Abstract
Objectives
Abdominal compartment syndrome (ACS) is poorly identified in surgery for ruptured abdominal aortic aneurysm and an early management is crucial. The aim of this study was to validate how many risk factors were needed to predict ACS. Secondary objectives were to assess its prevalence and the 30-day mortality.
Methods
All patients operated for ruptured abdominal aortic aneurysm during 5 years were included. An independent committee performed a retrospective diagnosis of ACS. Eight criteria were selected from the literature, and corresponded to pre- and intraoperative period: anemia (hemoglobin lower than 10 g/dL), prolonged shock (systolic blood pressure <90 mmHg more than 18 min), preoperative cardiac arrest, obesity (body mass index > 30), massive fluid resuscitation (≥3500 mL per hour for at least 1 h) and transfusions (>10 units packed blood red cell since the beginning of the treatment), severe hypothermia (≤33℃), acidosis (pH < 7.2). Sensitivity and specificity were assessed for each number of criteria.
Results
Eight patients were ACS+ and 28 ACS-, with three criteria for ACS+ and 1.5 for ACS- (p = 0.002). Three criteria among the eight selected criteria have the best cutoff for sensitivity and specificity (75% and 82%) with a positive predictive value of 54% and a negative predictive value of 92%. The prevalence of ACS was 17%. The 30-day mortality in ACS+ tended to be higher than in ACS- (p = 0.108).
Conclusion
The present results suggest that patients with an ACS seemed to have higher mortality and the threshold of three factors among eight specific factors is enough to predict this.
Introduction
Despite the development of new techniques in the endovascular approach, the perioperative mortality of ruptured abdominal aortic aneurysm (rAAA) is still dramatically high. 1 If the preoperative deaths are mainly due to uncontrolled bleeding and its consequences, the postoperative deaths are rather related to multiple organ failure (MOF) that could be explained by a misdiagnosis or a delayed diagnosis of an occurring abdominal compartment syndrome (ACS). The ACS has been well described in literature but not specifically investigated in vascular surgery, particularly in the case of rAAA. Nevertheless, it is clearly identified as being one of the main causes of mortality after a hemorrhagic shock, such as myocardial infarction and pulmonary complications. 2
Abdominal compartment syndrome in aortic surgery.
rAAA: ruptured abdominal aortic aneurysm; os: open surgery; es: endovascular surgery; ACS abdominal compartment syndrome.
The aim of this study was to assess the number of criteria to predict the occurrence of ACS after endovascular or open surgical repair of rAAA. The secondary objectives were to determine the prevalence of ACS in the studied population and to assess its impact on the 30-day mortality.
Material and methods
The present work is a retrospective, one center, observational study performed in the Department of Vascular Surgery of the University Hospital of Besançon, France. All patients aged 18 years or older who underwent rAAA repair from January 2007 to December 2011 were included. rAAA was defined as an extra arterial extravasation of blood or contrast next to an AAA on preoperative radiologic imaging, or as a rupture of an AAA discovered during surgery. Noninclusion criteria were a known bladder disease that might affect the validity of the intra-bladder assessment of the IAP, and death before arriving at the operating room or during surgery.
Data collected
Demographic data (i.e. age and sex) and relevant past medical history were collected from the medical files: cardiovascular risk factors (i.e. hypertension, defined as taking antihypertensive medication; diabetes mellitus, defined as taking a statin therapy; current smoking, defined as an active smoking reported by the medical files or by the patient, or a smoking withdrawal <6 months; and obesity, defined as a body mass index ≥30 kg·m−2), coronary heart disease (defined as a medical history of myocardial infarction, angina, or coronary lesions), lower limb peripheral arterial disease (defined as claudicant, previous vascular lower limb surgery, or critical limb ischemia), history of an aortic aneurysm, chronic renal failure (defined as a creatinine clearance <60 mL·min−1), antiplatelet or anticoagulant therapy, and previous abdominal surgery. Clinical presentation on admission in the operating room was recorded: abdominal or back pain, faintness, cardiac arrest, preoperative unconsciousness, orotracheal intubation required before the admission in the operating room, and preoperative use of intra-aortic occlusion balloon. The type of surgery (open or endovascular surgery), the operating time, the clamping time (total clamping time, supra renal, and supra celiac clamping time), the intraoperative blood loss, and the intraoperative total amount of fluids and blood products were extracted from the computerized anaesthesia records and from the operative reports. All patients were admitted post-operatively in the surgical intensive care unit (ICU). All patients had a urinary catheter on admission in the ICU. Patients were sedated and paralyzed using neuromuscular blocking agents until the improvement in their status was considered sufficient to try to awake them. Clinical parameters (heart rate, arterial pressure, body temperature, pulse oximetry, blood loss) and medication were checked at least every hour, and urine output at least every 3 h. Clinical parameters, medication, and urine output were reported on ICU nurse sheets. Data on outcome were collected from the medical files until day 30 after surgery.
