SEARCH

SEARCH BY CITATION

Keywords:

  • fine-needle aspiration;
  • cytopathology;
  • consultation;
  • diagnostic error

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

BACKGROUND:

The importance of interinstitutional consultation (IC) has been documented across a variety of surgical pathology organ systems. However, to the authors' knowledge, few studies exist regarding this practice within cytopathology and specifically within fine-needle aspiration cytology (FNAC).

METHODS:

All FNAC cases between September 2002 and January 2007 were reviewed. Original diagnoses and second opinion diagnoses (SODs) were categorized as either no diagnostic disagreement, or minor diagnostic disagreement, or major diagnostic disagreement, and the latter was defined as either a 2-step deviation on a scale of “unsatisfactory, benign, atypical, suspicious, and malignant” or a change in treatment and/or prognosis. Outcome was determined by a review of the electronic medical record.

RESULTS:

Among 742 FNAC cases from outside laboratories, there were minor disagreements in 132 cases (17.8%) and major disagreements in 69 cases (9.3%) compared with the SODs from the authors' laboratory. Follow-up was available for 60 of 69 major discrepancies. The SOD was supported on follow-up in 65% of major discrepancies, and the initial diagnosis was supported better in 33% of major discrepancies. In 55% of cases in which the original institution diagnosis was supported better, either no case slides were received for examination or the slides contained material that was considered nondiagnostic by the authors. An SOD prompted a change in clinical management in 32 of 742 patients (4.3%). Aspirates that were most prone to change in management or therapy were from the thyroid (13 cases), neck (soft tissue and lymph nodes; 9 cases), salivary gland (2 cases), and liver (2 cases). Of 60 major diagnostic disagreements, board-certified cytopathologists rendered an SOD in 44 cases, and 75% of the diagnoses were supported better by follow-up, whereas pathologists who were not board certified in cytopathology had only 38% of SODs supported.

CONCLUSIONS:

Of 742 FNAC cases, 9.3% had major diagnostic disagreements; and, in 4.3%, patient management and therapy were altered. These results were similar to studies in surgical pathology emphasizing the importance of IC in FNAC. The authors concluded that FNAC IC benefits patient care. Cancer (Cancer Cytopathol) 2009. © 2009 American Cancer Society.

Pathology consultations generally are divided into either personal consultation or interinstitutional consultation (IC).1 IC is a practice whereby pathology material from an initiating hospital/laboratory is reviewed by at least a second laboratory before patients receive definitive therapy. This may be because the patient is sent for a major procedure at the referral hospital or because the patient seeks a second opinion from a physician at another clinic/hospital before or even after initiating a treatment program. The objective of IC in pathology is to detect major diagnostic errors that potentially could lead to improper or unnecessary patient management or therapy.

Several authors, including the Association of Directors of Anatomic and Surgical Pathology (ADASP), recommend a mandatory review of pathology reports and slides before definitive therapy is undertaken or withheld.1-4 Thus, many hospitals across the United States have policies that either require or encourage a mandatory second opinion of outside surgical pathology material before treatment is initiated. A study by Gupta and Layfield5 determined that, of the institutions that responded to their survey, 50% required pathology review of pathologic material before therapy, and an additional 38% encouraged the practice. Academic institutions were more likely than others to mandate or encourage this practice. At Ohio State University Medical Center, there are no hospital bylaws or mandates requiring such review, but the practice is strongly encouraged and accepted.

The topic of IC has been discussed and debated in the scientific literature as well as in the lay press.6-9 The ADASP estimates that 2% is an acceptable threshold for a clinically significant disagreement after arbitration when applied to those cases for which it is decided that the interpretation from the outside institution is the correct interpretation.2 Previous studies in general surgical pathology have indicated that major diagnostic error rates by referring laboratories range from 1.4% to 9.1%.3, 10, 11 In the past 2 decades, most studies have focused on identifying body sites that are more prone to diagnostic error. These include, but are not limited to, skin,12 soft tissue,13 gastrointestinal tract,14 gynecologic,15-20 nervous system,21, 22 lymph nodes,23 and genitourinary tract.4, 24

To our knowledge, few studies exist that analyze the effect of IC within cytopathology and specifically for fine-needle aspiration cytology (FNAC). In those studies that evaluated cytologic material, the posted results were similar to what has been reported for surgical pathology.7, 25-27 However, some authors have hinted that the rate of discrepant diagnoses in cytology may be higher than that of surgical pathology, possibly as high as 21%.10

