Abstract
Introduction
To establish a decompression sickness (DCS) model in New Zealand white rabbits and a quantitative method for assessing bubble load in DCS by ultrasound.
Methods
Fifteen male New Zealand white rabbits were exposed in a hyperbaric chamber for 60 min with air compressed to 500 kPa (absolute pressure) at 100 kPa/min, followed by decompression at a rate of 200 kPa/min back to normal pressure. Behavioral changes were observed 10 min after removing the rabbits from the chamber to assess whether a model with DCS was successfully established. Bubbles in the inferior vena cava (IVC) and right ventricle (RV) were detected by ultrasound and semi-quantitatively graded using the Eftedal-Brubakk (EB) grade. One hour after exiting the chamber, the rabbits underwent autopsy to observe macroscopic bubbles in vessels and tissues for postmortem (PM) scoring. Correlations between EB grading by ultrasound and PM bubble scores were analyzed.
Results
The decompression protocol yielded a 100% DCS incidence (15/15) with 13.3% mortality (2/15) within 1 h after the rabbits were removed from the chamber. Ultrasound revealed bubble-like hyperechoic foci in the IVC and RV after decompression, with significantly higher EB grades in the IVC (p < 0.05). IVC's EB grades exhibited a stronger correlation with its PM bubble scores (r = 0.921, p < 0.01) compared to RV grades (r = 0.573, p < 0.05).
Conclusion
A reliable DCS model was established in New Zealand white rabbits. EB grading of the IVC and RV by ultrasound demonstrated a robust correlation with their PM bubble scores, suggesting potential for clinical translation in assessing bubble loads in DCS.
Introduction
Decompression sickness (DCS) is a systemic disorder caused by the formation of bubbles in blood or tissues when the previously dissolved gas exceeds the local saturation due to inadequate decompression after deep diving.1,2 Clinically, DCS is classified into Type I and Type II. Type I decompression sickness included pain, cutaneous manifestations, and constitutional symptoms, whereas Type II manifestations included numbness, tingling, paresthesia, muscle weakness, paralysis, and mental or motor abnormalities. 3 Studies have indicated a positive correlation between bubble load and the risk of developing DCS,4,5 making bubble load an objective indicator for evaluating DCS after diving.
Ultrasound, a noninvasive technique based on the propagation and reflection of high-frequency sound waves, can detect bubbles in blood and tissues. 6 Currently, the widely acknowledged method for quantifying bubble load is the Eftedal-Brubakk (EB) scale proposed in 1997, 7 which some scholars consider more objective and reliable compared to other methods. 8 However, despite the EB scale providing an effective means of quantifying bubbles, the selection of the site to be monitored remains a critical issue in practical application. Previous selection of locations for detecting bubbles has primarily focused on the subclavian vein or right ventricle (RV).9,10 Nevertheless, the RV is susceptible to interference from the heart valve and other tissue movements, while bubbles in the subclavian vein originate solely from the upper limb, which cannot accurately reflect the systemic bubble load. Therefore, there is an urgent need to further investigate the diagnostic value of using the EB scale to grade different vascular sites detected by ultrasound for assessing the macroscopic bubbles in DCS.
In this study, we used ultrasound to compare macroscopic bubble quantities in the inferior vena cava (IVC) and RV of New Zealand white rabbits using the EB grade. Postmortem, we graded the detectable bubbles using the gas score proposed by Bernaldo de Quirós et al. 11 Our analyses aimed to assess the diagnostic efficacy of ultrasound in grading DCS bubbles and to determine whether the IVC or RV is more effective for bubble load detection.
Materials and Methods
This prospective study was approved by the Ethics Committee of our hospital.
Animals
Fifteen male New Zealand white rabbits, weighing 2–2.5 kg, were purchased from the Suibei Experimental Animal Breeding Farm [Animal Production License: No. SCXK 2020-0050].
