Background

Total knee arthroplasty (TKA) is a highly effective treatment method of end-stage osteoarthritis and most patients experience pain relief within 3-6 months. However, about 20% of the patients are not satisfied nor pain free. The causes for recurrent pain after TKA are manifold and range from knee joint-related factors such as infection, arthrofibrosis, patellofemoral problems, malposition or malalignment, loosening or instability to non-knee joint-related causes such as psychological disorders, vascular pathologies, back or hip problems. Hence, the diagnostic process is demanding. The primary aim was to assess characteristics of pain in patients with ongoing pain after TKA and link the identified pain patterns to underlying pathologies. The secondary aim was to investigate the position of TKA components and evaluate bone tracer uptake (BTU) using pre-revision SPECT/CT and correlate these findings with the pain characteristics.

Methods

A prospectively collected cohort of 83 painful primary TKA patients was retrospectively evaluated. All patients followed a standardised diagnostic algorithm including 99m-Tc-HDP-SPECT/CT, which led to a diagnosis indicating revision surgery. Pain character, location, dynamics and radiation were systematically assessed as well as TKA component position in 3D-reconstructed CT. BTU was anatomically localised and quantified using a validated localization-scheme. Component positioning and BTU were correlated with pain characteristics using nonparametric Spearman correlations (p<0.05).

Results

Most frequent pain characters were pricking/lancinating (45.7%), pinching/crushing and dull/heavy (38.6%); 89.5% of all patients localized their knee pain anteriorly; 48.1% reported pain aggravations by descending stairs. Radiating pain was reported in 14% of the patients. Patella-related problems (56.7%) and instability (52.6%) were the most frequent pathologies. Significant correlations were found between pain and patients characteristics and SPECT/CT findings resulting in nine specific patterns. The most outstanding ones include: Pattern 1: More flexion in the femoral component correlated with tender/splitting pain and patella-related pathologies. Pattern 3: More varus in the femoral component correlated with dull/heavy and tingling/stinging pain during descending stairs, unloading and long-sitting in patients with high BMI and unresurfaced patella. Pattern 6: More posterior slope in the tibial component correlated with constant pain.

Conclusion

The results of this study involving specific pain patterns in unhappy TKA patients, help to place component positioning in the overall context of the "painful knee arthroplasty". Furthermore, the findings further differentiate and define the clinical picture of a painful knee after TKA. Knowing these patterns enables a prediction of the cause of the pain to be made as early as possible in the diagnostic process before the state of pain becomes chronic.