PAIN OUT: the making of an international acute pain registry.

BACKGROUND: About 240 million patients undergo surgery every year, worldwide. Roughly 50% of these patients report clinically significant pain. Numerous barriers impede provision of adequate management. Lack of evidence about appropriateness and effectiveness of interventions is one. A registry can provide such information, eventually facilitating better management. This paper reports the development and feasibility of PAIN OUT, the first international acute pain registry, established with funds from the European Commission, and presents preliminary analysis to illustrate the nature of investigations that registry data make possible. METHODS: On the first postoperative day, 6347 adult patients undergoing orthopaedic or general surgery, in 11 medical centres in Europe and Israel, provided Patient Reported Outcomes (PROs) using a validated questionnaire. Clinical data were abstracted from the patient's chart. RESULTS: Feasibility worked well. Over a period of 1 year, surveyors accrued targeted data sets and entered them into an online browser. Collaborators could receive online feedback comparing their findings about PROs against anonymized findings from other centres. Missing data for the majority of variables were low. Despite considerable variability between institutions, a large number of patients were treated according to the generic, evidence-based recommendations we assessed. However, this was not sufficient to result in acceptable outcomes for the majority of patients. CONCLUSION: The initial development of PAIN OUT has been achieved. From 2013, it continues as a not-for-profit academic project, open to clinicians and researchers worldwide. The International Association for Study of Pain and PAIN OUT will work together to maintain, disseminate and develop the registry.

Epidural vs. intravenous fentanyl during colorectal surgery using a double-blind, double-dummy design.

BACKGROUND: The overall therapeutic effectiveness of epidural fentanyl vs. the intravenous route is controversial. The present work describes a randomized, controlled, double-blind, double-dummy study of the intraoperative requirements of fentanyl administered by the intravenous or epidural routes during open colon surgery. METHODS: Thirty patients were randomized to receive intraoperative analgesia with boluses of fentanyl administered by either the epidural or intravenous route (2 µg/kg). The first fentanyl bolus was administered 10 min before incision, and repeated boluses were given when mean arterial pressure or heart rate increased more than 20% over basal values. General anaesthesia was maintained with a propofol infusion. Intraoperative fentanyl and propofol requirements, time to awakening, time to analgesia request, and incidence of adverse effects were recorded. RESULTS: Median [interquartile range (range)] fentanyl requirements in the epidural and intravenous groups were 0.81 [0.65 (0.47-2.61)] and 2.5 [1.08 (1.07-4.85)] µg/kg/h, respectively (P < 0.001). The epidural group had a shorter time to awakening, with a median of 8 min [4.5 (3-18)] compared with 20 min [12.5 (7-34)] for the intravenous group (P < 0.001). There were no significant differences in propofol requirements. The time to analgesia request was also delayed in the epidural group, with a median of 5 h [5.5 (1-16)] vs. 2 h [1 (1-5)] when fentanyl was administered intravenously (P < 0.001). The incidence of adverse effects was similar in both groups. CONCLUSION: During major abdominal surgery, epidural administration requires lower doses of intraoperative fentanyl when compared with the intravenous route. Epidural fentanyl also facilitates early awakening and residual analgesia without increasing adverse events.

Effects of surgery and/or remifentanil administration on the expression of pERK1/2, c-Fos and dynorphin in the dorsal root ganglia in mice.

Tissue injury and/or opioids induce plastic changes in the spinal cord resulting in pain hypersensitivity; the contribution of the dorsal root ganglia (DRG) is poorly understood. We evaluated DRG phenotypic changes induced by surgery and/or remifentanil in a mice model of postoperative pain using as neuronal markers ERK1/2 and c-Fos; prodynorphin mRNA and dynorphin levels were also determined. We hypothesized that a correlation between nociception and DRG reactivity would occur. Surgery and/or remifentanil induced mechanical hypersensitivity, correlated with ERK1/2 phosphorylation and c-Fos expression in the DRG; changes were greater in the remifentanil + incision group and still present on day 14 (p < 0.01 vs. control). Intrathecal PD98059 (ERK1/2 inhibitor) partially reversed the mechanical hypersensitivity (44%, p < 0.05) observed in the remifentanil + incision group. In this group, significant increases in prodynorphin mRNA (at 2, 7, and 14 days, p < 0.01) roughly coincided with increases in dynorphin (days 2 and 14, p < 0.001) in the DRG. Remifentanil or incision (alone) also induced an up-regulation in prodynorphin mRNA expression on days 7 and 14 (p < 0.01, p < 0.05, respectively), partially correlating with dynorphin levels. On day 21, all molecular changes returned to control levels in all experimental conditions, concurring with the complete recovery of nociceptive thresholds. Surgery and/or remifentanil induce up-regulation of c-Fos and pERK in the DRG, approximately correlating with nociceptive behavior, also associated with an increased expression of prodynorphin/dynorphin. These changes support the role of the DRG in the development and maintenance of pain hypersensitivity after surgery. The findings could contribute to the development of new therapeutic agents focused on peripheral targets.

