- Email: [email protected]
The Association Between Anxiety, Depression, and Locus of Control With Patient Outcomes Following Total Knee Arthroplasty
The Journal of Arthroplasty xxx (2019) 1e5
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
The Association Between Anxiety, Depression, and Locus of Control With Patient Outcomes Following Total Knee Arthroplasty Joshua Xu, BSc (Adv) a, b, *, Joshua Twiggs, PhD c, David Parker, MBBS b, Jonathan Negus, PhD, MBBS a, b a b c
Sydney Medical School, University of Sydney, Sydney, New South Wales, Australia Sydney Orthopaedic Research Institute, Chatswood, New South Wales, Australia 360 Knee Systems Pty Ltd, Pymble, New South Wales, Australia
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 August 2019 Received in revised form 29 September 2019 Accepted 13 October 2019 Available online xxx
Background: We aimed to determine how preoperative anxiety, depression, and locus of control (LoC) might predict patient outcomes following total knee arthroplasty (TKA). Methods: Patients undergoing TKA were prospectively recruited over an 18-month period. The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) was used to assess TKA outcomes. The Short Form-12, Hospital Anxiety and Depression Score, and LoC surveys were completed by the patients to assess their psychosocial state. These scores were collected preoperatively and at 6 weeks, 18 weeks, and 1 year postoperation. Results: The final cohort consisted of 136 patients. Greater preoperative depression (P ¼ .004) and anxiety (P ¼ .001) scores were correlated with worse total WOMAC score at 6 weeks and 18 weeks postoperatively, respectively. A poorer preoperative Short Form-12 mental score was also correlated with a worse total WOMAC score at 6 weeks postoperatively (P ¼ .007). Greater tendency toward an internal LoC preoperatively was correlated with better WOMAC pain (P < .001) and function (P ¼ .003) scores at 18 weeks postoperatively. However, there was no correlation between preoperative external LoC and postoperative WOMAC score. There was also no correlation between any of the preoperative psychosocial measures and WOMAC score at 1 year postoperatively. Conclusion: We identified a group of patients whose psychosocial markers predicted them to have worse outcomes in the short to medium term even though they normalized to satisfactory outcomes at 1 year postoperatively. Identifying this group could allow for targeted intervention with an adjustment of expectations and thus more effective recovery. © 2019 Elsevier Inc. All rights reserved.
Keywords: anxiety depression locus knee arthroplasty knee replacement predictors
Total knee arthroplasty (TKA) is an effective method for the treatment of severe osteoarthritis of the knee. Although the resolution of pain and improvement in knee function occurs in most cases, 11%-18% of patients undergoing TKA may be dissatisfied with their outcomes [1]. These patients seem to be dissatisfied despite adequate radiographic outcomes and objective improvements in range of motion. A plethora of factors have been linked with these
One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to https://doi.org/10.1016/j.arth.2019.10.022. * Reprint requests: Joshua Xu, BSc (Adv), The Gallery, Level 1/445, Victoria Avenue, Chatswood, NSW 2067, Australia. https://doi.org/10.1016/j.arth.2019.10.022 0883-5403/© 2019 Elsevier Inc. All rights reserved.
dissatisfied patients including low socioeconomic status, earlystage osteoarthritis, or significant comorbidities [2,3]. However, for some patients the dissatisfaction cannot be attributed to any medical or technical reasons [4]. Psychosocial characteristics may play a role in determining the outcome of TKA from the patient’s perspective. There have been studies indicating that psychological distress preoperatively can have detrimental effects on the outcome of TKA [5,6]. Depression and anxiety have been linked to exaggerated responses to pain, which may contribute to patient dissatisfaction [7]. Patients with an internal locus of control (LoC) are those that believe they are in control of their outcome and thus are more likely to achieve rehabilitation goals [8,9]. An external LoC is when a patient attributes their outcome to external factors such as doctors, other healthcare professionals, or chance, and is thought to have poorer
2
J. Xu et al. / The Journal of Arthroplasty xxx (2019) 1e5
outcomes [9]. These psychological factors may play a role in explaining a portion of unsuccessful TKA outcomes, which cannot be explained by known factors. With greater focus on patient-reported outcomes, psychosocial measures are being increasingly incorporated into clinical practice. Understanding the association between these psychosocial measures and TKA outcomes is an important step to eventually being able to predict patients who will have a poor outcome. The ability to identify psychosocial predictors of good or bad outcomes preoperatively would be important in guiding surgical decisions so that patient outcomes for TKA are optimized. In this study, we aimed to determine how preoperative anxiety, depression, and LoC might predict patient outcomes following TKA. We hypothesize that greater preoperative anxiety, depression, and an external LoC would be predictive of a poorer outcome following TKA.
of these scores. A higher SF-12 mental score reflects a better level of mental health. The Hospital Anxiety and Depression Scale (HADS) was used to assess anxiety and depression. The HADS is a 12-item questionnaire with each item scored on a 4-point Likert scale ranging from 0 to 3. Half the questions assessed anxiety and the other half depression, producing 2 scores. A higher anxiety or depression score reflects worse anxiety or depression respectively. A modified LoC questionnaire was used to determine if a patient had an internal or external LoC [11]. The LoC questionnaire contains 18 questions scored from 1 to 6, with each question assessing the magnitude of either internal or external LoC. For each patient this produced 2 scores, an internal and external score, which were both standardized to 100. The net difference of the standardized internal and external LoC scores was used to determine the predominating LoC (external or internal).
