Comparative Evaluation of Clinical, Cytological and Microbiological Profile in Abdominal vs. Cervical Lymph Nodal Tuberculosis with Special Emphasis on Utility of Auramine-O Staining.

CONTEXT: Extrapulmonary tuberculosis (EPTB) especially abdominal lymph nodal tuberculosis (LNTB) poses a unique diagnostic challenge. The clinical, cytological, and microbiological profiles, especially with respect to the use and role of Auramine -O (AO) stain, are not as well characterized in abdominal LNTB as cervical LNTB and were evaluated in the present comparative study. SUBJECTS AND METHODS: This study was conducted in the Department of Pathology of a tertiary care hospital in Shillong, Meghalaya in 540 clinical suspected cases of tuberculosis who underwent FNAC. The smears were submitted for Leishman's stain for cytological analysis, along with ZN and Auramine O stain for demonstration of the organism, analyzed, and scored and the results were compared with culture wherever available. The results from abdominal and cervical lymph nodal tuberculosis were compared using Microsoft Excel and SPSS software.
RESULTS: Out of 540 cases, most were tuberculosis (266) followed by reactive lymphadenitis (162), malignancy, and acute necrotizing lesion. On comparing, abdominal lymph nodes (n = 163) were more likely to reveal cheesy/purulent material macroscopically, necrotizing lymphadenitis along with ZN stain and Auramine positivity (P < 0.05) while cervical lymph nodes (n = 66) revealed a higher proportion of granulomatous lymphadenitis and culture positivity (P < 0.05). The sensitivity, NPV, and diagnostic accuracy of AO stain (85.9%, 48.0%, and 62.3%) were higher as compared to ZN stain (47.4%, 39.3%, and 51.9%) with culture as the gold standard. The combined sensitivity of Ziehl Neelsen stain and Auramine stain was 92.05%.
CONCLUSION: Cytological and microbiologic features of abdominal LNTB differ from cervical LNTB. Moreover, AO stain increases the smear positivity, is almost twice as sensitive as ZN stain and should be used as an adjunct in cytological material wherever available. Copyright:

INTRODUCTION

Tuberculosis (TB) is one of the top ten leading causes of ill health and death globally, infecting about a quarter of the world's population.[1] According to the World Health Organization (WHO) world TB report, 2019, the global burden of the disease was 10.0 million cases in 2018. Out of these, India accounted for 27% (2.69 million) cases and 35% of global TB deaths with an estimated prevalence of 199 cases/1,00,000 population.[1]

Extrapulmonary tuberculosis (EPTB) is a form of tuberculosis that refers to TB involving organs other than the lungs like lymph nodes, pleura, genitourinary tract, skin, joints, and bones or meninges. It represented 15% of total TB cases globally in 2018.[1] However, data on EPTB from India is limited. One of the studies reported the incidence of EPTB to range from 15-20% of all TB cases in HIV negative cases and 40-50% of all new TB cases in HIV positive cases.[23]

Whereas the diagnosis of pulmonary TB (PTB) has become more streamlined in recent years, the diagnosis of EPTB poses multiple challenges. Some of the challenges include a presentation at obscure inaccessible sites with atypical symptoms; paucibacillary nature of the disease; lack of data and diagnostic algorithms; requirements of invasive procedure to achieve samples which are usually of low volume; and, the requirement of trained personnel and infrastructure.[45]

Many laboratory methods are available for the diagnosis of EPTB and have been reviewed extensively elsewhere.[6789] Out of these, Fine needle aspiration cytology (FNAC) is a simple, cheap, rapid, and outpatient diagnostic procedure for EPTB which has high sensitivity and specificity especially if combined with Ziehl-Neelson (ZN) stain, Auramine-O (AO) stains and submission of material for mycobacterial culture.[6789] However, unlike PTB, the use and role of AO stain have not been studied extensively in FNAC material in EPTB. Moreover, to the best of our knowledge, cytomorphologic and microbiological features in abdominal lymph nodal tuberculosis have not been well characterized. With this view in mind, we conducted the present comparative study to evaluate the clinical, cytological, and microbiological profile of patients presenting in our hospital for FNAC with clinical suspicion of EPTB in cervical and abdominal lymph nodes with special emphasis on the use of AO stain for diagnosis of lymph nodal EPTB.

