Timing of intravenous patient-controlled remifentanil bolus during early labour.

BACKGROUND: Remifentanil labour analgesia is superior to nitrous oxide but less potent than epidural analgesia. The short onset and offset times of effect suggest that the timing of the bolus in the contraction cycle could have importance. We hypothesised that administering a remifentanil bolus during contraction pause would improve analgesia in early labour. METHODS: With permission from the ethical committee and the National Authority of Medicines, 50 parturients with uncomplicated singleton pregnancies and informed consent participated in a double blind cross-over study. Intravenous doses of 0.4 µg/kg remifentanil with 1-min infusion times were used during two study periods lasting six to eight contractions. Remifentanil and saline syringes were attached to two patient-controlled devices, one of which administered the bolus immediately after a trigger and the other targeted to start 140 s before the next contraction. The parturients assessed contraction pain, pain relief, sedation and nausea. Oxygen saturation (SaO(2)) pulse and blood pressure were recorded. SaO(2)<95% was the indication for oxygen supplement. RESULTS: Forty-one parturients were included in the analyses. Because of the period effect, pain and pain relief scores were analysed separately for each of the study periods. The mean pain and pain relief scores were similar during the two different dosing regimens. Side effects, the need for supplemental oxygen, SaO(2) and haemodynamics were similar. In a subgroup with long and regular contractions, however, delayed boluses were associated with lower pain scores. CONCLUSIONS: Administering a remifentanil bolus during the uterine contraction pause does not improve pain relief.

A comparison of paracervical block with single-shot spinal for labour analgesia in multiparous women: a randomised controlled trial.

BACKGROUND: Epidural and spinal analgesia may be contraindicated or unavailable in labour. This randomised controlled study examined the suitability of paracervical block as an alternative method of labour analgesia. METHODS: Multiparous women in labour were randomised to receive either paracervical block or single-shot spinal analgesia. Pain was quantified using a numerical rating scale. Subsequent analgesia, progress of labour, and mode of delivery were noted. Fetal heart rate patterns were reviewed. Apgar scores and umbilical artery pH measurements were collected. Parturients' satisfaction and willingness to have the same method of labour analgesia again were recorded. RESULTS: 122 parturients were randomised with data available on 104. Median pain scores decreased significantly in both groups; this was greater with single-shot spinal analgesia (difference between means 2.7; 95% CI 1.9-3.5; P(g)<0.001). Parturients receiving paracervical block received subsequent analgesia more often (23/56 vs. 3/48, P<0.001). Progress of labour, instrumental delivery rates, detected abnormal decelerations in cardiotocography and neonatal outcome were similar between groups. Shivering (P<0.04) and pruritus (P<0.001) were more common with single-shot spinal analgesia. Parturients in the paracervical block group were less satisfied (median 7.0, IQR 3.0-8.0 vs. median 9.0, IQR 8.0-10.0; P<0.001) and less willing (28/55 vs. 39/48, P=0.002) to have the same labour analgesia again. CONCLUSIONS: Paracervical block was less effective than single-shot spinal analgesia. Both methods were associated with a low incidence of fetal bradycardia but maternal side effects were more common with single-shot spinal analgesia.

The effect of complement factor H Y402H polymorphism on the outcome of photodynamic therapy in age-related macular degeneration.

PURPOSE: Photodynamic therapy (PDT) has been widely used in the treatment of age-related macular degeneration (AMD). The complement cascade has an important role in the tissue reactions occurring after PDT. The Y402H polymorphism of the complement factor H (CFH) gene has been identified as a risk factor for AMD. Since CFH is central in the regulation of the complement system the authors wanted to analyze whether the CFH Y402H polymorphism modifies the PDT outcome in AMD. METHODS: A total of 88 patients having been treated with PDT and without further scheduled PDT sessions were analyzed. Depending on the situation at their final PDT session the patients were classified retrospectively as PDT-responders or PDT-nonresponders. All patients were genotyped for the CFH Y402H polymorphism. RESULTS: The proportion of PDT-responders was 18/26 (69.2%) in patients homozygous for the CFH Y402H risk allele, 34/50 (68.0%) in heterozygous, and 7/12 (58.3%) in patients with the normal genotype (p=0.520). The median number of PDT treatments of the PDT-responders was three for all the genotypes. CONCLUSIONS: The dysfunction of the CFH related to the risk of AMD and caused by the Y402H polymorphism does not modify the outcome of PDT. Genotyping for CFH Y402H cannot be used to select patients for this treatment.

