Headache attributed to ingestion of a cold stimulus (ICHD-3 beta 4.5.1) is also known as ice cream headache (ICH). This cross-sectional epidemiological study included 283 students (10-14-year-olds) attending a grammar school in Germany, their parents (n = 401), and 41 teachers. A self-administered questionnaire was used to analyze the prevalence and characteristics of ICH based on the ICHD classification. Additionally, the association between ICH and other headaches was investigated in students and parents. Prevalence of ICH in students was 62 % without gender difference. In adults, only 36 % of females and 22 % of males reported ICH. There was an increased risk for ICH in students when mother (OR 10.7) or father (OR 8.4) had ICH. Other headaches in parents had no influence on the prevalence of ICH in students. However, in the groups of students and parents itself there was a highly significant association between ICH and other headaches (students: OR 2.4, mothers: OR 2.9, fathers: OR 6.8). There was a decreased risk for ICH when parents and students had no headache at all (OR < 0.4). ICH in students clearly shows a familial disposition by both father and mother. There was also an association between ICH and other headaches within the student and adult groups. The absence of headache history seems to be a protective factor for ICH.

Background There are various studies on experimentally provoked 'ice-cream headache' or 'headache attributed to ingestion or inhalation of a cold stimulus' (HICS) using different provocation protocols. The aim of this study was to compare two provocation protocols. Methods Ice cubes pressed to the palate and fast ingestion of ice water were used to provoke HICS and clinical features were compared. Results The ice-water stimulus provoked HICS significantly more often than the ice-cube stimulus (9/77 vs. 39/77). Ice-water-provoked HICS had a significantly shorter latency (median 15 s, range 4-97 s vs. median 68 s, range 27-96 s). There was no difference in pain localisation. Character after ice-cube stimulation was predominantly described as pressing and after ice-water stimulation as stabbing. A second HICS followed in 10/39 (26%) of the headaches provoked by ice water. Lacrimation occurred significantly more often in volunteers with than in those without HICS. Discussion HICS provoked by ice water was more frequent, had a shorter latency, different pain character and higher pain intensity than HICS provoked by ice cubes. The finding of two subsequent HICS attacks in the same volunteers supports the notion that two types of HICS exist. Lacrimation during HICS indicates involvement of the trigeminal-autonomic reflex.