Introduction
Bleeding after ENT operations is a common and potentially serious complication, particularly in children because of the small total blood volume and physiological compensation for blood loss. In a large German audit of 720 hospitals, pre-operative routine coagulation estimation failed to identify children who subsequently bled after adenoidectomy or tonsillectomy.Reference Bidlingmaier, Olivieri, Stelter, Eberl, von Kries and Kurnik1 These tests could not routinely identify von Willebrand's disorder. A detailed history to exclude abnormal bleeding or bruising in the patient, or any significant family history of these, is crucial in identifying such patients. This requires detailed pre-operative assessment of clotting and coagulation. In that study of children undergoing tonsillectomy and/or adenoidectomy, bleeding occurred most commonly within the first 24 hours.
Pre-operatively identified Willebrand's disorder has been treated with tranexamic acid and/or 1-desamino-8-D-arginine vasopressin (desmopressin) to reduce the risk of intra- and post-operative bleeding. An increasing body of evidence suggests that tranexamic acid might be a useful drug for preventing bleeding following adenotonsillectomy, managing bleeding during sinus surgery and managing epistaxis.
Tranexamic acid (trans-4 amino methyl-cyclohexane carboxylic acid) is a well-established drug that has been available for over 40 years. While commonly used as an adjunct to desmopressin in patients with von Willebrand's disorder,Reference Choudhuri and Bolton-Maggs2 tranexamic acid has more recently been widely proposed as an important pharmaceutical tool to reduce blood loss during surgery and following major trauma.Reference Perel, Al-Shahi Salman, Kawahara, Morris, Prieto-Merino and Roberts3, Reference Gruen and Reade4 It is included in the current core World Health Organization List of Essential Medicines, which is defined as ‘a list of minimum medicine needs for a basic healthcare system, listing the most efficacious, safe and cost-effective medicines for priority conditions. Priority conditions are selected on the basis of current and estimated future public health relevance, and potential for safe and cost-effective treatment’.5
Tranexamic acid is a potentially useful drug for ENT surgeons in the prevention and reduction of both the frequency and severity of blood loss during and after tonsillectomy.Reference Verstraete6 While the benefits of tranexamic acid seem potentially high, the risks of thromboembolic events and mortality are unclear. In a meta-analysis of tranexamic acid used to manage surgical bleeding, fewer deaths were reported in the tranexamic acid group, but there was uncertainty about serious adverse effects.Reference Ker, Edwards, Perel, Shakur and Roberts7
Pharmacology and applications
Tranexamic acid is a synthetic lysine analogue that facilitates antifibrinolysis via a competitive reversible blockade of lysine-binding sites on plasminogen molecules and as a non-competitive inhibitor of plasmin. These actions inhibit the conversion of plasminogen to plasmin on the surface of the fibrin, producing stabilisation of fibrin clots.Reference Reid, Zimmerman, Laussen, Mayer, Gorlin and Burrows8 Tranexamic acid might be particularly effective at reducing bleeding following ENT surgery as the oral mucosa (and salivary glands) are rich in plasminogen activators and have a low concentration of plasminogen inhibitors.Reference Verstraete6
Tranexamic acid has been widely used parenterally in elective cardiac, orthopaedic and urological surgery to reduce peri-operative blood loss.Reference Mahdy and Webster9 In orthopaedic surgery, tranexamic acid has been shown to reduce the blood loss during knee arthroplasty by 50 per cent, mitigating the need for transfusion with no increase in the risk of thromboembolic complications.Reference Veien, Sorensen, Madsen and Juelsgaard10
A Cochrane review of tranexamic acid for reducing mortality in urgent and emergency surgery concluded that the administration of tranexamic acid reduced the probability that a patient would receive a blood transfusion by around 30 per cent. However, it was noted that further, larger studies were needed to assess any adverse effects in this group of patients.Reference Perel, Ker, Morales Uribe and Roberts11
