The effectiveness of inking needle core prostate biopsies for preventing patient specimen identification errors: a technique to address Joint Commission patient safety goals in specialty laboratories.
* Context.--The elimination or reduction of medical errors has been
a main focus of health care enterprises in the United States since the
year 2000. Elimination of errors in patient and specimen identification
is a key component of this focus and is the number one goal in the Joint
Commission's 2008 National Patient Safety Goals Laboratory Services
Objective.--To evaluate the effectiveness of using permanent inks to maintain specimen identity in sequentially submitted prostate needle biopsies.
Design.--For a 12-month period, a grossing technician stained each prostate core with permanent ink developed for inking of pathology specimens. A different color was used for each patient, with all the prostate cores from all vials for a particular patient inked with the same color. Five colors were used sequentially: green, blue, yellow, orange, and black. The ink was diluted with distilled water to a consistency that allowed application of a thin, uniform coating of ink along the edges of the prostate core. The time required to ink patient specimens comprising different numbers of vials and prostate biopsies was timed. The number and type of inked specimen discrepancies were evaluated.
Results.--The identified discrepancy rate for prostate biopsy patients was 0.13%. The discrepancy rate in terms of total number of prostate blocks was 0.014%. Diluted inks adhered to biopsy contours throughout tissue processing.
The tissue showed no untoward reactions to the inks. Inking did not affect staining (histochemical or immunohistochemical) or pathologic evaluation. On average, inking prostate needle biopsies increases grossing time by 20%.
Conclusions.--Inking of all prostate core biopsies with colored inks, in sequential order, is an aid in maintaining specimen identity. It is a simple and effective method of addressing Joint Commission patient safety goals by maintaining specimen identity during processing of similar types of gross specimens. This technique may be applicable in other specialty laboratories and high-volume laboratories, where many similar tissue specimens are processed.
(Arch Pathol Lab Med. 2009;133:295-297)
Prostate cancer (Diagnosis)
Biopsy, Needle (Usage)
Biopsy, Needle (Methods)
Raff, Lester J.
Beck, Kenneth R.
O'Brien, Andrea S.
Bauer, Meagan E.
|Publication:||Name: Archives of Pathology & Laboratory Medicine Publisher: College of American Pathologists Audience: Academic; Professional Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2009 College of American Pathologists ISSN: 1543-2165|
|Issue:||Date: Feb, 2009 Source Volume: 133 Source Issue: 2|
|Geographic:||Geographic Scope: United States Geographic Code: 1USA United States|
The elimination or reduction of medical errors has been a main
focus of health care enterprises in the United States since the
publication of To Err Is Human: Building a Safer Health Care System by
the Institute of Medicine in 2000. (1) Makary et al (2) recently
quantified the "incidence and type of specimen identification
errors in a surgical patient population." Elimination of errors in
patient and specimen identification is a key component of this focus and
is the number one goal in the Joint Commission's 2008 National
Patient Safety Goals Laboratory Services Program. (3)
The processing of anatomic specimens within the pathology laboratory is under the direct control of the pathologist, and all necessary steps must be taken to ensure that specimens are not inadvertently transferred from one patient to another. Good pathology practice suggests that similar specimen types should not be processed in sequential order. However, laboratories such as ours that specialize in a certain type of specimen, such as prostatic needle biopsies, have no choice but to process grossly and microscopically similar specimens sequentially. We noted a report by Renshaw et al (4) on their experience in inking breast biopsy core specimens to avoid specimen mix-up.
They report a 0.3% discrepancy rate but do not indicate if their breast biopsies were processed sequentially or as part of a general surgical pathology load. We report our findings in inking prostate core biopsies in a laboratory where 85% of all specimen vials processed are of prostatic tissue.
MATERIALS AND METHODS
UroPartners LLC Laboratory (Westchester, Ill) receives biopsy specimens from 45 urologists within a single multilocation practice. Specimens include prostate biopsies, bladder biopsies, and segments of vasa deferentia. Needle core prostate biopsy specimens are received in 30-mL commercially prepared formalin vials. The surgeons submit between 2 and 14 vials per patient.
Most vials contain 1 or 2 cores; specimens from transperineal template--guided (saturation) biopsy procedures may contain 5 or more cores in each vial.
Specimen Handling Before Inking Procedure
Each office has access to the laboratory information system via a secure Internet browser--based interface. Each specimen is accessioned at the individual office, with patient demographics and laboratory tests entered by the office staff. Requisitions and specimen container labels are printed in the office. Each specimen container label is bar coded and imprinted with the patient's name, specimen identification number, and biopsy site. The specimen labels are affixed by the office staff to each container; all specimens and the requisition for the patient are sealed in a clear plastic bag. Each bag then receives a unique bar-coded label for specimen tracking by the courier service. A specimen manifest is generated for each office, and a copy is transported with the specimens to the laboratory. Upon delivery to the laboratory, each specimen is checked into the laboratory information system by scanning of the bar-coded label.
