Steen, Evi R. MD; De Baerdemaeker, Luc E. C. MD, PhD, DEEA; Van Limmen, Jurgen MD; Wouters, Patrick MD, PhD
Increased serum lipids increase solubility (the blood/gas partition coefficient) of an inhaled anesthetic in blood, and thereby should slow the rate of changes in the anesthetic partial pressure in blood and central nervous system during induction of and recovery from anesthesia. We may encounter this issue resulting from the phospholipid contained in propofol.1 We had an unusual opportunity to test these predictions in a single patient, a child with severe lipodystrophy undergoing consecutive anesthetics before and after plasmapheresis, a procedure that restored the concentration of serum lipids to normal from greatly increased values.
We received approval from the Ethics Committee and informed consent from the parents for publication of this report.
A 9-year-old girl with a history of a chromosome 2q deletion (2q33.2-2q34) entered our hospital with red cell aplasia and hemolytic anemia after a mycoplasma infection. She had a dysmorphic and lipodystrophic appearance (22 kg, 133 cm; body mass index 12.4 kg/m2). Clinical examination revealed xanthelasmata, hepatomegaly, and normal lung/heart auscultation. Initial treatment with steroids and γ globulins was associated with the development of extreme hypertriglyceridemia (19,290 mg/dL), hypercholesterolemia (1291 mg/dL), and diabetes. Laboratory results included hemoglobin of 9.9 g/dL; platelet count 589 000/µL; normal electrolytes and increased serum glucose; and plasma transaminases (aspartate aminotransferase 399 U/L, alanine aminotransferase 476 U/L, lactate dehydrogenase 534 U/L).
General anesthesia was induced with remifentanil (0.15 mcg/kg/min)/thiopental and maintained with remifentanil (0.15–0.2 mcg/kg/min) and sevoflurane to permit insertion of a subclavian venous dialysis catheter for plasmapheresis. Thiopental (60 mg + 20 mg + 20 mg) to a total of 5 mg/kg produced loss of consciousness but did not provide relaxation sufficient for tracheal intubation. Sevoflurane in a fresh gas flow (FGF) of 10 L/min with a vaporizer setting of 8 vol% achieved adequate conditions for tracheal intubation 15 minutes after initiating induction, doing so without need for neuromuscular blocking drugs. The first measured end-tidal sevoflurane concentration was 3.5 vol%. After tracheal intubation, the FGF was set at 4 L/min, and the fractional delivered concentration (FD) was set at 4 vol%, resulting in end-tidal fractional concentrations (FE) of 2.6 to 2.8 vol% sevoflurane. The ratios of the FE to the inspired fractional sevoflurane concentrations (FI) during induction and maintenance are presented in Figure 1. The lungs were mechanically ventilated using a tidal volume of 220 mL at a rate of 20 breaths/min. Surgical time was 40 minutes. Blood aspirated from the subclavian vein had a pink appearance and an oily texture (Fig. 2). After 60 minutes of anesthesia, administration of remifentanil and sevoflurane ceased, and the FGF was set at 10 L/min. Spontaneous breathing resumed 5 minutes later (65 minutes after starting anesthesia). FGF was increased to 18 L/min. Consciousness returned (associated with a Bispectral Index[BIS] value >90), and the tracheal tube was removed 40 minutes after stopping anesthetic delivery. Pulse oximetry, heart rate, and arterial blood pressure values were stable during the entire procedure.
Occlusion of the subclavian catheter occurred after several weeks, again requiring general anesthesia for insertion of a Hickman catheter. Laboratory results now included hemoglobin of 5.9 g/dL, platelets 337 000/µL, normal electrolytes, serum glucose and plasma triglycerides/cholesterol, and nearly normal serum transaminases (aspartate aminotransferase 31 U/L, alanine aminotransferase 43 U/L, lactate dehydrogenase 871 U/L) with a c-reactive protein level of 0.3 mg/dL. After induction with sevoflurane (FGF 10 L/min; FD initially 8 vol %), an IV catheter was inserted, and remifentanil administration initiated at 0.2 mcg/kg/min, continuing with 0.15 to 0.2 mcg/kg/min. Ten minutes after beginning anesthesia, the trachea was intubated (sevoflurane FE 1.8 vol%; Fig. 1). After intubation, FGF was set to 4 L/min with FD 4 vol%, decreased to 3 vol% after 20 minutes, and decreased to 2 vol% after 37 minutes until the end of surgery. Surgical time was 77 minutes. Sevoflurane and remifentanil administration ceased after 87 minutes of anesthesia, and FGF was increased to 10 L/min. Ten minutes after stopping anesthetic administration, consciousness returned with BIS values higher than 90, and the tracheal tube was removed.
