Comparison of an Intravenous Bolus of Famotidine and Mylanta II for the Control of

Gastric pH in Critically I11 Patients Paul Wilson, FRCSEd, Geoffrey W. B. Clark, FRCSEd, Marco Anselmino, MD, Neil T. Welch, FRCSEd,

Swarnjit Singh, MO, Galen Perdikis, MD, Ronald A. Hinder, MD, FACS, Omaha, Nebraska

The effects of the intravenous bolus administration of famotidine venus the administration of Mylanta II liquid every 2 hours on the pH of the gastric an- trum, body, and fundus for 24 hours were com- pared in 10 critically ill patients admitted to the in- teusive care unit with isolated cranial trauma. Patients received 30 mL of Mylanta II every 2 hours via nasogastric tube for 24 hours, followed by administration of 20 mg of intravenous bolus famo- tidine every 12 hours for the subsequent 24-hour period, pH of the gastric antrmn, body, and fundus was monitored continuously using a three antimony pH electrode/nasogastrie tube assembly. Gastric pH data were analyzed for the percentage of time pH was less than 4 and median pH for the antrum, body, and fundus for each 24-hour period. The percentage of time pH was less than 4 was signifi- cantly less in the antrum and body of the stomach during famotidine therapy (8.9% 4- 3.6% and 24.9% 4- 6.9%, respectively) compared with My- lanta II (39.1% 4- 6.7% and 57.6% 4- 8.5%, re- spectively, both p <0 .005 ) , but was not significant- ly different in the fundus (famotidine: 25.3% 4- 7.8%; Mylanta H: 28.3% 4- 6.5%). Median gastric pH for 24 hours was significantly greater in the an- trum and body of the stomach during famotidine therapy (7.8 4- 0.2 and 6.8 4- 0.6, respectively) compared with Mylanta II (4.5 4- 0.6 and 3.7 4- 0.9, respectively, p <0.005 and p <0.01, respec- tively), but was not significantly different in the fundus (famotidine: 5.9 4- 0.8; Mylanta II: 5.4 4- O.7).

The data indicate that an intravenous bolus of famotidine every 12 hours is more effective than Mylanta II liquid every 2 hours administered via a nasogastrie tube in maintaining gastric pH above 4 in critically ill patients. Famotidine produces a uni- form increase in gastric pH throughout the stom- ach, whereas Mylanta II controls only proximal gas- trie pH, probably related to fundie pooling of antacid in the supine position.

From the Department of Surgery, Creighton University, Omaha, Ne- braska.

Requests for reprints should be addressed to Ronald A. Hinder, MD, FACS, Department of Surgery, Suite 3740, Creighton University, 601 North 30th Street, Omaha, Nebraska 68131.

Presented at the 45th Annual Meeting of the Southwestern Surgi- cal Congress, Monterey, California, April 18-21, 1993.

T he maintenance of intragastric pH above 4 is con- sidered necessary in preventing acute gastreduode-

nal stress ulceration in critically ill patients [1-9]. Hourly administration of antacids via a nasogastric tube, which is a popular form of prophylaxis therapy, may cause side effects such as vomiting, diarrhea, alkalosis, and hyper- magnesemia in patients with renal insufficiency. This technique is very labor intensive; therefore, H2-receptor antagonist therapy is becoming widely employed in inten- sive care units and has been shown to be as effective as regular antacid administration in the prevention of stress lesions [1,3,4,6]. Although it has been demonstrated that continuous infusion of ranitidine is superior to bolus ad- ministration in maintaining intragastric pH above 4 [10], a recent report reveals that 75% of patients in intensive care units are treated with bolus H2-receptor antagonists [11]. Famotidine, which has a duration of action of 10 to 12 hours, is reported to have fewer side effects and drug interactions than ranitidine or cimetidine. The aim of this study was therefore to evaluate the effects of an intrave- nous bolus of famotidine every 12 hours on the pH of the gastric antrum, body, and fundus in critically ill patients in comparison with standard stress ulcer prophylaxis us- ing 30-mL doses of Mylanta II liquid every 2 hours given via a nasogastric tube.

