[Telephone coaching for depression]. / Telefoncoaching bei Depression.

BACKGROUND: Depression is associated with a substantial utilization of resources in the German healthcare system. A typical symptom in depression is loss of drive, which possibly contributes to non-adherence and increased costs. OBJECTIVE: The study is based on routine healthcare data and tested the hypothesis that telephone coaching in cases of depression leads to a reduction in total healthcare costs. MATERIAL AND METHODS: Based on approximately 80 covariates and using propensity score matching, a total of 1586 persons who had received telephone coaching for depression and covered by a German statutory health insurance fund were matched to a comparable cohort of patients with depression to whom telephone coaching had not been provided. RESULTS: Within the study period of 12 months (3rd quarter 2012-4th quarter 2013) a positive program effect was observed for the intervention group by a significant reduction of total healthcare costs (2332 € vs. 2626 €, p = 0.0015) resulting in total savings to the statutory health insurance fund of 415,532 €. Investment costs amounted to 256,683.42 € leading to a return on investment of 1.62 € (total savings/total investment). The coaching program was well accepted by patients. CONCLUSION: Telephone coaching for depression was able to significantly reduce total healthcare expenditure and the intervention was well accepted by patients.

Hemodynamic responses to ECT in a patient with critical aortic stenosis.

We present a case study of a 46-year-old woman with a psychotic depressive illness of 2 months' duration with the coexisting medical diagnoses of critical aortic stenosis, severe labile hypertension, renal failure necessitating hemodialysis of 7-years' duration, and systemic lupus. Because of unresponsiveness to an antidepressant drug regimen, severe motor retardation, mutism, and refusal of food and fluids by mouth, an urgent indication for electroconvulsive therapy (ECT) was established. However, the patient refused ECT, and to allow its initiation, a court order was obtained. In view of the coexisting diagnoses of critical aortic stenosis, labile hypertension, and renal failure, ECT represented a substantially increased risk in this patient because of severe arterial hypertension and tachycardia. The patient was successfully managed during each ECT, using a combination of metoprolol by mouth, which was supplemented by i.v. esmolol immediately prior to the application of the ECT stimulus, and sodium nitroprusside, which was infused for several minutes prior to the seizure and thereafter to attenuate arterial hypertension. Nevertheless, sudden death, a well-known complication of critical aortic stenosis, occurred 96 hours after the fourth ECT.

Intracranial haemodynamics during attenuated responses to electroconvulsive therapy in the presence of an intracerebral aneurysm.

OBJECTIVES: This report describes successful anaesthesia and electroconvulsive therapy (ECT) in a patient with an unruptured basilar artery aneurysm. ECT is associated with a hyperdynamic state characterised by arterial hypertension, tachycardia, and considerably increased cerebral blood flow rate and velocity. These responses pose an increased risk for subarachnoid haemorrhage when an intracranial aneurysm coexists. METHODS: A 54 year old woman presented for ECT. She had a 20 year history of major depression which was unresponsive to three different antidepressant drugs. There was also an unruptured 5 mm saccular aneurysm at the basilar tip, which had been documented by cerebral angiography, but its size had remained unchanged for the previous four years. After she declined surgical intervention, she gave informed consent for ECT. During a series of seven ECT sessions middle cerebral artery flow velocity was recorded by a pulsed transcranial Doppler ultrasonography system. She was pretreated with 50 mg oral atenolol daily, continuing up to the day of the last ECT and immediately before each treatment, sodium nitroprusside was infused at a rate of 30 microg/min, to reduce systolic arterial pressure to 90-95 mm Hg. RESULTS: Systolic flow velocity during the awake state ranged from 62-75 cm/s, remaining initially unchanged with sodium nitroprusside infusion. After induction of anaesthesia (0.5 mg/kg methohexitone and 0.9 mg/kg succinylcholine), flow velocities decreased to 39-54 cm/s, reaching maximal values of 90 cm/s (only 20% above baseline) after ECT. These flow velocities recorded post-ECT were considerably below the more than twofold increase recorded when no attenuating drugs were used. Systolic arterial blood pressure reached maximal values of 110-140 mm Hg and heart rate did not exceed 66 bpm. Rapid awakening followed each treatment, no focal or global neurological signs were apparent, and the patient was discharged in remission. CONCLUSION: In a patient with major depression and a coexisting intracerebral saccular aneurysm who was treated with ECT, the combination of beta blockade with atenolol and intravenous infusion of sodium nitroprusside prevented tachycardia and hypertension, and greatly attenuated the expected increase in flow velocity in the middle cerebral artery.

