Effects of perfusion pressure on tissue perfusion in septic shock

David LeDoux, MD; Mark E. Astiz, MD, FCCM; Charles M. Carpati, MD; Eric C. Rackow, MD, FCCM

Objective: To measure the effects of increasing mean arterial Increasing the MAP from 65 to 85 mm Hg with norepinephrine
pressure (MAP) on systemic oxygen metabolism and regional resulted in increases in cardiac index from 4.7 ⴞ 0.5 L/min/m2 to
tissue perfusion in septic shock. 5.5 ⴞ 0.6 L/min/m2 (p < 0.03). Arterial lactate was 3.1 ⴞ 0.9
Design: Prospective study. mEq/L at a MAP of 65 mm Hg and 3.0 ⴞ 0.9 mEq/L at 85 mm Hg
Setting: Medical and surgical intensive care units of a tertiary (NS). The gradient between arterial PCO2 and gastric intramucosal
care teaching hospital. PCO2 was 13 ⴞ 3 mm Hg (1.7 ⴞ 0.4 kPa) at a MAP of 65 mm Hg
Patients: Ten patients with the diagnosis of septic shock who and 16 ⴞ 3 at 85 mm Hg (2.1 ⴞ 0.4 kPa) (NS). Urine output at 65
required pressor agents to maintain a MAP > 60 mm Hg after mm Hg was 49 ⴞ 18 mL/hr and was 43 ⴞ 13 mL/hr at 85 mm Hg
fluid resuscitation to a pulmonary artery occlusion pressure (NS). As the MAP was raised, there were no significant changes in
(PAOP) > 12 mm Hg. skin capillary blood flow or red blood cell velocity.
Interventions: Norepinephrine was titrated to MAPs of 65, 75, Conclusions: Increasing the MAP from 65 mm Hg to 85 mm Hg
and 85 mm Hg in 10 patients with septic shock. with norepinephrine does not significantly affect systemic oxygen
Measurements and Main Results: At each level of MAP, hemo- metabolism, skin microcirculatory blood flow, urine output, or
dynamic parameters (heart rate, PAOP, cardiac index, left ven- splanchnic perfusion. (Crit Care Med 2000; 28:2729 –2732)
tricular stroke work index, and systemic vascular resistance KEY WORDS: sepsis; sepsis syndrome; septic shock; norepineph-
index), metabolic parameters (oxygen delivery, oxygen consump- rine; systemic hypotension; regional blood flow; gastric tonome-
tion, arterial lactate), and regional perfusion parameters (gastric try; lactate; arterial pressure; tissue oxygenation; laser-Doppler
mucosal PCO2, skin capillary blood flow and red blood cell velocity,
urine output) were measured.

S eptic shock is characterized by vary considerably, ranging from 60 mm mechanical ventilation; c) MAP ⬍ 60 mm Hg
hypotension, which is not re- Hg, which represents the point at which despite fluid resuscitation to a pulmonary ar-
versed despite aggressive fluid organ perfusion becomes pressure depen- tery occlusion pressure (PAOP) ⬎12 mm Hg;
resuscitation. Nitric oxide (1, dent, to levels of 80 mm Hg to 90 mm Hg and d) the requirement for norepinephrine to
2), prostacyclin (3, 4), and bradykinins (10 –12). The benefit of titrating catechol- maintain MAP ⱖ 60 mm Hg (13). Patients
(5) are among the substances that cause amine infusions to higher levels of arte- with a need for varying doses of pressors or
decreases in arteriolar tone and vasodila- rial pressure has never been demon- inotropic agents during the study period,
tion during septic shock. In addition, ex- strated and may adversely affect organ other than norepinephrine, were excluded.
perimental studies suggest that sepsis re- perfusion by inducing excessive vasocon- Heart rate (HR) was monitored continu-
ously. Arterial pressure was monitored via an
sults in down-regulation of adrenergic striction. Accordingly, the purpose of this
arterial catheter in either the radial or femoral
receptors, which may attenuate the re- study was to examine the effect of titrat-
artery. All patients were catheterized with a
sponse to endogenous and exogenous cat- ing norepinephrine to different levels of
pulmonary artery catheter. Serial measure-
echolamines (6). Impaired vascular reac- MAP on systemic and regional indices of ments of HR, MAP, PAOP, and central venous
tivity has also been described in septic perfusion. pressure were made. Transducers were refer-
shock (7, 8), leading to the hypothesis enced to the midaxillary line and all pressures
that organ perfusion may be pressure de- were taken at end-expiration. Cardiac index
MATERIALS AND METHODS
pendent in this syndrome. (CI) was measured by thermodilution using
The primary vasoactive agents used in The protocol was approved by the Institu- measurements that varied by ⬍10%. Oxyhe-
clinical practice to augment vascular tional Research Board of Saint Vincents Hos- moglobin saturation and content were mea-
tone are catecholamines. Of these agents, pital and Medical Center, and written in- sured with a co-oximeter (IL-282, Instrumen-
norepinephrine has been reported to im- formed consent was obtained from the proxies tation Laboratories, Lexington, MA). Arterial,
prove renal and splanchnic perfusion in or closest relatives of all patients admitted to mixed venous, and tonometrically measured
patients with septic shock (9, 10). How- the study. The study population included 10 CO2 tensions were determined by a blood gas
ever, the recommended end points with patients admitted to the medical or surgical analyzer (Nova Biomedical Stat Profile 5,
regard to mean arterial pressure (MAP) intensive care unit who met the following in- Waltham, MA). Arterial lactate levels were de-
clusion criteria: a) an identified site of infec- termined by the enzymatic method (Vitros 950
tion; b) a systemic inflammatory response as lactate analyzer, Johnson & Johnson, Roches-
From Saint Vincents Hospital and Medical Center, indicated by a temperature ⬎38.2°C or ter, NY). Derived hemodynamic variables were
New York Medical College, New York, New York. ⬍36.8°C, a heart rate ⬎90 beats/min, a respi- calculated from standard formulae: systemic
Copyright © 2000 by Lippincott Williams & Wilkins ratory rate ⬎20 breaths/min, or the need for vascular resistance index (SVRI) (dyne䡠sec/

Crit Care Med 2000 Vol. 28, No. 8 2729

cm5䡠m2) ⫽ (MAP ⫺ central venous pressure / ences were considered significant at p ⬍ .05. Table 1. Characteristics of 10 patients with septic
CI) ⫻ 80; systemic oxygen delivery (ḊO2) (mL/ Data are presented as mean ⫾ SE. shock
min/m2) ⫽ arterial oxygen content ⫻ CI; sys-
temic oxygen consumption (V̇O2) (mL/min/ Age (yrs) 68 ⫾ 12
RESULTS APACHE II 29 ⫾ 2.7
m 2 ) ⫽ arteriovenous oxygen content M/F 7/3
difference ⫻ CI; left ventricular stroke work Seven men and three women were en- Dobutamine/dopamine 1/5
index (g䡠m/m2) ⫽ CI / HR ⫻ (MAP ⫺ PAOP) ⫻ tered into the study (Table 1). Four pa- Cause of sepsis
0.0136. Cholecystitis 3
tients had pneumonia and six had an in-
A tonometric nasogastric tube was inserted Pneumonia 4
tra-abdominal source of sepsis. Two Urosepsis 1
into the stomach (TRIP NGS Catheter, Tono- patients were anuric and on continuous Colitis 2
metrics, Worcester, MA), after which radio- venovenous hemofiltration during the
graphic confirmation of catheter position was Data for age and Acute Physiology and
study. Patients were studied within the
obtained. All patients were placed on intrave- Chronic Health Evaluation (APACHE) II score
first 24 hrs of their course but after initial
nous famotidine. Phosphate-buffered solution are presented as mean ⫾ SE.
hemodynamic stabilization. One patient
was used to improved the accuracy of the
was receiving dobutamine, and five pa-
measurements (14). Intraluminal PCO2 was
tients were receiving low-dose dopamine. increases in gastric intramucosal PCO2
measured from 1.5-mL samples that were as-
pirated from the catheter balloon anaerobi-
The average Acute Physiology and that were not statistically significant, and
cally after discarding the first 1 mL. The mu-
Chronic Health Evaluation (APACHE) II there were no changes in the arterial-
cosal PCO2 measurement was multiplied by the score was 29 ⫾ 3 at the time of entry. intramucosal PCO2 gradient at the differ-
appropriate equilibration factor provided by Three patients survived to the 30-day fol- ent levels of arterial pressure. No changes
the manufacturer. Laser-Doppler measure- low-up. in urinary output were observed as MAP
ments of skin red blood cell flow and velocity The dose of norepinephrine was 23 ⫾ was increased.
