NEPHROLOGY 2008; 13, 4550

doi:10.1111/j.1440-1797.2007.00890.x

Review Article

Adult nephrotic syndrome: Non-specic strategies for treatment

JOHN A CHARLESWORTH, DAVID M GRACEY and BRUCE A PUSSELL Department of Nephrology, Prince of Wales Hospital and The University of New South Wales, Sydney, New South Wales, Australia
SUMMARY: Irrespective of aetiology, the nephrotic syndrome presents a range of potentially serious complications. These include thrombo-embolism, infection and hyperlipidaemia. Despite the prevalence of the nephrotic state among renal patients, there has been little prospective analysis of the therapeutic approach to these potentially life-threatening events even though their pathogenesis has been examined in some detail. Most of these complications are more prevalent once the albumin concentration falls below 20 g/L and it is recognized that restoration of serum albumin signicantly diminishes their frequency. However, this may be difficult to achieve, especially in adults. The problems of thrombo-embolism and infection are of immediate concern but, in persistent cases, the additional issues of hyperlipidaemia and loss of bone density also require consideration for therapy. Thus, in addition to specic attempts to reduce proteinuria, it is recommended that high-risk nephrotic patients receive anticoagulation, pneumococcal vaccination and lipid lowering therapy. Strategies for the preservation of bone density should also be considered, particularly in patients who receive high-dose corticosteroids. Among a range of non-specic treatments for proteinuria, angiotensin-converting enzyme inhibitors appear best in terms of efficacy and safety. Prospective trials are required to clarify the longitudinal impact of these generic strategies on the protection of the persistently nephrotic patient. KEY WORDS: bone mineral density, hyperlipidaemia, infection, nephrotic syndrome, proteinuria, thrombo-embolism.

A signicant percentage of patients with the nephrotic syndrome will remain heavily proteinuric despite vigorous attempts to modify its course with specic, therapeutic regimens. In the event that proteinuria remains sufciently heavy to maintain the nephrotic state, with its inherent risks, then the nephrologist is required to consider additional strategies to protect the patient from escalating harm. Moreover, the prolonged use of potentially dangerous agents, such as steroids and cytotoxic agents, in the hope of achieving a satisfactory remission, provides further clinical risks for the patient. This review will not address the specic therapeutic regimens currently in use for the treatment of patients with nephrotic syndrome. Rather, it will examine those policies that are applicable to all adult nephrotic patients, irrespective of aetiology. There is little prospective

data on these protective strategies even though there have been major advances in the safety and efcacy of therapeutic agents. THE EVOLUTION OF THE NEPHROTIC STATE With the exception of minimal change nephropathy, the majority of patients with nephrotic syndrome face the prospect of high-level chronic protein loss. In adults for example, two of the commonest causes, diabetic nephropathy and membranous nephropathy, pose major challenges in terms of therapy. This continuing state of heavy proteinuria presents several risks to the patient. First, protein depletion per se inuences a wide range of metabolic and immunological functions.1 The resultant hypoalbuminaemia does not reect urinary protein loss alone; the altered metabolism of albumin, both locally and systemically has long been recognized (Table 1).2,3 Moreover, the quantity of urinary albumin does not correlate precisely with the serum albumin concentration. A signicant number of patients will maintain a normal serum albumin despite losses in excess of the usual nephrotic level.4 It is recognized that this latter group of patients will be protected, to a substantial extent, from

Correspondence: Professor John A Charlesworth, Department of Nephrology, Prince of Wales Hospital, Barker Street, Randwick, NSW 2031, Australia. Email: j.charlesworth@unsw.edu.au Accepted for publication 27 August 2007. 2008 The Authors Journal compilation 2008 Asian Pacic Society of Nephrology

46

JA Charlesworth et al.

