To summarize W’s case up to March 17, here were the salient features:
- Baseline signs of conduction system disease
- Progressive, and now severe, dyspnea on exertion
- Unexplained relative hypotension, not due to adrenal insufficiency
- Weight loss and early satiety
- Hypercalcemia, initially mild, now more pronounced, with suppression of PTH
- Markedly active urinary sediment with severe dipstick proteinuria, but also microscopic hematuria and calcium oxalate stones
- Worsening renal function, possibly pre-renal azotemia.
The 50-mile distance separating W from my office made frequent visits impractical, but from March 17 onward I was essentially on daily contact with the patient either by phone or email.
I could not tie everything together, but the thought occurred to me that he might have systemic sarcoidosis with cardiac involvement: hypercalcemia, heart block, shortness of breath, gastrointestinal symptoms. In fact, I was clearly hoping for this diagnosis as something potentially treatable in what otherwise looked like an ominous illness.
On March 19, however, a 2-view chest x-ray was normal and the light bulb that had gone off in my head a few days before was quickly burnt.
On March 20, a chart entry after a phone conversation noted that he felt overall the same (weak), but his appetite was perhaps better. Stool testing was negative for blood. Repeat blood and urine testing on March 25 are shown here.
He now had 3+ blood with only 4-10 rbc per hpf in his urine. I became very nervous as I recalled that I had put him on a high dose of atorvastatin just a few weeks before. He was not complaining of muscle aches. His urine sodium was low, his renal function was getting worse, his ACE level was normal, and I had no idea what I was dealing with.
I asked him to stop the atorvastatin. I asked the lab to add CK and urinary myoglobin measurements to the previously obtained samples. The test results came back to me on March 28, but on March 27 I received some more serious news.
On that day, his wife informed me that W had had a cardiac arrest at work and was being taken to a regional tertiary care center.
His cardiac arrest was due to ventricular tachycardia, but he was successfully resuscitated. However, he had another cardiac arrest in the ambulance, from which he was again successfully resuscitated. He was hypotensive, requiring low doses of intravenous pressors. He went into anuric renal failure.
He was admitted to the cardiology service and I had helpful exchanges with the team, as well as with the nephrology consultant, relaying to them his course over the previous months and all the data to date. I received periodic updates from them throughout his hospitalization.
He stabilized in the ICU and regained consciousness, but had recurrent episodes of ventricular tachycardia despite intravenous lidocaine and amiodarone. An echocardiogram showed severe concentric hypertrophy with preserved left ventricular systolic function. The team was puzzled by his clinical illness, and remained perplexed for several more days. Dialysis was initiated.
On April 1, he was stable enough to undergo coronary angiography which revealed normal coronary arteries. On April 5, he was extubated and was able to follow simple commands.
Immunologic studies sent on admission returned, showing no evidence of monoclonal gammopathy and no Bence-Jones proteins. Free kappa light chains levels were undetectable (<0.1) but free lamba light chains were elevated (29.9).
On April 6 he developed melena and evidence of gastrointestinal bleeding. He continued to improve neurologically.
On April 7, he underwent a right ventricular endomyocardial biopsy whose initial report revealed that:
H&E, trichrome and Congo red stained sections show three pieces of myocardium with interstitial, perimyocytic and perivascular deposits of amyloid. The iron stain is negative. No lymphocytes or plasmacytoid cells are seen. The single piece of myocardial tissue submitted in saline was snap frozen in OCT and prepared for antibody analysis by immunofluorescence. Appropriate controls were used. H&E stained slides reveal interstitial deposits of amyloid. The antibody profile showed: Amyloid A negative, prealbumin negative, kappa light chain negative, lambda light chain positive, positive control amyloid P positive. This supports AL type amyloidosis, lambda light chain restricted.
The electron microscopy report showed:
The single piece of myocardial tissue submitted in glutaraldehyde fixative shows excellent cytologic preservation. A total of 4 toluidine blue-stained thick sections were prepared from the corresponding EPON plastic blocks and show abundant amounts of amyloid deposition. This is confirmed at the ultrastructural level where the characteristic randomly branching and beaded fibrils of amyloid are observed.
The final diagnosis was amyloidosis, AL type, Lamba light chain restricted,
He continued to have episodes of fast ventricular tachycardia requiring shocks. He became more hypotensive and his mental faculty became more compromised. Following a family meeting on April 11, the decision was made to no longer provide resuscitative shocks. WW expired on April 12.
Here are some thoughts that have crossed my mind as I’ve reviewed this case:
1. Live and learn. I have rarely been as stomped as I was in this case. Cardiac AL amyloidosis is a zebra but I nevertheless exhibited some of the common cognitive biases that plague clinicians (premature closure, anchoring, etc.). To what extent were my biases accentuated by the fact that I knew that diagnostic work-ups would be limited? I could have expanded my differential diagnosis in the first couple of months of our encounter, as some unexplained abnormalities were certainly present, but I did not.
This excellent review by Dr. Rodney Falk summarizes all the important features of the disease (I don’t suppose I’ll have another opportunity to diagnose cardiac AL amyloidosis, but who knows?). In particular, he notes that: “Hypertension is unusual, and in patients with a history of hypertension, ‘spontaneous’ resolution of hypertension over the preceding few months is common.”
2. I could have asked him to go to the emergency department as early as December 2013 (4 months before he died). In all likelihood, it would not have made a difference in his outcome and, in retrospect, I’m glad I did not. But was I playing with fire trying to juggle his medical as well as financial interests?
3. For all the talk about the value of EHRs in assuring transfer of pertinent health information across clinical settings, it seems that simple phone conversations between physicians works just fine, even with very unusual and complicated cases.
4. I am not turned off to taking care of uninsured patients by this case, on the contrary. WW was a delightful man who enriched my life. As challenging as his financial limitations were, they also inspired me to be particularly attentive to his needs and helped foster a solid patient-doctor relationship.