Effects of tramadol and dexketoprofen on analgesia and gastrointestinal transit in mice.

The purpose of the present study was to evaluate the nature of the antinociceptive interaction among dexketoprofen (DEX), a mixed inhibitor of the cyclo-oxygenases, and tramadol (TRAM), a weak opioid with monoaminergic activity that inhibits norepinephrine and serotonin re-uptake. We assessed antinociception in the acetic acid writhing test, the tail flick and the formalin (FT) tests, and gastrointestinal transit (GIT) after the administration of a charcoal meal. The analysis of the interaction was carried out using isobolograms and interaction indexes or the fixed-dose method GIT. The administration of DEX or TRAM individually induced dose-dependent antinociception in all the algesiometric tests. In the three tests, TRAM was between 5.2 (FT, phase I) and 35 times (FT, Phase II) more potent than DEX. When testing combinations at different potency ratios (1 : 1, 1 : 3, 3 : 1), we could demonstrate synergy in all algesiometric tests, only when drugs were combined in a 1 : 1 proportion. Interestingly, the proportion of the drugs in the combination could change the type of interaction from synergy to antagonism. On the inhibition of GIT, a dose-related inhibition was established for TRAM, but not for DEX. Using a fixed-dose protocol, we could demonstrate antagonism between DEX and TRAM on the inhibition of GIT. The results of the present study suggest that a combination of DEX and TRAM in a 1 : 1 proportion could be adequate to use in future clinical trials in humans.

A procedure-specific systematic review and consensus recommendations for postoperative analgesia following laparoscopic cholecystectomy.

BACKGROUND: Laparoscopic cholecystectomy has advantages over the open procedure for postoperative pain. However, a systematic review of postoperative pain management in this procedure has not been conducted. METHODS: A systematic review was conducted according to the guidelines of the Cochrane Collaboration. Randomized studies examining the effect of medical or surgical interventions on linear pain scores in patients undergoing laparoscopic cholecystectomy were included. Qualitative and quantitative analyses were performed. Recommendations for patient care were derived from review of these data, evidence from other relevant procedures, and clinical practice observations collated by the Delphi method among the authors. RESULTS: Sixty-nine randomized trials were included and 77 reports were excluded. Recommendations are provided for preoperative analgesia, anesthetic and operative techniques, and intraoperative and postoperative analgesia. CONCLUSIONS: A step-up approach to the management of postoperative pain following laparoscopic cholecystectomy is recommended. This approach has been designed to provide adequate analgesia while minimizing exposure to adverse events.

Low- vs high-dose almitrine combined with nitric oxide to prevent hypoxia during open-chest one-lung ventilation.

BACKGROUND: Almitrine combined with inhaled nitric oxide (NO) can prevent hypoxia during one-lung ventilation (OLV). The optimal dose of almitrine that would provide therapeutic advantage with few side-effects during open-chest OLV has not been established. METHODS: Forty-two patients undergoing thoracotomy were randomly allocated to three groups: placebo, almitrine 4 microg kg(-1) min(-1) and inhaled NO 10 p.p.m. (ALM4+NO), and almitrine 16 microg kg(-1) min(-1) and inhaled NO 10 p.p.m. (ALM16+NO). Gas exchange, haemodynamic and respiratory variables and plasma concentrations of almitrine and lactate were monitored. Measurements were obtained with the patient awake (baseline), after induction of anaesthesia with two-lung ventilation (control 2LV), 20 min after treatment (2LV+T), and then at 10, 20 and 30 min of OLV (OLV10', OLV20' and OLV30') with FI(O2)1. RESULTS: In the placebo group, OLV impaired Pa(O2) and increased pulmonary shunt [16 (SD 7) kPa and 42 (10)% respectively]. These improved with ALM4+NO [26 (10) kPa and 31 (7)%; P<0.001]. ALM16+NO further improved PaO2) to 36 (13) kPa (P<0.0001) but gave no improvement in the shunt. Mean pulmonary artery pressure was similar in the placebo and ALM4+NO groups [20 (4) vs 23 (5) mm Hg], whereas it was increased in the ALM16+NO group to 28 (8) mm Hg (P<0.01). Plasma concentrations of almitrine and lactate were unaltered by the treatments. CONCLUSIONS: Low-dose almitrine (4 microg kg(-1) min(-1)) together with inhaled NO significantly improves oxygenation during open-chest OLV, without modifying pulmonary haemodynamics. An increased dose of almitrine (16 microg kg(-1) min(-1)) with inhaled NO further improves arterial oxygenation, but also increases mean pulmonary artery pressure.