Table 1 Factors contributing to hypoalbuminaemia 1. Loss of urinary protein 2. Albumin hypercatabolism 3. Restricted upregulation of albumin synthesis 4. Compartmental redistribution of albumin

Table 2 Abnormalities of coagulation in nephrotic patients 1. Factors IX and XI 2. Factors V and VIII 3. Anti-thrombin III 4. Fibrinolysis 5. Fibrinogen; brinogenbrin 6. Platelet aggregation, in vitro

the inherent risks of the nephrotic state. The physician must be mindful that even minimal change nephrotic patients, with their typically prompt response to corticosteroid therapy, still retain a high susceptibility to relapse; it is generally estimated that more than 60% of both adults and children with this condition will experience at least one relapse within 12 months of diagnosis.5 Hence, it is appropriate to anticipate the long-term need to protect nephrotic patients from the multiple medical risks associated with these chronic protein-depleting disorders. THROMBO-EMBOLISM Although there is considerable variation in the reported incidence of thrombo-embolic complications of nephrotic syndrome, it is recognized that this problem affects approximately 35% of patients.6 It is generally viewed that venous complications are more prevalent although coronary and peripheral arterial events also occur. Venous sequelae include deep vein thrombosis, renal vein thrombosis and pulmonary thrombo-embolism. These appear particularly prevalent in patients with membranous nephropathy,7,8 although the reason for this predilection remains unclear. The fact that such patients are older and may have preexisting arterial and venous disease could, at least in part, explain this high incidence. The development of a procoagulant state in patients with nephrotic syndrome is often thought to result from the loss of antithrombotic factors in the urine. This grossly over simplies the nature of the coagulation disturbance that occurs in these patients. While factors of small molecular weight may be lost in the urine, there is an increase in plasma pro-coagulant cofactors, factors V and VIII, and often grossly elevated brinogen levels (Table 2). Platelet reactivity is also increased although these changes are most demonstrable in vitro and appear ow dependent.6 These changes become more prevalent once the serum albumin level falls below 25 g/L. However, it must be emphasized that the association between coagulation abnormalities and thrombo-embolic events is inexact. Hence, it is still unknown which parameters best guide the physician as to which patients might require prophylactic anticoagulation. There is no dispute about this need once such complications become evident. However, many of these thrombotic events, particularly renal vein thrombosis but also pulmonary embolism, remain clinically silent. Recommendations: Anticoagulation should be introduced in high-risk patients (see Table 3) There are no long-term clinical trials to assist with decision making as to whether prophylactic anticoagulation should

Table 3 Thrombo-embolism in nephrotic patients: high-risk groups 1. Serum albumin <20 g/L 2. Clinical hypovolaemia 3. Bed rest/intercurrent illness 4. Membranous nephropathy

be used in patients with nephrotic syndrome. However, a Markov-based decision analysis model in patients with idiopathic membranous nephropathy concluded that the benets of prophylactic treatment outweighed the risks associated with bleeding.9 Certain readily accessible features may assist the physician with this decision. Patients with membranous nephropathy appear to have the highest risk of thrombo-embolism. The level of serum albumin (particularly, if below 20 g/L) and associated high-level proteinuria (i.e. >10 g/24 h) amplify the risk. Also it would seem appropriate to introduce anticoagulant therapy, such as subcutaneous heparin, during periods of intercurrent illness and enforced bed rest. While the use of antiplatelet agents may appear more convenient and perhaps less hazardous to the patient it is uncertain whether they play a signicant role in the prevention of nephrosis-induced thrombo-embolism.6 Treatment options for prophylaxis Subcutaneous heparin during enforced bed rest Warfarin for persistent serum albumin <20 g/L RISK OF INFECTION Serious bacterial infection has long been recognized as a major, potentially life-threatening complication of nephrotic syndrome. Prior to the advent of antibiotics and the use of corticosteroids, sepsis was responsible for the death of approximately one-third of patients, including children. Pneumococcal sepsis, in particular, seems prevalent in children and these patients have demonstrably lower titres of pneumococcal antibodies compared with healthy subjects of a comparable age.10 In adults, the problem of sepsis remains clinically important, although less frequent, and a wider range of bacterial complications (including Escherichia coli) have been reported. Despite modern therapies, overwhelming sepsis is still potentially life-threatening in severely nephrotic subjects. This particularly holds during periods of medical intervention, such as hospitalization, when
2008 The Authors Journal compilation 2008 Asian Pacic Society of Nephrology

