Ann Nucl Med DOI 10.1007/s12149-015-1005-3
ORIGINAL ARTICLE
Carbon-14 urea breath test: does it work in patients with partial gastric resection? Fuat Dede1,5 • Hu¨seyin Civen2 • Faysal Dane3 • Mehmet Aliustaoglu4 Serdar Turhal3 • Halil Turgut Turoglu1 • Sabahat Inanir1
•
Received: 3 March 2015 / Accepted: 8 July 2015 Ó The Japanese Society of Nuclear Medicine 2015
Abstract Objective The diagnostic value of Carbon-14 urea breath test (C-14 UBT) in the detection of Helicobacter pylori (H. pylori) infection in non-operated patients has been proved. However, the efficacy of C-14 UBT in patients with partial gastric resection (PGR) has not been evaluated yet. Herein, the results of the C-14 UBT and H. pylori stool antigen test (HpSAT) in this patient group were compared with the endoscopic findings. Methods Multi-breath samples C-14 UBT and HpSAT were performed in all patients on the same day. Histology was used as a gold standard for testing C-14 UBT and HpSAT diagnostic efficacies. Results 30 patients (mean age: 54.6 ± 11 year) with PGR were included. The sensitivity and specificity of standard C-14 UBT were 29 and 100 %, respectively. When breath samples were collected at 20th min, and [35 CPM was selected as radioactivity threshold, the sensitivity raised to 86 % without any loss of specificity. The
& Fuat Dede [email protected] 1
Department of Nuclear Medicine, Marmara University School of Medicine, Istanbul, Turkey
2
Nuclear Medicine Clinic, Kocaeli State Hospital, Kocaeli, Turkey
3
Department of Medical Oncology, Marmara University School of Medicine, Istanbul, Turkey
4
Internal Medicine Clinic, Kartal Dr. Lutfi Kirdar Research and Training Hospital, Istanbul, Turkey
5
Nukleer Tip Anabilim Dali, S.B. Marmara Universitesi Pendik Egitim ve Arastirma Hastanesi, -1 kat A1 Blok Fevzi Cakmak Mahallesi Mimar Sinan Caddesi No:41 Ustkaynarca, Pendik, Istanbul, Turkey
specificity and sensitivity of the HpSAT were 71 and 96 %, respectively. Conclusions The sensitivity of the standard C-14 UBT was very poor for patients with PGR, and results of HpSAT were superior in this population. Certain modifications are needed if C-14 UBT is to be used in PGR patients. Keywords Carbon-14 urea breath test Stool antigen test Partial gastrectomy Gastric cancer Helicobacter pylori
Introduction After the discovery of Helicobacter pylori (H. pylori) in 1982, epidemiologic studies have revealed that it is a very common pathogen in the society that has infected nearly half of the world’s population [1, 2]. Geographic area, age, race, and socioeconomic status determine the prevalence of H. pylori infection [3]. H. pylori plays an important role in the development of duodenal ulcer (responsible for 90–95 % of all duodenal ulcers), atrophic gastritis and gastric cancer (represents nearly 5.5 % of all cancers and 25 % of all infection-related cancers) [4–6]. Because of this, diagnosing this bacteria and thereafter starting multidrug eradication therapy is important. Invasive (endoscopy, histology, rapid urease test, and culture) and non-invasive [serology, urea breath test (UBT), and H. pylori stool antigen test (HpSAT)] methods are used to diagnose this microorganism [7]. UBT is a very successful method for both initial diagnosis of H. pylori and monitoring response to treatment [7]. The principles and mechanisms of the test are as follows: The labeled urea [with either non-radioactive carbon-13 (C-13) or radioactive carbon-14 (C-14)] in the test material
123
Ann Nucl Med
