Pneumothorax in a Thiel cadaver model of cardiopulmonary resuscitation

INTRODUCTION

There has been some previous research in the field of cardiopulmonary resuscitation (CPR) using cadaver models.[1] The Thiel method was developed by Professor Walter Thiel (Graz, Austria) and described in 1992 and 2002. It consists of both an intravascular injection of the embalming solution and submersion of the bodies in a tank with the same solution for a determined period.[2,3] Due to preserved tissue flexibility and elasticity and lifelike anatomical aspect, bodies embalmed by this method are used in training for medical procedures.[4-6] Respiratory mechanics and dynamic pressures during CPR in the Thiel cadaver model were found to be comparable with those of out-of-hospital cardiac arrest patients.[7]

Chest injuries are mostly inevitable when performing chest compressions for CPR. Several clinical studies have demonstrated the type and prevalence of CPR-associated injuries that can be potentially life-threatening.[8-10] This includes pneumothorax, an entry of air into the pleural space. Diagnosis is traditionally made by chest radiography but can also be done by lung ultrasound and computed tomography (CT).[11,12] The prevalence of CPR-associated pneumothorax varies widely in the current literature, and a review with pooled analysis states a total rate of 2.5%.[9] Studies using CT following CPR found higher rates of up to 20.1%.[10,13]

Tension pneumothorax is present when there is increased pressure in the pleural space due to a one-way valve mechanism that allows air to enter the pleural space with every breath but not to escape. In addition to respiratory impairment, there is also a collapse of major blood vessels and hemodynamic instability. Therefore, tension pneumothorax is a life-threatening emergency that requires immediate diagnosis and treatment. In European Resuscitation Council Guidelines for Resuscitation 2021, tension pneumothorax is among the potentially reversible causes of cardiac arrest.[14]

To the best of our knowledge, pneumothorax in simulated CPR using a cadaver model has not been previously described. In our study, we aimed to find out if, and how often pneumothorax and tension pneumothorax occur in simulated CPR using bodies embalmed by the Thiel method.

METHODS

The investigation of pneumothorax was conducted during the development and execution of a study protocol to evaluate the performance of three common transport ventilators during continuous chest compressions. In this study, MEDUMAT Standard² (WEINMANN Emergency Medical Technology GmbH+ Co. KG, Germany), Oxylog 3000 plus (Drägerwerk AG & Co. KGaA, Germany) and Monnal T60 Air Liquide Medical Systems (Antony Cedex, France) were tested to determine if they can achieve sufficient alveolar ventilation under simulated CPR conditions.[15]

Cadavers

Eleven cadavers were randomly selected, which did not show any signs of pre-existing thoracic injuries. Written informed consent was obtained from all body donors to donate their bodies for educational and scientific purposes. All bodies donated to science were investigated under the strict rule of the donation program of the Institute of Macroscopic and Clinical Anatomy of the Medical University of Graz and the Styrian burial law. Hence, no additional approval by the local ethics committee was needed.

Study procedure

The investigation followed a standardized protocol (Figure 1). First, cadavers were orally intubated by direct laryngoscopy. Bronchoscopy (aScope™ 4 Broncho Large, Ambu™, Denmark) was performed to verify tracheal tube positioning and to remove any fluid collections or residues of the embalming process.

Figure 1.

Experiment flowchart. The experiment was conducted following a standardized protocol which is shown in this illustration. Adapted from Orlob et al.[15] P_TOP: top pressure; V_T: tidal volume; f: frequency; PEEP: positive end-expiratory pressure; I:E: inspiration/expiration ratio; P_MAX: upper pressure limit; CPR: cardiopulmonary resuscitation.

Initially, an intensive care respirator (Hamilton-C6, Hamilton Medical, Switzerland) for ventilation was used. Body height was measured and predicted body weight (PBW) was calculated. Lung recruitment was carried out by two quasistatic inflation manoeuvres and a 15-minute period of volume-controlled ventilation. For the first three cadavers, the applied top pressure of the 2^nd inflation manoeuvre was 35 cmH₂O (1 cmH₂O=0.098 kPa) but was later reduced to 30 cmH₂O due to the development of tension pneumothorax in all three cadavers.

In the course of the ventilation study, we assessed mechanical properties in three 2-minute sequences (ventilation only, chest compressions only, combined). Chest compressions were performed in a standardized way using an automated chest compression piston device (Corpuls CPR, GS Elektromedizinische Geräte G. Stemple GmbH, Germany) with a frequency of 103/min and depth of 5 cm.

