[Treatment reduction in intensive care. "Allowing the patient to die" by conscious withdrawal of medical procedures]

The conversion of an "attempt to treat" to "prolongation of dying" represents an important problem in modern intensive care. If the actual or presumed will of the patient is unknown, the physician has to decide about the extent of treatment in a paternalistic manner. In these difficult decisions the physician has to consider prognosis, and certainty of prognosis and has to carefully balance between the right to live and the right to die. This decision about the extent of therapy is a very personal medical activity and can be taken off the physician's shoulders by nobody. Consultation with other physicians involved, relatives, nurses and clergy, however, is mandatory, as a joint decision should be sought. If the situation is hopeless and further medical interventions are futile, then allowing the patient to die by therapy reductions is not only a possibility but a mandatory act of humanity. In that case it does not matter whether new treatment modalities are abandoned or whether already instituted medical measures are withdrawn. In clinical practice, however, the "fine tuning" of therapy reduction has to be tailored to the individual case and largely depends on prognostic certainty.

Repeated administration of recombinant human insulin-like growth factor-I in patients after gastric surgery. Effect on metabolic and hormonal patterns

OBJECTIVE

The primary objective of this investigation was to evaluate the anticatabolic effects of repeated subcutaneous administration of recombinant human insulin-like growth factor-I (rhlGF-I) in patients after gastric surgery.

SUMMARY BACKGROUND DATA

The anabolic and protein-sparing effects of growth hormone are primarily mediated by IGF-I. Malnutrition and catabolic states result in increasing blood levels of growth hormone and decreasing levels of IGF-I. Experimental data showed that exogenous IGF-I could attenuate or reverse catabolism.

METHODS

After giving their written informed consent, 38 male and female patients undergoing gastrectomy (age 40-75 years, body mass index 17-30 kg/m2) were treated with 80 micrograms/kg body weight rhlGF-I or placebo in a prospective, randomized, double-blind study for 5 consecutive days. Patients received a standardized total parenteral nutritional regimen with 3 g/kg body weight glucose and 0.1 g/kg body weight nitrogen. Nitrogen balance and 3-methylhistidine excretion were measured daily. Hormone profiles (IGF-I, IGFBP1, IGFBP3, cortisol, insulin, glucagon, triiodothyronine [T3], levothyroxine [T4], and thyroxine-binding globulin) were taken.

[Glucose-xylitol 35% (1:1) versus glucose 40%. Effectiveness and metabolic effects after major surgery]

Injury and stress are accompanied by a characteristic hormonal response and altered energy utilisation. Hyperglycaemia and negative nitrogen (N) balance are the leading symptoms of the metabolic changes in the post-operative state. In a prospective, randomised study the efficacy and metabolic effects of glucose-xylitol (GX) 35% (1:1) versus glucose (G) 40% were investigated in patients undergoing major surgery. METHOD. Twenty-four patients undergoing abdomino-thoracic oesophageal cancer surgery were treated in a standardised manner. Total parenteral nutrition was administered over 6 days (kg body wt.-1/day): day of surgery 1-1.25 g carbohydrate (CH); 1st postoperative day (POD) 1.5 g CH, 1 g amino acids (AA); 2nd POD 3 g CH, 1.5 g AA, 1.0 g fat; from 3rd POD 3 g CH, 1.5 g AA, 1.5 g fat (CH GX35% (n = 12) or G40% (n = 12), AA Intrafusin 15%, fat Intralipid 20%). Daily and cumulative N balances, blood-G profiles, blood chemistry, and physical parameters were determined. Glucagon and insulin profiles, CH losses, and oxalic acid secretion were measured. RESULTS. Both groups were comparable for age, body mass index, clinical and physical parameters, and blood chemistry. Mean cumulative N balances after 6 days were -12.0 +/- 16.3 g N for GX35% and -5.6 +/- 19.4 g N for G40% (n.s.; Wilcoxon, P < 0.05). Blood G was similar for both groups with values ranging from 130 to 240 mg/dl on the day of surgery and below 150 mg/dl on the consecutive days. In each group 1 patient needed additional insulin therapy. Glucagon and insulin levels did not show a significant difference between the groups. CONCLUSION. No difference in tolerance and efficacy of nutritional support by GX versus G at a dose of 3 g.kg body wt.-1.d in oesophagectomised patients could be observed. Similar blood G profiles were in accordance with comparable glucagon and insulin levels. Because of the high standard deviations of N balances, differences in efficacy could not be proven. A significantly lower level of pseudocholinesterase (PCHE) for G40% on day 7 might indicate enhanced hepatic protein synthesis in the GX group.

