Effect of Isoflurane on Neutrophil Phagocytic Function During Pregnancy

Objective: General anesthesia has been considered an independent risk factor for postcesarean infection, but the mechanism for this association has not been delineated. The purpose of this prospective investigation was to determine if phagocytic response of neutrophils was impaired by in vitro exposure to isoflurane, a commonly used anesthetic. Methods: Twelve milliliter venous blood samples were withdrawn from 18 term patients during labor. Neutrophils were separated by Ficoll gradient centrifugation. Aliquots of 2 × 106 neutrophils/ml were exposed to anesthesia using an airtight modular incubator chamber through which a 0.5% isoflurane:50% N2O + 50% O2 mixture flowed at a rate of 4 l/min for 90 min at 37℃. Neutrophils were assayed for phagocytosis by incubation with Escherichia coli conjugated with fluorescein isothiocyanate for 30 min at 37℃. Phagocytosis was assessed by flow cytometry. Neutrophils from the same patient that were not exposed to anesthesia served as controls. Results: The mean percentage of phagocytizing neutrophils in the isoflurane-treated group was 82.8 ± 24 compared to 83.5 ± 22 in the control group. The difference between the two groups was not significant. Conclusions: In vitro exposure to the general anesthetic isoflurane for 90 min does not significantly alter the phagocytic capacity of neutrophils.

2 |tion following cesarean delivery. Of the clinical risk factors for infection, general anesthesia was identified by Green and Sarubbi 3 to be the most statistically significant variable. However, according to Gibbs, 2 general anesthesia was not a consistent determinant for increased rate of postoperative infection. Clearly, the role of anesthesia as a predisposing risk factor for infection following obstetric surgery is not yet well established.
Inhalational anesthetic agents have been shown to suppress the immune system. 4'5 Some agents cause impairment of phagocytic function, an important aspect of nonspecific host resistance. [6][7][8][9] In-creased susceptibility to bacterial infections has been closely associated with defective neutrophil function. The host depends initially on neutrophils to eliminate invading bacterial pathogens. In obstetric patients, the stresses of surgery and anesthesia during cesarean delivery may diminish the activity of phagocytes, thus predisposing the patient to postpartum infection.
The purpose of the present prospective study was to determine the effect of isoflurane on the phagocytic function of neutrophils collected from laboring patients. Isoflurane is an inhalation anesthetic commonly used in clinical obstetrics in the United States. The technique of flow cytometry was used to evaluate phagocytosis of fluoresceinconjugated Escherichia coli by purified neutrophils after exposure in vitro to isoflurane. Flow cytometry allowed single cell analysis of phagocytic activity. 10--12

MATERIALS AND METHODS
The study population consisted of 18 normal term patients between the ages of 18 and 40 years who were in labor when admitted to Shands Hospital, University of Florida, from September to December, 1991. Written informed consent was obtained from each patient in accordance with guidelines established by the Institutional Review Board.
Twelve-milliliter samples of peripheral venous blood were withdrawn into heparinized vacutainer tubes from each of the 18 patients. Granulocytes from whole blood were isolated using a modification of Boyum's Ficoll-I--Iypaque density gradient centrifugation. 13 Briefly, 3 ml ofHistopaque-1119 (Sigma Chemical Co., St. Louis, MO) was transferred into a 15 ml conical centrifuge tube, then layered with 3 ml of Histopaque-1077 (Sigma Chemical Co.). Six milliliters of whole blood was carefully overlayed onto the upper gradient and centrifuged in a swinging bucket rotor at room temperature for 30 min at 700g. At the end of centrifugation, a distinct layer of lymphocytes, platelets, and other mononuclear cells settled at the plasma/1077 interphase while a layer of predominantly granulocytes settled at the 1077/1119 interphase. Erythrocytes gravitated to the bottom. Each layer was aspirated; the granulocyte layer was saved and washed 3 10 9 cells/ml. One milliliter aliquots were stored at -80C and thawed to room temperature immediately before use.
One milliliter suspensions of purified neutrophils were distributed into 35 10 mm sterile polystyrene disposable suspension culture dishes (Corning Glass Works, Corning, NY). Neutrophils that were exposed to anesthesia were placed inside an airtight Modular Incubator Chamber (Billups-Rothenberg, Inc., Del Mar, CA) through which a 0.5% isoflurane (Forane USP, Anaquest, Madison, WI):50% N20 + 50% O 2 mixture flowed at a rate of 4 1/min for 90 min at 37C. The isoflurane vaporizer was calibrated using a Perkin-Elmer mass spectrometer (Perkin-Elmer, Norwalk, CT). Gas samples were taken from the circuit limb containing exhaust gas after chamber equilibration was achieved. For controls, cell suspensions from each patient were incubated at 37C but not exposed to the anesthetic. The 90 min exposure period was selected because it should be near the maximum duration of exposure to general anesthesia that a patient would encounter during cesarean delivery.
Phagocytosis of fluorescein-conjugated E. coli was measured in a reaction mixture containing 0.5 ml cell suspension, 0.1 ml of fluorescent E. coli, and 0.4 ml diluted pooled sera. The ratio of neutrophils to bacteria was routinely 1:25 to 1:60. The mixture was allowed to incubate with an end-overend rotation for 30 min at 37C. The reaction was terminated by the addition of 3 ml cold 3  for data collection and analyses. The number of neutrophils, the percent fluorescent neutrophils, as well as the mean fluorescence per neutrophil were calculated within the gated region. The percent fluorescent neutrophils represented those cells that phagocytized fluorescent E. coli. For each patient, the percent phagocytizing neutrophils after in vitro exposure to isoflurane was compared to control cells that were not exposed to isoflurane. Flow cytometric observations were confirmed by fluorescent microscopy (Nikon Optiphot, Nikon, Inc., Garden City, NY). To ensure that the measured fluorescence was due to actual phagocytosis, we determined the effect of sodium azide, a known inhibitor of phagocytosis. Sodium azide in 100 mM concentration was incorporated into the phagocytosis assay mixture. is To further distinguish fluorescence due to internalization of bacteria vs. fluorescence resulting from simple adherence of bacteria to the cell surface of the phagocytes, fluorescence-quenching experiments were performed with trypan blue at low pH. 16 After termination of the phagocytosis reaction, the cells were suspended in cold phosphate-buffered saline with 0.25 mg/ml trypan blue at pH 4.5, then analyzed by flow cytometry.
Diluted pooled sera in the phagocytosis reaction mixture were used for opsonization to improve the efficiency of phagocytosis. Serum was collected from normal pregnant women and frozen in aliquots. Immediately prior to use, an aliquot was thawed and diluted 1:4 with RPMI 1640. Statistical analysis was performed using the twotailed paired t-test. P < 0.05 was considered significant.