During the first anesthetic, an FE of 3.46 ± 1.50 vol% (mean ± SD) maintained BIS values between 40 and 60, in contrast to an FE of 2.23 ± 0.36 vol% which produced the same BIS values during the second anesthetic.
Lipodystrophies are acquired or inherited disorders (acquired more common than inherited)2 characterized by the partial or complete loss of adipose tissue. In acquired generalized lipodystrophy, the fat loss occurs during childhood and adolescence (predominantly females), affecting large areas of the body, and these patients have hypertriglyceridemia. Absence of subcutaneous fat is noted early, often at or soon after birth. Acanthosis nigricans and hepatomegaly due to fatty infiltration of the liver are other associated abnormalities. In this patient, the diagnosis of acquired generalized lipodystrophy was made and retained.
The pharmacodynamics of enflurane anesthesia in a patient with congenital lipodystrophy was described in 1987, and in this patient, washout was delayed.3 In our patient, the first anesthetic was complicated by extreme triglyceridemia and hypercholesterolemia. These prolonged induction and emergence times contrast to those in the second anesthetic when the lipid profile was normal (Fig. 1). This case report illustrates the influence of serum lipids on the uptake and washout of volatile drugs.4,5
The blood/gas partition coefficient (λ) is the primary determinant of the rate of washin and washout of a volatile anesthetic. It may be affected by age,6 hypothermia,7,8 cholesterol, and serum triglyceride concentrations.4,9 Malviya and Lerman4 demonstrated that serum cholesterol normally predicts 18% of sevoflurane’s blood/gas partition coefficient. Hu et al.9 observed that patients with coronary artery disease have a significantly larger plasma triglyceride concentration and associated greater blood/gas partition coefficients for desflurane, isoflurane, and halothane.
What might have been the blood/gas partition coefficient for our patient before plasmapheresis? The normal sevoflurane blood/gas partition coefficient is approximately 0.65,10 but this is in the absence of the enormous concentrations of triglycerides (19,290 mg/dL) and cholesterol (1291 mg/dL) found in this child’s blood before plasmapheresis. Assuming the triglyceride and cholesterol components possess a specific density of 0.91 g/mL, these values would translate to 22.6 mL fat per 100 mL blood. Sevoflurane has a oil/gas partition coefficient of 47,10 and thus the 22.6 mL fat (i.e., oil) alone might contribute a partition coefficient of 0.226 × 47 or 10.6. Add to that, the contribution of the 77.4 mL blood remaining after subtracting the 22.6 mL fat and a partition coefficient of 11 might be a good approximation. That is almost 12, the partition coefficient for diethyl ether,11 an anesthetic famous for the slowness with which it induces anesthesia and from which recovery occurs.
Our patient differed from the normal patient in more than the enormous blood/gas partition coefficient before plasmapheresis. Her body mass index of 12.4 kg/m2 which was part of her lipodystrophy indicates that her fat depots were minimal or effectively absent. Bulk fat depots can store nearly endless amounts of anesthetic with time constants for equilibration measured in days, but the blood flow to bulk fat is limited, and thus, its effect on both uptake and elimination is small. Of greater importance is the loss of the effect of fat on intertissue diffusion—movement of anesthetic from well-perfused tissues to a thin layer of adjacent fat (e.g., from kidney to perirenal fat; from intestine and liver to omental and mesenteric fat).12 Our patient had less of this capacity, and from that standpoint, her highly perfused tissues including her brain may have equilibrated with and recovered from anesthesia more rapidly.
However, the lipidemia may have affected (lipid might have infiltrated) some highly vascular tissues or even muscle. That large organ, the liver, may have suffered from fatty infiltration, as suggested by hepatomegaly and increased levels of serum transaminases. Figure 1 gives results consistent with the aforementioned considerations. During induction of anesthesia, the FE/FI was greater after correction of the lipidosis (second anesthetic). For both anesthetics, intubation of the trachea decreased the FE/FI ratio (after 4–5 minutes for the second anesthetic and after 10 minutes for the first anesthetic), probably because of the greater accuracy of sampling of end-tidal gases consequent to the marked decrease in dead space. With consistent sampling (from mask or from tracheal tube), the FE/FI value was consistently higher (washin was faster) for the second anesthetic.
What can be said of the accuracy with which the FE/FI ratio represents the arterial/inspired sevoflurane partial pressure ratio (Pa/PI) after intubation? Even after intubation, FE/FI exceeds Pa/PI because of the contamination of FE gas with FI gas. Because the difference between FE and FI increases with increasing solubility of anesthetic in blood, the FE/FI ratio increasingly deviates from the Pa/PI ratio as the solubility of the anesthetic in blood increases. This assumes that the correctness of this interpretation explains the greater FE required to sustain what was interpreted to be the same level of anesthesia in the first anesthetic (e.g., the same BIS values). To achieve that level of anesthesia during the first (greater lipid) anesthesia required a mean FE value of 3.46% ± 1.50% sevoflurane, a value 55% more than the 2.23% ± 0.36% sevoflurane to achieve the same effect during maintenance with the second (normal lipid) anesthetic. A BIS value of ±60 was found for both anesthetics at the end of anesthesia (Fig. 3). Either we must accept that the FE/FI for the first anesthesia considerably overestimated the Pa/PI value, or we must accept that a greatly increased lipid in the plasma considerably increased anesthetic requirement.
