| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
aw Woliñski2
1 Department of General and Gastrointestinal Surgery, Medical Centre of Postgraduate Education, Orlowski Hospital, Warsaw, Poland
2 The Kielanowski Institute of Animal Physiology and Nutrition, Polish Academy of Sciences, Jabonna, Poland
3 Department of Physiological Sciences, Faculty of Veterinary Medicine, Warsaw Agricultural University, Warsaw, Poland
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
0.05). Two weeks after surgery, the number of ICC located in the region of Auerbach's plexus was higher and adhesions in the abdominal cavity lower in the ‘uncut’ Roux group. In conclusion, in the pig model, preservation of smooth muscle and ICC network continuity in the proximal jejunum may play an important role in early postsurgical recovery.
(Received 23 July 2006;
accepted after revision 8 December 2006; first published online 14 December 2006)
Corresponding author A. Kiciak: Department of General Surgery and Gastroenterology, Medical Centre of Postgraduate Education, Orlowski Hospital, 231 Czerniakowska Street, 00-416 Warsaw, Poland. Email: adamkic{at}yahoo.com
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Roux-en-Y gastroenterostomy, designed by C. Roux in 1897 (Polk et al. 2001), is indicated in humans with stomach cancer, peptic and reflux disease. Construction of the Roux limb requires transection of the upper gut wall, which disrupts the continuity of smooth muscles, nerves and the network of ICC. It was hypothesized that the disruption of tissue continuity may be responsible for postsurgical motility disturbances (delayed gastric emptying, manifested by epigastric fullness, abdominal pain, nausea and vomiting) observed in 10–42% of patients (Noh, 2000). To avoid the development of postsurgical complications, an ‘uncut’ Roux gastroenterostomy has been proposed (Van Stiegmann & Gott, 1988). The ‘uncut’ Roux is fashioned from a loop gastrojejunostomy with the afferent limb occluded with staples, thus affecting tissue continuity to a much lesser degree in comparison with the Roux-en-Y procedure. In a canine study, the ‘uncut’ Roux operation abolished formation of ectopic pacemakers in the Roux limb and reduced the delay in gastric emptying found with the conventional Roux gastojejunostomy (Miedema & Kelly, 1992). Our clinical observations in patients suggested that the ‘uncut’ Roux procedure results in fewer gastrointestinal motility disorders and in shorter hospitalization compared with the Roux-en-Y procedure (Bielecki et al. 2003), possibly owing to earlier recovery of a normal electromyography pattern. It was also hypothesized that reduced postoperative gastrointestinal motility might facilitate the creation of intra-abdominal adhesions.
Human studies comparing the early postoperative electromyography or motility patterns of the upper gut in relation to the ICC network following Roux-en-Y and ‘uncut’ Roux procedures are lacking, mostly owing to ethical considerations. Previous gastrointestinal electromyography studies showed the usefulness of pigs as a model for humans, in particular when the telemetry technique of recording was employed (Gacsalyi et al. 2000; Yao et al. 2003). The electromyography parameters (electronic control activity (ECA) and electrical response activity duration and frequency) in the pig upper small intestine are similar to those reported in humans, though the duration of the three-phased myoelectrical migrating complex (MMC), the foremost myoelectrical/motility pattern presented in fasted and fed animals and humans (Szurszewski, 1969; Code & Marlett, 1975; Fleckenstein, 1978; Vantrappen et al. 1979), is shorter by 20–30 min in pigs (Groner et al. 1990).
The first aim of this study was to compare the electromyography pattern of the upper jejunum following the Roux-en-Y gastrojejunostomy with the ‘uncut’ Roux procedure using a pig model. The second aim was to compare the density of the ICC net and creation of intra-abdominal adhesions in the two surgical procedures. Interstitial cells of Cajal are positive for c-kit proto-oncogene, which encodes for a transmembrane tyrosine kinase receptor (CD-117) and can be detected by immunochemical techniques (Huizinga et al. 1995).