Comparative assessment of the effects of alfentanil, esmolol or clonidine when used as adjuvants during induction of general anaesthesia.

BACKGROUND AND OBJECTIVE: This randomized, double-blinded, prospective study compared the effects of clonidine, esmolol or alfentanil on the level of hypnosis and haemodynamic responses to intravenous induction of anaesthesia and endotracheal intubation. METHODS: Forty-five patients scheduled for elective surgery were allotted to one of three groups. They were given either alfentanil 3 microg kg(-1) min(-1) (n = 15); esmolol 1 mg kg(-1) min(-1) (n = 16) or clonidine 3 microg kg(-1) (n = 14) as a 10 min infusion. The infusions of alfentanil and esmolol, but not of clonidine, were maintained during endotracheal intubation. Anaesthesia was induced with midazolam (2 mg) and thiopental as required to suppress the eyelash reflex. Atracurium (0.5 mg kg(-1)) was given to produce neuromuscular block. Mean arterial pressure, heart rate, and bispectral index were recorded on arrival (baseline), after study drug infusion, after injecting midazolam and thiopental, as well as after endotracheal intubation. ANOVA and chi2-test were used for analysis. RESULTS: Blood pressure, heart rate and the bispectral index were unaltered by the study drugs, but thiopental requirements were reduced by alfentanil and clonidine (P < 0.014). Mean arterial pressure values (mean +/- standard error of mean) in the alfentanil, esmolol and clonidine groups were: baseline: 107.8 +/- 3.8; 106.6 +/- 3.9; 103.4 +/- 3.7 mmHg; after thiopental: 74.0 +/- 4.2; 85.6 +/- 4.3; 94.2 +/- 4.1 mmHg and after endotracheal intubation: 91.7 +/- 5.3; 114.1 +/- 6.9; 123.6 +/- 5.6 mmHg, respectively (two-way ANOVA, P < 0.001). Mean arterial pressure changed significantly after intubation from baseline (P < 0.001) after alfentanil (-15%) and clonidine (+20%) but not after esmolol (+7%), while the changes between pre- and postintubation values were similar in all groups (24-33% increase). The bispectral index indicated that all patients had an adequate level of hypnosis, but the variability was higher in the esmolol group (P < 0.002). CONCLUSIONS: None of the study drugs blocked the increase in mean arterial pressure induced by endotracheal intubation, but esmolol provided better overall haemodynamic stability. All groups had an adequate level of hypnosis.

Simple method to determine the size of the laryngeal mask airway in children.

BACKGROUND AND OBJECTIVE: The size of the laryngeal mask airway in children is determined by the patient's weight. However, in some instances an alternative method may be wanted. The aim was to search for a new method that would be easy to perform at the bedside. METHODS: The size of the laryngeal mask airway was determined in 183 children by choosing the laryngeal mask that best matched the combined widths of the patient's index, middle and ring fingers. The results were compared with the standard method recommended by the manufacturer's weight-related guidelines. The patients were classified in different groups depending on the laryngeal mask airway sizes determined by both methods. A kappa coefficient evaluated the agreement between both techniques. RESULTS: The kappa coefficient was 0.81, showing an 'excellent agreement' between both methods. The size was the same for both methods in 142 children (78%). The disagreement between both techniques was only of one size in the remaining 41 patients (22%). In such patients, the weight was a borderline value that would indicate a change in the size of the laryngeal mask airway using the classic method. CONCLUSIONS: This new approach is of valid and practical use in children, particularly as an alternative in those situations where the patient's weight is unknown, such as in emergency situations or in those borderline instances where an alternative measurement would be useful.

Inflammation enhances mu-opioid receptor transcription and expression in mice intestine.

