Effect of Long-Term Gastric Acid Suppressive Therapy on Serum Vitamin B12 Levels in Patients with Zollinger-Ellison Syndrome

Effect of Long-Term Gastric Acid Suppressive Therapy on Serum Vitamin B12 Levels in Patients with Zollinger-Ellison Syndrome Basel Termanini, MD, Fathia Gibril, MD, Vincent E. Sutliff, MD, Fang Yu, MD, David J. Venzon, PhD, Robert T. Jensen, MD BACKGROUND AND AIMS: Long-term treatment with H1K1-adenotriphosphatase (ATPase) inhibitors, such as omeprazole or lansoprazole, for severe gastroesophageal reflux disease is now widely used. Whether such treatment will result in vitamin B12 deficiency is controversial. We studied whether longterm treatment with omeprazole alters serum vitamin B12 levels in patients with Zollinger-Ellison syndrome. METHODS: In 131 consecutive patients treated with either omeprazole (n 5 111) or histamine H2-receptor antagonists (n 5 20), serum vitamin B12 and folate levels and complete blood counts were determined after acid secretion had been controlled for at least 6 months. These studies were repeated yearly. Serum vitamin B12 and folate levels were correlated with the type of antisecretory drug and the extent of inhibition of acid secretion. RESULTS: The mean duration of omeprazole treatment was 4.5 years, and for H2-receptor antagonists 10 years. Vitamin B12 levels, but not serum folate levels or any hematological param-

eter, were significantly (P 5 0.03) lower in patients treated with omeprazole, especially those with omeprazole-induced sustained hyposecretion (P 5 0.0014) or complete achlorhydria (P ,0.0001). In 68 patients with two determinations at least 5 years apart, vitamin B12 levels decreased significantly (30%; P 5 0.001) only in patients rendered achlorhydric. The duration of omeprazole treatment was inversely correlated with vitamin B12 levels (P 5 0.013), but not folate levels. Eight patients (6%) developed subnormal B12 levels during follow-up. CONCLUSIONS: Long-term omeprazole treatment leads to significant decreases in serum vitamin B12 but not folate levels. These results suggest patients with Zollinger-Ellison syndrome treated with H1-K1-ATPase inhibitors should have serum vitamin B12 levels monitored. Furthermore, these results raise the possibility that other patients treated chronically with H1-K1ATPase inhibitors may develop B12 deficiency. Am J Med. 1998;104:422– 430. q1998 by Excerpta Medica, Inc.

he H1-K1-adenotriphosphatase (ATPase) inhibitors (omeprazole, lansoprazole) are potent, longlasting inhibitors of gastric acid secretion (1–3). These drugs are widely used for the treatment of moderate to severe gastroesophageal reflux disease and for peptic ulcer disease (1– 4). Because gastroesophageal reflux disease recurs in 80% of patients within 6 months when H1-K1-ATPase inhibitors are stopped (5), long-term, continuous treatment is being increasingly used (4,6,7). The possible consequences of such long-term treatment with H1-K1-ATPase inhibitors are not known (8,9), especially those that result from the achlorhydria that these potent antisecretory agents frequently produce (8 –10). Although achlorhydria-induced hypergastrinemia causing gastric carcinoid tumors has been extensively studied (10,11), much less is known about the potential long-term metabolic effects of drug-induced achlorhydria or hypochlorhydria. Gastric acidity is important for

the absorption of nutrients such as vitamin B12 and iron (8,12,13). Most dietary vitamin B12 is tightly proteinbound. It is released in the stomach by the action of gastric acid and pepsin where it binds to salivary R proteins and subsequently to intrinsic factor (14,15). This complex remains intact until it binds to specific receptors in the terminal ileum, where vitamin B12 is absorbed. A portion undergoes enterohepatic recycling (15,16). By causing achlorhydria, prolonged omeprazole treatment may result in vitamin B12 deficiency if proteinbound vitamin B12 is not adequately released from food (8,12,13,17). Three studies (13,18,19) treated healthy volunteers with oral omeprazole for up to 2 weeks and found a significant decrease in vitamin B12 absorption. However, another study (20) showed that an intravenous infusion of omeprazole did not change absorption of protein-bound vitamin B12. In longer term (3 to 4 year) studies, there was either no change in hematological parameters (21) or a down-trend in serum vitamin B12 levels that did not reach significance (12). After 4 years of continuous omeprazole treatment, the mean serum vitamin B12 level was significantly lower than after 2 years of treatment (17). Although only recently approved for the treatment of gastroesophageal reflux, H1-K1-ATPase inhibitors have been used for many years to treat patients with Zollinger-

T

From the Digestive Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases (BT, FG, VES, FY, RTJ), and the Biostatistics and Data Management Section, National Cancer Institute (DJV), National Institutes of Health, Bethesda, Maryland. Requests for reprints should be addressed to Robert T. Jensen, MD, NIH/NIDDK/DDB, 10/9C103, 10 Center Dr MSC 1804, Bethesda, Maryland 20892-1804. Manuscript submitted July 21, 1997 and accepted in revised form February 25, 1998. 422

q1998 by Excerpta Medica, Inc. All rights reserved.

0002-9343/98/$19.00 PII S0002-9343(98)00087-4

Vitamin B12 Levels in Zollinger-Ellison Syndrome/Termanini et al

Ellison syndrome (22,23). This syndrome is usually caused by a gastroentero-pancreatic tumor that releases gastrin, thereby causing gastric acid hypersecretion (24), which requires daily, life-long gastric acid antisecretory treatment (23–25). A cohort of patients with ZollingerEllison syndrome followed up at the National Institutes of Health is well suited for assessment of the possible long-term effects of omeprazole on serum vitamin B12 levels. No patients underwent routine gastric resections, the duration of omeprazole treatment (mean 4.4 years, range 0.2 to 12) and of total gastric antisecretory drug therapy are long (mean 9.3 years, range 1 to 22), and yearly gastric acid studies were obtained. Some patients were treated with histamine H2-receptor antagonists, which cause less profound acid inhibition (23,25) and thus can be used for comparison. We analyzed data from 131 consecutive patients with Zollinger-Ellison syndrome followed up at the National Institutes of Health who were treated with long-term gastric acid antisecretory agents.

