Background

The spleen is the most frequently injured abdominal organ after blunt trauma [1]. Splenic injury is commonly graded using the American Association for the Surgery of Trauma (AAST) splenic injury scale, where high-grade injury is defined as AAST IV or V injury [2]. Conservative management of high-grade injury has reported re-bleed rates of as high as 75% in adults [3]. As a part of modernisation of multidisciplinary treatment, splenic artery embolisation (SAE) has become a vital treatment strategy in improving the rate of spleen preservation (aka splenic salvage) in patients with high-grade injury [4], and its use is thus increasing [1].

However, due to a lack of prospective data there remains contention amongst stakeholders as to whether SAE should be performed at the time of presentation (prophylactic or pSAE), or whether patients should be observed and SAE or splenectomy used only if a patient re-bleeds, according to haemodynamics and local resources [5, 6]. The only randomised and controlled trial to date comparing pSAE or a strategy of initial observation, showed no difference in mortality between the groups (0% vs 0.8%). However, the authors showed that pSAE resulted in 100% splenic salvage compared to an observation-first approach of 93.7%, where 32.3% of patients required SAE due to re-bleeding. The authors showed that pSAE resulted in shorter median hospital length of stay (9 vs 13 days) and fewer complications (29.2% vs 41.5%) [6]. Based on this and retrospective studies across the globe [7,8,9,10,11,12], splenic salvage of AAST IV and V injury in stable patients should be at least 90% at a major trauma centre as a minimum quality benchmark, and pSAE is an effective treatment to achieve this [1, 6, 13].

It is an accepted management practice that when heterogeneity exists in decision-making, agreeing on a treatment guideline has a role in fostering efficient workflow [14]. In many hospitals, guidelines now exist for almost all procedures. Publication of robust guidelines from major and/or notable organisations has immense value as they provide expert-based guidance (ideally based on high-level evidence) on which individual hospitals can mould their daily practice [15]. However, as evidence is available to all, the basis of major decision-making in guidelines should theoretically be similar [16, 17]. However, many institutions continue to propose management approaches based on their own preferences, team structure, and/or interpretation of the available data [18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42].

This study aimed to systematically review existing published splenic trauma management guidelines and assess whether they recommend pSAE in stable patients with high-grade injury, based on the overall guideline quality.

Methods

Registration

The study was registered on the PROSPERO database (record number CRD42023440729) and is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.

Data identification and collection

Medline, PubMed, Cochrane, Embase, and Google Scholar were searched by the study authors based against the PICO format (patient, intervention, comparator, outcome), using the following MeSH terms: “Embolization, Therapeutic/methods”[MeSH], “Abdominal Injuries/therapy”[MeSH], “wounds and injuries [MeSH]”, “Embolization, Therapeutic”[MeSH], “splenic artery” [MeSH], “spleen” [MeSH]. The following additional keywords were searched: trauma, embolization, embolisation, angioembolization, angioembolisation, guideline, protocol, pathway, nonoperative, NOM.

Inclusion and exclusion criteria

Enrolled publications included practice management guidelines from any organisation, for example societies, colleges, government bodies, and hospitals. This included publications where splenic treatment guidelines were presented even if the intention of the study was not specifically to discuss the guideline. All studies within the last 20 years were included (1 January 2003 to 1 January 2023). Duplicate studies were excluded. Studies in the paediatric population were excluded given evidence on trauma management in this context is not comparable to adult populations [43]. Studies were also excluded when they were not in English (n = 3), or when the full text could not be obtained (n = 1).

Guideline quality

The Appraisal of Guidelines for Research and Evaluation II (AGREE-II) instrument has been formulated specifically to assist with development and appraisal of guidelines to ensure that clinicians can measure a guideline’s quality before implementing it in daily use [15]. Guidelines were independently evaluated by 2 study investigators (WC and ML) using the 2017 update of the AGREE-II instrument [44] and included grading against 6 different domains. Studies were then graded for quality as high, moderate, or low. Studies were graded as high when they scored greater than or equal to 60% of the maximum score in 3 or more domains, including domain 3 (rigour of development). Studies were graded moderate when they scored greater than or equal to 60% of the maximum score in 3 or more domains, but not including domain 3. Studies were scored as low when they scored less than 60% in 2 or more domains and less than 50% in domain 3. This determination was according to precedent from previous similar studies which have used this tool [43, 45].

