The syndrome of sepsis is associated with a high mortality rate, ranging from 27% in patients with sepsis and greater than 50% in those progressing to septic shock (1). Despite improved intensive care therapies, prognosis has improved little during the past decades. The microcirculation, which is a complex network of resistance vessels, plays an important role in the pathophysiology of the ongoing insult in sepsis (2-4). In these resistance vessels, perfusion is dependent on various factors, such as arterial oxygen saturation, oxygen consumption, blood viscosity, and red and white blood cell deformability (5, 6). Many of the abovementioned components of the microcirculation are affected in sepsis, resulting in a regional mismatch of oxygen supply and demand. Flow is shunted away from the susceptible microcirculation (7) as injured endothelial cells fail to relax and release potent vasoconstrictors with ensuing cellular swelling. Furthermore, permeability is increased, and leukocytes and platelets are trapped in the microcirculation resulting in progressive loss of perfusion and further organ injury.
Over the past years, with the use of new techniques, the human microcirculation has been examined in a large variety of clinical settings. One of the most striking findings was the pathological heterogeneity of microcirculatory flow. Some vascular beds show a preserved functional capillary density, whereas others have a sluggish blood flow, and some display no flow at all. Near-infrared spectroscopy (NIRS) (8-12) and sidestream dark-field (SDF) imaging (2, 3, 13-15) are noninvasive bedside techniques for continuous, real-time monitoring of tissue oxygenation and microcirculatory flow, whereas forearm plethysmography has been used to investigate vascular reactivity in the resistance vessels of the forearm (16-19). Endothelial dysfunction has been demonstrated by attenuated vasodilatory response to the infusion of acetylcholine into the brachial artery, and venous occlusion plethysmography measurements of forearm blood flow (FBF), in various vascular diseases and sepsis (20).
Endotoxin administration to humans is a model to study the pathophysiology of inflammation, coagulation, and cardiovascular effects of sepsis. We previously demonstrated that LPS administrations on 5 consecutive days resulted in development of LPS tolerance (21). The present study was undertaken to determine the effect of endotoxin administration and the development of LPS tolerance on the microcirculation and vascular reactivity measured by three different methods of approach. In this study, we measured the baseline microvascular flow (SDF imaging), in addition to endothelium-dependent ischemia-mediated tissue oxygenation (NIRS) and pharmacologically induced endothelium-dependent vasodilatory response to acetylcholine (venous occlusion plethysmography) during endotoxemia and after the development of LPS tolerance.
MATERIALS AND METHODS
The study protocol was approved by the Ethics Committee of the Radboud University Nijmegen Medical Centre and complies with the Declaration of Helsinki including current revisions and the Good Clinical Practice guidelines. Nine healthy male volunteers gave written informed consent to participate in the experiments, which were part of a larger endotoxin tolerance trial (NCT00246714). Subjects taking prescription drugs were excluded. Screening of the subjects before the experiment revealed no abnormalities in medical history and physical examination. No abnormalities were seen in the routine laboratory tests and electrocardiogram. All subjects tested negative to HIV and hepatitis B. Eight hours before the experiment, subjects refrained from beverages and food.
US reference E. coli endotoxin (lot Ec-5; Centre for Biologic Evaluation and Research, Food and Drug Administration, Bethesda, Md) was used in this study. Ec-5 endotoxin, supplied as a lipophilized powder, was reconstituted in 5 mL saline 0.9% for injection and vortex mixed for at least 5 min after reconstitution. The endotoxin solution was administered as a single intravenous bolus injection during 1 min at t = 0.