[Safety and tolerance of a new dipeptide-containing amino acid solution DP-Gln 20 versus a conventional amino acid solution Vamin 18EF in patients after elective colon surgery]

OBJECTIVE AND METHODS

Recent studies have demonstrated the important role of glutamine in postoperative metabolism. Using dipeptide-containing amino acid solutions it is possible to enhance glutamine supply. Safety and tolerance of DP-Gln 20 (gly-gln, gly-tyr) and Vamin 18EF were investigated in an open, prospective, randomised study. 16 patients received isonitrogenous parenteral nutrition over 4 days.

RESULTS

There was no difference in clinical and biochemical patterns between both groups. Complications and adverse events due to the infused amino acid solutions were not observed.

[Autologous blood donation and isovolemic hemodilution--indications and practical implementation]

OBJECTIVE

The state of the art of autologous blood transfusion is described with special emphasis on safety aspects, indications and medicolegal implications.

DATA SOURCES AND SELECTION CRITERIA

Literature was retrieved using the MEDLINE literature database. Medical and legal expert opinions on autologous blood transfusion programmes are presented as well as the actual German jurisdiction. Guidelines for autologous predeposit and haemodilution used in the University of Münster are described.

RESULTS

In the past decade all forms of autologous transfusions gained increasing influence in haemotherapy due to the ongoing discussion on the safety of blood products. The German Federal Court has demanded that whenever homologous perioperative transfusion is considered likely, patients have to be offered autologous predeposit. Legal conditions for autologous programmes directed by anaesthetists not specialised in transfusion medicine are described. Whole-blood predeposit should be limited to two autologous units. In cases with minor blood loss, isovolaemic haemodilution may be performed instead of autologous predeposit. However, autologous transfusions have their specific risks that are either related to the patient or to the procedure of autologous predeposit, e.g., clerical error, contamination of blood products and technical faults. Standard procedures of the University of Münster to ensure low-risk autologous transfusion are presented. They consist in adequate handling and proper identification, testing of donor for virus infection markers, bacterial culture from blood products and a list of contraindications: anaemia, unstable angina, myocardial infarction within 3 months, decompensated heart insufficiency, aortic valve stenosis with angina, and cases with infection and fever.

CONCLUSION

The risks related with autologous transfusion should be lower compared to homologous transfusions. Well-defined standards concerning indications and techniques are required to reach this goal.

Continuous spinal anesthesia: mechanical and technical problems of catheter placement

BACKGROUND AND OBJECTIVE

Although continuous spinal anesthesia with microcatheters has a number of advantages, there are also some drawbacks: technical problems in advancing the catheter, the possibility of traumatizing neural structures, the development of cauda equina syndrome, and maldistribution of the local anesthetic.

METHODS

Spinaloscopy was performed with a 2-mm-diameter endoscope in fresh cadavers to visualize the fate of the catheters, as well as the distribution of the local anesthetic administered through these fine-bore catheters. Midline and paramedian approach achieved an easy insertion of the 28-gauge catheter as long as the 22-gauge needle was not advanced too far into the subarachnoid space, thereby making it impossible for the catheter to bend at the anterior wall of the dura mater.

