Int J Pharm Pharm Sci, Vol 17, Issue 3, 32-35Original Article

A PROSPECTIVE OPEN-LABEL STUDY TO COMPARE THE EFFICACY AND SAFETY OF TOPIRAMATE ALONE VERSUS TOPIRAMATE WITH VITAMIN D IN PROPHYLACTIC TREATMENT OF PATIENTS WITH MIGRAINE

ALOK DIXIT1*, MANIK BRAHEMI2, NASREEN FATMA KHAN3, RAMAKANT YADAV4, C. V. SINGH1

1Department of Pharmacology, Uttar Pradesh University of Medical Sciences, Saifai, Etawah. UP-206130, India. 2Department of Pharmacology, GMC, Budaun-243601, Uttar Pradesh, India. 3Department of Pharmacology, GSVM MC, Kanpur. UP-208002, India. 4Department of Neurology, University of Medical Sciences, Saifai, Etawah-206130, Uttar Pradesh, India
*Corresponding author: Alok Dixit; *Email: alkdxt@yahoo.co.in

Received: 18 Dec 2024, Revised and Accepted: 23 Jan 2025

ABSTRACT

Objective: Migraine affects 10% of the global population, especially women aged 25-55, causing severe headaches with light and sound sensitivity. While acute treatments are common, preventive therapies are underutilized. Recent studies suggest combining vitamin D with Topiramate, an anticonvulsant, could enhance prevention, reducing both attack frequency and duration. Therefore, the present study compares the safety and efficacy of Topiramate alone versus topiramate with Vitamin D.

Methods: A study was conducted on 100 migraine patients to compare the effect of topiramate alone versus topiramate with vitamin D. Pain severity and disability were measured using a Visual Analog Scale (VAS), Migraine Disability Assessment Score (MIDAS), and Migraine Severity (MIGSEV) scores over 12 w, excluding patients with other medical conditions.

Results: Both the Topiramate and Topiramate with Vitamin D groups showed significant reduction in VAS, MIDAS, and MIGSEV scores after 12 weeks. Migraine without aura was more common in both groups. The Topiramate with Vitamin D group also experienced increased serum calcium and Vitamin D levels. Both groups showed significant improvements in pain and disability, with the Topiramate with Vitamin D group showing better results.

Conclusion: Combining topiramate with vitamin D effectively reduces migraine frequency and disability.

Keywords: Pain severity, Migraine disability assessment score

© 2025 The Authors. Published by Innovare Academic Sciences Pvt Ltd. This is an open access article under the CC BY license (https://creativecommons.org/licenses/by/4.0/)
DOI: https://dx.doi.org/10.22159/ijpps.2025v17i3.53449 Journal homepage: https://innovareacademics.in/journals/index.php/ijpps

INTRODUCTION

The word “migraine” is from the Greek ἡμικρανία (hemikrania), “a pain on one side of the head,” from ἡμι-(hemi-), “half,” and κρανίον (kranion), “skull” [1]. Migraine and tension headaches are the most common primary headache disorders that affect 80% of people all over the world [2]. A global estimation of migraine headache prevalence showed that migraine affects 1 in 10 people worldwide [3]. Migraine is a disabling headache disorder characterized by unilateral pulsating headaches associated with photophobia, phonophobia, nausea, and occasionally transient neurological symptoms [4]. In contrast to trigeminal autonomic cephalalgias, where the headache is generally side-locked, a common characteristic of migraine is that the headache may switch sides from attack to attack. Migraines are self-limiting, usually associated with autonomic symptoms [1]. Migraine without aura is more common in women than migraine with aura [5].

Recurrent migraines are commonly diagnosed in people around the ages of 25 and 55 worldwide, affecting 6% of men and 18% of women [5]. Most migraine patients have been prescribed or treat themselves with acute attack/rescue medication. Unfortunately, prophylactic migraine treatment is not much more common. Numerous global recommendations have been released about drug selection, treatment objectives, and indications. While there is considerable variation among these guidelines, it is generally agreed upon that a thorough discussion with the patient should precede the initiation of preventive medication. The reasonable objectives for preventive therapy include reduced attack frequency, severity, and length; enhanced response to acute attack treatment; and enhanced quality of life [6].

Today, physicians are interested in using alternative medicine and non-pharmacological remedies for migraine prevention, and many nutraceuticals, such as magnesium, coenzyme Q10, riboflavin, butterbur, feverfew, melatonin, etc., are utilized for the prophylaxis of migraine in children and adults [7]. Much concern was directed toward the presence of a possible relationship between vitamin D and migraine.

Vitamin D was found to have a role in the pathways involved in the pathogenesis of migraine, including pain sensitization, inflammation, and immune dysfunction [8]. The brain has an abundance of vitamin D receptors, and there is evidence of a non-skeletal role of vitamin D in inflammation, immunity, and the metabolism of neurotransmitters [9, 10]. An association between the dysfunction of transporter proteins of vitamin D3 metabolites and migraine attacks was observed in a study by Nagata et al. [11], which concluded that higher levels of vitamin D3 might reduce the risk of migraine headaches.

Sufficient levels of circulating vitamin D are mandatory for the absorption of several minerals, particularly calcium, but also phosphorus and magnesium [12]. There are two major forms of vitamin D: D2 and D3. The first one is acquired from ultraviolet (UV) irradiation of the yeast sterol, ergosterol, while vitamin D3 is produced from 7-dehydrocholesterol in the skin after exposure to UV radiation [13]. Vitamin D3 deficiency (serum 25-OH vitamin D3 level<20 ng/ml) and insufficiency (25-OH vitamin D3<30 ng/ml) are also connected with pain disorders, including fibromyalgia and headaches [13, 14]. It is also a potent antioxidant, thus contributing to the vascular health of the brain.

Furthermore, vitamin D and its metabolites (as a steroid hormone) can influence many neurotransmitters, including dopamine, acetylcholine, and serotonin [14]. Vitamin D also reduces the production of nitric oxide (NO) by inhibiting the expression of NO synthase. NO is an important biological regulator that affects neurotransmission and vasodilation and is considered a key mediator in migraine [15]. During headache attacks, NO levels in jugular venous plasma increase; there is also evidence that NO synthase inhibitors are effective in treating migraine [16].

Current evidence suggests that the antiepileptic drug Topiramate is best in context to safety and efficacy for prophylactic treatment of migraine. Several studies show that prophylactic treatment of topiramate in migraine reduces the frequency of attacks and has fewer side effects. It has been observed that patients with migraine are suffering from vitamin D deficiency. Vitamin D in combination with antiepileptic drugs for prophylaxis of migraine has shown a significant reduction in the frequency and duration of attacks. Just 12.4% reported current use of daily preventive migraine medication [17]. This indicates that preventive treatment is still severely underutilized. With this background, the present study aims to compare the efficacy and safety of topiramate alone versus topiramate with vitamin D in episodic and chronic migraine patients.

MATERIALS AND METHODS

A study involving 100 migraine patients was conducted in the Uttar Pradesh University of Medical Sciences, Saifai Etawah, U. P. Department of Pharmacology, in collaboration with the Neurology Department. The patients were identified as having episodic, chronic migraine with aura and migraine without aura. Patients with a history of traumatic brain injury, chronic liver/kidney disease, hypertension, psychiatric disorders, significant CNS disorders, epilepsy or seizure history, and pregnant or lactating women were excluded. Patients were divided into two groups: those receiving topiramate 25/50 mg daily and those receiving vitamin D3 8000 IU along with topiramate 25/50 mg daily. The lowest dose of either medication was used for all patients, and their medical and dietary history along with sociodemographic details was recorded. The degree of pain intensity was assessed using a Visual Analog Scale [(VAS)-0 to 10 rating scale] at 0, 4, 8, and 12 w. To determine the pain severity, a 12 w Migraine Disability Assessment Score (MIDAS) measuring the impact of headaches on migraineurs life and Migraine Severity (MIGSEV) scale for diagnosis and management of migraineurs were evaluated at baseline and after 3 mo.

Ethical approval: The study was approved by the Institutional Ethics Committee (193/2020-21)

Statistical analysis

All the patients were assessed on the intention-to-treat principle. The primary and secondary efficacies, as well as the quantitative and qualitative data, were compared. Primary efficacy was evaluated by the percentage change in VAS and secondary efficacy by the percentage changes in MIDAS and MIGSEV scores at the end of the study. The quantitative data was evaluated using the student t-test and the qualitative data using the chi-square test. A P-value less than 0.05 was considered significant.

RESULTS

In the Topiramate group, the majority of patients, 16 (32%), were aged 25 years or younger, while at least 4 (8%) were aged 55 years or older. The mean age of patients was 35.3 years. In the Topiramate+Vitamin D Group, the majority of patients, 26 (52%), were between 26 and 40 years old, with 3 (6%) patients aged over 55. The mean age of patients in this group was 33.9 years. The distribution of patients in different age categories in both groups was found to be statistically non-significant, and the mean age of patients in all groups was also non-significant (P>0.05).

The majority of patients in all groups were married, and the correlation of patients' marital status in both groups was found to be statistically non-significant (p>0.05). Out of a total of 100 patients, 36 in the Topiramate group and 40 in the Topiramate+Vitamin D group had a positive family history of migraine. Based on the type of migraine, patients from both groups were categorized, and it was observed that migraine without aura was more common overall, covering more than 50% of patients in each group. The correlation between the type of migraine and the distribution of patients in both groups was found to be statistically non-significant (p>0.05) as depicted in table 1.

Table 1: Distribution of patients with migraine based on age, gender, and marital status

Age in years Topiramate (n = 50) Topiramate+Vitamin D (n =50) P-value
≤ 25 16 (32%) 13 (26%) 0.231
26-40 15 (30%) 26 (52%)
40 – 55 15 (30%) 8 (16%)
>55 4 (8%) 3 (6%)
mean Age 35.3±12.5 33.9±10.9 0.346
Male 18 (36%) 22 (46%) 0.563
Female 32 (64%) 28 (54%)
Married 36 (72%) 41 (82%) 0.068
Unmarried 14 (28%) 9 (18%)
Migraine with Aura 17 (34%) 18 (36%) 0.818
Migraine without Aura 33 (66%) 32 (64%)

The VAS scores of patients in both groups were measured, and it was observed that the VAS score significantly reduced in both groups from 0 w to the 12th w (P<0.05). The MIDAS score was also significantly reduced in both the groups from 0 w to the 12th w (p<0.05). In comparison to the baseline, the MIDAS score in both groups was found to be non-significant at the end of the study, whereas it was significant at the 12th w with a p-value less than 0.05. The MIGSEV score was significantly reduced in both groups from 0 w to 12th ws (p<0.05) but was found to be non-significant at 0 w, whereas it was statistically significant in both the groups at 12th w (P<0.05) compared to the baseline, as depicted in table 2.

Table 2: Comparison of different scores in patients with migraine in different groups

Patients VAS score MIDAS score MIGSEV score
0 W 12 W 0 W 12 W 0 W 12 W
Mean±SD Mean±SD Mean±SD Mean±SD Mean±SD Mean±SD
Topiramate 7.3±0.9 5.3±0.6* 30.9±2.6 25.1±3.7* 3.0±0.0 2.5±0.5*
Topiramate+Vitamin D 8.1±0.8 3.3±0.6* 30.5±2.2 20.1±1.5*# 3.0±0.0 1.3±0.4*#

*Denotes P<0.001; Significant values at 12th w in corresponding groups as compared to 0 w, #Denotes P<0.001; Topiramate alone versus topiramate and vitamin D at 12th w.

Additionally, as depicted in table 3, the serum calcium and Vitamin D levels were observed to have increased significantly in the Topiramate along with Vitamin D group from 0 w (9.6±1.0 and 9.8±2.9, respectively) to the 12th w (11.0±0.1 and 15.5±2.7, respectively) (p<0.001).

Table 3: Comparison of serum vitamin D and serum calcium among patients with migraine in different groups at 0 d and the end of the 12th w

Patients Serum calcium Vitamin D
0 W 12 W 0 W 12 W
Mean±SD Mean±SD Mean±SD Mean±SD
Topiramate (n=50) 8.1±1.4 9.6±1.0* 10.2±3.7 9.8±2.9*
Topiramate+Vitamin D (n=50) 10.0±1.0 11.0±0.1* 10.6±2.9 15.5±2.7*

*Denotes P<0.001; Significant values at 12th w in corresponding groups as compared to 0 w {Reference Normal Values: Serum vitamin D: (30-50 ng/ml); Serum calcium: (8.5-11 mg/dl)}

