Abstract

Aim: Describe the management of blunt splenic trauma in a 10-year period in our department and perform a systematic review of the relevant literature.

Materials and methods: Retrospective analysis of registry data from patients with splenic injury from 2014 to 2024 and systematic review of the publications in PubMed, Scopus and the Cochrane Library.

Results: Sixty-eight patients, 58 males and 10 females with a mean age of 42 years, were included. Sixteen were grade I, 14 grade II, 12 grade III, 6 grade IV, and 20 grade V. Forty-two patients were treated non-operatively, and 26 underwent splenectomy. Two patients failed the non-operative management (NOM) (95.2% success rate). An increasing tendency for the number of platelets was noted. A statistically significant difference on days 5 and 10 between grade II and grade V (p=0.00056 and p=0.0118, respectively) and grades III and V (p=0.01 and p=0.01) was found, with grade V platelet count peaking on day 10.

In our qualitative synthesis, 17 studies were included, with 5137 patients, mostly males with a mean age of 37 years. 1189 were grade I and II, and 2916 were grade ≥3. A significant shift towards NOM has been noted (60.5%, with success rates ranging from 76% to 100%). A preference for splenic preservation over splenectomy, which may be achieved with embolization, is indicated. Clinical and laboratory evaluation should be carefully conducted when considering splenic embolization.

Conclusion: A shift towards NOM is noticed with surgical intervention restricted for hemodynamically unstable patients. Splenic embolization is an effective alternative which may be used to further establish non-surgical management.

embolization, non-operative management, platelet count, splenectomy, splenic injury

Trauma is a leading cause of mortality and disability, universally accounting for 4.4 million deaths annually, according to the World Health Organization.^[1] One of the most commonly injured organs in blunt abdominal trauma is the spleen.^[2] Patients with splenic injuries may present with different clinical presentations depending on the severity of the injury. In hemodynamically unstable patients, Focused Assessment with Sonography for Trauma (FAST) is performed, and a positive FAST defines the need for immediate laparotomy. In hemodynamically stable patients, the gold standard for diagnosing splenic injuries is the abdominal CT scan. The World Society of Emergency Surgery (WSES) classification system for splenic injuries by the American Association for the Surgery of Trauma (AAST) has been established. This classification divides splenic trauma into 5 categories depending on the CT findings.^[3]

The treatment options traditionally consist of operative and non-operative management (NOM). The latter is a prevailing course of treatment in hemodynamically stable patients and includes close monitoring, overall patient support, and a follow-up CT scan within 24-72 hours, as the risk of delayed subcapsular hematoma rupture or pseudoaneurysm rupture remains.^[2,3] Splenic artery embolization is an emerging course of treatment that offers the advantage of splenic salvage and may further establish non-surgical management.^[2] In hemodynamically unstable patients or patients with clinical signs of peritonitis, operative management is the course of treatment. In splenic injuries of grade 3 and above or in case of NOM failure, surgery should be considered.^[3]

Our study aims to describe the management and outcomes in our hospital over a ten-year period, 2014-2024. We assessed the grade of the injury, the treatment strategies, the length of hospitalization, transfusion requirements, morbidity, and mortality. In addition, we performed a systematic review of the existing literature to evaluate the management of spleen injuries in various Trauma Centers and assess the outcomes of the different treatment strategies.

A retrospective analysis of registry data was conducted for patients with traumatic splenic injury admitted to the General Hospital of Nikaia, Greece, from 2014 to 2024. Patients were identified from the institutional trauma registry. Patients undergoing splenectomy for non-traumatic reasons were excluded. Demographic data, mechanism of injury, radiology reports, operative details, antiplatelet drug treatment, mortality outcomes, and transfusion requirements were extracted by two independent reviewers (AE and VP) from hospital electronic records. The classification of splenic injury was made according to the American Association for the Surgery of Trauma (AAST) splenic injury scale.^[3]

Patients were divided into a NOM group and an operative group. Due to limited availability of an interventional radiologist, embolization of the splenic artery was not used as a treatment option for splenic rupture in our patient population. Primary outcomes were mortality, platelet count, and possible initiation of anti-platelet treatment, red blood cell (RBC) and fresh frozen plasma (FFP) transfusions, course of treatment, and the length of hospital stay. Mortality was defined as in-hospital death during the index admission. Transfusion requirements included units of red blood cells and fresh frozen plasma administered during hospitalization. Our study was approved by the Hospital’s Ethics Committee and all included patients provided a written consent for the use of their medical files.