Serum IL-1beta levels are associated with the presence of erosions in recent onset rheumatoid arthritis.

OBJECTIVE: To study interleukin (IL)-1beta levels in recent onset RA patients treated either with combination DMARD therapy (sulfasalazine, methotrexate, hydroxychloroquine) or a single DMARD therapy. METHODS: Serum IL-1beta levels were measured before the treatment and 6 months after the institution of either single or combination DMARD therapy using a high sensitivity ELISA method. Radiographic evaluation of the hands and feet was performed at 0 and 24 months. RESULTS: Significant correlations (r = 0.28, 95% CI 0.10-0.45) were found between IL-1beta levels measured at 0 and 6 months. The IL-1beta levels at 0 months correlated significantly (r = 0.23, 95% CI 0.03-0.4, p= 0.021) with the baseline number of eroded joints at 0 months but not with radiographic joint damage at 24 months. The baseline level of IL-1beta was a better indicator for the presence of eroded joints than the baseline level of serum CRP. No significant changes in IL-1beta levels were observed during the first 6 months of anti-rheumatic treatment in either group. No statistically significant difference between IL-1beta levels in the patients with or without the shared epitope could be observed. CONCLUSIONS: The serum IL-1beta level is significantly associated with the presence of erosions at the onset of RA but its predictive value is attenuated or lost when single or combination DMARD medication is instituted. Measuring IL-1beta at the time of diagnosis in a single patient cannot be used to estimate the erosive nature of the disease or the prognosis.

Incisional and epidural analgesia after caesarean delivery: a prospective, placebo-controlled, randomised clinical study.

BACKGROUND: This study evaluated efficacy, safety and patient satisfaction with incisional analgesia with a subfascial catheter compared to epidural analgesia for pain relief following caesarean section. METHODS: Forty patients were randomised after elective caesarean section to receive either intermittent 10-mL boluses of 0.125% levobupivacaine into the epidural space and physiologic saline into the surgical wound or intermittent 10-mL boluses of 0.25% levobupivacaine into the wound and epidural saline with a repeated 10-dose regimen. Analgesic efficacy was evaluated by numerical pain scores (0-10, 0=no pain, 10=worst pain) and based on the consumption of supplemental opioid. Side effects, patient satisfaction and plasma concentrations of levobupivacaine were recorded. RESULTS: In the epidural group average pain scores were lower (1.8 vs. 3, P=0.006) and the consumption of local anaesthetic (29 mL vs. 38 mL, P=0.01) was smaller during the first four postoperative hours, after which both groups had pain scores of 3 or less at rest. All parturients were able to walk after the 24-h study period. The total consumption of rescue opioid oxycodone (32 vs. 37 mg, P=0.6) during the whole 72-h study period was low in both study groups. Side effects were mild and rare. Satisfaction scores were equally high in the two groups. Peak plasma concentrations of levobupivacaine were below the toxic range. CONCLUSION: Incisional local analgesia via a subfascial catheter provided satisfactory pain relief with patient satisfaction comparable to that seen with epidural analgesia. This technique may be a good alternative to the more invasive epidural technique following caesarean section as a component of multimodal pain management.

Serum levels of amidated gastrin-17 and pepsinogen I in atrophic gastritis: an observational case-control study.

