What is the most appropriate action to stop an occasional episode of epistaxis?

Cochrane Database Syst Rev. 2018; 2018(12): CD004328.

Monitoring Editor: Jonathan Joseph,

Royal National Throat, Nose and Ear Hospital, 330 Gray's Inn Road, LondonUK, WC1X 8DA

John Radcliffe Hospital ‐ West Wing, ENT Department, Headley Way, OxfordUK, OX3 9DU

Nuffield Department of Surgical Sciences, University of Oxford, Cochrane ENT, c/o Cochrane UK, 18 Middle Way, OxfordUK, OX2 7LG

Cochrane UK, Summertown Pavilion, 18 ‐ 24 Middle Way, OxfordUK, OX2 7LG

Abstract

Background

Epistaxis (nosebleed) most commonly affects children and the elderly. The majority of episodes are managed at home with simple measures. In more severe cases medical intervention is required to either cauterise the bleeding vessel, or to pack the nose with various materials. Tranexamic acid is used in a number of clinical settings to stop bleeding by preventing clot breakdown (fibrinolysis). It may have a role in the management of epistaxis as an adjunct to standard treatments, reducing the need for further intervention.

Objectives

To determine the effects of tranexamic acid (oral, intravenous or topical) compared with placebo, no additional intervention or any other haemostatic agent in the management of patients with epistaxis.

Search methods

The Cochrane ENT Information Specialist searched the Cochrane ENT Register (via CRS Web); Central Register of Controlled Trials (CENTRAL) (via CRS Web); PubMed; Ovid Embase; CINAHL; Web of Science; ClinicalTrials.gov; ICTRP and additional sources for published and unpublished trials. The date of the search was 29 October 2018.

Selection criteria

Randomised controlled trials (RCTs) of tranexamic acid (in addition to usual care) compared with usual care plus placebo, usual care alone or usual care plus any other haemostatic agent, to control epistaxis in adults or children.

Data collection and analysis

We used the standard methodological procedures expected by Cochrane. The primary outcomes were control of epistaxis: re‐bleeding (as measured by the proportion of patients re‐bleeding within a period of up to 10 days) and significant adverse effects (seizures, thromboembolic events). Secondary outcomes were control of epistaxis as measured by the time to stop initial bleeding (the proportion of patients whose bleeding is controlled within a period of up to 30 minutes); severity of re‐bleeding (as measured by (a) the proportion of patients requiring any further intervention and (b) the proportion of patients requiring blood transfusion); length of hospital stay and other adverse effects. We used GRADE to assess the quality of the evidence for each outcome; this is indicated in italics.

Main results

We included six RCTs (692 participants). The overall risk of bias in the studies was low. Two studies assessed oral administration of tranexamic acid, given regularly over several days, and compared it to placebo. In the other four studies, a single application of topical tranexamic acid was compared with placebo (one study) and a combination of epinephrine and lidocaine or phenylephrine (three studies). All participants were adults.

Tranexamic acid versus placebo

For our primary outcome, control of epistaxis: re‐bleeding (proportion re‐bleeding within 10 days), we were able to pool data from three studies. The pooled result demonstrated a benefit of tranexamic acid compared to placebo, the risk of re‐bleeding reducing from 67% to 47% (risk ratio (RR) 0.71, 95% confidence interval (CI) 0.56 to 0.90; three studies; 225 participants; moderate‐quality evidence).

When we compared the effects of oral and topical tranexamic acid separately the risk of re‐bleeding with oral tranexamic acid reduced from 69% to 49%, RR 0.73 (95% CI 0.55 to 0.96; two studies, 157 participants; moderate‐quality evidence) and with topical tranexamic acid it reduced from 66% to 43%, RR 0.66 (95% CI 0.41 to 1.05; single study, 68 participants). We rated the quality of evidence provided by the single study as low, therefore it is uncertain whether topical tranexamic acid is effective in stopping bleeding in the 10‐day period after a single application.

No study specifically sought to identify and report our primary outcome: significant adverse effects (i.e. seizures, thromboembolic events).

The secondary outcome time to stop initial bleeding (proportion with bleeding controlled within 30 minutes) was measured in one study using topical tranexamic acid and there was no evidence of a difference at 30 minutes (RR 0.79, 95% CI 0.56 to 1.11; 68 participants; low‐quality evidence).

No studies reported the proportion of patients requiring any further intervention (e.g. repacking, surgery, embolisation).

One study of oral tranexamic acid reported the proportion of patients requiring blood transfusion and found no difference between groups: 5/45 (11%) versus 6/44 (14%) (RR 0.81, 95% CI 0.27 to 2.48; 89 participants; low‐quality evidence).

Two studies reported hospital length of stay. One study reported a significantly shorter stay in the oral tranexamic acid group (mean difference (MD) ‐1.60 days, 95% CI ‐2.49 to ‐0.71; 68 participants). The other study found no evidence of a difference between the groups.

Tranexamic acid versus other haemostatic agents

When we pooled the data from three studies the proportion of patients whose bleeding stopped within 10 minutes was significantly higher in the topical tranexamic acid group compared to the group receiving another haemostatic agent (70% versus 30%: RR 2.35, 95% CI 1.90 to 2.92; 460 participants) (moderate‐quality evidence).

Adverse effects across all studies

Five studies recorded 'adverse effects' in a general way. None found any difference between the groups in the occurrence of minor adverse effects (e.g. mild nausea and diarrhoea, 'bad taste' of gel). In one study a patient developed a superficial thrombophlebitis of both legs following discharge, however it is not reported in which group this occurred. No "other serious adverse effect" was reported in any study.

Authors' conclusions

We found moderate‐quality evidence that there is probably a reduction in the risk of re‐bleeding with the use of either oral or topical tranexamic acid in addition to usual care in adult patients with epistaxis, compared to placebo with usual care. However, the quality of evidence relating solely to topical tranexamic acid was low (one study only), so we are uncertain whether or not topical tranexamic acid is effective in stopping bleeding in the 10‐day period after a single application. We found moderate‐quality evidence that topical tranexamic acid is probably better than other topical agents in stopping bleeding in the first 10 minutes.

There have been only three RCTs on this subject since 1995. Since then there have been significant changes in nasal cauterisation and packing techniques (for example, techniques including nasal endoscopy and more invasive approaches such as endoscopic sphenopalatine artery ligation). New trials would inform us about the effectiveness of tranexamic acid in light of these developments.

Plain language summary

Tranexamic acid to help treat nosebleeds (epistaxis)

Background

Nosebleeds are a very common condition, with the majority of those affected either children or those over the age of 60. They usually stop on their own or by simply compressing the nose with fingers, although a small number require medical attention. This will involve either cauterising (sealing) the bleeding vessel, if it can be seen, or packing the inside of the nose with a material to cause pressure to build up and stop the bleeding ('usual care'). Occasionally bleeding continues despite these measures, or it restarts having initially been controlled. This can lead to a prolonged hospital stay and the possibility of further procedures such as repacking with a different type of nasal pack or an operation.

Tranexamic acid is a drug that is known to help promote blood clotting by preventing a natural process called fibrinolysis (dissolution of a clot). It is already used in a number of situations where bleeding is a significant worry, such as after heart surgery or major trauma. It can be given by mouth (orally), directly to the bleeding site (topically) or by injection into a vein (intravenously).

Study characteristics

We searched for randomised controlled trials in patients of any age with nosebleed requiring intervention. Patients were treated with tranexamic acid (in addition to usual care) compared to placebo, no treatment or any other agent used to stop bleeding. We found six studies that met our inclusion criteria, with a total of 692 participants. Two studies used oral administration of tranexamic acid and four used topical administration. All participants in the studies were adults. Three of the six studies were conducted over 20 years ago.

Key results

Three studies measured re‐bleeding within 10 days. When we combined the results we found that fewer patients who were given either oral or topical tranexamic acid had further episodes of re‐bleeding following an initial nosebleed compared to those treated with usual care.

The time to stop initial bleeding (control of bleeding within 30 minutes) was measured in four studies. In three studies the proportion of patients whose bleeding stopped within 10 minutes was significantly higher in the group receiving topical tranexamic acid compared to the group receiving a different drug (topical epinephrine and lidocaine or phenylephrine). In the other study there was no significant difference at 30 minutes when topical tranexamic acid was compared with placebo.