DISCUSSION

In the present study, a total of 100 individuals diagnosed with migraine were enrolled and evenly distributed into two categories: one receiving only Topiramate and the other receiving a combination of Topiramate and Vitamin D. Within the group receiving Topiramate with Vitamin D, a majority (more than 50%) of the patients fell within the 26-40 y age range, showing a higher prevalence among females. Our findings align with Theeler BJ et al. [18], who noted a higher occurrence of migraine in females compared to males. Various studies have reported migraine prevalence rates ranging from 5% to 8% in males and 11% to 16% in females [5, 19], potentially linked to hormonal fluctuations during the menstrual cycle, as proposed in the "Estrogen withdrawal hypothesis of Somerville." Similar results were noted by Victor et al. [20] in a national health survey in the United States. According to a recent systematic analysis for the Global Burden of Disease Study 2016, the age-standardized prevalence rate of migraine worldwide was estimated to be 14%, with a significantly higher prevalence in females (19%) compared to males (10%) [21].

Migraine, whether with or without aura, exhibits distinct familial patterns and modes of inheritance, suggesting different etiologies [22]. Migraine with aura is linked to an increased risk of ischemic stroke, whereas this increased risk is not associated with migraine without aura, and other disorders are also connected to migraine with aura, but not to migraine without aura [23]. In our study, migraine without aura was more prevalent, comprising over 50% of the patients in the groups. Silberstein SD et al. [24] reported that 35.5% of subjects in the intention-to-treat population experienced migraine with aura. Russell MB et al. [22],in a comparison of clinical characteristics in 484 migraineurs from the general population, found a prevalence of 14.7% for migraine without aura, in contrast to 7.9% for migraine with aura.

To assess the efficacy of Topiramate alone versus Topiramate with Vitamin D, a comparison of VAS, MIDAS, and MIGSEV scores was done at 0 w and the end of the 12th w after administering the drugs to the respective groups. In this study, the VAS scores for pain measurement were calculated at 0 w and 12 w intervals for both the groups and the differences were statistically significant for both groups (p<0.05). At the end of the 12th w, the results showed statistical significance with P<0.001 in the VAS score when compared to the baseline.

The baseline MIDAS score at 0 w between patients of Topiramate and those of Topiramate with Vitamin D was insignificant (P<0.408). However, at the end of the 12th w, the MIDAS score exhibited statistical significance (P<0.001). To assess the drug efficacy in each group, the Migraine Disability Assessment (MIDAS) questionnaire was employed. High mean scores were observed at the treatment initiation among the groups (30.9±2.6 in the Topiramate group and 30.5±2.2 in the Topiramate and Vitamin D group). After 12 w of treatment, significant differences were observed in scores among the groups (p<0.05), with values of 25.1±3.7 and 20.1±1.5, respectively. The Topiramate and Vitamin D medication demonstrated the lowest score.

Similar results were reported by Moras K [25], where patients subjected to migraine prophylaxis with amitriptyline exhibited a mean MIDAS score reduction from 11.6 to 9.4 after treatment. Dwajani. S. et al. [26] observed a reduction in MIDAS score from 22.70±13.68 at enrollment to 10.44±4.25 after 3 mo of treatment with Topiramate. Similarly, Laghari Jamil M. et al.[27] reported significant reductions in headache frequency and MIDAS total score with topiramate treatment.

At baseline, the MIGSEV score at 0 w between patients of Topiramate and those of Topiramate with Vitamin D was insignificant. MIGSEV scores were significantly reduced in both groups from 0 w to the 12th w (p<0.05). The correlation of MIGSEV scores in both groups was non-significant at 0 w but came out to be statistically significant at the 12th w (p<0.05). El Hasnaoui A. et al. [28] stated that the MIGSEV questionnaire is proposed as a simple measure of severity for the diagnosis and management of migraineurs, suitable for use by both physicians and patients. The limitations of the present study were the lack of a placebo, the single-centered open-label study, and the lack of follow-up after discontinuation.

CONCLUSION

The present study concludes that the combination of topiramate and vitamin D3 proved more efficacious than topiramate alone in reducing both weekly frequency and disability scores in patients associated with migraine. Moreover, the use of vitamin D3 therapy emerges as a potentially safe and effective approach for prophylaxis against both episodic and chronic migraines. To establish the optimal dosage of vitamin D3 for migraine prevention, additional clinical trials with larger sample sizes, altitude control, and consideration of seasonal variations are warranted.

