Radiographic and other diagnostic tools used in tuberculous malnourished children in Kassala hospitals, Eastern Sudan

Received: 22-Sep-2021 Accepted: 07-Oct-2021 Published: 15-Oct-2021

Introduction

Tuberculosis is one of the most important public health problems in Sudan and the in the federal MOH reported TB as the fourth leading cause of health service utilization, the fourth most frequent cause of hospital deaths (16%) of all hospital deaths and the most frequent reason for hospital admission (11.6% of all hospital admissions) [1].

Tuberculosis in Sudan constitute 9% of new cases in North Africa and the Middle East region making the country the third affected one in that region [2]. In June 2000 the coverage of directly observed treatment was 75% with a cure rate of more than 87% [3].

Tuberculosis in Sudan is the average Annual Risk of Infection (ARI) which is the proportion of population that is likely to be infected over a period of one year. This is estimated in Sudan to be 1.8% putting Sudan among countries of high prevalence for TB in the East Mediterranean region. The incidence of all forms of TB is therefore about 180 per 100,000 with a detection rate of only 43.6% in the beginning of this century. Prevalence of infection between 35% and 40% with death rate of about 60 per 100,000 cases [1, 3-5]

Malnutrition increases the susceptibility and or severity of tuberculosis and exerts many detrimental effects on many aspects of host immune response against mycobacterial infection [6,7]. In children it is estimated that 1.3 million children are infected annually and the annual deaths due to TB in children reaches 450,000 worldwide [8]. TB may be responsible of about 10% of hospital admission of children in the developing countries and about 10% of hospital deaths [9].

The size of the Mantoux test was associated with the degree of malnutrition and malnutrition was associated with atypical presentation of pulmonary TBin one study [10]. BCG vaccination should not be considered in the interpretation of a positive PPD test [11]. In a Turkish study on adults and adolescents with TB the authors evaluate the results of simultaneous application of PPD and BCG tests in order to assess their efficacy. The conclusion was that BCG is more sensitive and specific than PPD in diagnosis of TB in the study group [12]. Diagnosis of TB in children is mainly epidemiologic and clinical with use of chest radiographs and skin testing bacteriological diagnosis is of lower yield in childhood tuberculosis even in developed countries and for most children with pulmonary Tb culture confirmation is not needed [13,14]. Use of 5 tuberculin units of PPD is the standard method to detect TB infection but host factors may affect and diminish the area of induration. The test should be read within 48 to 72 hours [13-16]. Radiology should be used in a positive TST child case and it may show evidence of lymphadenopathy or lung parenchymal changes pleural lesions and cavitation are not common in tuberculosis children [15]. DNA amplification tests with polymerase chain reaction are sensitive and specific but they do not replace the routine ZN staining for AFB or culture. PCR gave positive yield results in some studies ranging from 25% to 100% [17-19]. Despite a lot of studies serology has found little place in routine diagnosis of TB example is ELISA test [20].

Diagnostic BCG test in tuberculosis children the reaction to BCG is accelerated with papules, pustule and scar appearing in few days instead of 2 to 3 months with particular importance in malnourished children [11,12]. A study in Turkey showed positive accelerated BCG in healthy children to be 100% and in malnourished children to be 82% while the Mantoux was positive in 44.5% and 18 % only in the same group of children Also similar results obtained in India [21,22]. Clinical diagnostic criteria include the WHO TB score chart depending on the duration of illness the decrease of weight for age and presence of family or other close contact past or present. Then scoring for other clinical or skin test features. A score of 7 or more is considered suggestive and treatment is recommended [23]. The presence of 3 main WHO criteria has a positive predictive value of 63% and in a study in Zambia on 75 tuberculous children the sensitivity of the WHO chart was found to be 88% while the specificity was low and is only 20% [24]. IUATLD formed a task group from 10 countries to develop a score model for childhood TB and 5 clinical criteria were adopted as the most relevant disease predictors in children: History of contact with a case of TB, Positive skin test, Persistent cough, Low weight for age and unexplained prolonged fever.

The model is of low specificity of less than 70% but good positive predictive value of (60%-77%) [25]. The objective of this study was to evaluate the use of the WHO clinical score criteria and accelerated BCG test reaction in the diagnosis of tuberculosis in malnourished children attending Kassala hospitals.

Materials and Methods

Study design

This was across sectional hospital based study.

Study area

Kassala State in Kassala Teaching and Kuwaiti paediatric hospitals and refered patients from peripheral refugee camps and rural hospitals.

Study period

One year from January 2010 to December 2011.

Sample size

84 patients.

