Sublingual capnometry and indexes of tissue perfusion in patients with circulatory failure.

OBJECTIVE: To examine the relationship between sublingual PCO(2) (PslCO(2)) and other indexes of tissue perfusion. DESIGN: Prospective observational study. SETTING: Medical and coronary ICUs in a tertiary-care teaching hospital. SUBJECTS: Twenty-five patients with circulatory failure, 19 patients with sepsis, and 6 patients with cardiac failure. MEASUREMENTS AND MAIN RESULTS: PslCO(2), gastric intramucosal PCO(2) (PiCO(2)), arterial lactate concentration, systemic oxygen delivery, and systemic oxygen consumption were measured at baseline and at 1, 3, 6, 12, and 24 h after the beginning of the study. PslCO(2) and the PslCO(2)-PaCO(2) gradient were increased but not significantly different in nonsurvivors compared to survivors at baseline. At 24 h, the mean (+/- SE) PslCO(2) was 45 +/- 4 mm Hg in survivors and 61 +/- 4 mm Hg in nonsurvivors (p = 0.06), while the PslCO(2)-PaCO(2) gradient was 14 +/- 3 mm Hg in survivors and 29 +/- 4 mm Hg in nonsurvivors (p < 0.05). No other significant differences in survivors and nonsurvivors were observed in any other index of perfusion. For all patients, the correlations between PslCO(2) and PiCO(2) (r = 0.459; p < 0.05) and cardiac index (r = 0.285; p < 0.05) were observed. The PslCO(2)-PaCO(2) gradient also was correlated with the PiCO(2)-PaCO(2) gradient (r = 0.323; p < 0.05). When patients were placed into subsets of sepsis and cardiac failure, the strength of the correlations increased in the patients with cardiac failure (PslCO(2) vs lactate, r = 0.611 and p < 0.05; PslCO(2) vs PiCO(2), r = 0.613 and p < 0.05; PslCO(2) vs PiCO(2)-PaCO(2) gradient, r = 0.648 and p < 0.05). CONCLUSION: PslCO(2) correlated best with PiCO(2) and arterial lactate concentration in patients with cardiac failure. PslCO(2) and the PslCO(2)-PaCO(2) gradient may be useful as indexes of the severity of perfusion failure.

Monophosphoryl lipid A stimulated up-regulation of reactive oxygen intermediates in human monocytes in vitro.

The production of reactive oxygen and nitrogen intermediates is a common response to infectious challenge in vivo. These agents have been implicated in the modulation of cytokine responses and are produced in large amounts in response to endotoxins produced by a number of infectious agents. The antigen-presenting cell activation caused by these lipopolysacchardies (LPS) has been exploited in the use of these agents as adjuvants. In recent years, less-toxic derivatives have been sought. One such agent, monophosphoryl lipid A (MPL), has been used increasingly in vivo as an adjuvant and as a modulator of the inflammatory process. It is known that this agent modulates the inflammatory response and cytokine production. In addition, we have shown its effect on the production of reactive nitrogen intermediates. In this paper, we show that MPL stimulates the release of high levels of superoxide (O(2)(-)) and hydrogen peroxide (H(2)O(2)), the latter being greater than that seen with LPS and appearing to be related to the inability of MPL to stimulate catalase activity. When cells were pretreated with LPS or MPL and subsequently challenged with LPS, the production of O(2)(-) and H(2)O(2) was inhibited significantly by LPS and MPL. The concentration of MPL required to induce significant hyporesponsiveness to subsequent LPS challenge was 10 times lower than that of LPS. Hyporesponsiveness was greatest when induced by 10 microg/ml MPL, the same concentration that induced the maximum release of H(2)O(2) in primary stimulation. In addition, we have shown that following MPL pretreatment, LPS stimulation does not cause the loss of cytoplasmic IkappaBalpha, which occurs when human monocytes are cultured with LPS. From our results, we propose a model for the reduced toxicity of MPL.

Role of interleukin-10 in monocyte hyporesponsiveness associated with septic shock.

