Analysis of Addenda in Anatomic Pathology as a Quality Monitoring Initiative.
The central objective of this study is to audit surgical pathology, cytology, and autopsy addenda reports in the pathology laboratory of an academic tertiary health care center to identify instances where addenda were inappropriately used to convey information that changed a report's diagnostic, prognostic, or therapeutic meaning. It is our premise that such information that comes to light after an original surgical pathology report has been issued requires an amended report.
All addendum reports and the corresponding originally issued reports for autopsy, cytology, and surgical pathology cases at an academic tertiary care pathology laboratory in Toronto, Ontario, Canada, during a 36-month period, from January 1, 2008, through December 31, 2010, were retrieved, reviewed, and initially categorized by a pathologist in a nonblinded fashion. After review, each addendum report was classified by accession class (autopsy, cytology, surgical pathology), issuing pathologist, subspecialty category, indication for addendum, and whether or not the contents of the addendum represented a change in the report's diagnostic, prognostic, or therapeutic meaning. The indication of all surgical pathology and cytology amendments over the course of the defined 36-month period was also reviewed for comparison.
All staff pathologists from all subspecialty areas available at our institution were included and grouped according to the following subspecialty categories: gastrointestinal, genitourinary, medical renal, dermatopathology, breast, gynecologic, cardiovascular, oral, orthopaedic, pulmonary, lymphoma/hematopathology, neuropathology, sarcoma and soft tissue, head and neck, and placenta. The indications for addenda reports considered included additional hematoxylin-eosin (H&E) sections, decalcification, missing parts or missing data, reporting of special histochemical stains (i.e., gram stain, periodic acid-Schiff stain with diastase), reporting of diagnostic immunohistochemical studies, reporting of prognostic biomarkers, reporting of results of molecular studies, reporting of electron microscopy study, reporting of immunofluorescence study, reporting of cytogenetics, review of external archival material, review of internal archival material, external consultation, intradepartmental consultation, disposition of material, autopsy neuropathology procedure/report, microsatellite instability reporting, clinician-requested review, significant omission (omitting information that would change treatment or prognosis; eg, omitting reporting a positive resection margin in a skin resection specimen for melanoma), minor omission (omitting information that would have no impact on treatment or prognosis of the patient; eg, reporting an incidental seborrheic keratosis in a skin resection specimen for invasive squamous cell carcinoma), disagreement with consultant, and other. Cytology addenda were further stratified as gynecologic cytology cases, nongynecologic fine-needle aspiration (FNA) cases, and nongynecologic non-FNA cases.
A change in diagnostic meaning was considered to have occurred when a different diagnosis was stated in the addendum that contradicted the primary diagnosis in the original report (eg, change from a benign diagnosis to malignant diagnosis or vice versa). A change in prognosis was considered to occur when the supplemental information significantly impacted the prognosis of the original diagnosis (eg, diagnostic immunohistochemistry proves micrometastatic nodal disease in sentinel axillary nodes after the case was finalized and released as negative for lymph node metastases). A change in implication for treatment included situations where the supplemental information provided in the addenda required action from the clinical management team, such as further surgery, adjuvant therapy, or discontinuing an already initiated treatment (eg, after release of a surgical pathology report, lymphovascular invasion is identified upon receipt of additional diagnostic immunohistochemical studies, and adjuvant chemotherapy would need to be initiated solely on this additional information). All reports identified as suggesting a change in diagnostic meaning, prognosis, or implications for treatment were reviewed in conjunction with any available clinical information by a second expert surgical pathologist. Addendum reports were labeled as misclassified only after consensus agreement between both reviewing pathologists upon completion of the review and discussion of the content, clinical context, and clinical ramification of the reports. Originally issued final reports that contained clear information about pending ancillary studies, and indicating that a subsequent addenda report was going to be issued with the results of these studies, were considered to be reported appropriately.
A total of 7464 addendum reports were identified for the defined 36-month period. Autopsy cases accounted for 135 addenda that were appropriately reported to primarily include details of neuropathologic examination (70.9%), reporting of an external consultation (8.7%), and reporting of molecular studies (6.8%). In all, 2301 addenda were issued for cytology cases during the 36-month period, including 2183 gynecologic cytology cases, 50 nongynecologic FNA cases, and 68 nongynecologic non-FNA cases (eg, peritoneal fluid aspiration). The primary indications for cytology addenda were the reporting of human papillomavirus molecular studies (95.1%), reporting of diagnostic immunohistochemical studies (3.0%), and reporting of special histochemical stains (0.6%). All addenda for cytologic cases were appropriately reported. In all, 5028 addenda for surgical pathology cases were issued, primarily by the breast site group (44.1%), and followed by the gastrointestinal site group (12.1%), dermatopathology service (11.0%), and neuropathology service (8.3%). The primary indications for issuing a surgical pathology addendum (Table 1) were reporting of a prognostic biochemical marker (42.4%), reporting of molecular studies (15.7%), reporting of diagnostic immunohistochemical studies (11.8%), reporting of histochemical stains (4.2%), and reporting of external consultations (3.9%).
