Accuracy of the Ultra-Rapid Urease Test for diagnosis of Helicobacter pylori infection

Abstract

Background

Rapid Urease Test (RUT) is a simple, cheap and relatively fast method for diagnosing Helicobacter pylori infection. It is therefore the preferred method used for patients undergoing gastroscopy. Most kits require 24 h to give results. The new Ultra-Rapid Urease Test (URUT) kit by Biohit® requires less than 1 h.

Objective

To determine URUT's diagnostic accuracy.

Method

Prospective, blind, multi-centre study involving dyspeptic patients. One corpus biopsy and three antral biopsies were obtained during gastroscopy for standard histological analysis, RUT and URUT. The URUT result was checked after 1 min, 5 min, 30 min and 60 min and the RUT was checked over the course of 24 h. Histology was used as the gold standard test.

Results

144 patients were included, 68% female, with a mean age of 49 years old; 50% were H. pylori positive. RUT and URUT diagnoses were correct in 85.9% and 90% of the cases, respectively. The mean waiting time for a positive RUT result was 6 h. The sensitivity, specificity, and positive and negative predictive values for RUT were, respectively, 82%, 90%, 89% and 84%. The URUT's results were similar (85%, 94%, 94% and 87%). These figures improved when patients taking PPIs were excluded (RUT: 86%, 91%, 93% and 83%; URUT: 91%, 94%, 96% and 89%). No statistically significant differences were found when comparing RUT and URUT distributions of correct diagnoses (McNemar's Test, p = 0.3) but there was a tendency towards better results with the URUT.

Conclusion

The URUT is equivalent to (or slightly better than) the traditional RUT in diagnosing H. pylori infection, and provides results in less than an hour.

Resumen

Introducción

El test de la ureasa (TRU) es un método simple, barato y relativamente rápido para el diagnóstico de la infección por Helicobacter pylori (H. pylori). Por tanto, es el método de elección en pacientes sometidos a gastroscopia. La mayoría de los kits requieren 24 h para obtener un resultado. En nuevo test ultrarrápido de la ureasa (TURU) de Biohit requiere menos de una hora.

Objetivo

Determinar la exactitud diagnóstica del TURU.

Método

Estudio multicéntrico, prospectivo y ciego, en el que se incluyó a pacientes dispépticos. Se obtuvieron 3 biopsias de antro y una de corpus durante la gastroscopia para análisis histológico estándar, TRU y TURU. El resultado del TURU se comprobó a los 1, 5, 30 y 60 min, mientras que el TRU se evaluó a lo largo de 24 h. La histología se utilizó como patrón oro.

Resultados

Se incluyó a 144 pacientes, 68% mujeres, edad media 49 años, el 50% fueron positivos para H. pylori. TRU y TURU diagnosticaron correctamente el 85,9% y 90,0% de los casos, respectivamente. La duración media de espera para un resultado positivo del TRU fue 6 h. La sensibilidad, la especificidad y los valores predictivos negativo y positivo para el TRU fueron, respectivamente, del 82, el 90, el 89 y el 84%. Los resultados del TURU fueron equivalentes (el 85, el 94, el 94 y el 87%). Estos resultados mejoraron al excluir los pacientes que tomaban IBP (TRU: 86, 91, 93 y 83%; TURU: 91, 94, 96 y 89%). La comparación de distribución de diagnósticos correctos entre TRU y TURU no encontró diferencias estadísticamente significativas (test de McNemar p = 0,3) pero existe una tendencia a mejores resultados con el TURU.

Conclusión

El TURU es equivalente (o algo superior) al TRU tradicional en el diagnóstico de la infección por H. pylori y obtiene los resultados en menos de una hora.

Introduction

The management of Helicobacter pylori (H. pylori) infection requires accurate diagnostic methods. Several diagnostic methods, both direct (histology, culture and urease tests) and indirect (faecal antigen, serology and urea breath test), have been developed.^1–4 Direct or “invasive” methods require the biopsy sampling while indirect or “non-invasive” methods detect secondary characteristics of the bacteria.⁵

Patients undergoing gastroscopy for dyspeptic symptoms are generally tested for H. pylori infection. Rapid Urease Test (RUT) is usually the method of choice as it is simple, cheap, gives an accurate result in 24 h and the result can be observed in the endoscopic units without the involvement of other services.^3–6 Urease kits have a media containing urea and a pH-dependant colour indicator, in which the biopsy is placed. H. pylori urease activity alters the pH causing a change of colour of the media.⁷

Common RUTs require up to 24 h to obtain an accurate result, what limits its benefits and utility in clinical practice as the treatment prescription cannot be given in the moment, forcing the patient to come back for the diagnosis to the hospital's outpatient clinic.^1,8 A quicker urease kit would increase efficiency by reducing costs and incommodities to patients.

