Indian Journal of Pathology and Microbiology
Home About us Instructions Submission Subscribe Advertise Contact e-Alerts Login 
Users Online: 47
Print this page  Email this page Bookmark this page Small font sizeDefault font sizeIncrease font size


 
ORIGINAL ARTICLE Table of Contents   
Year : 2008  |  Volume : 51  |  Issue : 4  |  Page : 469-473
Correlation of intra-operative frozen section consultation with the final diagnosis at a referral center in Karachi, Pakistan


Department of Pathology and Microbiology, Aga Khan University, Stadium Road, Karachi 74800, Sindh, Pakistan

Click here for correspondence address and email
 

   Abstract 

Background: The correlation of intra-operative frozen section diagnosis with final diagnosis on permanent sections is an integral part of quality assurance in surgical pathology laboratories. However, there is scant data on this topic from Pakistan. Similarly, no local study has looked at frozen section turnaround times. Aims and Objectives: To analyze indications, discrepancies and deferrals for all frozen sections performed or received at our institution over a 1-year period and to determine the turnaround time for frozen section diagnoses in our cases. Design: A retrospective study, was undertaken, of all frozen sections reported at our institution between 1 st January 2006 and 31 st December 2006. The records of these cases were reviewed. The number and types of discrepancies, including sampling and interpretation errors were determined. The deferred cases and causes for deferral were also determined. The turnaround times of all cases were recorded. Agreement rates were calculated as percent agreement, sensitivity/specificity and positive and negative predictive values. Results: A total of 356 specimens were received. Out of these, 14 cases (3.93%) were deferred to permanent sections. Of the remaining 342 cases, the discordant diagnostic frequency was 2.92% while the concordant diagnostic frequency was 97.08%. The most common pathological processes encountered were presence/typing of neoplasm (51.12%) and assessment of surgical margins (27.53%). The average turn-around time for frozen section diagnosis was 23 minutes; 60% of the cases were reported in 20 minutes or less. Conclusions: The accuracy of frozen section diagnosis at AKUH pathology department can be interpreted as comparable with most international quality control statistics for frozen sections. The overall error and deferral rates are within the range of previously published errors in pathology. Deferrals and errors in some sub-specialties were higher than in others. The results suggest specific measures should be taken to reduce the number of discrepancies. The overall goal is to reduce errors, reduce the number of deferrals and improve frozen section diagnosis turnaround times.

Keywords: Concordance, discordance, frozen section, Surgical Pathology, Pakistan

How to cite this article:
Ahmad Z, Barakzai MA, Idrees R, Bhurgri Y. Correlation of intra-operative frozen section consultation with the final diagnosis at a referral center in Karachi, Pakistan. Indian J Pathol Microbiol 2008;51:469-73

How to cite this URL:
Ahmad Z, Barakzai MA, Idrees R, Bhurgri Y. Correlation of intra-operative frozen section consultation with the final diagnosis at a referral center in Karachi, Pakistan. Indian J Pathol Microbiol [serial online] 2008 [cited 2014 Jul 24];51:469-73. Available from: http://www.ijpmonline.org/text.asp?2008/51/4/469/43733



   Introduction Top


The correlation of intra-operative frozen section diagnosis with the final histopathological diagnosis on permanent sections should form an integral part of quality assurance activities in the surgical pathology laboratory. Intra-operative frozen section diagnosis is not a routine practice in most institutions in Pakistan, thus there are very few studies or data available from Pakistani institutions. [1]

Intra-operative frozen section examination has been shown as an excellent diagnostic test. Two categories of errors are responsible for the majority of the disagreements, which are most commonly false negative. [2] Internationally published studies have confirmed the overall accuracy of intra-operative frozen section examination. [3],[4],[5] A College of American Pathologists (CAP)-sponsored review of over 90,000 frozen sections at 461 institutions showed a concordance rate of 98.58%. Of the discordant cases, 67.8% were false - negative diagnoses. The study reasons that the main causes for the discrepancies were either misinterpretation of the original frozen section (31.8%), absence of diagnostic tissue in the material frozen but present in the unsampled material (31.4%), [6] or the absence of diagnostic tissue in the frozen section but present in the corresponding permanent section (30%). In terms of turn-around time, a CAP-sponsored study of over 30,000 frozen sections at 700 hospitals from various countries showed that 90% of the frozen sections were completed i.e., the frozen section diagnosis was communicated to the surgeon, within 20 minutes. [7]

The Aga Khan University Hospital (AKUH) is a referral center in Karachi, Pakistan. The pathology department has been performing diagnoses on frozen sections for more than 20 years and the annual number of cases has been increasing gradually.

Aims and Objectives

  • To analyze indications, discrepancies, and deferrals for all diagnoses performed on frozen sections or received at our institution over a 1-year period.
  • To determine the turn-around time for frozen section diagnoses in our cases.



