Reversing oliguria in critically ill patients

Oliguria is a common occurrence in the ICU setting. In patients with preserved renal function, fluid challenges or low doses of diuretics are generally successful. In patients with oliguric renal failure, it is still essential to ensure adequate intravascular fluid volume, especially in critically ill patients. Loop diuretics remain the mainstay of treatment. When diuretic resistance is encountered, physicians should consider further optimization of hemodynamics, alternative loop diuretics, and combined drug therapy. In some cases, continuous renal replacement therapy can be very effective. Yet, while these interventions can help reduce the morbidity of severe volume overload, they have not been shown to improve mortality rates.     

Loop diuretics in the management of acute renal failure: a prospective, double-blind, placebo-controlled, randomized study

BACKGROUND: Studies on the role of loop diuretics in patients with acute renal failure (ARF) are largely retrospective, anecdotal, and poorly controlled. We report the results of a prospective, randomized, placebo-controlled, double-blind study examining the effect of loop diuretics on renal recovery, dialysis, and death in patients with ARF. METHODS: Ninety-two patients with ARF were enrolled into the study. All received intravenous dopamine, 2 micrograms/kg body weight/min throughout, 20% mannitol, 100 ml every 6 h for the first 3 days, and, in a double-blind manner, either torasemide, frusemide, or placebo, 3 mg/kg body weight i.v. every 6 h for 21 days or until renal recovery or death. RESULTS: Renal recovery, the need for dialysis, and death were no different in the three groups. Patients given a loop diuretic had a significant rise in urine flow rate in the first 24 h compared to placebo (P = 0.02). Based on the urine flow rate during the first post-medication day patients were divided into two groups--oliguric (< 50 ml/h) and non-oliguric (> or = 50 ml/h). Non-oliguric patients had a significantly lower mortality than oliguric patients (43% vs 69%, P = 0.01). However, they were less ill (APACHE II score 17.2 vs 20.6, P = 0.008) and had less severe renal failure at entry (creatinine clearance 14 ml/min vs 4 ml/min, P < 0.0001). CONCLUSION: The use of loop diuretics in oliguric patients with ARF can result in a diuresis. There is no evidence that these drugs can alter outcome.     

Torsemide: a new loop diuretic

The pharmacology, pharmacokinetics, clinical efficacy, adverse effects, and dosage and administration of torsemide are reviewed. Torsemide belongs to the pyridine-sulfonylurea class of loop diuretics. Its primary site of activity is the thick ascending limb of the loop of Henle, where it blocks active reabsorption of sodium and chloride, resulting in diuresis, natriuresis, and other effects. Torsemide has high bioavailability, a relatively long half-life, and a prolonged duration of activity. It is highly protein bound. Clinical trials indicate that torsemide is effective in the treatment of hypertension and of edema and other symptoms in patients with chronic renal failure (CRF), hepatic dysfunction, or congestive heart failure (CHF). Torsemide has infrequent, mild, and transient adverse effects; among the most common are orthostatic hypotension, fatigue, dizziness, and nervousness. The recommended initial oral dosages of torsemide are 10-20 mg/day for CHF, 20 mg/day for CRF, 5 mg/day for hypertension, and 5-10 mg/day (in combination with a potassium-sparing diuretic or aldosterone antagonist) for hepatic cirrhosis. In most patients, the pharmacokinetic advantages of torsemide over other loop diuretics are unlikely to translate into a substantial edge in clinical outcomes, and in practice there may be no cost advantages. Although torsemide does not offer major advantages over other loop diuretics, it may be of benefit in patients who do not respond to or cannot tolerate other agents.     

