Comparison of fine-needle aspiration and fine-needle capillary sampling of thyroid nodules: A prospective study with emphasis on the influence of nodule size
The objective of this study was to compare the sampling efficiency of ultrasound-guided fine-needle aspiration (FNA) and fine-needle capillary (FNC) sampling in thyroid nodules, in which the authors specifically analyzed the influence of nodule size.
This study included 280 thyroid nodules in 275 consecutive patients. The nodules were divided into 4 size subgroups: ≤5.0 mm, from 5.1 to 10.0 mm, from 10.1 to 20.0 mm, and >20.0 mm. Each nodule was sampled by both FNA and FNC. The final cytopathologic findings were reported. The smears were scored and then categorized as diagnostically inadequate, adequate, or superior on the basis of 4 parameters, which included background clot or blood, the number of obtained cells, preserved tissue architecture, and cellular degeneration.
The κ scores for agreement of the cytopathologic results between FNA and FNC sampling in the 4 size subgroups were 0.377, 0.455, 0.751, and 0.352 for nodules that measured ≤5.0 mm, from 5.1 to 10.0 mm, from 10.1 to 20.0 mm, and >20.0 mm, respectively. The proportion of nondiagnostic of FNAs was significantly lower than the proportion of nondiagnostic FNC samples in nodules that measured >20.0 mm (P = .037). Scores for the 4 diagnostic parameters were significantly greater in FNAs than in FNC samples in nodules that measured from 5.1 to 10.0 mm and >20.0 mm (all P < .05); however, similar results were not observed in the nodules that measured ≤5.0 mm or from 10.1 to 20.0 mm (all P > .05). Also, FNA yielded significantly more diagnostically superior specimens than FNC sampling in nodules that measured from 5.1 to 10.0 mm and >20.0 mm (P < .05 for both).
The current findings indicated that FNA may be more suitable than FNC for sampling nodules that measure from 5.1 to 10.0 mm and >20.0 mm; whereas, for nodules that measure ≤5.0 mm and from 10.1 to 20.0 mm, the 2 techniques could yield specimens with similar quality. Cancer (Cancer Cytopathol) 2014;122:266–273. © 2013 American Cancer Society.
Thyroid fine-needle aspiration (FNA) is a safe, simple, relatively accurate, and first-line diagnostic tool in the evaluation of thyroid nodules.[1-3] There is no doubt that FNA has contributed toward avoiding a large number of unnecessary surgeries in the past several decades. However, it is also generally believed that this aspiration technique frequently leads to microscopic hemorrhages, which are an obstacle to proper cytologic interpretation. In an attempt to overcome this problem, fine-needle capillary (FNC) sampling, a technique without aspiration, was developed in the 1980s. It has been suggested that this nonaspiration sampling technique could reduce the amount of blood in samples and produce superior quality specimens.[4-6]
In the literature, many studies have focused on comparisons between thyroid FNA and FNC sampling. Some studies confirmed that FNC yielded sampling material with better quality cellularity and less blood stains compared with FNA,[6-12] whereas other reports indicated that there was no significant difference between the 2 techniques.[13, 14] There were some limitations in those studies. First, ultrasound (US) guidance, which is superior to palpation guidance for obtaining adequate material, was not used in most studies. Second, most of those studies were based on a small sample size. Third, the influence of nodule size on FNA and FNC sampling was not specifically addressed. Therefore, the current study was designed to compare the sampling efficiency of US-guided FNA and FNC sampling in 280 thyroid nodules in which the influence of nodule size was specifically analyzed.
MATERIALS AND METHODS
This prospective study was approved by our institutional ethics committee, and all patients provided written informed consent. The study was carried out from June 2012 to August 2012 and enrolled 275 patients, including 215 women and 60 men (mean age ± standard deviation [SD], 45.51 ± 13.83 years; age range, 14-78 years). In total, 280 thyroid nodules in the 275 consecutive patients were evaluated using US-guided cytologic sampling. The thyroid nodules were divided into 4 size groups based on the greatest nodule dimension on US: ≤5.0 mm, from 5.1 to 10.0 mm, from 10.1 to 20.0 mm, and >20.0 mm.
