The objective of this study was to compare the inflammatory response within the abdominal cavity between three surgical methods. The study comprised 45 cows with left displacement of the abomasum, which were allocated into three groups (n = 15). Right flank laparotomy and omentopexy (group R), left flank laparotomy and omentopexy (group L), and laparoscopic abomasopexy (group J) have been applied. Laparoscopic abomasopexy was the only technique that requires perforation of the abomasal wall. Blood and peritoneal fluid (PF) samples were obtained before, and on days 1, 2 and 3 after surgery. Macroscopic and microscopic evaluation of PF were performed. Cytological and biochemical parameters were analysed in blood and PF. No bacteria were present in PF after surgery. The number of PF leukocytes increased in all groups on day 1 after surgery with the highest value after laparoscopy (median, 1st quartile, 3rd quartile, R: 13.1, 6.4, 16.0; L: 13.6, 9.9, 17.4; J: 33.7, 21.1, 46.9 G/l). Laparotomy resulted in an increase of blood and PF CK on day 1 after surgery, whereas, laparoscopy caused an increased PF CK only. All groups had elevated PF D-dimer concentrations before surgery, with further increase in groups R and L on day 1 after surgery.
Statistics from Altmetric.com
Left displacement of the abomasum (LDA) is a frequent pathological condition in lactating dairy cows. Although the incidence varies considerably in different countries, under different management, and different feeding conditions, LDA seems to occur more frequently nowadays (Geishauser 1995, Stengarde and Pehrson 2002). Therefore, surgical correction of LDA is a common therapeutic procedure for cattle practitioners.
A variety of medical and surgical treatment methods of LDA has been described. In most cases, appropriate surgical fixation of the abomasum is necessary after the displacement has been corrected. Among the surgical methods, laparotomy and omentopexy in the right flank (Gabel and Heath 1969) and left flank laparotomy and right ventral omentopexy, which is also known as Utrecht method (Lagerweij and Numans 1962), are probably the most commonly used techniques. In addition, the laparoscopic-guided abomasopexy methods have also been introduced recently (Janowitz 1998, Christiansen 2004, Mulon and others 2006).
The general principle of surgical treatment of abomasal displacement is to induce a local sterile inflammation and adhesion between the abomasum or omentum and the abdominal wall, to fix the abomasum permanently in its normal position. The fixation of the abomasum can be performed by abomasopexy or omentopexy. Omentopexy is generally not associated with perforation of the abomasal wall. Abomasopexy can be done either with or without perforation of the abomasal wall depending on the surgical method. Percutaneous bar suture techniques (Grymer and Sterner 1982) and the majority of laparoscopic methods (Janowitz 1998, Christiansen 2004) require a perforation of the abomasal wall to introduce a toggle pin that is used to fix the abomasum in its appropriate position. It is still the subject of ongoing debate whether minimal invasive laparoscopic techniques result in a sufficient local inflammatory reaction, which is considered as prerequisite for permanent fixation of the abomasum, and whether intentional perforation of the abomasal wall may result in bacterial contamination of the abdominal cavity and subsequent septic peritonitis (Kehler and Stark 2002, Mulon and others 2006, Seeger and others 2006). Several studies described better clinical recovery, rapid increase of feed intake and rumen motility, and shorter duration of abomasal hypomotility after laparoscopy (Seeger and others 2006, Wittek and others 2009). By contrast, in a field study published by Roy and others 2008, laparoscopic correction of LDA in dairy cows did not affect clinical recovery time and rate (appetite and milk yield), compared with laparotomy.
The effect of laparotomy on peritoneal fluid (PF) measurements in cattle has already been described (Oehme and Noordsy 1970, Anderson and others 1994). However, only routinely used parameters (colour, specific gravity, total protein and leukocyte count and differentiation) have been used to characterise intra-abdominal inflammatory processes. Oehme and Noordsy (1970) found an increased percentage of neutrophils (70 per cent) in PF after exploratory laparotomy. Anderson and others (1994) compared PF parameters before, and one, two and six days after exploratory laparotomy and omentopexy in 16 healthy dairy cows. Total protein and the number of polymorphnuclear cells increased after surgery. To our best knowledge, effects of laparoscopic correction and fixation of the abomasum by abomasopexy on cytological and biochemical parameters in PF of cows with LDA have not been published.
