| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Departments of *Cardiothoracic Anesthesia and Intensive Care and Clinical Neurophysiology, Sahlgrenska University Hospital, Göteborg, Sweden
Address correspondence and reprint requests to Sven-Erik Ricksten, MD, PhD, Department of Cardiothoracic Anesthesia and Intensive Care, Sahlgrenska University Hospital, S-413 45 Göteborg, Sweden. Address e-mail to sven-erik.ricksten{at}aniv.gu.se
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Implications: Jejunal mucosal perfusion increases during mild hypothermic cardiopulmonary bypass (CPB). Intestinal laser Doppler flowmetry, gastric tonometry, and measurements of splanchnic lactate extraction could not reveal a local or global splanchnic ischemia during or after CPB. A mismatch between splanchnic oxygen delivery and demand was seen, particularly during rewarming and warm CPB.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Splanchnic ischemia during and after CPB, as a result of imbalance between splanchnic oxygen supply and demand, decreased splanchnic oxygen use, or both may contribute to the development of multiple organ failure. It has been thought that splanchnic ischemia during CPB may lead to disrupted intestinal mucosal barrier function and subsequent translocation of endotoxin and microorganisms, with a consequent release of proinflammatory cytokines (4,6–9).
Earlier reports using laser Doppler flowmetry (LDF) in humans have shown a marked reduction in gastric and rectal mucosal perfusion during CPB (3,4,9–11). There are no reports in humans on the effects of CPB on small intestinal mucosal perfusion, which might similarly be jeopardized and lead to hypoxic injury of the vulnerable mucosa (8). The aim of this study was therefore to elucidate, by the use of the endoluminal LDF technique, whether there is a change in jejunal mucosal perfusion (JMP) during CPB.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The patients were premedicated with IM morphine (5 mg), scopolamine (0.2 mg), and PO flunitrazepam (1 mg). All patients also received oral cisapride 15–20 mg to facilitate the placement of the LDF catheter (see below). Anesthesia was induced with 100–200 µg fentanyl and 100–200 mg propofol followed by 0.1 mg/kg pancuronium. The anesthesia was maintained prior to CPB with an infusion of propofol 200–600 mg/h until the LDF catheter was positioned and was then maintained with a combination of fentanyl (total dose, including the induction dose, 7–10 µg/kg) and enflurane. The patients were mechanically ventilated to normocapnia at a PO2 > 15 kPa.
The perfusion system consisted of a hollow fiber membrane oxygenator and a Sarns 9000 max pump (Sarns Inc, Ann Arbor, MI). The CPB circuit was primed with 1800 mL of Ringer acetate® and 200 mL of mannitol 150 mg/L. Hematocrit was maintained >20%. None of the patients received transfusion of erythrocytes during the experimental procedure. A nonpulsatile pump flow of 2.5 L · min-1 · m-2 was maintained during CPB. Arterial CO2 tension was maintained at 4.7–5.3 kPa and was analyzed at 37°C. Arterial oxygen tension was maintained at 15–20 kPa. The surgical procedure was performed during mild hypothermia (34°C). Anesthesia was maintained by propofol infusion of 200–400 mg/h during CPB.
The local JMP was measured with a laser Doppler flowmeter ( Periflux PF 4001TM ; Perimed AB, Järfälla, Sweden). A custom-built, previously described (12,13) LDF catheter with two incorporated probes was used. The two probes, each consisting of one emitting and two collecting light fibers, were situated 23 mm and 123 mm, respectively, from the tip of the catheter. After the induction of anesthesia and during fluoroscopic guidance, the LDF catheter was advanced endoluminally through the pylorus to the proximal jejunum, where the two probes were placed 10–40 cm distal to the ligament of Treitz. Because the jejunum is collapsed in fasted subjects, the probes are in close contact with the mucosa, and a pulse-synchronous JMP is continuously recorded during intestinal quiescence (12,13). The position of the LDF catheter was confirmed after surgery by fluoroscopy.
A wavelength of 780 nm, a time constant of 0.2 s, and a bandwidth of 20–25 kHz was used. The sampling frequency was 32 Hz. The perfusion value was expressed in units of relative perfusion (perfusion units [PU]). The zero value was defined by placing the LDF probe against a plastic disk and the level 250 PU was calibrated using the motility standard provided by Perimed. All the JMP values presented were calculated from periods of intestinal quiescence without peristalsis. Peristalsis is easily recognized in the laser Doppler recording because it induces motion artifacts (13,14). The mean local JMP was averaged from the two channels (32 measurements per second), and a mean value over each of the recording periods was calculated with the Perisoft software (Perimed).
