Biomedical subjects
Publications and source records attributed to C Ferrier.
Indications for laparoscopic surgery of digestive disease are increasing in scope and now include colorectal affections. A technique for subtotal colectomy with ileorectal anastomosis under laparoscopic control is described.
1. Muscle sympathetic nerve activity (MSA) was recorded in the peroneal nerve at the knee by microneurography in ten healthy subjects and determinations were made simultaneously of intra-arterial blood pressure, and whole-body and cardiac noradrenaline spillover to plasma. Measurements were made at rest, during isometric handgrip at 30% of maximum power and during stress induced by forced mental arithmetic. 2. At rest there were significant positive correlations between spontaneous MSA (expressed as number of sympathetic bursts min-1) and both spillover of noradrenaline from the heart and concentration of noradrenaline in coronary sinus venous plasma. 3. Both isometric handgrip and mental arithmetic led to sustained increases of blood pressure, heart rate and MSA. Plasma concentrations of noradrenaline and spillover of noradrenaline (total body and cardiac) increased. In general the effects were more pronounced during handgrip than during stress. 4. When comparing effects during handgrip and stress the ratio between the fractional increases of MSA and cardiac noradrenaline spillover were significantly greater during handgrip. 5. The data suggest (a) that there are proportional interindividual differences in the strength of resting sympathetic activity to heart and skeletal muscle which are determined by a common mechanism and (b) that handgrip and mental stress are associated with differences in balance between sympathetic outflows to heart and skeletal muscle.
In essential hypertension sympathetic nerve firing is commonly increased. A central nervous system origin has been presumed but not tested directly. To estimate cerebral norepinephrine release in essential hypertension, spillover of norepinephrine into the cerebrovascular circulation was measured by isotope dilution, with high internal jugular venous sampling. Norepinephrine was released into the cerebrovascular circulation in both hypertensive patients and healthy volunteers and was present after administration of the ganglion blocker trimethaphan and in patients with sympathetic nervous failure, indicating that brain neurons and not cerebrovascular sympathetic nerves were the probable source. Although differing among hypertensive patients, norepinephrine spillover on average was higher in the hypertensive patients (153 +/- 41 pmol/min) than in healthy subjects (59 +/- 12 pmol/min; p less than 0.05), and was elevated in six of 17 patients, in whom the accompanying whole body norepinephrine spillover rate was higher than in the remaining 11 patients (p less than 0.01). To test for a possible link between brain norepinephrine release and human sympathetic nervous function, the effect of the tricyclic antidepressant desipramine (0.3 mg/kg i.v.) on both brain and whole body norepinephrine spillover was measured in healthy volunteers. Desipramine lowered the cerebrovascular spillover of norepinephrine, its precursor dihydroxyphenylalanine, and its metabolite dihydroxyphenylglycol by 50-80% and produced a mean fall of 35% in whole body norepinephrine spillover. One interpretation of these results is that human sympathetic nerve firing is dependent on norepinephrine release within the brain and that increased cerebral norepinephrine release may possibly be present in some patients with essential hypertension, underlying their higher sympathetic nerve firing rates.
The purpose of the present study was to assess the efficacy and tolerability of diuretic-free antihypertensive therapy with a calcium antagonist and/or an angiotensin converting enzyme (ACE) inhibitor in patients with diabetes mellitus. 54 hypertensive [blood pressure (BP) above 140/90mm Hg] patients with diabetes mellitus type 1 (n = 7) or 2 (n = 47) and normal serum creatinine levels (mean 82 +/- 6 mumol/L) received either verapamil or enalapril after a 2-week washout and a 4-week placebo phase. If BP remained elevated, both agents were combined. Verapamil or enalapril alone normalised diastolic BP (to less than 90mm Hg) in 36 patients; verapamil decreased BP from 159/98 to 147/87mm Hg (n = 19, p < 0.001) and enalapril decreased BP from 166/99 to 146/88mm Hg (n = 17, p < 0.001). In 18 patients who remained hypertensive after 10 weeks of monotherapy, a combination of both drugs decreased BP from 169/104 to 151/90mm Hg (p < 0.001). Overall, 87% of patients achieved a target BP response at 30 weeks. Urinary albumin as related to creatinine excretion (UAE; micrograms albumin:mg creatinine) was on average not significantly changed after verapamil or enalapril treatment, alone or combined. Nevertheless, in patients with initial microalbuminuria, UAE decreased (p < 0.05) during enalapril treatment. Serum potassium, total lipids, high density lipoprotein cholesterol, low density lipoprotein cholesterol, glycosylated haemoglobin, serum C peptide and fructosamine levels were not significantly modified by treatment. Subjective tolerability of the drugs was also generally good. Thus, in hypertensive patients with diabetes, a diuretic-free therapy based on the calcium antagonist verapamil or the ACE inhibitor enalapril, alone or combined, can effectively decrease BP without adversely affecting carbohydrate and lipid metabolism.
