An Early Birthday Present…

31 03 2009

… and just like that, I’ve gotten the go-ahead on a virtual slide box for Albany Medical Center.

So here’s the story: a current faculty member at the college received a grant some time ago to purchase a virtual slide box solution.  He subsequently chose NeuroInformatica from MBF Bioscience (jury’s still out on how good of a solution it is, but I’m up for anything right now), but he never really used it.  Now, he’s giving my department the opportunity to beef up the service.  I couldn’t be more up for the challenge; this is exactly the kind of thing I am into, and fortunately for me, I’m off-service for the next month, so hopefully I will have the opportunity to sit down and start tinkering with it to see what it can do.  Awesome.

Here’s the URL for the site:

I haven’t touched anything yet, but I’ll be given admin rights tomorrow, hopefully… Off to the races!!!


Abstract: Overexpression of Notch-1 Correlates with Grade, Stage and Overall Survival in Colorectal Adenocarcinoma (CRC)

31 03 2009

KA Robstad, JD Choate, MN Donovan, MA DiMaio, CE Sheehan and JS Ross
Department of Pathology, Albany Medical College, Albany, NY

Background: Notch signaling is believed to play a crucial role in cell differentiation, proliferation and survival. Dysregulation of Notch signaling has been reported in a wide variety of human malignancies including breast, cervical, CNS, lung, and pancreatic cancer as well as melanoma and certain leukemias. There is recent evidence that Notch signaling contributes to the natural progression of CRC. However, the prognostic significance of Notch-1 expression in CRC has not been previously investigated.

Design: Formalin-fixed, paraffin-embedded sections from 125 colorectal adenocarcinomas (CRCs) were immunostained by an automated method (Ventana Medical Systems; Tucson, AZ) using polyclonal Notch-1 antibody (sc-6014; Santa Cruz Biotechnology, Santa Cruz, CA). Cytoplasmic immunoreactivity was semiquantitatively evaluated based on both intensity and distribution and results were correlated with histologic and prognostic variables.

Result: Intense, diffuse overexpression of Notch-1 was observed in 45% (57/125) of CRC cases and correlated significantly with increasing AJCC stage (24% of stage I, 58% of stage II, 35% of stage III, and 66% of stage IV; p=0.041); histologic grade (11% of grade 1; 52% of grade 2, and 50% of grade 3; p=0.006), and overall survival (27% in those alive, 53% in those expired; p=0.011). On multivariate analysis, only pathologic stage was an independent predictor of overall survival.

Conclusion: Notch-1 overexpression is associated with tumor aggressiveness in CRC and significantly correlates with increasing tumor grade, pathologic stage and overall survival. Notch 1 overexpression may be a valuable prognostic indicator that can be used to plan therapy in CRC. Further study of Notch-1 expression in CRC management and targeted therapy development appears warranted.


Poster View:

Official 2009 USCAP Poster Submission - K. Robstad

Official 2009 USCAP Poster Submission - K. Robstad

Abstract: Evaluation of Sentinel Lymph Node Intraoperative Frozen Section Diagnoses In Breast Cancer Patients

31 03 2009

Robstad, K. Department of Pathology. Albany Medical Center. Albany, NY

Objective: To evaluate the effectiveness of intraoperative frozen section diagnosis at appropriately staging patients undergoing breast cancer surgery at Albany Medical Center, and to compare these data to existing published reports.

Background: It has been established that selective axillary dissection guided by lymph node scintillography can reduce the morbidity associated with a full axillary lymph node dissection. The safety of this approach, however, is contingent upon accurate staging of the patient intraoperatively. False negative reporting may cause morbidity for the patient including additional surgery or disease recurrence.

Methods: A database search of pathology reports from 2003 to 2008 containing the keywords, “breast” and “sentinel node” was conducted using the SoftPath software suite. Diagnosis at the time of frozen section was compared to the final reported diagnosis. From this information, accuracies, sensitivities, and negative predictive values were calculated overall and by year for both cases as a whole and by total number of recovered sentinel lymph nodes. This data was then compared with figures reported in current literature.

