Although perioperative physicians are generally attentive to anticoagulant medications that can impact hemostasis, herbal and nutritional supplements that have anticoagulant properties continue to be overlooked by many medical providers and patients. This can result in coagulation abnormalities and the potential for increased intraoperative bleeding that is difficult to surgically control.
We report the case of a patient presenting for C2–T2 decompression and fusion with iliac crest bone graft harvest from the left hip who experienced a relatively high amount of blood loss over the course of his surgery despite having discontinued his aspirin 1 week prior. He denied the use of nonsteroidal anti-inflammatory medications or other anticoagulant medications. Despite a normal baseline international normalized ratio (INR) (0.95 2 weeks prior), the surgeon began requesting the administration of fresh frozen plasma (FFP) for “oozing” and inability to surgically control the bleeding. During a discussion with the wife about potential blood-thinning medications and supplements, the attending anesthesiologist discovered that the patient consumed large amounts of garlic up to the night before the surgery. The patient was treated with DDAVP and cryoprecipitate with resolution of bleeding.
CONSENT FOR PUBLICATION
The patient reviewed this article and gave written permission for the authors to publish the report.
A 71-year-old white man presented for C2–T2 decompression and fusion for cervical stenosis and pain that was unresponsive to epidural steroid injections. His medical history also included type 2 diabetes mellitus, asthma, vitamin D deficiency, hypertension, and depression on 100 mg sertraline daily. Aside from sertraline, the patient was also taking 10 mg amlodipine daily, 5 mg hydrocodone/325 mg acetaminophen every 6 hours as needed for pain, 40 mg lisinopril daily, 20 mg omeprazole twice per day, 10 mg pravastatin nightly, 5 mg terazosin nightly, and 50 mg tramadol twice per day as needed for pain. He was seen 2 weeks before his scheduled surgery in the preoperative assessment clinic and advised to continue all his medications through the night before surgery.
His preoperative laboratories showed a prothrombin time (PT) of 10.6 seconds, INR 0.95, and partial thromboplastin time (PTT) 31.7 seconds with a platelet count of 150K/mm3, hemoglobin 13.8 g/dL, hematocrit 42.3%, and white blood cell count 7.2. His basic metabolic profile was normal. The patient was typed and screened (O-RhD[−]), and a right radial arterial line and 2 large-bore intravenous lines were placed before induction. General anesthesia was induced with propofol and fentanyl. Rocuronium was administered to facilitate tracheal intubation. Mechanical ventilation was established, and general anesthesia was maintained with air, oxygen, sevoflurane, and a remifentanil and propofol infusion.
The laminectomy was performed via posterior approach, and near its conclusion, bleeding persisted at a rate of approximately 150 mL/h. The surgeon stated that he was uncomfortable closing with that rate of ongoing blood loss. There was no obvious source of surgical bleeding, so the surgeon requested FFP for ongoing “oozing.” Differential diagnosis included surgical bleeding, platelet dysfunction, anticoagulant effects of medications, and other factors such as temperature and acidosis, which may also affect coagulation. The attending anesthesiologist and surgeon discussed the risks and benefits of FFP transfusion, and both agreed to wait for laboratory testing to target treatment rather than administering unnecessary blood products. The attending anesthesiologist then ordered a coagulation panel that included PT, PTT, INR, fibrinogen, complete blood cell count, and platelet function assay (PFA) to further evaluate the cause of bleeding (Table). Intraoperative arterial blood gas showed a pH 7.33, PCO2 46.6 mmHg, PO2 105 mmHg, base excess −2 mmol/L, HCO3 25 mEq/L, sodium 141 mEq/L, potassium 3.6 mEq/L, calcium 1.17 mg/dL, glucose 130 mg/dL, hematocrit 36%, and hemoglobin 12 g/dL. This mild respiratory acidosis was an unlikely contributor to bleeding abnormalities, but was nonetheless corrected.
