Single-Dose Enoxaparin for Portomesenteric Venous Thrombosis Prophylaxis after Sleeve Gastrectomy

Abstract

Introduction:

Portomesenteric venous thrombosis (PMVT) may complicate sleeve gastrectomy. We believe that single dose of enoxaparin postoperatively can reduce the risk of PMVT.

Objective:

The objective was to study the outcomes of enoxaparin single dose compared to other perioperative prophylactic doses in preventing PMVT.

Methods:

Participants included 590 patients who underwent laparoscopic sleeve gastrectomy (LSG). These retrospective cohort data were collected from patient medical charts after bariatric surgery. Patients were followed up in the close postoperative period and at 1, 3, 6, 12, and 18 months. Descriptive statistical analysis was carried out. The objective was to estimate the incidence of PMVT with postoperative single 40 mg subcutaneous enoxaparin prophylactic regimen.

Results:

From January 2017 to December 2021, 590 patients with obesity underwent LSG. Five patients developed PMVT with an estimate incidence of 0.85%. Three patients had unexplained tachycardia and three patients had postoperative bleeding.

Conclusions:

Single-dose enoxaparin 40 mg is an effective thrombosis prophylaxis without increasing risk of bleeding.

INTRODUCTION

Bariatric surgical procedures provide greater weight loss and more significant relief of obesity-related comorbidities such as type 2 diabetes, hypertension, hyperlipidemia, arthralgia, and obstructive sleep apnea.[1] However, it may cause serious complications, such as leak, bleeding, malnutrition, venous thromboembolism (VTE), and portomesenteric venous thrombosis (PMVT). The incidence of thromboembolic complications after bariatric procedures is 0.2% to 2.4%.[2] The pooled overall incidence of PMVT complicating bariatric procedures is 0.419% (95% confidence interval: 0.341%–0.505%) with an estimated mortality rate of 1.33%.[3] It seems that there is a statistically unexplored PMVT increased incidence after laparoscopic sleeve gastrectomy (LSG) in comparison with other bariatric surgery operations.[4]

Obesity per se is a hypercoagulable state causing proinflammatory mediators release, clotting factors level increase, and fibrinolysis reduction.[5] Other risk factors for PMVT include portal hypertension, myelodysplastic disorders including malignancy, intra-abdominal inflammatory diseases,[6] intra-abdominal surgical manipulation, intra-abdominal pressure elevation, splanchnic endothelium damage,[7] and genetic disorders such as factor V Leiden, protein C, and protein S deficiency.[8]

The diagnosis of PMVT in the immediate postoperative period is usually difficult and requires a high index of suspicion as symptoms are nonspecific such as abdominal pain, nausea, vomiting, and distension and physical findings may include low-grade fever, abdominal tenderness, and paralytic ileus. The onset may appear as soon as the 1^st postoperative day, and the diagnosis is made at variable days or weeks postoperative.[9]

There is no link between VTE and portomesenteric venous thrombosis (PMVT) regarding pathogenesis, prophylaxis, management, and complications; therefore, the prophylaxis for deep venous thrombosis (DVT) used for surgical patients cannot be applied to PMVT at the same level of confidence.[9]

Recommended prophylactic measures against VTE range from nonpharmacologic management such as mechanical compression devices and early ambulation alone to the use of anticoagulants and inferior vena cava filters.[10] Low-molecular-weight heparin (LMWH) such as enoxaparin has a longer half-life, higher bioavailability, a more predictable anticoagulant response, and a lower risk of thrombocytopenia than unfractionated heparin.[11] Enoxaparin is therefore widely used in bariatric surgery for thromboprophylaxis.[12]

METHODS

This was a retrospective cohort study of all patients with obesity operated on for LSG from January 2017 to December 2021. Data were obtained from the hospital medical records. Exclusion criteria included patients operated previously for morbid obesity, patients operated with other bariatric procedures than LSG, or patients on anticoagulation treatment for any cause before surgery. Informed consent was obtained from all patients to take part in the study. All patient data remained confidential. Ethical approval was obtained from the Ethics Committee and Institutional Review Board (IRB) (Number: 62022).

All the patients underwent LSG under general anesthesia; the operations were performed by two surgical teams in both hospitals under similar conditions of preoperative evaluation, preparation, pharmacologic anticoagulation prophylaxis, surgical technique, postoperative care, and follow-up.

