Effect of Oxygen Pneumoperitoneum on Hypoxia Tolerance in Beagle Dogs: An Experimental Study CURRENT

Background: When an airway is difficult to manage during surgery, it can be fatal. Thus, peritoneum oxygenation may be theoretically feasible as an alternative route to supply oxygen. The objective of this study is to investigate whether artificial oxygen pneumoperitoneum combined with abdominal lifting and compression is effective in systemic oxygenation and hypoxia tolerance in a beagle dog model of asphyxia. Methods: Latin Square design was adopted. Eight adult female beagle dogs were divided into four subgroups; they received four treatments in sequence in four sessions at 1-week intervals. The pre-treatment procedures included general anesthesia and a pause in ventilation to establish an asphyxia model. Each subgroup was subjected to the treatments as follows: negative control(A), artificial oxygen pneumoperitoneum(B), artificial oxygen pneumoperitoneum combined with abdominal lifting and compression from the 1st minute after asphyxia(C), or the same as C but starting the method from the 5th minute(D). Vital signs, important organ functions, and arterial blood gas were monitored. Hypoxia tolerance was assessed by the asphyxia time (from pause to restoration of ventilation) and PaO2 levels at 0, 6, 7, 8, 9, and 10 minutes after asphyxia onset. Results: The physiological status of the beagles was approximately identical before each experiment and the treatments did not aggravate the impact of hypoxia. The asphyxia time for treatments A, B, C, and D were 9.65 ± 1.10 min, 11.18± 0.50min, 11.45 ± 0.79min, and 11.70 ± 0.99min, respectively; the time was lengthened in treatments B, C, and D, compared with that in treatment A (P<0.01). Moreover, PaO2 levels at 9 and 10 min were higher in treatments B, C, and D than in treatment A (P<0.01). Conclusions: Artificial oxygen pneumoperitoneum combined with abdominal lifting and compression appears safe and effective and may improve systemic oxygenation and


Background
During emergency rescue and anesthesia procedures, maintaining the air way is a top priority, as prompt and efficient oxygen supply is required to sustain the patient's life. [1,2] However, it can be challenging to administer oxygen when managing a difficult airway caused by morbid obesity, maxillofacial trauma, or tracheal tumors; life-threatening difficulties in oxygen administration can occur, especially when encountering an emergency airway-unanticipated difficulties in either intubation or ventilation [3,4] .Thus, extra-pulmonary oxygenation represents an important alternative route to supply oxygen [5][6][7] .Thus by far, extracorporeal membrane oxygenation, intravascular oxygenator, and percutaneous cardio-pulmonary support have been applied for extra-pulmonary oxygenation; however, these procedures are complex, time-consuming, labor-intensive, and associated with many complications, due to multiple transfusions and systemic heparin administration [6][7][8][9][10] . Hence, researchers have attempted to find an easy, safe, and effective method by which clinicians can improve oxygen supply.
The peritoneum is a biological semi-permeable membrane comprising a large, absorptive, highly permeable surface with rich vascular beds. Therefore, it may be used for extrapulmonary oxygenation [11] . Previously, peritoneal oxygenation has been shown to moderately increase systemic arterial partial pressure of oxygen in rabbit, rat, and pig models [5,[12][13][14] via direct abdominal ventilation or intra-peritoneal infusion with oxygen carriers, such as perfluorocarbon, red blood cells, or oxygen micro-bubbles [5,8,12,13,[15][16][17][18][19] .Hence, peritoneal oxygenation has long been desired for extra-pulmonary oxygenation, as it is easy to perform, compared to use of an extracorporeal circuit.
Based on the above findings, oxygenation status may be improved by artificial oxygen pneumoperitoneum. In cases of hypoxia, oxygen saturation in the splanchnic bed may be reduced considerably; hence, additional oxygen could be absorbed from the peritoneal cavity. Additionally, recent research has suggested that abdominal lifting and compression could promote blood circulation in abdominal viscera and facilitate breathing by enhancing diaphragm movement [20,21] . Thus, the objective of this study was to investigate the effect of artificial oxygen pneumoperitoneum combined with abdominal lifting and compression on systemic oxygen supply and hypoxia tolerance in a beagle dog model of asphyxia.

