Electrical Dysrhythmias in the Roux Jejunal Limb: Cause and Treatment

Peter Morrison, FRCS(I), Brent W. Miedema, MD, Lothar Kohler, MD, Keith A. Kelly, MD, Rochester, Minnesota

Electrical dysrhythmias in the Roux limb after Roux gastrojejunostomy are associated with upper gut stasis of food. The aim of this study was to de- termine the cause of the dysrhythmias and whether they could be eliminated with pacing. A set of four dogs (Group A) underwent three sequential opera- tions: placement of jejunal electrodes at sites corre- sponding to the Roux limb; construction of a Roux limb without vagotomy, gastrectomy, or gastroje- junostomy; and transthoracic trnncal vagotomy. A second set of five dogs (Group B) underwent trun- cal vagotomy, distal gastrectomy, and Roux gastro- jejunostomy with recording electrodes placed on the Roux limb and a pacing electrode situated at the proximal end of the limb. Electrical recordings were obtained on four separate occasions after each operation. In Group A dogs, orad and disordered propagation of jejunal pacesetter potentials oc- curred in the Roux limb 56 q- 5% of the time after limb construction but never before construction. The pattern was not changed with vagotomy. In Group B dogs, electrical dysrhythmias in the Roux limb also occurred and were corrected with electri- cal pacing. We concluded that electrical dysrhyth- mias in the canine Roux limb are secondary to the jejunal transection done during Roux limb con- struetion, and are not due to gastrectomy, gastroen- terostomy, or vagotomy. The dysrhythmias can be corrected with pacing.

" n recent years, it has been recognized that vagotomy, . distal gastrectomy and Roux-Y gastrojejunostomy

can lead to stasis of food in the upper gut [I]. Ectopic pacemakers that appear in the Roux limb after operation likely contribute in part to this Roux stasis syndrome. The pacemakers generate orally propagating pacesetter po- tentials (slow waves) that drive the contractions of the

From the Department of Surgery and the Digestive Disease Center, Mayo Clinic and Mayo Foundation, Rochester, Minnesota. This work was supported in part by Grants DK18278, DK34988, and TW03296 from the National Institutes of Health, Bethesda, Maryland, and by the Mayo Foundation, Rochester, Minnesota. A portion of this work was published in abstract form in Gastroenterology 1985; 88: 1508.

Requests for reprints should be addressed to Keith A. Kelly, MD, Digestive Disease Center, Mayo Clinic, 200 First Street Southwest, Rochester, Minnesota 55905.

Manuscript submitted March 15, 1990, and accepted March 29, 1990.

Roux limb in a reverse or orad direction [2,3]. It is un- clear what aspect of the Roux procedure produces the electrical abnormality. The jejunal transection, the trun- cal vagotomy, and the gastroenterostomy may all have a role.

The aim of this study was to determine what aspect of the Roux operation produces electrical abnormalities in the Roux limb. This was accomplished by doing a series of three operations in dogs. Jejunal electrical activity was first monitored with an intact jejunum, then after the Roux-Y procedure without gastroenterostomy, and once again after a truncal vagotomy. In addition, dogs with vagotomy, distal gastrectomy, and Roux-Y gastrojejun- ostomy served to confirm abnormalities and to determine whether the orally propagating pacesetter potentials could be suppressed with pacing.

MATERIAL AND METHODS Animal preparation: Two groups of female mongrel

dogs weighing 14 to 16 kg were studied. All procedures and studies were approved and performed according to criteria set by the Institutional Animal Care and Use Committee of the Mayo Clinic. The first group of four dogs (Group A) underwent three sequential operations. In the first procedure, five silver-silver chloride electrodes were attached to the serosal surface of the jejunum at 5- cm intervals starting 20 cm distal to the ligament of Treitz (Figure I). The electrodes were connected by insu- lated copper wires to a multipinned socket contained in a stainless steel cannula that traversed the abdominal wall. At a second operation, the jejunum was divided 15 cm distal to the ligament of Treitz, the distal end of the transected bowel was oversewn, and the proximal end was anastomosed to the jejunum 50 cm distal to the transec- tion in an end-to-side manner. In the third operation, the dogs underwent truncal vagotomy performed transtho- raclcally. The dogs were maintained on pulverized dog food.

A second group of five dogs (Group B) underwent truncal vagotomy, distal two-thirds gastrectomy, and Roux-Y gastrojejunostomy (Figure 2). The Roux limb was created by transecting the jejunum 15 cm distal to the ligament of Treitz. The proximal end of the transec- tion was anastomosed to the jejunum 50 cm distal to the transection. The distal end of the transection was over- sewn and a gastroenterostomy constructed immediately distal to the closure. A bipolar platinum stimulating elec- trode was attached to the proximal end of the Roux limb. Five monopOlar silver-silver chloride electrodes were su- tured to the serosal surface of the Roux limb at 5-era intervals beginning 5 cm distal to the gastrojejunostomy.

Conduct of the studies: At least a 2-week recovery period was allowed after each surgical procedure. The

252 THE AMERICAN JOURNAL OF SURGERY VOLUME 160 SEPTEMBER 1990

ELECTRICAL DYSRHYTHMIAS IN THE ROUX JEJUNAL LIMB

Vagus

E2 "2 ,~)

Jelunum :~ I /E /Z50c . ' -54cm 1 m

E 2 . . . . t ~( / ,

E4 "~

Figure 1. Sequential operative prepara- E~ tigris in Group A dogs. Placement of monopoiar jejunal recording electrodes (left) was followed by construction of a Roux-Y jejunal limb (center) and then by transthoracic truncal vagotomy (right). E ~ ~ = electrode.

dogs were then fasted overnight for 12 hours and were placed in a canvas sling where they rested quietly. Ele~ri- cal activity was recorded using a Brush Mark 200 pen recorder with a time constant of 1 second.

The dogs in Group A had recordings obtained for two complete cycles of the interdigestive migrating myoelec- tric complex (IMMC). The dogs in Group B also had electrical recordings obtained for two complete cycles of the IMMC. At the completion of the second cycle, the dogs underwent pacing of the Roux limb with a square pulse wave generator. A pulse of 5 mA for 50 msec was given at a rate of 18 pulses/minute for 20 minutes. All studies were repeated four times on separate days in every dog after each operation.

Data analysis: All tracings were analyzed for fre- quency of the pacesetter potentials and direction of prop- agation of the potentials. Aboral propagation of the pace- setter potential over all electrodes was considered the usual pattern of health (Figure 3). Two types of electrical dysrhythmias were noted. The first was oral pacesetter potentia! propagation over the segment of bowel moni- tored with the electrodes (Figure 4). Second, both oral and aboral propagation could be seen at the same time in the segment (Figure 5). The percent of electrical dysr- hythmias was determined in all tracings.

Statistical comparisons were done using the Student's t-test for paired or unpaired data as appropriate. Because the direction of pacesetter potential propagation in the intact jejunum and with pacing was always in the aboral direction, comparisons with this parameter were done with a single group t-test using a constant of 95%.

RESULTS Group A: In the intact canine jejunum, propagation

of the pacesetter potentials was always in the aboral di, rection (Figure 3). After creation of the Roux-Y limb of

l r I1

Roux-en-Y gastrojejunostomy

P Pacing - electrode

odenum

-_j "///ltl~ \ ,,xl!K

q ,

15cm

Gastric remnant

" - " 5cm

�9 . . . . " , ' 5cm

. " / ~ R o u x limb

unum

c~.

Figure 2. Operative preparation in Group B dogs, consisting of vagotomy, distal two-thirds gastrectomy, and Roux-Y gastroenter- ostomy with placement of a pacing electrode and evenly spaced recording electrodes (e) on the Roux limb.

THE AMERICAN JOURNAL OF SURGERY VOLUME 160 SEPTEMBER 1990 253

MORRISON ET AL

I

l t t

t t I

t

E3 11 i I I I I ! I I I t

I t I 7 1 1 i _1 2 mV 1

I I 2 see

Figure 3. Recording of jejunal electrical activity in healthy dogs showing consis- tent ab0ral propagation of pacesetter po- tentials (dotted I!ne$). E1 through E5 rep- resent serosal electrodes placed 5 cm apart on the jejunum proceeding proxi- mal to distal, with E1 located 20 cm distal to the ligament of Trietz.

I I I I

I !

I I 1j2 mY

, I 2 sec

Figure 4. Recording of canine jejunal electrical activity showing orally propa- gating pacesetter potentials ( d o t t e d lines) in a Roux jejunal limb. Electrodes (E) placed as in Figure 3.

jejunum, a drop in the frequency of the pacesetter poten- tials occurred in the Roux limb,i and oral or oral and aboral propagation of the potentials was found 56% of the time (Figures 4 and 5, Table I). The addition of vagotomy did not change the frequency of the pacesetter potentials or the incidence of abnormal pacesetter potential propa- gation.

