![]() | Only 14 pages are availabe for public view |
Abstract Lymphadenopathy is a common clinical finding in the pediatric age group. Since a large variety of disorders may lead to lymph node enlargement, determining the cause of lymphadenopathy in children can be challenging. During the present study an attempt was made to study the etiology of pediatric lymphadenopathy after correlating information from the history and clinical examination with those of the laboratory, radiological, pathological and microbiological findings. The demographic parameters, clinical presentation (s) and pathological findings of the various cases of pediatric lymphadenopathy were determined together with the identification of the role of pathology in the diagnostic workup. Finally, a simple algorithm for the primary diagnostic evaluation of children with lymphadenopathy was suggested based on the present findings and literature review. The present study was a descriptive cross-sectional study comprising 120 prospective cases of pediatric lymphadenopathy. Their age ranged from one month to 18 years. The cases were sorted into 2 broad categories: non-neoplastic and neoplastic. The non-neoplastic category was further divided into infectious and non-infectious sub-categories. Out of the 120 studied cases, 65.8 % were non-neoplastic and 34.2% were neoplastic. Out of the 79 non-neoplastic cases, 68.4% were of infectious etiology and 31.6 % were of non-infectious etiology. Bacterial lymphadenitis was the commonest etiology in the non-neoplastic category. Mycobacterial lymphadenitis was the commonest cause of granulomatous lymphadenitis. Within the neoplastic category, lymphomas were the commonest etiology and NHLs were more common than HLs. However, the etiology of lymphadenopathy could not be ascertained in 8.3% of cases even after detailed clinical and investigative profile. The demographics of the present study showed a male predominance of 70.8%. No specific sex predilection to a certain diagnostic category could be found. Most of the children (57.5%) presented in the preschool age group (less than 6 years). No significant difference regarding the age was found between any of the diagnostic categories. Most of the children came from urban areas (62.5%). A significant difference was found regarding the residence between the non-neoplastic and the neoplastic categories (P=0.002) where most of the cases within the neoplastic category (56.1%) came from rural areas. The cases were distributed nearly equal along the different seasons of the year. However, a significant seasonal variation was found between the infectious and non-infectious categories (P=0.003) where most of the cases within the infectious category (42.6%) presented in Spring. As regards the presenting symptom(s), the most frequent presenting symptom in the present study was swelling (55% of the cases). Abdominal complaints as well as the occurrence of multiple symptoms combined were significantly more frequently observed in the neoplastic categories (P= 0.003 and 0.021 respectively). Constitutional symptoms were present in 68.3% of cases. However, no significant difference in constitutional symptoms was found between any of the diagnostic categories. Most of the children had duration of symptoms more than one month but less than 6 months (46.7% of cases). A significant difference was found in the duration of symptoms between the non-neoplastic and neoplastic categories (P=0.009) and between the infectious and non-infectious categories (P=0.001). By comparing the LN status between the non-neoplastic and the neoplastic categories, a significant difference was found regarding the site (P=0.012), number (P=0.015), size (P=0.0009), character (P< 0.0001), consistency (P=0.029), mobility (P< 0.0001) and tenderness of LNs (P=0.0004) but no significant difference was found regarding the distribution and the skin changes. By comparing the LN status between the infectious and non-infectious sub-categories within the non-neoplastic group, a significant difference was found regarding the site (P=0.01), consistency (P=0.026) and tenderness of LNs (P=0.004) but no significant difference was found regarding the distribution, number, size, character and skin changes. By comparing the initial laboratory investigations between the non-neoplastic and the neoplastic categories, a significant difference was found regarding the RBCs count (P=0.002), platelet count (P=0.032), some items of the differential leukocytic count (neutropenia, P=0.034 and lymphocytopenia, P=0.008), some items in the blood film (toxic granules in neutrophils, P=0.017 and blast cells, P<0.0001) and LDH level (P<0.0001) but no significant difference was found regarding the WBCs count and ESR /CRP levels. By comparing the initial laboratory investigations between the infectious and non-infectious sub-categories within the non-neoplastic group, a significant difference was found regarding one of the items of the differential leukocytic count (lymphocytosis, P=0.045), some items in the blood film (activated lymphocytes, P=0.046 and toxic granules in neutrophils, P=0.027) and ESR/CRP level (P=0.0008) but no significant difference was found regarding the RBCs, WBCs and platelet counts. The results of this study showed the following to be factors useful in determining the risk of malignancy: presenting symptoms (abdominal and multiple symptoms), duration of lymph node enlargement (1-6 months), number of LN groups involved (generalized lymphadenopathy), number (multiple LNs), size (> 2 cm), character (amalgamated), consistency (hard), mobility(fixed) and tenderness (non-tender) , abnormal CBC findings (anemia, thrombocytopenia, neutropenia and lymphocytopenia), blast cells in blood film and elevated LDH level. A multivariate binary regression model was conducted to delineate the predictors of neoplastic lymphadenopathy. The result of the analysis showed the following 3 factors to be highly significant predictors of malignancy: presence of abdominal symptoms, multiple symptoms and LN size (more than 2 cm). In the present study, the factors associated with infectious lymphadenitis were: duration of symptoms (less than one month), number of LN groups involved (localized lymphadenopathy), consistency of LNs (cystic/soft), tenderness of LNs (tender), abnormal CBC findings (lymphocytosis), activated lymphocytes, toxic granules in neutrophils in blood film and increased ESR/CRP levels. Regarding the role of pathology in the workup, LN sampling was resorted to in 56 cases (46.7%). FNAC was resorted to in 25 cases. The outcome of FNAC was confirmed, either by subsequent excision biopsy (16 cases), follow-up of the clinical course (6 cases) or other tests (3 cases). Lymph node biopsy was done in 43 cases, out of which, 16 cases underwent both FNAC and subsequent excision biopsy and 27 cases underwent only excision biopsy without prior FNAC. The final etiologic diagnosis of lymphadenopathy was reached through pathological study of LNs in the following conditions: 13 out of the 37 cases of bacterial lymphadenitis (abscesses -mycobacterial lymphadenitis) (35.15%), 2 out of the 14 cases of viral lymphadenopathy (post-viral reactive lymphadenopathy) (14.3%), 2 out of the 3 cases of parasitic lymphadenopathy (toxoplasmosis) (66.7%), all the 8 cases with a specific histologic features and unknown etiology (100%), 5 out of the 10 non-diagnostic cases (50%) (ND/RH), all the 21 cases of lymphoma (100%) and the single case of Kaposi’s sarcoma, 2 out of the 14 cases of leukemia (14.3%) and 2 out of the 5 cases of neuroblastoma (40%). The sensitivity, specificity and diagnostic accuracy of FNAC in the diagnosis of neoplastic lymphadenopathy as compared to correct diagnosis based on excision biopsy were 90%, 75% and 85.7% respectively and as compared to the correct diagnosis based on excision biopsy, follow-up of the clinical course and other tests were 92.3%, 90% and 91.3% respectively. The present study demonstrated that primary diagnostic evaluation of childhood lymphadenopathy is mainly based on clinical grounds. In most cases, a small number of additionally performed laboratory tests allow to correctly identify the etiology of pediatric lymphadenopathy. |