Br. J. Surg. Vol. 66 (1979) 489-492
Absorption from the jejunum is increased by forward and backward pacing* JACK C OLLI N, K E I T H A. KELLY A N D S I D N E Y F. P H I L L I P S t SUMMARY
In 4 dogs a 75-cm segment of jejunum was isolated from the intestinal stream. Stimulating electrodes were implanted at each end of the segment and recording electrodes at intervals between. The two ends of the segment were fashioned as stomas on the anterior abdominal wall. After recovery, the electrical activity of the segment was continuously recorded, whilst a solution of 100 mmolll glucose and 90 mmolll NaCl was infused into the proximal stoma at 2.9mllmin. Efluent from the distal stoma was collected in 5-min aliquots for three consecutive 30-min periods. In the second period electrical stimuli were given via either the proximal or the distal bipolar electrodes to entrain pacesetter potentials. Both forward and backward electrical pacing decreased the output of water, glucose and sodium from the jejunal segment, but the decrease was significantly greater with backward pacing. We conclude that pacing a segment of jejunum backwards with electrical stimuli produces a greater increase in absorption from the segment than pacing it forwards, a finding which has therapeutic implications in patients with the short bowel syndrome.
PATIENTS who have undergone resection of large portions of small intestine have rapid intestinal transit and suffer from malabsorption, steatorrhea and excessive faecal output of water and electrolytes (Winawer et al., 1966). Treatment of this ‘short bowel syndrome’ is difficult and often unsuccessful. An effective, safe method of slowing transit through the small bowel and enhancing absorption would be of considerable benefit. An attractive hypothesis is that intestinal transit can be slowed and intestinal absorption increased by pacing the pacesetter potentials of the small intestine with electrical stimuli. The pacesetter potentials are omnipresent, regularly recurring, cyclical changes in electrical potential found in the small intestine of both the dog and man (Sarna, 1975). The cycles are generated by a pacemaker located within the first centimetre of the duodenum (Hermon-Taylor and Code, 1971), from where they propagate distally to the remaining small bowel. As they sweep distally, the pacesetter potentials phase the onset of a second type of electrical activity, called action potentials (Sarna, 19751, which in turn triggers intestinal contractions. In this way, the pacesetter potentials set the pattern of intestinal contractions and regulate the transit of content through the small bowel. Bunker et al. (1967) have shown that the frequency of the pacesetter potentials is identical in the proximal 10-30 per cent of the canine small intestine, while from the mid-jejunum to the terminal ileum the frequency gradually declines. Along the proximal plateau of identical frequency electrical stimuli can entrain pacesetter potentials, increase their frequency and even reverse their direction of propagation (Akwari et al., 1975). Such backward pacing, in turn, reverses 40
the direction of propagation of intestinal contractions along the plateau and slows the transit of content through the paced segment (Kelly and Code, 1977). Pacesetter potentials along the distal frequency gradient can also be entrained if the distal bowel is separated from the duodenal pacemaker by intestinal transection. We have previously shown (Collin et al., 1977) that backward electrical pacing of an isolated jejunal segment results in prolongation of transit and increased absorption from the segment. The aim of the present study was to determine whether a similar increase in absorption could be produced by forward electrical pacing and t o compare changes in absorption with forward and backward pacing. Methods Four female mongrel dogs of 13-17 kg were used in the study. In each dog, under pentothal anaesthesia (30mg/kg) and using aseptic operating technique, a 75-cm segment of jejunum located 50-125cm from the pylorus was isolated, and the anatomical continuity of the remaining bowel re-established by end-to-end anastomosis. A bipolar stimulating Ag-AgC1 electrode was sewn to the serosal surface of the isolated segment 7.5 cm from each end and monopolar recording electrodes were attached at 10-cm intervals between (Fig. 1). The electrodes were connected by insulated copper wires to a multipinned socket mounted within a stainless-steel cannula which was fixed to the abdominal wall. The two ends of the segment were fashioned as separate stomas on the anterior abdominal wall. The dogs were allowed a minimum of 7 days to recover from the operation.
I1
Electrodes
Fig. 1. Preparation of an isolated 75-cm segment of proximal jejunum.
* Based on a communication t o the Surgical Research Society, January 1979. t Departments of Surgery and Gastroenterology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota. Correspondence to: Keith A. Kelly, Department of Surgery.