LPS 2 ng · kg−1 · d−1 was administered intravenously on 5 consecutive days to induce endotoxin tolerance. Before every LPS administration, the subjects were prehydrated with 1,500 mL glucose/saline infusion (22), and a continuous intravenous drip was started at 150 mL · h−1 for 6 h after LPS administration. Symptom scores, temperature (tympanic thermometer, °C), heart rate (electrocardiogram), blood pressure (brachial artery), and circulating cytokine levels were obtained before LPS administration and serially during 6 h after the administration of LPS on days 1 and 5. The symptom score consisted of backache, shivering, nausea, headache, and muscle ache. The volunteers were asked to score each of these complaints ranging from "not present" (score 0) up to "most severe ever" (score 5). Each time point represents a cumulative symptom score. TNF-α, IL-6, and IL-10 were measured by use of the Luminex Assay (Bioplex-kits; BioRad Laboratories, Hercules, Calif) in accordance with the manufacturer's instructions. IL-1ra levels were determined using the enzyme-linked immunosorbent assay duoset of R&D Systems (Minneapolis, Minn). The lower detection limits were TNF-α: 36 pg · mL−1, IL-6: 8 pg · mL−1, IL-10: 8 pg · mL−1, and IL-1ra: 0.2 pg · mL−1. Before and after LPS administration on days 1 and 5, microvascular changes and vascular reactivity were measured with NIRS, SDF imaging (at t = 0, 2, and 4 h, n = 6), and forearm plethysmography (t = 0 and 4 h).
Tissue oxygenation measurements protocol
The Beer-Lambert law states that the transmission of light through a solution of a colored compound decreases exponentially as the concentration of the compound increases (23). Near-infrared light transmits through tissues such as skin, adipose tissue, and muscle without attenuation and allows for detection of changes in specific light-absorbing chromophores (e.g., hemoglobin) in humans in vivo. Because the absorption spectra of oxyhemoglobin and deoxyhemoglobin differ, their relative concentrations within tissues can be determined. The tissue probe used on skeletal muscle emits near-infrared light and collects light returning from tissue of an approximate depth of 23 mm. The NIRS signal from skeletal muscle is hence a quantification of the tissue hemoglobin oxygen saturation specifically in the microcirculation. Muscle tissue oxygen saturation (STO2) readings were obtained noninvasively using a NIRS probe (InSpectra Tissue Spectrometer; Hutchinson Technology, Hutchinson, Minn) placed over the skin of the thenar eminence. The positioning of the probe was chosen because of relatively low adiposity and consistency with other studies. STO2 values were continuously monitored and stored into a computer using InSpectra software analysis Program version 2.0 (Hutchinson Technology) running in MatLab 7.0 (Math Works Inc, Natick, Mass). Baseline STO2 was recorded for 5 min; sample measurements signals were updated every 3.5 s, after which ischemia was induced by inflating an upper-arm cuff placed above the elbow to 50 mmHg above systolic blood pressure. After a 90-s period of ischemia, the occluding cuff was rapidly deflated to 0 mmHg. The slope of increase was defined as the utmost rate of STO2 increase 1 s after cuff deflation (%/s) and is thought to be representative of endothelial function (10).
SDF imaging protocol
Sidestream dark-field imaging was performed using a handheld device that illuminates an area of interest with light emitted by a circle of light-emitting diodes. The reflected light is returned through the inner image-conducting core, which is optically isolated from the light-emitting diodes and caught on camera. Therefore, in SDF imaging, the illumination and reflectance light paths do not travel down and back the same pathway, which improves contrast and lowers surface reflectance and blurring. If a wavelength within the hemoglobin absorption spectrum is chosen (e.g., 548 nm), erythrocytes will appear dark, and leukocytes may be visible as refringent bodies. The vessel walls itself are not visualized directly, although faint contours can be identified, depending on the presence of intravascular erythrocytes. Sidestream dark-field imaging and semiquantitative analysis were performed as described in detail elsewhere (14, 15, 24-26). In short, video images (Microscan; Microvision Medical, Amsterdam, the Netherlands) were captured through fire-wire connection to a Toshiba laptop computer with fire-wire capturing. Clips of 20 s duration were used (27), and each image was divided into four equal quadrants. Quantification of flow (microvascular flow index) was scored per quadrant, for each cohort of microvessel diameter. The diameters of the microvessels were small (10-25 μm), medium (25-50 μm), and large (50-100 μm) with quantification of flow (0 = no flow, 1 = intermittent flow, 2 = sluggish flow, and 3 = continuous flow). Sidestream dark-field images were analyzed off-line and in a blinded fashion by one of the investigators (P.S.) unaware of the applied protocol.
