British Association of Dermatologists guidelines for the management of people with chronic urticaria 2021*

Plain language summary available online The guideline is presented as a detailed review with highlighted recommendations for practical use in primary, secondary and tertiary care, in addition to an updated patient information leaflet (PIL; available on the BAD Skin Health Information website, https://www.skinhealthinfo.org.uk/a-z-conditions-treatments). Other than providing background information, the guideline does not cover angio-oedema without weals [other than idiopathic, which is now classified as part of chronic spontaneous urticaria (CSU)], hereditary angio-oedema, autoinflammatory syndromes or differential diagnosis. Additionally, the guideline focuses on chronic rather than acute urticaria. This set of guidelines has been developed using the BAD's recommended methodology.1 Further information can be found in Appendix J (see Supporting Information) with reference to the AGREE II instrument (www.agreetrust.org)2 and GRADE.3 The recommendations were developed for implementation in the UK National Health Service (NHS). The Guideline Development Group (GDG), which consisted of eight consultant dermatologists managing adults, children and young people; a consultant immunologist; a consultant psychodermatologist; a drug allergy specialist; two patient representatives and a technical team (consisting of an information scientist, guideline research fellows and a project manager providing methodological and technical support), established a number of clinical questions pertinent to the scope of the guideline and two sets of outcome measures of importance to patients, ranked according to the GRADE methodology (section 2.1; and Appendix A; see Supporting Information). A systematic literature search of the PubMed, MEDLINE, Embase and Cochrane databases was conducted to identify key articles on urticaria from January 2007 to March 2020. An additional, targeted literature search for the antihistamines acrivastine and bilastine was also carried out (from January 1980 to March 2020). Subsequently published papers known to the GDG were included. The final literature searches were run ahead of journal submission in 2021 to ensure currency. The search terms and strategies are detailed in Appendix K (see Supporting Information). Additional references relevant to the topic were also isolated from citations in the reviewed literature and the previous version of the guideline.4 Evidence from the included studies was graded according to the GRADE system (high, moderate, low or very low certainty). The recommendations are based on evidence drawn from systematic reviews of the literature pertaining to the clinical questions identified, following discussions with the entire GDG and factoring in all four factors that would affect their strength rating according to the GRADE approach (i.e. balance between desirable and undesirable effects, quality of evidence, patient values and preferences, and resource allocation). All GDG members contributed towards drafting and/or reviewing the narratives and information in the guideline and Supporting Information documents. When there was insufficient evidence from the literature, informal consensus was reached based on the experience of medical and patient GDG members. The summary of findings with forest plots (Appendix B), tables Linking the Evidence To the Recommendations (LETR; Appendix C), GRADE evidence profiles indicating the quality of evidence (Appendix D), summary of included studies (Appendix E), narrative findings for noncomparative studies (Appendix F), PRISMA flow diagram (Appendix G), list of excluded studies (Appendix H) and AMSTAR 2 appraisal of the included systematic reviews (Appendix I) are detailed in the Supporting Information. The strength of recommendation is expressed by the wording and symbols shown in Table 1. The GDG established the following clinical questions pertinent to the scope of the guideline. Review question 1: investigation Do tests, such as blood tests and the autologous serum skin test, alter the management of urticaria? Review question 2: treatment What is the clinical effectiveness of any H1-antihistamine compared with another for the treatment of urticaria? Review question 3: treatment Would changing from one H1-antihistamine to another lead to benefit in the treatment of urticaria? Review question 4: treatment Would adding an H2-antihistamine to an H1-antihistamine lead to benefit in the treatment of urticaria? Review question 5: treatment What is the effectiveness of leukotriene receptor antagonists in the treatment of urticaria? Review question 6: treatment What are the effectiveness and safety of increasing doses of H1-antihistamines? Review question 7: treatment Would adding other therapies to an H1-antihistamine lead to benefit in the treatment of urticaria, including sulfasalazine, dapsone, thyroxine, tricyclic antidepressants, hydroxychloroquine, methotrexate, danazol, tranexamic