Extravascular lung water in patients with septic shock during a fluid regimen guided by cardiac index

The Netherlands Journal of Medicine 2000;57:82–93

Original article

Extravascular lung water in patients with septic shock during a fluid regimen guided by cardiac index Alexander J.G.H. Bindels *,1 , Johannes G. van der Hoeven 2 , Arend E. Meinders Leiden University Medical Center, Department of General Internal Medicine, Medical Intensive Care Unit, P.O. Box 9600, 2300 RC Leiden, The Netherlands Received 25 April 2000; accepted 6 June 2000

Abstract Background: To elucidate whether patients with a septic shock develop pulmonary edema in a treatment protocol in which volume loading is guided by its effect on the cardiac output, rather than by preset values of pulmonary artery wedge pressure (PAWP). Methods: 15 consecutive patients with the diagnosis of septic shock were studied in a prospective observational study. Cardiac output, PAWP and extravascular lung water index (EVLWI) were determined at regular intervals during the first 24 h of treatment. Fluid challenges were given if MAP was , 80 mmHg and / or CI was , 4.5 l / min / m 2 , and PAWP was , 16 mmHg. Further fluid challenges were only given if the preceding fluid challenge resulted in an increase in CI of more than 10% and PAWP was still , 16 mmHg. Results: EVLWI was slightly above normal (10.461.2 ml / kg) and did not change during the treatment protocol. One third of the patients had an initial PAWP . 16 mmHg. In these patients, EVLWI was significantly higher than in patients with an initial PAWP , 16 mmHg (14.161.1 ml / kg versus 10.060.9 ml / kg, P 5 0.026). No significant correlation was found between PAWP and EVLWI. Conclusion: In this study, patients with septic shock did not develop pulmonary edema during the first 24 h of treatment, when their fluid regimen was guided by the effects on cardiac output.  2000 Elsevier Science B.V. All rights reserved. Keywords: Fluid challenge; Intrathoracic blood volume; Pulmonary artery wedge pressure; Pulmonary edema; Septic shock; Thermal-dye dilution

Introduction Fluid resuscitation in sepsis is one of the daily returning challenges for intensivists. The goal of

*Corresponding author. Tel.: 131-40-2399-111; fax: 131-402397-229. E-mail address: [email protected] (A.J.G.H. Bindels). 1 Current address: Catharina Hospital Eindhoven, Department of Intensive Care, Eindhoven, The Netherlands. 2 Current address: Bosch Medical Center, Department of Intensive Care, ’s-Hertogenbosch, The Netherlands.

fluid resuscitation is to find the optimal intravascular volume, resulting in an adequate tissue oxygenation, while at the same time fluid overload is prevented. Several studies have shown an association between a positive fluid balance or weight gain and a poor outcome [1–4]. Unfortunately, up to now, it is unclear which variables should be monitored to be informed about a patient’s volume status. In the initial stage of fluid resuscitation it is reasonable to use clinical parameters, such as blood pressure, heart rate, urine output, respiratory rate and mental status as indicators of adequate intravascular volume resto-

0300-2977 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0300-2977( 00 )00056-5

A. J.G.H. Bindels et al. / Extravascular lung water

ration. Many patients, however, do not respond to the initial fluid challenges or even deteriorate, or have underlying diseases limiting their physiologic reserves. In these patients, invasive monitoring to guide further fluid resuscitation is warranted, usually with a pulmonary artery catheter. A pre-set value of the pulmonary artery wedge pressure (PAWP), e.g. $ 18 mmHg, is often used as indicator of adequate fluid replacement. It has been shown, however, that PAWP does not accurately reflect the effective circulating volume [5]. The primary goal of fluid resuscitation is to restore adequate tissue oxygenation, which is largely determined by the cardiac output. Therefore, it seems a more physiological approach to stop volume loading at the point where the cardiac output no longer increases, rather than to strive for a pre-set goal of PAWP. Such an approach could possibly limit excessive fluid resuscitation with subsequent accumulation of edema in the lungs and other organs. Two studies showed that with a strategy of fluid resuscitation guided by PAWP, patients developed excessive pulmonary edema, were longer on mechanical ventilation and had a longer length of stay in the intensive care unit, than patients in whom fluid resuscitation was guided by the amount of extravascular lung water (EVLW) [6,7]. In our intensive care unit we use a treatment protocol in septic shock in which volume loading is guided by the effects on the cardiac output. PAWP is only used with an upper limit of 16 mmHg. If PAWP exceeds this value, no fluid challenges are allowed (for further details see Methods). In a prospective observational study design, we studied the amount of EVLW in patients with septic shock, who were treated according to our protocol. We hypothesised that with our approach the amount of EVLW would be limited. Furthermore, we were interested whether the effects of a fluid challenge could be predicted by the intrathoracic blood volume (ITBV), which has been claimed to be a better reflection of a patient’s intravascular volume than PAWP [5,8].