Epidemiological evidence suggests that dietary calcium and 1,25-dihydroxyvitamin D3 (1,25-(OH)2D3) might have a protective effect against colorectal cancers. Since the presence of receptors is required for steroid action, specific 1,25-(OH)2D3 receptors (RD3) were investigated in biopsies taken at different levels of the colon. The study involved 90 biopsies from patients operated on for colorectal adenocarcinoma. They were paired biopsies from adenocarcinoma tissue and adjacent normal mucosa. In addition, 26 normal intestinal mucosa biopsies from patients without cancer were examined. RD3 receptors were assayed in tissue extract by the dextran-coated charcoal technique and also characterized by sucrose density gradient sedimentation. Scatchard analysis showed a single class of specific high affinity-low capacity sites binding for 1,25-(OH)2D3. The incidence of RD3 was 86 per cent in normal mucosa (n = 77) and lower in carcinoma (n = 34), for which the incidence decreased significantly (P less than 0.001) from right colon (58 per cent) to left colon (37 per cent) and rectum (19 per cent). These data suggest that the normal colon is a potential target organ for 1,25-(OH)2D3 which might modulate calcium transport in the colon. Loss of receptivity to 1,25-(OH)2D3 is associated with malignant transformation.
1. Plasma concentrations of [3H]dihydroxyphenylglycol, the intraneuronal metabolite of noradrenaline, were examined during intravenous infusion of [3H]noradrenaline in 43 subjects, to assess the nature of its formation. Noradrenaline re-uptake by sympathetic nerves was estimated in 11 subjects from the effects of neuronal uptake blockade with desipramine on noradrenaline clearance and plasma concentrations of [3H]dihydroxyphenylglycol and endogenous dihydroxyphenylglycol. In seven subjects noradrenaline re-uptake and spillover into plasma were examined before and during mental arithmetic or handgrip exercise. 2. During infusion of [3H]noradrenaline, plasma [3H]dihydroxyphenylglycol increased progressively, indicating its formation from previously stored [3H]noradrenaline leaking from vesicles as well as from [3H]noradrenaline metabolism immediately after removal into sympathetic nerves. Thus, to estimate noradrenaline re-uptake, the amount of [3H]dihydroxyphenylglycol derived from [3H]noradrenaline metabolized immediately after removal into the sympathetic axoplasm must be isolated from that derived from [3H]noradrenaline sequestered into vesicles. 3. At rest in the supine position the rate of noradrenaline re-uptake was 474 +/- 122 pmol min-1 kg-1, 9.5-fold higher than the rate of spillover of noradrenaline into plasma (49.6 +/- 6.4 pmol min-1 kg-1). Noradrenaline re-uptake and spillover into plasma were both increased during mental arithmetic and isometric handgrip exercise.
