Printer Friendly

Postoperative plasma paracetamol levels following oral or intravenous paracetamol administration: a double-blind randomised controlled trial.

Paracetamol has widespread perioperative use and a well-established safety profile. A standard dose of 1.0 g has a number needed to treat of 3.6 when used as a single dose for postoperative pain (1).

The optimal timing and method of administration of paracetamol in the perioperative period is unclear. It has previously been demonstrated that postoperative intravenous paracetamol provides faster onset of analgesia than a similar dose given orally (2) and, more recently, that only two-thirds of patients given an oral dose of 1.0 g paracetamol preoperatively achieved therapeutic plasma concentrations at any point compared to 96% in those given 1.0 g intravenously preoperatively (3).

Antipyretic plasma concentrations of paracetamol are widely quoted to lie between 66 and 132 [micro]mol/l (10 to 20 mg/l) (4). However the original data describing this range have not been published. In post-tonsillectomy pain, postoperative plasma levels of paracetamol above 70 [micro]mol/l resulted in satisfactory analgesia in 75% of children (5). Although the pharmacokinetics of paracetamol have been demonstrated to be similar in adults and children (6), studies in adults have failed to a consistent concentration response correlation (6,7). Hahn et al found that there was no increasing analgesic benefit above a theoretical maximal plasma concentration at t=0 of 14 mg/l (92.4 [micro]mol/l).

To our knowledge, however, there is no published literature comparing the two most common clinical options of preoperative oral paracetamol and intraoperative intravenous paracetamol. This is an important question, given that the intravenous preparation is significantly more expensive and its administration more complex.

We performed a double-blind randomised controlled trial comparing these two practices to test the hypothesis that intraoperative intravenous administration of paracetamol would be more effective in achieving potentially therapeutic plasma concentrations in the early postoperative period than the same dose given orally preoperatively. Secondary outcomes included postoperative pain scores, rescue analgesia requirements and duration of stay in the recovery room.



Thirty patients undergoing day case arthroscopy of the knee in a single surgical centre were included in the study. Patients were at least 18 years of age, American Society of Anesthesiologists physical status I to II. Patients were excluded if they had any contraindication to paracetamol, body mass index over 35 kg/[m.sup.2] or were unsuitable for laryngeal mask anaesthesia. The study was approved by the Upper South A Regional Ethics Committee and registered with the Australian New Zealand Clinical Trials Registry (ACTRN 12609000908235, registered 20/10/09). Written informed consent was obtained from all patients.

Randomisation and administration of paracetamol

Patients were enrolled by a study investigator and then randomised using a web-based allocation sequence ( to receive either 1.0 g oral paracetamol (Pharmacare immediate release 500 mg tablets) 3 to 60 minutes preoperatively or an identical placebo. All tablets were administered from sealed numbered envelopes and both the patients and the preoperative staff were blinded to the allocation group. Paracetamol was the only drug given preoperatively. Those patients who received the placebo oral tablets were then given 1.0 g intravenous paracetamol intraoperatively (Perfalgan, Bristol-Myers Squibb, New York, NY, USA). This was given as a 15-minute infusion with completion of the infusionas close as possible to the end of the surgical procedure. The infusion was administered by the attending anaesthetist (non-blinded) who took no further part in the study. A non-blinded study investigator was responsible for confidentially informing the anaesthetist which patients required intravenous administration of paracetamol.

Anaesthesia and surgery

Anaesthesia was induced with a combination of fentanyl, midazolam and propofol with doses at the anaesthetist's discretion. A laryngeal mask was inserted and anaesthesia maintained with sevoflurane in an air/oxygen mixture. Boluses of intravenous fentanyl were used for intraoperative analgesia with no other analgesics used. Antiemetics were used as required.

Patients were all planned for day case knee arthroscopies. All operations were performed by two surgeons. At the end of the procedure 20 ml of 0.75% ropivacaine was injected into the knee joint.


Primary outcome

A blood sample for plasma paracetamol level was taken 30 minutes after each patient's arrival in the recovery room from the arm contralateral to the intravenous line. Plasma paracetamol concentrations were determined with an enzymatic, colorimetric assay that uses the enzyme acylamidase to hydrolyse paracetamol to p-aminophenol (Cambridge Life Sciences Enzyme Assay Kit, in an Architect C8000 Analyser, Abbott Diagnostics, Illinois, USA). The assay is linear up to 2500 [micro]mol/l with sensitivity of 10 [micro]mol/l.