Animals
Healthy adult female beagle dogs (n=8) weighing 8-10kg were purchased from the Academy of Military Medicine and Science of Chinese People's Liberation Army(PLA) (Beijing, China: Permit Number: SCXK(JUN) 2014-0001). All beagles were housed individually in accordance with guidelines for indoor pets at the Experimental Animal Unit of PLA General Hospital, and were given access to standard food and water at scheduled times in an air-conditioned room set at 24±1°C with 50±10% humidity, under a standard 12-12 hour light-dark cycle(light on from 7 AM to 7 PM). Animals were housed in the same cage throughout the experimental period. Adequate measures were taken to minimize animal discomfort.

Study Design
We investigated four treatment modalities in this study:(A)control, in which dogs received abdominal puncture alone with a Veress needle, but no additional treatments after the asphyxia model;(B) pneumoperitoneum (Pneu), in which dogs received artificial oxygen pneumoperitoneum beginning at1 minute after the asphyxia model;(C) early pneumoperitoneum plus abdominal lifting and compression(P+1), in which dogs received artificial oxygen pneumoperitoneum combined with abdominal lifting and compression, which was started at 1 minute after asphyxia modeling began; (D) late pneumoperitoneum plus abdominal lifting and compression(P+5), in which dogs received artificial oxygen pneumoperitoneum combined with abdominal lifting and compression beginning at 5 minutes after asphyxia modeling was started. Beagle dogs were numbered by weight and then divided into four subgroups based on weight-ranging from light to heavy. During the 4-week experimental period, each dog within each subgroup received all four treatments in a different sequence from one another; between treatments, a 1-week washout period was used to allow physical recovery to baseline. To investigate the effects of experimental time, individual dog, and treatment modality, a 4×4 Latin Square design was utilized (Table 1). vital signs such as SpO 2 , heart rate(HR), invasive arterial blood pressure, and cardiac output(CO), were monitored continuously throughout the procedure(V24E, Philips Inc., Amsterdam, The Netherlands; Vigileo-FloTrac™ system, Edwards LifeSciences Inc., Irvine, CA, USA). After a baseline ventilation period of 5 minutes, the asphyxia model was established by clamping the endotracheal tube. The dogs were then subjected to treatments A, B, C, or D, as described above. The artificial oxygen pneumoperitoneum was established in accordance with the clinical practice used in laparoscopic surgery [22] , and the pressure was set at 12cmH 2 O (model: PG120, Aesculap Inc., Tuttlingen, Germany).
Abdominal wall lifting and compression [23] was performed at a frequency of 60bpm, a ratio of 1:1, and a strength of 50N. Arterial blood gas (ABG) analyses were performed at baseline and at 6, 7, 8, 9, and 10 minutes, respectively, after the endotracheal tube was clamped, as well as at the time of asphyxia termination(model: GEM Premier 3000, Instrumentation Laboratory, Lexington, MA, USA). Asphyxia was terminated by administering 100% oxygen when one of the following criteria was met: 1)SpO 2 was <50%, (2)mean arterial pressure(MAP) was twice the baseline value, or (3) malignant arrhythmia occurred [24] . Asphyxia time was recorded, as was recovery time-the time at which the 6 vital signs were stabilized. Then, the pneumoperitoneum was deflated and anesthesia was discontinued. Dogs were returned to their cages upon extubation and awakening. Venous blood was collected before the experiment, as well as at 2 and 24 hours after the experiment, and was analyzed to assess hepatic and renal function, as well as oxidative stress-related enzymes. After a 1-week washout period, the dogs underwent the next round of the experiment, as shown in Table 1. After all rounds of the experiments, the dogs were euthanized. Placed in a quiet and comfortable position, the dogs were deeply anesthetized and monitored as above, and then euthanasia solution(Beuthanasia-D Special, 86mg/kg) was injected. Measures were taken to minimize the stress.