Furthermore, a systematic review of the relevant publications was conducted in three databases, PubMed, Scopus, and the Cochrane Library, until July 22, 2025, and reported in accordance with the PRISMA 2020 guidelines. Keywords used were “traumatic,” “spleen,” “injury,” “splenectomy,” “blunt splenic trauma,” and “management.” Eligible studies for inclusion complied with the following inclusion criteria: More specifically, they should be published after 2000 so as to be closer to modern data and guidelines and describe the management of traumatic spleen injury. The studies should include a non-operative management and an operative management group of patients. Studies in which an angioembolization group was included were also accepted. Randomized controlled trials (RCTs), cross-sectional, cohort, interventional, observational follow-up, and longitudinal studies, as well as prospective and retrospective studies, were all eligible for inclusion. Case reports, letters to the editor, systematic reviews, case series, and books were excluded.

Two independent reviewers (AE and VP) screened the titles and abstracts based on the predefined inclusion and exclusion criteria. After the initial screening, a full-text assessment of the articles was performed. Disagreements were settled through discussion, but in the case of an inconclusive decision, a third reviewer (IK) resolved the conflict. The study selection process is shown in the PRISMA 2020 flow chart presented in Fig. 1.

Figure 1.

Flow diagram of study selection.

Data extraction of the included articles was conducted in a data extraction sheet, which presented the following information: first author, publication year, study design, aim of the study, primary and secondary outcome, study population, age and sex, mechanism of injury, grade of injury, treatment, transfusion requirements, hospital stay, mortality and complications.

The risk of bias of the included studies was assessed with the ROBINS-I V2 tool for non-randomized studies. The following seven domains were evaluated: risk of bias due to confounding, risk of bias in classification of interventions, risk of bias in selection of participants into the study (or into the analysis), risk of bias due to deviations from intended interventions, risk of bias due to missing data, risk of bias arising from measurement of the outcome, and risk of bias in selection of the reported result. Possible risk of bias in judgments was low, moderate, serious, and critical. Two reviewers (AE, VP) independently rated all the included studies.

During the period of our study, 68 patients presented in the emergency department of our hospital with splenic injury, with the most common cause being road traffic accidents, and were all included in our study. 58 patients were males and 10 were females with a mean age of 42 years (range from 17 to 82 years). All patients underwent a CT scan on admission, and according to the AAST grading, 16 were grade I, 14 were grade II, 12 were grade III, 6 were grade IV, and 20 were grade V. Demographics, clinical characteristics, and complications are demonstrated in Table 1.

Operative management

Splenectomy was performed in 28 patients, 26 at presentation and 2 after failure of NOM as mentioned above. The mean age was 46.71; 26 were males, and 2 were females. The majority (n=20) of patients were classified as grade V, 2 as grade III, and 6 as grade IV. Mean transfusion requirements amounted to 2 RBCs and 2 FFPs.

An increasing tendency of the platelet count was also observed in the operative management group, with 3 patients exceeding 106, and antiplatelet treatment was initiated during their hospitalization. The mean length of hospital stay was 11.8 days. Overall in-hospital mortality was 0%. However, one patient was still hospitalized in the Intensive Care Unit at the time of the study due to comorbidities. This patient was excluded from the calculation of the length of stay. Discharged patients received recommendations for a follow-up CT scan in 15 days and a CBC (Complete Blood Cell) count at 5 days to monitor the platelet count and initiate treatment if needed. At follow-up CT, 6 patients demonstrated a pseudoaneurysm of the splenic artery.

A statistical analysis of the probability of blood and FFP transfusion depending on the grade of injury was performed based on the transfusion requirements of our patients. The need for blood transfusions has been found to increase with the grade of injury (grade I: 14%, grade II 29%, grade III 40%, grade IV 50%, and grade V 60%). A similar tendency was noted for the need for FFP transfusion (grade I: 14%, grade II: 14%, grade III: 40%, grade IV: 50%, and grade V: 10%).