PATIENTS AND METHODS Ten consecutive patients (8 men and 2 women, mean

age: 29 years, range: 20 to 62 years) admitted to the intensive care unit of St. Joseph Hospital, Omaha, Ne- braska, were entered into the study after informed con- sent had been obtained from a close relative, All patients were unconscious following isolated cranial trauma and were recruited within 24 hours of admission to the hospi- tal. Exclusion criteria were: (1) upper digestive tract hemorrhage on admission, (2) previous treatment with H2 blockers or omeprazole, (3) previous upper digestive tract surgery, or (4) concomitant intrathoracic, intra- abdominal, or pelvic trauma.

Patients received 30 mL of Mylanta II liquid (alumi- num hydroxide 400 mg/5 mL, magnesium hydroxide 400 mg/5 mL, simethicone 40 mg/5 mL, Johnson & John- son, Merck Consumer Pharmaceuticals, Washington, PA) every 2 hours (acid neutralizing capacity: 150 mEq in 30 mL) via a nasogastric tube during the initial 24- hour period. The nasogastric tube was otherwise clamped. This was substituted by an 20 mg intravenous bolus of famotidine every 12 hours, with the nasogastric tube left on continuous free drainage over the subsequent 24-hour period.

THE AMERICAN JOURNAL OF SURGERY VOLUME 166 DECEMBER 1993 621

WILSON ET AL

8 ?

5 4 3 2 1

17:eo

M M M M M M M M M M M M M M

21:80 81:88 85:88 89188 13:118 17:08

[ Famot~dine Famot~dine

?

6 5

4

3 2 !

14:88 10:08 22:08 82:88 86108 18:~ 14:88

Figure 1. Top. Typical 24-hour gastric pH profile for the antrum (thin line), body of stomach (thick line), and fun- dus (dotted Une) In a critically ill pa- tient receiving Mylanta II liquid (M) 30 mL every 2 hours via a nasogastric tube. Bottom. Subsequent 24-hour gastric pH profile in the same patient, during treatment with famotidine 20 mg every 12 hours as an intravenous bolus.

% time pHc4 70

60

50

40

30

20

10

0 ANTRUM

P<O.O05

m Famotidine

BODY FUNDUS

Figure 2. Mean (-I-SEM) percentage of time gastric pH was less than 4 in the antrum, body, and fundus of the 10 critically ill patients treated with 30 mL of Mylanta II every 2 hours for 24 hours, followed by 20 mg of famotidine in an intravenous bolus every 12 hours for 24 hours (p values: Wilcoxon signed-rank test for pairs).

Patients received no oral or nasogastric feeds during the study period. Concomitant administration of other drugs with known antacid effects such as antacids, sucral- fate, or omeprazole was not permitted.

Intragastric oH monitoring: The pH of the gastric antrum, body, and fundus were determined by three anti- mony electrodes (8 cm spacing between electrodes) con- nected to the external surface of a 16F diameter Salem sump nasogastric tube (Zinetics Medical, Salt Lake City, UT), which was positioned fluoroscopically so that the antimony electrodes lay 5, 13, and 21 cm below the car- dioesophageal junction. The position of the cardioesopha- geal junction was determined by the pH change between the esophagus and the stomach. Electrodes were calibrat-

ed with pH 7.0 and pH 1.0 buffers before and after each study. A difference of less than 0.3 pH units between the two readings was considered acceptable. The antimony and cutaneous reference electrodes were connected to portable digitrappers (Synectics, Irving, Texas), which store data every 4 seconds. Monitoring was performed for a minimum of 21 hours in each 24-hour period, after which data was transferred to an IBM-AT personal com- puter for analysis using appropriate software (Gastrosoft, Synectics, Irving, TX).

Data analysis: Data were analyzed for the percentage of time pH was less than 4 and median gastric pH for the 24 hours for each of the three probe positions and are expressed as mean 4- SEM. The Wilcoxon signed-rank test for paired data was used to compare successive 24- hour treatment periods. Fisher's exact test was used to compare the proportion of patients in each group with median gastric pH of less than 4. The Mann-Whitney U- test was used to compare gastric pH gradients. A p-value of <0.05 was considered significant.

