Heparin-binding protein (HBP) in critically ill patients with influenza A(H1N1) infection.

Heparin-binding protein (HBP) is an inducer of vascular endothelial leakage in severe infections. Fluid accumulation into alveoli is a general finding in acute respiratory distress syndrome (ARDS). Severe acute respiratory failure with ARDS is a complication of influenza A(H1N1) infection. Accordingly, we studied the HBP levels in critically ill patients with infection of influenza A(H1N1).Critically ill patients in four intensive care units (ICUs) with polymerase chain reaction (PCR) confirmed infection of influenza A(H1N1) were prospectively evaluated. We collected clinical data and blood samples at ICU admission and on day 2. Twenty-nine patients participated in the study. Compared with normal plasma levels, the HBP concentrations were highly elevated at baseline and at day 2: 98 ng/mL (62-183 ng/mL) and 93 ng/mL (62-271 ng/mL) (p 0.876), respectively. HBP concentrations were correlated with the lowest ratio of partial pressure of oxygen in arterial blood to fraction of inspired oxygen (PF ratio) during the ICU stay (rho = -0.321, p <0.05). In patients with and without invasive mechanical ventilation, the baseline HBP levels were 152 ng/mL (72-237 ng/mL) and 83 ng/mL (58-108 ng/mL) (p 0.088), respectively. The respective values at day 2 were 223 ng/mL (89-415 ng/mL) and 81 ng/mL (55-97 ng/mL) (p <0.05). The patients with septic shock/severe sepsis (compared with those without) did not have statistically significant differences in HBP concentrations at baseline or day 2. HBP concentrations are markedly elevated in all critically ill patients with influenza A(H1N1) infection. The increase in HBP concentrations seems to be associated with more pronounced respiratory dysfunction.

Corticosteroid therapy in intensive care unit patients with PCR-confirmed influenza A(H1N1) infection in Finland.

OBJECTIVE: To evaluate the incidence, treatment, and outcome of influenza A(H1N1) in Finnish intensive care units (ICUs) with special reference to corticosteroid treatment. METHODS: During the H1N1 outbreak in Finland between 11 October and 31 December 2009, we prospectively evaluated all consecutive ICU patients with high suspicion of or confirmed pandemic influenza A(H1N1) infection. We assessed severity of acute disease and daily organ dysfunction. Ventilatory support and other concomitant treatments were evaluated and recorded daily throughout the ICU stay. The primary outcome was hospital mortality. RESULTS: During the 3-month period altogether 132 ICU patients were tested polymerase chain reaction-positive for influenza A(H1N1). Of these patients, 78% needed non-invasive or invasive ventilatory support. The median (interquartile) length of ICU stay was 4 [2-12] days. Hospital mortality was 10 of 132 [8%, 95% confidence interval (CI) 3-12%]. Corticosteroids were administered to 72 (55%) patients, but rescue therapies except prone positioning were infrequently used. Simplified Acute Physiology Score II and Sequential Organ Failure Assessment scores in patients with and without corticosteroid treatment were 31 [24-36] and 6 [2-8] vs. 22 [5-30] and 3 [2-6], respectively. The crude hospital mortality was not different in patients with corticosteroid treatment compared to those without: 8 of 72 (11%, 95% CI 4-19%) vs. 2 of 60 (3%, 95% CI 0-8%) (P = 0.11). CONCLUSIONS: The majority of H1N1 patients in ICUs received ventilatory support. Corticosteroids were administered to more than half of the patients. Despite being more severely ill, patients given corticosteroids had comparable hospital outcome with patients not given corticosteroids.

Serum zinc in critically ill adult patients with acute respiratory failure.