Group B: All dogs had periods of abnormal propaga- tion of pacesetter potentials in the Roux limb. Overall, electrical dysrhythmias were present 43% of the time (Table I). The dysrhythmias Of the paeesetter potentials found in the Roux limbs of the Group B dogs with vagoto- my, gastric resection, and a Roux gastroenterostomy

were similar to those found in Group A dogs with Roux limb construction only. The disturbed pattern of paceset, ter potentials in the Roux limb of the Group B dogs could be consistently converted to a healthy pattern with aboral propagation of pace.setter potentials by pacing the proxi- mal end of the Roux limb. Pacing returned the frequency of the Roux pacesetter potentials to that found in the intact jejunum without pacing (about 18 cycles/min) and restored the aboral direction of propagation.

COMMENTS The major finding in this study was that abnormal

electric patterns in the canine Roux limb were due to

2 5 4 T H E A M E R I C A N J O U R N A L OF S U R G E R Y V O L U M E 160 SEPTEMBER 1990

ELECTRICAL DYSRHYTHMIAS IN THE ROUX JEJUNAL LIMB

Figure 5. Recording of canine jejunal electrical activity showing pacesetter potentials propagating orally and aborally from an ectopic pacemaker located be- tween E3 and Es in a Roux jejunal limb. Electrodes (E) placed as in Figure 3.

E3

E4

I I

! ! I I

I I

i V v

I

2 sec

jejunal transection and not to vagotomy, gastrectomy, or gastrojejunostomy. The transection of the jejunum dur- ing Roux limb construction disconnected the limb from the influence of the pacemaker of the small intestine located in the proximal duodenum [4]. This decreased the pacesetter potential frequency in the Roux limb and al- lowed ectopic pacemakers to arise within it. The pace- makers led to oral or disordered propagation of the Roux limb pacesetter potentials. The disordered electrical pat- tern in the Roux limb likely contributed to stasis in the Roux limb and caused the slow gastric emptying found in earlier tests [3].

Others have postulated that the jejunojejunostomy made during construction of the Roux limb may allow pacesetter potentials from the duodenum to propagate across the anastomosis and drive the limb backwards [5]. This seems unlikely, because the frequency of the Roux limb pacesetter potentials (14 cycles/min) was not as fast as the frequency of the duodenal pacesetter potentials (18 cycles/min). Also, sometimes the ectopic pacemakers in the Roux limb appeared near the proximal or mid-portion of the limb (Figure 5) and not at its distal end near the anastomosis. Lastly, convincing evidence of propagation

of pacesetter potentials across an enteric anastomosis has not as yet been shown.

The addition of vagotomy did not change the electri- cal activity of the Roux limb. This confirms previous work in the opossum and man [6,7]. The retention of solids in the canine Roux limb noted by others after vagotomy cannot be explained by a vagal-induced change in the jejunal electrical pattern [8]. The influence of vagotomy on the Roux stasis syndrome in humans is controversial. A large study showed no difference in the incidence of Roux stasis with or without vagotomy [9]. Others have implicated vagotomy as a contributor to the Roux stasis syndrome. Vagal denervation of the gastric remnant most likely does produce electrical abnormali- ties of the stomach and gastric atony that could contrib- ute to the Roux stasis syndrome [I0].

The hormonal changes induced by distal gastrectomy and gastroenterostomy have also been implicated as pos- sible mediators of disordered Roux motility [11]. In our study, however, the frequency or direction of the paceset- ter potentials in the Roux limb was not altered by the addition of distal gastrectomy and gastroenterostomy. Nonetheless, resection of the antrum will decrease gastrin

T A B L E I Characteristics of the Jejunal Pacesetter Potential (PP)

Group A (n = 4 dogs) Group B (n = 5 dogs) Vagotomy and Roux

Intact Roux Jejunal Roux Limb Gastrectomy Jejunal Parameter* Jejunum Limb Only and Vagotomy No Pacing Roux Pacing

PP frequency (cycles/min) 18.2 4- 0.5 14,2 4- 0.61 13.9 4- 0.4 t 14.5 4- 0.31 18.0 4- 0 Electrical 0 56.0 4- 5.0 t 53.0 4- 3.0 t 43.0 4- 9.0 + 0

dysrhythmias (% of PPs)

* Mean 4- SEM. t p <0,05 vs intact jejunum.

THE AMERICAN JOURNAL OF SURGERY VOLUME 160 SEPTEMBER I990 2 5 5

MORRISON ET AL

release. Furthermore, duodenal release of gastrointesti- nal hormones is bypassed with the conventional Roux procedure, and this could also affect Roux limb motility.