Secondary outcomes

Once the patient was awake a 100 mm visual analogue scale was used to assess postoperative pain at 10-minute intervals until discharge from the recovery area. Intravenous fentanyl was the only analgesic used as rescue analgesia postoperatively, prescribed as 20 [micro]g increments and given by a blinded recovery nurse if the visual analogue scale exceeded 30 mm. Total fentanyl requirements were then summed for each patient. Length of stay was calculated for each patient. Patients were discharged from the recovery area according to standard discharge criteria. Measurements were performed by a blinded recovery nurse.

Statistical analysis

Based on estimates of paracetamol administration effect from a previous study (8) (showing highly variable plasma paracetamol concentration following oral administration compared with intravenous) a sample of 10 patients in the intravenous group and 16 in the oral group was calculated to detect a clinically significant difference of 18 [micro]mol/l at an alpha value of 5% with 80% power. Numbers in the oral group were increased to 20 to allow for 20% of these patients having concentrations below the level of detection. The clinically significant difference selected was based on a previous study also sampling paracetamol levels 30 minutes postoperatively in children undergoing tonsillectomy (5) where an increase in plasma paracetamol level of approximately 20 [micro]mol/l (between 50, 70 and 90 [micro]mol/l) resulted in a 6% drop in the number of patients with unsatisfactory postoperative analgesia.

The Wilcoxon rank sum test and Fisher's exact test were used to assess patient groups to account for the modest sample size. There was no evidence that paracetamol levels were not normally distributed (Shapiro-Wilkes P=0.13) hence t-tests were used to calculate P values and 95% confidence interval. Linear regression analyses with effects for dose and group were used to assess the difference between oral and intravenous groups accounting for a linear dose effect. Repeated measures analysis of variance was used to assess the trend in pain score between groups over time. Results are reported as mean [+ or -] standard error unless otherwise stated. Differences are considered significant at P <0.05. All analysis was carried out using the statistical software package R 2.8.19.


Patient characteristics

Thirty-four patients were recruited, four did not meet inclusion criteria and therefore 30 were randomised between January and April 2009. Baseline patient characteristics and operative data were similar for both groups (Table 1) and in particular the dose of intraoperative fentanyl and the duration of surgery. Two patients smoked and both were in the oral group. Twenty-seven patients underwent simple medial meniscectomy, one patient in the intravenous group underwent medial meniscectomy with synovial biopsy and two patients in the oral group underwent synovial biopsy and removal of loose body. There were no adverse events recorded.

As expected from the study design, blood samples for paracetamol were collected closer to the time of dosing in the intravenous group at a mean of 39 minutes than in the oral group who were sampled at a mean of 119 minutes.

Plasma paracetamol levels

Mean plasma paracetamol levels 30 minutes after arrival in the recovery area were 88.6 [+ or -] 16.3 [micro]mol/l for the intravenous group and 53.2 [+ or -] 19.1 [micro]mol/l for the oral group (Figure 1); a significant difference of 35.5 [micro]mol/l (P=0.0005, confidence [interval.sub.95] 21.5 to 49.4). The highest level of 117 [micro]mol/l (from a patient in the intravenous group) was well below the level associated with paracetamol toxicity.

For both the intravenous and oral dosing there was a similar increase in plasma paracetamol levels with increasing dose adjusted for the patient's body weight. Each 1 mg/kg increase in dose resulted in plasma levels increasing by 6.03 [micro]mol/l.


Secondary outcomes

Patients in the oral group required more rescue analgesia in the recovery area than those in the intravenous group (Table 2) and stayed on average 12 minutes longer although these results did not reach statistical significance.

Pain scores were obtained only while the patient remained in the recovery area. Fifty minutes after arrival in the recovery area one-third of each group had been discharged. There was a statistically significant difference between pain scores only at this point for the remaining patients, 30.8 [+ or -] 5.8 mm oral versus 11.6 [+ or -] 2.8 mm intravenous (P=0.025).

Subsequent analysis concerning the seven patients in the oral group who did achieve plasma levels above 66 [micro]mol/l showed they required 84 [micro]g rescue fentanyl in recovery compared with 14 [micro]g in the intravenous group (P=0.16) and length of recovery stay was 79 minutes compared with 56 minutes in the intravenous group (P=0.064).


Main findings

We found that in this orthopaedic day surgery population intraoperative intravenous administration of 1.0 g paracetamol more consistently produced therapeutic plasma levels in the early postoperative period compared to the same dose given orally before surgery.

We also found a trend towards reduced length of recovery room stay and improved analgesia with intraoperative intravenous dosing.