Experimental Equipment
We used a DWC-II animal laboratory chamber (purchased from Weifang Huayao Oxygen Industry Co., Ltd), a GE portable color Doppler diagnostic ultrasound (Vivid q), and a linear array probe with a frequency range of 4–9 MHz (9L-RS)
Decompression Protocol
All rabbits were kept under standard laboratory conditions before the experiment to ensure sufficient food, water, and appropriate ambient temperature and were fasted 12 h before the experiment. Based on a previous study by Meng 12 (Table 1), the rabbits were placed in the hyperbaric chamber, where the pressure was gradually increased to 500 kPa (absolute pressure) at 100 kPa/min. Physiological and behavioral changes in the rabbits, including respiration and activity, were continuously observed and recorded in the hyperbaric environment for 1 h and then followed by decompression at a rate of 200 kPa/min back to normal atmospheric pressure (approximately 101.325 kPa). The chamber was maintained with continuous ventilation (flow rate of 1L/min with 21% oxygen and <0.3% carbon dioxide), a temperature of 22–24 °C, and a relative humidity of 65–75%.
Overview table for the study of the rabbit decompression sickness model.
Indicators for Evaluating DCS
After removal from the chamber, the rabbits were placed on the ground and observed for the development of DCS signs for 10 min. Like human DCS, animal DCS can be categorized into Type I and Type II according to signs. However, since animals lack the ability to express themselves, the differentiation is based on objective manifestations. Type I signs are mild, mainly showing flexion, restlessness, abnormal licking and gnawing movements, and limb pain, the latter of which is mainly deduced from physical signs including limb lifting or foot curling and vocalization; Type II signs are more serious, showing paralysis of one or more limbs, severe cardiopulmonary symptoms, and even death.13,14 For this study, DCS signs included hindlimb motor dysfunction and respiratory dysfunction (Table 2): (1) Hindlimb motor function was evaluated using the modified Tarlov scoring scale 15 : 0 = hindlimb paralysis; 1 = perceptible movement of hindlimb joints; 2 = brisk movement of hindlimb but unable to stand; 3 = able to stand; 4 = weak hop but not agile; 5 = normal ambulation. (2) Respiratory function was assessed with the modified Atkins grade scale 16 : 0 = respiratory failure, death; 1 = severely labored breathing, recumbent posture; 2 = restlessness, labored breathing; 3 = tachypnea, mildly labored breathing; 4 = normal respiration.
DCS signs of different severity in New Zealand white rabbits.
2D Ultrasound Detection
Ten minutes after removal from the chamber, an ultrasound examination was performed on the rabbit models with DCS to observe the 2D ultrasound images of the RV and IVC. Before ultrasonography, all rabbits were anesthetized with 0.05 mL/kg of intramuscular injections of xylazine hydrochloride (Dunhua Shengda Animal Medicine Co., Ltd, Qingdao, China), along with 0.05 mL/kg of tiletamine hydrochloride and zolazepam hydrochloride (Virbac Inc., Carros, France). The rabbits were then fixed in a stent to fully expose the thorax and abdomen, and the amount of air bubbles in the RV and IVC was observed by ultrasound. Bubble loads were semi-quantitatively graded using the EB scale 7 : 0 = no bubbles; 1 = occasional bubbles; 2 = at least 1 bubble per 4 cardiac cycles; 3 = at least 1 bubble per cardiac cycle; 4 = one bubble per cm2 in every image; 5 = “white-out,” single bubble cannot be discriminated.
Analysis of Bubble Load in Tissues
One hour after exiting the chamber, rabbits were euthanized using an overdose of xylazine hydrochloride, tiletamine hydrochloride, and zolazepam hydrochloride. Immediately after euthanasia, a necropsy was performed and the required tissue samples were collected for bubble analysis. Gas scores ranging from 0 to 6 were given to quantify bubble load in different blood vessels (subcutaneous, mesenteric, femoral, and coronary veins as well as the right atrium and inferior vena cava), and a gas score from 0 to 3 was used to assess the presence of gas beneath the capsule of different tissues (subcapsular emphysema) and within adipose tissues. 11 Details are shown in Tables 3 and 4. These scores were used to obtain a new gas score index to express the total gas score of each rabbit. Thus, the total gas score for each rabbit ranged from 0 to 42 (Table 4).
Definition of gas score index for postmortem examinations following Bernaldo de Quirós et al (2016).
Calculation of total gas score for each animal following Bernaldo de Quirós et al (2016).
Statistical Analysis
Statistical analysis was performed using SPSS 25 software. All quantitative data were presented as mean ± SD. For non-normally distributed data, the Wilcoxon rank-sum test was employed for comparisons between the two groups. The correlation between EB grades of the RV and IVC and their PM scores was analyzed using the Pearson correlation coefficient. Statistical significance was set at p < 0.05.