BACKGROUND: Helicobacter pylori infection is often diagnosed with non-endoscopic methods, such as serology or breath or antigen stool tests. These tests provide information on the presence or absence of the H. pylori gastritis only. We investigated whether atrophic gastritis can be diagnosed and typed non-endoscopically if the serum levels of pepsinogen I (S-PGI) and gastrin-17 (S-G-17) are assayed in connection with H. pylori testing. METHODS: The present investigation is an observational case-control study comprising 100 selected dyspeptic outpatients with (cases) or without (controls) advanced (moderate or severe) atrophic gastritis. Before the blood tests, all patients underwent a diagnostic gastroscopy with multiple biopsies. The series of cases includes 56 patients. Eight had an advanced antrum limited atrophic gastritis, 13 had resected antrum (in two of whom the corpus mucosa in the stump was atrophic), and 30 had corpus-limited atrophic gastritis. Four patients had an advanced atrophic gastritis in both the antrum and corpus (multifocal atrophic gastritis), and the whole stomach was removed in one patient. Twenty of the 44 controls had a non-atrophic H. pylori gastritis. Both the antrum and corpus were normal and healthy in 24 patients. The S-PGI and S-G-17 were determined with EIA methods using monoclonal antibodies to PGI and amidated G-17. Postprandial S-G-17 (S-G-17prand) was measured 20 min after a protein-rich drink. The H. pylori antibodies were assayed with a polyclonal EIA method. RESULTS: A low S-PGI (<25 microg/l; an empirical cut-off with best discrimination) was found in 31 of 37 patients (84%) with and in 3 of 63 patients (5%) without corpus atrophy in the biopsy specimens. A low S-G-17prand (<5 pmol/l) was found in all 8 patients with H. pylori-associated antral atrophy and in 11 of 14 patients (79%) with resected antrum but in 3 of 20 control patients (15%) with H. pylori-related non-atrophic gastritis. Median and mean values of both S-G-17prand and S-PGI decreased with increasing grade of antral and corpus atrophy, respectively. Among all patients with atrophic gastritis (multifocal atrophic gastritis, or atrophic gastritis limited to antrum or corpus) or resected stomach, 50 of 56 patients (89%; Cl 95%: 81%-97%) had a low S-PGI and/or a low S-G-17prand with positive H. pylori serology. Such low values werc found in 3 of the 44 control patients (7%; CI 95%: 0%-14%). CONCLUSIONS: Low serum levels of G-17prand and PGI are conceivable biomarkers of atrophic antral and corpus gastritis, respectively. A low S-G-17prand is a sign of the multifocal or antrum-limited atrophic gastritis in patients infected with H. pylori.

Improving the quality of postoperative pain relief.

A review of the literature shows a constant need to improve the quality of postoperative pain management. The objective of this study was to decrease the intensity and variation of postoperative pain by developing a nurse-based pain service on the ward. An acute pain nurse was appointed and an educational programme with detailed algorithms was started. Regular pain intensity measurements were implemented. Postoperative pain intensity, treatments and side-effects were assessed both before and after the introduction of the new system in 400 patients divided into two consecutive groups of equal size. The number of patients with inadequately treated pain (actual pain > 3/10) dropped by 64% after major gynaecological surgery (25 vs 9%, 95% CI for differences 7-24%; p<0.001 for pain scores). On an average, inadequate pain relief (retrospective average pain > 3/10) on the first postoperative day was more frequent on the ward before than after the reform (47 vs. 21%; 95% CI for differences 15-35%; p<0.001 for pain scores). The incidence of side-effects was similar in both groups (p> 0.05). The intensity and variation of postoperative pain on the ward decreased by developing a nurse-based pain service with an acute pain nurse, an educational programme and regular pain intensity measurements.

Postoperative nausea and vomiting after sevoflurane with or without ondansetron compared with propofol in female patients undergoing breast surgery.

IMPLICATIONS: We studied 180 female patients undergoing breast surgery. The patients were randomly allocated to receive one of three anesthetic techniques. Compared with either propofol or sevoflurane alone, sevoflurane with ondansetron resulted in a decreased incidence of postoperative nausea and vomiting. Sevoflurane with ondansetron prophylaxis is a good alternative to propofol with respect to avoiding postoperative nausea and vomiting.

Retinal breaks before and after neodymium:YAG posterior capsulotomy.

PURPOSE: To determine the frequency of asymptomatic retinal breaks before and after neodymium:YAG (Nd:YAG) laser posterior capsulotomy. SETTING: Department of Ophthalmology, Helsinki University Central Hospital, Helsinki, Finland. METHODS: Of 350 consecutive patients referred for their first laser Nd:YAG posterior capsulotomy, 235 eligible eyes were enrolled and 220 eyes completed the study. A vitreoretinal surgeon looked for retinal breaks using binocular indirect ophthalmoscopy with scleral indentation. The eyes were examined 1 week before and 1 hour and 1 month after the posterior capsulotomy. The number, type, and location of retinal breaks were recorded. RESULTS: The median age of eligible patients who did not participate in the study was higher than that of enrolled patients (79.6 versus 74.4 years; P =.0005). The mean axial length, median time from cataract surgery, and the course of cataract surgery were comparable in both groups. Before the posterior capsulotomy, an untreated retinal break was diagnosed in 4 of the 235 eyes (1.7%; 95% confidence interval [CI] 0 to 4) scheduled for surgery and an undiagnosed retinal detachment was present in 2 additional eyes (0.9%; 95% CI 0 to 3). An asymptomatic retinal break was also present in 4 fellow eyes (1.7%, 95% CI 0 to 4). No new breaks developed during Nd:YAG posterior capsulotomy using a median total energy of 51 mJ (range 10 to 901 mJ) and a median number of 22 applications (range 4 to 341 applications) and resulting in an opening with a median largest diameter of 3.4 mm (range 2.0 to 4.6 mm). In 1 treated eye (0.4%; 95% CI 0 to 2), a new retinal break had developed by 1 month postoperatively. CONCLUSIONS: The observed 2.1% frequency of asymptomatic retinal breaks that had escaped the attention of the referring ophthalmologist or had developed by 1 month after Nd:YAG posterior capsulotomy can be contrasted with the 0.5% to 2.0% frequency of retinal detachment reported in the literature. However, it is not known which proportion of such asymptomatic breaks, if any, will progress to detachment after Nd:YAG laser posterior capsulotomy.