Various beta-blockers possessing similar antihypertensive potency have been found to differ widely with regard to their influence on blood pressure-regulating factors such as cardiac output and plasma levels of renin or norepinephrine. Recently, beta-blocker-induced stimulation of circulating atrial natriuretic factor (ANF) was reported. Blood pressure is determined not only by levels of vasoconstrictive factors but also by tissue reactivity. To investigate these aspects, we assessed the cardiovascular responsiveness to norepinephrine and angiotensin II, plasma levels of catecholamines, angiotensin II, ANF and aldosterone and the body sodium-blood volume state of 15 patients with essential hypertension (mean age +/- s.e.m., 42 +/- 3 years) and 12 normal control subjects (41 +/- 5 years), first on placebo and then after 4 weeks of intervention with carteolol, a non-selective beta-adrenergic antagonist with intrinsic sympathomimetic activity. Compared with placebo, carteolol decreased resting plasma norepinephrine in both groups while plasma norepinephrine-blood pressure response curves were shifted to the left, their slopes increased and norepinephrine pressor doses decreased (P less than 0.05 to less than 0.001). Chronotropic responses to isoproterenol were abolished but negative chronotropic responses to a norepinephrine-induced 20 mmHg rise in diastolic blood pressure were unaltered. Plasma norepinephrine clearance in the supine position was slightly decreased in hypertensive and unchanged in normal subjects. Supine and upright blood pressure was lowered (P less than 0.05 to 0.001) in the hypertensive while upright systolic blood pressure only decreased in the normal group.(ABSTRACT TRUNCATED AT 250 WORDS)
The pathogenesis of essential hypertension may possibly involve a deficiency in, or a decreased response to, endogenous vasodilator and natriuretic factor(s). Searching for hereditary or familial defects, it is plausible to evaluate blood pressure (BP) regulating factors in (yet) normotensive offspring of hypertensive parents (OHyp), some of whom are in fact in a stage of prehypertension. Studies by our group demonstrated that compared with healthy offspring of normotensive parents, OHyp have plasma atrial natriuretic (ANF) factor levels that are unaltered on a low salt intake but often fail to increase normally in response to a high salt intake. Plasma levels of cyclic GMP, the presumed second messenger of ANF, also may tend to be decreased in certain OHyp. On the other hand, renal excretory responses of cyclic GMP and electrolytes to ANF infused in "physiological" dose were unchanged in some OHyp tested so far. In borderline to moderate, uncomplicated essential hypertension, plasma ANF levels are often "normal." This may be inappropriately low relative to the existing BP, although the relationship of circulating ANF to atrial pressures in essential hypertension remains to be clarified. A conversion to higher plasma ANF values may occur with cardiac complications such as left ventricular hypertrophy, enlargement, dysfunction, or overt heart failure. Acute or short-term elevation of circulating ANF within the physiological and pathophysiological range by ANF infusion produces an exaggerated natriuresis and lowers BP in essential hypertensive patients. We postulate a syndrome of ANF deficiency, characterized by an impaired response of circulating ANF to high salt intake and by low cyclic GMP levels in certain yet normotensive offspring of essential hypertensive parents and by inappropriately "normal" plasma ANF in some patients with uncomplicated essential hypertension. At the stage of prehypertension, a disturbance in the ANF - cyclic GMP pathway may be expressed primarily at the circulatory rather than at the renal level. Hypertension-prone humans also tend to have an exaggerated vascular reactivity to norepinephrine. Whether the two disturbances may be interrelated is presently unknown. Both defects may potentially predispose to the development of essential hypertension. Relative ANF deficiency, an enhanced natriuretic response to ANF, and a sustained antihypertensive effect of infused ANF may represent a rational basis for treatment of essential hypertension with agents that activate the ANF system.
Radiotracer measures of norepinephrine overflow to plasma are well suited for studying both human sympathetic nervous system responses to mental stress and sympathetic nervous pathophysiology in human hypertension. With an experimental laboratory stressor (cognitive challenge), we noted a preferential activation of the cardiac sympathetic outflow; however, in fainting reactions ("vasovagal syncope"), which occur infrequently during the course of central venous catheter placement under local anesthesia, the converse was seen--an almost total withdrawal of cardiac sympathetic activity. In primary human hypertension (particularly in younger patients), a differentiated activation of the sympathetic outflow to the heart and kidneys is present, based on measurements of norepinephrine spillover to plasma. It is uncertain whether this is attributable to behavioral factors and represents a component of the defense reaction. We previously reported overflow of norepinephrine into the cerebrovascular circulation (with high internal jugular venous sampling) in humans. Because this is resistant to ganglion blockade, brain neurons--not the cerebrovascular sympathetics--are the presumed source. In a preliminary study, we found higher rates of norepinephrine spillover into the cerebrovascular circulation in patients with essential hypertension than in healthy subjects, suggesting that an underlying increase in central nervous system norepinephrine turnover may be the basis for the increased sympathetic outflow.