The molecular mechanism of temperature enhancement of proton magnetic relaxation rates in methaemoprotein solutions. III. The effect of sixth ligand in high-spin derivatives

Levels of cefazolin were determined in plasma, urine, bile, and cerebrospinal fluid in humans after a bolus intravenous injection and during a controlled, continuous intravenous infusion. All the patients were studied in a steady-state and crossover fashion. In plasma, the mean peak level after bolus injection (1.5 g) studied in 12 patients was 206.5 mug/ml; during continuous infusion (6 g daily), the mean level remained stable at 52.6 mug/ml. With bolus injection and continuous infusion, respectively, 89.7 and 86.3% of the administered dose of cefazolin were excreted in the urine of nine patients over the 6-h period considered. The levels of cefazolin in common bile duct bile were studied in six cholecystectomized patients. In bile collected during the two 3-h periods of the experiment, the mean concentration of the drug in the bile after bolus injection was 66.9 and 22.0 mug/ml, respectively; during continuous infusion, the corresponding biliary levels were 50.7 and 51.3 mug/ml, respectively. In four neurosurgical patients with an intraventricular catheter, neither bolus injection nor continuous infusion resulted in a demonstrable concentration of cefazolin in the cerebrospinal fluid. The continuous intravenous administration of cefazolin might have some advantage over the intravenous bolus intermittent injections. In plasma, the area under the curve is greater with continuous infusion than with bolus injection. In bile, the levels of cefazolin are more sustained with continuous infusion than with bolus injection. This approach to intravenous administration of cefazolin deserves more pharmacological and clinical trials.     

Toxicity management in patients receiving low-dose aldesleukin therapy

OBJECTIVE: To review the pathophysiology and subsequent treatment options for low-dose aldesleukin-induced toxicity when administered via intravenous bolus infusion, continuous intravenous infusion, or subcutaneous injection. BACKGROUND: The adverse events associated with high-dose aldesleukin therapy (600,000 IU per kg i.v. every 8 h for a maximum of 14 doses) are well documented in the literature; however, the adverse event profile of lower doses and alternative administration routes are less well described. An understanding of the adverse event profile associated with these alternative regimens can enhance management of toxicity. DATA SOURCES: English-language clinical studies, abstracts, and review articles pertaining to low-dose intravenous, continuous intravenous infusion, or subcutaneous injection of aldesleukin, as well as aldesleukin-induced adverse events. STUDY SELECTION AND DATA EXTRACTION: Relevant studies were selected that assist with understanding the pathophysiology, clinical management, diagnosis, and management of aldesleukin-induced adverse events. CONCLUSIONS: Aldesleukin therapy initiates a cytokine-mediated proinflammatory process resulting in a toxicity profile that is different from traditional nonbiologic chemotherapeutic agents. The frequency and severity of adverse events associated with aldesleukin administration are dependent upon dose, route, and administration schedule. In addition, most adverse reactions are self-limiting. Alleviation of aldesleukin-induced adverse effects can usually be achieved on an outpatient basis with agents such as antiemetics, antipyretics, and topical creams or lotions, as well as nonmedication interventions. Aggressive and proactive management of aldesleukin associated toxicities can help facilitate completion of therapy.     

The dobutamine-dopamine combination versus amrinone in congestive heart failure with a marked edematogenic sign complicated by functional kidney failure. A comparison between 2 different models of inotropic stimulation and diuresis potentiation

