 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Medical Forum / General / General / May 2005C-reactive protein question
As you know I am being tested for lymphoma next week. My c reactice protein was high last week, a 54. It was in the normal range this week. Does that mean it might not be lymphoma? > As you know I am being tested for lymphoma next week. My c reactice > protein was high last week, a 54. It was in the normal range this week. > Does that mean it might not be lymphoma? It means you have some inflamation from something. Could be an infection. Without knowing your full history, the rest of your labs and examining you, we don't know what is going on. Jeff But, with it being normal, a 1.1, would that mean the infection is gone? I went to see the doctor for recurrent fevers I have been having every 3 months or so. They come quickly with flu like symptoms and go quickly. My bloodwork initially showed a high c reactive protein level while I was sick. It also should my Lymphocytes to be 14.1 which was low according to the range on the test. I was refered to an infectious disease doctor this week. She took more blood and scheduled me for a Gallium scan. I just found out my c reactive protein was 1.1 with normal being 0.0 to 4 something. She did not do another cbc. On the first test pretty much everythign was normal but the lymphocytes as i said and something that says "GR" was high. White blood cells were 9.8 which was in range and red blood cells were 4.78 which was in range. The first doctor checked me for Rocky Mountain Spotted fever, EBV,CMV,liver function, thyroid, etc. All were fine. > But, with it being normal, a 1.1, would that mean the infection is > gone? It would *seem* so. But, I don't know your full history. I have not examined you. And, I have not seen your all your lab results. I don't even know if you had an infection. I think seeing an infectious disease doctor is a very good idea. And the test she has ordered seems appropriate. Jeff > I went to see the doctor for recurrent fevers I have been having every > 3 months or so. They come quickly with flu like symptoms and go [quoted text clipped - 18 lines] > The first doctor checked me for Rocky Mountain Spotted fever, > EBV,CMV,liver function, thyroid, etc. All were fine. If one's c reactive protein level was high due to lymphoma, wouldnt it stay high as long as the lymphoma was present? > But, with it being normal, a 1.1, would that mean the infection is > gone? [quoted text clipped - 5 lines] > My bloodwork initially showed a high c reactive protein level while I > was sick. As it should. High CRP means you're sick ! > It also should my Lymphocytes to be 14.1 which was low > according to the range on the test. 14.1 what ? Grams per litre ? Pints ? Pence per gallon ? > I was refered to an infectious disease doctor this week. She took more > blood and scheduled me for a Gallium scan. Lymphoma isn't considered an infectious disease. > I just found out my c reactive protein was 1.1 with normal being 0.0 to > 4 something. She did not do another cbc. C reactive protein isn't part of a cbc > On the first test pretty much everythign was normal but the lymphocytes > as i said and something that says "GR" was high. > > White blood cells were 9.8 which was in range and red blood cells were > 4.78 which was in range. If your white cell count is 9.8 I think it's rather unlikely that your lymphocytes would be 14.1. Again, what are the units. I'm reminded of the apochrypal tale of the medical student with Hodgkin's lymphoma who committed suicide for fear of the lymphome. At post mortem it transpired there was nothing wrong with him. Talk to your doctor. > > But, with it being normal, a 1.1, would that mean the infection is > > gone? [quoted text clipped - 37 lines] > > Talk to your doctor. Gnostic nonsense. Pungent fragrance. Slathered sandwish. Mayonaise. Grey Poupon. Pampered pillow. This is what the cbc looks like: Patient Limits 3 Wbc 9.8 x10^3/uL 4.5-10.5 LY 14.1 L % 20.5-51.1 MO 2.7 3 % 1.7-9.3 GR 83.2 3H % 42.2-75.2 LY# 1.4 x10^3/UL 1.2-3.4 MO# 0.3 3 x10^3/uL 0.1-0.6 GR# 8.2 3H x10^3/uL 1.4-6.5 RBC 4.78 x10^6/uL 4.0-6.0 Hgb 14.1 g/dl 11.0-18.0 Hct 42.1 % 35.0-60.0 MCV 87.9 fl 80.0-99.9 MCH 29.5 pg 27.0-31.0 MCHC 33.5 g/dl 33.0-37.0 RDW 13.7 % 11.6-13.7 PLT 171 x10^3/uL 150-450 MPV 7.9 fL 7.8-11.0 Any help is appreciated! Yes, But what are standard ranges of norm, For each measure? I assume the range of numbers under "Patient limits". What is on there is exactly as it is on my sheet. stry...@hotmail.com wrote: > I assume the range of numbers under "Patient limits". What is on there > is exactly as it is on my sheet. You need a professionl to interpret those numbers, for norm range, if you are not familiar with medical research, eg, using a Merck Manual etc. Best to seek an expert. Is there somewhere or someone online that could help me? > Is there somewhere or someone online that could help me? Here is a link to an article discussing CRP. CRP is only one indicator, and not conclusive of anything. The article discusses issues which I believe you have encountered before. http://www.yourhealthbase.com/heart_CRP.htm Comparing to the lab norms for my drs. lab: You seem to be in normal range for all but GR# where you are apparently 8.2 and norm is 1.4 to 6.5 I do not know what this means. Jackie > This is what the cbc looks like: > [quoted text clipped - 17 lines] > > Any help is appreciated! http://web2.iadfw.net/uthman/lab_test.html Scroll down to the section on Hemoglobin/Hematocrit, etc. This website is by a clinical pathologist and has lots of information regarding lab testing. Your total WBC (white cell count) is 9.8, which is on the high side, but according to the lab normal posted, still in normal range. LY = Lymphocytes, which are at 14.1%. That means that, of the 9.8 X 10^3, 14.1% of your white cells are lymphocytes. The percentages of each type of cell should add up to 100%. MO = Monocytes GR = Granulocytes (which include segmented neutrophils, eosinophils, and basophils. Most blood counting instruments lump these all together). The numbers that follow are the ABSOLUTE counts of each, not percentages. Dr. Uthman explains all this on this page of his: http://web2.iadfw.net/uthman/blood_cells.html Absolute counts are actually more diagnostic than percentage counts, and can vary based on diseases. All of these parameters can help diagnose different diseases that can present in a routine blood count. Hematology is really really complicated. I know basic stuff, and that's about it. The other numbers are red cell indices, indicating size of red cell. They are calculations based on the hemoglobin. RDW is red cell distribution width. If you have lots of different sizes of RBC's, your RDW will be abnormal. PLT is platelet count. Yours is normal. I don't know what MPV is. The reason that you can't get anyone to commit to a diagnosis on this group is that this group is peopled by LAB TECHNOLOGISTS. We run the tests, but it is NOT OUR BUSINESS to make diagnoses. We furnish the doctors the raw data from the tests ordered BY THE PHYSICIANS and then the doc, based on lab results, his history, and physical exam, the latter two of which the lab is NOT privy to, in most cases, makes a diagnosis and a recommendation for treatment. I am confused as to why you are assuming you have lymphoma. Are you cashing in all your chips on one elevated CRP result? There must be something in your history that prompted your seeing an ID doctor. http://www.c-rptest.com/about_crp.asp Here is some information on the different types of CRP testing. I have been in microbiology exclusively since 1988, so don't know much about this testing. I was always taught that CRP was an indicator of inflammation. I know they are using it now with cardiac testing, but don't know much about it. http://www.questdiagnostics.com/hcp/topics/cardiocrp/cardiocrp.html We can only help you so much on this newsgroup, as we are, for the most part, NOT physicians. Hope you do understand our predicament. Judy Dilworth, M.T. (ASCP) Microbiology (but did some basic Chemistry/Hematology way back when for awhile....:-)) > This is what the cbc looks like: > [quoted text clipped - 17 lines] > > Any help is appreciated! > http://web2.iadfw.net/uthman/lab_test.html > [quoted text clipped - 12 lines] > GR = Granulocytes (which include segmented neutrophils, eosinophils, and > basophils. Most blood counting instruments lump these all together). His granulocyte count was high enough to call for a manual differential. I don't understand why that lab did not do it. It would be interesting to see if he had any immature neutrophils. Many really modern instruments (including the new coulter series with the laser flow cytometer) differentiate between Segs, eos, and baso's now. The 3 dimentional scattergrams can also show you if there is a population of immature cells and which cell line they are in. The thing is, the reduced lymphocyte percentage does not necessarily mean anything. The elevation of the granulacyte count and skew that and may even indicate bacterial infection. The fever is low grade tho'. He said he had some gastrointestinal problems earlier in his history. There might be something going on there! > The numbers that follow are the ABSOLUTE counts of each, not > percentages. Dr. Uthman explains all this on this page of his: [quoted text clipped - 11 lines] > RDW will be abnormal. PLT is platelet count. Yours is normal. I don't > know what MPV is. Mean platelet volume. Indicates platelet size. > The reason that you can't get anyone to commit to a diagnosis on this > group is that this group is peopled by LAB TECHNOLOGISTS. We run the [quoted text clipped - 3 lines] > latter two of which the lab is NOT privy to, in most cases, makes a > diagnosis and a recommendation for treatment. Very true. ;-) It's a standing joke in the lab field that we know just enough to be dangerous. <lol> Diagnosis cannot be made on the basis of lab results alone! A patient history must also be considered along with other diagnostic tests outside of the laboratory. Has he been checked for malaria? How severe are the fevers? Has he ever been outside of the country? What is his eosinophil count? There are other blood parasites that can do this. This is why I pointed him to google. There are dozens of possible causes for intermittant fevers! (fever of unknown origin). I'd not panic