Kidney International, Vol. 63, Supplement 83 (2003), pp. S17–S21

Hypertensive risk factors in kidney disease inAfrican Americans

AGNES B. FOGO

Department of Pathology, Vanderbilt University Medical Center, Nashville, TN.

Hypertensive risk factors in kidney disease in African Ameri- ated nephrosclerosis has been associated with geneticallycans. African Americans with hypertension more commonly and/or environmentally induced amplification of profi-develop renal insufficiency compared to Caucasians. The Afri- brotic mechanisms, such as the renin-angiotensin systemcan American Study of Kidney Disease (AASK) included a re-

(RAS) and transforming growth factor-� (TGF-�) [10–13].nal biopsy pilot study that demonstrated that the clinical diag-The AASK study importantly confirmed that renalnosis of so-called hypertensive nephrosclerosis in these African

American patients indeed was accurate. This biopsy study dem- biopsies in African Americans with a clinical diagno-onstrated extensive global glomerulosclerosis, far exceeding sis of hypertensive nephrosclerosis show morphologicalthat expected for patients’ age. We further compared our clini- lesions consistent with this diagnosis. The typical mor-cally indicated renal biopsies to determine if any phenotypic dif-

phologic features of hypertensive nephrosclerosis areferences were present in hypertensive nephrosclerosis in Afri-vascular wall medial thickening with frequent arteriolarcan Americans versus Caucasians. These studies point to an

excess of the solidified type of global glomerulosclerosis (GS) hyaline deposits, a varying degree of intimal fibrosis and(also called “decompensated benign nephrosclerosis”) in Afri- focal glomerular ischemic changes with variable thick-can Americans compared to more obsolescent type GS in Cau-

ening and wrinkling of the basement membrane, and tu-casians. We speculate that these phenotypic differences mightbular atrophy and interstitial fibrosis, with varying footreflect differing mechanisms of tissue injury in hypertensive

African Americans versus Caucasians, and discuss possible process effacement. These changes of hypertensive neph-contributors to this injury. rosclerosis were widespread in nearly all cases in the

AASK biopsy study [14]. The extent of arterio- andarteriolosclerosis did not correlate with age in this study,

Hypertensive nephrosclerosis is one of the most com- supporting the thesis that a disease process rather thanmon causes of end-stage renal disease (ESRD) world- aging per se contributed to the changes. For obviouswide [1, 2]. This incidence is especially high in the USA ethical reasons, a control population is not included inin African Americans [2]. This difference cannot be ex- the AASK study. Therefore, we also studied our clinicalplained by the higher prevalence and greater severity renal biopsies to compare hypertensive nephrosclerosisof hypertension in African Americans nor by socioeco- lesions in biopsied African Americans and Caucasiansnomic status, nor less access to health care [3, 4]. Despite [15]. Clearly, in this retrospective study, clinical criteriasimilar blood pressure (BP), African Americans with hy- for renal biopsy determined the study population. Per-pertension and renal insufficiency have a more rapid fall neger et al have elegantly demonstrated that the pre-in renal function than Caucasians [5, 6], suggesting that sumptive, clinical diagnosis of hypertensive nephroscle-a lower level of BP control may be even more important rosis is made much more commonly in African Americansin African Americans than in Caucasians in slowing the than in Caucasians [16]. Therefore, it is expected that Afri-progression of renal disease [7, 8]. Indeed, the recent up- can Americans with hypertensive and renal insufficiencydate on the AASK intervention trial showed that BP would only be biopsied when the clinical situation wascontrol with an angiotensin I converting enzyme (ACE)- atypical for hypertension nephrosclerosis or was moreinhibitor could slow progressive loss of the glomerular severe. All cases from our renal biopsy files with a clinico-filtration rate (GFR) in African Americans with hyper- pathologic diagnosis of hypertension nephrosclerosis intensive nephrosclerosis [9]. Conversely, excess end organ the absence of other disease were reviewed. Cases withdamage in African Americans with hypertension-associ- hypertensive changes and in addition segmental glomer-

ulosclerosis lesions were included when primary FSGSwas excluded by usual diagnostic criteria. Sixty-two renalKey words: hypertensive nephrosclerosis, hypertension, race, global

glomerulosclerosis, AASK. biopsies met the criteria (19 African American and 43 Cau-casian). Global glomerulosclerosis was classified as either 2003 by the International Society of Nephrology

