Telecytopathology for Immediate Evaluation of Fine-Needle Aspiration Specimens

Telecytopathology for Immediate Evaluation of Fine-Needle Aspiration Specimens Mariam Alsharif, MD; Jamie Carlo-Demovich, MD; Caroline Massey, MD; James E. Madory, DO; David Lewin, MD; Ana-Maria Medina, MD; Rosemary Recavarren, MD; Patricia M. Houser, MHS, CT (ASCP); and Jack Yang, MD BACKGROUND: On-site evaluation of fine-needle aspiration (FNA) specimens by a pathologist is essential to obtain adequate samples and provide a preliminary diagnosis. Distance from the laboratory can make this difficult. The authors present their experience with on-site evaluation using telecytopathology. METHODS: Dynamic images of cytology smears were captured and processed with a Nikon digital camera system for microscopy and transmitted via Ethernet. A pathologist accessed the real-time images on a computer and interpreted them while communicating with on-site operators over the telephone. Sample adequacy and accuracy of preliminary diagnosis were compared with those obtained by regular on-site evaluation. RESULTS: A total of 429 telecytopathology cases and 363 conventional on-site cases were compared. Specimens were mainly from the pancreas, gastrointestinal tract, liver, and lymph nodes. Adequacy rate was 94.0% for telecytopathology and 97.7% for conventional cases. Preliminary diagnoses of unsatisfactory, adequate (defer), negative/benign, atypical, neoplasm, suspicious, and positive for malignancy were 6.3%, 13.5%, 14.9%, 17.9%, 7.2%, 8.6%, and 31.5% for telecytopathology and 3.9%, 30.6%, 21.5%, 9.6%, 5.0%, 5.2%, and 24.2% for conventional cases. Preliminary and final diagnoses were discrepant in 7 (1.8%) of 371 telecytopathology cases, and in 8 (3.1%) of 252 conventional cases. Difficulty was encountered in some cases in distinguishing pancreatic endocrine neoplasm from lymphoid proliferations, and low grade pancreatic tumors from chronic pancreatitis via telecytopathology. CONCLUSIONS: On-site evaluation of FNA specimens via telecytopathology assures sample adequacy and accurate preliminary diagnosis compared with the conventional method. It allows pathologists to use their time more efficiently and makes on-site evaluations at remote locations possible. Cancer (Cancer Cytopathol) 2010;118:119 26. VC 2010 American Cancer Society. KEY WORDS: fine-needle aspiration, telecytopathology, telecytology, telepathology, rapid on-site evaluation, adequacy assessment, preliminary diagnosis, endoscopic ultrasound guided fine-needle aspiration cytology, pancreas cytology. Corresponding author: Jack Yang, MD, Department of Cytopathology, Medical University of South Carolina, 165 Ashley Avenue, Suite 309, Charleston, SC 29425; Fax: (843) 792-9653; yangja@musc.edu Department of Pathology and Laboratory Medicine, Medical University of South Carolina, Charleston, South Carolina See editorial on pages 115-8, this issue. Received: December 15, 2009; Revised: January 13, 2010; Accepted: January 14, 2010 Published online: May 24, 2010 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/cncy.20074, www.interscience.wiley.com Cancer Cytopathology June 25, 2010 119

On-site evaluation of fine-needle aspiration (FNA) specimens by a pathologist is important to confirm sample adequacy, to triage specimens for other ancillary tests, and to establish a preliminary cytologic interpretation. Distance from the pathology laboratory and/ or pathologists offices and multiple locations for FNA procedures may prevent a timely immediate assessment and can make evaluation by an on-site pathologist difficult and time-consuming. Telepathology systems enable the electronic transmission of microscopic images online for consultation and discussion. They represent a new option for communication among pathologists and between pathologists and interventional radiologists located at distant sites. Dynamic telepathology, in which live images are transmitted and viewed electronically in real time at a remote site, showed encouraging results when used for frozen section consultation compared with original frozen diagnosis made on-site. 1,2 Progress in telepathology technology has made it feasible for pathologists to view cytologic smears remotely over an Internet connection to provide immediate assessment. A diagnostic concordance of 97% and an accuracy of 99% were reported in a recent study that compared the findings using a telepathology system with the original assessment of adequacy, and preliminary and final cytologic diagnoses, making it feasible to use this system for consultation between senior residents and pathologists located at a distance in academic institutions. 