Biopsies from each vial undergo gross examination and are grossed and submitted for processing in separate prelabeled (VCP Millennium Printer, Surgipath, Richmond, Ill) tissue cassettes with blue sponge pads presoaked in formalin. Extra cassettes are used if more than 3 cores are present in a vial. Processing is performed on a Tissue-Tek VIP 5 processor (Sakura Finetek US, Torrance, Calif), followed by hand embedding and cutting, automated hematoxylin-eosin staining (Leica Autostainer XL, Leica Microsystems, Nussloch, Germany), and automated coverslipping (LjungTech GCS 600 Glass Coverslipper, Surgipath). Biopsies were cut at 4 [micro]m, with 3 levels on each of 3 prelabeled (VSP Millennium Printer) slides. Slides 1 and 3 were routinely stained, with slide 2 reserved for special stains, if needed.
Following cutting, all blocks and slide accession numbers were compared by a second individual to verify accuracy. The water bath surface was cleared with tissue (Kimwipes, Kimberly-Clark Professional, Dallas, Tex) between cutting each block.
Specimen Handling With Inking Procedure
For a 12-month period beginning in April 2007, the grossing technician stained each prostate core with permanent ink developed for inking of pathology specimens (Tissue Marking System, Cancer Diagnostics Inc, Birmingham, Mich). A different color was used for each patient, with all biopsies from all vials for a particular patient inked with the same color. Five colors were used sequentially: green, blue, yellow, orange, and black. The ink was diluted with distilled water to a consistency that allowed application of a thin, uniform coating of ink along the edges of the biopsies (Table 1). Red was evaluated but discarded because of pathologists' concerns that the red ink color might lead to misinterpretation of immunohistochemistry staining using Vulcan Fast Red (Vulcan Fast Red Chromogen Kit 2, BiocareMedical, Concord, Calif) as the chromogen. Purple was also evaluated but was too granular for consistent staining.
Working (diluted) solutions of ink were stored in 20-mL plastic squeeze bottles (http://www.blickart.com). Prior to inking a patient's biopsies, a small amount of working solution of the appropriate color was placed in a plastic boat (plastic disposable square weigh boat, 4.5 X 4.5 cm; Cardinal Health, Dublin, Ohio), and the ink color was dictated for transcription. The ink was applied to the biopsies using a flat wooden toothpick (purchased in any market or grocery). A new toothpick was used for each biopsy vial. After all biopsies were inked, the ink boat and any residual ink were discarded. The process was repeated for each patient using the next color ink in the series and a new boat and toothpicks. Inked biopsies were then placed between blue biopsy pads that were presoaked in Bouin Fixative (Newcomer Supply, Middleton, Wis), which was used as a mordant; excess ink was absorbed by the pads.
The time required to ink patient specimens comprising different numbers of vials and biopsies was timed during a 2-day period. While the grossing technician was working, a second technician recorded total time to complete the patients' specimens and the time required to ink the biopsies (Table 2).
At time of embedding, the histotechnologists checked each core macroscopically to verify that it was the same color as all other cores for the particular patient. The pathologist reading the microscopic slides was responsible for comparing the ink color staining the tissue microscopically with the ink color dictated by the grossing technician.
During the study period, surgical biopsy specimens were received for 4068 patients. Of these, 2251 (55%) were prostate biopsy cases. Of 25 764 blocks processed, 21 877 (85%) were prostate biopsies. Histotechnologists did not detect any misplaced specimens. Pathologists identified 2 small prostate tissue fragments, "floaters," with ink colors that did not match either the dictated ink color for the patient or the color of the other biopsies as seen microscopically.
The miscolored fragments were on 1 level of 1 slide and were most likely water bath floaters. The pathologists felt they would have identified the fragments as floaters without the ink, but the inking removed any doubt. Additionally, a mismatch was detected between the color indicated on the preliminary gross protocol and the ink color on the biopsies. Review of this case revealed a transcription error; the dictated color on the gross dictation tape matched the biopsy color. In no other cases was tissue with the improper ink color identified microscopically. Thus, the identified discrepancy rate was 3 of 2251 prostate biopsy patients, or 0.13%. The discrepancy rate in terms of total number of prostate blocks was 3 discrepancies in 21 877 prostate blocks, or 0.014%.
Diluted inks adhered to biopsy contours throughout tissue processing. The tissue showed no untoward reactions to the inks. The only unanticipated complication of inking was staining of the interior walls of the formalin reservoirs of our tissue processor. The staining is permanent; bleaching was unable to remove the staining. Otherwise, the tissue processor does not appear to have suffered any adverse consequences from the inks.
Maintaining identification of patient specimens throughout the course of processing is crucial, yet is difficult to prove when processing many specimens from the same organ with similar gross appearance. Meticulous attention to detail, such as checking slide accession numbers against block accession numbers and clearing the surface of the water bath following cutting of each block, is necessary but not foolproof. Using core inking adds another level of verification.