Renal vascular hyperresponsiveness to elevated ionized calcium in spontaneously hypertensive rat kidneys.

OBJECTIVES: Calcium may be indicated in critically ill patients for hemodynamic support. Its well-known action includes peripheral vasoconstriction. Vascular effects of calcium are unknown, however, in the presence of hypertension or in combination with calcium channel blocking drugs, commonly prescribed in the treatment of hypertension. The renal vessels of the spontaneously hypertensive rat (SHR) represent a suitable study model, because their vascular reactivity closely agrees with that in hypertensive humans. The present study should clarify (a) are the renal vessels of SHR responsive to high and low ionized calcium ([Ca+2] within the clinical ranges? (b) because release of nitric oxide is calcium ion dependent, are renal vascular responses altered after inhibition of NO synthase? (c) are vascular responses of SHR to hypercalcemia altered by the calcium channel blocking drug verapamil? ANIMALS AND INTERVENTIONS: We compared isolated kidneys of SHR and those of two strains of age-matched normotensive rats (NTR) in their responses to high and low [Ca+2]. They were perfused with oxygenated, warmed (37 degrees C) albumin containing Krebs-Henseleit buffer. In protocol A (n = 8 for each rat strain) steady state high [Ca+2] (1.88 mmol/l) and low [Ca+2] (0.55 mmol/l) were instituted in randomized order. In protocol B (n = 8 for each rat strain) interventions identical to those of protocol A were instituted after inhibition of NO synthase with NG monomethyl-L-arginine (L-NMMA). In protocol C, high and low [Ca+2] levels were instituted in SHR after verapamil pretreatment. At each [Ca+2] we measured changes in renal flow at constant perfusion pressures of 100 and 150 mm Hg. RESULTS: In SHR (perfusion pressure 100 mm Hg), high [Ca+2] induced a decrease in renal flow (-11.8 +/- 1.8% of control), which was significantly greater (p < 0.05) than the change (-6.1 +/- 1.5 and -6.9 +/- 1.4% of control) recorded in the two normotensive strains. In SHR (perfusion pressure 150 mmHg), high [Ca+2] induced a decrease in renal flow (-12 +/- 1.3% of control), also significantly greater (p < 0.05) than the changes (-6.2 +/- 1.1 and -5.8 +/- 1.7% of control) in the two normotensive strains. Similar differences and significances were again observed after L-NMMA pretreatment. In SHR, verapamil prevented renal vascular responses in SHR to both high and low [Ca+2]. CONCLUSIONS: First, renal vascular responses to high [Ca+2] in SHR are exaggerated. At the upper end of the hypercalcemia range the observed changes in renal flow at constant perfusion pressure were modest, however, and with lesser degrees of hypercalcemia they may be anticipated to be even less pronounced. Second, effects of high [Ca+2] were abolished after verapamil. If these findings are clinically applicable, they are of interest when calcium is infused in patients with hypertension.

Renovascular responses to high and low perfusate calcium steady-state experiments in the isolated perfused rat kidney with baseline vascular tone.

Acute hypercalcemia is commonly observed in surgical patients after calcium infusion while acute hypocalcemia is common during rapid citrated blood transfusion. Although high and low ionized calcium ([Ca2+]) within the clinical range produce an increase or decrease in cardiac performance and systemic vessel resistance, respectively, their effects on renal vessels have not been quantified. A possible renal vasoconstriction that might occur with high [Ca2+] is of clinical interest because it is a factor which may contribute to impaired renal circulation and decreased function. In this study we examined the renovascular responses to [Ca2+], which was varied within the clinical range under hemodynamically controlled conditions. We instituted high and low [Ca2+] in the per fusate, which consisted of Krebs-Henseleit buffer containing albumin, 60-65 g/liter. Stable high (n = 10) or low (n = 7) [Ca2+] (1.93 +/- 0.02 and 0.59 +/- 0.01 mM, respectively) was instituted for 10 min and preceded and followed by normal [Ca2+] of the same duration. In a separate protocol (n = 8) verapamil (10(-5) M) was added to the perfusate 10 min before high [Ca2+] was tested. We measured changes in renal flow at a constant perfusion pressure of 110 mm Hg and also characterized the renal vessels over a range of pressures by pressure vs flow plots. High [Ca2+] was associated with a small decrease in flow (from 28.8 +/- 2.4 to 26.9 +/- 2.6 ml/min/g, P < 0.02), indicating a small vasopressor effect. This effect was also shown by a leftward shift in the pressure vs flow plots. These changes were prevented by verapamil. GFR decreased (from 0.35 +/- 0.04 to 0.28 +/- 0.06 ml/min/ g, P < 0.01) without a significant change in sodium excretion or fractional sodium excretion. Low [Ca2+] was associated with increased renal flow (from 30.8 +/- 2.1 to 35.2 +/- 2.7 ml/min/g, P < 0.02), indicating a vasodilator effect. This effect was also shown by a rightward displacement of the pressure vs flow plots. GFR increased from 0.51 +/- 0.03 to 0.56 +/- 0.04 ml/min/ g, P < 0.01, as did sodium excretion (from 2.32 +/- 0.22 to 3.87 +/- 0.49 microEq/min, P < 0.01) and fractional sodium excretion (from 2.33 +/- 0.26 to 3.61 +/- 0.49%, P < 0.01). We conclude, first, that in the isolated perfused rat kidney, high [Ca2+] is a weak vasopressor while low [Ca2+] has vasodilator action. Second, high [Ca2+] effects are abolished by verapamil pretreatment. These findings illuminate mechanisms of high [Ca2+] effects on renovascular tone.