Statistical analysis
All data were analyzed by an observer who was blinded to the study protocol. Statistical calculations were performed using SPSS 17.0 (SPSS, Inc., Chicago, IL, USA). Data were expressed as the mean ± standard deviation. Intergroup comparisons were conducted using Latin Square ANOVA, followed by the least significant difference-t post hoc test to determine significant differences among treatments. Data regarding ABG analysis, HR, MAP, and other factors were analyzed using two-way ANOVA(treatment × trial time) with repeated measures followed by a Bonferroni post hoc test to analyze differences among treatment groups. P values <0.05 were considered statistically significant.

Results
3.1. Physiological and biochemical statuses in each group were nearly identical before each round of the experiment As described in the experimental protocol, the dogs were subjected to a severe hypoxic attack-nearly lethal-which resulted in damage to important organs, such as the heart, liver, and kidneys. Therefore, we included a 1-week interval between each round of the experiment to allow for physical recovery from the hypoxic attacks and to eliminate any lingering effects of the previous experiment. According to the Latin Square ANOVA results, all dogs were in a good state of health; no significant differences were observed among treatment groups with regard to HR, MAP, SpO 2 , CO, PaO 2 , PaCO 2 , pH, other indexes of ABG, or hepatic and renal functions before each round of experiments (Table 2). These findings suggested that the physiological and biochemical statuses of the dogs returned to baseline during the1-week intervals; additionally, there were no significant differences between experimental times or between individual dogs in terms of these indexes (Table   7 2).  1D).

Artificial oxygen pneumoperitoneum combined with abdominal lifting and compression
increased PaO 2 PaO 2 is a reliable indicator of oxygen content. Based on our pre-experiments, PaO 2 declined at a similar velocity among groups during the first 5 minutes of asphyxia, and the termination of asphyxia standard was met in some dogs after 10 minutes of asphyxia.
Hence, PaO 2 and PaCO 2 were detected and compared at every minute beginning from the 6th to the 10th minute in this study ( Fig. 2A and 2D). In Figure 2A and 2D, repeated measures two-way ANOVA revealed a difference in PaO 2 levels among groups(P<0.05),and no difference in PaCO 2 levels among groups(P>0.05). PaO 2 reduction in the Control group was greater than that in other groups beginning from8 minutes after asphyxia; PaCO 2 11 increased in a nearly linear fashion in all groups( Fig. 2A and 2D). Accordingly, we compared PaO 2 values at the 9th and 10th minutes ( Fig. 2B and 2C)  These findings suggested that artificial oxygen pneumoperitoneum could increase PaO 2 levels and that abdominal lifting and compression could further increase PaO 2 levels during asphyxia.

Changes in HR and MAP during asphyxia
To investigate the safety and effect of this method on hemodynamics, we compared changes in HR and MAP in each group every 2 minutes, from baseline to the 10th min after asphyxia (Fig. 3). HR appeared to initially accelerate, then slowed (Fig.3A). No differences were observed among the groups (Fig. 3A and 3C),although HR in the Control group tended to be lower than that in other groups at 10min (Fig. 3A and 3C). MAP appeared to increase in the Pneu, P+1, and P+5 groups; in the Control group, it tended to initially increase, then decrease. Although no differences were observed among the groups (Fig. 3B and 3D), MAP at 10min in the Control group tended to be lower than that in other groups ( Fig. 3B and 3D).

Artificial oxygen pneumoperitoneum combined with abdominal lifting and compression
did not alleviate or aggravate the impact of hypoxia To investigate the effects of hypoxia on hepatic and renal function, we assessed enzymes related to hepatic and renal function and oxidative stress at baseline, as well as at 2h and 24h after each experiment (Fig. 4). There were no differences in changes in ALT, AST, Crea, BUN, LDH, CK-MB, SOD, or MAO levels among groups. BUN, LDH, and SOD exhibited differences over time (Fig. 4D, 4E, 4G). After the experiment, BUN and LDH increased, while SOD decreased; however, all were within normal ranges. This indicates that the dogs had experienced hypoxic injuries, all of which were reversible. Moreover, the method did not appear to alleviate or aggravate this effect of hypoxia.

