Papillary thyroid carcinoma presenting with severe Guillain-Barré syndrome.

BACKGROUND: The Guillan-Barré syndrome (GBS) is the most common acute disease of the peripheral nervous system, often necessitating ICU care. Although no clear paraneoplastic syndrome has been defined, it has been known to occur together with several types of cancer. METHODS: We present the case of a 35-year-old man with acute flaccid paralysis and cervical lymphadenopathy. RESULTS: the patient was diagnosed with a severe presentation of GBS and papillary thyroid cancer. After surgical treatment for his cancer, his condition improved and he reached a nearly full recovery. CONCLUSION: our case concerns a young man in whom a severe presentation of GBS led to the diagnosis of thyroid cancer. To our knowledge, this represents the first reported case of this association.

Atriobronchial Fistula Complicated by Septic Cerebral Air Emboli After Pulmonary Vein Ablation.

OBJECTIVE: To describe a case of an infected atriobronchial fistula as a late complication after pulmonary vein ablation, leading to septic air emboli and requiring urgent cardiac surgery. DATA SOURCES: Clinical observation. STUDY SELECTION: Case report. DATA EXTRACTION: Relevant clinical information. PubMed was searched for relevant literature. DATA SYNTHESIS: Given its high success and low complication rate, pulmonary vein isolation is expected to be increasingly performed worldwide. Despite its success, some of its rare complications are potentially devastating and are difficult to diagnose early. In this report, we present the case of a 32-year-old woman, who was readmitted to hospital 2 months after pulmonary vein ablation. The clinical picture resembled meningococcemia with spreading petechiae on legs and arms raising concern for Waterhouse-Friderichsen syndrome. Further echocardiographic investigation led to the discovery of massive amounts of intracardiac air which demanded urgent lung isolation and sternotomy. Intraoperatively a small infected left atrial perforation was oversewn and a fistula to the right main bronchus was closed by means of an autologous pericardial patch. One month later, still revalidating, she could be discharged home with only minor neurologic sequelae. CONCLUSIONS: Clinicians should be aware of the dramatic complications of invasive antiarrhythmic procedures and their atypical and late presentations. Better preprocedural appreciation of cardiac wall thickness, early echocardiographic diagnosis, and swift referral for cardiac surgery might impact outcome dramatically.

A comparison of two insulin infusion protocols in the medical intensive care unit by continuous glucose monitoring.

BACKGROUND: Achieving good glycemic control in intensive care units (ICU) requires a safe and efficient insulin infusion protocol (IIP). We aimed to compare the clinical performance of two IIPs (Leuven versus modified Yale protocol) in patients admitted to medical ICU, by using continuous glucose monitoring (CGM). This is a pooled data analysis of two published prospective randomized controlled trials. CGM monitoring was performed in 57 MICU patients (age 64 ± 12 years, APACHE-II score 28 ± 7, non-diabetic/diabetic: 36/21). The main outcome measures were percentage of time in normoglycemia (80-110 mg/dl) and in hypoglycemia (<60 mg/dl), and glycemic variability (standard deviation, coefficient of variation, mean amplitude of glucose excursions, mean of daily differences). RESULTS: Twenty-two subjects were treated using the Leuven protocol and 35 by the Yale protocol; >63,000 CGM measurements were available. The percentage of time in normoglycemia (80-110 mg/dl) was higher (37 ± 15 vs. 26 ± 11%, p = 0.001) and percentage of time spent in hypoglycemia was lower (0[0-2] vs. 5[1-8]%, p = 0.001) in the Yale group. Median glycemia did not differ between groups (118[108-128] vs. 128[106-154] mg/dl). Glycemic variability was less pronounced in the Yale group (median SD 28[21-37] vs. 47[31-71] mg/dl, p = 0.001; CV 23[19-31] vs. 36[26-50]%, p = 0.001; MODD 35[26-41] vs. 60[33-94] mg/dl, p = 0.001). However, logistic regression could not identify type of IIP, diabetes status, age, BMI, or APACHE-II score as independent parameters for strict glucose control. CONCLUSIONS: The Yale protocol provided better average glycemia, more time spent in normoglycemia, less time in hypoglycemia, and less glycemic variability than the Leuven protocol, but was not independently associated with strict glycemic control.