Method Study Participants
Statistical Analysis
Patients requiring TKA due to end-stage knee osteoarthritis were prospectively recruited from a single orthopedic center in Sydney, Australia between March 2011 and September 2012. The decision to undergo TKA was determined after potential patients were screened by the operating orthopedic surgeon and found to be appropriate for TKA, and subsequently elected to proceed with surgery. The inclusion criteria for this analysis were patients over 18 years old, primary unilateral or bilateral TKA, and English speakers. Patients were excluded if they had previous septic joint, revision surgery, dementia, or were unable to return for all extra follow-up visits. Ethics approval was obtained prior to patient recruitment from the Northern Sydney Central Coast Health Human Research Ethics Committee (RESP/17/316/HREC/10/HAWKE/123).
Statistical analysis was performed of trends over time in each of the primary and psychosocial outcome scores, with repeated measures analysis of variance tests used to determine differences between group scores. WOMAC scores were normally distributed and correlations between psychosocial predictors and primary WOMAC outcomes were also assessed and checked for statistically significant difference from no correlation. Patients were further subgrouped by their LoC classification (internal or external) and investigated with t-tests for trends in WOMAC outcomes. In all analyses, statistical significance was set to P ¼ .05. All analyses were performed using R v3.4.2.
Assessment of Outcomes
Patient Demographics
Patient assessments were conducted preoperatively when the participants were first recruited into the study. The initial assessment was repeated closer to the operation if the recruitment date was more than 3 months before the operation. The questionnaires were administered to the patients via paper copy. Demographic information such as age, height, and weight was also collected preoperatively. Postoperatively, the assessments were conducted at 6 weeks, 18 weeks, and 1 year.
The final cohort consisted of 136 (63 male, 73 female) patients with a mean age of 71.5 years (standard deviation [SD] ¼ 7.6) and a mean BMI of 30.3 kg/m2 (SD ¼ 4.7).
Primary Outcome The primary TKA outcome was measured by the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) [10]. This questionnaire contains 24 items and provides a total score that is composed of pain (5 items), stiffness (2 items), and function (17 items) subscales. Each question is scored on a 5-point Likert scale ranging from 0 to 4, with 0 being no pain, stiffness, or difficulty and 4 being maximum pain, stiffness, or difficulty. For our analysis, we standardized the scores to 100 with a higher score reflecting a better outcome. Psychosocial Outcomes The Short Form (SF)-12 mental score is a health-related quality of life questionnaire that was used as a measure of overall mental health. The questionnaire consists of 12 items, each scored from 1 to 5, with the overall SF-12 mental score determined by an analysis
Results
WOMAC Outcomes The total WOMAC continued to significantly improve at all subsequent follow-up points postoperatively (P < .001). The greatest progression in total WOMAC was immediate postoperatively, improving from 63.8 (SD ¼ 15.4) preoperatively to 75.5 (SD ¼ 16.6) at 6 weeks and 87.2 (SD ¼ 12.1) at 18 weeks postoperation. From 18 weeks to 1 year, the improvement was more gradual, with a mean total WOMAC of 91.9 (SD ¼ 9.8) at 1 year postoperation (Fig. 1). All patients eventually improved from their baseline preoperative WOMAC score following TKA. Patients with a high preoperative anxiety or depression score (>7) had a significantly worse preoperative total WOMAC score than those with a low preoperative anxiety (P < .001) or depression (P ¼ .006) score (7) respectively. However, there was an improvement in total WOMAC for all these groups postoperatively. There was no significant difference in total WOMAC score between the high and low anxiety and depression groups at any postoperative interval. Also, when comparing the patients that had an internal LoC and external LoC preoperatively, there was no significant difference in total WOMAC score at any preoperative or postoperative interval. These results are summarized in Table 1.
J. Xu et al. / The Journal of Arthroplasty xxx (2019) 1e5
3
For patients with a preoperative internal LoC, those that maintained an internal LoC through follow-up had a better total WOMAC at 1 year than those that were initially internal but developed an external LoC during follow-up (P ¼.04). However, for patients with an external LoC initially, there was no difference in WOMAC outcomes for those that became internal vs those that remained external. Discussion
Fig. 1. WOMAC score for all patients preoperatively and at 6 wk, 18 wk, and 1 y post operation. TKR, total knee replacement; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Index.