Results: Between 2003 and 2008, there were 581 surgical cases in which intraoperative consultation was obtained regarding sentinel lymph node involvement by breast cancer; in these cases, a total of 1,647 lymph nodes were evaluated. The overall diagnostic accuracy at time of frozen was 96.28%. The overall sensitivity of detecting macrometastases was 80.34%. The negative predictive value was 95.62%. Data by year is reported in figures one through three.

Discussion: The data from the researched reports compares favorably with previously-published figures. Langer et al (2008) and Van de Vrande et al (2008) both report an accuracy of 90% compared with 96% at Albany Medical Center (AMC). Van de Vrande also documented a sensitivity of 84% in identifying macrometastases which compared to AMC’s 80% value. AlSahaf et al (2008) and Ali et al (2008) both report a negative predictive value of 97% which compares to AMC’s 95.6% value. Micrometastases were excluded from analysis because they have been found to have no impact on a patient’s prognosis; there were, however, 66 instances in which a case was reported as negative for tumor on frozen section that was  later designated to have involvement by either a focus of cancer measuring less than 0.2mm or isolated tumor cells (resulting in a N0i+ designation); only 23 cases were reported at AMC in which permanent sectioning revealed sentinel node involvement greater than 0.2mm which was not seen on frozen sections.

It is of note, that most published data compares frozen section diagnosis to the outcome of subsequent total lymph node dissection and not subsequent permanent fixation of the same specimen evaluated using frozen sectioning. Also, most existing publications fail to delineate the difference and micro- and macrometastases, making interpretation of existing data a complicated process. Finally, it is noteworthy that 12% of AMC case reports do not explicitly include a measurement of carcinoma involvement in the lymph node when not observed during intraoperative consultation.



































































Table 1. Accuracies, sensitivities and negative predictive values for overall cases and for individual nodes from 2003 to 2008.



1.       Ali R, Hanly AM, Naughton P, Castineira CF, Landers R, Cahill RA, Watson RG. Intraoperative frozen section assessment of sentinel lymph nodes in the operative management of women with symptomatic breast cancer.  World J Surg Oncol. 2008 Jun 26;6:69.

2.       AlSahaf M, AlShaban B, Mulsow J, Power C, Leen E, Walsh TN. Intra-operative examination of the sentinel node in breast cancer.  Ir Med J. 2008 Apr;101(4):120-2.

3.      Davit FE, Gatmaitan P, Garguilo G. Sentinel node mapping for breast cancer: the operative experience of a breast surgeon in a rural community.

4.       Holwitt DM, Gillanders WE, Aft RL, Eberlein TJ, Margenthaler JA. Sentinel lymph node biopsy in patients with multicentric/multifocal breast cancer: low false-negative rate and lack of axillary recurrence.  Am J Surg. 2008 Oct;196(4):562-5. Epub 2008 Aug 29.

5.       Langer I, Guller U, Berclaz G, Koechli OR, Moch H, Schaer G, Fehr MK, Hess T, Oertli D, Bronz L, Schnarwyler B, Wight E, Uehlinger U, Infanger E, Burger D, Zuber M. Accuracy of frozen section of sentinel lymph nodes: a prospective analysis of 659 breast cancer patients of the Swiss multicenter study.  Breast Cancer Res Treat. 2009 Jan;113(1):129-36. Epub 2008 Feb 23.

6.       Van de Vrande S, Meijer J, Rijnders A, Klinkenbijl JH. The value of intraoperative frozen section examination of sentinel lymph nodes in breast cancer. Eur J Surg Oncol. 2008 Sep 9.

7.       Van Wely BJ, Smidt ML, de Kievit IM, Wauters CA, Strobbe LJ. False-negative sentinel lymph node biopsy. Br J Surg. 2008 Nov;95(11):1352-5.

8.     Varga Z, Rageth C, Saurenmann E, Honegger C, von Orelli S, Fehr M, Fink D, Seifert B, Moch H, Caduff R. Loss of metastatic deposits in breast sentinel lymph nodes during intra-operative frozen section analysis.  Verh Dtsch Ges Pathol. 2007;91:221-4.