ACKNOWLEDGMENT

We are very grateful to the Uttar Pradesh University of Medical Sciences (UPUMS) and the associated hospital, Saifai for providing Support, Guidance, and Necessities.

FUNDING

Nil

AUTHORS CONTRIBUTIONS

All the authors Dr. Alok Dixit, Dr. Manik Brahemi, Dr. Nasreen Fatma Khan, Dr. Ramakant Yadav, and Dr. C. V. Singh, have equally made a substantial contribution to the conception, acquisition of data, and interpretation of data and drafting the article and agreed to be held accountable for all aspects of the work.

CONFLICTS OF INTERESTS

The researchers received no financial support from any pharmaceutical companies. The authors declared no conflicts of interest.

REFERENCES

  1. Ommurugan B, Priya A, Bairy LK. A complete review of migraine. Asian J Pharm Clin Res. 2017 Oct;10(10):57-62. doi: 10.22159/ajpcr.2017.v10i10.19207.

  2. Pietrobon D. Insights into migraine mechanisms and CaV2.1 calcium channel function from mouse models of familial hemiplegic migraine. J Physiol. 2010;588(11):1871-8. doi: 10.1113/jphysiol.2010.188003, PMID 20194127.

  3. Nattagh Eshtivani E, Sani MA, Dahri M, Ghalichi F, Ghavami A, Arjang P. The role of nutrients in the pathogenesis and treatment of migraine headaches: review. Biomed Pharmacother. 2018 Jun;102:317-25. doi: 10.1016/j.biopha.2018.03.059, PMID 29571016.

  4. Mulder EJ, Van Baal C, Gaist D, Kallela M, Kaprio J, Svensson DA. Genetic and environmental influences on migraine: a twin study across six countries. Twin Res. 2003;6(5):422-31. doi: 10.1375/136905203770326420, PMID 14624726.

  5. Sangermani R, Boncimino A. The use of nutraceutics in childrens and adolescents headache. Neurol Sci. 2017;38 Suppl 1:121-4. doi: 10.1007/s10072-017-2896-7, PMID 28527085.

  6. Evers S. Treatment of migraine with prophylactic drugs. Expert Opin Pharmacother. 2008;9(15):2565-73. doi: 10.1517/14656566.9.15.2565, PMID 18803445.

  7. Silberstein SD. Migraine. Lancet. 2004;363(9406):381-91. doi: 10.1016/S0140-6736(04)15440-8, PMID 15070571.

  8. Lukacs M, Tajti J, Fulop F, Toldi J, Edvinsson L, Vecsei L. Migraine neurogenic inflammation drug development pharmacochemical aspects. Curr Med Chem. 2017;24(33):3649-65. doi: 10.2174/0929867324666170712163437, PMID 28707585.

  9. Eyles DW, Burne TH, MC Grath JJ. Vitamin D effects on brain development adult brain function and the links between low levels of vitamin D and neuropsychiatric disease. Front Neuroendocrinol. 2013;34(1):47-64. doi: 10.1016/j.yfrne.2012.07.001, PMID 22796576.

  10. WU Z, Malihi Z, Stewart AW, Lawes CM, Scragg R. The association between vitamin D concentration and pain: a systematic review and meta-analysis. Public Health Nutr. 2018;21(11):2022-37. doi: 10.1017/S1368980018000551, PMID 29559013.

  11. Nagata E, Fujii N, Hosomichi K, Mitsunaga S, Suzuki Y, Mashimo Y. Possible association between dysfunction of vitamin D binding protein (GC globulin) and migraine attacks. Plos One. 2014;9(8):e105319. doi: 10.1371/journal.pone.0105319, PMID 25147936.

  12. Holick MF, Binkley NC, Bischoff Ferrari HA, Gordon CM, Hanley DA, Heaney RP. Evaluation treatment and prevention of vitamin D deficiency: an endocrine society clinical practice guideline. J Clin Endocrinol Metab. 2011;96(7):1911-30. doi: 10.1210/jc.2011-0385, PMID 21646368.

  13. Autier P, Boniol M, Pizot C, Mullie P. Vitamin D status and ill health: a systematic review. Lancet Diabetes Endocrinol. 2014;2(1):76-89. doi: 10.1016/S2213-8587(13)70165-7, PMID 24622671.

