Prospective step sections for small skin biopsies.
Objective.--To examine the utility of these "prospective" step sections in improving diagnostic accuracy and turnaround time.
Design.--Five hundred consecutive cases, in which step sections had been cut prior to slide review, were studied. For each specimen, 3 slides, each consisting of 1 ribbon of tissue containing 4 to 6 sections, were obtained at 50-[micro]m intervals from the paraffin block.
Results.--Fifty-eight biopsies (12%) were nondiagnostic using slide 1 alone. Step sections provided a diagnosis in 19 of 58 cases. In an additional 15 cases (3%) in which a diagnosis was possible using slide 1, deeper levels resulted in a change in diagnosis. Thus, in 34 (7%) of 500 biopsies, deeper levels resulted in improved diagnostic accuracy. In addition, the pathologist would have ordered step sections in a further 117 cases (23%) to clarify the diagnosis rendered on level 1 or to exclude other lesions. Thus, 30% of small skin biopsies would have required deeper levels if step sections had not been obtained prior to slide review.
Conclusions.--In our laboratory, the use of prospective step sections is essentially cost-neutral and case turnaround time is improved by 9% to 45%. Step sections result in a changed diagnosis in 7% of small skin biopsy specimens.
Sampling error, because of incomplete sectioning through the tissue block, is a known cause of false-negative diagnosis in pathology. Deeper sectioning of small biopsy specimens, such as needle core and endoscopic biopsies, is commonly used to enhance diagnostic sensitivity and accuracy. (1-4) In dermatopathology, studies suggest that deeper levels provide a more accurate diagnosis in about one third of skin biopsy specimens. Additional levels are most helpful in the diagnosis of skin cancer. (5-7) Standards for handling of small skin biopsy specimens have not been established, (8) and indeed there is great variability in how dermatopathology laboratories deal with the problem of nondiagnostic initial slides. (9)
Diagnostic laboratories must balance the utility of deeper levels with the additional time required and expense incurred and the anticipated impact on patient care. Traditionally, deeper levels are obtained, at the request of the pathologist, after the original slides have been reviewed. The usefulness of preparing these levels prior to receipt of slides by the pathologist has not been rigorously assessed. In this study, we examine the hypothesis that for small skin biopsies or curettings, preparation of 3 slides prior to review by the pathologist ("prospective" step sections) results in increased diagnostic accuracy and improves turnaround time (TAT).
MATERIALS AND METHODS
We examined 500 consecutive skin biopsy specimens for which prospective step sections had been prepared. Specimens were limited to curettings and small biopsies, typically punch biopsies or small shave specimens, that could not be bisected ([less than or equal to] 0.3 cm in diameter). This represented 23% of cases submitted to our community-based dermatopathology laboratory during a 1-month period. For each specimen, 3 slides, each consisting of 1 ribbon of tissue containing 4 to 6 sections, were obtained at 50-[micro]m intervals from the paraffin block. The use of 3 slides is arbitrary but represents the current practice in our laboratory.
A diagnosis was rendered using slide 1 only, and subsequently slides 2 and 3 were reviewed. The utility of slides 2 and 3 was classified as no additional information, more accurate diagnosis, malignant neoplasm excluded, benign neoplasm or inflammatory condition diagnosed, or malignant neoplasm diagnosed.
We used a time-motion, task-analysis approach to assess the technical labor costs of prospective step sections versus conventional, "retrospective" step sections (prepared at the request of the pathologist after the original slide has been examined). Costs for materials and capital equipment were also estimated.
Fifty-eight of 500 biopsies (12%) were nondiagnostic using slide 1 alone. Deeper levels provided a diagnosis for 19 (33%) of 58 originally nondiagnostic biopsies (Table 1). Figure 1 illustrates nondiagnostic material on the original (Figure 1, a) and second (Figure 1, b) slide, with basal cell carcinoma seen only on the third slide (Figure 1, c). Fifteen (3%) of 442 biopsies, which were diagnostic on slide 1, had a change in diagnosis after step sections (Table 2). Thus, step sections resulted in improved diagnostic accuracy in 34 biopsies (7%). For most of these biopsies (25/34), the improved diagnostic accuracy was achieved using slide 2 alone. Slide 3 was required in 9 of 34 biopsies (Tables 1 and 2).