Methods Patients From January 1995 until June 1996, 15 patients with the diagnosis septic shock were enrolled in the

83

study protocol. Sepsis was defined as a proven site of infection with at least two of the following clinical manifestations: body temperature . 388C or , 368C, heart rate . 90 beats / min, tachypnea, WBC count . 12 000 cells / mm 3 or , 4000 cells / mm 3 , or the presence of . 10% immature neutrophils (‘bands’). Shock was defined as a systolic blood pressure , 90 mmHg, or a reduction of the systolic blood pressure . 40 mmHg from its baseline. Exclusion criteria for the present study were a known allergy for iodine or known significant stenoses or prostheses in the tract of the femoral arteries. Informed consent was given by each patient or his / her next of kin. The study protocol was approved by the Local Ethics Committee.

Treatment protocol After the patients had been enrolled in the study, a normal response pulmonary artery catheter was placed as soon as possible. Before and during placement of the pulmonary artery catheter hypotension was treated, preferably with fluid therapy. The treating physician was allowed to use inotropes or vasopressors in this stage, if the systolic blood pressure was extremely low ( , 70 mmHg) or clinical evidence for pulmonary edema existed. After the pulmonary artery catheter was properly placed, checked by adequate pressure curves on the monitor and a bedside chest X-ray, treatment followed a stepwise algorithm that is outlined in Fig. 1. This algorithm is based on general recommendations [9,10], and was already implemented in the standard treatment of septic shock in our intensive care unit before the start of the current study protocol. The two main goals of this protocol were to achieve a mean arterial pressure (MAP) . 80 mmHg and a cardiac index (CI) $ 4.5 l / min / m 2 . If these criteria were not met and PAWP was , 16 mmHg, a fluid challenge of 500 ml of colloids was given. After this fluid challenge CI and PAWP were measured. If CI increased by 10% or more and PAWP was still , 16 mmHg, another fluid challenge was given. If CI increased , 10%, or PAWP was . 16 mmHg and CI was still , 4.5 l / min / m 2 , dobutamine was given up to 20 mg / kg / min. If CI was . 4.5 l / min / m 2 and MAP was still , 80 mmHg, norepinephrine was

The Netherlands Journal of Medicine 2000;57:82 – 93

84

A. J.G.H. Bindels et al. / Extravascular lung water

Fig. 1. Algorithm used in this study for the treatment of septic shock.

given. If CI was still , 4.5 l / min / m 2 with 20 mg / kg / min of dobutamine and MAP was . 80 mmHg, nitroprusside could be added. Whenever the primary goals (CI $ 4.5 l / min / m 2 and MAP . 80 mmHg) were achieved, further interventions were omitted. Whenever hematocrit was , 0.3, a transfusion of two units of packed red cells was allowed. Measurements Besides a pulmonary artery catheter (7.5-F Swan Ganz-catheter, Model VS1721, Ohmeda, Swindon, UK), a 4-F fiberoptic catheter (Pulsiocath PV2024, Pulsion, Munich, Germany) was placed in the descending aorta through a 6-F introducer sheath (Model 616150A, Ohmeda) in one of the femoral arteries. Mean arterial pressure (MAP) was recorded

via the side port of the introducer sheath. The pulmonary artery catheter was used for measurements of the central venous pressure (CVP), the mean pulmonary artery pressure (MPAP), and the pulmonary artery wedge pressure (PAWP), with the midchest level as zero reference. Thermodilution cardiac output (CO) was measured with injection of 10 ml ice-cold saline at random during the respiratory cycle. The mean value of three consecutive measurements was used for analysis. the heart rate was recorded continuously with one of the standard leads of the electrocardiogram. The CI was calculated according to the standard formula. Measurements of extravascular lung water index (EVLWI) and intrathoracic blood volume index (ITBVI) were obtained with the thermal-dye dilution technique (Cold Z-021 system, Pulsion). Measure-