BACKGROUND AND AIMS: Zinc deficiency leads to susceptibility to infections and may affect pulmonary epithelial cell integrity. Low zinc levels have also been associated with a degree of organ failure and decreased survival in critically ill children. Accordingly, the purpose of the study was to assess serum zinc in adult patients with acute respiratory failure, its association with ventilatory support time, intensive care unit (ICU) length of stay (LOS), organ dysfunction and 30-day mortality. METHODS: We included consecutive patients with acute respiratory failure during an eight-week prospective, observational multicentre study (the FINNALI-study). Acute respiratory failure was defined as a need for either non-invasive or invasive positive pressure ventilation for >6 h regardless of the underlying cause or risk factors. After informed consent, a sample for zinc measurement was drawn at 6 h after the start of treatment and analysed from 551 of these patients. RESULTS: Low serum zinc was frequent (95.8%) at the onset acute respiratory failure. The median interquartile range [IQR] was 4.7 [3.0-6.9] µmol/l. The median [IQR] serum zinc levels in non-infectious, sepsis and septic shock patients were 5.0 [3.1-7.1], 5.1 [3.5-7.3] and 3.8 [2.6-5.9] µmol/l, respectively, P<0.01. Baseline zinc levels were not associated with ventilatory support time (P=0.98) or ICU LOS (P=0.053). The area under curve in receiver operating characteristics analysis for serum zinc regarding 30-day mortality was 0.55 (95% CI 0.49-0.60). CONCLUSIONS: Serum zinc on initiation of ventilation had no predictive value for 30-day mortality, ventilatory support time or intensive care unit LOS.

Vasopressor agents after experimental brain death: effects of dopamine and vasopressin on vitality of the small gut.

BACKGROUND: Both congenital and acquired short bowel syndrome frequently leads to the necessity for long-term parenteral nutrition, which in turn may lead to any of several complications or death. Transplantation of the small bowel from brain-dead organ donors has been successfully performed over the last years. However, systemic blood pressure and blood perfusion to the splanchnic area decrease rapidly after brain death, which comprises the vitality of the small bowel. OBJECTIVE: To evaluate the differences between dopamine and low-dose vasopressin on perfusion and vitality of the small bowel after brain death. METHODS: Fifteen pigs were randomized into 3 groups: vasopressin (n = 6), dopamine (n = 6), or control (n = 3). Brain death was induced via stepwise filling of an epidural balloon. When the hypotensive phase was achieved, vasopressin, maximum dose of 0.04 IU/kg/h, or dopamine, maximum dose of 20 µg/kg/min, was administered for 5 hours with the objective of increasing mean arterial blood pressure by 15 mm Hg. RESULTS: Target blood pressure was achieved in the vasopressin group but not the dopamine group. Vasopressin reduced cardiac output, superior mesenteric artery (SMA) blood flow and oxygen delivery, and systemic oxygen delivery and consumption, and increased oxygen extraction. Dopamine increased SMA blood flow, and had no effect on systemic oxygen delivery or consumption. CONCLUSIONS: Vasopressin reversed hypotension but compromised both the systemic and SMA blood flow. Vasopressin was associated with inadequate oxygen delivery, estimated from decreased oxygen delivery and increased oxygen extraction. These adverse effects were not observed with dopamine.

Effects of levosimendan on indocyanine green plasma disappearance rate and the gastric mucosal-arterial pCO2 gradient in abdominal aortic aneurysm surgery.

