Because Food and Drug Administration (FDA) approved the use of IV anesthetic propofol in 1989, it has become the most popular IV anesthetic in the United States and Europe. The ultra-short duration of action also made propofol the most widely used sedating drug for diagnostic and short duration surgical procedures. However, prolonged use of propofol infusions in intensive care units has been reported to result in the development of propofol infusion syndrome (PRIS), a disorder characterized by lactic acidosis, myocardial dysfunction, and high mortality. The mechanism of PRIS remains unclear, its presentation can be protean, and the risk factors remain poorly defined, although they include high dose, long duration of propofol infusion, and young age.
Although typical clinical signs of the syndrome include lactic acidosis, rhabdomyolysis, renal failure, lipemia, and fatty infiltration of liver,1,2 metabolic acidosis and fatal cardiac failure are thus far considered to be the cardinal features of PRIS.3,4 According to a FDA report of deaths related to propofol use, 86% of patients suffered progressive myocardial dysfunction and 71% had metabolic acidosis.4 On the other hand, when PRIS was defined as progressive cardiac failure associated with metabolic acidosis and/or rhabdomyolysis, in only 31% of cases was the clinical picture consistent with the definition of PRIS.4
The onset time of PRIS also varies. PRIS was initially described as a fatal syndrome, developing after prolonged (>48 h) high-dose (>5 mg/kg) propofol infusion in critically ill patients.1–3,5–7 However, case reports of PRIS developing even after short-term infusion in children and adults during anesthesia in operating rooms or sedation in intensive care units have been sporadically appearing in the literature.8–17 In the majority of these patients, an increase in lactate and development of metabolic acidosis was apparent early, within 2–5 h after initiation of the propofol infusion. Moreover, the shortest onset of PRIS was reported to occur after propofol sedation for less than 1 h with unexplained severe metabolic acidosis, which resolved after cessation of propofol.14 Four cases of metabolic acidosis during propofol anesthesia in adults without known risk factors for PRIS have been reported11,12,17 in which blood lactate levels or base deficit increased beginning at 4 h after initiation of propofol infusion.12 A report by Cravens et al.18 suggested that metabolic acidosis could even develop instantly with subanesthetic doses of propofol, which are usually used for sedation as diagnostic and nonsurgical procedures.
Most experts on PRIS believe that early recognition and treatment of lactic acidosis play a crucial role in preventing the full-blown development of PRIS.13,14,16,19,20 Because PRIS had been reported to occur even in patients who received infusion of propofol for short duration, we hypothesized that lactic acidosis may develop during total IV anesthesia (TIVA) with high-dose propofol infusion administered for surgical procedures of long duration. The primary goal of our study was to investigate and compare the changes in lactate levels and pH during TIVA with propofol infusion with that during volatile anesthesia (VA) after 8 h of spine surgery.
Study Design and Participants
After IRB approval, a retrospective chart review of anesthesia and medical records of patients who underwent spine surgery between January 2004 and February 2006 at Harborview Medical Center (Seattle, WA) was performed. At our institution, TIVA with continuous infusion of propofol is the preferred anesthetic technique during spine surgery, when intraoperative monitoring of motor-evoked potentials is required21 by the surgeon. Otherwise, these procedures are usually performed under VA. Consequently, a large number of prolonged anesthetics with either propofol or VA were available to comprise the study population.
Potentially eligible patients were identified from the anesthesia records. Spine surgery cases were identified based on Current Procedural Terminology (CPT) code (## 22212, 22327, 22556, 22558, 22610, 22612, 22630, 22840-3, 22851-2). Study eligibility criteria included 1) age ≥18 yr; 2) elective surgery; and 3) anesthesia time of 8 h or longer. Exclusion criteria were 1) emergency or urgent surgery; 2) trauma and critically ill patients; and 3) use of combined propofol infusion and VA. To maximize the study power, for every patient who received propofol infusion as a part of TIVA (propofol group), two patients who received VA (VA group) were randomly selected. To enhance comparability between the two groups, restriction criteria were applied based on 1) a duration of anesthesia (anesthesia time [AT]) within 30 min of a matching TIVA subject, and 2) intraoperative blood loss (BL) ±500 mL, to select patients in the VA group.