While awaiting the laboratory results, the attending anesthesiologist had a discussion with the patient’s wife. She discovered that, although he had not taken any nonsteroidal anti-inflammatory drugs or aspirin products in the past week, the patient did like to chew on ginger and eat “a ton of garlic.” The patient’s wife estimated that he ate approximately 1 teaspoon of crushed garlic with every meal in the weeks leading up to the surgery and had an excessively large amount (1 tablespoon) for dinner the night before. It is estimated that this was equivalent to approximately 4 g of garlic per meal regularly and as much as 12 g of powdered garlic the evening just before surgery.
It was assumed that the patient’s platelet function may have been affected by his chronic garlic consumption, and the decision was made to administer 1 unit of apheresis platelets. Unfortunately, the blood blank had no donor-identical platelets readily available. Thus, DDAVP was administered at a rate of 0.3 μg/kg over 15 minutes. The laboratory results were a normal PFA, low fibrinogen level, and normal PT, PTT, and INR. The surgeon was still not satisfied with intraoperative control of the bleeding 30 minutes after the DDAVP infusion, and blood loss continued despite anesthetically induced hypotension. Five units of cryoprecipitate were administered to correct the hypofibrinogenemia and to potentially augment clot strength. This improved hemostasis and the surgeons then closed. The patient was successfully extubated and taken to the postanesthesia care unit. Total estimated blood loss (EBL) for the case was 850 mL. Although a small amount of output from the Jackson-Pratt drain remained overnight, no further hemostatic blood products were administered.
The patient’s pain was controlled postoperatively with a morphine patient-controlled analgesia. The sequential compression devices (SCDs), which were placed in the operating room, remained on for deep vein thrombosis prophylaxis given the decision to avoid pharmacologic prophylaxis with anticoagulants because of the unusual operating room blood loss. The patient’s hemoglobin/hematocrit remained stable over the next 4 days with output from the Jackson-Pratt drain of 10 to 20 mL per day until it was removed at discharge.
The typical EBL for a laminectomy and fusion at our institution is 500 to 600 mL, which is consistent with that reported in the literature.1 The EBL for the present case, at 850 mL, was approximately 50% higher than normal and not deemed to be surgically related. Although the “oozing” noted by the surgeon may have been largely subjective, and although there may have been other factors affecting coagulation for this patient, this case raises interesting points regarding the effects of herbal supplements on coagulation. Between 5% and 20% of patients presenting for surgery regularly take some sort of herbal supplementation, of which garlic is the most common, accounting for more than one-fifth of the supplements consumed.2,3 This is probably because of its reported potential benefits on cardiovascular health including reduction of cholesterol, suppression of low-density lipoprotein oxidation, and normalization of blood pressure.4 These effects are attributed to the organosulfur compounds contained within garlic, most notably allicin (Figure 1), which also provide its flavor and potent odor. Garlic is quickly absorbed after oral administration with a peak effect between 30 and 120 minutes.5 The minimum absorption rate has been estimated at 65% for allicin.6 The known elimination half-lives of the 2 primary components are 5 and 7 hours for allylmercapturic acid7 and 6 to 11 days for allicin (in vitro),8 although more research is needed to truly delineate the pharmacokinetics in humans.
Platelets are activated by a variety of substances including thrombin, arachidonic acid, collagen, serotonin, and adenosine diphosphate (ADP). Once activated, platelets produce thromboxane, a potent vasoconstrictor and inducer of platelet aggregation, and begin mobilizing a complex set of calcium-dependent processes. These include changing from the resting discoid shape into a sphere with long “pseudopod” extension arms, the release of serotonin granules, and expressing glycoprotein IIb/IIIa receptors to bind fibrinogen. Garlic is thought to interfere with platelet function by altering thromboxane production, preventing degranulation, and interfering with the binding of fibrinogen with glycoprotein IIa/IIIa (Figure 2).9
Impaired coagulation as a result of garlic supplements has been described in multiple settings including urologic, plastic, and ophthalmologic surgery.10–12 Of particular importance in these cases was the preparation and amount of garlic supplements taken. Normal levels of dietary garlic intake at 4 g per day have not been shown to impair platelet function as assessed by thromboelastography (TEG).13 However, there has been a case report of a spontaneous epidural hematoma attributed to 4 cloves per day in an octogenarian,14 and there is insufficient evidence to evaluate the possible additive effects of garlic in combination with platelet-inhibiting drugs or other herbs.