   Material and Methods Top


During this study, epidemiological data and pathology slides of 356 cases of frozen sections reported at the AKUH pathology department between 1 st January 2006 and 31 st December 2006 were reviewed. The selection criteria were surgical pathology cases where intra-operative frozen sections and permanent paraffin fixed tissue were both assessed at the AKUH pathology department.

Information regarding surgical pathology cases was extracted by the hospital Information System Department (ISD). The reported epidemiological data were rechecked. Cases were categorized by tumor site, indications of frozen section, discordance and concordance status and deferral status. Variables recorded were the hospital patient-number, date of registration, and topography. Manual and computerized validity checks for the data were performed to ensure reliability and to avoid duplication of the data. Each case had already been assigned a single accession number, the medical record (MR) number.

The frozen section specimens, as per laboratory routine, had been evaluated using Hematoxylin and Eosin (H&E) stains. Frozen sections were cut on a Shandon cryotome FSE machine. Subsequently, for the permanent section, the specimen was fixed overnight in 10% buffered formalin, grossed and adequate representative sections were taken according to the standardized guidelines. The permanent sections were initially evaluated on H&E stained sections. Special stains and immuno-histochemical stains were performed for confirmation of diagnosis and typing of tumors. For this study, already recorded data were retrieved; however, to ensure quality control, random slides, both frozen and permanent, were reviewed.

The data was computerized and analyzed using SPSS, Version 15.0. The number and type of discrepancies were compared, including sampling and interpretation errors. The deferred cases and causes for deferral were also determined. We also looked at the turn-around times by reviewing the records of all the cases. The agreement rates were calculated as percent agreement, sensitivity and specificity, and positive and negative predictive values.


   Results Top


The total number of surgical pathology cases received in the histopathology section during the study period was approximately 37,000. A total of 356 specimens were received for intra-operative consultation (frozen section), which is approximately 1 specimen for every day of the year. Out of these, 14 cases were deferred to permanent sections (3.93%). Of the remaining 342 cases, 10 cases were discordant, giving a discordant diagnostic frequency of 2.92% while 332 cases were concordant with a concordant diagnostic frequency of 97.08%.

The most common pathological processes encountered were verification and categorization of neoplasm (182 cases, 51.12%) and assessment of tumor margins (98 cases, 27.53%) [Table 1]. The third largest group was lymph nodes for metastases (59 cases, 16.57%). The most common source of deferred cases was lymph nodes whilst the most common source of discordant cases were ovarian masses and assessment of margins [Table 2]. The highest number of concordance frequency was observed for tumor margins (95 cases, 28.61%). Of the 10 discordant cases, there were 5 falsepositive cases and 5 false-negative cases. In these cases, the tumor was either found in deeper levels or in the additional tissue that had remained unsampled at the time of frozen section. In 2 cases, there was a misinterpretation of the original frozen section. In cases of breast carcinoma, breast, ovary, CNS, and bone were among the most common tissues sent for frozen section diagnosis [Table 2], excluding lymph nodes, for the presence of tumor metastases from various tissues or sentinel lymph nodes. Margins were most commonly sent from carcinomas of upper respiratory tract followed by breast and urinary bladder. The approximate turn-around time of frozen section diagnosis in our cases was 23 minutes with 60% of the cases reported in 20 minutes or less. The sensitivity of frozen section as a diagnostic test was 98.52% and the positive predictive value was also 98.52%.


   Discussion Top


The accuracy of frozen section diagnosis at the AKUH pathology department can be interpreted as comparable with most international quality control statistics for frozen sections. The deferral frequency was 3.93%, the discordant diagnostic frequency was 2.92%, and the concordant diagnostic frequency was 97.08%. The cause of discordance is interpretive; see [Figure 1(a)] and [Figure 1(b)]. Gross sampling errors or inadequate deeper levels at the time of frozen section reporting are shown in [Figure 2(a)] and [Figure 2(b)].

These findings are comparable with published CAP studies by Zarbo, et al. 1991 and Novis, et al. 1996 [8],[9] , which show deferral rates of 4.2% and 4.6%, discordant rates of 1.7% and 1.8%, and concordant rates of 98.3% and 98.2%, respectively. The reasons given for discordant diagnoses included gross tissue sampling (44.8%), misinterpretation (40%), sectioning (12.7%), inadequate history (5.6%), staining (1.5%), labeling (0.5%), and others (3%). More recent studies published in 2007 also concluded that disagreements were mostly due to interpretive and gross sampling errors. In one of these studies, false-negative disagreements were 3 times as common as false-positive disagreements. [2] It must, however, be emphasized that all the above cited studies looked at a much larger number of cases compared to our study.

