah S.A.
      • Alvand A.
      • Knight R.
      • Beard D.J.
      • Price A.J.
      Patient-reported function and quality of life after revision total knee arthroplasty: an analysis of 10,727 patients from the NHS PROMs program.
      ,
      • Baker P.
      • Cowling P.
      • Kurtz S.
      • Jameson S.
      • Gregg P.
      • Deehan D.
      Reason for revision influences early patient outcomes after aseptic knee revision.
      ,
      • Malviya A.
      • Bettinson K.
      • Kurtz S.M.
      • Deehan D.J.
      When do patient-reported assessments peak after revision knee arthroplasty?.
      ]. We reported higher values in our study. Index values differ among populations and the average EQ-5D Index of the Danish population is 0.9; thus, the revised patients had a worse quality of life than expected of Danish citizens [
      • Jensen M.B.
      • Jensen C.E.
      • Gudex C.
      • Pedersen K.M.
      • Sørensen S.S.
      • Ehlers L.H.
      Danish population health measured by the EQ-5D-5L.
      ].
      A larger proportion of patients revised for unexplained pain (41%) had a flexion deficit than patients revised for aseptic loosening (26%) estimated by the Copenhagen Knee ROM. A study investigating a cohort of patients revised for unexplained pain also found a large proportion of pain patients with a decreased ROM [
      • Hossain F.
      • Patel S.
      • Haddad F.S.
      Midterm assessment of causes and results of revision total knee arthroplasty.
      ]. The study found poor results of revisions for unexplained pain, especially for those with normal ROM. We did not detect differences in PROMs 1-2 years versus > 2-3 years after revision in both indication groups. This was expected because PROMs after revision have been shown to peak and stabilize after 1 year for most patients [
      • Greidanus N.V.
      • Peterson R.C.
      • Masri B.A.
      • Garbuz D.S.
      Quality of life outcomes in revision versus primary total knee arthroplasty.
      ]. The PROMs were similar among the surgical subgroups. This could indicate a stronger influence of the indication for revision on the outcomes than the influence of the surgical subgroups.

      Satisfaction

      Satisfaction rates of 72%-88% after aseptic revisions have been reported [
      • Sabah S.A.
      • Alvand A.
      • Knight R.
      • Beard D.J.
      • Price A.J.
      Patient-reported function and quality of life after revision total knee arthroplasty: an analysis of 10,727 patients from the NHS PROMs program.
      ,
      • Arndt K.B.
      • Schrøder H.M.
      • Troelsen A.
      • Lindberg-Larsen M.
      Prosthesis survival after revision knee arthroplasty for “pain without loosening” versus “aseptic loosening”: a Danish nationwide study.
      ]. Baker et al reported satisfaction rates after revision of 58% versus 72% of patients revised for unexplained pain versus aseptic loosening [
      • Baker P.
      • Cowling P.
      • Kurtz S.
      • Jameson S.
      • Gregg P.
      • Deehan D.
      Reason for revision influences early patient outcomes after aseptic knee revision.
      ]. Sabah et al concluded that two-thirds of the patients achieved a clinically meaningful improvement in joint function [
      • Sabah S.A.
      • Alvand A.
      • Knight R.
      • Beard D.J.
      • Price A.J.
      Patient-reported function and quality of life after revision total knee arthroplasty: an analysis of 10,727 patients from the NHS PROMs program.
      ]. These results correspond to those of our study and it seems that patients revised because of pain are generally less satisfied.

      Strengths and Limitations

      This was a nationwide study contributing with important information on PROs after knee arthroplasty revisions. Our study provides valuable data on what can be achieved after revision for unexplained pain and aseptic loosening.
      We did not have prerevision PROs available for this study, as these are not captured in the DKR; thus, the delta change in PROs may actually not be different between groups, which is a major limitation. In addition, it is unknown if the indication groups differed at baseline or improved/worsened equally. The additional questions asked in this study do bring some information of the patient self-reported improvements or worsening, although memory bias might influence the answers.
      The response rate of 68% may be acceptable in a nationwide study; however, results were less certain when responses were not complete. There are missing data in this study, as all patients did not complete or answer every item of each questionnaire. We accounted for this in the data analyses but it could potentially skew the results.

      Conclusion

      Patients undergoing TKA revision for the indication of unexplained pain had worse results on PROMs than those revised for aseptic loosening. Likewise, patients revised for unexplained pain were less satisfied compared to patients revised for aseptic loosening. This information is valuable to both surgeons and patients when candidates for revision surgery are selected, to obtain the best possible outcomes.

