D. Grimaud

Low exogenous lactate clearance as an early predictor of mortality in normolactatemic critically ill septic patients

ObjectiveTo evaluate the prognostic value of lactate clearance and lactate production in severely ill septic patients with normal or mildly elevated blood lactate concentration. DesignProspective, observational study. SettingNineteen-bed mixed medicosurgical intensive care unit. PatientsFifty-six patients with severe sepsis and blood lactate concentration <3 mmol/L. Measurements and Main ResultsLactate metabolism was evaluated in all patients. Lactate clearance was measured by modeling the change in arterial blood lactate over time induced by an infusion of 1 mmol/kg sodium lactate for 15 mins. Lactate production was calculated as the product of lactate clearance times the blood lactate concentration before the infusion. Outcome was taken to be mortality at 28 days after the beginning of the septic episode. A logistic regression model taking into account different risk factors was constructed. Among the 56 patients, 17 (30.3%) died before the 28th day. Basal blood lactate concentration was not different between survivors and nonsurvivors, whereas lactate clearance and production were higher in survivors (0.86 ± 0.32 vs. 0.58 ± 0.18 L/hr/kg, p < .005, and 1.19 ± 0.63 vs. 0.89 ± 0.24 mmol/hr/kg, p = .055, respectively). An increase in blood lactate 45 mins after the end of the lactate infusion (&Dgr;lact-T60) ≥0.6 mmol/L was predictive of 28-day mortality with 53% sensitivity and 90% specificity. Multivariate analysis showed that only three factors were independently and significantly correlated with 28-day mortality: presence of more than two organ failures (odds ratio, 27;p = .04), age >70 yrs (odds ratio, 5.7;p = .032), and &Dgr;lact-T60 ≥0.6 mmol/L (odds ratio, 14.2;p = .042). ConclusionLow lactate clearance in severely ill septic patients with normal or mildly elevated blood lactate is predictive of poor outcome independently of other known risk factors such as age and number of organ failures.

Effect of continuous venovenous hemofiltration with dialysis on lactate clearance in critically ill patients

OBJECTIVE To evaluate the effect of continuous venovenous hemofiltration with dialysis on lactate elimination by critically ill patients. DESIGN Prospective, clinical study. SETTING Surgical intensive care unit of a university hospital. PATIENTS Ten critically ill patients with acute renal failure and stable blood lactate concentrations. INTERVENTIONS Two-stage investigation: a) measurement of lactate concentrations in samples of serum and ultradiafiltrate from patients receiving continuous venovenous hemofiltration with dialysis to calculate lactate clearance by the hemofilter; b) evaluation of total plasma lactate clearance by infusing sodium L-lactate (1 mmol/kg of body weight) over 15 mins. MEASUREMENTS AND MAIN RESULTS Arterial lactate concentration was determined before, during, and after the infusion. Lactate elimination variables were calculated from the plasma curve using model-independent and model-dependent estimates (by software). At the end of the infusion, median blood lactate concentration increased from 1.4 mmol/L (range 0.8 to 2.6) to 4.8 mmol/L (range 2.4 to 5.7) and returned to 1.6 mmol/L (range 0.9 to 3.4) 60 mins later. The median total plasma lactate clearance was 1379 mL/min (range 753.7 to 1880.7) and the median filter lactate clearance was 24.2 mL/min (range 7.1 to 35.6). Thus, filter lactate clearance accounted for < 3% of total lactate clearance. CONCLUSIONS Continuous venovenous hemofiltration with dialysis cannot mask lactate overproduction, and its blood concentration remains a reliable marker of tissue oxygenation in patients receiving this renal replacement technique.

An evaluation of the brachial plexus block at the humeral canal using a neurostimulator (1417 patients): the efficacy, safety, and predictive criteria of failure.

