Should we treat all patients of glomerulonephritis?MK Mani
Apollo Hospital, Madras.
Keywords: Clinical Trials, Decision Making, Glomerulonephritis, drug therapy,Human, Steroids, adverse effects,therapeutic use,
The answer to this question is simple. I do not know. A review of the literature is confusing. Let me start with the Oxford Textbook of Nephrology , which has an extensive discussion of the subject. The figures in [Table - 1] are reproduced from the Oxford Textbook, but 1 did the statistical analysis.
From this it seems clear that membranoproliferative glomerulonephritis and membranous nephropathy both do better with treatment, and in membranous nephropathy the addition of cytotoxic agents yields better results than steroids alone. It is odd that Cameron, the editor of this book, in another article. Sting, Chronic Rejection of Transplant Kidney says he would use steroids as the first choice for membranous nephropathy. That suggests that things are not really so cleat. The complications of treatment have not been taken into account.
The International Study of Kidney Disease in Children (ISKDC) reported on the treatment of membranoproliferative glomerulonephritis . In this paper, 50% deicrement in the glomerular filtration rate (GFR) was taken as the end point. Patients were given 40 mg prednisolone per sq m body surface area on alternate days. The treated group had a better result, 95% surviving (ie. remaining above 50% of their original GFR) against 57% in the placebo group. However, this was not statistically significant. Further, many patients had severe toxic reactions necessitating withdrawal of steroids and when this was done, they lost renal function, so that ultimately the survival was 55% in the treatment group as against 53% in the placebo. The inference is that, though steroids may slow deterioration, the benefit is negated by toxicity. I want to stress that this is in the West and we would see worse side effects of steroids here.
There are other ways of looking at the problem. Donadio and Offord reviewed five randomised clinical trials, which they selected as being the most significant. No treatment proved effective. A point they made which we usually do not consider is the fact that the patient has had the disease for some time before he comes to the nephrologist for treatment. If a patient has had the disease for four years before he comes to the unit and gets included in a trial that is already four years survival without therapy. However, in the analysis, survival, whether as a member of the treatment group or as a control, is taken from the day the trial starts. They found a high drop out rate due to drug toxicity, and called for better planned studies concentrating on the prevention of renal failure, and considering side effects, quality of life, cost, and compliance, ail factors of the utmost importance in our country.
My study of the complications of steroid treatment of nephrotics shows, with our far less satisfactory hygiene, that 31% of patients treated had serious complications in just 8 weeks of steroid therapy. I have not analysed the complications of cytotoxic agents, but it is clear that there would be still more infection when they are used.
We need to fall back on the old aphorism: first, do no harm. Let me turn now to IgA nephropathy. In a review article, John found the consensus of opinion was that persistent microscopic haematuria, proteinuria, hypertension, impaired renal function, and the presence of glomerular hyalinosis and sclerosis, or many crescents on biopsy carried a bad prognosis. We would all agree with that. Ten year cumulative renal survival was 83.3%, and good results with treatment were found only when renal histology showed minimal change disease with lgA deposits. Many of the studies showed no significant difference between controls and the treatment group at 6 months, 1 year and 5 years.
Nephrology Dialysis Transplantation devoted an entire supplement to the subject of membranous nephropathy. Figures of 21 to 24% of patients having underlying causes like drugs or malignancy are mentioned. My own experience is nowhere near this figure, with almost all being idiopathic. Treatment is recommended only if protein excretion is more than 3.5 g in 24 hours. I am not sure that would be valid for our conditions. If the serum albumin is very low, and protein intake is also poor, protein excretion in the urine may be much less than that. 1 would therefore take the clinical diagnosis of the nephrotic syndrome as the criterion. While Cameron prefers steroids, many of the other authors prefer chlorambucil. Hypertension should be controlled, and if nothing else works, proteinuria can be reduced with the use of angiotensin converting enzyme (ACE) inhibitors.
Cameron's oft quoted study reported that around 50% of patients with membranous nephropathy were either in a stable nephrotic state or in remission within 15 years. Schiepatti et al followed 100 patients with idiopathic membranous nephropathy. 65% had partial or complete remission of proteinufla in 5 years. 14% reached the end state in 71 months (SD 32). With this natural history, they point out that to achieve statistical significance using a 50% increase in median survival you need 869 patients followed for 3 years, or 571 for 5 years. No single unit is likely to have that number of patients of any one disease, and that makes it essential that we should do multicentre studies. Remuzzi et al point out a high incidence of malignancy, and a tendency to relapse after stopping drugs. My own fear is more of infection. D'Amico looked at what he called the eleven most significant studies of IgA nephropathy, and found 10 year survival of 80 to 87% with or without treatment. When results are so good without treatment, giving a drug will make no difference to the outcome. This is in the West, but for us, it is more likely to do active harm because of the risk of infection. In 6 of 7 studies of membranous nephropathy, "no better survival related to treatment could be demonstrated." And with membranoproliferative glomerulonephritis, "in three of the four studies, survival curves of treated and untreated patients did riot show better survival in treated patients."
Omar Khayyam, a Persian poet of around 1200 AD, had an apt description for our present predicament. He said:
"Myself when young did eagerly frequent
Doctor and Saint, and heard great Argument
About it and about: but evermore
Came out by the same Door as in I went."
You can find literature to support any view you have. You can also find individual patients who respond dramatically to treatment with no problems. However, it is clear that if you review the entire literature, and if you look at all the patients you have treated, treatment does not seem to make a difference.
We need to be able to identify those patients who do show dramatic responses and to treat only those. That would call for multi-centre studies, which should be taken up under the aegis of the Indian Society of Nephrology. We must be sure we are talking about the same condition, so there should be clear criteria laid down, and some authorities on pathology like Dr. Date and Dr. Bhuyan should independently review all the biopsy material to make sure we are all talking the same language. The country's leading centres between them could easily assemble the number of patients needed to reach a significant result.
Meanwhile, I must make some recommendations to you. Make a diagnosis always, and do not treat blindly. If the patient has minimal lesion nephropathy, systemic lupus erythematosus, or vasculitis, there is no dispute. Results of treatment are good and you should treat with steroids or cytotoxic drugs. Endocapillary glomerulonephritis should not be treated, as spontaneous recovery is the rule. However, one should watch out for deterioration as some of the patients might go into crescentic glomerulonephritis. We are not yet sure of the treatment of extracapillary glomerulonephritis, but the situation is desperate, and it is probably worth trying some treatment. The use of plasma exchange will be an economic rather than a professional decision. As for the others, mesanglai proliferative glomerulonephritis, IgA nephropathy, membranoproliferative glomerulonephritis and idiopathic membranous nephropathy, I would recommend only symptomatic therapy and ACE inhibitors to reduce proteinuria. Renal biopsy is essential. Some patients have drug induced glomerulonephritis with severe acute interstitial changes. Their natural history is uncertain. Some recover spontaneously, but others rapidly progress into chronic interstitial nephritis and terminal renal failure. Some of them do well with steroids if treated early, and therefore if recovery is not rapid, steroids should be instituted.
Let me stress again that treatment with toxic and potentially lethal agents need not be instituted just because we have nothing else to offer. It is better to do nothing than to do active harm.
[Table - 1]