      Acknowledgments

      We acknowledge Odense Patient data Explorative Network (OPEN) for their contribution of data management and REDCap services to this study.
      Supplementary Table 1Patient-Reported Outcome Measures Presented as Total Scores for Unexplained Pain and Aseptic Loosening and as Scores in Two Time Periods Defined as Time From Revision to Data Collection (1-2 Y Versus 3 to < 4 Y) for Each Indication.
      PROM Outcome MeasureUnexplained PainP (1-2Y Versus > 2-3)Aseptic LooseningP (1-2Y Versus > 2-3Y)P (Pain Versus AL)
      Total n (%) n = 561-2 Y n (%)>2-3 Y n (%)Total n (%)1-2 Y n (%)>2-3 Y n (%)
      n = 23n = 33n = 206n = 104n = 102
      Oxford Knee Score
      Median (IQR).
      25 (15)26 (16)24 (12)0.667
      Wilcoxon Rank-Sum test.
      31 (18)31 (19)31 (17)0.338
      Wilcoxon Rank-Sum test.
      0.009
      Wilcoxon Rank-Sum test.
      EQ-5D-5L Index
      Median (IQR).
      0.6 (0.4)0.7 (0.3)0.6 (0.4)0.311
      Wilcoxon Rank-Sum test.
      0.8 (0.3)0.8 (0.4)0.8 (0.3)0.176
      Wilcoxon Rank-Sum test.
      0.009
      Wilcoxon Rank-Sum test.
      EQ VAS
      Median (IQR).
      ; 100 best health imaginable-0 worst health imaginable
      61 (37)65 (32)50 (41)0.306
      Wilcoxon Rank-Sum test.
      60 (40)62 (32)53 (40)0.257
      Wilcoxon Rank-Sum test.
      0.276
      Wilcoxon Rank-Sum test.
      FJS
      Median (IQR).
      50 (7)50 (9)48 (10)0.112
      Wilcoxon Rank-Sum test.
      50 (16)50 (13)48 (16)0.61
      Wilcoxon Rank-Sum test.
      0.905
      Wilcoxon Rank-Sum test.
      Copenhagen Knee ROM
       Flexion
      Median (IQR).
      5 (1)4 (2)5 (2)0.122
      Wilcoxon Rank-Sum test.
      5 (2)5 (1)5 (2)0.706
      Wilcoxon Rank-Sum test.
      0.035
      Wilcoxon Rank-Sum test.
       Flexion deficit (0-4)23 (41%)12 (52%)11 (33%)0.24254 (26%)24 (23%)30 (29%)0.3540.035
       Extension
      Median (IQR).
      4 (1)4 (1)4 (1)0.998
      Wilcoxon Rank-Sum test.
      4 (2)4 (1)4 (2)0.647
      Wilcoxon Rank-Sum test.
      0.083
      Wilcoxon Rank-Sum test.
       Extension deficit (0-3)21 (42%)8 (47%)13 (38%)0.67561 (29%)31 (29%)30 (28%)0.9090.258
      Additional questions
       What do you think was the reason for reoperation of your knee?
      Pain23 (43%)12 (52%)11 (35%)51 (27%)29 (30%)22 (23%)
      Loosening of the components7 (13%)2 (9%)5 (16%)83 (43%)42 (43%)41 (43%)
      Instability8 (15%)3 (13%)5 (16%)21 (11%)6 (6%)15 (16%)
      Decreased range of motion13 (24%)6 (26%)7 (23%)16 (8%)9 (9%)7 (7%)
      Other3 (5%)0 (0%)3 (10%)22 (11%)12 (12%)10 (11%)
       What was your average pain level the last month on a 0-100 scale; 0 = no pain; 100 = worst pain imaginable
      Median (IQR).
      62 (41)50 (38)72 (31)0.061
      Wilcoxon Rank-Sum test.
      45 (56)50 (59)35 (55)0.228
      Wilcoxon Rank-Sum test.
      0.008
      Wilcoxon Rank-Sum test.
       How satisfied are you with the result of the surgery on a 0-100 scale; 0 = very satisfied; 100 = not satisfied
      Median (IQR).
      75 (38)73 (30)81 (42)0.343
      Wilcoxon Rank-Sum test.
      50 (73)50 (82)35 (69)0.141
      Wilcoxon Rank-Sum test.
      < 0.001
      Wilcoxon Rank-Sum test.
       How are your knee problems now compared to prior to the operation?0.042
      Better, an important improvement13 (25%)5 (23%)8 (26%)87 (45%)42 (43%)45 (48%)
      Somewhat better, but enough to be an important improvement12 (23%)8 (36%)4 (13%)29 (15%)15 (16%)14 (15%)
      Very small change, not enough to be an important improvement4 (8%)1 (4%)3 (10%)14 (7%)9 (9%)5 (5%)
      About the same5 (9%)0 (0%)5 (16%)23 (12%)9 (9%)14 (15%)
      Very small change, not enough to be an important improvement3 (6%)1 (5%)2 (6%)7 (4%)4 (4%)3 (3%)
      Somewhat worse, but enough to be an important deterioration9 (17%)5 (23%)4 (13%)11 (6%)4 (4%)7 (7%)
      Worse, an important deterioration7 (13%)2 (9%)5 (16%)21 (11%)14 (15%)7 (7%)