MATERIALS AND METHODS All patients followed up at the National Institutes of Health with a diagnosis of Zollinger-Ellison syndrome were eligible (26,27). Patients underwent a determination of basal acid output and maximal acid output (28). Tumor extent was assessed using imaging studies including selective celiac angiography, computed tomography scan, transabdominal ultrasonography, and magnetic resonance imaging (29). Fasting serum gastrin levels were determined by Bioscience Laboratories, and all samples were diluted into the normal range for accurate determination of higher values (30). Patients were evaluated for multiple endocrine neoplasia-type 1 (31). All patients were maintained on a maintenance dose of either omeprazole or a histamine H2-receptor antagonist for control of gastric acid hypersecretion. The maintenance dose of drug was established as described previously (22,23,32,33). The onset of the disease was defined as the time of onset of continuous symptoms due to gastric hypersecretion (34).

Study Design After having their gastric acid hypersecretion controlled for at least 6 months, consecutive patients with ZollingerEllison syndrome were enrolled in the study. A medical history and physical examination, blood for a complete blood count, serum vitamin B12, and fasting serum gastrin levels were obtained and were repeated yearly. Patients were questioned about the use of oral vitamin B12 preparations; none needed to be excluded for taking such medications. Serum folate levels were obtained on the last 126 patients. Serum vitamin B12 and serum folate were measured by radioimmunoassays (Ciba-Corning Co.,

Medford, Massachusetts). The interassay variation for the serum vitamin B12 and serum folate for four different standards was as follows: vitamin B12 140 pg/mL (8%), 387 pg/mL (6%), 681 pg/mL (8%), and 964 pg/mL (10%); and for serum folate levels of 1.5 ug/mL (11%), 3 ug/mL (9.8%), 5 ug/mL (6.4%), and 11 ug/mL (7.5%). In those patients who were found to have low vitamin B12 serum levels, antiparietal and anti-intrinsic factor antibodies were obtained and a Schilling test was performed. Gastric acid secretory control on the current antisecretory treatment was determined by measuring gastric acid output for 1 hour prior to the next drug dose. Results from the prior years’ assessments of acid control on antisecretory drug were reviewed. Patients were classified as having complete basal achlorhydria if they had taken omeprazole or lansoprazole for at least 3 years and if all acid control measurements showed no acid secretion during each of the 3 previous yearly evaluations. A patient was classified as having sustained acid hyposecretion if gastric acid secretion for the hour prior to the next dose of antisecretory drug was less than 0.2 mEq/h for .50% of all yearly assessments. Patients were classified as having acid secretion .1 mEq/h or ,1 mEq/h based on the average acid output on the yearly evaluations on gastric antisecretory drugs taken 1 hour before the next dosage of antisecretory drug. The most recent serum vitamin B12 level, complete blood count, and serum folate determination were compared in patients, based on antisecretory drug status, duration of acid secretory treatment, type of antisecretory drug, and other clinical parameters. Statistical analysis was performed using the Student’s t test for paired and unpaired values, the Mann-Whitney-Wilcoxon test and the Fisher’s exact test. We calculated the correlations (35) between serum folate and vitamin B12 levels and the duration of treatment with omeprazole. P values ,0.05 were considered significant. All continuous variables are reported as mean 6 SD.

RESULTS One hundred and thirty-one consecutive patients with Zollinger-Ellison syndrome were studied. Patients were usually middle-aged with a slight male predominance (Table 1). At the time of the serum vitamin B12 level measurements 111 patients (85%) were taking omeprazole for a mean of 4.5 years (range 0.2 to 12). Prior to omeprazole treatment the patients had been treated with H2-receptor antagonists (ranitidine, cimetidine, or famotidine) alone or with an anticholinergic agent. One hundred and twenty-eight patients had been treated at some time with H2receptor antagonists for a mean of 5.0 years (range 0.2 to 18). At the time of this study, 20 patients continued to May 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104 423

Vitamin B12 Levels in Zollinger-Ellison Syndrome/Termanini et al

Table 1. Clinical and Laboratory Characteristics of the 131 Patients Studied Characteristic

Number

Age (yrs), mean 6 SD (range) Male gender Fasting serum gastrin level (pg/mL) median (range) Basal acid output (mEq/h)* mean 6 SD (range) Maximal acid output (mEq/h)* mean 6 SD (range) Disease duration before study (yrs)† mean 6 SD (range) Multiple endocrine neoplasia-type I Metastatic liver disease‡ Previous partial gastrectomy§

54 6 11 (17–76) 83 (63%) 314 (20–154000) 27.2 6 19.9 (0–96) 55 6 31 (2–144) 13.6 6 8.1 (1–41) 23 (18%) 29 (22%) 9 (7%)

* Basal and maximal acid output are from the 121 patients without previous gastric acid-reducing surgery. Two patients did not have a basal and maximal acid output because they could not be withdrawn from antisecretory drugs. † Time of onset of continuous symptoms due to acid-peptic disease until the present study. ‡ Histologically proven metastases to the liver. § Includes a previous Billroth I or Billroth II gastric resection.