The Netherlands Journal of Medicine 2000;57:82 – 93

A. J.G.H. Bindels et al. / Extravascular lung water

ments were made by injection of 10 ml ice-cold indocyanin green solution (2 mg / ml). The mean value of two measurements was used for analysis. For details concerning the thermal-dye dilution technique we refer to the literature [11,12]. Briefly, the method uses two indicators, i.e. ice-cold water and indocyanin green (ICG). Cold distributes to both intra- and extravascular volume, whereas ICG remains intravascular. Both indicators are injected into the right atrium, and concentration changes in time are recorded in the descending aorta. Thus, dilution curves result for both indicators. From the thermodilution curve CI is determined. From each indicator’s dilution curve a mean transit time (MTT) can be derived. MTT is composed of the appearance time ATi, which is the time until the first indicator particle has arrived at the point of detection, and the mean time difference between the occurrence of the first particle and all the following particles. The product of CO and MTT is the volume between the site of injection and the site of detection. The following volumes can be calculated Intrathoracic thermal volume (ITTV) (ml) 5 CO 3 MTT aorta (thermal) ITBV (ml) 5 CO 3 MTT aorta (ICG) EVLW (ml) 5 ITTV2 ITBV Indexes are calculated by dividing the absolute value by body surface area (ITBVI) and body weight (EVLWI). Together with each hemodynamic measurement arterial and mixed venous blood samples were taken for determination of hemoglobin (Medfield, MA, USA) and blood gas analysis (Ciba-Cormig 288 blood gas analyser). Oxygen delivery (DO 2 ) was calculated according to the standard formula. The first measurement was carried out immediately after placement of the catheters. The second measurement was carried out at the time of optimal intravascular volume, defined as ,10% increase of CI after a fluid challenge or PAWP .16 mmHg. In patients in whom the recorded values of the first measurement allowed no further fluid challenge (i.e. PAWP.16 mmHg), measurement 1 was considered to be the measurement with optimal intravascular

85

volume, and this measurement was then labelled as measurement 2. Measurements 3, 4, 5, 6 and 7 were carried out, respectively 3, 6, 12, 18, and 24 h after measurement 1. Statistical analysis Data are presented as mean6SEM unless noted otherwise. Data were analysed with analysis of variance for repeated measurements and a t-test for paired samples with Bonferroni correction where appropriate. Comparisons between groups of patients were made by means of an unpaired t-test or with the Kruskal–Wallis test. P values less than 0.05 were considered to be statistically significant.

Results Patient data are shown in Table 1. In ten out of fifteen patients (67%) blood cultures showed pathogens. Six patients (40%) survived and were discharged from hospital. Nonsurvivors all developed failure of one or more organ systems. Two elderly patients (patients number 3 and 15) had a myocardial infarction during the septic episode. Both patients eventually died of refractory cardiogenic shock. Patient 11 survived the septic episode but had fatal postanoxic encephalopathy after cardiac arrest. Five patients had a PAWP .16 mmHg by the time of the first measurement. They did not receive fluid therapy at that point. In these patients measurement 1 was regarded as the point of optimal intravascular volume. These patients however, did receive colloids before the start of the protocol with a mean volume 2000 ml of colloids (S.D. 547 ml). In the patients with a PAWP,16 mmHg a maximum of 580 ml of colloids was infused between measurement 1 and measurement 2. Also in this group, the main part of the volume infusion (mean 1466 ml, S.D. 843 ml) was given before the start of the protocol. The amount of infused colloids at the point of optimal volume status (i.e. ,10% increase in CI after fluid challenge or PAWP .16 mmHg) did not differ between patients with an initial PAWP.16 mmHg and patients with an initial PAWP,16 mmHg (20006547 versus 19746840 ml, not significant). Fig. 2 shows that the treatment protocol induced a

The Netherlands Journal of Medicine 2000;57:82 – 93

A. J.G.H. Bindels et al. / Extravascular lung water

86 Table 1 Clinical and procedural patient data a Pat. no.

Diagnosis with underlying disease

Blood culture

Sex

Age (years)

