Hemorrhagic Stroke

Review of the Literature

A study of stroke trends in Minnesota found that the incidence rate of ICH declined from 209 to 115 per 100,000 population between 1945 and 1975, but then increased again by 17% between 1975 and 1984.The onset of the initial decline coincided with the introduction of antihypertensive therapy, while the subsequent increase in the incidence coincided with the introduction of computed tomography (CT).CT most likely increased the detection of less severe strokes.ICH mortality rates declined from 91% to 48% from 1945 to 1984.This decline occurred during the 10 years when CT was first introduced (1976) and was due to the identification of smaller hemorrhages.(1)

Similarly, a Swedish population study between 1971 and 1987 found an increase in ICH incidence and a decrease in ICH mortality, most likely due to better detection of ICH with CT.The decline in mortality was credited to decreases in smoking habits and blood pressure.During this time, there was an increase in the percent of middle-aged men and some women in Sweden taking antihypertensive medications.(2)

In Germany, related trends in ICH mortality and incidence have been shown.A decrease by ½ of hospital mortality was seen between 1975-1990.Scientists believe that modern imaging techniques like CT allowed for easier and better diagnosis of smaller hematomas.As a result, patients were entering the hospital with less severe initial symptoms and, thus, more likely to survive.(3)

A 24-year follow-up study of Japanese men in Hawaii, the Honolulu Heart Program, showed a sharp decline in stroke mortality since the late 1960s.ICH incidence declined 4.2% and mortality fell significantly.The 30-day case fatality rate for ICH fell from 75% to 29%.The Honolulu Heart Program found that the control of hypertension was the most likely reason for the decline, since it was the strongest and most consistent risk factor.Decreases in smoking and improvements in diagnosis and treatment were also thought to account for some of the decreases.(4,5)

A 9-year study of 28-day case fatality rates of all strokes in Finland found comparable results to the Honolulu study.Fatality rates between 1983 and 1992 fell yearly by 3.6% in men, which was significant, and 2.6% in women, not significant.ICH fatality rates were 49% in women and 42% in men, a significant decline for men and a trend for women.Declines were attributed in part to a decrease in blood pressure due to treatment with antihypertensive medications.(6)

Seasonal trends have also been studied since it was hypothesized that low temperatures could influence blood pressure fluctuations.A study in Finland from 1982-1992 showed a 28% greater ICH rate among men and a 33% greater ICH rate among women in winter compared to summer.(7)An 8-year retrospective study in Brussels showed that the highest incidence of ICH occurred from November to December (23%), and the lowest occurred from July to August (10%).However, this study found no difference between hypertensive and normotensive patients.Therefore, it can be concluded that there may be seasonal variation, but it is most likely not due to the influence of the temperature on blood pressure.(8)More studies need to be done to determine the cause of the seasonal trends.

National and State Mortality Trends

I. Time Trends

Figure 1 shows age adjusted mortality rates of ICH in the United States and South Carolina between 1979 and 1998.US mortality rates slightly declined and then stabilized during this time.This trend is most likely due to better and earlier detection of ICH with computed tomography.Mortality rates in South Carolina are higher than in the US but have also declined since 1979.In 1979 the ICH mortality rate in South Carolina was approximately 25 per 100,000, compared to the US rate of approximately 17 per 100,000.In 1998 the ICH mortality rate in South Carolina was approximately 15 per 100,000, compared to the US rate of approximately 12 per 100,000.

II. Race and Sex Trends

Figures 2 and 3 demonstrate race and sex trends for the US and SC, respectively, between 1979-1998.Note that the scales on these two figures differ, as SC has higher ICH mortality rates compared to the US.In both the US and SC there has been little change in mortality of white males and females.A sharp decline in mortality among black females was seen in the US and has a similar rate to white females in 1998.Black males have consistently higher rates than all other groups in both the US and SC.In 1979 black males in the US were almost twice as likely to die of ICH compared to white males.In 1998, black males were approximately 1.5 times more likely to die of ICH compared to white males.Black males in SC in 1979 were approximately four times more likely to die of ICH compared to white males. In 1998, black males were about 1.5 times more likely to die from ICH compared to white males.So there has been a decline in mortality for all race and sex groups, although this decrease was less dramatic for white males and females.And, perhaps more importantly, there has been a decrease in the racial discrepancies, especially in South Carolina.

III. Age Trends

Figure 4 shows mortality rates per 100,00 by age.Hemorrhagic stroke rates increase with age.It is important to note the consistently higher rates are seen South Carolina.It is also interesting to see that the mortality rates for the 35-44 age group in SC is almost twice as high compared to the US rates.This is most likely due to the large population of African Americans in SC and to the high prevalence of hypertension in African Americans in SC.

Figure 5 demonstrates the mortality rates by age, sex, and race in the US and in SC.As noted before, black males have the highest mortality rates from hemorrhagic stroke in the US and SC, followed by black females, white females, and, finally, white males.More importantly, however, is the distribution of mortality in the various age groups.Black males have much higher mortality rates at younger ages in SC compared to whites.Studies should look at possible reasons for this early onset.Note the high mortality rates among black men aged 35-44, 45-54, and 55-64 in South Carolina.There are essentially no white men between 35-54 dying from ICH.Again, this trend is most likely due to the high prevalence of hypertension among African American men in these age groups.