Predictive criteria of occurrence of ACS
Eight clinical and biological criteria have been found in the literature and correspond to the promoting factors of ACS.3,15,17,19 The preoperative criteria were: a severe anemia (hemoglobin < 10 g/dL), a prolonged shock (a systolic blood pressure < 90 mmHg during at least 18 min), a preoperative cardiac arrest, and a medical history of obesity (body mass index > 30 kg·m−2). The intraoperative criteria were: a massive fluid resuscitation (≥3500 mL per hour during at least 1 h), multiple transfusions (>10 units packed blood red cell from the beginning of the treatment), hypothermia (body temperature ≤ 33℃), and a metabolic acidosis (pH < 7.20).4,12,17 This data were obtained from the medical and laboratory files. Each criterion was considered as equal to the others.
Definitions
Clinical and laboratory data from all patients were reviewed anonymously by a multidisciplinary committee during a common presentation session. The multidisciplinary committee comprised two vascular surgeons, two visceral surgeons, and two intensive care physicians. All members of the committee were experienced in the management of ACS. At the beginning of the presentation session, each member received an individual sheet mentioning the anonymized list of patients included. After the presentation of each case, each member of the committee was invited to report on their individual sheet if the patient had presented or not an ACS. Their individual decisions were blinded to other member’s diagnosis and blinded to the value of the eight predictive criteria of ACS. Members of the committee were not allowed to discuss with each other for avoiding any suggestion bias. Finally, patients were assigned to four groups, and the decision was based on the majority of the assessments: the ACS + group, including patients for whom the diagnosis of ACS was confirmed by at least four members; the ACS- group, including patients for whom at least four members had ruled out the diagnosis of ACS; the hemorrhagic shock group, including patients who died from a hemorrhage within the first postoperative hours; and the undetermined status group including patients who died within the first six post-operative hours as well, but for whom the committee was not able to state concerning the cause of the death. Patients of the last two groups were excluded from the statistical analysis.
Statistical analysis
Quantitative variables, expressed as median (interquartile range), were compared between the ACS+ and the ACS- groups using the Mann–Whitney U test. The intergroup comparisons for qualitative data, expressed as the number of patients (percentage), were done by using the Fisher exact test. Sensitivity, specificity, positive, and negative predictive values of each number of criteria were calculated and the receiver operating characteristic (ROC) curve was drawn. All p-values were two-sided and p < 0.05 was considered to be statistically significant. The statistical analysis was performed using SAS 9.3 software (SAS institute, Cary, NC, USA).
Results
Overall population
Forty-seven patients underwent rAAA repair during the study period (Figure 1). Thirty-five patients were treated by open surgery whereas 12 patients were treated by endovascular surgery. Description of the surgical procedure and of the anesthetic management is presented in Table 2. Three patients initially treated by endovascular surgery needed a surgical conversion during the procedure (two failures of the system and one major proximal endoleak). Forty-two patients were given a general anesthesia. Low-pressure monitoring during the preoperative management was applied to all the patients. A cell-saver device was used for all patients operated by open surgery. The prevalence of ACS in our series was 17% [6.3; 27.7]. Forty-three patients presented postoperative complications, especially an acute renal failure for 32 patients among which 20 patients needed a temporary dialysis, and pulmonary complications for 28 patients. The average length of stay in the ICU was about 3.0 [1.0; 17.0] days and the total hospital stay lasted for 14.0 [3.0; 30.0] days. In all, 17 patients died. The main causes for death were hemorrhagic shock and MOF.
