Changes in lung mechanics and ventilation-perfusion match: comparison of pulmonary air- and thromboembolism in rats

This experimental protocol was included in a research project approved by the National Food Chain Safety and Animal Health Directorate of Csongrád County, Hungary (no. XXXII./2096/2018) on September 24, 2018. The experimental procedures were performed according to the guidelines of the Scientific Committee of Animal Experimentation of the Hungarian Academy of Sciences (updated Law and Regulations on Animal Protection: 40/2013. [II. 14.], Government of Hungary) and European Union Directive 2010/63/EU on the protection of animals used for scientific purposes. All methods are reported in accordance with ARRIVE guidelines for the reporting of animal experiments [11].

Nine male Wistar rats (362 ± 25 g) were anesthetized by the intraperitoneal injection of sodium pentobarbital (45 mg/kg; Sigma-Aldrich, Budapest, Hungary). The left femoral artery and vein were cannulated by 23 G catheters to administer medications, monitor blood pressure, and collect blood samples. Anesthesia was maintained with intravenous sodium pentobarbital (5 mg/kg) every 30 min, and muscle relaxation was obtained by neuromuscular blockade with repeated intravenous boluses of pipecuronium (0.2 mg/kg every 30 min; Arduan, Richter-Gedeon, Budapest, Hungary). The body temperature of the rats was maintained at 37 °C ± 0.5 °C throughout the experiment using a heating pad equipped with a rectal thermometer (model 507223F; Harvard Apparatus, Holliston, MA, USA).

The animals were tracheostomized after additional local anesthesia with subcutaneous lidocaine (2–4 mg/kg), and the trachea was cannulated with an 18 G uncuffed endotracheal tube. Volume-controlled mechanical ventilation with a tidal volume (VT) of 7 mL/kg and positive end-expiratory pressure (PEEP) of 5 cmH₂O was applied using a small animal ventilator (Model 683; Harvard Apparatus, South Natick, MA, USA). The ventilation frequency was set to 55–60 breaths/min to maintain an end-tidal CO₂ (ETCO₂) level within the normal range (35–45 mmHg). Midline thoracotomy was then performed to access the left pulmonary artery. The mean arterial pressure (MAP) and heart rate (HR) were monitored and recorded using a data collection and acquisition system (Powerlab 8/35 and Labchart, ADInstruments, Dunedin, New Zealand).

The mechanical parameters of the airway and respiratory tissues were determined by measuring the input impedance of the respiratory system (Zrs) using the wave-tube approach of the forced oscillation technique [12]. Briefly, the tracheal cannula was attached to a loudspeaker-in-box system, which generated a small-amplitude pseudorandom forcing signal (amplitudes < 1.5 cmH₂O with 23 non-integer multiple-frequency components in the range of 0.5–20.75 Hz) through a wave-tube (polyethylene; length, 100 cm; internal diameter, 2 mm). During these measurements, ventilation was briefly suspended (8 s) at end-expiration, and the oscillatory pressures were measured with two identical miniature differential pressure transducers (Honeywell Differential Pressure Sensor model 24PCEFA6D; Honeywell, Charlotte, NC, USA) at the loudspeaker and tracheal ends of the wave-tube. After a minimum of four technically acceptable measurements in each protocol phase, Zrs values were calculated as the load impedance of the wave-tube [12]. The mechanical properties of the respiratory system were obtained by fitting the well-established and validated constant phase model to the ensemble-averaged Zrs spectra (16). This model comprised frequency-independent airway resistance (Raw) and airway inertance in series with a viscoelastic constant phase tissue unit that includes tissue elastance (H) and tissue damping (G) [13]. Hysteresivity, reflecting the coupling of the resistive and elastic properties within the viscoelastic respiratory tissues, was calculated as η = G/H [14].

Samples of 0.15 mL arterial blood were collected for blood gas analyses, from which the arterial partial pressures of oxygen (PaO₂) and carbon dioxide (PaCO₂) were determined using a point-of-care blood analyzer system (Epoc Reader and Host; Epocal, Inc., Ottawa, ON, Canada).