Forearm venous occlusion plethysmography protocol
Forearm blood flow was determined in both forearms with venous occlusion plethysmography (Filtrass Angio; Domed, Munich, Germany) as previously described (18). In short, the brachial artery of the nondominant arm was cannulated, and 30-min supine rest was taken before the start of the experiment in a temperature-stable room. Venous occlusion was achieved by inflating the upper-arm cuffs to 45 mmHg. Strain gauges were placed on the forearms and connected to plethysmograph to measure changes in forearm volume in response to inflation of the venous-congesting cuffs. Forearm blood flow measurements were adjusted for systemic changes unrelated to the local stimulus by expressing flow in the experimental arm as a ratio of concurrent flow in the noninfused (control) arm (FBF ratio). The effect of acetylcholine infusion is expressed as ratio flow change compared with baseline flow. Forearm blood flow data are the mean of measurements obtained during the last 3 min of each infusion (steady state). Forearm blood flow and acetylcholine dose were normalized for forearm volume as measured with the water displacement method and expressed in milliliters per min per deciliter forearm volume (mL · min−1 · dL−1). After instrumentation of both arms and an equilibration period of 15 min, the vasodilator response to infusion of acetylcholine into the brachial artery was quantified. Each dose-response curve started with a 5-min period of baseline FBF measurements, followed by 5-min intra-arterial acetylcholine infusion at 0.5, 2, and 8 µg · min−1 · dL−1. Four hours after LPS administration, the acetylcholine dose-response curve was repeated to determine the LPS-induced effects on acetylcholine-mediated vasodilation. On day 5, the measurements were repeated to determine the effects of endotoxin tolerance on endothelium-dependent vasodilation.
LPS-induced changes on different time points were tested for significance using repeated-measures ANOVA using SPSS 14.0.1 for Windows (SPSS Inc, Chicago, Ill). The results of NIRS and SDF imaging and the effect of acetylcholine infusion in the forearm were tested for significance with repeated-measures ANOVA and paired data with the use of Wilcoxon signed rank test for nonparametric data. Normally distributed data are expressed as mean ± SE, and not normally distributed data as median (interquartile range). P < 0.05 was considered statistically significant.
Baseline characteristics of the volunteers
The age, weight, height, heart rate, and blood pressure of the nine volunteers were 22 ± 2 years, 71 ± 8 kg, 181 ± 7 cm, 65 ± 16 beats/min, and 127 ± 18/73 ± 10 mmHg, respectively.
LPS administration induced the expected and transient flulike symptoms, characterized by a significant increase in symptom score, temperature, and heart rate and a decrease in blood pressure (P < 0.001 for all parameters). During subsequent LPS administrations, changes in temperature, heart rate, and blood pressure were significantly less pronounced, indicating the development of endotoxin tolerance (P < 0.001 for all parameters, but blood pressure P = 0.015; Table 1). During the 5 days of the experiment, all measured cytokines, both proinflammatory and anti-inflammatory, were significantly attenuated, also indicating the presence of tolerance (P < 0.001 for all measured cytokines; Table 1).
Tissue oxygenation during endotoxemia
To assess endothelial function, the slope of increase of STO2 after reperfusion was measured. Two hours after the first administration of LPS, the slope of increase was significantly attenuated with 79% (62%-92%) (P = 0.04) and restored to levels above baseline 2 h later. On day 5, the slope of increase after reperfusion showed no difference between time points after LPS administration when endotoxin tolerance was present (P = 0.72; Fig. 1).
On day 1, the microvascular flows of the medium and large microvessels were attenuated (33% [14%-40%] and 30% [10%-33%], respectively) 2 h after the first administration of LPS (P = 0.07 and 0.04, respectively). Two hours later, the response was restored to levels above baseline for the medium microvessels. On day 5, no statistically significant changes in flow were observed after the administration of LPS (P = 0.47 for both medium and large microvessels). Two and 4 h after the induction of endotoxemia and endotoxin tolerance, sublingual flow in small microvessels did not differ from baseline values (P = 0.50 and 0.22, respectively, at t = 2 and 4 h; Fig. 2).