take H2-receptor antagonists for a mean of 10 years (range 3 to 17). In all patients, acid hypersecretion had been treated with some gastric acid antisecretory drug for a mean of 9.3 years (range 1 to 22). Serum vitamin B12 and folate levels varied over a wide range. The mean serum vitamin B12 level was 478 6 239 pg/mL and the mean serum folate level was 9.2 6 4.8 ng/mL. Eight patients (6%) had a low serum vitamin B12 level, 4 patients (3%) had a low serum folate level, and 1 patient had both a low serum level of vitamin B12 and folate. At the time serum vitamin levels were measured, all patients had had their acid secretion well controlled for at least 6 months with acid secretory values ,10 mEq/h (Table 2). The mean gastric acid output for the hour prior to the next antisecretory drug dose was significantly lower (P 5 0.001) in patients treated with omeprazole) than in those treated with H2-receptor antagonists (1.2 6 1.8 versus 3.0 6 2.5 mEq/h, respectively; Table 2). This level of acid control represents the highest level of acid secretion during the day and is equal to a 94% decrease in acid secretion from the pretreatment value in patients on omeprazole and a 59% decrease for patients taking H2-receptor antagonists (P ,0.001). The percentage of patients with stringent acid control (acid output ,1 mEq/h) was significantly higher in the patients treated with omeprazole compared with the patients treated with H2-receptor antagonists. Serum vitamin B12 levels were significantly (P 5 0.04) lower in patients treated with omeprazole than in patients treated with H2-receptor antagonists (Table 3; Figure 1, left panel). In contrast, there were no significant differences in the serum folate, mean corpuscular volume, he424

May 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104

matocrit, or leukocyte values between these two groups of patients (Table 3). Serum vitamin B12 levels were also significantly lower (P 5 0.001) in the patients who had sustained gastric acid hyposecretion compared with reduced gastric acid secretion (Table 3; Figure 1, right panel). There were no significant differences in serum folate, mean corpuscular volumes, hematocrit, or leukocyte values between these two groups of patients (Table 3). To compare further the effects of the extent of acid suppression on serum vitamin B12 levels, we compared the change in serum vitamin B12 levels in 68 patients who had been treated with gastric antisecretory agents for at least 5 years and who had a serum vitamin B12 level at least 5 years before the value used in Table 3. In the 21 (30%) achlorhydric patients, there was a 27% (P ,0.001) decrease in the mean serum vitamin B12 levels measured 6.3 6 1.1 years apart (Figure 2, left panel). In contrast, in the 47 (70%) patients who were not achlorhydric there was no significant change in the mean serum vitamin B12 level after an identical time (Figure 2, right panel). Furthermore, 18 of 21 (86%) of the achlorhydric patients, but only 22 of 47 (46%) of the patients without achlorhydria had a decline in serum vitamin B12 levels between the two determinations (P 5 0.002). In the 47 patients without achlorhydria, 10 patients met the criteria of sustained hypochlorhydria. There was a significant difference (P 5 0.03) in the mean change in serum vitamin B12 levels in these two groups, with a mean decrease of 59 6 131 pg/mL in the hypochlorhydric patients and a mean increase of 57 6 148 pg/mL in the 37 patients without hypochlorhydria. Serum vitamin B12 levels did not differ when patients were stratified by age, gender, disease duration, basal acid

Table 2. Comparison of Control of Gastric Acid Output in Patients with Zollinger-Ellison Syndrome Taking Omeprazole or Ranitidine Ranitidine Omeprazole n 5 20 n 5 111 Acid control (mEq/h)* Mean 6 SD Median (Range) Distribution of acid control ,1 mEq/h 1–4 mEq/h .4 mEq/h

3.0 6 2.5 2.6 (0–6.9) 6 (30%) 7 (35%) 7 (35%)

P†

1.2 6 1.8 0.5 (0–8.9) 0.001 67 (60%) 36 (32%) 8 (7%)

0.002

* The acid control value is the gastric acid output for 1 hour before the next dose of either omeprazole or ranitidine. Previous studies (39 – 41) show adequate control is ,10 mEq/h in patients without, and ,5 mEq/h in patients with, moderate to severe esophageal disease or previous gastric acid-reducing surgery. † The P value for acid control is determined using the Mann-Whitney-Wilcoxon test because of the large number of zero values. The P value for the distribution of acid control was determined using a Fisher’s exact test for three variables.

Vitamin B12 Levels in Zollinger-Ellison Syndrome/Termanini et al

Table 3. Comparison of Serum Levels of Vitamin B12 and Folate and Various Hematological Results Between Patients Being Treated with or Without Omeprazole, and Patients With or Without Achlorhydria Sustained Acid Hyposecretion Present*

Omeprazole Therapy Parameter Number of patients (%) Vitamin B12 level (pg/mL) Mean 6 SD Median (range) Serum folate level (ng/mL) Mean 6 SD Median (range) Hematological values (Mean 6 SD) Hematocrit (%) Mean corpuscular volume (fL) Leukocytes (3103/mm3)

Yes 111 (85%)

No 20 (15%)

P† —

Yes 43 (36%)

No 78 (64%)

P† —

459 6 225 408 (102–1367)

582 6 285 538 (153–1344)

0.04

405 6 203 354 (133–1199)

532 6 253 509 (102–1367)

0.001

9.2 6 5.0 7.8 (1–26)

9.5 6 4.1 8.4 (4.4–19.8)

0.55

9.6 6 4.9 9.4 (1–20)

8.8 6 4.7 7.2 (2.8–26)

0.27

40 6 5 89 6 7 6.6 6 2.3

41 6 3 90 6 6 6.6 6 1.9

0.46 0.37 0.51

39 6 5 89 6 9 6.7 6 2.3

41 6 5 89 6 6 6.5 6 2.0

0.10 0.88 0.95

* Sustained acid hyposecretion was defined as having .50% of the yearly antisecretory drug control values below ,0.2 mEq/h before the next dose of drug. Ten patients were not included because the follow-up was too short to determine the sustained acid secretory category. † P values determined using the Mann-Whitney-Wilcoxon test. Values P ,0.05 were considered significant.