about Lymphoma until it was proven without a doubt to be the problem... > I am confused as to why you are assuming you have lymphoma. Are you > cashing in all your chips on one elevated CRP result? There must be > something in your history that prompted your seeing an ID doctor. > > http://www.c-rptest.com/about_crp.asp Wonder if they ran a sed rate... Just kidding. ;-) I think he is just scared. People tend to presume the worst. I think the doctor that even told him of that possibility with no absolute proof of it was, well, irresponsible at best. > Here is some information on the different types of CRP testing. I have > been in microbiology exclusively since 1988, so don't know much about [quoted text clipped - 11 lines] > (but did some basic Chemistry/Hematology way back when for > awhile....:-)) I read his CBC and I don't see how anyone can attempt to make a diagnosis based on that alone. AFAIK, CRP is not specific for lymphoma either? I also wonder if anyone even looked at the peripheral smear, and was it reviewed by a pathologist??? K. MT (ASCP) former hematology supervisor..... > > This is what the cbc looks like: > > [quoted text clipped - 18 lines] > > Any help is appreciated! >  SignatureK. Sprout the MungBean to reply "I don't like to commit myself about heaven and hell‹you see, I have friends in both places." --Mark Twain Thanks for the information. This is very educational and interesting. Have not been out of the country. Not sure what eosinophil count is. In December of 2003 I had 5 days of really bad flu. SInce then, I have had fevers and flu like symptoms come suddenly every 3 months or so. My fever goes up to about 100 to 102, I get chills and achs. I go to bed and usually feel a little better when I get up and by the next day feel fine. Starting in December 2003 I have lost some weight but I also went on a diet at this same time. I have lost abotu 30 lbs since 2003. I was on Atkins for a long time. (Not the full blown atkins) I have been off it for awhile and have gained some back but not much. For the last week I have noticed in the afternoons my temperature goes to around 99.2. In the morning it is 97.5 or so when I first get up. It usually goes to 98.7 or so right before bed. Two years ago I had a swollen lympth node under my arm that was painless. Went to the family doctor, he did a cbc which was normal. The node went back to normal in about a week. I have had stomach bloating and constipation for the last 3 years. I take 4 table spoons of Milk of Magnesia every night. By the way, this low grade fever has lasted for about 2 weeks or so. It used to go away but is kind of lingering. It is funny, It got up to 99.2 or so after work yesterday when I got home and I panicked because I was sure it was lymphoma. I called my uncle and had a nice conversation with him for about an hour on the phone untill bedtime. I checked my temperature before going to bed and it was 98.6 or so. Stryped, what you're describing is a classic "fever of unknown origin" case. These are usually referred to infectious disease doctors, who tailor their workup to your history. The fact that you have repeated fevers indicate to me that you need to visit one SOONER than later. Have your doc give you a referral. Family docs and internal medicine people usually don't have the investigatory skills to work out the many causes of this diagnosis. ID docs completed internal medicine first - before they specialize in infectious disease. Therefore, if your problem doesn't turn out to be caused by an infection, they should be able to handle getting you to the correct doctor. Eosinophils are a type of blood cell (granulocyte) that can become extremely elevated in parasitic infections, or mildly elevated in an allergy situation. In micro, we get the results of FUO workups. I'm sure Katra sees the hematology part of the workup, but micro and serology sees the serological and culture side. The ID doc will take a good history from you (hope you have good insurance, because an FUO workup can cost some bucks). He/she should include a detailed travel history, even if it was not out of the country. Certain parts of the US (are you in the US???) and I'm sure other countries are "notorious" for certain diseases and they will want to know that. Body temperatures cycle up and down during the day (ask any woman who's checking temperatures for ovulation). A 99.2 temp down to a 98.6 isn't really much. A temp of 100-102 is, however. http://www.wrongdiagnosis.com/sym/fuo.htm I don't know anything about this website (found it on Google) but this should keep you busy for awhile.... This may be a better website, and includes an "official" definition of FUO: http://www.5mcc.com/Assets/SUMMARY/TP0341.html Judy Dilworth, M.T. (ASCP) Microbiology > Thanks for the information. This is very educational and interesting. I worked for a private laboratory at a small draw station years ago. We had strict guidelines when a manual differential should be run, and when the tube should be sent up to the "main lab" for more intensive review. My guidelines were pretty strict, which was good, I think, as I was by myself and had no one else to run questions by when diffs got weird. I was pretty much limited to signing out extremely normal CBC's on my own. This was good as my main background up until that point was microbiology (although I think I could still recognize a good Dohle body :-)). I would think the elevated granulocytes alone would flag. I'm also surprised a manual differential wasn't done. Stryped - eosinophils are lumped into the granulocyte count. Granulocyte count is a total of neutrophils, eosinophils, and basophils. If grans are elevated, usually a manual slide is made and looked at - but it depends on your lab's guidelines for making manual slides, and the scattergram results you get in the lab on your printout, along with lots of other factors (that Katra knows about and I do not.) Katra - where do the monos go on a three part diff? Just curious. I never worked with one of these BIG :-) counters. My experience goes back to some weirdo Corning instrument that was old in 1988 (last time I did Hematology) and an old Coulter S I used in the mid-80's (a castoff they brought to my draw station lab). The scattergram technology was just starting in the mid-80's, and I never got to work with it. Yes, I agree - how did we get to lymphoma from one CBC and CRP result? Judy Dilworth, M.T. (ASCP) Microbiology > I worked for a private laboratory at a small draw station years ago. We > had strict guidelines when a manual differential should be run, and when [quoted text clipped - 7 lines] > I would think the elevated granulocytes alone would flag. I'm also > surprised a manual differential wasn't done. Three part differentials are no less accurate then 5 part differentials. In some instances they are more accurate meaning the problems associated with getting an accurate basophil eos count with the 5 part is common. Flags are important but if no flags are present than the mere presence of increased granulocyte percent is not important. Bands are not important and immature in the myelocte metamyeolocyte would generate a flag for manual diff. The total WBC is often a better predictor of infection in many cases. > Stryped - eosinophils are lumped into the granulocyte count. Granulocyte > count is a total of neutrophils, eosinophils, and basophils. If grans > are elevated, usually a manual slide is made and looked at - but it > depends on your lab's guidelines for making manual slides, and the When the granulocyte percent is very high in the 90% range with the three part diff is common to generate a manual diff but usually the trigger is a flag before it reaches that range. > scattergram results you get in the lab on your printout, along with lots > of other factors (that Katra knows about and I do not.) > > Katra - where do the monos go on a three part diff? Just curious. I Granulocytes mononuclears lymphocytes. LY 14.1 L % 20.5-51.1 MO 2.7 3 % 1.7-9.3 GR 83.2 3H % 42.2-75.2 > never worked with one of these BIG :-) counters. My experience goes back These are not big counters. 2 ft by 2 1/2ft front view. The 5 part ones are much larger. > to some weirdo Corning instrument that was old in 1988 (last time I did > Hematology) and an old Coulter S I used in the mid-80's (a castoff they > brought to my draw station lab). The scattergram technology was just > starting in the mid-80's, and I never got to work with it. > > Yes, I agree - how did we get to lymphoma from one CBC and CRP result? My impression was the doctor told him the possibility of that as he was referred to an infectious disease doctor and working down the differential diagnosis. Work in progress. > Judy Dilworth, M.T. (ASCP) > Microbiology ~ * Work In Progress, Indeed ~ ! * Woe Be Gone. ~ * ~ ________________________________________ ~ * Just One of The Morning Wood Astonishing Moments * ~ ________________________________________ ~ A Symphonic Composition ~ "Descend, Not Really a Sin..." ...steps, Of night ~ Morning shines, Yes, In Morning Wood... ~ la musique des foi ~ O that shimmering ~ The box of I! ... ... Our voices return To Eco ~ The posterior music Of demounting sure black Of The Night. ~ * ~ ~ ~ _________________________________________ * Dogging Arts * Fogging Minds * It's a Star * _________________________________________ * ~ * ~ Blog, or dog? Who knows. But if you see my lost pup, please bring him home! I got Leon a brand-new bone. _________________ http://journals.aol.com/virginiaz/DreamingofLeonardo > ~ * Work In Progress, > Indeed ~ ! * Sorry about the wording there as there is a working diagnosis until a definitive diagnosis is reached after the studies to rule in or rule out the working diagnosis. I forgot to mention that any eosinophilia would be evident in a histogram of the granulocytes as atypical and thus a flag would be generated even though the percent granulocytes may be normal in a three part diff. The basophils are also registered in the mononuclear fraction with the monocytes. > > ~ * Work In Progress, > > Indeed ~ ! * [quoted text clipped - 6 lines] > the percent granulocytes may be normal in a three part diff. The basophils > are also registered in the mononuclear fraction with the monocytes. Uh, not on any Sysmex or Coulter instrument I've ever worked on! Basophils are included in the 3 part diff granulocyte count. 