BACKGROUND: Levosimendan has a dual mechanism of action: it improves myocardial contractility and causes vasodilatation without increasing myocardial oxygen demand. In a laboratory setting, it selectively increases gastric mucosal oxygenation in particular and splanchnic perfusion in general. The aim of our study was to describe the effects of levosimendan on systemic and splanchnic circulation during and after abdominal aortic surgery. METHODS: Twenty abdominal aortic aneurysm surgery patients were randomized to receive either levosimendan (n=10) or placebo (n=10) in a double-blinded manner. Both the mode of anaesthesia and the surgical procedures were performed according to the local guidelines. Automatic gas tonometry was used to measure the gastric mucosal partial pressure of carbon dioxide. Systemic indocyanine green clearance plasma disappearance rate (ICG-PDR) was used to estimate the total splanchnic blood flow. RESULTS: The immediate post-operative recovery was uneventful in the two groups with a comparable, overnight length of stay in the intensive care unit. Cumulative doses of additional vasoactive drugs were comparable between the groups, with a tendency towards a higher cumulative dose of noradrenaline in the levosimendan group. After aortic clamping, the cardiac index was higher [4(3.8-4.7) l/min/m(2) vs. 2.6(2.3-3.6) l/min/m(2); P<0.05] and the gastric mucosal-arterial pCO(2) gradient was lower in levosimendan-treated patients [0.9(0.6-1.2) kPa vs. 1.7(1.2-2.1) kPa; (P<0.05)]. However, the total splanchnic blood flow, estimated by ICG-PDR, was comparable [29(21-29)% vs. 20(19-25)%; NS]. Organ dysfunction scores (sequential organ dysfunction assessment) were similar between the groups on the fifth post-operative day. CONCLUSION: Levosimendan favours gastric perfusion but appears not to have a major effect on total splanchnic perfusion in patients undergoing an elective aortic aneurysm operation.

Community-acquired septic shock: early management and outcome in a nationwide study in Finland.

AIM: To determine how the early treatment guidelines were adopted, and what was the impact of early treatment on mortality in septic shock in Finland. METHODS: This study was a sub-analysis of a prospective observational investigation of severe sepsis and septic shock in Finland (Finnsepsis). All patients with severe sepsis over 4 months in 24 intensive care units were included in the Finnsepsis study. Patients with community-acquired septic shock, admitted directly from the emergency department to the intensive care unit, were included in the sub-study. The following treatment targets were evaluated: measurement of lactate during the first 6 h; analysis of blood culture before antibiotics; commencement of antibiotics within 3 h; attainment of a mean arterial pressure of > or =65 mmHg, central venous pressure of > or =8 mmHg and central venous oxygen saturation of > or =70% or mixed venous oxygen saturation of > or =65% during the first 6 h. RESULTS: Of the 92 patients who fulfilled the inclusion criteria, six reached all treatment targets and 33 reached four or more targets (group > or =4). The hospital mortality of group > or =4 was 24% (8/33), compared with 42% (25/59) for those who reached three or fewer targets (group < or =3) (P= 0.08). The 1-year mortality rates of group > or =4 and group < or =3 were 36% and 59% (P= 0.04), respectively. In logistic regression analysis, a delayed initiation of antimicrobials was associated with an unfavourable outcome (P= 0.04). CONCLUSIONS: Compliance with international guidelines for the early treatment of septic shock was poor in Finnish hospitals. A failure to diagnose early and to start appropriate treatment was reflected in the high mortality. The delayed start of antibiotics was the most important individual variable leading to a high mortality in this nationwide study.

Intensive care of the elderly in Finland.

BACKGROUND: The ageing of the population will increase the demand for health care resources. The aim of this study was to determine how age affects resource consumption and outcome of intensive care in Finland. METHODS: Data on 79,361 admissions to 26 Finnish intensive care units (ICUs) during the years 1998-2004 were analysed. The severity of illness was measured using Simplified Acute Physiology II scores and the intensity of care using Therapeutic Intervention Scoring System scores. RESULTS: The median age was 62 years; 8.9% of patients were aged 80 years or over. The hospital mortality rate was 16.2% in the overall patient population, but 28.4% in patients aged 80 years or over. Old age was an independent risk factor for hospital mortality. The mean intensity of care was at its highest in the age groups 60-69, 70-74 and 75-79 years. It was notably lower for patients aged 80 years or over. If the need for intensive care remains unchanged in each age group, the change in the age distribution of the Finnish population will increase the demand for ICU beds by 19% by the year 2020 and by 25% by the year 2030. CONCLUSION: The hospital mortality rate increases with increasing age. The mean intensity of care is lower for the oldest patients than for patients aged less than 80 years. The ageing of the population will probably cause a remarkable increase in the need for intensive care in the near future.