Comparison of ondansetron and tropisetron combined with droperidol for the prevention of emesis in women with a history of post-operative nausea and vomiting.

The anti-emetic efficacy of prophylactic ondansetron and tropisetron in combination with a low dose of droperidol in patients with high probability for post-operative nausea and vomiting undergoing gynaecological laparoscopy was compared. Patients were randomly allocated in a double-blind manner to receive either ondansetron 8 mg (n = 45) or tropisetron 5 mg (n = 43) at the end of surgery. A standardized general anaesthetic technique was used, including droperidol 0.75 mg. The incidence of nausea was 36% and 49% (P = 0.28), and vomiting occurred in 13% and 14% of the patients in the ondansetron and tropisetron groups, respectively. The onset time for rescue medication was significantly sooner after tropisetron than ondansetron (3 h 18 min vs. 6 h 25 min; P = 0.007). There were no statistically significant differences in efficacy between prophylactic ondansetron and tropisetron combined with droperidol in a high-risk population. However, ondansetron appeared to be more effective in preventing post-operative nausea and vomiting in the early hours after surgery compared with tropisetron.

Osteotomy site healing following mandibular sagittal split osteotomy and rigid fixation with polylactide biodegradable screws.

Follow-up of 47 patients, treated using mandibular bilateral sagittal split osteotomy and self-reinforced poly-L-lactide acid (SR-PLLA) screws for rigid internal fixation, is presented. The focus was on clinical and radiological osteotomy healing. The average follow-up time was 2.1 years (range 0.5-5 years). Clinical recovery and radiological osteotomy healing during follow-up were uneventful. Osteolytic changes were seen around the SR-PLLA screws in 27% of cases. The majority of the screw canals remained as radiolucent shadows without bony filling.

Retinal detachment in pseudophakic eyes with and without Nd:YAG laser posterior capsulotomy.

OBJECTIVE: This study aimed to determine whether the characteristics of pseudophakic retinal detachment (RD) differ in eyes with and without a neodymium (Nd):YAG laser capsulotomy to identify features that might help to treat capsulotomy patients better. DESIGN: The study design was a retrospective cohort study of 129 consecutive eyes, 61 with a capsulotomy and 68 control eyes with an intact posterior capsule, in which a primary rhegmatogenous RD developed after a strictly uncomplicated extracapsular cataract extraction and intraocular lens implantation. SETTING: The study was conducted in a primary referral center. MAIN OUTCOME MEASURES: Number, type, and location of retinal breaks and type and extent of RD were measured. RESULTS: Risk factors predisposing to RD did not differ between the capsulotomy and control group. The median interval from cataract surgery to RD was expectedly longer in the capsulotomy group (4.1 years vs. 1.5 years, P < 0.001). The mean number of retinal breaks was larger in the capsulotomy group (1.7 vs. 1.1, P = 0.05), especially in upper quadrants (82 of 103 vs. 48 of 77, P = 0.024). Capsulotomy eyes also had a trend to have more atrophic holes that caused detachment relative to horseshoe breaks (34 of 103 vs. 15 of 77, P = 0.062). The type and extent of RD were similar in the two groups. CONCLUSIONS: This pilot study suggests that atrophic holes, particularly in the superior quadrants, may lead to RD preferentially after posterior capsulotomy. Early identification and treatment of such breaks might decrease the number of pseudophakic detachments after capsulotomy.

Comparison of ondansetron and droperidol in the prevention of postoperative nausea and vomiting after laparoscopic surgery in women. A randomised, double-blind, placebo-controlled trial.

