Sun, Nicholas C. H. MB BS*; Wong, Andrew Y. C. MB, ChB, FHKAM*; Irwin, Michael G. MB, ChB, MD, DA, FRCA, FHKCA, FHKAM†
Department of Anaesthesiology, *Queen Mary Hospital; and †The University of Hong Kong, Queen Mary Hospital, Pokfulam Road, Hong Kong
This work was funded by the Department of Anaesthesiology, The University of Hong Kong.
Accepted for publication January 12, 2005.
Address correspondence and reprint requests to Dr. Michael G. Irwin, Department of Anaesthesiology, University of Hong Kong, Room 424, K Block, Queen Mary Hospital, Pokfulam Road, Hong Kong. Address e-mail to firstname.lastname@example.org.
Propofol (2,6-diisopropylphenol) is one of the most commonly used IV anesthetics. Since 1982, it has been formulated in a concentration of 10 mg/mL in a fat emulsion consisting of 10% soybean oil (long-chain triglycerides) and marketed as Diprivan® 1% (AstraZeneca, Cheshire, UK). When used for anesthetic induction, Diprivan® causes pain or discomfort on injection in 28%–90% of patients with many factors affecting the incidence and severity (1–3). The large concentration of free propofol in the aqueous phase of Diprivan® is thought to be particularly associated with injection pain (4).
In a newer formulation of propofol, Propofol-Lipuro® 1% (B. Braun, Melshungen AG, Germany), the oil phase consists of long- and medium-chain triglycerides. Such a composition results in a smaller concentration of free propofol in the aqueous phase. An improved tolerability with Propofol-Lipuro® on injection compared with Diprivan® has been claimed and there are studies that show reduced pain intensity with Propofol-Lipuro® (5,6). These studies, however, have limitations such as lack of control over site, speed of injection, and patient and temporal variability. Consequently, we devised a randomized, double-blind, crossover study to compare pain on injection between Diprivan® and Propofol-Lipuro®.
After obtaining approval from our local IRB and written informed patient consent, 60 ASA physical status I or II patients were recruited. They were older than 18 yr and scheduled for elective gynecological or maxillofacial procedures under general anesthesia. Excluded from the study were patients with recent trauma, a neurological deficit in the nondominant hand, or known allergies to propofol, egg lecithin, soybean oil, or EMLA® (eutectic mixture of local anesthetics of 5% lidocaine and prilocaine) cream.
All patients had EMLA® cream applied to the dorsum of the nondominant hand 1–2 h before surgery. In the induction room, after the EMLA had been removed, a small vein on the dorsum of this hand was cannulated with a 22-gauge IV catheter (Angiocath, BD; Infusion Therapy System Inc., Sandy, UT). The pain verbal rating score (VRS 0–10) at cannulation was recorded (7). One milliliter of each propofol (Diprivan® and Propofol-Lipuro®) was prepared and labeled A or B randomly by an independent person outside the room. Because Diprivan® and Propofol-Lipuro® have the same appearance, both the investigator and patient were blinded to the study drugs.
Three minutes after venous cannulation, study drug A (either Diprivan® or Propofol-Lipuro®) was injected at a constant speed over 20 s using a syringe pump (Terumo TE-312; Terumo Corporation, Tokyo, Japan). VRS was recorded during injection (T0) and at 1 (T1), 2 (T2), 3 (T3), and 5 min (T5) after injection. The cannula was flushed with 10 mL of normal saline. After a further 10 min, the same procedure was repeated for study drug B. Consequently, in 30 patients, Diprivan® was given first followed by Propofol-Lipuro® (group D-P) and in another 30 patients, the sequence was reversed: Propofol-Lipuro® followed by Diprivan® (group P-D).
To detect a within-patient mean difference of area under the curve pain scores between the 2 preparations of propofol of no less than 5, the required sample size with 80% power was 51 at the two-sided 0.05 level of significance. For integrity, we recruited 60 patients (30 in each treatment sequence). Student’s t-test, χ2 test, and Mann-Whitney U-test were used to analyze demographic data. VRS were analyzed by two-way crossover analysis of variance (8). The statistical calculations were performed by the SAS System for Windows release 8.02 (SAS Institute Inc., Cary, NC). A value of P < 0.05 was taken as significant.