Propofol causes more hypotension than etomidate in patients with severe aortic stenosis: a double-blind, randomized study comparing propofol and etomidate.

BACKGROUND: Vasodilatation and hypotension are thought to be harmful in patients with severe aortic stenosis. Etomidate is preferred to propofol for anaesthesia induction in haemodynamically unstable patients, but may disturb cortisol synthesis. We assessed the haemodynamic effects of etomidate vs. propofol as anaesthesia induction agents, and the effects of these drugs on cortisol concentrations, in patients with severe aortic stenosis. The main end-point of the study was the incidence of hypotension. METHODS: Sixty-six patients with severe aortic stenosis scheduled for elective aortic valve replacement were enrolled in the study. The patients were randomized to receive either propofol or etomidate for induction of anaesthesia. Haemodynamic parameters, i.e. mean arterial pressure (MAP), pulmonary capillary wedge pressure (PCWP) and cardiac index (CI), were measured. If MAP decreased below 70 mmHg for more than 30 s, phenylephedrine was administered. Serum cortisol concentrations were also measured. RESULTS: MAP decreased in all patients (P < 0.001). MAP decreased to a greater extent in patients receiving propofol than in those receiving etomidate (P = 0.006). Patients receiving propofol needed phenylephedrine more often than those receiving etomidate (20/30 vs. 8/30, P = 0.002). CI and PCWP decreased in both groups (P < 0.001), with no difference between the groups. Patients receiving etomidate had a lower serum cortisol concentration immediately after the operation than those receiving propofol (P < 0.001), but no differences between the groups were observed on the first post-operative morning. CONCLUSION: Propofol is twice as likely as etomidate to evoke hypotension in anaesthesia induction of patients with severe aortic stenosis; however, etomidate transiently decreases post-operative serum cortisol concentrations.

Excess mortality in winter in Finnish intensive care.

BACKGROUND: In the general population, mortality from acute myocardial infarctions, strokes and respiratory causes is increased in winter. The winter climate in Finland is harsh. The aim of this study was to find out whether there are seasonal variations in mortality rates in Finnish intensive care units (ICUs). METHODS: We analysed data on 31,040 patients treated in 18 Finnish ICUs. We measured severity of illness with acute physiology and chronic health evaluation II (APACHE II) scores and intensity of care with therapeutic intervention scoring system (TISS) scores. We assessed mortality rates in different months and seasons and used logistic regression analysis to test the independent effect of various seasons on hospital mortality. We defined 'winter' as the period from December to February, inclusive. RESULTS: The crude hospital mortality rate was 17.9% in winter and 16.4% in non-winter, P = 0.003. Even after adjustment for case mix, winter season was an independent risk factor for increased hospital mortality (adjusted odds ratio 1.13, 95% confidence interval 1.04-1.22, P = 0.005). In particular, the risk of respiratory failure was increased in winter. Crude hospital mortality was increased during the main holiday season in July. However, the severity of illness-adjusted risk of death was not higher in July than in other months. An increase in the mean daily TISS score was an independent predictor of increased hospital mortality. CONCLUSION: Severity of illness-adjusted hospital mortality for Finnish ICU patients is higher in winter than in other seasons.

The proportion of intensive care unit admissions related to alcohol use: a prospective cohort study.

