Preoperative fasting in children: review of existing guidelines and recent developments.

The current guidelines for preoperative fasting recommend intervals of 6, 4, and 2 h (6-4-2) of fasting for solids, breast milk, and clear fluids, respectively. The objective is to minimize the risk of pulmonary aspiration of gastric contents, but also to prevent unnecessarily long fasting intervals. Pulmonary aspiration is rare and associated with nearly no mortality in paediatric anaesthesia. The incidence may have decreased during the last decades, judging from several audits published recently. However, several reports of very long fasting intervals have also been published, in spite of the implementation of the 6-4-2 fasting regimens. In this review, we examine the physiological basis for various fasting recommendations, the temporal relationship between fluid intake and residual gastric content, and the pathophysiological effects of preoperative fasting, and review recent publications of various attempts to reduce the incidence of prolonged fasting in children. The pros and cons of the current guidelines will be addressed, and possible strategies for a future revision will be suggested.

Safety of glucose-containing solutions during accidental hyperinfusion in piglets.

BACKGROUND: Errors in fluid management can lead to significant morbidity in children. We conducted an experimental animal study to determine the margin of safety in accidental hyperinfusion of different glucose and electrolyte containing solutions. METHODS: Fifteen piglets [bodyweight 12.1 (sd 2.0) kg] were randomly assigned to receive either 100 ml kg⁻¹ of balanced electrolyte solution with glucose 1% (BS-G1), hypotonic electrolyte solution with glucose 5% (HE-G5), or glucose 40% solution (G40) over 1 h. Blood electrolytes, glucose, and osmolality and intracranial pressure (ICP) were measured before, during, and after fluid administration. RESULTS: Hyperinfusion of BS-G1 led to moderate hyperglycaemia [baseline 3.4 (sd 1.3) mmol litre⁻¹, study end 12.6 (1.8) mmol litre⁻¹], but no other relevant pathophysiological alterations. Hyperinfusion of HE-G5 produced marked hyperglycaemia [baseline 3.9 (1.2) mmol litre⁻¹, study end 48.6 (4.3) mmol litre⁻¹, P < 0.05] and hyponatraemia [baseline 136.4 (1.3) mmol litre(-1), study end 119.6 (2.1) mmol litre⁻¹, P < 0.05], whereas osmolality remained stable during the course of the study. Hyperinfusion of G40 induced acute hyperglycaemic/hyperosmolar decompensation with an extreme decrease in serum electrolytes [e.g. sodium baseline 138 (1.1) mmol litre⁻¹, 30 min 87.8 (6.4) mmol litre⁻¹, P < 0.01], leading to cardiac arrest after infusion of 50-75 ml kg⁻¹. ICP remained within a physiological range in all groups. CONCLUSIONS: In an animal model of accidental hyperinfusion, BS-G1 showed the widest margin of safety and can therefore be expected to enhance patient safety in perioperative fluid management in children; HE-G5 proved significantly less safe; and G40 was found to be outright hazardous.

Reliability of continuous cardiac output determination by pulse-contour analysis in porcine septic shock.

BACKGROUND: Pulse-contour analysis represents a technique for cardiac output (CO)-measurement and allows continuously monitoring trends in CO. We evaluated reliability of pulse-contour CO (COpc) in septic shock. METHODS: Seventeen anaesthetized and mechanically ventilated pigs were investigated. After baseline measurements, 14 animals received 0.75 g/kg body weight faeces into the abdominal cavity to induce sepsis and were observed over 9 h, three animals served as controls. A central venous catheter was inserted into the jugular vein and an arterial catheter for thermodilution was inserted into the femoral artery. Two bedside computers were used for COpc. After induction of sepsis, COpc-computer No. 1 (COpcCAL) was recalibrated hourly. No further calibrations were performed in computer No. 2 (COpcNoCAL). We directly compared COpcCAL hourly before recalibration with COpcNoCAL. One hundred and seventy parallel triplicate determinations of CO were analysed using the method of Bland-Altman. RESULTS: Three hours after sepsis induction, correlation between recalibrated and non-recalibrated CO was r = 0.74, P < 0.01, at 5 h r = 0.59, P < 0.05 and 9 h r = 0.02, NS. Three hours after sepsis induction, bias +/- SD (limits of agreement) between both groups was 1.6 +/- 15.5 (-29.4-32.6) ml/kg/min, at 5 h -15.0 +/- 24.3 (-63.6-33.7) ml/kg/min and at 9 h -87.0 +/- 90.8 (-268.5-94.6) ml/kg/min. CONCLUSION: Continuous CO determination using pulse-contour analysis is a reliable method of assessing CO up to 5 h without recalibration in porcine septic shock. Thus, COpc may be a useful tool for assessment of unpredictable haemodynamic changes in sepsis.