RESULTS

Injection of methylene blue-colored hyperbaric local anesthetic through the catheter revealed an inhomogenous distribution with pooling in the caudal segments. After the catheter tip leaves the needle, the catheter should be advanced only 2-3 cm to avoid coiling, possible damage of the nerve roots, or malpositioning in preformed pouches.

CONCLUSION

To take advantage of continuous spinal anesthesia, a meticulous technique is required.

Mechanical ventilation in the prone position for acute respiratory failure after cardiac surgery

Ten patients with acute respiratory failure (ARF) after coronary artery bypass grafting were studied during conventional mechanical ventilation in the supine and in the prone position. Impaired gas exchange was defined as an inspired oxygen fraction (FIO2) greater than 0.5 to maintain an arterial oxygen tension (PaO2) > or = 70 mmHg, an alveolar-arterial PaO2 gradient (PA-aO2) > 200 mmHg and a venous admixture (QVA/QT) > 15% during mechanical ventilation with a tidal volume (VT) = 10 to 12 mL/kg, frequency (f) = 10 to 15 VT/min, inspiratory-expiratory (I:E) ratio = 0.5, and positive end-expiratory pressure (PEEP) of 5 to 7.5 cm H2O. In the supine position, systemic and pulmonary hemodynamics were in the normal range, but oxygenation was severely impaired. In all patients thoracic computed tomography scans were obtained and revealed crest-shaped bilateral densities in the dependent lung regions. FIO2 of 0.67 +/- 0.22 was required to maintain a PaO2 greater than 70 mmHg during mechanical ventilation in the supine position. Under these conditions PA-aO2 and QVA/QT were 362 +/- 153 mmHg and 32.5 +/- 8.3%, respectively. CO2 elimination was not severely affected. The patients were turned into the prone position after an average of 30.6 +/- 5.4 hours postoperatively and ventilated with unchanged VT, f, PEEP, and inspiratory-expiratory ratio for 26.7 +/- 11.7 hours (range, 10 to 42 hours). A second cardiopulmonary status was obtained within 2 to 5 hours of ventilation in the prone position.(ABSTRACT TRUNCATED AT 250 WORDS)

Effective pulmonary capillary pressure in experimental myocardial ischaemia

Effective pulmonary capillary pressure and extravascular lung water were investigated in dogs (n = 9) with normal heart function and after development of acute myocardial ischaemia. During control, no impairment of cardiopulmonary performance was observed. Extravascular lung water was in the normal range (8.1 +/- 2.8 ml.kg-1) and the effective pulmonary capillary pressure accounted for 1.36 +/- 0.53 kPa (10.2 +/- 4 mmHg). No correlation between extravascular lung water and effective pulmonary capillary pressure was observed (r2 = 0.347, P = 0.06). Arterial (RPA) and venous pulmonary resistance (RPV) were 70 +/- 15% and 30 +/- 6%, respectively. Acute myocardial ischaemia was induced by one stage occlusion of the left anterior descending (LAD) coronary artery; measurements during the ischaemia phase were performed 60 min following LAD occlusion. Myocardial ischaemia resulted in moderate changes of cardiac output, heart rate and left ventricular end-diastolic pressure. Oxygenation deteriorated, but no hypoxaemia occurred in any animal and CO2 elimination remained unchanged. Extravascular lung water was elevated (16.5 +/- 7.9 ml.kg-1, P < or = 0.01), and effective pulmonary capillary pressure was higher when compared with the control state (2.32 +/- 1.05 kPa (17.4 +/- 7.9 mmHg), P < or = 0.01). There was a significant correlation between both parameters (r2 = 0.528, P < or = 0.05). Longitudinal distribution of pulmonary vascular resistance was altered, and RPA decreased to 60 +/- 13% (P < or = 0.05), while RPV increased to 40 +/- 8% (P < or = 0.05). It is concluded that development of lung oedema is related to elevated effective pulmonary capillary pressure in dogs with acute myocardial ischaemia.(ABSTRACT TRUNCATED AT 250 WORDS)