BACKGROUND: Alcohol abuse is a risk factor for serious illnesses, and a history of chronic alcohol abuse adversely affects the outcome of critically ill patients. It is not known what proportion of intensive care unit (ICU) admissions is related to alcohol use. Therefore, we investigated the proportion of emergency admissions related to alcohol. METHODS: A prospective cohort study was conducted in a university hospital ICU. All adult patients (n = 893) who underwent emergency admission to our ICU during a period of 1 year were studied. RESULTS: The admitting physician determined whether there was a relationship between alcohol use and admission. ICU and hospital mortality and ICU length of stay (LOS) were recorded. The Therapeutic Intervention Scoring System (TISS) was used for ICU resource use estimation. There was a relationship between alcohol use and admission in 24% (215/893) of admissions and, in 156/893 admissions (17.5%), this seemed to be definite. ICU LOS was 1.2 days (0.7; 2.3) (median; interquartile range) for alcohol-related and 1.8 days (0.9; 3.6) for other admissions (P < 0.001). Patients with alcohol-related admissions consumed 17.8% of ICU patient-days and 18.7% of all accumulated TISS scores. ICU (8.8 vs. 10.5%, P = 0.603) and hospital (19.1 vs. 20.2%, P = 0.769) mortalities were no different between alcohol-related and other admissions. CONCLUSION: ICU admission is very often related to long-term chronic and/or occasional alcohol use.

Impact of gender on treatment and outcome of ICU patients.

BACKGROUND: Gender modifies immunologic responses caused by severe trauma or critical illness. The aim of this study was to investigate the impact of gender on hospital mortality, length of intensive care unit (ICU) stay, and intensity of care of patients treated in ICUs. METHODS: Data on 24,341 ICU patients were collected from a national database. We measured severity of illness with Acute Physiology and Chronic Health Evaluation II (APACHE II) scores and intensity of care with Therapeutic Intervention Scoring System (TISS) scores. We used logistic regression analysis to test the independent effect of gender on hospital mortality. We compared the lengths of ICU stay and the intensity of care of men and women. RESULTS: Male gender was associated with increased hospital mortality among postoperative ICU patients [adjusted odds ratio 1.33 (95% confidence interval 1.12-1.58, P = 0.001)] but not among medical patients [adjusted odds ratio 1.02 (95% confidence interval 0.92-1.13, P = 0.74)]. Male gender was associated with an increased risk of death particularly in the oldest age group (75 years or older) and among the patients with relatively low APACHE II scores (<16). Mean length of ICU stay was 3.2 days for men and 2.6 days for women (P < 0.001). Male patients comprised 61.7% of the study population but consumed 66.0% of days in intensive care. CONCLUSION: Male gender contributes to poor outcome in postoperative ICU patients. Approximately two-thirds of ICU resources are consumed by male patients.

Lack of effect of ranitidine on gastric luminal pH and mucosal PCO2 during the first day in the ICU.

BACKGROUND: Histamine(2) (H(2))-blocking agents can attenuate intragastric CO(2)-production by reducing gastric acid secretion and preventing the interaction between H(+) and bicarbonate. However, gastric acid production may be impaired in acute circulatory failure due to poor mucosal perfusion, and H(2)-blockade could further impair mucosal perfusion. METHODS: Forty patients with acute circulatory and/or respiratory failure, age 61 +/- 16 years (mean +/- SD), APACHE II score 21 +/- 7, and SOFA score 8 +/- 3, received randomly either ranitidine, 50 mg (R) or placebo (P) every 8 h. Gastric intraluminal pH (gpH; antimony probe with external reference electrode) and mucosal pCO(2) (prCO(2), semicontinuous air-tonometry) were measured during 24 h, and blood gases were taken at 6-h intervals. RESULTS: Gastric intraluminal pH was 4.3 +/- 2.4 in P and 5.1 +/- 1.6 in R (NS). Mean prCO(2) was 6.8 +/- 2.7 kPa in P and 7.4 +/- 2.1 kPa in R, and mucosal-arterial pCO(2) gradient (Delta pCO(2)) was 2.2 +/- 2.9 kPa and 2.4 +/- 2.4 kPa, respectively (NS). Within-patient variabilities of gpH and prCO(2) were not influenced by ranitidine. A posthoc analysis revealed that non-survival in R was associated with a low mucosal pHi after 24 h (P = 0.002). This was explained by a low arterial pH but not by differences in gpH or prCO(2). CONCLUSION: In acute respiratory and circulatory failure, H(2) blockade has an inconsistent impact on gpH and does not reduce variabilities of gpH or prCO(2).

