Clinical Science and Molecular Medicine (1975) 49, 5 19-521.
SHORT CDMMUNICATI0N
The effect of frusemide on water and electrolyte absorption from the human jejunum J. F. MAcKENZIE, K. M. C O C H R A N A N D R. I. RUSSELL Gastroenterology Unit,Royal Infirmary, Glasgow, Scotland
(Received 10 July 1975)
Summary 1. Frusemide, in a dose of 120 nmol (40 mg) administered intravenously, significantly reduced the absorption of water and electrolytes from the human jejunum, a double-lumen perfusion system with proximal occluding balloons being used. Net secretion of water and electrolytes occurred in some subjects. 2. No significant change in water and electrolyte absorption was observed with 60 nmol (20 mg) of frusemide. 3. These findings may explain the diarrhoea which may be induced by frusemide in some patients.
ethacrynic acid has been shown to inhibit water absorption by hamster gut sacs, secretion occurring with higher dosage (Binder, Katz, Spencer & Spiro, 1966). A reduction of sodium absorption has also been noted in isolated rabbit ileum after mucosal application of ethacrynic acid (Al-Awquati, Field & Greenough, 1974). If frusemide were to alter the water absorption mechanism in this way in the human jejunum, a possible explanation of frusemideinduced diarrhoea would be found.
Methods and subjects The study was performed by the technique of smallintestinal perfusion with a double-lumen tube with proximal occluding balloons and a mixing segment 30 cm long. Immediately proximal to this were two occluding balloons and above this was a further collection point for removal of the fluid collecting above the occluding balloons. The tube was passed under X-ray control, with an image intensifier, until the balloons were at the ligament of Treitz. Lack of contamination by the perfusate of fluid aspirated above the balloons was demonstrated in the ensuing 10-15 min by measuring marker concentration. The perfusing solution contained sodium (140 mmolll), potassium (5 mmolll), chloride (100 mmoll l), bicarbonate (45 mmolll) and glucose (11.2 mmol/l). The osmolality was 300 mosmol/l. Phenol Red (150 mgll) was used as a non-absorbable, water-soluble marker. In some subjects, I4Clabelled polyethylene glycol 4000 (1 gll) was used as a second marker.
Key words: electrolyte absorption, frusemide, smallintestinal perfusion, water absorption.
Introduction
The administration of frusemide is occasionally associated with diarrhoea as an unwanted side effect (Cooperman & Rubin, 1973; Kennedy, 1974). As frusemide is considered to be an inhibitor of renal transport mechanisms (Knox, Wright, Howards & Berliner, 1969), it is possible that intestinal transport mechanisms may also be altered by the drug. The principal action of frusemide on the kidney is similar to that of ethacrynic acid, namely inhibition of sodium reabsorption in the ascending loop of Henle (Cannon & Kilcoyne, 1969). Amongst other actions, Correspondence: Dr R. I. Russell, GastroenterologyUnit, Royal Infirmary, Glasgow G4 OSF, Scotland.
519
J. F. MacKenzie, K . M . Cochran and R. I. Russell
520
The solution was infused at a rate of 15 ml/min at 37°C. An equilibration period of 50 or 60 rnin was allowed to develop 'steady-state' conditions. Three separate collections (10 rnin each) were taken before injection of frusemide and then for the ensuing 50 or 60 min. The study was performed on twelve normal volunteer subjects, who gave informed consent for the procedure to be carried out. Five subjects received 60 nmol of frusemide and seven subjects received 120 nmol of frusemide. Small-intestinal transmucosal electrical potential difference was measured in some subjects. The technique used was that described by Cochran, MacKenzie & Russell (1975). For this study, however, the probe electrode, containing saturated potassium chloride in agar, was incorporated into the double-lumen perfusion system with the distal end of the probe at the mid-point of the test segment. Throughout the investigation EPD'') was recorded continuously. Water, sodium, potassium and chloride absorption was calculated from the change in concentration of the non-absorbable marker, Phenol Red, and polyethylene glycol. The statistical significance of the results was assessed by the paired t-test. Results
There was no significant change in the absorption of (I)
Abbreviation: EPD, electrical potential difference.
