| Abstract|| |
Background and Aim: The natural history of cervical cancer is unique that it is preceded by a precancerous condition for a long time. Morphometry as a tool can be used in early and accurate diagnosis of these precancerous and cancerous lesions. The present study aims at assessing the utility of cellular and nuclear morphometry in differentiating squamous cell abnormality from benign conditions and also differentiating the categories of squamous cell abnormalities. Materials and Methods: Forty-eight diagnosed cases of squamous epithelial cell abnormality, that is, 10 cases each of atypical squamous cell of undetermined significance ( ASC-US), low-grade squamous intraepithelial lesion (LSIL), high-grade SIL (HSIL), and squamous cell carcinoma (SCC) and eight cases of ASC-H (ASC cannot exclude HSIL), were made the sample population and compared with a control population of 10 cases of negative for intraepithelial lesion or malignancy (NILM). Parameters like nuclear area (NA), nuclear perimeter (NP), nuclear diameter (ND), nuclear compactness (NC), cellular area (CA), cellular diameter (CD), cellular perimeter (CP), and nucleocytoplasmic (N/C) ratio were used. Results: There was a significant difference in the six groups of squamous cell abnormality based on NA, NP, ND, CA, CP, and CD (P < 0.05) using one-way analysis of variance. Nuclear morphometry parameters like NA, NP, and ND were found to be the maximum for HSIL, followed by LSIL, ASC-H, ASC-US, SCC, and NILM groups in decreasing order. The mean CA, CP, and CD were found to be the maximum for NILM, followed by LSIL, ASC-US, HSIL, ASC-H, and SCC in decreasing order. On post hoc analysis, the lesions can be divided into three groups: NILM/normal; ASC-US and LSIL; and ASC-H, HSIL, and SCC, based on N/C ratio. Conclusion: In cervical lesions, holistic parameter of cytonucleomorphometry should be taken rather than taking nuclear morphometry only. N/C ratio is a highly statistically significant parameter that can differentiate between low-grade lesions and high-grade lesions.
Keywords: Carcinoma cervix, cervical cancer screening, cytonucleomorphometry, morphometry, squamous intraepithelial lesion, the Bethesda system
|How to cite this URL:|
Mishra S, Khurana U, Kapoor N, Joshi A, Joshi D. Evaluation of the cytonucleomorphometric parameters for cases diagnosed as squamous cell abnormality on conventional cervico-vaginal pap smears. J Cytol [Epub ahead of print] [cited 2023 Feb 5]. Available from: https://www.jcytol.org/preprintarticle.asp?id=364980
| Introduction|| |
Cervical cancer is the second most common type of cancer among Indian women, with an estimated 1,23,907 new cases diagnosed and mortality of 77,348 reported in 2020. The natural history of cervical cancer is unique that it is preceded by a precancerous condition for a long time. This latent period is the time that can be utilized for screening and early detection of these precancerous conditions and can lead to prevention of cervical cancer. There is a need for accurate categorization of these lesions as the management and recommendations vary for these lesions. Cytology forms one of the most common and trusted screening tools, and this can be a conventional Pap smear More Details or liquid-based cytology (LBC). The Bethesda system (TBS) is followed for uniform reporting of cervico-vaginal cytology., TBS categorizes the cases into negative for intraepithelial lesion or malignancy (NILM), epithelial cell abnormality, and others. The epithelial cell abnormality can be either squamous or glandular type. The squamous cell abnormality is further categorized into atypical squamous cell of undetermined significance (ASC-US), ASC cannot exclude high-grade squamous intraepithelial lesion (ASC-H), low-grade squamous intraepithelial lesion (LSIL), high-grade squamous epithelial lesion (HSIL), and squamous cell carcinoma (SCC). LBC preparations are not readily available in developing countries like India, and conventional smears form the cornerstone for cervical cancer screening strategy. There is considerable literature on morphometry in LBC preparations, but not on its applicability on conventional smears.