Our routine anesthesia practice during major spine surgery includes narcotic infusions of either remifentanil or fentanyl depending on the attending anesthesiologist’s preference with both TIVA and VA, direct arterial blood pressure monitoring, and frequent routine sampling of blood gases, which includes blood lactate levels, for monitoring of the level of hematocrit.
The following variables were retrieved and recorded: 1) demographic data: age, gender, body weight, ASA classification, medical history, chronic medications; and 2) intraoperative data: surgical time, type of surgery, number of vertebral levels instrumented, anesthesia duration, dose of anesthesia medications, systolic blood pressure (SBP) every 5 min (SBP was recorded by manual entry every 5 min on the anesthesia chart, and the value was retrieved and analyzed), intraoperative use of other medications, fluid balance, core temperature, laboratory data of arterial blood gases (ABL 725, Radiometer, Copenhagen, 700 series), where blood lactate level was measured by enzymatic amperemeter (normal reference 0 · 4–1 mmol/L). Base excess (BE) was calculated using a standard formula.22 For the collection of laboratory variables (lactate, pH, Paco2, bicarbonate, and glucose), the blood sample needed to be withdrawn within ±30 min of the specified study point. If the collection was outside this time window, no data were recorded. For the purpose of the study, we defined a priori the index time for all measurements as the time 8 h after “start of anesthesia” (T8) ± 60 min. Baseline was defined as the first value collected within 2 h of anesthesia start time.
Definition of Arterial Hypotension
Baseline SBP was defined as SBP measured preoperatively or retrieved from the records of the preoperative anesthesia clinic. Arterial hypotension was defined as SBP <90 mm Hg at any recorded time point (every 5 min). Episodes of arterial hypotension were identified and recorded.
Data are presented as mean ± sd. An initial exploratory data analysis was conducted to evaluate and describe the distribution of the baseline characteristics and the completeness of the data collection. We evaluated the distribution of baseline characteristics in the propofol and VA group using two-sample Student’s t-test or frequency tables, as appropriate. Two sample Student’s t-test was also used to compare the mean change in blood lactate levels from baseline to T8 between the propofol and VA groups. Complete case analysis was performed. In the main analysis, to account for unbalanced baseline characteristics other than AT and intraoperative BL, we used multivariable linear regression to adjust for potential confounders and precision variables including baseline blood lactate levels. For this analysis, lactate level at T8 was used as a response variable. In secondary analyses, we compared blood pH, bicarbonates, and glucose (response variables) between treatment groups, using the same regression model. A P < 0.05 was considered significant. P values are two-sided. The STATA statistical software, version 9.2 (Stata Corporation, College Station, TX) was used for all analyses.
The screening process for identifying eligible patients is summarized in Figure 1. Of 246 cases, 50 propofol cases were identified and matched to 100 VA cases, by AT (propofol: 618 ± 98 min vs VA: 589 ± 85 min, P = 0.9) and BL (propofol: 1955 ± 1409 mL vs VA: 1801 ± 1543 mL, P = 0.9). Ten patients in the propofol group and 28 patients in the VA group were excluded because lactate collection data were missing either at baseline or at T8, leaving 112 patients available for the analysis (n = 40 in the propofol group and n = 72 in the VA group, 10 patients received sevoflurane and 62 isoflurane).
Demographic and Baseline Characteristics
Patients who received VA were slightly older, had greater body weight, and slightly higher baseline SBP, when compared with patients who received propofol (Table 1). However, there was no difference between the groups in gender, severity of illness, use of chronic antihypertensive medications, and history of chronic arterial hypertension (Table 1).