SUBJECTS AND METHODS

This study was conducted in the Department of Pathology of a tertiary care hospital in Shillong; Meghalaya for a period from August 2017 to December 2019 in 540 clinical suspected cases of tuberculosis who underwent Fine Needle aspiration cytology (FNAC). Ethical Committee clearance was obtained from the institutional ethics committee. Due written consent was taken from the patient regarding the procedure as well as for participation in the study. FNAC was performed by a pathologist using either palpation-guided or USG guided procedure in accordance with the standard protocol which is followed in our institution. The macroscopic appearance of the specimen is noted and smears are made. If some sample is available, it is submitted for MTB culture. Culture is performed using MGIT 320 system. The smears are submitted for Leishman's stain for cytological analysis, along with ZN and Auramine-O stain for demonstration of an organism using standard protocols for staining.[101112] For every batch of tests, positive control slides are stained and checked before proceeding for analysis.

Clinical, radiological and demographic profile of patients were retrieved from the Electronic medical records. The slides were retrieved and Leishman's stained slides were examined for the presence of necrosis and granulomas. Based on microscopic examination, the cases were diagnosed as Necrotizing lymphadenitis, Necrotizing Granulomatous lymphadenitis, and Granulomatous lymphadenitis. Ziehl-Neelson (ZN) stained smears were examined for the presence of Acid-fast bacilli (AFB) and counted under an oil immersion field (OIF). Similarly, Auramine-O (AO) stained smears were examined for the presence of Fluorescent bacilli (FB) and counted under the high power field (HPF). All the above parameters were graded according to the criteria given in Table 1 and were modified from Revised National Tuberculosis Control Programme (RNTCP) guidelines.[1112]

Table 1

Criteria for grading various parameters like necrosis, granulomas, presence of Acid Fast Bacilli and fluorescent Bacilli in Leishman’s stain, Ziehl Neelsen stain and Auramine O stain stained smears respectively

Grade	Necrosis	Granulomas	ZNa stain (modified from RNTCP guidelines)	Auramine –O stain (modified from RNTCP guidelines)
0	No Necrosis	No Granulomas	No AFBb seen.	No FBc seen
1	Focal Necrosis	Few singly scattered epitheloid cells/ill form granulomas	Occasional AFBb seen (Scanty - 1-9 AFBb in 100 OIF)	Occasional FBc seen (Scanty- 1-19/1 length)
2	More than focal necrosis	Fair number of well formed granulomas	Few AFBb(1+- 10-99 AFBb in 100 OIF; 2+ - 1-10 AFBb in 50 OIF)	Few FBc seen(1+ - 20-199 FBc/1 Length; 2+ - 5-50/1 field))
3	Predominantly Necrotic smears	Numerous well formed granulomas	Numerous AFBb seen.(3+ - 10 or more AFBb seen/OIF)	Numerous FBc seen(3+ ->50 FB/1 field)

aZiehl Neelsen; bAcid Fast bacilli; cFluorescent Bacilli

Only the cases which were culture positive and/or staining positive and/or cytologically positive with strong clinical evidence of tuberculosis infection like a response to ATT or raised ESR provided no other identifiable cause of granulomatous lymphadenitis, were included in the study. Non-cooperative patients and patients who did not fulfill the above criteria were excluded from the study.