3 part diff machines are primitive at best these days......... ;-) Even the old Coulter HMX does a 5 part diff and it's considered OLD. Don't you upgrade every 5 years???? Any ANY granulocyte count over 80% should get a manual diff!!!!! To not do so it just sheer laziness. And most MD's would NOT agree with you that bands are not important. Depends on how many there are....... SignatureK. Sprout the MungBean to reply "I don't like to commit myself about heaven and hell‹you see, I have friends in both places." --Mark Twain "Katra" <KatraMungBean@Centurytel.net> wrote in message > > I forgot to mention that any eosinophilia would be evident in a histogram of > > the granulocytes as atypical and thus a flag would be generated even though > > the percent granulocytes may be normal in a three part diff. The basophils [quoted text clipped - 3 lines] > > Basophils are included in the 3 part diff granulocyte count. We used the Sysmex instrument several years back right after the coulter S and I thought we called them mononuclear fraction instead of monocytes but I might have messed that up as it has been years. > 3 part diff machines are primitive at best these days......... ;-) They are fine for low volumne. > Even the old Coulter HMX does a 5 part diff and it's considered OLD. Well it's not so much the machine as the flags and cuttoffs for smear review that are important. If you have an overly sensitive machine that flags everything including things that are not there then it becomes more of a problem. According to the technical reps any abnormal parameter and or flag present should call for a smear review. So that doesn't change whether it is a three part diff or 5. You also have the doctors who do not care for an auto diff no matter what it is. They simply want a manual one so the machine is worthless for that. > Don't you upgrade every 5 years???? Not every five years, no. We have CD 4K and CD 3200's at our place. The 3200 gives band flags all the time when they are not there and the 4K doesn't see them when they are there. > Any ANY granulocyte count over 80% should get a manual diff!!!!! That varies from institution to instiution and instrument. We do not do smear reviews because of any granulocyte percentage on the Cell Dyn's. We have smear reviews for Total WBC and white cell flags including immature band blast atypical lymphs etc. Years ago with the Sysmex three part diffs we did have on greater than 90% granulocytes. > To not do so it just sheer laziness. We have smear reviews or scans on those above and if a manual diff is needed then it is done otherwise the auto diff is reported with a comment that it was smear reviewed. We are a teaching hospital of 500 beds and yeah we are lazy. The interns still order manual diffs and want them on all newborns and elderly regardless of the autodiff. > And most MD's would NOT agree with you that bands are not important. > Depends on how many there are....... I agree with you that most MD's would not agree with bands. They have done many studies concerning the tech variation of scoring bands and found CV's over 20% making it unreliable. I might get 2 and somebody next to me gets 20%. The literature is out there and it will go the way of the bleeding time. I bet the same doctors think the bleeding time is great also. The Mayo Clinic laboratories does not report bands for that reason. Because of the unreliability of the counts the band is a poor predictor of anything and like the PFA in replacing the bleeding time, if you can get a machine that gives an accurate band percentage that is reproduceable then that might be revisited. in both places." --Mark Twain > "Katra" <KatraMungBean@Centurytel.net> wrote in message > > I forgot to > mention that any eosinophilia would be evident in a histogram of [quoted text clipped - 11 lines] > and I thought we called them mononuclear fraction instead of monocytes but I > might have messed that up as it has been years. No comment on that one. ;-) Basophils are in the granulocytic cells... Trust me. > > 3 part diff machines are primitive at best these days......... ;-) > They are fine for low volumne. [quoted text clipped - 5 lines] > everything including things that are not there then it becomes more of a > problem. I have to agree with you on that! Even then, Suspect flags still get diffs. Talking with the Coulter hematology applications specialist, even she, working with the medical staff and the pathologists on differential review criteria, has only been able to get manual slide reviews to 20% of CBC's at best in larger facilities. We are only a 120 bed hospital with a large out patient population and OP surgical center. I had the criteria from several different hospitals faxed to me to help us decide on our numerical cutoffs. Those are combined with both/either suspect or definitive flags. Numbers alone are _far_ from the only criteria, but there are common sense numerical settings AFAIK. The thing is, if your machine is flagging things that are not there _frequently_, you need to have a service rep. check the electronic settings. Keep a copy of the CBC's and the manual diff reviews (and do 200 cell diffs since machines tend to be more accurate, especially if a tech is counting the wrong area of the slide, or you are creating poorly made/too thick slides). There are OH so many factors to winnowing out problems! But when it comes to patient care, it's better to be safe than sorry! > According to the technical reps any abnormal parameter and or flag present > should call for a smear review. So that doesn't change whether it is a three > part diff or 5. True. > You also have the doctors who do not care for an auto diff > no matter what it is. They simply want a manual one so the machine is > worthless for that. Heh. Glad to see THAT problem is universal! <lol> > > Don't you upgrade every 5 years???? > > Not every five years, no. We have CD 4K and CD 3200's at our place. The 3200 > gives band flags all the time when they are not there and the 4K doesn't see > them when they are there. The LH sometimes flags bands or blasts too when they are not there... Believe it or not, sometimes that is more of a problem with mixing and stabilization of the cells with the EDTA anticoagulant. We see it more in fresher specimens from the ER. Additional mixing and a re-run often get rid of erroneous flags. With morning run samples that sit in the phlebotomy tray for any length of time, it's not usually as much of a problem. > > Any ANY granulocyte count over 80% should get a manual diff!!!!! > > That varies from institution to instiution and instrument. Yes, but, there are STILL national accepted standards. When I had to set up the criteria, I got input (as suggested by both Coulter and CAP) from several hospitals using the same instrument. But, think about it! 80% segs is kinda high. We are dealing with people's LIVES here! Do you really want to take silly chances??? > We do not do smear reviews because of any granulocyte percentage on the Cell > Dyn's. We have smear reviews for Total WBC and white cell flags including > immature band blast atypical lymphs etc. Indeed. :-) > Years ago with the Sysmex three part diffs we did have on greater than 90% > granulocytes. Sorry. That is too high IMHO. > > To not do so it just sheer laziness. > [quoted text clipped - 4 lines] > The interns still order manual diffs and want them on all newborns and > elderly regardless of the autodiff. We do manual diff's as requested, and also ALL pedi's 12 and under get a manual diff. Liability on pediatric patients is too much of an issue with our litiginous society to take any risks. And it keesp the pediatricians off of our backs. <lol> It's really not that much trouble, even with a high workload. > > And most MD's would NOT agree with you that bands are not important. > > Depends on how many there are....... > > I agree with you that most MD's would not agree with bands. > They have done many studies concerning the tech variation of scoring bands > and found CV's over 20% making it unreliable. C'mon dear! CV values are MEANINGLESS on small cell populations! Try doing CV calculations sometime on Digoxin control results. <lol> The smaller the numerical value, the less reliable CV calculations are. Ever looked at the CV's of a 5 part diff on Eos and Baso control numbers? > I might get 2 and somebody next to me gets 20%. That is an intralaboratory training issue!!! With proper training and in-servicing, you can correct that problem. I know. I had to work on it in our own lab to get better consistancy on reporting. Annual tech reviews help a LOT on that, and it's a CAP requirement. > The literature is out there > and it will go the way of the bleeding time. I bet the same doctors think > the bleeding time is great also. <grins> For the vast majority of laboratories that do not posses a machine for Platelet function studies, it's currently the simplest and most reliable plt function test. I'm not overly fond of it, but it's all we have for now. And it's not ordered frequently thank goodness! > The Mayo Clinic laboratories does not report bands for that reason. Because > of the unreliability of the counts the band is a poor predictor of anything > and like the PFA in replacing the bleeding time, if you can get a machine > that gives an accurate band percentage that is reproduceable then that might > be revisited. Give Coulter time....... ;-) The thing is, we still have to serve the physicians needs. Not wanting to do manual slides.......... Well, I think I made my feelings on that abundantly clear. Thanks for the input! This is fun! SignatureK. Sprout the MungBean to reply "I don't like to commit myself about heaven and hell‹you see, I have friends in both places." --Mark Twain > > "Katra" <KatraMungBean@Centurytel.net> wrote in message > > I forgot to > > mention that any eosinophilia would be evident in a histogram of [quoted text clipped - 14 lines] > No comment on that one. ;-) > Basophils are in the granulocytic cells... Trust me. That wasn't the point I was making. I am aware that basophils are granulocytes. I am also aware that machines think that blast are sometimes basophils or that blast are monocytes. You need to talk to the machine about that. Trust me on that one the machines don't always get it right. Sometimes you don't even know what the machine is looking at. > > > 3 part diff machines are primitive at best these days......... ;-) > > They