The plasma concentrations of paracetamol required to produce antipyresis and analgesia are still uncertain. The widely quoted range for antipyresis in children of 10 to 20 mg/l (66 to 132 [micro]mol/l) has no published data to support it (4). In children undergoing tonsillectomy, Anderson et al showed that a plasma concentration of over 70 [micro]mol/l measured at 30 minutes postoperatively was associated with superior analgesia (5). In adults, Hahn et al (7) suggested that adequate analgesia following laparoscopic sterilisation was achieved with initial peak plasma concentrations of over 14 mg/l (92.4 [micro]mol/l) following postoperative administration of intravenous paracetamol. They failed to find any additional analgesic benefit above this level. Beck et al (6) investigated rectal paracetamol in women undergoing hysterectomy with similar findings. Despite the larger 40 mg/kg dose resulting in sustained plasma levels above the 66 [micro]mol/l level there was no additional analgesia compared to the lower 20 mg/kg dose which resulted in sub-therapeutic levels.

Van der Westhuizen recently showed that preoperative intravenous paracetamol gave higher plasma concentrations than preoperative oral paracetamol in the same 1.0 g dose and therapeutic levels were not reached at all in 33% of patients in the oral group (3). Our data supports the superiority of the intravenous preparation, finding 1.0 g of intravenous paracetamol given intraoperatively reliably resulted in therapeutic plasma paracetamol levels. Even given the short duration of the surgical procedures in our study, the majority of our patients receiving preoperative oral paracetamol had subtherapeutic levels by the time they reached the recovery room.

Paracetamol dose selection

Previous studies have demonstrated increasing magnitude and duration of analgesia with doses of up to 2 g of intravenous paracetamol in both postoperative patients (10) and in a volunteer pain stimulus model (11). Interestingly, this dose-response relationship has not been observed with oral paracetamol. The Cochrane Review examining single dose oral paracetamol for postoperative pain provides a number needed to treat of 3.5 for a dose of 0.5 g, and 3.6 for 1.0 g (1). There are limited data for doses above 1.0 g orally but four small randomised controlled trials failed to demonstrate any increased analgesic efficacy for 1.5 g over 1.0 g (12), 2.0 g over 1.0 g (13,14), or 90 mg/kg over 60 mg/kg (15).

It is possible that the apparent analgesic ceiling effect with oral paracetamol reflects the relatively slow rate of rise in plasma levels and thus relatively modest increases in peak plasma concentration achieved despite the increase in area under the concentration--time curve. Nielsen et al (16) found that a 2 g dose of sustained release paracetamol did not provide any detectable analgesia whereas lower doses of immediate release paracetamol were superior to placebo. This was despite the sustained release preparation resulting in superior later plasma levels.

With a time to maximum plasma level (Tmax) for oral paracetamol of around 30 minutes to one hour it is likely that the mean sampling time of 119 minutes after paracetamol administration in the oral group missed the maximum plasma level in these patients. When comparing our data to that of pharmacokinetic studies in healthy volunteers (17,18), we find the plasma levels of both our groups of patients fit with the plasma concentration versus time graphs of oral administration only. This suggests that it is the mismatch between Tmax and dosing with oral paracetamol that may make it less suitable for perioperative administration, not inferior bioavailability. It should be noted however that oral paracetamol absorption may be unreliable in the early postoperative period following even minor surgery (8) and, after major abdominal surgery, Kennedy et al (19) showed that peak plasma levels were only 50% of baseline on the second postoperative day.

We showed (Figure 2) a linear correlation between plasma paracetamol levels and the dose of paracetamol administered per kg body weight. The slope was 6.03 [+ or -] 0.91 [micro]mol/l per mg/kg dose. This was true for both the oral and intravenous preparations. This knowledge may guide target dosing of paracetamol in individual patients to achieve therapeutic analgesic levels in the recovery area. Our results suggest that to target therapeutic paracetamol levels in the recovery area, at least 15 mg/kg is required orally. This is consistent with other research suggesting a preoperative oral dose of 20 mg/kg is required to reach a plasma concentration of 19 mg/l (125 [mciro]mol/l), the median Tmax reached by the same dose given intravenously pre-induction (3). As previously discussed, the clinical benefit of this pharmacokinetic target is not firmly established, and the evidence to date has shown no dose response above 1.0 g of oral paracetamo (l1,12-15).


With 30 patients, this was a study powered to investigate postoperative plasma paracetamol levels. Our results are consistent with a trend towards improved analgesia and reduced length of stay in recovery with intraoperative intravenous dosing but larger numbers of patients would be required to demonstrate this at an acceptable level of significance.


Although many institutions use a higher dose in practice, a dose of 1.0 g paracetamol was selected for the oral group to enable direct comparison with the same dose given intravenously. Furthermore, there is evidence of lack of increased analgesic effect with doses larger than 1.0 g orally.