OBJECTIVES: The purpose of this study was to examine the pattern of tumor necrosis factor (TNF)-alpha and interleukin (IL)-10 release in endotoxin-stimulated septic monocytes and to determine the role of IL-10 and transforming growth factor (TGF)-beta in monocyte hyporesponsiveness during septic shock. DESIGN: Monocytes isolated from ten healthy controls and ten patients with septic shock were incubated with endotoxin and cytokine release was assessed. Next, normal monocytes were incubated with either normal or septic serum and stimulated with endotoxin. Finally, normal monocytes were incubated with septic serum either with anti-IL-10 antibodies or anti-TGF-beta antibodies and then stimulated with endotoxin. MEASUREMENTS: TNF-alpha, IL-10, and TGF-beta levels were measured in the serum and in culture supernatants by enzyme-linked immunosorbent assay. SETTING: Research laboratory. MAIN RESULTS: IL-10 and TNF-alpha levels were significantly increased in septic serum, whereas TGF-beta levels were not different from controls. Normal monocytes increased TNF-alpha and IL-10 release in response to endotoxin. In contrast, septic monocyte TNF-alpha release was attenuated in response to endotoxin (1.8 +/- 0.5 vs. 1.0 +/- 0.4 ng/mL, stimulated vs. baseline), whereas IL-10 release increased significantly from baseline (173 +/- 91 vs. 8 +/- 4 pg/mL, stimulated vs. baseline). Incubation of normal monocytes with septic serum attenuated TNF-alpha release in response to endotoxin (32% +/- 8% of normal serum; p < .01), whereas IL-10 release was increased (285% +/- 84% of normal serum; p < .05). When normal monocytes were incubated with septic serum combined with anti-IL-10 antibodies, TNF-alpha release increased significantly to 75% +/- 17% of normal serum (p < .05 vs. septic serum alone). Incubation of normal monocytes with anti-TGF-beta antibodies did not significantly affect either TNF-alpha or IL-10 release in response to endotoxin. CONCLUSION: Monocytes from patients with septic shock exhibit persistent IL-10 release at a time when TNF-alpha release is downregulated. The continued release of IL-10 may contribute to impairment of monocyte proinflammatory cytokine release and the development of immune dysfunction in septic shock.

Effects of perfusion pressure on tissue perfusion in septic shock.

OBJECTIVE: To measure the effects of increasing mean arterial pressure (MAP) on systemic oxygen metabolism and regional tissue perfusion in septic shock. DESIGN: Prospective study. SETTING: Medical and surgical intensive care units of a tertiary care teaching hospital. PATIENTS: Ten patients with the diagnosis of septic shock who required pressor agents to maintain a MAP > or = 60 mm Hg after fluid resuscitation to a pulmonary artery occlusion pressure (PAOP) > or = 12 mm Hg. INTERVENTIONS: Norepinephrine was titrated to MAPs of 65, 75, and 85 mm Hg in 10 patients with septic shock. MEASUREMENTS AND MAIN RESULTS: At each level of MAP, hemodynamic parameters (heart rate, PAOP, cardiac index, left ventricular stroke work index, and systemic vascular resistance index), metabolic parameters (oxygen delivery, oxygen consumption, arterial lactate), and regional perfusion parameters (gastric mucosal Pco2, skin capillary blood flow and red blood cell velocity, urine output) were measured. Increasing the MAP from 65 to 85 mm Hg with norepinephrine resulted in increases in cardiac index from 4.7+/-0.5 L/min/m2 to 5.5+/-0.6 L/min/m2 (p < 0.03). Arterial lactate was 3.1+/-0.9 mEq/L at a MAP of 65 mm Hg and 3.0+/-0.9 mEq/L at 85 mm Hg (NS). The gradient between arterial P(CO2) and gastric intramucosal Pco2 was 13+/-3 mm Hg (1.7+/-0.4 kPa) at a MAP of 65 mm Hg and 16+/-3 at 85 mm Hg (2.1+/-0.4 kPa) (NS). Urine output at 65 mm Hg was 49+/-18 mL/hr and was 43+/-13 mL/hr at 85 mm Hg (NS). As the MAP was raised, there were no significant changes in skin capillary blood flow or red blood cell velocity. CONCLUSIONS: Increasing the MAP from 65 mm Hg to 85 mm Hg with norepinephrine does not significantly affect systemic oxygen metabolism, skin microcirculatory blood flow, urine output, or splanchnic perfusion.

