Postgrad Med J 1997; 73: 230 - 233 (C The Fellowship of Postgraduate Medicine, 1997

The significance of arterial at the Postgrad Med J: first published as 10.1136/pgmj.73.858.230 on 1 April 1997. Downloaded from onset of clinical nephritis

Indiran P Naiker, Vivien Chrystal, Ismail GH Randeree, Yackoob K Seedat

Summary surrogate measure of underlying glomerulone- The prognostic importance of hyperten- phritis. Furthermore there has been a report of sion at the onset ofclinical a dissociation of hypertension and lupus is not well established. We studied retro- nephritis in patients with systemic lupus spectively 44 patients with lupus nephritis erythematosus,5 and it is well known that in order to ascertain the prevalence of hypertension may be absent even in patients hypertension at presentation and to in- with aggressive lupus histology and significant vestigate a possible association between renal functional impairment.5 This retrospec- hypertension and renal functional impair- tive study was therefore undertaken in order to ment. A correlation was also sought ascertain the prevalence of hypertension at the between hypertension and histological onset of clinical lupus nephritis. We also class of lupus nephritis. Hypertension investigated a possible association between was graded as mild (diastolic 95 - 99 hypertension and renal vascular lesions thereof mmHg), moderate (100-114) or severe with renal functional impairment. Finally, we (>115). Impaired renal function (creati- searched for a correlation between hyperten- nine >120 umoMl) was graded as mild sion and histological class of lupus nephritis. (120-200 pmolll), moderate (200-350 It was decided to explore these associations jimol/l), or severe (>350 jymol/l). Histolo- on presentation ofthe disease, as the prognostic gical class and the presence of hyperten- importance of hypertension at the onset of sive renal vascular lesions was recorded. nephritis has not been well established.4 The prevalence of hypertension was Furthermore, while the detrimental effect of 38%. There were 17 hypertensives and 27 hypertension on renal function late in the normotensives. The incidence of renal course of lupus nephritis has been recog- impairment was greater in the hyperten- nised,6-9 it is nevertheless difficult to draw firm sives, 47% vs 18.5% (p=0.04). Mean serum conclusions about its prognostic importance at creatinine was also higher higher in this this stage of the disease, as it may be both a group (p=0.02). The presence of hyper- cause and consequence of renal insufficiency.4 tensive renal vascular lesions identified a high-risk subgroup who had a higher Patients and methods http://pmj.bmj.com/ incidence of renal functional impairment and worse renal function than the hyper- Between 1988 and 1993, all patients who had tensive group as a whole. Even at an early fulfilled the 1982 American Rheumatism As- stage, hypertension and hypertensive re- sociation Revised Criteria for the classification nal vascular lesions correlated well with of systemic lupus erythematosus'0 and who had renal functional impairment. Aggressive undergone renal biopsies at King Edward VIII treatment of hypertension is therefore Hospital were retrospectively assessed. Patients on October 2, 2021 by guest. Protected copyright. essential in early lupus nephritis in order who had no evidence of lupus nephritis University of Natal, to prevent further deterioration of renal according to the World Health Oganisation Renal Unit, King function as the disease evolves. (WHO) classification" were Edward VIII excluded, as Hospital, Durban, were patients who were on corticosteroid South Africa Keywords: hypertension, lupus nephritis, glomerulo- therapy prior to the renal biopsy. None of the Department of nephritis, systemic lupus erythematosus remaining patients had any evidence of other Medicine identifiable secondary causes of hypertension. IP Naiker A total of 44 patients, (39 women) satisfied YK Seedat The development of arterial hypertension in these criteria and were selected for the study. IGH Randeree patients with systemic Department of lupus erythematosus is The mean age was 33 years (range 15-55); 29 Anatomical generally believed to be secondary to under- were Blacks and 15 Indians. Patients were Pathology lying lupus . 1,2 Indeed, Ro- V Chrystal senbaum et aP found that the incidence of hypertension in patients with lupus nephritis Correspondence to Causes of hypertension in systemic Dr Indiran P Naiker, increased with both histological activity and lupus erythematosus Department of Medicine, loss of renal function. However, Ward and Medical School, * University of Natal, PO Studenski,4 who followed up 160 patients with proliferative lupus nephritis Box 17039, Congella, lupus nephritis for a median of 6.4 years found * corticosteroid therapy 4013, Durban, South that hypertension was an important indepen- * advanced renal functional impairment Africa dent risk factor for the development of end Accepted 27 June 1996 stage renal disease and was not merely a Box 1 Arterial hypertension in lupus nephritis 231

