Surgical Endoscopy

, Volume 28, Issue 2, pp 543–551

Clinical markers of the hypercoagulable state by rotational thrombelastometry in obese patients submitted to bariatric surgery

Authors

    • Department of Anesthesiology, Institute of Digestive and Metabolic Diseases, Hospital ClinicUniversity of Barcelona
  • Eva Rivas
    • Department of Anesthesiology, Institute of Digestive and Metabolic Diseases, Hospital ClinicUniversity of Barcelona
  • Graciela Martinez-Palli
    • Department of Anesthesiology, Institute of Digestive and Metabolic Diseases, Hospital ClinicUniversity of Barcelona
  • Annabel Blasi
    • Department of Anesthesiology, Institute of Digestive and Metabolic Diseases, Hospital ClinicUniversity of Barcelona
  • Juan Carlos Holguera
    • Department of Anesthesiology, Institute of Digestive and Metabolic Diseases, Hospital ClinicUniversity of Barcelona
  • Jaume Balust
    • Department of Anesthesiology, Institute of Digestive and Metabolic Diseases, Hospital ClinicUniversity of Barcelona
  • Salvadora Delgado
    • Department of Gastrointestinal Surgery, Institute of Digestive and Metabolic Diseases, Hospital ClinicUniversity of Barcelona
  • Antonio M. Lacy
    • Department of Gastrointestinal Surgery, Institute of Digestive and Metabolic Diseases, Hospital ClinicUniversity of Barcelona
Article

DOI: 10.1007/s00464-013-3203-1

Cite this article as:
Taura, P., Rivas, E., Martinez-Palli, G. et al. Surg Endosc (2014) 28: 543. doi:10.1007/s00464-013-3203-1

Abstract

Background

Metabolic and inflammatory disturbances associated with obesity are considered important trigger factors for venous thromboembolism (VTE). Identification of clinical markers indicating a hypercoagulability state could define a group of high-risk patients in the surgical setting. This study aimed to identify these markers using rotational thrombelastometry (ROTEM) analysis, an established method for hemostasis testing that can detect hyperfunctional stages of the clotting process.

Methods

From June to December 2010, this study investigated 109 consecutive obese patients (28 women and 22 men, mean age 46 years, body mass index 46.6 ± 7 kg/m2) with no history of VTE who were candidates for bariatric surgery. Preoperative clinical and metabolic characteristics and ROTEM analysis were recorded. Hypercoagulable risk was defined when patients showed a clot strength (G) of ≥11 dynes/cm2.

Results

Of the 109 patients, 20 (18 %) were hypercoagulable according to ROTEM analysis. Metabolic/inflammatory biomarkers such as leptin, C-reactive protein, fibrinogen levels, and platelet count were significantly higher in the high-risk patients. In the multivariate analysis, fibrinogen was an independent predictor of G ≥ 11 dynes/cm2 [odds ratio (OR) 2.92, 95 % confidence interval (CI) 1.80–5.21, p = 0.023]. After adjustment to other data, only waist circumference affected the prediction [OR 4.42, 95 % CI 2.27–6.71, p = 0.009]. Receiver operating characteristic curve analysis showed that 3.95 g/l was the best cutoff point for fibrinogen predictability (sensitivity 100 %, specificity 41 %).

Conclusion

A hypercoagulability state in obese patients is associated with central obesity and high fibrinogen levels, which should be considered clinical hallmarks of this state. More aggressive perioperative prophylaxis for VTE should be recommended when these hallmarks are present in obese patients.

Keywords

Metabolic syndromeObesityThrombelastometry analysisThrombophilia

Studies have shown obesity to be a major risk factor for cardiovascular disease and venous thromboembolism (VTE) among medical and surgical patients. Dysfunctional adipose tissue secretes adipokines and proinflammatory cytokines that may play a role in the production of clotting factors, contributing to an enhanced prothrombotic state [16]. These disturbances in hemostatic and fibrinolytic systems are mainly mediated by increased production of clotting factors (factor VIII and von Willebrand factor), plasminogen activator inhibitor-1 (PAI-1), and platelet activation.

Besides the aforementioned hemostasis abnormalities, the individual features of metabolic syndrome (MetS) are considered important triggering factors for VTE. Recent evidence suggests a positive association between central obesity and thrombin generation, which is the central event for the quality and stability of the formed clot that causes thrombosis [7, 8].

