S 386

14. Pharmacology/Toxicology - Posters 132

185

INTRAVENOJS HALCPERIDC^ DCES NOr CHANGE RENAL FUNC11CN

PROPHYLAXIS AND TREATMENT OF POST-ANAESTHETIC SHIVERING AND UNTOWARD THERMOGENESIS WITH TRAMADOL NC Jovanovid, D Lonbar-Stoiilikovic , T Marenovic

PARAMETERS INPATIENTS RECEI 1 NG LAN DOSE DCPAM I NE.

A Lboez Rodriguez , L Lopez Sanchez, JA Julia.

CBJECi1VES: Dopamine at doses <3 µg/Kgtmin (LDD) increase urine output

(UO). IV haloperidol (IH) is widely used to control agitation in critically ill patients and it is known to possess antidopaminergic actions. We ask if IH could interfere with the renal actions of LDD. DESIC^ ( Prospective study in an University Hospital ICU. SUBJECTS. Data were recorded in 11 consecutive patients with oliguria

(UO<50 mU/h): 7 patients had plasma creatinine (PC)<1 mg/dL (group A) and

4 had PC?2 mg/dL (group B). None received diuretics in the preceding 24 hours, systolic BP was >100 mm Hg and PWP was >10 mm Hg. METHODS. At time 0, LDD was initiated and six hours later a bolus of IH (25 mg) was administered. Renal function parameters (plasma and urine creatinine, sodium, potasium and urea nitrogen and urine specific gravity) were measured at 0 (before LDD), 6 (after LDD, before IH) and 12 (after LDD and IH) hours intervals. RESULTS AND STATISTICAL ANALYSIS: No significant differences (p<0.05) were found in UO and renal function parameters at the recorded intervals in

group A (A0, A6 and Al2) and in group B (B0, B6 and B12). All data is presented as Mean±SD.

Pna Una Pk

Uk BUN UUN Per

Ucr sg UO

138±5 94±40

3.7±0.6

138±5 103±74 4±0,7

140±6 81±53

3.8±0.5

149±8 60±17 4±0.6

150±7 44±18 4±0.3

149±6 38±15

3.7±0.7

77±55 39±30 45±36 47±20 53±33 53±31 30±15 28±14 32±19 238±141 247±148 246±151 1190±766 929±601 1255±546 853±214 1206±605 1183±312

0.66±0.16 0.67±0.1

109±87 1015±7

52±38 1014±6 945±466

0.63±0.1

4.7±2.2

90±62 1015±5 603±429

74±55 1021±6

4.8±2.2

4.8±2.2

70±53 74±45 1016±4 1017±5 445±304 517±189

Pna and Una plasma and urine Na, Pk and Uk: plasma and urine K (mM); BUN and UUN: blood and urinary urea nitrogen, Per and Ucr: plasma and urine creatinine (mg/dL); sg: specific gravity; UO: total urine output (mL) in the interval considered. CCNCLUSI CNS: IH does not change UO, sg, plasma and urine Na, K, cr and urea nitrogen in patients with normal or altered renal function receiving LDD

suggesting that IH does not have a peripheral antidopaminergic action. UCI. Hospital Infanta Cristina. Badajoz 06080, Spain.

OBJECTIVES: Muscular thermogenesis, which accompanies muscular shivering, occurs during the emergence from general anaesthesia, in regional anaesthetic procedures and in post-transfusion pyretic reactions. It is associated with up to 500% increased oxygen consumption, which jeopardises coronary blood supply. Among the drugs used in treatment of muscular thermogenesis tramadol seems to he most efficient and almost devoid of adverse effects. The aim of the study was to determine prophylactic and therapeutic properties of tramadol in different situations associated with shivering. DESIGN: Open, controlled, clinical trial of prophylactic and therapeutic efficacy and tolerability of tramadol in patients with muscular shivering. SUBJECTS: Out of total of 74 patients 20 of them were treated with tramadol prophylactically and 54 therapeutically. Tramadol was prophylactically and therapeutically administered 1 mg/kg iv and in 1-4 it boluses of 25 mg each, respectively. METHODS: Muscular thermogenesis was assessed clinically. The dose of tramadol needed for cessation of shivering and time interval from its administration to cessation of shivering were also recorded. RESULTS AND STATISTICAL ANALYSES: Prophylactic administration of tramadol was 100% successful. Results of therapeutic use of tramadol are given in the following table.

Indication

Average trama-

Time of cessation

dol dose (mg)

of shivering (min)

caesarean section

25.0

immediately

other operations

68.5 ± 20.2

2.7 ± 0.7

post-transfusion

100.0

5.3 ± 0.9

The dose of tramadol used and the time until the cessation of shivering depend on the duration of previous hypothermia. CONCLUSION: Both prophylaxis and therapy with tramadol are effective against muscular thermogenesis and shivering following emergence from general anaesthesia, during regional anaesthesia and after blood transfusions. Department of Anaesthesiology, Institute of Oncology and Radiology, Pasterova 14, YU-11000 Belgrade, Serbia, Federal Republic of Yugoslavia

184

271

ESMOLOL IN THE TREATMENT OF PERIOPERATIVE HYPERTENSION AND TACHYCARDIA D Loncar-Stoiilikovid , M P Stojiljkovic, M Markovid, B Djordjevi6, D Boskovic, S Marenovi/, T Marenovic

THE EFFECT OF SEDATION ON WEANING AFTER CORONARY ARTERY BYPASS GRAFTING: PROPOFOL VERSUS OXYCODONE-THIOPENTAL K Leino , S Nunes, P Valta, I Takala

OBJECTIVES: Esmolol is a novel cardioselective, titratable, intravenous beta-blocker with ultrashort half-life (9 min). The aim of the present study was to investigate therapeutic efficacy and safety of esmolol infusion in the treatment of perioperative tachycardia and hypertension.