Specific PF enzymes (LDH, ALP and CK) are known to be suitable to describe cell damage within the abdominal cavity. Furthermore, increased L-lactate was considered to be indicative for ischaemia. D-dimer could be used as a marker for inflammation and ischaemia, and translocation of bacteria through a damaged intestinal wall into the abdominal cavity is characterised by decreased PF glucose (Wittek and others 2010a, b). The effect of LDA and RDA on composition of PF has been described recently (Grosche and others 2012). However, this study did not compare changes of biochemical parameters in PF depending on the surgical methods used for correction of abomasal displacement.
The objectives of the following study are to evaluate postoperative clinical recovery of dairy cows after surgical correction of LDA by using left or right flank laparotomy and omentopexy, or laparoscopy and abomasopexy, and to compare the inflammatory response within, and release of bacteria into the abdominal cavity between these three surgical methods. We hypothesise that all treatment methods result in cytological and biochemical changes of PF indicating cell damage and inflammation, but the severity of these changes will be less after laparoscopy, despite perforation of the abomasum.
Material and methods
The study comprised 45 lactating dairy cows (Holstein-Frisian or Holstein-Frisian German black pied cross) which were referred to the veterinary hospital (Leipzig University) for treatment of LDA. The cows had a mean age of 3.8 ± 2.0 years, and they were on average (21, 14, 34; median, 1st and 3rd quartile) days in milk. The animals were thoroughly examined at time of admission. LDA was diagnosed by simultaneous percussion and auscultation of the left abdomen, and later confirmed during surgery. Results of physical and laboratory exams before surgery and on three consecutive days after surgery were included in the study: day 0, before surgery; and days 1, 2 and 3, after surgery. Physical examinations were performed twice daily, including body temperature, heart and respiration rate, and rumen motility (number of rumen contractions per three minutes). Food intake was assessed by visual appraisal of consumption (percentage of normal ration per day). The ration for each cow was calculated based on a 20 kg milk yield per day, consisting of grass silage, concentrates and hay. Additionally, cows could feed on straw which was used as bedding material.
An intravenous catheter (14 G, 20 cm; Walter Veterinar-Instrumente, Rietzneuendorf, Germany) was placed in the right jugular vein before surgery, and was used for blood sampling, and application of intravenous fluids and drugs.
Cows were alternately allocated into three treatment groups: group R, right flank laparotomy and omentopexy; group L, left flank laparotomy and omentopexy at the right ventral abdomen (Utrecht method); group J, two-step laparoscopy (Janowitz method) and abomasopexy at the right ventral abdomen. All three surgical procedures were performed between 9:00 and 12:00 by one experienced surgeon (TW). Before surgery, cows received an antibiotic (oxytetracyclin, 5 mg/kg intravenous twice a day; Ursocyclin, Serumwerk Bernburg, Bernburg, Germany) and anti-inflammatory treatment (metamizol sodium, 20 mg/kg intravenous twice a day; Metapyrin, Serumwerk Bernburg, Bernburg Germany). These treatments were continued twice daily until the end of the study. Intravenous fluids (20 litres of 0.9 per cent NaCl solution, and 4 litres of 40 per cent glucose solution; Serumwerk Bernburg, Bernburg, Germany) were administered one hour before surgery, and continued over a period of 24 hours postsurgery. Depending on the surgical method, local and/or regional anaesthesia of the left or right flank, or of the ventral abdomen, was performed with procaine hydrochloride (Isocain 2 per cent, Selectavet, Weyarn-Holzolling, Germany) 15 minutes before surgery.
Briefly, right flank laparotomy was performed after the region has been surgically prepared and anaesthetised with a combination of distal paravertebral anaesthesia and reverse L-block. The abdominal cavity was palpated thoroughly, and the displaced abomasum was located on the left side. After decompression, the abomasum was pulled underneath the rumen from the left to the right ventral abdomen. The greater omentum was exteriorised and the ‘pig ear’ and pylorus were identified. The area of the greater omentum, 10 cm caudal of the pylorus was fixed to the right abdominal wall ventral of the abdominal incision by sutures resulting in a right lateral omentopexy.