The partial pressure of carbon dioxide (PCO₂) in expired gas and ventilation airflow (V′) were simultaneously recorded using a rodent sidestream volumetric capnograph (Harvard Capnograph Type 340 for small rodents) at a sampling frequency of 256 Hz. Volumetric capnogram curves were generated for each expiratory cycle from the PCO₂ tracing and the volume signals obtained by integrating the corresponding V′ data. The capnogram phase boundaries were determined, and the shape factors were calculated based on previously described concepts [15, 16]. Briefly, the inflection point of phase 2 was determined as the maximum of the first derivative of the volumetric capnogram curve, and the phase 2 slope (S2V) was measured as the rate of change of PCO₂ around this inflection point. The start and end points of phase 2, which reflect the mixed emptying of airway-alveolar spaces, were identified as the maxima of the third derivative before and after this inflection point in both the time and volumetric domains. The phase 3 capnogram slope (S3V), reflecting the dynamics of the alveolar gas compartment emptying, was determined by fitting a linear regression line to the middle third of phase 3. Normalized phase 2 (Sn2V) and 3 (Sn3V) slopes were calculated by dividing S2V and S3V by the respective ETCO₂ values. This normalization allows a more objective comparison of changes in the capnogram shape due to changes in ETCO₂, such as observed in the embolism models used in the present study [17‐19].

Volumetric capnography also enables the calculation of Fowler’s anatomic dead space (VDF), Bohr’s physiological dead space (VDB), and Enghoff’s modified physiological dead space (VDE). VDF was defined as the exhaled gas volume until the inflection point in phase 2 [20]. VDB was calculated as [21]:where the mean alveolar partial pressure of CO₂ (PACO₂) was defined as the CO₂ concentration at the midpoint of phase 3 in the volumetric capnography curve, and the mixed expired CO₂ partial pressure (PĒCO₂) was estimated by dividing the integrated capnogram curve by VT in each expiratory cycle.

VDE, which also comprises nonventilated but perfused alveoli, also known as intrapulmonary shunting, was obtained as follows [22, 23]:where PaCO₂ is the partial pressure of CO₂ in an arterial blood sample.

The alveolar dead space fraction (AVDSf) was also determined as (PaCO₂-ETCO₂)/PaCO₂ [24].

In these preliminary experiments, 4 and 2 animals were used to examine the hemodynamic and respiratory responses to air injections of 0.2 ml and 0.15 ml, respectively. However, the responses observed in these pilot rats were excessively severe, resulting in lethal outcomes for 3 rats. These preliminary findings were the basis of the decision to use a 0.1 ml air bubble in the main protocol group.

Following anesthesia induction and surgical preparations, including the thoracotomy, the rats were ventilated as described above with a PEEP of 5 cmH₂O throughout the study. After stabilizing vital parameters, a lung recruitment maneuver was performed by closing the expiratory limb of the ventilator tubing until the next expiration to standardize lung volume history. Baseline data were collected 1–2 minutes later by recording capnograph curves and forced oscillation data, along with blood gas analyses of arterial blood samples. Pulmonary air embolism (AE) was then induced by injecting 0.1 mL of air (0.5% of total blood volume) mixed with 0.9 mL of saline into the femoral vein while continuously monitoring the changes in systemic hemodynamics and the capnogram. This injected air volume was titrated in pilot experiments to exert similar changes in ETCO₂ compared to those observed after unilateral pulmonary artery occlusion, yielding comparable gas exchange defects assessed in the expired gas (see above). When a peak decrease in ETCO₂ was observed (10–15 seconds after the air injection), an additional data set was recorded (condition AE). The effects of AE on systemic hemodynamics and gas exchange parameters lasted approximately 2–4 minutes and then returned to the normal range after approximately 5 minutes, as verified by the continuous monitoring of relevant vital signs. After a 10–15-min stabilization period, the left pulmonary artery was clamped using a small-sized surgical clip, and a third data set, including capnography, forced oscillations, and blood gas analyses, was collected under this condition, i.e. during left pulmonary artery occlusion (LPAO).

Data are presented as the median with interquartile ranges in each boxplot chart. The Shapiro–Wilk test was used to evaluate the data distributions for normality. For each parameter studied, one-way repeated-measures ANOVA with Holm–Šidák post hoc tests was applied to assess significant differences between the results obtained in the protocol phases. Statistical analyses were performed with the SigmaPlot software package (Version 14, Systat Software, Inc., Chicago, IL, USA).

Since changes in lung tissue stiffness were anticipated after embolism, H was considered the primary outcome variable for estimating sample size using one-way repeated-measures ANOVA, with an effect size (f) of 0.4, power of 0.8, and significance level of 0.05. This analysis performed using G*Power (version 3.1.9.7, Universität Düsseldorf, Germany) indicated that at least nine animals were required to detect a 15% significant difference in the primary outcome when considering the previously observed variability [25]. Statistical analyses were performed at a significance level of P < 0.05.