3 In this study, we summarize our experience using telecytopathology for immediate assessment of FNA specimens and compare it to conventional rapid on-site evaluation by a pathologist. MATERIALS AND METHODS Telecytopathology technology using a digital camera system for microscopy was introduced at our institution in the spring of 2008. We use it primarily for remote adequacy assessment and preliminary diagnosis of FNA biopsies obtained at our new Ashley-Rutledge Tower hospital located 2 blocks from the main medical university hospital, where the cytology laboratory and attending cytopathologist offices are located. FNA procedures are performed at different locations there, including interventional radiology, the endoscopy suite, and the operating room. A portable microscope with attached camera and monitor is used on location to evaluate these samples during the procedure. Cytologic smears are prepared on-site by the cytotechnologist and cytopathology fellow. Preparations for each pass include an air-dried Diff-Quik stained slide and an alcohol-fixed smear that is later stained with the Papanicolaou method in the laboratory. Additional material is obtained for cell-block, liquid-based preparation, and for further ancillary studies as needed. Our telecytopathology system is comprised of a digital camera, monitor, microscope, computer, and Internet connection. Dynamic images of the Diff-Quik smears are captured and processed with a Nikon digital camera system, Digital Sight series (Nikon Instruments, Melville, NY). The Digital Sight system uses a high-definition color camera head (DS-Fi1) that features a 5-megapixel charge-coupled device that can capture microstructures at a high resolution of 2560 1920 pixels, and a control unit (DS-L2) that features a built-in 8.4-inch liquid crystal display monitor. The standalone unit can be connected to a personal computer and functions as Web server for image distribution. The microscope is an Olympus CX 41 and uses objectives from 4 to 40 magnification (Fig. 1). The images are transmitted via ethernet and are accessible from any computer with Internet access (Figs. 2 and 3). A pathologist/cytopathologist interprets the cytologic images on a computer screen in their office in realtime while communicating with the operator over the telephone. The operator, usually a cytopathology fellow or senior resident, controls the microscope. He/she provides essential clinical information to the pathologist, and moves the slides on the stage to show the diagnostic fields. On the basis of this cooperation between them, the pathologist provides adequacy assessment and preliminary diagnoses that are conveyed by the fellow to the radiologist performing the procedure and are documented in the requisition form. The rates of sample adequacy, deferral, and preliminary and final diagnoses for cases diagnosed by telecytopathology were compiled in an Excel spreadsheet, and compared with those obtained by pathologists using conventional on-site microscopy at the main university hospital before telecytopathology. Two cytopathologists (J.Y and M.A.) performed the initial and final interpretations for the majority of the telecytopathology 120 Cancer Cytopathology June 25, 2010

Telecytopathology for On-Site Evaluation/Alsharif et al FIGURE 2. Adenocarcinoma of the pancreas is shown in a fine-needle aspiration specimen from a patient with a pancreatic mass, viewed remotely on a computer screen using the DS-L2 system (Diff-Quik). FIGURE 1. The telecytopathology Digital Sight DS-L2 system is shown, comprised of a Nikon camera and a monitor with a microscope. cases. The conventional cases were diagnosed by 5 other faculty pathologists. Diagnostic categories that were used in on-site evaluation included: unsatisfactory, negative for malignancy or benign, adequate/defer, atypical (including favor reactive), neoplasm (including spindle cell tumor, mucinous cystic neoplasm, and endocrine tumor), suspicious, and positive for malignancy. Accuracy was defined by agreement between the preliminary and final interpretation in the 2 groups. Cases