The abnormal electrical patterns in the Roux limb were consistently changed to a normal pattern by pacing the proximal end of the Roux limb. Pacing of the Roux limb in the rat can improve gastric emptying after Roux gastrojejunostomy [12]. Improved gastric emptying of liquids had been documented with pacing of the canine Roux limb [3]. Initial attempts to pace the Roux limb in humans have been unsuccessful [13]. Pacing more proxi- mally in the Roux limb, or changing the pacing stimulus parameters, may overcome these difficulties.

A consistent drop in the frequency of the pacesetter potential was s~n in the canine Roux limb compared with the intact bowel. This drop in frequency has also b~n observed in humans but to a lesser degr~ [13]. It is conceivable that this drop in frequency may contribute to upper gut stasis after Roux gastrojejunostomy. However, no data are currently available addressing this point.

In conclusion, orally propagating pacesetter potentials appear in the canine Roux limb after operation. The electrical dysrhythmia is due to the jejunal transection, and not to the vagotomy, the gastrectomy, or the gastroje- junostomy. The electrical abnormality probably contrib- utes, at least in part, to the upper gut stasis of food found in dogs and perhaps humans after Roux gastrojejunos- tomy. Pacing the proximal end of the Roux limb with electrical stimuli abolishes the electrical dysrhythmia in dogs. Pacing may have a role in the treatment of the Roux stasis syndrome in humans.

Electrical dysrhythmias in the Roux limb can be corrected with pacing. These data indicate that this clini- cally important syndrome is not due to gastrectomy, gastroenterostomy, or vagotomy.

REFERENCES 1. Mathias JR, Fernandez A, Sninsky CA, Clench MH, Davis RH. Nausea, vomiting and abdominal pain after Roux-en-Y anastomo- sis: motility of the jejunal limb. Gastroenterology 1985; 88:101-7. 2. Karistrom LH, Soper N J, Kelly KA, Phillips SF. Ectopic jejunal pacemakers and entcrogastric reflux after Roux gastrectomy: effect of intestinal pacing. Surgery 1989; 106: 486-95. 3. Karlstrom L, Kelly KA. Ectopic jejunal pacemakers and gastric emptying after Roux gastrectomy: effect of intestinal pacing. Sur- gery 1989; 106: 867-71. 4. Hermon-Taylor J, Code CF. Localization of the duodenal pace- maker and its role in the organization of duodenal myoelectric activity. Gut 1971; 12: 40-7. 5. Vantrappen G, Coremanns G, Janssens J, Penninckx F, Kerre- roans R. Inversion of the slow wave frequency gradient in symptom- atic patients with Roux-en-Y anastomosis [Abstract]. Gastroenter- ology 1988; 94: A477. 6. Gidda JS, Goyal RK. Influence of vagus nerves on electrical activity of opossum small intestine. Am J Physiol 1980; 239: G406- G410. 7. Papasova M, Atanassova E. Adaptation to surgical perturba- tions. In: Schultz SG, eel. Handbook of physiology. Section 6: the gastrointestinal system (vol. 1, part 2). Bethesda: American Physio- logical Society, 1989: 1199-224. 8. Vogel SB, Vair DB, Woodward ER. Alterations in gastrointesti- nal emptying of 99m-technetium-labelext solids following sequential antrectomy, truncal vagotomy and Roux-Y gastrcenterostomy. Ann Surg 1983; 198: 506-15. 9. Gustavsson S, Ilstrup DM, Morrison P, Kelly KA. Roux-Y stasis syndrome after gastrectomy. Am J Surg 1988; 155: 490-4. 10. Kelly KA, Code CF. Effect of transthoracic vagotomy on canine gastric electric activity. Gastrcenterology 1969; 57: 51-8. 11. Hocking MP, Vogel SB, Falasea CA, Woodward ER. Delayed gastric emptying of liquids and solids following Roux-en-Y biliary diversion. Ann Surg 1981; 194: 494-501. 12. Sawchuk A, Canal D, Grosfeld JL, et al. Electrical pacing of the Roux limb resolves delayext gastric emptying. J Surg Res 1987; 42: 635-41. 13. Richter HM III, Kelly KA. Effect of transection and pacing on human jejunal pacesetter potentials. Gastroenterology 1986; 91: 1380-5.

256 THE AMERICAN JOURNAL OF SURGERY VOLUME 160 SEPTEMBER 1990