Diagnostic efficacy of Widal slide agglutination test against Widal tube agglutination test in enteric fever

Introduction: Enteric fever is an endemic disease in India and warrants rapid and affordable diagnosis. The Widal slide agglutination test is a commonly used rapid screening test for this purpose. The literature available on its diagnostic ability in comparison to the tube agglutination test is however scanty. Hence, this study aims to evaluate the effi cacy of the Widal slide agglutination test and the tube agglutination test for the diagnosis of enteric fever. Materials and Methods: A total of 1470 sera were collected during the study period of one year from patients having pyrexia of unknown origin. All the samples were tested for the presence of anti O and anti H agglutinins against S. typhi and S. paratyphi A by semi quantitative slide and quantitative tube agglutination tests as per standard protocols. The titers of 1:80 (O agglutinins) and 1:160 (H agglutinins) were taken as the signifi cant titer for the diagnosis of enteric fever. The results of the slide agglutination test were compared with the tube agglutination test and analyzed using Fisher’s exact test. The sensitivity, specifi city, positive and negative predictive values of slide agglutination were calculated using the tube agglutination method as a standard for comparison. Results: Of the 294 slide positive samples, 209 (71.1%) samples tested negative by the tube agglutination test. The sensitivity, specifi city, positive and negative predictive values for the slide agglutination test were observed to be 100 % (CI 95.75-100%), 84.91% (CI 82.93-86.73%), 28.91% (CI 23.84-34.45%) and 100% (CI 99.69-100%) respectively. Conclusion: Serological diagnosis of enteric fever should always be confi rmed by the tube agglutination test rather than depending solely upon the rapid slide agglutination test results.


INTRODUCTION
Enteric fever is an important cause of morbidity in many regions of the world, with an estimated 13 million cases occurring annually in Asia alone. [1]Estimates suggest an incidence rate of more than 21.5 million cases globally in the year 2000. [2] India, enteric fever is caused by Salmonella enterica serotype Typhi and Paratyphi A. Serotypes B and C are very rare.Laboratory diagnosis mainly depends upon isolation of causative agents from specimens like blood and bone marrow.A blood culture gives positive results in 73-97% cases before the use of antibiotics. [3]However, the availability of microbiological culturing facilities is often limited in many typhoid endemic regions and blood cultures can be negative when patients have received prior antibiotic therapy. [1]Hence, serological diagnosis using Widal test is relied upon in many cases.International  agglutination test requires greater technical work than the more rapid slide test, and is a macroscopic test.It is useful to clarify erratic or equivocal agglutination reactions obtained by the more rapid slide test. [4]However, since it takes 18-24 hours to get the results; practically the diagnosis is often formed on the basis of the slide agglutination results which are available within minutes.Using slide agglutination as a screening test without confi rming the results by tube agglutination may lead to a false diagnosis of enteric fever and an avoidable introduction of antibiotic therapy.
The present study was therefore designed to assess the effi cacy of the slide agglutination technique in diagnosing enteric fever accurately compared to the tube agglutination technique.

MATERIALS AND METHODS
The cross-sectional study study was conducted over a period of one year from May 2011 to April 2012 after obtaining approval from the institutional ethics committee.
All the blood samples requisitioned and received in the microbiology laboratory for Widal test from patients suspected as having enteric fever during the study period were included in the study.The serum was separated from each blood sample following all standard precautions.The sera were then subjected to the Widal test by the slide agglutination method.The test was performed as per the manufacturer's instructions (Span diagnostics Ltd.).Briefl y, 50 μl of antigen was placed upon the slide provided in the kit followed by addition of 50 μl of serum.The slide was rocked gently for mixing.Since observation of agglutination in the form of visible clumps may have observer's bias, the result of the agglutination reaction was scored as 0 (no agglutination), 1+ (25% agglutination), 2+ (50% agglutination), 3+ (75% agglutination) or 4+ (100% agglutination).The sample was labeled as positive if the serum exhibited ≥ 2+ or 50% agglutination. [4]The positive results in the neat sample were titrated for the amount of antibodies by using the semi-quantitative Widal test as per manufacturer's recommendations.The titers of 1:80 and 1:160 were taken as cut off titers to determine positivity towards O and H antigens.
All the samples were then subjected to the tube agglutination test to fi nd out exact titer of antibodies.The serum was diluted in doubling dilutions.0.5 ml of each dilution was then added to a row of Felix tubes containing the same quantity of S. typhi O antigen and two rows of Dreyers' tubes containing the same quantity of S. typhi H and S. paratyphi A antigens.The rack containing all the tubes was then incubated at 37°C in a water bath overnight.Macroscopic agglutination was noted and recorded on the following day after keeping the rack at room temperature.The highest dilution of serum giving visible agglutination was calculated and matched against the currently used local cut off titer as mentioned above, to confi rm positivity.At the end, the results of both slide and tube agglutination test were compared and analyzed using Fisher's exact test.

RESULTS
A total of 1470 samples were received during the study period.The samples were fi rst subjected to the slide agglutination test.Of these, 1176 (80%) samples were negative while 294 (20%) samples were found to be positive.All the samples were then subjected to the tube agglutination test.All slide negative samples were confi rmed as negative by tube agglutination as none of them showed signifi cant titer for both O and H antibodies on tube agglutination test.
Out of 294 (20%) slide positive samples (with signifi cant titers on semiquantitative test), 85 (28.9%) samples were tube agglutination positive, whereas the remaining 209 (71.1%) had titers below signifi cant level on tube agglutination.It can thus be inferred that actual positivity by both slide and tube test was seen in only 5.78% samples while 14.21% samples were incorrectly labeled as positive by slide agglutination tests since they actually had titers less than the signifi cant levels [Table 1].

DISCUSSION
The defi nitive diagnosis of enteric fever depends on the isolation of organisms from blood, bone marrow or other body fl uids.The role of the Widal test has been to increase the index of suspicion for the presence of enteric fever by demonstrating positive agglutination during the acute and convalescent period of infection with evidence of a four-fold rise of antibody titer. [4]The slide Widal test has the advantage of being highly practicable and low cost. [7]The tube agglutination test requires at least overnight incubation and more technical work than the rapid slide test.
According to Hoffman et al., [8] the results of a single Widal test, tube dilution, micro-agglutination or slide agglutination are virtually un-interpretable unless the sensitivity and specifi city as well as the predictive values of the test for the specifi c laboratory and patient population are known.
In our study, the slide agglutination test performed well as a screening test (P < 0.001) since we observed very good overall sensitivity (100%) and NPV (100%).However, the specifi city was relatively low (84.91%), more so when TO antigen alone was considered.For H agglutination, the specifi city was better than O agglutination (92.31% for TH and 99.58% for AH).We did not test for S. paratyphi B antigen since enteric fever due to S. paratyphi B is not a common occurrence in our geographical location.
It has been argued that the positive predictive value (PPV) is the most important measure of a clinical diagnostic method since it represents the proportion of patients with positive test results that are correctly diagnosed. [9]In our study, the PPV was signifi cantly low for all the antigens (TO-28.91%,TH-39.8%,AH-50%).This clearly proves that, although the slide agglutination test is useful in screening out negative samples, the positivity expressed by it is not always helpful in diagnosing the disease correctly.
Keddy et al., [10] has compared four different techniques for the diagnosis of enteric fever, of which the slide agglutination test performed the worst.It had a very poor specifi city (O-3.6%, and H-50%) and low PPV (O-25%, H-53.8%) and NPV (O-68.65,H-77.8%) even though it was performed under optimal conditions in a National Reference Laboratory.This poor performance was further compounded by substantial inter-test variability, which suggests that in a fi eld situation, results would not be comparable between study sites.Hence, Keddy et al., [10] has suggested that the slide agglutination test should not be used as a diagnostic tool.Although the sensitivity and specifi city of the H slide agglutination test appeared to be greater, this was offset by the inconsistent results obtained with the O slide agglutination.This was obvious in the present study as well.
High false positivity exhibited by slide agglutination in this study can be attributed to the endemicity of the disease in the study area and possible presence of cross reacting antibodies of other bacterial and non-bacterial infections.If the slide agglutination test is solely relied upon for diagnosis, a signifi cant number of samples would have been falsely labeled as positive and clinicians would have started antimicrobial chemotherapy in otherwise healthy individuals.This could lead to serious consequences by unnecessarily pressuring the normal gut fl ora to develop antibiotic resistance.Hence, the use of slide agglutination test for diagnosis of enteric fever remains questionable.
It has been rightly stated by Welch et al., [6] "No Widal test is infallible and it is not likely that any will be developed that will lower the validity of the isolation of the etiological agent."A culture is and will remain as the gold standard for the diagnosis of enteric fever.
Unfortunately, in our study we could not use it as a standard for comparison because very few results of blood culture were available out of all the samples.However, it has been proved by many studies that the Widal agglutination test is still of signifi cant diagnostic value, particularly in an area where there is a reasonably high suspicion (prior probability) of enteric fever, provided judicious interpretation of the test is made. [11,12]At the same time, it is also true that if going for the slide agglutination test instead of the standard tube Widal test, it should always be interpreted with reference to the results of the tube agglutination test and clinical data available. [13]

CONCLUSION
The high false positivity and low positive predictive value shown by the slide agglutination test is to caution that the results of slide agglutination should not be solely relied upon for diagnosis and treatment of enteric fever.Performing slide agglutination test is actually a common practice in many resource-constrained laboratories hence if at all it is to be used, the results should always be confi rmed by the tube agglutination test and interpreted with reference to clinical data.
Journal of Medicine and Public Health | Jul-Sep 2014 | Vol 4 | Issue 3

Table 1 : Distribution of positive samples of the slide agglutination test in comparison to tube agglutination test
International Journal of Medicine and Public Health | Jul-Sep 2014 | Vol 4 | Issue 3

Table 2 : Sensitivity, specifi city, predictive values of slide agglutination test and individual antigens used in Widal test in comparison with the tube agglutination test
PPV = Positive predictive value, NPV = Negative predictive value, CI = Confi dence interval