Left flank regional anaesthesia and laparotomy were performed similar to the right flank approach. The left displaced abomasum was visible in the left flank after laparotomy. Stay sutures were placed at the greater omentum close to the greater curvature. After decompression, the abomasum was pushed underneath the rumen to the right ventral abdomen. The stay sutures were threaded to needles which were protected by the surgeon's hand, and taken to the ventral abdomen where they were stabbed through the ventral abdominal wall at the right site. Tightening the sutures from outside by a non-sterile assistant resulted in a right ventral omentopexy.
The two-step laparoscopy required the introduction of a special toggle pin into the left displaced abomasum via a trocar and cannula unit placed in the left flank as step one, with the cow in standing position. In a second step, the cow was placed on her back, and two trocars (endoscope and pliers) were introduced through the abdominal wall in the right ventral abdomen. The suture of the toggle pin was located, exteriorised through the trocar/cannula unit and tightened, resulting in an abomasopexy.
Blood samples were taken from the jugular vein before surgery (day 0) and 24 hours (day 1), 48 hours (day 2), and 72 hours (day 3) after surgery. Simultaneously, PF samples were obtained by abdominocentesis in the right caudal abdomen as described elsewhere (Wittek and others 2010a, Grosche and others 2012). Volume and turbidity of PF samples were assessed. Biochemical parameters (total protein, albumin, glucose, fibrinogen, CK, ALP and LDH) in blood and PF were assayed using an automatic analyser (Hitachi 912, Roche Diagnostics, Basel, Switzerland). Total bilirubin, urea, cholesterol and β-hydroxybutyrate were measured in serum only. Plasma L-lactate was analysed spectrophotometrically using a commercially available test kit (Sigma-Aldrich Deutschland GmbH, Taufkirchen, Germany). The concentration of the fibrin degradation product D-dimer was measured by a commercially available immunoassay (Nycocard D-Dimer assay and reader, Axis Shield AS, Oslo, Norway). The samples for D-dimer and L-lactate assays were stored at -21°C for up to one month before analysing. Recently published physiological ranges of biochemical measures in PF of dairy cows were used as reference values (Wittek and others 2010a, b).
White and red blood cell counts (WBC and RBC) were measured using an automatic analyser (Advia 120, Bayer Health Care, Division Diagnostica, Fernwald, Germany). Leukocyte differentiation (QuickDiff staining) in venous blood and PF was performed by one author (AG) who was not aware of the treatment protocol. Peritoneal fluid was also examined microscopically, and after Gram staining to detect bacteria.
Because data were mostly not normally distributed, results were expressed as median, 1st quartile, and 3rd quartile. Normality was tested with the Kolmogorov Smirnov Test. Data were log-transformed before parametric tests were applied. Analysis of variance (ANOVA) was used to compare between the treatment groups, and a repeated measure ANOVA was performed to compare between the experimental days of the study. If indicated, the Bonferroni t test was applied as posthoc test. A P value of <0.05 indicated statistical significant differences. A statistical programme SPSS 12 (SPSS 12.0, SPSS Inc, Chicago, USA) was used for all statistical analyses.
At time of admission, cows had a mildly increased body temperature (39.2, 38.4, 39.8°C; median, 1st and 3rd quartile), and decreased rumen motility (1, 0, 3 contractions per three minutes), but a normal respiratory (24, 18, 30 breaths per minute) and heart rate (72, 64, 82 beats per minute). Food intake was decreased (30/10/70 per cent of ration) before surgery. Cows of all groups were hyperbilirubinaemic (17.8, 14.4, 24.2 - mol/l), hyperketonaemic (1.83, 0.40, 3.44 mmol/l) and hypocholesterolaemic (1.59, 1.37, 2.44 mmol/l) at admission. Results of biochemical parameters in venous blood, and PF obtained before surgery (day 0), are summarised in Table 1. There were no differences of clinical and laboratory parameters between the groups before surgery. After surgical replacement of the abomasum, food intake improved gradually in all groups from 30 per cent (30, 10, 70 per cent on day 0, to 70 per cent (70, 50, 100 per cent) of the ration on day 3. In addition, rumen contractions increased in all groups over time (Fig 1) with a significantly faster recovery in group J. Consequently, serum β-hydroxybutyrate and bilirubin concentrations decreased in all groups, but they did not differ between the groups (data not shown).