Microvascular response in patients with cardiogenic shock.

OBJECTIVE: To examine the mechanisms contributing to decreased microvascular blood flow in cardiogenic shock by comparing patients with cardiogenic shock with critically ill controls and with patients with septic shock. DESIGN: Prospective, consecutive entry of patients meeting the criteria for septic shock, cardiogenic shock, and critical illness without coexisting infection or shock. SETTING: University hospital, medical intensive care unit, coronary care unit, and respiratory care unit. PATIENTS: Eight patients with cardiogenic shock secondary to acute myocardial infarction, six critically ill controls, and six patients with septic shock. MEASUREMENTS AND MAIN RESULTS: Forearm blood flow was measured at rest and during reactive hyperemia by venous air plethysmography. Red cell deformability was determined by filtration. Leukocyte aggregation was detected by the leukergy test. Neutrophil CD11b/CD18 expression and soluble intercellular adhesion molecule-1 levels were also measured. In cardiogenic shock, forearm arterial resistance was significantly increased at rest and during reactive hyperemia compared with controls and patients with septic shock. The response to reactive hyperemia was attenuated in cardiogenic and septic shock patients, as measured by the absolute change in forearm blood flow from baseline, which was significantly less as compared with controls (p < .01). The percent change in forearm blood flow during reactive hyperemia compared with forearm blood flow at rest was significantly lower in cardiogenic shock (60+/-10) and in septic shock (50+/-11) compared with controls at baseline (145+/-20; p < .01). Red cell deformability was significantly decreased in cardiogenic shock (1.2+/-0.2 mL/min; p < .05) and septic shock (1.1+/-0.2 mL/min; p < .05), compared with controls (1.8+/-0.1 mL/min). Neutrophil CD11b/CD18 expression, leukergy, and serum intercellular adhesion molecule-1 levels in cardiogenic shock patients were not significantly different from controls. CONCLUSION: These data suggest that the response to reactive hyperemia is attenuated in cardiogenic shock. This appears to reflect increased vasoconstriction and an impaired capacity for vasodilation. Decreased erythrocyte deformability may also be important in limiting systemic microvascular flow. However, evidence supporting a role for neutrophil-endothelial cell interactions was not observed.

The first decade of the American Board of Internal Medicine certification in critical care medicine: an overview of examinees and certificate holders from 1987 through 1996.

OVERVIEW: This study reviews the first decade of critical care medicine (CCM) certification by the American Board of Internal Medicine (1987-1996). Included are the characteristics of examinee and certificate-holder groups; examination performances from different underlying disciplines of internal medicine, with or without formal CCM training; and the influence of background and a training program as correlates of examination performance. DATA SOURCES: The CCM certification examination has been offered biennially since November 1987. Performance data on the American Board of Internal Medicine examinations in internal medicine and its subspecialties and added qualifications were available for candidates taking the CCM examinations. For examinees with formal CCM training, residency program director ratings, and information regarding the program characteristics of size and percentage of United States and Canadian medical graduates were also available. STUDY SELECTION: All examinees who ever attempted certification were included in this study. The study cohort for each of the five examination administrations consists of all first-time takers. CONCLUSIONS: Cohort sizes have decreased since formal training became an admission requirement in 1993. Percentages of International Medical Graduates and women attempting and achieving certification have increased steadily. Examination performance was positively associated with formal training, internal medicine examination performance, recent medical training, and pulmonary disease certification. For those with formal training, performance was also positively associated with training program director ratings of overall clinical competence and completion of a training program with a higher proportion of United States and Canadian medical graduates.

Influence of rheologic changes and platelet-neutrophil interactions on cell filtration in sepsis.