Psychosocial Outcomes Depression (P < .001) and anxiety (P < .001) scores continued to improve at all subsequent follow-up points after surgery (Fig. 2). However, the SF-12 mental score did not change following the procedure. Preoperatively, the majority of the cohort had an internal LoC (90%). From these initially internal LoC patients, the majority of them remained with an internal LoC through follow-up. However, for patients with a preoperative external LoC, 83.3% of these patients shifted to a predominately internal LoC by 1 year postoperation (Fig. 3). The magnitude of the internal LoC scores varied significantly during follow-up (P < .001). However, the magnitude of the external LoC remained unchanged over the follow-up period. Psychosocial Predictors A greater preoperative depression (P ¼ .004) and anxiety (P ¼ .001) score was correlated with a worse total WOMAC score at 6 weeks and 18 weeks postoperation. A poorer preoperative SF-12 mental score also correlated with a worse total WOMAC score at 6 weeks postoperation (P ¼ .007). A greater tendency toward an internal LoC score preoperatively was correlated with a better WOMAC pain (P < .001) and function (P ¼ .003) score at 18 weeks postoperation. A greater tendency toward an external LoC score preoperatively was not correlated with a worse postoperative WOMAC score.
We found that anxiety and depression prior to TKA was correlated with a worse outcome in the short to medium term which normalized to satisfactory outcomes at 1 year. A greater tendency toward an internal LoC was also predictive of a better outcome in the medium term although there were no correlations between external LoC and WOMAC outcomes. For patients with an internal LoC preoperatively, a shift toward an external LoC during follow-up may also be linked with a worse outcome at 1 year postoperation. Being able to preoperatively identify psychosocial factors that may predict a worse TKA outcome would be important in optimizing patient selection for TKA. Along with that it may assist in recognizing patients that would benefit from targeted interventions during the early postoperative period. Untreated depression and anxiety have been widely shown to cause poorer surgical outcomes [1,7,9,12]. Preoperative anxiety and depression are associated with an increased sensitivity to pain, a major factor that leads to patient dissatisfaction following surgery [7]. Untreated depression has also independently been linked with chronic pain and reduced function after TKA [12,13]. Furthermore, those with depression may not be as active and thus not participate in rehabilitation as actively [9]. This may explain the worse short and medium outcomes with preoperative depression. Other studies have also found that anxiety is only associated with poorer outcomes in the early postoperative period but not the ultimate outcome [12]. Although we found that anxiety was no longer detrimental to outcomes at 1 year postoperation, some studies have reported that it may take up to 5 years for the association to no longer be significant [12]. Thus, reducing anxiety and depression preoperatively may help improve short-term outcomes. This can potentially be done via preoperative screening tool such as the HADS. Those with a high anxiety and depression score should be referred to a psychiatrist or other mental health specialist to be treated before undergoing TKA. An internal LoC indicates that a patient believes they themselves are more in control of their outcomes, while an external LoC indicates that the patient tends to rely on, and praise or blame, external factors such as doctors, other people, or chance. From our study, an increased internal LoC was linked to better outcomes at 18 months postoperation. This has been shown in other studies, with higher internal LoC associated with more rapid achievement of a straight leg raise [8]. Greater dissatisfaction following TKA has also
Table 1 Follow-Up of Mean Total WOMAC Scores for Patients With Preoperative Internal and External LoC Along With High (>7) and Low (7) Preoperative Anxiety and Depression Scores. Mean Total WOMAC (SD)
High anxiety score Low anxiety score High depression score Low depression score Internal LoC External LoC
Preop
6 wk
18 wk
1y
48.4 64.3 50.8 63.6 61.7 62.7
69.6 77.0 71.4 76.5 76.9 71.2
83.1 87.8 84.2 87.5 87.2 87.9
88.0 92.4 88.1 92.2 92.0 91.9
(11.8) (13.1) (9.8) (14.2) (13.6) (15.5)
(15.2) (14.0) (11.7) (14.5) (14.1) (14.7)
LoC, locus of control; SD, standard deviation; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Index.
(11.1) (12.1) (11.1) (12.2) (12.1) (11.7)
(13.8) (9.1) (13.0) (9.5) (9.5) (20.3)
4
J. Xu et al. / The Journal of Arthroplasty xxx (2019) 1e5
Fig. 2. Depression and anxiety score for all patients preoperatively and at 6 wk, 18 wk, and 1 y post operation based on the Hospital Anxiety and Depression Scale questionnaire.
