Background The afferent lymphatic network and the draining lymph node (LN) are fixed environments acting as remote complementary check-points through progressive phases of the inflammatory response. LN Power Doppler ultrasonography (PDUS) is a non-invasive and sensitive imaging technique commonly adopted in the preoperative diagnostic work-up of cancers. Whether the draining LN undergoes PDUS-sensitive dynamic modifications during chronic inflammation in humans and whether these modifications can be tracked to improve disease assessment is currently unclear.
Objectives To investigate axillary LN sono-morphology and Doppler activity in established RA, defining the existence of alterations compared to healthy individuals and dissecting their relationship with synovitis in peripheral joints.
Methods 40 RA patients refractory to conventional synthetic DMARDs (DAS28≥3.2) were evaluated through complete clinical and PDUS examination in hands/wrists and axillary LN. 20 healthy individuals served as controls. 31 patients starting anti-TNF treatment were followed-up prospectively with further assessments at weeks 4 and 24. Axillary LN PDUS characteristics were assessed according to 0–3 semi-quantitative scores for global LN volume, structure of the lymphocyte-rich cortex and local perfusion. Patient-related indices for each parameter were constructed for comparative, correlative, response-to-treatment and outcome analyses. Synovitis degree (PD and gray scale) was quantified according to 0–3 semi-quantitative scales and 12-joint cumulative indices.
Results Baseline assessment of perfusion and sono-morphologic characteristics showed the existence of quantitative changes in a subset of RA axillary LN, collectively leading to a distinct sonotype differentiating patients and controls. Combining data derived from morpho-structural and vascularity indices, evidence for parameters exceeding the threshold of controls was restricted to 17 out of 40 patients (42.5%), indicating variable levels of LN response despite established active disease. No relationships were observed between LN parameters and ACPA-RF status, DAS28, acute phase reactants, tender-swollen joint counts. Significant correlations (at systemic and ipsilateral level) were instead consistently detected when sensitive PDUS imaging of the synovium was applied (LN-joint PD index correlation: r=0.35, p=0.03). Plasticity of the observed LN alterations was confirmed by responsiveness analyses showing clear-cut reduction at week 24 of both structural and perfusion scores in patients with LN alterations at baseline (LN PD index: p<0.002, LN cortical index: p<0.02, LN volume index: p<0.03). Patient stratification according to EULAR response at 6 months demonstrated the existence of specific differences in baseline LN scores (but not in clinical features and joint US parameters) with significant decrease in the moderate/non responder group (LN PD index median [IQR] 0 [0–1] vs 2 [1–4.25], p=0.01). At its lowest detectable threshold, the LN perfusion index was the only negative predictor of achievement of good response (OR=0.08, p=0.005).
Conclusions PDUS assessment of axillary LN allows detection of inflammatory-related changes in a subset of patients with active established RA. Lack of signs of LN challenge appears a negative predictor of clinical response.
Disclosure of Interest None declared