BACKGROUND: We evaluated the diuretic output in patients with decompensated chronic heart failure (CHF), previously treated by i.v. infusion with dobutamine and dopamine (dob-dop) or with amrinone (amr). Our target was to identify the possible discrepancies in urinary output perhaps linked to the different type of inotropic stimulation in the two subsets. METHODS: Adjunctive therapy with dob-dop or amr was chosen because the administration of diuretics only, without cardiac support, as tested in previous hospitalizations, had been demonstrated to produce unfavourable results, mainly expressed by finding of a low output syndrome in 50% of cases or more. The administration of i.v. infusion was maintained during 17 hours (1000 min approximatively), and included infusion in separate pumps of the two amines, dobutamine at dose of 5 micrograms/kg/min and dopamine at dose of 2.8 micrograms/kg/min or, alternatively, i.v. infusion of amr, administered at dose of 7 micrograms/kg/min. Infusion volumes were similar in the two subsets. The two subsets were homogeneous relatively to renal impairment, i.e. to the parameters (urinary Na, U/P creatinine, U/P urea, urinary osmolality) we fixed as markers idoneous to demonstrate the occurrence of organic renal damage (acute tubular necrosis). RESULTS: The diuresis was recovered in all 24 patients, and the urine volume resulted more pronounced in the subset attributed to the dob-dop at both the 8th and the 17th hour readings. We found no harmful alterations in HR and AP, whereas renal function parameters have been shown to enhance in both the dob-dop and amr arms. The diuretic effectiveness of the SIEV obtained by catecholamine implementation exercised a synergistic, favourable effect on diuresis, renal flow, glomerular filtration rate, and sodium post-proximal delivery. Amr resulted less effective then dob-dop simultaneous administration relatively to the diuretic effect. No remarkable differences were found in the two subsets as regards the heart rate, whereas a decrease in arterial pressure was found after amr. A persistent shift towards a condition of chronic renal failure, was identified in 4/24 patients, the two groups despite of the prolonged treatment at optimized doses: no remarkable side effects were reported. CONCLUSIONS: Thus, the selective effect upon renal hemodynamics, as exercised by dob-dop infusion low doses of dop, together with the enhanced renal output due to dob, has been shown to be more effective than amr influence: thus, the catecholamine therapeutical approach has been demonstrated to possess the best effectiveness in excitation of diuresis, among the CHF oliguric patients.     

Phenotype variability of the dominant beta-thalassemia induced in four Dutch families by the rare cd121 (G-->T) mutation

BACKGROUND: Initial clinical experience with recombinant factor VIIa (rVIIa) for treatment of haemophilia patients with inhibitors against factor VIII or IX has been obtained by administration of rVIIa by repeated intravenous bolus injections. However, continuous infusion of rVIIa may be a more appropriate administration method if prolonged treatment is indicated. METHODS: We have surveyed and analysed the initial experience with continuous infusion of rVIIa in the Netherlands and Belgium. RESULTS: Five hospitals treated 7 haemophilia patients with inhibitors on 9 different occasions (4 bleedings, 5 surgical interventions) by continuous infusion of rVIIa over a total of 59 days. Haemostatic coverage was considered effective in 8 out of 9 cases and partially effective in 1 case. Continuous infusion of rVIIa was aimed at rVIIa target plasma levels of 10 U/ml and a decrease in prothrombin time (PT) of 3 s compared to control levels. This was obtained by an initial bolus injection of 90 micrograms/kg prior to continuous infusion of rVIIa at doses between 30-6 micrograms/kg/h (mean 17.5 micrograms/kg/h). A conventional one-stage factor VII coagulation assay, often used in combination with a PT, was satisfactory in monitoring rVIIa treatment. The additional clinical value of anti-fibrinolytic and anti-thrombophlebitic treatment was unclear. CONCLUSION: In our experience, rVIIa appeared to be efficacious and safe when administered by continuous infusion. Continuous infusion of rVIIa is more convenient than bolus injections or rVIIa, easy to monitor and provides a cost reduction of > 50%. These advantages make continuous infusion an attractive administration method for prolonged treatment with rVIIa.     

Two-chain factor VIIa generated in the pericardium during surgery with cardiopulmonary bypass : relationship to increased thrombin generation and heparin concentration