been found in patients with a low internal LoC [7,9]. The better outcomes with an internal LoC are likely due to higher motivation and more self-directed activities in these people. Internal LoC patients have better compliance with rehabilitation and are more independent with managing the symptoms of their knee [8,14]. Greater self-control and positive expectations have also been shown to produce significant functional improvements [8]. Interestingly, we found that an increased external LoC was not linked with a worse outcome. This may be explained by the fact that all patients in this study underwent their TKA through the private healthcare system. These patients inherently have greater access to personalized care, rehabilitation, and home help. As a result, greater reliance on external factors in this population may not have a detrimental impact on the outcome. Additionally, we found that the internal LoC score varied, but the external LoC remained unchanged through the postoperative period. Thus, internal LoC may be a good psychosocial trait to monitor and target with intervention. It is important to address the limitations of this study. First, all patients recruited were managed in the private healthcare system and were mostly from a higher socioeconomic background. As a result, the patients included in this study may not be representative of the whole population. Further studies in the public hospital system would be required to determine if these findings are consistent across different populations and socioeconomic groups. With the inclusion of only 136 patients, the correlations and potential psychosocial predictors, while statistically significant, are based off a relatively small group of patients. Larger multicenter studies with increased patient numbers are required to further
substantiate the correlations identified in our study. There may also be other confounding variables we have not taken into account for this study, including surgeon, type of implant, and surgical technique. Other psychosocial factors such as patient expectations, social support, and personality were also not specifically assessed and thus may have had an effect on the outcome. Furthermore, there are now more reimbursement systems in the United States determining payment based on patient-reported outcome measures (PROMs). By identifying factors that may predict poorer patientreported outcome measures, there may be an effect on patient selection, hindering certain patients from receiving a TKA that may still be beneficial despite poorer psychosocial health.
Conclusions This study identified a group of patients whose psychosocial markers predicted worse outcomes in the short to medium term, even though they normalized to satisfactory outcomes at 1 year postoperation. However, 18 weeks of pain postoperatively is a significant amount of time for patients to experience, with potential ongoing negative implications. Thus, early recognition of patients whose results are more likely to degrade over time could allow for targeted preoperative psychological or pain management intervention, including an adjustment of expectations, and more effective rehabilitation postoperatively. Moving forward, larger multicenter studies are required to elucidate stronger correlations and develop clinically applicable preoperative prediction models.
References
Percentage. of patients
100% 80% Internal LoC remaining internal
60%
External LoC remaining external
40% 20% 0% Pre-operative
6 week
18 weeks
1 year
Follow up period
Fig. 3. Trends in the percentage of initially internal LoC patients (n ¼ 111) maintaining an internal LoC and the initially external LoC patients (n ¼ 25) maintaining an external LoC. LoC, locus of control.
[1] Fritsch BA, Dhurve K, Scholes C, El-Tawil S, Molloy A, Parker DA, et al. Multifactorial analysis of patient dissatisfaction after TKRdthe influence of psychological factors. Orthop J Sports Med 2017;5(5 suppl5). 2325967117S00168. [2] Schnurr C, Jarrous M, Gudden I, Eysel P, Konig DP. Pre-operative arthritis severity as a predictor for total knee arthroplasty patients’ satisfaction. Int Orthop 2013;37:1257e61. [3] Scott CE, Howie CR, MacDonald D, Biant LC. Predicting dissatisfaction following total knee replacement: a prospective study of 1217 patients. J Bone Joint Surg Br 2010;92:1253e8. [4] Wylde V, Dieppe P, Hewlett S, Learmonth ID. Total knee replacement: is it really an effective procedure for all? Knee 2007;14:417e23. [5] Blackburn J, Qureshi A, Amirfeyz R, Bannister G. Does preoperative anxiety and depression predict satisfaction after total knee replacement? Knee 2012;19:522e4. rez-Prieto D, Gil-Gonz L, [6] Pe alez S, Pelfort X, Leal-Blanquet J, Puig-Verdie Hinarejos P. Influence of depression on total knee arthroplasty outcomes. J Arthroplasty 2014;29:44e7.
J. Xu et al. / The Journal of Arthroplasty xxx (2019) 1e5 [7] Dhurve K, Scholes C, El-Tawil S, Shaikh A, Weng LK, Levin K, et al. Multifactorial analysis of dissatisfaction after primary total knee replacement. Knee 2017;24:856e62. [8] Kendell K, Saxby B, Farrow M, Naisby C. Psychological factors associated with shortterm recovery from total knee replacement. Br J Health Psychol 2001;6:41e52. [9] Lopez-Olivo MA, Landon GC, Siff SJ, Edelstein D, Pak C, Kallen MA, et al. Psychosocial determinants of outcomes in knee replacement. Ann Rheum Dis 2011;70:1775e81. [10] McConnell S, Kolopack P, Davis AM. The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC): a review of its utility and measurement properties. Arthritis Care Res 2001;45:453e61.
5
[11] Rotter JB. Generalized expectancies for internal versus external control of reinforcement. Psychol Monogr 1966;80:1e28. [12] Brander V, Gondek S, Martin E, Stulberg SD. Pain and depression influence outcome 5 years after knee replacement surgery. Clin Orthop Relat Res 2007;464:21e6. [13] Fisher DA, Dierckman B, Watts MR, Davis K. Looks good but feels bad: factors that contribute to poor results after total knee arthroplasty. J Arthroplasty 2007;22(6 Suppl 2):39e42. [14] Wallston KA, Strudler Wallston B, DeVellis R. Development of the Multidimensional Health Locus of Control (MHLC) scales. Health Educ Monogr 1978;6: 160e70.