Gastric mucosal PCO2 (PiCO2) was measured every 10 min by using a standard orogastric or nasogastric tube ( TRIPTM , Tonometry Catheter 14F; Tonometrics, Helsinki, Finland) connected to an automated gas analyzer ( TonocapTM TC-200; Datex, Helsinki, Finland) (5). The correct position of the tonometry tube was confirmed by fluoroscopy. Gastric pH (pHi)was calculated with the following formula (15): pHi = 6.1 + log (arterial bicarbonate concentration · 0.03-1 · PiCO2-1 · k-1).
After placement of the LDF probes and the tonometry tube, 7.0F Baxter pulmonary artery catheters (Baxter, Santa Ana, CA) were inserted in the right hepatic vein and in the pulmonary artery under fluoroscopic guidance. Cardiac index (CI) was measured in triplicate by the thermodilution technique with ice-cold boluses of 10-mL saline. Mean arterial blood pressure (MAP), pulmonary artery pressure, central venous pressure, and heart rate were continuously measured and stored, using the analog output from the hemodynamic monitor ( SirecustTM ; Siemens, Danvers, MA) and the Perisoft program (Perimed). Systemic oxygen delivery (DO2), uptake (&OV0312;O2), and extraction ratio, as well as systemic vascular resistance (SVR), were calculated from the following standard formulas, where CVP = central venous pressure, CO = cardiac output, CaO2 = arterial oxygen content, CvO2 = mixed venous oxygen content, and SaO2 and SvO2 are arterial and mixed venous oxygen saturation, respectively:equationequationequation
andequation
Arterial (AL) and hepatic venous (HVL) lactate concentrations were measured enzymatically (Hitachi 917, Roche, Stockholm, Sweden), and splanchnic lactate and oxygen extraction ratios were calculated with the following formulas, where SaO2 and ShvO2 are arterial and hepatic vein oxygen saturation, respectively:equation
andequation
Data were collected from the following seven 10-min measurement periods: (a) before start of surgery, (b) 10 min after sternotomy, (c) 10 min after the achievement of a core temperature of 34°C during CPB, (d) at a temperature of 34°C immediately before rewarming of the patient, (e) during rewarming at a body temperature of 35.5°C (RW), (f) at a body temperature of 36.5°C (W), and (g) 60 min after weaning from CPB (CPB + 60').
|
|
|
|
|
|
|
|
|
|
|
|
LDF measurements were performed continuously, and JMP was averaged from the two channels for each 10-min measurement period. PiCO2 was averaged from two consecutive values for each measurement period. Core temperature, CI, and pulmonary artery occlusion pressure (off CPB) were determined at the end of each 10-min measurement period. Blood samples for blood gas analyses and hematocrit were drawn from arterial, pulmonary (off CPB), and hepatic catheters, whereas lactate samples were drawn from arterial and hepatic catheters at the end of each measurement period. During CPB, blood samples for measurement of mixed venous oxygen saturation were drawn from the venous side of the CPB circuit.
Five patients received ephedrine (5–25 mg) after the induction of anesthesia but before CPB. One patient received bolus doses of phenylephrine (total dose 400 µg) during CPB to maintain a MAP >45 mm Hg. Two patients received nitroglycerin infusion (0.2–0.4 µg · kg-1 · min-1) to prevent vasospasm of the radial artery graft at the end of and after CPB. Three patients received sodium nitroprusside infusion (0.1–1 µg · kg-1 · min-1) for treatment of arterial hypertension during or after weaning from CPB, but only two received sodium nitroprusside (0.3–0.5 µg · kg-1 · min-1) during the experimental procedure. One of these received sodium nitroprusside only at the last recording period, and the second patient received sodium nitroprusside at a constant infusion rate of 0.4 µg · kg-1 · min-1 during hypothermia, RW, and W CPB.
Analysis of variance for repeated measurements was used to assess changes over time of the various variables. A post hoc contrast analysis was performed to compare the two weighted control periods (before and after surgery) with the remaining measurement periods. The relationship between JMP and splanchnic lactate extraction ( Fig. 4) was tested by using linear regression analysis. A P value <0.05 was considered statistically significant. Values are presented as mean ± SEM.
|
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
See Table 1 for data regarding central hemodynamics, systemic oxygen transport and uptake, body temperature, and hematocrit. MAP did not change during the experimental procedure. CI increased and SVR decreased during CPB. CI was significantly increased 1 h after the end of CPB compared with the control. Systemic oxygen delivery decreased to a minor extent during the first hypothermic measurement period but was unchanged during the remaining periods. Systemic oxygen uptake was higher during RW, during W CPB, and 1 h after CPB compared with control. Mixed venous oxygen saturation decreased and systemic oxygen extraction increased during RW, and W CPB and were also decreased and increased, respectively, 1 h after the end of CPB compared with control.