Recently, new kits have been developed trying to improve the utility and speed of the urease method.^5–7,9 Biohit^® commercializes a new urease method that, according to its technical information, is able to give accurate results in 30 min. Previous studies have evaluated the accuracy of different commercial kits obtaining encouraging results.^10–17 It seems that Biohit^®’s Ultra-Rapid Urease Test (URUT) is a promising diagnostic method but wider and more diverse evidence is still needed to recommend the systematic use URUT in clinical practice. Therefore, the aim of the present study was to evaluate and compare URUTs diagnostic accuracy for the diagnosis of H. pylori in the Spanish population.

Methods

Patients

In this prospective, blind, multicenter study, a total of 144 patients (68% female, mean age 49 years) who attended digestive services for upper gastrointestinal endoscopy were consecutively enrolled. Inclusion criteria were: patients over 18 years of age suffering from dyspepsia. Exclusion criteria were: presence of hepatic, renal, lung, endocrine, metabolic, haematological or malignant diseases; previous H. pylori eradication treatment; history of alcohol or drug abuse; and pregnancy or nursing.

Proton pump inhibitor (PPI) treatment was not considered an exclusion criterion as, in clinical practice, most patients undergoing upper gastrointestinal endoscopy are taking these drugs prior to the procedure. However, sub-analyses depending on PPI intake were performed for all calculations.

The investigator in charge of each diagnostic test was blinded to the results of the other tests.

Biopsies

Three antrum and one corpus biopsies were obtained. One antrum and the corpus biopsies were used for the standard histological analysis. Biopsies were fixed in 10% formalin and separately embedded in paraffin blocks. The sections, serially cut and stained with haematoxylin & eosin, were examined with light microscopy for the histological assessment of H. pylori infection by a pathologist. One antrum biopsy was used for the diagnosis of H. pylori infection with the traditional RUT. The other biopsy was used for diagnosis with URUT.

Urease kits

RUT was performed using one out of the three most common kits in Spanish hospitals (CLO test^®, Jatrox-test^® and Gut plus^®). Each hospital used the kit of their own routine clinical practice. The kit was checked out during 24 h. Biohit^®’s URUT (Biohit^®, Helsinki, Finland) was checked 1 min, 5 min, 30 min and 60 min after the biopsy was included in the kit's media. Flowchart of the patients is present in Fig. 1 (STARD flowchart).

Figure 1.

Patient STARD flowchart.

Ethical issues

This study was performed with the approval and follow up by the hospitals’ Ethics Committees. The design and development followed the WMA Helsinki Declaration of 1964 and its revisions and all applicable regulations. All patients signed an informed consent.

Statistical analysis

Mean and standard deviation were calculated for quantitative variables, and percentage and 95% confidence interval (95%CI) for qualitative variables. Sensitivity, specificity, positive and negative predictive values and positive and negative likelihood ratios were calculated for all urease kits. Z-test was used to compare diagnostic success rates. McNemar test was used to compare the distribution of results for both tests. Significance was considered for p < 0.05. In order to demonstrate equivalence of RUT and URUT, a sample size of 140 patients was calculated [significance level (alpha) 5%, power (beta) 90% and margin of equivalence (d) 5%].

Results

• Study population characteristics

One hundred and forty-four dyspeptic patients were included in 11 hospitals all over Spain in 2009. Average age was 49, and 69% were women. Forty-two percent of patients were taking PPIs. Fifty percent were H. pylori positive by histology. Mean waiting time for positive RUT result was 6.2 h.

• Overall diagnostic success of RUT and URUT

RUT and URUT diagnosis were correct in 85.9% (95%CI = 80–92%) (Table 1) and 90.0% (95%CI = 85–95%) (Table 2) of the cases respectively.