The total dose of propofol was 8.8 ± 2 mg · kg−1 · h−1 for up to the 8th hour among patients in the propofol group (Table 2), which translates approximately to the target of 4 μg/mL after 1 h or 4.5 μg/mL after 8 h of anesthesia, if target control infusion is used. There were no differences in number of operated vertebral segments (Table 2), intraoperative BL, and fluid balance between the groups (Table 3). When defined as SBP <90 mm Hg, there was no difference between the groups in the incidence of intraoperative hypotension. Use of steroids, insulin, vasopressors, and inotropes were similar in both groups. Patients in the propofol group had lower glucose during surgery than in the VA group (Table 2). There was no difference between the groups in intraoperative arterial Paco2, pH, bicarbonate, BE, and rate of insulin administration (Table 2).
There were no patients who received IV bicarbonate during surgery. After adjusting for baseline BE, age, weight, and baseline SBP, the propofol group tended to have higher BE than the VA group at 8 h (P = 0.089).
A baseline lactate level was normal in both groups but higher in the VA group (Table 2). A significant increase in arterial lactate level over time was observed in both groups: by 0.48 ± 0.7 mmol/L in the propofol group and by 1.2 ± 1.2 mmol/L in the VA group (Table 2). The 8-h change in lactate was significantly greater in patients given VA compared with patients receiving propofol anesthesia (unadjusted difference: 0.72, 95% CI: 1.15–0.0; P < 0 · 001). After adjusting for baseline lactate levels and unbalanced baseline characteristics (age, body weight, baseline glucose, and SBP), the absolute lactate change from baseline (to 8 h) remained statistically significant between the two groups (adjusted difference: 0 · 61, 95% CI: 1.1–0.1; P = 0 · 013).
In the whole cohort of 150 patients (Fig. 2), an incidence of lactic acidosis, when determined as pH <7.35 and lactate >2.5 mm/L at any time point during surgery, was found to be 7% in the VA group and 0% in the propofol group.
Because metabolic acidosis is currently considered to be one of the most important pathognomonic signs of PRIS, early recognition of metabolic/lactic acidosis in patients receiving high-dose propofol infusion may play a crucial role in the interruption or prevention of further development of the syndrome.17,20 We hypothesized that prolonged propofol infusion may cause an increase in lactate levels, possibly as a subclinical manifestation of PRIS. Instead, we observed higher lactate levels during VA when compared with propofol anesthesia, and not a single patient in our series developed lactic acidosis after 8 h of high-dose propofol infusion, suggesting that the propofol-based general anesthesia is less likely to cause lactic acidosis than VA.
In this regard, our data are at variance to the observations made by Cravens et al.18 who reported a higher incidence of subclinical metabolic acidosis in patients receiving low-dose propofol infusion (about half of the dose used in our patients), when compared with the control group. These observations, if confined, would raise alarm and appropriate concern whenever propofol is administered. However, these findings are difficult to interpret because of study flaws such as their rather liberal definition of “metabolic acidosis” (defined as a BE ≤−2 mEq/L only, thus including the lower limit of the normal range), unclear indications for withdrawing of blood gases, and a questionable control group.
Usually, a differential diagnosis of lactic acidosis during anesthesia is not complicated.23 Decreased oxygen delivery due to hypovolemia, hypotension, and/or anemia during anesthesia may contribute to hypoperfusion and hypoxemia of peripheral tissues, causing anaerobic metabolism and, as a result, an increase in blood lactate. In our study, the incidence of intraoperative hypotension was similar. Although there was a small difference in baseline SBP between the propofol and VA groups, this could potentially be spurious, because the incidence of chronic arterial hypertension and its treatment were similar in both groups. Patients in the VA group were slightly older (58 ± 14 yr) than in the propofol group (52 ± 15 yr). However, age per se is not known to affect blood lactate levels.24 In a comparative study examining the influence of age, similar levels of perioperative blood lactate were observed in patients after hemodilution and deliberate hypotension under isoflurane anesthesia, when a much older group than ours (72 ± 5 yr) was compared with a much younger group than ours (47 ± 11 yr).24
Because multiorgan failure secondary to systemic inflammatory response syndrome is a major cause of death in critically ill patients suffering profound inflammatory and immune imbalance,38 the host response was believed to play an important role in the development of PRIS.2 Most probably, the syndrome manifests as an idiosyncratic reaction to propofol occurring in susceptible individuals with a mild mitochondrial myopathy. Association with central nervous system and respiratory disease, treatment with catecholamines and corticosteroids, and genetic predisposition involving an impairment of mitochondrial fatty acid metabolism have been considered as risk factors for development of PRIS.20 Patients in our study were not critically ill, had undergone elective surgery, and both groups had a similar incidence of corticosteroid treatment during the surgery. However, we cannot comment on specificities of lipid metabolism or genetics in our population.