Haemodynamic, acid-base and blood volume changes during prolonged low pressure pneumoperitoneum in rabbits.

BACKGROUND: The anaesthetic management of small infants during advanced laparoscopic surgery can be complicated by the major pathophysiological effects of increased intra-abdominal pressure. In this study haemodynamic, acid-base and blood volume changes were investigated during pneumoperitoneum in a small animal model. METHODS: Ten fasted, anaesthetized, mechanically ventilated and multi-catheterized New Zealand rabbits were randomized to carbon dioxide pneumoperitoneum (PP, duration 210 min, pressure 8 mm Hg) or control group. Cardiac index was determined using trans-cardiopulmonary thermodilution and total blood volume was measured by thermal-dye dilution with indocyanine green using a fibreoptic monitor system. RESULTS: In PP cardiac index (CI), central venous oxygen saturation (Scv(O(2))), total blood volume (TBV) and base excess (BE) decreased significantly during the study whereas all variables remained constant in the control group. After release of PP the measured variables did not return to baseline within 30 min [PP, baseline vs study end: CI 108 (22) vs 85 (14) ml kg(-1) min(-1), Scv(O(2)) 81.4 (8.9) vs 56.7 (9.8)%, TBV 318 (69) vs 181 (54) ml, BE -1.9 (2.7) vs -8.7 (1.8) mmol litre(-1); P<0.01]. CONCLUSION: Our animal model suggests that a decrease in CI, metabolic acidosis and hypovolaemia could occur after prolonged low pressure pneumoperitoneum in small infants, which is possibly not detectable by the standard monitor setting. Therefore, the routine use of an extended monitoring including measurement of central venous oxygen saturation and acid-base parameters should be considered during and soon after operation, when pneumoperitoneum will last longer than 2 h.

Anuria during pneumoperitoneum in infants and children: a prospective study.

INTRODUCTION: Transient oliguria during laparoscopic surgery is a known phenomenon. Currently, no data on the impact of pneumoperitoneum on renal function in children are available. PATIENTS AND METHODS: Thirty children with normal kidney function, who underwent laparoscopic surgery, were included in a prospective study. A transurethral catheter was placed to measure urine output during and 6 hours after operation. Renal blood flow (resistive index) was evaluated by Doppler ultrasound of a segmental renal artery before surgery, every 15 minutes during laparoscopy, and after 24 hours. Blood and urine samples were studied before and 24 hours after surgery. Hemodynamic parameters were monitored continuously during standardized anesthesia, including a standardized intravenous infusion regimen. RESULTS: Urine output decreased within 45 minutes of pneumoperitoneum in all patients. Of 8 children younger than 1 year, 7 (88%) developed anuria vs 3 of 22 (14%) children aged 1 to 15 years (P < .001). Nine children 1 year and older (32%) developed oliguria. There was a significant recovering in the mean urine output until 5 to 6 hours after pneumoperitoneum in both age groups. No significant alterations of the renal blood flow (resistive index) and the serum and urine levels of cystatin C, creatinine, and urea nitrogen were evident until 24 hours postoperatively. The volume of infusion during pneumoperitoneum did not correlate with urine output. CONCLUSION: Pneumoperitoneum leads to anuria in most children younger than 1 year and to oliguria in about one third of older children. This is a completely reversible phenomenon. Urine output should not be taken into consideration for calculating intravenous fluid administration during pneumoperitoneum in children.

Comparison of cardiac output measurements by arterial trans-cardiopulmonary and pulmonary arterial thermodilution with direct Fick in septic shock.

