Effects of EDTA- and Sulfite-containing Formulations of Propofol on Respiratory System Resistance after Tracheal Intubation in Smokers

The Institutional Review Board at the Vanderbilt University Medical Center approved the study protocol and written informed consent was obtained from patients undergoing general anesthesia requiring tracheal intubation for elective surgery. The study population included patients classified as American Society of Anesthesiologists status 2–3. Only patients with reactive airways secondary to a long history of cigarette smoking (>1 pack/day, >5 yr) were enrolled into the study. All patients were actively smoking at the time of the investigation.

Patients with unstable cardiac or pulmonary disease, requiring rapid sequence intubation, patients under the age of 18 or geriatric patients above 80 yr of age, patients with sulfite allergy or treated with steroids or cholinolytics during the immediate perioperative period or in whom the protocol was otherwise deemed unsuitable by the primary anesthesiologists, were excluded from the study. After obtaining informed consent, the patients were assigned to receive 1% injectable emulsion of either ECP (Diprivan®, containing 0.005% of disodium edetate, manufactured for AstraZeneca Pharmaceuticals Ltd, Wilmington, DE, by AstraZeneca S.p.A., Caponago, Italy) (n = 20) or SCP (containing 0.25 mg/ml of sodium metabisulfite, manufactured for Baxter Healthcare Corporation, Deerfield, IL, by Gensia Sicor Pharmaceuticals, Irvine, CA) (n = 20). Propofol formulations (originally provided in 50 ml vials) were randomized, blinded, and dispensed in the opaque 50-ml syringes by the local hospital research pharmacy. Premedication with midazolam (0.04–0.05 mg/kg) was allowed. General anesthesia was induced with the investigated propofol 2 mg/kg bolus and fentanyl 0.002 mg/kg followed by vecuronium 0.1 mg/kg to accomplish muscle relaxation for the duration of the study. Intravenous injection of lidocaine (before or throughout propofol injection) was not used. Anesthesia was maintained with the same propofol formulation as used for induction at 0.15 mg · kg⁻¹· min⁻¹and supplemented by fentanyl (0.002 mg · kg⁻¹· h⁻¹).

Prior to tracheal intubation, complete muscle paralysis was confirmed using a nerve stimulation (train-of-four) of the ulnar nerve. All patients were intubated with a 7.5-mm endotracheal tube (ETT). The ETT was secured at 21 cm in women and at 23 cm in men. Correct placement of the ETT was confirmed by the presence of EtCO²on the capnograph and auscultation of equal bilateral breath sounds. Mechanical ventilation was started immediately after intubation with 50% oxygen, using tidal volumes of 10 ml/kg at 10 breaths/min, inspiratory:expiratory ratio (I:E) 1:1.5, and using square form inspiratory flow from Ohmeda Anesthesia Ventilator (Datex-Ohmeda, Madison, WI). These settings were maintained for the duration of the study. The presence or absence of clinical signs of bronchospasm (by auscultation) was also noted preoperatively, pre- and postintubation by an anesthesiologist blinded to the investigated propofol formulation.

The respiratory measurements were obtained using the noninvasive cardiac output monitor NICO²® System (Novametrix Medical Systems, Wallingford, CT) containing the respiratory profile module identical to the CO²SMO® and VenTrak® lung mechanics analyzers from the same manufacturer, extensively investigated previously in the area of lung research. 27Measurements were obtained by inserting the disposable flow/pressure adapter probe between the ETT outlet and the ventilator Y tubing, and included total respiratory system resistance (Rrs), peak inspiratory pressure (PIP), and dynamic compliance (Cdyn) obtained from the continuously recorded flow-volume loops. Heart rate, noninvasive blood pressure, and pulse oximetry (Spo²) were also recorded.

Flow and pressure measurements in the NICO²® monitor are made by a fixed orifice differential pressure pneumotachometer. The NICO²® monitor software compensations allow accurate flow and volume measurements, thus gas density and viscosity do not cause significant errors in flow measurement. The least-squares fitting method for calculating of Rrs has been used by the NICO²® system and it is based on the measurement of patient airflow, volume and airway pressure, as specified by the manufacturer. 28,29Briefly, the least squares fitting method assumes a specific model for the respiratory system and fits the waveform data to that model expressed by equation:

where dP is the pressure difference, R is respiratory resistance, F is airway flow, V is volume sample and C is compliance term. The dP is the pressure relative to a baseline level. The pulmonary end-expiratory pressure (PEEP) of the prior breath is used as the baseline level. The least square method minimizes the sum of squares between the observed pressure (dP observed ) and the best fit curve, dP bestfit :

To minimize the error between the best fit and observed pressures, the partial derivatives of S respective to R and C are computed, set to 0 and solved for R and C. This results in expressions for R and C consisting of cross-products of volume and flow, pressure and volume and flow themselves. The summations of these cross-products are accumulated throughout the inspiratory and expiratory portions of the breath from which dynamic compliance and resistance values are calculated. These calculations that are computationally intensive for a microprocessor-based system are computed real-time throughout the breath cycle using running summations.