is degraded to carbon dioxide (CO2) and ammonia with the presence of urease, an enzyme that is synthesized by H. pylori. The labeled CO2 is then absorbed from gastric mucosa and exhaled. The detection of the labeled CO2 in the exhaled breath confirms the diagnosis of H. pylori. C-14 UBT is a cheap and rapid test that does not require a test meal [8]. On the other hand, the need for authorized centers for handling radioactive material and transport problems limit its usage. Since C-13 is a stable isotope, it can be safely used in children and childbearing women. The major disadvantage of C-13 UBT is its higher cost. The risk of developing cancer in the residual stomach is increased in patients who underwent partial gastric resection (PGR) due to benign (ulcer, etc.) or malignant causes [9–11]. The incidence of gastric stump cancer reached up to 2 % in this group [12]. As in non-operated patients, atrophic gastritis, intestinal metaplasia, dysplasia, and finally gastric cancer are closely related with the H. pylori infection in PGR patients [1]. Therefore, these high-risk cases should be screened for H. pylori, and eradication therapy should be given when the infection is detected [9– 11]. The importance of UBT in patients without history of gastric surgery is undebatable [13, 14]. However, studies with C-13 UBT showed that with its relatively low sensitivity and specificity rates (77 and 89 %, respectively), the test failed to meet the expectations after PGR [15–18]. Decreased gastric volume, decreased bacterial load, rapid gastric emptying, and increased gastric pH could be responsible for the failure of the C-14 UBT in partially gastrectomized patients [17, 19, 20]. Certain modifications were recommended in order to suppress the effects of these factors. On the other hand, the diagnostic performance of C-14 UBT in this patient group has not been fully investigated. The aim of this study was to evaluate the performance of the standard and modified C-14 UBT and to compare results with HpSAT and endoscopy in patients with PGR.
Patients who received eradication therapy for H. pylori after surgery and patients who were found to have taken medication (bismuth, antibiotics, proton pump inhibitors, H2 blockers, and antacids) 4 weeks before the diagnostic tests were excluded from the study. This study was approved by the institutional ethics committee and all patients gave informed consent for participation in the study. Histopathological analysis All patients underwent postoperative routine fiber-optic esophagogastroscopy 15 days to 4 weeks (median 2.5 weeks) before the non-invasive tests (C-14 UBT and HpSAT). During procedure, multiple mucosal biopsy samples were obtained and stained with hematoxylin and eosin and modified Giemsa. Specimens were examined for the presence of H. pylori. Histology was used as a gold standard for testing C-14 UBT and HpSAT diagnostic efficacies. Carbon-14 Urea breath test After an overnight fast, C-14 UBT (HeliprobeÒ System, Kibion AB, Uppsala, Sweden) was performed for all enrolled patients. After ingestion of 37 kBq (1 lCi) C-14 Urea capsule with 50 mL water, the breath samples were collected at 10th, 20th, and 30th min after ingestion. During this 30-minute C-14 UBT urea reaction period, patients lay on the left side horizontally. The results were expressed as both counts per minute (CPM), and grading [0–1 (negative for H. pylori infection, CPM B 50), and 2 (positive for H. pylori infection, CPM [ 50)] as suggested by the manufacturer. H. pylori stool antigen test (HpSAT) For all cases, HpSAT was performed simultaneously with C-UBT on the same day. H. pylori antigens in feces were investigated by monoclonal antibodies with one-step colored immunochromatography (ImmunoCard STAT! HpSA, Meridian Diagnostics Inc., Cincinnati, Ohio, USA) technique according to the standard manufacturer recommendations. 14
Materials and methods Patients A total of 30 gastric cancer patients (F/M = 8/22; mean age = 54.6 ± 11 years; range = 30–74 years) with PGR were included in this prospective study. Billroth II procedure was performed in 26 patients (87 %), Roux-en-Y anastomosis in 3 patients (10 %), and wedge resection was done in one case (3 %). The pathology was reported as adenocarcinoma in all but two patients (94 %). Gastrointestinal stromal tumor was the diagnosis in one patient (3 %) and maltoma in the other one (3 %).