[FIGURE 1 OMITTED]
In 117 (23%) of 442 biopsies for which a diagnosis was possible on slide 1, pathologists would have ordered additional levels to confirm or clarify the diagnosis. These additional levels were not absolutely required for interpretation but provided the pathologist with additional information resulting in increased diagnostic certainty.
Therefore, to improve diagnostic sensitivity or to confirm/clarify the initial diagnosis rendered using slide 1, 151 (30%) of 500 biopsies required additional deeper levels.
The use of prospective step sections would be expected to improve TAT, because there is no requirement to wait until further slides are prepared. We use overnight tissue processing and our intralaboratory TAT for routine skin biopsies is 20 to 28 hours. Retrospective step sections, when requested on a routine, nonurgent basis, add 16 to 22 hours to the TAT of an individual case. Thus, the TAT for cases requiring retrospective skin biopsies is 36 to 50 hours. If all small biopsies required step sections, then the range in TAT improvement realized by using prospective sections would be 29% (36 hours vs 28 hours) to 250% (50 hours vs 20 hours). However, because, in our laboratory, only 30% of small skin specimens require deeper levels, the range of expected TAT improvement using prospective step sections is 9% to 45%. Anticipated TAT improvements therefore depend on laboratory TAT for normal specimens and the time required to produce retrospective step sections. A model for this relationship is shown in Figure 2.
[FIGURE 2 OMITTED]
We compared the technical time required to prepare step sections in a prospective versus a retrospective manner. Preparation of retrospective levels required 1.63 times longer per biopsy (13 minutes vs 8 minutes), because additional time was needed to retrieve the block, label slides, and mount and section the block. We calculated technical costs for each step of the process using relevant salary rates for laboratory assistants and laboratory technologists.
Because only 30% of small biopsies/curettings require step sections for definitive diagnosis, for 500 cases, retrospective step sections would be ordered on only 150 cases.
In our laboratory, prospective step sections on 500 cases have a technical cost of $1595, whereas if retrospective sections are used, the total cost is $1540. Thus, prospective sections only incur a small (3.5%) increase in technical labor costs.
Apart from labor costs, generation of additional slides does incur added supply costs. These include costs for solutions and stains, equipment use, glass slides, cover slips, and requirement for increased slide storage space. We estimated that the processing, materials, and storage cost for prospective step sections (500 biopsies; 1500 slides) versus retrospective sections (500 biopsies; 800 slides) is $145.
Deeper sections are routinely used in most dermatopathology laboratories. (9) Our study confirms previously published results; approximately 30% of small skin biopsies require deeper levels for maximum diagnostic accuracy. (5-7)
In some laboratories, step sections (or "deeper levels") are prepared prior to receipt of the slides by the pathologist (prospective step sections). The use of prospective step sections may improve diagnostic accuracy and TAT, but to our knowledge, this approach has not been formally evaluated. Our results show that prospective step sections, on small, nonbisectable skin biopsies or curetted material, result in very minor increased technical costs and can improve report TAT.
Step sections improved diagnostic sensitivity and accuracy in 7% of cases, but, in our practice, we routinely use additional levels in up to 30% of small skin biopsies to confirm, clarify, or rule out certain diagnoses. The use of additional sections for biopsies on which a diagnosis is already possible on slide 1 depends significantly on clinical suspicion, biopsy quality, and pathologist experience and will likely vary significantly among laboratories and among pathologists. The potential cost effectiveness of prospective step sections therefore depends on the current use of retrospective step sections in a given laboratory. If step sections are rarely ordered, then prospective step sections on all small skin biopsies will significantly add to technical workload and slide volumes. If step sections are requested very frequently, then prospective sections are likely to result in a significant cost saving.
If the original sections are not obtained from deep enough within the paraffin block, then further deeper levels (step sections) would be expected to improve diagnostic yield and accuracy. In the study by Carag et al (6) step sections (10 slides at 50-[micro]m intervals) revealed new diagnostic findings in 33% of biopsies with an initial diagnosis of actinic keratosis. In our study, step sections (3 slides at 50-[micro]m intervals) resulted in new findings in 7% of small skin biopsies or curettings. This result suggests that our technologists are cutting initial sections from sufficient depth in the block. There was no significant difference observed in the size of tissue section in slide 1 compared with slide 3.
