Clinical implications of epidural fat in the spinal canal. A scanning electron microscopic study.

BACKGROUND AND OBJECTIVES: This review of articles summarizes recent developments in relation to fat located in the epidural space and also in dural sleeves of spinal nerve roots in order to improve our understanding of the clinical effects of the epidural blockade. METHOD: Medline search was carried cross-matching of the following words: "epidural fat", "epidural space", "adipose tissue" and "fat cells" from 1966 to 2008 in which articles referring to different pathologies that alter the epidural fat were also reviewed. Techniques used by different authors included the use of samples from dissections, cryomicrotome sections, as well as light and electron microscopy. RESULTS: Fat in the epidural space has a metameric distribution along the spinal canal that can be altered in some pathological conditions. Epidural fat is not evenly distributed. At cervical level fat is absent while in the lumbar region, fat in the anterior and posterior aspects of the epidural space forms two unconnected structures. Fat cells are found also in the thickness of dural sleeves enveloping spinal nerve roots but not in the region of the dural sac. Epidural lipomatosis is characterized by an increase in epidural fat content. When a patient has a combination of kyphosis and scoliosis of the spine, the epidural fat distributes asymmetrically. Spinal stenosis is frequently accompanied by a reduction in the amount of epidural fat around the stenotic area. CONCLUSIONS: The epidural space contains abundant epidural fat that distributes along the spinal canal in a predictable pattern. Fat cells are also abundant in the dura that forms the sleeves around spinal nerve roots but they are not embedded within the laminas that form the dura mater of the dural sac. Drugs stored in fat, inside dural sleeves, could have a greater impact on nerve roots than drugs stored in epidural fat, given that the concentration of fat is proportionally higher inside nerve root sleeves than in the epidural space, and that the distance between nerves and fat is shorter. Similarly, changes in fat content and distribution caused by different pathologies may alter the absorption and distribution of drugs injected in the epidural space.

Buprenorphine added to the local anesthetic for brachial plexus block to provide postoperative analgesia in outpatients.

BACKGROUND AND OBJECTIVES: Over the past 10 years, several studies have suggested that the addition of certain opiates to the local anesthetic used for brachial block may provide effective, long-lasting postoperative analgesia. One of these studies indicated that the agonist-antagonist, buprenorphine, added to bupivacaine provided a longer period of postoperative analgesia than the traditional opiates, but in this study, it is impossible to determine the relative contributions of the local anesthetic and the opiate to the postoperative analgesia because of the extremely long duration of the anesthesia provided by the local anesthetic, bupivacaine. By repeating the study using a local anesthetic of a shorter duration, the present study delineates more clearly the contribution of the buprenorphine to postoperative analgesia when added to a shorter-acting local anesthetic. METHODS: Forty, healthy, consenting adult patients scheduled for upper extremity surgery were enrolled in the study. Premedication was provided by intravenous midazolam 2 mg/70 kg and anesthesia by a subclavian perivascular brachial plexus block. The patients were assigned randomly to 1 of 2 equal groups based on the agents used for the blocks. The patients in group I received 40 mL of a local anesthetic alone, while those in group II received the same local anesthetic plus buprenorphine 0.3 mg. The study was kept double-blind by having 1 anesthesiologist prepare the solutions, a second anesthesiologist perform the blocks, and a third anesthesiologist monitor the anesthesia and analgesia thereafter, up to and including the time of the first request for an analgesic medication. The data were reported as means (+/- SEM), and differences between groups were determined using repeated measures of analysis of variance (ANOVA) and chi(2), followed by the Fisher exact test for post hoc comparison. A P value of less than.05 was considered to be statistically significant. RESULTS: The mean duration of postoperative pain relief following the injection of the local anesthetic alone was 5.3 (+/- 0.15) hours as compared with 17.4 (+/- 1.26) hours when buprenorphine was added, a difference that was statistically (and clinically) significant (P <.0001). CONCLUSIONS: The addition of buprenorphine to the local anesthetic used for brachial plexus block in the present study provided a 3-fold increase in the duration of postoperative analgesia, with complete analgesia persisting 30 hours beyond the duration provided by the local anesthetic alone in 75% of the patients. This practice can be of particular benefit to patients undergoing ambulatory upper extremity surgery by providing prolonged analgesia after discharge from the hospital.