LSG was performed in the standard French position using five ports and an intra-abdominal CO2 insufflation pressure of 15 mmHg. The first team used a liver retractor (Mediflex Surgical Products, Islandia, NY, USA) only when the left liver lobe was bulky, while the second team used a liver retractor in all patients. Devascularization of the stomach greater curvature was carried out starting 4 cm from the pylorus up to the angle of His using electrocautery LigaSure (Covidien, Boulder, CO). A 36-Fr calibrating bougie was then passed to the stomach and the duodenum before making the gastric sleeve with a linear stapler. The calibrating bougie was then pulled proximally and methylene blue was instilled into it to check for leaks. No reinforcing sutures were inserted over the stabling line. No intra-abdominal drains were inserted. Intraoperative VTE prophylaxis used were calf pneumatic compression devices. No complications occurred during the procedures and no conversion from laparoscopic to open surgery was needed. To prevent DVT, all patients were instructed to ambulate early, preserve hydrated, and wear elastic stockings. Single-dose enoxaparin 40 mg given subcutaneously 12 h after surgery was used for thromboprophylaxis in all cases. No complications were observed during hospital stay. All the patients were discharged 24–48 h after surgery taking adequate oral fluids. Patients were followed up at 1, 3, 6, 12, and 18 months.

PMVT was defined as total or near-total obstruction of blood flow to the liver secondary to thrombus formation. This thrombus may extend proximally to the liver and involve intrahepatic portal veins or may propagate distally involving the splenic or mesenteric veins. Abdominal ultrasound scanning (USS) or computed tomography (CT) scanning was used to establish the diagnosis.

The main outcome measure was the occurrence of PMVT in patients with obesity who underwent LSG. Descriptive statistical analysis was carried out. The incidence of intraoperative and early postoperative bleeding from the surgical site and unexplained tachycardia was estimated.

RESULTS

There were 590 patients in this study, 204 (28%) males and 534 (72%) females, aged 12-62 years with a mean of 41.5 (+10 years) who underwent LSG from January 2017 to December 2021. The average BMI was 47.5 with a range of 36.9–86.2 kg/m². Five patients (3 females, 2 males) developed PMVT with an estimated incidence of 0.85%. The median age was 35 years (range: 27–46). The median interval to diagnosis of PMVT was 5 days (limits: 2 to 15 days) postsurgery. In the first surgeons team group, one patient was found to have protein S deficiency and another patient found to be heterozygous for mutant factor V Leiden deficiency postoperatively, liver retractor was not deployed intraoperatively in both cases to avoid liver operative trauma. The patient characteristics are detailed in Table 1. In this study, PMVT sites were found to include splenic vein as seen in Figure 1, right portal vein, confluence of splenic, portal, superior mesenteric vein (SMV), and right portal vein [Figure 2], SMV, and confluence between the splenic and portal veins [Figure 3].

Table 1.

Patient characteristics

Age	Sex	BMI	Comorbidities	Postoperative day of Dx	Clinical presentation	Dx	Investigated for hypercoagulation	Site of thrombosis
43	Female	43	5	Vague abdominal pain	CT	Yes	Splenic vein [Figure 1]
46	Male	52	HT	5	Abdominal pain	CT	Yes	Right portal vein
35	Male	38	DM	3	Abdominal pain	CT	No	Confluence of splenic, portal, SMV, and right portal vein [Figure 2]
27	Female	49.8	-	15	Abdominal pain	US	No	Superior mesenteric vein
30	Female	42	-	2	Abdominal pain	CT	No	Confluence between n the splenic and portal veins [Figure 3]
Age	Type	Management	Clavien–Dindo grade	Days of hospitalization
43	Protein S deficiency	Clexane 40 mg SC for 1 week then warfarin 5 mg for 6 months	II	2
46	Heterozygous for mutant factor V Leiden deficiency	Warfarin	II	5
35	-	Anticoagulation	II	3
27	-	Anticoagulation	II	2
30	-	Anticoagulation	II	5

BMI=Body mass index, CT=Computed tomography, HT=Hypertension, DM=Diabetes mellitus, US=Ultrasound, SC=Subcutaneous

Figure 1.

Computed tomography scan of patient no. 1 showing central filling defect (arrow) which is noted in the splenic vein causing its incomplete obstruction

Figure 2.

Computed tomography scan of patient no. 3 showing filling defect (arrow) at the confluence of the splenic, portal, superior mesenteric vein extending to the right portal vein which also appears to be distended

Figure 3.

Computed tomography scan of patient no 5 showing a large filling defect (arrow) in the main portal vein which appears to be distended and dilated

No patients developed DVT or pulmonary embolism (PE). Three patients had unexplained tachycardia and three patients had postoperative bleeding.

All patients with PMVT presented with variable severity of vague central abdominal pain with and without associated symptoms such as malaise, nausea, vomiting, and fever. All but one was diagnosed by abdominal CT scan with intravenous and oral contrast while one patient was diagnosed with abdominal USS; his diagnosis was delayed relative to the other patients due to subtle presentation (mild abdominal pain). All the patients were treated with anticoagulation successfully and no intravascular thrombolysis was needed. Follow-up by abdominal CT showed complete recanalization, and the patients had no more complaints.