We examined the role of erythrocyte (red blood cell; RBC) aggregation and deformability, neutrophil (polymorphonuclear neutrophil; PMN) deformability, whole-blood viscosity, and platelet-neutrophil interactions on cell filtration in subjects who were critically ill with sepsis (CIS), critically ill noninfected subjects (CINS), and healthy controls (C). We assessed cell deformability by filtration through filters of 5-microm pore size. Whole blood, RBC, PMN, and combinations of PMN and RBC were studied. Viscometry was done on isolated RBC. Platelet-PMN interactions were assessed with monoclonal antibodies to CD41 and activated CD63 platelet receptors, and to CD66b PMN receptors. Filtration pressure (Pi) for CIS was significantly greater than for C and CINS at both high and low PMN and RBC concentrations. Viscometry confirmed decreases in RBC deformability and demonstrated significant increases in RBC aggregation in CIS. Increments in Pi were significantly greater with PMN and PMN-RBC combinations suspended in platelet rich plasma (PRP) than in platelet poor plasma (PPP) for CIS as compared with CINS or C. Flow cytometry confirmed significantly greater platelet activation in CIS than in CINS or C (mean fluorescence: 39 +/- 9 lfu versus 18.7 +/- 4.0 lfu and 17.1 +/- 2.3 lfu, respectively) and greater platelet-PMN aggregation (mean fluorescence: 44.7 +/- 3.6 lfu versus 23 +/- 4.1 lfu, respectively) in CIS than in C. We conclude that decreased filtration of whole blood in CIS is related to decreases in RBC and PMN deformability, increases in RBC aggregation, and increased platelet-PMN interactions. Of these, the formation of platelet-PMN aggregates appeares to have the greatest effect in impairing cell filtration. These rheologic abnormalities may contribute to impaired microvascular blood flow in patients with sepsis.

Decreased response to recall antigens is associated with depressed costimulatory receptor expression in septic critically ill patients.

Anti-inflammatory substances are released during septic shock that modulate monocyte function. Decreased monocyte responsiveness to bacterial toxins and decreased expression of human-leukocyte-associated antigen-DR (HLA-DR) have been reported during septic shock and critical illness. Impaired antigen presentation has been inferred from these observations but has not been demonstrated. We assessed antigen presentation and costimulatory molecule expression in 12 age-matched control subjects, 10 noninfected critically ill patients (CINS), and 17 critically ill patients with sepsis (CIS). Antigen presentation was assessed by using in vitro lymphocyte 5-bromo-2-deoxyuridine (BrdU) incorporation in response to tetanus toxoid. The expression of HLA-DR and the costimulatory molecules CD28, CD86, and CTLA-4 was assessed by flow cytometry. Serum interleukin-10 (IL-10) was also measured by enzyme-linked immunosorbent assay. Serum IL-10 levels were significantly elevated in CIS patients (91 +/- 38 pg/mL) as compared with levels in control subjects (5 +/- 4 pg/mL)(P < .05). Lymphocyte BrdU incorporation increased by 710% +/- 243% in control subjects but by only 144% +/- 62% in CIS patients and 76% +/- 31% in CINS patients (P < .01 vs control). Monocyte HLA-DR expression, monocyte CD86 expression, and lymphocyte CD28 expression were significantly decreased in CIS patients (P < .01) as compared with control subjects. Conversely, lymphocyte CTLA-4 expression was significantly increased in CIS patients (P < .05 vs control). Monocyte CD86 expression was also significantly decreased in CINS patients as compared with control subjects. These data indicate that antigen presentation is decreased in critically ill patients with sepsis. This appears in part related to decreased expression of HLA-DR and the costimulatory molecules CD86 and CD28. Increased expression of the negative signal receptor CTLA-4 may also impair antigen presentation in patients with sepsis.

Effects of proinflammatory cytokines and bacterial toxins on neutrophil rheologic properties.