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

We reviewed our files for all FNAC specimens that were sent to the cytology laboratory for IC from September 2002 to January 2007. Our hospital does not have mandatory review of slides before therapy, but the practice is strongly encouraged and generally followed. Cytopathology reports from the original hospital and from our laboratory were compared. We excluded outside cases if paperwork was missing and if the initial diagnosis (ID) could not be determined. Also, no cases were included in this study for which an ambiguous diagnosis was submitted by the outside laboratory. This meant excluding cases in which the original diagnosis was blank, in which the outside diagnosis stated “case sent for outside opinion or pending consultation,” in which only a differential diagnosis was given (eg, lymphoma vs small cell carcinoma), in which a diagnosis posed a question (eg, “adenocarcinoma?”), and in which the diagnosis was incomplete and thus was not possible to classify into 1 of 5 main categories (unsatisfactory, benign, atypical, suspicious, and malignant). In addition, we excluded those “personal” cases sent for a specific pathologist's opinion. Most IC cases came from community hospitals in Ohio and were sent to us because the patient was being referred to a specific surgeon or oncologist at our medical center either for treatment or for a second opinion before treatment. Five different attending pathologists rotated in the cytopathology laboratory during this period and rendered second opinion diagnoses (SODs) on these outside cases. Three of those pathologists were board certified specifically in cytopathology, and 2 were not, although they were board certified in anatomic pathology. The referring hospital ID was compared with our in-house diagnosis; then, cases were divided into 1 of 3 categories: 1) no diagnostic disagreement, 2) minor diagnostic disagreement (MIDD), and 3) major diagnostic disagreement (MADD). Our definition of MADD was similar to that of Lueck et al27; that is, either a 2-step deviation on a scale of “unsatisfactory, benign, atypical, suspicious, and malignant” or a diagnostic disagreement that, according to the authors, affected patient treatment or prognosis. An MIDD was defined as a 1-step deviation on this same scale without a change in patient treatment or prognosis. Clinicopathologic correlations were analyzed by a review of the electronic medical record with a minimum of 1-year follow-up to determine the clinical impact, if any, of diagnostic discrepancies, and whether a modification in treatment or prognosis occurred.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

We recovered 742 FNAC outside reports that were sent for IC from a 52-month timeframe (September 2002 to January 2007) and compared them with the in-house SODs. From these, 541 cases (72.9%) had no diagnostic disagreement (Table 1). MIDDs occurred in 132 cases (17.8%), and MADDs occurred in 69 cases (9.3%). Clinical/pathologic/radiologic follow-up was available for 60 of these 69 MADD cases. Our SOD was supported better on follow-up in 39 cases (65%), and the ID was supported better in 20 cases (33%).

Table 1. Comparison of Initial and Second Opinion Fine-Needle Aspiration Cytology Diagnoses and Relevance to Change in Patient Management and Therapy
ComparisonNo. of FNAC Diagnoses (%)No. With Probable Change in Patient Management or Therapy on Follow-Up Based on SOD (%)
  1. FNAC indicates fine-needle aspiration cytology; ID, initial diagnosis; SOD, second opinion diagnosis; NA, not applicable.

ID and SOD agree541 (72.9)0 (0)
ID and SOD agree, but follow-up supports neither10 (0)
ID and SOD with minor disagreement132 (17.8)0 (0)
ID and SOD with major disagreement, but follow-up supports SOD39 (5.3)32 (4.3)
ID and SOD with major disagreement, but follow-up supports ID20 (2.7)NA
ID and SOD with major disagreement, but follow-up supports neither1 (0.1)0 (0)
ID and SOD with major disagreement, but no follow-up9 (1.2)0 (0)
Total74232 (4.3)

In 1 MADD case (1.6%), neither the ID nor the SOD was correct. This example was an FNAC of a neck lymph node in which the ID was “squamous cell carcinoma,” and the SOD was “atypical squamous cells.” Subsequent neck dissection revealed metastatic malignant melanoma. In another case in which there was no disagreement, tissue follow-up supported neither the ID nor the SOD. Both sets of pathologists rendered specific malignant diagnoses on a parotid FNAC case, and both were incorrect. The ID was “malignant, suggestive of acinic cell carcinoma,” and the SOD was “nonsmall cell carcinoma, favor salivary duct carcinoma.” Surgical resection revealed another example of malignant melanoma.

