Effect of Anthelmintic Treatment on the Agreement Between Real-Time Polymerase Chain Reaction (RT-PCR) and Kato–Katz Microscopic Technique in the Diagnosis of Soil-Transmitted Helminth Infections

Abstract

Soil-transmitted helminths (STHs), including Ascaris lumbricoides, hookworm species, and Trichuris trichiura, cause significant morbidity worldwide. For an effective proper control of their morbidity, accurate diagnosis method is needed. To this end, a polymerase chain reaction (PCR) method has been developed, but its use remains limited due to the high cost of its implementation, the resources required, and the lack of qualified technicians. The objective of the present analysis is to assess the agreement between the Kato–Katz microscopy method and quantitative real-time PCR (RT-PCR) in the diagnostic of STHs before and after treatment, to decipher the usefulness of either technique for evaluation of the treatment. Methods: Stool samples were collected before and after three- or six-weeks post-treatment from study participants and analyzed using Kato–Katz and RT-PCR methods for the diagnosis of STHs infections. The cure rate (CR) was estimated according to each diagnostic method. Agreement between CRs was tested using the Kappa statistical test. Results: Agreement between Kato–Katz and RT-PCR methods varied regarding the STH species targeted and was different after treatment compared with before treatment. At baseline, the two diagnostic methods showed a moderate agreement (0.45 < K < 0.5) for all STH species, whereas after treatment, concordance decreased slightly (0.11 ≤ K ≤ 0.14) for A. lumbricoides, remained moderate (0.5 ≤ K ≤ 0.53) for T. trichiura, and went from moderate to absent for hookworms. Conclusion: Our findings showed basically a moderate agreement between the Kato–Katz method and RT-PCR. There is a likely association with a moderate proportion of microscopy-positive cases. Reciprocally, a decrease in agreement after treatment was observed with low microscopy-positive cases after treatment, whereby RT-PCR was more likely to detect positive cases than microscopy. Therefore, the agreement is positively associated with an increasing in the number of samples testing positive.

1. Background

Soil-transmitted helminths (STHs) are helminths transmitted to humans via contact with soil. People become infected when exposed to infective form of parasite. The disease is transmitted either through skin penetration of infective larval stages of parasite, or by oral ingestion of fertile eggs. The most common STH species are Trichuris trichiura, Ascaris lumbricoides, and hookworm species including Ancylostoma duodenale and Necator americanus. STHs remain a public health problem in endemic areas, infecting nearly 1.5 billion people mainly in tropical and subtropical areas. STH infections are often chronic, resulting in malnutrition as result of malabsorption of macronutrients, stunted growth, and impaired mental and cognitive development in infected children [1]. Pre-school- and school-aged children are known to experience the morbidity of STH diseases most [2] and are therefore the target population for their control. Indeed, one of the World Health Organization (WHO) strategies for control of STH morbidity is treatment administration, through targeted treatment or large-scale administration of albendazole or mebendazole to at-risk populations [3]. To be efficient and because light infection may remain following drug administration and then constitute a reservoir for transmission, diagnosis methods should be highly sensitive.

The diagnosis of STH infections is performed under microscope using the Kato–Katz (KK) technique recommended by the WHO. The method aims to detect STH eggs in stool samples [4]. However, this technique has considerable shortcomings, such as substantial variation in readings resulting from uneven distribution of eggs in a single stool sample, daily fluctuations in egg excretion, and the skills and experience of the readers [5,6]. More importantly, the KK method may miss low-level infections, leading to underestimation of the true prevalence or, in the case of efficacy trials, could inflate the cure rates (CRs) because of residual low egg counts not detected after treatment [7]. Kato–Katz technique is a simple and affordable method for the diagnosis of STH, reason why question of assessing its usefulness remains, especially in poor or developing countries.