water and electrolytes from :he test segment after intravenous injection of 60 nmol of frusemide. A diuresis occurred in all five subjects. The mean net water and electrolyte absorption after 120 nmol of frusemide is shown in Table 1. The reduction in net absorption of water is statistically significant at 20, 30,40 and 50 min. The reduction in net sodium absorption is significant at 30,40 and 50 rnin and the reduction in net potassium absorption significant at 40 and 50 min. There was a net secretion of chloride in some subjects and a rather low absorption (in relation to the sodium absorption) in the remainder-the effect of frusemide in a dose of 120 nmol is to increase the net secretion significantly at 40 and 50 min. The mean urinary water excretion ( 2 SEM) before and after 120 nmol of frusemide was 137.6 k28.5 ml/h and 1378.6 f 176-1 ml/h respectively, and sodium excretion before and after the same dose of frusemide was 10.9 _+ 3.1 mmol/h and 156-3 f 21a2 mmol/h respectively. The change in absorption after frusemide was statistically significant whether polyethylene glycol or the Phenol Red data were used to calculate the results. It was noted, however, that the water absorption calculated from polyethylene glycol data was consistently less than when Phenol Red data were used. The mean EPD for the three subjects studied, recorded during the perfusion, was -2.35 mV (range+ 1.43 to -4.8 mV). After the administration of frusemide, the mean EPD was -2.84 mV (range - 1.22 to -4.3 mV).
TABLE 1. Water and electrolyte absorption before and after 120 nmol of frusemide
Mean values k SEM are shown. Significance o f results: * P< 0.05; ** P< 0.005. After frusernide Before frusemide
10 min
20 niin
40 rnin
30 min
50 min
60 rnin
~
Mean sodium absorption (mmol/h) Mean potassium absorption (rnrnol/h) Mean chloride absorption (rnmol/h) Mean water absorption (rnmol/h)
14.80
k 4.83 0.73 f 0.23
-0.92
k 4.63 140.41 f39.09
11.13 25.17 0.53 f0.23 - 1.73 4.43 1 1 1.39 2 40.5
*
9.06
2.32*
2 5.55
k7 5 1
0.49 2 0.27
0.37 0.35 - 5.89 k 5.42 56.2* f52.03
- 3.63
2 4.72 94.87* 242.19
1.89* *-7.57 0.16** f0.36
-9.42* k 6.24
*
18.17** 53.53
- 7.35** 5 8.25 007* f0.37 - 12.45* k 7.37 15.50**
k 61.91
- 1.72 f8.19 0.37 f0.32
- 8.14
2 6.07
-
Water and electrolyte absorption
521
Discussion
Acknowledgments
It appears from this study that frusemide alters transport mechanisms in the human jejunum but the mechanism is not clear. It has been suggested that sodiumlpotassium-dependentadenosine triphosphatase may play a role in the active transport of sodium across the gut mucosa (Katz & Epstein, 1968), and both frusemide and ethacrynic acid are potent inhibitors of this enzyme system in oivo and in vitro (Knox et al., 1969; Duggan & Noll, 1965). It is possible that this effect may be responsible for inhibition of sodium transport by frusemide. Measurement of water and electrolyte absorption in the jejunum by perfusion is a well-established technique (Schedl & Clifton, 1963; Phillips & Summerskill, 1966) and particularly lends itself to the present study in which each subject acted as his or her own control. Although we found a good correlation between the absorptive changes calculated from 14C-labelled polyethylene glycol results and from those with Phenol Red, it was noted that these specific figures obtained were consistently lower with the former results. This may be due to falsely high readings of absorbance obtained for Phenol Red solutions in the presence of mucoproteins (MacGregor & Meyer, 1974). The recordings of the small-intestinal EPD measured during the basal rate in our subjects were similar to those found by Geall (1970). Loss of sodium into the bowel lumen and a normal EPD have been shown to occur in coeliac disease (Russell, Allan, Gerskowitch & Robertson, 1972; R. I. Russell & K. M. Cochran, unpublished work). Similar EPD values have been reported in the ileum of patients with cholera, where there is also sodium loss. In these two conditions the sodium loss is thought to be accompanied by an equivalent chloride loss; thus a change in EPD does not occur. Our results in this study suggest therefore that sodium loss after frusemide is accompanied by an equivalent loss of anions. These results are of importance in relation to the mechanism of action of frusemide and its tendency to cause diarrhoea when given in high dosage.
The authors thank Professor A. C. Kennedy, who drew our attention to the problem, Dr H. J. Dargie for clinical help, and Miss J. Toms and Staff Nurse A. Brown for technical and nursing assistance.