[Cardiopulmonary effects of CPPV (continuous positive pressure ventilation) and IRV (inverse ratio ventilation) in experimental myocardial ischemia]

Continuous positive pressure ventilation (CPPV) is an established therapy for treatment of acute respiratory failure (ARF). However, cardiac performance may be severely disturbed due to elevated intrathoracic pressure, inducing a decrease in cardiac output (CO) and oxygen delivery (DO2). Alternatively, mechanical ventilation with prolonged inspiratory to expiratory duration ratio (inversed ratio ventilation IRV) has been successfully used in ARF. No data are available about IRV in acute haemodynamic oedema. Thus, the cardiopulmonary effects of CPPV (positive end-expiratory pressure [PEEP] = 10 cm H2O) and IRV (inspiration to expiration duration ratio [I:E] = 3.0) were studied in nine dogs (body weight 29.9 +/- 4.3 kg) before and after induction of myocardial ischaemia. METHODS. Continuous intravenous anaesthesia and muscle paralysis were provided by 1.2 mg.kg-1 x h-1 piritramide and 0.08 mg.kg-1 x h-1 pancuronium, and the animals were ventilated with intermittent positive pressure ventilation (IPPV) as reference method. Cardiocirculatory performance was determined by means of heart rate (HR), mean arterial pressure (MAP), mean pulmonary arterial pressure (MPAP), central venous pressure (CVP), pulmonary artery occlusion pressure (PAOP) and left ventricular end-diastolic pressure (LVEDP). Cardiac output (CO) was determined by thermodilution method. Systemic vascular resistance (SVR) was calculated. Pulmonary function was assessed by arterial and mixed venous blood gas tension for oxygen (PaO2, PvO2) and carbon dioxide (PaCO2). Functional residual lung capacity (FRC) was measured by means of the foreign gas wash-in method using helium as inert gas, and determination of extravascular lung water (EVLW) using the thermal-dye indicator technique. CPPV and IRV were studied in random sequence in the control phase and 60 min after induction of acute left ventricular ischaemia, which was achieved by occlusion of the ramus interventricularis anterior. RESULTS. During the control phase CPPV induced an increase in MPAP (P < 0.05), CVP (P < 0.05) and PAOP (P < 0.05). HR and MAP remained unchanged, whereas CO decreased by 16% (P < 0.05). FRC was elevated by 25 ml.kg-1 (P < 0.01), but not EVLW (9.1 +/- 3.5 ml.kg-1). There was no improvement in oxygenation; instead, oxygen delivery (DO2) decreased (P < 0.05). During inversed ratio ventilation MPAP, CVP, PAOP increased, but less than during CPPV. FRC was elevated mu 7.0 ml.kg-1 (P < 0.05), which was significantly less than during CPPV (P < 0.05). EVLW revealed no differences. During IPPV in the ischaemia phase cardiopulmonary performance deteriorated significantly. CO decreased by 19% (P < 0.05), whereas HR, MPAP, CVP and PAOP increased (P < 0.05). PaO2 was lower (P < 0.05) and alveolo-arterial PO2 gradient (PAaO2) increased (P < 0.05). All animals revealed moderate pulmonary oedema (EVLW = 15.1 +/- 8.4 ml.kg-1) (P < 0.01) and a lower FRC. Mechanical ventilation with PEEP significantly improved oxygenation and FRC; however, DO2 was slightly lower than during IPPV (not significant). IRV elevated PaO2, FRC and DO2, since CO was not depressed when compared with IPPV. CONCLUSIONS. CPPV and IRV may induce a recruitment of collapsed or hypoventilated lung areas, which is more pronounced during CPPV. During both modes of ventilation, oxygenation was improved without apparent changes in EVLW. Haemodynamic performance was more impaired during CPPV, and no improvement of left ventricular function secondary to an elevated intrathoracic pressure was observed. Occlusion of the RIVA coronary artery typically induces an infarction of 35% of left ventricular muscle mass; however, non-ischaemic myocardium reveals an unchanged or increased contractility. Thus, a reduction of left ventricular preload secondary to CPPV mainly contributes to haemodynamic depression, which is less pronounced during IRV due to a lower peak inspiratory airway pressure and mean airway pressure. IRV may be useful for mechanical ventCntCo

