Lysine Clonixinate vs Naproxen Sodium for the Acute Treatment of Migraine: A Double-Blind, Randomized, Crossover Study

Abstract and Introduction

Abstract

Background and Objectives

The process of inflammation is crucial in migraine, and several nonsteroidal anti-inflammatory drugs (NSAIDs) are effective in the treatment of migraine attacks. Despite their efficacy, the routine use of NSAIDs is limited by side effects as well as incomplete efficacy in some patients. Among the available options, lysine clonixinate (LC) and naproxen sodium (NS) have proved effective in migraine. The aim of this study was to compare the efficacy and tolerability of oral formulations of LC and NS in the treatment of moderate or severe migraine attacks, with a double-blind, crossover design.

Methods

Seventy subjects (62 women, 8 men) between ages 18 and 71 years (mean age, 41) with migraine according to the criteria of the International Headache Society were prospectively enrolled. The patients were randomized into 2 groups and each participant treated 2 migraine attacks. Group 1 treated the first attack with LC and the second attack with NS. Group 2 treated 2 attacks in a counterbalanced order. Doses were 250 mg of LC or 550 mg of NS, which were encapsulated for equal appearance. Headache intensity, nausea, photophobia, and side effects were evaluated at baseline, 1 hour, and 2 hours after drug administration. Rescue drugs were allowed after 2 hours for those who didn't respond, and this was also compared between groups.

Results

Sixty patients (54 women, 6 men) completed the study. At 1 hour, 13.6% patients who used LC were pain-free compared with 11.9% who used NS (P = .78). At 2 hours, 35.6% patients who took LC and 32.2% who took NS were pain-free (P = .69). At baseline, 52.5% of the patients randomized to group 1 reported nausea, compared with 33.9% in group 2, and both drugs eliminated nausea: At both 1 hour and 2 hours, nausea diminished significantly for those taking LC, but only after 2 hours for those who took NS (P < .0001). Both drugs eliminated photophobia at 1 hour and 2 hours; however, LC was superior to NS in reducing photophobia at 2 hours (P = .027). Ten patients who took LC and 8 who took NS required rescue drugs after 2 hours. Twelve patients who used LC and 16 who took NS reported side effects.

Comments

Although this study did not include a placebo arm, which impairs any definitive efficacy claims, we found LC and NS to be similarly effective and well tolerated in patients presenting moderate or severe attacks of migraine.

Introduction

Migraine is a highly prevalent neurologic disorder that typically manifests as moderate or severe intermittent headache attacks with associated symptoms.[1–3] Migraine attacks worsen with routine physical activities and may last 4-72 hours if not properly treated.[4] The burden of migraine is severe, resulting in considerable economic and social losses.[5]

Newer agents for the acute treatment of migraine include the triptans,[6] but common analgesics, nonsteroidal anti-inflammatory drugs (NSAIDs), and even opioids continue to be the basic components of the therapeutic arsenal in most countries and specialized centers.[7] In addition, clinical experience suggests that not all patients using a triptan reach a pain-free or sustained pain-free status. Headache recurrence within 24 hours, side effects, and even high costs may also represent limiting factors for the use of triptans.[8]

Some patients find relief with simple or combination analgesics[9,10] and/or NSAIDs. Most NSAIDs have proved effective for treating migraine and are still widely used in many countries despite the potential for gastrointestinal side effects and the current availability of more specific agents.[8–14] Different drug options and formulations are available. The choice for a specific type of medication depends on individual patient characteristics, including pain intensity, speed of headache development, the presence of associated symptoms, degree of incapacitation, and individual patient response.[11] In addition, in developing countries, such as Brazil, the choice of drugs may reflect cost issues.

Lysine clonixinate (LC) is an anthranilic derivative of nicotinic acid that resembles the chemical structure of flufenamic acid.[15,16] Several controlled and open-label trials support the use of LC for acute migraine treatment.[17–19] The agent's structural formula (2-[3-chloro-o-toluidine]piridino-3-carboxilate) allows fast absorption. LC is 96% to 98% protein-bound, and hepatic metabolism results in 4 inactive metabolites. Excretion is 75% renal and 25% fecal.[20] LC costs approximately 20 cents per 125-mg tablet.