Our inking method is not flawless. It is intended to identify mix-ups of biopsies from different patients during tissue processing, which includes tissues released during processing due to improperly closed or damaged cassettes, embedding errors, and floaters. It will not detect transfer of a core from one block to another in the same patient. It also will not detect transfers to another patient in whom the same color is used. This risk is minimized by using the maximum number of colors (5 in our study) and using colors in sequential order so that at least 4 patients separate the repeat of a particular color. Although there is a 20% increase in grossing time, we feel this added time is justified by the increased confidence that a malignant diagnosis, even one based on a single core out of multiple cores in multiple vials, is being assigned to the correct patient.
It might be argued that the above procedure is unnecessary, since uncertainties related to patient identity of processed specimens can be resolved using genetic analysis of fixed tissue. (5--7) Genetic analysis, however, is only useful once an uncertainty has been raised. This uncertainty may not be voiced until therapeutic action has already been taken based on the diagnosis of a mismatched specimen. Our procedure is intended to question specimen identity prior to any therapeutic action. In such situations, a genetic identification procedure could be used to definitively identify patient specimens prior to starting definitive therapy.
We feel inking can be used in other specialty laboratories, particularly laboratories specializing in gastrointestinal biopsies. High-volume laboratories that receive large numbers of prostate, bladder, uterine cervix, and gastrointestinal biopsies may find inking of these biopsies helpful to maintain specimen identity. This would be particularly true if many similar tissue specimens are processed sequentially or similar specimens are batched for processing.
In our opinion, in-office inking (ie, inking at the biopsy site) by more than a dozen different office staffs (in our case) is impractical and unmanageable. Quality assurance would be difficult to maintain in unwilling and harried office personnel. An essential part of our inking procedure is sequencing the colors to improve security. We feel sequencing the colors is best handled in the laboratory, where confusion is kept to a minimum. Additionally, adding ink to specimen vials after the biopsies have been placed in them would obscure already small biopsies, making retrieval in their entirety more difficult.
Our inking procedure would be expected to have less of an impact on our tissue processor compared with what is routinely done for specimens processed in hospital laboratories (eg, breast biopsy, thyroidectomy, and gastrectomy), because our inks are applied diluted, whereas those used in hospital laboratories are applied undiluted.
Based on our experience, we have made the inking of all prostatic core biopsies a standard operating procedure.
We join Andrew A. Renshaw, MD, in advocating inking as an effective method of addressing Joint Commission patient safety goals by maintaining specimen identity during processing of similar types of gross specimens.
Accepted for publication July 16, 2008.
(1.) Corrigan J, Kohn LT, Donaldson MS, eds. To Err is Human: Building a Safer Health Care System. Washington, DC: National Academy Press; 2000.
(2.) Makary MA, Epstein J, Pronovost PJ, Millman EA, Hartmann EC, Freischlag JA. Surgical specimen identification errors: a new measure of quality in surgical care. Surgery. 2007;141:450-455.
(3.) The Joint Commission. National patient safety goals. Available at: http://www.jointcommission.org/PatientSafety/NationalPatientSafetyGoals/ 08_lab_npsgs.htm.
(4.) Renshaw AA, Kish R, Gould EW. The value of inking breast cores to reduce specimen mix-up. Am J Clin Pathol. 2007;127:271-272.
(5.) Shibata D. Identification of mismatched fixed specimens with a commercially available kit based on the polymerase chain reaction. Am J Clin Pathol. 1993;100:592-593.
(6.) Suba EJ, Pfeifer JD, Raab SS. Patient identification error among prostate needle core biopsy specimens--are we ready for a DNA time-out? J Urol. 2007;178: 1245-1248.
(7.) Cao D, Hafez M, Berg K, Murphy K, Epstein JI. Little or no residual prostate cancer at radical prostatectomy: vanishing cancer or switched specimen?: a microsatellite analysis of specimen identity. Am J Surg Pathol. 2005;29:467-473.
Lester J. Raff, MD; George Engel, MD; Kenneth R. Beck, MD; Andrea S. O'Brien, HT, ASCP; Meagan E. Bauer, BA
From the Department of Anatomic Pathology, UroPartners LLC Laboratory, Westchester, Ill.
The authors have no relevant financial interest in the products or companies described in this article.
Reprints: Lester J. Raff, MD, Anatomic Pathology, UroPartners LLC Laboratory, 2225 Enterprise Dr, Suite 2511, Westchester, IL 60154 (e-mail: firstname.lastname@example.org).
Table 1. Color Ink Dilutions * Ink Distilled Water Green 1 1 Blue 1 1 Yellow 1 2 Orange 1 4 Black 1 0.5 * Proportion of ink to distilled water used in diluting each color is shown. Dilutions determined by trial and error to achieve an even, thin coating of ink around the edges of each biopsy. Table 2. Inking as Percentage of Grossing Time * 6 Vials/ 8 Vials/ 12 Vials/ Case Case Case Total No. of patients 3 4 6 13 No. of cores/fragments 41/18 81/14 73/25 195/57 Total time to ink, s 166 329 484 979 Total grossing time, s 883 1639 2362 4884 % Time inking 18.8 20.1 20.5 20.0 * Core: [greater than or equal to] 5-mm length; most cores, [less than or equal to] 10-mm length. Fragment: <5-mm length.
|Gale Copyright:||Copyright 2009 Gale, Cengage Learning. All rights reserved.|