Results
Successful Establishment of a DCS Model in New Zealand White Rabbits
A DCS model was prepared using 15 New Zealand white rabbits, with an incidence rate of 100% (15/15) and a mortality rate of 13.3% (2/15) 40 min after the rabbits were removed from the chamber. Type I DCS was observed in 11 rabbits (73.3%, 11/15), characterized by restlessness and accelerated respiration. Four rabbits (26.7%, 4/15) showed typical Type II DCS, including hind limb paralysis, severe respiratory distress, and rapid death due to respiratory failure. The assessment of respiratory and motor functions in the rabbit model with DCS is shown in Table 2.
Assessment of DCS Models Using Ultrasound and EB Scale
For Type I DCS, relatively few bubbles were seen in the IVC and RV. In the case of Type II DCS, more bubbles were present in the IVC and RV, appearing as substantial aggregations of bubbles (Figure 1). Subsequently, bubbles in the IVC and RV were graded by EB scale (Table 5). The IVC of 5 rabbits (3/15, 20%) had an EB grade of 0–3, while the RV of 13 rabbits (13/15, 86.7%) had an EB grade of 0–3. The EB grade of 10 rabbits’ IVC (12/15, 80%) was 4 or 5, and the RV of 2 rabbits (2/15, 13.3%) was graded at 4 or 5 on the EB scale. Overall, the EB grades of the IVC were significantly higher than those of the RV (P < 0.05).
Ultrasound images of the right ventricle (RV) and inferior vena cava (IVC) in New Zealand white rabbits with DCS of different severity. Ultrasound images of Type I DCS showed occasional bubble formations in the RV (A) and IVC (C), as indicated by the white arrows pointing to the scattered small bubbles. In contrast, ultrasound images of Type II DCS revealed substantial bubble aggregations in the RV (B) and IVC (D), with the white arrows highlighting the dense curtains of bubbles present.
Comparison of the EB grade of the RV and IVC in the New Zealand white rabbit model with DCS.
EB Grade Measured by Ultrasound and PM Bubble Score
The New Zealand white rabbits with DCS were autopsied, and PM bubble scores were assigned according to the criteria established by Bernaldo de Quirós et al. 11 Three rabbits (3/15, 20%) had PM bubble scores of 0–10, 5 rabbits (5/15, 33.3%) were graded at 11–20, and the PM bubble scores of 7 rabbits (7/15, 46.7%) were 21–30 (Table 6). The gross anatomy of the rabbit whose EB grade and PM bubble scores were 4 and 29, respectively, is displayed in Figure 2. Afterward, a correlative analysis of the EB grades of the IVC and RV with their PM bubble scores demonstrated that both the EB grades of the RV and IVC were correlated with their PM bubble scores (r = 0.573, 0.921, p < 0.05, 0.01), and the correlation between the EB grade and the PM bubble score of the IVC was more significant (Figure 3).
Bubbles in different veins and adipose tissues of a New Zealand white rabbit model with DCS after autopsy: Diaphragmatic vein (A), IVC (B), femoral vein (C), right atrium (D), coronary vein (E), abdominal adipose tissue (F) (the presence of bubbles indicated by white arrows).
Correlation between EB grades of IVC and RV and their PM gas scores.
Sign scores, EB grade, and PM bubble scores for different levels of decompression in New Zealand rabbits.
Discussion
Ultrasound examination is a noninvasive, real-time method for gas bubble detection, 6 but its accuracy in monitoring bubbles in DCS remains unknown. Based on the successful establishment of a New Zealand white rabbit model with DCS, this study further explored the diagnostic efficacy of using ultrasound to grade systemic bubbles induced by DCS according to the EB grade.
Fan et al 17 demonstrated that the faster the rate of decompression, the higher the morbidity rate under the same conditions of stabilizing pressure and time. In order to obtain a DCS model with an appropriate morbidity rate, the decompression protocol used in this study was based on a previous study by Meng. 12 Compared with Meng's study, the results of our study showed higher morbidity and lower mortality, which may be related to the individual differences in experimental animals and the preexperimental husbandry conditions, etc.