The SOD prompted a change in clinical management in 32 of 742 FNAC cases (4.3%) that were submitted for IC (Table 2). These included thyroid (13 cases), neck (soft tissue and lymph node; 9 cases), salivary gland (2 cases), liver (2 cases), and other lymph node, pancreas, lung, breast, kidney/adrenal, and mediastinum (1 case each). The SOD in 24 of these 32 cases was either malignant (20 cases) or suspicious for malignancy (4 cases) (Table 3). The IDs that were issued in these cases were unsatisfactory (1 case), malignant (4 cases), negative (6 cases), and atypical (14 cases). In 8 cases, the SOD was deemed insufficient for diagnosis (2 cases), negative (5 cases), or atypical (1 case). These corresponded to IDs that were classified as neoplastic (1 case), suspicious (6 cases), or malignant (1 case). Analysis of the 20 cases in which the ID was supported better revealed that, in 11 cases (55%), either not all slides for the case were received (3 cases), or the submitted material was deemed hypocellular and was insufficient for diagnosis (8 cases) in our laboratory (Table 4).

Table 2. Number of Fine-Needle Aspiration Cytology Diagnoses by Organ Sites With Major Diagnostic Disagreements and Changes in Patient Management and Therapy
FNA SiteNo. of FNAC Diagnoses (%)
Total No.No. of Major DisagreementsProbable Change in Patient Management or Therapy Based on Second Opinion
  1. FNAC indicates fine-needle aspiration cytology; FNA, fine-needle aspiration.

Thyroid173 (23.3)28 (16.2)13 (7.5)
Neck: Lymph nodes and soft tissue136 (18.3)13 (9.6)9 (6.6)
Lung84 (11.3)5 (6)1 (1.2)
Liver78 (10.5)7 (9)2 (2.6)
Salivary gland70 (9.4)8 (11.4)2 (2.9)
Breast64 (8.6)1 (1.6)1 (1.6)
Soft tissue/bone56 (7.5)1 (1.8)0 (0)
Lymph nodes other than cervical lymph nodes45 (6.1)1 (2.2)1 (2.2)
Pancreas24 (3.2)2 (8.3)1 (4.1)
Kidney/adrenal6 (0.8)2 (33.3)1 (16.7)
Mediastinum6 (0.8)1 (16.7)1 (16.7)
Total74269 (9.3)32 (4.3)
Table 3. Major Diagnostic Disagreements in Which the Second Opinion Diagnosis Was Supported Better on Follow-Up and Led to a Probable Change in Management/Therapy (n=32 Patients)
Body SiteIDSODBC/ NBCOutcomeChange in Management/Therapy
  1. ID indicates initial diagnosis; SOD, second opinion diagnosis; BC, board certified in cytopathology; NBC, not board certified in cytopathology; PTC, papillary thyroid carcinoma; FNA, fine-needle aspiration; NNG, non-neoplastic goiter; SI, surgical intervention.

ThyroidMultinodular goiterPTCBCThyroidectomy with PTCSI
ThyroidFollicular neoplasmInsufficient for DiagnosisNBCMultiple subsequent FNAs all negativeRebiopsy
ThyroidSuspicious for follicular lesionBenignNBCClinically believed to be NNG; no surgery doneManagement
ThyroidAtypical cellsPTCBCThyroidectomy with PTCSI
ThyroidAtypical follicular cellsPTCBCPatient treated with radioactive iodine for PTCTherapy
ThyroidAtypical cellsPTCBCThyroidectomy with PTCSI
ThyroidSuspicious for follicular neoplasmNNGBCClinically believed to be NNG, no surgery doneManagement
ThyroidAtypical cellsAnaplastic thyroid carcinomaBCPatient sent to hospiceManagement
ThyroidSuspicious for follicular neoplasmHashimoto thyroiditisBCClinically believed to be thyroiditis and surgery withheldManagement
ThyroidAtypical cellsPTCBCThyroidectomy with PTCSI
ThyroidAtypical cellsPTCBCThyroidectomy with PTCSI
ThyroidAtypical cellsPTCBCThyroidectomy with PTCSI
ThyroidBenignSuspicious for PTCBCThyroidectomy with PTCSI
NeckHighly suspicious for malignancyInsufficient for diagnosisBCAll subsequent FNAs and tissue biopsies were benignRebiopsy
NeckAtypical parakeratosisSquamous cell carcinomaNBCNeck dissection with squamous cell carcinomaSI
NeckLymphoproliferative disorderAlveolar rhabdomyosarcomaBCResection with alveolar rhabdomyosarcomaSI
NeckEvaluation limited because of obscuring bloodSuspicious for carcinomaBCResection with squamous cell carcinomaSI
NeckRare atypical cellsSquamous cell carcinomaBCNeck resection with squamous cell carcinomaSI
NeckNo malignant cells identifiedSuspicious for small cell neuroendocrine carcinomaBCSubsequent FNA with immunohistochemistry supporting diagnosis of small cell neuroendocrine carcinomaRebiopsy
NeckNegativeSquamous cell carcinomaBCResection squamous cell carcinomaSI
NeckAtypical cellsSquamous cell carcinomaBCResection squamous cell carcinomaSI
NeckAtypical cellsSquamous cell carcinomaBCResection squamous cell carcinomaSI
Salivary glandAtypical cellsLarge cell carcinomaBCMucoepidermoid carcinomaSI
Salivary glandAtypicalNonsmall cell carcinoma, favor high-grade mucoepidermoid carcinomaBCHigh-grade mucoepidermoid carcinomaSI
LungNonsmall cell carcinoma, favor squamous cell carcinomaMetastatic melanoma, spindled typeBCTreated for metastatic melanomaTherapy
LiverNegative for malignancySuspicious for neuroendocrine carcinomaBCPatient considered unresectable because of multiple liver lesionsManagement
LiverAdenocarcinomaWell differentiated neuroendocrine carcinomaBCTissue biopsy revealed well differentiated neuroendocrine carcinomaRebiopsy
KidneyRenal cell carcinomaSingle atypical cell of undetermined significance; extensive, air-drying artifactNBCTissue biopsy revealed benign fibroconnective tissueRebiopsy
Lymph nodeNegative for malignancySuspicious for lymphomaBCBiopsy and flow confirmed lymphomaRebiopsy
PancreasAtypical signet ring cells, suspicious for malignancyNegative for malignancyBCClinically not believed to be malignant, patient improving with clinical follow-upManagement
BreastSuspiciousNegative for malignancy; fibrocystic changesBCClinically believed to be fibrocystic change; subsequent FNA benignManagement
Mediastinal massSlight atypiaSquamous cell carcinomaBCTreated with chemotherapyTherapy
Table 4. Major Diagnostic Disagreements in Which the Initial Diagnosis Was Supported Better on Follow-Up (n=20 Patients)
Body SiteIDSODDiagnostic Comment on SODBC/NBCFollow-Up
  1. ID indicates initial diagnosis; SOD, second opinion diagnosis; BC, board certified in cytopathology; NBC, not board certified in cytopathology; NNG, non-neoplastic goiter; PTC, papillary thyroid carcinoma; NNG, non-neoplastic goiter; DLBL, diffuse large B-cell lymphoma.

ThyroidSuspicious for follicular neoplasmNNG BCThyroidectomy with Hurthle cell adenoma
ThyroidSuspicious for PTCBenign follicular cells and colloid; scant cellularity precludes definitive interpretation.Scant cellularity; entire case slides not receivedBCThyroidectomy with PTC
ThyroidSuspicious for follicular neoplasmInsufficient for diagnosisInsufficient cellular materialNBCThyroidectomy with follicular carcinoma
ThyroidPTCAtypical NBCThyroidectomy with PTC
ThyroidPTCAtypical NBCThyroidectomy with PTC
ThyroidSuspicious for follicular neoplasmSuboptimal and insufficient for diagnosisInsufficient cellular materialNBCThyroidectomy with PTC
ThyroidFollicular lesionFew groups of benign follicular cells; not entirely satisfactoryInsufficient cellular materialBCThyroidectomy with minimally invasive follicular carcinoma
ThyroidSuspicious for neoplasmHashimoto thyroiditis BCThyroidectomy with PTC
NeckSquamous cell carcinomaAtypical squamous cells BCTissue biopsy revealed squamous cell carcinoma
NeckSuspicious for malignancyInsufficientInsufficient cellular material; only 2 of 4 slides were receivedBCTissue biopsy with DLBL
Salivary glandMalignant neoplasm with squamous featuresAtypicalOnly 1 of 4 slides receivedNBCResection with squamous cell carcinoma
Salivary glandSuggestive of pleomorphic adenomaBenign salivary gland; insufficient for diagnosisInsufficient cellular materialNBCResection with Warthin tumor
Salivary glandAcinic cell carcinomaInsufficient for diagnosisInsufficient cellular materialNBCResection with acinic cell carcinoma
LungAdenocarcinomaAtypical BCIt was believed clinically that the patient had metastatic adenocarcinoma to lung, and the patient was started on chemotherapy
LungSuspicious for nonsmall cell lung carcinomaNegative BCBiopsy with well differentiated neuroendocrine carcinoma
LungAtypical cellsNondiagnosticInsufficient cellular materialBCRebiopsy revealed Hodgkin lymphoma
LiverMetastatic adenocarcinomaAtypical NBCTissue biopsy with metastatic adenocarcinoma of pancreatic origin
LiverSmall cell neuroendocrine carcinomaPoorly differentiated adenocarcinoma BCTissue biopsy with small cell neuroendocrine carcinoma
AdrenalSuspicious for lymphomaInsufficient for diagnosisInsufficient cellular materialBCTissue biopsy with follicular lymphoma
RetroperitonealSuspicious for lymphomaNondiagnosticInsufficient cellular materialNBCTissue biopsy with DLBL

During the study period, 5 attending pathologists had rotating clinical assignments in cytology and were involved in the sign-out of outside FNAC cases. Three of these individuals had additional board certification in cytopathology, and 2 were board certified only in anatomic pathology. Of the 60 MADDs that had clinical/pathologic/radiologic follow-up, board-certified cytopathologists rendered an SOD in 44 cases, and 33 of those diagnoses (75%) were supported better by follow-up. Pathologists who were not board certified in cytopathology rendered an SOD in 16 cases and had only 6 (38%) of those diagnoses supported. In this group of 60 major disagreements, board-certified cytopathologists issued an SOD of negative, neoplastic, suspicious, or malignant in 35 of 44 cases (79.5%), an insufficient diagnosis in 7 cases (15.9%), and an atypical diagnosis in 2 cases (4.5%). Pathologists who were not board certified in cytopathology rendered an SOD of negative, neoplastic, suspicious, or malignant in 3 of 16 cases (18.8%), an insufficient diagnosis in 7 cases (43.8%), and an atypical diagnosis in 6 cases (37.5%).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

The recent interest in IC was sparked by a 1999 report released by the Institute of Medicine that focused on errors in medicine. It mandated a 50% reduction in the incidence of medical errors across all subspecialties in the 5 years after its publication.28 The report was a catalyst that generated tremendous interest in medical errors and led to discussions and debates in the scientific literature as well as the media.6-9 IC is a complex issue with many challenges and no clear-cut answers. It is intended to reduce error by a strategy known as redundancy, which is used in both medical and nonmedical fields.29 At the forefront of obstacles in this strategy is the subject of added cost. In today's era of decreasing reimbursement and cost containment, third-party payers are likely to be reluctant to finance a second opinion.

Many studies have demonstrated the importance of IC. Some have demonstrated that the cost associated with IC is less than the cost of improper or unnecessary therapeutic intervention and potential human morbidity.4, 10 Several organizations have developed guidelines for IC with studies to review the body sites and organ systems most prone to error.1-3, 10-24 By identifying these sites, IC may have a significant impact on patient management and therapy. Then, clearer guidelines regarding consultation and review may be synthesized focusing on these organ systems. In this regard, thyroid, breast, lymph node, and lung are common sites at which a cytology diagnosis alone may lead to surgical intervention or chemotherapy/radiation therapy.26 Because most of our IC cases originated from community hospitals, this sampling may have been biased toward superficial targets rather than deep-seated lesions for which FNAC may be used less often. Our study demonstrated that thyroid FNAC (7.5% of total thyroid cases) and neck FNAC (soft tissue and lymph node; 6.6% of total neck cases) were the principal sites of MADD (Table 2). The 2007 National Cancer Institute-sponsored State of the Science Thyroid Conference (available at: http://thyroidfna.cancer.gov accessed March 2009) highlighted the importance and exalted status that thyroid FNAC recently has acquired. Widespread adoption of the diagnostic terminology, classification scheme, and adequacy statements published by that conference undoubtedly should help to diminish MADD cases for thyroid FNAC among pathologists as we go forward.30 It also is noteworthy that breast aspirates composed such a low number of IC cases, and only 1 case produced an MADD. Our community hospital referral base seems to mirror our own in-house experience with breast FNAC, in that the breast has become a minor component of our FNAC volume during the past several years. A study of cases from a 5-year period before 2002 might reveal a greater number of aspirates and discrepancies from this site.

Cost plays a role in a clinician's attempt to secure a diagnosis. Clinicians try to obtain a pathologic diagnosis in the least invasive, most cost-effective manner, which often results in increased pressure placed on pathologists to issue a diagnosis with minimal tissue or cellular material. This may contribute to a further increase in error rates. Another issue remains with the definition of error. Many authors express the difficulty and challenges of defining error in anatomic pathology.7, 29, 31-33 This stems from a lack of scientifically validated gold standards with which to compare differing diagnoses in anatomic pathology. Sirota pointed out that reliance on precision serves as an indication of the correct diagnosis.31 That is, precision, or low interobserver variability, serves as a surrogate for accuracy. However, as many investigators have reported, unfortunately, diagnostic variability is equated with error in surgical pathology.7, 29, 31, 33 Nonetheless, most diagnostic variation does not lead to a significant change in patient management or therapy. Our current study supports this notion. Although 132 SODs (17.8%) produced minor disagreements with the ID, none had an effect on patient management or therapy. In addition, even major discrepancies may not lead to a significant change in management. Of the 60 MADDs that had follow-up available, 28 (47%) did not significantly affect patient management or therapy regardless of whether the ID or the SOD was the correct diagnosis.

In more instances, the SOD was more accurate, because it was supported better on tissue or clinical follow-up. However, the results of the current study demonstrate that this is not always the case. In our series, the ID was supported better in approximately 33% of cases, whereas the SOD was supported better almost twice as often (65% of cases) when a MADD existed. Rarely, both the ID and the SOD are incorrect even if they are in complete agreement or if they reveal major disagreement. The current series included a single example of each of these circumstances in which metastatic malignant melanoma foiled all attempts at a correct diagnosis in 2 separate instances.

Published error rates in surgical pathology range between 1.4% and 9.1%.3, 10-11 The study by Abt et al suggests that cytology may have error rates as high as 21%.10 Our FNAC series uncovered MADDs in 9.3% of cases. In 4.3% of 742 cases, a change in patient management and therapy occurred because of the MADD. These numbers are similar to those from the surgical pathology literature. When analyzing the MADDs that led to a change in patient management and therapy, an SOD was malignant in 20 cases, whereas the most likely corresponding diagnosis issued by the originating institution was “atypical” (14 cases) (Table 3). In only 1 case did the SOD issue a diagnosis of “atypical” without further qualification. Thus, the originating laboratory was more likely to be reticent to issue a definitive diagnosis for this group. Nevertheless, no consistent pattern of diagnostic thresholds emerges when comparing our single academic center with community laboratories in this study. Table 4 shows that, in contrast to Table 3, the ID was more likely to issue atypical, suspicious, or definitively malignant diagnoses compared with the ID in 95% of MADD cases. Combining both tables illustrates that, in only slightly more than half of all MADD cases (27 of 52; 52%), the ID was listed as atypical or greater compared with the SOD. Therefore, our study does not strongly support a broad conclusion that a community hospital-based cytology laboratory tends to have a lower diagnostic threshold and, thus, is more likely to make an atypical/suspicious/definitive malignant diagnosis than an academic medical center-based cytology laboratory. Clinical input may have influenced the issuance of more examples of suspicious/definitively malignant diagnoses in the ID cases from Table 4, as discussed below.

FNAC provides unique challenges, particularly in determining adequacy for different body sites and the challenges of separating different diagnostic categories (ie, unsatisfactory, benign, atypical, suspicious, and malignant), especially in 1-step differences (ie, atypical vs suspicious). Only a few organ sites have well established diagnostic criteria for defining these terms, although, as stated above, the National Cancer Institute's terminology for thyroid should help in this regard. Moreover, FNAC is a procedure that is highly dependent on the skill of the individual who is performing the procedure and the ability of that individual to craft readable, well preserved smears that contain sufficient numbers of diagnostic cellular material that lacks obscuring debris, excessive blood, air-drying artifacts, and other technical artifacts. In the current study, there were 20 cases from the MADD category in which the diagnosis from the originating referral institution was supported better (Table 4). In 11 of those cases (55%), either a review of the original pathology report demonstrated that the total number of slides that existed in the case was less than our laboratory received to review (3 cases), or we deemed that the slides had insufficient cellularity for diagnosis (8 cases). It is unknown whether these “missing” slides contained either all diagnostic cells or more of cells that were necessary to substantiate the ID. It is possible that, if all slides were received or if an adequate specimen was obtained, then a better supported SOD would have been rendered. The ADASP has established recommendations for IC, and 1 of these is for the original institution to send material adequate to solve the diagnostic problem.1 In addition, and perhaps more noteworthy, clinicians/radiologists at the originating institution may have communicated vital clinical or radiologic information to the initial pathologist to enable a more specific and definitive ID on material that we considered of borderline or insufficient cellularity or on smears that we believed were prepared poorly and were not satisfactory for definitive interpretation. It is highly likely that such information was not transferred to us along with the slides, stressing the importance of close communication not only between the pathologist and the clinician but also between pathologists from different institutions rendering diagnoses on the same material—especially when faced with an MADD. It is this discussion that may help reduce error and uncertainty by addressing whether repeat biopsy and segregation for ancillary testing (immunohistochemistry, flow cytometry, molecular diagnostics) would provide additional benefit or whether the current material is adequate in light of the clinical situation. That stated, we have no record to determine whether our laboratory made a concerted effort to contact and speak with the pathologist involved in the ID when a case seemed to have a less than complete set of slides or when our SOD was nondiagnostic.