RESULTS Figure 1 shows a typical patient's gastric pH profile

for the antrum, body, and fundus for the successive 24- hour treatment periods. Administration of Mylanta II produced only short-lived elevations of gastric pH in the antrum, body, and fundus, whereas famotidine produced a consistent elevation of pH in all stomach compart- ments. There was a significant difference in the percent- age of time pH was less than 4 between successive treat- ment periods for the antrum (39.1% 4- 6.7% for Mylanta II versus 8.9% 4- 3.6% for famotidine) and body of the stomach (57.6% 4- 8.5% for Mylanta II versus 24.9% 4- 6.9% for famotidine) (both p <0.005, Figure 2). There was no difference in the percentage of time pH was less than 4 in the fundus (28.3% 4- 6.5% for Mylanta II versus 25.3% 4- 7.8% for famotidine). Similarly, there was a significant difference in the median pH for successive

622 THE AMERICAN JOURNAL OF SURGERY VOLUME 166 DECEMBER 1993

CONTROL OF GASTRIC pH

treatment periods in the antrum (4.5 4- 0.6 for Mylanta II versus 7.8 4- 0.2 for famotidine) and body of the stomach (3.7 4- 0.9 for Mylanta II versus 6.8 4- 0.6 for famotidine) (p <0.005 and p <0.01, respectively, Figure 3), but no difference in the median pH was recorded in the fundus (5.4 4- 0.7 for Mylanta II versus 5.9 4- 0.8 for famoti- dine). A significant pH gradient existed between the an- trum and fundus during famotidine therapy (p <0.05), which was not present during Mylanta II therapy (Figure 3). A significantly greater number of patients had a medi- an pH above 4 in the antrum and body of stomach during famotidine therapy compared with Mylanta II therapy (Table I). The 24-hour median pH in the fundus was above 4 in 80% of patients during both famotidine and Mylanta II therapy.

No drug-associated side effects were observed during the study. Upper digestive tract hemorrhage did not oc- cur in any of the patients.

COMMENTS Acute gastroduodenal stress lesions commonly occur

in critically ill patients. Such lesions, which may vary from mucosal erythema to frank ulceration, have been reported in up to 100% of patients in some series [12]. These lesions may be clinically silent or manifest as mas- sive gastrointestinal hemorrhage or perforation [13,14]. A number of factors are involved in the pathogenesis of stress lesions, of which intragastric hydrogen ion concen- tration and mucosal ischemia are major contributors. Mucosal ischemia, which weakens the mucosal defense mechanisms, may also cause hydrogen ion back-diffusion with the development of acute mucosal lesions [13,14]. The control of intragastric hydrogen ion concentration, by raising gastric pH to values above 4, is thought to be an important measure in the prophylaxis of stress lesions in critically ill patients [1-9,13,15].

Famotidine was chosen as the H2 blocker for this study since ranitidine and cimetidine have a duration of action of up to 9 hours and three bolus injections may be required over 24 hours. The longer acting H2 blocker famotidine requires only bolus injections every 12 hours, has been shown to be as effective as ranitidine in healing duodenal ulcers [16,17], and is superior to ranitidine in maintaining gastric pH above 4 [I8,19]. In the preven- tion of postoperative stress ulceration, famotidine has been shown to be superior to cimetidine [20].

In this study, we found that famotidine administered every 12 hours as a 20-mg bolus injection was effective at maintaining the gastric pH of the antrum, body, and fundus above 4 for more than 75% of the time in critically ill patients, maintained median gastric pH above 6 in the antrum and body of the stomach, and was superior to Mylanta II in the control of gastric pH. During famoti- dine therapy, a significantly greater number of patients had a median 24-hour gastric antral and body pH greater than 4 compared with that observed during Mylanta II treatment.

A significant pH gradient was observed between the antrum and fundus during famotidine therapy, with the pH being highest in the gastric antrum. This was not

Median pH

PcO.O05 8

7

6

5

4

3

2

1

0

IBm Mylanta n BIB Famot|dJna

P<0.01

BODY FUNDUS

i ANTRUM

Figure 3. Mean (4-SEM) median 24-hour gastric pH in the antrum, body. and fundus in the 10 critically ill patients treated with Mylanta II followed by famotidine (p values: Wilcoxon signed-rank test for pairs, NS = not significant).

TABLE 1 Percentage of Patients With Median 24-Hour Gastric pH

of Greater Than 4

Mylanta II (%) Famotidine (%) p Value*

Antrum 40 100 < 0.05 Body 30 90 < 0.05 Fundus 80 80 NS

NS = not significant. *p Value was calculated using Fisher's exact test.

apparent during Mylanta II therapy. Mylanta II exerted its greatest influence on pH in the fundus of the stomach, which may be related to pooling of the liquid antacid in the fundus of recumbent patients.