are fine for low volumne. [quoted text clipped - 8 lines] > I have to agree with you on that! Even then, Suspect flags still get > diffs. At your place maybe. I suggest you look at scan criteria. Me thinks you do not do scans but manual diffs on everything. We have criteria on when to reject the autodiff and scan the smear to see if the autodiff is valid and if not then a manual diff is done or accept the autodiff or criteria in rejecting the autodiff all together and only reporting a manual differential. Two sets of criteria. Talking with the Coulter hematology applications specialist, even > she, working with the medical staff and the pathologists on differential > review criteria, has only been able to get manual slide reviews to 20% > of CBC's at best in larger facilities. We are only a 120 bed hospital > with a large out patient population and OP surgical center. At a 120 bed hospital you don't even need a autodiff. I am sure the Coulter specialist can put all the medical staff in one room and talk to them. That's good. > I had the criteria from several different hospitals faxed to me to help > us decide on our numerical cutoffs. Those are combined with both/either > suspect or definitive flags. They all have coulters and the same machine? We don't have the same criteria for the CD 3200 and CD 4K. How could you as they are different machines and different technology lasers. Are you using their same reference range also. > Numbers alone are _far_ from the only criteria, but there are common > sense numerical settings AFAIK. > > The thing is, if your machine is flagging things that are not there > _frequently_, you need to have a service rep. check the electronic > settings. They do frequently and we reset frquently and we recalibrate frequently. These are high volumne output regardless of the instrument used they take a beating. The whole arguement on their side is that you want it sensitive so you don't miss anything. > Keep a copy of the CBC's and the manual diff reviews (and do 200 cell > diffs since machines tend to be more accurate, especially if a tech is The number of cells counted on the manual count is WBC count dependent with a 300 count in the very high range. > counting the wrong area of the slide, or you are creating poorly > made/too thick slides). There are OH so many factors to winnowing out > problems! I am saying that not all autodiffs are accurate and not all autodiffs are inaccurate. That all depends on reliance of validation studies that take into account all the stuff. > But when it comes to patient care, it's better to be safe than sorry! Nobody is implying otherwise. > > According to the technical reps any abnormal parameter and or flag present > > should call for a smear review. So that doesn't change whether it is a three [quoted text clipped - 7 lines] > > Heh. Glad to see THAT problem is universal! <lol> That really isn't a problem. There are clincial situations where not only the white cell diff is important to them but the red cell morphology is also and we have to scan a smear to be able to enter the RBC morph. > > > Don't you upgrade every 5 years???? > > [quoted text clipped - 6 lines] > stabilization of the cells with the EDTA anticoagulant. We see it more > in fresher specimens from the ER. We have our own ER lab and that is true sometimes but the technology is different for both instruments in not only diffs but in H and Hs MCHC also. > Additional mixing and a re-run often get rid of erroneous flags. > [quoted text clipped - 8 lines] > When I had to set up the criteria, I got input (as suggested by both > Coulter and CAP) from several hospitals using the same instrument. THat is your national standard. Then I misunderstood your national standards. CAP inspections are regional. You can set anything you want as long as you have validation studies. If those are not working well then you can reset to diffenent levels with validation studies. > But, think about it! 80% segs is kinda high. > We are dealing with people's LIVES here! > Do you really want to take silly chances??? So a 79% is OK and a 81% is not OK. Do you really have confidence in your autodiffs? I don't think so when you talk like that. It is not the relative percent that is important and you seem to make a mistake made by many non professionals over the intenet and that is look solely at the relative percent and not the absolute count. Penias and philias are defined clinically by the absolute counts. Please look up and show me where you see an 84% granulocytes as clinical neutrophilia? You are not going to save any lives by making that mistake. > > We do not do smear reviews because of any granulocyte percentage on the Cell > > Dyn's. We have smear reviews for Total WBC and white cell flags including [quoted text clipped - 6 lines] > > Sorry. That is too high IMHO. Just trying to remember but I think that was right or maybe 85% not really sure now but other criteria would have caught any significant abnormalities. > > > And most MD's would NOT agree with you that bands are not important. > > > Depends on how many there are....... [quoted text clipped - 5 lines] > C'mon dear! > CV values are MEANINGLESS on small cell populations! Those are clinical studies of CAP interests. The range of reporting is the problem. It is a problem and that problem like any test that is not performed properly is worthless clinically. Again the hallmark of any test is it's use clinically. Take one person from every state in the union and you will see a spread on reporting bands. That's a fact out there. > Try doing CV calculations sometime on Digoxin control results. <lol> > > The smaller the numerical value, the less reliable CV calculations are. There are clinical studies out there in sepsis and markers for sepsis and you will find that absolute white cell count vs granulocyte count vs band count that bands show poorly. Not only clinically but in performance. > Ever looked at the CV's of a 5 part diff on Eos and Baso control numbers? > > > I might get 2 and somebody next to me gets 20%. > > That is an intralaboratory training issue!!! Yes. It is an intralaboratory, interlaboratory and more importantly a clinical issue that has not gone away. > With proper training and in-servicing, you can correct that problem. CAP has been trying to do it for years and has failed. > I know. I had to work on it in our own lab to get better consistancy on > reporting. Annual tech reviews help a LOT on that, and it's a CAP > requirement. > > The literature is out there > > and it will go the way of the bleeding time. I bet the same doctors think [quoted text clipped - 4 lines] > Platelet function studies, it's currently the simplest and most reliable > plt function test. Again the way a test is appraised is not on how simple a test is but it's clinical utility. It is not reliable clinically which is why CAP has pretty much clamped down on when to order one to the point that they are rarely down anymore. Thanks for catching me up on technology in the hematology department. I forgot that many instruments are not nearly as huge as they were back in the 80's. I didn't realize that these instruments were that sensitive to abnormal cells. It makes sense that eos would give a different scattergram than other granulocytes. I received a cursory onceover with this technology almost twenty years ago, so thanks for the review. When I started working in labs back in 1971, I was a secretary for two years before I went into MT training. I remember that they were working with Coulter F's and my lab upgraded to an S right before I went into training in June of 1973. This was a HUGE BIG DEAL back then, believe me. The idea that you could hold a tube of blood up to an instrument and it would spit out an entire hemogram was just an amazing thing. Platelet counting didn't come until later, and the S did not include a platelet count. It did include all the indices, though. When I was a secretary, my tiny office was right off of the Hematology lab, so I could hear the hematology conversation all day. It was quite interesting. They were still doing platelets manually, and therefore only did them if one was ordered, although we always reported a platelet estimate on every slide. Reticulocyte counts were also done manually. Those are automated now, aren't they? My room was the file room, and I filed tons of reports, as we were not computerized back then. Everything was on paper and charted by hand. I still had to perform manual RBC and WBC counts and compare to the Coulter when I was a student. I don't know how long they required that to be done, but it sure was a waste of time (although it taught you to hate manual counts - talk about tedious!). When I worked for a private lab in the 80's, we had a small counter that was similar to a Coulter F that required cup dilutions for the counts, then lysing for the hemoglobins. It was a nice little counter, though - I don't remember the brand name. They bought me that after the old Coulter S kind of died. The latter ate a lot of expensive reagents. Also sorry I didn't remember the first part of the thread. I have my newsgroup reader set to only show new entries and didn't go back to review before I wrote. Thanks for all the great information. Judy Dilworth, M.T. (ASCP) Microbiology [lots of interesting stuff :-)] > Thanks for catching me up on technology in the hematology department. I > forgot that many instruments are not nearly as huge as they were back in [quoted text clipped - 42 lines] > > [lots of interesting stuff :-)] Meant to say yes retics are automated and routinely used and some instruments such as Cell dyn 4000 can be used in flow cytometry. The laser instrument has been adapted from other flow cytometry for use in CD4 counts for example or fetal cell detection for Rhogam testing or immunological platelet counts with monoclonal antibodies. > I worked for a private laboratory at a small draw station years ago. We > had strict guidelines when a manual differential should be run, and when [quoted text clipped - 4 lines] > This was good as my main background up until that point was microbiology > (although I think I could still recognize a good Dohle body :-)). <grins> I know what you mean... It's nice to be able to run weird stuff by other co-workers! Same goes for Gram stains, (especially on postive blood cultures) and O&P's if you have a poor specimen... > I would