In late 2019, a novel coronavirus named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spread to become a global pandemic, which is now termed coronavirus disease 2019 (COVID-19).1 2 3 4 5 At the time of writing, more than 12 million people worldwide have been infected, with a crude mortality ratio of approximately 5%.6

In both of the erstwhile SARS and MERS epidemics, there were reported cases of transmission to healthcare workers.7 8 Many physicians and allied health staff have contracted SARS-CoV, some of whom have died of the infection. Endotracheal intubation is considered as one of the highest risk procedures in the context of disease transmission.9 10 During the SARS outbreak, 21% of cases worldwide were among healthcare workers.7 Caputo et al11 estimated that 9% of the interviewed healthcare workers who performed an intubation ended up contracting SARS. Intubation is a high-risk aerosol-generating procedure, and the odds ratio to contract SARS for healthcare workers performing tracheal intubation is 6.6 compared with those who did not perform this procedure.12

In the setting of COVID-19, SARS-CoV-2 is a highly contagious, novel virus with many still-unknown characteristics. Recent literature on healthcare-related transmission of COVID-19 revealed that up to 20% of infections among healthcare workers were in those caring for patients with COVID-19, and that hospital-related transmission was suspected in 41% of patients.13 14 Amidst the COVID-19 pandemic, it is essential for paediatricians and intensivists to be equipped with clear intubation guidelines.15 16

In January 2020, we promptly developed a novel approach and protocol in preparation for the management of critically ill paediatric patients with COVID-19. Although several adult intensive care unit guidelines exist in the public domain, there are currently no standard or unified paediatric-specific guidelines in the literature.17 18 19 Adult intubation guidelines cannot be applied directly to paediatric practice, as children can present at wide age and weight ranges. In addition, there is a difference in physiology: children have higher rates of oxygen consumption and smaller functional residual capacity. Therefore, we tested different circuit configurations for endotracheal intubation and propose variations for the following hypothetical scenarios: (1) elective intubation for respiratory failure; and (2) emergency intubation during cardiopulmonary resuscitation.

Any paediatric patient who is suspected or confirmed to be infected with SARS-CoV-2 should be cared for in a single airborne infection isolation room with negative pressure relative to the atmosphere, and the outgoing air should be passed through a high-efficiency particulate air (HEPA) filter.20 All healthcare workers should wear personal protective equipment appropriate for airborne pathogens when caring for the patient, including a fit-tested N95 respirator, goggles, a face shield, a fluid-resistant non-sterile gown, and a pair of clean non-sterile gloves. Staff should receive training and practice donning and doffing of personal protective equipment according to standard protocols. Although SARS-CoV-2 is supposedly transmitted by droplets instead of being airborne, healthcare workers are advised to adopt airborne precautions in face of this rapidly transmitting and novel pathogen with uncertain viral characteristics.21 Ideally, the most experienced healthcare professional should perform the intubation with two other assistants. In the event that cricoid pressure is not given, it may be possible to have one less assistant to reduce potential exposure.

Age- and weight-appropriate airway equipment should ideally be prepared ahead of time and made readily available by the bedside (Fig 1). Single-use or dedicated equipment must be used. A cuffed endotracheal tube (ETT) is preferable. As bag-mask ventilation prior to intubation can generate aerosols, the practice should be avoided as far as practicable.22 We suggest setting up a closed-circuit bagging system that can be connected to a bedside scavenger to reduce the risk of viral transmission by employing unidirectional gas flow directed away from the patient’s side (Fig 2). A similar scavenger system has been used in ventilator and neonatal resuscitation circuits for suspected cases of SARS and COVID-19.23 24 This circuit can also be configured with a modified Ayre’s T-piece and either a 0.5-L bag for children weighing ≤20 kg or a 2-L bag for children weighing >20 kg. Furthermore, a bacterial and viral filter (preferably a HEPA filter) should be added to the system to reduce the risk of spreading the airborne pathogens to the breathing system and the ambient air.25 However, the choices for HEPA filters for paediatric patients are limited, as most HEPA filters on the market are designed for adult patients (ie, they require a tidal volume of at least 200-300 mL). A filter should always be placed between the face mask and the connecting tube of the T-piece circuit to filter all expired gas, and we further recommend placing an additional filter between the reservoir breathing bag and the suction tube to filter all scavenging gas, which would increase the filtration efficacy and reduce the risk of viral transmission to the ambient air. The manufacturers’ published filter specifications could be different from the actual filtration performance, as filtration efficiency can depend on temperature and humidity.26 27 Theoretically, based on the size of SARS-CoV-2 (0.06-0.2 μm), it can be removed by HEPA filters. However, manufacturers and suppliers might not have validated these filters for removal of the SARS-CoV-2 virus; therefore, we suggest that extra precautions are necessary.27 28 29 Oral suction during intubation is unavoidably an open suction procedure, and it is also regarded as an aerosol-generating procedure. Thus, it should be performed only after the administration of adequate muscle relaxant and only when there is genuine need (eg, when secretions obscure the larynx).

If available, video laryngoscopy should be used to increase the chance of successful intubation on the first pass, with the screen ideally located at the operator’s eye level. Another advantage of video laryngoscopy is that a longer distance can be maintained between the intubation field and the operator, reducing the risk of transmission. A ventilator circuit with heat and moisture exchanger filter (HMEF), closed suction system, and mainstream end-tidal CO₂ monitor (preferably disposable airway adaptor) should be prepared and connected to the ETT in one piece, which minimises any further disconnection of the circuit after successful intubation (Fig 1). We recommend the use of HMEF for paediatric patients as water bath humidifiers may increase the risk of infection dissemination. Although there might be increased risk of sputum retention with the use of HMEF, especially in small children, close monitoring and regular close suction can reduce this risk. Previous studies showed that water bath humidifiers can produce aerosols containing bacteria, and they could also potentially carry SARS-CoV-2.30 31

Patients should be pre-oxygenated by spontaneous breathing using the modified Ayre’s T-piece system with a tight-fitting face mask using the two-hand technique of bag-mask ventilation. Oxygen flow should be started at about 3 times the estimated minute volume, and the flow of oxygen and suction should be adjusted to keep the reservoir bag partially inflated. The operator’s attention should be directed to the reservoir bag’s inflation, as this provides an indication of whether a tight seal, adequate positive end expiratory pressure (PEEP), and an intact circuit are maintained. Manual bagging should be avoided as much as possible. Rapid sequence induction should be used to optimise the chance of success on the first attempt and to minimise any coughing or gag reflex during intubation.22 After 5 minutes of pre-oxygenation, apart from other sedatives for rapid sequence induction, an adequate muscle relaxant (eg, rocuronium 1.2 mg/kg) can be administered to the patient.32 Cricoid pressure can then be applied. After adequate muscle paralysis, the patient can be intubated with a cuffed ETT. Once the ETT has been passed though the cords to the proper depth, the cuff should be immediately inflated. After successful intubation and cuff inflation, the clinician must ensure that there is no leakage around the cuff. Finally, the ETT can be connected to the prepared ventilator, and ventilation should be started immediately to avoid bagging as far as possible.

If the intubation attempt is not successful, or if severe desaturation persists despite pre-oxygenation without bagging, a laryngeal mask airway can be inserted. Manual ventilation should be continued using the laryngeal mask airway and the modified T-piece system with adequate oxygen flow until help arrives. For subsequent intubation attempts, clinicians can consider using the ETT stylet or Bougie techniques.

Generally, the patient should be intubated as soon as possible because bagging with a face mask risks generating aerosols.10 Other than using the two-hand technique for tight mask fitting, leaking can also be minimised by using laryngeal mask airway during cardiopulmonary resuscitation, particularly if a difficult airway is anticipated or if an experienced operator is not available.

After successful intubation, clinicians should inflate the cuff and connect the ETT to the ventilator immediately while adjusting the ventilator settings in accordance with the patient’s age and weight. It is also important to make sure that end-tidal CO₂ monitoring is available, as it is crucial to ensure correct placement of ETT and the return of spontaneous circulation.

All of the equipment mentioned in this proposed intubation approach should be commercially available in most hospitals worldwide that provide paediatric services. Our proposed system provides multiple key functions, including respiratory pattern monitoring, spontaneous respiration oxygenation, apnoeic oxygenation, manual ventilation, and scavenging.

Our approach has several advantages. It is a closed breathing system, provided that the face mask has a tight seal and clinicians use the two-hand technique, which can minimise the risk of contaminated secretions leaking out. It allows contaminated gases to flow in the direction of the scavenging system, and manual ventilation is possible by squeezing the reservoir bag if necessary. Positive end expiratory pressure can be maintained by partially occluding the tail of the reservoir bag. This is especially important in the paediatric population, as children have a low functional residual capacity, and the addition of PEEP during pre-oxygenation can potentially minimise or prevent functional residual capacity reduction, airway closure, and subsequent atelectasis. The patient’s breathing pattern can be monitored by the movement of the reservoir bag. Finally, a manometer can be connected to the circuit if the operator prefers to monitor pressure in the circuit.33

Users should be aware of some potential limitations of our approach. First, the use of an Ayre’s T-piece with Jackson-Rees modification requires training. Second, adequate pre-oxygenation is required, as there is no oxygen supply during the period between removing the face mask from the patient and successful endotracheal intubation. Third, CO₂ rebreathing may occur if the patient’s minute volume is high. This can be mitigated by using a T-piece with fresh gas flow configured to 2 to 3 times the minute volume. For example, if the fresh oxygen flow is 15 L/min, CO₂ rebreathing may occur if the patient’s minute volume is more than 5 to 7.5 L/min; Fourth, the system cannot be used in areas with no oxygen supply. Finally, the whole system might be under negative pressure during the entire expiratory phase, which may lead to alveolar collapse. The application of a low level of PEEP to the anaesthetic bag would prevent this from occurring. Further, clinical and radiological evaluations are indicated to ensure that there is no significant atelectasis.

With adequate training and practice, our proposed approach to intubation should be safe and effective. We would welcome further research in a laboratory setting to test the system’s integrity and quantitatively measure the reductions in aerosols that it generates.

Amidst the current pandemic, in which much remains unknown about the coronavirus and COVID-19 disease, healthcare workers should take the highest levels of precautions and protection when providing care to patients with suspected SARS-CoV-2 infection and respiratory failure. By sharing the approach we have developed for endotracheal intubation, we aim to raise awareness of the precautions that need to be taken during intubation to reduce the risk of healthcare-related transmission, not only in the current COVID-19 pandemic, but also in future outbreaks of airborne pathogens.34 A review of all published paediatric airway and intubation protocols is now underway to evaluate this important management facet in paediatrics. Paediatricians should be as well prepared as physicians who care for adults.

All authors contributed to the concept or design of the study, acquisition of the data, analysis or interpretation of the data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

As an editor of the journal, KL Hon was not involved in the peer review process. Other authors have no conflicts of interest to disclose.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

1. Zhu N, Zhang D, Wang W, et al. A novel coronavirus from patients with pneumonia in China, 2019. N Engl J Med 2020;382:727-33. Crossref

2. Huang C, Wang Y, Li X, et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet 2020;395:497-506. Crossref

3. Hon KL, Leung KK. Severe acute respiratory symptoms and suspected SARS again 2020. Hong Kong Med J 2020;26:78-9. Crossref

4. World Health Organization. Naming the coronavirus disease (COVID-19) and the virus that causes it. Available from: https://www.who.int/emergencies/diseases/novel-coronavirus-2019/technical-guidance/naming-the-coronavirus-disease-(covid-2019)-and-the-virus-that-causes-it. Accessed 24 Feb 2020.

5. Hon KL, Leung KK, Leung AK, et al. Overview: the history and pediatric perspectives of severe acute respiratory syndromes: novel or just like SARS. Pediatr Pulmonol 2020;55:1584-91. Crossref

6. Coronavirus Resource Center, Johns Hopkins University. COVID-19 dashboard by the Center for Systems Science and Engineering at Johns Hopkins University. 2020. Available from: https://coronavirus.jhu.edu/map.html. Accessed 10 Jul 2020.

7. World Health Organization. Summary of probable SARS cases with onset of illness from 1 November 2002 to 31 July 2003. Available from: https://www.who.int/csr/sars/country/table2004_04_21/en/. Accessed 4 Jan 2020.

8. Al-Tawfiq JA, Auwaerter PG. Healthcare-associated infections: the hallmark of Middle East respiratory syndrome coronavirus with review of the literature. J Hosp Infect 2019;101:20-9. Crossref

9. Fowler RA, Guest CB, Lapinsky SE, et al. Transmission of severe acute respiratory syndrome during intubation and mechanical ventilation. Am J Respir Crit Care Med 2004;169:1198-202. Crossref

10. Chan MT, Chow BK, Lo T, et al. Exhaled air dispersion during bag-mask ventilation and sputum suctioning—implications for infection control. Sci Rep 2018;8:198. Crossref

11. Caputo KM, Byrick R, Chapman MG, Orser BA, Orser BJ. Intubation of SARS patients: infection and perspectives of healthcare workers. Can J Anesth 2006;53:122-9. Crossref

12. Tran K, Cimon K, Severn M, Pessoa-Silva CL, Conly J. Aerosol generating procedures and risk of transmission of acute respiratory infections to healthcare workers: a systematic review. PLoS One 2012;7:e35797. Crossref

13. Remuzzi A, Remuzzi G. COVID-19 and Italy: what next? Lancet 2020;395:1225-8. Crossref

14. Wang D, Hu B, Hu C, et al. Clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus-infected pneumonia in Wuhan, China. JAMA 2020;323:1061-9. Crossref

15. Li Q, Guan X, Wu P, et al. Early transmission dynamics in Wuhan, China, of novel coronavirus-infected pneumonia. N Engl J Med 2020;382:119-207. Crossref

16. Wu JT, Leung K, Leung GM. Nowcasting and forecasting the potential domestic and international spread of the 2019-nCoV outbreak originating in Wuhan, China: a modelling study. Lancet 2020;395:689-97. Crossref

17. Cheung JC, Ho LT, Cheng JV, Cham EY, Lam KN. Staff safety during emergency airway management for COVID-19 in Hong Kong. Lancet Respir Med 2020;8:e19. Crossref

18. Yao W, Wang T, Jiang B, et al. Emergency tracheal intubation in 202 patients with COVID-19 in Wuhan, China: lessons learnt and international expert recommendations. Br J Anaesth 2020;125:e28-37.

19. Brewster DJ, Chrimes N, Do TB, et al. Consensus statement: Safe Airway Society principles of airway management and tracheal intubation specific to the COVID-19 adult patient group. Med J Aust 2020;212:472-81. Crossref

20. American Society of Anesthesiologists. COVID-19 Information for health care professionals. 2020. Available from: https://www.asahq.org/about-asa/governance-and-committees/asa-committees/committee-on-occupational-health/coronavirus. Accessed 6 Feb 2020.

21. World Health Organization. Infection prevention and control during health care when novel coronavirus (nCoV) infection is suspected. Interim guidance. Available from: https://www.who.int/publications/i/item/10665-331495. Accessed 9 Feb 2020. Crossref

22. Wax RS, Christian MD. Practical recommendations for critical care and anesthesiology teams caring for novel coronavirus (2019-nCoV) patients. Can J Anaesth 2020;67:568-76. Crossref

23. Ng PC, So KW, Leung TF, et al. Infection control for SARS in a tertiary neonatal centre. Arch Dis Child Fetal Neonatal Ed 2003;88:F405-9. Crossref

24. Trevisanuto D, Moschino L, Doglioni N, Roehr CC, Gervasi MT, Baraldi E. Neonatal resuscitation where the mother has a suspected or confirmed novel coronavirus (SARS-CoV-2) infection: suggestion for a pragmatic action plan. Neonatology 2020;117:133-40. Crossref

25. Heuer JF, Crozier TA, Howard G, Quintel M. Can breathing circuit filters help prevent the spread of influenza A (H1N1) virus from intubated patients? GMS Hyg Infect Control 2013;8:Doc09.

26. Dellamonica J, Boisseau N, Goubaux B, Raucoules-Aimé M. Comparison of manufacturers’ specifications for 44 types of heat and moisture exchanging filters. Br J Anaesth 2004;93:532-9. Crossref

27. Fisher & Paykel Healthcare. Viral & bacterial filtration efficiency of Fisher & Paykel healthcare filters and F&P EvaquaTM 2 circuits. 2020. Available from: https://www. fphcare.com/us/covid-19/filters-evaqua-circuits-covid- 19/#tested-covid. Accessed 13 Jul 2020.

28. Hamilton Medical. Efficiency of HEPA filters. 2020. Available from: https://www.hamilton-medical.com/fr_CH/E-Learning-and-Education/Knowledge-Base/Knowledge-Base-Detail~2020-03-18~Efficiency-of-HEPA-filters~d5358f88-753e-4644-91c6-5c7b862e941f~.html. Accessed 13 Jul 2020.

29. Smiths Medical. HEPA filter information. 23 Mar 2020. Available from: https://www.smiths-medical.com/company-information/news-and-events/news/2020/march/23/hepa-filter-letter. Accessed 13 Jul 2020.

30. Gilmour IJ, Boyle MJ, Streifel A, McComb RC. The effects of circuit and humidifier type on contamination potential during mechanical ventilation: a laboratory study. Am J Infect Control 1995;23:65-72. Crossref

31. Al Ashry HS, Modrykamien AM. Humidification during mechanical ventilation in the adult patient. Biomed Res Int 2014;2014:715434. Crossref

32. World Health Organization. Clinical management of severe acute respiratory infection when novel coronavirus (nCoV) infection is suspected: interim guidance, 28 January 2020. Available from: https://apps.who.int/iris/handle/10665/330893. Accessed 11 Feb 2020.

33. Trachsel D, Svendsen J, Erb TO, von Ungern-Sternberg BS. Effects of anaesthesia on paediatric lung function. Br J Anaesth 2016;117:151-63. Crossref

34. Hon KL. Just like SARS. Pediatr Pulmonol 2009;44:1048-9. Crossref

The most recent wave of coronavirus disease 2019 (COVID-19) infections in July to August 2020 resulted in more than 100 new cases per day in Hong Kong for a prolonged period. The stress experienced by individual intensive care units (ICUs) in Hong Kong was demonstrated by the need for an unusually large number of patient transfers between units to maximise the available ICU capacity, despite the implementation of surge strategies. Admission decisions to ICUs resulting from the added pressure for ICU beds, as well as social dimensions that were triggered by the COVID-19 outbreak, resulted in an urgent requirement for contingencies to inform admission triage practices in the face of overwhelming ICU demand. Professional bodies have recommended that triage protocols (clinical decision support systems), rather than clinical judgement alone, be used in triage whenever possible,1 and that such protocols be available to assist frontline doctors.1 2 Such protocols should be locally relevant and prepared in advance of the need for implementation.