123
Statistics Statistical analysis was performed by GraphPad InStat Version 3.00 (GraphPad Software Inc, Sandiego, California, USA) and MedCalc Version 11.6.1.0 (MedCalc Software, Mariakerke, Belgium). Based on analyzed data, Kruskal–Wallis test (non-parametric ANOVA), the Mann– Whitney U-test, Spearman non-parametric correlation,
Ann Nucl Med
ROC curve analysis, and comparison of ROC curves were used. A p value less than 0.05 was considered significant.
Results The interval between the surgery and UBT ranged from 27 days to 21 years (mean = 27.7 ± 47 months; median = 14 months). H. pylori was detected in 7 patients (23.3 %) at follow-up endoscopy. The mean ages of H. pylori positive and negative patients were 53 ± 7 and 55 ± 23 years, respectively. No statistically significant difference in terms of age was detected between these two groups (Mann–Whitney U-test). UBT results With standard C-14 UBT criteria, H. pylori was detected in 2 patients at 10th min, 4 patients at 20th min, and 5 patients at 30th min. The sensitivity rates for each time point were 29, 57, and 71 %, respectively, while the specificity was 100 % for all (Table 1). UBT radioactivity counts (CPM) The C-14 UBT radioactivity counts in non-operated H. pylori (?) patients were reported to change between 69 CPM and 770 CPM (median and mean 269 and 300 CPM, respectively) [21]. When the radioactivity counts in H. pylori positive and negative patients were analyzed separately, we found that the breath sampling time did not have a statistically significant impact on radioactivity count rates (Kruskal–Wallis test, P [ 0.05, Table 2). When endoscopy Table 1 Results of C-14 UBT and HpSAT
ta
was assumed to be the gold standard, the radioactivity thresholds for 10th, 20th, and 30th min were found as [23, [35, and [29 CPM, respectively, by ROC curve analysis (Fig. 1). The sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), and accuracy are presented in Table 1. Although no statistically significant difference was found between three ROC curves, the best performance was achieved with 20th min breath sampling and radioactivity threshold [35 CPM. HpSAT results It gave true positive results in 5 patients and false-positive result in one patient. The sensitivity and specificity of HpSAT in our study population were 71.4 and 95.7 %, respectively (Table 1). Elapsed time after surgery Although not statistically significant, inverse relation between the prevalence of H. pylori and the elapsed time after surgery was found in partially gastrectomized patients (Spearman non-parametric correlation, r: -0.69, p [ 0.05, Fig. 2).
Discussion Regardless of the type and cause of the surgery, patients with PGR are prone to developing gastric cancer [9–11]. Enterogastric reflux and H. pylori colonization in the residual stomach are the main risk factors for the occurrence of malignancy [16]. In this group of subjects,
Test Thresholdb
Sensitivity
Specificity
NPV
PPV
Accuracy
10
[50c
28.6
100
82.1
100
83.3
20 30
[50c [50c
57.1 71.4
100 100
88.5 92
100 100
90 93
10
[23
85.7
86.9
95.2
66.7
87
20
[35
85.7
100
95.8
100
97
30
[29
71.4
100
92
100
93
C-14 UBT
HpSAT
71.4
95.7
91.7
83
90
C-14 UBTd and HpSAT
85.7
95.7
95.7
85.7
93.3
Bold values indicate better results NPV negative predictive value, PPV positive predictive value, UBT urea breath test, HpSAT H. pylori stool antigen test a
breath sampling time (minute)
b
counts per minute (CPM)
c
standard radioactivity threshold
d
Standard C-14 UBT
123
Ann Nucl Med Table 2 Comparison of radioactivity counts at 10th, 20th, and 30th min in H. Pylori (?) and (-) patients
10th min
20th min
30th min
p:
H. pylori (?) (n:7)
65.1 ± 72 CPM
77 ± 66 CPM
72 ± 58 CPM
0.84*
H. pylori (-) (n:23)
15.3 ± 10 CPM
14.3 ± 9 CPM
15.4 ± 8 CPM
0.80*
p:
0.0011**
ORIGINAL ARTICLE