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Treatments and experiments were conducted according to the European Community regulations concerning the welfare of experimental animals. The 3rd Local Ethics Committee in Warsaw approved all the animal treatments. Twelve male weaned Polish White pigs, 8 weeks old and with an initial body weight between 10 and 13 kg, were used. The animals were weaned when they were 35 days old, then kept in individual cages (0.8 m x 1.2 m) in a temperature-controlled room (20°C) with a light on between 07.00 and 21.00 h, and fed with a commercial feed at 09.00 and 20.00 h. Feed was given at a rate of 2% of body weight for each meal. Water was allowed ad libitum. Food was reduced by half (morning portion) the day before surgery, and totally withdrawn on the evening before the surgery and during the next two postoperative days. Starting from the third postoperative day, the animals received water orally, and from the fourth postoperative day, oral feeding was introduced. Hydration was maintained intravenously prior to water being received orally. All clinical aspects evaluated in the postoperative period were blinded and monitored regularly.
First surgery: implantation of electrodes and telemetry transmitter
The pigs underwent surgery to implant three pairs of silver bipolar electrodes on the jejunum, a telemetry transmitter in the abdominal muscles and silicone tubing in the right external jugular vein according to Gacsalyi et al. (2000). Azaperone (4 mg (kg body weight)–1, I.M., Stressnil, Janssen & Cilag Pharma, Vienna, Austria) was given as premedication. General anaesthesia was induced with 4% halothane (Narcotan 0.01%, Leciva, Czech Republic) mixed with oxygen plus N2O (1:2, 2 dm3 min–1 gas flow), and was maintained with halothane at a concentration of 1.5–2% during the entire surgical procedure. Right flank laparotomy was performed. The electrodes were sutured on the greater curvature of the jejunum (10, 15 and 25 cm distal to the Treitz ligament). The electrode needles were orientated to record spiking activity of the circular muscles and reduce slow wave amplitude (Szurszewski, 1969). A three-channel telemetry transmitter implant (D70EEE, DSI, Oregon, MN, USA) was fixed extraperitoneally in a pocket between the abdominal muscles. A stainless-steel earth electrode was fixed to the abdominal muscles. The pigs were allowed 10 days for postoperative recovery, during which they were intramuscularly injected with antibiotics once every other day (15 mg (kg body weight)–1, amoxycilin, ClamoxylTM L.A., Pfizer, UK). For analgesia, Ketonal (ketoprofen, inj. 100 mg (2 ml)–1, Lek Pharmaceuticals, Ljubljana, Slovenia) at a dose of 1 mg (kg body weight)–1 was injected intramuscularly every 12 h after the first and second surgery for 3 days and whenever needed later.
Second surgery: Roux-en-Y limb or ‘uncut’ Roux procedure
The second operation was performed 10 days after the implantation of electrodes and implants, and consisted of either a Roux-en-Y limb (12–15 cm long) or ‘uncut’ Roux limb formation. The Roux-en-Y limb procedure consisted of an incision of the jejunum just distal to the ligament of Treitz, connecting the distal remnant with the stomach, and connecting the proximal (duodenal) remnant with the jejunal loop by an end-to-side anastomosis between electrodes II and III (Fig. 1). The ‘uncut’ Roux procedure involved side-to-side anastomosis of the first jejunal loop with the stomach, locking the proximal loop with a transversal stapler (Proximate TX linear stapler, Johnson & Johnson, USA) close to the gastrojejunal anastomosis, and fashioning a jejunojejunostomy of the proximal part (closer to the ligament of Treitz) with an afferent loop (Fig. 1). Two-layered continuous sutures with monofilament synthetic absorbable suture (Maxon 0.2-0.24 mm or 0.15-0.19 mm, SynetureTM, Tyco Healthcare, Norwalk, CT, USA) were used for stomach and intestine anastomoses. The pigs were given analgesics and antibiotics for up to 72 h if no risk factors occurred. If the surgery lasted longer than 2 h or contamination of the peritoneal cavity with bile or intestinal contents occurred, regular antibiotic (amoxicillin) treatment was introduced for six days.
|
All measurements taken were blind. A 24 h electromyography recording session was performed 2–3 days before the second surgery to obtain the daily control MMC characteristics. The electromyography recordings started immediately after the second surgery and were continued for 2 weeks. After that, the animals were killed intravenously by an overdose of pentobarbiturate (Thiopental, Biochemie GMbH, Austria), the position of the electrodes was verified, whole-tissue intestine samples were harvested for microscopy study, and the telemetry implants were removed.