Jack Collin et al.
490
PP Proximal
No
frequency 0.22 Hr
pacing
4
Distal
Pacing
at
0.27
4
4
4
4
4
4
4
4
S1
Stimuli
4
4
Hz
4
4
4
4
4 . 4
10 s
Fig. 2. Electrical pacing at both proximal and distal sites of stimulation (S) increases the frequency of the pacesetter potentials (PP), but the PP propagate distally during proximal pacing and proximally during distal pacing. E, recording electrode; dotted lines show direction of PP propagation; arrows, stimulus artefact. T*
8
$-/
ta",:
6
Pacing Before
.t F ?
2
SE
During
After
4
2
0
30
60 90
0
30
60 90
B
F Minutes
Fig. 3. The effects of forward and backward electrical pacing on 111svolume output from the distal stoma of the jejunal segments. F, forward pacing; B, backward pacing. * Mean differs from that during pacing. Comparison of F and B, P < 0.001.
In each experiment 50 g of raw liver were fed to the fasted, conscious dog to induce a fed pattern of intestinal electrical activity and to abolish the interdigestive myoelectric complex (Mroz and Kelly, 1977). A 16-French Foley catheter with a 5-ml balloon was passed into the oral end of the segment through its proximal stoma and the balloon inflated in order t o prevent loss of fluid from the proximal stoma during the experiment. The segment was next gently irrigated with 100 ml
of isotonic sodium chloride to clear the lumen of debris. A solution containing 100 mmol/l glucose and 90 mmol/l sodium chloride (osmolality, 280 mosmol/l, pH 5.3) was then continuously infused through the catheter into the oral end of the segment at 2.9 ml/min. Thirty minutes of infusion were allowed for steady state conditions to be established, after which emuent from the distal stoma was collected in 5-min aliquots for three consecutive 30-min periods. The electrical activity of the jejunal segment was recorded throughout the experiment using a Gould Brush recorder, model 260, with the AC amplifiers set at a time constant of 1 s. In six experiments in each dog in the second 30-min period electrical stimuli (frequency 0.27 Hz, duration 50 ms, xnplitude 4 mA) were given via the proximal bipolar electrode $0 drive the isolated segment forward. In a further six experiments in each dog electrical stimuli were given via the distal bipolar electrode t o drive the segment backward. Stimuli were generated by a model 800 digital pulse generator (WP Instruments). The volume of each 5-min aliquot of effluent was recorded and the glucose, sodiumand potassium concentrations measured. The mean volume and output of glucose, sodium and potassium per 5 min were calculated for each 30-min period. Statistical analyses were carried out using the paired t test. Differences between the means of the three periods were analysed separately for the two groups of experiments. The mean outputs with forward and with backward pacing were then compared. In 18 experiments involving forward pacing and in 18 experiments involving backward pacing a bolus of 14C polyethylene glycol (PEG, mol. wt 4000) was injected in the period before pacing and a further bolus during pacing. The first bolus had passed completely through the loop by the time the second bolus was given. The time taken for 20 per cent excretion of each bolus of I4C PEG was determined and used as a measure of transit. N
Electrical pacing and absorption from jejunum Results
0.75~
In all experiments electrical stimulation at the proximal or the distal bipolar electrode during the second 30-min period consistently entrained the pacesetter potentials of the jejunal segment. Before pacing, the rhythm of the pacesetter potentials of the segment was irregular, their frequency variable (range 0.220.26 Hz, mean k s.e. mean 0.24 f 0.02 Hz) and the direction of their propagation inconsistent (Fig. 2). However, during pacing the rhythm of the pacesetter potentials was regular, their frequency was steady and increased to that of the stimulation frequency (0.27 Hz) and their direction of propagation was consistent. When pacing at the proximal site, pacesetter potentials were propagated in a distal direction-from the site of stimulation to the aboral end of the segment. However, when pacing at the distal site, the pacesetter potentials were propagated in the reverse directionfrom the site of stimulation to the orad end of the segment. The volume of output from the distal stoma during the 30 min of forward pacing ( ~ e a 2n s.e. mean = 6.0 k 0.5 m1/5 min) was slightly decreased compared to the 30 min before pacing (7-4 +_ 0.4 m1/5 min) (P< 0401). Glucose output during forward pacing (0.48 0.004 mmol/5 min) was also slightly decreased (P
Absorption from the jejunum is increased by forward and backward pacing* JACK C OLLI N, K E I T H A. KELLY A N D S I D N E Y F. P H I L L I P S t SUMMARY
In 4 dogs a 75-cm segment of jejunum was isolated from the intestinal stream. Stimulating electrodes were implanted at each end of the segment and recording electrodes at intervals between. The two ends of the segment were fashioned as stomas on the anterior abdominal wall. After recovery, the electrical activity of the segment was continuously recorded, whilst a solution of 100 mmolll glucose and 90 mmolll NaCl was infused into the proximal stoma at 2.9mllmin. Efluent from the distal stoma was collected in 5-min aliquots for three consecutive 30-min periods. In the second period electrical stimuli were given via either the proximal or the distal bipolar electrodes to entrain pacesetter potentials. Both forward and backward electrical pacing decreased the output of water, glucose and sodium from the jejunal segment, but the decrease was significantly greater with backward pacing. We conclude that pacing a segment of jejunum backwards with electrical stimuli produces a greater increase in absorption from the segment than pacing it forwards, a finding which has therapeutic implications in patients with the short bowel syndrome.