Eigenblutspende und isovolämische Hämodilution – Indikationen und praktische Durchführung

<i>Ziel:</i> Derzeit gültige Standards für die autologe Bluttransfusion werden beschrieben mit rechtlichen Grundlagen, Indikationen und Hinweisen zur sicheren Anwendung. <i>Quellen und Auswahlkriterien:</i> Die Literaturauswahl erfolgte mit der MEDLINE-Datenbank. Aktuelle Stellungnahmen aus medizinischer und juristischer Sicht sowie die Rechtsprechung zum Thema wurden berücksichtigt. Richtlinien für die Durchführung von Eigenblutspende und isovolämischer Hämodilution, wie sie für die Westfälische Wilhelms-Universität gültig sind, werden beschrieben. <i>Ergebnisse:</i> Die autologe Hämotherapie hat im Gefolge der Diskussionen um die Virussicherheit der homologen Transfusion an Bedeutung gewonnen. Rechtliche Vorgaben für die Durchführung von Eigenblutspendeprogrammen werden beschrieben. Die autologe Vollblutspende sollte sich auf zwei Einheiten beschränken. Bei geringem zu erwartenden Blutverlust kann eine isovolämische Hämodilution die Eigenblutspende ersetzen. Auch Eigenblutverfahren sind nicht frei von spezifischen Risiken. Sie bestehen aus patientenbezogenen Risiken im Zusammenhang mit der Grundkrankheit sowie aus verfahrensbezogenen Komplikationen wie Verwechslung, Kontamination und mangelhafter Durchführung. Standards für Eigenblutspende und Hämodilution müssen sich auf Identifikationsmerkmale, Ausschluß von Virusinfektion, Sicherung von Sterilität und definierte Ausschlußkriterien beziehen. <i>Schlußfolgerung:</i> Autologe Transfusionsverfahren sind nur dann sinnvoll, wenn im Vergleich zur homologen Transfusion eine tatsächliche Risikominderung eintritt. Mit einer Standardisierung von Indikation und Methode der Eigenblutspende und der isovolämischen Hämodilution ist dieses Ziel zu erreichen.

[Subdural intra-arachnoid spread of local anesthetics. A complication of spinal anesthesia]

Accidental subdural injections and catheterisations are a complication of epidural and spinal anaesthesia. The incidence of subdural spread in myelographies is estimated to be over 10% by the spinal technique. With spinaloscopy in an anatomic human model, we analysed the puncture process and the influence of different needle types on the incidence of subdural injection. We compared 22-gauge Sprotte, Quincke, and 18-gauge Tuohy needles in median and paramedian approaches with various bevel orientations. METHOD. The studies were performed in a preserved and recently expired cadaver donated to the Institut für Anatomie, Westfälische Wilhelms-Universität, Münster. The spinal column from T12 to S1, together with the back musculature (in order to preserve the normal curvature of the spine), were removed from the cadaver. Spinaloscopy was performed with a 4-mm endoscope with a 0 degree optic (Storz, Tuttlingen, Germany). All observations were made in the lumbosacral region of the dissected preparation. The endoscope was inserted from the caudal end of the spinal canal and, depending on the observations being made, the spinal canal was filled with air or artificial cerebrospinal fluid (CSF). To obtain information on the distribution of local anaesthetics injected into the subarachnoid space, 0.5% bupivacaine was coloured with a small amount of 1% methylene blue. The distribution of the coloured anaesthetic was clearly visible during and after injection. RESULTS. Needle insertion: Multiple observations were made using median or paramedian advancement of the needle into the spinal canal. With all needles, including the pencil-point, we saw an unexpected inward movement of the dura to the epidural space before penetration. This dural movement was independent of the direction of the dural fibres in the lumbar area. Distribution of local anaesthetics: Our observations indicate that difficulty with injecting drugs occurred when needle insertion was stopped too close to the dura, especially with the Sprotte needle. After manually registered penetration of the dura, the lateral opening of the needle only partially penetrates the dura. This allows CSF to appear in the needle hub, and injection into the vertical subdural space is possible. In all cases with the Sprotte needle, we could reproduce deposition of methylene-blue-coloured local anaesthetics into the subdural space. With the Quincke and Thuohy needles, it was not possible to deposit local anaesthetics into the subdural space in this model. CONCLUSION. Spinaloscopy was done in a non-fixated anatomic preparation of a spinal column with a 4-mm, 0 degree endoscope. From these observations we conclude that both manually registered penetration of the dural and the appearance of CSF in the needle hub can mimic correct needle position. Especially with the lateral opening of the Sprotte needle, deposition of local anaesthetics in the subdural space is possible.