OBJECTIVE: To examine the changes in neutrophil deformability, aggregation, and adherence in response to stimulation with proinflammatory cytokines and bacterial toxins. DESIGN: Prospective, randomized trial. SETTING: Research laboratory. SUBJECTS: Neutrophils isolated from healthy volunteers. INTERVENTIONS: Neutrophils were exposed to tumor necrosis factor (TNF)-alpha, interleukin (IL)-1beta, IL-8, their combination, endotoxin (LPS), lipoteichoic acid (LTA), and staphyloccocal enterotoxin B (SEB). Neutrophil deformability was measured as percent neutrophils filtered through 5-microm diameter filters. Aggregation was measured using a platelet aggregometer. Adherence was determined by examining the binding of neutrophils to albumin-coated latex beads. MEASUREMENTS AND MAIN RESULTS: Exposure to TNF-alpha and IL-1beta led to significant decreases in neutrophil filterability, which was attenuated by cytochalasin D pretreatment. LPS and LTA also decreased deformability, suggesting that these toxins directly stimulated neutrophils independent of cytokines. IL-8 and SEB did not significantly affect neutrophil deformability. TNF-alpha and LPS were associated with significant neutrophil aggregation, which was inhibited by pretreatment with anti-CD18 antibodies. Neutrophil aggregation was not affected by IL-1beta, LTA, or SEB. TNF-alpha, IL-8, and LPS increased neutrophil adherence, which also was attenuated by pretreatment with anti-CD18 antibodies. IL-1beta, LTA, and SEB did not significantly affect neutrophil adherence. CONCLUSIONS: Cytokines and bacterial toxins differ in their effects on neutrophil deformability, aggregation, and adherence. Of the cytokines examined, TNF-alpha appears to have the greatest direct effects on neutrophil rheology. Similarly, endotoxin appears to have greater direct effects on neutrophil rheology than the Gram-positive bacterial toxins, LTA, and staphylococcal enterotoxins.

Monophosphoryl lipid A stimulated up-regulation of nitric oxide synthase and nitric oxide release by human monocytes in vitro.

Monophosphoryl lipid A (MPL) is a derivative of lipopolysaccharide (LPS) with reduced toxicity which has been shown to modulate various immune functions in monocytes. We examined whether human monocytes can be stimulated to produce nitric oxide (NO) and its catalytic enzyme nitric oxide synthase (NOS). Monocytes were stimulated with LPS or MPL and both NOS and NO (as nitrite) production were measured. MPL at high doses (> 100 micrograms/ml) stimulated monocytes to release NO that was significantly greater than both the control and LPS-treated monocytes (p < 0.05). NO release by control cells and the LPS treated cells was not significantly different. Both arginase and N-monomethyl arginine (NMLA) inhibited the MPL stimulated release of NO (p < 0.01). MPL significantly increased inducible NOS (iNOS) expression as measured by both fluorescent microscopy and flow cytometry (p < 0.05). Similarly, both soluble NOS (sNOS) and particulate NOS (pNOS) activity were significantly up-regulated by MPL (p < 0.05). Significant correlations were found between pNOS expression and sNOS release (r = 0.72, p < 0.0001) and between 12 h NO release and sNOS production (r = 0.44, p < 0.005). These experiments confirm that human monocytes can be stimulated with MPL to produce NO in vitro and suggest that up-regulation of pNOS does not preclude NO release.

Effect of crystalloid and colloid solutions on blood rheology in sepsis.