Referring to the previous study by Bernaldo de Quirós et al, 21 we used ultrasound to detect air bubbles in the vessels at the end of decompression and found many bubble-like strong echogenic spots in the vessels. Compared to the previous study, we specifically focused on assessing bubble load in the inferior vena cava and right ventricle. By semiquantitatively grading the bubbles detected by ultrasound, we found that the EB grade of the inferior vena cava was significantly higher than that of the right ventricle, a finding that has not been reported in related studies. The reasons for this were analyzed: first, the inferior vena cava, as one of the important venous blood vessels in the body, is responsible for collecting the return of blood from the lower limbs, pelvis, and most of the organs in the abdominal cavity. This means that when air bubbles enter the circulation through the veins of the lower extremities, they first converge in the inferior vena cava. As a result, the inferior vena cava is anatomically located closer to the source of these bubbles, as opposed to the right ventricle, thus potentially capturing more of them. Second, the inferior vena cava is located behind the liver, which, as a relatively homogeneous solid organ, provides a good background for the acoustic window, which helps to minimize the scattering and attenuation of the ultrasound signal and enhances the clarity and contrast of the image. As a result, air bubbles are more easily captured, which in turn improves the detection rate of air bubbles. Finally, right ventricular bubble detection is susceptible to cardiac pulsations, leading to increased difficulty in ultrasound detection and lower detection rates, and is susceptible to interference from gases and ribs, resulting in poor resolution of ultrasound images compared to the inferior vena cava.
We also found a correlation between EB grades and DCS signs, where EB grades of 0–3 were associated with milder signs, and those at 4–5 indicated more severe signs. The result is consistent with previous research, 21 suggesting a coupling between EB grade, bubble load, and DCS sign severity, which further validates the feasibility of using ultrasound for assessing DCS severity. Moreover, individual variability was observed in the EB grades and clinical sign severity among the rabbits in this model, despite being subjected to the same decompression exposure protocol. Such inter-individual differences may be attributable to factors like body mass, activity levels, and ambient temperatures, all of which can modulate the incidence of DCS. DCS incidence is significantly elevated in animals with greater body mass. 17 Research has revealed a complex interplay between ambient temperature, animal activity levels, and DCS incidence. Exercising before or during hyperbaric exposure has been shown to increase susceptibility to DCS, 22 while the cold temperatures following decompression can heighten the incidence rate. 23
The postmortem bubble score performed by Bernaldo de Quirós et al 11 is considered to be the most accurate method currently available for assessing decompression sickness bubbles. In our experiments, since the analyses were performed within a very short period of time after death (less than 12 h), we can be sure that the bubbles detected were mainly fresh bubbles 11 —that is, bubbles resulting from decompression rather than from the putrefaction process. Our findings revealed a consistent increasing trend between ultrasound EB grades and PM bubble scores, with a statistically significant correlation (p < 0.05). Notably, the EB grades of the IVC exhibited an even stronger correlation with its PM bubble scores. This observation further substantiates the accuracy and reliability of using the EB grading of IVC as a measure for assessing bubble load in DCS.
In summary, this study successfully established a New Zealand white rabbit model with DCS and demonstrated that ultrasound evaluation of bubbles according to the EB scale can effectively reflect macroscopic bubble loads in animal models. Notably, the EB grading of the IVC provides a more accurate assessment of bubble load in DCS, underscoring its potential for clinical translation.
Footnotes
Acknowledgments
We sincerely thank the Shenzhen People's Hospital for providing the necessary facilities and help for animal experiments.
Author Contribution(s)
Declaration of Conflicting Interests
The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article.
Ethical Approval
Approval was granted by the Experimental Animal Ethics Committee of Zhuhai Bestest Biotechnology Co., Ltd (BST24W433).
Funding
This study has received funding from the National Defense Science Research Key Development-project of Sun Yat-sen University [89000–12230011]; National Natural Science Foundation of China [82371975]; Core talent fund of the Fifth Affiliated Hospital of Sun Yat-sen University [310103050302–220904094228]; Excellent Young Researchers Program of the Fifth Affiliated Hospital of Sun Yat-sen University [WYYXQN-2021010].
Guarantor
The scientific guarantor of this publication is Zhongzhen Su.
Informed Consent
Approval from the institutional animal care committee was obtained.