Although it was limited to 5 individuals, the results our study suggest that completing a cytology fellowship and passing the subspecialty board examination may provide an additional level of diagnostic accuracy/confidence. Boiko and Piepkorn34 demonstrated a difference between general pathologists and dermatopathologists in recognizing skin lesions such as early melanoma. A similar situation may hold true for cytopathology and for FNAC in particular. The 3 pathologists in our group who were board certified in cytopathology were more likely to issue a more definitive diagnosis than those who were not board certified in cytopathology (79.5% vs 18.8%, respectively) and were more likely to use the “atypical” category far less frequently (4.5% vs 37.5%, respectively). In general, it appears that pathologists who are not board certified in cytopathology are more likely to issue a generic diagnosis rather than a specific diagnosis. In addition, our laboratory does not have a policy mandating that 2 or more cytopathologists independently review all outside cases when there is disagreement with the outside cytologic diagnosis. If such a policy had been in place and had been enforced consistently, then this may have increased our accuracy in SOD. Additional studies are needed to assess whether and to what extent having subspecialty board certification and requiring “multiple cytopathologists review” of outside FNAC cases would help decrease error rates.

In summary, of 742 FNAC cases, 9.3% resulted in an MADD; and, in 4.3%, patient management and therapy were altered. These results were similar to studies in surgical pathology emphasizing the importance of IC in FNAC. We believe that IC of FNAC material benefits patient care.