The effect of changes in medullary extracellular tonicity on mRNA expression for aldose reductase (AR), sorbitol dehydrogenase (SDH), Na+/Cl-/betaine (BGT) and Na+/myo-inositol (SMIT) cotransporter in different kidney zones was studied using Northern blot analysis and non-radioactive in situ hybridization in four groups of rats: controls, acute diuresis (the loop diuretic furosemide was administered), chronic diuresis (5 days of diuresis), and antidiuresis [5 days of diuresis followed by 24 h deamino-Cys1, d-Arg8 vasopressin (dDAVP)]. Acute administration of the loop diuretic furosemide significantly reduced AR, SMIT and BGT gene expression in the inner and outer medulla compared with controls. Administration of dDAVP to chronically diuretic rats raised the expression of these three mRNAs in the inner but not the outer medulla compared with the chronically diuretic rats. None of these alterations in medullary tonicity significantly changed SDH expression. The in situ hybridization studies showed AR, BGT and SMIT mRNAs to be expressed in both epithelial and non-epithelial cells of the outer and inner medulla. The various cell types (epithelial, endothelial and interstitial cells) differed in their expression pattern and intensity of AR, SDH, BGT and SMIT mRNA, but the inner medullary cells responded uniformly to a decrease in extracellular tonicity with a reduction, and to an increase with enhancement of their AR, BGT and SMIT expression.     

Prediction of spinal cord ischemia with a retrievable stent graft on endovascular treatment for a case of thoracic aortic aneurysm

OBJECTIVE: To determine the pharmacokinetics and absolute bioavailability of ciprofloxacin in 12 critically ill patients receiving continuous enteral feeding. Design: a prospective, cross-over study. SETTING: 12-bed surgical intensive care unit in a University Hospital. PATIENTS: 12 stable critically ill patients on mechanical ventilation and receiving continuous enteral feeding (Normoreal fibres) without diarrhea or excessive residual gastric contents ( < 200 ml/4 h). None had gastro-intestinal disease, renal insufficiency (estimated creatinine clearance > or = 50 ml/min) or was receiving medications that could interfere with ciprofloxacin absorption or metabolism. MEASUREMENTS AND MAIN RESULTS: The study was carried out after the fourth (steady state) b. i. d. intravenous (i. v.) 1-h infusion of 400 mg and the second b. i. d. nasogastric (NG) dose of 750 mg (crushed tablet in suspension). Plasma concentrations were measured by high-performance liquid chromatography. The median (range) peak concentration after i. v. infusion was 4.1 (1.5-7.4) mg/l, and that after NG administration was 2.3 (0.7-5.8) mg/l, occurring 1.25 (0.75-3.33) h after dosing. The median [range] areas under plasma concentration-time curves were similar for the two administration routes (10.3 [3.3-34.6] and 8.4 [3.6-53.4] for i.v. infusion and NG administration, respectively). Ciprofloxacin bioavailability ranges from 31 to 82 % (median, 44%). CONCLUSIONS: In tube-fed critically ill patients, a switch to the NG ciprofloxacin after initial i. v. therapy to simplify the treatment of severe infections is restricted to those for whom serial assessments of ciprofloxacin levels are routinely available.     

Effect of administration and subsequent cessation of buserelin on cancellous bone of female rats

OBJECTIVE: Low-dose dopamine has been used in critically ill patients to minimize renal dysfunction without sufficient data to support its use. The aim of this study was to determine whether low-dose dopamine improves renal function, and whether dobutamine, a nondopaminergic inotrope, improves renal function. DESIGN: Prospective, randomized, double-blind trial. PATIENTS: Twenty-three patients at risk for renal dysfunction were entered into the study. Five patients were later withdrawn. Study data for the remaining 18 patients were: mean age 55 yrs; mean Acute Physiology and Chronic Health Evaluation (APACHE) II score of 18; mean weight 71 kg). The following conditions were present: mechanical ventilation (n = 17 [inverse-ratio ventilation, n = 6]); inotrope administration (n = 11); sepsis (n = 13); and adult respiratory distress syndrome or multiple organ failure syndrome (n = 9). INTERVENTIONS: The study patients were administered dopamine (200 micrograms/min), dobutamine (175 micrograms/min), and placebo (5% dextrose) over 5 hrs each in a randomized order. Ventilator settings, fluid management, and preexisting inotropic support were not altered during the study. MEASUREMENTS AND MAIN RESULTS: Systemic hemodynamic values and indices of renal function (4-hr urine volume, fractional excretion of sodium, and creatinine clearance) were measured during the last 4 hrs of each infusion. Dopamine produced a diuresis (145 +/- 148 mL/hr) compared with placebo (90 +/- 44 mL/hr; p < .01) without a change in creatinine clearance. Conversely, dobutamine caused a significant increase in creatinine clearance (97 +/- 54 mL/min) compared with placebo (79 +/- 38 mL/min; p < .01), without an increase in urine output. CONCLUSIONS: In stable critically ill patients, dopamine acted primarily as a diuretic and did not improve creatinine clearance. Dobutamine improved creatinine clearance without a significant change in urine output.     