Although not well described in clinical trials, animal studies have shown that processed garlic can also decrease fibrinogen levels at doses equivalent to 25 mg of lyophilized garlic/kg body weight.15 This ability to decrease fibrinogen levels may be dependent on the type of garlic, the dose, and the amount of processing that occurs before consumption. In our patient, it is possible that the consumption of large doses of garlic preoperatively resulted in lowered fibrinogen levels, although his intraoperative level also could have been decreased because of consumption.
One platelet-inhibiting drug on the patient’s prescribed medication list is sertraline, a selective serotonin reuptake inhibitor (SSRI). SSRIs have been associated with increased risk of gastrointestinal bleeding.15 SSRIs with the highest degree of serotonin reuptake inhibition (sertraline, paroxetine, and fluoxetine) are associated with the most frequent coagulation abnormalities related to decreased platelet aggregation and activity with prolongation of bleeding time.15 Serotonin in the blood activates 5-HT2A receptors on platelet membranes to enhance aggregation, and it has been shown that SSRIs can decrease intraplatelet levels of 5-HT.16 This results in less serotonin for platelets to release when activated at the site of injury.17
Unfortunately, it is difficult to diagnose platelet dysfunction in the perioperative setting. The PFA-100 works by directly stimulating platelets with either epinephrine or ADP and measuring the time it takes for blood flow through an aperture to cease (termed closure time). Drugs such as tirofiban and abciximab, which target the platelet αIIBβ3 receptor, will cause prolonged closure times.18 However, the sensitivity of the PFA-100 in detecting platelet granule defects such as those caused by SSRIs is poor.19 Positive predictive value for postoperative bleeding in cardiac surgical patients is only 18%.20 In our particular case, the PFA-100 results were normal (closure times <180 seconds and <120 seconds on epinephrine and ADP, respectively) and could not quantify any platelet defect. The use of other point-of-care coagulation testing such as TEG or rotational thromboelastometry may have provided more insight given their ability to assess fibrinogen as well as platelet contribution to clot strength. However, viscoelastic testing still may not be sensitive enough to detect subtle abnormalities caused by garlic or other nutritional supplements.21
In this particular patient’s case, the intake of excessive amounts of dietary garlic in both crushed and powdered forms combined with sertraline, another platelet-inhibiting agent, may have contributed to increased bleeding in the operating room. This case is unique in that the patient consumed large amounts of garlic in excess of usual dietary dosages (12 g the night before surgery), resulting in increased intraoperative bleeding. As previously discussed, 4 g per day of garlic does not affect platelet function on TEG13; however, 25 mg/kg in rats has been shown to lower fibrinogen levels (equivalent to approximately 1.7 g in a 70-kg human). For the practicing anesthesiologist, it is important to preoperatively assess patients not only for herb and supplement use, but also for dosages consumed. Furthermore, it would be advisable to warn patients regarding the increased risk of bleeding associated with certain common supplements such as garlic. Based on our findings and review of the literature, more than a teaspoon of garlic powder per day may be enough to lower fibrinogen levels and lead to bleeding abnormalities. For patients who regularly consume garlic, it may be advisable to discontinue or decrease their intake to less than this amount in the 1 to 2 weeks preceding high-risk surgery, although further human studies are warranted.
1. Ghogawala Z, Dziura J, Butler WE, et al. Laminectomy plus fusion versus laminectomy alone for lumbar spondylolisthesis. N Engl J Med. 2016;374:1424–1434.