DESIGN: Single-blind, randomised, controlled, comparative clinical trial of therapeutic efficacy and tolerability of esmolol plus opioid versus opioid-based technique of management of perioperative hypertension and tachycardia. SUBJECTS: A total of 42 patients (aged 48-68 years) of both sexes, scheduled f o r ophthalmic surgery under balanced anaesthesia, were enrolled in the study. They were

randomly assigned to two equal groups. Half of them received esmolol (infusion 0.3 mg/kg/min during the first 5 min and 0. 1 mg/kg/min thereafter), while remaining 21 patients served as a control group.

METHODS: Systolic and diastolic blood pressure and heart rate were recorded during the perioperative period, and especially during: induction to anaesthesia, intubation of trachea, first incision, surgical manipulations, suture and extubation of trachea. Clinically significant increase in either of these parameters (by 20% of their preinduction values or more) was in both groups treated in the same way - by means of iv bolus of fentanyl (0. 1 mg), either alone, or along with droperidol (5 mg) or, in case of the failure, by means of isoflurane inhalation. Use of all drugs was also recorded.

RESULTS AND STATISTICAL ANALYSES: In the esmolol group of patients systolic and diastolic blood pressure and heart rate were significantly lower than in control group during the extubation of trachea. At the end of operation systolic and diastolic blood pressure were significantly lower after esmolol, as well as diastolic pressure and

OBJECTIVES: To assess the effect of two different sedation methods, propofol (P) (RecofoP, Leiras, Finland) infusion and oxycodone (0) -thiopental (T) bolus dosage, on weaning following coronary artery bypass grafting (CABG). DESIGN: An open, prospective and randonrised study. SUBJECTS: Thirty (fifteen in both groups) male patients going through an elective CABG were studied All the patients had normal lung, liver and kidney functions. METHODS: Sedation was started in the intensive care unit (ICU) after a standardised high dose fentanyl anaesthesia. The goal was to keep adequate but not too heavy sedation (Ramsey's scoring, level 4-5), during which the patient was asleep and had a brisk or sluggish response to loud auditory stimulus or light glabellar tap. Sedation was initiated when patients had recovered to the level 4. P group recieved 0.5 mg/kg bolus followed by an initial infusion of 1.0 mg/kg/h. Too light sedation was met by increasing the infusion rate with 25 % after a 0.5 mg/kg bolus. The infusion rate was lowered 25 %, if the sedation was too deep. In the O-T group the patients recieved 0.1 mg/kg 0 and 0.7 mg/kg T boluses simultaneously. The sedation was maintained by giving the initial boluses again when needed Weaning was performed using pressure support (PS) trials of 5 minutes followed by 15 minutes of syncronised intermittent mandatory ventilation (SIMV) if necessary (end tidal CO2 >6 %). Extubation was done according to an arterial blood sample analysis. RESULTS AND STATISTICAL ANALYSES: There were no statistical differences (unpaired t-test) in operation times, anaesthetics given intraoperatively (mg/kg) or patient characteristics between the groups. Main results are presented in the table.

Table. Weaning data (mean ± SD). Unpaired t-test and Wilcoxon's test were used.

heart rate during the intubation of trachea (by 24%). At the same time, the control patients needed by 54% and 113% larger doses of droperidol and pancuronium, respectively. Isoflurane was added by 32% more frequently in control than in esmolol

group. Patients treated with esmolol were able to open their eyes upon verbal command in 4.2 times shorter time interval than those in the control group. CONCLUSION: Esmolol assures better cardiovascular stability during critical phases of anaesthesia and operation in ophthalmic surgery. This control of reflex sympathetic

hyperactivity in esmolol group is not associated with excessive analgesic and anaesthetic medication and therefore emergence from anaesthesia in these patients is shorter.

Department of Anaesthesiology and Intensive Care Therapy, Military Medical Academy Crnotrayska 17, YU-I 1000 Belgrade, Serbia, Federal Republic of Yugoslavia

Time from stopping sedation to extubation (min) Time from starting weaning to extubation (min) Pressure support trials (n)

Proptifol group

Oxycodone-thiopental group

p

125±28

191±88

0.013

63±24

112±63

0.011

1.7± 1.0

4.0±3.1

0.046

CONCLUSION: We compared the effect of two different sedation methods on weaning after CABG. The use of propofol infusion resulted in a shorter weaning period and quicker extubation compared to the oxycodone-tiopenthal bolus dosage. ICU, Kuopio University Hospital, P.O.B. 1777, FIN-70211 Kuopio, Finland