Because we observed a statistically significant increase in lactate levels with VA, compared with propofol, an alternative explanation for the difference in blood lactate levels between VA and propofol groups may be a different effect of these anesthetics on lactate metabolism, particularly on 1) lactate production, 2) lactate clearance via the liver or kidney, or 3) imbalance between both. Various factors can affect lactate metabolism during surgery and anesthesia. Release of catecholamines increases lactate production and decreases lactate clearance by decreasing hepatic blood flow and activation of Na-K-ATP pumps, provoking glycolysis.25 Although we did not measure catecholamine levels in the blood, we can assume that patients in both groups in our study had similarly extensive surgical stimulation, because there was no difference in the number of operated vertebral levels and BL. A potential explanation of an increase in lactate over time with isoflurane and sevoflurane would be an impaired liver clearance of lactate, because both VAs cause a dose-dependent decrease in hepatic blood flow.26–32 On the other hand, propofol might be expected to decrease lactate levels as it contains 10% soybean oil (Diprivan, Willmington, Zeneca, Inc.), which is a 10% intralipid emulsion containing a mixture of triglycerides, predominantly unsaturated fatty acids. An infusion of free fatty acids for 2 h has been reported to inhibit glycolysis and stimulate neoglycogenesis in healthy men and diabetic patients.33,34 Similarly, an increase of free fatty acids in blood with intralipids has been demonstrated to profoundly inhibit intrahepatic glycogenolysis and stimulate neoglycogenesis, thereby increasing intrahepatic uptake and use of lactate.35,36 In this regard, even a short-term (up to 160 min) infusion of propofol has been shown to significantly increase triglyceride levels in blood.37 Contrary to the mentioned short-term effect of propofol, we observed an increase in lactate after 8 h of anesthesia.
Because our patients were not given any glucose-containing solutions in either group, the VA group, compared with the propofol group, may be considered a “fasting group” in which increased glycolysis and glycogenolysis, subsequently leading to an increase in lactate production, may be expected. In contrast, in the propofol group, patients received about one-half of the daily caloric intake (an average of 8 · 8 mg · kg−1 · h−1 of propofol during 10 h equals 9 kcal/kg [1 mL of 10% intralipids = 1.1 kcal]). This significant caloric intake, compared with the “fasting” VA group, may play a positive role in neoglucogenesis and use of lactate, thereby maintaining blood lactate levels within the normal range.
Although the incidence of insulin treatment was similar, at 8 h of anesthesia both groups were mildly hyperglycemic, with a slightly, but not statistically significantly higher level of glucose in the VA group. This could potentially contribute to a higher lactate level.
In conclusion, the results of our study show that prolonged anesthesia with VAs isoflurane or sevoflurane is associated with an increase in blood lactate levels that are more than lactate levels observed during high-dose propofol anesthesia. However, although the difference in lactate levels between VA and propofol groups was statistically significant, the clinical importance of this finding is not clear because very few patients in the VA group developed lactic acidosis. This is the first study to document that high-dose propofol infusion for surgical anesthesia for 8 h causes neither significant hyperlactatemia nor lactic acidosis. The influence of the anesthetics on the lactate clearance and/or production might explain differences in blood lactate between the anesthetics in our study. However, the exact mechanisms of our findings cannot be ascertained, and prospective studies are needed to elucidate the physiology of lactate production and elimination, and the role of propofol as a nutrient during anesthesia of long duration. Nevertheless, our data strongly contrast the Cravens et al.18 study and do not support the contention that lactic acidosis develops frequently during low-dose propofol infusion and may be a subclinical manifestation of PRIS.