Molecular diagnostic methods have been thought to have high performance in the diagnosis of intestinal helminths, including STHs. Polymerase chain reaction (PCR)-based tests for detection of helminth DNA or ribosomal RNA in fecal samples are the most widely used molecular methods [8,9]. Higher specificity and sensitivity of molecular diagnostics are generally observed in studies comparing KK with molecular methods, mainly RT-PCR [10]. When assessing drug efficacy using such RT-PCR methods, studies reported a lower efficacy in the treatment of STH infections as compared to the KK method [11]. However, assessments of drug efficacy using molecular approaches remain rare, although monitoring of drug efficacy of the utmost importance in making treatment recommendations for new therapies and in light of possible future anthelmintic resistance [12]. This recalls the question of usefulness of the microscopy technique for the diagnosis of STH infections after drug administration contexts such as individual patient treatment or effectiveness of mass drug administrations. To this end, the present analysis aimed to assess the agreement between the RT-PCR method and the KK technique in diagnosing STH infection before and after treatment.

2. Methods

2.1. Study Site

The study took place at the Centre de Recherches Médicales de Lambaréné (CERMEL) [13] and was conducted in Lambaréné and surrounding areas. Lambaréné is an urban/semi-urban area of Gabon located 60 km south of the equator and is surrounded by a set of villages located along the National 1 road Southwest (rural 1) and Northwest (rural 2) of Lambaréné (Figure 1). The study area is known to be endemic for STH infections, as well as other helminths such as those causing schistosomiasis and filariasis [14].

2.2. Study Population and Study Design

The study was designed as a longitudinal study in which children aged from 2 to 17 years old were included. Stool samples were collected to determine STH infection status at baseline, then at three (C1 visit) and six weeks (C2 visit) post-treatment. The diagnosis of STH infection was made using the both KK technique and RT-PCR. Three treatments were considered: Albendazole 400 mg (Lincoln Pharmaceutical Ltd., B/h Satyam Complex, Science City Road, Sola, Ahmedabad-380060, Gujarat, India) administered over three consecutive days, or Albendazole 400 mg on the first and the third day alternated on the second day with either Mebendazole (500 mg) (Cadila Pharmaceutical Ltd., Dholka, Gujarat, Inde) or Pyrantel (125 mg) per kg of body weight (Innothera Chouzy 1 r Rene Chantereau, 41150 Chouzy sur Cisse, France). Participants were randomized to one of three treatment groups in 1:1:1 ratio.

2.3. Stool Sample Collection

A screw-top stool container was given to the participants who were asked to provide fresh stool sample in the morning of the next day. Stool samples were collected by the study team every morning at each participant’s home between 6 am and 11 am and transported in an icebox immediately to the CERMEL parasitology laboratory where the KK test was immediately performed. Aliquots of each stool sample were preserved in 2 mL microtubes and were stored in a −20 °C freezer for later molecular analysis.

2.4. Kato–Katz Test

A total of 41.7 mg of stool were used from each stool sample to perform the KK analysis [15,16]. Basically, as depicted in Supplementary Figure S2, the smears were performed in duplicate for egg load quantification of A. lumbricoides, T. trichiura, and hookworm infections [16]. Slides were examined by light microscopy within 60 min following preparation, and arithmetic means were calculated to determine egg loads of A. lumbricoides, T. trichiura, and hookworms. In case of qualitative discrepancy results or a quantitative difference of more than 20% of two positive results, a third reading was requested from a third independent reader. In this case, the two closer results were considered for the calculation of egg count. The results are given in eggs per gram (EPG) of stool.

2.5. Nucleic Acid Isolation

Total nucleic acids were extracted from 250 mg of frozen stool sample spiked with an internal extraction and PCR control (Phocid herpesvirus (PhHV-1) using the PowerSoil kit from Qiagen (Strasse 1, 40724, reference 12888-100, Hilden Germany). All steps of the DNA isolation were performed following the manufacturer’s instructions and the nucleic acids were eluted in a 100 μL volume and stored in aliquots at −20 °C.