Discussion
In this study, we investigated a method of extra-pulmonary oxygenation via peritoneum: artificial oxygen pneumoperitoneum combined with abdominal lifting and compression.
Overall, artificial oxygen pneumoperitoneum could prolong the hypoxia tolerance time of asphyxiated beagle dogs (Fig. 1B and 1C); its combination with abdominal lifting and compression could further increase PaO 2 ( Fig. 2A-C). Thus, this method may safely improve whole-body oxygenation status( Fig. 3 and 4).
Emergency airway and unpredictable difficult airway are not uncommon during anesthesia induction; these situations are extremely critical and difficult to manage appropriately. In addition, several case studies have reported the occurrence of complete airway obstruction during awake intubation [25] . Therefore, we aimed to identify other methods to supply oxygen promptly, particularly when encountered with above situations, in order to facilitate resuscitation, thereby providing additional time for tracheotomy.
Previous studies have demonstrated that oxygen supply via peritoneum is feasible by direct abdominal ventilation or intraperitoneal infusion with oxygen carriers [5,8,12,13,[15][16][17][18][19] .However, these methods are complex and cannot be performed immediately; therefore, they are not suitable for emergency resuscitation in the operating room.
Our method has both advantages and disadvantages compared with other peritoneal oxygenation methods. First, artificial pneumoperitoneum is a common technique that can be established rapidly in most hospitals, and abdominal lifting and compression is an easy-to-learn resuscitation procedure; this combination could be rapidly implemented to contribute to the establishment of invasive airway. Second, our experimental animals were beagle dogs, whose abdominal physiology is similar to that of humans; thus, the study is clinically applicable. Third, we adopted the Latin Square design, which enabled observation of the effects of different experimental times and individual dogs on the results. However, a disadvantage is that this method provided limited oxygen delivery, such that it increased PaO 2 by 13mmHg at the 10th min of asphyxia (Fig. 2C). However, a 13-mmHg increase in PaO 2 in critically hypoxic patients may be critical for resuscitation.
In pre-experiments, we explored appropriate pressures for the artificial pneumoperitoneum, and found that excessive or minimal pressures could not improve the dogs' hypoxia tolerance. Hence, we chose a moderate pressure of 12cmH 2 O in this study.
This pressure was appropriate for improving oxygenation status (Fig. 1C);possibly due to improved systemic oxygenation, our method may delay the onset of severe cardiac complications( Fig. 3A and 3B).
In this study, we used several criteria to determine when to terminate asphyxia and restore mechanical ventilation, which included SpO 2 <50%, MAP twice that of the baseline, or the occurrence of malignant arrhythmia. These were used to ensure thorough investigation and determine accurate pathological reversible time. To evaluate the suitability of these criteria, the recovery time and PaO 2 at the time of asphyxia termination were analyzed; the results showed no differences among the four treatments( Fig. 1A and 1D). Furthermore, we assessed several biochemical parameters, including enzymes related to hepatic and renal function. We found that LDH, SOD, and BUN levels changed after anoxia; notably, these indexes were restored to normal within 24 hours (Fig. 4). Therefore, all dogs might have experienced a similar level of hypoxia, and the different asphyxia times could reflect variations in hypoxic tolerance.
There were some limitations in our study. To ensure animal safety, we only compared PaO 2 at 9 and 10 min after asphyxia; we did not observe peritoneal pathological changes.
Additionally, we did not explore methods to reducePaCO 2 , and did not compare the occurrence of hypercapnia. Future studies should include investigation of the specific mechanism by which PaO 2 was increased, methods to further improve PaO 2 , and the feasibility of this method in clinical practice.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analyzed during this study are presented within the additional supporting files.

Competing interests
The authors declare that they have no competing interests.

Funding
This study was conducted with the support of the National Natural Science Foundation of