  14. Prakash S, Kumar M, Belani P, Susvirkar A, Ahuja S. Interrelationships between chronic tension-type headache musculoskeletal pain and vitamin D deficiency: is osteomalacia responsible for both headache and musculoskeletal pain? Ann Indian Acad Neurol. 2013;16(4):650-8. doi: 10.4103/0972-2327.120487, PMID 24339599.

  15. Garcion E, Sindji L, Nataf S, Brachet P, Darcy F, Montero Menei CN. Treatment of experimental autoimmune encephalomyelitis in rat by 1,25-dihydroxyvitamin D3 leads to early effects within the central nervous system. Acta Neuropathol. 2003;105(5):438-48. doi: 10.1007/s00401-002-0663-0, PMID 12677443.

  16. Messlinger K, Lennerz JK, Eberhardt M, Fischer MJ. CGRP and NO in the trigeminal system: mechanisms and role in headache generation. Headache. 2012;52(9):1411-27. doi: 10.1111/j.1526-4610.2012.02212.x, PMID 22788114.

  17. Minton GC, Miller AD, Bookstaver PB, Love BL. Topiramate: safety and efficacy of its use in the prevention and treatment of migraine. J Cent Nerv Syst Dis. 2011;3:155-68. doi: 10.4137/JCNSD.S4365, PMID 23861645.

  18. Theeler BJ, Kenney K, Prokhorenko OA, Fideli US, Campbell W, Erickson JC. Headache triggers in the US military. Headache. 2010;50(5):790-4. doi: 10.1111/j.1526-4610.2009.01571.x, PMID 19925623.

  19. Somerville BW. The role of estradiol withdrawal in the etiology of menstrual migraine. Neurology. 1972;22(4):355-65. doi: 10.1212/wnl.22.4.355, PMID 5062827.

  20. Victor TW, HU X, Campbell J, White RE, Buse DC, Lipton RB. Association between migraine anxiety and depression. Cephalalgia. 2010;30(5):567-75. doi: 10.1111/j.1468-2982.2009.01944.x, PMID 19614684.

  21. GBD 2016 Headache Collaborators. Global regional and national burden of migraine and tension-type headache 1990-2016: a systematic analysis for the global burden of disease study 2016. Lancet Neurol. 2018;17(11):954-76. doi: 10.1016/S1474-4422(18)30322-3, PMID 30353868.

  22. Russell MB, Rasmussen BK, Fenger K, Olesen J. Migraine without aura and migraine with aura are distinct clinical entities: a study of four hundred and eighty-four male and female migraineurs from the general population. Cephalalgia. 1996;16(4):239-45. doi: 10.1046/j.1468-2982.1996.1604239.x, PMID 8792035.

  23. Stang PE, Carson AP, Rose KM, MO J, Ephross SA, Shahar E. Headache cerebrovascular symptoms and stroke: the atherosclerosis risk in communities study. Neurology. 2005;64(9):1573-7. doi: 10.1212/01.WNL.0000158326.31368.04, PMID 15883318.

  24. Silberstein SD, Hulihan J, Karim MR, WU SC, Jordan D, Karvois D. Efficacy and tolerability of topiramate 200 mg/d in the prevention of migraine with/without aura in adults: a randomized placebo-controlled double blind 12-week pilot study. Clin Ther. 2006;28(7):1002-11. doi: 10.1016/j.clinthera.2006.07.003, PMID 16990078.

  25. Moras K, Nischal H. Impact of amitryptiline on migraine disability assessment score. J Clin Diagn Res. 2014;8(9):KC01-2. doi: 10.7860/JCDR/2014/7546.4854, PMID 25386467.

  26. Dwajani S, Ugargol AP. Assessment of quality of life in patients with migraine receiving topiramate. Indian J Pharm Pharmacol. 2014;1(1):30-4.

  27. Laghari MJ, Qazi N, Jamali S, Mustafa G. Role of topiramate in improving quality of life in migraine sufferers versus gold standard propranolol. JLUMHS. 2012;11(1):11-4.

  28. El Hasnaoui A, Vray M, Richard A, Nachit Ouinekh F, Boureau F, MIGSEV Group. Assessing the severity of migraine: development of the MIGSEV scale. Headache. 2003;43(6):628-35. doi: 10.1046/j.1526-4610.2003.03105.x, PMID 12786922.