Oscillometric blood pressure measurements by different devices are not interchangeable.

Blood pressure (BP) is frequently measured in patients by noninvasive blood pressure (NIBP) monitors. Values obtained by oscillometric devices of different brands may appear in one patient's record as if they were interchangeable; their concordance, however, has not been established. In 25 patients with major depression who were treated with electroconvulsive therapy (ECT) BP was measured on either arm by devices manufactured by SpaceLabs (SpL, 12 patients, 182 data points) and Marquette (Marq, 13 patients, 193 data points), respectively, and comparisons were made with simultaneous measurements on the opposite arm by Dinamap 1846SX (DIN), during the awake state and at 1-min intervals up to 5-7 min after ECT. Because ECT is associated with an intense, but short-lasting hyperdynamic state, comparisons of BP values could be made over a wide range of pressures. Bland-Altman plots were constructed to show the distribution of pressure differences at all pressures. Agreements between two instruments were judged according to guide lines by the American Association for Advancement of Medical Instrumentation (AAMI). The standard deviation of the difference (SDD) between two DIN devices was 7 mm Hg for systolic (SBP) and 6.3 mm Hg for diastolic blood pressure (DBP), whereas mean differences were 0.9 and 0.2 mm Hg, respectively (P = not significant [NS]), thus showing reproducibility. Corresponding SDD values SpL versus DIN were 9.1 for SBP and 8.3 mm Hg for DBP, while the mean differences were 1.6 (P = 0.026) and 7.3 (P = 0.0001) mm Hg, respectively. Corresponding SDD values for Marq versus DIN were 11.8 and 9.7 mm Hg with mean differences of 0.8 (P = NS) and 0.3 (P = NS) mm Hg. Whereas SBP differences DIN versus DIN exceeded 10 mm Hg in only 10% of observations, they exceeded that threshold in 31% and 32% of observations for SpL versus DIN and Marq versus DIN, respectively. In view of the variability that exceeds the AAMI guidelines and the one out of three occurrence of individual SBP differences exceeding 10 mm Hg for comparisons of SpL or Marq versus DIN, measurements by these three oscillometric devices are not interchangeable.

Renal vascular responses to high and low ionized calcium: influence of norepinephrine in the isolated perfused rat kidney.