Naproxen sodium (NS) is an arylpropionic acid derivative with a potency similar to aspirin and indomethacin but with a more favorable tolerability profile. This agent is a fully orally absorbed NSAID with peak plasma concentrations developing within 2-4 hours and a half-life of 14 hours; it is almost completely (99%) bound to plasma proteins. Roughly 30% of the drug undergoes 6-demethylation, with most of this metabolite excreted as a glucuronide or other conjugates.[21] NS has proved effective for the treatment of migraine attacks.[22] The cost of a 550-mg tablet is approximately 65 cents.

The goal of this study was to compare the efficacy and tolerability of the oral formulations of LC and NS for the treatment of moderate or severe migraine attacks.

Patients and Methods

From April to December 2004, we prospectively studied 70 patients (62 women, 8 men) between ages 18 and 71 years (mean age, 41) with the diagnosis of migraine without aura according to the International Headache Society criteria.[23] Participants were recruited from outpatient headache units from 2 neurology hospitals in Rio de Janeiro and Brasilia, Brazil. All patients were receiving headache prophylaxis treatment with drugs other than NSAIDs. An institutional review board approved the study, and all patients gave written informed consent.

We conducted a double-blind, crossover study in which patients treated 2 attacks of moderate or severe migraine intensity with either 250 mg of LC or 550 mg of NS. Patients were randomized to 2 groups through arrival order. They treated attacks in counterbalanced order. The first patient who was recruited took LC for the first attack and NS for the second attack. The next patient was randomized to take NS for the first attack and LC for the second attack. To ensure against a sequence effect on treatment, the patients were clearly oriented to take the study medications only after a minimal interval of 2 days. Those patients with persisting moderate or severe head pain 2 hours after taking the medication were allowed to use 100 mg of rectal indomethacin. The drugs were encapsulated to appear the same and maintain blindedness throughout the study. All patients received the drugs in boxes marked A and B, each containing 1 capsule. The severity of the migraine attack was measured with a verbal categoric scale in which 0 = no pain; 1 = mild headache; 2 = moderate headache; and 3 = severe headache.

Headache severity and the presence of nausea, photophobia, and side effects were evaluated at baseline (time of intake), 1 hour, and 2 hours after the drug administration. All patients completed an attack report for each of the treated attacks (Figure 1). The use of rescue medications after 2 hours was also compared . Patients receiving treatment for any other medical conditions, women of childbearing age not using contraceptive methods, and those who had used any type of symptomatic migraine medications within the previous 6 hours were excluded.

Figure 1.

Attack report.

We performed statistical comparisons by the nonparametric chi-square test and the Friedman analysis of variance to verify the association differences between treatment responses at 1 hour and 2 hours, consumption of rescue medication at 2 hours, and the incidence of side effects found in the 2 patient groups. To separately analyze time to treatment response in each of the groups, we performed an analysis of variance for repeated measures of categoric data (McNemar test corrected at the 1.7% level). We used the CATMOD² procedure by SAS System software to calculate a 2-tailed P value < .05 for significance levels.

Results

Of the 70 patients enrolled, 60 (54 women, 6 men) completed the study. Six patients did not fill out the protocol and/or treated the attacks when mild, and 4 were lost to follow-up. LC was used to treat 16 severe and 43 moderate attacks; NS was used to treat 19 severe and 39 moderate attacks.

At 1 hour, 13.8% of attacks treated with LC and 11.9% of attacks treated with NS were pain-free (P < .0001 compared with baseline for both drugs). At 2 hours, 35.6% of the attacks treated with LC and 35.2% of the attacks treated with NS were pain-free (P < .0001 compared with baseline for both drugs). Table 1 compares pain-free measures between groups, and Figure 2 shows the pain evolution with time in the 2 groups.

Table 1.

Statistical Analysis of Pain-free Measures Between Drugs LC and NS

	1 Hour After Baseline			2 Hours After Baseline
LC	NS	P Value*	LC	NS	P Value
Pain-free (%)	13.6	11.9	.78	35.6	32.2	.69

LC = lysine clonixinate; NS = naproxen sodium

^*P value of difference in efficacy between agents

Figure 2.