This document is built on the existing Admission, Discharge, and Triage Guidelines that inform routine clinical practice in Hospital Authority ICUs. The purpose of this COVID-19 Crisis Triage Tool is to provide structured guidance to frontline clinicians to assist with triage decision making should ICU resources become overwhelmed by patients requiring ICU care in Hong Kong.

The Admission, Discharge, and Triage Guidelines for Adult Intensive Care Services for use in day-to-day ICU operations in Hong Kong were recently updated in an internal operations circular in 2018. A brief summary of this guideline follows. Hong Kong ICUs provide a high standard of intensive care by international benchmarks.3 However, because of the expensive nature of intensive care resources, there are a limited number of ICU beds available in Hong Kong. Hong Kong has approximately 7.1 critical care beds per 100 000 population, a low number compared with other high-income regions in Asia (Singapore: 11.4/100 000, Taiwan: 29/100 000), North America (Canada: 12/100 000, United States: 20/100 000), and Europe (Germany: 25/100 000, Belgium: 20/100 000).4 5 Therefore, ICU beds in Hong Kong are generally reserved for patients with reversible medical conditions who have reasonable prospects of substantial recovery, and triage (prioritisation) decisions are routinely necessary.6 7 The existing admission and triage guidelines are designed to help optimise the use of ICU services to achieve the largest possible benefit for the most patients within available resources, a modified utilitarian ethical approach that is recommended by ICU professional bodies internationally.8 9 10 Briefly, patients who require ICU care are referred to the ICU team for admission screening. All ICU admission triage decisions are supervised by a senior, experienced ICU doctor and implemented according to individual unit policy. In principle, all triage decisions should be based on the patient’s medical condition and the benefits likely to be derived from ICU admission (in comparison with a lower level of care). Non-medical factors such as gender, race, religion, education level, and social status should not be considered when making triage decisions. The existing broad-based framework that guides individual unit policy (Fig 1) was used to inform the relevant components of the new COVID-19 Crisis Triage Tool.

Prioritisation for ICU admission in the form of admission triage can only be justified once all efforts to maximise available resources have been exhausted. The Hospital Authority’s existing infectious disease contingency plan dictates that the number of available ICU beds be increased and is based on certain key principles: first, that the standard of intensive care should be maintained at a standard similar to that usually provided by Hospital Authority ICUs, and second, that infection control procedures that provide a high level of protection against staff cross-infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) should be maintained. Maintaining these standards requires the provision of appropriately trained staff in adequate numbers. Failure to adhere to these principles may have devastating consequences, as occurred during the SARS outbreak in 2003.11 The requirement to maintain these standards necessarily results in a relatively limited surge capacity12 that may be incapable of meeting all the demands of a large COVID-19 transmission surge in the community. Thus, the overall increase (68 beds) in ICU airborne infection isolation room beds from Stage I (48 beds) to Stage III (116 beds) will likely be insufficient.

The Hong Kong Government has plans to construct a number of temporary community hospitals for a potentially large surge. However, this initiative does not include a provision for ICU beds, and independent preparations will be required to maximise ICU capacity. In the event that individual hospitals need to increase ICU capacity beyond that specified by the existing Contingency Plan Stage III provisions, a number of key principles should be adopted (Table 1).

Despite some previous attempts, no single objective, evidence-based triage tool in the form of a score or combination of scores has been shown to effectively determine appropriate ICU admission priority.13 14 15 Development of a guidance tool was thus initiated using an iterative process in which possible combinations of predictive scoring components were progressively evaluated for face validity by experienced intensive care specialists. The local development of this triage tool to accompany the Admission, Discharge, and Triage Guidelines for Hong Kong (outlined above) was led by 10 senior ICU clinical specialists who are currently practising in ICUs in Hong Kong. The participants included at least one representative of each of Hong Kong’s hospital cluster regions. Additionally, each participant routinely performs triage as a consequence of chronic ICU bed resource limitations. An initial meeting was held online, at which all key issues were discussed, and a draft document of the consensus view prepared by one author (GMJ). After circulation, several disagreements were documented. These were resolved by online voting, with a majority vote used to resolve persistent disagreement. Three rounds of online voting resulted in a finalised and universally supported document. A decision was made to respect and use the principles laid down in the pre-existing triage framework but to provide further detailed clinical guidance to frontline ICU doctors in Hong Kong regarding COVID-19. The starting point was to adapt and modify a recently published COVID-19 triage prioritisation tool developed by an international expert group.14 This tool took the form of a decision-making algorithm based on established ICU prognostic scoring systems that could inform bedside decision making in the event that ICU bed capacity becomes overwhelmed by patients with COVID-19. Therefore, the specific aim of the triage tool is to provide explicit and uniform guidance to all frontline doctors charged with the responsibility of triaging ICU admissions. This guidance should improve the objectivity and consistency of triage decision making across Hong Kong. The triage tool was designed to be easily understood, rapidly implemented, and of high utility. A list of the major considerations addressed to achieve this goal is provided in Table 2.

The first inclusion and exclusion criteria chosen for the triage tool (Fig 2) were those that, when answered, would rapidly finalise the decision without the need to proceed further, and thus prevent excessive use of valuable medical team time. Therefore, we created clear clinical exclusion criteria. Patients who are too ill to gain substantial incremental benefits from ICU care and those who refuse ICU admission on the basis of the perceived benefits and burdens of ICU care are excluded.

The explicit exclusion criteria are the same ones generally used in Hong Kong ICUs under normal circumstances. We chose general rather than specific diagnoses, as has been proposed previously, as specific diagnoses require the construction of long (but not exhaustive) lists.

The inclusion criteria are also directly comparable with the major inclusion criteria for ICU admission during ‘normal’ conditions: they reflect the need to admit patients who require ICU care to derive a survival benefit. Thus, patients who are ‘too well’ (ie, they can be reasonably treated in the ward) are excluded.

When a patient meets the inclusion criteria and does not meet any exclusion criteria, they become a potential ICU admission, and further priority is determined. Patients are subsequently chosen for admission based on their priority rank, ranging from 1—high priority to 3—low priority. A prioritisation score was developed by including variables that predict short–medium-term mortality (3-6 months) in the first instance and are the most compatible with the principle of ‘quick and clear’ decision making. The clinical frailty scale (CFS) [Fig 3],16 a modified American Society of Anesthesiologists (ASA) score17 to assess co-morbidity (Table 3), and last, the clinical assessment of the number of current organ system failures (OSF), that has previously been well established as an indicator to assist the prediction of mortality.

Outcome prognostication by the senior supervising ICU doctor is largely dependent on knowledge of the factors associated with poor outcomes and clinical experience. Although key relevant factors are captured by the tool, because COVID-19 is a new condition, we provide a table summarising mortality risk factors in patients with COVID-19 who are admitted to ICU to further aid prognostication (Table 4).18 19 20 The data were adapted from countries with ICU practices that are considered generally similar to those in Hong Kong.

Finally, it has been previously recommended that time-limited trials may be adopted at the time of admission.21 A time-limited trial establishes an agreement between the healthcare team and the patient/surrogate to apply necessary intensive care treatment for a pre-determined period of time. The ICU team keeps the family informed of patient progress, and when the pre-agreed time limit is reached, life support therapies are either continued if the patient has responded positively or withdrawn if therapy is failing. Setting an appropriate time period for the trial requires great care,22 and in the setting of COVID-19, care should be taken to allow sufficient time for the patient to respond to therapy. The median number of days of mechanical ventilation and the length of stay have been reported for patients with COVID-19 (10 days, and 9 to 12 days, respectively), whose ICU stays are longer than those of patients with other viral pneumonias.18 19

The existing triage framework has been circulated for comment and feedback from relevant clinical specialty leadership groups in the Hospital Authority. Further, the current accompanying tool has been reviewed by the ad-hoc Hospital Authority Clinical Ethics Committee Core Group and finalised after incorporating relevant suggestions for change. After implementation, the triage working group will review the need for adjustment and updating of the guidelines according to local circumstances.

The conceptual algorithm recommended herein broadly follows the existing recommended framework for individual triage decisions in that the inclusion criteria are based on a low likelihood of survival without ICU care (5%-10% or less), if met. Priorities for admission can then be allocated on the basis of agreed-upon criterion thresholds for survival, as established in the accompanying triage tool and adjusted for Hong Kong’s circumstances at a specific time. Thus, an incremental benefit of at least 40% to 45% would be required to meet the criteria for priority level 3, but one of at least 70% to 75% would be required to meet the criteria for priority level 1. Admission would depend on available resources after safe maximisation of surge capacity and the number of patients queuing for admission (eg, stricter incremental benefit thresholds may be required during the peak of the pandemic, and less strict thresholds may be implemented at the beginning and towards the end). The use of predicted incremental benefit for decision making has been previously endorsed by expert consensus groups when triage is required, both under outbreak and non-outbreak conditions.1 2 14

The CFS, which has nine variables, was chosen as the appropriate general health performance metric, as it meets local practice requirements: familiarity, ease of use, and having been validated as a predictor of short- and medium-term ICU outcomes.23 24 25 26 The well-established ASA score was chosen for modification to guide assessment of co-morbidities, as its descriptions are clear, concise, and logically presented (ASA 2019).17 Further, the relationship between increasing OSF scores and higher mortality is well established.27 28 A simple bedside clinical assessment of organ failure to decide the number of OSF is recommended, rather than attempting to determine the SOFA (Sequential Organ Failure Assessment) score, which requires additional calculations from clinical variables, assessment of missing variables, and then further prioritisation.29 30 We suggest using a clinical judgement for assessing end-stage organ failure of the noted organs (eg, brain, heart, lungs). However, individual units may choose to use the SOFA score if its calculation is considered achievable under local circumstances.

After deliberating at length, the group concluded that the indicative mortalities of the three chosen variables for determining the priority scores (general well-being [CFS], co-morbidities [ASA], and number of OSF) are such that in combination they are likely to correspond to the subjective predicted outcomes and survival percentages noted at the bottom of the notation for each priority score. The noted predicted survival percentages were calibrated with the recommendations of previous consensus expert groups, one who decided to define ‘a minimal acceptable incremental ICU benefit’ in a resource-limited setting as a 15% to 25% difference in mortality,10 and the second who adjusted this difference to be substantially larger (50%) to account for the increased pressure anticipated in an outbreak setting.14 The use of the tool to guide the clinical estimation of likely benefits (outcome of ICU admission compared with outcome expected if the patient remained on the ward/other care area) is necessary for prioritisation of patients who will benefit most from ICU treatment. Nevertheless, because individual patients may have overriding characteristics not captured by the individual or combined scores, the final decision regarding likely incremental benefit and subsequent prioritisation should be made by the senior supervising triage doctor.

If there is more than one patient judged to be within the same priority group, and there is anticipated queuing for the remaining available beds, further prioritisation by incremental ICU benefit, such as saving the most life-years (evaluating mortality from both acute and chronic disorders) should be considered. If a tie for ICU admission candidates remains after these progressive steps, we recommend that admission be determined by the first-come, first-served principle.

Because of the complexity of the decision-making process and the multiple factors that require careful consideration, final decisions are best made by an experienced ICU doctor. However, should uncontrollable circumstances dictate that decisions need be made by a more junior colleague, the tool can still provide assistance to guide and enhance consistent and justifiable decision making. To prepare for this possibility, preparatory education should be provided to more junior colleagues regarding triage decision making to facilitate appropriate interpretation of this tool.

This tool specifically addresses the triage of patients for ICU admission. However, when available ICU resources are overwhelmed, enhanced levels of care within the ward or available high care areas should be used for the treatment of cases denied ICU admission. This could optimise patient outcomes within the constraints of available alternatives. In this regard, both invasive and non-invasive mechanical ventilation is routine practice in the wards and high-care areas of many hospitals in Hong Kong. This fact can potentially be harnessed for the treatment of COVID-19 cases. Patients denied ICU admission on the basis of triage should be preferentially considered for diversion to such resources. Hospital-level coordination and close liaison between hospital facilities management and those who manage ICU resources is required to facilitate the appropriate use of all potentially available resources.31 Although this tool is designed specifically to guide ICU admission triage decisions, other users may consider using the priority assigned by the ICU triage officer to a refused case to allocate the patient to an appropriate next level of care.

Many bedside operational factors are part of the triage process but are not specifically embedded in this crisis tool. Nevertheless, they are substantially addressed in the current Admission, Discharge, and Triage Guidelines, of which the COVID-19 Crisis Triage Tool is an extension. These include the need for clear, empathic communication with patients and surrogates and the implementation of the appropriate best care plan, including palliation of symptoms when appropriate, to patients refused ICU admission. Clear and transparent communication with referring medical teams, mechanisms for audit and oversight, and channels for feedback and reassessment are also required.

Important limitations must be acknowledged. The current guideline is based on the consensus of experienced Hong Kong clinicians with a history of performing bedside triage and not high-level, published medical evidence. Although the prognostic systems chosen have been well demonstrated to align with survival prognosis and functional outcomes, prognostic uncertainty in intensive care cannot be overcome by a single scoring system. All prognostic scoring systems, including the CFS,32 have limitations, and for this reason, the simultaneous use of multiple scoring methods, as used in this tool, has been recommended.15

The referral of a patient for ICU care triggers a complex triage (prioritisation) decision that must be made when ICU beds are limited. It is expected that this triage tool, in the form of a detailed decision aid algorithm, should increase objectivity and transparency in triage decision making and help to enhance consistency between doctors both within and across ICUs in Hong Kong. However, this tool is an aid rather than a complete substitute for the carefully considered judgement of an experienced intensive care clinician.

Concept or design: All authors.
Acquisition of data: All authors.
Analysis or interpretation of data: All authors.
Drafting of the manuscript: GM Joynt.
Critical revision of the manuscript for important intellectual content: All authors.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

1. Christian MD, Sprung CL, King MA, et al. Triage: care of the critically ill and injured during pandemics and disasters: CHEST consensus statement. Chest 2014;146(4 Suppl):e61S-74S. Crossref

2. Joynt GM, Gopalan DP, Argent AA, et al. The Critical Care Society of Southern Africa Consensus Statement on ICU Triage and Rationing (ConICTri). S Afr Med J 2019;109:613-29. Crossref

3. Ling L, Ho CM, Ng PY, et al. Characteristics and outcomes of patients admitted to adult intensive care units in Hong Kong: a population retrospective cohort study from 2008 to 2018. J Intensive Care 2021;9:2. Crossref

4. Phua J, Faruq MO, Kulkarni AP, et al. Critical care bed capacity in Asian countries and regions. Crit Care Med 2020;48:654-62. Crossref

5. Murthy S, Wunsch H. Clinical review: international comparisons in critical care—lessons learned. Crit Care 2012;16:218. Crossref

6. Joynt GM, Gomersall CD, Tan P, Lee A, Cheng CA, Wong EL. Prospective evaluation of patients refused admission to an intensive care unit: triage, futility and outcome. Intensive Care Med 2001;27:1459-65. Crossref

7. Shum HP, Chan KC, Lau CW, Leung AK, Chan KW, Yan WW. Triage decisions and outcomes for patients with Triage Priority 3 on the Society of Critical Care Medicine scale. Crit Care Resusc 2010;12:42-9.

8. Nates JL, Nunnally M, Kleinpell R, et al. ICU Admission, Discharge, and Triage Guidelines: a framework to enhance clinical operations, development of institutional policies, and further research. Crit Care Med 2016;44:1553-602. Crossref

9. Blanch L, Abillama FF, Amin P, et al. Triage decisions for ICU admission: report from the task force of the World Federation of Societies of Intensive and Critical Care Medicine. J Crit Care 2016;36:301-5. Crossref

10. Joynt GM, Gopalan DP, Argent AA, et al. The Critical Care Society of Southern Africa Consensus Guideline on ICU Triage and Rationing (ConICTri). S Afr Med J 2019;109:630-62. Crossref

11. Legislative Council, HKSAR Government. Report of the Select Committee to inquiry into the handling of the severe acute respiratory syndrome outbreak by the Government and the Hospital Authority. 2004. Available from: https://www.legco.gov.hk/yr03-04/english/sc/sc_sars/reports/sars_rpt.htm. Accessed 15 Sep 2020.

12. Gomersall CD, Tai DY, Loo S, et al. Expanding ICU facilities in an epidemic: recommendations based on experience from the SARS epidemic in Hong Kong and Singapore. Intensive Care Med 2006;32:1004-13. Crossref

13. Guidet B, Hejblum G, Joynt G. Triage: what can we do to improve our practice? Intensive Care Med 2013;39:2044-6. Crossref

14. Sprung CL, Joynt GM, Christian MD, Truog RD, Rello J, Nates JL. Adult ICU triage during the coronavirus disease 2019 pandemic: who will live and who will die? Recommendations to improve survival. Crit Care Med 2020;48:1196-202. Crossref

15. Flaatten H, Beil M, Guidet B. Prognostication in older ICU patients: mission impossible? Br J Anaesth 2020;125:655-7. Crossref

16. Rockwood K, Song X, MacKnight C, et al. A global clinical measure of fitness and frailty in elderly people. CMAJ 2005;173:489-95. Crossref

17. American Society of Anaesthesiologists. ASA Physical Status Classification System. 2019. Available from: https:// www.asahq.org/standards-and-guidelines/asa-physical-status-classification-system. Accessed 16 Sep 2020.

