Design

The effects of NMB and pre-stretching were investigated in an established porcine model for abdominal insufflation [20]. Before each of the 30 experiments, subjects were randomized to three groups: complete NMB, moderate NMB, or no NMB (control). The sample size of ten animals per group was determined by establishing a detection limit of 20% or 500 mL gain in abdominal workspace volume. This calculation was based on a two-sided t test with a significance level of 0.05 and a statistical power of 80%, assuming a normal distribution. Initially considering a group size of 9, it was subsequently adjusted to 10 per group to safeguard against the potential loss of statistical significance resulting from unforeseen events, such as the death of an animal during the study.

In each animal, the abdomen was insufflated three times: initial insufflation and two repetitions. During each repeated insufflation, the insufflation pressure was increased stepwise to create detailed compliance curves. The IAP was sustained at 0, 3.75, 6, 7.5, 9, 10.5, 12, 13.5, and 15 mmHg, respectively. This range included 0 mmHg to acquire a baseline, and 3.75 mmHg to investigate what happens when the insufflation pressure is equal to the PEEP set on the mechanical ventilator (3.75 mmHg = 5 cmH2O). The other values were chosen to systematically cover the range of insufflation pressures used in clinical practice. At every step, the insufflation pressure was sustained for 3 min to allow the subject time to adjust to the new insufflation pressure. After this accommodation period, a CT scan was obtained during an end-expiratory breath-hold.

Rocuronium was titrated according to the NMB level defined by randomization. The level of NMB was maintained based on both train of four (TOF) and post-tetanic count (PTC) measurements. TOF and PTC were measured by stimulating the adductor muscles of the lower extremities. Post-tetanic depletion of the neuromuscular junction was avoided by measuring TOF and PTC on two different extremities. Rocuronium was infused at the jugular and femoral veins.

Simultaneous infusion at both veins was preferred because pilot experiments showed a decreased NMB measured at the lower extremities during abdominal insufflation, likely due to compression of the lower vasculature and subsequently decreased perfusion.

The following levels of NMB were used according to the definition described by Biro et al. [21]:

(1)

Complete NMB: TOF = 0/4 PTC < 1/10

(2)

Moderate NMB: TOF = 1/4–3/4 and PTC = 10/10

(3)

Control without NMB: TOF = 4/4 and PTC = 10/10

The level of NMB was verified after every CT scan using TOF. For every two CT scans, the PTC level was verified. If needed, the infusion rate of rocuronium was adjusted to keep the level of NMB within the predefined range.

Subjects

Measurements were performed in a 20-kg female Landrace porcine model. The animals were obtained from a specific pathogen-free farm. An enriched environment was provided and, if logistics allowed, animals were kept in groups. During a one-week accommodation period, the institute’s animal facility provided care and animals had free access to water and food. On the day of the experiment, the animals only had access to water. Animals were excluded if the cardiorespiratory physiology was affected by anatomic abnormalities. This study was registered at the Dutch Central Authority for Scientific Procedures on Animals and registered under license number AVD101002015180. Institutional approval was given by the Animal Welfare Body of Erasmus MC, University Medical Center Rotterdam, protocol number 15-180-02.

Preparations

Initial sedation was provided via intramuscular injection with midazolam (40 mg/kg), ketamine (1 mg/kg), and atropine (0.03 mg/kg). Fifteen minutes were given to ensure the onset of sedation. Instrumentation started by cannulation of the marginal ear vein (20 gauge). For mechanical ventilation, the animal was intubated using an endotracheal tube (size ~ 7 mm), and lidocaine spray was used to suppress the cough reflex. The animal was weighed before placement in a supine position onto the measurement CT slide. A mechanical ventilator (Fabian HFO, ACUTRONIC Medical Systems AG, Hirzel, Switzerland) was connected and set to volume guarantee mode with the tidal volume set to 7.5–8.0 mL/kg. For maintenance of anesthesia during instrumentation, propofol and sufentanyl were provided through the cannulated ear vein.

To ensure repeatable measurement conditions, mechanical ventilation was managed using an automated system throughout the measurement protocol [22]. The automated system was compatible with tidal volume guarantee mode. Oxygenation was maintained by adapting the fraction of inspired oxygen. The carbon dioxide levels were managed by altering the respiratory rate to target an end-tidal pCO2 of 7.0 kPa (52.5 mmHg).