For the 34 specimens (7%) in which diagnostic accuracy was improved with deeper levels, new histologic findings were apparent on slide 2 in 25 biopsies, and slide 3 was only required in 9 biopsies (Tables 1 and 2). Nevertheless if slide 3 had not been obtained, several clinically important diagnoses would have been missed (squamous cell carcinomas in situ , dysplastic nevus , and a severely dysplastic nevus ). The study design did not specifically examine the question of how many deeper levels are required for diagnosis on small skin biopsies. Indeed, in 39 biopsies, a diagnosis was not possible using 3 slides. If clinically indicated, additional slides (usually 3-10) were cut from the block. In most cases (26/39), examination of the additional slides did not allow a specific diagnosis. These deeper levels revealed 2 cases of actinic keratosis, 2 cases of spongiotic dermatitis, and a variety of benign lesions including viral wart, seborrheic keratosis, solar lentigo, and benign fibrous papule. There were no malignant lesions discovered on examination of the additional material in these 39 biopsies. We have not incorporated the impact of such cases in our model for TAT assessment. Clearly, such biopsies are problematic and best handled on a case-by-case basis incorporating information on suspected clinical diagnosis, biopsy size, and histologic features observed in the original levels.
Of the 117 biopsies for which a diagnosis was possible on slide 1 and for which pathologists would have ordered additional levels, none absolutely required these extra levels to make a diagnosis. Instead, the levels provided the pathologist with additional information resulting in increased diagnostic certainty, often allowing a more definitive diagnostic statement as opposed to "consistent with" or "suspicious for" in the diagnostic line. We did not formally evaluate the utility of slide 2 versus slide 3 in this category of biopsies.
The use of 3 prospective step sections resulted in improved TAT. This improvement is directly dependent on the overall TAT for skin biopsy specimens and the time required to prepare retrospective step sections. In laboratories with rapid TAT and deeper levels available within hours, the impact of prospective sections is less significant. However, in other settings the improvement in TAT may be considerable. For example, in a laboratory with a routine TAT of 24 hours, the requirement for an additional 16 hours (overnight TAT for step section requests) on 30% of small skin biopsies would result in a 20% increase in overall TAT.
The cost-effectiveness of this approach depends on the amount of extra time spent by the technologist to produce conventional, retrospective step sections (ie, computer login of workload, retrieval of the paraffin block, and labeling and cutting of the sections). Prospective step sections obviously do not require any of these additional procedures. Our analysis shows that this approach is likely to be very close to cost-neutral in terms of labor costs. However, there are increased supply costs for prospective sections (staining materials, glass slides, etc) and the costs for an increased storage volume (glass slides) must be considered. The modest increase in operational costs must be balanced against the improved diagnostic accuracy and potential improvement in report TAT. Improved diagnostic accuracy using step sections is most often obtained using slide 2 (25/34 biopsies), but we have not compared the costs of preparing 2 slides prospectively versus 3 slides. Labor costs are expected to be similar, with a minor reduction in material costs. We have not rigorously assessed the effect of prospective step sections on the time a pathologist needs to report a skin biopsy case. In our experience, the increased time required to examine 3 slides versus 1 slide is minimal, and indeed the use of prospective step sections results in increased diagnostic confidence and efficiency.
In summary, step sections on small skin biopsies and curettings increase diagnostic accuracy and prospective sections result in an improved TAT, with minimal cost increase.
Financial support was provided by a grant from the Calgary Laboratory Services Health Services Research Funding Competition.
Accepted for publication June 9, 2006.
(1.) Lane RB Jr, Lane CG, Mangold KA, Johnson MH, Allsbrook WC Jr. Needle biopsies of the prostate: what constitutes adequate sampling? Arch Pathol Lab Med. 1998;122:833-835.
(2.) Renshaw AA. Adequate sampling of breast core needle biopsies. Arch Pathol Lab Med. 2001;125:1055-1057.
(3.) Luo YV, Prihoda TJ, Sharkey FE. Number of levels needed for diagnosis of cervical biopsies. Arch Pathol Lab Med. 2002;126:1205-1208.
(4.) Chitkara YK, Eyre CL. Evaluation of initial and deeper sections of esophageal biopsy specimens for detection of intestinal metaplasia. Am J Clin Pathol. 2005; 123:886-888.
(5.) Maingi CP, Helm KF. Utility of deeper sections and special stains for dermatopathology specimens. J Cutan Pathol. 1997;24:171-175.
(6.) Carag HR, Prieto VG, Yballe LS, Shea CR. Utility of step sections: demonstration of additional pathological findings in biopsy samples initially diagnosed as actinic keratosis. Arch Dermatol. 2000;136:471-475.