The electromyography signals modulated into radio waves were received by a receiver antenna (RMC-1, DSI), which was placed under the animal's cage. The receiver was coupled to an analog output (DL10, DSI); each of three signal channels were filtered (high cut-off, 50 Hz; low cut-off, 10 Hz) and amplified (BioAmp, ADInstruments, Melbourne, Victoria, Australia) as described elsewhere (Gacsalyi et al. 2000). A four-channel PowerLab/4e (ADInstruments) and a PC computer were used to record, display and analyse the electromyography data. The 24 h consecutive recordings were analysed off-line using Chart v4.1 software (ADInstruments). The recordings were inspected visually for MMC. The duration (in s) of the electrical response activity (ERA), duration of MMC phases (in min), the root mean square (r.m.s.; in mV), which expresses the ‘power’ of the electromyography signal (Application Notes, ADInstruments), and the migration velocity of the ERA and phase III of MMC (in cm min–1) along the jejunum were analysed (Husebye et al. 1994; Gacsalyi et al. 2000). The MMC phases were classified according to Code & Marlett (1975). The phase IV is short in the pig proximal small intestine (several seconds) and therefore not included in our analysis. In pigs fed twice a day, after each feeding the MMC is replaced or masked by a ‘feeding pattern’ of 130 ± 15 min duration (Yao et al. 2003).
Analysis of ICC
Whole-tissue samples were taken from the three electrode sites and the gastrojejunal anastomosis area. The tissue samples were fixed in 9% formalin, embedded in paraffin blocks and sectioned. Prior to the immunochemistry, tissue sections were subjected to heat-induced epitope retrieval and adhered to silanized slides. The Cajal cell staining procedure included both the incubation protocol provided by DAKO® Retrieval Solution and, with polyclonal rabbit antihuman CD117, c-kit application directions in the Staining Procedure section (DAKO® NP-SERIES, DAKO Corporation, Carpinteria, CA, USA). In this study, the anti-c-kit was applied exclusively for quantitative assessment by light microscopy of the presence of Cajal cells, owing to their specific morphology and location. All measurements were blinded. A minimum of four slides was analysed from each tissue sample by light microscopy, and the number of ICC associated with Auerbach's plexus (ICC-AP) counted per field of vision (magnification x40). The following slides were used as controls: a negative control from the area of electrode I from pigs that were not operated with the Roux-en-Y or ‘uncut’ Roux procedure, and human fusocellular gastrointestinal stromal tumor (GIST) rich in CD-117-positive cells.
Intra-abdominal adhesions
All measurements taken were blind. No cross-reactivity with human matrix metalloproteinase-9 (MMP-9) (Amersham Biosciencies, UK) and human total tissue inhibitor of matrix metalloproteinase-1 (TIMP-1) (Amersham Biosciencies) was found in the samples of peritoneal cavity fluid from the experimental pigs (data not shown); therefore the adhesions in the peritoneal cavity were assessed using the four-degree macroscopic classification proposed by Zühlke et al. (1990). Accordingly, degree I is characterized as bluntly detachable organ adhesion, whereas degree IV is characterized as solid organ adhesions over broad areas, causing injuries during detachment.
Statistical analyses
The electromyography data were extracted from Chart onto a spreadsheet file and statistically analysed. The data were expressed as the means ± S.E.M. Student's t test or the Mann–Whitney U test, and one-way ANOVA followed by Tukey's post hoc test or Kruskal–Wallis test followed by Dunn's post hoc test were used to test the statistical differences between the control recordings and those following the treatments (GraphPad Prism v.4.1, GraphPad Software, San Diego, CA, USA). P < 0.05 was taken as the level of significance.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The animals recovered smoothly after the first surgery. In general, it took 2–3 days for them to reach normal food intake. The pigs lost 0.5 kg body weight on average and began gaining weight between days 5 and 6 after the first surgery. Within the first week after the second surgery, the daily food intake in the Roux-en-Y pigs was lower by ca 30% compared with ‘uncut’ Roux. At the end of the second week, the piglets from the ‘uncut’ Roux group had increased their weight by 18.0%, whereas those from the Roux-en-Y group had increased it by only 7.3% (P < 0.05). These results were consistent with the clinical observations, in that the piglets from the ‘uncut’ Roux group returned to their normal physical movement, felt hungry and looked for food within 2 days, whereas those from the Roux-en-Y group recovered within 3.5 days.