It is very important to differentiate the above-mentioned categories from each other and from reactive conditions as the management differs among the categories. The morphologic assessment of cytological smears depends on the expertise of cytopathologist, and cytological changes caused by infections, hormonal fluctuations, atrophy, and air drying can closely resemble epithelial cell abnormality. Nuclear morphometry has been studied extensively in oral buccal smears, and has been considered as an important tool to differentiate benign from malignant in breast cytology and effusion cytology and is being utilized in ocular cytology and also in cervical cytology. There are not many studies on the use of morphometry in conventional cervico-vaginal Pap smears. The present study aims at assessing the utility of cellular and nuclear morphometry in differentiating squamous cell abnormality from benign conditions and differentiating among the categories of squamous cell abnormalities on conventional smears.
| Materials and Methods|| |
The study was carried out in a tertiary care center from central India as a retrospective lab investigational study on conventional cervico-vaginal smears. The smears were stained with conventional Papanicolaou stain with the staining protocol being the same for all the smears, and reporting was done according to TBS for reporting cervical cytology, 2014. The archives of cytopathology (from 2018 to 2015) were taken out, and already diagnosed 10 cases each of ASC-US, LSIL, HSIL, and SCC were made the sample population. In the category of ASC-H, only eight cases were available. The sample size was constrained by the availability of cases. The diagnosis was confirmed according to the 2014 Bethesda system and also taking the clinical and histopathologic correlation and the human papillomavirus (HPV) status wherever available. The cases where there was cyto-histological miscorrelation were excluded from the study. The slides were viewed again to check quality and mark the appropriate areas. Appropriate remounting was done where required. Patient demographic data were noted from the requisition forms. The sample population was compared with the control population, which was formed by 10 cases diagnosed as NILM/or reactive cellular changes associated with inflammation.
High-quality digital images of the field were taken by a camera on the Leica DFC 295 microscope using a 10× ocular and a 40× objective. Image size was 3027 × 2304 pixels, that is, 293.14 × 223.13 μm (516.3 pixels = 50 μm). Multiple fields were photographed from each slide containing the group-specific abnormalities based on microscopic analysis manually. Twenty cells were selected from these multiple fields for each slide sample. The criteria for this selection were nonoverlapping, unfolded, and best possible visible cells which could be analyzed with confidence. The cases were analyzed using ImageJ 1.52 morphometric software for image processing and analysis (JAVA) developed by the National Institutes of Health (NIH), USA. All measurements were done manually using the free-hand outline and measurement functionality in ImageJ without any filter or plugin.
Details about the parameters used are as follows:
- Nuclear diameter (ND): Diameter was the diameter of the circle with the same area as the outlined nucleus.
- Nuclear area (NA): Area within outlined nuclear perimeter (NP)
- NP: Distance around nuclear border
- Nuclear compactness (NC): Perimeter2/area (NC is a measure of the shape and regularity of the nucleus. For a fixed NA, lower numerical value of compactness points toward a more rounded and regular nucleus. The regularity is lost as it undergoes neoplastic changes, and the value of compactness should thereby increase to suggest the same.)
- Cell diameter (CD): Diameter was the diameter of the circle with the same area as the outlined cell.
- Cell area (CA): Area within outlined cellular perimeter (CP)
- CP: Distance around cellular border
- Nucleocytoplasmic ratio (N/C ratio): Area of nucleus/(area of cell – area of nucleus)
NA, NP, CA, and CP are measured parameters; while ND, NC, CD, and N/C ratio are calculated. The nuclear and cellular diameters are mathematically calculated “diameters of perfect circles which have same area as the respective measured nuclear and cellular areas.”
Ethical clearance was obtained from the Institutional Ethics Committee before carrying out the study.
The cytonucleomorphometric parameters obtained were compared with the control group and between the groups. The Statistical Package for the Social Sciences (SPSS) software calculated the mean, standard deviation, and range for all the nuclear features and cytoplasmic features. One-way analysis of variance (ANOVA) was used. The intergroup comparisons were done using post hoc test, that is, Bonferroni multiple comparisons test. A P value < 0.05 was considered significant.
| Results|| |
Ten cases each of SCC, HSIL, LSIL, and ASC-US and eight cases of ASC-H category were taken as the sample population (48 cases). Ten cases of NILM/Reactive cellular changes associated with inflammation (RCC) were taken as the control population.
The average age distribution observed in SCC, HSIL, ASC-H, LSIL, ASC-US and NILM/RCC was 47.9, 46, 50.75, 39.3, 36.7, and 36.8 years, respectively.
Out of the 48 cases, the total number of menopausal women was 15 (31%): SCC (7/10), HSIL (2/10), ASC-H (4/8), LSIL (1/10), and ASC-US (1/10).
