Surgery Today

, Volume 43, Issue 11, pp 1281–1285

Criteria for the glucagon provocative test in the diagnosis of gastrinoma

Authors

    • Division of Biological Regulation and Oncology, Department of SurgeryTohoku University School of Medicine
  • Masayuki Kakyo
    • Division of Biological Regulation and Oncology, Department of SurgeryTohoku University School of Medicine
  • Makoto Kinouchi
    • Division of Biological Regulation and Oncology, Department of SurgeryTohoku University School of Medicine
  • Naoki Tanaka
    • Division of Biological Regulation and Oncology, Department of SurgeryTohoku University School of Medicine
  • Koh Miura
    • Division of Biological Regulation and Oncology, Department of SurgeryTohoku University School of Medicine
  • Takeshi Naitoh
    • Division of Biological Regulation and Oncology, Department of SurgeryTohoku University School of Medicine
  • Hitoshi Ogawa
    • Division of Biological Regulation and Oncology, Department of SurgeryTohoku University School of Medicine
  • Fuyuhiko Motoi
    • Division of Biological Regulation and Oncology, Department of SurgeryTohoku University School of Medicine
  • Shinichi Egawa
    • Division of Biological Regulation and Oncology, Department of SurgeryTohoku University School of Medicine
  • Tatsuya Ueno
    • Department of SurgerySouth Miyagi Medical Center
  • Hiroo Naito
    • Department of SurgerySouth Miyagi Medical Center
  • Michiaki Unno
    • Division of Biological Regulation and Oncology, Department of SurgeryTohoku University School of Medicine
Original Article

DOI: 10.1007/s00595-012-0334-2

Cite this article as:
Shibata, C., Kakyo, M., Kinouchi, M. et al. Surg Today (2013) 43: 1281. doi:10.1007/s00595-012-0334-2

Abstract

Introduction

The glucagon provocative test is useful for the diagnosis of gastrinoma. The aim of this study was to determine the criteria for the glucagon provocative test.

Methods

This study reviewed 8 patients that underwent the glucagon provocative test preoperatively and in whom the diagnosis was confirmed as gastrinoma histologically. The glucagon provocative test was performed by administering glucagon (20 μg/kg) intravenously, followed by 20 μg/kg h for the next 30 min, and plasma gastrin levels were measured 3 and 1 min before and 3, 5, 7, 10, 15, 20, and 30 min after the administration of glucagon. This study evaluated the peak value of plasma gastrin and the time required to reach the peak.

Results

Two of the 8 patients had multiple endocrine neoplasm type 1. The basal plasma gastrin levels ranged from 524 to 10,300 pg/ml. The time required to reach the peak was 3–10 min for all patients. The increase in the peak from the basal value was 235–8,920 pg/ml, and the percentage of increase was 38–337 %.

Conclusions

These results suggest that a diagnosis of gastrinoma should thus be made when plasma gastrin levels peak within 10 min after glucagon administration, with an increase of greater than 200 pg/ml and greater than 35 % of the basal value.

Keywords

GastrinomaGlucagonPancreaticoduodenectomy

Introduction

Gastrinomas in patients with Zollinger-Ellison syndrome are often very small (<1 cm) and difficult to localize. There are various causes of hypergastrinemia including the use of proton pump inhibitors and histamine-2 receptor antagonists, and chronic gastritis associated with chronic Helicobacter pylori infection [1]; therefore, it is important to distinguish between gastrinoma and other causes of hypergastrinemia. Although plasma gastrin levels decrease after intravenous injection of secretin in normal subjects, the plasma gastrin levels increase paradoxically in patients with gastrinoma [2]. The secretin provocative test, which utilizes this phenomenon of ‘paradoxical increase’, is a well-established diagnostic test for gastrinoma [3, 4]. It is, however, of concern that secretin is difficult to obtain not only in this country but also in other countries [5], and calcium is regarded as a candidate for an alternative secretagogue to secretin [6]. A previous report showed the glucagon provocative test to be useful in the diagnosis of gastrinoma in 3 patients, because a paradoxical increase in plasma gastrin levels was also seen after intravenous glucagon administration [7]. The diagnostic criteria for the glucagon provocative test, however, could not be determined because the study did not include a sufficient number of patients. It is also possible that some patients with gastrinoma do not respond to intravenous glucagon. Increased experience with patients with gastrinoma demonstrated the need to determine the diagnostic criteria for the glucagon provocative test.