None of the 135 autopsy or 2301 cytology addenda were deemed to have changed the initial diagnostic meaning, prognosis, or treatment implications of the primary report.
Thirty-three surgical pathology addenda over the defined time period were identified and deemed to contain information that warranted an amended report. The rate of surgical pathology faux addenda was 0.28 per 1000 cases for 2008, 0.37 per 1000 cases for 2009, and 0.37 per 1000 cases for 2010. Eight of these 33 addenda (24.2%) were by consensus determined to be a change in diagnosis. Thirty of the surgical pathology addenda contained supplemental information that would significantly alter patient management and 31 of the addenda had additional information that would significantly change the prognosis. Breast cases accounted for 30.6% of the faux addenda (Table 2), while 17.6% were dermatopathology cases, with gynecologic and gastrointestinal cases accounting for 15.7% each. The primary indications for the faux addenda (Table 3) included reporting of diagnostic immunohistochemical studies (12 of 33 faux addenda), reporting a significant omission (11 of 33 faux addenda), reporting results of additional H&E sections (4 of 33 faux addenda), reporting of decalcified sections (1 of 33 faux addenda), and reporting of special histochemical studies (1 of 33 faux addenda). Our review identifies that 17 of 33 (51.5%) faux addenda were issued before receipt and review of diagnostic immunohistochemical studies (12 of 33 faux addenda), additional H&E sections (4 of 33 faux addenda), or reporting of special histochemical staining (1 of 33 faux addenda). Although 11 individual pathologists had issued faux addenda, a single pathologist was responsible for 8 of the 33 (24.2%) misclassified cases. Examples of faux addenda are provided in Table 4.
During a 36-month period encompassing January 1, 2008, through December 31, 2010, 499 surgical pathology amendments were issued at a rate of 0.5 amendments per 1000 surgical pathology cases, including in-house and consultation cases. Two hundred eighty-six amendments (57.3%) were issued to change the final diagnosis or comment. Thirty one (6.2%) surgical pathology amendments were issued to report a change in the macroscopic description, and 58 (11.6%) were submitted to report a change in the microscopic description. One hundred nine (21.8%) amendments were found to be issued to correct clinical information and 15 (3.0%) to correct patient identification. During the same 36-month period, 58 amendments were issued for cytology class specimens at a rate of 0.69 amendments per 1000 cytology cases including both gynecologic and nongynecologic cases. Thirty-five of the 58 amendments (60.3%) were issued to change the diagnosis or comment and 14 (24.1%) were issued to correct clinical information. The remaining cytology amendments were issued to change the macroscopic description (3 amendments, 5.1%), to change the microscopic description (5 amendments, 8.6%), and to correct patient identification (1 amendment, 1.7%).
This review of 7464 autopsy, cytology, and surgical pathology addendum reports issued during a continuous 36-month period is one of the largest documented analyses of its kind. Our study documents the pattern of use and frequency of autopsy and cytology addendum reports and verifies that they are truly supplemental in nature. More importantly, it demonstrates that surgical pathology addenda may contain key diagnostic, prognostic, or treatment-related information that differs from the primary report. This study advances the concept that the usual practice of monitoring only reports designated as amendments likely underestimates true amendment rates, because some amendments masquerade as addenda.
Documenting the amendment rate of surgical pathology reports is a standard component of internal quality assurance programs. In a College of American Pathologists Q-Probes study, Nakhleh et al (1) quantified the number of amended reports reported by laboratories and suggested that laboratories should aim to reach the "best practice" benchmark of 0.22 amended reports per 1000 surgical pathology reports. In our study, the rate of surgical pathology addenda that met the common criterion for an amendment was 0.28 per 1000 cases for 2008, 0.37 per 1000 cases for 2009, and 0.37 per 1000 cases for 2010. The data from this study argue that amendment rates per 1000 cases are understated if addenda are not monitored. Addenda need to be monitored to detect faux addenda that are really amendments. These changes must be added to numerators of amendment rates.