References AL-AwQuAn,Q., FIELD, M. & GREENOUGH, W.B., 3rd (1974) Reversal of cyclic AMP-mediated intestinal secretion by ethacrynic acid. Journal of Clinical Investigation, 53, 687692. BINDER,H.J., KATZ, L.A., SPENCER, R.P. & SPIRO,H.M. (1966) The effect of inhibitors of renal transport on the small intestine. Journal of Clinical Investigation, 45, 18541858. CANNON, P.J. & KILCOYNE, M.M. (1969) Ethacrynic acid and frusemide: renal pharmacology and clinical use. Progress in Cardiovascular Disease, 12, 99-1 18. COCHRAN, K.M., MACKENZLE, J.F. & RUSSELL, R.I. (1975) Role of taurocholic acid in production of gastric mucosal damage after ingestion of aspirin. British Medical Journal, i, 183-185. COOPERMAN, L.B. & RUBIN,L. (1973) Toxicity of ethacrynic acid and frusemide. American Heart Journal, 85, 831-834. DUGGAN, D.E. & NOLL,R.M. (1965) Effects of ethacrynic acid and cardiac glycosides on membrane ATPase of renal cortex. Archives of Biochemistry, 109, 388-391. GEALL,M. (1970) The T-mural electrical potential of the human gastrointestinal tract. M.D. thesis, University of London. KATZ, A.I. & EPSTEIN,F.N. (1968) Physiologic role of sodium/potassium activated adenosine triphosphate in the transport of cations across biologic membranes. New England Journal of Medicine, 278, 253-261. KENNEDY, A.C. (1974) Diuretics and renal failure. Tenth Symposium on Advanced Medicine, Royal College of Physicians of London, pp. 380-387. Ed. by Ledingham, J.G.G. Pitman, London. KNOX,F.G., WRIGHT,F.S., HOWARDS, S.S. & BERLINER, R.W. (1969) Effect of frusernide on sodium reabsorption by proximal tubule of the dog. American Journal of Physiob g y , 217, 192-198. MACGREGOR, I.L. & MEYER,J.H. (1974) Non-absorbable indicators; the effect of protein on phenol red and polyethylene glycol determination. American Journal of Digestive Diseases, 19, 361-365. PHILLIPS, S.F. & SUMMERSKILL, W.H.J. (1966) Occlusion of the jejunum for intestinal perfusion in man. Proceedings of the Mayo Clinic, 51, 224-23 1. RUSSELL, R.I., ALLAN,J.G., GERSKOWITCH, V.P. & ROBERTSON, J.W.K. (1972) A study by perfusion techniques of the absorption abnormalities in the jejunum in adult coeliac disease. Clinical Science, 42, 735-741. W.& CLIFTON,J.A. (1963) Solute and water absorpSCHEDL, tion by the human small intestine. Nature (London), 199, 1264-1 267.
SHORT CDMMUNICATI0N
The effect of frusemide on water and electrolyte absorption from the human jejunum J. F. MAcKENZIE, K. M. C O C H R A N A N D R. I. RUSSELL Gastroenterology Unit,Royal Infirmary, Glasgow, Scotland
(Received 10 July 1975)
Summary 1. Frusemide, in a dose of 120 nmol (40 mg) administered intravenously, significantly reduced the absorption of water and electrolytes from the human jejunum, a double-lumen perfusion system with proximal occluding balloons being used. Net secretion of water and electrolytes occurred in some subjects. 2. No significant change in water and electrolyte absorption was observed with 60 nmol (20 mg) of frusemide. 3. These findings may explain the diarrhoea which may be induced by frusemide in some patients.
ethacrynic acid has been shown to inhibit water absorption by hamster gut sacs, secretion occurring with higher dosage (Binder, Katz, Spencer & Spiro, 1966). A reduction of sodium absorption has also been noted in isolated rabbit ileum after mucosal application of ethacrynic acid (Al-Awquati, Field & Greenough, 1974). If frusemide were to alter the water absorption mechanism in this way in the human jejunum, a possible explanation of frusemideinduced diarrhoea would be found.
Methods and subjects The study was performed by the technique of smallintestinal perfusion with a double-lumen tube with proximal occluding balloons and a mixing segment 30 cm long. Immediately proximal to this were two occluding balloons and above this was a further collection point for removal of the fluid collecting above the occluding balloons. The tube was passed under X-ray control, with an image intensifier, until the balloons were at the ligament of Treitz. Lack of contamination by the perfusate of fluid aspirated above the balloons was demonstrated in the ensuing 10-15 min by measuring marker concentration. The perfusing solution contained sodium (140 mmolll), potassium (5 mmolll), chloride (100 mmoll l), bicarbonate (45 mmolll) and glucose (11.2 mmol/l). The osmolality was 300 mosmol/l. Phenol Red (150 mgll) was used as a non-absorbable, water-soluble marker. In some subjects, I4Clabelled polyethylene glycol 4000 (1 gll) was used as a second marker.
Key words: electrolyte absorption, frusemide, smallintestinal perfusion, water absorption.