Our findings cast doubt on the occurrence of subclinical PRIS during prolonged propofol infusion in adults, and the incidence of PRIS or subclinical PRIS during propofol infusion remains unknown.
1. Kang TM. Propofol infusion syndrome in critically ill patients. Ann Pharmacother 2002;36:1453–6
2. Vasile B, Rasulo F, Candiani A, Latronico N. The pathophysiology of propofol infusion syndrome: a simple name for a complex syndrome. Intensive Care Med 2003;29:1417–25
3. Cremer OL, Moons KG, Bouman EA, Kruijswijk JE, de Smet AM, Kalkman CJ. Long-term propofol infusion and cardiac failure in adult head-injured patients. Lancet 2001;357:117–8
4. Wysowski DK, Pollock ML. Reports of death with use of propofol (Diprivan) for nonprocedural (long-term) sedation and literature review. Anesthesiology 2006;105:1047–51
5. Parke TJ, Stevens JE, Rice AS, Greenaway CL, Bray RJ, Smith PJ, Waldmann CS, Verghese C. Metabolic acidosis and fatal myocardial failure after propofol infusion in children: five case reports. BMJ 1992;305:613–6
6. Bray RJ. Propofol infusion syndrome in children. Paediatr Anaesth 1998;8:491–9
7. Perrier ND, Baerga-Varela Y, Murray MJ. Death related to propofol use in an adult patient. Crit Care Med 2000;28:3071–4
8. Mehta N, DeMunter C, Habibi P, Nadel S, Britto J. Short-term propofol infusions in children. Lancet 1999;354:866–7
9. Marinella MA. Lactic acidosis associated with propofol. Chest 1996;109:292
10. Liolios A, Guerit JM, Scholtes JL, Raftopoulos C, Hantson P. Propofol infusion syndrome associated with short-term large-dose infusion during surgical anesthesia in an adult. Anesth Analg 2005;100:1804–6
11. Burow BK, Johnson ME, Packer DL. Metabolic acidosis associated with propofol in the absence of other causative factors. Anesthesiology 2004;101:239–41
12. Salengros JC, Velghe-Lenelle CE, Bollens R, Englelman E, Barvais L. Lactic acidosis during propofol-remifentanil anesthesia in an adult. Anesthesiology 2004;101:241–3
13. Koch M, De Backer D, Vincent JL. Lactic acidosis: an early marker of propofol infusion syndrome? Intensive Care Med 2004;30:522
14. Haase R, Sauer H, Eichler G. Lactic acidosis following short-term propofol infusion may be an early warning of propofol infusion syndrome. J Neurosurg Anesthesiol 2005;17:122–3
15. Hermanns H, Lipfert P, Ladda S, Stevens MF. Propofol infusion syndrome during anaesthesia for scoliosis surgery in an adolescent with neonatal progeroid syndrome. Acta Anaesthesiol Scand 2006;50:392–4
16. Kill C, Leonhardt A, Wulf H. Lacticacidosis after short-term infusion of propofol for anaesthesia in a child with osteogenesis imperfecta. Paediatr Anaesth 2003;13:823–6
17. Chukwuemeka A, Ko R, Ralph-Edwards A. Short-term low-dose propofol anaesthesia associated with severe metabolic acidosis. Anaesth Intensive Care 2006;34:651–5
18. Cravens GT, Packer DL, Johnson ME. Incidence of propofol infusion syndrome during noninvasive radiofrequency ablation for atrial flutter or fibrillation. Anesthesiology 2007;106:1134–8
19. Corbett SM, Moore J, Rebuck JA, Rogers FB, Greene CM. Survival of propofol infusion syndrome in a head-injured patient. Crit Care Med 2006;34:2479–83
20. Kam PC, Cardone D. Propofol infusion syndrome. Anaesthesia 2007;62:690–701
