Introduction: The postoperative rehabilitation after a primary knee arthroplasty may be infiuenced by a variety of factors. Nevertheless, only a few studies evaluated the effect of various factors on patients’ short-term outcome. This prospective study was conducted to evaluate the effect of different factors on patients’ function six weeks after the surgery.

Patients and methods. We prospectively recruited 107 patients with osteoarthritis who underwent an uncomplicated total knee arthroplasty, using the same prosthesis and operative technique. Following variables were collected before and after the surgery: age, BMI, visual analogue pain score at rest and during activity, preoperative range of knee motion, involvement of other joints, comorbidities (Katz index), self assessed health status, admission and discharge hemoglobin levels, amount of blood transfusions and intensity of postoperative physiotherapy.

In order to quantify patients’ level of functioning, we used a timed up and go test (TUG) and the Oxford knee score that were collected before and after the surgery. To eliminate the infiuence of postoperative weakness on rehabilitation, hand grip measurements were performed as well. A multivariate regression analysis was performed to examine the infiuence of different peri-operative variables on the outcome measures. Adjusted R2 was measured to estimate the explanatory power of infiuence of these variables.

Results: There was no significant difference between preoperative and postoperative hand grip force measurements, indicating that the general strength of the patients did not deteriorate. A postoperative TUG was worse with higher preoperative TUG and higher rest pain score (adjusted R2=0.53). The amount of improvement in TUG was better only with lower rest pain score (adjusted R2=0.06). A postoperative Oxford hip score was better only with lower rest pain score (adjusted R2=0.30). The amount of improvement in the Oxford score was not infiuenced by any of the variables (adjusted R2=0.01). Only significant infiuences (p< 0.05) are mentioned.

Discussion: Most of preoperative and postoperative measured variables, including age, BMI, comorbidities, hemoglobin concentration and amount of physiotherapy had no significant effect on patient’s functional status after uncomplicated knee arthroplasty. Only the pain at rest had infiuence on the functional result. These results suggest that patient personality has a most significant effect on knee arthroplasty results, either through pain perception or otherwise.