Carbon-14 urea breath test: does it work in patients with partial gastric resection? Fuat Dede1,5 • Hu¨seyin Civen2 • Faysal Dane3 • Mehmet Aliustaoglu4 Serdar Turhal3 • Halil Turgut Turoglu1 • Sabahat Inanir1
•
Received: 3 March 2015 / Accepted: 8 July 2015 Ó The Japanese Society of Nuclear Medicine 2015
Abstract Objective The diagnostic value of Carbon-14 urea breath test (C-14 UBT) in the detection of Helicobacter pylori (H. pylori) infection in non-operated patients has been proved. However, the efficacy of C-14 UBT in patients with partial gastric resection (PGR) has not been evaluated yet. Herein, the results of the C-14 UBT and H. pylori stool antigen test (HpSAT) in this patient group were compared with the endoscopic findings. Methods Multi-breath samples C-14 UBT and HpSAT were performed in all patients on the same day. Histology was used as a gold standard for testing C-14 UBT and HpSAT diagnostic efficacies. Results 30 patients (mean age: 54.6 ± 11 year) with PGR were included. The sensitivity and specificity of standard C-14 UBT were 29 and 100 %, respectively. When breath samples were collected at 20th min, and [35 CPM was selected as radioactivity threshold, the sensitivity raised to 86 % without any loss of specificity. The
& Fuat Dede [email protected] 1
Department of Nuclear Medicine, Marmara University School of Medicine, Istanbul, Turkey
2
Nuclear Medicine Clinic, Kocaeli State Hospital, Kocaeli, Turkey
3
Department of Medical Oncology, Marmara University School of Medicine, Istanbul, Turkey
4
Internal Medicine Clinic, Kartal Dr. Lutfi Kirdar Research and Training Hospital, Istanbul, Turkey
5
Nukleer Tip Anabilim Dali, S.B. Marmara Universitesi Pendik Egitim ve Arastirma Hastanesi, -1 kat A1 Blok Fevzi Cakmak Mahallesi Mimar Sinan Caddesi No:41 Ustkaynarca, Pendik, Istanbul, Turkey
specificity and sensitivity of the HpSAT were 71 and 96 %, respectively. Conclusions The sensitivity of the standard C-14 UBT was very poor for patients with PGR, and results of HpSAT were superior in this population. Certain modifications are needed if C-14 UBT is to be used in PGR patients. Keywords Carbon-14 urea breath test Stool antigen test Partial gastrectomy Gastric cancer Helicobacter pylori
Introduction After the discovery of Helicobacter pylori (H. pylori) in 1982, epidemiologic studies have revealed that it is a very common pathogen in the society that has infected nearly half of the world’s population [1, 2]. Geographic area, age, race, and socioeconomic status determine the prevalence of H. pylori infection [3]. H. pylori plays an important role in the development of duodenal ulcer (responsible for 90–95 % of all duodenal ulcers), atrophic gastritis and gastric cancer (represents nearly 5.5 % of all cancers and 25 % of all infection-related cancers) [4–6]. Because of this, diagnosing this bacteria and thereafter starting multidrug eradication therapy is important. Invasive (endoscopy, histology, rapid urease test, and culture) and non-invasive [serology, urea breath test (UBT), and H. pylori stool antigen test (HpSAT)] methods are used to diagnose this microorganism [7]. UBT is a very successful method for both initial diagnosis of H. pylori and monitoring response to treatment [7]. The principles and mechanisms of the test are as follows: The labeled urea [with either non-radioactive carbon-13 (C-13) or radioactive carbon-14 (C-14)] in the test material
123
Ann Nucl Med
