There are four basic ways to classify hypertensive individuals. The first uses only the level of BP—DBP, SBP, or both. Such a system is based on the increase in the relative risk (the proportional likelihood) of cardiovascular events occurring as BP rises. Estimates of relative risk have been obtained from observational studies, such as the Framingham Heart Study.

The risks and ultimate outcomes in hypertensive individuals, however, are much more closely related to the presence of other cardiovascular risk factors and/or TOD than to BP level.5 Therefore, other systems based on absolute risk have emerged.Absolute risk—the actual odds that a patient (or population) will develop an event—reflects the sum of all the factors that contribute to the risk of cardiovascular disease. Such classification systems use BP level but also account for the effect of TOD and/or other cardiovascular risk factors on prognosis.

A third type of system classifies hypertensive individuals on the basis of the expected benefit of treatment as demonstrated in clinical trials of antihypertensive therapy. Although most systems rely heavily on such data, none use it exclusively. This approach might be more cost-effective but would be exceedingly difficult to validate and would not necessarily be superior to a system based on relative or absolute risk. It would be more complicated to use and may not be any more generalizable than current systems. Many populations with hypertension have not been adequately represented in clinical trials, and a substantial amount of extrapolation of the relative value of treating these individuals would still be necessary. Furthermore, clinical trials are usually efficacy studies (data obtained from a nonrepresentative cohort treated in an idealized fashion) rather than effectiveness studies (treatment of “real” patients under “real-world” conditions), and so the data obtained may not apply to the general hypertensive population.

The final way to classify hypertensive individuals would be to use some genetic, biochemical, or physiological parameter such as plasma renin activity or the renin-sodium index to denote those at high risk.10 Although in some studies a high plasma renin activity or renin-sodium index was shown to confer a poor prognosis, the use of such parameters for the selection of high-risk hypertensive individuals and choice of therapy has not fulfilled its promise and has not replaced more traditional classification systems. Whether genetic markers, such as angiotensin-converting enzyme gene polymorphism, will significantly improve our ability to select which people are at risk remains to be proved.

Since no classification system in common use relies only on clinical trial–derived evidence of benefit or on clinical, biochemical, or genetic markers of risk, these approaches will not be discussed further.

Systems Based on Relative Risk (Use of BP Level Alone)

Since we were first able to accurately measure BP noninvasively, we have defined and classified hypertension on the basis of BP level alone. This is appropriate because the rate of cardiovascular events rises as BP rises. In 1977, the initial report of the Joint National Committee on the Detection, Evaluation, and Treatment of High Blood Pressure (JNC-I) classified hypertension only by DBP. In JNC-II (1980), hypertension was first classified in the United States as mild (DBP from 90 to 104 mm Hg), moderate (DBP from 105 to 114), or severe (DBP ≥115). SBP was again not considered. JNC-III in 1984 refined the classification system further by introducing the term high-normal for those individuals with DBP values from 85 to 89 mm Hg, recognizing the excess risk for cardiovascular disease and premature mortality in individuals with this DBP level. JNC-III also introduced the terms isolated systolic hypertension and borderline isolated systolic hypertension for individuals with SBP ≥160 mm Hg and 140 to 159 mm Hg, respectively, and DBP <90 mm Hg. This belatedly acknowledged the evidence that individuals with elevated SBP were at increased risk of cardiovascular disease regardless of DBP level. We had known for some time that SBP was a better predictor than DBP of mortality and virtually all of the cardiovascular complications attributable to hypertension.

Systems Based on Absolute Risk (Use of BP and/or Comorbidity)

Since Janeway, authors have classified hypertensive individuals primarily by the amount of TOD noted, particularly in the optic fundus and kidney. Keith et al divided hypertensive individuals into four groups, now called grades, on the basis of retinal arteriolar changes and the presence or absence of exudates, hemorrhages, or papilledema. This system was very effective at defining prognosis. Individuals with group IV retinopathy (hemorrhages, exudates, and papilledema) had a dramatically worse prognosis, with a mortality rate of 76% at 1 year, than those in groups I and II (with only arterial tortuosity or arteriovenous nicking), in which only 10% to 12% died in 1 year.

In 1993, JNC-V modified the classification systems from earlier reports in two very important ways. It not only provided arbitrary cut points for both SBP and DBP, including both or either in the definition of hypertension, but also added individuals with SBPs of 130 to 139 mm Hg to the group with high-normal BP and specified <120 mm Hg SBP and <80 mm Hg DBP as optimal BP. The data from the Multiple Risk Factor Intervention Trial (MRFIT) screenees made it obligatory to recognize SBP and highlight the risk in individuals with SBP from 120 to 139 mm Hg.