S-17

Fogo: Hypertensive nephrosclerosis and raceS-18

solidified or obsolescent type. Obsolescent glomeruli merulosclerosis in African American patients with hyper-tensive nephrosclerosis? We speculate that the coexistencewere defined as glomeruli in which Bowman’s space was

occupied by collagenous, periodic acid Schiff (PAS)-pos- of solidified glomeruli with segmental glomerulosclerosiscould reflect an increased vasoreactivity in susceptibleitive material, and the tuft was retracted, as first de-

scribed by McManus in 1960 as “ischemic obsolescence” populations. African Americans tend to show a larger in-crease in blood pressure level than Caucasians during[17]. Solidified glomeruli were defined as glomeruli in

which the entire tuft was solidified, in the absence of the stress, particularly to stimuli that elicit a vasoconstrictorresponse [20–22]. Autoregulatory responses to hyperten-collagenous change in the capsular space, as described

first by Fahr in 1925, and subsequently revisited by Bohle sion, resulting in vasospasm, could be excessive in patientssusceptible to developing hypertension, as is the case inand Ratschek, using the term “decompensated benign

nephrosclerosis” [18, 19]. some African Americans [23]. Thus, we speculate thatglobal vasoconstrictive ischemia could result in completeThe patients in this retrospective analysis were on

average 58.7 � 2.0 years. African Americans were younger collapse and solidification. Further, we hypothesize thatadaptation and segmental resistance to collapse and/orthan Caucasians, and had higher serum creatinine values.

Proteinuria and hypertension were similar in the two reflow, with possible attendant segmental hypertension/hyperfiltration, could give rise to segmental sclerotic le-populations. Total global glomerular sclerosis was exten-

sive in both African Americans and Caucasians (all pa- sions associated with solidification in those patients. Thus,vasospasm could reflect primary microvascular injurytients 42 � 3%; African Americans 50 � 5% vs. Cauca-

sians 39 � 4%, P � NS). This severity of overall global and endothelial dysfunction that then cause hyperten-sion, or occur in response to hypertension in susceptibleglomerulosclerosis was similar to that seen in the AASK

biopsy study, where patients on average were aged patients. It is possible that a primary renal microvasculardisease results in hypertension, glomerulosclerosis, and53.0 � 11.0 years, and had on average 43 � 26% global

glomerulosclerosis [14]. Surprisingly, the percent of so- interstitial fibrosis [4].In regards to a possible role of hypertension in thelidified glomeruli was significantly increased in African

Americans (all patients 13 � 2%; African Americans lesions, Kincaid-Smith has noted that blood vessels inbenign nephrosclerosis “do not show the type of lesions25 � 6% vs. Caucasians 8 � 2%, P � 0.01). In contrast,

the percent of obsolescent glomeruli was not different which deprives nephrons of their blood supply” [24].Interestingly, even in non-hypertensive renal disease, sig-between African-Americans and Caucasians (all patients

29 � 3%; African Americans 24 � 5% vs. Caucasians nificant vascular sclerosis was present in the majority ofpatients in a large renal biopsy series. The most severe31 � 4%, P � NS). The percent of solidified glomeruli

did not increase with aging, while the percent of obsoles- vascular sclerosis was present in patients with focal seg-mental glomerulosclerosis (FSGS), followed by membra-cent glomeruli did. Solidified glomeruli were associated

with segmental glomerulosclerosis, and with clinically nous glomerulopathy, and IgA nephropathy. Surpris-ingly, 76.5% of patients with chronic interstitial nephritisworse disease.