Low dosage dopamine improves kidney function: current status of knowledge and evaluation of a controversial topic

The ability of low dose dopamine (1-3 micrograms x kg-1 x min-1) to cause selective renal vasodilation, to increase glomerular filtration rate, urine output, and natriuresis, is intuitively considered favourable. Dopamine, therefore, continues to be used in critically ill patients to preserve or improve renal function. Despite its application in a wide variety of disease states and in patients at risk of acute renal failure or with already decreased renal function, there is no conclusive evidence that "renal doses" of dopamine prevented acute renal failure or had any positive effect on patient outcome although it increased urine output and natriuresis consistently. Data from many clinical studies, however, are difficult to interpret due to small numbers of patients, the absence of control groups, the inability to exclude changes in cardiac output or to separate a diuretic effect of dopamine in the tubulus system from specific increases of glomerular filtration rate, and because of the variability of methods to determine renal performance (i.e. creatinine clearance, urine output, natriuresis, fractional excretion of sodium). Moreover, the routine use of dopamine is not innovous, since it may worsen gut ischaemia and suppress certain hormonal systems. Those who believe in the clinical benefits of "renal dose" dopamine argue that, even in the absence of an improvement in renal function, the maintenance of urine output could be useful in patients unresponsive to diuretics. Again, the clinical benefit of this diuretic action still needs to be shown. In conclusion, there is little justification for the routine administration of low-dose dopamine in patients at risk of renal failure. Large controlled clinical studies are urgently needed to determine whether dopamine improves renal function or prevents acute renal failure in patients at risk.     

Trace element and vitamin concentrations and losses in critically ill patients treated with continuous venovenous hemofiltration

OBJECTIVES: To measure the blood concentrations of a number of trace elements and vitamins in critically ill patients and examine their elimination by continuous venovenous hemofiltration (CVVH). SETTING: Intensive care unit of a tertiary institution. DESIGN: Prospective, controlled, clinical study. PATIENTS: Eight critically ill patients requiring renal replacement therapy, nine patients requiring intensive care treatment but not requiring renal replacement therapy, and nine healthy controls. INTERVENTIONS: Measurement of trace element and vitamin concentrations in blood and ultrafiltrate. MEASUREMENTS AND MAIN RESULTS: Compared with normal volunteers, critically ill patients requiring CVVH had significantly lower median blood concentrations of vitamin C, vitamin E, selenium, and zinc. During the first 24 hrs of CVVH, there were no changes in the trace element and vitamin concentrations in blood, nor were there differences between pre- and postfilter samples. Micronutrient losses in the ultrafiltrate were small or undetectable except for Vitamin C, chromium, and copper. Compared with normal volunteers, critically ill patients not requiring CVVH also had significantly lower median blood concentrations of vitamin C, vitamin E, selenium, and zinc. There were no differences between the two critically ill groups. CONCLUSIONS: The clinical significance of the reductions in blood concentrations of selenium, zinc, vitamin C, and vitamin E in critically ill patients and the ultrafiltrate losses of Vitamin C, copper, and chromium remains unclear.     

Failing diuretics in severe chronic heart failure

Three patients with chronic heart failure, men aged 29, 78 and 69 years, developed severe dyspnoea and oedema in spite of reduced sodium and fluid intake and medication including furosemide. Heart failure may become 'resistant to diuretics' due to pharmacokinetic and pharmacodynamic causes. High-dose continuous intravenous administration of a loop diuretic may afford relief in such cases, if necessary in combination with a thiazide derivative, an ACE inhibitor, an inotropic agent or an extracorporal technique. Monitoring and correction of the state of hydration of a patient with chronic heart failure may improve the prognosis and the quality of life.     