The prevalence of obesity continues to rise, and bariatric surgery is increasingly used as the gold standard of treatment. Although mortality after bariatric surgery is an uncommon event, morbidity during the perioperative period is commonly related to thromboembolic events ranging from simple deep venous thrombosis to pulmonary embolism [9, 10]. Curiously, large series of prospective studies investigating patients undergoing open or laparoscopic bariatric surgery have shown that the risk for postoperative VTE seems to be very low (0.05–1 %), with a major bleeding rate of 0.7–5.9 % [1115]. This finding reflects the pressing need to detect high-risk thrombotic patients and balance VTE prevention with the increased risk of bleeding. Specific identification of a high-risk group of obese patients with no history of VTE and a hypercoagulable state would facilitate appropriate thromboprophylaxis to avoid the potential adverse effects of overtreatment.

Routine plasma-based laboratory assays, prothrombin time, and activated partial thromboplastin time are unable to detect a hypercoagulable state because they can be monitored only until fibrin formation and do not entirely reflect overall hemostatic balance [16]. Dynamic clot formation evaluated by viscoelastic monitoring may overcome some limitations of routine laboratory coagulation tests and is increasingly used to detect coagulopathy and hypercoagulable states [17, 18].

More interestingly, findings have shown a hypercoagulability state identified by rotational thromboelastometry to be predictive of thromboembolic events in surgical patients [18]. Whole-blood rotational thrombelastometry (ROTEM; Pentapharm GmbH, Munich, Germany) provides a dynamic assessment of the clotting process, reflecting the interactions between the platelets, the coagulation and fibrinolytic factors, the activators, and the inhibitors that may affect the clotting process [19]. In addition, ROTEM provides further information about total clot strength (G) (inclusive of both platelet and enzymatic contribution) and thrombus generation velocity curves described by Sorensen et al. [20] that closely correlated with plasma thrombogram assays, the gold standard test of thrombin generation [21, 22]. Hence, ROTEM waveform information can be a reasonable surrogate marker of thrombin generation that is more sensitive for testing thrombophilia than the commonly used tests.

This prospective study aimed to use ROTEM analysis to identify the incidence of hypercoagulability and the clinical and biochemical markers predicting hypercoagulability risk in obese patients who are candidates for bariatric surgery. An additional aim was to determine whether the standard thromboprophylaxis could modify the thrombophilia risk step evaluated by ROTEM.

Patients and methods

After institutional review approval and signed informed consent, from June to December 2010, all consecutive obese patients undergoing laparoscopic bariatric surgery (Roux-en-Y gastric bypass or sleeve gastrectomy) were enrolled in a prospective study. We excluded patients who had any known risk factor for VTE other than age or morbid obesity and those receiving oral contraceptives, pharmacologic VTE prophylaxis, and oral antiplatelet therapy, as well as patients who had renal or hepatic disease. Patients with an earlier history of VTE, venous edema with ulcerations, or a known hypercoagulable state who were managed with temporary inferior vena cava filters and anticoagulants also were excluded from the study.

The study enrolled 109 patients. All the patients underwent preoperative evaluation the week before surgery, and the following clinical data were recorded: (1) anthropometric data including weight, height, body mass index (BMI), waist circumference, and the waist-to-hip ratio (WTH); and (2) a medical history including arterial hypertension, smoker activity, and obstructive sleep apnea syndrome (OSAS) confirmed by polysomnography and treated by continuous positive airway pressure (CPAP).

The clinical diagnosis of MetS was performed according to National Cholesterol Education Program adenosine triphosphate (ATP 3 criteria) [23]. Laboratory tests were designed to determine (1) a proinflammatory state (inflammatory adipokines [interleukin-6 (IL6), tumer necrosis factor-alpha (TNFα), and leptin], C-reactive protein [CRP] and fibrinogen); (2) dyslipidemia (triglyceride, cholesterol [high-density lipoprotein and low-density lipoprotein]); (3) insulin resistance (fasting insulin levels > 20 mU/L and triglyceride > 150 mg/dl); and (4) clotting factors (fibrinogen, platelets, activated partial thromboplastin time, and international normalized ratio). Surgical time, immediate postoperative complications, and length of hospital stay also were recorded.