To evaluate the efficacy and safety of the multiple peripheral nerve block technique at the humeral canal (humeral block) with the use of a neurostimulator, we prospectively studied 1417 patients undergoing upper limb surgery with a brachial plexus block at the humeral canal (1468 blocks). The success rate (defined as sensory block [in all nerve distributions] and/or the absence of another anesthetic technique required to allow surgery) was 95%. The threshold of minimal stimulation used to locate each nerve before injecting the anesthetic solution was the unique predictive factor for identified failure. For the median nerve, the threshold was 0.8 mA with a relative risk of failure (RRf: relative risk evaluated by series of Taylor with a 95% confidence interval) = 1.49 (P = 0.04), for the radial nerve the threshold was 0.6 mA (RRf 1.3, P = 0.02), and 0.7 mA for the ulnar nerve (RRf 1.36, P = 0.04). For any equal or higher stimulation level, the risk of failure of the humeral block increased. For the musculocutaneous nerve, we did not observe a significant stimulation threshold for the risk of failure; although beyond 0.7 mA, the RRf was always more than 1.3. Adverse events occurred in 7% of all cases and were usually minor (nausea/vomiting, anxiety, local pain). Our study provides supplementary information on the efficacy and safety of this technique. Stimulation thresholds are clinically identified for the first time as the main factor linked to the failure of a technique using a neurostimulator. We conclude that the humeral block is a reliable peripheral block allowing good success rates results with minor complications, which can be used as an alternative to the axillary block. Implications We prospectively evaluated the feasability and the factors causing failure of a peripheral nerve block technique (humeral block) using neurostimulation in a large number of patients. The importance of the level of stimulation for the success of the block was evaluated for the first time.

Effects of dopamine and norepinephrine on systemic and hepatosplanchnic hemodynamics, oxygen exchange, and energy balance in vasoplegic septic patients

Dopamine is widely used to improve systemic and hepatosplanchnic hemodynamics and oxygenation during sepsis. However, some studies have suggest that norepinephrine may have beneficial effects on regional blood flow and metabolism, whereas dopamine might have deleterious effects related to redistribution of blood flow away from the intestinal mucosa or by decreasing directly the cell redox state. In 12 vasoplegic septic patients, we compared the effects of norepinephrine and dopamine on systemic and hepatosplanchnic hemodynamics, oxygenation, and energy metabolism. Catecholamines were administered in a crossover randomized order to maintain mean arterial pressure (MAP) at 80 mmHg. Hepatosplanchnic blood flow (Qspl) was determined using a continuous infusion of indocyanine green dye. Despite a similar MAP, the cardiac index was higher with dopamine than with norepinephrine (6.3 [5.3-7.3] vs. 4.3 [3.8-4.9] L·min−1·m−2) (P < 0.001). Qspl was similar with both catecholamines, but the ratio of Qspl to cardiac output was significantly lower with dopamine (23.9% [17.5-33.5]) than with norepinephrine (33.5% [25.8-37]) (P < 0.05). Although global O2 delivery and O2 consumption were higher with dopamine (782 [707-859] vs. 553 [512-629] mL·min−1·m−2, P < 0.001 and 164 [134-192] vs. 128 [111-149] mL·min−1·m−2, P < 0.001, respectively), hepatosplanchnic O2 delivery and consumption were not different. Hepatic lactate uptake was lower (0.47 [0.3-0.89] vs. 1.01 [0.69-1.34] mmol·min−1) (P < 0.01), and hepatic venous lactate-to-pyruvate ratio was higher (15.3 [7.6-21.1] vs. 11.2 [6.6-15.1], P < 0.05) with dopamine than with norepinephrine. In vasoplegic septic patients, maintaining mean arterial pressure, hepatosplanchnic hemodynamics, and oxygen exchange with dopamine requires a consequent increased cardiac output, which is responsible for an increased global oxygen demand when compared with norepinephrine. In addition, dopamine impairs the hepatic energy balance. Its position as a preferential treatment compared with norepinephrine in this context may therefore be questionable.

Prolonged low-dose dopamine infusion induces a transient improvement in renal function in hemodynamically stable, critically ill patients: a single-blind, prospective, controlled study.