JNC-V also adopted the word “stage” to categorize hypertension severity, abandoning the designations of borderline, mild, moderate, and severe. The committee felt that the term “mild” hypertension implied to both providers and patients that hypertensive individuals with BPs in that range had a benign condition that did not require excessive concern. Although the risk of complications from an elevated BP is certainly higher and more immediate in people with stage III or IV hypertension (moderate and severe hypertension in the old nomenclature), the majority of events occur in those with stage I or II hypertension (and in those with high-normal BP readings) because so many more people have this BP level.  The data from the MRFIT screenees reiterated the importance of other risk factors and TOD on absolute risk. Although JNC-V did note the importance of comorbidity and detailed the need for specific mention of whether TOD or other risk factors were present, such was not emphasized in that classification system and few clinicians appear to have adopted that recommendation.

The 1993 revision of the WHO/ISH guidelines retained the terms “mild,” “moderate,” and “severe” and also “borderline” hypertension and expanded the staging system that had been used since 1962. Individuals with stage I hypertension have no objective signs of TOD. Those with stage II have one or more of the following: grade I or II retinopathy, proteinuria, and/or plasma creatinine levels between 1.2 and 2.0 mg/dL and/or evidence of atherosclerosis. Stage III includes individuals with symptoms and signs of more advanced TOD than those with stage II. This system is valuable because it forces the clinician to consider TOD and other risk factors rather than just BP, but it is too complicated for general use and may require extensive and expensive testing for establishment of the degree of TOD. Furthermore, the WHO/ISH system classifies hypertensive individuals by both BP level and TOD and does not provide adequate guidance as to how to use either in risk assessment.

The National Advisory Committee on Core Health and Disability Support Services from New Zealand tied estimates of absolute risk to therapeutic recommendations and strongly advised that hypertension treatment be based on absolute risk rather than BP level. Individuals with an absolute risk for a cardiovascular event of more than 20% in 10 years (for example, a 60-year-old man with a BP of 160/95 mm Hg and two other risk factors) should be treated, whereas someone whose risk is lower (a 40-year-old woman whose BP is 170/100 mm Hg with one other risk factor) should not. The risk predictions are based on data from the Framingham Heart Study that demonstrate the substantial increase in the absolute risk of cardiovascular disease events in individuals with SBP >=170 mm Hg and/or DBP >=100 and also indicate the effect that other cardiovascular risk factors, TOD, and symptomatic cardiovascular disease have on prognosis. Unfortunately, the value of this classification system and the recommendations on treatment that result are weakened because Framingham Heart Study risk predictions may not apply to other populations and are not universally accepted or widely enough known to be used in individualized therapy.

References
#  Rose G. Epidemiology. In: Marshall AJ, Barritt DW, eds. The Hypertensive Patient. Kent, UK: Pitman Medical; 1980:1-21.
# Alderman MH, Cushman WC, Hill MN, Krakoff LR. International round table discussion of national guidelines for the detection, evaluation, and treatment of hypertension. Am J Hypertens. 1993;6:974-981.[Medline] [Order article via Infotrieve]
# MacMahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J, Abbott R, Godwin J, Dyer A, Stamler J. Blood pressure, stroke, and coronary heart disease, part 1: prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet. 1990;335:765-774.[Medline] [Order article via Infotrieve]
# Kannel WB, Schwartz MJ, McNamara PM. Blood pressure and risk of coronary heart disease: The Framingham Study. Dis Chest. 1969;56:43-52.
# Jackson R, Barham P, Bills J, Birch L, McLennan L, MacMahon S, Maling T. Management of raised blood pressure in New Zealand: a discussion document. BMJ. 1993;307:107-110.
# Joint National Committee on the Detection, Evaluation and Treatment of High Blood Pressure. The fifth report of the Joint National Committee on the detection, evaluation, and treatment of high blood pressure (JNC-V). Arch Intern Med. 1993;153:154-183.[Medline] [Order article via Infotrieve]

Henry R. Black; Jong-Yoon Yi
the Department of Preventive Medicine, Rush–Presbyterian–St Luke’s Medical Center, Rush Medical College of Rush University, Chicago, Ill.

Chobanian AV, Bakris GL, Black HR, et al. Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension. 2003;42(6):1206-52. [Medline].