Randomized Evaluation of Glycemic Control in the Medical Intensive Care Unit Using Real-Time Continuous Glucose Monitoring (REGIMEN Trial).

BACKGROUND AND OBJECTIVE: Hyperglycemia occurs commonly in patients admitted to medical intensive care units (MICUs). Whether real-time (RT) continuous glucose monitoring (CGM) improves glycemic control and variability and reduces hypoglycemia in severely ill MICU patients with an Acute Physiology and Chronic Health Evaluation II (APACHE-II) score of ≥20 has not been studied. SUBJECTS AND METHODS: Thirty-five patients (66 ± 10 years of age; APACHE-II score, 28 ± 6) were randomly assigned to RT-CGM (n = 16) using the GlucoDay(®)S (A. Menarini Diagnostics, Florence, Italy) device or to blinded CGM. Insulin was infused using a modified Yale protocol targeting a blood glucose level between 80 and 120 mg/dL. Outcome measures were percentage of time in normoglycemia (80-110 mg/dL) and in hypoglycemia (<60 mg/dL), glycemic variability (SD, coefficient of variation, mean amplitude of glucose excursions, and mean of daily differences), and CGM accuracy (error grid analyses, Bland-Altman bias plot, and mean absolute relative deviation). RESULTS: During 96 h of monitoring, glycemia reached target (80-110 mg/dL) in 37 ± 15%, was between 70 and 180 mg/dL in 91 ± 10%, and <60 mg/dL in 2 ± 2% of the time. In the RT-CGM group glycemia averaged 119 ± 17 mg/dL versus 122 ± 11 mg/dL in the control group. Parameters of glucose variability and percentages of time at target glycemia and in hypoglycemia were similar between groups. GlucoDayS values and arterial glycemia correlated well, with 98.6% of data falling in Zones A and B of the error grid analysis. Mean absolute relative devation was 11.2%. CONCLUSIONS: RT-CGM did not ameliorate glucose control or variability; neither did it reduce the number of hypoglycemic events, but our insulin infusion protocol led to overall good glucose control without a significant hypoglycemia risk, making further improvement difficult.

Glucose control and use of continuous glucose monitoring in the intensive care unit: a critical review.

Stress hyperglycemia recently became a major therapeutic target in the Intensive Care Unit (ICU) since it occurs in most critically ill patients and is associated with adverse outcome, including increased mortality. Intensive insulin therapy to achieve normoglycemia may reduce mortality, morbidity and the length of ICU and in-hospital stay. However, obtaining normoglycemia requires extensive efforts from the medical staff, including frequent glucose monitoring and adjustment of insulin dose. Current insulin titration is based upon discrete glucose measurements, which may miss fast changes in glycemia and which does not give a full picture of overall glycemic control. Recent evidence suggests that continuous monitoring of glucose levels may help to signal glycemic excursions and eventually to optimize insulin titration in the ICU. In this review we will summarise monitoring and treatment strategies to achieve normoglycemia in the ICU, with special emphasis on the possible advantages of continuous glucose monitoring.

Beneficial effects of recombinant human activated protein C in a ewe model of septic shock.

OBJECTIVE: To investigate the effects of activated protein C (APC) in a clinically relevant animal model of septic shock. DESIGN: Prospective, randomized, controlled study. SETTING: University medical center research laboratory. SUBJECTS: Eighteen female sheep (body weight, 27-35 kg). INTERVENTIONS: Animals were fasted, anesthetized, invasively monitored, and mechanically ventilated before receiving 0.5 g/kg body weight of feces intraperitoneally to induce sepsis. Fluid resuscitation with Ringer lactate was titrated to maintain pulmonary artery occlusion pressure at baseline levels. No vasoactive agents or antibiotics were used. Two hours after the induction of sepsis, animals were randomized to receive an infusion of APC (24 microg x kg(-1) x hr(-1), n = 9) or an equivalent volume of vehicle (n = 9) throughout the experimental period. MEASUREMENTS AND MAIN RESULTS: The APC-treated animals had significantly higher arterial pressure, urine output, PaO2/FIO2 ratios, and thoracopulmonary compliance than the control animals. They had lower pulmonary arterial pressure and arterial lactate concentrations than the control animals. Plasma colloid oncotic pressure was better maintained in the APC-treated group than in the control group (p < .05). Prothrombin time and activated partial thromboplastin time were altered less, and plasma D-dimer concentrations were significantly lower in the APC-treated group than in the control group (p < .05). The blood protein C concentration and platelet count were maintained better in the APC-treated group than in the control group (p < .05). APC administration was associated with significantly longer survival (median, 27 hrs vs. 20 hrs; p < .05). At postmortem examination, the lung wet/dry ratio was significantly lower in the APC group than in the control group (6.3 +/- 0.7 vs. 7.1 +/- 1.2, p < .05). CONCLUSIONS: In this clinically relevant model of septic shock due to fecal peritonitis, administration of APC had beneficial effects on hemodynamic variables, gas exchange, lactic acidosis, and coagulation abnormalities. Higher colloid oncotic pressures and lower lung wet/dry ratios at autopsy suggest preserved endothelial integrity. APC administration resulted in prolonged survival.