with a preliminary diagnosis of benign or negative and a final diagnosis of neoplasm, suspicious, or positive for malignancy were considered discrepant. A preliminary interpretation of atypical cells with a final diagnosis of benign and/or reactive and conversely, an initial interpretation of benign with a final diagnosis of atypical cells, favor or probably reactive were not considered discordant when the clinical significance of the 2 results was the same, because the term atypical also encompassed reactive cellular changes. FIGURE 3. A pancreatic endocrine tumor is shown in an endoscopic ultrasound-guided fine-needle aspiration specimen from a patient with a pancreatic lesion (Diff-Quik). RESULTS There were 429 (250 men, 179 women; aged 22-93 years) consecutive cases evaluated by telecytopathology from March 1, 2008 to July 30 2009, and 363 (209 men, 154 women; aged 25-95 years) consecutive cases with conventional on-site evaluation from January 2 2007 to Cancer Cytopathology June 25, 2010 121

Table 1. Preliminary Cytology Diagnosis Made by Conventional Onsite Evaluation and Telecytopathology Preliminary Cytologic Diagnosis Conventional On-Site No. of Cases (%) Telecytopathology P Unsatisfactory 14 (3.9) 27 (6.3) >.05 Adequate, deferred 111 (30.6) 58 (13.5) <.001 Negative/benign 78 (21.5) 64 (14.9) <.05 Atypical epithelial cells 35 (9.6) 77 (17.9) <.05 Neoplasm (spindle 18 (5.0) 31 (7.2) >.05 cell/endocrine) Suspicious for malignancy 19 (5.2) 37 (8.6) >.05 Positive for malignant cells 88 (24.2) 135 (31.5) >.05 Total 363 (100) 429 (100) December 31, 2007. The anatomic sites sampled were similar in the 2 groups: pancreas was the predominant specimen, with the remainder coming from mediastinal and abdominal lymph nodes, stomach, liver, duodenum, bile duct, esophagus, adrenal gland, and rectum. In both groups, the preliminary cytologic diagnostic categories were similar (Table 1). Final cytologic diagnoses included unsatisfactory, negative for malignancy/benign, atypical, neoplasm (spindle cell, mucinous, or endocrine tumor), suspicious, and positive for malignancy (Table 2). The deferral rate was 13.5 % (58 of 429) for telecytopathology and 30.5% (111 of 363) for the conventional cases. When eliminating the deferred cases from the preliminary cytodiagnosis, the negative/benign category was lower in telecytopathology compared with the conventional group (64 of 371 [17.2%] vs 78 of 252 [30.9%]; P <.001). There were no significant differences in the final cytology diagnoses between the 2 groups, although there were slightly more unsatisfactory specimens in telecytopathology (Table 2), with a sample adequacy rate of 94.0% for telecytopathology and 97.8% for conventional microscopy. The discrepancy rate between the preliminary and final diagnosis was 7 (1.8%) of 371 for telecytopathology and 8 (3.1%) of 252 for the conventional group. Telecytopathology cases with discordant preliminary and final results are shown in Table 3. One discrepant case involved an endocrine tumor from the pancreas originally interpreted as negative for malignancy by telecytopathology because of scant neoplastic cells in the airdried smears evaluated on-site that resembled lymphocytes. On review, epithelioid cells with small to moderate amounts of granular cytoplasm and hyperchromatic Table 2. Final Cytology Diagnosis in Conventional and Telecytopathology Cases Final Cytologic Diagnosis Conventional On-Site No. of Cases (%) Telecytopathology Unsatisfactory 8 (2.2) 26 (6.0) >.05 Negative/benign 124 (34.1) 99 (23.1) <.05 Atypical cells 38 (10.4) a 43 (10.0) >.05 Neoplasm 33 (9.1) 58 (13.5) >.05 Suspicious for 10 (2.8) 14 (3.3) >.05 malignancy Positive for 150 (41.3) 189 (44.1) >.05 malignancy Total 363 (100) 429 (100) a Includes 2 cases with atypical lymphoid cells, suspicious for lymphoma. The remaining 36 cases were atypical epithelial cells. nuclei without nucleoli were present in the cell block sections, and these cells were reactive for cytokeratin and for the endocrine markers synaptophysin and chromogranin. A second discrepant case was a pancreas specimen with a preliminary diagnosis of benign that was interpreted as suspicious for malignancy in the final diagnosis. On review, highly atypical cells suspicious for malignancy were present in the alcohol-fixed Papanicolaou-stained slide, and only small numbers of these cells were identified in the rapid Diff-Quik stained slide from 1 pass. The remaining 3 passes from this aspirate showed benign gastrointestinal epithelial contaminant. Histological follow-up for this case showed adenocarcinoma in the pancreatic resection specimen. A third discordant case involved a celiac lymph node interpreted on-site as reactive lymph node, with a final P 122 Cancer Cytopathology June 25, 2010

