Cardiac arrest in intensive care unit: Case report and future recommendations.

Initiation of hemofiltration in a patient in septic shock can cause hemodynamic compromise potentially leading to cardiac arrest. We propose that the standard '4Hs and 4Ts' approach to the differential diagnosis of a cardiac arrest should be supplemented in critically ill patients with anaphylaxis and human and technical errors involving drug administration (the 5(th) H and T). To illustrate the point, we report a case where norepinephrine infused through a central venous catheter (CVC) was being removed by the central venovenous hemofiltration (CVVH) catheter causing the hemodynamic instability. CVVH has this potential of interfering with the systemic availability of drugs infused via a closely located CVC.

Hyperglycemia during hypothermic cardiopulmonary bypass does not alter postbypass vascular endothelial responses in dogs.

BACKGROUND: Hyperglycemia during hypothermic cardiopulmonary bypass (CPB) may alter intrinsic vasomotion by reducing endothelial-dependent vasorelaxation. Using a canine model of hypothermic CPB, this study tested whether hyperglycemia altered the vasodilator response to acetylcholine (ACh) and the vasoconstrictor response to phenylephrine (Phe). METHODS: In 20 anesthetized dogs, the left femoral arteries were excised and placed in gassed (95% O2-5% CO2) cold Krebs's solution. The animals were randomized into two groups undergoing 120 minutes of 28 degrees C CPB using membrane oxygenators. A hyperglycemic group (n = 10) received a continuous infusion of 50% dextrose to maintain blood glucose level greater than 500 mg/dL; a normoglycemic group (n = 10) received 0.9% saline. After rewarming and discontinuing CPB, the right femoral arteries were excised. Vessel rings were placed in a suffusion bath, and changes in isometric tension were measured. Dose-response relationships (ACh: 10(-9) to 10(-6)M; Phe: 3 x 10(-8) to 10(-4)M) and -log ED50 sensitivity to ACh and Phe before and after CPB were compared. RESULTS: Serum glucose during hypothermic CPB was significantly greater in glucose-treated dogs (525 +/- 9 mg/dL) than controls (109 +/- 5 mg/dL; p < 0.05). After CPB, -log ED50 values for ACh changed from 7.7 +/- 0.1 to 7.5 +/- 0.2 (p < 0.05) in normoglycemic dogs and from 7.8 +/- 0.1 to 7.6 +/- 0.1 (p < 0.05) in hyperglycemic animals, indicating similar and significant rightward shifts of the dose-response relationship to ACh after CPB in both groups. Neither hyperglycemia nor CPB altered the vasoconstrictor response to Phe. CONCLUSIONS: The reduction in ACh-mediated vasorelaxation after CPB did not differ between hyperglycemic and normoglycemic animals, indicating that hyperglycemia does not contribute to impaired vasorelaxation after CPB. Because Phe-induced vasoconstriction was unaffected, hyperglycemia during hypothermic CPB does not appear to increase the potential for postbypass vasospasm.

Volume expansion increases right ventricular infarct size in dogs by reducing collateral perfusion.