Fluid resuscitation in severe sepsis and septic shock: albumin, hydroxyethyl starch, gelatin or ringer's lactate-does it really make a difference?

The aim of this study was to investigate whether the type of i.v. fluid administered has an impact on outcome in an animal model of septic shock. The study included 28 anesthetized, invasively monitored, mechanically ventilated female sheep (29.5 +/- 4.0 kg), which received 0.5 g/kg body weight of feces into the abdominal cavity to induce peritonitis. During the surgical operation and 4 h after feces spillage, only Ringer's lactate (RL) was administered in all animals. Thereafter, animals were randomized to receive continuous infusions of RL (n = 7) alone or combined with either 20% albumin (n = 7, volume ratio to RL 1:10) or 6% hydroxyethyl starch (HES) (n = 7, volume ratio to RL 1:1), or gelatin alone (n= 7, no volume limitation). Fluid resuscitation was titrated to maintain pulmonary artery occlusion pressure at baseline levels throughout the experiment. No antibiotics or vasoactive drugs were administered, and animals were monitored until their spontaneous death. Hemodynamic variables were better with HES and albumin than with the other fluids, as reflected by higher stroke volume, cardiac index, and oxygen delivery (all P < 0.05). Hydroxyethyl-starch-treated animals also had lower arterial lactate concentrations (P < 0.01). However, times to develop hypotension and oliguria were similar in all groups. Blood interleukin (IL) 6 concentrations were significantly increased in all groups. The mean survival time was similar in all groups. In this clinically relevant model of prolonged septic shock, albumin and HES solution resulted in higher cardiac output, oxygen delivery, and lower blood lactate levels than gelatin and RL; however, the choice of i.v. fluid did not affect outcome.

Intensive insulin therapy in the intensive care unit: assessment by continuous glucose monitoring.

OBJECTIVE: Hyperglycemia occurs in most critically ill patients. Using continuous glucose monitoring (CGM), we investigated whether intensive insulin therapy based on discontinuous glucose monitoring can achieve normoglycemia (80-110 mg/dl) in a medical intensive care unit (MICU). RESEARCH DESIGN AND METHODS: Fifty adults (men/women 31/19, age 62 +/- 16 years, nondiabetic/diabetic 30/20, intravenous/subcutaneous insulin 22/28, and Acute Physiology and Chronic Health Evaluation II score 22 +/- 7) were prospectively recruited. Forty-eight-hour CGM was performed using a subcutaneous glucose sensor (GlucoDay) and compared with arterial glycemia. Main outcome measures were percent of time in normoglycemia and accuracy/applicability of CGM. RESULTS: During 48-h CGM, glycemia reached target (80-110 mg/dl) in only 22 +/- 18%, was >140 mg/dl in 39 +/- 27%, and was <60 mg/dl in 5 +/- 10% of the time. Patients on subcutaneous versus intravenous insulin had more glycemia readings >110 mg/dl (P = 0.016). Glycemia was higher in diabetic patients (170 +/- 77 vs. 129 +/- 35 mg/dl, P = 0.013). BMI was an independent determinant for bad glycemic control (beta = 0.73, P < 0.0001). Diabetic state (beta = 0.47, P < 0.0001), septic shock (beta = 0.22, P = 0.045), sequential organ failure assessment score (beta = 0.40, P = 0.001), and use of corticoids (beta = 0.28, P = 0.014) and inotropics (beta = -0.24, P = 0.035) were independent determinants of insulin dose. GlucoDay values and arterial glycemia correlated well (r = 0.85, P < 0.0001, n = 555 after six-point calibration), with 97% of data falling in regions A and B of error grid analysis. There were no adverse events using GlucoDay. CONCLUSIONS: GlucoDay, a well-tolerated 48-h CGM system, revealed that normoglycemia was only achieved 22% of the time in MICU patients. Further studies should investigate whether application of CGM to titrate insulin therapy can improve patient outcome.