FNA and FNC Procedures
Both FNA and FNC were performed by radiologists using a commercially available scanner (Esaote MyLab 90; Esaote, Genoa, Italy) equipped with a high-resolution small-parts probe (LA523; 4-13 MHz). Each nodule was sampled by 1 FNA pass and 1 FNC pass. FNA was performed first followed by FNC in 140 nodules, and FNC was performed first followed by FNA in the other 140 nodules. In cases of multinodular goiter, cytologic sampling was performed in the nodules that had suspicious US features. If no suspicious nodules were present, then the dominant nodules were selected for sampling. In cases of mixed cystic-solid nodules, the solid component was sampled. Patients who had purely cystic nodules were not enrolled in the current study.
All samplings were performed with 22-gauge needles under US guidance. During FNA, the needle attached to 10-mL syringe was inserted into the target nodule and moved back and forth within the nodule. Suction was applied when the needle was advanced into the nodule and halted before the needle was removed from the nodule. In FNC, the needle, without syringe attached, was introduced into the nodule and moved rapidly back and forth while it was angled in different directions within the nodule. In both FNA and FNC sampling procedures, the needle was withdrawn when the sample material appeared in the hub of the needle.
Interpretation of Smears
The obtained sample material was placed on glass slides. Smears were prepared, fixed in alcohol, and then stained with hematoxylin and eosin. The adequacy of the smears was assessed using the scoring system reported by Haddadi-Nezhad et al. The 2 sampling techniques were compared using the following 4 parameters: background clot or blood, number of obtained cells, preserved tissue architecture, and cellular degeneration (Table 1). A cumulative score between 0 and 2 was categorized as inadequate for diagnosis, scores between 3 and 5 were categorized as adequate for diagnosis, and scores between 6 and 8 were categorized as superior for diagnosis.
Table 1. Scoring System Applied to Specimen Quality Analysis
|Background blood or clot|| |
|Excessive: Diagnosis not possible||0|
|Intermediate: Diagnosis possible||1|
|Minimal. Diagnosed easily||2|
|No. of obtained cells|| |
|Minimal: Diagnosis not possible||0|
|Intermediate: Diagnosis possible||1|
|Abundant: Diagnosed easily||2|
|Preserved tissue architecture|| |
|Minimal to absent: Diagnosis not possible||0|
|Intermediate: Some preservation of, eg, follicle, papillae, etc||1|
|Well preserved: Diagnosed easily||2|
|Cellular degeneration|| |
|Severe: Diagnosis not possible||0|
|Intermediate: Diagnosis possible||1|
|Minimal: Diagnosed easily||2|
The final cytopathologic finding was reported by using the Bethesda criteria, in which a sample is considered adequate if it contains a minimum of 6 groups of well observed follicular cells, with at least 10 cells per group. The cytopathologic results were reported as follows: nondiagnostic, benign, indeterminate (including atypia of undetermined significance, follicular lesion of undetermined significance, suspicious for a follicular neoplasm, and follicular neoplasm), suspicious for malignancy, and malignancy. All smears were evaluated and interpreted by a single cytopathologist who was blinded to the sampling method used.
The degree of agreement between the final cytopathologic findings obtained by FNA and FNC sampling was examined using the Cohen κ statistic. The κ values were interpreted as follows: κ values from 0.00 to 0.20 indicated slight agreement; κ values from 0.21 to 0.40, fair agreement; κ values from 0.41 to 0.60, moderate agreement; κ values from 0.61 to 0.80, substantial agreement; and κ values from 0.80 to 1.00, almost perfect agreement. Chi-square tests or exact Fisher tests were used to evaluate the differences between FNA and FNC regarding the nondiagnostic proportion and the overall diagnostic performance. The Mann-Whitney U test was used to compare FNA/FNC scores between first-performed and second-performed procedures. The Wilcoxon signed-rank test was used to compare scores between FNA and FNC. The Kruskal-Wallis test was applied to compare FNA and FNC scores among the 4 size subgroups, and multiple comparisons between paired subgroups were assessed with the Nemenyi test. Statistical analyses were performed using the SPSS statistical software package (version 17.0; SPSS Inc., Chicago, Ill). P values < .05 were considered statistically significant.