When technology fails, is it your fault?

31 03 2009

ZDNet has an interesting article about whether or not med-tech companies should be held liable if their product fails, causing injury to a patient.  It’s a very interesting question, and it seems as if a precident had been set in the 2008 Riegel vs. Medtronic case which barred lawsuits from ‘challenging the safety or effectiveness of a medical device,’ as long as the device is marketed in a form that received premarket approval from the FDA.”

So does this apply to pathology?  Should it?  If it does, how would it?

In my narrow experience, my initial impression is that it seems like really murky waters. First, you have to consider the approval portion the ruling.  There is FDA approval for many laboratory test, but others, done in-house, may not be approved by the FDA.  As per protocol, reports generated from these results are mandated to indicate in writing that the test is not FDA approved, and has been developed and validated in-house.

Which brings me to my next point: validation.  

Each laboratory needs to validate their testing procedures.  So, if an in-house validated, non-FDA approved test fails or malfunctions, does the doctor or the hospital become liable for damages? Well, I guess I’d draw a parallel to the conventional practice of medicine, sans technology.  If the risk of a certain surgery is death, and this is well documented, and then the patient dies, then there is no malpractice, no deviation from the standard of practice that directly harms the patient occurred. So the same should be true of medical technology.  Can one expect all hardware and software to be 100% accurate all the time?  I suppose one could make the argument that machines are more consistent than humans, but you, and I both know that nothing is 100%.  This is one of those murky areas – if the specificity of a test is 99.999% and there is ever a false negative should the company that developed the test be held liable?  Personally, I’d have to say no, that is ridiculous.

Or take an example from our lab- for IHC, we used slides with a red-painted stripe to separate the control from the test sample.  For whatever reason, recently the antibody would stain the control but not roll over to the sample.  I suppose this is the fault of the lab; or is it the slide maker?  Or is it the automated stainer?  Or should the pathologist recognize a negative stain when the control works (assuming there’s no internal control in the sample).  If the slide and automated stainer have been validated to work with each other, who is to blame; or is there no blame, no fault?  I suppose I would have to put my foot down on my last supposition.

Nobody and no thing is perfect. Sometimes machines fail, and sometimes people fail; While we strive for perfection, we never will be. Without documentable negligence you have to be let off the hook, whether you are a doctor or the CEO of Medtronic.


If I am misguided, let me know. Also, please share your own opinion on the matter; perhaps I’m preaching to the choir, perhaps not…


[ZD-Net Health via Medgadget]

Other Medical & Pathology Blogs

23 03 2009

Here’s a list of the sites I get my information/inspiration from:


Doctors can have a sense of humor…

22 03 2009

Below is a collection of some of my favorite doctor/med school-related YouTube videos. Some of them are my own, others I’ve found elsewhere (including Respectful Insolence).

Sclerosing Mesenteritis



Indian Pathologists (from our Grand Rounds)


Feel free to add your own in the comment sections below…

USCAP 2009: BioImagene does it right.

22 03 2009

(First off, this post is not paid for or endorsed by BioImagene, USCAP or anyone for that matter. It is but a collection of my thoughts!)

I have to admit, before I went to USCAP this year, I had never heard of the company before, but, after the conference, I’ll never forget them. It’s funny, actually; I was introduced to the lead pathologist of BioImagene the night before their sponsored Informatics lecture, and I didn’t really know how big of a deal it was until the next day when I sat for their lecture and saw their booth at the show.

They are like the Apple of pathology. Clever ideas and fanfare. I loved it. Check out the videos below for an idea of what I’m talking about. My only problem with them is that they totally stole my ideas… Good on you BioImagene… but also thank you, because this is EXACTLY what I want to do with my career. If only I will be so lucky some day! Maybe some day I will invent the iMicroscope, lol…


Announcing the new Fluro Scope


Touchscreen Nav on an HP Touchsmart ‘Kitchen’ PC (where’s multitouch!?)


Actual ‘Pushing Glass’ Prototype

%d bloggers like this: