Keywords
Typhoid fever, Widal's test
This article is included in the Emerging Diseases and Outbreaks gateway.
The study aimed at determining the predictive values of Widal’s test against culture and the antimicrobial sensitivity profile of Salmonella typhi (S. typhi) isolates from patients attending private health facilities in Mbale city, Uganda. This was to provide evidence-based information to policy implementers for guideline enforcement.
This was a quantitative cross-sectional study in which 172 participants were recruited. Blood and stool samples collected were cultured; and to each stool sample the following tests were done; Xylose Lysine Deoxycholate (XLD), Salmonella Shigella agar (SSA), and MacConkey. Isolates suggestive for S. typhi were examined with triple sugar iron (TSI) and SIM (Sulphur, indole, and motile) and antibiotic sensitivity test (Ciprofloxacin, Imipenem, Ceftriaxone, Chloramphenicol and Augmentin for sensitivity testing) conducted with Mueller Hinton respectively. Generated data was entered in the Epidemiological information (Epi Info) software (version 7) and analyzed using the contingency table to determine the sensitivity, specificity, positive predictive value and the negative predictive value. The analysis was then presented in form of graphs and tables.
The positive predictive value (PPV) for Widal’s test was 6.8% while the negative predictive value (NPV) was 100%. Three isolates were sensitive to Ciprofloxacin (3, 100%) and Imipenem (3, 100%), intermediate to Ceftriaxone (2, 67%) and resistant to Chloramphenicol (2, 67%) and Augmentin (2, 67%).
In comparison with culture, Widal’s test is highly sensitive (100%) with a low specificity of 76% and PPV (6.8%) and NPV (100%). It should therefore be followed up with a more specific confirmatory test. S. typhi was highly sensitive to both Ciprofloxacin and Imipenem; and resistant Chloramphenicol and Augmentin.
Typhoid fever, Widal's test
The second version supersedes the first version following revision of the manuscript. Changes have been made in captions of figures, title of tables, additional information on how sensitivity, specificity, positive predictive value and negative predictive value were calculated. Added included explanations/justifications on why both blood and stool were considered in this study; and what gap was to be filled by this paper's findings. We also included possible reasons why findings of this study were disagreeing with other study's findings.
See the authors' detailed response to the review by Letemichael Negash Welekidan
The global burden of typhoid fever was estimated at 17 million cases with 178,000 deaths in 2015 with more cases being reported in South Asia, South East Asia, and Sub-Saharan Africa.1 Whereas typhoid fever appears to be less common or under ascertained in Africa compared to Asia, more recent studies confirm that typhoid fever incidence is high in some parts of Africa.2
The prevalence of the fever in Uganda was reported to be 7.6% during the February, 2015 disease outbreak that started from Kampala City3 and by the month of December 2015, the cases were reported by nearly all districts in Uganda with central region reporting the highest numbers followed by Northern, Western and Eastern respectively. In Mbale specifically, the prevalence of typhoid fever was reported at 1.3%.4
The recommended diagnosis for typhoid fever in Uganda is by use of culture, rapid antibody tests, Polymerase Chain Reaction (PCR) and relative leukopenia tests and not Widal’s test5; and the antibiotics used for treatment include Ciprofloxacin, Chloramphenicol, Ceftriaxone, and or Amoxicillin depending on the fever’s severity.5 Private health facilities in Mbale city are still depending on the Widal’s test that is no longer recommended to diagnose typhoid fever which isn’t the case in public health facilities. The test is not recommended because of its low positive predictive value (60%) and low negative predictive value (75.2%).6 The non-compliance to the diagnostic guidelines of diagnosing typhoid fever is causing inappropriate drug prescription and multidrug resistance development against antibiotics like ampicillin, chloramphenicol and ciprofloxacin.7
Since accurate diagnosis is crucial to the management of typhoid fever,7 the research aimed at determining the predictive values of Widal’s test against culture and the antimicrobial sensitivity profile of S. typhi isolates from patients attending private health facilities in Mbale city, Uganda; and give information about how typhoid fever is being managed in private health facilities, act as a basis for developing or re-enforcing policies on typhoid fever management, and also act as a reference to other researchers adventuring in the same field of study.