Inclusion criteria

All newly diagnosed children as cases of tuberculosis of age less than 60 months and were classified as malnutrition or underweight according to the clinical Welcome-Trust classification.

Exclusion criteria

Neonates, age more than 60 months, parental or care giver refusal to enrol in the study, concomitant chronic medical disease.

Research tools and techniques

A questionnaire was designed to contain full information about the nutritional social and medical history of the patient. Also examination findings anthropometric measurements and relevant investigation findings records for every patient were made. Anthropometric measurements done were the weight, height or length, and the head and mid-upper arm circumference and these were plotted against the suitable centile charts and the results were recorded. weight of every child as unclothed and in grams.

Investigations include

The Mantoux testing (use of PPD of 5 TU strength to give an intradermal injection of 0.1 ml to produce a wheal of about 5 mm on upper third of flexor surface of left forearm. Results were read in 48 hours-72 hours), Radiology; chest x-ray, E.S.R (Westergren device), Haemoglobin concentration (calorimeter), and WBC count and differential using a haemocytometer, BCG test IN Mantoux negative patients giving freeze dried BCG of concentration 0.1 mg/0.1 ml to inject intradermal 0.1 ml on lateral aspect of upper left forearm. Test considered accelerated when papule appears within 2 days, pustule within the first week and/or scab within the first 2 weeks. ZN stain for gastric aspirate and some further investigations as relevant.

Ethical consideration

Ethical approval from the hospital authority was requested and obtained prior to the start of the study (dated on 15.11.2009).

Also obtaining informed consent for each patient from parents or guardian direct interview for each patient performed by the author to fill the questionnaires.

Data entry and statistics

Data obtained was entered in a computer program (SPSS) and appropriate tests for significance levels were used.

Results

A total of 84 tuberculosis malnourished children were enrolled in this cross sectional study. The males constituted 51.2% and the females 48.8%. They were of different local ethnic groups and the mean age was 28.8 months ranging from 5 to 59 months. The different patterns of malnutrition classifications of study group are shown in Figure 1 and the different pattern of TB presentation in the study group is shown in Figure 2.

Figure 1: The Welcome’s classification of malnutrition in the study group (n = 84)

Figure 2: The pattern of tuberculosis disease in the study group (n=84)

The positive chest x-ray features in the study group was seen 74 patients (88%) with hilar shadows in 22 patients (26.1%) consolidation in 10 patients (12%) miliary shadows in 7 children (8.3%) and consolidation-collapse shadows in 5 patients (6%) as in Figure 3. There was no significant statistical association between x-ray features and either of the WHO clinical score (p value>0.811) nor with the IUATLD score (P value>0.775).

Figure 3: The chest x-ray findings in the study group (n=84)

The reaction of Mantoux test was 10 mm and above in 18 patients (21.4%) as shown in Table 1 with no significant correlation with the type of TB disease but was significantly associated with the classification of malnutrition (P>0.001) (Table 2).

Tab. 1. Mantoux skin test size the study group (n=84)

Tab. 2. Comparison between Mantoux test size and PEM classification in the study group (n=84)

The BCG test was done in the 66 patients who were considered negative for Mantoux test (78.6% of the total patients) and the test was accelerated in 51 of these (60.7% of the total number of patients). This diagnostic BCG test showed no significant relation with

The classification of malnutrition, the WHO score chart, the IUATLD score test, nor with the age of the child in months. Statistical relationship to the age of the child is shown in Table 3.

Tab. 3. Co-relation between diagnostic BCG reaction and age in months in the study group (n=84)

The ESR was moderately or highly elevated in 67 patients (79.7%) and 40 of the children showed anaemia of less than 8 g/dl. ZN stain for gastric aspirate was done for 24 patients and positive results obtained in 3 of them (3.6% of total patients). The WHO TB score chart was positive of ≥ 7 in 66 patients (78.6%) shown in Table 4 and was statistically significantly associated with the chest x-ray features (p>0.027) but not associated significantly with type of TB, history of measles, nor the size of Mantoux test. The IUATLD clinical score was positive of ≥ 5 in 75 patients (89.3%). This score was significantly associated with the duration of symptoms, with presence of cough and with x-ray features of TB (p>0.001) but no significant association with type of TB disease, the positive BCG test, the Mantoux size nor the previous history of measles.

Tab. 4. The WHO tuberculosis score chart (TSC) in the study group (n=84)

Discussion

Mantoux test was negative in 95.1% in the study of Adil, et al. but the test size had no significant association with the mortality [26]. Similar results to this study were obtained by Go CA, et al. in Turkey in their study on in which out of 22 malnourished tuberculosis children only 4 (18%) had a positive Mantoux test [21]. Studies on non-severely malnourished children suffering from tuberculosis on Sudanese children in Khartoum by Zien, et al., Magda, et al., and Moniem, et al. showed highly positive Mantoux of 88%, 92.8% and 81% respectively [27-29].

In a large Turkish study by Gocmen, et al. on 2205 62% have a positive Mantoux reaction [30].

BCG test was done in 66 patients in this study and was accelerated in 51 of them (77.3% of the total patients) several studies recommended the use of accelerated BCG as a simple method in the diagnosis of TB particularly in children [31].

Indian studies by Chanchuri, et al. and Chandra, et al. have shown that the positivity of accelerated BCG in diagnosing TB in is much higher than the Mantoux test [22,32]

Similar results were obtained by the studies done in other parts as in Turkey by Gocmen A et al and sonomez et al [12,30].

In my study the combined use of Mantoux test and of BCG in Mantoux negative patients gave a positive skin test reaction in 82.1% of the patients in the study.

In Sudanese studies on tuberculosis children radiography was positive in in 75% to 88% of the cases similar to the results obtained in this study.

Microscopy and ZN staining of gastric aspirate usually gives a low yield in children as in this study (only 3 patients out of 24 tested that is 3.6% of the total patients). This was also the case in studies in Sudan and other African studies. The large study by Dias on 627 TB children showed positive AFB culture in only 5% of the patients [33].

All lymph node biopsies in previous studies on Sudanese children suspected for TB were suggestive as in my study [27,29].

Although the ESR was moderately or highly elevated in 78 patients (92.8) but prevalence of anaemia and associated infections may affect the results and their interpretation.

Sixty patients (78.6%) had a WHO TSC score of ≥ 7 and 75 patients (89.3%) had a positive IUATLD score of ≥ 5 indicating that the sensitivity of both scores is high, but their specificity evaluation needs further sophisticated diagnostic tools which are not feasible such as PCR. Criticism had been directed to the use of these scores in diagnosing paediatric TB in chronically debilitated children as in malnourished children [34].

A study by Rheenen in Zambia showed high rate of false positive WHO score chart for TB in malnourished children and in spite of the high sensitivity of the score of 88% the specificity was only 25% [24].

A large study to evaluate the IUATLD scoring criteria for TB diagnosis in children Done in India by Suryan on 607 children concluded that malnutrition gives large number of false positive results and recommended that objectivity increased by adding radiological evidence to the score, the cut off point for positivity fixed higher, the weight of symptoms scaled down and the test used for screening only [35].

Conclusion and Recommendations

From this study done on malnourished tuberculosis suffered children in Kassala hospitals, it was concluded that pulmonary TB was the prevalent form of disease in this tuberculosis malnourished group of children and that the Mantoux skin test is not of good diagnostic value in this group as it is only positive in about one fifth of the patients, but the accelerated BCG test was of much help in these patients (was positive in near eighty percent of the tested children) the greatest skin test positivity and benefit to diagnosis could be obtained in these affected children when both tests used together.

The WHO score and the IUATLD score in tests were found to be sensitive in diagnosing these patients but nonspecific parameters are used in them and they generally appear non-specific

The followings are recommended:

➢ The use of BCG test in Mantoux negative malnourished children suspected for TB Will raise the skin test positivity (82.1% in this study )

➢ Adoption of the recommended scores modifications for debilitated children including that the objectivity increased by adding radiological evidence to the score, the cut off point for fixed higher, the weight of symptoms scaled down and the test used for screening only.

➢ Human development programmes and efforts in the area reinforced in improving education levels and better nutrition, housing conditions. Vaccination coverage and sanitation.

References

1. Asma IE. Manual of the national TB control programme in Sudan 2nd ed. Sudan. Federal MOH; 2000:1-51.
2. Dye C, Scheele S, Dolin P, Pathania V, Raviglione MC. Consensus statement. global burden of tuberculosis estimated incidence, prevalence and mortality by country. WHO global surveillance and monitoring project. JAMA. 1999;282:677-686.
3. Federal MOH. National TB control Programme background. Sudan. 1998:1-10
4. Federal Ministry of Health. Directorate of preventive and social medicine. National Tuberculosis Control Programme Annual Reports. 2000.
5. Federal Ministry of Health. National tuberculosis control programme. Progress report. Sudan. January-June 2000.
6. McMurray DN, Bartow RA, Mintzer CL, Frontera EH. Micronutrient status and immune function in tuberculosis. Ann N Y Acad Sci 1990;587:59-69.
7. McMurray DN. Cell mediated immunity in Nutritional deficiency. Prog Food Nutr Sci. 1984;8:193-228.
8. Donald PR. Children and Tuberculosis Protecting the next generation Lancet interactive.Tuberculosis progress report. 1999.
9. Starke JK, Munoz F. Tuberculosis. In: Beherman RE, Kliegman RM, Jenson HB, editors. Nelson,s Text book of paediatrics 16th edition Philadelphia. 2000.
10. Madebo TNY. Saeter G, Lindtrn B. HIV infection and malnutrition change the clinical and radiological features of pulmonary tuberculosis. Scand J Infect Dis. 1997; 29:355-359.
11. Giesielski SD. BCG vaccination and the PPD test; what the clinician need to know. J Fam Pract 1995;45:76-80.
12. Sonmez E, Yakinci CA, Aladag M, Evliyaoğlu E, Yologlu S, et al.  Diagnosis of tuberculosis PPD or BCG test. J Trop Med 1998;44:40-42.
13. Dunlap NE, John B, Robert C. Diagnostic standards and classifications of tuberculosis in adults and children. 2001;9:781-800.
14. Shinnick T, Good R. Diagnostic mycobacteriology laboratory practices. Clin Infect Dis. 1995;21:291-299.
15. Khan EA, Starke JR. Diagnosis of tuberculosis in children increased need for better methods. Emerg Infect Dis. 1995;1:115-123.
16. Krugman S. Samuel LK. Infectious diseases of children 7th ed. Philadelphia: Mosby company;1981:427-477.
17. Starke JR, Ong LT, Eisenach K, Smith KC, Denby M. Detection of M. Tuberculosis in gastric aspirate samples from children using polymerase chain reaction. Pediatrics. 1996;97:155-160.
18. Janata SK, Nair MN, Lahiri KK, Sarin NP. Polymefrase chain reaction in the diagnosis of tuberculosis. Indian Pediatr. 200;37:375-382.
19. Montenegro SH, Robert HG, Rosa C, Sheen P, Hopper T, et al. Improved detection of mycobacterium tuberculosis in Peruvian children by use of Heminested IS 6110 polymerase chain reaction assay. Clin Infect Dis. 2003;36:16-23.
20. Delacourt C, Gobin J, Gaillard JL, Blic JD, Veron M, et al. Value of ELISA using antigen 60 for the diagnosis of tuberculosis in children. Chest. 1993;104:393-398.
21. Go CA, Kiper N, Ertan U, Kalaycì O, Ozçelik U. Is the BCG of diagnostic value in tuberculosis. Tuber Lung Dis .1994;75:54-57.
22. Chandra K, Uperndra C.Evaluation of Mantoux and BCG test in the diagnosis of childhood TB .Indian Pediatr 1997;2:99-102.
23. WHO report. ICMR/WHO scientific group on vaccination against TB. 1980.
24. Rheenen PV. The use of pediatric tuberculosis score chart in an HIV endemic area. Trop Med Int Health. 2002;7:435-441.
25. Fouri PB, Becker PJ, Festenstien F, Migliori GB, Alcaide J, et al. Procedtures for developing a simple scoring method based on unsophisticated criteria for screening children for tuberculosis. Int J Tuberc Lung Dis. 1998;2:116-123.
26. Babiker AA. Clinical biochemical and some cell mediated immune responses in Sudanese children with protein energy malnutrition. University of Khartoum, Sudan 1994.
27. Ibrahim SA, Karrar ZA, Bashar S. The pattern of childhood tuberculosis in Kartoum. Sudan Med J. 1993;31:1-5.
28. Kamal -Eldin MA. A controlled clinical trial of short term chemotherapy in treatment of childhood pulmonary tuberculosis. University of Khartoum, Sudan. 1996
29. Elhag AA. Tuberculosis in Sudanese children, a hospital based study. University of Khartoum, Sudan. 2000.
30. Gocmen A, Kiper N, Ozcelik U, Senuyar R, Cengizlier R, Childhood tuberculosis a report of 2205 cases. Turk J Pediatr. 1997;39:149-158.
31. Rajajee S. Tuberculin and BCG test. Indian J Pediatr. 2000;67:9-13.
32. Chanchuri N, Shaw BP. Value of Mantoux test in the diagnosis of tuberculosis in chronic illness. Indian J Pub Health. 1993;37:37-41.
33. Senuyar
34. Hesseling AC, Schaaf HS, Gie RP, Starke JR, Beyers N. A critical review of the diagnostic approaches used in the diagnosis of childhood tuberculosis. Int J Tuberc Lung Dis. 2002;6:1038-1045.
35. Surynara L, Jagannatha PS. Scoring method for diagnosis of tuberculosis in children. National Conference of TB, Patna, India. 1999.