Tranexamic acid has been used topically as a mouthwash following dental and oral surgery.Reference Sindet-Pedersen, Ramström, Bernvil and Biombäck12 It has been recommended parenterally during cleft palate and lip surgery to help minimise blood loss at the time of operation, particularly following vomerine flap surgery.Reference Watson, James and Walker13 Tranexamic acid mouthwashes used after oral surgery in patients taking anticoagulants (typically 1 g three times daily for 5 days) reduces the rate and volume of bleeding.Reference Senghore and Harris14 For patients undergoing dentoalveolar surgery, without a known clotting disorder, a single dose of peri-operative tranexamic acid administered at 25 mg/kg prevented excessive post-operative bleeding, thereby facilitating day-case discharge.Reference Senghore and Harris14
A randomised, controlled trial of 73 patients undergoing elective bimaxillary osteotomy showed a significant reduction in blood loss in the treatment group following a single dose of tranexamic acid (20 mg/kg) just prior to surgery. The mean total blood loss was 422 ml greater in the placebo group than in the treatment group.Reference Choi, Irwin and Samman15
Prescribed orally, tranexamic acid is part of the management of menorrhagia, and in the UK it can be purchased over the counter to self-manage heavy periods.Reference Dunn and Goa16, 17
As a topical preparation, tranexamic acid is also used as a skin whitener, and for managing skin hyperpigmentation and melasma. The action of tranexamic acid reducing epidermal pigmentation is not fully understood, but is likely to be due to the blocking of plasminogen to keratinocyte binding, with resultant decreased melanocyte tyrosine activity.Reference Na, Choi, Yang, Choi, Kang and Park18–20
In the UK, tranexamic acid is usually prescribed orally at 1 g three times daily, as a 10-day course, for therapeutic indications. In the USA, the tablets are produced in 650 mg strength with a higher recommended daily dose range of 3–6 g. The most common adverse effects of tranexamic acid are gastrointestinal side effects, including nausea, diarrhoea and abdominal cramping, which are generally mild, but uncommon. A current US patent describes a proposed modified release formulation, which could minimise these unwanted effects.21
Renal toxicity is a potential concern in those with reduced renal function. When tranexamic acid is used on a long-term basis in patients with angioneurotic oedema, regular liver function testing and assessment of vision is recommended. Caution is advised when administering tranexamic acid intravenously, as too rapid a bolus injection can produce significant hypotension.
In trauma patients, the benefit of tranexamic acid in reducing blood loss has been demonstrated without a clear increased risk of morbidity or mortality due to thromboembolic sequelae.Reference Perel, Al-Shahi Salman, Kawahara, Morris, Prieto-Merino and Roberts3, Reference Gruen and Reade4 However, in 2011, a report described ischaemic stroke in two younger women receiving tranexamic acid over 3 days for gynaecological bleeding. Following genetic testing, both women were found to have a metabolic genetic abnormality that produced a synergetic effect on the plasminogen inhibitory action of tranexamic acid.Reference Nardi, Pelone, Bartolo, Di, Storto and Notte22 This genetic mutation of methylenetetrahydrofolate reductase is an independent risk factor for occlusive stroke, present in about 10 per cent of the US population, highest in the Hispanic ethnic group and lowest in the Afro-Caribbean group.Reference Schneider, Rees, Liu and Clegg23
In high-dose administration, tranexamic acid with a loading dose of 4 g over 1 hour, and then 1 g per hour for 6 hours, was shown to reduce blood loss and maternal morbidity in women who suffered a postpartum haemorrhage with blood loss of more than 800 ml. The side effects were mild and transient, but the authors acknowledge the study was not adequately powered to address safety issues.Reference Ducloy-Bouthors, Jude, Duhamel, Broisin, Huissoud and Keita-Meyer24 In developing countries, tranexamic acid is likely to be the most cost-effective intervention for postpartum haemorrhage. Of the 500 000 maternal deaths worldwide each year, approximately 99 per cent occur in the Third World.Reference Ronsmans and Graham25 The lifetime risk of mortality from postpartum haemorrhage for a woman living in Northern Europe is 1:30 000, while in the poorest nations the risk is as high as 1:6.Reference Snelgrove26
In ultra-high doses (equal to or greater than 100 mg/kg) administered during cardiopulmonary bypass, the risk of seizure within 24 hours of bypass was related to tranexamic acid administered in a dose-dependent fashion.Reference Kalavrouziotis, Voisine, Mohammadi, Dionne and Dagenais27
