Background
Abdominal compartment syndrome (ACS) is defined by an intra-abdominal pressure (IAP) of 20 mmHg or higher that is accompanied by newly developed organ dysfunction [
1]. Although the first cases of ACS were described decades ago, interest has recently increased exponentially. Nevertheless, a recent survey demonstrated that knowledge among physicians is still suboptimal toward clinical management and awareness of ACS [
2].
There is a wide range in the incidence of ACS in intensive care unit (ICU) patients, with patients in surgical ICUs more likely to develop ACS. Incidence rates of intra-abdominal hypertension (IAH) and ACS upon admission to the ICU are reported to be around 27.7% and 2.7%, respectively [
3,
4]. The prevalence of ACS is between 4.2% and 14% in patients admitted to the ICU after trauma; in general ICUs, it is estimated to be around 1%. With increasing insight into the risk factors for ACS, the introduction of guidelines, and the availability of strategies to limit progression from IAH to ACS, ACS seems to be decreasing in most ICUs [
5,
6].
As in other compartment syndromes, surgical decompression has been considered the definitive therapy for ACS for a long time, particularly in primary ACS. In case of secondary ACS, surgical intervention may no longer be the treatment of choice. The IAH/ACS management algorithm, as developed by WSACS (The Abdominal Compartment Society, formerly known as the World Society of the Abdominal Compartment Syndrome), was recently updated and recommends medical treatment options to reduce IAP before surgical decompression is needed [
7,
8]. The cornerstone of medical management in patients with IAH (defined as IAP ≥ 12 mmHg) is perfusion support and optimized fluid management, with several noninvasive methods used to reduce IAP, such as nasogastric decompression or percutaneous drainage of fluid collections. Only when these noninvasive medical treatment options fail to lower the IAP and organ failure persists is decompressive laparotomy recommended [
1,
9‐
11].
The effect of decompressive laparotomy on organ function in patients with ACS has been poorly described. Because mortality rates remain as high as 49% even after decompression [
12], further investigation is needed into the use of decompressive laparotomy to understand who could benefit the most. From a mechanistic point of view, for example, patients with decreased abdominal wall compliance would benefit the most, but this has been largely ignored. Our goal in this study was to investigate the effect of decompressive laparotomy on organ function, particularly on hemodynamics and respiratory and renal function, in a broad ICU population.
Methods
Search strategy
The literature was reviewed for studies reporting on the effect of decompressive laparotomy in patients with ACS published between 1995 and September 2017. The search terms (“abdominal compartment syndrome” or “intra-abdominal hypertension” or “intra-abdominal pressure” and “decompression” or “decompressive surgery” or “decompressive laparotomy”) were used to search three databases (MEDLINE, Embase, and Web of Science). Restrictions were applied to the search query. To minimize publication bias, case reports were excluded, as were case series or reviews describing fewer than five patients. In addition, animal studies were not considered. The search was limited to literature published in the English, French, German, Dutch, and Spanish languages. The bibliographies of the included articles were examined for relevant publications that might have been overlooked otherwise.
Inclusion and exclusion criteria
Eligible studies were assessed on the basis of predefined inclusion and exclusion criteria. Studies that described adult and pediatric patients who developed ACS and required decompressive laparotomy were included in the analysis.
Studies were included only if they reported the IAP at least before the procedure, or if they defined ACS, in addition to patient outcome for every patient who underwent decompressive laparotomy. Hemodynamic (blood pressure [BP], heart rate [HR], systemic vascular resistance index [SVRI], cardiac index [CI], pulmonary capillary wedge pressure [PCWP], and/or central venous pressure [CVP]), renal (urinary output [UO]), and/or respiratory (ratio of partial pressure arterial oxygen and fraction of inspired oxygen [P/F ratio], peak expiratory end pressure [PEEP], and/or peak inspiratory pressure [PIP]) parameters were required to be reported to measure the effect on organ dysfunction. Patients who underwent decompressive laparotomy needed to be clearly identified, and the data of these patients had to be discussed separately. The time frame of measuring these parameters had to be less than 72 hours after the intervention.
Decompressive laparotomy was defined as a vertical, midline, full-thickness abdominal incision aimed at reducing the IAP. This may or may not have been followed by a temporary abdominal closure. Other decompression techniques, such as subcutaneous fasciotomy or subcostal laparotomy, were not included.
Data extraction and outcome measures
Baseline characteristics of the patients as well as of the studies were extracted. These included first author; year of publication; number of patients; study design; and basic patient characteristics, such as age, gender, and cause of ACS. Studies were categorized according to whether the WSACS definition [
1] was used in the diagnosis of ACS. Each article was then categorized according to which type of ACS it described, namely primary, secondary, or combined. Studies reporting pediatric patients were analyzed separately. Hemodynamic, renal, and/or respiratory parameters were extracted and analyzed.