OBJECTIVE AND DESIGN: The aim of this study was to examine the influence of norepinephrine (NE) on renal vascular responses to high (1.88 mmol/L) and low (0.56 mmol/L) perfusate-ionized calcium ([Ca2+]) in the isolated perfused kidney of the rat. High and low [Ca2+] encompassed the clinical concentration range in this multiexperiment, randomized trial. MATERIALS AND METHODS: Rats (n = 25), ranging in age from 3 to 4 months, were anesthetized and the ureter and renal artery were cannulated. The right kidney was perfused with oxygenated, warmed albumin (67 g/L) containing Krebs-Henseleit buffer and placed in a thermostated chamber without interruption of flow. In protocol A (n = 7), steady-state high [Ca2+] (1.88 mmol/L) and low [Ca2+] (0.56 mmol/L) were instituted in randomized order in each experiment under basal conditions. In protocol B (n = 9), the same interventions were instituted during constant rate NE infusion. Changes in renal flow were measured at constant perfusion pressure (110 mm Hg), and renal vascular resistance (RVR) was calculated. Renal function was assessed by clearance of [14C]inulin and by fractional excretion of sodium. With NE-induced preconstriction, the increase in RVR observed during high [Ca2+] was +17.8 +/- 1.8% of control, and the decrease in RVR observed during low [Ca2+] was -35.9 +/- 8.2% of control. Both values were greater by a factor of 2 than corresponding results obtained under basal conditions (7 +/- 2.1% vs. -13.5 +/- 4.1% of control, respectively, p < 0.05). Whereas the decrease in glomerular filtration rate with high [Ca2+] was not significantly influenced by NE pretreatment (-9 +/- 1.8% of control with high [Ca2+] in combination with NE vs. 4.1 +/- 0.7% of control under basal conditions), the increase in glomerular filtration rate with low [Ca2+] was significantly greater in the presence of NE (12 +/- 0.7 vs. 102 +/- 8.5% of control, p < 0.01). CONCLUSIONS: Whereas under basal conditions renal vascular effects of high and low [Ca2+] (varied within the clinical concentration range) are small, the changes recorded with the same interventions after NE pretreatment are increased by a factor of > 2. Hypercalcemia-induced renovascular constriction and decreased function are unfavorable, especially in patients who are at risk for renal dysfunction from other causes.

Nitric oxide dose-response study in the isolated perfused rat kidney after inhibition of endothelium-derived relaxing factor synthesis: the role of serum albumin.

The dose-response relationship of steady-state nitric oxide (NO) administration on renal vascular resistance in isolated rat kidneys (IPRK) perfused at constant pressure was investigated after inhibition of NO synthesis with NG-monomethyl-L-arginine (L-NMMA). To study the influence of biological thiols on renovascular NO effects, experiments were carried out with Krebs-Henseleit (KH) perfusate solutions alone, and in combination with bovine serum albumin (KH-ALB). Steady-state administration of NO by gassing the perfusate with 0 to 340 ppm NO led to graded decreases in renovascular tone. The minimal effective NO perfusate concentration in the absence of endogenous NO synthesis was about 6 to 8 nM, whereas a near-maximal effect was observed with approximately 200 nM. The presence of albumin reduced the speed of onset of renal vasodilation and the maximal effect at a given concentration of NO. After termination of NO administration, NO-induced vasodilation persisted in KH-ALB perfused kidneys for 30 min, whereas KH-perfused kidneys showed a rapid reconstriction. These findings suggest that the prolonged, potent renal vasodilation was caused by a reaction of bovine serum albumin (BSA) with oxides of nitrogen to form S-nitroso-BSA. Nitrosothiol levels in the KH-ALB perfusate were found to be proportional to the concentration of NO administered. The above-mentioned findings, confirmed in identical experiments with diethylamine NONOate, a novel NO-liberating substance, support the biological importance of S-nitrosothiols (RS-NO) in the action and metabolism of endothelium-derived relaxing factor (EDRF) in the IPRK.

Comparative effects of esmolol and labetalol to attenuate hyperdynamic states after electroconvulsive therapy.

We studied 18 patients (age range, 53-90 yr) with at least one cardiovascular risk factor who were treated with electroconvulsive therapy (ECT) and compared effects of five pretreatments: no drug; esmolol, 1.3 or 4.4 mg/kg; or labetalol, 0.13 or 0.44 mg/kg. Each patient received all five treatments, during a series of five ECT sessions. Pretreatment was administered as a bolus within 10 s of induction or anesthesia. Doses of methohexital and succinylcholine were constant for the series of treatments and the assignment to no drug or to drug and dose was determined by randomized block design. Measurements of systolic and diastolic blood pressure (SBP, DBP) and heart rate (HR) were recorded during the awake state and 1, 3, 5, and 10 min after the seizure. The deviation of ST segments from baseline was measured by an electrocardiogram (ECG) monitor equipped with ST-segment analysis software. The results (mean +/- SEM) show that without pretreatment, there were significant (P < 0.05) peak increases in SBP and HR (55 +/- 5 mm Hg and 37 +/- 6 bpm, respectively), recorded 1 min after the seizure. Comparable reductions (by approximately 50%) in these peak values were achieved after esmolol (1.3 mg/kg) or labetalol (0.13 mg/kg), and cardiovascular responses were nearly eliminated after the same drugs in doses of 4.4 and 0.44 mg/kg, respectively. The deviation of ST-segment values from baseline in any lead was not measurably influenced by either antihypertensive drug. SBP values were lower after labetalol 10 min after the seizure, but not after esmolol. Asystolic time after the seizure was not significantly longer with either drug.(ABSTRACT TRUNCATED AT 250 WORDS)

Pulmonary vascular responses to hypercalcemia and hypocalcemia in the dog.