       Improvement0.1690.0850.356
      Importantly improved25 (69%)13 (81%)12 (60%)116 (77%)57 (71%)59 (83%)
      Not importantly improved11 (31%)3 (19%)8 (40%)35 (23)23 (29%)12 (17%)
       Do you find your present situation acceptable considering your daily level of function?0.2720.9150.005
      Yes19 (35%)10 (43%)9 (29%)109 (57%)56 (57%)53 (56%)
      No35 (65%)13 (57%)22 (71%)83 (43%)42 (43%)41 (44%)
       The question was only asked to patients replying no to the above: Do you think the treatment has failed?0.6320.8130.790
      Yes21 (64%)7 (58%)14 (67%)53 (66%)27 (67%)26 (65%)
      No12 (36%)5 (42%)7 (33%)27 (34%)13 (33%)14 (35%)
       Would you go through the surgery again?0.2200.1250.263
      Yes19 (35%)6 (26%)13 (42%)91 (47%)51 (52%)40 (42%)
      Maybe21 (39%)12 (52%)9 (29%)67 (35%)35 (35%)32 (34%)
      No14 (26%)5 (22%)9 (29%)36 (18%)13 (13%)23 (24%)
      If no statistical test is mentioned for P values, Chi-squared test was used.
      PROM, Patient-reported outcome measure; IQR, interquartile range; EQ VAS, EuroQol Visual Analogue Scale where 100 = best health imaginable and 0 = worst health imaginable; Copenhagen Knee ROM, Copenhagen Knee Range of Motion.
      a Median (IQR).
      b Wilcoxon Rank-Sum test.
      Supplementary Table 2Patient-Reported Outcomes From Patients Revised for the Indication Unexplained Pain Versus Aseptic Loosening by Surgical Subgroups 1 and 2.
      PROMUnexplained PainAseptic Loosening
      N = 56N = 206
      Surgical Subgroup1. TKA-TKA2. Partial Revision1. TKA-TKA2. Partial Revision
      n = 49n = 7n = 174n = 32
      OKS (median [IQR])26 (13)19 (9)31 (16)24.5 (23)
      EQ-5D-5L Index (median [IQR])0.7 (0.4)0.5 (0.5)0.8 (0.4)0.7 (0.5)
      EQ VAS (median [IQR])62 (37)40 (50)60 (31)40 (39.5)
      FJS (median [IQR])50 (8)46 (13)50 (17)48 (13)
      Copenhagen Knee ROM
       Flexion (median [IQR])5 (2)5 (1)5 (1)5 (2)
       Flexion deficit (0-4)20 (41%)3 (43%)40 (23%)14 (44%)
       Extension (median [IQR])4 (1)3 (1)4 (2)4 (1)
       Extension deficit (0-3)17 (35%)4 (57%)49 (28%)12 (38%)
      PROM, Patient-reported outcome measure; OKS, Oxford Knee Score; IQR, interquartile range; EQ-5D-5L, a value of 1 indicates the best quality of life and 0 indicate the worst; EQ VAS, EuroQol Visual Analogue Scale where 100 = best health imaginable and 0 = worst health imaginable; Copenhagen Knee ROM, Copenhagen Knee Range of Motion.
      Supplementary Table 3Questions on Pain and Satisfaction for Patients Revised for the Indication Unexplained Pain Versus Aseptic Loosening by Surgical Subgroups 1 and 2.
      QuestionsUnexplained Pain n = 56Aseptic Loosening n = 206
      Surgical Subgroup1. TKA-TKA n = 492. Partial Revision n = 71. TKA-TKA n = 1742. Partial Revision n = 32
      Pain
       What was your average pain level the last month on a 0-100 scale; 0 = no pain; 100 = worst pain imaginable (median [IQR])56 (48)70 (20)35 (57)38 (35)
      Satisfaction
       How satisfied are you with the result of the surgery on a 0-100 scale; 0 = very satisfied; 100 = not satisfied (median [IQR])73 (37)80 (32)33 (70)73 (64)
       Improvement. How are your knee problems now compared to prior to the operation?
      Importantly improved22 (71%)3 (69%)104 (81%)12 (55%)
      Not importantly improved9 (29%)2 (40%)25 (19%)10 (45%)
       Do you find your present situation acceptable considering your daily level of function?
      Yes17 (36%)2 (29%)97 (60%)12 (40%)
      No30 (64%)5 (71%)65 (40%)18 (60%)
       The question was only asked to patients replying no to the above: Do you think the treatment has failed?
      Yes18 (62%)3 (75%)36 (58%)17 (94%)
      No11 (38%)1 (25%)26 (42%)1 (6%)
       Would you go through the surgery again?
      Yes18 (38%)1 (14%)77 (47%)14 (47%)
      Maybe17 (36%)4 (57%)57 (35%)10 (33%)
      No12 (26%)2 (29%)30 (18%)6 (20%)