is degraded to carbon dioxide (CO2) and ammonia with the presence of urease, an enzyme that is synthesized by H. pylori. The labeled CO2 is then absorbed from gastric mucosa and exhaled. The detection of the labeled CO2 in the exhaled breath confirms the diagnosis of H. pylori. C-14 UBT is a cheap and rapid test that does not require a test meal [8]. On the other hand, the need for authorized centers for handling radioactive material and transport problems limit its usage. Since C-13 is a stable isotope, it can be safely used in children and childbearing women. The major disadvantage of C-13 UBT is its higher cost. The risk of developing cancer in the residual stomach is increased in patients who underwent partial gastric resection (PGR) due to benign (ulcer, etc.) or malignant causes [9–11]. The incidence of gastric stump cancer reached up to 2 % in this group [12]. As in non-operated patients, atrophic gastritis, intestinal metaplasia, dysplasia, and finally gastric cancer are closely related with the H. pylori infection in PGR patients [1]. Therefore, these high-risk cases should be screened for H. pylori, and eradication therapy should be given when the infection is detected [9– 11]. The importance of UBT in patients without history of gastric surgery is undebatable [13, 14]. However, studies with C-13 UBT showed that with its relatively low sensitivity and specificity rates (77 and 89 %, respectively), the test failed to meet the expectations after PGR [15–18]. Decreased gastric volume, decreased bacterial load, rapid gastric emptying, and increased gastric pH could be responsible for the failure of the C-14 UBT in partially gastrectomized patients [17, 19, 20]. Certain modifications were recommended in order to suppress the effects of these factors. On the other hand, the diagnostic performance of C-14 UBT in this patient group has not been fully investigated. The aim of this study was to evaluate the performance of the standard and modified C-14 UBT and to compare results with HpSAT and endoscopy in patients with PGR.
Patients who received eradication therapy for H. pylori after surgery and patients who were found to have taken medication (bismuth, antibiotics, proton pump inhibitors, H2 blockers, and antacids) 4 weeks before the diagnostic tests were excluded from the study. This study was approved by the institutional ethics committee and all patients gave informed consent for participation in the study. Histopathological analysis All patients underwent postoperative routine fiber-optic esophagogastroscopy 15 days to 4 weeks (median 2.5 weeks) before the non-invasive tests (C-14 UBT and HpSAT). During procedure, multiple mucosal biopsy samples were obtained and stained with hematoxylin and eosin and modified Giemsa. Specimens were examined for the presence of H. pylori. Histology was used as a gold standard for testing C-14 UBT and HpSAT diagnostic efficacies. Carbon-14 Urea breath test After an overnight fast, C-14 UBT (HeliprobeÒ System, Kibion AB, Uppsala, Sweden) was performed for all enrolled patients. After ingestion of 37 kBq (1 lCi) C-14 Urea capsule with 50 mL water, the breath samples were collected at 10th, 20th, and 30th min after ingestion. During this 30-minute C-14 UBT urea reaction period, patients lay on the left side horizontally. The results were expressed as both counts per minute (CPM), and grading [0–1 (negative for H. pylori infection, CPM B 50), and 2 (positive for H. pylori infection, CPM [ 50)] as suggested by the manufacturer. H. pylori stool antigen test (HpSAT) For all cases, HpSAT was performed simultaneously with C-UBT on the same day. H. pylori antigens in feces were investigated by monoclonal antibodies with one-step colored immunochromatography (ImmunoCard STAT! HpSA, Meridian Diagnostics Inc., Cincinnati, Ohio, USA) technique according to the standard manufacturer recommendations. 14
Materials and methods Patients A total of 30 gastric cancer patients (F/M = 8/22; mean age = 54.6 ± 11 years; range = 30–74 years) with PGR were included in this prospective study. Billroth II procedure was performed in 26 patients (87 %), Roux-en-Y anastomosis in 3 patients (10 %), and wedge resection was done in one case (3 %). The pathology was reported as adenocarcinoma in all but two patients (94 %). Gastrointestinal stromal tumor was the diagnosis in one patient (3 %) and maltoma in the other one (3 %).