Telecytopathology for On-Site Evaluation/Alsharif et al Table 3. Discrepancies Between Preliminary Interpretation and Final Cytologic Diagnosis for Telecytopathology Case No. Site Preliminary Interpretation Final Diagnosis Review and Additional Follow-up 1 Pancreas Benign Consistent with neoplasm, favor endocrine tumor (synaptophysinþ, chromograninþ, CKþ). Pancreatic endocrine tumor; no histology followup. 2 Pancreas Negative Suspicious for adenocarcinoma. GI epithelial contaminant with rare malignant cells in Diff-Quik, more in alcohol-fixed smear. Pancreas resection, adenocarcinoma. 3 Celiac lymph node Negative, reactive lymph node Rare large atypical cells in reactive lymph node. 4 Esophageal wall Nondiagnostic Rare atypical keratinized squamous cells. 5 Periduodenal Blood only, Renal cell carcinoma in cell block mass nondiagnostic and rare cells in ThinPrep slide. RCCþ, CD10þ, CKþ. 6 Peripancreas Rare spindle cells, nondiagnostic Rare mucinous cells with suggestion of papillary arrangements, cannot exclude mucinous neoplasm. Few single atypical cells may represent poorly differentiated carcinoma from stomach primary tumor. Biopsy, benign squamous mucosa. Concurrent core needle biopsy demonstrated renal cell carcinoma. Mucinous epithelium likely represents gastric mucosa. Repeat FNA was nondiagnostic. Patient has soft tissue neurofibromas; peripancreatic lesion is clinically suspected to be a soft tissue tumor. 7 Common bile duct Nondiagnostic Suspicious for adenocarcinoma Highly suspicious for adenocarcinoma in alcohol-fixed slide, rare in Diff-Quik slide; periportal lymph node demonstrated atypical glandular cells against a background of lymphoid tissue suspicious for metastatic adenocarcinoma. þ indicates positive; CK, cytokeratin; GI, gastrointestinal; RCC, renal cell carcinoma marker; FNA, fine-needle aspiration. diagnosis of rare atypical cells, cannot further characterize, in a background of reactive lymphoid cells. This patient had a history of signet-ring cell carcinoma of the duodenal bulb diagnosed in an outside biopsy. Follow-up histology of a stomach biopsy showed poorly differentiated adenocarcinoma with signet-ring cell features. On review of this case, the authors agreed that the rare atypical large cells in the aspirate from the celiac lymph node represent metastasis from the patients poorly differentiated adenocarcinoma of the stomach, presenting as single tumor cells admixed with background lymphoid tissue. Four additional cases that were considered nondiagnostic on-site by telecytopathology showed rare diagnostic cells in 1 or more of the accompanying cytology preparations, ThinPrep slides, and/or cell block sections at the time of case sign-out (Table 3). In 1 of these cases from an esophageal stricture, a few atypical parakeratotic squamous cells were identified in the final cytology. The slides from this case were not available for review, but a biopsy of the esophageal stricture showed benign squamous mucosa. The patient had dysphagia and a clinical concern for eosinophilic esophagitis. In another nondiagnostic sample from a periduodenal mass that showed blood only on-site, the cell block from this case revealed rare neoplastic cells with cytologic features consistent with renal cell carcinoma. This patient also had a concurrent core biopsy from the same site that showed renal cell carcinoma (Table 3). A third sample from a peripancreatic lesion showed rare spindle cells, nondiagnostic on-site with a final interpretation of mucinous epithelium, cannot exclude a mucinous neoplasm, recommend repeat FNA. The original material was not available for review, and a repeat FNA of the same lesion was nondiagnostic. There is no histology follow-up for this case. The fourth case with a preliminary diagnosis of nondiagnostic by telecytopathology was from a common bile-duct stricture with rare atypical glandular epithelium, suspicious for adenocarcinoma in the final cytology. On review, the representative cells were scant and mostly present in 1 alcohol-fixed Papanicolaou-stained smear. On a concurrent endoscopic ultrasound (EUS)-guided FNA of a periportal lymph node from this patient, both the preliminary and the final diagnoses were suspicious for metastatic adenocarcinoma. Cancer Cytopathology June 25, 2010 123