Donor blood for transfusion is a scarce resource and the transfusion itself is not without risk. The authors of an authoritative meta-analysis comment that tranexamic acid reduces the need for blood transfusion.Reference Ker, Edwards, Perel, Shakur and Roberts7 In 2011, 1 unit of blood cost the National Health Service (NHS) £125. Worldwide, most do not have access to safe blood, with the inherent risk of transmission of syphilis, human immunodeficiency virus, hepatitis B and C, and variant Creutzfeldt–Jacob disease. In Sub-Saharan Africa, an average of 5 units of blood per 1000 population are donated annually, compared with 30–60 units per 1000 population in First World countries.Reference Guerriero, Cairns, Jayaraman, Roberts, Perel and Shakur28 In those countries where there is a blood shortage and the risk of blood-borne infection is higher, it has been estimated that the cost per life saved using tranexamic acid is US$83–93.Reference Guerriero, Cairns, Jayaraman, Roberts, Perel and Shakur28 The authors conclude that a modest increase in the risk of thromboembolic events could outweigh the benefits of reduced blood use. Nevertheless, the Clinical Randomisation of an Antifibrinolytic in Significant Haemorrhage (‘CRASH-2’) study showed a significant reduction in mortality with no increase in such events, and a statistically significant reduction in the risk of myocardial infarction.Reference Perel, Al-Shahi Salman, Kawahara, Morris, Prieto-Merino and Roberts3
While the published literature supports the use of tranexamic acid for tonsillectomy as a means of reducing blood loss during surgery, and in those with defects of clotting or coagulation, there is little evidence to date to indicate that tranexamic acid could be useful in reducing primary or secondary haemorrhage. In September 2013, the protocol for a Cochrane review of tranexamic acid for the prevention and treatment of tonsillectomy-related haemorrhage in adults was published.Reference Vallamkondu, Ah-See, Rocke and Bannister29 This Cochrane review will investigate primary outcomes, including intra-operative blood loss, post-operative haemorrhage (primary and secondary), and re-admission and return to operating theatre rates. Secondary outcomes will include side effects of tranexamic acid (such as thromboembolic events) and length of hospital stay related to post-operative bleeding.
Tonsillectomy
A number of studies have published variable results on bleeding after tonsillectomy using different types of antifibrinolytics. One study showed no significant benefit from the routine use of tranexamic acid during tonsillectomy,Reference Mazauric, Boudon and Rheingold30 while others have confirmed a significant reduction in blood loss during surgery using conventional dissection techniques.Reference Verstraete, Tyberghein, De Greef, Daems and Van Hoof31, Reference Castelli and Vogt32 In a further randomised, controlled trial, a single dose of intravenous, peri-operative tranexamic acid at 10 mg/kg was associated with a mean blood loss of 36.64 ml, compared with 66.32 ml in the control group.Reference George, Kumar, Kumar and Shetty33 The authors reported no adverse effects from tranexamic acid. Three patients (3 per cent) developed primary haemorrhage, but none of those patients required operative intervention.
In a double-blind, randomised trial from Brazil, in which 95 children underwent adenotonsillectomy for hyperplasia (sic), the authors concluded that tranexamic acid administered at 10 mg/kg pre-operatively, and at 8 and 16 hours post-operatively, produced no significant reduction in blood loss at the time of surgery, nor any reduction in the rate of secondary bleeding at 10 days following the operation.Reference Brum, Miura, de Castro, Machado, Lima and Faibes34 Eight surgeons operated on the 95 children, using cold dissection with absorbable sutures in the tonsillar fossae (sic) for haemostasis. Two surgeons did not complete their randomisation blocks. In different sections of the paper, the authors report either 7 or 12 patients not receiving the second and third doses of tranexamic acid. Seventeen patients outside of the sample calculation were included to complete an intention-to-treat analysis, possibly producing an underestimate of any benefit from tranexamic acid. Given the short half-life of tranexamic acid, it is surprising that the authors anticipated any effect of tranexamic acid administered only on the day of surgery in preventing bleeding 10 days later.