Small intestinal electromyography
Before the second surgery, a total of 16.7 ± 3.8 MMC cycles were recorded daily. There were no differences in the number of MMC cycles recorded during the day (7.1 ± 2.3) and night (8.2 ± 1.5; n.s.). No significant day–night differences were found in the duration of phases I and III of the MMC; however, the duration of phase II at night was significantly shorter compared with daytime. The signal power (r.m.s.) of phases I and II recorded during the night was significantly lower compared with daytime (Table 1).
|
|
|
|
|
|
|
Identification of ICC and classification of adhesions
The ICC cells were characterized according to the cell morphology, characteristics of immunoreactivity and location of cells. The number of CD-117-positive cells in the upper jejunum in the Roux-en-Y group was lower than in the ‘uncut’ Roux group (3.50 ± 0.71 versus 8.21 ± 1.29 cells per field of vision, P = 0.012). Figure 6 shows examples of different quantities of Cajal cells at the first electrode site in Roux-en-Y animals. There were substantial animal-to-animal variations in the number of CD-117-positive cells, but the intra-animal variation was relatively low.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Transection of the jejunum during the construction of the Roux limb separates the limb from the natural intestinal pacemaker located in the duodenum. Ectopic pacemakers then appear in the limb and trigger retrograde contractions in its proximal portion. These contractions slow the transit through the limb and result in delayed gastric emptying (Tu & Kelly, 1994). This was reflected in our electromyography traces as retrograde migration of spikes in the Roux limb. Moreover, the MMC in the Roux limb of patients after total gastrectomy with early dumping syndrome failed to show phase III (Tomita et al. 2003). This pattern was also frequently observed in our study. In contrast, the differences in the organization of the pacemaker between ‘uncut’ Roux and Roux-en-Y procedures resulted in totally different postoperative outcomes when focusing on the clinical features. Body weight gain differed significantly between the groups; during the two postoperative weeks, the pigs from the ‘uncut’ Roux group gained more than those from the Roux-en-Y group.
A good postoperative course in pigs after the ‘uncut’ Roux procedure was observed despite pacemaker potential transmission being slowed down through the staple line. It was concluded that stapling did not disrupt the ICC network or myenteric plexus continuity and therefore preserved a superior pacemaker in the duodenum domination. This was confirmed by regular (in direction and speed) propagation between electrodes II and III. Moreover, the speed of propagation was not affected by either staple line or side-to-side anastomosis. Richardson et al. (2000) observed duodenal pacemaker transmission recovery within 1 week postoperatively in five of 14 investigated mongrel dogs after placing an uncut staple line 25 cm beyond the Treitz ligament. In the examined pigs, the migration velocity of spike bursts did not completely recover within 14 postoperative days. In contrast, the increase in the power signal (r.m.s.) observed on electrode III during the whole experiment may suggest acceleration of motility in the upper jejunum. The mechanism remains unknown, but may be related to the presence of staples in the tissues; it may, however, be of help in compensating for the motility disturbances related to the creation of the intestinal loop. Obviously, the Roux-en-Y procedure lacked this mechanism.