The role of H2-receptor antagonists in the pathogenesis of nosocomial pneumonia in mechanically ventilated patients

We conducted a computerized MEDLINE search and selected controlled studies and meta-analyses correlating the frequency of nosocomial pneumonia (NP) with the administration of histamine2-receptor antagonists (H2RAs) as stress ulcer prophylaxis in critically ill patients. Although such a correlation does exist, the literature supports the theory that gastric bacterial overcolonization through alkalinization by H2RAs is not a risk factor for NP in critically ill patients. Further well-designed studies are necessary to resolve the issue and clarify the role of H2RAs in the pathogenesis of NP.     

Predischarge screening of very low birthweight infants by click evoked otoacoustic emissions

Idiopathic edema patients abusing diuretics are occasionally becoming dependent to such a degree on increasing doses of diuretics that their withdrawal results in severe cardiorespiratory failure, occasionally even pulmonary edema. Two such patients are described and 1 is investigated in depth as to the mechanism of the diuretic abuse-induced excessive tubular avidity for sodium. An extreme diuretic-induced secondary hyperaldosteronism and atrial natriuretic factor suppression, although tapering off when diuretics are stopped, results in a continuous tubular sodium hyper-reabsorption. Since the most affected patient was deprived of the benefits of converting enzyme inhibitors because of their side effects, the only way to partially overcome this condition was a generous combination of several diuretics acting at several segments of the nephron. This contrasted with a similar patient who was relatively well controlled by a converting enzyme inhibitor combined with lower dose diuretics. Diuretic abuse-induced secondary hyperaldosteronism and diuretic resistance are apparently best prevented by converting enzyme inhibitors. When nonpharmacological preventive measures fail, converting enzyme inhibitors are preferable to diuretics as the first-choice treatment of idiopathic edema patients.     

Continuous cardiac output and mixed venous oxygen saturation monitoring

Continuous assessment of cardiac output and SVO2 in the critically ill may be helpful in both the monitoring variations in the patient's cardiovascular state and in determining the efficacy of therapy. Commercially available continuous cardiac output (CCO) monitoring systems are based on the pulsed warm thermodilution technique. In vitro validation studies have demonstrated that this method provides higher accuracy and greater resistance to thermal noise than standard bolus thermodilution techniques. Numerous clinical studies comparing bolus with continuous thermodilution techniques have shown this technique similarly accurate to track each other and to have negligible bias between them. The comparison between continuous thermal and other cardiac output methods also demonstrates good precision of the continuous thermal technique. Accuracy of continuous oximetry monitoring using reflectance oximetry via fiberoptics has been assessed both in vitro and in vivo. Most of the studies testing agreement between continuous SVO2 measurements and pulmonary arterial blood samples measured by standard oximetry have shown good correlation. Continuous SVO2 monitoring is often used in the management of critically ill patients. The most recently designed pulmonary artery catheters are now able to simultaneously measure either SVO2 and CCO or SVO2 and right ventricular ejection fraction. This ability to view simultaneous trends of SVO2 and right ventricular performance parameters will probably allow the clinician to graphically see the impact of volume loading or inotropic therapy over time, as well as the influence of multiple factors, including right ventricular dysfunction, on SVO2. However, the cost-effectiveness of new pulmonary artery catheters application remains still questionable because no established utility or therapeutic guidelines are available.     

