Epidemiology of Stroke in the Young
To the Editor:
We read with interest the article by Jacobs et al1 on stroke in the young. Their epidemiological work was carried out within the frame of the Northern Manhattan Stroke Study, a population-based registry, with comparisons between younger and older patients on demographic characteristics, risk factors, and prognosis, and with computation of the selective contribution of the younger patients to the burden of stroke. We conducted between 1994 and 1998 a similar study on 89 patients under 45 years of age with a first-ever stroke documented by brain neuroimaging, in a defined population of central Italy.2 Despite the inclusion in our study only of white subjects, we would like to make some comments and comparisons between the 2 studies.
The overall stroke incidence in the white residents of the Northern Manhattan area was 10/100 000 per year. This figure was identical to that found in our study (10.2/100 000) and within the range of studies on subjects under 45 years of age (8.8 to 15/100 000) quoted by Jacobs et al. However, if we refer to the studies quoted in our article, that were all performed with a comparable methodology, the range was even tighter (9.3 to 14.8/100 000) and, if those are the true figures of stroke incidence in young whites, the much higher rates observed among Hispanics (26/100 000) and blacks (25/100 000) appeared even more impressive. A higher migration rate of Hispanic and black residents in the study area together with a higher fertility rate with respect to the white residents might have contributed to a disproportionate increase of younger age groups in those populations. In fact, we already showed that any variation of the age distribution may affect stroke incidence not only as a predictable consequence of changes in the proportion of elderly individuals within that population but also because of a variation of the stroke risk in the different age groups.3 However, since patients eligibility relied on residency within the study area at least for 3 months before the event, while the study referral population was that reported by the 1990 US census, a biased estimation of the rate denominator might also have occurred for Hispanics and blacks. Degree of stability over time of the dynamic population should have been considered, accordingly.
The proportion of strokes observed in whites under 45 years of age by Jacobs et al (2.5%) was almost identical to that found in our study (2.6%), while the corresponding proportions were 12.4% for Hispanics and 4.4% for blacks. Whichever was the factor increasing stroke incidence among younger Hispanics and blacks, it should have accounted for an anticipation of stroke occurrence rather than for an increase of the global stroke risk which should have been equally distributed among the younger and older age groups.
Unexpectedly, Jacobs et al found similar 30-day case fatality rates in patients under (17%) and over (16%) 45 years of age, while in our study, the 30-day case fatality rate was lower (P=0.005) in subjects under 45 years of age (11.2%) than in those over 45 years of age (26.0%). Since the case fatality rate is notoriously higher for intracerebral and subarachnoid hemorrhage with respect to ischemic stroke4 and the proportion of ischemic stroke was lower in patients under 45 years of age than in those over 45, a higher incidence of hemorrhagic stroke in the study of Jacobs et al might have contributed to the lack of any difference in case fatality between younger and older patients.
Jacobs BS, Boden-Albala B, Lin I-F, Sacco RL. Stroke in the young in the northern Manhattan stroke study. Stroke. 2002; 33: 2789–2793.
Marini C, Totaro R, De Santis F, Ciancarelli I, Baldassarre M, Carolei A. Stroke in young adults in the community-based L’Aquila registry: incidence and prognosis. Stroke. 2001; 32: 52–56.
Leys D, Bandu L, Henon H, Lucas C, Mounier-Vehier F, Rondepierre P, Godefroy O. Clinical outcome in 287 consecutive young adults (15 to 45 years) with ischemic stroke. Neurology. 2002; 59: 26–33.
We appreciate the interest of Drs Marini and Carolei in our manuscript describing stroke in the young in Northern Manhattan.1 They correctly point out that instability of a population over time may lead to incorrect incidence rates. Unfortunately, we were unable to account for the possibility of an effect of population change in our study, particularly specific to race-ethnicity. While the incident strokes were surveyed from the years 1993 to 1997, the 1990 United States census of the population was used in determining incidence rates. This was the most accurate method available to use, particularly since there has not yet been available follow-up census data of the specific areas of Northern Manhattan by race-ethnicity. While this may soon become available from the year 2000 census, it may not be identically comparable due to some changes in the manner of census definitions. However, we do believe relying on the 1990 census data are the best representation of the population available at this time to calculate the incidence of stroke in this population. Drs Marini and Carolei raise concern of an effect of a “higher migration rate of Hispanic and black residents in the study area together with a higher fertility rate with respect to the white residents.” Although such an effect may alter race-ethnic specific incidence rates, without census data we cannot assume such an effect has occurred within the specified time period.
They also indicate that young patients represented a greater proportion of all strokes in blacks and Hispanics compared with whites. Pondering such proportions does not take into account the population size of the older and younger groups in which these events occurred. Based on these proportions, we cannot conclude a disproportionately greater incidence of stroke in these younger groups nor make conclusions regarding a disproportionately increased risk in these younger groups. Even if the incidences in older blacks and Hispanics were disproportionately lower compared with whites, this could simply be consistent with less race-ethnic disparity in the older age group.
We agree that the higher proportion of hemorrhagic stroke among all strokes in the young in our study may have contributed to the similar case fatality rates found in the younger and older age groups. However, the case fatality rates of the young stroke patients need to be interpreted with caution since the number of deaths in this group was so small.