PATIENTS who have undergone resection of large portions of small intestine have rapid intestinal transit and suffer from malabsorption, steatorrhea and excessive faecal output of water and electrolytes (Winawer et al., 1966). Treatment of this ‘short bowel syndrome’ is difficult and often unsuccessful. An effective, safe method of slowing transit through the small bowel and enhancing absorption would be of considerable benefit. An attractive hypothesis is that intestinal transit can be slowed and intestinal absorption increased by pacing the pacesetter potentials of the small intestine with electrical stimuli. The pacesetter potentials are omnipresent, regularly recurring, cyclical changes in electrical potential found in the small intestine of both the dog and man (Sarna, 1975). The cycles are generated by a pacemaker located within the first centimetre of the duodenum (Hermon-Taylor and Code, 1971), from where they propagate distally to the remaining small bowel. As they sweep distally, the pacesetter potentials phase the onset of a second type of electrical activity, called action potentials (Sarna, 19751, which in turn triggers intestinal contractions. In this way, the pacesetter potentials set the pattern of intestinal contractions and regulate the transit of content through the small bowel. Bunker et al. (1967) have shown that the frequency of the pacesetter potentials is identical in the proximal 10-30 per cent of the canine small intestine, while from the mid-jejunum to the terminal ileum the frequency gradually declines. Along the proximal plateau of identical frequency electrical stimuli can entrain pacesetter potentials, increase their frequency and even reverse their direction of propagation (Akwari et al., 1975). Such backward pacing, in turn, reverses 40
the direction of propagation of intestinal contractions along the plateau and slows the transit of content through the paced segment (Kelly and Code, 1977). Pacesetter potentials along the distal frequency gradient can also be entrained if the distal bowel is separated from the duodenal pacemaker by intestinal transection. We have previously shown (Collin et al., 1977) that backward electrical pacing of an isolated jejunal segment results in prolongation of transit and increased absorption from the segment. The aim of the present study was to determine whether a similar increase in absorption could be produced by forward electrical pacing and t o compare changes in absorption with forward and backward pacing. Methods Four female mongrel dogs of 13-17 kg were used in the study. In each dog, under pentothal anaesthesia (30mg/kg) and using aseptic operating technique, a 75-cm segment of jejunum located 50-125cm from the pylorus was isolated, and the anatomical continuity of the remaining bowel re-established by end-to-end anastomosis. A bipolar stimulating Ag-AgC1 electrode was sewn to the serosal surface of the isolated segment 7.5 cm from each end and monopolar recording electrodes were attached at 10-cm intervals between (Fig. 1). The electrodes were connected by insulated copper wires to a multipinned socket mounted within a stainless-steel cannula which was fixed to the abdominal wall. The two ends of the segment were fashioned as separate stomas on the anterior abdominal wall. The dogs were allowed a minimum of 7 days to recover from the operation.
I1
Electrodes
Fig. 1. Preparation of an isolated 75-cm segment of proximal jejunum.
* Based on a communication t o the Surgical Research Society, January 1979. t Departments of Surgery and Gastroenterology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota. Correspondence to: Keith A. Kelly, Department of Surgery.
Jack Collin et al.