S 387

279

426

IS AMIKACIN EMPIRICAL DOSAGE SCHEME RATIONAL FOR ICU PATIENTS ? F.Stathoulopoulou 2 , E.Pavlou', Ch.Delaki l , H.Thymianou 2 , E..Ioannidou' Alms-Objectives: Current practice for amikacin dosage for all pts with good renal function, including ICU pts is either 500mg BD or 7.5 mg/ kg/12 hours. We hypothesized that in ICU we would aim to higher peak concentrations in accordance to both higher MICs due to greater incidence of resistance patterns, and to the pathophysiological changes often met in ICU pts that could influence the clinical pharmacokinetics of amikacin . Methods : In this prospective, non comparative study, 14 ICU pts were enrolled, all mechanically ventilated, hemodynamically stable, with good renal function, all with documented infection and identified pathogen. Pts. were given the dose of (6.25- 7.64) mg/kg /12h as a 30 min infusion. On the 2nd day of treatment serial blood samples were taken: First sample just:

A COMPARISON OF SERUM CIPROFLOXACIN CONCENTRATIONS IN INTENSIVE CARE PATIENTS WITH NASOGASTRICALLY AND INTRAVENOUSLY ADMINISTERED CIPROFLOXACIN. L J Giles, Dr S P Bass, Or M A Stockwell

before the dose and 30 min, 1 h , 2h, 3h, 6h, 8h, 10h, 12h, post dose.

Pharmacokinetic profile is analyzed by using a two compartment model and evaluated with respect to: 1) The predicted pharmacokinetic parameters using calculations based on population data 2) The reported MICs. Results : Fourteen pts, 8 males 6 females, median age 49.5 yrs Range (21-80) yrs with comparable ISS, GCS, Apache II scores. Predicted parameters Actual parameters n 2:x3±8Raige(5.539) pmA01. (Vd) meen166 L±185Rarge(143-2a9). 6.64LIh±286 Ratge(227-9.16) DNS (Cl) 5.8LM±17RarW(2&8'16) (T1/2) 2.14 h ± 0.69 Range 1.3-4.2 3.6 h ±1.38 Range (2.5-6) (Kel) 0.349 h'±0.09 R'(0.-0.5z38 022h'±009 Rarge(0.1547-0.39 Observed values Range (52-220.3) 128mg x h x L - ' ± 57.81 AUC Range (13-39) 25mg /L ± 7.61 Peak concentration (C o „ k ) mean 10.6 mg/ L ± 4.82 Range (5-18.4) Average concentration (C.,) Trough concentrations (C 1 h ^ 10 pts (71 %) < 5 mg IL, 4 pts (29%) < 10 mg /L The majority of MICs for Ce ram (-) bacteria were > 16 mg/L Discussion-Conclusion: The differences in Half Life (T1/2), Elimination Constant (K,,) are attributed to the high differences in the Volume of Distribution (Vd) . The high interpatient variability in the actual Vd accounts for the differences in the observed Cpeek , C ss , Ctrw. We would aim to higher Cpeak because of the higher MICs reported for our ICU pts , considering also the fact that the average concentration is at the lower limit of 10.6 mg/L , low even for the general population. Observed troughs are not considered as predisposing factor for nephrotoxicity , but lower ones. would be desirable because of concomitant nephrotoxic drugs.There is a need to review our current dosage practice even to consider the once-a-day administration based on the individualization of the daily dosage in order to overpass the high inter and intrapatient variability . ICU' and Pharmacy Department 2 KAT General Hospital ,Nikis 2 Kifissia,

OBJECTIVE: To assess the serum Ciprofloxacin concentrations in intensive care (ICU) patients treated with the maximun licenced doses of nasogastrically (NO) (750mg b.d.) and intravenously (IV) (400mg b.d.) administered Ciprofloxacin. SUBJECTS: Twenty ICU patients receiving Ciprofloxacin for presumed or actual Pseudomonas infection. Serum Ciprofloxacin levels were determined for IV dosing in 11 patients and for NO dosing in 10 patients. All patients had established NG feeding and were receiving NO ranitidine 150 mg b.d. All patients included had normal renal function with serum creatinines of <12ommols/l. METHODS: After 48 hours (steady state) of IV Ciprofloxacin samples were taken one hour preadministration (trough) and one hour post infusion (peak). Patients were changed to HG Ciprofloxacin 750 mg b.d. when clinically indicated. Trough serum samples were taken and peak levels two hours after NO administration. Samples were analysed by high performance liquid chromatography. RESULTS: Analysis showed mean intravenous trough levels of 0.25+1-0.07mg/I and peak levels of 1.51+/-0.14mg/I (n=11). Nasogastric trough levels were 0.8+/0.18mg/I and peak levels of 2.92+/-0.5mg/I (n=10). Deviation from the mean is expressed as standard error of the mean. CONCLUSION: This data suggests that intravenously administered Ciprofloxacin 400 mg b.d. provides inadequate trough and peak levels for Pseudomonas aeruginosa eradication. Adequate serum concentrations are thought to be a trough of >0.5mg/I and a peak of >2.5mg/I. NO administration provides adequate peak and trough levels in patients concurrently receiving NO feeding and H2 antagonists. NO administered Ciprofloxacin is an alternative to the intravenous route and may have substantial cost saving implications. Further investigation is needed to clarify the correct dosing regimes (NO and IV) for effective antipseudomonal activity of ciprofloxacin. Intensive Care Unit, St Helier's Hospital, Carshalton, Surrey. SM5 IAA. U.K.