On day 1, the acetylcholine-induced vasodilatory response was significantly attenuated during endotoxemia compared with baseline (67% [45%-72%], P = 0.01, ANOVA repeated-measures over the complete dose-response curve). On day 5, this LPS-induced attenuated vasodilatory effect of acetylcholine infusion into the forearm circulation was not observed (P = 0.25; Fig. 3).
Our study shows that systemic inflammation in humans induced by experimental endotoxemia results in microvascular endothelial dysfunction that can be detected by three different methods of approach currently used in (micro)vascular studies. We demonstrate that, during acute endotoxemia, baseline microvascular flow is decreased (SDF imaging), and endothelial dysfunction is present, which can be detected either by an attenuated endothelium-dependent ischemia-mediated increase in tissue oxygenation (NIRS) or by an attenuated pharmacologically induced endothelium-dependent vasodilatory response to acetylcholine (venous occlusion plethysmography). Furthermore, we show a restored flow pattern when LPS tolerance develops and a restored endothelial function after subsequent LPS administrations related to the development of LPS tolerance.
Altered microvascular blood flow occurs frequently in patients with sepsis and may also play a role in the evolution from sepsis to septic shock. In addition, its presence is associated with outcome, as several studies have shown that microvascular blood flow was more severely altered in nonsurvivors than in survivors (2, 3, 28). Indeed, microcirculatory alterations improved rapidly in septic shock survivors but not in patients dying of multiple organ failure regardless of whether shock had resolved or not. Different patterns of microvascular alterations in septic shock could therefore characterize outcome (4). Interestingly, the severity of the endothelial dysfunction detected by NIRS was related to the severity of the septic state and related to outcome in a previous study, as the slope of increase after ischemia was lower in patients with septic shock compared with patients with sepsis (10). The pathophysiological mechanism by which endothelial dysfunction is sustained in patients with sepsis is not clear, but because our study shows that endothelial function is restored during subsequent administrations of LPS, it seems likely that the presence of circulating LPS itself does not account for this persistent dysfunction.
A consistent feature of the pathology in severe sepsis and multiple organ dysfunction is the focal nature of their distribution. It is thought that the endothelium is an important determinant of this focal response in sepsis. Typically, patients develop dysfunction only in a limited number of organs. Furthermore, distribution within organs is nonhomogeneous. In accordance, the endothelium displays remarkable heterogeneity in health and disease states, integrating systemic changes in inflammation and coagulation in ways that differ in time and from organ to organ. This may be an explanation for the observed normalized endothelial function and flow with NIRS and SDF imaging 4 h after LPS administration on day 1, whereas it is known that endothelial dysfunction is still present at that time in other vascular beds (29), as adequately shown with the occlusion plethysmography in previous studies (30) and confirmed in the present experiment. More research is needed to further unravel the mediators and kinetics of endothelial dysfunction during systemic inflammation.
Because of the presence of LPS receptors on the endothelium (31), it was thought that endotoxin may exert a direct effect on endothelial function. Indeed, after LPS administration, an upregulation of the Toll-like receptor (TLR) 4 receptor is observed with subsequent increased production of various cytokines. However, during endotoxin tolerance, when an identical dose of LPS is administered on day 5, endothelial dysfunction does not occur; this can either be due to downregulation of the TLR-4 receptor as the endothelial cell becomes tolerant or due to intracellular changes with normal regulated TLR expression on the endothelial cell. The other hypothesis, that LPS-induced cytokines or other inflammatory mediators are primarily responsible for endothelial dysfunction, is supported by the fact that, in humans, Salmonella typhi vaccination is associated with short-term impairment of endothelium-dependent dilation in conduit and resistance vessels that parallels the inflammatory response (32).