No studies reported the proportion of patients requiring any further intervention (e.g. repacking, surgery).

Only one study of oral tranexamic acid reported the proportion of patients requiring a blood transfusion and there was no evidence of a difference between the groups.

Length of hospital stay was reported in two studies. One study reported a significantly shorter stay in the oral tranexamic acid group, while the other found no evidence of a difference.

Five studies mention recording "adverse effects". None found any difference between the groups in the occurrence of minor adverse effects (e.g. mild nausea and diarrhoea, 'bad taste' of gel). In one study a patient did develop a superficial thrombophlebitis (inflammation and a blood clot in a vein near the surface of the skin) of both legs following discharge, but the study did not report in which treatment group this happened. No serious adverse event was seen in any of the studies.

Quality of the evidence and conclusions

Overall, the risk of bias in the six studies was low. We graded the quality of the evidence for the main outcome (control of epistaxis: re‐bleeding within 10 days) as moderate, which means that further research is likely to have an important impact on our confidence in the estimate of the effect and is likely to change the estimate. In light of this and the fact that 'usual care' has changed, with the development of more modern nasal cauterisation and packing techniques, since three of the included studies were carried out, there remains uncertainty about the role of tranexamic acid in the treatment of patients with epistaxis. Newer research into the effect of tranexamic acid as a treatment for nosebleeds would inform future management decisions for this condition.

The evidence in this review is up to date to October 2018.

Summary of findings

Background

Description of the condition

Epistaxis (nosebleed) is extremely common. It has a pronounced bimodal distribution, being common in childhood and then becoming less frequent before the incidence rises again in the sixth decade (McGarry 2008). Its prevalence in random samples of the population was found in one study to be 10% to 12% at any given time (Shaheen 1967). The cause of epistaxis is unknown in 70% to 80% of cases (idiopathic epistaxis) (Stell 1977). However, epistaxis may be secondary to a number of causes such as surgery, trauma, hypertension, coagulation abnormalities, hereditary haemorrhagic telangiectasia and the use of medications such as aspirin and warfarin (McGarry 2008).

Most epistaxis is self‐limiting or settles with simple measures such as compression (pinching the nose for example), and is ideally managed at home without further medical interference. Patients with more severe bleeding may require medical attention. Two basic methods are used: cauterisation to seal the bleeding vessel if it can be seen and, when this is not possible, various materials may be used to pack the nose to arrest the blood flow. In anterior epistaxis, which is the most common type, a bleeding point is often visible and can be cauterised under local anaesthetic by either a locally applied chemical such as silver nitrate or by electrocautery (hot wire). In some cases this is not sufficient and nasal packing is required with materials such as ribbon gauze, nasal tampons, balloons or inflatable packs. Posterior epistaxis is more common in older adults with hypertension and arteriosclerosis. In these cases a bleeding point is often not identified and nasal packing is necessary. This packing usually stays in for one to two days and applies pressure to the bleeding site thereby stopping the bleeding.

Re‐bleeding may occur after initial management, which often results in a longer hospital stay and further treatments including repeat nasal packing, surgery or embolisation.

If bleeding persists or recurs despite nasal packing then surgical intervention may be necessary and if the blood loss is great, transfusion may be necessary.

Patients admitted with a severe epistaxis, whether anterior or posterior, may face significant morbidity. Nasal packing is uncomfortable. There is also a very small risk of death from a number of causes including the epistaxis itself (when uncontrollable) and the effects of treatments. For example, nasal packing can lead to hypoxia and, if a general anaesthetic is required as part of the treatment, this also carries small but significant risks.

Description of the intervention

Tranexamic acid is an anti‐fibrinolytic agent, used to prevent or treat bleeding in a wide variety of clinical situations. It can be administered orally or intravenously, or applied topically. It is contra‐indicated in patients with thromboembolic disease such as stroke and heart attack.

How the intervention might work

Tranexamic acid stabilises blood clots by competitively inhibiting the binding of plasminogen to fibrin preventing fibrinolysis. This antifibrinolytic effect means that it is routinely used to reduce excessive bleeding and to prevent re‐bleeding in many clinical situations (Ker 2012; Roberts 2013). For example, it is used following cardiac surgery when cardiopulmonary bypass is used and in acute upper gastrointestinal bleeding (Dunn 1999).

Tranexamic acid may therefore have a role to play in the management of epistaxis as an adjunct to usual therapies. It may have roles in (a) stopping initial bleeding, and (b) preventing or reducing the frequency and/or severity of re‐bleeding. This will reduce the need for further interventions with all their attendant risks.

Tranexamic acid to stop or minimise initial bleeding

In patients with epistaxis, tranexamic acid may help stop bleeding at the time of initial administration (or very shortly thereafter), preventing the need for more invasive treatments such as cautery and nasal packing. It is uncertain whether a single dose (or a very short period of treatment) might result in a more rapid resolution of the patient's condition with fewer re‐bleeding episodes and the need for fewer interventions in a 'recovery' period lasting several days.

The initiation of a course of tranexamic acid treatment destined to last several days may have a similar effect.

Tranexamic acid to prevent re‐bleeding

Tranexamic acid may be prescribed regularly, over a number of days after the initial epistaxis, in order to achieve similar ends: that is, more rapid resolution of the patient's condition with fewer re‐bleeding episodes and the need for fewer interventions in that period.

Adverse effects

The most common adverse events are gastrointestinal (for example, nausea, diarrhoea and abdominal cramping), which are mild and uncommon (Robb 2014). However, thromboembolic events (Nishihara 2015) and seizures (Sharma 2014) have been reported (although these two studies do not evaluate patients with epistaxis or receiving tranexamic acid in doses similar to patients with epistaxis).

Why it is important to do this review

There is uncertainty about the role of tranexamic acid in the management of patients with epistaxis. A recent audit of epistaxis management in the UK analysed the data from 1122 patients and found that tranexamic acid was used in 8.2% of cases (UK Epistaxis Audit 2017). Time to haemostasis was longer and re‐bleeding rates were higher in these patients; however, the number of patients was small and they had a "higher degree of illness". A consensus document on the hospital management of epistaxis (BRS 2017) highlighted the inconsistent findings from a systematic review (Williams 2017) and therefore did not recommend the use of either oral or topical tranexamic acid, except in relevant cases (i.e. following national guidelines on the use of tranexamic acid in major haemorrhage). Further studies are, however, ongoing and their findings may expand the evidence base (ISRCTN34153772; NCT02930941; NCT03360045).

An up‐to‐date Cochrane Review, which can be updated as further studies are completed, is warranted.

Objectives

To determine the effects of tranexamic acid (oral, intravenous or topical) compared with placebo, no additional intervention or any other haemostatic agent in the management of patients with epistaxis.

Methods

Criteria for considering studies for this review

Types of studies

We included studies with the following design characteristics:

• Randomised controlled trials (RCTs) including cluster‐randomised trials.

• Patients were followed up for at least seven days.

Types of participants

Patients of any age with epistaxis requiring intervention by a healthcare professional.

Exclusion criteria

• Patients with a clotting or bleeding disorder, sinonasal malignancy and chronic inflammatory nasal conditions.

Types of interventions

Tranexamic acid in any formulation, delivered orally, intravenously or topically.

If other interventions were used at the start of treatment, when tranexamic acid or placebo was given or first started, these should have been used in both treatment arms. Allowed co‐interventions included cautery and nasal packing.

The comparators were:

• placebo;

• no treatment;

• any other haemostatic agent.

After the start of treatment, patients may require further interventions to control re‐bleeding. These include: cauterisation, nasal packing, surgery and embolisation.

The main comparison pair was:

• tranexamic acid versus placebo or no treatment.

A second comparison pair was:

• tranexamic versus any other haemostatic agent.

For a more detailed description of how we planned to group the interventions together for analysis see Subgroup analysis and investigation of heterogeneity.

Types of outcome measures

We analysed the following outcomes in the review, but we did not use them as a basis for including or excluding studies.

Primary outcomes

• Control of epistaxis: re‐bleeding, as measured by the proportion of patients re‐bleeding within a period of up to 10 days (minimum five days).

• Significant adverse effects: seizures, thromboembolic events.