The data was analyzed using Microsoft Excel and SPSS software. Numeric data were expressed as mean, standard deviation and analyzed using two-tailed unpaired student t-test. Categorical data were expressed as percentages and proportions and analyzed using Chi-squared (χ2) analysis (if more than two variables) or Fisher exact (FE) test (if two variables). The odds ratio was calculated wherever it was required. A P value of <0.05 was considered significant. Taking culture as the gold standard, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV), and diagnostic accuracy (DA) were calculated.

RESULTS

A total of 540 consecutive clinically suspected cases of tuberculosis undergoing FNAC were analyzed. Out of these, 492 cases were benign with most cases being diagnosed cytologically and microbiologically as tuberculosis (266, 49%), while other cases included reactive lymphadenitis (162, 30%) and acute necrotizing lesion (19, 04%). Malignancy was reported in 48 (09%) cases.

Only the cases which were culture positive; microbiological staining positive; or cytologically showing granulomas with strong clinical evidence of tuberculosis infection and no other recognizable cause of granulomatous lymphadenitis were included in the study. This included 241 cases.

The age range of the study population was 1-89 years (mean - 28.9 yrs; SD-12.7) with most patients (40%) presenting in the age group of 20-30 years. Fourteen cases were in the paediatric age group. The majority (60.2%) of our patients were females with a male to female ratio of 1: 1.7.

Most commonly i.e., 163 (67.6%) of our patients presented with cervical lymphadenitis, followed by 66 (27.3%) patients with abdominal lymph nodes along with eight (3.3%) cases from axillary lymph node, three cases from the chest wall (1.2%), and one case each of sub umbilical swelling and pleural (0.4%). The cervical lymph nodes included the submandibular, submental, and supraclavicular lymph nodes. Since the majority of our cases presented with cervical or abdominal lymphadenopathy, only these groups were analyzed statistically and are presented in Table 2. Only a minority of our cases (35%) required ultrasound guidance that too predominantly those who presented with abdominal lymphadenitis and some other non -palpable or deep-seated sites.

Table 2

Clinical, cytological and microbiological data along with site wise distribution of cases (cervical and abdominal) and their statistical significance

Variables Analyzed		Overall cases (n=241)	Cervical Lymph Nodes (n=163)	Abdominal Lymph Nodes (n=66)	Pa (Test)	Odds Ratio (OR)
Clinical features
Age in years (Mean±SD)		28.9±12.7	29.1±13.7	29.0±10.5	0.96, (t- testb)
Sex (M:F ratio)		1:1.7	1:1.7	1:1.8	1, (FE test)
Palpation vs USG guided	Palpation guided	157	145	0	<<0.05, (FE- testc)
	USG Guided	84	18	66
Macroscopic features (n=224)
	Blood mixed serous	70 (31.2%)	61 (40.7%)	09 (14.8%)
	Total cheesy/purulent	154 (68.8%)	89 (59.3%)	52 (85.2%)	0.0002, (FE-testc)	3.96
	Cheesy/purulent	138	83	42
	Blood mixed cheesy/purulent	16	06	10
Microscopic features
Granulomas (n=241)	Granulomas Present (%)	120 (50%)	93 (57.05%)	20 (30.3%)	0.0003, (FE-testc)	3.06
	Grade wise distribution (Grade 1, Grade 2, Grade 3)	61,39,20	44,32,17	14,04,02	0.18, (χ2=3.39)
Necrosis (n=241)	Necrosis Present (%)	201 (83.4%)	131 (80.9%)	62 (93.9%)	0.051, (FE- testc)	3.08
	Grade wise Distribution(Grade 1, Grade 2, Grade 3)	24,52,125	22,32,77	2,17,43	0.03, (χ2=7.14)
Cytological diagnosis (n=241)	Nec. Lymphadenitis	121 (50%)	70	46	0.0001, (χ2=13.83)
	Nec. Gran. Lymphadenitis	86 (36%)	67	16
	Gran. lymphadenitis	34 (14%)	26	4
Microbiologic features
ZNd Stain (n=231)	AFBe present (%)	91 (39.4%)	49 (31%)	40 (62.5%)	<<0.05, (FE-testc)	3.67
	Grade wise Distribution (Grade 1, Grade 2, Grade 3)	66,20,06	36,10,03	27,10,03	0.83, (χ2=0.38)
Auramine-O stain (n=200)	FBf present (%)	155 (77.5%)	91 (69.5%)	56 (91.8%)	0.0005, (FE-testc)	4.92
	Grade wise Distribution (Grade 1, Grade 2, Grade 3)	80,64,11	53,34,04	21,28,07	0.0258, (χ2=7.32)
MTBg Culture (n=156)	Culture positive (%)	97/156 (62.2%)	71/100 (71.0%)	23/46 (50.0%)	0.0164, (FE-testc)	2.45
INHh and Rifampicin Sensitivity (n=38)	Sensitivity (%)	29/38 (76.3%)