BACKGROUND: Women undergoing laparoscopic surgery are susceptible to postoperative nausea and vomiting (PONV). Ondansetron and droperidol are useful antiemetics. This study was designed to ascertain primarily the relative difference in efficacy of ondansetron and droperidol and secondarily between these drugs and placebo in the prevention of PONV after laparoscopic surgery. METHODS: The prophylactic antiemetic efficacy of ondansetron and droperidol was compared in a prospective, randomised, double-blind, placebo-controlled trial of 439 female inpatients scheduled for laparoscopic surgery. During induction of standardised general anaesthesia the patients received intravenously either ondansetron 8 mg (n=195), droperidol 1.25 mg (n=193) or placebo (n=51). The occurrence of nausea, vomiting, side-effects and the need for rescue antiemetic medication were recorded for 24 h postoperatively. RESULTS: The proportion of patients with nausea was 48%, 50% and 67% in the ondansetron, droperidol and placebo groups, respectively; with a significant difference when both ondansetron (P=0.02) and droperidol (P=0.04) were compared with placebo. Vomiting occurred in 18%, 26% and 37% of the patients in the three groups, respectively (P=0.05 between ondansetron and droperidol, P=0.004 between ondansetron and placebo, P=0.16 between droperidol and placebo). The proportion of patients given rescue medication was 34%, 28% and 49%, respectively (P=0.23 for ondansetron and droperidol, P=0.07 for ondansetron and placebo, P=0.007 for droperidol and placebo). During early recovery the patients treated with ondansetron were significantly more alert than after droperidol. Serious side-effects were not observed. Headache was significantly more common after ondansetron than after droperidol treatment. CONCLUSIONS: The efficacy of prophylactic ondansetron and droperidol in reducing postoperative nausea associated with laparoscopic surgery in female inpatients was similar, but ondansetron appeared to be slightly more efficient than droperidol in preventing vomiting. Ondansetron and droperidol were both significantly better than placebo in the prophylaxis of PONV.

The intensity of labor pain in grand multiparas.

BACKGROUND: The purpose of the current study was to investigate differences in pain experiences and requirements for pain control in parturients with different birth experience. Previous studies have shown that the labor pain experience is significantly less intense in multiparas than in primiparas and requirements for analgesic treatment have therefore been assumed to be minimal in this group. METHODS: The pain experience during labor was assessed by 70 consecutive grand multiparas (a minimum of five previous deliveries) compared to that of primiparous (n=70) and II-V parous women (n=70). Pain intensity was repeatedly assessed on a pain intensity scale (0-10) according to the progress of cervical dilatation at the first and second stage of labor. RESULTS: During the latent phase of cervical dilatation (0-3 cm) grand multiparas had a median pain score of 3 compared to 4 in II-V paras and 6 in primiparas (p<0.001). At the end of the first stage and during the second stage the intensity of pain in grand multiparas was significantly higher compared to that in primiparas (median scores 7 to 8 vs 6 to 7, p<0.05). Epidural blocks were administered to 40% of primiparas, 3% of II-V paras and to no grand multiparas (p<0.0001). Twenty-one percent of grand multiparas rated their pain as intolerable (pain scores 9-10) during the second stage compared to 10% of primiparas (p<0.05%). On the third day after delivery, 47% of grand multiparas regarded their analgesia insufficient. CONCLUSIONS: The majority of parturients, including grand multiparas, suffered from intense pain during labor. After delivery, a significant number of grand multiparas felt that they had received insufficient pain relief.

The effect of maternal obesity on labour and labour pain.

Consecutive obese (n = 53) and non-obese parturients (n = 609) were prospectively evaluated during labour to analyse the influence of maternal obesity on labour pain and outcome. Excessive pre-pregnancy weight was classified as a body mass index of 30 kg.m-2 or more. Pain intensity was measured using an 11 point visual scoring scale. Obese parturients had more complicated pregnancies (hypertension and diabetes) and their babies weighed significantly more (3865 g versus 3592 g, p < 0.001). These differences did not affect labour pain experience, or the duration or mode of delivery. Eighty-five percent of the obese parturients and 83% of the controls had high maximal pain scores during the first stage (> 7). Both groups received similar analgesia. More technical problems (p = 0.03) were experienced in establishing epidural analgesia for obese parturients, but this did not influence the success of pain treatment. After delivery, obese women were significantly more content with the pain relief received; only 12% vs 23% in the control group complained of poor pain control (p = 0.03). In this study, obesity and increased fetal size did not complicate labour or its outcome. Critical patient assessment should be emphasised, however, due to the physiological and medical problems present in obese parturients.

Paracervical block--a viable alternative for labor pain relief?