Rotational thrombelastometry

The basis of the thrombelastography technique has been previously described [19]. It measures the viscoelastic properties of clotting blood, graphically representing aspects of clot formation and lysis.

For ROTEM analysis, fresh venous blood was collected into citrated vacutainer tubes (3.2 % sodium citrate), and reagents were used to the manufacturer’s recommendation. All ROTEM assays were performed within 10 min after extraction.

An extrinsically activated assay with recombinant tissue factor (EXTEM) was performed, and the following standard variables of interest were determined at 37 °C: (1) clotting time (s), rate of initial fibrin polymerization; (2) α-angle (α in degree), rate of clot growth, reflecting the rate of thrombin generation; and (3) maximum clot firmness (MCF in mm), reflecting the end result of maximal platelet–fibrin interaction via the GP2b-3a receptors. Additionally, computer-generated parametric measures of total clot strength derived from MCF (5,000×MCF/100-MCF G in dynes/cm2) and area under the curve (AUC) or total thrombus generation (dynes/cm2) obtained from the first derivative waveform [20] also were recorded (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00464-013-3203-1/MediaObjects/464_2013_3203_Fig1_HTML.gif
Fig. 1

Thrombelastometry tracing. Clotting time (CT), the time taken for a clot to begin forming, is established when the trace amplitude reaches 2 mm. Maximum clot firmness (MCF) or clot shear elasticity reflect the contribution of fibrin to clot strength. Angle α (α degree) is the rate of clot growth reflecting the rate of thrombin generation. Area under the curve (AUC mm) or total thrombus generation is obtained from the first derivative waveform

Definition of hypercoagulability risk

Although a decrease in the time to clot initiation and an increase in the speed of clot propagation (α) and in clot strength (MCF, G, AUC) indicate an enhancement of hemostasis, G and AUC provide a more realistic representation of overall clot strength. Currently, only G has been validated as a marker of risk for VTE events. On the basis of previous data [18, 24], a patient in this study was declared to be at high hypercoagulable risk when G ≥ 11 dynes/cm2.

Study design

None of the patients received low-molecular-weight-heparin (LMWH) before surgery. Before induction of anesthesia, a venous blood sample was withdrawn, and thrombelastometry analysis was performed. All patients were treated with a multimodality VTE prophylaxis protocol including graduated compression stockings, intermittent sequential compression devices placed before the procedure began and kept in place until the surgical procedure was completed or the patient was ambulatory, and LMWH (enoxaparin 40 mg/s) once a day, starting the next morning after surgery. Also, early ambulation was enforced during the first 24 h after surgery.

Patients considered to be at high risk for thrombosis (G ≥ 11) were managed with extended postdischarge thromboprophylaxis enoxaparin 40 mg once a day up to 3 weeks after surgery. To assess the impact of standard VTE prophylaxis on hypercoagulable risk, the ROTEM parameters of 58 patients were determined daily ~4 to 6 h after the dose of enoxaparin was administered.

Statistical analysis

All data are presented as mean ± standard deviation. Categorical data are expressed as numbers and percentages. Quantitative variables were compared by paired t tests and qualitative variables by either Chi square or Fisher’s exact test as appropriate. For further comparisons between the two groups, the Mann–Whitney U test was used. The Spearman rank correlation was calculated for correlation between the ROTEM parameters and the demographic and biologic patient data.

The multivariate logistic regression model (backward logistic regression) using preoperative variables found to be significant in bivariate analysis or variables clinically relevant was considered for inclusion in the multivariate model to identify the variables capable of predicting G ≥ 11 dynes/cm2.

A receiver operating characteristic (ROC) curve was constructed to determine the highest accuracy of all variables predicting G ≥ 11 dynes/cm2. A maximum value of the Youden index was calculated and used as the best cutoff point to predict G ≥ 11 dynes/cm2 (area under the curve 0.5). All analyses were performed using the Statistical Package for the Social Sciences (SPSS; SPSS, Chicago, IL, USA), version 14.0. Statistical significance was set at a p value lower than 0.05.