18. Gupta S, Hayek SS, Wang W, et al. Factors associated with death in critically ill patients with coronavirus disease 2019 in the US. JAMA Intern Med 2020;180:1-12.Crossref

19. Grasselli G, Greco M, Zanella A, et al. Risk factors associated with mortality among patients with COVID-19 in intensive care units in Lombardy, Italy. JAMA Intern Med 2020;180:1345-55. Crossref

20. Cummings MJ, Baldwin MR, Abrams D, et al. Epidemiology, clinical course, and outcomes of critically ill adults with COVID-19 in New York City: a prospective cohort study. Lancet 2020;395:1763-70. Crossref

21. Quill TE, Holloway R. Time-limited trials near the end of life. JAMA 2011;306:1483-4. Crossref

22. Vink EE, Azoulay E, Caplan A, Kompanje EJ, Bakker J. Time-limited trial of intensive care treatment: an overview of current literature. Intensive Care Med 2018;44:1369-77. Crossref

23. Bagshaw M, Majumdar SR, Rolfson DB, Ibrahim Q, McDermid RC, Stelfox HT. A prospective multicenter cohort study of frailty in younger critically ill patients. Crit Care 2016;20:175. Crossref

24. Bagshaw SM, Stelfox HT, McDermid RC, et al. Association between frailty and short- and long-term outcomes among critically ill patients: a multicentre prospective cohort study. CMAJ 2014;186:E95-102. Crossref

25. Brummel NE, Bell SP, Girard TD, et al. Frailty and subsequent disability and mortality among patients with critical illness. Am J Respir Crit Care Med 2017;196:64-72. Crossref

26. Flaatten H, De Lange DW, Morandi A, et al. The impact of frailty on ICU and 30-day mortality and the level of care in very elderly patients (≥80 years). Intensive Care Med 2017;43:1820-8. Crossref

27. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. Prognosis in acute organ-system failure. Ann Surg 1985;202:685-93. Crossref

28. Peres Bota D, Melot C, Lopes Ferreira F, Nguyen Ba V, Vincent JL. The Multiple Organ Dysfunction Score (MODS) versus the Sequential Organ Failure Assessment (SOFA) score in outcome prediction. Intensive Care Med 2002;28:1619-24. Crossref

29. Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996;22:707-10. Crossref

30. Ferreira FL, Bota DP, Bross A, Mélot C, Vincent JL. Serial evaluation of the SOFA score to predict outcome in critically ill patients. JAMA 2001;286:1754-8. Crossref

31. Joynt GM, Loo S, Taylor BL, et al. Chapter 3. Coordination and collaboration with interface units. Recommendations and standard operating procedures for intensive care unit and hospital preparations for an influenza epidemic or mass disaster. Intensive Care Med 2010;36(Suppl 1):S21-31. Crossref

32. Darvall JN, Bellomo R, Bailey M, et al. Frailty and outcomes from pneumonia in critical illness: a population-based cohort study. Br J Anaesth 2020;125:730-8. Crossref

In 2019, Hong Kong overtook Japan as the region with the world’s longest life expectancy, with the life expectancy of Hong Kong Chinese men at 82.38 years.1 Benign prostatic hyperplasia (BPH) is the most common prostate problem for men older than 50 years. The occurrence of lower urinary tract symptoms (LUTS) due to BPH increases with age. A 1984 autopsy study showed that the prevalence of BPH rose with each decade after age 40, peaking at 88% in men in their 80s.2 Since the 1980s, medical therapy has been prescribed for patients with bothersome LUTS that negatively affects their quality of life. Moreover, the number of co-morbid diseases also increases with age. Co-morbidity increases from 10% at ages up to 19 years to 80% at ages 80 years and older.3 Co-morbidity also leads to polypharmacy and drug-drug interactions, which may result in serious adverse effects.

There is no established local consensus regarding the management of elderly male patients with LUTS. Thus, the Hong Kong Geriatrics Society (HKGS) and the Hong Kong Urological Association (HKUA) formed a working group with the aim of providing insights to clinicians involved in the medical management of male patients with LUTS through a consensus article.

The causes of male LUTS can be multifactorial. Detailed history, appropriate questionnaires, physical examination, and investigation not only help clinicians to reach a diagnosis and identify some alarming conditions (eg, prostate cancer and bladder cancer) but also guide treatment options and give prognostic information for patients’ counselling.

It is useful to determine the most predominant and bothersome LUTS to guide their management, eg, voiding symptoms (weak stream, intermittency, hesitancy, incomplete emptying) and storage symptoms (urgency, frequency, nocturia). The severity of symptoms can be categorised by the International Prostate Symptoms Score as mild (0-7), moderate (8-19), or severe (20-35). It is a validated tool for the assessment of symptoms and quality of life in patients with LUTS (online supplementary Appendix) and allows objective monitoring of treatment response.

Focused histories of the presence of neurological diseases, diabetes mellitus, medication, drinking habits, and prior lower urinary tract procedures are useful to identify causes of LUTS other than BPH (eg, neurogenic bladder, polydipsia, urethral stricture).

Referral to geriatricians should be considered in elderly patients with history of postural hypotension, delirium, dementia, frequent falling, or polypharmacy, as these patients have a higher risk of adverse effects from medical treatment of LUTS, and comprehensive geriatric assessment may be necessary.

Alarming symptoms should raise suspicion of pathologies other than BPH, eg, gross haematuria or unexplained dysuria may imply underlying neoplastic or inflammatory causes, or bedwetting may imply underlying chronic urinary retention with overflow incontinence. Prompt referral to urologists is preferable in the presence of such symptoms.

Digital rectal examination is used to assess prostate size, consistency, the presence of prostatic nodules, and anal tone. In addition, focused examination of the abdomen, external genitalia, and lower limbs is important. Palpable bladder, phimosis, penile mass, and abnormal neurological signs are important to notice when considering referral to appropriate specialists.

A rough estimation of prostate size by number of finger breaths on digital rectal examination is acceptable and may guide the use of 5α-reductase inhibitors (5ARi). Imaging assessment by ultrasound can be considered if more accurate assessment is preferred.

Most patients with LUTS have slow deterioration of symptoms, and very few develop complications over a 5-year period.4 In the primary care setting, the aim of initial evaluation is to detect non-BPH causes, and urinalysis should be included. Prostate-specific antigen (PSA) can be measured after proper counselling, and serum creatinine should be checked when renal impairment is suspected. Numerous additional investigations are also possible, such as flow rate measurement, post-void residual urine volume, renal ultrasonography, prostate sizing, or urodynamic study. However, these additional investigations are optional and need not be routinely performed at the initial evaluation, as they are not cost-effective. Selected patients with appropriate indications (eg, LUTS with poor response to medical treatment, presence of alarming symptoms, impaired renal function) benefit the most from these tests, and input from specialists is preferred in these circumstances.

Urinalysis (dipstick or sediment count) should be included in the primary evaluation of any patients presenting with LUTS to search for urinary tract infections, microscopic haematuria, and diabetes mellitus. If abnormal findings are detected, further tests are recommended.5

One of the differential diagnoses of male LUTS is prostate cancer. Prostate-specific antigen is organ-specific but not cancer-specific. There is substantial overlap in values between men with benign and malignant prostate disease. Hence, elevated PSA levels should be interpreted with caution.

For patients with abnormal DRE, checking PSA can increase the detection rate of prostate cancer. However, for patients with normal DRE, PSA should be checked only when the detection of prostate cancer will cause the disease’s management to be modified. In general, in patients with life expectancy of <10 years or with multiple co-morbidities, checking of PSA to detect prostate cancer might not be beneficial to the patient and should only be performed with special justification after proper counselling.

Assessment of renal function should be considered in patients with high risk of renal impairment (eg, those with multiple co-morbidities and polypharmacy).

The majority of patients with LUTS have slow progression of symptoms, with fewer than 2% developing urinary retention and fewer than 10% requiring BPH surgery over a 5-year period.4 Patients not bothered by their symptoms can be safely managed conservatively with education and lifestyle changes.6 Examples of lifestyle changes include reduction of fluid intake before bedtime to lessen nocturia, avoidance of caffeinated beverages or alcohol to reduce frequency and urgency, urethral milking to prevent post-micturition dribbling, and optimising the timing of medication, especially diuretics.

In addition to education and lifestyle changes, medical treatment can be considered for patients with bothersome symptoms. Voiding symptoms can be regarded as the manifestation of underlying bladder outlet obstruction resulting from BPH, which underpins the rationale of using alpha-1 adrenoceptor antagonists (α1-blockers). Storage symptoms can be attributed to either underlying obstruction-induced change in bladder function or overactive bladder without bladder outlet obstruction. The choice of agent depends on the predominant type of symptoms (ie, voiding vs storage symptoms). For patients with predominant voiding symptoms, the first-line medical treatment is α1-blockers, which have been shown to improve both voiding and storage symptoms.7 For patients with predominant storage symptoms or residual storage symptoms after a trial of α1-blockers, antimuscarinics and beta-3 adrenoceptor agonist (β3 agonist) can be considered. For patients with large prostate (eg, >40 cc), 5ARi can be used to reduce the prostate size, improving symptoms and preventing disease progression in terms of acute urinary retention and future need of BPH surgery. It is important to consider adverse effects before starting medical treatment, especially in older patients with multiple co-morbidities and polypharmacy.

Surgical treatment can be considered for patients who develop BPH complications (eg, urinary retention, bladder stones, obstructive uropathy, recurrent urinary tract infection, haematuria) or symptoms refractory to medical treatment. However, surgery is associated with potential morbidities and mortality, especially in frail geriatric patients.

Frailty is a syndrome characterised by reduced physiological reserve and increased vulnerability to adverse outcomes. Even minor stressor events such as surgery can trigger disproportionate worsening of health status in frail elderly people. The most frequently used model to identify frailty is the phenotype described by Fried et al8 in 2001, which comprises five variables: unintentional weight loss, self-reported exhaustion, low energy expenditure, slow gait speed, and weak grip strength. The definition of polypharmacy has no universally agreed cut-off point with regard to the number of medications. Different researchers have arbitrarily chosen various cut points. In the late 1990’s, the United States Centers for Medicare and Medicaid Services implemented a quality indicator measure that targets patients taking nine or more concurrent medications. An alternative definition of polypharmacy is the use of more medications than are medically necessary.9

After commencement of medical therapy, apart from the monitoring of treatment response and adverse drug reactions, it is also crucial to review medical conditions and identify the new occurrence of geriatric red flags (eg, frailty, polypharmacy) as patients age. The consensus algorithm on male LUTS care flow in the primary care setting by HKGS and HKUA is outlined in Figure 1.

The use of α1-blockers has been shown to be effective at reducing LUTS associated with BPH.5 The α1-blockers relax smooth muscle tone at the bladder neck and prostate by blocking the action of endogenously released noradrenaline.10 They are usually considered as the first-line therapy for male LUTS because of their good efficacy on symptomatic relief but do not alter the natural progression of the disease.

Currently available α1-blockers include prazosin, terazosin, doxazosin, alfuzosin, tamsulosin and silodosin. They have different uroselectivity, pharmacokinetic properties, and formulations (Table 1). Prazosin is a short-acting drug that requires multiple dosing schedules and was the earliest drug to be used for treatment of BPH. However, the 2003 American Urological Association Guidelines concluded that there was insufficient support for recommending prazosin as a treatment option for LUTS secondary to BPH.11 These and the European Association of Urology Guidelines regard prazosin as a nonstandard treatment.

Although different α1-blockers have similar efficacy in improving symptoms and uroflow at appropriate doses, uroselective agents (α-1A blockers) and long-acting preparations appeared to be better tolerated. The differences between the tolerability of various α1-blockers can be explained by the differences in the expression and distribution of receptor subtypes (alpha 1A and 1B) in the body (Fig 2).

Major adverse effects of α1-blocker use include dizziness, asthenia, postural hypotension, and syncope, which can result in falling (odds ratio [OR]=1.14) and fractures (OR=1.16), especially in elderly people,12 the majority of whom cannot tolerate these drugs at the higher adult dose range. Studies have consistently demonstrated that uroselective agents including tamsulosin and silodosin have the least effect on blood pressure and the lowest risk of developing vascular-related events.7 13 14 However, a minor but significant change in blood pressure and heart rate was observed with tamsulosin, whereas no significant change was demonstrated with silodosin compared with placebo in a randomised controlled trial.15

There have been reports concerning the association of α1-blockers (especially tamsulosin) with intraoperative floppy iris syndrome,16 leading to a high rate of complications during cataract surgery, and it is suggested that ophthalmologists be reminded so that they can take precautions. Sexually active patients should be informed of the adverse effect of abnormal ejaculation, which was another adverse reaction more commonly related with tamsulosin (OR=8.58) and silodosin (OR=32.5), and patients should be informed of the potential implications.17

Patients who are naïve to α-blockers may develop postural hypotension, known as the “first dose phenomenon,” which is more pronounced with non-selective α1-blockers during the first 8 weeks of treatment. However, it should not be overlooked with uroselective agents, and special precautions should be taken, especially in elderly patients. In addition, swallowing difficulties are not uncommon in elderly patients, in whom modified release preparations are inappropriate.

The pitfalls of prescribing α1-blockers are summarised in Table 2.18 Their most troublesome adverse effect is postural hypotension. The situation is even more complicated if the patient has concomitant hypertension or is taking multiple medications with hypotensive effects for various indications. It is estimated that more than 25% of men aged >60 years have concomitant BPH and hypertension,19 which poses a significant challenge in the prescription of α1-blockers. Non-selective α1-blockers have been available as antihypertensive agents for over 40 years. They reduce blood pressure by blocking postsynaptic alpha (mainly alpha-1B) receptors, thereby inhibiting noradrenaline release that induces vasoconstriction, resulting in dilatation of arterioles and venules. Among all α1-blockers, prazosin, terazosin, and doxazosin are approved for the management of hypertension, whereas alfuzosin, tamsulosin, and silodosin have minimal effect on blood pressure.

Because certain α1-blockers are approved treatments for hypertension, they are a reasonable choice for treatment of hypertensive men with LUTS. However, with advances in hypertensive treatment over past decades, the role of α1-blockers in this context has changed, especially after the introduction of uroselective agents. A consensus was reached regarding revision of the use of available safety data on α1-blockers in patients with hypertension (Table 3).20 21 22 23 24 25 26 27 28 29

For hypertensive LUTS patients who are not taking any antihypertensives, we do not recommend the use of non-selective α1-blockers for treatment of BPH and hypertension together (first-line treatment of hypertension). This recommendation is based on the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial Study.20 That study showed that compared with chlorthalidone (diuretics), the use of doxazosin is associated with significantly higher risk of stroke, congestive heart failure, peripheral vascular disease, angina, and cardiovascular disease requiring coronary revascularisation. Multiple guidelines (The Eighth Joint National Committee (JNC8) guideline30, the European Society of Cardiology/the European Society of Hypertension guideline31, the American College of Cardiology/American Heart Association guideline32 and Hypertension Canada’s 2017 guideline for diagnosis33) do not recommend α1-blockers as the first-line therapy for hypertension. Therefore, we recommended treating LUTS with uroselective agents, and hypertension should be treated with another class of antihypertensives according to existing hypertension guidelines.

For hypertensive LUTS patients who have suboptimal blood pressure control but are taking antihypertensive treatment, the addition of non-selective α1-blockers (doxazosin gastrointestinal therapeutic system [GITS] and terazosin) for treatment of hypertension as second- or third-line agents seems to be a reasonable option for achieving optimal blood pressure control.23 24 25 26 However, postural hypotension is a significant concern in the treatment group. Therefore, although non-selective α1-blockers are effective at reducing blood pressure as add-on therapy, the risks and benefits of this approach should be balanced and individualised. This is the case especially when other classes of antihypertensives may have additional benefits in certain patients in whom treatment of LUTS with uroselective agents and hypertension separately with another class of antihypertensive agents might be advisable.

Among normotensive male patients aged >40 years with LUTS who were not taking antihypertensives, a multicentre study showed that doxazosin GITS improved the International Prostate Symptoms Score and the Quality of Life index with minimal effect on blood pressure.21 Another review of α1-blockers’ effects on blood pressure showed no significant changes in blood pressure in normotensive patients irrespective of the type of α1-blockers used (tamsulosin, alfuzosin, doxazosin GITS, or terazosin).22 Therefore, it is reasonable to consider either non-selective or uroselective α1-blockers in this group.

The safety data on non-selective α1-blockers in normotensive LUTS patients who are taking antihypertensives with optimal blood pressure control are largely based on post-hoc analysis of randomised controlled trials assessing standard versus intensive blood pressure control, which involves non-selective α1-blockers as third- or fourth-line antihypertensives. In the Action to Control Cardiovascular Risk in Diabetes trial, which involved 10 251 high-risk participants with type 2 diabetes mellitus at 77 centres, non-selective α1-blockers were significantly associated with postural hypotension, which was associated with higher mortality and rates of heart failure and hospitalisation.27 Another European study, the Systolic Blood Pressure Intervention trial, which involved 9361 patients with increased cardiovascular risk but without diabetes, found that non-selective α1-blockers are associated with higher risk of syncope and falling, although no significant hypotensive events were demonstrated.28 Therefore, we recommend the use of uroselective agents for management of LUTS in this group of patients.

Apart from the risk of postural hypotension, the selection between non-selective and uroselective agents should also be based on other factors, including the risk of falling, polypharmacy, co-morbidities, and specific situations where the use of uroselective agents is advisable (Table 4).

Nevertheless, α1-blockers are effective in relieving LUTS and improving quality of life for patients with BPH. However, treatment decisions should be individualised and based on comprehensive assessment of patients with different needs, especially in frail elderly patients, as they tend to have accumulated co-morbidities, disabilities, and polypharmacy that often interact with each other.

Antimuscarinics are commonly used as pharmacological treatments for overactive bladder. This class of drug can also be used in predominant or mixed storage LUTS. These drugs increase bladder capacity and reduce urgency by blockading the muscarinic receptor during bladder storage.34 Antimuscarinics have shown a modest benefit over placebo in reducing urgency incontinence in women.35 36 The efficacy of all the antimuscarinics is similar.37 However, there is lack of head-to-head comparison, and not all antimuscarinics have been tested in elderly men. These drugs often require higher doses to achieve the optimal effects, and we recommend starting with the lowest dose and titrating up as needed if the patient has insufficient response and minimal adverse effects. Antimuscarinics should be avoided if the patient has clinically palpable bladder. These drugs can be associated with increased post-void residual urine volume after therapy, but acute retention is rare.5 Follow-up is recommended at 4 to 6 weeks to assess therapeutic response and determine whether a change in medication is necessary. Men should be advised to discontinue medication if they develop voiding difficulty, urinary infection, or worsening LUTS after initiation of therapy.

All antimuscarinics exert peripheral anticholinergic effects that may limit drug tolerability and dose escalation.35 Common adverse events include dry mouth (up to 16%), constipation (up to 4%), dizziness (up to 5%), micturition difficulty (up to 2%), blurred vision for near objects, tachycardia, drowsiness, and worsened cognitive function.5 Up to two-thirds of patients discontinue these medications beyond 1 year.38 Constipation and compensatory fluid intake for dry mouth may exacerbate urinary incontinence. Patients with dementia are more vulnerable to the adverse effects of antimuscarinics.39 40 Antimuscarinics should be avoided in patients with uncontrolled tachyarrhythmia, myasthenia gravis, and narrow angle-closure glaucoma. The adverse effects of antimuscarinics can be explained by the distribution of muscarinic acetylcholine receptor subtypes throughout the body (Fig 3). The differences in tolerability between antimuscarinics can be explained by their differences in selectivity for receptor subtypes and tissue penetration.

Antimuscarinics may have additive adverse effects when combined with other medications that have strong anticholinergic effects. They should be used with caution or preferably avoided if elderly patients are concomitantly taking other medications with high anticholinergic potency, eg, first-generation H1 antihistamines (chlorpheniramine, hydroxyzine, diphenhydramine), anti-Parkinson’s drugs (benztropine, trihexyphenidyl), spasmolytics (atropine, hyoscine), anti-emetics (promethazine), muscle relaxants, antipsychotics (chlorpromazine, fluphenazine, trifluoperazine, clozapine), and tricyclic antidepressants (amitriptyline, clomipramine, doxepin, imipramine, nortriptyline).41 42 43

The antimuscarinics registered in Hong Kong include oxybutynin, solifenacin, tolterodine, trospium, darifenacin, and fesoterodine. Solifenacin, darifenacin, and trospium may have less impact on the central nervous system (Table 5).