The cases were classified on the basis of TBS 2014 [Figure 1].
|Figure 1: A composite of photomicrographs. (a) A case of negative for intraepithelial lesion or malignancy. (b) A case of ASC-US; the arrow is pointing to Trichomonas vaginalis. (c) A case of LSIL. (d) A case of ASC-H. (e) A case of HSIL. (f) A case of squamous cell carcinoma. (Calibration 516.3 pixels = 50 μm, scale bar is 50 μm). ASC-H = ASC cannot exclude high-grade squamous intraepithelial lesion, ASC-US = atypical squamous cell of undetermined significance, HSIL = high-grade squamous intraepithelial lesion, LSIL = low-grade squamous intraepithelial lesion|
Click here to view
NILM [Figure 1]a: This along with the reactive cellular changes associated with inflammation was taken as a control group.
ASC-US [Figure 1]b: TBS recommends that the nuclei should be 2.5–3 times the area of a normal intermediate cell nucleus or twice the area of a metaplastic cell and there should be minimal hyperchromasia or irregularity. Atypical parakeratosis/atypical repair patterns can be seen. The cases in the indexed study showed superficial to intermediate type of squamous cell/metaplastic cells generally arranged singly or in loose sheets. The cytoplasmic borders were distinct. The nuclei were enlarged and dyskeratosis was seen in some cases. One case of ASC-US also showed coexisting organism Trichomonas vaginalis.
LSIL cervical intraepithelial neoplasia(CIN)I [Figure 1]c: The criteria by TBS for categorizing as LSIL are numerous. One criterion mentions that the NA is generally more than 3 times the size of an intermediate cell nucleus. The indexed cases showed superficial to intermediate type of squamous cells generally arranged singly and to a lesser extent in clusters and sheets. The cytoplasmic borders were distinct. Some cases showed keratohyalin granules. The nuclei were larger and showed mild hyperchromasia. Binucleation and perinuclear halo were seen in some cases. Anisonucleosis was commonly observed. One case was labeled as LSIL with ASC-H like areas.
ASC-H [Figure 1]d: Only eight cases of ASC-H were retrieved. Singly dispersed or loose aggregates of small to medium-sized cells with high N/C ratio, hyperchromatic nuclei, and scant cytoplasm were seen. The cells were insufficient or sparse on the smear for a definitive diagnosis of HSIL. Finding the requisite number of 20 cells for morphometry was also difficult in these cases.
HSIL (CIN II, CIN III) [Figure 1]e: Sheets, hyperchromatic crowded groups (HCG), or singly dispersed smaller cells with high N/C ratio, hyperchromatic nuclei, and scant cytoplasm. Three cases showed coexisting LSIL area. However, morphometry was attempted from higher-grade lesion. In the cases showing HCG like areas, performing cytonucleomorphometry was difficult. HSIL cells are characterized by a small size compared to normal cells and a nucleus forming three-fourths of the CA.
SCC [Figure 1]f: Tumor diathesis was seen in 70% cases, showed some evidence of keratinization in 90% cases and prominent nucleoli in 30% cases, and one case showed cytophagocytosis. Spindled and caudate type of cells were seen in 50% of the cases.
The NA, NP, and ND were found to be maximum for HSIL, followed by LSIL, ASC-H, ASC-US, SCC, and NILM groups in decreasing order [Table 1], [Figure 2], [Figure 3], [Figure 4]. The CA, CP, and CD were found to be maximum for NILM, followed by LSIL, ASC-US, HSIL, ASC-H, and SCC in decreasing order. Using one-way ANOVA, the nuclear parameters like NA, NP, ND, NC, and N/C ratio and the cellular parameters like CA, CP, and CD were found to be statistically significant (P < 0.05). The F values that signify the variation between intergroup means and intragroup mean values were found to be the most significant for N/C ratio followed by CD and was the least significant for NC compared to the other parameters.