Results

Clinical, demographic, and laboratory testing data of all the study participants are shown in Table 1. In the study population, the prevalence of MetS according to ATP 3 criteria was 39.4 %. For 74 patients, OSAS was diagnosed (67.8 %) and found to be severe in 52 patients (47.7 %), who were treated by CPAP. Weak smoker activity was present in 33 % of the patients.
Table 1

Patient characteristics of all obese patients stratified according to clot strength

 

All patients (n = 109)

G < 11 (n = 89)

G ≥ 11 (n = 20)

p value

Sex/female n (%)

82 (75)

70 (79)

17 (85)

0.347

Age, years (range)

46 (21–70)

47 (22–67)

41 (21–70)

0.059

BMI, kg/m2 (range)

46.6 (39–66)

46.5 (39–66)

46.4 (40–59)

0.742

Waist circumference, cm (range)

88.4 (79–114)

86.2 ± 14.3

105.6 ± 9.4

0.069

WTH ratio, n (%)

0.95 (0.72–1.12)

0.94 ± 0.8

1.01 ± 0.8

0.062

Smokers, n (%)

36 (33)

29 (33)

7 (35)

0.627

OSAS (CPAP), n (%)

52 (48)

41 (46)

11 (55)

0.132

MetS, n (%)

43 (39)

34 (38)

9 (45)

0.381

Leptin, ng/ml (range)

59.4 (28.3–113.7)

55.4 (28.3–104)

89.2 (42.4–113.7)

0.006

Glucose (mmol/l)

114.8 ± 49

117.7 ± 46

121.5 ± 64

0.759

Insulin (μg/ml)

25.5 ± 17

22.1 ± 18

23.4 ± 15

0.323

LDL cholesterol (mg/dl)

113 ± 30

114 ± 31

121 ± 16

0.619

HDL cholesterol (mg/dl)

43 ± 11

44 ± 10

38 ± 10

0.125

Triglycerides (mmol/l)

138.0 ± 72

134.2 ± 67

177.7 ± 147

0.672

TNFα, pg/ml (range)

7.5 (2–58)

7.2 (6–39)

9.5 (2–58)

0.720

CRP, mg/l (range)

1.1 (0.3–4.7)

1.1 (0.3–2.8)

2.4 (0.8–4.7)

0.009

Fibrinogen (g/l)

4.2 ± 0.9

4.1 ± 0.4

5.4 ± 0.8

0.000

Platelet count

287 ± 64

277 ± 66

333 ± 75

0.014

aPTT

31.2 ± 7

32.6 ± 6

30.4 ± 7

0.464

INR (%)

1.03 ± 0.04

1.98 ± 0.04

1.01 ± 0.08

0.783

Data are presented as n (range), median (%), or mean ± standard deviation. Statistical results between patients with G < 11 and patients with G ≥ 11 dynes/cm2 are shown

G clot strength (dynes/cm2), BMI body mass index, WTH waist-to-hip ratio, OSAS obstructive sleep apnea syndrome, CPAP continuous positive airway pressure, MetS metabolic syndrome, LDL low-density lipoprotein, HDL high-density lipoprotein, TNFα tumor necrosis factor-alpha, CRP C-reactive protein, aPTT activated partial thromboplastin time INR international normalized ratio

Considering all the patients, we found no significant correlations between clot strength and the presence of MetS or any of its components. Moreover, classical cardiovascular risk factors such as smoking, OSAS, and dyslipidemia had no impact on clot kinetics. However, the metabolic/inflammatory biomarkers (leptin, CRP, fibrinogen, and platelets) correlated significantly with clot strength and total clot formation (Fig. 2). This cohort of obese patients showed a strong correlation (r = 0.968, p < 0.000) between clot strength and total clot formation (G/AUC) (Fig. 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs00464-013-3203-1/MediaObjects/464_2013_3203_Fig2_HTML.gif
Fig. 2

Linear regression analysis of the relationship of several inflammatory biomarkers to clot strength and total thrombus generation (rate of thrombin generation). A Correlation between clot strength and fibrinogen, leptin, C-reactive protein (CRP), and platelet count. B Correlation between the area under the curve (AUC) and fibrinogen, leptin, CRP, and platelet count

https://static-content.springer.com/image/art%3A10.1007%2Fs00464-013-3203-1/MediaObjects/464_2013_3203_Fig3_HTML.gif
Fig. 3

Linear correlation analysis of the relation between clot strength (G) and the area under the curve (AUC)

A total of 20 patients (18.3 %, 17 women and 3 men) met the criteria for high hypercoagulability risk (G ≥ 11). Table 1 shows the demographic and anthropometrical data. Comparison with the remaining patients showed no significant differences. Metabolic/inflammatory biomarkers such as leptin, CRP, fibrinogen levels, and platelet counts were significantly higher among high-risk patients (Table 1).

