ONCOTIC PRESSURE AND EDEMA FORMATION IN HYPOALBUMINEMIC HIV-INFECTED PATIENTS WITH PROTEINURIA

Human immunodeficiency virus nephropathy (HIVN) continues to challenge nephrologic consultative services at major urban institutions. Althou gh noted in the literature, the decreased incidence of peripheral edem a in HIVN has been unexplained to date. In HIV patients, total protein s frequently are found to be elevated due to an elevated globulin frac tion, The impact that plasma proteins, specifically globulins, have on the total oncotic pressure has not been reported in HIVN, but may pla y a role in the paucity of edema noted in this proteinuric population, To evaluate the contributions of serum globulin to the total oncotic pressure and the presence or absence of edema in HIVN, we randomly sel ected 27 patients with proteinuria greater than 2.5 g/24 hr and serum albumin less than 3.1 g/dL from patients presenting to the nephrology outpatient clinic at the University of Miami/Jackson Memorial Hospital . Seventeen of the patients (63%) had a known diagnosis of HIV infecti on (group 1). These patients were subdivided into two subgroups: those presenting with clinically evident edema on physical examination (n = 7 [41%]; group 1A) and those who had an absence of edema (n = 10 [59% ]; group 1B). Conversely, group 2 comprised 10 patients without known HIV infection, of whom six (60%) had edema (group 2A) and four (40%) d id not (group 2B). Blood pressures were noted, and mean arterial press ure was calculated using standard formulas. Serum albumin, serum total proteins, and urine total proteins were measured using standard labor atory methods, Oncotic pressures for albumin (alpha), globulin (beta), and total protein (c) were calculated using the following formula: CO PpI = alpha(2.8c + 0.18c(2) + 0.012c(3)) + beta(0.9c + 0.12c(2) + 0.00 4c(3)). We used Student's t-test to analyze the data. There is no sign ificant difference between the albumin concentrations of HIV patients without edema (group 1B) and non-HIV patients with edema (group 2A), w ith mean concentrations of 2.3 +/- 0.1 g/dL versus 2.3 +/- 0.15 g/dL, respectively (P = NS). Group 1B, however, has a total oncotic pressure of 17.1 +/- 1.5 mm Hg, whereas both groups with edema (groups 1A and 2A) have statistically significant lower total oncotic pressures (12.1 +/- 2.3 mm Hg and 12.9 +/- 1.1 mm Hg, respectively; P < 0.05). The gl obulin oncotic pressures may account for some of the differences in to tal oncotic pressures, being significantly higher for those patients w ithout edema in group 1B compared with group 2A (7.1 +/- 0.9 mm Hg v 3 .9 +/- 0.4 mm Hg, respectively; P < 0.05). In patients with HIV, howev er, the presence or absence of edema is mandated by albumin concentrat ion because both groups have similar globulin concentrations (group 1A 3.1 +/- 0.1 g/dL v group 1B 3.8 +/- 0.3 g/dL; P = NS). Mean arterial pressure does not play a role in edema formation in this study because the HIV patients without edema had the higher blood pressures (group 1B 97.8 +/- 4.7 mm Hg v group 2A 84.7 +/- 5.5 mm Hg; P < 0.05), We con clude that globulins play an important role in maintaining oncotic pre ssure in low albumin states. HIVN patients with increased serum immune globulin may benefit from higher globulin oncotic pressure, delaying the onset of clinical edema in the setting of proteinuria. (C) 1997 by the National Kidney Foundation, Inc.