think the elevated granulocytes alone would flag. I'm also > surprised a manual differential wasn't done. Indeedy! > Stryped - eosinophils are lumped into the granulocyte count. Granulocyte > count is a total of neutrophils, eosinophils, and basophils. If grans > are elevated, usually a manual slide is made and looked at - but it > depends on your lab's guidelines for making manual slides, and the > scattergram results you get in the lab on your printout, along with lots > of other factors (that Katra knows about and I do not.) Heh. When we got the Coulter LH-750, our lab was still doing manual diffs on about 80% or more of CBC's. It was ridiculous! I was able to work with the pathologist to re-set our standards based on nation wide standards set by other LH users. We now are down to 20% to 25% of CBC's which is why it ticks me off when some people get "lazy" on us. <sigh> I finally discussed the issue with the new Hem. supervisor. I advised her that if she and others chose to make "exceptions" to the rules, it needed to be documented and approved by the medical staff, then signed off by the path. in charge. I'm still training her it seems. If she does not document the "exceptions" in the diff procedure, she is gonna get nailed by CAP on the next inspection! I can't help but wonder if someone got "lazy" on his CBC and chose not to do a manual diff. when it should have been! > Katra - where do the monos go on a three part diff? Just curious. I > never worked with one of these BIG :-) counters. My experience goes back > to some weirdo Corning instrument that was old in 1988 (last time I did > Hematology) and an old Coulter S I used in the mid-80's (a castoff they > brought to my draw station lab). The scattergram technology was just > starting in the mid-80's, and I never got to work with it. Oh my! The new Coulter laser technology is a real DREAM! Even the last Sysmex we had did a 5 part diff and flagged if there were bands/immature cells. The newer 5 part diff technology is well over 10 years old now. Unlike the Coulter, the Sysmex used lyse-specific reagents. The Coulter uses Laser flow cytometry technology. Much better and fewer boxes to move around <lol> I'm not sure I understand your question? Are you asking about Scattergrams or Histograms? They are on the lower part of the scattergram and are well grouped. The new LH series has a 3 dimentional scattergram and you can remove "cell groups" to visualize each one separately. It's AWEsome! The cell groups are also different colors. If you remove Eos, Baso's, Seg's, Lymph's and unlysed RBC groups, you can see the Mono group by itself and look for a separate grouping for immature cells. Pro-Monocytes and Mono-Blasts are rare unless you run into the rare Monocytic Leukemias... > Yes, I agree - how did we get to lymphoma from one CBC and CRP result? > > Judy Dilworth, M.T. (ASCP) > Microbiology :-) SignatureK. Sprout the MungBean to reply "I don't like to commit myself about heaven and hell‹you see, I have friends in both places." --Mark Twain > > I worked for a private laboratory at a small draw station years ago. We > > had strict guidelines when a manual differential should be run, and when [quoted text clipped - 35 lines] > not document the "exceptions" in the diff procedure, she is gonna get > nailed by CAP on the next inspection! Maybe you can show me where in the CAP inspection it says that specimens that have a greater than 80% granulocytes need a manual diff. That may be in your lab but that in now way is a CAP standard. This is from the CAP website checklist. HEM.34200 Phase II N/A YES NO Has the laboratory established criteria for checking and reviewing leukocyte differential counter data, histograms, and/or blood films for clinically important results flagged by the automated differential counter? COMMENTARY: The laboratory must have defined protocols for validation and review of automated WBC differential counter results for clinically significant findings. These include pathologic quantities of normal cell types and abnormal cells. Flagging mechanisms include those within the particular instrument, inspection of histographic/cytographic displays, laboratory criteria based on local experience, and awareness of published evaluations. REFERENCES: 1) Rümke CL. The statistically expected variability in differential leukocyte counts. In: Differential leukocyte counting, CAP conference/Aspen. Northfield, IL: CAP, 1977:39-45; 2) Payne BA, Pierre RV. Using the three-part differential: part II. Implementation of the system. Lab Med. 1986;17:517-522; 3) Kalish RJ, Becker K. Evaluation of the Coulter S-Plus V three-part differential in a community hospital, including criteria for its use. Am J Clin Pathol. 1986;86:751-755; 4) Ross DW, Bentley SA. Evaluation of an automated hematology system (Technicon H-1). Arch Pathol Lab Med. 1986;110:803-808; 5) NCCLS. Reference leukocyte differential count (proportional) and evaluation of instrumental methods; approved standard H20-A. Wayne, PA: NCCLS, 1991; 6) Miers MK, et al. White blood cell differentials as performed by the Technicon H-1; evaluation and implementation in a tertiary care hospital. Lab Med. 1991;22:99-106; 7) Hallawell R, et al. An evaluation of the Sysmex NE-8000 hematology analyzer. Am J Clin Pathol. 1991;96:594-601; 8) Cornbleet PJ, et al. Evaluation of the Cell-Dyn 3000 differential. Am J Clin Pathol. 1992;98:603-614; 9) NCCLS. Method comparison and bias estimation using patient samples; tentative guideline EP9-T. Wayne, PA: NCCLS, 1993; 10) Krause JR. The automated white blood cell differential. A current perspective. Hematol Oncol Clin North Am. 1994;8:605-16; 11) Goyzueta FG, et al. Automated differential white blood cell counts in the young pediatric population. Lab Med. 1996;27:48-52; 12) Gulati GL, et al. Suspect flags and regional flags on the Coulter-STKS. An assessment. Lab Med. 1999;30:675-680. > I can't help but wonder if someone got "lazy" on his CBC and chose not > to do a manual diff. when it should have been! Each site performs a validation study on site with that equipment. Come to our lab and expect to find an 80% cuttoff for manual diff and you will be laughed at. We do not do manual diffs or even make a smear with a patient who has a 14000 WBC count and there are no white cell flags and nobody cares about the percent granulocytes as a flag. The autodiff contains the granulocytes and diffs. It is a differential. They get a WBC and autodiff. That is based on validation studies at our site. I am curious to see what you would find in that smear, you being an ex hematology supervisor and the all the techs in your present bench are lazy? Please show me a published criteria that says manual diffs must be perfomed when granulocytes are over 80%. Let's see your hem supervisor is poor and the people who did that CBC above should have done a manual diff to see what? Please tell me what you are looking for in his blood smear with the above CBC results. > > I finally discussed the issue with the new Hem. supervisor. I advised > > her that if she and others chose to make "exceptions" to the rules, it [quoted text clipped - 7 lines] > your lab but that in now way is a CAP standard. This is from the CAP website > checklist. NO NO NO!!! You mis-understood me!!! Some of the tech's were deciding when and when not to do manual diff's based on the type of patient (most commonly labor and post-partum patients) and NOT following the written review criteria! ;-) THAT is a CAP violation, randomly not following written protocal! IF you are going to have exceptions to your written, documented criteria, then those exceptions must be decided on, approved by the lab director and documented in the procedure. > HEM.34200 Phase II N/A YES NO > [quoted text clipped - 8 lines] > instrument, inspection of histographic/cytographic displays, laboratory > criteria based on local experience, and awareness of published evaluations. Like I said, you obviously mis-read my commentary. Your laboratory is free to set your references as you see fit based on your population. Are other local hospitals with the same instrumentation set as loose? Whether I agree or disagree with your numbers is a personal opinion and I believe I stated it as such. ;-) <snipped references> What was your 90% value based upon? You must have the documentation for CAP inspectors in your implementation manual. You can't just randomly select numbers. I does not work that way. We based our 80% on a number of factors, AND it had to be approved by the hospital medical staff as well. We are not allowed to arbitrarily set numbers to reduce our workload. > > I can't help but wonder if someone got "lazy" on his CBC and chose not > > to do a manual diff. when it should have been! [quoted text clipped - 6 lines] > the granulocytes and diffs. It is a differential. They get a WBC and > autodiff. That is based on validation studies at our site. Whatever works for you. ;-) Just hope a defense lawyer never gets ahold of you in a lawsuit. > I am curious to see what you would find in that smear, you being an ex > hematology supervisor and the all the techs in your present bench are lazy? > Please show me a published criteria that says manual diffs must be perfomed > when granulocytes are over 80%. Let's just say that the majority of diff's that we do are positive, (we keep track of that to monitor over-flagging by the machine), so at this point, the 80% appears to be justified. Oh, and I was not fired as supervisor. For a number of complicated reasons, I resigned the position. Something to do with being forced to change shifts if I kept it, for a $7,000 per year paycut. :-P Lab politics. I was supervisor for over 12 years thru 3 different instruments. > Let's see your hem supervisor is poor and the people who did that CBC above > should have done a manual diff to see what? > Please tell me what you are looking for in his blood smear with the above > CBC results. Left shift and immature cells. They seem to be significant enough to our medical staff, and we are there to serve THEIR needs. Not our own. From the reviews I've done, we are only having a laziness problem with 3 out of the 20 people performing CBC's It's being dealt with. Cheers! SignatureK. Sprout the MungBean to reply "I don't like to commit myself about heaven and hell‹you see, I have friends in both places." --Mark Twain > > > I finally discussed the issue with