Secondary outcomes

• Control of epistaxis: time to stop initial bleeding (as measured by the proportion of patients whose bleeding is controlled within 30 minutes).

• Severity of re‐bleeding, as measured by:

  • the proportion of patients requiring any further intervention (e.g. repacking, surgery, embolisation) within 10 days;

  • the proportion of patients requiring blood transfusion within 10 days.

• Length of hospital stay in days.

• Other adverse effects.

Search methods for identification of studies

The Cochrane ENT Information Specialist conducted systematic searches for randomised controlled trials and controlled clinical trials. There were no language, publication year or publication status restrictions. The date of the search was 29 October 2018.

Electronic searches

The Information Specialist searched:

• the Cochrane ENT Register (searched via CRS Web 29 October 2018);

• the Cochrane Central Register of Controlled Trials (CENTRAL) (searched via CRS Web 29 October 2018);

• PubMed (1946 to 29 October 2018);

• Ovid EMBASE (1974 to 29 October 2018);

• Ovid CAB Abstracts (1910 to 29 October 2018);

• EBSCO CINAHL (1982 to 29 October 2018);

• LILACS, lilacs.bvsalud.org (searched 29 October 2018);

• KoreaMed, www.koreamed.org (searched 29 October 2018);

• IndMed, www.indmed.nic.in (searched 29 October 2018);

• PakMediNet, www.pakmedinet.com (searched 29 October 2018);

• Web of Knowledge, Web of Science (1945 to 29 October 2018);

• CNKI, www.cnki.com.cn (searched via Google Scholar 29 October 2018);

• ClinicalTrials.gov, (searched via CRS Web 29 October 2018);

• World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP), www.who.int/ictrp (searched 29 October 2018).

In searches prior to 2012, we also searched BIOSIS Previews 1926 to November 2012.

The Information Specialist modelled subject strategies for databases on the search strategy designed for CENTRAL. Where appropriate, they were combined with subject strategy adaptations of the highly sensitive search strategy designed by Cochrane for identifying randomised controlled trials and controlled clinical trials (as described in the Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0, Box 6.4.b. (Handbook 2011). Search strategies for major databases including CENTRAL are provided in Appendix 1.

Searching other resources

We scanned the reference lists of identified publications for additional trials and contacted trial authors where necessary. In addition, the Information Specialist searched PubMed to retrieve existing systematic reviews relevant to this systematic review, so that we could scan their reference lists for additional trials. The Information Specialist also ran non‐systematic searches of Google Scholar to retrieve grey literature and other sources of potential trials.

Data collection and analysis

Selection of studies

Two authors (JJ and PMD or JB and MB) independently scanned the search results to identify studies that loosely met the inclusion criteria. We then independently reviewed the full texts of the retrieved trials and applied the inclusion criteria. We resolved any differences between the authors in study selection through discussion and reaching a consensus. We documented our search and study selection process and depicted this graphically in a PRIMSA flowchart.

Data extraction and management

We extracted data from the studies using standardised data forms, allowing for an intention‐to‐treat analysis. The main data items that we extracted were: date, duration and setting for the study, number of participants, participant (baseline) characteristics (age, gender, other), type of 'usual care' provided, method of delivery and dosage of tranexamic acid and type of outcome measures used (see Appendix 2). Where data were missing we attempted to contact the authors of the study to request the missing information.

Assessment of risk of bias in included studies

Two authors (JJ and PMD) undertook independent assessment of the risk of bias of the included studies. The following were taken into consideration, as guided by the Cochrane Handbook for Systematic Reviews of Interventions (Handbook 2011):

• sequence generation;

• allocation concealment;

• blinding;

• incomplete outcome data;

• selective outcome reporting; and

• other sources of bias.

We used the Cochrane 'Risk of bias' tool in RevMan 5 (RevMan 2014), which involves describing each of these domains as reported in the study and then assigning a judgement about the adequacy of each entry: low, high or unclear (or unknown) risk of bias.

Measures of treatment effect

We summarised the effects of dichotomous outcomes (e.g. proportion of patients with re‐bleeding) as risk ratios (RR) with CIs. For the key outcomes that we presented in the 'Summary of findings' table, we also expressed the results as absolute numbers based on the pooled results and compared to the assumed risk. The assumed baseline risk is typically either (a) the median of the risks of the control groups in the included studies, this being used to represent a 'medium‐risk population' or, alternatively, (b) the average risk of the control groups in the included studies is used as the 'study population' (Handbook 2011). If a large number of studies were available, and where appropriate, we also planned to present additional data based on the assumed baseline risk in (c) a low‐risk population and (d) a high‐risk population.

For continuous outcomes, we expressed treatment effects as a mean difference (MD) with standard deviation (SD).

Unit of analysis issues

If we had found cluster‐randomised trials, we would have analysed these according to the methods in section 16.3.3 of the Cochrane Handbook for Systematic Reviews of Interventions (Handbook 2011).

Dealing with missing data

We planned to undertake an intention‐to‐treat analysis where possible.

We planned to contact study authors via email whenever the outcome of interest was not reported, if the methods of the study suggested that the outcome had been measured. We planned to do the same if not all data required for meta‐analysis had been reported, unless the missing data were standard deviations. If standard deviation data were not available, we planned to approximate these using the standard estimation methods from P values, standard errors or 95% CIs if these were reported as detailed in the Cochrane Handbook for Systematic Reviews of Interventions (Handbook 2011). If it was impossible to estimate these, we would have contacted the study authors.

Apart from imputations for missing standard deviations, we did not plan to conduct any other imputations.

Assessment of heterogeneity

We assessed clinical heterogeneity (which may be present even in the absence of statistical heterogeneity) by examining the included studies for potential differences between them in the types of participants recruited, interventions or controls used or outcomes measured. In this review we anticipated that the intervention might have been used in two different ways: as a single treatment at the time of bleeding, or over a period following initial treatment to prevent re‐bleeding, or a combination of both.

We assessed statistical heterogeneity using the Chi² test with P < 0.05 indicating significance. We also used the I² statistic with the following thresholds for assessing the impact of the heterogeneity the pooled analyses:

• 0% to 40%: might not be important;

• 30% to 60%: may represent moderate heterogeneity;

• 50% to 90%: may represent substantial heterogeneity;

• 75% to 100%: considerable heterogeneity.

Assessment of reporting biases

We planned to assess reporting bias as between‐study publication bias (using funnel plots if sufficient studies were available) and as within‐study outcome reporting bias. There was sometimes insufficient information to judge the risk of selective reporting bias: we noted this as an 'unclear' risk of bias.

Data synthesis

We conducted all meta‐analyses using Review Manager 5.3 (RevMan 2014). For dichotomous data, we analysed treatment differences as a risk ratio (RR) calculated using the Mantel‐Haenszel method.

For continuous outcomes, we planned to pool mean values obtained at follow‐up and report a mean difference (MD).

When statistical heterogeneity is low, random‐effects versus fixed‐effect methods yield trivial differences in treatment effects. However, when statistical heterogeneity is high, the random‐effects method provides a more conservative estimate of the difference.

Subgroup analysis and investigation of heterogeneity

We planned to conduct the following subgroup analyses regardless of whether statistical heterogeneity was observed, as these may be effect modifiers:

• mode and timing of administration;

• patient age (children versus adults);

• setting (inpatient versus outpatient);

• site of nosebleed (anterior versus posterior).

The various modes of administration of tranexamic acid are likely to affect the primary outcome in different ways. Topical administration is likely to have a faster effect but with a shorter duration. Oral and intravenous delivery will have a delayed onset (greater with oral administration than intravenous) but if repeated doses are given, the duration of effect will last at least as long as therapeutic doses are being administered.

Sensitivity analysis

We planned to carry out sensitivity analyses to determine whether the findings were robust to the decisions made in the course of identifying, screening and analysing the trials. We planned to conduct sensitivity analysis for the following factors, whenever possible, and where applicable:

• impact of model chosen: fixed‐effect versus random‐effects model;

• risk of bias of included studies: excluding studies with high risk of bias (we defined these as studies that had a high risk of allocation concealment bias and a high risk of attrition bias (overall loss to follow‐up of 20%, differential follow‐up observed).

If any of these investigations found a difference in the size of the effect or heterogeneity, we had planned to mention this in the Effects of interventions section.