1,001 subclavian perivascular brachial plexus blocks: success with a nerve stimulator.

BACKGROUND AND OBJECTIVES: Among the supraclavicular approaches to the brachial plexus, the subclavian perivascular technique is a well-established method of anesthesia of the upper extremity. Ever since Kulenkampf described his technique, eliciting a paresthesia has been almost mandatory ("no paresthesia, no anesthesia"). Lately, nerve stimulators have become more popular. However, up to the present time, clinical studies involving the nerve stimulator have failed to show success rates comparable to paresthesia techniques. METHODS: Data from 1,001 consecutive, subclavian perivascular blocks were prospectively gathered over 2.5 years. All blocks were performed according to Winnie's technique, but using a nerve stimulator instead of a paresthesia. When an adequate response was obtained, 35 to 40 mL of local anesthetic solution was injected. RESULTS: Nine hundred seventy-three blocks (97.2%) were completely successful; 16 blocks (1.6%) were incomplete and needed supplementation; and 12 blocks (1.2%) failed and required general anesthesia, giving a success rate for regional anesthesia of 98.8%. CONCLUSIONS: The subclavian perivascular block consistently provides an effective block for surgery on the upper extremity. At the site of injection with this technique, the plexus is reduced to its smallest components and the sheath is reduced to its smallest volume, which explains in great part the success obtained with this block. We believe that we have demonstrated a nerve stimulator technique that is both highly successful and safe; no clinical pneumothorax was found nor did any other major complications develop.

Intragenomic distribution and stability of transposable elements in euchromatin and heterochromatin of Drosophila melanogaster: non-LTR retrotransposon.

The intragenomic location of the elements of the I, G, jockey, F, and Doc transposon families has been studied by the Southern blot analysis, in 12 laboratory Drosophila melanogaster stocks. Elements located in euchromatin, heterochromatin, and on the Y chromosome are identified, and their stability has been assessed by comparing the autoradiographs detected in different stocks and analysis of individual flies. Evidence is shown suggesting that preferential location in euchromatin or heterochromatin and the distribution within heterochromatin are distinctive of transposon families. Elements located in heterochromatin can be unstable. These results are discussed in the context of the relationship between transposable elements and the host genome.

Management of arterial injuries produced by percutaneous femoral procedures.

BACKGROUND: A significant number of vascular injuries occur with the use of percutaneous diagnostic and therapeutic procedures. This study was done to indicate the types of these injuries and their management. METHODS: Over a 30-month period, 55 patients required operation for vascular complications after percutaneous femoral procedures including infrarenal arteriography (six patients) and angioplasty (22 patients), coronary angiography (16 patients) and angioplasty (five patients), and aortic balloon pump insertion (six patients). RESULTS: The 14 iliac and 41 femoral artery injuries included 29 pseudoaneurysms, six lacerations with persistent bleeding, seven dissections, six occlusions, three ruptures, two arteriovenous fistulas, and two large hematomas. Control for all femoral and distal external iliac artery lesions was obtained solely through a groin incision in 45 (82%) patients. Our technique for exposure of the external iliac artery through the groin is described. A separate retroperitoneal incision was necessary in 10 patients because of proximal injury, massive pseudoaneurysm, morbid obesity, or heavily scarred groins. In this series 34 lateral suture repairs, 11 interposition or bypass grafts, four patch angioplasties, one endarterectomy, three thrombectomies, and two hematoma evacuations were performed. Although no limb loss occurred, we encountered nine wound complications, five myocardial infarctions, and two deaths. CONCLUSIONS: This experience shows the wide variety of vascular complications caused by percutaneous procedures and the different techniques necessary for their management.

Comparison of heated-probe laser Doppler and transcutaneous oxygen measurements for predicting outcome of ischemic wounds.