      Appendix A. Supplementary Data

      References

        • DKR
        The Danish knee arthroplasty register, annual report 2021.
        ([accessed 30.09.22])
        • Lewis G.N.
        • Rice D.A.
        • McNair P.J.
        • Kluger M.
        Predictors of persistent pain after total knee arthroplasty: a systematic review and meta-analysis.
        Br J Anaesth. 2015; 114: 551-561
        • Petersen K.K.
        • Simonsen O.
        • Laursen M.B.
        • Nielsen T.A.
        • Rasmussen S.
        • Arendt-Nielsen L.
        Chronic postoperative pain after primary and revision total knee arthroplasty.
        Clin J Pain. 2015; 31: 1-6
        • Vince K.G.
        The problem total knee replacement: systematic, comprehensive and efficient evaluation.
        Bone Joint J. 2014; 96-b: 105-111
        • Reichel F.
        • Innmann M.
        • Gotterbarm T.
        • Schiltenwolf M.
        • Merle C.
        Predictors for persistent pain and dissatisfaction after total knee arthroplasty.
        Schmerz. 2019; 33: 185-190
        • Lindberg-Larsen M.
        • Jørgensen C.C.
        • Bæk Hansen T.
        • Solgaard S.
        • Odgaard A.
        • Kehlet H.
        Re-admissions, re-operations and length of stay in hospital after aseptic revision knee replacement in Denmark: a two-year nationwide study.
        Bone Joint J. 2014; 96-b: 1649-1656
        • Roman M.D.
        • Russu O.
        • Mohor C.
        • Necula R.
        • Boicean A.
        • Todor A.
        • et al.
        Outcomes in revision total knee arthroplasty (Review).
        Exp Ther Med. 2022; 23: 29
        • Lindberg-Larsen M.
        • Petersen P.B.
        • Corap Y.
        • Gromov K.
        • Jørgensen C.C.
        • Kehlet H.
        Fast-track revision knee arthroplasty.
        Knee. 2022; 34: 24-33
        • Arndt K.B.
        • Schrøder H.M.
        • Troelsen A.
        • Lindberg-Larsen M.
        Opioid and analgesic use before and after revision knee arthroplasty for the indications "pain without loosening" versus "aseptic loosening" - a Danish nationwide study.
        J Arthroplasty. 2022; 37: 1618-1625.e3
        • Sabah S.A.
        • Alvand A.
        • Knight R.
        • Beard D.J.
        • Price A.J.
        Patient-reported function and quality of life after revision total knee arthroplasty: an analysis of 10,727 patients from the NHS PROMs program.
        J Arthroplasty. 2021; 36: 2887-2895.e7
        • Baker P.
        • Cowling P.
        • Kurtz S.
        • Jameson S.
        • Gregg P.
        • Deehan D.
        Reason for revision influences early patient outcomes after aseptic knee revision.
        Clin Orthop Relat Res. 2012; 470: 2244-2252
        • Gagnier J.J.
        • Lai J.
        • Mokkink L.B.
        • Terwee C.B.
        COSMIN reporting guideline for studies on measurement properties of patient-reported outcome measures.
        Qual Life Res. 2021; 30: 2197-2218
        • Pedersen A.B.
        • Mehnert F.
        • Odgaard A.
        • Schrøder H.M.
        Existing data sources for clinical epidemiology: the Danish Knee Arthroplasty Register.
        Clin Epidemiol. 2012; 4: 125-135
        • Harris P.A.
        • Taylor R.
        • Thielke R.
        • Payne J.
        • Gonzalez N.
        • Conde J.G.
        Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support.
        J Biomed Inform. 2009; 42: 377-381
        • Harris P.A.
        • Taylor R.
        • Minor B.L.
        • Elliott V.
        • Fernandez M.
        • O'Neal L.
        • et al.
        The REDCap consortium: building an international community of software platform partners.