Outcomes

The primary outcome was to assess whether the study recommended the use of pSAE after high grade splenic trauma in stable patients, defined as embolisation (either proximal or distal) for splenic injury of AAST grade IV or V, regardless of the presence of a vascular lesion (active bleeding, arteriovenous fistula, or pseudoaneurysm), in stable patients. Secondary outcomes included stratification of the guideline quality and recommendations according to the type of institution authoring the guideline as well as the region of origin.

Statistical analysis

Numerical data were presented as percentage when calculated using the AGREE-II instrument, or number (percentage). Where relevant, assessment for differences of proportion between high, moderate, and low-quality guidelines was performed using the Chi Square test in Stata (Version 17.0-BE, StataCorp, Texas, USA). Probability values of less than 0.05 were deemed statistically significant.

Results

Search results were identified from the initial query including 951 from databases plus an additional 55 from Google Scholar. After applying exclusion criteria, a total of 28 published guidelines were included in the analysis as shown in Fig. 1. This included 5 guidelines from medical societies, 1 from a government institution, and 22 from individual hospital networks. The majority of the guidelines were from Europe and Asia (9 each, 32%), followed by North America (8, 29%) and Australia (2, 7%).

Fig. 1
figure 1

Flow diagram of study selection

Full size image

Table 1 shows the assessment of guideline quality, where 9 studies were of high quality, 9 studies of moderate quality, and 10 of low quality. As shown in Table 2, 27 out of 28 studies recommended the use of SAE in the management of splenic trauma, the only exception being the study from Koca et al. which did not mention embolisation in their flow chart or text, however, did allude to the concept elsewhere in the document without it being specifically mentioned [36]. In terms of the primary study endpoint, 15 guidelines (54%) recommended pSAE for high-grade injury. This included 6 out of 9 guidelines that were high quality (66.7%), 4 out of 9 guidelines that were moderate quality (44.4%), and 3 out of 10 (30%) guidelines that were low quality, p = 0.275.

Table 1 Assessment of guideline quality according to the AGREE-II instrument. All scores shown as a percentage of the maximum score
Full size table
Table 2 Summary of the guideline contents including inclusion of prophylactic splenic artery embolisation
Full size table

Table 3 shows the demographics of high-quality guidelines and compares them to moderate and low-quality guidelines. High quality guidelines were more commonly published by societies (44.4% vs 0% vs 10%) while both moderate and low-quality guidelines were more likely to be published by hospitals (55.6% vs 88.9% vs 90.0%), p = 0.074. In addition, high-quality guidelines were more likely to arise from the North American continent (66.7% vs 0% vs 20%) while both moderate and low-quality guidelines were more likely to arise from Europe (22.2% vs 33.3% vs 40%) or Asia (11.1% vs 44.4% vs 40%), p = 0.030. High, moderate, and low-quality guidelines were of similar likelihood to be published in the last 5 years (55.6% vs 55.6% vs 30%, p = 0.430).

Table 3 Demographics of splenic artery embolisation guidelines according to quality
Full size table

Discussion

This systematic review identified 28 guidelines on the treatment of blunt splenic injury in adults and of these, only 9 were of high quality according to the AGREE-II instrument. The incorporation of pSAE was seen in 54% which is modest, despite the available evidence [6,7,8,9,10,11,12].

In terms of the primary outcome, high-quality guidelines had a higher incorporation of pSAE (66.7%) and while the difference between high, moderate, and low-quality was not statistically significant, it is likely due to a type 2 error from a small sample. High-quality guidelines were also more likely to be written by societies while moderate and low-quality guidelines were more likely written by individual hospitals. Societies may be more likely to consider the value of their brand endorsement, and thus likely to consider the importance of stakeholder engagement and input during development. High-quality guidelines were also likely to arise from North America, however this may be confounded as this was also the location of many leading trauma societies, and the origin of the AGREE-II Enterprise. This may reflect the matured systems of trauma within a continent where trauma has a high prevalence. The authors strongly recommend that anyone who develops or updates a clinical practice guideline considers the AGREE-II instrument, or other similarly validated tools, to ensure that the standard of their recommendations are transparent and robust. From the results of this study, it is felt unlikely that any of the guidelines have specifically used such tools.