Patient characteristics, EMLA® duration, number of cannulation attempts, and VRS during cannulation did not differ between groups (Table 1). VRS with each preparation and differences in VRS between the 2 propofol preparations at different time points in each patient were analyzed (Table 2). In both preparations, the median VRS peaked at T1 and then showed a continuous decrease thereafter. In group D-P, the pain was significantly less with Propofol-Lipuro® compared with Diprivan® during initial injection (median difference in VRS = 2 [interquartile range 0–2], P = 0.002) and at 1 min (3 [0–4], P < 0.001). Interestingly, however, when Propofol-Lipuro® was administered first (group P-D), no significant differences in VRS were shown at any time points. Median VRS had decreased to 0 in both groups by T3. The change in median VRS period difference over time in the 2 groups is depicted in Figure 1. There was a significant difference in the area under the curve between the 2 groups (P = 0.002).
Most patients experienced only very mild, if any, sedative effects from each propofol injection and were fully conscious at the time of the injection of drug B.
Pain during IV bolus injection of propofol is a common problem (1–3). The incidence and intensity of the pain is affected by many factors, e.g., cannula size and site of insertion, volume, temperature and speed of injection, and concentration of propofol in the aqueous phase of the preparation (4). In Propofol-Lipuro®, more propofol is dissolved in the oil phase, which consists of long- and medium-chain triglycerides, and the resultant reduction in propofol concentration in the aqueous phase has been claimed to reduce pain on injection (4–6). Considering the many factors affecting injection pain with propofol and that pain is a subjective feeling easily influenced by personal and environmental factors, these studies claiming reduced pain with Propofol-Lipuro® have several limitations.
The study by Doenicke et al. (5) is the only one with a crossover design, but the 2 injections of different propofol preparations were separated by 14 days with no information provided on local anesthesia for cannulation, cannulation sites, and size of cannula. In the study by Rau et al. (6), the patients received lorazepam premedication, which may obviously have affected perception. In all theses studies, an anesthetic induction dose of propofol was used, making delayed VRS measurement impossible. We devised this crossover study with the 2 injections separated by 10 minutes, no premedication, EMLA applied to minimize the cannulation pain, and a subanesthetic dose of propofol allowing continuous measurement of VRS. It is possible that prior injection of one form of the drug may have influenced the perception of pain during injection of the subsequent one; however, we have minimized this bias by randomly alternating between Diprivan® and Propofol-Lipuro® as the first drug, flushing the cannula with normal saline, and allowing a reasonable time for attenuation of any central nervous system effect between doses. The broad range of VRS from T0 to T2 (0–10) reflects wide interindividual differences in perception of pain and emphasizes the importance of a good crossover design. Pretreatment with EMLA has been shown to have no effect on pain during propofol injection (9).
The immediate pain on propofol injection is attributed to a direct irritant effect of the drug (3). Propofol-Lipuro® did not cause immediate pain on injection in most patients (median VRS = 0) and this may be related to the reduction in propofol concentration in the aqueous phase. In both preparations, the VRS peaked at T1 which reflects the “delayed” pain of propofol injection. This pain has an onset latency of between 10 and 20 seconds (10), and is probably an indirect effect mediated via the kinin cascade. Propofol belongs to the group of phenols that irritate the skin, mucous membranes, and venous intima (11). By an indirect action on the endothelium, it activates the kallikrein-kinin system and releases bradykinin, thereby producing venous dilation and hyperpermeability, increasing the contact between the aqueous phase of propofol and the free nerve endings, resulting in pain (12). Propofol is now widely used for both anesthesia and sedation and has also been successfully used in subhypnotic doses for antiemesis (13). Pain on injection is obviously not important enough to negate its pharmacokinetic and pharmacodynamic advantages over other drugs that have led to this popularity. It is, however, troublesome and unpleasant, particularly during sedation. The clinical significance of this is reflected in the myriad techniques that have been tried in an attempt to reduce the incidence of pain on propofol injection. They differ in complexity and have produced variable results. Examples include adding lidocaine, cooling or warming propofol, diluting the propofol solution, injection of propofol into a large vein, and pretreatment with IV lidocaine, ondansetron, metoclopramide, an opioid, magnesium, or thiopental with or without application of a tourniquet (3,11,14–19). These interventions, however, are inconvenient and may be inappropriate during sedation. Our study shows that the use of Propofol-Lipuro® is associated with virtually no immediate pain on injection and a much reduced delayed pain without resorting to such techniques. The reason for this pain reduction was not investigated in the current study, although we speculate that it may be related to the smaller concentration of propofol in the aqueous phase of the formulation.