Opioid receptors (ORs) and their mRNA are present in the central and peripheral nervous systems of mammals and in different peripheral tissues, including the gut. Using a model of croton oil-induced (CO) intestinal inflammation in mice, we have shown a 6-fold increase in the potency of the antitransit and antisecretory effects of mu-OR agonists, mediated by peripheral ORs. We postulate that the enhanced effects are mediated by an increase in the expression of intestinal OR. We used jejunum (stripped of the mucosal layer) from mice with CO-induced intestinal inflammation and, as control subjects, saline-treated animals (SS). We evaluated the quantity of mu-OR mRNA determined by a competitive reverse-transcriptase polymerase chain reaction; the levels of mu-OR protein by Western blot immunoassay, and the localization and number of cells expressing mu-OR using immunohistochemistry. The results show a significant increase of mu-OR mRNA (7.7-fold) and receptor protein (3-fold) during intestinal inflammation. Inflammation also induced a 64.3% increase in the number of neurons expressing mu-OR immunoreactivity in the myenteric plexus but not in the submucosal plexus. Our results show that intestinal inflammation enhances the transcription and translation of mu-OR mRNA, thus explaining the increased potency of mu-opioids during inflammation.

Management of postoperative pain in Spain.

BACKGROUND: The management of postoperative pain is suboptimal world-wide. This survey was carried out to determine current management in Spanish hospitals. METHODS: Spanish hospitals were divided into two groups: <200 beds (n=346) and >200 beds (n=186). A structured questionnaire was mailed to the heads of the anaesthesiology services of a random sample of 150 of hospitals of <200 beds, and all larger hospitals. RESULTS: Only 19% of hospitals with <200 beds responded and further analysis of this group was not possible; 53% of hospitals with >200 beds responded. In this sample (>200 beds), 45% of patients receive information on postoperative pain, given by the anaesthesiologist in 78% of cases. Over 70% of the hospitals do not have an acute pain unit and responsibility for postoperative pain lies with the anaesthesiology team in 71%, with the surgical team in 40% and with nursing in 33%, with an overlap of pain caretakers. Pain is measured on the surgical wards in 36% and recorded with the vital signs in 34%. On post-anaesthetic recovery wards, analgesia is most frequently given by intravenous (83%), epidural (78%) and continuous intravenous (66%) administration. On surgical wards, the most frequent routes that are available are epidural (72%), intravenous (69%) and intramuscular (58%). Only 28% of the anaesthesiology services are satisfied with the pain treatment carried out in their hospitals. No significant differences on postoperative pain management were observed between teaching and non-teaching institutions. CONCLUSION: The survey shows that the management of postoperative pain in hospitals with >200 beds in Spain is suboptimal and this is associated with dissatisfaction among many anaesthesiologists.

Intestinal inflammation enhances the inhibitory effects of opioids on intestinal permeability in mice.

The inhibitory effects of central and peripherally acting opioid agonists on intestinal permeability (PER) were evaluated during acute and chronic intestinal inflammation in mice. Inflammation was induced by the intragastric (p.o.) administration of one (acute) or two (chronic) doses of croton oil (CO), whereas controls received saline (SS). Intestinal PER was assessed by the blood-to-lumen transfer of 51Cr-ethylenediaminetetraacetate (51Cr-EDTA). CO significantly increased PER during acute (2.5 times) and chronic (3.2 times) inflammation. The potency of s.c. morphine-inhibiting PER was enhanced 3.8 and 8.7 times in acute and chronic CO, whereas that of s.c. fentanyl was increased 2.0 and 4.3 times, respectively, compared with SS. Similarly, s.c. [D-Pen(2,5)]-enkephalin was 4.7 and 11.1 times more potent during acute and chronic CO, and the E(max) values of the dose-response curves increased 35% during inflammation. The potency of s.c. U50,488H was 5.6 (acute) and 6.7 times (chronic) greater compared with SS. All effects were reversed by specific antagonists. The i.p. administration of beta-funaltrexamine differentially blocked morphine effects during acute and chronic CO, suggesting that the effects are mediated by different populations of functional mu-opioid receptors (OR). The increase in potencies of s.c. PL017 and ICI-204,448 during CO were comparable to those observed with fentanyl and U50,488H and their effects were antagonized by s.c. naloxone methiodide. Moreover, the potency of the agonists during inflammation was unaltered when administered i.c.v. The results show that intestinal inflammation enhances the effects of delta- > mu- > kappa-opioid agonists on PER by activation of peripheral OR.

Antisense oligodeoxynucleotides to mu- and delta-opioid receptor mRNA block the enhanced effects of opioids during intestinal inflammation.