Only 5 societies worldwide have chosen to publish a guideline, and some are well overdue for modernisation. An example is the guideline from the Cardiovascular and Interventional Society of Europe (CIRSE), the largest IR society in the world. The current guideline is now over 10 years old and devoid of robust evidence, detail, clarity, and applicability [41], not in keeping with the quality usually seen from such an influential organisation. The low overall uptake of pSAE will be improved if IRs and IR societies such as CIRSE and the Society of Interventional Radiology (SIR), take a larger role in the governance of trauma, integrating themselves in trauma networks, and aligning themselves with local IR and trauma societies in different regions. It is also recommended that societies with sufficient infrastructure consider developing and regularly updating guidelines to remain relevant with constantly changing literature.

In general, most guidelines performed poorest in describing their stakeholder engagement (even those which were high quality), rarely involving a patient advocate, and often missing input from a wider multidisciplinary team. This opens the guideline to bias with reader interpretation and thus implementation of the recommendations. While this may have been acceptable in the early days of trauma which was typically run by surgical specialties, modern trauma management involves centralised, co-ordinated, tertiary care and involves a range of key stakeholders including emergency, surgery, interventional radiology, diagnostic radiology, anaesthetics, intensive care, and many others. In general, guidelines also performed generally poorly on applicability, often presenting an ideal pathway, but omitting measures to ensure that it is feasible, ways to overcome challenges to feasibility, costs, and auditing. It should be acknowledged that some studies presenting a guideline as a smaller component of a wider clinical investigation may not have felt the need to describe their guideline development in detail, and they were still included in this analysis. However, the robust development of a guideline has direct relevance on the downstream utility and it should be encouraged that authors present this vital data moving forward.

The domain of clarity of presentation had a somewhat dichotomous result. Those that published a clear and relevant flow chart and/or used an executive summary at the beginning generally scored highly. However, those that described their treatment without a specific chart scored lower. In addition, editorial independence was generally transparent through mandated reporting standards in journals. Those that did not publish their guideline in a journal often did not choose to voluntarily report any potential conflicts of interest.

A major component of the interpretation bias of guidelines is the failure for treatment pathways to adequately consider splenic function and instead place a weighted focus on mortality as the only endpoint [4]. This is also in part because it is difficult to measure treatment success in trauma and most studies in this context are retrospective with several inherent biases, leading to scepticism of results and lack of applicability of approaches where trauma systems aren’t replicable [6, 8]. In addition to these challenges, interventional radiology is still a young specialty, and as such training and governance structures vary significantly between different hospitals, regions, and countries [5]. This means that access to expert skills with appropriate training is also not universal.

Given SAE was mentioned in all guidelines, trauma networks should endeavour to include a sustainable IR service to be involved in the management of major trauma. This should involve expansion of the current IR workforce so that smaller hospitals have timely IR services available, or for patients to be sent to a centre with resources to prioritise splenic salvage in addition to mortality prevention. The benefits of salvage are vast and include the avoidance of a laparotomy with associated morbidity, prevention of overwhelming post-splenectomy infection (OPSI), reduced need for future vaccinations, reduced need for future prophylactic antibiotics, and cost savings to the individual and society [46,47,48,49].

The authors acknowledge that this study is limited by the interpretation and assessment of guideline quality and while this is an objective validated tool, still requires individual interpretation. In addition, guidelines published within a clinical cohort study may not have provided the full breadth of information to allow for their interpretation as discussed earlier. However, the authors chose to present all guidelines rather than limit the analysis to purely those from major societies given the very small sample. The statistical analysis in Table 3 is also based on a small sample. The guidelines also cross the 2018 update to the AAST injury grading criteria [2] and this may theoretically affect management decisions moving forward for those with an earlier guideline based on the 1994 AAST definitions. There is also variation in how stakeholders may view and define “prophylactic embolisation” with differing opinions on the relative importance of parenchymal injury and vascular lesions.

Conclusions

This systematic review showed that recommendation of pSAE is more common in guidelines which are of high quality. However, there is vast heterogeneity of recommended practice guidelines, likely based on individual trauma systems. This reflects biases with interpretation of data, and lack of multidisciplinary system inputs from IRs. More societies should publish guidelines, ensuring they are high-quality by conforming to existing validated reporting standards. Centres or countries which do not have the infrastructure to support pSAE are encouraged to embed IR within their trauma governance structure and overcome barriers to implementation to improve quality of care, rather than using a self-generated treatment algorithm which may not provide their patients with the established short- and long-term benefits of splenic salvage.