Headache evolution with the time after drug intake.

The evolution of nausea is shown in Table 2 and Table 3. At baseline, 52.5% of the attacks treated with LC and 34% of the attacks treated with NS were associated with nausea. After 1 hour, nausea was reduced significantly for both those who used LC (P < .0001) and those who took NS (P = .014). At 2 hours, relief of nausea was significant for both groups (P < .0001). However, compared with NS, significantly more attacks treated with LC were nausea-free at 1 hour (P < .0005) but not at 2 hours (P = .25).

Table 2.

Longitudinal Analysis of the Nausea and Photophobia for Drug A (LC)

Symptoms	End Points			P Value	Significant Differences*
	Baseline	1 Hour	2 Hours
Nausea (%)	52.5	22.0	20.3	< .0001	Baseline not equal to 1 hr
					Baseline not equal to 2 hr
Photophobia (%)	42.4	18.6	6.8	< .0001	Baseline not equal to 1 hr
					Baseline not equal to 2 hr

LC = lysine clonixinate

CATMOD

^*Significant differences according to McNemar test, at level of 1.7%

Table 3.

Longitudinal Analysis of the Nausea and Photophobia for Drug B (NS)

Symptoms	End Points			P Value	SignificantDifferences*
Baseline	1 Hour	2 Hours
Nausea (%)	33.9	30.5	18.6	.014	Baseline not equal to 2 hr
Photophobia (%)	37.2	30.5	16.9	.0001	Baseline not equal to 2 hr
					1 hr not equal to 2 hr

NS = naproxen sodium

CATMOD

^*Significant differences according to McNemar test, at level of 1.7%

Table 2 and Table 3 show results related to photophobia. Photophobia was significantly reduced at 1 hour for attacks treated with LC (P < .0001) but not with NS, and differences between the groups reached statistical significance (P < .016). At 2 hours, photophobia had improved significantly for attacks treated with either medication (P < .0001), but photophobia-free attacks differed between groups (P = .027). Figures 3 and 4 show the nausea and photophobia evolution for both LC- and NS-treated attacks.

Figure 3.

Evolution of nausea relief for LC and NS.

Figure 4.

Evolution of photophobia improvement for LC and NS.

Among the 59 attacks treated with LC, 8 required rescue medication after 2 hours, compared with 10 of the 58 attacks treated with NS. For attacks that were pain-free at 2 hours, sustained pain-free rates at 24 hours were 55% for LC and 60.1% for NS. In all, 45% of attacks treated with LC and 38.9% of attacks treated with NS recurred (defined as the return of the headache within 24 hours after achieving pain-free status at 2 hours)

Twelve patients using LC and 18 using NS reported side effects, including vomiting (2 attacks with LC and 3 attacks with NS), dizziness (3 attacks with each drug), heartburn or stomachache (4 attacks with LC and 7 attacks with NS), and malaise or asthenia (5 attacks with LC and 7 attacks with NS). Few patients reported more than 1 side effect with either drug. Finally, 25 patients expressed a preference for LC and 25 patients expressed a preference for NS. Eight patients reported no preference whatsoever and 2 did not express preferences.

Discussion

This double-blind, randomized, crossover study demonstrated that both LC and NS are moderately effective in reducing or eliminating the headache of a migraine attack within 2 hours, and in eliminating nausea and photophobia in the same time frame. The methodology that we used involved assessing pain intensity and the presence of nausea and photophobia at baseline, 1 hour, and 2 hours (in which the pain intensity was expressed by a categoric visual scale from 0 to 3), and rescue medication use at 2 hours, as suggested by the International Headache Society guidelines.[24] However, because we did not use a placebo arm, which impairs true efficacy claims, our results should be viewed with caution. Furthermore, we didn't assess differences in outcomes based on original headache intensity, and we examined sustained pain-free rates only for those who were pain-free at 2 hours and not all patients treated. However, we conducted this study because developing countries, such as Brazil, need inexpensive medications to use in the public health system, which is free for everyone. We chose to include LC because it is 2-2.5 times less expensive than NS, an accepted and effective agent for migraine attacks.