were recorded from the ventral surface of one 22 ␮g/min to maintain a MAP of 65 mm
forearm (ALF21R Advance Laser Flowmeter, Hg, 31 ⫾ 25 ␮g/min to maintain a MAP DISCUSSION
Advance Company, Tokyo, Japan). To measure of 75 mm Hg, and 47 ⫾ 39 ␮g/min to
urine output, drainage from a Foley catheter maintain a MAP of 85 mm Hg (Table 2). Norepinephrine is frequently used to
was emptied before and after the measure- The highest dose of norepinephrine re- improve arterial pressure in patients with
ment periods. quired by an individual patient to main- septic shock who remain hypotensive af-
Ventilator settings were adjusted to a tidal tain a MAP of 65 mm Hg was 53 ␮g/min, ter fluid infusion. The dose of norepi-
volume of 8 –10 mL/kg at a rate set to meet the to maintain a MAP of 75 mm Hg was 65 nephrine used in our study is in the mid-
patients’ ventilatory requirements with an ␮g/min, and to maintain a MAP of 85 mm range of infusion rates reported in
oxygen saturation of ⬎90%. All patients were Hg was 115 ␮g/min. There was an in- previous studies, where MAP levels of
treated with 250-mL aliquots of 5% albumin crease in CI from 4.7 ⫾ 0.5 L/min/m2 to ⬎80 mm Hg were achieved (10, 15, 16).
or 6% hetastarch to attain a PAOP of 12 mm 5.5 ⫾ 0.6 L/min/m2 as the norepineph- Both ␤- and ␣-adrenergic receptors are
Hg. Patients were transfused with packed red rine dose was increased to attain a MAP of stimulated with norepinephrine. How-
blood cells if, at study admission, the hemo- 85 mm Hg. This was not a statistically ever, the primary hemodynamic effect of
globin was ⬍9 g/dL. Patients were sedated significant difference by ANOVA, but it this drug is to increase SVR and thereby
with lorazepam or propofol; however, no ad- did exhibit a significant linear trend (p ⬍ increase MAP. Increases in cardiac output
ditional sedation, antipyretics, or vasoactive 0.03; Table 2). As the norepinephrine in- are usually modest but may occur, as was
drugs were administered during the study pe- fusion was increased, SVRI increased observed in our study, and can contribute
riod. from 998 ⫾ 94 dyne䡠sec/cm5䡠m2 at 65 to an increase in MAP as also evidenced in
After entering into the study, the patient’s mm Hg to 1065 ⫾ 101 dyne䡠sec/cm5䡠m2 our study (17, 18).
norepinephrine dose was adjusted to attain a at 75 mm Hg and to 1216 ⫾ 159 dyne䡠sec/ Oxygen delivery increased in parallel
MAP of 65 mm Hg. After a titration and equil-
cm5䡠m2 at 85 mm Hg, which exhibited a with the increase in CI. There was a ten-
ibration period of 45 mins, an additional 60
significant linear trend (p ⬍ 0.046). In- dency to increased oxygen consumption
mins then elapsed during which no alterations
creasing the dose of norepinephrine was in our study, but this may have reflected
in other vasoactive medications were permit-
associated with significant changes in HR primarily the thermogenic effect of cat-
ted. Hemodynamic data were then collected
but not in PAOP. A significant increase in echolamines on metabolic rate (19, 20).
and arterial lactate, blood gases, urine output,
tonometric data, and laser-Doppler measure-
left ventricular stroke work index was ob- Alternatively, the increase in the calcu-
ments were recorded. The norepinephrine served that primarily reflected an in- lated oxygen consumption may reflect
dose was then increased to attain a MAP of 75 crease in pressure work. the interaction of dependent variables
mm Hg during a 45-min period, and measure- Increases in oxygen delivery and con- (21). Lactate levels did not change, which
ments were repeated after an additional 60- sumption at each level of MAP were ob- suggests that increasing the degree of
min period of hemodynamic stability. Finally, served, but the linear trend was only sig- vasoconstriction as MAP was increased
the norepinephrine dose was increased to at- nificant for oxygen delivery (Table 3). from 65 mm Hg to 85 mm Hg did not
tain a MAP of 85 mm Hg, and data were again Mixed venous oxygen saturation and ar- adversely affect systemic tissue perfusion.