The length of time from receipt of a specimen to issuance of a final surgical pathology report is understandably constantly monitored as a standard quality indicator in surgical pathology practice. With the increasing integration of immunohistochemical biomarker testing and diagnostic molecular techniques into everyday practice, ancillary tools for diagnosis are more frequently requested. In a high-volume tertiary care center, ancillary studies, such as immunohistochemical studies, decalcification of tissue blocks, and molecular techniques, may take many days to complete. To avoid negatively impacting the time to diagnosis, surgical pathology reports are often issued with additional studies still pending. In our review, it was found that 12 of 33 faux addenda (36.4%) were issued for the reporting of a diagnostic immunohistochemical study and 4 of 33 faux surgical pathology addenda (12.1%) were issued for additional H&E sections. In surgical pathology cases where diagnostic immunohistochemical markers or additional sections have been requested, when the outcome has the potential to change either the fundamental diagnosis, prognosis, or treatment plan, in our practice it is advised that the final surgical pathology report not be released until these studies have been completed and reviewed. The data from our internal review identify that 17 of the 33 unmonitored misclassified supplemental reports (51.5%) could have been prevented by the practice of holding the release of the final surgical pathology report until receipt and complete review of all diagnostic immunohistochemical studies, special histochemical stains, and additional H&E sections. The additional wait time for additional immunohistochemical markers, histochemical stains, and H&E sections on a small subset of cases will in most circumstances not have a significant impact on immediate patient care or a pathologist's global turnaround time. It is recognized, however, that situations arise where rapidity of diagnosis is paramount. In this circumstance, strong communication with the clinical management team is required, and caution should be used when wording the preliminary report and diagnosis to ideally incorporate the working differential diagnosis, a listing of pending ancillary tests, and notification of a pending supplemental report with the results of the pending ancillary tests.
As clinical decisions or actions may have been initiated from the initial final diagnosis, it is required under the College of American Pathologists Commission on Laboratory Accreditation Laboratory Accreditation Program (GEN.41310, phase II) (1) that previous information be immediately available for comparison with the revised information in the supplemental report. Under this requirement, the initial report and the supplemental report must be identified as such, and the original data must be present in the revised report, or linked electronically or logically to the revised report. It is vital that pathologists be aware of how the laboratory information system interfaces with the hospital electronic patient record and specifically, how supplemental reports are presented to the clinical management team. In many systems, addenda appear at the bottom of the report, and thus may easily go undetected by the reader who does not look beyond the "final diagnosis" section of the report. In contrast, amended reports typically replace the content of the original report with the content of the corrected report. In addition, many departments have additional policies in place to ensure prompt notification of clinicians when amended reports are issued, but such policies are generally not applied to addendum reports. Given the observed frequency of faux addenda, it is plausible that suboptimal communication of supplemental addendum reports may be a preventable source of postanalytic error that may result in delay of appropriate treatment or initiation of inappropriate therapy.
Our study corroborates the findings of Finkelstein et al (3) in that surgical pathology addenda may contain information that significantly alters or contradicts the original report. The current study encompassed 3 consecutive years of addenda, which included all specimen classes (surgical pathology, cytology, and autopsy), while the study of Finkelstein et al (3) examined surgical pathology addenda during 4 discontinuous months spread across more than a decade. Our expanded data set suggests that faux addenda appear to be an issue primarily in surgical pathology and thus, monitoring efforts should focus on this division of laboratory medicine. Within surgical pathology, our rates of faux addenda were significantly lower than those found by Finkelstein et al, (3) but still higher than the "best practice" benchmark of 0.22 amended reports per 1000 surgical pathology reports suggested by a College of American Pathologists Q-Probes study. (1) The difference in rates of faux addenda between the 2 studies may reflect institutional differences or differences in practice patterns in different time periods, as a segment of the data of Finkelstein et al (3) encompassed months within a period spanning more than a decade.
Similar to the opinion put forward by Finkelstein et al, (3) it is also our impression that addenda reports receive less urgent attention by pathologists in comparison to final reports. Causes of this pattern of behavior may be the failure of pending addenda reports to be prioritized on a pathologist's work list created by our institutions' laboratory information system, and the fact that incoming workload requiring evaluation may be given priority and addendum report turnaround times are not monitored as a routine quality indicator. Given the inherent limitation of a single institution study and variability of practice patterns among institutions, the source of this trend will likely vary between laboratories.