Beta-3 agonist is a new class of pharmacological treatment used to relieve storage symptoms (urgency, urinary frequency, and urge urinary incontinence) associated with overactive bladder. It acts by binding to the β3 adrenergic receptors on the bladder smooth muscle causing bladder relaxation during the storage phase. Mirabegron is currently the only approved β3 agonist for treatment of overactive bladder.

ln a phase III clinical trial, mirabegron 50 mg daily resulted in a 50% reduction in the number of urgency episodes per 24 hours and a 128% increase in the mean volume voided per micturition compared with placebo.44 Unlike antimuscarinics, it has better tolerability with less dry mouth. In the study, mirabegron’s incidence of dry mouth was similar to that of placebo.45

Mirabegron has no influence on bladder contraction during the voiding phase. In the clinical trial, the incidence rate of acute urinary retention was the lowest in mirabegron-treated patients compared with the tolterodine and placebo groups (0.1%, 0.6%, and 0.2%, respectively).44 The same trial showed that mirabegron did not increase intraocular pressure, and it is therefore not contra-indicated in patients with glaucoma.

Regarding cardiovascular safety, the review and real-world data on mirabegron did not show any increased risk compared with conventional antimuscarinics or in those with coexisting cardiovascular disease.46 47 The European Association of Urology guideline recommends β3 agonist as a first-line medication for men with moderate-to-severe LUTS who have predominantly bladder storage symptoms.48

Beta-3 agonist can be considered when antimuscarinic adverse effects and high anticholinergic burden are concerns, especially in elderly adults with multiple co-morbidities and cognitive impairment. Several studies have shown that mirabegron is safe and effective in older patients.49 50 51

The recommended dosage of mirabegron is 50 mg daily. For patients with renal impairment (estimated glomerular filtration rate 15-29 mL/min/1.73 m²), Child–Pugh class B liver impairment, and those aged ≥80 years with multiple co-morbidities, 25 mg daily should be considered. Mirabegron is not recommended in patients with poorly controlled hypertension (systolic blood pressure >180 mm Hg or diastolic blood pressure >110 mm Hg), severe renal impairment (estimated glomerular filtration rate <15 mL/min/1.73 m²), or Child–Pugh class C liver impairment. The most common adverse effects reported were hypertension, nasopharyngitis, and headache but the overall adverse event rates were similar to those with placebo.52

5α-reductase is responsible for conversion of testosterone to dihydrotestosterone, which has an important role in prostate growth and the development of BPH.53 There are two isoforms of 5α-reductase: type 1—the predominant enzyme in extraprostatic tissue such as skin and liver; and type 2—the predominant enzyme in prostate (>90%), which is critical to development of BPH.

The 5ARi drugs inhibit conversion of testosterone to dihydrotestosterone, inducing apoptosis and atrophy of prostatic epithelial cells.54 It results in reduction of prostate volume and hence relief of bladder outflow obstruction. There are two types of 5ARi: finasteride, which acts only on type 2 5α-reductase, and dutasteride, which acts on both types. Meta-analysis has shown no differences in efficacy or safety among these two drugs.55 56 There are a few registered 5ARi drugs: Proscar (finasteride 5 mg), Avodart (dutasteride 0.5 mg), and Duodart (combination of dutasteride 0.5 mg and tamsulosin 0.4 mg).

Long-term 5ARi treatment in patients with moderate to severe LUTS and prostate volume >40 cc has been shown to reduce the symptoms score, risk of urinary retention, and risk of BPH-related surgery. In a landmark study, patients taking finasteride had improvement in symptoms and uroflow, their prostate size reduced by 20%, their risk of acute urinary retention reduced by 57%, and their risk of BPH-related surgery reduced by 55% compared with placebo after 4 years of treatment.56

There are some practical tips for prescribing 5ARi. First, the patient should have an enlarged prostate >40 cc on ultrasound imaging. If ultrasound is not readily available, it is acceptable to start 5ARi treatment when the prostate size is greater than two finger breadths on DRE. Second, it is important to inform the patient that 5ARi have a slow onset of action (3-6 months), as time is required for prostate volume reduction. Continuous long-term treatment should be expected. Third, the effects of 5ARis on PSA levels should be explained to patients. The PSA level is expected to be reduced by 50% after 6 to 12 months of treatment,56 and therefore, good drug compliance is required for proper interpretation of the PSA level in prostate cancer screening. A persistent PSA rise from the nadir in a patient on long-term 5ARi treatment is an indicator for prostate biopsy, and urological referral should be considered.57 Finally, although some studies have suggested a higher incidence of high-grade prostate cancer in patients taking long-term 5ARi, no causal relationship has been proven, and there is no difference in long-term survival.58 The common adverse effects are sexual dysfunction, such as decreased libido, erectile dysfunction, and ejaculatory problems in around 4% to 8% and breast enlargement and tenderness in 1% of patients.56

Male LUTS is a common presentation to primary care practitioners. Focused history and physical examination are essential to differentiate BPH from other causes of male LUTS and to guide its management. Patients with minimal symptoms can be managed conservatively, and pharmacological treatments can be considered if symptoms are bothersome. For patients with symptoms refractory to pharmacological treatments or who have complications (eg, urinary retention, obstructive uropathy), surgical intervention can be performed after assessment by urologists. With an ageing population, geriatricians are adopting an increasing role in the management of male patients with LUTS in the era of multiple co-morbidities and polypharmacy, as these patients are at higher risk of adverse effects from pharmacological treatments and are not optimal for surgical intervention. A consensus has been reached by the HKGS and HKUA regarding the diagnosis, evaluation, management, and referral mechanism for LUTS in the primary care setting. With collaboration between primary care practitioners, geriatricians and urologists, we hope that more holistic care can be provided to male patients with LUTS in Hong Kong.

All authors contributed to the concept or design of the study, acquisition of the data, analysis or interpretation of the data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

We thank Dr CH Cheng for drawing the images in Figures 2 and 3; and Dr CW Man for his assistance in editing the manuscript.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

1. United Nations Development Programme. Human Development Report 2019. Beyond income, beyond averages, beyond today: inequalities in human development in the 21st century. 2019. Available from: http://hdr.undp.org/en/content/human-development-report-2019. Accessed 7 Jul 2020.

2. Berry SJ, Coffey DS, Walsh PC, Ewing LL. The development of human benign prostatic hyperplasia with age. J Urol 1984;132:474-9. Crossref

3. van den Akker M, Buntinx F, Metsemakers JF, Roos S, Knottnerus JA. Multimorbidity in general practice: prevalence, incidence, and determinants of co-occurring chronic and recurrent diseases. J Clin Epidemiol 1998;51:367-75. Crossref

4. Ball AJ, Feneley RC, Abrams PH. The natural history of untreated “prostatism”. Br J Urol 1981;53:613-6. Crossref

5. Gravas S, Cornu JN, Gacci M, et al. EAU Guidelines on management of non-neurogenic male LUTS. Arnhem, The Netherlands: EAU Guidelines Office; 2020.

6. Brown CT, Yap T, Cromwell DA, et al. Self management for men with lower urinary tract symptoms: randomised controlled trial. BMJ 2007;334:25. Crossref

7. Djavan B, Marberger M. A meta-analysis on the efficacy and tolerability of alpha1-adrenoceptor antagonists in patients with lower urinary tract symptoms suggestive of benign prostatic obstruction. Eur Urol 1999;36:1-13. Crossref

8. Fried LP, Tangen CM, Walston J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001;56:M146-56. Crossref

9. Tjia J, Velten SJ, Parsons C, Valluri S, Briesacher BA. Studies to reduce unnecessary medication use in frail older adults: a systematic review. Drugs Aging 2013;30:285-307. Crossref

10. Michel MC, Vrydag W. Alpha1-, alpha2- and beta-adrenoceptors in the urinary bladder, urethra and prostate. Br J Pharmacol 2006;147 Suppl 2:S88-119. Crossref

11. AUA Practice Guidelines Committee. AUA guideline on management of benign prostatic hyperplasia (2003). Chapter 1: Diagnosis and treatment recommendations. J Urol 2003;170:530-47. Crossref

12. Welk B, McArthur E, Fraser LA, et al. The risk of fall and fracture with the initiation of a prostate-selective α antagonist: a population based cohort study. BMJ 2015;351:h5398. Crossref

13. Djavan B, Chapple C, Milani S, Marberger M. State of the art on the efficacy and tolerability of alpha1-adrenoceptor antagonists in patients with lower urinary tract symptoms suggestive of benign prostatic hyperplasia. Urology 2004;64:1081-8. Crossref

14. Morgia G. Does the use of silodosin to treat benign prostatic hyperplasia really offer something new? Eur Urol 2011;59:353-5. Crossref

15. Chapple CR, Montorsi F, Tammela TL, et al. Silodosin therapy for lower urinary tract symptoms in men with suspected benign prostatic hyperplasia: results of an international, randomized, double-blind, placebo- and active-controlled clinical trial performed in Europe. Eur Urol 2011;59:342-52. Crossref

16. Storr-Paulsen A, Nørregaard JC, Børme KK, Larsen AB, Thulesen J. Intraoperative floppy iris syndrome (IFIS): a practical approach to medical and surgical considerations in cataract extractions. Acta Ophthalmol 2009;87:704-8. Crossref

17. Gacci M, Ficarra V, Sebastianelli A, et al. Impact of medical treatments for male lower urinary tract symptoms due to benign prostatic hyperplasia on ejaculatory function: a systematic review and meta-analysis. J Sex Med 2014;11:1554-66. Crossref

18. United States Department of Veterans Affairs. Pharmacy Benefits Management Services, Medical Advisory Panel, VISN Pharmacist Executives. Alfuzosin, silodosin, tamsulosin/clinically uroselective alpha1-adrenergic blockers: recommendations for use. 2010. Available from: https://www.pbm.va.gov/PBM/clinicalguidance/clinicalrecommendations/Alpha_Blockers_(Clinically_Uroselective_Alfuzosin_Silodosin_Tamsulosin)_Clinical_Recommendations.pdf. Accessed 27 Nov 2019.

19. Boyle P, Napalkov P. The epidemiology of benign prostatic hyperplasia and observations on concomitant hypertension. Scand J Urol Nephrol Suppl 1995;168:7-12. Crossref

20. ALLHAT Collaborative Research Group [editorial]. Major cardiovascular events in hypertensive patients randomized to doxazosin vs chlorthalidone: the antihypertensive and lipid-lowering treatment to prevent heart attack trial (ALLHAT). JAMA 2000;283:1967-75. Crossref

21. Dahlöf B, Sever PS, Poulter NR, et al. Prevention of cardiovascular events with an antihypertensive regimen of amlodipine adding perindopril as required versus atenolol adding bendroflumethiazide as required, in the Anglo-Scandinavian Cardiac Outcomes Trial-Blood Pressure Lowering Arm (ASCOT-BPLA): a multicentre randomised controlled trial. Lancet 2005;366:895-906. Crossref

22. Lowe FC, Olson PJ, Padley RJ. Effects of terazosin therapy on blood pressure in men with benign prostatic hyperplasia concurrently treated with other antihypertensive medications. Urology 1999;54:81-5. Crossref

23. Chung BH, Hong SJ. Long-term follow-up study to evaluate the efficacy and safety of the doxazosin gastrointestinal therapeutic system in patients with benign prostatic hyperplasia with or without concomitant hypertension. BJU Int 2006;97:90-5. Crossref

24. Lee SH, Park KK, Mah SY, Chung BH. Effects of α-blocker ‘add on’ treatment on blood pressure in symptomatic BPH with or without concomitant hypertension. Prostate Cancer Prostatic Dis 2010;13:333-7. Crossref

25. Black HR, Keck M, Meredith P, Bullen K, Quinn S, Koren A. Controlled-release doxazosin as combination therapy in hypertension: the GATES study. J Clin Hypertens (Greenwich) 2006;8:159-66. Crossref

26. de Alvaro F, Hernández-Presa MA, ASOCIA Study. Effect of doxazosin gastrointestinal therapeutic system on patients with uncontrolled hypertension: the ASOCIA Study. J Cardiovasc Pharmacol 2006;47:271-6. Crossref

27. Fleg JL, Evans GW, Margolis KL, et al. Orthostatic hypotension in the ACCORD (Action to Control Cardiovascular Risk in Diabetes) blood pressure trial: prevalence, incidence, and prognostic significance. Hypertension 2016;68:888-95. Crossref

28. SPRINT Research Group; Wright JT Jr, Williamson JD, et al. A randomized trial of intensive versus standard blood-pressure control. N Engl J Med 2015;373:2103-16. Crossref

29. Hiremath S, Ruzicka M, Petrcich W, et al. Alpha-blocker use and the risk of hypotension and hypotension-related clinical events in women of advanced age. Hypertension 2019;74:645-51. Crossref

30. James PA, Oparil S, Carter BL, et al. 2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA 2014;311:507-20. Crossref

31. Williams B, Mancia G, Spiering W, et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension: The Task Force for the management of arterial hypertension of the European Society of Cardiology and the European Society of Hypertension: The Task Force for the management of arterial hypertension of the European Society of Cardiology and the European Society of Hypertension. J Hypertens 2018;36:1953-2041. Crossref

32. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/ AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension 2018;71:e13-e115. Crossref

33. Leung AA, Daskalopoulou SS, Dasgupta K, et al. Hypertension Canada’s 2017 Guidelines for Diagnosis, Risk Assessment, Prevention, and Treatment of Hypertension in Adults. Can J Cardiol 2017;33:557-76. Crossref

34. Finney SM, Andersson KE, Gillespie JI, Stewart LH. Antimuscarinic drugs in detrusor overactivity and the overactive bladder syndrome: motor or sensory actions? BJU Int 2006;98:503-7. Crossref

35. Shamliyan T, Wyman JF, Ramakrishnan R, Sainfort F, Kane RL. Benefits and harms of pharmacologic treatment for urinary incontinence in women: a systematic review. Ann Intern Med 2012;156:861-74. Crossref

36. Reynolds WS, McPheeters M, Blume J, et al. Comparative effectiveness of anticholinergic therapy for overactive bladder in women: a systematic review and meta-analysis. Obstet Gynecol 2015;125:1423-32. Crossref

37. Qaseem A, Dallas P, Forciea MA, et al. Nonsurgical management of urinary incontinence in women: a clinical practice guideline from the American College of Physicians. Ann Intern Med 2014;161:429-40. Crossref

38. Brostrøm S, Hallas J. Persistence of antimuscarinic drug use. Eur J Clin Pharmacol. 2009;65:309-14. Crossref

39. Coupland CA, Hill T, Dening T, Morriss R, Moore M, Hippisley-Cox J. Anticholinergic drug exposure and the risk of dementia: a nested case-control study. JAMA Intern Med 2019;179:1084-93. Crossref

40. Fox C, Richardson K, Maidment ID, et al. Anticholinergic medication use and cognitive impairment in the older population: the medical research council cognitive function and ageing study. J Am Geriatr Soc 2011;59:1477-83. Crossref

41. Durán CE, Azermai M, Vander Stichele RH. Systematic review of anticholinergic risk scales in older adults. Eur J Clin Pharmacol 2013;69:1485-96. Crossref

42. Salahudeen MS, Hilmer SN, Nishtala PS. Comparison of anticholinergic risk scales and associations with adverse health outcomes in older people. J Am Geriatr Soc 2015;63:85-90. Crossref

43. American Geriatrics Society 2015 Beers Criteria Update Expert Panel. American Geriatrics Society 2015 updated Beers criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc 2015;63:2227-46. Crossref

44. Chapple CR, Cardozo L, Nitti VW, Siddiqui E, Michel MC. Mirabegron in overactive bladder: a review of efficacy, safety, and tolerability. Neurourol Urodyn 2014;33:17-30. Crossref

45. Khullar V, Amarenco G, Angulo JC, et al. Efficacy and tolerability of mirabegron, a β(3)-adrenoceptor agonist, in patients with overactive bladder: results from a randomised European-Australian phase 3 trial. Eur Urol 2013;63:283- 95. Crossref

46. Rosa GM, Ferrero S, Nitti VW, Wagg A, Saleem T, Chapple CR. Cardiovascular safety of β3-adrenoceptor agonists for the treatment of patients with overactive bladder syndrome. Eur Urol 2016;69:311-23. Crossref

47. Katoh T, Kuwamoto K, Kato D, Kuroishi K. Real-world cardiovascular assessment of mirabegron treatment in patients with overactive bladder and concomitant cardiovascular disease: results of a Japanese post-marketing study. Int J Urol 2016;23:1009-15. Crossref

48. Burkhard FC, Bosch JL, Cruz F, et al. EAU Guidelines on urinary incontinence in adults. Arnhem, The Netherlands: EAU Guidelines Office; 2016.