2.6. Parasite DNA Detection by RT-PCR Triplex

The presence of A. lumbricoides, T. trichuria, and hookworm (N. americanus) DNA was screened using two different multiplex real-time PCR detection panels as described previously [6] Panel I targets A. lumbricoides, T. trichuria, and PhHV-1; and Panel II targets N. americanus, Strongyloides stercoralis, and PhHV-1. In short, DNA amplification was performed in a LightCycler^® 480 Roche Instrument II thermocycler (F. Hoffmann–La Roche, Serial number 30314, Basel, Switzerland) in 20 μL reactions using the QuantiTech Multiplex PCR NoROX enzyme (Qiagen, Hilden, Germany), and 0.1 mg/mL bovine serum albumin (BSA). Freeze-dried tubes of primers and molecular probes for the target regions of each species were obtained through the manufacturer (Eurofin Genomics, www.eurofingenomics.eu, accessed on 8 October 2024). All stock solutions were prepared at 100 µM, while working solutions were prepared at 10 µM, and 5 µL of extracted parasite DNA was used for the detection. All sequences for each primer and probe from each species targeted are indicated in Supplementary Table S1. All samples were tested in duplicate for the presence of helminths employing the following cycling protocol: one cycle at 95 °C for 15 min (polymerase activation), followed by 50 cycles of 95 °C for 15 s (denaturation) and 60 °C for 30 s (annealing). Samples with a cycle threshold (Ct) value of ≤40 were considered as positive.

2.7. Statistical Considerations

Statistical analysis was performed using R software (R177 Core Team, R Foundation for Statistical Computing, Vienna, Austria) version 4.1.0 and IBM SPSS statistic 26 (IBM Corp. Released 2019. IBM SPSS Statistics for Windows, Armonk, NY, USA: IBM Corp) version 26.0. Positive microscopy result of stool examination was based on the presence of at least one egg and the load of STH species infection was subsequently determined. The intensity of infection was defined following the WHO threshold [17]. RT-PCR results were based on a characteristic amplification curve, and a Ct-value of ≤40 was considered as positive. As described by Maylasari et al. [18], we arbitrarily extrapolated the Ct-values to estimate the quantity of targeted DNA: low quantity was defined as Ct ˃ 35, Moderate quantity as a Ct-values between 30 and 35, and high quantity as Ct-value < 30. In statistical analysis, qualitative data were summarized as proportions with a 95% confidence interval (95% CI). The binom exact function of the Binom R package was used to calculate the 95% CI of the CR. The difference in CR was determined under the assumption that no overlapping confidence intervals indicate statistical significance. The non-parametric Spearman’s rank correlation (using a scatterplot) was applied to measure the relationship between PCR Ct-values, nucleic acid load obtained after extraction, and egg load determined via microscopy. The kappa statistic test was used to assess agreement between microscopy and real-time PCR and was interpreted as suggested by Cohen [19]: poor for K < 0.20; fair for K from 0.21 to 0.40; moderate for K between 0.41 and 0.60; good for K from 0.61 to 0.80; and very good from 0.81 to 1.00. The significance of statistical tests was set at 5% and extrapolated where the confidence intervals did not overlap.

2.8. Ethical Consideration

The study protocol was approved by the institutional ethics committee of the Centre de Recherches Médicales de Lambaréné (CERMEL) under the registered number CEI-007/2019, and the National Ethic Committee of Gabon (PROT N°0084/2019/PR/SG/CNER). Parents or legal representative of each participant provided written informed consent before any study procedure. The study was conducted in line with the Good Clinical Practice principles of the International Conference on Harmonization [20] and the Declaration of Helsinki [21].

3. Results

3.1. Study Population Characteristics

A total of 280 participants out of 654 volunteers tested were diagnosed as positive for the presence of any STH eggs in stool and were included in the analysis (Supplementary Figure S2). Their mean (SD) age was 7.7 (3.5) years old with a 1.0 female-to-male sex ratio. Most participants (97%) came from rural areas, while only eight (3%) participants came from Lambaréné, a semi-urban area of Gabon. No difference in study population characteristics was observed between the three-study time points, i.e., baseline, three- and six-weeks post-treatment (

Table 1. Study population characteristics of the 280 participants and distribution of soil-transmitted helminth infections.

Variables	Baseline		3 Weeks Post-Treatment		6 Weeks Post-Treatment
	N	%	N	%	N	%
Total (N)	280	-	211	-	165	-
Sex
Female	140	50.0	106	50.2	81	40.0
Male	140	50.0	105	49.7	84	50.9
Female-to-male sex ratio	1.0	-	1.0	-	0.9	-
Mean age (SD)	7.7 (3.5)	-	7.5 (3.4)	-	7.5 (3.2)	-
Age group (in years)
2–5	95	33.9	72	34.1	54	36.2
6–10	112	40.0	88	41.7	74	49.6
11–17	73	26.0	51	24.1	37	24.6
Locality
Semi-urban	8	2.8	5	2.3	6	4.0
Rural area 1	122	43.5	97	45.9	79	47.8
Rural area 2	150	53.3	109	51.6	80	48.4

SD: standard deviation; %: Percentage.

).

3.2. Distribution of the Proportion of STH Infections Using Microscopy and RT-PCR Triplex

Table 2. Distribution of main soil-transmitted helminth infections by study time point and by diagnostic method.

	Baseline		Three Weeks Post-Treatment		Six Weeks Post-Treatment
	n	%	n	%	n	%
Total (N)	280		211		165
STH species distribution and infection intensity using microscopy
Ascaris lumbricoïdes	144	51.4	6	2.8	4	2.4
Eggs per gram (mean)	(15,025.5)	-	(6.9)	-	(15.4)
Light	59	41.0	6	100	4	100
Moderate	60	41.7	-	-	-	-
Heavy	25	17.4	-	-	-	-
T. trichiura	213	76.1	65	30.8	62	37.6
Eggs per gram	(518.5)	-	(45.7)	-	(152.8)	-
Light	176	82.6	64	98.5	57	91.9
Moderate	36	16.9	1	1.5	5	8.1
Heavy	1	0.5	-	-	-	-
Hookworms	57	26.8	2	0.9	0	0.0
Eggs per gram (mean)	(63.8)	-	(0.30)	-	(0.0)	-
Light	55	96.5	4	100	-	-
Moderate	2	3.5	-	-	-	-
Heavy	-	-	-	-	-	-
STH species distribution and infection intensity using RT-PCR
A. lumbricoïdes	197	70.4	61	28.9	49	30.0
Ct value (mean)	(29.3)	-	(34.6)	-	(35.1)
Low	29	14.7	28	45.9	24	49.0
Moderate	61	31.0	27	44.3	22	44.9
High	107	54.3	6	9.8	3	6.1
T. trichiura	222	79.3	97	46.0	81	49.1
Ct value (mean)	(32.0)		(35.6)	-	(36.5)	-
Low	57	25.7	54	55.7	57	70.4
Moderate	85	38.3	36	37.1	20	24.7
High	80	36.0	7	7.2	4	4.9
Hookworms	58	20.7	5	2.4	8	4.8
Ct value (mean)	(35.2)	-	(34.6)	-	(38.9)
Low	22	37.9	1	20.0	8	100.0
Moderate	23	39.6	4	80.0	-	-
High	13	22.4	-	-	-	-

Ct: cycle threshold, RT-PCR: Real-time polymerase chain reaction, STH: Soil-transmitted helminth, %: Percentage.

shows the distribution of STH species at the three time points of the study, as well as the intensity of the infection using microscopy and molecular (RT-PCR) methods, respectively. We observed an increase in the proportions of positive cases with RT-PCR at each time point compared with microscopy results. Microscopy revealed that the most prevalent STH species was T. trichiura (76%), followed by A. lumbricoides (51%) and hookworm (27%). After treatment, and compared to baseline, the trend in the distribution of STH species from RT-PCR results remained the same, although a significant reduction in the proportion of positive cases was observed (with both methods).

3.3. Cure Rate (CR) Assessment Using Microscopy and RT-PCR)

As presented in

Table 3. Cure rate (CR) distribution for ascariasis, trichuriasis, and ancylostomiasis at three and six weeks post-treatment based on the Kato–Katz and RT-PCR methods among the study population.

	Three Weeks Post-Treatment				Six Weeks Post-Treatment
	N	n	CR in %	95% CI (CR)	N	n	CR in %	95% CI (CR)
Microscopy method
A. lumbricoides	102	99	97.0	91.7–98.9	87	84	96.5	90.3–98.8
T. trichiura	165	105	63.6	56.0–70.5	130	70	53.8	45.2–62.1
Hookworm	42	39	92.8	80.9–97.5	29	29	100.0	88.3–100.0
RT-PCR method
A. lumbricoides	146	111	76.0	68.4–82.2	118	87	73.8	65.1–80.8
T. trichiura	170	99	58.2	50.7–65.2	135	56	41.4	33.5–49.9
Hookworm	42	39	92.8	80.9–97.5	31	29	93.5	79.2–98.2

N: number of participants infected with the respective STH species at baseline and seen at the control visit; n: Number of children cured; CR: Cure rate; CI: confidence interval; %: Percentage.

the CR using RT-PCR was statistically significantly lower for ascariasis at three weeks by RT-PCR as compared to microscopy (76%; 95% CI: 68–82% vs. 97%; 95% CI: 92–99%), while the difference was not statistically significant for trichuriasis and hookworm. Comparing the two time points, a decrease in the CR was observed only for trichuriasis at six weeks as compared to three weeks post-treatment but was statistically significant only by RT-PCR (58%; 95% CI: 51–65% vs. 41%; 95% CI: 33–50%) and not by microscopy (64%; 95% CI: 56–70% vs. 54%; 95% CI: 45–62%).

3.4. Association between Egg Load by Microscopy (Log EPG) and RT-PCR Ct-Values

Figure 2 shows the association between egg load and the Ct values obtained with RT-PCR. The correlation degree was significantly (p ≤ 0.001) moderate, at baseline and six weeks post-treatment (Spearman rho = 0.49), while it was weak (Spearman rho = 0.39) between egg load and Ct values from RT-PCR for T trichiura. Regarding A. lumbricoides, we observed a significant moderate and negative degree of correlation only at baseline (Spearman rho = 0.51, p ≤ 0.001). No significant correlation was observed from hookworms at baseline or at the two visits post-treatment.

3.5. Agreement Between Microscopy and RT-PCR Test Results

Analysis of agreement between both diagnostic methods is presented in

Table 4. Two by two table showing the kappa test statistic for the agreement between Kato–Katz and RT-PCR triplex techniques at the three main study time points.

	RT-PCR Triplex
	Baseline (N = 280)				Three Weeks Post-Treatment (N = 211)				Six Weeks Post-Treatment (N = 149)
Microscopy (Kato–Katz)	Positive n, (%)	Negative n, (%)	Total n, (%)	K Statistic, 95% CI (K), p-Value	Positive n, (%)	Negative n, (%)	Total n, (%)	K Statistic, 95% CI (K), p-Value	Positive n, (%)	Negative n, (%)	Total n, (%)	K Statistic, 95% CI (K), p-Value
A. lumbricoides
Positive	133, (47.5)	11, (3.9)	144, (51.4)	0.46,	6, (2.8)	0, (0.0)	6, (2.8)	0.14,	4, (2.4)	0, (0.0)	4, (2.4)	0.11
Negative	64, (22.9)	72, (25.7)	136, (48.5)	[0.37–0.56]	54, (26.0)	151, (72.9)	205, (97.1)	[0.07–0.20]	45, (27.2)	116, (70.3)	161, (97.5)	[0.06–0.18]
Total	197, (70.3)	83, (29.6)	280	p < 0.001	60, (21.4)	151, (70.0)	211	p < 0,0001	49, (29.6)	116, (70.3)	165	p = 0.002
T. trichiura
Positive	193, (68.9)	20, (7.1)	213, (76.0)	0.50,	57, (27.0)	8, (3.7)	65, (30.8)	0.52,	54, (32.7)	8, (4.8)	62, (37.5)	0.57,
Negative	29, (10.3)	38, (13.5)	67, (23.7)	[0.39–0.58]	41, (19.4)	105 (49.7)	146, (69.1)	[0.42–0.61]	27, (16.3)	76, (46.0.)	103, (62.4)	[0.47–0.66]
Total	222, (79.2)	57, (20.7)	280	p < 0.001	98, (46.4)	113, (53.5)	211	p < 0.001	81, (49.0)	84, (50.9)	165	p < 0.001
Hookworm
Positive	34, (12.1)	28, (8.5)	57, (20.3)	0.45,	2, (0.9)	0, (0.0)	2, (0.9)	0.49	0, (0)	0, (0)	0, (0)	0.0
Negative	24, (10.0)	194, (69.2)	218, (77.8)	[0.34–0.53]	4, (1.9)	205, (97.2)	209, (99.1)	[0.39–0.58]	8, (2.0%)	157, (94.6)	165, (100.0)	[0.00–0.02]
Total	58, (20.7)	222, (79.2)	280	p < 0.001	6, (2.8)	205, (97.2)	211	p < 0001	8, (2.0%)	157, (94.6)	165	p = 1

%: Percentage.

. Basically, the kappa test was statistically significant (p < 0.001) at all-time points and for all STH species, except for hookworm infection at six weeks post-treatment (C2 visit) (p = 1). At baseline, the kappa test indicated a moderate agreement between both tests for all STH species: A. lumbricoides (K = 0.46; 95% CI: 0.37–0.56), T. trichiura (K = 0.50; 95% CI: 0.39–0.58), and hookworm (K = 0.45; 95% CI: 0.34–0.53) infections. After treatment, the kappa test decreased to slight agreement only for A. lumbricoides at three weeks post-treatment (C1 visit) (K = 0.14; 95% CI: 0.07–0.20) and C2 (K = 0.11; 95% CI: 0.06–0.18) visits, but remained moderate for T. trichiura at C1 (K = 0.57, 95% CI: 0.42–0.61) and C2 (K = 0.57; 95% CI: 0.47–0.66) and for hookworm infection at C1 (K = 0.49; 95% CI: 0.39–0.58).

4. Discussion

The main objective of the present analysis was to assess the agreement between microscopy (KK method) and RT-PCR in the diagnosis of STH infections before and after treatment. The results showed a moderate agreement between both methods at baseline for all STH species targeted. After anthelmintic treatment, we obtained slight agreement between both methods for A. lumbricoides, while the agreement remained moderate for T. trichiura and hookworm.

The moderate agreement we found between PCR and microscopy at baseline indicates the difference between the two methods in the diagnosis of STH infections: RT-PCR was more likely to be positive where KK was negative for all STH species; T. trichiura (10%), hookworm (10%), and particularly A. lumbricoides (23%). This reflects the higher sensitivity of the PCR method in the diagnosis of STHs as compared to KK. This situation was expected and is explained mainly by the distribution of STH eggs in stool, particularly in cases of low infection intensity. The KK method is used to detect STH eggs in stool samples, while the RT-PCR method is used to detect parasite DNA obtained from eggs in stool sample. The capture of eggs in stool sample in the process of both methods can be influenced by some factors, particularly the number of STH eggs excreted in stool [22], which is very often unevenly distributed in the sample. In case of light infections, assuming a low number of eggs as was mainly the case for the three STH species, at baseline in our cohort; eggs could escape the technician during the processing of the stool for microscopy examination [22,23]. While for the PCR technique, a small number of eggs can easily provide a sufficient quantity of nuclei acids detectable by the RT-PCR technique [22,24]. Similarly, the proportions of participants found positive by microscopy but negative by PCR for A. lumbricoides (4%), T. trichiura (7%), and hookworm (8%) may also be explained by the same fact. Indeed, in that case, STH eggs can be present in the portion of stool sample used for microscopy and be absent, altered, or not sufficient enough in the portion of stool used for nucleic acid extraction to provide a sufficient quantity of nucleic acid to be detectable using RT-PCR. Although the proportion of discordance was higher for A. lumbricoides (27%) than for hookworm (18%) and T. trichiura (17%), RT- PCR was in general more likely to be positive where microscopy reported a negative result. The similar proportion of STH infection reported by both methods for T. trichiura, hookworm, and, less importantly, A. lumbricoides suggests that microscopy remains a relevant tool to assess STH prevalence when PCR cannot be implemented.

The treatment of STH is aimed to cure the infection or, in the worst case, reduce the intensity of infection. In case of mass drug administration, one can thus expect a reduction of infection prevalence and morbidity in the population. Our results show a significant decrease in egg burden following treatment, particularly for A. lumbricoides and hookworms (Table 2), with all infection intensity becoming light. This reduction also leads to an increase in discrepancies between the two techniques, as we observed three and six weeks after treatment. Indeed, during the post-treatment visits, we observed a slight agreement between the two methods for A. lumbricoides, while due to the very low number of positive cases, we were not able to assess the concordance for hookworm infection particularly at six weeks post-treatment. The discrepancy between the two methods after treatment is mainly characterized by the number of positive tests by PCR but negative by microscopy. This result reinforces the idea that RT-PCR, or molecular diagnostics in general, should be applied as a diagnostic tool for monitoring programs or studies where the control and elimination of STH in general is targeted, but where particularly hookworm or ascariasis infections, are being evaluated [11,25,26].

T. trichiura was one of the three STHs of interest in our work. In contrast to A. lumbricoides and hookworm infection, a different dynamic of egg count was observed for T. trichiura after anthelmintic treatment. Indeed, after an increase in the CR observed three weeks after treatment as a consequence of treatment as we presented elsewhere [27], the CR decreased from three weeks to six weeks even when assessed by both methods. Consequently, the agreement between both methods after treatment remained moderate at the two time points as the proportion of trichuriasis cases remained moderate at the two time points post-treatment, namely, 31% and 38%, respectively. This result supports our hypothesis that the discrepancy between the two diagnostic methods depends on the prevalence of the infection. In addition, because no difference was observed in the CR between the two methods and between the two time points, our results also suggest that the microscopy technique can remain of high interest for the assessment of drug efficacy in the treatment of those two STH species.

The original study was not designed for the present analysis. The sample size calculation did not aim to address the objective of the present analysis. We therefore recognize that this may affect our outcome and therefore influence our conclusions. However, as we included more than the recommended number of participants needed for drug efficacy assessment as recommended by the WHO particularly at baseline and at week three for A. lumbricoides and T. trichiura [3,28], we are confident in the reliability of our analysis. In addition, three different regimens of anthelmintic drugs were used for the treatment of STH infection in the study population and were not considered in our analysis. Although we were mainly interested in the infectious status of our study participants, we recognize the use of anthelmintic drugs could somehow affect our results and should be further investigated.

5. Conclusions

In the present study, we have evaluated the agreement between the Kato–Katz technique and RT-PCR in the diagnosis of STH infection before and after treatment. After anthelmintic treatment, we obtained slight agreement between the two methods for A. lumbricoides, while the agreement was moderate for T. trichiura and hookworm. Our findings indicate that the necessity to use RT-PCR will depend on the STH species, and on the outcome of interest. As such, Kato–Katz alone could remain a relevant tool to assess prevalence of soil-transmitted helminths in areas of moderate or high endemicity level, whereas in a context where appropriate assessment of STH infectious status is needed, such as intervention in the disease transmission, elimination, or in the context of clinical trials, RT-PCR should be applied, particularly in post-intervention.