output, previous history of partial gastrectomy, presence of severe gastroesophageal reflux disease, presence of liver metastases, or the need for high omeprazole maintenance doses (Table 4). Though serum vitamin B12 levels were significantly lower (P ,0.0001) in patients rendered achlorhydric for at least the last 3 years (Table 4), there was no difference in serum vitamin B12 levels between patients whose acid secretion while taking antisecretory medication averaged above or below 1 mEq/h, which represents at least an 80% decrease in basal acid secretion. Longer durations of omeprazole treatment were correlated with lower serum and vitamin B12 serum levels (r 5 20.24, P 5 0.013). There was no significant correlation between the duration of omeprazole therapy and serum folate levels (r 5 20.086, P 5 0.38; Figure 3). Eight of the 9 patients with low vitamin B12 levels had a previous serum vitamin B12 level measured prior to starting omeprazole or gastric antisecretory treatment (mean 6 years, range 2 months to 8.8 years). In each patient, this previous serum vitamin B12 level was normal (346 6 74 pg/mL, range 235 to 453 pg/mL) compared with mean values of 150 6 26 pg/mL (range 102 to 188 pg/mL) after omeprazole treatment (P 5 0.0004). None of the vitamin B12 deficient patients had anemia or peripheral neuropathy; 1 patient had macrocytosis. Vitamin B12 absorption was normal during a stage I Schilling test (using crystalline vitamin B12) in 6 of these patients. No patient with low vitamin B12 levels had intrinsic factor or parietal cell antibodies.

DISCUSSION Long-term omeprazole therapy is associated with a decline in serum vitamin B12 levels, which were below the

limit of normal in 7% of patients in this study. Several observations suggest that this association is causal. Patients treated with omeprazole had significantly lower serum vitamin B12 levels than patients treated with histamine H2 receptor antagonists. Serum vitamin B12 levels, but not folate levels, were correlated with the duration of omeprazole treatment. Prior to starting omeprazole, 8 of the patients with low vitamin B12 levels had normal levels. Omeprazole caused significantly greater acid suppression, which was the only factor that affected the serum vitamin B12 levels. Finally, there was a highly significant 30% decrease in serum vitamin B12 levels drawn 6 years apart in patients rendered achlorhydric by gastric antisecretory drugs, whereas there was no significant change in patients not rendered achlorhydric over a similar time period. Several different mechanisms may account for our findings. These include an omeprazole-induced decrease in intrinsic factor secretion; the concomitant development of pernicious anemia or atrophic gastritis with prolonged omeprazole treatment; the contribution of factors related to the Zollinger-Ellison syndrome (such as the effects of hypergastrinemia, malabsorption due to gastric acid hypersecretion, previous gastric acid-reducing surgery, the presence of metastatic disease causing cachexia, or associated with antitumor treatment); bacterial overgrowth due to drug-induced hypochlorhydria; and a direct effect of the achlorhydria on the absorption of the dietary vitamin B12. A possible role for most of these factors can be excluded by results in the present study or previous studies. Patients with Zollinger-Ellison syndrome are hypersecretors of intrinsic factor (15), and several studies demonstrate that omeprazole does not alter intrinsic factor secretion (20,36). In the present study, May 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104 425

Vitamin B12 Levels in Zollinger-Ellison Syndrome/Termanini et al

factors that are risk factors for developing vitamin B12 deficiency, such as subtotal gastrectomy (38,39), age (40,41), the presence of advanced disease causing tumorinduced cachexia, or associated with use of antitumor agents can be excluded because none of these factors affected serum vitamin B12 levels in the present study. Omeprazole-induced hypochlorhydria is the likely cause of the decreasing serum vitamin B12 levels with long-term treatment. In elderly patients with acid hyposecretion (13,15) and in patients post-subtotal gastrectomy or vagotomy (15,42), low vitamin B12 levels can occur because of malabsorption of protein-bound vitamin B12. Hypochlorhydria induces malabsorption of vitamin B12 by interfering with the acid-activated proteolytic digestion of dietary protein-bound vitamin B12 in the stomach (15,42,43) and by promoting bacterial overgrowth in the intestine (13,44). Hypochlorhydria or achlorhydria caused by omeprazole has been shown to cause the same effects, including bacterial overgrowth

Figure 1. Comparison of the serum vitamin B12 levels in patients treated or not treated with omeprazole (left panel) and patients with or without drug-induced sustained hypochlorhydria (right panel). The mean is shown by the horizontal bar, and the vertical line represents 6 SD. The dotted line represents the lower limit of normal for the assay. One hundred eleven patients were treated with omeprazole and 20 patients were treated with histamine H2-receptor antagonists. Fortythree patients had sustained acid hyposecretion and 78 patients did not.

there was no evidence that patients developed pernicious anemia. No patient developed achlorhydria when the gastric antisecretory drugs were stopped. In patients who developed abnormally low levels of serum vitamin B12, none had antiparietal or anti-intrinsic factor antibodies; Schilling tests were normal in 6 of these patients. Specific disease-related factors due to the Zollinger-Ellison syndrome can be excluded because there were no correlations between disease duration, hypergastrinemia (data not shown), or level of basal acid secretion and serum vitamin B12 levels. During the period of the study, all patients had their gastric acid hypersecretion controlled, and therefore acid-induced hypersecretion causing malabsorption (25,37) could not contribute to the decrease in serum vitamin B12 levels. Other specific disease-related 426

May 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104

Figure 2. Comparison of serum vitamin B12 levels at least 5 years apart in patients with or without drug-induced achlorhydria. The mean is shown by the horizontal bar, and the vertical line represents 6 SD. The earliest (initial) value is compared with the most recent (current) value in 68 patients. Twenty-one patients (left panel) had achlorhydria and 47 patients did not (right panel). The time between the initial and current values in the achlorhydric patients was 6.3 6 1.1 years; in the nonachlorhydric patients, it was 6.3 6 1.1 years.