acid, mycophenolate mofetil, intravenous immunoglobulins (IVIg) and anticoagulants? Review question 8: treatment What is the effectiveness of taking systemic corticosteroids for the treatment of urticaria? Review question 9: treatment What is the effectiveness of dietary exclusions or supplements for the treatment of urticaria? Review question 10: treatment What is the effectiveness of Helicobacter pylori eradication for the treatment of urticaria? Review question 11: treatment What is the effectiveness of avoiding nonsteroidal anti-inflammatory drugs (NSAIDs) for the treatment of urticaria? Review question 12: treatment What is the effectiveness of ciclosporin for the treatment of urticaria and are there any long-term benefits? Review question 13: treatment What is the effectiveness of omalizumab for the treatment of urticaria? Review question 14: treatment Is the response to treatment for inducible urticarias (e.g. symptomatic dermographism, delayed pressure urticaria, cold urticaria, cholinergic urticaria, vibratory angio-oedema, localized heat urticaria) the same as for CSU? Review question 15: treatment Do any other specific interventions work for inducible urticarias, such as antibiotics in cold urticaria, sulfasalazine in delayed pressure urticaria, phototherapy in dermographism, plasmapheresis in solar urticaria and anticholinergics in cholinergic urticaria? Review question 16: treatment Which H1-antihistamines can be used in pregnancy? Review question 17: treatment Which H1-antihistamines can be used during breastfeeding? Review question 18: treatment What are the key differences in the diagnosis and management of paediatric compared with adult urticaria (if there are any)? Review question 19: treatment What is the clinical effectiveness of miscellaneous monotherapies compared with each other for the treatment of urticaria? The GDG also established two sets of outcome measures of importance to patients (for treatment and investigation), ranked according to the GRADE methodology,5 by the patient representatives (for the full review protocol see Appendix A in the Supporting Information). In the investigation protocol, the outcome is either 'yes' or 'no'. The treatment outcomes (Table 2) use a nine-point scale; outcomes ranked 9, 8 or 7 are critical for decision making; those ranked 6, 5 or 4 are important but not critical for decision making; and those ranked 3, 2 or 1 are the least important for decision making. The following recommendations and ratings were agreed upon unanimously by members of the GDG and patient representatives. For further information on the wording used for recommendations and strength of recommendation ratings see Table 1. The evidence on which the recommendations are based is featured and discussed in Appendices B–E (see Supporting Information). The GDG is aware of the lack of high-quality evidence for many of these recommendations, therefore, strong recommendations with an asterisk (*) are based on available evidence, as well as informal consensus and specialist experience of medical and patient GDG members. Good practice point (GPP) recommendations are derived from informal consensus. Licensing information, dosages and monitoring requirements for specific drugs are not included. However, of note, apart from H1-antihistamines, oral steroids and omalizumab, none of the other treatment options discussed are licensed in the UK for use in urticaria. Except where otherwise stated, we recommend adherence to published guidelines, for example by the manufacturer, the BAD or, in the UK, the British National Formulary (www.bnf.org). In particular, note the licensed dosages for people aged < 14 years (also see Table 3). The recommendations relate to chronic spontaneous and inducible urticarias. Acute urticaria, angio-oedema without weals (other than idiopathic, which is now classified as part of CSU), hereditary angio-oedema and autoinflammatory diseases are not covered. For clarity, we have divided the management options into sections (general treatment, first-, second- and third-line options). However, depending on disease severity, disease fluctuation, comorbidities and the national criteria for use of drugs, the order and combinations of treatment may vary and may change during the course of each person's disease. We note that, since submission of this article for publication, a new international guideline on the management of urticaria has been published.6 Broadly, the recommendations are similar, except that the international guideline favours omalizumab over ciclosporin for CSU. General management for people with all types of chronic urticaria The most important step is to take a detailed clinical history, with examination supplemented by people's own photographs. In most cases this will provide an accurate clinical diagnosis (section 5.2), which will guide management. Disease pathogenesis may also be important in management (section 5.3). R1 (↑) Only consider