RESULTS
Different cell types were discriminated by their combined forward light scatter which measured cell size and side scatter which measured granularity. Figure  termined within the gated region was considered the percent phagocytizing neutrophils. The optimal ratio of phagocytes to bacteria and length of incubation for routine phagocytosis assay were first determined. Figure 2 demonstrates the rate of increase in percentage of phagocytizing neutrophils with length of incubation, as well as with increasing neutrophil to E. coli ratios. Maximal phagocytosis was achieved after 30 min incubation, where approximately 90% of neutrophils were fluorescent. Nonsignificant increases in phagocytosis occurred with longer incubation periods of up to 60 min. The rate of ingestion increased with bacterial concentration and reached saturation at a neutrophil to E. coli ratio of 1:25. There was no further increase in the rate of uptake when the number of bacteria was 50, 100, and 150 times the number of neutrophils. Based on these findings, a neutrophil to E. coli ratio of 1:25 to 1:60 with 30 min incubation time was routinely used for the phagocytosis assay.
Quenching with trypan blue at pH 4.5 did not alter the percent fluorescent neutrophils after 30 min of phagocytosis, indicating that at the end of this period fluorescent bacteria had been internalized. Sodium azide at 100 mM concentration decreased phagocytosis to 5-7% in both the control and isoflurane-exposed groups. Panels A in Figure 3 represent two-parameter dot displays of the gated neutrophil population of control cells and isoflurane-treated cells at times 0

DISCUSSION
The response of neutrophils to bacterial invasion includes chemotaxis, phagocytosis, oxidative and hydrolytic intracellular killing, and release of lysosomal components. Several studies have demonstrated that inhalation anesthetics depress certain phases of the phagocytic response. Significant inhibition of microbicidal capacity of human neutrophils has been observed with the volatile anesthetic halothane. 6'7 Nakagawara and colleagues 8 studied the effects of halothane, isoflurane, and enflurane on phagocytosis, superoxide production, and intracellular calcium mobilization on neutrophils obtained from healthy adult volunteers. Their results showed that these volatile agents caused a decrease in superoxide production that appeared to be due, at least in part, to inhibition of intracellular calcium mobilization.
Welch 9 reported that enflurane depressed bacterial killing and chemiluminescence only in neutrophils that were stressed by high bacterial challenge, but this inhibition was reversed by exposure to air for 30 min. Earlier investigations demonstrated that halothane, trichloroethylene, diethyl ether, and methoxyflurane inhibited the migration of human neutrophils toward a chemoattractant, casein. 17 Neutrophil chemotaxis was similarly depressed by nitrous oxide and enflurane, but not by enflurane's chemical isomer, isoflurane. 18 In fact, isoflurane stimulated chemotaxis of elicited rabbit neutrophils in vitro. 19 Therefore, different anesthetic agents appear to have different mechanisms by which they affect neutrophil function.
In our experiment, the profound decrease in percentage of fluorescent neutrophils in the presence of an inhibitor of phagocytosis such as sodium azide confirms that the amount of fluorescence in both the control cells and those exposed to isoflurane was due to actual engulfment of the bacteria by phagocytosis. Additional proof of actual internalization of fluorescent E. coli was the fact that no further fluorescence quenching was seen with addition of trypan blue in acid pH after 30 min of phagocytosis. The dye would quench fluorescence from bacteria that were attached externally to the cell surface but not those inside the cell. 16 Our study demonstrates that the ability to phagocytize bacteria remains active in neutrophils of normal pregnant women after exposure to isoflurane.
Although this investigation did not examine microbial killing by phagocytes, Welch 9 previously reported that the ability of neutrophils of normal healthy adults to kill E. coli, Klebsiella pneumoniae, or Staphylococcus aureus was not significantly altered when exposed to 1-3 % isoflurane for h. Furthermore, no inhibition of microbicidal activity of human neutrophils was observed with 70% nitrous oxide and 30% oxygen, alone, or in combination with isoflurane. INFECTIOUS  Therefore, bacterial infections occurring immediately after cesarean delivery do not appear to be the direct result of use of a general anesthetic such as isoflurane. The increased risk of infection is more likely due to the complex emergencies that create the need for general anesthesia for delivery.