Of the 280 thyroid nodules from 275 patients (greatest maximum dimension: range, 3.0-48.0 mm; mean ± SD; 13 ± 7.84 mm), there were 22 nodules ≤5.0 mm, 106 nodules that measured from 5.1 to 10.0 mm, 102 nodules that measured from 10.1 to 20.0 mm, and 50 nodules >20.0 mm. The cytopathologic results from the different size groups for FNA and FNC sampling are provided in Table 2. The κ scores for the agreement between the 2 techniques according to nodule size were 0.553 for all nodules, 0.377 for nodules ≤5.0 mm, 0.455 for nodules from 5.1 to 10.0 mm, 0.751 for nodules from 10.1 to 20.0 mm, and 0.352 for nodules >20.0 mm. The proportion of nondiagnostic FNAs was 15% in all nodules, 27.3% in nodules ≤5.0 mm, 15.1% in nodules from 5.1 to 10.0 mm, 14.7% in nodules from 10.1 to 20.0 mm, and 10% in nodules >20.0 mm; and the corresponding values for FNC sampling were 20.7%, 31.8%, 24.5%, 11.8%, and 26%, respectively. A comparison indicated that the difference in the nondiagnostic proportion between the 2 techniques was significant only in nodules >20.0 mm (P = .037).
Table 2. Cytologic Results of Fine-Needle Aspiration and Fine-Needle Capillary Sampling According to the Bethesda Criteria
|Nondiagnostic||42 (15)||58 (20.7)||0.553||6 (27.3)||7 (31.8)||0.377||16 (15.1)||26 (24.5)||0.455||15 (14.7)||12 (11.8)||0.751||5 (10)||13 (26)||0.352|
|Benign||156 (55.7)||150 (53.6)|| ||11 (50)||10 (45.5)|| ||52 (49.1)||52 (49.1)|| ||55 (53.9)||57 (55.9)|| ||38 (76)||31 (62)|| |
|Indeterminate||9 (3.2)||9 (3.2)|| ||0 (0)||0 (0)|| ||0 (0)||0 (0)|| ||6 (5.9)||7 (6.9)|| ||3 (6)||2 (4)|| |
|Suspicious for malignancy||12 (4.3)||4 (1.4)|| ||2 (9.1)||1 (4.5)|| ||5 (4.7)||0 (0)|| ||5 (4.9)||3 (2.9)|| ||0 (0)||0 (0)|| |
|Malignancy||61 (21.8)||59 (21.1)|| ||3 (13.6)||4 (18.2)|| ||33 (31.1)||28 (26.4)|| ||21 (20.6)||23 (22.5)|| ||4 (8)||4 (8)|| |
There were no significant differences in scores for the 4 diagnostic parameters or in the cumulative score when FNA or FNC was used first or second in the total group (all P > .05). Table 3 indicates that scores for the 4 diagnostic parameters and the cumulative score for sampling adequacy were significantly higher in FNAs than in FNC samples for all nodules, for nodules that measured from 5.1 to 10.0 mm, and for nodules that measured >20.0 mm (all P < .05), but not for nodules that measured ≤5.0 mm or for nodules that measured from 10.1 to 20.0 mm (all P > .05).
Table 3. Comparison of Mean Scores in Assessment Parameters Between Fine-Needle Aspiration and Fine-Needle Capillary Sampling
|Background blood or clot||1.41 ± 0.60||1.30 ± 0.63||.000||1.05 ± 0.65||1.05 ± 0.65||1.000||1.42 ± 0.62||1.27 ± 0.66||.003||1.46 ± 0.57||1.40 ± 0.56||.134||1.42± 0.57||1.24 ± 0.66||.013|
|No. of obtained cells||1.43 ± 0.71||1.27 ± 0.73||.000||1.14 ± 0.83||1.05 ± 0.79||.644||1.44 ± 0.72||1.19 ± 0.73a||.001||1.46 ± 0.68||1.46 ± 0.67||.991||1.48 ± 0.65||1.16 ± 0.77||.007|
|Preserved tissue architecture||1.46 ± 0.71||1.30 ± 0.74||.000||1.05 ± 0.84||1.05 ± 0.79b||1.000||1.51 ± 0.71||1.23 ± 0.76c||.000||1.48 ± 0.69||1.49 ± 0.66||.862||1.50 ± 0.65||1.18 ± 0.77||.005|
|Cellular degeneration||1.36 ± 0.71||1.21 ± 0.73||.000||1.00 ± 0.82||1.05 ± 0.79||.782||1.38 ± 0.71||1.13 ± 0.73d||.000||1.39 ± 0.71||1.38 ± 0.69||.852||1.42 ± 0.67||1.12 ± 0.72||.004|
|Total score||5.66 ± 2.57||5.08 ± 2.66||.000||4.23 ± 2.99||4.18 ± 2.84e||.949||5.75 ± 2.58||4.82 ± 2.71||.000||5.79 ± 2.51||5.74 ± 2.37||.828||5.82 ± 2.38||4.70 ± 2.80||.003|