        J Biomed Inform. 2019; 95: 103208
        • Dawson J.
        • Fitzpatrick R.
        • Murray D.
        • Carr A.
        Questionnaire on the perceptions of patients about total knee replacement.
        J Bone Joint Surg Br. 1998; 80: 63-69
        • Anne Mørup-Petersen M.K.
        • Nielsen R.
        • Paulsen A.
        • Anders O.
        Translation and classical test theory validation of the Danish version of the Oxford Knee Score.
        ([accessed 30.09.22])
        • Murray D.W.
        • Fitzpatrick R.
        • Rogers K.
        • Pandit H.
        • Beard D.J.
        • Carr A.J.
        • et al.
        The use of the Oxford hip and knee scores.
        J Bone Joint Surg Br. 2007; 89: 1010-1014
        • Jensen M.B.
        • Jensen C.E.
        • Gudex C.
        • Pedersen K.M.
        • Sørensen S.S.
        • Ehlers L.H.
        Danish population health measured by the EQ-5D-5L.
        Scand J Public Health. 2021; : 1-9https://doi.org/10.1177/14034948211058060
        • van Hout B.
        • Janssen M.F.
        • Feng Y.S.
        • Kohlmann T.
        • Busschbach J.
        • Golicki D.
        • et al.
        Interim scoring for the EQ-5D-5L: mapping the EQ-5D-5L to EQ-5D-3L value sets.
        Value Health. 2012; 15: 708-715
        • Behrend H.
        • Giesinger K.
        • Giesinger J.M.
        • Kuster M.S.
        The “forgotten joint” as the ultimate goal in joint arthroplasty: validation of a new patient-reported outcome measure.
        J Arthroplasty. 2012; 27: 430-436.e1
        • Thomsen M.G.
        • Latifi R.
        • Kallemose T.
        • Barfod K.W.
        • Husted H.
        • Troelsen A.
        Good validity and reliability of the forgotten joint score in evaluating the outcome of total knee arthroplasty.
        Acta Orthop. 2016; 87: 280-285
        • Mørup-Petersen A.
        • Holm P.M.
        • Holm C.E.
        • Klausen T.W.
        • Skou S.T.
        • Krogsgaard M.R.
        • et al.
        Knee osteoarthritis patients can provide useful estimates of passive knee range of motion: development and validation of the copenhagen knee ROM scale.
        J Arthroplasty. 2018; 33: 2875-2883.e3
        • EuroQol
        EQ-5D user guides.
        ([accessed 16.06.22])
        • Leta T.H.
        • Lygre S.H.
        • Skredderstuen A.
        • Hallan G.
        • Gjertsen J.E.
        • Rokne B.
        • et al.
        Outcomes of unicompartmental knee arthroplasty after aseptic revision to total knee arthroplasty: a comparative study of 768 TKAs and 578 UKAs revised to TKAs from the Norwegian arthroplasty register (1994 to 2011).
        J Bone Joint Surg Am. 2016; 98: 431-440
        • Malviya A.
        • Bettinson K.
        • Kurtz S.M.
        • Deehan D.J.
        When do patient-reported assessments peak after revision knee arthroplasty?.
        Clin Orthop Relat Res. 2012; 470: 1728-1734
        • Hossain F.
        • Patel S.
        • Haddad F.S.
        Midterm assessment of causes and results of revision total knee arthroplasty.
        Clin Orthop Relat Res. 2010; 468: 1221-1228
        • Greidanus N.V.
        • Peterson R.C.
        • Masri B.A.
        • Garbuz D.S.
        Quality of life outcomes in revision versus primary total knee arthroplasty.
        J Arthroplasty. 2011; 26: 615-620
        • Arndt K.B.
        • Schrøder H.M.
        • Troelsen A.
        • Lindberg-Larsen M.
        Prosthesis survival after revision knee arthroplasty for “pain without loosening” versus “aseptic loosening”: a Danish nationwide study.
        Acta Orthop. 2022; 93: 103-110