A systematic review and meta-analysis of the use of tranexamic acid in tonsillectomy confirmed that the use of tranexamic acid reduced blood loss during surgery, but failed to demonstrate any reduction in the rate of post-tonsillectomy haemorrhage.Reference Chan, Chan and Tanweer35 The authors note that of 38 citations, an extensive literature search yielded only 7 that could be included in their final analysis. These studies varied enormously in terms of the age range of patients, the dosage of tranexamic acid, and the schedule and duration of treatment. One such paper reported the topical application of a tranexamic acid paste that resulted in a higher bleeding rate than the placebo group.Reference Falbe-Hansen, Jacobsen and Lorenzen36 The small number of studies included in this meta-analysis were so heterogeneous that it is difficult to draw meaningful conclusions regarding the applicability of tranexamic acid in reducing post-operative bleeding after tonsillectomy. A helpful conclusion of this meta-analysis is the safety of tranexamic acid, with only 1 of the 7 studies reporting adverse effects, in 3 out of 40 patients. All the adverse effects were minor, including headache, dizziness and vomiting.
In one pilot study, 476 children aged between 3 and 16 years underwent elective day-case tonsillectomy for recurrent tonsillitis or sleep disordered breathing (within current guidance and commissioning restrictions37, Reference Robb, Bew, Kubba, Murphy, Rollin and Tremlett38), with or without adenoidectomy and/or ventilation tube insertion. In that study, tranexamic acid was administered as a single, slow intravenous peri-operative dose of 10–15 mg/kg.Reference Robb and Thorning39 The same surgeon operated upon all the children, using Coblation™ dissection and haemostasis for the tonsillectomy, and suction diathermy for the adenoidectomy.40 Operative blood loss was not measured in that study, but was clinically negligible, with swabs rarely required for haemostasis. The children were anaesthetised using an adapted version of a previously published day-case anaesthetic protocol.Reference Robb and Ewah41 The protocol was modified to substitute intravenous paracetamol at 15 mg/kg for rectal paracetamol. Following the publication of a recent Medicines and Healthcare Products Regulatory Agency drug safety update, post-operative codeine as a rescue analgesic has since been excluded from the protocol.42 In that study, tranexamic acid appeared to reduce the risk of primary haemorrhage on the day of surgery to 0.4 per cent, facilitating same-day discharge. No bleeding was recorded during the first 24 hours following tonsillectomy. The rate of secondary haemorrhage was not observed. Tranexamic acid is routinely available only in tablet form, which is unsuitable for paediatric administration. The short half-life of tranexamic acid means that no effect on secondary haemorrhage would be expected from a single peri-operative dose.
In a further prospective, observational study of adults undergoing day-case Coblation tonsillectomy between 2007 and 2013, 111 adults were given 1 g of tranexamic acid by slow intravenous injection peri-operatively, and then 1 g 3 times per day orally for 10 days post-operatively.Reference Robb and Ewah43 None of these patients experienced post-operative nausea, vomiting or bleeding on the day of surgery or within the 24 hours following tonsillectomy. Two patients (1.8 per cent) given tranexamic acid post-operatively developed nausea, with one stopping the treatment after a few days. This latter patient had no bleeding. While there were no episodes of significant post-tonsillectomy haemorrhage, three patients (2.7 per cent) spat up some blood or had some bloodstained saliva during the two weeks after surgery; these were reported as minor episodes for which none of the patients sought medical advice. (On the 0–4 Windfuhr classification of post-tonsillectomy haemorrhage, 97.3 per cent of cases were category 0 and 2.7 per cent were category 1.Reference Windfuhr and Seehafer44)
Adenoidectomy
There is almost no literature on the use of tranexamic acid for adenoidectomy without tonsillectomy. A solitary double-blind, prospective, randomised, controlled study of 400 children compared the topical application of tranexamic acid with saline following curettage adenoidectomy. The tranexamic acid (at 1 g in 10 ml of normal saline) was irrigated through a catheter into the nasopharynx and left in situ for 5 minutes before suction clearance of the nose, nasopharynx and pharynx. The volume of blood loss at operation and frequency of primary haemorrhage, with or without postnasal packing or transfusion, was significantly higher in the placebo group.Reference Albirmawy, Saafan, Shehata, Basuni and Eldaba45 It is, however, likely that with the increasing use of direct vision techniques for adenoidectomy (suction diathermy or Coblation), post-adenoidectomy bleeding will become exceptional.