Interestingly, Propofol-Lipuro® also seemed to attenuate the pain of subsequent Diprivan® injection. It may be that Propofol-Lipuro® can somehow inhibit pain mediators or their release; however, it would be expected not to produce any pain at all if this was the case. This has been postulated as a mechanism for the efficacy of pretreatment with drugs such as thiopental (16), although, in that study, pretreatment was given with venous occlusion to maximize drug contact with the vessel wall. It could simply be that, even with our relatively small doses and a 10-minute washout period, propofol still has some subtle effect on pain perception (20).
In conclusion, the reduction in immediate pain with Propofol-Lipuro® administration may be particularly useful during anesthesia induction, because obviously injection pain is not a problem once unconscious, and coupled with a reduced delayed pain would be an advantage during sedation.
1. Klement W, Arndt JO. Pain on injection of propofol: effects of concentration and diluent. Br J Anaesth 1991;67:281–4.
2. Lyons B, Lohan D, Flynn C, McCarroll M. Modification of pain on injection of propofol. Anaesthesia 1996;51:394–5.
3. Tan CH, Onsiong MK. Pain on injection of propofol. Anaesthesia 1998;53:468–76.
4. Doenicke AW, Roizen MF, Rau J, et al. Reducing pain during propofol injection: the role of the solvent. Anesth Analg 1996;82:472–4.
5. Doenicke AW, Roizen MF, Rau J, et al. Pharmacokinetics and pharmacodynamics of propofol in a new solvent. Anesth Analg 1997;85:1399–403.
6. Rau J, Roizen MF, Doenicke AW, et al. Propofol in an emulsion of long- and medium-chain triglycerides: the effect on pain. Anesth Analg 2001;93:382–4.
7. Katz J, Melzack R. Measurement of pain. Int Anesthesiol Clin 1992;10:229–45.
8. Jones B, Kenward MG. Design and analysis of cross-over trials. 2nd ed. New York: Chapman & Hall, 2003.
9. McCluskey A, Currer BA, Sayeed I. The efficacy of 5% lidocaine-prilocaine (EMLA) cream on pain during intravenous injection of propofol. Anesth Analg 2003;97:713–4.
10. Briggs LP, Clarke RSJ, Dundee JW, et al. Use of diisopropyl phenol as main agent for short procedures. Br J Anaesth 1981;53:1197–201.
11. Ambesh SP, Dubey PK, Sinha PK. Ondansetron pretreatment to alleviate pain on propofol injection: a randomized, controlled, double-blind study. Anesth Analg 1999;89:197–9.
12. Coderre TJ, Katz J, Vaccarino AL, Melcack R. Contribution of central neuroplasticity to pathological pain: review of clinical and experimental evidence. Pain 1993;52:259–85.
13. Scott RP, Saunders DA, Norman J. Propofol: clinical strategies for preventing the pain of injection. Anaesthesia 1988;43:492–4.
14. Numazaki M, Fujii Y. Reduction of emetic symptoms during cesarean delivery with antiemetics: propofol at subhypnotic dose versus traditional antiemetics. J Clin Anesth 2003;15:423–7.
15. Picard P, Tramer MR. Prevention of pain on injection with propofol: a quantitative systematic review. Anesth Analg 2000;90:963–9.
16. Agarwal A, Ansari MF, Gupta D, et al. Pretreatment with thiopental for prevention of pain associated with propofol injection. Anesth Analg 2004;98:683–6.
17. Roehm KD, Piper SN, Maleck WH, Boldt J. Prevention of propofol-induced injection pain by remifentanil: a placebo-controlled comparison with lidocaine. Anaesthesia 2003;58:165–70.
18. McCrirrick A, Hunter S. Pain on injection of propofol: the effect of injectate temperature. Anaesthesia 1990;45:443–4.
19. Memis D, Turan A, Karamanlioglu B, et al. The use of magnesium sulfate to prevent pain on injection of propofol. Anesth Analg 2002;95:606–8.
20. Anker-Moller E, Spangsberg N, Arendt-Nielsen L, et al. Subhypnotic doses of thiopentone and propofol cause analgesia to experimentally induced acute pain. Br J Anaesth 1991;66:185–8.