Blood warming during hemofiltration can improve hemodynamics and outcome in ovine septic shock.

BACKGROUND: This study was designed to evaluate the effects of blood warming during hemofiltration on global and regional hemodynamics, plasma lactate, and 24-h survival during septic shock. METHODS: Twenty anesthetized and mechanically ventilated sheep underwent induction of peritonitis and, 4 h later, were treated by hemofiltration with (n = 10) or without (n = 10) blood warming. RESULTS: In the group without blood warming, body temperature decreased after starting hemofiltration and remained below baseline. In the other animals, body temperature stabilized at baseline level during hemofiltration and increased to a maximum of 40.8 degrees C thereafter. The group without warming experienced a decrease in blood pressure (from 90 mmHg to 38 mmHg) and cardiac output (from 4.0 l/min to 2.3 l/min). Metabolic acidosis and the increase in lactate were less marked when temperature was maintained. None of the animals without warming but all of the animals with warming survived to 16 h. CONCLUSIONS: Differences in temperature during hemofiltration resulted in striking differences in hemodynamics, metabolic acidosis, and survival rate in this clinically relevant experimental model of septic shock.

High-volume hemofiltration in septic shock.

In the past decade we have learned a lot about the pathophysiology of septic shock. A lot of experimental research has been performed in vitro and in vivo, showing that hemofiltration can improve hemodynamics and survival. With modern machines, hemofiltration is becoming a sepsis treatment in patients.

Fever control in septic shock: beneficial or harmful?

The beneficial effects of interventions to control fever in sepsis are controversial. We investigated whether the use of acetaminophen and external cooling is beneficial to control fever in septic shock. We studied 24 fasted, anesthetized, invasively monitored, mechanically ventilated female sheep (27.0 +/- 4.6 kg) that received 0.5 g/kg body weight of feces into the abdominal cavity to induce sepsis. Ringer's lactate (RL) was titrated to maintain pulmonary artery occlusion pressure (PAOP) at baseline levels throughout the experimental period. During the 2 h after the surgical operation, animals were placed in the hypothermia group if their temperature fell below 36.0 degrees C; the other animals were randomized to three groups: high fever (T > 39.0 degrees C), mild fever (37.5 degrees C < T < 38.5 degrees C), and normothermia (36.0 degrees C < T < 37.0 degrees C). The administration of 25 mg/kg acetaminophen every 4 to 6 h combined with external cooling (ice pad) was used to control core temperature in these three groups. The PaO2/FiO2 ratio was higher and blood lactate concentration was lower in the high fever than in the other groups (P < 0.01 and 0.05, respectively). Survival time was longer in the high fever group (25.2 +/- 3.0 h) than in the mild fever (17.7 +/- 3.5 h), normothermia (16.0 +/- 1.9 h), and hypothermia (18.5 +/- 2.5 h) groups (P < 0.05 for all). Plasma heat shock protein (HSP) 70 levels were higher in the two fever groups than in the other groups (P < 0.05). In this clinically relevant septic shock model, the febrile response thus resulted in better respiratory function, lower blood lactate concentration, and prolonged survival time. Antipyretic interventions including acetaminophen and external cooling were associated with lower circulating HSP70 levels. These data challenge the temperature control practices often used routinely in acutely ill patients.

Low-dose vasopressin in the treatment of septic shock in sheep.

After induction of cecal perforation, 20 anesthetized sheep were randomized to be treated, when arterial blood pressure fell below 75 mm Hg, with vasopressin (fixed dose of 0.02 U/minute), norepinephrine (0.5-5 microg/kg/minute titrated to maintain mean arterial pressure between 75 and 85 mm Hg), vasopressin + norepinephrine (vasopressin at fixed dose 0.01 U/minute plus norepinephrine titrated as for norepinephrine only group), or no vasopressor (Ringer's lactate [control]). Mean arterial pressure was well maintained in all treatment groups. Superior mesenteric arterial blood flow was significantly lower in the vasopressin + norepinephrine group than in the vasopressin group. Vasopressin alone or combined with norepinephrine limited the increase in blood lactate concentration and ileal PCO2-gap compared with control and norepinephrine groups. Urine output was higher in the vasopressin group than in control and norepinephrine groups. Survival time was longer in the vasopressin (30 +/- 6 hours) and vasopressin + norepinephrine (30 +/- 3 hours) groups than in the norepinephrine group (20 +/- 1 hours, p < 0.05) and in all treatment groups than in the control group (17 +/- 2 hours, p < 0.05). Tissue injury was less severe in the vasopressin and vasopressin + norepinephrine groups than in the others. In this clinically relevant model of septic shock due to peritonitis, vasopressin administration (alone or with norepinephrine) can prolong survival.

Comparison of polyacrylonitrile (AN69) and polysulphone membrane during hemofiltration in canine endotoxic shock.

OBJECTIVE: This study was designed to compare the effects of continuous venovenous hemofiltration (CVVH) with two different membranes, polysulphone and polyacrylonitrile (AN69), on global and regional hemodynamics, plasma lactate, tumor necrosis factor-alpha levels, and plasma nitrite/nitrate during endotoxic shock in dogs. METHODS: Fifteen pentobarbital anesthetized and mechanically ventilated dogs were randomized into three groups of five dogs each. One group served as an endotoxin alone, time matching group and, 1 hr after endotoxin administration, the two other groups received CVVH at 3 L/hr for 270 mins, with either a polysulphone membrane or an polyacrylonitrile membrane. RESULTS: At 90 mins after endotoxin administration, dogs receiving CVVH with polyacrylonitrile membranes had a higher cardiac output, stroke volume, and left-ventricular stroke work index than the endotoxin alone and the polysulphone groups. CVVH with either polyacrylonitrile or polysulphone membranes prevented the rise in pulmonary artery pressure and pulmonary vascular resistance compared with the endotoxin alone group. Plasma lactate levels were not significantly altered, but the fall in bicarbonate seen in the endotoxin alone group did not occur in the two CVVH groups. Tumor necrosis factor levels in the plasma were not significantly altered by CVVH and remained very low (<50 pg/mL) in the ultrafiltrate fluid. CONCLUSION: In this acute endotoxic shock model, CVVH with the polyacrylonitrile membrane improved cardiac performance when compared with the polysulphone membrane. These effects could be caused by a more effective adsorption of inflammatory mediators other than tumor necrosis factor. Whether the polyacrylonitrile membrane should be preferred over the polysulphone membrane for CVVH in severe sepsis warrants further experimental and clinical study.

Optimal adrenergic support in septic shock due to peritonitis.

BACKGROUND: The authors evaluated optimal adrenergic support using norepinephrine, dopamine, and dobutamine in a clinically relevant model of septic shock. METHODS: Twenty-eight mature, female, anesthetized sheep (weight, 30.5 +/- 3.6 kg) underwent cecal ligation and perforation and were randomized into four groups of seven animals to be treated with norepinephrine, dopamine-norepinephrine, dobutamine-norepinephrine, or no adrenergic agent. In all groups, lactated Ringer's solution was administered to restore cardiac filling pressures to baseline. In the norepinephrine group, norepinephrine (0.5-5 microg. kg(-1). min(-1)) was titrated to maintain mean arterial pressure between 75-85 mmHg. In the dopamine-norepinephrine group, dopamine was given first, and norepinephrine was added only when mean arterial pressure remained below 75 mmHg despite the infusion of 20 microg. kg(-1). min(-1) dopamine. In the dobutamine-norepinephrine group, dobutamine was started at the same time as norepinephrine and titrated up to 20 microg. kg(-1). min(-1) to get a 15% increase in cardiac output. RESULTS: The dobutamine-norepinephrine group had greater cardiac output; superior mesenteric blood flow, oxygen delivery (Do(2)), and oxygen consumption ([OV0312]o(2)); and lower blood lactate concentration and partial pressure of carbon dioxide (Pco(2)) gap than the controls did. Cumulative urine output was significantly higher in the dobutamine-norepinephrine group than in the other groups. Survival time was significantly longer in the dobutamine-norepinephrine (24 +/- 4 h), dopamine- norepinephrine (24 +/- 6 h), and norepinephrine (20 +/- 1 h) groups than the control group (17 +/- 2 h; P < 0.05 vs. other groups), and significantly longer in the combined dopamine-norepinephrine and dobutamine-norepinephrine groups (24 +/- 5 h) than in the norepinephrine alone group (P < 0.05). Histologic examination of lung biopsies revealed less severe lesions in the dobutamine-norepinephrine group than in the control and norepinephrine alone groups. Anatomic alterations in the lung, liver, and small intestine were less severe in the dobutamine-norepinephrine group than in the other groups. CONCLUSIONS: In this prolonged septic shock model, association of norepinephrine with either dopamine or dobutamine resulted in the longest survival and the least severe pulmonary lesions. The combination of dobutamine with norepinephrine was associated with a better myocardial performance, greater Do(2) and [OV0312]o(2), lower blood lactate concentration and Pco(2) gap, and less anatomic injury.