BACKGROUND AND OBJECTIVE: The aim of this study was to compare cardiac output (CO) measurements by arterial trans-cardiopulmonary thermodilution (ATD) and pulmonary arterial thermodilution (PATD) with CO estimated on the basis of the Fick calculation via a metabolic monitor in septic shock. METHODS: In a prospective animal study 20 anaesthetized, ventilated pigs (20.9 +/- 1.9 kg) were investigated. Septic shock was induced with faecal peritonitis. A pulmonary artery catheter was used for conventional measurement of CO. Simultaneously ATD was measured with a thermistor tipped catheter inserted into right carotid artery. Whole body oxygen consumption was measured by indirect calorimetry. Eighty data pairs of simultaneous CO measurements were analysed. RESULTS: CO measured with Fick and that measured with PATD were significantly correlated (r = 0.94, r = 0.87, P < 0.001). Mean CO measured by PATD was 94.3 +/- 40.1 mL min(-1) kg(-1). Bias was 10.1 mL min(-1) kg(-1) (95% confidence interval (CI): 6.0-14.2 mL min(-1) kg(-1)) with limits of agreement of -26.8 to 47.0 mL min(-1) kg(-1). Correlation between Fick derived CO estimation and ATD CO was similar (r = 0.91, r2 = 0.83, P < 0.001). Mean CO measured by trans-cardiopulmonary thermodilution was 104.3 +/- 43.2 mL min(-1) kg(-1). Bias was 0.75 mL min(-1) kg(-1) (95% CI: -3.8 to 5.3 mL min(-1) kg(-1)) with limits of agreement of -39.7 to 41.2 mL min(-1) kg(-1). CONCLUSIONS: Even during haemodynamic instability in septic shock the correlation of arterial trans-cardiopulmonary thermodilution and PATD derived CO with direct Fick was good. As arterial trans-cardiopulmonary thermodilution is less invasive than PATD, the former may offer practical advantages.

Ringer's solution but not hydroxyethyl starch or modified fluid gelatin enhances platelet microvesicle formation in a porcine model of septic shock.

BACKGROUND: Sepsis is associated with volume deficit and clotting system activation. Platelet activation in sepsis results in an increased formation of microvesicles, which in turn, have been associated with increased mortality. We hypothesized an effect of different volume replacement solutions on platelet-derived microvesicle formation in septic shock. METHODS: Anaesthetized, mechanically ventilated and multi-catheterized pigs received 1 g kg(-1) body weight faeces into the abdominal cavity to induce sepsis and were observed over 8 h. Five animals in each group received volume replacement therapy with modified fluid gelatin 4% or 8% (MFG4%, MFG8%), 6% hydroxyethylstarch (HES) 200/0.5 or Ringer's solution (RS) to maintain a central venous pressure of 12 mm Hg. Flow cytometry was used for determination of microvesicles before induction of sepsis (baseline) and after 8 h. Platelets and microvesicles were identified with an anti-platelet monoclonal Ab and a secondary antibody. Microvesicles were determined as the smallest 1-3% positive cells in forward scatter. Intergroup comparisons were performed using Wilks-Lambda and Ryan-Einot-Gabriel-Welsh F-test. Differences within groups were compared using a two-tailed Student's t-test. RESULTS: Baseline values were considered as 100%. While microvesicle formation was reduced in HES (73 (sd 19)%), MFG4% (63 (41)%) and MFG8% groups (53 (17)%), an increase in the RS-group (210 (121)%) was observed. Eight hours after induction of sepsis, formation of microvesicles was significantly higher in the RS group compared to all colloid-treated groups. CONCLUSION: In this porcine septic shock model the formation of platelet-derived microvesicles was significantly increased by volume replacement with Ringer's solution in comparison to colloid solutions.

[Massive transfusion of washed red blood cells: acid-base and electrolyth changes for different wash solutions]. / Massivtransfusionen mit gewaschenen Erythrozyten: Beeinflussung des Säure-Basen-Elektrolyt-Haushalts durch verschiedene Waschlösungen.