Several limitations to our study should be acknowledged. Because of the LPS-induced burden for the subjects and laborious and invasive nature of the experiments, LPS studies are usually performed in small groups of subjects. Nevertheless, in the nine subjects studied, we were able to demonstrate the effects of systemic inflammation on various aspects of endothelial function by the use of three methods currently used in microcirculation research. Second, NIRS and SDF are limited by the semiquantitative approach, the intrauser variability, movement artifacts, and the indirect way of measurements that can all influence the obtained results. These limitations are, at least partially, obviated by the fact that serial observations per subject were obtained and analyzed in a blinded fashion. Although we demonstrated that the used methods were able to detect endothelial dysfunction during systemic inflammation, and the kinetics of endothelial function may be of importance in the critically ill patient, the clinical applicability of these methods is not yet established. Despite advances in our understanding of endothelial cell biology, this has not been translated into effective clinical therapies, although stabilizing the endothelium may be key to halting disease progression. It seems that organ dysfunction is often preceded by one unseen failing organ, the endothelium. Our study shows that bedside tests to identify endothelial dysfunction during systemic inflammation are possible, but whether endothelial dysfunction could become a treatment goal is not clear.
In conclusion, systemic inflammation in humans induced by experimental endotoxemia results in microvascular and endothelial dysfunction, which can be detected by NIRS, SDF imaging, and venous occlusion plethysmography. These microcirculatory alterations were no longer present after subsequent administrations of LPS, indicating that endothelial cells develop LPS tolerance either by downregulation of the LPS receptor or by modification of the TLR-4 signaling pathway, or that the LPS-induced cytokine response is responsible for the endothelial dysfunction. Studies focusing on surface receptor expression and analysis of markers for the intracellular pathways need to be conducted to differentiate between these putative pathophysiological mechanisms.
An increased understanding of endothelial dysfunction during sepsis may result in novel effective treatment regimens. Whether induction of endotoxin tolerance before elective interventions associated with a high risk of developing endothelial dysfunction improves patient outcome is worth investigating and needs to be established.
The authors thank Trees Jansen for cytokine measurements and our research nurse, Tijn Bouw, for his help with the conduct of the experiments.
1. Vincent JL, Sakr Y, Sprung CL, Ranieri VM, Reinhart K, Gerlach H, Moreno R, Carlet J, Le G Jr, Payen D: Sepsis in European intensive care units: results of the SOAP study. Crit Care Med
2. De Backer D, Creteur J, Preiser JC, Dubois MJ, Vincent JL: Microvascular blood flow is altered in patients with sepsis. Am J Respir Crit Care Med
3. Spronk PE, Ince C, Gardien MJ, Mathura KR, Oudemans-van Straaten HM, Zandstra DF: Nitroglycerin in septic shock after intravascular volume resuscitation. Lancet
4. Sakr Y, Dubois MJ, De BD, Creteur J, Vincent JL: Persistent microcirculatory alterations are associated with organ failure and death in patients with septic shock. Crit Care Med
5. Elbers PW, Ince C: Mechanisms of critical illness-classifying microcirculatory flow abnormalities in distributive shock. Crit Care
6. Spronk PE, Zandstra DF, Ince C: Bench-to-bedside review: sepsis is a disease of the microcirculation
. Crit Care
7. Ince C, Sinaasappel M: Microcirculatory oxygenation and shunting in sepsis and shock. Crit Care Med
8. De Blasi RA, Palmisani S, Alampi D, Mercieri M, Romano R, Collini S, Pinto G: Microvascular dysfunction and skeletal muscle oxygenation assessed by phase-modulation near-infrared spectroscopy
in patients with septic shock. Intensive Care Med
9. Doerschug KC, Delsing AS, Schmidt GA, Haynes WG: Impairments in microvascular reactivity are related to organ failure in human sepsis. Am J Physiol Heart Circ Physiol