Epistaxis
The most common ENT emergencies are usually the result of bleeding from local trauma and/or infection arising from Kiesselbach's plexus on the antero-inferior part of the nasal septum or from the retro-columellar vein. A detailed history can, however, point to more serious generalised disorders such as leukaemia or bleeding diatheses, typically von Willebrand's disorder.Reference Bennett and Giangrande46 A Cochrane protocol looking at tranexamic acid for epistaxis, first published in 2003, was updated in 2010 with new authors redrafting the protocol; it remains in progress.Reference Joseph, Martinez-Devesa and Perera47
In a randomised, controlled trial, a single intravenous injection of tranexamic acid (500 mg) was compared with anterior nasal packing using a tetracycline and vasoconstrictor solution. In the tranexamic acid group, the bleeding stopped within 10 minutes in 71 per cent of patients, with re-bleeding within 24 hours in 4.7 per cent. In the nasal packing group, the bleeding stopped within 10 minutes in only 31.2 per cent of patients and the re-bleed rate was 11.0 per cent. The authors also reported increased patient satisfaction in the tranexamic acid group and recommended this as an initial treatment for anterior epistaxis.Reference Zahed, Moharamzadeh, Alizadeharasi, Ghasemi and Saeedi48
Endoscopic sinus surgery
In a double-blind, randomised, controlled trial of 28 patients receiving either intravenous tranexamic acid or saline during endoscopic sinus surgery for chronic rhinosinusitis with or without polyposis, there was no measurable difference in intra-operative blood loss. The authors do not report whether the tranexamic acid group had a lower post-operative bleeding rate or less need of nasal packing.Reference Langille, Chiarella, Côté, Mulholland, Sowerby and Dziegielewski49
In a larger comparative analysis of 200 patients undergoing endoscopic sinus surgery, individuals were randomly assigned to a peri-operative intravenous tranexamic acid (500 mg) group or a non-treatment group. In the treatment group, blood loss was reduced by 72.48 per cent (p < 0.05).Reference Chhapola and Matta50 The authors note that the antifibrinolytic effect of tranexamic acid remains in some tissues for about 17 hours and in serum for up to 7–8 hours, potentially reducing the risk of bleeding in the early post-operative period.
A literature review of measures to control bleeding in endonasal, endoscopic skull base surgery concluded that low-dose, topical tranexamic acid (100 mg peri-operatively) provided improved haemostasis and surgical field quality at 2, 4 and 6 minutes after application, with no adverse effects.Reference Thongrong, Kasemsiri, Carrau and Bergese51
Conclusion
Among routine ENT operations, tonsillectomy is an intervention where blood loss during and bleeding after operation are frequent and potentially serious; tranexamic acid could significantly reduce the incidence of haemorrhage. In addition to the medical and psychological morbidity associated with bleeding following tonsil surgery, there is a financial cost, currently excluded from the NHS tariff and borne by the provider organisation.
For a typical general hospital, the annual cost of managing post-tonsillectomy haemorrhage could reach nearly £40 000.Reference Parmar, Hanks, Muir and Evans52 The number of hospitals carrying out tonsillectomy is difficult to confirm, but the best estimate for England is about 200.53 This equates to a financial burden of £8 million annually to manage post-tonsillectomy bleeding. In contrast, tranexamic acid at 1 g three times daily for 10 days costs £2.60.54 With approximately 450 tonsillectomies performed annually in a typical general hospital, the cost of tranexamic acid would be approximately £1170 per hospital per year, equalling £0.23 million nationally, with a potential saving of up to £7.77 million each year from avoidable re-admissions to hospital for post-tonsillectomy bleeding.
There are, to date, no large, high quality prospective studies investigating the applications of tranexamic acid for routine ENT surgery. The few studies published indicate a potential benefit, both in terms of morbidity and cost efficiency. The Cochrane review, currently in progress, is likely to conclude that large multicentre, randomised, prospective trials are needed to determine the efficacy of tranexamic acid applied to ENT surgery. The circumstantial evidence to date is, however, encouraging.