|Figure 2: (a) Photomicrograph shows superficial cells from a case of NILM. The NA and NP values of cells are seen on the right side. The mean NA of NILM cases was 39.94 ± 15.39. (b) Nuclear morphometry of a case of ASC-US. The nuclei are enlarged, but hyperchromasia and nuclear irregularity are not sufficient to call LSIL. The mean nuclear area of ASC-US cases was 84.19 ± 19.77. ASC-US = atypical squamous cell of undetermined significance, LSIL = low-grade squamous intraepithelial lesion, NA = nuclear area, NILM = negative for intraepithelial lesion or malignancy, NP = nuclear perimeter|
Click here to view
|Figure 3: (a) Nuclear morphometry in a case of LSIL. The mean nuclear area of LSIL cases was 104.32 ± 46.58. (b) Nuclear morphometry in a case of ASC-H. The mean nuclear area of ASC-H cases was 100.08 ± 39.53. ASC-H = atypical squamous cell cannot exclude high-grade squamous intraepithelial lesion, LSIL = low-grade squamous intraepithelial lesion|
Click here to view
|Figure 4: (a) Nuclear morphometry in a case of HSIL. The mean nuclear area of HSIL cases was 108.29 ± 39.07. (b) Morphometry in a case of SCC. Background shows tumor diathesis. The mean nuclear area of SCC cases was 78.94 ± 24.52. HSIL = high-grade squamous intraepithelial lesion, SCC = squamous cell carcinoma|
Click here to view
For comparisons between the individual groups, post hoc test, that is, Bonferroni multiple comparisons test, was also carried out. NA can be used to differentiate between HSIL and NILM (P = 0.001), ASC-H and NILM (P = 0.01), and LSIL and NILM (P = 0.002). NP can be used to differentiate between SCC and NILM (P = 0.023), HSIL and NILM (P < 0.001), ASC-H and NILM (P = 0.002), LSIL and NILM (P<0.001), and ASC-US and NILM (P = 0.01). ND can be used to differentiate between SCC and NILM (P = 0.02), HSIL and NILM (P < 0.001), ASC-H and NILM (P = 0.001), LSIL and NILM (P < 0.001), and ASC-US and NILM (P = 0.007). NC can be used to differentiate between ASC-H and NILM (P = 0.007). It is not useful for other group comparisons.
CA can be used to differentiate between NILM and all other groups (P < 0.001), ASC-US and SCC (P = 0.0270), and L-SIL and SCC (P = 0.0170). CP can be used to differentiate between NILM and all other groups (P < 0.001); ASC-US and SCC (P = 0.002), ASC-US and ASC-H (P = 0.0130); and L-SIL and SCC (P < 0.001), LSIL and HSIL (P = 0.0340), and LSIL and ASC-H (P = 0.0030). CD can be used to differentiate between NILM and all other groups (P < 0.001); ASC-US and SCC (P < 0.001), ASC-US and ASC-H (P = 0.0150); and LSIL and SCC (P < 0.001) and LSIL and HSIL (P=0.0380).
N/C ratio: Mean N/C ratios were maximum for SCC, followed by ASC-H, HSIL, LSIL, ASC-US, and NILM. It can be used to differentiate between NILM and all other groups (P < 0.001); ASC-US and SCC (P < 0.001), ASC-US and H-SIL (P < 0.001), ASC-US and ASC-H (P < 0.001); and L-SIL and SCC (P < 0.001), LSIL and HSIL (P < 0.001), LSIL and ASC-H (P < 0.001). N/C ratio cannot differentiate between ASC-US and LSIL and ASC-H, HSIL, and SCC. [Table 2] shows the results of Bonferroni multiple comparisons test for N/C ratio among the groups.
|Table 2: Results of Bonferroni multiple comparisons test for N/C ratio among groups|
Click here to view
| Discussion|| |
Cervical cancer is the fourth most common cancer affecting women worldwide after breast, colorectal, and lung cancers. It is also the fourth most common cause of cancer death (3,41,831 deaths in 2020) in women worldwide. Approximately one-fifth of all new cases are diagnosed in India, it being the second most common cancer in women. The global burden of this disease can be reduced by screening for neoplastic and preneoplastic lesions. The various screening modalities available for this are Lugol's iodine examination, visual inspection with acetic acid, pap smear screening (conventional and LBC), HPV testing by hybrid capture assay, colposcopy, and so on. Out of all these, pap smear screening forms the cornerstone modality. TBS is the most commonly recommended classification system for interpretation of cytological smears of cervix. Optimum categorization and subtyping of epithelial cell abnormality according to TBS requires expertise in the field. Morphometry helps in giving objective parameters that assist in proper categorization and further appropriate treatment. Morphometry is the measurement of various cell parameters microscopically by using image analysis. Various image analysis software are available for morphometry and they have been used in various studies, for example, ImageJ 1.44,, ImageJ 1.52, dot slide system from Olympus, DEBEL cytoscan, and Nikon Imaging software.
