Sunagawa and coworkers have proposed an ejecting, single beat, cosine curve matching method for the estimation of the maximum potential left ventricular pressure during an isovolumic contraction. We tested the hypothesis that this same information could be obtained from maximum dp/dt and the length of the ventricular contraction curve using a simple formula. The method was tested using Sunagawa's reported data as well as on data obtained from five dogs over a wide range of afterloads, and from three rabbits, four pigs, and four sheep. A high correlation coefficient between the two methods (.85 to .97), based on a linear model, was obtained in all the experiments. The formula method gave slightly higher values than the cosine curve matching procedure, was much simpler, and involved few of the curve matching assumptions. The maximum left ventricular pressure was also highly correlated with stroke work. These results imply that the maximum potential left ventricular pressure or energy is represented in, and can be predicted from, the ventricular pressure during normal ejection at a given end diastolic volume. When combined with measurements of end systolic pressure and stroke volume, reliable beat-to-beat estimates of the slope of the end systolic pressure-volume relationship (end systolic elastance) might be obtainable from single, left ventricular beats.

The use of end-systolic elastance as a parameter of left ventricular contractility is based on a theory put forward by Sagawa (1) and is limited by an assumed linearity of the model for a time varying compliance. A major problem is the contractility-dependent curvilinearity of the end-systolic pressure-volume relationship (ESPVR), which may result in intercept volumes that are below a passive unstretched volume. Based on the experimental data of Burkhoff et al. (2), we demonstrate the reasons for the lack of correlation of the slope (k) of the end-systolic: pressure (P-es)-volume (V-es) relationship with ventricular contractility. A parabolic relation, P-es = a . V-es(2) + b . V-es + c, was used for approximation of the ESPVRs, as proposed by Burkhoff et al., and the slope (k = 2 . a . V-es + b), together with the volume axis intercepts (V-0), calculated for the tangents for all P-es/V-es points of the ESPVRs. The results demonstrate the volume range dependence of the slope and V-0 of linear regression lines. However, intercepts of the ESPVRs tangents with a midrange constant pressure line (e.g., 80 mmHg or 100 mmHg) justify the use of the shift of the end-systolic pressure-volume relationship as a parameter of left ventricular contractility. This method appears to be valid over a much wider range of end-systolic volumes than is the use of the slope alone.

There is an ongoing discussion whether the heart is the primary target organ responsible for the development of cardiovascular failure during septic shock as well as its onset. We tried to study the reaction of the heart to sepsis in the early phase of 8 h, using a sublethal model of sepsis in six awake cross-bred Austrian mountain sheep. Sepsis was induced by infusion of a live Escherichia coli suspension at a dose of 5 x 10(7) colony-forming units per kg body weight over 8 h. Standard hemodynamic, hematologic and serum tumor necrosis factor (TNF) measurements were obtained. For evaluation of left ventricular performance we used the following methods, tested in five pilot experiments: 1) The shift of the end-systolic pressure-diameter relation. This was characterized by the calculated shift of the transverse external end-systolic diameter of the left ventricle at a ''midrange'' end-systolic pressure of 100 mmHg (end-systolic ventricular diameter deviation, ESVDD100). Calculations were performed using a second order regression function of the end-systolic pressure diameter points obtained by variation of afterload by a cuff occluder on the aorta; 2) The shift of the (dP/dt)max over end-diastolic diameter ratio compared to control values estimated by a graphical approach. Mean pulmonary pressure increased from 21 +/- 1 to 36 +/- 2 mmHg in the first hour after starting the E. coli infusion and remained elevated during the entire 8 h observation period. Serum TNF was found to peak 1 hour after start of E. coli infusion and was hardly detectable after 3 hours of bacteremia. Mean aortic pressure showed minor changes (maximum 105 +/- 3 mmHg, minimum 91 +/- 2 mmHg) and there were no statistically significant alterations of the cardiac index. ESVDD100 showed an ''oscillatory'' reaction in the first phase and a statistically significant decrease of contractility in the second phase (at 4 h). This was confirmed by the graphical method of the (dP/dt)max over end-diastolic diameter ratio. We may therefore conclude that there is no early depression of myocardial function or if so, it may be masked by adrenergic stimulation. In the later phase of the 8 h experiment there is a significantly decreased contractility of the heart. This may be compensated (e.g., ''Starling'' mechanism or heart rate increase) in this sublethal model.

Article Physikalische Untersuchungen zur Leitfähigkeitsmethode der Ventrikelvolumenmessung was published on January 1, 1993 in the journal Biomedical Engineering / Biomedizinische Technik (volume 38, issue s1).

In order to shed light on the controversy surrounding the choice of resuscitative fluids in shock, we used a canine model which we feel to be a superior mimic of human traumatic shock, combining hemorrhage (to a mean arterial pressure of 50 mmHg), fracture of both femora, and soft tissue crush. After 90 min, animals were resuscitated by reinfusion of shed blood, supplemented by 5% albumin (n = 8) or lactated Ringer's solution (n = 8). Plasma colloid osmotic pressure (COP), transcapillary escape rate for albumin (TER), total lung water and extravascular lung water (EVLW) were measured. COP fell in both groups, but remained above 9 mmHg in the albumin recipients, while falling below 7 in those receiving crystalloid (P < 0.05). Overall, the increase in EVLW averaged 20%; albumin recipients fared better (9.7%) than Ringer's recipients (31.1%), but wide inter-animal variation precluded statistical significance (P = 0.095). TER rose 30% per hour, without difference between groups. Quality of resuscitation (achieved blood pressure and cardiac output) was somewhat better in the albumin group. We conclude that this model allows study of the early microvascular leakage seen in shock; within the time-frame studied (maximum 4.5 h following shock), colloid and crystalloid resuscitation were approximately equivalent.

Bioassays using isolated animal hearts are important tools for the investigation of cardiac behaviour, but to obtaine accurate results a proper perfusion circuit has to be designed. In particular, biophysical studies of contractile and vascular behaviour require a perfusion circuit which permits the adjustment of several experimental parameters within wide ranges. It must also be able to maintain the stability of these parameters when the behaviour of the isolated organ undergoes major changes. To meet this requirement, we have developed a perfusion circuit which makes it possible to control either the perfusion pressure or the coronary flow, with a high degree of precision. There is an electronic controller which satisfies the requirements of a variety of safety and experimental requirements and guarantees a well-defined perfusion system. Computer simulation of the interaction between the perfusion circuit and the heart identified the basic elements of this time-variable, nonlinear system.