A comparison between the platelet count and the day of hospitalization was conducted. Platelet count was measured on presentation day, day 5, and day 10 for all patients hospitalized for splenic injury. The results are shown in Fig. 2. It is of note that only two patients of the NOM group with grades I and II splenic injury received antiplatelet drug therapy prior to the trauma. An increasing tendency for the number of platelets in all grades of injury was noted. On presentation day, no statistically significant difference (p=0.4049) between the grade of injury and the number of platelets was reported. However, on day 5 and day 10 a statistically significant correlation (p=0.0006 on day 5 and p=0.0007 on day 10) between the platelet count and the grade of injury was found. In addition, a comparison between the different grades of injury separately on day 5 and day 10 was performed. On day 5, a statistically significant difference between grade II and grade V (p=0.00056) and grades III and V (p=0.01) was found. Similarly, on day 10, a statistically significant difference between grades II and V (p=0.0118) and grades III and V (p=0.01) was noted. The difference in the platelet count among the different groups of splenic injury is mainly because of grade V injuries, which seem to differ significantly, especially from grade II and grade III. The platelet count of grade V splenic injuries is expected to peak on day 10. When excluding all patients treated operatively, the increase in the platelet count, although noted, was not statistically significant. Grade I injuries were used as a reference group.

Figure 2.

Comparison of the platelet count, day of hospitalization and grade of injury.

Results from the review of the literature

Overall, seventeen studies were included in our qualitative synthesis. Table 2 summarizes the characteristics and outcomes of the included studies.^[4-20] The total number of patients included in these studies was 5137, with a mean age of 37, with the majority being males. Regarding the grade of splenic injury of the studies providing this information, 1189 were grade 1 and 2, and 2916 were grade ≥3. As in our study, most splenic injuries were caused by motor vehicle crashes.

Table 2.

Download as

CSV

XLSX

Baseline characteristics of the final full-text articles

Author, year

Title

Aim

Main outcome

Secondary outcome

Study design

Number of patients

Sex M/F

Mean age

Mechanism of injury

Grade of injury

Treatment

Transfusion

Hospital stay

Mortality

Complications

1

2

3

4

5

Conservative management

Angioembolization

Operative management

RBC

FFP

Frandon et al. 2015[4]

Blunt splenic injury: are early adverse events related to trauma, nonoperative management, or surgery?

Compare clinical outcomes and early adverse events of OM, NOM, and NOM with SAE in BSI and identify the prognostic factors.

Patients with BSI present worse outcome and more adverse events in OM, but this is related to the severity of injury. The main predictor of adverse events remains the severity of injury

More infectious complications and mortality reported in the OM group , infections pleural effusion rate, pancreatitis were related to the severity of the injury

Retrospective study

136

n/a

25

n/a

77 patients with grade ≥3

61

50

25

n/a

n/a

12

4

Secondary bleeding 26 (13 NOM vs. 6 SAE vs. 7 OM) splenectomy 10 (6 NOM vs. 4 SAE), ACS, infection, pleural effusion, ARDS, acute renal failure, pancreatitis, thrombosis, splenic pseudocyst, splenic secondary vascular lesion

Bagaria et al. 2019[5]

Changing aspects in the management of splenic injury patients: experience of 129 isolated splenic injury patients at level 1 Trauma Center from India

Compare NOM and OM for isolated splenic injury in an Indian level 1 trauma center.

Careful selection is the most important factor dictating the success of NOM and patients with higher grades of splenic injury are more likely to require operative intervention

Patients with higher grade injuries are characterized by need of more blood transfusion, prolonged ICU and hospital stay, and higher mortality in comparison to patients of nonoperative group

Retrospective study

129

103/26

25.4

Road traffic accidents

6

19

44

41

19

70

25

34

n/a

n/a

NOM 5 days, OM 6 days

2

Three patients failed postembolization and underwent splenectomy; Six patients needed readmission

Chastang et al. 2015[6]

Is non-operative management of severe blunt splenic injury safer than embolization or surgery? Results from a French prospective multicenter study

Evaluate the morbidity and mortality of splenic injury according to severity and management (surgery, embolization, non-operative management)

Embolization should not be performed as a prophylactic measure, but only in presence of clinical or laboratory signs of bleeding.