Introduction
The administration of frusemide is occasionally associated with diarrhoea as an unwanted side effect (Cooperman & Rubin, 1973; Kennedy, 1974). As frusemide is considered to be an inhibitor of renal transport mechanisms (Knox, Wright, Howards & Berliner, 1969), it is possible that intestinal transport mechanisms may also be altered by the drug. The principal action of frusemide on the kidney is similar to that of ethacrynic acid, namely inhibition of sodium reabsorption in the ascending loop of Henle (Cannon & Kilcoyne, 1969). Amongst other actions, Correspondence: Dr R. I. Russell, GastroenterologyUnit, Royal Infirmary, Glasgow G4 OSF, Scotland.
519
J. F. MacKenzie, K . M . Cochran and R. I. Russell
520
The solution was infused at a rate of 15 ml/min at 37°C. An equilibration period of 50 or 60 rnin was allowed to develop 'steady-state' conditions. Three separate collections (10 rnin each) were taken before injection of frusemide and then for the ensuing 50 or 60 min. The study was performed on twelve normal volunteer subjects, who gave informed consent for the procedure to be carried out. Five subjects received 60 nmol of frusemide and seven subjects received 120 nmol of frusemide. Small-intestinal transmucosal electrical potential difference was measured in some subjects. The technique used was that described by Cochran, MacKenzie & Russell (1975). For this study, however, the probe electrode, containing saturated potassium chloride in agar, was incorporated into the double-lumen perfusion system with the distal end of the probe at the mid-point of the test segment. Throughout the investigation EPD'') was recorded continuously. Water, sodium, potassium and chloride absorption was calculated from the change in concentration of the non-absorbable marker, Phenol Red, and polyethylene glycol. The statistical significance of the results was assessed by the paired t-test. Results
There was no significant change in the absorption of (I)
Abbreviation: EPD, electrical potential difference.
water and electrolytes from :he test segment after intravenous injection of 60 nmol of frusemide. A diuresis occurred in all five subjects. The mean net water and electrolyte absorption after 120 nmol of frusemide is shown in Table 1. The reduction in net absorption of water is statistically significant at 20, 30,40 and 50 min. The reduction in net sodium absorption is significant at 30,40 and 50 rnin and the reduction in net potassium absorption significant at 40 and 50 min. There was a net secretion of chloride in some subjects and a rather low absorption (in relation to the sodium absorption) in the remainder-the effect of frusemide in a dose of 120 nmol is to increase the net secretion significantly at 40 and 50 min. The mean urinary water excretion ( 2 SEM) before and after 120 nmol of frusemide was 137.6 k28.5 ml/h and 1378.6 f 176-1 ml/h respectively, and sodium excretion before and after the same dose of frusemide was 10.9 _+ 3.1 mmol/h and 156-3 f 21a2 mmol/h respectively. The change in absorption after frusemide was statistically significant whether polyethylene glycol or the Phenol Red data were used to calculate the results. It was noted, however, that the water absorption calculated from polyethylene glycol data was consistently less than when Phenol Red data were used. The mean EPD for the three subjects studied, recorded during the perfusion, was -2.35 mV (range+ 1.43 to -4.8 mV). After the administration of frusemide, the mean EPD was -2.84 mV (range - 1.22 to -4.3 mV).
TABLE 1. Water and electrolyte absorption before and after 120 nmol of frusemide
Mean values k SEM are shown. Significance o f results: * P< 0.05; ** P< 0.005. After frusernide Before frusemide
10 min
20 niin
40 rnin
30 min
50 min
60 rnin
~
Mean sodium absorption (mmol/h) Mean potassium absorption (rnrnol/h) Mean chloride absorption (rnmol/h) Mean water absorption (rnmol/h)
14.80
k 4.83 0.73 f 0.23
-0.92
k 4.63 140.41 f39.09
11.13 25.17 0.53 f0.23 - 1.73 4.43 1 1 1.39 2 40.5
*
9.06
2.32*
2 5.55
k7 5 1
0.49 2 0.27
0.37 0.35 - 5.89 k 5.42 56.2* f52.03
- 3.63
2 4.72 94.87* 242.19
1.89* *-7.57 0.16** f0.36
-9.42* k 6.24
*
18.17** 53.53
- 7.35** 5 8.25 007* f0.37 - 12.45* k 7.37 15.50**
k 61.91
- 1.72 f8.19 0.37 f0.32
- 8.14
2 6.07
-
Water and electrolyte absorption
521
Discussion
Acknowledgments
It appears from this study that frusemide alters transport mechanisms in the human jejunum but the mechanism is not clear. It has been suggested that sodiumlpotassium-dependentadenosine triphosphatase may play a role in the active transport of sodium across the gut mucosa (Katz & Epstein, 1968), and both frusemide and ethacrynic acid are potent inhibitors of this enzyme system in oivo and in vitro (Knox et al., 1969; Duggan & Noll, 1965). It is possible that this effect may be responsible for inhibition of sodium transport by frusemide. Measurement of water and electrolyte absorption in the jejunum by perfusion is a well-established technique (Schedl & Clifton, 1963; Phillips & Summerskill, 1966) and particularly lends itself to the present study in which each subject acted as his or her own control. Although we found a good correlation between the absorptive changes calculated from 14C-labelled polyethylene glycol results and from those with Phenol Red, it was noted that these specific figures obtained were consistently lower with the former results. This may be due to falsely high readings of absorbance obtained for Phenol Red solutions in the presence of mucoproteins (MacGregor & Meyer, 1974). The recordings of the small-intestinal EPD measured during the basal rate in our subjects were similar to those found by Geall (1970). Loss of sodium into the bowel lumen and a normal EPD have been shown to occur in coeliac disease (Russell, Allan, Gerskowitch & Robertson, 1972; R. I. Russell & K. M. Cochran, unpublished work). Similar EPD values have been reported in the ileum of patients with cholera, where there is also sodium loss. In these two conditions the sodium loss is thought to be accompanied by an equivalent chloride loss; thus a change in EPD does not occur. Our results in this study suggest therefore that sodium loss after frusemide is accompanied by an equivalent loss of anions. These results are of importance in relation to the mechanism of action of frusemide and its tendency to cause diarrhoea when given in high dosage.
The authors thank Professor A. C. Kennedy, who drew our attention to the problem, Dr H. J. Dargie for clinical help, and Miss J. Toms and Staff Nurse A. Brown for technical and nursing assistance.
References AL-AwQuAn,Q., FIELD, M. & GREENOUGH, W.B., 3rd (1974) Reversal of cyclic AMP-mediated intestinal secretion by ethacrynic acid. Journal of Clinical Investigation, 53, 687692. BINDER,H.J., KATZ, L.A., SPENCER, R.P. & SPIRO,H.M. (1966) The effect of inhibitors of renal transport on the small intestine. Journal of Clinical Investigation, 45, 18541858. CANNON, P.J. & KILCOYNE, M.M. (1969) Ethacrynic acid and frusemide: renal pharmacology and clinical use. Progress in Cardiovascular Disease, 12, 99-1 18. COCHRAN, K.M., MACKENZLE, J.F. & RUSSELL, R.I. (1975) Role of taurocholic acid in production of gastric mucosal damage after ingestion of aspirin. British Medical Journal, i, 183-185. COOPERMAN, L.B. & RUBIN,L. (1973) Toxicity of ethacrynic acid and frusemide. American Heart Journal, 85, 831-834. DUGGAN, D.E. & NOLL,R.M. (1965) Effects of ethacrynic acid and cardiac glycosides on membrane ATPase of renal cortex. Archives of Biochemistry, 109, 388-391. GEALL,M. (1970) The T-mural electrical potential of the human gastrointestinal tract. M.D. thesis, University of London. KATZ, A.I. & EPSTEIN,F.N. (1968) Physiologic role of sodium/potassium activated adenosine triphosphate in the transport of cations across biologic membranes. New England Journal of Medicine, 278, 253-261. KENNEDY, A.C. (1974) Diuretics and renal failure. Tenth Symposium on Advanced Medicine, Royal College of Physicians of London, pp. 380-387. Ed. by Ledingham, J.G.G. Pitman, London. KNOX,F.G., WRIGHT,F.S., HOWARDS, S.S. & BERLINER, R.W. (1969) Effect of frusernide on sodium reabsorption by proximal tubule of the dog. American Journal of Physiob g y , 217, 192-198. MACGREGOR, I.L. & MEYER,J.H. (1974) Non-absorbable indicators; the effect of protein on phenol red and polyethylene glycol determination. American Journal of Digestive Diseases, 19, 361-365. PHILLIPS, S.F. & SUMMERSKILL, W.H.J. (1966) Occlusion of the jejunum for intestinal perfusion in man. Proceedings of the Mayo Clinic, 51, 224-23 1. RUSSELL, R.I., ALLAN,J.G., GERSKOWITCH, V.P. & ROBERTSON, J.W.K. (1972) A study by perfusion techniques of the absorption abnormalities in the jejunum in adult coeliac disease. Clinical Science, 42, 735-741. W.& CLIFTON,J.A. (1963) Solute and water absorpSCHEDL, tion by the human small intestine. Nature (London), 199, 1264-1 267.
