Obesity Surgery

, Volume 18, Issue 7, pp 791–796

Post Discharge Prophylactic Anticoagulation in Gastric Bypass Patient—How Safe?


    • Department of SurgeryHospital of Saint Raphael
    • Department of SurgeryVCUMC, MIS Center
  • Bolanle Asiyanbola
    • Department of SurgeryHospital of Saint Raphael
  • Elmer Valin
    • Department of SurgeryHospital of Saint Raphael
  • Randolph Reinhold
    • Department of SurgeryHospital of Saint Raphael
Research Article

DOI: 10.1007/s11695-007-9382-x

Cite this article as:
Ojo, P., Asiyanbola, B., Valin, E. et al. OBES SURG (2008) 18: 791. doi:10.1007/s11695-007-9382-x



It is becoming an increasingly common practice to discharge gastric bypass (GBP) patients on prophylactic anticoagulation. This is because pulmonary embolism (PE) is a common cause of mortality postoperatively. This study was undertaken to: (1) determine the incidence of major bleeding in GBP patients discharged on prophylactic low molecular weight heparin (LMWH)—enoxaparin and, (2) correlate the bleeding risk to the dose used.


Retrospective chart review of all open GBP operation from June 2004 to August 2005. One hundred and twenty seven patients were sent home on LMWH for 2 weeks. Indications: Body mass index (BMI) ≥50 kg/m2 with chronic venous stasis and/or obstructive sleep apnea, previous history of PE or deep vein thrombosis (DVT) or BMI ≥60 kg/m2. The study group was divided into two subgroups: 40 mg twice daily (bid) and 60 mg bid LMWH. Statistical analysis was done with the chi-square. The primary outcome measure was major bleeding; defined as bleeding during the period of LMWH use associated with symptomatic decrease in hematocrit (HCT), necessitating stopping LMWH administration before the end of the study period (2 weeks), bleeding-related readmission, blood transfusion, or intervention. Excluded were patients on warfarin or treated with therapeutic LMWH.


The groups were similarly matched for age, body mass index, and risk factors. No episode of major bleeding after discharge occurred in either group.


The use of low molecular weight heparin for prophylactic anticoagulation after open gastric bypass is not associated with risk of major bleeding.


Morbid obesitySuper obesityBariatric surgeryGastric bypassDeep vein thrombosisPulmonary embolismMajor bleedingEnoxaparinLow molecular weight heparin


Gastric bypass is one of the surgical procedures for managing morbid obesity. Excluding anastomotic leaks, postoperative pulmonary embolism (PE) is the leading cause of death in bariatric surgical patients [1]. Although the incidence of pulmonary embolism is quoted as between 1 to 3%, autopsy findings after perioperative death for morbid obesity found pathological evidence of pulmonary emboli in 80% of the patients studied whereas it was clinically suspected in only 20% [2, 3]. Most pulmonary embolism occurs in the first 30 days after gastric bypass [4].

Gastric bypass patients have a higher risk of pulmonary embolism if they have a body mass index (BMI) ≥60 kg/m2, history of pulmonary embolism (PE), or deep vein thrombosis (DVT), venous stasis disease, truncal obesity, obstructive sleep apnea, or have limited ambulatory capabilities [1, 5].

To date no standard for perioperative DVT/PE prophylaxis in the morbidly obese population exists in the literature [6]. Thirty-eight (38%) of members of the American Society for Bariatric Surgery surveyed recently used a combination of two or more prophylactic methods [6, 7]. The common modalities in use are preoperative low molecular weight (LMWH) or unfractionated heparin (UFH), preoperatively deployed inferior vena cava (IVC) filters, intraoperative and postoperative compression devices, postoperative early ambulation, and postoperative subcutaneous LMWH [6, 7].

It is becoming increasingly common to discharge gastric bypass patients on LMWH. Although Scholten et al. [8] had suggested that 40 mg twice daily enoxaparin administration may be more effective in the obese patients rather than the 30 mg used in the nonobese, neither its effectiveness nor safety had been shown [4]. Nonobese patients discharged on LMWH have 3% incidence of postoperative bleeding, but the incidence of major bleeding in post gastric bypass patients discharged on LMWH has not been studied.

The purpose of this study was to determine the incidence of major bleeding in high-risk superobese patients discharged on prophylactic LMWH after gastric bypass and to analyze any correlation to dosage.

Materials and Methods

A retrospective chart review was performed of all open gastric bypasses from June 2004 to August 2005. Charts from medical records during inpatient admission and physician office were reviewed. Of the 338 patients that underwent open gastric bypass during the 15 months study period, 127 were discharged on prophylactic anticoagulation with enoxaparin, a low molecular weight heparin (LMWH) for 2 weeks. According to departmental protocol, 40 mg or 60 mg twice daily (bid) enoxaparin was started postoperatively and continued until 2 weeks after discharge. The dosage of the LMWH used was at the discretion of the operating surgeon. Statistical analysis was done using the Chi-square test.