Figure 3 presents results that are readily interpreted. The great fat content in blood (perhaps to some extent spilling over into tissues such as liver) during the first anesthetic dramatically delayed elimination of the sevoflurane despite the shorter duration of the first anesthetic, doing so because of the greatly increased capacity of the lipid-laden blood and tissue to hold anesthetic. Long ago, Severinghaus13 noted that solubility was crucial to % clearance at the lungs, showing that % clearance = 1/[1 + (Q/VA) × λ ], where Q is cardiac output, VA is alveolar ventilation, and λ is the blood/gas partition coefficient. Assuming Q equals 5 L/min and VA 4 L/min, if λ equals 0.65, then clearance = 55%. But if λ equals 11, then clearance = 6.8%. Thus, it takes longer to eliminate the sevoflurane taken up during the first anesthetic; the BIS value returns to normal far more slowly, and time to tracheal extubation is delayed (40 vs 10 minutes). Why after the first anesthetic did the BIS value remain unchanged for 20 minutes after discontinuing sevoflurane administration? One answer might be that BIS is not linearly related to sevoflurane concentrations.14 That is, at anesthetizing concentrations, the BIS value correlates poorly with changes in effect-site concentrations.
What does this case report tell us? We learn again (in this case in a single patient) that solubility plays a key role in inhaled anesthetic pharmacokinetics. Indeed, the uniqueness of this case report is that there is but 1 individual with only blood solubility materially changed. The case report is also unique in the absence or near absence of bulk fat to affect uptake and elimination, reducing the groups affecting uptake to blood, vital tissues (the vessel rich group), and muscle (the muscle group).
1. Yang XL, Ma HX, Yang ZB, Liu AJ, Luo NF, Zhang WS, Wang L, Jiang XH, Li J, Liu J. Comparison of minimum alveolar concentration between intravenous isoflurane lipid emulsion and inhaled isoflurane in dogs. Anesthesiology. 2006;104:482–7
2. Garg A. Acquired and inherited lipodystrophies. N Engl J Med. 2004;350:1220–34
3. Koga Y, Sakuma N, Iwatsuki N, Hashimoto Y, Yamada Y. Anesthesia for a patient with total lipodystrophy–a clinical report. J Anesth. 1987;1:112–4
4. Malviya S, Lerman J. The blood/gas solubilities of sevoflurane, isoflurane, halothane, and serum constituent concentrations in neonates and adults. Anesthesiology. 1990;72:793–6
5. Saraiva RA, Willis BA, Steward A, Lunn JN, Mapleson WW. Halothane solubility in human blood. Br J Anaesth. 1977;49:115–9
6. Lerman J, Gregory GA, Willis MM, Eger EI 2nd. Age and solubility of volatile anesthetics in blood. Anesthesiology. 1984;61:139–43
7. Eger RR, Eger EI 2nd. Effect of temperature and age on the solubility of enflurane, halothane, isoflurane and methoxyflurane in human blood. Anesth Analg. 1985;64:640–2
8. Lockwood GG, Sapsed-Byrne SM, Smith MA. Effect of temperature on the solubility of desflurane, sevoflurane, enflurane and halothane in blood. Br J Anaesth. 1997;79:517–20
9. Hu P, Zhou JX, Liu J. Blood solubilities of volatile anesthetics in cardiac patients. J Cardiothorac Vasc Anesth. 2001;15:560–2
10. Eger EI 2nd, Eisenkraft JB, Weiskopf RB. The Pharmacology of Inhaled Anesthetics. 2002 Chicago, IL Healthcare Press:45
11. Eger EI 2nd, Shargel R, Merkel G. Solubility of diethyl ether in water, blood and oil. Anesthesiology. 1963;24:676–8
12. Eger EI IIEger EI 2nd, Eisenkraft JB, Weiskopf RB. Pharmacokinetics. The Pharmacology of Inhaled Anesthetics. 2002 Chicago, IL Healthcare Press:43–71
13. Severinghaus JWPapper E, Kitz R. Role of lung Factors. Uptake and Distribution of Anesthetic Agents. 1963 New York, NY McGraw-Hill:59–71
14. Baars JH, Kalisch D, Herold KF, Hadzidiakos DA, Rehberg B. Concentration-dependent suppression of F-waves by sevoflurane does not predict immobility to painful stimuli in humans. Br J Anaesth. 2005;95:789–97