regarded as being hypertensive if they had a Table 2 WHO histological classification of sitting blood pressure equal to or greater than hypertensives (HT) compared to normoten- 140/95 mmHg on at least two different occa- sives (NT). Percentages are given in parenth- sions. The recorded blood pressure was the eses mean diastolic pressure of the two determina- Postgrad Med J: first published as 10.1136/pgmj.73.858.230 on 1 April 1997. Downloaded from tions. None of the patients was receiving WHO class HT (n=17) NT (n=27) antihypertensive medication at the time of I - - assessment. Hypertension was classified as II 2(12) 7(26) being mild (diastolic 95-99 mmHg), moder- III 1(6) - ate (100- 114 mmHg) or severe IV 6(35) 4(15) (> 115 mmHg). Impaired renal function was V 7(41) 16(59) defined as a serum creatinine equal to or VI 1(6) - greater than 120 ,mol/l, and was graded as mild (120-200 iimol/l), moderate (200- 350 ,umol/l) or severe (>350 yimol/l). The mean serum creatinine was also higher in the histological classification was obtained from hypertensives compared to the normotensives: the pathologist's reports which utilise the 283.4 ,umol/l vs 100.2 /imol/l, respectively standard WHO classification of glomerular (p=0.02). Hypertensive renal vascular lesions involvement in systemic lupus erythemato- were identified in 1 1 patients, all ofwhom were sus.1' All histology was subsequently re-eval- hypertensive. Details of renal functional im- uated by a single senior pathologist who pairment and mean diastolic blood pressure in searched specifically for hypertensive renal these patients compared to those without vascular lesions. The latter were defined as hypertensive renal vascular lesions are de- the presence of one or more of the following scribed in table 1. The WHO histological abnormalities in the arterioles and small classification of the hypertensives compared arteries: hyperplasia of the media, fibro-elas- to the normotensives is depicted in table 2. tosis of the intima, and hyalinosis and sclerosis Although there appeared to be a greater of the vessel wall. prevalence of the milder histological classes in the normotensives (26% class II vs 12% and Statistical analysis 59% Class V vs 41%) and more patients with Means and standard deviations were calculated Class IV histology in the hypertensive group and compared using either the student's (35% vs 15%), these differences were not unpaired t-test for normal distributions, or statistically significant. However, five of the the Wilcoxon 2-sample test (the non-para- seven hypertensives and six of the 16 normo- metric equivalent), for skewed distributions. tensives with Class V histology had prolifera- The chi-square test was used to compare tive lesions (either Class Vc or Vd). categorical data. Spearman's correlation co- Proliferative lesions (excluding patients with efficient was utilised to determine a possible Class II histology) were therefore present in 12 correlation between the severity of hyperten- of the 17 hypertensives (70%) compared to sion and both renal function and histological only 10 of the 27 normotensives (37%), class. p=0.03. There was no relationship between http://pmj.bmj.com/ race and the prevalence of any of the histolo- Results gical classes. In the hypertensive group there was no correlation between the severity of Hypertension was diagnosed in 17 of the 44 hypertension and either renal function patients (38%) at the onset of clinical lupus (r=0.09) or histological class (r=0.08). nephritis. The majority (12 patients) had moderate hypertension. Mild and severe hy- Discussion on October 2, 2021 by guest. Protected copyright. pertension was present in three and two patients, respectively. The mean age of the Systemic lupus erythematosus was previously hypertensives (35.4 years) did not differ regarded as an uncommon disease in significantly from that of the 27 normotensive Africa. 12-14 Subsequently, however, perhaps patients (32.4 years). The race of the patients consequent upon an increased medical aware- did not influence the prevalence of hyperten- ness and wider use of screening tests,'5 it has sion. been recognised more frequently in Southern The incidence of impaired renal function Africa.'5-17 Furthermore, the disease has was significantly higher in the hypertensives (8/ serious consequences in South African Blacks, 17 patients, 47%) compared to the normoten- being associated with a high mortality.'8 Renal sives (5/27 patients, 18.5%) (p=0.04). The disease in our patients with systemic lupus erythematosus is common, with a reported prevalence of 65% and is a leading cause of Table 1 Renal function and blood pressure in patients with and mortality.'8 A recent study conducted at our without hypertensive renal vascular lesions institution reported a mortality of 35% in patients with lupus nephritis."5 Lesions Lesions This high mortality, largely due to the present (n=11) absent (n=33) development of end stage renal failure,'5 is of Mean diastolic blood pressure (mmHg) 104.5 84.2 (p=0.0001) particular concern to us, as the majority of our Mean serum creatinine (,umol/l) 392.7 97.0 (p=0.01) patients are socio-economically deprived Prevalence of renal functional impairment (%) 63.6 18.1 (p=0.Ol) Blacks in whom renal replacement therapy is not feasible because most of them reside either 232 Naiker, Chrystal, Randeree, Seedat

