Evaluation of Point-of-care Ultrasound in Children From 6 Months to 15 Years With Presumptive Tuberculosis

This study has as main objective to describe the findings obtained by POCUS methodology for pulmonary and extra-pulmonary tuberculosis in children per TB category. As secondary objectives: 1)To establish the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of POCUS positive (any of 8 pre-established sonographic signs) as a composite against a composite reference of the final diagnosis (confirmed TB, unconfirmed TB or unlikely TB). 2)To describe the sonographic signs according to HIV status (positive, negative) and to nutritional status (SAM or not SAM). 3)To determine the proportion of children with improvement / resolution of sonographic signs and the proportion without improvement of sonographic signs at 2 months of follow-up in children with and without TB. 4)To describe operational aspects of POCUS: compliance with POCUS examination; quality of POCUS; duration of POCUS; mean time to definitive diagnostic in POCUS positive results compared to POCUS negative results; proportion of children who started TB treatment on basis of POCUS positive results.

This is a multicentric study in 2 different sites: Malakal (South Sudan) and Bissau (Guinea Bissau). The methodology for the evaluation of POCUS for the diagnosis of TB in children will be longitudinal case series with diagnostic test evaluation.

Study participants: The research population are children with presumptive TB in Malakal (town and POC hospitals) or in Bissau (Simão Mendes National Hospital).

Case with diagnosis of tuberculosis:

• Confirmed TB: patients with Xpert MTB/RIF positive (or TB culture positive when available) in a respiratory sample (pulmonary TB) or in other sample, such as CSF, lymph node or other tissue/liquid (extra pulmonary TB); TB LAM positive in HIV-infected patients.
• Unconfirmed TB: no microbiological confirmation but clinical diagnosis of TB and started on anti TB treatment (as established in Appendix 4).
• Unlikely TB: good response to other treatments (antibiotic, nutritional support, etc.) and no suggestive symptoms after 1 week of admission.

Identification of participants:

Eligible children will be identified upon admission to paediatric ward, ITFC, PICU/NICU (only in Bissau) or as suggestive cases in OPD. The identification of suspected TB cases will be mainly based on clinical criteria. A diagnostic algorithm (Appendix 4) will be applied to all suspected cases. The populations of interest for this study are hospitalized children, severely malnourished (SAM) and HIV positive or exposed children.

All TB cases (confirmed or unconfirmed) will be enrolled in TB care in MSF program (Malakal) or MoH program (Bissau). MSF will ensure that all patients to be started on adequate weight-based dose fix combinations and followed up during the length of treatment. All children diagnosed of DRTB will be initiated on DRTB treatment regimen according to national recommendation.

All patients included in the study (including those with unlikely TB) will provide telephone number (if available) and address. Whenever possible and if accepted by parent or guardian, a community health worker (CHW) will accompany the patient home after discharge. This CHW will be responsible for tracing the patients not attending the follow-up appointment. The CHWs are part of the Health Promotion / Community Engagement team and routinely involved in tracing of defaulters, including HIV/TB patients.

Safety Considerations : There are safety considerations related to infection control of TB, even if children are only considered contagious after they acquire the capacity of expectorating (> 8 years old). For older children, isolation procedures for coughing patients, as well as protection measures for medical staff, need to be implemented. Lab staff needs to wear respirator and gloves when manipulating samples for Xpert MTB/RIF. Safety procedures related to Xpert MTB/RIF need to be in place.

Diagnostic procedures are routinely done in all children with suspected TB (on not specifically for the study). Gastric lavage and naso-pharyngeal aspirate can be uncomfortable. For gastric lavage, there's a small risk of spasms of vocal cords, tube entering in the airway instead of the oesophagus, minor bleeding or aspiration pneumonia. Regarding naso-pharyngeal aspirate, the actual risks for participants are still smaller; it can cause nausea or local annoyance. Lymph node puncture is a routine procedure as well and is only performed in palpable lymph nodes (≥ 2 cm). The risks are small and can comprise blood vessel punction or accidental exposure to blood for the staff. All three are common procedures which will only be performed by trained staff. Any possible complication will be treated by MSF. No specific additional safety considerations related to POCUS.

Data Management and Analysis: The number of children tested by the program for suspected tuberculosis, the number of eligible children and the number of cases included in the study during the recruitment period will be noted. The number of parent or guardian of eligible children who refuse their participation will also be registered. Demographic, clinical and diagnostic characteristics at baseline and during patient follow-up under the program will also be recorded.

Data will be collected in the files by the study team members. Data will be coded and entered into a RedCap database and analysed with SPSS software. Continuous variables (e.g. age, distance) will be summarized using mean and standard deviation or median and interquartile range as appropriate and may also be expressed as ordinal categories with frequencies; frequencies will be reported with corresponding 95% confidence intervals. The main outcome of interest for POCUS study is to describe the findings obtained by POCUS methodology for pulmonary and extra pulmonary tuberculosis in children.

Bivariate analysis is planned for TB (confirmed and unconfirmed) vs unlikely TB, HIV positive vs. HIV negative, and for SAM vs. not SAM. Children with HIV and SAM will be considered as a confounder. Comparisons of these categorical variables across groups will use Chi-squared or Fisher's exact test. All estimates will be presented with respective 95% confidence intervals. Cohen's kappa coefficient will be used to determine the inter reader agreement of TB-focused POCUS positive, negative and per sign between the operator and second reader, and third reader in case of disagreement.

The number of children screened for suspected tuberculosis, the number of eligible children and the number of cases included in the study during the recruitment period will be noted. The number of parents or guardians of eligible children who refuse their participation will also be registered. Demographic, clinical and diagnostic characteristics at baseline and during patient follow-up under the program will also be recorded. Data analysis of the study will be conducted by the Epidemiologist/Statistician involved in the study, in collaboration with the PI.

Ethical Considerations:

The study will be conducted in accordance with the Declaration of Helsinki on Ethical Principles of Medical Research on Human Subjects, and in accordance with the rules of the National Ethics Committee of South Sudan and Guinea Bissau. The protocol will be submitted for approval to the National Committee for Ethics of South Sudan and Guinea Bissau, and the MSF Ethics Committee.

Informed consent: Participation in the study is on a voluntary basis and requires prior informed consent. Trained staff will provide eligible parents or guardians with detailed information on the study's objectives and procedures, as well as a clear explanation of the risks and benefits of participation. A written information sheet will be given to the parents or guardians, in English or in local language (appendix 1). The study staff will assist the parents or guardians on all the points indicated in the form and answer all the questions. If the parent or guardian is not able to read the information sheet, staff should read it aloud.

All children will be briefed on the study and the information will be adjusted to the age range and an assent process will take place for children older than 10 years old, but following national legislature, the consent must be signed by an adult. All questions will be answered by medical staff responsible for the study.

Parents or guardians agreeing to have their child participating in the study will need to sign an informed consent, which will also be signed by the study staff. For illiterate parents or guardians, a witness who can read and write, designated by the parent or guardian himself and who will not be a staff member of the study, will be present throughout the information process and will co-sign the form to confirm that the parents or guardians has understood and willingly accepts participation in the study.

The process of informed consent will be done in an area of the clinic that offers enough privacy to make a voluntary decision without constraint. Parents or guardians will be explicitly informed that they have the right to refuse participation or withdraw from the study at any time during the research. Copies of both forms (ie. the study information sheet and informed consent, in English or in the local language) will be given to the parents or guardians.

Confidentiality: All information collected during the study will be recorded and identified by a specifically assigned number for the study and the number assigned routinely for the program. Patient or parents or guardians names will not be recorded in the study form or electronic databases. Access to the study information (in paper or electronic format) is restricted to authorized staff.

Home visits are part of the regular work of CHWs, who have specifically been trained for this task. To guarantee confidentiality, they are not informed about the reason of the visit and just deliver a request to attend the clinic as soon as possible. This information has been shared with the local authorities and is routinely explained to all patients prior to asking for consent for home visits.

Informed consent forms will be kept in a secure and confidential place in a locked cabinet by the study coordinator throughout the study. The consent forms will be archived for a minimum of 5 years after the end of the research, in a secure and locked archive by the investigator at MSF OCBA, in Barcelona. After this period, all documents will be destroyed. If the Ministry of Health requests to archive copies of the informed consents, they will be given at the end of the research study.