Dobutamine compensates deleterious hemodynamic and metabolic effects of vasopressin in the splanchnic region in endotoxin shock.

BACKGROUND: Vasopressin is a potent vasopressor in septic shock, but it may impair splanchnic perfusion. We compared the effects of vasopressin alone and in combination with dobutamine on systemic and splanchnic circulation and metabolism in porcine endotoxin shock. METHODS: Twelve pigs were randomized to receive either vasopressin (VASO, n = 6) or vasopressin in combination with dobutamine (DOBU, n = 6) during endotoxin shock (E. coli endotoxin infusion). Endotoxin infusion rate was increased to induce hypotension after which vasoactive drugs were started. We aimed to keep systemic mean arterial pressure (MAP) >70 mmHg by vasopressin; the goal of dobutamine infusion was to prevent decrease in cardiac output often associated with vasopressin infusion. Regional blood flows, oxygen delivery and consumption, arterial and regional lactate concentrations were measured. RESULTS: Mean arterial pressure >70 mmHg was achieved in both the VASO and DOBU groups. After the primary decrease of cardiac output by vasopressin, systemic blood flow remained stable in vasopressin-treated animals. However, vasopressin as a monotherapy decreased portal venous blood flow. This was prevented by dobutamine. Vasopressin also induced splanchnic lactate release and arterial hyperlactatemia, which were not observed when dobutamine was combined with vasopressin. CONCLUSION: Dobutamine prevents adverse hemodynamic and metabolic effects of vasopressin in septic shock.

Economic evaluation of the use of PCR assay in diagnosing pulmonary TB in a low-incidence area.

To determine whether polymerase chain reaction (PCR) testing in the initial diagnosis of pulmonary tuberculosis (TB) is cost-effective in a low-prevalence population, an economic evaluation was carried out between the smear and culture (NOPCR) and smear, culture and PCR (+PCR) strategies. A decision tree model based on retrospective laboratory data was developed to assess the strategies of testing patients with suspicion of TB. Direct healthcare costs prior to confirmation of TB or nontuberculous mycobacteria by PCR or culture were included. Effectiveness was measured by the probability of correct treatment and isolation decisions. In the baseline situation NOPCR costs Euro 29.50 less than the +PCR strategy per patient tested. According to sensitivity analyses, reducing PCR test price, shortening test performance time or increasing the proportion of smear-positive patients in the tested population would contribute to cost savings with the +PCR strategy. Routine polymerase chain reaction testing of all specimens from suspected tuberculosis patients in a low-prevalence population was not cost-saving. When the polymerase chain reaction assay was applied only to smear-positive sputum specimens, the smear and culture strategy was clearly dominated by it, i.e. the polymerase chain reaction smear-positive sputum strategy was less costly and more effective in producing correct treatment decisions and isolations.

Results and costs of intensive care in a tertiary university hospital from 1996-2000.

BACKGROUND: Costs of intensive care may be 20% of all hospital costs. Population aging likely increases the demand for intensive care services, while health care has financial limitations. Therefore data about outcome and costs of intensive care are needed. We studied changes in patient characteristics, outcome, intensity of care and costs of intensive care in a tertiary university hospital in Finland. METHODS: We analyzed retrospectively data of patients admitted to the ICU between 1 January 1996 and 31 December 2000 using the patient data management system. Postoperative and ICU patients were analyzed separately. Data included age, Apache II score, cause of intensive care admission, length and intensity of ICU care. ICU, hospital and 6-month mortality were analyzed. Intensity of care was assessed by TISS points and the annual costs of intensive care were evaluated. RESULTS: The number of ICU admissions from 1996-2000 was 11,323. The proportions of ICU and postoperative patients were 39% and 61%, respectively. The mean age of the patients did not change. The mean Apache II score increased over time both in the ICU and postoperative patients. There was no change in crude hospital mortality. Total ICU costs decreased from 8,660,000 euros (in 1997) to 7,480,000 euros (in 2000). In the ICU patients, the costs of hospital survival decreased towards the end of the study period. CONCLUSIONS: We treated more severely ill patients with unchanged outcome but at lower costs towards the end of the study period. Costs of intensive care are not necessarily increasing.