Objective: To evaluate the length of the effects of long‐term (48 hrs), low‐dose dopamine infusion on both renal function and systemic hemodynamic variables in stable nonoliguric critically ill patients. Design: Prospective, single‐blind, controlled clinical study. Setting: University hospital, 19‐bed multidisciplinary intensive care unit. Patients: Eight hemodynamically stable, critically ill patients with a mild nonoliguric renal impairment (creatinine clearance between 30 and 80 mL/min). Interventions: Each patient consecutively received 4 hrs of placebo, followed by a 3 μg/kg/min dopamine infusion during 48 hrs, then a new 4‐hr placebo period. We measured cardiac output and other hemodynamic variables by using a pulmonary artery catheter. The bladder was emptied to determine urine volume and to collect urine samples. Measurements were performed at six times: after the initial control of 4 hrs of placebo (C1); after 4 hrs (H4), 8 hrs (H8), 24 hrs (H24), and 48 hrs (H48) of dopamine infusion; and after the second control of 4 hrs of placebo (C2). Measurements and Main Results: We saw no significant change in systemic hemodynamic variables with dopamine at all times of infusion. Diuresis, creatinine clearance, and the fractional excretion of sodium (FeNa) at C1 and C2 were not different. Urine flow, creatinine clearance, and FeNa increased significantly 4 hrs after starting dopamine (for all these changes, p < .01 vs. C1 and C2). The maximum changes were obtained at H8, with an increase of 50% for diuresis, 37% for creatinine clearance, and 85% for FeNa (for all these changes, p < .01 vs. C1 and C2). But these effects waned progressively from H24, and both creatinine clearance and FeNa at H48 did not differ from control values. Conclusions: In stable critically ill patients, preventive low‐dose dopamine increased creatinine clearance, diuresis, and the fractional excretion of sodium without concomitant hemodynamic change. These effects reached a maximum during 8 hrs of dopamine infusion. But despite a slight persistent increase in diuresis, improvement in creatinine clearance and FeNa disappeared after 48 hrs. According to these data, it is likely that tolerance develops to dopamine‐receptor agonists in critically ill patients at risk of developing acute renal failure.

Initial effect of sodium bicarbonate on intracellular pH depends on the extracellular nonbicarbonate buffering capacity.

ObjectiveThe effect of sodium bicarbonate on intracellular pH under conditions close to those in vivo, with both bicarbonate and nonbicarbonate buffering systems, is unknown. We postulated that this effect depends on the nonbicarbonate buffering capacity because the alkali-induced back-titration of these buffers results in a concentration-dependent release of CO2 in the extracellular space, leading to a decrease in intracellular pH. DesignThe study was conducted in two stages. First, human hepatocytes were perfused with pH 7 bicarbonate-buffered medium (5 mM HCO3−, 20 torr Pco2) containing no nonbicarbonate buffer or small amounts (5 mM 4-[2-hydroxyethyl]-1-piperazineethanesulfonic acid [HEPES]) or large amounts (20 mM HEPES) of nonbicarbonate buffer. Second, the changes in intracellular pH of hepatocytes placed in acidotic human blood (pH 7, 5 mM HCO3−, 20 torr Pco2) at three hematocrits (40%, 20%, and 5%) were measured. SettingResearch laboratory at a medical university. SubjectsCryopreserved human hepatocytes thawed the day before the experiments. InterventionsSodium bicarbonate was infused for 10 mins to increase the HCO3− concentration from 5 to 30 mM. In the second part, 20 mM sodium bicarbonate was added directly to the blood bathing the cells. Measurements and Main Results The intracellular pH was measured with the pH-sensitive fluorescent dye bis-carboxyethyl carboxy-fluorescein in its esterified form, acetoxy-methyl ester, by using a single-cell imaging technique. Gas analyses were performed before and during the sodium bicarbonate load. Sodium bicarbonate caused a decrease in intracellular pH with all media except the artificial medium containing no HEPES. This decrease was small in media with low nonbicarbonate buffering capacity (5 mM HEPES and 5% hematocrit blood) and large in media with high nonbicarbonate buffering capacity (20 mM HEPES and 40% hematocrit blood). The change in intracellular pH was linked closely to the change in Pco2 caused by the sodium bicarbonate. ConclusionsThe effect of sodium bicarbonate on intracellular pH depends on changes in Pco2 in the medium bathing the cells. The increase in Pco2 is correlated with the extracellular nonbicarbonate buffering capacity because of the release of H+ ions coming from the back-titration of these buffers. We conclude that sodium bicarbonate may exacerbate cell acidosis under buffering conditions close to those in vivo and that the initial changes in cell pH caused by sodium bicarbonate depend on the extracellular nonbicarbonate buffering capacity.