Knowledge of the effect-site concentration of paracetamol in our study would undoubtedly be ideal but sampling cerebrospinal fluid is clearly not appropriate in this study population. Given that plasma paracetamol concentrations have been linked to effect, measurement of these is a clinical tool that can be linked to effect site by a simple pharmacokinetic/pharmacodynamic model.

The timing of our measurement of plasma paracetamol level is likely to coincide with the decline in plasma paracetamol levels in both groups and may not reflect peak plasma levels (18). However, this single measurement of the plasma paracetamol level was selected as a clinically relevant point. Ensuring adequate analgesia in the early postoperative period is particularly important in the day case surgery setting.


We have confirmed our hypothesis that intraoperative administration of 1.0 g of intravenous paracetamol more reliably achieved proposed therapeutic levels for analgesia in recovery than the same dose given by mouth preoperatively. Oral paracetamol should have excellent bioavailability in a preoperative patient and these differences in plasma levels largely reflect the increased flexibility in the timing of intravenous administration. For short procedures a preoperative oral dose of 15 mg/kg would be expected to achieve the proposed therapeutic levels in the recovery phase of the patient's stay. Longer procedures would be better managed with the intravenous formulation. Our results also suggest a trend towards improved postoperative analgesia and length of stay in the recovery room for the intravenous group compared with the oral group although larger numbers of patients would be required to investigate this further.


Thanks to the staff of Burwood Hospital Operating Theatres, Christchurch.

Conflict of interest

Bristol-Myers Squibb (manufacturers of Perfalgan) were aware of the trial and had offered to pay for the costs of performing the plasma paracetamol assays, but this was not required. They had no input into the study design, or into any part of the study conduct. Costs were covered by the Department of Anaesthesia at Christchurch Public Hospital.


(1.) Toms L, McQuay HJ, Derry S, Moore RA. Single dose oral paracetamol (acetaminophen) for postoperative pain in adults. Cochrane Database Syst Rev 2008; 8:CD004602.

(2.) Moller PL, Sindet-Pedersen S, Petersen CT, Juhl GI, Dillenschneider A, Skoglund LA. Onset of actaminophen analgesia: comparison of oral and intravenous routes after third molar surgery. Br J Anaesth 2005; 94:642-648.

(3.) Van der Westhuizen J, Kuo PY, Reed PW, Holder K. Randomised controlled trial comparing oral and intravenous paracetamol (acetaminophen) plasma levels when given as preoperative analgesia. Anaesth Intensive Care 2011; 39:242-246.

(4.) Rumack BH. Aspirin versus acetaminophen: a comparative view. Pediatrics 1978; 62:943-646.

(5.) Anderson BJ, Kanagasundarum S, Woollard G. Analgesic efficacy of paracetamol in children using tonsillectomy as a pain model. Anaesth Intensive Care 1996; 24:669-673.

(6.) Beck DH, Schenk MR, Hagemann K, Doepfmer UR, Kox WJ. The pharmacokinetics and analgesic efficacy of larger dose rectal acetaminophen (40mg/kg) in adults: a double-blinded, randomized study. Anesth Analg 2000; 90:431-436.

(7.) Hahn TW, Mogensen T, Lund C, Jacobsen LS, Hjortsoe N-C, Rasmussen SN et al. Analgesic effect of i.v. paracetamol: possible ceiling effect of paracetamol in postoperative pain. Acta Anaesthesiol Scand 2003; 47:138-145.

(8.) Pettersson PH, Owall A, Jakobsson J. Early bioavailability of paracetamol after oral or intravenous administration. Acta Anaesthesiol Scand 2004; 48:867-870.

(9.) R Foundation for Statistical Computing. R Development Core Team review: A language and environment for statistical computing. From Accessed March 2011.

(10.) Juhl GI, Norholt SE, Tonnesen E, Hiesse-Provost O, Jensen TS. Analgesic efficacy and safety of intravenous paracetamol (acetaminophen) administered as a 2g starting dose following third molar surgery. Eur J Pain 2006; 10:371-377.

(11.) Piguet V, Desmeules J, Dayer P. Lack of acetaminophen ceiling effect on R-III nociceptive flexion reflex. Eur J Clin Pharmacol 1998; 53:321-324.

(12.) Laska EM, Sunshine A, Zighelboim I, Roure C, Marrero I, Wanderling J et al. Effect of caffeine on acetaminophen analgesia. Clin Pharmacol Ther 1983; 33:498-509.