Flowchart. Demographic analysis and frequency of preoperative comorbid disease, initial presentation, and surgical data. *Median (interquartile range).
ACS+ and ACS- groups
Characteristics of patients with or without abdominal compartment syndrome.
*Median (interquartile range).
Pre- and intraoperative criteria in patients with or without abdominal compartment syndrome.

ROC curve analysis.
Discussion
In this pilot study, we found that with the threshold of three among the eight selected criteria, we would predict an ACS occurrence with a 54% positive predictive value and a 92% negative predictive value. Those criteria are pre- and intraoperative ones and easy to assess even in the operating room. The prevalence of ACS was found to be 17% and the mortality rate in the ACS+ group was 37.5%.
We did not separate endovascular from open surgery whereas they are obviously two different populations. But in the literature, the proportion of ACS is similar in both.4,5,8–13 The risk of ACS still exists for endovascular repair of rAAA and reports suggest that up to 20% of patients develop ACS requiring decompression.13,14,20,21 If patients with rAAA had a primary closure, it is necessary to monitor IAP during the ICU stay with bladder pressure monitoring, which was not routinely done in our experience. A protocol to monitor the IAP during the first 48 h after endovascular rAAA repair must be established. 22 If not, the early diagnosis will be missed and the treatment will be inadequate or too late. 23 In our study, all the risk factors applied were extracted from the literature and particularly from vascular studies. Six factors among the eight chosen were selected from the work of Rasmussen et al. 4 Furthermore, we decided to add two additional independent factors highlighted by Carr et al. 17 We have also chosen factors from WSACS who could be present in the operative room.3,18,19
The strength of the methodology is based on the independent committee and this fact needs to be discussed. The diagnosis of ACS can be easily assessed by bladder pressure monitoring, but we wanted to assess how many risk factors were required in order to develop one. Creating a committee was the only way to obtain an objective opinion concerning retrospective diagnosis of ACS. Our results are comparable to the literature in terms of prevalence of ACS and mortality rate.4,10,12,13 This suggests that the multidisciplinary approach, the anonymous presentation of the cases, and the decision based on the majority of the assessments were enough precautions for obtaining an objective result.
In our experience, we found a mortality three times higher in the ACS+ group than in the ACS- group (not significant), in accordance with the literature where the mortality rate was significantly higher for the patients with ACS compared with patients without ACS.4,10,12 Furthermore, delays in performing decompressive laparotomy are associated with an excessive morbidity and mortality rates up to 88%. 24 When the authors examined outcomes of patients for whom primary closure was not possible, they noted that the very early mortality was reduced and there was a decreasing trend in the global mortality (from 73% in primary closure to 50% with delayed closure). 25 The only way to significantly reduce the mortality seems to be to treat prophylactically the ACS that is to say, before its occurrence. Some management algorithms about IAH and ACS were described in the studies.4,26 Our study seeks to validate the threshold number of necessary criteria leading to early diagnosis of ACS, which is not known to our knowledge. This represents a preliminary approach, based on an observational study. The next step is to validate this threshold of three criteria by a prospective multicentric study, and to assess the importance of each one on the impact of the ACS occurrence.
Decompressive laparotomy is the only treatment for ACS. There are possible complications related to maintaining open abdomen such as incisional hernia or infection. 27 The benefits of an early decompression in a situation of ACS are evident, with a lower risk of organ impairment, and most likely increased overall survival.28–30 Presumptive decompression should be considered at the time of laparotomy for patients who demonstrate risk factors for ACS. The final aim would be to screen ACS in the operative room and manage ACS early.
The limitations of our study are its retrospective manner, the small number of patients, and single-center nature. But this study represents the necessary pilot study before a multicentric one, based on its conclusions. The absence of intra-bladder pressure measure could be another limitation but in our experience the committee allows us to obtain objective results.
Conclusion
The present results suggest that patients with an ACS after rAAA repair seemed to have higher mortality and complications rate. In our experience, the threshold of three factors among eight specific factors is enough to predict this. A multicentric prospective study is clearly needed to confirm these results (ethics committee registration number of our hospital 16/05).
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