This study shows that measurement of fundal pH alone, 5 cm below the cardioesophageal junction, which is a common site for positioning during gastric pH studies, may be misleading in the assessment of the global gastric pH response to antacids in stress ulcer prophylaxis.

Famotidine given every 12 hours as a 20-mg intrave- nous bolus is effective at controlling gastric pH in critical- ly ill patients and is more effective than Mylanta II given as a bolus every 2 hours. Famotidine is very effective at maintaining the gastric pH above 4 and, therefore, should be useful in prophylaxis against stress ulceration in criti- cally ill patients.

REFERENCES 1. Weigelt JA, Aurbakken CM, Gewertz BL, Snyder WH. Cimeti- dine versus antacid in prophylaxis for stress ulceration. Arch Surg 1981; 116: 597-601. 2. Halloran GL, Zfass AM, Gayle WE, Wheeler CB, Miller JD. Prevention of acute gastrointestinal complications after severe head injury: a controlled trial of cimetidine prophylaxis. Am J Surg 1980; 139: 44-8. 3. Stothert JC, Simonowitz DA, Dellinger EP. Randomized pro-

THE AMERICAN JOURNAL OF SURGERY VOLUME 166 DECEMBER 1993 623

WILSON ET AL

spective evaluation of cimetidine and antacid control of gastric pH in the critically ill. Ann Surg 1980; 192: 169-74. 4. Basso N, Bagarani M, Materia A, Fiorani S, Lunardi P, Speran- za V. Cimctidine and antacid prophylaxis of acute upper gastroin- testinal bleeding in high risk patients. Am J Surg 1981; 141: 339-41. 5. Herrmann V, Kaminski DL. Evaluation of intragastfic pH in acutely ill patients. Arch Surg 1979; 114: 511-4. 6. Poleski MH, Spanier AH. Cimetidine versus antacids in the prevention of stress erosions in critically ill patients. Am J Gas- troenterol 1986; 81: 107-11. 7. More DG, Raper RF, Munro IA, Watson C J, Boutagy JS, Shenfield GM. Randomized, prospective trial of cimetidine and ranitidine for control of intragastric pH in the critically ill. Surgery 1985; 97: 215-23. 8. Borrero E, Morgolis IB, Bank S, Shulman N, Chardavoyne R. Antacid versus sucralfate in preventing acute gastrointestinal bleeding. A randomized trial in 100 critically ill patients. Am J Surg 1984; 148: 809-12. 9. Peura DA, Johnson LF. Cimetidine for prevention and treatment of gastroduodenal mucosal lesions in patients in an intensive care unit. Ann [ntern Med 1985; 103: 173-7. 10. Fiorucei S, Clausi GC, Farinelli M, Santucci L, Pelli MA, Morelli A. Intragastric pH monitoring during antisccretory thera- py in patients with gastrointestinal bleeding. Am J Gastrocnterol 1989; 84: 1416-20. 11. Ballesteros MA, Hogan DL, Koss MA, Isenberg JI. Bolus or intravenous infusion of ranitidine: effects on gastric pH and acid secretion. Ann Intern Med 1990; 112: 334-9. 12. Lucas CE, Sugawa C, Friend W, Waly AJ. Therapeutic impli- cations of disturbed gastric physiology in patients with stress ulcer- ations. Am J Surg 1972; 123: 25-33. 13. Hinder RA. Stress ulceration: importance of blood flow and acid-base balance. In: Nyhus LM, editor. Surgery annual. Nor- walk: Appleton-Century-Crofts, 1986; 18: 95-110. 14. Fiddian-Groen RG, McGough E, Pittenger G, Rothman E. Predictive value of intramural pH and other risk factors for massive bleeding from stress ulceration. Gastroenterology 1983; 85: 613-20. 15. Stress ulcer prophylaxis in critically ill patients [editorial]. Lancet 1989; 2: 1255-6. 16. Bianchi Porro G, Lazzaroni M, Barbara L, et al. Famotidine versus ranitidine h.s. in acute duodenal ulcer. A multicentre endo- scopic trial. Ital J Gastrcenterol 1991; 23: 65-9. 17. al-Freihi HM, al-Hamdan A, al-Qurain A, et al. A single nocturnal dose of famotidine for the treatment of duodenal ulcer: evaluation of efficacy and safety. Trop Gastroenterol 1991; 12: 77-82. 18. Savarino V. Continuous intragastric pH monitoring: a real progress in the assessment of antisecretory drugs. Ital J Gastroen- terol 1990; 22 (Suppl 2): 20-3. 19. Fullarton GM, Macdonald AM, Mann SG, McColl KE. Con- trolled study of the effects of intravenous famotidine on intragastric pH in bleeding peptic ulcers. Aliment Pharmacol Ther 1991; 5: 77-84. 20. Lamothe PH, Rao E, Serra A J, et al. Comparative efficacy of r famotidine, ranitidine, and mylanta in postoperative stress ulcers. Gastric pH control and ulcer prevention in patients undergoing coronary artery bypass graft surgery. Gastroenterology 1991; 6: 515-20.