aProbability value; bStudent t-test; cFisher’s Exact Test; dZiehl Neelsen; eAcid Fast bacilli; fFluorescent Bacilli; gMycobacterium Tuberculosis; hIsoniazid

Macroscopically, most of our cases yielded cheesy/purulent aspirate (154; 68.8%) as compared to blood-mixed serous aspirate (70; 31.2%). In tuberculosis patients abdominal lymph nodes were statistically more likely (OR 3.96) to yield cheesy/purulent material as composed to cervical lymph nodes, (FE test, P = 0.0002).

On microscopic examination, granulomas were seen composed of epitheloid cells with the background showing either reactive lymphoid cells (Granulomatous lymphadenitis) or necrotic debris (Necrotizing Granulomatous lymphadenitis) as shown in Figure 1a and b. Other smears showed only necrotic debris without the presence of granulomas and the diagnosis was confirmed by microbiological staining (Necrotizing lymphadenitis). Granulomas were present in 50% of cases whereas necrosis was present in 83.4% of cases. The cervical lymph nodes were significantly more likely to reveal granulomas than abdominal lymph nodes, while abdominal lymph nodes were more likely than cervical lymph nodes to reveal necrosis, Table 2. Most of our cases were classified as necrotizing lymphadenitis (121, 50%) followed by necrotizing granulomatous lymphadenitis (86, 36%) and granulomatous lymphadenitis (34, 14%). This difference was found to be statistically significant on comparing the cervical and abdominal lymph nodes.

Figure 1

(a) Smear showing granuloma composed of epitheloid cells, granulomatous lymphadenitis (Leishman's Stain, 40X). (b) Smear showing predominantly granular necrotic debris with a granuloma composed of epitheloid cells, Necrotizing granulomatous lymphadenitis (Leishman's Stain, 10X). (c) Smear showing predominantly granular necrotic debris with presence of few acid fast bacilli, Necrotizing Lymphadenitis (Zeihl Neelson Stain, 100X, Grade 2+). (d) Smear showing fair number of fluorescent bacilli (Auramine O Stain, 40X, Grade 3+)

ZN stain was performed on 231/241 cases whereas AO stain was performed on 200/241 cases. Red, non-refractile, bacilli-shaped organisms identified in oil immersion were considered positive on ZN stained smears [Figure 1c] and fluorescent green bacilli shaped organisms identified in high power view on fluorescent microscopy, were considered positive on AO stained smears [Figure 1d].

Overall ZN positivity was 39.4% (91/231) and AO positivity was 77.5%. ZN positivity as well as AO positivity was more prevalent in abdominal lymph nodes as compared to cervical lymph nodes and this was found to be statistically significant, Table 2. On grade-wise distribution of cases, for AO stain, the abdominal lymph nodes were more likely to yield higher grades of AO positivity as compared to cervical lymph nodes. It is important to note that AO staining was able to diagnose almost double the number of cases as compared to ZN stain (77.5% vs 39.4%) and a higher number of cases showed a higher Grade in AO stain as compared to ZN stain. Statistically, if we compare ZN stain (39.4%) and AO stain (77.5%) positivity, AO stain was able to detect many more cases as compared to ZN stain (OR = 5.3) and was found to be statistically significant [Table 3]. Moreover, if we compare the grade-wise distribution of cases, AO stain was able to detect more cases in higher grades as compared to ZN stain.