(7.) Guillen DR, Cockerell CJ. Accurate diagnosis of cutaneous keratinocytic neoplasms: the importance of histological step sections (and other factors). Arch Dermatol. 2000;136:535-537.
(8.) Penneys NS. Histopathologist: to step section or not? Arch Dermatol. 2001; 137:375-376.
(9.) Rabinowitz AD, Silvers DN. Dermatopathology standards. J Cutan Pathol. 1996;23:194-196.
Andrea K. Bruecks, MD; Jill M. Shupe, MLT; Martin J. Trotter, MD, PhD
From the Calgary Laboratory Services, Department of Pathology and Laboratory Medicine, University of Calgary, Calgary, Alberta.
The authors have no relevant financial interest in the products or companies described in this article.
Previously presented in abstract form at the annual meeting of the United States and Canadian Academy of Pathology Meeting, Chicago, Ill, February 2002 (Mod Pathol. 2002;15:97A).
Reprints: Andrea K. Bruecks, MD, Calgary Laboratory Services, 9-3535 Research Rd NW, Calgary, Alberta, Canada T2L 2K8 (e-mail: firstname.lastname@example.org).
Table 1. Change From Nondiagnostic to Diagnostic Case No. Age, y/Sex Site Diagnosis 4 17/F Cheek Folliculitis 10 55/F Leg Basal cell carcinoma 38 37/F Cheek Fibrous papule 45 38/F Back Solar lentigo 105 75/F Arm Scar 113 69/M Finger Actinic keratosis 125 40/F Unknown Basal cell carcinoma 160 69/M Canthus Ruptured cyst 189 44/F Chest Hemangioma 225 70/F Nose Seborrheic keratosis 296 64/M Forehead Basal cell carcinoma 307 46/M Forehead Basal cell carcinoma 345 72/M Breast Verrucous keratosis 346 77/F Breast Benign lichenoid keratosis 407 67/F Cheek Actinic keratosis 432 44/F Neck Syringoma 447 37/F Shoulder Benign lichenoid keratosis 472 56/F Nose Fibrous papule 479 42/F Shoulder Nevus Case No. Diagnostic Slide Category 4 3 Benign 10 2 Malignant 38 3 Benign 45 2 Benign 105 2 Benign 113 3 Exclude malignant 125 2 Malignant 160 2 Benign 189 2 Benign 225 2 Benign 296 2 Malignant 307 2 Malignant 345 2 Exclude malignant 346 2 Benign 407 3 Exclude malignant 432 3 Benign 447 2 Benign 472 2 Benign 479 2 Benign Table 2. Diagnosis Changed * Case No. Age, y/Sex Site Slide 1 11 70/M Nose AK 29 75/M Ear Inverted follicular keratosis 63 27/F Nose Cyst 72 73/F Cheek AK 84 50/F Scalp SK, inflamed 95 19/M Face Lentigo 147 62/F Face AK 203 32/M Trunk Nevus 215 55/F Nose Fibrous papule 279 42/M Forearm Nevus 306 60/M Scrotum Epidermal hyperplasia 310 45/F Thigh Lentigo 395 69/M Shoulder AK 459 79/F Nose AK 464 83/M Ear AK Case No. Slide 2 Slide 3 11 AK SCC in situ 29 Trichilemmoma Trichilemmoma 63 SCC SCC 72 AK SCC in situ 84 BCC BCC 95 Nevus Nevus 147 SCC in situ SCC in situ 203 Nevus Dysplastic nevus 215 BCC BCC 279 Dysplastic nevus Dysplastic nevus (severe) 306 Hemangioma Hemangioma 310 Nevus Nevus 395 SCC in situ SCC in situ 459 BCC BCC 464 SCC in situ SCC in situ * AK indicates actinic keratosis; SCC, squamous cell carcinoma; SK, seborrheic keratosis; and BCC, basal cell carcinoma.
|Printer friendly Cite/link Email Feedback|
|Author:||Bruecks, Andrea K.; Shupe, Jill M.; Trotter, Martin J.|
|Publication:||Archives of Pathology & Laboratory Medicine|
|Article Type:||Clinical report|
|Date:||Jan 1, 2007|
|Previous Article:||Renal cell carcinoma with rhabdoid features: an aggressive neoplasm with overexpression of p53.|
|Next Article:||Application of a combined protocol for rational request and utilization of antibody assays improves clinical diagnostic efficacy in autoimmune...|