Serum total protein and albumin did not change after surgery. Total protein and albumin in PF were slightly increased after surgery in all groups, but changes were not significant (Table 2). Fibrinogen concentrations in peripheral blood were not affected by the surgical procedures (data not shown). Peritoneal fluid fibrinogen increased slightly immediately after surgery, and decreased on days 2 and 3. However, the concentrations varied considerably, thus, no significant differences could be determined between the experimental days and the groups (Table 2). Surgical correction of LDA did not affect venous blood D-dimer concentrations in all groups (data not shown). However, D-dimer in PF exceeded the upper reference value of 0.6 mg/l18 on day 0 in all groups (Fig 2). In groups R and L, the concentration of D-dimer in PF increased further immediately after surgery, but decreased on day 3, where it was significantly lower compared with days 0 and 1 (Fig 2). By contrast, D-dimer in group J did not change during the study (Fig 2). A number of cows with LDA (31/45) which were equally distributed within the three groups had increased L-lactate concentrations in PF before surgery that decreased significantly immediately after surgery (day 1) in all groups (Table 2).
Venous blood and PF CK activities were increased on day 1 after surgery in groups R and L, but did not change in group J (Figs 3 and 4). Venous blood LDH did not change in any of the groups, however, in cows of groups R and L, PF LDH was significantly increased on day 1 (Fig 5). Activities of ALP in venous blood and PF were not affected by any of the surgical methods used for correction of LDA in the present study (data not shown).
The number of erythrocytes in PF increased significantly on the first day after surgery in group R (day 0: 0.03, 0.00, 0.09 T/l; day 1: 0.27, 0.15, 0.31 T/l) and group L (day 0: 0.03, 0.00, 0.09 T/l; day 1: 0.16, 0.07, 0.21 T/l), but not in group J (day 0: 0.03, 0.00, 0.09 T/l; day 1: 0.07, 0.00, 0.12 T/l).
On day 0, the number and differentiation of leukocytes in PF of all cows was normal (2.85, 1.65, 4.14 G/l) with 55.8 per cent (55.8, 49.8, 70.0 per cent) polymorphnuclear cells, and 34.5 per cent (34.5, 24.0, 41.7 per cent) lymphocytes. Epithelial cells, monocytes, eosinophils and basophils were present in very low percentages (data not shown). PF leukocytes increased immediately after surgery in all groups, however, group J had a significantly higher leukocyte number compared with groups R and L on day 1 (Fig 6). Then, leukocytes gradually decreased from day 1 to days 2 and 3 in all groups, but remained increased during the study period (Fig 6). The percentage of polymorphnuclear cells was increased immediately after surgery in all groups (80.5, 74.5, 87.0 per cent), whereas, the percentage of lymphocytes (15.5, 9.5, 20.0 per cent) decreased. The percentage of polymorphnuclear cells started to decrease gradually from day 2 (70.0, 58.5, 72.0 per cent) to day 3 (67.0, 52.5, 73.0 per cent), which was accompanied by a corresponding increase of lymphocytes in all groups (day 2: 20.5, 18.5, 32.0 per cent; day 3: 23.0, 18.5, 29.0 per cent). The number of leukocytes in venous blood was highly variable among the groups (8.40, 4.30, 10.50 G/l), thus, the number and differentiation in venous blood leukocytes did not differ between the groups at any time during the study period. By contrast with leucocytes in PF, the number of leukocytes in venous blood was not influenced by surgery.
In none of the groups could bacteria be detected in PF before or after surgery. This was associated with normal glucose concentrations in PF of all three groups after surgical correction and fixation of the displaced abomasum (Table 2).
The results of the study demonstrate that all surgical procedures for correction and fixation of LDA resulted in an aseptic inflammatory response within the abdominal cavity indicated by increased PF leukocytes, D-dimer concentrations and enzyme activities. In all experimental groups, the increased blood and peritoneal L-lactate concentrations decreased immediately after surgery once the displaced abomasum has returned to its normal anatomical position indicating mild ischaemia and increased abomasal L-lactate synthesis (Wittek and others 2004) that is known to occur with LDA.