Comparison With Triptans

Triptans are expensive and contraindicated in patients with vascular disease. In addition, many patients using a triptan report side effects (up to 89% in one study) or limited efficacy; up to 31% of those taking sumatriptan discontinue it because of poor efficacy.[25–28] Moreover, the efficacy parameters of triptans, particularly pain-free measures, are modest, with a therapeutic gain (pain relief) of roughly 29% to 30%.[29] Clinical comparisons of performance have suggested that some NSAIDs, such as lysine acetylsalicylic acid and tolfenamic acid, present efficacy outcomes similar to sumatriptan, although sumatriptan studies include more patients with severe pain, which may bias comparisons and impede definitive conclusions. However, NSAIDs in adequate dosages and in combination with antiemetic drugs may produce efficacy measures similar to some of the oral triptans.[9,30–32]

In fact, systematic reviews and meta-analyses comparing triptans with other interventions, including NSAIDs, have demonstrated that although clinical practice and expert opinion suggest the superiority of triptans, well-designed clinical trials do not support such observations.[31,33–35] This discrepancy may result from inadequate statistical analysis; differences in response rates favoring patients from clinical practice compared with those who take part in trials; and reliance on end points at 2 hours in clinical trials, which may not fully capture the benefits of triptans relative to other therapies. Finally, waiting to treat a headache until it has become moderate or severe may disadvantage the triptans, because early treatment with these 5HT agonists (triptans) improves benefits compared with other pharmacologic agents.[36]

Recent evidence suggests that combining an NSAID with a triptan may improve efficacy and sustained pain-free rates without increasing side effects.[37–39] Although this has not been demonstrated with LC, it has been seen with NS. In one study, the combination of NS (500 mg) and sumatriptan (50 mg) was superior to the single use of either NS or sumatriptan.[40] Finally, the evidence that migraine seems to involve different pathophysiologic mechanisms, including impaired function of serotonergic pathways, neurogenic inflammation, and dopaminergic hypersensitivity, implies that addressing only one of the involved biological systems is less than ideal; therefore, examining new options, including traditional agents (such as NS and LC), may enhance our opportunities to treat migraine patients, or may provide evidence for using these traditional medications.[41]

Mechanism of Action of NSAIDs

Although the precise mechanism by which NSAIDs function in the relief of migraine and other headaches remains controversial, these agents seem to inhibit the synthesis of prostaglandins, free radicals, and superoxide, and promote partial inhibition of platelet aggregation secondary to the inhibition of thromboxane A2.[42–44] In addition, NSAIDs inhibit serotonin release from platelets and exert a central pharmacologic action in the thalamus and spinal cord.[42,44] In animal studies, LC has been shown to interact directly with the central serotonergic system and indirectly with opioid receptor systems in the thalamic nuclei, dentate gyrus, and layers of parietal cortex.[6,16,17,45] Furthermore, some of the NSAIDs that are effective in migraine treatment reveal high-affinity binding to nociceptive structures in the dorsal horn and brainstem nuclei.[46] Thus, LC and the other NSAIDs may act in migraine through a combination of central and peripheral mechanisms.

Conclusion

Despite the limitations of a study with no placebo arm, this trial demonstrated that LC and NS may be similarly effective to treat moderate or severe migraine attacks. In addition, both drugs presented reasonable tolerability and may be good candidates to be used alone in certain subsets of patients and combined with other classes of pharmacologic agents, such as triptans, in order to improve the efficacy of both. Moreover, LC may represent an attractive option for the treatment of acute migraine attacks in outpatients because it provides an efficacy and tolerability profile that is similar to that of NS, but is considerably less expensive than NS. Patients from developing countries, especially those, such as Brazil, where the health plans or government authorities do not provide medication for those not hospitalized, may benefit from this particular medication as well as others not yet studied.

Readers are encouraged to respond to George Lundberg, MD, Editor of MedGenMed, for the editor's eye only or for possible publication via email: glundberg@medscape.net