Table 3

Comparative statistics for overall positivity and grade-wise distribution of Ziehl Neelsen and Auramine- O stains

	ZNa stain (n,%)	AOb stain	P c	Odds Ratio
Positive	91 (39.4%)	155 (77.5%)	<<0.05, (FE testd)	5.3
Negative	120	45
Grade 1	66 (72.5%)	80 (51.6%)	0.005, {χ2 (2)=10.473}
Grade 2	20 (21.9%)	64 (41.3%)
Grade 3	06 (6.6%)	11 (7.1%)
Overall sensitivity	39.4%	77.5%

aZiehl Neelsen; bAuramine –O; cProbability value; dFisher Exact Test

The sensitivity, specificity, PPV, NPV, and diagnostic accuracy (DA) of cytological features and microbiological microscopic tests were evaluated against MTB culture as the gold standard. The AO stain was twice as sensitive (85.9% vs. 47.4%) with higher NPV (48.0 vs. 39.3%) and DA (68.3% vs. 51.9%) as compared to ZN stain in the detection of Mycobacterium Tuberculosis. However, Specificity (22.2% vs. 60.0%) and PPV (65.3% vs. 67.6%) of ZN stain was higher. The submission of smears for both the stains may not be identical with respect to the amount of material present on the smears which may have affected the smear positivity and thus, statistical analysis. All these biases can be overcome by repeating FNAC and submitting multiple samples. The combined sensitivity of Ziehl Neelsen stain and Auramine stain was 92.05% when the two tests are taken in parallel.

MTB culture was sent in 156 cases and the rate of culture positivity was 62.2% (97/156). Culture positivity was more prevalent in cervical lymph nodes as compared to abdominal lymph nodes (71% vs. 50%, OR = 2.45) and this was found to be statistically significant. This was in contrast to smear positivity which was higher in abdominal lymph nodes. This can be explained by the fact that purulent/cheesy material was aspirated from abdominal lymph nodes which are more likely to reveal the organism on morphological analysis. However, since most of these organisms within the center of the caseous or necrotic center are dead, the evaluation by culture yielded lower results.

A drug sensitivity pattern was available in a total of 38 cases, where 76.3% (29/38) cases were sensitive to Isoniazid (INH) and rifampicin. Eight of the resistant cases were resistant to both INH and rifampicin (multi-drug resistant) whereas one case was resistant to INH but sensitive to rifampicin (mono-drug resistant).

DISCUSSION

EPTB is a subset of TB which involves organs other than the lung and represents 15-30% of cases of TB worldwide.[123] Clinically, epidemiologically, morphologically, and microbiologically, EPTB is a heterogeneous disease involving multiple organ systems which require high clinical suspicion and multiple diagnostic modalities for arriving at a correct diagnosis.[45678] Among EPTB, lymph nodal tuberculosis (LNTB) is the most common subtype in India and worldwide and accounts for 60-90% cases of EPTB cases.[13] Cervical lymph nodes are the most commonly studied site probably due to ease of approachability.[13141516171819] Similarly, abdominal TB has been studied separately in some studies, however, the number of lymph nodes included in the studies is few, with predominant cases being peritoneal and luminal TB.[20212223] Moreover, to the best of our knowledge, there are no comparative studies evaluating the cervical and abdominal lymph nodes. In the present study, we endeavored to study the clinical, cytological, and microbiological features in abdominal lymph nodes and compare them with cervical lymph nodes. The comparison of clinicopathological data from various studies with our study has been given in Table 4.