GRADE and 'Summary of findings' table

Two authors (MB, JB) independently used the GRADE approach to rate the overall quality of evidence. The quality of evidence reflects the extent to which we are confident that an estimate of effect is correct and we applied this in the interpretation of results. There are four possible ratings: high, moderate, low and very low. A rating of high quality of evidence implies that we are confident in our estimate of effect and that further research is very unlikely to change our confidence in the estimate of effect. A rating of very low quality implies that any estimate of effect obtained is very uncertain.

The GRADE approach rates evidence from RCTs which do not have serious limitations as high quality. However, several factors can lead to the downgrading of the evidence to moderate, low or very low. The degree of downgrading is determined by the seriousness of these factors:

• study limitations (risk of bias);

• inconsistency;

• indirectness of evidence;

• imprecision; and

• publication bias.

We included 'Summary of findings' tables for the comparisons of tranexamic acid versus placebo or no treatment and tranexamic acid versus other haemostatic agent, constructed according to the recommendations described in Chapter 10 of the Cochrane Handbook for Systematic Reviews of Interventions (Handbook 2011).

We planned to include the following seven outcomes in the 'Summary of findings' table, but only those where data were available were actually included:

• Primary outcomes

  • Control of epistaxis: re‐bleeding (the proportion of patients re‐bleeding within a period of up to 10 days (minimum five days)).

  • Significant adverse effects (seizures, thromboembolic events).

• Secondary outcomes

  • Control of epistaxis: time to stop initial bleeding (the proportion of patients whose bleeding is controlled within 30 minutes).

  • Severity of re‐bleeding, as measured by:

    • the proportion of patients requiring any further intervention (e.g. repacking, surgery, embolisation) within 10 days);

    • the proportion of patients requiring blood transfusion within 10 days).

  • Length of hospital stay in days.

  • Other adverse effects.

Results

Description of studies

Results of the search

A total of 680 references were retrieved by the searches in October 2018. This reduced to 431 following de‐duplication and removal of clearly irrelevant references. We discarded 401 records in first‐level screening, leaving 30 references for further consideration.

We excluded 21 studies (see Excluded studies below and Characteristics of excluded studies for the reasons for exclusion).

We identified three ongoing studies (see Characteristics of ongoing studies). No studies are awaiting assessment.

We selected six studies for inclusion in the review.

A PRISMA flow diagram depicting our search and selection process is shown in Figure 1.

Process for sifting search results and selecting studies for inclusion.

Included studies

We included six studies with a total of 692 participants (Petruson 1974; Tibbelin 1995; White 1988; Zahed 2013; Zahed 2018). Full details of the included studies are shown in the Characteristics of included studies table.

Sample sizes

The number of participants in each study ranged from 68 to 216.

Interventions

Oral tranexamic acid for 10 days

Both the inpatient studies used the same intervention, 1 g tranexamic acid tablets, three times daily, to begin within one hour of admission and to continue for 10 days, or placebo. They differed in what was termed 'usual care':

Petruson 1974 used packing alone, comprising a posterior Foley catheter in all cases with the addition of an anterior gauze tampon if necessary. Twelve to 24 hours after bleeding was stopped, the balloon was deflated and the catheter removed three to six hours later. If there was no further bleeding the gauze was removed until fresh blood was seen on it. Every three to six hours the gauze was removed in a similar fashion until it was all out. If there was a new severe bleed, the patients were re‐packed as above.

White 1988 treated the patients with a combination of nasal cautery and packing. Packing was removed 24 hours after bleeding had stopped and patients were discharged home a further 24 hours later if there was no further bleeding.

Topical tranexamic acid (gel or liquid)

Tibbelin 1995 treated the study participants with either 15 mL tranexamic acid gel (10%) or placebo gel applied locally to fill the nasal cavity. If bleeding was not arrested 30 minutes after gel application, traditional methods were used, which constituted 'usual care'.

In the intervention group in both Zahed 2013 and Zahed 2018 a 15 cm cotton pledget soaked in the injectable form of tranexamic acid (500 mg in 5 mL) was placed in the nostril of the bleeding side. It was removed after bleeding was arrested. If "rescue treatment" was needed (presumably meaning that if bleeding did not stop) "routine anterior nasal packing and cautery" was used. The study participants in the control groups of both studies were treated with a cotton pledget soaked in epinephrine (1:100,000) and lidocaine (2%) to "decongest" the nose for 10 minutes followed by packing with several cotton pledgets coated in tetracycline on the bleeding side for three days. "Rescue treatment" for the control group in both studies consisted of nasal cautery.

There are differences (or potential differences) between the intervention and control arms of the Zahed 2013 and Zahed 2018 studies other than the application of tranexamic acid after the first 10 minutes of treatment. In the first 10 minutes, the two groups in both studies had either tranexamic acid or epinephrine/lidocaine; thereafter, the control groups were all packed and the intervention groups may or may not have been. This means that from 10 minutes to three days, one comparator group ‐ the control group ‐ comprised participants all of whose noses were cauterised and packed, and the other group had an unknown proportion of participants (< 100%) who were cauterised and packed. For this reason, Zahed 2013 and Zahed 2018 therefore contributed data only to the comparison tranexamic versus any other haemostatic agent, and only to one epistaxis‐related outcome (see below) and to adverse effects.

In Atabaki 2017, the intervention group received 1 cc tranexamic acid (injection solution, 500 mg/5 mL) poured onto a cotton ball and inserted into each nasal cavity. The control group received 1 cc phenylephrine (phenylephrine HCL, nasal drops, 0.5%) administered in the same way.

Outcomes

Control of epistaxis: re‐bleeding, as measured by the proportion of patients re‐bleeding within a period of up to 10 days (minimum five days)

Three studies recorded the number of patients with bleeding episodes over a 7‐ to 10‐day period (Petruson 1974; Tibbelin 1995; White 1988).

Significant adverse effects: seizures, thromboembolic events

Five of the studies referred to recording "adverse effects" and in Tibbelin 1995, White 1988, Zahed 2013 and Zahed 2018 mention is made of looking for "severe" adverse effects but these were not pre‐defined. Separate data relating to any adverse effects in the first 10 minutes of the Zahed 2013 and Zahed 2018 studies were not available. Atabaki 2017 did not report these.

Control of epistaxis: time to stop initial bleeding (as measured by the proportion of patients whose bleeding is controlled within 30 minutes)

This outcome was evaluated in four studies: Tibbelin 1995 (30 minutes) and Atabaki 2017, Zahed 2013 and Zahed 2018 (10 minutes).

Severity of re‐bleeding: as measured by a) the proportion of patients requiring any further intervention (e.g. repacking, surgery, embolisation) within 10 days and b) the proportion of patients requiring blood transfusion within 10 days

None of the studies reported the proportion of patients requiring any further intervention. The proportion of patients requiring blood transfusion within 10 days was reported only in White 1988.

Other adverse effects

As mentioned above, five studies mention recording "adverse effects".

Excluded studies

We excluded 21 studies. The reasons for exclusion are detailed in the Characteristics of excluded studies table. Several studies specifically studied tranexamic acid in participants with bleeding abnormalities such as hereditary haemorrhagic telangiectasia or in patients undergoing surgery (such as functional endoscopic sinus surgery or rhinoplasty).

Risk of bias in included studies

Please see the 'Risk of bias' table for each of the included studies (Characteristics of included studies), Figure 2 for a 'Risk of bias' graph (our judgements about each risk of bias item presented as percentages across all included studies) and Figure 3 for a 'Risk of bias' summary (our judgements about each risk of bias item for each included study).

'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.

As the Figures show, the overall risk of bias in the included studies is low. We found White 1988, Petruson 1974 and Tibbelin 1995 to have a low risk of bias overall while Zahed 2013 and Zahed 2018 at high risk due to the absence of blinding. In White 1988, the risk of attrition bias was unclear, but the number of participants lost to follow‐up was small. Atabaki 2017 had an unclear risk of bias in several domains.

Allocation

Sequence generation

We assessed all studies to be at low risk of selection bias with regard to sequence generation: all used randomisation, although two studies did not fully define how this was done (Petruson 1974; Tibbelin 1995).