Specific morbidity related to the management was 10% after NOM vs. 47% after embolization (p=0.02). Specific morbidity after surgery was 15%.

Prospective multicenter study

91

64/27

39

n/a

37

54

59

15

17+4

n/a

n/a

14

0

Four patients underwent splenectomy, 1 surgery for perforation, and 1 second embolization

Zurita Saavedra et al. 2020[7]

Management of splenic injuries utilizing a multidisciplinary protocol in 110 consecutive patients at a level II hospital

Assess the failure rate of the NOM after implantation of a multidisciplinary protocol for splenic injuries comparing the results with the literature

Conservative management and preservation of the spleen is preferred

High hospital stay

Retrospective study

110

9/2

37

n/a

14

24

34

37

56

-

54

n/a

n/a

22.8

1

Ten patients with postoperative complications

Rosati et al. 2015[8]

Management of splenic trauma: a single institution’s 8-year experience

Determine the outcomes of splenic injuries based on management strategies over an 8-year period at our Level 1 Trauma Center

Increase in nonoperative management by an increased application of angiography and embolization

Utilization of SE in patients with appropriate indications will result in low failure rates and improved mortality

Retrospective study

926

649/257

36.4

Motor vehicle accidents

463

224

239

677

129

120

319

n/a

10.5

8

Organ failure 107, shock 44, infection 96, post hemorrhagic anemia 204, respiratory support 143, laparotomy 65, tracheostomy 65

Gill et al. 2024[9]

Splenic artery embolization in the treatment of blunt splenic injury: single level 1 Trauma Center experience

Describe the experience of a single level 1 Trauma Center in the management of blunt splenic injuries

The endovascular management of BSI is safe and effective. The overall splenic salvage rate was 90.3%.

Distal embolization with Gelfoam® was not associated with higher rates of splenic infarction when compared with proximal embolization with coils. Combined proximal and distal embolization was associated with a higher incidence of splenic infarction and splenic abscess formation

Retrospective study

450

SAE patients 47/25

SAE patients 43.2

n/a

-

8

25

34

5

297

72 + 1

80

1.9

n/a

10.9

n/a

Five patients from NOM group failed due to bleeding and underwent SAE, Seven patients after SAE required splenectomy due to excessive bleeding; 4 readmissions in a 30-day period

Saurabh et al. 2011[10]

Splenic trauma - our experience at a level I Trauma Center

Identify the effect of non -operative management on splenic trauma patients and its implications at our level I Trauma Center between January 2007 and June 2008

NOMof splenic trauma can be performed with an acceptable outcome

All the penetrating splenic injuries underwent splenectomy

Retrospective study

67

n/a

25

Motor vehicle crash

21

13

16

14

3

42

-

25

4.04 in operative 1.4 in NOM group

n/a

12.8 for operative group 9.9 for NOM group

n/a

One patient failed NOM and required splenectomy

Jesani et al. 2019[11]

Splenic trauma, the way forward in reducing splenectomy: our 15-year experience

Study radiological assessment, management and outcome of traumatic splenic injury over 15 years in a UK district general hospital

A significant shift towards selective non-operative management within our institution over the past 15 years

A high proportion of low-grade splenic traumas were managed with splenectomy

Retrospective study

126

3/1

45

Motor vehicle crash

6

13

42

16

26

67

5

54

n/a

n/a

n/a

13

1 failed splenorrhaphy

Author, year

Title

Aim

Main outcome

Secondary outcome

Study design

Number of patients

Sex M/F

Mean age

Mechanism of injury

Grade of injury

Treatment

Transfusion

Hospital stay

Mortality

Complications

Wahl et al. 2004[12]

Blunt splenic injury: operation versus angiographic embolization

Whether AE is effective and safe in the management of blunt splenic injuries and how this intervention may affect patient outcomes and to evaluate whether AE could be compared to operative intervention for patients who went to the OR after CT

Angiographic embolization is safe and associated with fewer complications

The spleen AIS, heart rate, age, and GCS did not correlate with the need for an operation. Higher ISS, lower blood pressure, lower pH, and increased number of packed red blood cell transfusions were better indicators of the need for an operation versus embolization

Retrospective study

164

29/12

55.51

n/a

n/a

n/a

n/a

n/a

n/a

104

24

36

0.96

n/a

16.74 for AE and OR

10

Three patients underwent splenectomy at 24 to 72 hours after admission for observation; three patients underwent embolization due to decrease in hematocrit 3 days after admission wound infection, fluid collection, pancreatic leak, small bowel obstruction, intra-abdominal abscess, pleural effusion, abdominal compartment syndrome, pneumonia, atrial fibrillation, myocardial infarction, thromboembolic events, acute renal, failure, and bacteremia

Margary et al. 2018[13]

Emergency CT for assessment and management of blunt traumatic splenic injuries at a Level 1 Trauma Center: 13-year study

Determine the relationship between multidetector computed tomography findings, management strategies, and ultimate clinical outcomes in patients with splenic injuries secondary to blunt trauma

Multiphase MDCT is important in splenic trauma, both in the initial assessment and in subsequent follow-up

A positive association with mortality (p<0.05) was associated with prehospital cardiac arrest, systolic blood pressure <90mmHg at the ED admission, need for transfusion of blood components, OM of splenic trauma, presence of high-grade OIS with associated head, and chest and hepatic lesions. F-NOM was not associated with an increased mortality

Single-center, retrospective, longitudinal, and observational cohort study

351 263 underwent CT scan

289/62

35.6

Motor vehicle accidents

37

67

107

45

7

178

13

160

n/a

n/a

n/a

53

23 patients failed NOM and 14 underwent splenectomy

Cadeddu et al. 2006[14]

Management of spleen injuries in the adult trauma population: a ten-year experience

Compare clinical outcomes between operative and nonoperative managements in adult patients with blunt splenic injuries and assess the changes in the patterns of managing splenic injuries in the past 10 years

The rate of nonoperative management of spleen injuries was significantly increased

The rate of mortality and complications was slightly higher in operative patients (9.3% and 47.9%), compared with those who were managed medically (6.8% and 37.9%)

Retrospective study

266

183/83

38.11 OM 37 NOM 39

n/a

n/a

n/a

n/a

n/a

n/a

148

-

118

3.22 OM 6 NOM 1

n/a

15.6 OM 21 NOM 14

21

10 failed NOM

Heuer et al. 2010[15]

No further incidence of sepsis after splenectomy for severe trauma: a multi-institutional experience of the Trauma Registry of the DGU with 1,630 patients

Determine if splenectomy is an independent risk factor for the development of posttraumatic sepsis and multi-organ failure

Non-operative management leads to lower systemic infection rates and mortality in adult patients with moderate blunt splenic injury (grade 1-3) and should therefore be advocated.

Patients with grade 4 and 5 injury, patients with massive transfusion of red blood cells and unstable patients should be managed operatively

Retrospective study

1630

1163/467

35.37

n/a

295

457

393

872

872

-

758

6.62

n/a

29.9

24.8% for splenectomy group, 22.2% for NOM group

Organ failure, sepsis, mortality infection

Bruce et al. 2011[16]

Nonsurgical management of blunt splenic injury: is it cost effective?

Analyze outcomes and cost of splenic embolization compared with surgery for the management of blunt splenic injury

Nonsurgical treatment of blunt splenic injury is safe and cost effective

Angioembolization was statistically similar to surgical therapy regarding cost

Retrospective study

236

164/72

NOM 31.2 embolization 39 surgery 53.8

Motor vehicle collision

NOM 2.5 embolization 3.6 surgery 3.9

190

31

15

Embolization 2.7 surgery 6.7

Embolization 0.3 surgery 1.6

Embolization 5.7 surgery 7.7

2

5 from NOM 3 from embolization; 1 from surgery group failed treatment

Armstrong et al. 2018[17]

Successful non-operative management of hemodynamically unstable traumatic splenic injuries: 4-year case series in a UK major Trauma Center

Report the experience of managing traumatic splenic injury at a regional major trauma center in the UK over a 4-year period, with a particular focus on the NOM of hemodynamically unstable patients

Safe application of NOM to hemodynamically unstable patients with traumatic splenic injury, particularly in those with low-grade injuries

The cohort contained only seven unstable patients with high-grade injuries, the majority of whom did undergo operative intervention. Two such patients were successfully managed non-operatively, but the small sample size precluded further analysis