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
The Association Between Anxiety, Depression, and Locus of Control With Patient Outcomes Following Total Knee Arthroplasty Joshua Xu, BSc (Adv) a, b, *, Joshua Twiggs, PhD c, David Parker, MBBS b, Jonathan Negus, PhD, MBBS a, b a b c
Sydney Medical School, University of Sydney, Sydney, New South Wales, Australia Sydney Orthopaedic Research Institute, Chatswood, New South Wales, Australia 360 Knee Systems Pty Ltd, Pymble, New South Wales, Australia
a r t i c l e i n f o
a b s t r a c t
Article history: Received 15 August 2019 Received in revised form 29 September 2019 Accepted 13 October 2019 Available online xxx
Background: We aimed to determine how preoperative anxiety, depression, and locus of control (LoC) might predict patient outcomes following total knee arthroplasty (TKA). Methods: Patients undergoing TKA were prospectively recruited over an 18-month period. The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) was used to assess TKA outcomes. The Short Form-12, Hospital Anxiety and Depression Score, and LoC surveys were completed by the patients to assess their psychosocial state. These scores were collected preoperatively and at 6 weeks, 18 weeks, and 1 year postoperation. Results: The final cohort consisted of 136 patients. Greater preoperative depression (P ¼ .004) and anxiety (P ¼ .001) scores were correlated with worse total WOMAC score at 6 weeks and 18 weeks postoperatively, respectively. A poorer preoperative Short Form-12 mental score was also correlated with a worse total WOMAC score at 6 weeks postoperatively (P ¼ .007). Greater tendency toward an internal LoC preoperatively was correlated with better WOMAC pain (P < .001) and function (P ¼ .003) scores at 18 weeks postoperatively. However, there was no correlation between preoperative external LoC and postoperative WOMAC score. There was also no correlation between any of the preoperative psychosocial measures and WOMAC score at 1 year postoperatively. Conclusion: We identified a group of patients whose psychosocial markers predicted them to have worse outcomes in the short to medium term even though they normalized to satisfactory outcomes at 1 year postoperatively. Identifying this group could allow for targeted intervention with an adjustment of expectations and thus more effective recovery. © 2019 Elsevier Inc. All rights reserved.
Keywords: anxiety depression locus knee arthroplasty knee replacement predictors
Total knee arthroplasty (TKA) is an effective method for the treatment of severe osteoarthritis of the knee. Although the resolution of pain and improvement in knee function occurs in most cases, 11%-18% of patients undergoing TKA may be dissatisfied with their outcomes [1]. These patients seem to be dissatisfied despite adequate radiographic outcomes and objective improvements in range of motion. A plethora of factors have been linked with these
One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to https://doi.org/10.1016/j.arth.2019.10.022. * Reprint requests: Joshua Xu, BSc (Adv), The Gallery, Level 1/445, Victoria Avenue, Chatswood, NSW 2067, Australia. https://doi.org/10.1016/j.arth.2019.10.022 0883-5403/© 2019 Elsevier Inc. All rights reserved.
dissatisfied patients including low socioeconomic status, earlystage osteoarthritis, or significant comorbidities [2,3]. However, for some patients the dissatisfaction cannot be attributed to any medical or technical reasons [4]. Psychosocial characteristics may play a role in determining the outcome of TKA from the patient’s perspective. There have been studies indicating that psychological distress preoperatively can have detrimental effects on the outcome of TKA [5,6]. Depression and anxiety have been linked to exaggerated responses to pain, which may contribute to patient dissatisfaction [7]. Patients with an internal locus of control (LoC) are those that believe they are in control of their outcome and thus are more likely to achieve rehabilitation goals [8,9]. An external LoC is when a patient attributes their outcome to external factors such as doctors, other healthcare professionals, or chance, and is thought to have poorer
2
J. Xu et al. / The Journal of Arthroplasty xxx (2019) 1e5
outcomes [9]. These psychological factors may play a role in explaining a portion of unsuccessful TKA outcomes, which cannot be explained by known factors. With greater focus on patient-reported outcomes, psychosocial measures are being increasingly incorporated into clinical practice. Understanding the association between these psychosocial measures and TKA outcomes is an important step to eventually being able to predict patients who will have a poor outcome. The ability to identify psychosocial predictors of good or bad outcomes preoperatively would be important in guiding surgical decisions so that patient outcomes for TKA are optimized. In this study, we aimed to determine how preoperative anxiety, depression, and LoC might predict patient outcomes following TKA. We hypothesize that greater preoperative anxiety, depression, and an external LoC would be predictive of a poorer outcome following TKA.
of these scores. A higher SF-12 mental score reflects a better level of mental health. The Hospital Anxiety and Depression Scale (HADS) was used to assess anxiety and depression. The HADS is a 12-item questionnaire with each item scored on a 4-point Likert scale ranging from 0 to 3. Half the questions assessed anxiety and the other half depression, producing 2 scores. A higher anxiety or depression score reflects worse anxiety or depression respectively. A modified LoC questionnaire was used to determine if a patient had an internal or external LoC [11]. The LoC questionnaire contains 18 questions scored from 1 to 6, with each question assessing the magnitude of either internal or external LoC. For each patient this produced 2 scores, an internal and external score, which were both standardized to 100. The net difference of the standardized internal and external LoC scores was used to determine the predominating LoC (external or internal).