A multiple linear regression analysis was performed showed vascular sclerosis, with no difference in sclerosisin those with hypertension versus those with normalto examine if the clinical parameters of proteinuria, mean

arterial pressure (MAP), serum creatinine and age could blood pressure, implicating injury mechanisms otherthan systemic hypertension in this lesion [25].predict the severity of structural injury observed. Of

note, total global sclerosis and vascular sclerosis were Vascular sclerosis and glomerulosclerosis increase innormal populations with aging, but the rate varies widelypoorly predicted by this model (R2 � 0.16). MAP did

not contribute to the prediction by the model of any in different populations. The extent of global sclerosisin normal adults under the age of 50 years is estimatedmorphologic lesions, and proteinuria accounted only

minimally for the variability of global glomerulosclerosis. to be 1 to 3% of glomeruli, increasing up to 30% by age80 years in the USA [26, 27]. The Mexican populationIn summary, in this study, age and blood pressure were

not good predictors of the severity of structural injuries. showed less aortic fibroplasia and renal sclerosis at allages compared to the US population [28]. First and sec-Previously, analysis of the AASK data also showed that

the extensive global glomerulosclerosis (which has not ond generation Hispanic immigrants to the USA hadincreasing renal vascular sclerosis to intermediate levelsyet been analyzed in terms of specific type of GS) did not

correlate with vascular thickening, nor did the vascular between Mexico City natives and other USA residents[29]. In a similar morphologic study of Bolivian Indians,lesions correlate with the degree of hypertension [19].

Thus, both our retrospective biopsy study and the AASK only minimal vessel wall changes occurred with aging,even up to 80 years [30]. Elderly Japanese hypertensiverenal biopsy data do not support a simple linear relation-

ship between hypertension, vascular sclerosis, and ulti- patients showed only minimal glomerulosclerosis, evenless than expected for age in the USA Caucasian popula-mately global glomerulosclerosis.

What then could account for the extensive global glo- tion [31, 32]. Within the USA, vessel wall thickening in-

Fogo: Hypertensive nephrosclerosis and race S-19

creased with age in African Americans and was even subsequent renal and vascular sclerosis in adulthoodcould thus be linked to polymorphisms that also resultmore severe than in the Caucasian population [33].

Higher blood pressure was present in the African Ameri- in a developmental environment that affects glomerulardevelopment and growth. The RAS is one of severalcan population compared to the Caucasians in the New

Orleans region examined in this study, but these differ- mediators with effects on both embryonic and matura-tional development of the kidney, and response to injury.ences, if extrapolated to the autopsied patients in that

area, would not completely account for the greater vascu- Specific polymorphisms in angiotensin I converting en-zyme (ACE), angiotensinogen, and the angiotensin typelar wall thickening in the African American population.

These observations point to possible differences between 1 receptor genes have been linked with cardiovasculardisorders, including hypertension, myocardial infarctionpopulations, both in “normal” aging and in response to

injury, suggesting both environmental and genetic contri- and cardiac hypertrophy [41]. Familial clustering ofchronic renal failure has been noted in African Ameri-butions to the manifestations of disease.

Low birth weight, defined as birth weight less than cans [3, 42], and angiotensinogen gene mutations wereincreased in African Americans with essential hyperten-2.5 kg at term birth, may have both environmental and

genetic components. Low birth weight is more common sion, compared to Caucasian hypertensive populations[12]. The insertion/deletion (I/D) polymorphism of thein African Americans than in Caucasians. A linkage of

low birth weight to lower nephron number has been ACE gene also has been studied extensively. The Dpolymorphism has been associated with higher RAS ac-postulated to be a predictor of hypertension, cardiovas-

cular and renal disease in adulthood [34, 35]. The mecha- tivity, augmented response to angiotensin II, and pro-gression [43]. This D allele is significantly more frequentnism has been hypothesized to involve greater hemody-

namic stress on the fewer nephrons. In addition, when in African American populations compared to Cauca-sians [44, 45]. Interestingly, a study from Spain showednephrons are decreased in number, an increase in glo-

merular size, that is, glomerular hypertrophy, may occur. that hypertensive patients with impaired GFR more fre-quently were homozygous DD, compared to normalGlomerular hypertrophy is a sensitive marker for patho-

genic processes that lead to sclerosis [36]. Thus, we found ACE genotype distribution in hypertensive patients withpreserved renal function [46].that in patients with apparent minimal change disease

who subsequently showed overt focal segmental glomer- Pharmacological inhibition of the RAS inhibits matu-rational glomerular growth in animal models [47]. Angio-ulosclerosis on rebiopsy, glomerular size was markedly

larger than in age-matched patients with the typical tensinogen knockout mice have increased arteriosclero-sis, despite low blood pressures, and also have decreasedcourse of steroid-resistant minimal change disease [37].