STUDY OBJECTIVE: Plasma volume expansion is frequently recommended to correct the low output state resulting from right ventricular (RV) infarction. However, any subsequent increase in pericardial and RV filling pressures from volume expansion could impair RV collateral blood flow. We examined whether volume expansion in dogs before right coronary ligation reduced collateral perfusion and worsened the extent of RV necrosis. DESIGN: Randomized experimental study. SETTING: Animal research laboratory in university medical center. PARTICIPANTS: Forty anesthetized, closed-chest dogs were randomly assigned to normovolemic, pericardium opened (n = 10) or intact (n = 10) groups, and hypervolemic, pericardium opened (n = 10) or intact (n = 10) groups. INTERVENTIONS: Hypervolemic animals received 24 mL/kg of 6% hetastarch. All animals underwent 90 min right coronary ligation, followed by 120 min reperfusion. Collateral coronary blood flow (radioactive microspheres) and area of necrosis (An) were determined in the area at risk (Ar). MEASUREMENTS AND RESULTS: Stroke volume decreased in all groups with ischemia but remained 25 to 40% greater in both hypervolemic groups than in normovolemic animals (p < 0.05). In hypervolemic animals with intact pericardium, RV end-diastolic pressure increased to 10.4 +/- 2.1 mm Hg (mean +/- SD), a value that significantly exceeded those of the other three groups. During RV ischemia, collateral perfusion in the Ar was similar in both normovolemic groups and in hypervolemic animals with opened pericardium (mean range, 12.9 +/- 8.8 to 13.8 +/- 7.6 mL/min/100 g; p = NS), and the An/Ar varied from 11.8 +/- 6.3 to 18.6 +/- 17.4% (p = NS). In contrast, in hypervolemic animals with intact pericardium, collateral perfusion decreased to 7.2 +/- 3.5 mL/min/100 g and the An/Ar was increased to 38.2 +/- 18.6% (p < 0.05 compared with other groups, respectively). Overall, An/Ar was inversely related to collateral blood flow in the Ar (r = -0.46; p < 0.05) and correlated positively with RV end-diastolic pressure (r = 0.61; p < 0.05). CONCLUSIONS: Volume expansion preserved stroke volume during RV ischemia, independent of pericardial integrity. However, volume expansion in animals with an intact pericardium increased RV infarct size by twofold to threefold secondary to reduced periischemic collateral perfusion. This detrimental effect of volume expansion on infarct size was prevented by opening the pericardium.

Cerebral metabolic consequences of hypotensive challenges in hemodiluted pigs with and without cardiopulmonary bypass.

We tested the hypothesis that progressive aortic hypotension with bicarotid occlusion produces greater reductions in cerebral blood flow (CBF) and more flow-metabolism mismatching with hemodilution during cardiopulmonary bypass (CPB) than with hemodilution alone. In Yorkshire pigs randomized to hemodilution with CPB (n = 10) or hemodilution without CPB (control; n = 9), the effects of bicarotid ligation and graded hypotension on CBF (microspheres), the electroencephalogram (EEG), and cortical energy metabolites were examined. After bicarotid ligation, systemic flow was reduced for 15-min intervals of 80, 60, and 40 mm Hg aortic pressure, followed by a cortical brain biopsy. At baseline, CBF was lower in CPB (58 +/- 3 mL.100g-1.min-1) than control (90 +/- 3 mL.100 g-1.min-1., P < 0.05) animals, as was cerebral oxygen metabolism (3.1 +/- 0.1 vs 4.2 +/- 0.2 mL.min-1.100g-1; P < 0.05). Although CBF remained 40% lower at each level of hypotension in CPB than control animals (P < 0.05), EEG scores showed no intergroup differences, indicating similar flow-metabolism matching. Brain metabolites were similar between CPB and control groups (adenosine triphosphate, 9.6 +/- 2.4 vs 12.4 +/- 1.9 mumol/g; adenosine diphosphate, 6.0 +/- 0.7 vs 6.3 +/- 0.4 mumol/g; adenosine monophosphate, 4.8 +/- 0.9 vs 3.8 +/- 0.8 mumol/g; creatine phosphate, 8.3 +/- 1.8 vs 7.9 +/- 1.0 mumol/g; and lactate, 178.4 +/- 20.2 vs 150.8 +/- 13.9 mumol/g). Thus, despite significantly lower CBF during hypotension with bicarotid occlusion in hemodiluted animals during normothermic CPB, cortical electrical activity and the balance between flow and metabolism did not differ from those in control animals without CPB.

Systemic gaseous microemboli during left atrial catheterization: a common occurrence?