DISCUSSION Frederick A, Moore (Denver, CO): Dr. Wilson and

his colleagues have shown that famotidine administered at 12-hour intervals is more effective than antacids used

every 2 hours in controlling gastric pH. From this obser- vation, they conclude that famotidine should be useful as a prophylaxis against stress ulceration. I would like to first comment on two methodologic issues. You have chosen to do this study in patients with head injuries. In contradistinction to other types of critically ill patients, brain-injured patients are known to be hypersecretors of gastric acid. Can the observations made in these patients be extrapolated to the generic intensive care unit patient? The second issue is that, in your study protocol, antacids were always administered for the first 24 hours and then famotidine for the second 24 hours. The basic assumption is that gastric acid secretion is the same throughout this initial 48-hour admission to the hospital. Is this true? In regard to the clinical relevance of this study, do you think stress gastritis is a significant problem in intensive care units today? Clearly, in the late 1960s and early 1970s, severe hemorrhage from stress gastritis was common in surgical intensive care units. Today, it is rare that some- one actually dies as a result of this problem. We assume this is because of aggressive prophylaxis, but an equally tenable explanation is that the natural history of stress gastritis has changed. Do we really need to have a stron- ger prophylactic agent? To put it in another way, is it necessary that the gastric pH be consistently maintained at greater than 47 If we use your data, the answer is "no." You have shown that antacids failed in achieving this goal. Yet antacids have been proven to be effective as a prophylactic agent against stress gastritis. Likewise, bo- lus cimetidine does not consistently maintain the pH above 4. Yet, this has also been proven to be effective prophylaxis. Do you think a gastric pH consistently above 4 could be potentially harmful? I 'm sure you know that recent trials comparing sucralfate to antacids in stress ulcer prophylaxis has shown an association between high gastric pH and the occurrence of nosocomial pneumo- nias.

David Felleiano (Atlanta, GA): I have a methodolo- gic question. Why did you choose the dosing intervals with Mylanta II when there is reasonably good endoscop- ic evidence that the antacids disappear within 20 to 30 minutes in intensive care unit patients? Secondly, how did you decide on your modest dose of Mylanta II? Did you think about modifying the dose in order to keep the pH below 4 just to see how much you would need, or did you feel that would be repetitious with the studies that John Skillman did many years ago? I think that the Mylanta II group is at a disadvantage in the way the study was designed.

Paul Wilson (closing): Regarding the question re- garding head injured patients having gastric acid hyper- secretion, we feel that these data can be extrapolated to other intensive care unit patients. Not only gastric pH, but other factors such as mucosal ischemia may play a role. As to whether gastric acid secretion changed over the study period, [ agree that this is a problem with the study design. I do not know if gastric acid secretion should vary from the first 24 hours to the second 24 hours; however, we were forced to always use famotidine after Mylanta II. Is stress gastritis a significant problem to-

624 THE AMERICAN JOURNAL OF SURGERY VOLUME 166 DECEMBER 1993

CONTROL OF GASTRIC pH

day? Should we be aggressive in prophylaxis? It still occurs despite prophylaxis and is significantly decreased in patients receiving prophylaxis. Is it necessary to main- tain the gastric pH above 4? There are some concerns about this since it may increase the risk of nosocomial lung sepsis. This is, however, not as great a problem as

was once imagined. In regard to the disappearance of antacids in 20 minutes and the dosing of Mylanta II, we chose this dose because it is the standard treatment dose in the intensive care unit in our hospital, It is the dose used in several other studies. We wanted to compare famofi- dine with a standard antacid treatment.

THE AMERICAN JOURNAL OF SURGERY VOLUME 166 DECEMBER 1993 625