Vitamin B12 Levels in Zollinger-Ellison Syndrome/Termanini et al

Table 4. Comparison of Mean (6SD) Vitamin B12 Levels in Patients with Zollinger-Ellison Syndrome, Stratified by Demographic, Clinical, and Treatment Characteristics Vitamin B12 Values (pg/mL) Characteristic*

Characteristic Present

Characteristic Absent

P

Acid control Achlorhydria for $3 years Acid control ,1 mEq/h Age .50 years Male gender Duration of disease .10 years† Basal acid output .15 mEq/h Previous partial gastrectomy Severe gastroesophageal reflux disease Liver metastases Omeprazole daily dose .40 mg/d Multiple endocrine neoplasia type-I

321 6 110 458 6 232 478 6 245 478 6 187 485 6 257 471 6 237 555 6 358 482 6 210 472 6 250 487 6 250 398 6 172

534 6 230 500 6 245 477 6 229 479 6 265 466 6 207 501 6 244 472 6 229 477 6 246 480 6 237 472 6 232 495 6 248

,0.0001 0.21 0.88 0.33 0.87 0.43 0.69 0.75 0.78 0.77 0.10

* For all variables except complete achlorhydria, data from all 131 patients were analyzed. The numbers of patients in the different groups were acid output ,1 mEq/h, 69 patients; age .50 years, 84 patients; male gender, 83 patients; duration of disease .10 years, 80 patients; basal acid output .15 mEq/h, 88 patients; previous partial gastrectomy, 9 patients; severe gastroesophageal reflux disease, 22 patients; liver metastases present, 27 patients; omeprazole daily dose .40 mg/day, 55 patients; multiple endocrine neoplasia-I present, 23 patients until the present study. For achlorhydria, complete data for 3 years were available on 76 patients, 24 of whom did not have acid present. † Duration of disease is the time from onset of continuous symptoms of acid-peptic disease as defined previously (32,44).

(13,45,46) and decreased absorption of protein-bound, but not crystalline, vitamin B12 (8,13,15,18,19), which can be reversed by giving acidic solutions with the meal (13). H2-receptor antagonists also decrease vitamin B12 absorption by a similar mechanism (47–50). However, in contrast to omeprazole or lansoprazole, which cause prolonged, profound hypochlorhydria (1,3,23), the inhibitory effect of H2-receptor antagonists is short lived and reversed by eating, therefore reducing the effects of the transient hypochlorhydria (51). The mechanism of omeprazole’s effect in causing low vitamin B12 levels in patients with Zollinger-Ellison syndrome is similar to that reported in studies of patients without Zollinger-Ellison syndrome. We found that significantly lower serum vitamin B12 levels occurred only when sustained, profound hypochlorhydria (ie, ,0.2 mEq/h) was present. Among patients with slightly higher acid secretory rates, omeprazole had no effect on serum vitamin B12 levels. A decrease in serum vitamin B12 levels was not seen with antisecretory treatment with H2-receptor antagonists, even though the mean duration of treatment was greater than 5 years and acid secretion was inhibited more than 85%. This result is consistent with the known potency of H2-receptor antagonists in patients with Zollinger-Ellison syndrome in whom even very high, frequent doses rarely produce sustained acid hyposecretion (33,52–54), or inhibit acid secretion to ,1 mEq/h for .6 to 8 hours (53,54). In contrast, the

Figure 3. Correlation of the serum vitamin B12 (left panel) and serum folate (right panel) levels with the duration of omeprazole treatment. Each point represents a single patient. Included are data from 108 patients who had both serum vitamin B12 and folate levels. May 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104 427

Vitamin B12 Levels in Zollinger-Ellison Syndrome/Termanini et al

H1-K1-ATPase inhibitors have been shown to be highly effective and long-acting in patients with Zollinger-Ellison syndrome (22,23,55,56). Omeprazole and lansoprazole have a duration of action of .3 days in patients with Zollinger-Ellison syndrome (23) and are unusually effective, frequently controlling acid secretion at similar doses as in patients with peptic ulcer disease (1,3,22,32,57), and reducing acid secretion to 1 mEq/h or less (22,32,33). In the present study, omeprazole reduced acid secretion to significantly lower levels than the H2-receptor antagonists and caused a significantly greater proportion of patients to have acid secretion ,1 mEq/h. Only sustained profound hypochlorhydria or achlorhydria, which occurred almost entirely in the omeprazole-treated patients, was associated with lower vitamin B12 levels. The results of this study raise a number of practical questions. First, should serum vitamin B12 levels be measured in all patients with Zollinger-Ellison syndrome, and if so, how often? Second, at what serum vitamin B12 level should patients with Zollinger-Ellison syndrome with decreasing serum vitamin B12 levels be started on supplemental vitamin B12? Third, should serum vitamin B12 levels be measured in patients being treated long term with H1-K1-ATPase inhibitors for gastroesophageal reflux disease? Our results show that the mean time for the average vitamin B12 level to decrease below the lower limit of normal is 18 years. This long time period is similar to that frequently seen in strict vegetarians (16) who develop decreased vitamin B12 intake but, like patients with Zollinger-Ellison syndrome, have intrinsic factor secretion and thus have intact reabsorption of vitamin B12 from the enterohepatic cycle. Hence, a longer time period is needed for vitamin B12 deficiency than the 5 to 6 years usually seen in patients with pernicious anemia (15,16). Furthermore, in a study of patients with H2-inhibitorresistant peptic disease (17), it took 3.5 years of continuous omeprazole treatment to cause a significant decrease in serum vitamin B12 levels. These results show that prolonged treatment with H1-K1-ATPase inhibitors is needed to develop low levels of vitamin B12. However, there was considerable variation among patients in time to develop low serum vitamin B12 levels in the present study. The average delay in the diagnosis of ZollingerEllison syndrome is 6 years (25) during which time malabsorption of vitamin B12 may occur (37), resulting in decreased vitamin B12 body stores. Many patients with Zollinger-Ellison syndrome gain weight when the acid hypersecretion is first discovered, probably because of prior malabsorption. This malabsorption of fat, vitamin B12 and other nutrients may explain why some patients in the present study developed vitamin B12 deficiency with only 2 years of omeprazole treatment. Except for complete achlorhydria, no clinical feature predicted who would develop low serum vitamin B12 lev428