We therefore examined whether glomerular size differed glomerular maturation and hypoplastic papillae, point-ing to a crucial role of the RAS in development [48].in normal African American and Caucasian populations

[38]. Data were gathered from age- and body mass index Excess RAS is equally deleterious to the kidney as itsabsence, pointing to a crucial role for a finely balanced(BMI)-matched sudden/traumatic deaths in patients

without cardiovascular disease. Glomerular volume was local RAS in kidney development and injury. In trans-genic mice with increased copy number of the ACE gene,significantly greater in African Americans versus Cauca-

sians, and the distribution pattern was different in the and thus increased RAS activity, abnormal structuraldevelopment, with small glomeruli and vascular sclerosis,two groups. The African American population showed

a Gaussian distribution of glomerular size. In contrast was also present [49]. The RAS also modulates the scle-rosis that occurs in association with aging in experimentalthe Caucasian population had a bi-modal distribution,

with a subgroup with enlarged glomerular size compara- models: chronic angiotensin inhibition treatment evenregressed age related sclerosis in rats in glomeruli andble to that in the African American population [38].

One can hypothesize that these larger glomeruli reflect blood vessels [50].In summary, renal biopsy data suggest that there maya decreased nephron number at birth. Of interest in

this regard are Spencer, Wang and Hoy’s findings of be differing phenotypes of glomerulosclerosis in AfricanAmericans versus Caucasians. The precise mechanismsmarkedly low birth weight in the Australian Aboriginal

population, not accounted for by prematurity [39]. Stud- underlying the severe global glomerulosclerosis have notbeen established. Potential mechanisms include decreasedies in Aboriginal patients with renal insufficiency have

revealed varied histopathology, with frequent lesions re- nephron number possibly related to genetically influencedgrowth factor profiles that also could influence fibroticlated to previous infection. However, of note, marked

glomerular enlargement is present in those biopsies that responses in adulthood, abnormal acceleration of aging,abnormal vasoreactivity or other primary microvascularshow sclerotic lesions [40].

Maturation and number of glomeruli could be influ- diseases. Whether altered genotypes for renin-angioten-sin affect vascular/glomerular structure from the earliestenced by functional polymorphisms of genes involved

both in renal/glomerular development and in scarring stages in utero to aging in the hypertensive AfricanAmerican population remains to be elucidated.mechanisms. Augmented fibrotic responses to injury and

Fogo: Hypertensive nephrosclerosis and raceS-20

Reprint requests to Agnes B. Fogo, M.D., MCN C3310, Department 21. Walker AJ, Bassett DR, Duey WJ, et al: Cardiovascular andof Pathology, Vanderbilt University Medical Center, Nashville, Tennes- plasma catecholamine responses to exercise in blacks and whites.see 37232-2561, USA. Hypertension 20:542–548, 1992E-mail: agnes.fogo@mcmail.vanderbilt.edu 22. Teiber FA, Musante L, Braden D, et al: Racial differences in

hemodynamic responses to the cold face stimulus in children andadults. Psychosom Med 53:286–296, 1990REFERENCES

23. Luke RG: Hypertensive nephrosclerosis: Pathogenesis and preva-lence. Nephrol Dial Transplant 14:2271–2278, 19991. Valderrabano F, Gomez-Campdera F, Jones EH: Hypertension

24. Kincaid-Smith P: Clinical diagnosis of hypertensive nephrosclero-as a cause of end-stage renal disease: Lessons from internationalsis. Nephrol Dial Transplant 14:2255–2267, 1999registries. Kidney Int 54(Suppl 68):S60–66, 1998

25. Meyrier A, Simon P, Montseny JJ, et al: Vascular nephropathies2. U.S. Renal Data System: USRDS 1999 Annual Data Report. Amand nephrosclerosis: Epidemiology and pathophysiology. AdvJ Kidney Dis 34(Suppl 1):S40–S151, 1999Nephrol Necker Hosp 26:207–245, 19973. Freedman BI, Spray BJ, Tuttle AB, Buckalew VM: The familial