49. Wagg A, Staskin D, Engel E, Herschorn S, Kristy RM, Schermer CR. Efficacy, safety, and tolerability of mirabegron in patients aged ≥65yr with overactive bladder wet: a phase IV, double-blind, randomised, placebo-controlled study (PILLAR). Eur Urol 2020;77:211-20. Crossref

50. Makhani A, Thake M, Gibson W. Mirabegron in the treatment of overactive bladder: safety and efficacy in thehttps://doi.org/10.2147/CIA.S174402 very elderly patient. Clin Interv Aging 2020;15:575-81. Crossref

51. Lee YK, Kuo HC. Safety and therapeutic efficacy of mirabegron 25 mg in older patients with overactive bladder and multiple comorbidities. Geriatr Gerontol Int 2018;18:1330-3. Crossref

52. Tubaro A, Batista JE, Nitti VW, et al. Efficacy and safety of daily mirabegron 50 mg in male patients with overactive bladder: a critical analysis of five phase III studies. Ther Adv Urol 2017;9:137-54. Crossref

53. Carson C 3rd, Rittmaster R. The role of dihydrotestosterone in benign prostatic hyperplasia. Urology 2003;61(4 Suppl 1):2-7. Crossref

54. Rittmaster RS, Norman RW, Thomas LN, Rowden G. Evidence for atrophy and apoptosis in the prostates of men given finasteride. J Clin Endocrinol Metab 1996;81:814-9. Crossref

55. Jun JE, Kinkade A, Tung AC, Tejani AM. 5α-reductase inhibitors for treatment of benign prostatic hyperplasia: a systematic review and meta-analysis. Can J Hosp Pharm 2017;70:113-9. Crossref

56. Roehrborn CG, Boyle P, Bergner D, et al. Serum prostatespecific antigen and prostate volume predict long-term changes in symptoms and flow rate: results of a four-year, randomized trial comparing finasteride versus placebo. PLESS Study Group. Urology 1999;54:662-9. Crossref

57. Marks LS, Andriole GL, Fitzpatrick JM, Schulman CC, Roehrborn CG. The interpretation of serum prostate specific antigen in men receiving 5alpha-reductase inhibitors: a review and clinical recommendations. J Urol 2006;176:868-74. Crossref

58. Thompson IM Jr, Goodman PJ, Tangen CM, et al. Longterm survival of participants in the prostate cancer prevention trial. N Engl J Med 2013;369:603-10. Crossref

Women need special attention in epilepsy care. They face various challenges related to their reproductive cycles, including pregnancy, breastfeeding, contraception, menopause, and catamenial epilepsy. Antiepileptic drug (AED) options have increased exponentially in recent decades. New data on epilepsy management have emerged since the 2009 publication of the Hong Kong Epilepsy Guideline by the Hong Kong Epilepsy Society.1 This article aims to update the Society’s guideline with a focus on epilepsy management in women. This project received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Women with epilepsy, preferably with their partner or parents if appropriate, should receive counselling on contraception, conception, pregnancy, breastfeeding, and childcare. Preconception counselling is especially important for women who will become pregnant, as it may help to ensure good maternal and fetal outcomes.2 They should be reassured that most women with epilepsy have uneventful pregnancies and deliveries.2

The risk of epilepsy in the offspring of women with epilepsy is highly dependent on the parents’ epilepsy syndrome status.3 In general, the risk is slightly higher compared with that of the background population. In a large population-based study, the overall cumulative risk of epilepsy up to age 40 in individuals with epileptic parents was 4.5%, which is about 3-fold higher than that of the general population.4 Referral to appropriate specialists such as geneticists is appropriate if genetic counselling is indicated.

Animal studies have shown that AEDs can decrease the level of serum folate, increasing the risk of fetal neural tube defects.5 Neonatal malformations have also been associated with low maternal serum folate levels in humans.6 All women with epilepsy of childbearing age may be offered folate supplements while on AEDs.7 8 Although there is no definite consensus about folate dosage, it is reasonable to prescribe oral folate 5 mg daily to women with childbearing potential.1 2 9

Many AEDs have the ability to cross the placenta.7 Data have emerged on the potential risks of in-utero AED exposure to offspring. The risk of major congenital malformations (MCMs) such as hypospadias, congenital heart defects, club foot, cleft lip or palate, and spina bifida in children born to women with epilepsy treated with AEDs during pregnancy is 2 to 3 times higher than the 2% to 5% occurrence rate in the general population.10 11 Polytherapy probably has a higher risk of MCMs and future cognitive adverse drug effects compared with monotherapy.12 13 14 The risk of MCMs may increase with higher dosages of AEDs.12 Antenatal screening including fetal ultrasound scan can help to detect major fetal malformations.2 However, prenatal screening has limitations in terms of malformation detection and cannot provide information about neurodevelopmental complications.15

The North American Antiepileptic Drug Pregnancy Registry (NAAPR) reported that phenobarbital monotherapy was associated with an increased MCM rate of 6.5% (95% confidence interval [CI], 2.1%-14.5%) compared with the background rate of 1.6%.16 The same registry later reported that the MCM risk associated with phenobarbital was 5.5% in a larger group of 199 pregnancies.17 The teratogenic effects of phenobarbital are probably related to dosage. Data from the European Registry of Antiepileptic Drugs and Pregnancy (EURAP) showed that phenobarbital had an MCM risk of 5.4% (95% CI, 2.51%-10.04%) in the offspring of women taking daily doses <150 mg. The risk increased to 13.7% (95% CI, 5.70%-26.26%) at daily doses of 150 mg per day or above. Phenobarbital may also have deleterious effects on the cognition of offspring.12 18

A meta-analysis found that phenytoin had an MCM rate of 7.4% (95% CI, 3.60%-11.11%).14 The EURAP reported that the MCM rate associated with phenytoin was approximately 6.4% (95% CI, 2.8%-12.2%).19 The NAAPR reported that phenytoin was associated with a 2.9% risk of MCMs (compared with a background rate of 1.1%).17 The UK and Ireland Epilepsy and Pregnancy Register reported that the risk of MCMs associated with phenytoin was 3.7%.20 21 22 There is also evidence showing that in-utero phenytoin exposure is associated with an increased risk of impaired cognition.23 24 25

Valproate use during pregnancy may be associated with MCMs and long-term negative effects on the cognitive and neurological functions of offspring. Neural tube defects, facial clefts, and hypospadias may be related to in-utero valproate exposure.1 13 27 28 29 A meta-analysis showed that the overall risk of MCMs associated with in-utero exposure to valproate was 10.73% (95% CI, 8.16%-13.29%).14 The European Surveillance of Congenital Anomalies study found that valproate monotherapy was associated with significantly increased risks of six specific malformations, with the following odds ratios (ORs): spina bifida, 12.7-fold increase (95% CI, 7.7-20.7); atrial septal defect, 2.5-fold (1.4-4.4); cleft palate, 5.2-fold (2.8-9.9); hypospadias, 4.8-fold (2.9-8.1); polydactyly, 2.2-fold (1.0-4.5); and craniosynostosis, 6.8-fold (1.8-18.8).30 Adverse effects in terms of mental and motor development have been shown to be associated with in-utero valproate exposure in children.31 32 33 34 35 In 2011, the US Food and Drug Administration (FDA) issued a warning about valproate use during pregnancy because interim results from the NEAD (Neurodevelopmental Effects of Antiepileptic Drugs) study and other epidemiological studies showed lower cognitive function in children with in-utero valproate exposure.31 36 37 The NEAD study was a prospective multicentre cohort study conducted in the US and UK. Interim cognitive assessment at age 3 years showed that children exposed to valproate had intelligence quotient (IQ) scores 9 points lower than those exposed to lamotrigine (95% CI, 3.1-14.6; P=0.009), 7 points lower than those exposed to phenytoin (95% CI, 0.2-14.0; P=0.04), and 6 points lower than those exposed to carbamazepine (95% CI, 0.6-12.0; P=0.04). The association between valproate use and detrimental effects on IQ was dose-dependent.31 When 244 newborn children subsequently completed 6 years of follow-up, mean IQ at age 6 years was lower in children exposed to valproate (97; 95% CI, 94-101) than to carbamazepine (105; 95% CI, 102-108; P=0.0015), lamotrigine (108; 95% CI, 105-110; P=0.0003), or phenytoin (108; 95% CI, 104-112; P=0.0006).32 Children exposed to valproate were inferior in terms of verbal and memory abilities to those exposed to the other AEDs and in terms of non-verbal and executive functions to those exposed to lamotrigine. High doses of valproate were negatively associated with IQ, verbal ability, non-verbal ability, memory, and executive function.32 Valproate use during pregnancy may also increase the risk of childhood autism in offspring (adjusted hazard ratio: 2.9; 95% CI, 1.4-6.0).38 39 In 2013, the FDA strengthened its warning regarding valproate use: it suggested that valproate should only be used in epileptic pregnant women if other medications are ineffective or unacceptable, and it is contra-indicated for migraines during pregnancy. In women with epilepsy who are currently not pregnant but have the potential to become pregnant, valproate should not be considered as the first-line treatment, especially for focal epilepsies, unless there is no other alternative. For cases in which valproate is considered as an appropriate option, like certain idiopathic generalised epilepsies, the dose should be maintained at the lowest effective dose, preferably not exceeding 500 to 800 mg per day.15 In special circumstances, the use of valproate as initial treatment may be justifiable. For example, it could be used in epilepsies with high likelihood of remission and drug withdrawal before puberty or in patients with severe disabilities that make future pregnancy extremely unlikely.15 The American Academy of Neurology has recommended that valproate be avoided during the first trimester of pregnancy if possible.40 However, the use of valproate in women with epilepsy should be individualised with consideration of the dosage administered, seizure control, and potential effects on offspring. In certain cases, especially those of specific epilepsy syndromes such as juvenile myoclonic epilepsy and juvenile absence epilepsy, valproate may be the most suitable choice after balancing its seizure control properties with the potential adverse effects on both the patient and her offspring. Valproate may also be used when clear communication with the patient about the risk-benefit ratio has been undertaken.

Although the number of women with childbearing potential who take valproate is decreasing, there are persistent concerns about inadequate information provided to valproate users about its possible adverse effects on their offspring. Authorities from various countries or regions, such as the UK and the European Union, have adopted specific risk minimisation measures surrounding valproate usage among women with epilepsy. Examples include the provision of information leaflets highlighting the potential teratogenic and neurodevelopmental impacts of valproate exposure in utero, pregnancy tests before valproate prescription, a risk acknowledgement form filled upon valproate initiation and renewed annually by healthcare professionals, and alert cards provided to women with epilepsy with childbearing potential. These could serve as references to healthcare providers in different clinical settings. Direct adoption of these measures may not be feasible in some local situations. Certain practices may be advisable to enhance communication between prescribers and potential childbearing female valproate users. The indications and potential adverse effects of valproate could be communicated to patients in different formats, such as on paper or electronically. The plan of valproate therapy should be reviewed with women with epilepsy with pregnancy potential when clinically indicated, but not necessarily on a yearly basis. Other medical professionals (eg, nurses or pharmacists) could take a role in AED counselling, including advising patients regarding their potential adverse effects. Materials could be provided to patients regarding the drugs’ potential implications on pregnancy and fetal outcomes. These could be in the form of leaflets, warning messages, or QR codes on drug packaging. Good documentation of the communication between the clinician and patient, preferably including the family, cannot be overemphasised.

The NAAPR reported that carbamazepine monotherapy during pregnancy had a 2.9% overall risk of MCMs.17 The European Surveillance of Congenital Anomalies Database reported that the OR of spina bifida related to carbamazepine monotherapy was 2.6 versus no AEDs.41 The EURAP has also demonstrated a dose-dependent effect of carbamazepine. Carbamazepine doses of >400 mg per day were associated with a significantly higher risk of MCMs (5.3% for ≥400 to <1000 mg; 8.7% for ≥1000 mg) [Table 1 17 19 21 42 43].

A 2008 meta-analysis showed that lamotrigine monotherapy had an MCM risk of 2.91% (95% CI, 2.00%-3.82%).14 The NAAPR has indicated that lamotrigine monotherapy carried a 1.9% risk of MCMs.44 The EURAP pregnancy registry data show a relatively low risk of malformations when using lamotrigine monotherapy at doses <300 mg per day as compared with other AEDs (Table 1).42 When used as polytherapy, the combination of lamotrigine and valproate may have a relatively higher risk of MCMs (9.1%; OR=5.0; 95% CI, 1.5%-14%) compared with combinations of lamotrigine and other AEDs (2.9%; OR=1.5; 95% CI, 0.7%-3.0%).44

The UK and Ireland Epilepsy and Pregnancy Register reported that levetiracetam monotherapy has a 0.70% risk of MCMs (95% CI, 0.1%-2.51%).21 The MCM risk of levetiracetam as polytherapy was relatively higher at 6.47% (95% CI, 4.31%-9.60%). The MCM rates of polytherapy including levetiracetam varied with different regimens: the combination of levetiracetam and lamotrigine had a risk of 1.77% (95% CI, 0.49%-6.22%) compared with 6.90% (95% CI, 1.91%-21.96%) for the combination of levetiracetam and valproate and 9.38% (95% CI, 4.37%-18.98%) for the combination of levetiracetam and carbamazepine.

The FDA issued a warning regarding an increased risk of oral cleft in children born to mothers taking topiramate based on data from the NAAPR and the UK Epilepsy and Pregnancy Register. The UK registry reported the risk of oral cleft as 29 per 1000 (95% CI, 5-91 per 1000), which is more than 10 times the background risk.22 The NAAPR reported that the risk of cleft lip associated with topiramate monotherapy during pregnancy was 14 per 1000 (95% CI, 5.1-31 per 1000).17 The overall MCM rate of topiramate monotherapy has been reported to be 4.2% (95% CI, 2.4%-6.8%).

Pregabalin use in human pregnancy is relatively less studied. As pregabalin is also used for controlling neuralgic pain, clinical studies on pregabalin may involve confounding factors, such as concomitant medications and co-morbidities, which may be different from those of patients with epilepsy only. This may cause confounding factors in clinical studies, as the concomitant medications and disease spectrum may be different from those of patients with epilepsy only. A multicentre study found that the risk of MCMs associated with pregabalin use in the first trimester of pregnancy was 6.0%, compared with an unexposed risk of 2.1% (OR=3.0; 95% CI, 1.2%-7.9%).45 However, another larger study did not find any significant association between pregabalin use in the first trimester and MCMs. The exposed risk was 5.9%, compared with an unexposed risk of 3.3% (OR=1.78; 95% CI, 1.26%-2.58%) and the adjusted OR was 1.16 (95% CI, 0.81-1.67) according to that study.43

Oxcarbazepine is similar to carbamazepine in terms of chemical structure. The NAAPR reported that oxcarbazepine use in pregnancy was associated with a 2.2% risk of MCMs (95% CI, 0.6%-5.5%).17 The EURAP registry data show a 3.0% MCM risk associated with in-utero oxcarbazepine exposure (95% CI, 1.4%-5.4%).19

Most women with epilepsy have uneventful pregnancies.2 However, they should be informed about their risks of pregnancy complications. For example, the risks of such complications as needing Caesarean section, spontaneous abortion, pre-eclampsia or pregnancy-induced hypertension, pregnancy-related bleeding complications, fetal growth restriction, and premature uterine contractions, labour, and delivery may be higher than those of women without epilepsy, especially if they are on AEDs.9 46 47 48 Seizure freedom for 9 months to 1 year prior to pregnancy is associated with a high likelihood of seizure freedom during pregnancy.2 9 49 50 51 Neither increased seizure frequency nor status epilepticus has been substantially associated with pregnancy.9 There is no evidence that focal seizures with or without impairment of consciousness, absence, or myoclonic seizures affect pregnancy or the developing fetus adversely unless a fall causes injury or other serious complications. Generalised tonic-clonic seizures may carry a relatively higher risk to the fetus, although their absolute risk level remains very low, and their risk may depend on seizure frequency.52 The majority of women with epilepsy experience no change in seizure frequency during pregnancy, whereas a minority of them experience either an increase or a decrease in attack frequency.14 53 The increase in seizure frequency may be caused by a pregnancy-induced drop in AED concentration, sleep deprivation, or lack of drug adherence. Women who contemplate going without AEDs during pregnancy should undergo discussions about the potential risk of inadvertent status epilepticus or sudden unexpected death in epilepsy.9 52 A balanced discussion may be held with the patient to facilitate informed decision making and provide access to various options. Early and serial fetal ultrasound scans should be offered to pregnant women who are taking AEDs to screen for fetal structural anomalies.2 54 Additional screening or diagnostic tests, such as maternal serum tests and fetal echocardiography, should proceed according to clinical indications.2 55 Pregnancy in women with epilepsy optimally involves collaborative, multidisciplinary planning and management. The multidisciplinary team should involve specialists, such as obstetricians and neurologists, who have experience providing care for pregnant women with epilepsy.56 Monitoring of AED levels during pregnancy may help to guide dosage adjustment. For example, both total and free clearance of lamotrigine may increase substantially during pregnancy, with a peak in the third trimester.57 58 A decreased lamotrigine level could result in increased seizure frequency. Pregnancy may also affect serum levels of carbamazepine, phenytoin, levetiracetam, and oxcarbazepine to various extents.7 59 60 61 Serum level monitoring may be helpful to inform dosage adjustment during pregnancy. In pregnant women presenting with seizures in the second half of pregnancy that cannot be clearly attributed to epilepsy alone, consideration is usually given to pre-eclampsia, in which case definitive diagnosis should be sought by further assessment.2

Fetal complications may occur at slightly higher frequencies in pregnancies of women with epilepsy. A systematic review and meta-analysis that included over 2.8 million pregnancies found that spontaneous miscarriage (OR=1.54; 95% CI, 1.02-2.32), preterm birth before 37 weeks of gestation (OR=1.16; 95% CI, 1.01-1.34), and fetal growth restriction (OR=1.26; 95% CI, 1.20-1.33) were more common in women with epilepsy.48 However, the risks of early preterm birth before 34 weeks, gestational diabetes, fetal death or stillbirth, perinatal death, or admission to neonatal intensive care unit did not differ between women with and without epilepsy. In another large US cohort study, the risks of preterm labour (OR=1.54; 95% CI, 1.50-1.57) and stillbirth (adjusted OR=1.27; 95% CI, 1.17-1.38) were also higher in pregnancies of women with epilepsy.47 A 1-minute Apgar score of <7 in the neonate may be associated with maternal AED use.12 62

The interactions between AEDs and oral contraceptive pills should be discussed with all women with epilepsy of childbearing age. Enzyme-inducing AEDs can hinder the effectiveness of oral contraceptives63 (Table 2 64 65 66 67 68 69 70 71 72 73), and women with epilepsy on enzyme-inducing AEDs should avoid combined oral contraceptive pills, combined contraceptive patches, progestogen-only pills, progestogen-only implants, and vaginal ring contraceptives.2 74 75 Older-generation AEDs such as phenytoin, carbamazepine, phenobarbital, and primidone may belong to the enzyme-inducing category, but some newer-generation AEDs such as oxcarbazepine also have enzyme-inducing properties.76 It may be advisable for women with epilepsy to use other contraceptive methods than the above-mentioned hormone-containing pills and devices because of the risk of contraception failure. If the use of oral contraceptives is unavoidable, hormonal contraception may be adjusted in women taking enzyme-inducing AEDs, although there is little evidence supporting this (Box 75 77). As different cases are unique, it is advisable to consult specialists who have experience managing these conditions. Copper intrauterine devices, the levonorgestrel-releasing intrauterine system, and medroxyprogesterone acetate injections are less affected by enzyme-inducing AEDs.2 Some opinions suggest more frequent injections of medroxyprogesterone acetate (every 10 weeks instead of every 12 weeks) in women with epilepsy taking enzyme-inducing AEDs because of potential interactions.64 For emergency contraception, a copper intrauterine device should be advised. If the intrauterine device is not suitable or acceptable, a double dose of a total of 3 mg levonorgestrel can be given.2 78 79 In contrast, serum lamotrigine levels can be reduced by simultaneous use of any oestrogen-based contraceptive, leading to deteriorated seizure control.2 When the concomitant use of the contraceptive is stopped, the lamotrigine dose may need to be adjusted.46