OBJECTIVE: The composition of normal saline (NaCl), the standard wash solution for cell saver autotransfusion, is considerably different from physiologic plasma values in small infants. Therefore, we investigated acid-base and electrolyte changes during massive cell saver autotransfusion with different wash solutions in young pigs. METHODS: After approval by the animal protection authorities 15 young pigs (weight 10.6 +/- 1.1 kg, blood volume 848 +/- 88 ml, mean+/-SD) underwent 15 cycles of cell saver autotransfusion (Haemolite 2plus, Haemonetics). For each cycle, 100 ml arterial blood was withdrawn, washed with NaCl, physiologic multielectrolyte solution (PME, V Infusionslösung 296 mval Elektrolyte, Baxter) or physiologic erythrocyte protection solution (PEP, 3.2 % gelatine, pH 7.40, cHCO3 24 mmol/l), and then retransfused. Analyses of acid-base, electrolyte, and hematologic parameters were performed for systemic and washed blood samples. RESULTS: For NaCl there was a progressive decrease in systemic pH, HCO3 and base excess (BE) and an increase in chloride values (Cl) (p < 0.05). Use of PME slightly decreased pH (n. s.), whereas HCO3, BE and Cl remained stable. PEP slightly increased pH, HCO3 and BE, and decreased Cl (n. s.). Free hemoglobin increased in NaCl and PME (p < 0.05) and was below baseline in PEP (n. s.). Lactic acid course was comparable in all groups. CONCLUSIONS: The use of NaCl as wash solution for massive autotransfusion resulted in metabolic acidosis caused by dilution of HCO3 and increased Cl values. Fewer systemic acid-base and electrolyte changes were observed, when blood was washed with PME or PEP. The decreased hemoglobin release with PEP is possibly due to a gelatine specific electrostatic surface coating of erythrocyte membranes. For massive transfusion of washed red blood cells, physiologic multielectrolyte solution and physiologic erythrocyte protection solution should be preferred to NaCl, especially for small infants.

Hydroxyethyl starch and modified fluid gelatin maintain plasma volume in a porcine model of septic shock with capillary leakage.

OBJECTIVE: To compare the effects of different volume replacement therapies on maintenance of plasma volume in septic shock and capillary leakage syndrome. DESIGN AND SETTING: Prospective randomized, controlled animal laboratory study in a university animal laboratory. MEASUREMENTS AND RESULTS: Twenty-five fasted, anaesthetized, mechanically ventilated and multi-catheterized pigs (20.8+/-1.8 kg) received 1 g/kg body weight faeces into abdominal cavity to induce sepsis and were observed over 8 h. Five animals each received volume replacement therapy with modified fluid gelatin 4% or 8% (MFG4%, MFG8%), 6% HES 200/0.5, or Ringer's solution and were compared to controls receiving 6% HES 200/0.5. Infusion rate was titrated to maintain a central venous pressure of 12 mmHg. Plasma volume was determined using (51)Cr-labelled erythrocytes and standard formulae. Albumin escape rate was calculated using technetium (99m)Tc-labelled albumin. Colloid osmotic pressure, systemic haemodynamics and oxygenation were obtained before and 4 and 8 h after induction of sepsis. Plasma volume was reduced in the Ringer's solution group (-46%) but was maintained in HES (+/-0%), MFG4% (+4%), MFG8% (+23%) groups. Albumin escape rate increased in HES (+52%), MFG4% (+47%), MFG8% (+54%) and the Ringer's solution group (+41%) compared to controls. CONCLUSION: In this porcine septic shock model with concomitant capillary leakage syndrome, confirmed by an increased albumin escape rate, the artificial colloids HES, MFG4%, and MFG8% maintained plasma volume and colloid osmotic pressure. These results suggest the intravascular persistency of artificial colloids in the presence of albumin leakage. An editorial regarding this article can be found in the same issue (http://dx.doi.org/10.1007/s00134-002-1283-9)

Acid-base, electrolyte and metabolite concentrations in packed red blood cells for major transfusion in infants.

METHODS: Acid-base, electrolyte and metabolite concentrations were determined in 100 U of packed red blood cells (RBC) preserved in extended-storage media to be used for major transfusion in paediatric and cardiac surgery. RESULTS: In fresh RBC, low pH, bicarbonate (cHCO3-), base excess (BE) and high glucose values were observed all outside the physiological range. With lengthening storage duration, values of pH, cHCO3-, BE, sodium and glucose decreased and carbon dioxide, potassium and lactic acid concentrations increased [mean +/- SD (range): storage duration 6.7 +/- 3.8 (1-17) days, pH 6.79 +/- 0.1 (6.53-6.99); mmol.l-1: cHCO3- 11.1 +/- 1.5 (6.2-14.5), BE - 29.2 +/- 4.1 ([-39.4] - [-20.9]), potassium 20.5 +/- 7.8 (4.2-43.6), glucose 24.1 +/- 6.1 (16.7-29.2), lactic acid 9.4 +/- 4 (4.3-21.4)]. CONCLUSION: Massive and rapid transfusion of RBC may lead to a severe burden of hydrogen ions, carbon dioxide, potassium, glucose and lactic acid and this can be avoided by cell saver blood processing, when autologous erythrocytes from the operative field are saved and substrate load and storage lesions from packed red blood cells are minimized in one step by washing.

Protective effects of plasma replacement fluids on erythrocytes exposed to mechanical stress.

Haemoglobin release from 40 suspensions of packed red blood cells in modified fluid gelatin, 4% albumin solution, 6% hydroxyethyl starch and normal saline was investigated in vitro during circulation with a roller pump from a heart-lung machine for 120 min at a flow rate of 2.5 l.min-1 at room temperature. The lowest haemoglobin release was obtained with erythrocytes in modified fluid gelatin, whereas free haemoglobin concentrations became progressively higher with albumin, hydroxyethyl starch and normal saline [median free haemoglobin (interquartile range) after 120 min circulation: gelatin 493 (360-601) mg.l-1, albumin 692 (590-1111) mg.l-1, hydroxyethyl starch 1121 (692-1518) mg.l-1, normal saline 1178 (881-1757) mg.l-1, p < 0.001]. Modified fluid gelatin appears to have potent erythrocyte protective properties similar to those of albumin. This effect could decrease mechanical haemolysis during extracorporeal circulation or cell saver autotransfusion if modified fluid gelatin is used as part of a priming solution or as an additive in wash solutions.

Haemodynamic, acid-base and electrolyte changes during plasma replacement with hydroxyethyl starch or crystalloid solution in young pigs.

We investigated haemodynamic, acid-base and electrolyte changes during almost total plasma replacement with hydroxyethyl starch (HES) and physiological balanced electrolyte solution (PBE) by using a cell saver in ten young pigs. In the PBE group an additional 3550 (444) ml crystalloid solution [Mean (SD)] was infused over the course of the study in order to maintain pulmonary capillary wedge pressure. Plasma protein levels decreased in both groups and the colloid osmotic pressure increased in HES and decreased in PBE. At the end of the study, body weight [HES 10.4 (1), PBE 13.1 (1.4) kg, P < 0.01] and lactic acid concentration [HES 0.9 (0.3), PBE 2.9 (1. 3) mmol.l -1, P < 0.01] was higher and tissue oxygen delivery [HES 327 (22), PBE 89 (29) ml.min.m2, P < 0.01] was lower in the PBE group. There were only moderate acid-base changes in both groups, but at the end, anion gap was significant lower in HES. In conclusion, maintenance of colloid osmotic pressure close to the physiological range of infants seems to be advantageous during major paediatric surgery.

Haemoconcentration by gelatin-induced acceleration of erythrocyte sedimentation rate.

Erythrocyte sedimentation rates from 40 suspensions of packed red blood cells in modified fluid gelatin, 4% albumin solution, 6% hydroxyethyl starch and normal saline were measured at room temperature using Westergren's method. The erythrocyte sedimentation rate was extremely high in gelatin and this increase was significant after 10-60 min when compared with the other fluids. Erythrocyte sedimentation rates in albumin, hydroxyethyl starch and normal saline were low and there were no differences between these fluids [erythrocyte sedimentation after 60 min, median (interquartile range): gelatin 128 (111.2-130.0) mm, albumin 2 (1.5-2.0) mm, hydroxyethyl starch 1.5 (1.0-1.6) mm, normal saline 2 (1.9-2.5) mm, p < 0.0001]. The addition of twice the volume of modified fluid gelatin to a volume of red blood cells leads to rapid acceleration of the erythrocyte sedimentation rate. This is caused by increased erythrocyte aggregation, and in clinical practice this effect may be useful for the haemoconcentration of diluted blood from cardiopulmonary bypass circuits or cell-saver autotransfusion in paediatric surgery.