Our results also revealed that there were no significant differences in the scores for FNA among different size groups (all P ≤ .05). For FNC sampling, scores for the number of obtained cells, preserved tissue architecture, and cellular degeneration as well as the cumulative score differed significantly among the different size groups (P = .010, P = .010, P = .030, and P = .022, respectively); whereas the score for background clot or blood did not differ significantly (P = .103). Multiple comparisons between the 2 size groups for FNC revealed that the score for the number of obtained cells was significantly higher in nodules that measured from 10.1 to 20.0 mm (mean score ± SD, 1.46 ± 0.67) than in nodules that measured from 5.1 to 10.0 mm (1.19 ± 0.73; P = .024); the score for preserved tissue architecture was significantly higher in nodules that measured from 10.1 to 20.0 mm (1.49 ± 0.66) than in nodules that measured ≤5.0 mm (1.05 ± 0.79; P = .042) or in nodules that measured from 5.1 to 10.0 mm (1.23 ± 0.76; P = .041); the score for cellular degeneration was significantly higher in nodules that measured from 10.1 to 20.0 mm (1.38 ± 069) than in nodules that measured from 5.1 to 10.0 mm (1.13 ± 0.73; P = .043); and the cumulative score was significantly higher in nodules that measured from 10.1 to 20.0 mm (5.74 ± 2.37) than in nodules that measured ≤5.0 mm (4.18 ± 2.84; P = .048) (Table 3).
Table 4 provides data on the diagnostic performance of both techniques. FNA yielded significantly more diagnostically superior specimens than FNC in the total nodules (P = .011), in nodules that measured from 5.1 to 10.0 mm (P = .013), and in nodules that measured >20.0 mm (P = .045). However, there were no significant differences between FNA and FNC with regard to diagnostically adequate specimens. The number of diagnostically inadequate specimens were greater in FNC in the total nodules, in the nodules that measured from 5.1 to 10.0 mm, and in the nodules that measured >20.0 mm compared with FNA; however, these differences were not statistically significant (P = .098, P = .096, and P = .050, respectively).
Table 4. Comparison of Diagnostic Performance Between Fine-Needle Aspiration and Fine-Needle Capillary Sampling
|Superior||159 (56.8)||129 (46.1)||.011||8 (36.4)||7 (31.8)||1.000||63 (59.4)||45 (42.5)||.013||60 (58.8)||59 (57.8)||.887||28 (56)||18 (36)||.045|
|Adequate||86 (30.7)||102 (36.4)||.152||7 (31.8)||9 (40.9)||.755||30 (28.3)||39 (36.8)||.187||31 (30.4)||33 (32.4)||.763||18 (36)||21 (42)||.539|
|Inadequate||35 (12.5)||49 (17.5)||.098||7 (31.8)||6 (27.3)||1.000||13 (12.3)||22 (20.8)||.096||11 (10.8)||10 (9.8)||.818||4 (8)||11 (22)||.050|
FNA cytology is a widely accepted method for the evaluation and diagnosis of thyroid nodules. However, this technique commonly encounters the problem of blood-stained samples, which may result in diagnostic difficulties with the cytologic interpretation. Unlike the FNA procedure, which uses fairly high suction pressures, FNC sampling is a nonaspiration technique that relies on the property of capillary force in narrow channels, and it has been suggested that FNC may produce samples of better quality by decreasing the dilution of cells by blood.[5, 6] However, the effectiveness of these 2 sampling techniques in thyroid nodules has not been fully established in the literature.[6-14, 18-21] Table 5 summarizes the inconsistent results from previous studies that made comparisons between FNA and FNC based on diagnostic performance categorized as superior, adequate, and inadequate. Therefore, the decision to choose either FNA or FNC sampling of thyroid nodules was mainly based on individual anecdotal experience. Thus, prospective studies with larger samples are necessary to better assess the potential advantages of FNA and FNC.
