Optimism and pessimism are antithetically associated with post-operative knee function in patients’ undergoing total knee arthroplasty

After Institutional Review Board approval, all patients older than 18 years that met standardized criteria for surgery [17] and were scheduled for primary TKA due to end stage OA were eligible for inclusion. After informed consent, they were enrolled in the study. Exclusion criteria were German language inability, noncompliance of signing a written consent, life expectancy less than 1 year as judged by the treating physician, and any health factors that would preclude elective surgery (i.e., decompensated cardiac insufficiency, symptomatic anemia, etc.). Patient withdrawal was accepted at any time during the trial.

Patients were followed for 12 months post-operatively during the regular administrative procedures of the participating hospitals. Evaluation started in 05/2018 and ended in 01/2021. Baseline data (age, age at surgery, sex, type of surgery, ASA score, BMI, marital status, education level and income) for the eligible patients were elicited before surgery. Patients’ personality dimensions were queried pre-operatively using the German version of the revised Life Orientation Test (LOT-R) (Fig. 1) [18]. The LOT-R is a widely used instrument to measure dispositional optimism and pessimism according to the model of Scheier and Carver published in 1985 [12]. The score consists of 10 items, with three items (1, 4 and 10) assessing optimism, three items (3, 7 and 9) assessing pessimism and four filler items. Answers are given on a 5-point Likert scale, response categories range from strongly agree to strongly disagree. The scores of the optimism and pessimism sub-scales are the sum of the scores of the corresponding items. A total score can be calculated, adding the optimism and the inverted pessimism score [19]. Its reliability and validity have been established [20] and norm values have been determined [21], while it measures concepts with a high test–retest reliability [22]. In this study, in case of the regression analysis, the LOT-R Score variable was used as a separate, continuously bi-dimensional variable for both pessimism and optimism, which is the preferred application of this test [20, 21]. High score levels on the separate scales of the both variables “optimism” and “pessimism” indicated “high optimism” or “low pessimism”, respectively. In case of the descriptive statistics, LOT-R was categorized. “Optimistic” was defined as the following: 0–6 points = not optimistic, 7–9 points = somewhat optimistic, 10–12 points = very optimistic, whereas “pessimistic” was defined as the following: 0–6 points = very pessimistic, 7–9 points = somewhat pessimistic, 10–12 points = not pessimistic. Functional outcome was measured via the 12-Item Knee Osteoarthritis Outcome Score (KOOS-12) pre-operative and at 3-, 6- and 12-months post-operative. The KOOS-12 score evaluates symptoms and function using items from three subscales: “pain”, “function”, daily living”, and “quality of life”. Items are scored on a 4-point Likert scale and scores range from 0 to 100, with higher scores indicating better results. The score and its German version have been validated and checked for reliability, and the minimum clinically important difference (MCID) was detected at 11.1 points [23‐25].

The surveys (LOT-R, KOOS-12) were handed to included patients after informed consent during the last visit at the outpatient clinic prior to surgery (range 1–8 weeks), and post-operative at the above-described follow-ups. Patients completed the surveys on their own, with qualified study nurses on-site on demand if questions arose. Collected data was pseudonymized and stored electronically.

The data of 933 TKA patients were included in the study. At last follow-up, 740 patients with complete data could be analyzed. Baseline characteristics are shown in Table 1.

Approval from all ethics committees in the participating states in Germany was obtained prior to the study [Institutional Review Board approval: Submission No.: 837.533.17 (11367), Ethics Committee at the State Chambers of Physicians of Rhineland-Palatinate; B-F-2018-042, Ethics Committee at the State Chambers of Physicians of Baden-Wuerttemberg; MC 84/2018, Ethics Committee at the State Chambers of Physicians of Hesse].

To quantify the possible impact of LOT-R optimism/pessimism on post-operative outcomes, several linear regression models were fitted, in which the respective outcomes (KOOS-12 post-operative) were regressed on the study variable LOT-R optimism/pessimism. In addition to the respective study variable, each regression model included the possible confounders: age, ASA score, BMI, pre-operative KOOS-12 at baseline, EQ5D score, HSS expectation score, education background (low vs. high) and sex as covariates.

To obtain the regression equation for the association of LOT-R pessimism with the respective KOOS-12 score, the LOT-R- optimism score was replaced with the LOT-R-pessimism score of a given patient.

To prevent an attrition bias, all regression models were based on 10 datasets that were obtained by means of multiple imputation (using predictive mean matching as the imputation method). The imputation model included all recorded variables that were significantly correlated with LOT-R optimism/pessimism and the KOOS-variable.

All statistical analyses were performed using R [R version 3.5.1, Core Team (2017)] [26]. The level of significance was set at 0.05. While continuous variables were summarized with their mean and standard deviation, categorical variables were summarized with numbers and percentages.

Mean LOT-R optimism was 10.5 (SD 2.12). Mean LOT-R pessimism was 9.4 (SD 2.03). Distributions of LOT-R optimism and LOT-R pessimism are shown in Figs. 2 and 3.

Mean KOOS Scores for pre- and post-operative follow-ups in association with the respective optimistic and pessimistic categories are shown in Tables 2 and 3.

Optimistic patients showed significantly higher pre-operative KOOS-12 scores compared to their not optimistic counterparts (p ≤ 0.001) (Fig. 4). Pessimistic patients showed significantly lower pre-operative KOOS-12 scores compared to their not pessimistic counterparts (p ≤ 0.001) (Fig. 5).

Associations of LOT-R optimism and LOT-R pessimism scores with absolute post-operative KOOS-12 scores (3, 6 and 12 months) in the multivariate analysis are shown in Table 4, Figs. 6 and 7). Further analysis of the association of post-operative function and optimism/pessimism in the univariate analysis is shown in Table 5. This analysis showed an overall low positive correlation of LOT-R optimism and LOT-R pessimism with KOOS-12 values, respectively (R²-values shown in Tables 4 and 5).

This study has some limitations. First, other psychological variables known to interfere with post-operative outcome, such as depression and anxiety [6, 34], could not be taken into account, as they were not individually recorded in this database. Second, due to the character of an observational study without intervention, only associations and no causations could be investigated. Third, only PROMs and no objective outcome measures were used for measurement of post-operative functional outcome. Fourth, potential selection bias exists, as only patients already planned to undergo TKA were examined in this study.

However, due to the multi-center design, a large study sample size matching the general population of TKA patients in Germany [35] could be analyzed. It is notable that the optimism/pessimism survey took place close to surgery date (maximum 8 weeks pre-operative), suggesting high correlation with the patients’ actual current personality trait, especially as optimism/pessimism is known as a stable trait with high test–retest reliability [22]. Furthermore, treatment pathway and follow-up treatment were standardized in all participating study centers.