In the conventional group, 8 cases from the pancreas initially interpreted as benign had a final diagnosis of atypical glandular cells, not otherwise specified (NOS) in 3, mucinous neoplasm, NOS in 1, suspicious for adenocarcinoma in 2, and positive for adenocarcinoma in 2. The diagnostic cells were rare in the atypical and suspicious cases; they were mainly seen in the Thin- Prep and alcohol-fixed Papanicolaou-stained slides, respectively, in the 2 positive cases. Two of the cases with atypical glandular cells had follow-up histology that showed adenocarcinoma. A Whipple resection performed in the case of the mucinous neoplasm revealed adenocarcinoma. DISCUSSION Advancements in telepathology technology have made it possible to obtain pathology consultations remotely. Telepathology has proven to be a useful aid for intradepartmental frozen-section consultation. 4 When retrospective telepathologic examination of frozen sections representing various organ systems was compared with original glass slide diagnosis, agreements of 100% were obtained, validating this method for use in intraoperative diagnosis. 5 An analysis of published trials of telepathology in the areas of intraoperative frozen sections in routine and referral cases showed a diagnostic accuracy comparable to conventional microscopic diagnosis. 6 Wireless telepathology has also been used successfully for neuropathology subspecialty remote consultation. 7 In a study of 25 breast aspirates, intraobserver concordance rates of 80% to 96% were demonstrated between telecytology, using compressed digital images, and conventional glass slide diagnoses. 8 Alli et al found that intraobserver disagreement between digital and glass slide diagnosis of cervical smears was less than interobserver differences in reviewing glass slides; however, greater interobserver disagreements were observed in interpreting digital images. 9 Remote cytology consultation on 99 cases at the Armed Forces Institute of Pathology, using video microscopic images, showed 48% agreement with the contributors opinion, and only minor discrepancies were found between diagnoses made with telecytology and subsequent direct slide examination by the same consultant. 10 These data show that although static systems are affordable, they appear to be less accurate compared with dynamic systems. 11 This appears to be particularly true for telecytology, where the cytology specimen is thicker and has an increased depth of focus that would require a similar experience to observing the specimen under a microscope, which can be better achieved with video imaging in contrast to still images, thus improving the accuracy of diagnosis by this method. 11 Most studies on telepathology systems, however, including static, robotic, and more recently dynamic modes have centered on their application to remote frozen section diagnosis or expert opinions in histopathology, with few reports of their feasibility for use for cytologic specimens, and no data on their application for initial real-time on-site diagnosis in cytology. In the present study, we performed initial on-site evaluation of FNA specimens via a dynamic, live, and remotely operated telecytopathology system. We demonstrated that it provides similar results in assuring sample adequacy and diagnostic accuracy compared with onsite evaluation by a pathologist using conventional microscopy. We found a lower deferral rate with telecytopathology compared with the conventional method. A likely explanation for this is that the on-site pathologists in the conventional group verbally communicated their interpretations to clinicians, and such communications were not always transcribed in the requisition forms. Also, there might be some variations in practice seen between pathologists, even in the same group. Discrepancies in telecytopathology between on-site preliminary evaluation and final diagnosis were mostly minor, and they were mainly because of paucity of diagnostic material and the interpretive issues associated with characteristics of the EUS-guided FNA method and the resulting samples. In the case of the EUS-FNA from the pancreas interpreted as negative on-site and suspicious for adenocarcinoma in the final cytology, more careful screening by the trainee and oversight from the pathologist may have prevented missing the suspicious cells. In the case of the pancreatic endocrine tumor, misinterpretation resulted from scant numbers of tumor cells and a resemblance to lymphoid cells on initial evaluation. Distinguishing reactive or atypical lymphoid proliferations from endocrine tumors of the pancreas can be 124 Cancer Cytopathology June 25, 2010