the new Hem. supervisor. I advised > > > her that if she and others chose to make "exceptions" to the rules, it [quoted text clipped - 14 lines] > based on the type of patient (most commonly labor and post-partum > patients) and NOT following the written review criteria! ;-) So you assumed that the lab in this case did the same. I prefer to assume it was performed properly until proven otherwise. > THAT is a CAP violation, randomly not following written protocal! That is a lab problem that should be monitored by reviewing printouts. Most patients have multiple CBC's and deltas on diffs can be seen. > Like I said, you obviously mis-read my commentary. > > Your laboratory is free to set your references as you see fit based on > your population. Are other local hospitals with the same > instrumentation set as loose? We have a hospital chain so we have the same criteria. > Whether I agree or disagree with your numbers is a personal opinion and > I believe I stated it as such. ;-) [quoted text clipped - 5 lines] > implementation manual. You can't just randomly select numbers. > I does not work that way. Never said it wasn't validated but as I said I thought it was 90 and it might have been lower. > We based our 80% on a number of factors, AND it had to be approved by > the hospital medical staff as well. We are not allowed to arbitrarily > set numbers to reduce our workload. ??? If validation studies have been done to "validate" those results without flags etc that 85% granulocytes are in fact granulocytes than where is the problem? The medical staff is consulted but what does that mean with a large institution? Last I heard doctors were able to see a differntial and tell if it was a three part differential or a manual diff. They were and aren't prevented from ordering a manual diff in the first place. If they didn't like the autodiff they can order a manual one afterward. It happens all the time. I think you are over reacting on the cuttoffs. If it is valid and the medical chiefs agree and the doctors always have a right to use clinical judgements to overide, I don't see where you say that people are dying because of the 85 or 90% cuttoff. Just to let you know that bands, metas and myelos are granulocytes. They don't report out neutrophils but granulocytes. The abnormal valleys on the graph would flag any substantial immatures via validation studies. This is the instrument we are using and see what you think about flags. Notice they said that its performance is insufficient for detecting clinically relevant abnormal lymphocytes. Clin Lab Haematol. 2004 Feb;26(1):9-13. Related Articles, Books, LinkOut Diagnostic performance of the variant lymphocyte flag of the Abbott Cell-Dyn 4000 haematology analyser. Hoffmann JJ, Hoedemakers RM. Department of Clinical Laboratories, Catharina Hospital, Eindhoven, The Netherlands. hans.hoffmann@cze.nl BACKGROUND: In addition to differential cell counts, modern haematology analysers generate suspect flags if abnormal cells are detected. Reports on validation of suspect flags are scarce. We have routine experience with the Abbott Cell-Dyn 4000 analyser for over 5 years and have previously demonstrated the utility of the blast flag. Here we report a similar study on the performance of the analyser's Variant Lymphocyte (VL) flag. AIM OF THE STUDY: Evaluation of the diagnostic performance of the Cell-Dyn 4000 VL flag, as compared with lymphocyte morphology in blood smears. In addition, we investigated the usefulness of the numerical VL flag confidence index as provided by the analyser. MATERIALS AND METHODS: All samples generating a VL flag were reviewed over a 5-month period. We also reviewed smears from patients with known lymphoid disorders, even if the analyser did not flag the sample. Two experienced investigators assessed lymphocyte morphology independently. RESULTS: In total, 187 samples were included in the study, of which 183 had a VL flag and four had not. Of the 183 flagged samples, 83 appeared to have abnormal lymphocyte morphology and 100, normal lymphocyte morphology. The sensitivity of the VL flag for detecting abnormal lymphocytes was 0.95 and the positive predictive value was 0.44. Using ROC analysis of the VL flag confidence index, the area under the ROC curve was 0.58 (95% confidence interval 0.50-0.65). CONCLUSIONS: The Cell-Dyn VL flag has reasonable sensitivity but a high false-positive rate. In addition, its performance is insufficient for detecting clinically relevant abnormal lymphocytes. As the VL flag appeared to rely mainly on numerical criteria, it has no added value over numerical criteria defined by the laboratory. PMID: 14738431 [PubMed - indexed for MEDLINE] You have to be realistic on instruments and clinical interpretation. > > > I can't help but wonder if someone got "lazy" on his CBC and chose not > > > to do a manual diff. when it should have been! [quoted text clipped - 9 lines] > Whatever works for you. ;-) > Just hope a defense lawyer never gets ahold of you in a lawsuit. What would that be for? I just quoted the CAP requirements for you and each medical case is treated regionally based on the standards of care for that region. As a regional trauma center and teaching hospital that standard is not the same as in a 50 bed hospital out in the woods. I suspect that you have an attitude that because everybody doesn't it do it your way then they are lazy incompetent and can be sued for it. > > I am curious to see what you would find in that smear, you being an ex > > hematology supervisor and the all the techs in your present bench are lazy? [quoted text clipped - 4 lines] > keep track of that to monitor over-flagging by the machine), so at this > point, the 80% appears to be justified. You are seeing what in those cases in which you do diff? So what you are saying is that your machine is so insensitive that it can not pick up immature granulocytes on the autodiff without getting any other flag or abnormality such as H and H plt white count blast flag or anything that would flag you to look at a smear? that is way you have to rely on %'s I have found quite the opposite with the celldyns. The very high neutrophil percentage cases with no immature flags have in the overwhelming of cases been mature neutrophils. In about half of the cases with immature flags have had in fact immature white cells. Those are just approximations based on hundreds if diffs. > Oh, and I was not fired as supervisor. For a number of complicated > reasons, I resigned the position. [quoted text clipped - 11 lines] > > Left shift and immature cells. In a patient with a fever of a year and a half. A left shift. A normal CRP in which neonates have relied on for sepsis because immature granulocytes have a poor postivie predictor. I wonder if the infectious disease specialists that sees the granuloyctes, monocytes and lymphocytes will know that it is an auto diff and run to the hematology lab to have a manual diff done? It makes sense now considering your instrument does not pick up immature cells with your machine. OK use anything that works. > They seem to be significant enough to our medical staff, and we are > there to serve THEIR needs. I never said that immatures excluding bands were not signficant to be reported on diff. I am saying because of the poor reproduction nation wide and through clinical studies that reliance on the bands have taken them out of clinical relevance in evaluationg anything but in a few selected cases that not everyone in even those select cases agrees on. The article below goes into deal with studies involving every clinical situation bands have traditionally been used and the ROC on how well it did in that setting. Clinics in Laboratory Medicine vol 22 issue 1 pages 101-136 P Joanne Cornbleet Abstract The band count is a nonspecific, inaccurate, and imprecise laboratory test for bacterial infection. Review of the literature provides little support for the clinical utility of the band count in patients over 3 months of age. In the near future, rapid analysis of inflammatory factors, adhesion molecules, cytokines, neutrophil surface antigens, or even bacterial DNA may be superior alternative tests for the early diagnosis of sepsis. So why not do manual diffs instead of relying on instruments that do not report out immatures. I have yet to see an autodiff with bands, meta etc and you have admitted that the only way to pick up immatures is a manual diff? Don't you get nervous missing those bands? > Not our own. > [quoted text clipped - 4 lines] > > Cheers! in both places." --Mark Twain While I'm reading the discussions on Hematology methods between Katra and Robert, it is very interesting to see that Hematology has lots of different viewpoints, just like micro. I come from a very anal retentive lab background, and now am working in a lab where things are much more laissez faire with regards to interpretation. I think it's good, however, that I came from a "rules" background as it helps me make better decisions in this looser environment. With all due respect, I as a patient would like someone to look at my slide if I were running 80% granulocytes. :-) Small story: my mother-in-law, who is now deceased, presented at my little outpatient lab back in 1985 for her "yearly" blood work (yes they did that stuff back then). She hadn't been feeling that great, but otherwise was supposed to be in good health. I sent her CBC up to the main lab as I really didn't want to get involved in her reporting. Good thing I did. She had a "normal" (in the 8-9000 range) WBC count, but her absolute lymphocytes were extremely elevated. The hematologist looked at her slide (yes, it must have been flagged) and referred it to the pathologist. Diagnosis was probable early CLL, confirm with bone marrow examination. Of course, my mother-in-law called me for the results (back in pre-HIPAA days) and I hedged and said they weren't back yet. I got the printout and called her physician with the results. He reordered another count a week later that came back the same. She went on to have a bone marrow biopsy which confirmed the CLL diagnosis. She didn't really have lots of problems with it until around 1995, when her platelets went haywire (down) and she had to have open heart surgery to replace a damaged heart valve. Apparently it was damaged when she had had rheumatic fever in childhoos and just wore out. She had a massive stroke right after