Method Study Participants
Statistical Analysis
Patients requiring TKA due to end-stage knee osteoarthritis were prospectively recruited from a single orthopedic center in Sydney, Australia between March 2011 and September 2012. The decision to undergo TKA was determined after potential patients were screened by the operating orthopedic surgeon and found to be appropriate for TKA, and subsequently elected to proceed with surgery. The inclusion criteria for this analysis were patients over 18 years old, primary unilateral or bilateral TKA, and English speakers. Patients were excluded if they had previous septic joint, revision surgery, dementia, or were unable to return for all extra follow-up visits. Ethics approval was obtained prior to patient recruitment from the Northern Sydney Central Coast Health Human Research Ethics Committee (RESP/17/316/HREC/10/HAWKE/123).
Statistical analysis was performed of trends over time in each of the primary and psychosocial outcome scores, with repeated measures analysis of variance tests used to determine differences between group scores. WOMAC scores were normally distributed and correlations between psychosocial predictors and primary WOMAC outcomes were also assessed and checked for statistically significant difference from no correlation. Patients were further subgrouped by their LoC classification (internal or external) and investigated with t-tests for trends in WOMAC outcomes. In all analyses, statistical significance was set to P ¼ .05. All analyses were performed using R v3.4.2.
Assessment of Outcomes
Patient Demographics
Patient assessments were conducted preoperatively when the participants were first recruited into the study. The initial assessment was repeated closer to the operation if the recruitment date was more than 3 months before the operation. The questionnaires were administered to the patients via paper copy. Demographic information such as age, height, and weight was also collected preoperatively. Postoperatively, the assessments were conducted at 6 weeks, 18 weeks, and 1 year.
The final cohort consisted of 136 (63 male, 73 female) patients with a mean age of 71.5 years (standard deviation [SD] ¼ 7.6) and a mean BMI of 30.3 kg/m2 (SD ¼ 4.7).
Primary Outcome The primary TKA outcome was measured by the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) [10]. This questionnaire contains 24 items and provides a total score that is composed of pain (5 items), stiffness (2 items), and function (17 items) subscales. Each question is scored on a 5-point Likert scale ranging from 0 to 4, with 0 being no pain, stiffness, or difficulty and 4 being maximum pain, stiffness, or difficulty. For our analysis, we standardized the scores to 100 with a higher score reflecting a better outcome. Psychosocial Outcomes The Short Form (SF)-12 mental score is a health-related quality of life questionnaire that was used as a measure of overall mental health. The questionnaire consists of 12 items, each scored from 1 to 5, with the overall SF-12 mental score determined by an analysis
Results
WOMAC Outcomes The total WOMAC continued to significantly improve at all subsequent follow-up points postoperatively (P < .001). The greatest progression in total WOMAC was immediate postoperatively, improving from 63.8 (SD ¼ 15.4) preoperatively to 75.5 (SD ¼ 16.6) at 6 weeks and 87.2 (SD ¼ 12.1) at 18 weeks postoperation. From 18 weeks to 1 year, the improvement was more gradual, with a mean total WOMAC of 91.9 (SD ¼ 9.8) at 1 year postoperation (Fig. 1). All patients eventually improved from their baseline preoperative WOMAC score following TKA. Patients with a high preoperative anxiety or depression score (>7) had a significantly worse preoperative total WOMAC score than those with a low preoperative anxiety (P < .001) or depression (P ¼ .006) score (7) respectively. However, there was an improvement in total WOMAC for all these groups postoperatively. There was no significant difference in total WOMAC score between the high and low anxiety and depression groups at any postoperative interval. Also, when comparing the patients that had an internal LoC and external LoC preoperatively, there was no significant difference in total WOMAC score at any preoperative or postoperative interval. These results are summarized in Table 1.
J. Xu et al. / The Journal of Arthroplasty xxx (2019) 1e5
3
For patients with a preoperative internal LoC, those that maintained an internal LoC through follow-up had a better total WOMAC at 1 year than those that were initially internal but developed an external LoC during follow-up (P ¼.04). However, for patients with an external LoC initially, there was no difference in WOMAC outcomes for those that became internal vs those that remained external. Discussion
Fig. 1. WOMAC score for all patients preoperatively and at 6 wk, 18 wk, and 1 y post operation. TKR, total knee replacement; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Index.