Transcutaneous oxygen (TcPO2) measurement has proven to be an accurate means of predicting healing of ischemic wounds. This study compares the ability of TcPO2 and laser Doppler, modified by the addition of a heated probe (LDHP), to assess wound outcome. TcPO2 and LDHP measurements were made at the same site for 80 wounds, which consisted of 51 amputations (25 above knee, 6 below knee, 20 forefoot) and 29 ulcerations. Healing was defined as complete wound closure. Failure to heal was defined by the necessity for proximal amputation in 22 wounds (6 amputations, 16 ulcers). Outcome criteria were chosen to maximize accuracy and either positive or negative predictive values. Criteria with the greatest accuracy and positive predictive value for wound healing were > or = 11 mmHg for TcPO2 and > or = 50 mv for LDHP range. Criteria with the most appropriate accuracy and negative predictive value for wound failure were < 5 mmHg for TcPO2 and < 35 mv LDHP range. All wounds whose LDHP range was < 35 mv failed to heal, whereas some wounds with a TcPO2 of 0-1 mmHg healed successfully. An absolute prediction of wound healing (100% specificity and negative predictive value) was offered when either LDHP range was > or = 125 mv or TcPO2 was > or = 33 mmHg, although accuracy of either measurement at this criteria was unacceptable for more general application. We conclude that TcPO2 or LDHP will assess wound outcome with similar overall accuracy, although each test may be better for predicting a specific outcome.

Superoxide anion production by leukocytes exposed to post-ischemic skeletal muscle.

Superoxide anion (O2-) and polymorphonuclear leukocytes (PMNs) have been implicated in the genesis of skeletal muscle ischemia-reperfusion (I-R) injury, but the source of (O2-) has not been established. We studied PMNs as a potential source of O2- using a ferricytochrome reduction assay in 5 anesthetized dogs. Using a gracilis muscle model of I-R, 6 hours of ischemia was followed by 2 hours of reperfusion. The contralateral muscle served as control. Prior to ischemia and after 0.5 and 2.0 hours of reperfusion, PMNs were separated from the gracilis venous effluent of ischemic (I) and control (C) muscles. Central venous samples were also obtained prior to surgical preparation and after reperfusion. Assays for O2- were performed with and without zymosan (Z) activation. Results are expressed as nmol O2-/2 x 10(6) PMNs +/- SEM. Baseline production of O2- was 0.49 +/- 0.54 in central venous samples; Z increased the values to 6.77 +/- 2.13. After 2 hrs of reperfusion, central O2- was 1.57 +/- 0.75, which increased to 7.1 +/- 1.04 with Z. Gracilis venous samples O2- values with and without Z are reported in Table I. One way measures of analysis of variance showed no significant (p > 0.05) differences between samples. Our results demonstrate that PMNs are not the sole source of O2- in the pathophysiology of skeletal muscle I-R injury. PMN associated injury may be mediated by mechanisms other than O2- production.

A prospective evaluation of arterial intimal injuries in an experimental model.

The management of arterial intimal defects remains controversial because of uncertainty concerning their natural history. We developed an experimental canine model to prospectively evaluate posterior wall intimal flaps in the superficial femoral artery. Arterial intimal flaps were constructed in 20 anesthetized dogs (40 arteries) and evaluated by arteriography, and angioscopy, and intravascular ultrasound. Postoperative patency rates at 1 (n = 20) and 3 weeks (n = 20) were compared with a control group of ten animals (n = 20, arteriotomy without intimal flap). Acute thromboses occurred in five experimental arteries with thromboses of eight additional experimental arteries at followup. Control patency was 100%, while experimental group patencies were 75% (p less than 0.05) at 1 week and 60% (p less than 0.009) at 3 weeks. All thrombosed arteries had intimal flaps with greater than 75% luminal stenosis. We conclude that intimal injuries cause arterial thromboses acutely and during subsequent followup. Intimal flaps with stenosis greater than 75% as determined arteriographically are at greatest risk for thrombosis. Angioscopy and intravascular ultrasound characterize arterial intimal defects and may delineate injuries requiring surgical or endovascular repair.

Carotid endarterectomy for asymptomatic carotid stenosis: a ten-year experience with 120 procedures in a fellowship training program.

Performance of carotid endarterectomy for asymptomatic carotid stenosis has been restricted during recent years because of concern of reported complications in as high as 10-15% of patients, as well as limited long-term data on stroke protection. During the last 10 years, we have studied immediate and long-term results of carotid endarterectomy for asymptomatic disease in 120 patients. Operations were performed by a clinical vascular fellow with a staff surgeon in attendance in 113 (94%) cases with the remainder performed by the staff surgeon. Patients' mean age was 66 years; 82% were men. Risk factors included hypertension (56%), smoking (52%), coronary artery disease (32%), diabetes (24%), and hypercholesterolemia (6%). Arteriographic severity of stenoses was 80-99% in 74%, 60-79% in 22%, and 40-59% in 4% of cases. Postoperative complications included two transient neurological events (1.7%). No permanent strokes or deaths occurred. Using the life table method, cumulative stroke rate was 4.5% for ipsilateral events and 7.3% for contralateral events, confirming the high degree of stroke protection afforded by carotid endarterectomy in this population. Since these results were accomplished in a fellowship training program, we regard adequacy of this experience as the most influential factor in accomplishing this record. Surgeons who are unable to achieve comparable results should consider abandonment of the procedure or an extended period of additional training.