Treatment of metabolic acidosis.

Metabolic acidosis is characterized by a decrease of the blood pH associated with a decrease in the bicarbonate concentration. This may be secondary to a decrease in the strong ion difference or to an increase in the weak acids concentration, mainly the inorganic phosphorus. From a conceptual point of view, two types of nontoxic metabolic acidosis must be differentiated: the mineral metabolic acidosis that reveals the presence of an excess of nonmetabolizable anions, and the organic metabolic acidosis that reveals an excess of metabolizable anions. Significance and consequences of these two types of acidosis are radically different. Mineral acidosis is not caused by a failure in the energy metabolic pathways, and its treatment is mainly symptomatic by correcting the blood pH (alkali therapy) or accelerating the elimination of excessive mineral anions (renal replacement therapy). On the other hand, organic acidosis gives evidence that a severe underlying metabolic distress is in process. No reliable argument exists to prove that this acidosis is harmful under these conditions in humans. Experimental data even show that hypoxic cells are able to survive only if the medium is kept acidic. The management of an acute organic metabolic acidosis is therefore primarily based on the cause of the acidosis, and no scientific argument exists to justify the correction of the acid–base imbalance in this context.

Reliability of anion gap as an indicator of blood lactate in critically ill patients

Objective: To evaluate the sensitivity, specificity, and predictive values of an elevated anion gap as an indicator of hyperlactatemia and to assess the contribution of blood lactate to the serum anion gap in critically ill patients. Design: Prospective study. Setting: General intensive care unit of a university hospital. Patients: 498 patients, none with ketonuria, severe renal failure or aspirin, glycol, or methanol intoxication. Measurements and results: The anion gap was calculated as [Na+] − [Cl−] − [TCO2]. Hyperlactatemia was defined as a blood lactate concentration above 2.5 mmol/l. The mean blood lactate concentration was 3.7 ± 3.2 mmol/l and the mean serum anion gap was 14.3 ± 4.2 mEq/l. The sensitivity of an elevated anion gap to reveal hyperlactatemia was only 44 % [95 % confidence interval (CI) 38 to 50], whereas specificity was 91 % (CI 87 to 94) and the positive predictive value was 86 % (CI 79 to 90). As expected, the poor sensitivity of the anion gap increased with the lactate threshold value, whereas the specificity decreased [for a blood lactate cut-off of 5 mmol/l: sensitivity = 67 % (CI 58 to 75) and specificity = 83 % (CI 79 to 87)]. The correlation between the serum anion gap and blood lactate was broad (r2 = 0.41, p < 0.001) and the slope of this relationship (0.48 ± 0.026) was less than 1 (p < 0.001). The serum chloride concentration in patients with a normal anion gap (99.1 ± 6.9 mmol/l) was comparable to that in patients with an elevated anion gap (98.8 ± 7.1 mmol/l). Conclusions: An elevated anion gap is not a sensitive indicator of moderate hyperlactatemia, but it is quite specific, provided the other main causes of the elevated anion gap have been eliminated. Changes in blood lactate only account for about half of the changes in anion gap, and serum chloride does not seem to be an important factor in the determination of the serum anion gap.

Comparison of the renal effects of low to high doses of dopamine and dobutamine in critically ill patients: a single-blind randomized study.