Alterations in nuclear structure are the morphologic hallmarks of cancer diagnosis.,,,,,,,, The studies have mainly focused on nuclear morphometry in breast, cervical, and oral tumors. A large number of parameters have been studied by morphometry, but the nuclear parameters related to size like NA, NP, and ND have consistently been found to be significant both in histology and cytology in distinguishing benign versus malignant lesions. Mean NA is the most consistent of them all and was found to be significant in the study. The NA, NP, and ND were found to be maximum for HSIL, followed by LSIL, ASC-H, ASC-US, SCC, and NILM groups in decreasing order. TBS mentions that usually, the NA is maximum for LSIL, but it also mentions that some HSIL may have the same degree of nuclear enlargement as seen in LSIL, which may be one of the explanations for NA being the maximum for HSIL in the indexed study and the other being smaller sample size. HSIL cells can have wide range of NA, with some cases having NA as large as LSIL cells and at times as small as intermediate cell nucleus.
We added newer segments by adding cytoplasmic parameters as and when we reviewed the literature. N/C ratio and CD are very important in differentiating low-grade lesions from high-grade lesions. The CA, CP, and CD were found to be maximum for NILM, followed by LSIL, ASC-US, HSIL, ASC-H, and SCC in decreasing order and this was found to be in line with few other studies available in literature [Table 3]. The study by Wesola et al, also mentioned that tumor cells were the smallest among the cell groups measured. Also, the given table compares the finding with one other study whereby trends are similar but the values are much higher, which may be due to usage of different software (Image Analyzer Software Motic Image Plus 3.0 at a magnification of 400×). Also, their number of cases in different categories is variable.
|Table 3: Comparison between indexed study and the studies by Wesola et al.and Tiwari et al.|
Click here to view
Nuclear morphometry has been found to be useful in squamous cell epithelial abnormality as well as glandular cell abnormality of cervix.,,,, In the indexed study on morphometry of squamous epithelial cell abnormality, the nuclear parameters like area, perimeter, and diameter were found to be significant in differentiating NILM or reactive cellular changes associated with inflammation from intraepithelial lesions and carcinomas. Also, nuclear morphometry is helpful in differentiating between the groups. The cytonucleomorphometric analysis incorporating N/C ratio, CD, CP, and CA was also found to be more statistically significant in differentiating between the groups. Based on the analysis, morphometry is not able to differentiate between ASC-H and HSIL, which is true as the cells of ASC-H are like HSIL only, but fewer in number., The morphometric analysis was not able to differentiate SCC from HSIL as the main differentiating features of SCC are prominent nucleoli, tumor diathesis, and cells of caudate and spindly type which are not assessed by morphometry alone. Thus, cytonucleomorphometry can broadly classify lesions into NILM/normal; ASC-US and LSIL; ASC-H, HSIL, and SCC. The limitation of this study is its small sample size. However, in a study, nuclear morphometry itself was found to be statistically significant in differentiating LSIL from HSIL and from malignant ones.
However, there are limitations to morphometry, such as nuclear irregularities cannot be commented on in the morphometric analysis and need the presence of a trained cytopathologist. However, in a study, nuclear concave points were used in the morphometry of breast lesions, which was taken as a measure of nuclear irregularity. Similar parameters can be used in further studies on cervical lesions' morphometry. Morphometry becomes difficult and unreliable when the cells form syncytia and HCG, and this may be seen in HSIL and SCC.
To conclude, diagnosis and classification of the various cervical neoplasia is highly subjective and skill based. It requires a trained pathologist time to parse for significant identification features. The technique of morphometry aims to make this process more objective with the distinction being made on specifically selected parameters curated from studies like our own. The significant data gathered from manual cytonucleomorphometric studies can be used to automate the process and to generate and validate plugins/programs that can process the raw image data and classify clinical cases while consuming less time. This technique can also act as an adjunct for use by the clinical pathologist or for providing provisional diagnosis in remote areas with unavailability of trained pathologists.
In the end, the authors would like to again emphasize that in cervical lesions, cytonucleomorphometry should be taken rather than taking nuclear morphometry only and N/C ratio is a highly statistically significant parameter that can differentiate between low-grade lesions and high-grade lesions.