Popliteal artery agenesis: a new anatomic variant.

Agenesis of the popliteal artery has not been described as a vascular anomaly in the lower extremity. This case report describes congenital absence of the popliteal artery discovered during operative exploration after a traumatic injury to the lower extremity. The preoperative arteriogram suggested acute occlusion of the popliteal artery. Intraoperative exploration and arteriography were consistent with agenesis of the popliteal arterial segment, and postmortem examination confirmed these arteriographic and intraoperative observations. Embryologically, failure of the middle portion of the sciatic artery to persist would account for this anomaly. A review of series reporting congenital anomalies of the lower extremity vasculature failed to discover previous mention of this particular abnormality. Agenesis of the popliteal artery should be included among those vascular anomalies that may affect management of lower extremity vascular disease.

Unsuspected inflow disease in candidates for axillofemoral bypass operations: a prospective study.

Routine arteriography of the axillary, subclavian, and innominate arteries before axillofemoral bypass surgery has not been advocated because of the presumed rarity of stenosis of these inflow vessels. However, we have noted in this patient population with extensive atherosclerosis that inflow disease can cause axillofemoral graft failure despite normal preoperative clinical and noninvasive parameters. We prospectively determined the incidence of unsuspected inflow stenosis with arteriography in 40 consecutive candidates for primary (28) or secondary (12) axillofemoral bypass surgery. A new arteriographic technique with a single translumbar puncture was developed to safely and clearly visualize the potential inflow and outflow tracts. Ten of the 40 patients (25%) exhibited inflow stenosis greater than 50% of luminal diameter (unilateral in eight patients and bilateral in two patients). Seven were on the left side (five subclavian, two axillary) and five were on the right side (three subclavian, one axillary, one innominate). In eight of the 40 patients (20%) the stenotic inflow lesion was ipsilateral to the ischemic leg and therefore proximal to the preferred inflow site of an axillofemoral bypass. Upper extremity arterial pressure measurements suggested potential inflow artery stenosis in only three of the 12 (25%) instances. These findings show that equal arterial pressures in the upper extremities do not ensure adequate inflow for an axillofemoral graft. Routine arteriographic assessment of inflow intacts is recommended before axillofemoral bypass surgery.

A modified method for management of prosthetic graft infections involving an anastomosis to the common femoral artery.

In the last 10 years we have treated 28 patients with 33 groin infections involving a common femoral artery anastomosis of prosthetic arterial grafts (2 aortic Dacron grafts, 31 peripheral polytetrafluoroethylene grafts). Management included complete graft preservation for patent infected grafts (11 cases), subtotal excision of occluded infected grafts leaving an oversewn 2 to 3 mm graft remnant attached to a patent artery critical for limb survival (16 cases), and total graft excision with arterial oversewing or ligation for anastomotic bleeding (6 cases). Essential treatment adjuncts included (1) radical operative wound debridement, and (2) secondary revascularization by means of bypasses tunneled via lateral uninfected routes, and unusual approaches to uninvolved patent outflow arteries (i.e., the distal superficial or deep femoral or popliteal arteries) after isolation of the infected wound. Follow-up averaged 3 years (1 to 10 years). This plan of treatment resulted in an 11% (3/28) hospital mortality and an amputation rate of 13% (4/30 threatened limbs). Of the 25 survivors with 30 infected groin grafts, 87% (26) of the wounds healed uneventfully by secondary intention within 1 to 8 weeks (mean, 4 weeks) and have remained healed. One infected groin wound did not heal and required delayed total graft excision. Three patients had late anastomotic disruption with hemorrhage at 8 months, 2 years, and 4 years after initial treatment. This selected use of complete or partial graft preservation and other essential treatment adjuncts are proposed as a safer, easier method for managing infected prosthetic arterial grafts in the groin.

Resting gated pool ejection fraction: a poor predictor of perioperative myocardial infarction in patients undergoing vascular surgery for infrainguinal bypass grafting.

Ventricular ejection fraction is widely regarded as a prognostic indicator of perioperative myocardial infarction. To evaluate this premise the prevalence of perioperative myocardial infarction or cardiac death was analyzed in relation to preoperative resting gated pool ejection fraction in 85 patients undergoing vascular surgery for infrainguinal bypass grafting. Patients were divided into three groups on the basis of ejection fraction. Group I consisted of 50 patients with ejection fractions of 56% to 92%. Nine (18%) perioperative myocardial infarctions occurred in group I, and there were no cardiac deaths. Group II consisted of 20 patients with ejection fractions of 37% to 55%. Three (15%) myocardial infarctions occurred in this group, and there were no cardiac deaths. Group III included 15 patients with ejection fractions of 20% to 35%. Three (20%) cardiac events occurred in group III including one nonfatal myocardial infarction and two (13%) cardiac deaths. Statistical analysis showed no significant difference in prevalence of cardiac events between any group. These results suggest that resting ejection fraction is a poor predictor of perioperative myocardial infarction in patients undergoing vascular surgery. Patients with normal ejection fractions, but underlying coronary artery disease, are still at significant risk for a perioperative cardiac event.

Acute ischemia-reperfusion injury in the canine hindlimb.

A canine arterial ligation preparation was used to produce whole limb ischemia-reperfusion injury. Alterations in the distribution of arterial blood flow as well as the morphology of skeletal muscle ischemia-reperfusion have not been investigated completely in this setting. Five anesthetized adult mongrel dogs underwent multiple infrarenal aortic branch ligations; one randomly selected hindlimb was subjected to six hours of ischemia and two hours of reperfusion, while the opposite limb served as control. Distribution of arterial blood flow was analyzed by injection of radiolabeled microspheres. Electromagnetically measured femoral arterial blood flow was 92 +/- 10 ml/min during control, and increased significantly (p less than 0.05) to 254 +/- 94 ml/min during reperfusion. Flow distribution to skin, muscle, and bone was 9 +/- 2%, 68 +/- 7%, and 8 +/- 1% during control, and 7 +/- 3%, 65 +/- 8%, and 9 +/- 4% after reperfusion, which did not represent significant changes. Arteriovenous shunting was 11 +/- 4% during control, and was 13 +/- 5% during reperfusion, which was not significantly different. Subcellular injury in the ischemic and reperfused hindlimb was demonstrated by light and electron microscopy. These findings further characterize whole limb ischemia-reperfusion injury in the canine hindlimb.

Technetium 99m pyrophosphate quantitation of skeletal muscle ischemia and reperfusion injury.

The study of ischemia and reperfusion injury in the extremity has been hampered by lack of an accurate method of measuring skeletal muscle injury. We used a bilateral isolated in vivo canine gracilis muscle model in 15 anesthetized dogs. The experimental muscles had 4, 6, or 8 hours of ischemia and 1 hour of reperfusion. The contralateral gracilis muscle served as a control. Technetium 99m pyrophosphate (99mTc-PYP), an agent which localizes in injured muscle cells, was used to quantitate canine skeletal muscle damage. After 6 hours of ischemia and 1 hour of reperfusion, there was a significant increase of 215% of 99mTc-PYP uptake in the experimental vs the control muscle. Experimental muscle uptake was 8% greater than control after 4 hours and 405% more after 8 hours of ischemia and reperfusion. Segmental distribution of 99mTc-PYP uptake showed localization to be greatest in the middle of the muscle at the entry site of the gracilis artery. Electron microscopic evaluation also documented this area to have undergone the most severe injury. Distal portions of the muscle did not show increased damage. Our results show that 99mTc-PYP effectively quantitates skeletal muscle ischemia and reperfusion injury. The pattern of 99mTc-PYP uptake suggests that considerable injury is caused during reperfusion.

Alcohol increases mortality in murine head injury.

Head injury is a major factor in the mortality of traumatized patients, accounting for about 50 percent of the resulting fatalities. Alcohol intoxication is frequently (25 to 50 percent) associated with head injuries. This study was undertaken to investigate the effects of alcohol on head trauma in a standardized animal model. Swiss Webster mice (25 ± 2 g) were given intraperitoneally 0.2 mL of either saline or 50 percent ethanol in saline. Thirty minutes later, under light ether anesthesia, severe concussion was produced by dropping a 39.5-g lead weight from a height of 30 cm. The trauma was centered on the midskull by channeling the weight through a vertical tube, 1.2 cm in diameter. Animals were observed daily for eight days. Among the controls, 12 of 12 mice, (100 percent) survived for four days and 8 of 12 (67 percent) survived eight days. In the alcohol recipients, there were 10 of 21 survivors (48 percent) at four days and only one survivor (5 percent) at eight days. This study clearly demonstrates that alcohol increases the lethality of standardized head trauma in mice. The mechanism by which alcohol modifies the effects of craniocerebral trauma remains to be elucidated.

A comparison of isobutyl 2-cyanoacrylate glue, fibrin adhesive, and oxidized regenerated cellulose for control of needle hole bleeding from polytetrafluoroethylene vascular prostheses.

Prolonged anastomotic and needle hole bleeding from synthetic vascular prostheses is a common surgical problem in heparinized patients and in the patient with a coagulopathy. The relative effectiveness of various hemostatic agents has not previously been determined by controlled comparisons. In this study 6 mm polytetrafluoroethylene (PTFE) vascular prostheses were used to perform carotid-carotid and femorofemoral bypasses in eight heparinized dogs. The relative effectiveness of isobutyl 2-cyanoacrylate, fibrin adhesive, and oxidized regenerated cellulose was determined by application of these hemostatic agents to three sources of bleeding: arterial-PTFE anastomoses, PTFE-PTFE anastomoses, and 18-gauge needle holes in PTFE vascular prostheses. The mean time to hemostasis (MTH) was determined for each hemostatic agent. Although application of isobutyl 2-cyanoacrylate resulted in the shortest MTH in each experimental group, the reported inflammatory response that it induces and its possible carcinogenicity limits its availability for clinical use. Application of fibrin adhesive prepared from single-donor hepatitis-screened plasma resulted in a significantly shorter MTH in each experimental group when compared with oxidized regenerated cellulose. We recommend clinical use of fibrin adhesive to control needle hole and anastomotic bleeding from PTFE vascular prostheses.

Effect of splenectomy on acute inflammation.

It is becoming increasingly evident that platelets play a significant role in the inflammatory process. Heterogeneity has been demonstrated among platelets, and evidence has been presented that the spleen retains the youngest and hemostatically most effective platelets. The present study was designed to determine what effect splenectomy, by removing a special population of platelets, would have on acute inflammation as manifested by carrageenan-induced murine foot pad edema. Swiss white mice (25 +/- 2 g) were anesthetized and subjected to splenectomy (Group A, n = 13) or to celiotomy only (Group B, n = 15). One week later, carrageenan (0.1 cc of 1% solution) was injected into the right hind foot pad of each animal. The mice were killed 3.5 hr later, and both hind paws amputated at the tibio-tarsal joint and weighed individually. The change in weight of the experimental hind paw relative to control was expressed as edema index for each animal. The inflammatory reaction, as manifested by carrageenan edema, was significantly (P less than 0.001) decreased in those animals subjected to splenectomy. The mean edema index in Group A was 62.8 per cent compared to 80.9 per cent in Group B. The results of this study indicate that splenectomy decreases inflammation as manifested by carrageenan-induced edema. Splenectomy eliminates a population of platelets having greater functional capabilities than those found in the peripheral circulation and impairs the organism's ability to respond to injury.

Thromboelastographic evaluation of mini-dose heparin in minor surgical procedures.

Five dogs weighing approximately 10 kg were anesthetized with pentobarbital sodium 30 mg/kg intravenously (IV) and an initial sample of blood was drawn by jugular venipuncture for control values of thromboelastography (TEG). Immediately after venipuncture, mini-dose heparin (80 units/kg) was administered subcutaneously. One hour later, 0.5 ml blood was obtained by (A) jugular venipuncture for TEG, and an indwelling catheter immediately introduced into a femoral vein by cut-down. Blood samples for TEG were obtained via cut-down at (B) 15 min (C) 30 min, and (D) 60 min after cut-down. Low-dose heparin had no effect on the hemostatic parameters as measured by TEG and did not prevent the accelerated coagulation associated with surgery.