Spontaneous vaginal delivery is not absolutely contra-indicated in most women with epilepsy. Only about 1% to 2% of women with epilepsy develop generalised tonic-clonic seizures during labour.2 The EURAP registry reported seizure occurrence in 3.5% of women with epilepsy in labour.80 An epilepsy diagnosis per se is not an indication for elective Caesarean section or labour induction.2 Antiepileptic drugs should be continued orally or intravenously. Measures such as adequate analgesia, including transcutaneous electrical nerve stimulation, nitrous oxide and oxygen, and regional analgesia, are recommended to reduce pain and emotional distress, but they may trigger seizures.2 Pethidine, also known as meperidine, should be used with caution for analgesia during labour in women with epilepsy. Its metabolites may reduce the seizure threshold.81 Caesarean section may be necessary if seizures are frequent or prolonged.2 Seizures during labour should be terminated as quickly as possible to avoid maternal and fetal complications. Benzodiazepines are the drugs of choice,2 and phenytoin can also be given intravenously.82 Magnesium sulphate is a treatment option if a diagnosis of eclampsia is made.2 Some clinicians may prescribe vitamin K to women with epilepsy on enzyme-inducing AEDs in the last month of pregnancy to prevent haemorrhagic disease in the newborn. The usual dose is about 10 to 20 mg/day of oral vitamin K. There are insufficient data to fully support or refute maternal vitamin K prophylaxis.2 7 83 It has been recommended that newborns of women with epilepsy be given 1 mg of vitamin K1 parenterally at delivery.2

There is no absolute contra-indication to breastfeeding by women with epilepsy who are taking AEDs.2 84 However, mothers who take AEDs should be counselled about the potential risks of breastfeeding. Many AEDs are excreted in breast milk. The amount of drug absorbed by the infant depends on various factors, including the maternal plasma concentration, the degree of drug transfer to breast milk, and the amount of breast milk consumed by the infant. The degree of drug transfer to breast milk is inversely dependent on the drug’s protein binding ability.76 Primidone, levetiracetam, barbiturates, benzodiazepines, lamotrigine, gabapentin, topiramate, ethosuximide, and zonisamide probably have significant penetration into breast milk.7 76 85 86 However, there is no evidence to show that indirect AED exposure through breastfeeding has clinically significant effects on offspring.2 87 88 The potential risks of breastfeeding should be balanced by the benefits of breastfeeding for both the neonate and the mother.84

Women with epilepsy who have just given birth might face significant anxiety concerning the risk of accidents should they develop breakthrough seizures when they take care of their infants. While there is risk of injury to the infant, reassurance should be provided to the parents, as the actual risk level is probably low.89 Safety precautions that are simple to implement could further reduce such risks significantly.2 Mothers should breastfeed their babies while sitting on floor cushions to avoid dropping their babies should a seizure occur. If there is an aura, the mother should lay the baby down.2 The mother bathing the baby in a bathtub by herself could potentially carry some drowning risk if a seizure occurs. The same risk also applies to the mother carrying the baby single-handedly.46 Supervision by the partner or a caregiver may help to prevent injury to the baby should a seizure occur.2 Information may also be given to parents early to facilitate preparation or home modification to provide an optimal environment to care for the newborn. Seizure frequency may be exacerbated during the postpartum period for reasons such as lack of sleep and drug non-compliance.90 Serial serum AED level monitoring of the mother may be useful, as there may be physiological changes during the puerperal period, especially with certain agents such as lamotrigine.57 91 Maintaining the serum drug level at approximately the pre-conception level is advisable, provided that the woman had good seizure control before the pregnancy.51 92 Dosages of AEDs should be adjusted accordingly to avoid postpartum toxicity or deteriorated seizure control.2

With advancements in the field of reproduction technology, clinicians may receive questions on such technologies from women with epilepsy. Reproductive intervention may have implications for women with epilepsy and their offspring. For example, ovarian stimulation with gonadotropin may induce seizure exacerbation in women with epilepsy.93 Epilepsy is a heterogeneous disease entity, and although genetics play a part in its aetiology, it can have implications for many bodily functions. Single gene (ie, Mendelian) disorders account for only a small proportion of patients. The implications of these genetic factors in the offspring’s life and development are often variable and complicated by many different factors. Reproductive options and application of reproductive technology, such as pre-implantation diagnosis, remain controversial. Whenever possible, cases should be referred to appropriate specialists, including geneticists and obstetricians, for proper counselling.

Hormonal profile changes can affect seizure control in women with epilepsy.94 Clinicians should remind patients that hormone replacement therapy can significantly increase seizure frequency during menopause, particularly in women with catamenial epilepsy.95 Hormone replacement therapy (which entails the administration of exogenous female sex hormones) and contraceptive pills probably have similar drug interactions with AEDs.96

Epileptic seizures may be triggered by serum hormonal changes. Catamenial epilepsy is characterised by an increased seizure frequency during certain phase(s) of the menstrual cycle compared with baseline. However, there is no uniform definition for it. The seizures can cluster in different phases, such as the perimenstrual, periovulatory, and luteal phases of the menstrual cycle.97 Recognition of the catamenial seizure pattern can be challenging, as irregular menstrual cycles and anovulatory cycles are common. Women whose seizure occurrence is suspected to be related to the menstrual cycle should be advised to record the seizure attacks together with the phase of the menstrual cycle in their seizure diaries. Prolonged observation may be needed before a catamenial epilepsy pattern is recognised. Treatments include intermittent benzodiazepines (eg, clobazam) during the susceptible phase of the menstrual cycle.98 99 Increasing the anticonvulsant dosage during susceptible periods of the menstrual cycle may also be feasible. Hormonal therapy has not yet been proven to be effective in general.100 It may be beneficial in women with more frequent seizures around the perimenstrual period, especially in women with regular menstrual cycles.101 102 However, further studies are probably needed before routine clinical use can be suggested.100 Acetazolamide, used either daily or perimenstrually, may have benefits for catamenial epilepsy, although the evidence to support its use is limited.93 94 103 104

New information has accumulated recently, providing new insight into the care of women with epilepsy and the implications for their offspring (Table 3). Thorough discussions with women with epilepsy may increase patients’ levels of comfort and care, especially regarding major treatment decision on the issues of drug teratogenicity, neurobehavioral adverse effects on offspring, pregnancy, breastfeeding, contraception, reproduction technology, menopause, and catamenial epilepsy.

Concept or design: All authors.
Acquisition of data: RSK Chang.
Analysis and interpretation of data: RSK Chang.
Drafting of the manuscript: RSK Chang, KHK Lui, AWY Yung.
Critical revision of the manuscript for important intellectual content: RSK Chang, H Leung.

All authors had full access to the data, contributed to the study, approved the final version for publication and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

This project was supported in part by an unrestricted grant from the Hong Kong Epilepsy Society. All authors are members of the Hong Kong Epilepsy Society.

This consensus statement is designed to assist clinicians by providing an analytical framework for treatment of women with epilepsy. It is not intended to establish a community standard of care, replace a clinician’s medical judgement, or establish a protocol for all patients.

1. Guideline Development Group, Hong Kong Epilepsy Society. The Hong Kong Epilepsy Guideline 2009. Hong Kong Med J 2009;15 Suppl 5:6-28.

2. Royal College of Obstetricians and Gynaecologists. Epilepsy in Pregnancy (Green-top Guideline No. 68). Available from: https://www.rcog.org.uk/en/guidelines-research-services/guidelines/gtg68/. Accessed 19 Jul 2020.

3. Ottman R, Hirose S, Jain S, et al. Genetic testing in the epilepsies—report of the ILAE Genetics Commission. Epilepsia 2010;51:655-70. Crossref

4. Peljto AL, Barker-Cummings C, Vasoli VM, et al. Familial risk of epilepsy: a population-based study. Brain 2014;137:795-805. Crossref

5. Billings RE. Decreased hepatic 5, 10-methylenetetrahydrofolate reductase activity in mice after chronic phenytoin treatment. Mol Pharmacol 1984;25:459-66.

6. Kaaja E, Kaaja R, Hiilesmaa V. Major malformations in offspring of women with epilepsy. Neurology 2003;60:575-9. Crossref

7. Harden CL, Pennell PB, Koppel BS, et al. Management issues for women with epilepsy—focus on pregnancy (an evidence-based review): III. Vitamin K, folic acid, blood levels, and breast-feeding: report of the Quality Standards Subcommittee and Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology and the American Epilepsy Society. Epilepsia 2009;50:1247-55. Crossref

8. Betts T, Fox C. Proactive pre-conception counselling for women with epilepsy—is it effective? Seizure 1999;8:322-7. Crossref

9. Harden CL, Hopp J, Ting TY, et al. Management issues for women with epilepsy-focus on pregnancy (an evidence-based review): I. Obstetrical complications and change in seizure frequency: report of the Quality Standards Subcommittee and Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology and the American Epilepsy Society. Epilepsia 2009;50:1229-36. Crossref

10. Shorvon SD, Tomson T, Cock HR. The management of epilepsy during pregnancy—progress is painfully slow. Epilepsia 2009;50:973-4. Crossref

11. Fried S, Kozer E, Nulman I, Einarson TR, Koren G. Malformation rates in children of women with untreated epilepsy: a meta-analysis. Drug Saf 2004;27:197-202. Crossref

12. Harden CL, Meador KJ, Pennell PB, et al. Management issues for women with epilepsy-focus on pregnancy (an evidence-based review): II. Teratogenesis and perinatal outcomes: report of the Quality Standards Subcommittee and Therapeutics and Technology Subcommittee of the American Academy of Neurology and the American Epilepsy Society. Epilepsia 2009;50:1237-46. Crossref

13. Morrow J, Russell A, Guthrie E, et al. Malformation risks of antiepileptic drugs in pregnancy: a prospective study from the UK Epilepsy and Pregnancy Register. J Neurol Neurosurg Psychiatry 2006;77:193-8. Crossref

14. Meador K, Reynolds MW, Crean S, Fahrbach K, Probst C. Pregnancy outcomes in women with epilepsy: a systematic review and meta-analysis of published pregnancy registries and cohorts. Epilepsy Res 2008;81:1-13. Crossref

15. Tomson T, Marson A, Boon P, et al. Valproate in the treatment of epilepsy in girls and women of childbearing potential. Epilepsia 2015;56:1006-19. Crossref

16. Holmes LB, Wyszynski DF, Lieberman E. The AED (antiepileptic drug) pregnancy registry: a 6-year experience. Arch Neurol 2004;61:673-8. Crossref

17. Hernández-Díaz S, Smith CR, Shen A, et al. Comparative safety of antiepileptic drugs during pregnancy. Neurology 2012;78:1692-9. Crossref

18. Reinisch JM, Sanders SA, Mortensen EL, Rubin DB. In utero exposure to phenobarbital and intelligence deficits in adult men. JAMA 1995;274:1518-25. Crossref

19. Tomson T, Battino D, Bonizzoni E, et al. Comparative risk of major congenital malformations with eight different antiepileptic drugs: a prospective cohort study of the EURAP registry. Lancet Neurol 2018;17:530-8. Crossref

20. Campbell E, Kennedy F, Russell A, et al. Malformation risks of antiepileptic drug monotherapies in pregnancy: updated results from the UK and Ireland Epilepsy and Pregnancy Registers. J Neurol Neurosurg Psychiatry 2014;85:1029-34. Crossref

21. Mawhinney E, Craig J, Morrow J, et al. Levetiracetam in pregnancy: results from the UK and Ireland epilepsy and pregnancy registers. Neurology 2013;80:400-5. Crossref

22. Hunt S, Russell A, Smithson WH, et al. Topiramate in pregnancy: preliminary experience from the UK Epilepsy and Pregnancy Register. Neurology 2008;71:272-6. Crossref

23. Vanoverloop D, Schnell RR, Harvey EA, Holmes LB. The effects of prenatal exposure to phenytoin and other anticonvulsants on intellectual function at 4 to 8 years of age. Neurotoxicol Teratol 1992;14:329-35. Crossref

24. Scolnik D, Nulman I, Rovet J, et al. Neurodevelopment of children exposed in utero to phenytoin and carbamazepine monotherapy. JAMA 1994;271:767-70. Crossref

25. Wide K, Henning E, Tomson T, Winbladh B. Psychomotor development in preschool children exposed to antiepileptic drugs in utero. Acta Paediatr 2002;91:409-14. Crossref

26. Arpino C, Brescianini S, Robert E, et al. Teratogenic effects of antiepileptic drugs: use of an International Database on Malformations and Drug Exposure (MADRE). Epilepsia 2000;41:1436-43. Crossref

27. Bertollini R, Mastroiacovo P, Segni G. Maternal epilepsy and birth defects: a case-control study in the Italian Multicentric Registry of Birth Defects (IPIMC). Eur J Epidemiol 1985;1:67-72. Crossref

28. Artama M, Auvinen A, Raudaskoski T, Isojärvi I, Isojärvi J. Antiepileptic drug use of women with epilepsy and congenital malformations in offspring. Neurology 2005;64:1874-8. Crossref

29. Samrén EB, van Duijn CM, Christiaens GC, Hofman A, Lindhout D. Antiepileptic drug regimens and major congenital abnormalities in the offspring. Ann Neurol 1999;46:739-46. Crossref

30. Jentink J, Loane MA, Dolk H, et al. Valproic acid monotherapy in pregnancy and major congenital malformations. N Engl J Med 2010;362:2185-93. Crossref

31. Meador KJ, Baker GA, Browning N, et al. Cognitive function at 3 years of age after fetal exposure to antiepileptic drugs. N Engl J Med 2009;360:1597-605. Crossref

32. Meador KJ, Baker GA, Browning N, et al. Fetal antiepileptic drug exposure and cognitive outcomes at age 6 years (NEAD study): a prospective observational study. Lancet Neurol 2013;12:244-52. Crossref

33. Meador KJ, Baker GA, Browning N, et al. Effects of fetal antiepileptic drug exposure: outcomes at age 4.5 years. Neurology 2012;78:1207-14. Crossref

34. Gaily E, Kantola-Sorsa E, Hiilesmaa V, et al. Normal intelligence in children with prenatal exposure to carbamazepine. Neurology 2004;62:28-32. Crossref

35. Bromley R, Weston J, Adab N, et al. Treatment for epilepsy in pregnancy: neurodevelopmental outcomes in the child. Cochrane Database Syst Rev 2014;(10):CD010236. Crossref

36. Adab N, Jacoby A, Smith D, Chadwick D. Additional educational needs in children born to mothers with epilepsy. J Neurol Neurosurg Psychiatry 2001;70:15-21. Crossref

37. Adab N, Kini U, Vinten J, et al. The longer term outcome of children born to mothers with epilepsy. J Neurol Neurosurg Psychiatry 2004;75:1575-83. Crossref

38. Bromley RL, Mawer GE, Briggs M, et al. The prevalence of neurodevelopmental disorders in children prenatally exposed to antiepileptic drugs. J Neurol Neurosurg Psychiatry 2013;84:637-43. Crossref

39. Christensen J, Grønborg TK, Sørensen MJ, et al. Prenatal valproate exposure and risk of autism spectrum disorders and childhood autism. JAMA 2013;309:1696-703. Crossref

40. Harden CL, Meador KJ, Pennell PB, et al. Practice parameter update: management issues for women with epilepsy—focus on pregnancy (an evidence-based review): teratogenesis and perinatal outcomes: report of the Quality Standards Subcommittee and Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology and American Epilepsy Society. Neurology 2009;73:133-41. Crossref

41. Jentink J, Dolk H, Loane MA, et al. Intrauterine exposure to carbamazepine and specific congenital malformations: systematic review and case-control study. BMJ 2010;341:c6581. Crossref

42. Tomson T, Battino D, Bonizzoni E, et al. Dose-dependent risk of malformations with antiepileptic drugs: an analysis of data from the EURAP epilepsy and pregnancy registry. Lancet Neurol 2011;10:609-17. Crossref

43. Patorno E, Bateman BT, Huybrechts KF, et al. Pregabalin use early in pregnancy and the risk of major congenital malformations. Neurology 2017;88:2020-5. Crossref

44. Holmes LB, Mittendorf R, Shen A, Smith CR, Hernandez- Diaz S. Fetal effects of anticonvulsant polytherapies: different risks from different drug combinations. Arch Neurol 2011;68:1275-81. Crossref

45. Winterfeld U, Merlob P, Baud D, et al. Pregnancy outcome following maternal exposure to pregabalin may call for concern. Neurology 2016;86:2251-7. Crossref

46. Datta S, Muñoz-Largacha JA, Li L, Zhao GQ, Litle VR. Subcutaneous metastases from early stage esophageal adenocarcinoma case report. Int J Surg Case Rep 2016;29:108-12. Crossref

47. MacDonald SC, Bateman BT, McElrath TF, Hernández-Diáz S. Mortality and morbidity during delivery hospitalization among pregnant women with epilepsy in the United States. JAMA Neurol 2015;72:981-8. Crossref

48. Viale L, Allotey J, Cheong-See F, et al. Epilepsy in pregnancy and reproductive outcomes: a systematic review and meta-analysis. Lancet 2015;386:1845-52. Crossref

49. Vajda FJ, Hitchcock A, Graham J, O’Brien T, Lander C, Eadie M. Seizure control in antiepileptic drug-treated pregnancy. Epilepsia 2008;49:172-6. Crossref

50. Gjerde IO, Strandjord RE, Ulstein M. The course of epilepsy during pregnancy: a study of 78 cases. Acta Neurol Scand 1988;78:198-205. Crossref

51. Harden CL, Hopp J, Ting TY, et al. Practice parameter update: management issues for women with epilepsy—focus on pregnancy (an evidence-based review): obstetrical complications and change in seizure frequency: report of the Quality Standards Subcommittee and Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology and American Epilepsy Society. Neurology 2009;73:126-32. Crossref

52. Pennell PB. Antiepileptic drugs during pregnancy: what is known and which AEDs seem to be safest? Epilepsia 2008;49 Suppl 9:43-55. Crossref

53. European Registry of Antiepileptic Drugs in Pregnancy (EURAP). Seizure control and treatment in pregnancy: observations from the EURAP epilepsy pregnancy registry. Neurology. 2006;66:354-60. Crossref

54. National Institute for Health and Care Excellence, UK Government. Epilepsies: diagnosis and management. Clinical guideline [CG137]. 2013. Available from: https://www.nice.org.uk/guidance/cg137/chapter/1- Guidance#women-and-girls-with-epilepsy. Accessed 19 Jul 2020.