Animal models of acute tubular necrosis.

The evaluation and management of acute renal failure in the ICU patient remains a formidable task because of the complexity of this condition. Clinical and physiologic assessment and complementing laboratory and imaging tests are currently insufficient to differ between true renal parenchymal damage (acute tubular necrosis; it is important to realize that this term does not necessarily imply widespread injury, because whole organ dysfunction in humans has often been associated with very limited parenchymal cellular necrosis) and prerenal azotemia (decreased renal blood flow with altered glomerular hemodynamics and subsequently diminished glomerular filtration, without significant epithelial cell injury). Moreover, tubular damage and altered glomerular hemodynamics may coexist or lead to each other, and their relative contribution to the evolving renal dysfunction has not been unequivocally established. The limited data regarding the renal pathology of such patients and the scant information about human morphologic and functional correlates further undermine our knowledge about diagnostic and therapeutic approaches to these patients. Advanced techniques are critically needed to establish noninvasively the dynamic status of renal parenchymal microcirculation and the distribution of intrarenal oxygenation and to identify evolving cellular energy depletion and tubular cell damage. A few technologies are potentially promising, such as blood oxygen level dependent magnetic resonance imaging, positron emission tomography, and kidney injury molecule-1 detection in patients' urine. Because of the difficulties in analyzing the pathophysiology in humans, clinicians continue to rely largely on animal models to guide understanding and rationale for the identification of therapeutic targets. Data from such animal studies are complemented by studies in isolated perfused kidneys, isolated tubules, and tubular epithelial cell cultures. In this report, we summarize some concepts of acute tubular necrosis that have evolved as a result of these studies, evaluate available animal models, and underscore controversies regarding experimental acute tubular necrosis.

Comparison of continuous thermodilution and bolus cardiac output measurements in septic shock.

OBJECTIVE: To compare continuous (CCO) and bolus (BCO) thermodilution cardiac output measurement techniques over a wide range of cardiac outputs and blood temperatures in a septic sheep model. DESIGN AND SETTING: Prospective experimental study in a university intensive care laboratory. SUBJECTS: Thirty-five anesthetized sheep. INTERVENTIONS: Pulmonary artery catheters allowing measurement of CCO and BCO were placed through the external jugular vein. Cecal ligation and perforation was performed to induce septic shock. In 14 sheep two femoral venous catheters were placed and connected to a hemofiltration system to alter blood temperature. MEASUREMENTS: CCO and BCO were measured every hour during the experiment. Three 10-ml bolus injections of iced normal saline were given through a closed injectate system and then averaged. The CCO readings were collected just before the BCO measurements. The relationship between CCO and BCO was assessed using Bland and Altman's method. RESULTS: In 465 paired data the temperature ranged between 34.0 degrees and 40.9 degrees C, CCO between 1.4 and 17.0 l/min, and BCO between 1.1 and 17.4 l/min. There was a highly significant correlation between CCO and BCO ( r=0.97). The bias (difference between CCO and BCO) was -0.19 l/min, the SD of the difference 0.45 l/min, and the limits of agreement -1.08/0.71 l/min. There were also highly significant correlations between CCO and BCO at the different temperatures (extreme values: 34.0-34.9 degrees C, r=0.90; 40.0-40.9 degrees C, r=0.98). CONCLUSIONS: Thermodilution measurements of CCO are reliable, when compared to BCO measurements, over a large range of cardiac outputs and blood temperatures.