baseline investigations if clinically indicated (see section 6.0). R2 (GPP) Consider using appropriate validated scoring systems to assess disease activity and impact, for example Dermatology Life Quality Index (DLQI), weekly Urticaria Activity Score 7 (UAS7), Angioedema Activity Score (AAS) and/or Urticaria Control Test (UCT). R3 (GPP) Provide educational material or a patient information leaflet on urticaria or angio-oedema (https://www.skinhealthinfo.org.uk/a-z-conditions-treatments). R4 (GPP) Offer access to support and treatment for anxiety, depression and the psychosocial impact of the disease, where appropriate. The psychological impact can be assessed using, for example, the Hospital Anxiety and Depression Scale (HADS). R5 (GPP) Consider topical antipruritic agents, such as a menthol-containing emollient. R6 (GPP) Advise avoidance of identified triggers or exacerbating factors, such as drugs, and in particular triggers for inducible urticarias. R7 (↑↑) Stop angiotensin-converting enzyme (ACE) inhibitors in people with angio-oedema without weals. General management for people with chronic spontaneous urticaria R8 (↑↑) Avoid NSAIDs in people whose CSU appears to be exacerbated by this class of drugs. R9 (↑) Consider switching NSAID treatment to a selective cyclooxygenase-2 inhibitor, if tolerated and not contraindicated, when there is a history of acute exacerbation of CSU after NSAID intake for inflammation. However, evidence of benefit from switching low-dose aspirin when taken as an antithrombotic to an alternative antiplatelet drug is lacking. Refer to the National Institute for Health and Care Excellence,7 British Society of Allergy and Clinical Immunology (BSACI)8 or European Academy of Allergy and Clinical Immunology (EAACI) guidance9 if reactivity to NSAIDs is suspected. R10 (GPP) Do not advise dietary exclusion routinely. If, from a detailed history, food appears to play a role, investigate appropriately. Θ1 There is insufficient evidence to recommend routine screening for vitamin D deficiency. Θ2 There is insufficient evidence to make a recommendation on dietary supplementation. R11 (↓↓) Do not offer routine screening for Helicobacter pylori. First-line treatment options for people with chronic spontaneous urticaria R12 (↑↑) Offer a second-generation H1-antihistamine, using a regular daily licensed dose (Table 4). R13 (↓↓) Do not offer first-generation H1-antihistamines routinely, unless there is no alternative, due to concerns about their short- and long-term effects on the central nervous system. R14 (↑↑) Offer updosing (i.e. increasing the dose above the licensed dose) of a single second-generation H1-antihistamine, by up to fourfold the licensed dose, to people whose symptoms are inadequately controlled by the standard licensed dose, provided it is tolerated and there is no caution or contraindication (see section 7.2 and Appendix C – LETR narratives). Attempt stepwise dose reduction following complete symptom control. There is no evidence to guide the optimum duration of updosing or speed of dose reduction. R15 (↓↓) Do not updose mizolastine (see section 7.2). R16 (GPP) Consider switching from one second-generation H1-antihistamine to another in people whose symptoms do not respond adequately to, or who do not tolerate, the first drug at standard or increased doses. Θ3 There is insufficient evidence to make a recommendation on using two different second-generation H1-antihistamines at the same time. R17 (↓↓) Do not updose first-generation H1-antihistamines (see R13). R18 (↑) Consider montelukast, in addition to a second-generation H1-antihistamine, in people whose symptoms are inadequately controlled by standard and increased doses of second-generation H1-antihistamines. R19 (↑↑) Offer* progression of therapy, through first-line treatment options (see R12–R18) every 2–4 weeks (every 2 weeks in severe treatment-resistant disease). Θ4 There is insufficient evidence to recommend routine addition of H2-antihistamines to second-generation H1-antihistamines for people whose symptoms are inadequately controlled by the latter. However, they may be considered if urticaria is associated with dyspepsia, although dyspepsia should be investigated appropriately. See section 7.4. R20 (↑) Consider oral prednisolone (e.g. 0·5 mg kg−1) for short, infrequent courses of a few days as rescue treatment to control severe exacerbations, in addition to continued use of a second-generation H1-antihistamine. R21 (↓↓) Do not offer* long-term systemic corticosteroids unless there is no other option. Use the lowest effective dose for the shortest possible period.10 Second-line treatment options for people with chronic spontaneous urticaria For people with CSU with an inadequate response to first-line treatment, the following additional investigations may be relevant when considering the next treatment options. R22 (↓↓) Do not offer autologous serum skin tests (ASSTs) or autologous plasma skin tests (APSTs) routinely. R23 (↑) Consider measuring total IgE levels: a high total IgE level may indicate a higher probability of early disease responsiveness to omalizumab, whereas a normal total IgE level may indicate disease responsiveness to ciclosporin (section 6 and Appendix C – LETR narratives). R24 (↑) If available, consider a basophil histamine release assay (BHRA), although it is not yet subject to a national quality assurance scheme. A positive BHRA may indicate a higher probability of disease responsiveness to ciclosporin and slower or delayed response to omalizumab, whereas a negative BHRA may indicate a higher probability of disease responsiveness to omalizumab (section 6 and Appendix C – LETR narratives). Note: total IgE levels (R23) and BHRAs (R24) are only indicative and may not reflect actual clinical responsiveness in all patients. R25 (↑↑) Offer omalizumab, in addition to a second-generation H1-antihistamine, to people whose symptoms are inadequately controlled by first-line options. R26 (↑↑) Offer* ciclosporin for 3–6 months, in addition to a second-generation H1-antihistamine, to people whose symptoms are inadequately controlled by first-line options. R27 (↑↑) Avoid long-term use of ciclosporin where possible; if not, use at the lowest effective dose, interrupt treatment periodically to confirm continued requirement, and consider alternative agents (see R25, R28 and Θ5). Third-line treatment options for people with chronic spontaneous urticaria Treatment options for inducible urticarias There is much less evidence available than for CSU, but for people with all types of inducible urticaria the following are recommended (based mainly on small case series and anecdotal evidence). First-line treatment options for people with all types of inducible urticaria R29 (↑↑) Offer* a second-generation H1-antihistamine, using a regular daily licensed dose (Table 4). R30 (↓↓) Do not offer* first-generation H1-antihistamines routinely, unless there is no alternative, due to concerns about their short- and long-term effects on the central nervous system. R31 (↑↑) Offer* updosing of a single second-generation H1-antihistamine by up to fourfold the licensed dose to people whose symptoms are inadequately controlled by the standard licensed dose, provided it is tolerated and there is no caution or contraindication (section 7.2 and Appendix C – LETR narratives). Attempt stepwise dose reduction following complete symptom control. There is no evidence to guide the optimum duration of updosing or speed of dose reduction. R32 (↓↓) Do not updose mizolastine (see section 7.2). R33 (GPP) Consider switching from one second-generation H1-antihistamine to another in people whose symptoms do not respond adequately to, or who do not tolerate, the first drug at the standard or increased dose. Θ6 There is insufficient evidence to make a recommendation on using two different second-generation H1-antihistamines at the same time. R34 (↓↓) Do not updose first-generation H1-antihistamines (see R30). Θ7 There is insufficient evidence to recommend routine use of montelukast, although there is some evidence to support its use in some subtypes of inducible urticaria. Second-line treatment options for people with all types of inducible urticaria R35 (↑) Consider omalizumab, in addition to a second-generation H1-antihistamine, in people whose symptoms are inadequately controlled by first-line options, subject to licensing and funding. R36 (GPP) Offer self-injectable adrenaline, if appropriate, for those at risk of anaphylaxis, for example in association with cold or cholinergic urticaria. Third-line treatment options for people with all types of inducible urticaria Consider the following options, in addition to second-generation H1-antihistamines, in people with specific types of inducible urticaria, whose symptoms are inadequately responsive to first- and second-line treatment options, or where the latter are contraindicated or inappropriate. Cholinergic urticaria R37 (GPP) Consider anticholinergic drugs (e.g. oxybutynin), or beta blockers (e.g. propranolol), or danazol, or possibly phototherapy. Cold urticaria R38 (GPP) Consider ciclosporin. Θ8 There is insufficient evidence to recommend routine use of antibiotics (e.g. penicillin or tetracyclines). R39 (GPP) Do not offer cold desensitization. Delayed pressure urticaria R40 (↑) Consider dapsone or sulfasalazine. Solar urticaria R41 (GPP) Offer advice about sun avoidance and sun protection. R42 (↑) Consider UV prophylactic phototherapy using the wavelength of light relevant to the individual person, only following photoinvestigation and obtaining advice from a dermatologist at a specialist photodermatology centre. Θ9 There is limited evidence to recommend plasmapheresis or IVIg for people with solar urticaria. Symptomatic dermographism R43 (↑) Consider narrowband UVB (typically a course of around 30 treatments, repeated after 12 months, if necessary, but not for continual treatment). R44 (GPP) Consider psoralen plus UVA (similarly, not for continual treatment). R45 (GPP) Consider narrowband UVB for other forms of inducible urticaria. Considerations Θ10 There is insufficient evidence to make a recommendation about the safety of use of antihistamines during pregnancy and breastfeeding. However, in active disease and after counselling the female patient with any type of urticaria, where necessary, consider cetirizine or loratadine (see the individual drug summary of product characteristics11 for information on safety during pregnancy) and the discussion in Appendix C (LETR narratives). Future research recommendations The following list outlines future research recommendations (FRRs). FRR1 Further investigation of the genetic predisposition and/or mechanistic factors that drive the development of all types of urticaria and/or angio-oedema, including the new theory of IgE-mediated 'autoallergy' and characterization of the roles of basophils, eosinophils and lymphocytes. FRR2 Better characterization of, and comparisons between, basophil-based assays as predictors of drug responses. FRR3 Development of better biomarkers to predict responsiveness to anti-IgE and other therapies. FRR4 Utilizing the results from FRR1–3 to address the possibility of personalized therapy and whether new biological targets might offer new therapeutic options. FRR5 Randomized controlled trials (RCTs) evaluating the safety and efficacy of updosing one second-generation H1-antihistamine compared with using two different second-generation H1-antihistamines at the same time in people with CSU. FRR6 RCTs evaluating the safety and efficacy of omalizumab in people with all subtypes of inducible urticaria. FRR7 RCTs evaluating the safety and efficacy of other treatment options (as featured in the treatment algorithm) in people with all subtypes of inducible urticaria. FRR8 RCTs evaluating the safety and efficacy of emerging treatments for people with all types of urticaria, including the new high-affinity, humanized monoclonal anti-IgE antibody ligelizumab, and potential new treatment options such as tyrosine kinase inhibitors, dupilumab, histamine H4 receptor antagonists, C5a receptor antagonists and drugs targeting inhibitory mast cell receptors (see section 7.8). FRR9 Better characterization of the optimum duration of the various treatment options available to people with all types of urticaria. FRR10 Investigations into disease incidence and prevalence, the predictive value of laboratory investigations (such as total IgE levels or basophil-based assays), and the safety and efficacy of the various current and potential future treatment options in children and young people with urticaria and/or angio-oedema of all types. The recommendations, discussions in the LETRs (Appendix C) and consensus specialist experience were used to produce management pathways for adults with chronic urticaria – see Figure 1. Urticaria consists of transient weals, angio-oedema or both. Weals are usually itchy, whereas the swellings of angio-oedema are often not. Depending on the disease subtype, angio-oedema or weals may be painful. Urticaria is usually divided into acute and chronic forms, becoming chronic when daily or almost daily weals continue for 6 weeks or more, although many attacks of acute urticaria settle much more quickly. Some forms of urticaria may be accompanied by systemic symptoms, such as arthralgia, gastrointestinal disturbance, malaise, lethargy, or wheeze and/or mucosal involvement. Acute urticaria may be a presenting sign of anaphylaxis. The reported lifetime prevalence rate of urticaria varies from 8% to 24%, with a lifetime prevalence rate of about 1–2% for chronic urticaria.12-14 The point prevalence of chronic urticaria varies from 0·1% in North America to 1·4% in Asia.15 The disease is slightly more common in females, except in young children. People suffer greatly if they have any form of urticaria, and chronic disease may have a significant effect on quality of life.16, 17 Even though the initial treatment for many types of urticaria is similar, there are some important exceptions. Therefore, accurate clinical categorization, based on a detailed history and examination (section 5.2) and an understanding of disease pathogenesis (section 5.3), are essential to guide investigation and management. Of note, different forms of urticaria commonly occur together. No cause is identified in more than 50% of people with acute urticaria (causes when identified include drugs, infections including COVID-19, and type 1 hypersensitivity reactions) and in many of those with CSU (see section 5.3.2. for pathogenesis of CSU). Weals generally last for up to 24 h, but the swellings of angio-oedema may last for up to 72 h. Spontaneous urticaria Inducible urticarias (reproducibly induced by the same physical stimulus) Angio-oedema without weals Diseases presenting with urticaria-like rashes These urticarias are usually chronic. Weals are reproducibly induced by the same physical stimulus. Weals