490
PP Proximal
No
frequency 0.22 Hr
pacing
4
Distal
Pacing
at
0.27
4
4
4
4
4
4
4
4
S1
Stimuli
4
4
Hz
4
4
4
4
4 . 4
10 s
Fig. 2. Electrical pacing at both proximal and distal sites of stimulation (S) increases the frequency of the pacesetter potentials (PP), but the PP propagate distally during proximal pacing and proximally during distal pacing. E, recording electrode; dotted lines show direction of PP propagation; arrows, stimulus artefact. T*
8
$-/
ta",:
6
Pacing Before
.t F ?
2
SE
During
After
4
2
0
30
60 90
0
30
60 90
B
F Minutes
Fig. 3. The effects of forward and backward electrical pacing on 111svolume output from the distal stoma of the jejunal segments. F, forward pacing; B, backward pacing. * Mean differs from that during pacing. Comparison of F and B, P < 0.001.
In each experiment 50 g of raw liver were fed to the fasted, conscious dog to induce a fed pattern of intestinal electrical activity and to abolish the interdigestive myoelectric complex (Mroz and Kelly, 1977). A 16-French Foley catheter with a 5-ml balloon was passed into the oral end of the segment through its proximal stoma and the balloon inflated in order t o prevent loss of fluid from the proximal stoma during the experiment. The segment was next gently irrigated with 100 ml
of isotonic sodium chloride to clear the lumen of debris. A solution containing 100 mmol/l glucose and 90 mmol/l sodium chloride (osmolality, 280 mosmol/l, pH 5.3) was then continuously infused through the catheter into the oral end of the segment at 2.9 ml/min. Thirty minutes of infusion were allowed for steady state conditions to be established, after which emuent from the distal stoma was collected in 5-min aliquots for three consecutive 30-min periods. The electrical activity of the jejunal segment was recorded throughout the experiment using a Gould Brush recorder, model 260, with the AC amplifiers set at a time constant of 1 s. In six experiments in each dog in the second 30-min period electrical stimuli (frequency 0.27 Hz, duration 50 ms, xnplitude 4 mA) were given via the proximal bipolar electrode $0 drive the isolated segment forward. In a further six experiments in each dog electrical stimuli were given via the distal bipolar electrode t o drive the segment backward. Stimuli were generated by a model 800 digital pulse generator (WP Instruments). The volume of each 5-min aliquot of effluent was recorded and the glucose, sodiumand potassium concentrations measured. The mean volume and output of glucose, sodium and potassium per 5 min were calculated for each 30-min period. Statistical analyses were carried out using the paired t test. Differences between the means of the three periods were analysed separately for the two groups of experiments. The mean outputs with forward and with backward pacing were then compared. In 18 experiments involving forward pacing and in 18 experiments involving backward pacing a bolus of 14C polyethylene glycol (PEG, mol. wt 4000) was injected in the period before pacing and a further bolus during pacing. The first bolus had passed completely through the loop by the time the second bolus was given. The time taken for 20 per cent excretion of each bolus of I4C PEG was determined and used as a measure of transit. N
Electrical pacing and absorption from jejunum Results
0.75~
In all experiments electrical stimulation at the proximal or the distal bipolar electrode during the second 30-min period consistently entrained the pacesetter potentials of the jejunal segment. Before pacing, the rhythm of the pacesetter potentials of the segment was irregular, their frequency variable (range 0.220.26 Hz, mean k s.e. mean 0.24 f 0.02 Hz) and the direction of their propagation inconsistent (Fig. 2). However, during pacing the rhythm of the pacesetter potentials was regular, their frequency was steady and increased to that of the stimulation frequency (0.27 Hz) and their direction of propagation was consistent. When pacing at the proximal site, pacesetter potentials were propagated in a distal direction-from the site of stimulation to the aboral end of the segment. However, when pacing at the distal site, the pacesetter potentials were propagated in the reverse directionfrom the site of stimulation to the orad end of the segment. The volume of output from the distal stoma during the 30 min of forward pacing ( ~ e a 2n s.e. mean = 6.0 k 0.5 m1/5 min) was slightly decreased compared to the 30 min before pacing (7-4 +_ 0.4 m1/5 min) (P< 0401). Glucose output during forward pacing (0.48 0.004 mmol/5 min) was also slightly decreased (P