surgery, ended up in a nursing home, and passed away from a bowel obstruction in March, 1997. The point of this story is that flags and absolute counts in hematology are everything. With a normal WBC count, many labs without the advanced technology that was available at the "big lab" at the time would never have caught this diagnosis at such an early date. We owed a lot to the astute Hematology supervisor at our reference lab. I have no idea whether I would have caught this if I had done just a hemogram and manual diff at my little lab, as my limited hematology background would probably have missed the smudge cells and the "look" of the diff. Gotta go to work. Judy Dilworth, M.T. (ASCP) Microbiology Several points, actually, Not least ~ Privacy & Respect & Bereavement & Dignity. Back later, Gotta spin, Gotta shower, Gotta get outta here ... CRP, or peptide marker? See me? Hey ~ Please Hold my hand, Swim with me, and O, so much more ~ ! Come, lovely Capsicum ... > While I'm reading the discussions on Hematology methods between Katra > and Robert, it is very interesting to see that Hematology has lots of [quoted text clipped - 6 lines] > With all due respect, I as a patient would like someone to look at my > slide if I were running 80% granulocytes. :-) Include that in your protocol but don't assume everybody else is doing it and that the tech messed up or missed it because it wasn't done. That was the whole point I was making. It is the abosolute count that is important and not the relative percent that is important. You also have machines that have flags. If you don't trust the machine then there is no need in reporting out autodiffs or there is something wrong with the protocols or the doctor considers it important enough that a manual diff be performed. As far as who looks at blood smears, I have never seen in all my years in doing this an infectious disease specialist go look at a blood smear. If they thought that seeing bands or immatures are important they would look at the smear or order a manual diff regardless of what the diff shows. That doesn't happen. Clinical hematologist are always there and required to look at the blood smear regardless of what is reported on the differential by the tech. > Small story: my mother-in-law, who is now deceased, presented at my > little outpatient lab back in 1985 for her "yearly" blood work (yes they [quoted text clipped - 3 lines] > thing I did. She had a "normal" (in the 8-9000 range) WBC count, but her > absolute lymphocytes were extremely elevated. Again you mention absolute count and not relative percent so that is in the right direction. We do scan a slide with an absolute lymph count greater than 6 K or a variant lymph flag or a fragile cell percentage. The hematologist looked at > her slide (yes, it must have been flagged) and referred it to the > pathologist. Diagnosis was probable early CLL, confirm with bone marrow [quoted text clipped - 4 lines] > a bone marrow biopsy which confirmed the CLL diagnosis. She didn't > really have lots of problems with it until around 1995, They did nothing for her early on and the clinical implications would have been the same if it had been missed back then. Like any progression she probably had it for years before it was flagged. when her > platelets went haywire (down) and she had to have open heart surgery to > replace a damaged heart valve. Apparently it was damaged when she had > had rheumatic fever in childhoos and just wore out. She had a massive > stroke right after surgery, ended up in a nursing home, and passed away > from a bowel obstruction in March, 1997. Well sorry to hear about that but her clinical course was not effected by when it was discovered she had CLL. > The point of this story is that flags and absolute counts in hematology > are everything. Not everything as you can never catch all the abnormals based on above. It is not sensitive enough as I mentioned all hematologist look at smears for very minute clues not mentioned on reports. Blood smear interpretation is different from doing and reporting a differential. With a normal WBC count, many labs without the advanced > technology that was available at the "big lab" at the time would never > have caught this diagnosis at such an early date. You said it was the absolute lymphs that sent it in for review so it wasn't really the technology. If it had been done by a machine without a autodiff and a manual diff performed then it would still have the same abnormal morphology and sent in for review. The treatment protocols for CLL are very poor and the course is indolent so in the case of CLL the early aspects of diagnosis does not apply. Treatment guidelines require A and B symptoms some of which include parameters of the CBC. We owed a lot to the > astute Hematology supervisor at our reference lab. I have no idea > whether I would have caught this if I had done just a hemogram and > manual diff at my little lab, as my limited hematology background would > probably have missed the smudge cells and the "look" of the diff. In the case of CLL I think you worried needlessly as the platelet count etc would have manifested itself and treatment rendered then. Good luck > Gotta go to work. > > Judy Dilworth, M.T. (ASCP) > Microbiology > > While I'm reading the discussions on Hematology methods between Katra > > and Robert, it is very interesting to see that Hematology has lots of [quoted text clipped - 74 lines] > guidelines require A and B symptoms some of which include parameters of the > CBC. <lol> But the total WBC count was normal... Ok, try this one Robert. WBC count, 4.5 Neutrophil % 45% Lymphocyte % 50% Monos 5% Patient age: 27 Presents with high fever, vomiting and diarrhea. Based on the numbers, would you have done a diff? Machine was too primitive at the time to flag immature/band cells I'll tell you her diagnosis after I see your answer. ;-) > >We owed a lot to the > > astute Hematology supervisor at our reference lab. I have no idea [quoted text clipped - 5 lines] > would have manifested itself and treatment rendered then. > Good luck Later. MUCH later. Usually, the earlier CLL gets treated, the better chances of recovery. But, since CLL can often go on for years without showing symptoms, elderly patients are often left untreated. I once saw a 92 year old in the ER that had a CLL. Her WBC cound was running around 12,000. I was told that yes, she had a history of it and she had not even been told! They felt that old age would kill her before the CLL would. > > Gotta go to work. > > > > Judy Dilworth, M.T. (ASCP) > > Microbiology > SignatureK. Sprout the MungBean to reply "I don't like to commit myself about heaven and hell‹you see, I have friends in both places." --Mark Twain > > > While I'm reading the discussions on Hematology methods between Katra > > > and Robert, it is very interesting to see that Hematology has lots of [quoted text clipped - 76 lines] > > <lol> But the total WBC count was normal... As Judy mentioned her abosolute lymphs were high. I think you are over thinking there. Doctors are able to distinguish between all the CBC results. Do you want to be diagnosed early with CLL? Not me thank you as they won't do anything different. > Ok, try this one Robert. > [quoted text clipped - 6 lines] > > Presents with high fever, vomiting and diarrhea. I don't really make decisions on who gets a smear review based on symptoms etc. There are procotols established that people follow. People should not assume that those protocols are universal. What one lab scans another might not. What others consider as signficant such as bands there are others who do not. I report out bands in our lab but other labs do not. Some labs report out percentage of reactive lymphs and others only use few mod etc. Some call them reactive lymphs other places call them atypical lymphs. Some labs count mesothelial cells in their WBC count cell counts and others do not. So to answer your question I would follow the protocols of that lab and expect the doctors who are familiar with that lab to evaluate all the results. You age going under the assumption that doctors do not know how to interpret a diff. > Based on the numbers, would you have done a diff? It is never decided that way. We do not make individual decisions on cuttoff. IF it meets the criteria then it will have a scan. > Machine was too primitive at the time to flag immature/band cells Not important with the same answer. > I'll tell you her diagnosis after I see your answer. ;-) We don't make the diagnosis only perform the laboratory testing according to accepted practices. I have worked in many laboratories and each has pretty much differing procedures for everything. I basicly follow procedures. > > >We owed a lot to the > > > astute Hematology supervisor at our reference lab. I have no idea [quoted text clipped - 7 lines] > > Later. MUCH later. And treatment usually follows later MUCH later. > Usually, the earlier CLL gets treated, the better chances of recovery. Recovery? Oh really. Why don't we ask Judy if she was treated early? > But, since CLL can often go on for years without showing symptoms, > elderly patients are often left untreated. ?? Most are elderly and treatment rarely works. I once saw a 92 year old in > the ER that had a CLL. Her WBC cound was running around 12,000. > > I was told that yes, she had a history of it and she had not even been > told! They felt that old age would kill her before the CLL would. Twain When my mother-in-law had her bone marrow examination following two CBC's that indicated that she probably had early CLL, I don't believe much was done right away. I know she had to have follow-ups with the hematologist, and eventually went to see an oncologist. Her WBC eventually went up above 50,000, and her platelets started dropping. They had her on some sort of chemotherapy in the 90's, which helped. Everything went downhill when her heart valve gave out. In order for her to have the valve replacement surgery, they had to give her platelet transfusions. Then, when they were weaning her off heparin and onto Coumadin, in preparation for sending her home, she stroked out. The doctor had just been in her room, and the nurse came in and it happened. She was in the step-down unit. They rushed her to Neuro ICU. It took her a couple of days to sort of come around, but she was never the same. She was paralyzed on the right side and couldn't