Parwani, et al. [10] confirmed that higher deferral rates were found in some sub-specialties compared with others with the highest rates found in homeopath followed by ENT specimens, dermatopathology, and neuropathology cases. A study looking at pediatric cases found a discrepancy rate of 4% and a high deferral rate of 25%. [11] The authors reasoned that the high deferral rate in children was cause-specific; frozen sections being diagnosed less often for detection of lymph node metastases, adequacy of tumor margins, or to identify normal or unknown tissue; and more often for tumor classification and detection and for possible Hirchsprung disease.

A CAP study conducted in 1996 evaluated the reasons (indications) for and the immediate intra-operative surgical results (outcomes) associated with pathology intra-operative consultation. [8] This study included institutions in multiple countries and evaluated almost 10,000 cases. It concluded that the five most common indications for intra-operative consultation were verification or confirmation of diagnosis to determine type or extent of operation (51%); confirmation of adequacy of margins (16%); confirmation of tissue type (10%); expedition of diagnosis to inform family or patient (8%), and lastly to confirm sufficient tissue was submitted to secure a diagnosis in permanent section (8%). The study confirmed that intra-operative frozen sections, regardless of the initial indications, influence immediate patient care decisions, resulting in changed surgical procedures in almost 39% of all operative cases. Studies from other countries, including developing countries, show similar percentages of deferral, discordance and concordance; and similar causes for disagreements. [12],[13]

We compared the performance of our institution with prior performances working on the hypothesis that long-term monitoring of frozen-permanent section correlation is associated with improved performance. A 1993 study from our institution had looked at 1031 frozen sections performed during a 6-year period (1986-1991) with deferral, discordant, and concordant rates of 3.58%, 1.74%, and 94.66%, respectively. [1] Compared with this study, the concordance rate of our department has improved from 94.66% to 97.08% over two decades. This correlates with a recent study that looked at the effectiveness of the long-term monitoring of errors detected by frozen section-permanent section correlation and concluded that long-term monitoring of frozen-permanent section correlation is associated with sustained improvement in performance. This study showed that discordant frequencies decrease with longer participation in Q-Tracks monitor program, as well as with lower deferred diagnostic frequencies. [14]

The average turn-around time of frozen section diagnosis in our center was 23 minutes; 60% of the cases were reported in 20 minutes or less. Western studies have shown that 85-90% of cases are reported in 20 minutes or less. [7] The reason for the delay in our cases is the availability of only one cryostat machine and the limited number of skilled technologists versus the multiple tissues sent per case. To name a few examples, multiple margins or multiple lymph nodes are sent for sampling especially in the case of breast, skin, prostate, and head and neck malignancies, which are serially processed. Similarly, at one time more than one frozen section specimen may be sent from different specialties, which due to the reasons discussed above cannot be processed simultaneously. It is hoped that with the acquisition of more equipment and with more trained staff, the turn-around time will be greatly improved. It must also be emphasized, that where the number of specimens in any case were two or three, the overwhelming majority of cases were reported in 20 minutes or less.

It is recommended that if analyses on frozen sections are performed by several individuals on a rotation basis, it is important for a senior pathologist to review the material periodically to ensure quality of the sections and to ensure that the agreement between the frozen section diagnosis and final diagnosis remain at an acceptable level. These periodic reviews are also useful in pointing out patterns of use and misuse of the frozen section procedure by the various departments and their individual members. [15]


   Conclusion Top


The overall error rates and deferral rates in the current study are within the range of previously published errors in pathology. Errors and deferrals in some sub-specialties were higher than in others. The results suggest that specific measures should be taken to reduce the number of discrepancies. The overall goal is to reduce errors, reduce the number of deferrals, and improve frozen section diagnosis turn-around times, which would require more equipment and more trained staff.