OBJECTIVE: Gaseous microemboli during cardiac surgery have been implicated as a potential cause of postoperative neurologic injury. Any monitoring technique that exposes the systemic circulation to atmospheric pressure could introduce gaseous microemboli, causing cerebral microembolization. The incidence of carotid artery gaseous microemboli was studied during left atrial catheter insertion. DESIGN: Prospective clinical study. SETTING: Tertiary care university hospital. PARTICIPANTS: Twelve patients undergoing elective cardiac surgery. INTERVENTIONS: Perioperatively, a 5-MHz continuous wave Doppler probe was positioned over the left carotid artery to maximally record blood flow signals. The criteria used for detecting a gaseous microembolus were a sudden increase in the amplitude of the visual signal by 30% and a characteristic audible sound. MEASUREMENTS AND MAIN RESULTS: Numbers of microemboli at three timepoints (before and during left atrial catheter insertion and during catheter flushing) were assessed using the Friedman test. No emboli were detected before left atrial catheter insertion. When compared with the preinsertion time period, statistically (p < 0.05) significant numbers of gaseous microemboli were found in six patients during catheter insertion (3 +/- 1 microemboli; range 1 to 7 microemboli) and in five patients during catheter flushing (5 +/- 2 microemboli; range 1 to 12 microemboli). There was a tendency for patients with lower filling pressures to entrain more microemboli during insertion (r = 0.44; p = 0.149). No patient showed evidence of gross neurologic dysfunction postoperatively, although sensitive neurologic testing was not performed. CONCLUSIONS: Left atrial catheter insertion and flushing can cause systemic gaseous microemboli in more than 50% of patients. Although the number of microemboli introduced is relatively small, extreme care should be used during left atrial catheter insertion.

Increasing organ blood flow during cardiopulmonary bypass in pigs: comparison of dopamine and perfusion pressure.

OBJECTIVE: To determine whether low-dose dopamine infusion (5 micrograms/kg/min) during cardiopulmonary bypass selectively increases perfusion to the kidney, splanchnic organs, and brain at low (45 mm Hg) as well as high (90 mm Hg) perfusion pressures. DESIGN: Randomized crossover trial. SETTING: Animal research laboratory in a university medical center. SUBJECTS: Ten female Yorkshire pigs (weight 29.9 +/- 1.2 kg). INTERVENTION: Anesthetized pigs were placed on normothermic cardiopulmonary bypass at a 100-mL/kg/min flow rate. After baseline measurements, the animal was subjected, in random sequence, to 15-min periods of low perfusion pressure (45 mm Hg), low perfusion pressure with dopamine (5 micrograms/kg/min), high perfusion pressure (90 mm Hg), and high perfusion pressure with dopamine. Regional perfusion (radioactive microspheres) was measured in tissue samples (2 to 10 g) from the renal cortex (outer two-third and inner one-third segments), stomach, duodenum, jejunum, ileum, colon, pancreas, and cerebral hemispheres. MEASUREMENTS AND MAIN RESULTS: Systemic perfusion pressure was altered by adjusting pump flow rate (r2 = .61; p < .05). In the kidney, cortical perfusion pressure increased from 178 +/- 16 mL/min/100 g at the low perfusion pressure to 399 +/- 23 mL/min/100 g at the high perfusion pressure (p < .05). Perfusion pressure augmentation increased the ratio of outer/inner renal cortical blood flow from 0.9 +/- 0.1 to 1.2 +/- 0.1 (p < .05). At each perfusion pressure, low-dose dopamine had no beneficial effect on renal perfusion or flow distribution. Similar results were found in the splanchnic organs, where regional perfusion was altered by perfusion pressure but not by dopamine. In contrast, neither changing perfusion pressure nor adding low-dose dopamine altered blood flow to the cerebral cortex. CONCLUSIONS: These data indicate that the lower autoregulatory limits of perfusion to the kidneys and splanchnic organs differ from those limits to the brain during normothermic bypass. Selective vasodilation from low-dose dopamine was not found in renal, splanchnic, or cerebral vascular beds. Increasing the perfusion pressure by pump flow, rather than by the addition of low-dose dopamine, enhanced renal and splanchnic but not cerebral blood flows during cardiopulmonary bypass.

Hyperglycemia during hypothermic canine cardiopulmonary bypass increases cerebral lactate.

BACKGROUND: Hyperglycemia frequently occurs during cardiopulmonary bypass (CPB), although its direct effects on cerebral perfusion and metabolism are not known. Using a canine model of hypothermic CPB, we tested whether hyperglycemia alters cerebral blood flow and metabolism and cerebral energy charge. METHODS: Twenty anesthetized dogs were randomized into hyperglycemic (n = 10) and normoglycemic (n = 10) groups. The hyperglycemic group received an infusion of D50W, and the normoglycemic animals received an equal volume of 0.9% NaCl. Both groups underwent 120 min of hypothermic (28 degrees C) CPB using membrane oxygenators, followed by rewarming and termination of CPB. Cerebral blood flow (radioactive microspheres) and the cerebral metabolic rate for oxygen were measured intermittently during the experiment and brain tissue metabolites were obtained after bypass. RESULTS: Before CPB, the glucose-treated animals had higher serum glucose levels (534 +/- 12 mg/dL; mean +/- SE) than controls (103 +/- 4 mg/dL; P < 0.05), and this difference was maintained throughout the study. Cerebral blood flow and metabolism did not differ between groups at any time during the experiment. Sagittal sinus pressure was comparable between groups throughout CPB. Tissue high-energy phosphates and water contents were similar after CPB, although cerebral lactate levels were greater in hyperglycemic (37.2 +/- 5.7 mumol/g) than normoglycemic animals (19.7 +/- 3.7 mumol/g; P < 0.05). After CPB, pH values of cerebrospinal fluid for normoglycemic (7.33 +/- 0.01) and hyperglycemic (7.34 +/- 0.01) groups were similar. CONCLUSIONS: Hyperglycemia during CPB significantly increases cerebral lactate levels without adversely affecting cerebral blood flow and metabolism, cerebrospinal fluid pH, or cerebral energy charge.

Cardiopulmonary bypass impairs vascular endothelial relaxation: effects of gaseous microemboli in dogs.

Gaseous microemboli during hypothermic cardiopulmonary bypass (CPB) may injure the vascular endothelium and interfere with intrinsic vasomotion. We tested whether gaseous microemboli reduced the vasodilator response to acetylcholine (ACh, 10(-9)-10(-6) M) and potentiated the vasoconstrictor response to norepinephrine (NE, 3 x 10(-8)-10(-4) M). Arteries from 18 dogs were excised before and after 120 min 28 degrees C CPB using membrane (n = 9) and bubble (n = 9) oxygenators to produce microemboli, which were quantitated by Doppler. Five nonbypassed dogs were controls. In isolated vessel rings, the 50% effective dose (ED50) values for ACh (10(-8) M) and NE (10(-7) M) responses were calculated. Mean microemboli count per minute was 0 +/- 0 in the control group, 1.0 +/- 0.4 in the membrane group (P < 0.05 vs. controls), and 46.9 +/- 8.4 in the bubble group (P < 0.05 vs. control and membrane groups). ACh ED50 values did not change in controls but increased in the membrane group from 4.01 +/- 1.52 to 5.66 +/- 1.39 (P < 0.05) and in the bubble group from 2.32 +/- 0.56 to 7.21 +/- 1.90 (P < 0.05). The change in ED50 was greater for bubble than for membrane animals (P < 0.05) but did not correlate with microemboli number (bubble: r = 0.392, P = 0.297; membrane: r = 0.058, P = 0.802). NE responses were similar in all groups. Hypothermic CPB reduces ACh-induced dilation of the canine femoral artery independent of the incidence of gaseous microemboli.(ABSTRACT TRUNCATED AT 250 WORDS)

Clinical audit in intensive care--the Meath Intensive Care Audit.

Audit is now recognised as being an essential component of clinical practice. We report on the first year of the Meath Intensive Care Audit (MICA). This audit was instituted to investigate the activity of the unit, to assess the feasibility of continuous audit in our ICU and to provide data for future development of ICU facilities. Two hundred and fifty four patients were admitted between July 1st 1990 and June 30th 1991. The mean age at admission was 58 years and the mean length of stay 5.2 days. The mean APACHE II score was 16. Thirty four patients (13.4%) died in the ICU and 17 patients died in hospital following discharge from the unit bringing the hospital mortality rate to 20%. The audit proved feasible to implement and data collection is now accepted as a routine part of our ICU work.