Objective: The renal effects of dopamine in critically ill patients remain controversial. Low‐dose dobutamine has been reported to improve renal function. We compared the effects of various doses of dopamine and dobutamine on renal function in critically ill patients. Design: Prospective, single‐blind, randomized study. Setting: University hospital, 19‐bed multidisciplinary intensive care unit. Patients: Twelve hemodynamically stable patients with mild nonoliguric renal impairment. Interventions: Each patient randomly received four different doses of dopamine and dobutamine (placebo, 3, 7, and 12 μg/kg/min). Each infusion lasted for 4 hrs. Cardiac output and systemic hemodynamic variables were measured using a pulmonary arterial catheter at the beginning (HO) and the end (H4) of each infusion. The bladder was emptied at HO and H4 to determine urine volume and to collect samples. Measurements and Main Results: The cardiac index increased significantly with both dopamine and dobutamine (p < .001). Mean arterial pressure (MAP) increased, with the maximum effect of 20% seen with 12‐μg/kg/min dopamine infusion (p < .01). No change in MAP was seen with dobutamine. Dobutamine infusions did not change any renal variables. Conversely, all dopamine infusions significantly increased diuresis, creatinine clearance, and the fractional excretion of sodium (p < .01). Creatinine clearance increased from 61 ± 16.9 (SD) mL/min to a maximum of 85.7 ± 30 mL/min at the 7‐μg/kg/min dose; fractional excretion of sodium increased from 0.26% ± 0.28% to a maximum of 0.62% ± 0.51% at the 12‐μg/kg/min dose (p < .01). During dopamine infusions, there was a significant relationship between MAP and creatinine clearance (p = .018). Conclusions: At all doses studied, 4‐hr infusions of dopamine significantly increased creatinine clearance, diuresis, and the fractional excretion of sodium in stable critically ill patients. Conversely, dobutamine did not modify these variables. Although the level of MAP might partially contribute to the improvement in renal variables, it is more likely that the activation of renal dopamine receptors played a prominent role.

The increase in CO2 production induced by NaHCO3 depends on blood albumin and hemoglobin concentrations.

Objective: To evaluate the origin of H+ ions participating in the generation of CO2 coming from sodium bicarbonate infusion during metabolic acidosis. We hypothesized that these H+ ions come from a back-titration of the main non-bicarbonate buffers present in the blood, i. e. the hemoglobin and the albumin, and thus postulated that the rate of CO2 release from a bicarbonate load is dependent on the concentration of these buffers. Design: Prospective clinical and experimental study. Setting: Surgical intensive care unit of a university hospital Patientsand material: (1) Sixteen stable sedated and artificially ventilated critically ill patients with a mild base deficit. (2) Acidotic human blood (bicarbonate 5 mM, pH 7.0) of hematocrit 5, 10, 20 and 40 % regenerated from a mixture of frozen fresh plasma and packed red blood cells. Interventions: Patients: infusion of 1.5 mmol/kg sodium bicarbonate over 5 min. Regenerated blood: 25 mM sodium bicarbonate load. Measurements and results: Patients: continuous measurement of CO2 production (VCO2) on the expired gas using a metabolic monitor and arterial blood gas analysis before (T0), at the end (T5) and at 10, 30 and 60 min after the beginning of the bicarbonate infusion. The increase in VCO2 was 18 ± 7 % leading to a rise in PaCO2 from 39.6 ± 2.3 at T0 to 46.2 ± 2.7 mmHg at T5. The increases in VCO2 and in PaCO2 were significantly correlated to the albumin (r = 0.73, p < 0.005 and r = 0.70, p < 0.005, respectively) and to the hemoglobin (r = 0.51, p < 0.05 and r = 0.65, p < 0.01, respectively) concentrations. Regenerated blood: gas analysis 1 min after the bicarbonate load. The increase in PCO2 was closely related to the hematocrit (Ht) of the blood as it was 15.9 ± 7.5 mmHg for Ht 5 %, 29.0 ± 9.6 for Ht 10 %, 44.2 ± 5.9 for Ht 20 % and 71.0 ± 3.5 for Ht 40 % (n = 5 for each, p < 0.001). Conclusions: The importance of the release of CO2 from a bicarbonate load is dependent on the concentration of the blood non-bicarbonate buffers. It is therefore likely that the adverse effects of bicarbonate therapy linked to the CO2 generation are more important in patients with high blood albumin and hemoglobin concentrations.