Sepsis is associated with altered blood rheology. Fluid infusion is an essential component of therapy for septic shock. The purpose of this study was to compare the rheologic changes associated with saline, albumin, and hydroxyethyl starch in sepsis. Whole blood was obtained from five normal controls and five patients with severe sepsis. The samples were centrifuged, and the erythrocytes were resuspended in autologous plasma or autologous plasma plus the buffy coat at an hematocrit (Hct) of 40%. The sample was diluted to an Hct of 30%, 20%, and 10% with saline, albumin, or hydroxyethyl starch. Viscosity was measured at low and high shear rates and erythrocyte aggregation was measured by the ratio of viscosity at low to high shear rates. Erythrocyte deformability was assessed by filtration. The viscosity of hydroxyethyl starch was greater than saline, albumin, or autologous plasma (p < .01). Erythrocyte viscosity was greater (p < .01) and deformability less (p < .01) in septic blood compared with normals. Dilution with hydroxyethyl starch increased erythrocyte viscosity as compared with saline (p < .01) and albumin (p < .01). Erythrocyte deformability was decreased with both hydroxyethyl starch (p < .001) and albumin (p < .05) compared with saline. Increased erythrocyte aggregation was also observed with hydroxyethyl starch (p < .05) and albumin (NS) in septic cells when compared with saline. These data indicate that hydroxyethyl starch increases blood viscosity, decreases erythrocyte deformability, and increases erythrocyte aggregation when compared with saline. These changes are less significant with albumin. In patients with sepsis, these effects may further compromise the already altered erythrocyte rheology.

Neutrophil rheologic changes in septic shock.

Decreases in effective capillary blood flow during septic shock may be related to changes in neutrophil rheology which contribute to microvascular occlusion. The purpose of this study was to examine neutrophil deformability, adherence, and aggregation in patients with severe sepsis and septic shock. Neutrophils were isolated from six patients with septic shock (SS), 12 patients with severe sepsis (S), six noninfected critically ill patients (CINS), and nine normal volunteers (N). Neutrophil deformability was determined by examining filtration through 5-microm filters. Neutrophil aggregation was measured by aggregometry and leukergy. Neutrophil adherence was examined by assessing the binding of latex beads to neutrophils. Patients with S and SS demonstrated decreased neutrophil filterability of 27 +/- 2% and 16 +/- 5%, respectively (p < 0.01), in comparison with N subjects, 55 +/- 4% and CINS patients, 58 +/- 2%. Preincubation of neutrophils from S and SS patients with cytochalasin D significantly increased the percent filtration of neutrophils. Neutrophil aggregation, measured by aggregometry, was increased in SS patients, 16 +/- 4% (p < 0.01) compared with N subjects, 1 +/- 0.2% and CINS patients, 1 +/- 0.2%. Incubation of neutrophils of SS patients with anti-CD11/CD18 significantly increased the filtration of isolated neutrophils to 46 +/- 3% (p < 0.01) and decreased aggregation to 7 +/- 2%. Neutrophil adherence was not increased in S or SS patients. These observations suggest that neutrophil deformability is decreased in patients with S and SS. Increased leukoaggregation may also contribute to decreased filterability of neutrophils in SS patients. These mechanisms may play a role in impaired microvascular flow in septic shock.

Non-ICU care of hemodynamically stable mechanically ventilated patients.

OBJECTIVE: To analyze a 4 1/2-year experience caring for hemodynamically stable mechanically ventilated patients on a nonmonitored respiratory care floor (RCF) for therapeutic outcome, utilization, and costs. DESIGN: A retrospective medical records review. SETTING: ICUs and an RCF of a university-affiliated tertiary care center. PARTICIPANTS: Two hundred twenty-four patients requiring more than 24 h of mechanical ventilation cared for on the RCF. RESULTS: The mean age of patients was 67 +/- 17 years. Of the admissions, 58% were from the medical ICU, 28% were from surgical ICUs, and 9.4% were from general medical floors. Patients spent 50 +/- 66 days mechanically ventilated on the RCF. Overall survival was 50.4% with 93.8% of surviving patients successfully weaned from mechanical ventilation. Survival by diagnostic group demonstrated highest probability of survival in patients with trauma and lowest in patients with multisystem failure. Of the survivors, 39% were discharged home, 34% to a rehabilitation unit, and 24% to a skilled nursing facility. Savings based on differential of costs between the ICU and RCF, primarily from reduced staffing requirements, were estimated at $4.1 million. CONCLUSION: Use of a nonmonitored RCF for the care of hemodynamically stable mechanically ventilated patients yields acceptable therapeutic outcomes while providing the institution with increased flexibility in critical care bed management and significant financial savings.