55. International Society of Ultrasound in Obstetrics and Gynecology, Carvalho JS, et al. ISUOG Practice Guidelines (updated): sonographic screening examination of the fetal heart. Ultrasound Obstet Gynecol 2013;41:348-59. Crossref

56. Bhatia M, Adcock JE, Mackillop L. The management of pregnant women with epilepsy: a multidisciplinary collaborative approach to care. The Obstet Gynaecol 2017;19:279-88. Crossref

57. de Haan GJ, Edelbroek P, Segers J, et al. Gestation-induced changes in lamotrigine pharmacokinetics: a monotherapy study. Neurology 2004;63:571-3. Crossref

58. Pennell PB, Peng L, Newport DJ, et al. Lamotrigine in pregnancy: clearance, therapeutic drug monitoring, and seizure frequency. Neurology 2008;70:2130-6. Crossref

59. Pennell PB. Antiepileptic drug pharmacokinetics during pregnancy and lactation. Neurology 2003;61:S35-42. Crossref

60. Westin AA, Nakken KO, Johannessen SI, Reimers A, Lillestølen KM, Brodtkorb E. Serum concentration/dose ratio of topiramate during pregnancy. Epilepsia 2009;50:480-5. Crossref

61. Reisinger TL, Newman M, Loring DW, Pennell PB, Meador KJ. Antiepileptic drug clearance and seizure frequency during pregnancy in women with epilepsy. Epilepsy Behav 2013;29:13-8. Crossref

62. Viinikainen K, Heinonen S, Eriksson K, Kälviäinen R. Community-based, prospective, controlled study of obstetric and neonatal outcome of 179 pregnancies in women with epilepsy. Epilepsia 2006;47:186-92. Crossref

63. World Health Organization. Epilepsy: a manual for physicians. 2004. Available from: https://apps.who.int/ iris/bitstream/handle/10665/205014/B0769.pdf Accessed 19 Jul 2020.

64. Reddy DS. Clinical pharmacokinetic interactions between antiepileptic drugs and hormonal contraceptives. Expert Rev Clin Pharmacol 2010;3:183-92. Crossref

65. Perucca E. Clinically relevant drug interactions with antiepileptic drugs. Br J Clin Pharmacol 2006;61:246-55. Crossref

66. Tolou-Ghamari Z, Zare M, Habibabadi JM, Najafi MR. A quick review of carbamazepine pharmacokinetics in epilepsy from 1953 to 2012. J Res Med Sci 2013;18(Suppl 1):S81-5.

67. Sabers A, Ohman I, Christensen J, Tomson T. Oral contraceptives reduce lamotrigine plasma levels. Neurology 2003;61:570-1. Crossref

68. Patsalos PN. The clinical pharmacology profile of the new antiepileptic drug perampanel: a novel noncompetitive AMPA receptor antagonist. Epilepsia 2015;56:12-27. Crossref

69. Rosenfeld WE, Doose DR, Walker SA, Nayak RK. Effect of topiramate on the pharmacokinetics of an oral contraceptive containing norethindrone and ethinyl estradiol in patients with epilepsy. Epilepsia 1997;38:317-23. Crossref

70. Eldon MA, Underwood BA, Randinitis EJ, Sedman AJ. Gabapentin does not interact with a contraceptive regimen of norethindrone acetate and ethinyl estradiol. Neurology 1998;50:1146-8. Crossref

71. Cawello W, Rosenkranz B, Schmid B, Wierich W. Pharmacodynamic and pharmacokinetic evaluation of coadministration of lacosamide and an oral contraceptive (levonorgestrel plus ethinylestradiol) in healthy female volunteers. Epilepsia 2013;54:530-6. Crossref

72. Cawello W. Clinical pharmacokinetic and pharmacodynamic profile of lacosamide. Clin Pharmacokinet 2015;54:901- 14. Crossref

73. Ragueneau-Majlessi I, Levy RH, Janik F. Levetiracetam does not alter the pharmacokinetics of an oral contraceptive in healthy women. Epilepsia 2002;43:697-702. Crossref

74. Curtis KM, Tepper NK, Jatlaoui TC, et al. U.S. medical eligibility criteria for contraceptive use, 2016. MMWR Recomm Rep 2016;65:1-103. Crossref

75. Gooneratne IK, Wimalaratna M, Ranaweera AK, Wimalaratna S. Contraception advice for women with epilepsy. BMJ 2017;357:j2010. Crossref

76. Pennell PB, Gidal BE, Sabers A, Gordon J, Perucca E. Pharmacology of antiepileptic drugs during pregnancy and lactation. Epilepsy Behav 2007;11:263-9. Crossref

77. The Faculty of Sexual & Reproductive Healthcare of the Royal College of Obstetricians & Gynaecologists. FSRH CEU guidance: drug interactions with hormonal contraception. Available from: https://www.fsrh.org/standards-and-guidance/documents/ceu-clinical-guidance-drug-interactions-with-hormonal/. Accessed 11 Oct 2018.

78. The Faculty of Sexual & Reproductive Healthcare of the Royal College of Obstetricians & Gynaecologists. FSRH Guideline Emergency Contraception. Available from: https://www.fsrh.org/standards-and-guidance/documents/ceu-clinical-guidance-emergency-contraception-march-2017/. Accessed 10 Oct 18.

79. Black KI, Hussainy SY. Emergency contraception: oral and intrauterine options. Aust Fam Physician 2017;46:722-6.

80. Pennell PB. EURAP outcomes for seizure control during pregnancy: useful and encouraging data. Epilepsy Curr 2006;6:186-8. Crossref

81. Marinella MA. Meperidine-induced generalized seizures with normal renal function. South Med J 1997;90:556-8. Crossref

82. ACOG educational bulletin. Seizure disorders in pregnancy. Number 231, December 1996. Committee on Educational Bulletins of the American College of Obstetricians and Gynecologists [editorial]. Int J Gynaecol Obstet 1997;56:279-86.

83. Yamasmit W, Chaithongwongwatthana S, Tolosa JE. Prenatal vitamin K1 administration in epileptic women to prevent neonatal hemorrhage: is it effective? J Reprod Med 2006;51:463-6.

84. Veiby G, Bjørk M, Engelsen BA, Gilhus NE. Epilepsy and recommendations for breastfeeding. Seizure 2015;28:57-65. Crossref

85. Ohman I, Vitols S, Tomson T. Lamotrigine in pregnancy: pharmacokinetics during delivery, in the neonate, and during lactation. Epilepsia 2000;41:709-13. Crossref

86. Davanzo R, Dal Bo S, Bua J, Copertino M, Zanelli E, Matarazzo L. Antiepileptic drugs and breastfeeding. Ital J Pediatr 2013;39:50. Crossref

87. Meador KJ, Baker GA, Browning N, et al. Effects of breastfeeding in children of women taking antiepileptic drugs. Neurology 2010;75:1954-60. Crossref

88. Meador KJ, Baker GA, Browning N, et al. Breastfeeding in children of women taking antiepileptic drugs: cognitive outcomes at age 6 years. JAMA Pediatr 2014;168:729-36. Crossref

89. Fox C, Betts T. How much risk does a woman with active epilepsy pose to her newborn child in the puerperium? A pilot study. Seizure 1999;8:367-9. Crossref

90. Thomas SV, Syam U, Devi JS. Predictors of seizures during pregnancy in women with epilepsy. Epilepsia 2012;53:e85-8. Crossref

91. Tran TA, Leppik IE, Blesi K, Sathanandan ST, Remmel R. Lamotrigine clearance during pregnancy. Neurology 2002;59:251-5. Crossref

92. Tomson T, Landmark CJ, Battino D. Antiepileptic drug treatment in pregnancy: changes in drug disposition and their clinical implications. Epilepsia 2013;54:405-14. Crossref

93. Mostacci B, Esposto R, Lello S, Bisulli F, Licchetta L, Tinuper P. Estrogen-related seizure exacerbation following hormone therapy for assisted reproduction in women with epilepsy. Seizure 2018;61:200-2. Crossref

94. Taubøll E, Sveberg L, Svalheim S. Interactions between hormones and epilepsy. Seizure 2015;28:3-11. Crossref

95. Harden CL, Pulver MC, Ravdin L, Jacobs AR. The effect of menopause and perimenopause on the course of epilepsy. Epilepsia 1999;40:1402-7. Crossref

96. Reimers A. Hormone replacement therapy with estrogens may reduce lamotrigine serum concentrations: a matched case-control study. Epilepsia 2017;58:e6-9. Crossref

97. Herzog AG. Catamenial epilepsy: definition, prevalence pathophysiology and treatment. Seizure 2008;17:151-9. Crossref

98. Feely M, Gibson J. Intermittent clobazam for catamenial epilepsy: tolerance avoided. J Neurol Neurosurg Psychiatry 1984;47:1279-82. Crossref

99. Foldvary-Schaefer N, Falcone T. Catamenial epilepsy: pathophysiology, diagnosis, and management. Neurology 2003;61(6 Suppl 2):S2-15. Crossref

100. Herzog AG, Fowler KM, Smithson SD, et al. Progesterone vs placebo therapy for women with epilepsy: a randomized clinical trial. Neurology 2012;78:1959-66. Crossref

101. Herzog AG. Catamenial epilepsy: update on prevalence, pathophysiology and treatment from the findings of the NIH Progesterone Treatment Trial. Seizure 2015;28:18-25. Crossref

102. Navis A, Harden C. A treatment approach to catamenial epilepsy. Curr Treat Options Neurol 2016;18:30. Crossref

103. Reddy DS, Gould J, Gangisetty O. A mouse kindling model of perimenstrual catamenial epilepsy. J Pharmacol Exp Ther 2012;341:784-93. Crossref

104. Lim LL, Foldvary N, Mascha E, Lee J. Acetazolamide in women with catamenial epilepsy. Epilepsia 2001;42:746-9. Crossref

Clinical blood transfusion remains an essential and irreplaceable part of modern medicine, either as an independent therapeutic modality or an additional support to other clinical therapies. Anaemia, a serious disease with a worldwide burden on both hospitalised patients and society,1 2 3 is often managed with blood transfusion as part of the treatment. Without a reliable substitute, sourcing of the blood used in transfusion relies solely on donations from voluntary, non-remunerated blood donors. Because blood is a biological substance, it is impossible to completely eliminate adverse outcomes during and after transfusion. Worldwide, particularly in developed countries, the ageing of the population and emerging infectious diseases are the two most important and ongoing threats to the sustainability of the safe blood supply. Ageing populations tend to have increased numbers of complex surgeries and cancer treatments requiring increased blood transfusions.4 In 2016, the mean per capital blood use in high-income countries was 32 units of red cell components per 1000 population. Moreover, the most frequently transfused patient group is aged >60 years, accounting for up to 79% among these transfusions.5 Infectious pathogens continue to emerge rapidly, which could adversely affect transfusion safety both directly (if the pathogen is transmitted through blood transfusion) and indirectly (if outbreaks reduce the pool of available donors).6 Recent examples include the Zika virus outbreak in South America and the dengue, hepatitis E, and chikungunya virus outbreaks in Southeast Asia.7 Therefore, maintenance of a sustainable and safe blood supply continues to be a challenging task that requires a substantial amount of effort and resources.

There is evidence associating blood transfusion with less favourable clinical outcomes.8 9 This evidence includes a higher incidence of recurrence in cancer surgeries, higher operative mortality, failure to rescue from sepsis, and other serious complications like renal, neurological, cardiac, and pulmonary dysfunction.10 Patient blood management (PBM) is a patient-centred and multidisciplinary framework that has been rapidly developing throughout the last decade in Western countries to improve the treatment outcomes of patients who may need blood transfusions during their treatment. It refers to evidence-based medical

and surgical concepts designed to improve patient outcomes. Commonly, PBM employs a three-pillar approach: (1) optimise red blood cell mass; (2) minimise blood loss; and (3) manage anaemia (Table 111 12 13 14). Indeed, PBM optimises the patient’s condition before, during, and after the procedure and only recommends transfusion when indicated. It directly addresses the triad of independent risk factors that can affect patient outcomes: anaemia, blood loss, and transfusion. Anaemia is appropriately and timely managed according to its aetiology instead of being bluntly corrected by blood transfusion. Thereby, blood loss is minimised, and the harm associated with inappropriate transfusions is avoided.15 Therefore, countries that have implemented PBM have shown improvement of patients’ outcomes, such as overall survival, disease recurrence, infection rate, length of stay in intensive care unit and hospital, cost, and blood utilisation.16 17 18 19 20 21 22 During the Sixty-third World Health Assembly in 2010, member states were urged to establish or strengthen systems for the safe and rational use of blood products and to provide training for all staff involved in clinical transfusion, to implement potential solutions to minimise transfusion errors and promote patient safety, and to promote the availability of transfusion alternatives including, where appropriate, autologous transfusion and PBM.23

Although Hong Kong has a long history of self-sufficiency in terms of blood supply, population ageing has brought a significant increase in demand for blood over the last decade.24 Since 2015, the Hong Kong Blood Transfusion Service has experienced excessive difficulties at mobilising citizens to maintain a stable, safe blood supply. As a result, the Blood Transfusion Service faces a number of blood safety and donor issues that affect the blood supply. Emerging infectious diseases like Zika virus and low pre-donation haemoglobin due to iron deficiency are typical examples that may prevent apparently healthy persons from donating blood.

As blood is irreplaceable and has a limited shelf life, securing a sustainable and safe blood supply is of paramount importance in the modern healthcare system to ensure that patients’ transfusion needs are met promptly and appropriately. Strategies to enhance new donor recruitment and existing donor retention should be undertaken by the Blood Transfusion Service to increase the blood supply. However, demand control measures should be simultaneously implemented to reduce the pressure on the supply side. In Hong Kong, there has been increasing awareness of the concept of PBM beginning in the past 2 years, and small-scale projects have been initiated. One local project was able to increase the preoperative haemoglobin concentration and reduce the transfusion rate after implementation of PBM.25 In the UK, the National Institute for Health and Care Excellence recommended consideration of single-unit transfusions for adults without active bleeding in November 2015.26 On the basis of this recommendation, some medical departments in Hong Kong have implemented single-unit blood transfusions over the past 2 years, and unpublished audit results demonstrate an overall reduction of red blood cell transfusions in general medical in-patients over that period.

With the objective of improving patients’ outcomes and better managing transfusion demand, a group of experienced clinicians from different specialties and hospitals in Hong Kong has established the Hong Kong Society of Clinical Blood Management to continuously promote PBM in Hong Kong. The Society aims to discuss and make recommendations regarding the implementation of PBM in Hong Kong. Below are three areas of focus that the Society intends to address.

Anaemia is a serious disease burden in both hospitalised patients and society.1 2 3 Haematopoiesis and anaemia management are important modifiable risk factors for adverse outcomes.9 27 28 Beneficial outcomes in this important pillar of PBM are seen in not only surgical patients but also other patient groups, such as those with underlying medical, obstetric, or gynaecological problems. Therefore, clinical guidelines have recommended that anaemia be promptly recognised and the underlying causes identified and managed appropriately.29 Because some surgical patients have an increased risk of bleeding, haemoglobin measurement well before operation in all patients could provide adequate time to manage any anaemia before surgery and improve outcomes.29

Red blood cell transfusion should be restricted to the minimal amount necessary to achieve clinical stability and to patients presenting with severe iron deficiency anaemia and alarming symptoms (eg, haemodynamic instability) and/or risk criteria (eg, coronary heart disease).30 31 As iron deficiency (whether absolute or functional) is commonly found in anaemic patients, its correction should be promptly instituted. Oral iron supplements, provided as ferrous or ferric salts, are usually the first line of treatment for uncomplicated iron deficiency anaemia because of their availability, ease of administration, and relatively low cost. However, because of these supplements’ notorious gastrointestinal adverse effects, intravenous iron should be considered in patients with intolerance to oral iron and when more rapid restoration of the iron store is expected. ‘Newer’ intravenous iron formulations with safer profiles, such as ferric carboxymaltose or iron isomaltoside, which allow for a short-time (15-60 min) infusion of high iron doses (≥1000 mg), are now available for use in both in-patients and out-patients. Such intravenous iron formulations can rapidly correct iron deficiency and anaemia within a few weeks (vs the few months needed for correction via oral iron). At the University Hospitals Plymouth, UK, intravenous iron is given to successfully treat iron deficiency anaemia when surgery with anticipated blood loss of >500 mL is anticipated within 6 weeks.32 As a result, intravenous iron has become an important component of PBM management strategies.

In Hong Kong, the Blood Transfusion Service and the Hong Kong Medical Association have recently issued a simple algorithm to aid general practitioners with early and prompt recognition of anaemia and its management.33 Further work is required to enhance the general population’s awareness of anaemia and iron deficiency issues, their diagnosis, and improving their management.

Reducing or minimising blood loss in hospitalised patients is another approach to reduce the need for blood transfusion and improve patients’ outcomes. Some might consider that this type of planning should only occur in surgical or operative settings, but reducing iatrogenic blood loss in non-operative settings has also been shown to improve patients’ outcomes. Table 234 35 36 37 38 39 40 41 42 43 highlights the measures that have been shown to be effective at minimising blood loss. These measures are safe and have not affected organ function or caused other complications. Instead, they reduce iatrogenic blood loss and avoid blood transfusions.

Temporary cessation of antiplatelet and anticoagulant medications in the perioperative period may lead to reduced blood loss and transfusion requirements if the risks of perioperative thromboembolic events and bleeding are balanced. Meticulous surgical techniques such as performing minimally invasive surgery, judicious use of electrocautery, tourniquets, topical haemostatic agents, and intra-operative blood salvage can minimise surgery-related blood loss.44 45 46 47 48 49

A number of anaesthetic techniques can also help to reduce blood loss. Permissive hypotension refers to the lowering of mean arterial pressure to values between 50 and 65 mm Hg with the goal of reducing blood flow to the surgical field, thereby reducing blood loss and improving visibility in the surgical field.50 Studies have shown that permissive hypotension during anaesthesia reduced blood loss in spinal surgery, radical prostatectomy, functional endoscopic sinus surgery, and orthopaedic surgery.51 52 53 It can also reduce blood loss and blood product utilisation in adult trauma patients with haemorrhagic shock.54 Organ hypoperfusion is the major drawback, and therefore, this strategy may not be suitable for patients with coronary artery disease, cerebrovascular disease, traumatic brain injury, or spinal injury.

Prevention of perioperative hypothermia is another strategy that can help to reduce blood loss. Hypothermia is defined as a core temperature <36°C and is a common consequence of anaesthesia.55 Even mild hypothermia, defined as a core temperature

between 35°C and 36°C, significantly increases perioperative blood loss and augments the transfusion requirement.56 Therefore, measures should be taken to prevent inadvertent hypothermia, including identification of high-risk patients, pre-warming before surgery, intra-operative monitoring of body temperature, using warm intravenous/irrigation fluid and forced-air warming devices, and avoidance of unnecessary body exposure.57

Another method to minimise blood loss is acute normovolaemic haemodilution. Acute normovolaemic haemodilution involves withdrawal of whole blood with concurrent infusion of fluids to maintain normovolaemia.58 The autologous blood is re-infused at the conclusion of the surgery. This method has been shown to significantly reduce the incidence and volume of allogeneic blood transfusion, and its use should be considered in adult patients who undergo surgery in which substantial blood loss is anticipated.44 However, relatively profound anaemia is expected during the surgery, which may induce tissue ischaemia, particularly in the myocardium.45 Furthermore, the effects of normovolaemic haemodilution on morbidity and mortality are uncertain.