Psychosocial Outcomes Depression (P < .001) and anxiety (P < .001) scores continued to improve at all subsequent follow-up points after surgery (Fig. 2). However, the SF-12 mental score did not change following the procedure. Preoperatively, the majority of the cohort had an internal LoC (90%). From these initially internal LoC patients, the majority of them remained with an internal LoC through follow-up. However, for patients with a preoperative external LoC, 83.3% of these patients shifted to a predominately internal LoC by 1 year postoperation (Fig. 3). The magnitude of the internal LoC scores varied significantly during follow-up (P < .001). However, the magnitude of the external LoC remained unchanged over the follow-up period. Psychosocial Predictors A greater preoperative depression (P ¼ .004) and anxiety (P ¼ .001) score was correlated with a worse total WOMAC score at 6 weeks and 18 weeks postoperation. A poorer preoperative SF-12 mental score also correlated with a worse total WOMAC score at 6 weeks postoperation (P ¼ .007). A greater tendency toward an internal LoC score preoperatively was correlated with a better WOMAC pain (P < .001) and function (P ¼ .003) score at 18 weeks postoperation. A greater tendency toward an external LoC score preoperatively was not correlated with a worse postoperative WOMAC score.
We found that anxiety and depression prior to TKA was correlated with a worse outcome in the short to medium term which normalized to satisfactory outcomes at 1 year. A greater tendency toward an internal LoC was also predictive of a better outcome in the medium term although there were no correlations between external LoC and WOMAC outcomes. For patients with an internal LoC preoperatively, a shift toward an external LoC during follow-up may also be linked with a worse outcome at 1 year postoperation. Being able to preoperatively identify psychosocial factors that may predict a worse TKA outcome would be important in optimizing patient selection for TKA. Along with that it may assist in recognizing patients that would benefit from targeted interventions during the early postoperative period. Untreated depression and anxiety have been widely shown to cause poorer surgical outcomes [1,7,9,12]. Preoperative anxiety and depression are associated with an increased sensitivity to pain, a major factor that leads to patient dissatisfaction following surgery [7]. Untreated depression has also independently been linked with chronic pain and reduced function after TKA [12,13]. Furthermore, those with depression may not be as active and thus not participate in rehabilitation as actively [9]. This may explain the worse short and medium outcomes with preoperative depression. Other studies have also found that anxiety is only associated with poorer outcomes in the early postoperative period but not the ultimate outcome [12]. Although we found that anxiety was no longer detrimental to outcomes at 1 year postoperation, some studies have reported that it may take up to 5 years for the association to no longer be significant [12]. Thus, reducing anxiety and depression preoperatively may help improve short-term outcomes. This can potentially be done via preoperative screening tool such as the HADS. Those with a high anxiety and depression score should be referred to a psychiatrist or other mental health specialist to be treated before undergoing TKA. An internal LoC indicates that a patient believes they themselves are more in control of their outcomes, while an external LoC indicates that the patient tends to rely on, and praise or blame, external factors such as doctors, other people, or chance. From our study, an increased internal LoC was linked to better outcomes at 18 months postoperation. This has been shown in other studies, with higher internal LoC associated with more rapid achievement of a straight leg raise [8]. Greater dissatisfaction following TKA has also
Table 1 Follow-Up of Mean Total WOMAC Scores for Patients With Preoperative Internal and External LoC Along With High (>7) and Low (7) Preoperative Anxiety and Depression Scores. Mean Total WOMAC (SD)
High anxiety score Low anxiety score High depression score Low depression score Internal LoC External LoC
Preop
6 wk
18 wk
1y
48.4 64.3 50.8 63.6 61.7 62.7
69.6 77.0 71.4 76.5 76.9 71.2
83.1 87.8 84.2 87.5 87.2 87.9
88.0 92.4 88.1 92.2 92.0 91.9
(11.8) (13.1) (9.8) (14.2) (13.6) (15.5)
(15.2) (14.0) (11.7) (14.5) (14.1) (14.7)
LoC, locus of control; SD, standard deviation; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Index.
(11.1) (12.1) (11.1) (12.2) (12.1) (11.7)
(13.8) (9.1) (13.0) (9.5) (9.5) (20.3)
4
J. Xu et al. / The Journal of Arthroplasty xxx (2019) 1e5
Fig. 2. Depression and anxiety score for all patients preoperatively and at 6 wk, 18 wk, and 1 y post operation based on the Hospital Anxiety and Depression Scale questionnaire.