|
See Figures 1–4 and Table 2 for data regarding JMP, gastric-arterial PCO2 gradient, splanchnic oxygen, and lactate extraction. JMP increased during the second hypothermic measurement period (26%), during RW (31%), and during W CPB (38%) and was also increased 1 h after the end of CPB (42%) compared with control ( Fig. 1, 3). The gastric-arterial PCO2 gap (PiCO2 - PACO2) increased and pHi decreased during the RW period and during W CPB. One hour after the end of CPB, pHi was decreased and gastric-arterial PCO2 gap was increased compared with control (Fig. 1). Splanchnic oxygen extraction increased ( Fig. 2) and ShvO2 decreased during hypothermic CPB (Table 2). There was a further increase in splanchnic oxygen extraction and a decrease in ShvO2 during RW. One hour after the end of CPB, ShvO2 was decreased and splanchnic oxygen extraction was increased compared with control.
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Splanchnic (hepato-mesenteric) blood flow in humans has previously been measured during CPB by using the indocyanine green clearance technique. Thus, splanchnic blood flow is not significantly affected by normothermic or hypothermic CPB at a pump flow of 2.4 L · min-1 · m-2 compared with the prebypass condition (16,17). The increase in splanchnic oxygen extraction during hypothermic CPB, as shown in the present study, indicating a splanchnic oxygen supply/demand mismatch, was therefore most likely caused by a decrease in splanchnic oxygen delivery, in turn caused by a decrease in hemoglobin concentration caused by hemodilution. The systemic oxygen extraction was not significantly affected by CPB and hypothermia, indicating that the splanchnic region might be more susceptible to a decrease in oxygen delivery by hemodilution, compared with other organs. Previous studies on CPB in humans have also indicated a greater increase in splanchnic versus systemic oxygen extraction (6,17).
During RW and W CPB, splanchnic oxygen extraction increased further, and this was accompanied by a significant increase in systemic oxygen extraction when compared with the prebypass level. This increase in both splanchnic and systemic oxygen extractions was most likely caused by the combination of decreased hemoglobin levels and an increase in systemic and splanchnic oxygen consumption. The difference between systemic and splanchnic oxygen extraction was even more pronounced during the RW phase and W CPB when compared with hypothermia.
LDF has been used extensively in the gastrointestinal setting (3,4,9–14,18), but there are only a few human studies that use the endoluminal approach in the intestines (12–14). The LDF signal is proportional to the number and the mean velocity of red blood cells in the tissue volume of interest and thus can provide useful information on tissue oxygen delivery. Although LDF yields no absolute blood flow values, the LDF values from the mucosal side of the jejunum correlate strongly with simultaneously obtained absolute mucosal blood flow measured by hydrogen gas clearance and microsphere techniques (19). One advantage of the LDF technique is that it is possible to measure mucosal perfusion continuously, whereas the drawback is that measurements can be performed only at a local site and during intestinal quiescence, because peristalsis causes motion artifacts. However, peristalsis is a smaller problem in patients during anesthesia compared with awake volunteers (12,13).
A mucosal hypoperfusion of the jejunum by using the LDF technique has previously been described during CPB in animals despite well preserved intestinal blood flow (18). Furthermore, hypoperfusion of the gastric and rectal mucosa has been described in patients undergoing CPB (3,4,9–11). It has been suggested that this gastrointestinal mucosal hypoperfusion (4,9) might explain the increase in gut permeability seen after CPB (4,8,9) that is caused by a proposed hypoxic insult of the intestinal mucosa. However, in this study we demonstrated an increase in JMP both during hypothermia (26%) and particularly so during RW (31%) (Fig. 3) and W CPB (38%). The increase in JMP during hypothermia is most likely caused by a decrease in blood viscosity (9,20). The further increase in JMP during RW may be explained by an autoregulatory vasodilatation in the mucosal and submucosal layers in turn caused by increased metabolism. In humans, approximately 70%–75% of the resting intestinal blood flow is distributed to the mucosal and submucosal layers (21). It is therefore likely that the 30% to 40% increase in JMP, as shown in this study, also reflects an increase in intestinal blood flow. A previously described increase in portal venous blood flow, measured with color Doppler sonography, in patients during RW from cold CPB, supports these findings (22). Thus, despite an increase in mucosal and probably also intestinal perfusion during CPB, splanchnic oxygen extraction increased, indicating that the decrease in arterial blood oxygen content and the increase in splanchnic metabolism during RW were not fully matched by a proportional increase in splanchnic blood flow. The previously described decrease in gastric and rectal mucosal perfusion during CPB in humans, which contrasts with our results, may be explained by the use of a more pronounced hypothermia (15–32°C) (3,4,9–11) or lower pump flow (1.7–1.8 L · min-1 · m-2) (4,9) compared with the present study (34°C and 2.5 L · min-1 · m-2). However, it cannot be excluded that gastric, rectal, and intestinal mucosal perfusion does not change in parallel during CPB.