Statistical analysis
For statistical analysis, the Comprehensive Meta-Analysis software package (Biostat, Englewood, NJ, USA) was used. The outcome parameters retrieved from the studies (described as mean, SD, and sample size) were entered into the program to calculate the standardized mean difference (SMD) of each parameter as well as the
p value. When articles described median and range, an estimation of mean and SD was made using the formula described by Hozo et al. [
13]. Heterogeneity across studies was evaluated using Cochran’s
Q statistic. A random effects model was used when heterogeneity was present, as suggested by DerSimonian and Laird, to reduce bias [
14]. A
p value < 0.1 for the Cochran’s
Q statistic was considered to represent significant between-study heterogeneity. The
p value was considered statistically significant when this was < 0.05. Hedges’
g was used to examine the SMD, because the
p value provides information about only the presence of an effect and not the size of the effect. The following cutoffs were considered to estimate the effect size of the intervention on the reported parameter: 0.2–0.5, a small effect size; 0.5–0.8, a medium effect size; and > 0.8, a large effect size.
Quality assessment
A funnel plot was used to assess the presence of publication bias visually and was quantified by the Egger test. The threshold for bias was a
p value < 0.10. Two validated checklists were conducted to assess the methodological quality of all included studies. First, the methodological index for nonrandomized studies (MINORS) [
15] was used, which contained a checklist consisting of 8 criteria for noncomparative studies and 12 items for comparative studies. Downs and Black [
16] drafted a checklist for not only nonrandomized but also randomized studies. With this list, 27 items were verified to assess the quality of the studies. To facilitate assessment, combining both studies [32, 33], each score was converted to a 0–10 scale, and an average score was then calculated. The lower the score, the higher the risk of bias, and vice versa.
Discussion
In this systematic review and meta-analysis, we found that decompressive laparotomy results in a significant decrease in IAP. Decompressive laparotomy also had a measurable effect on organ failure, especially on respiratory function as well as on kidney function. There was a small effect on hemodynamics, which could be seen mainly in grade IV ACS. Mortality remains high; 49.7% of adults did not survive, underlining the severity of the illness. In children, the mean mortality rate was as high as 60.8% following decompressive laparotomy.
When laparotomy is performed, most hemodynamic, respiratory, and renal parameters will improve. Therefore, the results of this meta-analysis confirm the recommendation that decompressive laparotomy should be considered when medical options fail. Clearly, though, there is still room for improvement, and new options should be explored. It is incompletely understood which patients would benefit most from decompressive laparotomy or what is the most optimal timing for the intervention. As demonstrated in this analysis, there was a correlation between timing of decompressive laparotomy and mortality, although the correlation was weak and the clinical relevance limited. However, it is worth mentioning that in the studies in this analysis, some centers may have been reluctant to perform decompressive laparotomy, whereas others operated more rapidly, potentially leading to bias. Furthermore, we did not have individual-patient data available. When considering who would be the ideal candidates for decompressive laparotomy, it can be assumed that patients with low abdominal wall compliance are most likely to develop ACS and would probably benefit most from decompressive laparotomy. If there were methods available to easily identify these patients, a more individualized treatment approach would be possible [
19,
20].
Overall, organ failure is poorly defined in the reviewed articles, and many studies fail to report the true markers of organ failure. The use of the Sequential Organ Failure Assessment (SOFA) scoring system has been recommended before, in the WSACS guidelines of 2013 [
1]; the two most recent studies that were included in the present review did report SOFA before as well as after decompression. However, the SOFA scoring system has its own limitations. Minor changes (in any of the organ systems) that can be beneficial in patients might be overlooked because of the broad range within gradation scores; for example, the P/F ratio may increase from 110 to 190, but still the SOFA score would remain unchanged.
A positive trend can be seen in the baseline IAP, where pressures over 40 mmHg have become rare because IAP measurement is now more frequent and IAH and ACS are detected earlier. Even though there is a lower baseline IAP, the mortality rate did not improve over the years. Mortality was higher in grade IV ACS (52%) than in grades III and IV ACS combined (42%).
Because of the detrimental effects of ACS and the need for well-researched therapeutic options, articles need to describe not only IAP but also hemodynamic, respiratory, and renal parameters. Measuring protocols should be standardized to allow for a more complete description of data. Currently, different scoring systems and parameters are used to describe the results regarding multiple organ failure, whereas when describing outcome, to get a better understanding of the effect of an intervention, uniformity is key. Because mortality remains high, it is of the essence to report the cause of death when possible.
This study has several limitations. Before all the data could be analyzed, some of the data had to be converted from SEM to SD. Bessel’s correction was used to correct bias in the estimation of the population variance. Nevertheless, the mean SE will often be raised in these approximations, and therefore results could vary. Furthermore, some articles reported only median and range data. The formula described by Hozo et al. to convert median values to mean and SD was used [
13]. This might have affected outcomes, specifically in variance. Nonetheless, the alternative of not using these articles might be less favorable than using the estimated mean.