May 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104

els, suggesting that these levels should be measured in all patients with Zollinger-Ellison syndrome who required long-term treatment with H1-K1-ATPase inhibitors. Because only 30% of patients with Zollinger-Ellison syndrome who undergo attempted curative resection are cured long term (58,59), this includes most patients. At present, vitamin B12 levels in these patients should be generally assessed yearly because of the slow rate of change in these levels. This study cannot answer the question of whether some patients with Zollinger-Ellison syndrome should be routinely treated with vitamin B12. None of the patients in the present study developed anemia or neurological complications, and only 1 patient with a low vitamin B12 level had macrocytosis, even though 6% of patients developed serum vitamin B12 levels below the lower limit of normal. These results may suggest that patients do not need to be treated with vitamin B12 until manifestations of deficiency occur. However, several recent studies would question this course. Neurologic sequelae may occur without hematologic manifestations in up to 28% of patients (60 – 62). In a recent study in elderly patients (40) with vitamin B12 deficiency documented by low serum vitamin B12 levels, elevated serum methylmalonic acid, or total homocysteine concentrations, only 41% had anemia or macrocytosis. In another recent study, only 21% of patients with proven vitamin B12 deficiency after gastric resection had macrocytosis (39). Furthermore, a recent study (63) demonstrates that elderly patients with low serum vitamin B12 levels may have impaired antibody responses to pneumococcal vaccine even in the absence of hematological manifestations. In addition, vitamin B12 deficient cells produce lower immunoglobulin levels in vitro (64) and hypogammaglobulinemia has been described in vitamin B12 deficient patients (65,66). Because patients with Zollinger-Ellison syndrome frequently require surgical procedures and may have complications from the tumor, immunological impairment from subtle vitamin B12 deficiency (67) could be an important problem. Therefore, it would seem reasonable to either recommend vitamin B12 supplementation for patients with Zollinger-Ellison syndrome with low serum vitamin B12 levels, or to measure serum methylmalonic acid or total homocysteine levels to determine if the low vitamin B12 level is associated with metabolic changes (39,40,67). Should vitamin B12 levels be measured in patients without Zollinger-Ellison syndrome treated with H1K1-ATPase inhibitors for other indications, such as gastroesophageal reflux disease? The evidence suggests that the mechanism of omeprazole-induced decreases in serum vitamin B12 levels is the same in patients with and without Zollinger-Ellison syndrome. Recent studies (6,68) report that almost all patients treated long-term with omeprazole develop hypergastrinemia, demonstrating that most have sustained hypochlorhydria or achlor-

Vitamin B12 Levels in Zollinger-Ellison Syndrome/Termanini et al

hydria. Thus, a proportion of these patients may develop low serum vitamin B12 levels. Until it is determined which patients will develop low serum vitamin B12 levels, it would be reasonable to monitor serum vitamin B12 levels periodically in patients who require long-term treatment with H1-K1-ATPase inhibitors, especially in patients whose serum vitamin B12 levels are low at the start of therapy, such as elderly patients. In conclusion, our study demonstrates that serum vitamin B12 levels, but not serum folate levels, are significantly lower in patients with Zollinger-Ellison syndrome treated long term with omeprazole. Omeprazole treatment was associated with significantly lower acid secretory levels, and low acid secretory level was the only factor found to be associated with serum vitamin B12 levels. Six percent of patients eventually developed vitamin B12 levels below normal. These data support the conclusion that serum vitamin B12 levels should be monitored in all patients with Zollinger-Ellison syndrome maintained on H1-K1-ATPase inhibitors. These results raise the important question of whether some patients without Zollinger-Ellison syndrome, such as the elderly with gastroesophageal reflux disease treated with H1-K1-ATPase inhibitors, should have serum vitamin B12 levels monitored.

REFERENCES 1. Clissold SP, Campoli-Richards DM. Omeprazole: a preliminary review of its pharmacodynamic and pharmacokinetic properties and therapeutic potential on peptic ulcer disease and Zollinger-Ellison syndrome. Drugs. 1986;32:15– 47. 2. Wilde MI, McTavish D. Omeprazole. An update of its pharmacology and therapeutic use in acid-related disorders. Drugs. 1994;48: 91–132. 3. Spencer CM, Faulds D. Lansoprazole. A reappraisal of its pharmacodynamic and pharmacokinetic properties, and its therapeutic efficacy in acid-related disorders. Drugs. 1994;48:404 – 430. 4. Kahrilas PJ, Hogan WJ. Gastroesophageal reflux disease. In: Sleisenger MH, Fordtran JS, eds. Gastrointestinal Disease. 5th ed. Philadelphia, Pa: WB Saunders;1993;378 – 401. 5. Hetzel DJ, Dent J, Reed WD, et al. Healing and relapse of severe peptic esophagitis after treatment with omeprazole. Gastroenterology. 1988;95:903–912. 6. Jansen JB, Klinkenberg-Knol EC, Meuwissen SG, et al. Effect of long-term treatment with omeprazole on serum gastrin and serum group A and C pepsinogens in patients with reflux esophagitis. Gastroenterology. 1990;99:621– 628. 7. Lamberts R, Creutzfeldt W, Struber HG, et al. Long-term omeprazole therapy in peptic ulcer disease: gastrin, endocrine cell growth, and gastritis. Gastroenterology. 1993;104:1356 –1370. 8. Dutta SK. Vitamin B12 malabsorption and omeprazole therapy. J Am Coll Nutr. 1994;13:544 –545. 9. Maton P. Risks of hypergastrinemia. Dig Dis Sci. 1995;13:1–2. 10. Maton PN, Dayal Y. Clinical implications of hypergastrinemia. In: Zakim D, Dannenberg AJ, eds. Peptic Ulcer Disease and Other Acidrelated Disorders. Armonk, NY: Academic Research Associates; 1991;213–246. 11. Ekman L, Hansson E, Havu N, et al. Toxicological studies on omeprazole. Scand J Gastroenterol. 1985;20 (suppl 108):53– 69.