Involvement of ICC
Takayama et al. (2002) suggested that the pacemaker system plays a crucial role in maintaining gut motility. Hudson et al. (2006), using rabbit-raised polyclonal antiserum to c-kit, were the first to identify the ICC in the porcine ileum. In the present study, similarly, the morphology, characteristics of immunoreactivity and location of the cells made it most likely that they were c-kit-immunoractive ICC. A high variability in the expression of these cells in the jejunum was observed. Although the ICC network density varied across the intestine, random sampling from the same intestine segments made the summarized results more reliable. Indeed, in the present study, the differences in the number of ICC-AP within one pig were significantly smaller in comparison with the differences between pigs. In both experimental groups, there were pigs expressing high as well as low numbers of ICC, in line with previous observations (Faussone-Pellegrini & Thuneberg, 1999). In the post mortem examination, we excluded the possibility of anastomosis retraction (postoperative oedema) or stenosis (operative mistake), which might reduce the ICC network density, as shown previously in murine small intestine (Chang et al. 2001). We found a correspondence between the number of ICC-AP in the intestine wall and quicker restoration of regular MMC cycles within a group. Both in the Roux-en-Y and ‘uncut’ Roux groups, the fastest MMC restoration was observed in those individuals with the largest number of ICC-AP, and the average number of ICC was higher in the ‘uncut’ Roux group compared with the Roux-en-Y group. Our data suggest that the recovery of MMC phases, and in turn gut motility, depends on the number of intestinal pacemaker cells. Moreover, we observed a relationship between the number of ICC and postoperative adhesions in the abdominal cavity, as assessed by macroscopic classification (Zühlke et al. 1990). It is speculated that quicker restoration of gut motility associated with an abundance of ICC-AP limits formation of abdominal adhesions. It is worth mentioning that gut motility affects local blood flow (and thereby healing); however, the circulatory response reflects mostly the nature of the intervention used to activate motility (Walus & Jacobson, 1981). Though no specific studies could be found, there are supporting examples from other areas, for instance, the repair of Achilles tendon rupture (Mortensen et al. 1999; Ertem et al. 2002). In contrast, early postoperative mobility after major abdominal surgery did not reduce the duration of postoperative ileus (Waldhausen & Schirmer, 1990).
In conclusion, our results obtained in this animal model suggest that surgical procedures that spare the ICC network in the gut provide a benefit in earlier restoration of the intestinal MMC and creation of fewer intraperitoneal adhesions. Since the ICC are fundamental with respect to the electrical control activity of myocytes, these cells are obvious targets for future rational and effective therapy of gut disorders.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
d Chirurgiczny 75, 732–740.Chang IY, Glasgow NJ, Takayama I, Horiguchi K, Sanders KM & Wa SM (2001). Loss of interstitial cells of Cajal and development of electrical dysfunction in murine small bowel obstruction. J Physiol 15, 555–568.
Code CF & Marlett JA (1975). The interdigestive myo-electric complex of the stomach and small bowel of dogs. J Physiol 246, 289–309.
Ertem K, Elmali N, Kaygusuz MA, Inan M, Ayan I, Guner G & Karakaplan M (2002). The effect of continuous passive motion after repair of Achilles tendon ruptures: an experimental study in rabbits [in Turkish]. Acta Orthop Traumatol Turc 36, 141–146.[Medline]
Faussone-Pellegrini MS & Thuneberg L (1999). Guide to the identification of interstitial cells of Cajal. Microsc Res Tech 15, 246–266.
Fintl C, Hudson NP, Mayhew IG, Edwards GB, Proudman CJ & Pearson GT (2004). Interstitial cells of Cajal (ICC) in equine colic: an immunohistochemical study of horses with obstructive disorders of the small and large intestines. Equine Vet J 36, 474–479.[Medline]
Fleckenstein P (1978). Migrating electrical spike activity in fasting human small intestine. Am J Dig Dis 23, 769–775.[CrossRef][Medline]
Gacsalyi U, Zabielski R & Pierzynowski SG (2000). Telemetry facilitates long-term recording of gastrointestinal myoelectrical activity in pigs. Exp Physiol 85, 239–241.[Abstract]
Groner JI, Altschuler SM & Ziegler MM (1990). The newborn piglet: a model of neonatal gastrointestinal motility. J Pediatr Surg 25, 315–318.[CrossRef][Medline]
Hudson NP, Mayhew IG & Pearson GT (2001). A reduction in interstitial cells of Cajal in horses with equine dysautonomia (grass sickness). Auton Neurosci 92, 37–44.[CrossRef][Medline]
Hudson NP, Mayhew IG & Pearson GT (2006). Interstitial cells of Cajal and electrical activity of smooth muscle in porcine ileum. Acta Physiol (Oxf) 187, 391–397.