OBJECTIVE: To assess the efficacy and tolerance of a diuretic-free antihypertensive therapy with a Ca2+ antagonist and an angiotensin-converting enzyme (ACE) inhibitor in patients with non-insulin-dependent diabetes mellitus (NIDDM). RESEARCH DESIGN AND METHODS: After a 2-wk washout and a 4-wk placebo phase, 47 hypertensive patients with NIDDM randomly received verapamil or enalapril alone and, if blood pressure remained elevated, both agents combined over 30 wk. RESULTS: Verapamil or enalapril alone normalized blood pressure to less than 90 mmHg diastolic in 30 patients; verapamil decreased mean +/- SE blood pressure from 159/98 +/- 3/1 to 146/87 +/- 3/2 mmHg (n = 18, P less than 0.001) and enalapril from 166/99 +/- 5/2 to 146/86 +/- 3/1 mmHg (n = 12, P less than 0.001). In 17 patients who were still hypertensive after 10 wk of monotherapy, combination of both drugs decreased blood pressure from 170/104 +/- 4/2 to 152/90 +/- 4/2 mmHg (P less than 0.001). Fasting plasma glucose, glycosylated hemoglobin, serum fructosamine, total lipids, high-density and low-density lipoprotein cholesterol, apolipoproteins A-I and B, creatinine, and urinary albumin-creatinine ratio were not significantly modified. CONCLUSIONS: In hypertensive patients with NIDDM, a diuretic-free therapy based on the Ca2+ antagonist verapamil and/or the ACE inhibitor enalapril can effectively decrease blood pressure without adversely affecting carbohydrate and lipid metabolism.
Plasma concentrations of dihydroxyphenylalanine (DOPA) were measured before and during isometric handgrip exercise or mental stress and after coffee drinking or intravenous infusion of desipramine to examine the influence of sympathetic nervous activity on DOPA formation. Sympathetic activity was assessed by the spillover of norepinephrine into plasma. Turnover of norepinephrine was assessed by the plasma concentration of its intraneuronal metabolite, dihydroxyphenylglycol (DHPG). In normal subjects the resting plasma concentration of DOPA was 6.05 +/- 0.16 nmol/L (n = 42). Plasma DOPA level was increased by stimulation of the sympathetic nervous system; handgrip exercise caused a 0.49 +/- 0.07 nmol/L increase (n = 15), mental stress a 0.25 +/- 0.10 nmol/L increase (n = 34), and coffee drinking a 0.85 +/- 0.19 nmol/L increase (n = 9). Desipramine decreased plasma DOPA level by 0.25 +/- 0.06 nmol/L (n = 23). The small but consistent changes in plasma DOPA level during manipulations of sympathetic activity were positively correlated with changes in norepinephrine spillover (r = 0.55, n = 81) and plasma DHPG level (r = 0.66, n = 81). Percentage increases in plasma DOPA level during sympathetic activation were similar to those in plasma DHPG but were a sixth of the percentage increases in norepinephrine spillover. The similar increases in plasma DOPA and DHPG levels indicated that production of DOPA was related to the turnover of norepinephrine in sympathetic nerves. The smaller percentage increases in plasma DOPA (smaller than those in norepinephrine spillover) were consistent with the partial contribution of exocytotic neurotransmitter release to the turnover of norepinephrine in sympathetic nerves.
To evaluate the hypothesis of an atrial natriuretic factor (ANF) deficiency in hypertension-prone humans, we investigated plasma ANF and other variables in 116 white offspring of normotensive parents (ONorm) or essential hypertensive parents (OHyp). Ten ONorm and 10 OHyp, all men matched for age and body habitus, were studied after 4 days of low (70 mmol/day) and high (350 mmol/day) dietary sodium intake. After mild sodium restriction, plasma ANF did not differ between ONorm and OHyp (9.7 +/- 0.7 vs. 9.0 +/- 1.3 fmol/L). On high sodium intake, plasma ANF increased in ONorm, but not in OHyp (to 18.3 +/- 1.7 vs. 11.7 +/- 1.7 fmol/L; P less than 0.001). On the other hand, acute responses of plasma immunoreactive ANF (irANF) to saline loading or a norepinephrine-induced rise in blood pressure did not differ significantly between 8 ONorm and 8 OHyp. Fifty-one additional ONorm and 45 OHyp were evaluated during liberal sodium intake. Groups were further subdivided according to whether 24-h urinary sodium excretion was 91 mmol/m2 or less (modest salt intake) or more than 91 mmol/m2 (high salt intake). Twenty-four-hour urinary sodium was similar in the 26 ONorm and 21 OHyp on a modest salt intake (121 +/- 6 vs. 116 +/- 9 mmol) and in the 25 ONorm and the 24 OHyp on a high salt intake (226 +/- 10 vs. 221 +/- 9 mmol). However, compared with ONorm, plasma irANF in OHyp was slightly lower on modest sodium intake (7.7 +/- 0.7 vs. 5.3 +/- 0.7 fmol/L; P less than 0.05) and markedly reduced on high sodium intake (15.0 +/- 1.3 vs. 8.0 +/- 1.3 fmol/L; P less than 0.001). Moreover, the slope of the relationship between plasma irANF and 24-h urinary sodium was flatter in OHyp than in ONorm (z test = 2.4). We postulate a new endocrine syndrome characterized by a relative plasma ANF deficiency during high sodium intake in some hypertension-prone humans. This functional defect becomes apparent during chronic, rather than acute, stimulation of ANF release. It occurs as a familial disturbance and may potentially predispose to the development of hypertension.