Table 1.

CrossTable of Rapid Urease Tests (by commercial brand) vs. histology in the diagnosis of H. pylori infection.

		Histology		Total
		Negative	Positive
All kits	Negative	62	12	74
	Positive	7	54	61
	Total	69	66	135
CLO-test^®	Negative	29	3	32
	Positive	3	22	25
	Total	32	25	57
Jatrox-test^®	Negative	21	4	25
	Positive	1	22	23
	Total	22	26	48
Gut-plus^®	Negative	11	5	16
	Positive	4	10	14
	Total	15	15	30

Table 2.

CrossTable of Ultra Rapid Urease Test (URUT) (by time of assessment) vs. histology in the diagnosis of Helicobacter pylori infection.

		All Kits			Excluding Gut-Plus^®
		Histology		Total	Histology		Total
		Negative	Positive		Negative	Positive
URUT 1 min	Negative	68	41	109	54	28	82
	Positive	2	24	26	1	21	22
	Total	70	65	135	55	49	104
URUT 5 min	Negative	67	27	94	54	17	71
	Positive	3	41	44	1	35	36
	Total	70	68	138	55	52	107
URUT 30 min	Negative	67	15	82	54	10	64
	Positive	4	53	57	2	43	45
	Total	71	68	139	56	53	109
URUT 1 h	Negative	67	10	77	54	6	60
	Positive	4	58	62	2	47	49
	Total	171	68	139	56	53	109

• RUT by commercial brand

CLO-test^®, Jatrox-test^® and Gut-plus^® were used for the RUT in 37%, 41% and 22% the patients respectively. Comparing their results with the gold standard, CLO-test^® and Jatrox-test^® each diagnosed correctly 91% of patients but Gut-plus^® only 70% (Z-test = 2.55, p < 0.01). Therefore sub-analyses excluding those patients diagnosed with Gut-plus^® were performed (Table 1). Mean time for positive result was 6.2 ± 9.3 h (CLO-test^® 9.2 h; Jatrox-test^® 3.0 h; Gut-plus^® 7.3 h).

• URUT by time checked

One minute after biopsies were placed inside the kit's media, URUT correctly diagnosed H. pylori infection in 68% of patients, 78% at 5 min, 86% at 30 min and 90% at 60 min (Table 2).

• Accuracy depending on PPI intake

All the previous analyses were performed twice, considering all data and excluding those patients under PPI treatment. RUT and URUT (after 60 min) diagnosis was correct in 93% and 97% of patients, including only those patients where the traditional RUT was not Gut-Plus^® (Table 3).

Table 3.

CrossTable of Rapid and Ultra Rapid Urease Tests (RUT and URUT) vs. histology in the diagnosis of Helicobacter pylori infection in patients not taking proton pump inhibitor treatment.

		All kits			Excluding Gut-plus^®
		Histology		Total	Histology		Total
		Negative	Positive		Negative	Positive
RUT	Negative	30	6	36	27	3	30
	Positive	3	38	41	1	30	31
	Total	33	44	77	28	33	61
URUT	Negative	32	4	36	28	1	29
	Positive	2	42	45	1	34	35
	Total	34	47	81	29	35	64

• URUT and RUT's accuracy.

Accuracy calculations were performed for RUT and URUT (at checkpoints 30 and 60 min). PPI treatment and RUT brand sub-analyses were also studied and data is shown in Table 4. URUT (at 30 and 60 min) and RUT's diagnostic successes were not significantly different (Z-test p > 0.05). McNemar tests could not demonstrate statistically significant differences when comparing the distribution of success between RUT and URUT.

Table 4.

Rapid and Ultra Rapid Urease Test's (RUT and URUT) accuracy for the diagnosis of Helicobacter pylori infection in all subanalyses.

		Sensitivity	Specificity	PPV	NPV	PLR	NLR
RUT	All kits	82% (75–88%)	90% (85–95%)	89% (83–94%)	84% (78–90%)	8.6 (4.0–16.4)	0.20 (0.12–0.34)
	No Gut-plus^®	86% (80–93%)	94% (90–99%)	94% (89–98%)	88% (82–94%)	16 (5–47)	0.15 (0.07–0.29)
URUT	All kits	85% (79–91%)	94% (91–98%)	94% (89–98%)	87% (81–93%)	15 (6–39)	0.16 (0.09–0.28)
	No Gut-plus^®	89% (83–95%)	96% (93–100%)	96% (92–100%)	90% (84–96%)	25 (6–97)	0.12 (0.06–0.25)
RUT No PPI	All kits	86% (79–94%)	91% (84–97%)	93% (87–99%)	83% (75–92%)	10 (3–28)	0.15 (0.07–0.32)
	No Gut-plus^®	91% (84–98%)	96% (92–100%)	97% (92–100%)	90% (82–98%)	25 (4–175)	0.09 (0.03–0.28)
URUT 1 h No PPI	All kits	91% (85–97%)	94% (89–99%)	96% (91–100%)	89% (82–96%)	16 (4–60)	0.09 (0.04–0.24)
	No Gut-plus^®	97% (93–100%)	97% (92–100%)	97% (93–100%)	97% (92–100%)	28 (4–193)	0.03 (0–0.2)

PPV: positive predictive value; NPV: negative predictive value; PLR: positive likelihood ratio; NLR: negative likelihood ratio.

95% confidence interval is shown in parentheses.

Discussion

The present study evaluated the accuracy of RUT and URUT in the diagnosis of H. pylori infection and showed equivalent results for RUT kits (24 h result) and Biohit^®’s URUT at minutes 30 and 60.

Three RUT kits were studied; Jatrox-test^® and CLO-test^® offered equivalent results but Gut-plus^® had significantly lower accuracy in the studied population. Although this lower accuracy may be due to a low sample size beta error, all the subsequent analyses were calculated both including and excluding patients diagnosed with Gut-plus^® kit. Jatrox-test^® was able to offer positive results quicker than the other two kits, although the 24 h waiting period was still needed in order to avoid false negative results.

The accuracy of Biohit^®’s URUT diagnosis at 30 min was statistically equivalent to the traditionally used RUT kits. Even though the kit's manual clearly establishes the waiting period in 30 min, increasing it to one hour improved sensitivity (better than RUT) in all sub-analyses and did not reduce specificity, thus it may be recommendable to wait 60 min before checking the URUT result.

In accordance to previously published data,^1,7,8 the intake of PPI drugs the two weeks prior to endoscopy reduced the accuracy of all urease kits in our study. Therefore, the withdrawal of PPIs before endoscopy should be recommended to patients. The results of RUT (Jatrox-test^® and CLO-test^®) and URUT in our study in patients not taking PPI drugs were excellent, especially in the case of URUT's diagnosis after 60 min.

Finally, URUT and RUT kits’ incorrect diagnosis mostly occurred in the same patients and no statistically significant differences were found when comparing the distribution of correct and incorrect diagnosis of both methods. This implies that traditional RUTs limitations regarding diagnostic results,^1,4–8 probably due to low bacterial urease activity, are not overcome by URUT.

The main limitation of this study derives from mimicking clinical practice in Spain where there is no systematic implementations of recommendations regarding biopsy collection during endoscopy which did not allow deeper analysis of our patients.

In summary, previous studies comparing diverse ultra rapid urease tests have shown that the accuracy of ultra rapid tests is equivalent to that of traditional ones.^10–15 The present study seems to indicate that the URUT version developed by Biohit^® is equivalent to the standard RUT used in Spain, reaching approximately 90% sensitivity and specificity in patients naïve to eradication therapy. In general, the use of a validated URUT, as it offers equivalent results than 24 h kits, may be recommended in order to start treatment earlier and avoid doubling visits to the outpatient clinic which would reduce overall costs in the management of H. pylori infection and minimize inconveniences to patients.

Conflict of interests

Dr. Gisbert has served as speaker, consultant and advisory member for or has received research funding from Almirall, Nycomed, AstraZeneca, Casen Recordati, and Allergan. Dr. McNicholl has received retribution from Allergan for formative actions. Dr. Castro-Fernandez has received retribution from Allergan for formative actions. Dr. Perez-Aísa has received retribution from Allergan and Mylan for formative actions.

Acknowledgements

CIBERehd is funded by the Instituto de Salud Carlos III. This study was not funded by any Pharmaceutical Company.