A comparison of open and endovascular treatment strategies for the management of splenic artery aneurysms

Abstract

Splenic artery aneurysms are rare with an incidence of less than 0.8%. Evidence to support an endovascular management strategy over open surgical repair for SAA is limited. We used the Nationwide Inpatient Sample to compare open to endovascular SAA repair by assessing postoperative outcomes, length of hospital stay, and mortality. Multivariate logistic regression analysis was done to determine predictors of postoperative complications. There were 2316 admissions with a diagnosis code for SAA [347 (14.9%) endovascular repair and 112 (4.8%) open surgery]. There was a statistically significant lower rate of cardiac (2.3% vs 6.9%, P = 0.05) and pulmonary (8.9% vs 16.1%, P = 0.05) complications for the endovascular repair group. The risk of surgical site infection was also lower (0.6% vs 5.1%, P = 0.01) in the endovascular group. Median in-hospital LOS was greater for open repairs (6 vs. 4 days, P = 0.01). There were no statistically significant differences across procedures for renal complications (8.9%, P = 0.88) or in-hospital mortality (3%, P = 0.99). Regression analysis established procedure type to be independent predictor of postoperative complications. Endovascular repair of SAA is therefore associated with a lower complication rate and less resource utilization but no difference in mortality peri-operatively. This may justify an endovascular first treatment strategy in the management of SAA.

Keywords

Splenic artery aneurysm pseudoaneurysm aneurysm/therapy surgery

Introduction

Splenic artery aneurysms (SAA) are the most common visceral artery aneurysms, yet they are relatively rare with an overall estimated incidence of 0.8%.^1,2 The potential for fatal complications associated with aneurysm rupture may necessitate repair when discovered. In fact, the mortality associated with splenic artery aneurysm rupture has been reported to be as high as 30%, with an even higher rate reported in pregnant populations where the maternal mortality approaches 75% and the fetal mortality 95%.^3–5 Repair is therefore usually considered for symptomatic aneurysms, aneurysms greater than 2 cm in size, those uncovered in patients undergoing liver transplantation, or any SAA found in women of child-bearing age.^6–10

Traditionally, open surgical techniques have been used to treat or exclude SAAs. More recently, endovascular treatment has been widely accepted as an alternative for the management of these aneurysms.^11–13 However, due to the relatively low incidence of this disease, the available literature is limited to case reports and small series and there are few comparative studies. In a recently published meta-analyses of the literature by Hogendoorn et al., endovascular repair of SAA was found to have better short-term results and a lower perioperative mortality when compared to open repair although open SAA repair was associated with fewer re-interventions and better long-term outcomes.¹⁴ The purpose of our review is to compare open to endovascular SAA repair using a large nationwide database to assess postoperative outcomes, length of hospital stay, and mortality to better clarify treatment strategies.

Materials and methods

Data were collected from the Nationwide Inpatient Sample (NIS) database from the Health Care Cost and Utilization Project from 2008 to 2011. The NIS database is sponsored by the Agency for Healthcare Research and Quality, and represents a random 20% sample of all admissions to hospitals in states that require reporting. This currently represents over 1000 hospitals in 46 states, making NIS the largest all-payer administrative database in the USA. This database contains information regarding patient demographics, severity and comorbidity measures, primary and secondary diagnosis and procedures, length of stay, and discharge status.

We extrapolated our cohort via the International Classification of Diseases 9th Revision Clinical Modification (ICD-9-CM) codes. Hospitalized patients were identified with a diagnosis code for SAA (442.83) with or without hemoperitoneum upon admission (568.81). We selected those undergoing open repair with procedural codes 38.86, 38.36, 39.56, 39.57, 39.58, 39.51, and 39.52 and those undergoing endovascular repair with procedural codes 39.79 and 39.7 (Table 1). Data regarding demographics and comorbidities were collected. Postoperative complications (Table 2) were queried as secondary diagnosis and included the following systems: cardiac, pulmonary and renal. In addition, postoperative hemorrhage and surgical site infections were identified (Table 2).

Table 1.

ICD9 codes for SAA and procedure type (open vs endovascular).

442.3	Aneurysm of splenic artery
568.81	Splenic artery aneurysm with hemoperitonium(non-traumatic)
38.86	Other surgical occlusion of vessels, abdominal arteries
38.36	Resection of vessel with anastomosis, abdominal arteries
39.56	Repair of blood vessel with tissue patch graft
39.57	Repair of blood vessel with synthetic patch graft
39.58	Repair of blood vessel with unspecified type of patch graft
39.51	Clipping of aneurysm
39.52	Repair of aneurysm
39.79	Other endovascular procedures on other vessels
39.7	Endovascular repair of vessel

Table 2.

ICD9 codes for complications and other outcome measures.

