Recommendations are made on the basis of limited expert opinion, feasibility, and what CKS considers to be good clinical practice.

• An update for GPs written by an expert on behalf of the Raynaud's and Scleroderma Association states that [Raynaud's & Scleroderma Association, 2008]:
• If secondary Raynaud's phenomenon is suspected, referral to a specialist centre should be made 'if necessary'.
• 'Investigations can be arranged by the GP, or a referral can be made to a specialist'.
• Treatments for severe Raynaud's and digital ischaemia can only feasibly be given in secondary care.
• Intravenous iloprost (a prostaglandin) reduces the frequency and severity of attacks compared with placebo in people with Raynaud's phenomenon secondary to systemic sclerosis [Herrick, 2008a]. It may also be used for acute severe ischaemia with threatened digital loss [Pope, 2007b].
• Bosentan (a dual endothelin-1 receptor antagonist) may reduce new digital ulcer formation compared with placebo in people with Raynaud's phenomenon secondary to systemic sclerosis (and with previous digital ulcers in the last 12 months) [Herrick, 2008a], but should only be used under specialist supervision [BNF 56, 2008].
• Surgical interventions include:
• Digital (palmar) sympathectomy although its role has not been fully established [Herrick, 2008c].
• Stellate ganglion blocks, lumbar sympathetic blocks, and local or regional sympathectomy may be used for acute severe ischaemia with threatened digital loss [Pope, 2007b].
• Debridement of infected or necrotic tissue [Herrick, 2008c].

Lifestyle measures

• In the absence of good evidence from randomized, controlled trials (RCTs) on the efficacy of the recommended lifestyle measures, recommendations are made on the basis of expert opinion in review articles [Kaufman and All, 1996; Landry et al, 1996; Waller, 1997; Wigley, 2002; Bakst et al, 2008]. Expert external reviewers remarked that, in their experience, a drop in core temperature, not just temperature in the hands or feet, can precipitate an attack.

Nifedipine for primary Raynaud's phenomenon

• The recommendation to offer a trial of nifedipine for primary Raynaud's phenomenon is based on limited evidence from a systematic review (search date: April 2003) [Thompson and Pope, 2005] and a BMJ Clinical Evidence review (search date: October 2006) [Pope, 2007a] that nifedipine reduces the frequency and severity of primary Raynaud's attacks when compared with placebo, but is associated with high rates of adverse effects.
• The systematic review included RCTs of both immediate-release and modified-release preparations of nifedipine 20 mg to 60 mg daily [Thompson and Pope, 2005].
• The use of modified-release preparations is advocated by a large body of expert opinion on the basis of fewer adverse effects [Wigley, 2002; Pope, 2007b; Herrick, 2008c; Raynaud's & Scleroderma Association, 2008].

Nifedipine for secondary Raynaud's phenomenon

• The recommendation to offer a trial of nifedipine for secondary Raynaud's phenomenon is based on limited evidence from a BMJ Clinical Evidence review (search date: May 2007) that nifedipine reduces the frequency and severity of attacks compared with placebo in people with Raynaud's phenomenon secondary to systemic sclerosis, but is associated with high rates of adverse effects [Herrick, 2008a].

Other recommendations on nifedipine prescribing

• Doses and prescriptions are recommended on the basis of recommendations from the British National Formulary, convenience, and cost (at the time of publication) [BNF 56, 2008].
• Intermittent use is advocated on the basis of limited expert opinion [Pope, 2007b].
• The recommendation to periodically stop treatment is based on study findings that remission (defined as either no attacks for two cold seasons, or 12 months without symptoms) occurs in between 3% and 33% of people with primary Raynaud's phenomenon after 7–14 years [Suter et al, 2005; Carpentier et al, 2006; Hirschl et al, 2006].
• The recommendation to consider prescribing another calcium-channel blocker if nifedipine is not tolerated is based on expert opinion [Pope, 2007b] in the absence of published evidence of efficacy.

Treatments not recommended for primary Raynaud's phenomenon

• There is no good evidence to recommend the use of drugs other than calcium-channel blockers for the treatment of primary Raynaud's phenomenon. This is on the basis of a Cochrane systematic review of oral vasodilators for primary Raynaud's phenomenon (which excluded calcium-channel blockers) [Vinjar and Stewart, 2008]; a BMJ Clinical Evidence review [Pope, 2007a]; literature reviews of phosphodiesterase inhibitors, angiotensin-converting enzyme (ACE) inhibitors, and angiotensin-receptor blockers for Raynaud's phenomenon [Levien, 2006; Wood and Ernst, 2006]; and two controlled trials [Davinroy and Mosnier, 1993; Friedman et al, 2007].
• There is insufficient good evidence to recommend the use of thermal biofeedback, Chinese acupuncture, evening primrose oil, or Ginkgo biloba extract for the treatment of primary Raynaud's phenomenon.

Treatments not recommended for use in primary care for secondary Raynaud's phenomenon

• There is no good evidence to recommend the use in primary care of drugs other than calcium-channel blockers for the treatment of secondary Raynaud's phenomenon [Herrick, 2008a].
• Bosentan (a dual endothelin-1 receptor antagonist) may reduce new digital ulcer formation compared with placebo in people with Raynaud's phenomenon secondary to systemic sclerosis (and with previous digital ulcers in the last 12 months) [Herrick, 2008a], but should only be used under specialist supervision [BNF 56, 2008].
• There is no evidence to recommend the use of biofeedback or relaxation therapy for the treatment of secondary Raynaud's phenomenon [Herrick, 2008a].