Increased PF leukocytes and increased percentages of polymorphnuclear cells have been previously described in cattle after laparotomy (Oehme and Noordsy 1970, Anderson and others 1994). By contrast with Anderson and others (1994) who also found increased protein concentrations in PF after laparotomy and omentopexy, the present study did not show significant changes of PF protein after surgical correction of LDA. A possible reason could be differences in duration of the surgery, mechanical manipulation/invasiveness, and/or experience of the surgical personnel. Abdominal surgery and repositioning of LDA described by Anderson and others (1994) was performed during a surgical class by veterinary students, whereas, all surgical procedures in the present study were performed by an experienced surgeon. However, similar postoperative changes in PF (increased leukocyte number and percentages of polymorphnuclear cells, decreased percentages of lymphocytes, and increased concentration of total protein) have been found in other species after laparotomy as indicators of abdominal inflammation (Santschi and others 1988, Dehghani and others 2000).
The extent of tissue damage and bleeding caused by surgical intervention and manipulation was characterised by increased CK and LDH activities, and increased RBC in PF of all groups with higher values in PF of groups R and L after laparotomy compared with group J after laparoscopy. Together with the increased blood CK activities in groups R and L, the results confirm the hypothesis that laparoscopic correction of abomasal displacement can be considered less invasive than laparotomy.
In previous studies, PF D-dimer has been shown to be a precise parameter of intestinal ischaemia and peritoneal inflammation (Collatos and others 1995, Acosta and Bjork 2003, Altinyollar and others 2006). The mean concentrations of D-dimer in PF at day 0 in the present study were slightly increased in all cows with LDA based on the upper physiological limit of 0.6 mg/l (Wittek and others 2010a, b), which confirms findings that LDA can cause activation of the coagulation cascade and fibrinolysis in response to ischaemia and inflammation (Wittek and others 2004, Grosche and others 2012). Further increase of PF D-dimer present after laparotomy (groups R and L), but not after laparoscopy (group J), could be a result of the different extent of mechanical manipulation during surgery, which is considered minimal during laparoscopy.
Minimal invasive surgical techniques are known to result in a less severe and shorter duration of postoperative ileus in human patients and animals (Bohm and others 1995, Davies and others 1997, Hotokezaka and others 1997), which has also been demonstrated in cows after surgical correction of LDA (Wittek and others 2009). The severity of gastrointestinal hypomotility is related to the time and extent of surgical manipulation and subsequent inflammatory response (Poncet and Ivan 1984, Kalff and others 1998). Thus, immediate clinical recovery observed in cows after laparoscopic reposition of LDA and abomasopexy is most likely the result of prompt restoration of normal rumen and abomasal motility compared with right flank laparotomy and omentopexy (Seeger and others 2006, Wittek and others 2009). It has also been shown that repeated abdominocentesis had no impact on cytological and biochemical composition of PF in horses and cattle (Schumacher and others 1985, Juzwiak and others 1991, Burton and others 1997, Mendes and others 2005). Therefore, laboratory changes in PF in the present study are most likely a response to abomasal displacement and manipulation during surgery, and not caused by repeated abdominocentesis.
Although the abomasum was punctured to fix the abomasum to the ventral abdominal wall during laparoscopy, no bacteria were found in PF of any of the animals after surgery. However, the disproportional high number of PF leukocytes in cows of group J could indicate that a small amount of abomasal content was possibly leaking from the site of abomasopexy causing an increased accumulation of neutrophils. However, PF glucose as an indicator of bacterial presence did not decrease in group J after surgery, and the concentrations did not differ between the experimental groups. Neutrophils are innate immune cells that become activated immediately after acute injury to fight invading bacteria, and to protect the tissue from further damage. A similar inflammatory response characterised by an increase of PF leukocytes has been described after enterocentesis in horses (Schumacher and others 1985). In addition, local trauma and inflammation in cattle are often characterised by accumulation of fibrinogen and formation of fibrin covering the defect within minutes after injury (Radostits and others 2000) and preventing the surrounding peritoneum from contamination with abomasal content and/or bacteria.
In summary, the results of the present study prove that all three surgical methods for correction and fixation of LDA result in an aseptic peritonitis and activation of the coagulation system, which is the prerequisite for the fibrinous and later fibrous adhesion at the site of fixation. The two-step laparoscopic-guided abomasopexy caused significantly less tissue damage compared with laparotomy techniques. If correctly performed, the risk of bacterial contamination and subsequent septic peritonitis by using laparoscopic-guided reposition and abomasopexy is very low.
Provenance: not commissioned; externally peer reviewed
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.