In the univariate analysis, waist circumference, CRP, platelet count, and fibrinogen were significantly associated with clot strength (G ≥ 11 dynes/cm2) (Table 2). In the multivariate logistic regression analysis, only fibrinogen level remained as an independent predictor of G ≥ 11 dynes/cm2 [odds ratio (OR) 2.92, 95 % confidence interval (CI) 1.80–5.21, p = 0.023]. After adjustment of age, gender, BMI, waist circumference, OSAS, and smoker data, the dominant variable affecting the association between fibrinogen and G ≥ 11 was WTH (OR 3.42, 95 % CI 2.27–6.71, p = 0.009). The best cutoff point for fibrinogen to predict G ≥ 11was 3.95 g/l, with a sensitivity of 100 %, a specificity of 41 %, and an area under the ROC curve of 0.802 (CI 0.70–0.90, p = 0.000) (Fig. 4). Thrombelastometry data showed that, ruling out clotting time, all parameters analyzed were significantly higher in the high-risk patients (Table 3).
Table 2

Multivariate analysis of risk factors associated with clot strength

 

Univariate model

Multivariate model

OR (95 % CI)

p value

OR (95 % CI)

p value

BMI (kg/m2)

1.78 (0.41–1.95)

0.108

  

Waist circumference (cm)

1.07 (1.03–4.61)

0.047

  

Age (years)

2.17 (2.31–5.62)

0.071

  

CRP (mg/l)

1.89 (1.88–2.93)

0.013

  

Fibrinogen (g/l)

3.31 (1.16–5.09)

0.007

2.92 (1.80–5.21)

0.023

Platelet count

2.08 (1.02–3.04)

0.038

  

Leptine (ng/ml)

1.08 (0.78–3.81)

0.067

  

Risk factors associated with preoperative clot strength (G ≥ 11 dynes/cm2) in a univariate model. Multivariate model of risk factors that independently predict preoperative clot strength (G ≥ 11 dynes/cm2)

OR odds ratio, CI confidence interval, BMI body mass index, CRP C-reactive protein, G clot strength

https://static-content.springer.com/image/art%3A10.1007%2Fs00464-013-3203-1/MediaObjects/464_2013_3203_Fig4_HTML.gif
Fig. 4

Relative operating characteristics (ROC) curve of plasma fibrinogen levels to predict clot strength (G) ≥ 11 dynes/cm2 [area under the curve (AUC) = 0.802, 95 % confidence interval (CI) 0.70–0.90, p = 0.000]. The best cutoff point for fibrinogen to predict G ≥ 11 was 3.95 g/l (sensitivity of 100 %, specificity of 41 %)

Table 3

Basal thrombelastometry analysis of all patients stratified according to clot strength

 

All patients (n = 109)

G < 11 dynes/cm2 (n = 89)

G ≥ 11 dynes/cm2 (n = 20)

p value

EXTEM

    

CT (s)

70.4 ± 14

70.2 ± 14

64.7 ± 10

0.059

α (G°)

73.4 ± 3

72.5 ± 3

77.4 ± 2

0.003

MCF (mm)

65.4 ± 4.5

62.9 ± 3.2

72.2 ± 1.7

0.004

G (dynes/cm2) (range)

9.8 ± 1.8

8.1 ± 2.0 (5.2–10.8)

13.2 ± 0.8 (11.2–15)

0.000

AUC (dynes/cm2)

6567 ± 448

6502 ± 253

7096 ± 149

0.004

Data are presented as mean ± standard deviation. Statistical results of the Mann–Whitney U test between patients with G < 11 and patients with G ≥ 11 dynes/cm2 are presented

G clot strength, EXTEM extrinsically activated assay with recombinant tissue factor, CT clotting time, MCF maximum clot firmness, AUC area under curve

Postoperative course

All the patients were managed with the same protocol, and all of them were successfully extubated in the operating room. Laparoscopic gastric bypass was performed in 89 patients (81.6 %) and sleeve gastrectomy in the remaining 23 patients. The mean average operation time was 105 ± 12 min (range 72–135 min), and the hospital length of stay was 2.43 ± 0.6 days. There was no need for reconversion. Overall, one patient (in the low-risk group) had clinically postoperative bleeding requiring endoscopic intervention.