Huizinga JD, Thuneberg L, Kluppel M, Malysz J, Mikkelsen HB & Bernstein A (1995). W/kit gene required for interstitial cells of Cajal and for intestinal pacemaker activity. Nature 373, 347–349.[CrossRef][Medline]
Husebye E, Hellstrom PM & Midtvedt T (1994). Intestinal microflora stimulates myoelectric activity of rat small intestine by promoting cyclic initiation and aboral propagation of migrating myoelectric complex. Dig Dis Sci 39, 946–956.[CrossRef][Medline]
Miedema BW & Kelly KA (1992). The Roux stasis syndrome. Treatment by pacing and prevention by use of an ‘uncut’ Roux limb. Arch Surg 127, 295–300.[Abstract]
Mortensen HM, Skov O & Jensen PE (1999). Early motion of the ankle after operative treatment of a rupture of the Achilles tendon. A prospective, randomized clinical and radiographic study. J Bone Joint Surg Am 81, 983–990.
Noh SM (2000). Improvement of the Roux limb function using a new type of ‘uncut’ Roux limb. Am J Surg 180, 37–40.[CrossRef][Medline]
Otterson MF & Sarr MG (1993). Normal physiology of small intestinal motility. Surg Clin North Am 73, 1173–1192.[Medline]
Polk HC, Cheadle WG & Franklin GA (2001). Principles of operative surgery. In Sabiston textbook of surgery, 16th edn, ed. Townsend CM, pp. 163–170. WB Saunders, Philadelphia.
Richardson WS, Spivak H, Hudson JE, Budacz MA & Hunter JG (2000). Teflon buttress inhibits recanalization of uncut stapled bowel. J Gastrointest Surg 4, 424–429.[CrossRef][Medline]
Suzuki N, Prosser CL & Dahms V (1986). Boundary cells between longitudinal and circular layers: essential for electrical slow waves in cat intestine. Am J Physiol Gastrointest Liver Physiol 250, G287–G294.
Szurszewski JH (1969). A migrating electric complex of canine small intestine. Am J Physiol 217, 1757–1763.
Takayama I, Horiguchi K, Daigo Y, Mine T, Fujino MA & Ohno S (2002). The interstitial cells of Cajal and a gastroenteric pacemaker system. Arch Histol Cytol 65, 1–26.[CrossRef][Medline]
Tomita R, Fujisaki S & Tanjoh K (2003). Studies on gastrointestinal hormone and jejunal interdigestive migrating motor complex in patients with or without early dumping syndrome after total gastrectomy with Roux-en-Y reconstruction for early gastric cancer. Am J Surg 185, 354–359.[CrossRef][Medline]
Tu BN & Kelly KA (1994). Motility disorders after Roux-en-Y gastrojejunostomy. Obes Surg 4, 219–226.[CrossRef][Medline]
Van Stiegmann G & Gott JS (1988). An alternative to Roux-en-Y for treatment of bile reflux gastritis. Surg Gynecol Obstet 166, 69–70.[Medline]
Vantrappen G, Janssens J, Peeters TL, Bloom SR, Christofides ND & Hellemans J (1979). Motilin and the interdigestive migrating motor complex in man. Dig Dis Sci 24, 497–500.[CrossRef][Medline]
Waldhausen JH & Schirmer BD (1990). The effect of ambulation on recovery from postoperative ileus. Ann Surg 212, 671–677.[Medline]
Walus KM & Jacobson ED (1981). Relation between small intestinal motility and circulation. Am J Physiol Gastrointest Liver Physiol 241, G1–G15.
Yao G, Wolinski J, Korczynski W & Zabielski R (2003). Daily changes in antroduodenal myoelectric activity in weaned pigs. Anim Sci 76, 273–281.
Zabielski R & Naruse S (1999). Neurohormonal regulation of the exocrine pancreas during postnatal development. In Biology of the Pancreas in Growing Aimals, ed. Pierzynowski SG & Zabielski R, pp. 151–191. Elsevier Science BV, Amsterdam.
Zühlke HV, Lorenz EM, Straub EM & Savvas V (1990). Pathophysiology and classification of adhesions. Langenbecks Arch Chir, Suppl. II, Verh Dtsch Ges Chir, 1009–1016.
|
Acknowledgements |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