IV. Geographic Trends

Figures 6-10 show mortality rates by state.Crude rates and age-adjusted rates for age and race groups are shown.In each situation, mortality rates are highest in the Southeast region.For black males and females, the numbers of deaths outside the Southeast and a few other states (shown in white) were so small as to not have a valid mortality rate for ICH.Some explanation seen for the geographic trends could be that these areas have a higher population of African Americans and a higher prevalence of hypertension.

Figure 6:1998 Crude ICH Mortality Rates per 100,000 for US, Ages 25 and older

Figure 7:1998 Age-Adjusted ICH Mortality Rates per 100,000 for White Males, 25 and Older

Figure 8:1998 Age-Adjusted ICH Mortality Rates per 100,000 for White Females, 25 and Older

Figure 9:1998 Age-Adjusted ICH Mortality Rates per 100,000 for Black Males, 25 and Older

Figure 10:1998 Age-Adjusted ICH Mortality Rates per 100,000 per Black Females, 25 and Older

V. South Carolina Morbidity Data

 

Total Discharge Rate of ICH in SC
1998	285/100,000
1999	278/100,000

Figure 11 demonstrates that the incidence of ICH in South Carolina has not changed much in the past 2 years and is higher than the rates shown in the Minnesota study of 1985.South Carolina continues to have high rates of hemorrhagic stroke.Although mortality may have declined, morbidity remains an important issue.

 

1999 Discharge Rate of ICH in SC, by Race and Type of Insurance Coverage, for ages 18-64
	Black	White
HMO	301/100,000	170/100,000
Medicaid	168/100,000	88/100,0,00
Medicare	220/100,000	95/100,000
Self/Indigent	529/100,000	256/100,000

Figure 11 shows the various trends in incidence according to type of insurance.This could answer the question of whether socioeconomic status plays a role in the incidence of ICH.For both blacks and whites, indigent and self-pay patients have the highest hospitalization rates.Black rates are higher for every category of insurance.These trends give credence to the hypothesis that SES is a risk factor for ICH.It is likely that this category is mostly indigent, although this information was not provided.

 

1999 Discharge Rate of ICH in SC, by Health Service Area, for ages 18-64
HSA1	205/100,000
HSA2	219/100,000
HSA3	150/100,000
HSA4	222/100,000

Figure 12 provides incidence rates for the various health service areas in South Carolina.HSA1 consists of the Appalachia / Upstate counties and include Anderson, Cherokee, Greenville, Oconee, Pickens, Spartanburg and Union Counties.HSA2 consists of the Three Rivers / Midlands counties, including Abbeville, Aiken, Chester, Edgefield, Fairfield, Greenwood, Kershaw, Lancaster, Laurens, Lexington, McCormick, Newberry, Richland, Saluda, and York Counties.HAS3 consists ofPee Dee, Chesterfield, Clarendon, Darlington, Dillon, Florence, Georgetown, Horry, Lee, Marion, Marlboro, Sumter and Williamsburg Counties. Finally, HSA4 includes the Palmetto / Low Country counties of Allendale, Bamberg, Barnwell, Beaufort, Berkeley, Calhoun, Charleston, Colleton, Dorchester, Hampton, Jasper, and Orangeburg Counties. Since there are no major differences between these areas, it cannot be concluded that there are major geographic differences within the state.Therefore, something else is occurring in South Carolina to account for the higher mortality and incidence rates compared to the US.

Conclusion

References

1.Broderick JP.Stroke Trends in Rochester, Minnesota, during 1945 to 1984.Annals of Epidemiology.3(5):476-9, 1993

2.Harmsen P, Tsipogianni A, and Wilhelmsen L.Stroke incidence rates were unchanged, while fatality rates declined, during 1971-1987 in Goteborg, Sweden.Stroke.23(10):1410-5, 1992.

3.Schutz H.Spontaneous intracerebral hemorrhage.A disease in transition. [Review]Nervenarzt.63(2):63-73, 1992.

4.Yano K, Popper JS, Kagan A, Chyou PH, and Grove JS.Epidemiology of stroke among Japanese men in Hawaii during 24 years of follow-up:the Honolulu Heart Program.Health Reports.6(1):28-38, 1994.

5.Kagan A, Poper J, Reed DM, MacLean CJ, and Grove JS.Trends in stroke incidence and mortality in Hawaiian Japanese men.Stroke.25(6):1170-5, 1994.

6.Immonen-Raiha P, Mahonen M, Tuomilehto J, et al.Trends in case-fatality of stroke in Finland during 1983 to 1992.Stroke.28(12):2493-9, 1997.

7.Jakovljevic D, Salomaa V, Sivenius J, et al.Seasonal variation in the occurrence of stroke in a Finnish adult population.The FINMONICA Stroke Register.Finnish Monitoring Trends and Determinants in Cardiovascular Disease.Stroke.27(10):1774-9, 1996.

8.Capon A, Demeurisse G, Zheng L.Seasonal variation of cerebral hemorrhage in 236 consecutive cases in Brussels.Stroke.23(1):24-7, 1992.