usually appear within a few minutes of the stimulus and last for < 2 h, the exception being delayed pressure urticaria, where weals may take 30 min to 12 h to develop, and then last for a few days. The shape and size of the weals may aid diagnosis, for example linear weals in dermographism, or papular weals surrounded by a red flare in cholinergic or aquagenic urticarias. Some inducible urticarias present as a spectrum of symptoms from pruritus, urticaria and angio-oedema to anaphylaxis. Inducible urticarias can be confirmed on provocation testing (see section 6.0). Disease severity may be reduced through the avoidance of triggers, although this can be difficult and disabling. Inducible urticarias tend to be underdiagnosed. Usually no cause is identified. However, it is important not to miss uncommon cases of drug-induced angio-oedema where the culprit drug must be withdrawn (especially ACE inhibitors, where the angio-oedema may occur soon or many years after drug initiation),18, 19 or rare cases of C1 esterase inhibitor deficiency (see section 6.4). Both may cause life-threatening airway swelling and neither respond to the usual treatment for angio-oedema. Angio-oedema of the gastrointestinal tract is common in C1 esterase inhibitor deficiency. Like inducible urticarias, this is characterized by a weal-and-flare response at the site of contact of a trigger, anaphylaxis may occur, the onset is rapid (within minutes) and reactions usually last for < 2 h. However, the disease is acute, not chronic, and the trigger is not physical but instead may be any of a large variety of substances, for example food, plants, animals, fragrances or preservatives. Weals are usually of prolonged duration, may be painful rather than itchy and sometimes leave residual bruising or postinflammatory change. It can be difficult to differentiate urticarial vasculitis from delayed pressure urticaria. However, in urticarial vasculitis, there are often marked systemic symptoms, there may be joint or renal involvement, an association with other underlying diseases and high inflammatory markers. A skin biopsy is needed to confirm the diagnosis (see section 6.5).20 Autoinflammatory syndromes presenting with urticaria-like rashes include cryopyrin-associated periodic syndrome (CAPS) (usually with onset in childhood, although late-onset acquired disease is recognized) and Schnitzler syndrome (acquired with adult onset). CAPS consists of three overlapping conditions: familial cold autoinflammatory syndrome, Muckle–Wells syndrome and neonatal-onset multisystem inflammatory disorder. These diseases are rare (for pathogenesis see section 5.3.3). They differ in associated organ involvement, but are all usually accompanied by fever, malaise and high levels of inflammatory markers.21, 22 Some cases of acute or episodic urticaria and some cases of contact urticaria are due to mast cell degranulation caused by allergens crosslinking antigen-specific IgE (type 1 hypersensitivity). Approximately 33% of people with CSU have functional, histamine-releasing IgG autoantibodies. These either directly crosslink high-affinity IgE receptors (FcεRI) or bind IgE.23 A new theory is emerging of type 1 autoimmunity or 'autoallergy', in which IgE antibodies are directed at an element of self. Antigens may then crosslink IgE on mast cells or basophils, causing degranulation. It has been observed that there are fast and slow responders to treatment with omalizumab (an anti-IgE antibody). This has led to the proposal that the rapid response may be due to omalizumab rapidly binding free IgE autoantibodies against autoallergens, whilst the slow responses may be due to the slower loss of mast cell (or basophil) membrane-bound IgE and downregulation of FcεRI, thus reducing IgG-mediated activation.24 The functional importance of IgE 'autoallergic' autoantibodies is under investigation. Thus far, there is some evidence that IgE anti-thyroid peroxidase antibodies and possibly IgE anti-interleukin-24 antibodies may play a role in some patients with CSU.25, 26 These are characterized by dysregulation of innate immunity. Persistent uncontrolled inflammation occurs in the absence of triggers and is mediated by excessive cytokine production. Most cases of CAPS are due to autosomal dominant or de novo mutations in the NLRP3 gene, resulting in increased activity of the NLRP-3 inflammasome and increased secretion of interleukin-1β.27, 28 Late-onset CAPS is now thought to be due to acquired somatic (mosaic) mutations in NLRP3,29 but these have not been identified in Schnitzler syndrome, even though the clinical presentation is identical.30 Instead, Schnitzler syndrome is defined by the presence of a monoclonal gammopathy of unknown significance (MGUS), usually IgM, but sometimes IgG.30 Many people with CSU find that nonspecific factors including heat, alcohol, infections and stress exacerbate or trigger their urticaria, but the underlying mechanisms are poorly understood. Drugs may precipitate u