Table 5. Main Results From the Comparison Between Fine-Needle Aspiration and Fine-Needle Capillary Sampling in Previous Studies and the Current Series
|Santos & Leiman 1988||50||Palpation||22||Palpable nodules||8||44||<.001||78||50||.004||14||6||.182|
|Ramachandra 2011||69||ND||22 or 23||ND||7.3||20.3||.026||68||68||1.000||24.6||11.6||.047|
|Rizvi 2005||150||Palpation||23 or 24||Palpable nodules||20||44.7||<.001||73.3||53.3||<.001||6.7||2||.047|
|Kamal 2002||200||Palpation||23 or 24||Palpable nodules||44||49.5||.270||23||27||.356||33||23.5||.035|
|Mahajan & Sharma 2010||50||ND||ND||Palpable nodules||60||72||.205||24||8||.029||16||20||.603|
|de Carvalho 2009||260||Palpation||23||Palpable nodules||78.8||79.6||.829||13.1||12.3||.792||8.1||8.1||1.000|
|Haddadi-Nezhad 2003||200||ND||25||10-40 mm||ND||ND||>.05a||ND||ND||>.05a||ND||ND||>.05a|
|Tauro 2012||50||Palpation||23||Palpable nodules||30||18||.160||64||70||.523||6||12||.485|
|Kashi 2011||302||ND||25||Mainly 10-40 mm||40.4||32.5||.042||49.3||54.3||.222||10.3||13.2||.256|
|Current study||280||Ultrasound||22||3-48 mm||56.8||46.1||.011||30.7||36.4||.152||12.5||17.5||.098|
In the current prospective study, we attempted to compare the diagnostic adequacy between the 2 techniques, both of which were performed under US guidance. It is generally accepted that US-guided sampling of thyroid nodules has many advantages over palpation-guided sampling.[22-24] By monitoring the tip of the needle in real time, the sampling procedure can be very accurate and effective, even in thyroid nodules measuring <5 mm.[25, 26] Our study sample consisted of 280 nodules consecutively sized from 3.0 to 48.0 mm, among which 22 nodules measured <5 mm. The use of US guidance made it possible to investigate the influence of nodule size on FNA and FNC sampling, which was mostly ignored in previous studies.
Our results indicated a moderate degree of agreement in cytopathology results based on the Bethesda criteria between FNA and FNC sampling in the total group. In the size subgroups, the level of agreement was increased along with increasing size from nodules measuring ≤5.0 mm to those measuring 10.1 to 20.0 mm. However, the lowest κ value was observed for the nodules that measured >20.0 mm. The low κ values in smaller nodules may have been caused by the relatively lower sampling accuracy, whereas the low κ value in nodules measuring >20.0 mm may have been related to fibrotic and cystic changes in larger nodules, which could make it more difficult to obtain diagnostic material. Previous studies revealed that the nondiagnostic proportion was similar between FNA and FNC.[13, 18] Our study demonstrated that the nondiagnostic proportion was lower for FNA than for FNC in the total group and in all size subgroups except for nodules that measured from 10.1 to 20.0 mm; however, the difference was significant only for nodules that measured >20.0 mm (P = .037).
Usually, smear adequacy is evaluated using a scoring system based on the amount of cells and blood and on preservation of the material.[7, 14, 19] For our study, we used the scoring system reported by Haddadi-Nezhad et al, in which background clot or blood, the number of obtained cells, preserved tissue architecture, and cellular degeneration were assessed. For the purpose of evaluating whether traumatic disruption induced by the first sampling procedure reduced the quality of the second sampling material, half of the nodules underwent FNA followed by FNC, and vice versa. The results indicated that there were no significant differences in scores for the 4 diagnostic parameters or for the cumulative score when FNA or FNC was used first or second in the total group (all P > .05). Our result was consistent with that reported by Kamal et al but was inconsistent with the results reported by Romitelli et al, who demonstrated that the negative pressure used in FNA could disrupt the nodule structure; thus, the performance of FNC after FNA resulted in an increased frequency of inadequate specimens.
It is generally assumed that more cellular material may be obtained with FNA, whereas FNC may yield higher quality specimens with less blood stains by eliminating the suction effect. Our study indicated that less blood stains, more cells, better preserved tissue architecture, and less cellular degeneration were obtained in FNAs than in FNC samples in the total group. Such results were in discordance with most previous studies, which demonstrated that there were no significant differences between the 2 techniques with regard to scores for the individual parameters and total scores.[12-14, 21] However, our results were somewhat in accordance with those reported by Kashi et al, who observed more cellular material and better architecture in FNAs but less blood in FNC samples, whereas Kamal et al observed that FNC sampling yielded more cellular material than FNA. In the size subgroups, our data revealed that FNA scored better than FNC in nodules that measured from 5.1 to 10.0 mm and in those >20.0 mm (all P < .05), but not in nodules ≤5.0 mm or those that measured from 10.1 to 20.0 mm (all P > .05). Therefore, our results indicate that FNA may be more suitable than FNC sampling for nodules that measure from 5.1 to 10.0 mm and for those >20.0 mm.