In the intervention phase, the three ventilators were tested in a three-period crossover design. Each period consisted of two 2-minute cycles of simulated CPR. The order in which the ventilators were applied was randomly assigned by sealed envelopes to the individual cadaver.

Testing for pneumothorax, defined by the absence of lung sliding, was performed by bilateral lung ultrasound (Sonosite M-Turbo, FUJIFILM Sonosite GmbH, Frankfurt, Germany) by two clinicians trained and experienced in eFAST (extended focused assessment with sonography for trauma). This was performed after the initial recruitment phase and after every 2-minute cycle of chest compressions. We used a phased array probe that was positioned on the highest point of the anterior chest wall in a sagittal position, usually represented by the 3^rd or 4^th intercostal space in the mid-clavicular line.[16]

The whole course of the experiment consisted of 17 min of ventilation only, 2 min of chest compressions only and 14 min of ventilation and chest compression combined. Finally, bilateral finger thoracostomy was performed with particular attention given to the occurrence of tension pneumothorax, defined by the escape of air when performing thoracostomy.

Statistical analysis

Statistical analysis was performed using IBM SPSS Statistics 27. Continuous data were tested for conformation with a normal distribution using the Kolmogorow-Smirnow test and were compared using an independent samples t-test. Categorical variables were compared using Fisher’s exact test. A P-value ≤0.05 was considered statistically significant.

RESULTS

In total, we were able to detect pneumothorax in 8 out of 11 cadavers (72.7%), 4 bilaterally (36.4%), 2 left-sided (18.2%) and 2 right-sided (18.2%). Pneumothoraces were detected in different phases of the protocol, and the median cumulative chest compression time until detection of the first pneumothorax was 12 min (interquartile range 9.5, min–max 0–16) (Table 1).

Table 1.

Characteristics and results of the individual cadavers

Statistical testing for associations between cadaver variables (height, recruitment pressure, sex) and outcomes (any pneumothorax, tension pneumothorax) resulted in only one statistically significant finding. Tension pneumothorax was observed in all 3 cadavers which were subject to 35 cmH₂O recruitment pressure. In contrast, we had this finding in only 1 out of 8 cases with the 30 cmH₂O recruitment pressure. This association was found to be statistically significant (P=0.024).

DISCUSSION

Our results showed a pneumothorax rate of 72.7%, which is considerably higher compared to clinical data which show rates of up to 20.1%.^[8–10,13] As previously demonstrated we found cadavers embalmed by the Thiel method suitable for simulating CPR to study mechanical and respiratory properties.[7] Additionally, lung ultrasound proved to be a suitable diagnostic method for pneumothorax in the Thiel cadaver model.

Bodies embalmed by the Thiel method are known for their close to in-vivo characteristics of tissues.[1-3] Based on our results it seems plausible that embalming makes the lung less resilient against the mechanical stress from chest compressions and ventilation. Thus, the transferability of our results to the clinical setting is highly limited. Recruitment pressure also seems to play a role. The 3 cadavers that were subjected to a higher recruitment pressure (35 cmH₂O instead of 30 cmH₂O) showed bilateral tension pneumothorax, with 2 of them being detectable after the 15-minute recruitment phase with no chest compressions. This led to a reduction in the maximum applied recruitment pressure in the study protocol. We are aware that this presents a limitation of the current study, however, we also consider this to be one of the key lessons learned from this experiment.

Additionally, the study protocol was primarily developed for the ventilator study which explains some more limitations (e.g., 4 different ventilators used, sample size, cross-over design).[15] There is insufficient evidence to support or refute the use of a mechanical ventilator for cardiopulmonary resuscitation, and current guidelines recommend the use of a bag-valve mask device.[17] However, we found it important to use standardized ventilatory settings in our experiment, which were chosen in accordance with applicable guidelines and experts’ opinions.[14,18]

We were unable to evaluate the medical history of the body donors which may have included pulmonary conditions. A mechanical chest compression device was used to standardize chest compressions; however, there is some literature stating a higher incidence of CPR-associated injuries when using mechanical chest compression devices.[19,20] This could have contributed to the higher occurrence of pneumothorax observed in our experiment. We are also aware that CT has superior sensitivity in detecting pneumothorax compared to lung ultrasound.[21] In our experimental setting, CT was not available.

CONCLUSIONS

CPR-associated pneumothorax is a phenomenon that occurs in simulated CPR using the Thiel cadaver model. The pneumothorax rate was found to be substantially higher compared to clinical data. Higher recruitment pressure seems to contribute to the development of tension pneumothorax. Future studies using Thiel cadavers for simulated CPR should account for these findings in their study protocols.