in a rural environment or in peri-urban Summary/learning points settlements lacking basic ablution and sanitary facilities. l9 * the prevalence of hypertension at the onset of The role of hypertension at the onset of clinical lupus nephritis was 38% clinical lupus nephritis as a risk factor for * the incidence of renal functional impairment Postgrad Med J: first published as 10.1136/pgmj.73.858.230 on 1 April 1997. Downloaded from progression to renal disease is and mean serum creatinine levels were higher end-stage con- in the hypertensives than the normotensives troversial.5 Ward and Studenski4 found that * patients with hypertensive renal vascular hypertension and cigarette smoking at the lesions represented a high-risk subgroup with a onset ofnephritis were important, independent higher incidence of renal functional risk factors for progression to end-stage renal impairment and worse renal function than the disease. Appel et al,8 however, reported that hypertensive group as a whole hypertension at the time of the initial * severe (proliferative) lupus histology was more renal prevalent in hypertensive patients biopsy did not significantly influence renal or * in addition to immunosuppressive therapy, patient survival. Similarly, Nossent et alP0 in a effective treatment of hypertension in patients prospective study, found no relation between with early lupus nephritis is essential as the presence of hypertension and severity of hypertension is an important, potentially disease course. They commented, however, modifiable, risk factor for progression to end that the early detection and effective treatment stage renal failure in these patients of arterial hypertension in their patients prob- ably eliminated its contribution to subsequent Box 2 renal damage. The prevalence of hypertension at the onset of clinical lupus nephritis in our patients was tients. Finally, in the hypertensive group the 38%, and was not influenced by age or race. absence of a correlation between the severity of This is identical to the prevalence reported by hypertension and both renal function and Ward and Studenski,' and is considerably histological class is of doubtful significance as higher than the prevalence of hypertension in there were only 17 patients in this group. our adult Black and Indian population, viz; As discussed previously lupus nephritis in 25% and 14.2%, respectively.2' None of our our patients with systemic lupus erythematosus patients were receiving corticosteroids at the is common'8 and is associated with significant time of analysis and the incidence of hyperten- mortality, with approximately 25% of patients sion would be expected to increase with the dying from end-stage renal failure.'5 As we are evolution of the disease and initiation of unable to support many of these patients on corticosteroids. Even at this early stage of the renal replacement therapy programmes it is illness, however, hypertension appeared to be important to be able to modify the disease at an functionally significant in that both the inci- early stage. In addition to the adverse prog- dence of renal functional impairment and nosis associated with the more severe histolo- mean serum creatinine levels were significantly gical forms of lupus nephritis (especially WHO higher in the hypertensives than in the normo- Class IV and high activity and chronicity tensives. Furthermore, the presence of hyper- indices),622-2'4 arterial hypertension is also an tensive renal vascular lesions identified a high- important, potentially modifiable risk factor for http://pmj.bmj.com/ risk subgroup of patients who had a higher progression to end-stage renal disease.4'5 We incidence of renal functional impairment and are unable to draw any conclusions about the worse renal function than the hypertensive prognostic significance of hypertension at the group as a whole (table 1). All our patients with onset of clinical lupus nephritis in our patients hypertensive renal vascular lesions were hyper- as this would require a longer period of tensive. We were unable to demonstrate a observation. Nevertheless we were able to correlation between hypertension and histolo- demonstrate an association between hyperten- on October 2, 2021 by guest. Protected copyright. gical class of lupus nephritis (table 2). sion and impaired renal function on presenta- Nevertheless, proliferative lesions (including tion. Furthermore, the observation that the patients with WHO Class Vc and Vd histology presence of hypertensive renal vascular lesions and excluding those with Class II histology) identified a subgroup of patients with signifi- were significantly more prevalent in the hyper- cantly worse renal function suggests that tensives compared to the normotensives, sug- hypertension is functionally significant at an gesting that aggressive lupus histology may play early stage of lupus nephritis. We therefore a role in the pathogenesis of hypertension in believe that, in addition to appropriate im- systemic lupus erythematosus. In this context, munosuppressive therapy, early and effective Appel et a18 reported that their patients with treatment of arterial hypertension in patients WHO Class IIa and IIb histology had a more with early lupus nephritis is essential to prevent favourable renal and patient survival at five and deterioration of renal function. 