The NICO²® system was connected through the serial port to personal computer allowing data collection of all respiratory parameters at an average of 6 s. Means were calculated for every parameter by averaging 10 single data (60 s period) at 1 min through 10 min postintubation.

The primary outcome from this study includes Rrs after tracheal intubation in the investigated groups. The secondary outcomes include PIP, Cdyn and the incidence of wheezing after induction and intubation. The format of the study was expressed by two hypotheses: the null hypothesis (H0), which states that there is no statistically significant difference in the postintubation Rrs among the two groups and the alternate hypothesis (Ha), which states that there is a difference. A minimum sample size of 20 patients in each group (40 patients total) was calculated as needed to be enrolled and analyzed to detect a clinically relevant difference in the primary outcome of Rrs according to the power analysis based on the following parameters: type II error rate (β= 0.15), type I error rate (α= 0.05, difference in the population means (ς) = 4 cm H²O · l⁻¹· sec⁻¹, within group SD (δ) = 4 cm H²O · l⁻¹· sec⁻¹). Categorical data were compared using Fischer exact test. A repeated measures analysis of covariance procedure was used to examine differences between the treatment groups over time. The analysis is an extension of a general linear model. The test statistics was adjusted for the repeated measures over subjects that introduce a new source of variability (within subject variability). This model allows for the formation of the variance-covariance matrices consisting of the within subject (repeated measures) and between subject sources of variance (treatment group). The model, as we specified it, fit an overall line for each treatment group including a term for treatment group, and a treatment group by time interaction term. Using this specification, the group by time interaction can be looked at as the slope of the response variable over time for each group, and the group term can be looked at the intercept (also interpreted as the “mean of the outcome”) for each group. These analyses were performed using the SAS statistical package (SAS Institute, Cary, North Carolina) and utilized the mixed model procedure (Proc Mixed) described by Little et al.  30A P  value less than 0.05 was considered statistically significant and chosen for rejection of the null hypothesis.

The patients’ demographics are shown in table 1. There were no significant differences between the two groups regarding age, sex, weight, height, ASA classification, proportion of smokers, pack-years (PPY) smoked, or any other analyzed demographic parameters. No statistically significant differences in the noninvasive blood pressure, heart rate, SO2p, and temperature were noted between the study groups at any time during the study period (preoperatively, immediately after induction and for 10 min period after intubation, data not shown). The end-tidal carbon dioxide concentrations were similar before induction and did not differ between treatment groups during any recording period during the 10 min study period (average end-tidal carbon dioxide 30 ± 4 mmHg and 31 ± 3.5 mmHg for ECP and SCP induction, respectively). The duration of time from induction to intubation was similar in both treatment groups, 4.5 ± 0.6 and 4.3 ± 0.5 min, for ECP and SCP groups, respectively.

Figure 1shows the mean airway resistance (Rrs) calculated for each study group at 1 min through 10 min postintubation. The repeated Rrs measurements for the next 10 min revealed trend consisting of higher Rrs in the SCP group when compared both the ECP group. Figure 2shows the generalized least squares fitted lines for the two treatments over time. The repeated measures analysis of covariance found that there is a statistically significant difference in intercepts between the two treatment groups (P = 0.0437); this was based on estimating and testing the group variable in the model. There was also significant decrease in mean Rrs for both groups over time (P = 0.0294) but the slopes are essentially equal (P = 0.4991), as determined by estimating and testing the group by time interaction. There was more variation between subjects in the SCP treatment group, compared with the ECP group (ς^B-ECP2= 36.7111%vs. ς^B-SCP2= 47.7633%); the correlation of the measurements for individuals (that is, how similar were the measurements taken on each subject over time) was very similar for both treatments (ρ^ECP= 0.8987 and ρ^SCP= 0.9056).

Postinduction wheezing was found (by auscultation during mask ventilation) in 3 patients, from which 2 were induced with SCP and 1 with ECP. Five patients showed wheezing postintubation, from which one patient remained wheezing in ECP group and 4 patients had wheezing in SCP group.