The effect of carbon dioxide, respiratory rate and tidal volume on human heart rate variability.

BACKGROUND: Heart rate variability (HRV) has been used for assessment of depth of anesthesia. Alterations in respiratory rate and tidal volume modulate the sympatovagal neural drive to the heart. The changes in PaCO2 that accompany changes in breathing pattern may, through chemoreceptors in the brainstem, independently influence the autonomic control of the heart and modulate HRV. METHODS: We measured the effects of PaCO2, tidal volume and respiratory rate on HRV during spontaneous and mechanical ventilation in 22 healthy volunteers and in 25 mechanically ventilated anesthetized patients. RESULTS: Adding CO2 to the inspiratory gas increased high frequency (HF) and low frequency (LF) components of HRV in awake volunteers both during spontaneous and mechanical ventilation, while this effect of CO2 was abolished in patients during anesthesia. Increase of tidal volume increased HF component of HRV only in volunteers during spontaneous ventilation. On the other hand, when respiratory rate was reduced, the balance of HF and LF power moved toward LF power in all study groups. Breathing frequency altered HRV independent on PaCO2, tidal volume and the level of consciousness. In contrast, the effect of PaCO2 appeared to be related to normal level of consciousness, suggesting that a cortical modulation of the autonomic nervous activity contributes to the effects of PaCO2 on HRV. CONCLUSIONS: PaCO2, tidal volume and respiratory rate should be controlled when HRV power spectrum is measured in conscious patients or volunteers, while in anesthetized patients small changes in end-tidal CO2 or tidal volume do not modulate HRV if respiratory rate remains unchanged.

Myocardial function and haemodynamics in extensive burn trauma: evaluation by clinical signs, invasive monitoring, echocardiography and cytokine concentrations. A prospective clinical study.

BACKGROUND: The objectives of this study were to (1). describe the haemodynamic profile of patients with extensive burns during the early fluid resuscitation phase, (2). evaluate myocardial performance by invasive monitoring and echocardiography and (3). analyze the relations between serum cytokine (IL-6, IL-8, TNF) and natriuretic peptide (ANP, BNP) concentrations and myocardial function in these patients. METHODS: Prospective, clinical study in a tertiary care burn centre. Invasive haemodynamic measurements including a pulmonary artery catheter, echocardiography, blood samples for cytokine and atriopeptide analyses. The follow-up time was up to 72 h postinjury. RESULTS: According to echocardiography, patients were hypovolaemic despite aggressive (median 7,9 ml kg(-1) h(-1), range 3.3-11.7) fluid resuscitation and adequate urine output (median 0.9 ml kg(-1) h(-1), range 0.46-1.35) during the first day postinjury. There were no consistent findings of hyperlactatemia, metabolic acidosis or low mixed venous oxygen saturations. Daily highest and lowest values of cardiac index and stroke volume index increased and the lowest and highest values of systemic vascular resistance decreased. Cardiac performance (stroke volume index) improved during the study period even though there were no initial signs of myocardial depression in echocardiography. Three patients received a dobutamine infusion based on clinical judgement. There was no consistent association between haemodynamic changes and plasma cytokine concentrations. CONCLUSION: Persisting hypovolaemia is evident in the resuscitation phase of extensive burns despite aggressive fluid therapy and the lack of classic signs of hypoperfusion. Cardiac performance improves during the first days after extensive burn injury without association with plasma cytokine profile.