10 years than did patients in the other WHO classes (III, IV and V). They noted that the long-term prognosis for Class V patients was less favourable than originally suggested. From the point of view of hypertension, however, the presence of two hypertensive patients with Class II The authors acknowledge Ms Eleanor Gouws of the histology and four normnotensives with Biostatistics Department of the Medical Research Class IV histology (table 2) suggests that factors Council for her assistance with the statistical other than lupus histology must be involved in analysis, and Mrs D Mackrory for her secretarial the pathogensis of hypertension in these pa- assistance. Arterial hypertension in lupus nephritis 233

1 Budman DR, Steinberg AD. Hypertension and renal 14 Jessop S, Meyers OL. Systemic lupus erythematosus in disease in systemic lupus erythematosus. Arch Intern Med Cape Town. S Afr Med J 1973; 47: 222 - 5. 1976; 136: 1003-7. 15 Seedat YK, Parag KB, Ramsaroop R. Systemic lupus 2 Walker WG, Solez K. Renal involvement in disorders of erythematosus and renal involvement. A South African connective tissue. In: Strauss, Welt, eds, Disease of the experience. 1994; 66: 426-30. , 3rd edn. Boston: Little, Brown and Co, 1979; pp 16 Taylor HG, Stein CM. Systemic lupus erythematosus in 1259-88. Zimbabwe: Ann Rheum Dis 1986; 45: 645-8. Postgrad Med J: first published as 10.1136/pgmj.73.858.230 on 1 April 1997. Downloaded from 3 Rosenbaum JL, Akhtar M, Kramer MS. Evaluation of 17 Dessein PHMC, Gledhill RF, Rossouw DS. Systemic lupus clearance studies in lupus nephritis. Clin Nephrol 1974; 2: erythematosus in Black South Africans. S Afr Med J 1988; 47-51. 74: 387-9. 4 Ward MM, Studenski S. Clinical prognostic factors in lupus 18 Moody GM, Parag KB, Nathoo BC, Pudifin DJ, Duursma nephritis. The importance of hypertension and smoking. J, Seedat YK. High mortality with systemic lupus erythe- Arch Intern Med 1992; 152: 2082- 8. matosus in hospitalized African Blacks. Br J Rheumatol 5 Petrin J, Rozman B, Dolenc P, et al. The dissociation of 1994; 33: 1151-3. arterial hypertension and lupus glomerulonephritis in 19 Naiker IP, Naicker S, Botden L, Haffejee AA, Seedat YK. systemic lupus erythematosus. Blood Pressure 1993; 2: Continuous ambulatory peritoneal dialysis - A South 108- 12. African experience. Int _Artif Organs 1993; 16: 806-8. 6 Baldwin DS, Lowenstein J, Rothfield NF, Gallo G 20 Nossent JC, Bronsveld W, Swaak AJG. Systemic lupus McCluskey RF. The clinical course of the proliferative eythematosus. III. Observations on clinical renal involve- and membranous forms of lupus nephritis. Ann Intern Med ment and follow up of renal function: Dutch experience 1970; 7: 929-42. with 110 patients studied prospectively. Ann Rheum Dis 7 Baldwin DS, Gluck MC, Lowenstein J, Gallo GR. Lupus 1989; 48: 810-6. nephritis: clinical course as related to morphologic forms 21 Seedat YK. Race, environment and blood pressure: the and their transitions. Am JMed 1977; 62: 12-30. South African experience. J Hypertens 1993; 1: 7- 12. 8 Appel GB, Cohen DJ, Pirani CL, Meltzer J, Estes D. Long- 22 Nossent HC, Henzen-Logmans SC, Vroom JM, Berden term follow-up of patients with lupus nephritis: a study JHM, Swaak TJG. Contribution of renal biopsy data in based on the classification of the World Health Organiza- predicting outcome in lupus nephritis: analysis of 116 tion. Am 7 Med 1987; 83: 877 - 85. patients. Arthritis Rheum 1990; 33: 970 - 7. 9 Leaker B, Fairley KF, Dowling J, Kincaid-Smith P. Lupus 23 Austin HA, Muenz IR, Joyce KM, Antonovych H, Balow nephritis: clinical and pathological correlation. Q J Med JF. Diffuse proliferative lupus nephritis: identification of 1987; 62: 163 - 79. specific pathological features affecting renal outcome Kidney 10 Tan EM, Cohen AS, Fries JF, et al. The 1982 revised Int 1984; 25: 689-95. criteria for the classification of systemic lupus erythemato- 24 Schwartz MM, Bernstein J, Hill GS, Holley K, Phillips FA sus. Arthritis Rheum 1982; 25: 1271 - 7. and the Lupus Nephritis Collaborative Study Group. 11 Churg J, Sobin LH. Renal disease classification and atlas of Predictive value of renal pathology in diffuse proliferative glomerudar disease. Tokyo: Igaku-Shoin, 1982; pp 127 - 49. lupus glomerulonephritis. Kidney Int 1989; 36: 891-6. 12 Shaper AG. Systemic lupus erythematosus. E Afr Med J 1961; 31: 134-44. 13 Greenwood BM. and parasitic infec- tions. Lancet 1968; ii: 380-2. http://pmj.bmj.com/ on October 2, 2021 by guest. Protected copyright.