Telecytopathology for On-Site Evaluation/Alsharif et al likewise problematic in routine microscopy smears, especially when the material is scant. Pancreas aspirates with rare or mildly atypical epithelial cells related to chronic pancreatitis or mild dysplasia of ductal epithelium can be difficult to interpret when the majority of the pancreatic malignancies are well-differentiated adenocarcinomas. This difficulty was also described in another study of lowgrade adenocarcinoma and chronic pancreatitis evaluated using static telecytopathology of EUS-FNA biopsies of the pancreas, even among experienced pathologists. 12 Specimens with abundant gastric mucosal contaminants are difficult to distinguish from the often low-grade mucinous neoplasms of the pancreas, which is a well-known pitfall. Therefore, special attention should be paid to the needle approach being used (transgastric or transduodenal). In 1 specimen from a peripancreatic lesion that was nondiagnostic, with rare spindled cells on-site, a mucinous neoplasm could not be excluded in the final cytology because of the identification of mucinous epithelium. Repeat FNA was also nondiagnostic. The clinical and radiographic impression in that case was that the peripancreatic lesion represented a soft tissue tumor such as a neurofibroma. Although no further histological follow-up is available for that case, the mucinous epithelium described most likely represented gastric foveolar cells on review. In addition to the presence of sufficient diagnostic material, experience with cytologic findings in EUSguided FNA specimens from the pancreas combined with knowledge of clinical and radiographic data are important factors in avoiding such diagnostic pitfalls. We did not encounter any false-positive cases using telecytopathology in our series. The on-site operator, usually a cytopathology fellow, plays a critical role in the success of telecytopathology in our experience. He/she scans the slides, conveys relevant clinical data to the pathologist, and shows them key findings during their communication over the telephone so that a timely and correct interpretation is reached during the FNA procedure. Slightly more cases with misinterpretations in telecytopathology were observed when the fellows were in the early part of their cytopathology fellowship training year compared with later in the year. These were mainly because of the on-site operators insufficient experience in cytology and on-site evaluation, especially in the pancreas EUS-FNA cases with scant cellularity or abundant gastrointestinal mucosal contaminants. Pathologist may need to spend more time in examining smears when working with a less experienced on-site operator, although in the present study we did not find a significant difference in performance among individuals because of an overall low error rate. There was no significant issue regarding the equipment, and most pathologists became acquainted with the real-time online images quickly. Similar observations were made in a study of 40 pancreatic lesions obtained by EUS-FNA when retrospectively re-evaluated by dynamic telecytopathology. 