Only patients at a high risk for deep venous thrombosis or pulmonary embolisms were treated. The inclusion criteria were: previous history of pulmonary embolism or deep vein thrombosis; BMI ≥60 kg/m2; or BMI ≥50 kg/m2 with any of these three risk factors: venous stasis disease; obstructive sleep apnea, or severe ambulation limitation. Excluded were patients with previous history of bleeding, and those discharged on therapeutic LMWH dosages or warfarin. Inferior vena cava (IVC) filters were also inserted in patients with prior deep vein thrombosis or pulmonary embolism.

The primary outcome measure was major bleeding. This was defined as bleeding occurring during the period of the LMWH use associated with symptomatic decrease in hematocrit necessitating stopping of the LMWH administration before the end of the study period, bleeding-related readmission, blood transfusion, or intervention to stop the bleeding.


One hundred and twenty-seven (127) open gastric bypass patient constituted the study group. The male/female ratio was similar. Their BMI ranged from 44 to 84 as shown in Fig. 1. Most were however between a BMI of 50 and 60 kg/m2. Patients with BMI <50 kg/m2 were discharged on LMWH because they had previous history of pulmonary embolism or deep vein thrombosis. The most common reason for discharging patients on enoxaparin was BMI ≥50 kg/m2 with any of these three risk factors: venous stasis disease, obstructive sleep apnea, or severe limitation of ambulation (Fig. 2). Patients with prior DVT and PE only were a small group of 3% and 1% of the study group, respectively (Fig. 2). This subset with previous DVT and/or PE also had an IVC filter placed as the effectiveness of LMWH in preventing DVT and/or PE is still controversial.
Fig. 1

The graph shows the BMI ranges. Most had a BMI between 50 and 60 kg/m2

Fig. 2

The pie chart depicts the reasons for discharging patients on enoxaparin. The most common reason was a BMI ≥50 kg/m2 with any of these three risk factors: venous stasis disease, obstructive sleep apnea, or severe limitation of ambulation

The 127 patients were divided into two subgroups based on the dose of subcutaneous LMWH given: 40 mg bid enoxaparin in 59 patients and 60 mg bid enoxaparin in 68 patients.

Table 1 shows the subgroups division by enoxaparin dosage and BMI greater or less than 60 kg/m2. The anticoagulation dosages were similar irrespective of BMI (Tables 1 and 2) and indications for prophylactic anticoagulation postdischarge (Fig. 3).
Table 1

Enoxaparin dosing By BMI


BMI <60

BMI ≥60


P value

40 mg Enoxaparin




0.714 (NS)

60 mg Enoxaparin








Table 2

Characteristics of subgroup



P value

40 mg

60 mg

Age Ranges




0.979 (NS)













Mean (SD)

48 (11.1)

46 (10.4)





0.001 (NS)








0.714 (NS)




Mean (SD)



HCT Changes




0.511 (NS)




Major Bleeding



Fig. 3

The graph reveals that the anticoagulation dosages were similar irrespective of the indications for prophylactic anticoagulation post discharge

More males had 60 mg bid enoxaparin for prophylactic anticoagulation because a prior study at our institution had shown a higher risk of pulmonary embolism in males with BMI ≥60 kg/m2 (Fig. 1).

None of these gastric bypass patients with a high risk for pulmonary embolism that had to be discharged on prophylactic anticoagulation demonstrated any major bleeding (Fig. 4).
Fig. 4

A graph showing the changes in hematocrit level with enoxaparin usage. There was no significant reduction in the hematocrit, and none of the patients presented with complaints, symptoms, or signs of bleeding


Morbid obesity (especially when BMI ≥60 kg/m2) and gastric bypass are high-risk factors for thromboembolic disease. Twenty to fifty percent of the mortality after GBP is caused by pulmonary embolism [1, 9]. Increasing numbers of GBP surgeries are done and more surgeons are discharging patients on anticoagulation, the effectiveness of which is still unconvincing.

Low molecular weight heparin (LMWH) has become the standard for thrombophylaxis in most European countries and Canada and, since their introduction to the USA, have rapidly gained acceptance by American physicians [10]. Wu et al. [7] actually revealed LMWH as the second most frequent anticoagulant used in GBP in the USA, second only to UFH. LMWH has been shown to offer the same protection against DVT as UFH in postoperative patients with a similar or possibly lower risk of bleeding [11, 12].