collected after the two periods. terial lactate levels were unchanged Similarly, there were no significant ad-
Group differences were analyzed using re- throughout the study. Changes in skin verse effects on microcirculatory blood
peated-measures analysis of variance (ANOVA) capillary blood flow were not significant, flow in the subcutaneous tissues as mea-
with the Greenhouse-Geisser adjustment; be- and erythrocyte velocity also did not sured by laser-Doppler. The lack of any
cause of the small sample size, an extension change with increasing doses of norepi- adverse effects related to vasoconstriction
for linear trends was also examined. Differ- nephrine (Table 4). There were modest probably results from the fact that al-

2730 Crit Care Med 2000 Vol. 28, No. 8

though vascular tone increased as the splanchnic mucosal perfusion improved not significantly redirect blood flow away
norepinephrine infusion was increased, (9, 23), whereas in others, deterioration from the splanchnic circulation in septic
SVRI remained well within the normal has been observed (18, 24). In our study, shock (25).
range. This physiology contrasts with there was no significant change in gastric There also was no change in urinary
forms of shock in which cardiac output is intramucosal P CO 2 and the arterial- output observed in our study as the MAP
decreased, where the primary effect of intramucosal PCO2 gradient as the norepi- was increased with norepinephrine. This
norepinephrine is to increase vascular re- nephrine dose was increased, suggesting pattern contrasts with several previous
sistance to levels in excess of the normal that vasoconstriction associated with in- reports suggesting that norepinephrine
range and to potentially compromise tis- creasing the norepinephrine infusion did improved renal function as measured by
sue perfusion (22). not redistribute blood away from the urinary flow and creatinine clearance (10,
Norepinephrine has been reported to splanchnic circulation. This observation 16, 26). The difference between our study
have variable effects on splanchnic blood is consistent with experimental studies and those reports demonstrating im-
flow in septic shock. In some cases, which suggest that adrenergic agents do proved renal function primarily reflects
that fact that our baseline was a MAP of
65 mm Hg rather than the more hypo-
Table 2. Changes in hemodynamic parameters as mean arterial pressure (MAP) is increased from 65
tensive levels that characterized the pre-
mm Hg to 85 mm Hg with norepinephrine
vious studies. Indeed, in many of the pre-
MAP vious reports the initial levels of MAP
were in the range of 55 mm Hg, levels
65 mm Hg 75 mm Hg 85 mm Hg F/LT below the autoregulatory threshold for
renal blood flow (27).
HR (beats/min) 97 ⫾ 4 101 ⫾ 4 105 ⫾ 5 .02/.02 The goal of vasopressor therapy is to
MAP (mm Hg) 65 ⫾ 0.5 75 ⫾ 0.4 86 ⫾ 0.4 .0001/.0001
CI (L/min/m2) 4.7 ⫾ 0.5 5.3 ⫾ 0.6 5.5 ⫾ 0.6 .07/.03 improve arterial pressure while avoiding
PAOP (mm Hg) 14 ⫾ 1 15 ⫾ 1 16 ⫾ 1 .18/.16 excessive vasoconstriction. The mini-
LVSWI (g䡠m/m2) 45 ⫾ 3 52 ⫾ 5.5 63 ⫾ 7 .01/.01 mum acceptable level of MAP that is most
SVRI (dyne䡠sec/m2䡠cm5) 998 ⫾ 94 1065 ⫾ 101 1216 ⫾ 159 .09/.046 commonly cited is 60 mm Hg. This level
Norepinephrine (␮g/min) 23 ⫾ 22 31 ⫾ 25 47 ⫾ 39 .02/.016
of MAP represents the point at which
F, p value for repeated-measures analysis of variance (ANOVA) as MAP is increased from 65 mm Hg autoregulatory control of blood flow to
to 85 mm Hg; LT, p value for extension of ANOVA for linear trend; HR, heart rate; CI, cardiac index; the heart, kidneys, and brain ceases, re-
PAOP, pulmonary artery occlusion pressure; LVSWI, left ventricular stroke work index; SVRI, systemic sulting in pressure-dependent organ
vascular resistance index. blood flow (28 –30). In patients with hy-
Data are presented as mean ⫾ SE. pertensive disease or atherosclerotic dis-
ease, the autoregulatory curve may be
Table 3. Indices of systemic oxygen metabolism as mean arterial pressure (MAP) is increased from 65 shifted to the right, requiring higher
mm Hg to 85 mm Hg pressures to maintain organ perfusion.