997.1	Cardiac complication not elsewhere classified
410.00–410.92	Acute myocardial infarction
427.5	Cardiac arrest
997.31	Ventilator associated pneumonia
997.39	Other respiratory complication
507.0	Pneumonitis due to inhalation of food or vomitus
512.1	Iatrogenic pnemothorax
518.51	Acute respiratory failure following trauma and surgery
518.82	Other pulmonary insufficiency not elsewhere specified
480.0–480.9	Viral pneumonia…
481	Pneumococcal pneumonia
482.0–482.9	Bacterial pneumonia…
483.0, 483.1, 483.9	Pneumonia due to Mycoplasma, Chlamydia, other specified organism
484.1–484.8	Pneumonia in infectious diseases
485	Bronchopneumonia organism unspecified
486	Pneumonia organism specified
997.5	Urinary complication not elsewhere specified
584.5–584.9	Acute kidney failure…
593.81	Vascular disorders of kidney
998.11	Hemorrhage complicating the procedure
998.12	Hematoma complicating the procedure
39.41	Control of hemorrhage following vascular surgery
996.62	Infection and inflammation. Reaction due to other vascular device implant and graft
998.31	Disruption of internal operation (surgical) wound
998.32	Disruption of external operation (surgical) wound
998.51	Infection of postoperative seroma
98.59	Other post-operative infection
999.30	Disruption of wound, unspecified
39.4	Revision of vascular procedure
39.49	Other revision of vascular procedure

Demographics, comorbidities and outcomes were compared across patients undergoing open and endovascular repairs. A Chi-square test was used to analyze categorical variables, while continuous variables were analyzed using the Student t-test or Mann–Whitney U statistic. Multivariate logistic regression analysis based on age, comorbidities, and procedure type (open vs endovascular) was done to predict the presence of postoperative complications. Statistical significance was considered at a P-value of less than 0.05. All statistical analyses were performed using SPSS software (SPSS, Chicago, IL, USA).

Results

There were 2316 admissions during the study period with a diagnosis code present for SAA. Among these, 347 (14.9%) patients underwent endovascular repair and 112 (4.8%) patients had open repair. Therefore, the majority of patients admitted with a diagnosis of SAA were discharged without intervention. Of the 459 patients who underwent repair, 33 (7.2%) had a diagnosis of hemoperitoneum, with 20 of these undergoing endovascular repairs and 13 open surgery. There were three mortalities in this subgroup presenting with rupture, 1 (5%) in the endovascular group and 2 (15%) in the open group, which was a non-significant difference (P = 0.31).

Table 3 presents patient demographic characteristics, comorbidities and year of hospital discharge broken down by procedure type. There were no statistically significant differences with respect to age, race, or gender between the groups of patients undergoing open and endovascular repairs. The number of patients receiving endovascular repair was significantly higher than that undergoing open repair throughout the study period. The relative percentage of endovascular repairs did not differ based on the presence of comorbid conditions except for patients with peripheral vascular disease, who preferentially underwent endovascular repair.

Table 3.

Demographic characteristics by procedure type.

	Type of Surgical Repair
	Endovascular N = 347	Open N = 112	p-Value
Age (years; mean ± SD)	54.2 ± 15.1	56.4 ± 16.3	0.21
Demographics: N (%)
Sex:			0.18
Male	164 (47)	46 (41)
Female	183 (53)	66 (59)
Race:			0.89
White	237 (80)	74 (81)
Black	19 (6)	4 (4)
Hispanic	25 (8)	7 (8)
Asian/Pacific Islander	8 (3)	3 (3)
Native American	1 (0)	1 (1)
Other race	7 (2)	2 (2)
Year of Hospital Discharge:			0.002
2008	81 (23)	32 (29)
2009	75 (22)	38 (34)
2010	77 (22)	24 (21)
2011	114 (33)	18 (16)
Comorbidities:
Congestive Heart Failure	2 (3)	0 (0)	0.37
COPD	15 (19)	8 (25)	0.44
Diabetes-uncomplicated	11 (14)	4 (13)	0.88
Diabetes-chronic complications	1 (1)	0 (0)	0.53
Hypertension	34 (42)	12 (38)	0.66
Obesity	7 (9)	4 (13)	0.53
Peripheral vascular disease	31 (38)	2 (6)	0.001
Renal failure	4 (5)	1 (1)	0.67
Valvular disease	1 (1)	0 (0)	0.53

Statistically significant outcomes differences were observed between the two surgical groups, as shown in Table 4. Specifically, there was a significantly higher percentage of cardiac and pulmonary complications as well as a higher overall percentage of one of more of any of the complications under consideration among patients undergoing open surgical repairs than among those undergoing endovascular repairs. Additionally, there was a lower risk of surgical site infection in endovascularly treated SAA. There were no differences between groups when comparing renal or postoperative hemorrhagic complications. The median in-hospital LOS was greater for open repairs than endovascular repairs but there were no differences across procedures in the percentage of in-hospital deaths either within or beyond 30 day LOS.

Table 4.

Outcomes measures by procedure type.

	Type of surgical repair
	Endovascular N = 347	Open N = 112
Complications	N (%)	N (%)	p-value
Cardiac	8 (2.3)	7 (6.3)	0.05
Pulmonary	31 (8.9)	18 (16.1)	0.05
Renal	31 (8.9)	10 (8.9)	0.88
Postop hemorrhage	10 (2.9)	4 (3.6)	0.46
Surgical site infection	2 (0.6)	6 (5.1)	0.001
One or more of above	64 (18.4)	31 (26.5)	0.06
In-hospital deaths	9 (3)	3 (3)	0.99
In-hospital deaths (within 30 days)	8 (2)	3(3)	0.90
Hospital length of stay	4 (0–75)	6 (0–70)	<0.001
Median days (range)

Logistic regression analysis established age, CHF, renal failure, and procedure type to be independent predictors of postoperative complications (Table 5).

Table 5.

Multiple logistic regression analysis predicting presence of any complication based on age, co-morbidities, and procedure type.

Predictor variables in the regression equation	Odds ratio (95% Confidence interval)	p-Value
Congestive heart failure	8.14 (1.45, 45.80)	0.017
Obesity	1.74 (0.86, 3.49)	0.121
Renal failure	3.21 (1.32, 7.80)	0.010
Preipheral vascular disease	1.58 (0.95, 2.63)	0.075
Age (years)	1.02 (1.00, 1.03)	0.044
Endovascular procedure	0.57 (0.34, 0.98)	0.041

Discussion

SAA are the most common of all visceral artery aneurysms, but are still rare and infrequently encountered in clinical practice.¹⁵ More often, they are asymptomatic and found incidentally on abdominal imaging. SAA may present as a signet ring calcification in the LUQ on plain abdominal radiographs although most are found on abdominal CT.^16,17 These aneurysms can be associated with renal artery aneurysms and other visceral artery aneurysms, findings which have been reported 14% and 3% of the time, respectively.² SAA tend to be more common in women, with a 4:1 female to male ratio^5,18 although we found a near equal male to female distribution.⁵

The most common risk factors for SAA are portal hypertension and multiple pregnancies.⁸ Multiple pregnancies are believed to be a strong risk factor for SAA because of the hormonal changes and hepatic congestion that occurs during pregnancy.¹⁷ Other risk factors include splenomegaly, medial fibrodysplasia, atherosclerosis, autoimmune diseases, collagen vascular diseases, α-1 antitrypsin deficiency, and pancreatitis.^10,2,18,20 The majority of SAA are small and saccular and located in the middle or distal splenic artery.² If the aneurysm were to rupture, as occurs in approximately 3–10% of the cases, presentation tends to be abdominal pain with hypovolemic shock.²¹ In our study, 7.2% of the patients presented with hemoperitoneum and greater than 60% of these patients were treated endovascularly, making endovascular SAA repair a preferred modality even for emergent situations. Data concerning ruptured SAA and type of repair could not be built into multivariate analysis because the number of patients with ruptured SAA was too small to allow for any meaningful statistical analysis.

A topic of debate in recent years has been on the approach taken to treat these SAA, or if they should be treated all. Studies have shown that SAA less than 2 cm in size that were left untreated had very few complications with a survival rate of 95%.¹⁹ Hence, treatment for SAA is generally reserved for larger or symptomatic aneurysms, or for those found in liver transplant patients or women of child-bearing age. As demonstrated by our study, the move in recent years has been towards an endovascular approach for SAA treatment. For the endovascular approach, these aneurysms may either be embolized or stented.¹⁷ Embolization tends to be the preferred modality as stent grafts may be harder to insert and deploy within tortuous arteries.¹⁷ In this study, procedure type was an independent predictor of postoperative complications. In fact, we found endovascularly treated SAA’s to have a nearly 50% lower risk of pulmonary and cardiac complications, a lower risk of having more than one complication per patient, and a shorter hospital stay than SAA treated via open surgical repair. Also, we found a five-fold higher risk of surgical site infection when treating via open repair as opposed to an endovascular approach. Another factor favoring endovascular treatment is that this can be accomplished using only local anesthesia and sedation as an outpatient procedure, while open surgical repair necessitates general anesthesia and hospitalization.

As with other endovascular procedures, durability has been questioned, and data regarding long-term outcomes for endovascular approaches are limited. Our study focused on short-term outcomes and therefore does not help establish the durability of endovascular SAA repair. However, a recent report established 100% technical success rate for patients undergoing endovascular SAA repair with no aneurysm-related deaths on a median follow-up period of 45 months.²² In another publication with a mean follow-up of 15 months, aneurysm exclusion was seen on imaging studies in 97% of cases with no reports of sac growth.¹³ This study represents a retrospective analysis of the NIS database and is limited by all the inherent problems associated with the use of large databases. Commonly recognized in any administrative database, the data collection is subject to entry errors and selection and reporting bias. Additionally, we did not include any complications beyond the initial hospitalization period as these data are not included in NIS, and this might underestimate the morbidity associated with SAA repair. An additional limitation of this database is that is does not include any detail regarding the etiology or anatomy of the SAA undergoing treatment, nor are any technical aspects of repair provided. Despite these limitations, we believe this study adds support to the growing body of literature favoring endovascular treatment as the preferred approach in the management of SAA.

Conclusions

SAA are a rare but potentially fatal condition. While observation may be a reasonable approach, select patients may require open or endovascular repair. In this study, endovascular SAA repair was associated with a lower peri-operative complication rate and less resource utilization when compared to an open approach to SAA repair. We would favor endovascular treatment for patient with SAA requiring repair.

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.

References

Moore

Guida

Schumacher

. Splenic artery aneurysm. Bull Soc Int Chir 1970; 29: 210–218.

Shanley

Weinberger

. Acute abdominal vascular emergencies. Med Clin North Am 2008; 92.3: 627–647.

Abbas

Stone

Fowl

. Splenic artery aneurysms: Two decades experience at Mayo Clinic. Ann Vasc Surg 2002; 16: 442–449.

Tulsyan

Kashyap

Greenberg

. The endovascular management of visceral artery aneurysms and pseudoaneurysms. J Vasc Surg 2007; 45: 276–283.

Herbeck

Horbach

Putzenlechner

. Ruptured splenic artery aneurysm during pregnancy: A rare case report with both maternal and fetal survival. Am J Obstet Gynecol 1999; 181: 763–763.

Loffroy

Guiu

Cercueil

. Transcatheter arterial embolisation of splenic artery aneurysms and pseudoaneurysms: Short and long term results. Ann Vasc Surg 2008; 22: 618–626.

Gourgiotis

Alfaras

Salemis

. Spontaneous rupture of splenic artery aneurysm in pregnancy: A case report. Adv Med Sci 2008; 53: 341–343.

Carr

Mahvi

Hoch

. Visceral artery aneurysm rupture. J Vasc Surg 2001; 33: 806–811.

Pulli

Dorigo

Troisi

. Surgical treatment of visceral artery aneurysms: A 25-year experience. J Vasc Surg 2008; 48: 334–342.

10.

Sadat

Dar

Walsh

. Splenic artery aneurysms in pregnancy–a systematic review. Int J Surg 2008; 6: 261–265.

11.

Kasirajan

Greenberg

Clair

. Endovascular management of visceral artery aneurysm. J Endovasc Ther 2001; 8: 150–155.

12.

Saltzberg

Maldonado

Lamparello

. Is endovascular therapy the preferred treatment for all visceral artery aneurysms? Ann Vasc Surg 2005; 19: 507–515.

13.

Yamamoto

Hirota

Maeda

. Transcatheter coil embolization of splenic artery aneurysms: Therapeutic strategy using original classification. Cardiovasc Intervent Radiol 2010; 33: 292–292.

14.

Hogendoorn

Lavida

. Open repair, endovascular repair, and conservative management of true splenic artery aneurysms. J Vasc Surg 2014; 60: 1667–76.e1.

15.

Messina

Shanley

. Visceral artery aneurysms. Surg Clin North Am 1997; 77: 425–442.

16.

de Vries

Schattenkerk

Malt

. Complications of splenic artery aneurysm other than intraperitoneal rupture. Surgery 1982; 91: 200–204.

17.

Berceli

. Hepatic and splenic artery aneurysms. Semin Vasc Surg 2005; 18: 196–201.

18.

Stanley

Fry

. Pathogenesis and clinical significance of splenic artery aneurysms. Surgery 1974; 76: 898–909.

19.

Ulka

S-O

. Visceral artery aneurysms: Review of current management options. Mount Sinai J Med: J Translat Personaliz Med 2010; 77.3: 296–303.

20.

Lakin

Bena

Sarac

. The contemporary management of splenic artery aneurysms. J Vasc Surg 2010; 53.4: 958–965.

21.

Elamurugan

Kumar

Muthukumarassamy

. Splenic artery aneurysm presenting as extrahepatic portal vein obstruction: A case report. Case Rep Gastrointest Med 2011; 2011: 1–3.

22.

Kagaya

Miyata

Koshina

. Long-term results of endovascular treatment for splenic artery aneurysms. Int Angiol 2011; 30: 359–65.