BACKGROUND: Two hundred and forty-eight consecutive deliveries with a 0.25% bupivacaine paracervical block (PCB) using a superficial injection technique were studied prospectively during the progress of labor in the three month study period. Parturients with signs of fetal distress were excluded. Pain level and the maternal, fetal and neonatal effects related to PCB were evaluated. METHODS: Pain intensity level was assessed on a visual scoring scale (0-10). The intrapartum fetal heart rate patterns were recorded and reviewed to correlate the incidence of paracervical fetal bradycardia. RESULTS: Parturients with PCB were provided a significant reduction (p < 0.001) in pain levels and the change in pain scores could be maintained until the beginning of the second stage. After the block, bradycardia patterns were noted in five (2.0%) fetuses, a fetal heart rate ranged from 60 to 105 beats per minute with an onset time of 2-20 minutes after the block and a duration of 4-12 minutes. All newborns in the bradycardia group were delivered vaginally without signs of asphyxia. CONCLUSIONS: PCB in selected parturients with a low-dose superficial technique has been shown to be an effective analgesic modality with minimal fetal and neonatal side-effects, but post-blockade fetal bradycardia cannot be wholly eliminated.

Maternal expectations and experiences of labour pain--options of 1091 Finnish parturients.

A prospective survey of 1091 Finnish parturients was conducted in order to ascertain mothers' expectations for labour pain relief, to measure the actual pain during all three stages of labour and to question their satisfaction and the adequacy of pain relief on the third day following delivery. Antenatal expectations for pain relief were surveyed. Mothers were questioned on pain levels in the delivery room and 3 days after giving birth. Pain levels were ascertained using a visual pain score method. Antenatally, 90% of all parturients anticipated a need for pain relief during labour. In the delivery room over 80% of all parturients described their pain as very severe to intolerable, only 4% of the multiparous had low pain scores (0-2). After pain treatment 50% of multiparous women still had pain scores from 8 to 10, which reflects a lack of effective pain relief. Dissatisfaction with the childbirth experience was very low, and was associated with instrumental deliveries, but not with the usage of analgesia. 51% of all parturients complained of inadequate pain relief during childbirth, which, in multiparous women, was significantly associated with the second stage of labour.

Parturients' assessment of water blocks, pethidine, nitrous oxide, paracervical and epidural blocks in labour.

This prospective study evaluated the extent to which the analgesic procedures available resulted in efficient pain relief in 833 Finnish parturients. Their pain ratings were recorded according to their own assessment of pain intensity during labour before and after pain management and according to their total pain experience recalled on the third day after delivery. Only regional blocks significantly diminished pain scores during labour and a striking decrease in pain level (P < 0.0001) was maintained until the second stage by epidural analgesia (EDA) alone. The EDA-group consisted mainly of the primiparous (71%) with induced (17%) and most painful deliveries. After delivery 60% of parturients in this group recalled labour pain as being very severe or intolerable. This was partly due to delays in receiving epidural blockade and partly due to the parturient's reticence in requesting analgesia. Pain relief was rated as excellent or good by 94% of the EDA group but by only 50% of the remaining patients.

Effect of a non-steroidal anti-inflammatory drug, diclofenac, on haemostasis in patients undergoing total hip replacement.

Haemostasis was studied in patients receiving diclofenac for postoperative pain relief. Intravenous diclofenac 75 mg over 60 min, followed first by an infusion of 5 mg/h for 15 h and then by 50 mg every 8 h orally was administered to 20 patients undergoing total hip replacement. Eighteen patients receiving a placebo infusion and dextropropoxyfen per os served as controls. The results showed no statistically significant differences between the groups in blood loss, bleeding time (IVY), partial activated thromboplastin time and prothrombin complex assay or in platelet count. The measurements were performed preoperatively, 3 h postoperatively and on the fourth and tenth postoperative days. Plasma concentrations were also determined in ten patients undergoing knee arthroscopy. An i.v. diclofenac infusion of 75 mg over a period of 15 min was administered either once (to half of the patients) or twice. The mean diclofenac concentrations were 28 +/- 5 nmol/ml (+/- s.d.) after 15 min and 36 +/- 12 nmol/ml after the second infusion. The bleeding time in the arthroscopy patients receiving one or two bolus infusions of 75 mg diclofenac remained at the control level. It is concluded that diclofenac given as an intravenous infusion of 75 mg in 60 min, then 5 mg/h for 15 h, followed by 50 mg every 8 h orally, is a safe as dextropropoxyfen for pain relief in patients undergoing major orthopaedic surgery as far as coagulation data are concerned.