For FNA, no significant differences in scores were observed among the different size groups in our study (all P ≥ .05), indicating that FNA was able to obtain material with similar quality in nodules of various sizes. For FNC sampling, however, we observed that FNC was able to provide better sample material in nodules that measured from 10.1 to 20.0 mm than in nodules that measured ≤5.0 mm or from 5.1 to 10.0 mm with respect to scores for the number of obtained cells, preserved tissue architecture, and cellular degeneration, indicating that FNC sampling could be applied preferentially to nodules that measured from 10.1 to 20.0 mm.
Many previous studies demonstrated that diagnostically superior specimens were obtained more frequently by FNC than by FNA,[6, 8, 9] as indicated in Table 5, and that FNA produced more adequate material[6, 9, 12]; whereas other investigators observed that the proportion of superior or adequate samples was similar using both techniques.[10, 13, 19, 21] For insufficient samples, some studies indicated that FNA yielded more insufficient samples than FNC,[8-10] whereas others reported no significant differences between the 2 techniques. Our results were somewhat in discordance with previous studies. Our findings revealed that the proportion of adequate samples was the similar using both techniques for the total group and for all size subgroups. However, consistent with the results obtained by Kashi et al, we demonstrated that more diagnostically superior specimens were obtained by FNA than by FNC sampling in the total nodules (P = .011), in the nodules that measured from 5.1 to 10.0 mm (P = .013), and in the nodules that measured >20.0 mm (P = .045). Our study also revealed that more diagnostically inadequate specimens were yielded by FNC sampling in these 3 groups when compared with FNA; however, these differences were not statistically significant (P = .098, P = .096, and P = .050, respectively). Thus, once again, our results indicated that FNA may be more suitable than FNC sampling for nodules that measure from 5.1 to 10.0 mm and those >20.0 mm.
It is obvious that there were many inconsistencies in the efficiency of FNA and FNC between our results and previous findings. Previous studies mainly consisted of reports with small sample sizes, in which both FNA and FNC were under the guidance of palpation but not US. Therefore, those studies generally only enrolled patients who had palpable thyroid nodules. However, FNA biopsy was recommended for nodules <10 mm under several conditions. Our study sample consisted of nodules that were sized consecutively from 3 to 48 mm, which means that we were able to include both palpable and nonpalpable nodules. In addition, under US guidance, a more accurate biopsy could be obtained in our study compared with previous studies. Consequently, our results may be more suitable for routine clinical practice.
There were some limitations to this study. First, interobserver variability among radiologists who performed the sampling procedures was not assessed. Second, although the sample size was up to 280 nodules in the total group, the sample sizes in some size subgroups were relatively small, especially in the nodules that measured ≤5.0 mm. Third, 22-gauge to 27-gauge needles generally are used for thyroid biopsy.[2, 29] The results of a comparison between FNA and FNC may vary according to various needle sizes. However, we performed the study only using 22-gauge needles. Thus, additional studies with larger sample sizes and using different needle sizes will be necessary to obtain more reliable results regarding the efficiency of FNA and FNC sampling in thyroid nodules.
In conclusion, the current study demonstrated that fewer blood stains, more cells, better preserved tissue architecture, and less cellular degeneration could be obtained in FNA samples than in FNC samples for the total group, in nodules that measured from 5.1 to 10.0 mm, and in nodules that measured >20.0 mm. For diagnostic performance, more diagnostically superior specimens were obtained by FNA than by FNC in all nodules together, in nodules that measured from 5.1 to 10.0 mm, and in nodules that measured >20.0 mm. Therefore, FNA may be more suitable than FNC sampling for nodules that measure from 5.1 to 10.0 mm and >20.0 mm; whereas, for nodules that measure ≤5.0 mm and from 10.1 to 20.0 mm, the 2 techniques may yield specimens with similar quality.
This research was supported by the Science and Research Project of Shanghai Municipal Health Bureau (No. 2011012).
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.