I'm a bit confused by the addendum - so in practice I typically do 1:4 and sometimes go to 1:10 due to a vethive CE on SAT. I've never gone higher. Although I haven't had MANY IMHA cases that weren't just referred due to how ill they were and need for transfusion. To streamline our work since GP is always very busy, would you recommend to start at 1:10 and maybe do a second at 1:49 if positive at 1:10? I imagine at 1:10 most rouleaux would disperse by then. Or alternatively, if negative for clumping at 1:10, consider a dilution of 1:4 with the notion that some rouleaux may still be present and to definitely base a diagnosis off the Coomb's test? Ty
These are great, very deep cut questions. What I would say about the addendum from Urs Giger is that while he is super smart and knows his stuff hematology wise, he has fairly extreme opinions on a few things. One of those idiosyncratic views is that he more or less thinks you can't diagnose IMHA without a DAT/Coombs + result, and his critique of that study essentially flows from that. I think DAT can be useful to support IMHA in ambiguous cases, but in a case where it is checking every box (classic signalment, marked regen anemia, spheres, agglutination, evidence of hemolysis, increased bili etc) it seems overkill. It is interesting/curious that he critiques the Sensitivity and Specificity stats based on the small n, because those two parameters do NOT depend on disease prevalence (positive and negative predictive values DO).
To get back to the practical points, I will confess that a 1:49 dilution is pretty hardcore. I think a 1:10 is likely a good starting point, and that's what I learned in residency. To me, the biggest mistakes people make with the SAT are:
* Performing an SAT in inappropriate scenarios, such as patients who do not have any evidence of agglutination or Rouleaux (I know a lot of clinicians who tie their decision-making to HCT thresholds)
* Doing the old school 1:1 dilution, which is clearly insufficient
* Not making an actual wet mount, placing a coverslip on, and examining under the microscope (people just eye ball it grossly and call it +/-)
I'm a bit confused by the addendum - so in practice I typically do 1:4 and sometimes go to 1:10 due to a vethive CE on SAT. I've never gone higher. Although I haven't had MANY IMHA cases that weren't just referred due to how ill they were and need for transfusion. To streamline our work since GP is always very busy, would you recommend to start at 1:10 and maybe do a second at 1:49 if positive at 1:10? I imagine at 1:10 most rouleaux would disperse by then. Or alternatively, if negative for clumping at 1:10, consider a dilution of 1:4 with the notion that some rouleaux may still be present and to definitely base a diagnosis off the Coomb's test? Ty
These are great, very deep cut questions. What I would say about the addendum from Urs Giger is that while he is super smart and knows his stuff hematology wise, he has fairly extreme opinions on a few things. One of those idiosyncratic views is that he more or less thinks you can't diagnose IMHA without a DAT/Coombs + result, and his critique of that study essentially flows from that. I think DAT can be useful to support IMHA in ambiguous cases, but in a case where it is checking every box (classic signalment, marked regen anemia, spheres, agglutination, evidence of hemolysis, increased bili etc) it seems overkill. It is interesting/curious that he critiques the Sensitivity and Specificity stats based on the small n, because those two parameters do NOT depend on disease prevalence (positive and negative predictive values DO).
To get back to the practical points, I will confess that a 1:49 dilution is pretty hardcore. I think a 1:10 is likely a good starting point, and that's what I learned in residency. To me, the biggest mistakes people make with the SAT are:
* Performing an SAT in inappropriate scenarios, such as patients who do not have any evidence of agglutination or Rouleaux (I know a lot of clinicians who tie their decision-making to HCT thresholds)
* Doing the old school 1:1 dilution, which is clearly insufficient
* Not making an actual wet mount, placing a coverslip on, and examining under the microscope (people just eye ball it grossly and call it +/-)
Hope that helps!
yes that is very helpful and makes a lot of sense. ty