Athens 14561, Greece.

361

442

DOES METOCLOPRAMIDE ENHANCE GASTRIC EMPTYING IN CRITICALLY ILL PATIENTS? C.A. Jooste J.E. Mustoe, W. Bradbury, G.G. Collee

NEAR FATAL CENTRAL ANTICHOLINERGIC SYNDROME AFTER INSTILLATION OF TROPICAMIDECONTAINING EYEDROPS - REPORT OF A CASE GA Brunner A Lueger, P Kaufmann, KH Smolle, GJ Krejs

We investigated the efficacy of metoclopramide on gastric emptying in enterally fed critically ill patients. Metoclopramide is known to display prokinetic properties on the gastrointestinal tract, but these effects have not been authenticated in critically ill patients. Ten patients in our intensive care unit were studied. The patients were randomised to receive intravenous metoclopramide 10mg (treatment) or saline (control) on day one, and saline or metoclopramide on day two. Each patient was studied over two consecutive days thereby acting as his/her own control. Gastric emptying was measured indirectly using small bowel absorption of a single bolus dose of 1.5g of paracetamol administered via the nasogastric tube on both the treatment and control days. [1] Blood samples for analysis were taken at time zero and at 15, 30, 45, 60, 90 and 120 minutes after the administration of intragastric paracetamol and intravenous saline/metoclopramide. The plasma concentrations of paracetamol were estimated by using fluorescence polarisation immunoassay.

Pt No

Metoclopramide

Control

1 2 3 4 5 6 7 8 9 10

1102.50 1596.00 1136.25 568.50 2543.25 1942.50 1652.25 1053.00 1039.50 1236.75

355.00 861.00 1621.50 404.25 2163.75 1356.00 417.75 295.50 534.75 1431.00

_

Results are presented as area under the paracetamol absorption curve in the table. Statistical analysis using the Wilcoxan signed rank test gave a p value of 0.04, indicating a significant increase in gastric emptying following metoclopramide administration.

CONCLUSION: In our study, the administration of metoclopramide to a heterogeneous group of critically ill patients improved gastric emptying in eight of the ten patients studied. Metoclopramide may facilitate early establishment of enteral feeding in critically ill patients. Reference 1: Clements et al: Clin Pharmacol Ther 1978; 24:420-431 Dept. of Anaesthesia, The Royal Free Hospital, Pond St. London, NW3 2QG, UK

We report the first case of a near fatal anticholinergic intoxication after instillation of tropicamide 0.5% containing eyedrops, which have been regarded as completely safe so far. A 62-year old Caucasian male with an uneventful medical history received eyedrops containing tropicamide 0.5% and phenylephrine 1% for routine fundoscopy. During examination he complained about nausea and a dry feeling in his mouth. Half an hour later he experienced a series of two generalized seizures, fell into a deep coma (Glasgow Coma Scale 3) with bilateral positive Babinski sign and required mechanical ventilaton. Upon admission at our ICU physical examination revealed maximal mydriasis with no reaction to light, a flushed and dry skin and tachycardia at 123 bpm. His abdomen was soft with no bowel sounds.. Blood pressure, blood sugar level, oxygen saturation and electrolytes were normal, his temperature was not recorded. We suspected a central anticholinergic intoxication behind these symptoms and administered 2 mg of physostigmine intravenously over a 5-minute period. All symptoms except tachycardia disappeared within a few minutes and the patient could be extubated (Glasgow Coma Scale I5). As his condition remained stable, he did not require another administration of physostigmine. The patient was transferred to the general floor 24 hours later and discharged home after a 72 hour hospital stay. All physicians using tropicamide-containing mydriatics should be aware of this very rare but possibly fatal side effect of this agent. Intensive Care Unit, Karl Franzens University, 8036 Graz, Austria.

S 388

474

512

ROCURONIUM INFUSIONS IN THE INTENSIVE CARE UNIT RI Ra , A. Hadbavny, J. Hoyt

CONTINUOUS INFUSION OF OBIDOXIME AND THE IMPORTANCE OF PLASMA LEVELS MONITORING IN ORGANOPHOSPHATE POISONING

OBJECTIVES: We evaluated the neuromuscular blocking agent rocuronium in the intensive care unit (ICU). We hypothesized that doses could be minimiz ed and acceptable recovery times achieved when train of four (TOF) monitoring was used. DESIGN: -This was a prospective descriptive study. SUBJECTS: Patients hospitalized in the medical-surgical intensive care unit who received rocuronium infusions were enrolled. METHODS: Patients received a bolus of 50 mg (approximately 0.6 mg/kg). An infusion was then titrated to maintain a minimum of two of four twitches on TOF monitoring which was performed hourly. RESULTS AND STATISTICAL ANALYSES: The average dose was 0.28 mg/kg/hr. Recovery times were acceptable except in one case where TOF was intentionally kept at 0/4 due to instability. Two patients achieved clinically acceptable paralysis although 4/4 twitches were maintained. Addition of theophylline raised the dose of rocuronium required. P atientAPACHE II Total timeAvera geDose Recovery Time core (hours) (mg&af t) rl .13J2B0) EC* 40 45.25 0.18 0.5 AZ" 26.67 0.17 1.5 19 AT 19 88.68 0.30 NA JGt 7.75 19 85 0.17 14.75 DD** 20 266 0.29 JG* 45 52 0.09 NA JP"" 23 0.62 143.25 0.83 CW 16 48 0.40 2.0 *renal and hepatic insufficiency; "received neostigmine when 3/4 after rocuronium discontinued,- (renal failure patient very sensitive to rocuronium and tended to be 0/4 or 1/4; **maintained at 0/4; "'received theophylline; NA-not applicable: pt with TOF 4/4 when rocuronium discontinued CONCLUSIONS: The average dose of rocuronium required was half the recommended dose of the manufacturer (0.6-0.72 mg/kg/hr). Patients with renal or hepatic impairment required lower doses. Theophylline antagonized the effects of rocuronium. Maintenance of at least 2/4 twitches with TOF monitoring allows an adequate degree of neuromuscular blockade and acceptable recovery times.

I. Araaao#, M.E. Soares*, M.L.Bastos*, M.Lopes# # Intensive Care Unit (UCIP) - Hospital Santo Antonio - Porto, Portugal * Laboratory of Toxicology, Faculty of Pharmacy - Porto, Portugal

Department of Critical Care Medicine, St. Francis Medical Center, 400 45th Street, Pittsburgh, Pennsylvania 15201, USA

OBJECTIVES: Oximes are the elected antidotes used in humans intoxicated with organophosphate insecticides. In Portugal obidoxime chloride is the most commonly drug used as cholinesterase reactivator. This study's objective is to validate its administration in a continuous intravenous perfusion. DESIGN: This prospective study included the worst patients intoxicated with organophosphate insecticides admitted in the ICU between March through November 1995. These patients were submitted to the conventional treatment plus a continuous infusion of Obidoxime, with the purpose of stable plasma levels of this antidote. A reliable method for quantification of obidoxime in plasma of patients was first developed in Portugal in this study. SUBJECTS: Five patients intoxicated with organophosphate insecticides with suicidal purposes assisted in the Intensive Care Unit of Hospital Santo Antonio were included in this study because they meet the following criteria: 1) Plasma Cholinestares < 25% the normal value, 2) Need of endo-thraqueal intubation and artificial ventilation. The mean SAPS value of these patients was 51 (23-74). The ingested toxics were in 4 cases from Class I (ometoato, paratiao, metidatiao) and in 1 case Class II (quinatfus). METHODS: Besides the classical treatment of these intoxications (oxigenation and secure the airway, reversion of the muscarinic action with atropine administration, prevention of gastric absorption by means of gastric pro-kinetic emptying and lavage) the Obidoxime treatment included: an initial bolus of 250 mg during 1 hour, followed by continuous infusion of 25-20 mg/kg/day during 48-96 hours. Plasma samples were collected at different occasions after the beginning of the perfusion. Chromatographic analysis by HPLC-UV was developed for plasma drug levels monitoring. RESULTS: 2 groups of Obidoxime plasma levels were found: 1) within the Clark therapeutic levels (7,5-23,7 sg/ml), 2) toxic levels (patient #4 35 µg/ml, patient #5 28 gg/ml). It was observed that plasma cholinesterases level did not raise significantly more in spite of the reactivator use. The eritrocitary cholinesterases did not raise as fast as described in the literature. CONCLUSION: 2 cases of hepatocelular dysfunction preceded by renal impairment were observed with plasma levels over 25 isg/ml leading to the recommendation of reduction or suspension of Obidoxime administration. The different drug plasma levels found in spite of the same dosage regimens, justify the need for close monitoring in order to readjust the therapeutic doses. The therapeutic and toxic plasma levels for Obidoxime are still not universally accepted. Larger studies are needed in order to establish the optimal dosage regimens. Irene Araaao-UCIP Hospital Geral Santo Antonio LgProf.Abg1s glazar. 4050 Podo, Portugal

475

535

A NEW MODEL OF HOW RESPIRED HUMIDITY AFFECTS AIRWAY FUNCTION. R Williams*, P Seakins*, N Rankin**, T Smith**, D Geller**

PHOSPHODIESTERASE INHIBITORS IMPROVE CARDIAC OUTPUT IN CATECHOLAMINE RESISTANT SEPTIC SHOCK S J Mackenzie , I S Grant

OBJECTIVES: Intubated patients undergoing mechanical ventilation require'. humidification of their respired gases to avoid airway injury. It is uncertain^l l what level of delivered humidity is optimal. The objective of this study is to develop a model of the interaction of humidity and temperature with airway function, in order to determine optimal humidity levels. METHOD: Relevant indicators of the efficacy of delivered humidity were: The literature on how the function of these indicators is affected by, respired humidity and temperature was then reviewed. A conceptual model of the physiology of this interaction was developed. The models ability to predict the dysfunction sequence, which proceeds from non-optimal humidity, was!, tested by comparing it to existing observational data. RESULTS: The model successfully provides a rational explanation for the dysfunction sequence. Each of the indicators was found to be optimised when, gases were conditioned to the patient's core temperature and saturated withi water vapour (molecular not aerosol). Existing data were classified by dysfunction and subject to linear discriminant analysis, with 77% cross) validation success. Any deviation of delivered humidity from core temperature. saturated will cause progressive airway dysfunction over the exposure period.: The greater the deviation the faster the dysfunction sequence proceeds (seer table).