With the advancement and integration of ancillary techniques into routine practice, addenda in pathology reporting have not only increased in frequency but also evolved from supplementing purely diagnostic information to including clinically relevant prognostic and therapeutic information. The lack of standardized practice for issuance of supplemental reports has created a source of confusion for practicing pathologists and a genuine opportunity to misclassify a supplemental report. While most addenda are truly supplemental in nature, a subset of surgical pathology addenda contains important information that should replace some components of the original surgical pathology report in the form of an amended report. Our study demonstrates that faux addenda in surgical pathology occur at a frequency that is comparable to accepted amendment rates per 1000 cases. It is noteworthy that most of these reports were necessary because pathologists signed out reports before receiving all slides and special stains associated with the case. Addressing the issue of amendments that masquerade as addenda is of importance owing to the differing ways in which these results are communicated. Educational efforts should focus on reviewing all sections and stains before issuing the final report; policies should define when an amendment is necessary, and adherence to this policy should be monitored to provide clarity and ensure patient safety.
(1.) Nakhleh RE, Zarbo RJ. Amended reports in surgical pathology and implications for diagnostic error detection and avoidance: a College of American Pathologists Q-Probes study of 1,667,547 accessioned cases in 359 laboratories. Arch Pathol Lab Med. 1998; 122(4):303-309.
(2.) Goldsmith JD, Siegal GP, Suster S, et al. Reporting guidelines for clinical laboratory reports in surgical pathology. Arch Pathol Lab Med. 2008; 132(10): 1608-1616.
(3.) Finkelstein A, Levy GH, Cohen P, et al. Addenda in pathology reports: trends and their implications. Am J Clin Pathol. 2012; 137(4):606-611.
(4.) Roy JE, Hunt JL. Detection and classification of diagnostic discrepancies in surgical pathology. Adv Anat Pathol. 2010; 17(5):359-365.
(5.) Meier FA, Zarbo RJ, Varney RC, et al. Amended reports: development and validation of a taxonomy of defects. Am J Clin Pathol. 2008; 130(2):238-246.
(6.) Cooper K. Errors and error rates in surgical pathology: an Association of Directors of Anatomic and Surgical Pathology survey. Arch Pathol Lab Med. 2006; 130(5):607-609.
(7.) Stastny JF, Geisinger KR, Michael CW, et al. Another quality assurance issue: amended reports: what do we really know about them? Diagn Cytopathol. 1998; 18(1):67-70.
(8.) College of American Pathologists. Commission on Laboratory Accreditation: Laboratory General and Anatomic Pathology Checklists. Northfield, IL: College of American Pathologists; 2005.
Jesse Paul Babwah, MD; Mahmoud Khalifa, MSc, MBBCh, PhD; Corwyn Rowsell, MD
Accepted for publication January 24, 2014.
From the Department of Laboratory Medicine and Pathobiology, The University of Toronto, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada.
The authors have no relevant financial interest in the products or companies described in this article.
Reprints: Jesse Paul Babwah, MD, Department of Laboratory Medicine and Pathobiology, The University of Toronto, Sunnybrook Health Sciences Centre, 2075 Bayview Ave, Room E4-32, Toronto, ON M4N 3M5, Canada (e-mail: firstname.lastname@example.org).
Table 1. Breakdown of Surgical Pathology Addenda per Year Indication 2008 2009 2010 Total No. No. No. No. of of of of Cases Cases Cases Cases Additional H&E sections 43 36 32 111 Decalcification 48 36 41 125 Missing parts or missing data 28 14 32 74 Reporting of special histochemical 49 67 96 212 stains (ie, PASD) Reporting of IHC studies (diagnostic) 208 190 197 595 Reporting of prognostic biomarker 786 782 555 2123 Reporting of results of molecular 204 194 394 792 studies Reporting of electron microscopy study 61 71 50 182 Reporting of immunofluorescence study 12 33 61 106 Reporting of cytogenetics 28 29 49 106 Review of archival material (external) 2 1 10 13 Review of archival material (internal) 0 1 0 1 External consultation 72 66 59 197 Intradepartmental consultation 9 22 21 52 Disposition of material 3 0 3 6 Autopsy-neuropathology procedure/ 0 0 23 23 report Fetal autopsy 9 9 7 25 Microsatellite instability IHC 25 27 28 80 Clinician-requested review 14 14 6 34 Significant omission 4 10 2 16 Minor omission 31 19 2 52 Disagreement with consultant 0 0 0 0 Other 28 21 54 103 Total surgical pathology addenda by 1664 1642 1722 5028 indication