Patients and methods

The study reviewed the medical records of 8 patients with a plasma gastrin concentration above the normal range and that underwent resection with a preoperative diagnosis of gastrinoma. All of the patients had a neuroendocrine tumor (NET) that was stained positive for gastrin by immunohistochemistry in the histological examination. The glucagon provocative test was performed before the operation in all of these patients. The study analyzed the age, sex, past history, operation performed, primary sites of gastrinoma, and the results of the glucagon provocative test. The glucagon provocative test was performed by injecting 20 μg/kg glucagon (Glucagon-G Novo, Eisai Co., Ltd., Tokyo, Japan) into a subcutaneous vein in the forearm followed by an infusion at 20 μg/kg h for 30 min. Blood samples were drawn from the contralateral cubital vein at 3 and 1 min before and 3, 5, 7, 10, 15, 20, and 30 min after the administration of glucagon. Plasma gastrin levels were estimated with a specific radioimmunoassay (SRL Inc., Tokyo, Japan). The following parameters of plasma gastrin levels were evaluated: (1) the average of values 3 and 1 min before glucagon administration as the basal value (pg/ml), (2) the time required to reach the peak (peak time), (3) peak (pg/ml), (4) the increase of peak (pg/ml) calculated by subtracting basal value from the peak, and (5) the percentage of increase in the peak calculated as the increase of peak/mean basal value × 100. The correlation between the basal value, peak, and increase of peak was analyzed using Pearson’s correlation coefficient and p values less than 0.05 were considered to be significant.

The glucagon provocative test was conducted in the same manner in several patients with an increased fasting plasma gastrin level but with no tumor detected by imaging modalities.

Results

A summary of the data from the 8 patients (5 males and 3 females) is shown in Table 1. The results of patients 1–3 were reported in a previous study [7]. Two of the 8 patients (patients 5 and 8) had proven multiple endocrine neoplasia (MEN) type 1. The subjects’ ages ranged from 42 to 74 with a mean of 62. Patient 1 had undergone an antrectomy for a perforated duodenal ulcer, but none of the 7 remaining patients had a history of abdominal surgery. Subtotal stomach-preserving pancreaticoduodenectomy was performed in 4 patients, and conventional pancreaticoduodenectomy, distal gastrectomy, resection of the gastric remnant with the tumor, and enucleation of the tumor in the pancreatic uncinate were performed in 1 patient each. Patient 7 underwent simultaneous partial hepatic resection due to 3 liver metastases at the time of diagnosis. The primary sites of gastrinoma were the pancreatic head in 2 patients, duodenum in 2 patients, and the pancreatic uncinate, the duodenum and pancreatic body, stomach, and lesser omentum in 1 patient each.
Table 1

Summary of the 8 patients with gastrinomas

NO.

Age, Sex

Previous operation

Performed procedure

Primary site

Basal value (pg/ml)

Peak time (min)

Peak (pg/ml)

Increase in peak (pg/ml)

Percent increase

1

74, F

Distal gastrectomy

Resection of the gastric remnant with tumor

Lesser omentum

1,475

10

2,270

795

54

2

53, M

None

SSPPDb

Pancreatic head

10,300

3

15,800

5,500

53

3

55, F

None

Enucleation of the tumor

Pancreatic uncinate

804

7

3,507

2,703

337

4

61, M

None

PDc

Stomach

9,570

10

14,237

4,667

49

5a

52, M

None

SSPPDb

Pancreatic head

2,390

5

3,566

1,176

49

6

74, M

None

Distal gastrectomy

Duodenum

625

3

860

235

38

7

61, F

None

SSPPDb + Partial hepatectomy

Duodenum

3,623

5

12,543

8,920

246

8a

67, M

None

SSPPDb

Duodenum, Pancreatic body

524

5

761

237

45

aPatient with MEN type 1

bSubtotal stomach-preserving pancreaticoduodenectomy

cPancreaticoduodenectomy

The time-concentration relationship in the glucagon provocative test in 8 patients is shown in Fig. 1. Although the plasma gastrin levels before and after administration varied in each patient, a rapid increase in plasma gastrin levels was observed in all 8 patients. The results of the glucagon provocative test are also summarized in Table 1. The basal plasma gastrin levels before administration were above the normal range (less than 200 pg/ml) in all patients, with a minimum value of 524 pg/ml (patient 8) and a maximum value of 10,300 pg/ml (patient 2). Plasma gastrin levels increased after glucagon administration in all 8 patients and reached a peak level at 3 min in 2 patients, 5 min in 3 patients, 7 min in 1 patient, and 10 min in 2 patients. The peak value was the lowest in patient 8 (761 pg/ml) and highest in patient 2 (15,800 pg/ml). The increase to the peak was the least in patient 6 (235 pg/ml) and highest in patient 7 (8,920 pg/ml). There was a significant correlation between the basal and peak values (p = 0.0004) but not between the peak values and the increase in the peak (p = 0.8831) and between the basal value and the increase in the peak (p = 0.588). The percentage of increase in the peak value was the least in patient 6 (38 %) and the greatest in patient 3 (337 %).
https://static-content.springer.com/image/art%3A10.1007%2Fs00595-012-0334-2/MediaObjects/595_2012_334_Fig1_HTML.gif
Fig. 1

Plasma gastrin concentration after the intravenous administration of glucagon in 8 patients. Plasma gastrin concentration increased rapidly after the administration of glucagon. Arrows indicate the time of glucagon administration at a dose of 20 μg/kg + 20 μg/kg h for 30 min

Figure 2 shows the results of the glucagon provocative test in a patient with increased plasma gastrin levels in the fasting state but with no tumor detected by imaging modalities. This patient was not diagnosed to have gastrinoma, because the plasma gastrin levels decreased after the intravenous administration of glucagon.
https://static-content.springer.com/image/art%3A10.1007%2Fs00595-012-0334-2/MediaObjects/595_2012_334_Fig2_HTML.gif
Fig. 2

Plasma gastrin concentration after the intravenous administration of glucagon in a patient without gastrinoma. The arrow indicates the time of glucagon administration at a dose of 20 μg/kg + 20 μg/kg h for 30 min

Discussion

The effect of glucagon in increasing plasma gastrin levels has been reported previously in patients with gastrinoma [7, 8]. The plasma gastrin levels in all 8 patients histologically confirmed to have a diagnosis of gastrinoma increased rapidly in response to the administration of intravenous glucagon. An analysis of the time to reach the peak value, the increase in the peak value, and the percentage of increase over the basal value suggested that patients should be diagnosed to have a gastrinoma if an increase in plasma gastrin levels greater than 200 pg/ml and 35 % is observed within 10 min after the administration of glucagon. The secretin provocative test has been considered the most powerful tool in the differential diagnosis of gastrinoma, and a standard diagnostic criterion for this test is a post-administration increase in plasma gastrin levels of greater than 200 pg/ml in one study [1] and 110 pg/ml in another study [9]. Berna et al. [10] suggested that the standard criterion for a positive secretin test should be an increase in plasma gastrin levels greater than 120 pg/ml rather than 200 pg/ml. These levels of plasma gastrin for the diagnosis of gastrinoma in the secretin test are comparable to those with the glucagon provocative test in the current series.

One study, however, reported that the plasma gastrin levels do not always increase in response to intravenous glucagon [11]. It is possible that there are some patients with gastrinoma whose gastrin levels do not reach the diagnostic criteria according to the current study with the glucagon provocative test. In addition, some patients with a gastrinoma show a negative response to secretin.

The use of the glucagon provocative test may prove important in patients with gastrinoma to determine the completeness of resection during surgery and possibly in diagnosing a postoperative recurrence. The plasma gastrin levels do not increase in response to glucagon after the complete resection of gastrinomas [7]. The criteria for plasma gastrin levels proposed in this study might also be applicable to these intraoperative and postoperative tests. The identification of a secretagogue for preoperative, intraoperative, and postoperative provocative testing could also be used with a selective arterial secretagogue injection (SASI) test, in which the regional localization of the gastrinoma could be determined by selective administration to various anatomic regions and evaluation of blood samples from the hepatic vein [12]. A glucagon at a dose of 100 μg is an appropriate dose for SASI [7]. The criteria for changes in the plasma gastrin levels during the SASI test with glucagon thus need to be determined in further studies.

Roy et al. [13] reported that the primary site of the gastrinoma could not be determined based on surgery and imaging modalities in 61 of 261 (23 %) patients. A patient with gastrinoma had a tumor that could not be localized by imaging modalities, but the feeding artery for the tumor was suggested to be a gastroduodenal artery by a SASI test with secretin. Secretin, however, has not been available since 2004 in this country [14]. These facts indicate that an adequate secretagogue alternative to secretin is important in the diagnosis and localization of gastrinomas. Glucagon has an advantage in comparison to calcium, because glucagon induces a rapid and sharp increase of plasma gastrin levels. The term ‘paradoxical increase’ is applicable to the increase in plasma gastrin levels after glucagon administration in patients with gastrinoma, because the intravenous administration of glucagon decreases plasma gastrin levels in normal subjects [15]. Secretin receptors present in gastrinomas and the expression of the total secretin receptor are correlated with the results of the secretin test [16, 17]. Therefore, the direct binding of secretin to secretin receptors on gastrinoma cells must be the mechanism by which secretin induces the ‘paradoxical increase.’ A similar mechanism could be considered for glucagon, and it is important to identify the glucagon receptors on gastrinoma cells by either immunohistochemistry or other methods.

Acknowledgments

The authors thank Dr. Michael G. Sarr, Department of Surgery, Mayo Clinic, MN, USA, for reviewing this manuscript.

Copyright information

© Springer 2012