Intestinal inflammation enhances the inhibitory effects of mu- and delta-opioids in the gut, possibly related to an increased receptor expression. We evaluated the effects of opioids after intraperitoneal administration of antisense oligodeoxynucleotides to mu- and/or delta-opioid receptor mRNA. Inflammation was induced in mice by intragastric administration of croton oil; gastrointestinal transit was assessed with charcoal and permeability with [51Cr]etylenediaminetetraacetate ([51Cr]EDTA). Baseline values were unaltered after antisense oligodeoxynucleotides. In controls, antisense oligodeoxynucleotides to mu-opioid receptor mRNA decreased the antitransit effects of morphine (27%) and [N-MePhe3D-Pro4]morphiceptin (PL017) (26%), and the reduction was significantly greater during inflammation (50% and 47%). A similar effect was observed on permeability (control: 41-21% decrease; inflamed: 66-45%). In both assays, antisense oligodeoxynucleotides to delta-opioid receptor mRNA also reduced the effects of [D-Pen2,5]enkephalin (DPDPE) in a higher percentage during inflammation (43-32% controls, 60-49% inflamed). We show that antisense oligodeoxynucleotides to mu- and/or delta-opioid receptor mRNA are efficiently blocking the intestinal effects of opioids during inflammation, suggesting that an increased transcription of these receptors in the gut mediates the enhanced effects of opioids during inflammation.

Use of intravenous patient-controlled analgesia for the documentation of synergy between tramadol and metamizol.

The quantification of the synergistic interactions (beneficial and adverse) of analgesic drug combinations in humans has been elusive. We propose a new procedure based on analgesic requirements (i.v.-PCA) and pain intensity (VAS-PI). One hundred and one post-hysterectomy patients received at the time of analgesia request (TAR) tramadol (100 mg, group I) or metamizol (1.2 g, group II) alone, or combined in 1:1 (III), 1:0.3 (IV) or 1:3 ratio (V). After 15 min, they received the same treatment by PCA. VAS-PI, analgesic consumption and adverse effects were assessed at TAR, and periodically for 24 h. Data were analysed using interaction indexes and isobolograms. All treatments produced equivalent VAS-PI, per cent efficacy and adverse effects. When drugs were combined in a 1:1 ratio, synergy was present for the analgesic and adverse effects; all other treatments were additive.

Injury to the abdominal aorta during laparoscopic surgery: an unusual presentation.

IMPLICATIONS: Laparoscopic cholecystectomy is a very common surgical procedure, and vascular injuries account for one third of major complications during this surgery. We describe an unusual presentation of an abdominal aorta injury.

Intestinal inflammation and morphine tolerance alter the interaction between morphine and clonidine on gastrointestinal transit in mice.

BACKGROUND: Morphine and clonidine show synergy or antagonism inhibiting gastrointestinal transit depending on their proportion and level of effect. Their interaction during morphine tolerance and intestinal inflammation were assessed. METHODS: Gastrointestinal transit in mice was evaluated with charcoal and antitransit effects expressed as percent mean values +/- SEM. Tolerance was induced with a morphine pellet (75 mg) implanted for 72 h, and inflammation with intragastric croton oil. Dose-response curves for morphine and clonidine alone and combined at a 1:1 potency ratio were obtained, and doses producing a 50% and 60% inhibition were calculated (ED50, ED60). Interaction was established by isobolograms, interaction indexes, and analysis of variance. RESULTS: In naive and tolerant mice, the combination induced linear dose-response curves up to the ED60 and then reached a plateau. In naive mice, ED50 values were as follows: morphine 1.52 +/- 0.15 mg/kg, clonidine 0.09 +/- 0.008 mg/kg, and combined 0.506 +/- 0.084 mg/kg (0.478 +/- 0.08 mg/kg morphine plus 0.028 +/- 0.004 mg/kg clonidine). During tolerance, ED50 values were as follows: morphine 9.73 +/- 0.8 mg/kg, clonidine 0.09 +/- 0.007 mg/kg, combination 0.131 +/- 0.09 mg/kg (morphine 0. 13 +/- 0.09 mg/kg plus clonidine 0.0013 +/- 0.0005 mg/kg). In both groups, the interaction was synergistic up to the ED60 and antagonistic thereafter; synergy was enhanced during tolerance. During inflammation, ED50 values were as follows: morphine 0.17 +/- 0.04 mg/kg, clonidine 0.015 +/- 0.006 mg/kg, combined 0.62 +/- 0.04 mg/kg (morphine 0.568 +/- 0.04 mg/kg plus clonidine 0.052 +/- 0.004 mg/kg); thus, potencies of morphine and clonidine increased 9.3 and 7.1 times, while the combination remained unaltered. Moreover, inflammation transformed synergy into antagonism. CONCLUSIONS: The interaction between morphine and clonidine was significantly altered during tolerance and inflammation. During tolerance, synergy was present up to 60% effect and then became antagonistic. Inflammation converted synergy to antagonism. A common pathway in signal transduction could partially explain the results.