Retrospective study

106

79/27

37.8

Motor vehicle crash

86

16

84

7

14

n/a

n/a

n/a

3

2 patients failed NOM and underwent SAE

Author, year

Title

Aim

Main outcome

Secondary outcome

Study design

Number of patients

Sex M/F

Mean age

Mechanism of injury

Grade of injury

Treatment

Transfusion

Hospital stay

Mortality

Complications

Fugazzola et al. 2018[18]

The need for red blood cell transfusions in the emergency department as a risk factor for failure of non-operative management of splenic trauma: a multicenter prospective study

Assess whether the hemodynamic status of the patient represents a risk factor for fNOM and if this may be considered a relevant factor in the decision-making process, especially in Centers where AE, intensive monitoring and 24-h operating room are not available. Furthermore, the presence of additional risk factors for fNOM was investigated.

Contraindication to NOM in case of hemodynamically instability in case of splenic trauma

Attention should be paid to patients with transient hemodynamic stability, including patients who require transfusion of RBC in the ED. These patients could benefit from AE; in centers where AE, intensive monitoring and an 24-h operating room are not available, this particular subgroup of patients should probably be treated with operative management

Multicenter prospective observational study

124

91/33

50

n/a

3

48

34

30

5

44

22

58

6.44

n/a

n/a

n/a

8 patients failed NOM

Davies et al. 2018[19]

Splenic artery embolization in trauma: a five-year single-center experience at a UK major Trauma Centre

Assess SAE in the center over the last five years

NOM is the mainstay of treatment for the majority of splenic injury patients

Crossover from splenectomy to SAE as the primary treatment modality for an injured spleen

Retrospective study

176

n/a

n/a

n/a

n/a

n/a

n/a

n/a

n/a

122

26

28

n/a

n/a

n/a

n/a

8 patients failed NOM, partial infarct after distal SAE (n=5, followup time: 0.12–4.6 years) and two with partial infarcts after proximal SAE (n=5, followup time: 0.03–4.9 years). No infarcts required subsequent intervention.

Norrman et al. 2008[20]

Nonoperative management of blunt splenic trauma: also feasible and safe in centers with low trauma incidence and in the presence of established risk factors

Evaluate the results following the management of blunt splenic injury in adults in a Swedish university hospital with a low blunt abdominal trauma incidence

Most patients in this study were managed conservatively with a low failure rate of NOM.

CT scan and ultrasonographic examinations during follow-up did not contribute to a change in patient management

Retrospective study

49

36/12

30.4

Traffic accident

n/a

n/a

n/a

n/a

n/a

39

n/a

10

4.8

n/a

11.2

1

3 patients failed NOM (1 was operated and 2 underwent selective embolization); 1 patient died due to major head injury

AE: angioembolization; OM: operative management, NOM: nonoperative management; fNOM: failure of NOM; SAE: splenic artery embolization; BSI: blunt splenic injury

Over the last years, a significant shift towards non-operative management of splenic trauma has been noted.‌^[10,11,14] Frandon et al. report that patients undergoing surgery present with more adverse effects, mainly due to the higher severity of their injury compared to the NOM group, thus suggesting the severity of injury as a predictor of adverse effects. ^[4] Heuer et al. even describe lower infection and mortality rates in NOM patients with grades I-III splenic injuries.^[15] Furthermore, Wahl et al. noted that patients with higher Injury Severity Score (ISS), lower Glasgow Coma Scale (GCS), and more hemodynamically unstable were more likely to be managed operatively than with angioembolization. AE was also opted for more elderly patients so as to diminish the risk of surgery and anesthesia. Moreover, the need for blood transfusions in the OR group was significantly higher than the AE group (3.4 vs. 1.5, p<0.05).^[12]

Bagaria et al. confirm the need for operative management in higher-grade injuries and highlight the need for careful patient selection for the NOM group.^[5] Armstrong et al. describe the safe application of NOM treatment even in hemodynamically unstable patients with low-grade injuries.^[17] Studies indicate the preference for splenic preservation over splenectomy^[7,9], which may be achieved with embolization^[8]. Fugazzola et al. highlight that patients in transient hemodynamic stability, even in need of blood transfusions, may benefit from splenic embolization rather than operative management.^[18] Davies et al. even remark on a crossover from splenectomy to embolization of the injured spleen as a primary course of treatment.^[19] However, splenic embolization is not a practice without complications. Excessive bleeding and infarcts post-embolization have been described in the literature.^[9,19] Clinical and laboratory evaluation should, therefore, be carefully conducted when considering splenic embolization.^[6]