Buprenorphine, (Temgesic), a N-cyclopropylmethyl derivative of oripavine, is both an agonist and antagonist of morphine. Its effect on the motility of the sphincter of Oddi (SO) in humans have been investigated by endoscopic manometry (EM). Buprenorphine leads to a significant decrease in the amplitude of SO contraction waves without altering other parameters. Thus, it has no morphine-like effect on SO motility, but it does act like a partial antagonist of morphine in reducing the amplitude of SO contraction waves.
The heart is the major source of atrial natriuretic peptides (ANP). A propeptide is stored in atrial myocytes. In normal humans, atrial distension secondary to volume overload and/or increased atrial pressures are thought to stimulate the secretion of biologically active alpha-ANP (ANF-[99-126], 28 amino residues) into the circulation. Plasma immunoreactive ANP (irANP) rises in response to acute sodium-volume loading, the central shift of volume produced by lying down or by immersion, acute increases in blood pressure (BP), dynamic exercise, or the administration of glucocorticoids or mineralocorticoids. Plasma irANP also rises with aging. Synthetic alpha-ANP infused acutely i.v. can lower BP, reduce plasma volume by an extravascular shift, cause baroreflex-mediated sympathetic activation, directly inhibit adrenal steroidogenesis and lower plasma aldosterone and cortisol, directly inhibit renal renin release, elevate plasma insulin; diuresis, free water clearance and natriuresis increase already in response to low alpha-ANP doses that raise plasma irANP within the physiological-pathological range. It follows that in addition to direct influences on cardiovascular and renal function, the ANP system may comprise a cardio-adrenal feedback mechanism and perhaps also modulate insulin and the release of ADH. The major although yet unproven physiological role of the ANP system may be the protection of the heart against volume and/or pressure overload. The pathophysiological, diagnostic and therapeutic aspects of elevated plasma irANP values, ANP measurements, or administration of synthetic ANP, respectively, in various diseases are currently under intense study and of great potential interest.
UNLABELLED: To investigate the pathogenetic constellation and its modification by calcium channel blockade in hypertension associated with chronic nonoliguric renal failure, blood pressure (BP), various pressor factors or correlates, cardiovascular responsiveness, and plasma atrial natriuretic peptide (ANP) were assessed in 15 hypertensive patients (serum creatinine 160-715 mumol/l) before and after 6 weeks of intervention with the agent nitrendipine. On placebo, these patients had a lower plasma angiotensin II (AngII) clearance and higher values of supine plasma AngII, aldosterone, norepinephrine (NE), and heart rate than healthy humans. Acute responses of BP to AngII and of heart rate to isoproterenol were blunted in the patients (p less than 0.05-0.001). Plasma ANP was elevated, correlated positively with systolic BP, and rose in response to NE pressor infusion (p less than 0.05-0.001). Exchangeable sodium and blood volume did not differ significantly from normal values. Nitrendipine reduced the cardiovascular responses to AngII, NE, and isoproterenol and lowered supine BP from 173/102 +/- 5/2 to 146/81 +/- 3/3 mm Hg and upright BP from 170/105 +/- 5/2 to 145/86 +/- 4/3 mm Hg (p less than 0.05-0.001); except for slightly increased plasma AngII, the levels of other endocrine variables, exchangeable sodium, blood volume, and creatinine clearance were not significantly modified. CONCLUSIONS: Hypertension accompanying chronic nonoliguric renal impairment seems to be strongly AngII and probably also NE dependent. Circulating ANP levels are high in this setting. Calcium channel blockade with nitrendipine effectively reduces cardiovascular AngII and NE dependence and BP.