Blinding

Three studies are described as 'double‐blind' and were at low risk of bias (Petruson 1974; Tibbelin 1995; White 1988); however, only White 1988 fully defined how this was achieved. Atabaki 2017 is described as a "double‐blind" study but no further details are provided (unclear risk of bias).

It was not possible to blind the participants and those administering treatment in Zahed 2013 and Zahed 2018 due to clear differences in the methods of administration in the two groups (high risk of bias). Those analysing the data were blinded making these single‐blinded studies, however as only data from the first 10 minutes are used in this review, this is the only relevant time period for outcome assessment and evaluators at this point were not blind to the intervention.

Incomplete outcome data

The risk of attrition bias in four studies was low (Atabaki 2017; Tibbelin 1995; Zahed 2013; Zahed 2018). There is an unclear risk of bias due to a lack of data in one study (Petruson 1974). In the other included study seven of 96 patients (7%) did not complete the course of treatment with no explanation given or comment on the distribution between groups (White 1988). The data for these patients were removed from the analysis, which leads to an unclear risk of bias in this study.

Selective reporting

All planned outcomes from the methods section in each study were reported on in Atabaki 2017, Petruson 1974, Tibbelin 1995, Zahed 2013 and Zahed 2018. Zahed 2013 and Zahed 2018 had low risk of bias due to robust outcome reporting strategies. There was some lack of clarity in the other three studies due to unclear methods of reporting of efficacy variables in one (Tibbelin 1995), failure to report the number of units required for blood transfusion in another (Petruson 1974), and general poor reporting of patient allocation in the last (White 1988). In addition, several outcomes were not presented in the methods section of White 1988, but are reported in the outcomes section (length of stay, number of blood transfusions and amount of blood transfused).

Other potential sources of bias

There is an unclear risk of other bias. Petruson 1974 noted bleeding resulting from the removal of nasal tampons but did not record these bleeds in their results. The overall effect of this is unknown. Tibbelin 1995 noted that the baseline bleeding intensity was not equal between the treatment and control groups. They had applied a linear logistic model to data affected by this. Both Zahed 2013 and Zahed 2018 noted a difference between the treatment and control groups in the history of epistaxis. There were no concerns in the other study and therefore the risk of other bias is low (White 1988). We detected no other sources of bias in Atabaki 2017, but our assessments are based on information extracted by a translator of this study, which is published in Farsi, so the judgement is unclear.

None of the studies gave details of study funding. Only one study provided a statement about conflicts of interest and declared that there were none (Zahed 2013).

Effects of interventions

See: Table 1; Table 2

Summary of findings for the main comparison

Tranexamic acid compared to placebo plus usual care or usual care alone for patients with nasal haemorrhage (epistaxis)

Summary of findings 2

Tranexamic acid compared to other haemostatic agent for patients with nasal haemorrhage (epistaxis)

Tranexamic acid versus placebo or usual care alone

Primary outcomes

Control of epistaxis: re‐bleeding (proportion of patients re‐bleeding within a period of up to 10 days)

See Table 1.

The primary outcome measure for this review was the effectiveness of tranexamic acid in the control of epistaxis. We measured this primarily as the proportion of patients who had an episode of re‐bleeding within the first 10 days of treatment. Combining the results of three studies with a total of 225 participants, we found moderate‐quality evidence that tranexamic acid probably reduces the risk of re‐bleeding (risk ratio (RR) 0.71, 95% confidence interval (CI) 0.56 to 0.90, I² = 0%) (Analysis 1.1; Figure 4). In the three studies tranexamic acid was used in two very different ways: as an oral treatment, given regularly over several days (Petruson 1974; White 1988) and as a 'one‐off' topical intervention at the patients' initial presentation (Tibbelin 1995). Whilst the intervention may be very similar in terms of its composition and cellular mechanism of action, it is reasonable to assume that, overall, the way in which it is working as an agent to reduce bleeding is different. Therefore the subgroup analysis is especially pertinent.

Forest plot of comparison: 1 Tranexamic acid versus placebo plus usual care or usual care alone, outcome: 1.1 Control of epistaxis: episodes of re‐bleeding over 10 days.

Analysis

Comparison 1 Tranexamic acid versus placebo plus usual care or usual care alone, Outcome 1 Control of epistaxis: episodes of re‐bleeding over 10 days.

When we compared the effects of oral tranexamic acid (RR 0.73, 95% CI 0.55 to 0.96; two studies, 157 participants, Analysis 1.1.1) (Petruson 1974; White 1988) and the effects of topical tranexamic acid (RR 0.66, 95% CI 0.41 to 1.05; single study, 68 participants, Analysis 1.1.2) (Tibbelin 1995), both showed an effect similar to the overall result, although for the topical tranexamic acid group the 95% confidence interval included unity. We rated the quality of the evidence provided by this single study as low, and it is therefore uncertain whether or not topical tranexamic acid is effective in stopping bleeding in the period 10 days after a single application. A formal test of subgroup differences did not indicate any difference between the treatment effects of the two different methods of administration.

We were unable to carry out any of our other planned subgroup analyses: patient age (children versus adults), setting (inpatient versus outpatient) or site of nosebleed (anterior versus posterior).

Significant adverse effects: seizures, thromboembolic events

No study specifically sought to identify and report these particular adverse events, although all recorded 'adverse effects' in a general way. No significant adverse effects were reported in any study.

Secondary outcomes

Control of epistaxis: time to stop initial bleeding (as measured by the proportion of patients whose bleeding is controlled within 30 minutes)

We also measured control of epistaxis in terms of the proportion of participants who had stopped bleeding within a specified period of up to 30 minutes of treatment. Only Tibbelin 1995 provided data for this comparison. The study found no evidence of a difference between the topical tranexamic acid and control groups at 30 minutes (RR 0.79, 95% CI 0.56 to 1.11; single study; 68 participants) (Analysis 1.2). The quality of this evidence was low.

Analysis

Comparison 1 Tranexamic acid versus placebo plus usual care or usual care alone, Outcome 2 Control of epistaxis: time to stop initial bleeding (proportion with bleeding controlled within 30 minutes).

Severity of re‐bleeding: as measured by a) the proportion of patients requiring any further intervention (e.g. repacking, surgery, embolisation) within 10 days and b) the proportion of patients requiring blood transfusion within 10 days

No studies reported the proportion of patients requiring any further intervention.

When comparing oral tranexamic acid with placebo White 1988 found no evidence of a difference in the proportion of patients requiring blood transfusion: 5/45 versus 6/44 (RR 0.81, 95% CI 0.27 to 2.48; 89 participants) (Analysis 1.3) (low‐quality evidence).

Analysis

Comparison 1 Tranexamic acid versus placebo plus usual care or usual care alone, Outcome 3 Severity of re‐bleeding: proportion of patients requiring blood transfusion within 10 days.

Length of hospital stay in days

Petruson 1974 reported a significantly shorter stay in the oral tranexamic acid group (mean difference (MD) ‐1.60 days, 95% CI ‐2.49 to ‐0.71; 68 participants). White 1988 found no difference (MD 0.40 days, 95% CI ‐0.84 to 1.64; 89 participants) (Analysis 1.4). When we attempted to combine the data from these studies, the heterogeneity was high (I2= 85%).

Analysis

Comparison 1 Tranexamic acid versus placebo plus usual care or usual care alone, Outcome 4 Length of hospital stay.

Other adverse effects

All five studies recorded 'adverse effects' in a general way. Petruson 1974 identified no "symptoms which might be taken for general side effects". Two patients (one in each group) experienced small synechiae, likely to be related to the nasal packing rather than the intervention. Three patients in each group in the Tibbelin 1995 study reported that the gel had a 'bad taste' but no serious adverse event was recorded. White 1988 reported some complaints of mild nausea and diarrhoea, but these were experienced equally each group. One patient did develop a "a superficial thrombophlebitis of both legs following discharge from hospital”, however it is not reported in which group this occurred.

Tranexamic acid versus other haemostatic agents

See Table 2.

Atabaki 2017, Zahed 2013 and Zahed 2018 evaluated this comparison and data were only available for two outcomes.