Appropriate patient positioning during the intra-operative period may also help to reduce surgery-related blood loss. Elevation of the surgical site above the right atrium facilitates venous return and reduces venous engorgement. For example, the reverse Trendelenburg position has been shown to reduce intra-operative blood loss in endoscopic sinus surgery.46

Using the wide pad support widths of the Wilson frame, when compared with narrow pad support widths, significantly decreased intra-abdominal pressure and intra-operative blood loss in patients undergoing spine surgery in the prone position.47 Pharmacological agents can also be used to facilitate haemostasis. Tranexamic acid has been studied extensively in a wide range of surgeries and has been shown to reduce blood loss effectively without increasing the risk of thromboembolic events.48 49 In case of significant haemorrhage that is refractory to standard treatment, the use of recombinant factor VIIa should also be considered.59

Diagnostic phlebotomy for laboratory testing can also be a significant source of blood loss, especially in critically ill patients.60 Such blood loss has been associated with the development of anaemia and the need for transfusion.61 Therefore, blood tests should be ordered only when necessary, and the volume of blood collected should be the minimum required. Paediatric bottles can be used to minimise the blood volume collected for testing, which in turn reduces iatrogenic blood loss and transfusion requirements.60 Point-of-care testing devices require smaller blood volumes for analysis and serve as an alternative to traditional laboratory testing. Blood sampling from arterial and central venous lines traditionally involves discarding the initial blood sample. The method of returning the initial blood sample back to the patients has been used to significantly reduce iatrogenic blood loss,62 and this measure should be considered.

As blood transfusion is not without risks, consideration should be given to the balance of benefits against risks. Most would advocate the adoption of a quality clinical transfusion process, ie, “transfusion of the right number of units of blood to the right patient at the right time, in the right conditions, and according to appropriate guidelines”.63 Thus, clinicians should proceed through a chain of related events by making appropriate decisions (Fig).

On the basis of the above three areas for consideration, as well as advice from the Joint United Kingdom (UK) Blood Transfusion and Tissue Transplantation Services Professional Advisory Committee64 and the World Health Organization,65 the Hong Kong Society of Clinical Blood Management makes the following recommendations:
1. A PBM framework, covering primary, hospital, research, audit, and public health measures, should be developed for use in Hong Kong after engagement of different stakeholders;
2. Healthcare professionals, patients, and the public should be educated on the appropriateness of blood transfusion, and PBM programmes; and
3. A PBM framework should be developed for application in Hong Kong, including early recognition and better management of anaemia and iron deficiency in patients and the general population; optimisation of patients’ haematopoiesis and correction of coagulation before surgical procedures; and application of various blood-saving technologies/techniques and point-of-care testing to optimise patients’ outcomes with less transfusion.

The proposed PBM framework is a multi-pronged approach that encompasses a wide range of sectors, disciplines, specialties, and departments. Its implementation will include hospitals, clinics, healthcare facilities, and public health measures to provide care to in-patients, out-patients, and the public of Hong Kong. However, a number of barriers exist that may hamper PBM implementation in Hong Kong.66 67 These include misconceptions related to blood transfusion and difficulties accessing contemporary evidence and data about PBM. There are also existing cultural pressures to retain the status quo, with inadequate incentive for change, as blood is currently delivered freely and efficiently to receivers in Hong Kong. Resources may be inadequate or unequally allocated, such as ferritin assays and intravenous iron preparations for early diagnosis and effective treatment of anaemia, or cell savers and active patient warming equipment for minimising blood loss and conserving blood during surgery. Logistical complexities such as timely investigation and treatment of preoperative anaemia before elective surgery and establishment of point-of-care testing coagulation management programmes may also present obstacles. Finally, PBM lacks specific established quality mechanisms, such as associated policies, standards, guidelines, documentation, performance indicators, coordination, monitoring, evaluation, and feedback.

To overcome these barriers, strong leadership with central steering and empowerment of PBM advocates is required to reinforce and coordinate the current piecemeal and uncoordinated efforts of PBM promotion. The goal of PBM is not simply to reduce the amount of blood transfusion. It is a continuing programme of quality improvement that has the goal of improving patient outcomes via its different measures. With reference to other countries’ experiences and the barriers and challenges that could limit the implementation of PBM in clinical practice, an appropriate framework with local interest should be developed to implement PBM practices at the hospital and territory level.11 68 69

On the basis of the scientific evidence on the successful implementation of PBM and its improvement of patient outcomes, the Hong Kong Society of Clinical Blood Management strongly recommends that Hong Kong implement PBM as soon as possible. The Society will continue to work with relevant professional bodies, patients, and stakeholders to facilitate the local implementation of PBM.

All authors contributed to the concept or design of the study, acquisition and analysis or interpretation of data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

1. GBD 2016 Disease and Injury Incidence and Prevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 2017;390:1211-59. Crossref

2. Kassebaum NJ, Jasrasaria R, Naghavi M, et al. A systematic analysis of global anemia burden from 1990 to 2010. Blood 2014;123:615-24. Crossref

3. Kulier A, Gombotz H. Perioperative Anemia [in German]. Anaesthesist 2001;50:73-86. Crossref

4. Williamson LM, Devine DV. Challenges in the management of the blood supply. Lancet 2013;381:1866-75. Crossref

5. Global Status Report on Blood Safety and Availability. World Health Organization; 2017.

6. Marks PW, Epstein JS, Borio LL. Maintaining a safe blood supply in an era of emerging pathogens. J Infect Dis 2016;213:1676-7. Crossref

7. Faddy HM, Viennet E, Flower RL. Transfusion risk from emerging pathogens in the Asia-Pacific region. ISBT Science Series 2016;11:143-8. Crossref

8. Hans G, Zacharowski K, Spahn DR. Patient blood management. 2016. Available from: http://www.thieme-connect.de/products/ebooks/lookinside/10.1055/b-0036-129702. Accessed 15 Nov 2019.

9. Musallam KM, Tamim HM, Richards T, et al. Preoperative anaemia and postoperative outcomes in non-cardiac surgery: a retrospective cohort study. Lancet 2011;378:1396-407. Crossref

10. Ferraris VA, Hochstetler M, Martin JT, Mahan A, Saha SP. Blood transfusion and adverse surgical outcomes: the good and the bad. Surgery 2015;158:608-17. Crossref

11. Australia National Blood Authority. Patient blood management. Available from: https://www.blood.gov.au/patient-blood-management-pbm. Accessed 8 Mar 2019.

12. Isbister JP. The three-pillar matrix of patient blood management—an overview. Best Pract Res Clin Anaesthesiol 2013;27:69-84. Crossref

13. Spahn DR, Goodnough LT. Alternatives to blood transfusion. Lancet 2013;381:1855-65. Crossref

14. Hofmann A, Farmer S, Towler SC. Strategies to preempt and reduce the use of blood products: an Australian perspective. Curr Opin Anaesthesiol 2012;25:66-73. Crossref

15. Frietsch T, Shander A, Faraoni D, Hardy JF. Patient blood management is not about blood transfusion: it is about patients’ outcomes. Blood Transfus 2019;17:331-3.

16. Morden Healthcare. Hospitals see improved outcomes, lower costs as blood transfusions drop. 2017. Available from: http://www.modernhealthcare.com/article/20170721/NEWS/170729977. Accessed 15 Nov 2019.

17. Althoff FC, Neb H, Herrmann E, et al. Multimodal patient blood management program based on a three-pillar strategy: a systematic review and meta-analysis. Ann Surg 2019;269:794-804. Crossref

18. Leahy MF, Hofmann A, Towler S, et al. Improved outcomes and reduced costs associated with a health-system-wide patient blood management program: a retrospective observational study in four major adult tertiary-care hospitals. Transfusion 2017;57:1347-58. Crossref

19. Muñoz M, Acheson AG, Auerbach M, et al. International consensus statement on the peri-operative management of anaemia and iron deficiency. Anaesthesia 2017;72:233-47. Crossref

20. Shander A, Van Aken H, Colomina MJ, et al. Patient blood management in Europe. Br J Anaesth 2012;109:55-68. Crossref

21. Vaglio S, Gentili S, Marano G, et al. The Italian regulatory guidelines for the implementation of patient blood management. Blood Transfus 2017;15:325-8.

22. Vaglio S, Prisco D, Biancofiore G, et al. Recommendations for the implementation of a patient blood management programme. Application to elective major orthopaedic surgery in adults. Blood Transfus 2016;14:23-65.

23. World Health Organization. Sixty-third World Health Assembly: availability, safety and quality of blood products. 21 May 2010. Available from: http://apps.who.int/gb/ebwha/pdf_files/wha63/a63_r12-en.pdf. Accessed 8 Mar 2019.

24. Hong Kong Red Cross Blood Transfusion Service. Fact sheet of blood collection and use. Available from: https://www5.ha.org.hk/rcbts/fs2016?lang=en. Accessed 29 Feb 2020.

25. Fong CM. Hospital Authority Convention 2019 masterclass 1.4 patient blood management: experience from Tseung Kwan O Hospital. 2019. Available from: https://dryfta-assets.s3.eu-central1.amazonaws.com/assets/haconvention2019/editorimages/1562116855M1.4PatientBloodManagementv.2.pdf. Accessed 29 Feb 2020.

26. National Institute for Health and Care Excellence. Blood transfusion. NICE guideline 24. 2015. Available from: https://www.nice.org.uk/guidance/ng24. Accessed 20 Apr 2020.

27. Fowler AJ, Ahmad T, Phull MK, Allard S, Gillies MA, Pearse RM. Meta- analysis of the association between preoperative anaemia and mortality after surgery. Br J Surg 2015;102:1314-24. Crossref

28. Mueller MM, Van Remoortel H, Meybohm P, et al. Patient blood management: recommendations from the 2018 Frankfurt Consensus Conference. JAMA 2019;321:983-97. Crossref

29. De Hert S, Staender S, Fritsch G, et al. Pre-operative evaluation of adults undergoing elective noncardiac surgery: updated guideline from the European Society of Anaesthesiology. Eur J Anaesthesiol 2018;35:407-65. Crossref

30. Carson JL, Guyatt G, Heddle NM, et al. Clinical practice guidelines from the AABB: red blood cell transfusion thresholds and storage. JAMA 2016;316:2025-35. Crossref

31. Franchini M, Marano G, Mengoli C, et al. Red blood cell transfusion policy: a critical literature review. Blood Transfus 2017;15:307-17.

32. Cleland SR, Thomas W. Iron homeostasis and perioperative management of iron deficiency. BJA Education 2019;19:390-7. Crossref

33. Lau CW, Lee CK. A practical guide of anaemia management for general practitioners. HKMA CME Bulletin July 2018: 2-9.

34. Australia Victoria State Government. Reducing iatrogenic blood loss—clinical practice guideline template. Available from: https://www2.health.vic.gov.au/Api/downloadmedia/%7BF18A2239-E96F-4A49-98F1-C9124E81163F%7D. Accessed 8 Mar 2019.

35. Fischer DP, Zacharowski KD, Meybohm P. Savoring every drop—vampire or mosquito? Crit Care 2014;18:306. Crossref

36. Stefanini M. Iatrogenic anemia (can it be prevented?). J Thromb Haemost 2014;12:1591. Crossref

37. Ullman AJ, Keogh S, Coyer F, Long DA, New K, Rickard CM. ‘True blood’ the critical care story: an audit of blood sampling practice across three adult, paediatric and neonatal intensive care settings. Aust Crit Care 2016;29:90-5. Crossref

38. Squire Y, Laxton C. Blood conservation techniques. Tutorial 390. Available from: https://www.wfsahq.org/components/com_virtual_library/media/8d6712e97e44cd3c3c2af5cc1cbffc00-atow-390-00-01.pdf. Accessed 15 Nov 2019.

39. Jamnicki M, Kocian R, van der Linden P, Zaugg M, Spahn DR. Acute normovolemic hemodilution: physiology,limitations, and clinical use. J Cardiothoracic Vasc Anesth 2003;17:747-54. Crossref

40. Pasternak J, Nikolic D, Milosevic D, Popovic V, Markovic V. An analysis of the influence of intra-operative blood salvage and autologous transfusion on reducing the need for allogeneic transfusion in elective infrarenal abdominal aortic aneurysm repair. Blood Transfus 2014;12 Suppl 1:s182-6.

41. Ker K, Prieto-Merino D, Roberts I. Systematic review, meta-analysis and meta-regression of the effect of tranexamic acid on surgical blood loss. Br J Surg 2013;100:1271-9. Crossref

42. Alshryda S, Sukeik M, Sarda P, Blenkinsopp J, Haddad FS, Mason JM. A systematic review and meta-analysis of the topical administration of tranexamic acid in total hip and knee replacement. Bone Joint J 2014;96-B:1005-15. Crossref

43. Cheriyan T, Maier SP 2nd, Bianco K, et al. Efficacy of tranexamic acid on surgical bleeding in spine surgery: a meta-analysis. Spine J 2015;15:752-61. Crossref

44. National Blood Authority Australia. Patient Blood Management Guidelines: Module 2: Perioperative. Available from: https://www.blood.gov.au/pubs/pbm/module2/3-clinical-guidance/3.6.5-acute-normovolemic-haemodilution. html. Accessed 27 Apr 2020.

45. Murray D. Acute normovolemic hemodilution. Eur Spine J 2004;13 Suppl 1:S72-5. Crossref

46. Ko MT, Chuang KC, Su CY. Multiple analyses of factors related to intraoperative blood loss and the role of reverse Trendelenburg position in endoscopic sinus surgery. Laryngoscope 2008;118:1687-91. Crossref

47. Park CK. The effect of patient positioning on intraabdominal pressure and blood loss in spinal surgery. Anesth Analg 2000;91:552-7. Crossref

48. Franchini M, Mengoli C, Marietta M, et al. Safety of intravenous tranexamic acid in patients undergoing major orthopaedic surgery: a meta-analysis of randomised controlled trials. Blood Transfus 2018;16:36-43.

49. Houston BL, Uminski K, Mutter T, et al. Efficacy and safety of tranexamic acid in major non-cardiac surgeries at high risk for transfusion: a systematic review and meta-analysis. Transfus Med Rev 2020;34:51-62. Crossref

50. Shah A, Palmer AJ, Klein AA. Strategies to minimize intraoperative blood loss during major surgery. Br J Surg 2020;107:e26-38. Crossref

51. Albertin A, La Colla L, Gandolfi A, et al. Greater peripheral blood flow but less bleeding with propofol versus sevoflurane during spine surgery: a possible physiologic model? Spine (Phila Pa 1976) 2008;33:2017-22. Crossref

52. Boonmak P, Boonmak S, Laopaiboon M. Deliberate hypotension with propofol under anaesthesia for functional endoscopic sinus surgery (FESS). Cochrane Database Syst Rev 2016;10(10):CD006623. Crossref

53. Paul JE, Ling E, Lalonde C, Thabane L. Deliberate hypotension in orthopedic surgery reduces blood loss and transfusion requirements: a meta-analysis of randomized controlled trials. Can J Anaesth 2007;54:799-810. Crossref

54. Tran A, Yates J, Lau A, Lampron J, Matar M. Permissive hypotension versus conventional resuscitation strategies in adult trauma patients with hemorrhagic shock: A systematic review and meta-analysis of randomized controlled trials. J Trauma Acute Care Surg 2018;84:802-8. Crossref

55. Harper CM, Andrzejowski JC, Alexander R. NICE and warm. Br J Anaesth 2008;101:293-5. Crossref

56. Reynolds L, Beckmann J, Kurz A. Perioperative complications of hypothermia. Best Pract Res Clin Anaesthesiol 2008;22:645-57. Crossref

57. National Institute for Health and Care Excellence. Clinical guideline CG65. Hypothermia: prevention and management in adults having surgery. Available from: https://www.nice.org.uk/guidance/cg65. Accessed 20 Apr 2020.

58. Segal JB, Blasco-Colmenares E, Norris EJ, Guallar E. Preoperative acute normovolemic hemodilution: a metaanalysis. Transfusion 2004;44:632-44. Crossref

59. Hedner U. Recombinant activated factor VII: 30 years of research and innovation. Blood Rev 2015;29 Suppl 1:S4-8. Crossref

60. Tinmouth AT, McIntyre LA, Fowler RA. Blood conservation strategies to reduce the need for red blood cell transfusion in critically ill patients. CMAJ 2008;178(1):49-57. Crossref

61. AuBuchon JP, Puca K, Saxena S, Shulman IA, Waters JH. Getting started in patient blood management. 2011. Available from: https://www.aabb.org/pbm/Documents/112024DB.pdf. Accessed 8 Mar 2019.

62. Gleason E, Grossman S, Campbell C. Minimizing diagnostic blood loss in critically ill patients. Am J Crit Care 1992;1:85-90. Crossref

63. European Directorate for the Quality of Medicines & HealthCare. Guide to the preparation, use and quality assurance of blood components. Recommendation No. R (95) 15. Available from: http://www.ipst.pt/files/IPST/INFORMACAO_DOCUMENTACAO/EDQM_Blood_transfusion_guide_19ed_2017_pub_PUBSD-89.pdf. Accessed 8 Mar 2019.

64. Joint United Kingdom (UK) Blood Transfusion and Tissue Transplantation Services Professional Advisory Committee. Transfusion Handbook. 4: Safe transfusion—right blood, right patient, right time and right place. Available from: https://www.transfusionguidelines.org/transfusionhandbook/4-safe-transfusion-right-blood-right-patient-right-time-and-right-place. Accessed 8 Mar 2019.

65. World Health Organization. Developing a national policy and guidelines on the clinical use of blood: recommendations. Available from: https://www.who.int/bloodsafety/clinical_use/en/who_bct_bts_01_3.pdf?ua=1. Accessed 8 Mar 2019.

66. Mbanya D. Barriers and enablers to introducing comprehensive patient blood management in the hospital. Biologicals 2012;40:205-8. Crossref

67. Delaforce A, Duff J, Munday J, Hardy J. Overcoming barriers to evidence-based patient blood management: a restricted review. Implement Sci 2020;15:6. Crossref

68. European Commission. Building national programmes of patient blood management (PBM) in the EU—a guide for health authorities. Available from: https://ec.europa.eu/health/sites/health/files/blood_tissues_organs/docs/2017_eupbm_authorities_en.pdf. Accessed 8 Mar 2019.

69. European Commission. Supporting Patient Blood Management (PBM) in the EU—A practical implementation guide for hospitals. Mar 2017. Available from: https://ec.europa.eu/health/sites/health/files/blood_tissues_organs/docs/2017_eupbm_hospitals_en.pdf. Accessed 8 Mar 2019.