[Quality comparison of modified neurolept-, balanced and intravenous anesthesia. 1. Study design and patient analysis of the Krefelder study 1992].

The choice of appropriate anaesthesia in a more or less seriously ill patient requires detailed information on the risk and tolerance of each specific anaesthetic regimen. The objective of this prospective, randomised clinical trial was to test the hypothesis that three regimens of general anaesthesia--neurolept-(NLA), balanced (BAL), and intravenous propofol anaesthesia (IVA)--differ with regard to safety and comfort. The criteria for the intraoperative safety and postoperative comfort of the patients were the incidents, events and complications (IEC) that required medical treatment as well as the evaluation of postoperative complaints by the patients according to the IEC list and patient questionnaires of the German Society of Anaesthesia and Critical Care Medicine (DGAI). METHODS. The study duration was about 4 months, from January to April 1992. During this period the patients of all nine operative departments of the hospital received strictly randomised NLA, BAL, or IVA. Patients who had regional anaesthesia or were not capable of understanding the German language, were nonco-operative, or were seriously ill (ASA class IV to V) as well as children under 18 years of age did not participate in the study. All eligible patients provided their informed consent. ANAESTHESIA. For premedication 10 mg chlorazepate was administered the night before and on the day of surgery. Anaesthesia was conducted under normoventilation using a mixture of 70% nitrous oxide and 30% oxygen. NLA patients were induced intravenously with 0.2 mg/kg body weight etomidate and received 0.005 mg/kg fentanyl and 0.07 mg/kg droperidol before the start of surgery. The repetition dose was 0.2 mg fentanyl and 2.5 mg droperidol. In the BAL patients the dose of fentanyl and droperidol was reduced to 50% due to the addition of isoflurane up to 1 vol. %. IVA patients received 2 mg/kg propofol over 3 min followed by an infusion of 3-5 mg/kg per hour together with 0.2 mg fentanyl/h. Neuromuscular blockade was accomplished with vecuronium 0.1 mg/kg. If the blood pressure and heart rate increased by more than 20% of preoperative values, analgesia was reinforced by an additional fentanyl dose. Anaesthesia was subsequently enhanced by increasing the neurolept/propofol/isoflurane dose by up to 50%. DATA COLLECTION. The following parameters were registered: patients' personal data and physical condition according to ASA classification; the grade of risk according to the Munich risk checklist; the frequency of IEC during surgery; the patients' permanent medications; postanaesthetic vigilance and recovery; the acceptance of the assigned anaesthetic by the physician; the cost of the anaesthetic used; and pre- and post-operative complaints as well as the assessment of anaesthesia by the patient. The statistical evaluation was performed using the chi-square test. RESULTS. A total of 1,346 patients were enrolled in the study; 28 (2%) were excluded because the treatment protocol was changed by the anaesthesiologist. Seventy per cent were recruited from general, gynaecologic, or otorhinolaryngologic surgery. The three anaesthetic regimens (NLA, BAL, and IVA) were used in other departments with the same frequency with the exception of ophthalmology and urology (P > 0.1) (Fig. 1). Of the 1,318 eligible patients, 443 received NLA, 443 BAL, and 432 IVA (P = 0.8). The distribution of the various parameters was surprisingly similar among the three groups: the average age was 50 years (P = 0.91), body weight 71 kg (P = 0.33), reference or initial blood pressure 130/80 mm Hg (P = 0.36), average time of anaesthesia 103 min (P = 0.82), and all had the same risk score (P = 0.42). Sixty per cent were female. An average of 85% of the 18- to 89-year-old patients were considered to be healthy according to the ASA risk classification (P = 0.42). However, on applying the Munich risk checklist the average number of healthy individuals was 5% to 10% lower than that of the ASA risk classification.