 
   References Top

1.Aijaz F, Muzaffar S, Hussainy AS, Pervez S, Hasan SH, Sheikh H. Intraoperative Frozen section consultation: An analysis of accuracy in a teaching hospital. J Pak Med Assoc 1993;43:253-5.  Back to cited text no. 1    
2.White VA, Trotter MJ. Quality assurance in anatomic pathology: Correlation of intraoperative consultation with final diagnosis in 2812 specimens. Abstract presented at the 96 th annual meeting of the United States and Canadian Academy of Pathology, 2007.   Back to cited text no. 2    
3.Oneson RH, Minke JA, Silverberg SG. Intraoperative pathologic consultation: An audit of 1,000 recent consecutive cases. Am J Surg Pathol 1989;13:237-43.  Back to cited text no. 3    
4.Howanitz PJ, Hoffman GG, Zarbo RJ. The accuracy of frozen section diagnoses in 34 hospitals. Arch Pathol Lab Med 1990;114:355-9.  Back to cited text no. 4    
5.Rogers C, Klatt EC, Chandrasoma P. Accuracy of frozen section diagnosis in a teaching hospital. Arch Pathol Lab Med 1987;111:514-7.  Back to cited text no. 5    
6.Gephardt GN, Zarbo RJ. Interinstitutional comparison of frozen section consultations: A College of American Pathologists Q-probes study of 90,538 cases in 461 institutions. Arch Pathol Lab Med 1997;120:804-9.  Back to cited text no. 6    
7.Novis DA, Zarbo RJ. Interinstitutional comparison of frozen section turnaround time: A College of American Pathologists Q-probes study of 32868 frozen sections in 700 hospitals. Arch Pathol Lab Med 1997;121:559-67.  Back to cited text no. 7    
8.Zarbo RJ, Hoffman GG, Howanitz PJ. Interinstitutional comparison of frozen - section consultation: A College of American Pathologists Q-Probe study of 79,647 consultations in 297 North American institutions. Arch Pathol Lab Med 1991;115:1187-94.  Back to cited text no. 8    
9.Novis DA, Gephardt GN, Zarbo RJ; College of American Pathologists. Interinstitutional comparison of frozen section consultation in small hospitals: A College of American Pathologists Q-probes study of 18,532 frozen section consultation diagnoses in 233 small hospitals. Arch Pathol Lab Med 1996;120:1087-93.  Back to cited text no. 9    
10.Parwani AV, Vrbin C, Raab S. Types of Frozen section errors and deferral rates in a sub-speciality based Academic Surgical Pathology Practice. Abstract presented at the 96 th annual Meeting of the United States and Canadian Academy of Pathology 2007.   Back to cited text no. 10    
11.Coffin CM, Spilker K, Zhou H, Lowichik A, Pysher TJ. Frozen section diagnosis in pediatric surgical pathology: A decade's experience in a children's hospital. Arch Pathol Lab Med 2005;129:1619-25.  Back to cited text no. 11    
12.Khoo JJ. An audit of intraoperative frozen section in Johor. Med J Malaysia 2004;59:50-5.  Back to cited text no. 12    
13.Nigrisoli E, Gardini G. Quality control of intraoperative diagnosis: Annual review of 1490 frozen sections. Pathologica 1994;86:191-5.   Back to cited text no. 13    
14.Intraoperative consultation (frozen section). In: Rosai J, editor. Rosai and Ackerman's Surgical Pathology. 9 th ed. vol 1. Mosby; 2004.p. 9-12.   Back to cited text no. 14    
15.Raab SS, Tworek JA, Soures R, Zarbo RJ. The value of monitoring frozen section - permanent section correlation data over time. Arch Pathol Lab Med 2006;130:337-42.  Back to cited text no. 15    

Top
Correspondence Address:
Yasmin Bhurgri
Department of Pathology and Microbiology, Aga Khan University, Stadium Road, P.O. Box 3500, Karachi 74800
Pakistan
Login to access the Email id


DOI: 10.4103/0377-4929.43733

PMID: 19008568

Get Permissions



    Figures

  [Figure 1(a)], [Figure 1(b)], [Figure 2(a)], [Figure 2(b)]
 
 
    Tables

  [Table 1], [Table 2]

This article has been cited by
1 Feasibility and clinical utility of intraoperative consultation with frozen section in osseous lesions
Murat Sezak,Basak Doganavsargil,Burcin Kececi,Mehmet Argin,Dundar Sabah
Virchows Archiv. 2012; 461(2): 195
[Pubmed]
2 Frozen section analysis of margins for head and neck tumor resections: reduction of sampling errors with a third histologic level
Stephen M Olson, Mohammad Hussaini, James S Lewis
Modern Pathology. 2011;
[VIEW]
3 Accuracy of frozen section diagnosis: a retrospective analysis of 4785 cases : ACCURACY OF FROZEN SECTION DIAGNOSIS
CHARLOTTE WINTHER, NIELS GRAEM
Apmis. 2011; 119(4-5): 259
[VIEW]
4 Intraoperative imprint cytological assessment of the subareolar tissue of the nipple areola complex (NAC)
Tomasovic-Loncaric, C., Milanović, R., Lambaša, S., Križanac, S., Štoos-Veić, T., Kaić, G., Ostović, K.T.
Collegium Antropologicum. 2010; 34(2): 431-435
[Pubmed]



 

Top
 
  Search
 
  
  
    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Email Alert *
    Add to My List *
* Registration required (free)  


    Abstract
    Introduction
    Material and Methods
    Results
    Discussion
    Conclusion
    References
    Article Figures
    Article Tables

 Article Access Statistics
    Viewed3237    
    Printed122    
    Emailed3    
    PDF Downloaded391    
    Comments [Add]    
    Cited by others 4    

Recommend this journal