Survival

Organ failure

Volume infused before measurement 1 (ml)

Volume infused between measurement 1 and 2 (ml)

Inotropes (mg/kg/min) measurement 1

1

Urosepsis with hepatic cirrhosis

E. coli

M

65

NS

K, DIC

2000

–

Norepinephrine 0.4

2

Pneumococcal sepsis after splenectomy

Streptococcus pneumoniae

M

44

NS

K

1500

–

–

3

Pneumonia

b-Hemolytic Streptococcus group A

M

83

NS

K, MI

1500

–

Dobutamine 10 Norepinephrine 1.0

4

Pneumonia

b-Hemolytic Streptococcus group B

M

68

NS

K, ARDS, DIC

2000

–

Dobutamine 5 Norepinephrine 0.02

5

Pyogenic arthritis knee

b-Hemolytic Streptococcus group A

M

59

S

–

3000

–

–

6

Abdominal sepsis after mesenterial thrombosis

Serratia Marcescens

M

59

NS

K,L, ARDS

1000

500

Dobutamine 10

7

Pneumococcal pneumonia

Negative

F

62

S

ARDS

3000

500

–

8

Intravenous catheter sepsis

Staphylococcus aureus

M

50

S

–

1000

500

–

9

ARA-C vasculitis with Burkitt lymphoma

E. coli

F

48

S

–

3000

500

Norepinephrine 0.6

10

Sinusitis maxillaris with M. Kahler

Streptococcus mitis

F

51

NS

K

1000

500

–

11

Aspiration after cardiac arrest

Negative

M

76

NS

ARDS

500

500

Dobutamine 10

12

Aspiration with motorneuron disease

Negative

M

69

S

–

1000

500

Dopamine 5

13

Acute necrotizing pancreatitis

Negative

F

80

NS

K

1160

580

Dobutamine 10

14

Pseudomonas pneumonia

Negative

M

55

S

–

1000

500

Dobutamine 10 Norepinephrine 0.14

15

Pneumococcal pneumonia

Streptococcus pneumoniae

F

84

NS

ARDS, MI

2000

500

–

a

K, kidney; L, liver; ARDS, acute respiratory distress syndrome; DIC, diffuse intravascular coagulation; MI, myocardial infarction.

The Netherlands Journal of Medicine 2000;57:82 – 93

A. J.G.H. Bindels et al. / Extravascular lung water

87

Fig. 2. Course in systemic oxygen delivery (DO 2 ), cardiac index (CI), hemoglobin (Hb) and mean arterial pressure (MAP) in 15 patients. Fluid chal., the measurement after patients had received maximal fluid challenges. The Netherlands Journal of Medicine 2000;57:82 – 93

88

A. J.G.H. Bindels et al. / Extravascular lung water

25% increase in DO 2 after 18 h, and a 28% increase in MAP, although neither change was statistically significant. CI increased somewhat after the installation of inotropes (measurement after 3 h), but no consistent trend was noted. The pre-set goal of CI$ 4.5 l / min / m 2 was achieved in only five patients. Hemoglobin levels increased significantly from 5.660.3 to 6.860.2 mmol / l (P50.005) after 6 h. EVLWI was slightly above the normal range (10.461.2 ml / kg, normal values 5 –10 ml / kg) and did not change during the treatment protocol (Fig. 3, upper panel). Patients with EVLWI above normal reached their maximum values at the first measurement. Fluid challenge did not increase EVLWI (baseline (10.461.2 ml / kg) versus optimal volume measurement (10.060.9 ml / kg) in the patient group with an initial PAWP,16 mmHg, Fig. 3, middle panel). There was no significant correlation between PAWP and EVLWI (Fig. 4). Patients with an initial PAWP .16 mmHg, however, had higher values of EVLWI than patients with an initial PAWP ,16 mmHg at optimal volume status (14.161.1 ml / kg versus 10.060.9 ml / kg, P50.026). EVLWI did not differ between survivors and nonsurvivors throughout the study protocol (Fig. 3, lower panel). Reviewing the total group, ITBVI was normal at baseline (917669 ml / m 2 , normal values 800–1000 ml / m 2 ) and increased during the study period (Fig. 5, upper panel). Patients with PAWP,16 mmHg had a somewhat lower ITBVI at optimal volume status than patients with PAWP.16 mmHg (1018664 versus 12576114 ml / m 2 , P50.15, Fig. 5, middle panel). At baseline nonsurvivors had a lower ITBVI than survivors (783641 versus 10516102 ml / m 2 , P50.08, Fig. 5, lower panel). This difference subsided during further treatment. Ten patients had a PAWP,16 mmHg, and they were actually challenged with colloids between measurement 1 and 2. Only three of them showed a response with an increase in CI .10%. The nine patients who had an increase in CI ,10% after a fluid challenge, had lower ITBVI than patients who responded to a fluid challenge with an increase in CI of .10%, and patients with an initial PAWP.16 mmHg (815645, 11566123 and 12576114 ml / m 2 , respectively). These differences did not reach statistical significance. Likewise, there was a non-significant difference in EVLWI between responders, nonresponders

and patients with PAWP.16 mmHg (8.9562.1 ml / kg, 11.061.6 ml / kg and 14.162.7 ml / kg, respectively). The mean fluid balance after 24 h was 11880 ml (S.D.: 2100 ml), and the mean change in weight after 24 h was 11.9 kg (S.D.: 2.4 kg).

Discussion The main purpose of this study was to find out whether EVLW could be limited in patients with septic shock when treated with a fluid regimen guided by the effects on the cardiac output, with an upper limit of PAWP of 16 mmHg. In general, we found that our patients had EVLWI values slightly above normal from the start of the study, which may reflect increased capillary permeability during septic shock. These values, however, remained unchanged during the rest of the study period. Furthermore, patients with EVLWI-values above normal, all had the maximum level at the start of the study protocol, with decreasing trends afterwards. We found no correlation between PAWP and EVLWI (Fig. 4), which, again, may reflect the increased capillary permeability during sepsis. It also points to the shortcomings of PAWP as indicator of the amount of pulmonary edema. This is apprehensible from the physiological knowledge that PAWP is only a reflection of the real pulmonary capillary hydrostatic pressure [13], which is just one of the determinants of the Starling equation describing the factors influencing fluid transport across capillary membranes [14]. Despite the fact that there was no correlation between PAWP and EVLWI, EVLWI was significantly higher in patients with PAWP.16 mmHg than in patients with PAWP,16 mmHg. This finding indicates that an increased hydrostatic pressure still is a major determinant for the formation of pulmonary edema, even in the presence of a decreased reflection coefficient indicating increased capillary permeability, as is the case in septic shock. Although a PAWP of 16 mmHg generally is considered to be safe regarding the development of pulmonary edema, our findings indicate that this value cannot under all circumstances be accepted as such. This finding might also explain part of the results of two earlier

The Netherlands Journal of Medicine 2000;57:82 – 93

A. J.G.H. Bindels et al. / Extravascular lung water

89

Fig. 3. Course of extravascular lung water index (EVLWI) in the total group (upper panel), PAWP,16 mmHg versus .16 mmHg (middle panel), and survivors versus nonsurvivors (lower panel). [, P,0.05.

The Netherlands Journal of Medicine 2000;57:82 – 93

90

A. J.G.H. Bindels et al. / Extravascular lung water

Fig. 4. Correlation between pulmonary artery wedge pressure (PAWP) and extravascular lung water index (EVLWI) in 15 septic patients.

randomised studies that investigated fluid resuscitation according to EVLW versus PAWP [6,7]. In these studies, a decreased length of stay in the intensive care unit, a shorter time on mechanical ventilation, a less positive fluid balance, and less pulmonary edema was found in the EVLW group. Some of these results may be explained by the fact that in the PAWP group fluid challenges were not stopped until PAWP was 18 mmHg or more, regardless of the effect of a fluid challenge on CI. From our results, it is clear that even with a lower PAWP patients may develop pulmonary edema. In our opinion, future comparisons between fluid regimens based on PAWP and EVLW should therefore take the effects on CI into account, and should also use lower upper limits of PAWP. An optimum left ventricular filling pressure not exceeding a PAWP of 12 mmHg was earlier suggested by Packman and Rackow [15]. In contrast to PAWP, ITBVI has been shown to correlate to CI. So, ITBVI may be a better reflection of cardiac filling than PAWP [5,8]. In our study, ITBVI values were in the normal range at baseline, which is in agreement with the finding that only a minority of patients responded to a fluid challenge. An unexpected finding was the fact that patients who responded to fluid challenges with an increase in CI .10% had values of ITBVI above normal. This may reflect the persisting ability of responders to dilate their ventricles, a mechanism that has been suggested as an important adaptive mechanism in maintaining CO in the early phase of sepsis. Patients who do not

show ventricular dilation are supposed to have more severe myocardial edema [16]. Our study shows that treatment of septic shock with a fluid regimen based on the effect on CI induces an increase in DO 2 and in MAP. The former trend seems attributable to an increase in hemoglobin level. Our finding of only a marginal increase in CI after fluid challenge is probably due to the fact that all patients received the main part of fluids before the first measurement. Supranormal values were not achieved in the patient population under study. From several studies it is known, that although treatment aims at supranormal values, these pre-set values are reached only in a minority of patients [17–19]. Moreover, although critically ill patients who survive mostly have supranormal values of DO 2 [20–22], recent studies have shown that aggressive efforts to reach supranormal levels of CI and DO 2 , as a routine treatment in all critically ill patients, are not beneficial and may even be detrimental [23,24]. A striking result of our study was the fact that both the group with PAWP.16 mmHg and the group with PAWP,16 mmHg had received a total volume of 2000 ml at the point of optimal volume status. Only 30% of the patients who were actually challenged with fluids showed an increase in CI of more than 10%. This indicates that 80% of our patients had received an optimal volume load before the start of invasive hemodynamic monitoring. This finding, at least, puts some question marks to the additional value of invasive monitoring, although invasive

The Netherlands Journal of Medicine 2000;57:82 – 93

A. J.G.H. Bindels et al. / Extravascular lung water

91

Fig. 5. Course of intrathoracic blood volume index (ITBVI) in the total group (upper panel), PAWP,16 mmHg versus .16 mmHg (middle panel), and survivors versus nonsurvivors (lower panel).

The Netherlands Journal of Medicine 2000;57:82 – 93

92

A. J.G.H. Bindels et al. / Extravascular lung water

monitoring earlier was found to be superior to clinical judgement with regard to the hemodynamic assessment of critically ill patients [25]. In conclusion, the present study showed that patients with septic shock did not develop pulmonary edema during the first 24 h of treatment, when their fluid regimen was guided by its effect on the cardiac output, and PAWP was used only with an upper limit of 16 mmHg, rather than with a pre-set value which should be aimed at. More studies will be necessary to clarify whether such an approach actually affects patient outcome.

[9]

[10]

[11] [12]

Acknowledgements [13]

The authors would like to thank Professor JeanLouis Vincent for his valuable comments on earlier versions of this manuscript.

[14]

[15]

References

[16]

[1] Simmons RS, Berdine GG, Seidenfeld JJ, Prihoda TJ, Harris GD, Smith JD, Gilbert TJ, Mota E, Johanson Jr. WG. Fluid balance and the acute respiratory distress syndrome. Am Rev Respir Dis 1987;135:924–9. [2] Lowell JA, Schifferdecker C, Driscoll DF, Benotti PN, Bistrian BR. Postoperative fluid overload: not a benign problem. Crit Care Med 1990;18:728–33. [3] Humphrey H, Hall J, Sznajder I, Silverstein M, Wood L. Improved survival in ARDS patients with a reduction in pulmonary capillary wedge pressure. Chest 1990;97:1176– 80. [4] Schuller D, Mitchell JP, Calandrino FS, Schuster DP. Fluid balance during pulmonary edema. Is fluid gain a marker or a cause of poor outcome? Chest 1991;100:1068–75. [5] Lichtwarck-Aschoff M, Zeravik J, Pfeiffer UJ. Intrathoracic blood volume accurately reflects circulatory volume status in critically ill patients with mechanical ventilation. Intens Care Med 1992;18:142–7. [6] Eisenberg PR, Hansbrough JR, Anderson D, Schuster DP. A prospective study of lung water measurements during patient management in an intensive care unit. Am Rev Respir Dis 1987;136:662–8. [7] Mitchell JP, Schuller D, Calandrino FS, Schuster DP. Improved outcome based on fluid management in critically ill patients requiring a pulmonary artery catheter in an intensive care unit. Am Rev Respir Dis 1992;145:990–8. [8] Bindels AJGH, van der Hoeven JG, Meinders AE. Intrathoracic blood volume as indicator of cardiac preload in