OBJECTIVES: Patients in septic shock often require inotropic support. Catecholamines are widely and generally effectively used, but a minority of patients respond inadequately to them. We have studied the haemodynamic effect of phosphodiesterase inhibitors (PDEIs) in these patients. DESIGN: Unblinded, uncontrolled pilot study. SUBJECTS: Ten patients (3 male), suffering from septic shock who failed to achieve or maintain haemodynamic goals after fluid and catecholamine therapy. Patients aged between 44 and 81 years (mean 64) and APACHE II scores 211033 (mean 27). METHODS: All patients were mechanically ventilated and had indwelling arterial and thermodilution pulmonary artery catheters. All had an inadequate cardiac output despite receiving catecholamine infusions (adrenaline in 9 patients, noradrenaline in 5, dopamine in 2, dobutamine in 2). A phosphodiesterase inhibitor was introduced (enoximone in 7 cases, milrinone in 3) and haemodynamic measurements were repeated. RESULTS: Data were analysed using a T test. The results are shown in the table. Before Significance After (Mean±SEM) (Mean±SEM) Heart Rate 113±7.2 NS 119±7.0 MAP (mmHg) 87±4.8 85±5.6 NS CVP (mm Hg) NS 15±1.2 14±1.5 PAOP (mm Hg) 15±1.5 NS 12±1.8

Absolute Humidity below core mH2O/L Time to Mucociliary Dysfunction Hrs Time to Cell Damage Hrs

0

-10

-20

-30

-40

-44

>24

>24

0.5

<0.1

<0.1

<0.1

>24

>24

18

4

1

0.5

CONCLUSION: Inappropriate humidity levels delivered to the patient will initiate a sequence of airway dysfunction. A model of the relationship between delivered humidity, exposure duration and airway dysfunction has been derived. Existing data supports the model's predictions. Airway functions may be optimised when respired gases are delivered to the patient conditioned to the patients core temperature and saturated with water vapour. *Fisher & Paykel Healthcare, Carbine Rd, Panmure, Auckland, New Zealand **Intensive Care Unit, Middlemore Hospital, Otahuhu, Auckland, New Zealand

CI (I/min/m 2 )

SVR (dynes.s/cros) 1401±155

2.7±0.2

4.0±0.3 967±166

p<0.01 p<0.05

PVR (dynes.s/cme) 342±40.7

241329.8

NS

D02I (ml/min/m 2 ) 405±33.1

566±45.4

p<0.05

V0 2 I (ml/min/m 2 ) 132±10.7 Qs/Qt (%) 23.5±2.4

143±9.7 27.7±2.2

NS NS

Lactate (pmoUl) 1.36±0.2 NS 1.36±0.2 Some alterations to catecholamine infusions were required. No serious adverse effects were seen. Four patients survived. CONCLUSION: Phosphodiesterase inhibitors can be used to augment cardiac output in patients with catecholamine resistant septic shock. Serious adverse effects were not seen. Controlled trials are required to confirm both efficacy and safety. Intensive Therapy Unit, Western General Hospital, Edinburgh, EH4 2XU, UK.

S 389

555 THE EFFEUT OF FENTANYL ON CORONARY PERFUS ON PRESSURE IN ICU PATIENTS L.Tourkochoriti,E.Kerabatsos , B.Zidianakis, G.Filippatos, E.Boutzouka, G.Baltopoulos.

Intensive Care Unit, "Agioi Anargyroi" Cancer Hospital, 14564 Nea Kifissia, Greece. The use of a bolus dose of sedative drugs in critically ill patients is very common in the intensive care unit(ICU).The aim of this study was to estimate the effect of fentanyl on coronary perfusion pressure in ICU patients. Twelve severely ill patients, who were intubated and mechanically ventilated and had right cardiac catheter in place were studied.They were given 5y/kgr fentanyl i.v. bolus for sedation.Measurements of mean arterial pressure(MBP),cardiac index(CI) and coronary perfusion pressure(CPP) were obtained before and 3, 15 and 30 minutes after the administration of the drug. Statistical analysis was done using ANOVA. The results are expressed as mean value ±1 SD and are shown in the next table. MBP mmHg

0 min 100±14

3 min 77±14*

CI

3.7±1.1

3.0±0.9*

3.3±0.9*

3.3±1.0*

57±16

43±13*

45±10*

48±11*

lit/min/m'

CPP mmHg

15 min 81±10*

30 min 91±16

*P<0.05 versus 0 min values. The results indicate that fentanyl significantly decreases MBP, CI and CPP 3 minutes after the injection. There is a slight improvement in values 15 and 30 minutes after the administration of the drug but the difference remained statistically significant in correlation with zero values.