26. Kappel B, Olsen S: Cortical interstitial tissue and sclerosed glo-risk of end stage renal disease in African Americans. Am J Kidneymeruli in the normal human kidney, related to age and sex. ADis 21:387–393, 1993quantitative study. Virchows Arch 387:271–277, 19804. Freedman BI, Iskandar SS, Appel RG: The link between hyper-

27. Smith SM, Hoy WE, Cobb L: Low incidence of glomerulosclerosistension and nephrosclerosis. Am J Kidney Dis 25:207–221, 1995in normal kidneys. Arch Pathol Lab Med 113:1253–1256, 19895. Schulman NB, Ford CE, Hall WD, et al: Prognostic value of

28. Tracy RE, Berenson GS, Cueto-Garcia L, et al: Nephrosclerosisserum creatinine and effect of treatment of hypertension on renaland aortic atherosclerosis from age 6 to 70 years in the Unitedfunction. Results from the hypertension detection and follow-upStates and Mexico. Virchows Arch A Pathol Anat Histopatholprogram. Hypertension Detection and Follow-up Program Coop-420:479–488, 1992erative Group. Hypertension 13(Suppl 1):80–93, 1989

29. Tracy RE, Rios-Dalenz JL: Renovasculopathies of hypertension6. Klag MJ, Whelton PK, Randall BL, et al: End stage renalin Hispanic residents of Dallas, Texas. Arch Med Res 30:40–48,disease in African-American and white men. 16 year MRFIT find-1999ings. JAMA 277:1293–1298, 1997

30. Tracy RE, Rios-Dalenz JL: Rarity of hypertensive stigmata in7. Rostand SG, Brown G, Kirk KA, et al: Renal insufficiency inaging renocortical arteries of Bolivians. Virchows Arch 424:307–treated essential hypertension. N Engl J Med 320:684–688, 1989314, 19948. Hebert LA, Kusek JW, Greene T, et al: Effects of blood pressure

31. Tamura T: Histologic features of renal biopsies from patients withcontrol on progressive renal disease in Blacks and Whites. Modifi-essential hypertension and from the aged. Japanese Circ J 30:829–cation of Diet in Renal Disease Study Group. Hypertension 30:428–862, 1966435, 1997

32. Katafuchi R, Takebayashi S: Morphometrical and functional cor-9. Agodoa LY, Appel L, Bakris GL, et al: Effect of ramipril vsrelations in benign nephrosclerosis. Clin Nephrol 28:238–243, 1987amlodipine on renal outcomes in hypertensive nephrosclerosis: A

33. Tracy RE: Renovasculopathies of hypertension and the rise ofrandomized controlled trial. African American Study of Kidneyblood pressure with age in blacks and whites. Semin NephrolDisease and Hypertension (AASK) Study Group. JAMA16:126–133, 1996285:2719–2728, 2001

34. Brenner BM, Chertow GM: Congenital oligonephropathy and10. Suthanthiran M, Khanna A, Cukran D, et al: Transformingthe etiology of adult hypertension and progressive renal injury.growth factor-beta 1 hyperexpression in African American end-Am J Kidney Dis 23:171–175, 1994stage renal disease patients. Kidney Int 53:639–644, 1998

35. Hack M, Merkatz IR: Preterm delivery and low birth weight –11. Li B, Khanna A, Sharma V, et al: TGF-beta1 DNA polymor-A dire legacy. N Eng J Med 333:1772–1773, 1995phisms, protein levels, and blood pressure. Hypertension 33:271–

36. Fogo AB: Glomerular hypertension, abnormal glomerular growth,275, 1999and progression of renal diseases. Kidney Int 57(Suppl 75):S15–S21,12. Bloem LJ, Manatunga AK, Tewksbury DA, Pratt JH: The serum2000angiotensinogen concentration and variants of the angiotensinogen

37. Fogo A, Hawkins EP, Berry PL, et al: Glomerular hypertrophygene in white and black children. J Clin Invest 95:948–953, 1995in minimal change disease predicts subsequent progression to focal13. Duru K, Farrow S, Wang J-M, et al: Frequency of a deletionglomerular sclerosis. Kidney Int 38:115–123, 1990polymorphism in the gene for angiotensin converting enzyme is