[17]

[18]

[19]

[20]

[21]

[22] [23]

critically ill patients (Abstract). Intens Care Med 1995;21:S216. Dhainaut JF, Edwards JD, Grootendorst AF, Nightingale P, Pinsky MR, Reinhart K, Shoemaker WC, Vincent JL. Practical aspects of oxygen transport: conclusions and recommendations of the roundtable conference. Intens Care Med 1990;16(Suppl 2):S179–180. Fiddian-Green RG, Haglund U, Gutierrez G, Shoemaker WC. Goals for the resuscitation of shock. Crit Care Med 1993;21:S25–31. Lewis FR, Elings VB. Bedside measurement of lung water. J Surg Res 1979;27:250–61. ¨ ¨ Pfeiffer UJ, Backus G, Blumel G, Eckart J, Muller P, Winkler P, Zeravik J, Zimmerman GJ. A fiberoptics based system for integrated monitoring of cardiac output, intravascular blood volume, extravascular lung water, O 2 -saturation and a–v differences. In: Lewis FR, Pfeiffer UJ, editors, Practical applications of fiberoptics in critical care monitoring, Berlin: Springer Verlag, 1990, pp. 114–25. Versprille A. The pulmonary circulation during mechanical ventilation. Acta Anesthesiol Scan 1990;34(Suppl):51–62. Nunn JF. Pulmonary oedema. In: Nunn JF, editor, Nunn’s applied respiratory physiology, Oxford: Butterworth– Heinemann, 1993, pp. 484–93. Packman MI, Rackow EC. Optimal left heart filling pressure during fluid resuscitation of patients with hypovolemia and septic shock. Crit Care Med 1983;11:165–9. Parker MM, Shelhamer JH, Bacharach SL, Green MV, Natanson C, Frederick TM, Damske BA, Parillo JE. Profound but reversible myocardial depression in patients with septic shock. Ann Intern Med 1984;100:483–90. Tuchschmidt J, Fried J, Astiz M, Rackow E. Elevation of cardiac output and oxygen delivery improves outcome in septic shock. Chest 1992;102:216–20. Boyd O, Grounds RM, Bennett ED. A randomized clinical trial of the effect of deliberate perioperative increase of oxygen delivery on mortality in high-risk surgical patients. J Am Med Assoc 1993;270:2699–707. Yu M, Levy MM, Smith P, Takiguchi SA, Miyasaki A, Myers SA. Effect of maximizing oxygen delivery on morbidity and mortality rates in critically ill patients: a prospective, randomized, controlled study. Crit Care Med 1993;21:830–8. Shoemaker WC, Appel PL, Waxman K, Schwartz S, Chang P. Clinical trial of survivors’ cardiorespiratory patterns as therapeutic goals in critically ill postoperative patients. Crit Care Med 1982;10:398–403. Shoemaker WC, Appel PL, Kram HB. Measurement of tissue perfusion by oxygen transport patterns in experimental shock and in high-risk surgical patients. Intens Care Med 1990;16(Suppl 2):S135–144. Tuchschmidt J, Oblitas D, Fried JC. Oxygen consumption in sepsis and septic shock. Crit Care Med 1991;19:664–71. Hayes MA, Timmins AC, Yau EHS, Palazzo M, Hinds ChJ, Watson D. Elevation of systemic oxygen delivery in the treatment of critically ill patients. New Engl J Med 1994;330:1717–22.

The Netherlands Journal of Medicine 2000;57:82 – 93

A. J.G.H. Bindels et al. / Extravascular lung water [24] for the SvO 2 collaborative group, Gattinoni L, Brazzi L, Pelosi P, Latini R, Tognoni G, Pesenti A, Fumagalli R. A trial of goal-oriented hemodynamic therapy in critically ill patients. New Engl J Med 1995;333:1025–32.

93

[25] Eisenberg PR, Jaffe AS, Schuster DP. Clinical evaluation compared to pulmonary artery catheterization in the hemodynamic assessment of critically ill patients. Crit Care Med 1984;12:549–53.

The Netherlands Journal of Medicine 2000;57:82 – 93