[Intraoperative changes in arterial end tidal CO2 partial pressure difference in interventions with constant ventilation-perfusion ratio]. / Intraoperative Veränderungen der arterio-endtidalen CO2-Partialdruckdifferenz bei Eingriffen mit konstanten Ventilations-Perfusions-Verhältnissen.

During general anaesthesia, the endtidal CO2 pressure serves as an estimate of the arterial CO2 pressure to regulate the ventilator setting. Important arterial to end-tidal carbon dioxide tension differences (P(a-et)CO2) have been observed among patients undergoing procedures which have substantial impact on the ventilation-perfusion ratio (V/Q). Data on the P(a-et)CO2 for procedures in which the V/Q-ratio remains constant are lacking. Repeated measurements of P(a-et)CO2 in twelve patients with chronic obstructive lung disease (COLD) and nine pulmonary healthy patients undergoing jaw surgery were performed. The P(a-et)CO2 in the pulmonary healthy subjects (5.96 +/- 1.68 mmHg) was lower than in the COLD patients (9.05 +/- 3.49 mmHg) (p < 0.01). A clinically significant P(a-et)CO2 > or = 8 mmHg was observed in 52% of the measurements in patients with COLD compared with 11% in the pulmonary healthy subjects (p < 0.01). Both patient groups showed only minimal intraoperative changes of P(a-et)CO2. The deviation of all subsequent P(a-et)CO2 values from the initial P(a-et)CO2 was 2.17 +/- 1.52 mmHg in the pulmonary healthy patients and 2.02 +/- 1.49 mmHg in the patients with COLD (p = 0.76). Intraoperative changes of the P(a-et)CO2 are small during procedures with no major alterations of the V/Q ratio. For these procedures an initial measurement of the P(a-et)CO2 in patients with lung disease should be sufficient. In pulmonary healthy subjects the P(a-et)CO2 seems to be negligible.

Effect of magnetic resonance imaging on human respiratory burst of neutrophils.

It is known that low intensity magnetic fields increase superoxide anion production during the respiratory burst of rat peritoneal neutrophils in vitro. We investigated whether the high intensity magnetic fields (1.5 T) during magnetic resonance imaging can influence the human neutrophil function under in vivo conditions. Blood samples were obtained from 12 patients immediately before and after magnetic resonance imaging (mean time 27.6(+/-11.4 min)). The induced respiratory burst was investigated by the intracellular oxidative transformation of dihydrorhodamine 123 to the fluorescent dye rhodamine 123 via flow cytometry. The respiratory burst was induced either with phorbol 12-myristate 13-acetate, Escherichia coli, N-formyl-methionyl-leucylphenylalanine or priming with tumor necrosis factor followed by FMLP stimulation. There was no significant difference between the respiratory burst before and after magnetic resonance imaging, irrespective of the stimulating agent. Short time exposure to a high intensity magnetic field during magnetic resonance imaging seems not to influence the production of radical species in living neutrophils.

Alteration of anion gap during almost total plasma replacement with synthetic colloids in piglets.

In ten piglets (body weight 8.2-11.6 kg), acid base, electrolyte and anion gap changes were investigated during almost total plasma replacement with hydroxyethyl starch (HES) and modified fluid gelatin (GEL) in saline solution using a cell saver autotransfusion technique. During the study, there were only moderate acid base changes, but marked disturbances in anion balance. At study end, the mean chloride concentration was significantly higher (mmol/l: normal values 97-108, HES 116 +/- 1.5, GEL 108 +/- 1.1, p < 0.01) and the mean anion gap was significantly lower in the HES group in comparison to the GEL group (mmol/l: normal values 5-14, HES 3 +/- 1.7, GEL 11.9 +/- 0.9, p < 0.01). It is concluded that plasma replacement with electroneutral HES, but not with negatively charged GEL, can lower the anion gap irrespective of the underlying disease. This can be misleading when the anion gap is used for differential diagnosis of metabolic acidosis in patients after large volume infusion of synthetic colloids.