10. Creteur J, Carollo T, Soldati G, Buchele G, De BD, Vincent JL: The prognostic value of muscle STO2
in septic patients. Intensive Care Med
11. De Blasi RA, Ferrari M, Antonelli M, Conti G, Almenrader N, Gasparetto A: O2
delivery relationship and arteriolar resistance in the forearm of critically ill patients measured by near infrared spectroscopy. Shock
12. Pareznik R, Knezevic R, Voga G, Podbregar M: Changes in muscle tissue oxygenation during stagnant ischemia in septic patients. Intensive Care Med
13. Trzeciak S, Dellinger RP, Parrillo JE, Guglielmi M, Bajaj J, Abate NL, Arnold RC, Colilla S, Zanotti S, Hollenberg SM: Early microcirculatory perfusion derangements in patients with severe sepsis and septic shock: relationship to hemodynamics, oxygen transport, and survival. Ann Emerg Med
14. Boerma EC, Mathura KR, van der Voort PHJ, Spronk PE, Ince C: Quantifying bedside-derived imaging of microcirculatory abnormalities in septic patients: a prospective validation study. Crit Care
15. Boerma EC, van der Voort PHJ, Spronk PE, Ince C: Relationship between sublingual and intestinal microcirculatory perfusion in patients with abdominal sepsis. Crit Care Med
16. Pickkers P, Dorresteijn MJ, Bouw MP, van der Hoeven JG, Smits P: In vivo
evidence for nitric oxide-mediated calcium-activated potassium-channel activation during human endotoxemia
17. Petrie JR, Ueda S, Morris AD, Murray LS, Elliott HL, Connell JM: How reproducible is bilateral forearm plethysmography
? Br J Clin Pharmacol
18. Leslie SJ, Attina T, Hultsch E, Bolscher L, Grossman M, Denvir MA, Webb DJ: Comparison of two plethysmography systems in assessment of forearm blood flow. J Appl Physiol
19. Calver A, Collier J, Moncada S, Vallance P: Effect of local intra-arterial NG-monomethyl-l-arginine in patients with hypertension: the nitric oxide dilator mechanism appears abnormal. J Hypertens
20. Kienbaum P, Prante C, Lehmann N, Sander A, Jalowy A, Peters J: Alterations in forearm vascular reactivity in patients with septic shock. Anaesthesia
21. Draisma A, Pickkers P, Bouw MP, Hoeven van der H: Development of endotoxin tolerance
in humans in vivo
[abstract]. Intensive Care Med
22. Dorresteijn MJ, van Eijk LT, Netea MG, Smits P, van der Hoeven JG, Pickkers P: Iso-osmolar prehydration shifts the cytokine response towards a more anti-inflammatory balance in human endotoxemia
. J Endotoxin Res
23. Ward KR, Ivatury RR, Barbee RW, Terner J, Pittman R, Filho IP, Spiess B: Near infrared spectroscopy for evaluation of the trauma patient: a technology review. Resuscitation
24. Groner W, Winkelman JW, Harris AG, Ince C, Bouma GJ, Messmer K, Nadeau RG: Orthogonal polarization spectral imaging: a new method for study of the microcirculation
. Nat Med
25. Cerny V, Turek Z, Parizkova R: Orthogonal polarization spectral imaging. Physiol Res
26. Harris AG, Sinitsina I, Messmer K: The Cytoscan Model E-II, a new reflectance microscope for intravital microscopy: comparison with the standard fluorescence method. J Vasc Res
27. De Backer D, Hollenberg S, Boerma C, Goedhart P, Buchele G, Ospina-Tascon G, Dobbe I, Ince C: How to evaluate the microcirculation
: report of a round table conference. Crit Care
28. Kirschenbaum LA, Aziz M, Astiz ME, Saha DC, Rackow EC: Influence of rheologic changes and platelet-neutrophil interactions on cell filtration in sepsis. Am J Respir Crit Care Med
29. Suffredini AF, Fromm RE, Parker MM, Brenner M, Kovacs JA, Wesley RA, Parrillo JE: The cardiovascular response of normal humans to the administration of endotoxin. N Engl J Med
30. Pleiner J, Mittermayer F, Schaller G, MacAllister RJ, Wolzt M: High doses of vitamin C reverse Escherichia coli endotoxin-induced hyporeactivity to acetylcholine in the human forearm. Circulation
31. Faure E, Thomas L, Xu H, Medvedev A, Equils O, Arditi M: Bacterial lipopolysaccharide and IFN-gamma induce Toll-like receptor 2 and Toll-like receptor 4 expression in human endothelial cells: role of NF-kappa B activation. J Immunol
32. Hingorani AD, Cross J, Kharbanda RK, Mullen MJ, Bhagat K, Taylor M, Donald AE, Palacios M, Griffin GE, Deanfield JE, et al.: Acute systemic inflammation impairs endothelium-dependent dilatation in humans. Circulation