talk any more. I know that CLL was a factor in this, as her blood picture was just weird enough to have these clots happen. I think if she hadn't had CLL, she wouldn't have needed the platelet transfusions ahead of time (it was touch and go whether she could even have the surgery). As I recall, her original white count was in the 8.5-9.0 range, and her lymph percentage was elevated - sort of like a kid's differential. On first glance, it could look like a post viral infection diff. The key to it, as it was explained to me, was the absolutely lymphocyte elevation, along with the smudge cells. At that time I believe her platelets were in the low normal range - nothing to get excited about. The hematologist estimated that she had probably had the condition for at least two years prior to the CBC being done. That would put her onset at around 1983. She lived until 1997 and died at age 73. The surgeons would not operate on her bowel obstruction due to the CLL, I think, as she was still having platelet problems then. She was out in Arizona by that time, at my brother-in-laws house (they agreed to have her live with them), so we're not sure exactly what happened. Her quality of life was poor, but it was due to the stroke. I'm sure interpretation of a blood smear is something learned over many years. I am in charge of our gram stain proficiency testing, and trying to get generalists to read and interpret smears correctly is quite a challenge. There are many nuances you learn over many years, and you only learn them by looking at thousands of slides. I think you, Robert, and you, Katra, both have valid arguments for doing what you're doing. Remember, a lot of what is done in laboratories is determined by the pathologists in charge of the departments, and also the type of patient seen. Besides those parameters, there is the instrument, the technologists involved, and the workload, and how many FTE's are allotted to the department. If you got ten hematology supervisors in a circle that represented ten different labs (from large to medium to small, mostly outpatient to mostly inpatient) you would probably end up with a LOT of different interpretations as to when diffs should be read manually vs. staying with the machine produced results. I think a lot of it is just your gut feeling. Sometimes you just have to go with that, regardless of numerical results. That's why experience is so important, and why we must pass all this on to our younger counterparts. Hopefully, there will be more counterparts soon :-). Judy Dilworth, M.T. (ASCP) Microbiology <snipped> > I'm sure interpretation of a blood smear is something learned over many > years. I am in charge of our gram stain proficiency testing, and trying > to get generalists to read and interpret smears correctly is quite a > challenge. There are many nuances you learn over many years, and you > only learn them by looking at thousands of slides. Oh yeah... ;-) One thing about our lab, section supervisors still perform as generalists. I get to read gram stains too. <lol> And I'm still learning more after 18 years! > I think you, Robert, and you, Katra, both have valid arguments for doing > what you're doing. Remember, a lot of what is done in laboratories is [quoted text clipped - 12 lines] > must pass all this on to our younger counterparts. Hopefully, there will > be more counterparts soon :-). Excellent observation, and I concur! Unfortunately, there are also duffers in every field. I'm tempted to start a new thread on "laboratory horror stories". What is the worst thing you've ever seen happen in a lab you worked in due to tech negligence, carelessness or just plain over-worked staff? I've got at least a couple of doozies. :-P "Sink testing" is the greatest of sins....... > Judy Dilworth, M.T. (ASCP) > Microbiology Cheers! ;-) SignatureK. Sprout the MungBean to reply "I don't like to commit myself about heaven and hell‹you see, I have friends in both places." --Mark Twain > When my mother-in-law had her bone marrow examination following two > CBC's that indicated that she probably had early CLL, I don't believe > much was done right away. I know she had to have follow-ups with the > hematologist, and eventually went to see an oncologist. Her WBC > eventually went up above 50,000, and her platelets started dropping. > They had her on some sort of chemotherapy in the 90's, which helped. CLL is a disorder of apoptosis in that cells don't die when they should so the count increases not by rapidly dividing cells that are most sensitive to chemotherapy and radiation but because of the longivity of the cells. New agents are directed at inducing apoptosis. > Everything went downhill when her heart valve gave out. In order for her > to have the valve replacement surgery, they had to give her platelet [quoted text clipped - 3 lines] > She was in the step-down unit. They rushed her to Neuro ICU. It took her > a couple of days to sort of come around, but she was never the same. She Heart valve replacement involves the use of animal porcine valves in which the person does not need coumadin or by artifical valves in which they need to take coumadin for the rest of their lives. Porcine valves do not last as long as artivical valves so it is a hard decision to make as to which one to pick. > was paralyzed on the right side and couldn't talk any more. > > I know that CLL was a factor in this, as her blood picture was just > weird enough to have these clots happen. I think if she hadn't had CLL, > she wouldn't have needed the platelet transfusions ahead of time (it was > touch and go whether she could even have the surgery). That's true and now they have special tests that can be used to predict aspirin resistence and COX inhibitors resistence such as plavix in preventing some cases of stoke and heart attacks. People are taking these drugs and not getting the benefits of the antiplatelet function. > As I recall, her original white count was in the 8.5-9.0 range, and her > lymph percentage was elevated - sort of like a kid's differential. On > first glance, it could look like a post viral infection diff. The typical morphology of CLL cells are pretty unique even without the smudge cells. The problem is the number there as they will be mixed with normal lymphs at an early presentation. The problem as you state is that there are cases in which a lymphocytosis is present without the pleomorphic polyclonal morphology of a reactive viral lymphocytosis. I avoid the term reactive lymphs or atypical lymphs when reporting the diff in that case and let the pathogist make the decision on the wording. They ususally put reactive lymphocytosis or simply lymphocytosis present. Downy cells are more typical of viral infections or reactive lymphs. The key to > it, as it was explained to me, was the absolutely lymphocyte elevation, > along with the smudge cells. Smudge are not specific for CLL and can be seen in atypical lymphs, blasts and in CLL or even in neutrophils in fatty smears. The key is to get rid of them to see what they are. At that time I believe her platelets were > in the low normal range - nothing to get excited about. The hematologist > estimated that she had probably had the condition for at least two years > prior to the CBC being done. It takes years to get it into the bone marrow for it to reduce platelets. I would guess more than two years. That would put her onset at around 1983. > She lived until 1997 and died at age 73. The surgeons would not operate > on her bowel obstruction due to the CLL, I think, as she was still > having platelet problems then. She was out in Arizona by that time, at > my brother-in-laws house (they agreed to have her live with them), so > we're not sure exactly what happened. Her quality of life was poor, but > it was due to the stroke. Some conditons are still best dealt with at home and modern medicine lags. > I'm sure interpretation of a blood smear is something learned over many > years. I am in charge of our gram stain proficiency testing, and trying > to get generalists to read and interpret smears correctly is quite a > challenge. There are many nuances you learn over many years, and you > only learn them by looking at thousands of slides. That is correct and there is no substitute for experience. Both are similar in that you are looking at minute differences. > I think you, Robert, and you, Katra, both have valid arguments for doing > what you're doing. Remember, a lot of what is done in laboratories is [quoted text clipped - 9 lines] > with the machine produced results. I think a lot of it is just your gut > feeling. We do have 6 hospitals in our system and the supervisors for chemistry or hem or blood bank get together in their own meetings to make policies and procedures. They are not uniform because of the medical staffs etc as you mention above. Sometimes you just have to go with that, regardless of > numerical results. That's why experience is so important, and why we > must pass all this on to our younger counterparts. Hopefully, there will > be more counterparts soon :-). > > Judy Dilworth, M.T. (ASCP) > Microbiology We have specific pathologist review criteria apart from the smear criteria but as you said that Instrument makers market their products as accurate and we validate that accuracy in house. Coming from a background of doing 200 diffs a day department in the early years to where it is now is not really a problem. I don't need an autodiff to tell me what I am looking at. I do need an autodiff to tell me what not to look at. That is a fine tuning process but it takes a few minutes. The question of which valve to put in came up before my mother-in-law's surgery. They decided on the porcine valve, even though it wouldn't last as long, and she knew this going in. I think it was because they didn't want her to have to take Coumadin along with having CLL, but I can't be sure. Her platelets were unstable at that point. They had to do the valve replacement, however, as her mitral valve was in very bad shape, and there was no way she could live with the valve in the condition that she was in - the veritable rock and hard place scenario. She had collapsed in the back yard and was having atrial fibrillation - a condition that did not go away after the replacement, however. Thanks for the in-depth on CLL. Micro is my department, and I know e |
Enable EMail Alerts
Start New Thread
Reply to this Message