As shown in Table 4, in spite of prolonged clotting time, standard VTE prophylaxis did not modify clot strength at 24 and 48 h postoperatively. It is important to note that all patients showed a trend toward a progressive increase in clot strength and total clot formation (or thrombin generation).
Table 4

Evolution of thrombelastometry analysis during the postoperative period of 58 patients submitted to laparoscopic bariatric surgery with conventional pharmacologic thromboprophylaxis

  

1 h after surgery

24 h after surgery

p value

48 h after surgery

p valuea

G < 11

CT (s)

68.2 ± 14

75 ± 10

NS

87.3 ± 9

0.032

G > 11

61.8 ± 9

70.4 ± 5

NS

84.3 ± 14

0.024

G < 11

G

8.9 ± 1

9.1 ± 0.8

NS

9.6 ± 0.5

0.091

G ≥ 11

12.5 ± 1.3b

12.2 ± 0.8b

NS

13.3 ± 1.1b

0.064

G < 11

AUC

6371 ± 301

6512 ± 204

NS

6550 ± 129

0.108

G ≥ 11

7101 ± 128b

7010 ± 68b

NS

7296 ± 108b

0.068

Rotational thrombelastometry (ROTEM) data of 58 patients (G < 11, n = 49; G ≥ 11, n = 9) 1, 24, and 48 h after surgery. Data are presented as mean ± standard deviation

G clot strength, CT clotting time, NS nonsignificant difference, AUC area under curve

ap statistical results of Mann–Whitney U test between patients 1 h after surgery and 48 h after surgery

bp < 0.01 between G groups

Discussion

In this study, metabolic/inflammatory biomarkers and central obesity were found to be predictors of a hypercoagulable state evaluated by ROTEM analysis in an obese population. These results reinforce the emerging concept that obesity is a state of chronic low-grade inflammation associated with profound metabolic disturbances that may disrupt regulatory mechanisms of hemostasis.

Excess visceral adipose tissue may play the role of metabolically active secretory cells leading to an inflammatory picture analogous to low-grade sepsis. As a result of exposure to an inflammatory stimulus, circulating blood cells express tissular factor and PAI-1, fibrinogen release, and platelet activation leading to intravascular clotting, vessel occlusion, and thrombotic pathology. Fibrinogen, an acute-phase reactant like CRP, rises in response to a high cytokine state. Thus, prothrombotic and proinflammatory states may be metabolically interconnected.

Metabolic syndrome, defined on the basis of combined central obesity, impaired glucose metabolism, dyslipidemia, and arterial hypertension, seems to be a powerful and prevalent predictor of cardiovascular morbidity. Moreover, recent data support the hypothesis that obesity without MetS also develops impairments in several markers of cardiovascular disease [25, 26].

In our study cohort, MetS did not correlate with clot firmness, and only inflammatory markers showed a significant correlation with clot strength and clot generation. Central distribution of fat mass is a source of cytokines and adipokines involved in endothelial dysfunction, chronic inflammation, and oxidative stress.

In our patients, central obesity rather than BMI was associated with clot strength. This finding is in agreement with a recent study by Livingston [27], who demonstrated that fat distribution is a better predictor of VTE complications than the degree of obesity itself, whereas previous studies found that a BMI higher than 60 kg/m2 increases the relative risk of VTE threefold [28].

Among the cardiovascular risk factors, OSAS and smoking are directly correlated with hemostatic activation. Previous studies demonstrated that leptin levels are elevated in most obese patients with OSAS and that these high leptin levels promote platelet activation, thrombin generation, and high levels of PAI-1 [29].

The absence of a significant correlation between OSAS and ROTEM hypercoagulability parameters in our study likely was due to CPAP treatment during at least 3 months for all of our patients with a diagnosis of OSAS. Previous studies have suggested that CPAP treatment for more than 3 months decreases PAI-1 and platelet activation and consequently might diminish exaggerated coagulant activity [30]. Otherwise, the direct effect of cigarette smoking on coagulation activity, mediated by platelet activation and thrombin generation [31], also could be attenuated by cessation or reduction of smoking 6 months before surgery, which is part of our institutional protocol.

