Use of the Valsalva manoeuvre to identify haemodialysis patients at risk of congestive heart failure.

BACKGROUND In the presence of elevated cardiac filling pressures, the decline of blood pressure (BP) during the straining phase of a Valsalva manoeuvre is blunted or absent. We compared the use of non-invasively measured BP response to a Valsalva manoeuvre with clinical assessment and bioimpedance measurements to identify haemodialysis patients at risk of acute congestive heart failure (CHF). METHODS Continuous BP response (Finapres) to a Valsalva manoeuvre, clinical assessment by nephrologists, and bioimpedance estimations of extracellular fluid volume were determined before and after haemodialysis, once every week during a 5-week period. Acute CHF was defined according to preset clinical and radiological criteria. RESULTS Participants (age 60+/-19 years, six females, nine males) had an average predialysis weight of 66.8+/-11.8 kg. Patients were dialysed for 3.8+/-0.8 h with a mean ultrafiltration of 2.4+/-1.1 litres. Valsalva systolic BP ratios (phase 2 to 1) decreased significantly during dialysis from 0.81+/-0.11 to 0.73+/-0.10 (P<0.05). Five patients experienced an episode of acute CHF. The Valsalva BP ratios for these patients before and after dialysis (0.89+/-0.05 and 0.78+/-0.05 respectively) were higher than for the remaining ten patients (0.77+/-0.10 and 0.70+/-0.11, respectively) (P<0.05). A cutoff Valsalva BP ratio of 0.82 resulted in positive and negative predictive values for CHF of 62 and 100% respectively. No differences in clinical assessment or bioimpedance parameters were found, with the exception of postdialysis diastolic BP and predialysis ankle oedema. After treatment of CHF, Valsalva BP ratios decreased significantly without changes in the other hydration parameters. CONCLUSIONS Non-invasive assessment of the BP response to a Valsalva manoeuvre appears to be a potential tool for identifying patients at risk of acute CHF during maintenance haemodialysis.