Control of epistaxis: time to stop initial bleeding (as measured by the proportion of patients whose bleeding is controlled within 30 minutes)

When we pooled the data from the three studies the proportion of patients whose bleeding stopped within 10 minutes was significantly higher in the topical tranexamic acid group (70% versus 30%; RR 2.35, 95% CI 1.90 to 2.92, three studies, 460 participants) (Analysis 2.1; Figure 5). There was no heterogeneity in the pooled analysis (I2 = 0%).

Forest plot of comparison: 2 Tranexamic acid versus other haemostatic agent, outcome: 2.1 Control of epistaxis: time to stop initial bleeding (proportion with bleeding controlled within 10 minutes).

Analysis

Comparison 2 Tranexamic acid versus other haemostatic agent, Outcome 1 Control of epistaxis: time to stop initial bleeding (proportion with bleeding controlled within 10 minutes).

Adverse events

There was no difference between the groups in complications (nausea/vomiting and intolerance) and no serious adverse event was observed (Zahed 2013; Zahed 2018). Atabaki 2017 did not report adverse events.

No data were available for any of our other outcomes.

Discussion

Summary of main results

See Table 1.

The primary outcome measure for this review was the effectiveness of tranexamic acid in the control of epistaxis as measured by the proportion of patients who had an episode of re‐bleeding within the first 10 days of treatment.

We found moderate‐quality evidence that tranexamic acid probably reduces the risk of re‐bleeding when compared to placebo or no treatment. Tranexamic acid was used in two very different ways in the three studies on which this finding is based: as an oral treatment, given regularly over several days (Petruson 1974; White 1988) and as a 'one‐off' topical intervention at the patients' initial presentation (Tibbelin 1995). Subgroup analysis to compare the effects of oral and topical tranexamic acid showed a similar effect to the overall result, although for the topical tranexamic acid group the 95% confidence interval included unity. The quality of evidence provided by this single study is low therefore it is uncertain whether or not topical tranexamic acid is effective in stopping bleeding in the period 10 days after a single application.

We also measured control of epistaxis in terms of the proportion of participants who had stopped bleeding within 30 minutes of treatment. For the comparison of tranexamic acid versus placebo we found low‐quality evidence from a single study and determined that tranexamic acid may lead to little or no difference (Tibbelin 1995). This outcome was also addressed in three studies comparing tranexamic acid versus another haemostatic agent (epinephrine/lidocaine combination or phenylephrine) and moderate‐quality evidence indicates that tranexamic acid probably increases the chance of bleeding stopping in the first 10 minutes after application (Atabaki 2017; Zahed 2013; Zahed 2018).

No study specifically sought to identify and report seizures or thromboembolic events (significant adverse effects).

No studies reported the proportion of patients requiring any further intervention (e.g. repacking, surgery, embolisation) within 10 days.

Blood transfusion requirements were only recorded in one study with no significant differences between the oral tranexamic acid and control groups (White 1988).

One study reported a significantly shorter length of hospital stay in the oral tranexamic acid group (Petruson 1974), while another found no difference (White 1988).

Five included studies mentioned recording adverse effects but none identified any significant differences between groups.

Overall completeness and applicability of evidence

Whilst the limited amount of evidence identified in this review is applicable to the review question, the paucity of it means that the evidence is far from complete. The identified studies also highlight a clear difference between the two ways in which tranexamic acid has been used: as a short‐term topical treatment and a longer‐term oral one. In neither case is there a clear and unequivocal answer about the merits of tranexamic acid.

The adult participants recruited to the studies are representative of those seen in day‐to‐day practice.

The definition of 'usual care' encompasses most of the standard treatments used, but such 'usual care' varies around the world and is in a continuous state of evolution as new types of packs and packing materials are introduced.

Quality of the evidence

The body of evidence included in this review (six studies with a total of 692 participants) is insufficient to allow robust conclusions to be drawn. The quality of evidence for the outcomes assessed was moderate or low; we downgraded the evidence because of imprecision and study limitations (risk of bias).

Potential biases in the review process

Three of the most recent included studies took place in Iran (Atabaki 2017; Zahed 2013; Zahed 2018). Two were conducted by the same research group (Zahed 2013; Zahed 2018); however, patient numbers, characteristics and recruitment periods in the three studies were clearly different.

The search for relevant studies for this review encompassed all the main databases as outlined in the Search methods for identification of studies section. The search terms used should have identified all randomised controlled trials comparing the use of tranexamic acid in epistaxis to either placebo or 'usual care'. We do not believe that the methodology of the review is likely to have introduced any bias into the review process.

Agreements and disagreements with other studies or reviews

A number of systematic reviews have evaluated the use of tranexamic acid in patients undergoing nasal surgery but we are only aware of two other systematic reviews of its use in patients with epistaxis (Kamhieh 2016; Williams 2017). Kamhieh 2016 does not appear to have pre‐specified the outcome measures of interest, does not include the studies Atabaki 2017, Petruson 1974 and Zahed 2018, and provides a narrative description of the studies and the results. The authors do not seem to have noted the additional differences (other than the topical treatment applied at the outset) between the two arms of the Zahed 2013 study, and the implications of this. Williams 2017 identified four of the same randomised controlled trials that are included in our review and drew comparable conclusions, noting the small numbers of participants and study limitations and proposing the need for further studies. A recent UK consensus document reflects these findings and does not recommend the use of tranexamic acid in the management of epistaxis, beyond its use as defined in national guidelines for major haemorrhage (BRS 2017).

Authors' conclusions

Implications for practice

Tranexamic acid probably reduces the risk of patients with epistaxis having further bleeding episodes. We found moderate‐quality evidence that there may be a reduction in the risk of re‐bleeding from 67% to 47% with the use oral or topical tranexamic acid in addition to usual care in adult patients with epistaxis, compared to placebo with usual care. Low‐quality evidence suggests that topical tranexamic acid may make little or no difference to the control of bleeding in the first 30 minutes after its application. In three studies, the proportion of patients whose bleeding stopped within 10 minutes was significantly higher in the topical tranexamic acid group compared with other haemostatic agents (topical epinephrine/lidocaine combination or phenylephrine) (moderate‐quality evidence). No significant adverse effects of treatment were reported in the included studies.

Implications for research

This review has found evidence to suggest that there is probably a reduction in re‐bleeding episodes when using oral or topical tranexamic acid in addition to 'usual care'. However, although the latest randomised controlled trials were published between 2013 and 2018, all the other trials were conducted before 1995. Further randomised controlled trials, on similar cohorts of patients, attending hospital with epistaxis, using modern nasal cauterisation and packing techniques (for example, techniques including nasal endoscopy and more invasive approaches such as endoscopic sphenopalatine artery ligation), and investigating the use of tranexamic acid in light of these advances, compared to 'usual care' alone, would be informative and improve our certainty about any treatment effect. Trials may specifically look at the effectiveness and safety of topical tranexamic acid. It has been suggested that the use of this formulation may be associated with fewer adverse events (due to an absence of systemic absorption).

History

Protocol first published: Issue 3, 2003
Review first published: Issue 12, 2018

Acknowledgements

We would like to thank Dr Ali Yadollahpour for translating a study from Farsi for this review.

This project was supported by the National Institute for Health Research, via Cochrane Infrastructure, Cochrane Programme Grant or Cochrane Incentive funding to Cochrane ENT. The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the Systematic Reviews Programme, NIHR, NHS or the Department of Health.

Appendices

Appendix 1. Search strategies

Appendix 2. Data extraction

Methods

• (Double‐/single‐/non‐) blinded, (cluster‐/cross‐over/ parallel ‐group/within‐patient/quasi‐/non‐) randomised controlled trial with x duration of treatment and y duration of follow‐up

Participants

• Setting

• Sample size:

  • Number randomised:

  • Number completed:

• Participant (baseline) characteristics:

  • Age

  • Gender

  • Other characteristics (risk factors)

• Inclusion criteria

• Exclusion criteria

Interventions

• Intervention group (method of delivery and dosage of tranexamic acid)

• Comparator group (placebo; type of ‘usual care’ provided)

• Use of additional interventions

Outcomes

• Primary and secondary outcomes and time points

Funding sources

Declarations of interest

Notes

Data and analyses

Comparison 1

Tranexamic acid versus placebo plus usual care or usual care alone

Comparison 2

Tranexamic acid versus other haemostatic agent

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Characteristics of excluded studies [ordered by study ID]

Characteristics of ongoing studies [ordered by study ID]

Differences between protocol and review

• The title has been changed from ‘Antifibrinolytic agent tranexamic acid for nasal haemorrhage (epistaxis)’ to ‘Tranexamic acid for patients with nasal haemorrhage (epistaxis)’ in line with another Cochrane ENT protocol on this drug (Ravesloot 2017).