[Spinal endoscopy in the detection of problems caused by continuous spinal anesthesia]

Continuous spinal anaesthesia has a number of advantages, but there are a number of drawbacks as well: difficulties in threading the catheter, distribution of the local anaesthetics and the development of cauda equina syndrome. Spinaloscopy was done to visualize the fate of catheters during and after their insertion, as well as the distribution of local anaesthetics injected through these fine-bore catheters. METHOD. The studies were conducted in preserved and fresh cadavers donated to the Anatomic Institute for Medical Studies. The spinal column from T12 to S1, together with the back musculature (in order to preserve the normal curvature of the spine) were removed from the cadaver. Spinaloscopy was done with a 4 mm endoscope with a 0 degree optic (Storz, Tutlingen, FRG). All observations were made from the lumbosacral region of the dissected preparation. In this fashion, it was possible to observe the insertion of the spinal needle used to introduce the catheter into the subdural space. The distribution of local anaesthetics injected through a 22-gauge spinal needle or a 28-gauge catheter was shown by injecting 0.5% hyperbaric bupivacaine colored with a small amount of 1% methylene blue. Pictures were taken 15, 30 and 45 s after beginning the injection. RESULTS. Difficulty in threading the catheter: our observations indicate that the difficulty in inserting microbore catheters is most likely due to inserting the needle too far. It is impossible for the catheter to bend and be inserted into the subarachnoid space. In many cases the catheter encountered the anterior wall of the spinal canal and would slide along various structures. Distribution of the drug: the injection is better dispersed with a 22-gauge needle and it completely fills the subarachnoid space. The local anaesthetics injected through the 28-gauge nylon catheter (Kendall Healthcare, Mansfield, Mass.) are distributed in the dependent portions of the spinal canal. If high doses and a high concentration are injected, the distribution pattern may result in an overconcentration in some parts of the subarachnoid space. Possibility of trauma: the catheter stretches around the roots, the potential for trauma is that untoward stress may be applied to the root, either during full insertion of the catheter or during its withdrawal. CONCLUSION. Spinaloscopy was done in a non-fixated anatomic spinal column preparation with a 4 mm 0 degree endoscope (Storz, Tuttlingen, FRG). Based on our observations, we conclude: The catheter should only be inserted 2 cm into the subarachnoid space. This may decrease the risk of malpositioning. After the tip of the catheter has reached the subarachnoid space, the stylet should be with drawn 2 or 3 cm to minimize the risk of nerve injury and/or bleeding.

[Tubal malpositioning as a cause of tubal cuff hernia]

Airway obstruction of endotracheal tubes may occur during general anaesthesia. A case is reported where endobronchial intubation and inflation of the cuff caused herniation into the left bronchus before the malpositioning had been corrected. This led to a cuff herniation that caused airway obstruction at the end of anaesthesia. The regional overinflation of the tubes cuff could be reproduced by a model by other tubes of the same type. Malpositioning of endotracheal tubes may cause cuff herniation that potentially leads to acute airway obstruction.