Arterial and central venous lines were placed for hemodynamic monitoring. After this, both NMB monitors were placed on both lower extremities (Dräger TOFScan, Drägerwerk AG & Co. KGaA, Lübeck, Germany).

A 10-mm trocar (VersaOne™, Medtronic, Minneapolis, USA) was placed at the subumbilical midline. After insertion, intraperitoneal placement was verified endoscopically. For creating the pneumoperitoneum, a commercially available CO2 insufflator was used (Endoflator 40, Karl Storz SE & Co. KG, Tüttlingen, Germany) with an inline custom-built device for high-frequency measurement and additional pressure control.

MeasurementsIntra-abdominal volume

CT scans were obtained using a Somatom Force scanner (Siemens Healthcare GmbH, Erlangen, Germany) and reconstructed with a 1 mm slice thickness. The IAV was measured using Myrian imaging software (Version 2.6.5 Research Edition, Intrasense, Montpellier, France). For this, the surgical workspace was automatically segmented, visually checked, and manually corrected.

Respiratory pressures

To analyze changes in respiratory mechanics, PEEP and tidal volume were kept constant such that the peak inspiratory pressure (PIP) reflects changes in respiratory compliance. The hemodynamic response was evaluated based on heart rate, blood pressure, and cardiac output. These were monitored using a hemodynamic patient monitor (PulsioFlex monitor, Getinge AB, Göteborg, Sweden). Samples from the mechanical ventilator and hemodynamic patient monitor were recorded at a one-second interval. The sample at 5 s before the end-expiratory breath-hold needed for the CT scan was used for further analysis.

Arterial pressure, heart rate, and cardiac output

The effects of NMB, repetition, and the insufflation step on the circulatory system were investigated using mean arterial blood pressure (MAP), measuring heart rate (HR), and cardiac output (CO). These parameters were sampled simultaneous to the respiratory pressures.

AnalysisAbdominal mechanics

An existing model for the evaluation of abdominal compliance was used for the analysis [23]. This model requires the baseline pressure of 0 mmHg to be omitted. Figure 1 shows an example of the measured IAV and the estimated abdominal pressure–volume curve. The model is based on the assumption that there are parameters which defines the pressure–volume relation:

1.

The maximum IAV, Vmax rises, this shifts the horizontal asymptote up.

2.

The abdominal cavity opens up at a lower insufflation pressure. The opening pressure, p0, shifts to the left.

3.

The curvature of the pressure–volume relation changes, λexp. A steeper increase in IAV takes place due to which the asymptotic Vmax is reached at a lower IAP.

Fig. 1

Abdominal pressure–volume curves in three different subjects (▲, ●, and ■) and estimated parameters. The corresponding models (–) are extrapolated to provide a visual explanation. Vmax relates to the horizontal asymptote and p0 relates to the location on the horizontal axis when the volume equals zero. The curvature is described by λexp

In a previous study [23], parameters Vmax, p0, and λexp were estimated using an empirical model:

$$IAV\left(p\right)=_-\frac_}^_\bullet (IAP-_)}}$$

This model was fitted to the pressure–volume curves of each insufflation run using Matlab (R2023b, Mathworks, Natick, Massachusetts, U.S.). Each parameter was tested using a linear mixed model with NMB group and insufflation repetition as fixed effects and subject number as random effects. The analysis was performed in R Studio (2022.07.2, R Foundation for Statistical Computing, Vienna, Austria). Linear mixed models were estimated for each of the three outcome parameters Vmax, p0, and λexp. The independent variables (fixed effects) in the linear mixed models were NMB group (control, moderate, or complete) and insufflation run (REP). A random intercept of individual animals was included in the model to account for the within-subject correlations.

Cardiorespiratory effects

The cardiorespiratory effects investigated are PIP, MAP, HR, and CO. For each animal, these variables were repeatedly measured per insufflation run (REP) and pressure level at every insufflation step (STEP). For the analysis of the peak inspiratory pressure, an insufflation pressure of 0 mmHg was included in the analysis. To account for the structure of the data, a linear mixed effects model was developed with as independent variables NMB group, insufflation run and pressure level (as a categorical variable), and all two-way interactions between these three variables. A random intercept and random slope of the pressure level (as a continuous variable) were included for each animal and each insufflation run of each animal. The model was run separately for each cardiorespiratory effect. The results of the model are presented using the estimated marginal means, which are the predicted values of the outcome after adjusting for the effects of independent variables.