This study was approval by the Mbale regional referral hospital research ethical committee (MRRH REC) with a reference of MRRH-REC IN-COM 006/2018 before data collection. MRRH REC is accredited by UNST with a registration number of UG-REC-011. Permission was sought from Mbale city authorities and private health facility authorities before conducting the study. A written consent was sought from the participants prior to recruitment into the study. Codes instead of names were used as sample identification numbers for confidentiality purposes. Results obtained were availed to the clinicians and the research participants for health management purposes. The S. typhi isolates and the data summary was kept safely in the Mbale regional referral hospital microbiology laboratory.
This was a quantitative cross-sectional study that recruited 172 patients clinically suspected for typhoid fever and seeking treatment from private clinics in Mbale city between December, 2017 to January, 2018. A systematic sampling technique was used to identify and enroll four participants (every 1st, 3rd, 5th and 7th) from each private health facility into the study. The sample size was used to obtain the number of private facilities to be enrolled into the study (i.e., sample size/number of expected participants per facility). The study included typhoid suspects seeking diagnosis and treatment in private health facilities during the period of study and consented in writing to participate in the study. Those who were on antibiotics within one month of the study were excluded. Both blood samples (10 ml and 2 ml) and stool samples (2 g) were collected from the research participants for study and treatment purposes.
Using a vacuum extraction blood-sampling system, 10 mls of blood was aseptically collected from the research participant and dispensed directly into BD blood culture bottle. 2 ml syringe was used to aseptically collect 2 ml of blood from the research participant, discarded the needle, opened the plan vacutainer tube to dispense the 2 ml volume of collected blood; this was used for Widal’s test by the laboratory staff at the private health facility as routine. The private health facility continued with their usual technique of tile rocking as they observed for agglutination while performing the Widal’s screening test using serum as a sample. To widen the scope of the investigation, and following orientation, research participant scooped 2 g of fresh stool samples using the spoon in the clean leak-proof containers that were free from disinfectant or detergent residue and recapped with tight-fitting, leak-proof lids; and delivered to the private health facility laboratory where it was inoculated into a Cary Blair media for transportation to the microbiology laboratory for culture. Both blood and stool samples including the request form were triple packaged and transported to the microbiology Laboratory Unit at Microbiology Department in Mbale regional referral hospital. Other information collected from the research participants included demographic data such age and gender and the Widal’s test result at the private health facility.
BD blood culture bottles inoculated with blood samples were inserted into the Bactec machine for incubation set at 37°C for up to seven days. While, Stool samples were inoculated on XLD, SS agar (placed in the incubator set at 37°C for 24 hours) and MacConkey was first placed into anerobic gar before placing it in the incubator to limit the background of environmental flora and allow only S. typhi to grow. Gram reaction testing, triple iron sugar (TSI), and SIM biochemical testing were performed on isolates for identification purposes. The susceptibility of each bacteria to each antibiotic disk (Ciprofloxacin, Ceftriaxone, Azithromycin, Amoxiclave, Imipenem and Chloramphenicol) was done using the disc diffusion method and results recorded as resistant (R), Intermediate (I) and Sensitive (S).
Data generated from the study was entered in the Epidemiological information (Epi Info) software (version 7) and analyzed using the contingency table to determine the sensitivity, specificity, positive predictive value (PPV) and the negative predictive value (NPV). Sensitivity was calculated as number of research participants with culture positive and Widal reactive test results divided by total number of those with culture positive results and multiplied by 100. Specificity was calculated as number of research participants with culture negative and Widal non-reactive test results divided by total number of those with culture positive results and multiplied by 100. PPV was calculated as number of research participants with Widal reactive and culture positive test results divided by the total number of those with Widal reactive test results and multiplied by 100. The NPV was calculated as number of research participants with Widal non-reactive and culture negative test results divided by the total number of those with Widal non-reactive test results and multiplied by 100. The analysis was then presented in form of graphs and tables.
The personal protective equipment (examination gloves, lab coats and closed shoes) was put on during sample collection and sample processing procedures. The media (except blood culture broth) were prepared aseptically, and incubated to rule out contamination. All Laboratory procedures were done following the standard operation procedures (SOPs) guide lines of the Mbale regional referral hospital microbiology laboratory.
A total of 172 participants, 55% female and 45% male, aged between <17 to >50 years were enrolled for the study. Majority of the participants were aged between 18 and 34 years and few were those above 50 years of age (Figure 1).
Occurrence of S. typhi positive participants
Out of the 172 Widal’s tests that were done in the private health facility, 44 (25.6%) were reactive and 128 (74.4%) were non-reactive. When compared to the gold standard, culture had 3 (1.7%) S. typhi isolates and 169 (98.3%) did not have any significant growth (Table 1).
The samples were drawn from 43 private health facilities of which only three (7%) had S. typhi isolated from their referred samples (Figure 2).
S. typhi was isolated more from the male participants (67%, 2/3) as compared to the female counterparts (33%, 1/3). These isolates were mainly obtained from two age groups namely 18-34 years and 35-49 years with more amongst the 18-34 years age group (67%, 2/3). The S. typhi was isolated from three (03) stool samples and none from blood samples.
Diagnostic accuracy study
This study showed the Widal’s test diagnostic accuracy to be as follows: sensitivity (100%), specificity (76%), positive predictive value (6.8%) and negative predictive value (100%) (Table 1).
Drug sensitivity patterns
This study showed that the drugs i.e., Chloramphenicol and Augmentin demonstrated resistance each at 67% (2/3). Those that were highly sensitive included Ciprofloxacin and Imipenem each at 100% (3/3). Ceftriaxone 67% (2/3) was intermediate (Table 2).
This study revealed Widal’s test PPV as 6.8% suggestive of a possibility of not all participants with reactive Widal’s test result had typhoid fever.8,9 This finding was similar to most of studies reviewed whose PPV was below 60% with exception of Minjibir (2020) whose PPV was 68.2%.6,10–15 Widal’s test may not be not dependable for diagnosis of typhoid fever.14 This study revealed Widal’s test NPV as 100% suggestive of a possibility of all participants with non-reactive Widal’s test result as not having typhoid fever.8,9 This finding was similar to most of the studies reviewed whose NPV was above 90% with exception of Mengist (2017) and Mawazo (2019) who obtained 75.2% and 89.7% respectively.6,10–15 The difference with Mengist (2017) and Mawazo (2019) could be because of different brand of Widal’s test reagent used. Therefore, the Widal’s test result can serve as a good indicator of the absence of typhoid fever.11 This study revealed that the Widal’s test had a sensitivity of 100%. This indicated that, the Widal’s test had the ability to rule out the presence of typhoid fever and this surpasses that of Widal’s test acceptable of 70%.16 This finding was similar to most of studies with exception of Yousif (2018) whose value was 26.5%.6,10–15 This study also revealed that the Widal’s test has a specificity of 76%. This indicated that Widal’s test had the ability to identify positive cases although with a high false positive rate of 93.2% (41/44). This was so probably because typhoid fever is uncommon among the participants hence giving a large proportion of false positive test results and this explains partly why this study’s positive predictive value was low (6.8%).17 This finding was similar to most of the studies reviewed whose value was above 70% specificity with exception of Deksissa and Gebremedhin (2019), Mawazo (2019) and Muthoni (2016) whose values were 44.4%, 18.3% and 47.7% respectively.6,10–15 The difference could be due to different brand of Widal’s test reagent used. As described by many scholars, false diagnosis of typhoid fever can lead to an inappropriate treatment which eventually leads to antibiotic resistance development, a serious threat to the global public health which if not addressed in time, may lead to mortality.18,19 This study described the antibiotic sensitivity profile of S. typhi. Chloramphenicol was 67% (2/3) resistant, a finding was similar to Wasihum et al (2015) and Deksissa and Gebremedhi (2019) who reported 76.1% and 20% resistance respectively.11,20 Augmentin was 67% (2/3) resistant, a finding similar to Wasihum et al (2015) and; Deksissa and Gebremedhi (2019) also reported 2.6% and 100% resistance.11,21 This study’s finding was however contrary to Chaudhary et al (2016) who reported Augmentin as 94.3% sensitive.10 This study also revealed that Ciprofloxacin was 100% (3/3) sensitive which finding was contrary to Wasihum et al (2015) who reported (76.1%).21 This study’s finding of Ceftriaxone being 67% (2/3) intermediate was contrary to Deksissa and Gebremedhi (2019) who reported 100% sensitivity.11
The limitation of this study was the very few S. typhi isolates since this did not provide adequate information on the antibiotic resistance profile of Mbale city. Future studies would increase the scope and study to obtain more S. typhi isolates such that adequate information on the antibiotic profile is obtained.
In comparison with culture, the gold standard, Widal’s test is highly sensitive (100%) with a high negative predictive value (100%) but low specificity (76%) and PPV (6.8%) S. typhi was highly sensitive to both Ciprofloxacin and Imipenem; and resistant Chloramphenicol and Augmentin. This urgent need for policy implementers to enforce adherence to Uganda Clinical Guidelines on typhoid fever diagnosis in Mbale city.
This study is significant in that, it gives information about how typhoid fever is being managed in private health facilities, act as a basis to develop or re-enforce policies on typhoid fever management, and act as a reference to other researchers adventuring in the same field of study. Inappropriate typhoid diagnosis in Mbale city private health facilities could lead to inappropriate antibiotic prescription, antibiotic resistance development and death. Conducting this comparative study and determining the antimicrobial sensitivity profile will provide information on typhoid fever management in private health facilities in Mbale city.
The data that support the findings of this study are available on request from the corresponding author, [SG]. The data are not publicly available because they contain information that could compromise the privacy of research participants. If the reader needs access to the said data, they could do so by writing an email to the corresponding author, [SG] detailing justification for the primary data source request.
The author(s) would like to thank Makerere University for technical support while conducting the study; and with permission, the author(s) acknowledges the administrative support from Rt. Rev. Patrick Gidudu, DR. Can. Hannah Gidudu, Rev. Prof. Dr. Jeanette Meadway and Mr. John Meadway that facilitated the study. The views expressed in the submitted article are author(s) own and not an official position of the institution.
Views | Downloads | |
---|---|---|
F1000Research | - | - |
PubMed Central
Data from PMC are received and updated monthly.
|
- | - |
Is the work clearly and accurately presented and does it cite the current literature?
Yes
Is the study design appropriate and is the work technically sound?
Yes
Are sufficient details of methods and analysis provided to allow replication by others?
Partly
If applicable, is the statistical analysis and its interpretation appropriate?
Yes
Are all the source data underlying the results available to ensure full reproducibility?
Yes
Are the conclusions drawn adequately supported by the results?
Partly
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Epidemiology, genomics of Enteric pathogens and AMR
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Microbiologist
Is the work clearly and accurately presented and does it cite the current literature?
Partly
Is the study design appropriate and is the work technically sound?
Partly
Are sufficient details of methods and analysis provided to allow replication by others?
Partly
If applicable, is the statistical analysis and its interpretation appropriate?
Partly
Are all the source data underlying the results available to ensure full reproducibility?
Partly
Are the conclusions drawn adequately supported by the results?
Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Microbiologist
Alongside their report, reviewers assign a status to the article:
Invited Reviewers | ||
---|---|---|
1 | 2 | |
Version 2 (revision) 26 Feb 24 |
read | read |
Version 1 17 Mar 22 |
read |
Provide sufficient details of any financial or non-financial competing interests to enable users to assess whether your comments might lead a reasonable person to question your impartiality. Consider the following examples, but note that this is not an exhaustive list:
Sign up for content alerts and receive a weekly or monthly email with all newly published articles
Already registered? Sign in
The email address should be the one you originally registered with F1000.
You registered with F1000 via Google, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Google account password, please click here.
You registered with F1000 via Facebook, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Facebook account password, please click here.
If your email address is registered with us, we will email you instructions to reset your password.
If you think you should have received this email but it has not arrived, please check your spam filters and/or contact for further assistance.
Comments on this article Comments (0)