(13.) Skoglund LA, Pettersen N. Effects of acetaminophen after bilateral oral surgery: double dose twice daily versus standard dose four times daily. Pharmacotherapy 1991; 11:370-375.

(14.) Skoglund LA, Skjelbred P, Fyllingen G. Analgesic efficacy of acetaminophen 1000 mg, acetaminophen 2000 mg, and the combination of acetaminophen 1000 mg and codeine phosphate 60 mg versus placebo in acute postoperative pain. Pharmacotherapy 1991; 11:364-369.

(15.) Zacharias M, De Silva RK, Hickling J, Medlicott NJ, Reith DM. Comparative safety and efficacy of two high dose regimens of oral paracetamol in healthy adults undergoing third molar surgery under local anaesthesia. Anaesth Intensive Care 2007; 35:544-549.

(16.) Nielsen JC, Bjerring P, Arendt-Nielsen L, Petterson K-J. Analgesic efficacy of immediate and sustained release paracetamol and plasma concentration of paracetamol. Double-blind, placebo-controlled evaluation using painful laser stimulation. Eur J Clin Pharmacol 1992; 42:261-264.

(17.) Sevilla-Tirado FJ, Gonzalez-Vallejo EB, Leary AC, Breedt HJ, Hyde VJ, Fernandez-Hernando N. Bioavailability of two new formulations of paracetamol, compared with three marketed formulations, in healthy volunteers. Methods Find Exp Clin Pharmacol 2003; 25:531-535.

(18.) Bannwarth B, Pehourcq F. Pharmcological Rationale for the Clinical Use of Paracetamol: Phamacokinetic and Pharmacodynamic Issues. Drugs 2003; 63:5-13.

(19.) Kennedy JM, van Rij AM. Drug absorption from the small intestine in immediate postoperative patients. Br J Anaesth 2006; 97:171-180.

C. N. BRETT *, S. G. BARNETT ([dagger]), J. PEARSON ([double dagger])

Department of Anaesthesia, Christchurch Hospital, Christchurch, New Zealand

This study was presented in part at the New Zealand Anaesthesia Annual Scientific Meeting, 5 November 2009 in Rotorua, New Zealand.

* FANZCA, Consultant Anaesthetist.

([dagger]) MB, BS, Anaesthetic Registrar, Dunedin Hospital, Dunedin.

([double dagger]) PhD, Consultant Biostatistician, Department of Pathology, University of Otago, Dunedin.

Address for correspondence: Dr C. Brett, Department of Anaesthesia, Christchurch Public Hospital, Riccarton Avenue, Private Bag 4710, Christchurch, New Zealand. Email:

Accepted for publication on August 27, 2011.
Patient characteristics and operative data (mean [+ or -] SE)

 Intravenous, Oral, P value *
 n=10 n=20

Age, y 50.7 50.3 0.42
 (13.3) (14.7)
Gender 0.70
 Female 3 8
 Male 7 12

BMI 27.5 (4.4) 27.9 (4.1) 0.72

ASA physical status 1.00
 I 7 14
 II 3 6

Smoking 0.54
 Yes 0 2
 No 10 18

Intraoperative fentanyl, [micro]g 165 (38) 158 (43) 0.65

Operation duration, min 23 (4) 0.10

Time from paracetamol 38 (7) 119 (29)
administration to testing
plasma levels, min

SE=standard error, BMI=body mass index, ASA=American
Society of Anesthesiologists. * P value from Wilcoxon rank
sum test for continuous variables from Fisher's exact test for
categorical values.

Outcomes. Plasma paracetamol levels, postoperative fentanyl
requirements, length of stay in recovery, mean pain scores
(mean [+ or -] SE)

 Intravenous, Oral, P value *
 n=10 n=20

Paracetamol level, 88.6 (16.3) 53.15 (19.1) 0.0005

Postoperative fentanyl, 14 (19.6) 45.5 (85.0) 0.53

Total time in recovery, 56 (8 64 (25) 0.61

Mean pain score, mm 15.3 (3.1) 23.1 (4.7) 0.17

SE=standard error. * P value from Wilcoxon rank sum test.
COPYRIGHT 2012 Australian Society of Anaesthetists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Brett, C.N.; Barnett, S.G.; Pearson, J.
Publication:Anaesthesia and Intensive Care
Article Type:Report
Geographic Code:8AUST
Date:Jan 1, 2012
Previous Article:Determinants of family satisfaction in the intensive care unit.
Next Article:Audit of extrapleural local anaesthetic infusion in neonates following repair of tracheo-oesophageal fistulae and oesophageal atresia via thoracotomy.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters