Hrev_master [Emergency Care Journal 2017; 13:6753] [page 37] Percutaneous artery embolization of bleeding rectus sheath hematomas in hemodynamically unstable patients: Outcomes of 43 patients in a tertiary referral hospital Alberto Cereda,1,2 Paolo Aseni,3 Lucio De Capitani,4 Marco Cereda,3 Ruggero Vercelli,5 Cristina Giannattasio,1,2 Antonio Gaetano Rampoldi5 1Cardiology IV, Cardiotoracovascular A. De Gasperis Department, Niguarda Cà Granda Hospital, Milan; 2School of Medicine and Surgery, University of Milano-Bicocca, Milan; 3Emergency Department, Niguarda Cà granda Hospital, Milan; 4Department of Statistics and Quantitative Methods, Bicocca University, Milan; 5Interventional Radiology, Niguarda Cà Granda Hospital, Milan, Italy Abstract Rectus sheath hematoma (RSH) is an uncommon cause of abdominal pain that can lead to life-threatening bleeding, particularly in elderly patients receiving anticoagulation therapy. Type III RSHs, based on computer tomography (CT) evaluation, is character- ized by active bleeding with intramuscular or intraperitoneal extension and hemodynamic instability. Medical contemporary knowl- edge of this condition is contentious and overall 20% of acute mortality has been reported. The purpose of this study was to retrospectively review our experience and outcomes in the management of RSHs treat- ed with percutaneous arterial embolization of the epigastric vessels. We retrospectively analyzed 43 patients with RSH type III, sub- mitted to percutaneous epigastric artery embolization from 2007 to 2015. Percutaneous arterial embolization was fea- sible and successful in patients with a high burden of comorbidities and receiving anti- coagulation therapy. There was no acute mortality and a late mortality at 3 months of 9.1% (4/43) was not directly related to RSHs or arterial embolization. Patients with late mortality had lower ejection fraction, pro- longed PTT, greater RDW and warfarin in overlapping with low-molecular-weight heparins. Transcatheter arterial embolization of the epigastric vessels was safe, feasible and effective in stopping the bleeding. Despite anti-platelets/anti-coagulation thera- py and a high burden of comorbidities, there was no acute mortality related to RSH. Introduction Rectus sheath hematoma (RSH) is an uncommon condition resulting from blood accumulating within the sheath of the rectus abdominis secondary to the disruption of the epigastric arteries.1 Coughing spells, iatrogenic injuries and trauma in the back- ground context of an elderly anticoagulated patient with muscle tissue atrophy and frailty due to comorbidities, can result in a relentless bleeding.2-4 Based on radiological evaluation by CT, Berna et al.5,6 differenti- ate RSH into three types: type I and type II hematomas, are intramuscular and self-lim- iting. Type III is characterized by a marked intramuscular or intraperitoneal hematoma with active bleeding and hemodynamic instability. Treatment depends on the RSH classification type, patient’s underlying medical conditions and identification of potential complications. Conservative treat- ment such as rest, analgesics, and reversal of anticoagulation are utilized for nonex- panding and small hematomas (RSH type I and II), while blood products and fluid resuscitation are needed for RSH type III. The mainstay of treatment for RSH unre- sponsive to conservative therapy has been surgical exploration and ligation of the bleeding vascular lesions. However, the mortality rate of elderly patients who under- go abdominal surgery is very high with a risk of re-bleeding from impaired hemosta- sis.7 In 1980, Levy first reported the trans- catheter gelfoam embolization technique in the treatment of rectus sheath hematoma.8 This interventional therapy can produce hemostasis of the bleeding vessels, reduce the size of the hematoma and decrease the need for blood transfusions. Arterial embolization is a promising therapeutic option for bleeding from a variety of sources, mainly in the abdomen and pelvis and embolization through a coaxial micro- catheter enables selective distal emboliza- tion of small-vessel bleeding points. Although the mainstay of treatment for rec- tus sheath hematomas with hemodynamic compromise has historically been the surgi- cal option, trans-catheter arterial emboliza- tion of the bleeding arteries has been sug- gested as a fast, less invasive and resource- ful technique.9-11 Aim of the study The purpose of this retrospective study was to analyze our experience and out- comes of patients with RSHs type III sub- mitted to percutaneous arterial embolization of the epigastric vessels. Materials and Methods We retrospectively analyzed the clinical records of 43 consecutive patients with RSH affected by RSHs and evaluated at our hospital, a tertiary referral center for trauma and medical emergency, from January 2007 to December 2015. We reviewed all the interventional radiological procedures of percutaneous epigastric artery embolization in patients with RSH type III. The inclusion criteria were the followings: clinical diag- nosis of RSH type III associated with hypotension or hemodynamic instability not manageable only with conservative medical therapy, acute bleeding necessitating fluid resuscitation and/or blood components, documentation of active bleeding at abdom- inal CT scan. The following clinical and radiological variables were collected and recorded in a dedicated dataset: age, etiolo- gy of RSH (spontaneous, trauma, post-para- centhesis, post endovascular trans-femoral procedures, heparin related), left ventricular ejection fraction (EF), HASBLED score (a cardiac risk score used to estimate the 1- year risk for major bleeding in patients with atrial fibrillation), laboratory values just before embolization: hemoglobin, hemat- ocrit, red cell distribution width (RDW%), platelets, renal function, coagulation (PT and PTT), albumin and total protein. RSH was defined spontaneous in absence of a specific cause and in absence of a history of trauma. Heparin-related RSH were defined when one or more RSH developed at the Emergency Care Journal 2017; volume 13:6753 Correspondence: Alberto Cereda, Cardiology IV, Cardiotoracovascular A. De Gasperis Department, Niguarda Cà Granda Hospital, Milan, Italy. Tel: +39.3200883375. E-mail: tskcer@hotmail.it Key words: rectus sheath hematoma; per- cuteanous arterial embolization; epigastric artery; life threatening abdominal bleeding; anticoagulation. Received for publication: 15 April 2017. Accepted for publication: 25 May 2017. This work is licensed under a Creative Commons Attribution 4.0 License (by-nc 4.0). ©Copyright A. Cereda et al., 2017 Licensee PAGEPress, Italy Emergency Care Journal 2017; 13:6753 doi:10.4081/ecj.2017.6753 Non co mmerc ial us e o nly site of low-molecular-weight heparin (LWMH) injections and other possible causes were excluded. All past medical his- tory were reviewed and the following comorbidities were taken into account: car- diac disease (hypertensive, ischemic, valvu- lar, pulmonary embolism, history of atrial fibrillation), chronic liver disase in patients with liver failure requiring recurrent para- centhesis, chronic renal disease, evidence of oncological disease during the past five years. Digital subtraction angiography was performed through the common femoral artery after administration of local anesthet- ic on the contralateral side to the hematoma. Vascular access was obtained by placement of a 5-French sheath. The choice of the diagnostic catheter was based on the anato- my of iliac vessels. A supraselective catheterization of the affected artery through a 3-French coaxial microcatheter system was used to embolize the distal artery with synthetic glue and/or spiral coils to stop the bleeding. Statistical analyses were performed using R software. Results are expressed as mean±standard deviation (SD). Measurements are standardized for body surface area. Data are presented as mean±SD and categorical data are present- ed as number and percentage. The differ- ences of the mean of continuous variables of patients survived and dead after three months have been compared by using the exact Fisher-Pitman permutation test. All the performed statistical tests are one-tailed and P<0.05 was chosen to establish statisti- cal significance. Results The mean age of the population with RSH was 70±13 with a slight female preva- lence (55%). Patients had a mean HAS- BLED score of 3 (Table 1). There was no acute mortality related to RSH, survival at 3 month was 91.9% (39/43 patients). Cause of death in four patients within 3 months were related to comorbidi- ties and not directy to RSHs or the embolization procedure. Deceased patients had a lower ejection fraction (EF 40.5±20.6% versus 56.1±3.9%, P=0.006), a prolonged PTT (PTT 71.2 secs versus 41.8±18 secs, p value 0.011), valvular car- diac disease (75% versus 17.9%, P=0.04) unrelated to the type of cardiac prosthesis (biologic or mechanical). All deceased patients were on warfarin and 3 out of 4 were on bridge therapy with LWMH plus warfarin. The etiologies of RSH were the follow- ing: endovascular procedures (4.7%), para- centhesis (9.3%), subcutaneous heparin injection (13.9%), post-surgery (16.3%), trauma (18.6%) and spontaneous (37.2%). Our population is characterized by a high burden of comorbidities, 32.6% have a chronic liver disease, 20.9% an oncological disease, 55.8% were suffering from heart disease. The majority of patients (60.4%) were in therapy with antiplatelets (aspirin or clopidogrel) and/or OACs. Six out of 43 patients were in therapy with both heparin and warfarin as a bridge therapy (Table 2). Six patients had a bilateral epigastric artery involvement and there were no differ- ences between right and left epigastric ves- sels involvement. Embolization was per- formed with gelfoam in all patients and in 11 cases spiral coils were added to obtain a more stable hemostasis. In 4 patients a sur- gical approach was chosen: in two patients, surgery treatment as first choice was unsuc- cessful and embolization was used as rescue therapy, in only two patients percutaneous embolization failed to stop the bleeding and a surgical approach was performed as sal- vage therapy (Table 3). Discussion RSH is an uncommon and misdiag- nosed potentially life-threatening condition.13 Our patients with RSH have an high prevalence of liver disease (32.6%), oncological disease (20.9%) and have a car- diac condition from which derived in most cases the indication to antiplatelets and/or OACs. None of our patients were in therapy with new oral anticoagulants.14 Despite comorbidities, in our work frailty is empha- sized by a low level of albumin and a higher values of RDW%. Frailty is a biological syndrome that not solely results in deterioration of one single system, but multiple systems such as mus- culoskeletal, cardiovascular, metabolic or immunological systems. Serum albumin is the main protein synthesized by the liver and is considered a good surrogate of frailty and low plasma albumin is associated with sarcopenia.15 Nevertheless, there was no difference in albumin levels between the survived and dead (Albumin: 3.27±1.03 g/dL in survived versus 3.75±0.5 in deceased, ns). Instead of albumin, RDW was higher in patients with RSH that do not survived (RDW: 16.1±2.8 versus 20.5±4.1%, P=0.011). RDW% is a quantitative measure of variability in the size of circulating ery- throcytes with higher values reflecting greater heterogeneity in cell sizes. RDW has been shown to predict mortality and adverse outcomes in various clinical condi- tions, chronic and acute disease. It not clear however if higher RDW values in decesead patients may be explained by acute hemor- rhage or the chronic health status, both of them potentially linked to mortality.16,17 All non-surviving patients at three months after RSH (4 patients) were in ther- Article Table 1. Demographics and etiology of patients with rectus sheath hematomas. All patients Alive after 3 months Deceased within 3 months P Patients (n) 43 39 4 Age (years) 70.3±13.6 69.7±14.2 75.5±3.4 ns Male/female ratio 19/24 18/21 1/3 ns Cardiac EF (%) 54.7±8 56.1±3.9 40.5±20.6 ns HAS-BLED score 3 (1-5) 3 (1-5) 3 (3-5) ns Etiology of RSH Spontanous 16/43 (37.2%) 15/39 (38.5%) 1/4 (25%) ns Traumatic 8/43 (18.6%) 8/39 (20.5%) - - Post abdominal surgery 7/43 (16.3%) 7/39 (17.9%) - - Heparin (LMWHs) 6/43 (13.9%) 4/39 (10%) 2/4 (10%) ns Paracenthesis 4/43 (9.3%) 4/39 (10.3%) - - Femoral endovascular procedure 2/43 (4.7%) 1/39 (2.6%) 1/4 (25%) ns EF, ejection fraction; RSH, rectus sheath hematoma; LMWHs, low-molecular-weight heparins. [page 38] [Emergency Care Journal 2017; 13:6753] Non co mmerc ial us e o nly [Emergency Care Journal 2017; 13:6753] [page 39] apy with at least one antiplatelets and/or OACs. There was a great prevalence in the usage of warfarin and LWMHs plus war- farin in the non-surviving patients. In our study, we observed that bridge therapy with LMWHs and warfarin was used in 3/4 patients that died within 3 months (LWMHs + Warfarin: 3/39 in surviving and 3/4 in decesead patients, P=0.006). Bridging ther- apy with heparin is controversial, although many guidelines recommend interruption of oral anticoagulation therapy and the use of bridging therapy with subcutaneous low- molecular-weight heparin around the time of many kinds of surgeries, this strategy is associated with an increased risk of bleed- ing. Although anticoagulation is not proba- bly the cause of RSH but only a possible precipitating risk factors. Oral anticoagula- tion and in particular LMWHs in overlap- ping with warfarin, could be a sword of Damocles hanging over frail and elderly patients injured by the double-edged sword of anticoagulation. Due to the paucity of medical literature, morbidity and mortality rates of RSH are difficult to estimate. The mortality rate for iatrogenic rectus sheath hematoma is reported to be 18%, whereas the mortality rate for patients undergoing anticoagulation therapy is reported to be 25% regardless of embolization and/or sur- gery and higher in presence of hemodynam- ic instability. In our work, in an high risk population of elderly patients, with RSH type III hematomas treated with percuta- neous transcatheter embolization, the mor- tality was lower than that reported in litera- ture, only 4 patients over 43 died within Article Table 2. Laboratory values and comorbidities of patients with rectus sheath hematomas. All patients Alive after 3 months Deceased within 3 months P Laboratory values Hemoglobin (g/dL) 8.9±1.4 9.1±1.3 8.25±1.7 ns Hematocrit (%) 27.7±3.6 27.6%±3.8 28.25±2.6 ns Platelets (103/µL) 172.5±87.3 174±87 163±97 ns RDW (%) 16.7±3.4 16.1±2.8 20.5±4.1 0.011 Creatinine (mg/dL) 1.14±0.5 1,20±0.57 1.24±0.59 ns Azotemia (mg/dL) 58.73±42.8 58±44.5 63.6±33.3 ns PT (sec) 1.6±0.6 1.45±0.6 1.75±0.9 ns PTT (sec) 46.6±26.1 41.85±18.2 71.25±46 0.04 Albumin (g/dL) 3.24±0.8 3.27±1.03 3.75±0.5 ns Total sieric protein (g/dL) 6.1±0.8 6.0±0.8 6.25±0.5 ns Comorbidities Chronic liver disease 14/43 (32.6%) 13/39 (33.3%) 1/4 (25%) ns Neoplasia 9/43 (20.9%) 9/39 (23.1%) - - Cardiac disease 24/43 (55.8%) 20/39 (51.3%) 4/4 (100%) ns Type of cardiac disease Hypertensive 7/43 (16.3%) 7/39 (17.9%) - - Ischemic 4/43 (9.3%) 3/39 (7.7%) 1/4 (25%) ns Valvular 10/43 (23.3%) 7/39 (17.9%) 3/4 (75%) 0.03 Pulmonary embolism 3/43 (7%) 3/39 (7.7%) - - Hystory of atrial fibrillation 16/43 (37.2%) 14/39 (35.9%) 2/4 (50%) ns Cardiac valvular prosthesis 5/43 (11.6%) 14/39 (35.9%) 1/4 (25%) ns Mechanical cardiac prosthesis 1/43 (2.3%) 0/39 1/4 (25%) ns RDW, red cell distribution width; PT, prothrombin time; PTT, partial thromboplastin time; ns, not significant. Table 3. Antiplatelets and/or anticoagulants therapy of patients with rectus sheath hematomas. All patients Alive after 3 months Deceased within 3 months P Antiplatelets and anticoagulants Aspirin 4/43 (9.3%) 3/39 (7.7%) 1/4 (25%) ns P2Y antagonist (Clopidogrel) 4/43 (9.3%) 4/39 (10.3%) - ns Heparin (LMWH) 16/43 (37.2%) 13/39 (33.3%) 3/4 (75%) ns Warfarin 13/43 (30.2%) 9/39 (23.1%) 4/4 (100%) 0.005 LMWH + Warfarin 6/43 (13.9%) 3/39 (7.7%) 3/4 (75%) 0.006 Epigastric artery involvement Right epigastry artery 23/43 (53.5%) 22/39 (56.4%) 1/4 (25%) ns Left epigastric artery 14/43 (32.5%) 13/39 (33.3%) 1/4 (25%) ns Bilateral (left and right) 6/43 (14%) 4/39 (10.3%) 2/4 (50%) 0.03 Embolisation with hemostatic glue 43/43 (100%) 39/39 (100%) 4/4 (100%) ns Embolisation with spiral coil 11/43 (25.6%) 10/39 (25.6%) 1/4 (25%) ns Surgical treatment* 4/43 (9.3%) 3/39 (7.6%) 1/4 (25%) ns Acute mortality 0/43 0/43 0/43 - Late mortality within 3 months 4/43 (9.3%) 0/39 4/4 (100%) ns LMWH, low-molecular-weight heparin. Embolization was performed in all patients with hemostatic glue, in 11 patients spiral coils were added to obtain a safer hemostasis. *In 2 patients surgery was chosen as first treatment, a subsequent percutaneous embolization was needed. In 2 patients percutaneous embolization failed to stop the bleeding and a surgical approach was performed as rescue therapy. Non co mmerc ial us e o nly [page 40] [Emergency Care Journal 2017; 13:6753] three months (late mortality 9.3%). Specifically, none of the four patients died directly from RSH: one patients #1 (with a mechanical prosthetic mitro-aortic valve demanding treatment with warfarin despite the high risk of re-bleeding) died of a disseminated intravascular coagulation from a gastro-intestinal bleeding 14 days after a successful embolization of RSH, the second patient #2 developed an abdominal compartmental syndrome from RSH and died 45 days later for a multi-organ failure (Figure 1), the third patient #3 died of septic shock after a prolonged hospital stay after two months, the fourth patient died for heart failure after three months. The percutaneous transcatether embolization with hemostatic gelfoam and/or spiral coils was technically feasible and effective in stopping the bleeding in 41/43 patients, in only two patients surgery was deemed to stop the bleeding. In conclusion, although surgical evacu- ation of hematoma and ligation of bleeding vessels are considered the standard surgical procedure for severe RSH, our retrospective study suggests that percutaneous emboliza- tion could be a safe and feasible therapeutic option associated with a low mortality in an high risk frail population with RSH type III and could be the first line therapy followed by surgery in case of failure (suggested flow-chart algorithm; Figure 2). Limitations of the study Our case-control retrospective study is limited by a relative small number of patients sampled, though in the context of an unusual and underreported condition, without a group of control. Moreover, refer- ral center bias should be also considered as patients were evaluated in a tertiary referral center hospital. Conclusions In 43 patients evaluated a RSH type III in a tertiary referral hospital, transcatheter arterial embolization of the epigastric ves- sels was safe, feasible and effective in stop- ping the bleeding. Despite anti- platelets/anti-coagulation therapy and a high burden of comorbidities, there was no acute mortality related to RSH, survival at 3 month was 91.9% (39/43 patients). Deaths before 3 months were related to comorbidi- ties, not directly to RSH. Patients who died within 3 months (4/43, mortality 9.1%) have a lower ejection fraction, prolonged PTT, an increased RDW and were more often on warfarin in overlapping with LMWHs. Article Figure 1. Computed tomography (CT) imaging of the bilateral rectus sheath hematomas (RSHs) of patient #2. He was 72 years old with a known cardiac ischemic disease (ejection fraction 25%), and he was admitted for a persistent atrial fibrillation. Low-molecular- weight heparins (LWMHs) and warfarin were initiated. After 2 days there were perium- bilical and flank ecchymosis at the sites of LMWHs injections due to RSHs (A with intraperitoneal extension (white arrowhead). RSHs were complicated from an abdominal compartment syndrome and acute renal failure. CT images of RSHs without and with contrast (A and B). Ultrasound image of one RSH (C). Abdominal sagittal plane (D) and intraperitoneal expansion of RSH (E). Angiography of the right epigastric artery with several bleeding points encircled (F). Figure 2. Proposed and flowchart of rectus sheath hematoma (RSH) treatment. Clinical examination is crucial for diagnosis, risk stratification and hemodynamic stability. In unstable patient, after fluid and blood resuscitation, computed tomography (CT) with contrast should be indicated to rule out the etiology bleeding and confirming the diag- nosis of RSH. In a case of active bleeding recognized from contrast blushing on CT, we suggest percutaneous arterial embolization as first approach, if not successful emboliza- tion could be repeated otherwise referring patients to surgery. Non co mmerc ial us e o nly [Emergency Care Journal 2017; 13:6753] [page 41] References 1. Teske JM. Hematoma of the rectus abdominis muscle: report of a case and analysis of 100 cases from the litera- ture. Am J Surg 1946;71:689-95. 2. Zainea GG, Jordan F. Rectus sheath hematomas: their pathogenesis, diagno- sis, and management. Am Surg 1988; 54:630-3. 3. Cherry WB, Mueller PS. Rectus sheath hematoma: review of 126 cases at a sin- gle institution. Medicine (Baltimore) 2006;85:105-10. 4. Levine MN, Raskob G, Landefeld S, et al. Hemorrhagic complications of anti- coagulant treatment. Chest 2001; 119:108S. 5. Berna JD, Garcia-Medina V, Guirao J, et al. Rectus sheath hematoma: diagnos- tic classification by CT. Abdom Imaging 1996;21:62-4. 6. Berna JD, Zuazu I, Madrigal M, et al. Conservative treatment of large rectus sheath hematoma in patients undergo- ing anticoagulant therapy. Abdom Imaging 2000;25:230-4. 7. Hildreth DH. Anticoagulation therapy and rectus sheath hematoma. Am J Surg 1972;124:80-6. 8. Levy JM, Gordon HW, Pitha NR, et al. Gelfoam embolisation for control of bleeding from rectus sheath haematoma. AJR Am J Roentgenol 1980;135:1283-4. 9. Rimola J, Perendreu J, Falco J, et al. Percutaneous arterial embolization in the management of rectus sheath hematoma. AJR Am J Roentgenol 2007;188:497-502. 10. Zissin R, Gayer G, Kots E, et al. Transcatheter arterial embolisation in anticoagulant-related haematoma – a current therapeutic option: a report of four patients and review of the litera- ture. Int J Clin Pract 2007;61:1321-7. 11. Wong JH, Ng SS, Ho SS, et al. Transcatheter arterial embolization of spontaneous rectus sheath haematoma in a Chinese woman. Asian J Surg 2008;31:36-9. 12. Fitzgerald JEF, Fitzgerald LA, Anderson FE, et al. The changing nature of rectus sheath haematoma: case series and literature review. Int J Surg 2009;7:150-4. 13. Hoefflinghaus T, Landolt L, Bachli EB. Rectus sheath hematoma - a rare and dangerous complication of anticoagula- tion therapy. A case report and review of current literature. Open J Int Med 2014;4:87-94. 14. Sheth HS, Kumar R, Di Nella J, et al. Sheth evaluation of risk factors for rec- tus sheath hematoma. Clin Appl Thromb Hemost 2016;22:292-6. 15. Wysokinski WE, McBane RD. Periprocedural bridging management of anticoagulation. Circulation 2012;126: 486-90. 16. Wagner D, De Marco MN, Amini N, et al. Role of frailty and sarcopenia in pre- dicting outcomes among patients under- going gastrointestinal surgery. World J Gastrointest Surg 2016;27:27-40. 17. Zhang Z, Xu X, Ni H, et al. Red cell dis- tribution width is associated with hospi- tal mortality in unselected critically ill patients. J Thorac Dis 2013;5:730-6. Article Non co mmerc ial us e o nly