It is still unclear whether the prophylactic effect of LMWH in morbid obese individuals is weight related [1316]. A minimum of 40 mg bid is recommended in the obese population compared to 30-mg bid dosage for prophylaxis in the nonobese.

In a nonrandomized retrospective study, Scholten et al. [8] reported that enoxaparin 40 mg every 12 h (q12h) lowered the incidence of DVT to 0.6% compared with 5.4% for enoxaparin 30 mg q12h [8]. There remains a lack of a consensus regarding the appropriate dose, timing, and duration of thrombophylaxis in obese patients because of the paucity of data from randomized controlled trials evaluating LMWH dosing in severely obese surgical patients [7].

Kalfarentzos et al. [17] randomly assigned 60 gastric bypass patients to two different doses of nadroparin, a LMWH, and noted that the lower dose of 5700 IU was as effective as the 9500 IU dose, but was associated with fewer bleeding complications.

Two studies have questioned the use of a fixed dose as opposed to a weight-based (mg/kg) dose of enoxaparin in obese patients [18]. Using volunteers, Frederick et al. [13] demonstrated a strong negative correlation between body weight and the anticoagulant effect of a fixed dose of enoxaparin (40 mg), measured using anti-factor Xa activity. Similarly, Hamad et al. [19] showed that enoxaparin at a fixed dose of 30 mg q12h achieved subtherapeutic anti-Xa activity in bariatric surgical patients.

We used a dose of 40 mg or 60 mg bid in our study to decrease the risk of PE in the very high risk GBP without any risk of major bleeding. Sixty milligrams bid enoxaparin was used more in the male population with BMI ≥60 kg/m2 because they are at a higher risk of DVT/PE [1, 5].

Enoxaparin is a LMWH given as a subcutaneous injection. It has the advantages of convenience (fixed dosing without need for monitoring or intravenous infusion), a modest excess of bleeding, ease of administration and it immediately establishes adequate levels of anticoagulation. LMWH are better absorbed, have a more predictable dose response and a longer half-life than UFH [20].

This study was undertaken because a 5-year review of pulmonary embolism in our open gastric bypass patients population had revealed an incidence of 0.9% (11 of 1225 patients) with one death (lethal PE incidence of 0.08%). Symptoms and diagnosis were within the first 2 weeks postoperatively. PE in that review was found to be more common in morbidly obese individuals with BMI >50 with chronic venous stasis diseases and men with BMI ≥60 kg/m2 without any other risk factors [21]. The enoxaparin dosage of 60 mg bid was used for these higher risk patients and individuals with previous history of DVT and PE because of their increased risk.

In our study population of 127 high-risk GBP patients discharged on 2 weeks prophylactic LMWH anticoagulation, there was no incidence of symptomatic decrease in hematocrit, bleeding-related readmission, blood transfusion, or intervention. There was, however, a death after the study period, but within a month of surgery in a male with BMI of 60 kg/m2 despite 60 mg enoxaparin bid and IVC filter placement from PE. The autopsy on this patient revealed very extensive thrombosis into and beyond the IVC filter. The effectiveness of postdischarge anticoagulation is an unanswered question, as the incidence of death from PE before and after institution of postdischarge anticoagulation in our institution was the same.

A study by Miller et al. [6] using subcutaneous heparin 8 hourly showed a postoperative hemorrhage in 2.4% of the 255 patients studied. Hamad et al. [18] had published a major bleeding risk of 0.6% in her multicenter study of 668 patients on enoxaparin. This bleeding, however, occurred during their hospital admissions.

The risk of heparin-induced thrombocytopenia has been shown to be lower with LMWH than with UFH [22]. A potential disadvantage of using LMWH is the need for repeated subcutaneous injections leading to increased patient discomfort and injection hematoma. However, self-injection has been shown to be generally well accepted and safe in previous studies [23].

The limitations of this study are its lack of a control group and its retrospective nature, which does not allow us to adequately look at the prevalence of minor bleeding in these high-risk patients. None of our studied patients developed symptomatic decrease in hematocrit, nor required the stopping of enoxaparin administration before the end of the study period (2 weeks), bleeding-related readmission, blood transfusion, or intervention to stop the bleeding.


Gastric bypass patients at very high risk for thromboembolism because of a high BMI and previous history of DVT/PE can be safely discharged on enoxaparin for anticoagulation with minimal or no increased risk of bleeding. Either the 40 mg bid or the 60 mg bid regimen is safe from the standpoint of postoperative bleeding in the setting of extended prophylaxis. The clinical efficacy of enoxaparin, 40 vs 60 mg bid, in preventing DVT or PE would, however, require further studies.

Copyright information

© Springer Science + Business Media B.V. 2007