743 EFFECT OF DOPEXAMINE ON VENTILATION - PERFUSION (VA/Q) - DISTRIBUTION AND GAS EXCHANGE IN HUMANS T. Hachenbere , S. Karmann, H. Thomas, B. Pfeiffer, M. Griindling, M. Wendt OBJECTIVES: Dopexamine is a dopaminergic (DA - 1) and 02 adrenoceptor agonist which

increases splanchnic blood flow in patients with chronic congestive heart failure (1) or

sepsis (2). Infusion of dopexamine may improve splanchnic perfusion during cardiac or non - cardiac surgery, however, the effects on ventilation - perfusion (VA/Q) distribution and gas exchange have not been studied. DESIGN: Prospective longitudinal study SUBJECTS: Eight patients (height 170 ± 8 cm, weight 70 ± 6 kg, age 57 ± 11 years) without cardiopulmonary diseases undergoing elective major abdominal surgery METHODS: The study had been approved by the ethics committee and informed consent had been obtained from each patient. Monitoring included heart rate (HR), mean arterial (MAP) and pulmonary arterial pressure (MPAP), cardiac output (CO), arterial 02 - and CO2 - partial pressure (Pa02, PaCO2). Intrapulmonary shunt (QS/QT), perfusion of areas with 0.005 < VA/Q < 0.1 [VA/Ql ow ], ventilation of areas with 10 < VA/Q < 100 [VA/Qhigh] and dead space ventilation (VW/VT) were calculated from the retention and elimination of six inert gases (3). Data were obtained during controlled mechanical ventilation (F102 = 40 % in N2) before and during infusion of dopexamine (1.0 pg/kg/min and 2.0 pg/kg/min) for a period of 30 min. The results were analyzed by multiple analysis of variance and post - hoc Dunnett's- test. P < 0.05 was considered as significant.

RESULTS: HR [beats/min) MAP [kPa] MPAP [kPa] CO [L/min] Pa02 [kPa] [kPa] PaCO2 Q s/Qi [%of Co] V .q/Ql ow [% of CO] V.4/Qhigh [%of V g] V D/VT [%of Vg]

Control

Dopexamine

Dopexamine

92 ± 10 11.1 ± 1.6 2.3 ± 0.5 5.5 ± 0.7 17.6 ± 6.3 4.8 t 0.8 14±10 3±5 5±7 34±12

104 ± 12 ** 11.9 ± 2.5 2.0 ± 0.4 6.3±1.0* 20.9 ± 6.7 * 4.8 ± 1.0 5±5* 4±7 7±6 34± 14

106 ± 9 -* 12.1 ± 2.7 2.5 ± 0.3 7.2 ± 1.9 18.9 ± 5.6 4.8 ± 0.7 11±12 1±1 2±3 37±9

1.0 µg/kg/min 2.0 pg/kg/min

Mean ± standard deviation * = p < 0.05 an = p < 0.01 CONCLUSION: In anaesthetized patients without cardiopulmonary diseases infusion of dopexamine (1.0 Itg/kg/min) significantly increased HR (+ 11 %), CO (+ 14 %) and Pa02 (+ 19 %). QS/QT decreased by 155 %, possibly due to a redistribution of pulmonary blood flow towards better ventilated alveoli. CO increased further during infusion of 2.0 pg/kg/min dopexamine , but this caused a deterioration of V A /Q - distribution. REFERENCES: (1) Am J Cardiol 62: 30C - 36C, 1988 (2) Crit Care Med 22: 789 - 795, 1994 (3) J Appl Physiol 36: 588-599, 1974

L. Tourkochoriti, Ifassiou 49, 15772 Zografou, Athens, GREECE

Department of Anaesthesiology, Ernst - Moritz - Arndt Universitat Greifswald, Germany

626

764

THE PHARMACOKINETICS OF MORPHINE IN VENTILATED TRAUMA PATIENTS

A POSSIBLE INTERACTION BETWEEN 546C88 AND DOBUTAMINE IN HUMAN SEPTIC SHOCK G Wray , R Skinner, D Rees and D Watson.

H Berkenstadt , E Segal, H Mayan, S Almog, S Dani, A Perel, D Ezra.

Morphine (MO) is commonly used for the sedation and pain relief in trauma patients. The pharmacokinetics of the drug may be affected by factors such as hepatic injury, changes in cardiac output, mechanical ventilation and drug-drug interaction. The aim of the present study was to define the pharmacokinetics of MO in severe trauma patients, and to find the correlation between the pharmacokinetics and several clinical variables. Methods: Eight ventilated trauma patients, ages 24-91 years, APACHEII score 28-37 were studied during the first 48 hours of their admission to the ICU. Blood samples for plasma MO concentration (determined by RIA technique) were drawn through an arterial line: 1,3,5,10,15,30,45,60,75,90, 105,120,180,240 300 min. after its IV injection in a dose of 0.025 mg/kg. Results: Pharmacokinetic data using a two compartments model are described in the table. In comparison to data from anesthetized (1) and healthy volunteers (2), MO clearance was found to be lower, the volume of the central compartment (V 1) higher, the apparent volume of distribution (Vss) lower, and the mean elimination half life (Tl/20) similar, Inter patients variability was high, and could not be explained on the basis of the patient's age, severity of injury, presence of organ failure, hemodynamic and respiratory status. Vss V1 (1/kg) T1/20 (min) Clearance (mi/min/kg) (Vkg) MEAN±SD Normal Values (anesthesia) (1) Normal Values

(volunteers) (2)

6.4±2.9 23±3.1 24110

0.23± 0.14 0.13± 0.06

1.46±1

106±46

3.4±0.6

104±15

3.3±0.9

114±30

Conclusion: MO clearance is. significantly decreased in severe trauma patients with apparently stable hernodynamic status. Hepatic blood flow decrease might be the main reason. References: 1. Anesthesiology 54:187-192,1981 2. Goodman and Gillman eighth edition pp 1182.. Departments of Anesthesiology, Critical Care and Clinical Pharmacology, Sheba Medical Ce nter Tel Ha^hsmgr, Ismel 52621