38. Pesce C, Schmidt K, Fogo A, et al: Glomerular size and the inci-increased in African Americans with hypertension. Am J Hypertensdence of renal disease in African Americans and Caucasians. J7:759–762, 1994Nephrol 7:355–358, 199814. Fogo A, Breyer JA, Smith MC, et al: Accuracy of the diagnosis

39. Spencer J, Wang Z, Hoy W: Low birth weight and reduced renalof hypertensive nephrosclerosis in African Americans: A reportvolume in Aboriginal children. Am J Kidney Dis 37:915–920, 2001from the African American Study on Kidney Disease (AASK)

Trial. AASK Pilot Study Investigators. Kidney Int 51:244–252, 1997 40. Young RJ, Hoy WE, Kincaid-Smith P, et al: Glomerular size andglomerulosclerosis in Australian Aborigines. Am J Kidney Dis15. Marcantoni C, Ma L-J, Federspiel C, Fogo AB: Hypertensive

nephrosclerosis in African Americans versus Caucasians. Kidney 36:481–489, 200041. Teio KK: Angiotensin converting enzyme genotypes and disease.Int 62:172–180, 2002

16. Perneger TV, Whelton PK, Klag MJ, Rossiter KD: Diagnosis BMJ 311:76–77, 199542. Ferguson R, Grim CE, Opgenorth TJ: A familial risk of chronicof hypertensive end stage renal disease: Effect of patients’ race.

Am J Epidemiol 141:10–15, 1995 renal failure among blacks on dialysis. J Clin Epidemiol 41:1189–1196, 198817. Mcmanus JFA, Lupton CH Jr: Ischaemic obsolescence of renal

glomeruli: The natural history of the lesions and their relation to 43. Hunley TE, Julian BA, Phillips JA, et al: Angiotensin convertingenzyme gene polymorphism: Potential silencer motif and impacthypertension. Lab Invest 9:413–434, 1960

18. Fahr T: Pathologische Anatomie des Morbus Brightii, in Hand- on progression in IgA nephropathy. Kidney Int 49:571–577, 199644. Rutledge DR, Kubilis P, Browe CS, Ross EA: Polymorphism ofbuch de Speziellen Pathologische Anatomie und Histologie (vol 6,

part 1), edited by Henke F, Lubarsch O, Berlin, Springer, 1925, the angiotensin I converting enzyme gene in essential hypertensivepatients. Biochem Mol Biol Int 35:661–668, 1995p 156�/6�

19. Bohle A, Ratschek M: The compensated and the decompensated 45. Duru K, Farrow S, Wang J-M, et al: Frequency of a deletionpolymorphism in the gene for angiotensin converting enzyme isform of benign nephrosclerosis. Pathol Pract Res 174:357–367, 1982

20. Pickering TG: Hypertension in blacks. Curr Opin Nephrol Hyper- increased in African Americans with hypertension. Am J Hypertens7:759–762, 1994tens 3:207–212, 1994

Fogo: Hypertensive nephrosclerosis and race S-21

46. Fernandez-Llama P, Poch E, Oriola J, et al: Angiotensin con- maintenance of kidney morphology and growth factor regulation.J Clin Invest 96:2947–2954, 1995verting enzyme gene I/D polymorphism in essential hypertension

and nephroangiosclerosis. Kidney Int 53:1743–1747, 1998 49. Krege JH, Kim HS, Moyer JS, et al: Angiotensin-converting en-zyme gene mutations, blood pressures, and cardiovascular homeo-47. Fogo A, Yoshida Y, Yared A, Ichikawa I: Importance of angio-

genic action of angiotensin II in the glomerular growth of maturing stasis. Hypertension 29:150–157, 199750. Ma L-J, Nakamura S, Whitsitt JS, et al: Regression of sclerosiskidneys. Kidney Int 38:1068–1074, 1990

48. Niimura F, Labosky PA, Kakuchi J, et al: Gene targeting in mice in aging by an angiotensin inhibition-induced decrease in PAI-1.Kidney Int 58:2425–2436, 2000reveals a requirement for angiotensin in the development and