12. Koop H, Bachem MG. Serum iron, ferritin, and vitamin B12 during prolonged omeprazole therapy. J Clin Gastroenterol. 1992;14:288 – 292. 13. Saltzman JR, Kemp JA, Golner BB, et al. Effect of hypochlorhydria due to omeprazole treatment or atrophic gastritis on proteinbound vitamin B12 absorption. J Am Coll Nutr. 1994;13:584 –591. 14. Seetharam B. Gastrointestinal absorption and transport of cobalamin (vitamin B12). In: Johnson LR, Alpers DH, Christensen J, et al, eds. Physiology of the Gastrointestinal Tract. 3rd ed. New York: Raven Press; 1994:1997–2026. 15. Festen HP. Intrinsic factor secretion and cobalamin absorption. Physiology and pathophysiology in the gastrointestinal tract. Scand J Gastroenterol. 1991;26:1–7. 16. Babior BM. Metabolic aspects of folic acid and cobalamin. In: Beutler E, Lichtman MA, Coller BS, Kipps TJ, eds. William’s Hematology. 5th ed. New York: McGraw-Hill; 1996:380 –393. 17. Koop H. Metabolic consequences of long-term inhibition of acid secretion by omeprazole. Aliment Pharmacol Ther. 1992;6:399 – 406. 18. Marcuard SP, Albernaz L, Khazanie PG. Omeprazole therapy causes malabsorption of cyanocobalamin (Vitamin B12). Ann Intern Med. 1994;120:211–215. 19. Festen H, Tertoolen J. Malabsorption of protein-bound but not unbound cobalamin during treatment with omeprazole. Gastroenterology. 1988;94:A125. Abstract. 20. Kittang E, Aadland E, Schjonsby H, Rohss K. The effect of omeprazole on gastric acidity and the absorption of liver cobalamins. Scand J Gastroenterol. 1987;22:156 –160. 21. Maton PN, Vinayek R, Frucht H, et al. Long-term efficacy and safety of omeprazole in patients with Zollinger-Ellison syndrome: a prospective study. Gastroenterology. 1989;97:827– 836. 22. Metz DC, Strader DB, Orbuch M, et al. Use of omeprazole in Zollinger-Ellison: a prospective nine-year study of efficacy and safety. Aliment Pharmacol Ther. 1993;7:597– 610. 23. Metz DC, Jensen RT. Advances in gastric antisecretory therapy in Zollinger-Ellison syndrome. In: Mignon M, Jensen RT, eds. Endocrine Tumors of the Pancreas: Recent Advances in Research and Management. Frontiers of Gastrointestinal Research. Basel, Switzerland: S Karger; 1995:240 –257. 24. Jensen RT, Fraker DL. Zollinger-Ellison syndrome: advances in treatment of the gastric hypersecretion and the gastrinoma. JAMA. 1994;271:1–7. 25. Jensen RT, Gardner JD. Gastrinoma. In: Go VLW, DiMagno EP, Gardner JD, et al, eds. The Pancreas: Biology, Pathobiology and Disease. 2nd ed. New York, NY: Raven Press; 1993:931–978. 26. Fishbeyn VA, Norton JA, Benya RV, et al. Assessment and prediction of long-term cure in patients with Zollinger-Ellison syndrome: the best approach. Ann Intern Med. 1993;119:199 –206. 27. Frucht H, Howard JM, Slaff JI, et al. Secretin and calcium provocative tests in the Zollinger-Ellison syndrome: a prospective study. Ann Intern Med. 1989;111:713–722. 28. Metz DC, Pisegna JR, Fishbeyn VA, et al. Control of gastric acid hypersecretion in the management of patients with Zollinger-Ellison syndrome. World J Surg. 1993;17:468 – 480. 29. Orbuch M, Doppman JL, Strader DB, et al. Imaging for pancreatic endocrine tumor localization: recent advances. In: Mignon M, Jensen RT, eds. Endocrine Tumors of the Pancreas: Recent Advances in Research and Management. Frontiers of Gastrointestinal Research. Basel, Switzerland: S Karger; 1995:268 –281. 30. Cherner JA, Doppman JL, Norton JA, et al. Selective venous sampling for gastrin to localize gastrinomas. A prospective study. Ann Intern Med. 1986;105:841– 847. 31. Benya RV, Metz DC, Venzon DJ, et al. Zollinger-Ellison syndrome can be the initial endocrine manifestation in patients with multiple endocrine neoplasia-type 1. Am J Med. 1994;97:436 – 444. May 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104 429