Statement of ethics
This study protocol was reviewed and approved by the institutional human ethics committee, AIIMS Bhopal, with the approval number IHEC-LOP/2018/STS0139. Waiver of consent was granted as patient identifier was not used and confidentiality was assured.
The study was carried out under the Indian council for medical research-Short term studentship scheme (ICMR STS) scheme.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Wright TC, Ronett BM, Kurman RJ, Ferency A. Precancerous lesion of the cervix. In: Kurman RJ, Ellenson LH, Ronnett BM, editors. Blaustein's Pathology of the Female Genital Tract. 6th
ed. New York: Springer; 2011. p. 193-253.
Nayar R, Wilbur DC. The Bethesda System for Reporting Cervical Cytology. 3rd
ed. Switzerland: Springer; 2015.
Nayar R, Wilbur DC. The Pap test and Bethesda 2014. Acta Cytol 2015;59:121-32.
Punit VP, Sheela K, Veerendra K, Vidya GD. Quantitative cytomorphometric analysis of exfoliated normal gingival cells. J Cytol 2011;28:66-72.
Nivia M, Sunil SN, Rathy R, Anilkumar TV. Comparative cytomorphometric analysis of oral mucosal cells in normal, tobacco users, oral leukoplakia and oral squamous cell carcinoma. J Cytol 2015;32:253-60.
] [Full text]
Skjørten F, Kaaresen R, Jacobsen U, Skaane P, Amlie E. Nuclear morphometry of benign and malignant breast lesions. Eur J Surg Oncol 1991;17:350-3.
Arora B, Setia S, Rekhi B. Role of computerized morphometric analysis in diagnosis of effusion specimens. Diagn Cytopathol 2006;34:670–5.
Doughty MJ. Assessment of size and nucleo-cytoplasmic characteristics of the squamous cells of the corneal epithelium. Clin Exp Optom 2015;98:218-23.
Rani MND, Narasimha A, Kumar MLH, SR S. Evaluation of pre-malignant and malignant lesions in cervico vaginal (PAP) smears by nuclear morphometry. J Clin Diagn Res 2014;8:FC16-9.doi: 10.7860/JCDR/2014/10367.5221
Panwar H, Khurana U, Gupta K, Shrivastava S, Jain A, Sankhla K. A clinic pathological study of neoplastic and preneoplastic lesions of cervix along with HPV genotyping in biopsy proven preneoplastic lesions. Indian J Obstet Gynaecol Res 2016;3:400-6.
Narasimha A, Vasavi B, Harendra Kumar ML. Significance of nuclear morphometry in benign and malignant breast aspirates. Int J App Basic Med Res 2013;3:22-6.
] [Full text]
Wesoła M, Lipiński A, Jeleńe M. Morphometry in the cytological diagnosis of cervical smears. Adv Clin Exp Med 2014;23:289–93.
Kalhan S, Dubey S, Sharma S, Dudani S, Preeti, Dixit M. Significance of nuclear morphometry in cytological aspirates of breast masses. J Cytol 2010;27:16-21.
] [Full text]
Kashyap A, Jain M, Shukla S, Andley M. Study of nuclear morphometry on cytology specimens of benign and malignant breast lesions: A study of 122 cases. J Cytol 2017;34:105.
] [Full text]
Tiwari AK, Khare A, Grover SC, Bansal R, Sharma S. Role of nuclear morphometry in screening of cervical pap smear. J Clin Diagn Res 2019;13:EC13-6.
Conceição T, Braga C, Rosado L, Vasconcelos MJM. A review of computational methods for cervical cells segmentation and abnormality classification. Int J Mol Sci 2019;20:5114.
Cho JD, Park BT, Lee JD. Morphometric study on cervical intraepithelial neoplasia. Korean J Pathol 1990;24:267-74.
Swartz R, West L, Boiko I, Malpica A, MacAulay C, Carraso A, et al
. Use of nuclear morphometry characteristics to distinguish between normal and abnormal cervical glandular histologies. Anal Cell Pathol 2003;25:193-200.
Assistant Professor, Department of Pathology and Lab Medicine, AIIMS Bhopal, Madhya Pradesh - 462 024
Source of Support: None, Conflict of Interest: None
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3]