The ability to autoregulate blood flow
MAP
may also be altered by organ injury as has
65 mm Hg 75 mm Hg 85 mm Hg F/LT been demonstrated in acute renal failure
and in head injury (31, 32). Alterations in
ḊO2 (mL/min/m2) 620 ⫾ 59 670 ⫾ 59 703 ⫾ 74 .07/.02 vascular reactivity with septic shock may
V̇O2 (mL/min/m2) 119 ⫾ 12 138 ⫾ 20 153 ⫾ 20 .24/.11 also potentially alter the relationship be-
Sv̄O2 (%) 76 ⫾ 3 76 ⫾ 2 70 ⫾ 2 .94/.76 tween blood flow and organ perfusion,
Lactate (mEq/L) 3.1 ⫾ 0.9 2.9 ⫾ 0.8 3.0 ⫾ 0.9 .55/.77
resulting in a wider range of pressure-
F, p value for repeated-measures ANOVA as MAP is increased from 65 mm Hg to 85 mm Hg; LT, dependent perfusion. These concerns
p value for extension of ANOVA for linear trend; ḊO2, oxygen delivery; ȮO2, oxygen consumption; Sv̄O2, have led to therapeutic protocols in
venous oxygen saturation. which pressor therapy is titrated to levels
Data are presented as mean ⫾ SE. of MAP between 80 mm Hg and 90 mm
Hg (10, 11). Our study does not support
Table 4. Indices of regional perfusion as MAP is increased from 65 mm Hg to 85 mm Hg the hypothesis that tissue perfusion is
pressure dependent over the range of 65
MAP mm Hg to 85 mm Hg and suggests little
benefit in increasing the MAP to ⬎65 mm
65 mm Hg 75 mm Hg 85 mm Hg F/LT
Hg. Conversely, our data also suggest
Urinary output (mL) 49 ⫾ 18 56 ⫾ 21 43 ⫾ 13 .60/.71 that in vasodilated patients with septic
Capillary blood flow (mL/min/100 g) 6.0 ⫾ 1.6 5.8 ⫾ 1.2 5.3 ⫾ 0.9 .59/.55 shock, the use of vasopressors to increase
Red cell velocity (au) 0.42 ⫾ 0.06 0.44 ⫾ 0.06 0.42 ⫾ 0.06 .74/.97 vascular tone toward more normal values
PiCO2 (mm Hg) 41 ⫾ 2 47 ⫾ 2 46 ⫾ 2 .11/.12 of vascular resistance does not adversely
Pa-PiCO2 (mm Hg) 13 ⫾ 3 17 ⫾ 3 16 ⫾ 3 .27/.40
affect tissue perfusion.