• The Background has been redrafted and updated.

• We have reworded the Objectives to show the review comparisons clearly.

• We have defined Types of studies more fully, including the minimum follow‐up period.

• Types of participants has been set out more concisely and brought into line with the other Cochrane ENT tranexamic acid protocol (Ravesloot 2017).

• Types of interventions has been amended to define the review comparisons more clearly, in line with other recent ENT reviews.

• In Types of outcome measures, the primary outcome 'Effectiveness in control of epistaxis. Frequency and/or severity of rebleeds with a measure of blood loss' has been split into two separate outcomes: 'Control of epistaxis: re‐bleeding, as measured by the proportion of patients re‐bleeding within a period of up to 10 days (primary outcome)' and 'Control of epistaxis: time to stop initial bleeding (as measured by the proportion of patients whose bleeding is controlled within 30 minutes)' (secondary outcome). 'Adverse and/or side effects' has been similarly split into two outcomes: 'Significant adverse effects (seizures, thromboembolic events)' (primary) and 'Other adverse effects', to bring the review in line with other recent ENT reviews and the related tranexamic acid protocol (Ravesloot 2017). We have clarified the wording of other outcomes.

Contributions of authors

Jonathan Joseph: screened search results and selected studies, carried out 'Risk of bias' assessment and statistical analysis, wrote the text of the review.
Pablo Martinez‐Devesa: screened search results and selected studies, carried out 'Risk of bias' assessment and statistical analysis, wrote the text of the review.
Jenny Bellorini: screened search results and selected studies (October 2018), carried out 'Risk of bias' assessment and statistical analysis, carried out GRADE assessment and edited the text of the review.
Martin J Burton: screened search results and selected studies (October 2018), carried out 'Risk of bias' assessment and statistical analysis, carried out GRADE assessment and wrote/edited the text of the review.

Sources of support

Internal sources

• No sources of support supplied

External sources

• National Institute for Health Research, UK.

  Infrastructure funding for Cochrane ENT

Declarations of interest

Jonathan Joseph: none known.
Pablo Martinez‐Devesa: none known.
Jenny Bellorini: Jenny Bellorini is Managing Editor of Cochrane ENT, but had no role in the editorial sign‐off process for this review.
Martin J Burton: Professor Martin Burton is joint Co‐ordinating Editor of Cochrane ENT, but had no role in the editorial sign‐off process for this review.

References

References to studies included in this review

Atabaki 2017 {published data only}

• Atabaki P, Samarei R, Aribi MS, Soheili A, Mehryar HR. A comparative study on the effect of topical phenylephrine with topical tranexamic acid in management of epistaxis. Journal of the Urmia Nursing and Midwifery Faculty 2017;15(7):488‐96. [Google Scholar]

Petruson 1974 {published data only}

• Petruson B. A double blind study to evaluate the effect on epistaxis with oral administration of the antifibrinolytic drug tranexamic acid (Cyklokapron®). Acta Otolaryngologica 1974;77(Suppl 317):57‐61. [Google Scholar]

Tibbelin 1995 {published data only}

• Tibbelin A, Aust R, Bende M, Holgersson M, Petruson B, Rundcrantz H, et al. Effect of local tranexamic acid gel in the treatment of epistaxis. ORL; Journal for Oto‐rhino‐laryngology and its Related Specialties 1995;57(4):207‐9. [PubMed] [Google Scholar]

White 1988 {published data only}

• White A, O'Reilly BF. Oral tranexamic acid in the management of epistaxis. Clinical Otolaryngology and Allied Sciences 1998;13(1):11‐6. [PubMed] [Google Scholar]

Zahed 2013 {published data only}

• Saeedi M, Arasi S. Effectiveness of local tranexamic acid on epistaxis control. Iranian Registry of Clinical Trials April 2012.
• Zahed R, Moharamzadeh P, Alizadeh Arasi S, Ghasemi A, Saeedi M. A new and rapid method for epistaxis treatment using injectable form of tranexamic acid topically: a randomized controlled trial. American Journal of Emergency Medicine 2013;31(9):1389‐92. [PubMed] [Google Scholar]

Zahed 2018 {published data only}

• Zahed R, Mousavi Jazayeri M H, Naderi A, Naderpour Z, Saeedi M. Topical tranexamic acid compared with anterior nasal packing for treatment of epistaxis in patients taking antiplatelet drugs: randomized controlled trial. Academic Emergency Medicine 2018;25(3):261‐6. [PubMed] [Google Scholar]

References to studies excluded from this review

Alimian 2011 {published data only}

• Alimian M, Mohseni M. The effect of intravenous tranexamic acid on blood loss and surgical field quality during endoscopic sinus surgery: a placebo‐controlled clinical trial. Journal of Clinical Anesthesia 2011;23:611‐5. [PubMed] [Google Scholar]

ATERO 2014 {published data only}

• Gaillard S, Dupuis‐Girod S, Boutitie F, Rivière S, Morinière S, Hatron PY, et al. ATERO Study Group. Tranexamic acid for epistaxis in hereditary hemorrhagic telangiectasia patients: a European cross‐over controlled trial in a rare disease. Journal of Thrombosis and Haemostasis 2014;12(9):1494‐502. [PubMed] [Google Scholar]

Athanasiadis 2007 {published data only}

• Athanasiadis T, Beule A, Wormald P. Effects of topical antifibrinolytics in endoscopic sinus surgery: a pilot randomised controlled trial. American Journal of Rhinology 2007;21(6):737‐42. [PubMed] [Google Scholar]

Baradaranfar 2017 {published data only}

• Baradaranfar MH, Dadgarnia MH, Mahmoudi H, Behniafard N, Atighechi S, Zand V, et al. The effect of topical tranexamic acid on bleeding reduction during functional endoscopic sinus surgery. Iranian Journal of Otorhinolaryngology 2017;29(91):69‐74. [PMC free article] [PubMed] [Google Scholar]

Beikaei 2015 {published data only}

• Beikaei M, Ghazipour A, Derakhshande V, Saki N, Nikakhlagh S. Evaluating the effect of intravenous tranexamic acid on intraoperative bleeding during elective rhinoplasty surgery. Biomedical and Pharmacology Journal 2015;8SE:753‐9. [Google Scholar]

Chhapola 2011 {published data only}

• Chhapola S, Matta I. Short‐term use of tranexamic acid to reduce blood loss in endoscopic nasal surgeries. Clinical Rhinology 2011;4:79‐81. [Google Scholar]

Eftekharian 2016 {published data only}

• Eftekharian HR, Rajabzadeh Z. The efficacy of preoperative oral tranexamic acid on intraoperative bleeding during rhinoplasty. Journal of Craniofacial Surgery 2016;27(1):97‐100. [PubMed] [Google Scholar]

Fernandez‐L 2007 {published data only}

• Fernandez‐L A, Garrido‐Martin EM, Sanz‐Rodriguez F, Ramirez J‐R, Morales‐Angulo C, Zarrabeita R, et al. Therapeutic action of tranexamic acid in hereditary haemorrhagic telangiectasia (HHT): regulation of ALK‐a/endoglin pathway in endothelial cells. Thrombosis Haemostasis 2007;97(2):254‐62. [PubMed] [Google Scholar]

Geisthoff 2014 {published data only}

• Geisthoff UW, Seyfert U, Konig J, Kubler M, Bieg B, Plinkert PK. Hereditary hemorrhagic telangiectasia and tranexamic acid: a double‐blind placebo‐controlled crossover study. European Archive of Otorhinolaryngology 2004;261(1):41. [Google Scholar]

Ghavimi 2017 {published data only}

• Ghavimi MA, Taheri Talesh K, Ghoreishizadeh A, Chavoshzadeh MA, Zarandi A. Efficacy of tranexamic acid on side effects of rhinoplasty: a randomized double‐blind study. Journal of cranio‐maxillofacial surgery. Journal of Cranio‐maxillo‐facial Surgery 2017;45(6):897‐902. [PubMed] [Google Scholar]

Gossage 2015 {published data only}

• Gossage JR, Whitehead KJ, Sautter NB, McWilliams JP, Chakinala M, Merlo C, et al. North American study of epistaxis in HHT (nose trial). Angiogenesis 2015;18(4):541‐2. [Google Scholar]

IRCT2014122520434N1 {published data only}

• IRCT2014122520434N1. Comparison of the effect of tranexamic acid and dexmedetomidine on intra‐operative bleeding in patients candidate for rhinoplasty. https://en.irct.ir/trial/18125 (first received 16 July 2015). [IRCT2014122520434N1]

IRCT201509088872N9 {published data only}

• IRCT201509088872N9. Effectiveness of local tranexamic acid on epistaxis control in patients with antiplateletes therapy (ASA, Plavix). http://www.irct.ir/ (first received 25 September 2015).