Effects of enflurane on inducibility of ventricular tachycardia

The effects of enflurane on cardiac electrophysiologic parameters and on inducibility of ventricular tachycardia (VT) by programmed stimulation were studied in 12 patients (11 men, 1 woman, mean age +/- standard deviation 55 +/- 8 years) with drug refractory sustained monomorphic VT who underwent transcatheter ablation with high-energy direct-current shocks. One catheter ablation procedure was performed in 10 patients, whereas 2 ablation sessions were necessary in 2 patients. Programmed ventricular stimulation was performed on 2 separate days (mean interval 19). There were 2 baseline studies, 1 several days before ("baseline study I") and the second at the beginning of the ablation procedure ("baseline study II") while the patient was awake and nonsedated. The third programmed stimulation study was done 15 to 30 minutes after administration of anesthesia with enflurane, oxygen and nitrous oxide ("enflurane study"). Rate of sinus rhythm, QRS duration, PQ interval and ventricular effective refractory period were unaltered, whereas QTc interval increased significantly after initiation of anesthesia. Before and after induction of general anesthesia, clinical VT was inducible in all patients. However, in 1 patient, induction of VT was only possible by pacing in the left ventricle after enflurane administration. Based on these data, it is concluded that general anesthesia with enflurane, oxygen and nitrous oxide has no marked influence on inducibility of clinical VTs. Therefore, this type of anesthesia may be useful for nonpharmacologic, ablative procedures requiring general anesthesia.

Development of intensive care medicine in Germany

As of 1991, intensive care medicine in Germany is not an independent medical specialty but a part of other main medical specialities such as anesthesiology, internal medicine, surgery, and pediatrics. Accordingly, there is neither formal training nor a separate board examination in intensive care medicine. As in other countries, intensive care units (ICUs) were established during the 1950s and 1960s, triggered by the positive experience with consolidation of polio victims in special respiratory care units. Surgical (or operative) ICUs predominantly are operated by anesthesiology departments, as anesthesiologists' expertise in respiratory and hemodynamic support qualifies them for the management of the critically ill patient in the perioperative phase. This article gives a brief review of the development of intensive care medicine in Germany, thereby providing the historical background for its present national and regional organization, facilities, and education and training programs.

Effect of selective flora suppression on colonization, infection, and mortality in critically ill patients: a one-year, prospective consecutive study

OBJECTIVE

To study the effect of enterally administered polymyxin E, tobramycin, and amphotericin B (selective flora suppression) on bacterial colonization, infection, resistance, and mortality rate.

DESIGN

Prospective, consecutive crossover controlled study.

SETTING

Two surgical ICUs in a university hospital; ICU I with ten beds, ICU II with eight beds.

PATIENTS

Two hundred patients entered the 1-yr trial. Fifty of 111 patients received selective flora suppression during the first 6 months in ICU I (test group), while 61 of 111 patients served as the control group in the following 6 months. In ICU II, 49 of 89 patients received no selective flora suppression in the first 6 months (control group), followed by 40 of 89 patients receiving selective flora suppression during the second 6-month period (test group).

INTERVENTIONS

The test group got a mixture of nonabsorbable antibiotics (paste and suspension) in the digestive tract. The control group received paste and suspension without antimicrobial agents. All 200 patients received cefotaxime during the first 4 days.

MEASUREMENTS AND MAIN RESULTS

With the use of selective flora suppression, colonization with aerobic Gram-negative bacilli was significantly (p less than .01) reduced. There was also a significant reduction in nosocomial bronchopulmonary (ICU I and II; p less than .001) and urinary tract (ICU II; p less than .001) infections. The difference in mortality was not significant. There was no development of resistance against the antibiotics used during the limited period evaluated.

CONCLUSIONS

Selective flora suppression is effective in reducing secondary colonization by aerobic Gram-negative bacilli. Reduction of bronchopulmonary and urinary tract infections most likely occurs with colonization prevention.