123
Statistics Statistical analysis was performed by GraphPad InStat Version 3.00 (GraphPad Software Inc, Sandiego, California, USA) and MedCalc Version 11.6.1.0 (MedCalc Software, Mariakerke, Belgium). Based on analyzed data, Kruskal–Wallis test (non-parametric ANOVA), the Mann– Whitney U-test, Spearman non-parametric correlation,
Ann Nucl Med
ROC curve analysis, and comparison of ROC curves were used. A p value less than 0.05 was considered significant.
Results The interval between the surgery and UBT ranged from 27 days to 21 years (mean = 27.7 ± 47 months; median = 14 months). H. pylori was detected in 7 patients (23.3 %) at follow-up endoscopy. The mean ages of H. pylori positive and negative patients were 53 ± 7 and 55 ± 23 years, respectively. No statistically significant difference in terms of age was detected between these two groups (Mann–Whitney U-test). UBT results With standard C-14 UBT criteria, H. pylori was detected in 2 patients at 10th min, 4 patients at 20th min, and 5 patients at 30th min. The sensitivity rates for each time point were 29, 57, and 71 %, respectively, while the specificity was 100 % for all (Table 1). UBT radioactivity counts (CPM) The C-14 UBT radioactivity counts in non-operated H. pylori (?) patients were reported to change between 69 CPM and 770 CPM (median and mean 269 and 300 CPM, respectively) [21]. When the radioactivity counts in H. pylori positive and negative patients were analyzed separately, we found that the breath sampling time did not have a statistically significant impact on radioactivity count rates (Kruskal–Wallis test, P [ 0.05, Table 2). When endoscopy Table 1 Results of C-14 UBT and HpSAT
ta
was assumed to be the gold standard, the radioactivity thresholds for 10th, 20th, and 30th min were found as [23, [35, and [29 CPM, respectively, by ROC curve analysis (Fig. 1). The sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), and accuracy are presented in Table 1. Although no statistically significant difference was found between three ROC curves, the best performance was achieved with 20th min breath sampling and radioactivity threshold [35 CPM. HpSAT results It gave true positive results in 5 patients and false-positive result in one patient. The sensitivity and specificity of HpSAT in our study population were 71.4 and 95.7 %, respectively (Table 1). Elapsed time after surgery Although not statistically significant, inverse relation between the prevalence of H. pylori and the elapsed time after surgery was found in partially gastrectomized patients (Spearman non-parametric correlation, r: -0.69, p [ 0.05, Fig. 2).
Discussion Regardless of the type and cause of the surgery, patients with PGR are prone to developing gastric cancer [9–11]. Enterogastric reflux and H. pylori colonization in the residual stomach are the main risk factors for the occurrence of malignancy [16]. In this group of subjects,
Test Thresholdb
Sensitivity
Specificity
NPV
PPV
Accuracy
10
[50c
28.6
100
82.1
100
83.3
20 30
[50c [50c
57.1 71.4
100 100
88.5 92
100 100
90 93
10
[23
85.7
86.9
95.2
66.7
87
20
[35
85.7
100
95.8
100
97
30
[29
71.4
100
92
100
93
C-14 UBT
HpSAT
71.4
95.7
91.7
83
90
C-14 UBTd and HpSAT
85.7
95.7
95.7
85.7
93.3
Bold values indicate better results NPV negative predictive value, PPV positive predictive value, UBT urea breath test, HpSAT H. pylori stool antigen test a
breath sampling time (minute)
b
counts per minute (CPM)
c
standard radioactivity threshold
d
Standard C-14 UBT
123
Ann Nucl Med Table 2 Comparison of radioactivity counts at 10th, 20th, and 30th min in H. Pylori (?) and (-) patients
10th min
20th min
30th min
p:
H. pylori (?) (n:7)
65.1 ± 72 CPM
77 ± 66 CPM
72 ± 58 CPM
0.84*
H. pylori (-) (n:23)
15.3 ± 10 CPM
14.3 ± 9 CPM
15.4 ± 8 CPM
0.80*
p:
0.0011**