Thrombin, the primary driving force of the coagulation system, plays a significant role in the processes of initiation and propagation of a thrombus. The main substrate for thrombin generation is fibrinogen, which also is essential for platelet aggregation leading to fibrinolysis inhibition. Fibrinogen clearly is associated with a variety of conventional risk factors, and several polymorphisms of the fibrinogen gene may predispose to elevated fibrinogen in response to various environmental stimuli.

In a thrombophilia screening study of obese patients assessed by laboratory assays, Overby et al. [32] found that high levels of fibrinogen are the cause of thrombophilia in 45 % of patients presenting for laparoscopic bariatric surgery. In our study, the plasma fibrinogen level was significantly higher in high-risk patients and also was an independent predictor of G ≥ 11 dynes/cm2 (OR 2.92, 95 % CI 1.80–5.21, p = 0.023). A ROC curve analysis allowed us to establish a cutoff point of 3.95 mg/l for the prognosis of G ≥ 11 dynes/m2 to identify all patients considered at high risk (sensitivity 100 %). Thus, considering the strong correlation between thrombin generation (AUC) and clot strength (G) (Fig. 3), it could be concluded that in obese patients, fibrinogen level is a significant marker of thrombin generation and risk of thrombotic disorders. These results suggest that targeted VTE prophylaxis methods and extended pharmacologic prophylaxis guided by the plasma fibrinogen levels could minimize the potential risk of bleeding associated with the use of LMWH.

Thrombelastometry provides a rapid and inexpensive assessment of in vitro coagulation. It is used increasingly to screen patients being studied for a suspected prothrombotic state [17, 18, 24, 33]. Thromboelastometry is a very attractive system because it reflects hemostasis in a dynamic mode with a graphic representation of the clotting process. It has been introduced in clinical practice to monitor the LMWH antithrombotic effect [34, 35]. Although LMWH may influence all basic ROTEM parameters, clotting time is the most suitable parameter for monitoring LMWH because it has been correlated with anti-Xa plasma levels, the gold standard for assessing the antithrombotic effect of LMWH [36].

In our cohort of obese patients, clotting time increased significantly 48 h after the start of standard prophylactic treatment, reflecting the enoxaparin effect, but thrombus strength and thrombus generation did not change or show a trend to increase (Table 4). These findings sustain the evidence that a significant percentage of VTE can occur long after hospital discharge subsequent to bariatric surgery [10].

Clinical studies of patients monitored by thromboelastometry and treated with different LMWH dosing regimens initiated pre- or postoperatively are needed to assess the effect of pharmacologic prophylaxis on thrombus strength and thrombin generation. In patients with no history of thrombophilia or VTE, no consensus currently exists on a weight-based LMWH dosing regimen. Also, indication for extended-duration prophylaxis beyond the period of hospitalization remains controversial [3739].

Some limitations of this prospective study must be considered. First, we did not perform conventional thrombin generation tests to correlate with thrombus generation (AUC). Second, we did not measure anti-Xa to corroborate the effect of LMWH on the ROTEM thrombophilia parameters. Third, because the incidence of VTE was very low in this group of patients (only 1 patient), we could not correlate the risk stratification index with outcome.

In summary, our results suggest that preoperative ROTEM evaluation of obese patients presenting for laparoscopic bariatric surgery without known thrombophilia risk is useful for detecting patients at risk. On the basis of our findings, central obesity and preoperative inflammatory markers are associated with thrombophilia risk, and fibrinogen appears to be the pivotal risk factor among them. Fibrinogen associated with central obesity was an independent predictor of clot strength. Targeted prophylaxis based on the presence of these clinical markers in obese candidates for surgery may minimize the potential bleeding/thrombosis risk associated with LWMH prophylaxis.

Finally, it seems that standard thromboprophylaxis did not significantly modify the postoperative thrombophilia risk assessed by ROTEM. Large randomized clinical trials of patients monitored preoperatively with thrombelastometry and treated with different antithrombotic regimens are needed to correlate risk prediction accuracy with outcome.

Disclosures

Pilar Taura, Eva Rivas, Graciela Martinez-Palli, Annabel Blasi, Juan Carlos Holguera, Jaume Balust, Salvadora Delgado, and Antonio M. Lacy has no conflicts of interest or financial ties to disclose.

Copyright information

© Springer Science+Business Media New York 2013