Antinociceptive/anti-edema effects of liposomal morphine during acute inflammation of the rat paw.

We evaluated the anti-edema/antinociceptive effects of subcutaneous free and liposomal morphine in rats with carrageenan-induced inflammation of the paw. We assessed antinociception by the paw pressure test and edema by plethysmography. Unilamellar liposomes (150-200 nm) with 0.3% morphine hydrochloride were used; encapsulation significantly reduced the rate for release of morphine in vitro. During inflammation, the antinociceptive potency of free, but not liposomal morphine increased 2.5 times; moreover, duration of the effects was prolonged by encapsulation (p < 0.001). The anti-edema effects of liposomal morphine were more pronounced (p < 0. 001) and of longer duration (p < 0.05). All the effects were reversed by naloxone. The results show that morphine encapsulation enhances the anti-edema effects and prolongs antinociception.

Effects of mu-opioid receptor agonists on intestinal secretion and permeability during acute intestinal inflammation in mice.

We evaluated and compared the effects of mu-opioid receptor agonists on mucosal fluid transport and permeability, during acute intestinal inflammation. We hypothesized that inflammation would sensitize mu-opioid receptors in the submucosal plexus and/or enterocytes enhancing the effects of mu-opioid receptor agonists. Inflammation was induced by intragastric administration of croton oil, whereas controls received saline. Fluid transport was assessed by enteropooling, and intestinal permeability by blood-to-lumen passage of [51Cr] etylenediaminetetraacetate ([51Cr] EDTA). Intestinal inflammation induced a significant increase in enteropooling (1.9 times) and permeability (2.5 times). In saline- and croton oil-treated animals, mu-opioid receptor agonists produced dose-related inhibitions of enteropooling and intestinal permeability. During inflammation, the potency of morphine increased 4.8 and 3.7 times, inhibiting enteropooling and intestinal permeability, respectively; the potencies of fentanyl and PL017 similarly increased by approximately three (enteropooling) and two times (permeability) in croton oil animals. All effects were reversed by naloxone and naloxone methiodide. The results show that inflammation increases the inhibitory potency of mu-opioid receptor agonists on secretion and permeability, suggesting a sensitization of peripheral mu-opioid receptors.

Antibodies and antisense oligodeoxynucleotides to mu-opioid receptors, selectively block the effects of mu-opioid agonists on intestinal transit and permeability in mice.

1. We have studied the effects of mu and delta opioids on intestinal function (permeability, PER; gastrointestinal transit, GIT), and their antagonism after the intracerebroventricular (i.c.v.) administration of specific antibodies (ABs) or antisense oligodeoxynucleotides (ODN) to mu-receptors (OR). Central versus peripheral site/s of action of subcutaneous (s.c.) mu-opioids, were also assessed. 2. Male Swiss CD-1 mice were used. GIT was measured with charcoal and PER by the passage of 51Cr-EDTA from blood to lumen. 3. Morphine and fentanyl (i.c.v. and s.c.) inhibited GIT and PER in a dose-related manner; they were more potent by i.c.v. route, both on GIT and PER (70 and 17 times for morphine and fentanyl). They also had a greater effect on GIT than PER (4.3 and 1.6 times). DPDPE had a lower potency than mu-agonists in all experiments, and no dose-response could be obtained after s.c. administration on GIT. 4. Pretreatment with i.c.v. ABs (24 h) or antisense ODN (5 days), decreased the effects (GIT and PER) of i.c.v. morphine and fentanyl, while those of DPDPE remained unchanged. The ABs did not alter the peripheral effects of mu-opioids. 5. The results show that (i.c.v. or s.c.) mu opioids produce dose-related inhibitions of PER and GIT, being more potent by the i.c.v. route. Delta-opioids had a greater effect on PER than GIT, while the opposite occurred for mu-agonists. Pretreatment with ABs or ODN to mu-OR, blocked the central effects of mu (but not delta) agonists on GIT and PER.