Choice of drug

• Medication is often used for prolonged hiccups [Cymet, 2002], but there is no strong evidence to guide the choice of drug. CKS only identified one randomized controlled trial of treatment for hiccups (a crossover trial of four people treated with either baclofen or placebo) [Ramirez and Graham, 1992]. There are, however, numerous case reports of the effectiveness of drugs in the treatment of hiccups. CKS has recommended the following drugs taking into account the licensed indication and the number of case reports of effectiveness.
• A proton pump inhibitor may be effective in relieving symptoms if hiccups are related to gastro-oesophageal reflux disease [Smith and Busracamwongs, 2003; Schuchmann and Browne, 2007].
• Chlorpromazine is licensed for the treatment of intractable hiccups. It may be less effective when initiated in primary care, as the main report demonstrating effectiveness used intravenous administration until the hiccups were controlled before switching to oral treatment [Friedgood and Ripstein, 1955]. Adverse effects (particularly sedation) frequently limit its usefulness, particularly in the palliative care situation [Regnard, 2004].
• Haloperidol is licensed for the treatment of intractable hiccups. Most reports of its effectiveness described intramuscular administration, followed by oral treatment [Ives et al, 1985]. It is better tolerated than chlorpromazine and can be used in palliative care [Finnish Medical Society, 2007].
• Baclofen (unlicensed indication) is commonly regarded as a first-line drug [DTB, 1990; Regnard, 2004]. There are numerous case reports documenting its effectiveness, particularly when other drugs have failed [Burke et al, 1988; Lance and Bassil, 1989; Yaqoob et al, 1989; Bhalotra, 1990; Ramirez and Graham, 1992; Fodstad and Nilsson, 1993; Guelaud et al, 1995; Johnson and Kriel, 1996; D'Alessandro and Dever, 1997; Marien and Havlak, 1997; Marino, 1998; Walker et al, 1998; Oneschuk, 1999; Katsinelos et al, 2000; Lewis, 2000; Hadjiyannacos et al, 2001]. Review guidelines [Finnish Medical Society, 2007; ICSI, 2007] and the Palliative Care Formulary [Twycross and Wilcock, 2007] also suggest that baclofen has a role in people with hiccups in palliative care.
• Metoclopramide (unlicensed indication) has been reported to stop hiccups regardless of the underlying cause [Middleton, 1973; Madanagopolan, 1975], but may be particularly helpful in people with gastric stasis or distension [Regnard, 2004]. One case report found it to be effective for migraine-associated hiccups [Gupta, 2006]. Metoclopramide is also an option for use in palliative care [Finnish Medical Society, 2007].
• Gabapentin (unlicensed indication) has been reported to be effective in the treatment of hiccups, particularly if they are thought to have a neurological cause (e.g. stroke) [Petroianu et al, 2000; Porzio et al, 2003; Hernandez et al, 2004; Moretti et al, 2004; Alonso-Navarro et al, 2007]. A case series illustrated the role of gabapentin when adverse effects limited the use of other medications [Schuchmann and Browne, 2007]. A review including three of these case reports concluded that gabapentin may be also be useful in palliative care because it is not metabolized by the liver (so therefore could be considered for people with liver failure), has few adverse effects, and does not cause sedation [Tegeler and Baumrucker, 2008].
• Parenteral midazolam may be appropriate in the terminal phase of advanced cancer if other treatments are unsuccessful and if the person is very distressed by the hiccups, but should only be used on the advice of a specialist [Wilcock and Twycross, 1996; Regnard, 2004; Regnard and Hockley, 2004; Moro et al, 2005; Twycross and Wilcock, 2007].

Drugs not included

• Other (unlicensed) options have been mentioned in case reports including amitriptyline [Stalnikowicz et al, 1986; Parvin et al, 1988; Peabody et al, 1988], carbamazepine [McFarling and Susac, 1974], valproic acid [Jacobson et al, 1981], sodium valproate [Masand et al, 1990; Regnard and Hockley, 2004], and nifedipine [Mukhopadhyay et al, 1986; Lipps et al, 1990; Brigham and Bolin, 1992]. However, given the relatively smaller body of evidence to support their use and their comparative adverse effect profiles, CKS does not recommend that they are used to treat intractable hiccups in primary care.

Dose

• The recommended dosages of the licensed drugs (chlorpromazine and haloperidol) are based on the British National Formulary [BNF 55, 2008], the Summary of Product Characteristics for each drug [ABPI Medicines Compendium, 2005; ABPI Medicines Compendium, 2007], and the Palliative Care Formulary [Twycross and Wilcock, 2007].
• For the unlicensed drugs, there is no manufacturer information, therefore CKS has based recommended dosages on those used in the relevant case reports. For baclofen and midazolam, recommended dosages were also guided by the Palliative Care Formulary [Twycross and Wilcock, 2007].

Duration of treatment

• The duration of treatment reported in the literature varies considerably, ranging from 1 week to several months.
• Many people who were treated for only short periods, with cessation of their hiccups, remained free of hiccups on discontinuation of treatment. However, there are also reports of people who relapsed as soon as treatment was withdrawn and who had to go back on treatment, often having to stay on treatment for several months.
• The strategy recommended by CKS is pragmatic advice, allowing the clinician to adjust the dose and duration according to the person's response.