21. Sloan TB, Heyer EJ. Anesthesia for intraoperative neurophysiologic monitoring of the spinal cord. J Clin Neurophysiol 2002;19:430–43
22. Siggaard-Andersen O. An acid-base chart for arterial blood with normal and pathophysiological reference areas. Scand J Clin Lab Invest 1971;27:239–45
23. Funston JS, Prough DS. Two reports of propofol anesthesia associated with metabolic acidosis in adults. Anesthesiology 2004;101:6–8
24. Fukusaki M, Nakamura T, Hara T, Fukushima H, Hasuo H, Sumikawa K. Splanchnic perfusion during controlled hypotension with haemodilution under isoflurane anaesthesia in elderly patients. Eur J Anaesthesiol 1999;16:519–25
25. Gladden LB. Lactate metabolism: a new paradigm for the third millennium. J Physiol 2004;558(Pt 1):5–30
26. Bernard JM, Doursout MF, Wouters P, Hartley CJ, Merin RG, Chelly JE. Effects of sevoflurane and isoflurane on hepatic circulation in the chronically instrumented dog. Anesthesiology 1992;77:541–5
27. Gelman S, Fowler KC, Smith LR. Liver circulation and function during isoflurane and halothane anesthesia. Anesthesiology 1984;61:726–30
28. Gelman S, Fowler KC, Smith LR. Regional blood flow during isoflurane and halothane anesthesia. Anesth Analg 1984;63:557–65
29. Goldfarb G, Debaene B, Ang ET, Roulot D, Jolis P, Lebrec D. Hepatic blood flow in humans during isoflurane-N2
O and halothane-N2
O anesthesia. Anesth Analg 1990;71:349–53
30. Fujita Y, Kimura K, Hamada H, Takaori M. Comparative effects of halothane, isoflurane, and sevoflurane on the liver with hepatic artery ligation in the beagle. Anesthesiology 1991;75:313–8
31. Frink EJ Jr, Morgan SE, Coetzee A, Conzen PF, Brown BR Jr. The effects of sevoflurane, halothane, enflurane, and isoflurane on hepatic blood flow and oxygenation in chronically instrumented greyhound dogs. Anesthesiology 1992;76:85–90
32. Nishiyama T, Yokoyama T, Hanaoka K. Effects of sevoflurane and isoflurane anesthesia on arterial ketone body ratio and liver function. Acta Anaesthesiol Scand 1999;43:347–51
33. Boden G, Jadali F. Effects of lipid on basal carbohydrate metabolism in normal men. Diabetes 1991;40:686–92
34. Saloranta C, Franssila-Kallunki A, Ekstrand A, Taskinen MR, Groop L. Modulation of hepatic glucose production by non-esterified fatty acids in type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia 1991;34:409–15
35. Chu CA, Sherck SM, Igawa K, Sindelar DK, Neal DW, Emshwiller M, Cherrington AD. Effects of free fatty acids on hepatic glycogenolysis and gluconeogenesis in conscious dogs. Am J Physiol Endocrinol Metab 2002;282:E402–E411
36. Chu CA, Galassetti P, Igawa K, Sindelar DK, Neal DW, Burish M, Cherrington AD. Interaction of free fatty acids and epinephrine in regulating hepatic glucose production in conscious dogs. Am J Physiol Endocrinol Metab 2003;284:E291–E301
37. Gottschling S, Meyer S, Krenn T, Kleinschmidt S, Reinhard H, Graf N, Shamdeen GM. Effects of short-term propofol administration on pancreatic enzymes and triglyceride levels in children. Anaesthesia 2005;60:660–3
© 2009 International Anesthesia Research Society
38. Bone RC. Immunologic dissonance: a continuing evolution in our understanding of the systemic inflammatory response syndrome (SIRS) and the multiple organ dysfunction syndrome (MODS). Ann Intern Med 1996;125:680–7