Vitamin B12 Levels in Zollinger-Ellison Syndrome/Termanini et al 32. Metz DC, Pisegna JR, Fishbeyn VA, et al. Currently used doses of omeprazole in Zollinger-Ellison syndrome are too high. Gastroenterology. 1992;103:1498 –1508. 33. Miller LS, Vinayek R, Frucht H, et al. Reflux esophagitis in patients with Zollinger-Ellison syndrome. Gastroenterology. 1990;98:341– 346. 34. Weber HC, Venzon DJ, Lin JT, et al. Determinants of metastatic rate and survival in patients with Zollinger-Ellison syndrome: a prospective long-term study. Gastroenterology. 1995;108:1637– 1649. 35. Dawson-Saunders B, Trapp RG. Basic and Clinical Biostatistics. Norwalk, Conn: Appleton & Lange; 1990:1–329. 36. Kittang E, Aadlang E, Schonsby H. Effect of omeprazole on the secretion of intrinsic factor, gastric acid and pepsin in man. Gut. 1985;26:594 –598. 37. Shimoda SS, Saunders DR, Rubin CE. The Zollinger-Ellison syndrome with steatorrhea. II. The mechanism of fat and vitamin B 12 malabsorption. Gastroenterology. 1968;55(6):705–723. 38. Shafer RB, Ripley D, Swaim WR, et al. Hematologic alterations following partial gastrectomy. Am J Med Sci. 1973;266:240 –248. 39. Sumner AE, Chin MM, Abrahm JL, et al. Elevated methylmalonic acid and total homocysteine levels show high prevalence of vitamin B12 deficiency after gastric surgery. Ann Intern Med. 1996;124:469 – 476. 40. Lindenbaum J, Rosenberg IH, Wilson PW, et al. Prevalence of cobalamin deficiency in the Framingham elderly population. Am J Clin Nutr. 1994;60:2–11. 41. Joosten E, van den Berg A, Riezler R, et al. Metabolic evidence that deficiencies of vitamin B-12 (cobalamin), folate, and vitamin B-6 occur commonly in elderly people. Am J Clin Nutr. 1993;58:468 – 476. 42. Doscherholmen A, Swaim WR. Impaired assimilation of egg CO57 vitamin B12 in patients with hypochlorhydria and achlorhydria and after gastric resection. Gastroenterology. 1973;64:913–919. 43. King CE, Leibach J, Tosekes PP. Clinically significant vitamin B12 deficiency secondary to malabsorption of protein-bound vitamin B12. Dig Dis Sci. 1979;24:397– 402. 44. Suter P, Golner B, Goldin B, et al. Reversal of protein-bound vitamin B12 malabsorption with antibiotics in atrophic gastritis. Gastroenterology. 1991;101:1039 –1045. 45. Saltzman JR, Kowdley KV, Pedrosa MC, et al. Bacterial overgrowth without clinical malabsorption in elderly hypochlorhydric subjects. Gastroenterology. 1994;106:615– 623. 46. Thorens J, Froehlich F, Schwizer W, et al. Bacterial overgrowth during treatment with omeprazole compared with cimetidine: a prospective randomised double blind study. Gut. 1996;39:54 –59. 47. Streeter AM, Goulston KJ, Bathur FA, et al. Cimetidine and malabsorption of cobalamin. Dig Dis Sci. 1982;27:13–16. 48. Steinberg WM, King CE, Toskes PP. Malabsorption of proteinbound cobalamin but not unbound cobalamin during cimetidine administration. Dig Dis Sci. 1980;25:188 –191. 49. Salom IL, Silvis SE, Doscherholmen A. Effect of cimetidine on the absorption of vitamin B12. Scand J Gastroenterol. 1982;17:129 –131. 50. Belaiche J, Cattan D, Zittoun J, et al. Effect of ranitidine on cobalamin absorption. Dig Dis Sci. 1983;28:667– 668. Letter. 51. Lanzon-Miller S, Pounder RE, Hamilton MR, et al. Twenty-four-

430

May 1998

THE AMERICAN JOURNAL OF MEDICINEt

Volume 104

52. 53.

54.

55.

56. 57.

58.

59. 60.

61.

62.

63.

64.

65.

66.

67. 68.

hour intragastric acidity and plasma gastrin concentration before and during treatment with either ranitidine or omeprazole. Aliment Pharmacol Ther. 1987;1:239 –251. Vinayek R, Howard JM, Maton PN, et al. Famotidine in the therapy of gastric hypersecretory states. Am J Med. 1986;81:49 –59. Jensen RT, Collen MJ, McArthur KE, et al. Comparison of the effectiveness of ranitidine and cimetidine in inhibiting acid secretion in patients with gastric hypersecretory states. In: Misiewicz JJ, Wood JR, eds. Ranitidine: Therapeutic Advances. Amsterdam, Netherlands: Excerpta Medica; 1984:168 –198. McArthur KE, Raufman JP, Seaman JJ, et al. Cimetidine pharmacokinetics in patients with Zollinger-Ellison syndrome. Gastroenterology. 1987;93:69 –76. Bardram L, Stadil F. Effects of omeprazole on acid secretion and acid related symptoms in patients with Zollinger-Ellison syndrome. Scand J Gastroenterol. 1989;166 (suppl):95–100. Frucht H, Maton PN, Jensen RT. Use of omeprazole in patients with the Zollinger-Ellison syndrome. Dig Dis Sci. 1991;36:394 – 404. Jensen RT, Metz DC, Koviack PD, Feigenbaum KM. Prospective study of the long-term efficacy and safety of lansoprazole in patients with Zollinger-Ellison syndrome. Aliment Pharmacol Ther. 1993;7 (Suppl 1):41–50. Norton JA, Doppman JL, Jensen RT. Curative resection in Zollinger-Ellison syndrome: results of a 10-year prospective study. Ann Surg. 1992;215:8 –18. Jensen RT. Gastrin-producing tumors. Cancer Treat Res. 1997;89: 293–334. Lindenbaum J, Healton EB, Savage DG, et al. Neuropsychiatric disorders caused by cobalamin deficiency in the absence of anemia or macrocytosis. NEJM. 1988;318:1720 –1728. Carmel R. Pernicious anemia. The expected findings of very low serum cobalamin levels, anemia, and macrocytosis are often lacking. Arch Intern Med. 1988;148:1712–1714. Carmel R, Sinow RM, Karnaze DS. Atypical cobalamin deficiency. Subtle biochemical evidence of deficiency is commonly demonstrable in patients without megaloblastic anemia and is often associated with protein-bound cobalamin malabsorption. J Lab Clin Med. 1987;109:454 – 463. Fata FT, Herzlich BC, Schiffman G, Ast AL. Impaired antibody responses to pneumococcal polysaccharide in elderly patients with low serum vitamin B12 levels. Ann Intern Med. 1996;124:299 –304. Hall CA, Green-Colligan PD, Begley JA. The role of cobalamin in synthesis of protein and immunoglobulins by human peripheral lymphocytes. Nutr Res. 1981;1:349 –361. Van Dommelen CK, Slagboom G, Meester GT, Wadman SK. Reversible hypogammaglobulinaemia in cyanocobalamin (B12) deficiency. Acta Med Scand. 1963;174:193–200. Kafetz K. Immunoglobulin deficiency responding to vitamin B12 in two elderly patients with megaloblastic anaemia. Postgrad Med J. 1985;61:1065–1066. Carmel R. Subtle cobalamin deficiency. Ann Intern Med. 1996;124: 338 –340. Lamberts R, Creutzfeldt W, Stockmann F, et al. Long term omeprazole treatment in man: effects on gastric endocrine cell populations. Digestion. 1988;39:126 –135.