The pulmonary artery responses in the isolated whole-blood perfused canine lung to ionized calcium ([Ca++]) were quantified over a range of hypercalcemia and hypocalcemia values ([Ca++] = 0.23-1.88 mM) under conditions of controlled pulmonary blood flow and constant mean aortic and left atrial pressures. Calcium chloride, administered as bolus doses in the clinical range (5-15 mg.kg-1) at initial normocalcemia and without interventions producing vasoconstriction did not influence mean pulmonary artery pressure at constant pulmonary blood flow. Stable hypercalcemia ([Ca++] = 1.88 +/- 0.05 mM) did not influence the slope of the pulmonary artery pressure-flow plot. Because normal pulmonary vasomotor tone is low and cannot readily be lowered further, the possible vasodilator action of hypocalcemia was assessed by its ability to decrease the slope of the mean pulmonary artery pressure-flow plot, which had been first increased by alveolar hypoxia (AHX) or infusion of the prostaglandin endoperoxide analog U46619 (PG). During AHX (n = 5), a graded reduction from normocalcemia ([Ca++] = 1.08 +/- 0.02 mM) to moderate hypocalcemia ([Ca++] = 0.8 and 0.5 mM) did not alter the pulmonary artery pressure-flow plot, but severe hypocalcemia ([Ca++] = 0.26 +/- 0.01 mM) decreased the slope by 13 +/- 0.9 mmHg.l-1.min-1. The comparison of severe hypocalcemia ([Ca++] = 0.23-0.27 mM) versus a high dose of nifedipine (bolus of 10 micrograms/kg followed by continuous infusion at 40 micrograms.kg-1.h-1) on pulmonary vascular tone increased by either AHX or PG infusion indicated that both hypocalcemia and nifedipine decreased the slope of the relationship between mean pulmonary artery pressure and flow (during AHX: -16.1 +/- 1.38 and -23.3 +/- 1.73 mmHg.l-1.min-1, both P = 0.0001 vs. AHX alone, and during PG: -17.05 +/- 1.95 and -8.4 +/- 1.78 mmHg.l-1.min-1, P = 0.0001 vs. PG alone). Two principal conclusions emerge. First, the pulmonary vessels are minimally sensitive to changes in ionized calcium throughout the clinical hypercalcemia and hypocalcemia ranges; extreme hypocalcemia is required to produce vasodilation, which was reversed with calcium infusion. Second, whereas the pulmonary vasodilator effects of extreme hypocalcemia were independent of the intervention inducing pulmonary vasoconstriction (AHX vs. PG), those of nifedipine were much more pronounced with AHX.

Methohexital Dosage in Electroconvulsive Therapy.

A comparison of methohexital at 0.5 mg/kg versus 1 mg/kg iv as anesthesia for unilateral brief pulse electroconvulsive therapy showed no difference in systolic or diastolic blood pressure before, immediately after, or 5 or 10 min after the seizure. Mean seizure duration was not significantly shorter using the higher methohexital dosage. These results show that methohexital can be safely used in a broad dosage range without undue effects on blood pressure or seizure duration.

Arterial hypertension and multiple cerebral aneurysms in a patient treated with electroconvulsive therapy.

The authors successfully instituted two courses of ECT at a 1-year interval for drug-resistant major depression in a patient with arterial hypertension and intracranial aneurysms. Both ECT courses required arterial and central venous cannulas, but the first course was complicated by an unusual and excessive degree of hypertension, which was not appropriately responsive to high doses of sodium nitroprusside (9 micrograms/kg/minute). Appropriate responsiveness to nitroprusside was established after therapy with timolol. The combination of sodium nitroprusside and timolol proved effective throughout the second course of ECT.

Ionized hypocalcemia after fresh frozen plasma administration to thermally injured children: effects of infusion rate, duration, and treatment with calcium chloride.