Despite statistically significant differences in the postintubation Rrs, only slight differences in the PIP values were recorded between the study groups (data not shown). The PIP values were about 10% higher in SCP group at 2 and 5 min after intubation, and these differences were not statistically significant (P > 005). No statistically differences (P > 0.05) in the Cdyn measurements were recorded in both investigated groups during 10 min after intubation (data not shown).

Our results demonstrate that tracheal intubation in smokers after induction and maintenance of anesthesia with SCP produces higher postintubation Rrs than when ECP was used. This effect was found immediately (1 min) after intubation and lasted for at least 10 min. After statistical analysis of the data with repeated measures ANOVA it is concluded that both analyzed variables: treatment group factor (i.e. , SCP vs.  ECP) and time (i.e. , time after intubation) significantly affected the overall Rrs in investigated subjects.

In the reported investigations we used the pneumotachometric technique to measure Rrs throughout the duration of the study. The measurement of Rrs includes, besides airway caliber associated resistance, components of resistance related to the chest wall, and additional components of lung resistance related to lung parenchyma. The inclusion of other components might make this parameter a less sensitive index to changes in airway caliber. Our measurements of respiratory resistance levels were however comparable with those of other studies and similar study populations. 8,27,31 

Our findings support the investigation of Brown et al.  6who showed that the change in the formulation of propofol abolishes the attenuation of induced bronchoconstriction in sheep model. In contrast to their study, our results revealed significant differences in the respiratory resistance after intubation and anesthesia induction with SCP compared to ECP, whereas Brown et al.  6found that SCP caused only a slight but not significant increase in Rrs to methacholine-induced bronchoconstriction; however, ECP attenuated methacholine-induced bronchoconstriction.

The effects of propofol on attenuation of airway muscle contraction induced by intubation or exposure to variety of physical and chemical factors have been confirmed in several in vitro  studies, 10,32as well as studies in animals 5,6and humans. 7–11As shown in previous in vitro  studies, 10propofol is able to inhibit Ca²⁺release from intracellular stores and also attenuates Ca²⁺influx via  voltage dependent channels. In addition, in vitro  propofol has been shown to produce relaxation of tracheal smooth muscle with spontaneous tone or contraction induced by acetylcholine, carbachol, histamine, prostaglandin F2α, and potassium. 10,13Our results are consistent with these previous studies in that they demonstrate that ECP, but not SCP, was able to attenuate intubation-induced bronchoconstriction.

Sulfite sensitivity in the form of bronchospasm and asthma is known to occur in the population of patients with reactive airways, and bronchoconstriction was reported previously after inhalation, enteric or intravenous administration of sulfite-containing products. 13–25The incidence of true sulfite allergy is estimated at 1/1000 patient. 33Sulfite-sensitive patients with asthma will not necessarily react after ingestion of sulfite containing food or drugs. 22,23However, it has also been shown that nonsulfite sensitive patients with asthma have a high incidence of bronchospasm after a metabisulfite challenge test. 22 

A number of studies have indicated that 5–10% of all chronic asthmatics are sulfite hypersensitive. 17Incidence of metabisulfite-induced bronchospasm in patients with reactive airways secondary to a long history of smoking has never been investigated. Eames et al.  8measured the airway resistance in response to intubation after induction with different anesthetics. The groups were subdivided into smokers and nonsmokers. Smokers showed a greater Rrs overall with a significant difference between propofol and other anesthetics. There was no significant difference among nonsmokers.

Based on the presented data we can consider that susceptibility to sulfite preservative in propofol is increased in at least some patients with previously altered tracheobronchial systems (e.g. , asthma or history of smoking). The exact mechanism of sulfite-induced bronchoconstriction is unknown and might include: 1) IgE-mediated antigen reaction 22; 2) Nonreagenic anaphylactoid reactions, caused when sulfite is combined with membranes of mast cells and basophiles 22; and 3) Activation of tracheobronchial irritant receptors and stimulation of a cholinergic reflex. 33It is unclear at the present time which of the above mechanisms might be responsible (if any) for the increased total respiratory system resistance after tracheal intubation in patients anesthetized with SCP. The other hypothesis implicates the adverse respiratory effects of some secondary products of propofol, metabisulfite, and lipids present in the SCP preparation. It was reported previously that metabisulfite content in the intralipid vehicle of propofol might promote formation various toxic oxidative metabolites of lipids, including malondialdehyde (MDA), and promote oxidative stress response. 34–36The hypothetical role of the sulfite and/or other potentially toxic compounds in the SCP solution in the mechanism of the observed adverse pulmonary effect remains to be elucidated.

In summary, SCP significantly increases intubation-induced bronchoconstriction in patients with a long history of smoking when compared to induction of general anesthesia with ECP. The clinical significance of this effect is uncertain and requires further investigations.