Abbreviations: H&E, hematoxylin-eosin stain; IHC, immunohistochemistry; PASD, periodic acid--Schiff diastase stain. Table 2. List of Addenda Better Reported as Amendments by Site Group per Year Site Group 2008 2009 2010 Total No. No. No. No. of of of of Cases Cases Cases Cases Gastrointestinal and hepatobiliary 2 2 1 5 Genitourinary (surgical) 1 2 0 3 Skin 1 3 2 6 Breast 3 3 4 10 Female genital tract 2 1 2 5 Lymphoma 0 0 1 1 Head and neck 0 1 2 3 Total flagged addenda by site group 9 12 12 33 (surgical pathology) Table 3. List of Faux Addenda by Indication per Year 2008 2009 2010 Total General Criteria No. No. No. No. of of of of Cases Cases Cases Cases Additional H&E sections 0 0 4 4 Decalcification 0 0 1 1 Reporting of special histochemical 0 1 0 1 stains (ie, PASD) Reporting of IHC studies 4 5 3 12 (diagnostic) Intradepartmental consultation 0 0 1 1 Clinician-requested review 1 0 0 1 Significant omission 3 6 2 11 Other 1 0 1 2 Total flagged surgical pathology 9 12 12 33 addenda by indication Abbreviations: H&E, hematoxylin-eosin stain; IHC, immunohistochemistry; PASD, periodic acid-Schiff diastase stain. Table 4. Examples of Addenda Better Reported as Amendments Specimen Type Original Final Addendum Text Diagnosis Prostatectomy Adenocarcinoma, Tumor is present Gleason score 4 + 3 = focally at the left 7 (with synoptic peripheral margin at report) the base of the prostate. Cervix biopsy Endometrioid The tumor cells are adenocarcinoma, high positive for CDX-2 and grade CK20, negative for CK7 and ER. The staining pattern favors metastatic colorectal carcinoma. Total Nodular hyperplasia Occult papillary thyroidectomy thyroid carcinoma (0.8 cm). Resection of Acute inflammatory Colonic adenocarcinoma ileorectal infiltrate on serosal (with synoptic anastomosis surface; 8 lymph nodes report). with no evidence of malignancy Sentinel lymph Tiny fragment of Recut slides and IHC node dissection crushed lymphoid studies (S100 and for melanoma tissue, negative for MART-1) showed malignancy positivity for malignant melanoma. Breast core Infiltrating lobular E-cadherin test was biopsy carcinoma, performed and the intermediate nuclear small focus of grade (2/3) invasive carcinoma shows membranous staining. This pattern supports the ductal, not lobular, type. The diagnosis is finalized. Colon Nonspecific chronic Poorly differentiated endoscopic inflammation and edema carcinoma, consistent biopsy with metastatic carcinoma. Random gastric Small-bowel mucosa Amyloidosis biopsy with no significant histologic abnormality Specimen Type Indication for Comment Addendum Prostatectomy Significant omission The positive margin status was not reported in the initial pathology report. Cervix biopsy Reporting of The initial report was diagnostic IHC signed out before studies receiving diagnostic IHC. The initial report did not indicate that IHC results, which potentially can change the diagnosis, were pending. Total Pending decalcified The initial report had thyroidectomy sections a benign final diagnosis and did not indicate decalcified sections were pending. Resection of Clinician-requested Discordance with the ileorectal review clinical presentation anastomosis prompted a request by the clinician to review the specimen grossly and microscopically. Sentinel lymph Reporting of The initial report was node dissection additional H&E signed out as negative for melanoma slides and IHC for malignancy before studies receiving additional H&E slides and additional diagnostic IHC. Additional studies demonstrate metastatic disease. Breast core Reporting of The initial report was biopsy diagnostic IHC signed out as invasive studies lobular carcinoma. There was no mention on the initial report of pending diagnostic IHC (E-cadherin). Colon Reporting of The history of a endoscopic diagnostic IHC gastric mass prompted biopsy studies IHC for the colon biopsy sample. The initial report was issued before receiving IHC results. There was no indication of pending studies. Random gastric Reporting of The initial diagnosis biopsy histochemical did not report an stains abnormality. Abbreviations: CDX-2, homeobox protein CDX-2; CK, cytokeratin; ER, estrogen receptor; H&E, hematoxylin-eosin stain; IHC, immunohistochemistry; MART-1, melanoma antigen recognized by T-cells 1.
|Printer friendly Cite/link Email Feedback|
|Author:||Babwah, Jesse Paul; Khalifa, Mahmoud; Rowsell, Corwyn|
|Publication:||Archives of Pathology & Laboratory Medicine|
|Date:||Nov 1, 2014|
|Previous Article:||Keeping Pace With New Frontiers in Pathology.|
|Next Article:||Stromal Tumor of Uncertain Malignant Potential of the Prostate.|