F, p value for repeated-measures analysis of variance (ANOVA) as MAP is increased from 65 mm Hg A limitation to our study is the small
to 85 mm Hg; LT, p value for extension of ANOVA for linear trend; au, arbitrary units; PiCO2, gastric number of patients and short infusion
intramucosal PCO2; Pa-PiCO2, arterial-gastric intramucosal CO2 gradient. period. Despite the sample size, signifi-
Data are presented as mean ⫾ SE. cant changes in hemodynamic variables

Crit Care Med 2000 Vol. 28, No. 8 2731

were observed that were not evident in human septic shock: Association with myo- 22. Gunnar RM, Cruz A, Boswell J, et al: Myo-
the perfusion-related variables. The infu- cardial hyporesponsiveness to cat- cardial infarction with shock. Hemodynamic
sion period was purposefully kept short to echolamines. Crit Care Med 1993; 21:31–39 studies and results of therapy. Circulation
7. Auclair MC, Svinarell P, Schmitt H: Vascular al- 1966; 32:753–762
avoid the background effects of changes
pha-adrenoceptor blockade by E. coli endotoxin 23. Meier-Hellman A, Specht M, Hannemann L,
in the patients’ underlying conditions. Al-
in the rat. Eur J Pharmacol 1986; 127:121–124 et al: Splanchnic blood flow is greater in
though changes in lactate may take 8. Martin CM, Yaghi A, Sibbald WJ, et al: Dif- septic shock treated with norepinephrine
longer to equilibrate between interven- ferential impairment of vascular reactivity of than in severe sepsis. Intensive Care Med
tions, it should be noted that there were small pulmonary and systemic arteries in 1996; 22:1354 –1359
no significant changes in lactate over the hyperdynamic sepsis. Am Rev Respir Dis 24. Meier-Hellman A, Reinhart K: Influence of
entire 6-hr study period (33, 34). Simi- 1993; 148:164 –172 catecholamines on regional perfusion and
larly, although the temporal relationship 9. Marik PE, Mohedin M: The contrasting ef- tissue oxygenation in septic shock patients.
between changes in splanchnic blood fects of dopamine and norepinephrine on Sepsis. In: Current Perspectives in Patho-
flow and changes in arterial-intramucosal systemic and splanchnic oxygen utilization physiology and Therapy. Reinhart K, Eyrich
in hyperdynamic sepsis. JAMA 1994; 272: K, Sprung C (Eds). Berlin, Springer-Verlag,
PCO2 gradient is unclear, there was also
1354 –1357 1994, pp. 274 –291
no significant change in this parameter
10. Martin C, Papazian L, Perrin G, et al: Nor- 25. Bersten AD, Hersch M, Cheung J, et al: The
over the time of the study. epinephrine or dopamine for the treatment effect of various sympathomimetics on the
It is also important to note that the of hyperdynamic septic shock. Chest 1993; regional circulations in hyperdynamic sepsis.
average age of patients in our study was 103:1826 –1831 Surgery 1992; 112:549 –561
68 yrs, suggesting that our results should 11. Hayes MA, Timmins AC, Yau E, et al: Eleva- 26. Desjars P, Pinaud M, Bugnon D, et al: Nor-
be applicable to an older, sicker patient tion of systemic oxygen delivery in the treat- epinephrine therapy has no deleterious renal
population. However, only three patients ment of critically ill patients. N Engl J Med effects in human septic shock. Crit Care Med
had a previous history of hypertension, 1994; 330:1717–1722 1989; 17:426 – 429
and it is possible that this group of pa- 12. Astiz M, Rackow E: Update on the clinical 27. Robertson CR, Deen WM, Troy JL, et al: Dy-
approach to septic shock. Lancet 1998; 351:
tients may benefit from higher levels of namics of glomerular ultrafiltration in the
1501–1505
arterial pressure. A confounding variable rat. III. Hemodynamics and autoregulation.
13. American College of Chest Physicians/
in our study was that many of the pa- Am J Physiol 1972; 223:1191–1200
Society of Critical Care Medicine Consensus
tients were receiving dopamine. This 28. Johnson P: Autoregulation of blood flow.
Conference: Definitions for sepsis and organ
agent has been observed to significantly Circ Res 1986; 59:483– 495
failure and guidelines for the use of innova-
29. Mueller H, Ayres S, Giannelli S, et al: Effect
enhance renal blood flow in the face of tive therapies in sepsis. Crit Care Med 1992;
of isoproterenol, l-norepinephrine, and in-
adrenergic vasoconstriction and may 20:864 – 874
traaortic counterpulsation on hemodynam-
have mitigated some of the effects of in- 14. Knichwitz G, Mertes N, Kuhmann M: Im-
ics and myocardial metabolism in shock fol-
creased vasoconstriction associated with proved PCO2 measurement in six standard
lowing acute myocardial infarction.
higher doses of norepinephrine (35, 36). blood gas analysers using a phosphate-
Circulation 1972; 16:335–350
buffered solution for gastric tonometry.
Finally, our observations may not be ap- 30. Strandgaard S, Sengupta E, MacKenzie E:
Anaesthesia 1995; 50:532–534
plicable in patients in whom vascular re- The lower and upper limits of autoregulation
15. Schreuder WO, Schneider AJ, Groeneveld J,
sistance is increased with norepinephrine et al: Effect of dopamine vs norepinephrine
of cerebral blood flow. In: Cerebral Circula-
in excess of the normal range in an effort on hemodynamics in septic shock. Chest tion and Metabolism. Langfitt T, McHenry
to increase arterial pressure. 1989; 95:1282–1288 LC Jr, Reivich M, et al. (Eds). New York,
16. Redl-Wenzl EM, Armbruster C, Edelmann G, Springer-Verlag, 1975, pp. 3– 6
et al: The effects of norepinephrine on hemo- 31. Kelleher SP, Robinette JB, Conger JD: Sym-
REFERENCES pathetic nervous system in the loss of auto-
dynamics and renal function in severe septic
1. Kilbourn RG, Jubran A, Gross SS, et al: Re- shock states. Intensive Care Med 1993; 19: regulation in acute renal failure. Am J
versal of endotoxin-mediated shock by NG- 151–154 Physiol 1984; 246:F379 –F386
methyl-L-arginine, an inhibitor of nitric 17. Winslow EJ, Loeb HS, Rahimtoola SH, et al: 32. Cold GE: Cerebral blood flow in the acute
oxide synthesis. Biochem Biophys Res Hemodynamic studies and results of therapy phase after head injury. Part 2: Correlation to
Commun 1990; 172:1132–1138 in 50 patients with bacteremic shock. Am J intravascular pressure (IVP), cerebral perfu-
2. Lorente JA, Landin L, DePablo R, et al: L- Med 1973; 54:421– 432 sion pressure (CPP), PaCO2, ventricular fluid
arginine pathway in the sepsis syndrome. 18. Ruokonen E, Takala J, Kari A, et al: Regional lactate, lactate/pyruvate ratio and pH. Acta
Crit Care Med 1993; 21:1287–1295 blood flow and oxygen transport in septic Anaesthesiol Scand 1981; 25:332–335
3. Slotman GJ, Quinn JV, Burchard KW, et al: shock. Crit Care Med 1993; 21:1296 –1303 33. Vincent JL, Dufaye P, Berre J, et al: Serial
Thromboxane, prostacyclin, and the hemo- 19. Chiolero R, Flatt JP, Revelly JP, et al: Effects lactate determinations during circulatory
dynamic effects of graded bacteremic shock. of catecholamines on oxygen consumption shock. Crit Care Med 1983; 11:449 – 451
Circulatory Shock 1985; 16:395– 404 and oxygen delivery in critically ill patients. 34. Falk JL, Rackow EC, Leavy J, et al: Delayed
4. Slotman GJ, Burchard KW, Williams JJ, et al: Chest 1991; 100:1676 –1684 lactate clearance in patients surviving circu-
Interaction of prostaglandin, activated com- 20. Ensinger H, Weichel T, Lindner KH, et al: latory shock. Acute Care 1985; 11:212–215
plement, and granulocytes in clinical sepsis Effects of norepinephrine, epinephrine, and 35. Hoogenberg K, Smit A, Girbes A: Effects of
and hypotension. Surgery 1986; 99:744 –751 dopamine infusions on oxygen consumption low dose dopamine on renal and systemic
5. Pixley RA, De La Cadena RA, Page JD, et al: in volunteers. Crit Care Med 1993; 21: hemodynamics during incremental norepi-
Activation of the contact system in lethal 1502–1508 nephrine infusion in healthy volunteers. Crit
hypotensive bacteremia in a baboon model. 21. Ronco JJ, Fenwick JC, Wiggs BR, et al: Oxy- Care Med 1998; 26:260 –265
Am J Pathol 1992; 140:897–906 gen consumption is independent of increases 36. Richer M, Robert S, Leble M: Renal hemody-
6. Silverman HJ, Penaranda R, Orens JB, et al: in oxygen delivery by dobutamine in septic namics during norepinephrine and low dose
Impaired beta-adrenergic receptor stimula- patients who have normal or increased plasma dopamine infusions in man. Crit Care Med
tion of cyclic adenosine monophosphate in lactate. Am Rev Respir Dis 1993; 147:25–31 1996; 24:1150 –1156

2732 Crit Care Med 2000 Vol. 28, No. 8