Treatment of adult nephrotic syndrome

47

intravenous access and exposure to more resistant organisms is more likely to occur. Breaches in the skin of the grossly oedematous patient are common and must be treated with a high degree of suspicion. Numerous studies document abnormalities of immunity in patients with nephrosis. IgG concentrations, for example, are signicantly reduced during exacerbation and may persist following remission.11 By contrast, IgM levels may be elevated. Nephrotic patients may also show disturbances of complement, particularly factor B of the alternative pathway, which provides protection against encapsulated organisms such as pneumacoccus.12 Some patients with low IgG levels will also have reduction in C1q. These changes in factor B and C1q predominantly reect variations in metabolism rather than urinary protein loss. Disturbances of neutrophil phagocytosis and T lymphocyte function in vitro have been documented during exacerbations of the nephrotic syndrome. The introduction of steroids, with or without cytotoxic agents, increases the risk of infection. While the intention of such therapy is to achieve remission, its duration may be sufcient to induce an even higher susceptibility to a wider range of infections. Hence, the actively treated nephrotic patient requires a level of surveillance comparable with that where specic treatment is not used (e.g. diabetic nephropathy). Recommendations: vigilance; pneumococcal vaccine A high level of suspicion of underlying infection must be maintained in all patients who remain nephrotic. While pneumococcal sepsis has become less frequent, the question arises as to whether patients should be vaccinated if heavy proteinuria and hypoalbuminaemia persist. Hesitation might exist because of concerns as to whether such patients are responsive to such vaccines, particularly in the presence of immunosuppressive therapy. Although there is evidence, as stated above, that children with minimal change nephropathy have lower than normal titres of pneumococcal antibodies, these data suggest that their response to vaccination is essentially normal. However, despite this satisfactory rise in titre, there is a rapid decline in antibody levels within 6 months such that <50% of patients maintain adequate immunization by 1 year.10 There is no evidence to support the use of prophylactic antibiotics in nephrotic patients. HYPERLIPIDAEMIA Hyperlipidaemia is a signicant vascular risk factor in the general population and in many nephrotic patients the lipid prole will be comparable to that observed in subjects with primary vascular disease. Hence, without effective reduction in proteinuria, some nephrotic patients may face similar long-term risks. However, this is difcult to prove in the presence of other vascular risk factors, such as hypertension. Nevertheless, there is general recognition that nephrotic patients face a several-fold increase in the risk of a signicant coronary event (see Table 4). (This risk does not apply
2008 The Authors Journal compilation 2008 Asian Pacic Society of Nephrology

Table 4 Risk factors for vascular disease in nephrotic syndrome 1. Hyperlipidaemia: LDL; VLDL; Lp(a) or normal trigs 2. Hypertension 3. Uraemia 4. Hypercoagulable state
A role for hypercoagulability in pathogenesis is suspected but not conrmed. LDL, low-density lipoproteins; Lp(a), lipoprotein(a); VLDL, very-low-density lipoproteins.

to patients with minimal change nephropathy.) The basis for the lipid abnormalities in such patients remains somewhat unresolved although it would appear that both synthetic and catabolic mechanisms are defective (see Wheeler and Bernard13 for review). Lipolytic factors have been identied in the urine of nephrotic animals14 and such factors have been suspected but not proven to be lost in the urine of nephrotic patients. It is recognized that low plasma albumin levels play a role in promoting hyperlipidaemia and IV infusion of albumin induces short-term reduction in lipid levels.15 Also, restoration of plasma volume with nonproteinaceous products similarly achieves an improved lipid prole.13 Prior to the development of HMG-CoA reductase inhibitors, the treatment of hyperlipidaemia in nephrotic patients was poorly tolerated. However, current therapy offers a safe opportunity to improve lipid levels and this would seem appropriate in cases where the nephrotic state is sustained. Certain diagnostic categories, such as diabetic nephropathy, would be expected to not only resist attempts at reduction in proteinuria but also be associated with a high risk of vascular damage. The safety and efcacy of statin therapy for hypercholesterolaemia in patients with heavy proteinuria has been demonstrated.16 However, the longterm effect of such treatment on the progress of renal disease is less certain. Other methods of lipid lowering have also been considered in patients with the nephrotic syndrome. These include the use of soy protein diets,17 sh oils and the use of angiotensin-converting enzyme inhibitors (ACE). Similarly, low dose aspirin has an established role in the long-term prevention of vascular disease in the general population. However, there are no substantial data to conrm its benet in the persistently nephrotic patient. Recommendations: Use a HMG-CoA reductase inhibitor in high-risk patients Unlike the problems of thrombo-embolism and infection, the need to treat lipid abnormalities in nephrotic patients would appear less urgent. Nevertheless, in patients with continuing heavy proteinuria, particularly in the presence of additional risk factors, it would seem logical to introduce specic therapy. The use of statins is a proven safe and effective approach for such patients. The other modes of therapy mentioned above may also be of benet but, given the potential severity of the lipid abnormalities, would be unlikely to be of sufcient efcacy when used alone.

48

JA Charlesworth et al.

PROTEIN INTAKE In the presence of albumin hypercatabolism and restricted synthesis, it would appear logical to increase the dietary protein of patients with nephrotic syndrome. In the carefully supervised hospital situation, this has been shown to improve nitrogen balance. However, in practice, patients are frequently anorexic, nauseous and prematurely satiable. These symptoms are presumed to be a consequence of gastrointestinal oedema and the presence of ascites. By contrast, dietary protein restriction may have a favourable effect on proteinuria and dyslipidaemia but, potentially at the expense of adequate nutrition. Hence, many physicians elect to maintain a relatively normal protein intake which may well strike the best compromise between nutritional requirements and palatability. In serious circumstances, such as clinical hypovolaemia, with or without deterioration in kidney function, the intravenous administration of salt-free albumin has potential benets. However, the value of such strategies are understandably short lived: IV supplementation of albumin may untowardly affect albumin synthesis and, in most cases, a high percentage is rapidly lost in the urine. The expense of IV infusions is well documented. Recommendations: Normal dietary protein intake There would appear to be no convincing evidence for modifying the protein intake of nephrotic patients. Both increases and decreases may prove to be clinically problematic. The maintenance of appetite and overall nutrition may be difcult in the early stages of treatment and modication of protein intake may worsen these difculties. The use of IV albumin preparations would seem unjustied on both medical and nancial grounds except in the presence of serious haemodynamic complications (including acute renal failure) and/or the presence of severe resistant oedema. In these latter cases, short-term use of IV albumin may prove to be a useful supplement to sequential diuretic administration and sodium restriction. REDUCING PROTEINURIA: NON-SPECIFIC STRATEGIES Safe methods of reducing proteinuria are fundamental to the effective treatment of the nephrotic syndrome. Given the limited success of immunological regimens, which still lack selectivity in the majority of cases, it has been necessary to seek generic strategies, particularly in patients with resistant forms of the disorder. Three approaches have been used: dietary protein restriction (as discussed), non-steroidal antiinammatory drugs and ACE inhibition. It is generally thought that the lowering of urinary protein with nonsteroidal agents exceeds what might be expected from changes in glomerular ltration alone; nevertheless, the potentially unfavourable effect on glomerular blood ow and the ever-present possibility of serious non-renal side-effects have restricted the wide-spread use of these

compounds. By contrast, many physicians consider the introduction of ACE inhibitors as an integral part of their treatment of nephrotic patients. Their benecial effect on proteinuria and, hence on plasma lipids, appears additional to their inuence on glomerular blood ow and lags behind their blood pressure-lowering effect.18,19 Many physicians will introduce these agents irrespective of blood pressure status. However, in the clinical context of hypovolaemia, which is most prevalent in patients with severe hypoalbuminaemia and high level proteinuria (i.e. where ACE inhibition may appear especially applicable), they need to be introduced with caution because of the potential to reduce glomerular ltration. Recommendations: Use ACE inhibitor or AII-receptor antagonist Until more effective specic therapies are developed for the various forms of nephrotic syndrome apart from minimal change nephropathy, it will be necessary to consider nonspecic strategies for reducing proteinuria. ACE inhibitors would appear the most appropriate agents for this purpose. It would be anticipated that they have comparable benets to other strategies and are superior in regard to long-term maintenance of kidney function. In using these agents, the clinician will need to adjust dosage according to the anticipated changes in blood pressure, proteinuria and renal function. It would appear appropriate to introduce ACE inhibition in low dosage. CONTROL OF OEDEMA The nephrotic syndrome is an important cause of secondary hyperaldosteronism and urinary sodium excretion may, initially at least, be very low. Hence, dietary sodium restriction is widely practised in such patients and would appear an appropriate, safe starting point in the management of oedema. Again, as with perturbations in dietary protein, extreme reduction in salt can lead to reduced palatability and worsening of anorexia. Also, clinical assessment of the patient will leave some degree of uncertainty as to the actual sodium and uid status of the individual patient. Diuretics are best introduced in a stepwise fashion to minimize the risk of hypovolaemia. Nevertheless, several factors make nephrotic patients relatively resistant to such treatment. In the oridly oedematous patient, intestinal oedema may hinder absorption while impaired glomerular ltration will reduce transfer to the tubular lumen. High levels of urinary binding proteins may further restrict diuretic efcacy. Generally, patients with gross oedema are best treated in hospital until progress is made with this problem. A loss of 12 kg in weight per day is generally considered appropriate. Many patients will require sequential nephron blockade to achieve this purpose. The combination of a loop diuretic with an aldosterone antagonist is widely used. The addition of a third agent, such as a thiazide, may be considered in difcult cases. Physical measures including control of uid intake, support stockings and
2008 The Authors Journal compilation 2008 Asian Pacic Society of Nephrology

Treatment of adult nephrotic syndrome

49

supervised bed rest should be considered as part of overall care of the nephrotic patient. Recommendations Restriction of sodium intake to 5070 mmol/day provides a reasonable contribution to oedema control in most adult patients. Diuretics should be introduced with caution in order to achieve a weight loss of 12 kg/day in the rst instance. It should be anticipated that many patients will be resistant to such agents despite a relatively normal GFR. However, the variability of response among individuals makes it appropriate to introduce diuretics in a stepwise fashion in order to determine appropriate dosage. BONE PRESERVATION IN NEPHROTIC PATIENTS There is little mention of the impact of the nephrotic state on bone integrity in older literature. This reects both the inability to assess bone mineral density (BMD) and to offer effective therapy. However, the evaluation of BMD and treatment for osteoporosis now make it appropriate to consider bone status in patients with either ongoing or relapsing nephrotic syndrome. Moreover, it is recognized that changes in bone density may occur within weeks or months of introduction of high doses of corticosteroids. The administration of a catabolic steroid to patients with heavy proteinuria and demonstrable albumin hypercatabolism represents a combination of factors likely to be severely detrimental to bone density. Several publications document the relatively rapid loss of bone mineral content following the introduction of high-dose daily steroids for nephrotic syndrome. Gluati et al. prospectively studied 100 consecutive children with idiopathic nephrotic syndrome who were initially treated with prednisone, 60 mg/m2 per day for 6 weeks followed by 40 mg on alternate days for a further 6 weeks.20 Relapses were treated in a similar manner. Comparisons with children where persistence of nephrosis was less problematic showed a signicantly lower BMD in the relapsing/ persistent group. Similar, although smaller, studies have been undertaken in adults. All groups have concluded that steroid treatment of nephrotic syndrome provides a signicant risk for early development of osteoporosis which warrants both surveillance by modern techniques and, if necessary, introduction of specic treatment Recommendations In the event that the nephrotic state requires steroid therapy for three or more months, then serious consideration should be given to formal assessment of BMD and, in the event of signicant bone loss, specic therapy for osteoporosis (see Editorial).21 CONCLUSION The nephrotic syndrome produces a wide range of clinical risks, irrespective of cause. In many cases, these may persist
2008 The Authors Journal compilation 2008 Asian Pacic Society of Nephrology

for months or even years. Hence, a wide range of protective strategies need to be implemented to protect patients from potentially serious complications while specic attempts are made to reduce or abolish proteinuria. Thrombo-embolism represents the highest life-threatening risk to such patients and, despite difculties in implementation and monitoring, anticoagulation must be considered in the persistently nephrotic patient. With safer therapies for dyslipidaemia, osteopenia and reduction in proteinuria, it would also seem appropriate to include these agents in the therapeutic regimen.

REFERENCES
1. Bernard DB. Extrarenal complications of the nephrotic syndrome. Kidney Int. 1988; 33: 1184202. 2. Bernard DB. Metabolic abnormalities in nephrotic syndrome: pathophysiology and complications. In: Brenner GM, Stein JH (eds). Nephrotic Syndrome. New York: Churchill Livingstone, 1982; 85120. 3. Kaysen G, Gambertoglio J, Felts J, Hutchison F. Albumin synthesis, albuminuria and hyperlipidemia in nephrotic syndrome. Kidney Int. 1987; 31: 136876. 4. Praga M, Borstein B, Andres A et al. Nephrotic proteinuria without hypoalbuminemia: Clinical characteristics and response to angiotensin-converting enzyme inhibitor. Am. J. Kidney Dis. 1991; 17: 3308. 5. Ponticelli C, Passerini P. Treatment of the nephrotic syndrome associated with primary glomerulonephritis. Kidney Int. 1994; 46: 595604. 6. Rabelink TJ, Zwaginga JJ, Koomans HA, Sixma JJ. Thrombosis and hemostasis in renal disease. Kidney Int. 1994; 46: 28796. 7. Robert A, Olmer M, Sampol J, Gugliotta JE, Casanova P. Clinical correlation between hypercoagulability and thrombo-embolic phenomena. Kidney Int. 1987; 31: 8305. 8. Bellomo R., Wood C, Wagner I et al. Idiopathic membranous nephropathy in an Australian population: The incidence of thrombo-embolism and its impact on the natural history. Nephron 1993; 63: 2401. 9. Sarasin FP, Schifferli JA. Prophylactic oral anticoagulation in nephrotic patients with idiopathic membranous nephropathy. Kidney Int. 1994; 45: 57885. 10. Fuchshuber A, Khnemund O, Keuth B, Ltticken R, Michalk D, Querfeld U. Pneumococcal vaccine in children and young adults with chronic renal disease. Nephrol. Dial. Transplant. 1996; 11: 46873. 11. Giangiacomo J, Cleary TG, Cole BR, Hoffsten P, Robson AM. Serum immunoglobulins in the nephrotic syndrome. N Engl. J. Med. 1975; 293: 812. 12. McLean RH, Forsgren A, Bjrkstn B, Kim Y, Quie PG, Michael AF. Decreased serum factor B concentration associated with decreased opsonization of Escherichia coli in the idiopathic nephrotic syndrome. Pediatr. Res. 1977; 11: 910. 13. Wheeler DC, Bernard DB. Lipid abnormalities in nephrotic syndrome. Am. J. Kidney Dis. 1994; 23: 33146. 14. Staphrans J, Anderson CD, LurZ. FW, Felts JM. Separation of a lipoprotein lilpase cofactor from the alpha I-acid glycoprotein fraction from urine of nephrotic patients. Biochim. Biophys. Acta 1980; 617: 51423. 15. Kaysen GA, Don B, Schambelan M. Proteinuria, albumin synthesis and hyperlipidemia in the nephrotic syndrome. Nephrol. Dial. Transplant. 1991; 6: 1419.

50

JA Charlesworth et al.

16. Thomas ME, Harris KPG, Ramaswamy C et al. Simvastatin therapy for hypercholesterolemic patients with nephrotic syndrome or signicant proteinuria. Kidney Int. 1993; 44: 11249. 17. DAmico G, Gentile MG, Mann G et al. Effect of vegetarian soy diet on hyperlipidaemia in nephrotic syndrome. Lancet 1992; 339: 11314. 18. Gansevoort RT, de Zeeuw D, de Jong PE. Dissociation between the course of the hemodynamic antiproteinuric effects of angiotensin I converting enzyme inhibition. Kidney Int. 1993; 44: 57984.

19. Gansevoort RT, Sluiter WI, Hemmelder MH, de Zeeuw D, de Jong PE. Antiproteinuric effect of blood-pressure-lowering agents: A meta-analysis of comparative trials. Nephrol. Dial. Transplant. 1995; 10: 196374. 20. Gulati S, Godbole M, Singh U, Gulati K, Srivastava A. Are children with idiopathic nephrotic syndrome at risk for metabolic bone disease? Am. J. Kidney Dis. 2003; 41: 11639. 21. Elder G. Nephrotic Syndrome dont forget the bones! Nephrology 2008; 13: 389.

2008 The Authors Journal compilation 2008 Asian Pacic Society of Nephrology