been found in patients with a low internal LoC [7,9]. The better outcomes with an internal LoC are likely due to higher motivation and more self-directed activities in these people. Internal LoC patients have better compliance with rehabilitation and are more independent with managing the symptoms of their knee [8,14]. Greater self-control and positive expectations have also been shown to produce significant functional improvements [8]. Interestingly, we found that an increased external LoC was not linked with a worse outcome. This may be explained by the fact that all patients in this study underwent their TKA through the private healthcare system. These patients inherently have greater access to personalized care, rehabilitation, and home help. As a result, greater reliance on external factors in this population may not have a detrimental impact on the outcome. Additionally, we found that the internal LoC score varied, but the external LoC remained unchanged through the postoperative period. Thus, internal LoC may be a good psychosocial trait to monitor and target with intervention. It is important to address the limitations of this study. First, all patients recruited were managed in the private healthcare system and were mostly from a higher socioeconomic background. As a result, the patients included in this study may not be representative of the whole population. Further studies in the public hospital system would be required to determine if these findings are consistent across different populations and socioeconomic groups. With the inclusion of only 136 patients, the correlations and potential psychosocial predictors, while statistically significant, are based off a relatively small group of patients. Larger multicenter studies with increased patient numbers are required to further
substantiate the correlations identified in our study. There may also be other confounding variables we have not taken into account for this study, including surgeon, type of implant, and surgical technique. Other psychosocial factors such as patient expectations, social support, and personality were also not specifically assessed and thus may have had an effect on the outcome. Furthermore, there are now more reimbursement systems in the United States determining payment based on patient-reported outcome measures (PROMs). By identifying factors that may predict poorer patientreported outcome measures, there may be an effect on patient selection, hindering certain patients from receiving a TKA that may still be beneficial despite poorer psychosocial health.
Conclusions This study identified a group of patients whose psychosocial markers predicted worse outcomes in the short to medium term, even though they normalized to satisfactory outcomes at 1 year postoperation. However, 18 weeks of pain postoperatively is a significant amount of time for patients to experience, with potential ongoing negative implications. Thus, early recognition of patients whose results are more likely to degrade over time could allow for targeted preoperative psychological or pain management intervention, including an adjustment of expectations, and more effective rehabilitation postoperatively. Moving forward, larger multicenter studies are required to elucidate stronger correlations and develop clinically applicable preoperative prediction models.
References
Percentage. of patients
100% 80% Internal LoC remaining internal
60%
External LoC remaining external
40% 20% 0% Pre-operative
6 week
18 weeks
1 year
Follow up period
Fig. 3. Trends in the percentage of initially internal LoC patients (n ¼ 111) maintaining an internal LoC and the initially external LoC patients (n ¼ 25) maintaining an external LoC. LoC, locus of control.
[1] Fritsch BA, Dhurve K, Scholes C, El-Tawil S, Molloy A, Parker DA, et al. Multifactorial analysis of patient dissatisfaction after TKRdthe influence of psychological factors. Orthop J Sports Med 2017;5(5 suppl5). 2325967117S00168. [2] Schnurr C, Jarrous M, Gudden I, Eysel P, Konig DP. Pre-operative arthritis severity as a predictor for total knee arthroplasty patients’ satisfaction. Int Orthop 2013;37:1257e61. [3] Scott CE, Howie CR, MacDonald D, Biant LC. Predicting dissatisfaction following total knee replacement: a prospective study of 1217 patients. J Bone Joint Surg Br 2010;92:1253e8. [4] Wylde V, Dieppe P, Hewlett S, Learmonth ID. Total knee replacement: is it really an effective procedure for all? Knee 2007;14:417e23. [5] Blackburn J, Qureshi A, Amirfeyz R, Bannister G. Does preoperative anxiety and depression predict satisfaction after total knee replacement? Knee 2012;19:522e4. rez-Prieto D, Gil-Gonz L, [6] Pe alez S, Pelfort X, Leal-Blanquet J, Puig-Verdie Hinarejos P. Influence of depression on total knee arthroplasty outcomes. J Arthroplasty 2014;29:44e7.
J. Xu et al. / The Journal of Arthroplasty xxx (2019) 1e5 [7] Dhurve K, Scholes C, El-Tawil S, Shaikh A, Weng LK, Levin K, et al. Multifactorial analysis of dissatisfaction after primary total knee replacement. Knee 2017;24:856e62. [8] Kendell K, Saxby B, Farrow M, Naisby C. Psychological factors associated with shortterm recovery from total knee replacement. Br J Health Psychol 2001;6:41e52. [9] Lopez-Olivo MA, Landon GC, Siff SJ, Edelstein D, Pak C, Kallen MA, et al. Psychosocial determinants of outcomes in knee replacement. Ann Rheum Dis 2011;70:1775e81. [10] McConnell S, Kolopack P, Davis AM. The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC): a review of its utility and measurement properties. Arthritis Care Res 2001;45:453e61.
5
[11] Rotter JB. Generalized expectancies for internal versus external control of reinforcement. Psychol Monogr 1966;80:1e28. [12] Brander V, Gondek S, Martin E, Stulberg SD. Pain and depression influence outcome 5 years after knee replacement surgery. Clin Orthop Relat Res 2007;464:21e6. [13] Fisher DA, Dierckman B, Watts MR, Davis K. Looks good but feels bad: factors that contribute to poor results after total knee arthroplasty. J Arthroplasty 2007;22(6 Suppl 2):39e42. [14] Wallston KA, Strudler Wallston B, DeVellis R. Development of the Multidimensional Health Locus of Control (MHLC) scales. Health Educ Monogr 1978;6: 160e70.