Plasma immunoreactive atrial natriuretic factor (irANF) levels and the effects of alpha-human ANF (alpha-hANF) infusion were investigated in 7 patients with liver cirrhosis and ascites. Under basal conditions, supine blood pressure (BP) averaged 136/76 +/- 9/4 mm Hg (mean +/- SEM). Plasma irANF concentrations (124 +/- 33 pg/ml) were higher (p less than 0.01) than those in age-matched normal subjects (47 +/- 5 pg/ml). Plasma renin activity (PRA 5.9 +/- 2.2 ng/ml/h), aldosterone (18 +/- 7 ng/dl) and norepinephrine (NE, 66 +/- 5 ng/dl) levels were also elevated compared to the age-related normal range. Alpha-hANF infusion for 60 min at 0.036 micrograms/kg/min decreased the mean BP (-14%; p less than 0.05), increased PRA (+179%; p less than 0.05) and plasma NE (+24%; p less than 0.05). Glomerular filtration rate (GFR), effective renal plasma flow (ERPF), diuresis and natriuresis were not modified. A subsequent 60-min infusion of alpha-hANF at 0.067 micrograms/kg/min produced a marked fall in mean BP (-26%; p less than 0.001), hemoconcentration (hematocrit +6%; p less than 0.001) despite stable body fluid balance and a further increase in PRA (+350%, p less than 0.005). GFR and ERPF were severely reduced (-55 and -56%, respectively; p less than 0.001), while diuresis and natriuresis were not modified. Plasma aldosterone was unaltered during, but rose (+72%; p less than 0.01) after the cessation of alpha-hANF infusion. Variations in natriuresis during alpha-hANF infusion correlated positively with BP (r = 0.47; p less than 0.01), ERPF (r = 0.53; p less than 0.01) or GFR (r = 0.51; p less than 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)
The effects of cirrhosis on the elimination kinetics and plasma protein binding of sufentanil were evaluated in 12 anesthetized patients with uncomplicated cirrhosis and these findings were compared with data from age-matched control anesthetized patients with normal hepatic and renal function. Sufentanil 3 micrograms/kg was given intravenously as a bolus injection and venous plasma concentrations were measured at intervals up to 10 hrs. The average (+/- SD) elimination half life was 3.5 +/- 0.9 hrs in controls and did not differ in cirrhotics: 4.1 +/- 0.6 hrs. The plasma clearance did not differ between the two groups: 11.3 +/- 2.5 ml.min-1.kg-1 in controls and 10.8 +/- 4.6 ml.min-1.kg-1 in cirrhotic patients. The sufentanil free fraction was also similar in controls (8.3 +/- 1.5%) and in cirrhotic patients (9.6 +/- 1.8%). These data suggest that sufentanil in a single dose should have a similar duration of action in patients with uncomplicated cirrhosis and in normal patients.
Most anaesthetic agents cause cardiac depression possibly hazardous in the elderly, especially in presence of a poor cardiac reserve. Ninety-three patients undergoing non cardiac surgery lasting more than 90 min. were entered in a double-blind multicentre randomized trial. They were 65 year old or more and unaffected by evolutive angina pectoris. After insertion of a Swan Ganz catheter and an arterial cannula, anaesthesia was induced by thiopentone, fentanyl and 02/nitrous oxide (50%). Forty-five patients were infused dobutamine 7 micrograms.kg-1.min-1 (group D) from 10 min. after induction till the completion of surgery. Forty-eight patients received a placebo (group P). Haemodynamic parameters were recorded throughout anaesthesia and at its emergence. After induction, heart rate, pulmonary capillary wedge pressure and mean pulmonary artery pressure did not change significantly; mean arterial pressure, cardiac index, stroke index and left ventricular stroke work index decreased by 21, 33, 28 and 42% respectively (p less than 0.001); systemic and pulmonary arterial resistances increased by 12 and 37% respectively (p less than 0.001). In group P, these changes persisted throughout the procedure but, 30 min after extubation, cardiac index returned to control levels due to a 25% increase in heart rate; in this group 4 patients presented with both perioperative low cardiac output and persistent postoperative confusion. With dobutamine, haemodynamic parameters returned to preoperative values and heart rate increased by 12 b.min-1. More arrhythmias and hypertensive episodes but less hypotensions occurred in group D. Substantial haemodynamic changes occur during anaesthesia and surgery in elderly patients. Dobutamine corrects the peroperative decrease in cardiac output and blood pressure, and might prevent postoperative neurological disorders.