We report a possible interaction between the nitric oxide synthase (NOS) inhibitor 546C88 (L-N ° -Methylarginine HCI) and dobutamine(DB) in septic shock. A patient in established septic shock was entered into the open-label, dose escalation, pharmacokinetic and pharmacodynamic study of 546C88. The patient received an 8 hour infusion of 546C88 at a fixed dose of 10 mg/kg/hr. Mean arterial pressure (MAP was maintained between 70-90 mmHg, and systemic vascular resistance sustained while noradrenaline (NA) requirement was reduced from 0.39 µg/kg/min to 0.18 pg/kg/min. DB (5gg/kg/min) was introduced to increase cardiac index (CI) 285 minutes (t=12.45) after the end of the 546C88 infusion. At this time the plasma level of 546C88 was 24 gg/mL. Within 10 minutes the blood pressure rose to 300/150 mmHg, SVRI 3240 dyne.sec/cro s/m 2 and CI 3.43 L/min/m z . Both DB and NA infusions were stopped, blood pressure returned to normal, and there were no further hvoertensive surees. Haemodvnamic events are summarised in the followine table. Pre End 546C88 During Post Post Post Infusion

t=0

t=4

t=8

t=12

t=12.45

t=12.55

Heart Rate

127

109

93

78

81

81

MAP

76

79

82

77

154

108

PAOP/CVP

15/11

13/12

11/11

11/10

17/15

15/11

CI

5.5

4.9

2.7

2.9

3.4

3.4 2276

SVRI

867

1147

PVRI

2375

60

1984

3240

108

367

137

147

101

NA/DB rate 546088 rate

0.39/0 0

0.21/0 10

0.18/0 10

0.08/0 0

0.06/5 0

0/0 0

(PAOP-pulmonary arterial occlusion pressure& CVPcentral venous pressure both mmHg; S/PVRI-systemic/pulmonary vascular resistance index; dyne.sec/cro s/m 2 , NA/DB rate pg/kg/min and 546C88 rate mg/kg/hr) Discussion : A synergistic action between methylarginine and DB on MAP has been reported in endotoxaemic dogs'. Dobutamine is an agonist at 11, (i2 and a adrenoceptors with an inovasodilator profile. The observed pressor effect in this report suggests the emergence of a predominant a mediated vasoconstriction during NOS inhibition. This may have important implications for the use of 546C88 together with DB in the treatment of septic shock. Reference : (1) Kilbourn RG et al. Crit Care Med 1994; 22(11):1835-40. Intensive Care Unit, The Homerton Hospital, Homerton Row, London E9 6SR.

S 390 768 CHANGES TO PLASMA IONISED CALCIUM AND MAGNESIUM LEVELS FOLLOWING THE RAPID INFUSION OF ALBUMIN M. Marearson , F. Noormohamed, N. Soni OBJECTIVES: Albumin has binding sites with high affinity for divalent cations such as calcium and magnesium. In vitro work has shown that the level of ionised calcium in plasma is reduced following the addition of increasing amounts of albumin. We have previously described a bedside technique to measure vascular permeability, (the Albumin Distribution Index, ADI) which requires the rapid administration of a 40g bolus of albumin given as 200m1 of a 20% solution. In order to determine the effect of this albumin bolus on plasma calcium and magnesium levels in vivo, the following study was carried out. SUBJECTS: Eleven patients with sepsis syndrome on the intensive care unit. METHOD: Subjects were given 200ml of 20% human albumin solution (BPL, Elstree) as a rapid infusion over 2 minutes via a central venous cannula, as part of the ADI technique of permeability assessment. Arterial blood specimens were collected into sodium heparin (Vacutainer 45) tubes prior to, and 1 minute, 5 minutes and 30 minutes after the albumin bolus. Specimens were separated immediately, and plasma analysed on a Kone Microlyte 6 Ion Selective Analyser, with correction to pH 7.4. Statistical analysis was by ANOVA for repeated measures followed by Newman-Keuls test. RESULTS:

I minute Baseline Albumin (g/I) 24.9±6.5 14.0±5.1 (% chan a vs. baseline) +78%/0 0.302 Haematocrit 0.291 (% change vs. baseline) ±0.045 (-4±2%) Ca" (mmol/1) 1.44±0.24 1.34±0.21 (% change vs. baseline) (-7±3%) Mg" (mmol/1) 0.78±0.24 0.7410.21 (% change vs. baseline) (-5±3%)

5 minutes 23.8±6.4 +70% 0.284 (-6±2%) 1.36±0.22 (-6±4%) 0.73±0.21 (-5±5%)

30 minutes 22.4±6.3 (+60%) 0.278 (-8±3%) 1.39±0.23 (-3±3%) 0.76±0.23 (-3±5%)

CONCLUSION: The rapid infusion of 200m1 of 20% albumin solution causes a transient reduction in ionised calcium and magnesium levels only marginally greater than that due to dilution alone. Departments of Intensive Care and Academic Therapeutics, Chelsea and Westminster Hospital, London SW I O 9NH