Postural orthostatic tachycardia syndrome explained using a baroreflex response model

Figures

• 

  Figure 1.. (a) Haemodynamics is controlled by the baroreflex system, which senses changes in carotid arterial pressure predicted as a function of upper body arterial pressure. Afferent signals from baroreceptor neurons are integrated into the nucleus solitary tract (NTS) and transmitted via sympathetic and parasympathetic neurons regulating HR, peripheral vascular resistance, and ventricular contraction. The systemic circulation is represented by compartments lumping upper (au) and lower (al) body arteries, upper (vu) and lower (vl) body veins and the left heart (lh). The upper body compartment contains organs above the lower abdomen, including the abdominal splanchnic vessels, and the lower body compartment contains organs below the lower abdomen. Flow (Q) through the aortic valve (av) is transported from the left heart to the upper body arteries. From here, it is transported to the lower body arteries and through the upper body peripheral vasculature (up) to the upper body veins. A parallel connection transports flow through the lower body peripheral vasculature (lp). From the lower body venous flow is transported to the upper body veins and finally via the mitral valve (mv) back to the…

• 

  Figure 2.. Head-up tilt (HUT) test. Patients are tilted, head up, from 0 to 60∘ over 7 s. The shaded regions illustrate the blood volume distribution.

• 

  Figure 3.. Left panels show the decreasing (a) and increasing (c) Hill functions (light blue, grey, light pink) for characteristic values of kX and P2X. Model predictions using these functions are superimposed in (blue, black, pink). (b,d) The time series predicted from the differential equation (2.11) using the Hill functions in (a,c).

• 

  Figure 4.. Process of obtaining amplitude of the ∼0.1 Hz component of a heart rate signal, the same process is used for blood pressure. Starting from the left, the solution obtained using a variable step size solver is interpolated at 100 Hz. The last 200 s are marked (black trace in (a)). Second, the discrete Fourier transform is applied to obtain the amplitude of the frequencies. We examine the frequency range corresponding to the baroreflex, 0.05–0.15 Hz (black spectrum in (b)). Third, we find the maximum of the amplitude in this range (black column in (c)).

• 

  Figure 5.. (a) From left to right: heart rate (H, top) and upper arterial pressure (Pau, bottom) predictions for kH=10, 20 and 30. Note the large amplitude oscillations in the right panel are higher than values observed in patient data but are included to illustrate the behaviour of the model. (b) From left to right: maximum and minimum values for varying values of kH, the amplitude of the ∼0.1 Hz region response, and frequency of oscillations. Enlarged red dotes show denote measurements corresponding to kH=10,20,30.

• 

  Figure 6.. Two-dimensional parameter analysis of kR versus kH. (a) Amplitudes of peak heart rate (H) oscillation (left) and peak upper arterial blood pressure (Pau) oscillation (right) at the ∼0.1 Hz frequency band for values of kR and kH at rest (top) and head-up tilt (HUT, bottom). (b) The same information as (a) but with 2% noise. Average measurements from data [7] are marked for control patients at rest (CR), and POTS patients during head-up tilt (PH). Note that the physiologically possible oscillations correspond to the green regions. (c) Heart rate predictions during HUT for red dots on lower panels of (a); ith panel from top corresponds to ith dot from the left in (a). (d) Similar information as (c) but pertaining to (b).

• 

  Figure 7.. Two-dimensional parameter analysis of blood volume (BV) versus kH. (a) Amplitudes of peak heart rate (H) oscillation (left) and peak upper arterial blood pressure (Pau) oscillation (right) at the ∼0.1 Hz frequency band for values of BV and kH at rest (top) and head-up tilt (HUT, bottom). (b) The same information as (a) but with 2% noise. Average measurements from data [7] are marked for control patients at rest (CR), and POTS patients during head-up tilt (PH).

• 

  Figure 8.. Results of simulation with HUT at t=75. Row 1: heart rate (H, bps), left ventricle pressure (Plv, mmHg) row 2: upper arterial pressure (Pau, mmHg), lower arterial pressure (Pal, mmHg) row 3: upper venous pressure (Pvu, mmHg), lower venous pressure (Pvl, mmHg).

• 

  Figure 9.. Characteristic data for a control and hyperadrenergic POTS patient (black) and model predictions (red and green) of heart rate (H), carotid artery blood pressure (Pc) and mean pressure. Simulations with 4500 ml of blood are in the top row with simulations with 3500 ml of blood in the bottom row. Cau is decreased after HUT for all simulations to represent constriction of vasculature upon HUT. In control P2H is decreased (left), kH and P2H are increased after HUT to replicate hyperadrenergic POTS (middle) and kH increased, RlpM and Rlpm decreased to replicate neuropathic POTS (right). To compare model predictions H data are scaled such that the baseline is 1 bps and blood pressure vary from 80 to 120 mmHg.