References

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References
  • 1
    Association of Directors of Anatomic and Surgical Pathology. Consultations in surgical pathology. Am J Surg Pathol. 1993; 17: 743-745.
  • 2
    Association of Directors of Anatomic and Surgical Pathology. Recommendations on quality control and quality assurance in anatomic pathology. Am J Surg Pathol. 1991; 15: 1007-1009.
  • 3
    Kronz JD, Westra WH, Epstein JI. Mandatory second opinion surgical pathology at a large referral hospital. Cancer. 1999; 86: 2426-2435.
  • 4
    Epstein JI, Walsh PC, Sanfilippo F. Clinical and cost impact of second-opinion pathology. Review of prostate biopsies prior to radical prostatectomy. Am J Surg Pathol. 1996; 20: 851-857.
  • 5
    Gupta D, Layfield LJ. Prevalence of inter-institutional anatomic pathology slide review. Am J Surg Pathol. 2000; 24: 280-284.
  • 6
    Parker-Pope T. Risk of error may justify second opinion on pathology reports. Wall Street Journal. April 13, 2001: 1.
  • 7
    Frable WF. Surgical pathology—second reviews, institutional reviews, audits, and correlations. Arch Pathol Lab Med. 2006; 130: 620-625.
  • 8
    Parker-Pope T. Why it's hard to get a second opinion (and how to make sure you get one). Wall Street Journal. February 1, 2005: 1.
  • 9
    Tarkan L. Value of second opinions is underscored in study of biopsies [Health Section]. New York Times. April 4, 2000: D7 20.
  • 10
    Abt AB, Abt LG, Olt GJ. The effect of interinstitution anatomic pathology consultation on patient care. Arch Pathol Lab Med. 1995; 119: 514-517.
  • 11
    Manion E, Cohen MB, Weydert J. Mandatory second opinion in surgical pathology referral material: clinical consequences of major disagreements. Am J Surg Pathol. 2008; 32: 732-737.
  • 12
    Ackerman AB. Discordance among expert pathologists in diagnosis of melanocytic neoplasms. Hum Pathol. 1996; 27: 1115-1116.
  • 13
    Arbiser ZK, Folpe AL, Weiss SW. Consultative (expert) second opinions in soft tissue pathology. Analysis of problem-prone diagnostic situations. Am J Clin Pathol. 2001; 116: 473-476.
  • 14
    Hahm GK, Niemann TH, Lucas JG, et al. The value of second opinion in gastrointestinal and liver pathology. Arch Pathol Lab Med. 2001; 125: 736-739.
  • 15
    Jacques SM, Qureshi F, Munkarah A, et al. Interinstitutional surgical pathology review in gynecologic oncology. I. Cancer in endometrial curettings and biopsies. Int J Gynecol Pathol. 1998; 17: 36-41.
  • 16
    Jacques SM, Qureshi F, Munkarah A, et al. Interinstitutional surgical pathology review in gynecologic oncology. II. Endometrial cancer in hysterectomy specimens. Int J Gynecol Pathol. 1998; 17: 42-45.
  • 17
    Khalifa MA, Dodge J, Covens A, et al. Slide review in gynecologic oncology ensures completeness of reporting and diagnostic accuracy. Gynecol Oncol. 2003; 90: 425-430.
  • 18
    Chafe S, Honore L, Pearcey R, et al. An analysis of the impact of pathology review in gynecologic cancer. Int J Radiat Oncol Biol Phys. 2000; 48: 1433-1438.
  • 19
    McGowan L, Norris HJ. The mistaken diagnosis of carcinoma of the ovary. Surg Gynecol Obstet. 1991; 173: 211-215.
  • 20
    Selman AE, Niemann TH, Fowler JM, et al. Quality assurance of second opinion pathology in gynecologic oncology. Obstet Gynecol. 1999; 94(1 suppl): 302-306.
  • 21
    Inouye L, Langford LA, Fuller GN, Bruner JM. Diagnostic discrepancies in a neuropathology consultation practice: 500 consecutive cases. Mod Pathol. 1996; 9(1 Suppl): 165A.
  • 22
    Bruner JM, Inouye L, Fuller GN, Langford LA. Diagnostic discrepancies and their clinical impact in a neuropathology referral practice. Cancer. 1997; 79: 796-803.
  • 23
    LaCasce AS, Kho ME, Friedberg JW, et al. Comparison of referring and final pathology for patients with non-Hodgkin's lymphoma in the National Comprehensive Cancer Network. J Clin Oncol. 2008; 26: 1-6.
  • 24
    Wurzer JC, Al-Saleem TI, Hanlon AL, et al. Histopathologic review of prostate biopsies from patients referred to a comprehensive cancer center. Cancer. 1998; 83: 753-759.
  • 25
    Layfield LJ, Jones C, Rowe L, et al. Institutional review of outside cytology materials: a retrospective analysis of 2 institutions' experiences. Diagn Cytopathol. 2002; 26: 45-48.
  • 26
    Baloch ZW, Hendreen S, Gupta PK. Interinstitutional review of thyroid fine-needle aspirations: impact on clinical management of thyroid nodules. Diagn Cytopathol. 2001; 25: 231-234.
  • 27
    Lueck N, Jensen CS, Cohen MB, et al. Mandatory second opinion in cytopathology referral material [abstract]. Mod Pathol. 2008; 21(1 suppl): 356A.
  • 28
    KohnLT, CorriganJM, DonaldsonMS, eds. To Err is Human: Building a Safer Health System. Washington, DC: National Academy Press; 1999.
  • 29
    Tomaszewski JE, Bear HD, Connally JA, et al. Consensus conference on second opinions in diagnostic anatomic pathology. Who, what, and when. Am J Clin Pathol. 2000; 114: 329-335.
  • 30
    Baloch ZW, LiVolsi VA, Asa SL, et al. Diagnostic terminology and morphologic criteria for cytologic diagnosis of thyroid lesions: a synopsis of the National Cancer Institute thyroid fine-needle aspiration state of the science conference. Diagn Cytopathol. 2008; 36: 425-437.
  • 31
    Sirota RL. Defining error in anatomic pathology. Arch Pathol Lab Med. 2006; 130: 604-606.
  • 32
    Murphy WM, Rivera-Ramirez I, Luciani LG, Wajsman Z. Second opinion of anatomical pathology: a complex issue not easily reduced to matters of right and wrong. J Urol. 2001; 165( 6 pt 1): 1957-1959.
  • 33
    Foucar E. Error in anatomic pathology. Am J Clin Pathol. 2001; 116( suppl): S34-S46.
  • 34
    Boiko PE, Piepkorn MW. Reliability of skin biopsy pathology. J Am Board Fam Pract. 1994; 7: 371-374.