In my previous blog post on cytology quality improvement, you and I had an assignment to read this article, Arch Pathol Lab Med. 2011 Nov;135(11):1387-90, Quality improvement in cytology: where do we go from here? by Andrew Renshaw, MD. The time has come to followup on this initial post and discuss the article and provide some conclusions.
Objective: Cytology Quality Improvement
"...To review the ways to use new information to improve the quality of cytology."
Review and Discussion
The article begins with a statement about the high level of cytology quality currently found among various practice settings, followed by a question:
"...Given this success, one may ask, how can cytology improve on its current level of quality, or where should we go from here?"
The author admits that this is a "limited review"...and on that point, I definitely agree...it just scratches the surface. The topics discussed in the article which the author describes as "new information" aren't necessarily "new", but they are certainly changing with the times, ie. cytology-histology correlation, the electronic medical record (EMR), monitoring of gyn-cytology workload reporting, correlating prior aspirate history in thyroid aspirate evaluations, and correlation of molecular studies with urine cytology.
On this topic, Renshaw suggests that over time, as access to clinical history has improved (in theory) due to EMRs, and as patient involvement in their care has increased, the need for "real[-]time" review of cytology-histology correlations and 5 year look-backs is apparent, with more prompt correction of variances. The correlation is often discussed in the biopsy report. Interestingly, with respect to "new" information, I think something that Renshaw did not discuss, but probably could have been included here is to include the correlation with HPV testing results as well. This would be particularly important in the event of negative cervical cytology/+hr-HPV followed by biopsy. An improvement in cytology quality could incorporate reviewing those cytology cases which were found to have a SIL+ lesion, to determine whether the cytology was a true negative or a false negative. These metrics would lend themselves well to monitoring. And in fact, this has just recently been published in Diagnostic Cytopathology and in J Cell Mol Med (and you thought your homework was over!). It would also be interesting to compare these numbers with those described here, to see what the breakdown of biopsy followup shows.
Before the EMR, there was the card file...yup, I am old enough to remember that...and I bet there are quite a few of you out there who do as well. Renshaw discusses the unclear nature of the amount of sleuthing that the cytologist or pathologist should undertake to assure they have all relevant clinical and past cytology history. The reality is, that this is likely an impossible task until the time where we have a universal data repository of patient information that every lab in the country would be hooked up to, and have the ability to search prior history performed anywhere. The government is definitely moving in this direction, with the emphasis on the EMR, and Google I am sure, will be in the loop. Next up will be a unique patient identifier in order to identify the patient in the country wide system. There is not sufficient man-power to perform such a task on every pap test/biopsy received in the laboratory, and the clinician's office will not be very happy with the number of calls and requests into their offices for that information on a daily basis. So, at this time, a reasonable-effort should be made, as well as client education to provide proper history with the sample. LIS modifications to provide the history-file electronically at time of sign out would be highly desirable, and is a component of most new LIS systems on the market. The question of PHI does come up with respect to access to clinical history in this new environment, and will likely be further addressed in the future.
Gynecologic Cytology-How Fast? (or How Many?)
In Renshaw's article, he alludes to suggested methods to assess cytology screening sensitivity, specifically, the Epithelial Cell Abnormality (ECA)-adjusted workload (ECA rate x workload) for automated screening. The author recently published an article using this methodology here. In addition, the suggestion in this abbreviated review of cytology quality, alluded to the 70 slides/day average laboratory maximum cytotechnologist workload article also recently published. It remains to be determined what effect this may or may not have on cytology practices and screening quality.
Thyroid Fine-Needle Aspiration and Prior Aspirate History
A huge step in the improvement of thyroid cytology quality was the advent of TBSRTC, which I recently discussed here and here. Not only does this system attempt to standardize terminology, it also provides an assessment of risk for malignancy based on diagnostic category that can assist with individualized patient management. Renshaw discusses incorporation of prior aspirate results in the FNA report, which can be somewhat problematic when the repeat is benign. Not mentioned in the article is the importance of communication between the cytopathologist and the clinician, which in these discordant scenarios could be important. Certainly, this depends on the desires and relationship between the clinician and pathologist...believe it or not, no two are alike! Unfortunately, this article preceded several articles recently published in Cancer Cytopathology which discuss suggested thyroid cytology quality metrics, which are summarized here. Expected rates for diagnostic categories are discussed as well as other proposed metrics, such as the AUS/FLUS:M ratio. Even though TBSRTC is relatively new, it is fairly easy to recategorize historical data as needed in order to obtain useful comparisons.
Urine Cytology and Molecular Correlations
I agree with the author that many cytologists find urine cytology challenging. The exfoliated cells are suspended in a toxic environment that definitely affects morphology. Atypical cells with degenerative changes can often be identified in this toxic environment, and the ability to reliably distinguish these degenerative changes in normal cells from degenerating malignant cells is very challenging. The advent of UroVysion testing has assisted in improving the sensitivity of bladder cancer screening compared to cytology, although at much greater cost. This expensive test also suffers from lack of standardization in methodology, and in particular patients, specificity of the test. As a result, there is a trend towards providing a correlative report combining both methodologies, in contrast to using either test independently in the appropriate situations. The author raises the suggestion of a consensus conference to evaluate performance data because of the difficulty in defining atypia and in standardization of UroVysion, similar to TBS for gynecologic and thyroid cytology. Inter-laboratory surveys have been discussed by others as a means to define UroVysion precision. I would guess, that since several items referenced earlier in the article, have since come to pass, that this is very likely...maybe just another hint of what's to come? So until the advent of inter-laboratory surveys, or a new Bethesda Conference on Urine Cytology, what quality tools should be utilized to improve quality? Although not mentioned, it would make sense to monitor those cases with both cytology and molecular testing, and follow the results of each, tracking when concordant or discordant, and obtaining biopsy followup to assist with determining test performance.
What About the Breast?
Although the author admittedly indicated that this was a "limited review", there was not discussion regarding the possible use of new molecular techniques on FNAC samples of breast. Certainly, this is a really BIG deal with tissue biopsy sampling, so it should be a similar BIG deal with FNAC samples. Information on quality measures for HER-2 testing can be found on the CAP website. You may find my previous blog posts on potential uses of FNAC in breast of interest, regarding HER-2 testing and IHC staining.
I think Renshaw's brief article represents a nice introduction to the topic of cytology quality improvement. The message to any cytologist is "adaptability". Times are a changin', and with increasing frequency. Staying up to date with the latest technological advances, and how to utilize them to improve the quality of care provided to patients is the unique challenge facing all of us in the field. Our level of cytology quality remains high, yet we need to be able to fine tune our processes. Quality assessment and monitoring includes pre-analytical, analytical and post-analytical factors. We can offer many cytology quality improvements that can affect patient satisfaction, and can affect personalized patient management, while at the same time even enhancing diagnostic accuracy. The field of cytology has significant opportunity as less invasive methods with new molecular testing may represent a cost-effective alternative to more aggressive procedures. Hmmm....is it just me, or doesn't that seem like what cytology was originally supposed to be about? Just ask the Europeans.