DISCUSSION

PMVT is not encountered frequently. However, it is associated with high morbidity (mesenteric ischemia in 5%–15% of cases) and mortality rates (20%–50%).[13] Since 1991, PMVT have been reported after different laparoscopic surgeries some of which had no direct intervention on the portal venous system.[14] PMVT has been reported after splenectomy, liver transplantation, and Roux-en-Y gastric bypass (RYGB).[15,16] PMVT as a complication of LSG was first reported as a thrombophilic patient by Berthet et al. in 2009.[17]

The incidence of PMVT after LSG is higher than that after RYGB and laparoscopic adjustable gastric banding.[5,18] Possible causes for this include liver congestion due to prolonged use of liver retractor,[19] gastric and splenic venous reflux after ligation of short gastric vessels and gastroepiploic vessels, and mechanical or thermal damage to the splenic vein by an energy dividing and sealing device during release of the greater omentum and posterior gastric tissue from the stomach.[20]

Data from general surgery studies are used for comparison of LMWH use with no pharmacologic thromboprophylaxis after obesity surgery since comparative studies in the latter group are limited.

In this series, the incidence of PMVT after LSG was 0.85%. It is congruent with the findings of other series with a PMVT range of 0.078%–1.0% shown in Table 2 and using different types of pharmacologic and nonpharmacologic VTE prophylaxis including different doses and duration of enoxaparin pre- and postoperatively. Many systematic reviews postulated that the incidence of PMVT after bariatric surgeries varied from 0.13% to 1%.[37]

Table 2.

Some previous studies show the incidence of portomesenteric venous thrombosis and the type and dose of thromboprophylaxis after laparoscopic sleeve gastrectomy

Study year (reference)	Study years	Study type	Number of patients	Patients developed PMVT (%)	Thromboprophylaxis
Salinas et al. 2014[21]	2005–2011	Retrospective On prospective database	1713	17 (1)	Enoxaparin 40 mg SC 12 h after surgery until discharge
Bellanger et al., 2010[22]	-	Case series	>800	3 (0.38)	N/A
Wysocki et al., 2020[23]	2014–2018	Retrospective	1192	2 (0.17)	ERAS protocol
Godoroja et al., 2019[24]	2014–2019	Retrospective	3861	3 (0.078)	DVT prophylaxis which continued 21 days postoperatively or anticoagulation treatment followed by bridging to oral therapy when indicated
Bucheeri and Abulsel, 2019[25]	2012–2017	Retrospective	769	2 (0.26)	Pre- and postoperative 5000 IU heparin was started 8 h after the operation and administered every 8 h switched to enoxaparin 40 mg on the first postoperative day
Johari et al., 2020[26]	2008–2015	Retrospective	259	2 (0.78)	N/A
Bani Hani et al., 2019[27]	2010–2019	Retrospective	4900	7 (0.14)	Subcutaneous enoxaparin (40 mg) (7–10) days
Tan et al., 2018[19]	2007–2016	Retrospective	5951	18 (0.30)	Prophylactic doses of LMWH or unfractionated heparin with mechanical thromboprophylaxis
Lalezari et al., 2018[28]	2011–2015	Retrospective	821	1 (0.12)	Only SCD
Bassiouny and Chalabi, 2020[29]	2014–2018	Retrospective	84	10 (11.1)	N/A
Dakour Aridi et al., 2018[30]	2008–2013	Retrospective	400	4 (1.0)	N/A
Parikh et al., 2017[31]	2006–2016	Retrospective	9749	40 (0.41)	Heparin 5000–7500 units subcutaneous or enoxaparin perioperatively
Casillas et al., 2017[32]	2010–2015	Retrospective	252	1 (0.40)	N/A
AlSabah et al., 2017[9]	2011–2016	Retrospective	2280	9 (0.39)	BMI of <50, they start with 40 mg enoxaparin twice daily for 2 weeks, which is then reduced to 40 mg once daily for another week. If the patient’s BMI is >50, they are started on 60 mg enoxaparin twice daily for 2 weeks, followed by 60 mg once daily for another week
Al Jarallah et al., 2017[33]	2009–2015	Retrospective	808	2 (0.25)	<50 kg/m2 BMI patients: 3500 IU tinzaparin 2 h before surgery and then daily 3500 IU for 10 days >50 kg/m2 BMI patients: 40 mg enoxaparin 2 h before surgery and then 40 mg daily for 2 weeks
Villagrán et al., 2016[20]	2009–2015	Retrospective	1236	5 (0.40)	Enoxaparin 40 mg per day was initiated intraoperatively and continued for 7 days/modified in 2015 to 14 days
Burchett et al., 2015[34]	2007–2012	Retrospective	182	1 (0.55)	N/A
Alsina et al., 2014[35]	2007–2013	Prospective	100	1 (1.0)	Enoxaparin (0.5 mg/kg/day) was prescribed for 12 h preoperatively and for maintenance administration of 30 days postoperatively
Goitein et al., 2013[5]	2007–2012	Retrospective	4355	16 (0.36)	Subcutaneous enoxaparin sodium was administered once daily, but it varied slightly with regard to length of treatment (2 centers continued enoxaparin after the patients were discharged)
Belnap et al., 2016[36]	2012–2015	Retrospective	646	4 (0.62)	Perioperative heparin 5000 units subcutaneously q 8 h and SCD for 2 days

N/A=Not available, ERAS=Enhanced recovery after surgery, DVT=Deep venous thrombosis, LMWH=Low–molecular-weight heparin, SCD=Sequential compression device, BMI=Body mass index

Prophylactic anticoagulation for PMVT after bariatric surgery is a controversial issue. The practice varied between using multiple doses over many days or weeks perioperatively[18,38,39] to no prophylactic doses at all.[5] No consensus regarding the appropriate dose, timing, and duration of thromboprophylaxis in patients with obesity because of the limited and inconsistent data from level I evidence randomized controlled trials of LMWH use in bariatric surgery.[3] Scholten et al.[12] compared two consecutive patient groups who were administered different enoxaparin dose: 30 mg subcutaneously twice daily in 92 patients and 40 mg subcutaneous (SC) twice daily in 389 patients. The higher dose group had a lower VTE incidence (0.6% vs. 5.4%, P = 0.01) with no significant difference in bleeding.

Hamad and Choban[40] compared postoperative VTE rates in different LMWH prophylaxis regimens used in five bariatric centers. One center used 40 mg enoxaparin doses with either once or twice daily dose. The centers were heterogeneous regarding VTE risk factors and operative conditions, prohibiting valid comparative analysis. Three regimens of adjusted-dose enoxaparin were studied by Chlysta et al.[41] and compared with respect to bleeding, thromboembolism, and mortality. There was no significant VTE or bleeding rate difference among groups. One PE occurred in the fixed preoperative enoxaparin dosing group.[41] Raftopoulos et al.[42] compared enoxaparin 30 mg twice daily until hospital discharge only in 132 patients and the addition of enoxaparin 40 mg daily for 10 days postdischarge in 176 patients. Of the group who received inhospital thromboprophylaxis only, 4.5% had a VTE incident within 30 days postoperatively compared with none in the postdischarge group (P = 0.006). Becattini et al.’s review of 19 studies found little benefit and increased bleeding from body weight adjusted LMWH dose.[43] Other studies concluded that pharmacologic thromboprophylaxis is not mandatory when nonpharmacologic prophylaxes such as early ambulation, use of sequential compression devices, and adequate hydration are undertaken in patients who have no previous history of VTE or at high risk for any other causes.[44,45] Hussain et al.[46] found no significant difference between patients who received or did not receive thromboprophylaxis (0.7% and 1.5%, respectively; P = 0.58), and their overall VTE incidence was 0.9%.

There are significant limitations to all published literature including small number of participants that is not enough to confirm clinically and statistically significant result differences between regimen groups. Randomization of subjects into different treatment groups is not found in all papers, and bias is introduced into any findings due to clearly noticeable differences between patient groups. Finally, although risk factors specific to LSG procedure itself were noticed by many studies, their effect on ideal VTE and PMVT prophylaxis has not been evaluated prospectively.

Optimal enoxaparin thromboprophylaxis in bariatric surgery has not yet been determined by clinical trials.[47] However, the American Association of Clinical Endocrinologists, the Obesity Society, and the American Society for Metabolic and Bariatric Surgery (ASMBS) together guidelines for bariatric surgery stated that anticoagulant doses are not specified, but posthospital discharge pharmacologic thromboprophylaxis is recommended for high-risk groups, such as those with history of DVT, taking in consideration the risk of bleeding. The ASMBS also recommends prolonged duration of chemoprophylaxis but with no specific dose or duration recommendations.[48]

CONCLUSIONS

PMVT is an uncommon but serious complication of bariatric surgery, especially after LSG. The risk factors for PMVT seem to be related patient thrombophilia and the procedure itself, namely the use of cautery devices in the vicinity of the portal and mesenteric veins. At present, there is no consensus as to which thromboprophylaxis regimen is best for bariatric surgery patient perioperatively. The best regimen would be a balance between preventing perioperative VTE and PMVT without increasing the risk of major bleeding. The length of postdischarge prophylaxis is debatable and further research is needed order to establish the optimal duration. The use of enoxaparin in a single 40 mg SC dose is as effective as prolonged and frequent doses, especially in low-risk patient groups. An individualized assessment of bleeding should be considered before deciding on a specific enoxaparin dose.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.