Calculation of gastric mucosal pH before, during, and after cardiac surgery has been used extensively to assess gastric mucosal perfusion and oxygenation (5). Estimation of PiCO2 with gastric air tonometry instead of saline has simplified the tonometric procedure without changing the reliability (5,23). Calculation of the measured gastric-arterial PCO2 difference (PCO2 gap) is considered to be a better marker of matching between gastric mucosal metabolism and perfusion than calculated pHi (5,15,23). In this study, the PCO2 gap increased slightly but significantly during RW and W CPB. It has been suggested that a PCO2 gap in the range 1.3–3.3 kPa reflects stagnant flow and that a gap >3.3 kPa would indicate anaerobic metabolism in the gastric mucosa (15,23). The mean PCO2 gap increased to 0.57 ± 0.12 kPa during warm CPB, and none of the patients had a PCO2 gap >3.3 kPa. The most likely explanation for this small but significant increase in PCO2 gap would be an increase in gastric mucosal metabolism, probably also reflecting an increase in global splanchnic metabolism during RW and W CPB; this increase was not fully matched by an increase in gastric mucosal perfusion.
Splanchnic lactate metabolism reflects the adequacy of splanchnic perfusion in cardiac surgical patients (24). The increase in splanchnic lactate extraction during CPB in this study indicates that splanchnic perfusion was adequate to support aerobic metabolism. An increase in splanchnic lactate extraction could be explained by increased levels of AL, an increase in splanchnic metabolic turnover, or both, especially during RW and W CPB. None of the patients had increased HVL levels or produced lactate during CPB. However, two patients produced lactate during surgery before CPB (Fig. 4). In these two patients, the corresponding LDF values were among the lowest obtained (119 PU and 135 PU, respectively). The significant increase in AL levels during CPB were thus probably not of intestinal origin but could instead be due to lactate production outside the splanchnic area or a decreased peripheral metabolism of pyruvate (2,17). An increase in pyruvate together with an increase in lactate can be seen if activity of the enzyme pyruvate dehydrogenase is impaired (2) or if the rate of glycolysis is increased. This can be seen in response to hyperglycemia and increased adrenergic stimulation (25). Increased levels of glucose and catecholamines are seen during CPB (2,25).
Four patients received nitroglycerin or sodium nitroprusside at low infusion rates during the experimental protocol. We do not believe that the use of these vasodilators influenced the overall results of this study, as three of these patients received the vasodilator only during the last recording period and one patient had nitroprusside infusion at a constant infusion rate during a major part of the experimental period.
In this study on mild hypothermic CPB in humans, there was a progressive increase in JMP, a finding that is in striking contrast to the previously described hypoperfusion of the gastric and rectal mucosa during CPB in humans. The increase in jejunal blood flow during CPB, as reflected by the 30%–40% increase in JMP in this study, could not compensate for the decrease in blood oxygen carrying capacity (hemodilution) and the increase in splanchnic metabolic demand during RW, thus probably explaining why splanchnic oxygen extraction and the gastric PCO2 gap increased during CPB.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Nygren, A. Thoren, E. Houltz, and S.-E. Ricksten Autoregulation of human jejunal mucosal perfusion during cardiopulmonary bypass. Anesth. Analg., June 1, 2006; 102(6): 1617 - 1622. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Hessel II Abdominal Organ Injury After Cardiac Surgery Seminars in Cardiothoracic and Vascular Anesthesia, September 1, 2004; 8(3): 243 - 263. [Abstract] [PDF]
|
|
|
|
 |
|
 R. Rokyta Jr, M. Matejovic, A. Krouzecky, K. Opatrny Jr, J. Ruzicka, and I. Novak Effects of continuous venovenous haemofiltration-induced cooling on global haemodynamics, splanchnic oxygen and energy balance in critically ill patients Nephrol. Dial. Transplant., March 1, 2004; 19(3): 623 - 630. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Okano, S. Miyoshi, R. Owada, N. Fujita, Y. Kadoi, S. Saito, F. Goto, and T. Morita Impairment of Hepatosplanchnic Oxygenation and Increase of Serum Hyaluronate During Normothermic and Mild Hypothermic Cardiopulmonary Bypass Anesth. Analg., August 1, 2002; 95(2): 278 - 286. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Okano, H. Hiraoka, R. Owada, N. Fujita, Y. Kadoi, S. Saito, F. Goto, and T. Morita Hepatosplanchnic oxygenation is better preserved during mild hypothermic than during normothermic cardiopulmonary bypass : [L'oxygenation hepatosplanchnique est mieux preservee pendant la circulation extracorporelle sous legere hypothermie que sous normothermie] Can J Anesth, November 1, 2001; 48(10): 1011 - 1014. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