13 In that study, there was no significant difference in agreement between regular on-site microscopy and telecytopathology diagnoses, and fewer discordant cases occurred later within the study, when cytopathologists became more familiar with the system being used. 13 In the event that the fellow is held up at a location or absent or there is a second concomitant procedure at another location that requires immediate assessment, a senior resident or cytotechnologist could serve in their place. We think that residents with enough cytology training can efficiently use this system, and experienced cytotechnologists can become adept if adequately trained as well. It can also be used by most pathologists with experience in cytology and in particular EUS-FNA cytology. By using this technology, a pathologist may consult a colleague in the same vicinity on a difficult cytology case, during a procedure, by inviting them to view the images simultaneously. This may help to reach a consensus and minimize diagnostic errors. This modality also facilitates on-site evaluations at multiple sites by different pathologists, and it allows them to use their time more efficiently. We did not record the time spent for initial examination of FNA specimens via telecytopathology. In our experience, the time required for preliminary telecytodiagnosis is acceptable and comparable to that for regular on-site microscopy. In a recent study of 40 cases that were re-evaluated retrospectively by Kim et al, an average of 7.5 minutes (range, 2-20 minutes) was spent by the cytopathologist performing telecytopathologic diagnoses. 13 In that study, a robotic microscope was used and operated by the cytopathologist from an offsite location. Additional studies using similar telecytopathology systems as a substitute for on-site examination of FNA cytology specimens are needed. In summary, initial assessment of FNA samples by telecytopathology has a similar diagnostic accuracy to Cancer Cytopathology June 25, 2010 125

conventional on-site microscopy. Considering the low cost of the equipment (approximately $6000, excluding the microscope), telecytopathology is a cost-effective alternative that can be integrated into mainstream diagnostic cytopathology for adequacy assessment and preliminary diagnosis of FNA specimens to help support trainees and pathologists located at distant or multiple sites. CONFLICT OF INTEREST DISCLOSURES The authors made no disclosures. REFERENCES 1. Frierson HF Jr, Galgano MT. Frozen-section diagnosis by wireless telepathology and ultra portable computer: use in pathology resident/faculty consultation. Hum Pathol. 2007;38: 1330-1334. 2. Baak JP, van Diest PJ, Meijer GA. Experience with a dynamic inexpensive video-conferencing system for frozen section telepathology. Anal Cell Pathol. 2000;21:169-175. 3. Kerr SE, Bellizzi AM, Stelow EB, Frierson HF Jr, Policarpio-Nicholas ML. Initial assessment of fine-needle aspiration specimens by telepathology: validation for use in pathology resident-faculty consultations. Am J Clin Pathol. 2008;130:409-413. 4. Liang WY, Hsu CY, Lai CR, Ho DM, Chiang IJ. Lowcost telepathology system for intraoperative frozen-section consultation: our experience and review of the literature. Hum Pathol. 2008;39:56-62. 5. Kaplan KJ, Burgess JR, Sandberg GD, Myers CP, Bigott TR, Greenspan RB. Use of robotic telepathology for frozen-section diagnosis: a retrospective trial of a telepathology system for intraoperative consultation. Mod Pathol. 2002;15: 1197-1204. 6. Cross SS, Dennis T, Start RD. Telepathology: current status and future prospects in diagnostic histopathology. Histopathology. 2002;41:91-109. 7. WeinsteinRS,GrahamAR,RichterLC,etal.Overview of telepathology, virtual microscopy, and whole slide imaging: prospects for the future. Hum Pathol. 2009;40: 1057-1069. 8. Briscoe D, Adair CF, Thompson LD, et al. Telecytologic diagnosis of breast fine needle aspiration biopsies. Intraobserver concordance. Acta Cytol. 2000;44:175-180. 9. Alli PM, Ollayos CW, Thompson LD, et al. Telecytology: intraobserver and interobserver reproducibility in the diagnosis of cervical-vaginal smears. Hum Pathol. 2001;32: 1318-1322. 10. Allen EA, Ollayos CW, Tellado MV, et al. Characteristics of a telecytology consultation service. Hum Pathol. 2001;32: 1323-1326. 11. Yamashiro K, Taira K, Matsubayashi S, et al. Comparison between a traditional single still image and a multiframe video image along the z-axis of the same microscopic field of interest in cytology: which does contribute to telecytology? Diagn Cytopathol. 2009;37:727-731. 12. Marchevsky AM, Nelson V, Martin SE, et al. Telecytology of fine-needle aspiration biopsies of the pancreas: a study of well-differentiated adenocarcinoma and chronic pancreatitis with atypical epithelial repair changes. Diagn Cytopathol. 2003;28:147-152. 13. Kim B, Chieng DC, Crowe DR, et al. Dynamic telecytopathology of on site rapid cytology diagnoses for pancreatic carcinoma. Cytojournal. 2006;3:27. 126 Cancer Cytopathology June 25, 2010