Jabalameli 2006 {published data only}

• Jabalameli M, Zakeri K. Evaluation of topical tranexamic acid on intraoperative bleeding in endoscopic sinus surgery. Iranian Journal of Medical Science 2006;31(4):221‐3. [Google Scholar]

Keiani Motlagh 2003 {published data only}

• Keiani Motlagh K, Loeb I, Legrand W, Daelemans P, Reck J. Prevention of postoperative bleeding in patients taking oral anticoagulants. Effects of tranexamic acid [Prevention des saignements postoperatoires chez des patients sous anticoagulants oraux; effets de l'acide tranexamique]. Revue de Stomatologie et de Chirurgie Maxillo‐Faciale 2003;104(2):77‐9. [PubMed] [Google Scholar]

Kulkarni 2018 {published data only}

• Kulkarni VR. A comparative study of tranexamic acid and ehamsylate for control of blood loss in functional endoscopic sinus surgery. Paripex ‐ Indian Journal of Research 2018;7(6):58‐63. [Google Scholar]

Mehdizadeh 2018 {published data only}

• Mehdizadeh M, Ghassemi A, Khakzad M, Mir M, Nekoohesh L, Moghadamnia A, et al. Comparison of the effect of dexamethasone and tranexamic acid, separately or in combination on post‐rhinoplasty edema and ecchymosis. Aesthetic Plastic Surgery 2018;42(1):246‐52. [PubMed] [Google Scholar]

NOSE 2012 {published data only}

• Gossage J. North American study of epistaxis in HHT. http://clinicaltrials.gov/ct2/show/NCT01408030 (accessed 6 August 2012). [NCT01408030]

Sabba 2001 {published data only}

• Sabba C, Gallitelli M, Palasciano G. Efficacy of unusually high doses of tranexamic acid for the treatment of epistaxis in hereditary hemorrhagic telangiectasia. New England Journal of Medicine 2001;345(12):926. [PubMed] [Google Scholar]

Whitehead 2016 {published data only}

• Whitehead KJ, Sautter NB, McWilliams JP, Chakinala MM, Merlo CA, Johnson MH, et al. Effect of topical intranasal therapy on epistaxis frequency in patients with hereditary hemorrhagic telangiectasia: a randomized clinical trial. JAMA 2016;316(9):943‐51. [PubMed] [Google Scholar]

Yaniv 2006 {published data only}

• Yaniv E, Shvero J, Hadar T. Hemostatic effect of tranexamic acid in elective surgery. American Journal of Rhinology 2006;20(2):227‐9. [PubMed] [Google Scholar]

References to ongoing studies

ISRCTN34153772 {published data only}

• ISRCTN34153772. Novel use of TXA to reduce the need for nasal packing in epistaxis. http://www.isrctn.com/ISRCTN34153772 (first received 1 June 2017).

NCT02930941 {published data only}

• NCT02930941. Topical intranasal tranexamic acid for epistaxis in the emergency department. https://clinicaltrials.gov/ct2/show/NCT02930941 (first received 12 October 2016).

NCT03360045 {published data only}

• NCT03360045. The evaluation of effectiveness of nasal compression with tranexamic acid compared to simple nasal compression and Merocel packing. https://clinicaltrials.gov/ct2/show/NCT03360045 (first received 2 December 2017). [NCT03360045]

Additional references

BRS 2017

• INTEGRATE (The National ENT Trainee Research Network), National ENT Trainee Research Network. The British Rhinological Society multidisciplinary consensus recommendations on the hospital management of epistaxis. Journal of Laryngology and Otology 2017;131(12):1142‐56. [PubMed] [Google Scholar]

Dunn 1999

• Dunn CJ, Goa KL. Tranexamic acid: a review of its use in surgery and other indications. Drugs 1999;57(6):1005‐32. [PubMed] [Google Scholar]

Handbook 2011

• Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from www.cochrane‐handbook.org.

Kamhieh 2016

• Kamhieh Y, Fox H. Tranexamic acid in epistaxis: a systematic review. Clinical Otolaryngology 2016;41(6):771‐6. [PubMed] [Google Scholar]

Ker 2012

• Ker K, Edwards P, Perel P, Shakur H, Roberts I. Effect of tranexamic acid on surgical bleeding: systematic review and cumulative meta‐analysis. BMJ 2012;344:e3054. [PMC free article] [PubMed] [Google Scholar]

McGarry 2008

• McGarry GW. Epistaxis. In: Gleeson M, Browning GG, Burton MJ, Clarke R, Hibbert J, Jones NS, et al. editor(s). Scott‐Brown's Otolaryngology, Head and Neck Surgery. 7th Edition. Vol. 2, Chapter 126, London: Edward Arnold, 2008:1596‐606. [Google Scholar]

Nishihara 2015

• Nishihara S, Hamada M. Does tranexamic acid alter the risk of thromboembolism after total hip arthroplasty in the absence of routine chemical thromboprophylaxis?. The Bone & Joint Journal 2015;97‐B(4):458‐62. [PubMed] [Google Scholar]

Ravesloot 2017

• Ravesloot MJL, Lourijsen E, Avdeeva K, Pundir V, Fokkens W. Tranexamic acid for the reduction of bleeding during functional endoscopic sinus surgery. Cochrane Database of Systematic Reviews 2017, Issue 11. [DOI: 10.1002/14651858.CD012843] [CrossRef] [Google Scholar]

RevMan 2014 [Computer program]

• The Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan). Version 5.3. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014.

Robb 2014

• Robb PJ. Tranexamic acid ‐ a useful drug in ENT surgery?. Journal of Laryngology and Otology 2014;128:574‐9. [PubMed] [Google Scholar]

Roberts 2013

• Roberts I, Shakur H, Coats T, Hunt B, Balogun E, Barnetson L, et al. The CRASH‐2 trial: a randomised controlled trial and economic evaluation of the effects of tranexamic acid on death, vascular occlusive events and transfusion requirement in bleeding trauma patients. Health Technology Assessment (Winchester, England) 2013;17(10):1‐79. [PMC free article] [PubMed] [Google Scholar]

Shaheen 1967

• Shaheen OH. Epistaxis in the Middle Aged and Elderly [Thesis]. London: University of London, 1967. [Google Scholar]

Sharma 2014

• Sharma V, Katznelson R, Jerath A, Garrido‐Olivares L, Carroll J, Rao V, et al. The association between tranexamic acid and convulsive seizures after cardiac surgery: a multivariate analysis in 11 529 patients. Anaesthesia 2014;69(2):124‐30. [PubMed] [Google Scholar]

UK Epistaxis Audit 2017

• INTEGRATE (The National ENT Trainee Research Network), National ENT Trainee Research Network. Epistaxis 2016: national audit of management. Journal of Laryngology and Otology 2017;131(12):1131‐41. [DOI: 10.1017/S002221511700202X; PUBMED: 29280693] [PubMed] [CrossRef] [Google Scholar]

Williams 2017

• Williams A, Biffen A, Pilkington N, Arrick L, Williams RJ, Smith ME, et al. Haematological factors in the management of adult epistaxis: systematic review. Journal of Laryngology and Otology 2017;131(12):1093‐107. [DOI: 10.1017/S0022215117002067; PUBMED: 29280698] [PubMed] [CrossRef] [Google Scholar]

Articles from The Cochrane Database of Systematic Reviews are provided here courtesy of Wiley