A number of cardiac arrests and severe hypotensive episodes have been witnessed associated with the intravenous infusion of fresh frozen plasma (FFP). To clarify the possible role of hypocalcemia, 28 thermally injured anesthetized pediatric patients with massive blood loss were studied to examine the cardiovascular responses (mean arterial pressure [MAP], heart rate, ECG) to 49 infusions of FFP. Rapid, statistically significant reductions in ionized calcium ([Ca2+]) followed each of four rates (1.0, 1.5, 2.0, and 2.5 ml.kg-1.min-1 for 5 minutes) of FFP infusion (P less than 0.0001). The slowest rate resulted in significantly less reduction in [Ca2+] than did the higher infusion rates (P less than 0.002). In five children MAP decreased greater than or equal to 20% below baseline levels, but this was not correlated with rate of FFP administration or decrease in [Ca2+]. The decreases in [Ca2+] and MAP were inversely related to age and unrelated to anesthetic technique. Changes in the Q-oTc interval were not related to [Ca2+]. Adverse cardiovascular responses and reduced [Ca2+] were not significantly different between 5- and 10-minute FFP infusions. Fewer fluctuations in MAP occurred when calcium chloride (CaCl2) was administered; the least fluctuation in [Ca2+] occurred when CaCl2 was administered during the plasma infusion. It is concluded that in thermally injured children 1-17 years old: 1) Rapid infusions of FFP produce sudden but evanescent decreases in [Ca2+]; more rapid infusions result in greater reductions in [Ca2+]. 2) There is no correlation between [Ca2+] and systemic hypotension. 3) Clinically important decreases in MAP occasionally accompany the rapid infusion of FFP.(ABSTRACT TRUNCATED AT 250 WORDS)

Calcium chloride versus calcium gluconate: comparison of ionization and cardiovascular effects in children and dogs.

A randomized prospective study in both children and dogs compared ionization of calcium chloride and calcium gluconate. Five conditioned dogs under halothane anesthesia received calcium chloride (4, 8, 12 mg/kg) and calcium gluconate (14, 28, 42 mg/kg) intravenously. Ten children scheduled for burn wound excision and grafting received both calcium chloride (2.5 mg/kg) and calcium gluconate (7.5 mg/kg) injected through a central venous cannula. Ionized calcium was measured at 0, 0.5, 1, 3, 5, and 10 min in the children, and 0, 0.5, 1, 2, 3, 4, 5, 10, 20, and 45 min in the dogs. The authors conclude that equal elemental calcium doses of calcium gluconate (10%) and calcium chloride (10%) (approximately 3:1), injected over the same period of time, are equivalent in their ability to raise [Ca++] during normocalcemic states in children and dogs; the changes in [Ca++] following calcium administration are short-lived (minutes); rapidity of ionization seems to exclude hepatic metabolism as an important factor in the dissociation of calcium gluconate; and equivalent rises in [Ca++] produced by calcium gluconate or calcium chloride resulted in equivalent cardiovascular effects. The authors feel that either form of calcium salt would be satisfactory if indicated during cardiopulmonary resuscitation or for the treatment of ionized hypocalcemia due to massive blood transfusion.

Effect of preload on ischaemic and non-ischaemic left ventricular regional function.

The response to preload of ischaemic and non-ischaemic regions of the left ventricle was studied in 14 dogs undergoing right heart bypass with mean aortic pressure and heart rate held constant. Regional function was measured by sonomicrometry before and after coronary artery occlusion. In the ischaemic region, as expected, there was paradoxical systolic lengthening (that is, systolic shortening was negative) but as stroke volume was progressively increased end diastolic length increased, whereas end systolic length changed little; thus systolic lengthening decreased (systolic shortening increased). Ischaemic regions that were dyskinetic at low stroke volumes were virtually akinetic at high stroke volumes. Additional studies showed that this response was not attributable to increased regional blood flow at high preloads and occurred over a wide range of heart rates and mean aortic pressures. Plots of systolic shortening against end diastolic length, expressing the regional Frank-Starling relation, were well described by linear regression in both ischaemic and non-ischaemic regions, although a few of these relations were better described by higher order polynomials. The slopes of these relations in the ischaemic region were 0.86(0.05) before and 0.83(0.06) after ligation, reflecting a small effect of preload on end systolic length. The data suggest that when contractility and afterload are constant preload determines the magnitude and in certain instances the sign of systolic shortening. In any ischaemic regions incapable of developing force the positive slope of the Frank-Starling relation is attributable to myocardial passive elastic properties. Paradoxical lengthening does not, however, necessarily indicate the absence of active force development; positive and negative values of systolic shortening describe a continuous spectrum of regional contractility. Thus the effects of preload and contractility on systolic shortening in ischaemic as well as non-ischaemic myocardium require differentiation.