This case is not really that much of a diagnostic dilemma, but it reminded me of an old case I had many many years ago. I’ll talk about that below. However, this appearance certainly caused me a bit of a pause when I was a baby pathologist, and it might be a good reminder for the younger tykes among us. It’s also a reminder to focus more on low power morphology than high power morphology. As one of my teachers, Fred Askin, told me when I was a resident, I should look at cases with a “low power objective and high power intellect.”
So, take a look at these cells from the lung:
Looks a bit like a mass of anaplastic cells, desu ne? Big variable nuclei, some folded, variable small “nucleoli”, lots of pyknosis and some cells that are at that “can’t tell if it’s pyknosis or funny looking mitosis” level. If one focuses on the high power appearance of this, it can be a bit problematic.
If you pull out a bit, they clearly in something, and if you scout around, there are a few fragments of respiratory epithelium. It’s clearly a bunch of cells in a bronchiole:
Still, it has a mass-like appearance, and you can get wacky fragments of things in the lungs…
But, if you move around more at low power, you see the more classic bronchopneumonia pattern, regardless of the cytolologic appearance:
In some of these areas, the inflammatory cells are a bit better preserved, and some “real” polymorphonuclear leukocytes are still recognizable:
So, this is just a run-of-the-mill bronchopneumonia in a person who’s been dead awhile. The dysplastic/anaplastic-appearing cells are degenerating leukocytes, probably lymphocytes and other mononuclear cells. There’s a pretty good literature on this progression. Polys fall apart first, and lymphocytes last significantly longer. So, if you are looking at a collection of cells you might not see any polys, but a bunch of atypical or barely recognizable lymphs. It kinda looks lymphoma-like to me on high power. It’s possible to make that mistake if you focus too much on high power appearance in isolation.
This came up in one of my cases many years ago when an outside consultant reviewed one of my autopsies. It was a baby found dead at home. I don’t actually remember what the child died of, but I think it was head trauma. The baby was taken to the hospital and, even thought the infant was clearly dead, a CBC was performed. It showed a white cell count of almost 100,000. On histology, the cells had this atypical appearance. The consultant claimed that I had missed a lymphoma.
But no. First, CBCs in the postmortem period are *almost* useless. The blood settles after death, and the counts are a function of where on the body you sample. In particular, if you sample just right, you can get the equivalent of a sample from the buffy coat. There was a study done back in the 1980s, I think that sampled blood from cadavers at various sites. They found large differences in hct, wbc, etc. I no longer have that reference, so if any of you old folk reading this remember it, send it to me. Second, if there is a significant postmortem interval, then doing a differential may be almost impossible.to do a differential. There are a large number of studies looking at the degradation of both red and white cells in the postmortem period. Trying to quantify that degradation is yet another windmill for time-of-death enthusiasts to tilt at.
UPDATE: I still can’t find that old article, but someone sent me a reference to postmortem CT that indicates that the hct in dependent areas rises to 80% with hypostasis: Takahashi, N., Satou, C., Higuchi, T., Shiotani, M., Maeda, H. and Hirose, Y., 2010. Quantitative analysis of intracranial hypostasis: comparison of early postmortem and antemortem CT findings. American Journal of Roentgenology, 195(6), pp.W388-W393.
As always, the pics are free for use with or without attribution, though attribution is appreciated.