Table 4

Comparison of clinico-pathological data from various studies with our study

Details of the study	Age (Mean/Age grp with max pts)	Sex M:F	Site of EPTB & Number of Patients with TBa	Incidence (%)	Gross Features		Microscopic features			ZNb stain (%)
					Ch (%)	BM (%)	Necr (%)	NecrGran (%)	Gran (%)
Present study	28.9 yr/20-30 yrs	1:1.5	Total -241Cer.– 163Abd. - 66	49.0	68.8	31.2	50	36	14	39.4
Umar et al.,[15]	11-30 yrs	0.7:1	Total -612Cer. – 507Abd. - 04	-	-	-	17.3	75.8	6.9	37.8
Mitra et al.,[16]	25.1 yr/11-40 yrs	1:1.5	Total – 180Cer.-144Abd- Nil	45.0	66.5	33.5	30.5	40.0	29.5	51.6
Patel et al.,[17]	11-30 yrs	1:1.04	Total – 173Cer. – 339/351Abd.- Nil	49.3	-	-	27.4	71.4	-	33.9
Chand et al.[18]	35.2 yr/11-30 yrs	1:1.4	Total – 550Cervical –499Abdominal-Nil	56.9	83.6	16.4	49.8	21.8	28.4	44.5
Thakur et al.,[19]	10-30 yr	1.2:1	Total – 90Cervical-75Abdominal-Nil	62.1	46.7	53.3	-	-	-	26.7
Singh et al.,[21]	20-60 yrs	1.07:1	Total - 58Abd LN-18	-	-	-	-	-	-	-
Rana et al.,[22]	29.6 yrs/21-30 yrs	1:1.32	Total-300Abd LN- 21	-	81.0	-	47.6	19.0	33.4	-
Cho et al.,[23]	53.0 years	1:2.3	Total -139Abd LN-07	-	-	-	-	85.0	-	28.6

aTuberculosis; bZiehl Neelsen

Our population of 241 patients presented with a mean age of 28.9 years with maximum patients in the age group of 20-30 years. This was in accordance with most other studies that have reported that most patients present within the second or third decade.[14151617181922] However, some studies show presentation in the higher age groups with a mean age of 39.2 years and 59.5 years.[1320] Female preponderance was noted in our study with a male to female ratio of 1: 1.7. This was also in accordance with most of the Indian studies like Gautam et al.,[13] Monga et al.[14] and others.[1516171822] In some studies, however, male preponderance was reported.[1920]

Among the FNACs done in lymph nodes, the incidence of tuberculosis in various studies ranged from 45-62.1% which was in accordance with our study population i.e., 49%.[16171819] Other diagnoses were reactive lymphadenitis, acute lymphadenitis, metastatic malignancy, and smaller percentages of lymphomas. Macroscopically, our patient population presented predominantly with cheesy/purulent material which was similar to other studies.[161822] In our study, in addition, we found that abdominal lymph nodes were more likely to reveal cheesy or purulent material than cervical lymph nodes. Only one study previously reported cheesy material in 85% cases in abdominal lymph nodes similar to our cases which were 85.2%.[23]

Microscopically, in our study, most cases presented with necrotizing lymphadenitis (50%) and necrotizing granulomatous lymphadenitis (36%). Purely granulomatous lymphadenitis without necrosis was seen in 14% of cases. The prevalence of purely granulomatous inflammation was higher in most other studies ranging from 21.8-34.4%.[16171822] Only one study reported incidence of pure granulomatous inflammation, lower than our study, in 6.9% of their cases.[15] This feature can be explained by the bias in the above studies which predominantly represent data from cervical lymph nodes whereas our study population included a sizeable population with abdominal lymph nodes. To corroborate, in our study, the cervical lymph node were more likely to show granulomatous inflammation than abdominal lymphadenopathy which presented with predominantly necrotizing inflammation.

ZN stain positivity was found in 39.4% of cases in our study which is in accordance with most studies, however, few of the studies like those by Mitra et al.[16] and Chand et al.[18] reported higher positivity rates of 44.5 and 51.6% respectively.[1723] In addition, in our study abdominal lymph nodes showed higher positivity (3.67 times) than cervical lymph nodes (62.5% vs 31%) which was found to be highly significant.

AO stain has been used in pulmonary tuberculosis samples like sputum, bronchoalveolar lavage (BAL) fluid, bronchial brushings, bronchial washings, and post bronchoscopic sputum specimens.[2425] It has replaced ZN stain, wherever the resources are available, as a screening modality for evaluation of sputum samples, as a part of the Revised National Tuberculosis Control Programme (RNTCP), which is now National Tuberculosis Elimination Programme (NTEP).[12] This is because it has higher sensitivity, requires less time, presents higher contrast as compared to ZN stain, and is examined in high power instead of oil immersion field resulting in less fatigue and faster results. Some of the disadvantages include higher initial cost; use of toxic reagents and requirement of a fluorescent microscope. Most of these studies have reported higher sensitivity of AO staining as compared to ZN stain in the demonstration of Mycobacterium tuberculosis when compared to Gene expert/culture positivity.[2425] Moreover, they have reported faster reporting (up to 2-3 times), better diagnostic accuracy, and better value in detecting paucibacillary cases.[122425]

Despite its advantages, AO stain has been studied only recently in a few cases of EPTB, which too was limited to cervical lymph nodes, as shown in Table 5.[26272829] In these cases also similar findings have been found, highlighting the utility of AO stain. The sensitivity for ZN stain ranged from 26.2 to 44% whereas the sensitivity of AO stain ranged from 56.3 to 81.82% highlighting the higher positivity by AO stain with its inherent advantages. However, one study showed a slightly lower sensitivity of AO stain as compared to ZN stain.[30] Various diagnostic parameters of the tests like Sensitivity, Specificity, PPV, NPV, and DA showed different values in different studies depending on what was taken as the standard of diagnosis.[192930] When compared with a study using culture as the gold standard in lymph nodal tuberculosis, our results were similar.[19]

Table 5

Comparison of data on Auramine-O stain and Ziehl Neelsen stain from various studies on Fine Needle Aspiration material with our study

Details of the study	No. of Samples & Site	ZNa Stain	Auramine-O Stain
Present study	Total - 241	39.4%	77.5%
	Cervical -163	31%	69.5%
	Abdominal -66	62.5%	91.8%
Akhter et al.,[26]	Total 42 (Cervical-36, Adb-01)	26.2%	64.2%
Krishna et al.,[27]	Total- 88 (Cervical-57, Abd-None)	37.5%	81.82%
Lokeshwara et al.,[28]	Total- 50 (Cervical-40, Abd-None	44%	80%
Aalmeen et al.,[29]	Total 151 (Cervical-131, Abd-03)	40.4%	56.3%

aZiehl Neelsen

To conclude, abdominal lymph nodal tuberculosis is cytologically and microbiologically different from cervical lymph nodal tuberculosis. The abdominal lymph nodes are more likely to reveal cheesy/purulent material macroscopically, necrotizing lymphadenitis rather than granulomatous lymphadenitis along with higher ZN stain and Auramine positivity than cervical lymph nodes. Cervical lymph nodes, on the other hand, revealed a higher number of granulomatous lymphadenitis and culture positivity. This has clinical implications in patients having both cervical and abdominal lymphadenopathy, FNAC from abdominal lymph nodes are more likely to reveal diagnostic material for cytological diagnosis as well as smear positivity whereas cervical lymph nodes may be a preferred site for submitting the material for culture. Moreover, AO stain should be used as an adjunct in cytological material wherever available, which increases the positivity in smear-negative cases and is twice more sensitive than ZN stain. More such studies are required to help to delineate and characterize these differences further and modify the way we approach these entities clinically.

Declaration of Patient Consent

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Financial Support and Sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.