Some of the People All of the Time, All of the People None of the Time

Article Outline

• Copyright

If you put a hundred nephrologists in a room, told them they had nephrotic syndrome, and allowed them to choose their diagnosis, they'd all probably choose minimal change disease, I said, borrowing a line I'd seen used, repeatedly and with great success, by a senior attending. The patient, an anxious man in his early 70s who'd developed new-onset lower extremity edema just a month earlier and now was hearing the report of his renal biopsy, smiled broadly and let out an audible sigh of relief. “That's because the disease is responsive to steroids about 75% of the time,” I explained.

The scene was different just a few days earlier with another patient. A woman in her 50s who'd been referred for a second opinion on fibrillary glomerulonephritis was wiping away her tears as we reviewed her history. After a year of worsening kidney function, she was biopsied with a creatinine of 3 mg/dL and found to have fibrillary glomerulonephritis. Her nephrologist had not recommended anything more than conservative therapy and had already discussed with her and her family the eventual need for renal replacement therapy.

The patient worked her way through a box of tissues and asked me if I agreed with her nephrologist that nothing could be done at this point. “My husband and I searched the internet and found something on rituximab working for this disease,” she said. Her husband chimed in at this point, relaying a story about a cousin whose lymphoma had responded to rituximab. “Why can't we try that for her?” he asked. I explained that the data on rituximab therapy for fibrillary glomerulonephritis was all anecdotal, from case reports and case series. “Those patients had much lower creatinine levels than you,” I said, “which means the disease was treated at a much earlier state.” She asked me what were the chances that rituximab would work for her. “It's just a guess,” I answered, “but probably no better than 10%.” I meant to convey pessimism, almost certain failure. Her husband spoke first. “That's better than zero percent,” he said. “I'd like to try the therapy,” she said.

The same senior attending who'd provided the line about nephrologists choosing minimal change disease as their preferred etiology of nephrotic syndrome had another line he used with patients after a course of treatment failed. “We have plenty of other treatments to offer,” he said to a patient with membranous nephropathy, running through a list of other immunomodulatory therapies. “That's the thing about membranous: there are lots of available therapies, and they work for some of the people all the time, but all of the people none of the time.”

Treating glomerular diseases feels different from treating other, more common diseases, in that there seems to be an inherent skill in choosing therapies, in knowing beforehand which therapy will work for which patient. To some extent, this is indeed true. The experts at treating glomerular diseases are those who've seen the most cases and have the experience to say to themselves, “This patient reminds me of that other patient who responded to this specific therapy.” Yet even the best nephrologists have patients who don't respond, whose treatment instincts fail them. And of those nephrologists who choose minimal change disease as their diagnosis, about a quarter of them won't respond to steroids.

While the body of data from randomized trials of glomerular diseases is growing, the bulk of evidence in this field still comes from observational studies. The art of treating glomerular diseases, therefore, still remains an art—decisions often are not easy or obvious, and clinical acumen remains a prized possession for both patients and their nephrologists.

I've had the fortunate opportunity to train at 2 centers renowned for their approach to glomerular diseases, and this experience has led to an appreciation of what distinguishes the top glomerular specialists. First, they are as up-to-date with the literature as humanly possible, knowing every case report, pilot study, and clinical trial, even those that have yet to be published. When a patient says she's found something on the internet about a promising new therapy for her disease, they know exactly what she's talking about.

Second, while they are more likely than others to initially choose the best therapy, they are truly marked by their keen instinct to recognize when a treatment is not working. Even when the literature suggested otherwise, these glomerular specialists would sometimes stop a therapy before the end of a recommended course and change directions. Their decisions were typically based on gradations of response rather than a black-and-white definition, an impression, for example, that a patient's drop in proteinuria just wasn't enough to justify keeping her on prednisone for another 3 months.

Third, they are exceptionally strong communicators, able to convey to patients why they think a treatment will work, why they recognize that a treatment is not working, and, in some cases, why aggressive treatment would be worse than no treatment at all.

Having the first 2 qualities of an enormous knowledge base and clinical acumen doesn't suffice, as patients need to have faith that their nephrologists are making sound decisions in the face of uncertain outcomes. A simple statement like, “There are lots of therapies that work for some of the people all of the time, but all of the people none of the time,” can go a long way. It is very easy to offer a therapy to a patient when it has a high likelihood of success; it is far more difficult to embark upon a treatment course when the odds are stacked in favor of treatment failure. Communication throughout either of these processes is crucial, of course, but much more so in the latter arena, which is where so many of the glomerular diseases fall.

Thus the art of treating glomerular disease also rests heavily on how data is presented to patients and how prognoses are conveyed. I stole the line about the 100 nephrologists because I saw how effectively it conveyed a good prognosis. My attempt at downplaying case reports and giving a low chance of success for rituximab to the women with fibrillary glomerulonephritis was unsuccessful, as she left the office with what I felt was an unreasonable amount of faith in the therapy. Daniel Kahneman and Amos Tversky, renowned psychologists who developed the prospect theory of behavioral economics, have shown that people's attitudes toward risks concerning gains are usually quite different from their attitudes toward risks concerning losses. Their theories hold true for nephrologists in our instinctive approach to risk/benefit discussions. For example, the claim about the 100 nephrologists is more effective at inspiring confidence if we focus on all choosing the same diagnosis rather than 25 of them failing therapy. I may have better conveyed my pessimism about rituximab if I'd told my patient the treatment had a 90% chance of failure rather than a 10% chance of success.

I started the man with minimal change disease on steroids and the woman with fibrillary glomerulonephritis on rituximab. In one case, I felt as if I'd performed as well as a glomerular expert, while in the other, I doubted my decision and how it was made. The patients' outcomes were unexpected. While the woman responded to rituximab, the man did not respond to steroids or subsequent second-line therapies, eventually requiring dialysis. Unfortunately, he died soon thereafter. Although the outcome was bad, I tried my best to communicate all decision making processes with him and his family throughout his treatment course. Because of this communication, despite the unfortunate outcome, I still felt that I'd performed as well as the experts who'd trained me.

While we work towards finding therapies that will help all of the people all of the time, we, along with our patients, will ride the uncertain pathway of treatments that only work for some of the people. If you put 100 nephrologists in a room, told them they had nephrotic syndrome, and then told them they would fail therapy, all of them would want the kind of doctor who'd quickly recognize treatment failure and just as quickly and clearly communicate to them how next to proceed, even if it's on an equally uncertain path.