Disturbance of continuous, pump administration of cardiovascular drugs by hydrostatic pressure

Administration of highly concentrated, highly potent, and therefore highly dangerous drugs with syringe pumps is common in modern anaesthesia as well as in intensive care and emergency medicine. Because of their exact flow rates down to < 1 ml/h, these pumps are predestined for delivery of drugs with short half-lives, such as catecholamines and vasodilators. But intravenous application of drugs with syringe pumps is not without problems. While it is well known that syringes not fixed correctly into the pump can empty themselves by the influence of gravity, it seems not to be known that hydrostatic pressure can influence the flow rate of a correctly connected system even during continuous infusion. In this situation a change of height of the syringe pump in relation to the patient's position can have tremendous effects on hemodynamics due to unintended acceleration or deceleration of the flow rate. This case report demonstrates that the elevation of a connected epinephrine pump while moving a cardiac surgery patient after ACB operation from the operation table into his bed led to critical increases of heart rate, blood pressure and left atrial pressure. In order to quantify the problem we repeated the situation experimentally. It could be demonstrated that the elevation of the syringe pump by 80-100 cm delivers an additional bolus of 4-5 drops as the central venous catheter outlet. Lowering the pump consecutively leads to the opposite effect. In the case reported, the accidentally administered bolus of epinephrine was 12-15 micrograms (we use a concentration of 60 micrograms/ml epinephrine for continuous infusion with syringe pumps). From this accidental observation the following conclusion can be drawn: The change of height, in relation to the patient's position, of a running syringe pump during continuous infusion of highly concentrated cardiovascular drugs may cause considerable, even life-threatening hemodynamic disorders. Even in a closed infusion system (syringe-extension-central venous catheter), hydrostatic pressure influences infusion rate. Elevation of the pump leads to unintended bolus administration, and lowering of the pump is followed by an interruption of the infusion. In the knowledge of this phenomenon, unexpected hemodynamic reactions during transport of critically ill patients cannot always be interpreted as a result of inadequate anesthesia or volume load, but may be a consequence of incorrect handling of the syringe pumps as described in this report.     

MRI and articular cartilage

There are considerable laboratory data and information from animal and continuous culture in vitro models to support continuous infusion therapy for penicillins and cephalosporins, but, as yet, the only existing clinical data relate to cephalosporins. Penicillins do not exert concentration-dependent killing in the therapeutic range but have a post-antibiotic effect (PAE) against Gram-positive cocci but not Gram-negative rods. Animal models indicate the time (T) during which the serum concentrations exceed the minimum inhibitory concentration (MIC) of the pathogen [T > MIC] determines outcomes. Pharmacokinetic studies in humans indicate that continuous infusion with penicillins is possible but there are no clinical data on efficacy. Cephalosporins have similar pharmacodynamic properties to penicillins; T > MIC determines outcome. Data related to ceftazidime indicate that the drug concentration at steady-state (Css) should exceed the pathogen MIC by > 1-fold and perhaps by 4- to 5-fold or more. Human pharmacokinetics of ceftazidime administered by continuous infusion to a wide variety of patient groups indicates that Css of > 20 mg/L can easily be achieved using conventional daily doses. Clinical data indicate increased effectiveness of a continuous regimen in neutropenic patients with Gram-negative infection. Furthermore cefuroxime administration by continuous infusion has resulted in lower doses and shorter course durations. Little is known of the pharmacodynamics of monobactams and there are few clinical data on continuous infusion therapy. Carbapenems have different pharmacodynamics to other beta-lactams as they have concentration-dependent killing and a PAE with both Gram-positive and Gram-negative bacteria. While T > MIC has a role in determining outcomes, the proportion of the dosing interval for which serum drug concentrations should exceed the pathogen MIC is less than for other beta-lactams. In vitro models have shown that continuous infusion is effective, as is less frequent dosing. There are few data on continuous infusion of carbapenems but some patients have been treated with once-daily dosing. Clinically, continuous infusion therapy with penicillins and cephalosporins should be considered in patients infected with susceptible Gram-negative rods not responding to conventional therapy. As an approximation, the same total daily dose should be given but a bolus intravenous injection should be give at the start of continuous infusion to ensure Css is reached rapidly. The Css may be difficult to predict and determination of serum drug concentrations may be indicated. Ideally, the Css should be calculated based on the MIC of the potential pathogen and may be higher or lower than the Css achieved by a conventional daily dose.