Moreover, a forest plot of the success rates for NOM (conservative management without embolization) in the included studies is presented in Fig. 3. The plot reveals a wide range of success rates across the studies, from 76% to 100%. Bagaria et al. demonstrated a perfect 100% success rate of NOM; however, the sample size is relatively small, thus a wider confidence interval.^[5] Rosati et al.^[8] and Gill et al.^[9] presented with great statistical precision, with a 0.986 and 0.983 proportion of success, respectively, and a relatively big sample size, showing a narrow confidence interval. Chatting et al. reported the lowest success rate (0.762).^[6] Taking into consideration the fact that most studies reflected high proportions and tight confidence intervals, we can conclude that NOM represents a consistent and effective first-line strategy. The heterogeneity of the success rates may be attributed to injury severity and surveillance strategy.

Figure 3.

Forest plot of the success rate of NOM in the included articles.

Quality assessment

Risk of bias for the non-randomized studies was determined as low for 9 studies^{[5,7-11,13,15,18]} and as moderate in 8 ^{[4,6,12,14,16,17,19,20]}. Risk-of-bias summary is demonstrated in Fig. 4. Nearly half of the included studies are deemed of moderate bias, which may influence the conclusions. However, in our opinion, in non-randomized retrospective studies, a certain degree of confounding is to be expected.

Figure 4.

Risk of bias for non-randomized studies

Our study describes the management of traumatic splenic injuries presented in our hospital over the last decade. NOM was applied to most of our patients, following the international guidelines. Operative management was chosen for the hemodynamically unstable patients with high-grade injuries. The option of splenic embolization was not available in our hospital. In addition, our study also captures the management of traumatic splenic injuries as presented in the literature. A significant shift towards NOM is noted, and splenic embolization seems an important tool to salvage the organ.

In our experience, most traumatic spleen injuries are a result of blunt trauma in motor vehicle accidents or falls. Lower-grade injuries are admitted to the surgical department for close monitoring, daily routine blood examination, and follow-up CT at 72 hours. Higher-grade injuries are usually treated operatively, especially in the case of an unstable patient when a positive FAST is performed in the emergency department and the patient is driven directly to the operating table for exploratory laparotomy and splenectomy. In the case of a stable patient, a CT scan is performed, and if active bleeding of the splenic artery or a ruptured spleen is found, the treatment of choice is surgery. In our opinion, this is the case when an angioembolization of the actively bleeding vessel would be an acceptable, even preferable, option to surgery, but unfortunately, an interventional radiologist is not always available in our hospital. Still, the proportion of NOM treatment in our hospital is similar to that of the literature (61.7% in our hospital vs. 60.5% in the literature), with only 2 patients in our study failing NOM treatment and requiring surgical intervention (95.2% success rate of NOM). Perhaps, if a permanent interventional radiologist were available, these two patients with grade III injuries could have been treated with embolization rather than operatively. This highlights that even non-trauma centers, where also the option of an interventional radiologist may not be available, may provide adequate management in splenic trauma patients with good outcomes, in accordance with the international guidelines and the results from the literature.^[20]

The literature search is consistent towards a NOM of splenic trauma, especially in low-grade splenic trauma. Bontrager et al. pointed out the low rate of surgical intervention in low-grade blunt splenic trauma^[21], and age does not appear to be a restrictive factor for NOM application^[22]. Ruhnke et al. further highlight the low failure rate of NOM management and outline active bleeding and hemoglobin decrease as the sole indicators for operative treatment.^[23] Conservative management has even been applied to penetrating trauma patients, although spleen-preserving surgery or splenectomy is more commonly opted for, presenting similar mortality rates.^[24] Splenic repair surgery has also been suggested for patients with blunt splenic trauma and has been associated with lower mortality rates.^[25] However, still spleen-preserving surgeries are fewer than splenectomies even in low-grade injuries.^[26]

Splenic embolization is widely used even for higher-grade splenic injuries with good results. Cinquantini et al. report an overall 91% success rate of non-surgical treatment that is the combination of NOM and embolization.‌^[27] The use of splenic embolization is a safe option even for higher-grade splenic injuries, with the advantage of splenic preservation.^[12,28] Bhullar et al. reported a higher tendency for NOM treatment and a simultaneous lower tendency for OM in their ten-year cohort study. When comparing the OM and NOM groups, the former presented higher ISS and mortality scores. When comparing the two NOM groups, patients undergoing AE had higher ISS scores and higher grades of splenic injury. In addition, they reported a decreased failure of NOM treatment in higher-grade splenic injuries (grade IV-V). In their study the indications for AE were contrast blush in the CT and decreasing hemoglobin, as these were the reasons for conservative management failure.^[29] Clements et al. propose angiography for possible vascular injuries in grade III patients who are at risk for re-bleed, as in their study 18 patients with occult vascular injuries were not detected at initial CT.^[28]

However, the literature is not entirely consistent on the results of angioembolization, as it has also been described as overused^[30], and it is not described with precision when surgery or angioembolization should be performed. After all, angioembolization is an interventional approach and may lead to complications such as rebleeds, splenic abscess, infarcts, and access site pseudoaneurysms, which may require further treatment and hospitalization such as endovascular stent placement, guided drainage, and even surgery.^[31,32] Duschene et al. have even described a fourfold higher Adult Respiratory Distress Syndrome (ARDS) incidence rate in patients post embolization than post splenectomy, though the mortality rates were similar.^[33] Still, angioembolization remains a minimally invasive technique with the additional benefit of preserving the organ.

Another interesting remark in our study is the positive correlation of the number of platelets and the grade of splenic injury. In our study, two grade I and II patients received antiplatelet therapy prior to their injury and one patient during hospitalization. In the operative management group, three patients initiated antiplatelet therapy post-splenectomy. An increase in the number of platelets was found in all splenic injuries but was statistically significant only in patients with grade V splenic injuries. Thrombocytosis after splenectomy has already been established in the literature, with many patients initiating antiplatelet drug therapy.^[34,35] Chia et al. found that the platelet count seems to gradually increase in a period of 16-17 days after the splenic injury, independently of the course of treatment. Thrombocytosis is more likely to occur in patients with total splenectomy than spleen-preserving management, and it has been found to be independent of the length of stay, grade of injury, ISS score, age, and need for blood transfusion.^[34] Pommerening et al. noted that traumatic splenectomy is a hypercoagulable state that initiates at 48 hours after injury and continues for at least 5 days.^[35] Kovar et al. noticed an increasing tendency in the platelet count in all patients with splenic injuries, but this differs depending on the intervention type regardless of the possible infection.‌^[36] Platelet count should thus be closely monitored, and antiplatelet drug therapy should be initiated when needed. The possibility of protective initiation of antiplatelet drug therapy depending on the grade of injury should also be examined in future studies.

The present study is subject to several limitations. In our hospital, an interventional radiologist is not always available for splenic embolization, and thus, many patients with higher-grade injuries or with active bleeding undergo operation instead of NOM. Secondly, our sample size is small, thus limiting generalization of the results. However, according to our research, this is the first study describing the management of blunt splenic injuries in our country and is able to provide valuable insight into the outcomes of non-operative management in a real-world clinical setting. Thirdly, the risk of bias assessment of the included studies in our systematic review shows moderate bias in eight studies, which may affect the conclusions. In our opinion, though, in non-randomized retrospective studies, a certain degree of confounding is to be expected. Our study presents a few strengths. To our knowledge, this is the first systematic review of the management of blunt splenic trauma and may be used to illustrate the application of the existing guidelines of blunt splenic trauma in trauma centers internationally. Moreover, it highlights the lack of specific guidelines for splenic embolization, which may present as an effective alternative.

A shift towards NOM worldwide is noticed. Surgical intervention is restricted solely for hemodynamically unstable patients. Splenic embolization is an effective alternative and may be used to further establish non-surgical management while it also offers the benefit of splenic salvage. However, more specific guidelines regarding the selection of patients undergoing embolization are needed. Close monitoring of the platelet count and initiation of antiplatelet drug therapy are recommended.