2. Tsen LC, Segal S, Pothier M, Bader AM. Alternative medicine use in presurgical patients. Anesthesiology. 2000;93:148–151.
3. Skinner CM, Rangasami J. Preoperative use of herbal medicines: a patient survey. Br J Anaesth. 2002;89:792–795.
4. Khatua TN, Adela R, Banerjee SK. Garlic and cardioprotection: insights into the molecular mechanisms. Can J Physiol Pharmacol. 2013;91:448–458.
5. Wang CZ, Moss J, Yuan CS. Commonly used dietary supplements on coagulation function during surgery. Medicines (Basel). 2015;2:157–185.
6. Lachmann G, Lorenz D, Radeck W, Steiper M. [The pharmacokinetics of the S35 labeled garlic constituents alliin, allicin and vinyldithiine] [in German]. Arzneimittelforschung. 1994;44:734–743.
7. de Rooij BM, Boogaard PJ, Rijksen DA, Commandeur JN, Vermeulen NP. Urinary excretion of N-acetyl-S-allyl-L-cysteine upon garlic consumption by human volunteers. Arch Toxicol. 1996;70:635–639.
8. Fujisawa H, Suma K, Origuchi K, Kumagai H, Seki T, Ariga T. Biological and chemical stability of garlic-derived allicin. J Agric Food Chem. 2008;56:4229–4235.
9. Rahman K. Effects of garlic on platelet biochemistry and physiology. Mol Nutr Food Res. 2007;51:1335–1344.
10. Gravas S, Tzortzis V, Rountas C, Melekos MD. Extracorporeal shock-wave lithotripsy and garlic consumption: a lesson to learn. Urol Res. 2010;38:61–63.
11. Burnham BE. Garlic as a possible risk for postoperative bleeding. Plast Reconstr Surg. 1995;95:213.
12. Carden SM, Good WV, Carden PA, Good RM. Garlic and the strabismus surgeon. Clin Experiment Ophthalmol. 2002;30:303–304.
13. Scharbert G, Kalb ML, Duris M, Marschalek C, Kozek-Langenecker SA. Garlic at dietary doses does not impair platelet function. Anesth Analg. 2007;105:1214–1218.
14. Rose KD, Croissant PD, Parliament CF, Levin MB. Spontaneous spinal epidural hematoma with associated platelet dysfunction from excessive garlic ingestion: a case report. Neurosurgery. 1990;26:880–882.
15. Jastrzebski Z, Leontowicz H, Leontowicz M, et al. The bioactivity of processed garlic (Allium sativum
L.) as shown in vitro and in vivo studies on rats. Food Chem Toxicol. 2007;45:1626–1633.
16. Halperin D, Reber G. Influence of antidepressants on hemostasis. Dialogues Clin Neurosci. 2007;9:47–59.
17. De Clerck F. The role of serotonin in thrombogenesis. Clin Physiol Biochem. 1990;8(suppl 3):40–49.
18. Hayward CP, Harrison P, Cattaneo M, Ortel TL, Rao AK; Platelet Physiology Subcommittee of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Platelet function analyzer (PFA)-100 closure time in the evaluation of platelet disorders and platelet function. J Thromb Haemost. 2006;4:312–319.
19. McCloskey DJ, Postolache TT, Vittone BJ, et al. Selective serotonin reuptake inhibitors: measurement of effect on platelet function. Transl Res. 2008;151:168–172.
20. Slaughter TF, Sreeram G, Sharma AD, El-Moalem H, East CJ, Greenberg CS. Reversible shear-mediated platelet dysfunction during cardiac surgery as assessed by the PFA-100 platelet function analyzer. Blood Coagul Fibrinolysis. 2001;12:85–93.
21. Bagge A, Schött U, Kander T. Effects of naturopathic medicines on Multiplate and ROTEM: a prospective experimental pilot study in healthy volunteers. BMC Complement Altern Med. 2016;16:64.